From c799ba6eab7a1bf02792e2ced933be3ae5f8daa2 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 11 Dec 2015 15:31:10 +0100
Subject: cfg80211: remove CFG80211_REG_DEBUG

Instead of having this Kconfig option, which just *floods* the
kernel log,
 * remove the per-channel prints that are fairly useless anyway
 * convert the conditional printing to pr_debug()

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/Kconfig b/net/wireless/Kconfig
index da72ed3..ec3bf30 100644
--- a/net/wireless/Kconfig
+++ b/net/wireless/Kconfig
@@ -61,19 +61,6 @@ config CFG80211_DEVELOPER_WARNINGS
 	  on it (or mac80211).
 
 
-config CFG80211_REG_DEBUG
-	bool "cfg80211 regulatory debugging"
-	depends on CFG80211
-	default n
-	---help---
-	  You can enable this if you want to debug regulatory changes.
-	  For more information on cfg80211 regulatory refer to the wireless
-	  wiki:
-
-	  http://wireless.kernel.org/en/developers/Regulatory
-
-	  If unsure, say N.
-
 config CFG80211_CERTIFICATION_ONUS
 	bool "cfg80211 certification onus"
 	depends on CFG80211 && EXPERT
diff --git a/net/wireless/reg.c b/net/wireless/reg.c
index 3b0ce1c..bc76b28 100644
--- a/net/wireless/reg.c
+++ b/net/wireless/reg.c
@@ -60,13 +60,6 @@
 #include "regdb.h"
 #include "nl80211.h"
 
-#ifdef CONFIG_CFG80211_REG_DEBUG
-#define REG_DBG_PRINT(format, args...)			\
-	printk(KERN_DEBUG pr_fmt(format), ##args)
-#else
-#define REG_DBG_PRINT(args...)
-#endif
-
 /*
  * Grace period we give before making sure all current interfaces reside on
  * channels allowed by the current regulatory domain.
@@ -178,12 +171,10 @@ enum nl80211_dfs_regions reg_get_dfs_region(struct wiphy *wiphy)
 	if (wiphy_regd->dfs_region == regd->dfs_region)
 		goto out;
 
-	REG_DBG_PRINT("%s: device specific dfs_region "
-		      "(%s) disagrees with cfg80211's "
-		      "central dfs_region (%s)\n",
-		      dev_name(&wiphy->dev),
-		      reg_dfs_region_str(wiphy_regd->dfs_region),
-		      reg_dfs_region_str(regd->dfs_region));
+	pr_debug("%s: device specific dfs_region (%s) disagrees with cfg80211's central dfs_region (%s)\n",
+		 dev_name(&wiphy->dev),
+		 reg_dfs_region_str(wiphy_regd->dfs_region),
+		 reg_dfs_region_str(regd->dfs_region));
 
 out:
 	return regd->dfs_region;
@@ -541,7 +532,7 @@ static DECLARE_DELAYED_WORK(crda_timeout, crda_timeout_work);
 
 static void crda_timeout_work(struct work_struct *work)
 {
-	REG_DBG_PRINT("Timeout while waiting for CRDA to reply, restoring regulatory settings\n");
+	pr_debug("Timeout while waiting for CRDA to reply, restoring regulatory settings\n");
 	rtnl_lock();
 	reg_crda_timeouts++;
 	restore_regulatory_settings(true);
@@ -583,7 +574,7 @@ static int call_crda(const char *alpha2)
 
 	if (!is_world_regdom((char *) alpha2))
 		pr_debug("Calling CRDA for country: %c%c\n",
-			alpha2[0], alpha2[1]);
+			 alpha2[0], alpha2[1]);
 	else
 		pr_debug("Calling CRDA to update world regulatory domain\n");
 
@@ -1130,42 +1121,6 @@ const char *reg_initiator_name(enum nl80211_reg_initiator initiator)
 }
 EXPORT_SYMBOL(reg_initiator_name);
 
-static void chan_reg_rule_print_dbg(const struct ieee80211_regdomain *regd,
-				    struct ieee80211_channel *chan,
-				    const struct ieee80211_reg_rule *reg_rule)
-{
-#ifdef CONFIG_CFG80211_REG_DEBUG
-	const struct ieee80211_power_rule *power_rule;
-	const struct ieee80211_freq_range *freq_range;
-	char max_antenna_gain[32], bw[32];
-
-	power_rule = &reg_rule->power_rule;
-	freq_range = &reg_rule->freq_range;
-
-	if (!power_rule->max_antenna_gain)
-		snprintf(max_antenna_gain, sizeof(max_antenna_gain), "N/A");
-	else
-		snprintf(max_antenna_gain, sizeof(max_antenna_gain), "%d mBi",
-			 power_rule->max_antenna_gain);
-
-	if (reg_rule->flags & NL80211_RRF_AUTO_BW)
-		snprintf(bw, sizeof(bw), "%d KHz, %d KHz AUTO",
-			 freq_range->max_bandwidth_khz,
-			 reg_get_max_bandwidth(regd, reg_rule));
-	else
-		snprintf(bw, sizeof(bw), "%d KHz",
-			 freq_range->max_bandwidth_khz);
-
-	REG_DBG_PRINT("Updating information on frequency %d MHz with regulatory rule:\n",
-		      chan->center_freq);
-
-	REG_DBG_PRINT("(%d KHz - %d KHz @ %s), (%s, %d mBm)\n",
-		      freq_range->start_freq_khz, freq_range->end_freq_khz,
-		      bw, max_antenna_gain,
-		      power_rule->max_eirp);
-#endif
-}
-
 static uint32_t reg_rule_to_chan_bw_flags(const struct ieee80211_regdomain *regd,
 					  const struct ieee80211_reg_rule *reg_rule,
 					  const struct ieee80211_channel *chan)
@@ -1240,20 +1195,19 @@ static void handle_channel(struct wiphy *wiphy,
 		if (lr->initiator == NL80211_REGDOM_SET_BY_DRIVER &&
 		    request_wiphy && request_wiphy == wiphy &&
 		    request_wiphy->regulatory_flags & REGULATORY_STRICT_REG) {
-			REG_DBG_PRINT("Disabling freq %d MHz for good\n",
-				      chan->center_freq);
+			pr_debug("Disabling freq %d MHz for good\n",
+				 chan->center_freq);
 			chan->orig_flags |= IEEE80211_CHAN_DISABLED;
 			chan->flags = chan->orig_flags;
 		} else {
-			REG_DBG_PRINT("Disabling freq %d MHz\n",
-				      chan->center_freq);
+			pr_debug("Disabling freq %d MHz\n",
+				 chan->center_freq);
 			chan->flags |= IEEE80211_CHAN_DISABLED;
 		}
 		return;
 	}
 
 	regd = reg_get_regdomain(wiphy);
-	chan_reg_rule_print_dbg(regd, chan, reg_rule);
 
 	power_rule = &reg_rule->power_rule;
 	bw_flags = reg_rule_to_chan_bw_flags(regd, reg_rule, chan);
@@ -1391,18 +1345,15 @@ static bool ignore_reg_update(struct wiphy *wiphy,
 		return true;
 
 	if (!lr) {
-		REG_DBG_PRINT("Ignoring regulatory request set by %s "
-			      "since last_request is not set\n",
-			      reg_initiator_name(initiator));
+		pr_debug("Ignoring regulatory request set by %s since last_request is not set\n",
+			 reg_initiator_name(initiator));
 		return true;
 	}
 
 	if (initiator == NL80211_REGDOM_SET_BY_CORE &&
 	    wiphy->regulatory_flags & REGULATORY_CUSTOM_REG) {
-		REG_DBG_PRINT("Ignoring regulatory request set by %s "
-			      "since the driver uses its own custom "
-			      "regulatory domain\n",
-			      reg_initiator_name(initiator));
+		pr_debug("Ignoring regulatory request set by %s since the driver uses its own custom regulatory domain\n",
+			 reg_initiator_name(initiator));
 		return true;
 	}
 
@@ -1413,10 +1364,8 @@ static bool ignore_reg_update(struct wiphy *wiphy,
 	if (wiphy_strict_alpha2_regd(wiphy) && !wiphy->regd &&
 	    initiator != NL80211_REGDOM_SET_BY_COUNTRY_IE &&
 	    !is_world_regdom(lr->alpha2)) {
-		REG_DBG_PRINT("Ignoring regulatory request set by %s "
-			      "since the driver requires its own regulatory "
-			      "domain to be set first\n",
-			      reg_initiator_name(initiator));
+		pr_debug("Ignoring regulatory request set by %s since the driver requires its own regulatory domain to be set first\n",
+			 reg_initiator_name(initiator));
 		return true;
 	}
 
@@ -1697,7 +1646,7 @@ static void reg_check_chans_work(struct work_struct *work)
 {
 	struct cfg80211_registered_device *rdev;
 
-	REG_DBG_PRINT("Verifying active interfaces after reg change\n");
+	pr_debug("Verifying active interfaces after reg change\n");
 	rtnl_lock();
 
 	list_for_each_entry(rdev, &cfg80211_rdev_list, list)
@@ -1779,8 +1728,8 @@ static void handle_channel_custom(struct wiphy *wiphy,
 	}
 
 	if (IS_ERR(reg_rule)) {
-		REG_DBG_PRINT("Disabling freq %d MHz as custom regd has no rule that fits it\n",
-			      chan->center_freq);
+		pr_debug("Disabling freq %d MHz as custom regd has no rule that fits it\n",
+			 chan->center_freq);
 		if (wiphy->regulatory_flags & REGULATORY_WIPHY_SELF_MANAGED) {
 			chan->flags |= IEEE80211_CHAN_DISABLED;
 		} else {
@@ -1790,8 +1739,6 @@ static void handle_channel_custom(struct wiphy *wiphy,
 		return;
 	}
 
-	chan_reg_rule_print_dbg(regd, chan, reg_rule);
-
 	power_rule = &reg_rule->power_rule;
 	bw_flags = reg_rule_to_chan_bw_flags(regd, reg_rule, chan);
 
@@ -2522,7 +2469,7 @@ static void restore_alpha2(char *alpha2, bool reset_user)
 	if (is_user_regdom_saved()) {
 		/* Unless we're asked to ignore it and reset it */
 		if (reset_user) {
-			REG_DBG_PRINT("Restoring regulatory settings including user preference\n");
+			pr_debug("Restoring regulatory settings including user preference\n");
 			user_alpha2[0] = '9';
 			user_alpha2[1] = '7';
 
@@ -2532,24 +2479,24 @@ static void restore_alpha2(char *alpha2, bool reset_user)
 			 * back as they were for a full restore.
 			 */
 			if (!is_world_regdom(ieee80211_regdom)) {
-				REG_DBG_PRINT("Keeping preference on module parameter ieee80211_regdom: %c%c\n",
-					      ieee80211_regdom[0], ieee80211_regdom[1]);
+				pr_debug("Keeping preference on module parameter ieee80211_regdom: %c%c\n",
+					 ieee80211_regdom[0], ieee80211_regdom[1]);
 				alpha2[0] = ieee80211_regdom[0];
 				alpha2[1] = ieee80211_regdom[1];
 			}
 		} else {
-			REG_DBG_PRINT("Restoring regulatory settings while preserving user preference for: %c%c\n",
-				      user_alpha2[0], user_alpha2[1]);
+			pr_debug("Restoring regulatory settings while preserving user preference for: %c%c\n",
+				 user_alpha2[0], user_alpha2[1]);
 			alpha2[0] = user_alpha2[0];
 			alpha2[1] = user_alpha2[1];
 		}
 	} else if (!is_world_regdom(ieee80211_regdom)) {
-		REG_DBG_PRINT("Keeping preference on module parameter ieee80211_regdom: %c%c\n",
-			      ieee80211_regdom[0], ieee80211_regdom[1]);
+		pr_debug("Keeping preference on module parameter ieee80211_regdom: %c%c\n",
+			 ieee80211_regdom[0], ieee80211_regdom[1]);
 		alpha2[0] = ieee80211_regdom[0];
 		alpha2[1] = ieee80211_regdom[1];
 	} else
-		REG_DBG_PRINT("Restoring regulatory settings\n");
+		pr_debug("Restoring regulatory settings\n");
 }
 
 static void restore_custom_reg_settings(struct wiphy *wiphy)
@@ -2661,14 +2608,14 @@ static void restore_regulatory_settings(bool reset_user)
 	list_splice_tail_init(&tmp_reg_req_list, &reg_requests_list);
 	spin_unlock(&reg_requests_lock);
 
-	REG_DBG_PRINT("Kicking the queue\n");
+	pr_debug("Kicking the queue\n");
 
 	schedule_work(&reg_work);
 }
 
 void regulatory_hint_disconnect(void)
 {
-	REG_DBG_PRINT("All devices are disconnected, going to restore regulatory settings\n");
+	pr_debug("All devices are disconnected, going to restore regulatory settings\n");
 	restore_regulatory_settings(false);
 }
 
@@ -2716,10 +2663,10 @@ int regulatory_hint_found_beacon(struct wiphy *wiphy,
 	if (!reg_beacon)
 		return -ENOMEM;
 
-	REG_DBG_PRINT("Found new beacon on frequency: %d MHz (Ch %d) on %s\n",
-		      beacon_chan->center_freq,
-		      ieee80211_frequency_to_channel(beacon_chan->center_freq),
-		      wiphy_name(wiphy));
+	pr_debug("Found new beacon on frequency: %d MHz (Ch %d) on %s\n",
+		 beacon_chan->center_freq,
+		 ieee80211_frequency_to_channel(beacon_chan->center_freq),
+		 wiphy_name(wiphy));
 
 	memcpy(&reg_beacon->chan, beacon_chan,
 	       sizeof(struct ieee80211_channel));
@@ -2798,8 +2745,7 @@ bool reg_supported_dfs_region(enum nl80211_dfs_regions dfs_region)
 	case NL80211_DFS_JP:
 		return true;
 	default:
-		REG_DBG_PRINT("Ignoring uknown DFS master region: %d\n",
-			      dfs_region);
+		pr_debug("Ignoring uknown DFS master region: %d\n", dfs_region);
 		return false;
 	}
 }
-- 
cgit v0.10.2


From 23a1f8d44c0bca48f04fc2a2f1edafd826ce6133 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 8 Dec 2015 16:04:31 +0200
Subject: mac80211: process and save VHT MU-MIMO group frame

The Group ID Management frame is an Action frame of
category VHT. It is transmitted by the AP to assign
or change the user position of a STA for one or more
group IDs.
Process and save the group membership data. Notify
underlying driver of changes.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 452c0b0..d9ddb89 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -843,6 +843,8 @@ enum ieee80211_vht_opmode_bits {
 };
 
 #define WLAN_SA_QUERY_TR_ID_LEN 2
+#define WLAN_MEMBERSHIP_LEN 8
+#define WLAN_USER_POSITION_LEN 16
 
 /**
  * struct ieee80211_tpc_report_ie
@@ -991,6 +993,11 @@ struct ieee80211_mgmt {
 				} __packed vht_opmode_notif;
 				struct {
 					u8 action_code;
+					u8 membership[WLAN_MEMBERSHIP_LEN];
+					u8 position[WLAN_USER_POSITION_LEN];
+				} __packed vht_group_notif;
+				struct {
+					u8 action_code;
 					u8 dialog_token;
 					u8 tpc_elem_id;
 					u8 tpc_elem_length;
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 7c30faf..8da483b 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -298,6 +298,7 @@ struct ieee80211_vif_chanctx_switch {
  *	note that this is only called when it changes after the channel
  *	context had been assigned.
  * @BSS_CHANGED_OCB: OCB join status changed
+ * @BSS_CHANGED_MU_GROUPS: VHT MU-MIMO group id or user position changed
  */
 enum ieee80211_bss_change {
 	BSS_CHANGED_ASSOC		= 1<<0,
@@ -323,6 +324,7 @@ enum ieee80211_bss_change {
 	BSS_CHANGED_BEACON_INFO		= 1<<20,
 	BSS_CHANGED_BANDWIDTH		= 1<<21,
 	BSS_CHANGED_OCB                 = 1<<22,
+	BSS_CHANGED_MU_GROUPS		= 1<<23,
 
 	/* when adding here, make sure to change ieee80211_reconfig */
 };
@@ -436,6 +438,19 @@ struct ieee80211_event {
 };
 
 /**
+ * struct ieee80211_mu_group_data - STA's VHT MU-MIMO group data
+ *
+ * This structure describes the group id data of VHT MU-MIMO
+ *
+ * @membership: 64 bits array - a bit is set if station is member of the group
+ * @position: 2 bits per group id indicating the position in the group
+ */
+struct ieee80211_mu_group_data {
+	u8 membership[WLAN_MEMBERSHIP_LEN];
+	u8 position[WLAN_USER_POSITION_LEN];
+};
+
+/**
  * struct ieee80211_bss_conf - holds the BSS's changing parameters
  *
  * This structure keeps information about a BSS (and an association
@@ -477,6 +492,7 @@ struct ieee80211_event {
  * @enable_beacon: whether beaconing should be enabled or not
  * @chandef: Channel definition for this BSS -- the hardware might be
  *	configured a higher bandwidth than this BSS uses, for example.
+ * @mu_group: VHT MU-MIMO group membership data
  * @ht_operation_mode: HT operation mode like in &struct ieee80211_ht_operation.
  *	This field is only valid when the channel is a wide HT/VHT channel.
  *	Note that with TDLS this can be the case (channel is HT, protection must
@@ -535,6 +551,7 @@ struct ieee80211_bss_conf {
 	s32 cqm_rssi_thold;
 	u32 cqm_rssi_hyst;
 	struct cfg80211_chan_def chandef;
+	struct ieee80211_mu_group_data mu_group;
 	__be32 arp_addr_list[IEEE80211_BSS_ARP_ADDR_LIST_LEN];
 	int arp_addr_cnt;
 	bool qos;
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index b84f6aa..747402d 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1714,6 +1714,8 @@ ieee80211_vht_cap_ie_to_sta_vht_cap(struct ieee80211_sub_if_data *sdata,
 enum ieee80211_sta_rx_bandwidth ieee80211_sta_cap_rx_bw(struct sta_info *sta);
 enum ieee80211_sta_rx_bandwidth ieee80211_sta_cur_vht_bw(struct sta_info *sta);
 void ieee80211_sta_set_rx_nss(struct sta_info *sta);
+void ieee80211_process_mu_groups(struct ieee80211_sub_if_data *sdata,
+				 struct ieee80211_mgmt *mgmt);
 u32 __ieee80211_vht_handle_opmode(struct ieee80211_sub_if_data *sdata,
                                   struct sta_info *sta, u8 opmode,
 				  enum ieee80211_band band);
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index c9e325d..33ae3c8 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -1271,6 +1271,16 @@ static void ieee80211_iface_work(struct work_struct *work)
 				}
 			}
 			mutex_unlock(&local->sta_mtx);
+		} else if (ieee80211_is_action(mgmt->frame_control) &&
+			   mgmt->u.action.category == WLAN_CATEGORY_VHT) {
+			switch (mgmt->u.action.u.vht_group_notif.action_code) {
+			case WLAN_VHT_ACTION_GROUPID_MGMT:
+				ieee80211_process_mu_groups(sdata, mgmt);
+				break;
+			default:
+				WARN_ON(1);
+				break;
+			}
 		} else if (ieee80211_is_data_qos(mgmt->frame_control)) {
 			struct ieee80211_hdr *hdr = (void *)mgmt;
 			/*
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 1c342e2..31d5881 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2079,6 +2079,13 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	memset(&ifmgd->ht_capa_mask, 0, sizeof(ifmgd->ht_capa_mask));
 	memset(&ifmgd->vht_capa, 0, sizeof(ifmgd->vht_capa));
 	memset(&ifmgd->vht_capa_mask, 0, sizeof(ifmgd->vht_capa_mask));
+
+	/* reset MU-MIMO ownership and group data */
+	memset(sdata->vif.bss_conf.mu_group.membership, 0,
+	       sizeof(sdata->vif.bss_conf.mu_group.membership));
+	memset(sdata->vif.bss_conf.mu_group.position, 0,
+	       sizeof(sdata->vif.bss_conf.mu_group.position));
+	changed |= BSS_CHANGED_MU_GROUPS;
 	sdata->flags &= ~IEEE80211_SDATA_MU_MIMO_OWNER;
 
 	sdata->ap_power_level = IEEE80211_UNSET_POWER_LEVEL;
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index bc08185..a5668b5 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2738,6 +2738,11 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 						    opmode, status->band);
 			goto handled;
 		}
+		case WLAN_VHT_ACTION_GROUPID_MGMT: {
+			if (len < IEEE80211_MIN_ACTION_SIZE + 25)
+				goto invalid;
+			goto queue;
+		}
 		default:
 			break;
 		}
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 3943d4b..f4b2c04 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -1928,6 +1928,9 @@ int ieee80211_reconfig(struct ieee80211_local *local)
 			  BSS_CHANGED_IDLE |
 			  BSS_CHANGED_TXPOWER;
 
+		if (sdata->flags & IEEE80211_SDATA_MU_MIMO_OWNER)
+			changed |= BSS_CHANGED_MU_GROUPS;
+
 		switch (sdata->vif.type) {
 		case NL80211_IFTYPE_STATION:
 			changed |= BSS_CHANGED_ASSOC |
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index c38b2f0..050de08 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -1,6 +1,9 @@
 /*
  * VHT handling
  *
+ * Portions of this file
+ * Copyright(c) 2015 Intel Deutschland GmbH
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
@@ -425,6 +428,28 @@ u32 __ieee80211_vht_handle_opmode(struct ieee80211_sub_if_data *sdata,
 	return changed;
 }
 
+void ieee80211_process_mu_groups(struct ieee80211_sub_if_data *sdata,
+				 struct ieee80211_mgmt *mgmt)
+{
+	struct ieee80211_bss_conf *bss_conf = &sdata->vif.bss_conf;
+
+	if (!(sdata->flags & IEEE80211_SDATA_MU_MIMO_OWNER))
+		return;
+
+	if (!memcmp(mgmt->u.action.u.vht_group_notif.position,
+		    bss_conf->mu_group.position, WLAN_USER_POSITION_LEN) &&
+	    !memcmp(mgmt->u.action.u.vht_group_notif.membership,
+		    bss_conf->mu_group.membership, WLAN_MEMBERSHIP_LEN))
+		return;
+
+	memcpy(mgmt->u.action.u.vht_group_notif.membership,
+	       bss_conf->mu_group.membership, WLAN_MEMBERSHIP_LEN);
+	memcpy(mgmt->u.action.u.vht_group_notif.position,
+	       bss_conf->mu_group.position, WLAN_USER_POSITION_LEN);
+
+	ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_MU_GROUPS);
+}
+
 void ieee80211_vht_handle_opmode(struct ieee80211_sub_if_data *sdata,
 				 struct sta_info *sta, u8 opmode,
 				 enum ieee80211_band band)
-- 
cgit v0.10.2


From f9cfa5f354b11e56cd8f019c12e14a42585586cd Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 8 Dec 2015 16:04:33 +0200
Subject: mac80211: add flag for duplication check

Add an option for driver to check for packet duplication
by itself.
This is needed for example by the iwlwifi driver which
parallelizes the RX path and does the duplication check
per queue.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 8da483b..ecab934 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1063,7 +1063,7 @@ enum mac80211_rx_flags {
 	RX_FLAG_HT_GF			= BIT(13),
 	RX_FLAG_AMPDU_DETAILS		= BIT(14),
 	RX_FLAG_PN_VALIDATED		= BIT(15),
-	/* bit 16 free */
+	RX_FLAG_DUP_VALIDATED		= BIT(16),
 	RX_FLAG_AMPDU_LAST_KNOWN	= BIT(17),
 	RX_FLAG_AMPDU_IS_LAST		= BIT(18),
 	RX_FLAG_AMPDU_DELIM_CRC_ERROR	= BIT(19),
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index a5668b5..fe675d7 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1099,6 +1099,9 @@ ieee80211_rx_h_check_dup(struct ieee80211_rx_data *rx)
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)rx->skb->data;
 	struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb);
 
+	if (status->flag & RX_FLAG_DUP_VALIDATED)
+		return RX_CONTINUE;
+
 	/*
 	 * Drop duplicate 802.11 retransmissions
 	 * (IEEE 802.11-2012: 9.3.2.10 "Duplicate detection and recovery")
-- 
cgit v0.10.2


From fad471860c097844432c7cf5d3ae6a0a059c2bdc Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 8 Dec 2015 16:04:34 +0200
Subject: mac80211: pass RX aggregation window size to driver

Currently mac80211 does not inform the driver of the window
size when starting an RX aggregation session.
To enable managing the reorder buffer in the driver or hardware
the window size is needed.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index ecab934..a990338 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -3047,9 +3047,11 @@ enum ieee80211_reconfig_type {
  * 	ieee80211_ampdu_mlme_action. Starting sequence number (@ssn)
  * 	is the first frame we expect to perform the action on. Notice
  * 	that TX/RX_STOP can pass NULL for this parameter.
- *	The @buf_size parameter is only valid when the action is set to
- *	%IEEE80211_AMPDU_TX_OPERATIONAL and indicates the peer's reorder
- *	buffer size (number of subframes) for this session -- the driver
+ *	The @buf_size parameter is valid only when the action is set to
+ *	%IEEE80211_AMPDU_RX_START or %IEEE80211_AMPDU_TX_OPERATIONAL and
+ *	indicates the reorder buffer size (number of subframes) for this
+ *	session.
+ *	When the action is set to %IEEE80211_AMPDU_TX_OPERATIONAL the driver
  *	may neither send aggregates containing more subframes than this
  *	nor send aggregates in a way that lost frames would exceed the
  *	buffer size. If just limiting the aggregate size, this would be
diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c
index 10ad4ac..7867273 100644
--- a/net/mac80211/agg-rx.c
+++ b/net/mac80211/agg-rx.c
@@ -323,7 +323,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 		__skb_queue_head_init(&tid_agg_rx->reorder_buf[i]);
 
 	ret = drv_ampdu_action(local, sta->sdata, IEEE80211_AMPDU_RX_START,
-			       &sta->sta, tid, &start_seq_num, 0, false);
+			       &sta->sta, tid, &start_seq_num, buf_size, false);
 	ht_dbg(sta->sdata, "Rx A-MPDU request on %pM tid %d result %d\n",
 	       sta->sta.addr, tid, ret);
 	if (ret) {
-- 
cgit v0.10.2


From 4352a4d7f6bfd0aed0276a13fa4993db35714db4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 8 Dec 2015 16:04:35 +0200
Subject: mac80211: document status.freq restrictions

It's not always necessary to set the status.freq field, for example
when this would be an expensive calculation. It must be set for all
management frames (as they might be reported to userspace), but for
data frames it's not really required. Document this.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index a990338..bdee1cc 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1108,6 +1108,8 @@ enum mac80211_rx_vht_flags {
  *	it but can store it and pass it back to the driver for synchronisation
  * @band: the active band when this frame was received
  * @freq: frequency the radio was tuned to when receiving this frame, in MHz
+ *	This field must be set for management frames, but isn't strictly needed
+ *	for data (other) frames - for those it only affects radiotap reporting.
  * @signal: signal strength when receiving this frame, either in dBm, in dB or
  *	unspecified depending on the hardware capabilities flags
  *	@IEEE80211_HW_SIGNAL_*
-- 
cgit v0.10.2


From a85a7e28f45f7217b9a2efc3ba323de5c0e5b056 Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliangtang@163.com>
Date: Fri, 1 Jan 2016 23:48:52 +0800
Subject: cfg80211/mac80211: use to_delayed_work

Use to_delayed_work() instead of open-coding it.

Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 31d5881..4af9b2b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1638,8 +1638,7 @@ void ieee80211_dynamic_ps_timer(unsigned long data)
 
 void ieee80211_dfs_cac_timer_work(struct work_struct *work)
 {
-	struct delayed_work *delayed_work =
-		container_of(work, struct delayed_work, work);
+	struct delayed_work *delayed_work = to_delayed_work(work);
 	struct ieee80211_sub_if_data *sdata =
 		container_of(delayed_work, struct ieee80211_sub_if_data,
 			     dfs_cac_timer_work);
diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c
index fb44fa3..ff32825 100644
--- a/net/wireless/mlme.c
+++ b/net/wireless/mlme.c
@@ -711,7 +711,7 @@ EXPORT_SYMBOL(cfg80211_rx_mgmt);
 
 void cfg80211_dfs_channels_update_work(struct work_struct *work)
 {
-	struct delayed_work *delayed_work;
+	struct delayed_work *delayed_work = to_delayed_work(work);
 	struct cfg80211_registered_device *rdev;
 	struct cfg80211_chan_def chandef;
 	struct ieee80211_supported_band *sband;
@@ -721,7 +721,6 @@ void cfg80211_dfs_channels_update_work(struct work_struct *work)
 	unsigned long timeout, next_time = 0;
 	int bandid, i;
 
-	delayed_work = container_of(work, struct delayed_work, work);
 	rdev = container_of(delayed_work, struct cfg80211_registered_device,
 			    dfs_update_channels_wk);
 	wiphy = &rdev->wiphy;
-- 
cgit v0.10.2


From 50ea05efaf3bed7dd34bcc2635a8b3f53bd0ccc1 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sarasharon1@gmail.com>
Date: Wed, 30 Dec 2015 16:06:04 +0200
Subject: mac80211: pass block ack session timeout to to driver

Currently mac80211 does not inform the driver of the session
block ack timeout when starting a rx aggregation session.
Drivers that manage the reorder buffer need to know this
parameter.
Seeing that there are now too many arguments for the
drv_ampdu_action() function, wrap them inside a structure.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 6146a29..368de5e 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -6366,12 +6366,13 @@ static u64 ath10k_get_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 
 static int ath10k_ampdu_action(struct ieee80211_hw *hw,
 			       struct ieee80211_vif *vif,
-			       enum ieee80211_ampdu_mlme_action action,
-			       struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			       u8 buf_size, bool amsdu)
+			       struct ieee80211_ampdu_params *params)
 {
 	struct ath10k *ar = hw->priv;
 	struct ath10k_vif *arvif = ath10k_vif_to_arvif(vif);
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
 
 	ath10k_dbg(ar, ATH10K_DBG_MAC, "mac ampdu vdev_id %i sta %pM tid %hu action %d\n",
 		   arvif->vdev_id, sta->addr, tid, action);
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
index fe1fd1a..639294a 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
@@ -1657,13 +1657,14 @@ static void ath9k_htc_reset_tsf(struct ieee80211_hw *hw,
 
 static int ath9k_htc_ampdu_action(struct ieee80211_hw *hw,
 				  struct ieee80211_vif *vif,
-				  enum ieee80211_ampdu_mlme_action action,
-				  struct ieee80211_sta *sta,
-				  u16 tid, u16 *ssn, u8 buf_size, bool amsdu)
+				  struct ieee80211_ampdu_params *params)
 {
 	struct ath9k_htc_priv *priv = hw->priv;
 	struct ath9k_htc_sta *ista;
 	int ret = 0;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
 
 	mutex_lock(&priv->mutex);
 	ath9k_htc_ps_wakeup(priv);
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index c1b33fd..cf58a30 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -1864,14 +1864,16 @@ static void ath9k_reset_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 
 static int ath9k_ampdu_action(struct ieee80211_hw *hw,
 			      struct ieee80211_vif *vif,
-			      enum ieee80211_ampdu_mlme_action action,
-			      struct ieee80211_sta *sta,
-			      u16 tid, u16 *ssn, u8 buf_size, bool amsdu)
+			      struct ieee80211_ampdu_params *params)
 {
 	struct ath_softc *sc = hw->priv;
 	struct ath_common *common = ath9k_hw_common(sc->sc_ah);
 	bool flush = false;
 	int ret = 0;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 
 	mutex_lock(&sc->mutex);
 
diff --git a/drivers/net/wireless/ath/carl9170/main.c b/drivers/net/wireless/ath/carl9170/main.c
index 19d3d64..4d1527a 100644
--- a/drivers/net/wireless/ath/carl9170/main.c
+++ b/drivers/net/wireless/ath/carl9170/main.c
@@ -1413,10 +1413,12 @@ static void carl9170_ampdu_work(struct work_struct *work)
 
 static int carl9170_op_ampdu_action(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
-				    enum ieee80211_ampdu_mlme_action action,
-				    struct ieee80211_sta *sta,
-				    u16 tid, u16 *ssn, u8 buf_size, bool amsdu)
+				    struct ieee80211_ampdu_params *params)
 {
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 	struct ar9170 *ar = hw->priv;
 	struct carl9170_sta_info *sta_info = (void *) sta->drv_priv;
 	struct carl9170_sta_tid *tid_info;
diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c
index 7c169ab..a27279c 100644
--- a/drivers/net/wireless/ath/wcn36xx/main.c
+++ b/drivers/net/wireless/ath/wcn36xx/main.c
@@ -857,12 +857,14 @@ static int wcn36xx_resume(struct ieee80211_hw *hw)
 
 static int wcn36xx_ampdu_action(struct ieee80211_hw *hw,
 		    struct ieee80211_vif *vif,
-		    enum ieee80211_ampdu_mlme_action action,
-		    struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-		    u8 buf_size, bool amsdu)
+		    struct ieee80211_ampdu_params *params)
 {
 	struct wcn36xx *wcn = hw->priv;
 	struct wcn36xx_sta *sta_priv = NULL;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 
 	wcn36xx_dbg(WCN36XX_DBG_MAC, "mac ampdu action action %d tid %d\n",
 		    action, tid);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
index bec2dc1..61ae276 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
@@ -818,13 +818,15 @@ brcms_ops_sta_add(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 static int
 brcms_ops_ampdu_action(struct ieee80211_hw *hw,
 		    struct ieee80211_vif *vif,
-		    enum ieee80211_ampdu_mlme_action action,
-		    struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-		    u8 buf_size, bool amsdu)
+		    struct ieee80211_ampdu_params *params)
 {
 	struct brcms_info *wl = hw->priv;
 	struct scb *scb = &wl->wlc->pri_scb;
 	int status;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u8 buf_size = params->buf_size;
 
 	if (WARN_ON(scb->magic != SCB_MAGIC))
 		return -EIDRM;
diff --git a/drivers/net/wireless/intel/iwlegacy/4965-mac.c b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
index fd38aa0..b75f4ef 100644
--- a/drivers/net/wireless/intel/iwlegacy/4965-mac.c
+++ b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
@@ -5982,12 +5982,14 @@ il4965_mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 int
 il4965_mac_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-			enum ieee80211_ampdu_mlme_action action,
-			struct ieee80211_sta *sta, u16 tid, u16 * ssn,
-			u8 buf_size, bool amsdu)
+			struct ieee80211_ampdu_params *params)
 {
 	struct il_priv *il = hw->priv;
 	int ret = -EINVAL;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 
 	D_HT("A-MPDU action on addr %pM tid %d\n", sta->addr, tid);
 
diff --git a/drivers/net/wireless/intel/iwlegacy/4965.h b/drivers/net/wireless/intel/iwlegacy/4965.h
index 8ab8706..e432715 100644
--- a/drivers/net/wireless/intel/iwlegacy/4965.h
+++ b/drivers/net/wireless/intel/iwlegacy/4965.h
@@ -182,9 +182,7 @@ void il4965_mac_update_tkip_key(struct ieee80211_hw *hw,
 				struct ieee80211_sta *sta, u32 iv32,
 				u16 *phase1key);
 int il4965_mac_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-			    enum ieee80211_ampdu_mlme_action action,
-			    struct ieee80211_sta *sta, u16 tid, u16 * ssn,
-			    u8 buf_size, bool amsdu);
+			    struct ieee80211_ampdu_params *params);
 int il4965_mac_sta_add(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta);
 void
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
index 29ea1c6..151721e 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
@@ -732,12 +732,15 @@ static inline bool iwl_enable_tx_ampdu(const struct iwl_cfg *cfg)
 
 static int iwlagn_mac_ampdu_action(struct ieee80211_hw *hw,
 				   struct ieee80211_vif *vif,
-				   enum ieee80211_ampdu_mlme_action action,
-				   struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-				   u8 buf_size, bool amsdu)
+				   struct ieee80211_ampdu_params *params)
 {
 	struct iwl_priv *priv = IWL_MAC80211_GET_DVM(hw);
 	int ret = -EINVAL;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
+	u8 buf_size = params->buf_size;
 	struct iwl_station_priv *sta_priv = (void *) sta->drv_priv;
 
 	IWL_DEBUG_HT(priv, "A-MPDU action on addr %pM tid %d\n",
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d70a171..1bd3f0b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -837,13 +837,16 @@ iwl_mvm_ampdu_check_trigger(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
-				    enum ieee80211_ampdu_mlme_action action,
-				    struct ieee80211_sta *sta, u16 tid,
-				    u16 *ssn, u8 buf_size, bool amsdu)
+				    struct ieee80211_ampdu_params *params)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int ret;
 	bool tx_agg_ref = false;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
+	u8 buf_size = params->buf_size;
 
 	IWL_DEBUG_HT(mvm, "A-MPDU action on addr %pM tid %d: action %d\n",
 		     sta->addr, tid, action);
diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index c32889a..e31a94f 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -1845,10 +1845,12 @@ static int mac80211_hwsim_testmode_cmd(struct ieee80211_hw *hw,
 
 static int mac80211_hwsim_ampdu_action(struct ieee80211_hw *hw,
 				       struct ieee80211_vif *vif,
-				       enum ieee80211_ampdu_mlme_action action,
-				       struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-				       u8 buf_size, bool amsdu)
+				       struct ieee80211_ampdu_params *params)
 {
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+
 	switch (action) {
 	case IEEE80211_AMPDU_TX_START:
 		ieee80211_start_tx_ba_cb_irqsafe(vif, sta->addr, tid);
diff --git a/drivers/net/wireless/marvell/mwl8k.c b/drivers/net/wireless/marvell/mwl8k.c
index 30e3aaa..088429d 100644
--- a/drivers/net/wireless/marvell/mwl8k.c
+++ b/drivers/net/wireless/marvell/mwl8k.c
@@ -5421,11 +5421,13 @@ static int mwl8k_get_survey(struct ieee80211_hw *hw, int idx,
 
 static int
 mwl8k_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-		   enum ieee80211_ampdu_mlme_action action,
-		   struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-		   u8 buf_size, bool amsdu)
+		   struct ieee80211_ampdu_params *params)
 {
-
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
+	u8 buf_size = params->buf_size;
 	int i, rc = 0;
 	struct mwl8k_priv *priv = hw->priv;
 	struct mwl8k_ampdu_stream *stream;
diff --git a/drivers/net/wireless/mediatek/mt7601u/main.c b/drivers/net/wireless/mediatek/mt7601u/main.c
index f715eee..e70dd95 100644
--- a/drivers/net/wireless/mediatek/mt7601u/main.c
+++ b/drivers/net/wireless/mediatek/mt7601u/main.c
@@ -334,11 +334,13 @@ static int mt7601u_set_rts_threshold(struct ieee80211_hw *hw, u32 value)
 
 static int
 mt76_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-		  enum ieee80211_ampdu_mlme_action action,
-		  struct ieee80211_sta *sta, u16 tid, u16 *ssn, u8 buf_size,
-		  bool amsdu)
+		  struct ieee80211_ampdu_params *params)
 {
 	struct mt7601u_dev *dev = hw->priv;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 	struct mt76_sta *msta = (struct mt76_sta *) sta->drv_priv;
 
 	WARN_ON(msta->wcid.idx > GROUP_WCID(0));
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c
index 9733b31..69c1c09 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c
@@ -7935,10 +7935,11 @@ u64 rt2800_get_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 EXPORT_SYMBOL_GPL(rt2800_get_tsf);
 
 int rt2800_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-			enum ieee80211_ampdu_mlme_action action,
-			struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			u8 buf_size, bool amsdu)
+			struct ieee80211_ampdu_params *params)
 {
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
 	struct rt2x00_sta *sta_priv = (struct rt2x00_sta *)sta->drv_priv;
 	int ret = 0;
 
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.h b/drivers/net/wireless/ralink/rt2x00/rt2800lib.h
index 440790b..83f1a44 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.h
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.h
@@ -218,9 +218,7 @@ int rt2800_conf_tx(struct ieee80211_hw *hw,
 		   const struct ieee80211_tx_queue_params *params);
 u64 rt2800_get_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif);
 int rt2800_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-			enum ieee80211_ampdu_mlme_action action,
-			struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			u8 buf_size, bool amsdu);
+			struct ieee80211_ampdu_params *params);
 int rt2800_get_survey(struct ieee80211_hw *hw, int idx,
 		      struct survey_info *survey);
 void rt2800_disable_wpdma(struct rt2x00_dev *rt2x00dev);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 6aed923..7d820c3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5375,13 +5375,13 @@ static int rtl8xxxu_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 static int
 rtl8xxxu_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
-		      enum ieee80211_ampdu_mlme_action action,
-		      struct ieee80211_sta *sta, u16 tid, u16 *ssn, u8 buf_size,
-		      bool amsdu)
+		      struct ieee80211_ampdu_params *params)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
 	struct device *dev = &priv->udev->dev;
 	u8 ampdu_factor, ampdu_density;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
 
 	switch (action) {
 	case IEEE80211_AMPDU_TX_START:
diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c
index 4ae421e..f250761 100644
--- a/drivers/net/wireless/realtek/rtlwifi/core.c
+++ b/drivers/net/wireless/realtek/rtlwifi/core.c
@@ -1371,11 +1371,13 @@ static void rtl_op_sta_notify(struct ieee80211_hw *hw,
 
 static int rtl_op_ampdu_action(struct ieee80211_hw *hw,
 			       struct ieee80211_vif *vif,
-			       enum ieee80211_ampdu_mlme_action action,
-			       struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			       u8 buf_size, bool amsdu)
+			       struct ieee80211_ampdu_params *params)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 
 	switch (action) {
 	case IEEE80211_AMPDU_TX_START:
diff --git a/drivers/net/wireless/rsi/rsi_91x_mac80211.c b/drivers/net/wireless/rsi/rsi_91x_mac80211.c
index b5bcc93..4df992d 100644
--- a/drivers/net/wireless/rsi/rsi_91x_mac80211.c
+++ b/drivers/net/wireless/rsi/rsi_91x_mac80211.c
@@ -659,29 +659,24 @@ static int rsi_mac80211_set_key(struct ieee80211_hw *hw,
  *				 informs the f/w regarding this.
  * @hw: Pointer to the ieee80211_hw structure.
  * @vif: Pointer to the ieee80211_vif structure.
- * @action: ieee80211_ampdu_mlme_action enum.
- * @sta: Pointer to the ieee80211_sta structure.
- * @tid: Traffic identifier.
- * @ssn: Pointer to ssn value.
- * @buf_size: Buffer size (for kernel version > 2.6.38).
- * @amsdu: is AMSDU in AMPDU allowed
+ * @params: Pointer to A-MPDU action parameters
  *
  * Return: status: 0 on success, negative error code on failure.
  */
 static int rsi_mac80211_ampdu_action(struct ieee80211_hw *hw,
 				     struct ieee80211_vif *vif,
-				     enum ieee80211_ampdu_mlme_action action,
-				     struct ieee80211_sta *sta,
-				     unsigned short tid,
-				     unsigned short *ssn,
-				     unsigned char buf_size,
-				     bool amsdu)
+				     struct ieee80211_ampdu_params *params)
 {
 	int status = -EOPNOTSUPP;
 	struct rsi_hw *adapter = hw->priv;
 	struct rsi_common *common = adapter->priv;
 	u16 seq_no = 0;
 	u8 ii = 0;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
+	u8 buf_size = params->buf_size;
 
 	for (ii = 0; ii < RSI_MAX_VIFS; ii++) {
 		if (vif == adapter->vifs[ii])
diff --git a/drivers/net/wireless/st/cw1200/sta.c b/drivers/net/wireless/st/cw1200/sta.c
index 06321c7..d0ddcde 100644
--- a/drivers/net/wireless/st/cw1200/sta.c
+++ b/drivers/net/wireless/st/cw1200/sta.c
@@ -2129,9 +2129,7 @@ void cw1200_mcast_timeout(unsigned long arg)
 
 int cw1200_ampdu_action(struct ieee80211_hw *hw,
 			struct ieee80211_vif *vif,
-			enum ieee80211_ampdu_mlme_action action,
-			struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			u8 buf_size, bool amsdu)
+			struct ieee80211_ampdu_params *params)
 {
 	/* Aggregation is implemented fully in firmware,
 	 * including block ack negotiation. Do not allow
diff --git a/drivers/net/wireless/st/cw1200/sta.h b/drivers/net/wireless/st/cw1200/sta.h
index bebb337..a0bacaa 100644
--- a/drivers/net/wireless/st/cw1200/sta.h
+++ b/drivers/net/wireless/st/cw1200/sta.h
@@ -109,9 +109,7 @@ void cw1200_bss_info_changed(struct ieee80211_hw *dev,
 			     u32 changed);
 int cw1200_ampdu_action(struct ieee80211_hw *hw,
 			struct ieee80211_vif *vif,
-			enum ieee80211_ampdu_mlme_action action,
-			struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			u8 buf_size, bool amsdu);
+			struct ieee80211_ampdu_params *params);
 
 void cw1200_suspend_resume(struct cw1200_common *priv,
 			  struct wsm_suspend_resume *arg);
diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index d1109c4..45662cf 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -5187,14 +5187,16 @@ out:
 
 static int wl1271_op_ampdu_action(struct ieee80211_hw *hw,
 				  struct ieee80211_vif *vif,
-				  enum ieee80211_ampdu_mlme_action action,
-				  struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-				  u8 buf_size, bool amsdu)
+				  struct ieee80211_ampdu_params *params)
 {
 	struct wl1271 *wl = hw->priv;
 	struct wl12xx_vif *wlvif = wl12xx_vif_to_data(vif);
 	int ret;
 	u8 hlid, *ba_bitmap;
+	struct ieee80211_sta *sta = params->sta;
+	enum ieee80211_ampdu_mlme_action action = params->action;
+	u16 tid = params->tid;
+	u16 *ssn = &params->ssn;
 
 	wl1271_debug(DEBUG_MAC80211, "mac80211 ampdu action %d tid %d", action,
 		     tid);
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index bdee1cc..6c9c559 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -2703,6 +2703,33 @@ enum ieee80211_ampdu_mlme_action {
 };
 
 /**
+ * struct ieee80211_ampdu_params - AMPDU action parameters
+ *
+ * @action: the ampdu action, value from %ieee80211_ampdu_mlme_action.
+ * @sta: peer of this AMPDU session
+ * @tid: tid of the BA session
+ * @ssn: start sequence number of the session. TX/RX_STOP can pass 0. When
+ *	action is set to %IEEE80211_AMPDU_RX_START the driver passes back the
+ *	actual ssn value used to start the session and writes the value here.
+ * @buf_size: reorder buffer size  (number of subframes). Valid only when the
+ *	action is set to %IEEE80211_AMPDU_RX_START or
+ *	%IEEE80211_AMPDU_TX_OPERATIONAL
+ * @amsdu: indicates the peer's ability to receive A-MSDU within A-MPDU.
+ *	valid when the action is set to %IEEE80211_AMPDU_TX_OPERATIONAL
+ * @timeout: BA session timeout. Valid only when the action is set to
+ *	%IEEE80211_AMPDU_RX_START
+ */
+struct ieee80211_ampdu_params {
+	enum ieee80211_ampdu_mlme_action action;
+	struct ieee80211_sta *sta;
+	u16 tid;
+	u16 ssn;
+	u8 buf_size;
+	bool amsdu;
+	u16 timeout;
+};
+
+/**
  * enum ieee80211_frame_release_type - frame release reason
  * @IEEE80211_FRAME_RELEASE_PSPOLL: frame released for PS-Poll
  * @IEEE80211_FRAME_RELEASE_UAPSD: frame(s) released due to
@@ -3046,15 +3073,9 @@ enum ieee80211_reconfig_type {
  * @ampdu_action: Perform a certain A-MPDU action
  * 	The RA/TID combination determines the destination and TID we want
  * 	the ampdu action to be performed for. The action is defined through
- * 	ieee80211_ampdu_mlme_action. Starting sequence number (@ssn)
- * 	is the first frame we expect to perform the action on. Notice
- * 	that TX/RX_STOP can pass NULL for this parameter.
- *	The @buf_size parameter is valid only when the action is set to
- *	%IEEE80211_AMPDU_RX_START or %IEEE80211_AMPDU_TX_OPERATIONAL and
- *	indicates the reorder buffer size (number of subframes) for this
- *	session.
+ *	ieee80211_ampdu_mlme_action.
  *	When the action is set to %IEEE80211_AMPDU_TX_OPERATIONAL the driver
- *	may neither send aggregates containing more subframes than this
+ *	may neither send aggregates containing more subframes than @buf_size
  *	nor send aggregates in a way that lost frames would exceed the
  *	buffer size. If just limiting the aggregate size, this would be
  *	possible with a buf_size of 8:
@@ -3065,9 +3086,6 @@ enum ieee80211_reconfig_type {
  *	buffer size of 8. Correct ways to retransmit #1 would be:
  *	 - TX:       1 or 18 or 81
  *	Even "189" would be wrong since 1 could be lost again.
- *	The @amsdu parameter is valid when the action is set to
- *	%IEEE80211_AMPDU_TX_OPERATIONAL and indicates the peer's ability
- *	to receive A-MSDU within A-MPDU.
  *
  *	Returns a negative error code on failure.
  *	The callback can sleep.
@@ -3409,9 +3427,7 @@ struct ieee80211_ops {
 	int (*tx_last_beacon)(struct ieee80211_hw *hw);
 	int (*ampdu_action)(struct ieee80211_hw *hw,
 			    struct ieee80211_vif *vif,
-			    enum ieee80211_ampdu_mlme_action action,
-			    struct ieee80211_sta *sta, u16 tid, u16 *ssn,
-			    u8 buf_size, bool amsdu);
+			    struct ieee80211_ampdu_params *params);
 	int (*get_survey)(struct ieee80211_hw *hw, int idx,
 		struct survey_info *survey);
 	void (*rfkill_poll)(struct ieee80211_hw *hw);
diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c
index 7867273..ec80db7 100644
--- a/net/mac80211/agg-rx.c
+++ b/net/mac80211/agg-rx.c
@@ -7,6 +7,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2007-2010, Intel Corporation
+ * Copyright(c) 2015 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -61,6 +62,14 @@ void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid,
 {
 	struct ieee80211_local *local = sta->local;
 	struct tid_ampdu_rx *tid_rx;
+	struct ieee80211_ampdu_params params = {
+		.sta = &sta->sta,
+		.action = IEEE80211_AMPDU_RX_STOP,
+		.tid = tid,
+		.amsdu = false,
+		.timeout = 0,
+		.ssn = 0,
+	};
 
 	lockdep_assert_held(&sta->ampdu_mlme.mtx);
 
@@ -78,8 +87,7 @@ void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid,
 	       initiator == WLAN_BACK_RECIPIENT ? "recipient" : "inititator",
 	       (int)reason);
 
-	if (drv_ampdu_action(local, sta->sdata, IEEE80211_AMPDU_RX_STOP,
-			     &sta->sta, tid, NULL, 0, false))
+	if (drv_ampdu_action(local, sta->sdata, &params))
 		sdata_info(sta->sdata,
 			   "HW problem - can not stop rx aggregation for %pM tid %d\n",
 			   sta->sta.addr, tid);
@@ -237,6 +245,15 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 {
 	struct ieee80211_local *local = sta->sdata->local;
 	struct tid_ampdu_rx *tid_agg_rx;
+	struct ieee80211_ampdu_params params = {
+		.sta = &sta->sta,
+		.action = IEEE80211_AMPDU_RX_START,
+		.tid = tid,
+		.amsdu = false,
+		.timeout = timeout,
+		.ssn = start_seq_num,
+	};
+
 	int i, ret = -EOPNOTSUPP;
 	u16 status = WLAN_STATUS_REQUEST_DECLINED;
 
@@ -275,6 +292,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 	/* make sure the size doesn't exceed the maximum supported by the hw */
 	if (buf_size > local->hw.max_rx_aggregation_subframes)
 		buf_size = local->hw.max_rx_aggregation_subframes;
+	params.buf_size = buf_size;
 
 	/* examine state machine */
 	mutex_lock(&sta->ampdu_mlme.mtx);
@@ -322,8 +340,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 	for (i = 0; i < buf_size; i++)
 		__skb_queue_head_init(&tid_agg_rx->reorder_buf[i]);
 
-	ret = drv_ampdu_action(local, sta->sdata, IEEE80211_AMPDU_RX_START,
-			       &sta->sta, tid, &start_seq_num, buf_size, false);
+	ret = drv_ampdu_action(local, sta->sdata, &params);
 	ht_dbg(sta->sdata, "Rx A-MPDU request on %pM tid %d result %d\n",
 	       sta->sta.addr, tid, ret);
 	if (ret) {
diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index ff75718..4932e9f 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -7,6 +7,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2007-2010, Intel Corporation
+ * Copyright(c) 2015 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -295,7 +296,14 @@ int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid,
 {
 	struct ieee80211_local *local = sta->local;
 	struct tid_ampdu_tx *tid_tx;
-	enum ieee80211_ampdu_mlme_action action;
+	struct ieee80211_ampdu_params params = {
+		.sta = &sta->sta,
+		.tid = tid,
+		.buf_size = 0,
+		.amsdu = false,
+		.timeout = 0,
+		.ssn = 0,
+	};
 	int ret;
 
 	lockdep_assert_held(&sta->ampdu_mlme.mtx);
@@ -304,10 +312,10 @@ int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid,
 	case AGG_STOP_DECLINED:
 	case AGG_STOP_LOCAL_REQUEST:
 	case AGG_STOP_PEER_REQUEST:
-		action = IEEE80211_AMPDU_TX_STOP_CONT;
+		params.action = IEEE80211_AMPDU_TX_STOP_CONT;
 		break;
 	case AGG_STOP_DESTROY_STA:
-		action = IEEE80211_AMPDU_TX_STOP_FLUSH;
+		params.action = IEEE80211_AMPDU_TX_STOP_FLUSH;
 		break;
 	default:
 		WARN_ON_ONCE(1);
@@ -330,9 +338,8 @@ int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid,
 		spin_unlock_bh(&sta->lock);
 		if (reason != AGG_STOP_DESTROY_STA)
 			return -EALREADY;
-		ret = drv_ampdu_action(local, sta->sdata,
-				       IEEE80211_AMPDU_TX_STOP_FLUSH_CONT,
-				       &sta->sta, tid, NULL, 0, false);
+		params.action = IEEE80211_AMPDU_TX_STOP_FLUSH_CONT;
+		ret = drv_ampdu_action(local, sta->sdata, &params);
 		WARN_ON_ONCE(ret);
 		return 0;
 	}
@@ -381,8 +388,7 @@ int ___ieee80211_stop_tx_ba_session(struct sta_info *sta, u16 tid,
 					WLAN_BACK_INITIATOR;
 	tid_tx->tx_stop = reason == AGG_STOP_LOCAL_REQUEST;
 
-	ret = drv_ampdu_action(local, sta->sdata, action,
-			       &sta->sta, tid, NULL, 0, false);
+	ret = drv_ampdu_action(local, sta->sdata, &params);
 
 	/* HW shall not deny going back to legacy */
 	if (WARN_ON(ret)) {
@@ -445,7 +451,14 @@ void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 	struct tid_ampdu_tx *tid_tx;
 	struct ieee80211_local *local = sta->local;
 	struct ieee80211_sub_if_data *sdata = sta->sdata;
-	u16 start_seq_num;
+	struct ieee80211_ampdu_params params = {
+		.sta = &sta->sta,
+		.action = IEEE80211_AMPDU_TX_START,
+		.tid = tid,
+		.buf_size = 0,
+		.amsdu = false,
+		.timeout = 0,
+	};
 	int ret;
 
 	tid_tx = rcu_dereference_protected_tid_tx(sta, tid);
@@ -467,10 +480,8 @@ void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 	 */
 	synchronize_net();
 
-	start_seq_num = sta->tid_seq[tid] >> 4;
-
-	ret = drv_ampdu_action(local, sdata, IEEE80211_AMPDU_TX_START,
-			       &sta->sta, tid, &start_seq_num, 0, false);
+	params.ssn = sta->tid_seq[tid] >> 4;
+	ret = drv_ampdu_action(local, sdata, &params);
 	if (ret) {
 		ht_dbg(sdata,
 		       "BA request denied - HW unavailable for %pM tid %d\n",
@@ -499,7 +510,7 @@ void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 
 	/* send AddBA request */
 	ieee80211_send_addba_request(sdata, sta->sta.addr, tid,
-				     tid_tx->dialog_token, start_seq_num,
+				     tid_tx->dialog_token, params.ssn,
 				     IEEE80211_MAX_AMPDU_BUF,
 				     tid_tx->timeout);
 }
@@ -684,18 +695,24 @@ static void ieee80211_agg_tx_operational(struct ieee80211_local *local,
 					 struct sta_info *sta, u16 tid)
 {
 	struct tid_ampdu_tx *tid_tx;
+	struct ieee80211_ampdu_params params = {
+		.sta = &sta->sta,
+		.action = IEEE80211_AMPDU_TX_OPERATIONAL,
+		.tid = tid,
+		.timeout = 0,
+		.ssn = 0,
+	};
 
 	lockdep_assert_held(&sta->ampdu_mlme.mtx);
 
 	tid_tx = rcu_dereference_protected_tid_tx(sta, tid);
+	params.buf_size = tid_tx->buf_size;
+	params.amsdu = tid_tx->amsdu;
 
 	ht_dbg(sta->sdata, "Aggregation is on for %pM tid %d\n",
 	       sta->sta.addr, tid);
 
-	drv_ampdu_action(local, sta->sdata,
-			 IEEE80211_AMPDU_TX_OPERATIONAL,
-			 &sta->sta, tid, NULL, tid_tx->buf_size,
-			 tid_tx->amsdu);
+	drv_ampdu_action(local, sta->sdata, &params);
 
 	/*
 	 * synchronize with TX path, while splicing the TX path
diff --git a/net/mac80211/driver-ops.c b/net/mac80211/driver-ops.c
index ca1fe55..c258f10 100644
--- a/net/mac80211/driver-ops.c
+++ b/net/mac80211/driver-ops.c
@@ -284,9 +284,7 @@ int drv_switch_vif_chanctx(struct ieee80211_local *local,
 
 int drv_ampdu_action(struct ieee80211_local *local,
 		     struct ieee80211_sub_if_data *sdata,
-		     enum ieee80211_ampdu_mlme_action action,
-		     struct ieee80211_sta *sta, u16 tid,
-		     u16 *ssn, u8 buf_size, bool amsdu)
+		     struct ieee80211_ampdu_params *params)
 {
 	int ret = -EOPNOTSUPP;
 
@@ -296,12 +294,10 @@ int drv_ampdu_action(struct ieee80211_local *local,
 	if (!check_sdata_in_driver(sdata))
 		return -EIO;
 
-	trace_drv_ampdu_action(local, sdata, action, sta, tid,
-			       ssn, buf_size, amsdu);
+	trace_drv_ampdu_action(local, sdata, params);
 
 	if (local->ops->ampdu_action)
-		ret = local->ops->ampdu_action(&local->hw, &sdata->vif, action,
-					       sta, tid, ssn, buf_size, amsdu);
+		ret = local->ops->ampdu_action(&local->hw, &sdata->vif, params);
 
 	trace_drv_return_int(local, ret);
 
diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h
index 154ce4b..18b0d65 100644
--- a/net/mac80211/driver-ops.h
+++ b/net/mac80211/driver-ops.h
@@ -585,9 +585,7 @@ static inline int drv_tx_last_beacon(struct ieee80211_local *local)
 
 int drv_ampdu_action(struct ieee80211_local *local,
 		     struct ieee80211_sub_if_data *sdata,
-		     enum ieee80211_ampdu_mlme_action action,
-		     struct ieee80211_sta *sta, u16 tid,
-		     u16 *ssn, u8 buf_size, bool amsdu);
+		     struct ieee80211_ampdu_params *params);
 
 static inline int drv_get_survey(struct ieee80211_local *local, int idx,
 				struct survey_info *survey)
diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h
index a6b4442..2b0a17e 100644
--- a/net/mac80211/trace.h
+++ b/net/mac80211/trace.h
@@ -80,7 +80,23 @@
 #define KEY_PR_FMT	" cipher:0x%x, flags=%#x, keyidx=%d, hw_key_idx=%d"
 #define KEY_PR_ARG	__entry->cipher, __entry->flags, __entry->keyidx, __entry->hw_key_idx
 
-
+#define AMPDU_ACTION_ENTRY	__field(enum ieee80211_ampdu_mlme_action,		\
+					ieee80211_ampdu_mlme_action)			\
+				STA_ENTRY						\
+				__field(u16, tid)					\
+				__field(u16, ssn)					\
+				__field(u8, buf_size)					\
+				__field(bool, amsdu)					\
+				__field(u16, timeout)
+#define AMPDU_ACTION_ASSIGN	STA_NAMED_ASSIGN(params->sta);				\
+				__entry->tid = params->tid;				\
+				__entry->ssn = params->ssn;				\
+				__entry->buf_size = params->buf_size;			\
+				__entry->amsdu = params->amsdu;				\
+				__entry->timeout = params->timeout;
+#define AMPDU_ACTION_PR_FMT	STA_PR_FMT " tid %d, ssn %d, buf_size %u, amsdu %d, timeout %d"
+#define AMPDU_ACTION_PR_ARG	STA_PR_ARG, __entry->tid, __entry->ssn,			\
+				__entry->buf_size, __entry->amsdu, __entry->timeout
 
 /*
  * Tracing for driver callbacks.
@@ -970,38 +986,25 @@ DEFINE_EVENT(local_only_evt, drv_tx_last_beacon,
 TRACE_EVENT(drv_ampdu_action,
 	TP_PROTO(struct ieee80211_local *local,
 		 struct ieee80211_sub_if_data *sdata,
-		 enum ieee80211_ampdu_mlme_action action,
-		 struct ieee80211_sta *sta, u16 tid,
-		 u16 *ssn, u8 buf_size, bool amsdu),
+		 struct ieee80211_ampdu_params *params),
 
-	TP_ARGS(local, sdata, action, sta, tid, ssn, buf_size, amsdu),
+	TP_ARGS(local, sdata, params),
 
 	TP_STRUCT__entry(
 		LOCAL_ENTRY
-		STA_ENTRY
-		__field(u32, action)
-		__field(u16, tid)
-		__field(u16, ssn)
-		__field(u8, buf_size)
-		__field(bool, amsdu)
 		VIF_ENTRY
+		AMPDU_ACTION_ENTRY
 	),
 
 	TP_fast_assign(
 		LOCAL_ASSIGN;
 		VIF_ASSIGN;
-		STA_ASSIGN;
-		__entry->action = action;
-		__entry->tid = tid;
-		__entry->ssn = ssn ? *ssn : 0;
-		__entry->buf_size = buf_size;
-		__entry->amsdu = amsdu;
+		AMPDU_ACTION_ASSIGN;
 	),
 
 	TP_printk(
-		LOCAL_PR_FMT VIF_PR_FMT STA_PR_FMT " action:%d tid:%d buf:%d amsdu:%d",
-		LOCAL_PR_ARG, VIF_PR_ARG, STA_PR_ARG, __entry->action,
-		__entry->tid, __entry->buf_size, __entry->amsdu
+		LOCAL_PR_FMT VIF_PR_FMT AMPDU_ACTION_PR_FMT,
+		LOCAL_PR_ARG, VIF_PR_ARG, AMPDU_ACTION_PR_ARG
 	)
 );
 
-- 
cgit v0.10.2


From a365e668e89d50230737ec20d43fe8fe18412730 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 11:26:44 +0530
Subject: dt: bindings: add bindings for ipq4019 wifi block

Add device tree binding documentation details for wifi block present
in Qualcomm IPQ4019 SoC into "qcom,ath10k.txt".

Right now, ath10k wireless driver has support for PCI based wlan devices.
There is a plan to extend ath10k driver to support wifi devices which are
connected over AHB as well (enumeration will happen via device tree node).
For AHB based devices (ie, ipq4019), all wifi properties are supplied via
device tree (including irq, reg addr, cal data,etc).

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Acked-by: Rob Herring <robh@kernel.org>
[kvalo@qca.qualcomm.com: fixed typos noticed by Rob]
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
index edefc26..96aae6b 100644
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.txt
@@ -1,17 +1,46 @@
 * Qualcomm Atheros ath10k wireless devices
 
-For ath10k devices the calibration data can be provided through Device
-Tree. The node is a child node of the PCI controller.
-
 Required properties:
--compatible : Should be "qcom,ath10k"
+- compatible: Should be one of the following:
+	* "qcom,ath10k"
+	* "qcom,ipq4019-wifi"
+
+PCI based devices uses compatible string "qcom,ath10k" and takes only
+calibration data via "qcom,ath10k-calibration-data". Rest of the properties
+are not applicable for PCI based devices.
+
+AHB based devices (i.e. ipq4019) uses compatible string "qcom,ipq4019-wifi"
+and also uses most of the properties defined in this doc.
 
 Optional properties:
+- reg: Address and length of the register set for the device.
+- resets: Must contain an entry for each entry in reset-names.
+          See ../reset/reseti.txt for details.
+- reset-names: Must include the list of following reset names,
+	       "wifi_cpu_init"
+	       "wifi_radio_srif"
+	       "wifi_radio_warm"
+	       "wifi_radio_cold"
+	       "wifi_core_warm"
+	       "wifi_core_cold"
+- clocks: List of clock specifiers, must contain an entry for each required
+          entry in clock-names.
+- clock-names: Should contain the clock names "wifi_wcss_cmd", "wifi_wcss_ref",
+               "wifi_wcss_rtc".
+- interrupts: List of interrupt lines. Must contain an entry
+	      for each entry in the interrupt-names property.
+- interrupt-names: Must include the entries for MSI interrupt
+		   names ("msi0" to "msi15") and legacy interrupt
+		   name ("legacy"),
+- qcom,msi_addr: MSI interrupt address.
+- qcom,msi_base: Base value to add before writing MSI data into
+		MSI address register.
 - qcom,ath10k-calibration-data : calibration data as an array, the
 				 length can vary between hw versions
 
+Example (to supply the calibration data alone):
 
-Example:
+In this example, the node is defined as child node of the PCI controller.
 
 pci {
 	pcie@0 {
@@ -28,3 +57,53 @@ pci {
 		};
 	};
 };
+
+Example (to supply ipq4019 SoC wifi block details):
+
+wifi0: wifi@a000000 {
+	compatible = "qcom,ipq4019-wifi";
+	reg = <0xa000000 0x200000>;
+	resets = <&gcc WIFI0_CPU_INIT_RESET>,
+		 <&gcc WIFI0_RADIO_SRIF_RESET>,
+		 <&gcc WIFI0_RADIO_WARM_RESET>,
+		 <&gcc WIFI0_RADIO_COLD_RESET>,
+		 <&gcc WIFI0_CORE_WARM_RESET>,
+		 <&gcc WIFI0_CORE_COLD_RESET>;
+	reset-names = "wifi_cpu_init",
+		      "wifi_radio_srif",
+		      "wifi_radio_warm",
+		      "wifi_radio_cold",
+		      "wifi_core_warm",
+		      "wifi_core_cold";
+	clocks = <&gcc GCC_WCSS2G_CLK>,
+		 <&gcc GCC_WCSS2G_REF_CLK>,
+		 <&gcc GCC_WCSS2G_RTC_CLK>;
+	clock-names = "wifi_wcss_cmd",
+		      "wifi_wcss_ref",
+		      "wifi_wcss_rtc";
+	interrupts = <0 0x20 0x1>,
+		     <0 0x21 0x1>,
+		     <0 0x22 0x1>,
+		     <0 0x23 0x1>,
+		     <0 0x24 0x1>,
+		     <0 0x25 0x1>,
+		     <0 0x26 0x1>,
+		     <0 0x27 0x1>,
+		     <0 0x28 0x1>,
+		     <0 0x29 0x1>,
+		     <0 0x2a 0x1>,
+		     <0 0x2b 0x1>,
+		     <0 0x2c 0x1>,
+		     <0 0x2d 0x1>,
+		     <0 0x2e 0x1>,
+		     <0 0x2f 0x1>,
+		     <0 0xa8 0x0>;
+	interrupt-names = "msi0",  "msi1",  "msi2",  "msi3",
+			  "msi4",  "msi5",  "msi6",  "msi7",
+			  "msi8",  "msi9",  "msi10", "msi11",
+			  "msi12", "msi13", "msi14", "msi15",
+			  "legacy";
+	qcom,msi_addr = <0x0b006040>;
+	qcom,msi_base = <0x40>;
+	qcom,ath10k-calibration-data = [ 01 02 03 ... ];
+};
-- 
cgit v0.10.2


From d742c969b3ad9c83ac57579dfd56c90dd437db37 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 13 Jan 2016 14:52:52 +0100
Subject: ath10k: prevent txrx running for too long

On multicore systems it was possible for the txrx
tasklet to keep on running for long periods of
time on a single CPU due to tx completion
processing. Another CPU could feed the running
tasklet for an indefinite amount of time.

The tasklet is now guaranteed to run a finite
amount of time and is limited by HTT CE Rx ring
depth.

This improves behavior when RPS is used on target
system and might improve TCP handling as well.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 91afa3a..3079434 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2156,10 +2156,18 @@ static void ath10k_htt_txrx_compl_task(unsigned long ptr)
 {
 	struct ath10k_htt *htt = (struct ath10k_htt *)ptr;
 	struct ath10k *ar = htt->ar;
+	struct sk_buff_head tx_q;
 	struct htt_resp *resp;
 	struct sk_buff *skb;
+	unsigned long flags;
 
-	while ((skb = skb_dequeue(&htt->tx_compl_q))) {
+	__skb_queue_head_init(&tx_q);
+
+	spin_lock_irqsave(&htt->tx_compl_q.lock, flags);
+	skb_queue_splice_init(&htt->tx_compl_q, &tx_q);
+	spin_unlock_irqrestore(&htt->tx_compl_q.lock, flags);
+
+	while ((skb = __skb_dequeue(&tx_q))) {
 		ath10k_htt_rx_frm_tx_compl(htt->ar, skb);
 		dev_kfree_skb_any(skb);
 	}
-- 
cgit v0.10.2


From de46c015c9b703bb8abb91856ed45e3639c7e7d2 Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:28 +0530
Subject: ath10k: add support for parsing per STA rx_duration for 10.2.4

Add support for parsing 'peer_rx_duration' as part
of 10.2.4 peer_stats. Also register PEER_STATS service
for parsing 'rx_duration' (and for any new fields to be added
as part of peer_stats). Have checks for backward compatibility with
older 10.2.4 firmware incase PEER_STATS service is not enabled

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 7840cf3..d47a085 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -159,6 +159,7 @@ struct ath10k_fw_stats_peer {
 	u32 peer_rssi;
 	u32 peer_tx_rate;
 	u32 peer_rx_rate; /* 10x only */
+	u32 rx_duration;
 };
 
 struct ath10k_fw_stats_vdev {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index a7c3d29..c610d17 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -2862,11 +2862,20 @@ static int ath10k_wmi_10_2_4_op_pull_fw_stats(struct ath10k *ar,
 	/* fw doesn't implement vdev stats */
 
 	for (i = 0; i < num_peer_stats; i++) {
-		const struct wmi_10_2_4_peer_stats *src;
+		const struct wmi_10_2_4_ext_peer_stats *src;
 		struct ath10k_fw_stats_peer *dst;
+		int stats_len;
+		bool ext_peer_stats_support;
+
+		ext_peer_stats_support = test_bit(WMI_SERVICE_PEER_STATS,
+						  ar->wmi.svc_map);
+		if (ext_peer_stats_support)
+			stats_len = sizeof(struct wmi_10_2_4_ext_peer_stats);
+		else
+			stats_len = sizeof(struct wmi_10_2_4_peer_stats);
 
 		src = (void *)skb->data;
-		if (!skb_pull(skb, sizeof(*src)))
+		if (!skb_pull(skb, stats_len))
 			return -EPROTO;
 
 		dst = kzalloc(sizeof(*dst), GFP_ATOMIC);
@@ -2876,6 +2885,9 @@ static int ath10k_wmi_10_2_4_op_pull_fw_stats(struct ath10k *ar,
 		ath10k_wmi_pull_peer_stats(&src->common.old, dst);
 
 		dst->peer_rx_rate = __le32_to_cpu(src->common.peer_rx_rate);
+
+		if (ext_peer_stats_support)
+			dst->rx_duration = __le32_to_cpu(src->rx_duration);
 		/* FIXME: expose 10.2 specific values */
 
 		list_add_tail(&dst->list, &stats->peers);
@@ -5517,6 +5529,9 @@ static struct sk_buff *ath10k_wmi_10_2_op_gen_init(struct ath10k *ar)
 	    test_bit(WMI_SERVICE_COEX_GPIO, ar->wmi.svc_map))
 		features |= WMI_10_2_COEX_GPIO;
 
+	if (test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map))
+		features |= WMI_10_2_PEER_STATS;
+
 	cmd->resource_config.feature_mask = __cpu_to_le32(features);
 
 	memcpy(&cmd->resource_config.common, &config, sizeof(config));
@@ -7126,6 +7141,9 @@ ath10k_wmi_fw_peer_stats_fill(const struct ath10k_fw_stats_peer *peer,
 			"Peer TX rate", peer->peer_tx_rate);
 	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
 			"Peer RX rate", peer->peer_rx_rate);
+	len += scnprintf(buf + len, buf_len - len, "%30s %u\n",
+			"Peer RX duration", peer->rx_duration);
+
 	len += scnprintf(buf + len, buf_len - len, "\n");
 	*length = len;
 }
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index d85ad78..e1bb02b 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -177,6 +177,7 @@ enum wmi_service {
 	WMI_SERVICE_BSS_CHANNEL_INFO_64,
 	WMI_SERVICE_EXT_RES_CFG_SUPPORT,
 	WMI_SERVICE_MESH,
+	WMI_SERVICE_PEER_STATS,
 
 	/* keep last */
 	WMI_SERVICE_MAX,
@@ -213,6 +214,7 @@ enum wmi_10x_service {
 	WMI_10X_SERVICE_BSS_CHANNEL_INFO_64,
 	WMI_10X_SERVICE_MESH,
 	WMI_10X_SERVICE_EXT_RES_CFG_SUPPORT,
+	WMI_10X_SERVICE_PEER_STATS,
 };
 
 enum wmi_main_service {
@@ -386,6 +388,7 @@ static inline char *wmi_service_name(int service_id)
 	SVCSTR(WMI_SERVICE_BSS_CHANNEL_INFO_64);
 	SVCSTR(WMI_SERVICE_EXT_RES_CFG_SUPPORT);
 	SVCSTR(WMI_SERVICE_MESH);
+	SVCSTR(WMI_SERVICE_PEER_STATS);
 	default:
 		return NULL;
 	}
@@ -463,6 +466,8 @@ static inline void wmi_10x_svc_map(const __le32 *in, unsigned long *out,
 	       WMI_SERVICE_MESH, len);
 	SVCMAP(WMI_10X_SERVICE_EXT_RES_CFG_SUPPORT,
 	       WMI_SERVICE_EXT_RES_CFG_SUPPORT, len);
+	SVCMAP(WMI_10X_SERVICE_PEER_STATS,
+	       WMI_SERVICE_PEER_STATS, len);
 }
 
 static inline void wmi_main_svc_map(const __le32 *in, unsigned long *out,
@@ -2417,6 +2422,7 @@ enum wmi_10_2_feature_mask {
 	WMI_10_2_RX_BATCH_MODE = BIT(0),
 	WMI_10_2_ATF_CONFIG    = BIT(1),
 	WMI_10_2_COEX_GPIO     = BIT(3),
+	WMI_10_2_PEER_STATS    = BIT(7),
 };
 
 struct wmi_resource_config_10_2 {
@@ -4230,6 +4236,12 @@ struct wmi_10_2_4_peer_stats {
 	__le32 unknown_value; /* FIXME: what is this word? */
 } __packed;
 
+struct wmi_10_2_4_ext_peer_stats {
+	struct wmi_10_2_peer_stats common;
+	__le32 peer_rssi_changed;
+	__le32 rx_duration;
+} __packed;
+
 struct wmi_10_4_peer_stats {
 	struct wmi_mac_addr peer_macaddr;
 	__le32 peer_rssi;
-- 
cgit v0.10.2


From 774e656e94cdd8db27a8831abb3d35fbb95d2bc6 Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:29 +0530
Subject: ath10k: fix naming Peer stats rssi_changed field in 10.2.4

Fix naming of peer stats rssi_changed field in 10.2.4 to make it
more readable. As of now this field is not used, but necessary to
pull in fw_stats with appropriate length.

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index e1bb02b..187f495 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -4233,7 +4233,7 @@ struct wmi_10_2_peer_stats {
 
 struct wmi_10_2_4_peer_stats {
 	struct wmi_10_2_peer_stats common;
-	__le32 unknown_value; /* FIXME: what is this word? */
+	__le32 peer_rssi_changed;
 } __packed;
 
 struct wmi_10_2_4_ext_peer_stats {
-- 
cgit v0.10.2


From 1fe374f5b07ed58e485fdbfa7410876e25ce1a78 Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:30 +0530
Subject: ath10k: cleanup setting pdev paramaters

Replace the local variable 'burst_enable' with 'param'
for mapping and setting pdev paraemeters and with this patch
pretty easy to extend support for new parameters adhering to
linux kernel coding guidelines

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 6146a29..0d7671c 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -3988,7 +3988,7 @@ static int ath10k_set_antenna(struct ieee80211_hw *hw, u32 tx_ant, u32 rx_ant)
 static int ath10k_start(struct ieee80211_hw *hw)
 {
 	struct ath10k *ar = hw->priv;
-	u32 burst_enable;
+	u32 param;
 	int ret = 0;
 
 	/*
@@ -4031,13 +4031,15 @@ static int ath10k_start(struct ieee80211_hw *hw)
 		goto err_power_down;
 	}
 
-	ret = ath10k_wmi_pdev_set_param(ar, ar->wmi.pdev_param->pmf_qos, 1);
+	param = ar->wmi.pdev_param->pmf_qos;
+	ret = ath10k_wmi_pdev_set_param(ar, param, 1);
 	if (ret) {
 		ath10k_warn(ar, "failed to enable PMF QOS: %d\n", ret);
 		goto err_core_stop;
 	}
 
-	ret = ath10k_wmi_pdev_set_param(ar, ar->wmi.pdev_param->dynamic_bw, 1);
+	param = ar->wmi.pdev_param->dynamic_bw;
+	ret = ath10k_wmi_pdev_set_param(ar, param, 1);
 	if (ret) {
 		ath10k_warn(ar, "failed to enable dynamic BW: %d\n", ret);
 		goto err_core_stop;
@@ -4053,8 +4055,8 @@ static int ath10k_start(struct ieee80211_hw *hw)
 	}
 
 	if (test_bit(WMI_SERVICE_BURST, ar->wmi.svc_map)) {
-		burst_enable = ar->wmi.pdev_param->burst_enable;
-		ret = ath10k_wmi_pdev_set_param(ar, burst_enable, 0);
+		param = ar->wmi.pdev_param->burst_enable;
+		ret = ath10k_wmi_pdev_set_param(ar, param, 0);
 		if (ret) {
 			ath10k_warn(ar, "failed to disable burst: %d\n", ret);
 			goto err_core_stop;
@@ -4072,8 +4074,8 @@ static int ath10k_start(struct ieee80211_hw *hw)
 	 * this problem.
 	 */
 
-	ret = ath10k_wmi_pdev_set_param(ar,
-					ar->wmi.pdev_param->arp_ac_override, 0);
+	param = ar->wmi.pdev_param->arp_ac_override;
+	ret = ath10k_wmi_pdev_set_param(ar, param, 0);
 	if (ret) {
 		ath10k_warn(ar, "failed to set arp ac override parameter: %d\n",
 			    ret);
@@ -4092,8 +4094,8 @@ static int ath10k_start(struct ieee80211_hw *hw)
 		}
 	}
 
-	ret = ath10k_wmi_pdev_set_param(ar,
-					ar->wmi.pdev_param->ani_enable, 1);
+	param = ar->wmi.pdev_param->ani_enable;
+	ret = ath10k_wmi_pdev_set_param(ar, param, 1);
 	if (ret) {
 		ath10k_warn(ar, "failed to enable ani by default: %d\n",
 			    ret);
-- 
cgit v0.10.2


From b4619ea2f72153df42d8ba2894b2af5e621d6df6 Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:31 +0530
Subject: ath10k: rename few function names of firmware stats

Prerequisite patch to address checkpatch errors for the next patch
in this series, this function names are bit too long, make it short

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 2bdf540..4fb1e0e 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -276,7 +276,7 @@ static const struct file_operations fops_wmi_services = {
 	.llseek = default_llseek,
 };
 
-static void ath10k_debug_fw_stats_pdevs_free(struct list_head *head)
+static void ath10k_fw_stats_pdevs_free(struct list_head *head)
 {
 	struct ath10k_fw_stats_pdev *i, *tmp;
 
@@ -286,7 +286,7 @@ static void ath10k_debug_fw_stats_pdevs_free(struct list_head *head)
 	}
 }
 
-static void ath10k_debug_fw_stats_vdevs_free(struct list_head *head)
+static void ath10k_fw_stats_vdevs_free(struct list_head *head)
 {
 	struct ath10k_fw_stats_vdev *i, *tmp;
 
@@ -296,7 +296,7 @@ static void ath10k_debug_fw_stats_vdevs_free(struct list_head *head)
 	}
 }
 
-static void ath10k_debug_fw_stats_peers_free(struct list_head *head)
+static void ath10k_fw_stats_peers_free(struct list_head *head)
 {
 	struct ath10k_fw_stats_peer *i, *tmp;
 
@@ -310,9 +310,9 @@ static void ath10k_debug_fw_stats_reset(struct ath10k *ar)
 {
 	spin_lock_bh(&ar->data_lock);
 	ar->debug.fw_stats_done = false;
-	ath10k_debug_fw_stats_pdevs_free(&ar->debug.fw_stats.pdevs);
-	ath10k_debug_fw_stats_vdevs_free(&ar->debug.fw_stats.vdevs);
-	ath10k_debug_fw_stats_peers_free(&ar->debug.fw_stats.peers);
+	ath10k_fw_stats_pdevs_free(&ar->debug.fw_stats.pdevs);
+	ath10k_fw_stats_vdevs_free(&ar->debug.fw_stats.vdevs);
+	ath10k_fw_stats_peers_free(&ar->debug.fw_stats.peers);
 	spin_unlock_bh(&ar->data_lock);
 }
 
@@ -391,9 +391,9 @@ free:
 	/* In some cases lists have been spliced and cleared. Free up
 	 * resources if that is not the case.
 	 */
-	ath10k_debug_fw_stats_pdevs_free(&stats.pdevs);
-	ath10k_debug_fw_stats_vdevs_free(&stats.vdevs);
-	ath10k_debug_fw_stats_peers_free(&stats.peers);
+	ath10k_fw_stats_pdevs_free(&stats.pdevs);
+	ath10k_fw_stats_vdevs_free(&stats.vdevs);
+	ath10k_fw_stats_peers_free(&stats.peers);
 
 	spin_unlock_bh(&ar->data_lock);
 }
-- 
cgit v0.10.2


From d57e7f2e7e042a7cef19702abe42f1e9b55d754c Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:32 +0530
Subject: ath10k: provision to support periodic peer stats update

Enable support for periodic peer stats update when peer stats
service is enabled. The host to update the peer stats received
from the firmware periodically, since firmware will reset this to zero
after sometime (due to memory constraints)

While enabling periodic peer / vdev stats cleanup the existing
list in debugfs if max limit is reached, so that the new stats is
updated.

Ideally speaking based on 'Michal Kazior's' suggestion
we need to completely move to periodic update of all the stats making
the 'ping - pong' method obselete. This requires a bit of re-work and
some testing as well, also confirmation regarding backward comptability
for various firmware and chipsets. Hence allow periodic update only
for peer_stats.

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 4fb1e0e..3b8da81 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -347,7 +347,8 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 	 *     delivered which is treated as end-of-data and is itself discarded
 	 */
 
-	if (ar->debug.fw_stats_done) {
+	if (ar->debug.fw_stats_done &&
+	    !test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map)) {
 		ath10k_warn(ar, "received unsolicited stats update event\n");
 		goto free;
 	}
@@ -372,11 +373,13 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 			/* Although this is unlikely impose a sane limit to
 			 * prevent firmware from DoS-ing the host.
 			 */
+			ath10k_fw_stats_peers_free(&ar->debug.fw_stats.peers);
 			ath10k_warn(ar, "dropping fw peer stats\n");
 			goto free;
 		}
 
 		if (num_vdevs >= BITS_PER_LONG) {
+			ath10k_fw_stats_vdevs_free(&ar->debug.fw_stats.vdevs);
 			ath10k_warn(ar, "dropping fw vdev stats\n");
 			goto free;
 		}
-- 
cgit v0.10.2


From 8351c052194b30f852db6225a568e48ee2a4947b Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:33 +0530
Subject: ath10k: enable periodic peer stats update

Register for 500ms as periodic peer stats update period,
and parameters like rx_duration that needs to be tracked
in host can be achieved by this provision, also periodic
stats update is the future of fw_stats and shall be extended
for pdev / vdev stats irrespecitive PEER_STATS service is enabled
or not

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
[kvalo@qca.qualcomm.com: fix a checkpatch warning]
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 0d7671c..ef0438d 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -4104,6 +4104,18 @@ static int ath10k_start(struct ieee80211_hw *hw)
 
 	ar->ani_enabled = true;
 
+	if (test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map)) {
+		param = ar->wmi.pdev_param->peer_stats_update_period;
+		ret = ath10k_wmi_pdev_set_param(ar, param,
+						PEER_DEFAULT_STATS_UPDATE_PERIOD);
+		if (ret) {
+			ath10k_warn(ar,
+				    "failed to set peer stats period : %d\n",
+				    ret);
+			goto err_core_stop;
+		}
+	}
+
 	ar->num_started_vdevs = 0;
 	ath10k_regd_update(ar);
 
-- 
cgit v0.10.2


From 856e7c3084399fb7e029628dbe43dcb9bb7d7b5b Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 13 Jan 2016 21:16:34 +0530
Subject: ath10k: add debugfs support for Per STA total rx duration

Add debugfs support for per client total rx duration, track this
via the report of Peer stats rx duration reported for every 500ms

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index d47a085..921b86a 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -316,6 +316,7 @@ struct ath10k_sta {
 #ifdef CONFIG_MAC80211_DEBUGFS
 	/* protected by conf_mutex */
 	bool aggr_mode;
+	u64 rx_duration;
 #endif
 };
 
diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 3b8da81..8d4148a 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -319,7 +319,7 @@ static void ath10k_debug_fw_stats_reset(struct ath10k *ar)
 void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 {
 	struct ath10k_fw_stats stats = {};
-	bool is_start, is_started, is_end;
+	bool is_start, is_started, is_end, peer_stats_svc;
 	size_t num_peers;
 	size_t num_vdevs;
 	int ret;
@@ -347,8 +347,8 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 	 *     delivered which is treated as end-of-data and is itself discarded
 	 */
 
-	if (ar->debug.fw_stats_done &&
-	    !test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map)) {
+	peer_stats_svc = test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map);
+	if (ar->debug.fw_stats_done && !peer_stats_svc) {
 		ath10k_warn(ar, "received unsolicited stats update event\n");
 		goto free;
 	}
@@ -384,6 +384,9 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 			goto free;
 		}
 
+		if (peer_stats_svc)
+			ath10k_sta_update_rx_duration(ar, &stats.peers);
+
 		list_splice_tail_init(&stats.peers, &ar->debug.fw_stats.peers);
 		list_splice_tail_init(&stats.vdevs, &ar->debug.fw_stats.vdevs);
 	}
diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h
index 814719c..f273478 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -153,6 +153,12 @@ ath10k_debug_get_new_fw_crash_data(struct ath10k *ar)
 #ifdef CONFIG_MAC80211_DEBUGFS
 void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			    struct ieee80211_sta *sta, struct dentry *dir);
+void ath10k_sta_update_rx_duration(struct ath10k *ar, struct list_head *peer);
+#else
+static inline void ath10k_sta_update_rx_duration(struct ath10k *ar,
+						 struct list_head *peer)
+{
+}
 #endif /* CONFIG_MAC80211_DEBUGFS */
 
 #ifdef CONFIG_ATH10K_DEBUG
diff --git a/drivers/net/wireless/ath/ath10k/debugfs_sta.c b/drivers/net/wireless/ath/ath10k/debugfs_sta.c
index 95b5c49..67ef75b 100644
--- a/drivers/net/wireless/ath/ath10k/debugfs_sta.c
+++ b/drivers/net/wireless/ath/ath10k/debugfs_sta.c
@@ -18,6 +18,23 @@
 #include "wmi-ops.h"
 #include "debug.h"
 
+void ath10k_sta_update_rx_duration(struct ath10k *ar, struct list_head *head)
+{	struct ieee80211_sta *sta;
+	struct ath10k_fw_stats_peer *peer;
+	struct ath10k_sta *arsta;
+
+	rcu_read_lock();
+	list_for_each_entry(peer, head, list) {
+		sta = ieee80211_find_sta_by_ifaddr(ar->hw, peer->peer_macaddr,
+						   NULL);
+		if (!sta)
+			continue;
+		arsta = (struct ath10k_sta *)sta->drv_priv;
+		arsta->rx_duration += (u64)peer->rx_duration;
+	}
+	rcu_read_unlock();
+}
+
 static ssize_t ath10k_dbg_sta_read_aggr_mode(struct file *file,
 					     char __user *user_buf,
 					     size_t count, loff_t *ppos)
@@ -232,6 +249,28 @@ static const struct file_operations fops_delba = {
 	.llseek = default_llseek,
 };
 
+static ssize_t ath10k_dbg_sta_read_rx_duration(struct file *file,
+					       char __user *user_buf,
+					       size_t count, loff_t *ppos)
+{
+	struct ieee80211_sta *sta = file->private_data;
+	struct ath10k_sta *arsta = (struct ath10k_sta *)sta->drv_priv;
+	char buf[100];
+	int len = 0;
+
+	len = scnprintf(buf, sizeof(buf),
+			"%llu usecs\n", arsta->rx_duration);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_rx_duration = {
+	.read = ath10k_dbg_sta_read_rx_duration,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
 void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			    struct ieee80211_sta *sta, struct dentry *dir)
 {
@@ -240,4 +279,6 @@ void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	debugfs_create_file("addba", S_IWUSR, dir, sta, &fops_addba);
 	debugfs_create_file("addba_resp", S_IWUSR, dir, sta, &fops_addba_resp);
 	debugfs_create_file("delba", S_IWUSR, dir, sta, &fops_delba);
+	debugfs_create_file("rx_duration", S_IRUGO, dir, sta,
+			    &fops_rx_duration);
 }
-- 
cgit v0.10.2


From 8b019fb0d0e75e2f682f17742b99c0ca970ca024 Mon Sep 17 00:00:00 2001
From: Yanbo Li <yanbol@qca.qualcomm.com>
Date: Thu, 14 Jan 2016 15:39:19 -0800
Subject: ath10k: remove the p2p notice of absence info from 10.2.4 FW beacon
 info

The p2p NOA never been supported at 10.2.4 FW, remove it to avoid SWBA
event parse error for multi beacon interval case.

Signed-off-by: Yanbo Li <yanbol@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index c610d17..0f01a8d 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -3256,6 +3256,50 @@ static int ath10k_wmi_op_pull_swba_ev(struct ath10k *ar, struct sk_buff *skb,
 	return 0;
 }
 
+static int ath10k_wmi_10_2_4_op_pull_swba_ev(struct ath10k *ar,
+					     struct sk_buff *skb,
+					     struct wmi_swba_ev_arg *arg)
+{
+	struct wmi_10_2_4_host_swba_event *ev = (void *)skb->data;
+	u32 map;
+	size_t i;
+
+	if (skb->len < sizeof(*ev))
+		return -EPROTO;
+
+	skb_pull(skb, sizeof(*ev));
+	arg->vdev_map = ev->vdev_map;
+
+	for (i = 0, map = __le32_to_cpu(ev->vdev_map); map; map >>= 1) {
+		if (!(map & BIT(0)))
+			continue;
+
+		/* If this happens there were some changes in firmware and
+		 * ath10k should update the max size of tim_info array.
+		 */
+		if (WARN_ON_ONCE(i == ARRAY_SIZE(arg->tim_info)))
+			break;
+
+		if (__le32_to_cpu(ev->bcn_info[i].tim_info.tim_len) >
+		     sizeof(ev->bcn_info[i].tim_info.tim_bitmap)) {
+			ath10k_warn(ar, "refusing to parse invalid swba structure\n");
+			return -EPROTO;
+		}
+
+		arg->tim_info[i].tim_len = ev->bcn_info[i].tim_info.tim_len;
+		arg->tim_info[i].tim_mcast = ev->bcn_info[i].tim_info.tim_mcast;
+		arg->tim_info[i].tim_bitmap =
+				ev->bcn_info[i].tim_info.tim_bitmap;
+		arg->tim_info[i].tim_changed =
+				ev->bcn_info[i].tim_info.tim_changed;
+		arg->tim_info[i].tim_num_ps_pending =
+				ev->bcn_info[i].tim_info.tim_num_ps_pending;
+		i++;
+	}
+
+	return 0;
+}
+
 static int ath10k_wmi_10_4_op_pull_swba_ev(struct ath10k *ar,
 					   struct sk_buff *skb,
 					   struct wmi_swba_ev_arg *arg)
@@ -7584,7 +7628,7 @@ static const struct wmi_ops wmi_10_2_4_ops = {
 	.pull_ch_info = ath10k_wmi_op_pull_ch_info_ev,
 	.pull_vdev_start = ath10k_wmi_op_pull_vdev_start_ev,
 	.pull_peer_kick = ath10k_wmi_op_pull_peer_kick_ev,
-	.pull_swba = ath10k_wmi_op_pull_swba_ev,
+	.pull_swba = ath10k_wmi_10_2_4_op_pull_swba_ev,
 	.pull_phyerr_hdr = ath10k_wmi_op_pull_phyerr_ev_hdr,
 	.pull_phyerr = ath10k_wmi_op_pull_phyerr_ev,
 	.pull_rdy = ath10k_wmi_op_pull_rdy_ev,
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 187f495..806d4bb 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -5454,6 +5454,16 @@ struct wmi_host_swba_event {
 	struct wmi_bcn_info bcn_info[0];
 } __packed;
 
+struct wmi_10_2_4_bcn_info {
+	struct wmi_tim_info tim_info;
+	/* The 10.2.4 FW doesn't have p2p NOA info */
+} __packed;
+
+struct wmi_10_2_4_host_swba_event {
+	__le32 vdev_map;
+	struct wmi_10_2_4_bcn_info bcn_info[0];
+} __packed;
+
 /* 16 words = 512 client + 1 word = for guard */
 #define WMI_10_4_TIM_BITMAP_ARRAY_SIZE 17
 
-- 
cgit v0.10.2


From b9a9693fd9aea43f50b107dfc8cbaea317f95a79 Mon Sep 17 00:00:00 2001
From: Kalle Valo <Janusz.Dziedzic@tieto.com>
Date: Fri, 27 Nov 2015 09:37:14 +0100
Subject: ath9k: request NOA update when chanctx active

Request NOA update when chanctx active, also in case
of STA.

Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/channel.c b/drivers/net/wireless/ath/ath9k/channel.c
index 50e614b..0bcb82a 100644
--- a/drivers/net/wireless/ath/ath9k/channel.c
+++ b/drivers/net/wireless/ath/ath9k/channel.c
@@ -1454,7 +1454,7 @@ static void ath9k_update_p2p_ps(struct ath_softc *sc, struct ieee80211_vif *vif)
 	if (!sc->p2p_ps_timer)
 		return;
 
-	if (vif->type != NL80211_IFTYPE_STATION || !vif->p2p)
+	if (vif->type != NL80211_IFTYPE_STATION)
 		return;
 
 	sc->p2p_ps_vif = avp;
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index c1b33fd..4ae63fd 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -978,7 +978,7 @@ static void ath9k_update_bssid_mask(struct ath_softc *sc,
 		if (ctx->nvifs_assigned != 1)
 			continue;
 
-		if (!avp->vif->p2p || !iter_data->has_hw_macaddr)
+		if (!iter_data->has_hw_macaddr)
 			continue;
 
 		ether_addr_copy(common->curbssid, avp->bssid);
@@ -1255,6 +1255,9 @@ static int ath9k_add_interface(struct ieee80211_hw *hw,
 	ath_dbg(common, CONFIG, "Attach a VIF of type: %d\n", vif->type);
 	sc->cur_chan->nvifs++;
 
+	if (vif->type == NL80211_IFTYPE_STATION && ath9k_is_chanctx_enabled())
+		vif->driver_flags |= IEEE80211_VIF_GET_NOA_UPDATE;
+
 	if (ath9k_uses_beacons(vif->type))
 		ath9k_beacon_assign_slot(sc, vif);
 
-- 
cgit v0.10.2


From aea57edf80c6e96d6dc24757599396af99c02b19 Mon Sep 17 00:00:00 2001
From: Alexander Tsoy <alexander@tsoy.me>
Date: Fri, 8 Jan 2016 01:26:03 +0300
Subject: ath9k_htc: add device ID for Toshiba WLM-20U2/GN-1080

This device is available under different marketing names:
WLM-20U2 - Wireless USB Dongle for Toshiba TVs
GN-1080 - Wireless LAN Module for Toshiba MFPs.

Signed-off-by: Alexander Tsoy <alexander@tsoy.me>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index 165dd20..8cbf490 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -55,6 +55,8 @@ static struct usb_device_id ath9k_hif_usb_ids[] = {
 	  .driver_info = AR9280_USB },  /* Buffalo WLI-UV-AG300P */
 	{ USB_DEVICE(0x04da, 0x3904),
 	  .driver_info = AR9280_USB },
+	{ USB_DEVICE(0x0930, 0x0a08),
+	  .driver_info = AR9280_USB },  /* Toshiba WLM-20U2 and GN-1080 */
 
 	{ USB_DEVICE(0x0cf3, 0x20ff),
 	  .driver_info = STORAGE_DEVICE },
-- 
cgit v0.10.2


From c5a157e4d7d7f1d39dbb382bc3ad7ea9f248ed7d Mon Sep 17 00:00:00 2001
From: Lior David <liord@qti.qualcomm.com>
Date: Sun, 17 Jan 2016 12:39:06 +0200
Subject: wil6210: fix privacy flag calculation in change_beacon

Currently the privacy flag calculation in change_beacon
checks for RSN IE inside proberesp_ies, but normally it
is not found there. It works today because of code inside
the function wil_fix_bcon, but wil_fix_bcon is not directly
related to change_beacon, and it may be changed or removed
in the future and break the calculation.
To fix this issue, change the privacy flag calculation
to check RSN IE inside the beacon itself. The new check is more
reliable and will not be sensitive to changes in wil_fix_bcon.

Signed-off-by: Lior David <qca_liord@qca.qualcomm.com>
Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 20d07ef..182f071 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -828,9 +828,9 @@ static int wil_cfg80211_change_beacon(struct wiphy *wiphy,
 		wil_print_bcon_data(bcon);
 	}
 
-	if (bcon->proberesp_ies &&
-	    cfg80211_find_ie(WLAN_EID_RSN, bcon->proberesp_ies,
-			     bcon->proberesp_ies_len))
+	if (bcon->tail &&
+	    cfg80211_find_ie(WLAN_EID_RSN, bcon->tail,
+			     bcon->tail_len))
 		privacy = 1;
 
 	/* in case privacy has changed, need to restart the AP */
-- 
cgit v0.10.2


From c100c883e785802ab333604237928a65ae62c6bc Mon Sep 17 00:00:00 2001
From: Lior David <qca_liord@qca.qualcomm.com>
Date: Sun, 17 Jan 2016 12:39:07 +0200
Subject: wil6210: use extra IEs from probe response

In the start_ap/change_beacon API, when we set up probe response
offloading, we only use the IE list from the probe response
template and not the IE list from the proberesp_ies argument.
As a result, we miss important IEs and it causes problems with
some scenarios such as P2P.
With this change, we merge the list of IEs from the probe response
template and proberesp_ies and send the merged list to the FW
for offloading. It is still FW responsibility to filter out
irrelevant IEs when sending probe response, based on the actual
contents of the probe request.
Also in case association response termplate is not provided,
we will use the merged list of IEs from probe response in the
association response as well.

Signed-off-by: Lior David <qca_liord@qca.qualcomm.com>
Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 182f071..774352f 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -696,6 +696,79 @@ static int wil_cancel_remain_on_channel(struct wiphy *wiphy,
 	return rc;
 }
 
+/**
+ * find a specific IE in a list of IEs
+ * return a pointer to the beginning of IE in the list
+ * or NULL if not found
+ */
+static const u8 *_wil_cfg80211_find_ie(const u8 *ies, u16 ies_len, const u8 *ie,
+				       u16 ie_len)
+{
+	struct ieee80211_vendor_ie *vie;
+	u32 oui;
+
+	/* IE tag at offset 0, length at offset 1 */
+	if (ie_len < 2 || 2 + ie[1] > ie_len)
+		return NULL;
+
+	if (ie[0] != WLAN_EID_VENDOR_SPECIFIC)
+		return cfg80211_find_ie(ie[0], ies, ies_len);
+
+	/* make sure there is room for 3 bytes OUI + 1 byte OUI type */
+	if (ie[1] < 4)
+		return NULL;
+	vie = (struct ieee80211_vendor_ie *)ie;
+	oui = vie->oui[0] << 16 | vie->oui[1] << 8 | vie->oui[2];
+	return cfg80211_find_vendor_ie(oui, vie->oui_type, ies,
+				       ies_len);
+}
+
+/**
+ * merge the IEs in two lists into a single list.
+ * do not include IEs from the second list which exist in the first list.
+ * add only vendor specific IEs from second list to keep
+ * the merged list sorted (since vendor-specific IE has the
+ * highest tag number)
+ * caller must free the allocated memory for merged IEs
+ */
+static int _wil_cfg80211_merge_extra_ies(const u8 *ies1, u16 ies1_len,
+					 const u8 *ies2, u16 ies2_len,
+					 u8 **merged_ies, u16 *merged_len)
+{
+	u8 *buf, *dpos;
+	const u8 *spos;
+
+	if (ies1_len == 0 && ies2_len == 0) {
+		*merged_ies = NULL;
+		*merged_len = 0;
+		return 0;
+	}
+
+	buf = kmalloc(ies1_len + ies2_len, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+	memcpy(buf, ies1, ies1_len);
+	dpos = buf + ies1_len;
+	spos = ies2;
+	while (spos + 1 < ies2 + ies2_len) {
+		/* IE tag at offset 0, length at offset 1 */
+		u16 ielen = 2 + spos[1];
+
+		if (spos + ielen > ies2 + ies2_len)
+			break;
+		if (spos[0] == WLAN_EID_VENDOR_SPECIFIC &&
+		    !_wil_cfg80211_find_ie(ies1, ies1_len, spos, ielen)) {
+			memcpy(dpos, spos, ielen);
+			dpos += ielen;
+		}
+		spos += ielen;
+	}
+
+	*merged_ies = buf;
+	*merged_len = dpos - buf;
+	return 0;
+}
+
 static void wil_print_bcon_data(struct cfg80211_beacon_data *b)
 {
 	print_hex_dump_bytes("head     ", DUMP_PREFIX_OFFSET,
@@ -712,49 +785,49 @@ static void wil_print_bcon_data(struct cfg80211_beacon_data *b)
 			     b->assocresp_ies, b->assocresp_ies_len);
 }
 
-static int wil_fix_bcon(struct wil6210_priv *wil,
-			struct cfg80211_beacon_data *bcon)
-{
-	struct ieee80211_mgmt *f = (struct ieee80211_mgmt *)bcon->probe_resp;
-	size_t hlen = offsetof(struct ieee80211_mgmt, u.probe_resp.variable);
-
-	if (bcon->probe_resp_len <= hlen)
-		return 0;
-
-/* always use IE's from full probe frame, they has more info
- * notable RSN
- */
-	bcon->proberesp_ies = f->u.probe_resp.variable;
-	bcon->proberesp_ies_len = bcon->probe_resp_len - hlen;
-	if (!bcon->assocresp_ies) {
-		bcon->assocresp_ies = bcon->proberesp_ies;
-		bcon->assocresp_ies_len = bcon->proberesp_ies_len;
-	}
-
-	return 1;
-}
-
 /* internal functions for device reset and starting AP */
 static int _wil_cfg80211_set_ies(struct wiphy *wiphy,
 				 struct cfg80211_beacon_data *bcon)
 {
 	int rc;
 	struct wil6210_priv *wil = wiphy_to_wil(wiphy);
+	u16 len = 0, proberesp_len = 0;
+	u8 *ies = NULL, *proberesp = NULL;
+
+	if (bcon->probe_resp) {
+		struct ieee80211_mgmt *f =
+			(struct ieee80211_mgmt *)bcon->probe_resp;
+		size_t hlen = offsetof(struct ieee80211_mgmt,
+				       u.probe_resp.variable);
+		proberesp = f->u.probe_resp.variable;
+		proberesp_len = bcon->probe_resp_len - hlen;
+	}
+	rc = _wil_cfg80211_merge_extra_ies(proberesp,
+					   proberesp_len,
+					   bcon->proberesp_ies,
+					   bcon->proberesp_ies_len,
+					   &ies, &len);
 
-	rc = wmi_set_ie(wil, WMI_FRAME_PROBE_RESP, bcon->proberesp_ies_len,
-			bcon->proberesp_ies);
 	if (rc)
-		return rc;
+		goto out;
 
-	rc = wmi_set_ie(wil, WMI_FRAME_ASSOC_RESP, bcon->assocresp_ies_len,
-			bcon->assocresp_ies);
+	rc = wmi_set_ie(wil, WMI_FRAME_PROBE_RESP, len, ies);
+	if (rc)
+		goto out;
+
+	if (bcon->assocresp_ies)
+		rc = wmi_set_ie(wil, WMI_FRAME_ASSOC_RESP,
+				bcon->assocresp_ies_len, bcon->assocresp_ies);
+	else
+		rc = wmi_set_ie(wil, WMI_FRAME_ASSOC_RESP, len, ies);
 #if 0 /* to use beacon IE's, remove this #if 0 */
 	if (rc)
-		return rc;
+		goto out;
 
 	rc = wmi_set_ie(wil, WMI_FRAME_BEACON, bcon->tail_len, bcon->tail);
 #endif
-
+out:
+	kfree(ies);
 	return rc;
 }
 
@@ -823,11 +896,6 @@ static int wil_cfg80211_change_beacon(struct wiphy *wiphy,
 	wil_dbg_misc(wil, "%s()\n", __func__);
 	wil_print_bcon_data(bcon);
 
-	if (wil_fix_bcon(wil, bcon)) {
-		wil_dbg_misc(wil, "Fixed bcon\n");
-		wil_print_bcon_data(bcon);
-	}
-
 	if (bcon->tail &&
 	    cfg80211_find_ie(WLAN_EID_RSN, bcon->tail,
 			     bcon->tail_len))
@@ -900,11 +968,6 @@ static int wil_cfg80211_start_ap(struct wiphy *wiphy,
 	wil_print_bcon_data(bcon);
 	wil_print_crypto(wil, crypto);
 
-	if (wil_fix_bcon(wil, bcon)) {
-		wil_dbg_misc(wil, "Fixed bcon\n");
-		wil_print_bcon_data(bcon);
-	}
-
 	rc = _wil_cfg80211_start_ap(wiphy, ndev,
 				    info->ssid, info->ssid_len, info->privacy,
 				    info->beacon_interval, channel->hw_value,
-- 
cgit v0.10.2


From 640751ac30b2ddfbe06ad712aa7d57a8d89a7eef Mon Sep 17 00:00:00 2001
From: Maya Erez <qca_merez@qca.qualcomm.com>
Date: Sun, 17 Jan 2016 12:39:08 +0200
Subject: wil6210: handle tx completions only if vring tx data is enabled

wil_vring_fini_tx is called in disconnect in order to free the
allocated vrings.
wil_vring_fini_tx is disabling the vring_tx_data before napi_synchronize
is called in order to avoid the tx handling of this vring, while
wil_vring_free is called only after napi finished the current handling
of the tx completed packets.
Due to that, in case of disconnect, wil6210_netdev_poll_tx can be called
when vring->va is not NULL but vring_tx_data[i]->enabled is already
disabled.

This patch checks vring_tx_data[i]->enabled in wil6210_netdev_poll_tx
to prevent handling of disabled vrings.

Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/netdev.c b/drivers/net/wireless/ath/wil6210/netdev.c
index 56aaa2d..ecc3c1b 100644
--- a/drivers/net/wireless/ath/wil6210/netdev.c
+++ b/drivers/net/wireless/ath/wil6210/netdev.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -108,8 +108,9 @@ static int wil6210_netdev_poll_tx(struct napi_struct *napi, int budget)
 	/* always process ALL Tx complete, regardless budget - it is fast */
 	for (i = 0; i < WIL6210_MAX_TX_RINGS; i++) {
 		struct vring *vring = &wil->vring_tx[i];
+		struct vring_tx_data *txdata = &wil->vring_tx_data[i];
 
-		if (!vring->va)
+		if (!vring->va || !txdata->enabled)
 			continue;
 
 		tx_done += wil_tx_complete(wil, i);
-- 
cgit v0.10.2


From b729aaf066b013458e95cd6c69a5e371eb9ce355 Mon Sep 17 00:00:00 2001
From: Maya Erez <qca_merez@qca.qualcomm.com>
Date: Sun, 17 Jan 2016 12:39:09 +0200
Subject: wil6210: find TX vring only if vring_tx_data is enabled

In TX vring allocation, vring_tx->va is allocated before WMI command to
configure the vring is sent to the device. As the WMI command can take
time to complete, it can lead to scenarios where vring_tx->va is not NULL
but vring is still not enabled.

This patch adds a check that vring_tx_data is enabled before returning
a valid TX vring.

Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index 7887e6c..c22ee8b 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -911,10 +911,11 @@ static struct vring *wil_find_tx_ucast(struct wil6210_priv *wil,
 			continue;
 		if (wil->vring2cid_tid[i][0] == cid) {
 			struct vring *v = &wil->vring_tx[i];
+			struct vring_tx_data *txdata = &wil->vring_tx_data[i];
 
 			wil_dbg_txrx(wil, "%s(%pM) -> [%d]\n",
 				     __func__, eth->h_dest, i);
-			if (v->va) {
+			if (v->va && txdata->enabled) {
 				return v;
 			} else {
 				wil_dbg_txrx(wil, "vring[%d] not valid\n", i);
@@ -935,6 +936,7 @@ static struct vring *wil_find_tx_vring_sta(struct wil6210_priv *wil,
 	struct vring *v;
 	int i;
 	u8 cid;
+	struct vring_tx_data *txdata;
 
 	/* In the STA mode, it is expected to have only 1 VRING
 	 * for the AP we connected to.
@@ -942,7 +944,8 @@ static struct vring *wil_find_tx_vring_sta(struct wil6210_priv *wil,
 	 */
 	for (i = 0; i < WIL6210_MAX_TX_RINGS; i++) {
 		v = &wil->vring_tx[i];
-		if (!v->va)
+		txdata = &wil->vring_tx_data[i];
+		if (!v->va || !txdata->enabled)
 			continue;
 
 		cid = wil->vring2cid_tid[i][0];
@@ -978,12 +981,14 @@ static struct vring *wil_find_tx_bcast_1(struct wil6210_priv *wil,
 					 struct sk_buff *skb)
 {
 	struct vring *v;
+	struct vring_tx_data *txdata;
 	int i = wil->bcast_vring;
 
 	if (i < 0)
 		return NULL;
 	v = &wil->vring_tx[i];
-	if (!v->va)
+	txdata = &wil->vring_tx_data[i];
+	if (!v->va || !txdata->enabled)
 		return NULL;
 	if (!wil->vring_tx_data[i].dot1x_open &&
 	    (skb->protocol != cpu_to_be16(ETH_P_PAE)))
@@ -1010,11 +1015,13 @@ static struct vring *wil_find_tx_bcast_2(struct wil6210_priv *wil,
 	u8 cid;
 	struct ethhdr *eth = (void *)skb->data;
 	char *src = eth->h_source;
+	struct vring_tx_data *txdata;
 
 	/* find 1-st vring eligible for data */
 	for (i = 0; i < WIL6210_MAX_TX_RINGS; i++) {
 		v = &wil->vring_tx[i];
-		if (!v->va)
+		txdata = &wil->vring_tx_data[i];
+		if (!v->va || !txdata->enabled)
 			continue;
 
 		cid = wil->vring2cid_tid[i][0];
-- 
cgit v0.10.2


From 0916d9f2b6bea53a830ae5061b63a515c04da62e Mon Sep 17 00:00:00 2001
From: Maya Erez <qca_merez@qca.qualcomm.com>
Date: Sun, 17 Jan 2016 12:39:10 +0200
Subject: wil6210: handle multiple connect/disconnect events

In the current solution wil6210 configures the vring in a worker
and holds only one pending CID. This implementation may lead to
race conditions between connect and disconnect events of multiple
stations or fast connect/disconnect events of the same station.

In order to allow the removal of the connect worker and handling of
WMI_VRING_CFG_DONE_EVENTID in the connect event, the WMI replies
that provide the reply in a given buffer needs to be handled
immediately in the WMI event interrupt thread.
To prevent deadlocks, WMI replies that requires additional
handling are still handled via the events list.

Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index b39f0bf..0652efe 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -155,7 +155,7 @@ __acquires(&sta->tid_rx_lock) __releases(&sta->tid_rx_lock)
 
 	if (sta->status != wil_sta_unused) {
 		if (!from_event)
-			wmi_disconnect_sta(wil, sta->addr, reason_code);
+			wmi_disconnect_sta(wil, sta->addr, reason_code, true);
 
 		switch (wdev->iftype) {
 		case NL80211_IFTYPE_AP:
@@ -195,8 +195,8 @@ static void _wil6210_disconnect(struct wil6210_priv *wil, const u8 *bssid,
 	struct wireless_dev *wdev = wil->wdev;
 
 	might_sleep();
-	wil_dbg_misc(wil, "%s(bssid=%pM, reason=%d, ev%s)\n", __func__, bssid,
-		     reason_code, from_event ? "+" : "-");
+	wil_info(wil, "%s(bssid=%pM, reason=%d, ev%s)\n", __func__, bssid,
+		 reason_code, from_event ? "+" : "-");
 
 	/* Cases are:
 	 * - disconnect single STA, still connected
@@ -258,13 +258,16 @@ static void wil_disconnect_worker(struct work_struct *work)
 static void wil_connect_timer_fn(ulong x)
 {
 	struct wil6210_priv *wil = (void *)x;
+	bool q;
 
-	wil_dbg_misc(wil, "Connect timeout\n");
+	wil_err(wil, "Connect timeout detected, disconnect station\n");
 
 	/* reschedule to thread context - disconnect won't
-	 * run from atomic context
+	 * run from atomic context.
+	 * queue on wmi_wq to prevent race with connect event.
 	 */
-	schedule_work(&wil->disconnect_worker);
+	q = queue_work(wil->wmi_wq, &wil->disconnect_worker);
+	wil_dbg_wmi(wil, "queue_work of disconnect_worker -> %d\n", q);
 }
 
 static void wil_scan_timer_fn(ulong x)
@@ -369,6 +372,32 @@ static int wil_find_free_vring(struct wil6210_priv *wil)
 	return -EINVAL;
 }
 
+int wil_tx_init(struct wil6210_priv *wil, int cid)
+{
+	int rc = -EINVAL, ringid;
+
+	if (cid < 0) {
+		wil_err(wil, "No connection pending\n");
+		goto out;
+	}
+	ringid = wil_find_free_vring(wil);
+	if (ringid < 0) {
+		wil_err(wil, "No free vring found\n");
+		goto out;
+	}
+
+	wil_dbg_wmi(wil, "Configure for connection CID %d vring %d\n",
+		    cid, ringid);
+
+	rc = wil_vring_init_tx(wil, ringid, 1 << tx_ring_order, cid, 0);
+	if (rc)
+		wil_err(wil, "wil_vring_init_tx for CID %d vring %d failed\n",
+			cid, ringid);
+
+out:
+	return rc;
+}
+
 int wil_bcast_init(struct wil6210_priv *wil)
 {
 	int ri = wil->bcast_vring, rc;
@@ -399,41 +428,6 @@ void wil_bcast_fini(struct wil6210_priv *wil)
 	wil_vring_fini_tx(wil, ri);
 }
 
-static void wil_connect_worker(struct work_struct *work)
-{
-	int rc, cid, ringid;
-	struct wil6210_priv *wil = container_of(work, struct wil6210_priv,
-						connect_worker);
-	struct net_device *ndev = wil_to_ndev(wil);
-
-	mutex_lock(&wil->mutex);
-
-	cid = wil->pending_connect_cid;
-	if (cid < 0) {
-		wil_err(wil, "No connection pending\n");
-		goto out;
-	}
-	ringid = wil_find_free_vring(wil);
-	if (ringid < 0) {
-		wil_err(wil, "No free vring found\n");
-		goto out;
-	}
-
-	wil_dbg_wmi(wil, "Configure for connection CID %d vring %d\n",
-		    cid, ringid);
-
-	rc = wil_vring_init_tx(wil, ringid, 1 << tx_ring_order, cid, 0);
-	wil->pending_connect_cid = -1;
-	if (rc == 0) {
-		wil->sta[cid].status = wil_sta_connected;
-		netif_tx_wake_all_queues(ndev);
-	} else {
-		wil_disconnect_cid(wil, cid, WLAN_REASON_UNSPECIFIED, true);
-	}
-out:
-	mutex_unlock(&wil->mutex);
-}
-
 int wil_priv_init(struct wil6210_priv *wil)
 {
 	uint i;
@@ -453,12 +447,10 @@ int wil_priv_init(struct wil6210_priv *wil)
 	init_completion(&wil->wmi_ready);
 	init_completion(&wil->wmi_call);
 
-	wil->pending_connect_cid = -1;
 	wil->bcast_vring = -1;
 	setup_timer(&wil->connect_timer, wil_connect_timer_fn, (ulong)wil);
 	setup_timer(&wil->scan_timer, wil_scan_timer_fn, (ulong)wil);
 
-	INIT_WORK(&wil->connect_worker, wil_connect_worker);
 	INIT_WORK(&wil->disconnect_worker, wil_disconnect_worker);
 	INIT_WORK(&wil->wmi_event_worker, wmi_event_worker);
 	INIT_WORK(&wil->fw_error_worker, wil_fw_error_worker);
@@ -844,7 +836,6 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw)
 	}
 
 	/* init after reset */
-	wil->pending_connect_cid = -1;
 	wil->ap_isolate = 0;
 	reinit_completion(&wil->wmi_ready);
 	reinit_completion(&wil->wmi_call);
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index c22ee8b..9680b97 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -794,6 +794,9 @@ int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
 	txdata->dot1x_open = false;
 	txdata->enabled = 0;
 	wil_vring_free(wil, vring, 1);
+	wil->vring2cid_tid[id][0] = WIL6210_MAX_CID;
+	wil->vring2cid_tid[id][1] = 0;
+
  out:
 
 	return rc;
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 235e205..1b8fa1d 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -581,12 +581,10 @@ struct wil6210_priv {
 	struct workqueue_struct *wmi_wq; /* for deferred calls */
 	struct work_struct wmi_event_worker;
 	struct workqueue_struct *wq_service;
-	struct work_struct connect_worker;
 	struct work_struct disconnect_worker;
 	struct work_struct fw_error_worker;	/* for FW error recovery */
 	struct timer_list connect_timer;
 	struct timer_list scan_timer; /* detect scan timeout */
-	int pending_connect_cid;
 	struct list_head pending_wmi_ev;
 	/*
 	 * protect pending_wmi_ev
@@ -756,7 +754,8 @@ int wmi_rx_chain_add(struct wil6210_priv *wil, struct vring *vring);
 int wmi_p2p_cfg(struct wil6210_priv *wil, int channel);
 int wmi_rxon(struct wil6210_priv *wil, bool on);
 int wmi_get_temperature(struct wil6210_priv *wil, u32 *t_m, u32 *t_r);
-int wmi_disconnect_sta(struct wil6210_priv *wil, const u8 *mac, u16 reason);
+int wmi_disconnect_sta(struct wil6210_priv *wil, const u8 *mac, u16 reason,
+		       bool full_disconnect);
 int wmi_addba(struct wil6210_priv *wil, u8 ringid, u8 size, u16 timeout);
 int wmi_delba_tx(struct wil6210_priv *wil, u8 ringid, u16 reason);
 int wmi_delba_rx(struct wil6210_priv *wil, u8 cidxtid, u16 reason);
@@ -807,6 +806,7 @@ void wil_rx_fini(struct wil6210_priv *wil);
 int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
 		      int cid, int tid);
 void wil_vring_fini_tx(struct wil6210_priv *wil, int id);
+int wil_tx_init(struct wil6210_priv *wil, int cid);
 int wil_vring_init_bcast(struct wil6210_priv *wil, int id, int size);
 int wil_bcast_init(struct wil6210_priv *wil);
 void wil_bcast_fini(struct wil6210_priv *wil);
diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c
index e3ea74c..e1a6cb8 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.c
+++ b/drivers/net/wireless/ath/wil6210/wmi.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -426,6 +426,7 @@ static void wmi_evt_connect(struct wil6210_priv *wil, int id, void *d, int len)
 	const size_t assoc_req_ie_offset = sizeof(u16) * 2;
 	/* capinfo(u16) + status_code(u16) + associd(u16) + IEs */
 	const size_t assoc_resp_ie_offset = sizeof(u16) * 3;
+	int rc;
 
 	if (len < sizeof(*evt)) {
 		wil_err(wil, "Connect event too short : %d bytes\n", len);
@@ -445,8 +446,8 @@ static void wmi_evt_connect(struct wil6210_priv *wil, int id, void *d, int len)
 	}
 
 	ch = evt->channel + 1;
-	wil_dbg_wmi(wil, "Connect %pM channel [%d] cid %d\n",
-		    evt->bssid, ch, evt->cid);
+	wil_info(wil, "Connect %pM channel [%d] cid %d\n",
+		 evt->bssid, ch, evt->cid);
 	wil_hex_dump_wmi("connect AI : ", DUMP_PREFIX_OFFSET, 16, 1,
 			 evt->assoc_info, len - sizeof(*evt), true);
 
@@ -468,20 +469,67 @@ static void wmi_evt_connect(struct wil6210_priv *wil, int id, void *d, int len)
 		assoc_resp_ielen = 0;
 	}
 
+	mutex_lock(&wil->mutex);
+	if (test_bit(wil_status_resetting, wil->status) ||
+	    !test_bit(wil_status_fwready, wil->status)) {
+		wil_err(wil, "status_resetting, cancel connect event, CID %d\n",
+			evt->cid);
+		mutex_unlock(&wil->mutex);
+		/* no need for cleanup, wil_reset will do that */
+		return;
+	}
+
 	if ((wdev->iftype == NL80211_IFTYPE_STATION) ||
 	    (wdev->iftype == NL80211_IFTYPE_P2P_CLIENT)) {
 		if (!test_bit(wil_status_fwconnecting, wil->status)) {
 			wil_err(wil, "Not in connecting state\n");
+			mutex_unlock(&wil->mutex);
 			return;
 		}
 		del_timer_sync(&wil->connect_timer);
-		cfg80211_connect_result(ndev, evt->bssid,
-					assoc_req_ie, assoc_req_ielen,
-					assoc_resp_ie, assoc_resp_ielen,
-					WLAN_STATUS_SUCCESS, GFP_KERNEL);
+	}
+
+	/* FIXME FW can transmit only ucast frames to peer */
+	/* FIXME real ring_id instead of hard coded 0 */
+	ether_addr_copy(wil->sta[evt->cid].addr, evt->bssid);
+	wil->sta[evt->cid].status = wil_sta_conn_pending;
 
+	rc = wil_tx_init(wil, evt->cid);
+	if (rc) {
+		wil_err(wil, "%s: config tx vring failed for CID %d, rc (%d)\n",
+			__func__, evt->cid, rc);
+		wmi_disconnect_sta(wil, wil->sta[evt->cid].addr,
+				   WLAN_REASON_UNSPECIFIED, false);
+	} else {
+		wil_info(wil, "%s: successful connection to CID %d\n",
+			 __func__, evt->cid);
+	}
+
+	if ((wdev->iftype == NL80211_IFTYPE_STATION) ||
+	    (wdev->iftype == NL80211_IFTYPE_P2P_CLIENT)) {
+		if (rc) {
+			netif_tx_stop_all_queues(ndev);
+			netif_carrier_off(ndev);
+			wil_err(wil,
+				"%s: cfg80211_connect_result with failure\n",
+				__func__);
+			cfg80211_connect_result(ndev, evt->bssid, NULL, 0,
+						NULL, 0,
+						WLAN_STATUS_UNSPECIFIED_FAILURE,
+						GFP_KERNEL);
+			goto out;
+		} else {
+			cfg80211_connect_result(ndev, evt->bssid,
+						assoc_req_ie, assoc_req_ielen,
+						assoc_resp_ie, assoc_resp_ielen,
+						WLAN_STATUS_SUCCESS,
+						GFP_KERNEL);
+		}
 	} else if ((wdev->iftype == NL80211_IFTYPE_AP) ||
 		   (wdev->iftype == NL80211_IFTYPE_P2P_GO)) {
+		if (rc)
+			goto out;
+
 		memset(&sinfo, 0, sizeof(sinfo));
 
 		sinfo.generation = wil->sinfo_gen++;
@@ -492,17 +540,21 @@ static void wmi_evt_connect(struct wil6210_priv *wil, int id, void *d, int len)
 		}
 
 		cfg80211_new_sta(ndev, evt->bssid, &sinfo, GFP_KERNEL);
+	} else {
+		wil_err(wil, "%s: unhandled iftype %d for CID %d\n",
+			__func__, wdev->iftype, evt->cid);
+		goto out;
 	}
-	clear_bit(wil_status_fwconnecting, wil->status);
-	set_bit(wil_status_fwconnected, wil->status);
 
-	/* FIXME FW can transmit only ucast frames to peer */
-	/* FIXME real ring_id instead of hard coded 0 */
-	ether_addr_copy(wil->sta[evt->cid].addr, evt->bssid);
-	wil->sta[evt->cid].status = wil_sta_conn_pending;
+	wil->sta[evt->cid].status = wil_sta_connected;
+	set_bit(wil_status_fwconnected, wil->status);
+	netif_tx_wake_all_queues(ndev);
 
-	wil->pending_connect_cid = evt->cid;
-	queue_work(wil->wq_service, &wil->connect_worker);
+out:
+	if (rc)
+		wil->sta[evt->cid].status = wil_sta_unused;
+	clear_bit(wil_status_fwconnecting, wil->status);
+	mutex_unlock(&wil->mutex);
 }
 
 static void wmi_evt_disconnect(struct wil6210_priv *wil, int id,
@@ -511,8 +563,8 @@ static void wmi_evt_disconnect(struct wil6210_priv *wil, int id,
 	struct wmi_disconnect_event *evt = d;
 	u16 reason_code = le16_to_cpu(evt->protocol_reason_status);
 
-	wil_dbg_wmi(wil, "Disconnect %pM reason [proto %d wmi %d]\n",
-		    evt->bssid, reason_code, evt->disconnect_reason);
+	wil_info(wil, "Disconnect %pM reason [proto %d wmi %d]\n",
+		 evt->bssid, reason_code, evt->disconnect_reason);
 
 	wil->sinfo_gen++;
 
@@ -727,6 +779,7 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 	void __iomem *src;
 	ulong flags;
 	unsigned n;
+	unsigned int num_immed_reply = 0;
 
 	if (!test_bit(wil_status_mbox_ready, wil->status)) {
 		wil_err(wil, "Reset in progress. Cannot handle WMI event\n");
@@ -736,6 +789,7 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 	for (n = 0;; n++) {
 		u16 len;
 		bool q;
+		bool immed_reply = false;
 
 		r->head = wil_r(wil, RGF_MBOX +
 				offsetof(struct wil6210_mbox_ctl, rx.head));
@@ -784,6 +838,13 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 			struct wil6210_mbox_hdr_wmi *wmi = &evt->event.wmi;
 			u16 id = le16_to_cpu(wmi->id);
 			u32 tstamp = le32_to_cpu(wmi->timestamp);
+			if (wil->reply_id && wil->reply_id == id) {
+				if (wil->reply_buf) {
+					memcpy(wil->reply_buf, wmi,
+					       min(len, wil->reply_size));
+					immed_reply = true;
+				}
+			}
 
 			wil_dbg_wmi(wil, "WMI event 0x%04x MID %d @%d msec\n",
 				    id, wmi->mid, tstamp);
@@ -799,15 +860,24 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 		wil_w(wil, RGF_MBOX +
 		      offsetof(struct wil6210_mbox_ctl, rx.tail), r->tail);
 
-		/* add to the pending list */
-		spin_lock_irqsave(&wil->wmi_ev_lock, flags);
-		list_add_tail(&evt->list, &wil->pending_wmi_ev);
-		spin_unlock_irqrestore(&wil->wmi_ev_lock, flags);
-		q = queue_work(wil->wmi_wq, &wil->wmi_event_worker);
-		wil_dbg_wmi(wil, "queue_work -> %d\n", q);
+		if (immed_reply) {
+			wil_dbg_wmi(wil, "%s: Complete WMI 0x%04x\n",
+				    __func__, wil->reply_id);
+			kfree(evt);
+			num_immed_reply++;
+			complete(&wil->wmi_call);
+		} else {
+			/* add to the pending list */
+			spin_lock_irqsave(&wil->wmi_ev_lock, flags);
+			list_add_tail(&evt->list, &wil->pending_wmi_ev);
+			spin_unlock_irqrestore(&wil->wmi_ev_lock, flags);
+			q = queue_work(wil->wmi_wq, &wil->wmi_event_worker);
+			wil_dbg_wmi(wil, "queue_work -> %d\n", q);
+		}
 	}
 	/* normally, 1 event per IRQ should be processed */
-	wil_dbg_wmi(wil, "%s -> %d events queued\n", __func__, n);
+	wil_dbg_wmi(wil, "%s -> %d events queued, %d completed\n", __func__,
+		    n - num_immed_reply, num_immed_reply);
 }
 
 int wmi_call(struct wil6210_priv *wil, u16 cmdid, void *buf, u16 len,
@@ -1184,7 +1254,8 @@ int wmi_get_temperature(struct wil6210_priv *wil, u32 *t_bb, u32 *t_rf)
 	return 0;
 }
 
-int wmi_disconnect_sta(struct wil6210_priv *wil, const u8 *mac, u16 reason)
+int wmi_disconnect_sta(struct wil6210_priv *wil, const u8 *mac, u16 reason,
+		       bool full_disconnect)
 {
 	int rc;
 	u16 reason_code;
@@ -1208,19 +1279,20 @@ int wmi_disconnect_sta(struct wil6210_priv *wil, const u8 *mac, u16 reason)
 		return rc;
 	}
 
-	/* call event handler manually after processing wmi_call,
-	 * to avoid deadlock - disconnect event handler acquires wil->mutex
-	 * while it is already held here
-	 */
-	reason_code = le16_to_cpu(reply.evt.protocol_reason_status);
-
-	wil_dbg_wmi(wil, "Disconnect %pM reason [proto %d wmi %d]\n",
-		    reply.evt.bssid, reason_code,
-		    reply.evt.disconnect_reason);
+	if (full_disconnect) {
+		/* call event handler manually after processing wmi_call,
+		 * to avoid deadlock - disconnect event handler acquires
+		 * wil->mutex while it is already held here
+		 */
+		reason_code = le16_to_cpu(reply.evt.protocol_reason_status);
 
-	wil->sinfo_gen++;
-	wil6210_disconnect(wil, reply.evt.bssid, reason_code, true);
+		wil_dbg_wmi(wil, "Disconnect %pM reason [proto %d wmi %d]\n",
+			    reply.evt.bssid, reason_code,
+			    reply.evt.disconnect_reason);
 
+		wil->sinfo_gen++;
+		wil6210_disconnect(wil, reply.evt.bssid, reason_code, true);
+	}
 	return 0;
 }
 
@@ -1348,14 +1420,11 @@ static void wmi_event_handle(struct wil6210_priv *wil,
 			    id, wil->reply_id);
 		/* check if someone waits for this event */
 		if (wil->reply_id && wil->reply_id == id) {
-			if (wil->reply_buf) {
-				memcpy(wil->reply_buf, wmi,
-				       min(len, wil->reply_size));
-			} else {
-				wmi_evt_call_handler(wil, id, evt_data,
-						     len - sizeof(*wmi));
-			}
-			wil_dbg_wmi(wil, "Complete WMI 0x%04x\n", id);
+			WARN_ON(wil->reply_buf);
+			wmi_evt_call_handler(wil, id, evt_data,
+					     len - sizeof(*wmi));
+			wil_dbg_wmi(wil, "%s: Complete WMI 0x%04x\n",
+				    __func__, id);
 			complete(&wil->wmi_call);
 			return;
 		}
-- 
cgit v0.10.2


From 424aecaa797901708e7f260b642d27b077cdb491 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Thu, 14 Jan 2016 03:15:13 +0100
Subject: ath9k_hw: add low power tx gain table for AR953x

Used in some newer TP-Link AR9533 devices.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_hw.c b/drivers/net/wireless/ath/ath9k/ar9003_hw.c
index 8b238c1..2fe12b0 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_hw.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_hw.c
@@ -698,6 +698,9 @@ static void ar9003_tx_gain_table_mode2(struct ath_hw *ah)
 	else if (AR_SREV_9340(ah))
 		INIT_INI_ARRAY(&ah->iniModesTxGain,
 			ar9340Modes_low_ob_db_tx_gain_table_1p0);
+	else if (AR_SREV_9531_11(ah))
+		INIT_INI_ARRAY(&ah->iniModesTxGain,
+			       qca953x_1p1_modes_no_xpa_low_power_tx_gain_table);
 	else if (AR_SREV_9485_11_OR_LATER(ah))
 		INIT_INI_ARRAY(&ah->iniModesTxGain,
 			ar9485Modes_low_ob_db_tx_gain_1_1);
diff --git a/drivers/net/wireless/ath/ath9k/ar953x_initvals.h b/drivers/net/wireless/ath/ath9k/ar953x_initvals.h
index 6fc0d07..c0b90da 100644
--- a/drivers/net/wireless/ath/ath9k/ar953x_initvals.h
+++ b/drivers/net/wireless/ath/ath9k/ar953x_initvals.h
@@ -757,6 +757,71 @@ static const u32 qca953x_1p1_modes_xpa_tx_gain_table[][2] = {
 	{0x00016448, 0x6c927a70},
 };
 
+static const u32 qca953x_1p1_modes_no_xpa_low_power_tx_gain_table[][2] = {
+	/* Addr      allmodes  */
+	{0x0000a2dc, 0xfff55592},
+	{0x0000a2e0, 0xfff99924},
+	{0x0000a2e4, 0xfffe1e00},
+	{0x0000a2e8, 0xffffe000},
+	{0x0000a410, 0x000050d6},
+	{0x0000a500, 0x00000069},
+	{0x0000a504, 0x0400006b},
+	{0x0000a508, 0x0800006d},
+	{0x0000a50c, 0x0c000269},
+	{0x0000a510, 0x1000026b},
+	{0x0000a514, 0x1400026d},
+	{0x0000a518, 0x18000669},
+	{0x0000a51c, 0x1c00066b},
+	{0x0000a520, 0x1d000a68},
+	{0x0000a524, 0x21000a6a},
+	{0x0000a528, 0x25000a6c},
+	{0x0000a52c, 0x29000a6e},
+	{0x0000a530, 0x2d0012a9},
+	{0x0000a534, 0x310012ab},
+	{0x0000a538, 0x350012ad},
+	{0x0000a53c, 0x39001b0a},
+	{0x0000a540, 0x3d001b0c},
+	{0x0000a544, 0x41001b0e},
+	{0x0000a548, 0x43001bae},
+	{0x0000a54c, 0x45001914},
+	{0x0000a550, 0x47001916},
+	{0x0000a554, 0x49001b96},
+	{0x0000a558, 0x49001b96},
+	{0x0000a55c, 0x49001b96},
+	{0x0000a560, 0x49001b96},
+	{0x0000a564, 0x49001b96},
+	{0x0000a568, 0x49001b96},
+	{0x0000a56c, 0x49001b96},
+	{0x0000a570, 0x49001b96},
+	{0x0000a574, 0x49001b96},
+	{0x0000a578, 0x49001b96},
+	{0x0000a57c, 0x49001b96},
+	{0x0000a600, 0x00000000},
+	{0x0000a604, 0x00000000},
+	{0x0000a608, 0x00000000},
+	{0x0000a60c, 0x00000000},
+	{0x0000a610, 0x00000000},
+	{0x0000a614, 0x00000000},
+	{0x0000a618, 0x00804201},
+	{0x0000a61c, 0x01408201},
+	{0x0000a620, 0x01408502},
+	{0x0000a624, 0x01408502},
+	{0x0000a628, 0x01408502},
+	{0x0000a62c, 0x01408502},
+	{0x0000a630, 0x01408502},
+	{0x0000a634, 0x01408502},
+	{0x0000a638, 0x01408502},
+	{0x0000a63c, 0x01408502},
+	{0x0000b2dc, 0xfff55592},
+	{0x0000b2e0, 0xfff99924},
+	{0x0000b2e4, 0xfffe1e00},
+	{0x0000b2e8, 0xffffe000},
+	{0x00016044, 0x044922db},
+	{0x00016048, 0x6c927a70},
+	{0x00016444, 0x044922db},
+	{0x00016448, 0x6c927a70},
+};
+
 static const u32 qca953x_2p0_baseband_core[][2] = {
 	/* Addr      allmodes  */
 	{0x00009800, 0xafe68e30},
-- 
cgit v0.10.2


From 344cd850b55e9dfd4f3ec1ba564f8edbc8136aca Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Fri, 15 Jan 2016 14:41:30 +0100
Subject: ath9k: Drop useless const on chanctx_event_delta() return type

drivers/net/wireless/ath/ath9k/channel.c:230: warning: type qualifiers ignored on function return type

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/channel.c b/drivers/net/wireless/ath/ath9k/channel.c
index 0bcb82a..319cb5f 100644
--- a/drivers/net/wireless/ath/ath9k/channel.c
+++ b/drivers/net/wireless/ath/ath9k/channel.c
@@ -226,7 +226,7 @@ static const char *chanctx_state_string(enum ath_chanctx_state state)
 	}
 }
 
-static const u32 chanctx_event_delta(struct ath_softc *sc)
+static u32 chanctx_event_delta(struct ath_softc *sc)
 {
 	u64 ms;
 	struct timespec ts, *old;
-- 
cgit v0.10.2


From c1cab1df9b651d999922e0f04d01dc046121a237 Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Mon, 18 Jan 2016 09:33:46 +0800
Subject: ath9k: avoid ANI restart if no trigger

Fixes commit 54da20d83f0e ("ath9k_hw: improve ANI processing and rx desensitizing parameters")

Call ath9k_ani_restart() only when the phy error rate reach the
ANI immunity threshold. Sync the logic with internal code base.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ani.c b/drivers/net/wireless/ath/ath9k/ani.c
index 25e45e4..aa82e13 100644
--- a/drivers/net/wireless/ath/ath9k/ani.c
+++ b/drivers/net/wireless/ath/ath9k/ani.c
@@ -450,7 +450,9 @@ void ath9k_hw_ani_monitor(struct ath_hw *ah, struct ath9k_channel *chan)
 		} else if (cckPhyErrRate > ah->config.cck_trig_high) {
 			ath9k_hw_ani_cck_err_trigger(ah);
 			aniState->ofdmsTurn = true;
-		}
+		} else
+			return;
+			
 		ath9k_ani_restart(ah);
 	}
 }
-- 
cgit v0.10.2


From 66b533d5adb3eb0cd111b07e516ab13a47ed34f5 Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Mon, 18 Jan 2016 09:33:47 +0800
Subject: ath9k: clean up ANI per-channel pointer checking

commit c24bd3620c50 ("ath9k: Do not maintain ANI state per-channel")
removed per-channel handling, the code to check 'curchan' also
should be removed as never used.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ani.c b/drivers/net/wireless/ath/ath9k/ani.c
index aa82e13..815efe9 100644
--- a/drivers/net/wireless/ath/ath9k/ani.c
+++ b/drivers/net/wireless/ath/ath9k/ani.c
@@ -126,12 +126,8 @@ static void ath9k_hw_update_mibstats(struct ath_hw *ah,
 
 static void ath9k_ani_restart(struct ath_hw *ah)
 {
-	struct ar5416AniState *aniState;
-
-	if (!ah->curchan)
-		return;
+	struct ar5416AniState *aniState = &ah->ani;
 
-	aniState = &ah->ani;
 	aniState->listenTime = 0;
 
 	ENABLE_REGWRITE_BUFFER(ah);
@@ -221,12 +217,7 @@ static void ath9k_hw_set_ofdm_nil(struct ath_hw *ah, u8 immunityLevel,
 
 static void ath9k_hw_ani_ofdm_err_trigger(struct ath_hw *ah)
 {
-	struct ar5416AniState *aniState;
-
-	if (!ah->curchan)
-		return;
-
-	aniState = &ah->ani;
+	struct ar5416AniState *aniState = &ah->ani;
 
 	if (aniState->ofdmNoiseImmunityLevel < ATH9K_ANI_OFDM_MAX_LEVEL)
 		ath9k_hw_set_ofdm_nil(ah, aniState->ofdmNoiseImmunityLevel + 1, false);
@@ -281,12 +272,7 @@ static void ath9k_hw_set_cck_nil(struct ath_hw *ah, u_int8_t immunityLevel,
 
 static void ath9k_hw_ani_cck_err_trigger(struct ath_hw *ah)
 {
-	struct ar5416AniState *aniState;
-
-	if (!ah->curchan)
-		return;
-
-	aniState = &ah->ani;
+	struct ar5416AniState *aniState = &ah->ani;
 
 	if (aniState->cckNoiseImmunityLevel < ATH9K_ANI_CCK_MAX_LEVEL)
 		ath9k_hw_set_cck_nil(ah, aniState->cckNoiseImmunityLevel + 1,
@@ -299,9 +285,7 @@ static void ath9k_hw_ani_cck_err_trigger(struct ath_hw *ah)
  */
 static void ath9k_hw_ani_lower_immunity(struct ath_hw *ah)
 {
-	struct ar5416AniState *aniState;
-
-	aniState = &ah->ani;
+	struct ar5416AniState *aniState = &ah->ani;
 
 	/* lower OFDM noise immunity */
 	if (aniState->ofdmNoiseImmunityLevel > 0 &&
@@ -329,7 +313,7 @@ void ath9k_ani_reset(struct ath_hw *ah, bool is_scanning)
 	struct ath_common *common = ath9k_hw_common(ah);
 	int ofdm_nil, cck_nil;
 
-	if (!ah->curchan)
+	if (!chan)
 		return;
 
 	BUG_ON(aniState == NULL);
@@ -416,14 +400,10 @@ static bool ath9k_hw_ani_read_counters(struct ath_hw *ah)
 
 void ath9k_hw_ani_monitor(struct ath_hw *ah, struct ath9k_channel *chan)
 {
-	struct ar5416AniState *aniState;
+	struct ar5416AniState *aniState = &ah->ani;
 	struct ath_common *common = ath9k_hw_common(ah);
 	u32 ofdmPhyErrRate, cckPhyErrRate;
 
-	if (!ah->curchan)
-		return;
-
-	aniState = &ah->ani;
 	if (!ath9k_hw_ani_read_counters(ah))
 		return;
 
-- 
cgit v0.10.2


From 3f6cc4e57d304f552abc2bdba716519c5e18b442 Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Mon, 18 Jan 2016 09:33:48 +0800
Subject: ath9k: do not reset while BB panic(0x4000409) on ar9561

BB panic(0x4000409) observed while AP enabling/disabling
bursting.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_phy.c b/drivers/net/wireless/ath/ath9k/ar9003_phy.c
index 201425e..abd9646 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_phy.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_phy.c
@@ -2071,7 +2071,8 @@ void ar9003_hw_attach_phy_ops(struct ath_hw *ah)
  *             to be disabled.
  *
  * 0x04000409: Packet stuck on receive.
- *             Full chip reset is required for all chips except AR9340.
+ *             Full chip reset is required for all chips except
+ *	       AR9340, AR9531 and AR9561.
  */
 
 /*
@@ -2100,7 +2101,7 @@ bool ar9003_hw_bb_watchdog_check(struct ath_hw *ah)
 	case 0x04000b09:
 		return true;
 	case 0x04000409:
-		if (AR_SREV_9340(ah) || AR_SREV_9531(ah))
+		if (AR_SREV_9340(ah) || AR_SREV_9531(ah) || AR_SREV_9561(ah))
 			return false;
 		else
 			return true;
-- 
cgit v0.10.2


From 1542bc37842d5d51e4aa71286a08be9615eba884 Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Mon, 18 Jan 2016 09:33:49 +0800
Subject: ath9k: fix inconsistent use of tab and space in indentation

Minor changes for indenting.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c
index 8b4561e..54ed2f7 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c
@@ -5485,11 +5485,11 @@ unsigned int ar9003_get_paprd_scale_factor(struct ath_hw *ah,
 			  AR9300_PAPRD_SCALE_1);
 	else {
 		if (chan->channel >= 5700)
-		return MS(le32_to_cpu(eep->modalHeader5G.papdRateMaskHt20),
-			  AR9300_PAPRD_SCALE_1);
+			return MS(le32_to_cpu(eep->modalHeader5G.papdRateMaskHt20),
+				  AR9300_PAPRD_SCALE_1);
 		else if (chan->channel >= 5400)
 			return MS(le32_to_cpu(eep->modalHeader5G.papdRateMaskHt40),
-				   AR9300_PAPRD_SCALE_2);
+				  AR9300_PAPRD_SCALE_2);
 		else
 			return MS(le32_to_cpu(eep->modalHeader5G.papdRateMaskHt40),
 				  AR9300_PAPRD_SCALE_1);
-- 
cgit v0.10.2


From 466b0f0208508b08eb59907fac0958d6da7fa3e2 Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Mon, 18 Jan 2016 09:33:50 +0800
Subject: ath9k: fix data bus error on ar9300 and ar9580

One crash issue be found on ar9300: RTC_RC reg read leads crash, leading
the data bus error, due to RTC_RC reg write not happen properly.

Warm Reset trigger in continuous beacon stuck for one of the customer for
other chip, noticed the MAC was stuck in RTC reset. After analysis noticed
DMA did not complete when RTC was put in reset.

So, before resetting the MAC need to make sure there are no pending DMA
transactions because this reset does not reset all parts of the chip.

The 12th and 11th bit of MAC _DMA_CFG register used to do that.
	12 cfg_halt_ack 0x0
		0 DMA has not yet halted
		1 DMA has halted
	11 cfg_halt_req 0x0
		0 DMA logic operates normally
		1 Request DMA logic to stop so software can reset the MAC

The Bit [12] of this register indicates when the halt has taken effect or
not. the DMA halt IS NOT recoverable; once software sets bit [11] to
request a DMA halt, software must wait for bit [12] to be set and reset
the MAC.

So, the same thing we implemented for ar9580 chip.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c
index 257f46e..e7a3101 100644
--- a/drivers/net/wireless/ath/ath9k/hw.c
+++ b/drivers/net/wireless/ath/ath9k/hw.c
@@ -1368,6 +1368,16 @@ static bool ath9k_hw_set_reset(struct ath_hw *ah, int type)
 	if (ath9k_hw_mci_is_enabled(ah))
 		ar9003_mci_check_gpm_offset(ah);
 
+	/* DMA HALT added to resolve ar9300 and ar9580 bus error during
+	 * RTC_RC reg read
+	 */
+	if (AR_SREV_9300(ah) || AR_SREV_9580(ah)) {
+		REG_SET_BIT(ah, AR_CFG, AR_CFG_HALT_REQ);
+		ath9k_hw_wait(ah, AR_CFG, AR_CFG_HALT_ACK, AR_CFG_HALT_ACK,
+			      20 * AH_WAIT_TIMEOUT);
+		REG_CLR_BIT(ah, AR_CFG, AR_CFG_HALT_REQ);
+	}
+
 	REG_WRITE(ah, AR_RTC_RC, rst_flags);
 
 	REGWRITE_BUFFER_FLUSH(ah);
diff --git a/drivers/net/wireless/ath/ath9k/reg.h b/drivers/net/wireless/ath/ath9k/reg.h
index caba54d..c8d35fe 100644
--- a/drivers/net/wireless/ath/ath9k/reg.h
+++ b/drivers/net/wireless/ath/ath9k/reg.h
@@ -34,8 +34,10 @@
 #define AR_CFG_SWRG          0x00000010
 #define AR_CFG_AP_ADHOC_INDICATION 0x00000020
 #define AR_CFG_PHOK          0x00000100
-#define AR_CFG_CLK_GATE_DIS  0x00000400
 #define AR_CFG_EEBS          0x00000200
+#define AR_CFG_CLK_GATE_DIS  0x00000400
+#define AR_CFG_HALT_REQ	     0x00000800
+#define AR_CFG_HALT_ACK	     0x00001000
 #define AR_CFG_PCI_MASTER_REQ_Q_THRESH         0x00060000
 #define AR_CFG_PCI_MASTER_REQ_Q_THRESH_S       17
 
-- 
cgit v0.10.2


From df94e702277e14889af5eef5fbfc3a33261c8b33 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 21 Jan 2016 14:13:23 +0100
Subject: ath10k: rename some HTT events

New names make a bit more sense.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt.c b/drivers/net/wireless/ath/ath10k/htt.c
index 3e6ba63..7561f22 100644
--- a/drivers/net/wireless/ath/ath10k/htt.c
+++ b/drivers/net/wireless/ath/ath10k/htt.c
@@ -131,12 +131,12 @@ static const enum htt_t2h_msg_type htt_10_4_t2h_msg_types[] = {
 	[HTT_10_4_T2H_MSG_TYPE_AGGR_CONF] = HTT_T2H_MSG_TYPE_AGGR_CONF,
 	[HTT_10_4_T2H_MSG_TYPE_TX_FETCH_IND] =
 				HTT_T2H_MSG_TYPE_TX_FETCH_IND,
-	[HTT_10_4_T2H_MSG_TYPE_TX_FETCH_CONF] =
-				HTT_T2H_MSG_TYPE_TX_FETCH_CONF,
+	[HTT_10_4_T2H_MSG_TYPE_TX_FETCH_CONFIRM] =
+				HTT_T2H_MSG_TYPE_TX_FETCH_CONFIRM,
 	[HTT_10_4_T2H_MSG_TYPE_STATS_NOUPLOAD] =
 				HTT_T2H_MSG_TYPE_STATS_NOUPLOAD,
-	[HTT_10_4_T2H_MSG_TYPE_TX_LOW_LATENCY_IND] =
-				HTT_T2H_MSG_TYPE_TX_LOW_LATENCY_IND,
+	[HTT_10_4_T2H_MSG_TYPE_TX_MODE_SWITCH_IND] =
+				HTT_T2H_MSG_TYPE_TX_MODE_SWITCH_IND,
 };
 
 int ath10k_htt_connect(struct ath10k_htt *htt)
diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index 47ca048..edd3b90 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -413,10 +413,10 @@ enum htt_10_4_t2h_msg_type {
 	HTT_10_4_T2H_MSG_TYPE_EN_STATS               = 0x14,
 	HTT_10_4_T2H_MSG_TYPE_AGGR_CONF              = 0x15,
 	HTT_10_4_T2H_MSG_TYPE_TX_FETCH_IND           = 0x16,
-	HTT_10_4_T2H_MSG_TYPE_TX_FETCH_CONF          = 0x17,
+	HTT_10_4_T2H_MSG_TYPE_TX_FETCH_CONFIRM       = 0x17,
 	HTT_10_4_T2H_MSG_TYPE_STATS_NOUPLOAD         = 0x18,
 	/* 0x19 to 0x2f are reserved */
-	HTT_10_4_T2H_MSG_TYPE_TX_LOW_LATENCY_IND     = 0x30,
+	HTT_10_4_T2H_MSG_TYPE_TX_MODE_SWITCH_IND     = 0x30,
 	/* keep this last */
 	HTT_10_4_T2H_NUM_MSGS
 };
@@ -449,8 +449,8 @@ enum htt_t2h_msg_type {
 	HTT_T2H_MSG_TYPE_TEST,
 	HTT_T2H_MSG_TYPE_EN_STATS,
 	HTT_T2H_MSG_TYPE_TX_FETCH_IND,
-	HTT_T2H_MSG_TYPE_TX_FETCH_CONF,
-	HTT_T2H_MSG_TYPE_TX_LOW_LATENCY_IND,
+	HTT_T2H_MSG_TYPE_TX_FETCH_CONFIRM,
+	HTT_T2H_MSG_TYPE_TX_MODE_SWITCH_IND,
 	/* keep this last */
 	HTT_T2H_NUM_MSGS
 };
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 3079434..84abe4f 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2125,8 +2125,8 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		break;
 	case HTT_T2H_MSG_TYPE_EN_STATS:
 	case HTT_T2H_MSG_TYPE_TX_FETCH_IND:
-	case HTT_T2H_MSG_TYPE_TX_FETCH_CONF:
-	case HTT_T2H_MSG_TYPE_TX_LOW_LATENCY_IND:
+	case HTT_T2H_MSG_TYPE_TX_FETCH_CONFIRM:
+	case HTT_T2H_MSG_TYPE_TX_MODE_SWITCH_IND:
 	default:
 		ath10k_warn(ar, "htt event (%d) not handled\n",
 			    resp->hdr.msg_type);
-- 
cgit v0.10.2


From 22e6b3bc5d9668dc711665d255efad89c527b4d6 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 21 Jan 2016 14:13:24 +0100
Subject: ath10k: add new htt definitions

These definitions are associated with some
improvements upcomming for 10.4 and QCA99X0.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index edd3b90..0c5628d 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -52,6 +52,7 @@ enum htt_h2t_msg_type { /* host-to-target */
 	/* This command is used for sending management frames in HTT < 3.0.
 	 * HTT >= 3.0 uses TX_FRM for everything. */
 	HTT_H2T_MSG_TYPE_MGMT_TX            = 7,
+	HTT_H2T_MSG_TYPE_TX_FETCH_RESP      = 11,
 
 	HTT_H2T_NUM_MSGS /* keep this last */
 };
@@ -1306,9 +1307,43 @@ struct htt_frag_desc_bank_id {
  * so we use a conservatively safe value for now */
 #define HTT_FRAG_DESC_BANK_MAX 4
 
-#define HTT_FRAG_DESC_BANK_CFG_INFO_PDEV_ID_MASK 0x03
-#define HTT_FRAG_DESC_BANK_CFG_INFO_PDEV_ID_LSB  0
-#define HTT_FRAG_DESC_BANK_CFG_INFO_SWAP         (1 << 2)
+#define HTT_FRAG_DESC_BANK_CFG_INFO_PDEV_ID_MASK		0x03
+#define HTT_FRAG_DESC_BANK_CFG_INFO_PDEV_ID_LSB			0
+#define HTT_FRAG_DESC_BANK_CFG_INFO_SWAP			BIT(2)
+#define HTT_FRAG_DESC_BANK_CFG_INFO_Q_STATE_VALID		BIT(3)
+#define HTT_FRAG_DESC_BANK_CFG_INFO_Q_STATE_DEPTH_TYPE_MASK	BIT(4)
+#define HTT_FRAG_DESC_BANK_CFG_INFO_Q_STATE_DEPTH_TYPE_LSB	4
+
+enum htt_q_depth_type {
+	HTT_Q_DEPTH_TYPE_BYTES = 0,
+	HTT_Q_DEPTH_TYPE_MSDUS = 1,
+};
+
+#define HTT_TX_Q_STATE_NUM_PEERS		(TARGET_10_4_NUM_QCACHE_PEERS_MAX + \
+						 TARGET_10_4_NUM_VDEVS)
+#define HTT_TX_Q_STATE_NUM_TIDS			8
+#define HTT_TX_Q_STATE_ENTRY_SIZE		1
+#define HTT_TX_Q_STATE_ENTRY_MULTIPLIER		0
+
+/**
+ * htt_q_state_conf - part of htt_frag_desc_bank_cfg for host q state config
+ *
+ * Defines host q state format and behavior. See htt_q_state.
+ *
+ * @record_size: Defines the size of each host q entry in bytes. In practice
+ *	however firmware (at least 10.4.3-00191) ignores this host
+ *	configuration value and uses hardcoded value of 1.
+ * @record_multiplier: This is valid only when q depth type is MSDUs. It
+ *	defines the exponent for the power of 2 multiplication.
+ */
+struct htt_q_state_conf {
+	__le32 paddr;
+	__le16 num_peers;
+	__le16 num_tids;
+	u8 record_size;
+	u8 record_multiplier;
+	u8 pad[2];
+} __packed;
 
 struct htt_frag_desc_bank_cfg {
 	u8 info; /* HTT_FRAG_DESC_BANK_CFG_INFO_ */
@@ -1316,6 +1351,114 @@ struct htt_frag_desc_bank_cfg {
 	u8 desc_size;
 	__le32 bank_base_addrs[HTT_FRAG_DESC_BANK_MAX];
 	struct htt_frag_desc_bank_id bank_id[HTT_FRAG_DESC_BANK_MAX];
+	struct htt_q_state_conf q_state;
+} __packed;
+
+#define HTT_TX_Q_STATE_ENTRY_COEFFICIENT	128
+#define HTT_TX_Q_STATE_ENTRY_FACTOR_MASK	0x3f
+#define HTT_TX_Q_STATE_ENTRY_FACTOR_LSB		0
+#define HTT_TX_Q_STATE_ENTRY_EXP_MASK		0xc0
+#define HTT_TX_Q_STATE_ENTRY_EXP_LSB		6
+
+/**
+ * htt_q_state - shared between host and firmware via DMA
+ *
+ * This structure is used for the host to expose it's software queue state to
+ * firmware so that its rate control can schedule fetch requests for optimized
+ * performance. This is most notably used for MU-MIMO aggregation when multiple
+ * MU clients are connected.
+ *
+ * @count: Each element defines the host queue depth. When q depth type was
+ *	configured as HTT_Q_DEPTH_TYPE_BYTES then each entry is defined as:
+ *	FACTOR * 128 * 8^EXP (see HTT_TX_Q_STATE_ENTRY_FACTOR_MASK and
+ *	HTT_TX_Q_STATE_ENTRY_EXP_MASK). When q depth type was configured as
+ *	HTT_Q_DEPTH_TYPE_MSDUS the number of packets is scaled by 2 **
+ *	record_multiplier (see htt_q_state_conf).
+ * @map: Used by firmware to quickly check which host queues are not empty. It
+ *	is a bitmap simply saying.
+ * @seq: Used by firmware to quickly check if the host queues were updated
+ *	since it last checked.
+ *
+ * FIXME: Is the q_state map[] size calculation really correct?
+ */
+struct htt_q_state {
+	u8 count[HTT_TX_Q_STATE_NUM_TIDS][HTT_TX_Q_STATE_NUM_PEERS];
+	u32 map[HTT_TX_Q_STATE_NUM_TIDS][(HTT_TX_Q_STATE_NUM_PEERS + 31) / 32];
+	__le32 seq;
+} __packed;
+
+#define HTT_TX_FETCH_RECORD_INFO_PEER_ID_MASK	0x0fff
+#define HTT_TX_FETCH_RECORD_INFO_PEER_ID_LSB	0
+#define HTT_TX_FETCH_RECORD_INFO_TID_MASK	0xf000
+#define HTT_TX_FETCH_RECORD_INFO_TID_LSB	12
+
+struct htt_tx_fetch_record {
+	__le16 info; /* HTT_TX_FETCH_IND_RECORD_INFO_ */
+	__le16 num_msdus;
+	__le32 num_bytes;
+} __packed;
+
+struct htt_tx_fetch_ind {
+	u8 pad0;
+	__le16 fetch_seq_num;
+	__le32 token;
+	__le16 num_resp_ids;
+	__le16 num_records;
+	struct htt_tx_fetch_record records[0];
+	__le32 resp_ids[0]; /* ath10k_htt_get_tx_fetch_ind_resp_ids() */
+} __packed;
+
+static inline void *
+ath10k_htt_get_tx_fetch_ind_resp_ids(struct htt_tx_fetch_ind *ind)
+{
+	return (void *)&ind->records[le16_to_cpu(ind->num_records)];
+}
+
+struct htt_tx_fetch_resp {
+	u8 pad0;
+	__le16 resp_id;
+	__le16 fetch_seq_num;
+	__le16 num_records;
+	__le32 token;
+	struct htt_tx_fetch_record records[0];
+} __packed;
+
+struct htt_tx_fetch_confirm {
+	u8 pad0;
+	__le16 num_resp_ids;
+	__le32 resp_ids[0];
+} __packed;
+
+enum htt_tx_mode_switch_mode {
+	HTT_TX_MODE_SWITCH_PUSH = 0,
+	HTT_TX_MODE_SWITCH_PUSH_PULL = 1,
+};
+
+#define HTT_TX_MODE_SWITCH_IND_INFO0_ENABLE		BIT(0)
+#define HTT_TX_MODE_SWITCH_IND_INFO0_NUM_RECORDS_MASK	0xfffe
+#define HTT_TX_MODE_SWITCH_IND_INFO0_NUM_RECORDS_LSB	1
+
+#define HTT_TX_MODE_SWITCH_IND_INFO1_MODE_MASK		0x0003
+#define HTT_TX_MODE_SWITCH_IND_INFO1_MODE_LSB		0
+#define HTT_TX_MODE_SWITCH_IND_INFO1_THRESHOLD_MASK	0xfffc
+#define HTT_TX_MODE_SWITCH_IND_INFO1_THRESHOLD_LSB	2
+
+#define HTT_TX_MODE_SWITCH_RECORD_INFO0_PEER_ID_MASK	0x0fff
+#define HTT_TX_MODE_SWITCH_RECORD_INFO0_PEER_ID_LSB	0
+#define HTT_TX_MODE_SWITCH_RECORD_INFO0_TID_MASK	0xf000
+#define HTT_TX_MODE_SWITCH_RECORD_INFO0_TID_LSB		12
+
+struct htt_tx_mode_switch_record {
+	__le16 info0; /* HTT_TX_MODE_SWITCH_RECORD_INFO0_ */
+	__le16 num_max_msdus;
+} __packed;
+
+struct htt_tx_mode_switch_ind {
+	u8 pad0;
+	__le16 info0; /* HTT_TX_MODE_SWITCH_IND_INFO0_ */
+	__le16 info1; /* HTT_TX_MODE_SWITCH_IND_INFO1_ */
+	u8 pad1[2];
+	struct htt_tx_mode_switch_record records[0];
 } __packed;
 
 union htt_rx_pn_t {
@@ -1340,6 +1483,7 @@ struct htt_cmd {
 		struct htt_oob_sync_req oob_sync_req;
 		struct htt_aggr_conf aggr_conf;
 		struct htt_frag_desc_bank_cfg frag_desc_bank_cfg;
+		struct htt_tx_fetch_resp tx_fetch_resp;
 	};
 } __packed;
 
@@ -1364,6 +1508,9 @@ struct htt_resp {
 		struct htt_rx_pn_ind rx_pn_ind;
 		struct htt_rx_offload_ind rx_offload_ind;
 		struct htt_rx_in_ord_ind rx_in_ord_ind;
+		struct htt_tx_fetch_ind tx_fetch_ind;
+		struct htt_tx_fetch_confirm tx_fetch_confirm;
+		struct htt_tx_mode_switch_ind tx_mode_switch_ind;
 	};
 } __packed;
 
-- 
cgit v0.10.2


From 9b783763aa9623dedeed77bec158f7377cb87750 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 21 Jan 2016 14:13:25 +0100
Subject: ath10k: add new FW_FEATURE_PEER_FLOW_CONTROL

This feature flag will be used for firmware which
supports pull-push model where host shares it's
software queue state with firmware and firmware
generates fetch requests telling host which queues
to dequeue tx from.

Primary function of this is improved MU-MIMO
performance with multiple clients.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index b41eb3f..48a2620 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -217,6 +217,7 @@ static const char *const ath10k_core_fw_feature_str[] = {
 	[ATH10K_FW_FEATURE_RAW_MODE_SUPPORT] = "raw-mode",
 	[ATH10K_FW_FEATURE_SUPPORTS_ADAPTIVE_CCA] = "adaptive-cca",
 	[ATH10K_FW_FEATURE_MFP_SUPPORT] = "mfp",
+	[ATH10K_FW_FEATURE_PEER_FLOW_CONTROL] = "peer-flow-ctrl",
 };
 
 static unsigned int ath10k_core_get_fw_feature_str(char *buf,
diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 921b86a..5067a0f 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -512,6 +512,15 @@ enum ath10k_fw_features {
 	/* Firmware supports management frame protection */
 	ATH10K_FW_FEATURE_MFP_SUPPORT = 12,
 
+	/* Firmware supports pull-push model where host shares it's software
+	 * queue state with firmware and firmware generates fetch requests
+	 * telling host which queues to dequeue tx from.
+	 *
+	 * Primary function of this is improved MU-MIMO performance with
+	 * multiple clients.
+	 */
+	ATH10K_FW_FEATURE_PEER_FLOW_CONTROL = 13,
+
 	/* keep last */
 	ATH10K_FW_FEATURE_COUNT,
 };
-- 
cgit v0.10.2


From 575fc89500d37839eb24ecd258982b24edefed6f Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 21 Jan 2016 14:13:26 +0100
Subject: ath10k: clean up cont frag desc init code

This makes the code easier to extend and re-use.

While at it fix _warn to _err. Other than that
there are no functional changes.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index b3adadb..28b8d7a 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -97,6 +97,41 @@ void ath10k_htt_tx_free_msdu_id(struct ath10k_htt *htt, u16 msdu_id)
 	idr_remove(&htt->pending_tx, msdu_id);
 }
 
+static void ath10k_htt_tx_free_cont_frag_desc(struct ath10k_htt *htt)
+{
+	size_t size;
+
+	if (!htt->frag_desc.vaddr)
+		return;
+
+	size = htt->max_num_pending_tx * sizeof(struct htt_msdu_ext_desc);
+
+	dma_free_coherent(htt->ar->dev,
+			  size,
+			  htt->frag_desc.vaddr,
+			  htt->frag_desc.paddr);
+}
+
+static int ath10k_htt_tx_alloc_cont_frag_desc(struct ath10k_htt *htt)
+{
+	struct ath10k *ar = htt->ar;
+	size_t size;
+
+	if (!ar->hw_params.continuous_frag_desc)
+		return 0;
+
+	size = htt->max_num_pending_tx * sizeof(struct htt_msdu_ext_desc);
+	htt->frag_desc.vaddr = dma_alloc_coherent(ar->dev, size,
+						  &htt->frag_desc.paddr,
+						  GFP_KERNEL);
+	if (!htt->frag_desc.vaddr) {
+		ath10k_err(ar, "failed to alloc fragment desc memory\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
 int ath10k_htt_tx_alloc(struct ath10k_htt *htt)
 {
 	struct ath10k *ar = htt->ar;
@@ -118,20 +153,12 @@ int ath10k_htt_tx_alloc(struct ath10k_htt *htt)
 		goto free_idr_pending_tx;
 	}
 
-	if (!ar->hw_params.continuous_frag_desc)
-		goto skip_frag_desc_alloc;
-
-	size = htt->max_num_pending_tx * sizeof(struct htt_msdu_ext_desc);
-	htt->frag_desc.vaddr = dma_alloc_coherent(ar->dev, size,
-						  &htt->frag_desc.paddr,
-						  GFP_KERNEL);
-	if (!htt->frag_desc.vaddr) {
-		ath10k_warn(ar, "failed to alloc fragment desc memory\n");
-		ret = -ENOMEM;
+	ret = ath10k_htt_tx_alloc_cont_frag_desc(htt);
+	if (ret) {
+		ath10k_err(ar, "failed to alloc cont frag desc: %d\n", ret);
 		goto free_txbuf;
 	}
 
-skip_frag_desc_alloc:
 	return 0;
 
 free_txbuf:
@@ -139,8 +166,10 @@ free_txbuf:
 			  sizeof(struct ath10k_htt_txbuf);
 	dma_free_coherent(htt->ar->dev, size, htt->txbuf.vaddr,
 			  htt->txbuf.paddr);
+
 free_idr_pending_tx:
 	idr_destroy(&htt->pending_tx);
+
 	return ret;
 }
 
@@ -174,12 +203,7 @@ void ath10k_htt_tx_free(struct ath10k_htt *htt)
 				  htt->txbuf.paddr);
 	}
 
-	if (htt->frag_desc.vaddr) {
-		size = htt->max_num_pending_tx *
-				  sizeof(struct htt_msdu_ext_desc);
-		dma_free_coherent(htt->ar->dev, size, htt->frag_desc.vaddr,
-				  htt->frag_desc.paddr);
-	}
+	ath10k_htt_tx_free_cont_frag_desc(htt);
 }
 
 void ath10k_htt_htc_tx_complete(struct ath10k *ar, struct sk_buff *skb)
-- 
cgit v0.10.2


From 9b15873628050fe59ebbfae200be7d50ce5e1491 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 21 Jan 2016 14:13:27 +0100
Subject: ath10k: implement basic support for new tx path firmware

This allows to use the new firmware which
implements the new tx data path. Without this
patch firmware supporting new tx path stops
responding shortly after booting.

This patch doesn't implement the entire pull-push
logic available in the new firmware. This will be
done in subsequent patches.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h
index 0c5628d..13391ea 100644
--- a/drivers/net/wireless/ath/ath10k/htt.h
+++ b/drivers/net/wireless/ath/ath10k/htt.h
@@ -1665,6 +1665,14 @@ struct ath10k_htt {
 		dma_addr_t paddr;
 		struct ath10k_htt_txbuf *vaddr;
 	} txbuf;
+
+	struct {
+		struct htt_q_state *vaddr;
+		dma_addr_t paddr;
+		u16 num_peers;
+		u16 num_tids;
+		enum htt_q_depth_type type;
+	} tx_q_state;
 };
 
 #define RX_HTT_HDR_STATUS_LEN 64
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index 84abe4f..cc957a6 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2123,10 +2123,12 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		break;
 	case HTT_T2H_MSG_TYPE_AGGR_CONF:
 		break;
-	case HTT_T2H_MSG_TYPE_EN_STATS:
 	case HTT_T2H_MSG_TYPE_TX_FETCH_IND:
 	case HTT_T2H_MSG_TYPE_TX_FETCH_CONFIRM:
 	case HTT_T2H_MSG_TYPE_TX_MODE_SWITCH_IND:
+		/* TODO: Implement pull-push logic */
+		break;
+	case HTT_T2H_MSG_TYPE_EN_STATS:
 	default:
 		ath10k_warn(ar, "htt event (%d) not handled\n",
 			    resp->hdr.msg_type);
diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c
index 28b8d7a..95acb72 100644
--- a/drivers/net/wireless/ath/ath10k/htt_tx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_tx.c
@@ -132,6 +132,50 @@ static int ath10k_htt_tx_alloc_cont_frag_desc(struct ath10k_htt *htt)
 	return 0;
 }
 
+static void ath10k_htt_tx_free_txq(struct ath10k_htt *htt)
+{
+	struct ath10k *ar = htt->ar;
+	size_t size;
+
+	if (!test_bit(ATH10K_FW_FEATURE_PEER_FLOW_CONTROL, ar->fw_features))
+		return;
+
+	size = sizeof(*htt->tx_q_state.vaddr);
+
+	dma_unmap_single(ar->dev, htt->tx_q_state.paddr, size, DMA_TO_DEVICE);
+	kfree(htt->tx_q_state.vaddr);
+}
+
+static int ath10k_htt_tx_alloc_txq(struct ath10k_htt *htt)
+{
+	struct ath10k *ar = htt->ar;
+	size_t size;
+	int ret;
+
+	if (!test_bit(ATH10K_FW_FEATURE_PEER_FLOW_CONTROL, ar->fw_features))
+		return 0;
+
+	htt->tx_q_state.num_peers = HTT_TX_Q_STATE_NUM_PEERS;
+	htt->tx_q_state.num_tids = HTT_TX_Q_STATE_NUM_TIDS;
+	htt->tx_q_state.type = HTT_Q_DEPTH_TYPE_BYTES;
+
+	size = sizeof(*htt->tx_q_state.vaddr);
+	htt->tx_q_state.vaddr = kzalloc(size, GFP_KERNEL);
+	if (!htt->tx_q_state.vaddr)
+		return -ENOMEM;
+
+	htt->tx_q_state.paddr = dma_map_single(ar->dev, htt->tx_q_state.vaddr,
+					       size, DMA_TO_DEVICE);
+	ret = dma_mapping_error(ar->dev, htt->tx_q_state.paddr);
+	if (ret) {
+		ath10k_warn(ar, "failed to dma map tx_q_state: %d\n", ret);
+		kfree(htt->tx_q_state.vaddr);
+		return -EIO;
+	}
+
+	return 0;
+}
+
 int ath10k_htt_tx_alloc(struct ath10k_htt *htt)
 {
 	struct ath10k *ar = htt->ar;
@@ -159,8 +203,17 @@ int ath10k_htt_tx_alloc(struct ath10k_htt *htt)
 		goto free_txbuf;
 	}
 
+	ret = ath10k_htt_tx_alloc_txq(htt);
+	if (ret) {
+		ath10k_err(ar, "failed to alloc txq: %d\n", ret);
+		goto free_frag_desc;
+	}
+
 	return 0;
 
+free_frag_desc:
+	ath10k_htt_tx_free_cont_frag_desc(htt);
+
 free_txbuf:
 	size = htt->max_num_pending_tx *
 			  sizeof(struct ath10k_htt_txbuf);
@@ -203,6 +256,7 @@ void ath10k_htt_tx_free(struct ath10k_htt *htt)
 				  htt->txbuf.paddr);
 	}
 
+	ath10k_htt_tx_free_txq(htt);
 	ath10k_htt_tx_free_cont_frag_desc(htt);
 }
 
@@ -292,7 +346,9 @@ int ath10k_htt_send_frag_desc_bank_cfg(struct ath10k_htt *htt)
 	struct ath10k *ar = htt->ar;
 	struct sk_buff *skb;
 	struct htt_cmd *cmd;
+	struct htt_frag_desc_bank_cfg *cfg;
 	int ret, size;
+	u8 info;
 
 	if (!ar->hw_params.continuous_frag_desc)
 		return 0;
@@ -310,14 +366,30 @@ int ath10k_htt_send_frag_desc_bank_cfg(struct ath10k_htt *htt)
 	skb_put(skb, size);
 	cmd = (struct htt_cmd *)skb->data;
 	cmd->hdr.msg_type = HTT_H2T_MSG_TYPE_FRAG_DESC_BANK_CFG;
-	cmd->frag_desc_bank_cfg.info = 0;
-	cmd->frag_desc_bank_cfg.num_banks = 1;
-	cmd->frag_desc_bank_cfg.desc_size = sizeof(struct htt_msdu_ext_desc);
-	cmd->frag_desc_bank_cfg.bank_base_addrs[0] =
-				__cpu_to_le32(htt->frag_desc.paddr);
-	cmd->frag_desc_bank_cfg.bank_id[0].bank_min_id = 0;
-	cmd->frag_desc_bank_cfg.bank_id[0].bank_max_id =
-				__cpu_to_le16(htt->max_num_pending_tx - 1);
+
+	info = 0;
+	info |= SM(htt->tx_q_state.type,
+		   HTT_FRAG_DESC_BANK_CFG_INFO_Q_STATE_DEPTH_TYPE);
+
+	if (test_bit(ATH10K_FW_FEATURE_PEER_FLOW_CONTROL, ar->fw_features))
+		info |= HTT_FRAG_DESC_BANK_CFG_INFO_Q_STATE_VALID;
+
+	cfg = &cmd->frag_desc_bank_cfg;
+	cfg->info = info;
+	cfg->num_banks = 1;
+	cfg->desc_size = sizeof(struct htt_msdu_ext_desc);
+	cfg->bank_base_addrs[0] = __cpu_to_le32(htt->frag_desc.paddr);
+	cfg->bank_id[0].bank_min_id = 0;
+	cfg->bank_id[0].bank_max_id = __cpu_to_le16(htt->max_num_pending_tx -
+						    1);
+
+	cfg->q_state.paddr = cpu_to_le32(htt->tx_q_state.paddr);
+	cfg->q_state.num_peers = cpu_to_le16(htt->tx_q_state.num_peers);
+	cfg->q_state.num_tids = cpu_to_le16(htt->tx_q_state.num_tids);
+	cfg->q_state.record_size = HTT_TX_Q_STATE_ENTRY_SIZE;
+	cfg->q_state.record_multiplier = HTT_TX_Q_STATE_ENTRY_MULTIPLIER;
+
+	ath10k_dbg(ar, ATH10K_DBG_HTT, "htt frag desc bank cmd\n");
 
 	ret = ath10k_htc_send(&htt->ar->htc, htt->eid, skb);
 	if (ret) {
-- 
cgit v0.10.2


From f52f517189dea69614bbee9ec5ab2fe366905b16 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:22 +0530
Subject: ath10k: make some of ath10k_pci_* func reusable

Some of static functions present in pci.c file are reusable
in ahb (qca4019) case. Remove static word for those reusable
functions and have those function prototype declaration in
pci.h file. So that, pci.h header file can be included in
ahb module and reused. There is no functionality changes done
in this patch.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index ee925c6..956e548 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -94,7 +94,6 @@ static const struct ath10k_pci_supp_chip ath10k_pci_supp_chips[] = {
 static void ath10k_pci_buffer_cleanup(struct ath10k *ar);
 static int ath10k_pci_cold_reset(struct ath10k *ar);
 static int ath10k_pci_safe_chip_reset(struct ath10k *ar);
-static int ath10k_pci_wait_for_target_init(struct ath10k *ar);
 static int ath10k_pci_init_irq(struct ath10k *ar);
 static int ath10k_pci_deinit_irq(struct ath10k *ar);
 static int ath10k_pci_request_irq(struct ath10k *ar);
@@ -687,7 +686,7 @@ void ath10k_pci_reg_write32(struct ath10k *ar, u32 addr, u32 val)
 	ath10k_pci_write32(ar, PCIE_LOCAL_BASE_ADDRESS + addr, val);
 }
 
-static bool ath10k_pci_irq_pending(struct ath10k *ar)
+bool ath10k_pci_irq_pending(struct ath10k *ar)
 {
 	u32 cause;
 
@@ -700,7 +699,7 @@ static bool ath10k_pci_irq_pending(struct ath10k *ar)
 	return false;
 }
 
-static void ath10k_pci_disable_and_clear_legacy_irq(struct ath10k *ar)
+void ath10k_pci_disable_and_clear_legacy_irq(struct ath10k *ar)
 {
 	/* IMPORTANT: INTR_CLR register has to be set after
 	 * INTR_ENABLE is set to 0, otherwise interrupt can not be
@@ -716,7 +715,7 @@ static void ath10k_pci_disable_and_clear_legacy_irq(struct ath10k *ar)
 				PCIE_INTR_ENABLE_ADDRESS);
 }
 
-static void ath10k_pci_enable_legacy_irq(struct ath10k *ar)
+void ath10k_pci_enable_legacy_irq(struct ath10k *ar)
 {
 	ath10k_pci_write32(ar, SOC_CORE_BASE_ADDRESS +
 			   PCIE_INTR_ENABLE_ADDRESS,
@@ -809,7 +808,7 @@ static void ath10k_pci_rx_post_pipe(struct ath10k_pci_pipe *pipe)
 	}
 }
 
-static void ath10k_pci_rx_post(struct ath10k *ar)
+void ath10k_pci_rx_post(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int i;
@@ -818,7 +817,7 @@ static void ath10k_pci_rx_post(struct ath10k *ar)
 		ath10k_pci_rx_post_pipe(&ar_pci->pipe_info[i]);
 }
 
-static void ath10k_pci_rx_replenish_retry(unsigned long ptr)
+void ath10k_pci_rx_replenish_retry(unsigned long ptr)
 {
 	struct ath10k *ar = (void *)ptr;
 
@@ -1007,8 +1006,8 @@ static int __ath10k_pci_diag_read_hi(struct ath10k *ar, void *dest,
 #define ath10k_pci_diag_read_hi(ar, dest, src, len)		\
 	__ath10k_pci_diag_read_hi(ar, dest, HI_ITEM(src), len)
 
-static int ath10k_pci_diag_write_mem(struct ath10k *ar, u32 address,
-				     const void *data, int nbytes)
+int ath10k_pci_diag_write_mem(struct ath10k *ar, u32 address,
+			      const void *data, int nbytes)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int ret = 0;
@@ -1263,8 +1262,8 @@ static void ath10k_pci_htt_rx_cb(struct ath10k_ce_pipe *ce_state)
 	ath10k_pci_process_rx_cb(ce_state, ath10k_pci_htt_rx_deliver);
 }
 
-static int ath10k_pci_hif_tx_sg(struct ath10k *ar, u8 pipe_id,
-				struct ath10k_hif_sg_item *items, int n_items)
+int ath10k_pci_hif_tx_sg(struct ath10k *ar, u8 pipe_id,
+			 struct ath10k_hif_sg_item *items, int n_items)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pci_pipe = &ar_pci->pipe_info[pipe_id];
@@ -1332,13 +1331,13 @@ err:
 	return err;
 }
 
-static int ath10k_pci_hif_diag_read(struct ath10k *ar, u32 address, void *buf,
-				    size_t buf_len)
+int ath10k_pci_hif_diag_read(struct ath10k *ar, u32 address, void *buf,
+			     size_t buf_len)
 {
 	return ath10k_pci_diag_read_mem(ar, address, buf, buf_len);
 }
 
-static u16 ath10k_pci_hif_get_free_queue_number(struct ath10k *ar, u8 pipe)
+u16 ath10k_pci_hif_get_free_queue_number(struct ath10k *ar, u8 pipe)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 
@@ -1406,8 +1405,8 @@ static void ath10k_pci_fw_crashed_dump(struct ath10k *ar)
 	queue_work(ar->workqueue, &ar->restart_work);
 }
 
-static void ath10k_pci_hif_send_complete_check(struct ath10k *ar, u8 pipe,
-					       int force)
+void ath10k_pci_hif_send_complete_check(struct ath10k *ar, u8 pipe,
+					int force)
 {
 	ath10k_dbg(ar, ATH10K_DBG_PCI, "pci hif send complete check\n");
 
@@ -1432,7 +1431,7 @@ static void ath10k_pci_hif_send_complete_check(struct ath10k *ar, u8 pipe,
 	ath10k_ce_per_engine_service(ar, pipe);
 }
 
-static void ath10k_pci_kill_tasklet(struct ath10k *ar)
+void ath10k_pci_kill_tasklet(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int i;
@@ -1446,8 +1445,8 @@ static void ath10k_pci_kill_tasklet(struct ath10k *ar)
 	del_timer_sync(&ar_pci->rx_post_retry);
 }
 
-static int ath10k_pci_hif_map_service_to_pipe(struct ath10k *ar, u16 service_id,
-					      u8 *ul_pipe, u8 *dl_pipe)
+int ath10k_pci_hif_map_service_to_pipe(struct ath10k *ar, u16 service_id,
+				       u8 *ul_pipe, u8 *dl_pipe)
 {
 	const struct service_to_pipe *entry;
 	bool ul_set = false, dl_set = false;
@@ -1491,8 +1490,8 @@ static int ath10k_pci_hif_map_service_to_pipe(struct ath10k *ar, u16 service_id,
 	return 0;
 }
 
-static void ath10k_pci_hif_get_default_pipe(struct ath10k *ar,
-					    u8 *ul_pipe, u8 *dl_pipe)
+void ath10k_pci_hif_get_default_pipe(struct ath10k *ar,
+				     u8 *ul_pipe, u8 *dl_pipe)
 {
 	ath10k_dbg(ar, ATH10K_DBG_PCI, "pci hif get default pipe\n");
 
@@ -1668,7 +1667,7 @@ static void ath10k_pci_buffer_cleanup(struct ath10k *ar)
 	}
 }
 
-static void ath10k_pci_ce_deinit(struct ath10k *ar)
+void ath10k_pci_ce_deinit(struct ath10k *ar)
 {
 	int i;
 
@@ -1676,7 +1675,7 @@ static void ath10k_pci_ce_deinit(struct ath10k *ar)
 		ath10k_ce_deinit_pipe(ar, i);
 }
 
-static void ath10k_pci_flush(struct ath10k *ar)
+void ath10k_pci_flush(struct ath10k *ar)
 {
 	ath10k_pci_kill_tasklet(ar);
 	ath10k_pci_buffer_cleanup(ar);
@@ -1711,9 +1710,9 @@ static void ath10k_pci_hif_stop(struct ath10k *ar)
 	spin_unlock_irqrestore(&ar_pci->ps_lock, flags);
 }
 
-static int ath10k_pci_hif_exchange_bmi_msg(struct ath10k *ar,
-					   void *req, u32 req_len,
-					   void *resp, u32 *resp_len)
+int ath10k_pci_hif_exchange_bmi_msg(struct ath10k *ar,
+				    void *req, u32 req_len,
+				    void *resp, u32 *resp_len)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pci_tx = &ar_pci->pipe_info[BMI_CE_NUM_TO_TARG];
@@ -1907,7 +1906,7 @@ static int ath10k_pci_get_num_banks(struct ath10k *ar)
 	return 1;
 }
 
-static int ath10k_pci_init_config(struct ath10k *ar)
+int ath10k_pci_init_config(struct ath10k *ar)
 {
 	u32 interconnect_targ_addr;
 	u32 pcie_state_targ_addr = 0;
@@ -2071,7 +2070,7 @@ static void ath10k_pci_override_ce_config(struct ath10k *ar)
 	target_service_to_ce_map_wlan[15].pipenum = __cpu_to_le32(1);
 }
 
-static int ath10k_pci_alloc_pipes(struct ath10k *ar)
+int ath10k_pci_alloc_pipes(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	struct ath10k_pci_pipe *pipe;
@@ -2102,7 +2101,7 @@ static int ath10k_pci_alloc_pipes(struct ath10k *ar)
 	return 0;
 }
 
-static void ath10k_pci_free_pipes(struct ath10k *ar)
+void ath10k_pci_free_pipes(struct ath10k *ar)
 {
 	int i;
 
@@ -2110,7 +2109,7 @@ static void ath10k_pci_free_pipes(struct ath10k *ar)
 		ath10k_ce_free_pipe(ar, i);
 }
 
-static int ath10k_pci_init_pipes(struct ath10k *ar)
+int ath10k_pci_init_pipes(struct ath10k *ar)
 {
 	int i, ret;
 
@@ -2453,7 +2452,7 @@ err_sleep:
 	return ret;
 }
 
-static void ath10k_pci_hif_power_down(struct ath10k *ar)
+void ath10k_pci_hif_power_down(struct ath10k *ar)
 {
 	ath10k_dbg(ar, ATH10K_DBG_BOOT, "boot hif power down\n");
 
@@ -2722,7 +2721,7 @@ static void ath10k_pci_free_irq(struct ath10k *ar)
 		free_irq(ar_pci->pdev->irq + i, ar);
 }
 
-static void ath10k_pci_init_irq_tasklets(struct ath10k *ar)
+void ath10k_pci_init_irq_tasklets(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int i;
@@ -2808,7 +2807,7 @@ static int ath10k_pci_deinit_irq(struct ath10k *ar)
 	return 0;
 }
 
-static int ath10k_pci_wait_for_target_init(struct ath10k *ar)
+int ath10k_pci_wait_for_target_init(struct ath10k *ar)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	unsigned long timeout;
diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h
index f91bf33..ae76131 100644
--- a/drivers/net/wireless/ath/ath10k/pci.h
+++ b/drivers/net/wireless/ath/ath10k/pci.h
@@ -253,6 +253,38 @@ u32 ath10k_pci_read32(struct ath10k *ar, u32 offset);
 u32 ath10k_pci_soc_read32(struct ath10k *ar, u32 addr);
 u32 ath10k_pci_reg_read32(struct ath10k *ar, u32 addr);
 
+int ath10k_pci_hif_tx_sg(struct ath10k *ar, u8 pipe_id,
+			 struct ath10k_hif_sg_item *items, int n_items);
+int ath10k_pci_hif_diag_read(struct ath10k *ar, u32 address, void *buf,
+			     size_t buf_len);
+int ath10k_pci_diag_write_mem(struct ath10k *ar, u32 address,
+			      const void *data, int nbytes);
+int ath10k_pci_hif_exchange_bmi_msg(struct ath10k *ar, void *req, u32 req_len,
+				    void *resp, u32 *resp_len);
+int ath10k_pci_hif_map_service_to_pipe(struct ath10k *ar, u16 service_id,
+				       u8 *ul_pipe, u8 *dl_pipe);
+void ath10k_pci_hif_get_default_pipe(struct ath10k *ar, u8 *ul_pipe,
+				     u8 *dl_pipe);
+void ath10k_pci_hif_send_complete_check(struct ath10k *ar, u8 pipe,
+					int force);
+u16 ath10k_pci_hif_get_free_queue_number(struct ath10k *ar, u8 pipe);
+void ath10k_pci_hif_power_down(struct ath10k *ar);
+int ath10k_pci_alloc_pipes(struct ath10k *ar);
+void ath10k_pci_free_pipes(struct ath10k *ar);
+void ath10k_pci_free_pipes(struct ath10k *ar);
+void ath10k_pci_rx_replenish_retry(unsigned long ptr);
+void ath10k_pci_ce_deinit(struct ath10k *ar);
+void ath10k_pci_init_irq_tasklets(struct ath10k *ar);
+void ath10k_pci_kill_tasklet(struct ath10k *ar);
+int ath10k_pci_init_pipes(struct ath10k *ar);
+int ath10k_pci_init_config(struct ath10k *ar);
+void ath10k_pci_rx_post(struct ath10k *ar);
+void ath10k_pci_flush(struct ath10k *ar);
+void ath10k_pci_enable_legacy_irq(struct ath10k *ar);
+bool ath10k_pci_irq_pending(struct ath10k *ar);
+void ath10k_pci_disable_and_clear_legacy_irq(struct ath10k *ar);
+int ath10k_pci_wait_for_target_init(struct ath10k *ar);
+
 /* QCA6174 is known to have Tx/Rx issues when SOC_WAKE register is poked too
  * frequently. To avoid this put SoC to sleep after a very conservative grace
  * period. Adjust with great care.
-- 
cgit v0.10.2


From 4ddb3299aa49ddeb40680d7427d3259d32aefa6d Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:23 +0530
Subject: ath10k: make ath10k_pci_read32/write32() ops more generic

ath10k_pci_read32/write32() does work more specific to
PCI by ensuring pci wake/sleep for every read and write.
There is a plan to use most of stuff available in pci.c
(irq stuff, copy engine, etc) for AHB case. Such kind
of pci wake/sleep for every read/write is not required
in AHB case (qca4019). All those reusable areas in pci.c
and ce.c calls ath10k_pci_read32/write32() for low level
read and write.

In fact, ath10k_pci_read32/write32() should do what it does
today for PCI case. But for AHB, it has to do differently.
To make ath10k_pci_read32/write32() more generic, new function
pointers are added in ar_pci for the function which does
operation more close to the bus. Later, corresponding bus
specific read and write function will be mapped to that.

ath10k_pci_read32/write32() are changed to call directly
those function pointers without worrying which bus underlying
to it. Also, the function to get number of bank is changed
in the same way.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 956e548..c5f6604 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -619,7 +619,7 @@ static void ath10k_pci_sleep_sync(struct ath10k *ar)
 	spin_unlock_irqrestore(&ar_pci->ps_lock, flags);
 }
 
-void ath10k_pci_write32(struct ath10k *ar, u32 offset, u32 value)
+static void ath10k_bus_pci_write32(struct ath10k *ar, u32 offset, u32 value)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	int ret;
@@ -641,7 +641,7 @@ void ath10k_pci_write32(struct ath10k *ar, u32 offset, u32 value)
 	ath10k_pci_sleep(ar);
 }
 
-u32 ath10k_pci_read32(struct ath10k *ar, u32 offset)
+static u32 ath10k_bus_pci_read32(struct ath10k *ar, u32 offset)
 {
 	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
 	u32 val;
@@ -666,6 +666,20 @@ u32 ath10k_pci_read32(struct ath10k *ar, u32 offset)
 	return val;
 }
 
+inline void ath10k_pci_write32(struct ath10k *ar, u32 offset, u32 value)
+{
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+
+	ar_pci->bus_ops->write32(ar, offset, value);
+}
+
+inline u32 ath10k_pci_read32(struct ath10k *ar, u32 offset)
+{
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+
+	return ar_pci->bus_ops->read32(ar, offset);
+}
+
 u32 ath10k_pci_soc_read32(struct ath10k *ar, u32 addr)
 {
 	return ath10k_pci_read32(ar, RTC_SOC_BASE_ADDRESS + addr);
@@ -1906,6 +1920,13 @@ static int ath10k_pci_get_num_banks(struct ath10k *ar)
 	return 1;
 }
 
+static int ath10k_bus_get_num_banks(struct ath10k *ar)
+{
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+
+	return ar_pci->bus_ops->get_num_banks(ar);
+}
+
 int ath10k_pci_init_config(struct ath10k *ar)
 {
 	u32 interconnect_targ_addr;
@@ -2017,7 +2038,7 @@ int ath10k_pci_init_config(struct ath10k *ar)
 	/* first bank is switched to IRAM */
 	ealloc_value |= ((HI_EARLY_ALLOC_MAGIC << HI_EARLY_ALLOC_MAGIC_SHIFT) &
 			 HI_EARLY_ALLOC_MAGIC_MASK);
-	ealloc_value |= ((ath10k_pci_get_num_banks(ar) <<
+	ealloc_value |= ((ath10k_bus_get_num_banks(ar) <<
 			  HI_EARLY_ALLOC_IRAM_BANKS_SHIFT) &
 			 HI_EARLY_ALLOC_IRAM_BANKS_MASK);
 
@@ -2988,6 +3009,12 @@ static bool ath10k_pci_chip_is_supported(u32 dev_id, u32 chip_id)
 	return false;
 }
 
+static const struct ath10k_bus_ops ath10k_pci_bus_ops = {
+	.read32		= ath10k_bus_pci_read32,
+	.write32	= ath10k_bus_pci_write32,
+	.get_num_banks	= ath10k_pci_get_num_banks,
+};
+
 static int ath10k_pci_probe(struct pci_dev *pdev,
 			    const struct pci_device_id *pci_dev)
 {
@@ -3038,6 +3065,7 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 	ar_pci->ar = ar;
 	ar->dev_id = pci_dev->device;
 	ar_pci->pci_ps = pci_ps;
+	ar_pci->bus_ops = &ath10k_pci_bus_ops;
 
 	ar->id.vendor = pdev->vendor;
 	ar->id.device = pdev->device;
diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h
index ae76131..41f3fac 100644
--- a/drivers/net/wireless/ath/ath10k/pci.h
+++ b/drivers/net/wireless/ath/ath10k/pci.h
@@ -157,6 +157,12 @@ struct ath10k_pci_supp_chip {
 	u32 rev_id;
 };
 
+struct ath10k_bus_ops {
+	u32 (*read32)(struct ath10k *ar, u32 offset);
+	void (*write32)(struct ath10k *ar, u32 offset, u32 value);
+	int (*get_num_banks)(struct ath10k *ar);
+};
+
 struct ath10k_pci {
 	struct pci_dev *pdev;
 	struct device *dev;
@@ -225,6 +231,8 @@ struct ath10k_pci {
 	 * on MMIO read/write.
 	 */
 	bool pci_ps;
+
+	const struct ath10k_bus_ops *bus_ops;
 };
 
 static inline struct ath10k_pci *ath10k_pci_priv(struct ath10k *ar)
-- 
cgit v0.10.2


From 90188f807f2a8bada8e165b932b56f7e03b0a9b9 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:24 +0530
Subject: ath10k: pull reusable code from pci probe and remove for ahb

Some of the code present in ath10k_pci_{probe|remove} are reusable
in ahb case too. To avoid code duplication, move reusable code to
new functions. Later, those new functions can be called from ahb
module's probe and exit functions.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index c5f6604..6ef878c 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -3009,6 +3009,37 @@ static bool ath10k_pci_chip_is_supported(u32 dev_id, u32 chip_id)
 	return false;
 }
 
+int ath10k_pci_setup_resource(struct ath10k *ar)
+{
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+	int ret;
+
+	spin_lock_init(&ar_pci->ce_lock);
+	spin_lock_init(&ar_pci->ps_lock);
+
+	setup_timer(&ar_pci->rx_post_retry, ath10k_pci_rx_replenish_retry,
+		    (unsigned long)ar);
+
+	if (QCA_REV_6174(ar))
+		ath10k_pci_override_ce_config(ar);
+
+	ret = ath10k_pci_alloc_pipes(ar);
+	if (ret) {
+		ath10k_err(ar, "failed to allocate copy engine pipes: %d\n",
+			   ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+void ath10k_pci_release_resource(struct ath10k *ar)
+{
+	ath10k_pci_kill_tasklet(ar);
+	ath10k_pci_ce_deinit(ar);
+	ath10k_pci_free_pipes(ar);
+}
+
 static const struct ath10k_bus_ops ath10k_pci_bus_ops = {
 	.read32		= ath10k_bus_pci_read32,
 	.write32	= ath10k_bus_pci_write32,
@@ -3072,34 +3103,25 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 	ar->id.subsystem_vendor = pdev->subsystem_vendor;
 	ar->id.subsystem_device = pdev->subsystem_device;
 
-	spin_lock_init(&ar_pci->ce_lock);
-	spin_lock_init(&ar_pci->ps_lock);
-
-	setup_timer(&ar_pci->rx_post_retry, ath10k_pci_rx_replenish_retry,
-		    (unsigned long)ar);
 	setup_timer(&ar_pci->ps_timer, ath10k_pci_ps_timer,
 		    (unsigned long)ar);
 
-	ret = ath10k_pci_claim(ar);
+	ret = ath10k_pci_setup_resource(ar);
 	if (ret) {
-		ath10k_err(ar, "failed to claim device: %d\n", ret);
+		ath10k_err(ar, "failed to setup resource: %d\n", ret);
 		goto err_core_destroy;
 	}
 
-	if (QCA_REV_6174(ar))
-		ath10k_pci_override_ce_config(ar);
-
-	ret = ath10k_pci_alloc_pipes(ar);
+	ret = ath10k_pci_claim(ar);
 	if (ret) {
-		ath10k_err(ar, "failed to allocate copy engine pipes: %d\n",
-			   ret);
-		goto err_sleep;
+		ath10k_err(ar, "failed to claim device: %d\n", ret);
+		goto err_free_pipes;
 	}
 
 	ret = ath10k_pci_force_wake(ar);
 	if (ret) {
 		ath10k_warn(ar, "failed to wake up device : %d\n", ret);
-		goto err_free_pipes;
+		goto err_sleep;
 	}
 
 	ath10k_pci_ce_deinit(ar);
@@ -3108,7 +3130,7 @@ static int ath10k_pci_probe(struct pci_dev *pdev,
 	ret = ath10k_pci_init_irq(ar);
 	if (ret) {
 		ath10k_err(ar, "failed to init irqs: %d\n", ret);
-		goto err_free_pipes;
+		goto err_sleep;
 	}
 
 	ath10k_info(ar, "pci irq %s interrupts %d irq_mode %d reset_mode %d\n",
@@ -3154,13 +3176,13 @@ err_free_irq:
 err_deinit_irq:
 	ath10k_pci_deinit_irq(ar);
 
-err_free_pipes:
-	ath10k_pci_free_pipes(ar);
-
 err_sleep:
 	ath10k_pci_sleep_sync(ar);
 	ath10k_pci_release(ar);
 
+err_free_pipes:
+	ath10k_pci_free_pipes(ar);
+
 err_core_destroy:
 	ath10k_core_destroy(ar);
 
@@ -3184,10 +3206,8 @@ static void ath10k_pci_remove(struct pci_dev *pdev)
 
 	ath10k_core_unregister(ar);
 	ath10k_pci_free_irq(ar);
-	ath10k_pci_kill_tasklet(ar);
 	ath10k_pci_deinit_irq(ar);
-	ath10k_pci_ce_deinit(ar);
-	ath10k_pci_free_pipes(ar);
+	ath10k_pci_release_resource(ar);
 	ath10k_pci_sleep_sync(ar);
 	ath10k_pci_release(ar);
 	ath10k_core_destroy(ar);
diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h
index 41f3fac..fcfdbca 100644
--- a/drivers/net/wireless/ath/ath10k/pci.h
+++ b/drivers/net/wireless/ath/ath10k/pci.h
@@ -292,6 +292,8 @@ void ath10k_pci_enable_legacy_irq(struct ath10k *ar);
 bool ath10k_pci_irq_pending(struct ath10k *ar);
 void ath10k_pci_disable_and_clear_legacy_irq(struct ath10k *ar);
 int ath10k_pci_wait_for_target_init(struct ath10k *ar);
+int ath10k_pci_setup_resource(struct ath10k *ar);
+void ath10k_pci_release_resource(struct ath10k *ar);
 
 /* QCA6174 is known to have Tx/Rx issues when SOC_WAKE register is poked too
  * frequently. To avoid this put SoC to sleep after a very conservative grace
-- 
cgit v0.10.2


From 0b523ced9a3cd05abf240e913b2d80e7d0fa3478 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:25 +0530
Subject: ath10k: add basic skeleton to support ahb

qca4019 uses ahb instead of pci where it slightly differs in device
enumeration, clock control, reset control, etc. Good thing is that
ahb also uses copy engine for the data transaction. So, the most of
the stuff implemented in pci.c/ce.c are reusable in ahb case too.

Device enumeration in ahb case comes through platform driver/device
model. All resource details like irq, memory map, clocks, etc for
qca4019 can be fetched from of_node of platform device.

Simply flow would look like,

 device tree => platform device (kernel) => platform driver (ath10k)

Device tree entry will have all qca4019 resource details and the same
info will be passed to kernel. Kernel will prepare new platform device
for that entry and expose DT info to of_node in platform device.
Later, ath10k would register platform driver with unique compatible name
and then kernels binds to corresponding compatible entry & calls ath10k
ahb probe functions. From there onwards, ath10k will take control of it
and move forward.

New bool flag CONFIG_ATH10K_AHB is added in Kconfig to conditionally
enable ahb support in ath10k. On enabling this flag, ath10k_pci.ko
will have ahb support. This patch adds only basic skeleton and few
macros to support ahb in the context of qca4019.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/Kconfig b/drivers/net/wireless/ath/ath10k/Kconfig
index 03aa35f..d7f207a 100644
--- a/drivers/net/wireless/ath/ath10k/Kconfig
+++ b/drivers/net/wireless/ath/ath10k/Kconfig
@@ -15,6 +15,12 @@ config ATH10K_PCI
 	---help---
 	  This module adds support for PCIE bus
 
+config ATH10K_AHB
+	bool "Atheros ath10k AHB support"
+	depends on ATH10K_PCI && OF
+	---help---
+	  This module adds support for AHB bus
+
 config ATH10K_DEBUG
 	bool "Atheros ath10k debugging"
 	depends on ATH10K
diff --git a/drivers/net/wireless/ath/ath10k/Makefile b/drivers/net/wireless/ath/ath10k/Makefile
index c04fb00..930fadd 100644
--- a/drivers/net/wireless/ath/ath10k/Makefile
+++ b/drivers/net/wireless/ath/ath10k/Makefile
@@ -25,5 +25,7 @@ obj-$(CONFIG_ATH10K_PCI) += ath10k_pci.o
 ath10k_pci-y += pci.o \
 		ce.o
 
+ath10k_pci-$(CONFIG_ATH10K_AHB) += ahb.o
+
 # for tracing framework to find trace.h
 CFLAGS_trace.o := -I$(src)
diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
new file mode 100644
index 0000000..129f4f4
--- /dev/null
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -0,0 +1,67 @@
+/*
+ * Copyright (c) 2016 Qualcomm Atheros, Inc. All rights reserved.
+ * Copyright (c) 2015 The Linux Foundation. All rights reserved.
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+#include <linux/module.h>
+#include "core.h"
+#include "debug.h"
+#include "ahb.h"
+
+static const struct of_device_id ath10k_ahb_of_match[] = {
+	/* TODO: enable this entry once everything in place.
+	 * { .compatible = "qcom,ipq4019-wifi",
+	 *   .data = (void *)ATH10K_HW_QCA4019 },
+	 */
+	{ }
+};
+
+MODULE_DEVICE_TABLE(of, ath10k_ahb_of_match);
+
+static int ath10k_ahb_probe(struct platform_device *pdev)
+{
+	return 0;
+}
+
+static int ath10k_ahb_remove(struct platform_device *pdev)
+{
+	return 0;
+}
+
+static struct platform_driver ath10k_ahb_driver = {
+	.driver         = {
+		.name   = "ath10k_ahb",
+		.of_match_table = ath10k_ahb_of_match,
+	},
+	.probe  = ath10k_ahb_probe,
+	.remove = ath10k_ahb_remove,
+};
+
+int ath10k_ahb_init(void)
+{
+	int ret;
+
+	printk(KERN_ERR "AHB support is still work in progress\n");
+
+	ret = platform_driver_register(&ath10k_ahb_driver);
+	if (ret)
+		printk(KERN_ERR "failed to register ath10k ahb driver: %d\n",
+		       ret);
+	return ret;
+}
+
+void ath10k_ahb_exit(void)
+{
+	platform_driver_unregister(&ath10k_ahb_driver);
+}
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
new file mode 100644
index 0000000..d1afe93
--- /dev/null
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -0,0 +1,41 @@
+/*
+ * Copyright (c) 2016 Qualcomm Atheros, Inc. All rights reserved.
+ * Copyright (c) 2015 The Linux Foundation. All rights reserved.
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+
+#ifndef _AHB_H_
+#define _AHB_H_
+
+#include <linux/platform_device.h>
+
+#ifdef CONFIG_ATH10K_AHB
+
+int ath10k_ahb_init(void);
+void ath10k_ahb_exit(void);
+
+#else /* CONFIG_ATH10K_AHB */
+
+static inline int ath10k_ahb_init(void)
+{
+	return 0;
+}
+
+static inline void ath10k_ahb_exit(void)
+{
+}
+
+#endif /* CONFIG_ATH10K_AHB */
+
+#endif /* _AHB_H_ */
diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 5067a0f..2e411b5 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -69,6 +69,7 @@ struct ath10k;
 
 enum ath10k_bus {
 	ATH10K_BUS_PCI,
+	ATH10K_BUS_AHB,
 };
 
 static inline const char *ath10k_bus_str(enum ath10k_bus bus)
@@ -76,6 +77,8 @@ static inline const char *ath10k_bus_str(enum ath10k_bus bus)
 	switch (bus) {
 	case ATH10K_BUS_PCI:
 		return "pci";
+	case ATH10K_BUS_AHB:
+		return "ahb";
 	}
 
 	return "unknown";
diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h
index f273478..6206edd 100644
--- a/drivers/net/wireless/ath/ath10k/debug.h
+++ b/drivers/net/wireless/ath/ath10k/debug.h
@@ -37,6 +37,7 @@ enum ath10k_debug_mask {
 	ATH10K_DBG_TESTMODE	= 0x00001000,
 	ATH10K_DBG_WMI_PRINT	= 0x00002000,
 	ATH10K_DBG_PCI_PS	= 0x00004000,
+	ATH10K_DBG_AHB		= 0x00008000,
 	ATH10K_DBG_ANY		= 0xffffffff,
 };
 
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 0678831..f885015 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -200,6 +200,7 @@ enum ath10k_hw_rev {
 	ATH10K_HW_QCA6174,
 	ATH10K_HW_QCA99X0,
 	ATH10K_HW_QCA9377,
+	ATH10K_HW_QCA4019,
 };
 
 struct ath10k_hw_regs {
@@ -253,6 +254,7 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey,
 #define QCA_REV_6174(ar) ((ar)->hw_rev == ATH10K_HW_QCA6174)
 #define QCA_REV_99X0(ar) ((ar)->hw_rev == ATH10K_HW_QCA99X0)
 #define QCA_REV_9377(ar) ((ar)->hw_rev == ATH10K_HW_QCA9377)
+#define QCA_REV_40XX(ar) ((ar)->hw_rev == ATH10K_HW_QCA4019)
 
 /* Known pecularities:
  *  - raw appears in nwifi decap, raw and nwifi appear in ethernet decap
diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 6ef878c..0e338b6 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -851,6 +851,7 @@ static u32 ath10k_pci_targ_cpu_to_ce_addr(struct ath10k *ar, u32 addr)
 		       0x7ff) << 21;
 		break;
 	case ATH10K_HW_QCA99X0:
+	case ATH10K_HW_QCA4019:
 		val = ath10k_pci_read32(ar, PCIE_BAR_REG_ADDRESS);
 		break;
 	}
@@ -1529,6 +1530,7 @@ static void ath10k_pci_irq_msi_fw_mask(struct ath10k *ar)
 				   CORE_CTRL_ADDRESS, val);
 		break;
 	case ATH10K_HW_QCA99X0:
+	case ATH10K_HW_QCA4019:
 		/* TODO: Find appropriate register configuration for QCA99X0
 		 *  to mask irq/MSI.
 		 */
@@ -1551,6 +1553,7 @@ static void ath10k_pci_irq_msi_fw_unmask(struct ath10k *ar)
 				   CORE_CTRL_ADDRESS, val);
 		break;
 	case ATH10K_HW_QCA99X0:
+	case ATH10K_HW_QCA4019:
 		/* TODO: Find appropriate register configuration for QCA99X0
 		 *  to unmask irq/MSI.
 		 */
@@ -3231,6 +3234,10 @@ static int __init ath10k_pci_init(void)
 		printk(KERN_ERR "failed to register ath10k pci driver: %d\n",
 		       ret);
 
+	ret = ath10k_ahb_init();
+	if (ret)
+		printk(KERN_ERR "ahb init failed: %d\n", ret);
+
 	return ret;
 }
 module_init(ath10k_pci_init);
@@ -3238,6 +3245,7 @@ module_init(ath10k_pci_init);
 static void __exit ath10k_pci_exit(void)
 {
 	pci_unregister_driver(&ath10k_pci_driver);
+	ath10k_ahb_exit();
 }
 
 module_exit(ath10k_pci_exit);
diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h
index fcfdbca..c2d4a79 100644
--- a/drivers/net/wireless/ath/ath10k/pci.h
+++ b/drivers/net/wireless/ath/ath10k/pci.h
@@ -22,6 +22,7 @@
 
 #include "hw.h"
 #include "ce.h"
+#include "ahb.h"
 
 /*
  * maximum number of bytes that can be handled atomically by DiagRead/DiagWrite
-- 
cgit v0.10.2


From 37a219a556b01b3236156ecf58717c1e41b552d9 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:26 +0530
Subject: ath10k: include qca4019 register map table

New register table is added for qca4019 to tell about it's
register mapping details.

Nothing much other than this.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 48a2620..1f4a278 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -1980,6 +1980,10 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev,
 		ar->regs = &qca99x0_regs;
 		ar->hw_values = &qca99x0_values;
 		break;
+	case ATH10K_HW_QCA4019:
+		ar->regs = &qca4019_regs;
+		ar->hw_values = &qca4019_values;
+		break;
 	default:
 		ath10k_err(ar, "unsupported core hardware revision %d\n",
 			   hw_rev);
diff --git a/drivers/net/wireless/ath/ath10k/hw.c b/drivers/net/wireless/ath/ath10k/hw.c
index 7b84d08..f544d48 100644
--- a/drivers/net/wireless/ath/ath10k/hw.c
+++ b/drivers/net/wireless/ath/ath10k/hw.c
@@ -109,6 +109,38 @@ const struct ath10k_hw_regs qca99x0_regs = {
 	.pcie_intr_clr_address			= 0x00000010,
 };
 
+const struct ath10k_hw_regs qca4019_regs = {
+	.rtc_soc_base_address                   = 0x00080000,
+	.soc_core_base_address                  = 0x00082000,
+	.ce_wrapper_base_address                = 0x0004d000,
+	.ce0_base_address                       = 0x0004a000,
+	.ce1_base_address                       = 0x0004a400,
+	.ce2_base_address                       = 0x0004a800,
+	.ce3_base_address                       = 0x0004ac00,
+	.ce4_base_address                       = 0x0004b000,
+	.ce5_base_address                       = 0x0004b400,
+	.ce6_base_address                       = 0x0004b800,
+	.ce7_base_address                       = 0x0004bc00,
+	/* qca4019 supports upto 12 copy engines. Since base address
+	 * of ce8 to ce11 are not directly referred in the code,
+	 * no need have them in separate members in this table.
+	 *      Copy Engine             Address
+	 *      CE8                     0x0004c000
+	 *      CE9                     0x0004c400
+	 *      CE10                    0x0004c800
+	 *      CE11                    0x0004cc00
+	 */
+	.soc_reset_control_si0_rst_mask         = 0x00000001,
+	.soc_reset_control_ce_rst_mask          = 0x00000100,
+	.soc_chip_id_address                    = 0x000000ec,
+	.fw_indicator_address                   = 0x0004f00c,
+	.ce_wrap_intr_sum_host_msi_lsb          = 0x0000000c,
+	.ce_wrap_intr_sum_host_msi_mask         = 0x00fff000,
+	.pcie_intr_fw_mask                      = 0x00100000,
+	.pcie_intr_ce_mask_all                  = 0x000fff00,
+	.pcie_intr_clr_address                  = 0x00000010,
+};
+
 const struct ath10k_hw_values qca988x_values = {
 	.rtc_state_val_on		= 3,
 	.ce_count			= 8,
@@ -136,6 +168,13 @@ const struct ath10k_hw_values qca99x0_values = {
 	.ce_desc_meta_data_lsb		= 4,
 };
 
+const struct ath10k_hw_values qca4019_values = {
+	.ce_count                       = 12,
+	.num_target_ce_config_wlan      = 10,
+	.ce_desc_meta_data_mask         = 0xFFF0,
+	.ce_desc_meta_data_lsb          = 4,
+};
+
 void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey,
 				u32 cc, u32 rcc, u32 cc_prev, u32 rcc_prev)
 {
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index f885015..f57a37b 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -233,6 +233,7 @@ struct ath10k_hw_regs {
 extern const struct ath10k_hw_regs qca988x_regs;
 extern const struct ath10k_hw_regs qca6174_regs;
 extern const struct ath10k_hw_regs qca99x0_regs;
+extern const struct ath10k_hw_regs qca4019_regs;
 
 struct ath10k_hw_values {
 	u32 rtc_state_val_on;
@@ -246,6 +247,7 @@ struct ath10k_hw_values {
 extern const struct ath10k_hw_values qca988x_values;
 extern const struct ath10k_hw_values qca6174_values;
 extern const struct ath10k_hw_values qca99x0_values;
+extern const struct ath10k_hw_values qca4019_values;
 
 void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey,
 				u32 cc, u32 rcc, u32 cc_prev, u32 rcc_prev);
-- 
cgit v0.10.2


From 7f8e79cdc2534559e3e38f5aee0b9f78e57e8fb3 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:27 +0530
Subject: ath10k: add helper functions in ahb.c for reg rd/wr

qca4019 deals with below register memory region to control the clock,
reset, etc.

        - Memory to control wifi core
        - gcc (outside of wifi)
        - tcsr (outside of wifi)

Add new helper functions to perform read/write in above registers
spaces. Actual ioremap for above registers are done in later patch.
Struct ath10k_ahb is introduced to maintain ahb specific info and
memory this struct will be allocated in the continuation of struct
ath10k_pci (again, memory ath10k_ahb is allocated in the later patch).

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index 129f4f4..eab4b47 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -17,6 +17,7 @@
 #include <linux/module.h>
 #include "core.h"
 #include "debug.h"
+#include "pci.h"
 #include "ahb.h"
 
 static const struct of_device_id ath10k_ahb_of_match[] = {
@@ -29,6 +30,60 @@ static const struct of_device_id ath10k_ahb_of_match[] = {
 
 MODULE_DEVICE_TABLE(of, ath10k_ahb_of_match);
 
+static inline struct ath10k_ahb *ath10k_ahb_priv(struct ath10k *ar)
+{
+	return &((struct ath10k_pci *)ar->drv_priv)->ahb[0];
+}
+
+static void ath10k_ahb_write32(struct ath10k *ar, u32 offset, u32 value)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	iowrite32(value, ar_ahb->mem + offset);
+}
+
+static u32 ath10k_ahb_read32(struct ath10k *ar, u32 offset)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	return ioread32(ar_ahb->mem + offset);
+}
+
+static u32 ath10k_ahb_gcc_read32(struct ath10k *ar, u32 offset)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	return ioread32(ar_ahb->gcc_mem + offset);
+}
+
+static void ath10k_ahb_tcsr_write32(struct ath10k *ar, u32 offset, u32 value)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	iowrite32(value, ar_ahb->tcsr_mem + offset);
+}
+
+static u32 ath10k_ahb_tcsr_read32(struct ath10k *ar, u32 offset)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	return ioread32(ar_ahb->tcsr_mem + offset);
+}
+
+static u32 ath10k_ahb_soc_read32(struct ath10k *ar, u32 addr)
+{
+	return ath10k_ahb_read32(ar, RTC_SOC_BASE_ADDRESS + addr);
+}
+
+static int ath10k_ahb_get_num_banks(struct ath10k *ar)
+{
+	if (ar->hw_rev == ATH10K_HW_QCA4019)
+		return 1;
+
+	ath10k_warn(ar, "unknown number of banks, assuming 1\n");
+	return 1;
+}
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index d1afe93..77d15af 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -20,6 +20,13 @@
 
 #include <linux/platform_device.h>
 
+struct ath10k_ahb {
+	struct platform_device *pdev;
+	void __iomem *mem;
+	void __iomem *gcc_mem;
+	void __iomem *tcsr_mem;
+};
+
 #ifdef CONFIG_ATH10K_AHB
 
 int ath10k_ahb_init(void);
diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h
index c2d4a79..249c73a 100644
--- a/drivers/net/wireless/ath/ath10k/pci.h
+++ b/drivers/net/wireless/ath/ath10k/pci.h
@@ -234,6 +234,12 @@ struct ath10k_pci {
 	bool pci_ps;
 
 	const struct ath10k_bus_ops *bus_ops;
+
+	/* Keep this entry in the last, memory for struct ath10k_ahb is
+	 * allocated (ahb support enabled case) in the continuation of
+	 * this struct.
+	 */
+	struct ath10k_ahb ahb[0];
 };
 
 static inline struct ath10k_pci *ath10k_pci_priv(struct ath10k *ar)
-- 
cgit v0.10.2


From 8beff219c528d19c66e9d2c40fa868d7e7c2e36f Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:28 +0530
Subject: ath10k: add clock ctrl related functions in ahb

pre qca4019 chipsets has/uses internal clock generator for
the operation. But, qca4019 uses external clocks supplied from
outside of target (ie, outside of wifi core). Three different clocks
(cmd clock, ref clock, rtc clock) comes into picture in qca4019.
All those clocks needs to configured with help of global clock
controller (gcc) to make qca4019 functioning.

Add functions for clock init/deinit, clock enable/disable in ahb.
This is just a preparation, functions added in this patch will be
used in later patches.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index eab4b47..11185c0 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -15,6 +15,7 @@
  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
 #include <linux/module.h>
+#include <linux/clk.h>
 #include "core.h"
 #include "debug.h"
 #include "pci.h"
@@ -84,6 +85,128 @@ static int ath10k_ahb_get_num_banks(struct ath10k *ar)
 	return 1;
 }
 
+static int ath10k_ahb_clock_init(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	struct device *dev;
+	int ret;
+
+	dev = &ar_ahb->pdev->dev;
+
+	ar_ahb->cmd_clk = clk_get(dev, "wifi_wcss_cmd");
+	if (IS_ERR_OR_NULL(ar_ahb->cmd_clk)) {
+		ath10k_err(ar, "failed to get cmd clk: %ld\n",
+			   PTR_ERR(ar_ahb->cmd_clk));
+		ret = ar_ahb->cmd_clk ? PTR_ERR(ar_ahb->cmd_clk) : -ENODEV;
+		goto out;
+	}
+
+	ar_ahb->ref_clk = clk_get(dev, "wifi_wcss_ref");
+	if (IS_ERR_OR_NULL(ar_ahb->ref_clk)) {
+		ath10k_err(ar, "failed to get ref clk: %ld\n",
+			   PTR_ERR(ar_ahb->ref_clk));
+		ret = ar_ahb->ref_clk ? PTR_ERR(ar_ahb->ref_clk) : -ENODEV;
+		goto err_cmd_clk_put;
+	}
+
+	ar_ahb->rtc_clk = clk_get(dev, "wifi_wcss_rtc");
+	if (IS_ERR_OR_NULL(ar_ahb->rtc_clk)) {
+		ath10k_err(ar, "failed to get rtc clk: %ld\n",
+			   PTR_ERR(ar_ahb->rtc_clk));
+		ret = ar_ahb->rtc_clk ? PTR_ERR(ar_ahb->rtc_clk) : -ENODEV;
+		goto err_ref_clk_put;
+	}
+
+	return 0;
+
+err_ref_clk_put:
+	clk_put(ar_ahb->ref_clk);
+
+err_cmd_clk_put:
+	clk_put(ar_ahb->cmd_clk);
+
+out:
+	return ret;
+}
+
+static void ath10k_ahb_clock_deinit(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->cmd_clk))
+		clk_put(ar_ahb->cmd_clk);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->ref_clk))
+		clk_put(ar_ahb->ref_clk);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->rtc_clk))
+		clk_put(ar_ahb->rtc_clk);
+
+	ar_ahb->cmd_clk = NULL;
+	ar_ahb->ref_clk = NULL;
+	ar_ahb->rtc_clk = NULL;
+}
+
+static int ath10k_ahb_clock_enable(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	struct device *dev;
+	int ret;
+
+	dev = &ar_ahb->pdev->dev;
+
+	if (IS_ERR_OR_NULL(ar_ahb->cmd_clk) ||
+	    IS_ERR_OR_NULL(ar_ahb->ref_clk) ||
+	    IS_ERR_OR_NULL(ar_ahb->rtc_clk)) {
+		ath10k_err(ar, "clock(s) is/are not initialized\n");
+		ret = -EIO;
+		goto out;
+	}
+
+	ret = clk_prepare_enable(ar_ahb->cmd_clk);
+	if (ret) {
+		ath10k_err(ar, "failed to enable cmd clk: %d\n", ret);
+		goto out;
+	}
+
+	ret = clk_prepare_enable(ar_ahb->ref_clk);
+	if (ret) {
+		ath10k_err(ar, "failed to enable ref clk: %d\n", ret);
+		goto err_cmd_clk_disable;
+	}
+
+	ret = clk_prepare_enable(ar_ahb->rtc_clk);
+	if (ret) {
+		ath10k_err(ar, "failed to enable rtc clk: %d\n", ret);
+		goto err_ref_clk_disable;
+	}
+
+	return 0;
+
+err_ref_clk_disable:
+	clk_disable_unprepare(ar_ahb->ref_clk);
+
+err_cmd_clk_disable:
+	clk_disable_unprepare(ar_ahb->cmd_clk);
+
+out:
+	return ret;
+}
+
+static void ath10k_ahb_clock_disable(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->cmd_clk))
+		clk_disable_unprepare(ar_ahb->cmd_clk);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->ref_clk))
+		clk_disable_unprepare(ar_ahb->ref_clk);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->rtc_clk))
+		clk_disable_unprepare(ar_ahb->rtc_clk);
+}
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index 77d15af..753b433 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -25,6 +25,10 @@ struct ath10k_ahb {
 	void __iomem *mem;
 	void __iomem *gcc_mem;
 	void __iomem *tcsr_mem;
+
+	struct clk *cmd_clk;
+	struct clk *ref_clk;
+	struct clk *rtc_clk;
 };
 
 #ifdef CONFIG_ATH10K_AHB
-- 
cgit v0.10.2


From 14854bfd9daaa20778db64f224a957c2abaab6fe Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:29 +0530
Subject: ath10k: add reset ctrl related functions in ahb

To perform reset on qca4019 wifi, multiple reset lines needs
to be toggled in a sequence with help of reset controller support
in the kernel. This patch adds functions to reset control init/deinit
and release reset.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/Kconfig b/drivers/net/wireless/ath/ath10k/Kconfig
index d7f207a..db1ca62 100644
--- a/drivers/net/wireless/ath/ath10k/Kconfig
+++ b/drivers/net/wireless/ath/ath10k/Kconfig
@@ -17,7 +17,7 @@ config ATH10K_PCI
 
 config ATH10K_AHB
 	bool "Atheros ath10k AHB support"
-	depends on ATH10K_PCI && OF
+	depends on ATH10K_PCI && OF && RESET_CONTROLLER
 	---help---
 	  This module adds support for AHB bus
 
diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index 11185c0..d1f1972 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -16,6 +16,7 @@
  */
 #include <linux/module.h>
 #include <linux/clk.h>
+#include <linux/reset.h>
 #include "core.h"
 #include "debug.h"
 #include "pci.h"
@@ -207,6 +208,143 @@ static void ath10k_ahb_clock_disable(struct ath10k *ar)
 		clk_disable_unprepare(ar_ahb->rtc_clk);
 }
 
+static int ath10k_ahb_rst_ctrl_init(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	struct device *dev;
+	int ret;
+
+	dev = &ar_ahb->pdev->dev;
+
+	ar_ahb->core_cold_rst = reset_control_get(dev, "wifi_core_cold");
+	if (IS_ERR_OR_NULL(ar_ahb->core_cold_rst)) {
+		ath10k_err(ar, "failed to get core cold rst ctrl: %ld\n",
+			   PTR_ERR(ar_ahb->core_cold_rst));
+		ret = ar_ahb->core_cold_rst ?
+			PTR_ERR(ar_ahb->core_cold_rst) : -ENODEV;
+		goto out;
+	}
+
+	ar_ahb->radio_cold_rst = reset_control_get(dev, "wifi_radio_cold");
+	if (IS_ERR_OR_NULL(ar_ahb->radio_cold_rst)) {
+		ath10k_err(ar, "failed to get radio cold rst ctrl: %ld\n",
+			   PTR_ERR(ar_ahb->radio_cold_rst));
+		ret = ar_ahb->radio_cold_rst ?
+			PTR_ERR(ar_ahb->radio_cold_rst) : -ENODEV;
+		goto err_core_cold_rst_put;
+	}
+
+	ar_ahb->radio_warm_rst = reset_control_get(dev, "wifi_radio_warm");
+	if (IS_ERR_OR_NULL(ar_ahb->radio_warm_rst)) {
+		ath10k_err(ar, "failed to get radio warm rst ctrl: %ld\n",
+			   PTR_ERR(ar_ahb->radio_warm_rst));
+		ret = ar_ahb->radio_warm_rst ?
+			PTR_ERR(ar_ahb->radio_warm_rst) : -ENODEV;
+		goto err_radio_cold_rst_put;
+	}
+
+	ar_ahb->radio_srif_rst = reset_control_get(dev, "wifi_radio_srif");
+	if (IS_ERR_OR_NULL(ar_ahb->radio_srif_rst)) {
+		ath10k_err(ar, "failed to get radio srif rst ctrl: %ld\n",
+			   PTR_ERR(ar_ahb->radio_srif_rst));
+		ret = ar_ahb->radio_srif_rst ?
+			PTR_ERR(ar_ahb->radio_srif_rst) : -ENODEV;
+		goto err_radio_warm_rst_put;
+	}
+
+	ar_ahb->cpu_init_rst = reset_control_get(dev, "wifi_cpu_init");
+	if (IS_ERR_OR_NULL(ar_ahb->cpu_init_rst)) {
+		ath10k_err(ar, "failed to get cpu init rst ctrl: %ld\n",
+			   PTR_ERR(ar_ahb->cpu_init_rst));
+		ret = ar_ahb->cpu_init_rst ?
+			PTR_ERR(ar_ahb->cpu_init_rst) : -ENODEV;
+		goto err_radio_srif_rst_put;
+	}
+
+	return 0;
+
+err_radio_srif_rst_put:
+	reset_control_put(ar_ahb->radio_srif_rst);
+
+err_radio_warm_rst_put:
+	reset_control_put(ar_ahb->radio_warm_rst);
+
+err_radio_cold_rst_put:
+	reset_control_put(ar_ahb->radio_cold_rst);
+
+err_core_cold_rst_put:
+	reset_control_put(ar_ahb->core_cold_rst);
+
+out:
+	return ret;
+}
+
+static void ath10k_ahb_rst_ctrl_deinit(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->core_cold_rst))
+		reset_control_put(ar_ahb->core_cold_rst);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->radio_cold_rst))
+		reset_control_put(ar_ahb->radio_cold_rst);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->radio_warm_rst))
+		reset_control_put(ar_ahb->radio_warm_rst);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->radio_srif_rst))
+		reset_control_put(ar_ahb->radio_srif_rst);
+
+	if (!IS_ERR_OR_NULL(ar_ahb->cpu_init_rst))
+		reset_control_put(ar_ahb->cpu_init_rst);
+
+	ar_ahb->core_cold_rst = NULL;
+	ar_ahb->radio_cold_rst = NULL;
+	ar_ahb->radio_warm_rst = NULL;
+	ar_ahb->radio_srif_rst = NULL;
+	ar_ahb->cpu_init_rst = NULL;
+}
+
+static int ath10k_ahb_release_reset(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	int ret;
+
+	if (IS_ERR_OR_NULL(ar_ahb->radio_cold_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->radio_warm_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->radio_srif_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->cpu_init_rst)) {
+		ath10k_err(ar, "rst ctrl(s) is/are not initialized\n");
+		return -EINVAL;
+	}
+
+	ret = reset_control_deassert(ar_ahb->radio_cold_rst);
+	if (ret) {
+		ath10k_err(ar, "failed to deassert radio cold rst: %d\n", ret);
+		return ret;
+	}
+
+	ret = reset_control_deassert(ar_ahb->radio_warm_rst);
+	if (ret) {
+		ath10k_err(ar, "failed to deassert radio warm rst: %d\n", ret);
+		return ret;
+	}
+
+	ret = reset_control_deassert(ar_ahb->radio_srif_rst);
+	if (ret) {
+		ath10k_err(ar, "failed to deassert radio srif rst: %d\n", ret);
+		return ret;
+	}
+
+	ret = reset_control_deassert(ar_ahb->cpu_init_rst);
+	if (ret) {
+		ath10k_err(ar, "failed to deassert cpu init rst: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index 753b433..2904b7b 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -29,6 +29,12 @@ struct ath10k_ahb {
 	struct clk *cmd_clk;
 	struct clk *ref_clk;
 	struct clk *rtc_clk;
+
+	struct reset_control *core_cold_rst;
+	struct reset_control *radio_cold_rst;
+	struct reset_control *radio_warm_rst;
+	struct reset_control *radio_srif_rst;
+	struct reset_control *cpu_init_rst;
 };
 
 #ifdef CONFIG_ATH10K_AHB
-- 
cgit v0.10.2


From 133df0f849bc7b5448c62272986af626f2bda6bd Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:30 +0530
Subject: ath10k: add chip and bus halt logic in ahb

Add function to perform chip halt sequence and function to halt axi
bus in ahb module. Mainly used in the scenario like driver unload.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index d1f1972..2305078 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -345,6 +345,119 @@ static int ath10k_ahb_release_reset(struct ath10k *ar)
 	return 0;
 }
 
+static void ath10k_ahb_halt_axi_bus(struct ath10k *ar, u32 haltreq_reg,
+				    u32 haltack_reg)
+{
+	unsigned long timeout;
+	u32 val;
+
+	/* Issue halt axi bus request */
+	val = ath10k_ahb_tcsr_read32(ar, haltreq_reg);
+	val |= AHB_AXI_BUS_HALT_REQ;
+	ath10k_ahb_tcsr_write32(ar, haltreq_reg, val);
+
+	/* Wait for axi bus halted ack */
+	timeout = jiffies + msecs_to_jiffies(ATH10K_AHB_AXI_BUS_HALT_TIMEOUT);
+	do {
+		val = ath10k_ahb_tcsr_read32(ar, haltack_reg);
+		if (val & AHB_AXI_BUS_HALT_ACK)
+			break;
+
+		mdelay(1);
+	} while (time_before(jiffies, timeout));
+
+	if (!(val & AHB_AXI_BUS_HALT_ACK)) {
+		ath10k_err(ar, "failed to halt axi bus: %d\n", val);
+		return;
+	}
+
+	ath10k_dbg(ar, ATH10K_DBG_AHB, "axi bus halted\n");
+}
+
+static void ath10k_ahb_halt_chip(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	u32 core_id, glb_cfg_reg, haltreq_reg, haltack_reg;
+	u32 val;
+	int ret;
+
+	if (IS_ERR_OR_NULL(ar_ahb->core_cold_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->radio_cold_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->radio_warm_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->radio_srif_rst) ||
+	    IS_ERR_OR_NULL(ar_ahb->cpu_init_rst)) {
+		ath10k_err(ar, "rst ctrl(s) is/are not initialized\n");
+		return;
+	}
+
+	core_id = ath10k_ahb_read32(ar, ATH10K_AHB_WLAN_CORE_ID_REG);
+
+	switch (core_id) {
+	case 0:
+		glb_cfg_reg = ATH10K_AHB_TCSR_WIFI0_GLB_CFG;
+		haltreq_reg = ATH10K_AHB_TCSR_WCSS0_HALTREQ;
+		haltack_reg = ATH10K_AHB_TCSR_WCSS0_HALTACK;
+		break;
+	case 1:
+		glb_cfg_reg = ATH10K_AHB_TCSR_WIFI1_GLB_CFG;
+		haltreq_reg = ATH10K_AHB_TCSR_WCSS1_HALTREQ;
+		haltack_reg = ATH10K_AHB_TCSR_WCSS1_HALTACK;
+		break;
+	default:
+		ath10k_err(ar, "invalid core id %d found, skipping reset sequence\n",
+			   core_id);
+		return;
+	}
+
+	ath10k_ahb_halt_axi_bus(ar, haltreq_reg, haltack_reg);
+
+	val = ath10k_ahb_tcsr_read32(ar, glb_cfg_reg);
+	val |= TCSR_WIFIX_GLB_CFG_DISABLE_CORE_CLK;
+	ath10k_ahb_tcsr_write32(ar, glb_cfg_reg, val);
+
+	ret = reset_control_assert(ar_ahb->core_cold_rst);
+	if (ret)
+		ath10k_err(ar, "failed to assert core cold rst: %d\n", ret);
+	msleep(1);
+
+	ret = reset_control_assert(ar_ahb->radio_cold_rst);
+	if (ret)
+		ath10k_err(ar, "failed to assert radio cold rst: %d\n", ret);
+	msleep(1);
+
+	ret = reset_control_assert(ar_ahb->radio_warm_rst);
+	if (ret)
+		ath10k_err(ar, "failed to assert radio warm rst: %d\n", ret);
+	msleep(1);
+
+	ret = reset_control_assert(ar_ahb->radio_srif_rst);
+	if (ret)
+		ath10k_err(ar, "failed to assert radio srif rst: %d\n", ret);
+	msleep(1);
+
+	ret = reset_control_assert(ar_ahb->cpu_init_rst);
+	if (ret)
+		ath10k_err(ar, "failed to assert cpu init rst: %d\n", ret);
+	msleep(10);
+
+	/* Clear halt req and core clock disable req before
+	 * deasserting wifi core reset.
+	 */
+	val = ath10k_ahb_tcsr_read32(ar, haltreq_reg);
+	val &= ~AHB_AXI_BUS_HALT_REQ;
+	ath10k_ahb_tcsr_write32(ar, haltreq_reg, val);
+
+	val = ath10k_ahb_tcsr_read32(ar, glb_cfg_reg);
+	val &= ~TCSR_WIFIX_GLB_CFG_DISABLE_CORE_CLK;
+	ath10k_ahb_tcsr_write32(ar, glb_cfg_reg, val);
+
+	ret = reset_control_deassert(ar_ahb->core_cold_rst);
+	if (ret)
+		ath10k_err(ar, "failed to deassert core cold rst: %d\n", ret);
+
+	ath10k_dbg(ar, ATH10K_DBG_AHB, "core %d reset done\n", core_id);
+}
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index 2904b7b..4761eeb 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -39,6 +39,21 @@ struct ath10k_ahb {
 
 #ifdef CONFIG_ATH10K_AHB
 
+#define ATH10K_AHB_WLAN_CORE_ID_REG          0x82030
+
+#define ATH10K_AHB_TCSR_WIFI0_GLB_CFG        0x49000
+#define ATH10K_AHB_TCSR_WIFI1_GLB_CFG        0x49004
+#define TCSR_WIFIX_GLB_CFG_DISABLE_CORE_CLK  BIT(25)
+
+#define ATH10K_AHB_TCSR_WCSS0_HALTREQ        0x52000
+#define ATH10K_AHB_TCSR_WCSS1_HALTREQ        0x52010
+#define ATH10K_AHB_TCSR_WCSS0_HALTACK        0x52004
+#define ATH10K_AHB_TCSR_WCSS1_HALTACK        0x52014
+
+#define ATH10K_AHB_AXI_BUS_HALT_TIMEOUT      10 /* msec */
+#define AHB_AXI_BUS_HALT_REQ                 1
+#define AHB_AXI_BUS_HALT_ACK                 1
+
 int ath10k_ahb_init(void);
 void ath10k_ahb_exit(void);
 
-- 
cgit v0.10.2


From 1c44fcb9234c80bbf9e4043deab8492e1c0dcf99 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:31 +0530
Subject: ath10k: include irq related functions in ahb

Add irq related functions to register,handle,release,disable interrupt.

qca4019 supports msi interrupt, but it has the problem. Until the issue
gets sorted out, only legacy interrupt model is enabled and used.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index 2305078..29d0b6c 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -458,6 +458,50 @@ static void ath10k_ahb_halt_chip(struct ath10k *ar)
 	ath10k_dbg(ar, ATH10K_DBG_AHB, "core %d reset done\n", core_id);
 }
 
+static irqreturn_t ath10k_ahb_interrupt_handler(int irq, void *arg)
+{
+	struct ath10k *ar = arg;
+	struct ath10k_pci *ar_pci = ath10k_pci_priv(ar);
+
+	if (!ath10k_pci_irq_pending(ar))
+		return IRQ_NONE;
+
+	ath10k_pci_disable_and_clear_legacy_irq(ar);
+	tasklet_schedule(&ar_pci->intr_tq);
+
+	return IRQ_HANDLED;
+}
+
+static int ath10k_ahb_request_irq_legacy(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	int ret;
+
+	ret = request_irq(ar_ahb->irq,
+			  ath10k_ahb_interrupt_handler,
+			  IRQF_SHARED, "ath10k_ahb", ar);
+	if (ret) {
+		ath10k_warn(ar, "failed to request legacy irq %d: %d\n",
+			    ar_ahb->irq, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void ath10k_ahb_release_irq_legacy(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	free_irq(ar_ahb->irq, ar);
+}
+
+static void ath10k_ahb_irq_disable(struct ath10k *ar)
+{
+	ath10k_ce_disable_interrupts(ar);
+	ath10k_pci_disable_and_clear_legacy_irq(ar);
+}
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index 4761eeb..97c40e4 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -26,6 +26,8 @@ struct ath10k_ahb {
 	void __iomem *gcc_mem;
 	void __iomem *tcsr_mem;
 
+	int irq;
+
 	struct clk *cmd_clk;
 	struct clk *ref_clk;
 	struct clk *rtc_clk;
-- 
cgit v0.10.2


From 704dc4e36769218145d30c41e81a5fbfed3977c7 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:32 +0530
Subject: ath10k: add resource init and deinit in ahb

Add function to gather resources required for qca4019 to operate
(memory, irq, dma setting, clock init , rest control init) and
function release those resources when it's not needed.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index 29d0b6c..d83a864 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -15,6 +15,8 @@
  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
 #include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <linux/clk.h>
 #include <linux/reset.h>
 #include "core.h"
@@ -502,6 +504,126 @@ static void ath10k_ahb_irq_disable(struct ath10k *ar)
 	ath10k_pci_disable_and_clear_legacy_irq(ar);
 }
 
+static int ath10k_ahb_resource_init(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	struct platform_device *pdev;
+	struct device *dev;
+	struct resource *res;
+	int ret;
+
+	pdev = ar_ahb->pdev;
+	dev = &pdev->dev;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		ath10k_err(ar, "failed to get memory resource\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	ar_ahb->mem = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(ar_ahb->mem)) {
+		ath10k_err(ar, "mem ioremap error\n");
+		ret = PTR_ERR(ar_ahb->mem);
+		goto out;
+	}
+
+	ar_ahb->mem_len = resource_size(res);
+
+	ar_ahb->gcc_mem = ioremap_nocache(ATH10K_GCC_REG_BASE,
+					  ATH10K_GCC_REG_SIZE);
+	if (!ar_ahb->gcc_mem) {
+		ath10k_err(ar, "gcc mem ioremap error\n");
+		ret = -ENOMEM;
+		goto err_mem_unmap;
+	}
+
+	ar_ahb->tcsr_mem = ioremap_nocache(ATH10K_TCSR_REG_BASE,
+					   ATH10K_TCSR_REG_SIZE);
+	if (!ar_ahb->tcsr_mem) {
+		ath10k_err(ar, "tcsr mem ioremap error\n");
+		ret = -ENOMEM;
+		goto err_gcc_mem_unmap;
+	}
+
+	ret = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		ath10k_err(ar, "failed to set 32-bit dma mask: %d\n", ret);
+		goto err_tcsr_mem_unmap;
+	}
+
+	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		ath10k_err(ar, "failed to set 32-bit consistent dma: %d\n",
+			   ret);
+		goto err_tcsr_mem_unmap;
+	}
+
+	ret = ath10k_ahb_clock_init(ar);
+	if (ret)
+		goto err_tcsr_mem_unmap;
+
+	ret = ath10k_ahb_rst_ctrl_init(ar);
+	if (ret)
+		goto err_clock_deinit;
+
+	ar_ahb->irq = platform_get_irq_byname(pdev, "legacy");
+	if (ar_ahb->irq < 0) {
+		ath10k_err(ar, "failed to get irq number: %d\n", ar_ahb->irq);
+		goto err_clock_deinit;
+	}
+
+	ath10k_dbg(ar, ATH10K_DBG_BOOT, "irq: %d\n", ar_ahb->irq);
+
+	ath10k_dbg(ar, ATH10K_DBG_BOOT, "mem: 0x%p mem_len: %lu gcc mem: 0x%p tcsr_mem: 0x%p\n",
+		   ar_ahb->mem, ar_ahb->mem_len,
+		   ar_ahb->gcc_mem, ar_ahb->tcsr_mem);
+	return 0;
+
+err_clock_deinit:
+	ath10k_ahb_clock_deinit(ar);
+
+err_tcsr_mem_unmap:
+	iounmap(ar_ahb->tcsr_mem);
+
+err_gcc_mem_unmap:
+	ar_ahb->tcsr_mem = NULL;
+	iounmap(ar_ahb->gcc_mem);
+
+err_mem_unmap:
+	ar_ahb->gcc_mem = NULL;
+	devm_iounmap(&pdev->dev, ar_ahb->mem);
+
+out:
+	ar_ahb->mem = NULL;
+	return ret;
+}
+
+static void ath10k_ahb_resource_deinit(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+	struct device *dev;
+
+	dev = &ar_ahb->pdev->dev;
+
+	if (ar_ahb->mem)
+		devm_iounmap(dev, ar_ahb->mem);
+
+	if (ar_ahb->gcc_mem)
+		iounmap(ar_ahb->gcc_mem);
+
+	if (ar_ahb->tcsr_mem)
+		iounmap(ar_ahb->tcsr_mem);
+
+	ar_ahb->mem = NULL;
+	ar_ahb->gcc_mem = NULL;
+	ar_ahb->tcsr_mem = NULL;
+
+	ath10k_ahb_clock_deinit(ar);
+	ath10k_ahb_rst_ctrl_deinit(ar);
+}
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
 	return 0;
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index 97c40e4..5bd01b4 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -23,6 +23,7 @@
 struct ath10k_ahb {
 	struct platform_device *pdev;
 	void __iomem *mem;
+	unsigned long mem_len;
 	void __iomem *gcc_mem;
 	void __iomem *tcsr_mem;
 
@@ -41,6 +42,12 @@ struct ath10k_ahb {
 
 #ifdef CONFIG_ATH10K_AHB
 
+#define ATH10K_GCC_REG_BASE                  0x1800000
+#define ATH10K_GCC_REG_SIZE                  0x60000
+
+#define ATH10K_TCSR_REG_BASE                 0x1900000
+#define ATH10K_TCSR_REG_SIZE                 0x80000
+
 #define ATH10K_AHB_WLAN_CORE_ID_REG          0x82030
 
 #define ATH10K_AHB_TCSR_WIFI0_GLB_CFG        0x49000
-- 
cgit v0.10.2


From 0d87c9208a17cb82fed66e29180078ab603086ea Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Wed, 27 Jan 2016 15:24:33 +0530
Subject: ath10k: expose hif ops for ahb

Like how pci.c exposes hif ops for the bus specific operation,
expose similar hif ops table for ahb with all required functions
linked to it. Many ath10k_pci_* functions are reused here in hif ops
table. If something is not sharable, new functions are added for ahb
and linked to hif ops table.

Finally, make ath10k_ahb_probe/remove() to perform what is expected
out of it.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index d83a864..bd62bc1 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -624,13 +624,284 @@ static void ath10k_ahb_resource_deinit(struct ath10k *ar)
 	ath10k_ahb_rst_ctrl_deinit(ar);
 }
 
+static int ath10k_ahb_prepare_device(struct ath10k *ar)
+{
+	u32 val;
+	int ret;
+
+	ret = ath10k_ahb_clock_enable(ar);
+	if (ret) {
+		ath10k_err(ar, "failed to enable clocks\n");
+		return ret;
+	}
+
+	/* Clock for the target is supplied from outside of target (ie,
+	 * external clock module controlled by the host). Target needs
+	 * to know what frequency target cpu is configured which is needed
+	 * for target internal use. Read target cpu frequency info from
+	 * gcc register and write into target's scratch register where
+	 * target expects this information.
+	 */
+	val = ath10k_ahb_gcc_read32(ar, ATH10K_AHB_GCC_FEPLL_PLL_DIV);
+	ath10k_ahb_write32(ar, ATH10K_AHB_WIFI_SCRATCH_5_REG, val);
+
+	ret = ath10k_ahb_release_reset(ar);
+	if (ret)
+		goto err_clk_disable;
+
+	ath10k_ahb_irq_disable(ar);
+
+	ath10k_ahb_write32(ar, FW_INDICATOR_ADDRESS, FW_IND_HOST_READY);
+
+	ret = ath10k_pci_wait_for_target_init(ar);
+	if (ret)
+		goto err_halt_chip;
+
+	return 0;
+
+err_halt_chip:
+	ath10k_ahb_halt_chip(ar);
+
+err_clk_disable:
+	ath10k_ahb_clock_disable(ar);
+
+	return ret;
+}
+
+static int ath10k_ahb_chip_reset(struct ath10k *ar)
+{
+	int ret;
+
+	ath10k_ahb_halt_chip(ar);
+	ath10k_ahb_clock_disable(ar);
+
+	ret = ath10k_ahb_prepare_device(ar);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int ath10k_ahb_wake_target_cpu(struct ath10k *ar)
+{
+	u32 addr, val;
+
+	addr = SOC_CORE_BASE_ADDRESS | CORE_CTRL_ADDRESS;
+	val = ath10k_ahb_read32(ar, addr);
+	val |= ATH10K_AHB_CORE_CTRL_CPU_INTR_MASK;
+	ath10k_ahb_write32(ar, addr, val);
+
+	return 0;
+}
+
+static int ath10k_ahb_hif_start(struct ath10k *ar)
+{
+	ath10k_dbg(ar, ATH10K_DBG_BOOT, "boot ahb hif start\n");
+
+	ath10k_ce_enable_interrupts(ar);
+	ath10k_pci_enable_legacy_irq(ar);
+
+	ath10k_pci_rx_post(ar);
+
+	return 0;
+}
+
+static void ath10k_ahb_hif_stop(struct ath10k *ar)
+{
+	struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar);
+
+	ath10k_dbg(ar, ATH10K_DBG_BOOT, "boot ahb hif stop\n");
+
+	ath10k_ahb_irq_disable(ar);
+	synchronize_irq(ar_ahb->irq);
+
+	ath10k_pci_flush(ar);
+}
+
+static int ath10k_ahb_hif_power_up(struct ath10k *ar)
+{
+	int ret;
+
+	ath10k_dbg(ar, ATH10K_DBG_BOOT, "boot ahb hif power up\n");
+
+	ret = ath10k_ahb_chip_reset(ar);
+	if (ret) {
+		ath10k_err(ar, "failed to reset chip: %d\n", ret);
+		goto out;
+	}
+
+	ret = ath10k_pci_init_pipes(ar);
+	if (ret) {
+		ath10k_err(ar, "failed to initialize CE: %d\n", ret);
+		goto out;
+	}
+
+	ret = ath10k_pci_init_config(ar);
+	if (ret) {
+		ath10k_err(ar, "failed to setup init config: %d\n", ret);
+		goto err_ce_deinit;
+	}
+
+	ret = ath10k_ahb_wake_target_cpu(ar);
+	if (ret) {
+		ath10k_err(ar, "could not wake up target CPU: %d\n", ret);
+		goto err_ce_deinit;
+	}
+
+	return 0;
+
+err_ce_deinit:
+	ath10k_pci_ce_deinit(ar);
+out:
+	return ret;
+}
+
+static const struct ath10k_hif_ops ath10k_ahb_hif_ops = {
+	.tx_sg                  = ath10k_pci_hif_tx_sg,
+	.diag_read              = ath10k_pci_hif_diag_read,
+	.diag_write             = ath10k_pci_diag_write_mem,
+	.exchange_bmi_msg       = ath10k_pci_hif_exchange_bmi_msg,
+	.start                  = ath10k_ahb_hif_start,
+	.stop                   = ath10k_ahb_hif_stop,
+	.map_service_to_pipe    = ath10k_pci_hif_map_service_to_pipe,
+	.get_default_pipe       = ath10k_pci_hif_get_default_pipe,
+	.send_complete_check    = ath10k_pci_hif_send_complete_check,
+	.get_free_queue_number  = ath10k_pci_hif_get_free_queue_number,
+	.power_up               = ath10k_ahb_hif_power_up,
+	.power_down             = ath10k_pci_hif_power_down,
+	.read32                 = ath10k_ahb_read32,
+	.write32                = ath10k_ahb_write32,
+};
+
+static const struct ath10k_bus_ops ath10k_ahb_bus_ops = {
+	.read32		= ath10k_ahb_read32,
+	.write32	= ath10k_ahb_write32,
+	.get_num_banks	= ath10k_ahb_get_num_banks,
+};
+
 static int ath10k_ahb_probe(struct platform_device *pdev)
 {
+	struct ath10k *ar;
+	struct ath10k_ahb *ar_ahb;
+	struct ath10k_pci *ar_pci;
+	const struct of_device_id *of_id;
+	enum ath10k_hw_rev hw_rev;
+	size_t size;
+	int ret;
+	u32 chip_id;
+
+	of_id = of_match_device(ath10k_ahb_of_match, &pdev->dev);
+	if (!of_id) {
+		dev_err(&pdev->dev, "failed to find matching device tree id\n");
+		return -EINVAL;
+	}
+
+	hw_rev = (enum ath10k_hw_rev)of_id->data;
+
+	size = sizeof(*ar_pci) + sizeof(*ar_ahb);
+	ar = ath10k_core_create(size, &pdev->dev, ATH10K_BUS_AHB,
+				hw_rev, &ath10k_ahb_hif_ops);
+	if (!ar) {
+		dev_err(&pdev->dev, "failed to allocate core\n");
+		return -ENOMEM;
+	}
+
+	ath10k_dbg(ar, ATH10K_DBG_BOOT, "ahb probe\n");
+
+	ar_pci = ath10k_pci_priv(ar);
+	ar_ahb = ath10k_ahb_priv(ar);
+
+	ar_ahb->pdev = pdev;
+	platform_set_drvdata(pdev, ar);
+
+	ret = ath10k_ahb_resource_init(ar);
+	if (ret)
+		goto err_core_destroy;
+
+	ar->dev_id = 0;
+	ar_pci->mem = ar_ahb->mem;
+	ar_pci->mem_len = ar_ahb->mem_len;
+	ar_pci->ar = ar;
+	ar_pci->bus_ops = &ath10k_ahb_bus_ops;
+
+	ret = ath10k_pci_setup_resource(ar);
+	if (ret) {
+		ath10k_err(ar, "failed to setup resource: %d\n", ret);
+		goto err_resource_deinit;
+	}
+
+	ath10k_pci_init_irq_tasklets(ar);
+
+	ret = ath10k_ahb_request_irq_legacy(ar);
+	if (ret)
+		goto err_free_pipes;
+
+	ret = ath10k_ahb_prepare_device(ar);
+	if (ret)
+		goto err_free_irq;
+
+	ath10k_pci_ce_deinit(ar);
+
+	chip_id = ath10k_ahb_soc_read32(ar, SOC_CHIP_ID_ADDRESS);
+	if (chip_id == 0xffffffff) {
+		ath10k_err(ar, "failed to get chip id\n");
+		goto err_halt_device;
+	}
+
+	ret = ath10k_core_register(ar, chip_id);
+	if (ret) {
+		ath10k_err(ar, "failed to register driver core: %d\n", ret);
+		goto err_halt_device;
+	}
+
 	return 0;
+
+err_halt_device:
+	ath10k_ahb_halt_chip(ar);
+	ath10k_ahb_clock_disable(ar);
+
+err_free_irq:
+	ath10k_ahb_release_irq_legacy(ar);
+
+err_free_pipes:
+	ath10k_pci_free_pipes(ar);
+
+err_resource_deinit:
+	ath10k_ahb_resource_deinit(ar);
+
+err_core_destroy:
+	ath10k_core_destroy(ar);
+	platform_set_drvdata(pdev, NULL);
+
+	return ret;
 }
 
 static int ath10k_ahb_remove(struct platform_device *pdev)
 {
+	struct ath10k *ar = platform_get_drvdata(pdev);
+	struct ath10k_ahb *ar_ahb;
+
+	if (!ar)
+		return -EINVAL;
+
+	ar_ahb = ath10k_ahb_priv(ar);
+
+	if (!ar_ahb)
+		return -EINVAL;
+
+	ath10k_dbg(ar, ATH10K_DBG_AHB, "ahb remove\n");
+
+	ath10k_core_unregister(ar);
+	ath10k_ahb_irq_disable(ar);
+	ath10k_ahb_release_irq_legacy(ar);
+	ath10k_pci_release_resource(ar);
+	ath10k_ahb_halt_chip(ar);
+	ath10k_ahb_clock_disable(ar);
+	ath10k_ahb_resource_deinit(ar);
+	ath10k_core_destroy(ar);
+
+	platform_set_drvdata(pdev, NULL);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ath/ath10k/ahb.h b/drivers/net/wireless/ath/ath10k/ahb.h
index 5bd01b4..d43e375 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.h
+++ b/drivers/net/wireless/ath/ath10k/ahb.h
@@ -48,6 +48,9 @@ struct ath10k_ahb {
 #define ATH10K_TCSR_REG_BASE                 0x1900000
 #define ATH10K_TCSR_REG_SIZE                 0x80000
 
+#define ATH10K_AHB_GCC_FEPLL_PLL_DIV         0x2f020
+#define ATH10K_AHB_WIFI_SCRATCH_5_REG        0x4f014
+
 #define ATH10K_AHB_WLAN_CORE_ID_REG          0x82030
 
 #define ATH10K_AHB_TCSR_WIFI0_GLB_CFG        0x49000
@@ -63,6 +66,8 @@ struct ath10k_ahb {
 #define AHB_AXI_BUS_HALT_REQ                 1
 #define AHB_AXI_BUS_HALT_ACK                 1
 
+#define ATH10K_AHB_CORE_CTRL_CPU_INTR_MASK   1
+
 int ath10k_ahb_init(void);
 void ath10k_ahb_exit(void);
 
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index f57a37b..4217bdb 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -605,6 +605,7 @@ enum ath10k_hw_4addr_pad {
 #define FW_INDICATOR_ADDRESS			ar->regs->fw_indicator_address
 #define FW_IND_EVENT_PENDING			1
 #define FW_IND_INITIALIZED			2
+#define FW_IND_HOST_READY			0x80000000
 
 /* HOST_REG interrupt from firmware */
 #define PCIE_INTR_FIRMWARE_MASK			ar->regs->pcie_intr_fw_mask
-- 
cgit v0.10.2


From 1286558e45fd90fd0faf5d41df35f511c51f25e7 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Fri, 22 Jan 2016 01:56:30 +0100
Subject: ath9k: do not limit the number of DFS channel interfaces to 1

I think this limit was added when CSA with multiple interfaces wasn't
working yet. It should no longer be necessary

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c
index ab7a1ac..d4e0ac9 100644
--- a/drivers/net/wireless/ath/ath9k/init.c
+++ b/drivers/net/wireless/ath/ath9k/init.c
@@ -751,14 +751,6 @@ static const struct ieee80211_iface_combination if_comb_multi[] = {
 
 #endif /* CONFIG_ATH9K_CHANNEL_CONTEXT */
 
-static const struct ieee80211_iface_limit if_dfs_limits[] = {
-	{ .max = 1,	.types = BIT(NL80211_IFTYPE_AP) |
-#ifdef CONFIG_MAC80211_MESH
-				 BIT(NL80211_IFTYPE_MESH_POINT) |
-#endif
-				 BIT(NL80211_IFTYPE_ADHOC) },
-};
-
 static const struct ieee80211_iface_combination if_comb[] = {
 	{
 		.limits = if_limits,
@@ -766,6 +758,11 @@ static const struct ieee80211_iface_combination if_comb[] = {
 		.max_interfaces = 2048,
 		.num_different_channels = 1,
 		.beacon_int_infra_match = true,
+#ifdef CONFIG_ATH9K_DFS_CERTIFIED
+		.radar_detect_widths =	BIT(NL80211_CHAN_WIDTH_20_NOHT) |
+					BIT(NL80211_CHAN_WIDTH_20) |
+					BIT(NL80211_CHAN_WIDTH_40),
+#endif
 	},
 	{
 		.limits = wds_limits,
@@ -774,18 +771,6 @@ static const struct ieee80211_iface_combination if_comb[] = {
 		.num_different_channels = 1,
 		.beacon_int_infra_match = true,
 	},
-#ifdef CONFIG_ATH9K_DFS_CERTIFIED
-	{
-		.limits = if_dfs_limits,
-		.n_limits = ARRAY_SIZE(if_dfs_limits),
-		.max_interfaces = 1,
-		.num_different_channels = 1,
-		.beacon_int_infra_match = true,
-		.radar_detect_widths =	BIT(NL80211_CHAN_WIDTH_20_NOHT) |
-					BIT(NL80211_CHAN_WIDTH_20) |
-					BIT(NL80211_CHAN_WIDTH_40),
-	}
-#endif
 };
 
 #ifdef CONFIG_ATH9K_CHANNEL_CONTEXT
-- 
cgit v0.10.2


From 4c1ce07bcf62536ed42a4ba43e5fb580be61ac48 Mon Sep 17 00:00:00 2001
From: Uri Mashiach <uri.mashiach@compulab.co.il>
Date: Wed, 30 Dec 2015 15:35:31 +0200
Subject: wlcore/wl12xx: spi: add power operation function

The power function uses a consumer regulator access to update the WiFi
enable GPIO value.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Tested-By: Sebastian Reichel <sre@kernel.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ti/wlcore/spi.c b/drivers/net/wireless/ti/wlcore/spi.c
index 44f059f..dec1512 100644
--- a/drivers/net/wireless/ti/wlcore/spi.c
+++ b/drivers/net/wireless/ti/wlcore/spi.c
@@ -30,6 +30,7 @@
 #include <linux/spi/spi.h>
 #include <linux/wl12xx.h>
 #include <linux/platform_device.h>
+#include <linux/regulator/consumer.h>
 
 #include "wlcore.h"
 #include "wl12xx_80211.h"
@@ -81,6 +82,7 @@
 struct wl12xx_spi_glue {
 	struct device *dev;
 	struct platform_device *core;
+	struct regulator *reg; /* Power regulator */
 };
 
 static void wl12xx_spi_reset(struct device *child)
@@ -318,11 +320,40 @@ static int __must_check wl12xx_spi_raw_write(struct device *child, int addr,
 	return 0;
 }
 
+/**
+ * wl12xx_spi_set_power - power on/off the wl12xx unit
+ * @child: wl12xx device handle.
+ * @enable: true/false to power on/off the unit.
+ *
+ * use the WiFi enable regulator to enable/disable the WiFi unit.
+ */
+static int wl12xx_spi_set_power(struct device *child, bool enable)
+{
+	int ret = 0;
+	struct wl12xx_spi_glue *glue = dev_get_drvdata(child->parent);
+
+	WARN_ON(!glue->reg);
+
+	/* Update regulator state */
+	if (enable) {
+		ret = regulator_enable(glue->reg);
+		if (ret)
+			dev_err(child, "Power enable failure\n");
+	} else {
+		ret =  regulator_disable(glue->reg);
+		if (ret)
+			dev_err(child, "Power disable failure\n");
+	}
+
+	return ret;
+}
+
 static struct wl1271_if_operations spi_ops = {
 	.read		= wl12xx_spi_raw_read,
 	.write		= wl12xx_spi_raw_write,
 	.reset		= wl12xx_spi_reset,
 	.init		= wl12xx_spi_init,
+	.power		= wl12xx_spi_set_power,
 	.set_block_size = NULL,
 };
 
@@ -353,6 +384,14 @@ static int wl1271_probe(struct spi_device *spi)
 	 * comes from the board-peripherals file */
 	spi->bits_per_word = 32;
 
+	glue->reg = devm_regulator_get(&spi->dev, "vwlan");
+	if (PTR_ERR(glue->reg) == -EPROBE_DEFER)
+		return -EPROBE_DEFER;
+	if (IS_ERR(glue->reg)) {
+		dev_err(glue->dev, "can't get regulator\n");
+		return PTR_ERR(glue->reg);
+	}
+
 	ret = spi_setup(spi);
 	if (ret < 0) {
 		dev_err(glue->dev, "spi_setup failed\n");
-- 
cgit v0.10.2


From 04654c386145239c8bcb35878b0b0537ce916766 Mon Sep 17 00:00:00 2001
From: Uri Mashiach <uri.mashiach@compulab.co.il>
Date: Wed, 30 Dec 2015 15:35:32 +0200
Subject: wlcore/wl12xx: spi: add device tree support

Add DT support for the wl1271 SPI WiFi.

Add documentation file for the wl1271 SPI WiFi.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Acked-by: Rob Herring <robh@kernel.org>
Tested-By: Sebastian Reichel <sre@kernel.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt b/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt
new file mode 100644
index 0000000..9180724
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt
@@ -0,0 +1,36 @@
+* Texas Instruments wl1271 wireless lan controller
+
+The wl1271 chip can be connected via SPI or via SDIO. This
+document describes the binding for the SPI connected chip.
+
+Required properties:
+- compatible :          Should be "ti,wl1271"
+- reg :                 Chip select address of device
+- spi-max-frequency :   Maximum SPI clocking speed of device in Hz
+- ref-clock-frequency : Reference clock frequency
+- interrupt-parent, interrupts :
+                        Should contain parameters for 1 interrupt line.
+                        Interrupt parameters: parent, line number, type.
+- vwlan-supply :        Point the node of the regulator that powers/enable the wl1271 chip
+
+Optional properties:
+- clock-xtal :          boolean, clock is generated from XTAL
+
+- Please consult Documentation/devicetree/bindings/spi/spi-bus.txt
+  for optional SPI connection related properties,
+
+Examples:
+
+&spi1 {
+	wl1271@1 {
+		compatible = "ti,wl1271";
+
+		reg = <1>;
+		spi-max-frequency = <48000000>;
+		clock-xtal;
+		ref-clock-frequency = <38400000>;
+		interrupt-parent = <&gpio3>;
+		interrupts = <8 IRQ_TYPE_LEVEL_HIGH>;
+		vwlan-supply = <&vwlan_fixed>;
+	};
+};
diff --git a/drivers/net/wireless/ti/wlcore/Kconfig b/drivers/net/wireless/ti/wlcore/Kconfig
index 969c9d7..8a8f1e7 100644
--- a/drivers/net/wireless/ti/wlcore/Kconfig
+++ b/drivers/net/wireless/ti/wlcore/Kconfig
@@ -13,7 +13,7 @@ config WLCORE
 
 config WLCORE_SPI
 	tristate "TI wlcore SPI support"
-	depends on WLCORE && SPI_MASTER
+	depends on WLCORE && SPI_MASTER && OF
 	select CRC7
 	---help---
 	  This module adds support for the SPI interface of adapters using
diff --git a/drivers/net/wireless/ti/wlcore/spi.c b/drivers/net/wireless/ti/wlcore/spi.c
index dec1512..020ac1a 100644
--- a/drivers/net/wireless/ti/wlcore/spi.c
+++ b/drivers/net/wireless/ti/wlcore/spi.c
@@ -30,6 +30,7 @@
 #include <linux/spi/spi.h>
 #include <linux/wl12xx.h>
 #include <linux/platform_device.h>
+#include <linux/of_irq.h>
 #include <linux/regulator/consumer.h>
 
 #include "wlcore.h"
@@ -357,6 +358,39 @@ static struct wl1271_if_operations spi_ops = {
 	.set_block_size = NULL,
 };
 
+static const struct of_device_id wlcore_spi_of_match_table[] = {
+	{ .compatible = "ti,wl1271" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, wlcore_spi_of_match_table);
+
+/**
+ * wlcore_probe_of - DT node parsing.
+ * @spi: SPI slave device parameters.
+ * @res: resource parameters.
+ * @glue: wl12xx SPI bus to slave device glue parameters.
+ * @pdev_data: wlcore device parameters
+ */
+static int wlcore_probe_of(struct spi_device *spi, struct wl12xx_spi_glue *glue,
+			   struct wlcore_platdev_data *pdev_data)
+{
+	struct device_node *dt_node = spi->dev.of_node;
+	int ret;
+
+	if (of_find_property(dt_node, "clock-xtal", NULL))
+		pdev_data->ref_clock_xtal = true;
+
+	ret = of_property_read_u32(dt_node, "ref-clock-frequency",
+				   &pdev_data->ref_clock_freq);
+	if (IS_ERR_VALUE(ret)) {
+		dev_err(glue->dev,
+			"can't get reference clock frequency (%d)\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int wl1271_probe(struct spi_device *spi)
 {
 	struct wl12xx_spi_glue *glue;
@@ -366,8 +400,6 @@ static int wl1271_probe(struct spi_device *spi)
 
 	memset(&pdev_data, 0x00, sizeof(pdev_data));
 
-	/* TODO: add DT parsing when needed */
-
 	pdev_data.if_ops = &spi_ops;
 
 	glue = devm_kzalloc(&spi->dev, sizeof(*glue), GFP_KERNEL);
@@ -392,6 +424,13 @@ static int wl1271_probe(struct spi_device *spi)
 		return PTR_ERR(glue->reg);
 	}
 
+	ret = wlcore_probe_of(spi, glue, &pdev_data);
+	if (IS_ERR_VALUE(ret)) {
+		dev_err(glue->dev,
+			"can't get device tree parameters (%d)\n", ret);
+		return ret;
+	}
+
 	ret = spi_setup(spi);
 	if (ret < 0) {
 		dev_err(glue->dev, "spi_setup failed\n");
@@ -409,7 +448,7 @@ static int wl1271_probe(struct spi_device *spi)
 	memset(res, 0x00, sizeof(res));
 
 	res[0].start = spi->irq;
-	res[0].flags = IORESOURCE_IRQ;
+	res[0].flags = IORESOURCE_IRQ | irq_get_trigger_type(spi->irq);
 	res[0].name = "irq";
 
 	ret = platform_device_add_resources(glue->core, res, ARRAY_SIZE(res));
@@ -447,10 +486,10 @@ static int wl1271_remove(struct spi_device *spi)
 	return 0;
 }
 
-
 static struct spi_driver wl1271_spi_driver = {
 	.driver = {
 		.name		= "wl1271_spi",
+		.of_match_table = of_match_ptr(wlcore_spi_of_match_table),
 	},
 
 	.probe		= wl1271_probe,
-- 
cgit v0.10.2


From 2fd5c6ed0b4fbf3c2d9314a24c82862ce5254d42 Mon Sep 17 00:00:00 2001
From: chunfan chen <jeffc@marvell.com>
Date: Wed, 6 Jan 2016 23:40:47 -0800
Subject: mwifiex: firmware download enhancements

Same chip is being used by WLAN as well as bluetooth
drivers. Each driver needs to check during initialisation
if firmware is already active or it needs to be freshly
downloaded. If one driver has started downloading the
firmware, other finds the winner flag as false.

mwifiex_check_fw_status() checks firmware status and also
check if WLAN is the winner for firmware downloading.

Once we detect that other interface is downloading
the firmware, we call this routine again with max
poll count to wait until firmware is ready.

This patch splits the routine to avoid checking
winner status unnecessarily multiple times and ensures
that correct messages are displayed to user.

Firmware status poll count is also increased to 150.

Signed-off-by: Chunfan Chen <jeffc@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h
index ced7af2..426e76a 100644
--- a/drivers/net/wireless/marvell/mwifiex/fw.h
+++ b/drivers/net/wireless/marvell/mwifiex/fw.h
@@ -96,7 +96,7 @@ enum KEY_TYPE_ID {
 #define WAPI_KEY_LEN			(WLAN_KEY_LEN_SMS4 + PN_LEN + 2)
 
 #define MAX_POLL_TRIES			100
-#define MAX_FIRMWARE_POLL_TRIES			100
+#define MAX_FIRMWARE_POLL_TRIES			150
 
 #define FIRMWARE_READY_SDIO				0xfedc
 #define FIRMWARE_READY_PCIE				0xfedcba00
diff --git a/drivers/net/wireless/marvell/mwifiex/init.c b/drivers/net/wireless/marvell/mwifiex/init.c
index 6f7876e..517653b 100644
--- a/drivers/net/wireless/marvell/mwifiex/init.c
+++ b/drivers/net/wireless/marvell/mwifiex/init.c
@@ -741,8 +741,6 @@ int mwifiex_dnld_fw(struct mwifiex_adapter *adapter,
 	u32 poll_num = 1;
 
 	if (adapter->if_ops.check_fw_status) {
-		adapter->winner = 0;
-
 		/* check if firmware is already running */
 		ret = adapter->if_ops.check_fw_status(adapter, poll_num);
 		if (!ret) {
@@ -750,13 +748,23 @@ int mwifiex_dnld_fw(struct mwifiex_adapter *adapter,
 				    "WLAN FW already running! Skip FW dnld\n");
 			return 0;
 		}
+	}
+
+	/* check if we are the winner for downloading FW */
+	if (adapter->if_ops.check_winner_status) {
+		adapter->winner = 0;
+		ret = adapter->if_ops.check_winner_status(adapter);
 
 		poll_num = MAX_FIRMWARE_POLL_TRIES;
+		if (ret) {
+			mwifiex_dbg(adapter, MSG,
+				    "WLAN read winner status failed!\n");
+			return ret;
+		}
 
-		/* check if we are the winner for downloading FW */
 		if (!adapter->winner) {
 			mwifiex_dbg(adapter, MSG,
-				    "FW already running! Skip FW dnld\n");
+				    "WLAN is not the winner! Skip FW dnld\n");
 			goto poll_fw;
 		}
 	}
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 2f7f478..c08be79 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -791,6 +791,7 @@ struct mwifiex_if_ops {
 	int (*init_if) (struct mwifiex_adapter *);
 	void (*cleanup_if) (struct mwifiex_adapter *);
 	int (*check_fw_status) (struct mwifiex_adapter *, u32);
+	int (*check_winner_status)(struct mwifiex_adapter *);
 	int (*prog_fw) (struct mwifiex_adapter *, struct mwifiex_fw_image *);
 	int (*register_dev) (struct mwifiex_adapter *);
 	void (*unregister_dev) (struct mwifiex_adapter *);
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index 6d0dc40..f448d7c 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2007,14 +2007,12 @@ done:
 
 /*
  * This function checks the firmware status in card.
- *
- * The winner interface is also determined by this function.
  */
 static int
 mwifiex_check_fw_status(struct mwifiex_adapter *adapter, u32 poll_num)
 {
 	int ret = 0;
-	u32 firmware_stat, winner_status;
+	u32 firmware_stat;
 	struct pcie_service_card *card = adapter->card;
 	const struct mwifiex_pcie_card_reg *reg = card->pcie.reg;
 	u32 tries;
@@ -2054,19 +2052,28 @@ mwifiex_check_fw_status(struct mwifiex_adapter *adapter, u32 poll_num)
 		}
 	}
 
-	if (ret) {
-		if (mwifiex_read_reg(adapter, reg->fw_status,
-				     &winner_status))
-			ret = -1;
-		else if (!winner_status) {
-			mwifiex_dbg(adapter, INFO,
-				    "PCI-E is the winner\n");
-			adapter->winner = 1;
-		} else {
-			mwifiex_dbg(adapter, ERROR,
-				    "PCI-E is not the winner <%#x,%d>, exit dnld\n",
-				    ret, adapter->winner);
-		}
+	return ret;
+}
+
+/* This function checks if WLAN is the winner.
+ */
+static int
+mwifiex_check_winner_status(struct mwifiex_adapter *adapter)
+{
+	u32 winner = 0;
+	int ret = 0;
+	struct pcie_service_card *card = adapter->card;
+	const struct mwifiex_pcie_card_reg *reg = card->pcie.reg;
+
+	if (mwifiex_read_reg(adapter, reg->fw_status, &winner)) {
+		ret = -1;
+	} else if (!winner) {
+		mwifiex_dbg(adapter, INFO, "PCI-E is the winner\n");
+		adapter->winner = 1;
+	} else {
+		mwifiex_dbg(adapter, ERROR,
+			    "PCI-E is not the winner <%#x,%d>, exit dnld\n",
+			    ret, adapter->winner);
 	}
 
 	return ret;
@@ -2675,6 +2682,7 @@ static struct mwifiex_if_ops pcie_ops = {
 	.init_if =			mwifiex_pcie_init,
 	.cleanup_if =			mwifiex_pcie_cleanup,
 	.check_fw_status =		mwifiex_check_fw_status,
+	.check_winner_status =          mwifiex_check_winner_status,
 	.prog_fw =			mwifiex_prog_fw_w_helper,
 	.register_dev =			mwifiex_register_dev,
 	.unregister_dev =		mwifiex_unregister_dev,
diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c
index 4c8cae6..33771d3 100644
--- a/drivers/net/wireless/marvell/mwifiex/sdio.c
+++ b/drivers/net/wireless/marvell/mwifiex/sdio.c
@@ -1039,19 +1039,14 @@ done:
 
 /*
  * This function checks the firmware status in card.
- *
- * The winner interface is also determined by this function.
  */
 static int mwifiex_check_fw_status(struct mwifiex_adapter *adapter,
 				   u32 poll_num)
 {
-	struct sdio_mmc_card *card = adapter->card;
 	int ret = 0;
 	u16 firmware_stat;
 	u32 tries;
-	u8 winner_status;
 
-	/* Wait for firmware initialization event */
 	for (tries = 0; tries < poll_num; tries++) {
 		ret = mwifiex_sdio_read_fw_status(adapter, &firmware_stat);
 		if (ret)
@@ -1065,16 +1060,25 @@ static int mwifiex_check_fw_status(struct mwifiex_adapter *adapter,
 		}
 	}
 
-	if (ret) {
-		if (mwifiex_read_reg
-		    (adapter, card->reg->status_reg_0, &winner_status))
-			winner_status = 0;
+	return ret;
+}
+
+/* This function checks if WLAN is the winner.
+ */
+static int mwifiex_check_winner_status(struct mwifiex_adapter *adapter)
+{
+	int ret = 0;
+	u8 winner = 0;
+	struct sdio_mmc_card *card = adapter->card;
+
+	if (mwifiex_read_reg(adapter, card->reg->status_reg_0, &winner))
+		return -1;
+
+	if (winner)
+		adapter->winner = 0;
+	else
+		adapter->winner = 1;
 
-		if (winner_status)
-			adapter->winner = 0;
-		else
-			adapter->winner = 1;
-	}
 	return ret;
 }
 
@@ -2620,6 +2624,7 @@ static struct mwifiex_if_ops sdio_ops = {
 	.init_if = mwifiex_init_sdio,
 	.cleanup_if = mwifiex_cleanup_sdio,
 	.check_fw_status = mwifiex_check_fw_status,
+	.check_winner_status = mwifiex_check_winner_status,
 	.prog_fw = mwifiex_prog_fw_w_helper,
 	.register_dev = mwifiex_register_dev,
 	.unregister_dev = mwifiex_unregister_dev,
-- 
cgit v0.10.2


From dc386ce76dedaeeaaf006fceb6ed8cf2e20ff026 Mon Sep 17 00:00:00 2001
From: chunfan chen <jeffc@marvell.com>
Date: Wed, 6 Jan 2016 23:40:48 -0800
Subject: mwifiex: fix IBSS data path issue.

The port_open flag is not applicable for IBSS mode. IBSS data
path was broken when port_open flag was introduced.
This patch fixes the problem by correcting the checks.

Fixes: 5c8946330abfa4c ("mwifiex: enable traffic only when port is open")
Signed-off-by: chunfan chen <jeffc@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/sta_event.c b/drivers/net/wireless/marvell/mwifiex/sta_event.c
index ff3ee9d..23bae87 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_event.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_event.c
@@ -607,11 +607,13 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv)
 
 	case EVENT_PS_AWAKE:
 		mwifiex_dbg(adapter, EVENT, "info: EVENT: AWAKE\n");
-		if (!adapter->pps_uapsd_mode && priv->port_open &&
+		if (!adapter->pps_uapsd_mode &&
+		    (priv->port_open ||
+		     (priv->bss_mode == NL80211_IFTYPE_ADHOC)) &&
 		    priv->media_connected && adapter->sleep_period.period) {
-				adapter->pps_uapsd_mode = true;
-				mwifiex_dbg(adapter, EVENT,
-					    "event: PPS/UAPSD mode activated\n");
+			adapter->pps_uapsd_mode = true;
+			mwifiex_dbg(adapter, EVENT,
+				    "event: PPS/UAPSD mode activated\n");
 		}
 		adapter->tx_lock_flag = false;
 		if (adapter->pps_uapsd_mode && adapter->gen_null_pkt) {
diff --git a/drivers/net/wireless/marvell/mwifiex/wmm.c b/drivers/net/wireless/marvell/mwifiex/wmm.c
index acccd67..499e5a7 100644
--- a/drivers/net/wireless/marvell/mwifiex/wmm.c
+++ b/drivers/net/wireless/marvell/mwifiex/wmm.c
@@ -475,7 +475,8 @@ mwifiex_wmm_lists_empty(struct mwifiex_adapter *adapter)
 		priv = adapter->priv[i];
 		if (!priv)
 			continue;
-		if (!priv->port_open)
+		if (!priv->port_open &&
+		    (priv->bss_mode != NL80211_IFTYPE_ADHOC))
 			continue;
 		if (adapter->if_ops.is_port_ready &&
 		    !adapter->if_ops.is_port_ready(priv))
@@ -1099,7 +1100,8 @@ mwifiex_wmm_get_highest_priolist_ptr(struct mwifiex_adapter *adapter,
 
 			priv_tmp = adapter->bss_prio_tbl[j].bss_prio_cur->priv;
 
-			if (!priv_tmp->port_open ||
+			if (((priv_tmp->bss_mode != NL80211_IFTYPE_ADHOC) &&
+			     !priv_tmp->port_open) ||
 			    (atomic_read(&priv_tmp->wmm.tx_pkts_queued) == 0))
 				continue;
 
-- 
cgit v0.10.2


From 17934b6a32b2fdfcc0a9e83d17c780f15627aa30 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Wed, 6 Jan 2016 23:40:49 -0800
Subject: mwifiex: add debugfs file to read chip information

This patch add 'verext' debugfs item, which can be used to
get detailed chip specific information from our firmware.

Examples:
echo "1" > /sys/kernel/debug/mwifiex/mlan0/verext
cat /sys/kernel/debug/mwifiex/mlan0/verext

Signed-off-by: Shengzhen Li <szli@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/README b/drivers/net/wireless/marvell/mwifiex/README
index 2f0f9b5..24e649b 100644
--- a/drivers/net/wireless/marvell/mwifiex/README
+++ b/drivers/net/wireless/marvell/mwifiex/README
@@ -237,4 +237,14 @@ device_dump
 
 	cat fw_dump
 
+verext
+	This command is used to get extended firmware version string using
+	different configuration parameters.
+
+	Usage:
+		echo "[version_str_sel]" > verext
+		cat verext
+
+		[version_str_sel]: firmware support several extend version
+				   string cases, include 0/1/10/20/21/99
 ===============================================================================
diff --git a/drivers/net/wireless/marvell/mwifiex/debugfs.c b/drivers/net/wireless/marvell/mwifiex/debugfs.c
index 0b9c580..df28836 100644
--- a/drivers/net/wireless/marvell/mwifiex/debugfs.c
+++ b/drivers/net/wireless/marvell/mwifiex/debugfs.c
@@ -95,8 +95,7 @@ mwifiex_info_read(struct file *file, char __user *ubuf,
 
 	mwifiex_drv_get_driver_version(priv->adapter, fmt, sizeof(fmt) - 1);
 
-	if (!priv->version_str[0])
-		mwifiex_get_ver_ext(priv);
+	mwifiex_get_ver_ext(priv, 0);
 
 	p += sprintf(p, "driver_name = " "\"mwifiex\"\n");
 	p += sprintf(p, "driver_version = %s", fmt);
@@ -583,6 +582,52 @@ done:
 	return ret;
 }
 
+/* debugfs verext file write handler.
+ * This function is called when the 'verext' file is opened for write
+ */
+static ssize_t
+mwifiex_verext_write(struct file *file, const char __user *ubuf,
+		     size_t count, loff_t *ppos)
+{
+	int ret;
+	u32 versionstrsel;
+	struct mwifiex_private *priv = (void *)file->private_data;
+	char buf[16];
+
+	memset(buf, 0, sizeof(buf));
+
+	if (copy_from_user(&buf, ubuf, min_t(size_t, sizeof(buf) - 1, count)))
+		return -EFAULT;
+
+	ret = kstrtou32(buf, 10, &versionstrsel);
+	if (ret)
+		return ret;
+
+	priv->versionstrsel = versionstrsel;
+
+	return count;
+}
+
+/* Proc verext file read handler.
+ * This function is called when the 'verext' file is opened for reading
+ * This function can be used read driver exteneed verion string.
+ */
+static ssize_t
+mwifiex_verext_read(struct file *file, char __user *ubuf,
+		    size_t count, loff_t *ppos)
+{
+	struct mwifiex_private *priv =
+		(struct mwifiex_private *)file->private_data;
+	char buf[256];
+	int ret;
+
+	mwifiex_get_ver_ext(priv, priv->versionstrsel);
+	ret = snprintf(buf, sizeof(buf), "version string: %s\n",
+		       priv->version_str);
+
+	return simple_read_from_buffer(ubuf, count, ppos, buf, ret);
+}
+
 /* Proc memrw file write handler.
  * This function is called when the 'memrw' file is opened for writing
  * This function can be used to write to a memory location.
@@ -940,6 +985,7 @@ MWIFIEX_DFS_FILE_OPS(histogram);
 MWIFIEX_DFS_FILE_OPS(debug_mask);
 MWIFIEX_DFS_FILE_OPS(timeshare_coex);
 MWIFIEX_DFS_FILE_WRITE_OPS(reset);
+MWIFIEX_DFS_FILE_OPS(verext);
 
 /*
  * This function creates the debug FS directory structure and the files.
@@ -968,6 +1014,7 @@ mwifiex_dev_debugfs_init(struct mwifiex_private *priv)
 	MWIFIEX_DFS_ADD_FILE(debug_mask);
 	MWIFIEX_DFS_ADD_FILE(timeshare_coex);
 	MWIFIEX_DFS_ADD_FILE(reset);
+	MWIFIEX_DFS_ADD_FILE(verext);
 }
 
 /*
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index c08be79..35ab052 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -616,6 +616,7 @@ struct mwifiex_private {
 	spinlock_t curr_bcn_buf_lock;
 	struct wireless_dev wdev;
 	struct mwifiex_chan_freq_power cfp;
+	u32 versionstrsel;
 	char version_str[128];
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *dfs_dev_dir;
@@ -1418,7 +1419,7 @@ int mwifiex_set_encode(struct mwifiex_private *priv, struct key_params *kp,
 
 int mwifiex_set_gen_ie(struct mwifiex_private *priv, const u8 *ie, int ie_len);
 
-int mwifiex_get_ver_ext(struct mwifiex_private *priv);
+int mwifiex_get_ver_ext(struct mwifiex_private *priv, u32 version_str_sel);
 
 int mwifiex_remain_on_chan_cfg(struct mwifiex_private *priv, u16 action,
 			       struct ieee80211_channel *chan,
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
index 6a4fc5d..210b257 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
@@ -1114,11 +1114,12 @@ int mwifiex_set_encode(struct mwifiex_private *priv, struct key_params *kp,
  * with requisite parameters and calls the IOCTL handler.
  */
 int
-mwifiex_get_ver_ext(struct mwifiex_private *priv)
+mwifiex_get_ver_ext(struct mwifiex_private *priv, u32 version_str_sel)
 {
 	struct mwifiex_ver_ext ver_ext;
 
 	memset(&ver_ext, 0, sizeof(struct host_cmd_ds_version_ext));
+	ver_ext.version_str_sel = version_str_sel;
 	if (mwifiex_send_cmd(priv, HostCmd_CMD_VERSION_EXT,
 			     HostCmd_ACT_GEN_GET, 0, &ver_ext, true))
 		return -1;
-- 
cgit v0.10.2


From 68f37e5d7a2e00306adab033fba6c3042b33e8e1 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Thu, 7 Jan 2016 14:28:50 +0100
Subject: wlcore: fix error handling in wlcore_event_fw_logger

wlcore_read/wlcore_write can return negative values so it should
be assigned to signed variable.

The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci [1].

[1]: http://permalink.gmane.org/gmane.linux.kernel/2120705

Fixes: 3719c17e1816 ("wlcore/wl18xx: fw logger over sdio")
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ti/wlcore/event.c b/drivers/net/wireless/ti/wlcore/event.c
index c964054..4b59f67 100644
--- a/drivers/net/wireless/ti/wlcore/event.c
+++ b/drivers/net/wireless/ti/wlcore/event.c
@@ -38,7 +38,7 @@
 
 int wlcore_event_fw_logger(struct wl1271 *wl)
 {
-	u32 ret;
+	int ret;
 	struct fw_logger_information fw_log;
 	u8  *buffer;
 	u32 internal_fw_addrbase = WL18XX_DATA_RAM_BASE_ADDRESS;
-- 
cgit v0.10.2


From 99074fc1e67b374b5c72406a23ac01fed806d634 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Mon, 11 Jan 2016 02:16:40 -0800
Subject: mwifiex: enable pcie MSIx interrupt mode support

Newer pcie chipsets (8997 onwards) support MSIx. This
patch enables it.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index f448d7c..5e15464 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2082,20 +2082,28 @@ mwifiex_check_winner_status(struct mwifiex_adapter *adapter)
 /*
  * This function reads the interrupt status from card.
  */
-static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter)
+static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter,
+				     int msg_id)
 {
 	u32 pcie_ireg;
 	unsigned long flags;
+	struct pcie_service_card *card = adapter->card;
 
 	if (!mwifiex_pcie_ok_to_access_hw(adapter))
 		return;
 
-	if (mwifiex_read_reg(adapter, PCIE_HOST_INT_STATUS, &pcie_ireg)) {
-		mwifiex_dbg(adapter, ERROR, "Read register failed\n");
-		return;
-	}
+	if (card->msix_enable && msg_id >= 0) {
+		pcie_ireg = BIT(msg_id);
+	} else {
+		if (mwifiex_read_reg(adapter, PCIE_HOST_INT_STATUS,
+				     &pcie_ireg)) {
+			mwifiex_dbg(adapter, ERROR, "Read register failed\n");
+			return;
+		}
+
+		if ((pcie_ireg == 0xFFFFFFFF) || !pcie_ireg)
+			return;
 
-	if ((pcie_ireg != 0xFFFFFFFF) && (pcie_ireg)) {
 
 		mwifiex_pcie_disable_host_int(adapter);
 
@@ -2106,20 +2114,23 @@ static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter)
 				    "Write register failed\n");
 			return;
 		}
-		spin_lock_irqsave(&adapter->int_lock, flags);
-		adapter->int_status |= pcie_ireg;
-		spin_unlock_irqrestore(&adapter->int_lock, flags);
-
-		if (!adapter->pps_uapsd_mode &&
-		    adapter->ps_state == PS_STATE_SLEEP &&
-		    mwifiex_pcie_ok_to_access_hw(adapter)) {
-				/* Potentially for PCIe we could get other
-				 * interrupts like shared. Don't change power
-				 * state until cookie is set */
-				adapter->ps_state = PS_STATE_AWAKE;
-				adapter->pm_wakeup_fw_try = false;
-				del_timer(&adapter->wakeup_timer);
-		}
+	}
+
+	spin_lock_irqsave(&adapter->int_lock, flags);
+	adapter->int_status |= pcie_ireg;
+	spin_unlock_irqrestore(&adapter->int_lock, flags);
+	mwifiex_dbg(adapter, INTR, "ireg: 0x%08x\n", pcie_ireg);
+
+	if (!adapter->pps_uapsd_mode &&
+	    adapter->ps_state == PS_STATE_SLEEP &&
+	    mwifiex_pcie_ok_to_access_hw(adapter)) {
+		/* Potentially for PCIe we could get other
+		 * interrupts like shared. Don't change power
+		 * state until cookie is set
+		 */
+		adapter->ps_state = PS_STATE_AWAKE;
+		adapter->pm_wakeup_fw_try = false;
+		del_timer(&adapter->wakeup_timer);
 	}
 }
 
@@ -2131,7 +2142,8 @@ static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter)
  */
 static irqreturn_t mwifiex_pcie_interrupt(int irq, void *context)
 {
-	struct pci_dev *pdev = (struct pci_dev *)context;
+	struct mwifiex_msix_context *ctx = context;
+	struct pci_dev *pdev = ctx->dev;
 	struct pcie_service_card *card;
 	struct mwifiex_adapter *adapter;
 
@@ -2151,7 +2163,11 @@ static irqreturn_t mwifiex_pcie_interrupt(int irq, void *context)
 	if (adapter->surprise_removed)
 		goto exit;
 
-	mwifiex_interrupt_status(adapter);
+	if (card->msix_enable)
+		mwifiex_interrupt_status(adapter, ctx->msg_id);
+	else
+		mwifiex_interrupt_status(adapter, -1);
+
 	mwifiex_queue_main_work(adapter);
 
 exit:
@@ -2171,7 +2187,7 @@ exit:
  * In case of Rx packets received, the packets are uploaded from card to
  * host and processed accordingly.
  */
-static int mwifiex_process_int_status(struct mwifiex_adapter *adapter)
+static int mwifiex_process_pcie_int(struct mwifiex_adapter *adapter)
 {
 	int ret;
 	u32 pcie_ireg;
@@ -2251,6 +2267,69 @@ static int mwifiex_process_int_status(struct mwifiex_adapter *adapter)
 	return 0;
 }
 
+static int mwifiex_process_msix_int(struct mwifiex_adapter *adapter)
+{
+	int ret;
+	u32 pcie_ireg;
+	unsigned long flags;
+
+	spin_lock_irqsave(&adapter->int_lock, flags);
+	/* Clear out unused interrupts */
+	pcie_ireg = adapter->int_status;
+	adapter->int_status = 0;
+	spin_unlock_irqrestore(&adapter->int_lock, flags);
+
+	if (pcie_ireg & HOST_INTR_DNLD_DONE) {
+		mwifiex_dbg(adapter, INTR,
+			    "info: TX DNLD Done\n");
+		ret = mwifiex_pcie_send_data_complete(adapter);
+		if (ret)
+			return ret;
+	}
+	if (pcie_ireg & HOST_INTR_UPLD_RDY) {
+		mwifiex_dbg(adapter, INTR,
+			    "info: Rx DATA\n");
+		ret = mwifiex_pcie_process_recv_data(adapter);
+		if (ret)
+			return ret;
+	}
+	if (pcie_ireg & HOST_INTR_EVENT_RDY) {
+		mwifiex_dbg(adapter, INTR,
+			    "info: Rx EVENT\n");
+		ret = mwifiex_pcie_process_event_ready(adapter);
+		if (ret)
+			return ret;
+	}
+
+	if (pcie_ireg & HOST_INTR_CMD_DONE) {
+		if (adapter->cmd_sent) {
+			mwifiex_dbg(adapter, INTR,
+				    "info: CMD sent Interrupt\n");
+			adapter->cmd_sent = false;
+		}
+		/* Handle command response */
+		ret = mwifiex_pcie_process_cmd_complete(adapter);
+		if (ret)
+			return ret;
+	}
+
+	mwifiex_dbg(adapter, INTR,
+		    "info: cmd_sent=%d data_sent=%d\n",
+		    adapter->cmd_sent, adapter->data_sent);
+
+	return 0;
+}
+
+static int mwifiex_process_int_status(struct mwifiex_adapter *adapter)
+{
+	struct pcie_service_card *card = adapter->card;
+
+	if (card->msix_enable)
+		return mwifiex_process_msix_int(adapter);
+	else
+		return mwifiex_process_pcie_int(adapter);
+}
+
 /*
  * This function downloads data from driver to card.
  *
@@ -2602,10 +2681,43 @@ static void mwifiex_pcie_cleanup(struct mwifiex_adapter *adapter)
 
 static int mwifiex_pcie_request_irq(struct mwifiex_adapter *adapter)
 {
-	int ret;
+	int ret, i, j;
 	struct pcie_service_card *card = adapter->card;
 	struct pci_dev *pdev = card->dev;
 
+	if (card->pcie.reg->msix_support) {
+		for (i = 0; i < MWIFIEX_NUM_MSIX_VECTORS; i++)
+			card->msix_entries[i].entry = i;
+		ret = pci_enable_msix_exact(pdev, card->msix_entries,
+					    MWIFIEX_NUM_MSIX_VECTORS);
+		if (!ret) {
+			for (i = 0; i < MWIFIEX_NUM_MSIX_VECTORS; i++) {
+				card->msix_ctx[i].dev = pdev;
+				card->msix_ctx[i].msg_id = i;
+
+				ret = request_irq(card->msix_entries[i].vector,
+						  mwifiex_pcie_interrupt, 0,
+						  "MWIFIEX_PCIE_MSIX",
+						  &card->msix_ctx[i]);
+				if (ret)
+					break;
+			}
+
+			if (ret) {
+				mwifiex_dbg(adapter, INFO, "request_irq fail: %d\n",
+					    ret);
+				for (j = 0; j < i; j++)
+					free_irq(card->msix_entries[j].vector,
+						 &card->msix_ctx[i]);
+				pci_disable_msix(pdev);
+			} else {
+				mwifiex_dbg(adapter, MSG, "MSIx enabled!");
+				card->msix_enable = 1;
+				return 0;
+			}
+		}
+	}
+
 	if (pci_enable_msi(pdev) != 0)
 		pci_disable_msi(pdev);
 	else
@@ -2613,8 +2725,10 @@ static int mwifiex_pcie_request_irq(struct mwifiex_adapter *adapter)
 
 	mwifiex_dbg(adapter, INFO, "msi_enable = %d\n", card->msi_enable);
 
+	card->share_irq_ctx.dev = pdev;
+	card->share_irq_ctx.msg_id = -1;
 	ret = request_irq(pdev->irq, mwifiex_pcie_interrupt, IRQF_SHARED,
-			  "MRVL_PCIE", pdev);
+			  "MRVL_PCIE", &card->share_irq_ctx);
 	if (ret) {
 		pr_err("request_irq failed: ret=%d\n", ret);
 		adapter->card = NULL;
@@ -2660,11 +2774,28 @@ static void mwifiex_unregister_dev(struct mwifiex_adapter *adapter)
 {
 	struct pcie_service_card *card = adapter->card;
 	const struct mwifiex_pcie_card_reg *reg;
+	struct pci_dev *pdev = card->dev;
+	int i;
 
 	if (card) {
-		mwifiex_dbg(adapter, INFO,
-			    "%s(): calling free_irq()\n", __func__);
-		free_irq(card->dev->irq, card->dev);
+		if (card->msix_enable) {
+			for (i = 0; i < MWIFIEX_NUM_MSIX_VECTORS; i++)
+				synchronize_irq(card->msix_entries[i].vector);
+
+			for (i = 0; i < MWIFIEX_NUM_MSIX_VECTORS; i++)
+				free_irq(card->msix_entries[i].vector,
+					 &card->msix_ctx[i]);
+
+			card->msix_enable = 0;
+			pci_disable_msix(pdev);
+	       } else {
+			mwifiex_dbg(adapter, INFO,
+				    "%s(): calling free_irq()\n", __func__);
+		       free_irq(card->dev->irq, &card->share_irq_ctx);
+
+			if (card->msi_enable)
+				pci_disable_msi(pdev);
+	       }
 
 		reg = card->pcie.reg;
 		if (reg->sleep_cookie)
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.h b/drivers/net/wireless/marvell/mwifiex/pcie.h
index 6fc2873..9700ac3 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.h
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.h
@@ -135,6 +135,7 @@ struct mwifiex_pcie_card_reg {
 	u16 fw_dump_ctrl;
 	u16 fw_dump_start;
 	u16 fw_dump_end;
+	u8 msix_support;
 };
 
 static const struct mwifiex_pcie_card_reg mwifiex_reg_8766 = {
@@ -166,6 +167,7 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8766 = {
 	.ring_tx_start_ptr = 0,
 	.pfu_enabled = 0,
 	.sleep_cookie = 1,
+	.msix_support = 0,
 };
 
 static const struct mwifiex_pcie_card_reg mwifiex_reg_8897 = {
@@ -200,6 +202,7 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8897 = {
 	.fw_dump_ctrl = 0xcf4,
 	.fw_dump_start = 0xcf8,
 	.fw_dump_end = 0xcff,
+	.msix_support = 0,
 };
 
 static const struct mwifiex_pcie_card_reg mwifiex_reg_8997 = {
@@ -231,6 +234,7 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8997 = {
 	.ring_tx_start_ptr = MWIFIEX_BD_FLAG_TX_START_PTR,
 	.pfu_enabled = 1,
 	.sleep_cookie = 0,
+	.msix_support = 1,
 };
 
 struct mwifiex_pcie_device {
@@ -290,6 +294,13 @@ struct mwifiex_pfu_buf_desc {
 	u32 reserved;
 } __packed;
 
+#define MWIFIEX_NUM_MSIX_VECTORS   4
+
+struct mwifiex_msix_context {
+	struct pci_dev *dev;
+	u16 msg_id;
+};
+
 struct pcie_service_card {
 	struct pci_dev *dev;
 	struct mwifiex_adapter *adapter;
@@ -327,6 +338,12 @@ struct pcie_service_card {
 	void __iomem *pci_mmap;
 	void __iomem *pci_mmap1;
 	int msi_enable;
+	int msix_enable;
+#ifdef CONFIG_PCI
+	struct msix_entry msix_entries[MWIFIEX_NUM_MSIX_VECTORS];
+#endif
+	struct mwifiex_msix_context msix_ctx[MWIFIEX_NUM_MSIX_VECTORS];
+	struct mwifiex_msix_context share_irq_ctx;
 };
 
 static inline int
-- 
cgit v0.10.2


From dc896b15fe94345912ff13fa8eb581b5c6acccde Mon Sep 17 00:00:00 2001
From: Shengzhen Li <szli@marvell.com>
Date: Tue, 12 Jan 2016 05:43:16 -0800
Subject: mwifiex: fix power state out of sync problem

It's been observed that driver's power state goes out
of sync with firmware in some corner cases. When this
happens, driver tries to download the data to firmware
assuming it's awake which causes Tx data timeout.

Main thread will process interrupt as soon as interrupt
handler fills 'int_status' variable.

This patch fixes the race issue by updating 'int_status'
at the end of mwifiex_interrupt_status().

Signed-off-by: Shengzhen Li <szli@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index 5e15464..918e049 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2116,11 +2116,6 @@ static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter,
 		}
 	}
 
-	spin_lock_irqsave(&adapter->int_lock, flags);
-	adapter->int_status |= pcie_ireg;
-	spin_unlock_irqrestore(&adapter->int_lock, flags);
-	mwifiex_dbg(adapter, INTR, "ireg: 0x%08x\n", pcie_ireg);
-
 	if (!adapter->pps_uapsd_mode &&
 	    adapter->ps_state == PS_STATE_SLEEP &&
 	    mwifiex_pcie_ok_to_access_hw(adapter)) {
@@ -2132,6 +2127,11 @@ static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter,
 		adapter->pm_wakeup_fw_try = false;
 		del_timer(&adapter->wakeup_timer);
 	}
+
+	spin_lock_irqsave(&adapter->int_lock, flags);
+	adapter->int_status |= pcie_ireg;
+	spin_unlock_irqrestore(&adapter->int_lock, flags);
+	mwifiex_dbg(adapter, INTR, "ireg: 0x%08x\n", pcie_ireg);
 }
 
 /*
-- 
cgit v0.10.2


From 0c9b7f22e8e1f3aa5b88d7530db8b3a7d647adb6 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Wed, 13 Jan 2016 01:26:52 -0800
Subject: mwifiex: add schedule scan support

This patch add sched scan support for mwifiex, include cfg80211
sched_scan_start/sched_scan_stop handler, corresponding bgscan
command path and event handler.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: chunfan chen <jeffc@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index e7adef7..c27c6cc 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -1962,6 +1962,9 @@ mwifiex_cfg80211_disconnect(struct wiphy *wiphy, struct net_device *dev,
 {
 	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
 
+	if (!mwifiex_stop_bg_scan(priv))
+		cfg80211_sched_scan_stopped_rtnl(priv->wdev.wiphy);
+
 	if (mwifiex_deauthenticate(priv, NULL))
 		return -EFAULT;
 
@@ -2217,6 +2220,9 @@ mwifiex_cfg80211_connect(struct wiphy *wiphy, struct net_device *dev,
 		    "info: Trying to associate to %s and bssid %pM\n",
 		    (char *)sme->ssid, sme->bssid);
 
+	if (!mwifiex_stop_bg_scan(priv))
+		cfg80211_sched_scan_stopped_rtnl(priv->wdev.wiphy);
+
 	ret = mwifiex_cfg80211_assoc(priv, sme->ssid_len, sme->ssid, sme->bssid,
 				     priv->bss_mode, sme->channel, sme, 0);
 	if (!ret) {
@@ -2420,6 +2426,9 @@ mwifiex_cfg80211_scan(struct wiphy *wiphy,
 		return -EBUSY;
 	}
 
+	if (!mwifiex_stop_bg_scan(priv))
+		cfg80211_sched_scan_stopped_rtnl(priv->wdev.wiphy);
+
 	user_scan_cfg = kzalloc(sizeof(*user_scan_cfg), GFP_KERNEL);
 	if (!user_scan_cfg)
 		return -ENOMEM;
@@ -2487,6 +2496,121 @@ mwifiex_cfg80211_scan(struct wiphy *wiphy,
 	return 0;
 }
 
+/* CFG802.11 operation handler for sched_scan_start.
+ *
+ * This function issues a bgscan config request to the firmware based upon
+ * the user specified sched_scan configuration. On successful completion,
+ * firmware will generate BGSCAN_REPORT event, driver should issue bgscan
+ * query command to get sched_scan results from firmware.
+ */
+static int
+mwifiex_cfg80211_sched_scan_start(struct wiphy *wiphy,
+				  struct net_device *dev,
+				  struct cfg80211_sched_scan_request *request)
+{
+	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
+	int i, offset;
+	struct ieee80211_channel *chan;
+	struct mwifiex_bg_scan_cfg *bgscan_cfg;
+	struct ieee_types_header *ie;
+
+	if (!request || (!request->n_ssids && !request->n_match_sets)) {
+		wiphy_err(wiphy, "%s : Invalid Sched_scan parameters",
+			  __func__);
+		return -EINVAL;
+	}
+
+	wiphy_info(wiphy, "sched_scan start : n_ssids=%d n_match_sets=%d ",
+		   request->n_ssids, request->n_match_sets);
+	wiphy_info(wiphy, "n_channels=%d interval=%d ie_len=%d\n",
+		   request->n_channels, request->scan_plans->interval,
+		   (int)request->ie_len);
+
+	bgscan_cfg = kzalloc(sizeof(*bgscan_cfg), GFP_KERNEL);
+	if (!bgscan_cfg)
+		return -ENOMEM;
+
+	if (priv->scan_request || priv->scan_aborting)
+		bgscan_cfg->start_later = true;
+
+	bgscan_cfg->num_ssids = request->n_match_sets;
+	bgscan_cfg->ssid_list = request->match_sets;
+
+	if (request->ie && request->ie_len) {
+		offset = 0;
+		for (i = 0; i < MWIFIEX_MAX_VSIE_NUM; i++) {
+			if (priv->vs_ie[i].mask != MWIFIEX_VSIE_MASK_CLEAR)
+				continue;
+			priv->vs_ie[i].mask = MWIFIEX_VSIE_MASK_BGSCAN;
+			ie = (struct ieee_types_header *)(request->ie + offset);
+			memcpy(&priv->vs_ie[i].ie, ie, sizeof(*ie) + ie->len);
+			offset += sizeof(*ie) + ie->len;
+
+			if (offset >= request->ie_len)
+				break;
+		}
+	}
+
+	for (i = 0; i < min_t(u32, request->n_channels,
+			      MWIFIEX_BG_SCAN_CHAN_MAX); i++) {
+		chan = request->channels[i];
+		bgscan_cfg->chan_list[i].chan_number = chan->hw_value;
+		bgscan_cfg->chan_list[i].radio_type = chan->band;
+
+		if ((chan->flags & IEEE80211_CHAN_NO_IR) || !request->n_ssids)
+			bgscan_cfg->chan_list[i].scan_type =
+						MWIFIEX_SCAN_TYPE_PASSIVE;
+		else
+			bgscan_cfg->chan_list[i].scan_type =
+						MWIFIEX_SCAN_TYPE_ACTIVE;
+
+		bgscan_cfg->chan_list[i].scan_time = 0;
+	}
+
+	bgscan_cfg->chan_per_scan = min_t(u32, request->n_channels,
+					  MWIFIEX_BG_SCAN_CHAN_MAX);
+
+	/* Use at least 15 second for per scan cycle */
+	bgscan_cfg->scan_interval = (request->scan_plans->interval >
+				     MWIFIEX_BGSCAN_INTERVAL) ?
+				request->scan_plans->interval :
+				MWIFIEX_BGSCAN_INTERVAL;
+
+	bgscan_cfg->repeat_count = MWIFIEX_BGSCAN_REPEAT_COUNT;
+	bgscan_cfg->report_condition = MWIFIEX_BGSCAN_SSID_MATCH |
+				MWIFIEX_BGSCAN_WAIT_ALL_CHAN_DONE;
+	bgscan_cfg->bss_type = MWIFIEX_BSS_MODE_INFRA;
+	bgscan_cfg->action = MWIFIEX_BGSCAN_ACT_SET;
+	bgscan_cfg->enable = true;
+
+	if (mwifiex_send_cmd(priv, HostCmd_CMD_802_11_BG_SCAN_CONFIG,
+			     HostCmd_ACT_GEN_SET, 0, bgscan_cfg, true)) {
+		kfree(bgscan_cfg);
+		return -EFAULT;
+	}
+
+	priv->sched_scanning = true;
+
+	kfree(bgscan_cfg);
+	return 0;
+}
+
+/* CFG802.11 operation handler for sched_scan_stop.
+ *
+ * This function issues a bgscan config command to disable
+ * previous bgscan configuration in the firmware
+ */
+static int mwifiex_cfg80211_sched_scan_stop(struct wiphy *wiphy,
+					    struct net_device *dev)
+{
+	struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev);
+
+	wiphy_info(wiphy, "sched scan stop!");
+	mwifiex_stop_bg_scan(priv);
+
+	return 0;
+}
+
 static void mwifiex_setup_vht_caps(struct ieee80211_sta_vht_cap *vht_info,
 				   struct mwifiex_private *priv)
 {
@@ -2848,6 +2972,9 @@ int mwifiex_del_virtual_intf(struct wiphy *wiphy, struct wireless_dev *wdev)
 	mwifiex_dev_debugfs_remove(priv);
 #endif
 
+	if (priv->sched_scanning)
+		priv->sched_scanning = false;
+
 	mwifiex_stop_net_dev_queue(priv->netdev, adapter);
 
 	skb_queue_walk_safe(&priv->bypass_txq, skb, tmp)
@@ -3701,6 +3828,8 @@ static struct cfg80211_ops mwifiex_cfg80211_ops = {
 	.set_cqm_rssi_config = mwifiex_cfg80211_set_cqm_rssi_config,
 	.set_antenna = mwifiex_cfg80211_set_antenna,
 	.del_station = mwifiex_cfg80211_del_station,
+	.sched_scan_start = mwifiex_cfg80211_sched_scan_start,
+	.sched_scan_stop = mwifiex_cfg80211_sched_scan_stop,
 #ifdef CONFIG_PM
 	.suspend = mwifiex_cfg80211_suspend,
 	.resume = mwifiex_cfg80211_resume,
@@ -3829,6 +3958,7 @@ int mwifiex_register_cfg80211(struct mwifiex_adapter *adapter)
 	wiphy->flags |= WIPHY_FLAG_HAVE_AP_SME |
 			WIPHY_FLAG_AP_PROBE_RESP_OFFLOAD |
 			WIPHY_FLAG_AP_UAPSD |
+			WIPHY_FLAG_SUPPORTS_SCHED_SCAN |
 			WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL |
 			WIPHY_FLAG_HAS_CHANNEL_SWITCH |
 			WIPHY_FLAG_PS_ON_BY_DEFAULT;
@@ -3847,6 +3977,10 @@ int mwifiex_register_cfg80211(struct mwifiex_adapter *adapter)
 				    NL80211_PROBE_RESP_OFFLOAD_SUPPORT_WPS2 |
 				    NL80211_PROBE_RESP_OFFLOAD_SUPPORT_P2P;
 
+	wiphy->max_sched_scan_ssids = MWIFIEX_MAX_SSID_LIST_LENGTH;
+	wiphy->max_sched_scan_ie_len = MWIFIEX_MAX_VSIE_LEN;
+	wiphy->max_match_sets = MWIFIEX_MAX_SSID_LIST_LENGTH;
+
 	wiphy->available_antennas_tx = BIT(adapter->number_of_antenna) - 1;
 	wiphy->available_antennas_rx = BIT(adapter->number_of_antenna) - 1;
 
diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h
index 426e76a..07bdc2a 100644
--- a/drivers/net/wireless/marvell/mwifiex/fw.h
+++ b/drivers/net/wireless/marvell/mwifiex/fw.h
@@ -144,6 +144,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 #define TLV_TYPE_WILDCARDSSID       (PROPRIETARY_TLV_BASE_ID + 18)
 #define TLV_TYPE_TSFTIMESTAMP       (PROPRIETARY_TLV_BASE_ID + 19)
 #define TLV_TYPE_RSSI_HIGH          (PROPRIETARY_TLV_BASE_ID + 22)
+#define TLV_TYPE_BGSCAN_START_LATER (PROPRIETARY_TLV_BASE_ID + 30)
 #define TLV_TYPE_AUTH_TYPE          (PROPRIETARY_TLV_BASE_ID + 31)
 #define TLV_TYPE_STA_MAC_ADDR       (PROPRIETARY_TLV_BASE_ID + 32)
 #define TLV_TYPE_BSSID              (PROPRIETARY_TLV_BASE_ID + 35)
@@ -177,6 +178,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 #define TLV_TYPE_TX_PAUSE           (PROPRIETARY_TLV_BASE_ID + 148)
 #define TLV_TYPE_COALESCE_RULE      (PROPRIETARY_TLV_BASE_ID + 154)
 #define TLV_TYPE_KEY_PARAM_V2       (PROPRIETARY_TLV_BASE_ID + 156)
+#define TLV_TYPE_REPEAT_COUNT       (PROPRIETARY_TLV_BASE_ID + 176)
 #define TLV_TYPE_MULTI_CHAN_INFO    (PROPRIETARY_TLV_BASE_ID + 183)
 #define TLV_TYPE_MC_GROUP_INFO      (PROPRIETARY_TLV_BASE_ID + 184)
 #define TLV_TYPE_TDLS_IDLE_TIMEOUT  (PROPRIETARY_TLV_BASE_ID + 194)
@@ -331,6 +333,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 #define HostCmd_CMD_802_11_MAC_ADDRESS                0x004D
 #define HostCmd_CMD_802_11D_DOMAIN_INFO               0x005b
 #define HostCmd_CMD_802_11_KEY_MATERIAL               0x005e
+#define HostCmd_CMD_802_11_BG_SCAN_CONFIG             0x006b
 #define HostCmd_CMD_802_11_BG_SCAN_QUERY              0x006c
 #define HostCmd_CMD_WMM_GET_STATUS                    0x0071
 #define HostCmd_CMD_802_11_SUBSCRIBE_EVENT            0x0075
@@ -523,6 +526,7 @@ enum P2P_MODES {
 #define EVENT_CHANNEL_REPORT_RDY        0x00000054
 #define EVENT_TX_DATA_PAUSE             0x00000055
 #define EVENT_EXT_SCAN_REPORT           0x00000058
+#define EVENT_BG_SCAN_STOPPED           0x00000065
 #define EVENT_REMAIN_ON_CHAN_EXPIRED    0x0000005f
 #define EVENT_MULTI_CHAN_INFO           0x0000006a
 #define EVENT_TX_STATUS_REPORT		0x00000074
@@ -733,6 +737,16 @@ struct mwifiex_ie_types_num_probes {
 	__le16 num_probes;
 } __packed;
 
+struct mwifiex_ie_types_repeat_count {
+	struct mwifiex_ie_types_header header;
+	__le16 repeat_count;
+} __packed;
+
+struct mwifiex_ie_types_bgscan_start_later {
+	struct mwifiex_ie_types_header header;
+	__le16 start_later;
+} __packed;
+
 struct mwifiex_ie_types_scan_chan_gap {
 	struct mwifiex_ie_types_header header;
 	/* time gap in TUs to be used between two consecutive channels scan */
@@ -1425,6 +1439,36 @@ struct mwifiex_user_scan_cfg {
 	u16 scan_chan_gap;
 } __packed;
 
+#define MWIFIEX_BG_SCAN_CHAN_MAX 38
+#define MWIFIEX_BSS_MODE_INFRA 1
+#define MWIFIEX_BGSCAN_ACT_GET     0x0000
+#define MWIFIEX_BGSCAN_ACT_SET     0x0001
+#define MWIFIEX_BGSCAN_ACT_SET_ALL 0xff01
+/** ssid match */
+#define MWIFIEX_BGSCAN_SSID_MATCH          0x0001
+/** ssid match and RSSI exceeded */
+#define MWIFIEX_BGSCAN_SSID_RSSI_MATCH     0x0004
+/**wait for all channel scan to complete to report scan result*/
+#define MWIFIEX_BGSCAN_WAIT_ALL_CHAN_DONE  0x80000000
+
+struct mwifiex_bg_scan_cfg {
+	u16 action;
+	u8 enable;
+	u8 bss_type;
+	u8 chan_per_scan;
+	u32 scan_interval;
+	u32 report_condition;
+	u8 num_probes;
+	u8 rssi_threshold;
+	u8 snr_threshold;
+	u16 repeat_count;
+	u16 start_later;
+	struct cfg80211_match_set *ssid_list;
+	u8 num_ssids;
+	struct mwifiex_user_scan_chan chan_list[MWIFIEX_BG_SCAN_CHAN_MAX];
+	u16 scan_chan_gap;
+} __packed;
+
 struct ie_body {
 	u8 grp_key_oui[4];
 	u8 ptk_cnt[2];
@@ -1470,6 +1514,20 @@ struct mwifiex_ie_types_bss_scan_info {
 	__le64 tsf;
 } __packed;
 
+struct host_cmd_ds_802_11_bg_scan_config {
+	__le16 action;
+	u8 enable;
+	u8 bss_type;
+	u8 chan_per_scan;
+	u8 reserved;
+	__le16 reserved1;
+	__le32 scan_interval;
+	__le32 reserved2;
+	__le32 report_condition;
+	__le16 reserved3;
+	u8 tlv[0];
+} __packed;
+
 struct host_cmd_ds_802_11_bg_scan_query {
 	u8 flush;
 } __packed;
@@ -2124,6 +2182,7 @@ struct host_cmd_ds_command {
 		struct host_cmd_ds_802_11_scan scan;
 		struct host_cmd_ds_802_11_scan_ext ext_scan;
 		struct host_cmd_ds_802_11_scan_rsp scan_resp;
+		struct host_cmd_ds_802_11_bg_scan_config bg_scan_config;
 		struct host_cmd_ds_802_11_bg_scan_query bg_scan_query;
 		struct host_cmd_ds_802_11_bg_scan_query_rsp bg_scan_query_resp;
 		struct host_cmd_ds_802_11_associate associate;
diff --git a/drivers/net/wireless/marvell/mwifiex/ioctl.h b/drivers/net/wireless/marvell/mwifiex/ioctl.h
index 4f0174c..6333b16 100644
--- a/drivers/net/wireless/marvell/mwifiex/ioctl.h
+++ b/drivers/net/wireless/marvell/mwifiex/ioctl.h
@@ -414,6 +414,7 @@ struct mwifiex_ds_mef_cfg {
 #define MWIFIEX_VSIE_MASK_SCAN     0x01
 #define MWIFIEX_VSIE_MASK_ASSOC    0x02
 #define MWIFIEX_VSIE_MASK_ADHOC    0x04
+#define MWIFIEX_VSIE_MASK_BGSCAN   0x08
 
 enum {
 	MWIFIEX_FUNC_INIT = 1,
diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c
index 79c16de..a99b72b 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.c
+++ b/drivers/net/wireless/marvell/mwifiex/main.c
@@ -746,6 +746,13 @@ int mwifiex_queue_tx_pkt(struct mwifiex_private *priv, struct sk_buff *skb)
 
 	mwifiex_queue_main_work(priv->adapter);
 
+	if (priv->sched_scanning) {
+		mwifiex_dbg(priv->adapter, INFO,
+			    "aborting bgscan on ndo_stop\n");
+		mwifiex_stop_bg_scan(priv);
+		cfg80211_sched_scan_stopped(priv->wdev.wiphy);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 35ab052..5f5bcf8 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -198,6 +198,11 @@ do {								\
 			       buf, len, false);		\
 } while (0)
 
+/** Min BGSCAN interval 15 second */
+#define MWIFIEX_BGSCAN_INTERVAL 15000
+/** default repeat count */
+#define MWIFIEX_BGSCAN_REPEAT_COUNT 6
+
 struct mwifiex_dbg {
 	u32 num_cmd_host_to_card_failure;
 	u32 num_cmd_sleep_cfm_host_to_card_failure;
@@ -641,6 +646,7 @@ struct mwifiex_private {
 	u32 mgmt_frame_mask;
 	struct mwifiex_roc_cfg roc_cfg;
 	bool scan_aborting;
+	u8 sched_scanning;
 	u8 csa_chan;
 	unsigned long csa_expire_time;
 	u8 del_list_idx;
@@ -1198,6 +1204,10 @@ int mwifiex_ret_802_11_scan_ext(struct mwifiex_private *priv,
 				struct host_cmd_ds_command *resp);
 int mwifiex_handle_event_ext_scan_report(struct mwifiex_private *priv,
 					 void *buf);
+int mwifiex_cmd_802_11_bg_scan_config(struct mwifiex_private *priv,
+				      struct host_cmd_ds_command *cmd,
+				      void *data_buf);
+int mwifiex_stop_bg_scan(struct mwifiex_private *priv);
 
 /*
  * This function checks if the queuing is RA based or not.
diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c
index c20017c..d4e2143 100644
--- a/drivers/net/wireless/marvell/mwifiex/scan.c
+++ b/drivers/net/wireless/marvell/mwifiex/scan.c
@@ -547,6 +547,61 @@ mwifiex_scan_create_channel_list(struct mwifiex_private *priv,
 	return chan_idx;
 }
 
+/* This function creates a channel list tlv for bgscan config, based
+ * on region/band information.
+ */
+static int
+mwifiex_bgscan_create_channel_list(struct mwifiex_private *priv,
+				   const struct mwifiex_bg_scan_cfg
+						*bgscan_cfg_in,
+				   struct mwifiex_chan_scan_param_set
+						*scan_chan_list)
+{
+	enum ieee80211_band band;
+	struct ieee80211_supported_band *sband;
+	struct ieee80211_channel *ch;
+	struct mwifiex_adapter *adapter = priv->adapter;
+	int chan_idx = 0, i;
+
+	for (band = 0; (band < IEEE80211_NUM_BANDS); band++) {
+		if (!priv->wdev.wiphy->bands[band])
+			continue;
+
+		sband = priv->wdev.wiphy->bands[band];
+
+		for (i = 0; (i < sband->n_channels) ; i++) {
+			ch = &sband->channels[i];
+			if (ch->flags & IEEE80211_CHAN_DISABLED)
+				continue;
+			scan_chan_list[chan_idx].radio_type = band;
+
+			if (bgscan_cfg_in->chan_list[0].scan_time)
+				scan_chan_list[chan_idx].max_scan_time =
+					cpu_to_le16((u16)bgscan_cfg_in->
+					chan_list[0].scan_time);
+			else if (ch->flags & IEEE80211_CHAN_NO_IR)
+				scan_chan_list[chan_idx].max_scan_time =
+					cpu_to_le16(adapter->passive_scan_time);
+			else
+				scan_chan_list[chan_idx].max_scan_time =
+					cpu_to_le16(adapter->
+						    specific_scan_time);
+
+			if (ch->flags & IEEE80211_CHAN_NO_IR)
+				scan_chan_list[chan_idx].chan_scan_mode_bitmap
+					|= MWIFIEX_PASSIVE_SCAN;
+			else
+				scan_chan_list[chan_idx].chan_scan_mode_bitmap
+					&= ~MWIFIEX_PASSIVE_SCAN;
+
+			scan_chan_list[chan_idx].chan_number =
+							(u32)ch->hw_value;
+			chan_idx++;
+		}
+	}
+	return chan_idx;
+}
+
 /* This function appends rate TLV to scan config command. */
 static int
 mwifiex_append_rate_tlv(struct mwifiex_private *priv,
@@ -2155,6 +2210,212 @@ int mwifiex_cmd_802_11_scan_ext(struct mwifiex_private *priv,
 	return 0;
 }
 
+/* This function prepares an background scan config command to be sent
+ * to the firmware
+ */
+int mwifiex_cmd_802_11_bg_scan_config(struct mwifiex_private *priv,
+				      struct host_cmd_ds_command *cmd,
+				      void *data_buf)
+{
+	struct host_cmd_ds_802_11_bg_scan_config *bgscan_config =
+					&cmd->params.bg_scan_config;
+	struct mwifiex_bg_scan_cfg *bgscan_cfg_in = data_buf;
+	u8 *tlv_pos = bgscan_config->tlv;
+	u8 num_probes;
+	u32 ssid_len, chan_idx, scan_type, scan_dur, chan_num;
+	int i;
+	struct mwifiex_ie_types_num_probes *num_probes_tlv;
+	struct mwifiex_ie_types_repeat_count *repeat_count_tlv;
+	struct mwifiex_ie_types_bgscan_start_later *start_later_tlv;
+	struct mwifiex_ie_types_wildcard_ssid_params *wildcard_ssid_tlv;
+	struct mwifiex_ie_types_chan_list_param_set *chan_list_tlv;
+	struct mwifiex_chan_scan_param_set *temp_chan;
+
+	cmd->command = cpu_to_le16(HostCmd_CMD_802_11_BG_SCAN_CONFIG);
+	cmd->size = cpu_to_le16(sizeof(*bgscan_config) + S_DS_GEN);
+
+	bgscan_config->action = cpu_to_le16(bgscan_cfg_in->action);
+	bgscan_config->enable = bgscan_cfg_in->enable;
+	bgscan_config->bss_type = bgscan_cfg_in->bss_type;
+	bgscan_config->scan_interval =
+		cpu_to_le32(bgscan_cfg_in->scan_interval);
+	bgscan_config->report_condition =
+		cpu_to_le32(bgscan_cfg_in->report_condition);
+
+	/*  stop sched scan  */
+	if (!bgscan_config->enable)
+		return 0;
+
+	bgscan_config->chan_per_scan = bgscan_cfg_in->chan_per_scan;
+
+	num_probes = (bgscan_cfg_in->num_probes ? bgscan_cfg_in->
+		      num_probes : priv->adapter->scan_probes);
+
+	if (num_probes) {
+		num_probes_tlv = (struct mwifiex_ie_types_num_probes *)tlv_pos;
+		num_probes_tlv->header.type = cpu_to_le16(TLV_TYPE_NUMPROBES);
+		num_probes_tlv->header.len =
+			cpu_to_le16(sizeof(num_probes_tlv->num_probes));
+		num_probes_tlv->num_probes = cpu_to_le16((u16)num_probes);
+
+		tlv_pos += sizeof(num_probes_tlv->header) +
+			le16_to_cpu(num_probes_tlv->header.len);
+	}
+
+	if (bgscan_cfg_in->repeat_count) {
+		repeat_count_tlv =
+			(struct mwifiex_ie_types_repeat_count *)tlv_pos;
+		repeat_count_tlv->header.type =
+			cpu_to_le16(TLV_TYPE_REPEAT_COUNT);
+		repeat_count_tlv->header.len =
+			cpu_to_le16(sizeof(repeat_count_tlv->repeat_count));
+		repeat_count_tlv->repeat_count =
+			cpu_to_le16(bgscan_cfg_in->repeat_count);
+
+		tlv_pos += sizeof(repeat_count_tlv->header) +
+			le16_to_cpu(repeat_count_tlv->header.len);
+	}
+
+	for (i = 0; i < bgscan_cfg_in->num_ssids; i++) {
+		ssid_len = bgscan_cfg_in->ssid_list[i].ssid.ssid_len;
+
+		wildcard_ssid_tlv =
+			(struct mwifiex_ie_types_wildcard_ssid_params *)tlv_pos;
+		wildcard_ssid_tlv->header.type =
+				cpu_to_le16(TLV_TYPE_WILDCARDSSID);
+		wildcard_ssid_tlv->header.len = cpu_to_le16(
+				(u16)(ssid_len + sizeof(wildcard_ssid_tlv->
+							 max_ssid_length)));
+
+		/* max_ssid_length = 0 tells firmware to perform
+		 * specific scan for the SSID filled, whereas
+		 * max_ssid_length = IEEE80211_MAX_SSID_LEN is for
+		 * wildcard scan.
+		 */
+		if (ssid_len)
+			wildcard_ssid_tlv->max_ssid_length = 0;
+		else
+			wildcard_ssid_tlv->max_ssid_length =
+						IEEE80211_MAX_SSID_LEN;
+
+		memcpy(wildcard_ssid_tlv->ssid,
+		       bgscan_cfg_in->ssid_list[i].ssid.ssid, ssid_len);
+
+		tlv_pos += (sizeof(wildcard_ssid_tlv->header)
+				+ le16_to_cpu(wildcard_ssid_tlv->header.len));
+	}
+
+	chan_list_tlv = (struct mwifiex_ie_types_chan_list_param_set *)tlv_pos;
+
+	if (bgscan_cfg_in->chan_list[0].chan_number) {
+		dev_dbg(priv->adapter->dev, "info: bgscan: Using supplied channel list\n");
+
+		chan_list_tlv->header.type = cpu_to_le16(TLV_TYPE_CHANLIST);
+
+		for (chan_idx = 0;
+		     chan_idx < MWIFIEX_BG_SCAN_CHAN_MAX &&
+		     bgscan_cfg_in->chan_list[chan_idx].chan_number;
+		     chan_idx++) {
+			temp_chan = chan_list_tlv->chan_scan_param + chan_idx;
+
+			/* Increment the TLV header length by size appended */
+			le16_add_cpu(&chan_list_tlv->header.len,
+				     sizeof(chan_list_tlv->chan_scan_param));
+
+			temp_chan->chan_number =
+				bgscan_cfg_in->chan_list[chan_idx].chan_number;
+			temp_chan->radio_type =
+				bgscan_cfg_in->chan_list[chan_idx].radio_type;
+
+			scan_type =
+				bgscan_cfg_in->chan_list[chan_idx].scan_type;
+
+			if (scan_type == MWIFIEX_SCAN_TYPE_PASSIVE)
+				temp_chan->chan_scan_mode_bitmap
+					|= MWIFIEX_PASSIVE_SCAN;
+			else
+				temp_chan->chan_scan_mode_bitmap
+					&= ~MWIFIEX_PASSIVE_SCAN;
+
+			if (bgscan_cfg_in->chan_list[chan_idx].scan_time) {
+				scan_dur = (u16)bgscan_cfg_in->
+					chan_list[chan_idx].scan_time;
+			} else {
+				scan_dur = (scan_type ==
+					    MWIFIEX_SCAN_TYPE_PASSIVE) ?
+					    priv->adapter->passive_scan_time :
+					    priv->adapter->specific_scan_time;
+			}
+
+			temp_chan->min_scan_time = cpu_to_le16(scan_dur);
+			temp_chan->max_scan_time = cpu_to_le16(scan_dur);
+		}
+	} else {
+		dev_dbg(priv->adapter->dev,
+			"info: bgscan: Creating full region channel list\n");
+		chan_num =
+			mwifiex_bgscan_create_channel_list(priv, bgscan_cfg_in,
+							   chan_list_tlv->
+							   chan_scan_param);
+		le16_add_cpu(&chan_list_tlv->header.len,
+			     chan_num *
+			     sizeof(chan_list_tlv->chan_scan_param[0]));
+	}
+
+	tlv_pos += (sizeof(chan_list_tlv->header)
+			+ le16_to_cpu(chan_list_tlv->header.len));
+
+	if (bgscan_cfg_in->start_later) {
+		start_later_tlv =
+			(struct mwifiex_ie_types_bgscan_start_later *)tlv_pos;
+		start_later_tlv->header.type =
+			cpu_to_le16(TLV_TYPE_BGSCAN_START_LATER);
+		start_later_tlv->header.len =
+			cpu_to_le16(sizeof(start_later_tlv->start_later));
+		start_later_tlv->start_later =
+			cpu_to_le16(bgscan_cfg_in->start_later);
+
+		tlv_pos += sizeof(start_later_tlv->header) +
+			le16_to_cpu(start_later_tlv->header.len);
+	}
+
+	/* Append vendor specific IE TLV */
+	mwifiex_cmd_append_vsie_tlv(priv, MWIFIEX_VSIE_MASK_BGSCAN, &tlv_pos);
+
+	le16_add_cpu(&cmd->size, tlv_pos - bgscan_config->tlv);
+
+	return 0;
+}
+
+int mwifiex_stop_bg_scan(struct mwifiex_private *priv)
+{
+	struct mwifiex_bg_scan_cfg *bgscan_cfg;
+
+	if (!priv->sched_scanning) {
+		dev_dbg(priv->adapter->dev, "bgscan already stopped!\n");
+		return 0;
+	}
+
+	bgscan_cfg = kzalloc(sizeof(*bgscan_cfg), GFP_KERNEL);
+	if (!bgscan_cfg)
+		return -ENOMEM;
+
+	bgscan_cfg->bss_type = MWIFIEX_BSS_MODE_INFRA;
+	bgscan_cfg->action = MWIFIEX_BGSCAN_ACT_SET;
+	bgscan_cfg->enable = false;
+
+	if (mwifiex_send_cmd(priv, HostCmd_CMD_802_11_BG_SCAN_CONFIG,
+			     HostCmd_ACT_GEN_SET, 0, bgscan_cfg, true)) {
+		kfree(bgscan_cfg);
+		return -EFAULT;
+	}
+
+	kfree(bgscan_cfg);
+	priv->sched_scanning = false;
+
+	return 0;
+}
+
 static void
 mwifiex_update_chan_statistics(struct mwifiex_private *priv,
 			       struct mwifiex_ietypes_chanstats *tlv_stat)
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c
index e486867..60f3ded 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c
@@ -1873,6 +1873,10 @@ int mwifiex_sta_prepare_cmd(struct mwifiex_private *priv, uint16_t cmd_no,
 	case HostCmd_CMD_802_11_SCAN:
 		ret = mwifiex_cmd_802_11_scan(cmd_ptr, data_buf);
 		break;
+	case HostCmd_CMD_802_11_BG_SCAN_CONFIG:
+		ret = mwifiex_cmd_802_11_bg_scan_config(priv, cmd_ptr,
+							data_buf);
+		break;
 	case HostCmd_CMD_802_11_BG_SCAN_QUERY:
 		ret = mwifiex_cmd_802_11_bg_scan_query(cmd_ptr);
 		break;
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c
index 9ac7aa2..4b23d3b 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c
@@ -1076,9 +1076,12 @@ int mwifiex_process_sta_cmdresp(struct mwifiex_private *priv, u16 cmdresp_no,
 		break;
 	case HostCmd_CMD_802_11_BG_SCAN_QUERY:
 		ret = mwifiex_ret_802_11_scan(priv, resp);
+		cfg80211_sched_scan_results(priv->wdev.wiphy);
 		mwifiex_dbg(adapter, CMD,
 			    "info: CMD_RESP: BG_SCAN result is ready!\n");
 		break;
+	case HostCmd_CMD_802_11_BG_SCAN_CONFIG:
+		break;
 	case HostCmd_CMD_TXPWR_CFG:
 		ret = mwifiex_ret_tx_power_cfg(priv, resp);
 		break;
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_event.c b/drivers/net/wireless/marvell/mwifiex/sta_event.c
index 23bae87..fd8061c 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_event.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_event.c
@@ -688,6 +688,13 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv)
 				       HostCmd_ACT_GEN_GET, 0, NULL, false);
 		break;
 
+	case EVENT_BG_SCAN_STOPPED:
+		dev_dbg(adapter->dev, "event: BGS_STOPPED\n");
+		cfg80211_sched_scan_stopped(priv->wdev.wiphy);
+		if (priv->sched_scanning)
+			priv->sched_scanning = false;
+		break;
+
 	case EVENT_PORT_RELEASE:
 		mwifiex_dbg(adapter, EVENT, "event: PORT RELEASE\n");
 		priv->port_open = true;
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
index 210b257..2cc1a32 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
@@ -504,6 +504,13 @@ int mwifiex_enable_hs(struct mwifiex_adapter *adapter)
 		}
 	}
 
+	priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA);
+	if (priv && priv->sched_scanning) {
+		dev_dbg(adapter->dev, "aborting bgscan!\n");
+		mwifiex_stop_bg_scan(priv);
+		cfg80211_sched_scan_stopped(priv->wdev.wiphy);
+	}
+
 	if (adapter->hs_activated) {
 		mwifiex_dbg(adapter, CMD,
 			    "cmd: HS Already activated\n");
-- 
cgit v0.10.2


From 5323b53d80fc17604b2c60b725af116827330b2c Mon Sep 17 00:00:00 2001
From: chunfan chen <jeffc@marvell.com>
Date: Wed, 13 Jan 2016 01:26:53 -0800
Subject: mwifiex: add wowlan info messages

This patch adds informative messages in wake up on
magic packet, disconnect, pattern configuration paths.

Signed-off-by: chunfan chen <jeffc@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index c27c6cc..6d36d08 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -3171,10 +3171,12 @@ static int mwifiex_set_wowlan_mef_entry(struct mwifiex_private *priv,
 				sizeof(byte_seq));
 		mef_entry->filter[filt_num].filt_type = TYPE_EQ;
 
-		if (first_pat)
+		if (first_pat) {
 			first_pat = false;
-		else
+			mwifiex_dbg(priv->adapter, INFO, "Wake on patterns\n");
+		} else {
 			mef_entry->filter[filt_num].filt_action = TYPE_AND;
+		}
 
 		filt_num++;
 	}
@@ -3200,6 +3202,7 @@ static int mwifiex_set_wowlan_mef_entry(struct mwifiex_private *priv,
 		mef_entry->filter[filt_num].offset = 56;
 		mef_entry->filter[filt_num].filt_type = TYPE_EQ;
 		mef_entry->filter[filt_num].filt_action = TYPE_OR;
+		mwifiex_dbg(priv->adapter, INFO, "Wake on magic packet\n");
 	}
 	return ret;
 }
@@ -3295,6 +3298,7 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 				    "Failed to set HS params\n");
 			return ret;
 		}
+		mwifiex_dbg(priv->adapter, INFO, "Wake on device disconnect\n");
 	}
 
 	return ret;
-- 
cgit v0.10.2


From 7d7f07d8c5d35ebfb207650b0dbcf437dad76bab Mon Sep 17 00:00:00 2001
From: chunfan chen <jeffc@marvell.com>
Date: Wed, 13 Jan 2016 01:26:54 -0800
Subject: mwifiex: add wowlan net-detect support

This patch adds support for wakeup when configured
network is detected.

Signed-off-by: chunfan chen <jeffc@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index 6d36d08..a1b8d89 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -3273,7 +3273,7 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 
 	priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA);
 
-	if (!priv->media_connected) {
+	if (!priv->media_connected && !wowlan->nd_config) {
 		mwifiex_dbg(adapter, ERROR,
 			    "Can not configure WOWLAN in disconnected state\n");
 		return 0;
@@ -3285,22 +3285,32 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 		return ret;
 	}
 
+	memset(&hs_cfg, 0, sizeof(hs_cfg));
+	hs_cfg.conditions = le32_to_cpu(adapter->hs_cfg.conditions);
+
+	if (wowlan->nd_config) {
+		mwifiex_dbg(adapter, INFO, "Wake on net detect\n");
+		hs_cfg.conditions |= HS_CFG_COND_MAC_EVENT;
+		mwifiex_cfg80211_sched_scan_start(wiphy, priv->netdev,
+						  wowlan->nd_config);
+	}
+
 	if (wowlan->disconnect) {
-		memset(&hs_cfg, 0, sizeof(hs_cfg));
-		hs_cfg.is_invoke_hostcmd = false;
-		hs_cfg.conditions = HS_CFG_COND_MAC_EVENT;
-		hs_cfg.gpio = adapter->hs_cfg.gpio;
-		hs_cfg.gap = adapter->hs_cfg.gap;
-		ret = mwifiex_set_hs_params(priv, HostCmd_ACT_GEN_SET,
-					    MWIFIEX_SYNC_CMD, &hs_cfg);
-		if (ret) {
-			mwifiex_dbg(adapter, ERROR,
-				    "Failed to set HS params\n");
-			return ret;
-		}
+		hs_cfg.conditions |= HS_CFG_COND_MAC_EVENT;
 		mwifiex_dbg(priv->adapter, INFO, "Wake on device disconnect\n");
 	}
 
+	hs_cfg.is_invoke_hostcmd = false;
+	hs_cfg.gpio = adapter->hs_cfg.gpio;
+	hs_cfg.gap = adapter->hs_cfg.gap;
+	ret = mwifiex_set_hs_params(priv, HostCmd_ACT_GEN_SET,
+				    MWIFIEX_SYNC_CMD, &hs_cfg);
+	if (ret) {
+		mwifiex_dbg(adapter, ERROR,
+			    "Failed to set HS params\n");
+		return ret;
+	}
+
 	return ret;
 }
 
@@ -3853,11 +3863,13 @@ static struct cfg80211_ops mwifiex_cfg80211_ops = {
 
 #ifdef CONFIG_PM
 static const struct wiphy_wowlan_support mwifiex_wowlan_support = {
-	.flags = WIPHY_WOWLAN_MAGIC_PKT | WIPHY_WOWLAN_DISCONNECT,
+	.flags = WIPHY_WOWLAN_MAGIC_PKT | WIPHY_WOWLAN_DISCONNECT |
+		WIPHY_WOWLAN_NET_DETECT,
 	.n_patterns = MWIFIEX_MEF_MAX_FILTERS,
 	.pattern_min_len = 1,
 	.pattern_max_len = MWIFIEX_MAX_PATTERN_LEN,
 	.max_pkt_offset = MWIFIEX_MAX_OFFSET_LEN,
+	.max_nd_match_sets = MWIFIEX_MAX_ND_MATCH_SETS,
 };
 #endif
 
diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h
index 07bdc2a..81b491a 100644
--- a/drivers/net/wireless/marvell/mwifiex/fw.h
+++ b/drivers/net/wireless/marvell/mwifiex/fw.h
@@ -543,6 +543,8 @@ enum P2P_MODES {
 
 #define MWIFIEX_MAX_PATTERN_LEN		40
 #define MWIFIEX_MAX_OFFSET_LEN		100
+#define MWIFIEX_MAX_ND_MATCH_SETS	10
+
 #define STACK_NBYTES			100
 #define TYPE_DNUM			1
 #define TYPE_BYTESEQ			2
diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c
index a99b72b..3cfa946 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.c
+++ b/drivers/net/wireless/marvell/mwifiex/main.c
@@ -132,6 +132,13 @@ static int mwifiex_unregister(struct mwifiex_adapter *adapter)
 		}
 	}
 
+	if (adapter->nd_info) {
+		for (i = 0 ; i < adapter->nd_info->n_matches ; i++)
+			kfree(adapter->nd_info->matches[i]);
+		kfree(adapter->nd_info);
+		adapter->nd_info = NULL;
+	}
+
 	vfree(adapter->chan_stats);
 	kfree(adapter);
 	return 0;
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 5f5bcf8..bf9b24a 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -1002,6 +1002,7 @@ struct mwifiex_adapter {
 	u8 active_scan_triggered;
 	bool usb_mc_status;
 	bool usb_mc_setup;
+	struct cfg80211_wowlan_nd_info *nd_info;
 };
 
 void mwifiex_process_tx_queue(struct mwifiex_adapter *adapter);
diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c
index d4e2143..2702bd9 100644
--- a/drivers/net/wireless/marvell/mwifiex/scan.c
+++ b/drivers/net/wireless/marvell/mwifiex/scan.c
@@ -2092,6 +2092,8 @@ int mwifiex_ret_802_11_scan(struct mwifiex_private *priv,
 	u8 is_bgscan_resp;
 	__le64 fw_tsf = 0;
 	u8 *radio_type;
+	struct cfg80211_wowlan_nd_match *pmatch;
+	struct cfg80211_sched_scan_request *nd_config = NULL;
 
 	is_bgscan_resp = (le16_to_cpu(resp->command)
 			  == HostCmd_CMD_802_11_BG_SCAN_QUERY);
@@ -2154,6 +2156,21 @@ int mwifiex_ret_802_11_scan(struct mwifiex_private *priv,
 					     (struct mwifiex_ie_types_data **)
 					     &chan_band_tlv);
 
+#ifdef CONFIG_PM
+	if (priv->wdev.wiphy->wowlan_config)
+		nd_config = priv->wdev.wiphy->wowlan_config->nd_config;
+#endif
+
+	if (nd_config) {
+		adapter->nd_info =
+			kzalloc(sizeof(struct cfg80211_wowlan_nd_match) +
+				sizeof(struct cfg80211_wowlan_nd_match *) *
+				scan_rsp->number_of_sets, GFP_ATOMIC);
+
+		if (adapter->nd_info)
+			adapter->nd_info->n_matches = scan_rsp->number_of_sets;
+	}
+
 	for (idx = 0; idx < scan_rsp->number_of_sets && bytes_left; idx++) {
 		/*
 		 * If the TSF TLV was appended to the scan results, save this
@@ -2172,6 +2189,23 @@ int mwifiex_ret_802_11_scan(struct mwifiex_private *priv,
 			radio_type = NULL;
 		}
 
+		if (chan_band_tlv && adapter->nd_info) {
+			adapter->nd_info->matches[idx] =
+				kzalloc(sizeof(*pmatch) +
+				sizeof(u32), GFP_ATOMIC);
+
+			pmatch = adapter->nd_info->matches[idx];
+
+			if (!pmatch) {
+				memset(pmatch, 0, sizeof(*pmatch));
+				if (chan_band_tlv) {
+					pmatch->n_channels = 1;
+					pmatch->channels[0] =
+						chan_band->chan_number;
+				}
+			}
+		}
+
 		ret = mwifiex_parse_single_response_buf(priv, &bss_info,
 							&bytes_left,
 							le64_to_cpu(fw_tsf),
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
index 2cc1a32..7277c24 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
@@ -505,10 +505,17 @@ int mwifiex_enable_hs(struct mwifiex_adapter *adapter)
 	}
 
 	priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA);
+
 	if (priv && priv->sched_scanning) {
-		dev_dbg(adapter->dev, "aborting bgscan!\n");
-		mwifiex_stop_bg_scan(priv);
-		cfg80211_sched_scan_stopped(priv->wdev.wiphy);
+#ifdef CONFIG_PM
+		if (!priv->wdev.wiphy->wowlan_config->nd_config) {
+#endif
+			mwifiex_dbg(adapter, CMD, "aborting bgscan!\n");
+			mwifiex_stop_bg_scan(priv);
+			cfg80211_sched_scan_stopped(priv->wdev.wiphy);
+#ifdef CONFIG_PM
+		}
+#endif
 	}
 
 	if (adapter->hs_activated) {
-- 
cgit v0.10.2


From 8de00f1b1c7f37e523430efba8e4257b2761df7c Mon Sep 17 00:00:00 2001
From: chunfan chen <jeffc@marvell.com>
Date: Wed, 13 Jan 2016 01:26:55 -0800
Subject: mwifiex: report wakeup reason to cfg80211

This patch adds code to report wakeup reason to cfg80211
when system is resumed.

Signed-off-by: chunfan chen <jeffc@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index a1b8d89..c80d9e2 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -3316,6 +3316,64 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 
 static int mwifiex_cfg80211_resume(struct wiphy *wiphy)
 {
+	struct mwifiex_adapter *adapter = mwifiex_cfg80211_get_adapter(wiphy);
+	struct mwifiex_private *priv =
+		mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA);
+	struct mwifiex_ds_wakeup_reason wakeup_reason;
+	struct cfg80211_wowlan_wakeup wakeup_report;
+	int i;
+
+	mwifiex_get_wakeup_reason(priv, HostCmd_ACT_GEN_GET, MWIFIEX_SYNC_CMD,
+				  &wakeup_reason);
+	memset(&wakeup_report, 0, sizeof(struct cfg80211_wowlan_wakeup));
+
+	wakeup_report.pattern_idx = -1;
+
+	switch (wakeup_reason.hs_wakeup_reason) {
+	case NO_HSWAKEUP_REASON:
+		break;
+	case BCAST_DATA_MATCHED:
+		break;
+	case MCAST_DATA_MATCHED:
+		break;
+	case UCAST_DATA_MATCHED:
+		break;
+	case MASKTABLE_EVENT_MATCHED:
+		break;
+	case NON_MASKABLE_EVENT_MATCHED:
+		if (wiphy->wowlan_config->disconnect)
+			wakeup_report.disconnect = true;
+		if (wiphy->wowlan_config->nd_config)
+			wakeup_report.net_detect = adapter->nd_info;
+		break;
+	case NON_MASKABLE_CONDITION_MATCHED:
+		break;
+	case MAGIC_PATTERN_MATCHED:
+		if (wiphy->wowlan_config->magic_pkt)
+			wakeup_report.magic_pkt = true;
+		if (wiphy->wowlan_config->n_patterns)
+			wakeup_report.pattern_idx = 1;
+		break;
+	case CONTROL_FRAME_MATCHED:
+		break;
+	case	MANAGEMENT_FRAME_MATCHED:
+		break;
+	default:
+		break;
+	}
+
+	if ((wakeup_reason.hs_wakeup_reason > 0) &&
+	    (wakeup_reason.hs_wakeup_reason <= 7))
+		cfg80211_report_wowlan_wakeup(&priv->wdev, &wakeup_report,
+					      GFP_KERNEL);
+
+	if (adapter->nd_info) {
+		for (i = 0 ; i < adapter->nd_info->n_matches ; i++)
+			kfree(adapter->nd_info->matches[i]);
+		kfree(adapter->nd_info);
+		adapter->nd_info = NULL;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/marvell/mwifiex/cmdevt.c b/drivers/net/wireless/marvell/mwifiex/cmdevt.c
index cb25aa7..a12adee 100644
--- a/drivers/net/wireless/marvell/mwifiex/cmdevt.c
+++ b/drivers/net/wireless/marvell/mwifiex/cmdevt.c
@@ -1657,3 +1657,16 @@ int mwifiex_ret_get_hw_spec(struct mwifiex_private *priv,
 
 	return 0;
 }
+
+/* This function handles the command response of hs wakeup reason
+ * command.
+ */
+int mwifiex_ret_wakeup_reason(struct mwifiex_private *priv,
+			      struct host_cmd_ds_command *resp,
+			      struct host_cmd_ds_wakeup_reason *wakeup_reason)
+{
+	wakeup_reason->wakeup_reason =
+		resp->params.hs_wakeup_reason.wakeup_reason;
+
+	return 0;
+}
diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h
index 81b491a..d293e48 100644
--- a/drivers/net/wireless/marvell/mwifiex/fw.h
+++ b/drivers/net/wireless/marvell/mwifiex/fw.h
@@ -373,6 +373,7 @@ enum MWIFIEX_802_11_PRIVACY_FILTER {
 #define HostCmd_CMD_MGMT_FRAME_REG                    0x010c
 #define HostCmd_CMD_REMAIN_ON_CHAN                    0x010d
 #define HostCmd_CMD_11AC_CFG			      0x0112
+#define HostCmd_CMD_HS_WAKEUP_REASON                  0x0116
 #define HostCmd_CMD_TDLS_CONFIG                       0x0100
 #define HostCmd_CMD_MC_POLICY                         0x0121
 #define HostCmd_CMD_TDLS_OPER                         0x0122
@@ -607,6 +608,20 @@ struct mwifiex_ie_types_data {
 #define MWIFIEX_RXPD_FLAGS_TDLS_PACKET      0x01
 #define MWIFIEX_TXPD_FLAGS_REQ_TX_STATUS    0x20
 
+enum HS_WAKEUP_REASON {
+	NO_HSWAKEUP_REASON = 0,
+	BCAST_DATA_MATCHED,
+	MCAST_DATA_MATCHED,
+	UCAST_DATA_MATCHED,
+	MASKTABLE_EVENT_MATCHED,
+	NON_MASKABLE_EVENT_MATCHED,
+	NON_MASKABLE_CONDITION_MATCHED,
+	MAGIC_PATTERN_MATCHED,
+	CONTROL_FRAME_MATCHED,
+	MANAGEMENT_FRAME_MATCHED,
+	RESERVED
+};
+
 struct txpd {
 	u8 bss_type;
 	u8 bss_num;
@@ -2159,6 +2174,10 @@ struct host_cmd_ds_robust_coex {
 	__le16 reserved;
 } __packed;
 
+struct host_cmd_ds_wakeup_reason {
+	u16  wakeup_reason;
+} __packed;
+
 struct host_cmd_ds_command {
 	__le16 command;
 	__le16 size;
@@ -2231,6 +2250,7 @@ struct host_cmd_ds_command {
 		struct host_cmd_sdio_sp_rx_aggr_cfg sdio_rx_aggr_cfg;
 		struct host_cmd_ds_multi_chan_policy mc_policy;
 		struct host_cmd_ds_robust_coex coex;
+		struct host_cmd_ds_wakeup_reason hs_wakeup_reason;
 	} params;
 } __packed;
 
diff --git a/drivers/net/wireless/marvell/mwifiex/ioctl.h b/drivers/net/wireless/marvell/mwifiex/ioctl.h
index 6333b16..14cfa37 100644
--- a/drivers/net/wireless/marvell/mwifiex/ioctl.h
+++ b/drivers/net/wireless/marvell/mwifiex/ioctl.h
@@ -271,6 +271,10 @@ struct mwifiex_ds_hs_cfg {
 	u32 gap;
 };
 
+struct mwifiex_ds_wakeup_reason {
+	u16  hs_wakeup_reason;
+};
+
 #define DEEP_SLEEP_ON  1
 #define DEEP_SLEEP_OFF 0
 #define DEEP_SLEEP_IDLE_TIME	100
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index bf9b24a..c755be54c 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -1599,6 +1599,12 @@ void mwifiex_drv_info_dump(struct mwifiex_adapter *adapter);
 void mwifiex_upload_device_dump(struct mwifiex_adapter *adapter);
 void *mwifiex_alloc_dma_align_buf(int rx_len, gfp_t flags);
 void mwifiex_queue_main_work(struct mwifiex_adapter *adapter);
+int mwifiex_get_wakeup_reason(struct mwifiex_private *priv, u16 action,
+			      int cmd_type,
+			      struct mwifiex_ds_wakeup_reason *wakeup_reason);
+int mwifiex_ret_wakeup_reason(struct mwifiex_private *priv,
+			      struct host_cmd_ds_command *resp,
+			      struct host_cmd_ds_wakeup_reason *wakeup_reason);
 void mwifiex_coex_ampdu_rxwinsize(struct mwifiex_adapter *adapter);
 void mwifiex_11n_delba(struct mwifiex_private *priv, int tid);
 int mwifiex_send_domain_info_cmd_fw(struct wiphy *wiphy);
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c
index 60f3ded..30f1526 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c
@@ -1813,6 +1813,22 @@ static int mwifiex_cmd_sdio_rx_aggr_cfg(struct host_cmd_ds_command *cmd,
 	return 0;
 }
 
+/* This function prepares command to get HS wakeup reason.
+ *
+ * Preparation includes -
+ *      - Setting command ID, action and proper size
+ *      - Ensuring correct endian-ness
+ */
+static int mwifiex_cmd_get_wakeup_reason(struct mwifiex_private *priv,
+					 struct host_cmd_ds_command *cmd)
+{
+	cmd->command = cpu_to_le16(HostCmd_CMD_HS_WAKEUP_REASON);
+	cmd->size = cpu_to_le16(sizeof(struct host_cmd_ds_wakeup_reason) +
+				S_DS_GEN);
+
+	return 0;
+}
+
 /*
  * This function prepares the commands before sending them to the firmware.
  *
@@ -2067,6 +2083,9 @@ int mwifiex_sta_prepare_cmd(struct mwifiex_private *priv, uint16_t cmd_no,
 		ret = mwifiex_cmd_sdio_rx_aggr_cfg(cmd_ptr, cmd_action,
 						   data_buf);
 		break;
+	case HostCmd_CMD_HS_WAKEUP_REASON:
+		ret = mwifiex_cmd_get_wakeup_reason(priv, cmd_ptr);
+		break;
 	case HostCmd_CMD_MC_POLICY:
 		ret = mwifiex_cmd_set_mc_policy(priv, cmd_ptr, cmd_action,
 						data_buf);
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c
index 4b23d3b..d96523e 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c
@@ -1236,6 +1236,9 @@ int mwifiex_process_sta_cmdresp(struct mwifiex_private *priv, u16 cmdresp_no,
 	case HostCmd_CMD_SDIO_SP_RX_AGGR_CFG:
 		ret = mwifiex_ret_sdio_rx_aggr_cfg(priv, resp);
 		break;
+	case HostCmd_CMD_HS_WAKEUP_REASON:
+		ret = mwifiex_ret_wakeup_reason(priv, resp, data_buf);
+		break;
 	case HostCmd_CMD_TDLS_CONFIG:
 		break;
 	case HostCmd_CMD_ROBUST_COEX:
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
index 7277c24..5cbee58 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
@@ -1465,3 +1465,19 @@ mwifiex_set_gen_ie(struct mwifiex_private *priv, const u8 *ie, int ie_len)
 
 	return 0;
 }
+
+/* This function get Host Sleep wake up reason.
+ *
+ */
+int mwifiex_get_wakeup_reason(struct mwifiex_private *priv, u16 action,
+			      int cmd_type,
+			      struct mwifiex_ds_wakeup_reason *wakeup_reason)
+{
+	int status = 0;
+
+	status = mwifiex_send_cmd(priv, HostCmd_CMD_HS_WAKEUP_REASON,
+				  HostCmd_ACT_GEN_GET, 0, wakeup_reason,
+				  cmd_type == MWIFIEX_SYNC_CMD);
+
+	return status;
+}
-- 
cgit v0.10.2


From fdcab083055d759325c8e0f8999d9e192417fc20 Mon Sep 17 00:00:00 2001
From: Ganapathi Bhat <gbhat@marvell.com>
Date: Wed, 13 Jan 2016 01:26:56 -0800
Subject: mwifiex: add RSSI support for net-detect

This patch adds support for waking up the device on
finding better RSSI. Threshold RSSI value will be
configured by application.

Signed-off-by: Ganapathi Bhat <gbhat@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index c80d9e2..8461553 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -2582,6 +2582,10 @@ mwifiex_cfg80211_sched_scan_start(struct wiphy *wiphy,
 	bgscan_cfg->bss_type = MWIFIEX_BSS_MODE_INFRA;
 	bgscan_cfg->action = MWIFIEX_BGSCAN_ACT_SET;
 	bgscan_cfg->enable = true;
+	if (request->min_rssi_thold != NL80211_SCAN_RSSI_THOLD_OFF) {
+		bgscan_cfg->report_condition |= MWIFIEX_BGSCAN_SSID_RSSI_MATCH;
+		bgscan_cfg->rssi_threshold = request->min_rssi_thold;
+	}
 
 	if (mwifiex_send_cmd(priv, HostCmd_CMD_802_11_BG_SCAN_CONFIG,
 			     HostCmd_ACT_GEN_SET, 0, bgscan_cfg, true)) {
diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h
index d293e48..4af9168 100644
--- a/drivers/net/wireless/marvell/mwifiex/fw.h
+++ b/drivers/net/wireless/marvell/mwifiex/fw.h
@@ -759,6 +759,11 @@ struct mwifiex_ie_types_repeat_count {
 	__le16 repeat_count;
 } __packed;
 
+struct mwifiex_ie_types_min_rssi_threshold {
+	struct mwifiex_ie_types_header header;
+	__le16 rssi_threshold;
+} __packed;
+
 struct mwifiex_ie_types_bgscan_start_later {
 	struct mwifiex_ie_types_header header;
 	__le16 start_later;
diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c
index 2702bd9..fc8d8ca 100644
--- a/drivers/net/wireless/marvell/mwifiex/scan.c
+++ b/drivers/net/wireless/marvell/mwifiex/scan.c
@@ -2260,6 +2260,7 @@ int mwifiex_cmd_802_11_bg_scan_config(struct mwifiex_private *priv,
 	int i;
 	struct mwifiex_ie_types_num_probes *num_probes_tlv;
 	struct mwifiex_ie_types_repeat_count *repeat_count_tlv;
+	struct mwifiex_ie_types_min_rssi_threshold *rssi_threshold_tlv;
 	struct mwifiex_ie_types_bgscan_start_later *start_later_tlv;
 	struct mwifiex_ie_types_wildcard_ssid_params *wildcard_ssid_tlv;
 	struct mwifiex_ie_types_chan_list_param_set *chan_list_tlv;
@@ -2310,6 +2311,20 @@ int mwifiex_cmd_802_11_bg_scan_config(struct mwifiex_private *priv,
 			le16_to_cpu(repeat_count_tlv->header.len);
 	}
 
+	if (bgscan_cfg_in->rssi_threshold) {
+		rssi_threshold_tlv =
+			(struct mwifiex_ie_types_min_rssi_threshold *)tlv_pos;
+		rssi_threshold_tlv->header.type =
+			cpu_to_le16(TLV_TYPE_RSSI_LOW);
+		rssi_threshold_tlv->header.len =
+			cpu_to_le16(sizeof(rssi_threshold_tlv->rssi_threshold));
+		rssi_threshold_tlv->rssi_threshold =
+			cpu_to_le16(bgscan_cfg_in->rssi_threshold);
+
+		tlv_pos += sizeof(rssi_threshold_tlv->header) +
+			le16_to_cpu(rssi_threshold_tlv->header.len);
+	}
+
 	for (i = 0; i < bgscan_cfg_in->num_ssids; i++) {
 		ssid_len = bgscan_cfg_in->ssid_list[i].ssid.ssid_len;
 
-- 
cgit v0.10.2


From a92277bc3bfe7c41cac13ca4a7d5070033543732 Mon Sep 17 00:00:00 2001
From: Amitkumar Karwar <akarwar@marvell.com>
Date: Wed, 13 Jan 2016 01:26:57 -0800
Subject: mwifiex: use SYNC flag for canceling host sleep

Host sleep is cancelled in sdio resume() handler.
Cfg80211's resume handler is immediately called after
this. SYNC flag here ensures that host sleep handshake
gets completed and we have valid "adapter->nd_config"
before we report host wakeup reason in cfg80211's
resume handler.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c
index 33771d3..abf15db 100644
--- a/drivers/net/wireless/marvell/mwifiex/sdio.c
+++ b/drivers/net/wireless/marvell/mwifiex/sdio.c
@@ -181,7 +181,7 @@ static int mwifiex_sdio_resume(struct device *dev)
 
 	/* Disable Host Sleep */
 	mwifiex_cancel_hs(mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA),
-			  MWIFIEX_ASYNC_CMD);
+			  MWIFIEX_SYNC_CMD);
 
 	return 0;
 }
-- 
cgit v0.10.2


From 3955525d5d17118ed8900e9e79753057120eb02a Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 14 Jan 2016 09:39:21 +0200
Subject: iwlwifi: pcie: buffer packets to avoid overflowing Tx queues

When the Tx queues are full above a threshold, we
immediately stop the mac80211's queue to stop getting new
packets. This worked until TSO was enabled.
With TSO, one single packet from mac80211 can use many
descriptors since a large send needs to be split into
several segments.
This means that stopping mac80211's queues is not enough
and we also need to ensure that we don't overflow the Tx
queues with one single packet from mac80211.
Add code to transport layer to do just that. Stop
mac80211's queue as soon as the queue is full above the
same threshold as before, and keep pushing the current
packet along with its segments on the queue, but check
that we don't overflow. If that would happen, buffer the
segments, and send them when there is room in the Tx queue
again. Of course, we first need to send the buffered
segments and only then, wake up mac80211's queues.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index cc3888e..2d8b415 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -280,6 +280,7 @@ struct iwl_txq {
 	bool ampdu;
 	bool block;
 	unsigned long wd_timeout;
+	struct sk_buff_head overflow_q;
 };
 
 static inline dma_addr_t
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index 5262028..b0b0fd9 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -571,6 +571,7 @@ static int iwl_pcie_txq_init(struct iwl_trans *trans, struct iwl_txq *txq,
 		return ret;
 
 	spin_lock_init(&txq->lock);
+	__skb_queue_head_init(&txq->overflow_q);
 
 	/*
 	 * Tell nic where to find circular buffer of Tx Frame Descriptors for
@@ -621,6 +622,13 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 		q->read_ptr = iwl_queue_inc_wrap(q->read_ptr);
 	}
 	txq->active = false;
+
+	while (!skb_queue_empty(&txq->overflow_q)) {
+		struct sk_buff *skb = __skb_dequeue(&txq->overflow_q);
+
+		iwl_op_mode_free_skb(trans->op_mode, skb);
+	}
+
 	spin_unlock_bh(&txq->lock);
 
 	/* just in case - this queue may have been stopped */
@@ -1052,8 +1060,41 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 
 	iwl_pcie_txq_progress(txq);
 
-	if (iwl_queue_space(&txq->q) > txq->q.low_mark)
-		iwl_wake_queue(trans, txq);
+	if (iwl_queue_space(&txq->q) > txq->q.low_mark &&
+	    test_bit(txq_id, trans_pcie->queue_stopped)) {
+		struct sk_buff_head skbs;
+
+		__skb_queue_head_init(&skbs);
+		skb_queue_splice_init(&txq->overflow_q, &skbs);
+
+		/*
+		 * This is tricky: we are in reclaim path which is non
+		 * re-entrant, so noone will try to take the access the
+		 * txq data from that path. We stopped tx, so we can't
+		 * have tx as well. Bottom line, we can unlock and re-lock
+		 * later.
+		 */
+		spin_unlock_bh(&txq->lock);
+
+		while (!skb_queue_empty(&skbs)) {
+			struct sk_buff *skb = __skb_dequeue(&skbs);
+			struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+			u8 dev_cmd_idx = IWL_TRANS_FIRST_DRIVER_DATA + 1;
+			struct iwl_device_cmd *dev_cmd =
+				info->driver_data[dev_cmd_idx];
+
+			/*
+			 * Note that we can very well be overflowing again.
+			 * In that case, iwl_queue_space will be small again
+			 * and we won't wake mac80211's queue.
+			 */
+			iwl_trans_pcie_tx(trans, skb, dev_cmd, txq_id);
+		}
+		spin_lock_bh(&txq->lock);
+
+		if (iwl_queue_space(&txq->q) > txq->q.low_mark)
+			iwl_wake_queue(trans, txq);
+	}
 
 	if (q->read_ptr == q->write_ptr) {
 		IWL_DEBUG_RPM(trans, "Q %d - last tx reclaimed\n", q->id);
@@ -2161,6 +2202,8 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 
 		csum = skb_checksum(skb, offs, skb->len - offs, 0);
 		*(__sum16 *)(skb->data + csum_offs) = csum_fold(csum);
+
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 
 	if (skb_is_nonlinear(skb) &&
@@ -2177,6 +2220,22 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 
 	spin_lock(&txq->lock);
 
+	if (iwl_queue_space(q) < q->high_mark) {
+		iwl_stop_queue(trans, txq);
+
+		/* don't put the packet on the ring, if there is no room */
+		if (unlikely(iwl_queue_space(q) < 3)) {
+			struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+
+			info->driver_data[IWL_TRANS_FIRST_DRIVER_DATA + 1] =
+				dev_cmd;
+			__skb_queue_tail(&txq->overflow_q, skb);
+
+			spin_unlock(&txq->lock);
+			return 0;
+		}
+	}
+
 	/* In AGG mode, the index in the ring must correspond to the WiFi
 	 * sequence number. This is a HW requirements to help the SCD to parse
 	 * the BA.
@@ -2281,12 +2340,6 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	 * At this point the frame is "transmitted" successfully
 	 * and we will get a TX status notification eventually.
 	 */
-	if (iwl_queue_space(q) < q->high_mark) {
-		if (wait_write_ptr)
-			iwl_pcie_txq_inc_wr_ptr(trans, txq);
-		else
-			iwl_stop_queue(trans, txq);
-	}
 	spin_unlock(&txq->lock);
 	return 0;
 out_err:
-- 
cgit v0.10.2


From 7848505416be4e96af14c3387775622d26a13146 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 14 Dec 2015 17:44:11 +0200
Subject: iwlwifi: pcie: add infrastructure for multi-queue rx

The 9000 series devices will support multi rx queues.
Current code has one static rx queue - change it to allocate
a number of queues per the device capability (pre-9000 devices
have the number of rx queues set to one).

Subsequent generalizations are:

Change the code to access an explicit numbered rx queue only
when the queue number is known - when handling interrupt, when
accessing the default queue and when iterating the queues.
The rest of the functions will receive the rx queue as a pointer.

Generalize the warning in allocation failure to consider the
allocator status instead of a single rx queue status.

Move the rx initial pool of memory buffers to be shared among
all the queues and allocated to the default queue on init.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 2d8b415..11ad87f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -57,7 +57,7 @@
 #define RX_POST_REQ_ALLOC 2
 #define RX_CLAIM_REQ_ALLOC 8
 #define RX_POOL_SIZE ((RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC) * RX_NUM_QUEUES)
-#define RX_LOW_WATERMARK 8
+#define RX_PENDING_WATERMARK 16
 
 struct iwl_host_cmd;
 
@@ -103,7 +103,6 @@ struct isr_statistics {
  * @rb_stts: driver's pointer to receive buffer status
  * @rb_stts_dma: bus address of receive buffer status
  * @lock:
- * @pool: initial pool of iwl_rx_mem_buffer for the queue
  * @queue: actual rx queue
  *
  * NOTE:  rx_free and rx_used are used as a FIFO for iwl_rx_mem_buffers
@@ -122,7 +121,6 @@ struct iwl_rxq {
 	struct iwl_rb_status *rb_stts;
 	dma_addr_t rb_stts_dma;
 	spinlock_t lock;
-	struct iwl_rx_mem_buffer pool[RX_QUEUE_SIZE];
 	struct iwl_rx_mem_buffer *queue[RX_QUEUE_SIZE];
 };
 
@@ -298,6 +296,7 @@ struct iwl_tso_hdr_page {
 /**
  * struct iwl_trans_pcie - PCIe transport specific data
  * @rxq: all the RX queue data
+ * @rx_pool: initial pool of iwl_rx_mem_buffer for all the queues
  * @rba: allocator for RX replenishing
  * @drv - pointer to iwl_drv
  * @trans: pointer to the generic transport area
@@ -324,7 +323,8 @@ struct iwl_tso_hdr_page {
  * @fw_mon_size: size of the buffer for the firmware monitor
  */
 struct iwl_trans_pcie {
-	struct iwl_rxq rxq;
+	struct iwl_rxq *rxq;
+	struct iwl_rx_mem_buffer rx_pool[RX_QUEUE_SIZE];
 	struct iwl_rb_allocator rba;
 	struct iwl_trans *trans;
 	struct iwl_drv *drv;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index ccafbd8..f557f3d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -173,10 +173,9 @@ int iwl_pcie_rx_stop(struct iwl_trans *trans)
 /*
  * iwl_pcie_rxq_inc_wr_ptr - Update the write pointer for the RX queue
  */
-static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans)
+static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans,
+				    struct iwl_rxq *rxq)
 {
-	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	u32 reg;
 
 	lockdep_assert_held(&rxq->lock);
@@ -207,18 +206,18 @@ static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans)
 static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
-
-	spin_lock(&rxq->lock);
-
-	if (!rxq->need_update)
-		goto exit_unlock;
+	int i;
 
-	iwl_pcie_rxq_inc_wr_ptr(trans);
-	rxq->need_update = false;
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
 
- exit_unlock:
-	spin_unlock(&rxq->lock);
+		if (!rxq->need_update)
+			continue;
+		spin_lock(&rxq->lock);
+		iwl_pcie_rxq_inc_wr_ptr(trans, rxq);
+		rxq->need_update = false;
+		spin_unlock(&rxq->lock);
+	}
 }
 
 /*
@@ -232,10 +231,8 @@ static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans)
  * also updates the memory address in the firmware to reference the new
  * target buffer.
  */
-static void iwl_pcie_rxq_restock(struct iwl_trans *trans)
+static void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
 {
-	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	struct iwl_rx_mem_buffer *rxb;
 
 	/*
@@ -272,7 +269,7 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans)
 	 * Increment device's write pointer in multiples of 8. */
 	if (rxq->write_actual != (rxq->write & ~0x7)) {
 		spin_lock(&rxq->lock);
-		iwl_pcie_rxq_inc_wr_ptr(trans);
+		iwl_pcie_rxq_inc_wr_ptr(trans, rxq);
 		spin_unlock(&rxq->lock);
 	}
 }
@@ -285,13 +282,9 @@ static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans,
 					   gfp_t priority)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	struct page *page;
 	gfp_t gfp_mask = priority;
 
-	if (rxq->free_count > RX_LOW_WATERMARK)
-		gfp_mask |= __GFP_NOWARN;
-
 	if (trans_pcie->rx_page_order > 0)
 		gfp_mask |= __GFP_COMP;
 
@@ -301,16 +294,13 @@ static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans,
 		if (net_ratelimit())
 			IWL_DEBUG_INFO(trans, "alloc_pages failed, order: %d\n",
 				       trans_pcie->rx_page_order);
-		/* Issue an error if the hardware has consumed more than half
-		 * of its free buffer list and we don't have enough
-		 * pre-allocated buffers.
+		/*
+		 * Issue an error if we don't have enough pre-allocated
+		  * buffers.
 `		 */
-		if (rxq->free_count <= RX_LOW_WATERMARK &&
-		    iwl_rxq_space(rxq) > (RX_QUEUE_SIZE / 2) &&
-		    net_ratelimit())
+		if (!(gfp_mask & __GFP_NOWARN) && net_ratelimit())
 			IWL_CRIT(trans,
-				 "Failed to alloc_pages with GFP_KERNEL. Only %u free buffers remaining.\n",
-				 rxq->free_count);
+				 "Failed to alloc_pages\n");
 		return NULL;
 	}
 	return page;
@@ -325,10 +315,10 @@ static struct page *iwl_pcie_rx_alloc_page(struct iwl_trans *trans,
  * iwl_pcie_rxq_restock. The latter function will update the HW to use the newly
  * allocated buffers.
  */
-static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority)
+static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
+				   struct iwl_rxq *rxq)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	struct iwl_rx_mem_buffer *rxb;
 	struct page *page;
 
@@ -386,41 +376,24 @@ static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority)
 	}
 }
 
-static void iwl_pcie_rxq_free_rbs(struct iwl_trans *trans)
+static void iwl_pcie_free_rbs_pool(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	int i;
 
-	lockdep_assert_held(&rxq->lock);
-
 	for (i = 0; i < RX_QUEUE_SIZE; i++) {
-		if (!rxq->pool[i].page)
+		if (!trans_pcie->rx_pool[i].page)
 			continue;
-		dma_unmap_page(trans->dev, rxq->pool[i].page_dma,
+		dma_unmap_page(trans->dev, trans_pcie->rx_pool[i].page_dma,
 			       PAGE_SIZE << trans_pcie->rx_page_order,
 			       DMA_FROM_DEVICE);
-		__free_pages(rxq->pool[i].page, trans_pcie->rx_page_order);
-		rxq->pool[i].page = NULL;
+		__free_pages(trans_pcie->rx_pool[i].page,
+			     trans_pcie->rx_page_order);
+		trans_pcie->rx_pool[i].page = NULL;
 	}
 }
 
 /*
- * iwl_pcie_rx_replenish - Move all used buffers from rx_used to rx_free
- *
- * When moving to rx_free an page is allocated for the slot.
- *
- * Also restock the Rx queue via iwl_pcie_rxq_restock.
- * This is called only during initialization
- */
-static void iwl_pcie_rx_replenish(struct iwl_trans *trans)
-{
-	iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL);
-
-	iwl_pcie_rxq_restock(trans);
-}
-
-/*
  * iwl_pcie_rx_allocator - Allocates pages in the background for RX queues
  *
  * Allocates for each received request 8 pages
@@ -444,6 +417,11 @@ static void iwl_pcie_rx_allocator(struct iwl_trans *trans)
 	while (pending) {
 		int i;
 		struct list_head local_allocated;
+		gfp_t gfp_mask = GFP_KERNEL;
+
+		/* Do not post a warning if there are only a few requests */
+		if (pending < RX_PENDING_WATERMARK)
+			gfp_mask |= __GFP_NOWARN;
 
 		INIT_LIST_HEAD(&local_allocated);
 
@@ -463,7 +441,7 @@ static void iwl_pcie_rx_allocator(struct iwl_trans *trans)
 			BUG_ON(rxb->page);
 
 			/* Alloc a new receive buffer */
-			page = iwl_pcie_rx_alloc_page(trans, GFP_KERNEL);
+			page = iwl_pcie_rx_alloc_page(trans, gfp_mask);
 			if (!page)
 				continue;
 			rxb->page = page;
@@ -561,38 +539,60 @@ static void iwl_pcie_rx_allocator_work(struct work_struct *data)
 static int iwl_pcie_rx_alloc(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
 	struct device *dev = trans->dev;
+	int i;
+
+	if (WARN_ON(trans_pcie->rxq))
+		return -EINVAL;
 
-	memset(&trans_pcie->rxq, 0, sizeof(trans_pcie->rxq));
+	trans_pcie->rxq = kcalloc(trans->num_rx_queues, sizeof(struct iwl_rxq),
+				  GFP_KERNEL);
+	if (!trans_pcie->rxq)
+		return -EINVAL;
 
-	spin_lock_init(&rxq->lock);
 	spin_lock_init(&rba->lock);
 
-	if (WARN_ON(rxq->bd || rxq->rb_stts))
-		return -EINVAL;
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
 
-	/* Allocate the circular buffer of Read Buffer Descriptors (RBDs) */
-	rxq->bd = dma_zalloc_coherent(dev, sizeof(__le32) * RX_QUEUE_SIZE,
+		spin_lock_init(&rxq->lock);
+		/*
+		 * Allocate the circular buffer of Read Buffer Descriptors
+		 * (RBDs)
+		 */
+		rxq->bd = dma_zalloc_coherent(dev,
+				      sizeof(__le32) * RX_QUEUE_SIZE,
 				      &rxq->bd_dma, GFP_KERNEL);
-	if (!rxq->bd)
-		goto err_bd;
+		if (!rxq->bd)
+			goto err;
 
-	/*Allocate the driver's pointer to receive buffer status */
-	rxq->rb_stts = dma_zalloc_coherent(dev, sizeof(*rxq->rb_stts),
-					   &rxq->rb_stts_dma, GFP_KERNEL);
-	if (!rxq->rb_stts)
-		goto err_rb_stts;
 
+		/*Allocate the driver's pointer to receive buffer status */
+		rxq->rb_stts = dma_zalloc_coherent(dev, sizeof(*rxq->rb_stts),
+						   &rxq->rb_stts_dma,
+						   GFP_KERNEL);
+		if (!rxq->rb_stts)
+			goto err;
+	}
 	return 0;
 
-err_rb_stts:
-	dma_free_coherent(dev, sizeof(__le32) * RX_QUEUE_SIZE,
-			  rxq->bd, rxq->bd_dma);
-	rxq->bd_dma = 0;
-	rxq->bd = NULL;
-err_bd:
+err:
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
+
+		if (rxq->bd)
+			dma_free_coherent(dev, sizeof(__le32) * RX_QUEUE_SIZE,
+					  rxq->bd, rxq->bd_dma);
+		rxq->bd_dma = 0;
+		rxq->bd = NULL;
+
+		if (rxq->rb_stts)
+			dma_free_coherent(trans->dev,
+					  sizeof(struct iwl_rb_status),
+					  rxq->rb_stts, rxq->rb_stts_dma);
+	}
+	kfree(trans_pcie->rxq);
 	return -ENOMEM;
 }
 
@@ -661,17 +661,12 @@ static void iwl_pcie_rx_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq)
 
 static void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq)
 {
-	int i;
-
 	lockdep_assert_held(&rxq->lock);
 
 	INIT_LIST_HEAD(&rxq->rx_free);
 	INIT_LIST_HEAD(&rxq->rx_used);
 	rxq->free_count = 0;
 	rxq->used_count = 0;
-
-	for (i = 0; i < RX_QUEUE_SIZE; i++)
-		list_add(&rxq->pool[i].list, &rxq->rx_used);
 }
 
 static void iwl_pcie_rx_init_rba(struct iwl_rb_allocator *rba)
@@ -709,15 +704,16 @@ static void iwl_pcie_rx_free_rba(struct iwl_trans *trans)
 int iwl_pcie_rx_init(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
+	struct iwl_rxq *def_rxq;
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
 	int i, err;
 
-	if (!rxq->bd) {
+	if (!trans_pcie->rxq) {
 		err = iwl_pcie_rx_alloc(trans);
 		if (err)
 			return err;
 	}
+	def_rxq = trans_pcie->rxq;
 	if (!rba->alloc_wq)
 		rba->alloc_wq = alloc_workqueue("rb_allocator",
 						WQ_HIGHPRI | WQ_UNBOUND, 1);
@@ -731,29 +727,42 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	iwl_pcie_rx_init_rba(rba);
 	spin_unlock(&rba->lock);
 
-	spin_lock(&rxq->lock);
-
 	/* free all first - we might be reconfigured for a different size */
-	iwl_pcie_rxq_free_rbs(trans);
-	iwl_pcie_rx_init_rxb_lists(rxq);
+	iwl_pcie_free_rbs_pool(trans);
 
 	for (i = 0; i < RX_QUEUE_SIZE; i++)
-		rxq->queue[i] = NULL;
+		def_rxq->queue[i] = NULL;
 
-	/* Set us so that we have processed and used all buffers, but have
-	 * not restocked the Rx queue with fresh buffers */
-	rxq->read = rxq->write = 0;
-	rxq->write_actual = 0;
-	memset(rxq->rb_stts, 0, sizeof(*rxq->rb_stts));
-	spin_unlock(&rxq->lock);
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
 
-	iwl_pcie_rx_replenish(trans);
+		spin_lock(&rxq->lock);
+		/*
+		 * Set read write pointer to reflect that we have processed
+		 * and used all buffers, but have not restocked the Rx queue
+		 * with fresh buffers
+		 */
+		rxq->read = 0;
+		rxq->write = 0;
+		rxq->write_actual = 0;
+		memset(rxq->rb_stts, 0, sizeof(*rxq->rb_stts));
 
-	iwl_pcie_rx_hw_init(trans, rxq);
+		iwl_pcie_rx_init_rxb_lists(rxq);
 
-	spin_lock(&rxq->lock);
-	iwl_pcie_rxq_inc_wr_ptr(trans);
-	spin_unlock(&rxq->lock);
+		spin_unlock(&rxq->lock);
+	}
+
+	/* move the entire pool to the default queue ownership */
+	for (i = 0; i < RX_QUEUE_SIZE; i++)
+		list_add(&trans_pcie->rx_pool[i].list, &def_rxq->rx_used);
+
+	iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL, def_rxq);
+	iwl_pcie_rxq_restock(trans, def_rxq);
+	iwl_pcie_rx_hw_init(trans, def_rxq);
+
+	spin_lock(&def_rxq->lock);
+	iwl_pcie_rxq_inc_wr_ptr(trans, def_rxq);
+	spin_unlock(&def_rxq->lock);
 
 	return 0;
 }
@@ -761,12 +770,14 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 void iwl_pcie_rx_free(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
+	int i;
 
-	/*if rxq->bd is NULL, it means that nothing has been allocated,
-	 * exit now */
-	if (!rxq->bd) {
+	/*
+	 * if rxq is NULL, it means that nothing has been allocated,
+	 * exit now
+	 */
+	if (!trans_pcie->rxq) {
 		IWL_DEBUG_INFO(trans, "Free NULL rx context\n");
 		return;
 	}
@@ -781,23 +792,28 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 	iwl_pcie_rx_free_rba(trans);
 	spin_unlock(&rba->lock);
 
-	spin_lock(&rxq->lock);
-	iwl_pcie_rxq_free_rbs(trans);
-	spin_unlock(&rxq->lock);
-
-	dma_free_coherent(trans->dev, sizeof(__le32) * RX_QUEUE_SIZE,
-			  rxq->bd, rxq->bd_dma);
-	rxq->bd_dma = 0;
-	rxq->bd = NULL;
+	iwl_pcie_free_rbs_pool(trans);
+
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
+
+		if (rxq->bd)
+			dma_free_coherent(trans->dev,
+					  sizeof(__le32) * RX_QUEUE_SIZE,
+					  rxq->bd, rxq->bd_dma);
+		rxq->bd_dma = 0;
+		rxq->bd = NULL;
+
+		if (rxq->rb_stts)
+			dma_free_coherent(trans->dev,
+					  sizeof(struct iwl_rb_status),
+					  rxq->rb_stts, rxq->rb_stts_dma);
+		else
+			IWL_DEBUG_INFO(trans,
+				       "Free rxq->rb_stts which is NULL\n");
+	}
 
-	if (rxq->rb_stts)
-		dma_free_coherent(trans->dev,
-				  sizeof(struct iwl_rb_status),
-				  rxq->rb_stts, rxq->rb_stts_dma);
-	else
-		IWL_DEBUG_INFO(trans, "Free rxq->rb_stts which is NULL\n");
-	rxq->rb_stts_dma = 0;
-	rxq->rb_stts = NULL;
+	kfree(trans_pcie->rxq);
 }
 
 /*
@@ -841,11 +857,11 @@ static void iwl_pcie_rx_reuse_rbd(struct iwl_trans *trans,
 }
 
 static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
+				struct iwl_rxq *rxq,
 				struct iwl_rx_mem_buffer *rxb,
 				bool emergency)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
 	struct iwl_txq *txq = &trans_pcie->txq[trans_pcie->cmd_queue];
 	bool page_stolen = false;
 	int max_len = PAGE_SIZE << trans_pcie->rx_page_order;
@@ -975,7 +991,7 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 static void iwl_pcie_rx_handle(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
+	struct iwl_rxq *rxq = &trans_pcie->rxq[0];
 	u32 r, i, j, count = 0;
 	bool emergency = false;
 
@@ -1000,7 +1016,7 @@ restart:
 		rxq->queue[i] = NULL;
 
 		IWL_DEBUG_RX(trans, "rxbuf: HW = %d, SW = %d\n", r, i);
-		iwl_pcie_rx_handle_rb(trans, rxb, emergency);
+		iwl_pcie_rx_handle_rb(trans, rxq, rxb, emergency);
 
 		i = (i + 1) & RX_QUEUE_MASK;
 
@@ -1043,7 +1059,7 @@ restart:
 				if (rxq->used_count < RX_QUEUE_SIZE / 3)
 					emergency = false;
 				spin_unlock(&rxq->lock);
-				iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC);
+				iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
 				spin_lock(&rxq->lock);
 			}
 		}
@@ -1055,7 +1071,7 @@ restart:
 		if (rxq->free_count >=  RX_CLAIM_REQ_ALLOC) {
 			rxq->read = i;
 			spin_unlock(&rxq->lock);
-			iwl_pcie_rxq_restock(trans);
+			iwl_pcie_rxq_restock(trans, rxq);
 			goto restart;
 		}
 	}
@@ -1077,7 +1093,7 @@ restart:
 	 * will be restocked by the next call of iwl_pcie_rxq_restock.
 	 */
 	if (unlikely(emergency && count))
-		iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC);
+		iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
 
 	if (trans_pcie->napi.poll)
 		napi_gro_flush(&trans_pcie->napi, false);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index d60a467..0302aed 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -2001,29 +2001,48 @@ static ssize_t iwl_dbgfs_rx_queue_read(struct file *file,
 {
 	struct iwl_trans *trans = file->private_data;
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
-	char buf[256];
-	int pos = 0;
-	const size_t bufsz = sizeof(buf);
-
-	pos += scnprintf(buf + pos, bufsz - pos, "read: %u\n",
-						rxq->read);
-	pos += scnprintf(buf + pos, bufsz - pos, "write: %u\n",
-						rxq->write);
-	pos += scnprintf(buf + pos, bufsz - pos, "write_actual: %u\n",
-						rxq->write_actual);
-	pos += scnprintf(buf + pos, bufsz - pos, "need_update: %d\n",
-						rxq->need_update);
-	pos += scnprintf(buf + pos, bufsz - pos, "free_count: %u\n",
-						rxq->free_count);
-	if (rxq->rb_stts) {
-		pos += scnprintf(buf + pos, bufsz - pos, "closed_rb_num: %u\n",
-			 le16_to_cpu(rxq->rb_stts->closed_rb_num) &  0x0FFF);
-	} else {
-		pos += scnprintf(buf + pos, bufsz - pos,
-					"closed_rb_num: Not Allocated\n");
+	char *buf;
+	int pos = 0, i, ret;
+	size_t bufsz = sizeof(buf);
+
+	bufsz = sizeof(char) * 121 * trans->num_rx_queues;
+
+	if (!trans_pcie->rxq)
+		return -EAGAIN;
+
+	buf = kzalloc(bufsz, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	for (i = 0; i < trans->num_rx_queues && pos < bufsz; i++) {
+		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
+
+		pos += scnprintf(buf + pos, bufsz - pos, "queue#: %2d\n",
+				 i);
+		pos += scnprintf(buf + pos, bufsz - pos, "\tread: %u\n",
+				 rxq->read);
+		pos += scnprintf(buf + pos, bufsz - pos, "\twrite: %u\n",
+				 rxq->write);
+		pos += scnprintf(buf + pos, bufsz - pos, "\twrite_actual: %u\n",
+				 rxq->write_actual);
+		pos += scnprintf(buf + pos, bufsz - pos, "\tneed_update: %2d\n",
+				 rxq->need_update);
+		pos += scnprintf(buf + pos, bufsz - pos, "\tfree_count: %u\n",
+				 rxq->free_count);
+		if (rxq->rb_stts) {
+			pos += scnprintf(buf + pos, bufsz - pos,
+					 "\tclosed_rb_num: %u\n",
+					 le16_to_cpu(rxq->rb_stts->closed_rb_num) &
+					 0x0FFF);
+		} else {
+			pos += scnprintf(buf + pos, bufsz - pos,
+					 "\tclosed_rb_num: Not Allocated\n");
+	}
 	}
-	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
+	ret = simple_read_from_buffer(user_buf, count, ppos, buf, pos);
+	kfree(buf);
+
+	return ret;
 }
 
 static ssize_t iwl_dbgfs_interrupt_read(struct file *file,
@@ -2188,7 +2207,8 @@ static u32 iwl_trans_pcie_dump_rbs(struct iwl_trans *trans,
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int max_len = PAGE_SIZE << trans_pcie->rx_page_order;
-	struct iwl_rxq *rxq = &trans_pcie->rxq;
+	/* Dump RBs is supported only for pre-9000 devices (1 queue) */
+	struct iwl_rxq *rxq = &trans_pcie->rxq[0];
 	u32 i, r, j, rb_len = 0;
 
 	spin_lock(&rxq->lock);
@@ -2438,11 +2458,12 @@ static struct iwl_trans_dump_data
 	len += sizeof(*data) + (FH_MEM_UPPER_BOUND - FH_MEM_LOWER_BOUND);
 
 	if (dump_rbs) {
+		/* Dump RBs is supported only for pre-9000 devices (1 queue) */
+		struct iwl_rxq *rxq = &trans_pcie->rxq[0];
 		/* RBs */
-		num_rbs = le16_to_cpu(ACCESS_ONCE(
-				      trans_pcie->rxq.rb_stts->closed_rb_num))
+		num_rbs = le16_to_cpu(ACCESS_ONCE(rxq->rb_stts->closed_rb_num))
 				      & 0x0FFF;
-		num_rbs = (num_rbs - trans_pcie->rxq.read) & RX_QUEUE_MASK;
+		num_rbs = (num_rbs - rxq->read) & RX_QUEUE_MASK;
 		len += num_rbs * (sizeof(*data) +
 				  sizeof(struct iwl_fw_error_dump_rb) +
 				  (PAGE_SIZE << trans_pcie->rx_page_order));
-- 
cgit v0.10.2


From 96a6497bc3ed1c19d877e5f4f95f2cfa27448abe Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 23 Dec 2015 15:10:03 +0200
Subject: iwlwifi: pcie: add 9000 series multi queue rx DMA support

The 9000 series introduces several changes in the device
DMA operation.
As the device now supports multi-queue rx, several DMA channels
should be configured.
The flows of providing the device with the allocated RBDs now
changes as well - the device maintains a separate table of used
and free table.

The hardware may use the free table to feed RBDs to any queue.
This requires maintaing a shared table to map returned RBDs to
the original RXB - for that purpose the VID is introduced - an
internal identifier of the RB placed in the lower 12 bits and
returned by HW in the used data.

Another change is the support of 64 bit DMA address.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
index ecbf482..4b93404 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
@@ -138,7 +138,8 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 	.smem_offset = IWL9000_SMEM_OFFSET,				\
 	.smem_len = IWL9000_SMEM_LEN,					\
 	.thermal_params = &iwl9000_tt_params,				\
-	.apmg_not_supported = true
+	.apmg_not_supported = true,					\
+	.mq_rx_supported = true
 
 const struct iwl_cfg iwl9260_2ac_cfg = {
 		.name = "Intel(R) Dual Band Wireless AC 9260",
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index f990481..dad5570 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -311,6 +311,7 @@ struct iwl_pwr_tx_backoff {
  * @dccm2_len: length of the second DCCM
  * @smem_offset: offset from which the SMEM begins
  * @smem_len: the length of SMEM
+ * @mq_rx_supported: multi-queue rx support
  *
  * We enable the driver to be backward compatible wrt. hardware features.
  * API differences in uCode shouldn't be handled here but through TLVs
@@ -362,6 +363,7 @@ struct iwl_cfg {
 	const u32 smem_len;
 	const struct iwl_tt_params *thermal_params;
 	bool apmg_not_supported;
+	bool mq_rx_supported;
 };
 
 /*
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
index 5cc6be92..4ab6682 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
@@ -6,6 +6,7 @@
  * GPL LICENSE SUMMARY
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
+ * Copyright(c) 2015 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -31,6 +32,7 @@
  * BSD LICENSE
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
+ * Copyright(c) 2015 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -312,6 +314,77 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl)
 #define FH_MEM_TFDIB_REG1_ADDR_BITSHIFT	28
 #define FH_MEM_TB_MAX_LENGTH			(0x00020000)
 
+/* 9000 rx series registers */
+
+#define RFH_Q0_FRBDCB_BA_LSB 0xA08000 /* 64 bit address */
+#define RFH_Q_FRBDCB_BA_LSB(q) (RFH_Q0_FRBDCB_BA_LSB + (q) * 8)
+/* Write index table */
+#define RFH_Q0_FRBDCB_WIDX 0xA08080
+#define RFH_Q_FRBDCB_WIDX(q) (RFH_Q0_FRBDCB_WIDX + (q) * 4)
+/* Read index table */
+#define RFH_Q0_FRBDCB_RIDX 0xA080C0
+#define RFH_Q_FRBDCB_RIDX(q) (RFH_Q0_FRBDCB_RIDX + (q) * 4)
+/* Used list table */
+#define RFH_Q0_URBDCB_BA_LSB 0xA08100 /* 64 bit address */
+#define RFH_Q_URBDCB_BA_LSB(q) (RFH_Q0_URBDCB_BA_LSB + (q) * 8)
+/* Write index table */
+#define RFH_Q0_URBDCB_WIDX 0xA08180
+#define RFH_Q_URBDCB_WIDX(q) (RFH_Q0_URBDCB_WIDX + (q) * 4)
+#define RFH_Q0_URBDCB_VAID 0xA081C0
+#define RFH_Q_URBDCB_VAID(q) (RFH_Q0_URBDCB_VAID + (q) * 4)
+/* stts */
+#define RFH_Q0_URBD_STTS_WPTR_LSB 0xA08200 /*64 bits address */
+#define RFH_Q_URBD_STTS_WPTR_LSB(q) (RFH_Q0_URBD_STTS_WPTR_LSB + (q) * 8)
+
+#define RFH_Q0_ORB_WPTR_LSB 0xA08280
+#define RFH_Q_ORB_WPTR_LSB(q) (RFH_Q0_ORB_WPTR_LSB + (q) * 8)
+#define RFH_RBDBUF_RBD0_LSB 0xA08300
+#define RFH_RBDBUF_RBD_LSB(q) (RFH_RBDBUF_RBD0_LSB + (q) * 8)
+
+/* DMA configuration */
+#define RFH_RXF_DMA_CFG 0xA09820
+/* RB size */
+#define RFH_RXF_DMA_RB_SIZE_MASK (0x000F0000) /* bits 16-19 */
+#define RFH_RXF_DMA_RB_SIZE_POS 16
+#define RFH_RXF_DMA_RB_SIZE_1K	(0x1 << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_2K	(0x2 << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_4K	(0x4 << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_8K	(0x8 << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_12K	(0x9 << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_16K	(0xA << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_20K	(0xB << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_24K	(0xC << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_28K	(0xD << RFH_RXF_DMA_RB_SIZE_POS)
+#define RFH_RXF_DMA_RB_SIZE_32K	(0xE << RFH_RXF_DMA_RB_SIZE_POS)
+/* RB Circular Buffer size:defines the table sizes in RBD units */
+#define RFH_RXF_DMA_RBDCB_SIZE_MASK (0x00F00000) /* bits 20-23 */
+#define RFH_RXF_DMA_RBDCB_SIZE_POS 20
+#define RFH_RXF_DMA_RBDCB_SIZE_8	(0x3 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_16	(0x4 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_32	(0x5 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_64	(0x7 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_128	(0x7 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_256	(0x8 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_512	(0x9 << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_1024	(0xA << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_RBDCB_SIZE_2048	(0xB << RFH_RXF_DMA_RBDCB_SIZE_POS)
+#define RFH_RXF_DMA_MIN_RB_SIZE_MASK (0x03000000) /* bit 24-25 */
+#define RFH_RXF_DMA_MIN_RB_SIZE_POS	24
+#define RFH_RXF_DMA_MIN_RB_4_8	(3 << RFH_RXF_DMA_MIN_RB_SIZE_POS)
+#define RFH_RXF_DMA_SINGLE_FRAME_MASK (0x20000000) /* bit 29 */
+#define RFH_DMA_EN_MASK (0xC0000000) /* bits 30-31*/
+#define RFH_DMA_EN_ENABLE_VAL BIT(31)
+
+#define RFH_RXF_RXQ_ACTIVE 0xA0980C
+
+#define RFH_GEN_CFG	0xA09800
+#define RFH_GEN_CFG_DEFAULT_RXQ_NUM_MASK 0xF00
+#define RFH_GEN_CFG_SERVICE_DMA_SNOOP BIT(0)
+#define RFH_GEN_CFG_RFH_DMA_SNOOP BIT(1)
+#define DEFAULT_RXQ_NUM 8
+
+/* end of 9000 rx series registers */
+
 /* TFDB  Area - TFDs buffer table */
 #define FH_MEM_TFDIB_DRAM_ADDR_LSB_MSK      (0xFFFFFFFF)
 #define FH_TFDIB_LOWER_BOUND       (FH_MEM_LOWER_BOUND + 0x900)
@@ -434,6 +507,10 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl)
  */
 #define FH_TX_CHICKEN_BITS_SCD_AUTO_RETRY_EN	(0x00000002)
 
+#define MQ_RX_TABLE_SIZE		512
+#define MQ_RX_TABLE_MASK		(MQ_RX_TABLE_SIZE - 1)
+#define MQ_RX_POOL_SIZE			MQ_RX_TABLE_MASK
+
 #define RX_QUEUE_SIZE                         256
 #define RX_QUEUE_MASK                         255
 #define RX_QUEUE_SIZE_LOG                     8
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 11ad87f..bdda702 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -56,7 +56,6 @@
 #define RX_NUM_QUEUES 1
 #define RX_POST_REQ_ALLOC 2
 #define RX_CLAIM_REQ_ALLOC 8
-#define RX_POOL_SIZE ((RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC) * RX_NUM_QUEUES)
 #define RX_PENDING_WATERMARK 16
 
 struct iwl_host_cmd;
@@ -64,9 +63,16 @@ struct iwl_host_cmd;
 /*This file includes the declaration that are internal to the
  * trans_pcie layer */
 
+/**
+ * struct iwl_rx_mem_buffer
+ * @page_dma: bus address of rxb page
+ * @page: driver's pointer to the rxb page
+ * @vid: index of this rxb in the global table
+ */
 struct iwl_rx_mem_buffer {
 	dma_addr_t page_dma;
 	struct page *page;
+	u16 vid;
 	struct list_head list;
 };
 
@@ -90,8 +96,12 @@ struct isr_statistics {
 
 /**
  * struct iwl_rxq - Rx queue
- * @bd: driver's pointer to buffer of receive buffer descriptors (rbd)
+ * @id: queue index
+ * @bd: driver's pointer to buffer of receive buffer descriptors (rbd).
+ *	Address size is 32 bit in pre-9000 devices and 64 bit in 9000 devices.
  * @bd_dma: bus address of buffer of receive buffer descriptors (rbd)
+ * @ubd: driver's pointer to buffer of used receive buffer descriptors (rbd)
+ * @ubd_dma: physical address of buffer of used receive buffer descriptors (rbd)
  * @read: Shared index to newest available Rx buffer
  * @write: Shared index to oldest written Rx packet
  * @free_count: Number of pre-allocated buffers in rx_free
@@ -103,18 +113,22 @@ struct isr_statistics {
  * @rb_stts: driver's pointer to receive buffer status
  * @rb_stts_dma: bus address of receive buffer status
  * @lock:
- * @queue: actual rx queue
+ * @queue: actual rx queue. Not used for multi-rx queue.
  *
  * NOTE:  rx_free and rx_used are used as a FIFO for iwl_rx_mem_buffers
  */
 struct iwl_rxq {
-	__le32 *bd;
+	int id;
+	void *bd;
 	dma_addr_t bd_dma;
+	__le32 *used_bd;
+	dma_addr_t used_bd_dma;
 	u32 read;
 	u32 write;
 	u32 free_count;
 	u32 used_count;
 	u32 write_actual;
+	u32 queue_size;
 	struct list_head rx_free;
 	struct list_head rx_used;
 	bool need_update;
@@ -126,7 +140,6 @@ struct iwl_rxq {
 
 /**
  * struct iwl_rb_allocator - Rx allocator
- * @pool: initial pool of allocator
  * @req_pending: number of requests the allcator had not processed yet
  * @req_ready: number of requests honored and ready for claiming
  * @rbd_allocated: RBDs with pages allocated and ready to be handled to
@@ -138,7 +151,6 @@ struct iwl_rxq {
  * @rx_alloc: work struct for background calls
  */
 struct iwl_rb_allocator {
-	struct iwl_rx_mem_buffer pool[RX_POOL_SIZE];
 	atomic_t req_pending;
 	atomic_t req_ready;
 	struct list_head rbd_allocated;
@@ -297,6 +309,7 @@ struct iwl_tso_hdr_page {
  * struct iwl_trans_pcie - PCIe transport specific data
  * @rxq: all the RX queue data
  * @rx_pool: initial pool of iwl_rx_mem_buffer for all the queues
+ * @global_table: table mapping received VID from hw to rxb
  * @rba: allocator for RX replenishing
  * @drv - pointer to iwl_drv
  * @trans: pointer to the generic transport area
@@ -324,7 +337,8 @@ struct iwl_tso_hdr_page {
  */
 struct iwl_trans_pcie {
 	struct iwl_rxq *rxq;
-	struct iwl_rx_mem_buffer rx_pool[RX_QUEUE_SIZE];
+	struct iwl_rx_mem_buffer rx_pool[MQ_RX_POOL_SIZE];
+	struct iwl_rx_mem_buffer *global_table[MQ_RX_TABLE_SIZE];
 	struct iwl_rb_allocator rba;
 	struct iwl_trans *trans;
 	struct iwl_drv *drv;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index f557f3d..a385f3c 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -140,8 +140,8 @@
  */
 static int iwl_rxq_space(const struct iwl_rxq *rxq)
 {
-	/* Make sure RX_QUEUE_SIZE is a power of 2 */
-	BUILD_BUG_ON(RX_QUEUE_SIZE & (RX_QUEUE_SIZE - 1));
+	/* Make sure rx queue size is a power of 2 */
+	WARN_ON(rxq->queue_size & (rxq->queue_size - 1));
 
 	/*
 	 * There can be up to (RX_QUEUE_SIZE - 1) free slots, to avoid ambiguity
@@ -149,7 +149,7 @@ static int iwl_rxq_space(const struct iwl_rxq *rxq)
 	 * The following is equivalent to modulo by RX_QUEUE_SIZE and is well
 	 * defined for negative dividends.
 	 */
-	return (rxq->read - rxq->write - 1) & (RX_QUEUE_SIZE - 1);
+	return (rxq->read - rxq->write - 1) & (rxq->queue_size - 1);
 }
 
 /*
@@ -160,6 +160,12 @@ static inline __le32 iwl_pcie_dma_addr2rbd_ptr(dma_addr_t dma_addr)
 	return cpu_to_le32((u32)(dma_addr >> 8));
 }
 
+static void iwl_pcie_write_prph_64(struct iwl_trans *trans, u64 ofs, u64 val)
+{
+	iwl_write_prph(trans, ofs, val & 0xffffffff);
+	iwl_write_prph(trans, ofs + 4, val >> 32);
+}
+
 /*
  * iwl_pcie_rx_stop - stops the Rx DMA
  */
@@ -200,7 +206,11 @@ static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans,
 	}
 
 	rxq->write_actual = round_down(rxq->write, 8);
-	iwl_write32(trans, FH_RSCSR_CHNL0_WPTR, rxq->write_actual);
+	if (trans->cfg->mq_rx_supported)
+		iwl_write_prph(trans, RFH_Q_FRBDCB_WIDX(rxq->id),
+			       rxq->write_actual);
+	else
+		iwl_write32(trans, FH_RSCSR_CHNL0_WPTR, rxq->write_actual);
 }
 
 static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans)
@@ -220,6 +230,51 @@ static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans)
 	}
 }
 
+static void iwl_pcie_rxq_mq_restock(struct iwl_trans *trans,
+				    struct iwl_rxq *rxq)
+{
+	struct iwl_rx_mem_buffer *rxb;
+
+	/*
+	 * If the device isn't enabled - no need to try to add buffers...
+	 * This can happen when we stop the device and still have an interrupt
+	 * pending. We stop the APM before we sync the interrupts because we
+	 * have to (see comment there). On the other hand, since the APM is
+	 * stopped, we cannot access the HW (in particular not prph).
+	 * So don't try to restock if the APM has been already stopped.
+	 */
+	if (!test_bit(STATUS_DEVICE_ENABLED, &trans->status))
+		return;
+
+	spin_lock(&rxq->lock);
+	while (rxq->free_count) {
+		__le64 *bd = (__le64 *)rxq->bd;
+
+		/* Get next free Rx buffer, remove from free list */
+		rxb = list_first_entry(&rxq->rx_free, struct iwl_rx_mem_buffer,
+				       list);
+		list_del(&rxb->list);
+
+		/* 12 first bits are expected to be empty */
+		WARN_ON(rxb->page_dma & DMA_BIT_MASK(12));
+		/* Point to Rx buffer via next RBD in circular buffer */
+		bd[rxq->write] = cpu_to_le64(rxb->page_dma | rxb->vid);
+		rxq->write = (rxq->write + 1) & MQ_RX_TABLE_MASK;
+		rxq->free_count--;
+	}
+	spin_unlock(&rxq->lock);
+
+	/*
+	 * If we've added more space for the firmware to place data, tell it.
+	 * Increment device's write pointer in multiples of 8.
+	 */
+	if (rxq->write_actual != (rxq->write & ~0x7)) {
+		spin_lock(&rxq->lock);
+		iwl_pcie_rxq_inc_wr_ptr(trans, rxq);
+		spin_unlock(&rxq->lock);
+	}
+}
+
 /*
  * iwl_pcie_rxq_restock - refill RX queue from pre-allocated pool
  *
@@ -248,6 +303,7 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
 
 	spin_lock(&rxq->lock);
 	while ((iwl_rxq_space(rxq) > 0) && (rxq->free_count)) {
+		__le32 *bd = (__le32 *)rxq->bd;
 		/* The overwritten rxb must be a used one */
 		rxb = rxq->queue[rxq->write];
 		BUG_ON(rxb && rxb->page);
@@ -258,7 +314,7 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
 		list_del(&rxb->list);
 
 		/* Point to Rx buffer via next RBD in circular buffer */
-		rxq->bd[rxq->write] = iwl_pcie_dma_addr2rbd_ptr(rxb->page_dma);
+		bd[rxq->write] = iwl_pcie_dma_addr2rbd_ptr(rxb->page_dma);
 		rxq->queue[rxq->write] = rxb;
 		rxq->write = (rxq->write + 1) & RX_QUEUE_MASK;
 		rxq->free_count--;
@@ -362,10 +418,6 @@ static void iwl_pcie_rxq_alloc_rbs(struct iwl_trans *trans, gfp_t priority,
 			__free_pages(page, trans_pcie->rx_page_order);
 			return;
 		}
-		/* dma address must be no more than 36 bits */
-		BUG_ON(rxb->page_dma & ~DMA_BIT_MASK(36));
-		/* and also 256 byte aligned! */
-		BUG_ON(rxb->page_dma & DMA_BIT_MASK(8));
 
 		spin_lock(&rxq->lock);
 
@@ -381,7 +433,7 @@ static void iwl_pcie_free_rbs_pool(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
-	for (i = 0; i < RX_QUEUE_SIZE; i++) {
+	for (i = 0; i < MQ_RX_POOL_SIZE; i++) {
 		if (!trans_pcie->rx_pool[i].page)
 			continue;
 		dma_unmap_page(trans->dev, trans_pcie->rx_pool[i].page_dma,
@@ -455,10 +507,6 @@ static void iwl_pcie_rx_allocator(struct iwl_trans *trans)
 				__free_pages(page, trans_pcie->rx_page_order);
 				continue;
 			}
-			/* dma address must be no more than 36 bits */
-			BUG_ON(rxb->page_dma & ~DMA_BIT_MASK(36));
-			/* and also 256 byte aligned! */
-			BUG_ON(rxb->page_dma & DMA_BIT_MASK(8));
 
 			/* move the allocated entry to the out list */
 			list_move(&rxb->list, &local_allocated);
@@ -542,6 +590,8 @@ static int iwl_pcie_rx_alloc(struct iwl_trans *trans)
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
 	struct device *dev = trans->dev;
 	int i;
+	int free_size = trans->cfg->mq_rx_supported ? sizeof(__le64) :
+						      sizeof(__le32);
 
 	if (WARN_ON(trans_pcie->rxq))
 		return -EINVAL;
@@ -557,16 +607,30 @@ static int iwl_pcie_rx_alloc(struct iwl_trans *trans)
 		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
 
 		spin_lock_init(&rxq->lock);
+		if (trans->cfg->mq_rx_supported)
+			rxq->queue_size = MQ_RX_TABLE_SIZE;
+		else
+			rxq->queue_size = RX_QUEUE_SIZE;
+
 		/*
 		 * Allocate the circular buffer of Read Buffer Descriptors
 		 * (RBDs)
 		 */
 		rxq->bd = dma_zalloc_coherent(dev,
-				      sizeof(__le32) * RX_QUEUE_SIZE,
-				      &rxq->bd_dma, GFP_KERNEL);
+					     free_size * rxq->queue_size,
+					     &rxq->bd_dma, GFP_KERNEL);
 		if (!rxq->bd)
 			goto err;
 
+		if (trans->cfg->mq_rx_supported) {
+			rxq->used_bd = dma_zalloc_coherent(dev,
+							   sizeof(__le32) *
+							   rxq->queue_size,
+							   &rxq->used_bd_dma,
+							   GFP_KERNEL);
+			if (!rxq->used_bd)
+				goto err;
+		}
 
 		/*Allocate the driver's pointer to receive buffer status */
 		rxq->rb_stts = dma_zalloc_coherent(dev, sizeof(*rxq->rb_stts),
@@ -582,7 +646,7 @@ err:
 		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
 
 		if (rxq->bd)
-			dma_free_coherent(dev, sizeof(__le32) * RX_QUEUE_SIZE,
+			dma_free_coherent(dev, free_size * rxq->queue_size,
 					  rxq->bd, rxq->bd_dma);
 		rxq->bd_dma = 0;
 		rxq->bd = NULL;
@@ -591,8 +655,15 @@ err:
 			dma_free_coherent(trans->dev,
 					  sizeof(struct iwl_rb_status),
 					  rxq->rb_stts, rxq->rb_stts_dma);
+
+		if (rxq->used_bd)
+			dma_free_coherent(dev, sizeof(__le32) * rxq->queue_size,
+					  rxq->used_bd, rxq->used_bd_dma);
+		rxq->used_bd_dma = 0;
+		rxq->used_bd = NULL;
 	}
 	kfree(trans_pcie->rxq);
+
 	return -ENOMEM;
 }
 
@@ -659,46 +730,82 @@ static void iwl_pcie_rx_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq)
 		iwl_set_bit(trans, CSR_INT_COALESCING, IWL_HOST_INT_OPER_MODE);
 }
 
-static void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq)
+static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq)
 {
-	lockdep_assert_held(&rxq->lock);
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	u32 rb_size, enabled = 0;
+	int i;
 
-	INIT_LIST_HEAD(&rxq->rx_free);
-	INIT_LIST_HEAD(&rxq->rx_used);
-	rxq->free_count = 0;
-	rxq->used_count = 0;
-}
+	switch (trans_pcie->rx_buf_size) {
+	case IWL_AMSDU_4K:
+		rb_size = RFH_RXF_DMA_RB_SIZE_4K;
+		break;
+	case IWL_AMSDU_8K:
+		rb_size = RFH_RXF_DMA_RB_SIZE_8K;
+		break;
+	case IWL_AMSDU_12K:
+		rb_size = RFH_RXF_DMA_RB_SIZE_12K;
+		break;
+	default:
+		WARN_ON(1);
+		rb_size = RFH_RXF_DMA_RB_SIZE_4K;
+	}
 
-static void iwl_pcie_rx_init_rba(struct iwl_rb_allocator *rba)
-{
-	int i;
+	/* Stop Rx DMA */
+	iwl_write_prph(trans, RFH_RXF_DMA_CFG, 0);
+	/* disable free amd used rx queue operation */
+	iwl_write_prph(trans, RFH_RXF_RXQ_ACTIVE, 0);
 
-	lockdep_assert_held(&rba->lock);
+	for (i = 0; i < trans->num_rx_queues; i++) {
+		/* Tell device where to find RBD free table in DRAM */
+		iwl_pcie_write_prph_64(trans, RFH_Q_FRBDCB_BA_LSB(i),
+				       (u64)(rxq->bd_dma));
+		/* Tell device where to find RBD used table in DRAM */
+		iwl_pcie_write_prph_64(trans, RFH_Q_URBDCB_BA_LSB(i),
+				       (u64)(rxq->used_bd_dma));
+		/* Tell device where in DRAM to update its Rx status */
+		iwl_pcie_write_prph_64(trans, RFH_Q_URBD_STTS_WPTR_LSB(i),
+				       rxq->rb_stts_dma);
+		/* Reset device indice tables */
+		iwl_write_prph(trans, RFH_Q_FRBDCB_WIDX(i), 0);
+		iwl_write_prph(trans, RFH_Q_FRBDCB_RIDX(i), 0);
+		iwl_write_prph(trans, RFH_Q_URBDCB_WIDX(i), 0);
+
+		enabled |= BIT(i) | BIT(i + 16);
+	}
 
-	INIT_LIST_HEAD(&rba->rbd_allocated);
-	INIT_LIST_HEAD(&rba->rbd_empty);
+	/* restock default queue */
+	iwl_pcie_rxq_mq_restock(trans, &trans_pcie->rxq[0]);
 
-	for (i = 0; i < RX_POOL_SIZE; i++)
-		list_add(&rba->pool[i].list, &rba->rbd_empty);
+	/*
+	 * Enable Rx DMA
+	 * Single frame mode
+	 * Rx buffer size 4 or 8k or 12k
+	 * Min RB size 4 or 8
+	 * 512 RBDs
+	 */
+	iwl_write_prph(trans, RFH_RXF_DMA_CFG,
+		       RFH_DMA_EN_ENABLE_VAL |
+		       rb_size | RFH_RXF_DMA_SINGLE_FRAME_MASK |
+		       RFH_RXF_DMA_MIN_RB_4_8 |
+		       RFH_RXF_DMA_RBDCB_SIZE_512);
+
+	iwl_write_prph(trans, RFH_GEN_CFG, RFH_GEN_CFG_RFH_DMA_SNOOP |
+					  RFH_GEN_CFG_SERVICE_DMA_SNOOP);
+	iwl_write_prph(trans, RFH_RXF_RXQ_ACTIVE, enabled);
+
+	/* Set interrupt coalescing timer to default (2048 usecs) */
+	iwl_write8(trans, CSR_INT_COALESCING, IWL_HOST_INT_TIMEOUT_DEF);
 }
 
-static void iwl_pcie_rx_free_rba(struct iwl_trans *trans)
+static void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq)
 {
-	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rb_allocator *rba = &trans_pcie->rba;
-	int i;
-
-	lockdep_assert_held(&rba->lock);
+	lockdep_assert_held(&rxq->lock);
 
-	for (i = 0; i < RX_POOL_SIZE; i++) {
-		if (!rba->pool[i].page)
-			continue;
-		dma_unmap_page(trans->dev, rba->pool[i].page_dma,
-			       PAGE_SIZE << trans_pcie->rx_page_order,
-			       DMA_FROM_DEVICE);
-		__free_pages(rba->pool[i].page, trans_pcie->rx_page_order);
-		rba->pool[i].page = NULL;
-	}
+	INIT_LIST_HEAD(&rxq->rx_free);
+	INIT_LIST_HEAD(&rxq->rx_used);
+	rxq->free_count = 0;
+	rxq->used_count = 0;
 }
 
 int iwl_pcie_rx_init(struct iwl_trans *trans)
@@ -706,7 +813,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rxq *def_rxq;
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
-	int i, err;
+	int i, err, num_rbds, allocator_pool_size;
 
 	if (!trans_pcie->rxq) {
 		err = iwl_pcie_rx_alloc(trans);
@@ -722,9 +829,8 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	spin_lock(&rba->lock);
 	atomic_set(&rba->req_pending, 0);
 	atomic_set(&rba->req_ready, 0);
-	/* free all first - we might be reconfigured for a different size */
-	iwl_pcie_rx_free_rba(trans);
-	iwl_pcie_rx_init_rba(rba);
+	INIT_LIST_HEAD(&rba->rbd_allocated);
+	INIT_LIST_HEAD(&rba->rbd_empty);
 	spin_unlock(&rba->lock);
 
 	/* free all first - we might be reconfigured for a different size */
@@ -736,6 +842,8 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	for (i = 0; i < trans->num_rx_queues; i++) {
 		struct iwl_rxq *rxq = &trans_pcie->rxq[i];
 
+		rxq->id = i;
+
 		spin_lock(&rxq->lock);
 		/*
 		 * Set read write pointer to reflect that we have processed
@@ -752,13 +860,29 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 		spin_unlock(&rxq->lock);
 	}
 
-	/* move the entire pool to the default queue ownership */
-	for (i = 0; i < RX_QUEUE_SIZE; i++)
-		list_add(&trans_pcie->rx_pool[i].list, &def_rxq->rx_used);
+	/* move the pool to the default queue and allocator ownerships */
+	num_rbds = trans->cfg->mq_rx_supported ?
+		     MQ_RX_POOL_SIZE : RX_QUEUE_SIZE;
+	allocator_pool_size = trans->num_rx_queues *
+		(RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC);
+	for (i = 0; i < num_rbds; i++) {
+		struct iwl_rx_mem_buffer *rxb = &trans_pcie->rx_pool[i];
+
+		if (i < allocator_pool_size)
+			list_add(&rxb->list, &rba->rbd_empty);
+		else
+			list_add(&rxb->list, &def_rxq->rx_used);
+		trans_pcie->global_table[i] = rxb;
+		rxb->vid = (u16)i;
+	}
 
 	iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL, def_rxq);
-	iwl_pcie_rxq_restock(trans, def_rxq);
-	iwl_pcie_rx_hw_init(trans, def_rxq);
+	if (trans->cfg->mq_rx_supported) {
+		iwl_pcie_rx_mq_hw_init(trans, def_rxq);
+	} else {
+		iwl_pcie_rxq_restock(trans, def_rxq);
+		iwl_pcie_rx_hw_init(trans, def_rxq);
+	}
 
 	spin_lock(&def_rxq->lock);
 	iwl_pcie_rxq_inc_wr_ptr(trans, def_rxq);
@@ -771,6 +895,8 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
+	int free_size = trans->cfg->mq_rx_supported ? sizeof(__le64) :
+					      sizeof(__le32);
 	int i;
 
 	/*
@@ -788,10 +914,6 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 		rba->alloc_wq = NULL;
 	}
 
-	spin_lock(&rba->lock);
-	iwl_pcie_rx_free_rba(trans);
-	spin_unlock(&rba->lock);
-
 	iwl_pcie_free_rbs_pool(trans);
 
 	for (i = 0; i < trans->num_rx_queues; i++) {
@@ -799,7 +921,7 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 
 		if (rxq->bd)
 			dma_free_coherent(trans->dev,
-					  sizeof(__le32) * RX_QUEUE_SIZE,
+					  free_size * rxq->queue_size,
 					  rxq->bd, rxq->bd_dma);
 		rxq->bd_dma = 0;
 		rxq->bd = NULL;
@@ -811,8 +933,14 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 		else
 			IWL_DEBUG_INFO(trans,
 				       "Free rxq->rb_stts which is NULL\n");
-	}
 
+		if (rxq->used_bd)
+			dma_free_coherent(trans->dev,
+					  sizeof(__le32) * rxq->queue_size,
+					  rxq->used_bd, rxq->used_bd_dma);
+		rxq->used_bd_dma = 0;
+		rxq->used_bd = NULL;
+	}
 	kfree(trans_pcie->rxq);
 }
 
@@ -1009,16 +1137,26 @@ restart:
 	while (i != r) {
 		struct iwl_rx_mem_buffer *rxb;
 
-		if (unlikely(rxq->used_count == RX_QUEUE_SIZE / 2))
+		if (unlikely(rxq->used_count == rxq->queue_size / 2))
 			emergency = true;
 
-		rxb = rxq->queue[i];
-		rxq->queue[i] = NULL;
+		if (trans->cfg->mq_rx_supported) {
+			/*
+			 * used_bd is a 32 bit but only 12 are used to retrieve
+			 * the vid
+			 */
+			u16 vid = (u16)le32_to_cpu(rxq->used_bd[i]);
+
+			rxb = trans_pcie->global_table[vid];
+		} else {
+			rxb = rxq->queue[i];
+			rxq->queue[i] = NULL;
+		}
 
 		IWL_DEBUG_RX(trans, "rxbuf: HW = %d, SW = %d\n", r, i);
 		iwl_pcie_rx_handle_rb(trans, rxq, rxb, emergency);
 
-		i = (i + 1) & RX_QUEUE_MASK;
+		i = (i + 1) & (rxq->queue_size - 1);
 
 		/* If we have RX_CLAIM_REQ_ALLOC released rx buffers -
 		 * try to claim the pre-allocated buffers from the allocator */
@@ -1056,7 +1194,7 @@ restart:
 			count++;
 			if (count == 8) {
 				count = 0;
-				if (rxq->used_count < RX_QUEUE_SIZE / 3)
+				if (rxq->used_count < rxq->queue_size / 3)
 					emergency = false;
 				spin_unlock(&rxq->lock);
 				iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
@@ -1071,7 +1209,10 @@ restart:
 		if (rxq->free_count >=  RX_CLAIM_REQ_ALLOC) {
 			rxq->read = i;
 			spin_unlock(&rxq->lock);
-			iwl_pcie_rxq_restock(trans, rxq);
+			if (trans->cfg->mq_rx_supported)
+				iwl_pcie_rxq_mq_restock(trans, rxq);
+			else
+				iwl_pcie_rxq_restock(trans, rxq);
 			goto restart;
 		}
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 0302aed..3581096 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -2403,7 +2403,8 @@ static struct iwl_trans_dump_data
 	u32 len, num_rbs;
 	u32 monitor_len;
 	int i, ptr;
-	bool dump_rbs = test_bit(STATUS_FW_ERROR, &trans->status);
+	bool dump_rbs = test_bit(STATUS_FW_ERROR, &trans->status) &&
+			!trans->cfg->mq_rx_supported;
 
 	/* transport dump header */
 	len = sizeof(*dump_data);
@@ -2562,7 +2563,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 	struct iwl_trans_pcie *trans_pcie;
 	struct iwl_trans *trans;
 	u16 pci_cmd;
-	int ret;
+	int ret, addr_size;
 
 	trans = iwl_trans_alloc(sizeof(struct iwl_trans_pcie),
 				&pdev->dev, cfg, &trans_ops_pcie, 0);
@@ -2600,11 +2601,17 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 				       PCIE_LINK_STATE_CLKPM);
 	}
 
+	if (cfg->mq_rx_supported)
+		addr_size = 64;
+	else
+		addr_size = 36;
+
 	pci_set_master(pdev);
 
-	ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(36));
+	ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(addr_size));
 	if (!ret)
-		ret = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(36));
+		ret = pci_set_consistent_dma_mask(pdev,
+						  DMA_BIT_MASK(addr_size));
 	if (ret) {
 		ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (!ret)
-- 
cgit v0.10.2


From b3ff1270566d41eb8ab2d67844bf17b7fa9fee78 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Wed, 6 Jan 2016 18:40:38 -0200
Subject: iwlwifi: pcie: add initial RTPM support for PCI

Add an initial implementation of runtime power management (RTPM) for
PCI devices.  With this patch, RTPM is only used when wifi is off
(i.e. the wifi interface is down).  This implementation is behind a
new Kconfig flag, IWLWIFI_PCIE_RTPM.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/Kconfig b/drivers/net/wireless/intel/iwlwifi/Kconfig
index 8660677..11932d5 100644
--- a/drivers/net/wireless/intel/iwlwifi/Kconfig
+++ b/drivers/net/wireless/intel/iwlwifi/Kconfig
@@ -99,6 +99,18 @@ config IWLWIFI_UAPSD
 
 	  If unsure, say N.
 
+config IWLWIFI_PCIE_RTPM
+       bool "Enable runtime power management mode for PCIe devices"
+       depends on IWLMVM && PM
+       default false
+       help
+         Say Y here to enable runtime power management for PCIe
+         devices.  If enabled, the device will go into low power mode
+         when idle for a short period of time, allowing for improved
+         power saving during runtime.
+
+	 If unsure, say N.
+
 menu "Debugging Options"
 
 config IWLWIFI_DEBUG
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 6261a68..676d239 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2007 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -66,6 +67,9 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/module.h>
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+#include <linux/pm_runtime.h>
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 #include <linux/pci.h>
 #include <linux/pci-aspm.h>
 #include <linux/acpi.h>
@@ -623,6 +627,13 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto out_free_drv;
 
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	pm_runtime_set_active(&pdev->dev);
+	pm_runtime_set_autosuspend_delay(&pdev->dev,
+					 iwlwifi_mod_params.d0i3_entry_delay);
+	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_allow(&pdev->dev);
+#endif
 	return 0;
 
 out_free_drv:
@@ -689,15 +700,58 @@ static int iwl_pci_resume(struct device *device)
 	return 0;
 }
 
-static SIMPLE_DEV_PM_OPS(iwl_dev_pm_ops, iwl_pci_suspend, iwl_pci_resume);
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+static int iwl_pci_runtime_suspend(struct device *device)
+{
+	struct pci_dev *pdev = to_pci_dev(device);
+	struct iwl_trans *trans = pci_get_drvdata(pdev);
+
+	IWL_DEBUG_RPM(trans, "entering runtime suspend\n");
+
+	/* For now we only allow D0I3 if the device is off */
+	if (test_bit(STATUS_DEVICE_ENABLED, &trans->status))
+		return -EBUSY;
+
+	trans->system_pm_mode = IWL_PLAT_PM_MODE_D0I3;
+
+	iwl_trans_d3_suspend(trans, false);
+
+	return 0;
+}
+
+static int iwl_pci_runtime_resume(struct device *device)
+{
+	struct pci_dev *pdev = to_pci_dev(device);
+	struct iwl_trans *trans = pci_get_drvdata(pdev);
+	enum iwl_d3_status d3_status;
+
+	IWL_DEBUG_RPM(trans, "exiting runtime suspend (resume)\n");
+
+	iwl_trans_d3_resume(trans, &d3_status, false);
+
+	trans->system_pm_mode = IWL_PLAT_PM_MODE_D3;
+
+	return 0;
+}
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
+
+static const struct dev_pm_ops iwl_dev_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(iwl_pci_suspend,
+				iwl_pci_resume)
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	SET_RUNTIME_PM_OPS(iwl_pci_runtime_suspend,
+			   iwl_pci_runtime_resume,
+			   NULL)
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
+};
 
 #define IWL_PM_OPS	(&iwl_dev_pm_ops)
 
-#else
+#else /* CONFIG_PM_SLEEP */
 
 #define IWL_PM_OPS	NULL
 
-#endif
+#endif /* CONFIG_PM_SLEEP */
 
 static struct pci_driver iwl_pci_driver = {
 	.name = DRV_NAME,
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 3581096..db94fe1 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -72,6 +72,9 @@
 #include <linux/bitops.h>
 #include <linux/gfp.h>
 #include <linux/vmalloc.h>
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+#include <linux/pm_runtime.h>
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 
 #include "iwl-drv.h"
 #include "iwl-trans.h"
@@ -1194,6 +1197,9 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 	if (hw_rfkill != was_hw_rfkill)
 		iwl_trans_pcie_rf_kill(trans, hw_rfkill);
 
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	pm_runtime_put_sync(trans->dev);
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 	/* re-take ownership to prevent other users from stealing the deivce */
 	iwl_pcie_prepare_card_hw(trans);
 }
@@ -1353,6 +1359,9 @@ static int _iwl_trans_pcie_start_hw(struct iwl_trans *trans, bool low_power)
 	/* ... rfkill can call stop_device and set it false if needed */
 	iwl_trans_pcie_rf_kill(trans, hw_rfkill);
 
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	pm_runtime_get_sync(trans->dev);
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 	return 0;
 }
 
@@ -1476,6 +1485,10 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	/* TODO: check if this is really needed */
+	pm_runtime_disable(trans->dev);
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 	synchronize_irq(trans_pcie->pci_dev->irq);
 
 	iwl_pcie_tx_free(trans);
@@ -1831,6 +1844,9 @@ void iwl_trans_pcie_ref(struct iwl_trans *trans)
 	spin_lock_irqsave(&trans_pcie->ref_lock, flags);
 	IWL_DEBUG_RPM(trans, "ref_counter: %d\n", trans_pcie->ref_count);
 	trans_pcie->ref_count++;
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	pm_runtime_get(&trans_pcie->pci_dev->dev);
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 	spin_unlock_irqrestore(&trans_pcie->ref_lock, flags);
 }
 
@@ -1849,6 +1865,11 @@ void iwl_trans_pcie_unref(struct iwl_trans *trans)
 		return;
 	}
 	trans_pcie->ref_count--;
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	pm_runtime_mark_last_busy(&trans_pcie->pci_dev->dev);
+	pm_runtime_put_autosuspend(&trans_pcie->pci_dev->dev);
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
+
 	spin_unlock_irqrestore(&trans_pcie->ref_lock, flags);
 }
 
@@ -2728,6 +2749,12 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 
 	trans_pcie->inta_mask = CSR_INI_SET_MASK;
 
+#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+	trans->runtime_pm_mode = IWL_PLAT_PM_MODE_D0I3;
+#else
+	trans->runtime_pm_mode = IWL_PLAT_PM_MODE_DISABLED;
+#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
+
 	return trans;
 
 out_free_ict:
-- 
cgit v0.10.2


From 4cbb8e50338a2f2453ba399ce52562e0a111fc1f Mon Sep 17 00:00:00 2001
From: Luciano Coelho <luciano.coelho@intel.com>
Date: Tue, 18 Aug 2015 16:02:38 +0300
Subject: iwlwifi: pcie: add RTPM support when wifi is enabled

Enable runtime power management (RTPM) for PCIe devices and implement
the corresponding functions to enable D0i3 mode when the device is
idle.

Additionally, remove some unnecessary #ifdef's because the RTPM code
will not be called if runtime PM is not configured.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 82fb3a9..fe170a3 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -736,6 +736,11 @@ enum iwl_plat_pm_mode {
 	IWL_PLAT_PM_MODE_D0I3,
 };
 
+/* Max time to wait for trans to become idle/non-idle on d0i3
+ * enter/exit (in msecs).
+ */
+#define IWL_TRANS_IDLE_TIMEOUT 2000
+
 /**
  * struct iwl_trans - transport common data
  *
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 676d239..16b579a5 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -67,9 +67,7 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/module.h>
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
 #include <linux/pm_runtime.h>
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 #include <linux/pci.h>
 #include <linux/pci-aspm.h>
 #include <linux/acpi.h>
@@ -627,13 +625,15 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto out_free_drv;
 
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
-	pm_runtime_set_active(&pdev->dev);
-	pm_runtime_set_autosuspend_delay(&pdev->dev,
+	/* if RTPM is in use, enable it in our device */
+	if (iwl_trans->runtime_pm_mode != IWL_PLAT_PM_MODE_DISABLED) {
+		pm_runtime_set_active(&pdev->dev);
+		pm_runtime_set_autosuspend_delay(&pdev->dev,
 					 iwlwifi_mod_params.d0i3_entry_delay);
-	pm_runtime_use_autosuspend(&pdev->dev);
-	pm_runtime_allow(&pdev->dev);
-#endif
+		pm_runtime_use_autosuspend(&pdev->dev);
+		pm_runtime_allow(&pdev->dev);
+	}
+
 	return 0;
 
 out_free_drv:
@@ -700,17 +700,90 @@ static int iwl_pci_resume(struct device *device)
 	return 0;
 }
 
+int iwl_pci_fw_enter_d0i3(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int ret;
+
+	if (test_bit(STATUS_FW_ERROR, &trans->status))
+		return 0;
+
+	set_bit(STATUS_TRANS_GOING_IDLE, &trans->status);
+
+	/* config the fw */
+	ret = iwl_op_mode_enter_d0i3(trans->op_mode);
+	if (ret == 1) {
+		IWL_DEBUG_RPM(trans, "aborting d0i3 entrance\n");
+		clear_bit(STATUS_TRANS_GOING_IDLE, &trans->status);
+		return -EBUSY;
+	}
+	if (ret)
+		goto err;
+
+	ret = wait_event_timeout(trans_pcie->d0i3_waitq,
+				 test_bit(STATUS_TRANS_IDLE, &trans->status),
+				 msecs_to_jiffies(IWL_TRANS_IDLE_TIMEOUT));
+	if (!ret) {
+		IWL_ERR(trans, "Timeout entering D0i3\n");
+		ret = -ETIMEDOUT;
+		goto err;
+	}
+
+	clear_bit(STATUS_TRANS_GOING_IDLE, &trans->status);
+
+	return 0;
+err:
+	clear_bit(STATUS_TRANS_GOING_IDLE, &trans->status);
+	iwl_trans_fw_error(trans);
+	return ret;
+}
+
+int iwl_pci_fw_exit_d0i3(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int ret;
+
+	/* sometimes a D0i3 entry is not followed through */
+	if (!test_bit(STATUS_TRANS_IDLE, &trans->status))
+		return 0;
+
+	/* config the fw */
+	ret = iwl_op_mode_exit_d0i3(trans->op_mode);
+	if (ret)
+		goto err;
+
+	/* we clear STATUS_TRANS_IDLE only when D0I3_END command is completed */
+
+	ret = wait_event_timeout(trans_pcie->d0i3_waitq,
+				 !test_bit(STATUS_TRANS_IDLE, &trans->status),
+				 msecs_to_jiffies(IWL_TRANS_IDLE_TIMEOUT));
+	if (!ret) {
+		IWL_ERR(trans, "Timeout exiting D0i3\n");
+		ret = -ETIMEDOUT;
+		goto err;
+	}
+
+	return 0;
+err:
+	clear_bit(STATUS_TRANS_IDLE, &trans->status);
+	iwl_trans_fw_error(trans);
+	return ret;
+}
+
 #ifdef CONFIG_IWLWIFI_PCIE_RTPM
 static int iwl_pci_runtime_suspend(struct device *device)
 {
 	struct pci_dev *pdev = to_pci_dev(device);
 	struct iwl_trans *trans = pci_get_drvdata(pdev);
+	int ret;
 
 	IWL_DEBUG_RPM(trans, "entering runtime suspend\n");
 
-	/* For now we only allow D0I3 if the device is off */
-	if (test_bit(STATUS_DEVICE_ENABLED, &trans->status))
-		return -EBUSY;
+	if (test_bit(STATUS_DEVICE_ENABLED, &trans->status)) {
+		ret = iwl_pci_fw_enter_d0i3(trans);
+		if (ret < 0)
+			return ret;
+	}
 
 	trans->system_pm_mode = IWL_PLAT_PM_MODE_D0I3;
 
@@ -729,7 +802,8 @@ static int iwl_pci_runtime_resume(struct device *device)
 
 	iwl_trans_d3_resume(trans, &d3_status, false);
 
-	trans->system_pm_mode = IWL_PLAT_PM_MODE_D3;
+	if (test_bit(STATUS_DEVICE_ENABLED, &trans->status))
+		return iwl_pci_fw_exit_d0i3(trans);
 
 	return 0;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index bdda702..7bc02e0 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -2,6 +2,7 @@
  *
  * Copyright(c) 2003 - 2015 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * Portions of this file are derived from the ipw3945 project, as well
  * as portions of the ieee80211 subsystem header files.
@@ -374,6 +375,7 @@ struct iwl_trans_pcie {
 	bool ucode_write_complete;
 	wait_queue_head_t ucode_write_waitq;
 	wait_queue_head_t wait_command_queue;
+	wait_queue_head_t d0i3_waitq;
 
 	u8 cmd_queue;
 	u8 cmd_fifo;
@@ -594,4 +596,7 @@ static inline int iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans)
 }
 #endif
 
+int iwl_pci_fw_exit_d0i3(struct iwl_trans *trans);
+int iwl_pci_fw_enter_d0i3(struct iwl_trans *trans);
+
 #endif /* __iwl_trans_int_pcie_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index db94fe1..cfdc7f6 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -72,9 +72,7 @@
 #include <linux/bitops.h>
 #include <linux/gfp.h>
 #include <linux/vmalloc.h>
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
 #include <linux/pm_runtime.h>
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 
 #include "iwl-drv.h"
 #include "iwl-trans.h"
@@ -1197,9 +1195,6 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 	if (hw_rfkill != was_hw_rfkill)
 		iwl_trans_pcie_rf_kill(trans, hw_rfkill);
 
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
-	pm_runtime_put_sync(trans->dev);
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 	/* re-take ownership to prevent other users from stealing the deivce */
 	iwl_pcie_prepare_card_hw(trans);
 }
@@ -1359,9 +1354,10 @@ static int _iwl_trans_pcie_start_hw(struct iwl_trans *trans, bool low_power)
 	/* ... rfkill can call stop_device and set it false if needed */
 	iwl_trans_pcie_rf_kill(trans, hw_rfkill);
 
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
-	pm_runtime_get_sync(trans->dev);
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
+	/* Make sure we sync here, because we'll need full access later */
+	if (low_power)
+		pm_runtime_resume(trans->dev);
+
 	return 0;
 }
 
@@ -1485,10 +1481,9 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
 	/* TODO: check if this is really needed */
 	pm_runtime_disable(trans->dev);
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
+
 	synchronize_irq(trans_pcie->pci_dev->irq);
 
 	iwl_pcie_tx_free(trans);
@@ -1844,9 +1839,7 @@ void iwl_trans_pcie_ref(struct iwl_trans *trans)
 	spin_lock_irqsave(&trans_pcie->ref_lock, flags);
 	IWL_DEBUG_RPM(trans, "ref_counter: %d\n", trans_pcie->ref_count);
 	trans_pcie->ref_count++;
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
 	pm_runtime_get(&trans_pcie->pci_dev->dev);
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 	spin_unlock_irqrestore(&trans_pcie->ref_lock, flags);
 }
 
@@ -1865,10 +1858,9 @@ void iwl_trans_pcie_unref(struct iwl_trans *trans)
 		return;
 	}
 	trans_pcie->ref_count--;
-#ifdef CONFIG_IWLWIFI_PCIE_RTPM
+
 	pm_runtime_mark_last_busy(&trans_pcie->pci_dev->dev);
 	pm_runtime_put_autosuspend(&trans_pcie->pci_dev->dev);
-#endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 
 	spin_unlock_irqrestore(&trans_pcie->ref_lock, flags);
 }
@@ -2536,6 +2528,22 @@ static struct iwl_trans_dump_data
 	return dump_data;
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int iwl_trans_pcie_suspend(struct iwl_trans *trans)
+{
+	if (trans->runtime_pm_mode == IWL_PLAT_PM_MODE_D0I3)
+		return iwl_pci_fw_enter_d0i3(trans);
+
+	return 0;
+}
+
+static void iwl_trans_pcie_resume(struct iwl_trans *trans)
+{
+	if (trans->runtime_pm_mode == IWL_PLAT_PM_MODE_D0I3)
+		iwl_pci_fw_exit_d0i3(trans);
+}
+#endif /* CONFIG_PM_SLEEP */
+
 static const struct iwl_trans_ops trans_ops_pcie = {
 	.start_hw = iwl_trans_pcie_start_hw,
 	.op_mode_leave = iwl_trans_pcie_op_mode_leave,
@@ -2546,6 +2554,11 @@ static const struct iwl_trans_ops trans_ops_pcie = {
 	.d3_suspend = iwl_trans_pcie_d3_suspend,
 	.d3_resume = iwl_trans_pcie_d3_resume,
 
+#ifdef CONFIG_PM_SLEEP
+	.suspend = iwl_trans_pcie_suspend,
+	.resume = iwl_trans_pcie_resume,
+#endif /* CONFIG_PM_SLEEP */
+
 	.send_cmd = iwl_trans_pcie_send_hcmd,
 
 	.tx = iwl_trans_pcie_tx,
@@ -2735,6 +2748,8 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 	/* Initialize the wait queue for commands */
 	init_waitqueue_head(&trans_pcie->wait_command_queue);
 
+	init_waitqueue_head(&trans_pcie->d0i3_waitq);
+
 	ret = iwl_pcie_alloc_ict(trans);
 	if (ret)
 		goto out_pci_disable_msi;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index b0b0fd9..c499345b 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -1,7 +1,8 @@
 /******************************************************************************
  *
  * Copyright(c) 2003 - 2014 Intel Corporation. All rights reserved.
- * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * Portions of this file are derived from the ipw3945 project, as well
  * as portions of the ieee80211 subsystem header files.
@@ -1727,6 +1728,20 @@ void iwl_pcie_hcmd_complete(struct iwl_trans *trans,
 		wake_up(&trans_pcie->wait_command_queue);
 	}
 
+	if (meta->flags & CMD_MAKE_TRANS_IDLE) {
+		IWL_DEBUG_INFO(trans, "complete %s - mark trans as idle\n",
+			       iwl_get_cmd_string(trans, cmd->hdr.cmd));
+		set_bit(STATUS_TRANS_IDLE, &trans->status);
+		wake_up(&trans_pcie->d0i3_waitq);
+	}
+
+	if (meta->flags & CMD_WAKE_UP_TRANS) {
+		IWL_DEBUG_INFO(trans, "complete %s - clear trans idle flag\n",
+			       iwl_get_cmd_string(trans, cmd->hdr.cmd));
+		clear_bit(STATUS_TRANS_IDLE, &trans->status);
+		wake_up(&trans_pcie->d0i3_waitq);
+	}
+
 	meta->flags = 0;
 
 	spin_unlock_bh(&txq->lock);
-- 
cgit v0.10.2


From 23ae61282b88873bec2d56c78fea531f8485146c Mon Sep 17 00:00:00 2001
From: Matti Gottlieb <matti.gottlieb@intel.com>
Date: Thu, 31 Dec 2015 18:18:02 +0200
Subject: iwlwifi: mvm: Do not switch to D3 image on suspend

Currently when the driver is configured with wowlan parameters, and enters
D3 mode, the driver switches the FW image to D3, and when it exists
suspend, it reloads the D0 image.

If the firmware supports the consolidation of the D0 & D3 images there is
no need to load the D3 image on suspend, and no need to reload the D0
image on resume.

Do not switch images on suspend / resume, for firmwares that support
consolidated images.

Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
index 29ea1c6..4db4cb7 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
@@ -396,7 +396,7 @@ static int iwlagn_mac_suspend(struct ieee80211_hw *hw,
 	iwl_write32(priv->trans, CSR_UCODE_DRV_GP1_SET,
 		    CSR_UCODE_DRV_GP1_BIT_D3_CFG_COMPLETE);
 
-	iwl_trans_d3_suspend(priv->trans, false);
+	iwl_trans_d3_suspend(priv->trans, false, true);
 
 	goto out;
 
@@ -469,7 +469,7 @@ static int iwlagn_mac_resume(struct ieee80211_hw *hw)
 	/* we'll clear ctx->vif during iwlagn_prepare_restart() */
 	vif = ctx->vif;
 
-	ret = iwl_trans_d3_resume(priv->trans, &d3_status, false);
+	ret = iwl_trans_d3_resume(priv->trans, &d3_status, false, true);
 	if (ret)
 		goto out_unlock;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 84f8aeb..2273908 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -297,6 +297,7 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t;
  * @IWL_UCODE_TLV_CAPA_DQA_SUPPORT: supports dynamic queue allocation (DQA),
  *	which also implies support for the scheduler configuration command
  * @IWL_UCODE_TLV_CAPA_TDLS_CHANNEL_SWITCH: supports TDLS channel switching
+ * @IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG: Consolidated D3-D0 image
  * @IWL_UCODE_TLV_CAPA_HOTSPOT_SUPPORT: supports Hot Spot Command
  * @IWL_UCODE_TLV_CAPA_DC2DC_SUPPORT: supports DC2DC Command
  * @IWL_UCODE_TLV_CAPA_CSUM_SUPPORT: supports TCP Checksum Offload
@@ -330,6 +331,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_QUIET_PERIOD_SUPPORT		= (__force iwl_ucode_tlv_capa_t)11,
 	IWL_UCODE_TLV_CAPA_DQA_SUPPORT			= (__force iwl_ucode_tlv_capa_t)12,
 	IWL_UCODE_TLV_CAPA_TDLS_CHANNEL_SWITCH		= (__force iwl_ucode_tlv_capa_t)13,
+	IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG		= (__force iwl_ucode_tlv_capa_t)17,
 	IWL_UCODE_TLV_CAPA_HOTSPOT_SUPPORT		= (__force iwl_ucode_tlv_capa_t)18,
 	IWL_UCODE_TLV_CAPA_DC2DC_CONFIG_SUPPORT		= (__force iwl_ucode_tlv_capa_t)19,
 	IWL_UCODE_TLV_CAPA_CSUM_SUPPORT			= (__force iwl_ucode_tlv_capa_t)21,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index fe170a3..e6a5e99 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -618,9 +618,9 @@ struct iwl_trans_ops {
 	void (*fw_alive)(struct iwl_trans *trans, u32 scd_addr);
 	void (*stop_device)(struct iwl_trans *trans, bool low_power);
 
-	void (*d3_suspend)(struct iwl_trans *trans, bool test);
+	void (*d3_suspend)(struct iwl_trans *trans, bool test, bool reset);
 	int (*d3_resume)(struct iwl_trans *trans, enum iwl_d3_status *status,
-			 bool test);
+			 bool test, bool reset);
 
 	int (*send_cmd)(struct iwl_trans *trans, struct iwl_host_cmd *cmd);
 
@@ -925,22 +925,23 @@ static inline void iwl_trans_stop_device(struct iwl_trans *trans)
 	_iwl_trans_stop_device(trans, true);
 }
 
-static inline void iwl_trans_d3_suspend(struct iwl_trans *trans, bool test)
+static inline void iwl_trans_d3_suspend(struct iwl_trans *trans, bool test,
+					bool reset)
 {
 	might_sleep();
 	if (trans->ops->d3_suspend)
-		trans->ops->d3_suspend(trans, test);
+		trans->ops->d3_suspend(trans, test, reset);
 }
 
 static inline int iwl_trans_d3_resume(struct iwl_trans *trans,
 				      enum iwl_d3_status *status,
-				      bool test)
+				      bool test, bool reset)
 {
 	might_sleep();
 	if (!trans->ops->d3_resume)
 		return 0;
 
-	return trans->ops->d3_resume(trans, status, test);
+	return trans->ops->d3_resume(trans, status, test, reset);
 }
 
 static inline void iwl_trans_ref(struct iwl_trans *trans)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index d3e21d9..78572ef 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016   Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016   Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -1023,14 +1025,18 @@ iwl_mvm_wowlan_config(struct iwl_mvm *mvm,
 		      struct ieee80211_sta *ap_sta)
 {
 	int ret;
+	bool unified_image = fw_has_capa(&mvm->fw->ucode_capa,
+					 IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG);
 
-	ret = iwl_mvm_switch_to_d3(mvm);
-	if (ret)
-		return ret;
+	if (!unified_image) {
+		ret = iwl_mvm_switch_to_d3(mvm);
+		if (ret)
+			return ret;
 
-	ret = iwl_mvm_d3_reprogram(mvm, vif, ap_sta);
-	if (ret)
-		return ret;
+		ret = iwl_mvm_d3_reprogram(mvm, vif, ap_sta);
+		if (ret)
+			return ret;
+	}
 
 	if (!iwlwifi_mod_params.sw_crypto) {
 		/*
@@ -1072,10 +1078,14 @@ iwl_mvm_netdetect_config(struct iwl_mvm *mvm,
 {
 	struct iwl_wowlan_config_cmd wowlan_config_cmd = {};
 	int ret;
+	bool unified_image = fw_has_capa(&mvm->fw->ucode_capa,
+					 IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG);
 
-	ret = iwl_mvm_switch_to_d3(mvm);
-	if (ret)
-		return ret;
+	if (!unified_image) {
+		ret = iwl_mvm_switch_to_d3(mvm);
+		if (ret)
+			return ret;
+	}
 
 	/* rfkill release can be either for wowlan or netdetect */
 	if (wowlan->rfkill_release)
@@ -1151,6 +1161,8 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 	};
 	int ret;
 	int len __maybe_unused;
+	bool unified_image = fw_has_capa(&mvm->fw->ucode_capa,
+					 IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG);
 
 	if (!wowlan) {
 		/*
@@ -1236,7 +1248,7 @@ static int __iwl_mvm_suspend(struct ieee80211_hw *hw,
 
 	clear_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status);
 
-	iwl_trans_d3_suspend(mvm->trans, test);
+	iwl_trans_d3_suspend(mvm->trans, test, !unified_image);
  out:
 	if (ret < 0) {
 		iwl_mvm_ref(mvm, IWL_MVM_REF_UCODE_DOWN);
@@ -1299,7 +1311,7 @@ int iwl_mvm_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan)
 		__set_bit(D0I3_DEFER_WAKEUP, &mvm->d0i3_suspend_flags);
 		mutex_unlock(&mvm->d0i3_suspend_mutex);
 
-		iwl_trans_d3_suspend(trans, false);
+		iwl_trans_d3_suspend(trans, false, false);
 
 		return 0;
 	}
@@ -2041,9 +2053,14 @@ static void iwl_mvm_d3_disconnect_iter(void *data, u8 *mac,
 static int __iwl_mvm_resume(struct iwl_mvm *mvm, bool test)
 {
 	struct ieee80211_vif *vif = NULL;
-	int ret;
+	int ret = 1;
 	enum iwl_d3_status d3_status;
 	bool keep = false;
+	bool unified_image = fw_has_capa(&mvm->fw->ucode_capa,
+					 IWL_UCODE_TLV_CAPA_CNSLDTD_D3_D0_IMG);
+
+	u32 flags = CMD_ASYNC | CMD_HIGH_PRIO | CMD_SEND_IN_IDLE |
+				    CMD_WAKE_UP_TRANS;
 
 	mutex_lock(&mvm->mutex);
 
@@ -2052,7 +2069,7 @@ static int __iwl_mvm_resume(struct iwl_mvm *mvm, bool test)
 	if (IS_ERR_OR_NULL(vif))
 		goto err;
 
-	ret = iwl_trans_d3_resume(mvm->trans, &d3_status, test);
+	ret = iwl_trans_d3_resume(mvm->trans, &d3_status, test, !unified_image);
 	if (ret)
 		goto err;
 
@@ -2095,17 +2112,28 @@ out_iterate:
 			iwl_mvm_d3_disconnect_iter, keep ? vif : NULL);
 
 out:
-	/* return 1 to reconfigure the device */
+	if (unified_image && !ret) {
+		ret = iwl_mvm_send_cmd_pdu(mvm, D0I3_END_CMD, flags, 0, NULL);
+		if (!ret) /* D3 ended successfully - no need to reset device */
+			return 0;
+	}
+
+	/*
+	 * Reconfigure the device in one of the following cases:
+	 * 1. We are not using a unified image
+	 * 2. We are using a unified image but had an error while exiting D3
+	 */
 	set_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status);
 	set_bit(IWL_MVM_STATUS_D3_RECONFIG, &mvm->status);
-
-	/* We always return 1, which causes mac80211 to do a reconfig
-	 * with IEEE80211_RECONFIG_TYPE_RESTART.  This type of
-	 * reconfig calls iwl_mvm_restart_complete(), where we unref
-	 * the IWL_MVM_REF_UCODE_DOWN, so we need to take the
-	 * reference here.
+	/*
+	 * When switching images we return 1, which causes mac80211
+	 * to do a reconfig with IEEE80211_RECONFIG_TYPE_RESTART.
+	 * This type of reconfig calls iwl_mvm_restart_complete(),
+	 * where we unref the IWL_MVM_REF_UCODE_DOWN, so we need
+	 * to take the reference here.
 	 */
 	iwl_mvm_ref(mvm, IWL_MVM_REF_UCODE_DOWN);
+
 	return 1;
 }
 
@@ -2122,7 +2150,7 @@ static int iwl_mvm_resume_d0i3(struct iwl_mvm *mvm)
 	enum iwl_d3_status d3_status;
 	struct iwl_trans *trans = mvm->trans;
 
-	iwl_trans_d3_resume(trans, &d3_status, false);
+	iwl_trans_d3_resume(trans, &d3_status, false, false);
 
 	/*
 	 * make sure to clear D0I3_DEFER_WAKEUP before
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 9a15642..0476e76 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -1109,7 +1109,7 @@ static int iwl_mvm_scan_umac(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	cmd->general_flags = cpu_to_le32(iwl_mvm_scan_umac_flags(mvm, params,
 								 vif));
 
-	if (type == IWL_MVM_SCAN_SCHED)
+	if (type == IWL_MVM_SCAN_SCHED || type == IWL_MVM_SCAN_NETDETECT)
 		cmd->flags = cpu_to_le32(IWL_UMAC_SCAN_FLAG_PREEMPTIVE);
 
 	if (iwl_mvm_scan_use_ebs(mvm, vif))
@@ -1351,7 +1351,7 @@ int iwl_mvm_sched_scan_start(struct iwl_mvm *mvm,
 
 	if (fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_UMAC_SCAN)) {
 		hcmd.id = iwl_cmd_id(SCAN_REQ_UMAC, IWL_ALWAYS_LONG_GROUP, 0);
-		ret = iwl_mvm_scan_umac(mvm, vif, &params, IWL_MVM_SCAN_SCHED);
+		ret = iwl_mvm_scan_umac(mvm, vif, &params, type);
 	} else {
 		hcmd.id = SCAN_OFFLOAD_REQUEST_CMD;
 		ret = iwl_mvm_scan_lmac(mvm, vif, &params);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 16b579a5..762e7c4 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -787,7 +787,7 @@ static int iwl_pci_runtime_suspend(struct device *device)
 
 	trans->system_pm_mode = IWL_PLAT_PM_MODE_D0I3;
 
-	iwl_trans_d3_suspend(trans, false);
+	iwl_trans_d3_suspend(trans, false, false);
 
 	return 0;
 }
@@ -800,7 +800,7 @@ static int iwl_pci_runtime_resume(struct device *device)
 
 	IWL_DEBUG_RPM(trans, "exiting runtime suspend (resume)\n");
 
-	iwl_trans_d3_resume(trans, &d3_status, false);
+	iwl_trans_d3_resume(trans, &d3_status, false, false);
 
 	if (test_bit(STATUS_DEVICE_ENABLED, &trans->status))
 		return iwl_pci_fw_exit_d0i3(trans);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index cfdc7f6..abe0916 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1219,11 +1219,12 @@ void iwl_trans_pcie_rf_kill(struct iwl_trans *trans, bool state)
 		_iwl_trans_pcie_stop_device(trans, true);
 }
 
-static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test)
+static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test,
+				      bool reset)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
-	if (trans->system_pm_mode == IWL_PLAT_PM_MODE_D0I3) {
+	if (!reset) {
 		/* Enable persistence mode to avoid reset */
 		iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
 			    CSR_HW_IF_CONFIG_REG_PERSIST_MODE);
@@ -1247,7 +1248,7 @@ static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test)
 	iwl_clear_bit(trans, CSR_GP_CNTRL,
 		      CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
 
-	if (trans->system_pm_mode == IWL_PLAT_PM_MODE_D3) {
+	if (reset) {
 		/*
 		 * reset TX queues -- some of their registers reset during S3
 		 * so if we don't reset everything here the D3 image would try
@@ -1261,7 +1262,7 @@ static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test)
 
 static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
 				    enum iwl_d3_status *status,
-				    bool test)
+				    bool test,  bool reset)
 {
 	u32 val;
 	int ret;
@@ -1296,7 +1297,7 @@ static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
 
 	iwl_pcie_set_pwr(trans, false);
 
-	if (trans->system_pm_mode == IWL_PLAT_PM_MODE_D0I3) {
+	if (!reset) {
 		iwl_clear_bit(trans, CSR_GP_CNTRL,
 			      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 	} else {
-- 
cgit v0.10.2


From 15ffd075e4088d3fc83ce8db914e7d3b6f31bc1c Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 24 Jan 2016 23:04:22 +0200
Subject: iwlwifi: various comments and code cleanups

No need to include net/ip6_checksum.h twice.
Remove TODOs.
Remove trailing space.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index e6a5e99..0ca0f13 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -506,7 +506,7 @@ struct iwl_trans_config {
 	bool sw_csum_tx;
 	const struct iwl_hcmd_arr *command_groups;
 	int command_groups_size;
- 
+
 	u32 sdio_adma_addr;
 };
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
index fb6d341..ab467cb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
@@ -287,16 +287,13 @@ enum iwl_rx_mpdu_status {
 	IWL_RX_MPDU_STATUS_KEY_ERROR		= BIT(4),
 	IWL_RX_MPDU_STATUS_ICV_OK		= BIT(5),
 	IWL_RX_MPDU_STATUS_MIC_OK		= BIT(6),
-	/* TODO - verify this is the correct value */
 	IWL_RX_MPDU_RES_STATUS_TTAK_OK		= BIT(7),
 	IWL_RX_MPDU_STATUS_SEC_MASK		= 0x7 << 8,
 	IWL_RX_MPDU_STATUS_SEC_NONE		= 0x0 << 8,
 	IWL_RX_MPDU_STATUS_SEC_WEP		= 0x1 << 8,
 	IWL_RX_MPDU_STATUS_SEC_CCM		= 0x2 << 8,
 	IWL_RX_MPDU_STATUS_SEC_TKIP		= 0x3 << 8,
-	/* TODO - define IWL_RX_MPDU_STATUS_SEC_EXT_ENC - this is a stub */
 	IWL_RX_MPDU_STATUS_SEC_EXT_ENC		= 0x4 << 8,
-	/* TODO - define IWL_RX_MPDU_STATUS_SEC_GCM - this is a stub */
 	IWL_RX_MPDU_STATUS_SEC_GCM		= 0x5 << 8,
 	IWL_RX_MPDU_STATUS_DECRYPTED		= BIT(11),
 	IWL_RX_MPDU_STATUS_WEP_MATCH		= BIT(12),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 89ea70d..6bf2bde 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -392,7 +392,6 @@ static const struct iwl_hcmd_arr iwl_mvm_groups[] = {
 	[PHY_OPS_GROUP] = HCMD_ARR(iwl_mvm_phy_names),
 };
 
-
 /* this forward declaration can avoid to export the function */
 static void iwl_mvm_async_handlers_wk(struct work_struct *wk);
 static void iwl_mvm_d0i3_exit_work(struct work_struct *wk);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index c499345b..837a7d5 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -34,7 +34,6 @@
 #include <linux/sched.h>
 #include <net/ip6_checksum.h>
 #include <net/tso.h>
-#include <net/ip6_checksum.h>
 
 #include "iwl-debug.h"
 #include "iwl-csr.h"
-- 
cgit v0.10.2


From bdf408eca877a2ae310e3df9f05e865ffe10e4de Mon Sep 17 00:00:00 2001
From: Hubert Tarasiuk <hubert.tarasiuk@gmail.com>
Date: Sun, 24 Jan 2016 13:35:06 +0100
Subject: iwlwifi: dvm: handle zero brightness for wifi LED

In order to have the LED being OFF constantly when the
brightness is set to 0, we need to pass IWL_LED_SOLID to
iwl_led_cmd as the off parameter, otherwise the led will
stay on constantly.

This fixes
https://bugzilla.kernel.org/show_bug.cgi?id=110551

Signed-off-by: Hubert Tarasiuk <hubert.tarasiuk@gmail.com>
[reworked the commit message]
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/led.c b/drivers/net/wireless/intel/iwlwifi/dvm/led.c
index 1aabb5e..1bbd17a 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/led.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/led.c
@@ -152,11 +152,14 @@ static void iwl_led_brightness_set(struct led_classdev *led_cdev,
 {
 	struct iwl_priv *priv = container_of(led_cdev, struct iwl_priv, led);
 	unsigned long on = 0;
+	unsigned long off = 0;
 
 	if (brightness > 0)
 		on = IWL_LED_SOLID;
+	else
+		off = IWL_LED_SOLID;
 
-	iwl_led_cmd(priv, on, 0);
+	iwl_led_cmd(priv, on, off);
 }
 
 static int iwl_led_blink_set(struct led_classdev *led_cdev,
-- 
cgit v0.10.2


From d1cae0fd22e9d0c55c97da29dcdfe28d99d96ea6 Mon Sep 17 00:00:00 2001
From: Rodrigo Freire <rfreire@redhat.com>
Date: Wed, 6 Jan 2016 20:24:01 -0200
Subject: iwlwifi: Document missing module options

This patch documents two missing module options in the internal
code comment block.

Signed-off-by: Rodrigo Freire <rfreire@redhat.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
index fd42f63..b88ecc7 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
@@ -108,6 +108,8 @@ enum iwl_amsdu_size {
  * @power_level: power level, default = 1
  * @debug_level: levels are IWL_DL_*
  * @ant_coupling: antenna coupling in dB, default = 0
+ * @nvm_file: specifies a external NVM file
+ * @uapsd_disable: disable U-APSD, default = 1
  * @d0i3_disable: disable d0i3, default = 1,
  * @d0i3_entry_delay: time to wait after no refs are taken before
  *	entering D0i3 (in msecs)
-- 
cgit v0.10.2


From 25657fec94f87308da19bb3d5063feef04d5ba20 Mon Sep 17 00:00:00 2001
From: Golan Ben-Ami <golan.ben.ami@intel.com>
Date: Wed, 2 Sep 2015 12:34:23 +0300
Subject: iwlwifi: mvm: add trigger for firmware dump upon TX response status

This will allow to collect the data when the firmware
sends a specific tx response status.

Signed-off-by: Golan Ben-Ami <golan.ben.ami@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h
index a5aaf68..8425e1a 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h
@@ -293,6 +293,8 @@ iwl_fw_error_next_data(struct iwl_fw_error_dump_data *data)
  * @FW_DBG_TX_LATENCY: trigger log collection when the tx latency goes above a
  *	threshold.
  * @FW_DBG_TDLS: trigger log collection upon TDLS related events.
+ * @FW_DBG_TRIGGER_TX_STATUS: trigger log collection upon tx status when
+ *  the firmware sends a tx reply.
  */
 enum iwl_fw_dbg_trigger {
 	FW_DBG_TRIGGER_INVALID = 0,
@@ -309,6 +311,7 @@ enum iwl_fw_dbg_trigger {
 	FW_DBG_TRIGGER_BA,
 	FW_DBG_TRIGGER_TX_LATENCY,
 	FW_DBG_TRIGGER_TDLS,
+	FW_DBG_TRIGGER_TX_STATUS,
 
 	/* must be last */
 	FW_DBG_TRIGGER_MAX,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 2273908..4050b4a 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -750,6 +750,19 @@ struct iwl_fw_dbg_trigger_tdls {
 } __packed;
 
 /**
+ * struct iwl_fw_dbg_trigger_tx_status - configures trigger for tx response
+ *  status.
+ * @statuses: the list of statuses to trigger the collection on
+ */
+struct iwl_fw_dbg_trigger_tx_status {
+	struct tx_status {
+		u8 status;
+		u8 reserved[3];
+	} __packed statuses[16];
+	__le32 reserved[2];
+} __packed;
+
+/**
  * struct iwl_fw_dbg_conf_tlv - a TLV that describes a debug configuration.
  * @id: conf id
  * @usniffer: should the uSniffer image be used
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 8bf48a7..5d73db2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -736,6 +736,37 @@ static void iwl_mvm_hwrate_to_tx_status(u32 rate_n_flags,
 	iwl_mvm_hwrate_to_tx_rate(rate_n_flags, info->band, r);
 }
 
+static void iwl_mvm_tx_status_check_trigger(struct iwl_mvm *mvm,
+					    u32 status)
+{
+	struct iwl_fw_dbg_trigger_tlv *trig;
+	struct iwl_fw_dbg_trigger_tx_status *status_trig;
+	int i;
+
+	if (!iwl_fw_dbg_trigger_enabled(mvm->fw, FW_DBG_TRIGGER_TX_STATUS))
+		return;
+
+	trig = iwl_fw_dbg_get_trigger(mvm->fw, FW_DBG_TRIGGER_TX_STATUS);
+	status_trig = (void *)trig->data;
+
+	if (!iwl_fw_dbg_trigger_check_stop(mvm, NULL, trig))
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(status_trig->statuses); i++) {
+		/* don't collect on status 0 */
+		if (!status_trig->statuses[i].status)
+			break;
+
+		if (status_trig->statuses[i].status != (status & TX_STATUS_MSK))
+			continue;
+
+		iwl_mvm_fw_dbg_collect_trig(mvm, trig,
+					    "Tx status %d was received",
+					    status & TX_STATUS_MSK);
+		break;
+	}
+}
+
 static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 				     struct iwl_rx_packet *pkt)
 {
@@ -784,6 +815,8 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 			break;
 		}
 
+		iwl_mvm_tx_status_check_trigger(mvm, status);
+
 		info->status.rates[0].count = tx_resp->failure_frame + 1;
 		iwl_mvm_hwrate_to_tx_status(le32_to_cpu(tx_resp->initial_rate),
 					    info);
-- 
cgit v0.10.2


From ee95ed3728cb3ed70b76fb770fff66133ff8b1fa Mon Sep 17 00:00:00 2001
From: Avri Altman <avri.altman@intel.com>
Date: Wed, 25 Nov 2015 13:17:10 +0200
Subject: iwlwifi: mvm: Add P2P client snoozing

Enable snoozing and U-APSD on P2P client. The firwmare will
support this only if the BSS vif is not associated.
Make this configurable by a constant variable and disable
it by default.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 4050b4a..dbf08c4 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -302,6 +302,7 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t;
  * @IWL_UCODE_TLV_CAPA_DC2DC_SUPPORT: supports DC2DC Command
  * @IWL_UCODE_TLV_CAPA_CSUM_SUPPORT: supports TCP Checksum Offload
  * @IWL_UCODE_TLV_CAPA_RADIO_BEACON_STATS: support radio and beacon statistics
+ * @IWL_UCODE_TLV_CAPA_P2P_STANDALONE_UAPSD: support p2p standalone U-APSD
  * @IWL_UCODE_TLV_CAPA_BT_COEX_PLCR: enabled BT Coex packet level co-running
  * @IWL_UCODE_TLV_CAPA_LAR_MULTI_MCC: ucode supports LAR updates with different
  *	sources for the MCC. This TLV bit is a future replacement to
@@ -336,6 +337,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_DC2DC_CONFIG_SUPPORT		= (__force iwl_ucode_tlv_capa_t)19,
 	IWL_UCODE_TLV_CAPA_CSUM_SUPPORT			= (__force iwl_ucode_tlv_capa_t)21,
 	IWL_UCODE_TLV_CAPA_RADIO_BEACON_STATS		= (__force iwl_ucode_tlv_capa_t)22,
+	IWL_UCODE_TLV_CAPA_P2P_STANDALONE_UAPSD		= (__force iwl_ucode_tlv_capa_t)26,
 	IWL_UCODE_TLV_CAPA_BT_COEX_PLCR			= (__force iwl_ucode_tlv_capa_t)28,
 	IWL_UCODE_TLV_CAPA_LAR_MULTI_MCC		= (__force iwl_ucode_tlv_capa_t)29,
 	IWL_UCODE_TLV_CAPA_BT_COEX_RRC			= (__force iwl_ucode_tlv_capa_t)30,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index b00c03f..959fc4d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -73,6 +73,7 @@
 #define IWL_MVM_WOWLAN_PS_RX_DATA_TIMEOUT	(10 * USEC_PER_MSEC)
 #define IWL_MVM_SHORT_PS_TX_DATA_TIMEOUT	(2 * 1024) /* defined in TU */
 #define IWL_MVM_SHORT_PS_RX_DATA_TIMEOUT	(40 * 1024) /* defined in TU */
+#define IWL_MVM_P2P_UAPSD_STANDALONE		0
 #define IWL_MVM_P2P_LOWLATENCY_PS_ENABLE	0
 #define IWL_MVM_UAPSD_RX_DATA_TIMEOUT		(50 * USEC_PER_MSEC)
 #define IWL_MVM_UAPSD_TX_DATA_TIMEOUT		(50 * USEC_PER_MSEC)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 5f3ac8c..5a92ab1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1005,6 +1005,14 @@ static inline bool iwl_mvm_is_mplut_supported(struct iwl_mvm *mvm)
 		IWL_MVM_BT_COEX_MPLUT;
 }
 
+static inline
+bool iwl_mvm_is_p2p_standalone_uapsd_supported(struct iwl_mvm *mvm)
+{
+	return fw_has_capa(&mvm->fw->ucode_capa,
+			   IWL_UCODE_TLV_CAPA_P2P_STANDALONE_UAPSD) &&
+		IWL_MVM_P2P_UAPSD_STANDALONE;
+}
+
 static inline bool iwl_mvm_has_new_rx_api(struct iwl_mvm *mvm)
 {
 	/* firmware flag isn't defined yet */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 9de159f..16b3e36 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -259,6 +259,26 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 		IWL_MVM_PS_HEAVY_RX_THLD_PERCENT;
 }
 
+static void iwl_mvm_p2p_standalone_iterator(void *_data, u8 *mac,
+					    struct ieee80211_vif *vif)
+{
+	bool *is_p2p_standalone = _data;
+
+	switch (ieee80211_vif_type_p2p(vif)) {
+	case NL80211_IFTYPE_P2P_GO:
+	case NL80211_IFTYPE_AP:
+		*is_p2p_standalone = false;
+		break;
+	case NL80211_IFTYPE_STATION:
+		if (vif->bss_conf.assoc)
+			*is_p2p_standalone = false;
+		break;
+
+	default:
+		break;
+	}
+}
+
 static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 				       struct ieee80211_vif *vif)
 {
@@ -268,9 +288,6 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 		    ETH_ALEN))
 		return false;
 
-	if (vif->p2p &&
-	    !(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_P2P_PS_UAPSD))
-		return false;
 	/*
 	 * Avoid using uAPSD if P2P client is associated to GO that uses
 	 * opportunistic power save. This is due to current FW limitation.
@@ -287,6 +304,22 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 	if (iwl_mvm_phy_ctx_count(mvm) >= 2)
 		return false;
 
+	if (vif->p2p) {
+		/* Allow U-APSD only if p2p is stand alone */
+		bool is_p2p_standalone = true;
+
+		if (!iwl_mvm_is_p2p_standalone_uapsd_supported(mvm))
+			return false;
+
+		ieee80211_iterate_active_interfaces_atomic(mvm->hw,
+					IEEE80211_IFACE_ITER_NORMAL,
+					iwl_mvm_p2p_standalone_iterator,
+					&is_p2p_standalone);
+
+		if (!is_p2p_standalone)
+			return false;
+	}
+
 	return true;
 }
 
-- 
cgit v0.10.2


From 33efe947e797c0f26b81bfe1b1eaf53501a6cd8e Mon Sep 17 00:00:00 2001
From: Golan Ben-Ami <golan.ben.ami@intel.com>
Date: Wed, 23 Dec 2015 17:53:27 +0200
Subject: iwlwifi: mvm: make collecting fw debug data optional

Slow platforms may have issues with dumping data upon
firmware assert. Make it easier to disable it for those
platform.

Signed-off-by: Golan Ben-Ami <golan.ben.ami@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index 959fc4d..4b560e4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -6,7 +6,8 @@
  * GPL LICENSE SUMMARY
  *
  * Copyright(c) 2013 - 2014 Intel Corporation. All rights reserved.
- * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2015        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -32,7 +33,8 @@
  * BSD LICENSE
  *
  * Copyright(c) 2013 - 2014 Intel Corporation. All rights reserved.
- * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2015        Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -108,6 +110,7 @@
 #define IWL_MVM_RS_80_20_FAR_RANGE_TWEAK	1
 #define IWL_MVM_TOF_IS_RESPONDER		0
 #define IWL_MVM_SW_TX_CSUM_OFFLOAD		0
+#define IWL_MVM_COLLECT_FW_ERR_DUMP		1
 #define IWL_MVM_RS_NUM_TRY_BEFORE_ANT_TOGGLE    1
 #define IWL_MVM_RS_HT_VHT_RETRIES_PER_RATE      2
 #define IWL_MVM_RS_HT_VHT_RETRIES_PER_RATE_TW   1
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c
index 0813f81..94ec3f0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c
@@ -435,6 +435,10 @@ void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm)
 	bool monitor_dump_only = false;
 	int i;
 
+	if (!IWL_MVM_COLLECT_FW_ERR_DUMP &&
+	    !mvm->trans->dbg_dest_tlv)
+		return;
+
 	lockdep_assert_held(&mvm->mutex);
 
 	/* there's no point in fw dump if the bus is dead */
-- 
cgit v0.10.2


From d8f7c5115d09e4b6b72b5b33414853aaa39bf837 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 14 Jan 2016 10:46:38 +0100
Subject: iwlwifi: mvm: remove shadowing variable

The outer scope has a perfectly suitable 'i' variable,
use it instead of adding a shadowing one in the inner
scope.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c
index 94ec3f0..4856eac 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c
@@ -644,8 +644,6 @@ void iwl_mvm_fw_error_dump(struct iwl_mvm *mvm)
 
 	/* Dump fw's virtual image */
 	if (mvm->fw->img[mvm->cur_ucode].paging_mem_size) {
-		u32 i;
-
 		for (i = 1; i < mvm->num_of_paging_blk + 1; i++) {
 			struct iwl_fw_error_dump_paging *paging;
 			struct page *pages =
-- 
cgit v0.10.2


From 27e070d3d3c2f6ab925ffd7c54814cabd95db51c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 14 Jan 2016 10:55:22 +0100
Subject: iwlwifi: mvm: fix debugfs signedness warning

Using kstrtouint() with a signed int isn't really right,
use kstrotoint() instead.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 90500e2..8059efa 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -983,7 +983,7 @@ static ssize_t iwl_dbgfs_cont_recording_write(struct iwl_mvm *mvm,
 	    trans->cfg->device_family != IWL_DEVICE_FAMILY_8000)
 		return -EOPNOTSUPP;
 
-	ret = kstrtouint(buf, 0, &rec_mode);
+	ret = kstrtoint(buf, 0, &rec_mode);
 	if (ret)
 		return ret;
 
-- 
cgit v0.10.2


From 7869318e4296e27d4cb97505e8739d62b48b2d58 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Sun, 27 Dec 2015 13:45:42 +0200
Subject: iwlwifi: mvm: add support for negative temperatures

The driver should support also negative temperatures.
So there is a need to separate between the return value and
temperature in order to be able to distinguish between
a negative temperature and error value.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 8059efa..5c0f939 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -261,17 +261,18 @@ static ssize_t iwl_dbgfs_nic_temp_read(struct file *file,
 {
 	struct iwl_mvm *mvm = file->private_data;
 	char buf[16];
-	int pos, temp;
+	int pos, ret;
+	s32 temp;
 
 	if (!mvm->ucode_loaded)
 		return -EIO;
 
 	mutex_lock(&mvm->mutex);
-	temp = iwl_mvm_get_temp(mvm);
+	ret = iwl_mvm_get_temp(mvm, &temp);
 	mutex_unlock(&mvm->mutex);
 
-	if (temp < 0)
-		return temp;
+	if (ret)
+		return -EIO;
 
 	pos = scnprintf(buf , sizeof(buf), "%d\n", temp);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 5a92ab1..216c8d6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1489,7 +1489,7 @@ void iwl_mvm_tt_handler(struct iwl_mvm *mvm);
 void iwl_mvm_tt_initialize(struct iwl_mvm *mvm, u32 min_backoff);
 void iwl_mvm_tt_exit(struct iwl_mvm *mvm);
 void iwl_mvm_set_hw_ctkill_state(struct iwl_mvm *mvm, bool state);
-int iwl_mvm_get_temp(struct iwl_mvm *mvm);
+int iwl_mvm_get_temp(struct iwl_mvm *mvm, s32 *temp);
 
 /* Location Aware Regulatory */
 struct iwl_mcc_update_resp *
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index fb76004..758d05a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -194,12 +194,12 @@ static int iwl_mvm_get_temp_cmd(struct iwl_mvm *mvm)
 	return iwl_mvm_send_cmd_pdu(mvm, cmdid, 0, sizeof(extcmd), &extcmd);
 }
 
-int iwl_mvm_get_temp(struct iwl_mvm *mvm)
+int iwl_mvm_get_temp(struct iwl_mvm *mvm, s32 *temp)
 {
 	struct iwl_notification_wait wait_temp_notif;
 	static u16 temp_notif[] = { WIDE_ID(PHY_OPS_GROUP,
 					    DTS_MEASUREMENT_NOTIF_WIDE) };
-	int ret, temp;
+	int ret;
 
 	if (!fw_has_api(&mvm->fw->ucode_capa, IWL_UCODE_TLV_API_WIDE_CMD_HDR))
 		temp_notif[0] = DTS_MEASUREMENT_NOTIFICATION;
@@ -208,7 +208,7 @@ int iwl_mvm_get_temp(struct iwl_mvm *mvm)
 
 	iwl_init_notification_wait(&mvm->notif_wait, &wait_temp_notif,
 				   temp_notif, ARRAY_SIZE(temp_notif),
-				   iwl_mvm_temp_notif_wait, &temp);
+				   iwl_mvm_temp_notif_wait, temp);
 
 	ret = iwl_mvm_get_temp_cmd(mvm);
 	if (ret) {
@@ -219,12 +219,10 @@ int iwl_mvm_get_temp(struct iwl_mvm *mvm)
 
 	ret = iwl_wait_notification(&mvm->notif_wait, &wait_temp_notif,
 				    IWL_MVM_TEMP_NOTIF_WAIT_TIMEOUT);
-	if (ret) {
+	if (ret)
 		IWL_ERR(mvm, "Getting the temperature timed out\n");
-		return ret;
-	}
 
-	return temp;
+	return ret;
 }
 
 static void check_exit_ctkill(struct work_struct *work)
@@ -233,6 +231,7 @@ static void check_exit_ctkill(struct work_struct *work)
 	struct iwl_mvm *mvm;
 	u32 duration;
 	s32 temp;
+	int ret;
 
 	tt = container_of(work, struct iwl_mvm_tt_mgmt, ct_kill_exit.work);
 	mvm = container_of(tt, struct iwl_mvm, thermal_throttle);
@@ -250,13 +249,13 @@ static void check_exit_ctkill(struct work_struct *work)
 		goto reschedule;
 	}
 
-	temp = iwl_mvm_get_temp(mvm);
+	ret = iwl_mvm_get_temp(mvm, &temp);
 
 	iwl_mvm_unref(mvm, IWL_MVM_REF_CHECK_CTKILL);
 
 	__iwl_mvm_mac_stop(mvm);
 
-	if (temp < 0)
+	if (ret)
 		goto reschedule;
 
 	IWL_DEBUG_TEMP(mvm, "NIC temperature: %d\n", temp);
-- 
cgit v0.10.2


From 0db056d32490181be48efc2ca7cfa7494b13e6b5 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 29 Dec 2015 11:07:15 +0200
Subject: iwlwifi: mvm: support beacon storing

Currently firmware is configured to filter out beacons. In case
a beacon was changed - it is waking the host.
However, some vendors change their IEs frequently without any
significant change, and redundant wakeups are triggered as a
result.
As a solution disable beacon filtering when entering d0i3.
Instead, firmware will store the latest beacon and upon exiting
d0i3 it will send it up to the host, so the host can act upon
changes (if there were any).
This beacon will arrive as a dedicated notification - support it
as well.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index dbf08c4..5152987 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -315,6 +315,8 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t;
  * @IWL_UCODE_TLV_CAPA_BT_MPLUT_SUPPORT: supports bt-coex Multi-priority LUT
  * @IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION: firmware will decide on what
  *	antenna the beacon should be transmitted
+ * @IWL_UCODE_TLV_CAPA_BEACON_STORING: firmware will store the latest beacon
+ *	from AP and will send it upon d0i3 exit.
  * @IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2: support LAR API V2
  *
  * @NUM_IWL_UCODE_TLV_CAPA: number of bits used
@@ -346,6 +348,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_SHORT_PM_TIMEOUTS		= (__force iwl_ucode_tlv_capa_t)65,
 	IWL_UCODE_TLV_CAPA_BT_MPLUT_SUPPORT		= (__force iwl_ucode_tlv_capa_t)67,
 	IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION		= (__force iwl_ucode_tlv_capa_t)71,
+	IWL_UCODE_TLV_CAPA_BEACON_STORING		= (__force iwl_ucode_tlv_capa_t)72,
 	IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2		= (__force iwl_ucode_tlv_capa_t)73,
 
 	NUM_IWL_UCODE_TLV_CAPA
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index 78572ef..3463761 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -853,7 +853,8 @@ iwl_mvm_get_wowlan_config(struct iwl_mvm *mvm,
 	wowlan_config_cmd->is_11n_connection =
 					ap_sta->ht_cap.ht_supported;
 	wowlan_config_cmd->flags = ENABLE_L3_FILTERING |
-		ENABLE_NBNS_FILTERING | ENABLE_DHCP_FILTERING;
+		ENABLE_NBNS_FILTERING | ENABLE_DHCP_FILTERING |
+		ENABLE_STORE_BEACON;
 
 	/* Query the last used seqno and set it */
 	ret = iwl_mvm_get_last_nonqos_seq(mvm, vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-d3.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-d3.h
index 62b9a0a..eec52c5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-d3.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-d3.h
@@ -251,6 +251,7 @@ enum iwl_wowlan_flags {
 	ENABLE_L3_FILTERING	= BIT(1),
 	ENABLE_NBNS_FILTERING	= BIT(2),
 	ENABLE_DHCP_FILTERING	= BIT(3),
+	ENABLE_STORE_BEACON	= BIT(4),
 };
 
 struct iwl_wowlan_config_cmd {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index 82049bb..b6b5727 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -280,11 +280,16 @@ enum iwl_phy_ops_subcmd_ids {
 	DTS_MEASUREMENT_NOTIF_WIDE = 0xFF,
 };
 
+enum iwl_prot_offload_subcmd_ids {
+	STORED_BEACON_NTF = 0xFF,
+};
+
 /* command groups */
 enum {
 	LEGACY_GROUP = 0x0,
 	LONG_GROUP = 0x1,
 	PHY_OPS_GROUP = 0x4,
+	PROT_OFFLOAD_GROUP = 0xb,
 };
 
 /**
@@ -1851,4 +1856,28 @@ struct iwl_shared_mem_cfg {
 	__le32 page_buff_size;
 } __packed; /* SHARED_MEM_ALLOC_API_S_VER_1 */
 
+#define MAX_STORED_BEACON_SIZE 600
+
+/**
+ * Stored beacon notification
+ *
+ * @system_time: system time on air rise
+ * @tsf: TSF on air rise
+ * @beacon_timestamp: beacon on air rise
+ * @phy_flags: general phy flags: band, modulation, etc.
+ * @channel: channel this beacon was received on
+ * @rates: rate in ucode internal format
+ * @byte_count: frame's byte count
+ */
+struct iwl_stored_beacon_notif {
+	__le32 system_time;
+	__le64 tsf;
+	__le32 beacon_timestamp;
+	__le16 phy_flags;
+	__le16 channel;
+	__le32 rates;
+	__le32 byte_count;
+	u8 data[MAX_STORED_BEACON_SIZE];
+} __packed; /* WOWLAN_STROED_BEACON_INFO_S_VER_1 */
+
 #endif /* __fw_api_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index bf1e5eb..62927f5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -7,7 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
- * Copyright(c) 2015 Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -34,7 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
- * Copyright(c) 2015 Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -1462,3 +1462,40 @@ void iwl_mvm_rx_missed_beacons_notif(struct iwl_mvm *mvm,
 						   iwl_mvm_beacon_loss_iterator,
 						   mb);
 }
+
+void iwl_mvm_rx_stored_beacon_notif(struct iwl_mvm *mvm,
+				    struct iwl_rx_cmd_buffer *rxb)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_stored_beacon_notif *sb = (void *)pkt->data;
+	struct ieee80211_rx_status rx_status;
+	struct sk_buff *skb;
+	u32 size = le32_to_cpu(sb->byte_count);
+
+	if (size == 0)
+		return;
+
+	skb = alloc_skb(size, GFP_ATOMIC);
+	if (!skb) {
+		IWL_ERR(mvm, "alloc_skb failed\n");
+		return;
+	}
+
+	/* update rx_status according to the notification's metadata */
+	memset(&rx_status, 0, sizeof(rx_status));
+	rx_status.mactime = le64_to_cpu(sb->tsf);
+	rx_status.device_timestamp = le32_to_cpu(sb->system_time);
+	rx_status.band =
+		(sb->phy_flags & cpu_to_le16(RX_RES_PHY_FLAGS_BAND_24)) ?
+				IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ;
+	rx_status.freq =
+		ieee80211_channel_to_frequency(le16_to_cpu(sb->channel),
+					       rx_status.band);
+
+	/* copy the data */
+	memcpy(skb_put(skb, size), sb->data, size);
+	memcpy(IEEE80211_SKB_RXCB(skb), &rx_status, sizeof(rx_status));
+
+	/* pass it as regular rx to mac80211 */
+	ieee80211_rx_napi(mvm->hw, skb, NULL);
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 216c8d6..b45cf4f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1192,6 +1192,8 @@ void iwl_mvm_rx_beacon_notif(struct iwl_mvm *mvm,
 			     struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_rx_missed_beacons_notif(struct iwl_mvm *mvm,
 				     struct iwl_rx_cmd_buffer *rxb);
+void iwl_mvm_rx_stored_beacon_notif(struct iwl_mvm *mvm,
+				    struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_mac_ctxt_recalc_tsf_id(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif);
 unsigned long iwl_mvm_get_used_hw_queues(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 6bf2bde..3c869ad 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -33,6 +33,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -267,6 +268,8 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 		   true),
 	RX_HANDLER(MFUART_LOAD_NOTIFICATION, iwl_mvm_rx_mfuart_notif, false),
 	RX_HANDLER(TOF_NOTIFICATION, iwl_mvm_tof_resp_handler, true),
+	RX_HANDLER_GRP(PROT_OFFLOAD_GROUP, STORED_BEACON_NTF,
+		       iwl_mvm_rx_stored_beacon_notif, false),
 
 };
 #undef RX_HANDLER
@@ -386,10 +389,18 @@ static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
 	HCMD_NAME(DTS_MEASUREMENT_NOTIF_WIDE),
 };
 
+/* Please keep this array *SORTED* by hex value.
+ * Access is done through binary search
+ */
+static const struct iwl_hcmd_names iwl_mvm_prot_offload_names[] = {
+	HCMD_NAME(STORED_BEACON_NTF),
+};
+
 static const struct iwl_hcmd_arr iwl_mvm_groups[] = {
 	[LEGACY_GROUP] = HCMD_ARR(iwl_mvm_legacy_names),
 	[LONG_GROUP] = HCMD_ARR(iwl_mvm_legacy_names),
 	[PHY_OPS_GROUP] = HCMD_ARR(iwl_mvm_phy_names),
+	[PROT_OFFLOAD_GROUP] = HCMD_ARR(iwl_mvm_prot_offload_names),
 };
 
 /* this forward declaration can avoid to export the function */
@@ -1195,7 +1206,7 @@ static void iwl_mvm_set_wowlan_data(struct iwl_mvm *mvm,
 	cmd->is_11n_connection = ap_sta->ht_cap.ht_supported;
 	cmd->offloading_tid = iter_data->offloading_tid;
 	cmd->flags = ENABLE_L3_FILTERING | ENABLE_NBNS_FILTERING |
-		ENABLE_DHCP_FILTERING;
+		ENABLE_DHCP_FILTERING | ENABLE_STORE_BEACON;
 	/*
 	 * The d0i3 uCode takes care of the nonqos counters,
 	 * so configure only the qos seq ones.
@@ -1216,8 +1227,7 @@ int iwl_mvm_enter_d0i3(struct iwl_op_mode *op_mode)
 	struct iwl_wowlan_config_cmd wowlan_config_cmd = {
 		.wakeup_filter = cpu_to_le32(IWL_WOWLAN_WAKEUP_RX_FRAME |
 					     IWL_WOWLAN_WAKEUP_BEACON_MISS |
-					     IWL_WOWLAN_WAKEUP_LINK_CHANGE |
-					     IWL_WOWLAN_WAKEUP_BCN_FILTERING),
+					     IWL_WOWLAN_WAKEUP_LINK_CHANGE),
 	};
 	struct iwl_d3_manager_config d3_cfg_cmd = {
 		.min_sleep_time = cpu_to_le32(1000),
@@ -1267,6 +1277,12 @@ int iwl_mvm_enter_d0i3(struct iwl_op_mode *op_mode)
 
 	/* configure wowlan configuration only if needed */
 	if (mvm->d0i3_ap_sta_id != IWL_MVM_STATION_COUNT) {
+		/* wake on beacons only if beacon storing isn't supported */
+		if (!fw_has_capa(&mvm->fw->ucode_capa,
+				 IWL_UCODE_TLV_CAPA_BEACON_STORING))
+			wowlan_config_cmd.wakeup_filter |=
+				cpu_to_le32(IWL_WOWLAN_WAKEUP_BCN_FILTERING);
+
 		iwl_mvm_wowlan_config_key_params(mvm,
 						 d0i3_iter_data.connected_vif,
 						 true, flags);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 16b3e36..0d03b3e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -7,7 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
- * Copyright(c) 2015        Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -34,7 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
- * Copyright(c) 2015        Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -903,9 +903,9 @@ static int iwl_mvm_update_beacon_abort(struct iwl_mvm *mvm,
 	return _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd, 0, false);
 }
 
-int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
-				  struct ieee80211_vif *vif,
-				  u32 flags)
+static int _iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
+					  struct ieee80211_vif *vif,
+					  u32 flags, bool d0i3)
 {
 	struct iwl_beacon_filter_cmd cmd = {};
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -916,12 +916,20 @@ int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 
 	ret = iwl_mvm_beacon_filter_send_cmd(mvm, &cmd, flags);
 
-	if (!ret)
+	/* don't change bf_enabled in case of temporary d0i3 configuration */
+	if (!ret && !d0i3)
 		mvmvif->bf_data.bf_enabled = false;
 
 	return ret;
 }
 
+int iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
+				  struct ieee80211_vif *vif,
+				  u32 flags)
+{
+	return _iwl_mvm_disable_beacon_filter(mvm, vif, flags, false);
+}
+
 static int iwl_mvm_power_set_ps(struct iwl_mvm *mvm)
 {
 	bool disable_ps;
@@ -1058,8 +1066,17 @@ int iwl_mvm_update_d0i3_power_mode(struct iwl_mvm *mvm,
 			IWL_BF_CMD_CONFIG_D0I3,
 			.bf_enable_beacon_filter = cpu_to_le32(1),
 		};
-		ret = _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd_bf,
-						    flags, true);
+		/*
+		 * When beacon storing is supported - disable beacon filtering
+		 * altogether - the latest beacon will be sent when exiting d0i3
+		 */
+		if (fw_has_capa(&mvm->fw->ucode_capa,
+				IWL_UCODE_TLV_CAPA_BEACON_STORING))
+			ret = _iwl_mvm_disable_beacon_filter(mvm, vif, flags,
+							     true);
+		else
+			ret = _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd_bf,
+							    flags, true);
 	} else {
 		if (mvmvif->bf_data.bf_enabled)
 			ret = iwl_mvm_enable_beacon_filter(mvm, vif, flags);
-- 
cgit v0.10.2


From b525d08125e61e262be63ea250e3e13e93b6a92a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 6 Jan 2016 10:01:41 +0100
Subject: iwlwifi: mvm: track low-latency sources separately

To be able to test low-latency behaviour properly, split the
different low-latency sources so that setting any one of them,
for example from debugfs, is sufficient; this avoids getting
the debug setting overwritten by other sources.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
index 9e0d463..c286a5f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -1255,6 +1257,7 @@ static ssize_t iwl_dbgfs_low_latency_write(struct ieee80211_vif *vif, char *buf,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm *mvm = mvmvif->mvm;
+	bool prev;
 	u8 value;
 	int ret;
 
@@ -1265,7 +1268,9 @@ static ssize_t iwl_dbgfs_low_latency_write(struct ieee80211_vif *vif, char *buf,
 		return -EINVAL;
 
 	mutex_lock(&mvm->mutex);
-	iwl_mvm_update_low_latency(mvm, vif, value);
+	prev = iwl_mvm_vif_low_latency(mvmvif);
+	mvmvif->low_latency_dbgfs = value;
+	iwl_mvm_update_low_latency(mvm, vif, prev);
 	mutex_unlock(&mvm->mutex);
 
 	return count;
@@ -1277,11 +1282,15 @@ static ssize_t iwl_dbgfs_low_latency_read(struct file *file,
 {
 	struct ieee80211_vif *vif = file->private_data;
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	char buf[2];
+	char buf[30] = {};
+	int len;
 
-	buf[0] = mvmvif->low_latency ? '1' : '0';
-	buf[1] = '\n';
-	return simple_read_from_buffer(user_buf, count, ppos, buf, sizeof(buf));
+	len = snprintf(buf, sizeof(buf) - 1,
+		       "traffic=%d\ndbgfs=%d\nvcmd=%d\n",
+		       mvmvif->low_latency_traffic,
+		       mvmvif->low_latency_dbgfs,
+		       mvmvif->low_latency_vcmd);
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
 }
 
 static ssize_t iwl_dbgfs_uapsd_misbehaving_read(struct file *file,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index b45cf4f..418aeff 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -346,8 +346,9 @@ struct iwl_mvm_vif_bf_data {
  * @pm_enabled - Indicate if MAC power management is allowed
  * @monitor_active: indicates that monitor context is configured, and that the
  *	interface should get quota etc.
- * @low_latency: indicates that this interface is in low-latency mode
- *	(VMACLowLatencyMode)
+ * @low_latency_traffic: indicates low latency traffic was detected
+ * @low_latency_dbgfs: low latency mode set from debugfs
+ * @low_latency_vcmd: low latency mode set from vendor command
  * @ps_disabled: indicates that this interface requires PS to be disabled
  * @queue_params: QoS params for this MAC
  * @bcast_sta: station used for broadcast packets. Used by the following
@@ -375,7 +376,7 @@ struct iwl_mvm_vif {
 	bool ap_ibss_active;
 	bool pm_enabled;
 	bool monitor_active;
-	bool low_latency;
+	bool low_latency_traffic, low_latency_dbgfs, low_latency_vcmd;
 	bool ps_disabled;
 	struct iwl_mvm_vif_bf_data bf_data;
 
@@ -1427,8 +1428,9 @@ static inline bool iwl_mvm_vif_low_latency(struct iwl_mvm_vif *mvmvif)
 	 * binding, so this has no real impact. For now, just return
 	 * the current desired low-latency state.
 	 */
-
-	return mvmvif->low_latency;
+	return mvmvif->low_latency_dbgfs ||
+	       mvmvif->low_latency_traffic ||
+	       mvmvif->low_latency_vcmd;
 }
 
 /* hw scheduler queue config */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 3a989f5..59453c1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -937,18 +937,16 @@ bool iwl_mvm_rx_diversity_allowed(struct iwl_mvm *mvm)
 }
 
 int iwl_mvm_update_low_latency(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			       bool value)
+			       bool prev)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int res;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	if (mvmvif->low_latency == value)
+	if (iwl_mvm_vif_low_latency(mvmvif) == prev)
 		return 0;
 
-	mvmvif->low_latency = value;
-
 	res = iwl_mvm_update_quotas(mvm, false, NULL);
 	if (res)
 		return res;
-- 
cgit v0.10.2


From a80c1cf9b4b2937cfe782aba7b54a5144fe7eec2 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 13 Jan 2016 15:01:00 +0100
Subject: iwlwifi: mvm: support setting minimum quota from debugfs

For debug purposes, allow setting minimum quota (for a single
virtual interface) from debugfs. This is an absolute minimum,
so it can only be set up to 95%.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
index c286a5f..1400445 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
@@ -1372,6 +1372,59 @@ static ssize_t iwl_dbgfs_rx_phyinfo_read(struct file *file,
 	return simple_read_from_buffer(user_buf, count, ppos, buf, sizeof(buf));
 }
 
+static void iwl_dbgfs_quota_check(void *data, u8 *mac,
+				  struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	int *ret = data;
+
+	if (mvmvif->dbgfs_quota_min)
+		*ret = -EINVAL;
+}
+
+static ssize_t iwl_dbgfs_quota_min_write(struct ieee80211_vif *vif, char *buf,
+					 size_t count, loff_t *ppos)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm *mvm = mvmvif->mvm;
+	u16 value;
+	int ret;
+
+	ret = kstrtou16(buf, 0, &value);
+	if (ret)
+		return ret;
+
+	if (value > 95)
+		return -EINVAL;
+
+	mutex_lock(&mvm->mutex);
+
+	mvmvif->dbgfs_quota_min = 0;
+	ieee80211_iterate_interfaces(mvm->hw, IEEE80211_IFACE_ITER_NORMAL,
+				     iwl_dbgfs_quota_check, &ret);
+	if (ret == 0) {
+		mvmvif->dbgfs_quota_min = value;
+		iwl_mvm_update_quotas(mvm, false, NULL);
+	}
+	mutex_unlock(&mvm->mutex);
+
+	return ret ?: count;
+}
+
+static ssize_t iwl_dbgfs_quota_min_read(struct file *file,
+					char __user *user_buf,
+					size_t count, loff_t *ppos)
+{
+	struct ieee80211_vif *vif = file->private_data;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	char buf[10];
+	int len;
+
+	len = snprintf(buf, sizeof(buf), "%d\n", mvmvif->dbgfs_quota_min);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
 #define MVM_DEBUGFS_WRITE_FILE_OPS(name, bufsz) \
 	_MVM_DEBUGFS_WRITE_FILE_OPS(name, bufsz, struct ieee80211_vif)
 #define MVM_DEBUGFS_READ_WRITE_FILE_OPS(name, bufsz) \
@@ -1395,6 +1448,7 @@ MVM_DEBUGFS_READ_WRITE_FILE_OPS(tof_range_req_ext, 32);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(tof_range_abort, 32);
 MVM_DEBUGFS_READ_FILE_OPS(tof_range_response);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(tof_responder_params, 32);
+MVM_DEBUGFS_READ_WRITE_FILE_OPS(quota_min, 32);
 
 void iwl_mvm_vif_dbgfs_register(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
@@ -1432,6 +1486,8 @@ void iwl_mvm_vif_dbgfs_register(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 				 S_IRUSR | S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE_VIF(rx_phyinfo, mvmvif->dbgfs_dir,
 				 S_IRUSR | S_IWUSR);
+	MVM_DEBUGFS_ADD_FILE_VIF(quota_min, mvmvif->dbgfs_dir,
+				 S_IRUSR | S_IWUSR);
 
 	if (vif->type == NL80211_IFTYPE_STATION && !vif->p2p &&
 	    mvmvif == mvm->bf_allowed_vif)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 418aeff..5a34808 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -433,6 +433,7 @@ struct iwl_mvm_vif {
 	struct iwl_dbgfs_pm dbgfs_pm;
 	struct iwl_dbgfs_bf dbgfs_bf;
 	struct iwl_mac_power_cmd mac_pwr_cmd;
+	int dbgfs_quota_min;
 #endif
 
 	enum ieee80211_smps_mode smps_requests[NUM_IWL_MVM_SMPS_REQ];
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/quota.c b/drivers/net/wireless/intel/iwlwifi/mvm/quota.c
index 0b762b4..2141db5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/quota.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/quota.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -74,6 +76,9 @@ struct iwl_mvm_quota_iterator_data {
 	int n_interfaces[MAX_BINDINGS];
 	int colors[MAX_BINDINGS];
 	int low_latency[MAX_BINDINGS];
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+	int dbgfs_min[MAX_BINDINGS];
+#endif
 	int n_low_latency_bindings;
 	struct ieee80211_vif *disabled_vif;
 };
@@ -129,6 +134,12 @@ static void iwl_mvm_quota_iterator(void *_data, u8 *mac,
 
 	data->n_interfaces[id]++;
 
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+	if (mvmvif->dbgfs_quota_min)
+		data->dbgfs_min[id] = max(data->dbgfs_min[id],
+					  mvmvif->dbgfs_quota_min);
+#endif
+
 	if (iwl_mvm_vif_low_latency(mvmvif) && !data->low_latency[id]) {
 		data->n_low_latency_bindings++;
 		data->low_latency[id] = true;
@@ -259,6 +270,11 @@ int iwl_mvm_update_quotas(struct iwl_mvm *mvm,
 
 		if (data.n_interfaces[i] <= 0)
 			cmd.quotas[idx].quota = cpu_to_le32(0);
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+		else if (data.dbgfs_min[i])
+			cmd.quotas[idx].quota =
+				cpu_to_le32(data.dbgfs_min[i] * QUOTA_100 / 100);
+#endif
 		else if (data.n_low_latency_bindings == 1 && n_non_lowlat &&
 			 data.low_latency[i])
 			/*
-- 
cgit v0.10.2


From 0c1c6e37137fc2333ef9898c24ea0cb9efbd8df0 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 16 Dec 2015 21:17:06 +0200
Subject: iwlwifi: mvm: change access to ieee80211_hdr

Make the code clearer.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 0c073e0..4cce372 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -294,7 +294,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 	struct ieee80211_rx_status *rx_status;
 	struct iwl_rx_packet *pkt = rxb_addr(rxb);
 	struct iwl_rx_mpdu_desc *desc = (void *)pkt->data;
-	struct ieee80211_hdr *hdr = (void *)(desc + 1);
+	struct ieee80211_hdr *hdr = (void *)(pkt->data + sizeof(*desc));
 	u32 len = le16_to_cpu(desc->mpdu_len);
 	u32 rate_n_flags = le32_to_cpu(desc->rate_n_flags);
 	struct ieee80211_sta *sta = NULL;
-- 
cgit v0.10.2


From 1c4e15a23d27e2efd56d2b9b7436bff060f126ef Mon Sep 17 00:00:00 2001
From: Max Stepanov <Max.Stepanov@intel.com>
Date: Mon, 4 Jan 2016 11:09:14 +0200
Subject: iwlwifi: mvm: add debug print if scan config is ignored

Print a debug message in iwl_mvm_config_scan() if a scan configuration
data is decided not to be sent to FW.

Signed-off-by: Max Stepanov <Max.Stepanov@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 0476e76..1e1ab9d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -930,8 +930,11 @@ int iwl_mvm_config_scan(struct iwl_mvm *mvm)
 	if (WARN_ON(num_channels > mvm->fw->ucode_capa.n_scan_channels))
 		return -ENOBUFS;
 
-	if (type == mvm->scan_type)
+	if (type == mvm->scan_type) {
+		IWL_DEBUG_SCAN(mvm,
+			       "Ignoring UMAC scan config of the same type\n");
 		return 0;
+	}
 
 	cmd_size = sizeof(*scan_config) + mvm->fw->ucode_capa.n_scan_channels;
 
-- 
cgit v0.10.2


From 837c4da98481d4e504b2aba077c8528fee1b6d13 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Thu, 7 Jan 2016 16:50:45 +0200
Subject: iwlwifi: mvm: change the check for ADD_STA status

The firmware will return the baid for BA session in the
ADD_STA command response.
This requires masking the check of the status, which is
actually only 8 bits, and not the whole 32 bits.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h
index 6fca4fb..1eb3983 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h
@@ -253,6 +253,9 @@ struct iwl_mvm_keyinfo {
 	__le64 hw_tkip_mic_tx_key;
 } __packed;
 
+#define IWL_ADD_STA_STATUS_MASK	0xFF
+#define IWL_ADD_STA_BAID_MASK	0xFF00
+
 /**
  * struct iwl_mvm_add_sta_cmd - Add/modify a station in the fw's sta table.
  * ( REPLY_ADD_STA = 0x18 )
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index b556e33..84b7fd1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -192,7 +192,7 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	if (ret)
 		return ret;
 
-	switch (status) {
+	switch (status & IWL_ADD_STA_STATUS_MASK) {
 	case ADD_STA_SUCCESS:
 		IWL_DEBUG_ASSOC(mvm, "ADD_STA PASSED\n");
 		break;
@@ -362,7 +362,7 @@ int iwl_mvm_drain_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 	if (ret)
 		return ret;
 
-	switch (status) {
+	switch (status & IWL_ADD_STA_STATUS_MASK) {
 	case ADD_STA_SUCCESS:
 		IWL_DEBUG_INFO(mvm, "Frames for staid %d will drained in fw\n",
 			       mvmsta->sta_id);
@@ -628,7 +628,7 @@ static int iwl_mvm_add_int_sta_common(struct iwl_mvm *mvm,
 	if (ret)
 		return ret;
 
-	switch (status) {
+	switch (status & IWL_ADD_STA_STATUS_MASK) {
 	case ADD_STA_SUCCESS:
 		IWL_DEBUG_INFO(mvm, "Internal station added.\n");
 		return 0;
@@ -851,7 +851,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	if (ret)
 		return ret;
 
-	switch (status) {
+	switch (status & IWL_ADD_STA_STATUS_MASK) {
 	case ADD_STA_SUCCESS:
 		IWL_DEBUG_INFO(mvm, "RX BA Session %sed in fw\n",
 			       start ? "start" : "stopp");
@@ -909,7 +909,7 @@ static int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	if (ret)
 		return ret;
 
-	switch (status) {
+	switch (status & IWL_ADD_STA_STATUS_MASK) {
 	case ADD_STA_SUCCESS:
 		break;
 	default:
-- 
cgit v0.10.2


From 81f02ba3dbe0c9133dc4490f0a7d3acd99a89aad Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 30 Dec 2015 23:58:29 +0200
Subject: iwlwifi: mvm: add tlv for multi queue rx support

Previous patches enabled the multi-queue rx path based on
iwl_mvm_has_new_rx_api() which returned false by default.
Change it to return the actual value based on the firmware
TLV which is now defined.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 5152987..e2dbc67 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -347,6 +347,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_EXTENDED_DTS_MEASURE		= (__force iwl_ucode_tlv_capa_t)64,
 	IWL_UCODE_TLV_CAPA_SHORT_PM_TIMEOUTS		= (__force iwl_ucode_tlv_capa_t)65,
 	IWL_UCODE_TLV_CAPA_BT_MPLUT_SUPPORT		= (__force iwl_ucode_tlv_capa_t)67,
+	IWL_UCODE_TLV_CAPA_MULTI_QUEUE_RX_SUPPORT	= (__force iwl_ucode_tlv_capa_t)68,
 	IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION		= (__force iwl_ucode_tlv_capa_t)71,
 	IWL_UCODE_TLV_CAPA_BEACON_STORING		= (__force iwl_ucode_tlv_capa_t)72,
 	IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2		= (__force iwl_ucode_tlv_capa_t)73,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 5a34808..f87aa97 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1017,8 +1017,8 @@ bool iwl_mvm_is_p2p_standalone_uapsd_supported(struct iwl_mvm *mvm)
 
 static inline bool iwl_mvm_has_new_rx_api(struct iwl_mvm *mvm)
 {
-	/* firmware flag isn't defined yet */
-	return false;
+	return fw_has_capa(&mvm->fw->ucode_capa,
+			   IWL_UCODE_TLV_CAPA_MULTI_QUEUE_RX_SUPPORT);
 }
 
 extern const u8 iwl_mvm_ac_to_tx_fifo[];
-- 
cgit v0.10.2


From c257d5fb52872213f33016c0750bc2334a781700 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 24 Jan 2016 15:28:43 +0100
Subject: iwlwifi: treat iwl_parse_nvm_data() MAC addr as little endian

The MAC address parameters passed to iwl_parse_nvm_data() are passed on
to iwl_set_hw_address_family_8000() which treats them as little endian.
Annotate them as such, and add the missing byte-swapping in mvm.

While at it, add the MAC address to the error to make debugging issues
with it easier.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 7b89bfc..50f4cc6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -539,7 +539,7 @@ static void iwl_set_hw_address_family_8000(struct device *dev,
 					   struct iwl_nvm_data *data,
 					   const __le16 *mac_override,
 					   const __le16 *nvm_hw,
-					   u32 mac_addr0, u32 mac_addr1)
+					   __le32 mac_addr0, __le32 mac_addr1)
 {
 	const u8 *hw_addr;
 
@@ -583,7 +583,8 @@ static void iwl_set_hw_address_family_8000(struct device *dev,
 
 		if (!is_valid_ether_addr(data->hw_addr))
 			IWL_ERR_DEV(dev,
-				    "mac address from hw section is not valid\n");
+				    "mac address (%pM) from hw section is not valid\n",
+				    data->hw_addr);
 
 		return;
 	}
@@ -597,7 +598,7 @@ iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
 		   const __le16 *nvm_calib, const __le16 *regulatory,
 		   const __le16 *mac_override, const __le16 *phy_sku,
 		   u8 tx_chains, u8 rx_chains, bool lar_fw_supported,
-		   u32 mac_addr0, u32 mac_addr1)
+		   __le32 mac_addr0, __le32 mac_addr1)
 {
 	struct iwl_nvm_data *data;
 	u32 sku;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
index 92466ee..4e8e0dc 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
@@ -79,7 +79,7 @@ iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
 		   const __le16 *nvm_calib, const __le16 *regulatory,
 		   const __le16 *mac_override, const __le16 *phy_sku,
 		   u8 tx_chains, u8 rx_chains, bool lar_fw_supported,
-		   u32 mac_addr0, u32 mac_addr1);
+		   __le32 mac_addr0, __le32 mac_addr1);
 
 /**
  * iwl_parse_mcc_info - parse MCC (mobile country code) info coming from FW
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
index 7a3da2d..c446e0d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
@@ -300,7 +300,7 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm)
 	struct iwl_nvm_section *sections = mvm->nvm_sections;
 	const __le16 *hw, *sw, *calib, *regulatory, *mac_override, *phy_sku;
 	bool lar_enabled;
-	u32 mac_addr0, mac_addr1;
+	__le32 mac_addr0, mac_addr1;
 
 	/* Checking for required sections */
 	if (mvm->trans->cfg->device_family != IWL_DEVICE_FAMILY_8000) {
@@ -337,8 +337,10 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm)
 		return NULL;
 
 	/* read the mac address from WFMP registers */
-	mac_addr0 = iwl_trans_read_prph(mvm->trans, WFMP_MAC_ADDR_0);
-	mac_addr1 = iwl_trans_read_prph(mvm->trans, WFMP_MAC_ADDR_1);
+	mac_addr0 = cpu_to_le32(iwl_trans_read_prph(mvm->trans,
+						    WFMP_MAC_ADDR_0));
+	mac_addr1 = cpu_to_le32(iwl_trans_read_prph(mvm->trans,
+						    WFMP_MAC_ADDR_1));
 
 	hw = (const __le16 *)sections[mvm->cfg->nvm_hw_section_num].data;
 	sw = (const __le16 *)sections[NVM_SECTION_TYPE_SW].data;
-- 
cgit v0.10.2


From 8812182e5df31e49cd888bafa36bcccf87787940 Mon Sep 17 00:00:00 2001
From: Avri Altman <avri.altman@intel.com>
Date: Sun, 24 Jan 2016 15:18:59 +0200
Subject: iwlwifi: mvm: Remove bf_vif from iwl_power_vifs

This member is actually not needed as beacon abort
is only allowed for a bss station.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 0d03b3e..83e686d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -577,7 +577,6 @@ void iwl_mvm_power_uapsd_misbehaving_ap_notif(struct iwl_mvm *mvm,
 
 struct iwl_power_vifs {
 	struct iwl_mvm *mvm;
-	struct ieee80211_vif *bf_vif;
 	struct ieee80211_vif *bss_vif;
 	struct ieee80211_vif *p2p_vif;
 	struct ieee80211_vif *ap_vif;
@@ -650,11 +649,6 @@ static void iwl_mvm_power_get_vifs_iterator(void *_data, u8 *mac,
 		if (mvmvif->phy_ctxt)
 			if (mvmvif->phy_ctxt->id < MAX_PHYS)
 				power_iterator->bss_active = true;
-
-		if (mvmvif->bf_data.bf_enabled &&
-		    !WARN_ON(power_iterator->bf_vif))
-			power_iterator->bf_vif = vif;
-
 		break;
 
 	default:
@@ -959,21 +953,19 @@ static int iwl_mvm_power_set_ps(struct iwl_mvm *mvm)
 }
 
 static int iwl_mvm_power_set_ba(struct iwl_mvm *mvm,
-				struct iwl_power_vifs *vifs)
+				struct ieee80211_vif *vif)
 {
-	struct iwl_mvm_vif *mvmvif;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	bool ba_enable;
 
-	if (!vifs->bf_vif)
+	if (!mvmvif->bf_data.bf_enabled)
 		return 0;
 
-	mvmvif = iwl_mvm_vif_from_mac80211(vifs->bf_vif);
-
 	ba_enable = !(!mvmvif->pm_enabled || mvm->ps_disabled ||
-		      !vifs->bf_vif->bss_conf.ps ||
+		      !vif->bss_conf.ps ||
 		      iwl_mvm_vif_low_latency(mvmvif));
 
-	return iwl_mvm_update_beacon_abort(mvm, vifs->bf_vif, ba_enable);
+	return iwl_mvm_update_beacon_abort(mvm, vif, ba_enable);
 }
 
 int iwl_mvm_power_update_ps(struct iwl_mvm *mvm)
@@ -994,7 +986,10 @@ int iwl_mvm_power_update_ps(struct iwl_mvm *mvm)
 	if (ret)
 		return ret;
 
-	return iwl_mvm_power_set_ba(mvm, &vifs);
+	if (vifs.bss_vif)
+		return iwl_mvm_power_set_ba(mvm, vifs.bss_vif);
+
+	return 0;
 }
 
 int iwl_mvm_power_update_mac(struct iwl_mvm *mvm)
@@ -1029,7 +1024,10 @@ int iwl_mvm_power_update_mac(struct iwl_mvm *mvm)
 			return ret;
 	}
 
-	return iwl_mvm_power_set_ba(mvm, &vifs);
+	if (vifs.bss_vif)
+		return iwl_mvm_power_set_ba(mvm, vifs.bss_vif);
+
+	return 0;
 }
 
 int iwl_mvm_update_d0i3_power_mode(struct iwl_mvm *mvm,
-- 
cgit v0.10.2


From 416ec4f3928544fcedba5a4d98866e74ce44f369 Mon Sep 17 00:00:00 2001
From: Avri Altman <avri.altman@intel.com>
Date: Sun, 24 Jan 2016 15:56:11 +0200
Subject: iwlwifi: mvm: Remove iwl_mvm_update_beacon_abort

It is only called from iwl_mvm_power_set_ba() so simplify things
by removing it.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 83e686d..f313910 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -877,26 +877,6 @@ int iwl_mvm_enable_beacon_filter(struct iwl_mvm *mvm,
 	return _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd, flags, false);
 }
 
-static int iwl_mvm_update_beacon_abort(struct iwl_mvm *mvm,
-				       struct ieee80211_vif *vif,
-				       bool enable)
-{
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	struct iwl_beacon_filter_cmd cmd = {
-		IWL_BF_CMD_CONFIG_DEFAULTS,
-		.bf_enable_beacon_filter = cpu_to_le32(1),
-	};
-
-	if (!mvmvif->bf_data.bf_enabled)
-		return 0;
-
-	if (mvm->cur_ucode == IWL_UCODE_WOWLAN)
-		cmd.ba_escape_timer = cpu_to_le32(IWL_BA_ESCAPE_TIMER_D3);
-
-	mvmvif->bf_data.ba_enabled = enable;
-	return _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd, 0, false);
-}
-
 static int _iwl_mvm_disable_beacon_filter(struct iwl_mvm *mvm,
 					  struct ieee80211_vif *vif,
 					  u32 flags, bool d0i3)
@@ -956,16 +936,23 @@ static int iwl_mvm_power_set_ba(struct iwl_mvm *mvm,
 				struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	bool ba_enable;
+	struct iwl_beacon_filter_cmd cmd = {
+		IWL_BF_CMD_CONFIG_DEFAULTS,
+		.bf_enable_beacon_filter = cpu_to_le32(1),
+	};
 
 	if (!mvmvif->bf_data.bf_enabled)
 		return 0;
 
-	ba_enable = !(!mvmvif->pm_enabled || mvm->ps_disabled ||
-		      !vif->bss_conf.ps ||
-		      iwl_mvm_vif_low_latency(mvmvif));
+	if (mvm->cur_ucode == IWL_UCODE_WOWLAN)
+		cmd.ba_escape_timer = cpu_to_le32(IWL_BA_ESCAPE_TIMER_D3);
+
+	mvmvif->bf_data.ba_enabled = !(!mvmvif->pm_enabled ||
+				       mvm->ps_disabled ||
+				       !vif->bss_conf.ps ||
+				       iwl_mvm_vif_low_latency(mvmvif));
 
-	return iwl_mvm_update_beacon_abort(mvm, vif, ba_enable);
+	return _iwl_mvm_enable_beacon_filter(mvm, vif, &cmd, 0, false);
 }
 
 int iwl_mvm_power_update_ps(struct iwl_mvm *mvm)
-- 
cgit v0.10.2


From 854c57057390e45041a56c3d3228eb805e980c99 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 26 Jan 2016 13:17:47 +0200
Subject: iwlwifi: mvm: add new ADD_STA command version

The 9000 hardware introduces the frame releaser, which
keeps track of the aggregation window and notifies host
of the window status. This requires in turn updating
the hardware with the RX BA session window size.
Firmware API was changed to enable that, update the driver
accordingly.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h
index 1eb3983..90d9113 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -257,6 +259,65 @@ struct iwl_mvm_keyinfo {
 #define IWL_ADD_STA_BAID_MASK	0xFF00
 
 /**
+ * struct iwl_mvm_add_sta_cmd_v7 - Add/modify a station in the fw's sta table.
+ * ( REPLY_ADD_STA = 0x18 )
+ * @add_modify: 1: modify existing, 0: add new station
+ * @awake_acs:
+ * @tid_disable_tx: is tid BIT(tid) enabled for Tx. Clear BIT(x) to enable
+ *	AMPDU for tid x. Set %STA_MODIFY_TID_DISABLE_TX to change this field.
+ * @mac_id_n_color: the Mac context this station belongs to
+ * @addr[ETH_ALEN]: station's MAC address
+ * @sta_id: index of station in uCode's station table
+ * @modify_mask: STA_MODIFY_*, selects which parameters to modify vs. leave
+ *	alone. 1 - modify, 0 - don't change.
+ * @station_flags: look at %iwl_sta_flags
+ * @station_flags_msk: what of %station_flags have changed
+ * @add_immediate_ba_tid: tid for which to add block-ack support (Rx)
+ *	Set %STA_MODIFY_ADD_BA_TID to use this field, and also set
+ *	add_immediate_ba_ssn.
+ * @remove_immediate_ba_tid: tid for which to remove block-ack support (Rx)
+ *	Set %STA_MODIFY_REMOVE_BA_TID to use this field
+ * @add_immediate_ba_ssn: ssn for the Rx block-ack session. Used together with
+ *	add_immediate_ba_tid.
+ * @sleep_tx_count: number of packets to transmit to station even though it is
+ *	asleep. Used to synchronise PS-poll and u-APSD responses while ucode
+ *	keeps track of STA sleep state.
+ * @sleep_state_flags: Look at %iwl_sta_sleep_flag.
+ * @assoc_id: assoc_id to be sent in VHT PLCP (9-bit), for grp use 0, for AP
+ *	mac-addr.
+ * @beamform_flags: beam forming controls
+ * @tfd_queue_msk: tfd queues used by this station
+ *
+ * The device contains an internal table of per-station information, with info
+ * on security keys, aggregation parameters, and Tx rates for initial Tx
+ * attempt and any retries (set by REPLY_TX_LINK_QUALITY_CMD).
+ *
+ * ADD_STA sets up the table entry for one station, either creating a new
+ * entry, or modifying a pre-existing one.
+ */
+struct iwl_mvm_add_sta_cmd_v7 {
+	u8 add_modify;
+	u8 awake_acs;
+	__le16 tid_disable_tx;
+	__le32 mac_id_n_color;
+	u8 addr[ETH_ALEN];	/* _STA_ID_MODIFY_INFO_API_S_VER_1 */
+	__le16 reserved2;
+	u8 sta_id;
+	u8 modify_mask;
+	__le16 reserved3;
+	__le32 station_flags;
+	__le32 station_flags_msk;
+	u8 add_immediate_ba_tid;
+	u8 remove_immediate_ba_tid;
+	__le16 add_immediate_ba_ssn;
+	__le16 sleep_tx_count;
+	__le16 sleep_state_flags;
+	__le16 assoc_id;
+	__le16 beamform_flags;
+	__le32 tfd_queue_msk;
+} __packed; /* ADD_STA_CMD_API_S_VER_7 */
+
+/**
  * struct iwl_mvm_add_sta_cmd - Add/modify a station in the fw's sta table.
  * ( REPLY_ADD_STA = 0x18 )
  * @add_modify: 1: modify existing, 0: add new station
@@ -285,6 +346,7 @@ struct iwl_mvm_keyinfo {
  *	mac-addr.
  * @beamform_flags: beam forming controls
  * @tfd_queue_msk: tfd queues used by this station
+ * @rx_ba_window: aggregation window size
  *
  * The device contains an internal table of per-station information, with info
  * on security keys, aggregation parameters, and Tx rates for initial Tx
@@ -313,7 +375,9 @@ struct iwl_mvm_add_sta_cmd {
 	__le16 assoc_id;
 	__le16 beamform_flags;
 	__le32 tfd_queue_msk;
-} __packed; /* ADD_STA_CMD_API_S_VER_7 */
+	__le16 rx_ba_window;
+	__le16 reserved;
+} __packed; /* ADD_STA_CMD_API_S_VER_8 */
 
 /**
  * struct iwl_mvm_add_sta_key_cmd - add/modify sta key
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index d70a171..01476f5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -884,10 +884,10 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 			ret = -EINVAL;
 			break;
 		}
-		ret = iwl_mvm_sta_rx_agg(mvm, sta, tid, *ssn, true);
+		ret = iwl_mvm_sta_rx_agg(mvm, sta, tid, *ssn, true, buf_size);
 		break;
 	case IEEE80211_AMPDU_RX_STOP:
-		ret = iwl_mvm_sta_rx_agg(mvm, sta, tid, 0, false);
+		ret = iwl_mvm_sta_rx_agg(mvm, sta, tid, 0, false, buf_size);
 		break;
 	case IEEE80211_AMPDU_TX_START:
 		if (!iwl_enable_tx_ampdu(mvm->cfg)) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 84b7fd1..4854e79 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2015 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2015 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -68,6 +70,18 @@
 #include "sta.h"
 #include "rs.h"
 
+/*
+ * New version of ADD_STA_sta command added new fields at the end of the
+ * structure, so sending the size of the relevant API's structure is enough to
+ * support both API versions.
+ */
+static inline int iwl_mvm_add_sta_cmd_size(struct iwl_mvm *mvm)
+{
+	return iwl_mvm_has_new_rx_api(mvm) ?
+		sizeof(struct iwl_mvm_add_sta_cmd) :
+		sizeof(struct iwl_mvm_add_sta_cmd_v7);
+}
+
 static int iwl_mvm_find_free_sta_id(struct iwl_mvm *mvm,
 				    enum nl80211_iftype iftype)
 {
@@ -187,7 +201,8 @@ int iwl_mvm_sta_send_to_fw(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		cpu_to_le32(mpdu_dens << STA_FLG_AGG_MPDU_DENS_SHIFT);
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(add_sta_cmd),
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA,
+					  iwl_mvm_add_sta_cmd_size(mvm),
 					  &add_sta_cmd, &status);
 	if (ret)
 		return ret;
@@ -357,7 +372,8 @@ int iwl_mvm_drain_sta(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta,
 	cmd.station_flags_msk = cpu_to_le32(STA_FLG_DRAIN_FLOW);
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA,
+					  iwl_mvm_add_sta_cmd_size(mvm),
 					  &cmd, &status);
 	if (ret)
 		return ret;
@@ -623,7 +639,8 @@ static int iwl_mvm_add_int_sta_common(struct iwl_mvm *mvm,
 	if (addr)
 		memcpy(cmd.addr, addr, ETH_ALEN);
 
-	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA,
+					  iwl_mvm_add_sta_cmd_size(mvm),
 					  &cmd, &status);
 	if (ret)
 		return ret;
@@ -819,7 +836,7 @@ int iwl_mvm_rm_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 #define IWL_MAX_RX_BA_SESSIONS 16
 
 int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
-		       int tid, u16 ssn, bool start)
+		       int tid, u16 ssn, bool start, u8 buf_size)
 {
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_add_sta_cmd cmd = {};
@@ -839,6 +856,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	if (start) {
 		cmd.add_immediate_ba_tid = (u8) tid;
 		cmd.add_immediate_ba_ssn = cpu_to_le16(ssn);
+		cmd.rx_ba_window = cpu_to_le16((u16)buf_size);
 	} else {
 		cmd.remove_immediate_ba_tid = (u8) tid;
 	}
@@ -846,7 +864,8 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 				  STA_MODIFY_REMOVE_BA_TID;
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA,
+					  iwl_mvm_add_sta_cmd_size(mvm),
 					  &cmd, &status);
 	if (ret)
 		return ret;
@@ -904,7 +923,8 @@ static int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	cmd.tid_disable_tx = cpu_to_le16(mvm_sta->tid_disable_agg);
 
 	status = ADD_STA_SUCCESS;
-	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, sizeof(cmd),
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA,
+					  iwl_mvm_add_sta_cmd_size(mvm),
 					  &cmd, &status);
 	if (ret)
 		return ret;
@@ -1640,7 +1660,8 @@ void iwl_mvm_sta_modify_ps_wake(struct iwl_mvm *mvm,
 	};
 	int ret;
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC, sizeof(cmd), &cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC,
+				   iwl_mvm_add_sta_cmd_size(mvm), &cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send ADD_STA command (%d)\n", ret);
 }
@@ -1731,7 +1752,7 @@ void iwl_mvm_sta_modify_sleep_tx_count(struct iwl_mvm *mvm,
 
 	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA,
 				   CMD_ASYNC | CMD_WANT_ASYNC_CALLBACK,
-				   sizeof(cmd), &cmd);
+				   iwl_mvm_add_sta_cmd_size(mvm), &cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send ADD_STA command (%d)\n", ret);
 }
@@ -1766,7 +1787,8 @@ void iwl_mvm_sta_modify_disable_tx(struct iwl_mvm *mvm,
 	};
 	int ret;
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC, sizeof(cmd), &cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, ADD_STA, CMD_ASYNC,
+				   iwl_mvm_add_sta_cmd_size(mvm), &cmd);
 	if (ret)
 		IWL_ERR(mvm, "Failed to send ADD_STA command (%d)\n", ret);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 39fdf52..e3b9446 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -401,7 +401,7 @@ void iwl_mvm_rx_eosp_notif(struct iwl_mvm *mvm,
 
 /* AMPDU */
 int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
-		       int tid, u16 ssn, bool start);
+		       int tid, u16 ssn, bool start, u8 buf_size);
 int iwl_mvm_sta_tx_agg_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta, u16 tid, u16 *ssn);
 int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-- 
cgit v0.10.2


From 6ad6c01fa74f649722eeca7308dc1fc4e43235cd Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Wed, 27 Jan 2016 00:10:15 +0200
Subject: iwlwifi: mvm: rs: fix TPC action decision algorithm

Decreasing Tx power is allowed only when success ratio is
above the threshold defined in the algorithm. Add this condition.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 7bb6fd0..986e12a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -2040,7 +2040,8 @@ static enum tpc_action rs_get_tpc_action(struct iwl_mvm *mvm,
 	}
 
 	/* try decreasing first if applicable */
-	if (weak != TPC_INVALID) {
+	if (sr >= RS_PERCENT(IWL_MVM_RS_TPC_SR_NO_INCREASE) &&
+	    weak != TPC_INVALID) {
 		if (weak_tpt == IWL_INVALID_VALUE &&
 		    (strong_tpt == IWL_INVALID_VALUE ||
 		     current_tpt >= strong_tpt)) {
-- 
cgit v0.10.2


From 43413a975d06e5e34016751fac27e29ec3d4d10f Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Thu, 31 Dec 2015 11:49:18 +0200
Subject: iwlwifi: mvm: support rss queues configuration command

9000 series supports multi-queue rx. The hardware needs
to be configured with the hash functions to perform and
indirection table that maps hash results to the relevant
CPUs\queues.
Support this configuration.
Add debugfs hook to configure the indirection table in
order to enable performance analysis. The configuration
is stateless, receives a partial or full pattern and sends
the command to the firmware.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 5c0f939..005cc09 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -943,6 +944,47 @@ iwl_dbgfs_scan_ant_rxchain_write(struct iwl_mvm *mvm, char *buf,
 	return count;
 }
 
+static ssize_t iwl_dbgfs_indirection_tbl_write(struct iwl_mvm *mvm,
+					       char *buf, size_t count,
+					       loff_t *ppos)
+{
+	struct iwl_rss_config_cmd cmd = {
+		.flags = cpu_to_le32(IWL_RSS_ENABLE),
+		.hash_mask = IWL_RSS_HASH_TYPE_IPV4_TCP |
+			     IWL_RSS_HASH_TYPE_IPV4_PAYLOAD |
+			     IWL_RSS_HASH_TYPE_IPV6_TCP |
+			     IWL_RSS_HASH_TYPE_IPV6_PAYLOAD,
+	};
+	int ret, i, num_repeats, nbytes = count / 2;
+
+	ret = hex2bin(cmd.indirection_table, buf, nbytes);
+	if (ret)
+		return ret;
+
+	/*
+	 * The input is the redirection table, partial or full.
+	 * Repeat the pattern if needed.
+	 * For example, input of 01020F will be repeated 42 times,
+	 * indirecting RSS hash results to queues 1, 2, 15 (skipping
+	 * queues 3 - 14).
+	 */
+	num_repeats = ARRAY_SIZE(cmd.indirection_table) / nbytes;
+	for (i = 1; i < num_repeats; i++)
+		memcpy(&cmd.indirection_table[i * nbytes],
+		       cmd.indirection_table, nbytes);
+	/* handle cut in the middle pattern for the last places */
+	memcpy(&cmd.indirection_table[i * nbytes], cmd.indirection_table,
+	       ARRAY_SIZE(cmd.indirection_table) % nbytes);
+
+	memcpy(cmd.secret_key, mvm->secret_key, ARRAY_SIZE(cmd.secret_key));
+
+	mutex_lock(&mvm->mutex);
+	ret = iwl_mvm_send_cmd_pdu(mvm, RSS_CONFIG_CMD, 0, sizeof(cmd), &cmd);
+	mutex_unlock(&mvm->mutex);
+
+	return ret ?: count;
+}
+
 static ssize_t iwl_dbgfs_fw_dbg_conf_read(struct file *file,
 					  char __user *user_buf,
 					  size_t count, loff_t *ppos)
@@ -1455,6 +1497,7 @@ MVM_DEBUGFS_READ_WRITE_FILE_OPS(d0i3_refs, 8);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(fw_dbg_conf, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_dbg_collect, 64);
 MVM_DEBUGFS_WRITE_FILE_OPS(cont_recording, 8);
+MVM_DEBUGFS_WRITE_FILE_OPS(indirection_tbl, 16);
 
 #ifdef CONFIG_IWLWIFI_BCAST_FILTERING
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(bcast_filters, 256);
@@ -1499,6 +1542,7 @@ int iwl_mvm_dbgfs_register(struct iwl_mvm *mvm, struct dentry *dbgfs_dir)
 	MVM_DEBUGFS_ADD_FILE(fw_dbg_collect, mvm->debugfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(send_echo_cmd, mvm->debugfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(cont_recording, mvm->debugfs_dir, S_IWUSR);
+	MVM_DEBUGFS_ADD_FILE(indirection_tbl, mvm->debugfs_dir, S_IWUSR);
 	if (!debugfs_create_bool("enable_scan_iteration_notif",
 				 S_IRUSR | S_IWUSR,
 				 mvm->debugfs_dir,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
index ab467cb..b45c61e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
@@ -7,7 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
- * Copyright(c) 2015        Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -34,7 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
- * Copyright(c) 2015        Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -362,4 +362,33 @@ struct iwl_frame_release {
 	__le16 nssn;
 };
 
+enum iwl_rss_hash_func_en {
+	IWL_RSS_HASH_TYPE_IPV4_TCP,
+	IWL_RSS_HASH_TYPE_IPV4_UDP,
+	IWL_RSS_HASH_TYPE_IPV4_PAYLOAD,
+	IWL_RSS_HASH_TYPE_IPV6_TCP,
+	IWL_RSS_HASH_TYPE_IPV6_UDP,
+	IWL_RSS_HASH_TYPE_IPV6_PAYLOAD,
+};
+
+#define IWL_RSS_HASH_KEY_CNT 10
+#define IWL_RSS_INDIRECTION_TABLE_SIZE 128
+#define IWL_RSS_ENABLE 1
+
+/**
+ * struct iwl_rss_config_cmd - RSS (Receive Side Scaling) configuration
+ *
+ * @flags: 1 - enable, 0 - disable
+ * @hash_mask: Type of RSS to use. Values are from %iwl_rss_hash_func_en
+ * @secret_key: 320 bit input of random key configuration from driver
+ * @indirection_table: indirection table
+ */
+struct iwl_rss_config_cmd {
+	__le32 flags;
+	u8 hash_mask;
+	u8 reserved[3];
+	__le32 secret_key[IWL_RSS_HASH_KEY_CNT];
+	u8 indirection_table[IWL_RSS_INDIRECTION_TABLE_SIZE];
+} __packed; /* RSS_CONFIG_CMD_API_S_VER_1 */
+
 #endif /* __fw_api_rx_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index b6b5727..f332497 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -213,6 +213,8 @@ enum {
 
 	MFUART_LOAD_NOTIFICATION = 0xb1,
 
+	RSS_CONFIG_CMD = 0xb3,
+
 	REPLY_RX_PHY_CMD = 0xc0,
 	REPLY_RX_MPDU_CMD = 0xc1,
 	FRAME_RELEASE = 0xc3,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 4ed5180..070e2af 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -107,6 +108,24 @@ static int iwl_send_tx_ant_cfg(struct iwl_mvm *mvm, u8 valid_tx_ant)
 				    sizeof(tx_ant_cmd), &tx_ant_cmd);
 }
 
+static int iwl_send_rss_cfg_cmd(struct iwl_mvm *mvm)
+{
+	int i;
+	struct iwl_rss_config_cmd cmd = {
+		.flags = cpu_to_le32(IWL_RSS_ENABLE),
+		.hash_mask = IWL_RSS_HASH_TYPE_IPV4_TCP |
+			     IWL_RSS_HASH_TYPE_IPV4_PAYLOAD |
+			     IWL_RSS_HASH_TYPE_IPV6_TCP |
+			     IWL_RSS_HASH_TYPE_IPV6_PAYLOAD,
+	};
+
+	for (i = 0; i < ARRAY_SIZE(cmd.indirection_table); i++)
+		cmd.indirection_table[i] = i % mvm->trans->num_rx_queues;
+	memcpy(cmd.secret_key, mvm->secret_key, ARRAY_SIZE(cmd.secret_key));
+
+	return iwl_mvm_send_cmd_pdu(mvm, RSS_CONFIG_CMD, 0, sizeof(cmd), &cmd);
+}
+
 static void iwl_free_fw_paging(struct iwl_mvm *mvm)
 {
 	int i;
@@ -894,6 +913,16 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	if (ret)
 		goto error;
 
+	/* Init RSS configuration */
+	if (iwl_mvm_has_new_rx_api(mvm)) {
+		ret = iwl_send_rss_cfg_cmd(mvm);
+		if (ret) {
+			IWL_ERR(mvm, "Failed to configure RSS queues: %d\n",
+				ret);
+			goto error;
+		}
+	}
+
 	/* init the fw <-> mac80211 STA mapping */
 	for (i = 0; i < IWL_MVM_STATION_COUNT; i++)
 		RCU_INIT_POINTER(mvm->fw_id_to_mac_id[i], NULL);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index f87aa97..747f7eb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -647,6 +647,7 @@ struct iwl_mvm {
 	atomic_t pending_frames[IWL_MVM_STATION_COUNT];
 	u32 tfd_drained[IWL_MVM_STATION_COUNT];
 	u8 rx_ba_sessions;
+	u32 secret_key[IWL_RSS_HASH_KEY_CNT];
 
 	/* configured by mac80211 */
 	u32 rts_threshold;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 3c869ad..325ff8a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -347,6 +347,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
 	HCMD_NAME(MAC_PM_POWER_TABLE),
 	HCMD_NAME(TDLS_CHANNEL_SWITCH_NOTIFICATION),
 	HCMD_NAME(MFUART_LOAD_NOTIFICATION),
+	HCMD_NAME(RSS_CONFIG_CMD),
 	HCMD_NAME(SCAN_ITERATION_COMPLETE_UMAC),
 	HCMD_NAME(REPLY_RX_PHY_CMD),
 	HCMD_NAME(REPLY_RX_MPDU_CMD),
@@ -651,6 +652,9 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 
 	iwl_mvm_tof_init(mvm);
 
+	/* init RSS hash key */
+	get_random_bytes(mvm->secret_key, ARRAY_SIZE(mvm->secret_key));
+
 	return op_mode;
 
  out_unregister:
-- 
cgit v0.10.2


From bce977310416a2b6c9ad774dfe2071eaacd3d90b Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 25 Jan 2016 18:14:49 +0200
Subject: iwlwifi: pcie: enable multi-queue rx path

Previous patches enabled new 9000 hardware DMA for one queue
only.
Enable the actual multi-queue path and configuration now.
This requires also per-queue NAPI struct.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 7bc02e0..2f95916 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -136,6 +136,7 @@ struct iwl_rxq {
 	struct iwl_rb_status *rb_stts;
 	dma_addr_t rb_stts_dma;
 	spinlock_t lock;
+	struct napi_struct napi;
 	struct iwl_rx_mem_buffer *queue[RX_QUEUE_SIZE];
 };
 
@@ -345,7 +346,6 @@ struct iwl_trans_pcie {
 	struct iwl_drv *drv;
 
 	struct net_device napi_dev;
-	struct napi_struct napi;
 
 	struct __percpu iwl_tso_hdr_page *tso_hdr_page;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index a385f3c..51314e56 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -2,6 +2,7 @@
  *
  * Copyright(c) 2003 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * Portions of this file are derived from the ipw3945 project, as well
  * as portions of the ieee80211 subsystem header files.
@@ -730,7 +731,7 @@ static void iwl_pcie_rx_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq)
 		iwl_set_bit(trans, CSR_INT_COALESCING, IWL_HOST_INT_OPER_MODE);
 }
 
-static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq)
+static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	u32 rb_size, enabled = 0;
@@ -759,13 +760,13 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq)
 	for (i = 0; i < trans->num_rx_queues; i++) {
 		/* Tell device where to find RBD free table in DRAM */
 		iwl_pcie_write_prph_64(trans, RFH_Q_FRBDCB_BA_LSB(i),
-				       (u64)(rxq->bd_dma));
+				       (u64)(trans_pcie->rxq[i].bd_dma));
 		/* Tell device where to find RBD used table in DRAM */
 		iwl_pcie_write_prph_64(trans, RFH_Q_URBDCB_BA_LSB(i),
-				       (u64)(rxq->used_bd_dma));
+				       (u64)(trans_pcie->rxq[i].used_bd_dma));
 		/* Tell device where in DRAM to update its Rx status */
 		iwl_pcie_write_prph_64(trans, RFH_Q_URBD_STTS_WPTR_LSB(i),
-				       rxq->rb_stts_dma);
+				       trans_pcie->rxq[i].rb_stts_dma);
 		/* Reset device indice tables */
 		iwl_write_prph(trans, RFH_Q_FRBDCB_WIDX(i), 0);
 		iwl_write_prph(trans, RFH_Q_FRBDCB_RIDX(i), 0);
@@ -808,6 +809,12 @@ static void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq)
 	rxq->used_count = 0;
 }
 
+static int iwl_pcie_dummy_napi_poll(struct napi_struct *napi, int budget)
+{
+	WARN_ON(1);
+	return 0;
+}
+
 int iwl_pcie_rx_init(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
@@ -857,6 +864,10 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 
 		iwl_pcie_rx_init_rxb_lists(rxq);
 
+		if (!rxq->napi.poll)
+			netif_napi_add(&trans_pcie->napi_dev, &rxq->napi,
+				       iwl_pcie_dummy_napi_poll, 64);
+
 		spin_unlock(&rxq->lock);
 	}
 
@@ -878,7 +889,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 
 	iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL, def_rxq);
 	if (trans->cfg->mq_rx_supported) {
-		iwl_pcie_rx_mq_hw_init(trans, def_rxq);
+		iwl_pcie_rx_mq_hw_init(trans);
 	} else {
 		iwl_pcie_rxq_restock(trans, def_rxq);
 		iwl_pcie_rx_hw_init(trans, def_rxq);
@@ -940,6 +951,9 @@ void iwl_pcie_rx_free(struct iwl_trans *trans)
 					  rxq->used_bd, rxq->used_bd_dma);
 		rxq->used_bd_dma = 0;
 		rxq->used_bd = NULL;
+
+		if (rxq->napi.poll)
+			netif_napi_del(&rxq->napi);
 	}
 	kfree(trans_pcie->rxq);
 }
@@ -1055,7 +1069,12 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 		index = SEQ_TO_INDEX(sequence);
 		cmd_index = get_cmd_index(&txq->q, index);
 
-		iwl_op_mode_rx(trans->op_mode, &trans_pcie->napi, &rxcb);
+		if (rxq->id == 0)
+			iwl_op_mode_rx(trans->op_mode, &rxq->napi,
+				       &rxcb);
+		else
+			iwl_op_mode_rx_rss(trans->op_mode, &rxq->napi,
+					   &rxcb, rxq->id);
 
 		if (reclaim) {
 			kzfree(txq->entries[cmd_index].free_buf);
@@ -1236,8 +1255,8 @@ restart:
 	if (unlikely(emergency && count))
 		iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
 
-	if (trans_pcie->napi.poll)
-		napi_gro_flush(&trans_pcie->napi, false);
+	if (rxq->napi.poll)
+		napi_gro_flush(&rxq->napi, false);
 }
 
 /*
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index abe0916..b796952 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1428,12 +1428,6 @@ static void iwl_trans_pcie_write_prph(struct iwl_trans *trans, u32 addr,
 	iwl_trans_pcie_write32(trans, HBUS_TARG_PRPH_WDAT, val);
 }
 
-static int iwl_pcie_dummy_napi_poll(struct napi_struct *napi, int budget)
-{
-	WARN_ON(1);
-	return 0;
-}
-
 static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 				     const struct iwl_trans_config *trans_cfg)
 {
@@ -1470,11 +1464,8 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 	 * As this function may be called again in some corner cases don't
 	 * do anything if NAPI was already initialized.
 	 */
-	if (!trans_pcie->napi.poll) {
+	if (trans_pcie->napi_dev.reg_state != NETREG_DUMMY)
 		init_dummy_netdev(&trans_pcie->napi_dev);
-		netif_napi_add(&trans_pcie->napi_dev, &trans_pcie->napi,
-			       iwl_pcie_dummy_napi_poll, 64);
-	}
 }
 
 void iwl_trans_pcie_free(struct iwl_trans *trans)
@@ -1498,9 +1489,6 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	pci_release_regions(trans_pcie->pci_dev);
 	pci_disable_device(trans_pcie->pci_dev);
 
-	if (trans_pcie->napi.poll)
-		netif_napi_del(&trans_pcie->napi);
-
 	iwl_pcie_free_fw_monitor(trans);
 
 	for_each_possible_cpu(i) {
-- 
cgit v0.10.2


From d56a7801b544d63b0d32bc8bca0c12a259b1d8e9 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 26 Jan 2016 12:35:13 +0200
Subject: iwlwifi: pcie: update iwl_mpdu_desc fields

Final API of iwl_mpdu_desc has a change in the order of
the fields and does not include energy from the third
antenna (which is perfectly fine, since we don't have one).
Update the structure accordingly.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
index b45c61e..df939f5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
@@ -347,11 +347,11 @@ struct iwl_rx_mpdu_desc {
 	/* DW8 */
 	__le32 filter_match;
 	/* DW9 */
-	__le32 gp2_on_air_rise;
-	/* DW10 */
 	__le32 rate_n_flags;
+	/* DW10 */
+	u8 energy_a, energy_b, channel, reserved;
 	/* DW11 */
-	u8 energy_a, energy_b, energy_c, channel;
+	__le32 gp2_on_air_rise;
 	/* DW12 & DW13 */
 	__le64 tsf_on_air_rise;
 } __packed;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 4cce372..615dea1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -201,25 +201,22 @@ static void iwl_mvm_get_signal_strength(struct iwl_mvm *mvm,
 					struct iwl_rx_mpdu_desc *desc,
 					struct ieee80211_rx_status *rx_status)
 {
-	int energy_a, energy_b, energy_c, max_energy;
+	int energy_a, energy_b, max_energy;
 
 	energy_a = desc->energy_a;
 	energy_a = energy_a ? -energy_a : S8_MIN;
 	energy_b = desc->energy_b;
 	energy_b = energy_b ? -energy_b : S8_MIN;
-	energy_c = desc->energy_c;
-	energy_c = energy_c ? -energy_c : S8_MIN;
 	max_energy = max(energy_a, energy_b);
-	max_energy = max(max_energy, energy_c);
 
-	IWL_DEBUG_STATS(mvm, "energy In A %d B %d C %d , and max %d\n",
-			energy_a, energy_b, energy_c, max_energy);
+	IWL_DEBUG_STATS(mvm, "energy In A %d B %d, and max %d\n",
+			energy_a, energy_b, max_energy);
 
 	rx_status->signal = max_energy;
 	rx_status->chains = 0; /* TODO: phy info */
 	rx_status->chain_signal[0] = energy_a;
 	rx_status->chain_signal[1] = energy_b;
-	rx_status->chain_signal[2] = energy_c;
+	rx_status->chain_signal[2] = S8_MIN;
 }
 
 static int iwl_mvm_rx_crypto(struct iwl_mvm *mvm, struct ieee80211_hdr *hdr,
-- 
cgit v0.10.2


From c89e333d5e4a27f736c960b29040b261c21563a5 Mon Sep 17 00:00:00 2001
From: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Date: Tue, 26 Jan 2016 18:12:28 +0200
Subject: iwlwifi: mvm: allow to disable beacon filtering for AP/GO interface

When in AP mode we need to filter in beacons from other APs to update HT
operation mode. As a power optimization the beacons are filtered out when
there are no associated stations. As a result, when there are no
associated stations, we will not update the HT operation mode until a
station connects.
Add a debugfs parameter that allows to disable this optimization.

Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 005cc09..c529e53 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -1548,6 +1548,9 @@ int iwl_mvm_dbgfs_register(struct iwl_mvm *mvm, struct dentry *dbgfs_dir)
 				 mvm->debugfs_dir,
 				 &mvm->scan_iter_notif_enabled))
 		goto err;
+	if (!debugfs_create_bool("drop_bcn_ap_mode", S_IRUSR | S_IWUSR,
+				 mvm->debugfs_dir, &mvm->drop_bcn_ap_mode))
+		goto err;
 
 #ifdef CONFIG_IWLWIFI_BCAST_FILTERING
 	if (mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_BCAST_FILTERING) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 62927f5..535134d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -744,7 +744,7 @@ static void iwl_mvm_mac_ctxt_cmd_common(struct iwl_mvm *mvm,
 		 * wake-ups.
 		 */
 		cmd->filter_flags |= cpu_to_le32(MAC_FILTER_IN_PROBE_REQUEST);
-		if (mvmvif->ap_assoc_sta_count) {
+		if (mvmvif->ap_assoc_sta_count || !mvm->drop_bcn_ap_mode) {
 			cmd->filter_flags |= cpu_to_le32(MAC_FILTER_IN_BEACON);
 			IWL_DEBUG_HC(mvm, "Asking FW to pass beacons\n");
 		} else {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 747f7eb..ebe37bb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -859,6 +859,12 @@ struct iwl_mvm {
 
 	u32 ciphers[6];
 	struct iwl_mvm_tof_data tof_data;
+
+	/*
+	 * Drop beacons from other APs in AP mode when there are no connected
+	 * clients.
+	 */
+	bool drop_bcn_ap_mode;
 };
 
 /* Extract MVM priv from op_mode and _hw */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 325ff8a..09a94a5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -492,6 +492,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	}
 	mvm->sf_state = SF_UNINIT;
 	mvm->cur_ucode = IWL_UCODE_INIT;
+	mvm->drop_bcn_ap_mode = true;
 
 	mutex_init(&mvm->mutex);
 	mutex_init(&mvm->d0i3_suspend_mutex);
-- 
cgit v0.10.2


From 6f68b002715460c7552e8688b6f289659d5f8046 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 6 Sep 2015 21:38:49 +0200
Subject: batman-adv: Fix kerneldoc parsing of structs/enums

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/multicast.h b/net/batman-adv/multicast.h
index 8f3cb04..685219f 100644
--- a/net/batman-adv/multicast.h
+++ b/net/batman-adv/multicast.h
@@ -23,7 +23,7 @@
 struct sk_buff;
 
 /**
- * batadv_forw_mode - the way a packet should be forwarded as
+ * enum batadv_forw_mode - the way a packet should be forwarded as
  * @BATADV_FORW_ALL: forward the packet to all nodes (currently via classic
  *  flooding)
  * @BATADV_FORW_SINGLE: forward the packet to a single node (currently via the
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index 0558e32..e80a8fc 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -158,7 +158,7 @@ enum batadv_tt_client_flags {
 };
 
 /**
- * batadv_vlan_flags - flags for the four MSB of any vlan ID field
+ * enum batadv_vlan_flags - flags for the four MSB of any vlan ID field
  * @BATADV_VLAN_HAS_TAG: whether the field contains a valid vlan tag or not
  */
 enum batadv_vlan_flags {
@@ -230,7 +230,7 @@ struct batadv_ogm_packet {
 #define BATADV_OGM_HLEN sizeof(struct batadv_ogm_packet)
 
 /**
- * batadv_icmp_header - common members among all the ICMP packets
+ * struct batadv_icmp_header - common members among all the ICMP packets
  * @packet_type: batman-adv packet type, part of the general header
  * @version: batman-adv protocol version, part of the genereal header
  * @ttl: time to live for this packet, part of the genereal header
@@ -256,7 +256,7 @@ struct batadv_icmp_header {
 };
 
 /**
- * batadv_icmp_packet - ICMP packet
+ * struct batadv_icmp_packet - ICMP packet
  * @packet_type: batman-adv packet type, part of the general header
  * @version: batman-adv protocol version, part of the genereal header
  * @ttl: time to live for this packet, part of the genereal header
@@ -282,7 +282,7 @@ struct batadv_icmp_packet {
 #define BATADV_RR_LEN 16
 
 /**
- * batadv_icmp_packet_rr - ICMP RouteRecord packet
+ * struct batadv_icmp_packet_rr - ICMP RouteRecord packet
  * @packet_type: batman-adv packet type, part of the general header
  * @version: batman-adv protocol version, part of the genereal header
  * @ttl: time to live for this packet, part of the genereal header
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 3437b66..8c3dda3 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -346,7 +346,7 @@ struct batadv_gw_node {
 };
 
 /**
- * batadv_hardif_neigh_node - unique neighbor per hard interface
+ * struct batadv_hardif_neigh_node - unique neighbor per hard interface
  * @list: list node for batadv_hard_iface::neigh_list
  * @addr: the MAC address of the neighboring interface
  * @if_incoming: pointer to incoming hard interface
-- 
cgit v0.10.2


From 42ea22fe2c5da2ebd7123518a1b2cd7597ebffd7 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 6 Sep 2015 21:38:50 +0200
Subject: batman-adv: Remove kerneldoc for missing parameters

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index 20d9282..fedddbe 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -242,7 +242,6 @@ err:
 /**
  * batadv_frag_merge_packets - merge a chain of fragments
  * @chain: head of chain with fragments
- * @skb: packet with total size of skb after merging
  *
  * Expand the first skb in the chain and copy the content of the remaining
  * skb's into the expanded one. After doing so, clear the chain.
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index e80a8fc..6d5b178 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -413,7 +413,6 @@ struct batadv_bcast_packet {
  * @packet_type: batman-adv packet type, part of the general header
  * @version: batman-adv protocol version, part of the genereal header
  * @ttl: time to live for this packet, part of the genereal header
- * @reserved: Align following fields to 2-byte boundaries
  * @first_source: original source of first included packet
  * @first_orig_dest: original destinal of first included packet
  * @first_crc: checksum of first included packet
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index e4f2646..54d6608 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -904,7 +904,6 @@ rx_success:
  * batadv_recv_unicast_tvlv - receive and process unicast tvlv packets
  * @skb: unicast tvlv packet to process
  * @recv_if: pointer to interface this packet was received on
- * @dst_addr: the payload destination
  *
  * Returns NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
  * otherwise.
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index cdfc85f..3ccede5 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -217,7 +217,6 @@ batadv_tt_global_entry_free_ref(struct batadv_tt_global_entry *tt_global_entry)
 
 /**
  * batadv_tt_global_hash_count - count the number of orig entries
- * @hash: hash table containing the tt entries
  * @addr: the mac address of the client to count entries for
  * @vid: VLAN identifier
  *
@@ -288,7 +287,6 @@ static void batadv_tt_local_size_dec(struct batadv_priv *bat_priv,
 /**
  * batadv_tt_global_size_mod - change the size by v of the local table
  *  identified by vid
- * @bat_priv: the bat priv with all the soft interface information
  * @vid: the VLAN identifier
  * @v: the amount to sum to the global table size
  */
@@ -2406,7 +2404,6 @@ static void batadv_tt_tvlv_generate(struct batadv_priv *bat_priv,
  * @orig_node: originator for which the CRCs have to be checked
  * @tt_vlan: pointer to the first tvlv VLAN entry
  * @num_vlan: number of tvlv VLAN entries
- * @create: if true, create VLAN objects if not found
  *
  * Return true if all the received CRCs match the locally stored ones, false
  * otherwise
@@ -3375,7 +3372,6 @@ out:
  * @tt_change: pointer to the first entry in the TT buffer
  * @tt_num_changes: number of tt changes inside the tt buffer
  * @ttvn: translation table version number of this changeset
- * @tt_crc: crc32 checksum of orig node's translation table
  */
 static void batadv_tt_update_orig(struct batadv_priv *bat_priv,
 				  struct batadv_orig_node *orig_node,
-- 
cgit v0.10.2


From e51f0397ded5bcbc2dd1694bc3d6f24cb2d42fa5 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 6 Sep 2015 21:38:51 +0200
Subject: batman-adv: Fix names in kerneldoc of functions

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index fe578f7..dfeebc8 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -375,7 +375,7 @@ out:
 
 /**
  * batadv_neigh_ifinfo_get - find the ifinfo from an neigh_node
- * @neigh_node: the neigh node to be queried
+ * @neigh: the neigh node to be queried
  * @if_outgoing: the interface for which the ifinfo should be acquired
  *
  * The object is returned with refcounter increased by 1.
@@ -408,7 +408,7 @@ batadv_neigh_ifinfo_get(struct batadv_neigh_node *neigh,
 
 /**
  * batadv_neigh_ifinfo_new - search and possibly create an neigh_ifinfo object
- * @neigh_node: the neigh node to be queried
+ * @neigh: the neigh node to be queried
  * @if_outgoing: the interface for which the ifinfo should be acquired
  *
  * Returns NULL in case of failure or the neigh_ifinfo object for the
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index 6d5b178..72fd5b0 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -494,7 +494,7 @@ struct batadv_tvlv_gateway_data {
  * struct batadv_tvlv_tt_data - tt data propagated through the tt tvlv container
  * @flags: translation table flags (see batadv_tt_data_flags)
  * @ttvn: translation table version number
- * @vlan_num: number of announced VLANs. In the TVLV this struct is followed by
+ * @num_vlan: number of announced VLANs. In the TVLV this struct is followed by
  *  one batadv_tvlv_tt_vlan_data object per announced vlan
  */
 struct batadv_tvlv_tt_data {
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 782fa33..3fbf791 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -211,7 +211,7 @@ static bool batadv_send_skb_prepare_unicast(struct sk_buff *skb,
  *  unicast 4addr header
  * @bat_priv: the bat priv with all the soft interface information
  * @skb: the skb containing the payload to encapsulate
- * @orig_node: the destination node
+ * @orig: the destination node
  * @packet_subtype: the unicast 4addr packet subtype to use
  *
  * Returns false if the payload could not be encapsulated or true otherwise.
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index ac4d08d..c923e3c 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -480,7 +480,7 @@ out:
 /**
  * batadv_softif_vlan_free_ref - decrease the vlan object refcounter and
  *  possibly free it
- * @softif_vlan: the vlan object to release
+ * @vlan: the vlan object to release
  */
 void batadv_softif_vlan_free_ref(struct batadv_softif_vlan *vlan)
 {
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 3ccede5..7d8fc90 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -3366,8 +3366,8 @@ out:
  * batadv_tt_update_orig - update global translation table with new tt
  *  information received via ogms
  * @bat_priv: the bat priv with all the soft interface information
- * @orig: the orig_node of the ogm
- * @tt_vlan: pointer to the first tvlv VLAN entry
+ * @orig_node: the orig_node of the ogm
+ * @tt_buff: pointer to the first tvlv VLAN entry
  * @tt_num_vlan: number of tvlv VLAN entries
  * @tt_change: pointer to the first entry in the TT buffer
  * @tt_num_changes: number of tt changes inside the tt buffer
-- 
cgit v0.10.2


From f34ac9d4ec1fcb02fc70aad8d1ce0c0641a90b12 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 6 Sep 2015 21:38:53 +0200
Subject: batman-adv: Drop invalid kerneldoc for variable batadv_vlan_attrs

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index fe87777..33772d9 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -620,9 +620,7 @@ static struct batadv_attribute *batadv_mesh_attrs[] = {
 
 BATADV_ATTR_VLAN_BOOL(ap_isolation, S_IRUGO | S_IWUSR, NULL);
 
-/**
- * batadv_vlan_attrs - array of vlan specific sysfs attributes
- */
+/* array of vlan specific sysfs attributes */
 static struct batadv_attribute *batadv_vlan_attrs[] = {
 	&batadv_attr_vlan_ap_isolation,
 	NULL,
-- 
cgit v0.10.2


From d1f6825067b9c167b3853c9af67e68d5412c2c63 Mon Sep 17 00:00:00 2001
From: Marek Lindner <mareklindner@neomailbox.ch>
Date: Sat, 15 Aug 2015 18:04:59 +0800
Subject: batman-adv: kernel doc readability updates

Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 8c3dda3..7f7a0f1 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -73,7 +73,7 @@ enum batadv_dhcp_recipient {
 #define BATADV_TT_SYNC_MASK	0x00F0
 
 /**
- * struct batadv_hard_iface_bat_iv - per hard interface B.A.T.M.A.N. IV data
+ * struct batadv_hard_iface_bat_iv - per hard-interface B.A.T.M.A.N. IV data
  * @ogm_buff: buffer holding the OGM packet
  * @ogm_buff_len: length of the OGM packet buffer
  * @ogm_seqno: OGM sequence number - used to identify each OGM
@@ -97,8 +97,8 @@ struct batadv_hard_iface_bat_iv {
  *  batman-adv for this interface
  * @soft_iface: the batman-adv interface which uses this network interface
  * @rcu: struct used for freeing in an RCU-safe manner
- * @bat_iv: BATMAN IV specific per hard interface data
- * @cleanup_work: work queue callback item for hard interface deinit
+ * @bat_iv: per hard-interface B.A.T.M.A.N. IV data
+ * @cleanup_work: work queue callback item for hard-interface deinit
  * @debug_dir: dentry for nc subdir in batman-adv directory in debugfs
  * @neigh_list: list of unique single hop neighbors via this interface
  * @neigh_list_lock: lock protecting neigh_list
@@ -125,7 +125,7 @@ struct batadv_hard_iface {
 /**
  * struct batadv_orig_ifinfo - originator info per outgoing interface
  * @list: list node for orig_node::ifinfo_list
- * @if_outgoing: pointer to outgoing hard interface
+ * @if_outgoing: pointer to outgoing hard-interface
  * @router: router that should be used to reach this originator
  * @last_real_seqno: last and best known sequence number
  * @last_ttl: ttl of last received packet
@@ -202,7 +202,7 @@ struct batadv_orig_node_vlan {
 
 /**
  * struct batadv_orig_bat_iv - B.A.T.M.A.N. IV private orig_node members
- * @bcast_own: set of bitfields (one per hard interface) where each one counts
+ * @bcast_own: set of bitfields (one per hard-interface) where each one counts
  * the number of our OGMs this orig_node rebroadcasted "back" to us  (relative
  * to last_real_seqno). Every bitfield is BATADV_TQ_LOCAL_WINDOW_SIZE bits long.
  * @bcast_own_sum: sum of bcast_own
@@ -346,10 +346,11 @@ struct batadv_gw_node {
 };
 
 /**
- * struct batadv_hardif_neigh_node - unique neighbor per hard interface
+ * struct batadv_hardif_neigh_node - unique neighbor per hard-interface
  * @list: list node for batadv_hard_iface::neigh_list
  * @addr: the MAC address of the neighboring interface
- * @if_incoming: pointer to incoming hard interface
+ * @if_incoming: pointer to incoming hard-interface
+ * @last_seen: when last packet via this neighbor was received
  * @refcount: number of contexts the object is used
  * @rcu: struct used for freeing in a RCU-safe manner
  */
@@ -369,7 +370,7 @@ struct batadv_hardif_neigh_node {
  * @addr: the MAC address of the neighboring interface
  * @ifinfo_list: list for routing metrics per outgoing interface
  * @ifinfo_lock: lock protecting private ifinfo members and list
- * @if_incoming: pointer to incoming hard interface
+ * @if_incoming: pointer to incoming hard-interface
  * @last_seen: when last packet via this neighbor was received
  * @refcount: number of contexts the object is used
  * @rcu: struct used for freeing in an RCU-safe manner
@@ -388,7 +389,7 @@ struct batadv_neigh_node {
 
 /**
  * struct batadv_neigh_ifinfo_bat_iv - neighbor information per outgoing
- *  interface for BATMAN IV
+ *  interface for B.A.T.M.A.N. IV
  * @tq_recv: ring buffer of received TQ values from this neigh node
  * @tq_index: ring buffer index
  * @tq_avg: averaged tq of all tq values in the ring buffer (tq_recv)
@@ -407,7 +408,7 @@ struct batadv_neigh_ifinfo_bat_iv {
 /**
  * struct batadv_neigh_ifinfo - neighbor information per outgoing interface
  * @list: list node for batadv_neigh_node::ifinfo_list
- * @if_outgoing: pointer to outgoing hard interface
+ * @if_outgoing: pointer to outgoing hard-interface
  * @bat_iv: B.A.T.M.A.N. IV private structure
  * @last_ttl: last received ttl from this neigh node
  * @refcount: number of contexts the object is used
@@ -783,8 +784,8 @@ struct batadv_softif_vlan {
  * @forw_bat_list_lock: lock protecting forw_bat_list
  * @forw_bcast_list_lock: lock protecting forw_bcast_list
  * @orig_work: work queue callback item for orig node purging
- * @cleanup_work: work queue callback item for soft interface deinit
- * @primary_if: one of the hard interfaces assigned to this mesh interface
+ * @cleanup_work: work queue callback item for soft-interface deinit
+ * @primary_if: one of the hard-interfaces assigned to this mesh interface
  *  becomes the primary interface
  * @bat_algo_ops: routing algorithm used by this mesh interface
  * @softif_vlan_list: a list of softif_vlan structs, one per VLAN created on top
-- 
cgit v0.10.2


From 62fe710f6e2fdf4c337763bc0774cb3232fa2139 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Tue, 15 Sep 2015 19:00:48 +0200
Subject: batman-adv: Fix kerneldoc parsing of return description

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index df625de..a6d389b 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -88,7 +88,7 @@ static void batadv_ring_buffer_set(u8 lq_recv[], u8 *lq_index, u8 value)
  * in the given ring buffer
  * @lq_recv: pointer to the ring buffer
  *
- * Returns computed average value.
+ * Return: computed average value.
  */
 static u8 batadv_ring_buffer_avg(const u8 lq_recv[])
 {
@@ -132,7 +132,7 @@ static void batadv_iv_ogm_orig_free(struct batadv_orig_node *orig_node)
  * @orig_node: the orig_node that has to be changed
  * @max_if_num: the current amount of interfaces
  *
- * Returns 0 on success, a negative error code otherwise.
+ * Return: 0 on success, a negative error code otherwise.
  */
 static int batadv_iv_ogm_orig_add_if(struct batadv_orig_node *orig_node,
 				     int max_if_num)
@@ -180,7 +180,7 @@ unlock:
  * @max_if_num: the current amount of interfaces
  * @del_if_num: the index of the interface being removed
  *
- * Returns 0 on success, a negative error code otherwise.
+ * Return: 0 on success, a negative error code otherwise.
  */
 static int batadv_iv_ogm_orig_del_if(struct batadv_orig_node *orig_node,
 				     int max_if_num, int del_if_num)
@@ -246,7 +246,7 @@ unlock:
  * @bat_priv: the bat priv with all the soft interface information
  * @addr: mac address of the originator
  *
- * Returns the originator object corresponding to the passed mac address or NULL
+ * Return: the originator object corresponding to the passed mac address or NULL
  * on failure.
  * If the object does not exists it is created an initialised.
  */
@@ -522,7 +522,7 @@ out:
  * @if_outgoing: interface for which the retransmission should be considered
  * @forw_packet: the forwarded packet which should be checked
  *
- * Returns true if new_packet can be aggregated with forw_packet
+ * Return: true if new_packet can be aggregated with forw_packet
  */
 static bool
 batadv_iv_ogm_can_aggregate(const struct batadv_ogm_packet *new_bat_ogm_packet,
@@ -1125,7 +1125,7 @@ out:
  * @if_incoming: interface where the packet was received
  * @if_outgoing: interface for which the retransmission should be considered
  *
- * Returns 1 if the link can be considered bidirectional, 0 otherwise
+ * Return: 1 if the link can be considered bidirectional, 0 otherwise
  */
 static int batadv_iv_ogm_calc_tq(struct batadv_orig_node *orig_node,
 				 struct batadv_orig_node *orig_neigh_node,
@@ -1269,7 +1269,7 @@ out:
  * @if_incoming: interface on which the OGM packet was received
  * @if_outgoing: interface for which the retransmission should be considered
  *
- * Returns duplicate status as enum batadv_dup_status
+ * Return: duplicate status as enum batadv_dup_status
  */
 static enum batadv_dup_status
 batadv_iv_ogm_update_seqnos(const struct ethhdr *ethhdr,
@@ -1929,7 +1929,7 @@ static void batadv_iv_neigh_print(struct batadv_priv *bat_priv,
  * @neigh2: the second neighbor object of the comparison
  * @if_outgoing2: outgoing interface for the second neighbor
  *
- * Returns a value less, equal to or greater than 0 if the metric via neigh1 is
+ * Return: a value less, equal to or greater than 0 if the metric via neigh1 is
  * lower, the same as or higher than the metric via neigh2
  */
 static int batadv_iv_ogm_neigh_cmp(struct batadv_neigh_node *neigh1,
@@ -1970,7 +1970,7 @@ out:
  * @neigh2: the second neighbor object of the comparison
  * @if_outgoing2: outgoing interface for the second neighbor
  *
- * Returns true if the metric via neigh1 is equally good or better than
+ * Return: true if the metric via neigh1 is equally good or better than
  * the metric via neigh2, false otherwise.
  */
 static bool
diff --git a/net/batman-adv/bitarray.c b/net/batman-adv/bitarray.c
index 25cbc36..06bab51 100644
--- a/net/batman-adv/bitarray.c
+++ b/net/batman-adv/bitarray.c
@@ -31,7 +31,7 @@ static void batadv_bitmap_shift_left(unsigned long *seq_bits, s32 n)
 
 /* receive and process one packet within the sequence number window.
  *
- * returns:
+ * Return:
  *  1 if the window was moved (either new or very old)
  *  0 if the window was not moved/shifted.
  */
diff --git a/net/batman-adv/bitarray.h b/net/batman-adv/bitarray.h
index 0226b22..cf2aeb0 100644
--- a/net/batman-adv/bitarray.h
+++ b/net/batman-adv/bitarray.h
@@ -24,7 +24,10 @@
 #include <linux/compiler.h>
 #include <linux/types.h>
 
-/* Returns 1 if the corresponding bit in the given seq_bits indicates true
+/**
+ * batadv_test_bit
+ *
+ * Return: 1 if the corresponding bit in the given seq_bits indicates true
  * and curr_seqno is within range of last_seqno. Otherwise returns 0.
  */
 static inline int batadv_test_bit(const unsigned long *seq_bits,
@@ -48,8 +51,10 @@ static inline void batadv_set_bit(unsigned long *seq_bits, s32 n)
 	set_bit(n, seq_bits); /* turn the position on */
 }
 
-/* receive and process one packet, returns 1 if received seq_num is considered
- * new, 0 if old
+/**
+ * batadv_bit_get_packet - receive and process one packet
+ *
+ * Return: 1 if received seq_num is considered new, 0 if old
  */
 int batadv_bit_get_packet(void *priv, unsigned long *seq_bits, s32 seq_num_diff,
 			  int set_mark);
diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index c24c481..90c7994 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -58,7 +58,11 @@ static void
 batadv_bla_send_announce(struct batadv_priv *bat_priv,
 			 struct batadv_bla_backbone_gw *backbone_gw);
 
-/* return the index of the claim */
+/**
+ * batadv_choose_claim
+ *
+ * Return: the index of the claim
+ */
 static inline u32 batadv_choose_claim(const void *data, u32 size)
 {
 	struct batadv_bla_claim *claim = (struct batadv_bla_claim *)data;
@@ -70,7 +74,11 @@ static inline u32 batadv_choose_claim(const void *data, u32 size)
 	return hash % size;
 }
 
-/* return the index of the backbone gateway */
+/**
+ * batadv_choose_backbone_gw
+ *
+ * Return: the index of the backbone gateway
+ */
 static inline u32 batadv_choose_backbone_gw(const void *data, u32 size)
 {
 	const struct batadv_bla_claim *claim = (struct batadv_bla_claim *)data;
@@ -145,8 +153,9 @@ static void batadv_claim_free_ref(struct batadv_bla_claim *claim)
  * @bat_priv: the bat priv with all the soft interface information
  * @data: search data (may be local/static data)
  *
- * looks for a claim in the hash, and returns it if found
- * or NULL otherwise.
+ * looks for a claim in the hash
+ *
+ * Return: claim if found or NULL otherwise.
  */
 static struct batadv_bla_claim
 *batadv_claim_hash_find(struct batadv_priv *bat_priv,
@@ -186,7 +195,7 @@ static struct batadv_bla_claim
  * @addr: the address of the originator
  * @vid: the VLAN ID
  *
- * Returns claim if found or NULL otherwise.
+ * Return: claim if found or NULL otherwise.
  */
 static struct batadv_bla_backbone_gw *
 batadv_backbone_hash_find(struct batadv_priv *bat_priv, u8 *addr,
@@ -666,7 +675,11 @@ static void batadv_bla_del_claim(struct batadv_priv *bat_priv,
 	batadv_claim_free_ref(claim);
 }
 
-/* check for ANNOUNCE frame, return 1 if handled */
+/**
+ * batadv_handle_announce - check for ANNOUNCE frame
+ *
+ * Return: 1 if handled
+ */
 static int batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr,
 				  u8 *backbone_addr, unsigned short vid)
 {
@@ -716,7 +729,11 @@ static int batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr,
 	return 1;
 }
 
-/* check for REQUEST frame, return 1 if handled */
+/**
+ * batadv_handle_request - check for REQUEST frame
+ *
+ * Return: 1 if handled
+ */
 static int batadv_handle_request(struct batadv_priv *bat_priv,
 				 struct batadv_hard_iface *primary_if,
 				 u8 *backbone_addr, struct ethhdr *ethhdr,
@@ -740,7 +757,11 @@ static int batadv_handle_request(struct batadv_priv *bat_priv,
 	return 1;
 }
 
-/* check for UNCLAIM frame, return 1 if handled */
+/**
+ * batadv_handle_unclaim - check for UNCLAIM frame
+ *
+ * Return: 1 if handled
+ */
 static int batadv_handle_unclaim(struct batadv_priv *bat_priv,
 				 struct batadv_hard_iface *primary_if,
 				 u8 *backbone_addr, u8 *claim_addr,
@@ -769,7 +790,11 @@ static int batadv_handle_unclaim(struct batadv_priv *bat_priv,
 	return 1;
 }
 
-/* check for CLAIM frame, return 1 if handled */
+/**
+ * batadv_handle_claim - check for CLAIM frame
+ *
+ * Return: 1 if handled
+ */
 static int batadv_handle_claim(struct batadv_priv *bat_priv,
 			       struct batadv_hard_iface *primary_if,
 			       u8 *backbone_addr, u8 *claim_addr,
@@ -809,7 +834,7 @@ static int batadv_handle_claim(struct batadv_priv *bat_priv,
  * This function also applies the group ID of the sender
  * if it is in the same mesh.
  *
- * returns:
+ * Return:
  *	2  - if it is a claim packet and on the same group
  *	1  - if is a claim packet from another group
  *	0  - if it is not a claim packet
@@ -880,7 +905,7 @@ static int batadv_check_claim_group(struct batadv_priv *bat_priv,
  *
  * Check if this is a claim frame, and process it accordingly.
  *
- * returns 1 if it was a claim frame, otherwise return 0 to
+ * Return: 1 if it was a claim frame, otherwise return 0 to
  * tell the callee that it can use the frame on its own.
  */
 static int batadv_bla_process_claim(struct batadv_priv *bat_priv,
@@ -1397,7 +1422,7 @@ out:
  *
  * Check if the originator is a gateway for the VLAN identified by vid.
  *
- * Returns true if orig is a backbone for this vid, false otherwise.
+ * Return: true if orig is a backbone for this vid, false otherwise.
  */
 bool batadv_bla_is_backbone_gw_orig(struct batadv_priv *bat_priv, u8 *orig,
 				    unsigned short vid)
@@ -1436,9 +1461,10 @@ bool batadv_bla_is_backbone_gw_orig(struct batadv_priv *bat_priv, u8 *orig,
  * @orig_node: the orig_node of the frame
  * @hdr_size: maximum length of the frame
  *
- * bla_is_backbone_gw inspects the skb for the VLAN ID and returns 1
- * if the orig_node is also a gateway on the soft interface, otherwise it
- * returns 0.
+ * bla_is_backbone_gw inspects the skb for the VLAN ID
+ *
+ * Return: 1 if the orig_node is also a gateway on the soft interface, otherwise
+ * it returns 0.
  */
 int batadv_bla_is_backbone_gw(struct sk_buff *skb,
 			      struct batadv_orig_node *orig_node, int hdr_size)
@@ -1498,8 +1524,9 @@ void batadv_bla_free(struct batadv_priv *bat_priv)
  *  * we have to race for a claim
  *  * if the frame is allowed on the LAN
  *
- * in these cases, the skb is further handled by this function and
- * returns 1, otherwise it returns 0 and the caller shall further
+ * in these cases, the skb is further handled by this function
+ *
+ * Return: 1 if handled, otherwise it returns 0 and the caller shall further
  * process the skb.
  */
 int batadv_bla_rx(struct batadv_priv *bat_priv, struct sk_buff *skb,
@@ -1592,11 +1619,12 @@ out:
  *  * a claim was received which has to be processed
  *  * the frame is allowed on the mesh
  *
- * in these cases, the skb is further handled by this function and
- * returns 1, otherwise it returns 0 and the caller shall further
- * process the skb.
+ * in these cases, the skb is further handled by this function.
  *
  * This call might reallocate skb data.
+ *
+ * Return: 1 if handled, otherwise it returns 0 and the caller shall further
+ * process the skb.
  */
 int batadv_bla_tx(struct batadv_priv *bat_priv, struct sk_buff *skb,
 		  unsigned short vid)
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index a49c705..2f13429 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -76,7 +76,7 @@ static void batadv_dat_entry_free_ref(struct batadv_dat_entry *dat_entry)
  * batadv_dat_to_purge - check whether a dat_entry has to be purged or not
  * @dat_entry: the entry to check
  *
- * Returns true if the entry has to be purged now, false otherwise.
+ * Return: true if the entry has to be purged now, false otherwise.
  */
 static bool batadv_dat_to_purge(struct batadv_dat_entry *dat_entry)
 {
@@ -151,7 +151,7 @@ static void batadv_dat_purge(struct work_struct *work)
  * @node: node in the local table
  * @data2: second object to compare the node to
  *
- * Returns 1 if the two entries are the same, 0 otherwise.
+ * Return: 1 if the two entries are the same, 0 otherwise.
  */
 static int batadv_compare_dat(const struct hlist_node *node, const void *data2)
 {
@@ -166,7 +166,7 @@ static int batadv_compare_dat(const struct hlist_node *node, const void *data2)
  * @skb: ARP packet
  * @hdr_size: size of the possible header before the ARP packet
  *
- * Returns the value of the hw_src field in the ARP packet.
+ * Return: the value of the hw_src field in the ARP packet.
  */
 static u8 *batadv_arp_hw_src(struct sk_buff *skb, int hdr_size)
 {
@@ -183,7 +183,7 @@ static u8 *batadv_arp_hw_src(struct sk_buff *skb, int hdr_size)
  * @skb: ARP packet
  * @hdr_size: size of the possible header before the ARP packet
  *
- * Returns the value of the ip_src field in the ARP packet.
+ * Return: the value of the ip_src field in the ARP packet.
  */
 static __be32 batadv_arp_ip_src(struct sk_buff *skb, int hdr_size)
 {
@@ -195,7 +195,7 @@ static __be32 batadv_arp_ip_src(struct sk_buff *skb, int hdr_size)
  * @skb: ARP packet
  * @hdr_size: size of the possible header before the ARP packet
  *
- * Returns the value of the hw_dst field in the ARP packet.
+ * Return: the value of the hw_dst field in the ARP packet.
  */
 static u8 *batadv_arp_hw_dst(struct sk_buff *skb, int hdr_size)
 {
@@ -207,7 +207,7 @@ static u8 *batadv_arp_hw_dst(struct sk_buff *skb, int hdr_size)
  * @skb: ARP packet
  * @hdr_size: size of the possible header before the ARP packet
  *
- * Returns the value of the ip_dst field in the ARP packet.
+ * Return: the value of the ip_dst field in the ARP packet.
  */
 static __be32 batadv_arp_ip_dst(struct sk_buff *skb, int hdr_size)
 {
@@ -219,7 +219,7 @@ static __be32 batadv_arp_ip_dst(struct sk_buff *skb, int hdr_size)
  * @data: data to hash
  * @size: size of the hash table
  *
- * Returns the selected index in the hash table for the given data.
+ * Return: the selected index in the hash table for the given data.
  */
 static u32 batadv_hash_dat(const void *data, u32 size)
 {
@@ -256,7 +256,7 @@ static u32 batadv_hash_dat(const void *data, u32 size)
  * @ip: search key
  * @vid: VLAN identifier
  *
- * Returns the dat_entry if found, NULL otherwise.
+ * Return: the dat_entry if found, NULL otherwise.
  */
 static struct batadv_dat_entry *
 batadv_dat_entry_hash_find(struct batadv_priv *bat_priv, __be32 ip,
@@ -440,7 +440,7 @@ static void batadv_dbg_arp(struct batadv_priv *bat_priv, struct sk_buff *skb,
  * @candidate: orig_node under evaluation
  * @max_orig_node: last selected candidate
  *
- * Returns true if the node has been elected as next candidate or false
+ * Return: true if the node has been elected as next candidate or false
  * otherwise.
  */
 static bool batadv_is_orig_node_eligible(struct batadv_dat_candidate *res,
@@ -558,7 +558,7 @@ static void batadv_choose_next_candidate(struct batadv_priv *bat_priv,
  * closest values (from the LEFT, with wrap around if needed) then the hash
  * value of the key. ip_dst is the key.
  *
- * Returns the candidate array of size BATADV_DAT_CANDIDATE_NUM.
+ * Return: the candidate array of size BATADV_DAT_CANDIDATE_NUM.
  */
 static struct batadv_dat_candidate *
 batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst)
@@ -602,7 +602,7 @@ batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst)
  * This function copies the skb with pskb_copy() and is sent as unicast packet
  * to each of the selected candidates.
  *
- * Returns true if the packet is sent to at least one candidate, false
+ * Return: true if the packet is sent to at least one candidate, false
  * otherwise.
  */
 static bool batadv_dat_send_data(struct batadv_priv *bat_priv,
@@ -831,7 +831,7 @@ out:
  * @skb: packet to analyse
  * @hdr_size: size of the possible header before the ARP packet in the skb
  *
- * Returns the ARP type if the skb contains a valid ARP packet, 0 otherwise.
+ * Return: the ARP type if the skb contains a valid ARP packet, 0 otherwise.
  */
 static u16 batadv_arp_get_type(struct batadv_priv *bat_priv,
 			       struct sk_buff *skb, int hdr_size)
@@ -904,8 +904,9 @@ out:
  * @skb: the buffer containing the packet to extract the VID from
  * @hdr_size: the size of the batman-adv header encapsulating the packet
  *
- * If the packet embedded in the skb is vlan tagged this function returns the
- * VID with the BATADV_VLAN_HAS_TAG flag. Otherwise BATADV_NO_FLAGS is returned.
+ * Return: If the packet embedded in the skb is vlan tagged this function
+ * returns the VID with the BATADV_VLAN_HAS_TAG flag. Otherwise BATADV_NO_FLAGS
+ * is returned.
  */
 static unsigned short batadv_dat_get_vid(struct sk_buff *skb, int *hdr_size)
 {
@@ -930,7 +931,7 @@ static unsigned short batadv_dat_get_vid(struct sk_buff *skb, int *hdr_size)
  * @bat_priv: the bat priv with all the soft interface information
  * @skb: packet to check
  *
- * Returns true if the message has been sent to the dht candidates, false
+ * Return: true if the message has been sent to the dht candidates, false
  * otherwise. In case of a positive return value the message has to be enqueued
  * to permit the fallback.
  */
@@ -1020,7 +1021,7 @@ out:
  * @skb: packet to check
  * @hdr_size: size of the encapsulation header
  *
- * Returns true if the request has been answered, false otherwise.
+ * Return: true if the request has been answered, false otherwise.
  */
 bool batadv_dat_snoop_incoming_arp_request(struct batadv_priv *bat_priv,
 					   struct sk_buff *skb, int hdr_size)
@@ -1143,7 +1144,7 @@ void batadv_dat_snoop_outgoing_arp_reply(struct batadv_priv *bat_priv,
  * @skb: packet to check
  * @hdr_size: size of the encapsulation header
  *
- * Returns true if the packet was snooped and consumed by DAT. False if the
+ * Return: true if the packet was snooped and consumed by DAT. False if the
  * packet has to be delivered to the interface
  */
 bool batadv_dat_snoop_incoming_arp_reply(struct batadv_priv *bat_priv,
@@ -1200,7 +1201,7 @@ out:
  * @bat_priv: the bat priv with all the soft interface information
  * @forw_packet: the broadcast packet
  *
- * Returns true if the node can drop the packet, false otherwise.
+ * Return: true if the node can drop the packet, false otherwise.
  */
 bool batadv_dat_drop_broadcast_packet(struct batadv_priv *bat_priv,
 				      struct batadv_forw_packet *forw_packet)
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index fedddbe..80eddf4 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -85,7 +85,7 @@ void batadv_frag_purge_orig(struct batadv_orig_node *orig_node,
 /**
  * batadv_frag_size_limit - maximum possible size of packet to be fragmented
  *
- * Returns the maximum size of payload that can be fragmented.
+ * Return: the maximum size of payload that can be fragmented.
  */
 static int batadv_frag_size_limit(void)
 {
@@ -107,7 +107,7 @@ static int batadv_frag_size_limit(void)
  *
  * Caller must hold chain->lock.
  *
- * Returns true if chain is empty and caller can just insert the new fragment
+ * Return: true if chain is empty and caller can just insert the new fragment
  * without searching for the right position.
  */
 static bool batadv_frag_init_chain(struct batadv_frag_table_entry *chain,
@@ -136,7 +136,7 @@ static bool batadv_frag_init_chain(struct batadv_frag_table_entry *chain,
  * Insert a new fragment into the reverse ordered chain in the right table
  * entry. The hash table entry is cleared if "old" fragments exist in it.
  *
- * Returns true if skb is buffered, false on error. If the chain has all the
+ * Return: true if skb is buffered, false on error. If the chain has all the
  * fragments needed to merge the packet, the chain is moved to the passed head
  * to avoid locking the chain in the table.
  */
@@ -246,7 +246,7 @@ err:
  * Expand the first skb in the chain and copy the content of the remaining
  * skb's into the expanded one. After doing so, clear the chain.
  *
- * Returns the merged skb or NULL on error.
+ * Return: the merged skb or NULL on error.
  */
 static struct sk_buff *
 batadv_frag_merge_packets(struct hlist_head *chain)
@@ -306,6 +306,9 @@ free:
  * There are three possible outcomes: 1) Packet is merged: Return true and
  * set *skb to merged packet; 2) Packet is buffered: Return true and set *skb
  * to NULL; 3) Error: Return false and leave skb as is.
+ *
+ * Return: true when packet is merged or buffered, false when skb is not not
+ * used.
  */
 bool batadv_frag_skb_buffer(struct sk_buff **skb,
 			    struct batadv_orig_node *orig_node_src)
@@ -343,7 +346,7 @@ out_err:
  * will exceed the MTU towards the next-hop. If so, the fragment is forwarded
  * without merging it.
  *
- * Returns true if the fragment is consumed/forwarded, false otherwise.
+ * Return: true if the fragment is consumed/forwarded, false otherwise.
  */
 bool batadv_frag_skb_fwd(struct sk_buff *skb,
 			 struct batadv_hard_iface *recv_if,
@@ -398,7 +401,7 @@ out:
  * passed mtu and the old one with the rest. The new skb contains data from the
  * tail of the old skb.
  *
- * Returns the new fragment, NULL on error.
+ * Return: the new fragment, NULL on error.
  */
 static struct sk_buff *batadv_frag_create(struct sk_buff *skb,
 					  struct batadv_frag_packet *frag_head,
@@ -432,7 +435,7 @@ err:
  * @orig_node: final destination of the created fragments
  * @neigh_node: next-hop of the created fragments
  *
- * Returns true on success, false otherwise.
+ * Return: true on success, false otherwise.
  */
 bool batadv_frag_send_packet(struct sk_buff *skb,
 			     struct batadv_orig_node *orig_node,
diff --git a/net/batman-adv/fragmentation.h b/net/batman-adv/fragmentation.h
index 8b9877e..c5476fe 100644
--- a/net/batman-adv/fragmentation.h
+++ b/net/batman-adv/fragmentation.h
@@ -42,7 +42,7 @@ bool batadv_frag_send_packet(struct sk_buff *skb,
  * batadv_frag_check_entry - check if a list of fragments has timed out
  * @frags_entry: table entry to check
  *
- * Returns true if the frags entry has timed out, false otherwise.
+ * Return: true if the frags entry has timed out, false otherwise.
  */
 static inline bool
 batadv_frag_check_entry(struct batadv_frag_table_entry *frags_entry)
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index e6c8382..8350775 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -456,7 +456,7 @@ static void batadv_gw_node_add(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @orig_node: originator announcing gateway capabilities
  *
- * Returns gateway node if found or NULL otherwise.
+ * Return: gateway node if found or NULL otherwise.
  */
 static struct batadv_gw_node *
 batadv_gw_node_get(struct batadv_priv *bat_priv,
@@ -655,13 +655,13 @@ out:
  * @chaddr: buffer where the client address will be stored. Valid
  *  only if the function returns BATADV_DHCP_TO_CLIENT
  *
- * Returns:
+ * This function may re-allocate the data buffer of the skb passed as argument.
+ *
+ * Return:
  * - BATADV_DHCP_NO if the packet is not a dhcp message or if there was an error
  *   while parsing it
  * - BATADV_DHCP_TO_SERVER if this is a message going to the DHCP server
  * - BATADV_DHCP_TO_CLIENT if this is a message going to a DHCP client
- *
- * This function may re-allocate the data buffer of the skb passed as argument.
  */
 enum batadv_dhcp_recipient
 batadv_gw_dhcp_recipient_get(struct sk_buff *skb, unsigned int *header_len,
@@ -776,11 +776,11 @@ batadv_gw_dhcp_recipient_get(struct sk_buff *skb, unsigned int *header_len,
  * server. Due to topology changes it may be the case that the GW server
  * previously selected is not the best one anymore.
  *
- * Returns true if the packet destination is unicast and it is not the best gw,
- * false otherwise.
- *
  * This call might reallocate skb data.
  * Must be invoked only when the DHCP packet is going TO a DHCP SERVER.
+ *
+ * Return: true if the packet destination is unicast and it is not the best gw,
+ * false otherwise.
  */
 bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
 			    struct sk_buff *skb)
diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c
index b51bfac..b287448 100644
--- a/net/batman-adv/gateway_common.c
+++ b/net/batman-adv/gateway_common.c
@@ -38,7 +38,7 @@
  * @description: text shown when throughput string cannot be parsed
  * @throughput: pointer holding the returned throughput information
  *
- * Returns false on parse error and true otherwise.
+ * Return: false on parse error and true otherwise.
  */
 static bool batadv_parse_throughput(struct net_device *net_dev, char *buff,
 				    const char *description, u32 *throughput)
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 01acccc..49e05d2 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -85,7 +85,7 @@ out:
  * This function recursively checks all the fathers of the device passed as
  * argument looking for a batman-adv soft interface.
  *
- * Returns true if the device is descendant of a batman-adv mesh interface (or
+ * Return: true if the device is descendant of a batman-adv mesh interface (or
  * if it is a batman-adv interface itself), false otherwise
  */
 static bool batadv_is_on_batman_iface(const struct net_device *net_dev)
@@ -136,7 +136,7 @@ static int batadv_is_valid_iface(const struct net_device *net_dev)
  *  interface
  * @net_device: the device to check
  *
- * Returns true if the net device is a 802.11 wireless device, false otherwise.
+ * Return: true if the net device is a 802.11 wireless device, false otherwise.
  */
 bool batadv_is_wifi_netdev(struct net_device *net_device)
 {
@@ -401,7 +401,8 @@ batadv_hardif_deactivate_interface(struct batadv_hard_iface *hard_iface)
  *
  * Invoke ndo_del_slave on master passing slave as argument. In this way slave
  * is free'd and master can correctly change its internal state.
- * Return 0 on success, a negative value representing the error otherwise
+ *
+ * Return: 0 on success, a negative value representing the error otherwise
  */
 static int batadv_master_del_slave(struct batadv_hard_iface *slave,
 				   struct net_device *master)
diff --git a/net/batman-adv/hash.h b/net/batman-adv/hash.h
index 3776262..071a87e 100644
--- a/net/batman-adv/hash.h
+++ b/net/batman-adv/hash.h
@@ -30,14 +30,17 @@
 struct lock_class_key;
 
 /* callback to a compare function.  should compare 2 element datas for their
- * keys, return 0 if same and not 0 if not same
+ * keys
+ *
+ * Return: 0 if same and not 0 if not same
  */
 typedef int (*batadv_hashdata_compare_cb)(const struct hlist_node *,
 					  const void *);
 
-/* the hashfunction, should return an index
- * based on the key in the data of the first
- * argument and the size the second
+/* the hashfunction
+ *
+ * Return: an index based on the key in the data of the first argument and the
+ * size the second
  */
 typedef u32 (*batadv_hashdata_choose_cb)(const void *, u32);
 typedef void (*batadv_hashdata_free_cb)(struct hlist_node *, void *);
@@ -96,7 +99,7 @@ static inline void batadv_hash_delete(struct batadv_hashtable *hash,
  *	@data: data passed to the aforementioned callbacks as argument
  *	@data_node: to be added element
  *
- *	Returns 0 on success, 1 if the element already is in the hash
+ *	Return: 0 on success, 1 if the element already is in the hash
  *	and -1 on error.
  */
 static inline int batadv_hash_add(struct batadv_hashtable *hash,
@@ -139,10 +142,11 @@ out:
 	return ret;
 }
 
-/* removes data from hash, if found. returns pointer do data on success, so you
- * can remove the used structure yourself, or NULL on error .  data could be the
- * structure you use with just the key filled, we just need the key for
- * comparing.
+/* removes data from hash, if found. data could be the structure you use with
+ * just the key filled, we just need the key for comparing.
+ *
+ * Return: returns pointer do data on success, so you can remove the used
+ * structure yourself, or NULL on error
  */
 static inline void *batadv_hash_remove(struct batadv_hashtable *hash,
 				       batadv_hashdata_compare_cb compare,
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 4b5d61f..70907f6f 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -233,7 +233,7 @@ void batadv_mesh_free(struct net_device *soft_iface)
  * @bat_priv: the bat priv with all the soft interface information
  * @addr: the address to check
  *
- * Returns 'true' if the mac address was found, false otherwise.
+ * Return: 'true' if the mac address was found, false otherwise.
  */
 bool batadv_is_my_mac(struct batadv_priv *bat_priv, const u8 *addr)
 {
@@ -262,7 +262,7 @@ bool batadv_is_my_mac(struct batadv_priv *bat_priv, const u8 *addr)
  *  function that requires the primary interface
  * @seq: debugfs table seq_file struct
  *
- * Returns primary interface if found or NULL otherwise.
+ * Return: primary interface if found or NULL otherwise.
  */
 struct batadv_hard_iface *
 batadv_seq_print_text_primary_if_get(struct seq_file *seq)
@@ -297,7 +297,7 @@ out:
  * batadv_max_header_len - calculate maximum encapsulation overhead for a
  *  payload packet
  *
- * Return the maximum encapsulation overhead in bytes.
+ * Return: the maximum encapsulation overhead in bytes.
  */
 int batadv_max_header_len(void)
 {
@@ -640,7 +640,7 @@ batadv_tvlv_handler_free_ref(struct batadv_tvlv_handler *tvlv_handler)
  * @type: tvlv handler type to look for
  * @version: tvlv handler version to look for
  *
- * Returns tvlv handler if found or NULL otherwise.
+ * Return: tvlv handler if found or NULL otherwise.
  */
 static struct batadv_tvlv_handler
 *batadv_tvlv_handler_get(struct batadv_priv *bat_priv, u8 type, u8 version)
@@ -688,7 +688,7 @@ static void batadv_tvlv_container_free_ref(struct batadv_tvlv_container *tvlv)
  * Has to be called with the appropriate locks being acquired
  * (tvlv.container_list_lock).
  *
- * Returns tvlv container if found or NULL otherwise.
+ * Return: tvlv container if found or NULL otherwise.
  */
 static struct batadv_tvlv_container
 *batadv_tvlv_container_get(struct batadv_priv *bat_priv, u8 type, u8 version)
@@ -720,7 +720,7 @@ static struct batadv_tvlv_container
  * Has to be called with the appropriate locks being acquired
  * (tvlv.container_list_lock).
  *
- * Returns size of all currently registered tvlv containers in bytes.
+ * Return: size of all currently registered tvlv containers in bytes.
  */
 static u16 batadv_tvlv_container_list_size(struct batadv_priv *bat_priv)
 {
@@ -826,7 +826,7 @@ void batadv_tvlv_container_register(struct batadv_priv *bat_priv,
  * @additional_packet_len: requested additional packet size on top of minimum
  *  size
  *
- * Returns true of the packet buffer could be changed to the requested size,
+ * Return: true of the packet buffer could be changed to the requested size,
  * false otherwise.
  */
 static bool batadv_tvlv_realloc_packet_buff(unsigned char **packet_buff,
@@ -862,7 +862,7 @@ static bool batadv_tvlv_realloc_packet_buff(unsigned char **packet_buff,
  * The ogm packet might be enlarged or shrunk depending on the current size
  * and the size of the to-be-appended tvlv containers.
  *
- * Returns size of all appended tvlv containers in bytes.
+ * Return: size of all appended tvlv containers in bytes.
  */
 u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv,
 				     unsigned char **packet_buff,
@@ -915,7 +915,7 @@ end:
  * @tvlv_value: tvlv content
  * @tvlv_value_len: tvlv content length
  *
- * Returns success if handler was not found or the return value of the handler
+ * Return: success if handler was not found or the return value of the handler
  * callback.
  */
 static int batadv_tvlv_call_handler(struct batadv_priv *bat_priv,
@@ -968,7 +968,7 @@ static int batadv_tvlv_call_handler(struct batadv_priv *bat_priv,
  * @tvlv_value: tvlv content
  * @tvlv_value_len: tvlv content length
  *
- * Returns success when processing an OGM or the return value of all called
+ * Return: success when processing an OGM or the return value of all called
  * handler callbacks.
  */
 int batadv_tvlv_containers_process(struct batadv_priv *bat_priv,
@@ -1190,8 +1190,8 @@ out:
  * @skb: the buffer containing the packet
  * @header_len: length of the batman header preceding the ethernet header
  *
- * If the packet embedded in the skb is vlan tagged this function returns the
- * VID with the BATADV_VLAN_HAS_TAG flag. Otherwise BATADV_NO_FLAGS is returned.
+ * Return: VID with the BATADV_VLAN_HAS_TAG flag when the packet embedded in the
+ * skb is vlan tagged. Otherwise BATADV_NO_FLAGS.
  */
 unsigned short batadv_get_vid(struct sk_buff *skb, size_t header_len)
 {
@@ -1218,7 +1218,7 @@ unsigned short batadv_get_vid(struct sk_buff *skb, size_t header_len)
  * @vid: the VLAN identifier for which the AP isolation attributed as to be
  *  looked up
  *
- * Returns true if AP isolation is on for the VLAN idenfied by vid, false
+ * Return: true if AP isolation is on for the VLAN idenfied by vid, false
  * otherwise
  */
 bool batadv_vlan_ap_isola_get(struct batadv_priv *bat_priv, unsigned short vid)
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 9dbd910..8a48d4a 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -273,9 +273,12 @@ static inline void _batadv_dbg(int type __always_unused,
 		pr_err("%s: " fmt, _netdev->name, ## arg);		\
 	} while (0)
 
-/* returns 1 if they are the same ethernet addr
+/**
+ * batadv_compare_eth - Compare two not u16 aligned Ethernet addresses
  *
  * note: can't use ether_addr_equal() as it requires aligned memory
+ *
+ * Return: 1 if they are the same ethernet addr
  */
 static inline bool batadv_compare_eth(const void *data1, const void *data2)
 {
@@ -287,7 +290,7 @@ static inline bool batadv_compare_eth(const void *data1, const void *data2)
  * @timestamp:		base value to compare with (in jiffies)
  * @timeout:		added to base value before comparing (in milliseconds)
  *
- * Returns true if current time is after timestamp + timeout
+ * Return: true if current time is after timestamp + timeout
  */
 static inline bool batadv_has_timed_out(unsigned long timestamp,
 					unsigned int timeout)
@@ -326,7 +329,11 @@ static inline void batadv_add_counter(struct batadv_priv *bat_priv, size_t idx,
 
 #define batadv_inc_counter(b, i) batadv_add_counter(b, i, 1)
 
-/* Sum and return the cpu-local counters for index 'idx' */
+/**
+ * batadv_sum_counter - Sum the cpu-local counters for index 'idx'
+ *
+ * Return: sum of all cpu-local counters
+ */
 static inline u64 batadv_sum_counter(struct batadv_priv *bat_priv,  size_t idx)
 {
 	u64 *counters, sum = 0;
diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c
index 75fa501..d984eee 100644
--- a/net/batman-adv/multicast.c
+++ b/net/batman-adv/multicast.c
@@ -55,7 +55,7 @@
  * Collect multicast addresses of the local multicast listeners
  * on the given soft interface, dev, in the given mcast_list.
  *
- * Returns -ENOMEM on memory allocation error or the number of
+ * Return: -ENOMEM on memory allocation error or the number of
  * items added to the mcast_list otherwise.
  */
 static int batadv_mcast_mla_softif_get(struct net_device *dev,
@@ -87,7 +87,7 @@ static int batadv_mcast_mla_softif_get(struct net_device *dev,
  * @mcast_addr: the multicast address to check
  * @mcast_list: the list with multicast addresses to search in
  *
- * Returns true if the given address is already in the given list.
+ * Return: true if the given address is already in the given list.
  * Otherwise returns false.
  */
 static bool batadv_mcast_mla_is_duplicate(u8 *mcast_addr,
@@ -195,8 +195,9 @@ static void batadv_mcast_mla_tt_add(struct batadv_priv *bat_priv,
  * batadv_mcast_has_bridge - check whether the soft-iface is bridged
  * @bat_priv: the bat priv with all the soft interface information
  *
- * Checks whether there is a bridge on top of our soft interface. Returns
- * true if so, false otherwise.
+ * Checks whether there is a bridge on top of our soft interface.
+ *
+ * Return: true if there is a bridge, false otherwise.
  */
 static bool batadv_mcast_has_bridge(struct batadv_priv *bat_priv)
 {
@@ -218,7 +219,7 @@ static bool batadv_mcast_has_bridge(struct batadv_priv *bat_priv)
  * Updates the own multicast tvlv with our current multicast related settings,
  * capabilities and inabilities.
  *
- * Returns true if the tvlv container is registered afterwards. Otherwise
+ * Return: true if the tvlv container is registered afterwards. Otherwise
  * returns false.
  */
 static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv)
@@ -289,8 +290,8 @@ out:
  * Checks whether the given IPv4 packet has the potential to be forwarded with a
  * mode more optimal than classic flooding.
  *
- * If so then returns 0. Otherwise -EINVAL is returned or -ENOMEM in case of
- * memory allocation failure.
+ * Return: If so then 0. Otherwise -EINVAL or -ENOMEM in case of memory
+ * allocation failure.
  */
 static int batadv_mcast_forw_mode_check_ipv4(struct batadv_priv *bat_priv,
 					     struct sk_buff *skb,
@@ -327,8 +328,7 @@ static int batadv_mcast_forw_mode_check_ipv4(struct batadv_priv *bat_priv,
  * Checks whether the given IPv6 packet has the potential to be forwarded with a
  * mode more optimal than classic flooding.
  *
- * If so then returns 0. Otherwise -EINVAL is returned or -ENOMEM if we are out
- * of memory.
+ * Return: If so then 0. Otherwise -EINVAL is or -ENOMEM if we are out of memory
  */
 static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
 					     struct sk_buff *skb,
@@ -366,8 +366,7 @@ static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv,
  * Checks whether the given multicast ethernet frame has the potential to be
  * forwarded with a mode more optimal than classic flooding.
  *
- * If so then returns 0. Otherwise -EINVAL is returned or -ENOMEM if we are out
- * of memory.
+ * Return: If so then 0. Otherwise -EINVAL is or -ENOMEM if we are out of memory
  */
 static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv,
 					struct sk_buff *skb,
@@ -398,7 +397,7 @@ static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @ethhdr: ethernet header of a packet
  *
- * Returns the number of nodes which want all IPv4 multicast traffic if the
+ * Return: the number of nodes which want all IPv4 multicast traffic if the
  * given ethhdr is from an IPv4 packet or the number of nodes which want all
  * IPv6 traffic if it matches an IPv6 packet.
  */
@@ -421,7 +420,7 @@ static int batadv_mcast_forw_want_all_ip_count(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @ethhdr: the ether header containing the multicast destination
  *
- * Returns an orig_node matching the multicast address provided by ethhdr
+ * Return: an orig_node matching the multicast address provided by ethhdr
  * via a translation table lookup. This increases the returned nodes refcount.
  */
 static struct batadv_orig_node *
@@ -436,7 +435,7 @@ batadv_mcast_forw_tt_node_get(struct batadv_priv *bat_priv,
  * batadv_mcast_want_forw_ipv4_node_get - get a node with an ipv4 flag
  * @bat_priv: the bat priv with all the soft interface information
  *
- * Returns an orig_node which has the BATADV_MCAST_WANT_ALL_IPV4 flag set and
+ * Return: an orig_node which has the BATADV_MCAST_WANT_ALL_IPV4 flag set and
  * increases its refcount.
  */
 static struct batadv_orig_node *
@@ -463,7 +462,7 @@ batadv_mcast_forw_ipv4_node_get(struct batadv_priv *bat_priv)
  * batadv_mcast_want_forw_ipv6_node_get - get a node with an ipv6 flag
  * @bat_priv: the bat priv with all the soft interface information
  *
- * Returns an orig_node which has the BATADV_MCAST_WANT_ALL_IPV6 flag set
+ * Return: an orig_node which has the BATADV_MCAST_WANT_ALL_IPV6 flag set
  * and increases its refcount.
  */
 static struct batadv_orig_node *
@@ -491,7 +490,7 @@ batadv_mcast_forw_ipv6_node_get(struct batadv_priv *bat_priv)
  * @bat_priv: the bat priv with all the soft interface information
  * @ethhdr: an ethernet header to determine the protocol family from
  *
- * Returns an orig_node which has the BATADV_MCAST_WANT_ALL_IPV4 or
+ * Return: an orig_node which has the BATADV_MCAST_WANT_ALL_IPV4 or
  * BATADV_MCAST_WANT_ALL_IPV6 flag, depending on the provided ethhdr, set and
  * increases its refcount.
  */
@@ -514,7 +513,7 @@ batadv_mcast_forw_ip_node_get(struct batadv_priv *bat_priv,
  * batadv_mcast_want_forw_unsnoop_node_get - get a node with an unsnoopable flag
  * @bat_priv: the bat priv with all the soft interface information
  *
- * Returns an orig_node which has the BATADV_MCAST_WANT_ALL_UNSNOOPABLES flag
+ * Return: an orig_node which has the BATADV_MCAST_WANT_ALL_UNSNOOPABLES flag
  * set and increases its refcount.
  */
 static struct batadv_orig_node *
@@ -543,7 +542,7 @@ batadv_mcast_forw_unsnoop_node_get(struct batadv_priv *bat_priv)
  * @skb: The multicast packet to check
  * @orig: an originator to be set to forward the skb to
  *
- * Returns the forwarding mode as enum batadv_forw_mode and in case of
+ * Return: the forwarding mode as enum batadv_forw_mode and in case of
  * BATADV_FORW_SINGLE set the orig to the single originator the skb
  * should be forwarded to.
  */
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index cc63b44..1a75d28 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -251,7 +251,7 @@ static void batadv_nc_packet_free(struct batadv_nc_packet *nc_packet)
  * @bat_priv: the bat priv with all the soft interface information
  * @nc_node: the nc node to check
  *
- * Returns true if the entry has to be purged now, false otherwise
+ * Return: true if the entry has to be purged now, false otherwise
  */
 static bool batadv_nc_to_purge_nc_node(struct batadv_priv *bat_priv,
 				       struct batadv_nc_node *nc_node)
@@ -267,7 +267,7 @@ static bool batadv_nc_to_purge_nc_node(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @nc_path: the nc path to check
  *
- * Returns true if the entry has to be purged now, false otherwise
+ * Return: true if the entry has to be purged now, false otherwise
  */
 static bool batadv_nc_to_purge_nc_path_coding(struct batadv_priv *bat_priv,
 					      struct batadv_nc_path *nc_path)
@@ -287,7 +287,7 @@ static bool batadv_nc_to_purge_nc_path_coding(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @nc_path: the nc path to check
  *
- * Returns true if the entry has to be purged now, false otherwise
+ * Return: true if the entry has to be purged now, false otherwise
  */
 static bool batadv_nc_to_purge_nc_path_decoding(struct batadv_priv *bat_priv,
 						struct batadv_nc_path *nc_path)
@@ -470,7 +470,7 @@ static void batadv_nc_hash_key_gen(struct batadv_nc_path *key, const char *src,
  * @data: data to hash
  * @size: size of the hash table
  *
- * Returns the selected index in the hash table for the given data.
+ * Return: the selected index in the hash table for the given data.
  */
 static u32 batadv_nc_hash_choose(const void *data, u32 size)
 {
@@ -489,7 +489,7 @@ static u32 batadv_nc_hash_choose(const void *data, u32 size)
  * @node: node in the local table
  * @data2: second object to compare the node to
  *
- * Returns 1 if the two entry are the same, 0 otherwise
+ * Return: 1 if the two entry are the same, 0 otherwise
  */
 static int batadv_nc_hash_compare(const struct hlist_node *node,
 				  const void *data2)
@@ -516,7 +516,7 @@ static int batadv_nc_hash_compare(const struct hlist_node *node,
  * @hash: hash table containing the nc path
  * @data: search key
  *
- * Returns the nc_path if found, NULL otherwise.
+ * Return: the nc_path if found, NULL otherwise.
  */
 static struct batadv_nc_path *
 batadv_nc_hash_find(struct batadv_hashtable *hash,
@@ -571,7 +571,7 @@ static void batadv_nc_send_packet(struct batadv_nc_packet *nc_packet)
  * timeout. If so, the packet is no longer kept and the entry deleted from the
  * queue. Has to be called with the appropriate locks.
  *
- * Returns false as soon as the entry in the fifo queue has not been timed out
+ * Return: false as soon as the entry in the fifo queue has not been timed out
  * yet and true otherwise.
  */
 static bool batadv_nc_sniffed_purge(struct batadv_priv *bat_priv,
@@ -610,7 +610,7 @@ out:
  * packet is no longer delayed, immediately sent and the entry deleted from the
  * queue. Has to be called with the appropriate locks.
  *
- * Returns false as soon as the entry in the fifo queue has not been timed out
+ * Return: false as soon as the entry in the fifo queue has not been timed out
  * yet and true otherwise.
  */
 static bool batadv_nc_fwd_flush(struct batadv_priv *bat_priv,
@@ -731,7 +731,7 @@ static void batadv_nc_worker(struct work_struct *work)
  * @orig_node: neighboring orig node which may be used as nc candidate
  * @ogm_packet: incoming ogm packet also used for the checks
  *
- * Returns true if:
+ * Return: true if:
  *  1) The OGM must have the most recent sequence number.
  *  2) The TTL must be decremented by one and only one.
  *  3) The OGM must be received from the first hop from orig_node.
@@ -772,7 +772,7 @@ static bool batadv_can_nc_with_orig(struct batadv_priv *bat_priv,
  *  (can be equal to orig_node)
  * @in_coding: traverse incoming or outgoing network coding list
  *
- * Returns the nc_node if found, NULL otherwise.
+ * Return: the nc_node if found, NULL otherwise.
  */
 static struct batadv_nc_node
 *batadv_nc_find_nc_node(struct batadv_orig_node *orig_node,
@@ -814,7 +814,7 @@ static struct batadv_nc_node
  *  (can be equal to orig_node)
  * @in_coding: traverse incoming or outgoing network coding list
  *
- * Returns the nc_node if found or created, NULL in case of an error.
+ * Return: the nc_node if found or created, NULL in case of an error.
  */
 static struct batadv_nc_node
 *batadv_nc_get_nc_node(struct batadv_priv *bat_priv,
@@ -932,7 +932,7 @@ out:
  * @src: ethernet source address - first half of the nc path search key
  * @dst: ethernet destination address - second half of the nc path search key
  *
- * Returns pointer to nc_path if the path was found or created, returns NULL
+ * Return: pointer to nc_path if the path was found or created, returns NULL
  * on error.
  */
 static struct batadv_nc_path *batadv_nc_get_path(struct batadv_priv *bat_priv,
@@ -1029,7 +1029,7 @@ static void batadv_nc_memxor(char *dst, const char *src, unsigned int len)
  * @nc_packet: structure containing the packet to the skb can be coded with
  * @neigh_node: next hop to forward packet to
  *
- * Returns true if both packets are consumed, false otherwise.
+ * Return: true if both packets are consumed, false otherwise.
  */
 static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 				   struct sk_buff *skb,
@@ -1228,7 +1228,7 @@ out:
  * Since the source encoded the packet we can be certain it has all necessary
  * decode information.
  *
- * Returns true if coding of a decoded packet is allowed.
+ * Return: true if coding of a decoded packet is allowed.
  */
 static bool batadv_nc_skb_coding_possible(struct sk_buff *skb, u8 *dst, u8 *src)
 {
@@ -1246,7 +1246,7 @@ static bool batadv_nc_skb_coding_possible(struct sk_buff *skb, u8 *dst, u8 *src)
  * @skb: data skb to forward
  * @eth_dst: next hop mac address of skb
  *
- * Returns true if coding of a decoded skb is allowed.
+ * Return: true if coding of a decoded skb is allowed.
  */
 static struct batadv_nc_packet *
 batadv_nc_path_search(struct batadv_priv *bat_priv,
@@ -1314,7 +1314,7 @@ batadv_nc_path_search(struct batadv_priv *bat_priv,
  * @eth_src: source mac address of skb
  * @in_nc_node: pointer to skb next hop's neighbor nc node
  *
- * Returns an nc packet if a suitable coding packet was found, NULL otherwise.
+ * Return: an nc packet if a suitable coding packet was found, NULL otherwise.
  */
 static struct batadv_nc_packet *
 batadv_nc_skb_src_search(struct batadv_priv *bat_priv,
@@ -1397,7 +1397,7 @@ static void batadv_nc_skb_store_before_coding(struct batadv_priv *bat_priv,
  * next hop that potentially sent a packet which our next hop also received
  * (overheard) and has stored for later decoding.
  *
- * Returns true if the skb was consumed (encoded packet sent) or false otherwise
+ * Return: true if the skb was consumed (encoded packet sent) or false otherwise
  */
 static bool batadv_nc_skb_dst_search(struct sk_buff *skb,
 				     struct batadv_neigh_node *neigh_node,
@@ -1451,7 +1451,7 @@ static bool batadv_nc_skb_dst_search(struct sk_buff *skb,
  * @neigh_node: next hop to forward packet to
  * @packet_id: checksum to identify packet
  *
- * Returns true if the packet was buffered or false in case of an error.
+ * Return: true if the packet was buffered or false in case of an error.
  */
 static bool batadv_nc_skb_add_to_path(struct sk_buff *skb,
 				      struct batadv_nc_path *nc_path,
@@ -1485,7 +1485,7 @@ static bool batadv_nc_skb_add_to_path(struct sk_buff *skb,
  * @skb: data skb to forward
  * @neigh_node: next hop to forward packet to
  *
- * Returns true if the skb was consumed (encoded packet sent) or false otherwise
+ * Return: true if the skb was consumed (encoded packet sent) or false otherwise
  */
 bool batadv_nc_skb_forward(struct sk_buff *skb,
 			   struct batadv_neigh_node *neigh_node)
@@ -1624,7 +1624,7 @@ void batadv_nc_skb_store_sniffed_unicast(struct batadv_priv *bat_priv,
  * @skb: unicast skb to decode
  * @nc_packet: decode data needed to decode the skb
  *
- * Returns pointer to decoded unicast packet if the packet was decoded or NULL
+ * Return: pointer to decoded unicast packet if the packet was decoded or NULL
  * in case of an error.
  */
 static struct batadv_unicast_packet *
@@ -1718,7 +1718,7 @@ batadv_nc_skb_decode_packet(struct batadv_priv *bat_priv, struct sk_buff *skb,
  * @ethhdr: pointer to the ethernet header inside the coded packet
  * @coded: coded packet we try to find decode data for
  *
- * Returns pointer to nc packet if the needed data was found or NULL otherwise.
+ * Return: pointer to nc packet if the needed data was found or NULL otherwise.
  */
 static struct batadv_nc_packet *
 batadv_nc_find_decoding_packet(struct batadv_priv *bat_priv,
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index dfeebc8..ed1d6d7 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -47,7 +47,11 @@ static struct lock_class_key batadv_orig_hash_lock_class_key;
 
 static void batadv_purge_orig(struct work_struct *work);
 
-/* returns 1 if they are the same originator */
+/**
+ * batadv_compare_orig
+ *
+ * Return: 1 if they are the same originator
+ */
 int batadv_compare_orig(const struct hlist_node *node, const void *data2)
 {
 	const void *data1 = container_of(node, struct batadv_orig_node,
@@ -61,7 +65,7 @@ int batadv_compare_orig(const struct hlist_node *node, const void *data2)
  * @orig_node: the originator serving the VLAN
  * @vid: the VLAN identifier
  *
- * Returns the vlan object identified by vid and belonging to orig_node or NULL
+ * Return: the vlan object identified by vid and belonging to orig_node or NULL
  * if it does not exist.
  */
 struct batadv_orig_node_vlan *
@@ -93,7 +97,7 @@ batadv_orig_node_vlan_get(struct batadv_orig_node *orig_node,
  * @orig_node: the originator serving the VLAN
  * @vid: the VLAN identifier
  *
- * Returns NULL in case of failure or the vlan object identified by vid and
+ * Return: NULL in case of failure or the vlan object identified by vid and
  * belonging to orig_node otherwise. The object is created and added to the list
  * if it does not exist.
  *
@@ -266,7 +270,7 @@ void batadv_neigh_node_free_ref(struct batadv_neigh_node *neigh_node)
  * @if_outgoing: the interface where the payload packet has been received or
  *  the OGM should be sent to
  *
- * Returns the neighbor which should be router for this orig_node/iface.
+ * Return: the neighbor which should be router for this orig_node/iface.
  *
  * The object is returned with refcounter increased by 1.
  */
@@ -298,7 +302,7 @@ batadv_orig_router_get(struct batadv_orig_node *orig_node,
  * @orig_node: the orig node to be queried
  * @if_outgoing: the interface for which the ifinfo should be acquired
  *
- * Returns the requested orig_ifinfo or NULL if not found.
+ * Return: the requested orig_ifinfo or NULL if not found.
  *
  * The object is returned with refcounter increased by 1.
  */
@@ -330,7 +334,7 @@ batadv_orig_ifinfo_get(struct batadv_orig_node *orig_node,
  * @orig_node: the orig node to be queried
  * @if_outgoing: the interface for which the ifinfo should be acquired
  *
- * Returns NULL in case of failure or the orig_ifinfo object for the if_outgoing
+ * Return: NULL in case of failure or the orig_ifinfo object for the if_outgoing
  * interface otherwise. The object is created and added to the list
  * if it does not exist.
  *
@@ -380,7 +384,7 @@ out:
  *
  * The object is returned with refcounter increased by 1.
  *
- * Returns the requested neigh_ifinfo or NULL if not found
+ * Return: the requested neigh_ifinfo or NULL if not found
  */
 struct batadv_neigh_ifinfo *
 batadv_neigh_ifinfo_get(struct batadv_neigh_node *neigh,
@@ -411,7 +415,7 @@ batadv_neigh_ifinfo_get(struct batadv_neigh_node *neigh,
  * @neigh: the neigh node to be queried
  * @if_outgoing: the interface for which the ifinfo should be acquired
  *
- * Returns NULL in case of failure or the neigh_ifinfo object for the
+ * Return: NULL in case of failure or the neigh_ifinfo object for the
  * if_outgoing interface otherwise. The object is created and added to the list
  * if it does not exist.
  *
@@ -459,7 +463,8 @@ out:
  *
  * Looks for and possibly returns a neighbour belonging to this originator list
  * which is connected through the provided hard interface.
- * Returns NULL if the neighbour is not found.
+ *
+ * Return: neighbor when found. Othwerwise NULL
  */
 static struct batadv_neigh_node *
 batadv_neigh_node_get(const struct batadv_orig_node *orig_node,
@@ -492,7 +497,7 @@ batadv_neigh_node_get(const struct batadv_orig_node *orig_node,
  * @hard_iface: the interface this neighbour is connected to
  * @neigh_addr: the interface address of the neighbour to retrieve
  *
- * Returns the hardif neighbour node if found or created or NULL otherwise.
+ * Return: the hardif neighbour node if found or created or NULL otherwise.
  */
 static struct batadv_hardif_neigh_node *
 batadv_hardif_neigh_create(struct batadv_hard_iface *hard_iface,
@@ -540,7 +545,7 @@ out:
  * @hard_iface: the interface this neighbour is connected to
  * @neigh_addr: the interface address of the neighbour to retrieve
  *
- * Returns the hardif neighbour node if found or created or NULL otherwise.
+ * Return: the hardif neighbour node if found or created or NULL otherwise.
  */
 static struct batadv_hardif_neigh_node *
 batadv_hardif_neigh_get_or_create(struct batadv_hard_iface *hard_iface,
@@ -562,7 +567,8 @@ batadv_hardif_neigh_get_or_create(struct batadv_hard_iface *hard_iface,
  * @neigh_addr: the address of the neighbour
  *
  * Looks for and possibly returns a neighbour belonging to this hard interface.
- * Returns NULL if the neighbour is not found.
+ *
+ * Return: neighbor when found. Othwerwise NULL
  */
 struct batadv_hardif_neigh_node *
 batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface,
@@ -594,7 +600,8 @@ batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface,
  * @neigh_addr: the mac address of the neighbour interface
  *
  * Allocates a new neigh_node object and initialises all the generic fields.
- * Returns the new object or NULL on failure.
+ *
+ * Return: neighbor when found. Othwerwise NULL
  */
 struct batadv_neigh_node *
 batadv_neigh_node_new(struct batadv_orig_node *orig_node,
@@ -656,7 +663,7 @@ out:
  * @seq: neighbour table seq_file struct
  * @offset: not used
  *
- * Always returns 0.
+ * Return: always 0
  */
 int batadv_hardif_neigh_seq_print_text(struct seq_file *seq, void *offset)
 {
@@ -820,7 +827,8 @@ void batadv_originator_free(struct batadv_priv *bat_priv)
  *
  * Creates a new originator object and initialise all the generic fields.
  * The new object is not added to the originator list.
- * Returns the newly created object or NULL on failure.
+ *
+ * Return: the newly created object or NULL on failure.
  */
 struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv,
 					      const u8 *addr)
@@ -937,7 +945,7 @@ batadv_purge_neigh_ifinfo(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @orig_node: orig node which is to be checked
  *
- * Returns true if any ifinfo entry was purged, false otherwise.
+ * Return: true if any ifinfo entry was purged, false otherwise.
  */
 static bool
 batadv_purge_orig_ifinfo(struct batadv_priv *bat_priv,
@@ -989,7 +997,7 @@ batadv_purge_orig_ifinfo(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @orig_node: orig node which is to be checked
  *
- * Returns true if any neighbor was purged, false otherwise
+ * Return: true if any neighbor was purged, false otherwise
  */
 static bool
 batadv_purge_orig_neighbors(struct batadv_priv *bat_priv,
@@ -1048,7 +1056,7 @@ batadv_purge_orig_neighbors(struct batadv_priv *bat_priv,
  * @orig_node: orig node which is to be checked
  * @if_outgoing: the interface for which the metric should be compared
  *
- * Returns the current best neighbor, with refcount increased.
+ * Return: the current best neighbor, with refcount increased.
  */
 static struct batadv_neigh_node *
 batadv_find_best_neighbor(struct batadv_priv *bat_priv,
@@ -1085,7 +1093,7 @@ batadv_find_best_neighbor(struct batadv_priv *bat_priv,
  * This function checks if the orig_node or substructures of it have become
  * obsolete, and purges this information if that's the case.
  *
- * Returns true if the orig_node is to be removed, false otherwise.
+ * Return: true if the orig_node is to be removed, false otherwise.
  */
 static bool batadv_purge_orig_node(struct batadv_priv *bat_priv,
 				   struct batadv_orig_node *orig_node)
@@ -1230,7 +1238,7 @@ int batadv_orig_seq_print_text(struct seq_file *seq, void *offset)
  * @seq: debugfs table seq_file struct
  * @offset: not used
  *
- * Returns 0
+ * Return: 0
  */
 int batadv_orig_hardif_seq_print_text(struct seq_file *seq, void *offset)
 {
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 54d6608..bcaa787 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -140,9 +140,12 @@ out:
 		batadv_neigh_node_free_ref(router);
 }
 
-/* checks whether the host restarted and is in the protection time.
- * returns:
- *  0 if the packet is to be accepted
+/**
+ * batadv_window_protected checks whether the host restarted and is in the
+ *  protection time.
+ *
+ * Return:
+ *  0 if the packet is to be accepted.
  *  1 if the packet is to be ignored.
  */
 int batadv_window_protected(struct batadv_priv *bat_priv, s32 seq_num_diff,
@@ -198,7 +201,7 @@ bool batadv_check_management_packet(struct sk_buff *skb,
  * @bat_priv: the bat priv with all the soft interface information
  * @skb: icmp packet to process
  *
- * Returns NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
+ * Return: NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
  * otherwise.
  */
 static int batadv_recv_my_icmp_packet(struct batadv_priv *bat_priv,
@@ -398,10 +401,11 @@ out:
  * @skb: packet to check
  * @hdr_size: size of header to pull
  *
- * Check for short header and bad addresses in given packet. Returns negative
- * value when check fails and 0 otherwise. The negative value depends on the
- * reason: -ENODATA for bad header, -EBADR for broadcast destination or source,
- * and -EREMOTE for non-local (other host) destination.
+ * Check for short header and bad addresses in given packet.
+ *
+ * Return: negative value when check fails and 0 otherwise. The negative value
+ * depends on the reason: -ENODATA for bad header, -EBADR for broadcast
+ * destination or source, and -EREMOTE for non-local (other host) destination.
  */
 static int batadv_check_unicast_packet(struct batadv_priv *bat_priv,
 				       struct sk_buff *skb, int hdr_size)
@@ -435,7 +439,7 @@ static int batadv_check_unicast_packet(struct batadv_priv *bat_priv,
  * @orig_node: the destination node
  * @recv_if: pointer to interface this packet was received on
  *
- * Returns the router which should be used for this orig_node on
+ * Return: the router which should be used for this orig_node on
  * this interface, or NULL if not available.
  */
 struct batadv_neigh_node *
@@ -648,7 +652,7 @@ out:
  * the new corresponding information (originator address where the destination
  * client currently is and its known TTVN)
  *
- * Returns true if the packet header has been updated, false otherwise
+ * Return: true if the packet header has been updated, false otherwise
  */
 static bool
 batadv_reroute_unicast_packet(struct batadv_priv *bat_priv,
@@ -805,7 +809,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv,
  * @skb: unicast tvlv packet to process
  * @recv_if: pointer to interface this packet was received on
  *
- * Returns NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
+ * Return: NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
  * otherwise.
  */
 int batadv_recv_unhandled_unicast_packet(struct sk_buff *skb,
@@ -905,7 +909,7 @@ rx_success:
  * @skb: unicast tvlv packet to process
  * @recv_if: pointer to interface this packet was received on
  *
- * Returns NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
+ * Return: NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
  * otherwise.
  */
 int batadv_recv_unicast_tvlv(struct sk_buff *skb,
@@ -959,7 +963,7 @@ int batadv_recv_unicast_tvlv(struct sk_buff *skb,
  * the assembled packet will exceed our MTU; 2) Buffer fragment, if we till
  * lack further fragments; 3) Merge fragments, if we have all needed parts.
  *
- * Return NET_RX_DROP if the skb is not consumed, NET_RX_SUCCESS otherwise.
+ * Return: NET_RX_DROP if the skb is not consumed, NET_RX_SUCCESS otherwise.
  */
 int batadv_recv_frag_packet(struct sk_buff *skb,
 			    struct batadv_hard_iface *recv_if)
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 3fbf791..d9b93c5 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -111,7 +111,7 @@ send_skb_err:
  * host, NULL can be passed as recv_if and no interface alternating is
  * attempted.
  *
- * Returns NET_XMIT_SUCCESS on success, NET_XMIT_DROP on failure, or
+ * Return: NET_XMIT_SUCCESS on success, NET_XMIT_DROP on failure, or
  * NET_XMIT_POLICED if the skb is buffered for later transmit.
  */
 int batadv_send_skb_to_orig(struct sk_buff *skb,
@@ -165,7 +165,7 @@ out:
  * @hdr_size: amount of bytes to push at the beginning of the skb
  * @orig_node: the destination node
  *
- * Returns false if the buffer extension was not possible or true otherwise.
+ * Return: false if the buffer extension was not possible or true otherwise.
  */
 static bool
 batadv_send_skb_push_fill_unicast(struct sk_buff *skb, int hdr_size,
@@ -196,7 +196,7 @@ batadv_send_skb_push_fill_unicast(struct sk_buff *skb, int hdr_size,
  * @skb: the skb containing the payload to encapsulate
  * @orig_node: the destination node
  *
- * Returns false if the payload could not be encapsulated or true otherwise.
+ * Return: false if the payload could not be encapsulated or true otherwise.
  */
 static bool batadv_send_skb_prepare_unicast(struct sk_buff *skb,
 					    struct batadv_orig_node *orig_node)
@@ -214,7 +214,7 @@ static bool batadv_send_skb_prepare_unicast(struct sk_buff *skb,
  * @orig: the destination node
  * @packet_subtype: the unicast 4addr packet subtype to use
  *
- * Returns false if the payload could not be encapsulated or true otherwise.
+ * Return: false if the payload could not be encapsulated or true otherwise.
  */
 bool batadv_send_skb_prepare_unicast_4addr(struct batadv_priv *bat_priv,
 					   struct sk_buff *skb,
@@ -265,7 +265,7 @@ out:
  * as packet_type. Then send this frame to the given orig_node and release a
  * reference to this orig_node.
  *
- * Returns NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
+ * Return: NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
  */
 int batadv_send_skb_unicast(struct batadv_priv *bat_priv,
 			    struct sk_buff *skb, int packet_type,
@@ -339,7 +339,7 @@ out:
  * BATADV_UNICAST_4ADDR was supplied as packet_type. Then send this frame
  * to the according destination node.
  *
- * Returns NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
+ * Return: NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
  */
 int batadv_send_skb_via_tt_generic(struct batadv_priv *bat_priv,
 				   struct sk_buff *skb, int packet_type,
@@ -373,7 +373,7 @@ int batadv_send_skb_via_tt_generic(struct batadv_priv *bat_priv,
  * Look up the currently selected gateway. Wrap the given skb into a batman-adv
  * unicast header and send this frame to this gateway node.
  *
- * Returns NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
+ * Return: NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
  */
 int batadv_send_skb_via_gw(struct batadv_priv *bat_priv, struct sk_buff *skb,
 			   unsigned short vid)
@@ -430,14 +430,16 @@ _batadv_add_bcast_packet_to_list(struct batadv_priv *bat_priv,
 			   send_time);
 }
 
-/* add a broadcast packet to the queue and setup timers. broadcast packets
- * are sent multiple times to increase probability for being received.
+/**
+ * batadv_add_bcast_packet_to_list
  *
- * This function returns NETDEV_TX_OK on success and NETDEV_TX_BUSY on
- * errors.
+ * add a broadcast packet to the queue and setup timers. broadcast packets
+ * are sent multiple times to increase probability for being received.
  *
  * The skb is not consumed, so the caller should make sure that the
  * skb is freed.
+ *
+ * Return: NETDEV_TX_OK on success and NETDEV_TX_BUSY on errors.
  */
 int batadv_add_bcast_packet_to_list(struct batadv_priv *bat_priv,
 				    const struct sk_buff *skb,
diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h
index 82059f2..3adc11f 100644
--- a/net/batman-adv/send.h
+++ b/net/batman-adv/send.h
@@ -69,7 +69,7 @@ int batadv_send_skb_via_gw(struct batadv_priv *bat_priv, struct sk_buff *skb,
  * header via the translation table. Wrap the given skb into a batman-adv
  * unicast header. Then send this frame to the according destination node.
  *
- * Returns NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
+ * Return: NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
  */
 static inline int batadv_send_skb_via_tt(struct batadv_priv *bat_priv,
 					 struct sk_buff *skb, u8 *dst_hint,
@@ -92,7 +92,7 @@ static inline int batadv_send_skb_via_tt(struct batadv_priv *bat_priv,
  * unicast-4addr header. Then send this frame to the according destination
  * node.
  *
- * Returns NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
+ * Return: NET_XMIT_DROP in case of error or NET_XMIT_SUCCESS otherwise.
  */
 static inline int batadv_send_skb_via_tt_4addr(struct batadv_priv *bat_priv,
 					       struct sk_buff *skb,
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index c923e3c..5ee794b 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -501,7 +501,7 @@ void batadv_softif_vlan_free_ref(struct batadv_softif_vlan *vlan)
  * @bat_priv: the bat priv with all the soft interface information
  * @vid: the identifier of the vlan object to retrieve
  *
- * Returns the private data of the vlan matching the vid passed as argument or
+ * Return: the private data of the vlan matching the vid passed as argument or
  * NULL otherwise. The refcounter of the returned object is incremented by 1.
  */
 struct batadv_softif_vlan *batadv_softif_vlan_get(struct batadv_priv *bat_priv,
@@ -530,7 +530,7 @@ struct batadv_softif_vlan *batadv_softif_vlan_get(struct batadv_priv *bat_priv,
  * @bat_priv: the bat priv with all the soft interface information
  * @vid: the VLAN identifier
  *
- * Returns 0 on success, a negative error otherwise.
+ * Return: 0 on success, a negative error otherwise.
  */
 int batadv_softif_create_vlan(struct batadv_priv *bat_priv, unsigned short vid)
 {
@@ -599,7 +599,7 @@ static void batadv_softif_destroy_vlan(struct batadv_priv *bat_priv,
  * Set up all the internal structures for handling the new vlan on top of the
  * mesh interface
  *
- * Returns 0 on success or a negative error code in case of failure.
+ * Return: 0 on success or a negative error code in case of failure.
  */
 static int batadv_interface_add_vid(struct net_device *dev, __be16 proto,
 				    unsigned short vid)
@@ -656,7 +656,7 @@ static int batadv_interface_add_vid(struct net_device *dev, __be16 proto,
  * Destroy all the internal structures used to handle the vlan identified by vid
  * on top of the mesh interface
  *
- * Returns 0 on success, -EINVAL if the specified prototype is not ETH_P_8021Q
+ * Return: 0 on success, -EINVAL if the specified prototype is not ETH_P_8021Q
  * or -ENOENT if the specified vlan id wasn't registered.
  */
 static int batadv_interface_kill_vid(struct net_device *dev, __be16 proto,
@@ -745,7 +745,7 @@ static void batadv_softif_destroy_finish(struct work_struct *work)
  * batadv_softif_init_late - late stage initialization of soft interface
  * @dev: registered network device to modify
  *
- * Returns error code on failures
+ * Return: error code on failures
  */
 static int batadv_softif_init_late(struct net_device *dev)
 {
@@ -847,7 +847,7 @@ free_bat_counters:
  * @dev: batadv_soft_interface used as master interface
  * @slave_dev: net_device which should become the slave interface
  *
- * Return 0 if successful or error otherwise.
+ * Return: 0 if successful or error otherwise.
  */
 static int batadv_softif_slave_add(struct net_device *dev,
 				   struct net_device *slave_dev)
@@ -872,7 +872,7 @@ out:
  * @dev: batadv_soft_interface used as master interface
  * @slave_dev: net_device which should be removed from the master interface
  *
- * Return 0 if successful or error otherwise.
+ * Return: 0 if successful or error otherwise.
  */
 static int batadv_softif_slave_del(struct net_device *dev,
 				   struct net_device *slave_dev)
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index 33772d9..7658e6e 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -64,7 +64,7 @@ static struct batadv_priv *batadv_kobj_to_batpriv(struct kobject *obj)
  * batadv_vlan_kobj_to_batpriv - convert a vlan kobj in the associated batpriv
  * @obj: kobject to covert
  *
- * Returns the associated batadv_priv struct.
+ * Return: the associated batadv_priv struct.
  */
 static struct batadv_priv *batadv_vlan_kobj_to_batpriv(struct kobject *obj)
 {
@@ -84,7 +84,7 @@ static struct batadv_priv *batadv_vlan_kobj_to_batpriv(struct kobject *obj)
  * batadv_kobj_to_vlan - convert a kobj in the associated softif_vlan struct
  * @obj: kobject to covert
  *
- * Returns the associated softif_vlan struct if found, NULL otherwise.
+ * Return: the associated softif_vlan struct if found, NULL otherwise.
  */
 static struct batadv_softif_vlan *
 batadv_kobj_to_vlan(struct batadv_priv *bat_priv, struct kobject *obj)
@@ -491,7 +491,7 @@ static ssize_t batadv_store_gw_bwidth(struct kobject *kobj,
  * @attr: the batman-adv attribute the user is interacting with
  * @buff: the buffer that will contain the data to send back to the user
  *
- * Returns the number of bytes written into 'buff' on success or a negative
+ * Return: the number of bytes written into 'buff' on success or a negative
  * error code in case of failure
  */
 static ssize_t batadv_show_isolation_mark(struct kobject *kobj,
@@ -511,7 +511,7 @@ static ssize_t batadv_show_isolation_mark(struct kobject *kobj,
  * @buff: the buffer containing the user data
  * @count: number of bytes in the buffer
  *
- * Returns 'count' on success or a negative error code in case of failure
+ * Return: 'count' on success or a negative error code in case of failure
  */
 static ssize_t batadv_store_isolation_mark(struct kobject *kobj,
 					   struct attribute *attr, char *buff,
@@ -681,7 +681,7 @@ void batadv_sysfs_del_meshif(struct net_device *dev)
  * @dev: netdev of the mesh interface
  * @vlan: private data of the newly added VLAN interface
  *
- * Returns 0 on success and -ENOMEM if any of the structure allocations fails.
+ * Return: 0 on success and -ENOMEM if any of the structure allocations fails.
  */
 int batadv_sysfs_add_vlan(struct net_device *dev,
 			  struct batadv_softif_vlan *vlan)
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 7d8fc90..af1d24c 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -68,7 +68,11 @@ static void batadv_tt_global_del(struct batadv_priv *bat_priv,
 				 unsigned short vid, const char *message,
 				 bool roaming);
 
-/* returns 1 if they are the same mac addr and vid */
+/**
+ * batadv_compare_tt
+ *
+ * Return: 1 if they are the same mac addr and vid
+ */
 static int batadv_compare_tt(const struct hlist_node *node, const void *data2)
 {
 	const void *data1 = container_of(node, struct batadv_tt_common_entry,
@@ -84,7 +88,7 @@ static int batadv_compare_tt(const struct hlist_node *node, const void *data2)
  * @data: pointer to the tt_common_entry object to map
  * @size: the size of the hash table
  *
- * Returns the hash index where the object represented by 'data' should be
+ * Return: the hash index where the object represented by 'data' should be
  * stored at.
  */
 static inline u32 batadv_choose_tt(const void *data, u32 size)
@@ -105,7 +109,7 @@ static inline u32 batadv_choose_tt(const void *data, u32 size)
  * @addr: the mac address of the client to look for
  * @vid: VLAN identifier
  *
- * Returns a pointer to the tt_common struct belonging to the searched client if
+ * Return: a pointer to the tt_common struct belonging to the searched client if
  * found, NULL otherwise.
  */
 static struct batadv_tt_common_entry *
@@ -150,7 +154,7 @@ batadv_tt_hash_find(struct batadv_hashtable *hash, const u8 *addr,
  * @addr: the mac address of the client to look for
  * @vid: VLAN identifier
  *
- * Returns a pointer to the corresponding tt_local_entry struct if the client is
+ * Return: a pointer to the corresponding tt_local_entry struct if the client is
  * found, NULL otherwise.
  */
 static struct batadv_tt_local_entry *
@@ -175,7 +179,7 @@ batadv_tt_local_hash_find(struct batadv_priv *bat_priv, const u8 *addr,
  * @addr: the mac address of the client to look for
  * @vid: VLAN identifier
  *
- * Returns a pointer to the corresponding tt_global_entry struct if the client
+ * Return: a pointer to the corresponding tt_global_entry struct if the client
  * is found, NULL otherwise.
  */
 static struct batadv_tt_global_entry *
@@ -220,7 +224,7 @@ batadv_tt_global_entry_free_ref(struct batadv_tt_global_entry *tt_global_entry)
  * @addr: the mac address of the client to count entries for
  * @vid: VLAN identifier
  *
- * Return the number of originators advertising the given address/data
+ * Return: the number of originators advertising the given address/data
  * (excluding ourself).
  */
 int batadv_tt_global_hash_count(struct batadv_priv *bat_priv,
@@ -433,7 +437,7 @@ unlock:
  * batadv_tt_len - compute length in bytes of given number of tt changes
  * @changes_num: number of tt changes
  *
- * Returns computed length in bytes.
+ * Return: computed length in bytes.
  */
 static int batadv_tt_len(int changes_num)
 {
@@ -444,7 +448,7 @@ static int batadv_tt_len(int changes_num)
  * batadv_tt_entries - compute the number of entries fitting in tt_len bytes
  * @tt_len: available space
  *
- * Returns the number of entries.
+ * Return: the number of entries.
  */
 static u16 batadv_tt_entries(u16 tt_len)
 {
@@ -456,7 +460,7 @@ static u16 batadv_tt_entries(u16 tt_len)
  *  size when transmitted over the air
  * @bat_priv: the bat priv with all the soft interface information
  *
- * Returns local translation table size in bytes.
+ * Return: local translation table size in bytes.
  */
 static int batadv_tt_local_table_transmit_size(struct batadv_priv *bat_priv)
 {
@@ -522,7 +526,7 @@ static void batadv_tt_global_free(struct batadv_priv *bat_priv,
  * @mark: the value contained in the skb->mark field of the received packet (if
  *  any)
  *
- * Returns true if the client was successfully added, false otherwise.
+ * Return: true if the client was successfully added, false otherwise.
  */
 bool batadv_tt_local_add(struct net_device *soft_iface, const u8 *addr,
 			 unsigned short vid, int ifindex, u32 mark)
@@ -722,7 +726,7 @@ out:
  * header made up by one tvlv_tt_data object and a series of tvlv_tt_vlan_data
  * objects, one per active VLAN served by the originator node.
  *
- * Return the size of the allocated buffer or 0 in case of failure.
+ * Return: the size of the allocated buffer or 0 in case of failure.
  */
 static u16
 batadv_tt_prepare_tvlv_global_data(struct batadv_orig_node *orig_node,
@@ -796,7 +800,7 @@ out:
  * header made up by one tvlv_tt_data object and a series of tvlv_tt_vlan_data
  * objects, one per active VLAN.
  *
- * Return the size of the allocated buffer or 0 in case of failure.
+ * Return: the size of the allocated buffer or 0 in case of failure.
  */
 static u16
 batadv_tt_prepare_tvlv_local_data(struct batadv_priv *bat_priv,
@@ -1038,7 +1042,7 @@ batadv_tt_local_set_pending(struct batadv_priv *bat_priv,
  * @message: message to append to the log on deletion
  * @roaming: true if the deletion is due to a roaming event
  *
- * Returns the flags assigned to the local entry before being deleted
+ * Return: the flags assigned to the local entry before being deleted
  */
 u16 batadv_tt_local_remove(struct batadv_priv *bat_priv, const u8 *addr,
 			   unsigned short vid, const char *message,
@@ -1238,10 +1242,13 @@ static void batadv_tt_changes_list_free(struct batadv_priv *bat_priv)
 	spin_unlock_bh(&bat_priv->tt.changes_list_lock);
 }
 
-/* retrieves the orig_tt_list_entry belonging to orig_node from the
+/**
+ * batadv_tt_global_orig_entry_find
+ *
+ * retrieves the orig_tt_list_entry belonging to orig_node from the
  * batadv_tt_global_entry list
  *
- * returns it with an increased refcounter, NULL if not found
+ * Return: it with an increased refcounter, NULL if not found
  */
 static struct batadv_tt_orig_list_entry *
 batadv_tt_global_orig_entry_find(const struct batadv_tt_global_entry *entry,
@@ -1266,8 +1273,12 @@ batadv_tt_global_orig_entry_find(const struct batadv_tt_global_entry *entry,
 	return orig_entry;
 }
 
-/* find out if an orig_node is already in the list of a tt_global_entry.
- * returns true if found, false otherwise
+/**
+ * batadv_tt_global_entry_has_orig
+ *
+ * find out if an orig_node is already in the list of a tt_global_entry.
+ *
+ * Return: true if found, false otherwise
  */
 static bool
 batadv_tt_global_entry_has_orig(const struct batadv_tt_global_entry *entry,
@@ -1339,7 +1350,7 @@ out:
  *
  * The caller must hold orig_node refcount.
  *
- * Return true if the new entry has been added, false otherwise
+ * Return: true if the new entry has been added, false otherwise
  */
 static bool batadv_tt_global_add(struct batadv_priv *bat_priv,
 				 struct batadv_orig_node *orig_node,
@@ -1497,7 +1508,7 @@ out:
  * @tt_global_entry: global translation table entry to be analyzed
  *
  * This functon assumes the caller holds rcu_read_lock().
- * Returns best originator list entry or NULL on errors.
+ * Return: best originator list entry or NULL on errors.
  */
 static struct batadv_tt_orig_list_entry *
 batadv_transtable_best_orig(struct batadv_priv *bat_priv,
@@ -2027,7 +2038,7 @@ _batadv_is_ap_isolated(struct batadv_tt_local_entry *tt_local_entry,
  * @addr: mac address of the destination client
  * @vid: VLAN identifier
  *
- * Returns a pointer to the originator that was selected as destination in the
+ * Return: a pointer to the originator that was selected as destination in the
  * mesh for contacting the client 'addr', NULL otherwise.
  * In case of multiple originators serving the same client, the function returns
  * the best one (best in terms of metric towards the destination node).
@@ -2102,7 +2113,7 @@ out:
  * because the XOR operation can combine them all while trying to reduce the
  * noise as much as possible.
  *
- * Returns the checksum of the global table of a given originator.
+ * Return: the checksum of the global table of a given originator.
  */
 static u32 batadv_tt_global_crc(struct batadv_priv *bat_priv,
 				struct batadv_orig_node *orig_node,
@@ -2179,7 +2190,7 @@ static u32 batadv_tt_global_crc(struct batadv_priv *bat_priv,
  * For details about the computation, please refer to the documentation for
  * batadv_tt_global_crc().
  *
- * Returns the checksum of the local table
+ * Return: the checksum of the local table
  */
 static u32 batadv_tt_local_crc(struct batadv_priv *bat_priv,
 			       unsigned short vid)
@@ -2285,7 +2296,7 @@ static void batadv_tt_req_purge(struct batadv_priv *bat_priv)
  * @bat_priv: the bat priv with all the soft interface information
  * @orig_node: orig node this request is being issued for
  *
- * Returns the pointer to the new tt_req_node struct if no request
+ * Return: the pointer to the new tt_req_node struct if no request
  * has already been issued for this orig_node, NULL otherwise.
  */
 static struct batadv_tt_req_node *
@@ -2320,7 +2331,7 @@ unlock:
  * @entry_ptr: to be checked local tt entry
  * @data_ptr: not used but definition required to satisfy the callback prototype
  *
- * Returns 1 if the entry is a valid, 0 otherwise.
+ * Return: 1 if the entry is a valid, 0 otherwise.
  */
 static int batadv_tt_local_valid(const void *entry_ptr, const void *data_ptr)
 {
@@ -2405,7 +2416,7 @@ static void batadv_tt_tvlv_generate(struct batadv_priv *bat_priv,
  * @tt_vlan: pointer to the first tvlv VLAN entry
  * @num_vlan: number of tvlv VLAN entries
  *
- * Return true if all the received CRCs match the locally stored ones, false
+ * Return: true if all the received CRCs match the locally stored ones, false
  * otherwise
  */
 static bool batadv_tt_global_check_crc(struct batadv_orig_node *orig_node,
@@ -2588,7 +2599,7 @@ out:
  * @req_src: mac address of tt request sender
  * @req_dst: mac address of tt request recipient
  *
- * Returns true if tt request reply was sent, false otherwise.
+ * Return: true if tt request reply was sent, false otherwise.
  */
 static bool batadv_send_other_tt_response(struct batadv_priv *bat_priv,
 					  struct batadv_tvlv_tt_data *tt_data,
@@ -2720,7 +2731,7 @@ out:
  * @tt_data: tt data containing the tt request information
  * @req_src: mac address of tt request sender
  *
- * Returns true if tt request reply was sent, false otherwise.
+ * Return: true if tt request reply was sent, false otherwise.
  */
 static bool batadv_send_my_tt_response(struct batadv_priv *bat_priv,
 				       struct batadv_tvlv_tt_data *tt_data,
@@ -2838,7 +2849,7 @@ out:
  * @req_src: mac address of tt request sender
  * @req_dst: mac address of tt request recipient
  *
- * Returns true if tt request reply was sent, false otherwise.
+ * Return: true if tt request reply was sent, false otherwise.
  */
 static bool batadv_send_tt_response(struct batadv_priv *bat_priv,
 				    struct batadv_tvlv_tt_data *tt_data,
@@ -2933,7 +2944,7 @@ static void batadv_tt_update_changes(struct batadv_priv *bat_priv,
  * @addr: the mac address of the client to check
  * @vid: VLAN identifier
  *
- * Returns true if the client is served by this node, false otherwise.
+ * Return: true if the client is served by this node, false otherwise.
  */
 bool batadv_is_my_client(struct batadv_priv *bat_priv, const u8 *addr,
 			 unsigned short vid)
@@ -3050,11 +3061,14 @@ static void batadv_tt_roam_purge(struct batadv_priv *bat_priv)
 	spin_unlock_bh(&bat_priv->tt.roam_list_lock);
 }
 
-/* This function checks whether the client already reached the
+/**
+ * batadv_tt_check_roam_count
+ *
+ * This function checks whether the client already reached the
  * maximum number of possible roaming phases. In this case the ROAMING_ADV
  * will not be sent.
  *
- * returns true if the ROAMING_ADV can be sent, false otherwise
+ * Return: true if the ROAMING_ADV can be sent, false otherwise
  */
 static bool batadv_tt_check_roam_count(struct batadv_priv *bat_priv, u8 *client)
 {
@@ -3453,7 +3467,7 @@ request_table:
  * @addr: the mac address of the client to check
  * @vid: VLAN identifier
  *
- * Returns true if we know that the client has moved from its old originator
+ * Return: true if we know that the client has moved from its old originator
  * to another one. This entry is still kept for consistency purposes and will be
  * deleted later by a DEL or because of timeout
  */
@@ -3479,7 +3493,7 @@ out:
  * @addr: the mac address of the local client to query
  * @vid: VLAN identifier
  *
- * Returns true if the local client is known to be roaming (it is not served by
+ * Return: true if the local client is known to be roaming (it is not served by
  * this node anymore) or not. If yes, the client is still present in the table
  * to keep the latter consistent with the node TTVN
  */
@@ -3608,7 +3622,7 @@ static void batadv_tt_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv,
  * @tvlv_value: tvlv buffer containing the tt data
  * @tvlv_value_len: tvlv buffer length
  *
- * Returns NET_RX_DROP if the tt tvlv is to be re-routed, NET_RX_SUCCESS
+ * Return: NET_RX_DROP if the tt tvlv is to be re-routed, NET_RX_SUCCESS
  * otherwise.
  */
 static int batadv_tt_tvlv_unicast_handler_v1(struct batadv_priv *bat_priv,
@@ -3689,7 +3703,7 @@ static int batadv_tt_tvlv_unicast_handler_v1(struct batadv_priv *bat_priv,
  * @tvlv_value: tvlv buffer containing the tt data
  * @tvlv_value_len: tvlv buffer length
  *
- * Returns NET_RX_DROP if the tt roam tvlv is to be re-routed, NET_RX_SUCCESS
+ * Return: NET_RX_DROP if the tt roam tvlv is to be re-routed, NET_RX_SUCCESS
  * otherwise.
  */
 static int batadv_roam_tvlv_unicast_handler_v1(struct batadv_priv *bat_priv,
@@ -3735,7 +3749,7 @@ out:
  * batadv_tt_init - initialise the translation table internals
  * @bat_priv: the bat priv with all the soft interface information
  *
- * Return 0 on success or negative error number in case of failure.
+ * Return: 0 on success or negative error number in case of failure.
  */
 int batadv_tt_init(struct batadv_priv *bat_priv)
 {
@@ -3773,7 +3787,7 @@ int batadv_tt_init(struct batadv_priv *bat_priv)
  * @addr: the mac address of the client
  * @vid: the identifier of the VLAN where this client is connected
  *
- * Returns true if the client is marked with the TT_CLIENT_ISOLA flag, false
+ * Return: true if the client is marked with the TT_CLIENT_ISOLA flag, false
  * otherwise
  */
 bool batadv_tt_global_is_isolated(struct batadv_priv *bat_priv,
-- 
cgit v0.10.2


From 7afcbbef6471130a1eb586fea0f5c06609b8341f Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 31 Oct 2015 12:29:29 +0100
Subject: batman-adv: Fix kerneldoc of main functions

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bitarray.c b/net/batman-adv/bitarray.c
index 06bab51..3404195 100644
--- a/net/batman-adv/bitarray.c
+++ b/net/batman-adv/bitarray.c
@@ -29,10 +29,16 @@ static void batadv_bitmap_shift_left(unsigned long *seq_bits, s32 n)
 	bitmap_shift_left(seq_bits, seq_bits, n, BATADV_TQ_LOCAL_WINDOW_SIZE);
 }
 
-/* receive and process one packet within the sequence number window.
+/**
+ * batadv_bit_get_packet - receive and process one packet within the sequence
+ *  number window
+ * @priv: the bat priv with all the soft interface information
+ * @seq_bits: pointer to the sequence number receive packet
+ * @seq_num_diff: difference between the current/received sequence number and
+ *  the last sequence number
+ * @set_mark: whether this packet should be marked in seq_bits
  *
- * Return:
- *  1 if the window was moved (either new or very old)
+ * Return: 1 if the window was moved (either new or very old),
  *  0 if the window was not moved/shifted.
  */
 int batadv_bit_get_packet(void *priv, unsigned long *seq_bits, s32 seq_num_diff,
diff --git a/net/batman-adv/bitarray.h b/net/batman-adv/bitarray.h
index cf2aeb0..2b64d7a 100644
--- a/net/batman-adv/bitarray.h
+++ b/net/batman-adv/bitarray.h
@@ -25,7 +25,11 @@
 #include <linux/types.h>
 
 /**
- * batadv_test_bit
+ * batadv_test_bit - check if bit is set in the current window
+ *
+ * @seq_bits: pointer to the sequence number receive packet
+ * @last_seqno: latest sequence number in seq_bits
+ * @curr_seqno: sequence number to test for
  *
  * Return: 1 if the corresponding bit in the given seq_bits indicates true
  * and curr_seqno is within range of last_seqno. Otherwise returns 0.
@@ -51,11 +55,6 @@ static inline void batadv_set_bit(unsigned long *seq_bits, s32 n)
 	set_bit(n, seq_bits); /* turn the position on */
 }
 
-/**
- * batadv_bit_get_packet - receive and process one packet
- *
- * Return: 1 if received seq_num is considered new, 0 if old
- */
 int batadv_bit_get_packet(void *priv, unsigned long *seq_bits, s32 seq_num_diff,
 			  int set_mark);
 
diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c
index 037ad0a..0fc9df5 100644
--- a/net/batman-adv/debugfs.c
+++ b/net/batman-adv/debugfs.c
@@ -281,6 +281,8 @@ static int batadv_originators_open(struct inode *inode, struct file *file)
  *  originator table of an hard interface
  * @inode: inode pointer to debugfs file
  * @file: pointer to the seq_file
+ *
+ * Return: 0 on success or negative error number in case of failure
  */
 static int batadv_originators_hardif_open(struct inode *inode,
 					  struct file *file)
@@ -329,6 +331,8 @@ static int batadv_bla_backbone_table_open(struct inode *inode,
  * batadv_dat_cache_open - Prepare file handler for reads from dat_chache
  * @inode: inode which was opened
  * @file: file handle to be initialized
+ *
+ * Return: 0 on success or negative error number in case of failure
  */
 static int batadv_dat_cache_open(struct inode *inode, struct file *file)
 {
@@ -483,6 +487,8 @@ void batadv_debugfs_destroy(void)
  * batadv_debugfs_add_hardif - creates the base directory for a hard interface
  *  in debugfs.
  * @hard_iface: hard interface which should be added.
+ *
+ * Return: 0 on success or negative error number in case of failure
  */
 int batadv_debugfs_add_hardif(struct batadv_hard_iface *hard_iface)
 {
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 70907f6f..9d34be6 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -599,6 +599,8 @@ int batadv_algo_seq_print_text(struct seq_file *seq, void *offset)
  *
  * payload_ptr must always point to an address in the skb head buffer and not to
  * a fragment.
+ *
+ * Return: big endian crc32c of the checksummed data
  */
 __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr)
 {
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 8a48d4a..34f56ef 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -275,6 +275,8 @@ static inline void _batadv_dbg(int type __always_unused,
 
 /**
  * batadv_compare_eth - Compare two not u16 aligned Ethernet addresses
+ * @data1: Pointer to a six-byte array containing the Ethernet address
+ * @data2: Pointer other six-byte array containing the Ethernet address
  *
  * note: can't use ether_addr_equal() as it requires aligned memory
  *
@@ -331,6 +333,8 @@ static inline void batadv_add_counter(struct batadv_priv *bat_priv, size_t idx,
 
 /**
  * batadv_sum_counter - Sum the cpu-local counters for index 'idx'
+ * @bat_priv: the bat priv with all the soft interface information
+ * @idx: index of counter to sum up
  *
  * Return: sum of all cpu-local counters
  */
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index ed1d6d7..f6ca4e5 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -48,7 +48,9 @@ static struct lock_class_key batadv_orig_hash_lock_class_key;
 static void batadv_purge_orig(struct work_struct *work);
 
 /**
- * batadv_compare_orig
+ * batadv_compare_orig - comparing function used in the originator hash table
+ * @node: node in the local table
+ * @data2: second object to compare the node to
  *
  * Return: 1 if they are the same originator
  */
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index 72fd5b0..fdb0163 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -209,6 +209,11 @@ struct batadv_bla_claim_dst {
  * @version: batman-adv protocol version, part of the genereal header
  * @ttl: time to live for this packet, part of the genereal header
  * @flags: contains routing relevant flags - see enum batadv_iv_flags
+ * @seqno: sequence identification
+ * @orig: address of the source node
+ * @prev_sender: address of the previous sender
+ * @reserved: reserved byte for alignment
+ * @tq: transmission quality
  * @tvlv_len: length of tvlv data following the ogm header
  */
 struct batadv_ogm_packet {
@@ -345,6 +350,7 @@ struct batadv_unicast_packet {
  * @u: common unicast packet header
  * @src: address of the source
  * @subtype: packet subtype
+ * @reserved: reserved byte for alignment
  */
 struct batadv_unicast_4addr_packet {
 	struct batadv_unicast_packet u;
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index bcaa787..1fb1be3 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -141,8 +141,13 @@ out:
 }
 
 /**
- * batadv_window_protected checks whether the host restarted and is in the
+ * batadv_window_protected - checks whether the host restarted and is in the
  *  protection time.
+ * @bat_priv: the bat priv with all the soft interface information
+ * @seq_num_diff: difference between the current/received sequence number and
+ *  the last sequence number
+ * @last_reset: jiffies timestamp of the last reset, will be updated when reset
+ *  is detected
  *
  * Return:
  *  0 if the packet is to be accepted.
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index d9b93c5..c188f46 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -431,7 +431,10 @@ _batadv_add_bcast_packet_to_list(struct batadv_priv *bat_priv,
 }
 
 /**
- * batadv_add_bcast_packet_to_list
+ * batadv_add_bcast_packet_to_list - queue broadcast packet for multiple sends
+ * @bat_priv: the bat priv with all the soft interface information
+ * @skb: broadcast packet to add
+ * @delay: number of jiffies to wait before sending
  *
  * add a broadcast packet to the queue and setup timers. broadcast packets
  * are sent multiple times to increase probability for being received.
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 5ee794b..6c65de9 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -594,6 +594,7 @@ static void batadv_softif_destroy_vlan(struct batadv_priv *bat_priv,
 /**
  * batadv_interface_add_vid - ndo_add_vid API implementation
  * @dev: the netdev of the mesh interface
+ * @proto: protocol of the the vlan id
  * @vid: identifier of the new vlan
  *
  * Set up all the internal structures for handling the new vlan on top of the
@@ -651,6 +652,7 @@ static int batadv_interface_add_vid(struct net_device *dev, __be16 proto,
 /**
  * batadv_interface_kill_vid - ndo_kill_vid API implementation
  * @dev: the netdev of the mesh interface
+ * @proto: protocol of the the vlan id
  * @vid: identifier of the deleted vlan
  *
  * Destroy all the internal structures used to handle the vlan identified by vid
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index 7658e6e..f38d7b7 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -82,6 +82,7 @@ static struct batadv_priv *batadv_vlan_kobj_to_batpriv(struct kobject *obj)
 
 /**
  * batadv_kobj_to_vlan - convert a kobj in the associated softif_vlan struct
+ * @bat_priv: the bat priv with all the soft interface information
  * @obj: kobject to covert
  *
  * Return: the associated softif_vlan struct if found, NULL otherwise.
-- 
cgit v0.10.2


From 672e797850422bdbbcd3c781a07fe777042ee068 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 31 Oct 2015 12:29:30 +0100
Subject: batman-adv: Fix kerneldoc of network-coding functions

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 1a75d28..3a752be 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -64,6 +64,8 @@ static int batadv_nc_recv_coded_packet(struct sk_buff *skb,
 
 /**
  * batadv_nc_init - one-time initialization for network coding
+ *
+ * Return: 0 on success or negative error number in case of failure
  */
 int __init batadv_nc_init(void)
 {
@@ -142,6 +144,8 @@ static void batadv_nc_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv,
 /**
  * batadv_nc_mesh_init - initialise coding hash table and start house keeping
  * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return: 0 on success or negative error number in case of failure
  */
 int batadv_nc_mesh_init(struct batadv_priv *bat_priv)
 {
@@ -989,6 +993,8 @@ static struct batadv_nc_path *batadv_nc_get_path(struct batadv_priv *bat_priv,
  * batadv_nc_random_weight_tq - scale the receivers TQ-value to avoid unfair
  *  selection of a receiver with slightly lower TQ than the other
  * @tq: to be weighted tq value
+ *
+ * Return: scaled tq value
  */
 static u8 batadv_nc_random_weight_tq(u8 tq)
 {
@@ -1781,6 +1787,9 @@ batadv_nc_find_decoding_packet(struct batadv_priv *bat_priv,
  *  resulting unicast packet
  * @skb: incoming coded packet
  * @recv_if: pointer to interface this packet was received on
+ *
+ * Return: NET_RX_SUCCESS if the packet has been consumed or NET_RX_DROP
+ * otherwise.
  */
 static int batadv_nc_recv_coded_packet(struct sk_buff *skb,
 				       struct batadv_hard_iface *recv_if)
@@ -1865,6 +1874,8 @@ void batadv_nc_mesh_free(struct batadv_priv *bat_priv)
  * batadv_nc_nodes_seq_print_text - print the nc node information
  * @seq: seq file to print on
  * @offset: not used
+ *
+ * Return: always 0
  */
 int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset)
 {
@@ -1927,6 +1938,8 @@ out:
 /**
  * batadv_nc_init_debugfs - create nc folder and related files in debugfs
  * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return: 0 on success or negative error number in case of failure
  */
 int batadv_nc_init_debugfs(struct batadv_priv *bat_priv)
 {
-- 
cgit v0.10.2


From 04e14be65bdb09f8ed87f75e0b5463ced6706949 Mon Sep 17 00:00:00 2001
From: Simon Wunderlich <sw@simonwunderlich.de>
Date: Fri, 6 Nov 2015 10:45:19 +0100
Subject: batman-adv: Update/repair bridge loop avoidance kerneldoc

Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 90c7994..131aca0 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -59,9 +59,11 @@ batadv_bla_send_announce(struct batadv_priv *bat_priv,
 			 struct batadv_bla_backbone_gw *backbone_gw);
 
 /**
- * batadv_choose_claim
+ * batadv_choose_claim - choose the right bucket for a claim.
+ * @data: data to hash
+ * @size: size of the hash table
  *
- * Return: the index of the claim
+ * Return: the hash index of the claim
  */
 static inline u32 batadv_choose_claim(const void *data, u32 size)
 {
@@ -75,9 +77,11 @@ static inline u32 batadv_choose_claim(const void *data, u32 size)
 }
 
 /**
- * batadv_choose_backbone_gw
+ * batadv_choose_backbone_gw - choose the right bucket for a backbone gateway.
+ * @data: data to hash
+ * @size: size of the hash table
  *
- * Return: the index of the backbone gateway
+ * Return: the hash index of the backbone gateway
  */
 static inline u32 batadv_choose_backbone_gw(const void *data, u32 size)
 {
@@ -90,7 +94,13 @@ static inline u32 batadv_choose_backbone_gw(const void *data, u32 size)
 	return hash % size;
 }
 
-/* compares address and vid of two backbone gws */
+/**
+ * batadv_compare_backbone_gw - compare address and vid of two backbone gws
+ * @node: list node of the first entry to compare
+ * @data2: pointer to the second backbone gateway
+ *
+ * Return: 1 if the backbones have the same data, 0 otherwise
+ */
 static int batadv_compare_backbone_gw(const struct hlist_node *node,
 				      const void *data2)
 {
@@ -108,7 +118,13 @@ static int batadv_compare_backbone_gw(const struct hlist_node *node,
 	return 1;
 }
 
-/* compares address and vid of two claims */
+/**
+ * batadv_compare_backbone_gw - compare address and vid of two claims
+ * @node: list node of the first entry to compare
+ * @data2: pointer to the second claims
+ *
+ * Return: 1 if the claim have the same data, 0 otherwise
+ */
 static int batadv_compare_claim(const struct hlist_node *node,
 				const void *data2)
 {
@@ -126,7 +142,10 @@ static int batadv_compare_claim(const struct hlist_node *node,
 	return 1;
 }
 
-/* free a backbone gw */
+/**
+ * batadv_compare_backbone_gw - free backbone gw
+ * @backbone_gw: backbone gateway to be free'd
+ */
 static void
 batadv_backbone_gw_free_ref(struct batadv_bla_backbone_gw *backbone_gw)
 {
@@ -134,14 +153,21 @@ batadv_backbone_gw_free_ref(struct batadv_bla_backbone_gw *backbone_gw)
 		kfree_rcu(backbone_gw, rcu);
 }
 
-/* finally deinitialize the claim */
+/**
+ * batadv_claim_release - release claim from lists and queue for free after rcu
+ *  grace period
+ * @ref: kref pointer of the claim
+ */
 static void batadv_claim_release(struct batadv_bla_claim *claim)
 {
 	batadv_backbone_gw_free_ref(claim->backbone_gw);
 	kfree_rcu(claim, rcu);
 }
 
-/* free a claim, call claim_free_rcu if its the last reference */
+/**
+ * batadv_claim_free_rcu - free a claim
+ * @claim: claim to be free'd
+ */
 static void batadv_claim_free_ref(struct batadv_bla_claim *claim)
 {
 	if (atomic_dec_and_test(&claim->refcount))
@@ -149,12 +175,10 @@ static void batadv_claim_free_ref(struct batadv_bla_claim *claim)
 }
 
 /**
- * batadv_claim_hash_find
+ * batadv_claim_hash_find - looks for a claim in the claim hash
  * @bat_priv: the bat priv with all the soft interface information
  * @data: search data (may be local/static data)
  *
- * looks for a claim in the hash
- *
  * Return: claim if found or NULL otherwise.
  */
 static struct batadv_bla_claim
@@ -190,12 +214,12 @@ static struct batadv_bla_claim
 }
 
 /**
- * batadv_backbone_hash_find - looks for a claim in the hash
+ * batadv_backbone_hash_find - looks for a backbone gateway in the hash
  * @bat_priv: the bat priv with all the soft interface information
  * @addr: the address of the originator
  * @vid: the VLAN ID
  *
- * Return: claim if found or NULL otherwise.
+ * Return: backbone gateway if found or NULL otherwise
  */
 static struct batadv_bla_backbone_gw *
 batadv_backbone_hash_find(struct batadv_priv *bat_priv, u8 *addr,
@@ -233,7 +257,10 @@ batadv_backbone_hash_find(struct batadv_priv *bat_priv, u8 *addr,
 	return backbone_gw_tmp;
 }
 
-/* delete all claims for a backbone */
+/**
+ * batadv_bla_del_backbone_claims - delete all claims for a backbone
+ * @backbone_gw: backbone gateway where the claims should be removed
+ */
 static void
 batadv_bla_del_backbone_claims(struct batadv_bla_backbone_gw *backbone_gw)
 {
@@ -381,14 +408,13 @@ out:
 }
 
 /**
- * batadv_bla_get_backbone_gw
+ * batadv_bla_get_backbone_gw - finds or creates a backbone gateway
  * @bat_priv: the bat priv with all the soft interface information
  * @orig: the mac address of the originator
  * @vid: the VLAN ID
  * @own_backbone: set if the requested backbone is local
  *
- * searches for the backbone gw or creates a new one if it could not
- * be found.
+ * Return: the (possibly created) backbone gateway or NULL on error
  */
 static struct batadv_bla_backbone_gw *
 batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, u8 *orig,
@@ -454,7 +480,13 @@ batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, u8 *orig,
 	return entry;
 }
 
-/* update or add the own backbone gw to make sure we announce
+/**
+ * batadv_bla_update_own_backbone_gw - updates the own backbone gw for a VLAN
+ * @bat_priv: the bat priv with all the soft interface information
+ * @primary_if: the selected primary interface
+ * @vid: VLAN identifier
+ *
+ * update or add the own backbone gw to make sure we announce
  * where we receive other backbone gws
  */
 static void
@@ -551,12 +583,9 @@ static void batadv_bla_send_request(struct batadv_bla_backbone_gw *backbone_gw)
 }
 
 /**
- * batadv_bla_send_announce
+ * batadv_bla_send_announce - Send an announcement frame
  * @bat_priv: the bat priv with all the soft interface information
  * @backbone_gw: our backbone gateway which should be announced
- *
- * This function sends an announcement. It is called from multiple
- * places.
  */
 static void batadv_bla_send_announce(struct batadv_priv *bat_priv,
 				     struct batadv_bla_backbone_gw *backbone_gw)
@@ -646,8 +675,11 @@ claim_free_ref:
 	batadv_claim_free_ref(claim);
 }
 
-/* Delete a claim from the claim hash which has the
- * given mac address and vid.
+/**
+ * batadv_bla_del_claim - delete a claim from the claim hash
+ * @bat_priv: the bat priv with all the soft interface information
+ * @mac: mac address of the claim to be removed
+ * @vid: VLAN id for the claim to be removed
  */
 static void batadv_bla_del_claim(struct batadv_priv *bat_priv,
 				 const u8 *mac, const unsigned short vid)
@@ -677,6 +709,10 @@ static void batadv_bla_del_claim(struct batadv_priv *bat_priv,
 
 /**
  * batadv_handle_announce - check for ANNOUNCE frame
+ * @bat_priv: the bat priv with all the soft interface information
+ * @an_addr: announcement mac address (ARP Sender HW address)
+ * @backbone_addr: originator address of the sender (Ethernet source MAC)
+ * @vid: the VLAN ID of the frame
  *
  * Return: 1 if handled
  */
@@ -731,6 +767,11 @@ static int batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr,
 
 /**
  * batadv_handle_request - check for REQUEST frame
+ * @bat_priv: the bat priv with all the soft interface information
+ * @primary_if: the primary hard interface of this batman soft interface
+ * @backbone_addr: backbone address to be requested (ARP sender HW MAC)
+ * @ethhdr: ethernet header of a packet
+ * @vid: the VLAN ID of the frame
  *
  * Return: 1 if handled
  */
@@ -759,6 +800,11 @@ static int batadv_handle_request(struct batadv_priv *bat_priv,
 
 /**
  * batadv_handle_unclaim - check for UNCLAIM frame
+ * @bat_priv: the bat priv with all the soft interface information
+ * @primary_if: the primary hard interface of this batman soft interface
+ * @backbone_addr: originator address of the backbone (Ethernet source)
+ * @claim_addr: Client to be unclaimed (ARP sender HW MAC)
+ * @vid: the VLAN ID of the frame
  *
  * Return: 1 if handled
  */
@@ -792,6 +838,11 @@ static int batadv_handle_unclaim(struct batadv_priv *bat_priv,
 
 /**
  * batadv_handle_claim - check for CLAIM frame
+ * @bat_priv: the bat priv with all the soft interface information
+ * @primary_if: the primary hard interface of this batman soft interface
+ * @backbone_addr: originator address of the backbone (Ethernet Source)
+ * @claim_addr: client mac address to be claimed (ARP sender HW MAC)
+ * @vid: the VLAN ID of the frame
  *
  * Return: 1 if handled
  */
@@ -823,7 +874,7 @@ static int batadv_handle_claim(struct batadv_priv *bat_priv,
 }
 
 /**
- * batadv_check_claim_group
+ * batadv_check_claim_group - check for claim group membership
  * @bat_priv: the bat priv with all the soft interface information
  * @primary_if: the primary interface of this batman interface
  * @hw_src: the Hardware source in the ARP Header
@@ -898,13 +949,11 @@ static int batadv_check_claim_group(struct batadv_priv *bat_priv,
 }
 
 /**
- * batadv_bla_process_claim
+ * batadv_bla_process_claim - Check if this is a claim frame, and process it
  * @bat_priv: the bat priv with all the soft interface information
  * @primary_if: the primary hard interface of this batman soft interface
  * @skb: the frame to be checked
  *
- * Check if this is a claim frame, and process it accordingly.
- *
  * Return: 1 if it was a claim frame, otherwise return 0 to
  * tell the callee that it can use the frame on its own.
  */
@@ -1036,7 +1085,13 @@ static int batadv_bla_process_claim(struct batadv_priv *bat_priv,
 	return 1;
 }
 
-/* Check when we last heard from other nodes, and remove them in case of
+/**
+ * batadv_bla_purge_backbone_gw - Remove backbone gateways after a timeout or
+ *  immediately
+ * @bat_priv: the bat priv with all the soft interface information
+ * @now: whether the whole hash shall be wiped now
+ *
+ * Check when we last heard from other nodes, and remove them in case of
  * a time out, or clean all backbone gws if now is set.
  */
 static void batadv_bla_purge_backbone_gw(struct batadv_priv *bat_priv, int now)
@@ -1084,7 +1139,7 @@ purge_now:
 }
 
 /**
- * batadv_bla_purge_claims
+ * batadv_bla_purge_claims - Remove claims after a timeout or immediately
  * @bat_priv: the bat priv with all the soft interface information
  * @primary_if: the selected primary interface, may be NULL if now is set
  * @now: whether the whole hash shall be wiped now
@@ -1133,12 +1188,11 @@ purge_now:
 }
 
 /**
- * batadv_bla_update_orig_address
+ * batadv_bla_update_orig_address - Update the backbone gateways when the own
+ *  originator address changes
  * @bat_priv: the bat priv with all the soft interface information
  * @primary_if: the new selected primary_if
  * @oldif: the old primary interface, may be NULL
- *
- * Update the backbone gateways when the own orig address changes.
  */
 void batadv_bla_update_orig_address(struct batadv_priv *bat_priv,
 				    struct batadv_hard_iface *primary_if,
@@ -1209,7 +1263,11 @@ void batadv_bla_status_update(struct net_device *net_dev)
 	batadv_hardif_free_ref(primary_if);
 }
 
-/* periodic work to do:
+/**
+ * batadv_bla_periodic_work - performs periodic bla work
+ * @work: kernel work struct
+ *
+ * periodic work to do:
  *  * purge structures when they are too old
  *  * send announcements
  */
@@ -1290,7 +1348,12 @@ out:
 static struct lock_class_key batadv_claim_hash_lock_class_key;
 static struct lock_class_key batadv_backbone_hash_lock_class_key;
 
-/* initialize all bla structures */
+/**
+ * batadv_bla_init - initialize all bla structures
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return: 0 on success, < 0 on error.
+ */
 int batadv_bla_init(struct batadv_priv *bat_priv)
 {
 	int i;
@@ -1345,7 +1408,7 @@ int batadv_bla_init(struct batadv_priv *bat_priv)
 }
 
 /**
- * batadv_bla_check_bcast_duplist
+ * batadv_bla_check_bcast_duplist - Check if a frame is in the broadcast dup.
  * @bat_priv: the bat priv with all the soft interface information
  * @skb: contains the bcast_packet to be checked
  *
@@ -1357,6 +1420,8 @@ int batadv_bla_init(struct batadv_priv *bat_priv)
  * with a good chance that it is the same packet. If it is furthermore
  * sent by another host, drop it. We allow equal packets from
  * the same host however as this might be intended.
+ *
+ * Return: 1 if a packet is in the duplicate list, 0 otherwise.
  */
 int batadv_bla_check_bcast_duplist(struct batadv_priv *bat_priv,
 				   struct sk_buff *skb)
@@ -1415,13 +1480,12 @@ out:
 }
 
 /**
- * batadv_bla_is_backbone_gw_orig
+ * batadv_bla_is_backbone_gw_orig - Check if the originator is a gateway for
+ *  the VLAN identified by vid.
  * @bat_priv: the bat priv with all the soft interface information
  * @orig: originator mac address
  * @vid: VLAN identifier
  *
- * Check if the originator is a gateway for the VLAN identified by vid.
- *
  * Return: true if orig is a backbone for this vid, false otherwise.
  */
 bool batadv_bla_is_backbone_gw_orig(struct batadv_priv *bat_priv, u8 *orig,
@@ -1456,13 +1520,11 @@ bool batadv_bla_is_backbone_gw_orig(struct batadv_priv *bat_priv, u8 *orig,
 }
 
 /**
- * batadv_bla_is_backbone_gw
+ * batadv_bla_is_backbone_gw - check if originator is a backbone gw for a VLAN.
  * @skb: the frame to be checked
  * @orig_node: the orig_node of the frame
  * @hdr_size: maximum length of the frame
  *
- * bla_is_backbone_gw inspects the skb for the VLAN ID
- *
  * Return: 1 if the orig_node is also a gateway on the soft interface, otherwise
  * it returns 0.
  */
@@ -1491,7 +1553,12 @@ int batadv_bla_is_backbone_gw(struct sk_buff *skb,
 	return 1;
 }
 
-/* free all bla structures (for softinterface free or module unload) */
+/**
+ * batadv_bla_init - free all bla structures
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * for softinterface free or module unload
+ */
 void batadv_bla_free(struct batadv_priv *bat_priv)
 {
 	struct batadv_hard_iface *primary_if;
@@ -1514,13 +1581,13 @@ void batadv_bla_free(struct batadv_priv *bat_priv)
 }
 
 /**
- * batadv_bla_rx
+ * batadv_bla_rx - check packets coming from the mesh.
  * @bat_priv: the bat priv with all the soft interface information
  * @skb: the frame to be checked
  * @vid: the VLAN ID of the frame
  * @is_bcast: the packet came in a broadcast packet type.
  *
- * bla_rx avoidance checks if:
+ * batadv_bla_rx avoidance checks if:
  *  * we have to race for a claim
  *  * if the frame is allowed on the LAN
  *
@@ -1610,12 +1677,12 @@ out:
 }
 
 /**
- * batadv_bla_tx
+ * batadv_bla_tx - check packets going into the mesh
  * @bat_priv: the bat priv with all the soft interface information
  * @skb: the frame to be checked
  * @vid: the VLAN ID of the frame
  *
- * bla_tx checks if:
+ * batadv_bla_tx checks if:
  *  * a claim was received which has to be processed
  *  * the frame is allowed on the mesh
  *
@@ -1698,6 +1765,13 @@ out:
 	return ret;
 }
 
+/**
+ * batadv_bla_claim_table_seq_print_text - print the claim table in a seq file
+ * @seq: seq file to print on
+ * @offset: not used
+ *
+ * Return: always 0
+ */
 int batadv_bla_claim_table_seq_print_text(struct seq_file *seq, void *offset)
 {
 	struct net_device *net_dev = (struct net_device *)seq->private;
@@ -1747,6 +1821,14 @@ out:
 	return 0;
 }
 
+/**
+ * batadv_bla_backbone_table_seq_print_text - print the backbone table in a seq
+ *  file
+ * @seq: seq file to print on
+ * @offset: not used
+ *
+ * Return: always 0
+ */
 int batadv_bla_backbone_table_seq_print_text(struct seq_file *seq, void *offset)
 {
 	struct net_device *net_dev = (struct net_device *)seq->private;
-- 
cgit v0.10.2


From fe13c2aadf208e6210ee125f550dcf725c48721b Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <a@unstable.cc>
Date: Tue, 17 Nov 2015 16:40:51 +0800
Subject: batman-adv: fix kerneldoc for DAT functions

Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 2f13429..60df823 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -741,6 +741,8 @@ static void batadv_dat_hash_free(struct batadv_priv *bat_priv)
 /**
  * batadv_dat_init - initialise the DAT internals
  * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return: 0 in case of success, a negative error code otherwise
  */
 int batadv_dat_init(struct batadv_priv *bat_priv)
 {
@@ -779,6 +781,8 @@ void batadv_dat_free(struct batadv_priv *bat_priv)
  * batadv_dat_cache_seq_print_text - print the local DAT hash table
  * @seq: seq file to print on
  * @offset: not used
+ *
+ * Return: always 0
  */
 int batadv_dat_cache_seq_print_text(struct seq_file *seq, void *offset)
 {
-- 
cgit v0.10.2


From d15cd6221c74027af005d5acbefa43f67ff3df37 Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <a@unstable.cc>
Date: Tue, 17 Nov 2015 16:40:52 +0800
Subject: batman-adv: fix kerneldoc for TT functions

Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index af1d24c..9fcf6bf 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -69,9 +69,13 @@ static void batadv_tt_global_del(struct batadv_priv *bat_priv,
 				 bool roaming);
 
 /**
- * batadv_compare_tt
+ * batadv_compare_tt - check if two TT entries are the same
+ * @node: the list element pointer of the first TT entry
+ * @data2: pointer to the tt_common_entry of the second TT entry
  *
- * Return: 1 if they are the same mac addr and vid
+ * Compare the MAC address and the VLAN ID of the two TT entries and check if
+ * they are the same TT client.
+ * Return: 1 if the two TT clients are the same, 0 otherwise
  */
 static int batadv_compare_tt(const struct hlist_node *node, const void *data2)
 {
@@ -221,6 +225,7 @@ batadv_tt_global_entry_free_ref(struct batadv_tt_global_entry *tt_global_entry)
 
 /**
  * batadv_tt_global_hash_count - count the number of orig entries
+ * @bat_priv: the bat priv with all the soft interface information
  * @addr: the mac address of the client to count entries for
  * @vid: VLAN identifier
  *
@@ -289,8 +294,9 @@ static void batadv_tt_local_size_dec(struct batadv_priv *bat_priv,
 }
 
 /**
- * batadv_tt_global_size_mod - change the size by v of the local table
- *  identified by vid
+ * batadv_tt_global_size_mod - change the size by v of the global table
+ *  for orig_node identified by vid
+ * @orig_node: the originator for which the table has to be modified
  * @vid: the VLAN identifier
  * @v: the amount to sum to the global table size
  */
@@ -721,7 +727,6 @@ out:
  *  function reserves the amount of space needed to send the entire global TT
  *  table. In case of success the value is updated with the real amount of
  *  reserved bytes
-
  * Allocate the needed amount of memory for the entire TT TVLV and write its
  * header made up by one tvlv_tt_data object and a series of tvlv_tt_vlan_data
  * objects, one per active VLAN served by the originator node.
@@ -1243,9 +1248,12 @@ static void batadv_tt_changes_list_free(struct batadv_priv *bat_priv)
 }
 
 /**
- * batadv_tt_global_orig_entry_find
+ * batadv_tt_global_orig_entry_find - find a TT orig_list_entry
+ * @entry: the TT global entry where the orig_list_entry has to be
+ *  extracted from
+ * @orig_node: the originator for which the orig_list_entry has to be found
  *
- * retrieves the orig_tt_list_entry belonging to orig_node from the
+ * retrieve the orig_tt_list_entry belonging to orig_node from the
  * batadv_tt_global_entry list
  *
  * Return: it with an increased refcounter, NULL if not found
@@ -1274,7 +1282,10 @@ batadv_tt_global_orig_entry_find(const struct batadv_tt_global_entry *entry,
 }
 
 /**
- * batadv_tt_global_entry_has_orig
+ * batadv_tt_global_entry_has_orig - check if a TT global entry is also handled
+ *  by a given originator
+ * @entry: the TT global entry to check
+ * @orig_node: the originator to search in the list
  *
  * find out if an orig_node is already in the list of a tt_global_entry.
  *
@@ -2519,6 +2530,8 @@ static void batadv_tt_global_update_crc(struct batadv_priv *bat_priv,
  * @num_vlan: number of tvlv VLAN entries
  * @full_table: ask for the entire translation table if true, while only for the
  *  last TT diff otherwise
+ *
+ * Return: true if the TT Request was sent, false otherwise
  */
 static int batadv_send_tt_request(struct batadv_priv *bat_priv,
 				  struct batadv_orig_node *dst_orig_node,
@@ -3062,7 +3075,9 @@ static void batadv_tt_roam_purge(struct batadv_priv *bat_priv)
 }
 
 /**
- * batadv_tt_check_roam_count
+ * batadv_tt_check_roam_count - check if a client has roamed too frequently
+ * @bat_priv: the bat priv with all the soft interface information
+ * @client: mac address of the roaming client
  *
  * This function checks whether the client already reached the
  * maximum number of possible roaming phases. In this case the ROAMING_ADV
-- 
cgit v0.10.2


From a14c131d8c056397b34bfde63b398b712637d60f Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <a@unstable.cc>
Date: Tue, 17 Nov 2015 16:40:53 +0800
Subject: batman-adv: add kernel doc for AP isolation attributes in bat_priv

Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 7f7a0f1..5e8c8df 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -772,6 +772,9 @@ struct batadv_softif_vlan {
  * @orig_interval: OGM broadcast interval in milliseconds
  * @hop_penalty: penalty which will be applied to an OGM's tq-field on every hop
  * @log_level: configured log level (see batadv_dbg_level)
+ * @isolation_mark: the skb->mark value used to match packets for AP isolation
+ * @isolation_mark_mask: bitmask identifying the bits in skb->mark to be used
+ *  for the isolation mark
  * @bcast_seqno: last sent broadcast packet sequence number
  * @bcast_queue_left: number of remaining buffered broadcast packet slots
  * @batman_queue_left: number of remaining OGM packet slots
-- 
cgit v0.10.2


From 600184b705ae040f5ff7442466d0ec2bbae5560c Mon Sep 17 00:00:00 2001
From: Simon Wunderlich <simon@open-mesh.com>
Date: Mon, 23 Nov 2015 19:57:21 +0100
Subject: batman-adv: add kerneldoc for batadv_iv_ogm_aggr_packet

Signed-off-by: Simon Wunderlich <simon@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index a6d389b..7377313 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -396,7 +396,14 @@ static u8 batadv_hop_penalty(u8 tq, const struct batadv_priv *bat_priv)
 	return new_tq;
 }
 
-/* is there another aggregated packet here? */
+/**
+ * batadv_iv_ogm_aggr_packet - checks if there is another OGM attached
+ * @buff_pos: current position in the skb
+ * @packet_len: total length of the skb
+ * @tvlv_len: tvlv length of the previously considered OGM
+ *
+ * Return: true if there is enough space for another OGM, false otherwise.
+ */
 static bool batadv_iv_ogm_aggr_packet(int buff_pos, int packet_len,
 				      __be16 tvlv_len)
 {
-- 
cgit v0.10.2


From ec9b83ca5fc1b39ed6e710b24484a272f9bcc3ae Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Tue, 5 Jan 2016 12:06:16 +0100
Subject: batman-adv: Fix kernel-doc for batadv_claim_free_ref

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 131aca0..b3a72e2 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -165,7 +165,8 @@ static void batadv_claim_release(struct batadv_bla_claim *claim)
 }
 
 /**
- * batadv_claim_free_rcu - free a claim
+ * batadv_claim_free_ref - decrement the claim refcounter and possibly
+ *  release it
  * @claim: claim to be free'd
  */
 static void batadv_claim_free_ref(struct batadv_bla_claim *claim)
-- 
cgit v0.10.2


From 0046b0402ac284a31940e3354614ac88c64881a3 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Fri, 1 Jan 2016 00:01:03 +0100
Subject: batman-adv: update copyright years for 2016

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index 21434ab..207e2af 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -1,5 +1,5 @@
 #
-# Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+# Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
 #
 # Marek Lindner, Simon Wunderlich
 #
diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h
index 4e59cf3..a7485d6 100644
--- a/net/batman-adv/bat_algo.h
+++ b/net/batman-adv/bat_algo.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2011-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 7377313..3266bcb 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/bitarray.c b/net/batman-adv/bitarray.c
index 3404195..b56bb00 100644
--- a/net/batman-adv/bitarray.c
+++ b/net/batman-adv/bitarray.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2006-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2016  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  *
diff --git a/net/batman-adv/bitarray.h b/net/batman-adv/bitarray.h
index 2b64d7a..3e41bb8 100644
--- a/net/batman-adv/bitarray.h
+++ b/net/batman-adv/bitarray.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2006-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2016  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  *
diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index b3a72e2..7791609 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2011-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2016  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich
  *
diff --git a/net/batman-adv/bridge_loop_avoidance.h b/net/batman-adv/bridge_loop_avoidance.h
index 7ea199b..579f0fa 100644
--- a/net/batman-adv/bridge_loop_avoidance.h
+++ b/net/batman-adv/bridge_loop_avoidance.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2011-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2016  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich
  *
diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c
index 0fc9df5..48253cf 100644
--- a/net/batman-adv/debugfs.c
+++ b/net/batman-adv/debugfs.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2010-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/debugfs.h b/net/batman-adv/debugfs.h
index 80ab8d6..1ab4e2e6 100644
--- a/net/batman-adv/debugfs.h
+++ b/net/batman-adv/debugfs.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2010-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 60df823..017fffe 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2011-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2016  B.A.T.M.A.N. contributors:
  *
  * Antonio Quartulli
  *
diff --git a/net/batman-adv/distributed-arp-table.h b/net/batman-adv/distributed-arp-table.h
index 26d4a52..813ecea 100644
--- a/net/batman-adv/distributed-arp-table.h
+++ b/net/batman-adv/distributed-arp-table.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2011-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2011-2016  B.A.T.M.A.N. contributors:
  *
  * Antonio Quartulli
  *
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index 80eddf4..55656e8 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2013-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2016  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll <martin@hundeboll.net>
  *
diff --git a/net/batman-adv/fragmentation.h b/net/batman-adv/fragmentation.h
index c5476fe..9ff77c7 100644
--- a/net/batman-adv/fragmentation.h
+++ b/net/batman-adv/fragmentation.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2013-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2013-2016  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll <martin@hundeboll.net>
  *
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 8350775..5950974 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2009-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/gateway_client.h b/net/batman-adv/gateway_client.h
index fa95277..582dd8c 100644
--- a/net/batman-adv/gateway_client.h
+++ b/net/batman-adv/gateway_client.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2009-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c
index b287448..5ee04f7 100644
--- a/net/batman-adv/gateway_common.c
+++ b/net/batman-adv/gateway_common.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2009-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/gateway_common.h b/net/batman-adv/gateway_common.h
index ab893e3..b583463 100644
--- a/net/batman-adv/gateway_common.h
+++ b/net/batman-adv/gateway_common.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2009-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 49e05d2..db90022 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/hard-interface.h b/net/batman-adv/hard-interface.h
index 7b12ea8..4d6b5e1 100644
--- a/net/batman-adv/hard-interface.h
+++ b/net/batman-adv/hard-interface.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/hash.c b/net/batman-adv/hash.c
index 2ea6a18..a0a0fdb 100644
--- a/net/batman-adv/hash.c
+++ b/net/batman-adv/hash.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2006-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2016  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  *
diff --git a/net/batman-adv/hash.h b/net/batman-adv/hash.h
index 071a87e..9bb57b8 100644
--- a/net/batman-adv/hash.h
+++ b/net/batman-adv/hash.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2006-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2006-2016  B.A.T.M.A.N. contributors:
  *
  * Simon Wunderlich, Marek Lindner
  *
diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c
index bcabb5e..a69da37 100644
--- a/net/batman-adv/icmp_socket.c
+++ b/net/batman-adv/icmp_socket.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/icmp_socket.h b/net/batman-adv/icmp_socket.h
index e937143..618d5de 100644
--- a/net/batman-adv/icmp_socket.h
+++ b/net/batman-adv/icmp_socket.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 9d34be6..5f319fd 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 34f56ef..a7dc41a 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c
index d984eee..155565e 100644
--- a/net/batman-adv/multicast.c
+++ b/net/batman-adv/multicast.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2014-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2014-2016  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing
  *
diff --git a/net/batman-adv/multicast.h b/net/batman-adv/multicast.h
index 685219f..80bceec 100644
--- a/net/batman-adv/multicast.h
+++ b/net/batman-adv/multicast.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2014-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2014-2016  B.A.T.M.A.N. contributors:
  *
  * Linus Lüssing
  *
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 3a752be..0b30c15 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2012-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2012-2016  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll, Jeppe Ledet-Pedersen
  *
diff --git a/net/batman-adv/network-coding.h b/net/batman-adv/network-coding.h
index 8f6d4ad..d6d7fb4 100644
--- a/net/batman-adv/network-coding.h
+++ b/net/batman-adv/network-coding.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2012-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2012-2016  B.A.T.M.A.N. contributors:
  *
  * Martin Hundebøll, Jeppe Ledet-Pedersen
  *
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index f6ca4e5..d4a30db 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2009-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2009-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index cf07304..745b4e4 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index fdb0163..e7f9151 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 1fb1be3..f4b60b1 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index 204bbe4..c776e96 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index c188f46..d8b03fd 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h
index 3adc11f..7ff95ca 100644
--- a/net/batman-adv/send.h
+++ b/net/batman-adv/send.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 6c65de9..4bf35b8 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
diff --git a/net/batman-adv/soft-interface.h b/net/batman-adv/soft-interface.h
index 8e82176..d17cfba 100644
--- a/net/batman-adv/soft-interface.h
+++ b/net/batman-adv/soft-interface.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index f38d7b7..964fc59 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2010-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/sysfs.h b/net/batman-adv/sysfs.h
index 6197442..c76021b 100644
--- a/net/batman-adv/sysfs.h
+++ b/net/batman-adv/sysfs.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2010-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2010-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner
  *
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 9fcf6bf..0dc8a5c 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich, Antonio Quartulli
  *
diff --git a/net/batman-adv/translation-table.h b/net/batman-adv/translation-table.h
index abd8e11..7c7e2c0 100644
--- a/net/batman-adv/translation-table.h
+++ b/net/batman-adv/translation-table.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich, Antonio Quartulli
  *
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 5e8c8df..8974bc0 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1,4 +1,4 @@
-/* Copyright (C) 2007-2015 B.A.T.M.A.N. contributors:
+/* Copyright (C) 2007-2016  B.A.T.M.A.N. contributors:
  *
  * Marek Lindner, Simon Wunderlich
  *
-- 
cgit v0.10.2


From 875e94392ad2be9776c8325d3573160eb1455a2b Mon Sep 17 00:00:00 2001
From: Maya Erez <qca_merez@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 19:24:02 +0200
Subject: wil6210: prevent access to vring_tx_data lock during its init

wil_tx_vring locks the vring_tx_data lock before accessing the TX
vring to check if it is enabled and valid for use.
In case of quick disconnect / connect events for the same station,
spin_lock(&txdata->lock) can be called during the lock initialization
in the vring init function.
To prevent such a race, the TX vrings spin lock should be initialized
once during wil6210 driver initialization.

Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 0652efe..712ebbf 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -438,6 +438,9 @@ int wil_priv_init(struct wil6210_priv *wil)
 	for (i = 0; i < WIL6210_MAX_CID; i++)
 		spin_lock_init(&wil->sta[i].tid_rx_lock);
 
+	for (i = 0; i < WIL6210_MAX_TX_RINGS; i++)
+		spin_lock_init(&wil->vring_tx_data[i].lock);
+
 	mutex_init(&wil->mutex);
 	mutex_init(&wil->wmi_mutex);
 	mutex_init(&wil->back_rx_mutex);
diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c
index 9680b97..6af2090 100644
--- a/drivers/net/wireless/ath/wil6210/txrx.c
+++ b/drivers/net/wireless/ath/wil6210/txrx.c
@@ -717,6 +717,21 @@ void wil_rx_fini(struct wil6210_priv *wil)
 		wil_vring_free(wil, vring, 0);
 }
 
+static inline void wil_tx_data_init(struct vring_tx_data *txdata)
+{
+	spin_lock_bh(&txdata->lock);
+	txdata->dot1x_open = 0;
+	txdata->enabled = 0;
+	txdata->idle = 0;
+	txdata->last_idle = 0;
+	txdata->begin = 0;
+	txdata->agg_wsize = 0;
+	txdata->agg_timeout = 0;
+	txdata->agg_amsdu = 0;
+	txdata->addba_in_progress = false;
+	spin_unlock_bh(&txdata->lock);
+}
+
 int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
 		      int cid, int tid)
 {
@@ -758,8 +773,7 @@ int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
 		goto out;
 	}
 
-	memset(txdata, 0, sizeof(*txdata));
-	spin_lock_init(&txdata->lock);
+	wil_tx_data_init(txdata);
 	vring->size = size;
 	rc = wil_vring_alloc(wil, vring);
 	if (rc)
@@ -791,8 +805,10 @@ int wil_vring_init_tx(struct wil6210_priv *wil, int id, int size,
 
 	return 0;
  out_free:
+	spin_lock_bh(&txdata->lock);
 	txdata->dot1x_open = false;
 	txdata->enabled = 0;
+	spin_unlock_bh(&txdata->lock);
 	wil_vring_free(wil, vring, 1);
 	wil->vring2cid_tid[id][0] = WIL6210_MAX_CID;
 	wil->vring2cid_tid[id][1] = 0;
@@ -834,8 +850,7 @@ int wil_vring_init_bcast(struct wil6210_priv *wil, int id, int size)
 		goto out;
 	}
 
-	memset(txdata, 0, sizeof(*txdata));
-	spin_lock_init(&txdata->lock);
+	wil_tx_data_init(txdata);
 	vring->size = size;
 	rc = wil_vring_alloc(wil, vring);
 	if (rc)
@@ -865,8 +880,10 @@ int wil_vring_init_bcast(struct wil6210_priv *wil, int id, int size)
 
 	return 0;
  out_free:
+	spin_lock_bh(&txdata->lock);
 	txdata->enabled = 0;
 	txdata->dot1x_open = false;
+	spin_unlock_bh(&txdata->lock);
 	wil_vring_free(wil, vring, 1);
  out:
 
@@ -894,7 +911,6 @@ void wil_vring_fini_tx(struct wil6210_priv *wil, int id)
 		napi_synchronize(&wil->napi_tx);
 
 	wil_vring_free(wil, vring, 1);
-	memset(txdata, 0, sizeof(*txdata));
 }
 
 static struct vring *wil_find_tx_ucast(struct wil6210_priv *wil,
-- 
cgit v0.10.2


From 78771d76f826202012201d700028a2d866d03fb3 Mon Sep 17 00:00:00 2001
From: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 19:24:03 +0200
Subject: wil6210: wait for disconnect completion

cfg80211_ops.disconnect() should wait for disconnect flow to
complete. If it does not, internal state becomes out of sync with
one in cfg80211. If one does stress test connect/disconnect
sequence, cfg80211 will issue next connect before disconnect
completed internally.

Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com>
Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 774352f..97ad91e 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -535,7 +535,18 @@ static int wil_cfg80211_disconnect(struct wiphy *wiphy,
 
 	wil_dbg_misc(wil, "%s(reason=%d)\n", __func__, reason_code);
 
-	rc = wmi_send(wil, WMI_DISCONNECT_CMDID, NULL, 0);
+	if (!(test_bit(wil_status_fwconnecting, wil->status) ||
+	      test_bit(wil_status_fwconnected, wil->status))) {
+		wil_err(wil, "%s: Disconnect was called while disconnected\n",
+			__func__);
+		return 0;
+	}
+
+	rc = wmi_call(wil, WMI_DISCONNECT_CMDID, NULL, 0,
+		      WMI_DISCONNECT_EVENTID, NULL, 0,
+		      WIL6210_DISCONNECT_TO_MS);
+	if (rc)
+		wil_err(wil, "%s: disconnect error %d\n", __func__, rc);
 
 	return rc;
 }
diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c
index 712ebbf..78ba6e0 100644
--- a/drivers/net/wireless/ath/wil6210/main.c
+++ b/drivers/net/wireless/ath/wil6210/main.c
@@ -23,9 +23,6 @@
 #include "wmi.h"
 #include "boot_loader.h"
 
-#define WAIT_FOR_DISCONNECT_TIMEOUT_MS 2000
-#define WAIT_FOR_DISCONNECT_INTERVAL_MS 10
-
 bool debug_fw; /* = false; */
 module_param(debug_fw, bool, S_IRUGO);
 MODULE_PARM_DESC(debug_fw, " do not perform card reset. For FW debug");
@@ -942,8 +939,7 @@ int wil_up(struct wil6210_priv *wil)
 
 int __wil_down(struct wil6210_priv *wil)
 {
-	int iter = WAIT_FOR_DISCONNECT_TIMEOUT_MS /
-			WAIT_FOR_DISCONNECT_INTERVAL_MS;
+	int rc;
 
 	WARN_ON(!mutex_is_locked(&wil->mutex));
 
@@ -967,22 +963,16 @@ int __wil_down(struct wil6210_priv *wil)
 	}
 
 	if (test_bit(wil_status_fwconnected, wil->status) ||
-	    test_bit(wil_status_fwconnecting, wil->status))
-		wmi_send(wil, WMI_DISCONNECT_CMDID, NULL, 0);
+	    test_bit(wil_status_fwconnecting, wil->status)) {
 
-	/* make sure wil is idle (not connected) */
-	mutex_unlock(&wil->mutex);
-	while (iter--) {
-		int idle = !test_bit(wil_status_fwconnected, wil->status) &&
-			   !test_bit(wil_status_fwconnecting, wil->status);
-		if (idle)
-			break;
-		msleep(WAIT_FOR_DISCONNECT_INTERVAL_MS);
+		mutex_unlock(&wil->mutex);
+		rc = wmi_call(wil, WMI_DISCONNECT_CMDID, NULL, 0,
+			      WMI_DISCONNECT_EVENTID, NULL, 0,
+			      WIL6210_DISCONNECT_TO_MS);
+		mutex_lock(&wil->mutex);
+		if (rc)
+			wil_err(wil, "timeout waiting for disconnect\n");
 	}
-	mutex_lock(&wil->mutex);
-
-	if (iter < 0)
-		wil_err(wil, "timeout waiting for idle FW/HW\n");
 
 	wil_reset(wil, false);
 
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 1b8fa1d..9502965 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -92,6 +92,7 @@ static inline u32 wil_mtu2macbuf(u32 mtu)
 #define WIL6210_FW_RECOVERY_RETRIES	(5) /* try to recover this many times */
 #define WIL6210_FW_RECOVERY_TO	msecs_to_jiffies(5000)
 #define WIL6210_SCAN_TO		msecs_to_jiffies(10000)
+#define WIL6210_DISCONNECT_TO_MS (2000)
 #define WIL6210_RX_HIGH_TRSH_INIT		(0)
 #define WIL6210_RX_HIGH_TRSH_DEFAULT \
 				(1 << (WIL_RX_RING_SIZE_ORDER_DEFAULT - 3))
-- 
cgit v0.10.2


From fe5c271e2878fb080f1b32fec5b4e3f7a6070152 Mon Sep 17 00:00:00 2001
From: Maya Erez <qca_merez@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 19:24:04 +0200
Subject: wil6210: protect synchronous wmi commands handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In case there are multiple WMI commands with the same reply_id,
the following scenario can occur:
- Driver sends the first command to the device
- The reply didn’t get on time and there is timeout
- Reply_id, reply_buf and reply_size are set to 0
- Driver sends second wmi command with the same reply_id as the first
- Driver sets wil->reply_id
- Reply for the first wmi command arrives and handled by wmi_recv_cmd
- As its ID fits the reply_id but the reply_buf is not set yet it is
handled as a reply with event handler, and WARN_ON is printed

This patch guarantee atomic setting of all the reply variables and
prevents the above scenario.

Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c
index e1a6cb8..493e721 100644
--- a/drivers/net/wireless/ath/wil6210/wmi.c
+++ b/drivers/net/wireless/ath/wil6210/wmi.c
@@ -838,6 +838,7 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 			struct wil6210_mbox_hdr_wmi *wmi = &evt->event.wmi;
 			u16 id = le16_to_cpu(wmi->id);
 			u32 tstamp = le32_to_cpu(wmi->timestamp);
+			spin_lock_irqsave(&wil->wmi_ev_lock, flags);
 			if (wil->reply_id && wil->reply_id == id) {
 				if (wil->reply_buf) {
 					memcpy(wil->reply_buf, wmi,
@@ -845,6 +846,7 @@ void wmi_recv_cmd(struct wil6210_priv *wil)
 					immed_reply = true;
 				}
 			}
+			spin_unlock_irqrestore(&wil->wmi_ev_lock, flags);
 
 			wil_dbg_wmi(wil, "WMI event 0x%04x MID %d @%d msec\n",
 				    id, wmi->mid, tstamp);
@@ -888,13 +890,16 @@ int wmi_call(struct wil6210_priv *wil, u16 cmdid, void *buf, u16 len,
 
 	mutex_lock(&wil->wmi_mutex);
 
+	spin_lock(&wil->wmi_ev_lock);
+	wil->reply_id = reply_id;
+	wil->reply_buf = reply;
+	wil->reply_size = reply_size;
+	spin_unlock(&wil->wmi_ev_lock);
+
 	rc = __wmi_send(wil, cmdid, buf, len);
 	if (rc)
 		goto out;
 
-	wil->reply_id = reply_id;
-	wil->reply_buf = reply;
-	wil->reply_size = reply_size;
 	remain = wait_for_completion_timeout(&wil->wmi_call,
 					     msecs_to_jiffies(to_msec));
 	if (0 == remain) {
@@ -907,10 +912,14 @@ int wmi_call(struct wil6210_priv *wil, u16 cmdid, void *buf, u16 len,
 			    cmdid, reply_id,
 			    to_msec - jiffies_to_msecs(remain));
 	}
+
+out:
+	spin_lock(&wil->wmi_ev_lock);
 	wil->reply_id = 0;
 	wil->reply_buf = NULL;
 	wil->reply_size = 0;
- out:
+	spin_unlock(&wil->wmi_ev_lock);
+
 	mutex_unlock(&wil->wmi_mutex);
 
 	return rc;
-- 
cgit v0.10.2


From ee5dfe0d240e21afe63bdd77bf1d4a386203aa6e Mon Sep 17 00:00:00 2001
From: Hamad Kadmany <qca_hkadmany@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 19:24:05 +0200
Subject: wil6210: TX vring optimization

Tx vring needs to be enlarged to get better
performance for traffic over 2Gbps.

Signed-off-by: Hamad Kadmany <qca_hkadmany@qca.qualcomm.com>
Signed-off-by: Maya Erez <qca_merez@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/wil6210/debugfs.c b/drivers/net/wireless/ath/wil6210/debugfs.c
index 97bc186..98c9148 100644
--- a/drivers/net/wireless/ath/wil6210/debugfs.c
+++ b/drivers/net/wireless/ath/wil6210/debugfs.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2012-2015 Qualcomm Atheros, Inc.
+ * Copyright (c) 2012-2016 Qualcomm Atheros, Inc.
  *
  * Permission to use, copy, modify, and/or distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above
@@ -68,13 +68,13 @@ static void wil_print_vring(struct seq_file *s, struct wil6210_priv *wil,
 		seq_puts(s, "???\n");
 	}
 
-	if (vring->va && (vring->size < 1025)) {
+	if (vring->va && (vring->size <= (1 << WIL_RING_SIZE_ORDER_MAX))) {
 		uint i;
 
 		for (i = 0; i < vring->size; i++) {
 			volatile struct vring_tx_desc *d = &vring->va[i].tx;
 
-			if ((i % 64) == 0 && (i != 0))
+			if ((i % 128) == 0 && (i != 0))
 				seq_puts(s, "\n");
 			seq_printf(s, "%c", (d->dma.status & BIT(0)) ?
 					_s : (vring->ctx[i].skb ? _h : 'h'));
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 9502965..8427d68 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -51,7 +51,7 @@ static inline u32 WIL_GET_BITS(u32 x, int b0, int b1)
 
 #define WIL_TX_Q_LEN_DEFAULT		(4000)
 #define WIL_RX_RING_SIZE_ORDER_DEFAULT	(10)
-#define WIL_TX_RING_SIZE_ORDER_DEFAULT	(10)
+#define WIL_TX_RING_SIZE_ORDER_DEFAULT	(12)
 #define WIL_BCAST_RING_SIZE_ORDER_DEFAULT	(7)
 #define WIL_BCAST_MCS0_LIMIT		(1024) /* limit for MCS0 frame size */
 /* limit ring size in range [32..32k] */
-- 
cgit v0.10.2


From 0b3d76e9f8b4a2b67358b454843ede30555f93a5 Mon Sep 17 00:00:00 2001
From: Peter Oh <poh@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 13:54:07 -0800
Subject: ath10k: rename Mesh related service names

WMI_10_4_SERVICE_MESH bit is for non IEEE802.11s Mesh.
Hence rename it to WMI_10_4_SERVICE_MESH_NON_11S.
Also add _11S as post-fix to each of WMI_SERVICE_MESH and
WMI_VDEV_SUBTYPE_MESH specifying the service is for 11s Mesh.
This will help users to distinguish 11s Mesh from non 11s Mesh.

Signed-off-by: Peter Oh <poh@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index ef0438d..1406b0f 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -4380,8 +4380,8 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		arvif->vdev_type = WMI_VDEV_TYPE_IBSS;
 		break;
 	case NL80211_IFTYPE_MESH_POINT:
-		if (test_bit(WMI_SERVICE_MESH, ar->wmi.svc_map)) {
-			arvif->vdev_subtype = WMI_VDEV_SUBTYPE_MESH;
+		if (test_bit(WMI_SERVICE_MESH_11S, ar->wmi.svc_map)) {
+			arvif->vdev_subtype = WMI_VDEV_SUBTYPE_MESH_11S;
 		} else if (!test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags)) {
 			ret = -EINVAL;
 			ath10k_warn(ar, "must load driver with rawmode=1 to add mesh interfaces\n");
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 806d4bb..0458bf6 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -176,7 +176,8 @@ enum wmi_service {
 	WMI_SERVICE_AUX_CHAN_LOAD_INTF,
 	WMI_SERVICE_BSS_CHANNEL_INFO_64,
 	WMI_SERVICE_EXT_RES_CFG_SUPPORT,
-	WMI_SERVICE_MESH,
+	WMI_SERVICE_MESH_11S,
+	WMI_SERVICE_MESH_NON_11S,
 	WMI_SERVICE_PEER_STATS,
 
 	/* keep last */
@@ -296,7 +297,7 @@ enum wmi_10_4_service {
 	WMI_10_4_SERVICE_AUX_CHAN_LOAD_INTF,
 	WMI_10_4_SERVICE_BSS_CHANNEL_INFO_64,
 	WMI_10_4_SERVICE_EXT_RES_CFG_SUPPORT,
-	WMI_10_4_SERVICE_MESH,
+	WMI_10_4_SERVICE_MESH_NON_11S,
 };
 
 static inline char *wmi_service_name(int service_id)
@@ -387,7 +388,8 @@ static inline char *wmi_service_name(int service_id)
 	SVCSTR(WMI_SERVICE_AUX_CHAN_LOAD_INTF);
 	SVCSTR(WMI_SERVICE_BSS_CHANNEL_INFO_64);
 	SVCSTR(WMI_SERVICE_EXT_RES_CFG_SUPPORT);
-	SVCSTR(WMI_SERVICE_MESH);
+	SVCSTR(WMI_SERVICE_MESH_11S);
+	SVCSTR(WMI_SERVICE_MESH_NON_11S);
 	SVCSTR(WMI_SERVICE_PEER_STATS);
 	default:
 		return NULL;
@@ -463,7 +465,7 @@ static inline void wmi_10x_svc_map(const __le32 *in, unsigned long *out,
 	SVCMAP(WMI_10X_SERVICE_BSS_CHANNEL_INFO_64,
 	       WMI_SERVICE_BSS_CHANNEL_INFO_64, len);
 	SVCMAP(WMI_10X_SERVICE_MESH,
-	       WMI_SERVICE_MESH, len);
+	       WMI_SERVICE_MESH_11S, len);
 	SVCMAP(WMI_10X_SERVICE_EXT_RES_CFG_SUPPORT,
 	       WMI_SERVICE_EXT_RES_CFG_SUPPORT, len);
 	SVCMAP(WMI_10X_SERVICE_PEER_STATS,
@@ -628,8 +630,8 @@ static inline void wmi_10_4_svc_map(const __le32 *in, unsigned long *out,
 	       WMI_SERVICE_BSS_CHANNEL_INFO_64, len);
 	SVCMAP(WMI_10_4_SERVICE_EXT_RES_CFG_SUPPORT,
 	       WMI_SERVICE_EXT_RES_CFG_SUPPORT, len);
-	SVCMAP(WMI_10_4_SERVICE_MESH,
-	       WMI_SERVICE_MESH, len);
+	SVCMAP(WMI_10_4_SERVICE_MESH_NON_11S,
+	       WMI_SERVICE_MESH_NON_11S, len);
 }
 
 #undef SVCMAP
@@ -4287,7 +4289,7 @@ enum wmi_vdev_subtype {
 	WMI_VDEV_SUBTYPE_P2P_CLIENT = 2,
 	WMI_VDEV_SUBTYPE_P2P_GO     = 3,
 	WMI_VDEV_SUBTYPE_PROXY_STA	= 4,
-	WMI_VDEV_SUBTYPE_MESH		= 5,
+	WMI_VDEV_SUBTYPE_MESH_11S   = 5,
 };
 
 /* values for vdev_subtype */
-- 
cgit v0.10.2


From e70e9ba9df8609908c34d042cf980c1a59c56900 Mon Sep 17 00:00:00 2001
From: Peter Oh <poh@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 13:54:08 -0800
Subject: ath10k: update 10.4 WMI service map

Update 10.4 WMI service map to sync to the latest 10.4 firmware
as of 1/20/2016.

Signed-off-by: Peter Oh <poh@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 0458bf6..031018a 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -179,6 +179,7 @@ enum wmi_service {
 	WMI_SERVICE_MESH_11S,
 	WMI_SERVICE_MESH_NON_11S,
 	WMI_SERVICE_PEER_STATS,
+	WMI_SERVICE_RESTRT_CHNL_SUPPORT,
 
 	/* keep last */
 	WMI_SERVICE_MAX,
@@ -298,6 +299,9 @@ enum wmi_10_4_service {
 	WMI_10_4_SERVICE_BSS_CHANNEL_INFO_64,
 	WMI_10_4_SERVICE_EXT_RES_CFG_SUPPORT,
 	WMI_10_4_SERVICE_MESH_NON_11S,
+	WMI_10_4_SERVICE_RESTRT_CHNL_SUPPORT,
+	WMI_10_4_SERVICE_PEER_STATS,
+	WMI_10_4_SERVICE_MESH_11S,
 };
 
 static inline char *wmi_service_name(int service_id)
@@ -391,6 +395,7 @@ static inline char *wmi_service_name(int service_id)
 	SVCSTR(WMI_SERVICE_MESH_11S);
 	SVCSTR(WMI_SERVICE_MESH_NON_11S);
 	SVCSTR(WMI_SERVICE_PEER_STATS);
+	SVCSTR(WMI_SERVICE_RESTRT_CHNL_SUPPORT);
 	default:
 		return NULL;
 	}
@@ -632,6 +637,12 @@ static inline void wmi_10_4_svc_map(const __le32 *in, unsigned long *out,
 	       WMI_SERVICE_EXT_RES_CFG_SUPPORT, len);
 	SVCMAP(WMI_10_4_SERVICE_MESH_NON_11S,
 	       WMI_SERVICE_MESH_NON_11S, len);
+	SVCMAP(WMI_10_4_SERVICE_RESTRT_CHNL_SUPPORT,
+	       WMI_SERVICE_RESTRT_CHNL_SUPPORT, len);
+	SVCMAP(WMI_10_4_SERVICE_PEER_STATS,
+	       WMI_SERVICE_PEER_STATS, len);
+	SVCMAP(WMI_10_4_SERVICE_MESH_11S,
+	       WMI_SERVICE_MESH_11S, len);
 }
 
 #undef SVCMAP
-- 
cgit v0.10.2


From 08c27be1db84dcdad14e86ec76d9c9915deb14eb Mon Sep 17 00:00:00 2001
From: Peter Oh <poh@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 13:54:09 -0800
Subject: ath10k: use vif->type and vif->p2p for P2P_GO check

Interface type P2P_GO can be checked by either arvif->vdev_type
and arvif->vdev_subtype or vif->type and vif->p2p.
Use later one to avoid more cpu consumption that could happen
when subtype abstraction layer change is introduced.

Signed-off-by: Peter Oh <poh@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 1406b0f..addf3cb 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -1358,10 +1358,7 @@ static int ath10k_mac_setup_bcn_p2p_ie(struct ath10k_vif *arvif,
 	const u8 *p2p_ie;
 	int ret;
 
-	if (arvif->vdev_type != WMI_VDEV_TYPE_AP)
-		return 0;
-
-	if (arvif->vdev_subtype != WMI_VDEV_SUBTYPE_P2P_GO)
+	if (arvif->vif->type != NL80211_IFTYPE_AP || !arvif->vif->p2p)
 		return 0;
 
 	mgmt = (void *)bcn->data;
@@ -3259,8 +3256,7 @@ static void ath10k_tx_h_add_p2p_noa_ie(struct ath10k *ar,
 	struct ath10k_vif *arvif = ath10k_vif_to_arvif(vif);
 
 	/* This is case only for P2P_GO */
-	if (arvif->vdev_type != WMI_VDEV_TYPE_AP ||
-	    arvif->vdev_subtype != WMI_VDEV_SUBTYPE_P2P_GO)
+	if (vif->type != NL80211_IFTYPE_AP || !vif->p2p)
 		return;
 
 	if (unlikely(ieee80211_is_probe_resp(hdr->frame_control))) {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 0f01a8d..5158716 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -3196,7 +3196,7 @@ static void ath10k_wmi_update_noa(struct ath10k *ar, struct ath10k_vif *arvif,
 				  struct sk_buff *bcn,
 				  const struct wmi_p2p_noa_info *noa)
 {
-	if (arvif->vdev_subtype != WMI_VDEV_SUBTYPE_P2P_GO)
+	if (!arvif->vif->p2p)
 		return;
 
 	ath10k_dbg(ar, ATH10K_DBG_MGMT, "noa changed: %d\n", noa->changed);
-- 
cgit v0.10.2


From 6e4de1a49aa3254f46f66db5aac530707c193cde Mon Sep 17 00:00:00 2001
From: Peter Oh <poh@qca.qualcomm.com>
Date: Thu, 28 Jan 2016 13:54:10 -0800
Subject: ath10k: add abstraction layer for vdev subtype

Abstraction layer for vdev subtype is added to solve
subtype mismatch and to give flexible compatibility
among different firmware revisions.

For instance, 10.2 and 10.4 firmware has different
definition of their vdev subtypes for Mesh.
10.4 defined subtype 6 for 802.11s Mesh while 10.2 uses 5.
Hence use the abstraction API to get right subtype to use.

Signed-off-by: Peter Oh <poh@qca.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index addf3cb..96e2f8a 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -4359,25 +4359,29 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		   bit, ar->free_vdev_map);
 
 	arvif->vdev_id = bit;
-	arvif->vdev_subtype = WMI_VDEV_SUBTYPE_NONE;
+	arvif->vdev_subtype =
+		ath10k_wmi_get_vdev_subtype(ar, WMI_VDEV_SUBTYPE_NONE);
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_P2P_DEVICE:
 		arvif->vdev_type = WMI_VDEV_TYPE_STA;
-		arvif->vdev_subtype = WMI_VDEV_SUBTYPE_P2P_DEVICE;
+		arvif->vdev_subtype = ath10k_wmi_get_vdev_subtype
+					(ar, WMI_VDEV_SUBTYPE_P2P_DEVICE);
 		break;
 	case NL80211_IFTYPE_UNSPECIFIED:
 	case NL80211_IFTYPE_STATION:
 		arvif->vdev_type = WMI_VDEV_TYPE_STA;
 		if (vif->p2p)
-			arvif->vdev_subtype = WMI_VDEV_SUBTYPE_P2P_CLIENT;
+			arvif->vdev_subtype = ath10k_wmi_get_vdev_subtype
+					(ar, WMI_VDEV_SUBTYPE_P2P_CLIENT);
 		break;
 	case NL80211_IFTYPE_ADHOC:
 		arvif->vdev_type = WMI_VDEV_TYPE_IBSS;
 		break;
 	case NL80211_IFTYPE_MESH_POINT:
 		if (test_bit(WMI_SERVICE_MESH_11S, ar->wmi.svc_map)) {
-			arvif->vdev_subtype = WMI_VDEV_SUBTYPE_MESH_11S;
+			arvif->vdev_subtype = ath10k_wmi_get_vdev_subtype
+						(ar, WMI_VDEV_SUBTYPE_MESH_11S);
 		} else if (!test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags)) {
 			ret = -EINVAL;
 			ath10k_warn(ar, "must load driver with rawmode=1 to add mesh interfaces\n");
@@ -4389,7 +4393,8 @@ static int ath10k_add_interface(struct ieee80211_hw *hw,
 		arvif->vdev_type = WMI_VDEV_TYPE_AP;
 
 		if (vif->p2p)
-			arvif->vdev_subtype = WMI_VDEV_SUBTYPE_P2P_GO;
+			arvif->vdev_subtype = ath10k_wmi_get_vdev_subtype
+						(ar, WMI_VDEV_SUBTYPE_P2P_GO);
 		break;
 	case NL80211_IFTYPE_MONITOR:
 		arvif->vdev_type = WMI_VDEV_TYPE_MONITOR;
diff --git a/drivers/net/wireless/ath/ath10k/wmi-ops.h b/drivers/net/wireless/ath/ath10k/wmi-ops.h
index 8f4f6a8..32ab34e 100644
--- a/drivers/net/wireless/ath/ath10k/wmi-ops.h
+++ b/drivers/net/wireless/ath/ath10k/wmi-ops.h
@@ -186,6 +186,8 @@ struct wmi_ops {
 							u8 enable,
 							u32 detect_level,
 							u32 detect_margin);
+	int (*get_vdev_subtype)(struct ath10k *ar,
+				enum wmi_vdev_subtype subtype);
 };
 
 int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb, u32 cmd_id);
@@ -1327,4 +1329,13 @@ ath10k_wmi_pdev_enable_adaptive_cca(struct ath10k *ar, u8 enable,
 				   ar->wmi.cmd->pdev_enable_adaptive_cca_cmdid);
 }
 
+static inline int
+ath10k_wmi_get_vdev_subtype(struct ath10k *ar, enum wmi_vdev_subtype subtype)
+{
+	if (!ar->wmi.ops->get_vdev_subtype)
+		return -EOPNOTSUPP;
+
+	return ar->wmi.ops->get_vdev_subtype(ar, subtype);
+}
+
 #endif
diff --git a/drivers/net/wireless/ath/ath10k/wmi-tlv.c b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
index 3b3a27b..1085932 100644
--- a/drivers/net/wireless/ath/ath10k/wmi-tlv.c
+++ b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
@@ -3483,6 +3483,7 @@ static const struct wmi_ops wmi_tlv_ops = {
 	.gen_tdls_peer_update = ath10k_wmi_tlv_op_gen_tdls_peer_update,
 	.gen_adaptive_qcs = ath10k_wmi_tlv_op_gen_adaptive_qcs,
 	.fw_stats_fill = ath10k_wmi_main_op_fw_stats_fill,
+	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
 };
 
 static const struct wmi_peer_flags_map wmi_tlv_peer_flags_map = {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 5158716..04799c7 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -7413,6 +7413,71 @@ unlock:
 		buf[len] = 0;
 }
 
+int ath10k_wmi_op_get_vdev_subtype(struct ath10k *ar,
+				   enum wmi_vdev_subtype subtype)
+{
+	switch (subtype) {
+	case WMI_VDEV_SUBTYPE_NONE:
+		return WMI_VDEV_SUBTYPE_LEGACY_NONE;
+	case WMI_VDEV_SUBTYPE_P2P_DEVICE:
+		return WMI_VDEV_SUBTYPE_LEGACY_P2P_DEV;
+	case WMI_VDEV_SUBTYPE_P2P_CLIENT:
+		return WMI_VDEV_SUBTYPE_LEGACY_P2P_CLI;
+	case WMI_VDEV_SUBTYPE_P2P_GO:
+		return WMI_VDEV_SUBTYPE_LEGACY_P2P_GO;
+	case WMI_VDEV_SUBTYPE_PROXY_STA:
+		return WMI_VDEV_SUBTYPE_LEGACY_PROXY_STA;
+	case WMI_VDEV_SUBTYPE_MESH_11S:
+	case WMI_VDEV_SUBTYPE_MESH_NON_11S:
+		return -ENOTSUPP;
+	}
+	return -ENOTSUPP;
+}
+
+static int ath10k_wmi_10_2_4_op_get_vdev_subtype(struct ath10k *ar,
+						 enum wmi_vdev_subtype subtype)
+{
+	switch (subtype) {
+	case WMI_VDEV_SUBTYPE_NONE:
+		return WMI_VDEV_SUBTYPE_10_2_4_NONE;
+	case WMI_VDEV_SUBTYPE_P2P_DEVICE:
+		return WMI_VDEV_SUBTYPE_10_2_4_P2P_DEV;
+	case WMI_VDEV_SUBTYPE_P2P_CLIENT:
+		return WMI_VDEV_SUBTYPE_10_2_4_P2P_CLI;
+	case WMI_VDEV_SUBTYPE_P2P_GO:
+		return WMI_VDEV_SUBTYPE_10_2_4_P2P_GO;
+	case WMI_VDEV_SUBTYPE_PROXY_STA:
+		return WMI_VDEV_SUBTYPE_10_2_4_PROXY_STA;
+	case WMI_VDEV_SUBTYPE_MESH_11S:
+		return WMI_VDEV_SUBTYPE_10_2_4_MESH_11S;
+	case WMI_VDEV_SUBTYPE_MESH_NON_11S:
+		return -ENOTSUPP;
+	}
+	return -ENOTSUPP;
+}
+
+static int ath10k_wmi_10_4_op_get_vdev_subtype(struct ath10k *ar,
+					       enum wmi_vdev_subtype subtype)
+{
+	switch (subtype) {
+	case WMI_VDEV_SUBTYPE_NONE:
+		return WMI_VDEV_SUBTYPE_10_4_NONE;
+	case WMI_VDEV_SUBTYPE_P2P_DEVICE:
+		return WMI_VDEV_SUBTYPE_10_4_P2P_DEV;
+	case WMI_VDEV_SUBTYPE_P2P_CLIENT:
+		return WMI_VDEV_SUBTYPE_10_4_P2P_CLI;
+	case WMI_VDEV_SUBTYPE_P2P_GO:
+		return WMI_VDEV_SUBTYPE_10_4_P2P_GO;
+	case WMI_VDEV_SUBTYPE_PROXY_STA:
+		return WMI_VDEV_SUBTYPE_10_4_PROXY_STA;
+	case WMI_VDEV_SUBTYPE_MESH_11S:
+		return WMI_VDEV_SUBTYPE_10_4_MESH_11S;
+	case WMI_VDEV_SUBTYPE_MESH_NON_11S:
+		return WMI_VDEV_SUBTYPE_10_4_MESH_NON_11S;
+	}
+	return -ENOTSUPP;
+}
+
 static const struct wmi_ops wmi_ops = {
 	.rx = ath10k_wmi_op_rx,
 	.map_svc = wmi_main_svc_map,
@@ -7472,6 +7537,7 @@ static const struct wmi_ops wmi_ops = {
 	.gen_addba_set_resp = ath10k_wmi_op_gen_addba_set_resp,
 	.gen_delba_send = ath10k_wmi_op_gen_delba_send,
 	.fw_stats_fill = ath10k_wmi_main_op_fw_stats_fill,
+	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
 	/* .gen_bcn_tmpl not implemented */
 	/* .gen_prb_tmpl not implemented */
 	/* .gen_p2p_go_bcn_ie not implemented */
@@ -7539,6 +7605,7 @@ static const struct wmi_ops wmi_10_1_ops = {
 	.gen_addba_set_resp = ath10k_wmi_op_gen_addba_set_resp,
 	.gen_delba_send = ath10k_wmi_op_gen_delba_send,
 	.fw_stats_fill = ath10k_wmi_10x_op_fw_stats_fill,
+	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
 	/* .gen_bcn_tmpl not implemented */
 	/* .gen_prb_tmpl not implemented */
 	/* .gen_p2p_go_bcn_ie not implemented */
@@ -7607,6 +7674,7 @@ static const struct wmi_ops wmi_10_2_ops = {
 	.gen_addba_set_resp = ath10k_wmi_op_gen_addba_set_resp,
 	.gen_delba_send = ath10k_wmi_op_gen_delba_send,
 	.fw_stats_fill = ath10k_wmi_10x_op_fw_stats_fill,
+	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
 	/* .gen_pdev_enable_adaptive_cca not implemented */
 };
 
@@ -7673,6 +7741,7 @@ static const struct wmi_ops wmi_10_2_4_ops = {
 	.fw_stats_fill = ath10k_wmi_10x_op_fw_stats_fill,
 	.gen_pdev_enable_adaptive_cca =
 		ath10k_wmi_op_gen_pdev_enable_adaptive_cca,
+	.get_vdev_subtype = ath10k_wmi_10_2_4_op_get_vdev_subtype,
 	/* .gen_bcn_tmpl not implemented */
 	/* .gen_prb_tmpl not implemented */
 	/* .gen_p2p_go_bcn_ie not implemented */
@@ -7739,6 +7808,7 @@ static const struct wmi_ops wmi_10_4_ops = {
 	/* shared with 10.2 */
 	.gen_request_stats = ath10k_wmi_op_gen_request_stats,
 	.gen_pdev_get_temperature = ath10k_wmi_10_2_op_gen_pdev_get_temperature,
+	.get_vdev_subtype = ath10k_wmi_10_4_op_get_vdev_subtype,
 };
 
 int ath10k_wmi_attach(struct ath10k *ar)
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 031018a..049c618 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -4295,12 +4295,40 @@ enum wmi_vdev_type {
 };
 
 enum wmi_vdev_subtype {
-	WMI_VDEV_SUBTYPE_NONE       = 0,
-	WMI_VDEV_SUBTYPE_P2P_DEVICE = 1,
-	WMI_VDEV_SUBTYPE_P2P_CLIENT = 2,
-	WMI_VDEV_SUBTYPE_P2P_GO     = 3,
-	WMI_VDEV_SUBTYPE_PROXY_STA	= 4,
-	WMI_VDEV_SUBTYPE_MESH_11S   = 5,
+	WMI_VDEV_SUBTYPE_NONE,
+	WMI_VDEV_SUBTYPE_P2P_DEVICE,
+	WMI_VDEV_SUBTYPE_P2P_CLIENT,
+	WMI_VDEV_SUBTYPE_P2P_GO,
+	WMI_VDEV_SUBTYPE_PROXY_STA,
+	WMI_VDEV_SUBTYPE_MESH_11S,
+	WMI_VDEV_SUBTYPE_MESH_NON_11S,
+};
+
+enum wmi_vdev_subtype_legacy {
+	WMI_VDEV_SUBTYPE_LEGACY_NONE      = 0,
+	WMI_VDEV_SUBTYPE_LEGACY_P2P_DEV   = 1,
+	WMI_VDEV_SUBTYPE_LEGACY_P2P_CLI   = 2,
+	WMI_VDEV_SUBTYPE_LEGACY_P2P_GO    = 3,
+	WMI_VDEV_SUBTYPE_LEGACY_PROXY_STA = 4,
+};
+
+enum wmi_vdev_subtype_10_2_4 {
+	WMI_VDEV_SUBTYPE_10_2_4_NONE      = 0,
+	WMI_VDEV_SUBTYPE_10_2_4_P2P_DEV   = 1,
+	WMI_VDEV_SUBTYPE_10_2_4_P2P_CLI   = 2,
+	WMI_VDEV_SUBTYPE_10_2_4_P2P_GO    = 3,
+	WMI_VDEV_SUBTYPE_10_2_4_PROXY_STA = 4,
+	WMI_VDEV_SUBTYPE_10_2_4_MESH_11S  = 5,
+};
+
+enum wmi_vdev_subtype_10_4 {
+	WMI_VDEV_SUBTYPE_10_4_NONE         = 0,
+	WMI_VDEV_SUBTYPE_10_4_P2P_DEV      = 1,
+	WMI_VDEV_SUBTYPE_10_4_P2P_CLI      = 2,
+	WMI_VDEV_SUBTYPE_10_4_P2P_GO       = 3,
+	WMI_VDEV_SUBTYPE_10_4_PROXY_STA    = 4,
+	WMI_VDEV_SUBTYPE_10_4_MESH_NON_11S = 5,
+	WMI_VDEV_SUBTYPE_10_4_MESH_11S     = 6,
 };
 
 /* values for vdev_subtype */
@@ -6471,5 +6499,7 @@ size_t ath10k_wmi_fw_stats_num_vdevs(struct list_head *head);
 void ath10k_wmi_10_4_op_fw_stats_fill(struct ath10k *ar,
 				      struct ath10k_fw_stats *fw_stats,
 				      char *buf);
+int ath10k_wmi_op_get_vdev_subtype(struct ath10k *ar,
+				   enum wmi_vdev_subtype subtype);
 
 #endif /* _WMI_H_ */
-- 
cgit v0.10.2


From b1a958c933c5b0f967844ec05e76ec2985450782 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@codeaurora.org>
Date: Fri, 29 Jan 2016 16:35:11 +0530
Subject: ath10k: add qca4019 hw params

Add a new entry in hw_params_list for qca4019 with list of
it's own details.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 1f4a278..7904387 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -201,6 +201,26 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 			.board_ext_size = QCA9377_BOARD_EXT_DATA_SZ,
 		},
 	},
+	{
+		.id = QCA4019_HW_1_0_DEV_VERSION,
+		.dev_id = 0,
+		.name = "qca4019 hw1.0",
+		.patch_load_addr = QCA4019_HW_1_0_PATCH_LOAD_ADDR,
+		.uart_pin = 7,
+		.otp_exe_param = 0x0010000,
+		.continuous_frag_desc = true,
+		.channel_counters_freq_hz = 125000,
+		.max_probe_resp_desc_thres = 24,
+		.hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE,
+		.fw = {
+			.dir = QCA4019_HW_1_0_FW_DIR,
+			.fw = QCA4019_HW_1_0_FW_FILE,
+			.otp = QCA4019_HW_1_0_OTP_FILE,
+			.board = QCA4019_HW_1_0_BOARD_DATA_FILE,
+			.board_size = QCA4019_BOARD_DATA_SZ,
+			.board_ext_size = QCA4019_BOARD_EXT_DATA_SZ,
+		},
+	},
 };
 
 static const char *const ath10k_core_fw_feature_str[] = {
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 4217bdb..291ca1f 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -106,6 +106,14 @@ enum qca9377_chip_id_rev {
 #define QCA9377_HW_1_0_BOARD_DATA_FILE "board.bin"
 #define QCA9377_HW_1_0_PATCH_LOAD_ADDR	0x1234
 
+/* QCA4019 1.0 definitions */
+#define QCA4019_HW_1_0_DEV_VERSION     0x01000000
+#define QCA4019_HW_1_0_FW_DIR          ATH10K_FW_DIR "/QCA4019/hw1.0"
+#define QCA4019_HW_1_0_FW_FILE         "firmware.bin"
+#define QCA4019_HW_1_0_OTP_FILE        "otp.bin"
+#define QCA4019_HW_1_0_BOARD_DATA_FILE "board.bin"
+#define QCA4019_HW_1_0_PATCH_LOAD_ADDR  0x1234
+
 #define ATH10K_FW_API2_FILE		"firmware-2.bin"
 #define ATH10K_FW_API3_FILE		"firmware-3.bin"
 
diff --git a/drivers/net/wireless/ath/ath10k/targaddrs.h b/drivers/net/wireless/ath/ath10k/targaddrs.h
index 05a421b..361f143 100644
--- a/drivers/net/wireless/ath/ath10k/targaddrs.h
+++ b/drivers/net/wireless/ath/ath10k/targaddrs.h
@@ -456,4 +456,7 @@ Fw Mode/SubMode Mask
 #define QCA99X0_BOARD_DATA_SZ	  12288
 #define QCA99X0_BOARD_EXT_DATA_SZ 0
 
+#define QCA4019_BOARD_DATA_SZ	  12064
+#define QCA4019_BOARD_EXT_DATA_SZ 0
+
 #endif /* __TARGADDRS_H__ */
-- 
cgit v0.10.2


From 5699a6f238ebfc8a6be3963a30d86d6f1e8aaaf9 Mon Sep 17 00:00:00 2001
From: Raja Mani <rmani@qti.qualcomm.com>
Date: Fri, 29 Jan 2016 16:35:12 +0530
Subject: ath10k: populate qca4019 fw specific wmi init params

Some of the parameter like tx/rx chain mask, number of htt tx desc,
qcache active peer count, etc goes via wmi init cmd to qca4019 firmware
are different.

To make use of 10.4 gen_init function for qca4019, change wmi service
ready handler and 10.4 wmi init functions to adapt qca4019 specific
init values.

Signed-off-by: Raja Mani <rmani@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 7904387..4d31764 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -156,6 +156,11 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.channel_counters_freq_hz = 150000,
 		.max_probe_resp_desc_thres = 24,
 		.hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE,
+		.num_msdu_desc = 1424,
+		.qcache_active_peers = 50,
+		.tx_chain_mask = 0xf,
+		.rx_chain_mask = 0xf,
+		.max_spatial_stream = 4,
 		.fw = {
 			.dir = QCA99X0_HW_2_0_FW_DIR,
 			.fw = QCA99X0_HW_2_0_FW_FILE,
@@ -212,6 +217,11 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.channel_counters_freq_hz = 125000,
 		.max_probe_resp_desc_thres = 24,
 		.hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE,
+		.num_msdu_desc = 2500,
+		.qcache_active_peers = 35,
+		.tx_chain_mask = 0x3,
+		.rx_chain_mask = 0x3,
+		.max_spatial_stream = 2,
 		.fw = {
 			.dir = QCA4019_HW_1_0_FW_DIR,
 			.fw = QCA4019_HW_1_0_FW_FILE,
@@ -1523,9 +1533,9 @@ static int ath10k_core_init_firmware_features(struct ath10k *ar)
 		ar->num_active_peers = TARGET_10_4_ACTIVE_PEERS;
 		ar->max_num_vdevs = TARGET_10_4_NUM_VDEVS;
 		ar->num_tids = TARGET_10_4_TGT_NUM_TIDS;
-		ar->htt.max_num_pending_tx = TARGET_10_4_NUM_MSDU_DESC;
+		ar->htt.max_num_pending_tx = ar->hw_params.num_msdu_desc;
 		ar->fw_stats_req_mask = WMI_STAT_PEER;
-		ar->max_spatial_stream = WMI_10_4_MAX_SPATIAL_STREAM;
+		ar->max_spatial_stream = ar->hw_params.max_spatial_stream;
 		break;
 	case ATH10K_FW_WMI_OP_VERSION_UNSET:
 	case ATH10K_FW_WMI_OP_VERSION_MAX:
diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 2e411b5..a62b62a 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -680,6 +680,12 @@ struct ath10k {
 		/* The padding bytes's location is different on various chips */
 		enum ath10k_hw_4addr_pad hw_4addr_pad;
 
+		u32 num_msdu_desc;
+		u32 qcache_active_peers;
+		u32 tx_chain_mask;
+		u32 rx_chain_mask;
+		u32 max_spatial_stream;
+
 		struct ath10k_hw_params_fw {
 			const char *dir;
 			const char *fw;
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 291ca1f..2dece8d 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -426,16 +426,11 @@ enum ath10k_hw_4addr_pad {
 #define TARGET_10_4_ACTIVE_PEERS		0
 
 #define TARGET_10_4_NUM_QCACHE_PEERS_MAX	512
-#define TARGET_10_4_QCACHE_ACTIVE_PEERS		50
 #define TARGET_10_4_NUM_OFFLOAD_PEERS		0
 #define TARGET_10_4_NUM_OFFLOAD_REORDER_BUFFS	0
 #define TARGET_10_4_NUM_PEER_KEYS		2
 #define TARGET_10_4_TGT_NUM_TIDS		((TARGET_10_4_NUM_PEERS) * 2)
 #define TARGET_10_4_AST_SKID_LIMIT		32
-#define TARGET_10_4_TX_CHAIN_MASK		(BIT(0) | BIT(1) | \
-						 BIT(2) | BIT(3))
-#define TARGET_10_4_RX_CHAIN_MASK		(BIT(0) | BIT(1) | \
-						 BIT(2) | BIT(3))
 
 /* 100 ms for video, best-effort, and background */
 #define TARGET_10_4_RX_TIMEOUT_LO_PRI		100
@@ -461,7 +456,6 @@ enum ath10k_hw_4addr_pad {
 #define TARGET_10_4_RX_SKIP_DEFRAG_TIMEOUT_DUP_DETECTION_CHECK 1
 #define TARGET_10_4_VOW_CONFIG			0
 #define TARGET_10_4_GTK_OFFLOAD_MAX_VDEV	3
-#define TARGET_10_4_NUM_MSDU_DESC		(1024 + 400)
 #define TARGET_10_4_11AC_TX_MAX_FRAGS		2
 #define TARGET_10_4_MAX_PEER_EXT_STATS		16
 #define TARGET_10_4_SMART_ANT_CAP		0
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 04799c7..1ce6742 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -4618,9 +4618,9 @@ static void ath10k_wmi_event_service_ready_work(struct work_struct *work)
 
 	if (test_bit(WMI_SERVICE_PEER_CACHING, ar->wmi.svc_map)) {
 		ar->max_num_peers = TARGET_10_4_NUM_QCACHE_PEERS_MAX +
-				    TARGET_10_4_NUM_VDEVS;
-		ar->num_active_peers = TARGET_10_4_QCACHE_ACTIVE_PEERS +
-				       TARGET_10_4_NUM_VDEVS;
+				    ar->max_num_vdevs;
+		ar->num_active_peers = ar->hw_params.qcache_active_peers +
+				       ar->max_num_vdevs;
 		ar->num_tids = ar->num_active_peers * 2;
 		ar->max_num_stations = TARGET_10_4_NUM_QCACHE_PEERS_MAX;
 	}
@@ -5602,8 +5602,8 @@ static struct sk_buff *ath10k_wmi_10_4_op_gen_init(struct ath10k *ar)
 			__cpu_to_le32(TARGET_10_4_NUM_OFFLOAD_REORDER_BUFFS);
 	config.num_peer_keys  = __cpu_to_le32(TARGET_10_4_NUM_PEER_KEYS);
 	config.ast_skid_limit = __cpu_to_le32(TARGET_10_4_AST_SKID_LIMIT);
-	config.tx_chain_mask  = __cpu_to_le32(TARGET_10_4_TX_CHAIN_MASK);
-	config.rx_chain_mask  = __cpu_to_le32(TARGET_10_4_RX_CHAIN_MASK);
+	config.tx_chain_mask  = __cpu_to_le32(ar->hw_params.tx_chain_mask);
+	config.rx_chain_mask  = __cpu_to_le32(ar->hw_params.rx_chain_mask);
 
 	config.rx_timeout_pri[0] = __cpu_to_le32(TARGET_10_4_RX_TIMEOUT_LO_PRI);
 	config.rx_timeout_pri[1] = __cpu_to_le32(TARGET_10_4_RX_TIMEOUT_LO_PRI);
@@ -5634,7 +5634,7 @@ static struct sk_buff *ath10k_wmi_10_4_op_gen_init(struct ath10k *ar)
 	config.vow_config = __cpu_to_le32(TARGET_10_4_VOW_CONFIG);
 	config.gtk_offload_max_vdev =
 			__cpu_to_le32(TARGET_10_4_GTK_OFFLOAD_MAX_VDEV);
-	config.num_msdu_desc = __cpu_to_le32(TARGET_10_4_NUM_MSDU_DESC);
+	config.num_msdu_desc = __cpu_to_le32(ar->htt.max_num_pending_tx);
 	config.max_frag_entries = __cpu_to_le32(TARGET_10_4_11AC_TX_MAX_FRAGS);
 	config.max_peer_ext_stats =
 			__cpu_to_le32(TARGET_10_4_MAX_PEER_EXT_STATS);
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 049c618..4d3cbc4 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -1818,7 +1818,6 @@ enum wmi_channel_change_cause {
 #define WMI_CHANNEL_CHANGE_CAUSE_CSA (1 << 13)
 
 #define WMI_MAX_SPATIAL_STREAM        3 /* default max ss */
-#define WMI_10_4_MAX_SPATIAL_STREAM   4
 
 /* HT Capabilities*/
 #define WMI_HT_CAP_ENABLED                0x0001   /* HT Enabled/ disabled */
-- 
cgit v0.10.2


From 7b5e73961915438d623cb80eea97868b3a66cc48 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 6 Dec 2015 16:15:07 +0100
Subject: batman-adv: Switch to HTTPS version of links

open-mesh.org and its subdomains can only be accessed via HTTPS. HTTP-only
requests are currently redirected automatically to HTTPS but references in
the source code should be only https.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/Documentation/networking/batman-adv.txt b/Documentation/networking/batman-adv.txt
index ff23b75..1b5e7a7 100644
--- a/Documentation/networking/batman-adv.txt
+++ b/Documentation/networking/batman-adv.txt
@@ -187,7 +187,7 @@ interfaces to the kernel module settings.
 
 For more information, please see the manpage (man batctl).
 
-batctl is available on http://www.open-mesh.org/
+batctl is available on https://www.open-mesh.org/
 
 
 CONTACT
diff --git a/MAINTAINERS b/MAINTAINERS
index f678c37..1e216aa 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2147,7 +2147,7 @@ M:	Marek Lindner <mareklindner@neomailbox.ch>
 M:	Simon Wunderlich <sw@simonwunderlich.de>
 M:	Antonio Quartulli <a@unstable.cc>
 L:	b.a.t.m.a.n@lists.open-mesh.org
-W:	http://www.open-mesh.org/
+W:	https://www.open-mesh.org/
 S:	Maintained
 F:	net/batman-adv/
 
diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index c6fc8f7..2dd40e5 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -12,7 +12,7 @@ config BATMAN_ADV
           B.A.T.M.A.N. (better approach to mobile ad-hoc networking) is
           a routing protocol for multi-hop ad-hoc mesh networks. The
           networks may be wired or wireless. See
-          http://www.open-mesh.org/ for more information and user space
+          https://www.open-mesh.org/ for more information and user space
           tools.
 
 config BATMAN_ADV_BLA
-- 
cgit v0.10.2


From 04d5a21d62887b9bc5383fc68ab2756287da4532 Mon Sep 17 00:00:00 2001
From: Kiran Patil <kiran.patil@intel.com>
Date: Wed, 9 Dec 2015 15:50:23 -0800
Subject: i40e: Add mac_filter_element at the end of the list instead of HEAD

Add MAC filter element to the end of the list in the given order,
just to be tidy, and just in case there are ever any ordering issues in
the future.

Change-ID: Idc15276147593ea9393ac72c861f9c7905a791b4
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 8f3b53e..d078a63 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1368,7 +1368,7 @@ struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi,
 		f->changed = true;
 
 		INIT_LIST_HEAD(&f->list);
-		list_add(&f->list, &vsi->mac_filter_list);
+		list_add_tail(&f->list, &vsi->mac_filter_list);
 	}
 
 	/* increment counter and add a new flag if needed */
-- 
cgit v0.10.2


From 6e35c04cf633e55648acb9ccabff42aa37bd4044 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Wed, 9 Dec 2015 15:50:24 -0800
Subject: i40e/i40evf: Fix RSS rx-flow-hash configuration through ethtool

This patch fixes the Hash PCTYPE enable for X722 since it supports
a broader selection of PCTYPES for TCP and UDP.

This patch also fixes a bug in XL710, X710, X722 support for RSS,
as of now we cannot reduce the (4)tuple for RSS for TCP/IPv4/IPV6 or
UDP/IPv4/IPv6 packets since this requires a product feature change
that comes in a later release.

A VF should never be allowed to change the tuples for RSS for any
PCTYPE since that's a global setting for the device in case of i40e
devices.

Change-ID: I0ee7203c9b24813260f58f3220798bc9d9ac4a12
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 29d5833..c8b9dca 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2166,8 +2166,7 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 	case TCP_V4_FLOW:
 		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 		case 0:
-			hena &= ~BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
-			break;
+			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
 			break;
@@ -2178,8 +2177,7 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 	case TCP_V6_FLOW:
 		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 		case 0:
-			hena &= ~BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
-			break;
+			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
 			break;
@@ -2190,9 +2188,7 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 	case UDP_V4_FLOW:
 		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 		case 0:
-			hena &= ~(BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
-				  BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
-			break;
+			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
@@ -2204,9 +2200,7 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 	case UDP_V6_FLOW:
 		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 		case 0:
-			hena &= ~(BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
-				  BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
-			break;
+			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
index a4c9feb..8906785 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
@@ -477,54 +477,30 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 
 	switch (nfc->flow_type) {
 	case TCP_V4_FLOW:
-		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
-		case 0:
-			hena &= ~BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
-			break;
-		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3))
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
-			break;
-		default:
+		else
 			return -EINVAL;
-		}
 		break;
 	case TCP_V6_FLOW:
-		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
-		case 0:
-			hena &= ~BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
-			break;
-		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3))
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
-			break;
-		default:
+		else
 			return -EINVAL;
-		}
 		break;
 	case UDP_V4_FLOW:
-		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
-		case 0:
-			hena &= ~(BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
-				  BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
-			break;
-		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
-			break;
-		default:
+		} else {
 			return -EINVAL;
 		}
 		break;
 	case UDP_V6_FLOW:
-		switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
-		case 0:
-			hena &= ~(BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
-				  BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
-			break;
-		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
-			break;
-		default:
+		} else {
 			return -EINVAL;
 		}
 		break;
-- 
cgit v0.10.2


From 48b1804ee3cdad7bf115666eb35edf12a734710f Mon Sep 17 00:00:00 2001
From: Catherine Sullivan <catherine.sullivan@intel.com>
Date: Wed, 9 Dec 2015 15:50:25 -0800
Subject: i40e: Replace X722 mac check in ethtool get_settings

100M SGMII is only supported on X722.  Replace the mac check with
a feature flag check that is only set for the X722 device.

Change-ID: I53452d9af6af8cd9dca8500215fbc6ce93418f52
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 68f2204..47f6c0a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -339,6 +339,7 @@ struct i40e_pf {
 #define I40E_FLAG_VEB_MODE_ENABLED		BIT_ULL(40)
 #define I40E_FLAG_GENEVE_OFFLOAD_CAPABLE	BIT_ULL(41)
 #define I40E_FLAG_NO_PCI_LINK_CHECK		BIT_ULL(42)
+#define I40E_FLAG_100M_SGMII_CAPABLE		BIT_ULL(43)
 #define I40E_FLAG_PF_MAC			BIT_ULL(50)
 
 	/* tracks features that get auto disabled by errors */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index c8b9dca..252a9dd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -340,7 +340,7 @@ static void i40e_get_settings_link_up(struct i40e_hw *hw,
 				  SUPPORTED_1000baseT_Full;
 		if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB)
 			ecmd->advertising |= ADVERTISED_1000baseT_Full;
-		if (pf->hw.mac.type == I40E_MAC_X722) {
+		if (pf->flags & I40E_FLAG_100M_SGMII_CAPABLE) {
 			ecmd->supported |= SUPPORTED_100baseT_Full;
 			if (hw_link_info->requested_speeds &
 			    I40E_LINK_SPEED_100MB)
-- 
cgit v0.10.2


From 209dc4daf23f92b3e0bc6d602411506c4083e421 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 9 Dec 2015 15:50:27 -0800
Subject: i40evf: allow channel bonding of VFs

In some modes, bonding would not enslave VF interfaces. This is due to
bonding calling change_mtu and the immediately calling open. Because of
the asynchronous nature of the admin queue mechanism, the VF returns
-EBUSY to the open call, because it knows the previous operation hasn't
finished yet. This causes bonding to fail with a less-than-useful error
message.

To fix this, remove the check for pending operations at the beginning of
open. But this introduces a new bug where the driver will panic on a
quick close/open cycle. To fix that, we add a new driver state,
__I40EVF_DOWN_PENDING, that the driver enters when down is called. The
driver finally transitions to a fully DOWN state when it receives
confirmation from the PF driver that all the queues are disabled. This
allows open to complete even if there is a pending mtu change, and
bonding is finally happy.

Change-ID: I06f4c7e435d5bacbfceaa7c3f209e0ff04be21cc
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index be1b72b..9e15f68 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -173,6 +173,7 @@ enum i40evf_state_t {
 	__I40EVF_RESETTING,		/* in reset */
 	/* Below here, watchdog is running */
 	__I40EVF_DOWN,			/* ready, can be opened */
+	__I40EVF_DOWN_PENDING,		/* descending, waiting for watchdog */
 	__I40EVF_TESTING,		/* in ethtool self-test */
 	__I40EVF_RUNNING,		/* opened, working */
 };
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 94da913..d1c4335 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1032,7 +1032,7 @@ void i40evf_down(struct i40evf_adapter *adapter)
 	struct net_device *netdev = adapter->netdev;
 	struct i40evf_mac_filter *f;
 
-	if (adapter->state == __I40EVF_DOWN)
+	if (adapter->state <= __I40EVF_DOWN_PENDING)
 		return;
 
 	while (test_and_set_bit(__I40EVF_IN_CRITICAL_TASK,
@@ -2142,7 +2142,8 @@ static int i40evf_open(struct net_device *netdev)
 		dev_err(&adapter->pdev->dev, "Unable to open device due to PF driver failure.\n");
 		return -EIO;
 	}
-	if (adapter->state != __I40EVF_DOWN || adapter->aq_required)
+
+	if (adapter->state != __I40EVF_DOWN)
 		return -EBUSY;
 
 	/* allocate transmit descriptors */
@@ -2197,14 +2198,14 @@ static int i40evf_close(struct net_device *netdev)
 {
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
 
-	if (adapter->state <= __I40EVF_DOWN)
+	if (adapter->state <= __I40EVF_DOWN_PENDING)
 		return 0;
 
 
 	set_bit(__I40E_DOWN, &adapter->vsi.state);
 
 	i40evf_down(adapter);
-	adapter->state = __I40EVF_DOWN;
+	adapter->state = __I40EVF_DOWN_PENDING;
 	i40evf_free_traffic_irqs(adapter);
 
 	return 0;
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index c1c5262..d3739cc 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -804,6 +804,8 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 	case I40E_VIRTCHNL_OP_DISABLE_QUEUES:
 		i40evf_free_all_tx_resources(adapter);
 		i40evf_free_all_rx_resources(adapter);
+		if (adapter->state == __I40EVF_DOWN_PENDING)
+			adapter->state = __I40EVF_DOWN;
 		break;
 	case I40E_VIRTCHNL_OP_VERSION:
 	case I40E_VIRTCHNL_OP_CONFIG_IRQ_MAP:
-- 
cgit v0.10.2


From 406e734aa893fa5841e67de6d4f688ba70a82e4f Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Thu, 10 Dec 2015 11:38:49 -0800
Subject: i40e: define function capabilities in only one place

The device capabilities were defined in two places, and neither had all
the definitions.  It really belongs with the AQ API definition, so this
patch removes the other set of definitions and fills out the missing item.

Change-ID: I273ba7d79a476cd11d2e0ca5825fec1716740de2
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index b22012a..256ce65 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -422,6 +422,7 @@ struct i40e_aqc_list_capabilities_element_resp {
 #define I40E_AQ_CAP_ID_LED		0x0061
 #define I40E_AQ_CAP_ID_SDP		0x0062
 #define I40E_AQ_CAP_ID_MDIO		0x0063
+#define I40E_AQ_CAP_ID_WSR_PROT		0x0064
 #define I40E_AQ_CAP_ID_FLEX10		0x00F1
 #define I40E_AQ_CAP_ID_CEM		0x00F2
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 6a034dd..4bdb08b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -2765,35 +2765,6 @@ i40e_aq_erase_nvm_exit:
 	return status;
 }
 
-#define I40E_DEV_FUNC_CAP_SWITCH_MODE	0x01
-#define I40E_DEV_FUNC_CAP_MGMT_MODE	0x02
-#define I40E_DEV_FUNC_CAP_NPAR		0x03
-#define I40E_DEV_FUNC_CAP_OS2BMC	0x04
-#define I40E_DEV_FUNC_CAP_VALID_FUNC	0x05
-#define I40E_DEV_FUNC_CAP_SRIOV_1_1	0x12
-#define I40E_DEV_FUNC_CAP_VF		0x13
-#define I40E_DEV_FUNC_CAP_VMDQ		0x14
-#define I40E_DEV_FUNC_CAP_802_1_QBG	0x15
-#define I40E_DEV_FUNC_CAP_802_1_QBH	0x16
-#define I40E_DEV_FUNC_CAP_VSI		0x17
-#define I40E_DEV_FUNC_CAP_DCB		0x18
-#define I40E_DEV_FUNC_CAP_FCOE		0x21
-#define I40E_DEV_FUNC_CAP_ISCSI		0x22
-#define I40E_DEV_FUNC_CAP_RSS		0x40
-#define I40E_DEV_FUNC_CAP_RX_QUEUES	0x41
-#define I40E_DEV_FUNC_CAP_TX_QUEUES	0x42
-#define I40E_DEV_FUNC_CAP_MSIX		0x43
-#define I40E_DEV_FUNC_CAP_MSIX_VF	0x44
-#define I40E_DEV_FUNC_CAP_FLOW_DIRECTOR	0x45
-#define I40E_DEV_FUNC_CAP_IEEE_1588	0x46
-#define I40E_DEV_FUNC_CAP_FLEX10	0xF1
-#define I40E_DEV_FUNC_CAP_CEM		0xF2
-#define I40E_DEV_FUNC_CAP_IWARP		0x51
-#define I40E_DEV_FUNC_CAP_LED		0x61
-#define I40E_DEV_FUNC_CAP_SDP		0x62
-#define I40E_DEV_FUNC_CAP_MDIO		0x63
-#define I40E_DEV_FUNC_CAP_WR_CSR_PROT	0x64
-
 /**
  * i40e_parse_discover_capabilities
  * @hw: pointer to the hw struct
@@ -2832,79 +2803,79 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 		major_rev = cap->major_rev;
 
 		switch (id) {
-		case I40E_DEV_FUNC_CAP_SWITCH_MODE:
+		case I40E_AQ_CAP_ID_SWITCH_MODE:
 			p->switch_mode = number;
 			break;
-		case I40E_DEV_FUNC_CAP_MGMT_MODE:
+		case I40E_AQ_CAP_ID_MNG_MODE:
 			p->management_mode = number;
 			break;
-		case I40E_DEV_FUNC_CAP_NPAR:
+		case I40E_AQ_CAP_ID_NPAR_ACTIVE:
 			p->npar_enable = number;
 			break;
-		case I40E_DEV_FUNC_CAP_OS2BMC:
+		case I40E_AQ_CAP_ID_OS2BMC_CAP:
 			p->os2bmc = number;
 			break;
-		case I40E_DEV_FUNC_CAP_VALID_FUNC:
+		case I40E_AQ_CAP_ID_FUNCTIONS_VALID:
 			p->valid_functions = number;
 			break;
-		case I40E_DEV_FUNC_CAP_SRIOV_1_1:
+		case I40E_AQ_CAP_ID_SRIOV:
 			if (number == 1)
 				p->sr_iov_1_1 = true;
 			break;
-		case I40E_DEV_FUNC_CAP_VF:
+		case I40E_AQ_CAP_ID_VF:
 			p->num_vfs = number;
 			p->vf_base_id = logical_id;
 			break;
-		case I40E_DEV_FUNC_CAP_VMDQ:
+		case I40E_AQ_CAP_ID_VMDQ:
 			if (number == 1)
 				p->vmdq = true;
 			break;
-		case I40E_DEV_FUNC_CAP_802_1_QBG:
+		case I40E_AQ_CAP_ID_8021QBG:
 			if (number == 1)
 				p->evb_802_1_qbg = true;
 			break;
-		case I40E_DEV_FUNC_CAP_802_1_QBH:
+		case I40E_AQ_CAP_ID_8021QBR:
 			if (number == 1)
 				p->evb_802_1_qbh = true;
 			break;
-		case I40E_DEV_FUNC_CAP_VSI:
+		case I40E_AQ_CAP_ID_VSI:
 			p->num_vsis = number;
 			break;
-		case I40E_DEV_FUNC_CAP_DCB:
+		case I40E_AQ_CAP_ID_DCB:
 			if (number == 1) {
 				p->dcb = true;
 				p->enabled_tcmap = logical_id;
 				p->maxtc = phys_id;
 			}
 			break;
-		case I40E_DEV_FUNC_CAP_FCOE:
+		case I40E_AQ_CAP_ID_FCOE:
 			if (number == 1)
 				p->fcoe = true;
 			break;
-		case I40E_DEV_FUNC_CAP_ISCSI:
+		case I40E_AQ_CAP_ID_ISCSI:
 			if (number == 1)
 				p->iscsi = true;
 			break;
-		case I40E_DEV_FUNC_CAP_RSS:
+		case I40E_AQ_CAP_ID_RSS:
 			p->rss = true;
 			p->rss_table_size = number;
 			p->rss_table_entry_width = logical_id;
 			break;
-		case I40E_DEV_FUNC_CAP_RX_QUEUES:
+		case I40E_AQ_CAP_ID_RXQ:
 			p->num_rx_qp = number;
 			p->base_queue = phys_id;
 			break;
-		case I40E_DEV_FUNC_CAP_TX_QUEUES:
+		case I40E_AQ_CAP_ID_TXQ:
 			p->num_tx_qp = number;
 			p->base_queue = phys_id;
 			break;
-		case I40E_DEV_FUNC_CAP_MSIX:
+		case I40E_AQ_CAP_ID_MSIX:
 			p->num_msix_vectors = number;
 			break;
-		case I40E_DEV_FUNC_CAP_MSIX_VF:
+		case I40E_AQ_CAP_ID_VF_MSIX:
 			p->num_msix_vectors_vf = number;
 			break;
-		case I40E_DEV_FUNC_CAP_FLEX10:
+		case I40E_AQ_CAP_ID_FLEX10:
 			if (major_rev == 1) {
 				if (number == 1) {
 					p->flex10_enable = true;
@@ -2920,38 +2891,38 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 			p->flex10_mode = logical_id;
 			p->flex10_status = phys_id;
 			break;
-		case I40E_DEV_FUNC_CAP_CEM:
+		case I40E_AQ_CAP_ID_CEM:
 			if (number == 1)
 				p->mgmt_cem = true;
 			break;
-		case I40E_DEV_FUNC_CAP_IWARP:
+		case I40E_AQ_CAP_ID_IWARP:
 			if (number == 1)
 				p->iwarp = true;
 			break;
-		case I40E_DEV_FUNC_CAP_LED:
+		case I40E_AQ_CAP_ID_LED:
 			if (phys_id < I40E_HW_CAP_MAX_GPIO)
 				p->led[phys_id] = true;
 			break;
-		case I40E_DEV_FUNC_CAP_SDP:
+		case I40E_AQ_CAP_ID_SDP:
 			if (phys_id < I40E_HW_CAP_MAX_GPIO)
 				p->sdp[phys_id] = true;
 			break;
-		case I40E_DEV_FUNC_CAP_MDIO:
+		case I40E_AQ_CAP_ID_MDIO:
 			if (number == 1) {
 				p->mdio_port_num = phys_id;
 				p->mdio_port_mode = logical_id;
 			}
 			break;
-		case I40E_DEV_FUNC_CAP_IEEE_1588:
+		case I40E_AQ_CAP_ID_1588:
 			if (number == 1)
 				p->ieee_1588 = true;
 			break;
-		case I40E_DEV_FUNC_CAP_FLOW_DIRECTOR:
+		case I40E_AQ_CAP_ID_FLOW_DIRECTOR:
 			p->fd = true;
 			p->fd_filters_guaranteed = number;
 			p->fd_filters_best_effort = logical_id;
 			break;
-		case I40E_DEV_FUNC_CAP_WR_CSR_PROT:
+		case I40E_AQ_CAP_ID_WSR_PROT:
 			p->wr_csr_prot = (u64)number;
 			p->wr_csr_prot |= (u64)logical_id << 32;
 			break;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index f5b2b36..0d3bc3b 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -419,6 +419,7 @@ struct i40e_aqc_list_capabilities_element_resp {
 #define I40E_AQ_CAP_ID_LED		0x0061
 #define I40E_AQ_CAP_ID_SDP		0x0062
 #define I40E_AQ_CAP_ID_MDIO		0x0063
+#define I40E_AQ_CAP_ID_WSR_PROT		0x0064
 #define I40E_AQ_CAP_ID_FLEX10		0x00F1
 #define I40E_AQ_CAP_ID_CEM		0x00F2
 
-- 
cgit v0.10.2


From 10311540fab76c7e5530bf5f0267a3d1b8d5818e Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 9 Dec 2015 15:50:30 -0800
Subject: i40evf: null out ring pointers on free

Since we check these ring pointers to make sure we don't double-allocate
or double-free the rings, we had better null them out after we free
them. In very rare cases this can cause a panic if the driver is removed
during reset recovery.

Change-ID: Ib06eb4910a3058275c8f7ec5ef7f45baa4674f96
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index d1c4335..81d9584 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1122,7 +1122,9 @@ static void i40evf_free_queues(struct i40evf_adapter *adapter)
 	if (!adapter->vsi_res)
 		return;
 	kfree(adapter->tx_rings);
+	adapter->tx_rings = NULL;
 	kfree(adapter->rx_rings);
+	adapter->rx_rings = NULL;
 }
 
 /**
-- 
cgit v0.10.2


From 8eed76fa4885f1ed9f19f4d3a16dd24cebf09c19 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Wed, 9 Dec 2015 15:50:31 -0800
Subject: i40e: Cleanup the code with respect to restarting autoneg

The restart-autoneg work around does not apply to X722.
Added a flag to set it only for the right MAC and right FW version
where the work around should be applied.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Change-ID: I942c3ff40cccd1e56f424b1da776b020fe3c9d2a
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 47f6c0a..53ed3bd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -340,6 +340,7 @@ struct i40e_pf {
 #define I40E_FLAG_GENEVE_OFFLOAD_CAPABLE	BIT_ULL(41)
 #define I40E_FLAG_NO_PCI_LINK_CHECK		BIT_ULL(42)
 #define I40E_FLAG_100M_SGMII_CAPABLE		BIT_ULL(43)
+#define I40E_FLAG_RESTART_AUTONEG		BIT_ULL(44)
 #define I40E_FLAG_PF_MAC			BIT_ULL(50)
 
 	/* tracks features that get auto disabled by errors */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index d078a63..1a7022c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -6889,8 +6889,7 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit)
 		wr32(hw, I40E_REG_MSS, val);
 	}
 
-	if (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 33)) ||
-	    (pf->hw.aq.fw_maj_ver < 4)) {
+	if (pf->flags & I40E_FLAG_RESTART_AUTONEG) {
 		msleep(75);
 		ret = i40e_aq_set_link_restart_an(&pf->hw, true, NULL);
 		if (ret)
@@ -8367,6 +8366,12 @@ static int i40e_sw_init(struct i40e_pf *pf)
 				 pf->hw.func_caps.fd_filters_best_effort;
 	}
 
+	if (((pf->hw.mac.type == I40E_MAC_X710) ||
+	     (pf->hw.mac.type == I40E_MAC_XL710)) &&
+	    (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 33)) ||
+	    (pf->hw.aq.fw_maj_ver < 4)))
+		pf->flags |= I40E_FLAG_RESTART_AUTONEG;
+
 	if (pf->hw.func_caps.vmdq) {
 		pf->num_vmdq_vsis = I40E_DEFAULT_NUM_VMDQ_VSI;
 		pf->flags |= I40E_FLAG_VMDQ_ENABLED;
@@ -10904,8 +10909,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		wr32(hw, I40E_REG_MSS, val);
 	}
 
-	if (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 33)) ||
-	    (pf->hw.aq.fw_maj_ver < 4)) {
+	if (pf->flags & I40E_FLAG_RESTART_AUTONEG) {
 		msleep(75);
 		err = i40e_aq_set_link_restart_an(&pf->hw, true, NULL);
 		if (err)
-- 
cgit v0.10.2


From 5afdaaa0555257f3c42b141908567d40aca0e1d1 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Thu, 10 Dec 2015 11:38:50 -0800
Subject: i40e: update features with right offload

Synchronize code bases and add SCTP offload support.

Change-ID: I9f99071f7176225479026930c387bf681a47494e
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 1a7022c..486ae16 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8947,11 +8947,11 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	np = netdev_priv(netdev);
 	np->vsi = vsi;
 
-	netdev->hw_enc_features |= NETIF_F_IP_CSUM	 |
-				  NETIF_F_RXCSUM	 |
-				  NETIF_F_GSO_UDP_TUNNEL |
-				  NETIF_F_GSO_GRE	 |
-				  NETIF_F_TSO;
+	netdev->hw_enc_features |= NETIF_F_IP_CSUM	  |
+				   NETIF_F_GSO_UDP_TUNNEL |
+				   NETIF_F_GSO_GRE	  |
+				   NETIF_F_TSO		  |
+				   0;
 
 	netdev->features = NETIF_F_SG		       |
 			   NETIF_F_IP_CSUM	       |
-- 
cgit v0.10.2


From 4f9e697ebbf40ba482ad0481da1e978440f2a53e Mon Sep 17 00:00:00 2001
From: Catherine Sullivan <catherine.sullivan@intel.com>
Date: Thu, 10 Dec 2015 11:38:51 -0800
Subject: i40e: bump version to 1.4.10

Bump.

Change-ID: Ic9a495feb9ab0606f953c3848b0acf67169d3930
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 486ae16..c88583e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -51,7 +51,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 8
+#define DRV_VERSION_BUILD 10
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
-- 
cgit v0.10.2


From 35dae51de3e99db10f355642f5fc67719b93f558 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Tue, 22 Dec 2015 14:25:03 -0800
Subject: i40e: add new device IDs for X722

Add the KX and QSFP device IDs for X722.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 4bdb08b..3b03a31 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -55,6 +55,8 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_20G_KR2_A:
 			hw->mac.type = I40E_MAC_XL710;
 			break;
+		case I40E_DEV_ID_KX_X722:
+		case I40E_DEV_ID_QSFP_X722:
 		case I40E_DEV_ID_SFP_X722:
 		case I40E_DEV_ID_1G_BASE_T_X722:
 		case I40E_DEV_ID_10G_BASE_T_X722:
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h
index 448ef4c..f7ce5c7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_devids.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h
@@ -41,6 +41,8 @@
 #define I40E_DEV_ID_10G_BASE_T4		0x1589
 #define I40E_DEV_ID_VF			0x154C
 #define I40E_DEV_ID_VF_HV		0x1571
+#define I40E_DEV_ID_KX_X722		0x37CE
+#define I40E_DEV_ID_QSFP_X722		0x37CF
 #define I40E_DEV_ID_SFP_X722		0x37D0
 #define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
 #define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index c88583e..b3e671b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -90,6 +90,8 @@ static const struct pci_device_id i40e_pci_tbl[] = {
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T4), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_X722), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_X722), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0},
-- 
cgit v0.10.2


From 95a73780921aecc5e66022e000fcf8aeecfb53cf Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Tue, 22 Dec 2015 14:25:04 -0800
Subject: i40e: Extend ethtool RSS hooks for X722

This patch adds another way to access the RSS keys and lut using the AQ
for X722 devices.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index b3e671b..bd81a97 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -7937,6 +7937,52 @@ static int i40e_vsi_config_rss(struct i40e_vsi *vsi)
 }
 
 /**
+ * i40e_get_rss_aq - Get RSS keys and lut by using AQ commands
+ * @vsi: Pointer to vsi structure
+ * @seed: Buffter to store the hash keys
+ * @lut: Buffer to store the lookup table entries
+ * @lut_size: Size of buffer to store the lookup table entries
+ *
+ * Return 0 on success, negative on failure
+ */
+static int i40e_get_rss_aq(struct i40e_vsi *vsi, const u8 *seed,
+			   u8 *lut, u16 lut_size)
+{
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	int ret = 0;
+
+	if (seed) {
+		ret = i40e_aq_get_rss_key(hw, vsi->id,
+			(struct i40e_aqc_get_set_rss_key_data *)seed);
+		if (ret) {
+			dev_info(&pf->pdev->dev,
+				 "Cannot get RSS key, err %s aq_err %s\n",
+				 i40e_stat_str(&pf->hw, ret),
+				 i40e_aq_str(&pf->hw,
+					     pf->hw.aq.asq_last_status));
+			return ret;
+		}
+	}
+
+	if (lut) {
+		bool pf_lut = vsi->type == I40E_VSI_MAIN ? true : false;
+
+		ret = i40e_aq_get_rss_lut(hw, vsi->id, pf_lut, lut, lut_size);
+		if (ret) {
+			dev_info(&pf->pdev->dev,
+				 "Cannot get RSS lut, err %s aq_err %s\n",
+				 i40e_stat_str(&pf->hw, ret),
+				 i40e_aq_str(&pf->hw,
+					     pf->hw.aq.asq_last_status));
+			return ret;
+		}
+	}
+
+	return ret;
+}
+
+/**
  * i40e_config_rss_reg - Configure RSS keys and lut by writing registers
  * @vsi: Pointer to vsi structure
  * @seed: RSS hash seed
@@ -8038,7 +8084,12 @@ int i40e_config_rss(struct i40e_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size)
  */
 int i40e_get_rss(struct i40e_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size)
 {
-	return i40e_get_rss_reg(vsi, seed, lut, lut_size);
+	struct i40e_pf *pf = vsi->back;
+
+	if (pf->flags & I40E_FLAG_RSS_AQ_CAPABLE)
+		return i40e_get_rss_aq(vsi, seed, lut, lut_size);
+	else
+		return i40e_get_rss_reg(vsi, seed, lut, lut_size);
 }
 
 /**
-- 
cgit v0.10.2


From 3d0da5b78262c1f86294419c7a70e4c837aca159 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Tue, 22 Dec 2015 14:25:05 -0800
Subject: i40e/i40evf: Fix for UDP/TCP RSS for X722

The PCTYPES for the X710 and X722 families are different. This patch
makes adjustments for that.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 252a9dd..8a3f93d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2168,6 +2168,10 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		case 0:
 			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
+				hena |=
+			   BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK);
+
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
 			break;
 		default:
@@ -2179,6 +2183,10 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		case 0:
 			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
+				hena |=
+			   BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK);
+
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
 			break;
 		default:
@@ -2190,6 +2198,11 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		case 0:
 			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
+				hena |=
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP);
+
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
 			break;
@@ -2202,6 +2215,11 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		case 0:
 			return -EINVAL;
 		case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+			if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE)
+				hena |=
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP);
+
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
 			break;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 63e62f9..86aacb9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1213,6 +1213,12 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg)
 		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_RSS_REG;
 	}
 
+	if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE) {
+		if (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+			vfres->vf_offload_flags |=
+				I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2;
+	}
+
 	if (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_RX_POLLING)
 		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_RX_POLLING;
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
index 8906785..bd1c272 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
@@ -459,6 +459,7 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 				   struct ethtool_rxnfc *nfc)
 {
 	struct i40e_hw *hw = &adapter->hw;
+	u32 flags = adapter->vf_res->vf_offload_flags;
 
 	u64 hena = (u64)rd32(hw, I40E_VFQF_HENA(0)) |
 		   ((u64)rd32(hw, I40E_VFQF_HENA(1)) << 32);
@@ -477,19 +478,34 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 
 	switch (nfc->flow_type) {
 	case TCP_V4_FLOW:
-		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3))
+		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+				hena |=
+			   BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK);
+
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
-		else
+		} else {
 			return -EINVAL;
+		}
 		break;
 	case TCP_V6_FLOW:
-		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3))
+		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+				hena |=
+			   BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK);
+
 			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
-		else
+		} else {
 			return -EINVAL;
+		}
 		break;
 	case UDP_V4_FLOW:
 		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+				hena |=
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP);
+
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
 		} else {
@@ -498,6 +514,11 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 		break;
 	case UDP_V6_FLOW:
 		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+				hena |=
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
+			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP);
+
 			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
 		} else {
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 81d9584..798f0de 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1456,7 +1456,11 @@ static int i40evf_init_rss(struct i40evf_adapter *adapter)
 	int ret;
 
 	/* Enable PCTYPES for RSS, TCP/UDP with IPv4/IPv6 */
-	hena = I40E_DEFAULT_RSS_HENA;
+	if (adapter->vf_res->vf_offload_flags &
+					I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+		hena = I40E_DEFAULT_RSS_HENA_EXPANDED;
+	else
+		hena = I40E_DEFAULT_RSS_HENA;
 	wr32(hw, I40E_VFQF_HENA(0), (u32)hena);
 	wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32));
 
@@ -2507,8 +2511,6 @@ static void i40evf_init_task(struct work_struct *work)
 	if (adapter->vf_res->vf_offload_flags &
 		    I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR)
 		adapter->flags |= I40EVF_FLAG_WB_ON_ITR_CAPABLE;
-	if (!RSS_AQ(adapter))
-		i40evf_init_rss(adapter);
 	err = i40evf_request_misc_irq(adapter);
 	if (err)
 		goto err_sw_init;
-- 
cgit v0.10.2


From f6d83d1376f496e6218080dd6eb663830672813f Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Tue, 22 Dec 2015 14:25:07 -0800
Subject: i40evf: add new write-back mode

Add write-back on interrupt throttle rate timer expiration support
for the i40evf driver, when running on X722 devices.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 86aacb9..659d782 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1222,6 +1222,12 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg)
 	if (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_RX_POLLING)
 		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_RX_POLLING;
 
+	if (pf->flags & I40E_FLAG_WB_ON_ITR_CAPABLE) {
+		if (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR)
+			vfres->vf_offload_flags |=
+					I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR;
+	}
+
 	vfres->num_vsis = num_vsis;
 	vfres->num_queue_pairs = vf->num_queue_pairs;
 	vfres->max_vectors = pf->hw.func_caps.num_msix_vectors_vf;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 7a00657..7d663fb 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -252,6 +252,22 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
 	tx_ring->q_vector->tx.total_bytes += total_bytes;
 	tx_ring->q_vector->tx.total_packets += total_packets;
 
+	if (tx_ring->flags & I40E_TXR_FLAGS_WB_ON_ITR) {
+		unsigned int j = 0;
+		/* check to see if there are < 4 descriptors
+		 * waiting to be written back, then kick the hardware to force
+		 * them to be written back in case we stay in NAPI.
+		 * In this mode on X722 we do not enable Interrupt.
+		 */
+		j = i40evf_get_tx_pending(tx_ring);
+
+		if (budget &&
+		    ((j / (WB_STRIDE + 1)) == 0) && (j > 0) &&
+		    !test_bit(__I40E_DOWN, &tx_ring->vsi->state) &&
+		    (I40E_DESC_UNUSED(tx_ring) != tx_ring->count))
+			tx_ring->arm_wb = true;
+	}
+
 	netdev_tx_completed_queue(netdev_get_tx_queue(tx_ring->netdev,
 						      tx_ring->queue_index),
 				  total_packets, total_bytes);
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 798f0de..615ad0f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2511,6 +2511,11 @@ static void i40evf_init_task(struct work_struct *work)
 	if (adapter->vf_res->vf_offload_flags &
 		    I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR)
 		adapter->flags |= I40EVF_FLAG_WB_ON_ITR_CAPABLE;
+
+	if (adapter->vf_res->vf_offload_flags &
+	    I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR)
+		adapter->flags |= I40EVF_FLAG_WB_ON_ITR_CAPABLE;
+
 	err = i40evf_request_misc_irq(adapter);
 	if (err)
 		goto err_sw_init;
-- 
cgit v0.10.2


From 2803b16c10ea7eec170c485388f5f26ae30e92fe Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Tue, 22 Dec 2015 14:25:08 -0800
Subject: i40e/i40evf: Use private workqueue

As done per ixgbe, use a private workqueue to avoid blocking the
system workqueue.  This avoids some strange side effects when
some other entity is depending on the system work queue.

Change-ID: Ic8ba08f5b03696cf638b21afd25fbae7738d55ee
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index bd81a97..3e482bc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -112,6 +112,8 @@ MODULE_DESCRIPTION("Intel(R) Ethernet Connection XL710 Network Driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 
+static struct workqueue_struct *i40e_wq;
+
 /**
  * i40e_allocate_dma_mem_d - OS specific memory alloc for shared code
  * @hw:   pointer to the HW structure
@@ -297,7 +299,7 @@ static void i40e_service_event_schedule(struct i40e_pf *pf)
 	if (!test_bit(__I40E_DOWN, &pf->state) &&
 	    !test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state) &&
 	    !test_and_set_bit(__I40E_SERVICE_SCHED, &pf->state))
-		schedule_work(&pf->service_task);
+		queue_work(i40e_wq, &pf->service_task);
 }
 
 /**
@@ -11470,6 +11472,16 @@ static int __init i40e_init_module(void)
 		i40e_driver_string, i40e_driver_version_str);
 	pr_info("%s: %s\n", i40e_driver_name, i40e_copyright);
 
+	/* we will see if single thread per module is enough for now,
+	 * it can't be any worse than using the system workqueue which
+	 * was already single threaded
+	 */
+	i40e_wq = create_singlethread_workqueue(i40e_driver_name);
+	if (!i40e_wq) {
+		pr_err("%s: Failed to create workqueue\n", i40e_driver_name);
+		return -ENOMEM;
+	}
+
 	i40e_dbg_init();
 	return pci_register_driver(&i40e_driver);
 }
@@ -11484,6 +11496,7 @@ module_init(i40e_init_module);
 static void __exit i40e_exit_module(void)
 {
 	pci_unregister_driver(&i40e_driver);
+	destroy_workqueue(i40e_wq);
 	i40e_dbg_exit();
 }
 module_exit(i40e_exit_module);
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 615ad0f..66964eb 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -69,6 +69,8 @@ MODULE_DESCRIPTION("Intel(R) XL710 X710 Virtual Function Network Driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 
+static struct workqueue_struct *i40evf_wq;
+
 /**
  * i40evf_allocate_dma_mem_d - OS specific memory alloc for shared code
  * @hw:   pointer to the HW structure
@@ -182,7 +184,7 @@ static void i40evf_tx_timeout(struct net_device *netdev)
 	if (!(adapter->flags & (I40EVF_FLAG_RESET_PENDING |
 				I40EVF_FLAG_RESET_NEEDED))) {
 		adapter->flags |= I40EVF_FLAG_RESET_NEEDED;
-		schedule_work(&adapter->reset_task);
+		queue_work(i40evf_wq, &adapter->reset_task);
 	}
 }
 
@@ -2895,6 +2897,11 @@ static int __init i40evf_init_module(void)
 
 	pr_info("%s\n", i40evf_copyright);
 
+	i40evf_wq = create_singlethread_workqueue(i40evf_driver_name);
+	if (!i40evf_wq) {
+		pr_err("%s: Failed to create workqueue\n", i40evf_driver_name);
+		return -ENOMEM;
+	}
 	ret = pci_register_driver(&i40evf_driver);
 	return ret;
 }
@@ -2910,6 +2917,7 @@ module_init(i40evf_init_module);
 static void __exit i40evf_exit_module(void)
 {
 	pci_unregister_driver(&i40evf_driver);
+	destroy_workqueue(i40evf_wq);
 }
 
 module_exit(i40evf_exit_module);
-- 
cgit v0.10.2


From 4ba40bcea7046a1fa9b57a8cb4bcb8776e86e7ed Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Tue, 22 Dec 2015 14:25:09 -0800
Subject: i40e: add new proxy-wol bit for X722

Add the new proxy-wake-on-lan capability bit available with the
new X722 device.

Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 256ce65..bff0995 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -402,6 +402,7 @@ struct i40e_aqc_list_capabilities_element_resp {
 #define I40E_AQ_CAP_ID_OS2BMC_CAP	0x0004
 #define I40E_AQ_CAP_ID_FUNCTIONS_VALID	0x0005
 #define I40E_AQ_CAP_ID_ALTERNATE_RAM	0x0006
+#define I40E_AQ_CAP_ID_WOL_AND_PROXY	0x0008
 #define I40E_AQ_CAP_ID_SRIOV		0x0012
 #define I40E_AQ_CAP_ID_VF		0x0013
 #define I40E_AQ_CAP_ID_VMDQ		0x0014
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 0d3bc3b..365a7d6 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -399,6 +399,7 @@ struct i40e_aqc_list_capabilities_element_resp {
 #define I40E_AQ_CAP_ID_OS2BMC_CAP	0x0004
 #define I40E_AQ_CAP_ID_FUNCTIONS_VALID	0x0005
 #define I40E_AQ_CAP_ID_ALTERNATE_RAM	0x0006
+#define I40E_AQ_CAP_ID_WOL_AND_PROXY	0x0008
 #define I40E_AQ_CAP_ID_SRIOV		0x0012
 #define I40E_AQ_CAP_ID_VF		0x0013
 #define I40E_AQ_CAP_ID_VMDQ		0x0014
-- 
cgit v0.10.2


From 6dfae3892473e91080accdb2da8476481393f769 Mon Sep 17 00:00:00 2001
From: Greg Bowers <gregory.j.bowers@intel.com>
Date: Tue, 22 Dec 2015 14:25:10 -0800
Subject: i40e: Limit DCB FW version checks to X710/XL710 devices

X710/XL710 devices require FW version checks to properly handle DCB
configurations from the FW.  Newer devices do not, so limit these checks
to X710/XL710.

Signed-off-by: Greg Bowers <gregory.j.bowers@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb.c b/drivers/net/ethernet/intel/i40e/i40e_dcb.c
index 2691277..582daa7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb.c
@@ -814,13 +814,15 @@ i40e_status i40e_get_dcb_config(struct i40e_hw *hw)
 	struct i40e_aqc_get_cee_dcb_cfg_resp cee_cfg;
 	struct i40e_aqc_get_cee_dcb_cfg_v1_resp cee_v1_cfg;
 
-	/* If Firmware version < v4.33 IEEE only */
-	if (((hw->aq.fw_maj_ver == 4) && (hw->aq.fw_min_ver < 33)) ||
-	    (hw->aq.fw_maj_ver < 4))
+	/* If Firmware version < v4.33 on X710/XL710, IEEE only */
+	if ((hw->mac.type == I40E_MAC_XL710) &&
+	    (((hw->aq.fw_maj_ver == 4) && (hw->aq.fw_min_ver < 33)) ||
+	      (hw->aq.fw_maj_ver < 4)))
 		return i40e_get_ieee_dcb_config(hw);
 
-	/* If Firmware version == v4.33 use old CEE struct */
-	if ((hw->aq.fw_maj_ver == 4) && (hw->aq.fw_min_ver == 33)) {
+	/* If Firmware version == v4.33 on X710/XL710, use old CEE struct */
+	if ((hw->mac.type == I40E_MAC_XL710) &&
+	    ((hw->aq.fw_maj_ver == 4) && (hw->aq.fw_min_ver == 33))) {
 		ret = i40e_aq_get_cee_dcb_config(hw, &cee_v1_cfg,
 						 sizeof(cee_v1_cfg), NULL);
 		if (!ret) {
-- 
cgit v0.10.2


From 5394f02f0c0553f97bb4c5596a34c9a7333c032b Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Tue, 22 Dec 2015 14:25:11 -0800
Subject: i40e: AQ Add Run PHY Activity struct

Add the AQ opcode and struct definitions for the Run PHY Activity command

Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index bff0995..9e340ca 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -220,6 +220,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_phy_wol_caps		= 0x0621,
 	i40e_aqc_opc_set_phy_debug		= 0x0622,
 	i40e_aqc_opc_upload_ext_phy_fm		= 0x0625,
+	i40e_aqc_opc_run_phy_activity		= 0x0626,
 
 	/* NVM commands */
 	i40e_aqc_opc_nvm_read			= 0x0701,
@@ -1825,6 +1826,18 @@ enum i40e_aq_phy_reg_type {
 	I40E_AQC_PHY_REG_EXERNAL_MODULE	= 0x3
 };
 
+/* Run PHY Activity (0x0626) */
+struct i40e_aqc_run_phy_activity {
+	__le16  activity_id;
+	u8      flags;
+	u8      reserved1;
+	__le32  control;
+	__le32  data;
+	u8      reserved2[4];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_run_phy_activity);
+
 /* NVM Read command (indirect 0x0701)
  * NVM Erase commands (direct 0x0702)
  * NVM Update commands (indirect 0x0703)
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 365a7d6..51d83c6 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -220,6 +220,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_get_phy_wol_caps		= 0x0621,
 	i40e_aqc_opc_set_phy_debug		= 0x0622,
 	i40e_aqc_opc_upload_ext_phy_fm		= 0x0625,
+	i40e_aqc_opc_run_phy_activity		= 0x0626,
 
 	/* NVM commands */
 	i40e_aqc_opc_nvm_read			= 0x0701,
@@ -1822,6 +1823,18 @@ enum i40e_aq_phy_reg_type {
 	I40E_AQC_PHY_REG_EXERNAL_MODULE	= 0x3
 };
 
+/* Run PHY Activity (0x0626) */
+struct i40e_aqc_run_phy_activity {
+	__le16  activity_id;
+	u8      flags;
+	u8      reserved1;
+	__le32  control;
+	__le32  data;
+	u8      reserved2[4];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_run_phy_activity);
+
 /* NVM Read command (indirect 0x0701)
  * NVM Erase commands (direct 0x0702)
  * NVM Update commands (indirect 0x0703)
-- 
cgit v0.10.2


From 5926425368ba8f1b186d45d96020c288e3bb9b8d Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Tue, 22 Dec 2015 14:25:12 -0800
Subject: i40e: AQ Geneve cloud tunnel type

Fix the name of the new cloud tunnel type from the place-holder NGE
name to the official Geneve.  Also fix the spelling of the VXLAN type.

Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 9e340ca..eab55ea 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -1260,9 +1260,9 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_SHIFT		9
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_MASK		0x1E00
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_XVLAN		0
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_VXLAN		0
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_NVGRE_OMAC		1
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_NGE			2
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
 
 	__le32	tenant_id;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 51d83c6..30b5a33 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -1257,9 +1257,9 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_SHIFT		9
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_MASK		0x1E00
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_XVLAN		0
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_VXLAN		0
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_NVGRE_OMAC		1
-#define I40E_AQC_ADD_CLOUD_TNL_TYPE_NGE			2
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
 
 	__le32	tenant_id;
-- 
cgit v0.10.2


From 5eb772f7ca86267565ef40c7b987c88405689b96 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Tue, 22 Dec 2015 14:25:13 -0800
Subject: i40e: AQ Add external power class to get link status

Add the new External Device Power Ability field to the get_link_status data
structure, using space from the reserved field at the end of the struct.

Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index eab55ea..0e608d2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -1758,7 +1758,12 @@ struct i40e_aqc_get_link_status {
 	u8	config;
 #define I40E_AQ_CONFIG_CRC_ENA		0x04
 #define I40E_AQ_CONFIG_PACING_MASK	0x78
-	u8	reserved[5];
+	u8	external_power_ability;
+#define I40E_AQ_LINK_POWER_CLASS_1	0x00
+#define I40E_AQ_LINK_POWER_CLASS_2	0x01
+#define I40E_AQ_LINK_POWER_CLASS_3	0x02
+#define I40E_AQ_LINK_POWER_CLASS_4	0x03
+	u8	reserved[4];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_get_link_status);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 30b5a33..578b178 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -1755,7 +1755,12 @@ struct i40e_aqc_get_link_status {
 	u8	config;
 #define I40E_AQ_CONFIG_CRC_ENA		0x04
 #define I40E_AQ_CONFIG_PACING_MASK	0x78
-	u8	reserved[5];
+	u8	external_power_ability;
+#define I40E_AQ_LINK_POWER_CLASS_1	0x00
+#define I40E_AQ_LINK_POWER_CLASS_2	0x01
+#define I40E_AQ_LINK_POWER_CLASS_3	0x02
+#define I40E_AQ_LINK_POWER_CLASS_4	0x03
+	u8	reserved[4];
 };
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_get_link_status);
-- 
cgit v0.10.2


From f8db54cc4df7b065b0028f8c919e2f47983f2043 Mon Sep 17 00:00:00 2001
From: Catherine Sullivan <catherine.sullivan@intel.com>
Date: Tue, 22 Dec 2015 14:25:14 -0800
Subject: i40e: add 100Mb ethtool reporting

Add some missing reporting/advertisement of 100Mb capability
for adapters that support it.

Change-ID: I8b8523fbdc99517bec29d90c71b3744db11542ac
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 8a3f93d..4549591 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -411,6 +411,10 @@ static void i40e_get_settings_link_down(struct i40e_hw *hw,
 		if (pf->hw.mac.type == I40E_MAC_X722) {
 			ecmd->supported |= SUPPORTED_100baseT_Full;
 			ecmd->advertising |= ADVERTISED_100baseT_Full;
+			if (pf->flags & I40E_FLAG_100M_SGMII_CAPABLE) {
+				ecmd->supported |= SUPPORTED_100baseT_Full;
+				ecmd->advertising |= ADVERTISED_100baseT_Full;
+			}
 		}
 	}
 	if (phy_types & I40E_CAP_PHY_TYPE_XAUI ||
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 3e482bc..320b049 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8453,6 +8453,7 @@ static int i40e_sw_init(struct i40e_pf *pf)
 			     I40E_FLAG_OUTER_UDP_CSUM_CAPABLE |
 			     I40E_FLAG_WB_ON_ITR_CAPABLE |
 			     I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE |
+			     I40E_FLAG_100M_SGMII_CAPABLE |
 			     I40E_FLAG_GENEVE_OFFLOAD_CAPABLE;
 	}
 	pf->eeprom_version = 0xDEAD;
-- 
cgit v0.10.2


From 296d48568042360d0e2a6e6e91b0130acb5ca738 Mon Sep 17 00:00:00 2001
From: Mahesh Bandewar <maheshb@google.com>
Date: Wed, 27 Jan 2016 23:33:28 -0800
Subject: ipvlan: inherit MTU from master device

When we create IPvlan slave; we use ether_setup() and that
sets up default MTU to 1500 while the master device may have
lower / different MTU. Any subsequent changes to the masters'
MTU are reflected into the slaves' MTU setting. However if those
don't happen (most likely scenario), the slaves' MTU stays at
1500 which could be bad.

This change adds code to inherit MTU from the master device
instead of using the default value during the link initialization
phase.

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
CC: Eric Dumazet <eric.dumazet@gmail.com>
CC: Tim Hockins <thockins@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index f94392d..7a3b414 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -468,6 +468,7 @@ static int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 	ipvlan->dev = dev;
 	ipvlan->port = port;
 	ipvlan->sfeatures = IPVLAN_FEATURES;
+	ipvlan_adjust_mtu(ipvlan, phy_dev);
 	INIT_LIST_HEAD(&ipvlan->addrs);
 
 	/* TODO Probably put random address here to be presented to the
-- 
cgit v0.10.2


From d01332f1acacc0cb43a61f4244dd2b846d4cd585 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Mon, 1 Feb 2016 08:19:56 +0100
Subject: tipc: fix link attribute propagation bug

Changing certain link attributes (link tolerance and link priority)
from the TIPC management tool is supposed to automatically take
effect at both endpoints of the affected link.

Currently the media address is not instantiated for the link and is
used uninstantiated when crafting protocol messages designated for the
peer endpoint. This means that changing a link property currently
results in the property being changed on the local machine but the
protocol message designated for the peer gets lost. Resulting in
property discrepancy between the endpoints.

In this patch we resolve this by using the media address from the
link entry and using the bearer transmit function to send it. Hence,
we can now eliminate the redundant function tipc_link_prot_xmit() and
the redundant field tipc_link::media_addr.

Fixes: 2af5ae372a4b (tipc: clean up unused code and structures)
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Reported-by: Jason Hu <huzhijiang@gmail.com>
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/link.c b/net/tipc/link.c
index 0c2944f..f156353 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -123,7 +123,6 @@ struct tipc_stats {
 struct tipc_link {
 	u32 addr;
 	char name[TIPC_MAX_LINK_NAME];
-	struct tipc_media_addr *media_addr;
 	struct net *net;
 
 	/* Management and link supervision data */
@@ -1261,26 +1260,6 @@ drop:
 	return rc;
 }
 
-/*
- * Send protocol message to the other endpoint.
- */
-static void tipc_link_proto_xmit(struct tipc_link *l, u32 msg_typ,
-				 int probe_msg, u32 gap, u32 tolerance,
-				 u32 priority)
-{
-	struct sk_buff *skb = NULL;
-	struct sk_buff_head xmitq;
-
-	__skb_queue_head_init(&xmitq);
-	tipc_link_build_proto_msg(l, msg_typ, probe_msg, gap,
-				  tolerance, priority, &xmitq);
-	skb = __skb_dequeue(&xmitq);
-	if (!skb)
-		return;
-	tipc_bearer_xmit_skb(l->net, l->bearer_id, skb, l->media_addr);
-	l->rcv_unacked = 0;
-}
-
 static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
 				      u16 rcvgap, int tolerance, int priority,
 				      struct sk_buff_head *xmitq)
@@ -2021,16 +2000,18 @@ msg_full:
 	return -EMSGSIZE;
 }
 
-void tipc_link_set_tolerance(struct tipc_link *l, u32 tol)
+void tipc_link_set_tolerance(struct tipc_link *l, u32 tol,
+			     struct sk_buff_head *xmitq)
 {
 	l->tolerance = tol;
-	tipc_link_proto_xmit(l, STATE_MSG, 0, 0, tol, 0);
+	tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, tol, 0, xmitq);
 }
 
-void tipc_link_set_prio(struct tipc_link *l, u32 prio)
+void tipc_link_set_prio(struct tipc_link *l, u32 prio,
+			struct sk_buff_head *xmitq)
 {
 	l->priority = prio;
-	tipc_link_proto_xmit(l, STATE_MSG, 0, 0, 0, prio);
+	tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, prio, xmitq);
 }
 
 void tipc_link_set_abort_limit(struct tipc_link *l, u32 limit)
diff --git a/net/tipc/link.h b/net/tipc/link.h
index b2ae0f4..b4ee9d6 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -112,8 +112,10 @@ char tipc_link_plane(struct tipc_link *l);
 int tipc_link_prio(struct tipc_link *l);
 int tipc_link_window(struct tipc_link *l);
 unsigned long tipc_link_tolerance(struct tipc_link *l);
-void tipc_link_set_tolerance(struct tipc_link *l, u32 tol);
-void tipc_link_set_prio(struct tipc_link *l, u32 prio);
+void tipc_link_set_tolerance(struct tipc_link *l, u32 tol,
+			     struct sk_buff_head *xmitq);
+void tipc_link_set_prio(struct tipc_link *l, u32 prio,
+			struct sk_buff_head *xmitq);
 void tipc_link_set_abort_limit(struct tipc_link *l, u32 limit);
 void tipc_link_set_queue_limits(struct tipc_link *l, u32 window);
 int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg,
diff --git a/net/tipc/node.c b/net/tipc/node.c
index fa97d96..f8a8255 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -1637,9 +1637,12 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info)
 	char *name;
 	struct tipc_link *link;
 	struct tipc_node *node;
+	struct sk_buff_head xmitq;
 	struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1];
 	struct net *net = sock_net(skb->sk);
 
+	__skb_queue_head_init(&xmitq);
+
 	if (!info->attrs[TIPC_NLA_LINK])
 		return -EINVAL;
 
@@ -1683,13 +1686,13 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info)
 			u32 tol;
 
 			tol = nla_get_u32(props[TIPC_NLA_PROP_TOL]);
-			tipc_link_set_tolerance(link, tol);
+			tipc_link_set_tolerance(link, tol, &xmitq);
 		}
 		if (props[TIPC_NLA_PROP_PRIO]) {
 			u32 prio;
 
 			prio = nla_get_u32(props[TIPC_NLA_PROP_PRIO]);
-			tipc_link_set_prio(link, prio);
+			tipc_link_set_prio(link, prio, &xmitq);
 		}
 		if (props[TIPC_NLA_PROP_WIN]) {
 			u32 win;
@@ -1701,7 +1704,7 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info)
 
 out:
 	tipc_node_read_unlock(node);
-
+	tipc_bearer_xmit(net, bearer_id, &xmitq, &node->links[bearer_id].maddr);
 	return res;
 }
 
-- 
cgit v0.10.2


From 817298102b0bc936b08dfcc5fbcc2213157050f2 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Mon, 1 Feb 2016 08:19:57 +0100
Subject: tipc: fix link priority propagation

Currently link priority changes isn't handled for active links. In
this patch we resolve this by changing our priority if the peer passes
a valid priority in a state message.

Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/link.c b/net/tipc/link.c
index f156353..6f4a6d9 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1458,6 +1458,12 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 		if (in_range(peers_tol, TIPC_MIN_LINK_TOL, TIPC_MAX_LINK_TOL))
 			l->tolerance = peers_tol;
 
+		if (peers_prio && in_range(peers_prio, TIPC_MIN_LINK_PRI,
+					   TIPC_MAX_LINK_PRI)) {
+			l->priority = peers_prio;
+			rc = tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
+		}
+
 		l->silent_intv_cnt = 0;
 		l->stats.recv_states++;
 		if (msg_probe(hdr))
-- 
cgit v0.10.2


From 9256645af09807bc52fa8b2e66ecd28ab25318c4 Mon Sep 17 00:00:00 2001
From: Jarod Wilson <jarod@redhat.com>
Date: Mon, 1 Feb 2016 18:51:04 -0500
Subject: net/core: relax BUILD_BUG_ON in netdev_stats_to_stats64

The netdev_stats_to_stats64 function copies the deprecated
net_device_stats format stats into rtnl_link_stats64 for legacy support
purposes, but with the BUILD_BUG_ON as it was, it wasn't possible to
extend rtnl_link_stats64 without also extending net_device_stats. Relax
the BUILD_BUG_ON to only require that rtnl_link_stats64 is larger, and
zero out all the stat counters that aren't present in net_device_stats.

CC: Eric Dumazet <edumazet@google.com>
CC: netdev@vger.kernel.org
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/dev.c b/net/core/dev.c
index 8cba3d8..65863e5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7253,24 +7253,31 @@ void netdev_run_todo(void)
 	}
 }
 
-/* Convert net_device_stats to rtnl_link_stats64.  They have the same
- * fields in the same order, with only the type differing.
+/* Convert net_device_stats to rtnl_link_stats64. rtnl_link_stats64 has
+ * all the same fields in the same order as net_device_stats, with only
+ * the type differing, but rtnl_link_stats64 may have additional fields
+ * at the end for newer counters.
  */
 void netdev_stats_to_stats64(struct rtnl_link_stats64 *stats64,
 			     const struct net_device_stats *netdev_stats)
 {
 #if BITS_PER_LONG == 64
-	BUILD_BUG_ON(sizeof(*stats64) != sizeof(*netdev_stats));
+	BUILD_BUG_ON(sizeof(*stats64) < sizeof(*netdev_stats));
 	memcpy(stats64, netdev_stats, sizeof(*stats64));
+	/* zero out counters that only exist in rtnl_link_stats64 */
+	memset((char *)stats64 + sizeof(*netdev_stats), 0,
+	       sizeof(*stats64) - sizeof(*netdev_stats));
 #else
-	size_t i, n = sizeof(*stats64) / sizeof(u64);
+	size_t i, n = sizeof(*netdev_stats) / sizeof(unsigned long);
 	const unsigned long *src = (const unsigned long *)netdev_stats;
 	u64 *dst = (u64 *)stats64;
 
-	BUILD_BUG_ON(sizeof(*netdev_stats) / sizeof(unsigned long) !=
-		     sizeof(*stats64) / sizeof(u64));
+	BUILD_BUG_ON(n > sizeof(*stats64) / sizeof(u64));
 	for (i = 0; i < n; i++)
 		dst[i] = src[i];
+	/* zero out counters that only exist in rtnl_link_stats64 */
+	memset((char *)stats64 + n * sizeof(u64), 0,
+	       sizeof(*stats64) - n * sizeof(u64));
 #endif
 }
 EXPORT_SYMBOL(netdev_stats_to_stats64);
-- 
cgit v0.10.2


From 6e7333d315a768170a59ac771297ee0551bdddbf Mon Sep 17 00:00:00 2001
From: Jarod Wilson <jarod@redhat.com>
Date: Mon, 1 Feb 2016 18:51:05 -0500
Subject: net: add rx_nohandler stat counter

This adds an rx_nohandler stat counter, along with a sysfs statistics
node, and copies the counter out via netlink as well.

CC: "David S. Miller" <davem@davemloft.net>
CC: Eric Dumazet <edumazet@google.com>
CC: Jiri Pirko <jiri@mellanox.com>
CC: Daniel Borkmann <daniel@iogearbox.net>
CC: Tom Herbert <tom@herbertland.com>
CC: Jay Vosburgh <j.vosburgh@gmail.com>
CC: Veaceslav Falico <vfalico@gmail.com>
CC: Andy Gospodarek <gospo@cumulusnetworks.com>
CC: netdev@vger.kernel.org
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 289c231..78a20ce 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1397,6 +1397,8 @@ enum netdev_priv_flags {
  *			do not use this in drivers
  *	@tx_dropped:	Dropped packets by core network,
  *			do not use this in drivers
+ *	@rx_nohandler:	nohandler dropped packets by core network on
+ *			inactive devices, do not use this in drivers
  *
  *	@wireless_handlers:	List of functions to handle Wireless Extensions,
  *				instead of ioctl,
@@ -1611,6 +1613,7 @@ struct net_device {
 
 	atomic_long_t		rx_dropped;
 	atomic_long_t		tx_dropped;
+	atomic_long_t		rx_nohandler;
 
 #ifdef CONFIG_WIRELESS_EXT
 	const struct iw_handler_def *	wireless_handlers;
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index a30b780..d3e90b9 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -35,6 +35,8 @@ struct rtnl_link_stats {
 	/* for cslip etc */
 	__u32	rx_compressed;
 	__u32	tx_compressed;
+
+	__u32	rx_nohandler;		/* dropped, no handler found	*/
 };
 
 /* The main device statistics structure */
@@ -68,6 +70,8 @@ struct rtnl_link_stats64 {
 	/* for cslip etc */
 	__u64	rx_compressed;
 	__u64	tx_compressed;
+
+	__u64	rx_nohandler;		/* dropped, no handler found	*/
 };
 
 /* The struct should be in sync with struct ifmap */
diff --git a/net/core/dev.c b/net/core/dev.c
index 65863e5..f128483 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4154,7 +4154,10 @@ ncls:
 			ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev);
 	} else {
 drop:
-		atomic_long_inc(&skb->dev->rx_dropped);
+		if (!deliver_exact)
+			atomic_long_inc(&skb->dev->rx_dropped);
+		else
+			atomic_long_inc(&skb->dev->rx_nohandler);
 		kfree_skb(skb);
 		/* Jamal, now you will not able to escape explaining
 		 * me how you were going to use this. :-)
@@ -7307,6 +7310,7 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
 	}
 	storage->rx_dropped += atomic_long_read(&dev->rx_dropped);
 	storage->tx_dropped += atomic_long_read(&dev->tx_dropped);
+	storage->rx_nohandler += atomic_long_read(&dev->rx_nohandler);
 	return storage;
 }
 EXPORT_SYMBOL(dev_get_stats);
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index b6c8a66..da7dbc2 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -574,6 +574,7 @@ NETSTAT_ENTRY(tx_heartbeat_errors);
 NETSTAT_ENTRY(tx_window_errors);
 NETSTAT_ENTRY(rx_compressed);
 NETSTAT_ENTRY(tx_compressed);
+NETSTAT_ENTRY(rx_nohandler);
 
 static struct attribute *netstat_attrs[] = {
 	&dev_attr_rx_packets.attr,
@@ -599,6 +600,7 @@ static struct attribute *netstat_attrs[] = {
 	&dev_attr_tx_window_errors.attr,
 	&dev_attr_rx_compressed.attr,
 	&dev_attr_tx_compressed.attr,
+	&dev_attr_rx_nohandler.attr,
 	NULL
 };
 
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index d735e85..20d7135 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -804,6 +804,8 @@ static void copy_rtnl_link_stats(struct rtnl_link_stats *a,
 
 	a->rx_compressed = b->rx_compressed;
 	a->tx_compressed = b->tx_compressed;
+
+	a->rx_nohandler = b->rx_nohandler;
 }
 
 static void copy_rtnl_link_stats64(void *v, const struct rtnl_link_stats64 *b)
-- 
cgit v0.10.2


From bb63daf9efb4f2bcb657d7179a53bd808f978dc9 Mon Sep 17 00:00:00 2001
From: Jarod Wilson <jarod@redhat.com>
Date: Mon, 1 Feb 2016 18:51:06 -0500
Subject: team: track sum of rx_nohandler for all slaves

CC: Jiri Pirko <jiri@resnulli.us>
CC: netdev@vger.kernel.org
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 718ceea..00558e1 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -758,6 +758,8 @@ static rx_handler_result_t team_handle_frame(struct sk_buff **pskb)
 		u64_stats_update_end(&pcpu_stats->syncp);
 
 		skb->dev = team->dev;
+	} else if (res == RX_HANDLER_EXACT) {
+		this_cpu_inc(team->pcpu_stats->rx_nohandler);
 	} else {
 		this_cpu_inc(team->pcpu_stats->rx_dropped);
 	}
@@ -1807,7 +1809,7 @@ team_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	struct team *team = netdev_priv(dev);
 	struct team_pcpu_stats *p;
 	u64 rx_packets, rx_bytes, rx_multicast, tx_packets, tx_bytes;
-	u32 rx_dropped = 0, tx_dropped = 0;
+	u32 rx_dropped = 0, tx_dropped = 0, rx_nohandler = 0;
 	unsigned int start;
 	int i;
 
@@ -1828,14 +1830,16 @@ team_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 		stats->tx_packets	+= tx_packets;
 		stats->tx_bytes		+= tx_bytes;
 		/*
-		 * rx_dropped & tx_dropped are u32, updated
-		 * without syncp protection.
+		 * rx_dropped, tx_dropped & rx_nohandler are u32,
+		 * updated without syncp protection.
 		 */
 		rx_dropped	+= p->rx_dropped;
 		tx_dropped	+= p->tx_dropped;
+		rx_nohandler	+= p->rx_nohandler;
 	}
 	stats->rx_dropped	= rx_dropped;
 	stats->tx_dropped	= tx_dropped;
+	stats->rx_nohandler	= rx_nohandler;
 	return stats;
 }
 
diff --git a/include/linux/if_team.h b/include/linux/if_team.h
index b84e49c..174f43f 100644
--- a/include/linux/if_team.h
+++ b/include/linux/if_team.h
@@ -24,6 +24,7 @@ struct team_pcpu_stats {
 	struct u64_stats_sync	syncp;
 	u32			rx_dropped;
 	u32			tx_dropped;
+	u32			rx_nohandler;
 };
 
 struct team;
-- 
cgit v0.10.2


From f344b0d940d2da88c23b864f818da43081ce300f Mon Sep 17 00:00:00 2001
From: Jarod Wilson <jarod@redhat.com>
Date: Mon, 1 Feb 2016 18:51:07 -0500
Subject: bond: track sum of rx_nohandler for all slaves

Sample output with this set applied for an active-backup bond:

$ cat /sys/devices/virtual/net/bond0/lower_p7p1/statistics/rx_nohandler
16568
$ cat /sys/devices/virtual/net/bond0/lower_p5p2/statistics/rx_nohandler
16583
$ cat /sys/devices/virtual/net/bond0/statistics/rx_nohandler
33151

CC: Jay Vosburgh <j.vosburgh@gmail.com>
CC: Veaceslav Falico <vfalico@gmail.com>
CC: Andy Gospodarek <gospo@cumulusnetworks.com>
CC: netdev@vger.kernel.org
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 56b5605..6587929 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -3309,6 +3309,7 @@ static struct rtnl_link_stats64 *bond_get_stats(struct net_device *bond_dev,
 		stats->rx_bytes += sstats->rx_bytes - pstats->rx_bytes;
 		stats->rx_errors += sstats->rx_errors - pstats->rx_errors;
 		stats->rx_dropped += sstats->rx_dropped - pstats->rx_dropped;
+		stats->rx_nohandler += sstats->rx_nohandler - pstats->rx_nohandler;
 
 		stats->tx_packets += sstats->tx_packets - pstats->tx_packets;;
 		stats->tx_bytes += sstats->tx_bytes - pstats->tx_bytes;
-- 
cgit v0.10.2


From 61d2bcae99f66a640b3dd9632180209143fb5512 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 1 Feb 2016 21:03:07 -0800
Subject: tcp: fastopen: accept data/FIN present in SYNACK message

RFC 7413 (TCP Fast Open) 4.2.2 states that the SYNACK message
MAY include data and/or FIN

This patch adds support for the client side :

If we receive a SYNACK with payload or FIN, queue the skb instead
of ignoring it.

Since we already support the same for SYN, we refactor the existing
code and reuse it. Note we need to clone the skb, so this operation
might fail under memory pressure.

Sara Dickinson pointed out FreeBSD server Fast Open implementation
was planned to generate such SYNACK in the future.

The server side might be implemented on linux later.

Reported-by: Sara Dickinson <sara@sinodun.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/tcp.h b/include/net/tcp.h
index f6f8f03..27f4c73 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1437,6 +1437,7 @@ void tcp_free_fastopen_req(struct tcp_sock *tp);
 
 extern struct tcp_fastopen_context __rcu *tcp_fastopen_ctx;
 int tcp_fastopen_reset_cipher(void *key, unsigned int len);
+void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb);
 struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb,
 			      struct request_sock *req,
 			      struct tcp_fastopen_cookie *foc,
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 55be6ac..467d3e9 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -124,6 +124,35 @@ static bool tcp_fastopen_cookie_gen(struct request_sock *req,
 	return false;
 }
 
+
+/* If an incoming SYN or SYNACK frame contains a payload and/or FIN,
+ * queue this additional data / FIN.
+ */
+void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (TCP_SKB_CB(skb)->end_seq == tp->rcv_nxt)
+		return;
+
+	skb = skb_clone(skb, GFP_ATOMIC);
+	if (!skb)
+		return;
+
+	skb_dst_drop(skb);
+	__skb_pull(skb, tcp_hdrlen(skb));
+	skb_set_owner_r(skb, sk);
+
+	tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
+	__skb_queue_tail(&sk->sk_receive_queue, skb);
+	tp->syn_data_acked = 1;
+
+	/* u64_stats_update_begin(&tp->syncp) not needed here,
+	 * as we certainly are not changing upper 32bit value (0)
+	 */
+	tp->bytes_received = skb->len;
+}
+
 static struct sock *tcp_fastopen_create_child(struct sock *sk,
 					      struct sk_buff *skb,
 					      struct dst_entry *dst,
@@ -132,7 +161,6 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk,
 	struct tcp_sock *tp;
 	struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
 	struct sock *child;
-	u32 end_seq;
 	bool own_req;
 
 	req->num_retrans = 0;
@@ -178,35 +206,11 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk,
 	tcp_init_metrics(child);
 	tcp_init_buffer_space(child);
 
-	/* Queue the data carried in the SYN packet.
-	 * We used to play tricky games with skb_get().
-	 * With lockless listener, it is a dead end.
-	 * Do not think about it.
-	 *
-	 * XXX (TFO) - we honor a zero-payload TFO request for now,
-	 * (any reason not to?) but no need to queue the skb since
-	 * there is no data. How about SYN+FIN?
-	 */
-	end_seq = TCP_SKB_CB(skb)->end_seq;
-	if (end_seq != TCP_SKB_CB(skb)->seq + 1) {
-		struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
-
-		if (likely(skb2)) {
-			skb_dst_drop(skb2);
-			__skb_pull(skb2, tcp_hdrlen(skb));
-			skb_set_owner_r(skb2, child);
-			__skb_queue_tail(&child->sk_receive_queue, skb2);
-			tp->syn_data_acked = 1;
-
-			/* u64_stats_update_begin(&tp->syncp) not needed here,
-			 * as we certainly are not changing upper 32bit value (0)
-			 */
-			tp->bytes_received = end_seq - TCP_SKB_CB(skb)->seq - 1;
-		} else {
-			end_seq = TCP_SKB_CB(skb)->seq + 1;
-		}
-	}
-	tcp_rsk(req)->rcv_nxt = tp->rcv_nxt = end_seq;
+	tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
+
+	tcp_fastopen_add_skb(child, skb);
+
+	tcp_rsk(req)->rcv_nxt = tp->rcv_nxt;
 	/* tcp_conn_request() is sending the SYNACK,
 	 * and queues the child into listener accept queue.
 	 */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 1c2a734..4add3eb 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5509,6 +5509,9 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack,
 	tp->syn_data_acked = tp->syn_data;
 	if (tp->syn_data_acked)
 		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENACTIVE);
+
+	tcp_fastopen_add_skb(sk, synack);
+
 	return false;
 }
 
-- 
cgit v0.10.2


From 9d691539eea2d977e3eb86766c389a19a9c13146 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 1 Feb 2016 21:03:08 -0800
Subject: tcp: do not enqueue skb with SYN flag

If we remove the SYN flag from the skbs that tcp_fastopen_add_skb()
places in socket receive queue, then we can remove the test that
tcp_recvmsg() has to perform in fast path.

All we have to do is to adjust SEQ in the slow path.

For the moment, we place an unlikely() and output a message
if we find an skb having SYN flag set.
Goal would be to get rid of the test completely.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 19746b3..c507577 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1466,8 +1466,10 @@ static struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off)
 
 	while ((skb = skb_peek(&sk->sk_receive_queue)) != NULL) {
 		offset = seq - TCP_SKB_CB(skb)->seq;
-		if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)
+		if (unlikely(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)) {
+			pr_err_once("%s: found a SYN, please report !\n", __func__);
 			offset--;
+		}
 		if (offset < skb->len || (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)) {
 			*off = offset;
 			return skb;
@@ -1657,8 +1659,10 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock,
 				break;
 
 			offset = *seq - TCP_SKB_CB(skb)->seq;
-			if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)
+			if (unlikely(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)) {
+				pr_err_once("%s: found a SYN, please report !\n", __func__);
 				offset--;
+			}
 			if (offset < skb->len)
 				goto found_ok_skb;
 			if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 467d3e9..6a6e11e 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -143,6 +143,9 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
 	__skb_pull(skb, tcp_hdrlen(skb));
 	skb_set_owner_r(skb, sk);
 
+	TCP_SKB_CB(skb)->seq++;
+	TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_SYN;
+
 	tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
 	__skb_queue_tail(&sk->sk_receive_queue, skb);
 	tp->syn_data_acked = 1;
-- 
cgit v0.10.2


From ba905f5e2f63d86ed4cfbd3d9096fb28d156f1ee Mon Sep 17 00:00:00 2001
From: Kim Jones <kim-marie.jones@intel.com>
Date: Tue, 2 Feb 2016 03:51:16 +0000
Subject: ethtool: Declare netdev_rss_key as __read_mostly.

netdev_rss_key is written to once and thereafter is read by
drivers when they are initialising. The fact that it is mostly
read and not written to makes it a candidate for a __read_mostly
declaration.

Signed-off-by: Kim Jones <kim-marie.jones@intel.com>
Signed-off-by: Alan Carey <alan.carey@intel.com>
Acked-by: Rami Rosen <rami.rosen@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 78a20ce..219f53c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3744,7 +3744,7 @@ void netdev_lower_state_changed(struct net_device *lower_dev,
 
 /* RSS keys are 40 or 52 bytes long */
 #define NETDEV_RSS_KEY_LEN 52
-extern u8 netdev_rss_key[NETDEV_RSS_KEY_LEN];
+extern u8 netdev_rss_key[NETDEV_RSS_KEY_LEN] __read_mostly;
 void netdev_rss_key_fill(void *buffer, size_t len);
 
 int dev_get_nest_level(struct net_device *dev,
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index daf0470..453c803 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -632,7 +632,7 @@ static int ethtool_copy_validate_indir(u32 *indir, void __user *useraddr,
 	return 0;
 }
 
-u8 netdev_rss_key[NETDEV_RSS_KEY_LEN];
+u8 netdev_rss_key[NETDEV_RSS_KEY_LEN] __read_mostly;
 
 void netdev_rss_key_fill(void *buffer, size_t len)
 {
-- 
cgit v0.10.2


From 824bd0ce6c7c43a9e1e210abf124958e54d88342 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 1 Feb 2016 22:39:53 -0800
Subject: bpf: introduce BPF_MAP_TYPE_PERCPU_HASH map

Introduce BPF_MAP_TYPE_PERCPU_HASH map type which is used to do
accurate counters without need to use BPF_XADD instruction which turned
out to be too costly for high-performance network monitoring.
In the typical use case the 'key' is the flow tuple or other long
living object that sees a lot of events per second.

bpf_map_lookup_elem() returns per-cpu area.
Example:
struct {
  u32 packets;
  u32 bytes;
} * ptr = bpf_map_lookup_elem(&map, &key);
/* ptr points to this_cpu area of the value, so the following
 * increments will not collide with other cpus
 */
ptr->packets ++;
ptr->bytes += skb->len;

bpf_update_elem() atomically creates a new element where all per-cpu
values are zero initialized and this_cpu value is populated with
given 'value'.
Note that non-per-cpu hash map always allocates new element
and then deletes old after rcu grace period to maintain atomicity
of update. Per-cpu hash map updates element values in-place.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index aa6f857..43ae40c 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -81,6 +81,7 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_ARRAY,
 	BPF_MAP_TYPE_PROG_ARRAY,
 	BPF_MAP_TYPE_PERF_EVENT_ARRAY,
+	BPF_MAP_TYPE_PERCPU_HASH,
 };
 
 enum bpf_prog_type {
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index c5b30fd..2be5f6e 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -31,21 +31,27 @@ struct bpf_htab {
 struct htab_elem {
 	struct hlist_node hash_node;
 	struct rcu_head rcu;
-	u32 hash;
+	union {
+		u32 hash;
+		u32 key_size;
+	};
 	char key[0] __aligned(8);
 };
 
 /* Called from syscall */
 static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 {
+	bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_HASH;
 	struct bpf_htab *htab;
 	int err, i;
+	u64 cost;
 
 	htab = kzalloc(sizeof(*htab), GFP_USER);
 	if (!htab)
 		return ERR_PTR(-ENOMEM);
 
 	/* mandatory map attributes */
+	htab->map.map_type = attr->map_type;
 	htab->map.key_size = attr->key_size;
 	htab->map.value_size = attr->value_size;
 	htab->map.max_entries = attr->max_entries;
@@ -77,24 +83,34 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 		 */
 		goto free_htab;
 
+	if (percpu && round_up(htab->map.value_size, 8) > PCPU_MIN_UNIT_SIZE)
+		/* make sure the size for pcpu_alloc() is reasonable */
+		goto free_htab;
+
 	htab->elem_size = sizeof(struct htab_elem) +
-			  round_up(htab->map.key_size, 8) +
-			  htab->map.value_size;
+			  round_up(htab->map.key_size, 8);
+	if (percpu)
+		htab->elem_size += sizeof(void *);
+	else
+		htab->elem_size += htab->map.value_size;
 
 	/* prevent zero size kmalloc and check for u32 overflow */
 	if (htab->n_buckets == 0 ||
 	    htab->n_buckets > U32_MAX / sizeof(struct bucket))
 		goto free_htab;
 
-	if ((u64) htab->n_buckets * sizeof(struct bucket) +
-	    (u64) htab->elem_size * htab->map.max_entries >=
-	    U32_MAX - PAGE_SIZE)
+	cost = (u64) htab->n_buckets * sizeof(struct bucket) +
+	       (u64) htab->elem_size * htab->map.max_entries;
+
+	if (percpu)
+		cost += (u64) round_up(htab->map.value_size, 8) *
+			num_possible_cpus() * htab->map.max_entries;
+
+	if (cost >= U32_MAX - PAGE_SIZE)
 		/* make sure page count doesn't overflow */
 		goto free_htab;
 
-	htab->map.pages = round_up(htab->n_buckets * sizeof(struct bucket) +
-				   htab->elem_size * htab->map.max_entries,
-				   PAGE_SIZE) >> PAGE_SHIFT;
+	htab->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT;
 
 	err = -ENOMEM;
 	htab->buckets = kmalloc_array(htab->n_buckets, sizeof(struct bucket),
@@ -148,7 +164,7 @@ static struct htab_elem *lookup_elem_raw(struct hlist_head *head, u32 hash,
 }
 
 /* Called from syscall or from eBPF program */
-static void *htab_map_lookup_elem(struct bpf_map *map, void *key)
+static void *__htab_map_lookup_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct hlist_head *head;
@@ -166,6 +182,13 @@ static void *htab_map_lookup_elem(struct bpf_map *map, void *key)
 
 	l = lookup_elem_raw(head, hash, key, key_size);
 
+	return l;
+}
+
+static void *htab_map_lookup_elem(struct bpf_map *map, void *key)
+{
+	struct htab_elem *l = __htab_map_lookup_elem(map, key);
+
 	if (l)
 		return l->key + round_up(map->key_size, 8);
 
@@ -230,65 +253,139 @@ find_first_elem:
 	return -ENOENT;
 }
 
+
+static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size,
+				     void __percpu *pptr)
+{
+	*(void __percpu **)(l->key + key_size) = pptr;
+}
+
+static inline void __percpu *htab_elem_get_ptr(struct htab_elem *l, u32 key_size)
+{
+	return *(void __percpu **)(l->key + key_size);
+}
+
+static void htab_percpu_elem_free(struct htab_elem *l)
+{
+	free_percpu(htab_elem_get_ptr(l, l->key_size));
+	kfree(l);
+}
+
+static void htab_percpu_elem_free_rcu(struct rcu_head *head)
+{
+	struct htab_elem *l = container_of(head, struct htab_elem, rcu);
+
+	htab_percpu_elem_free(l);
+}
+
+static void free_htab_elem(struct htab_elem *l, bool percpu, u32 key_size)
+{
+	if (percpu) {
+		l->key_size = key_size;
+		call_rcu(&l->rcu, htab_percpu_elem_free_rcu);
+	} else {
+		kfree_rcu(l, rcu);
+	}
+}
+
+static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
+					 void *value, u32 key_size, u32 hash,
+					 bool percpu)
+{
+	u32 size = htab->map.value_size;
+	struct htab_elem *l_new;
+	void __percpu *pptr;
+
+	l_new = kmalloc(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN);
+	if (!l_new)
+		return NULL;
+
+	memcpy(l_new->key, key, key_size);
+	if (percpu) {
+		/* round up value_size to 8 bytes */
+		size = round_up(size, 8);
+
+		/* alloc_percpu zero-fills */
+		pptr = __alloc_percpu_gfp(size, 8, GFP_ATOMIC | __GFP_NOWARN);
+		if (!pptr) {
+			kfree(l_new);
+			return NULL;
+		}
+
+		/* copy true value_size bytes */
+		memcpy(this_cpu_ptr(pptr), value, htab->map.value_size);
+		htab_elem_set_ptr(l_new, key_size, pptr);
+	} else {
+		memcpy(l_new->key + round_up(key_size, 8), value, size);
+	}
+
+	l_new->hash = hash;
+	return l_new;
+}
+
+static int check_flags(struct bpf_htab *htab, struct htab_elem *l_old,
+		       u64 map_flags)
+{
+	if (!l_old && unlikely(atomic_read(&htab->count) >= htab->map.max_entries))
+		/* if elem with this 'key' doesn't exist and we've reached
+		 * max_entries limit, fail insertion of new elem
+		 */
+		return -E2BIG;
+
+	if (l_old && map_flags == BPF_NOEXIST)
+		/* elem already exists */
+		return -EEXIST;
+
+	if (!l_old && map_flags == BPF_EXIST)
+		/* elem doesn't exist, cannot update it */
+		return -ENOENT;
+
+	return 0;
+}
+
 /* Called from syscall or from eBPF program */
 static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 				u64 map_flags)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	struct htab_elem *l_new, *l_old;
+	struct htab_elem *l_new = NULL, *l_old;
 	struct hlist_head *head;
-	struct bucket *b;
 	unsigned long flags;
-	u32 key_size;
+	struct bucket *b;
+	u32 key_size, hash;
 	int ret;
 
-	if (map_flags > BPF_EXIST)
+	if (unlikely(map_flags > BPF_EXIST))
 		/* unknown flags */
 		return -EINVAL;
 
 	WARN_ON_ONCE(!rcu_read_lock_held());
 
-	/* allocate new element outside of lock */
-	l_new = kmalloc(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN);
-	if (!l_new)
-		return -ENOMEM;
-
 	key_size = map->key_size;
 
-	memcpy(l_new->key, key, key_size);
-	memcpy(l_new->key + round_up(key_size, 8), value, map->value_size);
+	hash = htab_map_hash(key, key_size);
+
+	/* allocate new element outside of the lock, since
+	 * we're most likley going to insert it
+	 */
+	l_new = alloc_htab_elem(htab, key, value, key_size, hash, false);
+	if (!l_new)
+		return -ENOMEM;
 
-	l_new->hash = htab_map_hash(l_new->key, key_size);
-	b = __select_bucket(htab, l_new->hash);
+	b = __select_bucket(htab, hash);
 	head = &b->head;
 
 	/* bpf_map_update_elem() can be called in_irq() */
 	raw_spin_lock_irqsave(&b->lock, flags);
 
-	l_old = lookup_elem_raw(head, l_new->hash, key, key_size);
+	l_old = lookup_elem_raw(head, hash, key, key_size);
 
-	if (!l_old && unlikely(atomic_read(&htab->count) >= map->max_entries)) {
-		/* if elem with this 'key' doesn't exist and we've reached
-		 * max_entries limit, fail insertion of new elem
-		 */
-		ret = -E2BIG;
+	ret = check_flags(htab, l_old, map_flags);
+	if (ret)
 		goto err;
-	}
 
-	if (l_old && map_flags == BPF_NOEXIST) {
-		/* elem already exists */
-		ret = -EEXIST;
-		goto err;
-	}
-
-	if (!l_old && map_flags == BPF_EXIST) {
-		/* elem doesn't exist, cannot update it */
-		ret = -ENOENT;
-		goto err;
-	}
-
-	/* add new element to the head of the list, so that concurrent
-	 * search will find it before old elem
+	/* add new element to the head of the list, so that
+	 * concurrent search will find it before old elem
 	 */
 	hlist_add_head_rcu(&l_new->hash_node, head);
 	if (l_old) {
@@ -298,7 +395,6 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 		atomic_inc(&htab->count);
 	}
 	raw_spin_unlock_irqrestore(&b->lock, flags);
-
 	return 0;
 err:
 	raw_spin_unlock_irqrestore(&b->lock, flags);
@@ -306,10 +402,64 @@ err:
 	return ret;
 }
 
+static int htab_percpu_map_update_elem(struct bpf_map *map, void *key,
+				       void *value, u64 map_flags)
+{
+	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
+	struct htab_elem *l_new = NULL, *l_old;
+	struct hlist_head *head;
+	unsigned long flags;
+	struct bucket *b;
+	u32 key_size, hash;
+	int ret;
+
+	if (unlikely(map_flags > BPF_EXIST))
+		/* unknown flags */
+		return -EINVAL;
+
+	WARN_ON_ONCE(!rcu_read_lock_held());
+
+	key_size = map->key_size;
+
+	hash = htab_map_hash(key, key_size);
+
+	b = __select_bucket(htab, hash);
+	head = &b->head;
+
+	/* bpf_map_update_elem() can be called in_irq() */
+	raw_spin_lock_irqsave(&b->lock, flags);
+
+	l_old = lookup_elem_raw(head, hash, key, key_size);
+
+	ret = check_flags(htab, l_old, map_flags);
+	if (ret)
+		goto err;
+
+	if (l_old) {
+		/* per-cpu hash map can update value in-place */
+		memcpy(this_cpu_ptr(htab_elem_get_ptr(l_old, key_size)),
+		       value, htab->map.value_size);
+	} else {
+		l_new = alloc_htab_elem(htab, key, value, key_size,
+					hash, true);
+		if (!l_new) {
+			ret = -ENOMEM;
+			goto err;
+		}
+		hlist_add_head_rcu(&l_new->hash_node, head);
+		atomic_inc(&htab->count);
+	}
+	ret = 0;
+err:
+	raw_spin_unlock_irqrestore(&b->lock, flags);
+	return ret;
+}
+
 /* Called from syscall or from eBPF program */
 static int htab_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
+	bool percpu = map->map_type == BPF_MAP_TYPE_PERCPU_HASH;
 	struct hlist_head *head;
 	struct bucket *b;
 	struct htab_elem *l;
@@ -332,7 +482,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key)
 	if (l) {
 		hlist_del_rcu(&l->hash_node);
 		atomic_dec(&htab->count);
-		kfree_rcu(l, rcu);
+		free_htab_elem(l, percpu, key_size);
 		ret = 0;
 	}
 
@@ -352,7 +502,12 @@ static void delete_all_elements(struct bpf_htab *htab)
 		hlist_for_each_entry_safe(l, n, head, hash_node) {
 			hlist_del_rcu(&l->hash_node);
 			atomic_dec(&htab->count);
-			kfree(l);
+			if (htab->map.map_type == BPF_MAP_TYPE_PERCPU_HASH) {
+				l->key_size = htab->map.key_size;
+				htab_percpu_elem_free(l);
+			} else {
+				kfree(l);
+			}
 		}
 	}
 }
@@ -391,9 +546,35 @@ static struct bpf_map_type_list htab_type __read_mostly = {
 	.type = BPF_MAP_TYPE_HASH,
 };
 
+/* Called from eBPF program */
+static void *htab_percpu_map_lookup_elem(struct bpf_map *map, void *key)
+{
+	struct htab_elem *l = __htab_map_lookup_elem(map, key);
+
+	if (l)
+		return this_cpu_ptr(htab_elem_get_ptr(l, map->key_size));
+	else
+		return NULL;
+}
+
+static const struct bpf_map_ops htab_percpu_ops = {
+	.map_alloc = htab_map_alloc,
+	.map_free = htab_map_free,
+	.map_get_next_key = htab_map_get_next_key,
+	.map_lookup_elem = htab_percpu_map_lookup_elem,
+	.map_update_elem = htab_percpu_map_update_elem,
+	.map_delete_elem = htab_map_delete_elem,
+};
+
+static struct bpf_map_type_list htab_percpu_type __read_mostly = {
+	.ops = &htab_percpu_ops,
+	.type = BPF_MAP_TYPE_PERCPU_HASH,
+};
+
 static int __init register_htab_map(void)
 {
 	bpf_register_map_type(&htab_type);
+	bpf_register_map_type(&htab_percpu_type);
 	return 0;
 }
 late_initcall(register_htab_map);
-- 
cgit v0.10.2


From a10423b87a7eae75da79ce80a8d9475047a674ee Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 1 Feb 2016 22:39:54 -0800
Subject: bpf: introduce BPF_MAP_TYPE_PERCPU_ARRAY map

Primary use case is a histogram array of latency
where bpf program computes the latency of block requests or other
events and stores histogram of latency into array of 64 elements.
All cpus are constantly running, so normal increment is not accurate,
bpf_xadd causes cache ping-pong and this per-cpu approach allows
fastest collision-free counters.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 83d1926..141fb0d4 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -151,6 +151,7 @@ struct bpf_array {
 	union {
 		char value[0] __aligned(8);
 		void *ptrs[0] __aligned(8);
+		void __percpu *pptrs[0] __aligned(8);
 	};
 };
 #define MAX_TAIL_CALL_CNT 32
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 43ae40c..2ee0fde 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -82,6 +82,7 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_PROG_ARRAY,
 	BPF_MAP_TYPE_PERF_EVENT_ARRAY,
 	BPF_MAP_TYPE_PERCPU_HASH,
+	BPF_MAP_TYPE_PERCPU_ARRAY,
 };
 
 enum bpf_prog_type {
diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 89ebbc4..b9bf1d7 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -17,11 +17,39 @@
 #include <linux/filter.h>
 #include <linux/perf_event.h>
 
+static void bpf_array_free_percpu(struct bpf_array *array)
+{
+	int i;
+
+	for (i = 0; i < array->map.max_entries; i++)
+		free_percpu(array->pptrs[i]);
+}
+
+static int bpf_array_alloc_percpu(struct bpf_array *array)
+{
+	void __percpu *ptr;
+	int i;
+
+	for (i = 0; i < array->map.max_entries; i++) {
+		ptr = __alloc_percpu_gfp(array->elem_size, 8,
+					 GFP_USER | __GFP_NOWARN);
+		if (!ptr) {
+			bpf_array_free_percpu(array);
+			return -ENOMEM;
+		}
+		array->pptrs[i] = ptr;
+	}
+
+	return 0;
+}
+
 /* Called from syscall */
 static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 {
+	bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_ARRAY;
 	struct bpf_array *array;
-	u32 elem_size, array_size;
+	u64 array_size;
+	u32 elem_size;
 
 	/* check sanity of attributes */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
@@ -36,12 +64,16 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 
 	elem_size = round_up(attr->value_size, 8);
 
-	/* check round_up into zero and u32 overflow */
-	if (elem_size == 0 ||
-	    attr->max_entries > (U32_MAX - PAGE_SIZE - sizeof(*array)) / elem_size)
+	array_size = sizeof(*array);
+	if (percpu)
+		array_size += (u64) attr->max_entries * sizeof(void *);
+	else
+		array_size += (u64) attr->max_entries * elem_size;
+
+	/* make sure there is no u32 overflow later in round_up() */
+	if (array_size >= U32_MAX - PAGE_SIZE)
 		return ERR_PTR(-ENOMEM);
 
-	array_size = sizeof(*array) + attr->max_entries * elem_size;
 
 	/* allocate all map elements and zero-initialize them */
 	array = kzalloc(array_size, GFP_USER | __GFP_NOWARN);
@@ -52,12 +84,25 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 	}
 
 	/* copy mandatory map attributes */
+	array->map.map_type = attr->map_type;
 	array->map.key_size = attr->key_size;
 	array->map.value_size = attr->value_size;
 	array->map.max_entries = attr->max_entries;
-	array->map.pages = round_up(array_size, PAGE_SIZE) >> PAGE_SHIFT;
 	array->elem_size = elem_size;
 
+	if (!percpu)
+		goto out;
+
+	array_size += (u64) attr->max_entries * elem_size * num_possible_cpus();
+
+	if (array_size >= U32_MAX - PAGE_SIZE ||
+	    elem_size > PCPU_MIN_UNIT_SIZE || bpf_array_alloc_percpu(array)) {
+		kvfree(array);
+		return ERR_PTR(-ENOMEM);
+	}
+out:
+	array->map.pages = round_up(array_size, PAGE_SIZE) >> PAGE_SHIFT;
+
 	return &array->map;
 }
 
@@ -67,12 +112,24 @@ static void *array_map_lookup_elem(struct bpf_map *map, void *key)
 	struct bpf_array *array = container_of(map, struct bpf_array, map);
 	u32 index = *(u32 *)key;
 
-	if (index >= array->map.max_entries)
+	if (unlikely(index >= array->map.max_entries))
 		return NULL;
 
 	return array->value + array->elem_size * index;
 }
 
+/* Called from eBPF program */
+static void *percpu_array_map_lookup_elem(struct bpf_map *map, void *key)
+{
+	struct bpf_array *array = container_of(map, struct bpf_array, map);
+	u32 index = *(u32 *)key;
+
+	if (unlikely(index >= array->map.max_entries))
+		return NULL;
+
+	return this_cpu_ptr(array->pptrs[index]);
+}
+
 /* Called from syscall */
 static int array_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 {
@@ -99,19 +156,24 @@ static int array_map_update_elem(struct bpf_map *map, void *key, void *value,
 	struct bpf_array *array = container_of(map, struct bpf_array, map);
 	u32 index = *(u32 *)key;
 
-	if (map_flags > BPF_EXIST)
+	if (unlikely(map_flags > BPF_EXIST))
 		/* unknown flags */
 		return -EINVAL;
 
-	if (index >= array->map.max_entries)
+	if (unlikely(index >= array->map.max_entries))
 		/* all elements were pre-allocated, cannot insert a new one */
 		return -E2BIG;
 
-	if (map_flags == BPF_NOEXIST)
+	if (unlikely(map_flags == BPF_NOEXIST))
 		/* all elements already exist */
 		return -EEXIST;
 
-	memcpy(array->value + array->elem_size * index, value, map->value_size);
+	if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY)
+		memcpy(this_cpu_ptr(array->pptrs[index]),
+		       value, map->value_size);
+	else
+		memcpy(array->value + array->elem_size * index,
+		       value, map->value_size);
 	return 0;
 }
 
@@ -133,6 +195,9 @@ static void array_map_free(struct bpf_map *map)
 	 */
 	synchronize_rcu();
 
+	if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY)
+		bpf_array_free_percpu(array);
+
 	kvfree(array);
 }
 
@@ -150,9 +215,24 @@ static struct bpf_map_type_list array_type __read_mostly = {
 	.type = BPF_MAP_TYPE_ARRAY,
 };
 
+static const struct bpf_map_ops percpu_array_ops = {
+	.map_alloc = array_map_alloc,
+	.map_free = array_map_free,
+	.map_get_next_key = array_map_get_next_key,
+	.map_lookup_elem = percpu_array_map_lookup_elem,
+	.map_update_elem = array_map_update_elem,
+	.map_delete_elem = array_map_delete_elem,
+};
+
+static struct bpf_map_type_list percpu_array_type __read_mostly = {
+	.ops = &percpu_array_ops,
+	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
+};
+
 static int __init register_array_map(void)
 {
 	bpf_register_map_type(&array_type);
+	bpf_register_map_type(&percpu_array_type);
 	return 0;
 }
 late_initcall(register_array_map);
-- 
cgit v0.10.2


From 15a07b33814d14ca817887dbea8530728dc0fbe4 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 1 Feb 2016 22:39:55 -0800
Subject: bpf: add lookup/update support for per-cpu hash and array maps

The functions bpf_map_lookup_elem(map, key, value) and
bpf_map_update_elem(map, key, value, flags) need to get/set
values from all-cpus for per-cpu hash and array maps,
so that user space can aggregate/update them as necessary.

Example of single counter aggregation in user space:
  unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
  long values[nr_cpus];
  long value = 0;

  bpf_lookup_elem(fd, key, values);
  for (i = 0; i < nr_cpus; i++)
    value += values[i];

The user space must provide round_up(value_size, 8) * nr_cpus
array to get/set values, since kernel will use 'long' copy
of per-cpu values to try to copy good counters atomically.
It's a best-effort, since bpf programs and user space are racing
to access the same memory.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 141fb0d4..90ee6ab 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -183,6 +183,29 @@ int bpf_prog_new_fd(struct bpf_prog *prog);
 int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
 int bpf_obj_get_user(const char __user *pathname);
 
+int bpf_percpu_hash_copy(struct bpf_map *map, void *key, void *value);
+int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value);
+int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value,
+			   u64 flags);
+int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value,
+			    u64 flags);
+
+/* memcpy that is used with 8-byte aligned pointers, power-of-8 size and
+ * forced to use 'long' read/writes to try to atomically copy long counters.
+ * Best-effort only.  No barriers here, since it _will_ race with concurrent
+ * updates from BPF programs. Called from bpf syscall and mostly used with
+ * size 8 or 16 bytes, so ask compiler to inline it.
+ */
+static inline void bpf_long_memcpy(void *dst, const void *src, u32 size)
+{
+	const long *lsrc = src;
+	long *ldst = dst;
+
+	size /= sizeof(long);
+	while (size--)
+		*ldst++ = *lsrc++;
+}
+
 /* verify correctness of eBPF program */
 int bpf_check(struct bpf_prog **fp, union bpf_attr *attr);
 #else
diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index b9bf1d7..bd3bdf2 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -130,6 +130,32 @@ static void *percpu_array_map_lookup_elem(struct bpf_map *map, void *key)
 	return this_cpu_ptr(array->pptrs[index]);
 }
 
+int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value)
+{
+	struct bpf_array *array = container_of(map, struct bpf_array, map);
+	u32 index = *(u32 *)key;
+	void __percpu *pptr;
+	int cpu, off = 0;
+	u32 size;
+
+	if (unlikely(index >= array->map.max_entries))
+		return -ENOENT;
+
+	/* per_cpu areas are zero-filled and bpf programs can only
+	 * access 'value_size' of them, so copying rounded areas
+	 * will not leak any kernel data
+	 */
+	size = round_up(map->value_size, 8);
+	rcu_read_lock();
+	pptr = array->pptrs[index];
+	for_each_possible_cpu(cpu) {
+		bpf_long_memcpy(value + off, per_cpu_ptr(pptr, cpu), size);
+		off += size;
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
 /* Called from syscall */
 static int array_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 {
@@ -177,6 +203,44 @@ static int array_map_update_elem(struct bpf_map *map, void *key, void *value,
 	return 0;
 }
 
+int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value,
+			    u64 map_flags)
+{
+	struct bpf_array *array = container_of(map, struct bpf_array, map);
+	u32 index = *(u32 *)key;
+	void __percpu *pptr;
+	int cpu, off = 0;
+	u32 size;
+
+	if (unlikely(map_flags > BPF_EXIST))
+		/* unknown flags */
+		return -EINVAL;
+
+	if (unlikely(index >= array->map.max_entries))
+		/* all elements were pre-allocated, cannot insert a new one */
+		return -E2BIG;
+
+	if (unlikely(map_flags == BPF_NOEXIST))
+		/* all elements already exist */
+		return -EEXIST;
+
+	/* the user space will provide round_up(value_size, 8) bytes that
+	 * will be copied into per-cpu area. bpf programs can only access
+	 * value_size of it. During lookup the same extra bytes will be
+	 * returned or zeros which were zero-filled by percpu_alloc,
+	 * so no kernel data leaks possible
+	 */
+	size = round_up(map->value_size, 8);
+	rcu_read_lock();
+	pptr = array->pptrs[index];
+	for_each_possible_cpu(cpu) {
+		bpf_long_memcpy(per_cpu_ptr(pptr, cpu), value + off, size);
+		off += size;
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
 /* Called from syscall or from eBPF program */
 static int array_map_delete_elem(struct bpf_map *map, void *key)
 {
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 2be5f6e..fd5db8f 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -290,7 +290,7 @@ static void free_htab_elem(struct htab_elem *l, bool percpu, u32 key_size)
 
 static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 					 void *value, u32 key_size, u32 hash,
-					 bool percpu)
+					 bool percpu, bool onallcpus)
 {
 	u32 size = htab->map.value_size;
 	struct htab_elem *l_new;
@@ -312,8 +312,18 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 			return NULL;
 		}
 
-		/* copy true value_size bytes */
-		memcpy(this_cpu_ptr(pptr), value, htab->map.value_size);
+		if (!onallcpus) {
+			/* copy true value_size bytes */
+			memcpy(this_cpu_ptr(pptr), value, htab->map.value_size);
+		} else {
+			int off = 0, cpu;
+
+			for_each_possible_cpu(cpu) {
+				bpf_long_memcpy(per_cpu_ptr(pptr, cpu),
+						value + off, size);
+				off += size;
+			}
+		}
 		htab_elem_set_ptr(l_new, key_size, pptr);
 	} else {
 		memcpy(l_new->key + round_up(key_size, 8), value, size);
@@ -368,7 +378,7 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 	/* allocate new element outside of the lock, since
 	 * we're most likley going to insert it
 	 */
-	l_new = alloc_htab_elem(htab, key, value, key_size, hash, false);
+	l_new = alloc_htab_elem(htab, key, value, key_size, hash, false, false);
 	if (!l_new)
 		return -ENOMEM;
 
@@ -402,8 +412,9 @@ err:
 	return ret;
 }
 
-static int htab_percpu_map_update_elem(struct bpf_map *map, void *key,
-				       void *value, u64 map_flags)
+static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key,
+					 void *value, u64 map_flags,
+					 bool onallcpus)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct htab_elem *l_new = NULL, *l_old;
@@ -436,12 +447,25 @@ static int htab_percpu_map_update_elem(struct bpf_map *map, void *key,
 		goto err;
 
 	if (l_old) {
+		void __percpu *pptr = htab_elem_get_ptr(l_old, key_size);
+		u32 size = htab->map.value_size;
+
 		/* per-cpu hash map can update value in-place */
-		memcpy(this_cpu_ptr(htab_elem_get_ptr(l_old, key_size)),
-		       value, htab->map.value_size);
+		if (!onallcpus) {
+			memcpy(this_cpu_ptr(pptr), value, size);
+		} else {
+			int off = 0, cpu;
+
+			size = round_up(size, 8);
+			for_each_possible_cpu(cpu) {
+				bpf_long_memcpy(per_cpu_ptr(pptr, cpu),
+						value + off, size);
+				off += size;
+			}
+		}
 	} else {
 		l_new = alloc_htab_elem(htab, key, value, key_size,
-					hash, true);
+					hash, true, onallcpus);
 		if (!l_new) {
 			ret = -ENOMEM;
 			goto err;
@@ -455,6 +479,12 @@ err:
 	return ret;
 }
 
+static int htab_percpu_map_update_elem(struct bpf_map *map, void *key,
+				       void *value, u64 map_flags)
+{
+	return __htab_percpu_map_update_elem(map, key, value, map_flags, false);
+}
+
 /* Called from syscall or from eBPF program */
 static int htab_map_delete_elem(struct bpf_map *map, void *key)
 {
@@ -557,6 +587,41 @@ static void *htab_percpu_map_lookup_elem(struct bpf_map *map, void *key)
 		return NULL;
 }
 
+int bpf_percpu_hash_copy(struct bpf_map *map, void *key, void *value)
+{
+	struct htab_elem *l;
+	void __percpu *pptr;
+	int ret = -ENOENT;
+	int cpu, off = 0;
+	u32 size;
+
+	/* per_cpu areas are zero-filled and bpf programs can only
+	 * access 'value_size' of them, so copying rounded areas
+	 * will not leak any kernel data
+	 */
+	size = round_up(map->value_size, 8);
+	rcu_read_lock();
+	l = __htab_map_lookup_elem(map, key);
+	if (!l)
+		goto out;
+	pptr = htab_elem_get_ptr(l, map->key_size);
+	for_each_possible_cpu(cpu) {
+		bpf_long_memcpy(value + off,
+				per_cpu_ptr(pptr, cpu), size);
+		off += size;
+	}
+	ret = 0;
+out:
+	rcu_read_unlock();
+	return ret;
+}
+
+int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value,
+			   u64 map_flags)
+{
+	return __htab_percpu_map_update_elem(map, key, value, map_flags, true);
+}
+
 static const struct bpf_map_ops htab_percpu_ops = {
 	.map_alloc = htab_map_alloc,
 	.map_free = htab_map_free,
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 6373970..c95a753 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -239,6 +239,7 @@ static int map_lookup_elem(union bpf_attr *attr)
 	int ufd = attr->map_fd;
 	struct bpf_map *map;
 	void *key, *value, *ptr;
+	u32 value_size;
 	struct fd f;
 	int err;
 
@@ -259,23 +260,35 @@ static int map_lookup_elem(union bpf_attr *attr)
 	if (copy_from_user(key, ukey, map->key_size) != 0)
 		goto free_key;
 
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
+	    map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY)
+		value_size = round_up(map->value_size, 8) * num_possible_cpus();
+	else
+		value_size = map->value_size;
+
 	err = -ENOMEM;
-	value = kmalloc(map->value_size, GFP_USER | __GFP_NOWARN);
+	value = kmalloc(value_size, GFP_USER | __GFP_NOWARN);
 	if (!value)
 		goto free_key;
 
-	rcu_read_lock();
-	ptr = map->ops->map_lookup_elem(map, key);
-	if (ptr)
-		memcpy(value, ptr, map->value_size);
-	rcu_read_unlock();
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH) {
+		err = bpf_percpu_hash_copy(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
+		err = bpf_percpu_array_copy(map, key, value);
+	} else {
+		rcu_read_lock();
+		ptr = map->ops->map_lookup_elem(map, key);
+		if (ptr)
+			memcpy(value, ptr, value_size);
+		rcu_read_unlock();
+		err = ptr ? 0 : -ENOENT;
+	}
 
-	err = -ENOENT;
-	if (!ptr)
+	if (err)
 		goto free_value;
 
 	err = -EFAULT;
-	if (copy_to_user(uvalue, value, map->value_size) != 0)
+	if (copy_to_user(uvalue, value, value_size) != 0)
 		goto free_value;
 
 	err = 0;
@@ -298,6 +311,7 @@ static int map_update_elem(union bpf_attr *attr)
 	int ufd = attr->map_fd;
 	struct bpf_map *map;
 	void *key, *value;
+	u32 value_size;
 	struct fd f;
 	int err;
 
@@ -318,21 +332,30 @@ static int map_update_elem(union bpf_attr *attr)
 	if (copy_from_user(key, ukey, map->key_size) != 0)
 		goto free_key;
 
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
+	    map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY)
+		value_size = round_up(map->value_size, 8) * num_possible_cpus();
+	else
+		value_size = map->value_size;
+
 	err = -ENOMEM;
-	value = kmalloc(map->value_size, GFP_USER | __GFP_NOWARN);
+	value = kmalloc(value_size, GFP_USER | __GFP_NOWARN);
 	if (!value)
 		goto free_key;
 
 	err = -EFAULT;
-	if (copy_from_user(value, uvalue, map->value_size) != 0)
+	if (copy_from_user(value, uvalue, value_size) != 0)
 		goto free_value;
 
-	/* eBPF program that use maps are running under rcu_read_lock(),
-	 * therefore all map accessors rely on this fact, so do the same here
-	 */
-	rcu_read_lock();
-	err = map->ops->map_update_elem(map, key, value, attr->flags);
-	rcu_read_unlock();
+	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH) {
+		err = bpf_percpu_hash_update(map, key, value, attr->flags);
+	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
+		err = bpf_percpu_array_update(map, key, value, attr->flags);
+	} else {
+		rcu_read_lock();
+		err = map->ops->map_update_elem(map, key, value, attr->flags);
+		rcu_read_unlock();
+	}
 
 free_value:
 	kfree(value);
-- 
cgit v0.10.2


From e15596717948c95587a0b15363030283c126c23a Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Mon, 1 Feb 2016 22:39:56 -0800
Subject: samples/bpf: unit test for BPF_MAP_TYPE_PERCPU_HASH

A sanity test for BPF_MAP_TYPE_PERCPU_HASH.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/test_maps.c b/samples/bpf/test_maps.c
index 6299ee9..5f5fe53 100644
--- a/samples/bpf/test_maps.c
+++ b/samples/bpf/test_maps.c
@@ -89,6 +89,100 @@ static void test_hashmap_sanity(int i, void *data)
 	close(map_fd);
 }
 
+/* sanity tests for percpu map API */
+static void test_percpu_hashmap_sanity(int task, void *data)
+{
+	long long key, next_key;
+	int expected_key_mask = 0;
+	unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+	long long value[nr_cpus];
+	int map_fd, i;
+
+	map_fd = bpf_create_map(BPF_MAP_TYPE_PERCPU_HASH, sizeof(key),
+				sizeof(value[0]), 2);
+	if (map_fd < 0) {
+		printf("failed to create hashmap '%s'\n", strerror(errno));
+		exit(1);
+	}
+
+	for (i = 0; i < nr_cpus; i++)
+		value[i] = i + 100;
+	key = 1;
+	/* insert key=1 element */
+	assert(!(expected_key_mask & key));
+	assert(bpf_update_elem(map_fd, &key, value, BPF_ANY) == 0);
+	expected_key_mask |= key;
+
+	/* BPF_NOEXIST means: add new element if it doesn't exist */
+	assert(bpf_update_elem(map_fd, &key, value, BPF_NOEXIST) == -1 &&
+	       /* key=1 already exists */
+	       errno == EEXIST);
+
+	/* -1 is an invalid flag */
+	assert(bpf_update_elem(map_fd, &key, value, -1) == -1 &&
+	       errno == EINVAL);
+
+	/* check that key=1 can be found. value could be 0 if the lookup
+	 * was run from a different cpu.
+	 */
+	value[0] = 1;
+	assert(bpf_lookup_elem(map_fd, &key, value) == 0 && value[0] == 100);
+
+	key = 2;
+	/* check that key=2 is not found */
+	assert(bpf_lookup_elem(map_fd, &key, value) == -1 && errno == ENOENT);
+
+	/* BPF_EXIST means: update existing element */
+	assert(bpf_update_elem(map_fd, &key, value, BPF_EXIST) == -1 &&
+	       /* key=2 is not there */
+	       errno == ENOENT);
+
+	/* insert key=2 element */
+	assert(!(expected_key_mask & key));
+	assert(bpf_update_elem(map_fd, &key, value, BPF_NOEXIST) == 0);
+	expected_key_mask |= key;
+
+	/* key=1 and key=2 were inserted, check that key=0 cannot be inserted
+	 * due to max_entries limit
+	 */
+	key = 0;
+	assert(bpf_update_elem(map_fd, &key, value, BPF_NOEXIST) == -1 &&
+	       errno == E2BIG);
+
+	/* check that key = 0 doesn't exist */
+	assert(bpf_delete_elem(map_fd, &key) == -1 && errno == ENOENT);
+
+	/* iterate over two elements */
+	while (!bpf_get_next_key(map_fd, &key, &next_key)) {
+		assert((expected_key_mask & next_key) == next_key);
+		expected_key_mask &= ~next_key;
+
+		assert(bpf_lookup_elem(map_fd, &next_key, value) == 0);
+		for (i = 0; i < nr_cpus; i++)
+			assert(value[i] == i + 100);
+
+		key = next_key;
+	}
+	assert(errno == ENOENT);
+
+	/* Update with BPF_EXIST */
+	key = 1;
+	assert(bpf_update_elem(map_fd, &key, value, BPF_EXIST) == 0);
+
+	/* delete both elements */
+	key = 1;
+	assert(bpf_delete_elem(map_fd, &key) == 0);
+	key = 2;
+	assert(bpf_delete_elem(map_fd, &key) == 0);
+	assert(bpf_delete_elem(map_fd, &key) == -1 && errno == ENOENT);
+
+	key = 0;
+	/* check that map is empty */
+	assert(bpf_get_next_key(map_fd, &key, &next_key) == -1 &&
+	       errno == ENOENT);
+	close(map_fd);
+}
+
 static void test_arraymap_sanity(int i, void *data)
 {
 	int key, next_key, map_fd;
@@ -209,6 +303,7 @@ static void run_parallel(int tasks, void (*fn)(int i, void *data), void *data)
 static void test_map_stress(void)
 {
 	run_parallel(100, test_hashmap_sanity, NULL);
+	run_parallel(100, test_percpu_hashmap_sanity, NULL);
 	run_parallel(100, test_arraymap_sanity, NULL);
 }
 
@@ -282,6 +377,7 @@ static void test_map_parallel(void)
 int main(void)
 {
 	test_hashmap_sanity(0, NULL);
+	test_percpu_hashmap_sanity(0, NULL);
 	test_arraymap_sanity(0, NULL);
 	test_map_large();
 	test_map_parallel();
-- 
cgit v0.10.2


From df570f577231407d929bdc6f59ae2f53e0028e8a Mon Sep 17 00:00:00 2001
From: "tom.leiming@gmail.com" <tom.leiming@gmail.com>
Date: Mon, 1 Feb 2016 22:39:57 -0800
Subject: samples/bpf: unit test for BPF_MAP_TYPE_PERCPU_ARRAY

A sanity test for BPF_MAP_TYPE_PERCPU_ARRAY

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/test_maps.c b/samples/bpf/test_maps.c
index 5f5fe53..ad466ed 100644
--- a/samples/bpf/test_maps.c
+++ b/samples/bpf/test_maps.c
@@ -236,6 +236,94 @@ static void test_arraymap_sanity(int i, void *data)
 	close(map_fd);
 }
 
+static void test_percpu_arraymap_many_keys(void)
+{
+	unsigned nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+	unsigned nr_keys = 20000;
+	long values[nr_cpus];
+	int key, map_fd, i;
+
+	map_fd = bpf_create_map(BPF_MAP_TYPE_PERCPU_ARRAY, sizeof(key),
+				sizeof(values[0]), nr_keys);
+	if (map_fd < 0) {
+		printf("failed to create per-cpu arraymap '%s'\n",
+		       strerror(errno));
+		exit(1);
+	}
+
+	for (i = 0; i < nr_cpus; i++)
+		values[i] = i + 10;
+
+	for (key = 0; key < nr_keys; key++)
+		assert(bpf_update_elem(map_fd, &key, values, BPF_ANY) == 0);
+
+	for (key = 0; key < nr_keys; key++) {
+		for (i = 0; i < nr_cpus; i++)
+			values[i] = 0;
+		assert(bpf_lookup_elem(map_fd, &key, values) == 0);
+		for (i = 0; i < nr_cpus; i++)
+			assert(values[i] == i + 10);
+	}
+
+	close(map_fd);
+}
+
+static void test_percpu_arraymap_sanity(int i, void *data)
+{
+	unsigned nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+	long values[nr_cpus];
+	int key, next_key, map_fd;
+
+	map_fd = bpf_create_map(BPF_MAP_TYPE_PERCPU_ARRAY, sizeof(key),
+				sizeof(values[0]), 2);
+	if (map_fd < 0) {
+		printf("failed to create arraymap '%s'\n", strerror(errno));
+		exit(1);
+	}
+
+	for (i = 0; i < nr_cpus; i++)
+		values[i] = i + 100;
+
+	key = 1;
+	/* insert key=1 element */
+	assert(bpf_update_elem(map_fd, &key, values, BPF_ANY) == 0);
+
+	values[0] = 0;
+	assert(bpf_update_elem(map_fd, &key, values, BPF_NOEXIST) == -1 &&
+	       errno == EEXIST);
+
+	/* check that key=1 can be found */
+	assert(bpf_lookup_elem(map_fd, &key, values) == 0 && values[0] == 100);
+
+	key = 0;
+	/* check that key=0 is also found and zero initialized */
+	assert(bpf_lookup_elem(map_fd, &key, values) == 0 &&
+	       values[0] == 0 && values[nr_cpus - 1] == 0);
+
+
+	/* check that key=2 cannot be inserted due to max_entries limit */
+	key = 2;
+	assert(bpf_update_elem(map_fd, &key, values, BPF_EXIST) == -1 &&
+	       errno == E2BIG);
+
+	/* check that key = 2 doesn't exist */
+	assert(bpf_lookup_elem(map_fd, &key, values) == -1 && errno == ENOENT);
+
+	/* iterate over two elements */
+	assert(bpf_get_next_key(map_fd, &key, &next_key) == 0 &&
+	       next_key == 0);
+	assert(bpf_get_next_key(map_fd, &next_key, &next_key) == 0 &&
+	       next_key == 1);
+	assert(bpf_get_next_key(map_fd, &next_key, &next_key) == -1 &&
+	       errno == ENOENT);
+
+	/* delete shouldn't succeed */
+	key = 1;
+	assert(bpf_delete_elem(map_fd, &key) == -1 && errno == EINVAL);
+
+	close(map_fd);
+}
+
 #define MAP_SIZE (32 * 1024)
 static void test_map_large(void)
 {
@@ -305,6 +393,7 @@ static void test_map_stress(void)
 	run_parallel(100, test_hashmap_sanity, NULL);
 	run_parallel(100, test_percpu_hashmap_sanity, NULL);
 	run_parallel(100, test_arraymap_sanity, NULL);
+	run_parallel(100, test_percpu_arraymap_sanity, NULL);
 }
 
 #define TASKS 1024
@@ -379,6 +468,9 @@ int main(void)
 	test_hashmap_sanity(0, NULL);
 	test_percpu_hashmap_sanity(0, NULL);
 	test_arraymap_sanity(0, NULL);
+	test_percpu_arraymap_sanity(0, NULL);
+	test_percpu_arraymap_many_keys();
+
 	test_map_large();
 	test_map_parallel();
 	test_map_stress();
-- 
cgit v0.10.2


From 3059303f59cf90a84e7fdef154ff0b215bcfaa97 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 1 Feb 2016 22:39:58 -0800
Subject: samples/bpf: update tracex[23] examples to use per-cpu maps

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/tracex2_kern.c b/samples/bpf/tracex2_kern.c
index b32367c..09c1adc 100644
--- a/samples/bpf/tracex2_kern.c
+++ b/samples/bpf/tracex2_kern.c
@@ -70,7 +70,7 @@ struct hist_key {
 };
 
 struct bpf_map_def SEC("maps") my_hist_map = {
-	.type = BPF_MAP_TYPE_HASH,
+	.type = BPF_MAP_TYPE_PERCPU_HASH,
 	.key_size = sizeof(struct hist_key),
 	.value_size = sizeof(long),
 	.max_entries = 1024,
diff --git a/samples/bpf/tracex2_user.c b/samples/bpf/tracex2_user.c
index cd0241c..ab5b19e 100644
--- a/samples/bpf/tracex2_user.c
+++ b/samples/bpf/tracex2_user.c
@@ -37,6 +37,8 @@ struct hist_key {
 static void print_hist_for_pid(int fd, void *task)
 {
 	struct hist_key key = {}, next_key;
+	unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+	long values[nr_cpus];
 	char starstr[MAX_STARS];
 	long value;
 	long data[MAX_INDEX] = {};
@@ -49,7 +51,10 @@ static void print_hist_for_pid(int fd, void *task)
 			key = next_key;
 			continue;
 		}
-		bpf_lookup_elem(fd, &next_key, &value);
+		bpf_lookup_elem(fd, &next_key, values);
+		value = 0;
+		for (i = 0; i < nr_cpus; i++)
+			value += values[i];
 		ind = next_key.index;
 		data[ind] = value;
 		if (value && ind > max_ind)
diff --git a/samples/bpf/tracex3_kern.c b/samples/bpf/tracex3_kern.c
index bf337fb..9974c3d 100644
--- a/samples/bpf/tracex3_kern.c
+++ b/samples/bpf/tracex3_kern.c
@@ -20,7 +20,7 @@ struct bpf_map_def SEC("maps") my_map = {
 /* kprobe is NOT a stable ABI. If kernel internals change this bpf+kprobe
  * example will no longer be meaningful
  */
-SEC("kprobe/blk_mq_start_request")
+SEC("kprobe/blk_start_request")
 int bpf_prog1(struct pt_regs *ctx)
 {
 	long rq = PT_REGS_PARM1(ctx);
@@ -42,13 +42,13 @@ static unsigned int log2l(unsigned long long n)
 #define SLOTS 100
 
 struct bpf_map_def SEC("maps") lat_map = {
-	.type = BPF_MAP_TYPE_ARRAY,
+	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(u64),
 	.max_entries = SLOTS,
 };
 
-SEC("kprobe/blk_update_request")
+SEC("kprobe/blk_account_io_completion")
 int bpf_prog2(struct pt_regs *ctx)
 {
 	long rq = PT_REGS_PARM1(ctx);
@@ -81,7 +81,7 @@ int bpf_prog2(struct pt_regs *ctx)
 
 	value = bpf_map_lookup_elem(&lat_map, &index);
 	if (value)
-		__sync_fetch_and_add((long *)value, 1);
+		*value += 1;
 
 	return 0;
 }
diff --git a/samples/bpf/tracex3_user.c b/samples/bpf/tracex3_user.c
index 0aaa933..48716f7 100644
--- a/samples/bpf/tracex3_user.c
+++ b/samples/bpf/tracex3_user.c
@@ -20,11 +20,13 @@
 
 static void clear_stats(int fd)
 {
+	unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+	__u64 values[nr_cpus];
 	__u32 key;
-	__u64 value = 0;
 
+	memset(values, 0, sizeof(values));
 	for (key = 0; key < SLOTS; key++)
-		bpf_update_elem(fd, &key, &value, BPF_ANY);
+		bpf_update_elem(fd, &key, values, BPF_ANY);
 }
 
 const char *color[] = {
@@ -75,15 +77,20 @@ static void print_banner(void)
 
 static void print_hist(int fd)
 {
-	__u32 key;
-	__u64 value;
-	__u64 cnt[SLOTS];
-	__u64 max_cnt = 0;
+	unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
 	__u64 total_events = 0;
+	long values[nr_cpus];
+	__u64 max_cnt = 0;
+	__u64 cnt[SLOTS];
+	__u64 value;
+	__u32 key;
+	int i;
 
 	for (key = 0; key < SLOTS; key++) {
+		bpf_lookup_elem(fd, &key, values);
 		value = 0;
-		bpf_lookup_elem(fd, &key, &value);
+		for (i = 0; i < nr_cpus; i++)
+			value += values[i];
 		cnt[key] = value;
 		total_events += value;
 		if (value > max_cnt)
-- 
cgit v0.10.2


From 387b75f8b31437792e8334390fdf5cf060d1e3da Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 2 Feb 2016 07:47:14 +0100
Subject: bgmac: add helper checking for BCM4707 / BCM53018 chip id
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Chipsets with BCM4707 / BCM53018 ID require special handling at a few
places in the code. It's likely there will be more IDs to check in the
future. To simplify it add this trivial helper.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index 06f6cff..230f8e6 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -26,6 +26,17 @@ static const struct bcma_device_id bgmac_bcma_tbl[] = {
 };
 MODULE_DEVICE_TABLE(bcma, bgmac_bcma_tbl);
 
+static inline bool bgmac_is_bcm4707_family(struct bgmac *bgmac)
+{
+	switch (bgmac->core->bus->chipinfo.id) {
+	case BCMA_CHIP_ID_BCM4707:
+	case BCMA_CHIP_ID_BCM53018:
+		return true;
+	default:
+		return false;
+	}
+}
+
 static bool bgmac_wait_value(struct bcma_device *core, u16 reg, u32 mask,
 			     u32 value, int timeout)
 {
@@ -987,11 +998,9 @@ static void bgmac_mac_speed(struct bgmac *bgmac)
 static void bgmac_miiconfig(struct bgmac *bgmac)
 {
 	struct bcma_device *core = bgmac->core;
-	struct bcma_chipinfo *ci = &core->bus->chipinfo;
 	u8 imode;
 
-	if (ci->id == BCMA_CHIP_ID_BCM4707 ||
-	    ci->id == BCMA_CHIP_ID_BCM53018) {
+	if (bgmac_is_bcm4707_family(bgmac)) {
 		bcma_awrite32(core, BCMA_IOCTL,
 			      bcma_aread32(core, BCMA_IOCTL) | 0x40 |
 			      BGMAC_BCMA_IOCTL_SW_CLKEN);
@@ -1055,9 +1064,7 @@ static void bgmac_chip_reset(struct bgmac *bgmac)
 	}
 
 	/* Request Misc PLL for corerev > 2 */
-	if (core->id.rev > 2 &&
-	    ci->id != BCMA_CHIP_ID_BCM4707 &&
-	    ci->id != BCMA_CHIP_ID_BCM53018) {
+	if (core->id.rev > 2 && !bgmac_is_bcm4707_family(bgmac)) {
 		bgmac_set(bgmac, BCMA_CLKCTLST,
 			  BGMAC_BCMA_CLKCTLST_MISC_PLL_REQ);
 		bgmac_wait_value(bgmac->core, BCMA_CLKCTLST,
@@ -1193,8 +1200,7 @@ static void bgmac_enable(struct bgmac *bgmac)
 		break;
 	}
 
-	if (ci->id != BCMA_CHIP_ID_BCM4707 &&
-	    ci->id != BCMA_CHIP_ID_BCM53018) {
+	if (!bgmac_is_bcm4707_family(bgmac)) {
 		rxq_ctl = bgmac_read(bgmac, BGMAC_RXQ_CTL);
 		rxq_ctl &= ~BGMAC_RXQ_CTL_MDP_MASK;
 		bp_clk = bcma_pmu_get_bus_clock(&bgmac->core->bus->drv_cc) /
@@ -1472,14 +1478,12 @@ static int bgmac_fixed_phy_register(struct bgmac *bgmac)
 
 static int bgmac_mii_register(struct bgmac *bgmac)
 {
-	struct bcma_chipinfo *ci = &bgmac->core->bus->chipinfo;
 	struct mii_bus *mii_bus;
 	struct phy_device *phy_dev;
 	char bus_id[MII_BUS_ID_SIZE + 3];
 	int err = 0;
 
-	if (ci->id == BCMA_CHIP_ID_BCM4707 ||
-	    ci->id == BCMA_CHIP_ID_BCM53018)
+	if (bgmac_is_bcm4707_family(bgmac))
 		return bgmac_fixed_phy_register(bgmac);
 
 	mii_bus = mdiobus_alloc();
@@ -1539,7 +1543,6 @@ static void bgmac_mii_unregister(struct bgmac *bgmac)
 /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipattach */
 static int bgmac_probe(struct bcma_device *core)
 {
-	struct bcma_chipinfo *ci = &core->bus->chipinfo;
 	struct net_device *net_dev;
 	struct bgmac *bgmac;
 	struct ssb_sprom *sprom = &core->bus->sprom;
@@ -1620,8 +1623,7 @@ static int bgmac_probe(struct bcma_device *core)
 	bgmac_chip_reset(bgmac);
 
 	/* For Northstar, we have to take all GMAC core out of reset */
-	if (ci->id == BCMA_CHIP_ID_BCM4707 ||
-	    ci->id == BCMA_CHIP_ID_BCM53018) {
+	if (bgmac_is_bcm4707_family(bgmac)) {
 		struct bcma_device *ns_core;
 		int ns_gmac;
 
-- 
cgit v0.10.2


From c6140a299bdf2425d092383fafaaf2a4a3745989 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Tue, 2 Feb 2016 08:32:55 +0000
Subject: bonding: add slave device name for debug

netdev_dbg() will add bond device name, it will be helpful if we print
slave device name.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 6587929..7c9eb67 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -618,8 +618,8 @@ static void bond_hw_addr_swap(struct bonding *bond, struct slave *new_active,
 static void bond_set_dev_addr(struct net_device *bond_dev,
 			      struct net_device *slave_dev)
 {
-	netdev_dbg(bond_dev, "bond_dev=%p slave_dev=%p slave_dev->addr_len=%d\n",
-		   bond_dev, slave_dev, slave_dev->addr_len);
+	netdev_dbg(bond_dev, "bond_dev=%p slave_dev=%p slave_dev->name=%s slave_dev->addr_len=%d\n",
+		   bond_dev, slave_dev, slave_dev->name, slave_dev->addr_len);
 	memcpy(bond_dev->dev_addr, slave_dev->dev_addr, slave_dev->addr_len);
 	bond_dev->addr_assign_type = NET_ADDR_STOLEN;
 	call_netdevice_notifiers(NETDEV_CHANGEADDR, bond_dev);
-- 
cgit v0.10.2


From 4f61d4ef7088581997ad95ad37db2d8a459844e4 Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:08 +0100
Subject: tipc: remove incorrect check for subscription timeout value

Until now, during subscription creation we set sub->timeout by
converting the timeout request value in milliseconds to jiffies.
This is followed by setting the timeout value in the timer if
sub->timeout != TIPC_WAIT_FOREVER.

For a subscription create request with a timeout value of
TIPC_WAIT_FOREVER, msecs_to_jiffies(TIPC_WAIT_FOREVER)
returns MAX_JIFFY_OFFSET (0xfffffffe). This is not equal to
TIPC_WAIT_FOREVER (0xffffffff).

In this commit, we remove this check.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 69ee2ee..7e61a90 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -270,8 +270,7 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 	memcpy(&sub->evt.s, s, sizeof(*s));
 	atomic_inc(&tn->subscription_count);
 	setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub);
-	if (sub->timeout != TIPC_WAIT_FOREVER)
-		sub->timeout += jiffies;
+	sub->timeout += jiffies;
 	if (!mod_timer(&sub->timer, sub->timeout))
 		tipc_subscrb_get(subscriber);
 	*sub_p = sub;
-- 
cgit v0.10.2


From 3086523149ef4c15f5e75bb2ed9f43a8f3f95015 Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:09 +0100
Subject: tipc: remove filter and timeout elements from struct
 tipc_subscription

Until now, struct tipc_subscription has duplicate timeout and filter
attributes present:
1. directly as members of struct tipc_subscription
2. in struct tipc_subscr, which is contained in struct tipc_event

In this commit, we remove the references to these elements as
members of struct tipc_subscription and replace them with elements
from struct tipc_subscr.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 7e61a90..cef0b6b 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -110,7 +110,8 @@ void tipc_subscrp_report_overlap(struct tipc_subscription *sub, u32 found_lower,
 {
 	if (!tipc_subscrp_check_overlap(sub, found_lower, found_upper))
 		return;
-	if (!must && !(sub->filter & TIPC_SUB_PORTS))
+	if (!must &&
+	    !(htohl(sub->evt.s.filter, sub->swap) & TIPC_SUB_PORTS))
 		return;
 
 	tipc_subscrp_send_event(sub, found_lower, found_upper, event, port_ref,
@@ -222,6 +223,7 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 {
 	struct tipc_net *tn = net_generic(net, tipc_net_id);
 	struct tipc_subscription *sub;
+	u32 timeout, filter;
 	int swap;
 
 	/* Determine subscriber's endianness */
@@ -253,10 +255,8 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 	sub->seq.type = htohl(s->seq.type, swap);
 	sub->seq.lower = htohl(s->seq.lower, swap);
 	sub->seq.upper = htohl(s->seq.upper, swap);
-	sub->timeout = msecs_to_jiffies(htohl(s->timeout, swap));
-	sub->filter = htohl(s->filter, swap);
-	if ((!(sub->filter & TIPC_SUB_PORTS) ==
-	     !(sub->filter & TIPC_SUB_SERVICE)) ||
+	filter = htohl(s->filter, swap);
+	if (((filter & TIPC_SUB_PORTS) && (filter & TIPC_SUB_SERVICE)) ||
 	    (sub->seq.lower > sub->seq.upper)) {
 		pr_warn("Subscription rejected, illegal request\n");
 		kfree(sub);
@@ -265,13 +265,14 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 	spin_lock_bh(&subscriber->lock);
 	list_add(&sub->subscrp_list, &subscriber->subscrp_list);
 	spin_unlock_bh(&subscriber->lock);
+
 	sub->subscriber = subscriber;
 	sub->swap = swap;
 	memcpy(&sub->evt.s, s, sizeof(*s));
 	atomic_inc(&tn->subscription_count);
 	setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub);
-	sub->timeout += jiffies;
-	if (!mod_timer(&sub->timer, sub->timeout))
+	timeout = htohl(sub->evt.s.timeout, swap);
+	if (!mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)))
 		tipc_subscrb_get(subscriber);
 	*sub_p = sub;
 	return 0;
diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h
index 92ee18cc..9e69dbf 100644
--- a/net/tipc/subscr.h
+++ b/net/tipc/subscr.h
@@ -50,12 +50,9 @@ struct tipc_subscriber;
  * @subscriber: pointer to its subscriber
  * @seq: name sequence associated with subscription
  * @net: point to network namespace
- * @timeout: duration of subscription (in ms)
- * @filter: event filtering to be done for subscription
  * @timer: timer governing subscription duration (optional)
  * @nameseq_list: adjacent subscriptions in name sequence's subscription list
  * @subscrp_list: adjacent subscriptions in subscriber's subscription list
- * @server_ref: object reference of server port associated with subscription
  * @swap: indicates if subscriber uses opposite endianness in its messages
  * @evt: template for events generated by subscription
  */
@@ -63,8 +60,6 @@ struct tipc_subscription {
 	struct tipc_subscriber *subscriber;
 	struct tipc_name_seq seq;
 	struct net *net;
-	unsigned long timeout;
-	u32 filter;
 	struct timer_list timer;
 	struct list_head nameseq_list;
 	struct list_head subscrp_list;
-- 
cgit v0.10.2


From a4273c73ebf06515eab6b5a84817a175ad8acdc4 Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:10 +0100
Subject: tipc: remove struct tipc_name_seq from struct tipc_subscription

Until now, struct tipc_subscriber has duplicate fields for
type, upper and lower (as member of struct tipc_name_seq) at:
1. as member seq in struct tipc_subscription
2. as member seq in struct tipc_subscr, which is contained
   in struct tipc_event
The former structure contains the type, upper and lower
values in network byte order and the later contains the
intact copy of the request.
The struct tipc_subscription contains a field swap to
determine if request needs network byte order conversion.
Thus by using swap, we can convert the request when
required instead of duplicating it.

In this commit,
1. we remove the references to these elements as members of
   struct tipc_subscription and replace them with elements
   from struct tipc_subscr.
2. provide new functions to convert the user request into
   network byte order.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c
index 91fce70..777b979 100644
--- a/net/tipc/name_table.c
+++ b/net/tipc/name_table.c
@@ -418,6 +418,9 @@ static void tipc_nameseq_subscribe(struct name_seq *nseq,
 				   struct tipc_subscription *s)
 {
 	struct sub_seq *sseq = nseq->sseqs;
+	struct tipc_name_seq ns;
+
+	tipc_subscrp_convert_seq(&s->evt.s.seq, s->swap, &ns);
 
 	list_add(&s->nameseq_list, &nseq->subscriptions);
 
@@ -425,7 +428,7 @@ static void tipc_nameseq_subscribe(struct name_seq *nseq,
 		return;
 
 	while (sseq != &nseq->sseqs[nseq->first_free]) {
-		if (tipc_subscrp_check_overlap(s, sseq->lower, sseq->upper)) {
+		if (tipc_subscrp_check_overlap(&ns, sseq->lower, sseq->upper)) {
 			struct publication *crs;
 			struct name_info *info = sseq->info;
 			int must_report = 1;
@@ -722,9 +725,10 @@ int tipc_nametbl_withdraw(struct net *net, u32 type, u32 lower, u32 ref,
 void tipc_nametbl_subscribe(struct tipc_subscription *s)
 {
 	struct tipc_net *tn = net_generic(s->net, tipc_net_id);
-	u32 type = s->seq.type;
+	u32 type = tipc_subscrp_convert_seq_type(s->evt.s.seq.type, s->swap);
 	int index = hash(type);
 	struct name_seq *seq;
+	struct tipc_name_seq ns;
 
 	spin_lock_bh(&tn->nametbl_lock);
 	seq = nametbl_find_seq(s->net, type);
@@ -735,8 +739,9 @@ void tipc_nametbl_subscribe(struct tipc_subscription *s)
 		tipc_nameseq_subscribe(seq, s);
 		spin_unlock_bh(&seq->lock);
 	} else {
+		tipc_subscrp_convert_seq(&s->evt.s.seq, s->swap, &ns);
 		pr_warn("Failed to create subscription for {%u,%u,%u}\n",
-			s->seq.type, s->seq.lower, s->seq.upper);
+			ns.type, ns.lower, ns.upper);
 	}
 	spin_unlock_bh(&tn->nametbl_lock);
 }
@@ -748,9 +753,10 @@ void tipc_nametbl_unsubscribe(struct tipc_subscription *s)
 {
 	struct tipc_net *tn = net_generic(s->net, tipc_net_id);
 	struct name_seq *seq;
+	u32 type = tipc_subscrp_convert_seq_type(s->evt.s.seq.type, s->swap);
 
 	spin_lock_bh(&tn->nametbl_lock);
-	seq = nametbl_find_seq(s->net, s->seq.type);
+	seq = nametbl_find_seq(s->net, type);
 	if (seq != NULL) {
 		spin_lock_bh(&seq->lock);
 		list_del_init(&s->nameseq_list);
diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index cef0b6b..5f3100e 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -92,23 +92,39 @@ static void tipc_subscrp_send_event(struct tipc_subscription *sub,
  *
  * Returns 1 if there is overlap, otherwise 0.
  */
-int tipc_subscrp_check_overlap(struct tipc_subscription *sub, u32 found_lower,
+int tipc_subscrp_check_overlap(struct tipc_name_seq *seq, u32 found_lower,
 			       u32 found_upper)
 {
-	if (found_lower < sub->seq.lower)
-		found_lower = sub->seq.lower;
-	if (found_upper > sub->seq.upper)
-		found_upper = sub->seq.upper;
+	if (found_lower < seq->lower)
+		found_lower = seq->lower;
+	if (found_upper > seq->upper)
+		found_upper = seq->upper;
 	if (found_lower > found_upper)
 		return 0;
 	return 1;
 }
 
+u32 tipc_subscrp_convert_seq_type(u32 type, int swap)
+{
+	return htohl(type, swap);
+}
+
+void tipc_subscrp_convert_seq(struct tipc_name_seq *in, int swap,
+			      struct tipc_name_seq *out)
+{
+	out->type = htohl(in->type, swap);
+	out->lower = htohl(in->lower, swap);
+	out->upper = htohl(in->upper, swap);
+}
+
 void tipc_subscrp_report_overlap(struct tipc_subscription *sub, u32 found_lower,
 				 u32 found_upper, u32 event, u32 port_ref,
 				 u32 node, int must)
 {
-	if (!tipc_subscrp_check_overlap(sub, found_lower, found_upper))
+	struct tipc_name_seq seq;
+
+	tipc_subscrp_convert_seq(&sub->evt.s.seq, sub->swap, &seq);
+	if (!tipc_subscrp_check_overlap(&seq, found_lower, found_upper))
 		return;
 	if (!must &&
 	    !(htohl(sub->evt.s.filter, sub->swap) & TIPC_SUB_PORTS))
@@ -252,12 +268,9 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 
 	/* Initialize subscription object */
 	sub->net = net;
-	sub->seq.type = htohl(s->seq.type, swap);
-	sub->seq.lower = htohl(s->seq.lower, swap);
-	sub->seq.upper = htohl(s->seq.upper, swap);
 	filter = htohl(s->filter, swap);
 	if (((filter & TIPC_SUB_PORTS) && (filter & TIPC_SUB_SERVICE)) ||
-	    (sub->seq.lower > sub->seq.upper)) {
+	    (htohl(s->seq.lower, swap) > htohl(s->seq.upper, swap))) {
 		pr_warn("Subscription rejected, illegal request\n");
 		kfree(sub);
 		return -EINVAL;
diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h
index 9e69dbf..be60103 100644
--- a/net/tipc/subscr.h
+++ b/net/tipc/subscr.h
@@ -58,7 +58,6 @@ struct tipc_subscriber;
  */
 struct tipc_subscription {
 	struct tipc_subscriber *subscriber;
-	struct tipc_name_seq seq;
 	struct net *net;
 	struct timer_list timer;
 	struct list_head nameseq_list;
@@ -67,11 +66,14 @@ struct tipc_subscription {
 	struct tipc_event evt;
 };
 
-int tipc_subscrp_check_overlap(struct tipc_subscription *sub, u32 found_lower,
+int tipc_subscrp_check_overlap(struct tipc_name_seq *seq, u32 found_lower,
 			       u32 found_upper);
 void tipc_subscrp_report_overlap(struct tipc_subscription *sub,
 				 u32 found_lower, u32 found_upper, u32 event,
 				 u32 port_ref, u32 node, int must);
+void tipc_subscrp_convert_seq(struct tipc_name_seq *in, int swap,
+			      struct tipc_name_seq *out);
+u32 tipc_subscrp_convert_seq_type(u32 type, int swap);
 int tipc_topsrv_start(struct net *net);
 void tipc_topsrv_stop(struct net *net);
 
-- 
cgit v0.10.2


From 7c13c6224123a6424bd3bc60ef982759754501e9 Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:11 +0100
Subject: tipc: introduce tipc_subscrb_subscribe() routine

In this commit, we split tipc_subscrp_create() into two:
1. tipc_subscrp_create() creates a subscription
2. A new function tipc_subscrp_subscribe() adds the
   subscription to the subscriber subscription list,
   activates the subscription timer and subscribes to
   the nametable updates.

In future commits, the purpose of tipc_subscrb_rcv_cb() will
be to either subscribe or cancel a subscription.

There is no functional change in this commit.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 5f3100e..022a2f2 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -233,13 +233,13 @@ static void tipc_subscrp_cancel(struct tipc_subscr *s,
 	spin_unlock_bh(&subscriber->lock);
 }
 
-static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
-			       struct tipc_subscriber *subscriber,
-			       struct tipc_subscription **sub_p)
+static struct tipc_subscription *tipc_subscrp_create(struct net *net,
+						     struct tipc_subscr *s,
+						     struct tipc_subscriber *subscriber)
 {
 	struct tipc_net *tn = net_generic(net, tipc_net_id);
 	struct tipc_subscription *sub;
-	u32 timeout, filter;
+	u32 filter;
 	int swap;
 
 	/* Determine subscriber's endianness */
@@ -249,21 +249,21 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 	if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) {
 		s->filter &= ~htohl(TIPC_SUB_CANCEL, swap);
 		tipc_subscrp_cancel(s, subscriber);
-		return 0;
+		return NULL;
 	}
 
 	/* Refuse subscription if global limit exceeded */
 	if (atomic_read(&tn->subscription_count) >= TIPC_MAX_SUBSCRIPTIONS) {
 		pr_warn("Subscription rejected, limit reached (%u)\n",
 			TIPC_MAX_SUBSCRIPTIONS);
-		return -EINVAL;
+		return NULL;
 	}
 
 	/* Allocate subscription object */
 	sub = kmalloc(sizeof(*sub), GFP_ATOMIC);
 	if (!sub) {
 		pr_warn("Subscription rejected, no memory\n");
-		return -ENOMEM;
+		return NULL;
 	}
 
 	/* Initialize subscription object */
@@ -273,22 +273,36 @@ static int tipc_subscrp_create(struct net *net, struct tipc_subscr *s,
 	    (htohl(s->seq.lower, swap) > htohl(s->seq.upper, swap))) {
 		pr_warn("Subscription rejected, illegal request\n");
 		kfree(sub);
-		return -EINVAL;
+		return NULL;
 	}
-	spin_lock_bh(&subscriber->lock);
-	list_add(&sub->subscrp_list, &subscriber->subscrp_list);
-	spin_unlock_bh(&subscriber->lock);
 
-	sub->subscriber = subscriber;
 	sub->swap = swap;
 	memcpy(&sub->evt.s, s, sizeof(*s));
 	atomic_inc(&tn->subscription_count);
 	setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub);
-	timeout = htohl(sub->evt.s.timeout, swap);
+	return sub;
+}
+
+static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
+				   struct tipc_subscriber *subscriber)
+{
+	struct tipc_net *tn = net_generic(net, tipc_net_id);
+	struct tipc_subscription *sub = NULL;
+	u32 timeout;
+
+	sub = tipc_subscrp_create(net, s, subscriber);
+	if (!sub)
+		return tipc_conn_terminate(tn->topsrv, subscriber->conid);
+
+	spin_lock_bh(&subscriber->lock);
+	list_add(&sub->subscrp_list, &subscriber->subscrp_list);
+	spin_unlock_bh(&subscriber->lock);
+
+	sub->subscriber = subscriber;
+	timeout = htohl(sub->evt.s.timeout, sub->swap);
 	if (!mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)))
 		tipc_subscrb_get(subscriber);
-	*sub_p = sub;
-	return 0;
+	tipc_nametbl_subscribe(sub);
 }
 
 /* Handle one termination request for the subscriber */
@@ -302,14 +316,8 @@ static void tipc_subscrb_rcv_cb(struct net *net, int conid,
 				struct sockaddr_tipc *addr, void *usr_data,
 				void *buf, size_t len)
 {
-	struct tipc_subscriber *subscrb = usr_data;
-	struct tipc_subscription *sub = NULL;
-	struct tipc_net *tn = net_generic(net, tipc_net_id);
-
-	if (tipc_subscrp_create(net, (struct tipc_subscr *)buf, subscrb, &sub))
-		return tipc_conn_terminate(tn->topsrv, subscrb->conid);
-
-	tipc_nametbl_subscribe(sub);
+	tipc_subscrp_subscribe(net, (struct tipc_subscr *)buf,
+			       (struct tipc_subscriber *)usr_data);
 }
 
 /* Handle one request to establish a new subscriber */
-- 
cgit v0.10.2


From c8beccc67c833db4a01c621cbc6eb0a577286806 Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:12 +0100
Subject: tipc: fix connection abort during subscription cancellation

In 'commit 7fe8097cef5f ("tipc: fix nullpointer bug when subscribing
to events")', we terminate the connection if the subscription
creation fails.
In the same commit, the subscription creation result was based on
the value of subscription pointer (set in the function) instead of
the return code.

Unfortunately, the same function also handles subscription
cancellation request. For a subscription cancellation request,
the subscription pointer cannot be set. Thus the connection is
terminated during cancellation request.

In this commit, we move the subcription cancel check outside
of tipc_subscrp_create(). Hence,
- tipc_subscrp_create() will create a subscripton
- tipc_subscrb_rcv_cb() will subscribe or cancel a subscription.

Fixes: 'commit 7fe8097cef5f ("tipc: fix nullpointer bug when subscribing to events")'

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 022a2f2..5312272 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -235,22 +235,11 @@ static void tipc_subscrp_cancel(struct tipc_subscr *s,
 
 static struct tipc_subscription *tipc_subscrp_create(struct net *net,
 						     struct tipc_subscr *s,
-						     struct tipc_subscriber *subscriber)
+						     int swap)
 {
 	struct tipc_net *tn = net_generic(net, tipc_net_id);
 	struct tipc_subscription *sub;
-	u32 filter;
-	int swap;
-
-	/* Determine subscriber's endianness */
-	swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE));
-
-	/* Detect & process a subscription cancellation request */
-	if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) {
-		s->filter &= ~htohl(TIPC_SUB_CANCEL, swap);
-		tipc_subscrp_cancel(s, subscriber);
-		return NULL;
-	}
+	u32 filter = htohl(s->filter, swap);
 
 	/* Refuse subscription if global limit exceeded */
 	if (atomic_read(&tn->subscription_count) >= TIPC_MAX_SUBSCRIPTIONS) {
@@ -268,7 +257,6 @@ static struct tipc_subscription *tipc_subscrp_create(struct net *net,
 
 	/* Initialize subscription object */
 	sub->net = net;
-	filter = htohl(s->filter, swap);
 	if (((filter & TIPC_SUB_PORTS) && (filter & TIPC_SUB_SERVICE)) ||
 	    (htohl(s->seq.lower, swap) > htohl(s->seq.upper, swap))) {
 		pr_warn("Subscription rejected, illegal request\n");
@@ -284,13 +272,13 @@ static struct tipc_subscription *tipc_subscrp_create(struct net *net,
 }
 
 static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
-				   struct tipc_subscriber *subscriber)
+				   struct tipc_subscriber *subscriber, int swap)
 {
 	struct tipc_net *tn = net_generic(net, tipc_net_id);
 	struct tipc_subscription *sub = NULL;
 	u32 timeout;
 
-	sub = tipc_subscrp_create(net, s, subscriber);
+	sub = tipc_subscrp_create(net, s, swap);
 	if (!sub)
 		return tipc_conn_terminate(tn->topsrv, subscriber->conid);
 
@@ -299,7 +287,7 @@ static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
 	spin_unlock_bh(&subscriber->lock);
 
 	sub->subscriber = subscriber;
-	timeout = htohl(sub->evt.s.timeout, sub->swap);
+	timeout = htohl(sub->evt.s.timeout, swap);
 	if (!mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)))
 		tipc_subscrb_get(subscriber);
 	tipc_nametbl_subscribe(sub);
@@ -316,8 +304,20 @@ static void tipc_subscrb_rcv_cb(struct net *net, int conid,
 				struct sockaddr_tipc *addr, void *usr_data,
 				void *buf, size_t len)
 {
-	tipc_subscrp_subscribe(net, (struct tipc_subscr *)buf,
-			       (struct tipc_subscriber *)usr_data);
+	struct tipc_subscriber *subscriber = usr_data;
+	struct tipc_subscr *s = (struct tipc_subscr *)buf;
+	int swap;
+
+	/* Determine subscriber's endianness */
+	swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE));
+
+	/* Detect & process a subscription cancellation request */
+	if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) {
+		s->filter &= ~htohl(TIPC_SUB_CANCEL, swap);
+		return tipc_subscrp_cancel(s, subscriber);
+	}
+
+	tipc_subscrp_subscribe(net, s, subscriber, swap);
 }
 
 /* Handle one request to establish a new subscriber */
-- 
cgit v0.10.2


From cb01c7c8701a35866479753fe78d04bd9826dd1b Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:13 +0100
Subject: tipc: fix connection abort when receiving invalid cancel request

Until now, the subscribers endianness for a subscription
create/cancel request is determined as:
    swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE))
The checks are performed only for port/service subscriptions.

The swap calculation is incorrect if the filter in the subscription
cancellation request is set to TIPC_SUB_CANCEL (it's a malformed
cancel request, as the corresponding subscription create filter
is missing).
Thus, the check if the request is for cancellation fails and the
request is treated as a subscription create request. The
subscription creation fails as the request is illegal, which
terminates this connection.

In this commit we determine the endianness by including
TIPC_SUB_CANCEL, which will set swap correctly and the
request is processed as a cancellation request.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 5312272..24d2c81 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -309,7 +309,8 @@ static void tipc_subscrb_rcv_cb(struct net *net, int conid,
 	int swap;
 
 	/* Determine subscriber's endianness */
-	swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE));
+	swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE |
+			      TIPC_SUB_CANCEL));
 
 	/* Detect & process a subscription cancellation request */
 	if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) {
-- 
cgit v0.10.2


From d4091899c9bbfd6695449c6b09517ceb45bb379d Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:14 +0100
Subject: tipc: hold subscriber->lock for tipc_nametbl_subscribe()

Until now, while creating a subscription the subscriber lock
protects only the subscribers subscription list and not the
nametable. The call to tipc_nametbl_subscribe() is outside
the lock. However, at subscription timeout and cancel both
the subscribers subscription list and the nametable are
protected by the subscriber lock.

This asymmetric locking mechanism leads to the following problem:
In a SMP system, the timer can be fire on another core before
the create request is complete.
When the timer thread calls tipc_nametbl_unsubscribe() before create
thread calls tipc_nametbl_subscribe(), we get a nullptr exception.

This can be simulated by creating subscription with timeout=0 and
sometimes the timeout occurs before the create request is complete.

The following is the oops:
[57.569661] BUG: unable to handle kernel NULL pointer dereference at (null)
[57.577498] IP: [<ffffffffa02135aa>] tipc_nametbl_unsubscribe+0x8a/0x120 [tipc]
[57.584820] PGD 0
[57.586834] Oops: 0002 [#1] SMP
[57.685506] CPU: 14 PID: 10077 Comm: kworker/u40:1 Tainted: P OENX 3.12.48-52.27.1.     9688.1.PTF-default #1
[57.703637] Workqueue: tipc_rcv tipc_recv_work [tipc]
[57.708697] task: ffff88064c7f00c0 ti: ffff880629ef4000 task.ti: ffff880629ef4000
[57.716181] RIP: 0010:[<ffffffffa02135aa>]  [<ffffffffa02135aa>] tipc_nametbl_unsubscribe+0x8a/   0x120 [tipc]
[...]
[57.812327] Call Trace:
[57.814806]  [<ffffffffa0211c77>] tipc_subscrp_delete+0x37/0x90 [tipc]
[57.821357]  [<ffffffffa0211e2f>] tipc_subscrp_timeout+0x3f/0x70 [tipc]
[57.827982]  [<ffffffff810618c1>] call_timer_fn+0x31/0x100
[57.833490]  [<ffffffff81062709>] run_timer_softirq+0x1f9/0x2b0
[57.839414]  [<ffffffff8105a795>] __do_softirq+0xe5/0x230
[57.844827]  [<ffffffff81520d1c>] call_softirq+0x1c/0x30
[57.850150]  [<ffffffff81004665>] do_softirq+0x55/0x90
[57.855285]  [<ffffffff8105aa35>] irq_exit+0x95/0xa0
[57.860290]  [<ffffffff815215b5>] smp_apic_timer_interrupt+0x45/0x60
[57.866644]  [<ffffffff8152005d>] apic_timer_interrupt+0x6d/0x80
[57.872686]  [<ffffffffa02121c5>] tipc_subscrb_rcv_cb+0x2a5/0x3f0 [tipc]
[57.879425]  [<ffffffffa021c65f>] tipc_receive_from_sock+0x9f/0x100 [tipc]
[57.886324]  [<ffffffffa021c826>] tipc_recv_work+0x26/0x60 [tipc]
[57.892463]  [<ffffffff8106fb22>] process_one_work+0x172/0x420
[57.898309]  [<ffffffff8107079a>] worker_thread+0x11a/0x3c0
[57.903871]  [<ffffffff81077114>] kthread+0xb4/0xc0
[57.908751]  [<ffffffff8151f318>] ret_from_fork+0x58/0x90

In this commit, we do the following at subscription creation:
1. set the subscription's subscriber pointer before performing
   tipc_nametbl_subscribe(), as this value is required further in
   the call chain ex: by tipc_subscrp_send_event().
2. move tipc_nametbl_subscribe() under the scope of subscriber lock

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 24d2c81..e4ebbc1 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -284,13 +284,13 @@ static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
 
 	spin_lock_bh(&subscriber->lock);
 	list_add(&sub->subscrp_list, &subscriber->subscrp_list);
+	sub->subscriber = subscriber;
+	tipc_nametbl_subscribe(sub);
 	spin_unlock_bh(&subscriber->lock);
 
-	sub->subscriber = subscriber;
 	timeout = htohl(sub->evt.s.timeout, swap);
 	if (!mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)))
 		tipc_subscrb_get(subscriber);
-	tipc_nametbl_subscribe(sub);
 }
 
 /* Handle one termination request for the subscriber */
-- 
cgit v0.10.2


From f3ad288c56d21e3b8ec77e13ab9aa06dbbfa3577 Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:15 +0100
Subject: tipc: protect tipc_subscrb_get() with subscriber spin lock

Until now, during subscription creation the mod_time() &
tipc_subscrb_get() are called after releasing the subscriber
spin lock.

In a SMP system when performing a subscription creation, if the
subscription timeout occurs simultaneously (the timer is
scheduled to run on another CPU) then the timer thread
might decrement the subscribers refcount before the create
thread increments the refcount.

This can be simulated by creating subscription with timeout=0 and
sometimes the timeout occurs before the create request is complete.
This leads to the following message:
[30.702949] BUG: spinlock bad magic on CPU#1, kworker/u8:3/87
[30.703834] general protection fault: 0000 [#1] SMP
[30.704826] CPU: 1 PID: 87 Comm: kworker/u8:3 Not tainted 4.4.0-rc8+ #18
[30.704826] Workqueue: tipc_rcv tipc_recv_work [tipc]
[30.704826] task: ffff88003f878600 ti: ffff88003fae0000 task.ti: ffff88003fae0000
[30.704826] RIP: 0010:[<ffffffff8109196c>]  [<ffffffff8109196c>] spin_dump+0x5c/0xe0
[...]
[30.704826] Call Trace:
[30.704826]  [<ffffffff81091a16>] spin_bug+0x26/0x30
[30.704826]  [<ffffffff81091b75>] do_raw_spin_lock+0xe5/0x120
[30.704826]  [<ffffffff81684439>] _raw_spin_lock_bh+0x19/0x20
[30.704826]  [<ffffffffa0096f10>] tipc_subscrb_rcv_cb+0x1d0/0x330 [tipc]
[30.704826]  [<ffffffffa00a37b1>] tipc_receive_from_sock+0xc1/0x150 [tipc]
[30.704826]  [<ffffffffa00a31df>] tipc_recv_work+0x3f/0x80 [tipc]
[30.704826]  [<ffffffff8106a739>] process_one_work+0x149/0x3c0
[30.704826]  [<ffffffff8106aa16>] worker_thread+0x66/0x460
[30.704826]  [<ffffffff8106a9b0>] ? process_one_work+0x3c0/0x3c0
[30.704826]  [<ffffffff8106a9b0>] ? process_one_work+0x3c0/0x3c0
[30.704826]  [<ffffffff8107029d>] kthread+0xed/0x110
[30.704826]  [<ffffffff810701b0>] ? kthread_create_on_node+0x190/0x190
[30.704826]  [<ffffffff81684bdf>] ret_from_fork+0x3f/0x70

In this commit,
1. we remove the check for the return code for mod_timer()
2. we protect tipc_subscrb_get() using the subscriber spin lock.
   We increment the subscriber's refcount as soon as we add the
   subscription to subscriber's subscription list.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index e4ebbc1..7d226ec 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -284,13 +284,13 @@ static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
 
 	spin_lock_bh(&subscriber->lock);
 	list_add(&sub->subscrp_list, &subscriber->subscrp_list);
+	tipc_subscrb_get(subscriber);
 	sub->subscriber = subscriber;
 	tipc_nametbl_subscribe(sub);
 	spin_unlock_bh(&subscriber->lock);
 
 	timeout = htohl(sub->evt.s.timeout, swap);
-	if (!mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout)))
-		tipc_subscrb_get(subscriber);
+	mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout));
 }
 
 /* Handle one termination request for the subscriber */
-- 
cgit v0.10.2


From ae245557f87fffe2e1c39ba07524024e650e822b Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:16 +0100
Subject: tipc: donot create timers if subscription timeout = TIPC_WAIT_FOREVER

Until now, we create timers even for the subscription requests
with timeout = TIPC_WAIT_FOREVER.
This can be improved by avoiding timer creation when the timeout
is set to TIPC_WAIT_FOREVER.

In this commit, we introduce a check to creates timers only
when timeout != TIPC_WAIT_FOREVER.

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 7d226ec..22963ca 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -188,12 +188,14 @@ static struct tipc_subscriber *tipc_subscrb_create(int conid)
 static void tipc_subscrb_delete(struct tipc_subscriber *subscriber)
 {
 	struct tipc_subscription *sub, *temp;
+	u32 timeout;
 
 	spin_lock_bh(&subscriber->lock);
 	/* Destroy any existing subscriptions for subscriber */
 	list_for_each_entry_safe(sub, temp, &subscriber->subscrp_list,
 				 subscrp_list) {
-		if (del_timer(&sub->timer)) {
+		timeout = htohl(sub->evt.s.timeout, sub->swap);
+		if ((timeout == TIPC_WAIT_FOREVER) || del_timer(&sub->timer)) {
 			tipc_subscrp_delete(sub);
 			tipc_subscrb_put(subscriber);
 		}
@@ -217,13 +219,16 @@ static void tipc_subscrp_cancel(struct tipc_subscr *s,
 				struct tipc_subscriber *subscriber)
 {
 	struct tipc_subscription *sub, *temp;
+	u32 timeout;
 
 	spin_lock_bh(&subscriber->lock);
 	/* Find first matching subscription, exit if not found */
 	list_for_each_entry_safe(sub, temp, &subscriber->subscrp_list,
 				 subscrp_list) {
 		if (!memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) {
-			if (del_timer(&sub->timer)) {
+			timeout = htohl(sub->evt.s.timeout, sub->swap);
+			if ((timeout == TIPC_WAIT_FOREVER) ||
+			    del_timer(&sub->timer)) {
 				tipc_subscrp_delete(sub);
 				tipc_subscrb_put(subscriber);
 			}
@@ -267,7 +272,6 @@ static struct tipc_subscription *tipc_subscrp_create(struct net *net,
 	sub->swap = swap;
 	memcpy(&sub->evt.s, s, sizeof(*s));
 	atomic_inc(&tn->subscription_count);
-	setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub);
 	return sub;
 }
 
@@ -290,6 +294,10 @@ static void tipc_subscrp_subscribe(struct net *net, struct tipc_subscr *s,
 	spin_unlock_bh(&subscriber->lock);
 
 	timeout = htohl(sub->evt.s.timeout, swap);
+	if (timeout == TIPC_WAIT_FOREVER)
+		return;
+
+	setup_timer(&sub->timer, tipc_subscrp_timeout, (unsigned long)sub);
 	mod_timer(&sub->timer, jiffies + msecs_to_jiffies(timeout));
 }
 
-- 
cgit v0.10.2


From 06c8581f85e99bbf69723f76ad2a40fa8a8c8cdd Mon Sep 17 00:00:00 2001
From: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Date: Tue, 2 Feb 2016 10:52:17 +0100
Subject: tipc: use alloc_ordered_workqueue() instead of WQ_UNBOUND w/
 max_active = 1

Until now, tipc_rcv and tipc_send workqueues in server are allocated
with parameters WQ_UNBOUND & max_active = 1.
This parameters passed to this function makes it equivalent to
alloc_ordered_workqueue(). The later form is more explicit and
can inherit future ordered_workqueue changes.

In this commit we replace alloc_workqueue() with more readable
alloc_ordered_workqueue().

Acked-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/server.c b/net/tipc/server.c
index 922e04a..2446bfb 100644
--- a/net/tipc/server.c
+++ b/net/tipc/server.c
@@ -571,13 +571,13 @@ static void tipc_work_stop(struct tipc_server *s)
 
 static int tipc_work_start(struct tipc_server *s)
 {
-	s->rcv_wq = alloc_workqueue("tipc_rcv", WQ_UNBOUND, 1);
+	s->rcv_wq = alloc_ordered_workqueue("tipc_rcv", 0);
 	if (!s->rcv_wq) {
 		pr_err("can't start tipc receive workqueue\n");
 		return -ENOMEM;
 	}
 
-	s->send_wq = alloc_workqueue("tipc_send", WQ_UNBOUND, 1);
+	s->send_wq = alloc_ordered_workqueue("tipc_send", 0);
 	if (!s->send_wq) {
 		pr_err("can't start tipc send workqueue\n");
 		destroy_workqueue(s->rcv_wq);
-- 
cgit v0.10.2


From 1b3b36b044939b438c4463e8bde8cd644ba82032 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Sat, 16 Jan 2016 00:48:32 +0100
Subject: bcma: support identifying MX25L25635F serial flash
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's a Macronix 32 MiB flash found on board with BCM47189 SoC.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/driver_chipcommon_sflash.c b/drivers/bcma/driver_chipcommon_sflash.c
index 7e11ef4..04d706c 100644
--- a/drivers/bcma/driver_chipcommon_sflash.c
+++ b/drivers/bcma/driver_chipcommon_sflash.c
@@ -38,6 +38,7 @@ static const struct bcma_sflash_tbl_e bcma_sflash_st_tbl[] = {
 	{ "M25P32", 0x15, 0x10000, 64, },
 	{ "M25P64", 0x16, 0x10000, 128, },
 	{ "M25FL128", 0x17, 0x10000, 256, },
+	{ "MX25L25635F", 0x18, 0x10000, 512, },
 	{ NULL },
 };
 
-- 
cgit v0.10.2


From 7267bcda332e2782e21a559f3b1b859a35b4062d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Sat, 16 Jan 2016 00:48:52 +0100
Subject: bcma: identify bus cores (devices) found on BCM47189
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add missing defines and print proper names.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/scan.c b/drivers/bcma/scan.c
index df806b9..5ee7311 100644
--- a/drivers/bcma/scan.c
+++ b/drivers/bcma/scan.c
@@ -98,6 +98,9 @@ static const struct bcma_device_id_name bcma_bcm_device_names[] = {
 	{ BCMA_CORE_SHIM, "SHIM" },
 	{ BCMA_CORE_PCIE2, "PCIe Gen2" },
 	{ BCMA_CORE_ARM_CR4, "ARM CR4" },
+	{ BCMA_CORE_GCI, "GCI" },
+	{ BCMA_CORE_CMEM, "CNDS DDR2/3 memory controller" },
+	{ BCMA_CORE_ARM_CA7, "ARM CA7" },
 	{ BCMA_CORE_DEFAULT, "Default" },
 };
 
diff --git a/include/linux/bcma/bcma.h b/include/linux/bcma/bcma.h
index 3feb1b2..991ebb4 100644
--- a/include/linux/bcma/bcma.h
+++ b/include/linux/bcma/bcma.h
@@ -151,6 +151,8 @@ struct bcma_host_ops {
 #define BCMA_CORE_PCIE2			0x83C	/* PCI Express Gen2 */
 #define BCMA_CORE_USB30_DEV		0x83D
 #define BCMA_CORE_ARM_CR4		0x83E
+#define BCMA_CORE_GCI			0x840
+#define BCMA_CORE_CMEM			0x846	/* CNDS DDR2/3 memory controller */
 #define BCMA_CORE_ARM_CA7		0x847
 #define BCMA_CORE_SYS_MEM		0x849
 #define BCMA_CORE_DEFAULT		0xFFF
-- 
cgit v0.10.2


From 0c06f5d43e1f969bae3fa7832a94af46c75db9a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Sat, 16 Jan 2016 00:48:53 +0100
Subject: bcma: support chipsets with PMU and GCI cores (devices)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Both cores are another exceptions. They are not accessed in a standard
way and to they don't need or have wrapping addresses.
This fixes bus scanning after finding such core.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/scan.c b/drivers/bcma/scan.c
index 5ee7311..4a2d1b2 100644
--- a/drivers/bcma/scan.c
+++ b/drivers/bcma/scan.c
@@ -318,6 +318,8 @@ static int bcma_get_next_core(struct bcma_bus *bus, u32 __iomem **eromptr,
 		switch (core->id.id) {
 		case BCMA_CORE_4706_MAC_GBIT_COMMON:
 		case BCMA_CORE_NS_CHIPCOMMON_B:
+		case BCMA_CORE_PMU:
+		case BCMA_CORE_GCI:
 		/* Not used yet: case BCMA_CORE_OOB_ROUTER: */
 			break;
 		default:
-- 
cgit v0.10.2


From 67edf354faaf93156646e741483b2313bc756c0f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 19 Jan 2016 08:45:25 +0100
Subject: bcma: use _PMU_ in all names of PMU registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

PMU (Power Management Unit) seems to be a separated piece of hardware,
just accessed using ChipCommon core registers. In recent Broadcom
chipsets PMU is not bounded to CC but available as separated core.

To make code cleaner & easier to review (for a correct R/W access) use
clearer names.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/driver_chipcommon_pmu.c b/drivers/bcma/driver_chipcommon_pmu.c
index fe0d48c..472f39d 100644
--- a/drivers/bcma/driver_chipcommon_pmu.c
+++ b/drivers/bcma/driver_chipcommon_pmu.c
@@ -15,44 +15,44 @@
 
 u32 bcma_chipco_pll_read(struct bcma_drv_cc *cc, u32 offset)
 {
-	bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PLLCTL_ADDR);
-	return bcma_cc_read32(cc, BCMA_CC_PLLCTL_DATA);
+	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
+	return bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_pll_read);
 
 void bcma_chipco_pll_write(struct bcma_drv_cc *cc, u32 offset, u32 value)
 {
-	bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PLLCTL_ADDR);
-	bcma_cc_write32(cc, BCMA_CC_PLLCTL_DATA, value);
+	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
+	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, value);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_pll_write);
 
 void bcma_chipco_pll_maskset(struct bcma_drv_cc *cc, u32 offset, u32 mask,
 			     u32 set)
 {
-	bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PLLCTL_ADDR);
-	bcma_cc_maskset32(cc, BCMA_CC_PLLCTL_DATA, mask, set);
+	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
+	bcma_cc_maskset32(cc, BCMA_CC_PMU_PLLCTL_DATA, mask, set);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_pll_maskset);
 
 void bcma_chipco_chipctl_maskset(struct bcma_drv_cc *cc,
 				 u32 offset, u32 mask, u32 set)
 {
-	bcma_cc_write32(cc, BCMA_CC_CHIPCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_CHIPCTL_ADDR);
-	bcma_cc_maskset32(cc, BCMA_CC_CHIPCTL_DATA, mask, set);
+	bcma_cc_write32(cc, BCMA_CC_PMU_CHIPCTL_ADDR, offset);
+	bcma_cc_read32(cc, BCMA_CC_PMU_CHIPCTL_ADDR);
+	bcma_cc_maskset32(cc, BCMA_CC_PMU_CHIPCTL_DATA, mask, set);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_chipctl_maskset);
 
 void bcma_chipco_regctl_maskset(struct bcma_drv_cc *cc, u32 offset, u32 mask,
 				u32 set)
 {
-	bcma_cc_write32(cc, BCMA_CC_REGCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_REGCTL_ADDR);
-	bcma_cc_maskset32(cc, BCMA_CC_REGCTL_DATA, mask, set);
+	bcma_cc_write32(cc, BCMA_CC_PMU_REGCTL_ADDR, offset);
+	bcma_cc_read32(cc, BCMA_CC_PMU_REGCTL_ADDR);
+	bcma_cc_maskset32(cc, BCMA_CC_PMU_REGCTL_DATA, mask, set);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_regctl_maskset);
 
@@ -472,8 +472,8 @@ u32 bcma_pmu_get_cpu_clock(struct bcma_drv_cc *cc)
 static void bcma_pmu_spuravoid_pll_write(struct bcma_drv_cc *cc, u32 offset,
 					 u32 value)
 {
-	bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR, offset);
-	bcma_cc_write32(cc, BCMA_CC_PLLCTL_DATA, value);
+	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, value);
 }
 
 void bcma_pmu_spuravoid_pllupdate(struct bcma_drv_cc *cc, int spuravoid)
@@ -497,20 +497,20 @@ void bcma_pmu_spuravoid_pllupdate(struct bcma_drv_cc *cc, int spuravoid)
 		       bus->chipinfo.id == BCMA_CHIP_ID_BCM53572) ? 6 : 0;
 
 		/* RMW only the P1 divider */
-		bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR,
+		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR,
 				BCMA_CC_PMU_PLL_CTL0 + phypll_offset);
-		tmp = bcma_cc_read32(cc, BCMA_CC_PLLCTL_DATA);
+		tmp = bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
 		tmp &= (~(BCMA_CC_PMU1_PLL0_PC0_P1DIV_MASK));
 		tmp |= (bcm5357_bcm43236_p1div[spuravoid] << BCMA_CC_PMU1_PLL0_PC0_P1DIV_SHIFT);
-		bcma_cc_write32(cc, BCMA_CC_PLLCTL_DATA, tmp);
+		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, tmp);
 
 		/* RMW only the int feedback divider */
-		bcma_cc_write32(cc, BCMA_CC_PLLCTL_ADDR,
+		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR,
 				BCMA_CC_PMU_PLL_CTL2 + phypll_offset);
-		tmp = bcma_cc_read32(cc, BCMA_CC_PLLCTL_DATA);
+		tmp = bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
 		tmp &= ~(BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_MASK);
 		tmp |= (bcm5357_bcm43236_ndiv[spuravoid]) << BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_SHIFT;
-		bcma_cc_write32(cc, BCMA_CC_PLLCTL_DATA, tmp);
+		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, tmp);
 
 		tmp = BCMA_CC_PMU_CTL_PLL_UPD;
 		break;
diff --git a/drivers/net/wireless/broadcom/b43/main.c b/drivers/net/wireless/broadcom/b43/main.c
index ec013fb..c279211 100644
--- a/drivers/net/wireless/broadcom/b43/main.c
+++ b/drivers/net/wireless/broadcom/b43/main.c
@@ -1215,10 +1215,10 @@ void b43_wireless_core_phy_pll_reset(struct b43_wldev *dev)
 	case B43_BUS_BCMA:
 		bcma_cc = &dev->dev->bdev->bus->drv_cc;
 
-		bcma_cc_write32(bcma_cc, BCMA_CC_CHIPCTL_ADDR, 0);
-		bcma_cc_mask32(bcma_cc, BCMA_CC_CHIPCTL_DATA, ~0x4);
-		bcma_cc_set32(bcma_cc, BCMA_CC_CHIPCTL_DATA, 0x4);
-		bcma_cc_mask32(bcma_cc, BCMA_CC_CHIPCTL_DATA, ~0x4);
+		bcma_cc_write32(bcma_cc, BCMA_CC_PMU_CHIPCTL_ADDR, 0);
+		bcma_cc_mask32(bcma_cc, BCMA_CC_PMU_CHIPCTL_DATA, ~0x4);
+		bcma_cc_set32(bcma_cc, BCMA_CC_PMU_CHIPCTL_DATA, 0x4);
+		bcma_cc_mask32(bcma_cc, BCMA_CC_PMU_CHIPCTL_DATA, ~0x4);
 		break;
 #endif
 #ifdef CONFIG_B43_SSB
diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h
index db51a6f..96d8d56 100644
--- a/include/linux/bcma/bcma_driver_chipcommon.h
+++ b/include/linux/bcma/bcma_driver_chipcommon.h
@@ -351,12 +351,12 @@
 #define BCMA_CC_PMU_RES_REQTS		0x0640 /* PMU res req timer sel */
 #define BCMA_CC_PMU_RES_REQT		0x0644 /* PMU res req timer */
 #define BCMA_CC_PMU_RES_REQM		0x0648 /* PMU res req mask */
-#define BCMA_CC_CHIPCTL_ADDR		0x0650
-#define BCMA_CC_CHIPCTL_DATA		0x0654
-#define BCMA_CC_REGCTL_ADDR		0x0658
-#define BCMA_CC_REGCTL_DATA		0x065C
-#define BCMA_CC_PLLCTL_ADDR		0x0660
-#define BCMA_CC_PLLCTL_DATA		0x0664
+#define BCMA_CC_PMU_CHIPCTL_ADDR	0x0650
+#define BCMA_CC_PMU_CHIPCTL_DATA	0x0654
+#define BCMA_CC_PMU_REGCTL_ADDR		0x0658
+#define BCMA_CC_PMU_REGCTL_DATA		0x065C
+#define BCMA_CC_PMU_PLLCTL_ADDR		0x0660
+#define BCMA_CC_PMU_PLLCTL_DATA		0x0664
 #define BCMA_CC_PMU_STRAPOPT		0x0668 /* (corerev >= 28) */
 #define BCMA_CC_PMU_XTAL_FREQ		0x066C /* (pmurev >= 10) */
 #define  BCMA_CC_PMU_XTAL_FREQ_ILPCTL_MASK	0x00001FFF
-- 
cgit v0.10.2


From b3c47afbf54d86daa0473895e8ca9e8b663f5c1a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 19 Jan 2016 08:45:26 +0100
Subject: bcma: support PMU present as separated bus core
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On recent Broadcom chipsets PMU is present as separated core and it
can't be accessed using ChipCommon anymore as it fails with e.g.:
[    0.000577] Unhandled fault: external abort on non-linefetch (0x1008) at 0xf1000604

Solve it by using a new (PMU) core pointer set to ChipCommon or PMU
depending on the hardware capabilities.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/driver_chipcommon.c b/drivers/bcma/driver_chipcommon.c
index b7c8a8d..36ee221 100644
--- a/drivers/bcma/driver_chipcommon.c
+++ b/drivers/bcma/driver_chipcommon.c
@@ -185,7 +185,7 @@ u32 bcma_chipco_watchdog_timer_set(struct bcma_drv_cc *cc, u32 ticks)
 			ticks = 2;
 		else if (ticks > maxt)
 			ticks = maxt;
-		bcma_cc_write32(cc, BCMA_CC_PMU_WATCHDOG, ticks);
+		bcma_pmu_write32(cc, BCMA_CC_PMU_WATCHDOG, ticks);
 	} else {
 		struct bcma_bus *bus = cc->core->bus;
 
diff --git a/drivers/bcma/driver_chipcommon_pmu.c b/drivers/bcma/driver_chipcommon_pmu.c
index 472f39d..f1eb4d3 100644
--- a/drivers/bcma/driver_chipcommon_pmu.c
+++ b/drivers/bcma/driver_chipcommon_pmu.c
@@ -15,44 +15,44 @@
 
 u32 bcma_chipco_pll_read(struct bcma_drv_cc *cc, u32 offset)
 {
-	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
-	return bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_pmu_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
+	return bcma_pmu_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_pll_read);
 
 void bcma_chipco_pll_write(struct bcma_drv_cc *cc, u32 offset, u32 value)
 {
-	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
-	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, value);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_pmu_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, value);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_pll_write);
 
 void bcma_chipco_pll_maskset(struct bcma_drv_cc *cc, u32 offset, u32 mask,
 			     u32 set)
 {
-	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
-	bcma_cc_maskset32(cc, BCMA_CC_PMU_PLLCTL_DATA, mask, set);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_pmu_read32(cc, BCMA_CC_PMU_PLLCTL_ADDR);
+	bcma_pmu_maskset32(cc, BCMA_CC_PMU_PLLCTL_DATA, mask, set);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_pll_maskset);
 
 void bcma_chipco_chipctl_maskset(struct bcma_drv_cc *cc,
 				 u32 offset, u32 mask, u32 set)
 {
-	bcma_cc_write32(cc, BCMA_CC_PMU_CHIPCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PMU_CHIPCTL_ADDR);
-	bcma_cc_maskset32(cc, BCMA_CC_PMU_CHIPCTL_DATA, mask, set);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_CHIPCTL_ADDR, offset);
+	bcma_pmu_read32(cc, BCMA_CC_PMU_CHIPCTL_ADDR);
+	bcma_pmu_maskset32(cc, BCMA_CC_PMU_CHIPCTL_DATA, mask, set);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_chipctl_maskset);
 
 void bcma_chipco_regctl_maskset(struct bcma_drv_cc *cc, u32 offset, u32 mask,
 				u32 set)
 {
-	bcma_cc_write32(cc, BCMA_CC_PMU_REGCTL_ADDR, offset);
-	bcma_cc_read32(cc, BCMA_CC_PMU_REGCTL_ADDR);
-	bcma_cc_maskset32(cc, BCMA_CC_PMU_REGCTL_DATA, mask, set);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_REGCTL_ADDR, offset);
+	bcma_pmu_read32(cc, BCMA_CC_PMU_REGCTL_ADDR);
+	bcma_pmu_maskset32(cc, BCMA_CC_PMU_REGCTL_DATA, mask, set);
 }
 EXPORT_SYMBOL_GPL(bcma_chipco_regctl_maskset);
 
@@ -60,18 +60,18 @@ static u32 bcma_pmu_xtalfreq(struct bcma_drv_cc *cc)
 {
 	u32 ilp_ctl, alp_hz;
 
-	if (!(bcma_cc_read32(cc, BCMA_CC_PMU_STAT) &
+	if (!(bcma_pmu_read32(cc, BCMA_CC_PMU_STAT) &
 	      BCMA_CC_PMU_STAT_EXT_LPO_AVAIL))
 		return 0;
 
-	bcma_cc_write32(cc, BCMA_CC_PMU_XTAL_FREQ,
-			BIT(BCMA_CC_PMU_XTAL_FREQ_MEASURE_SHIFT));
+	bcma_pmu_write32(cc, BCMA_CC_PMU_XTAL_FREQ,
+			 BIT(BCMA_CC_PMU_XTAL_FREQ_MEASURE_SHIFT));
 	usleep_range(1000, 2000);
 
-	ilp_ctl = bcma_cc_read32(cc, BCMA_CC_PMU_XTAL_FREQ);
+	ilp_ctl = bcma_pmu_read32(cc, BCMA_CC_PMU_XTAL_FREQ);
 	ilp_ctl &= BCMA_CC_PMU_XTAL_FREQ_ILPCTL_MASK;
 
-	bcma_cc_write32(cc, BCMA_CC_PMU_XTAL_FREQ, 0);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_XTAL_FREQ, 0);
 
 	alp_hz = ilp_ctl * 32768 / 4;
 	return (alp_hz + 50000) / 100000 * 100;
@@ -127,8 +127,8 @@ static void bcma_pmu2_pll_init0(struct bcma_drv_cc *cc, u32 xtalfreq)
 		mask = (u32)~(BCMA_RES_4314_HT_AVAIL |
 			      BCMA_RES_4314_MACPHY_CLK_AVAIL);
 
-		bcma_cc_mask32(cc, BCMA_CC_PMU_MINRES_MSK, mask);
-		bcma_cc_mask32(cc, BCMA_CC_PMU_MAXRES_MSK, mask);
+		bcma_pmu_mask32(cc, BCMA_CC_PMU_MINRES_MSK, mask);
+		bcma_pmu_mask32(cc, BCMA_CC_PMU_MAXRES_MSK, mask);
 		bcma_wait_value(cc->core, BCMA_CLKCTLST,
 				BCMA_CLKCTLST_HAVEHT, 0, 20000);
 		break;
@@ -140,7 +140,7 @@ static void bcma_pmu2_pll_init0(struct bcma_drv_cc *cc, u32 xtalfreq)
 
 	/* Flush */
 	if (cc->pmu.rev >= 2)
-		bcma_cc_set32(cc, BCMA_CC_PMU_CTL, BCMA_CC_PMU_CTL_PLL_UPD);
+		bcma_pmu_set32(cc, BCMA_CC_PMU_CTL, BCMA_CC_PMU_CTL_PLL_UPD);
 
 	/* TODO: Do we need to update OTP? */
 }
@@ -195,9 +195,9 @@ static void bcma_pmu_resources_init(struct bcma_drv_cc *cc)
 
 	/* Set the resource masks. */
 	if (min_msk)
-		bcma_cc_write32(cc, BCMA_CC_PMU_MINRES_MSK, min_msk);
+		bcma_pmu_write32(cc, BCMA_CC_PMU_MINRES_MSK, min_msk);
 	if (max_msk)
-		bcma_cc_write32(cc, BCMA_CC_PMU_MAXRES_MSK, max_msk);
+		bcma_pmu_write32(cc, BCMA_CC_PMU_MAXRES_MSK, max_msk);
 
 	/*
 	 * Add some delay; allow resources to come up and settle.
@@ -269,23 +269,33 @@ static void bcma_pmu_workarounds(struct bcma_drv_cc *cc)
 
 void bcma_pmu_early_init(struct bcma_drv_cc *cc)
 {
+	struct bcma_bus *bus = cc->core->bus;
 	u32 pmucap;
 
-	pmucap = bcma_cc_read32(cc, BCMA_CC_PMU_CAP);
+	if (cc->core->id.rev >= 35 &&
+	    cc->capabilities_ext & BCMA_CC_CAP_EXT_AOB_PRESENT) {
+		cc->pmu.core = bcma_find_core(bus, BCMA_CORE_PMU);
+		if (!cc->pmu.core)
+			bcma_warn(bus, "Couldn't find expected PMU core");
+	}
+	if (!cc->pmu.core)
+		cc->pmu.core = cc->core;
+
+	pmucap = bcma_pmu_read32(cc, BCMA_CC_PMU_CAP);
 	cc->pmu.rev = (pmucap & BCMA_CC_PMU_CAP_REVISION);
 
-	bcma_debug(cc->core->bus, "Found rev %u PMU (capabilities 0x%08X)\n",
-		   cc->pmu.rev, pmucap);
+	bcma_debug(bus, "Found rev %u PMU (capabilities 0x%08X)\n", cc->pmu.rev,
+		   pmucap);
 }
 
 void bcma_pmu_init(struct bcma_drv_cc *cc)
 {
 	if (cc->pmu.rev == 1)
-		bcma_cc_mask32(cc, BCMA_CC_PMU_CTL,
-			      ~BCMA_CC_PMU_CTL_NOILPONW);
+		bcma_pmu_mask32(cc, BCMA_CC_PMU_CTL,
+				~BCMA_CC_PMU_CTL_NOILPONW);
 	else
-		bcma_cc_set32(cc, BCMA_CC_PMU_CTL,
-			     BCMA_CC_PMU_CTL_NOILPONW);
+		bcma_pmu_set32(cc, BCMA_CC_PMU_CTL,
+			       BCMA_CC_PMU_CTL_NOILPONW);
 
 	bcma_pmu_pll_init(cc);
 	bcma_pmu_resources_init(cc);
@@ -472,8 +482,8 @@ u32 bcma_pmu_get_cpu_clock(struct bcma_drv_cc *cc)
 static void bcma_pmu_spuravoid_pll_write(struct bcma_drv_cc *cc, u32 offset,
 					 u32 value)
 {
-	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
-	bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, value);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR, offset);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, value);
 }
 
 void bcma_pmu_spuravoid_pllupdate(struct bcma_drv_cc *cc, int spuravoid)
@@ -497,20 +507,20 @@ void bcma_pmu_spuravoid_pllupdate(struct bcma_drv_cc *cc, int spuravoid)
 		       bus->chipinfo.id == BCMA_CHIP_ID_BCM53572) ? 6 : 0;
 
 		/* RMW only the P1 divider */
-		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR,
+		bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR,
 				BCMA_CC_PMU_PLL_CTL0 + phypll_offset);
-		tmp = bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
+		tmp = bcma_pmu_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
 		tmp &= (~(BCMA_CC_PMU1_PLL0_PC0_P1DIV_MASK));
 		tmp |= (bcm5357_bcm43236_p1div[spuravoid] << BCMA_CC_PMU1_PLL0_PC0_P1DIV_SHIFT);
-		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, tmp);
+		bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, tmp);
 
 		/* RMW only the int feedback divider */
-		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR,
+		bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_ADDR,
 				BCMA_CC_PMU_PLL_CTL2 + phypll_offset);
-		tmp = bcma_cc_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
+		tmp = bcma_pmu_read32(cc, BCMA_CC_PMU_PLLCTL_DATA);
 		tmp &= ~(BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_MASK);
 		tmp |= (bcm5357_bcm43236_ndiv[spuravoid]) << BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_SHIFT;
-		bcma_cc_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, tmp);
+		bcma_pmu_write32(cc, BCMA_CC_PMU_PLLCTL_DATA, tmp);
 
 		tmp = BCMA_CC_PMU_CTL_PLL_UPD;
 		break;
@@ -646,7 +656,7 @@ void bcma_pmu_spuravoid_pllupdate(struct bcma_drv_cc *cc, int spuravoid)
 		break;
 	}
 
-	tmp |= bcma_cc_read32(cc, BCMA_CC_PMU_CTL);
-	bcma_cc_write32(cc, BCMA_CC_PMU_CTL, tmp);
+	tmp |= bcma_pmu_read32(cc, BCMA_CC_PMU_CTL);
+	bcma_pmu_write32(cc, BCMA_CC_PMU_CTL, tmp);
 }
 EXPORT_SYMBOL_GPL(bcma_pmu_spuravoid_pllupdate);
diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h
index 96d8d56..700d0c6 100644
--- a/include/linux/bcma/bcma_driver_chipcommon.h
+++ b/include/linux/bcma/bcma_driver_chipcommon.h
@@ -217,6 +217,11 @@
 #define	 BCMA_CC_CLKDIV_JTAG_SHIFT	8
 #define	 BCMA_CC_CLKDIV_UART		0x000000FF
 #define BCMA_CC_CAP_EXT			0x00AC		/* Capabilities */
+#define  BCMA_CC_CAP_EXT_SECI_PRESENT	0x00000001
+#define  BCMA_CC_CAP_EXT_GSIO_PRESENT	0x00000002
+#define  BCMA_CC_CAP_EXT_GCI_PRESENT	0x00000004
+#define  BCMA_CC_CAP_EXT_SECI_PUART_PRESENT		0x00000008    /* UART present */
+#define  BCMA_CC_CAP_EXT_AOB_PRESENT	0x00000040
 #define BCMA_CC_PLLONDELAY		0x00B0		/* Rev >= 4 only */
 #define BCMA_CC_FREFSELDELAY		0x00B4		/* Rev >= 4 only */
 #define BCMA_CC_SLOWCLKCTL		0x00B8		/* 6 <= Rev <= 9 only */
@@ -566,6 +571,7 @@
  * Check availability with ((struct bcma_chipcommon)->capabilities & BCMA_CC_CAP_PMU)
  */
 struct bcma_chipcommon_pmu {
+	struct bcma_device *core;	/* Can be separated core or just ChipCommon one */
 	u8 rev;			/* PMU revision */
 	u32 crystalfreq;	/* The active crystal frequency (in kHz) */
 };
@@ -660,6 +666,19 @@ struct bcma_drv_cc_b {
 #define bcma_cc_maskset32(cc, offset, mask, set) \
 	bcma_cc_write32(cc, offset, (bcma_cc_read32(cc, offset) & (mask)) | (set))
 
+/* PMU registers access */
+#define bcma_pmu_read32(cc, offset) \
+	bcma_read32((cc)->pmu.core, offset)
+#define bcma_pmu_write32(cc, offset, val) \
+	bcma_write32((cc)->pmu.core, offset, val)
+
+#define bcma_pmu_mask32(cc, offset, mask) \
+	bcma_pmu_write32(cc, offset, bcma_pmu_read32(cc, offset) & (mask))
+#define bcma_pmu_set32(cc, offset, set) \
+	bcma_pmu_write32(cc, offset, bcma_pmu_read32(cc, offset) | (set))
+#define bcma_pmu_maskset32(cc, offset, mask, set) \
+	bcma_pmu_write32(cc, offset, (bcma_pmu_read32(cc, offset) & (mask)) | (set))
+
 extern u32 bcma_chipco_watchdog_timer_set(struct bcma_drv_cc *cc, u32 ticks);
 
 extern u32 bcma_chipco_get_alp_clock(struct bcma_drv_cc *cc);
-- 
cgit v0.10.2


From 36e8072ebeb1981074c7ec1b63bae3ecc3295b4b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Wed, 20 Jan 2016 16:46:04 +0100
Subject: brcmfmac: fix setting primary channel for 80 MHz width
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

First of all it changes the way we calculate primary channel offset. If
we use e.g. 80 MHz channel with primary frequency 5180 MHz (which means
center frequency is 5210 MHz) it makes sense to calculate primary offset
as -30 MHz.
Then it fixes values we compare primary_offset with. We were comparing
offset in MHz against -2 or 2 which was resulting in picking a wrong
primary channel.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 7b01e4d..d00c5c1 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -247,7 +247,7 @@ static u16 chandef_to_chanspec(struct brcmu_d11inf *d11inf,
 	brcmf_dbg(TRACE, "chandef: control %d center %d width %d\n",
 		  ch->chan->center_freq, ch->center_freq1, ch->width);
 	ch_inf.chnum = ieee80211_frequency_to_channel(ch->center_freq1);
-	primary_offset = ch->center_freq1 - ch->chan->center_freq;
+	primary_offset = ch->chan->center_freq - ch->center_freq1;
 	switch (ch->width) {
 	case NL80211_CHAN_WIDTH_20:
 	case NL80211_CHAN_WIDTH_20_NOHT:
@@ -256,24 +256,21 @@ static u16 chandef_to_chanspec(struct brcmu_d11inf *d11inf,
 		break;
 	case NL80211_CHAN_WIDTH_40:
 		ch_inf.bw = BRCMU_CHAN_BW_40;
-		if (primary_offset < 0)
+		if (primary_offset > 0)
 			ch_inf.sb = BRCMU_CHAN_SB_U;
 		else
 			ch_inf.sb = BRCMU_CHAN_SB_L;
 		break;
 	case NL80211_CHAN_WIDTH_80:
 		ch_inf.bw = BRCMU_CHAN_BW_80;
-		if (primary_offset < 0) {
-			if (primary_offset < -CH_10MHZ_APART)
-				ch_inf.sb = BRCMU_CHAN_SB_UU;
-			else
-				ch_inf.sb = BRCMU_CHAN_SB_UL;
-		} else {
-			if (primary_offset > CH_10MHZ_APART)
-				ch_inf.sb = BRCMU_CHAN_SB_LL;
-			else
-				ch_inf.sb = BRCMU_CHAN_SB_LU;
-		}
+		if (primary_offset == -30)
+			ch_inf.sb = BRCMU_CHAN_SB_LL;
+		else if (primary_offset == -10)
+			ch_inf.sb = BRCMU_CHAN_SB_LU;
+		else if (primary_offset == 10)
+			ch_inf.sb = BRCMU_CHAN_SB_UL;
+		else
+			ch_inf.sb = BRCMU_CHAN_SB_UU;
 		break;
 	case NL80211_CHAN_WIDTH_80P80:
 	case NL80211_CHAN_WIDTH_160:
-- 
cgit v0.10.2


From 4c81acab3816c6941900c2abdb412df6cf34bc74 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Fri, 22 Jan 2016 18:02:54 +0100
Subject: bcma: init serial console directly from ChipCommon code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

UART is connected to and controlled over ChipCommon core. It doesn't
have much to do with MIPS core (where we initialize it currently)
except just existing on embedded systemms. There isn't point of such
cross-core initialization (and we needed #ifdef anyway) so just handle
it in ChipCommon.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/bcma_private.h b/drivers/bcma/bcma_private.h
index 38f1567..7e4ddfb 100644
--- a/drivers/bcma/bcma_private.h
+++ b/drivers/bcma/bcma_private.h
@@ -48,7 +48,6 @@ void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc);
 void bcma_core_chipcommon_init(struct bcma_drv_cc *cc);
 void bcma_chipco_bcm4331_ext_pa_lines_ctl(struct bcma_drv_cc *cc, bool enable);
 #ifdef CONFIG_BCMA_DRIVER_MIPS
-void bcma_chipco_serial_init(struct bcma_drv_cc *cc);
 extern struct platform_device bcma_pflash_dev;
 #endif /* CONFIG_BCMA_DRIVER_MIPS */
 
diff --git a/drivers/bcma/driver_chipcommon.c b/drivers/bcma/driver_chipcommon.c
index 36ee221..bdb73d9 100644
--- a/drivers/bcma/driver_chipcommon.c
+++ b/drivers/bcma/driver_chipcommon.c
@@ -15,6 +15,8 @@
 #include <linux/platform_device.h>
 #include <linux/bcma/bcma.h>
 
+static void bcma_chipco_serial_init(struct bcma_drv_cc *cc);
+
 static inline u32 bcma_cc_write32_masked(struct bcma_drv_cc *cc, u16 offset,
 					 u32 mask, u32 value)
 {
@@ -115,6 +117,8 @@ int bcma_chipco_watchdog_register(struct bcma_drv_cc *cc)
 
 void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc)
 {
+	struct bcma_bus *bus = cc->core->bus;
+
 	if (cc->early_setup_done)
 		return;
 
@@ -129,6 +133,9 @@ void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc)
 	if (cc->capabilities & BCMA_CC_CAP_PMU)
 		bcma_pmu_early_init(cc);
 
+	if (IS_BUILTIN(CONFIG_BCM47XX) && bus->hosttype == BCMA_HOSTTYPE_SOC)
+		bcma_chipco_serial_init(cc);
+
 	cc->early_setup_done = true;
 }
 
@@ -314,9 +321,9 @@ u32 bcma_chipco_gpio_pulldown(struct bcma_drv_cc *cc, u32 mask, u32 value)
 	return res;
 }
 
-#ifdef CONFIG_BCMA_DRIVER_MIPS
-void bcma_chipco_serial_init(struct bcma_drv_cc *cc)
+static void bcma_chipco_serial_init(struct bcma_drv_cc *cc)
 {
+#if IS_BUILTIN(CONFIG_BCM47XX)
 	unsigned int irq;
 	u32 baud_base;
 	u32 i;
@@ -358,5 +365,5 @@ void bcma_chipco_serial_init(struct bcma_drv_cc *cc)
 		ports[i].baud_base = baud_base;
 		ports[i].reg_shift = 0;
 	}
+#endif /* CONFIG_BCM47XX */
 }
-#endif /* CONFIG_BCMA_DRIVER_MIPS */
diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c
index 24424f3..a40a203 100644
--- a/drivers/bcma/driver_mips.c
+++ b/drivers/bcma/driver_mips.c
@@ -337,12 +337,9 @@ static void bcma_core_mips_flash_detect(struct bcma_drv_mips *mcore)
 
 void bcma_core_mips_early_init(struct bcma_drv_mips *mcore)
 {
-	struct bcma_bus *bus = mcore->core->bus;
-
 	if (mcore->early_setup_done)
 		return;
 
-	bcma_chipco_serial_init(&bus->drv_cc);
 	bcma_core_mips_flash_detect(mcore);
 
 	mcore->early_setup_done = true;
-- 
cgit v0.10.2


From 61dba73cdbba8ec5c01b31beaf9e2debc2d2f273 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Sun, 24 Jan 2016 16:37:33 +0100
Subject: bcma: add support for BCM47094
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's another SoC with 32 GPIOs and simplified watchdog handling. It was
tested on D-Link DIR-885L.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/driver_chipcommon.c b/drivers/bcma/driver_chipcommon.c
index bdb73d9..b0f44a2 100644
--- a/drivers/bcma/driver_chipcommon.c
+++ b/drivers/bcma/driver_chipcommon.c
@@ -197,6 +197,7 @@ u32 bcma_chipco_watchdog_timer_set(struct bcma_drv_cc *cc, u32 ticks)
 		struct bcma_bus *bus = cc->core->bus;
 
 		if (bus->chipinfo.id != BCMA_CHIP_ID_BCM4707 &&
+		    bus->chipinfo.id != BCMA_CHIP_ID_BCM47094 &&
 		    bus->chipinfo.id != BCMA_CHIP_ID_BCM53018)
 			bcma_core_set_clockmode(cc->core,
 						ticks ? BCMA_CLKMODE_FAST : BCMA_CLKMODE_DYNAMIC);
diff --git a/drivers/bcma/driver_gpio.c b/drivers/bcma/driver_gpio.c
index 504899a7..77b0738f 100644
--- a/drivers/bcma/driver_gpio.c
+++ b/drivers/bcma/driver_gpio.c
@@ -197,6 +197,7 @@ int bcma_gpio_init(struct bcma_drv_cc *cc)
 	case BCMA_CHIP_ID_BCM4707:
 	case BCMA_CHIP_ID_BCM5357:
 	case BCMA_CHIP_ID_BCM53572:
+	case BCMA_CHIP_ID_BCM47094:
 		chip->ngpio	= 32;
 		break;
 	default:
diff --git a/include/linux/bcma/bcma.h b/include/linux/bcma/bcma.h
index 991ebb4..0367c63 100644
--- a/include/linux/bcma/bcma.h
+++ b/include/linux/bcma/bcma.h
@@ -201,6 +201,7 @@ struct bcma_host_ops {
 #define  BCMA_PKG_ID_BCM4707	1
 #define  BCMA_PKG_ID_BCM4708	2
 #define  BCMA_PKG_ID_BCM4709	0
+#define BCMA_CHIP_ID_BCM47094	53030
 #define BCMA_CHIP_ID_BCM53018	53018
 
 /* Board types (on PCI usually equals to the subsystem dev id) */
-- 
cgit v0.10.2


From 515b399c9a204da70d54916d23c1748ff7083fb8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 26 Jan 2016 17:02:09 +0100
Subject: bcma: claim only 14e4:4365 PCI Dell card with SoftMAC BCM43142
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It seems 14e4:4365 pattern is too generic as there are two devices:
1) 14e4:4365 1028:0016 with SoftMAC BCM43142 chipset
2) 14e4:4365 14e4:4365 with FullMAC BCM4366 chipset
The later one was found in D-Link DIR-885L router and we want to let
brcmfmac handle it.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/host_pci.c b/drivers/bcma/host_pci.c
index 0856189..cae5385 100644
--- a/drivers/bcma/host_pci.c
+++ b/drivers/bcma/host_pci.c
@@ -294,7 +294,7 @@ static const struct pci_device_id bcma_pci_bridge_tbl[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4358) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4359) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4360) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4365) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_BROADCOM, 0x4365, PCI_VENDOR_ID_DELL, 0x0016) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x43a0) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x43a9) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x43aa) },
-- 
cgit v0.10.2


From a4f4abd037c1f8c6a834c95a46a2514d4b350888 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 26 Jan 2016 17:57:01 +0100
Subject: brcmfmac: analyze descriptors of current component only
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

So far we were looking for address descriptors without a check for
crossing current component border. In case of dealing with unsupported
descriptor or descriptor missing at all the code would incorrectly get
data from another component.

Consider this binary-described component from BCM4366 EROM:
4bf83b01	TAG==CI		CID==0x83b
20080201	TAG==CI		PORTS==0+1	WRAPPERS==0+1
18400035	TAG==ADDR	SZ_SZD		TYPE_SLAVE
00050000
18107085	TAG==ADDR	SZ_4K		TYPE_SWRAP

Driver was assigning invalid base address to this core:
brcmfmac:  [6 ] core 0x83b:32 base 0x18109000 wrap 0x18107000
which came from totally different component defined in EROM:
43b36701	TAG==CI		CID==0x367
00000201	TAG==CI		PORTS==0+1	WRAPPERS==0+0
18109005	TAG==ADDR	SZ_4K		TYPE_SLAVE

This change will also allow us to support components without wrapper
address in the future.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index 82e4382..e434e2a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -803,7 +803,14 @@ static int brcmf_chip_dmp_get_regaddr(struct brcmf_chip_priv *ci, u32 *eromaddr,
 				*eromaddr -= 4;
 				return -EFAULT;
 			}
-		} while (desc != DMP_DESC_ADDRESS);
+		} while (desc != DMP_DESC_ADDRESS &&
+			 desc != DMP_DESC_COMPONENT);
+
+		/* stop if we crossed current component border */
+		if (desc == DMP_DESC_COMPONENT) {
+			*eromaddr -= 4;
+			return 0;
+		}
 
 		/* skip upper 32-bit address descriptor */
 		if (val & DMP_DESC_ADDRSIZE_GT32)
-- 
cgit v0.10.2


From 44977b81823f4801d45d61fb9b278b4483923ae1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 26 Jan 2016 17:57:02 +0100
Subject: brcmfmac: allow storing PMU core without wrapper address
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Separated PMU core can be found in new devices and should be used for
accessing PMU registers (which were routed through ChipCommon so far).
This core is one of exceptions that doesn't have or need wrapper address
to be still safely accessible.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index e434e2a..55952d4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -883,7 +883,8 @@ int brcmf_chip_dmp_erom_scan(struct brcmf_chip_priv *ci)
 		rev = (val & DMP_COMP_REVISION) >> DMP_COMP_REVISION_S;
 
 		/* need core with ports */
-		if (nmw + nsw == 0)
+		if (nmw + nsw == 0 &&
+		    id != BCMA_CORE_PMU)
 			continue;
 
 		/* try to obtain register address info */
-- 
cgit v0.10.2


From 9befe9195a17856f8bf5ede7a5c067bd0a281be6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 26 Jan 2016 17:57:03 +0100
Subject: brcmfmac: read extended capabilities of ChipCommon core
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is an extra bitfield with info about some present hardware.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index 55952d4..f4a4d00 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -1025,6 +1025,9 @@ static int brcmf_chip_setup(struct brcmf_chip_priv *chip)
 	/* get chipcommon capabilites */
 	pub->cc_caps = chip->ops->read32(chip->ctx,
 					 CORE_CC_REG(base, capabilities));
+	pub->cc_caps_ext = chip->ops->read32(chip->ctx,
+					     CORE_CC_REG(base,
+							 capabilities_ext));
 
 	/* get pmu caps & rev */
 	if (pub->cc_caps & CC_CAP_PMU) {
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
index f6b5fee..cb9145f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
@@ -27,6 +27,7 @@
  * @chip: chip identifier.
  * @chiprev: chip revision.
  * @cc_caps: chipcommon core capabilities.
+ * @cc_caps_ext: chipcommon core extended capabilities.
  * @pmucaps: PMU capabilities.
  * @pmurev: PMU revision.
  * @rambase: RAM base address (only applicable for ARM CR4 chips).
@@ -38,6 +39,7 @@ struct brcmf_chip {
 	u32 chip;
 	u32 chiprev;
 	u32 cc_caps;
+	u32 cc_caps_ext;
 	u32 pmucaps;
 	u32 pmurev;
 	u32 rambase;
-- 
cgit v0.10.2


From e2b397f18cc6423c2dad87f9a633add762d869dd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 26 Jan 2016 17:57:04 +0100
Subject: brcmfmac: access PMU registers using standalone PMU core if available
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On recent Broadcom chipsets PMU is present as separated core and it
can't be accessed using ChipCommon anymore as it fails with e.g.:
[   18.198412] Unhandled fault: imprecise external abort (0x1406) at 0xb6da200f

Add a new helper function that will return a proper core that should be
used for accessing PMU registers.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
index f4a4d00..0e8f2a0 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c
@@ -1014,6 +1014,7 @@ static int brcmf_chip_setup(struct brcmf_chip_priv *chip)
 {
 	struct brcmf_chip *pub;
 	struct brcmf_core_priv *cc;
+	struct brcmf_core *pmu;
 	u32 base;
 	u32 val;
 	int ret = 0;
@@ -1030,9 +1031,10 @@ static int brcmf_chip_setup(struct brcmf_chip_priv *chip)
 							 capabilities_ext));
 
 	/* get pmu caps & rev */
+	pmu = brcmf_chip_get_pmu(pub); /* after reading cc_caps_ext */
 	if (pub->cc_caps & CC_CAP_PMU) {
 		val = chip->ops->read32(chip->ctx,
-					CORE_CC_REG(base, pmucapabilities));
+					CORE_CC_REG(pmu->base, pmucapabilities));
 		pub->pmurev = val & PCAP_REV_MASK;
 		pub->pmucaps = val;
 	}
@@ -1131,6 +1133,23 @@ struct brcmf_core *brcmf_chip_get_chipcommon(struct brcmf_chip *pub)
 	return &cc->pub;
 }
 
+struct brcmf_core *brcmf_chip_get_pmu(struct brcmf_chip *pub)
+{
+	struct brcmf_core *cc = brcmf_chip_get_chipcommon(pub);
+	struct brcmf_core *pmu;
+
+	/* See if there is separated PMU core available */
+	if (cc->rev >= 35 &&
+	    pub->cc_caps_ext & BCMA_CC_CAP_EXT_AOB_PRESENT) {
+		pmu = brcmf_chip_get_core(pub, BCMA_CORE_PMU);
+		if (pmu)
+			return pmu;
+	}
+
+	/* Fallback to ChipCommon core for older hardware */
+	return cc;
+}
+
 bool brcmf_chip_iscoreup(struct brcmf_core *pub)
 {
 	struct brcmf_core_priv *core;
@@ -1301,6 +1320,7 @@ bool brcmf_chip_sr_capable(struct brcmf_chip *pub)
 {
 	u32 base, addr, reg, pmu_cc3_mask = ~0;
 	struct brcmf_chip_priv *chip;
+	struct brcmf_core *pmu = brcmf_chip_get_pmu(pub);
 
 	brcmf_dbg(TRACE, "Enter\n");
 
@@ -1320,9 +1340,9 @@ bool brcmf_chip_sr_capable(struct brcmf_chip *pub)
 	case BRCM_CC_4335_CHIP_ID:
 	case BRCM_CC_4339_CHIP_ID:
 		/* read PMU chipcontrol register 3 */
-		addr = CORE_CC_REG(base, chipcontrol_addr);
+		addr = CORE_CC_REG(pmu->base, chipcontrol_addr);
 		chip->ops->write32(chip->ctx, addr, 3);
-		addr = CORE_CC_REG(base, chipcontrol_data);
+		addr = CORE_CC_REG(pmu->base, chipcontrol_data);
 		reg = chip->ops->read32(chip->ctx, addr);
 		return (reg & pmu_cc3_mask) != 0;
 	case BRCM_CC_43430_CHIP_ID:
@@ -1330,12 +1350,12 @@ bool brcmf_chip_sr_capable(struct brcmf_chip *pub)
 		reg = chip->ops->read32(chip->ctx, addr);
 		return reg != 0;
 	default:
-		addr = CORE_CC_REG(base, pmucapabilities_ext);
+		addr = CORE_CC_REG(pmu->base, pmucapabilities_ext);
 		reg = chip->ops->read32(chip->ctx, addr);
 		if ((reg & PCAPEXT_SR_SUPPORTED_MASK) == 0)
 			return false;
 
-		addr = CORE_CC_REG(base, retention_ctl);
+		addr = CORE_CC_REG(pmu->base, retention_ctl);
 		reg = chip->ops->read32(chip->ctx, addr);
 		return (reg & (PMU_RCTL_MACPHY_DISABLE_MASK |
 			       PMU_RCTL_LOGIC_DISABLE_MASK)) == 0;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
index cb9145f..dd0ec3e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.h
@@ -85,6 +85,7 @@ struct brcmf_chip *brcmf_chip_attach(void *ctx,
 void brcmf_chip_detach(struct brcmf_chip *chip);
 struct brcmf_core *brcmf_chip_get_core(struct brcmf_chip *chip, u16 coreid);
 struct brcmf_core *brcmf_chip_get_chipcommon(struct brcmf_chip *chip);
+struct brcmf_core *brcmf_chip_get_pmu(struct brcmf_chip *pub);
 bool brcmf_chip_iscoreup(struct brcmf_core *core);
 void brcmf_chip_coredisable(struct brcmf_core *core, u32 prereset, u32 reset);
 void brcmf_chip_resetcore(struct brcmf_core *core, u32 prereset, u32 reset,
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index dd66143..80b5d47 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -3615,7 +3615,6 @@ brcmf_sdio_drivestrengthinit(struct brcmf_sdio_dev *sdiodev,
 	const struct sdiod_drive_str *str_tab = NULL;
 	u32 str_mask;
 	u32 str_shift;
-	u32 base;
 	u32 i;
 	u32 drivestrength_sel = 0;
 	u32 cc_data_temp;
@@ -3658,14 +3657,15 @@ brcmf_sdio_drivestrengthinit(struct brcmf_sdio_dev *sdiodev,
 	}
 
 	if (str_tab != NULL) {
+		struct brcmf_core *pmu = brcmf_chip_get_pmu(ci);
+
 		for (i = 0; str_tab[i].strength != 0; i++) {
 			if (drivestrength >= str_tab[i].strength) {
 				drivestrength_sel = str_tab[i].sel;
 				break;
 			}
 		}
-		base = brcmf_chip_get_chipcommon(ci)->base;
-		addr = CORE_CC_REG(base, chipcontrol_addr);
+		addr = CORE_CC_REG(pmu->base, chipcontrol_addr);
 		brcmf_sdiod_regwl(sdiodev, addr, 1, NULL);
 		cc_data_temp = brcmf_sdiod_regrl(sdiodev, addr, NULL);
 		cc_data_temp &= ~str_mask;
@@ -3835,8 +3835,7 @@ brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 		goto fail;
 
 	/* set PMUControl so a backplane reset does PMU state reload */
-	reg_addr = CORE_CC_REG(brcmf_chip_get_chipcommon(bus->ci)->base,
-			       pmucontrol);
+	reg_addr = CORE_CC_REG(brcmf_chip_get_pmu(bus->ci)->base, pmucontrol);
 	reg_val = brcmf_sdiod_regrl(bus->sdiodev, reg_addr, &err);
 	if (err)
 		goto fail;
-- 
cgit v0.10.2


From f66ab2a7da8ffd78ba26048f8108577ba73a1828 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Tue, 26 Jan 2016 17:57:05 +0100
Subject: brcmfmac: add support for 14e4:4365 PCI ID with BCM4366 chipset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On Broadcom ARM routers BCM4366 cards are available with 14e4:4365 ID.
Unfortunately this ID was already used by Broadcom for cards with
BCM43142, a totally different chipset requiring SoftMAC driver. To avoid
a conflict between brcmfmac and bcma use more specific ID entry with
subvendor and subdevice specified.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 0480b70..d5f9ef4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -1951,6 +1951,9 @@ static const struct dev_pm_ops brcmf_pciedrvr_pm = {
 
 #define BRCMF_PCIE_DEVICE(dev_id)	{ BRCM_PCIE_VENDOR_ID_BROADCOM, dev_id,\
 	PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_NETWORK_OTHER << 8, 0xffff00, 0 }
+#define BRCMF_PCIE_DEVICE_SUB(dev_id, subvend, subdev)	{ \
+	BRCM_PCIE_VENDOR_ID_BROADCOM, dev_id,\
+	subvend, subdev, PCI_CLASS_NETWORK_OTHER << 8, 0xffff00, 0 }
 
 static struct pci_device_id brcmf_pcie_devid_table[] = {
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4350_DEVICE_ID),
@@ -1966,6 +1969,7 @@ static struct pci_device_id brcmf_pcie_devid_table[] = {
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4365_DEVICE_ID),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4365_2G_DEVICE_ID),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4365_5G_DEVICE_ID),
+	BRCMF_PCIE_DEVICE_SUB(0x4365, BRCM_PCIE_VENDOR_ID_BROADCOM, 0x4365),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4366_DEVICE_ID),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4366_2G_DEVICE_ID),
 	BRCMF_PCIE_DEVICE(BRCM_PCIE_4366_5G_DEVICE_ID),
-- 
cgit v0.10.2


From d1e61b86a2b1b99e5dcd4d41aa23d033b83542a1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Sun, 31 Jan 2016 12:14:34 +0100
Subject: brcmfmac: treat NULL character in NVRAM as separator
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Platform NVRAM (stored on a flash partition) has entries separated by a
NULL (\0) char. Our parsing code switches from VALUE state to IDLE
whenever it meets a NULL (\0). When that happens our IDLE handler should
simply consume it and analyze whatever is placed ahead.

This fixes harmless warnings spamming debugging output:
[  155.165624] brcmfmac: brcmf_nvram_handle_idle warning: ln=1:col=20: ignoring invalid character
[  155.180806] brcmfmac: brcmf_nvram_handle_idle warning: ln=1:col=44: ignoring invalid character
[  155.195971] brcmfmac: brcmf_nvram_handle_idle warning: ln=1:col=63: ignoring invalid character

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c
index 1365c12..7269056 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/firmware.c
@@ -93,7 +93,7 @@ static enum nvram_parser_state brcmf_nvram_handle_idle(struct nvram_parser *nvp)
 	c = nvp->data[nvp->pos];
 	if (c == '\n')
 		return COMMENT;
-	if (is_whitespace(c))
+	if (is_whitespace(c) || c == '\0')
 		goto proceed;
 	if (c == '#')
 		return COMMENT;
-- 
cgit v0.10.2


From 97f1a17109272b9f060cbd6d1d5be41528643000 Mon Sep 17 00:00:00 2001
From: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
Date: Mon, 25 Jan 2016 11:47:29 +0100
Subject: brcmfmac: sdio: Increase the default timeouts a bit

On a Radxa Rock2 board with a Ampak AP6335 (Broadcom 4339 core) it seems
the card responds very quickly most of the time, unfortunately during
initialisation it sometimes seems to take just a bit over 2 seconds to
respond.

This results intialization failing with message like:
  brcmf_c_preinit_dcmds: Retreiving cur_etheraddr failed, -52
  brcmf_bus_start: failed: -52
  brcmf_sdio_firmware_callback: dongle is not responding

Increasing the timeout to allow for a bit more headroom allows the
card to initialize reliably.

A quick search online after diagnosing/fixing this showed that Google
has a similar patch in their ChromeOS tree, so this doesn't seem
specific to the board I'm using.

Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Acked-by: Arend van Spriel <arend@broadcom.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index 80b5d47..6776e45 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -45,8 +45,8 @@
 #include "chip.h"
 #include "firmware.h"
 
-#define DCMD_RESP_TIMEOUT	msecs_to_jiffies(2000)
-#define CTL_DONE_TIMEOUT	msecs_to_jiffies(2000)
+#define DCMD_RESP_TIMEOUT	msecs_to_jiffies(2500)
+#define CTL_DONE_TIMEOUT	msecs_to_jiffies(2500)
 
 #ifdef DEBUG
 
-- 
cgit v0.10.2


From 48dc5fb3ba53b20418de8514700f63d88c5de3a3 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 28 Jan 2016 22:58:28 +0100
Subject: hostap: avoid uninitialized variable use in hfa384x_get_rid

The driver reads a value from hfa384x_from_bap(), which may fail,
and then assigns the value to a local variable. gcc detects that
in in the failure case, the 'rlen' variable now contains
uninitialized data:

In file included from ../drivers/net/wireless/intersil/hostap/hostap_pci.c:220:0:
drivers/net/wireless/intersil/hostap/hostap_hw.c: In function 'hfa384x_get_rid':
drivers/net/wireless/intersil/hostap/hostap_hw.c:842:5: warning: 'rec' may be used uninitialized in this function [-Wmaybe-uninitialized]
  if (le16_to_cpu(rec.len) == 0) {

This restructures the function as suggested by Russell King, to
make it more readable and get more reliable error handling, by
handling each failure mode using a goto.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/intersil/hostap/hostap_hw.c b/drivers/net/wireless/intersil/hostap/hostap_hw.c
index 6df3ee5..515aa3f 100644
--- a/drivers/net/wireless/intersil/hostap/hostap_hw.c
+++ b/drivers/net/wireless/intersil/hostap/hostap_hw.c
@@ -836,25 +836,30 @@ static int hfa384x_get_rid(struct net_device *dev, u16 rid, void *buf, int len,
 	spin_lock_bh(&local->baplock);
 
 	res = hfa384x_setup_bap(dev, BAP0, rid, 0);
-	if (!res)
-		res = hfa384x_from_bap(dev, BAP0, &rec, sizeof(rec));
+	if (res)
+		goto unlock;
+
+	res = hfa384x_from_bap(dev, BAP0, &rec, sizeof(rec));
+	if (res)
+		goto unlock;
 
 	if (le16_to_cpu(rec.len) == 0) {
 		/* RID not available */
 		res = -ENODATA;
+		goto unlock;
 	}
 
 	rlen = (le16_to_cpu(rec.len) - 1) * 2;
-	if (!res && exact_len && rlen != len) {
+	if (exact_len && rlen != len) {
 		printk(KERN_DEBUG "%s: hfa384x_get_rid - RID len mismatch: "
 		       "rid=0x%04x, len=%d (expected %d)\n",
 		       dev->name, rid, rlen, len);
 		res = -ENODATA;
 	}
 
-	if (!res)
-		res = hfa384x_from_bap(dev, BAP0, buf, len);
+	res = hfa384x_from_bap(dev, BAP0, buf, len);
 
+unlock:
 	spin_unlock_bh(&local->baplock);
 	mutex_unlock(&local->rid_bap_mtx);
 
-- 
cgit v0.10.2


From 44ca509cb9769f27ad8f7d662910e31329eed46f Mon Sep 17 00:00:00 2001
From: Nachiket Kukade <kukaden@marvell.com>
Date: Sun, 31 Jan 2016 23:44:46 -0800
Subject: mwifiex: fix bandwidth display problem

Instead of using HT info from beacon IEs, use HT info from
association response frame to update bandwidth in
cfg80211_get_channel handler.

Signed-off-by: Nachiket Kukade <kukaden@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index 8461553..f2dce81 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -3793,8 +3793,8 @@ static int mwifiex_cfg80211_get_channel(struct wiphy *wiphy,
 		freq = ieee80211_channel_to_frequency(curr_bss->channel, band);
 		chan = ieee80211_get_channel(wiphy, freq);
 
-		if (curr_bss->bcn_ht_oper) {
-			second_chan_offset = curr_bss->bcn_ht_oper->ht_param &
+		if (priv->ht_param_present) {
+			second_chan_offset = priv->assoc_resp_ht_param &
 					IEEE80211_HT_PARAM_CHA_SEC_OFFSET;
 			chan_type = mwifiex_sec_chan_offset_to_chan_type
 							(second_chan_offset);
diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h
index 4af9168..c134cf8 100644
--- a/drivers/net/wireless/marvell/mwifiex/fw.h
+++ b/drivers/net/wireless/marvell/mwifiex/fw.h
@@ -1063,7 +1063,7 @@ struct ieee_types_assoc_rsp {
 	__le16 cap_info_bitmap;
 	__le16 status_code;
 	__le16 a_id;
-	u8 ie_buffer[1];
+	u8 ie_buffer[0];
 } __packed;
 
 struct host_cmd_ds_802_11_associate_rsp {
diff --git a/drivers/net/wireless/marvell/mwifiex/join.c b/drivers/net/wireless/marvell/mwifiex/join.c
index cc09a81..62211fc 100644
--- a/drivers/net/wireless/marvell/mwifiex/join.c
+++ b/drivers/net/wireless/marvell/mwifiex/join.c
@@ -644,6 +644,8 @@ int mwifiex_ret_802_11_associate(struct mwifiex_private *priv,
 	struct mwifiex_bssdescriptor *bss_desc;
 	bool enable_data = true;
 	u16 cap_info, status_code, aid;
+	const u8 *ie_ptr;
+	struct ieee80211_ht_operation *assoc_resp_ht_oper;
 
 	assoc_rsp = (struct ieee_types_assoc_rsp *) &resp->params;
 
@@ -733,6 +735,19 @@ int mwifiex_ret_802_11_associate(struct mwifiex_private *priv,
 			= ((bss_desc->wmm_ie.qos_info_bitmap &
 				IEEE80211_WMM_IE_AP_QOSINFO_UAPSD) ? 1 : 0);
 
+	/* Store the bandwidth information from assoc response */
+	ie_ptr = cfg80211_find_ie(WLAN_EID_HT_OPERATION, assoc_rsp->ie_buffer,
+				  priv->assoc_rsp_size
+				  - sizeof(struct ieee_types_assoc_rsp));
+	if (ie_ptr) {
+		assoc_resp_ht_oper = (struct ieee80211_ht_operation *)(ie_ptr
+					+ sizeof(struct ieee_types_header));
+		priv->assoc_resp_ht_param = assoc_resp_ht_oper->ht_param;
+		priv->ht_param_present = true;
+	} else {
+		priv->ht_param_present = false;
+	}
+
 	mwifiex_dbg(priv->adapter, INFO,
 		    "info: ASSOC_RESP: curr_pkt_filter is %#x\n",
 		    priv->curr_pkt_filter);
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index c755be54c..89a1576 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -674,6 +674,8 @@ struct mwifiex_private {
 	struct mwifiex_ds_mem_rw mem_rw;
 	struct sk_buff_head bypass_txq;
 	struct mwifiex_user_scan_chan hidden_chan[MWIFIEX_USER_SCAN_CHAN_MAX];
+	u8 assoc_resp_ht_param;
+	bool ht_param_present;
 };
 
 
diff --git a/drivers/net/wireless/marvell/mwifiex/sta_event.c b/drivers/net/wireless/marvell/mwifiex/sta_event.c
index fd8061c..070bce4 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_event.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_event.c
@@ -92,6 +92,9 @@ mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason_code)
 	priv->is_data_rate_auto = true;
 	priv->data_rate = 0;
 
+	priv->assoc_resp_ht_param = 0;
+	priv->ht_param_present = false;
+
 	if ((GET_BSS_ROLE(priv) == MWIFIEX_BSS_ROLE_STA ||
 	     GET_BSS_ROLE(priv) == MWIFIEX_BSS_ROLE_UAP) && priv->hist_data)
 		mwifiex_hist_data_reset(priv);
-- 
cgit v0.10.2


From efdf0e393e76aa18e9c2ec0ca9b3a71ad410994b Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 2 Feb 2016 13:00:05 +0300
Subject: mwifiex: fix a reversed condition

The NULL test here is reversed.

Fixes: 7d7f07d8c5d3 ('mwifiex: add wowlan net-detect support')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c
index fc8d8ca..489f7a9 100644
--- a/drivers/net/wireless/marvell/mwifiex/scan.c
+++ b/drivers/net/wireless/marvell/mwifiex/scan.c
@@ -2196,7 +2196,7 @@ int mwifiex_ret_802_11_scan(struct mwifiex_private *priv,
 
 			pmatch = adapter->nd_info->matches[idx];
 
-			if (!pmatch) {
+			if (pmatch) {
 				memset(pmatch, 0, sizeof(*pmatch));
 				if (chan_band_tlv) {
 					pmatch->n_channels = 1;
-- 
cgit v0.10.2


From 0a7701b4defcebc7ce461355e6d9478df313b084 Mon Sep 17 00:00:00 2001
From: Andreas Kemnade <andreas@kemnade.info>
Date: Sat, 30 Jan 2016 18:01:51 +0100
Subject: libertas: fix pointer bugs for PS_MODE commands

struct cmd_ds_802_11_ps_mode
contains the command header and a pointer to it was
initialized with data points to the body which leads to
mis-interpretation of the cmd_ds_802_11_ps_mode.action member.
cmd[0] contains the header, &cmd[1] points beyond that.
cmdnode->cmdbuf is a pointer to the command buffer
This piece of code was unused since power saving was
not enabled.

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/cmd.c b/drivers/net/wireless/marvell/libertas/cmd.c
index 0387a5b..40467d6 100644
--- a/drivers/net/wireless/marvell/libertas/cmd.c
+++ b/drivers/net/wireless/marvell/libertas/cmd.c
@@ -957,7 +957,7 @@ static void lbs_queue_cmd(struct lbs_private *priv,
 
 	/* Exit_PS command needs to be queued in the header always. */
 	if (le16_to_cpu(cmdnode->cmdbuf->command) == CMD_802_11_PS_MODE) {
-		struct cmd_ds_802_11_ps_mode *psm = (void *) &cmdnode->cmdbuf;
+		struct cmd_ds_802_11_ps_mode *psm = (void *)cmdnode->cmdbuf;
 
 		if (psm->action == cpu_to_le16(PS_MODE_ACTION_EXIT_PS)) {
 			if (priv->psstate != PS_STATE_FULL_POWER)
@@ -1387,7 +1387,7 @@ int lbs_execute_next_command(struct lbs_private *priv)
 				 * PS command. Ignore it if it is not Exit_PS.
 				 * otherwise send it down immediately.
 				 */
-				struct cmd_ds_802_11_ps_mode *psm = (void *)&cmd[1];
+				struct cmd_ds_802_11_ps_mode *psm = (void *)cmd;
 
 				lbs_deb_host(
 				       "EXEC_NEXT_CMD: PS cmd, action 0x%02x\n",
diff --git a/drivers/net/wireless/marvell/libertas/cmdresp.c b/drivers/net/wireless/marvell/libertas/cmdresp.c
index e5442e8..701125f 100644
--- a/drivers/net/wireless/marvell/libertas/cmdresp.c
+++ b/drivers/net/wireless/marvell/libertas/cmdresp.c
@@ -123,7 +123,10 @@ int lbs_process_command_response(struct lbs_private *priv, u8 *data, u32 len)
 	priv->cmd_timed_out = 0;
 
 	if (respcmd == CMD_RET(CMD_802_11_PS_MODE)) {
-		struct cmd_ds_802_11_ps_mode *psmode = (void *) &resp[1];
+		/* struct cmd_ds_802_11_ps_mode also contains
+		 * the header
+		 */
+		struct cmd_ds_802_11_ps_mode *psmode = (void *)resp;
 		u16 action = le16_to_cpu(psmode->action);
 
 		lbs_deb_host(
-- 
cgit v0.10.2


From fae4f9f78ab11937bc45b0f3625efe2e35aed510 Mon Sep 17 00:00:00 2001
From: Andreas Kemnade <andreas@kemnade.info>
Date: Sat, 30 Jan 2016 18:01:52 +0100
Subject: libertas: check whether bus can do more than polling

If a sdio host does not support sdio irqs, polling is used
instead. That has an impact on performance. Some functionality
should not be enabled then. This add a variable in
libertas_priv to indicate that.

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/dev.h b/drivers/net/wireless/marvell/libertas/dev.h
index 6bd1608..edf710b 100644
--- a/drivers/net/wireless/marvell/libertas/dev.h
+++ b/drivers/net/wireless/marvell/libertas/dev.h
@@ -99,6 +99,7 @@ struct lbs_private {
 	/* Hardware access */
 	void *card;
 	bool iface_running;
+	u8 is_polling; /* host has to poll the card irq */
 	u8 fw_ready;
 	u8 surpriseremoved;
 	u8 setup_fw_on_resume;
diff --git a/drivers/net/wireless/marvell/libertas/if_sdio.c b/drivers/net/wireless/marvell/libertas/if_sdio.c
index 68fd3a9..13eae9f 100644
--- a/drivers/net/wireless/marvell/libertas/if_sdio.c
+++ b/drivers/net/wireless/marvell/libertas/if_sdio.c
@@ -1267,7 +1267,7 @@ static int if_sdio_probe(struct sdio_func *func,
 	priv->reset_card = if_sdio_reset_card;
 	priv->power_save = if_sdio_power_save;
 	priv->power_restore = if_sdio_power_restore;
-
+	priv->is_polling = !(func->card->host->caps & MMC_CAP_SDIO_IRQ);
 	ret = if_sdio_power_on(card);
 	if (ret)
 		goto err_activate_card;
diff --git a/drivers/net/wireless/marvell/libertas/if_usb.c b/drivers/net/wireless/marvell/libertas/if_usb.c
index dff08a2..aba0c99 100644
--- a/drivers/net/wireless/marvell/libertas/if_usb.c
+++ b/drivers/net/wireless/marvell/libertas/if_usb.c
@@ -267,6 +267,7 @@ static int if_usb_probe(struct usb_interface *intf,
 	priv->enter_deep_sleep = NULL;
 	priv->exit_deep_sleep = NULL;
 	priv->reset_deep_sleep_wakeup = NULL;
+	priv->is_polling = false;
 #ifdef CONFIG_OLPC
 	if (machine_is_olpc())
 		priv->reset_card = if_usb_reset_olpc_card;
-- 
cgit v0.10.2


From 57954b94cad77c3253beca805cf493861c7d92b5 Mon Sep 17 00:00:00 2001
From: Andreas Kemnade <andreas@kemnade.info>
Date: Sat, 30 Jan 2016 18:01:53 +0100
Subject: libertas: do not confirm sleep if commands are pending

If the main thread gets one PS AWAKE event and one PS SLEEP event
in one iteration over event_fifo there will never be checks for
commands to be processed, since psstate will always be
PS_STATE_SLEEP or PS_STATE_PRE_SLEEP

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/cmdresp.c b/drivers/net/wireless/marvell/libertas/cmdresp.c
index 701125f..c95bf6d 100644
--- a/drivers/net/wireless/marvell/libertas/cmdresp.c
+++ b/drivers/net/wireless/marvell/libertas/cmdresp.c
@@ -257,6 +257,10 @@ int lbs_process_event(struct lbs_private *priv, u32 event)
 			       "EVENT: in FULL POWER mode, ignoring PS_SLEEP\n");
 			break;
 		}
+		if (!list_empty(&priv->cmdpendingq)) {
+			lbs_deb_cmd("EVENT: commands in queue, do not sleep\n");
+			break;
+		}
 		priv->psstate = PS_STATE_PRE_SLEEP;
 
 		lbs_ps_confirm_sleep(priv);
-- 
cgit v0.10.2


From fada24a5477073dd8fa84baf25ead204dc1dcb18 Mon Sep 17 00:00:00 2001
From: Andreas Kemnade <andreas@kemnade.info>
Date: Sat, 30 Jan 2016 18:01:54 +0100
Subject: libertas: go back to ps mode without commands pending

Removes the old todo block and checks only whether ieee powersave
mode is requested. We still have to check for being connected as
this powersave mode includes logic for regularly waking up and
checking for packets which only makes sense when connected.
For not being connected, another mode is needed.

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/cmd.c b/drivers/net/wireless/marvell/libertas/cmd.c
index 40467d6..4ddd0e5 100644
--- a/drivers/net/wireless/marvell/libertas/cmd.c
+++ b/drivers/net/wireless/marvell/libertas/cmd.c
@@ -1428,40 +1428,14 @@ int lbs_execute_next_command(struct lbs_private *priv)
 		 * check if in power save mode, if yes, put the device back
 		 * to PS mode
 		 */
-#ifdef TODO
-		/*
-		 * This was the old code for libertas+wext. Someone that
-		 * understands this beast should re-code it in a sane way.
-		 *
-		 * I actually don't understand why this is related to WPA
-		 * and to connection status, shouldn't powering should be
-		 * independ of such things?
-		 */
 		if ((priv->psmode != LBS802_11POWERMODECAM) &&
 		    (priv->psstate == PS_STATE_FULL_POWER) &&
-		    ((priv->connect_status == LBS_CONNECTED) ||
-		    lbs_mesh_connected(priv))) {
-			if (priv->secinfo.WPAenabled ||
-			    priv->secinfo.WPA2enabled) {
-				/* check for valid WPA group keys */
-				if (priv->wpa_mcast_key.len ||
-				    priv->wpa_unicast_key.len) {
-					lbs_deb_host(
-					       "EXEC_NEXT_CMD: WPA enabled and GTK_SET"
-					       " go back to PS_SLEEP");
-					lbs_set_ps_mode(priv,
-							PS_MODE_ACTION_ENTER_PS,
-							false);
-				}
-			} else {
-				lbs_deb_host(
-				       "EXEC_NEXT_CMD: cmdpendingq empty, "
-				       "go back to PS_SLEEP");
-				lbs_set_ps_mode(priv, PS_MODE_ACTION_ENTER_PS,
-						false);
-			}
+		    (priv->connect_status == LBS_CONNECTED)) {
+			lbs_deb_host(
+				"EXEC_NEXT_CMD: cmdpendingq empty, go back to PS_SLEEP");
+			lbs_set_ps_mode(priv, PS_MODE_ACTION_ENTER_PS,
+					false);
 		}
-#endif
 	}
 
 	ret = 0;
-- 
cgit v0.10.2


From 0b8802dc5f59a4517e4130dc3606eb973760928a Mon Sep 17 00:00:00 2001
From: Andreas Kemnade <andreas@kemnade.info>
Date: Sat, 30 Jan 2016 18:01:55 +0100
Subject: libertas: fix ps-mode related removal problems

When the device is remove e.g. because of going to suspend
 mode with powersaving enabled, lbs_remove_card tries to exit
powersaving state even when already woken up. That command is
not processed properly in that situation, since the command
processing queue is already stopped, so it waits forever
for the command being processed, so disable it.

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/main.c b/drivers/net/wireless/marvell/libertas/main.c
index 8079560..b35b8bc 100644
--- a/drivers/net/wireless/marvell/libertas/main.c
+++ b/drivers/net/wireless/marvell/libertas/main.c
@@ -1060,7 +1060,12 @@ void lbs_remove_card(struct lbs_private *priv)
 
 	if (priv->psmode == LBS802_11POWERMODEMAX_PSP) {
 		priv->psmode = LBS802_11POWERMODECAM;
-		lbs_set_ps_mode(priv, PS_MODE_ACTION_EXIT_PS, true);
+		/* no need to wakeup if already woken up,
+		 * on suspend, this exit ps command is not processed
+		 * the driver hangs
+		 */
+		if (priv->psstate != PS_STATE_FULL_POWER)
+			lbs_set_ps_mode(priv, PS_MODE_ACTION_EXIT_PS, true);
 	}
 
 	if (priv->is_deep_sleep) {
-- 
cgit v0.10.2


From 143e49458424b80fd004963b6ba88ed583ea62da Mon Sep 17 00:00:00 2001
From: Andreas Kemnade <andreas@kemnade.info>
Date: Sat, 30 Jan 2016 18:01:56 +0100
Subject: libertas: add an cfg80211 interface for powersaving

This patch adds an interface for handling commands like
iwconfig wlanX power on/off. Such an interface formerly existed
when the driver used wext.

While performance with sdio in polling mode without using
powersave mode is quite bad, powersaving mode is unusable,
so do not enable it under such conditions.

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/cfg.c b/drivers/net/wireless/marvell/libertas/cfg.c
index 86955c4..2eea76a 100644
--- a/drivers/net/wireless/marvell/libertas/cfg.c
+++ b/drivers/net/wireless/marvell/libertas/cfg.c
@@ -2039,6 +2039,43 @@ static int lbs_leave_ibss(struct wiphy *wiphy, struct net_device *dev)
 
 
 
+int lbs_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev,
+		       bool enabled, int timeout)
+{
+	struct lbs_private *priv = wiphy_priv(wiphy);
+
+	if  (!(priv->fwcapinfo & FW_CAPINFO_PS)) {
+		if (!enabled)
+			return 0;
+		else
+			return -EINVAL;
+	}
+	/* firmware does not work well with too long latency with power saving
+	 * enabled, so do not enable it if there is only polling, no
+	 * interrupts (like in some sdio hosts which can only
+	 * poll for sdio irqs)
+	 */
+	if  (priv->is_polling) {
+		if (!enabled)
+			return 0;
+		else
+			return -EINVAL;
+	}
+	if (!enabled) {
+		priv->psmode = LBS802_11POWERMODECAM;
+		if (priv->psstate != PS_STATE_FULL_POWER)
+			lbs_set_ps_mode(priv,
+					PS_MODE_ACTION_EXIT_PS,
+					true);
+		return 0;
+	}
+	if (priv->psmode != LBS802_11POWERMODECAM)
+		return 0;
+	priv->psmode = LBS802_11POWERMODEMAX_PSP;
+	if (priv->connect_status == LBS_CONNECTED)
+		lbs_set_ps_mode(priv, PS_MODE_ACTION_ENTER_PS, true);
+	return 0;
+}
 
 /*
  * Initialization
@@ -2057,6 +2094,7 @@ static struct cfg80211_ops lbs_cfg80211_ops = {
 	.change_virtual_intf = lbs_change_intf,
 	.join_ibss = lbs_join_ibss,
 	.leave_ibss = lbs_leave_ibss,
+	.set_power_mgmt = lbs_set_power_mgmt,
 };
 
 
-- 
cgit v0.10.2


From e3e17b773bfe45462b7f3fae20c550025975cb13 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Sat, 6 Feb 2016 11:16:28 -0800
Subject: tcp: fastopen: call tcp_fin() if FIN present in SYNACK

When we acknowledge a FIN, it is not enough to ack the sequence number
and queue the skb into receive queue. We also have to call tcp_fin()
to properly update socket state and send proper poll() notifications.

It seems we also had the problem if we received a SYN packet with the
FIN flag set, but it does not seem an urgent issue, as no known
implementation can do that.

Fixes: 61d2bcae99f6 ("tcp: fastopen: accept data/FIN present in SYNACK message")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 27f4c73..479d535 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -568,6 +568,7 @@ void tcp_rearm_rto(struct sock *sk);
 void tcp_synack_rtt_meas(struct sock *sk, struct request_sock *req);
 void tcp_reset(struct sock *sk);
 void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, struct sk_buff *skb);
+void tcp_fin(struct sock *sk);
 
 /* tcp_timer.c */
 void tcp_init_xmit_timers(struct sock *);
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 6a6e11e..fdb286d 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -154,6 +154,9 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
 	 * as we certainly are not changing upper 32bit value (0)
 	 */
 	tp->bytes_received = skb->len;
+
+	if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
+		tcp_fin(sk);
 }
 
 static struct sock *tcp_fastopen_create_child(struct sock *sk,
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 4add3eb..8194a25 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3995,7 +3995,7 @@ void tcp_reset(struct sock *sk)
  *
  *	If we are in FINWAIT-2, a received FIN moves us to TIME-WAIT.
  */
-static void tcp_fin(struct sock *sk)
+void tcp_fin(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
-- 
cgit v0.10.2


From ac2b335e7149e7dd0c8dcbb2b158a17b78c480a1 Mon Sep 17 00:00:00 2001
From: Paul Mcquade <paulmcquad@gmail.com>
Date: Sat, 17 Oct 2015 21:04:39 +0100
Subject: net: wireless: rt2x00: Fixed Spacing issues

Removed empty spaces before/after parenthesis

Signed-off-by: Paul McQuade <paulmcquad@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ralink/rt2x00/rt61pci.h b/drivers/net/wireless/ralink/rt2x00/rt61pci.h
index 1442075..ab86415 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt61pci.h
+++ b/drivers/net/wireless/ralink/rt2x00/rt61pci.h
@@ -138,14 +138,14 @@
 #define PAIRWISE_TA_TABLE_BASE		0x1a00
 
 #define SHARED_KEY_ENTRY(__idx) \
-	( SHARED_KEY_TABLE_BASE + \
-		((__idx) * sizeof(struct hw_key_entry)) )
+	(SHARED_KEY_TABLE_BASE + \
+		((__idx) * sizeof(struct hw_key_entry)))
 #define PAIRWISE_KEY_ENTRY(__idx) \
-	( PAIRWISE_KEY_TABLE_BASE + \
-		((__idx) * sizeof(struct hw_key_entry)) )
+	(PAIRWISE_KEY_TABLE_BASE + \
+		((__idx) * sizeof(struct hw_key_entry)))
 #define PAIRWISE_TA_ENTRY(__idx) \
-	( PAIRWISE_TA_TABLE_BASE + \
-		((__idx) * sizeof(struct hw_pairwise_ta_entry)) )
+	(PAIRWISE_TA_TABLE_BASE + \
+		((__idx) * sizeof(struct hw_pairwise_ta_entry)))
 
 struct hw_key_entry {
 	u8 key[16];
@@ -180,7 +180,7 @@ struct hw_pairwise_ta_entry {
 #define HW_BEACON_BASE3			0x2f00
 
 #define HW_BEACON_OFFSET(__index) \
-	( HW_BEACON_BASE0 + (__index * 0x0100) )
+	(HW_BEACON_BASE0 + (__index * 0x0100))
 
 /*
  * HOST-MCU shared memory.
@@ -1287,9 +1287,9 @@ struct hw_pairwise_ta_entry {
 /*
  * DMA descriptor defines.
  */
-#define TXD_DESC_SIZE			( 16 * sizeof(__le32) )
-#define TXINFO_SIZE			( 6 * sizeof(__le32) )
-#define RXD_DESC_SIZE			( 16 * sizeof(__le32) )
+#define TXD_DESC_SIZE			(16 * sizeof(__le32))
+#define TXINFO_SIZE			(6 * sizeof(__le32))
+#define RXD_DESC_SIZE			(16 * sizeof(__le32))
 
 /*
  * TX descriptor format for TX, PRIO and Beacon Ring.
-- 
cgit v0.10.2


From b2cc2dd8ebb885ad1b2ee4c5d35b40d98db211fa Mon Sep 17 00:00:00 2001
From: Paul Mcquade <paulmcquad@gmail.com>
Date: Sat, 17 Oct 2015 22:06:43 +0100
Subject: net: wireless: rt2x00: Space issue

Removed empty spaces before/after parenthesis

Signed-off-by: Paul McQuade <paulmcquad@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
index 3282ddb..8c679f2 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
@@ -107,7 +107,7 @@
  * amount of bytes needed to move the data.
  */
 #define ALIGN_SIZE(__skb, __header) \
-	(  ((unsigned long)((__skb)->data + (__header))) & 3 )
+	(((unsigned long)((__skb)->data + (__header))) & 3)
 
 /*
  * Constants for extra TX headroom for alignment purposes.
@@ -128,14 +128,14 @@
 #define SLOT_TIME		20
 #define SHORT_SLOT_TIME		9
 #define SIFS			10
-#define PIFS			( SIFS + SLOT_TIME )
-#define SHORT_PIFS		( SIFS + SHORT_SLOT_TIME )
-#define DIFS			( PIFS + SLOT_TIME )
-#define SHORT_DIFS		( SHORT_PIFS + SHORT_SLOT_TIME )
-#define EIFS			( SIFS + DIFS + \
-				  GET_DURATION(IEEE80211_HEADER + ACK_SIZE, 10) )
-#define SHORT_EIFS		( SIFS + SHORT_DIFS + \
-				  GET_DURATION(IEEE80211_HEADER + ACK_SIZE, 10) )
+#define PIFS			(SIFS + SLOT_TIME)
+#define SHORT_PIFS		(SIFS + SHORT_SLOT_TIME)
+#define DIFS			(PIFS + SLOT_TIME)
+#define SHORT_DIFS		(SHORT_PIFS + SHORT_SLOT_TIME)
+#define EIFS			(SIFS + DIFS + \
+				  GET_DURATION(IEEE80211_HEADER + ACK_SIZE, 10))
+#define SHORT_EIFS		(SIFS + SHORT_DIFS + \
+				  GET_DURATION(IEEE80211_HEADER + ACK_SIZE, 10))
 
 enum rt2x00_chip_intf {
 	RT2X00_CHIP_INTF_PCI,
-- 
cgit v0.10.2


From 5b451715e94d43fdb8f3cca67d1ec72184f294f9 Mon Sep 17 00:00:00 2001
From: Paul Mcquade <paulmcquad@gmail.com>
Date: Sat, 17 Oct 2015 22:11:23 +0100
Subject: net: wireless: rt2x00: Space Required

Space needed before open parenthesis

Signed-off-by: Paul McQuade <paulmcquad@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c b/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c
index 90fdb02..25ee3cb 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c
@@ -629,7 +629,7 @@ static struct dentry *rt2x00debug_create_file_chipset(const char *name,
 	data += sprintf(data, "register\tbase\twords\twordsize\n");
 #define RT2X00DEBUGFS_SPRINTF_REGISTER(__name)			\
 {								\
-	if(debug->__name.read)					\
+	if (debug->__name.read)					\
 		data += sprintf(data, __stringify(__name)	\
 				"\t%d\t%d\t%d\n",		\
 				debug->__name.word_base,	\
@@ -699,7 +699,7 @@ void rt2x00debug_register(struct rt2x00_dev *rt2x00dev)
 
 #define RT2X00DEBUGFS_CREATE_REGISTER_ENTRY(__intf, __name)			\
 ({										\
-	if(debug->__name.read) {						\
+	if (debug->__name.read) {						\
 		(__intf)->__name##_off_entry =					\
 		debugfs_create_u32(__stringify(__name) "_offset",		\
 				       S_IRUSR | S_IWUSR,			\
-- 
cgit v0.10.2


From 6970cd446c25a4634f03c1adf0c12c010235bf3e Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:02 -0800
Subject: mwifiex: display right transmit packet delay

drv_pkt_delay_max should be assigned non-zero value, so that
packet delay can be accumulate in the right way.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 89a1576..0433977 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -298,6 +298,7 @@ struct mwifiex_tid_tbl {
 #define WMM_HIGHEST_PRIORITY		7
 #define HIGH_PRIO_TID				7
 #define LOW_PRIO_TID				0
+#define MWIFIEX_WMM_DRV_DELAY_MAX 510
 
 struct mwifiex_wmm_desc {
 	struct mwifiex_tid_tbl tid_tbl_ptr[MAX_NUM_TID];
diff --git a/drivers/net/wireless/marvell/mwifiex/wmm.c b/drivers/net/wireless/marvell/mwifiex/wmm.c
index 499e5a7..0eb2465 100644
--- a/drivers/net/wireless/marvell/mwifiex/wmm.c
+++ b/drivers/net/wireless/marvell/mwifiex/wmm.c
@@ -438,6 +438,7 @@ mwifiex_wmm_init(struct mwifiex_adapter *adapter)
 		mwifiex_set_ba_params(priv);
 		mwifiex_reset_11n_rx_seq_num(priv);
 
+		priv->wmm.drv_pkt_delay_max = MWIFIEX_WMM_DRV_DELAY_MAX;
 		atomic_set(&priv->wmm.tx_pkts_queued, 0);
 		atomic_set(&priv->wmm.highest_queued_prio, HIGH_PRIO_TID);
 	}
-- 
cgit v0.10.2


From af05148392f50490c662dccee6c502d9fcba33e2 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:03 -0800
Subject: mwifiex: process pcie io memory read failure case

It is observed that ioread32 may fail to read pcie register
in certain scenarios, this patch handles these cases.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index 918e049..b9fe181 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -323,6 +323,8 @@ static int mwifiex_read_reg(struct mwifiex_adapter *adapter, int reg, u32 *data)
 	struct pcie_service_card *card = adapter->card;
 
 	*data = ioread32(card->pci_mmap1 + reg);
+	if (*data == 0xffffffff)
+		return 0xffffffff;
 
 	return 0;
 }
-- 
cgit v0.10.2


From 0172404d79ec37e2c7d9dc90f6876dc314dc0fe5 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:04 -0800
Subject: mwifiex: skip firmware dump when read_regs() fails

If we are not able to read registers or PCIe memory,
it means PCIe device is in bad state. We will skip
firmware dump in this case.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index b9fe181..a18d944 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2366,9 +2366,13 @@ mwifiex_pcie_rdwr_firmware(struct mwifiex_adapter *adapter, u8 doneflag)
 {
 	int ret, tries;
 	u8 ctrl_data;
+	u32 fw_status;
 	struct pcie_service_card *card = adapter->card;
 	const struct mwifiex_pcie_card_reg *reg = card->pcie.reg;
 
+	if (mwifiex_read_reg(adapter, reg->fw_status, &fw_status))
+		return RDWR_STATUS_FAILURE;
+
 	ret = mwifiex_write_reg(adapter, reg->fw_dump_ctrl, FW_DUMP_HOST_READY);
 	if (ret) {
 		mwifiex_dbg(adapter, ERROR,
-- 
cgit v0.10.2


From 9a86232213b9498e0c0721fc508f760e7a089e59 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:05 -0800
Subject: mwifiex: keep original structure in decl header file

memory_type_mapping strucuture did not refer to other mwifiex
specific strture. A better software design method would keep
it in decl header file, which does not include other mwifiex
header file.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/decl.h b/drivers/net/wireless/marvell/mwifiex/decl.h
index d9c15cd..c744a70 100644
--- a/drivers/net/wireless/marvell/mwifiex/decl.h
+++ b/drivers/net/wireless/marvell/mwifiex/decl.h
@@ -270,4 +270,25 @@ struct mwifiex_11h_intf_state {
 	bool is_11h_enabled;
 	bool is_11h_active;
 } __packed;
+
+#define MWIFIEX_FW_DUMP_IDX		0xff
+#define MWIFIEX_DRV_INFO_IDX		20
+#define FW_DUMP_MAX_NAME_LEN		8
+#define FW_DUMP_HOST_READY      0xEE
+#define FW_DUMP_DONE			0xFF
+#define FW_DUMP_READ_DONE		0xFE
+
+struct memory_type_mapping {
+	u8 mem_name[FW_DUMP_MAX_NAME_LEN];
+	u8 *mem_ptr;
+	u32 mem_size;
+	u8 done_flag;
+};
+
+enum rdwr_status {
+	RDWR_STATUS_SUCCESS = 0,
+	RDWR_STATUS_FAILURE = 1,
+	RDWR_STATUS_DONE = 2
+};
+
 #endif /* !_MWIFIEX_DECL_H_ */
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 0433977..aea7aee 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -489,26 +489,6 @@ struct mwifiex_roc_cfg {
 	struct ieee80211_channel chan;
 };
 
-#define MWIFIEX_FW_DUMP_IDX		0xff
-#define MWIFIEX_DRV_INFO_IDX		20
-#define FW_DUMP_MAX_NAME_LEN		8
-#define FW_DUMP_HOST_READY		0xEE
-#define FW_DUMP_DONE			0xFF
-#define FW_DUMP_READ_DONE		0xFE
-
-struct memory_type_mapping {
-	u8 mem_name[FW_DUMP_MAX_NAME_LEN];
-	u8 *mem_ptr;
-	u32 mem_size;
-	u8 done_flag;
-};
-
-enum rdwr_status {
-	RDWR_STATUS_SUCCESS = 0,
-	RDWR_STATUS_FAILURE = 1,
-	RDWR_STATUS_DONE = 2
-};
-
 enum mwifiex_iface_work_flags {
 	MWIFIEX_IFACE_WORK_DEVICE_DUMP,
 	MWIFIEX_IFACE_WORK_CARD_RESET,
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.h b/drivers/net/wireless/marvell/mwifiex/pcie.h
index 9700ac3..8f190c4 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.h
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.h
@@ -26,6 +26,7 @@
 #include    <linux/pcieport_if.h>
 #include    <linux/interrupt.h>
 
+#include    "decl.h"
 #include    "main.h"
 
 #define PCIE8766_DEFAULT_FW_NAME "mrvl/pcie8766_uapsta.bin"
-- 
cgit v0.10.2


From 50632092dfda712dec2232d6ddf75e803c397c70 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:06 -0800
Subject: mwifiex: use an extensible framework for firmware dump solution

This patch apply an extensible firmware dump framework,
so that other chipset can be easily added as needed.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index a18d944..c94d31e 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -37,17 +37,6 @@ static struct mwifiex_if_ops pcie_ops;
 
 static struct semaphore add_remove_card_sem;
 
-static struct memory_type_mapping mem_type_mapping_tbl[] = {
-	{"ITCM", NULL, 0, 0xF0},
-	{"DTCM", NULL, 0, 0xF1},
-	{"SQRAM", NULL, 0, 0xF2},
-	{"IRAM", NULL, 0, 0xF3},
-	{"APU", NULL, 0, 0xF4},
-	{"CIU", NULL, 0, 0xF5},
-	{"ICU", NULL, 0, 0xF6},
-	{"MAC", NULL, 0, 0xF7},
-};
-
 static int
 mwifiex_map_pci_memory(struct mwifiex_adapter *adapter, struct sk_buff *skb,
 		       size_t size, int flags)
@@ -206,6 +195,8 @@ static int mwifiex_pcie_probe(struct pci_dev *pdev,
 		card->pcie.blksz_fw_dl = data->blksz_fw_dl;
 		card->pcie.tx_buf_size = data->tx_buf_size;
 		card->pcie.can_dump_fw = data->can_dump_fw;
+		card->pcie.mem_type_mapping_tbl = data->mem_type_mapping_tbl;
+		card->pcie.num_mem_types = data->num_mem_types;
 		card->pcie.can_ext_scan = data->can_ext_scan;
 	}
 
@@ -2373,7 +2364,8 @@ mwifiex_pcie_rdwr_firmware(struct mwifiex_adapter *adapter, u8 doneflag)
 	if (mwifiex_read_reg(adapter, reg->fw_status, &fw_status))
 		return RDWR_STATUS_FAILURE;
 
-	ret = mwifiex_write_reg(adapter, reg->fw_dump_ctrl, FW_DUMP_HOST_READY);
+	ret = mwifiex_write_reg(adapter, reg->fw_dump_ctrl,
+				reg->fw_dump_host_ready);
 	if (ret) {
 		mwifiex_dbg(adapter, ERROR,
 			    "PCIE write err\n");
@@ -2386,11 +2378,11 @@ mwifiex_pcie_rdwr_firmware(struct mwifiex_adapter *adapter, u8 doneflag)
 			return RDWR_STATUS_SUCCESS;
 		if (doneflag && ctrl_data == doneflag)
 			return RDWR_STATUS_DONE;
-		if (ctrl_data != FW_DUMP_HOST_READY) {
+		if (ctrl_data != reg->fw_dump_host_ready) {
 			mwifiex_dbg(adapter, WARN,
 				    "The ctrl reg was changed, re-try again!\n");
 			ret = mwifiex_write_reg(adapter, reg->fw_dump_ctrl,
-						FW_DUMP_HOST_READY);
+						reg->fw_dump_host_ready);
 			if (ret) {
 				mwifiex_dbg(adapter, ERROR,
 					    "PCIE write err\n");
@@ -2418,8 +2410,9 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 	if (!card->pcie.can_dump_fw)
 		return;
 
-	for (idx = 0; idx < ARRAY_SIZE(mem_type_mapping_tbl); idx++) {
-		struct memory_type_mapping *entry = &mem_type_mapping_tbl[idx];
+	for (idx = 0; idx < adapter->num_mem_types; idx++) {
+		struct memory_type_mapping *entry =
+				&adapter->mem_type_mapping_tbl[idx];
 
 		if (entry->mem_ptr) {
 			vfree(entry->mem_ptr);
@@ -2440,8 +2433,8 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 
 	/* Read the length of every memory which will dump */
 	for (idx = 0; idx < dump_num; idx++) {
-		struct memory_type_mapping *entry = &mem_type_mapping_tbl[idx];
-
+		struct memory_type_mapping *entry =
+				&adapter->mem_type_mapping_tbl[idx];
 		stat = mwifiex_pcie_rdwr_firmware(adapter, doneflag);
 		if (stat == RDWR_STATUS_FAILURE)
 			return;
@@ -2457,7 +2450,7 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 		if (memory_size == 0) {
 			mwifiex_dbg(adapter, MSG, "Firmware dump Finished!\n");
 			ret = mwifiex_write_reg(adapter, creg->fw_dump_ctrl,
-						FW_DUMP_READ_DONE);
+						creg->fw_dump_read_done);
 			if (ret) {
 				mwifiex_dbg(adapter, ERROR, "PCIE write err\n");
 				return;
@@ -2762,8 +2755,8 @@ static int mwifiex_register_dev(struct mwifiex_adapter *adapter)
 		return -1;
 
 	adapter->tx_buf_size = card->pcie.tx_buf_size;
-	adapter->mem_type_mapping_tbl = mem_type_mapping_tbl;
-	adapter->num_mem_types = ARRAY_SIZE(mem_type_mapping_tbl);
+	adapter->mem_type_mapping_tbl = card->pcie.mem_type_mapping_tbl;
+	adapter->num_mem_types = card->pcie.num_mem_types;
 	strcpy(adapter->fw_name, card->pcie.firmware);
 	adapter->ext_scan = card->pcie.can_ext_scan;
 
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.h b/drivers/net/wireless/marvell/mwifiex/pcie.h
index 8f190c4..8bc4390 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.h
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.h
@@ -136,6 +136,8 @@ struct mwifiex_pcie_card_reg {
 	u16 fw_dump_ctrl;
 	u16 fw_dump_start;
 	u16 fw_dump_end;
+	u8 fw_dump_host_ready;
+	u8 fw_dump_read_done;
 	u8 msix_support;
 };
 
@@ -203,6 +205,8 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8897 = {
 	.fw_dump_ctrl = 0xcf4,
 	.fw_dump_start = 0xcf8,
 	.fw_dump_end = 0xcff,
+	.fw_dump_host_ready = 0xee,
+	.fw_dump_read_done = 0xfe,
 	.msix_support = 0,
 };
 
@@ -238,12 +242,25 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8997 = {
 	.msix_support = 1,
 };
 
+static struct memory_type_mapping mem_type_mapping_tbl_w8897[] = {
+	{"ITCM", NULL, 0, 0xF0},
+	{"DTCM", NULL, 0, 0xF1},
+	{"SQRAM", NULL, 0, 0xF2},
+	{"IRAM", NULL, 0, 0xF3},
+	{"APU", NULL, 0, 0xF4},
+	{"CIU", NULL, 0, 0xF5},
+	{"ICU", NULL, 0, 0xF6},
+	{"MAC", NULL, 0, 0xF7},
+};
+
 struct mwifiex_pcie_device {
 	const char *firmware;
 	const struct mwifiex_pcie_card_reg *reg;
 	u16 blksz_fw_dl;
 	u16 tx_buf_size;
 	bool can_dump_fw;
+	struct memory_type_mapping *mem_type_mapping_tbl;
+	u8 num_mem_types;
 	bool can_ext_scan;
 };
 
@@ -262,6 +279,8 @@ static const struct mwifiex_pcie_device mwifiex_pcie8897 = {
 	.blksz_fw_dl = MWIFIEX_PCIE_BLOCK_SIZE_FW_DNLD,
 	.tx_buf_size = MWIFIEX_TX_DATA_BUF_SIZE_4K,
 	.can_dump_fw = true,
+	.mem_type_mapping_tbl = mem_type_mapping_tbl_w8897,
+	.num_mem_types = ARRAY_SIZE(mem_type_mapping_tbl_w8897),
 	.can_ext_scan = true,
 };
 
-- 
cgit v0.10.2


From 56486026c282983c390b264493a4619d997b86f1 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:07 -0800
Subject: mwifiex: dynamically increase preallocated firmware dump memory size

This patch increase firmware dump memory 4K each time, until
meet the demand.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/decl.h b/drivers/net/wireless/marvell/mwifiex/decl.h
index c744a70..a184f9f 100644
--- a/drivers/net/wireless/marvell/mwifiex/decl.h
+++ b/drivers/net/wireless/marvell/mwifiex/decl.h
@@ -122,6 +122,8 @@
 #define BLOCK_NUMBER_OFFSET		15
 #define SDIO_HEADER_OFFSET		28
 
+#define MWIFIEX_SIZE_4K 0x4000
+
 enum mwifiex_bss_type {
 	MWIFIEX_BSS_TYPE_STA = 0,
 	MWIFIEX_BSS_TYPE_UAP = 1,
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index c94d31e..a62a09b 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2402,7 +2402,8 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 	struct pcie_service_card *card = adapter->card;
 	const struct mwifiex_pcie_card_reg *creg = card->pcie.reg;
 	unsigned int reg, reg_start, reg_end;
-	u8 *dbg_ptr, *end_ptr, dump_num, idx, i, read_reg, doneflag = 0;
+	u8 *dbg_ptr, *end_ptr, *tmp_ptr, dump_num;
+	u8 idx, i, read_reg, doneflag = 0;
 	enum rdwr_status stat;
 	u32 memory_size;
 	int ret;
@@ -2485,11 +2486,21 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 				mwifiex_read_reg_byte(adapter, reg, dbg_ptr);
 				if (dbg_ptr < end_ptr) {
 					dbg_ptr++;
-				} else {
-					mwifiex_dbg(adapter, ERROR,
-						    "Allocated buf not enough\n");
-					return;
+					continue;
 				}
+				mwifiex_dbg(adapter, ERROR,
+					    "pre-allocated buf not enough\n");
+				tmp_ptr =
+					vzalloc(memory_size + MWIFIEX_SIZE_4K);
+				if (!tmp_ptr)
+					return;
+				memcpy(tmp_ptr, entry->mem_ptr, memory_size);
+				vfree(entry->mem_ptr);
+				entry->mem_ptr = tmp_ptr;
+				tmp_ptr = NULL;
+				dbg_ptr = entry->mem_ptr + memory_size;
+				memory_size += MWIFIEX_SIZE_4K;
+				end_ptr = entry->mem_ptr + memory_size;
 			}
 
 			if (stat != RDWR_STATUS_DONE)
-- 
cgit v0.10.2


From 03f0f7cdebeb0872076817ff835ea812a62b37c6 Mon Sep 17 00:00:00 2001
From: Amitkumar Karwar <akarwar@marvell.com>
Date: Tue, 2 Feb 2016 22:05:08 -0800
Subject: mwifiex: increase the priority of firmware dump message

Firmware dump operation takes few seconds. Hence it's
important to notify user in dmesg that firmware dump
has started or completed.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index a62a09b..14fe316 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2422,7 +2422,7 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 		entry->mem_size = 0;
 	}
 
-	mwifiex_dbg(adapter, DUMP, "== mwifiex firmware dump start ==\n");
+	mwifiex_dbg(adapter, MSG, "== mwifiex firmware dump start ==\n");
 
 	/* Read the number of the memories which will dump */
 	stat = mwifiex_pcie_rdwr_firmware(adapter, doneflag);
@@ -2512,7 +2512,7 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 			break;
 		} while (true);
 	}
-	mwifiex_dbg(adapter, DUMP, "== mwifiex firmware dump end ==\n");
+	mwifiex_dbg(adapter, MSG, "== mwifiex firmware dump end ==\n");
 }
 
 static void mwifiex_pcie_device_dump_work(struct mwifiex_adapter *adapter)
-- 
cgit v0.10.2


From 11e70824e75f2cfbad9ae066ca5b29e1c361f19e Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 2 Feb 2016 22:05:09 -0800
Subject: mwifiex: firmware dump support for w8997 chipset

Current firmware dump solution support w8897 chipset, this
patch extend the exist framework with support for w8997
chipset.

Trigger firmware dump using,
cat /sys/kernel/debug/mwifiex/mlan0/device_dump,
data can be obtain by
cat /sys/class/devcoredump/devcd*/data > data.txt after that.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/decl.h b/drivers/net/wireless/marvell/mwifiex/decl.h
index a184f9f..bec300b 100644
--- a/drivers/net/wireless/marvell/mwifiex/decl.h
+++ b/drivers/net/wireless/marvell/mwifiex/decl.h
@@ -274,6 +274,7 @@ struct mwifiex_11h_intf_state {
 } __packed;
 
 #define MWIFIEX_FW_DUMP_IDX		0xff
+#define MWIFIEX_FW_DUMP_MAX_MEMSIZE     0x160000
 #define MWIFIEX_DRV_INFO_IDX		20
 #define FW_DUMP_MAX_NAME_LEN		8
 #define FW_DUMP_HOST_READY      0xEE
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index 14fe316..cc07214 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2402,7 +2402,7 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 	struct pcie_service_card *card = adapter->card;
 	const struct mwifiex_pcie_card_reg *creg = card->pcie.reg;
 	unsigned int reg, reg_start, reg_end;
-	u8 *dbg_ptr, *end_ptr, *tmp_ptr, dump_num;
+	u8 *dbg_ptr, *end_ptr, *tmp_ptr, fw_dump_num, dump_num;
 	u8 idx, i, read_reg, doneflag = 0;
 	enum rdwr_status stat;
 	u32 memory_size;
@@ -2430,22 +2430,32 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 		return;
 
 	reg = creg->fw_dump_start;
-	mwifiex_read_reg_byte(adapter, reg, &dump_num);
+	mwifiex_read_reg_byte(adapter, reg, &fw_dump_num);
+
+	/* W8997 chipset firmware dump will be restore in single region*/
+	if (fw_dump_num == 0)
+		dump_num = 1;
+	else
+		dump_num = fw_dump_num;
 
 	/* Read the length of every memory which will dump */
 	for (idx = 0; idx < dump_num; idx++) {
 		struct memory_type_mapping *entry =
 				&adapter->mem_type_mapping_tbl[idx];
-		stat = mwifiex_pcie_rdwr_firmware(adapter, doneflag);
-		if (stat == RDWR_STATUS_FAILURE)
-			return;
-
 		memory_size = 0;
-		reg = creg->fw_dump_start;
-		for (i = 0; i < 4; i++) {
-			mwifiex_read_reg_byte(adapter, reg, &read_reg);
-			memory_size |= (read_reg << (i * 8));
+		if (fw_dump_num != 0) {
+			stat = mwifiex_pcie_rdwr_firmware(adapter, doneflag);
+			if (stat == RDWR_STATUS_FAILURE)
+				return;
+
+			reg = creg->fw_dump_start;
+			for (i = 0; i < 4; i++) {
+				mwifiex_read_reg_byte(adapter, reg, &read_reg);
+				memory_size |= (read_reg << (i * 8));
 			reg++;
+			}
+		} else {
+			memory_size = MWIFIEX_FW_DUMP_MAX_MEMSIZE;
 		}
 
 		if (memory_size == 0) {
diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.h b/drivers/net/wireless/marvell/mwifiex/pcie.h
index 8bc4390..29e58ce 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.h
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.h
@@ -239,6 +239,11 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8997 = {
 	.ring_tx_start_ptr = MWIFIEX_BD_FLAG_TX_START_PTR,
 	.pfu_enabled = 1,
 	.sleep_cookie = 0,
+	.fw_dump_ctrl = 0xcf4,
+	.fw_dump_start = 0xcf8,
+	.fw_dump_end = 0xcff,
+	.fw_dump_host_ready = 0xcc,
+	.fw_dump_read_done = 0xdd,
 	.msix_support = 1,
 };
 
@@ -253,6 +258,10 @@ static struct memory_type_mapping mem_type_mapping_tbl_w8897[] = {
 	{"MAC", NULL, 0, 0xF7},
 };
 
+static struct memory_type_mapping mem_type_mapping_tbl_w8997[] = {
+	{"DUMP", NULL, 0, 0xDD},
+};
+
 struct mwifiex_pcie_device {
 	const char *firmware;
 	const struct mwifiex_pcie_card_reg *reg;
@@ -289,7 +298,9 @@ static const struct mwifiex_pcie_device mwifiex_pcie8997 = {
 	.reg            = &mwifiex_reg_8997,
 	.blksz_fw_dl = MWIFIEX_PCIE_BLOCK_SIZE_FW_DNLD,
 	.tx_buf_size = MWIFIEX_TX_DATA_BUF_SIZE_4K,
-	.can_dump_fw = false,
+	.can_dump_fw = true,
+	.mem_type_mapping_tbl = mem_type_mapping_tbl_w8997,
+	.num_mem_types = ARRAY_SIZE(mem_type_mapping_tbl_w8997),
 	.can_ext_scan = true,
 };
 
-- 
cgit v0.10.2


From 0e715d6fbd2a4a1dcd215d6d51091346e6a3d3fa Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 2 Feb 2016 18:09:11 +0100
Subject: vxlan: cleanup types

include/net/vxlan.h is a kernel header, no need to prefix fixed size types
with double underscore.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 0fb8644..5c6425061 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -30,15 +30,15 @@
  * [0] https://tools.ietf.org/html/draft-smith-vxlan-group-policy
  */
 struct vxlanhdr_gbp {
-	__u8	vx_flags;
+	u8	vx_flags;
 #ifdef __LITTLE_ENDIAN_BITFIELD
-	__u8	reserved_flags1:3,
+	u8	reserved_flags1:3,
 		policy_applied:1,
 		reserved_flags2:2,
 		dont_learn:1,
 		reserved_flags3:1;
 #elif defined(__BIG_ENDIAN_BITFIELD)
-	__u8	reserved_flags1:1,
+	u8	reserved_flags1:1,
 		dont_learn:1,
 		reserved_flags2:2,
 		policy_applied:1,
@@ -138,10 +138,10 @@ struct vxlan_config {
 	int			remote_ifindex;
 	int			mtu;
 	__be16			dst_port;
-	__u16			port_min;
-	__u16			port_max;
-	__u8			tos;
-	__u8			ttl;
+	u16			port_min;
+	u16			port_max;
+	u8			tos;
+	u8			ttl;
 	u32			flags;
 	unsigned long		age_interval;
 	unsigned int		addrmax;
-- 
cgit v0.10.2


From 427bc465bf9fcdab749f6997ff7a4eecaef4ca40 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 2 Feb 2016 18:09:12 +0100
Subject: vxlan: remove duplicated macros

VNI_HASH_BITS and VNI_HASH_SIZE are defined twice. Remove the extra
definitions.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 5c6425061..234bf1e 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -9,9 +9,6 @@
 #include <linux/udp.h>
 #include <net/dst_metadata.h>
 
-#define VNI_HASH_BITS	10
-#define VNI_HASH_SIZE	(1<<VNI_HASH_BITS)
-
 /*
  * VXLAN Group Based Policy Extension:
  * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
-- 
cgit v0.10.2


From 828788ac99d5de6bae10b333d1e8ddf25928ac12 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 2 Feb 2016 18:09:13 +0100
Subject: vxlan: restructure vxlan.h definitions

RCO and GBP are VXLAN extensions, not specified in RFC 7348. Because of
that, they need to be explicitly enabled when creating vxlan interface. By
default, those extensions are not used and plain VXLAN header is sent and
received.

Reflect this in vxlan.h: first, the plain VXLAN header is defined. Following
it, RCO is documented and defined, and likewise for GBP.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 234bf1e..25bd919 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -9,14 +9,71 @@
 #include <linux/udp.h>
 #include <net/dst_metadata.h>
 
+/* VXLAN protocol (RFC 7348) header:
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |R|R|R|R|I|R|R|R|               Reserved                        |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |                VXLAN Network Identifier (VNI) |   Reserved    |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ *
+ * I = VXLAN Network Identifier (VNI) present.
+ */
+struct vxlanhdr {
+	__be32 vx_flags;
+	__be32 vx_vni;
+};
+
+/* VXLAN header flags. */
+#define VXLAN_HF_VNI BIT(27)
+
+#define VXLAN_N_VID     (1u << 24)
+#define VXLAN_VID_MASK  (VXLAN_N_VID - 1)
+#define VXLAN_VNI_MASK  (VXLAN_VID_MASK << 8)
+#define VXLAN_HLEN (sizeof(struct udphdr) + sizeof(struct vxlanhdr))
+
+#define VNI_HASH_BITS	10
+#define VNI_HASH_SIZE	(1<<VNI_HASH_BITS)
+#define FDB_HASH_BITS	8
+#define FDB_HASH_SIZE	(1<<FDB_HASH_BITS)
+
+/* Remote checksum offload for VXLAN (VXLAN_F_REMCSUM_[RT]X):
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |R|R|R|R|I|R|R|R|R|R|C|              Reserved                   |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |           VXLAN Network Identifier (VNI)      |O| Csum start  |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ *
+ * C = Remote checksum offload bit. When set indicates that the
+ *     remote checksum offload data is present.
+ *
+ * O = Offset bit. Indicates the checksum offset relative to
+ *     checksum start.
+ *
+ * Csum start = Checksum start divided by two.
+ *
+ * http://tools.ietf.org/html/draft-herbert-vxlan-rco
+ */
+
+/* VXLAN-RCO header flags. */
+#define VXLAN_HF_RCO BIT(21)
+
+/* Remote checksum offload header option */
+#define VXLAN_RCO_MASK  0x7f    /* Last byte of vni field */
+#define VXLAN_RCO_UDP   0x80    /* Indicate UDP RCO (TCP when not set *) */
+#define VXLAN_RCO_SHIFT 1       /* Left shift of start */
+#define VXLAN_RCO_SHIFT_MASK ((1 << VXLAN_RCO_SHIFT) - 1)
+#define VXLAN_MAX_REMCSUM_START (VXLAN_RCO_MASK << VXLAN_RCO_SHIFT)
+
 /*
- * VXLAN Group Based Policy Extension:
+ * VXLAN Group Based Policy Extension (VXLAN_F_GBP):
  * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * |1|-|-|-|1|-|-|-|R|D|R|R|A|R|R|R|        Group Policy ID        |
+ * |G|R|R|R|I|R|R|R|R|D|R|R|A|R|R|R|        Group Policy ID        |
  * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  * |                VXLAN Network Identifier (VNI) |   Reserved    |
  * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  *
+ * G = Group Policy ID present.
+ *
  * D = Don't Learn bit. When set, this bit indicates that the egress
  *     VTEP MUST NOT learn the source address of the encapsulated frame.
  *
@@ -24,7 +81,7 @@
  *     this packet. Policies MUST NOT be applied by devices when the
  *     A bit is set.
  *
- * [0] https://tools.ietf.org/html/draft-smith-vxlan-group-policy
+ * https://tools.ietf.org/html/draft-smith-vxlan-group-policy
  */
 struct vxlanhdr_gbp {
 	u8	vx_flags;
@@ -47,6 +104,9 @@ struct vxlanhdr_gbp {
 	__be32	vx_vni;
 };
 
+/* VXLAN-GBP header flags. */
+#define VXLAN_HF_GBP BIT(31)
+
 #define VXLAN_GBP_USED_BITS (VXLAN_HF_GBP | 0xFFFFFF)
 
 /* skb->mark mapping
@@ -59,44 +119,6 @@ struct vxlanhdr_gbp {
 #define VXLAN_GBP_POLICY_APPLIED	(BIT(3) << 16)
 #define VXLAN_GBP_ID_MASK		(0xFFFF)
 
-/* VXLAN protocol header:
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * |G|R|R|R|I|R|R|C|               Reserved                        |
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- * |                VXLAN Network Identifier (VNI) |   Reserved    |
- * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- *
- * G = 1	Group Policy (VXLAN-GBP)
- * I = 1	VXLAN Network Identifier (VNI) present
- * C = 1	Remote checksum offload (RCO)
- */
-struct vxlanhdr {
-	__be32 vx_flags;
-	__be32 vx_vni;
-};
-
-/* VXLAN header flags. */
-#define VXLAN_HF_RCO BIT(21)
-#define VXLAN_HF_VNI BIT(27)
-#define VXLAN_HF_GBP BIT(31)
-
-/* Remote checksum offload header option */
-#define VXLAN_RCO_MASK  0x7f    /* Last byte of vni field */
-#define VXLAN_RCO_UDP   0x80    /* Indicate UDP RCO (TCP when not set *) */
-#define VXLAN_RCO_SHIFT 1       /* Left shift of start */
-#define VXLAN_RCO_SHIFT_MASK ((1 << VXLAN_RCO_SHIFT) - 1)
-#define VXLAN_MAX_REMCSUM_START (VXLAN_RCO_MASK << VXLAN_RCO_SHIFT)
-
-#define VXLAN_N_VID     (1u << 24)
-#define VXLAN_VID_MASK  (VXLAN_N_VID - 1)
-#define VXLAN_VNI_MASK  (VXLAN_VID_MASK << 8)
-#define VXLAN_HLEN (sizeof(struct udphdr) + sizeof(struct vxlanhdr))
-
-#define VNI_HASH_BITS	10
-#define VNI_HASH_SIZE	(1<<VNI_HASH_BITS)
-#define FDB_HASH_BITS	8
-#define FDB_HASH_SIZE	(1<<FDB_HASH_BITS)
-
 struct vxlan_metadata {
 	u32		gbp;
 };
-- 
cgit v0.10.2


From 1a8496ba409132afb3b407599061c34847de42d3 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 2 Feb 2016 18:09:14 +0100
Subject: vxlan: consolidate output route calculation

The code for output route lookup is duplicated for ndo_start_xmit and
ndo_fill_metadata_dst. Move it to a common function.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 6543918..d0f7723 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1847,6 +1847,27 @@ static int vxlan_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *sk
 	return 0;
 }
 
+static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
+				      struct sk_buff *skb, int oif, u8 tos,
+				      __be32 daddr, __be32 *saddr)
+{
+	struct rtable *rt = NULL;
+	struct flowi4 fl4;
+
+	memset(&fl4, 0, sizeof(fl4));
+	fl4.flowi4_oif = oif;
+	fl4.flowi4_tos = RT_TOS(tos);
+	fl4.flowi4_mark = skb->mark;
+	fl4.flowi4_proto = IPPROTO_UDP;
+	fl4.daddr = daddr;
+	fl4.saddr = vxlan->cfg.saddr.sin.sin_addr.s_addr;
+
+	rt = ip_route_output_key(vxlan->net, &fl4);
+	if (!IS_ERR(rt))
+		*saddr = fl4.saddr;
+	return rt;
+}
+
 #if IS_ENABLED(CONFIG_IPV6)
 static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 					  struct sk_buff *skb, int oif,
@@ -1928,7 +1949,6 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	struct sock *sk;
 	struct rtable *rt = NULL;
 	const struct iphdr *old_iph;
-	struct flowi4 fl4;
 	union vxlan_addr *dst;
 	union vxlan_addr remote_ip;
 	struct vxlan_metadata _md;
@@ -1995,6 +2015,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	}
 
 	if (dst->sa.sa_family == AF_INET) {
+		__be32 saddr;
+
 		if (!vxlan->vn4_sock)
 			goto drop;
 		sk = vxlan->vn4_sock->sock->sk;
@@ -2009,15 +2031,9 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 				flags &= ~VXLAN_F_UDP_CSUM;
 		}
 
-		memset(&fl4, 0, sizeof(fl4));
-		fl4.flowi4_oif = rdst ? rdst->remote_ifindex : 0;
-		fl4.flowi4_tos = RT_TOS(tos);
-		fl4.flowi4_mark = skb->mark;
-		fl4.flowi4_proto = IPPROTO_UDP;
-		fl4.daddr = dst->sin.sin_addr.s_addr;
-		fl4.saddr = vxlan->cfg.saddr.sin.sin_addr.s_addr;
-
-		rt = ip_route_output_key(vxlan->net, &fl4);
+		rt = vxlan_get_route(vxlan, skb,
+				     rdst ? rdst->remote_ifindex : 0, tos,
+				     dst->sin.sin_addr.s_addr, &saddr);
 		if (IS_ERR(rt)) {
 			netdev_dbg(dev, "no route to %pI4\n",
 				   &dst->sin.sin_addr.s_addr);
@@ -2049,7 +2065,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 
 		tos = ip_tunnel_ecn_encap(tos, old_iph, skb);
 		ttl = ttl ? : ip4_dst_hoplimit(&rt->dst);
-		err = vxlan_xmit_skb(rt, sk, skb, fl4.saddr,
+		err = vxlan_xmit_skb(rt, sk, skb, saddr,
 				     dst->sin.sin_addr.s_addr, tos, ttl, df,
 				     src_port, dst_port, htonl(vni << 8), md,
 				     !net_eq(vxlan->net, dev_net(vxlan->dev)),
@@ -2390,31 +2406,6 @@ static int vxlan_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
-static int egress_ipv4_tun_info(struct net_device *dev, struct sk_buff *skb,
-				struct ip_tunnel_info *info,
-				__be16 sport, __be16 dport)
-{
-	struct vxlan_dev *vxlan = netdev_priv(dev);
-	struct rtable *rt;
-	struct flowi4 fl4;
-
-	memset(&fl4, 0, sizeof(fl4));
-	fl4.flowi4_tos = RT_TOS(info->key.tos);
-	fl4.flowi4_mark = skb->mark;
-	fl4.flowi4_proto = IPPROTO_UDP;
-	fl4.daddr = info->key.u.ipv4.dst;
-
-	rt = ip_route_output_key(vxlan->net, &fl4);
-	if (IS_ERR(rt))
-		return PTR_ERR(rt);
-	ip_rt_put(rt);
-
-	info->key.u.ipv4.src = fl4.saddr;
-	info->key.tp_src = sport;
-	info->key.tp_dst = dport;
-	return 0;
-}
-
 static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 {
 	struct vxlan_dev *vxlan = netdev_priv(dev);
@@ -2426,9 +2417,16 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 	dport = info->key.tp_dst ? : vxlan->cfg.dst_port;
 
 	if (ip_tunnel_info_af(info) == AF_INET) {
+		struct rtable *rt;
+
 		if (!vxlan->vn4_sock)
 			return -EINVAL;
-		return egress_ipv4_tun_info(dev, skb, info, sport, dport);
+		rt = vxlan_get_route(vxlan, skb, 0, info->key.tos,
+				     info->key.u.ipv4.dst,
+				     &info->key.u.ipv4.src);
+		if (IS_ERR(rt))
+			return PTR_ERR(rt);
+		ip_rt_put(rt);
 	} else {
 #if IS_ENABLED(CONFIG_IPV6)
 		struct dst_entry *ndst;
@@ -2441,13 +2439,12 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 		if (IS_ERR(ndst))
 			return PTR_ERR(ndst);
 		dst_release(ndst);
-
-		info->key.tp_src = sport;
-		info->key.tp_dst = dport;
 #else /* !CONFIG_IPV6 */
 		return -EPFNOSUPPORT;
 #endif
 	}
+	info->key.tp_src = sport;
+	info->key.tp_dst = dport;
 	return 0;
 }
 
-- 
cgit v0.10.2


From b4ed5cad24c1072033efbffa680c84c9ba19c798 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 2 Feb 2016 18:09:15 +0100
Subject: vxlan: consolidate csum flag handling

The flag for tx checksumming for tunneling over IPv4 and IPv6 is different.
Decide whether to do tx checksumming in vxlan_xmit_one and pass it on as
a separate flag. This will allow for tx path consolidation in the next
patch.

Unfortunately, gcc is not clever enough to see that udp_sum is always
initialized and gives an uninitialized variable warning. Set it to false to
silence the warning.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index d0f7723..fe3fd480 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1690,12 +1690,13 @@ static int vxlan6_xmit_skb(struct dst_entry *dst, struct sock *sk,
 			   struct net_device *dev, struct in6_addr *saddr,
 			   struct in6_addr *daddr, __u8 prio, __u8 ttl,
 			   __be16 src_port, __be16 dst_port, __be32 vni,
-			   struct vxlan_metadata *md, bool xnet, u32 vxflags)
+			   struct vxlan_metadata *md, bool xnet, u32 vxflags,
+			   bool udp_sum)
 {
 	struct vxlanhdr *vxh;
 	int min_headroom;
 	int err;
-	bool udp_sum = !(vxflags & VXLAN_F_UDP_ZERO_CSUM6_TX);
+	bool nocheck = !udp_sum;
 	int type = udp_sum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
 	u16 hdrlen = sizeof(struct vxlanhdr);
 
@@ -1763,8 +1764,7 @@ static int vxlan6_xmit_skb(struct dst_entry *dst, struct sock *sk,
 	skb_set_inner_protocol(skb, htons(ETH_P_TEB));
 
 	udp_tunnel6_xmit_skb(dst, sk, skb, dev, saddr, daddr, prio,
-			     ttl, src_port, dst_port,
-			     !!(vxflags & VXLAN_F_UDP_ZERO_CSUM6_TX));
+			     ttl, src_port, dst_port, nocheck);
 	return 0;
 err:
 	dst_release(dst);
@@ -1775,12 +1775,13 @@ err:
 static int vxlan_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb,
 			  __be32 src, __be32 dst, __u8 tos, __u8 ttl, __be16 df,
 			  __be16 src_port, __be16 dst_port, __be32 vni,
-			  struct vxlan_metadata *md, bool xnet, u32 vxflags)
+			  struct vxlan_metadata *md, bool xnet, u32 vxflags,
+			  bool udp_sum)
 {
 	struct vxlanhdr *vxh;
 	int min_headroom;
 	int err;
-	bool udp_sum = !!(vxflags & VXLAN_F_UDP_CSUM);
+	bool nocheck = !udp_sum;
 	int type = udp_sum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
 	u16 hdrlen = sizeof(struct vxlanhdr);
 
@@ -1842,8 +1843,7 @@ static int vxlan_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *sk
 	skb_set_inner_protocol(skb, htons(ETH_P_TEB));
 
 	udp_tunnel_xmit_skb(rt, sk, skb, src, dst, tos, ttl, df,
-			    src_port, dst_port, xnet,
-			    !(vxflags & VXLAN_F_UDP_CSUM));
+			    src_port, dst_port, xnet, nocheck);
 	return 0;
 }
 
@@ -1959,6 +1959,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	__u8 tos, ttl;
 	int err;
 	u32 flags = vxlan->flags;
+	bool udp_sum = false;
 
 	info = skb_tunnel_info(skb);
 
@@ -2007,6 +2008,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	if (info) {
 		ttl = info->key.ttl;
 		tos = info->key.tos;
+		udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM);
 
 		if (info->options_len)
 			md = ip_tunnel_info_opts(info);
@@ -2024,11 +2026,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		if (info) {
 			if (info->key.tun_flags & TUNNEL_DONT_FRAGMENT)
 				df = htons(IP_DF);
-
-			if (info->key.tun_flags & TUNNEL_CSUM)
-				flags |= VXLAN_F_UDP_CSUM;
-			else
-				flags &= ~VXLAN_F_UDP_CSUM;
+		} else {
+			udp_sum = !!(flags & VXLAN_F_UDP_CSUM);
 		}
 
 		rt = vxlan_get_route(vxlan, skb,
@@ -2069,7 +2068,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 				     dst->sin.sin_addr.s_addr, tos, ttl, df,
 				     src_port, dst_port, htonl(vni << 8), md,
 				     !net_eq(vxlan->net, dev_net(vxlan->dev)),
-				     flags);
+				     flags, udp_sum);
 		if (err < 0) {
 			/* skb is already freed. */
 			skb = NULL;
@@ -2119,18 +2118,14 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			return;
 		}
 
-		if (info) {
-			if (info->key.tun_flags & TUNNEL_CSUM)
-				flags &= ~VXLAN_F_UDP_ZERO_CSUM6_TX;
-			else
-				flags |= VXLAN_F_UDP_ZERO_CSUM6_TX;
-		}
+		if (!info)
+			udp_sum = !(flags & VXLAN_F_UDP_ZERO_CSUM6_TX);
 
 		ttl = ttl ? : ip6_dst_hoplimit(ndst);
 		err = vxlan6_xmit_skb(ndst, sk, skb, dev, &saddr, &dst->sin6.sin6_addr,
 				      0, ttl, src_port, dst_port, htonl(vni << 8), md,
 				      !net_eq(vxlan->net, dev_net(vxlan->dev)),
-				      flags);
+				      flags, udp_sum);
 #endif
 	}
 
-- 
cgit v0.10.2


From f491e56dba511d318f52efa4c226471bf5943e88 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 2 Feb 2016 18:09:16 +0100
Subject: vxlan: consolidate vxlan_xmit_skb and vxlan6_xmit_skb

There's a lot of code duplication. Factor out the duplicate code to a new
function shared between IPv4 and IPv6 xmit path.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index fe3fd480..65f5247 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1684,19 +1684,14 @@ static void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, u32 vxflags,
 	gbp->policy_id = htons(md->gbp & VXLAN_GBP_ID_MASK);
 }
 
-#if IS_ENABLED(CONFIG_IPV6)
-static int vxlan6_xmit_skb(struct dst_entry *dst, struct sock *sk,
-			   struct sk_buff *skb,
-			   struct net_device *dev, struct in6_addr *saddr,
-			   struct in6_addr *daddr, __u8 prio, __u8 ttl,
-			   __be16 src_port, __be16 dst_port, __be32 vni,
-			   struct vxlan_metadata *md, bool xnet, u32 vxflags,
+static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
+			   int iphdr_len, __be32 vni,
+			   struct vxlan_metadata *md, u32 vxflags,
 			   bool udp_sum)
 {
 	struct vxlanhdr *vxh;
 	int min_headroom;
 	int err;
-	bool nocheck = !udp_sum;
 	int type = udp_sum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
 	u16 hdrlen = sizeof(struct vxlanhdr);
 
@@ -1713,93 +1708,8 @@ static int vxlan6_xmit_skb(struct dst_entry *dst, struct sock *sk,
 		}
 	}
 
-	skb_scrub_packet(skb, xnet);
-
 	min_headroom = LL_RESERVED_SPACE(dst->dev) + dst->header_len
-			+ VXLAN_HLEN + sizeof(struct ipv6hdr)
-			+ (skb_vlan_tag_present(skb) ? VLAN_HLEN : 0);
-
-	/* Need space for new headers (invalidates iph ptr) */
-	err = skb_cow_head(skb, min_headroom);
-	if (unlikely(err)) {
-		kfree_skb(skb);
-		goto err;
-	}
-
-	skb = vlan_hwaccel_push_inside(skb);
-	if (WARN_ON(!skb)) {
-		err = -ENOMEM;
-		goto err;
-	}
-
-	skb = iptunnel_handle_offloads(skb, udp_sum, type);
-	if (IS_ERR(skb)) {
-		err = -EINVAL;
-		goto err;
-	}
-
-	vxh = (struct vxlanhdr *) __skb_push(skb, sizeof(*vxh));
-	vxh->vx_flags = htonl(VXLAN_HF_VNI);
-	vxh->vx_vni = vni;
-
-	if (type & SKB_GSO_TUNNEL_REMCSUM) {
-		u32 data = (skb_checksum_start_offset(skb) - hdrlen) >>
-			   VXLAN_RCO_SHIFT;
-
-		if (skb->csum_offset == offsetof(struct udphdr, check))
-			data |= VXLAN_RCO_UDP;
-
-		vxh->vx_vni |= htonl(data);
-		vxh->vx_flags |= htonl(VXLAN_HF_RCO);
-
-		if (!skb_is_gso(skb)) {
-			skb->ip_summed = CHECKSUM_NONE;
-			skb->encapsulation = 0;
-		}
-	}
-
-	if (vxflags & VXLAN_F_GBP)
-		vxlan_build_gbp_hdr(vxh, vxflags, md);
-
-	skb_set_inner_protocol(skb, htons(ETH_P_TEB));
-
-	udp_tunnel6_xmit_skb(dst, sk, skb, dev, saddr, daddr, prio,
-			     ttl, src_port, dst_port, nocheck);
-	return 0;
-err:
-	dst_release(dst);
-	return err;
-}
-#endif
-
-static int vxlan_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb,
-			  __be32 src, __be32 dst, __u8 tos, __u8 ttl, __be16 df,
-			  __be16 src_port, __be16 dst_port, __be32 vni,
-			  struct vxlan_metadata *md, bool xnet, u32 vxflags,
-			  bool udp_sum)
-{
-	struct vxlanhdr *vxh;
-	int min_headroom;
-	int err;
-	bool nocheck = !udp_sum;
-	int type = udp_sum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
-	u16 hdrlen = sizeof(struct vxlanhdr);
-
-	if ((vxflags & VXLAN_F_REMCSUM_TX) &&
-	    skb->ip_summed == CHECKSUM_PARTIAL) {
-		int csum_start = skb_checksum_start_offset(skb);
-
-		if (csum_start <= VXLAN_MAX_REMCSUM_START &&
-		    !(csum_start & VXLAN_RCO_SHIFT_MASK) &&
-		    (skb->csum_offset == offsetof(struct udphdr, check) ||
-		     skb->csum_offset == offsetof(struct tcphdr, check))) {
-			udp_sum = false;
-			type |= SKB_GSO_TUNNEL_REMCSUM;
-		}
-	}
-
-	min_headroom = LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len
-			+ VXLAN_HLEN + sizeof(struct iphdr)
+			+ VXLAN_HLEN + iphdr_len
 			+ (skb_vlan_tag_present(skb) ? VLAN_HLEN : 0);
 
 	/* Need space for new headers (invalidates iph ptr) */
@@ -1841,9 +1751,6 @@ static int vxlan_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *sk
 		vxlan_build_gbp_hdr(vxh, vxflags, md);
 
 	skb_set_inner_protocol(skb, htons(ETH_P_TEB));
-
-	udp_tunnel_xmit_skb(rt, sk, skb, src, dst, tos, ttl, df,
-			    src_port, dst_port, xnet, nocheck);
 	return 0;
 }
 
@@ -1960,6 +1867,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	int err;
 	u32 flags = vxlan->flags;
 	bool udp_sum = false;
+	bool xnet = !net_eq(vxlan->net, dev_net(vxlan->dev));
 
 	info = skb_tunnel_info(skb);
 
@@ -2064,16 +1972,14 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 
 		tos = ip_tunnel_ecn_encap(tos, old_iph, skb);
 		ttl = ttl ? : ip4_dst_hoplimit(&rt->dst);
-		err = vxlan_xmit_skb(rt, sk, skb, saddr,
-				     dst->sin.sin_addr.s_addr, tos, ttl, df,
-				     src_port, dst_port, htonl(vni << 8), md,
-				     !net_eq(vxlan->net, dev_net(vxlan->dev)),
-				     flags, udp_sum);
-		if (err < 0) {
-			/* skb is already freed. */
-			skb = NULL;
-			goto rt_tx_error;
-		}
+		err = vxlan_build_skb(skb, &rt->dst, sizeof(struct iphdr),
+				      htonl(vni << 8), md, flags, udp_sum);
+		if (err < 0)
+			goto xmit_tx_error;
+
+		udp_tunnel_xmit_skb(rt, sk, skb, saddr,
+				    dst->sin.sin_addr.s_addr, tos, ttl, df,
+				    src_port, dst_port, xnet, !udp_sum);
 #if IS_ENABLED(CONFIG_IPV6)
 	} else {
 		struct dst_entry *ndst;
@@ -2122,10 +2028,16 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			udp_sum = !(flags & VXLAN_F_UDP_ZERO_CSUM6_TX);
 
 		ttl = ttl ? : ip6_dst_hoplimit(ndst);
-		err = vxlan6_xmit_skb(ndst, sk, skb, dev, &saddr, &dst->sin6.sin6_addr,
-				      0, ttl, src_port, dst_port, htonl(vni << 8), md,
-				      !net_eq(vxlan->net, dev_net(vxlan->dev)),
-				      flags, udp_sum);
+		skb_scrub_packet(skb, xnet);
+		err = vxlan_build_skb(skb, ndst, sizeof(struct ipv6hdr),
+				      htonl(vni << 8), md, flags, udp_sum);
+		if (err < 0) {
+			dst_release(ndst);
+			return;
+		}
+		udp_tunnel6_xmit_skb(ndst, sk, skb, dev,
+				     &saddr, &dst->sin6.sin6_addr,
+				     0, ttl, src_port, dst_port, !udp_sum);
 #endif
 	}
 
@@ -2135,6 +2047,9 @@ drop:
 	dev->stats.tx_dropped++;
 	goto tx_free;
 
+xmit_tx_error:
+	/* skb is already freed. */
+	skb = NULL;
 rt_tx_error:
 	ip_rt_put(rt);
 tx_error:
-- 
cgit v0.10.2


From a5a773a54e59538d03fd1a20facd7214c030b1d3 Mon Sep 17 00:00:00 2001
From: Suresh Reddy <suresh.reddy@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:16 +0530
Subject: be2net: return error status from be_set_phys_id()

be_set_phys_id() returns 0 to ethtool when the command fails in the FW.

This patch fixes the set_phys_id() to return -EIO in case the FW cmd fails.

Signed-off-by: Suresh Reddy <suresh.reddy@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index a19ac44..2ff6916 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -720,29 +720,32 @@ static int be_set_phys_id(struct net_device *netdev,
 			  enum ethtool_phys_id_state state)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
+	int status = 0;
 
 	switch (state) {
 	case ETHTOOL_ID_ACTIVE:
-		be_cmd_get_beacon_state(adapter, adapter->hba_port_num,
-					&adapter->beacon_state);
-		return 1;	/* cycle on/off once per second */
+		status = be_cmd_get_beacon_state(adapter, adapter->hba_port_num,
+						 &adapter->beacon_state);
+		if (status)
+			return be_cmd_status(status);
+		return 1;       /* cycle on/off once per second */
 
 	case ETHTOOL_ID_ON:
-		be_cmd_set_beacon_state(adapter, adapter->hba_port_num, 0, 0,
-					BEACON_STATE_ENABLED);
+		status = be_cmd_set_beacon_state(adapter, adapter->hba_port_num,
+						 0, 0, BEACON_STATE_ENABLED);
 		break;
 
 	case ETHTOOL_ID_OFF:
-		be_cmd_set_beacon_state(adapter, adapter->hba_port_num, 0, 0,
-					BEACON_STATE_DISABLED);
+		status = be_cmd_set_beacon_state(adapter, adapter->hba_port_num,
+						 0, 0, BEACON_STATE_DISABLED);
 		break;
 
 	case ETHTOOL_ID_INACTIVE:
-		be_cmd_set_beacon_state(adapter, adapter->hba_port_num, 0, 0,
-					adapter->beacon_state);
+		status = be_cmd_set_beacon_state(adapter, adapter->hba_port_num,
+						 0, 0, adapter->beacon_state);
 	}
 
-	return 0;
+	return be_cmd_status(status);
 }
 
 static int be_set_dump(struct net_device *netdev, struct ethtool_dump *dump)
-- 
cgit v0.10.2


From fa5c867d4df275a2c11a455043848574e80a8b32 Mon Sep 17 00:00:00 2001
From: Suresh Reddy <suresh.reddy@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:17 +0530
Subject: be2net: check for INSUFFICIENT_PRIVILEGES error

The driver currently logs the message "VF is not privileged to issue
opcode" by checking only the base_status field for UNAUTHORIZED_REQUEST.
Add check to look for INSUFFICIENT_PRIVILEGES in the additional status
field also as not all cmds fail with that base status.

Signed-off-by: Suresh Reddy <suresh.reddy@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index b63d8ad..3b665f1 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -236,7 +236,8 @@ static int be_mcc_compl_process(struct be_adapter *adapter,
 
 	if (base_status != MCC_STATUS_SUCCESS &&
 	    !be_skip_err_log(opcode, base_status, addl_status)) {
-		if (base_status == MCC_STATUS_UNAUTHORIZED_REQUEST) {
+		if (base_status == MCC_STATUS_UNAUTHORIZED_REQUEST ||
+		    addl_status == MCC_ADDL_STATUS_INSUFFICIENT_PRIVILEGES) {
 			dev_warn(&adapter->pdev->dev,
 				 "VF is not privileged to issue opcode %d-%d\n",
 				 opcode, subsystem);
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index 241819b..f260ef3 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -68,7 +68,8 @@ enum mcc_addl_status {
 	MCC_ADDL_STATUS_TOO_MANY_INTERFACES = 0x4a,
 	MCC_ADDL_STATUS_INSUFFICIENT_VLANS = 0xab,
 	MCC_ADDL_STATUS_INVALID_SIGNATURE = 0x56,
-	MCC_ADDL_STATUS_MISSING_SIGNATURE = 0x57
+	MCC_ADDL_STATUS_MISSING_SIGNATURE = 0x57,
+	MCC_ADDL_STATUS_INSUFFICIENT_PRIVILEGES = 0x60
 };
 
 #define CQE_BASE_STATUS_MASK		0xFFFF
-- 
cgit v0.10.2


From 41dcdfbd0944a3ef08224a8c9ba3b2fdeae0dd86 Mon Sep 17 00:00:00 2001
From: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:18 +0530
Subject: be2net: Fix be_vlan_rem_vid() to check vlan id being removed

The driver decrements its vlan count without checking if it is really
present in its list. This results in an invalid vlan count and impacts
subsequent vlan add/rem ops. The function be_vlan_rem_vid() should be
updated to fix this.

Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index f99de36..09e6f2c 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1463,6 +1463,9 @@ static int be_vlan_rem_vid(struct net_device *netdev, __be16 proto, u16 vid)
 	if (lancer_chip(adapter) && vid == 0)
 		return 0;
 
+	if (!test_bit(vid, adapter->vids))
+		return 0;
+
 	clear_bit(vid, adapter->vids);
 	adapter->vlans_added--;
 
-- 
cgit v0.10.2


From ee9ad2802449c0a6811a74af2ce30500895d137b Mon Sep 17 00:00:00 2001
From: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:19 +0530
Subject: be2net: SRIOV Queue distribution should factor in EQ-count of VFs

The SRIOV resource distribution logic for RX/TX queue counts is not optimal
when a small number of VFs are enabled. It does not take into account the
VF's EQ count while computing the queue counts. Because of this, the VF
gets a large number of queues, though it doesn't have sufficient EQs,
resulting in wasted queue resources. And the PF gets a smaller share of
queues though it has more EQs. Fix this by capping the VF queue count at
its EQ count.

Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index cf83783..de88c30 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -89,6 +89,10 @@
 #define BE3_MAX_TX_QS		16
 #define BE3_MAX_EVT_QS		16
 #define BE3_SRIOV_MAX_EVT_QS	8
+#define SH_VF_MAX_NIC_EQS	3	/* Skyhawk VFs can have a max of 4 EQs
+					 * and at least 1 is granted to either
+					 * SURF/DPDK
+					 */
 
 #define MAX_RSS_IFACES		15
 #define MAX_RX_QS		32
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 09e6f2c..62f6fbb 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -3792,18 +3792,15 @@ static u16 be_calculate_vf_qs(struct be_adapter *adapter, u16 num_vfs)
 	struct be_resources res = adapter->pool_res;
 	u16 num_vf_qs = 1;
 
-	/* Distribute the queue resources equally among the PF and it's VFs
+	/* Distribute the queue resources among the PF and it's VFs
 	 * Do not distribute queue resources in multi-channel configuration.
 	 */
 	if (num_vfs && !be_is_mc(adapter)) {
-		/* If number of VFs requested is 8 less than max supported,
-		 * assign 8 queue pairs to the PF and divide the remaining
-		 * resources evenly among the VFs
-		 */
-		if (num_vfs < (be_max_vfs(adapter) - 8))
-			num_vf_qs = (res.max_rss_qs - 8) / num_vfs;
-		else
-			num_vf_qs = res.max_rss_qs / num_vfs;
+		 /* Divide the qpairs evenly among the VFs and the PF, capped
+		  * at VF-EQ-count. Any remainder qpairs belong to the PF.
+		  */
+		num_vf_qs = min(SH_VF_MAX_NIC_EQS,
+				res.max_rss_qs / (num_vfs + 1));
 
 		/* Skyhawk-R chip supports only MAX_RSS_IFACES RSS capable
 		 * interfaces per port. Provide RSS on VFs, only if number
-- 
cgit v0.10.2


From 2e365b1b80aa98655a5582dbb9bf6cf8c0ff268c Mon Sep 17 00:00:00 2001
From: Somnath Kotur <somnath.kotur@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:20 +0530
Subject: be2net: Don't run ethtool self-tests for VFs

The CMD_SUBSYSTEM_LOWLEVEL cmds need DEV_CFG Privilege to run
which VFs don't have by default.
Self-tests need to be issued only for PFs.

Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 3b665f1..7d51d47 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -65,7 +65,22 @@ static struct be_cmd_priv_map cmd_priv_map[] = {
 		CMD_SUBSYSTEM_COMMON,
 		BE_PRIV_LNKMGMT | BE_PRIV_VHADM |
 		BE_PRIV_DEVCFG | BE_PRIV_DEVSEC
-	}
+	},
+	{
+		OPCODE_LOWLEVEL_HOST_DDR_DMA,
+		CMD_SUBSYSTEM_LOWLEVEL,
+		BE_PRIV_DEVCFG | BE_PRIV_DEVSEC
+	},
+	{
+		OPCODE_LOWLEVEL_LOOPBACK_TEST,
+		CMD_SUBSYSTEM_LOWLEVEL,
+		BE_PRIV_DEVCFG | BE_PRIV_DEVSEC
+	},
+	{
+		OPCODE_LOWLEVEL_SET_LOOPBACK_MODE,
+		CMD_SUBSYSTEM_LOWLEVEL,
+		BE_PRIV_DEVCFG | BE_PRIV_DEVSEC
+	},
 };
 
 static bool be_cmd_allowed(struct be_adapter *adapter, u8 opcode, u8 subsystem)
@@ -3169,6 +3184,10 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num,
 	struct be_cmd_req_set_lmode *req;
 	int status;
 
+	if (!be_cmd_allowed(adapter, OPCODE_LOWLEVEL_SET_LOOPBACK_MODE,
+			    CMD_SUBSYSTEM_LOWLEVEL))
+		return -EPERM;
+
 	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
@@ -3214,6 +3233,10 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num,
 	struct be_cmd_resp_loopback_test *resp;
 	int status;
 
+	if (!be_cmd_allowed(adapter, OPCODE_LOWLEVEL_LOOPBACK_TEST,
+			    CMD_SUBSYSTEM_LOWLEVEL))
+		return -EPERM;
+
 	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
@@ -3260,6 +3283,10 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern,
 	int status;
 	int i, j = 0;
 
+	if (!be_cmd_allowed(adapter, OPCODE_LOWLEVEL_HOST_DDR_DMA,
+			    CMD_SUBSYSTEM_LOWLEVEL))
+		return -EPERM;
+
 	spin_lock_bh(&adapter->mcc_lock);
 
 	wrb = wrb_from_mccq(adapter);
-- 
cgit v0.10.2


From 1babbad46f4aba1c242e43e27c5b36a62311b2d4 Mon Sep 17 00:00:00 2001
From: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:21 +0530
Subject: be2net: Fix Lancer error recovery

After error is detected, wait for adapter to move to ready state
before destroying queues and cleanup of other resources. Also
skip performing any cleanup for non-Lancer chips and move debug
messages to correct routine.

Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 62f6fbb..6eb3aba 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -4859,21 +4859,27 @@ static int be_resume(struct be_adapter *adapter)
 
 static int be_err_recover(struct be_adapter *adapter)
 {
-	struct device *dev = &adapter->pdev->dev;
 	int status;
 
+	/* Error recovery is supported only Lancer as of now */
+	if (!lancer_chip(adapter))
+		return -EIO;
+
+	/* Wait for adapter to reach quiescent state before
+	 * destroying queues
+	 */
+	status = be_fw_wait_ready(adapter);
+	if (status)
+		goto err;
+
+	be_cleanup(adapter);
+
 	status = be_resume(adapter);
 	if (status)
 		goto err;
 
-	dev_info(dev, "Adapter recovery successful\n");
 	return 0;
 err:
-	if (be_physfn(adapter))
-		dev_err(dev, "Adapter recovery failed\n");
-	else
-		dev_err(dev, "Re-trying adapter recovery\n");
-
 	return status;
 }
 
@@ -4882,21 +4888,32 @@ static void be_err_detection_task(struct work_struct *work)
 	struct be_adapter *adapter =
 				container_of(work, struct be_adapter,
 					     be_err_detection_work.work);
-	int status = 0;
+	struct device *dev = &adapter->pdev->dev;
+	int recovery_status;
 
 	be_detect_error(adapter);
 
-	if (be_check_error(adapter, BE_ERROR_HW)) {
-		be_cleanup(adapter);
-
-		/* As of now error recovery support is in Lancer only */
-		if (lancer_chip(adapter))
-			status = be_err_recover(adapter);
+	if (be_check_error(adapter, BE_ERROR_HW))
+		recovery_status = be_err_recover(adapter);
+	else
+		goto reschedule_task;
+
+	if (!recovery_status) {
+		dev_info(dev, "Adapter recovery successful\n");
+		goto reschedule_task;
+	} else if (be_virtfn(adapter)) {
+		/* For VFs, check if PF have allocated resources
+		 * every second.
+		 */
+		dev_err(dev, "Re-trying adapter recovery\n");
+		goto reschedule_task;
+	} else {
+		dev_err(dev, "Adapter recovery failed\n");
 	}
 
-	/* Always attempt recovery on VFs */
-	if (!status || be_virtfn(adapter))
-		be_schedule_err_detection(adapter);
+	return;
+reschedule_task:
+	be_schedule_err_detection(adapter);
 }
 
 static void be_log_sfp_info(struct be_adapter *adapter)
-- 
cgit v0.10.2


From 972f37b424e65f9ef1ce143b8658d9ed50db9f42 Mon Sep 17 00:00:00 2001
From: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:22 +0530
Subject: be2net: Add retry in case of error recovery failure

Retry error recovery MAX_ERR_RECOVERY_RETRY_COUNT times in case of
failure during error recovery.

Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index de88c30..515e206 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -397,6 +397,10 @@ enum vf_state {
 #define BE_UC_PMAC_COUNT			30
 #define BE_VF_UC_PMAC_COUNT			2
 
+#define MAX_ERR_RECOVERY_RETRY_COUNT		3
+#define ERR_DETECTION_DELAY			1000
+#define ERR_RECOVERY_RETRY_DELAY		30000
+
 /* Ethtool set_dump flags */
 #define LANCER_INITIATE_FW_DUMP			0x1
 #define LANCER_DELETE_FW_DUMP			0x2
@@ -534,6 +538,7 @@ struct be_adapter {
 	u16 work_counter;
 
 	struct delayed_work be_err_detection_work;
+	u8 recovery_retries;
 	u8 err_flags;
 	u32 flags;
 	u32 cmd_privileges;
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 6eb3aba..d5286d3 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -4265,10 +4265,10 @@ static void be_schedule_worker(struct be_adapter *adapter)
 	adapter->flags |= BE_FLAGS_WORKER_SCHEDULED;
 }
 
-static void be_schedule_err_detection(struct be_adapter *adapter)
+static void be_schedule_err_detection(struct be_adapter *adapter, u32 delay)
 {
 	schedule_delayed_work(&adapter->be_err_detection_work,
-			      msecs_to_jiffies(1000));
+			      msecs_to_jiffies(delay));
 	adapter->flags |= BE_FLAGS_ERR_DETECTION_SCHEDULED;
 }
 
@@ -4890,6 +4890,7 @@ static void be_err_detection_task(struct work_struct *work)
 					     be_err_detection_work.work);
 	struct device *dev = &adapter->pdev->dev;
 	int recovery_status;
+	int delay = ERR_DETECTION_DELAY;
 
 	be_detect_error(adapter);
 
@@ -4899,6 +4900,7 @@ static void be_err_detection_task(struct work_struct *work)
 		goto reschedule_task;
 
 	if (!recovery_status) {
+		adapter->recovery_retries = 0;
 		dev_info(dev, "Adapter recovery successful\n");
 		goto reschedule_task;
 	} else if (be_virtfn(adapter)) {
@@ -4907,13 +4909,22 @@ static void be_err_detection_task(struct work_struct *work)
 		 */
 		dev_err(dev, "Re-trying adapter recovery\n");
 		goto reschedule_task;
+	} else if (adapter->recovery_retries++ <
+		   MAX_ERR_RECOVERY_RETRY_COUNT) {
+		/* In case of another error during recovery, it takes 30 sec
+		 * for adapter to come out of error. Retry error recovery after
+		 * this time interval.
+		 */
+		dev_err(&adapter->pdev->dev, "Re-trying adapter recovery\n");
+		delay = ERR_RECOVERY_RETRY_DELAY;
+		goto reschedule_task;
 	} else {
 		dev_err(dev, "Adapter recovery failed\n");
 	}
 
 	return;
 reschedule_task:
-	be_schedule_err_detection(adapter);
+	be_schedule_err_detection(adapter, delay);
 }
 
 static void be_log_sfp_info(struct be_adapter *adapter)
@@ -5309,7 +5320,7 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id)
 
 	be_roce_dev_add(adapter);
 
-	be_schedule_err_detection(adapter);
+	be_schedule_err_detection(adapter, ERR_DETECTION_DELAY);
 
 	/* On Die temperature not supported for VF. */
 	if (be_physfn(adapter) && IS_ENABLED(CONFIG_BE2NET_HWMON)) {
@@ -5376,7 +5387,7 @@ static int be_pci_resume(struct pci_dev *pdev)
 	if (status)
 		return status;
 
-	be_schedule_err_detection(adapter);
+	be_schedule_err_detection(adapter, ERR_DETECTION_DELAY);
 
 	if (adapter->wol_en)
 		be_setup_wol(adapter, false);
@@ -5476,7 +5487,7 @@ static void be_eeh_resume(struct pci_dev *pdev)
 	if (status)
 		goto err;
 
-	be_schedule_err_detection(adapter);
+	be_schedule_err_detection(adapter, ERR_DETECTION_DELAY);
 	return;
 err:
 	dev_err(&adapter->pdev->dev, "EEH resume failed\n");
-- 
cgit v0.10.2


From 3c0d49aaa67e1990f4c081892e9bb69070853919 Mon Sep 17 00:00:00 2001
From: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Date: Wed, 3 Feb 2016 09:49:23 +0530
Subject: be2net: Fix interval calculation in interrupt moderation

Interrupt moderation parameters need to be recalculated only
after a time interval of 1 ms. Interval calculation is wrong
when there is a rollover of jiffies. Using recommended way of interval
calculation using jiffies to fix this.

Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index d5286d3..9c1fc9d 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1917,8 +1917,7 @@ static u32 be_get_eq_delay_mult_enc(struct be_eq_obj *eqo)
 	if (!aic->enable)
 		return 0;
 
-	if (time_before_eq(now, aic->jiffies) ||
-	    jiffies_to_msecs(now - aic->jiffies) < 1)
+	if (jiffies_to_msecs(now - aic->jiffies) < 1)
 		eqd = aic->prev_eqd;
 	else
 		eqd = be_get_new_eqd(eqo);
-- 
cgit v0.10.2


From 22fae97d863679994b951799dd4bbe7afd95897b Mon Sep 17 00:00:00 2001
From: Paul Durrant <Paul.Durrant@citrix.com>
Date: Tue, 2 Feb 2016 11:55:05 +0000
Subject: xen-netback: implement dynamic multicast control

My recent patch to the Xen Project documents a protocol for 'dynamic
multicast control' in netif.h. This extends the previous multicast control
protocol to not require a shared ring reconnection to turn the feature off.
Instead the backend watches the "request-multicast-control" key in xenstore
and turns the feature off if the key value is written to zero.

This patch adds support for dynamic multicast control in xen-netback.

Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 0333ab0..1128252 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -251,6 +251,7 @@ struct xenvif {
 	unsigned int stalled_queues;
 
 	struct xenbus_watch credit_watch;
+	struct xenbus_watch mcast_ctrl_watch;
 
 	spinlock_t lock;
 
diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index 56ebd82..39a303d 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -327,7 +327,7 @@ static int netback_probe(struct xenbus_device *dev,
 			goto abort_transaction;
 		}
 
-		/* We support multicast-control. */
+		/* We support dynamic multicast-control. */
 		err = xenbus_printf(xbt, dev->nodename,
 				    "feature-multicast-control", "%d", 1);
 		if (err) {
@@ -335,6 +335,14 @@ static int netback_probe(struct xenbus_device *dev,
 			goto abort_transaction;
 		}
 
+		err = xenbus_printf(xbt, dev->nodename,
+				    "feature-dynamic-multicast-control",
+				    "%d", 1);
+		if (err) {
+			message = "writing feature-dynamic-multicast-control";
+			goto abort_transaction;
+		}
+
 		err = xenbus_transaction_end(xbt, 0);
 	} while (err == -EAGAIN);
 
@@ -683,7 +691,8 @@ static void xen_net_rate_changed(struct xenbus_watch *watch,
 	}
 }
 
-static int xen_register_watchers(struct xenbus_device *dev, struct xenvif *vif)
+static int xen_register_credit_watch(struct xenbus_device *dev,
+				     struct xenvif *vif)
 {
 	int err = 0;
 	char *node;
@@ -708,7 +717,7 @@ static int xen_register_watchers(struct xenbus_device *dev, struct xenvif *vif)
 	return err;
 }
 
-static void xen_unregister_watchers(struct xenvif *vif)
+static void xen_unregister_credit_watch(struct xenvif *vif)
 {
 	if (vif->credit_watch.node) {
 		unregister_xenbus_watch(&vif->credit_watch);
@@ -717,6 +726,75 @@ static void xen_unregister_watchers(struct xenvif *vif)
 	}
 }
 
+static void xen_mcast_ctrl_changed(struct xenbus_watch *watch,
+				   const char **vec, unsigned int len)
+{
+	struct xenvif *vif = container_of(watch, struct xenvif,
+					  mcast_ctrl_watch);
+	struct xenbus_device *dev = xenvif_to_xenbus_device(vif);
+	int val;
+
+	if (xenbus_scanf(XBT_NIL, dev->otherend,
+			 "request-multicast-control", "%d", &val) < 0)
+		val = 0;
+	vif->multicast_control = !!val;
+}
+
+static int xen_register_mcast_ctrl_watch(struct xenbus_device *dev,
+					 struct xenvif *vif)
+{
+	int err = 0;
+	char *node;
+	unsigned maxlen = strlen(dev->otherend) +
+		sizeof("/request-multicast-control");
+
+	if (vif->mcast_ctrl_watch.node) {
+		pr_err_ratelimited("Watch is already registered\n");
+		return -EADDRINUSE;
+	}
+
+	node = kmalloc(maxlen, GFP_KERNEL);
+	if (!node) {
+		pr_err("Failed to allocate memory for watch\n");
+		return -ENOMEM;
+	}
+	snprintf(node, maxlen, "%s/request-multicast-control",
+		 dev->otherend);
+	vif->mcast_ctrl_watch.node = node;
+	vif->mcast_ctrl_watch.callback = xen_mcast_ctrl_changed;
+	err = register_xenbus_watch(&vif->mcast_ctrl_watch);
+	if (err) {
+		pr_err("Failed to set watcher %s\n",
+		       vif->mcast_ctrl_watch.node);
+		kfree(node);
+		vif->mcast_ctrl_watch.node = NULL;
+		vif->mcast_ctrl_watch.callback = NULL;
+	}
+	return err;
+}
+
+static void xen_unregister_mcast_ctrl_watch(struct xenvif *vif)
+{
+	if (vif->mcast_ctrl_watch.node) {
+		unregister_xenbus_watch(&vif->mcast_ctrl_watch);
+		kfree(vif->mcast_ctrl_watch.node);
+		vif->mcast_ctrl_watch.node = NULL;
+	}
+}
+
+static void xen_register_watchers(struct xenbus_device *dev,
+				  struct xenvif *vif)
+{
+	xen_register_credit_watch(dev, vif);
+	xen_register_mcast_ctrl_watch(dev, vif);
+}
+
+static void xen_unregister_watchers(struct xenvif *vif)
+{
+	xen_unregister_mcast_ctrl_watch(vif);
+	xen_unregister_credit_watch(vif);
+}
+
 static void unregister_hotplug_status_watch(struct backend_info *be)
 {
 	if (be->have_hotplug_status_watch) {
@@ -1030,11 +1108,6 @@ static int read_xenbus_vif_flags(struct backend_info *be)
 		val = 0;
 	vif->ipv6_csum = !!val;
 
-	if (xenbus_scanf(XBT_NIL, dev->otherend, "request-multicast-control",
-			 "%d", &val) < 0)
-		val = 0;
-	vif->multicast_control = !!val;
-
 	return 0;
 }
 
-- 
cgit v0.10.2


From 67eb03318bc5fe170ae832423fda7a23b0d801cf Mon Sep 17 00:00:00 2001
From: David Ahern <dsa@cumulusnetworks.com>
Date: Tue, 2 Feb 2016 07:43:45 -0800
Subject: net: Add support for fill_slave_info to VRF device

Allows userspace to have direct access to VRF table association
versus looking up master device and its table.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 66addb7..76e1fc9 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -877,6 +877,24 @@ static int vrf_fillinfo(struct sk_buff *skb,
 	return nla_put_u32(skb, IFLA_VRF_TABLE, vrf->tb_id);
 }
 
+static size_t vrf_get_slave_size(const struct net_device *bond_dev,
+				 const struct net_device *slave_dev)
+{
+	return nla_total_size(sizeof(u32));  /* IFLA_VRF_PORT_TABLE */
+}
+
+static int vrf_fill_slave_info(struct sk_buff *skb,
+			       const struct net_device *vrf_dev,
+			       const struct net_device *slave_dev)
+{
+	struct net_vrf *vrf = netdev_priv(vrf_dev);
+
+	if (nla_put_u32(skb, IFLA_VRF_PORT_TABLE, vrf->tb_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
 static const struct nla_policy vrf_nl_policy[IFLA_VRF_MAX + 1] = {
 	[IFLA_VRF_TABLE] = { .type = NLA_U32 },
 };
@@ -890,6 +908,9 @@ static struct rtnl_link_ops vrf_link_ops __read_mostly = {
 	.validate	= vrf_validate,
 	.fill_info	= vrf_fillinfo,
 
+	.get_slave_size  = vrf_get_slave_size,
+	.fill_slave_info = vrf_fill_slave_info,
+
 	.newlink	= vrf_newlink,
 	.dellink	= vrf_dellink,
 	.setup		= vrf_setup,
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index d3e90b9..d452cea 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -405,6 +405,14 @@ enum {
 
 #define IFLA_VRF_MAX (__IFLA_VRF_MAX - 1)
 
+enum {
+	IFLA_VRF_PORT_UNSPEC,
+	IFLA_VRF_PORT_TABLE,
+	__IFLA_VRF_PORT_MAX
+};
+
+#define IFLA_VRF_PORT_MAX (__IFLA_VRF_PORT_MAX - 1)
+
 /* IPVLAN section */
 enum {
 	IFLA_IPVLAN_UNSPEC,
-- 
cgit v0.10.2


From 3575dbf2cbbc8e598f17ec441aed526dbea0e1bd Mon Sep 17 00:00:00 2001
From: David Herrmann <dh.herrmann@gmail.com>
Date: Tue, 2 Feb 2016 18:17:54 +0100
Subject: net: drop write-only stack variable

Remove a write-only stack variable from unix_attach_fds(). This is a
left-over from the security fix in:

    commit 712f4aad406bb1ed67f3f98d04c044191f0ff593
    Author: willy tarreau <w@1wt.eu>
    Date:   Sun Jan 10 07:54:56 2016 +0100

        unix: properly account for FDs passed over unix sockets

Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 49d5093..b374555 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1534,7 +1534,6 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)
 {
 	int i;
 	unsigned char max_level = 0;
-	int unix_sock_count = 0;
 
 	if (too_many_unix_fds(current))
 		return -ETOOMANYREFS;
@@ -1542,11 +1541,9 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)
 	for (i = scm->fp->count - 1; i >= 0; i--) {
 		struct sock *sk = unix_get_socket(scm->fp->fp[i]);
 
-		if (sk) {
-			unix_sock_count++;
+		if (sk)
 			max_level = max(max_level,
 					unix_sk(sk)->recursion_level);
-		}
 	}
 	if (unlikely(max_level > MAX_RECURSION_LEVEL))
 		return -ETOOMANYREFS;
-- 
cgit v0.10.2


From e662ca40de846e0a2be6326a7c4668326ddb194c Mon Sep 17 00:00:00 2001
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 2 Feb 2016 10:33:04 -0800
Subject: tcp: retransmit after recovery processing and congestion control

The retransmission and F-RTO transmission currently happen inside
recovery state processing (tcp_fastretrans_alert) but before
congestion control.  This refactoring moves the logic after both
s.t. we can determine how much to send (cwnd) before deciding what to
send.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 8194a25..84a4ab9 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -126,6 +126,10 @@ int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2;
 #define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH)
 #define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH))
 
+#define REXMIT_NONE	0 /* no loss recovery to do */
+#define REXMIT_LOST	1 /* retransmit packets marked lost */
+#define REXMIT_NEW	2 /* FRTO-style transmit of unsent/new packets */
+
 /* Adapt the MSS value used to make delayed ack decision to the
  * real world.
  */
@@ -2662,7 +2666,8 @@ static void tcp_enter_recovery(struct sock *sk, bool ece_ack)
 /* Process an ACK in CA_Loss state. Move to CA_Open if lost data are
  * recovered or spurious. Otherwise retransmits more on partial ACKs.
  */
-static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack)
+static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack,
+			     int *rexmit)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	bool recovered = !before(tp->snd_una, tp->high_seq);
@@ -2684,10 +2689,15 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack)
 				tp->frto = 0; /* Step 3.a. loss was real */
 		} else if (flag & FLAG_SND_UNA_ADVANCED && !recovered) {
 			tp->high_seq = tp->snd_nxt;
-			__tcp_push_pending_frames(sk, tcp_current_mss(sk),
-						  TCP_NAGLE_OFF);
-			if (after(tp->snd_nxt, tp->high_seq))
-				return; /* Step 2.b */
+			/* Step 2.b. Try send new data (but deferred until cwnd
+			 * is updated in tcp_ack()). Otherwise fall back to
+			 * the conventional recovery.
+			 */
+			if (tcp_send_head(sk) &&
+			    after(tcp_wnd_end(tp), tp->snd_nxt)) {
+				*rexmit = REXMIT_NEW;
+				return;
+			}
 			tp->frto = 0;
 		}
 	}
@@ -2706,7 +2716,7 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack)
 		else if (flag & FLAG_SND_UNA_ADVANCED)
 			tcp_reset_reno_sack(tp);
 	}
-	tcp_xmit_retransmit_queue(sk);
+	*rexmit = REXMIT_LOST;
 }
 
 /* Undo during fast recovery after partial ACK. */
@@ -2756,7 +2766,7 @@ static bool tcp_try_undo_partial(struct sock *sk, const int acked,
  */
 static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 				  const int prior_unsacked,
-				  bool is_dupack, int flag)
+				  bool is_dupack, int flag, int *rexmit)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -2831,7 +2841,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 		}
 		break;
 	case TCP_CA_Loss:
-		tcp_process_loss(sk, flag, is_dupack);
+		tcp_process_loss(sk, flag, is_dupack, rexmit);
 		if (icsk->icsk_ca_state != TCP_CA_Open &&
 		    !(flag & FLAG_LOST_RETRANS))
 			return;
@@ -2871,7 +2881,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 	if (do_lost)
 		tcp_update_scoreboard(sk, fast_rexmit);
 	tcp_cwnd_reduction(sk, prior_unsacked, fast_rexmit, flag);
-	tcp_xmit_retransmit_queue(sk);
+	*rexmit = REXMIT_LOST;
 }
 
 /* Kathleen Nichols' algorithm for tracking the minimum value of
@@ -3506,6 +3516,27 @@ static inline void tcp_in_ack_event(struct sock *sk, u32 flags)
 		icsk->icsk_ca_ops->in_ack_event(sk, flags);
 }
 
+/* Congestion control has updated the cwnd already. So if we're in
+ * loss recovery then now we do any new sends (for FRTO) or
+ * retransmits (for CA_Loss or CA_recovery) that make sense.
+ */
+static void tcp_xmit_recovery(struct sock *sk, int rexmit)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+
+	if (rexmit == REXMIT_NONE)
+		return;
+
+	if (unlikely(rexmit == 2)) {
+		__tcp_push_pending_frames(sk, tcp_current_mss(sk),
+					  TCP_NAGLE_OFF);
+		if (after(tp->snd_nxt, tp->high_seq))
+			return;
+		tp->frto = 0;
+	}
+	tcp_xmit_retransmit_queue(sk);
+}
+
 /* This routine deals with incoming acks, but not outgoing ones. */
 static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 {
@@ -3520,6 +3551,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	int prior_packets = tp->packets_out;
 	const int prior_unsacked = tp->packets_out - tp->sacked_out;
 	int acked = 0; /* Number of packets newly acked */
+	int rexmit = REXMIT_NONE; /* Flag to (re)transmit to recover losses */
 
 	sack_state.first_sackt.v64 = 0;
 
@@ -3616,7 +3648,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	if (tcp_ack_is_dubious(sk, flag)) {
 		is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP));
 		tcp_fastretrans_alert(sk, acked, prior_unsacked,
-				      is_dupack, flag);
+				      is_dupack, flag, &rexmit);
 	}
 	if (tp->tlp_high_seq)
 		tcp_process_tlp_ack(sk, ack, flag);
@@ -3634,13 +3666,14 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	if (icsk->icsk_pending == ICSK_TIME_RETRANS)
 		tcp_schedule_loss_probe(sk);
 	tcp_update_pacing_rate(sk);
+	tcp_xmit_recovery(sk, rexmit);
 	return 1;
 
 no_queue:
 	/* If data was DSACKed, see if we can undo a cwnd reduction. */
 	if (flag & FLAG_DSACKING_ACK)
 		tcp_fastretrans_alert(sk, acked, prior_unsacked,
-				      is_dupack, flag);
+				      is_dupack, flag, &rexmit);
 	/* If this ack opens up a zero window, clear backoff.  It was
 	 * being used to time the probes, and is probably far higher than
 	 * it needs to be for normal retransmission.
@@ -3664,7 +3697,8 @@ old_ack:
 		flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una,
 						&sack_state);
 		tcp_fastretrans_alert(sk, acked, prior_unsacked,
-				      is_dupack, flag);
+				      is_dupack, flag, &rexmit);
+		tcp_xmit_recovery(sk, rexmit);
 	}
 
 	SOCK_DEBUG(sk, "Ack %u before %u:%u\n", ack, tp->snd_una, tp->snd_nxt);
-- 
cgit v0.10.2


From 31ba0c10723e9eba378f96de1d1a9426129949e1 Mon Sep 17 00:00:00 2001
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 2 Feb 2016 10:33:05 -0800
Subject: tcp: move cwnd reduction after recovery state procesing

Currently the cwnd is reduced and increased in various different
places. The reduction happens in various places in the recovery
state processing (tcp_fastretrans_alert) while the increase
happens afterward.

A better sequence is to identify lost packets and update
the congestion control state (icsk_ca_state) first. Then base
on the new state, up/down the cwnd in one central place. It's
more clear to reason cwnd changes.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 84a4ab9..dc810df 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2471,14 +2471,12 @@ static void tcp_init_cwnd_reduction(struct sock *sk)
 	tcp_ecn_queue_cwr(tp);
 }
 
-static void tcp_cwnd_reduction(struct sock *sk, const int prior_unsacked,
-			       int fast_rexmit, int flag)
+static void tcp_cwnd_reduction(struct sock *sk, int newly_acked_sacked,
+			       int flag)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	int sndcnt = 0;
 	int delta = tp->snd_ssthresh - tcp_packets_in_flight(tp);
-	int newly_acked_sacked = prior_unsacked -
-				 (tp->packets_out - tp->sacked_out);
 
 	if (newly_acked_sacked <= 0 || WARN_ON_ONCE(!tp->prior_cwnd))
 		return;
@@ -2496,7 +2494,8 @@ static void tcp_cwnd_reduction(struct sock *sk, const int prior_unsacked,
 	} else {
 		sndcnt = min(delta, newly_acked_sacked);
 	}
-	sndcnt = max(sndcnt, (fast_rexmit ? 1 : 0));
+	/* Force a fast retransmit upon entering fast recovery */
+	sndcnt = max(sndcnt, (tp->prr_out ? 0 : 1));
 	tp->snd_cwnd = tcp_packets_in_flight(tp) + sndcnt;
 }
 
@@ -2541,7 +2540,7 @@ static void tcp_try_keep_open(struct sock *sk)
 	}
 }
 
-static void tcp_try_to_open(struct sock *sk, int flag, const int prior_unsacked)
+static void tcp_try_to_open(struct sock *sk, int flag)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
@@ -2555,8 +2554,6 @@ static void tcp_try_to_open(struct sock *sk, int flag, const int prior_unsacked)
 
 	if (inet_csk(sk)->icsk_ca_state != TCP_CA_CWR) {
 		tcp_try_keep_open(sk);
-	} else {
-		tcp_cwnd_reduction(sk, prior_unsacked, 0, flag);
 	}
 }
 
@@ -2720,8 +2717,7 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack,
 }
 
 /* Undo during fast recovery after partial ACK. */
-static bool tcp_try_undo_partial(struct sock *sk, const int acked,
-				 const int prior_unsacked, int flag)
+static bool tcp_try_undo_partial(struct sock *sk, const int acked)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
@@ -2736,10 +2732,8 @@ static bool tcp_try_undo_partial(struct sock *sk, const int acked,
 		 * can undo. Otherwise we clock out new packets but do not
 		 * mark more packets lost or retransmit more.
 		 */
-		if (tp->retrans_out) {
-			tcp_cwnd_reduction(sk, prior_unsacked, 0, flag);
+		if (tp->retrans_out)
 			return true;
-		}
 
 		if (!tcp_any_retrans_done(sk))
 			tp->retrans_stamp = 0;
@@ -2758,21 +2752,21 @@ static bool tcp_try_undo_partial(struct sock *sk, const int acked,
  * taking into account both packets sitting in receiver's buffer and
  * packets lost by network.
  *
- * Besides that it does CWND reduction, when packet loss is detected
- * and changes state of machine.
+ * Besides that it updates the congestion state when packet loss or ECN
+ * is detected. But it does not reduce the cwnd, it is done by the
+ * congestion control later.
  *
  * It does _not_ decide what to send, it is made in function
  * tcp_xmit_retransmit_queue().
  */
 static void tcp_fastretrans_alert(struct sock *sk, const int acked,
-				  const int prior_unsacked,
-				  bool is_dupack, int flag, int *rexmit)
+				  bool is_dupack, int *ack_flag, int *rexmit)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	int fast_rexmit = 0, flag = *ack_flag;
 	bool do_lost = is_dupack || ((flag & FLAG_DATA_SACKED) &&
 				    (tcp_fackets_out(tp) > tp->reordering));
-	int fast_rexmit = 0;
 
 	if (WARN_ON(!tp->packets_out && tp->sacked_out))
 		tp->sacked_out = 0;
@@ -2819,8 +2813,10 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 
 	/* Use RACK to detect loss */
 	if (sysctl_tcp_recovery & TCP_RACK_LOST_RETRANS &&
-	    tcp_rack_mark_lost(sk))
+	    tcp_rack_mark_lost(sk)) {
 		flag |= FLAG_LOST_RETRANS;
+		*ack_flag |= FLAG_LOST_RETRANS;
+	}
 
 	/* E. Process state. */
 	switch (icsk->icsk_ca_state) {
@@ -2829,7 +2825,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 			if (tcp_is_reno(tp) && is_dupack)
 				tcp_add_reno_sack(sk);
 		} else {
-			if (tcp_try_undo_partial(sk, acked, prior_unsacked, flag))
+			if (tcp_try_undo_partial(sk, acked))
 				return;
 			/* Partial ACK arrived. Force fast retransmit. */
 			do_lost = tcp_is_reno(tp) ||
@@ -2858,7 +2854,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 			tcp_try_undo_dsack(sk);
 
 		if (!tcp_time_to_recover(sk, flag)) {
-			tcp_try_to_open(sk, flag, prior_unsacked);
+			tcp_try_to_open(sk, flag);
 			return;
 		}
 
@@ -2880,7 +2876,6 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked,
 
 	if (do_lost)
 		tcp_update_scoreboard(sk, fast_rexmit);
-	tcp_cwnd_reduction(sk, prior_unsacked, fast_rexmit, flag);
 	*rexmit = REXMIT_LOST;
 }
 
@@ -3306,9 +3301,6 @@ static inline bool tcp_ack_is_dubious(const struct sock *sk, const int flag)
 /* Decide wheather to run the increase function of congestion control. */
 static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag)
 {
-	if (tcp_in_cwnd_reduction(sk))
-		return false;
-
 	/* If reordering is high then always grow cwnd whenever data is
 	 * delivered regardless of its ordering. Otherwise stay conservative
 	 * and only grow cwnd on in-order delivery (RFC5681). A stretched ACK w/
@@ -3551,6 +3543,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	int prior_packets = tp->packets_out;
 	const int prior_unsacked = tp->packets_out - tp->sacked_out;
 	int acked = 0; /* Number of packets newly acked */
+	int acked_sacked; /* Number of packets newly acked or sacked */
 	int rexmit = REXMIT_NONE; /* Flag to (re)transmit to recover losses */
 
 	sack_state.first_sackt.v64 = 0;
@@ -3647,15 +3640,20 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 
 	if (tcp_ack_is_dubious(sk, flag)) {
 		is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP));
-		tcp_fastretrans_alert(sk, acked, prior_unsacked,
-				      is_dupack, flag, &rexmit);
+		tcp_fastretrans_alert(sk, acked, is_dupack, &flag, &rexmit);
 	}
 	if (tp->tlp_high_seq)
 		tcp_process_tlp_ack(sk, ack, flag);
 
+	acked_sacked = prior_unsacked - (tp->packets_out - tp->sacked_out);
 	/* Advance cwnd if state allows */
-	if (tcp_may_raise_cwnd(sk, flag))
+	if (tcp_in_cwnd_reduction(sk)) {
+		/* Reduce cwnd if state mandates */
+		tcp_cwnd_reduction(sk, acked_sacked, flag);
+	} else if (tcp_may_raise_cwnd(sk, flag)) {
+		/* Advance cwnd if state allows */
 		tcp_cong_avoid(sk, ack, acked);
+	}
 
 	if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) {
 		struct dst_entry *dst = __sk_dst_get(sk);
@@ -3672,8 +3670,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 no_queue:
 	/* If data was DSACKed, see if we can undo a cwnd reduction. */
 	if (flag & FLAG_DSACKING_ACK)
-		tcp_fastretrans_alert(sk, acked, prior_unsacked,
-				      is_dupack, flag, &rexmit);
+		tcp_fastretrans_alert(sk, acked, is_dupack, &flag, &rexmit);
 	/* If this ack opens up a zero window, clear backoff.  It was
 	 * being used to time the probes, and is probably far higher than
 	 * it needs to be for normal retransmission.
@@ -3696,8 +3693,7 @@ old_ack:
 	if (TCP_SKB_CB(skb)->sacked) {
 		flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una,
 						&sack_state);
-		tcp_fastretrans_alert(sk, acked, prior_unsacked,
-				      is_dupack, flag, &rexmit);
+		tcp_fastretrans_alert(sk, acked, is_dupack, &flag, &rexmit);
 		tcp_xmit_recovery(sk, rexmit);
 	}
 
-- 
cgit v0.10.2


From ddf1af6fa00e772fdb67a7d22cb83fac2b8968a8 Mon Sep 17 00:00:00 2001
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 2 Feb 2016 10:33:06 -0800
Subject: tcp: new delivery accounting

This patch changes the accounting of how many packets are
newly acked or sacked when the sender receives an ACK.

The current approach basically computes

   newly_acked_sacked = (prior_packets - prior_sacked) -
                        (tp->packets_out - tp->sacked_out)

   where prior_packets and prior_sacked out are snapshot
   at the beginning of the ACK processing.

The new approach tracks the delivery information via a new
TCP state variable "delivered" which monotically increases
as new packets are delivered in order or out-of-order.

The reason for this change is that the current approach is
brittle that produces negative or inaccurate estimate.

   1) For non-SACK connections, an ACK that advances the SND.UNA
   could reset the DUPACK counters (tp->sacked_out) in
   tcp_process_loss() or tcp_fastretrans_alert(). This inflates
   the inflight suddenly and causes under-estimate or even
   negative estimate. Here is a real example:

                   before   after (processing ACK)
   packets_out     75       73
   sacked_out      23        0
   ca state        Loss     Open

   The old approach computes (75-23) - (73 - 0) = -21 delivered
   while the new approach computes 1 delivered since it
   considers the 2nd-24th packets are delivered OOO.

   2) MSS change would re-count packets_out and sacked_out so
   the estimate is in-accurate and can even become negative.
   E.g., the inflight is doubled when MSS is halved.

   3) Spurious retransmission signaled by DSACK is not accounted

The new approach is simpler and more robust. For SACK connections,
tp->delivered increments as packets are being acked or sacked in
SACK and ACK processing.

For non-sack connections, it's done in tcp_remove_reno_sacks() and
tcp_add_reno_sack(). When an ACK advances the SND.UNA, tp->delivered
is incremented by the number of packets ACKed (less the current
number of DUPACKs received plus one packet hole).  Upon receiving
a DUPACK, tp->delivered is incremented assuming one out-of-order
packet is delivered.

Upon receiving a DSACK, tp->delivered is incremtened assuming one
retransmission is delivered in tcp_sacktag_write_queue().

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index b386361..d909fee 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -256,6 +256,7 @@ struct tcp_sock {
 	u32	prr_delivered;	/* Number of newly delivered packets to
 				 * receiver in Recovery. */
 	u32	prr_out;	/* Total number of pkts sent during Recovery. */
+	u32	delivered;	/* Total data packets delivered incl. rexmits */
 
  	u32	rcv_wnd;	/* Current receiver window		*/
 	u32	write_seq;	/* Tail(+1) of data held in tcp send buffer */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index dc810df..2d690b3 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1214,6 +1214,7 @@ static u8 tcp_sacktag_one(struct sock *sk,
 		sacked |= TCPCB_SACKED_ACKED;
 		state->flag |= FLAG_DATA_SACKED;
 		tp->sacked_out += pcount;
+		tp->delivered += pcount;  /* Out-of-order packets delivered */
 
 		fack_count += pcount;
 
@@ -1825,8 +1826,12 @@ static void tcp_check_reno_reordering(struct sock *sk, const int addend)
 static void tcp_add_reno_sack(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
+	u32 prior_sacked = tp->sacked_out;
+
 	tp->sacked_out++;
 	tcp_check_reno_reordering(sk, 0);
+	if (tp->sacked_out > prior_sacked)
+		tp->delivered++; /* Some out-of-order packet is delivered */
 	tcp_verify_left_out(tp);
 }
 
@@ -1838,6 +1843,7 @@ static void tcp_remove_reno_sacks(struct sock *sk, int acked)
 
 	if (acked > 0) {
 		/* One ACK acked hole. The rest eat duplicate ACKs. */
+		tp->delivered += max_t(int, acked - tp->sacked_out, 1);
 		if (acked - 1 >= tp->sacked_out)
 			tp->sacked_out = 0;
 		else
@@ -3156,10 +3162,13 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 				flag |= FLAG_ORIG_SACK_ACKED;
 		}
 
-		if (sacked & TCPCB_SACKED_ACKED)
+		if (sacked & TCPCB_SACKED_ACKED) {
 			tp->sacked_out -= acked_pcount;
-		else if (tcp_is_sack(tp) && !tcp_skb_spurious_retrans(tp, skb))
-			tcp_rack_advance(tp, &skb->skb_mstamp, sacked);
+		} else if (tcp_is_sack(tp)) {
+			tp->delivered += acked_pcount;
+			if (!tcp_skb_spurious_retrans(tp, skb))
+				tcp_rack_advance(tp, &skb->skb_mstamp, sacked);
+		}
 		if (sacked & TCPCB_LOST)
 			tp->lost_out -= acked_pcount;
 
@@ -3541,9 +3550,9 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	bool is_dupack = false;
 	u32 prior_fackets;
 	int prior_packets = tp->packets_out;
-	const int prior_unsacked = tp->packets_out - tp->sacked_out;
+	u32 prior_delivered = tp->delivered;
 	int acked = 0; /* Number of packets newly acked */
-	int acked_sacked; /* Number of packets newly acked or sacked */
+	u32 acked_sacked; /* Number of packets newly acked or sacked */
 	int rexmit = REXMIT_NONE; /* Flag to (re)transmit to recover losses */
 
 	sack_state.first_sackt.v64 = 0;
@@ -3645,7 +3654,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	if (tp->tlp_high_seq)
 		tcp_process_tlp_ack(sk, ack, flag);
 
-	acked_sacked = prior_unsacked - (tp->packets_out - tp->sacked_out);
+	acked_sacked = tp->delivered - prior_delivered;
 	/* Advance cwnd if state allows */
 	if (tcp_in_cwnd_reduction(sk)) {
 		/* Reduce cwnd if state mandates */
-- 
cgit v0.10.2


From 3ebd88710584d494b670e54b2c339e3be290956c Mon Sep 17 00:00:00 2001
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 2 Feb 2016 10:33:07 -0800
Subject: tcp: refactor pkts acked accounting

A small refactoring that gets number of packets cumulatively acked
from tcp_clean_rtx_queue() directly.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2d690b3..40824b3 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3104,7 +3104,7 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb,
  * arrived at the other end.
  */
 static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
-			       u32 prior_snd_una,
+			       u32 prior_snd_una, int *acked,
 			       struct tcp_sacktag_state *sack)
 {
 	const struct inet_connection_sock *icsk = inet_csk(sk);
@@ -3277,6 +3277,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 		}
 	}
 #endif
+	*acked = pkts_acked;
 	return flag;
 }
 
@@ -3642,10 +3643,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 		goto no_queue;
 
 	/* See if we can take anything off of the retransmit queue. */
-	acked = tp->packets_out;
-	flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una,
+	flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una, &acked,
 				    &sack_state);
-	acked -= tp->packets_out;
 
 	if (tcp_ack_is_dubious(sk, flag)) {
 		is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP));
-- 
cgit v0.10.2


From 2d14a4def4fc87cb2d2712f7841b45189d75e301 Mon Sep 17 00:00:00 2001
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 2 Feb 2016 10:33:08 -0800
Subject: tcp: make congestion control more robust against reordering

This change enables congestion control to update cwnd based on
not only packet cumulatively acked but also packets delivered
out-of-order. This makes congestion control robust against packet
reordering because it may raise cwnd as long as packets are being
delivered once reordering has been detected (i.e., it only cares
the amount of packets delivered, not the ordering among them).

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 40824b3..d598ff4 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3660,7 +3660,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 		tcp_cwnd_reduction(sk, acked_sacked, flag);
 	} else if (tcp_may_raise_cwnd(sk, flag)) {
 		/* Advance cwnd if state allows */
-		tcp_cong_avoid(sk, ack, acked);
+		tcp_cong_avoid(sk, ack, acked_sacked);
 	}
 
 	if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) {
-- 
cgit v0.10.2


From d452e6caf8367cc70cf940c24a6a6cc2d521d3c1 Mon Sep 17 00:00:00 2001
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 2 Feb 2016 10:33:09 -0800
Subject: tcp: tcp_cong_control helper

Refactor and consolidate cwnd and rate updates into a new function
tcp_cong_control().

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index d598ff4..596c1cb 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3323,6 +3323,24 @@ static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag)
 	return flag & FLAG_DATA_ACKED;
 }
 
+/* The "ultimate" congestion control function that aims to replace the rigid
+ * cwnd increase and decrease control (tcp_cong_avoid,tcp_*cwnd_reduction).
+ * It's called toward the end of processing an ACK with precise rate
+ * information. All transmission or retransmission are delayed afterwards.
+ */
+static void tcp_cong_control(struct sock *sk, u32 ack, u32 acked_sacked,
+			     int flag)
+{
+	if (tcp_in_cwnd_reduction(sk)) {
+		/* Reduce cwnd if state mandates */
+		tcp_cwnd_reduction(sk, acked_sacked, flag);
+	} else if (tcp_may_raise_cwnd(sk, flag)) {
+		/* Advance cwnd if state allows */
+		tcp_cong_avoid(sk, ack, acked_sacked);
+	}
+	tcp_update_pacing_rate(sk);
+}
+
 /* Check that window update is acceptable.
  * The function assumes that snd_una<=ack<=snd_next.
  */
@@ -3553,7 +3571,6 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	int prior_packets = tp->packets_out;
 	u32 prior_delivered = tp->delivered;
 	int acked = 0; /* Number of packets newly acked */
-	u32 acked_sacked; /* Number of packets newly acked or sacked */
 	int rexmit = REXMIT_NONE; /* Flag to (re)transmit to recover losses */
 
 	sack_state.first_sackt.v64 = 0;
@@ -3653,16 +3670,6 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	if (tp->tlp_high_seq)
 		tcp_process_tlp_ack(sk, ack, flag);
 
-	acked_sacked = tp->delivered - prior_delivered;
-	/* Advance cwnd if state allows */
-	if (tcp_in_cwnd_reduction(sk)) {
-		/* Reduce cwnd if state mandates */
-		tcp_cwnd_reduction(sk, acked_sacked, flag);
-	} else if (tcp_may_raise_cwnd(sk, flag)) {
-		/* Advance cwnd if state allows */
-		tcp_cong_avoid(sk, ack, acked_sacked);
-	}
-
 	if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) {
 		struct dst_entry *dst = __sk_dst_get(sk);
 		if (dst)
@@ -3671,7 +3678,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 
 	if (icsk->icsk_pending == ICSK_TIME_RETRANS)
 		tcp_schedule_loss_probe(sk);
-	tcp_update_pacing_rate(sk);
+	tcp_cong_control(sk, ack, tp->delivered - prior_delivered, flag);
 	tcp_xmit_recovery(sk, rexmit);
 	return 1;
 
-- 
cgit v0.10.2


From 46fcc6ef9d39eb7b1becaa5ef5cba64d230f7c3f Mon Sep 17 00:00:00 2001
From: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Date: Tue, 2 Feb 2016 10:41:55 -0800
Subject: sunvnet: Add support for perf LDC event tracing

Add perf event macros for support of tracing and instrumentation
of LDC state machine

Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/trace/events/sunvnet.h b/include/trace/events/sunvnet.h
new file mode 100644
index 0000000..eb080b2
--- /dev/null
+++ b/include/trace/events/sunvnet.h
@@ -0,0 +1,139 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM sunvnet
+
+#if !defined(_TRACE_SUNVNET_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_SUNVNET_H
+
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(vnet_rx_one,
+
+	TP_PROTO(int lsid, int rsid, int index, int needs_ack),
+
+	TP_ARGS(lsid, rsid, index, needs_ack),
+
+	TP_STRUCT__entry(
+		__field(int, lsid)
+		__field(int, rsid)
+		__field(int, index)
+		__field(int, needs_ack)
+	),
+
+	TP_fast_assign(
+		__entry->lsid = lsid;
+		__entry->rsid = rsid;
+		__entry->index = index;
+		__entry->needs_ack = needs_ack;
+	),
+
+	TP_printk("(%x:%x) walk_rx_one index %d; needs_ack %d",
+		__entry->lsid, __entry->rsid,
+		__entry->index, __entry->needs_ack)
+);
+
+DECLARE_EVENT_CLASS(vnet_tx_stopped_ack_template,
+
+	TP_PROTO(int lsid, int rsid, int ack_end, int npkts),
+
+	TP_ARGS(lsid, rsid, ack_end, npkts),
+
+	TP_STRUCT__entry(
+		__field(int, lsid)
+		__field(int, rsid)
+		__field(int, ack_end)
+		__field(int, npkts)
+	),
+
+	TP_fast_assign(
+		__entry->lsid = lsid;
+		__entry->rsid = rsid;
+		__entry->ack_end = ack_end;
+		__entry->npkts = npkts;
+	),
+
+	TP_printk("(%x:%x) stopped ack for %d; npkts %d",
+		__entry->lsid, __entry->rsid,
+		__entry->ack_end, __entry->npkts)
+);
+DEFINE_EVENT(vnet_tx_stopped_ack_template, vnet_tx_send_stopped_ack,
+	     TP_PROTO(int lsid, int rsid, int ack_end, int npkts),
+	     TP_ARGS(lsid, rsid, ack_end, npkts));
+DEFINE_EVENT(vnet_tx_stopped_ack_template, vnet_tx_defer_stopped_ack,
+	     TP_PROTO(int lsid, int rsid, int ack_end, int npkts),
+	     TP_ARGS(lsid, rsid, ack_end, npkts));
+DEFINE_EVENT(vnet_tx_stopped_ack_template, vnet_tx_pending_stopped_ack,
+	     TP_PROTO(int lsid, int rsid, int ack_end, int npkts),
+	     TP_ARGS(lsid, rsid, ack_end, npkts));
+
+TRACE_EVENT(vnet_rx_stopped_ack,
+
+	TP_PROTO(int lsid, int rsid, int end),
+
+	TP_ARGS(lsid, rsid, end),
+
+	TP_STRUCT__entry(
+		__field(int, lsid)
+		__field(int, rsid)
+		__field(int, end)
+	),
+
+	TP_fast_assign(
+		__entry->lsid = lsid;
+		__entry->rsid = rsid;
+		__entry->end = end;
+	),
+
+	TP_printk("(%x:%x) stopped ack for index %d",
+		__entry->lsid, __entry->rsid, __entry->end)
+);
+
+TRACE_EVENT(vnet_tx_trigger,
+
+	TP_PROTO(int lsid, int rsid, int start, int err),
+
+	TP_ARGS(lsid, rsid, start, err),
+
+	TP_STRUCT__entry(
+		__field(int, lsid)
+		__field(int, rsid)
+		__field(int, start)
+		__field(int, err)
+	),
+
+	TP_fast_assign(
+		__entry->lsid = lsid;
+		__entry->rsid = rsid;
+		__entry->start = start;
+		__entry->err = err;
+	),
+
+	TP_printk("(%x:%x) Tx trigger for %d sent with err %d %s",
+		__entry->lsid, __entry->rsid, __entry->start,
+		__entry->err, __entry->err > 0 ? "(ok)" : " ")
+);
+
+TRACE_EVENT(vnet_skip_tx_trigger,
+
+	TP_PROTO(int lsid, int rsid, int last),
+
+	TP_ARGS(lsid, rsid, last),
+
+	TP_STRUCT__entry(
+		__field(int, lsid)
+		__field(int, rsid)
+		__field(int, last)
+	),
+
+	TP_fast_assign(
+		__entry->lsid = lsid;
+		__entry->rsid = rsid;
+		__entry->last = last;
+	),
+
+	TP_printk("(%x:%x) Skip Tx trigger. Last trigger sent was %d",
+		__entry->lsid, __entry->rsid, __entry->last)
+);
+#endif /* _TRACE_SOCK_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
-- 
cgit v0.10.2


From 365a10289fccbf769ad8b172c30716fac50fa278 Mon Sep 17 00:00:00 2001
From: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Date: Tue, 2 Feb 2016 10:41:56 -0800
Subject: sunvnet: perf tracepoint invocations to trace LDC state machine

Use sunvnet perf trace macros to monitor LDC message exchange state.

Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/sun/sunvnet.c b/drivers/net/ethernet/sun/sunvnet.c
index 23fa298..942a95d 100644
--- a/drivers/net/ethernet/sun/sunvnet.c
+++ b/drivers/net/ethernet/sun/sunvnet.c
@@ -17,6 +17,8 @@
 #include <linux/mutex.h>
 #include <linux/highmem.h>
 #include <linux/if_vlan.h>
+#define CREATE_TRACE_POINTS
+#include <trace/events/sunvnet.h>
 
 #if IS_ENABLED(CONFIG_IPV6)
 #include <linux/icmpv6.h>
@@ -540,6 +542,8 @@ static int vnet_walk_rx_one(struct vnet_port *port,
 	err = vnet_rx_one(port, desc);
 	if (err == -ECONNRESET)
 		return err;
+	trace_vnet_rx_one(port->vio._local_sid, port->vio._peer_sid,
+			  index, desc->hdr.ack);
 	desc->hdr.state = VIO_DESC_DONE;
 	err = put_rx_desc(port, dr, desc, index);
 	if (err < 0)
@@ -587,9 +591,15 @@ static int vnet_walk_rx(struct vnet_port *port, struct vio_dring_state *dr,
 		ack_start = ack_end = vio_dring_prev(dr, start);
 	if (send_ack) {
 		port->napi_resume = false;
+		trace_vnet_tx_send_stopped_ack(port->vio._local_sid,
+					       port->vio._peer_sid,
+					       ack_end, *npkts);
 		return vnet_send_ack(port, dr, ack_start, ack_end,
 				     VIO_DRING_STOPPED);
 	} else  {
+		trace_vnet_tx_defer_stopped_ack(port->vio._local_sid,
+						port->vio._peer_sid,
+						ack_end, *npkts);
 		port->napi_resume = true;
 		port->napi_stop_idx = ack_end;
 		return 1;
@@ -663,6 +673,8 @@ static int vnet_ack(struct vnet_port *port, void *msgbuf)
 	/* sync for race conditions with vnet_start_xmit() and tell xmit it
 	 * is time to send a trigger.
 	 */
+	trace_vnet_rx_stopped_ack(port->vio._local_sid,
+				  port->vio._peer_sid, end);
 	dr->cons = vio_dring_next(dr, end);
 	desc = vio_dring_entry(dr, dr->cons);
 	if (desc->hdr.state == VIO_DESC_READY && !port->start_cons) {
@@ -886,6 +898,9 @@ static int __vnet_tx_trigger(struct vnet_port *port, u32 start)
 	int retries = 0;
 
 	if (port->stop_rx) {
+		trace_vnet_tx_pending_stopped_ack(port->vio._local_sid,
+						  port->vio._peer_sid,
+						  port->stop_rx_idx, -1);
 		err = vnet_send_ack(port,
 				    &port->vio.drings[VIO_DRIVER_RX_RING],
 				    port->stop_rx_idx, -1,
@@ -908,6 +923,8 @@ static int __vnet_tx_trigger(struct vnet_port *port, u32 start)
 		if (retries++ > VNET_MAX_RETRIES)
 			break;
 	} while (err == -EAGAIN);
+	trace_vnet_tx_trigger(port->vio._local_sid,
+			      port->vio._peer_sid, start, err);
 
 	return err;
 }
@@ -1414,8 +1431,11 @@ static int vnet_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	 * producer to consumer announcement that work is available to the
 	 * consumer
 	 */
-	if (!port->start_cons)
-		goto ldc_start_done; /* previous trigger suffices */
+	if (!port->start_cons) { /* previous trigger suffices */
+		trace_vnet_skip_tx_trigger(port->vio._local_sid,
+					   port->vio._peer_sid, dr->cons);
+		goto ldc_start_done;
+	}
 
 	err = __vnet_tx_trigger(port, dr->cons);
 	if (unlikely(err < 0)) {
-- 
cgit v0.10.2


From 103a8ad1fa3b261c78dfc842cb315defe9d40be0 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Wed, 3 Feb 2016 04:04:36 +0100
Subject: ethtool: add speed/duplex validation functions

Add functions which check if the speed/duplex are defined.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 57fa390..b2e1801 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -1319,11 +1319,45 @@ enum ethtool_sfeatures_retval_bits {
 
 #define SPEED_UNKNOWN		-1
 
+static inline int ethtool_validate_speed(__u32 speed)
+{
+	switch (speed) {
+	case SPEED_10:
+	case SPEED_100:
+	case SPEED_1000:
+	case SPEED_2500:
+	case SPEED_5000:
+	case SPEED_10000:
+	case SPEED_20000:
+	case SPEED_25000:
+	case SPEED_40000:
+	case SPEED_50000:
+	case SPEED_56000:
+	case SPEED_100000:
+	case SPEED_UNKNOWN:
+		return 1;
+	}
+
+	return 0;
+}
+
 /* Duplex, half or full. */
 #define DUPLEX_HALF		0x00
 #define DUPLEX_FULL		0x01
 #define DUPLEX_UNKNOWN		0xff
 
+static inline int ethtool_validate_duplex(__u8 duplex)
+{
+	switch (duplex) {
+	case DUPLEX_HALF:
+	case DUPLEX_FULL:
+	case DUPLEX_UNKNOWN:
+		return 1;
+	}
+
+	return 0;
+}
+
 /* Which connector port. */
 #define PORT_TP			0x00
 #define PORT_AUI		0x01
-- 
cgit v0.10.2


From 16032be56c1f66770da15cb94f0eb366c37aff6e Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Wed, 3 Feb 2016 04:04:37 +0100
Subject: virtio_net: add ethtool support for set and get of settings

This patch allows the user to set and retrieve speed and duplex of the
virtio_net device via ethtool. Having this functionality is very helpful
for simulating different environments and also enables the virtio_net
device to participate in operations where proper speed and duplex are
required (e.g. currently bonding lacp mode requires full duplex). Custom
speed and duplex are not allowed, the user-supplied settings are validated
before applying.

Example:
$ ethtool eth1
Settings for eth1:
...
	Speed: Unknown!
	Duplex: Unknown! (255)
$ ethtool -s eth1 speed 1000 duplex full
$ ethtool eth1
Settings for eth1:
...
	Speed: 1000Mb/s
	Duplex: Full

Based on a patch by Roopa Prabhu.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 767ab11..c9fd52a 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -146,6 +146,10 @@ struct virtnet_info {
 	virtio_net_ctrl_ack ctrl_status;
 	u8 ctrl_promisc;
 	u8 ctrl_allmulti;
+
+	/* Ethtool settings */
+	u8 duplex;
+	u32 speed;
 };
 
 struct padded_vnet_hdr {
@@ -1376,6 +1380,58 @@ static void virtnet_get_channels(struct net_device *dev,
 	channels->other_count = 0;
 }
 
+/* Check if the user is trying to change anything besides speed/duplex */
+static bool virtnet_validate_ethtool_cmd(const struct ethtool_cmd *cmd)
+{
+	struct ethtool_cmd diff1 = *cmd;
+	struct ethtool_cmd diff2 = {};
+
+	/* advertising and cmd are usually set, ignore port because we set it */
+	ethtool_cmd_speed_set(&diff1, 0);
+	diff1.advertising = 0;
+	diff1.duplex = 0;
+	diff1.port = 0;
+	diff1.cmd = 0;
+
+	return !memcmp(&diff1, &diff2, sizeof(diff1));
+}
+
+static int virtnet_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+	u32 speed;
+
+	speed = ethtool_cmd_speed(cmd);
+	/* don't allow custom speed and duplex */
+	if (!ethtool_validate_speed(speed) ||
+	    !ethtool_validate_duplex(cmd->duplex) ||
+	    !virtnet_validate_ethtool_cmd(cmd))
+		return -EINVAL;
+	vi->speed = speed;
+	vi->duplex = cmd->duplex;
+
+	return 0;
+}
+
+static int virtnet_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+
+	ethtool_cmd_speed_set(cmd, vi->speed);
+	cmd->duplex = vi->duplex;
+	cmd->port = PORT_OTHER;
+
+	return 0;
+}
+
+static void virtnet_init_settings(struct net_device *dev)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+
+	vi->speed = SPEED_UNKNOWN;
+	vi->duplex = DUPLEX_UNKNOWN;
+}
+
 static const struct ethtool_ops virtnet_ethtool_ops = {
 	.get_drvinfo = virtnet_get_drvinfo,
 	.get_link = ethtool_op_get_link,
@@ -1383,6 +1439,8 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
 	.set_channels = virtnet_set_channels,
 	.get_channels = virtnet_get_channels,
 	.get_ts_info = ethtool_op_get_ts_info,
+	.get_settings = virtnet_get_settings,
+	.set_settings = virtnet_set_settings,
 };
 
 #define MIN_MTU 68
@@ -1855,6 +1913,8 @@ static int virtnet_probe(struct virtio_device *vdev)
 	netif_set_real_num_tx_queues(dev, vi->curr_queue_pairs);
 	netif_set_real_num_rx_queues(dev, vi->curr_queue_pairs);
 
+	virtnet_init_settings(dev);
+
 	err = register_netdev(dev);
 	if (err) {
 		pr_debug("virtio_net: registering device failed\n");
-- 
cgit v0.10.2


From 6fa251663069e05daadd1666cbf3b658bf840ea4 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:49 +0200
Subject: ipv4: Namespaceify tcp syn retries sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 2b7907a..b7b5bd6 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -98,6 +98,8 @@ struct netns_ipv4 {
 	int sysctl_tcp_keepalive_probes;
 	int sysctl_tcp_keepalive_intvl;
 
+	int sysctl_tcp_syn_retries;
+
 	struct ping_group_range ping_group_range;
 
 	atomic_t dev_addr_genid;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 479d535..825485c 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps;
 extern int sysctl_tcp_window_scaling;
 extern int sysctl_tcp_sack;
 extern int sysctl_tcp_fin_timeout;
-extern int sysctl_tcp_syn_retries;
 extern int sysctl_tcp_synack_retries;
 extern int sysctl_tcp_retries1;
 extern int sysctl_tcp_retries2;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 4d367b4..ae9dd88 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -292,15 +292,6 @@ static struct ctl_table ipv4_table[] = {
 		.extra2		= &ip_ttl_max,
 	},
 	{
-		.procname	= "tcp_syn_retries",
-		.data		= &sysctl_tcp_syn_retries,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &tcp_syn_retries_min,
-		.extra2		= &tcp_syn_retries_max
-	},
-	{
 		.procname	= "tcp_synack_retries",
 		.data		= &sysctl_tcp_synack_retries,
 		.maxlen		= sizeof(int),
@@ -960,6 +951,15 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
+	{
+		.procname	= "tcp_syn_retries",
+		.data		= &init_net.ipv4.sysctl_tcp_syn_retries,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &tcp_syn_retries_min,
+		.extra2		= &tcp_syn_retries_max
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index c507577..3dbb363 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2731,6 +2731,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	int val, len;
 
 	if (get_user(len, optlen))
@@ -2765,7 +2766,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 		val = keepalive_probes(tp);
 		break;
 	case TCP_SYNCNT:
-		val = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries;
+		val = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries;
 		break;
 	case TCP_LINGER2:
 		val = tp->linger2;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a4d5237..f746485 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2388,6 +2388,8 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_keepalive_probes = TCP_KEEPALIVE_PROBES;
 	net->ipv4.sysctl_tcp_keepalive_intvl = TCP_KEEPALIVE_INTVL;
 
+	net->ipv4.sysctl_tcp_syn_retries = TCP_SYN_RETRIES;
+
 	return 0;
 fail:
 	tcp_sk_exit(net);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index a4730a2..c5d51f5 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -22,7 +22,6 @@
 #include <linux/gfp.h>
 #include <net/tcp.h>
 
-int sysctl_tcp_syn_retries __read_mostly = TCP_SYN_RETRIES;
 int sysctl_tcp_synack_retries __read_mostly = TCP_SYNACK_RETRIES;
 int sysctl_tcp_retries1 __read_mostly = TCP_RETR1;
 int sysctl_tcp_retries2 __read_mostly = TCP_RETR2;
@@ -157,6 +156,7 @@ static int tcp_write_timeout(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	int retry_until;
 	bool do_reset, syn_set = false;
 
@@ -169,7 +169,7 @@ static int tcp_write_timeout(struct sock *sk)
 				NET_INC_STATS_BH(sock_net(sk),
 						 LINUX_MIB_TCPFASTOPENACTIVEFAIL);
 		}
-		retry_until = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries;
+		retry_until = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries;
 		syn_set = true;
 	} else {
 		if (retransmits_timed_out(sk, sysctl_tcp_retries1, 0, 0)) {
-- 
cgit v0.10.2


From 7c083ecb3ba4583a625d5ff9655d1a819e374493 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:50 +0200
Subject: ipv4: Namespaceify tcp synack retries sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index b7b5bd6..9e83084 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -99,6 +99,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_keepalive_intvl;
 
 	int sysctl_tcp_syn_retries;
+	int sysctl_tcp_synack_retries;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 825485c..05659e8 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps;
 extern int sysctl_tcp_window_scaling;
 extern int sysctl_tcp_sack;
 extern int sysctl_tcp_fin_timeout;
-extern int sysctl_tcp_synack_retries;
 extern int sysctl_tcp_retries1;
 extern int sysctl_tcp_retries2;
 extern int sysctl_tcp_orphan_retries;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 46b9c88..9b17c179 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -482,10 +482,6 @@ EXPORT_SYMBOL_GPL(inet_csk_route_child_sock);
 #define AF_INET_FAMILY(fam) true
 #endif
 
-/* Only thing we need from tcp.h */
-extern int sysctl_tcp_synack_retries;
-
-
 /* Decide when to expire the request and when to resend SYN-ACK */
 static inline void syn_ack_recalc(struct request_sock *req, const int thresh,
 				  const int max_retries,
@@ -557,6 +553,7 @@ static void reqsk_timer_handler(unsigned long data)
 {
 	struct request_sock *req = (struct request_sock *)data;
 	struct sock *sk_listener = req->rsk_listener;
+	struct net *net = sock_net(sk_listener);
 	struct inet_connection_sock *icsk = inet_csk(sk_listener);
 	struct request_sock_queue *queue = &icsk->icsk_accept_queue;
 	int qlen, expire = 0, resend = 0;
@@ -566,7 +563,7 @@ static void reqsk_timer_handler(unsigned long data)
 	if (sk_state_load(sk_listener) != TCP_LISTEN)
 		goto drop;
 
-	max_retries = icsk->icsk_syn_retries ? : sysctl_tcp_synack_retries;
+	max_retries = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_synack_retries;
 	thresh = max_retries;
 	/* Normally all the openreqs are young and become mature
 	 * (i.e. converted to established socket) for first timeout.
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index ae9dd88..bb682e3 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -292,13 +292,6 @@ static struct ctl_table ipv4_table[] = {
 		.extra2		= &ip_ttl_max,
 	},
 	{
-		.procname	= "tcp_synack_retries",
-		.data		= &sysctl_tcp_synack_retries,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "tcp_max_orphans",
 		.data		= &sysctl_tcp_max_orphans,
 		.maxlen		= sizeof(int),
@@ -960,6 +953,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.extra1		= &tcp_syn_retries_min,
 		.extra2		= &tcp_syn_retries_max
 	},
+	{
+		.procname	= "tcp_synack_retries",
+		.data		= &init_net.ipv4.sysctl_tcp_synack_retries,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index f746485..3146279 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2389,6 +2389,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_keepalive_intvl = TCP_KEEPALIVE_INTVL;
 
 	net->ipv4.sysctl_tcp_syn_retries = TCP_SYN_RETRIES;
+	net->ipv4.sysctl_tcp_synack_retries = TCP_SYNACK_RETRIES;
 
 	return 0;
 fail:
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index c5d51f5..ca25fdf 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -22,7 +22,6 @@
 #include <linux/gfp.h>
 #include <net/tcp.h>
 
-int sysctl_tcp_synack_retries __read_mostly = TCP_SYNACK_RETRIES;
 int sysctl_tcp_retries1 __read_mostly = TCP_RETR1;
 int sysctl_tcp_retries2 __read_mostly = TCP_RETR2;
 int sysctl_tcp_orphan_retries __read_mostly;
@@ -332,7 +331,7 @@ static void tcp_fastopen_synack_timer(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	int max_retries = icsk->icsk_syn_retries ? :
-	    sysctl_tcp_synack_retries + 1; /* add one more retry for fastopen */
+	    sock_net(sk)->ipv4.sysctl_tcp_synack_retries + 1; /* add one more retry for fastopen */
 	struct request_sock *req;
 
 	req = tcp_sk(sk)->fastopen_rsk;
-- 
cgit v0.10.2


From 12ed8244ed8b31b023ea6d2851fd8b15f2999e9b Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:51 +0200
Subject: ipv4: Namespaceify tcp syncookies sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 9e83084..ac000fc 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -101,6 +101,8 @@ struct netns_ipv4 {
 	int sysctl_tcp_syn_retries;
 	int sysctl_tcp_synack_retries;
 
+	int sysctl_tcp_syncookies;
+
 	struct ping_group_range ping_group_range;
 
 	atomic_t dev_addr_genid;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 05659e8..1fb23b7 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -243,7 +243,6 @@ extern int sysctl_tcp_fin_timeout;
 extern int sysctl_tcp_retries1;
 extern int sysctl_tcp_retries2;
 extern int sysctl_tcp_orphan_retries;
-extern int sysctl_tcp_syncookies;
 extern int sysctl_tcp_fastopen;
 extern int sysctl_tcp_retrans_collapse;
 extern int sysctl_tcp_stdurg;
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 643a86c..ba0dcff 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -19,8 +19,6 @@
 #include <net/tcp.h>
 #include <net/route.h>
 
-extern int sysctl_tcp_syncookies;
-
 static u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS] __read_mostly;
 
 #define COOKIEBITS 24	/* Upper bits store count */
@@ -307,7 +305,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	__u8 rcv_wscale;
 	struct flowi4 fl4;
 
-	if (!sysctl_tcp_syncookies || !th->ack || th->rst)
+	if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies || !th->ack || th->rst)
 		goto out;
 
 	if (tcp_synq_no_recent_overflow(sk))
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index bb682e3..d801425 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -341,15 +341,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
-#ifdef CONFIG_SYN_COOKIES
-	{
-		.procname	= "tcp_syncookies",
-		.data		= &sysctl_tcp_syncookies,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-#endif
 	{
 		.procname	= "tcp_fastopen",
 		.data		= &sysctl_tcp_fastopen,
@@ -960,6 +951,15 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
+#ifdef CONFIG_SYN_COOKIES
+	{
+		.procname	= "tcp_syncookies",
+		.data		= &init_net.ipv4.sysctl_tcp_syncookies,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+#endif
 	{ }
 };
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 596c1cb..b17aba4 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6163,9 +6163,10 @@ static bool tcp_syn_flood_action(const struct sock *sk,
 	struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
 	const char *msg = "Dropping request";
 	bool want_cookie = false;
+	struct net *net = sock_net(sk);
 
 #ifdef CONFIG_SYN_COOKIES
-	if (sysctl_tcp_syncookies) {
+	if (net->ipv4.sysctl_tcp_syncookies) {
 		msg = "Sending cookies";
 		want_cookie = true;
 		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDOCOOKIES);
@@ -6174,7 +6175,7 @@ static bool tcp_syn_flood_action(const struct sock *sk,
 		NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDROP);
 
 	if (!queue->synflood_warned &&
-	    sysctl_tcp_syncookies != 2 &&
+	    net->ipv4.sysctl_tcp_syncookies != 2 &&
 	    xchg(&queue->synflood_warned, 1) == 0)
 		pr_info("%s: Possible SYN flooding on port %d. %s.  Check SNMP counters.\n",
 			proto, ntohs(tcp_hdr(skb)->dest), msg);
@@ -6207,6 +6208,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 	__u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn;
 	struct tcp_options_received tmp_opt;
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	struct sock *fastopen_sk = NULL;
 	struct dst_entry *dst = NULL;
 	struct request_sock *req;
@@ -6217,7 +6219,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 	 * limitations, they conserve resources and peer is
 	 * evidently real one.
 	 */
-	if ((sysctl_tcp_syncookies == 2 ||
+	if ((net->ipv4.sysctl_tcp_syncookies == 2 ||
 	     inet_csk_reqsk_queue_is_full(sk)) && !isn) {
 		want_cookie = tcp_syn_flood_action(sk, skb, rsk_ops->slab_name);
 		if (!want_cookie)
@@ -6283,7 +6285,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 			}
 		}
 		/* Kill the following clause, if you dislike this way. */
-		else if (!sysctl_tcp_syncookies &&
+		else if (!net->ipv4.sysctl_tcp_syncookies &&
 			 (sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
 			  (sysctl_max_syn_backlog >> 2)) &&
 			 !tcp_peer_is_proven(req, dst, false,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3146279..98313d1 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -860,7 +860,6 @@ static void tcp_v4_reqsk_destructor(struct request_sock *req)
 	kfree(inet_rsk(req)->opt);
 }
 
-
 #ifdef CONFIG_TCP_MD5SIG
 /*
  * RFC2385 MD5 checksumming requires a mapping of
@@ -2391,6 +2390,8 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_syn_retries = TCP_SYN_RETRIES;
 	net->ipv4.sysctl_tcp_synack_retries = TCP_SYNACK_RETRIES;
 
+	net->ipv4.sysctl_tcp_syncookies = 0;
+
 	return 0;
 fail:
 	tcp_sk_exit(net);
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 75632a9..fadd8b9 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -27,9 +27,6 @@
 #include <net/inet_common.h>
 #include <net/xfrm.h>
 
-int sysctl_tcp_syncookies __read_mostly = 1;
-EXPORT_SYMBOL(sysctl_tcp_syncookies);
-
 int sysctl_tcp_abort_on_overflow __read_mostly;
 
 struct inet_timewait_death_row tcp_death_row = {
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 2906ef2..0e393ff 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -148,7 +148,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	struct dst_entry *dst;
 	__u8 rcv_wscale;
 
-	if (!sysctl_tcp_syncookies || !th->ack || th->rst)
+	if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies || !th->ack || th->rst)
 		goto out;
 
 	if (tcp_synq_no_recent_overflow(sk))
-- 
cgit v0.10.2


From 1043e25ff96a1efc7bd34d11f5f32203a28a3bd7 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:52 +0200
Subject: ipv4: Namespaceify tcp reordering sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index ac000fc..eb4cd0a 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -100,8 +100,8 @@ struct netns_ipv4 {
 
 	int sysctl_tcp_syn_retries;
 	int sysctl_tcp_synack_retries;
-
 	int sysctl_tcp_syncookies;
+	int sysctl_tcp_reordering;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 1fb23b7..7e9a147 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -961,9 +961,11 @@ static inline void tcp_enable_fack(struct tcp_sock *tp)
  */
 static inline void tcp_enable_early_retrans(struct tcp_sock *tp)
 {
+	struct net *net = sock_net((struct sock *)tp);
+
 	tp->do_early_retrans = sysctl_tcp_early_retrans &&
 		sysctl_tcp_early_retrans < 4 && !sysctl_tcp_thin_dupack &&
-		sysctl_tcp_reordering == 3;
+		net->ipv4.sysctl_tcp_reordering == 3;
 }
 
 static inline void tcp_disable_early_retrans(struct tcp_sock *tp)
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index d801425..7cd2057 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -456,13 +456,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec,
 	},
 	{
-		.procname	= "tcp_reordering",
-		.data		= &sysctl_tcp_reordering,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "tcp_max_reordering",
 		.data		= &sysctl_tcp_max_reordering,
 		.maxlen		= sizeof(int),
@@ -960,6 +953,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 #endif
+	{
+		.procname	= "tcp_reordering",
+		.data		= &init_net.ipv4.sysctl_tcp_reordering,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 3dbb363..f4db6b0 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -406,7 +406,7 @@ void tcp_init_sock(struct sock *sk)
 	tp->mss_cache = TCP_MSS_DEFAULT;
 	u64_stats_init(&tp->syncp);
 
-	tp->reordering = sysctl_tcp_reordering;
+	tp->reordering = sock_net(sk)->ipv4.sysctl_tcp_reordering;
 	tcp_enable_early_retrans(tp);
 	tcp_assign_congestion_control(sk);
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index b17aba4..5ee6fe0 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -80,9 +80,7 @@ int sysctl_tcp_timestamps __read_mostly = 1;
 int sysctl_tcp_window_scaling __read_mostly = 1;
 int sysctl_tcp_sack __read_mostly = 1;
 int sysctl_tcp_fack __read_mostly = 1;
-int sysctl_tcp_reordering __read_mostly = TCP_FASTRETRANS_THRESH;
 int sysctl_tcp_max_reordering __read_mostly = 300;
-EXPORT_SYMBOL(sysctl_tcp_reordering);
 int sysctl_tcp_dsack __read_mostly = 1;
 int sysctl_tcp_app_win __read_mostly = 31;
 int sysctl_tcp_adv_win_scale __read_mostly = 1;
@@ -1883,6 +1881,7 @@ void tcp_enter_loss(struct sock *sk)
 {
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	struct sk_buff *skb;
 	bool new_recovery = icsk->icsk_ca_state < TCP_CA_Recovery;
 	bool is_reneg;			/* is receiver reneging on SACKs? */
@@ -1933,9 +1932,9 @@ void tcp_enter_loss(struct sock *sk)
 	 * suggests that the degree of reordering is over-estimated.
 	 */
 	if (icsk->icsk_ca_state <= TCP_CA_Disorder &&
-	    tp->sacked_out >= sysctl_tcp_reordering)
+	    tp->sacked_out >= net->ipv4.sysctl_tcp_reordering)
 		tp->reordering = min_t(unsigned int, tp->reordering,
-				       sysctl_tcp_reordering);
+				       net->ipv4.sysctl_tcp_reordering);
 	tcp_set_ca_state(sk, TCP_CA_Loss);
 	tp->high_seq = tp->snd_nxt;
 	tcp_ecn_queue_cwr(tp);
@@ -2119,6 +2118,7 @@ static bool tcp_time_to_recover(struct sock *sk, int flag)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	__u32 packets_out;
+	int tcp_reordering = sock_net(sk)->ipv4.sysctl_tcp_reordering;
 
 	/* Trick#1: The loss is proven. */
 	if (tp->lost_out)
@@ -2133,7 +2133,7 @@ static bool tcp_time_to_recover(struct sock *sk, int flag)
 	 */
 	packets_out = tp->packets_out;
 	if (packets_out <= tp->reordering &&
-	    tp->sacked_out >= max_t(__u32, packets_out/2, sysctl_tcp_reordering) &&
+	    tp->sacked_out >= max_t(__u32, packets_out/2, tcp_reordering) &&
 	    !tcp_may_send_now(sk)) {
 		/* We have nothing to send. This connection is limited
 		 * either by receiver window or by application.
@@ -3317,7 +3317,7 @@ static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag)
 	 * new SACK or ECE mark may first advance cwnd here and later reduce
 	 * cwnd in tcp_fastretrans_alert() based on more states.
 	 */
-	if (tcp_sk(sk)->reordering > sysctl_tcp_reordering)
+	if (tcp_sk(sk)->reordering > sock_net(sk)->ipv4.sysctl_tcp_reordering)
 		return flag & FLAG_FORWARD_PROGRESS;
 
 	return flag & FLAG_DATA_ACKED;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 98313d1..10dfc8b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2389,8 +2389,8 @@ static int __net_init tcp_sk_init(struct net *net)
 
 	net->ipv4.sysctl_tcp_syn_retries = TCP_SYN_RETRIES;
 	net->ipv4.sysctl_tcp_synack_retries = TCP_SYNACK_RETRIES;
-
 	net->ipv4.sysctl_tcp_syncookies = 0;
+	net->ipv4.sysctl_tcp_reordering = TCP_FASTRETRANS_THRESH;
 
 	return 0;
 fail:
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index c8cbc2b..c26241f 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -369,6 +369,7 @@ void tcp_update_metrics(struct sock *sk)
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	struct dst_entry *dst = __sk_dst_get(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	struct tcp_metrics_block *tm;
 	unsigned long rtt;
 	u32 val;
@@ -473,7 +474,7 @@ void tcp_update_metrics(struct sock *sk)
 		if (!tcp_metric_locked(tm, TCP_METRIC_REORDERING)) {
 			val = tcp_metric_get(tm, TCP_METRIC_REORDERING);
 			if (val < tp->reordering &&
-			    tp->reordering != sysctl_tcp_reordering)
+			    tp->reordering != net->ipv4.sysctl_tcp_reordering)
 				tcp_metric_set(tm, TCP_METRIC_REORDERING,
 					       tp->reordering);
 		}
-- 
cgit v0.10.2


From ae5c3f406cffe15ffd2aa544961b7cd027468d46 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:53 +0200
Subject: ipv4: Namespaceify tcp_retries1 sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index eb4cd0a..dee6ba6 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -102,6 +102,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_synack_retries;
 	int sysctl_tcp_syncookies;
 	int sysctl_tcp_reordering;
+	int sysctl_tcp_retries1;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 7e9a147..da96b9a 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps;
 extern int sysctl_tcp_window_scaling;
 extern int sysctl_tcp_sack;
 extern int sysctl_tcp_fin_timeout;
-extern int sysctl_tcp_retries1;
 extern int sysctl_tcp_retries2;
 extern int sysctl_tcp_orphan_retries;
 extern int sysctl_tcp_fastopen;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 7cd2057..52853c6 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -320,14 +320,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "tcp_retries1",
-		.data		= &sysctl_tcp_retries1,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra2		= &tcp_retr1_max
-	},
-	{
 		.procname	= "tcp_retries2",
 		.data		= &sysctl_tcp_retries2,
 		.maxlen		= sizeof(int),
@@ -960,6 +952,14 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
+	{
+		.procname	= "tcp_retries1",
+		.data		= &init_net.ipv4.sysctl_tcp_retries1,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra2		= &tcp_retr1_max
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 10dfc8b..57fe3c6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2391,6 +2391,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_synack_retries = TCP_SYNACK_RETRIES;
 	net->ipv4.sysctl_tcp_syncookies = 0;
 	net->ipv4.sysctl_tcp_reordering = TCP_FASTRETRANS_THRESH;
+	net->ipv4.sysctl_tcp_retries1 = TCP_RETR1;
 
 	return 0;
 fail:
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index ca25fdf..6694e33 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -22,7 +22,6 @@
 #include <linux/gfp.h>
 #include <net/tcp.h>
 
-int sysctl_tcp_retries1 __read_mostly = TCP_RETR1;
 int sysctl_tcp_retries2 __read_mostly = TCP_RETR2;
 int sysctl_tcp_orphan_retries __read_mostly;
 int sysctl_tcp_thin_linear_timeouts __read_mostly;
@@ -171,7 +170,7 @@ static int tcp_write_timeout(struct sock *sk)
 		retry_until = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries;
 		syn_set = true;
 	} else {
-		if (retransmits_timed_out(sk, sysctl_tcp_retries1, 0, 0)) {
+		if (retransmits_timed_out(sk, net->ipv4.sysctl_tcp_retries1, 0, 0)) {
 			/* Some middle-boxes may black-hole Fast Open _after_
 			 * the handshake. Therefore we conservatively disable
 			 * Fast Open on this path on recurring timeouts with
@@ -180,7 +179,7 @@ static int tcp_write_timeout(struct sock *sk)
 			if (tp->syn_data_acked &&
 			    tp->bytes_acked <= tp->rx_opt.mss_clamp) {
 				tcp_fastopen_cache_set(sk, 0, NULL, true, 0);
-				if (icsk->icsk_retransmits == sysctl_tcp_retries1)
+				if (icsk->icsk_retransmits == net->ipv4.sysctl_tcp_retries1)
 					NET_INC_STATS_BH(sock_net(sk),
 							 LINUX_MIB_TCPFASTOPENACTIVEFAIL);
 			}
@@ -359,6 +358,7 @@ static void tcp_fastopen_synack_timer(struct sock *sk)
 void tcp_retransmit_timer(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	struct inet_connection_sock *icsk = inet_csk(sk);
 
 	if (tp->fastopen_rsk) {
@@ -489,7 +489,7 @@ out_reset_timer:
 		icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);
 	}
 	inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, TCP_RTO_MAX);
-	if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1, 0, 0))
+	if (retransmits_timed_out(sk, net->ipv4.sysctl_tcp_retries1 + 1, 0, 0))
 		__sk_dst_reset(sk);
 
 out:;
-- 
cgit v0.10.2


From c6214a97c86c660de4f7ddb8eed925192e646161 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:54 +0200
Subject: ipv4: Namespaceify tcp_retries2 sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index dee6ba6..d92c8e5 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -103,6 +103,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_syncookies;
 	int sysctl_tcp_reordering;
 	int sysctl_tcp_retries1;
+	int sysctl_tcp_retries2;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index da96b9a..a786cfa 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps;
 extern int sysctl_tcp_window_scaling;
 extern int sysctl_tcp_sack;
 extern int sysctl_tcp_fin_timeout;
-extern int sysctl_tcp_retries2;
 extern int sysctl_tcp_orphan_retries;
 extern int sysctl_tcp_fastopen;
 extern int sysctl_tcp_retrans_collapse;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 52853c6..8e339d4 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -320,13 +320,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "tcp_retries2",
-		.data		= &sysctl_tcp_retries2,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "tcp_fin_timeout",
 		.data		= &sysctl_tcp_fin_timeout,
 		.maxlen		= sizeof(int),
@@ -960,6 +953,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec_minmax,
 		.extra2		= &tcp_retr1_max
 	},
+	{
+		.procname	= "tcp_retries2",
+		.data		= &init_net.ipv4.sysctl_tcp_retries2,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 57fe3c6..0710e61 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2392,6 +2392,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_syncookies = 0;
 	net->ipv4.sysctl_tcp_reordering = TCP_FASTRETRANS_THRESH;
 	net->ipv4.sysctl_tcp_retries1 = TCP_RETR1;
+	net->ipv4.sysctl_tcp_retries2 = TCP_RETR2;
 
 	return 0;
 fail:
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index fda379c..7beb3f6 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3476,6 +3476,7 @@ void tcp_send_probe0(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	struct net *net = sock_net(sk);
 	unsigned long probe_max;
 	int err;
 
@@ -3489,7 +3490,7 @@ void tcp_send_probe0(struct sock *sk)
 	}
 
 	if (err <= 0) {
-		if (icsk->icsk_backoff < sysctl_tcp_retries2)
+		if (icsk->icsk_backoff < net->ipv4.sysctl_tcp_retries2)
 			icsk->icsk_backoff++;
 		icsk->icsk_probes_out++;
 		probe_max = TCP_RTO_MAX;
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 6694e33..09f4e02 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -22,7 +22,6 @@
 #include <linux/gfp.h>
 #include <net/tcp.h>
 
-int sysctl_tcp_retries2 __read_mostly = TCP_RETR2;
 int sysctl_tcp_orphan_retries __read_mostly;
 int sysctl_tcp_thin_linear_timeouts __read_mostly;
 
@@ -189,7 +188,7 @@ static int tcp_write_timeout(struct sock *sk)
 			dst_negative_advice(sk);
 		}
 
-		retry_until = sysctl_tcp_retries2;
+		retry_until = net->ipv4.sysctl_tcp_retries2;
 		if (sock_flag(sk, SOCK_DEAD)) {
 			const bool alive = icsk->icsk_rto < TCP_RTO_MAX;
 
@@ -303,7 +302,7 @@ static void tcp_probe_timer(struct sock *sk)
 		 (s32)(tcp_time_stamp - start_ts) > icsk->icsk_user_timeout)
 		goto abort;
 
-	max_probes = sysctl_tcp_retries2;
+	max_probes = sock_net(sk)->ipv4.sysctl_tcp_retries2;
 	if (sock_flag(sk, SOCK_DEAD)) {
 		const bool alive = inet_csk_rto_backoff(icsk, TCP_RTO_MAX) < TCP_RTO_MAX;
 
-- 
cgit v0.10.2


From c402d9beffb6141ab2e4d2ad8be71128803a28ca Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:55 +0200
Subject: ipv4: Namespaceify tcp_orphan_retries sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index d92c8e5..0802303 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -104,6 +104,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_reordering;
 	int sysctl_tcp_retries1;
 	int sysctl_tcp_retries2;
+	int sysctl_tcp_orphan_retries;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index a786cfa..71f840b 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -240,7 +240,6 @@ extern int sysctl_tcp_timestamps;
 extern int sysctl_tcp_window_scaling;
 extern int sysctl_tcp_sack;
 extern int sysctl_tcp_fin_timeout;
-extern int sysctl_tcp_orphan_retries;
 extern int sysctl_tcp_fastopen;
 extern int sysctl_tcp_retrans_collapse;
 extern int sysctl_tcp_stdurg;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 8e339d4..b7af633 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -420,13 +420,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec_jiffies,
 	},
 	{
-		.procname	= "tcp_orphan_retries",
-		.data		= &sysctl_tcp_orphan_retries,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "tcp_fack",
 		.data		= &sysctl_tcp_fack,
 		.maxlen		= sizeof(int),
@@ -960,6 +953,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
+	{
+		.procname	= "tcp_orphan_retries",
+		.data		= &init_net.ipv4.sysctl_tcp_orphan_retries,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 0710e61..1240dd6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2393,6 +2393,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_reordering = TCP_FASTRETRANS_THRESH;
 	net->ipv4.sysctl_tcp_retries1 = TCP_RETR1;
 	net->ipv4.sysctl_tcp_retries2 = TCP_RETR2;
+	net->ipv4.sysctl_tcp_orphan_retries = 0;
 
 	return 0;
 fail:
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 09f4e02..49bc474 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -22,7 +22,6 @@
 #include <linux/gfp.h>
 #include <net/tcp.h>
 
-int sysctl_tcp_orphan_retries __read_mostly;
 int sysctl_tcp_thin_linear_timeouts __read_mostly;
 
 static void tcp_write_err(struct sock *sk)
@@ -78,7 +77,7 @@ static int tcp_out_of_resources(struct sock *sk, bool do_reset)
 /* Calculate maximal number or retries on an orphaned socket. */
 static int tcp_orphan_retries(struct sock *sk, bool alive)
 {
-	int retries = sysctl_tcp_orphan_retries; /* May be zero. */
+	int retries = sock_net(sk)->ipv4.sysctl_tcp_orphan_retries; /* May be zero. */
 
 	/* We know from an ICMP that something is wrong. */
 	if (sk->sk_err_soft && !alive)
-- 
cgit v0.10.2


From 1e579caa18b96f9eb18f4f5416658cd15f37c062 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:56 +0200
Subject: ipv4: Namespaceify tcp_fin_timeout sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 0802303..de5ff43 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -105,6 +105,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_retries1;
 	int sysctl_tcp_retries2;
 	int sysctl_tcp_orphan_retries;
+	int sysctl_tcp_fin_timeout;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 71f840b..3f160c2 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -239,7 +239,6 @@ extern struct inet_timewait_death_row tcp_death_row;
 extern int sysctl_tcp_timestamps;
 extern int sysctl_tcp_window_scaling;
 extern int sysctl_tcp_sack;
-extern int sysctl_tcp_fin_timeout;
 extern int sysctl_tcp_fastopen;
 extern int sysctl_tcp_retrans_collapse;
 extern int sysctl_tcp_stdurg;
@@ -1249,7 +1248,7 @@ static inline u32 keepalive_time_elapsed(const struct tcp_sock *tp)
 
 static inline int tcp_fin_time(const struct sock *sk)
 {
-	int fin_timeout = tcp_sk(sk)->linger2 ? : sysctl_tcp_fin_timeout;
+	int fin_timeout = tcp_sk(sk)->linger2 ? : sock_net(sk)->ipv4.sysctl_tcp_fin_timeout;
 	const int rto = inet_csk(sk)->icsk_rto;
 
 	if (fin_timeout < (rto << 2) - (rto >> 1))
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index b7af633..8bd335a 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -320,13 +320,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "tcp_fin_timeout",
-		.data		= &sysctl_tcp_fin_timeout,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_jiffies,
-	},
-	{
 		.procname	= "tcp_fastopen",
 		.data		= &sysctl_tcp_fastopen,
 		.maxlen		= sizeof(int),
@@ -960,6 +953,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
+	{
+		.procname	= "tcp_fin_timeout",
+		.data		= &init_net.ipv4.sysctl_tcp_fin_timeout,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f4db6b0..014f18e 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -282,8 +282,6 @@
 #include <asm/unaligned.h>
 #include <net/busy_poll.h>
 
-int sysctl_tcp_fin_timeout __read_mostly = TCP_FIN_TIMEOUT;
-
 int sysctl_tcp_min_tso_segs __read_mostly = 2;
 
 int sysctl_tcp_autocorking __read_mostly = 1;
@@ -2330,6 +2328,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct inet_connection_sock *icsk = inet_csk(sk);
+	struct net *net = sock_net(sk);
 	int val;
 	int err = 0;
 
@@ -2526,7 +2525,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 	case TCP_LINGER2:
 		if (val < 0)
 			tp->linger2 = -1;
-		else if (val > sysctl_tcp_fin_timeout / HZ)
+		else if (val > net->ipv4.sysctl_tcp_fin_timeout / HZ)
 			tp->linger2 = 0;
 		else
 			tp->linger2 = val * HZ;
@@ -2771,7 +2770,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 	case TCP_LINGER2:
 		val = tp->linger2;
 		if (val >= 0)
-			val = (val ? : sysctl_tcp_fin_timeout) / HZ;
+			val = (val ? : net->ipv4.sysctl_tcp_fin_timeout) / HZ;
 		break;
 	case TCP_DEFER_ACCEPT:
 		val = retrans_to_secs(icsk->icsk_accept_queue.rskq_defer_accept,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 1240dd6..36c83c2 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2394,6 +2394,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_retries1 = TCP_RETR1;
 	net->ipv4.sysctl_tcp_retries2 = TCP_RETR2;
 	net->ipv4.sysctl_tcp_orphan_retries = 0;
+	net->ipv4.sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT;
 
 	return 0;
 fail:
-- 
cgit v0.10.2


From 4979f2d9f7262b9b180bc83de8d70f7a7721c085 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Wed, 3 Feb 2016 09:46:57 +0200
Subject: ipv4: Namespaceify tcp_notsent_lowat sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index de5ff43..4d6ec3f 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -106,6 +106,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_retries2;
 	int sysctl_tcp_orphan_retries;
 	int sysctl_tcp_fin_timeout;
+	unsigned int sysctl_tcp_notsent_lowat;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 3f160c2..9b2cb0c 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -267,7 +267,6 @@ extern int sysctl_tcp_thin_dupack;
 extern int sysctl_tcp_early_retrans;
 extern int sysctl_tcp_limit_output_bytes;
 extern int sysctl_tcp_challenge_ack_limit;
-extern unsigned int sysctl_tcp_notsent_lowat;
 extern int sysctl_tcp_min_tso_segs;
 extern int sysctl_tcp_min_rtt_wlen;
 extern int sysctl_tcp_autocorking;
@@ -1682,7 +1681,8 @@ void __tcp_v4_send_check(struct sk_buff *skb, __be32 saddr, __be32 daddr);
 
 static inline u32 tcp_notsent_lowat(const struct tcp_sock *tp)
 {
-	return tp->notsent_lowat ?: sysctl_tcp_notsent_lowat;
+	struct net *net = sock_net((struct sock *)tp);
+	return tp->notsent_lowat ?: net->ipv4.sysctl_tcp_notsent_lowat;
 }
 
 static inline bool tcp_stream_memory_free(const struct sock *sk)
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 8bd335a..44bb598 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -456,13 +456,6 @@ static struct ctl_table ipv4_table[] = {
 		.extra1		= &one,
 	},
 	{
-		.procname	= "tcp_notsent_lowat",
-		.data		= &sysctl_tcp_notsent_lowat,
-		.maxlen		= sizeof(sysctl_tcp_notsent_lowat),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
-	},
-	{
 		.procname	= "tcp_rmem",
 		.data		= &sysctl_tcp_rmem,
 		.maxlen		= sizeof(sysctl_tcp_rmem),
@@ -960,6 +953,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
+	{
+		.procname	= "tcp_notsent_lowat",
+		.data		= &init_net.ipv4.sysctl_tcp_notsent_lowat,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
 	{ }
 };
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 36c83c2..11ae706 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2395,6 +2395,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_retries2 = TCP_RETR2;
 	net->ipv4.sysctl_tcp_orphan_retries = 0;
 	net->ipv4.sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT;
+	net->ipv4.sysctl_tcp_notsent_lowat = UINT_MAX;
 
 	return 0;
 fail:
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 7beb3f6..7d2c7a4 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -62,9 +62,6 @@ int sysctl_tcp_tso_win_divisor __read_mostly = 3;
 /* By default, RFC2861 behavior.  */
 int sysctl_tcp_slow_start_after_idle __read_mostly = 1;
 
-unsigned int sysctl_tcp_notsent_lowat __read_mostly = UINT_MAX;
-EXPORT_SYMBOL(sysctl_tcp_notsent_lowat);
-
 static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
 			   int push_one, gfp_t gfp);
 
-- 
cgit v0.10.2


From 0aca737d4623b7b791b4b1e5f9b015fff5f7180c Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 8 Feb 2016 04:24:33 -0500
Subject: tcp: Fix syncookies sysctl default.

Unintentionally the default was changed to zero, fix
that.

Fixes: 12ed8244ed ("ipv4: Namespaceify tcp syncookies sysctl knob")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 11ae706..0d381fa 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2389,7 +2389,7 @@ static int __net_init tcp_sk_init(struct net *net)
 
 	net->ipv4.sysctl_tcp_syn_retries = TCP_SYN_RETRIES;
 	net->ipv4.sysctl_tcp_synack_retries = TCP_SYNACK_RETRIES;
-	net->ipv4.sysctl_tcp_syncookies = 0;
+	net->ipv4.sysctl_tcp_syncookies = 1;
 	net->ipv4.sysctl_tcp_reordering = TCP_FASTRETRANS_THRESH;
 	net->ipv4.sysctl_tcp_retries1 = TCP_RETR1;
 	net->ipv4.sysctl_tcp_retries2 = TCP_RETR2;
-- 
cgit v0.10.2


From d66bd9050bba153c35736e53b8a3ddbff328adc5 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Wed, 3 Feb 2016 02:02:32 +0000
Subject: bonding: trivial: style fixes

remove some redudant brackets, use sizeof(*) instead of sizeof(struct x).

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 7c9eb67..705cb01 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -928,11 +928,10 @@ void bond_select_active_slave(struct bonding *bond)
 		if (!rv)
 			return;
 
-		if (netif_carrier_ok(bond->dev)) {
+		if (netif_carrier_ok(bond->dev))
 			netdev_info(bond->dev, "first active interface up!\n");
-		} else {
+		else
 			netdev_info(bond->dev, "now running without any active interface!\n");
-		}
 	}
 }
 
@@ -1178,9 +1177,8 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb)
 		}
 	}
 
-	if (bond_should_deliver_exact_match(skb, slave, bond)) {
+	if (bond_should_deliver_exact_match(skb, slave, bond))
 		return RX_HANDLER_EXACT;
-	}
 
 	skb->dev = bond->dev;
 
@@ -1241,7 +1239,7 @@ static struct slave *bond_alloc_slave(struct bonding *bond)
 {
 	struct slave *slave = NULL;
 
-	slave = kzalloc(sizeof(struct slave), GFP_KERNEL);
+	slave = kzalloc(sizeof(*slave), GFP_KERNEL);
 	if (!slave)
 		return NULL;
 
-- 
cgit v0.10.2


From 157ede6784ba2837c7dc43f195418c75927f8488 Mon Sep 17 00:00:00 2001
From: Elad Raz <eladr@mellanox.com>
Date: Wed, 3 Feb 2016 09:57:04 +0100
Subject: bridge: mdb: add support for offloaded mdb entries

Add new bitmask member 'flags' to br_mdb_entry structure. Adding
MDB_FLAGS_OFFLOAD bit which indicates MDB entries is offloaded to hardware.

Signed-off-by: Elad Raz <eladr@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 18db144..ec35472 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -183,6 +183,8 @@ struct br_mdb_entry {
 #define MDB_TEMPORARY 0
 #define MDB_PERMANENT 1
 	__u8 state;
+#define MDB_FLAGS_OFFLOAD	(1 << 0)
+	__u8 flags;
 	__u16 vid;
 	struct {
 		union {
-- 
cgit v0.10.2


From 9d06b6d8a3fcba8eb8e1c62c92942452afea8057 Mon Sep 17 00:00:00 2001
From: Elad Raz <eladr@mellanox.com>
Date: Wed, 3 Feb 2016 09:57:05 +0100
Subject: bridge: mdb: Separate br_mdb_entry->state from
 net_bridge_port_group->state

Change net_bridge_port_group 'state' member to 'flags' and define new set
of flags internal to the kernel.

Signed-off-by: Elad Raz <eladr@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 30e105f..5312570 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -41,6 +41,14 @@ fail:
 	return -EMSGSIZE;
 }
 
+static void __mdb_entry_fill_flags(struct br_mdb_entry *e, unsigned char flags)
+{
+	e->state = flags & MDB_PG_FLAGS_PERMANENT;
+	e->flags = 0;
+	if (flags & MDB_PG_FLAGS_OFFLOAD)
+		e->flags |= MDB_FLAGS_OFFLOAD;
+}
+
 static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 			    struct net_device *dev)
 {
@@ -85,8 +93,8 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 					struct br_mdb_entry e;
 					memset(&e, 0, sizeof(e));
 					e.ifindex = port->dev->ifindex;
-					e.state = p->state;
 					e.vid = p->addr.vid;
+					__mdb_entry_fill_flags(&e, p->flags);
 					if (p->addr.proto == htons(ETH_P_IP))
 						e.addr.u.ip4 = p->addr.u.ip4;
 #if IS_ENABLED(CONFIG_IPV6)
@@ -254,7 +262,7 @@ errout:
 }
 
 void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port,
-		   struct br_ip *group, int type, u8 state)
+		   struct br_ip *group, int type, u8 flags)
 {
 	struct br_mdb_entry entry;
 
@@ -265,8 +273,8 @@ void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port,
 #if IS_ENABLED(CONFIG_IPV6)
 	entry.addr.u.ip6 = group->u.ip6;
 #endif
-	entry.state = state;
 	entry.vid = group->vid;
+	__mdb_entry_fill_flags(&entry, flags);
 	__br_mdb_notify(dev, &entry, type);
 }
 
@@ -568,7 +576,7 @@ static int __br_mdb_del(struct net_bridge *br, struct br_mdb_entry *entry)
 		if (p->port->state == BR_STATE_DISABLED)
 			goto unlock;
 
-		entry->state = p->state;
+		__mdb_entry_fill_flags(entry, p->flags);
 		rcu_assign_pointer(*pp, p->next);
 		hlist_del_init(&p->mglist);
 		del_timer(&p->timer);
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 03661d9..d156491 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -284,7 +284,7 @@ static void br_multicast_del_pg(struct net_bridge *br,
 		hlist_del_init(&p->mglist);
 		del_timer(&p->timer);
 		br_mdb_notify(br->dev, p->port, &pg->addr, RTM_DELMDB,
-			      p->state);
+			      p->flags);
 		call_rcu_bh(&p->rcu, br_multicast_free_pg);
 
 		if (!mp->ports && !mp->mglist &&
@@ -304,7 +304,7 @@ static void br_multicast_port_group_expired(unsigned long data)
 
 	spin_lock(&br->multicast_lock);
 	if (!netif_running(br->dev) || timer_pending(&pg->timer) ||
-	    hlist_unhashed(&pg->mglist) || pg->state & MDB_PERMANENT)
+	    hlist_unhashed(&pg->mglist) || pg->flags & MDB_PG_FLAGS_PERMANENT)
 		goto out;
 
 	br_multicast_del_pg(br, pg);
@@ -649,7 +649,7 @@ struct net_bridge_port_group *br_multicast_new_port_group(
 			struct net_bridge_port *port,
 			struct br_ip *group,
 			struct net_bridge_port_group __rcu *next,
-			unsigned char state)
+			unsigned char flags)
 {
 	struct net_bridge_port_group *p;
 
@@ -659,7 +659,7 @@ struct net_bridge_port_group *br_multicast_new_port_group(
 
 	p->addr = *group;
 	p->port = port;
-	p->state = state;
+	p->flags = flags;
 	rcu_assign_pointer(p->next, next);
 	hlist_add_head(&p->mglist, &port->mglist);
 	setup_timer(&p->timer, br_multicast_port_group_expired,
@@ -702,11 +702,11 @@ static int br_multicast_add_group(struct net_bridge *br,
 			break;
 	}
 
-	p = br_multicast_new_port_group(port, group, *pp, MDB_TEMPORARY);
+	p = br_multicast_new_port_group(port, group, *pp, 0);
 	if (unlikely(!p))
 		goto err;
 	rcu_assign_pointer(*pp, p);
-	br_mdb_notify(br->dev, port, group, RTM_NEWMDB, MDB_TEMPORARY);
+	br_mdb_notify(br->dev, port, group, RTM_NEWMDB, 0);
 
 found:
 	mod_timer(&p->timer, now + br->multicast_membership_interval);
@@ -975,7 +975,7 @@ void br_multicast_disable_port(struct net_bridge_port *port)
 
 	spin_lock(&br->multicast_lock);
 	hlist_for_each_entry_safe(pg, n, &port->mglist, mglist)
-		if (pg->state == MDB_TEMPORARY)
+		if (!(pg->flags & MDB_PG_FLAGS_PERMANENT))
 			br_multicast_del_pg(br, pg);
 
 	if (!hlist_unhashed(&port->rlist)) {
@@ -1454,7 +1454,7 @@ br_multicast_leave_group(struct net_bridge *br,
 			del_timer(&p->timer);
 			call_rcu_bh(&p->rcu, br_multicast_free_pg);
 			br_mdb_notify(br->dev, port, group, RTM_DELMDB,
-				      p->state);
+				      p->flags);
 
 			if (!mp->ports && !mp->mglist &&
 			    netif_running(br->dev))
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 216018c..67f4dce 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -150,6 +150,9 @@ struct net_bridge_fdb_entry
 	struct rcu_head			rcu;
 };
 
+#define MDB_PG_FLAGS_PERMANENT	BIT(0)
+#define MDB_PG_FLAGS_OFFLOAD	BIT(1)
+
 struct net_bridge_port_group {
 	struct net_bridge_port		*port;
 	struct net_bridge_port_group __rcu *next;
@@ -157,7 +160,7 @@ struct net_bridge_port_group {
 	struct rcu_head			rcu;
 	struct timer_list		timer;
 	struct br_ip			addr;
-	unsigned char			state;
+	unsigned char			flags;
 };
 
 struct net_bridge_mdb_entry
@@ -554,11 +557,11 @@ void br_multicast_free_pg(struct rcu_head *head);
 struct net_bridge_port_group *
 br_multicast_new_port_group(struct net_bridge_port *port, struct br_ip *group,
 			    struct net_bridge_port_group __rcu *next,
-			    unsigned char state);
+			    unsigned char flags);
 void br_mdb_init(void);
 void br_mdb_uninit(void);
 void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port,
-		   struct br_ip *group, int type, u8 state);
+		   struct br_ip *group, int type, u8 flags);
 void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port,
 		   int type);
 
-- 
cgit v0.10.2


From 9e8430f8d60d985c9f292b5eb5790cc13a87200c Mon Sep 17 00:00:00 2001
From: Elad Raz <eladr@mellanox.com>
Date: Wed, 3 Feb 2016 09:57:06 +0100
Subject: bridge: mdb: Passing the port-group pointer to br_mdb module

Passing the port-group to br_mdb in order to allow direct access to the
structure. br_mdb will later use the structure to reflect HW reflection
status via "state" variable.

Signed-off-by: Elad Raz <eladr@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 5312570..ac089286 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -217,7 +217,7 @@ static inline size_t rtnl_mdb_nlmsg_size(void)
 }
 
 static void __br_mdb_notify(struct net_device *dev, struct br_mdb_entry *entry,
-			    int type)
+			    int type, struct net_bridge_port_group *pg)
 {
 	struct switchdev_obj_port_mdb mdb = {
 		.obj = {
@@ -240,10 +240,13 @@ static void __br_mdb_notify(struct net_device *dev, struct br_mdb_entry *entry,
 #endif
 
 	mdb.obj.orig_dev = port_dev;
-	if (port_dev && type == RTM_NEWMDB)
-		switchdev_port_obj_add(port_dev, &mdb.obj);
-	else if (port_dev && type == RTM_DELMDB)
+	if (port_dev && type == RTM_NEWMDB) {
+		err = switchdev_port_obj_add(port_dev, &mdb.obj);
+		if (!err && pg)
+			pg->flags |= MDB_PG_FLAGS_OFFLOAD;
+	} else if (port_dev && type == RTM_DELMDB) {
 		switchdev_port_obj_del(port_dev, &mdb.obj);
+	}
 
 	skb = nlmsg_new(rtnl_mdb_nlmsg_size(), GFP_ATOMIC);
 	if (!skb)
@@ -261,21 +264,21 @@ errout:
 	rtnl_set_sk_err(net, RTNLGRP_MDB, err);
 }
 
-void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port,
-		   struct br_ip *group, int type, u8 flags)
+void br_mdb_notify(struct net_device *dev, struct net_bridge_port_group *pg,
+		   int type)
 {
 	struct br_mdb_entry entry;
 
 	memset(&entry, 0, sizeof(entry));
-	entry.ifindex = port->dev->ifindex;
-	entry.addr.proto = group->proto;
-	entry.addr.u.ip4 = group->u.ip4;
+	entry.ifindex = pg->port->dev->ifindex;
+	entry.addr.proto = pg->addr.proto;
+	entry.addr.u.ip4 = pg->addr.u.ip4;
 #if IS_ENABLED(CONFIG_IPV6)
-	entry.addr.u.ip6 = group->u.ip6;
+	entry.addr.u.ip6 = pg->addr.u.ip6;
 #endif
-	entry.vid = group->vid;
-	__mdb_entry_fill_flags(&entry, flags);
-	__br_mdb_notify(dev, &entry, type);
+	entry.vid = pg->addr.vid;
+	__mdb_entry_fill_flags(&entry, pg->flags);
+	__br_mdb_notify(dev, &entry, type, pg);
 }
 
 static int nlmsg_populate_rtr_fill(struct sk_buff *skb,
@@ -420,7 +423,8 @@ static int br_mdb_parse(struct sk_buff *skb, struct nlmsghdr *nlh,
 }
 
 static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port,
-			    struct br_ip *group, unsigned char state)
+			    struct br_ip *group, unsigned char state,
+			    struct net_bridge_port_group **pg)
 {
 	struct net_bridge_mdb_entry *mp;
 	struct net_bridge_port_group *p;
@@ -451,6 +455,7 @@ static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port,
 	if (unlikely(!p))
 		return -ENOMEM;
 	rcu_assign_pointer(*pp, p);
+	*pg = p;
 	if (state == MDB_TEMPORARY)
 		mod_timer(&p->timer, now + br->multicast_membership_interval);
 
@@ -458,7 +463,8 @@ static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port,
 }
 
 static int __br_mdb_add(struct net *net, struct net_bridge *br,
-			struct br_mdb_entry *entry)
+			struct br_mdb_entry *entry,
+			struct net_bridge_port_group **pg)
 {
 	struct br_ip ip;
 	struct net_device *dev;
@@ -487,7 +493,7 @@ static int __br_mdb_add(struct net *net, struct net_bridge *br,
 #endif
 
 	spin_lock_bh(&br->multicast_lock);
-	ret = br_mdb_add_group(br, p, &ip, entry->state);
+	ret = br_mdb_add_group(br, p, &ip, entry->state, pg);
 	spin_unlock_bh(&br->multicast_lock);
 	return ret;
 }
@@ -495,6 +501,7 @@ static int __br_mdb_add(struct net *net, struct net_bridge *br,
 static int br_mdb_add(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
 	struct net *net = sock_net(skb->sk);
+	struct net_bridge_port_group *pg;
 	struct net_bridge_vlan_group *vg;
 	struct net_device *dev, *pdev;
 	struct br_mdb_entry *entry;
@@ -524,15 +531,15 @@ static int br_mdb_add(struct sk_buff *skb, struct nlmsghdr *nlh)
 	if (br_vlan_enabled(br) && vg && entry->vid == 0) {
 		list_for_each_entry(v, &vg->vlan_list, vlist) {
 			entry->vid = v->vid;
-			err = __br_mdb_add(net, br, entry);
+			err = __br_mdb_add(net, br, entry, &pg);
 			if (err)
 				break;
-			__br_mdb_notify(dev, entry, RTM_NEWMDB);
+			__br_mdb_notify(dev, entry, RTM_NEWMDB, pg);
 		}
 	} else {
-		err = __br_mdb_add(net, br, entry);
+		err = __br_mdb_add(net, br, entry, &pg);
 		if (!err)
-			__br_mdb_notify(dev, entry, RTM_NEWMDB);
+			__br_mdb_notify(dev, entry, RTM_NEWMDB, pg);
 	}
 
 	return err;
@@ -628,12 +635,12 @@ static int br_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh)
 			entry->vid = v->vid;
 			err = __br_mdb_del(br, entry);
 			if (!err)
-				__br_mdb_notify(dev, entry, RTM_DELMDB);
+				__br_mdb_notify(dev, entry, RTM_DELMDB, NULL);
 		}
 	} else {
 		err = __br_mdb_del(br, entry);
 		if (!err)
-			__br_mdb_notify(dev, entry, RTM_DELMDB);
+			__br_mdb_notify(dev, entry, RTM_DELMDB, NULL);
 	}
 
 	return err;
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index d156491..8b6e424 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -283,8 +283,7 @@ static void br_multicast_del_pg(struct net_bridge *br,
 		rcu_assign_pointer(*pp, p->next);
 		hlist_del_init(&p->mglist);
 		del_timer(&p->timer);
-		br_mdb_notify(br->dev, p->port, &pg->addr, RTM_DELMDB,
-			      p->flags);
+		br_mdb_notify(br->dev, p, RTM_DELMDB);
 		call_rcu_bh(&p->rcu, br_multicast_free_pg);
 
 		if (!mp->ports && !mp->mglist &&
@@ -706,7 +705,7 @@ static int br_multicast_add_group(struct net_bridge *br,
 	if (unlikely(!p))
 		goto err;
 	rcu_assign_pointer(*pp, p);
-	br_mdb_notify(br->dev, port, group, RTM_NEWMDB, 0);
+	br_mdb_notify(br->dev, p, RTM_NEWMDB);
 
 found:
 	mod_timer(&p->timer, now + br->multicast_membership_interval);
@@ -1453,8 +1452,7 @@ br_multicast_leave_group(struct net_bridge *br,
 			hlist_del_init(&p->mglist);
 			del_timer(&p->timer);
 			call_rcu_bh(&p->rcu, br_multicast_free_pg);
-			br_mdb_notify(br->dev, port, group, RTM_DELMDB,
-				      p->flags);
+			br_mdb_notify(br->dev, p, RTM_DELMDB);
 
 			if (!mp->ports && !mp->mglist &&
 			    netif_running(br->dev))
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 67f4dce..302ab0a 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -560,8 +560,8 @@ br_multicast_new_port_group(struct net_bridge_port *port, struct br_ip *group,
 			    unsigned char flags);
 void br_mdb_init(void);
 void br_mdb_uninit(void);
-void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port,
-		   struct br_ip *group, int type, u8 flags);
+void br_mdb_notify(struct net_device *dev, struct net_bridge_port_group *pg,
+		   int type);
 void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port,
 		   int type);
 
-- 
cgit v0.10.2


From 5ee14e6d336f1daacf5ba73e831029c5ab7ae329 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Wed, 3 Feb 2016 13:17:01 +0100
Subject: bonding: 3ad: apply ad_actor settings changes immediately

Currently the bonding allows to set ad_actor_system and prio while the
bond device is down, but these are actually applied only if there aren't
any slaves yet (applied to bond device when first slave shows up, and to
slaves at 3ad bind time). After this patch changes are applied immediately
and the new values can be used/seen after the bond's upped so it's not
necessary anymore to release all and enslave again to see the changes.

CC: Jay Vosburgh <j.vosburgh@gmail.com>
CC: Veaceslav Falico <vfalico@gmail.com>
CC: Andy Gospodarek <gospo@cumulusnetworks.com>
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 4cbb8b2..ee94056 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -357,6 +357,14 @@ static u8 __get_duplex(struct port *port)
 	return retval;
 }
 
+static void __ad_actor_update_port(struct port *port)
+{
+	const struct bonding *bond = bond_get_bond_by_slave(port->slave);
+
+	port->actor_system = BOND_AD_INFO(bond).system.sys_mac_addr;
+	port->actor_system_priority = BOND_AD_INFO(bond).system.sys_priority;
+}
+
 /* Conversions */
 
 /**
@@ -1963,9 +1971,7 @@ void bond_3ad_bind_slave(struct slave *slave)
 		port->actor_admin_port_key = bond->params.ad_user_port_key << 6;
 		ad_update_actor_keys(port, false);
 		/* actor system is the bond's system */
-		port->actor_system = BOND_AD_INFO(bond).system.sys_mac_addr;
-		port->actor_system_priority =
-		    BOND_AD_INFO(bond).system.sys_priority;
+		__ad_actor_update_port(port);
 		/* tx timer(to verify that no more than MAX_TX_IN_SECOND
 		 * lacpdu's are sent in one second)
 		 */
@@ -2148,6 +2154,34 @@ out:
 }
 
 /**
+ * bond_3ad_update_ad_actor_settings - reflect change of actor settings to ports
+ * @bond: bonding struct to work on
+ *
+ * If an ad_actor setting gets changed we need to update the individual port
+ * settings so the bond device will use the new values when it gets upped.
+ */
+void bond_3ad_update_ad_actor_settings(struct bonding *bond)
+{
+	struct list_head *iter;
+	struct slave *slave;
+
+	ASSERT_RTNL();
+
+	BOND_AD_INFO(bond).system.sys_priority = bond->params.ad_actor_sys_prio;
+	if (is_zero_ether_addr(bond->params.ad_actor_system))
+		BOND_AD_INFO(bond).system.sys_mac_addr =
+		    *((struct mac_addr *)bond->dev->dev_addr);
+	else
+		BOND_AD_INFO(bond).system.sys_mac_addr =
+		    *((struct mac_addr *)bond->params.ad_actor_system);
+
+	spin_lock_bh(&bond->mode_lock);
+	bond_for_each_slave(bond, slave, iter)
+		__ad_actor_update_port(&(SLAVE_AD_INFO(slave)->port));
+	spin_unlock_bh(&bond->mode_lock);
+}
+
+/**
  * bond_3ad_state_machine_handler - handle state machines timeout
  * @bond: bonding struct to work on
  *
diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index 55e93b6..ed0bdae 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -1392,6 +1392,8 @@ static int bond_option_ad_actor_sys_prio_set(struct bonding *bond,
 		    newval->value);
 
 	bond->params.ad_actor_sys_prio = newval->value;
+	bond_3ad_update_ad_actor_settings(bond);
+
 	return 0;
 }
 
@@ -1418,6 +1420,8 @@ static int bond_option_ad_actor_system_set(struct bonding *bond,
 
 	netdev_info(bond->dev, "Setting ad_actor_system to %pM\n", mac);
 	ether_addr_copy(bond->params.ad_actor_system, mac);
+	bond_3ad_update_ad_actor_settings(bond);
+
 	return 0;
 
 err:
diff --git a/include/net/bond_3ad.h b/include/net/bond_3ad.h
index f1fbc3b..f358ad5 100644
--- a/include/net/bond_3ad.h
+++ b/include/net/bond_3ad.h
@@ -306,5 +306,6 @@ int bond_3ad_lacpdu_recv(const struct sk_buff *skb, struct bonding *bond,
 			 struct slave *slave);
 int bond_3ad_set_carrier(struct bonding *bond);
 void bond_3ad_update_lacp_rate(struct bonding *bond);
+void bond_3ad_update_ad_actor_settings(struct bonding *bond);
 #endif /* _NET_BOND_3AD_H */
 
-- 
cgit v0.10.2


From 16cc1400456a4d5d6df9f5e7f1b2c7cb8c50404b Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 3 Feb 2016 18:02:14 -0500
Subject: packet: move vnet_hdr code to helper functions

packet_snd and packet_rcv support virtio net headers for GSO.
Move this logic into helper functions to be able to reuse it in
tpacket_snd and tpacket_rcv.

This is a straighforward code move with one exception. Instead of
creating and passing a separate gso_type variable, reuse
vnet_hdr.gso_type after conversion from virtio to kernel gso type.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 992396a..bd3de7b 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -1960,6 +1960,64 @@ static unsigned int run_filter(struct sk_buff *skb,
 	return res;
 }
 
+static int __packet_rcv_vnet(const struct sk_buff *skb,
+			     struct virtio_net_hdr *vnet_hdr)
+{
+	*vnet_hdr = (const struct virtio_net_hdr) { 0 };
+
+	if (skb_is_gso(skb)) {
+		struct skb_shared_info *sinfo = skb_shinfo(skb);
+
+		/* This is a hint as to how much should be linear. */
+		vnet_hdr->hdr_len =
+			__cpu_to_virtio16(vio_le(), skb_headlen(skb));
+		vnet_hdr->gso_size =
+			__cpu_to_virtio16(vio_le(), sinfo->gso_size);
+
+		if (sinfo->gso_type & SKB_GSO_TCPV4)
+			vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+		else if (sinfo->gso_type & SKB_GSO_TCPV6)
+			vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+		else if (sinfo->gso_type & SKB_GSO_UDP)
+			vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP;
+		else if (sinfo->gso_type & SKB_GSO_FCOE)
+			return -EINVAL;
+		else
+			BUG();
+
+		if (sinfo->gso_type & SKB_GSO_TCP_ECN)
+			vnet_hdr->gso_type |= VIRTIO_NET_HDR_GSO_ECN;
+	} else
+		vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		vnet_hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+		vnet_hdr->csum_start = __cpu_to_virtio16(vio_le(),
+				  skb_checksum_start_offset(skb));
+		vnet_hdr->csum_offset = __cpu_to_virtio16(vio_le(),
+						 skb->csum_offset);
+	} else if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
+		vnet_hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID;
+	} /* else everything is zero */
+
+	return 0;
+}
+
+static int packet_rcv_vnet(struct msghdr *msg, const struct sk_buff *skb,
+			   size_t *len)
+{
+	struct virtio_net_hdr vnet_hdr;
+
+	if (*len < sizeof(vnet_hdr))
+		return -EINVAL;
+	*len -= sizeof(vnet_hdr);
+
+	if (__packet_rcv_vnet(skb, &vnet_hdr))
+		return -EINVAL;
+
+	return memcpy_to_msg(msg, (void *)&vnet_hdr, sizeof(vnet_hdr));
+}
+
 /*
  * This function makes lazy skb cloning in hope that most of packets
  * are discarded by BPF.
@@ -2347,6 +2405,84 @@ static void tpacket_set_protocol(const struct net_device *dev,
 	}
 }
 
+static int __packet_snd_vnet_parse(struct virtio_net_hdr *vnet_hdr, size_t len)
+{
+	unsigned short gso_type = 0;
+
+	if ((vnet_hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) &&
+	    (__virtio16_to_cpu(vio_le(), vnet_hdr->csum_start) +
+	     __virtio16_to_cpu(vio_le(), vnet_hdr->csum_offset) + 2 >
+	      __virtio16_to_cpu(vio_le(), vnet_hdr->hdr_len)))
+		vnet_hdr->hdr_len = __cpu_to_virtio16(vio_le(),
+			 __virtio16_to_cpu(vio_le(), vnet_hdr->csum_start) +
+			__virtio16_to_cpu(vio_le(), vnet_hdr->csum_offset) + 2);
+
+	if (__virtio16_to_cpu(vio_le(), vnet_hdr->hdr_len) > len)
+		return -EINVAL;
+
+	if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) {
+		switch (vnet_hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
+		case VIRTIO_NET_HDR_GSO_TCPV4:
+			gso_type = SKB_GSO_TCPV4;
+			break;
+		case VIRTIO_NET_HDR_GSO_TCPV6:
+			gso_type = SKB_GSO_TCPV6;
+			break;
+		case VIRTIO_NET_HDR_GSO_UDP:
+			gso_type = SKB_GSO_UDP;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		if (vnet_hdr->gso_type & VIRTIO_NET_HDR_GSO_ECN)
+			gso_type |= SKB_GSO_TCP_ECN;
+
+		if (vnet_hdr->gso_size == 0)
+			return -EINVAL;
+	}
+
+	vnet_hdr->gso_type = gso_type;	/* changes type, temporary storage */
+	return 0;
+}
+
+static int packet_snd_vnet_parse(struct msghdr *msg, size_t *len,
+				 struct virtio_net_hdr *vnet_hdr)
+{
+	int n;
+
+	if (*len < sizeof(*vnet_hdr))
+		return -EINVAL;
+	*len -= sizeof(*vnet_hdr);
+
+	n = copy_from_iter(vnet_hdr, sizeof(*vnet_hdr), &msg->msg_iter);
+	if (n != sizeof(*vnet_hdr))
+		return -EFAULT;
+
+	return __packet_snd_vnet_parse(vnet_hdr, *len);
+}
+
+static int packet_snd_vnet_gso(struct sk_buff *skb,
+			       struct virtio_net_hdr *vnet_hdr)
+{
+	if (vnet_hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
+		u16 s = __virtio16_to_cpu(vio_le(), vnet_hdr->csum_start);
+		u16 o = __virtio16_to_cpu(vio_le(), vnet_hdr->csum_offset);
+
+		if (!skb_partial_csum_set(skb, s, o))
+			return -EINVAL;
+	}
+
+	skb_shinfo(skb)->gso_size =
+		__virtio16_to_cpu(vio_le(), vnet_hdr->gso_size);
+	skb_shinfo(skb)->gso_type = vnet_hdr->gso_type;
+
+	/* Header must be checked, and gso_segs computed. */
+	skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
+	skb_shinfo(skb)->gso_segs = 0;
+	return 0;
+}
+
 static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 		void *frame, struct net_device *dev, int size_max,
 		__be16 proto, unsigned char *addr, int hlen)
@@ -2643,12 +2779,9 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	struct sockcm_cookie sockc;
 	struct virtio_net_hdr vnet_hdr = { 0 };
 	int offset = 0;
-	int vnet_hdr_len;
 	struct packet_sock *po = pkt_sk(sk);
-	unsigned short gso_type = 0;
 	int hlen, tlen;
 	int extra_len = 0;
-	ssize_t n;
 
 	/*
 	 *	Get and verify the address.
@@ -2686,53 +2819,9 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	if (sock->type == SOCK_RAW)
 		reserve = dev->hard_header_len;
 	if (po->has_vnet_hdr) {
-		vnet_hdr_len = sizeof(vnet_hdr);
-
-		err = -EINVAL;
-		if (len < vnet_hdr_len)
-			goto out_unlock;
-
-		len -= vnet_hdr_len;
-
-		err = -EFAULT;
-		n = copy_from_iter(&vnet_hdr, vnet_hdr_len, &msg->msg_iter);
-		if (n != vnet_hdr_len)
-			goto out_unlock;
-
-		if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) &&
-		    (__virtio16_to_cpu(vio_le(), vnet_hdr.csum_start) +
-		     __virtio16_to_cpu(vio_le(), vnet_hdr.csum_offset) + 2 >
-		      __virtio16_to_cpu(vio_le(), vnet_hdr.hdr_len)))
-			vnet_hdr.hdr_len = __cpu_to_virtio16(vio_le(),
-				 __virtio16_to_cpu(vio_le(), vnet_hdr.csum_start) +
-				__virtio16_to_cpu(vio_le(), vnet_hdr.csum_offset) + 2);
-
-		err = -EINVAL;
-		if (__virtio16_to_cpu(vio_le(), vnet_hdr.hdr_len) > len)
+		err = packet_snd_vnet_parse(msg, &len, &vnet_hdr);
+		if (err)
 			goto out_unlock;
-
-		if (vnet_hdr.gso_type != VIRTIO_NET_HDR_GSO_NONE) {
-			switch (vnet_hdr.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
-			case VIRTIO_NET_HDR_GSO_TCPV4:
-				gso_type = SKB_GSO_TCPV4;
-				break;
-			case VIRTIO_NET_HDR_GSO_TCPV6:
-				gso_type = SKB_GSO_TCPV6;
-				break;
-			case VIRTIO_NET_HDR_GSO_UDP:
-				gso_type = SKB_GSO_UDP;
-				break;
-			default:
-				goto out_unlock;
-			}
-
-			if (vnet_hdr.gso_type & VIRTIO_NET_HDR_GSO_ECN)
-				gso_type |= SKB_GSO_TCP_ECN;
-
-			if (vnet_hdr.gso_size == 0)
-				goto out_unlock;
-
-		}
 	}
 
 	if (unlikely(sock_flag(sk, SOCK_NOFCS))) {
@@ -2744,7 +2833,8 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	}
 
 	err = -EMSGSIZE;
-	if (!gso_type && (len > dev->mtu + reserve + VLAN_HLEN + extra_len))
+	if (!vnet_hdr.gso_type &&
+	    (len > dev->mtu + reserve + VLAN_HLEN + extra_len))
 		goto out_unlock;
 
 	err = -ENOBUFS;
@@ -2775,7 +2865,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 
 	sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags);
 
-	if (!gso_type && (len > dev->mtu + reserve + extra_len) &&
+	if (!vnet_hdr.gso_type && (len > dev->mtu + reserve + extra_len) &&
 	    !packet_extra_vlan_len_allowed(dev, skb)) {
 		err = -EMSGSIZE;
 		goto out_free;
@@ -2789,24 +2879,10 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	packet_pick_tx_queue(dev, skb);
 
 	if (po->has_vnet_hdr) {
-		if (vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
-			u16 s = __virtio16_to_cpu(vio_le(), vnet_hdr.csum_start);
-			u16 o = __virtio16_to_cpu(vio_le(), vnet_hdr.csum_offset);
-			if (!skb_partial_csum_set(skb, s, o)) {
-				err = -EINVAL;
-				goto out_free;
-			}
-		}
-
-		skb_shinfo(skb)->gso_size =
-			__virtio16_to_cpu(vio_le(), vnet_hdr.gso_size);
-		skb_shinfo(skb)->gso_type = gso_type;
-
-		/* Header must be checked, and gso_segs computed. */
-		skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
-		skb_shinfo(skb)->gso_segs = 0;
-
-		len += vnet_hdr_len;
+		err = packet_snd_vnet_gso(skb, &vnet_hdr);
+		if (err)
+			goto out_free;
+		len += sizeof(vnet_hdr);
 	}
 
 	skb_probe_transport_header(skb, reserve);
@@ -3177,51 +3253,10 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 		packet_rcv_has_room(pkt_sk(sk), NULL);
 
 	if (pkt_sk(sk)->has_vnet_hdr) {
-		struct virtio_net_hdr vnet_hdr = { 0 };
-
-		err = -EINVAL;
-		vnet_hdr_len = sizeof(vnet_hdr);
-		if (len < vnet_hdr_len)
-			goto out_free;
-
-		len -= vnet_hdr_len;
-
-		if (skb_is_gso(skb)) {
-			struct skb_shared_info *sinfo = skb_shinfo(skb);
-
-			/* This is a hint as to how much should be linear. */
-			vnet_hdr.hdr_len =
-				__cpu_to_virtio16(vio_le(), skb_headlen(skb));
-			vnet_hdr.gso_size =
-				__cpu_to_virtio16(vio_le(), sinfo->gso_size);
-			if (sinfo->gso_type & SKB_GSO_TCPV4)
-				vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
-			else if (sinfo->gso_type & SKB_GSO_TCPV6)
-				vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
-			else if (sinfo->gso_type & SKB_GSO_UDP)
-				vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_UDP;
-			else if (sinfo->gso_type & SKB_GSO_FCOE)
-				goto out_free;
-			else
-				BUG();
-			if (sinfo->gso_type & SKB_GSO_TCP_ECN)
-				vnet_hdr.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
-		} else
-			vnet_hdr.gso_type = VIRTIO_NET_HDR_GSO_NONE;
-
-		if (skb->ip_summed == CHECKSUM_PARTIAL) {
-			vnet_hdr.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
-			vnet_hdr.csum_start = __cpu_to_virtio16(vio_le(),
-					  skb_checksum_start_offset(skb));
-			vnet_hdr.csum_offset = __cpu_to_virtio16(vio_le(),
-							 skb->csum_offset);
-		} else if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
-			vnet_hdr.flags = VIRTIO_NET_HDR_F_DATA_VALID;
-		} /* else everything is zero */
-
-		err = memcpy_to_msg(msg, (void *)&vnet_hdr, vnet_hdr_len);
-		if (err < 0)
+		err = packet_rcv_vnet(msg, skb, &len);
+		if (err)
 			goto out_free;
+		vnet_hdr_len = sizeof(struct virtio_net_hdr);
 	}
 
 	/* You lose any data beyond the buffer you gave. If it worries
-- 
cgit v0.10.2


From 58d19b19cd99b438541eea4cdbf5c171900b25e5 Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 3 Feb 2016 18:02:15 -0500
Subject: packet: vnet_hdr support for tpacket_rcv

Support socket option PACKET_VNET_HDR together with PACKET_RX_RING.
When enabled, a struct virtio_net_hdr will precede the data in the
packet ring slots.

Verified with test program at
github.com/wdebruij/kerneltools/blob/master/tests/psock_rxring_vnet.c

  pkt: 1454269209.798420 len=5066
  vnet: gso_type=tcpv4 gso_size=1448 hlen=66 ecn=off
  csum: start=34 off=16
  eth: proto=0x800
  ip: src=<masked> dst=<masked> proto=6 len=5052

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index bd3de7b..b26df32 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2206,7 +2206,9 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		unsigned int maclen = skb_network_offset(skb);
 		netoff = TPACKET_ALIGN(po->tp_hdrlen +
 				       (maclen < 16 ? 16 : maclen)) +
-			po->tp_reserve;
+				       po->tp_reserve;
+		if (po->has_vnet_hdr)
+			netoff += sizeof(struct virtio_net_hdr);
 		macoff = netoff - maclen;
 	}
 	if (po->tp_version <= TPACKET_V2) {
@@ -2243,7 +2245,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 	h.raw = packet_current_rx_frame(po, skb,
 					TP_STATUS_KERNEL, (macoff+snaplen));
 	if (!h.raw)
-		goto ring_is_full;
+		goto drop_n_account;
 	if (po->tp_version <= TPACKET_V2) {
 		packet_increment_rx_head(po, &po->rx_ring);
 	/*
@@ -2262,6 +2264,14 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 	}
 	spin_unlock(&sk->sk_receive_queue.lock);
 
+	if (po->has_vnet_hdr) {
+		if (__packet_rcv_vnet(skb, h.raw + macoff -
+					   sizeof(struct virtio_net_hdr))) {
+			spin_lock(&sk->sk_receive_queue.lock);
+			goto drop_n_account;
+		}
+	}
+
 	skb_copy_bits(skb, 0, h.raw + macoff, snaplen);
 
 	if (!(ts_status = tpacket_get_timestamp(skb, &ts, po->tp_tstamp)))
@@ -2357,7 +2367,7 @@ drop:
 	kfree_skb(skb);
 	return 0;
 
-ring_is_full:
+drop_n_account:
 	po->stats.stats1.tp_drops++;
 	spin_unlock(&sk->sk_receive_queue.lock);
 
@@ -3587,7 +3597,8 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv
 		}
 		if (optlen < len)
 			return -EINVAL;
-		if (pkt_sk(sk)->has_vnet_hdr)
+		if (pkt_sk(sk)->has_vnet_hdr &&
+		    optname == PACKET_TX_RING)
 			return -EINVAL;
 		if (copy_from_user(&req_u.req, optval, len))
 			return -EFAULT;
-- 
cgit v0.10.2


From 8d39b4a6b83c141acaf0b5e8f90fd5e67721ff90 Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 3 Feb 2016 18:02:16 -0500
Subject: packet: parse tpacket header before skb alloc

GSO packet headers must be stored in the linear skb segment.
Move tpacket header parsing before sock_alloc_send_skb. The GSO
follow-on patch will later increase the skb linear argument to
sock_alloc_send_skb if needed for large packets.

The header parsing code does not require an allocated skb, so is
safe to move. Later pass to tpacket_fill_skb the computed data
start and length.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index b26df32..89377bf 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2494,14 +2494,13 @@ static int packet_snd_vnet_gso(struct sk_buff *skb,
 }
 
 static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
-		void *frame, struct net_device *dev, int size_max,
+		void *frame, struct net_device *dev, void *data, int tp_len,
 		__be16 proto, unsigned char *addr, int hlen)
 {
 	union tpacket_uhdr ph;
-	int to_write, offset, len, tp_len, nr_frags, len_max;
+	int to_write, offset, len, nr_frags, len_max;
 	struct socket *sock = po->sk.sk_socket;
 	struct page *page;
-	void *data;
 	int err;
 
 	ph.raw = frame;
@@ -2513,51 +2512,9 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 	sock_tx_timestamp(&po->sk, &skb_shinfo(skb)->tx_flags);
 	skb_shinfo(skb)->destructor_arg = ph.raw;
 
-	switch (po->tp_version) {
-	case TPACKET_V2:
-		tp_len = ph.h2->tp_len;
-		break;
-	default:
-		tp_len = ph.h1->tp_len;
-		break;
-	}
-	if (unlikely(tp_len > size_max)) {
-		pr_err("packet size is too long (%d > %d)\n", tp_len, size_max);
-		return -EMSGSIZE;
-	}
-
 	skb_reserve(skb, hlen);
 	skb_reset_network_header(skb);
 
-	if (unlikely(po->tp_tx_has_off)) {
-		int off_min, off_max, off;
-		off_min = po->tp_hdrlen - sizeof(struct sockaddr_ll);
-		off_max = po->tx_ring.frame_size - tp_len;
-		if (sock->type == SOCK_DGRAM) {
-			switch (po->tp_version) {
-			case TPACKET_V2:
-				off = ph.h2->tp_net;
-				break;
-			default:
-				off = ph.h1->tp_net;
-				break;
-			}
-		} else {
-			switch (po->tp_version) {
-			case TPACKET_V2:
-				off = ph.h2->tp_mac;
-				break;
-			default:
-				off = ph.h1->tp_mac;
-				break;
-			}
-		}
-		if (unlikely((off < off_min) || (off_max < off)))
-			return -EINVAL;
-		data = ph.raw + off;
-	} else {
-		data = ph.raw + po->tp_hdrlen - sizeof(struct sockaddr_ll);
-	}
 	to_write = tp_len;
 
 	if (sock->type == SOCK_DGRAM) {
@@ -2615,6 +2572,61 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 	return tp_len;
 }
 
+static int tpacket_parse_header(struct packet_sock *po, void *frame,
+				int size_max, void **data)
+{
+	union tpacket_uhdr ph;
+	int tp_len, off;
+
+	ph.raw = frame;
+
+	switch (po->tp_version) {
+	case TPACKET_V2:
+		tp_len = ph.h2->tp_len;
+		break;
+	default:
+		tp_len = ph.h1->tp_len;
+		break;
+	}
+	if (unlikely(tp_len > size_max)) {
+		pr_err("packet size is too long (%d > %d)\n", tp_len, size_max);
+		return -EMSGSIZE;
+	}
+
+	if (unlikely(po->tp_tx_has_off)) {
+		int off_min, off_max;
+
+		off_min = po->tp_hdrlen - sizeof(struct sockaddr_ll);
+		off_max = po->tx_ring.frame_size - tp_len;
+		if (po->sk.sk_type == SOCK_DGRAM) {
+			switch (po->tp_version) {
+			case TPACKET_V2:
+				off = ph.h2->tp_net;
+				break;
+			default:
+				off = ph.h1->tp_net;
+				break;
+			}
+		} else {
+			switch (po->tp_version) {
+			case TPACKET_V2:
+				off = ph.h2->tp_mac;
+				break;
+			default:
+				off = ph.h1->tp_mac;
+				break;
+			}
+		}
+		if (unlikely((off < off_min) || (off_max < off)))
+			return -EINVAL;
+	} else {
+		off = po->tp_hdrlen - sizeof(struct sockaddr_ll);
+	}
+
+	*data = frame + off;
+	return tp_len;
+}
+
 static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 {
 	struct sk_buff *skb;
@@ -2626,6 +2638,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	bool need_wait = !(msg->msg_flags & MSG_DONTWAIT);
 	int tp_len, size_max;
 	unsigned char *addr;
+	void *data;
 	int len_sum = 0;
 	int status = TP_STATUS_AVAILABLE;
 	int hlen, tlen;
@@ -2673,6 +2686,11 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 			continue;
 		}
 
+		skb = NULL;
+		tp_len = tpacket_parse_header(po, ph, size_max, &data);
+		if (tp_len < 0)
+			goto tpacket_error;
+
 		status = TP_STATUS_SEND_REQUEST;
 		hlen = LL_RESERVED_SPACE(dev);
 		tlen = dev->needed_tailroom;
@@ -2686,7 +2704,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 				err = len_sum;
 			goto out_status;
 		}
-		tp_len = tpacket_fill_skb(po, skb, ph, dev, size_max, proto,
+		tp_len = tpacket_fill_skb(po, skb, ph, dev, data, tp_len, proto,
 					  addr, hlen);
 		if (likely(tp_len >= 0) &&
 		    tp_len > dev->mtu + reserve &&
@@ -2694,6 +2712,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 			tp_len = -EMSGSIZE;
 
 		if (unlikely(tp_len < 0)) {
+tpacket_error:
 			if (po->tp_loss) {
 				__packet_set_status(po, ph,
 						TP_STATUS_AVAILABLE);
-- 
cgit v0.10.2


From 1d036d25e5609ba73fee6a88db01c306b140d512 Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 3 Feb 2016 18:02:17 -0500
Subject: packet: tpacket_snd gso and checksum offload

Support socket option PACKET_VNET_HDR together with PACKET_TX_RING.

When enabled, a struct virtio_net_hdr is expected to precede the data
in the ring. The vnet option must be set before the ring is created.

The implementation reuses the existing skb_copy_bits code that is used
when dev->hard_header_len is non-zero. Move this ll_header check to
before the skb alloc and combine it with a test for vnet_hdr->hdr_len.
Allocate and copy the max of the two.

Verified with test program at
github.com/wdebruij/kerneltools/blob/master/tests/psock_txring_vnet.c

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 89377bf..b7e7851 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2495,7 +2495,7 @@ static int packet_snd_vnet_gso(struct sk_buff *skb,
 
 static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 		void *frame, struct net_device *dev, void *data, int tp_len,
-		__be16 proto, unsigned char *addr, int hlen)
+		__be16 proto, unsigned char *addr, int hlen, int copylen)
 {
 	union tpacket_uhdr ph;
 	int to_write, offset, len, nr_frags, len_max;
@@ -2522,20 +2522,17 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 				NULL, tp_len);
 		if (unlikely(err < 0))
 			return -EINVAL;
-	} else if (dev->hard_header_len) {
-		if (ll_header_truncated(dev, tp_len))
-			return -EINVAL;
-
+	} else if (copylen) {
 		skb_push(skb, dev->hard_header_len);
-		err = skb_store_bits(skb, 0, data,
-				dev->hard_header_len);
+		skb_put(skb, copylen - dev->hard_header_len);
+		err = skb_store_bits(skb, 0, data, copylen);
 		if (unlikely(err))
 			return err;
 		if (!skb->protocol)
 			tpacket_set_protocol(dev, skb);
 
-		data += dev->hard_header_len;
-		to_write -= dev->hard_header_len;
+		data += copylen;
+		to_write -= copylen;
 	}
 
 	offset = offset_in_page(data);
@@ -2631,6 +2628,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 {
 	struct sk_buff *skb;
 	struct net_device *dev;
+	struct virtio_net_hdr *vnet_hdr = NULL;
 	__be16 proto;
 	int err, reserve = 0;
 	void *ph;
@@ -2641,7 +2639,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	void *data;
 	int len_sum = 0;
 	int status = TP_STATUS_AVAILABLE;
-	int hlen, tlen;
+	int hlen, tlen, copylen = 0;
 
 	mutex_lock(&po->pg_vec_lock);
 
@@ -2674,7 +2672,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	size_max = po->tx_ring.frame_size
 		- (po->tp_hdrlen - sizeof(struct sockaddr_ll));
 
-	if (size_max > dev->mtu + reserve + VLAN_HLEN)
+	if ((size_max > dev->mtu + reserve + VLAN_HLEN) && !po->has_vnet_hdr)
 		size_max = dev->mtu + reserve + VLAN_HLEN;
 
 	do {
@@ -2694,8 +2692,28 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 		status = TP_STATUS_SEND_REQUEST;
 		hlen = LL_RESERVED_SPACE(dev);
 		tlen = dev->needed_tailroom;
+		if (po->has_vnet_hdr) {
+			vnet_hdr = data;
+			data += sizeof(*vnet_hdr);
+			tp_len -= sizeof(*vnet_hdr);
+			if (tp_len < 0 ||
+			    __packet_snd_vnet_parse(vnet_hdr, tp_len)) {
+				tp_len = -EINVAL;
+				goto tpacket_error;
+			}
+			copylen = __virtio16_to_cpu(vio_le(),
+						    vnet_hdr->hdr_len);
+		}
+		if (dev->hard_header_len) {
+			if (ll_header_truncated(dev, tp_len)) {
+				tp_len = -EINVAL;
+				goto tpacket_error;
+			}
+			copylen = max_t(int, copylen, dev->hard_header_len);
+		}
 		skb = sock_alloc_send_skb(&po->sk,
-				hlen + tlen + sizeof(struct sockaddr_ll),
+				hlen + tlen + sizeof(struct sockaddr_ll) +
+				(copylen - dev->hard_header_len),
 				!need_wait, &err);
 
 		if (unlikely(skb == NULL)) {
@@ -2705,9 +2723,10 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 			goto out_status;
 		}
 		tp_len = tpacket_fill_skb(po, skb, ph, dev, data, tp_len, proto,
-					  addr, hlen);
+					  addr, hlen, copylen);
 		if (likely(tp_len >= 0) &&
 		    tp_len > dev->mtu + reserve &&
+		    !po->has_vnet_hdr &&
 		    !packet_extra_vlan_len_allowed(dev, skb))
 			tp_len = -EMSGSIZE;
 
@@ -2726,6 +2745,11 @@ tpacket_error:
 			}
 		}
 
+		if (po->has_vnet_hdr && packet_snd_vnet_gso(skb, vnet_hdr)) {
+			tp_len = -EINVAL;
+			goto tpacket_error;
+		}
+
 		packet_pick_tx_queue(dev, skb);
 
 		skb->destructor = tpacket_destruct_skb;
@@ -3616,9 +3640,6 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv
 		}
 		if (optlen < len)
 			return -EINVAL;
-		if (pkt_sk(sk)->has_vnet_hdr &&
-		    optname == PACKET_TX_RING)
-			return -EINVAL;
 		if (copy_from_user(&req_u.req, optval, len))
 			return -EFAULT;
 		return packet_set_ring(sk, &req_u, 0,
-- 
cgit v0.10.2


From 568b3ce7a8efdc23ad2f0b9ca8d82899cf68972d Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Wed, 10 Feb 2016 01:37:44 +0300
Subject: ravb: factor out register bit twiddling code

The  driver has often repeated pattern of reading a register,  AND'ing and/or
OR'ing some bits  and writing  the  value back. Factor the pattern out into
ravb_modify() -- this saves 260 bytes of code with ARM gcc 4.7.3.

While at it, update Cogent Embedded's copyrights.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h
index 9fbe92a..b2160d1 100644
--- a/drivers/net/ethernet/renesas/ravb.h
+++ b/drivers/net/ethernet/renesas/ravb.h
@@ -2,7 +2,7 @@
  *
  * Copyright (C) 2014-2015 Renesas Electronics Corporation
  * Copyright (C) 2015 Renesas Solutions Corp.
- * Copyright (C) 2015 Cogent Embedded, Inc. <source@cogentembedded.com>
+ * Copyright (C) 2015-2016 Cogent Embedded, Inc. <source@cogentembedded.com>
  *
  * Based on the SuperH Ethernet driver
  *
@@ -837,6 +837,8 @@ static inline void ravb_write(struct net_device *ndev, u32 data,
 	iowrite32(data, priv->addr + reg);
 }
 
+void ravb_modify(struct net_device *ndev, enum ravb_reg reg, u32 clear,
+		 u32 set);
 int ravb_wait(struct net_device *ndev, enum ravb_reg reg, u32 mask, u32 value);
 
 irqreturn_t ravb_ptp_interrupt(struct net_device *ndev);
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index ac43ed9..c936682 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -2,7 +2,7 @@
  *
  * Copyright (C) 2014-2015 Renesas Electronics Corporation
  * Copyright (C) 2015 Renesas Solutions Corp.
- * Copyright (C) 2015 Cogent Embedded, Inc. <source@cogentembedded.com>
+ * Copyright (C) 2015-2016 Cogent Embedded, Inc. <source@cogentembedded.com>
  *
  * Based on the SuperH Ethernet driver
  *
@@ -42,6 +42,12 @@
 		 NETIF_MSG_RX_ERR | \
 		 NETIF_MSG_TX_ERR)
 
+void ravb_modify(struct net_device *ndev, enum ravb_reg reg, u32 clear,
+		 u32 set)
+{
+	ravb_write(ndev, (ravb_read(ndev, reg) & ~clear) | set, reg);
+}
+
 int ravb_wait(struct net_device *ndev, enum ravb_reg reg, u32 mask, u32 value)
 {
 	int i;
@@ -59,8 +65,7 @@ static int ravb_config(struct net_device *ndev)
 	int error;
 
 	/* Set config mode */
-	ravb_write(ndev, (ravb_read(ndev, CCC) & ~CCC_OPC) | CCC_OPC_CONFIG,
-		   CCC);
+	ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG);
 	/* Check if the operating mode is changed to the config mode */
 	error = ravb_wait(ndev, CSR, CSR_OPS, CSR_OPS_CONFIG);
 	if (error)
@@ -72,13 +77,8 @@ static int ravb_config(struct net_device *ndev)
 static void ravb_set_duplex(struct net_device *ndev)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
-	u32 ecmr = ravb_read(ndev, ECMR);
 
-	if (priv->duplex)	/* Full */
-		ecmr |=  ECMR_DM;
-	else			/* Half */
-		ecmr &= ~ECMR_DM;
-	ravb_write(ndev, ecmr, ECMR);
+	ravb_modify(ndev, ECMR, ECMR_DM, priv->duplex ? ECMR_DM : 0);
 }
 
 static void ravb_set_rate(struct net_device *ndev)
@@ -131,13 +131,8 @@ static void ravb_mdio_ctrl(struct mdiobb_ctrl *ctrl, u32 mask, int set)
 {
 	struct ravb_private *priv = container_of(ctrl, struct ravb_private,
 						 mdiobb);
-	u32 pir = ravb_read(priv->ndev, PIR);
 
-	if (set)
-		pir |=  mask;
-	else
-		pir &= ~mask;
-	ravb_write(priv->ndev, pir, PIR);
+	ravb_modify(priv->ndev, PIR, mask, set ? mask : 0);
 }
 
 /* MDC pin control */
@@ -393,9 +388,9 @@ static int ravb_dmac_init(struct net_device *ndev)
 	ravb_ring_format(ndev, RAVB_NC);
 
 #if defined(__LITTLE_ENDIAN)
-	ravb_write(ndev, ravb_read(ndev, CCC) & ~CCC_BOC, CCC);
+	ravb_modify(ndev, CCC, CCC_BOC, 0);
 #else
-	ravb_write(ndev, ravb_read(ndev, CCC) | CCC_BOC, CCC);
+	ravb_modify(ndev, CCC, CCC_BOC, CCC_BOC);
 #endif
 
 	/* Set AVB RX */
@@ -418,8 +413,7 @@ static int ravb_dmac_init(struct net_device *ndev)
 	ravb_write(ndev, TIC_FTE0 | TIC_FTE1 | TIC_TFUE, TIC);
 
 	/* Setting the control will start the AVB-DMAC process. */
-	ravb_write(ndev, (ravb_read(ndev, CCC) & ~CCC_OPC) | CCC_OPC_OPERATION,
-		   CCC);
+	ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_OPERATION);
 
 	return 0;
 }
@@ -493,7 +487,7 @@ static void ravb_get_tx_tstamp(struct net_device *ndev)
 				break;
 			}
 		}
-		ravb_write(ndev, ravb_read(ndev, TCCR) | TCCR_TFR, TCCR);
+		ravb_modify(ndev, TCCR, TCCR_TFR, TCCR_TFR);
 	}
 }
 
@@ -613,13 +607,13 @@ static bool ravb_rx(struct net_device *ndev, int *quota, int q)
 static void ravb_rcv_snd_disable(struct net_device *ndev)
 {
 	/* Disable TX and RX */
-	ravb_write(ndev, ravb_read(ndev, ECMR) & ~(ECMR_RE | ECMR_TE), ECMR);
+	ravb_modify(ndev, ECMR, ECMR_RE | ECMR_TE, 0);
 }
 
 static void ravb_rcv_snd_enable(struct net_device *ndev)
 {
 	/* Enable TX and RX */
-	ravb_write(ndev, ravb_read(ndev, ECMR) | ECMR_RE | ECMR_TE, ECMR);
+	ravb_modify(ndev, ECMR, ECMR_RE | ECMR_TE, ECMR_RE | ECMR_TE);
 }
 
 /* function for waiting dma process finished */
@@ -812,8 +806,8 @@ static int ravb_poll(struct napi_struct *napi, int budget)
 
 	/* Re-enable RX/TX interrupts */
 	spin_lock_irqsave(&priv->lock, flags);
-	ravb_write(ndev, ravb_read(ndev, RIC0) | mask, RIC0);
-	ravb_write(ndev, ravb_read(ndev, TIC)  | mask,  TIC);
+	ravb_modify(ndev, RIC0, mask, mask);
+	ravb_modify(ndev, TIC,  mask, mask);
 	mmiowb();
 	spin_unlock_irqrestore(&priv->lock, flags);
 
@@ -852,8 +846,7 @@ static void ravb_adjust_link(struct net_device *ndev)
 			ravb_set_rate(ndev);
 		}
 		if (!priv->link) {
-			ravb_write(ndev, ravb_read(ndev, ECMR) & ~ECMR_TXF,
-				   ECMR);
+			ravb_modify(ndev, ECMR, ECMR_TXF, 0);
 			new_state = true;
 			priv->link = phydev->link;
 			if (priv->no_avb_link)
@@ -1393,7 +1386,7 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	desc--;
 	desc->die_dt = DT_FSTART;
 
-	ravb_write(ndev, ravb_read(ndev, TCCR) | (TCCR_TSRQ0 << q), TCCR);
+	ravb_modify(ndev, TCCR, TCCR_TSRQ0 << q, TCCR_TSRQ0 << q);
 
 	priv->cur_tx[q] += NUM_TX_DESC;
 	if (priv->cur_tx[q] - priv->dirty_tx[q] >
@@ -1468,15 +1461,10 @@ static void ravb_set_rx_mode(struct net_device *ndev)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	unsigned long flags;
-	u32 ecmr;
 
 	spin_lock_irqsave(&priv->lock, flags);
-	ecmr = ravb_read(ndev, ECMR);
-	if (ndev->flags & IFF_PROMISC)
-		ecmr |=  ECMR_PRM;
-	else
-		ecmr &= ~ECMR_PRM;
-	ravb_write(ndev, ecmr, ECMR);
+	ravb_modify(ndev, ECMR, ECMR_PRM,
+		    ndev->flags & IFF_PROMISC ? ECMR_PRM : 0);
 	mmiowb();
 	spin_unlock_irqrestore(&priv->lock, flags);
 }
@@ -1804,14 +1792,12 @@ static int ravb_probe(struct platform_device *pdev)
 
 	/* Set AVB config mode */
 	if (chip_id == RCAR_GEN2) {
-		ravb_write(ndev, (ravb_read(ndev, CCC) & ~CCC_OPC) |
-			   CCC_OPC_CONFIG, CCC);
+		ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG);
 		/* Set CSEL value */
-		ravb_write(ndev, (ravb_read(ndev, CCC) & ~CCC_CSEL) |
-			   CCC_CSEL_HPB, CCC);
+		ravb_modify(ndev, CCC, CCC_CSEL, CCC_CSEL_HPB);
 	} else {
-		ravb_write(ndev, (ravb_read(ndev, CCC) & ~CCC_OPC) |
-			   CCC_OPC_CONFIG | CCC_GAC | CCC_CSEL_HPB, CCC);
+		ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG |
+			    CCC_GAC | CCC_CSEL_HPB);
 	}
 
 	/* Set CSEL value */
@@ -1824,7 +1810,7 @@ static int ravb_probe(struct platform_device *pdev)
 		goto out_release;
 
 	/* Request GTI loading */
-	ravb_write(ndev, ravb_read(ndev, GCCR) | GCCR_LTI, GCCR);
+	ravb_modify(ndev, GCCR, GCCR_LTI, GCCR_LTI);
 
 	/* Allocate descriptor base address table */
 	priv->desc_bat_size = sizeof(struct ravb_desc) * DBAT_ENTRY_NUM;
diff --git a/drivers/net/ethernet/renesas/ravb_ptp.c b/drivers/net/ethernet/renesas/ravb_ptp.c
index 7a8ce92..57992cc 100644
--- a/drivers/net/ethernet/renesas/ravb_ptp.c
+++ b/drivers/net/ethernet/renesas/ravb_ptp.c
@@ -2,7 +2,7 @@
  *
  * Copyright (C) 2013-2015 Renesas Electronics Corporation
  * Copyright (C) 2015 Renesas Solutions Corp.
- * Copyright (C) 2015 Cogent Embedded, Inc. <source@cogentembedded.com>
+ * Copyright (C) 2015-2016 Cogent Embedded, Inc. <source@cogentembedded.com>
  *
  *  This program is free software; you can redistribute it and/or modify
  *  it under the terms of the GNU General Public License as published by
@@ -21,7 +21,7 @@ static int ravb_ptp_tcr_request(struct ravb_private *priv, u32 request)
 	if (error)
 		return error;
 
-	ravb_write(ndev, ravb_read(ndev, GCCR) | request, GCCR);
+	ravb_modify(ndev, GCCR, request, request);
 	return ravb_wait(ndev, GCCR, GCCR_TCR, GCCR_TCR_NOREQ);
 }
 
@@ -185,7 +185,6 @@ static int ravb_ptp_extts(struct ptp_clock_info *ptp,
 						 ptp.info);
 	struct net_device *ndev = priv->ndev;
 	unsigned long flags;
-	u32 gic;
 
 	if (req->index)
 		return -EINVAL;
@@ -195,12 +194,7 @@ static int ravb_ptp_extts(struct ptp_clock_info *ptp,
 	priv->ptp.extts[req->index] = on;
 
 	spin_lock_irqsave(&priv->lock, flags);
-	gic = ravb_read(ndev, GIC);
-	if (on)
-		gic |= GIC_PTCE;
-	else
-		gic &= ~GIC_PTCE;
-	ravb_write(ndev, gic, GIC);
+	ravb_modify(ndev, GIC, GIC_PTCE, on ? GIC_PTCE : 0);
 	mmiowb();
 	spin_unlock_irqrestore(&priv->lock, flags);
 
@@ -216,7 +210,6 @@ static int ravb_ptp_perout(struct ptp_clock_info *ptp,
 	struct ravb_ptp_perout *perout;
 	unsigned long flags;
 	int error = 0;
-	u32 gic;
 
 	if (req->index)
 		return -EINVAL;
@@ -248,9 +241,7 @@ static int ravb_ptp_perout(struct ptp_clock_info *ptp,
 		error = ravb_ptp_update_compare(priv, (u32)start_ns);
 		if (!error) {
 			/* Unmask interrupt */
-			gic = ravb_read(ndev, GIC);
-			gic |= GIC_PTME;
-			ravb_write(ndev, gic, GIC);
+			ravb_modify(ndev, GIC, GIC_PTME, GIC_PTME);
 		}
 	} else	{
 		spin_lock_irqsave(&priv->lock, flags);
@@ -259,9 +250,7 @@ static int ravb_ptp_perout(struct ptp_clock_info *ptp,
 		perout->period = 0;
 
 		/* Mask interrupt */
-		gic = ravb_read(ndev, GIC);
-		gic &= ~GIC_PTME;
-		ravb_write(ndev, gic, GIC);
+		ravb_modify(ndev, GIC, GIC_PTME, 0);
 	}
 	mmiowb();
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -331,7 +320,6 @@ void ravb_ptp_init(struct net_device *ndev, struct platform_device *pdev)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	unsigned long flags;
-	u32 gccr;
 
 	priv->ptp.info = ravb_ptp_info;
 
@@ -340,8 +328,7 @@ void ravb_ptp_init(struct net_device *ndev, struct platform_device *pdev)
 
 	spin_lock_irqsave(&priv->lock, flags);
 	ravb_wait(ndev, GCCR, GCCR_TCR, GCCR_TCR_NOREQ);
-	gccr = ravb_read(ndev, GCCR) & ~GCCR_TCSS;
-	ravb_write(ndev, gccr | GCCR_TCSS_ADJGPTP, GCCR);
+	ravb_modify(ndev, GCCR, GCCR_TCSS, GCCR_TCSS_ADJGPTP);
 	mmiowb();
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-- 
cgit v0.10.2


From b2b14d2f57df89fc2d57099ce13c5b513061d3a1 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Wed, 10 Feb 2016 01:38:28 +0300
Subject: sh_eth: factor out register bit twiddling code

The  driver has often repeated pattern of reading a register,  AND'ing and/or
OR'ing some bits  and writing  the  value back. Factor the pattern out into
sh_eth_modify() -- this saves  84 bytes of code with ARM gcc 4.7.3.

While at it, update Cogent Embedded's copyright.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index dfa9e59..0a150b2 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -3,7 +3,7 @@
  *  Copyright (C) 2014  Renesas Electronics Corporation
  *  Copyright (C) 2006-2012 Nobuhiro Iwamatsu
  *  Copyright (C) 2008-2014 Renesas Solutions Corp.
- *  Copyright (C) 2013-2014 Cogent Embedded, Inc.
+ *  Copyright (C) 2013-2016 Cogent Embedded, Inc.
  *  Copyright (C) 2014 Codethink Limited
  *
  *  This program is free software; you can redistribute it and/or modify it
@@ -428,6 +428,13 @@ static u32 sh_eth_read(struct net_device *ndev, int enum_index)
 	return ioread32(mdp->addr + offset);
 }
 
+static void sh_eth_modify(struct net_device *ndev, int enum_index, u32 clear,
+			  u32 set)
+{
+	sh_eth_write(ndev, (sh_eth_read(ndev, enum_index) & ~clear) | set,
+		     enum_index);
+}
+
 static bool sh_eth_is_gether(struct sh_eth_private *mdp)
 {
 	return mdp->reg_offset == sh_eth_offset_gigabit;
@@ -467,10 +474,7 @@ static void sh_eth_set_duplex(struct net_device *ndev)
 {
 	struct sh_eth_private *mdp = netdev_priv(ndev);
 
-	if (mdp->duplex) /* Full */
-		sh_eth_write(ndev, sh_eth_read(ndev, ECMR) | ECMR_DM, ECMR);
-	else		/* Half */
-		sh_eth_write(ndev, sh_eth_read(ndev, ECMR) & ~ECMR_DM, ECMR);
+	sh_eth_modify(ndev, ECMR, ECMR_DM, mdp->duplex ? ECMR_DM : 0);
 }
 
 static void sh_eth_chip_reset(struct net_device *ndev)
@@ -583,10 +587,10 @@ static void sh_eth_set_rate_r8a777x(struct net_device *ndev)
 
 	switch (mdp->speed) {
 	case 10: /* 10BASE */
-		sh_eth_write(ndev, sh_eth_read(ndev, ECMR) & ~ECMR_ELB, ECMR);
+		sh_eth_modify(ndev, ECMR, ECMR_ELB, 0);
 		break;
 	case 100:/* 100BASE */
-		sh_eth_write(ndev, sh_eth_read(ndev, ECMR) | ECMR_ELB, ECMR);
+		sh_eth_modify(ndev, ECMR, ECMR_ELB, ECMR_ELB);
 		break;
 	default:
 		break;
@@ -649,10 +653,10 @@ static void sh_eth_set_rate_sh7724(struct net_device *ndev)
 
 	switch (mdp->speed) {
 	case 10: /* 10BASE */
-		sh_eth_write(ndev, sh_eth_read(ndev, ECMR) & ~ECMR_RTM, ECMR);
+		sh_eth_modify(ndev, ECMR, ECMR_RTM, 0);
 		break;
 	case 100:/* 100BASE */
-		sh_eth_write(ndev, sh_eth_read(ndev, ECMR) | ECMR_RTM, ECMR);
+		sh_eth_modify(ndev, ECMR, ECMR_RTM, ECMR_RTM);
 		break;
 	default:
 		break;
@@ -924,8 +928,7 @@ static int sh_eth_reset(struct net_device *ndev)
 
 	if (sh_eth_is_gether(mdp) || sh_eth_is_rz_fast_ether(mdp)) {
 		sh_eth_write(ndev, EDSR_ENALL, EDSR);
-		sh_eth_write(ndev, sh_eth_read(ndev, EDMR) | EDMR_SRST_GETHER,
-			     EDMR);
+		sh_eth_modify(ndev, EDMR, EDMR_SRST_GETHER, EDMR_SRST_GETHER);
 
 		ret = sh_eth_check_reset(ndev);
 		if (ret)
@@ -949,11 +952,9 @@ static int sh_eth_reset(struct net_device *ndev)
 		if (mdp->cd->select_mii)
 			sh_eth_select_mii(ndev);
 	} else {
-		sh_eth_write(ndev, sh_eth_read(ndev, EDMR) | EDMR_SRST_ETHER,
-			     EDMR);
+		sh_eth_modify(ndev, EDMR, EDMR_SRST_ETHER, EDMR_SRST_ETHER);
 		mdelay(3);
-		sh_eth_write(ndev, sh_eth_read(ndev, EDMR) & ~EDMR_SRST_ETHER,
-			     EDMR);
+		sh_eth_modify(ndev, EDMR, EDMR_SRST_ETHER, 0);
 	}
 
 	return ret;
@@ -1285,7 +1286,7 @@ static int sh_eth_dev_init(struct net_device *ndev, bool start)
 	sh_eth_write(ndev, ndev->mtu + ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN,
 		     RFLR);
 
-	sh_eth_write(ndev, sh_eth_read(ndev, EESR), EESR);
+	sh_eth_modify(ndev, EESR, 0, 0);
 	if (start) {
 		mdp->irq_enabled = true;
 		sh_eth_write(ndev, mdp->cd->eesipr_value, EESIPR);
@@ -1532,15 +1533,13 @@ static int sh_eth_rx(struct net_device *ndev, u32 intr_status, int *quota)
 static void sh_eth_rcv_snd_disable(struct net_device *ndev)
 {
 	/* disable tx and rx */
-	sh_eth_write(ndev, sh_eth_read(ndev, ECMR) &
-		~(ECMR_RE | ECMR_TE), ECMR);
+	sh_eth_modify(ndev, ECMR, ECMR_RE | ECMR_TE, 0);
 }
 
 static void sh_eth_rcv_snd_enable(struct net_device *ndev)
 {
 	/* enable tx and rx */
-	sh_eth_write(ndev, sh_eth_read(ndev, ECMR) |
-		(ECMR_RE | ECMR_TE), ECMR);
+	sh_eth_modify(ndev, ECMR, ECMR_RE | ECMR_TE, ECMR_RE | ECMR_TE);
 }
 
 /* error control function */
@@ -1569,13 +1568,11 @@ static void sh_eth_error(struct net_device *ndev, u32 intr_status)
 				sh_eth_rcv_snd_disable(ndev);
 			} else {
 				/* Link Up */
-				sh_eth_write(ndev, sh_eth_read(ndev, EESIPR) &
-						   ~DMAC_M_ECI, EESIPR);
+				sh_eth_modify(ndev, EESIPR, DMAC_M_ECI, 0);
 				/* clear int */
-				sh_eth_write(ndev, sh_eth_read(ndev, ECSR),
-					     ECSR);
-				sh_eth_write(ndev, sh_eth_read(ndev, EESIPR) |
-						   DMAC_M_ECI, EESIPR);
+				sh_eth_modify(ndev, ECSR, 0, 0);
+				sh_eth_modify(ndev, EESIPR, DMAC_M_ECI,
+					      DMAC_M_ECI);
 				/* enable tx and rx */
 				sh_eth_rcv_snd_enable(ndev);
 			}
@@ -1765,9 +1762,7 @@ static void sh_eth_adjust_link(struct net_device *ndev)
 				mdp->cd->set_rate(ndev);
 		}
 		if (!mdp->link) {
-			sh_eth_write(ndev,
-				     sh_eth_read(ndev, ECMR) & ~ECMR_TXF,
-				     ECMR);
+			sh_eth_modify(ndev, ECMR, ECMR_TXF, 0);
 			new_state = 1;
 			mdp->link = phydev->link;
 			if (mdp->cd->no_psr || mdp->no_ether_link)
-- 
cgit v0.10.2


From aaa5672052cb0f18a5ab0ab19e9ef71c50e018a8 Mon Sep 17 00:00:00 2001
From: Jean Sacren <sakiwit@gmail.com>
Date: Tue, 9 Feb 2016 20:47:17 -0700
Subject: sxgbe: remove unused code

Remove the unused code of sxgbe_xpcs.

Reported-by: Julia Lawall <julia.lawall@lip6.fr>
Suggested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jean Sacren <sakiwit@gmail.com>
Cc: Byungho An <bh74.an@samsung.com>
Cc: Girish K S <ks.giri@samsung.com>
Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1601191918470.2531@hadrien
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/samsung/sxgbe/Makefile b/drivers/net/ethernet/samsung/sxgbe/Makefile
index dcc80b9..31e9685 100644
--- a/drivers/net/ethernet/samsung/sxgbe/Makefile
+++ b/drivers/net/ethernet/samsung/sxgbe/Makefile
@@ -1,4 +1,4 @@
 obj-$(CONFIG_SXGBE_ETH) += samsung-sxgbe.o
 samsung-sxgbe-objs:= sxgbe_platform.o sxgbe_main.o sxgbe_desc.o \
 		sxgbe_dma.o sxgbe_core.o sxgbe_mtl.o  sxgbe_mdio.o \
-		sxgbe_ethtool.o sxgbe_xpcs.o $(samsung-sxgbe-y)
+		sxgbe_ethtool.o $(samsung-sxgbe-y)
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_xpcs.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_xpcs.c
deleted file mode 100644
index 51c3219..0000000
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_xpcs.c
+++ /dev/null
@@ -1,91 +0,0 @@
-/* 10G controller driver for Samsung SoCs
- *
- * Copyright (C) 2013 Samsung Electronics Co., Ltd.
- *		http://www.samsung.com
- *
- * Author: Siva Reddy Kallam <siva.kallam@samsung.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/bitops.h>
-#include <linux/kernel.h>
-#include <linux/netdevice.h>
-#include <linux/phy.h>
-#include "sxgbe_common.h"
-#include "sxgbe_xpcs.h"
-
-static int sxgbe_xpcs_read(struct net_device *ndev, unsigned int reg)
-{
-	u32 value;
-	struct sxgbe_priv_data *priv = netdev_priv(ndev);
-
-	value = readl(priv->ioaddr + XPCS_OFFSET + reg);
-
-	return value;
-}
-
-static int sxgbe_xpcs_write(struct net_device *ndev, int reg, int data)
-{
-	struct sxgbe_priv_data *priv = netdev_priv(ndev);
-
-	writel(data, priv->ioaddr + XPCS_OFFSET + reg);
-
-	return 0;
-}
-
-int sxgbe_xpcs_init(struct net_device *ndev)
-{
-	u32 value;
-
-	value = sxgbe_xpcs_read(ndev, SR_PCS_MMD_CONTROL1);
-	/* 10G XAUI mode */
-	sxgbe_xpcs_write(ndev, SR_PCS_CONTROL2, XPCS_TYPE_SEL_X);
-	sxgbe_xpcs_write(ndev, VR_PCS_MMD_XAUI_MODE_CONTROL, XPCS_XAUI_MODE);
-	sxgbe_xpcs_write(ndev, VR_PCS_MMD_XAUI_MODE_CONTROL, value | BIT(13));
-	sxgbe_xpcs_write(ndev, SR_PCS_MMD_CONTROL1, value | BIT(11));
-
-	do {
-		value = sxgbe_xpcs_read(ndev, VR_PCS_MMD_DIGITAL_STATUS);
-	} while ((value & XPCS_QSEQ_STATE_MPLLOFF) == XPCS_QSEQ_STATE_STABLE);
-
-	value = sxgbe_xpcs_read(ndev, SR_PCS_MMD_CONTROL1);
-	sxgbe_xpcs_write(ndev, SR_PCS_MMD_CONTROL1, value & ~BIT(11));
-
-	do {
-		value = sxgbe_xpcs_read(ndev, VR_PCS_MMD_DIGITAL_STATUS);
-	} while ((value & XPCS_QSEQ_STATE_MPLLOFF) != XPCS_QSEQ_STATE_STABLE);
-
-	return 0;
-}
-
-int sxgbe_xpcs_init_1G(struct net_device *ndev)
-{
-	int value;
-
-	/* 10GBASE-X PCS (1G) mode */
-	sxgbe_xpcs_write(ndev, SR_PCS_CONTROL2, XPCS_TYPE_SEL_X);
-	sxgbe_xpcs_write(ndev, VR_PCS_MMD_XAUI_MODE_CONTROL, XPCS_XAUI_MODE);
-	value = sxgbe_xpcs_read(ndev, SR_PCS_MMD_CONTROL1);
-	sxgbe_xpcs_write(ndev, SR_PCS_MMD_CONTROL1, value & ~BIT(13));
-
-	value = sxgbe_xpcs_read(ndev, SR_MII_MMD_CONTROL);
-	sxgbe_xpcs_write(ndev, SR_MII_MMD_CONTROL, value | BIT(6));
-	sxgbe_xpcs_write(ndev, SR_MII_MMD_CONTROL, value & ~BIT(13));
-	value = sxgbe_xpcs_read(ndev, SR_PCS_MMD_CONTROL1);
-	sxgbe_xpcs_write(ndev, SR_PCS_MMD_CONTROL1, value | BIT(11));
-
-	do {
-		value = sxgbe_xpcs_read(ndev, VR_PCS_MMD_DIGITAL_STATUS);
-	} while ((value & XPCS_QSEQ_STATE_MPLLOFF) != XPCS_QSEQ_STATE_STABLE);
-
-	value = sxgbe_xpcs_read(ndev, SR_PCS_MMD_CONTROL1);
-	sxgbe_xpcs_write(ndev, SR_PCS_MMD_CONTROL1, value & ~BIT(11));
-
-	/* Auto Negotiation cluase 37 enable */
-	value = sxgbe_xpcs_read(ndev, SR_MII_MMD_CONTROL);
-	sxgbe_xpcs_write(ndev, SR_MII_MMD_CONTROL, value | BIT(12));
-
-	return 0;
-}
diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_xpcs.h b/drivers/net/ethernet/samsung/sxgbe/sxgbe_xpcs.h
deleted file mode 100644
index 6b26a50..0000000
--- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_xpcs.h
+++ /dev/null
@@ -1,38 +0,0 @@
-/* 10G controller driver for Samsung SoCs
- *
- * Copyright (C) 2013 Samsung Electronics Co., Ltd.
- *		http://www.samsung.com
- *
- * Author: Byungho An <bh74.an@samsung.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#ifndef __SXGBE_XPCS_H__
-#define __SXGBE_XPCS_H__
-
-/* XPCS Registers */
-#define XPCS_OFFSET			0x1A060000
-#define SR_PCS_MMD_CONTROL1		0x030000
-#define SR_PCS_CONTROL2			0x030007
-#define VR_PCS_MMD_XAUI_MODE_CONTROL	0x038004
-#define VR_PCS_MMD_DIGITAL_STATUS	0x038010
-#define SR_MII_MMD_CONTROL		0x1F0000
-#define SR_MII_MMD_AN_ADV		0x1F0004
-#define SR_MII_MMD_AN_LINK_PARTNER_BA	0x1F0005
-#define VR_MII_MMD_AN_CONTROL		0x1F8001
-#define VR_MII_MMD_AN_INT_STATUS	0x1F8002
-
-#define XPCS_QSEQ_STATE_STABLE		0x10
-#define XPCS_QSEQ_STATE_MPLLOFF		0x1c
-#define XPCS_TYPE_SEL_R			0x00
-#define XPCS_TYPE_SEL_X			0x01
-#define XPCS_TYPE_SEL_W			0x02
-#define XPCS_XAUI_MODE			0x00
-#define XPCS_RXAUI_MODE			0x01
-
-int sxgbe_xpcs_init(struct net_device *ndev);
-int sxgbe_xpcs_init_1G(struct net_device *ndev);
-
-#endif /* __SXGBE_XPCS_H__ */
-- 
cgit v0.10.2


From 140ed8e87ca8f4875c2b146cdb2cdbf0c9ac6080 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Tue, 5 Jan 2016 12:06:26 +0100
Subject: batman-adv: Drop reference to netdevice on last reference

The references to the network device should be dropped inside the release
function for batadv_hard_iface similar to what is done with the batman-adv
internal datastructures.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index db90022..b17e272 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -47,13 +47,16 @@
 #include "sysfs.h"
 #include "translation-table.h"
 
-void batadv_hardif_free_rcu(struct rcu_head *rcu)
+/**
+ * batadv_hardif_release - release hard interface from lists and queue for
+ *  free after rcu grace period
+ * @hard_iface: the hard interface to free
+ */
+void batadv_hardif_release(struct batadv_hard_iface *hard_iface)
 {
-	struct batadv_hard_iface *hard_iface;
-
-	hard_iface = container_of(rcu, struct batadv_hard_iface, rcu);
 	dev_put(hard_iface->net_dev);
-	kfree(hard_iface);
+
+	kfree_rcu(hard_iface, rcu);
 }
 
 struct batadv_hard_iface *
diff --git a/net/batman-adv/hard-interface.h b/net/batman-adv/hard-interface.h
index 4d6b5e1..9eb08ad 100644
--- a/net/batman-adv/hard-interface.h
+++ b/net/batman-adv/hard-interface.h
@@ -61,18 +61,18 @@ void batadv_hardif_disable_interface(struct batadv_hard_iface *hard_iface,
 void batadv_hardif_remove_interfaces(void);
 int batadv_hardif_min_mtu(struct net_device *soft_iface);
 void batadv_update_min_mtu(struct net_device *soft_iface);
-void batadv_hardif_free_rcu(struct rcu_head *rcu);
+void batadv_hardif_release(struct batadv_hard_iface *hard_iface);
 
 /**
  * batadv_hardif_free_ref - decrement the hard interface refcounter and
- *  possibly free it
+ *  possibly release it
  * @hard_iface: the hard interface to free
  */
 static inline void
 batadv_hardif_free_ref(struct batadv_hard_iface *hard_iface)
 {
 	if (atomic_dec_and_test(&hard_iface->refcount))
-		call_rcu(&hard_iface->rcu, batadv_hardif_free_rcu);
+		batadv_hardif_release(hard_iface);
 }
 
 static inline struct batadv_hard_iface *
-- 
cgit v0.10.2


From 81f02683504dc5590204c3fa059eac86fbfa0927 Mon Sep 17 00:00:00 2001
From: Simon Wunderlich <simon@open-mesh.com>
Date: Mon, 23 Nov 2015 19:57:22 +0100
Subject: batman-adv: add seqno maximum age and protection start flag
 parameters

To allow future use of the window protected function with different
maximum sequence numbers, add a parameter to set this value which
was previously hardcoded. Another parameter added for future use is a
flag to return whether the protection window has started.

While at it, also fix the kerneldoc.

Signed-off-by: Simon Wunderlich <simon@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 3266bcb..23ce90e 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1315,7 +1315,8 @@ batadv_iv_ogm_update_seqnos(const struct ethhdr *ethhdr,
 	/* signalize caller that the packet is to be dropped. */
 	if (!hlist_empty(&orig_node->neigh_list) &&
 	    batadv_window_protected(bat_priv, seq_diff,
-				    &orig_ifinfo->batman_seqno_reset)) {
+				    BATADV_TQ_LOCAL_WINDOW_SIZE,
+				    &orig_ifinfo->batman_seqno_reset, NULL)) {
 		ret = BATADV_PROTECTED;
 		goto out;
 	}
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index a7dc41a..32dfc9e 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -35,6 +35,9 @@
 /* Time To Live of broadcast messages */
 #define BATADV_TTL 50
 
+/* maximum sequence number age of broadcast messages */
+#define BATADV_BCAST_MAX_AGE 64
+
 /* purge originators after time in seconds if no valid packet comes in
  * -> TODO: check influence on BATADV_TQ_LOCAL_WINDOW_SIZE
  */
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index f4b60b1..1c8b35d 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -146,23 +146,29 @@ out:
  * @bat_priv: the bat priv with all the soft interface information
  * @seq_num_diff: difference between the current/received sequence number and
  *  the last sequence number
+ * @seq_old_max_diff: maximum age of sequence number not considered as restart
  * @last_reset: jiffies timestamp of the last reset, will be updated when reset
  *  is detected
+ * @protection_started: is set to true if the protection window was started,
+ *   doesn't change otherwise.
  *
  * Return:
  *  0 if the packet is to be accepted.
  *  1 if the packet is to be ignored.
  */
 int batadv_window_protected(struct batadv_priv *bat_priv, s32 seq_num_diff,
-			    unsigned long *last_reset)
+			    s32 seq_old_max_diff, unsigned long *last_reset,
+			    bool *protection_started)
 {
-	if (seq_num_diff <= -BATADV_TQ_LOCAL_WINDOW_SIZE ||
+	if (seq_num_diff <= -seq_old_max_diff ||
 	    seq_num_diff >= BATADV_EXPECTED_SEQNO_RANGE) {
 		if (!batadv_has_timed_out(*last_reset,
 					  BATADV_RESET_PROTECTION_MS))
 			return 1;
 
 		*last_reset = jiffies;
+		if (protection_started)
+			*protection_started = true;
 		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
 			   "old packet received, start protection\n");
 	}
@@ -1073,7 +1079,8 @@ int batadv_recv_bcast_packet(struct sk_buff *skb,
 
 	/* check whether the packet is old and the host just restarted. */
 	if (batadv_window_protected(bat_priv, seq_diff,
-				    &orig_node->bcast_seqno_reset))
+				    BATADV_BCAST_MAX_AGE,
+				    &orig_node->bcast_seqno_reset, NULL))
 		goto spin_unlock;
 
 	/* mark broadcast in flood history, update window position
diff --git a/net/batman-adv/routing.h b/net/batman-adv/routing.h
index c776e96..02a5caa 100644
--- a/net/batman-adv/routing.h
+++ b/net/batman-adv/routing.h
@@ -52,6 +52,7 @@ batadv_find_router(struct batadv_priv *bat_priv,
 		   struct batadv_orig_node *orig_node,
 		   struct batadv_hard_iface *recv_if);
 int batadv_window_protected(struct batadv_priv *bat_priv, s32 seq_num_diff,
-			    unsigned long *last_reset);
+			    s32 seq_old_max_diff, unsigned long *last_reset,
+			    bool *protection_started);
 
 #endif /* _NET_BATMAN_ADV_ROUTING_H_ */
-- 
cgit v0.10.2


From dded0692247c76f892adee3ba6878ff5f394e210 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 20 Dec 2015 09:04:03 +0100
Subject: batman-adv: Add lockdep assert for container_list_lock

The batadv_tvlv_container* functions state in their kernel-doc that they
require tvlv.container_list_lock. Add an assert to automatically detect
when this might have been ignored by the caller.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 5f319fd..a9b4f75 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -697,6 +697,8 @@ static struct batadv_tvlv_container
 {
 	struct batadv_tvlv_container *tvlv_tmp, *tvlv = NULL;
 
+	lockdep_assert_held(&bat_priv->tvlv.container_list_lock);
+
 	hlist_for_each_entry(tvlv_tmp, &bat_priv->tvlv.container_list, list) {
 		if (tvlv_tmp->tvlv_hdr.type != type)
 			continue;
@@ -729,6 +731,8 @@ static u16 batadv_tvlv_container_list_size(struct batadv_priv *bat_priv)
 	struct batadv_tvlv_container *tvlv;
 	u16 tvlv_len = 0;
 
+	lockdep_assert_held(&bat_priv->tvlv.container_list_lock);
+
 	hlist_for_each_entry(tvlv, &bat_priv->tvlv.container_list, list) {
 		tvlv_len += sizeof(struct batadv_tvlv_hdr);
 		tvlv_len += ntohs(tvlv->tvlv_hdr.len);
-- 
cgit v0.10.2


From 90f564dff43c9c62d68568c550c011825f14a899 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:40 +0100
Subject: batman-adv: Convert batadv_hardif_neigh_node to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index d4a30db..7710595 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -23,6 +23,7 @@
 #include <linux/fs.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
@@ -196,11 +197,15 @@ void batadv_neigh_ifinfo_free_ref(struct batadv_neigh_ifinfo *neigh_ifinfo)
 /**
  * batadv_hardif_neigh_release - release hardif neigh node from lists and
  *  queue for free after rcu grace period
- * @hardif_neigh: hardif neigh neighbor to free
+ * @ref: kref pointer of the neigh_node
  */
-static void
-batadv_hardif_neigh_release(struct batadv_hardif_neigh_node *hardif_neigh)
+static void batadv_hardif_neigh_release(struct kref *ref)
 {
+	struct batadv_hardif_neigh_node *hardif_neigh;
+
+	hardif_neigh = container_of(ref, struct batadv_hardif_neigh_node,
+				    refcount);
+
 	spin_lock_bh(&hardif_neigh->if_incoming->neigh_list_lock);
 	hlist_del_init_rcu(&hardif_neigh->list);
 	spin_unlock_bh(&hardif_neigh->if_incoming->neigh_list_lock);
@@ -216,8 +221,7 @@ batadv_hardif_neigh_release(struct batadv_hardif_neigh_node *hardif_neigh)
  */
 void batadv_hardif_neigh_free_ref(struct batadv_hardif_neigh_node *hardif_neigh)
 {
-	if (atomic_dec_and_test(&hardif_neigh->refcount))
-		batadv_hardif_neigh_release(hardif_neigh);
+	kref_put(&hardif_neigh->refcount, batadv_hardif_neigh_release);
 }
 
 /**
@@ -529,7 +533,7 @@ batadv_hardif_neigh_create(struct batadv_hard_iface *hard_iface,
 	hardif_neigh->if_incoming = hard_iface;
 	hardif_neigh->last_seen = jiffies;
 
-	atomic_set(&hardif_neigh->refcount, 1);
+	kref_init(&hardif_neigh->refcount);
 
 	if (bat_priv->bat_algo_ops->bat_hardif_neigh_init)
 		bat_priv->bat_algo_ops->bat_hardif_neigh_init(hardif_neigh);
@@ -584,7 +588,7 @@ batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface,
 		if (!batadv_compare_eth(tmp_hardif_neigh->addr, neigh_addr))
 			continue;
 
-		if (!atomic_inc_not_zero(&tmp_hardif_neigh->refcount))
+		if (!kref_get_unless_zero(&tmp_hardif_neigh->refcount))
 			continue;
 
 		hardif_neigh = tmp_hardif_neigh;
@@ -648,7 +652,7 @@ batadv_neigh_node_new(struct batadv_orig_node *orig_node,
 	spin_unlock_bh(&orig_node->neigh_list_lock);
 
 	/* increment unique neighbor refcount */
-	atomic_inc(&hardif_neigh->refcount);
+	kref_get(&hardif_neigh->refcount);
 
 	batadv_dbg(BATADV_DBG_BATMAN, orig_node->bat_priv,
 		   "Creating new neighbor %pM for orig_node %pM on interface %s\n",
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 8974bc0..ace7900 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -25,6 +25,7 @@
 #include <linux/bitops.h>
 #include <linux/compiler.h>
 #include <linux/if_ether.h>
+#include <linux/kref.h>
 #include <linux/netdevice.h>
 #include <linux/sched.h> /* for linux/wait.h */
 #include <linux/spinlock.h>
@@ -359,7 +360,7 @@ struct batadv_hardif_neigh_node {
 	u8 addr[ETH_ALEN];
 	struct batadv_hard_iface *if_incoming;
 	unsigned long last_seen;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From e7aed321b8c07d2019b003c3f51a067f2b4f78ab Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:41 +0100
Subject: batman-adv: Convert batadv_gw_node to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 5950974..1f3847c 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -28,6 +28,7 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/netdevice.h>
 #include <linux/rculist.h>
@@ -59,12 +60,29 @@
  */
 #define BATADV_DHCP_CHADDR_OFFSET	28
 
+/**
+ * batadv_gw_node_release - release gw_node from lists and queue for free after
+ *  rcu grace period
+ * @ref: kref pointer of the gw_node
+ */
+static void batadv_gw_node_release(struct kref *ref)
+{
+	struct batadv_gw_node *gw_node;
+
+	gw_node = container_of(ref, struct batadv_gw_node, refcount);
+
+	batadv_orig_node_free_ref(gw_node->orig_node);
+	kfree_rcu(gw_node, rcu);
+}
+
+/**
+ * batadv_gw_node_free_ref - decrement the gw_node refcounter and possibly
+ *  release it
+ * @gw_node: gateway node to free
+ */
 static void batadv_gw_node_free_ref(struct batadv_gw_node *gw_node)
 {
-	if (atomic_dec_and_test(&gw_node->refcount)) {
-		batadv_orig_node_free_ref(gw_node->orig_node);
-		kfree_rcu(gw_node, rcu);
-	}
+	kref_put(&gw_node->refcount, batadv_gw_node_release);
 }
 
 static struct batadv_gw_node *
@@ -77,7 +95,7 @@ batadv_gw_get_selected_gw_node(struct batadv_priv *bat_priv)
 	if (!gw_node)
 		goto out;
 
-	if (!atomic_inc_not_zero(&gw_node->refcount))
+	if (!kref_get_unless_zero(&gw_node->refcount))
 		gw_node = NULL;
 
 out:
@@ -118,7 +136,7 @@ static void batadv_gw_select(struct batadv_priv *bat_priv,
 
 	spin_lock_bh(&bat_priv->gw.list_lock);
 
-	if (new_gw_node && !atomic_inc_not_zero(&new_gw_node->refcount))
+	if (new_gw_node && !kref_get_unless_zero(&new_gw_node->refcount))
 		new_gw_node = NULL;
 
 	curr_gw_node = rcu_dereference_protected(bat_priv->gw.curr_gw, 1);
@@ -170,7 +188,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 		if (!router_ifinfo)
 			goto next;
 
-		if (!atomic_inc_not_zero(&gw_node->refcount))
+		if (!kref_get_unless_zero(&gw_node->refcount))
 			goto next;
 
 		tq_avg = router_ifinfo->bat_iv.tq_avg;
@@ -188,7 +206,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 				if (curr_gw)
 					batadv_gw_node_free_ref(curr_gw);
 				curr_gw = gw_node;
-				atomic_inc(&curr_gw->refcount);
+				kref_get(&curr_gw->refcount);
 			}
 			break;
 
@@ -203,7 +221,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 				if (curr_gw)
 					batadv_gw_node_free_ref(curr_gw);
 				curr_gw = gw_node;
-				atomic_inc(&curr_gw->refcount);
+				kref_get(&curr_gw->refcount);
 			}
 			break;
 		}
@@ -436,7 +454,7 @@ static void batadv_gw_node_add(struct batadv_priv *bat_priv,
 	gw_node->orig_node = orig_node;
 	gw_node->bandwidth_down = ntohl(gateway->bandwidth_down);
 	gw_node->bandwidth_up = ntohl(gateway->bandwidth_up);
-	atomic_set(&gw_node->refcount, 1);
+	kref_init(&gw_node->refcount);
 
 	spin_lock_bh(&bat_priv->gw.list_lock);
 	hlist_add_head_rcu(&gw_node->list, &bat_priv->gw.list);
@@ -469,7 +487,7 @@ batadv_gw_node_get(struct batadv_priv *bat_priv,
 		if (gw_node_tmp->orig_node != orig_node)
 			continue;
 
-		if (!atomic_inc_not_zero(&gw_node_tmp->refcount))
+		if (!kref_get_unless_zero(&gw_node_tmp->refcount))
 			continue;
 
 		gw_node = gw_node_tmp;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index ace7900..06c3888 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -342,7 +342,7 @@ struct batadv_gw_node {
 	struct batadv_orig_node *orig_node;
 	u32 bandwidth_down;
 	u32 bandwidth_up;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 6be4d30c18270fe275084c7cd30884535424266e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:42 +0100
Subject: batman-adv: Convert batadv_softif_vlan to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index 4bf35b8..d4490ff 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -30,6 +30,7 @@
 #include <linux/if_vlan.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
@@ -478,8 +479,26 @@ out:
 }
 
 /**
+ * batadv_softif_vlan_release - release vlan from lists and queue for free after
+ *  rcu grace period
+ * @ref: kref pointer of the vlan object
+ */
+static void batadv_softif_vlan_release(struct kref *ref)
+{
+	struct batadv_softif_vlan *vlan;
+
+	vlan = container_of(ref, struct batadv_softif_vlan, refcount);
+
+	spin_lock_bh(&vlan->bat_priv->softif_vlan_list_lock);
+	hlist_del_rcu(&vlan->list);
+	spin_unlock_bh(&vlan->bat_priv->softif_vlan_list_lock);
+
+	kfree_rcu(vlan, rcu);
+}
+
+/**
  * batadv_softif_vlan_free_ref - decrease the vlan object refcounter and
- *  possibly free it
+ *  possibly release it
  * @vlan: the vlan object to release
  */
 void batadv_softif_vlan_free_ref(struct batadv_softif_vlan *vlan)
@@ -487,13 +506,7 @@ void batadv_softif_vlan_free_ref(struct batadv_softif_vlan *vlan)
 	if (!vlan)
 		return;
 
-	if (atomic_dec_and_test(&vlan->refcount)) {
-		spin_lock_bh(&vlan->bat_priv->softif_vlan_list_lock);
-		hlist_del_rcu(&vlan->list);
-		spin_unlock_bh(&vlan->bat_priv->softif_vlan_list_lock);
-
-		kfree_rcu(vlan, rcu);
-	}
+	kref_put(&vlan->refcount, batadv_softif_vlan_release);
 }
 
 /**
@@ -514,7 +527,7 @@ struct batadv_softif_vlan *batadv_softif_vlan_get(struct batadv_priv *bat_priv,
 		if (vlan_tmp->vid != vid)
 			continue;
 
-		if (!atomic_inc_not_zero(&vlan_tmp->refcount))
+		if (!kref_get_unless_zero(&vlan_tmp->refcount))
 			continue;
 
 		vlan = vlan_tmp;
@@ -549,7 +562,7 @@ int batadv_softif_create_vlan(struct batadv_priv *bat_priv, unsigned short vid)
 
 	vlan->bat_priv = bat_priv;
 	vlan->vid = vid;
-	atomic_set(&vlan->refcount, 1);
+	kref_init(&vlan->refcount);
 
 	atomic_set(&vlan->ap_isolation, 0);
 
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index 964fc59..ab4382b 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -25,6 +25,7 @@
 #include <linux/fs.h>
 #include <linux/if.h>
 #include <linux/if_vlan.h>
+#include <linux/kref.h>
 #include <linux/kernel.h>
 #include <linux/netdevice.h>
 #include <linux/printk.h>
@@ -97,7 +98,7 @@ batadv_kobj_to_vlan(struct batadv_priv *bat_priv, struct kobject *obj)
 		if (vlan_tmp->kobj != obj)
 			continue;
 
-		if (!atomic_inc_not_zero(&vlan_tmp->refcount))
+		if (!kref_get_unless_zero(&vlan_tmp->refcount))
 			continue;
 
 		vlan = vlan_tmp;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 06c3888..868f6b5 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -746,7 +746,7 @@ struct batadv_softif_vlan {
 	atomic_t ap_isolation;		/* boolean */
 	struct batadv_vlan_tt tt;
 	struct hlist_node list;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 06e56ded867d1035b877aba28189736f41e85ab0 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:43 +0100
Subject: batman-adv: Convert batadv_bla_backbone_gw to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 7791609..41116e4 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -31,6 +31,7 @@
 #include <linux/jhash.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
@@ -143,14 +144,29 @@ static int batadv_compare_claim(const struct hlist_node *node,
 }
 
 /**
- * batadv_compare_backbone_gw - free backbone gw
+ * batadv_backbone_gw_release - release backbone gw from lists and queue for
+ *  free after rcu grace period
+ * @ref: kref pointer of the backbone gw
+ */
+static void batadv_backbone_gw_release(struct kref *ref)
+{
+	struct batadv_bla_backbone_gw *backbone_gw;
+
+	backbone_gw = container_of(ref, struct batadv_bla_backbone_gw,
+				   refcount);
+
+	kfree_rcu(backbone_gw, rcu);
+}
+
+/**
+ * batadv_backbone_gw_free_ref - decrement the backbone gw refcounter and
+ *  possibly release it
  * @backbone_gw: backbone gateway to be free'd
  */
 static void
 batadv_backbone_gw_free_ref(struct batadv_bla_backbone_gw *backbone_gw)
 {
-	if (atomic_dec_and_test(&backbone_gw->refcount))
-		kfree_rcu(backbone_gw, rcu);
+	kref_put(&backbone_gw->refcount, batadv_backbone_gw_release);
 }
 
 /**
@@ -247,7 +263,7 @@ batadv_backbone_hash_find(struct batadv_priv *bat_priv, u8 *addr,
 						&search_entry))
 			continue;
 
-		if (!atomic_inc_not_zero(&backbone_gw->refcount))
+		if (!kref_get_unless_zero(&backbone_gw->refcount))
 			continue;
 
 		backbone_gw_tmp = backbone_gw;
@@ -448,7 +464,8 @@ batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, u8 *orig,
 	ether_addr_copy(entry->orig, orig);
 
 	/* one for the hash, one for returning */
-	atomic_set(&entry->refcount, 2);
+	kref_init(&entry->refcount);
+	kref_get(&entry->refcount);
 
 	hash_added = batadv_hash_add(bat_priv->bla.backbone_hash,
 				     batadv_compare_backbone_gw,
@@ -664,7 +681,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,
 		batadv_backbone_gw_free_ref(claim->backbone_gw);
 	}
 	/* set (new) backbone gw */
-	atomic_inc(&backbone_gw->refcount);
+	kref_get(&backbone_gw->refcount);
 	claim->backbone_gw = backbone_gw;
 
 	spin_lock_bh(&backbone_gw->crc_lock);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 868f6b5..595f524 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -930,7 +930,7 @@ struct batadv_bla_backbone_gw {
 	atomic_t request_sent;
 	u16 crc;
 	spinlock_t crc_lock; /* protects crc */
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 71b7e3d3161f7b1922a2c3b5d39b7b451c3f8cf7 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:44 +0100
Subject: batman-adv: Convert batadv_bla_claim to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 41116e4..7781f39 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -174,8 +174,12 @@ batadv_backbone_gw_free_ref(struct batadv_bla_backbone_gw *backbone_gw)
  *  grace period
  * @ref: kref pointer of the claim
  */
-static void batadv_claim_release(struct batadv_bla_claim *claim)
+static void batadv_claim_release(struct kref *ref)
 {
+	struct batadv_bla_claim *claim;
+
+	claim = container_of(ref, struct batadv_bla_claim, refcount);
+
 	batadv_backbone_gw_free_ref(claim->backbone_gw);
 	kfree_rcu(claim, rcu);
 }
@@ -187,8 +191,7 @@ static void batadv_claim_release(struct batadv_bla_claim *claim)
  */
 static void batadv_claim_free_ref(struct batadv_bla_claim *claim)
 {
-	if (atomic_dec_and_test(&claim->refcount))
-		batadv_claim_release(claim);
+	kref_put(&claim->refcount, batadv_claim_release);
 }
 
 /**
@@ -219,7 +222,7 @@ static struct batadv_bla_claim
 		if (!batadv_compare_claim(&claim->hash_entry, data))
 			continue;
 
-		if (!atomic_inc_not_zero(&claim->refcount))
+		if (!kref_get_unless_zero(&claim->refcount))
 			continue;
 
 		claim_tmp = claim;
@@ -651,7 +654,8 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,
 		claim->lasttime = jiffies;
 		claim->backbone_gw = backbone_gw;
 
-		atomic_set(&claim->refcount, 2);
+		kref_init(&claim->refcount);
+		kref_get(&claim->refcount);
 		batadv_dbg(BATADV_DBG_BLA, bat_priv,
 			   "bla_add_claim(): adding new entry %pM, vid %d to hash ...\n",
 			   mac, BATADV_PRINT_VID(vid));
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 595f524..34c13b6 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -951,7 +951,7 @@ struct batadv_bla_claim {
 	unsigned long lasttime;
 	struct hlist_node hash_entry;
 	struct rcu_head rcu;
-	atomic_t refcount;
+	struct kref refcount;
 };
 #endif
 
-- 
cgit v0.10.2


From daf99b4810658d41d22be367fc3c1ae22505bef2 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:45 +0100
Subject: batman-adv: Convert batadv_nc_node to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 0b30c15..7a1216e 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -32,6 +32,7 @@
 #include <linux/jhash.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
@@ -209,23 +210,26 @@ void batadv_nc_init_orig(struct batadv_orig_node *orig_node)
 /**
  * batadv_nc_node_release - release nc_node from lists and queue for free after
  *  rcu grace period
- * @nc_node: the nc node to free
+ * @ref: kref pointer of the nc_node
  */
-static void batadv_nc_node_release(struct batadv_nc_node *nc_node)
+static void batadv_nc_node_release(struct kref *ref)
 {
+	struct batadv_nc_node *nc_node;
+
+	nc_node = container_of(ref, struct batadv_nc_node, refcount);
+
 	batadv_orig_node_free_ref(nc_node->orig_node);
 	kfree_rcu(nc_node, rcu);
 }
 
 /**
- * batadv_nc_node_free_ref - decrement the nc node refcounter and possibly
+ * batadv_nc_node_free_ref - decrement the nc_node refcounter and possibly
  *  release it
- * @nc_node: the nc node to free
+ * @nc_node: nc_node to be free'd
  */
 static void batadv_nc_node_free_ref(struct batadv_nc_node *nc_node)
 {
-	if (atomic_dec_and_test(&nc_node->refcount))
-		batadv_nc_node_release(nc_node);
+	kref_put(&nc_node->refcount, batadv_nc_node_release);
 }
 
 /**
@@ -797,7 +801,7 @@ static struct batadv_nc_node
 		if (!batadv_compare_eth(nc_node->addr, orig_node->orig))
 			continue;
 
-		if (!atomic_inc_not_zero(&nc_node->refcount))
+		if (!kref_get_unless_zero(&nc_node->refcount))
 			continue;
 
 		/* Found a match */
@@ -848,7 +852,8 @@ static struct batadv_nc_node
 	INIT_LIST_HEAD(&nc_node->list);
 	ether_addr_copy(nc_node->addr, orig_node->orig);
 	nc_node->orig_node = orig_neigh_node;
-	atomic_set(&nc_node->refcount, 2);
+	kref_init(&nc_node->refcount);
+	kref_get(&nc_node->refcount);
 
 	/* Select ingoing or outgoing coding node */
 	if (in_coding) {
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 34c13b6..05421af 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1067,7 +1067,7 @@ struct batadv_tt_roam_node {
 struct batadv_nc_node {
 	struct list_head list;
 	u8 addr[ETH_ALEN];
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 	struct batadv_orig_node *orig_node;
 	unsigned long last_seen;
-- 
cgit v0.10.2


From 727e0cd59e7be176edc7728cef883f912a29168f Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:46 +0100
Subject: batman-adv: Convert batadv_nc_path to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 7a1216e..371fc6b 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -233,14 +233,27 @@ static void batadv_nc_node_free_ref(struct batadv_nc_node *nc_node)
 }
 
 /**
- * batadv_nc_path_free_ref - decrements the nc path refcounter and possibly
- * frees it
- * @nc_path: the nc node to free
+ * batadv_nc_path_release - release nc_path from lists and queue for free after
+ *  rcu grace period
+ * @ref: kref pointer of the nc_path
+ */
+static void batadv_nc_path_release(struct kref *ref)
+{
+	struct batadv_nc_path *nc_path;
+
+	nc_path = container_of(ref, struct batadv_nc_path, refcount);
+
+	kfree_rcu(nc_path, rcu);
+}
+
+/**
+ * batadv_nc_path_free_ref - decrement the nc_path refcounter and possibly
+ *  release it
+ * @nc_path: nc_path to be free'd
  */
 static void batadv_nc_path_free_ref(struct batadv_nc_path *nc_path)
 {
-	if (atomic_dec_and_test(&nc_path->refcount))
-		kfree_rcu(nc_path, rcu);
+	kref_put(&nc_path->refcount, batadv_nc_path_release);
 }
 
 /**
@@ -545,7 +558,7 @@ batadv_nc_hash_find(struct batadv_hashtable *hash,
 		if (!batadv_nc_hash_compare(&nc_path->hash_entry, data))
 			continue;
 
-		if (!atomic_inc_not_zero(&nc_path->refcount))
+		if (!kref_get_unless_zero(&nc_path->refcount))
 			continue;
 
 		nc_path_tmp = nc_path;
@@ -972,7 +985,8 @@ static struct batadv_nc_path *batadv_nc_get_path(struct batadv_priv *bat_priv,
 	/* Initialize nc_path */
 	INIT_LIST_HEAD(&nc_path->packet_list);
 	spin_lock_init(&nc_path->packet_list_lock);
-	atomic_set(&nc_path->refcount, 2);
+	kref_init(&nc_path->refcount);
+	kref_get(&nc_path->refcount);
 	nc_path->last_valid = jiffies;
 	ether_addr_copy(nc_path->next_hop, dst);
 	ether_addr_copy(nc_path->prev_hop, src);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 05421af..b3d9b5b 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1087,7 +1087,7 @@ struct batadv_nc_node {
 struct batadv_nc_path {
 	struct hlist_node hash_entry;
 	struct rcu_head rcu;
-	atomic_t refcount;
+	struct kref refcount;
 	struct list_head packet_list;
 	spinlock_t packet_list_lock; /* Protects packet_list */
 	u8 next_hop[ETH_ALEN];
-- 
cgit v0.10.2


From 68a6722cc41f92ceffa8915bfe6a9c0307f2abeb Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:47 +0100
Subject: batman-adv: Convert batadv_dat_entry to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 017fffe..921e252 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -30,6 +30,7 @@
 #include <linux/in.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/rculist.h>
 #include <linux/rcupdate.h>
@@ -62,14 +63,27 @@ static void batadv_dat_start_timer(struct batadv_priv *bat_priv)
 }
 
 /**
+ * batadv_dat_entry_release - release dat_entry from lists and queue for free
+ *  after rcu grace period
+ * @ref: kref pointer of the dat_entry
+ */
+static void batadv_dat_entry_release(struct kref *ref)
+{
+	struct batadv_dat_entry *dat_entry;
+
+	dat_entry = container_of(ref, struct batadv_dat_entry, refcount);
+
+	kfree_rcu(dat_entry, rcu);
+}
+
+/**
  * batadv_dat_entry_free_ref - decrement the dat_entry refcounter and possibly
- * free it
- * @dat_entry: the entry to free
+ *  release it
+ * @dat_entry: dat_entry to be free'd
  */
 static void batadv_dat_entry_free_ref(struct batadv_dat_entry *dat_entry)
 {
-	if (atomic_dec_and_test(&dat_entry->refcount))
-		kfree_rcu(dat_entry, rcu);
+	kref_put(&dat_entry->refcount, batadv_dat_entry_release);
 }
 
 /**
@@ -281,7 +295,7 @@ batadv_dat_entry_hash_find(struct batadv_priv *bat_priv, __be32 ip,
 		if (dat_entry->ip != ip)
 			continue;
 
-		if (!atomic_inc_not_zero(&dat_entry->refcount))
+		if (!kref_get_unless_zero(&dat_entry->refcount))
 			continue;
 
 		dat_entry_tmp = dat_entry;
@@ -326,7 +340,8 @@ static void batadv_dat_entry_add(struct batadv_priv *bat_priv, __be32 ip,
 	dat_entry->vid = vid;
 	ether_addr_copy(dat_entry->mac_addr, mac_addr);
 	dat_entry->last_update = jiffies;
-	atomic_set(&dat_entry->refcount, 2);
+	kref_init(&dat_entry->refcount);
+	kref_get(&dat_entry->refcount);
 
 	hash_added = batadv_hash_add(bat_priv->dat.hash, batadv_compare_dat,
 				     batadv_hash_dat, dat_entry,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index b3d9b5b..9df8f39 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1230,7 +1230,7 @@ struct batadv_dat_entry {
 	unsigned short vid;
 	unsigned long last_update;
 	struct hlist_node hash_entry;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From f7157dd135012d9e9b4d74a9f8b2426ea92131f8 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:48 +0100
Subject: batman-adv: Convert batadv_tvlv_container to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index a9b4f75..4b6c258 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -29,6 +29,7 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/module.h>
@@ -670,14 +671,25 @@ static struct batadv_tvlv_handler
 }
 
 /**
+ * batadv_tvlv_container_release - release tvlv from lists and free
+ * @ref: kref pointer of the tvlv
+ */
+static void batadv_tvlv_container_release(struct kref *ref)
+{
+	struct batadv_tvlv_container *tvlv;
+
+	tvlv = container_of(ref, struct batadv_tvlv_container, refcount);
+	kfree(tvlv);
+}
+
+/**
  * batadv_tvlv_container_free_ref - decrement the tvlv container refcounter and
- *  possibly free it
+ *  possibly release it
  * @tvlv: the tvlv container to free
  */
 static void batadv_tvlv_container_free_ref(struct batadv_tvlv_container *tvlv)
 {
-	if (atomic_dec_and_test(&tvlv->refcount))
-		kfree(tvlv);
+	kref_put(&tvlv->refcount, batadv_tvlv_container_release);
 }
 
 /**
@@ -706,7 +718,7 @@ static struct batadv_tvlv_container
 		if (tvlv_tmp->tvlv_hdr.version != version)
 			continue;
 
-		if (!atomic_inc_not_zero(&tvlv_tmp->refcount))
+		if (!kref_get_unless_zero(&tvlv_tmp->refcount))
 			continue;
 
 		tvlv = tvlv_tmp;
@@ -814,7 +826,7 @@ void batadv_tvlv_container_register(struct batadv_priv *bat_priv,
 
 	memcpy(tvlv_new + 1, tvlv_value, ntohs(tvlv_new->tvlv_hdr.len));
 	INIT_HLIST_NODE(&tvlv_new->list);
-	atomic_set(&tvlv_new->refcount, 1);
+	kref_init(&tvlv_new->refcount);
 
 	spin_lock_bh(&bat_priv->tvlv.container_list_lock);
 	tvlv_old = batadv_tvlv_container_get(bat_priv, type, version);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 9df8f39..c55925b 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1266,7 +1266,7 @@ struct batadv_dat_candidate {
 struct batadv_tvlv_container {
 	struct hlist_node list;
 	struct batadv_tvlv_hdr tvlv_hdr;
-	atomic_t refcount;
+	struct kref refcount;
 };
 
 /**
-- 
cgit v0.10.2


From 32836f56f88c280d39fc302bcc0e0e3ed6bb412e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:49 +0100
Subject: batman-adv: Convert batadv_tvlv_handler to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 4b6c258..568c550 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -625,15 +625,27 @@ __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr)
 }
 
 /**
- * batadv_tvlv_handler_free_ref - decrement the tvlv handler refcounter and
- *  possibly free it
+ * batadv_tvlv_handler_release - release tvlv handler from lists and queue for
+ *  free after rcu grace period
+ * @ref: kref pointer of the tvlv
+ */
+static void batadv_tvlv_handler_release(struct kref *ref)
+{
+	struct batadv_tvlv_handler *tvlv_handler;
+
+	tvlv_handler = container_of(ref, struct batadv_tvlv_handler, refcount);
+	kfree_rcu(tvlv_handler, rcu);
+}
+
+/**
+ * batadv_tvlv_handler_free_ref - decrement the tvlv container refcounter and
+ *  possibly release it
  * @tvlv_handler: the tvlv handler to free
  */
 static void
 batadv_tvlv_handler_free_ref(struct batadv_tvlv_handler *tvlv_handler)
 {
-	if (atomic_dec_and_test(&tvlv_handler->refcount))
-		kfree_rcu(tvlv_handler, rcu);
+	kref_put(&tvlv_handler->refcount, batadv_tvlv_handler_release);
 }
 
 /**
@@ -659,7 +671,7 @@ static struct batadv_tvlv_handler
 		if (tvlv_handler_tmp->version != version)
 			continue;
 
-		if (!atomic_inc_not_zero(&tvlv_handler_tmp->refcount))
+		if (!kref_get_unless_zero(&tvlv_handler_tmp->refcount))
 			continue;
 
 		tvlv_handler = tvlv_handler_tmp;
@@ -1112,7 +1124,7 @@ void batadv_tvlv_handler_register(struct batadv_priv *bat_priv,
 	tvlv_handler->type = type;
 	tvlv_handler->version = version;
 	tvlv_handler->flags = flags;
-	atomic_set(&tvlv_handler->refcount, 1);
+	kref_init(&tvlv_handler->refcount);
 	INIT_HLIST_NODE(&tvlv_handler->list);
 
 	spin_lock_bh(&bat_priv->tvlv.handler_list_lock);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index c55925b..f89dd97 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1293,7 +1293,7 @@ struct batadv_tvlv_handler {
 	u8 type;
 	u8 version;
 	u8 flags;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 6e8ef69dd46044abcee3c3dbc5a42afce229ac48 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:50 +0100
Subject: batman-adv: Convert batadv_tt_orig_list_entry to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 0dc8a5c..68494a1 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -31,6 +31,7 @@
 #include <linux/jhash.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
@@ -346,22 +347,28 @@ static void batadv_tt_global_size_dec(struct batadv_orig_node *orig_node,
 /**
  * batadv_tt_orig_list_entry_release - release tt orig entry from lists and
  *  queue for free after rcu grace period
- * @orig_entry: tt orig entry to be free'd
+ * @ref: kref pointer of the tt orig entry
  */
-static void
-batadv_tt_orig_list_entry_release(struct batadv_tt_orig_list_entry *orig_entry)
+static void batadv_tt_orig_list_entry_release(struct kref *ref)
 {
+	struct batadv_tt_orig_list_entry *orig_entry;
+
+	orig_entry = container_of(ref, struct batadv_tt_orig_list_entry,
+				  refcount);
+
 	batadv_orig_node_free_ref(orig_entry->orig_node);
 	kfree_rcu(orig_entry, rcu);
 }
 
+/**
+ * batadv_tt_orig_list_entry_free_ref - decrement the tt orig entry refcounter
+ *  and possibly release it
+ * @orig_entry: tt orig entry to be free'd
+ */
 static void
 batadv_tt_orig_list_entry_free_ref(struct batadv_tt_orig_list_entry *orig_entry)
 {
-	if (!atomic_dec_and_test(&orig_entry->refcount))
-		return;
-
-	batadv_tt_orig_list_entry_release(orig_entry);
+	kref_put(&orig_entry->refcount, batadv_tt_orig_list_entry_release);
 }
 
 /**
@@ -1270,7 +1277,7 @@ batadv_tt_global_orig_entry_find(const struct batadv_tt_global_entry *entry,
 	hlist_for_each_entry_rcu(tmp_orig_entry, head, list) {
 		if (tmp_orig_entry->orig_node != orig_node)
 			continue;
-		if (!atomic_inc_not_zero(&tmp_orig_entry->refcount))
+		if (!kref_get_unless_zero(&tmp_orig_entry->refcount))
 			continue;
 
 		orig_entry = tmp_orig_entry;
@@ -1331,7 +1338,8 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global,
 	batadv_tt_global_size_inc(orig_node, tt_global->common.vid);
 	orig_entry->orig_node = orig_node;
 	orig_entry->ttvn = ttvn;
-	atomic_set(&orig_entry->refcount, 2);
+	kref_init(&orig_entry->refcount);
+	kref_get(&orig_entry->refcount);
 
 	spin_lock_bh(&tt_global->list_lock);
 	hlist_add_head_rcu(&orig_entry->list,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index f89dd97..17f97ff 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -1014,7 +1014,7 @@ struct batadv_tt_orig_list_entry {
 	struct batadv_orig_node *orig_node;
 	u8 ttvn;
 	struct hlist_node list;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 962c68328b14eb311edc950801cda2fceddad2df Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:51 +0100
Subject: batman-adv: Convert batadv_neigh_ifinfo to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 7710595..1c50abe 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -172,11 +172,14 @@ err:
 /**
  * batadv_neigh_ifinfo_release - release neigh_ifinfo from lists and queue for
  *  free after rcu grace period
- * @neigh_ifinfo: the neigh_ifinfo object to release
+ * @ref: kref pointer of the neigh_ifinfo
  */
-static void
-batadv_neigh_ifinfo_release(struct batadv_neigh_ifinfo *neigh_ifinfo)
+static void batadv_neigh_ifinfo_release(struct kref *ref)
 {
+	struct batadv_neigh_ifinfo *neigh_ifinfo;
+
+	neigh_ifinfo = container_of(ref, struct batadv_neigh_ifinfo, refcount);
+
 	if (neigh_ifinfo->if_outgoing != BATADV_IF_DEFAULT)
 		batadv_hardif_free_ref(neigh_ifinfo->if_outgoing);
 
@@ -190,8 +193,7 @@ batadv_neigh_ifinfo_release(struct batadv_neigh_ifinfo *neigh_ifinfo)
  */
 void batadv_neigh_ifinfo_free_ref(struct batadv_neigh_ifinfo *neigh_ifinfo)
 {
-	if (atomic_dec_and_test(&neigh_ifinfo->refcount))
-		batadv_neigh_ifinfo_release(neigh_ifinfo);
+	kref_put(&neigh_ifinfo->refcount, batadv_neigh_ifinfo_release);
 }
 
 /**
@@ -405,7 +407,7 @@ batadv_neigh_ifinfo_get(struct batadv_neigh_node *neigh,
 		if (tmp_neigh_ifinfo->if_outgoing != if_outgoing)
 			continue;
 
-		if (!atomic_inc_not_zero(&tmp_neigh_ifinfo->refcount))
+		if (!kref_get_unless_zero(&tmp_neigh_ifinfo->refcount))
 			continue;
 
 		neigh_ifinfo = tmp_neigh_ifinfo;
@@ -450,7 +452,8 @@ batadv_neigh_ifinfo_new(struct batadv_neigh_node *neigh,
 	}
 
 	INIT_HLIST_NODE(&neigh_ifinfo->list);
-	atomic_set(&neigh_ifinfo->refcount, 2);
+	kref_init(&neigh_ifinfo->refcount);
+	kref_get(&neigh_ifinfo->refcount);
 	neigh_ifinfo->if_outgoing = if_outgoing;
 
 	hlist_add_head_rcu(&neigh_ifinfo->list, &neigh->ifinfo_list);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 17f97ff..81d9f8d 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -420,7 +420,7 @@ struct batadv_neigh_ifinfo {
 	struct batadv_hard_iface *if_outgoing;
 	struct batadv_neigh_ifinfo_bat_iv bat_iv;
 	u8 last_ttl;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From a6ba0d340db63b95cd07484370cca69929d8f4e4 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:52 +0100
Subject: batman-adv: Convert batadv_orig_ifinfo to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 1c50abe..428983a 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -326,7 +326,7 @@ batadv_orig_ifinfo_get(struct batadv_orig_node *orig_node,
 		if (tmp->if_outgoing != if_outgoing)
 			continue;
 
-		if (!atomic_inc_not_zero(&tmp->refcount))
+		if (!kref_get_unless_zero(&tmp->refcount))
 			continue;
 
 		orig_ifinfo = tmp;
@@ -377,7 +377,8 @@ batadv_orig_ifinfo_new(struct batadv_orig_node *orig_node,
 	orig_ifinfo->batman_seqno_reset = reset_time;
 	orig_ifinfo->if_outgoing = if_outgoing;
 	INIT_HLIST_NODE(&orig_ifinfo->list);
-	atomic_set(&orig_ifinfo->refcount, 2);
+	kref_init(&orig_ifinfo->refcount);
+	kref_get(&orig_ifinfo->refcount);
 	hlist_add_head_rcu(&orig_ifinfo->list,
 			   &orig_node->ifinfo_list);
 out:
@@ -704,12 +705,15 @@ int batadv_hardif_neigh_seq_print_text(struct seq_file *seq, void *offset)
 /**
  * batadv_orig_ifinfo_release - release orig_ifinfo from lists and queue for
  *  free after rcu grace period
- * @orig_ifinfo: the orig_ifinfo object to release
+ * @ref: kref pointer of the orig_ifinfo
  */
-static void batadv_orig_ifinfo_release(struct batadv_orig_ifinfo *orig_ifinfo)
+static void batadv_orig_ifinfo_release(struct kref *ref)
 {
+	struct batadv_orig_ifinfo *orig_ifinfo;
 	struct batadv_neigh_node *router;
 
+	orig_ifinfo = container_of(ref, struct batadv_orig_ifinfo, refcount);
+
 	if (orig_ifinfo->if_outgoing != BATADV_IF_DEFAULT)
 		batadv_hardif_free_ref(orig_ifinfo->if_outgoing);
 
@@ -728,8 +732,7 @@ static void batadv_orig_ifinfo_release(struct batadv_orig_ifinfo *orig_ifinfo)
  */
 void batadv_orig_ifinfo_free_ref(struct batadv_orig_ifinfo *orig_ifinfo)
 {
-	if (atomic_dec_and_test(&orig_ifinfo->refcount))
-		batadv_orig_ifinfo_release(orig_ifinfo);
+	kref_put(&orig_ifinfo->refcount, batadv_orig_ifinfo_release);
 }
 
 /**
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 1c8b35d..4a5cd8b 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -25,6 +25,7 @@
 #include <linux/etherdevice.h>
 #include <linux/if_ether.h>
 #include <linux/jiffies.h>
+#include <linux/kref.h>
 #include <linux/netdevice.h>
 #include <linux/printk.h>
 #include <linux/rculist.h>
@@ -497,7 +498,7 @@ batadv_find_router(struct batadv_priv *bat_priv,
 
 	hlist_for_each_entry_rcu(cand, &orig_node->ifinfo_list, list) {
 		/* acquire some structures and references ... */
-		if (!atomic_inc_not_zero(&cand->refcount))
+		if (!kref_get_unless_zero(&cand->refcount))
 			continue;
 
 		cand_router = rcu_dereference(cand->router);
@@ -524,7 +525,7 @@ batadv_find_router(struct batadv_priv *bat_priv,
 		/* mark the first possible candidate */
 		if (!first_candidate) {
 			atomic_inc(&cand_router->refcount);
-			atomic_inc(&cand->refcount);
+			kref_get(&cand->refcount);
 			first_candidate = cand;
 			first_candidate_router = cand_router;
 		}
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 81d9f8d..6193c01 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -141,7 +141,7 @@ struct batadv_orig_ifinfo {
 	u32 last_real_seqno;
 	u8 last_ttl;
 	unsigned long batman_seqno_reset;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 77ae32e8989a26d24bd8007be3b9dc4a348175a9 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:53 +0100
Subject: batman-adv: Convert batadv_neigh_node to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 23ce90e..affcbb5 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -31,6 +31,7 @@
 #include <linux/init.h>
 #include <linux/jiffies.h>
 #include <linux/list.h>
+#include <linux/kref.h>
 #include <linux/netdevice.h>
 #include <linux/pkt_sched.h>
 #include <linux/printk.h>
@@ -1002,7 +1003,7 @@ batadv_iv_ogm_orig_update(struct batadv_priv *bat_priv,
 		neigh_addr = tmp_neigh_node->addr;
 		if (batadv_compare_eth(neigh_addr, ethhdr->h_source) &&
 		    tmp_neigh_node->if_incoming == if_incoming &&
-		    atomic_inc_not_zero(&tmp_neigh_node->refcount)) {
+		    kref_get_unless_zero(&tmp_neigh_node->refcount)) {
 			if (WARN(neigh_node, "too many matching neigh_nodes"))
 				batadv_neigh_node_free_ref(neigh_node);
 			neigh_node = tmp_neigh_node;
@@ -1161,7 +1162,7 @@ static int batadv_iv_ogm_calc_tq(struct batadv_orig_node *orig_node,
 		if (tmp_neigh_node->if_incoming != if_incoming)
 			continue;
 
-		if (!atomic_inc_not_zero(&tmp_neigh_node->refcount))
+		if (!kref_get_unless_zero(&tmp_neigh_node->refcount))
 			continue;
 
 		neigh_node = tmp_neigh_node;
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 428983a..9e3dbd8 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -229,15 +229,17 @@ void batadv_hardif_neigh_free_ref(struct batadv_hardif_neigh_node *hardif_neigh)
 /**
  * batadv_neigh_node_release - release neigh_node from lists and queue for
  *  free after rcu grace period
- * @neigh_node: neigh neighbor to free
+ * @ref: kref pointer of the neigh_node
  */
-static void batadv_neigh_node_release(struct batadv_neigh_node *neigh_node)
+static void batadv_neigh_node_release(struct kref *ref)
 {
 	struct hlist_node *node_tmp;
+	struct batadv_neigh_node *neigh_node;
 	struct batadv_hardif_neigh_node *hardif_neigh;
 	struct batadv_neigh_ifinfo *neigh_ifinfo;
 	struct batadv_algo_ops *bao;
 
+	neigh_node = container_of(ref, struct batadv_neigh_node, refcount);
 	bao = neigh_node->orig_node->bat_priv->bat_algo_ops;
 
 	hlist_for_each_entry_safe(neigh_ifinfo, node_tmp,
@@ -262,14 +264,13 @@ static void batadv_neigh_node_release(struct batadv_neigh_node *neigh_node)
 }
 
 /**
- * batadv_neigh_node_free_ref - decrement the neighbors refcounter
- *  and possibly release it
+ * batadv_neigh_node_free_ref - decrement the neighbors refcounter and possibly
+ *  release it
  * @neigh_node: neigh neighbor to free
  */
 void batadv_neigh_node_free_ref(struct batadv_neigh_node *neigh_node)
 {
-	if (atomic_dec_and_test(&neigh_node->refcount))
-		batadv_neigh_node_release(neigh_node);
+	kref_put(&neigh_node->refcount, batadv_neigh_node_release);
 }
 
 /**
@@ -298,7 +299,7 @@ batadv_orig_router_get(struct batadv_orig_node *orig_node,
 		break;
 	}
 
-	if (router && !atomic_inc_not_zero(&router->refcount))
+	if (router && !kref_get_unless_zero(&router->refcount))
 		router = NULL;
 
 	rcu_read_unlock();
@@ -491,7 +492,7 @@ batadv_neigh_node_get(const struct batadv_orig_node *orig_node,
 		if (tmp_neigh_node->if_incoming != hard_iface)
 			continue;
 
-		if (!atomic_inc_not_zero(&tmp_neigh_node->refcount))
+		if (!kref_get_unless_zero(&tmp_neigh_node->refcount))
 			continue;
 
 		res = tmp_neigh_node;
@@ -649,7 +650,8 @@ batadv_neigh_node_new(struct batadv_orig_node *orig_node,
 	neigh_node->orig_node = orig_node;
 
 	/* extra reference for return */
-	atomic_set(&neigh_node->refcount, 2);
+	kref_init(&neigh_node->refcount);
+	kref_get(&neigh_node->refcount);
 
 	spin_lock_bh(&orig_node->neigh_list_lock);
 	hlist_add_head_rcu(&neigh_node->list, &orig_node->neigh_list);
@@ -1084,7 +1086,7 @@ batadv_find_best_neighbor(struct batadv_priv *bat_priv,
 						best, if_outgoing) <= 0))
 			continue;
 
-		if (!atomic_inc_not_zero(&neigh->refcount))
+		if (!kref_get_unless_zero(&neigh->refcount))
 			continue;
 
 		if (best)
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 4a5cd8b..205310b 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -73,7 +73,7 @@ static void _batadv_update_route(struct batadv_priv *bat_priv,
 
 	rcu_read_lock();
 	curr_router = rcu_dereference(orig_ifinfo->router);
-	if (curr_router && !atomic_inc_not_zero(&curr_router->refcount))
+	if (curr_router && !kref_get_unless_zero(&curr_router->refcount))
 		curr_router = NULL;
 	rcu_read_unlock();
 
@@ -101,7 +101,7 @@ static void _batadv_update_route(struct batadv_priv *bat_priv,
 		batadv_neigh_node_free_ref(curr_router);
 
 	/* increase refcount of new best neighbor */
-	if (neigh_node && !atomic_inc_not_zero(&neigh_node->refcount))
+	if (neigh_node && !kref_get_unless_zero(&neigh_node->refcount))
 		neigh_node = NULL;
 
 	spin_lock_bh(&orig_node->neigh_list_lock);
@@ -505,7 +505,7 @@ batadv_find_router(struct batadv_priv *bat_priv,
 		if (!cand_router)
 			goto next;
 
-		if (!atomic_inc_not_zero(&cand_router->refcount)) {
+		if (!kref_get_unless_zero(&cand_router->refcount)) {
 			cand_router = NULL;
 			goto next;
 		}
@@ -524,7 +524,7 @@ batadv_find_router(struct batadv_priv *bat_priv,
 
 		/* mark the first possible candidate */
 		if (!first_candidate) {
-			atomic_inc(&cand_router->refcount);
+			kref_get(&cand_router->refcount);
 			kref_get(&cand->refcount);
 			first_candidate = cand;
 			first_candidate_router = cand_router;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 6193c01..68ca39e 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -384,7 +384,7 @@ struct batadv_neigh_node {
 	spinlock_t ifinfo_lock;	/* protects ifinfo_list and its members */
 	struct batadv_hard_iface *if_incoming;
 	unsigned long last_seen;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 7a659d5694b9b580615f0a813b1e6675f26c629e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:54 +0100
Subject: batman-adv: Convert batadv_hard_iface to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index affcbb5..bf0e7d6 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -644,10 +644,10 @@ static void batadv_iv_ogm_aggregate_new(const unsigned char *packet_buff,
 	unsigned char *skb_buff;
 	unsigned int skb_size;
 
-	if (!atomic_inc_not_zero(&if_incoming->refcount))
+	if (!kref_get_unless_zero(&if_incoming->refcount))
 		return;
 
-	if (!atomic_inc_not_zero(&if_outgoing->refcount))
+	if (!kref_get_unless_zero(&if_outgoing->refcount))
 		goto out_free_incoming;
 
 	/* own packet should always be scheduled */
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index b17e272..fb2d9c0 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -18,6 +18,7 @@
 #include "hard-interface.h"
 #include "main.h"
 
+#include <linux/atomic.h>
 #include <linux/bug.h>
 #include <linux/byteorder/generic.h>
 #include <linux/errno.h>
@@ -26,6 +27,7 @@
 #include <linux/if_ether.h>
 #include <linux/if.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/netdevice.h>
 #include <linux/printk.h>
@@ -50,10 +52,13 @@
 /**
  * batadv_hardif_release - release hard interface from lists and queue for
  *  free after rcu grace period
- * @hard_iface: the hard interface to free
+ * @ref: kref pointer of the hard interface
  */
-void batadv_hardif_release(struct batadv_hard_iface *hard_iface)
+void batadv_hardif_release(struct kref *ref)
 {
+	struct batadv_hard_iface *hard_iface;
+
+	hard_iface = container_of(ref, struct batadv_hard_iface, refcount);
 	dev_put(hard_iface->net_dev);
 
 	kfree_rcu(hard_iface, rcu);
@@ -67,7 +72,7 @@ batadv_hardif_get_by_netdev(const struct net_device *net_dev)
 	rcu_read_lock();
 	list_for_each_entry_rcu(hard_iface, &batadv_hardif_list, list) {
 		if (hard_iface->net_dev == net_dev &&
-		    atomic_inc_not_zero(&hard_iface->refcount))
+		    kref_get_unless_zero(&hard_iface->refcount))
 			goto out;
 	}
 
@@ -172,7 +177,7 @@ batadv_hardif_get_active(const struct net_device *soft_iface)
 			continue;
 
 		if (hard_iface->if_status == BATADV_IF_ACTIVE &&
-		    atomic_inc_not_zero(&hard_iface->refcount))
+		    kref_get_unless_zero(&hard_iface->refcount))
 			goto out;
 	}
 
@@ -206,7 +211,7 @@ static void batadv_primary_if_select(struct batadv_priv *bat_priv,
 
 	ASSERT_RTNL();
 
-	if (new_hard_iface && !atomic_inc_not_zero(&new_hard_iface->refcount))
+	if (new_hard_iface && !kref_get_unless_zero(&new_hard_iface->refcount))
 		new_hard_iface = NULL;
 
 	curr_hard_iface = rcu_dereference_protected(bat_priv->primary_if, 1);
@@ -434,7 +439,7 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface,
 	if (hard_iface->if_status != BATADV_IF_NOT_IN_USE)
 		goto out;
 
-	if (!atomic_inc_not_zero(&hard_iface->refcount))
+	if (!kref_get_unless_zero(&hard_iface->refcount))
 		goto out;
 
 	soft_iface = dev_get_by_name(&init_net, iface_name);
@@ -655,7 +660,8 @@ batadv_hardif_add_interface(struct net_device *net_dev)
 		hard_iface->num_bcasts = BATADV_NUM_BCASTS_WIRELESS;
 
 	/* extra reference for return */
-	atomic_set(&hard_iface->refcount, 2);
+	kref_init(&hard_iface->refcount);
+	kref_get(&hard_iface->refcount);
 
 	batadv_check_known_mac_addr(hard_iface->net_dev);
 	list_add_tail_rcu(&hard_iface->list, &batadv_hardif_list);
diff --git a/net/batman-adv/hard-interface.h b/net/batman-adv/hard-interface.h
index 9eb08ad..5cecc6b 100644
--- a/net/batman-adv/hard-interface.h
+++ b/net/batman-adv/hard-interface.h
@@ -20,8 +20,8 @@
 
 #include "main.h"
 
-#include <linux/atomic.h>
 #include <linux/compiler.h>
+#include <linux/kref.h>
 #include <linux/notifier.h>
 #include <linux/rcupdate.h>
 #include <linux/stddef.h>
@@ -61,18 +61,16 @@ void batadv_hardif_disable_interface(struct batadv_hard_iface *hard_iface,
 void batadv_hardif_remove_interfaces(void);
 int batadv_hardif_min_mtu(struct net_device *soft_iface);
 void batadv_update_min_mtu(struct net_device *soft_iface);
-void batadv_hardif_release(struct batadv_hard_iface *hard_iface);
+void batadv_hardif_release(struct kref *ref);
 
 /**
- * batadv_hardif_free_ref - decrement the hard interface refcounter and
- *  possibly release it
+ * batadv_hardif_free_ref - decrement the hard interface refcounter and possibly
+ *  release it
  * @hard_iface: the hard interface to free
  */
-static inline void
-batadv_hardif_free_ref(struct batadv_hard_iface *hard_iface)
+static inline void batadv_hardif_free_ref(struct batadv_hard_iface *hard_iface)
 {
-	if (atomic_dec_and_test(&hard_iface->refcount))
-		batadv_hardif_release(hard_iface);
+	kref_put(&hard_iface->refcount, batadv_hardif_release);
 }
 
 static inline struct batadv_hard_iface *
@@ -85,7 +83,7 @@ batadv_primary_if_get_selected(struct batadv_priv *bat_priv)
 	if (!hard_iface)
 		goto out;
 
-	if (!atomic_inc_not_zero(&hard_iface->refcount))
+	if (!kref_get_unless_zero(&hard_iface->refcount))
 		hard_iface = NULL;
 
 out:
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 9e3dbd8..bf27007 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -367,7 +367,7 @@ batadv_orig_ifinfo_new(struct batadv_orig_node *orig_node,
 		goto out;
 
 	if (if_outgoing != BATADV_IF_DEFAULT &&
-	    !atomic_inc_not_zero(&if_outgoing->refcount)) {
+	    !kref_get_unless_zero(&if_outgoing->refcount)) {
 		kfree(orig_ifinfo);
 		orig_ifinfo = NULL;
 		goto out;
@@ -447,7 +447,7 @@ batadv_neigh_ifinfo_new(struct batadv_neigh_node *neigh,
 	if (!neigh_ifinfo)
 		goto out;
 
-	if (if_outgoing && !atomic_inc_not_zero(&if_outgoing->refcount)) {
+	if (if_outgoing && !kref_get_unless_zero(&if_outgoing->refcount)) {
 		kfree(neigh_ifinfo);
 		neigh_ifinfo = NULL;
 		goto out;
@@ -524,7 +524,7 @@ batadv_hardif_neigh_create(struct batadv_hard_iface *hard_iface,
 	if (hardif_neigh)
 		goto out;
 
-	if (!atomic_inc_not_zero(&hard_iface->refcount))
+	if (!kref_get_unless_zero(&hard_iface->refcount))
 		goto out;
 
 	hardif_neigh = kzalloc(sizeof(*hardif_neigh), GFP_ATOMIC);
@@ -635,7 +635,7 @@ batadv_neigh_node_new(struct batadv_orig_node *orig_node,
 	if (!neigh_node)
 		goto out;
 
-	if (!atomic_inc_not_zero(&hard_iface->refcount)) {
+	if (!kref_get_unless_zero(&hard_iface->refcount)) {
 		kfree(neigh_node);
 		neigh_node = NULL;
 		goto out;
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 68ca39e..74cb263 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -111,7 +111,7 @@ struct batadv_hard_iface {
 	struct net_device *net_dev;
 	u8 num_bcasts;
 	struct kobject *hardif_obj;
-	atomic_t refcount;
+	struct kref refcount;
 	struct packet_type batman_adv_ptype;
 	struct net_device *soft_iface;
 	struct rcu_head rcu;
-- 
cgit v0.10.2


From 161a3be9324eaed3ecd1879c994d58b16bbf4732 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:55 +0100
Subject: batman-adv: Convert batadv_orig_node_vlan to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index bf27007..8b2ef8d 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -82,7 +82,7 @@ batadv_orig_node_vlan_get(struct batadv_orig_node *orig_node,
 		if (tmp->vid != vid)
 			continue;
 
-		if (!atomic_inc_not_zero(&tmp->refcount))
+		if (!kref_get_unless_zero(&tmp->refcount))
 			continue;
 
 		vlan = tmp;
@@ -123,7 +123,8 @@ batadv_orig_node_vlan_new(struct batadv_orig_node *orig_node,
 	if (!vlan)
 		goto out;
 
-	atomic_set(&vlan->refcount, 2);
+	kref_init(&vlan->refcount);
+	kref_get(&vlan->refcount);
 	vlan->vid = vid;
 
 	hlist_add_head_rcu(&vlan->list, &orig_node->vlan_list);
@@ -135,14 +136,27 @@ out:
 }
 
 /**
- * batadv_orig_node_vlan_free_ref - decrement the refcounter and possibly free
- *  the originator-vlan object
+ * batadv_orig_node_vlan_release - release originator-vlan object from lists
+ *  and queue for free after rcu grace period
+ * @ref: kref pointer of the originator-vlan object
+ */
+static void batadv_orig_node_vlan_release(struct kref *ref)
+{
+	struct batadv_orig_node_vlan *orig_vlan;
+
+	orig_vlan = container_of(ref, struct batadv_orig_node_vlan, refcount);
+
+	kfree_rcu(orig_vlan, rcu);
+}
+
+/**
+ * batadv_orig_node_vlan_free_ref - decrement the refcounter and possibly
+ *  release the originator-vlan object
  * @orig_vlan: the originator-vlan object to release
  */
 void batadv_orig_node_vlan_free_ref(struct batadv_orig_node_vlan *orig_vlan)
 {
-	if (atomic_dec_and_test(&orig_vlan->refcount))
-		kfree_rcu(orig_vlan, rcu);
+	kref_put(&orig_vlan->refcount, batadv_orig_node_vlan_release);
 }
 
 int batadv_originator_init(struct batadv_priv *bat_priv)
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 74cb263..504f3db 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -197,7 +197,7 @@ struct batadv_orig_node_vlan {
 	unsigned short vid;
 	struct batadv_vlan_tt tt;
 	struct hlist_node list;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 7c12439115da77389733d9bf313f64eedaec3b73 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:56 +0100
Subject: batman-adv: Convert batadv_orig_node to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 921e252..e326111 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -542,7 +542,7 @@ static void batadv_choose_next_candidate(struct batadv_priv *bat_priv,
 							  max_orig_node))
 				continue;
 
-			if (!atomic_inc_not_zero(&orig_node->refcount))
+			if (!kref_get_unless_zero(&orig_node->refcount))
 				continue;
 
 			max = tmp_max;
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 1f3847c..261866e 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -118,7 +118,7 @@ batadv_gw_get_selected_orig(struct batadv_priv *bat_priv)
 	if (!orig_node)
 		goto unlock;
 
-	if (!atomic_inc_not_zero(&orig_node->refcount))
+	if (!kref_get_unless_zero(&orig_node->refcount))
 		orig_node = NULL;
 
 unlock:
@@ -441,7 +441,7 @@ static void batadv_gw_node_add(struct batadv_priv *bat_priv,
 	if (gateway->bandwidth_down == 0)
 		return;
 
-	if (!atomic_inc_not_zero(&orig_node->refcount))
+	if (!kref_get_unless_zero(&orig_node->refcount))
 		return;
 
 	gw_node = kzalloc(sizeof(*gw_node), GFP_ATOMIC);
diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c
index 155565e..8caa2c7 100644
--- a/net/batman-adv/multicast.c
+++ b/net/batman-adv/multicast.c
@@ -30,6 +30,7 @@
 #include <linux/in.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
+#include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/lockdep.h>
 #include <linux/netdevice.h>
@@ -447,7 +448,7 @@ batadv_mcast_forw_ipv4_node_get(struct batadv_priv *bat_priv)
 	hlist_for_each_entry_rcu(tmp_orig_node,
 				 &bat_priv->mcast.want_all_ipv4_list,
 				 mcast_want_all_ipv4_node) {
-		if (!atomic_inc_not_zero(&tmp_orig_node->refcount))
+		if (!kref_get_unless_zero(&tmp_orig_node->refcount))
 			continue;
 
 		orig_node = tmp_orig_node;
@@ -474,7 +475,7 @@ batadv_mcast_forw_ipv6_node_get(struct batadv_priv *bat_priv)
 	hlist_for_each_entry_rcu(tmp_orig_node,
 				 &bat_priv->mcast.want_all_ipv6_list,
 				 mcast_want_all_ipv6_node) {
-		if (!atomic_inc_not_zero(&tmp_orig_node->refcount))
+		if (!kref_get_unless_zero(&tmp_orig_node->refcount))
 			continue;
 
 		orig_node = tmp_orig_node;
@@ -525,7 +526,7 @@ batadv_mcast_forw_unsnoop_node_get(struct batadv_priv *bat_priv)
 	hlist_for_each_entry_rcu(tmp_orig_node,
 				 &bat_priv->mcast.want_all_unsnoopables_list,
 				 mcast_want_all_unsnoopables_node) {
-		if (!atomic_inc_not_zero(&tmp_orig_node->refcount))
+		if (!kref_get_unless_zero(&tmp_orig_node->refcount))
 			continue;
 
 		orig_node = tmp_orig_node;
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 371fc6b..a4eb8ee 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -858,7 +858,7 @@ static struct batadv_nc_node
 	if (!nc_node)
 		return NULL;
 
-	if (!atomic_inc_not_zero(&orig_neigh_node->refcount))
+	if (!kref_get_unless_zero(&orig_neigh_node->refcount))
 		goto free;
 
 	/* Initialize nc_node */
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 8b2ef8d..eacd0e5 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -18,6 +18,7 @@
 #include "originator.h"
 #include "main.h"
 
+#include <linux/atomic.h>
 #include <linux/errno.h>
 #include <linux/etherdevice.h>
 #include <linux/fs.h>
@@ -775,14 +776,17 @@ static void batadv_orig_node_free_rcu(struct rcu_head *rcu)
 /**
  * batadv_orig_node_release - release orig_node from lists and queue for
  *  free after rcu grace period
- * @orig_node: the orig node to free
+ * @ref: kref pointer of the orig_node
  */
-static void batadv_orig_node_release(struct batadv_orig_node *orig_node)
+static void batadv_orig_node_release(struct kref *ref)
 {
 	struct hlist_node *node_tmp;
 	struct batadv_neigh_node *neigh_node;
+	struct batadv_orig_node *orig_node;
 	struct batadv_orig_ifinfo *orig_ifinfo;
 
+	orig_node = container_of(ref, struct batadv_orig_node, refcount);
+
 	spin_lock_bh(&orig_node->neigh_list_lock);
 
 	/* for all neighbors towards this originator ... */
@@ -812,8 +816,7 @@ static void batadv_orig_node_release(struct batadv_orig_node *orig_node)
  */
 void batadv_orig_node_free_ref(struct batadv_orig_node *orig_node)
 {
-	if (atomic_dec_and_test(&orig_node->refcount))
-		batadv_orig_node_release(orig_node);
+	kref_put(&orig_node->refcount, batadv_orig_node_release);
 }
 
 void batadv_originator_free(struct batadv_priv *bat_priv)
@@ -885,7 +888,8 @@ struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv,
 	batadv_nc_init_orig(orig_node);
 
 	/* extra reference for return */
-	atomic_set(&orig_node->refcount, 2);
+	kref_init(&orig_node->refcount);
+	kref_get(&orig_node->refcount);
 
 	orig_node->bat_priv = bat_priv;
 	ether_addr_copy(orig_node->orig, addr);
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 745b4e4..9950740 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -20,10 +20,10 @@
 
 #include "main.h"
 
-#include <linux/atomic.h>
 #include <linux/compiler.h>
 #include <linux/if_ether.h>
 #include <linux/jhash.h>
+#include <linux/kref.h>
 #include <linux/rculist.h>
 #include <linux/rcupdate.h>
 #include <linux/stddef.h>
@@ -115,7 +115,7 @@ batadv_orig_hash_find(struct batadv_priv *bat_priv, const void *data)
 		if (!batadv_compare_eth(orig_node, data))
 			continue;
 
-		if (!atomic_inc_not_zero(&orig_node->refcount))
+		if (!kref_get_unless_zero(&orig_node->refcount))
 			continue;
 
 		orig_node_tmp = orig_node;
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 68494a1..96b1571 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -1334,7 +1334,7 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global,
 		goto out;
 
 	INIT_HLIST_NODE(&orig_entry->list);
-	atomic_inc(&orig_node->refcount);
+	kref_get(&orig_node->refcount);
 	batadv_tt_global_size_inc(orig_node, tt_global->common.vid);
 	orig_entry->orig_node = orig_node;
 	orig_entry->ttvn = ttvn;
@@ -2097,7 +2097,7 @@ struct batadv_orig_node *batadv_transtable_search(struct batadv_priv *bat_priv,
 	/* found anything? */
 	if (best_entry)
 		orig_node = best_entry->orig_node;
-	if (orig_node && !atomic_inc_not_zero(&orig_node->refcount))
+	if (orig_node && !kref_get_unless_zero(&orig_node->refcount))
 		orig_node = NULL;
 	rcu_read_unlock();
 
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 504f3db..aea82a1 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -299,7 +299,7 @@ struct batadv_orig_node {
 	struct batadv_priv *bat_priv;
 	/* bcast_seqno_lock protects: bcast_bits & last_bcast_seqno */
 	spinlock_t bcast_seqno_lock;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 #ifdef CONFIG_BATMAN_ADV_NC
 	struct list_head in_coding_list;
-- 
cgit v0.10.2


From 92dcdf09a139e51b7b043b9443b8fd69de298dff Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sat, 16 Jan 2016 10:29:57 +0100
Subject: batman-adv: Convert batadv_tt_common_entry to kref

batman-adv uses a self-written reference implementation which is just based
on atomic_t. This is less obvious when reading the code than kref and
therefore increases the change that the reference counting will be missed.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 96b1571..1188279 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -142,7 +142,7 @@ batadv_tt_hash_find(struct batadv_hashtable *hash, const u8 *addr,
 		if (tt->vid != vid)
 			continue;
 
-		if (!atomic_inc_not_zero(&tt->refcount))
+		if (!kref_get_unless_zero(&tt->refcount))
 			continue;
 
 		tt_tmp = tt;
@@ -203,25 +203,59 @@ batadv_tt_global_hash_find(struct batadv_priv *bat_priv, const u8 *addr,
 	return tt_global_entry;
 }
 
+/**
+ * batadv_tt_local_entry_release - release tt_local_entry from lists and queue
+ *  for free after rcu grace period
+ * @ref: kref pointer of the nc_node
+ */
+static void batadv_tt_local_entry_release(struct kref *ref)
+{
+	struct batadv_tt_local_entry *tt_local_entry;
+
+	tt_local_entry = container_of(ref, struct batadv_tt_local_entry,
+				      common.refcount);
+
+	kfree_rcu(tt_local_entry, common.rcu);
+}
+
+/**
+ * batadv_tt_local_entry_free_ref - decrement the tt_local_entry refcounter and
+ *  possibly release it
+ * @tt_local_entry: tt_local_entry to be free'd
+ */
 static void
 batadv_tt_local_entry_free_ref(struct batadv_tt_local_entry *tt_local_entry)
 {
-	if (atomic_dec_and_test(&tt_local_entry->common.refcount))
-		kfree_rcu(tt_local_entry, common.rcu);
+	kref_put(&tt_local_entry->common.refcount,
+		 batadv_tt_local_entry_release);
 }
 
 /**
- * batadv_tt_global_entry_free_ref - decrement the refcounter for a
- *  tt_global_entry and possibly free it
- * @tt_global_entry: the object to free
+ * batadv_tt_global_entry_release - release tt_global_entry from lists and queue
+ *  for free after rcu grace period
+ * @ref: kref pointer of the nc_node
+ */
+static void batadv_tt_global_entry_release(struct kref *ref)
+{
+	struct batadv_tt_global_entry *tt_global_entry;
+
+	tt_global_entry = container_of(ref, struct batadv_tt_global_entry,
+				       common.refcount);
+
+	batadv_tt_global_del_orig_list(tt_global_entry);
+	kfree_rcu(tt_global_entry, common.rcu);
+}
+
+/**
+ * batadv_tt_global_entry_free_ref - decrement the tt_global_entry refcounter
+ *  and possibly release it
+ * @tt_global_entry: tt_global_entry to be free'd
  */
 static void
 batadv_tt_global_entry_free_ref(struct batadv_tt_global_entry *tt_global_entry)
 {
-	if (atomic_dec_and_test(&tt_global_entry->common.refcount)) {
-		batadv_tt_global_del_orig_list(tt_global_entry);
-		kfree_rcu(tt_global_entry, common.rcu);
-	}
+	kref_put(&tt_global_entry->common.refcount,
+		 batadv_tt_global_entry_release);
 }
 
 /**
@@ -633,7 +667,8 @@ bool batadv_tt_local_add(struct net_device *soft_iface, const u8 *addr,
 	tt_local->common.vid = vid;
 	if (batadv_is_wifi_netdev(in_dev))
 		tt_local->common.flags |= BATADV_TT_CLIENT_WIFI;
-	atomic_set(&tt_local->common.refcount, 2);
+	kref_init(&tt_local->common.refcount);
+	kref_get(&tt_local->common.refcount);
 	tt_local->last_seen = jiffies;
 	tt_local->common.added_at = tt_local->last_seen;
 
@@ -1415,7 +1450,8 @@ static bool batadv_tt_global_add(struct batadv_priv *bat_priv,
 		 */
 		if (flags & BATADV_TT_CLIENT_ROAM)
 			tt_global_entry->roam_at = jiffies;
-		atomic_set(&common->refcount, 2);
+		kref_init(&common->refcount);
+		kref_get(&common->refcount);
 		common->added_at = jiffies;
 
 		INIT_HLIST_HEAD(&tt_global_entry->orig_list);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index aea82a1..612de23 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -972,7 +972,7 @@ struct batadv_tt_common_entry {
 	struct hlist_node hash_entry;
 	u16 flags;
 	unsigned long added_at;
-	atomic_t refcount;
+	struct kref refcount;
 	struct rcu_head rcu;
 };
 
-- 
cgit v0.10.2


From 622190669403fb7a7557a8b618e57c0a3f703056 Mon Sep 17 00:00:00 2001
From: Ajit Khaparde <ajit.khaparde@broadcom.com>
Date: Wed, 10 Feb 2016 22:45:53 +0530
Subject: be2net: Request RSS capability of Rx interface depending on number of
 Rx rings

Currently we request RSS capability even if a single Rx ring is created.
As a result in few cases we unnecessarily consume an RSS capable interface
which is a limited resource in the chip.
This patch enables RSS on an interface only if more than one Rx ring
is created.

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 7d51d47..33cbbde 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -1513,34 +1513,25 @@ int be_cmd_if_create(struct be_adapter *adapter, u32 cap_flags, u32 en_flags,
 	return status;
 }
 
-/* Uses MCCQ */
+/* Uses MCCQ if available else MBOX */
 int be_cmd_if_destroy(struct be_adapter *adapter, int interface_id, u32 domain)
 {
-	struct be_mcc_wrb *wrb;
+	struct be_mcc_wrb wrb = {0};
 	struct be_cmd_req_if_destroy *req;
 	int status;
 
 	if (interface_id == -1)
 		return 0;
 
-	spin_lock_bh(&adapter->mcc_lock);
-
-	wrb = wrb_from_mccq(adapter);
-	if (!wrb) {
-		status = -EBUSY;
-		goto err;
-	}
-	req = embedded_payload(wrb);
+	req = embedded_payload(&wrb);
 
 	be_wrb_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
 			       OPCODE_COMMON_NTWK_INTERFACE_DESTROY,
-			       sizeof(*req), wrb, NULL);
+			       sizeof(*req), &wrb, NULL);
 	req->hdr.domain = domain;
 	req->interface_id = cpu_to_le32(interface_id);
 
-	status = be_mcc_notify_wait(adapter);
-err:
-	spin_unlock_bh(&adapter->mcc_lock);
+	status = be_cmd_notify_wait(adapter, &wrb);
 	return status;
 }
 
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 9c1fc9d..e4fc874 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -3365,6 +3365,7 @@ done:
 
 static void be_rx_qs_destroy(struct be_adapter *adapter)
 {
+	struct rss_info *rss = &adapter->rss_info;
 	struct be_queue_info *q;
 	struct be_rx_obj *rxo;
 	int i;
@@ -3391,6 +3392,12 @@ static void be_rx_qs_destroy(struct be_adapter *adapter)
 		}
 		be_queue_free(adapter, q);
 	}
+
+	if (rss->rss_flags) {
+		rss->rss_flags = RSS_ENABLE_NONE;
+		be_cmd_rss_config(adapter, rss->rsstable, rss->rss_flags,
+				  128, rss->rss_hkey);
+	}
 }
 
 static void be_disable_if_filters(struct be_adapter *adapter)
@@ -3511,20 +3518,21 @@ static int be_rx_qs_create(struct be_adapter *adapter)
 		if (!BEx_chip(adapter))
 			rss->rss_flags |= RSS_ENABLE_UDP_IPV4 |
 				RSS_ENABLE_UDP_IPV6;
+
+		netdev_rss_key_fill(rss_key, RSS_HASH_KEY_LEN);
+		rc = be_cmd_rss_config(adapter, rss->rsstable, rss->rss_flags,
+				       RSS_INDIR_TABLE_LEN, rss_key);
+		if (rc) {
+			rss->rss_flags = RSS_ENABLE_NONE;
+			return rc;
+		}
+
+		memcpy(rss->rss_hkey, rss_key, RSS_HASH_KEY_LEN);
 	} else {
 		/* Disable RSS, if only default RX Q is created */
 		rss->rss_flags = RSS_ENABLE_NONE;
 	}
 
-	netdev_rss_key_fill(rss_key, RSS_HASH_KEY_LEN);
-	rc = be_cmd_rss_config(adapter, rss->rsstable, rss->rss_flags,
-			       RSS_INDIR_TABLE_LEN, rss_key);
-	if (rc) {
-		rss->rss_flags = RSS_ENABLE_NONE;
-		return rc;
-	}
-
-	memcpy(rss->rss_hkey, rss_key, RSS_HASH_KEY_LEN);
 
 	/* Post 1 less than RXQ-len to avoid head being equal to tail,
 	 * which is a queue empty condition
@@ -4306,6 +4314,23 @@ err:
 	return status;
 }
 
+static int be_if_create(struct be_adapter *adapter)
+{
+	u32 en_flags = BE_IF_FLAGS_RSS | BE_IF_FLAGS_DEFQ_RSS;
+	u32 cap_flags = be_if_cap_flags(adapter);
+	int status;
+
+	if (adapter->cfg_num_qs == 1)
+		cap_flags &= ~(BE_IF_FLAGS_DEFQ_RSS | BE_IF_FLAGS_RSS);
+
+	en_flags &= cap_flags;
+	/* will enable all the needed filter flags in be_open() */
+	status = be_cmd_if_create(adapter, be_if_cap_flags(adapter), en_flags,
+				  &adapter->if_handle, 0);
+
+	return status;
+}
+
 int be_update_queues(struct be_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
@@ -4323,6 +4348,9 @@ int be_update_queues(struct be_adapter *adapter)
 		be_msix_disable(adapter);
 
 	be_clear_queues(adapter);
+	status = be_cmd_if_destroy(adapter, adapter->if_handle,  0);
+	if (status)
+		return status;
 
 	if (!msix_enabled(adapter)) {
 		status = be_msix_enable(adapter);
@@ -4330,6 +4358,10 @@ int be_update_queues(struct be_adapter *adapter)
 			return status;
 	}
 
+	status = be_if_create(adapter);
+	if (status)
+		return status;
+
 	status = be_setup_queues(adapter);
 	if (status)
 		return status;
@@ -4394,7 +4426,6 @@ static int be_func_init(struct be_adapter *adapter)
 static int be_setup(struct be_adapter *adapter)
 {
 	struct device *dev = &adapter->pdev->dev;
-	u32 en_flags;
 	int status;
 
 	status = be_func_init(adapter);
@@ -4427,10 +4458,7 @@ static int be_setup(struct be_adapter *adapter)
 		goto err;
 
 	/* will enable all the needed filter flags in be_open() */
-	en_flags = BE_IF_FLAGS_RSS | BE_IF_FLAGS_DEFQ_RSS;
-	en_flags = en_flags & be_if_cap_flags(adapter);
-	status = be_cmd_if_create(adapter, be_if_cap_flags(adapter), en_flags,
-				  &adapter->if_handle, 0);
+	status = be_if_create(adapter);
 	if (status)
 		goto err;
 
@@ -4803,7 +4831,7 @@ static void be_netdev_init(struct net_device *netdev)
 	netdev->hw_features |= NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
 		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM |
 		NETIF_F_HW_VLAN_CTAG_TX;
-	if (be_multi_rxq(adapter))
+	if ((be_if_cap_flags(adapter) & BE_IF_FLAGS_RSS))
 		netdev->hw_features |= NETIF_F_RXHASH;
 
 	netdev->features |= netdev->hw_features |
-- 
cgit v0.10.2


From 51d1f98a136fcb16fa3304c2386d9a5202335a63 Mon Sep 17 00:00:00 2001
From: Ajit Khaparde <ajit.khaparde@broadcom.com>
Date: Wed, 10 Feb 2016 22:45:54 +0530
Subject: be2net: Interpret and log new data that's added to the port
 misconfigure async event

>From FW version 11.0. onwards, the PORT_MISCONFIG event generated by the FW
will carry more information about the event in the "data_word1"
and "data_word2" fields. This patch adds support in the driver to parse the
new information and log it accordingly. This patch also changes some of the
messages that are being logged currently.

Signed-off-by: Suresh Reddy <suresh.reddy@broadcom.com>
Signed-off-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index 515e206..ab24f84 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -115,6 +115,8 @@
 #define	RSS_INDIR_TABLE_LEN	128
 #define RSS_HASH_KEY_LEN	40
 
+#define BE_UNKNOWN_PHY_STATE	0xFF
+
 struct be_dma_mem {
 	void *va;
 	dma_addr_t dma;
@@ -390,7 +392,7 @@ enum vf_state {
 #define BE_FLAGS_QNQ_ASYNC_EVT_RCVD		BIT(7)
 #define BE_FLAGS_VXLAN_OFFLOADS			BIT(8)
 #define BE_FLAGS_SETUP_DONE			BIT(9)
-#define BE_FLAGS_EVT_INCOMPATIBLE_SFP		BIT(10)
+#define BE_FLAGS_PHY_MISCONFIGURED		BIT(10)
 #define BE_FLAGS_ERR_DETECTION_SCHEDULED	BIT(11)
 #define BE_FLAGS_OS2BMC				BIT(12)
 
@@ -603,6 +605,7 @@ struct be_adapter {
 	u32 bmc_filt_mask;
 	u32 fat_dump_len;
 	u16 serial_num[CNTL_SERIAL_NUM_WORDS];
+	u8 phy_state; /* state of sfp optics (functional, faulted, etc.,) */
 };
 
 #define be_physfn(adapter)		(!adapter->virtfn)
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 33cbbde..66fa214 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -19,19 +19,25 @@
 #include "be.h"
 #include "be_cmds.h"
 
-static char *be_port_misconfig_evt_desc[] = {
-	"A valid SFP module detected",
-	"Optics faulted/ incorrectly installed/ not installed.",
-	"Optics of two types installed.",
-	"Incompatible optics.",
-	"Unknown port SFP status"
+char *be_misconfig_evt_port_state[] = {
+	"Physical Link is functional",
+	"Optics faulted/incorrectly installed/not installed - Reseat optics. If issue not resolved, replace.",
+	"Optics of two types installed – Remove one optic or install matching pair of optics.",
+	"Incompatible optics – Replace with compatible optics for card to function.",
+	"Unqualified optics – Replace with Avago optics for Warranty and Technical Support.",
+	"Uncertified optics – Replace with Avago-certified optics to enable link operation."
 };
 
-static char *be_port_misconfig_remedy_desc[] = {
-	"",
-	"Reseat optics. If issue not resolved, replace",
-	"Remove one optic or install matching pair of optics",
-	"Replace with compatible optics for card to function",
+static char *be_port_misconfig_evt_severity[] = {
+	"KERN_WARN",
+	"KERN_INFO",
+	"KERN_ERR",
+	"KERN_WARN"
+};
+
+static char *phy_state_oper_desc[] = {
+	"Link is non-operational",
+	"Link is operational",
 	""
 };
 
@@ -297,22 +303,56 @@ static void be_async_port_misconfig_event_process(struct be_adapter *adapter,
 {
 	struct be_async_event_misconfig_port *evt =
 			(struct be_async_event_misconfig_port *)compl;
-	u32 sfp_mismatch_evt = le32_to_cpu(evt->event_data_word1);
+	u32 sfp_misconfig_evt_word1 = le32_to_cpu(evt->event_data_word1);
+	u32 sfp_misconfig_evt_word2 = le32_to_cpu(evt->event_data_word2);
+	u8 phy_oper_state = PHY_STATE_OPER_MSG_NONE;
 	struct device *dev = &adapter->pdev->dev;
-	u8 port_misconfig_evt;
+	u8 msg_severity = DEFAULT_MSG_SEVERITY;
+	u8 phy_state_info;
+	u8 new_phy_state;
+
+	new_phy_state =
+		(sfp_misconfig_evt_word1 >> (adapter->hba_port_num * 8)) & 0xff;
+
+	if (new_phy_state == adapter->phy_state)
+		return;
+
+	adapter->phy_state = new_phy_state;
 
-	port_misconfig_evt =
-		((sfp_mismatch_evt >> (adapter->hba_port_num * 8)) & 0xff);
+	/* for older fw that doesn't populate link effect data */
+	if (!sfp_misconfig_evt_word2)
+		goto log_message;
 
+	phy_state_info =
+		(sfp_misconfig_evt_word2 >> (adapter->hba_port_num * 8)) & 0xff;
+
+	if (phy_state_info & PHY_STATE_INFO_VALID) {
+		msg_severity = (phy_state_info & PHY_STATE_MSG_SEVERITY) >> 1;
+
+		if (be_phy_unqualified(new_phy_state))
+			phy_oper_state = (phy_state_info & PHY_STATE_OPER);
+	}
+
+log_message:
 	/* Log an error message that would allow a user to determine
 	 * whether the SFPs have an issue
 	 */
-	dev_info(dev, "Port %c: %s %s", adapter->port_name,
-		 be_port_misconfig_evt_desc[port_misconfig_evt],
-		 be_port_misconfig_remedy_desc[port_misconfig_evt]);
-
-	if (port_misconfig_evt == INCOMPATIBLE_SFP)
-		adapter->flags |= BE_FLAGS_EVT_INCOMPATIBLE_SFP;
+	if (be_phy_state_unknown(new_phy_state))
+		dev_printk(be_port_misconfig_evt_severity[msg_severity], dev,
+			   "Port %c: Unrecognized Optics state: 0x%x. %s",
+			   adapter->port_name,
+			   new_phy_state,
+			   phy_state_oper_desc[phy_oper_state]);
+	else
+		dev_printk(be_port_misconfig_evt_severity[msg_severity], dev,
+			   "Port %c: %s %s",
+			   adapter->port_name,
+			   be_misconfig_evt_port_state[new_phy_state],
+			   phy_state_oper_desc[phy_oper_state]);
+
+	/* Log Vendor name and part no. if a misconfigured SFP is detected */
+	if (be_phy_misconfigured(new_phy_state))
+		adapter->flags |= BE_FLAGS_PHY_MISCONFIGURED;
 }
 
 /* Grp5 CoS Priority evt */
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index f260ef3..0b9f741 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -176,10 +176,53 @@ struct be_async_event_qnq {
 	u32 flags;
 } __packed;
 
-#define INCOMPATIBLE_SFP		0x3
+enum {
+	BE_PHY_FUNCTIONAL	= 0,
+	BE_PHY_NOT_PRESENT	= 1,
+	BE_PHY_DIFF_MEDIA	= 2,
+	BE_PHY_INCOMPATIBLE	= 3,
+	BE_PHY_UNQUALIFIED	= 4,
+	BE_PHY_UNCERTIFIED	= 5
+};
+
+#define PHY_STATE_MSG_SEVERITY		0x6
+#define PHY_STATE_OPER			0x1
+#define PHY_STATE_INFO_VALID		0x80
+#define	PHY_STATE_OPER_MSG_NONE		0x2
+#define DEFAULT_MSG_SEVERITY		0x1
+
+#define be_phy_state_unknown(phy_state) (phy_state > BE_PHY_UNCERTIFIED)
+#define be_phy_unqualified(phy_state)				\
+			(phy_state == BE_PHY_UNQUALIFIED ||	\
+			 phy_state == BE_PHY_UNCERTIFIED)
+#define be_phy_misconfigured(phy_state)				\
+			(phy_state == BE_PHY_INCOMPATIBLE ||	\
+			 phy_state == BE_PHY_UNQUALIFIED ||	\
+			 phy_state == BE_PHY_UNCERTIFIED)
+
+extern  char *be_misconfig_evt_port_state[];
+
 /* async event indicating misconfigured port */
 struct be_async_event_misconfig_port {
+ /* DATA_WORD1:
+  * phy state of port 0: bits 7 - 0
+  * phy state of port 1: bits 15 - 8
+  * phy state of port 2: bits 23 - 16
+  * phy state of port 3: bits 31 - 24
+  */
 	u32 event_data_word1;
+ /* DATA_WORD2:
+  * phy state info of port 0: bits 7 - 0
+  * phy state info of port 1: bits 15 - 8
+  * phy state info of port 2: bits 23 - 16
+  * phy state info of port 3: bits 31 - 24
+  *
+  * PHY STATE INFO:
+  * Link operability	 :bit 0
+  * Message severity	 :bit 2 - 1
+  * Rsvd			 :bits 6 - 3
+  * phy state info valid	 :bit 7
+  */
 	u32 event_data_word2;
 	u32 rsvd0;
 	u32 flags;
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index e4fc874..4ecd8c2 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -4089,6 +4089,7 @@ static void be_setup_init(struct be_adapter *adapter)
 	adapter->if_handle = -1;
 	adapter->be3_native = false;
 	adapter->if_flags = 0;
+	adapter->phy_state = BE_UNKNOWN_PHY_STATE;
 	if (be_physfn(adapter))
 		adapter->cmd_privileges = MAX_PRIVILEGES;
 	else
@@ -4961,11 +4962,13 @@ static void be_log_sfp_info(struct be_adapter *adapter)
 	status = be_cmd_query_sfp_info(adapter);
 	if (!status) {
 		dev_err(&adapter->pdev->dev,
-			"Unqualified SFP+ detected on %c from %s part no: %s",
-			adapter->port_name, adapter->phy.vendor_name,
+			"Port %c: %s Vendor: %s part no: %s",
+			adapter->port_name,
+			be_misconfig_evt_port_state[adapter->phy_state],
+			adapter->phy.vendor_name,
 			adapter->phy.vendor_pn);
 	}
-	adapter->flags &= ~BE_FLAGS_EVT_INCOMPATIBLE_SFP;
+	adapter->flags &= ~BE_FLAGS_PHY_MISCONFIGURED;
 }
 
 static void be_worker(struct work_struct *work)
@@ -5009,7 +5012,7 @@ static void be_worker(struct work_struct *work)
 	if (!skyhawk_chip(adapter))
 		be_eqd_update(adapter, false);
 
-	if (adapter->flags & BE_FLAGS_EVT_INCOMPATIBLE_SFP)
+	if (adapter->flags & BE_FLAGS_PHY_MISCONFIGURED)
 		be_log_sfp_info(adapter);
 
 reschedule:
-- 
cgit v0.10.2


From 086c653f5862591a9cfe2386f5650d03adacc33a Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:35 -0500
Subject: sock: struct proto hash function may error

In order to support fast reuseport lookups in TCP, the hash function
defined in struct proto must be capable of returning an error code.
This patch changes the function signature of all related hash functions
to return an integer and handles or propagates this return value at
all call sites.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index de2e3ad..554440e 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -208,7 +208,7 @@ void inet_hashinfo_init(struct inet_hashinfo *h);
 bool inet_ehash_insert(struct sock *sk, struct sock *osk);
 bool inet_ehash_nolisten(struct sock *sk, struct sock *osk);
 void __inet_hash(struct sock *sk, struct sock *osk);
-void inet_hash(struct sock *sk);
+int inet_hash(struct sock *sk);
 void inet_unhash(struct sock *sk);
 
 struct sock *__inet_lookup_listener(struct net *net,
diff --git a/include/net/phonet/phonet.h b/include/net/phonet/phonet.h
index 68e5097..039cc29 100644
--- a/include/net/phonet/phonet.h
+++ b/include/net/phonet/phonet.h
@@ -51,7 +51,7 @@ void pn_sock_init(void);
 struct sock *pn_find_sock_by_sa(struct net *net, const struct sockaddr_pn *sa);
 void pn_deliver_sock_broadcast(struct net *net, struct sk_buff *skb);
 void phonet_get_local_port_range(int *min, int *max);
-void pn_sock_hash(struct sock *sk);
+int pn_sock_hash(struct sock *sk);
 void pn_sock_unhash(struct sock *sk);
 int pn_sock_get_port(struct sock *sk, unsigned short sport);
 
diff --git a/include/net/ping.h b/include/net/ping.h
index ac80cb4..5fd7cc2 100644
--- a/include/net/ping.h
+++ b/include/net/ping.h
@@ -65,7 +65,7 @@ struct pingfakehdr {
 };
 
 int  ping_get_port(struct sock *sk, unsigned short ident);
-void ping_hash(struct sock *sk);
+int ping_hash(struct sock *sk);
 void ping_unhash(struct sock *sk);
 
 int  ping_init_sock(struct sock *sk);
diff --git a/include/net/raw.h b/include/net/raw.h
index 6a40c65..3e78900 100644
--- a/include/net/raw.h
+++ b/include/net/raw.h
@@ -57,7 +57,7 @@ int raw_seq_open(struct inode *ino, struct file *file,
 
 #endif
 
-void raw_hash_sk(struct sock *sk);
+int raw_hash_sk(struct sock *sk);
 void raw_unhash_sk(struct sock *sk);
 
 struct raw_sock {
diff --git a/include/net/sock.h b/include/net/sock.h
index f5ea148..255d3e0 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -984,7 +984,7 @@ struct proto {
 	void		(*release_cb)(struct sock *sk);
 
 	/* Keeping track of sk's, looking them up, and port selection methods. */
-	void			(*hash)(struct sock *sk);
+	int			(*hash)(struct sock *sk);
 	void			(*unhash)(struct sock *sk);
 	void			(*rehash)(struct sock *sk);
 	int			(*get_port)(struct sock *sk, unsigned short snum);
@@ -1194,10 +1194,10 @@ static inline void sock_prot_inuse_add(struct net *net, struct proto *prot,
 /* With per-bucket locks this operation is not-atomic, so that
  * this version is not worse.
  */
-static inline void __sk_prot_rehash(struct sock *sk)
+static inline int __sk_prot_rehash(struct sock *sk)
 {
 	sk->sk_prot->unhash(sk);
-	sk->sk_prot->hash(sk);
+	return sk->sk_prot->hash(sk);
 }
 
 void sk_prot_clear_portaddr_nulls(struct sock *sk, int size);
diff --git a/include/net/udp.h b/include/net/udp.h
index 2842541..92927f7 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -177,9 +177,10 @@ static inline struct udphdr *udp_gro_udphdr(struct sk_buff *skb)
 }
 
 /* hash routines shared between UDPv4/6 and UDP-Litev4/6 */
-static inline void udp_lib_hash(struct sock *sk)
+static inline int udp_lib_hash(struct sock *sk)
 {
 	BUG();
+	return 0;
 }
 
 void udp_lib_unhash(struct sock *sk);
diff --git a/net/ieee802154/socket.c b/net/ieee802154/socket.c
index a548be2..e0bd013 100644
--- a/net/ieee802154/socket.c
+++ b/net/ieee802154/socket.c
@@ -182,12 +182,14 @@ static int ieee802154_sock_ioctl(struct socket *sock, unsigned int cmd,
 static HLIST_HEAD(raw_head);
 static DEFINE_RWLOCK(raw_lock);
 
-static void raw_hash(struct sock *sk)
+static int raw_hash(struct sock *sk)
 {
 	write_lock_bh(&raw_lock);
 	sk_add_node(sk, &raw_head);
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock_bh(&raw_lock);
+
+	return 0;
 }
 
 static void raw_unhash(struct sock *sk)
@@ -462,12 +464,14 @@ static inline struct dgram_sock *dgram_sk(const struct sock *sk)
 	return container_of(sk, struct dgram_sock, sk);
 }
 
-static void dgram_hash(struct sock *sk)
+static int dgram_hash(struct sock *sk)
 {
 	write_lock_bh(&dgram_lock);
 	sk_add_node(sk, &dgram_head);
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock_bh(&dgram_lock);
+
+	return 0;
 }
 
 static void dgram_unhash(struct sock *sk)
@@ -1026,8 +1030,13 @@ static int ieee802154_create(struct net *net, struct socket *sock,
 	/* Checksums on by default */
 	sock_set_flag(sk, SOCK_ZAPPED);
 
-	if (sk->sk_prot->hash)
-		sk->sk_prot->hash(sk);
+	if (sk->sk_prot->hash) {
+		rc = sk->sk_prot->hash(sk);
+		if (rc) {
+			sk_common_release(sk);
+			goto out;
+		}
+	}
 
 	if (sk->sk_prot->init) {
 		rc = sk->sk_prot->init(sk);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 5c5db66..eade66d 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -370,7 +370,11 @@ lookup_protocol:
 		 */
 		inet->inet_sport = htons(inet->inet_num);
 		/* Add to protocol hash chains. */
-		sk->sk_prot->hash(sk);
+		err = sk->sk_prot->hash(sk);
+		if (err) {
+			sk_common_release(sk);
+			goto out;
+		}
 	}
 
 	if (sk->sk_prot->init) {
@@ -1142,8 +1146,7 @@ static int inet_sk_reselect_saddr(struct sock *sk)
 	 * Besides that, it does not check for connection
 	 * uniqueness. Wait for troubles.
 	 */
-	__sk_prot_rehash(sk);
-	return 0;
+	return __sk_prot_rehash(sk);
 }
 
 int inet_sk_rebuild_header(struct sock *sk)
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 9b17c179..12c8d38 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -734,6 +734,7 @@ int inet_csk_listen_start(struct sock *sk, int backlog)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct inet_sock *inet = inet_sk(sk);
+	int err = -EADDRINUSE;
 
 	reqsk_queue_alloc(&icsk->icsk_accept_queue);
 
@@ -751,13 +752,14 @@ int inet_csk_listen_start(struct sock *sk, int backlog)
 		inet->inet_sport = htons(inet->inet_num);
 
 		sk_dst_reset(sk);
-		sk->sk_prot->hash(sk);
+		err = sk->sk_prot->hash(sk);
 
-		return 0;
+		if (likely(!err))
+			return 0;
 	}
 
 	sk->sk_state = TCP_CLOSE;
-	return -EADDRINUSE;
+	return err;
 }
 EXPORT_SYMBOL_GPL(inet_csk_listen_start);
 
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index ccc5980..b6023b7 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -468,13 +468,15 @@ void __inet_hash(struct sock *sk, struct sock *osk)
 }
 EXPORT_SYMBOL(__inet_hash);
 
-void inet_hash(struct sock *sk)
+int inet_hash(struct sock *sk)
 {
 	if (sk->sk_state != TCP_CLOSE) {
 		local_bh_disable();
 		__inet_hash(sk, NULL);
 		local_bh_enable();
 	}
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(inet_hash);
 
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index c117b21..f6f93fc 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -145,10 +145,12 @@ fail:
 }
 EXPORT_SYMBOL_GPL(ping_get_port);
 
-void ping_hash(struct sock *sk)
+int ping_hash(struct sock *sk)
 {
 	pr_debug("ping_hash(sk->port=%u)\n", inet_sk(sk)->inet_num);
 	BUG(); /* "Please do not press this button again." */
+
+	return 0;
 }
 
 void ping_unhash(struct sock *sk)
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index bc35f18..d635251 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -93,7 +93,7 @@ static struct raw_hashinfo raw_v4_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(raw_v4_hashinfo.lock),
 };
 
-void raw_hash_sk(struct sock *sk)
+int raw_hash_sk(struct sock *sk)
 {
 	struct raw_hashinfo *h = sk->sk_prot->h.raw_hash;
 	struct hlist_head *head;
@@ -104,6 +104,8 @@ void raw_hash_sk(struct sock *sk)
 	sk_add_node(sk, head);
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock_bh(&h->lock);
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(raw_hash_sk);
 
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 9f5137c..b11c37c 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -235,7 +235,11 @@ lookup_protocol:
 		 * creation time automatically shares.
 		 */
 		inet->inet_sport = htons(inet->inet_num);
-		sk->sk_prot->hash(sk);
+		err = sk->sk_prot->hash(sk);
+		if (err) {
+			sk_common_release(sk);
+			goto out;
+		}
 	}
 	if (sk->sk_prot->init) {
 		err = sk->sk_prot->init(sk);
diff --git a/net/phonet/socket.c b/net/phonet/socket.c
index d575ef4..ffd5f22 100644
--- a/net/phonet/socket.c
+++ b/net/phonet/socket.c
@@ -140,13 +140,15 @@ void pn_deliver_sock_broadcast(struct net *net, struct sk_buff *skb)
 	rcu_read_unlock();
 }
 
-void pn_sock_hash(struct sock *sk)
+int pn_sock_hash(struct sock *sk)
 {
 	struct hlist_head *hlist = pn_hash_list(pn_sk(sk)->sobject);
 
 	mutex_lock(&pnsocks.lock);
 	sk_add_node_rcu(sk, hlist);
 	mutex_unlock(&pnsocks.lock);
+
+	return 0;
 }
 EXPORT_SYMBOL(pn_sock_hash);
 
@@ -200,7 +202,7 @@ static int pn_socket_bind(struct socket *sock, struct sockaddr *addr, int len)
 	pn->resource = spn->spn_resource;
 
 	/* Enable RX on the socket */
-	sk->sk_prot->hash(sk);
+	err = sk->sk_prot->hash(sk);
 out_port:
 	mutex_unlock(&port_mutex);
 out:
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 5ca2ebf..6427b9d 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -6101,9 +6101,10 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname,
 	return retval;
 }
 
-static void sctp_hash(struct sock *sk)
+static int sctp_hash(struct sock *sk)
 {
 	/* STUB */
+	return 0;
 }
 
 static void sctp_unhash(struct sock *sk)
-- 
cgit v0.10.2


From 496611d7b5eaf59c03440c8f2def1d9988ad2459 Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:36 -0500
Subject: inet: create IPv6-equivalent inet_hash function

In order to support fast lookups for TCP sockets with SO_REUSEPORT,
the function that adds sockets to the listening hash set needs
to be able to check receive address equality.  Since this equality
check is different for IPv4 and IPv6, we will need two different
socket hashing functions.

This patch adds inet6_hash identical to the existing inet_hash function
and updates the appropriate references.  A following patch will
differentiate the two by passing different comparison functions to
__inet_hash.

Additionally, in order to use the IPv6 address equality function from
inet6_hashtables (which is compiled as a built-in object when IPv6 is
enabled) it also needs to be in a built-in object file as well.  This
moves ipv6_rcv_saddr_equal into inet_hashtables to accomplish this.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h
index 7ff588c..b3c28a9 100644
--- a/include/net/inet6_hashtables.h
+++ b/include/net/inet6_hashtables.h
@@ -96,6 +96,8 @@ struct sock *inet6_lookup(struct net *net, struct inet_hashinfo *hashinfo,
 			  const struct in6_addr *saddr, const __be16 sport,
 			  const struct in6_addr *daddr, const __be16 dport,
 			  const int dif);
+
+int inet6_hash(struct sock *sk);
 #endif /* IS_ENABLED(CONFIG_IPV6) */
 
 #define INET6_MATCH(__sk, __net, __saddr, __daddr, __ports, __dif)	\
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 9c6d050..90a8269 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -993,7 +993,7 @@ static struct proto dccp_v6_prot = {
 	.sendmsg	   = dccp_sendmsg,
 	.recvmsg	   = dccp_recvmsg,
 	.backlog_rcv	   = dccp_v6_do_rcv,
-	.hash		   = inet_hash,
+	.hash		   = inet6_hash,
 	.unhash		   = inet_unhash,
 	.accept		   = inet_csk_accept,
 	.get_port	   = inet_csk_get_port,
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 21ace5a..072653d 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -274,3 +274,59 @@ int inet6_hash_connect(struct inet_timewait_death_row *death_row,
 				   __inet6_check_established);
 }
 EXPORT_SYMBOL_GPL(inet6_hash_connect);
+
+int inet6_hash(struct sock *sk)
+{
+	if (sk->sk_state != TCP_CLOSE) {
+		local_bh_disable();
+		__inet_hash(sk, NULL);
+		local_bh_enable();
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(inet6_hash);
+
+/* match_wildcard == true:  IPV6_ADDR_ANY equals to any IPv6 addresses if IPv6
+ *                          only, and any IPv4 addresses if not IPv6 only
+ * match_wildcard == false: addresses must be exactly the same, i.e.
+ *                          IPV6_ADDR_ANY only equals to IPV6_ADDR_ANY,
+ *                          and 0.0.0.0 equals to 0.0.0.0 only
+ */
+int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
+			 bool match_wildcard)
+{
+	const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2);
+	int sk2_ipv6only = inet_v6_ipv6only(sk2);
+	int addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr);
+	int addr_type2 = sk2_rcv_saddr6 ? ipv6_addr_type(sk2_rcv_saddr6) : IPV6_ADDR_MAPPED;
+
+	/* if both are mapped, treat as IPv4 */
+	if (addr_type == IPV6_ADDR_MAPPED && addr_type2 == IPV6_ADDR_MAPPED) {
+		if (!sk2_ipv6only) {
+			if (sk->sk_rcv_saddr == sk2->sk_rcv_saddr)
+				return 1;
+			if (!sk->sk_rcv_saddr || !sk2->sk_rcv_saddr)
+				return match_wildcard;
+		}
+		return 0;
+	}
+
+	if (addr_type == IPV6_ADDR_ANY && addr_type2 == IPV6_ADDR_ANY)
+		return 1;
+
+	if (addr_type2 == IPV6_ADDR_ANY && match_wildcard &&
+	    !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED))
+		return 1;
+
+	if (addr_type == IPV6_ADDR_ANY && match_wildcard &&
+	    !(ipv6_only_sock(sk) && addr_type2 == IPV6_ADDR_MAPPED))
+		return 1;
+
+	if (sk2_rcv_saddr6 &&
+	    ipv6_addr_equal(&sk->sk_v6_rcv_saddr, sk2_rcv_saddr6))
+		return 1;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ipv6_rcv_saddr_equal);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 006396e..d72bcfb 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1865,7 +1865,7 @@ struct proto tcpv6_prot = {
 	.sendpage		= tcp_sendpage,
 	.backlog_rcv		= tcp_v6_do_rcv,
 	.release_cb		= tcp_release_cb,
-	.hash			= inet_hash,
+	.hash			= inet6_hash,
 	.unhash			= inet_unhash,
 	.get_port		= inet_csk_get_port,
 	.enter_memory_pressure	= tcp_enter_memory_pressure,
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 22e28a4..ac4e7e0 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -37,6 +37,7 @@
 #include <linux/slab.h>
 #include <asm/uaccess.h>
 
+#include <net/addrconf.h>
 #include <net/ndisc.h>
 #include <net/protocol.h>
 #include <net/transp_v6.h>
@@ -77,49 +78,6 @@ static u32 udp6_ehashfn(const struct net *net,
 			       udp_ipv6_hash_secret + net_hash_mix(net));
 }
 
-/* match_wildcard == true:  IPV6_ADDR_ANY equals to any IPv6 addresses if IPv6
- *                          only, and any IPv4 addresses if not IPv6 only
- * match_wildcard == false: addresses must be exactly the same, i.e.
- *                          IPV6_ADDR_ANY only equals to IPV6_ADDR_ANY,
- *                          and 0.0.0.0 equals to 0.0.0.0 only
- */
-int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
-			 bool match_wildcard)
-{
-	const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2);
-	int sk2_ipv6only = inet_v6_ipv6only(sk2);
-	int addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr);
-	int addr_type2 = sk2_rcv_saddr6 ? ipv6_addr_type(sk2_rcv_saddr6) : IPV6_ADDR_MAPPED;
-
-	/* if both are mapped, treat as IPv4 */
-	if (addr_type == IPV6_ADDR_MAPPED && addr_type2 == IPV6_ADDR_MAPPED) {
-		if (!sk2_ipv6only) {
-			if (sk->sk_rcv_saddr == sk2->sk_rcv_saddr)
-				return 1;
-			if (!sk->sk_rcv_saddr || !sk2->sk_rcv_saddr)
-				return match_wildcard;
-		}
-		return 0;
-	}
-
-	if (addr_type == IPV6_ADDR_ANY && addr_type2 == IPV6_ADDR_ANY)
-		return 1;
-
-	if (addr_type2 == IPV6_ADDR_ANY && match_wildcard &&
-	    !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED))
-		return 1;
-
-	if (addr_type == IPV6_ADDR_ANY && match_wildcard &&
-	    !(ipv6_only_sock(sk) && addr_type2 == IPV6_ADDR_MAPPED))
-		return 1;
-
-	if (sk2_rcv_saddr6 &&
-	    ipv6_addr_equal(&sk->sk_v6_rcv_saddr, sk2_rcv_saddr6))
-		return 1;
-
-	return 0;
-}
-
 static u32 udp6_portaddr_hash(const struct net *net,
 			      const struct in6_addr *addr6,
 			      unsigned int port)
diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c
index a2c8747..6b54ff3 100644
--- a/net/l2tp/l2tp_ip6.c
+++ b/net/l2tp/l2tp_ip6.c
@@ -25,6 +25,7 @@
 #include <net/udp.h>
 #include <net/inet_common.h>
 #include <net/inet_hashtables.h>
+#include <net/inet6_hashtables.h>
 #include <net/tcp_states.h>
 #include <net/protocol.h>
 #include <net/xfrm.h>
@@ -718,7 +719,7 @@ static struct proto l2tp_ip6_prot = {
 	.sendmsg	   = l2tp_ip6_sendmsg,
 	.recvmsg	   = l2tp_ip6_recvmsg,
 	.backlog_rcv	   = l2tp_ip6_backlog_recv,
-	.hash		   = inet_hash,
+	.hash		   = inet6_hash,
 	.unhash		   = inet_unhash,
 	.obj_size	   = sizeof(struct l2tp_ip6_sock),
 #ifdef CONFIG_COMPAT
-- 
cgit v0.10.2


From d9b3fca27385eafe61c3ca6feab6cb1e7dc77482 Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:37 -0500
Subject: tcp: __tcp_hdrlen() helper

tcp_hdrlen is wasteful if you already have a pointer to struct tcphdr.
This splits the size calculation into a helper function that can be
used if a struct tcphdr is already available.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index d909fee..bcbf51d 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -29,9 +29,14 @@ static inline struct tcphdr *tcp_hdr(const struct sk_buff *skb)
 	return (struct tcphdr *)skb_transport_header(skb);
 }
 
+static inline unsigned int __tcp_hdrlen(const struct tcphdr *th)
+{
+	return th->doff * 4;
+}
+
 static inline unsigned int tcp_hdrlen(const struct sk_buff *skb)
 {
-	return tcp_hdr(skb)->doff * 4;
+	return __tcp_hdrlen(tcp_hdr(skb));
 }
 
 static inline struct tcphdr *inner_tcp_hdr(const struct sk_buff *skb)
-- 
cgit v0.10.2


From a583636a83ea383fd07517e5a7a2eedbc5d90fb1 Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:38 -0500
Subject: inet: refactor inet[6]_lookup functions to take skb

This is a preliminary step to allow fast socket lookup of SO_REUSEPORT
groups.  Doing so with a BPF filter will require access to the
skb in question.  This change plumbs the skb (and offset to payload
data) through the call stack to the listening socket lookup
implementations where it will be used in a following patch.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 47f52d3..730d856 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -87,6 +87,8 @@ int __ipv6_get_lladdr(struct inet6_dev *idev, struct in6_addr *addr,
 		      u32 banned_flags);
 int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr,
 		    u32 banned_flags);
+int ipv4_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
+			 bool match_wildcard);
 int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2,
 			 bool match_wildcard);
 void addrconf_join_solict(struct net_device *dev, const struct in6_addr *addr);
diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h
index b3c28a9..28332bd 100644
--- a/include/net/inet6_hashtables.h
+++ b/include/net/inet6_hashtables.h
@@ -53,6 +53,7 @@ struct sock *__inet6_lookup_established(struct net *net,
 
 struct sock *inet6_lookup_listener(struct net *net,
 				   struct inet_hashinfo *hashinfo,
+				   struct sk_buff *skb, int doff,
 				   const struct in6_addr *saddr,
 				   const __be16 sport,
 				   const struct in6_addr *daddr,
@@ -60,6 +61,7 @@ struct sock *inet6_lookup_listener(struct net *net,
 
 static inline struct sock *__inet6_lookup(struct net *net,
 					  struct inet_hashinfo *hashinfo,
+					  struct sk_buff *skb, int doff,
 					  const struct in6_addr *saddr,
 					  const __be16 sport,
 					  const struct in6_addr *daddr,
@@ -71,12 +73,12 @@ static inline struct sock *__inet6_lookup(struct net *net,
 	if (sk)
 		return sk;
 
-	return inet6_lookup_listener(net, hashinfo, saddr, sport,
+	return inet6_lookup_listener(net, hashinfo, skb, doff, saddr, sport,
 				     daddr, hnum, dif);
 }
 
 static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo,
-					      struct sk_buff *skb,
+					      struct sk_buff *skb, int doff,
 					      const __be16 sport,
 					      const __be16 dport,
 					      int iif)
@@ -86,13 +88,14 @@ static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo,
 	if (sk)
 		return sk;
 
-	return __inet6_lookup(dev_net(skb_dst(skb)->dev), hashinfo,
-			      &ipv6_hdr(skb)->saddr, sport,
+	return __inet6_lookup(dev_net(skb_dst(skb)->dev), hashinfo, skb,
+			      doff, &ipv6_hdr(skb)->saddr, sport,
 			      &ipv6_hdr(skb)->daddr, ntohs(dport),
 			      iif);
 }
 
 struct sock *inet6_lookup(struct net *net, struct inet_hashinfo *hashinfo,
+			  struct sk_buff *skb, int doff,
 			  const struct in6_addr *saddr, const __be16 sport,
 			  const struct in6_addr *daddr, const __be16 dport,
 			  const int dif);
diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 554440e..8240339 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -213,6 +213,7 @@ void inet_unhash(struct sock *sk);
 
 struct sock *__inet_lookup_listener(struct net *net,
 				    struct inet_hashinfo *hashinfo,
+				    struct sk_buff *skb, int doff,
 				    const __be32 saddr, const __be16 sport,
 				    const __be32 daddr,
 				    const unsigned short hnum,
@@ -220,10 +221,11 @@ struct sock *__inet_lookup_listener(struct net *net,
 
 static inline struct sock *inet_lookup_listener(struct net *net,
 		struct inet_hashinfo *hashinfo,
+		struct sk_buff *skb, int doff,
 		__be32 saddr, __be16 sport,
 		__be32 daddr, __be16 dport, int dif)
 {
-	return __inet_lookup_listener(net, hashinfo, saddr, sport,
+	return __inet_lookup_listener(net, hashinfo, skb, doff, saddr, sport,
 				      daddr, ntohs(dport), dif);
 }
 
@@ -299,6 +301,7 @@ static inline struct sock *
 
 static inline struct sock *__inet_lookup(struct net *net,
 					 struct inet_hashinfo *hashinfo,
+					 struct sk_buff *skb, int doff,
 					 const __be32 saddr, const __be16 sport,
 					 const __be32 daddr, const __be16 dport,
 					 const int dif)
@@ -307,12 +310,13 @@ static inline struct sock *__inet_lookup(struct net *net,
 	struct sock *sk = __inet_lookup_established(net, hashinfo,
 				saddr, sport, daddr, hnum, dif);
 
-	return sk ? : __inet_lookup_listener(net, hashinfo, saddr, sport,
-					     daddr, hnum, dif);
+	return sk ? : __inet_lookup_listener(net, hashinfo, skb, doff, saddr,
+					     sport, daddr, hnum, dif);
 }
 
 static inline struct sock *inet_lookup(struct net *net,
 				       struct inet_hashinfo *hashinfo,
+				       struct sk_buff *skb, int doff,
 				       const __be32 saddr, const __be16 sport,
 				       const __be32 daddr, const __be16 dport,
 				       const int dif)
@@ -320,7 +324,8 @@ static inline struct sock *inet_lookup(struct net *net,
 	struct sock *sk;
 
 	local_bh_disable();
-	sk = __inet_lookup(net, hashinfo, saddr, sport, daddr, dport, dif);
+	sk = __inet_lookup(net, hashinfo, skb, doff, saddr, sport, daddr,
+			   dport, dif);
 	local_bh_enable();
 
 	return sk;
@@ -328,6 +333,7 @@ static inline struct sock *inet_lookup(struct net *net,
 
 static inline struct sock *__inet_lookup_skb(struct inet_hashinfo *hashinfo,
 					     struct sk_buff *skb,
+					     int doff,
 					     const __be16 sport,
 					     const __be16 dport)
 {
@@ -337,8 +343,8 @@ static inline struct sock *__inet_lookup_skb(struct inet_hashinfo *hashinfo,
 	if (sk)
 		return sk;
 	else
-		return __inet_lookup(dev_net(skb_dst(skb)->dev), hashinfo,
-				     iph->saddr, sport,
+		return __inet_lookup(dev_net(skb_dst(skb)->dev), hashinfo, skb,
+				     doff, iph->saddr, sport,
 				     iph->daddr, dport, inet_iif(skb));
 }
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 5684e14..1e0c600 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -802,7 +802,7 @@ static int dccp_v4_rcv(struct sk_buff *skb)
 	}
 
 lookup:
-	sk = __inet_lookup_skb(&dccp_hashinfo, skb,
+	sk = __inet_lookup_skb(&dccp_hashinfo, skb, __dccp_hdr_len(dh),
 			       dh->dccph_sport, dh->dccph_dport);
 	if (!sk) {
 		dccp_pr_debug("failed to look up flow ID in table and "
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 90a8269..45cbe85 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -668,7 +668,7 @@ static int dccp_v6_rcv(struct sk_buff *skb)
 		DCCP_SKB_CB(skb)->dccpd_ack_seq = dccp_hdr_ack_seq(skb);
 
 lookup:
-	sk = __inet6_lookup_skb(&dccp_hashinfo, skb,
+	sk = __inet6_lookup_skb(&dccp_hashinfo, skb, __dccp_hdr_len(dh),
 			        dh->dccph_sport, dh->dccph_dport,
 				inet6_iif(skb));
 	if (!sk) {
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 6029157..50c0d96 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -357,18 +357,18 @@ struct sock *inet_diag_find_one_icsk(struct net *net,
 	struct sock *sk;
 
 	if (req->sdiag_family == AF_INET)
-		sk = inet_lookup(net, hashinfo, req->id.idiag_dst[0],
+		sk = inet_lookup(net, hashinfo, NULL, 0, req->id.idiag_dst[0],
 				 req->id.idiag_dport, req->id.idiag_src[0],
 				 req->id.idiag_sport, req->id.idiag_if);
 #if IS_ENABLED(CONFIG_IPV6)
 	else if (req->sdiag_family == AF_INET6) {
 		if (ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_dst) &&
 		    ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_src))
-			sk = inet_lookup(net, hashinfo, req->id.idiag_dst[3],
+			sk = inet_lookup(net, hashinfo, NULL, 0, req->id.idiag_dst[3],
 					 req->id.idiag_dport, req->id.idiag_src[3],
 					 req->id.idiag_sport, req->id.idiag_if);
 		else
-			sk = inet6_lookup(net, hashinfo,
+			sk = inet6_lookup(net, hashinfo, NULL, 0,
 					  (struct in6_addr *)req->id.idiag_dst,
 					  req->id.idiag_dport,
 					  (struct in6_addr *)req->id.idiag_src,
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index b6023b7..5e4290b 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -205,6 +205,7 @@ static inline int compute_score(struct sock *sk, struct net *net,
 
 struct sock *__inet_lookup_listener(struct net *net,
 				    struct inet_hashinfo *hashinfo,
+				    struct sk_buff *skb, int doff,
 				    const __be32 saddr, __be16 sport,
 				    const __be32 daddr, const unsigned short hnum,
 				    const int dif)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 0d381fa..3f872a6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -637,8 +637,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		 * Incoming packet is checked with md5 hash with finding key,
 		 * no RST generated if md5 hash doesn't match.
 		 */
-		sk1 = __inet_lookup_listener(net,
-					     &tcp_hashinfo, ip_hdr(skb)->saddr,
+		sk1 = __inet_lookup_listener(net, &tcp_hashinfo, NULL, 0,
+					     ip_hdr(skb)->saddr,
 					     th->source, ip_hdr(skb)->daddr,
 					     ntohs(th->source), inet_iif(skb));
 		/* don't send rst if it can't find key */
@@ -1581,7 +1581,8 @@ int tcp_v4_rcv(struct sk_buff *skb)
 	TCP_SKB_CB(skb)->sacked	 = 0;
 
 lookup:
-	sk = __inet_lookup_skb(&tcp_hashinfo, skb, th->source, th->dest);
+	sk = __inet_lookup_skb(&tcp_hashinfo, skb, __tcp_hdrlen(th), th->source,
+			       th->dest);
 	if (!sk)
 		goto no_tcp_socket;
 
@@ -1695,7 +1696,8 @@ do_time_wait:
 	switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) {
 	case TCP_TW_SYN: {
 		struct sock *sk2 = inet_lookup_listener(dev_net(skb->dev),
-							&tcp_hashinfo,
+							&tcp_hashinfo, skb,
+							__tcp_hdrlen(th),
 							iph->saddr, th->source,
 							iph->daddr, th->dest,
 							inet_iif(skb));
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 072653d..004345d 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -121,7 +121,9 @@ static inline int compute_score(struct sock *sk, struct net *net,
 }
 
 struct sock *inet6_lookup_listener(struct net *net,
-		struct inet_hashinfo *hashinfo, const struct in6_addr *saddr,
+		struct inet_hashinfo *hashinfo,
+		struct sk_buff *skb, int doff,
+		const struct in6_addr *saddr,
 		const __be16 sport, const struct in6_addr *daddr,
 		const unsigned short hnum, const int dif)
 {
@@ -177,6 +179,7 @@ begin:
 EXPORT_SYMBOL_GPL(inet6_lookup_listener);
 
 struct sock *inet6_lookup(struct net *net, struct inet_hashinfo *hashinfo,
+			  struct sk_buff *skb, int doff,
 			  const struct in6_addr *saddr, const __be16 sport,
 			  const struct in6_addr *daddr, const __be16 dport,
 			  const int dif)
@@ -184,7 +187,8 @@ struct sock *inet6_lookup(struct net *net, struct inet_hashinfo *hashinfo,
 	struct sock *sk;
 
 	local_bh_disable();
-	sk = __inet6_lookup(net, hashinfo, saddr, sport, daddr, ntohs(dport), dif);
+	sk = __inet6_lookup(net, hashinfo, skb, doff, saddr, sport, daddr,
+			    ntohs(dport), dif);
 	local_bh_enable();
 
 	return sk;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index d72bcfb..9977b6f 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -866,7 +866,8 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb)
 		 * no RST generated if md5 hash doesn't match.
 		 */
 		sk1 = inet6_lookup_listener(dev_net(skb_dst(skb)->dev),
-					   &tcp_hashinfo, &ipv6h->saddr,
+					   &tcp_hashinfo, NULL, 0,
+					   &ipv6h->saddr,
 					   th->source, &ipv6h->daddr,
 					   ntohs(th->source), tcp_v6_iif(skb));
 		if (!sk1)
@@ -1375,8 +1376,8 @@ static int tcp_v6_rcv(struct sk_buff *skb)
 	hdr = ipv6_hdr(skb);
 
 lookup:
-	sk = __inet6_lookup_skb(&tcp_hashinfo, skb, th->source, th->dest,
-				inet6_iif(skb));
+	sk = __inet6_lookup_skb(&tcp_hashinfo, skb, __tcp_hdrlen(th),
+				th->source, th->dest, inet6_iif(skb));
 	if (!sk)
 		goto no_tcp_socket;
 
@@ -1500,6 +1501,7 @@ do_time_wait:
 		struct sock *sk2;
 
 		sk2 = inet6_lookup_listener(dev_net(skb->dev), &tcp_hashinfo,
+					    skb, __tcp_hdrlen(th),
 					    &ipv6_hdr(skb)->saddr, th->source,
 					    &ipv6_hdr(skb)->daddr,
 					    ntohs(th->dest), tcp_v6_iif(skb));
diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c
index 3ab591e..7f4414d 100644
--- a/net/netfilter/xt_TPROXY.c
+++ b/net/netfilter/xt_TPROXY.c
@@ -105,19 +105,24 @@ tproxy_laddr4(struct sk_buff *skb, __be32 user_laddr, __be32 daddr)
  * belonging to established connections going through that one.
  */
 static inline struct sock *
-nf_tproxy_get_sock_v4(struct net *net, const u8 protocol,
+nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, void *hp,
+		      const u8 protocol,
 		      const __be32 saddr, const __be32 daddr,
 		      const __be16 sport, const __be16 dport,
 		      const struct net_device *in,
 		      const enum nf_tproxy_lookup_t lookup_type)
 {
 	struct sock *sk;
+	struct tcphdr *tcph;
 
 	switch (protocol) {
 	case IPPROTO_TCP:
 		switch (lookup_type) {
 		case NFT_LOOKUP_LISTENER:
-			sk = inet_lookup_listener(net, &tcp_hashinfo,
+			tcph = hp;
+			sk = inet_lookup_listener(net, &tcp_hashinfo, skb,
+						    ip_hdrlen(skb) +
+						      __tcp_hdrlen(tcph),
 						    saddr, sport,
 						    daddr, dport,
 						    in->ifindex);
@@ -169,19 +174,23 @@ nf_tproxy_get_sock_v4(struct net *net, const u8 protocol,
 
 #ifdef XT_TPROXY_HAVE_IPV6
 static inline struct sock *
-nf_tproxy_get_sock_v6(struct net *net, const u8 protocol,
+nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff, void *hp,
+		      const u8 protocol,
 		      const struct in6_addr *saddr, const struct in6_addr *daddr,
 		      const __be16 sport, const __be16 dport,
 		      const struct net_device *in,
 		      const enum nf_tproxy_lookup_t lookup_type)
 {
 	struct sock *sk;
+	struct tcphdr *tcph;
 
 	switch (protocol) {
 	case IPPROTO_TCP:
 		switch (lookup_type) {
 		case NFT_LOOKUP_LISTENER:
-			sk = inet6_lookup_listener(net, &tcp_hashinfo,
+			tcph = hp;
+			sk = inet6_lookup_listener(net, &tcp_hashinfo, skb,
+						   thoff + __tcp_hdrlen(tcph),
 						   saddr, sport,
 						   daddr, ntohs(dport),
 						   in->ifindex);
@@ -267,7 +276,7 @@ tproxy_handle_time_wait4(struct net *net, struct sk_buff *skb,
 		 * to a listener socket if there's one */
 		struct sock *sk2;
 
-		sk2 = nf_tproxy_get_sock_v4(net, iph->protocol,
+		sk2 = nf_tproxy_get_sock_v4(net, skb, hp, iph->protocol,
 					    iph->saddr, laddr ? laddr : iph->daddr,
 					    hp->source, lport ? lport : hp->dest,
 					    skb->dev, NFT_LOOKUP_LISTENER);
@@ -305,7 +314,7 @@ tproxy_tg4(struct net *net, struct sk_buff *skb, __be32 laddr, __be16 lport,
 	 * addresses, this happens if the redirect already happened
 	 * and the current packet belongs to an already established
 	 * connection */
-	sk = nf_tproxy_get_sock_v4(net, iph->protocol,
+	sk = nf_tproxy_get_sock_v4(net, skb, hp, iph->protocol,
 				   iph->saddr, iph->daddr,
 				   hp->source, hp->dest,
 				   skb->dev, NFT_LOOKUP_ESTABLISHED);
@@ -321,7 +330,7 @@ tproxy_tg4(struct net *net, struct sk_buff *skb, __be32 laddr, __be16 lport,
 	else if (!sk)
 		/* no, there's no established connection, check if
 		 * there's a listener on the redirected addr/port */
-		sk = nf_tproxy_get_sock_v4(net, iph->protocol,
+		sk = nf_tproxy_get_sock_v4(net, skb, hp, iph->protocol,
 					   iph->saddr, laddr,
 					   hp->source, lport,
 					   skb->dev, NFT_LOOKUP_LISTENER);
@@ -429,7 +438,7 @@ tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff,
 		 * to a listener socket if there's one */
 		struct sock *sk2;
 
-		sk2 = nf_tproxy_get_sock_v6(par->net, tproto,
+		sk2 = nf_tproxy_get_sock_v6(par->net, skb, thoff, hp, tproto,
 					    &iph->saddr,
 					    tproxy_laddr6(skb, &tgi->laddr.in6, &iph->daddr),
 					    hp->source,
@@ -472,7 +481,7 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
 	 * addresses, this happens if the redirect already happened
 	 * and the current packet belongs to an already established
 	 * connection */
-	sk = nf_tproxy_get_sock_v6(par->net, tproto,
+	sk = nf_tproxy_get_sock_v6(par->net, skb, thoff, hp, tproto,
 				   &iph->saddr, &iph->daddr,
 				   hp->source, hp->dest,
 				   par->in, NFT_LOOKUP_ESTABLISHED);
@@ -487,8 +496,8 @@ tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
 	else if (!sk)
 		/* no there's no established connection, check if
 		 * there's a listener on the redirected addr/port */
-		sk = nf_tproxy_get_sock_v6(par->net, tproto,
-					   &iph->saddr, laddr,
+		sk = nf_tproxy_get_sock_v6(par->net, skb, thoff, hp,
+					   tproto, &iph->saddr, laddr,
 					   hp->source, lport,
 					   par->in, NFT_LOOKUP_LISTENER);
 
diff --git a/net/netfilter/xt_socket.c b/net/netfilter/xt_socket.c
index 2ec08f0..49d14ec 100644
--- a/net/netfilter/xt_socket.c
+++ b/net/netfilter/xt_socket.c
@@ -112,14 +112,15 @@ extract_icmp4_fields(const struct sk_buff *skb,
  *     box.
  */
 static struct sock *
-xt_socket_get_sock_v4(struct net *net, const u8 protocol,
+xt_socket_get_sock_v4(struct net *net, struct sk_buff *skb, const int doff,
+		      const u8 protocol,
 		      const __be32 saddr, const __be32 daddr,
 		      const __be16 sport, const __be16 dport,
 		      const struct net_device *in)
 {
 	switch (protocol) {
 	case IPPROTO_TCP:
-		return __inet_lookup(net, &tcp_hashinfo,
+		return __inet_lookup(net, &tcp_hashinfo, skb, doff,
 				     saddr, sport, daddr, dport,
 				     in->ifindex);
 	case IPPROTO_UDP:
@@ -148,6 +149,8 @@ static struct sock *xt_socket_lookup_slow_v4(struct net *net,
 					     const struct net_device *indev)
 {
 	const struct iphdr *iph = ip_hdr(skb);
+	struct sk_buff *data_skb = NULL;
+	int doff = 0;
 	__be32 uninitialized_var(daddr), uninitialized_var(saddr);
 	__be16 uninitialized_var(dport), uninitialized_var(sport);
 	u8 uninitialized_var(protocol);
@@ -169,6 +172,10 @@ static struct sock *xt_socket_lookup_slow_v4(struct net *net,
 		sport = hp->source;
 		daddr = iph->daddr;
 		dport = hp->dest;
+		data_skb = (struct sk_buff *)skb;
+		doff = iph->protocol == IPPROTO_TCP ?
+			ip_hdrlen(skb) + __tcp_hdrlen((struct tcphdr *)hp) :
+			ip_hdrlen(skb) + sizeof(*hp);
 
 	} else if (iph->protocol == IPPROTO_ICMP) {
 		if (extract_icmp4_fields(skb, &protocol, &saddr, &daddr,
@@ -198,8 +205,8 @@ static struct sock *xt_socket_lookup_slow_v4(struct net *net,
 	}
 #endif
 
-	return xt_socket_get_sock_v4(net, protocol, saddr, daddr,
-				     sport, dport, indev);
+	return xt_socket_get_sock_v4(net, data_skb, doff, protocol, saddr,
+				     daddr, sport, dport, indev);
 }
 
 static bool
@@ -318,14 +325,15 @@ extract_icmp6_fields(const struct sk_buff *skb,
 }
 
 static struct sock *
-xt_socket_get_sock_v6(struct net *net, const u8 protocol,
+xt_socket_get_sock_v6(struct net *net, struct sk_buff *skb, int doff,
+		      const u8 protocol,
 		      const struct in6_addr *saddr, const struct in6_addr *daddr,
 		      const __be16 sport, const __be16 dport,
 		      const struct net_device *in)
 {
 	switch (protocol) {
 	case IPPROTO_TCP:
-		return inet6_lookup(net, &tcp_hashinfo,
+		return inet6_lookup(net, &tcp_hashinfo, skb, doff,
 				    saddr, sport, daddr, dport,
 				    in->ifindex);
 	case IPPROTO_UDP:
@@ -343,6 +351,8 @@ static struct sock *xt_socket_lookup_slow_v6(struct net *net,
 	__be16 uninitialized_var(dport), uninitialized_var(sport);
 	const struct in6_addr *daddr = NULL, *saddr = NULL;
 	struct ipv6hdr *iph = ipv6_hdr(skb);
+	struct sk_buff *data_skb = NULL;
+	int doff = 0;
 	int thoff = 0, tproto;
 
 	tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL);
@@ -362,6 +372,10 @@ static struct sock *xt_socket_lookup_slow_v6(struct net *net,
 		sport = hp->source;
 		daddr = &iph->daddr;
 		dport = hp->dest;
+		data_skb = (struct sk_buff *)skb;
+		doff = tproto == IPPROTO_TCP ?
+			thoff + __tcp_hdrlen((struct tcphdr *)hp) :
+			thoff + sizeof(*hp);
 
 	} else if (tproto == IPPROTO_ICMPV6) {
 		struct ipv6hdr ipv6_var;
@@ -373,7 +387,7 @@ static struct sock *xt_socket_lookup_slow_v6(struct net *net,
 		return NULL;
 	}
 
-	return xt_socket_get_sock_v6(net, tproto, saddr, daddr,
+	return xt_socket_get_sock_v6(net, data_skb, doff, tproto, saddr, daddr,
 				     sport, dport, indev);
 }
 
-- 
cgit v0.10.2


From fa463497679352c04d201631534955e6be66eef8 Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:39 -0500
Subject: soreuseport: Prep for fast reuseport TCP socket selection

Both of the lines in this patch probably should have been included
in the initial implementation of this code for generic socket
support, but weren't technically necessary since only UDP sockets
were supported.

First, the sk_reuseport_cb points to a structure which assumes
each socket in the group has this pointer assigned at the same
time it's added to the array in the structure.  The sk_clone_lock
function breaks this assumption.  Since a child socket shouldn't
implicitly be in a reuseport group, the simple fix is to clear
the field in the clone.

Second, the SO_ATTACH_REUSEPORT_xBPF socket options require that
SO_REUSEPORT also be set first.  For UDP sockets, this is easily
enforced at bind-time since that process both puts the socket in
the appropriate receive hlist and updates the reuseport structures.
Since these operations can happen at two different times for TCP
sockets (bind and listen) it must be explicitly checked to enforce
the use of SO_REUSEPORT with SO_ATTACH_REUSEPORT_xBPF in the
setsockopt call.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/filter.c b/net/core/filter.c
index 94d2620..2a6e956 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1181,7 +1181,7 @@ static int __reuseport_attach_prog(struct bpf_prog *prog, struct sock *sk)
 	if (bpf_prog_size(prog->len) > sysctl_optmem_max)
 		return -ENOMEM;
 
-	if (sk_unhashed(sk)) {
+	if (sk_unhashed(sk) && sk->sk_reuseport) {
 		err = reuseport_alloc(sk);
 		if (err)
 			return err;
diff --git a/net/core/sock.c b/net/core/sock.c
index 6c1c8bc9..46dc8ad 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1531,6 +1531,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
 			newsk = NULL;
 			goto out;
 		}
+		RCU_INIT_POINTER(newsk->sk_reuseport_cb, NULL);
 
 		newsk->sk_err	   = 0;
 		newsk->sk_priority = 0;
-- 
cgit v0.10.2


From c125e80b88687b25b321795457309eaaee4bf270 Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:40 -0500
Subject: soreuseport: fast reuseport TCP socket selection

This change extends the fast SO_REUSEPORT socket lookup implemented
for UDP to TCP.  Listener sockets with SO_REUSEPORT and the same
receive address are additionally added to an array for faster
random access.  This means that only a single socket from the group
must be found in the listener list before any socket in the group can
be used to receive a packet.  Previously, every socket in the group
needed to be considered before handing off the incoming packet.

This feature also exposes the ability to use a BPF program when
selecting a socket from a reuseport group.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 8240339..50f635c 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -207,7 +207,10 @@ void inet_hashinfo_init(struct inet_hashinfo *h);
 
 bool inet_ehash_insert(struct sock *sk, struct sock *osk);
 bool inet_ehash_nolisten(struct sock *sk, struct sock *osk);
-void __inet_hash(struct sock *sk, struct sock *osk);
+int __inet_hash(struct sock *sk, struct sock *osk,
+		int (*saddr_same)(const struct sock *sk1,
+				  const struct sock *sk2,
+				  bool match_wildcard));
 int inet_hash(struct sock *sk);
 void inet_unhash(struct sock *sk);
 
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 12c8d38..c16a2e6 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -24,6 +24,7 @@
 #include <net/tcp_states.h>
 #include <net/xfrm.h>
 #include <net/tcp.h>
+#include <net/sock_reuseport.h>
 
 #ifdef INET_CSK_DEBUG
 const char inet_csk_timer_bug_msg[] = "inet_csk BUG: unknown timer value\n";
@@ -67,7 +68,8 @@ int inet_csk_bind_conflict(const struct sock *sk,
 			if ((!reuse || !sk2->sk_reuse ||
 			    sk2->sk_state == TCP_LISTEN) &&
 			    (!reuseport || !sk2->sk_reuseport ||
-			    (sk2->sk_state != TCP_TIME_WAIT &&
+			     rcu_access_pointer(sk->sk_reuseport_cb) ||
+			     (sk2->sk_state != TCP_TIME_WAIT &&
 			     !uid_eq(uid, sock_i_uid(sk2))))) {
 
 				if (!sk2->sk_rcv_saddr || !sk->sk_rcv_saddr ||
@@ -132,6 +134,7 @@ again:
 					      sk->sk_state != TCP_LISTEN) ||
 					     (tb->fastreuseport > 0 &&
 					      sk->sk_reuseport &&
+					      !rcu_access_pointer(sk->sk_reuseport_cb) &&
 					      uid_eq(tb->fastuid, uid))) &&
 					    (tb->num_owners < smallest_size || smallest_size == -1)) {
 						smallest_size = tb->num_owners;
@@ -193,15 +196,18 @@ tb_found:
 		if (((tb->fastreuse > 0 &&
 		      sk->sk_reuse && sk->sk_state != TCP_LISTEN) ||
 		     (tb->fastreuseport > 0 &&
-		      sk->sk_reuseport && uid_eq(tb->fastuid, uid))) &&
-		    smallest_size == -1) {
+		      sk->sk_reuseport &&
+		      !rcu_access_pointer(sk->sk_reuseport_cb) &&
+		      uid_eq(tb->fastuid, uid))) && smallest_size == -1) {
 			goto success;
 		} else {
 			ret = 1;
 			if (inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, true)) {
 				if (((sk->sk_reuse && sk->sk_state != TCP_LISTEN) ||
 				     (tb->fastreuseport > 0 &&
-				      sk->sk_reuseport && uid_eq(tb->fastuid, uid))) &&
+				      sk->sk_reuseport &&
+				      !rcu_access_pointer(sk->sk_reuseport_cb) &&
+				      uid_eq(tb->fastuid, uid))) &&
 				    smallest_size != -1 && --attempts >= 0) {
 					spin_unlock(&head->lock);
 					goto again;
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 5e4290b..c0f9942 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -20,10 +20,12 @@
 #include <linux/wait.h>
 #include <linux/vmalloc.h>
 
+#include <net/addrconf.h>
 #include <net/inet_connection_sock.h>
 #include <net/inet_hashtables.h>
 #include <net/secure_seq.h>
 #include <net/ip.h>
+#include <net/sock_reuseport.h>
 
 static u32 inet_ehashfn(const struct net *net, const __be32 laddr,
 			const __u16 lport, const __be32 faddr,
@@ -215,6 +217,7 @@ struct sock *__inet_lookup_listener(struct net *net,
 	unsigned int hash = inet_lhashfn(net, hnum);
 	struct inet_listen_hashbucket *ilb = &hashinfo->listening_hash[hash];
 	int score, hiscore, matches = 0, reuseport = 0;
+	bool select_ok = true;
 	u32 phash = 0;
 
 	rcu_read_lock();
@@ -230,6 +233,15 @@ begin:
 			if (reuseport) {
 				phash = inet_ehashfn(net, daddr, hnum,
 						     saddr, sport);
+				if (select_ok) {
+					struct sock *sk2;
+					sk2 = reuseport_select_sock(sk, phash,
+								    skb, doff);
+					if (sk2) {
+						result = sk2;
+						goto found;
+					}
+				}
 				matches = 1;
 			}
 		} else if (score == hiscore && reuseport) {
@@ -247,11 +259,13 @@ begin:
 	if (get_nulls_value(node) != hash + LISTENING_NULLS_BASE)
 		goto begin;
 	if (result) {
+found:
 		if (unlikely(!atomic_inc_not_zero(&result->sk_refcnt)))
 			result = NULL;
 		else if (unlikely(compute_score(result, net, hnum, daddr,
 				  dif) < hiscore)) {
 			sock_put(result);
+			select_ok = false;
 			goto begin;
 		}
 	}
@@ -450,34 +464,74 @@ bool inet_ehash_nolisten(struct sock *sk, struct sock *osk)
 }
 EXPORT_SYMBOL_GPL(inet_ehash_nolisten);
 
-void __inet_hash(struct sock *sk, struct sock *osk)
+static int inet_reuseport_add_sock(struct sock *sk,
+				   struct inet_listen_hashbucket *ilb,
+				   int (*saddr_same)(const struct sock *sk1,
+						     const struct sock *sk2,
+						     bool match_wildcard))
+{
+	struct sock *sk2;
+	struct hlist_nulls_node *node;
+	kuid_t uid = sock_i_uid(sk);
+
+	sk_nulls_for_each_rcu(sk2, node, &ilb->head) {
+		if (sk2 != sk &&
+		    sk2->sk_family == sk->sk_family &&
+		    ipv6_only_sock(sk2) == ipv6_only_sock(sk) &&
+		    sk2->sk_bound_dev_if == sk->sk_bound_dev_if &&
+		    sk2->sk_reuseport && uid_eq(uid, sock_i_uid(sk2)) &&
+		    saddr_same(sk, sk2, false))
+			return reuseport_add_sock(sk, sk2);
+	}
+
+	/* Initial allocation may have already happened via setsockopt */
+	if (!rcu_access_pointer(sk->sk_reuseport_cb))
+		return reuseport_alloc(sk);
+	return 0;
+}
+
+int __inet_hash(struct sock *sk, struct sock *osk,
+		 int (*saddr_same)(const struct sock *sk1,
+				   const struct sock *sk2,
+				   bool match_wildcard))
 {
 	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
 	struct inet_listen_hashbucket *ilb;
+	int err = 0;
 
 	if (sk->sk_state != TCP_LISTEN) {
 		inet_ehash_nolisten(sk, osk);
-		return;
+		return 0;
 	}
 	WARN_ON(!sk_unhashed(sk));
 	ilb = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)];
 
 	spin_lock(&ilb->lock);
+	if (sk->sk_reuseport) {
+		err = inet_reuseport_add_sock(sk, ilb, saddr_same);
+		if (err)
+			goto unlock;
+	}
 	__sk_nulls_add_node_rcu(sk, &ilb->head);
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
+unlock:
 	spin_unlock(&ilb->lock);
+
+	return err;
 }
 EXPORT_SYMBOL(__inet_hash);
 
 int inet_hash(struct sock *sk)
 {
+	int err = 0;
+
 	if (sk->sk_state != TCP_CLOSE) {
 		local_bh_disable();
-		__inet_hash(sk, NULL);
+		err = __inet_hash(sk, NULL, ipv4_rcv_saddr_equal);
 		local_bh_enable();
 	}
 
-	return 0;
+	return err;
 }
 EXPORT_SYMBOL_GPL(inet_hash);
 
@@ -496,6 +550,8 @@ void inet_unhash(struct sock *sk)
 		lock = inet_ehash_lockp(hashinfo, sk->sk_hash);
 
 	spin_lock_bh(lock);
+	if (rcu_access_pointer(sk->sk_reuseport_cb))
+		reuseport_detach_sock(sk);
 	done = __sk_nulls_del_node_init_rcu(sk);
 	if (done)
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index be0b218..ac3cedb 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -356,8 +356,8 @@ EXPORT_SYMBOL(udp_lib_get_port);
  * match_wildcard == false: addresses must be exactly the same, i.e.
  *                          0.0.0.0 only equals to 0.0.0.0
  */
-static int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2,
-				bool match_wildcard)
+int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2,
+			 bool match_wildcard)
 {
 	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
 
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 36c3f01..532c3ef 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -26,6 +26,7 @@
 #include <net/ip6_route.h>
 #include <net/sock.h>
 #include <net/inet6_connection_sock.h>
+#include <net/sock_reuseport.h>
 
 int inet6_csk_bind_conflict(const struct sock *sk,
 			    const struct inet_bind_bucket *tb, bool relax)
@@ -48,6 +49,7 @@ int inet6_csk_bind_conflict(const struct sock *sk,
 			if ((!reuse || !sk2->sk_reuse ||
 			     sk2->sk_state == TCP_LISTEN) &&
 			    (!reuseport || !sk2->sk_reuseport ||
+			     rcu_access_pointer(sk->sk_reuseport_cb) ||
 			     (sk2->sk_state != TCP_TIME_WAIT &&
 			      !uid_eq(uid,
 				      sock_i_uid((struct sock *)sk2))))) {
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 004345d..70f2628 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -17,11 +17,13 @@
 #include <linux/module.h>
 #include <linux/random.h>
 
+#include <net/addrconf.h>
 #include <net/inet_connection_sock.h>
 #include <net/inet_hashtables.h>
 #include <net/inet6_hashtables.h>
 #include <net/secure_seq.h>
 #include <net/ip.h>
+#include <net/sock_reuseport.h>
 
 u32 inet6_ehashfn(const struct net *net,
 		  const struct in6_addr *laddr, const u16 lport,
@@ -131,6 +133,7 @@ struct sock *inet6_lookup_listener(struct net *net,
 	const struct hlist_nulls_node *node;
 	struct sock *result;
 	int score, hiscore, matches = 0, reuseport = 0;
+	bool select_ok = true;
 	u32 phash = 0;
 	unsigned int hash = inet_lhashfn(net, hnum);
 	struct inet_listen_hashbucket *ilb = &hashinfo->listening_hash[hash];
@@ -148,6 +151,15 @@ begin:
 			if (reuseport) {
 				phash = inet6_ehashfn(net, daddr, hnum,
 						      saddr, sport);
+				if (select_ok) {
+					struct sock *sk2;
+					sk2 = reuseport_select_sock(sk, phash,
+								    skb, doff);
+					if (sk2) {
+						result = sk2;
+						goto found;
+					}
+				}
 				matches = 1;
 			}
 		} else if (score == hiscore && reuseport) {
@@ -165,11 +177,13 @@ begin:
 	if (get_nulls_value(node) != hash + LISTENING_NULLS_BASE)
 		goto begin;
 	if (result) {
+found:
 		if (unlikely(!atomic_inc_not_zero(&result->sk_refcnt)))
 			result = NULL;
 		else if (unlikely(compute_score(result, net, hnum, daddr,
 				  dif) < hiscore)) {
 			sock_put(result);
+			select_ok = false;
 			goto begin;
 		}
 	}
@@ -283,7 +297,7 @@ int inet6_hash(struct sock *sk)
 {
 	if (sk->sk_state != TCP_CLOSE) {
 		local_bh_disable();
-		__inet_hash(sk, NULL);
+		__inet_hash(sk, NULL, ipv6_rcv_saddr_equal);
 		local_bh_enable();
 	}
 
-- 
cgit v0.10.2


From 4b2a6aed2115cd72faaffc92e03d6516e8113904 Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Wed, 10 Feb 2016 11:50:41 -0500
Subject: soreuseport: BPF selection functional test for TCP

Unfortunately the existing test relied on packet payload in order to
map incoming packets to sockets.  In order to get this to work with TCP,
TCP_FASTOPEN needed to be used.

Since the fast open path is slightly different than the standard TCP path,
I created a second test which sends to reuseport group members based
on receiving cpu core id.  This will probably serve as a better
real-world example use as well.

Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore
index 6fb2336..69bb3fc 100644
--- a/tools/testing/selftests/net/.gitignore
+++ b/tools/testing/selftests/net/.gitignore
@@ -2,3 +2,4 @@ socket
 psock_fanout
 psock_tpacket
 reuseport_bpf
+reuseport_bpf_cpu
diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 41449b5..c658792 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -4,7 +4,7 @@ CFLAGS = -Wall -O2 -g
 
 CFLAGS += -I../../../../usr/include/
 
-NET_PROGS = socket psock_fanout psock_tpacket reuseport_bpf
+NET_PROGS = socket psock_fanout psock_tpacket reuseport_bpf reuseport_bpf_cpu
 
 all: $(NET_PROGS)
 %: %.c
diff --git a/tools/testing/selftests/net/reuseport_bpf.c b/tools/testing/selftests/net/reuseport_bpf.c
index bec1b5d..96ba386 100644
--- a/tools/testing/selftests/net/reuseport_bpf.c
+++ b/tools/testing/selftests/net/reuseport_bpf.c
@@ -9,10 +9,12 @@
 
 #include <errno.h>
 #include <error.h>
+#include <fcntl.h>
 #include <linux/bpf.h>
 #include <linux/filter.h>
 #include <linux/unistd.h>
 #include <netinet/in.h>
+#include <netinet/tcp.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -169,9 +171,15 @@ static void build_recv_group(const struct test_params p, int fd[], uint16_t mod,
 		if (bind(fd[i], addr, sockaddr_size()))
 			error(1, errno, "failed to bind recv socket %d", i);
 
-		if (p.protocol == SOCK_STREAM)
+		if (p.protocol == SOCK_STREAM) {
+			opt = 4;
+			if (setsockopt(fd[i], SOL_TCP, TCP_FASTOPEN, &opt,
+				       sizeof(opt)))
+				error(1, errno,
+				      "failed to set TCP_FASTOPEN on %d", i);
 			if (listen(fd[i], p.recv_socks * 10))
 				error(1, errno, "failed to listen on socket");
+		}
 	}
 	free(addr);
 }
@@ -189,10 +197,8 @@ static void send_from(struct test_params p, uint16_t sport, char *buf,
 
 	if (bind(fd, saddr, sockaddr_size()))
 		error(1, errno, "failed to bind send socket");
-	if (connect(fd, daddr, sockaddr_size()))
-		error(1, errno, "failed to connect");
 
-	if (send(fd, buf, len, 0) < 0)
+	if (sendto(fd, buf, len, MSG_FASTOPEN, daddr, sockaddr_size()) < 0)
 		error(1, errno, "failed to send message");
 
 	close(fd);
@@ -260,7 +266,7 @@ static void test_recv_order(const struct test_params p, int fd[], int mod)
 	}
 }
 
-static void test_reuseport_ebpf(const struct test_params p)
+static void test_reuseport_ebpf(struct test_params p)
 {
 	int i, fd[p.recv_socks];
 
@@ -268,6 +274,7 @@ static void test_reuseport_ebpf(const struct test_params p)
 	build_recv_group(p, fd, p.recv_socks, attach_ebpf);
 	test_recv_order(p, fd, p.recv_socks);
 
+	p.send_port_min += p.recv_socks * 2;
 	fprintf(stderr, "Reprograming, testing mod %zd...\n", p.recv_socks / 2);
 	attach_ebpf(fd[0], p.recv_socks / 2);
 	test_recv_order(p, fd, p.recv_socks / 2);
@@ -276,7 +283,7 @@ static void test_reuseport_ebpf(const struct test_params p)
 		close(fd[i]);
 }
 
-static void test_reuseport_cbpf(const struct test_params p)
+static void test_reuseport_cbpf(struct test_params p)
 {
 	int i, fd[p.recv_socks];
 
@@ -284,6 +291,7 @@ static void test_reuseport_cbpf(const struct test_params p)
 	build_recv_group(p, fd, p.recv_socks, attach_cbpf);
 	test_recv_order(p, fd, p.recv_socks);
 
+	p.send_port_min += p.recv_socks * 2;
 	fprintf(stderr, "Reprograming, testing mod %zd...\n", p.recv_socks / 2);
 	attach_cbpf(fd[0], p.recv_socks / 2);
 	test_recv_order(p, fd, p.recv_socks / 2);
@@ -377,7 +385,7 @@ static void test_filter_no_reuseport(const struct test_params p)
 
 static void test_filter_without_bind(void)
 {
-	int fd1, fd2;
+	int fd1, fd2, opt = 1;
 
 	fprintf(stderr, "Testing filter add without bind...\n");
 	fd1 = socket(AF_INET, SOCK_DGRAM, 0);
@@ -386,6 +394,10 @@ static void test_filter_without_bind(void)
 	fd2 = socket(AF_INET, SOCK_DGRAM, 0);
 	if (fd2 < 0)
 		error(1, errno, "failed to create socket 2");
+	if (setsockopt(fd1, SOL_SOCKET, SO_REUSEPORT, &opt, sizeof(opt)))
+		error(1, errno, "failed to set SO_REUSEPORT on socket 1");
+	if (setsockopt(fd2, SOL_SOCKET, SO_REUSEPORT, &opt, sizeof(opt)))
+		error(1, errno, "failed to set SO_REUSEPORT on socket 2");
 
 	attach_ebpf(fd1, 10);
 	attach_cbpf(fd2, 10);
@@ -394,6 +406,32 @@ static void test_filter_without_bind(void)
 	close(fd2);
 }
 
+void enable_fastopen(void)
+{
+	int fd = open("/proc/sys/net/ipv4/tcp_fastopen", 0);
+	int rw_mask = 3;  /* bit 1: client side; bit-2 server side */
+	int val, size;
+	char buf[16];
+
+	if (fd < 0)
+		error(1, errno, "Unable to open tcp_fastopen sysctl");
+	if (read(fd, buf, sizeof(buf)) <= 0)
+		error(1, errno, "Unable to read tcp_fastopen sysctl");
+	val = atoi(buf);
+	close(fd);
+
+	if ((val & rw_mask) != rw_mask) {
+		fd = open("/proc/sys/net/ipv4/tcp_fastopen", O_RDWR);
+		if (fd < 0)
+			error(1, errno,
+			      "Unable to open tcp_fastopen sysctl for writing");
+		val |= rw_mask;
+		size = snprintf(buf, 16, "%d", val);
+		if (write(fd, buf, size) <= 0)
+			error(1, errno, "Unable to write tcp_fastopen sysctl");
+		close(fd);
+	}
+}
 
 int main(void)
 {
@@ -506,6 +544,71 @@ int main(void)
 		.recv_port = 8007,
 		.send_port_min = 9100});
 
+	/* TCP fastopen is required for the TCP tests */
+	enable_fastopen();
+	fprintf(stderr, "---- IPv4 TCP ----\n");
+	test_reuseport_ebpf((struct test_params) {
+		.recv_family = AF_INET,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8008,
+		.send_port_min = 9120});
+	test_reuseport_cbpf((struct test_params) {
+		.recv_family = AF_INET,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8009,
+		.send_port_min = 9160});
+	test_extra_filter((struct test_params) {
+		.recv_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8010});
+	test_filter_no_reuseport((struct test_params) {
+		.recv_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8011});
+
+	fprintf(stderr, "---- IPv6 TCP ----\n");
+	test_reuseport_ebpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8012,
+		.send_port_min = 9200});
+	test_reuseport_cbpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8013,
+		.send_port_min = 9240});
+	test_extra_filter((struct test_params) {
+		.recv_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8014});
+	test_filter_no_reuseport((struct test_params) {
+		.recv_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8015});
+
+	fprintf(stderr, "---- IPv6 TCP w/ mapped IPv4 ----\n");
+	test_reuseport_ebpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8016,
+		.send_port_min = 9320});
+	test_reuseport_cbpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8017,
+		.send_port_min = 9360});
 
 	test_filter_without_bind();
 
diff --git a/tools/testing/selftests/net/reuseport_bpf_cpu.c b/tools/testing/selftests/net/reuseport_bpf_cpu.c
new file mode 100644
index 0000000..b23d6f5
--- /dev/null
+++ b/tools/testing/selftests/net/reuseport_bpf_cpu.c
@@ -0,0 +1,258 @@
+/*
+ * Test functionality of BPF filters with SO_REUSEPORT.  This program creates
+ * an SO_REUSEPORT receiver group containing one socket per CPU core. It then
+ * creates a BPF program that will select a socket from this group based
+ * on the core id that receives the packet.  The sending code artificially
+ * moves itself to run on different core ids and sends one message from
+ * each core.  Since these packets are delivered over loopback, they should
+ * arrive on the same core that sent them.  The receiving code then ensures
+ * that the packet was received on the socket for the corresponding core id.
+ * This entire process is done for several different core id permutations
+ * and for each IPv4/IPv6 and TCP/UDP combination.
+ */
+
+#define _GNU_SOURCE
+
+#include <arpa/inet.h>
+#include <errno.h>
+#include <error.h>
+#include <linux/filter.h>
+#include <linux/in.h>
+#include <linux/unistd.h>
+#include <sched.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/epoll.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <unistd.h>
+
+static const int PORT = 8888;
+
+static void build_rcv_group(int *rcv_fd, size_t len, int family, int proto)
+{
+	struct sockaddr_storage addr;
+	struct sockaddr_in  *addr4;
+	struct sockaddr_in6 *addr6;
+	size_t i;
+	int opt;
+
+	switch (family) {
+	case AF_INET:
+		addr4 = (struct sockaddr_in *)&addr;
+		addr4->sin_family = AF_INET;
+		addr4->sin_addr.s_addr = htonl(INADDR_ANY);
+		addr4->sin_port = htons(PORT);
+		break;
+	case AF_INET6:
+		addr6 = (struct sockaddr_in6 *)&addr;
+		addr6->sin6_family = AF_INET6;
+		addr6->sin6_addr = in6addr_any;
+		addr6->sin6_port = htons(PORT);
+		break;
+	default:
+		error(1, 0, "Unsupported family %d", family);
+	}
+
+	for (i = 0; i < len; ++i) {
+		rcv_fd[i] = socket(family, proto, 0);
+		if (rcv_fd[i] < 0)
+			error(1, errno, "failed to create receive socket");
+
+		opt = 1;
+		if (setsockopt(rcv_fd[i], SOL_SOCKET, SO_REUSEPORT, &opt,
+			       sizeof(opt)))
+			error(1, errno, "failed to set SO_REUSEPORT");
+
+		if (bind(rcv_fd[i], (struct sockaddr *)&addr, sizeof(addr)))
+			error(1, errno, "failed to bind receive socket");
+
+		if (proto == SOCK_STREAM && listen(rcv_fd[i], len * 10))
+			error(1, errno, "failed to listen on receive port");
+	}
+}
+
+static void attach_bpf(int fd)
+{
+	struct sock_filter code[] = {
+		/* A = raw_smp_processor_id() */
+		{ BPF_LD  | BPF_W | BPF_ABS, 0, 0, SKF_AD_OFF + SKF_AD_CPU },
+		/* return A */
+		{ BPF_RET | BPF_A, 0, 0, 0 },
+	};
+	struct sock_fprog p = {
+		.len = 2,
+		.filter = code,
+	};
+
+	if (setsockopt(fd, SOL_SOCKET, SO_ATTACH_REUSEPORT_CBPF, &p, sizeof(p)))
+		error(1, errno, "failed to set SO_ATTACH_REUSEPORT_CBPF");
+}
+
+static void send_from_cpu(int cpu_id, int family, int proto)
+{
+	struct sockaddr_storage saddr, daddr;
+	struct sockaddr_in  *saddr4, *daddr4;
+	struct sockaddr_in6 *saddr6, *daddr6;
+	cpu_set_t cpu_set;
+	int fd;
+
+	switch (family) {
+	case AF_INET:
+		saddr4 = (struct sockaddr_in *)&saddr;
+		saddr4->sin_family = AF_INET;
+		saddr4->sin_addr.s_addr = htonl(INADDR_ANY);
+		saddr4->sin_port = 0;
+
+		daddr4 = (struct sockaddr_in *)&daddr;
+		daddr4->sin_family = AF_INET;
+		daddr4->sin_addr.s_addr = htonl(INADDR_LOOPBACK);
+		daddr4->sin_port = htons(PORT);
+		break;
+	case AF_INET6:
+		saddr6 = (struct sockaddr_in6 *)&saddr;
+		saddr6->sin6_family = AF_INET6;
+		saddr6->sin6_addr = in6addr_any;
+		saddr6->sin6_port = 0;
+
+		daddr6 = (struct sockaddr_in6 *)&daddr;
+		daddr6->sin6_family = AF_INET6;
+		daddr6->sin6_addr = in6addr_loopback;
+		daddr6->sin6_port = htons(PORT);
+		break;
+	default:
+		error(1, 0, "Unsupported family %d", family);
+	}
+
+	memset(&cpu_set, 0, sizeof(cpu_set));
+	CPU_SET(cpu_id, &cpu_set);
+	if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0)
+		error(1, errno, "failed to pin to cpu");
+
+	fd = socket(family, proto, 0);
+	if (fd < 0)
+		error(1, errno, "failed to create send socket");
+
+	if (bind(fd, (struct sockaddr *)&saddr, sizeof(saddr)))
+		error(1, errno, "failed to bind send socket");
+
+	if (connect(fd, (struct sockaddr *)&daddr, sizeof(daddr)))
+		error(1, errno, "failed to connect send socket");
+
+	if (send(fd, "a", 1, 0) < 0)
+		error(1, errno, "failed to send message");
+
+	close(fd);
+}
+
+static
+void receive_on_cpu(int *rcv_fd, int len, int epfd, int cpu_id, int proto)
+{
+	struct epoll_event ev;
+	int i, fd;
+	char buf[8];
+
+	i = epoll_wait(epfd, &ev, 1, -1);
+	if (i < 0)
+		error(1, errno, "epoll_wait failed");
+
+	if (proto == SOCK_STREAM) {
+		fd = accept(ev.data.fd, NULL, NULL);
+		if (fd < 0)
+			error(1, errno, "failed to accept");
+		i = recv(fd, buf, sizeof(buf), 0);
+		close(fd);
+	} else {
+		i = recv(ev.data.fd, buf, sizeof(buf), 0);
+	}
+
+	if (i < 0)
+		error(1, errno, "failed to recv");
+
+	for (i = 0; i < len; ++i)
+		if (ev.data.fd == rcv_fd[i])
+			break;
+	if (i == len)
+		error(1, 0, "failed to find socket");
+	fprintf(stderr, "send cpu %d, receive socket %d\n", cpu_id, i);
+	if (cpu_id != i)
+		error(1, 0, "cpu id/receive socket mismatch");
+}
+
+static void test(int *rcv_fd, int len, int family, int proto)
+{
+	struct epoll_event ev;
+	int epfd, cpu;
+
+	build_rcv_group(rcv_fd, len, family, proto);
+	attach_bpf(rcv_fd[0]);
+
+	epfd = epoll_create(1);
+	if (epfd < 0)
+		error(1, errno, "failed to create epoll");
+	for (cpu = 0; cpu < len; ++cpu) {
+		ev.events = EPOLLIN;
+		ev.data.fd = rcv_fd[cpu];
+		if (epoll_ctl(epfd, EPOLL_CTL_ADD, rcv_fd[cpu], &ev))
+			error(1, errno, "failed to register sock epoll");
+	}
+
+	/* Forward iterate */
+	for (cpu = 0; cpu < len; ++cpu) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	/* Reverse iterate */
+	for (cpu = len - 1; cpu >= 0; --cpu) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	/* Even cores */
+	for (cpu = 0; cpu < len; cpu += 2) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	/* Odd cores */
+	for (cpu = 1; cpu < len; cpu += 2) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	close(epfd);
+	for (cpu = 0; cpu < len; ++cpu)
+		close(rcv_fd[cpu]);
+}
+
+int main(void)
+{
+	int *rcv_fd, cpus;
+
+	cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	if (cpus <= 0)
+		error(1, errno, "failed counting cpus");
+
+	rcv_fd = calloc(cpus, sizeof(int));
+	if (!rcv_fd)
+		error(1, 0, "failed to allocate array");
+
+	fprintf(stderr, "---- IPv4 UDP ----\n");
+	test(rcv_fd, cpus, AF_INET, SOCK_DGRAM);
+
+	fprintf(stderr, "---- IPv6 UDP ----\n");
+	test(rcv_fd, cpus, AF_INET6, SOCK_DGRAM);
+
+	fprintf(stderr, "---- IPv4 TCP ----\n");
+	test(rcv_fd, cpus, AF_INET, SOCK_STREAM);
+
+	fprintf(stderr, "---- IPv6 TCP ----\n");
+	test(rcv_fd, cpus, AF_INET6, SOCK_STREAM);
+
+	free(rcv_fd);
+
+	fprintf(stderr, "SUCCESS\n");
+	return 0;
+}
-- 
cgit v0.10.2


From dc599f76c22b0de55a2a9141dfe52e70b32a1194 Mon Sep 17 00:00:00 2001
From: David Ahern <dsa@cumulusnetworks.com>
Date: Tue, 2 Feb 2016 08:17:07 -0800
Subject: net: Add support for filtering link dump by master device and kind

Add support for filtering link dumps by master device and kind, similar
to the filtering implemented for neighbor dumps.

Each net_device that exists adds between 1196 bytes (eth) and 1556 bytes
(bridge) to the link dump. As the number of interfaces increases so does
the amount of data pushed to user space for a link list. If the user
only wants to see a list of specific devices (e.g., interfaces enslaved
to a specific bridge or a list of VRFs) most of that data is thrown away.
Passing the filters to the kernel to have only relevant data returned
makes the dump more efficient.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 20d7135..62737f4 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1414,6 +1414,58 @@ static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = {
 	[IFLA_PORT_RESPONSE]	= { .type = NLA_U16, },
 };
 
+static const struct rtnl_link_ops *linkinfo_to_kind_ops(const struct nlattr *nla)
+{
+	const struct rtnl_link_ops *ops = NULL;
+	struct nlattr *linfo[IFLA_INFO_MAX + 1];
+
+	if (nla_parse_nested(linfo, IFLA_INFO_MAX, nla, ifla_info_policy) < 0)
+		return NULL;
+
+	if (linfo[IFLA_INFO_KIND]) {
+		char kind[MODULE_NAME_LEN];
+
+		nla_strlcpy(kind, linfo[IFLA_INFO_KIND], sizeof(kind));
+		ops = rtnl_link_ops_get(kind);
+	}
+
+	return ops;
+}
+
+static bool link_master_filtered(struct net_device *dev, int master_idx)
+{
+	struct net_device *master;
+
+	if (!master_idx)
+		return false;
+
+	master = netdev_master_upper_dev_get(dev);
+	if (!master || master->ifindex != master_idx)
+		return true;
+
+	return false;
+}
+
+static bool link_kind_filtered(const struct net_device *dev,
+			       const struct rtnl_link_ops *kind_ops)
+{
+	if (kind_ops && dev->rtnl_link_ops != kind_ops)
+		return true;
+
+	return false;
+}
+
+static bool link_dump_filtered(struct net_device *dev,
+			       int master_idx,
+			       const struct rtnl_link_ops *kind_ops)
+{
+	if (link_master_filtered(dev, master_idx) ||
+	    link_kind_filtered(dev, kind_ops))
+		return true;
+
+	return false;
+}
+
 static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
@@ -1423,6 +1475,9 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 	struct hlist_head *head;
 	struct nlattr *tb[IFLA_MAX+1];
 	u32 ext_filter_mask = 0;
+	const struct rtnl_link_ops *kind_ops = NULL;
+	unsigned int flags = NLM_F_MULTI;
+	int master_idx = 0;
 	int err;
 	int hdrlen;
 
@@ -1445,18 +1500,29 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
 
 		if (tb[IFLA_EXT_MASK])
 			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
+
+		if (tb[IFLA_MASTER])
+			master_idx = nla_get_u32(tb[IFLA_MASTER]);
+
+		if (tb[IFLA_LINKINFO])
+			kind_ops = linkinfo_to_kind_ops(tb[IFLA_LINKINFO]);
+
+		if (master_idx || kind_ops)
+			flags |= NLM_F_DUMP_FILTERED;
 	}
 
 	for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
 		idx = 0;
 		head = &net->dev_index_head[h];
 		hlist_for_each_entry(dev, head, index_hlist) {
+			if (link_dump_filtered(dev, master_idx, kind_ops))
+				continue;
 			if (idx < s_idx)
 				goto cont;
 			err = rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
 					       NETLINK_CB(cb->skb).portid,
 					       cb->nlh->nlmsg_seq, 0,
-					       NLM_F_MULTI,
+					       flags,
 					       ext_filter_mask);
 			/* If we ran out of room on the first message,
 			 * we're in trouble
-- 
cgit v0.10.2


From 1490d2bd1bbcf9ad127f7764b7f28b26a55e011b Mon Sep 17 00:00:00 2001
From: Wei Tang <tangwei@cmss.chinamobile.com>
Date: Thu, 4 Feb 2016 17:36:23 +0800
Subject: bpf_dbg: do not initialise statics to 0

This patch fixes the checkpatch.pl error to bpf_dbg.c:

ERROR: do not initialise statics to 0

Signed-off-by: Wei Tang <tangwei@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/tools/net/bpf_dbg.c b/tools/net/bpf_dbg.c
index 9a287be..4f254bc 100644
--- a/tools/net/bpf_dbg.c
+++ b/tools/net/bpf_dbg.c
@@ -129,16 +129,16 @@ struct bpf_regs {
 };
 
 static struct sock_filter bpf_image[BPF_MAXINSNS + 1];
-static unsigned int bpf_prog_len = 0;
+static unsigned int bpf_prog_len;
 
 static int bpf_breakpoints[64];
 static struct bpf_regs bpf_regs[BPF_MAXINSNS + 1];
 static struct bpf_regs bpf_curr;
-static unsigned int bpf_regs_len = 0;
+static unsigned int bpf_regs_len;
 
 static int pcap_fd = -1;
-static unsigned int pcap_packet = 0;
-static size_t pcap_map_size = 0;
+static unsigned int pcap_packet;
+static size_t pcap_map_size;
 static char *pcap_ptr_va_start, *pcap_ptr_va_curr;
 
 static const char * const op_table[] = {
@@ -1172,7 +1172,7 @@ static int cmd_breakpoint(char *subcmd)
 
 static int cmd_run(char *num)
 {
-	static uint32_t pass = 0, fail = 0;
+	static uint32_t pass, fail;
 	bool has_limit = true;
 	int pkts = 0, i = 0;
 
-- 
cgit v0.10.2


From ccad0993567812e88617f105c01b04f1528538b6 Mon Sep 17 00:00:00 2001
From: Siva Reddy Kallam <siva.kallam@broadcom.com>
Date: Thu, 4 Feb 2016 15:20:47 +0530
Subject: MAINTAINERS: Update tg3 maintainer

Signed-off-by: Siva Reddy Kallam <siva.kallam@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Acked-by: Prashant Sreedharan <prashant@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/MAINTAINERS b/MAINTAINERS
index 1e216aa..355e1c8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2424,6 +2424,7 @@ F:	include/linux/bcm963xx_nvram.h
 F:	include/linux/bcm963xx_tag.h
 
 BROADCOM TG3 GIGABIT ETHERNET DRIVER
+M:	Siva Reddy Kallam <siva.kallam@broadcom.com>
 M:	Prashant Sreedharan <prashant@broadcom.com>
 M:	Michael Chan <mchan@broadcom.com>
 L:	netdev@vger.kernel.org
-- 
cgit v0.10.2


From 12b74dfadb5a7a23baf4db941dc9fd9d371f249a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 4 Feb 2016 13:31:17 +0100
Subject: ipv4: add option to drop unicast encapsulated in L2 multicast

In order to solve a problem with 802.11, the so-called hole-196 attack,
add an option (sysctl) called "drop_unicast_in_l2_multicast" which, if
enabled, causes the stack to drop IPv4 unicast packets encapsulated in
link-layer multi- or broadcast frames. Such frames can (as an attack)
be created by any member of the same wireless network and transmitted
as valid encrypted frames since the symmetric key for broadcast frames
is shared between all stations.

Additionally, enabling this option provides compliance with a SHOULD
clause of RFC 1122.

Reviewed-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 73b36d7..d5910d6 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1216,6 +1216,13 @@ promote_secondaries - BOOLEAN
 	promote a corresponding secondary IP address instead of
 	removing all the corresponding secondary IP addresses.
 
+drop_unicast_in_l2_multicast - BOOLEAN
+	Drop any unicast IP packets that are received in link-layer
+	multicast (or broadcast) frames.
+	This behavior (for multicast) is actually a SHOULD in RFC
+	1122, but is disabled by default for compatibility reasons.
+	Default: off (0)
+
 
 tag - INTEGER
 	Allows you to write a number, which can be used as required.
diff --git a/include/uapi/linux/ip.h b/include/uapi/linux/ip.h
index 08f894d..584834f 100644
--- a/include/uapi/linux/ip.h
+++ b/include/uapi/linux/ip.h
@@ -165,6 +165,7 @@ enum
 	IPV4_DEVCONF_IGMPV2_UNSOLICITED_REPORT_INTERVAL,
 	IPV4_DEVCONF_IGMPV3_UNSOLICITED_REPORT_INTERVAL,
 	IPV4_DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN,
+	IPV4_DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
 	__IPV4_DEVCONF_MAX
 };
 
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index cebd9d3..dbbab28 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -2192,6 +2192,8 @@ static struct devinet_sysctl_table {
 					      "promote_secondaries"),
 		DEVINET_SYSCTL_FLUSHING_ENTRY(ROUTE_LOCALNET,
 					      "route_localnet"),
+		DEVINET_SYSCTL_FLUSHING_ENTRY(DROP_UNICAST_IN_L2_MULTICAST,
+					      "drop_unicast_in_l2_multicast"),
 	},
 };
 
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index d77eb0c..852002f 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -362,8 +362,31 @@ static int ip_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 	rt = skb_rtable(skb);
 	if (rt->rt_type == RTN_MULTICAST) {
 		IP_UPD_PO_STATS_BH(net, IPSTATS_MIB_INMCAST, skb->len);
-	} else if (rt->rt_type == RTN_BROADCAST)
+	} else if (rt->rt_type == RTN_BROADCAST) {
 		IP_UPD_PO_STATS_BH(net, IPSTATS_MIB_INBCAST, skb->len);
+	} else if (skb->pkt_type == PACKET_BROADCAST ||
+		   skb->pkt_type == PACKET_MULTICAST) {
+		struct in_device *in_dev = __in_dev_get_rcu(skb->dev);
+
+		/* RFC 1122 3.3.6:
+		 *
+		 *   When a host sends a datagram to a link-layer broadcast
+		 *   address, the IP destination address MUST be a legal IP
+		 *   broadcast or IP multicast address.
+		 *
+		 *   A host SHOULD silently discard a datagram that is received
+		 *   via a link-layer broadcast (see Section 2.4) but does not
+		 *   specify an IP multicast or broadcast destination address.
+		 *
+		 * This doesn't explicitly say L2 *broadcast*, but broadcast is
+		 * in a way a form of multicast and the most common use case for
+		 * this is 802.11 protecting against cross-station spoofing (the
+		 * so-called "hole-196" attack) so do it for both.
+		 */
+		if (in_dev &&
+		    IN_DEV_ORCONF(in_dev, DROP_UNICAST_IN_L2_MULTICAST))
+			goto drop;
+	}
 
 	return dst_input(skb);
 
-- 
cgit v0.10.2


From 97daf331455077645ae1f13438bebd3d1a2e94ee Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 4 Feb 2016 13:31:18 +0100
Subject: ipv4: add option to drop gratuitous ARP packets

In certain 802.11 wireless deployments, there will be ARP proxies
that use knowledge of the network to correctly answer requests.
To prevent gratuitous ARP frames on the shared medium from being
a problem, on such deployments wireless needs to drop them.

Enable this by providing an option called "drop_gratuitous_arp".

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index d5910d6..a53bbfa 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1223,6 +1223,12 @@ drop_unicast_in_l2_multicast - BOOLEAN
 	1122, but is disabled by default for compatibility reasons.
 	Default: off (0)
 
+drop_gratuitous_arp - BOOLEAN
+	Drop all gratuitous ARP frames, for example if there's a known
+	good ARP proxy on the network and such frames need not be used
+	(or in the case of 802.11, must not be used to prevent attacks.)
+	Default: off (0)
+
 
 tag - INTEGER
 	Allows you to write a number, which can be used as required.
diff --git a/include/uapi/linux/ip.h b/include/uapi/linux/ip.h
index 584834f..f291569 100644
--- a/include/uapi/linux/ip.h
+++ b/include/uapi/linux/ip.h
@@ -166,6 +166,7 @@ enum
 	IPV4_DEVCONF_IGMPV3_UNSOLICITED_REPORT_INTERVAL,
 	IPV4_DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN,
 	IPV4_DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
+	IPV4_DEVCONF_DROP_GRATUITOUS_ARP,
 	__IPV4_DEVCONF_MAX
 };
 
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 59b3e0e..c102eb5 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -735,6 +735,14 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 	    (!IN_DEV_ROUTE_LOCALNET(in_dev) && ipv4_is_loopback(tip)))
 		goto out;
 
+ /*
+  *	For some 802.11 wireless deployments (and possibly other networks),
+  *	there will be an ARP proxy and gratuitous ARP frames are attacks
+  *	and thus should not be accepted.
+  */
+	if (sip == tip && IN_DEV_ORCONF(in_dev, DROP_GRATUITOUS_ARP))
+		goto out;
+
 /*
  *     Special case: We must set Frame Relay source Q.922 address
  */
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index dbbab28..3d83531 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -2185,6 +2185,8 @@ static struct devinet_sysctl_table {
 					"igmpv3_unsolicited_report_interval"),
 		DEVINET_SYSCTL_RW_ENTRY(IGNORE_ROUTES_WITH_LINKDOWN,
 					"ignore_routes_with_linkdown"),
+		DEVINET_SYSCTL_RW_ENTRY(DROP_GRATUITOUS_ARP,
+					"drop_gratuitous_arp"),
 
 		DEVINET_SYSCTL_FLUSHING_ENTRY(NOXFRM, "disable_xfrm"),
 		DEVINET_SYSCTL_FLUSHING_ENTRY(NOPOLICY, "disable_policy"),
-- 
cgit v0.10.2


From abbc30436d39dfed8ebfca338d253f211ac7b094 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 4 Feb 2016 13:31:19 +0100
Subject: ipv6: add option to drop unicast encapsulated in L2 multicast

In order to solve a problem with 802.11, the so-called hole-196 attack,
add an option (sysctl) called "drop_unicast_in_l2_multicast" which, if
enabled, causes the stack to drop IPv6 unicast packets encapsulated in
link-layer multi- or broadcast frames. Such frames can (as an attack)
be created by any member of the same wireless network and transmitted
as valid encrypted frames since the symmetric key for broadcast frames
is shared between all stations.

Reviewed-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index a53bbfa..e0e7350 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1674,6 +1674,12 @@ stable_secret - IPv6 address
 
 	By default the stable secret is unset.
 
+drop_unicast_in_l2_multicast - BOOLEAN
+	Drop any unicast IPv6 packets that are received in link-layer
+	multicast (or broadcast) frames.
+
+	By default this is turned off.
+
 icmp/*:
 ratelimit - INTEGER
 	Limit the maximal rates for sending ICMPv6 packets.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 402753b..4a4c1ae 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -50,6 +50,7 @@ struct ipv6_devconf {
 	__s32		mc_forwarding;
 #endif
 	__s32		disable_ipv6;
+	__s32		drop_unicast_in_l2_multicast;
 	__s32		accept_dad;
 	__s32		force_tllao;
 	__s32           ndisc_notify;
diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h
index 38b4fef..4c41357 100644
--- a/include/uapi/linux/ipv6.h
+++ b/include/uapi/linux/ipv6.h
@@ -174,6 +174,7 @@ enum {
 	DEVCONF_USE_OIF_ADDRS_ONLY,
 	DEVCONF_ACCEPT_RA_MIN_HOP_LIMIT,
 	DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN,
+	DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 38eedde..23e325f 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -4711,6 +4711,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN] = cnf->ignore_routes_with_linkdown;
 	/* we omit DEVCONF_STABLE_SECRET for now */
 	array[DEVCONF_USE_OIF_ADDRS_ONLY] = cnf->use_oif_addrs_only;
+	array[DEVCONF_DROP_UNICAST_IN_L2_MULTICAST] = cnf->drop_unicast_in_l2_multicast;
 }
 
 static inline size_t inet6_ifla6_size(void)
@@ -5785,6 +5786,13 @@ static struct addrconf_sysctl_table
 			.proc_handler	= addrconf_sysctl_ignore_routes_with_linkdown,
 		},
 		{
+			.procname	= "drop_unicast_in_l2_multicast",
+			.data		= &ipv6_devconf.drop_unicast_in_l2_multicast,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= proc_dointvec,
+		},
+		{
 			/* sentinel */
 		}
 	},
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 9075acf..31ac3c5 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -134,6 +134,16 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 	    IPV6_ADDR_MC_SCOPE(&hdr->daddr) == 1)
 		goto err;
 
+	/* If enabled, drop unicast packets that were encapsulated in link-layer
+	 * multicast or broadcast to protected against the so-called "hole-196"
+	 * attack in 802.11 wireless.
+	 */
+	if (!ipv6_addr_is_multicast(&hdr->daddr) &&
+	    (skb->pkt_type == PACKET_BROADCAST ||
+	     skb->pkt_type == PACKET_MULTICAST) &&
+	    idev->cnf.drop_unicast_in_l2_multicast)
+		goto err;
+
 	/* RFC4291 2.7
 	 * Nodes must not originate a packet to a multicast address whose scope
 	 * field contains the reserved value 0; if such a packet is received, it
-- 
cgit v0.10.2


From 7a02bf892d8f1e5298af1676f001bee410509d80 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 4 Feb 2016 13:31:20 +0100
Subject: ipv6: add option to drop unsolicited neighbor advertisements

In certain 802.11 wireless deployments, there will be NA proxies
that use knowledge of the network to correctly answer requests.
To prevent unsolicitd advertisements on the shared medium from
being a problem, on such deployments wireless needs to drop them.

Enable this by providing an option called "drop_unsolicited_na".

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index e0e7350..24ce97f 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1680,6 +1680,13 @@ drop_unicast_in_l2_multicast - BOOLEAN
 
 	By default this is turned off.
 
+drop_unsolicited_na - BOOLEAN
+	Drop all unsolicited neighbor advertisements, for example if there's
+	a known good NA proxy on the network and such frames need not be used
+	(or in the case of 802.11, must not be used to prevent attacks.)
+
+	By default this is turned off.
+
 icmp/*:
 ratelimit - INTEGER
 	Limit the maximal rates for sending ICMPv6 packets.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 4a4c1ae..4b2267e 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -56,6 +56,7 @@ struct ipv6_devconf {
 	__s32           ndisc_notify;
 	__s32		suppress_frag_ndisc;
 	__s32		accept_ra_mtu;
+	__s32		drop_unsolicited_na;
 	struct ipv6_stable_secret {
 		bool initialized;
 		struct in6_addr secret;
diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h
index 4c41357..ec117b6 100644
--- a/include/uapi/linux/ipv6.h
+++ b/include/uapi/linux/ipv6.h
@@ -175,6 +175,7 @@ enum {
 	DEVCONF_ACCEPT_RA_MIN_HOP_LIMIT,
 	DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN,
 	DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
+	DEVCONF_DROP_UNSOLICITED_NA,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 23e325f..ac0ba9e 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -4712,6 +4712,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	/* we omit DEVCONF_STABLE_SECRET for now */
 	array[DEVCONF_USE_OIF_ADDRS_ONLY] = cnf->use_oif_addrs_only;
 	array[DEVCONF_DROP_UNICAST_IN_L2_MULTICAST] = cnf->drop_unicast_in_l2_multicast;
+	array[DEVCONF_DROP_UNSOLICITED_NA] = cnf->drop_unsolicited_na;
 }
 
 static inline size_t inet6_ifla6_size(void)
@@ -5793,6 +5794,13 @@ static struct addrconf_sysctl_table
 			.proc_handler	= proc_dointvec,
 		},
 		{
+			.procname	= "drop_unsolicited_na",
+			.data		= &ipv6_devconf.drop_unsolicited_na,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= proc_dointvec,
+		},
+		{
 			/* sentinel */
 		}
 	},
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 84afb9a..c245895 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -883,6 +883,7 @@ static void ndisc_recv_na(struct sk_buff *skb)
 				    offsetof(struct nd_msg, opt));
 	struct ndisc_options ndopts;
 	struct net_device *dev = skb->dev;
+	struct inet6_dev *idev = __in6_dev_get(dev);
 	struct inet6_ifaddr *ifp;
 	struct neighbour *neigh;
 
@@ -902,6 +903,14 @@ static void ndisc_recv_na(struct sk_buff *skb)
 		return;
 	}
 
+	/* For some 802.11 wireless deployments (and possibly other networks),
+	 * there will be a NA proxy and unsolicitd packets are attacks
+	 * and thus should not be accepted.
+	 */
+	if (!msg->icmph.icmp6_solicited && idev &&
+	    idev->cnf.drop_unsolicited_na)
+		return;
+
 	if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) {
 		ND_PRINTK(2, warn, "NS: invalid ND option\n");
 		return;
-- 
cgit v0.10.2


From 7f20cd252185702f951009e0a56778f870d50ca6 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Thu, 4 Feb 2016 17:42:28 +0100
Subject: bonding: 3ad: allow to set ad_actor settings while the bond is up

No need to require the bond down while changing these settings, the change
will be reflected immediately and the 3ad mode will sort itself out.
For faster convergence set port->ntt to true in order to generate new
LACPDUs immediately.

CC: Jay Vosburgh <j.vosburgh@gmail.com>
CC: Veaceslav Falico <vfalico@gmail.com>
CC: Andy Gospodarek <gospo@cumulusnetworks.com>
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index ee94056..b9304a2 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -2176,8 +2176,12 @@ void bond_3ad_update_ad_actor_settings(struct bonding *bond)
 		    *((struct mac_addr *)bond->params.ad_actor_system);
 
 	spin_lock_bh(&bond->mode_lock);
-	bond_for_each_slave(bond, slave, iter)
-		__ad_actor_update_port(&(SLAVE_AD_INFO(slave)->port));
+	bond_for_each_slave(bond, slave, iter) {
+		struct port *port = &(SLAVE_AD_INFO(slave))->port;
+
+		__ad_actor_update_port(port);
+		port->ntt = true;
+	}
 	spin_unlock_bh(&bond->mode_lock);
 }
 
diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index ed0bdae..577e57c 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -402,7 +402,6 @@ static const struct bond_option bond_opts[BOND_OPT_LAST] = {
 		.id = BOND_OPT_AD_ACTOR_SYS_PRIO,
 		.name = "ad_actor_sys_prio",
 		.unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_8023AD)),
-		.flags = BOND_OPTFLAG_IFDOWN,
 		.values = bond_ad_actor_sys_prio_tbl,
 		.set = bond_option_ad_actor_sys_prio_set,
 	},
@@ -410,7 +409,7 @@ static const struct bond_option bond_opts[BOND_OPT_LAST] = {
 		.id = BOND_OPT_AD_ACTOR_SYSTEM,
 		.name = "ad_actor_system",
 		.unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_8023AD)),
-		.flags = BOND_OPTFLAG_RAWVAL | BOND_OPTFLAG_IFDOWN,
+		.flags = BOND_OPTFLAG_RAWVAL,
 		.set = bond_option_ad_actor_system_set,
 	},
 	[BOND_OPT_AD_USER_PORT_KEY] = {
-- 
cgit v0.10.2


From 55cd48c821def1d3192f166b4ce48254cf2e5a69 Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:43 -0700
Subject: net: fec: stop the "rcv is not +last, " error messages

Setting the FTRL register will stop the fec from
trying to use multiple receive buffers.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index 2106d72..cc9677a 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -64,6 +64,7 @@
 #define FEC_R_FIFO_RSEM		0x194 /* Receive FIFO section empty threshold */
 #define FEC_R_FIFO_RAEM		0x198 /* Receive FIFO almost empty threshold */
 #define FEC_R_FIFO_RAFL		0x19c /* Receive FIFO almost full threshold */
+#define FEC_FTRL		0x1b0 /* Frame truncation receive length*/
 #define FEC_RACC		0x1c4 /* Receive Accelerator function */
 #define FEC_RCMR_1		0x1c8 /* Receive classification match ring 1 */
 #define FEC_RCMR_2		0x1cc /* Receive classification match ring 2 */
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 41c81f6..3e5b24a 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -988,6 +988,7 @@ fec_restart(struct net_device *ndev)
 			val &= ~FEC_RACC_OPTIONS;
 		writel(val, fep->hwp + FEC_RACC);
 	}
+	writel(PKT_MAXBUF_SIZE, fep->hwp + FEC_FTRL);
 #endif
 
 	/*
-- 
cgit v0.10.2


From 095098e194d1357a509c1d8e9a421bf825dddd2a Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:44 -0700
Subject: net: fec: fix rx error counts

On an overrun, the other flags are not
valid, so don't check them.

Also, don't pass bad frames up the stack.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 3e5b24a..162fa59 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -1408,37 +1408,31 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 			break;
 		pkt_received++;
 
-		/* Since we have allocated space to hold a complete frame,
-		 * the last indicator should be set.
-		 */
-		if ((status & BD_ENET_RX_LAST) == 0)
-			netdev_err(ndev, "rcv is not +last\n");
-
 		writel(FEC_ENET_RXF, fep->hwp + FEC_IEVENT);
 
 		/* Check for errors. */
+		status ^= BD_ENET_RX_LAST;
 		if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH | BD_ENET_RX_NO |
-			   BD_ENET_RX_CR | BD_ENET_RX_OV)) {
+			   BD_ENET_RX_CR | BD_ENET_RX_OV | BD_ENET_RX_LAST |
+			   BD_ENET_RX_CL)) {
 			ndev->stats.rx_errors++;
-			if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH)) {
+			if (status & BD_ENET_RX_OV) {
+				/* FIFO overrun */
+				ndev->stats.rx_fifo_errors++;
+				goto rx_processing_done;
+			}
+			if (status & (BD_ENET_RX_LG | BD_ENET_RX_SH
+						| BD_ENET_RX_LAST)) {
 				/* Frame too long or too short. */
 				ndev->stats.rx_length_errors++;
+				if (status & BD_ENET_RX_LAST)
+					netdev_err(ndev, "rcv is not +last\n");
 			}
-			if (status & BD_ENET_RX_NO)	/* Frame alignment */
-				ndev->stats.rx_frame_errors++;
 			if (status & BD_ENET_RX_CR)	/* CRC Error */
 				ndev->stats.rx_crc_errors++;
-			if (status & BD_ENET_RX_OV)	/* FIFO overrun */
-				ndev->stats.rx_fifo_errors++;
-		}
-
-		/* Report late collisions as a frame error.
-		 * On this error, the BD is closed, but we don't know what we
-		 * have in the buffer.  So, just drop this frame on the floor.
-		 */
-		if (status & BD_ENET_RX_CL) {
-			ndev->stats.rx_errors++;
-			ndev->stats.rx_frame_errors++;
+			/* Report late collisions as a frame error. */
+			if (status & (BD_ENET_RX_NO | BD_ENET_RX_CL))
+				ndev->stats.rx_frame_errors++;
 			goto rx_processing_done;
 		}
 
-- 
cgit v0.10.2


From 93c595f7b8c5fe98bd78bab5270a49c40a69f221 Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:45 -0700
Subject: net: fec: fix fec_enet_get_free_txdesc_num

When first initialized, cur_tx points to the 1st
entry in the queue, and dirty_tx points to the last.
At this point, fec_enet_get_free_txdesc_num will
return tx_ring_size -2. If tx_ring_size -2 entries
are now queued, then fec_enet_get_free_txdesc_num
should return 0, but it returns tx_ring_size instead.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 162fa59..adbddfd 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -296,7 +296,7 @@ static int fec_enet_get_free_txdesc_num(struct fec_enet_private *fep,
 	entries = ((const char *)txq->dirty_tx -
 			(const char *)txq->cur_tx) / fep->bufdesc_size - 1;
 
-	return entries > 0 ? entries : entries + txq->tx_ring_size;
+	return entries >= 0 ? entries : entries + txq->tx_ring_size;
 }
 
 static void swap_buffer(void *bufaddr, int len)
-- 
cgit v0.10.2


From 7355f2760620b385d1335f9cf73fbb1b640c4726 Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:46 -0700
Subject: net: fec: add struct bufdesc_prop

This reduces code and gains speed.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index cc9677a..53ec04f 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -448,33 +448,34 @@ struct bufdesc_ex {
 /* Controller supports RACC register */
 #define FEC_QUIRK_HAS_RACC		(1 << 12)
 
+struct bufdesc_prop {
+	int qid;
+	/* Address of Rx and Tx buffers */
+	struct bufdesc	*base;
+	struct bufdesc	*last;
+	struct bufdesc	*cur;
+	dma_addr_t	dma;
+	unsigned short ring_size;
+	unsigned char dsize;
+	unsigned char dsize_log2;
+};
+
 struct fec_enet_priv_tx_q {
-	int index;
+	struct bufdesc_prop bd;
 	unsigned char *tx_bounce[TX_RING_SIZE];
 	struct  sk_buff *tx_skbuff[TX_RING_SIZE];
 
-	dma_addr_t	bd_dma;
-	struct bufdesc	*tx_bd_base;
-	uint tx_ring_size;
-
 	unsigned short tx_stop_threshold;
 	unsigned short tx_wake_threshold;
 
-	struct bufdesc	*cur_tx;
 	struct bufdesc	*dirty_tx;
 	char *tso_hdrs;
 	dma_addr_t tso_hdrs_dma;
 };
 
 struct fec_enet_priv_rx_q {
-	int index;
+	struct bufdesc_prop bd;
 	struct  sk_buff *rx_skbuff[RX_RING_SIZE];
-
-	dma_addr_t	bd_dma;
-	struct bufdesc	*rx_bd_base;
-	uint rx_ring_size;
-
-	struct bufdesc	*cur_rx;
 };
 
 /* The FEC buffer descriptors track the ring buffers.  The rx_bd_base and
@@ -514,8 +515,6 @@ struct fec_enet_private {
 	unsigned long work_ts;
 	unsigned long work_mdio;
 
-	unsigned short bufdesc_size;
-
 	struct	platform_device *pdev;
 
 	int	dev_id;
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index adbddfd..b039288 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -217,86 +217,38 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address");
 
 #define IS_TSO_HEADER(txq, addr) \
 	((addr >= txq->tso_hdrs_dma) && \
-	(addr < txq->tso_hdrs_dma + txq->tx_ring_size * TSO_HEADER_SIZE))
+	(addr < txq->tso_hdrs_dma + txq->bd.ring_size * TSO_HEADER_SIZE))
 
 static int mii_cnt;
 
-static inline
-struct bufdesc *fec_enet_get_nextdesc(struct bufdesc *bdp,
-				      struct fec_enet_private *fep,
-				      int queue_id)
-{
-	struct bufdesc *new_bd = bdp + 1;
-	struct bufdesc_ex *ex_new_bd = (struct bufdesc_ex *)bdp + 1;
-	struct fec_enet_priv_tx_q *txq = fep->tx_queue[queue_id];
-	struct fec_enet_priv_rx_q *rxq = fep->rx_queue[queue_id];
-	struct bufdesc_ex *ex_base;
-	struct bufdesc *base;
-	int ring_size;
-
-	if (bdp >= txq->tx_bd_base) {
-		base = txq->tx_bd_base;
-		ring_size = txq->tx_ring_size;
-		ex_base = (struct bufdesc_ex *)txq->tx_bd_base;
-	} else {
-		base = rxq->rx_bd_base;
-		ring_size = rxq->rx_ring_size;
-		ex_base = (struct bufdesc_ex *)rxq->rx_bd_base;
-	}
-
-	if (fep->bufdesc_ex)
-		return (struct bufdesc *)((ex_new_bd >= (ex_base + ring_size)) ?
-			ex_base : ex_new_bd);
-	else
-		return (new_bd >= (base + ring_size)) ?
-			base : new_bd;
-}
-
-static inline
-struct bufdesc *fec_enet_get_prevdesc(struct bufdesc *bdp,
-				      struct fec_enet_private *fep,
-				      int queue_id)
-{
-	struct bufdesc *new_bd = bdp - 1;
-	struct bufdesc_ex *ex_new_bd = (struct bufdesc_ex *)bdp - 1;
-	struct fec_enet_priv_tx_q *txq = fep->tx_queue[queue_id];
-	struct fec_enet_priv_rx_q *rxq = fep->rx_queue[queue_id];
-	struct bufdesc_ex *ex_base;
-	struct bufdesc *base;
-	int ring_size;
-
-	if (bdp >= txq->tx_bd_base) {
-		base = txq->tx_bd_base;
-		ring_size = txq->tx_ring_size;
-		ex_base = (struct bufdesc_ex *)txq->tx_bd_base;
-	} else {
-		base = rxq->rx_bd_base;
-		ring_size = rxq->rx_ring_size;
-		ex_base = (struct bufdesc_ex *)rxq->rx_bd_base;
-	}
+static struct bufdesc *fec_enet_get_nextdesc(struct bufdesc *bdp,
+					     struct bufdesc_prop *bd)
+{
+	return (bdp >= bd->last) ? bd->base
+			: (struct bufdesc *)(((unsigned)bdp) + bd->dsize);
+}
 
-	if (fep->bufdesc_ex)
-		return (struct bufdesc *)((ex_new_bd < ex_base) ?
-			(ex_new_bd + ring_size) : ex_new_bd);
-	else
-		return (new_bd < base) ? (new_bd + ring_size) : new_bd;
+static struct bufdesc *fec_enet_get_prevdesc(struct bufdesc *bdp,
+					     struct bufdesc_prop *bd)
+{
+	return (bdp <= bd->base) ? bd->last
+			: (struct bufdesc *)(((unsigned)bdp) - bd->dsize);
 }
 
-static int fec_enet_get_bd_index(struct bufdesc *base, struct bufdesc *bdp,
-				struct fec_enet_private *fep)
+static int fec_enet_get_bd_index(struct bufdesc *bdp,
+				 struct bufdesc_prop *bd)
 {
-	return ((const char *)bdp - (const char *)base) / fep->bufdesc_size;
+	return ((const char *)bdp - (const char *)bd->base) >> bd->dsize_log2;
 }
 
-static int fec_enet_get_free_txdesc_num(struct fec_enet_private *fep,
-					struct fec_enet_priv_tx_q *txq)
+static int fec_enet_get_free_txdesc_num(struct fec_enet_priv_tx_q *txq)
 {
 	int entries;
 
-	entries = ((const char *)txq->dirty_tx -
-			(const char *)txq->cur_tx) / fep->bufdesc_size - 1;
+	entries = (((const char *)txq->dirty_tx -
+			(const char *)txq->bd.cur) >> txq->bd.dsize_log2) - 1;
 
-	return entries >= 0 ? entries : entries + txq->tx_ring_size;
+	return entries >= 0 ? entries : entries + txq->bd.ring_size;
 }
 
 static void swap_buffer(void *bufaddr, int len)
@@ -329,20 +281,20 @@ static void fec_dump(struct net_device *ndev)
 	pr_info("Nr     SC     addr       len  SKB\n");
 
 	txq = fep->tx_queue[0];
-	bdp = txq->tx_bd_base;
+	bdp = txq->bd.base;
 
 	do {
 		pr_info("%3u %c%c 0x%04x 0x%08x %4u %p\n",
 			index,
-			bdp == txq->cur_tx ? 'S' : ' ',
+			bdp == txq->bd.cur ? 'S' : ' ',
 			bdp == txq->dirty_tx ? 'H' : ' ',
 			fec16_to_cpu(bdp->cbd_sc),
 			fec32_to_cpu(bdp->cbd_bufaddr),
 			fec16_to_cpu(bdp->cbd_datlen),
 			txq->tx_skbuff[index]);
-		bdp = fec_enet_get_nextdesc(bdp, fep, 0);
+		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 		index++;
-	} while (bdp != txq->tx_bd_base);
+	} while (bdp != txq->bd.base);
 }
 
 static inline bool is_ipv4_pkt(struct sk_buff *skb)
@@ -373,7 +325,7 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 			     struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct bufdesc *bdp = txq->cur_tx;
+	struct bufdesc *bdp = txq->bd.cur;
 	struct bufdesc_ex *ebdp;
 	int nr_frags = skb_shinfo(skb)->nr_frags;
 	unsigned short queue = skb_get_queue_mapping(skb);
@@ -388,7 +340,7 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 
 	for (frag = 0; frag < nr_frags; frag++) {
 		this_frag = &skb_shinfo(skb)->frags[frag];
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue);
+		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 		ebdp = (struct bufdesc_ex *)bdp;
 
 		status = fec16_to_cpu(bdp->cbd_sc);
@@ -418,7 +370,7 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 
 		bufaddr = page_address(this_frag->page.p) + this_frag->page_offset;
 
-		index = fec_enet_get_bd_index(txq->tx_bd_base, bdp, fep);
+		index = fec_enet_get_bd_index(bdp, &txq->bd);
 		if (((unsigned long) bufaddr) & fep->tx_align ||
 			fep->quirks & FEC_QUIRK_SWAP_FRAME) {
 			memcpy(txq->tx_bounce[index], bufaddr, frag_len);
@@ -444,9 +396,9 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 
 	return bdp;
 dma_mapping_error:
-	bdp = txq->cur_tx;
+	bdp = txq->bd.cur;
 	for (i = 0; i < frag; i++) {
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue);
+		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 		dma_unmap_single(&fep->pdev->dev, fec32_to_cpu(bdp->cbd_bufaddr),
 				 fec16_to_cpu(bdp->cbd_datlen), DMA_TO_DEVICE);
 	}
@@ -468,7 +420,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	unsigned int index;
 	int entries_free;
 
-	entries_free = fec_enet_get_free_txdesc_num(fep, txq);
+	entries_free = fec_enet_get_free_txdesc_num(txq);
 	if (entries_free < MAX_SKB_FRAGS + 1) {
 		dev_kfree_skb_any(skb);
 		if (net_ratelimit())
@@ -483,7 +435,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	}
 
 	/* Fill in a Tx ring entry */
-	bdp = txq->cur_tx;
+	bdp = txq->bd.cur;
 	last_bdp = bdp;
 	status = fec16_to_cpu(bdp->cbd_sc);
 	status &= ~BD_ENET_TX_STATS;
@@ -493,7 +445,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	buflen = skb_headlen(skb);
 
 	queue = skb_get_queue_mapping(skb);
-	index = fec_enet_get_bd_index(txq->tx_bd_base, bdp, fep);
+	index = fec_enet_get_bd_index(bdp, &txq->bd);
 	if (((unsigned long) bufaddr) & fep->tx_align ||
 		fep->quirks & FEC_QUIRK_SWAP_FRAME) {
 		memcpy(txq->tx_bounce[index], skb->data, buflen);
@@ -544,7 +496,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 		ebdp->cbd_esc = cpu_to_fec32(estatus);
 	}
 
-	index = fec_enet_get_bd_index(txq->tx_bd_base, last_bdp, fep);
+	index = fec_enet_get_bd_index(last_bdp, &txq->bd);
 	/* Save skb pointer */
 	txq->tx_skbuff[index] = skb;
 
@@ -558,15 +510,15 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	bdp->cbd_sc = cpu_to_fec16(status);
 
 	/* If this was the last BD in the ring, start at the beginning again. */
-	bdp = fec_enet_get_nextdesc(last_bdp, fep, queue);
+	bdp = fec_enet_get_nextdesc(last_bdp, &txq->bd);
 
 	skb_tx_timestamp(skb);
 
 	/* Make sure the update to bdp and tx_skbuff are performed before
-	 * cur_tx.
+	 * txq->bd.cur.
 	 */
 	wmb();
-	txq->cur_tx = bdp;
+	txq->bd.cur = bdp;
 
 	/* Trigger transmission start */
 	writel(0, fep->hwp + FEC_X_DES_ACTIVE(queue));
@@ -697,13 +649,13 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
 	int total_len, data_left;
-	struct bufdesc *bdp = txq->cur_tx;
+	struct bufdesc *bdp = txq->bd.cur;
 	unsigned short queue = skb_get_queue_mapping(skb);
 	struct tso_t tso;
 	unsigned int index = 0;
 	int ret;
 
-	if (tso_count_descs(skb) >= fec_enet_get_free_txdesc_num(fep, txq)) {
+	if (tso_count_descs(skb) >= fec_enet_get_free_txdesc_num(txq)) {
 		dev_kfree_skb_any(skb);
 		if (net_ratelimit())
 			netdev_err(ndev, "NOT enough BD for TSO!\n");
@@ -723,7 +675,7 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 	while (total_len > 0) {
 		char *hdr;
 
-		index = fec_enet_get_bd_index(txq->tx_bd_base, bdp, fep);
+		index = fec_enet_get_bd_index(bdp, &txq->bd);
 		data_left = min_t(int, skb_shinfo(skb)->gso_size, total_len);
 		total_len -= data_left;
 
@@ -738,9 +690,8 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 			int size;
 
 			size = min_t(int, tso.size, data_left);
-			bdp = fec_enet_get_nextdesc(bdp, fep, queue);
-			index = fec_enet_get_bd_index(txq->tx_bd_base,
-						      bdp, fep);
+			bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
+			index = fec_enet_get_bd_index(bdp, &txq->bd);
 			ret = fec_enet_txq_put_data_tso(txq, skb, ndev,
 							bdp, index,
 							tso.data, size,
@@ -753,14 +704,14 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 			tso_build_data(skb, &tso, size);
 		}
 
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue);
+		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 	}
 
 	/* Save skb pointer */
 	txq->tx_skbuff[index] = skb;
 
 	skb_tx_timestamp(skb);
-	txq->cur_tx = bdp;
+	txq->bd.cur = bdp;
 
 	/* Trigger transmission start */
 	if (!(fep->quirks & FEC_QUIRK_ERR007885) ||
@@ -798,7 +749,7 @@ fec_enet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	if (ret)
 		return ret;
 
-	entries_free = fec_enet_get_free_txdesc_num(fep, txq);
+	entries_free = fec_enet_get_free_txdesc_num(txq);
 	if (entries_free <= txq->tx_stop_threshold)
 		netif_tx_stop_queue(nq);
 
@@ -819,32 +770,32 @@ static void fec_enet_bd_init(struct net_device *dev)
 	for (q = 0; q < fep->num_rx_queues; q++) {
 		/* Initialize the receive buffer descriptors. */
 		rxq = fep->rx_queue[q];
-		bdp = rxq->rx_bd_base;
+		bdp = rxq->bd.base;
 
-		for (i = 0; i < rxq->rx_ring_size; i++) {
+		for (i = 0; i < rxq->bd.ring_size; i++) {
 
 			/* Initialize the BD for every fragment in the page. */
 			if (bdp->cbd_bufaddr)
 				bdp->cbd_sc = cpu_to_fec16(BD_ENET_RX_EMPTY);
 			else
 				bdp->cbd_sc = cpu_to_fec16(0);
-			bdp = fec_enet_get_nextdesc(bdp, fep, q);
+			bdp = fec_enet_get_nextdesc(bdp, &rxq->bd);
 		}
 
 		/* Set the last buffer to wrap */
-		bdp = fec_enet_get_prevdesc(bdp, fep, q);
+		bdp = fec_enet_get_prevdesc(bdp, &rxq->bd);
 		bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP);
 
-		rxq->cur_rx = rxq->rx_bd_base;
+		rxq->bd.cur = rxq->bd.base;
 	}
 
 	for (q = 0; q < fep->num_tx_queues; q++) {
 		/* ...and the same for transmit */
 		txq = fep->tx_queue[q];
-		bdp = txq->tx_bd_base;
-		txq->cur_tx = bdp;
+		bdp = txq->bd.base;
+		txq->bd.cur = bdp;
 
-		for (i = 0; i < txq->tx_ring_size; i++) {
+		for (i = 0; i < txq->bd.ring_size; i++) {
 			/* Initialize the BD for every fragment in the page. */
 			bdp->cbd_sc = cpu_to_fec16(0);
 			if (txq->tx_skbuff[i]) {
@@ -852,11 +803,11 @@ static void fec_enet_bd_init(struct net_device *dev)
 				txq->tx_skbuff[i] = NULL;
 			}
 			bdp->cbd_bufaddr = cpu_to_fec32(0);
-			bdp = fec_enet_get_nextdesc(bdp, fep, q);
+			bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 		}
 
 		/* Set the last buffer to wrap */
-		bdp = fec_enet_get_prevdesc(bdp, fep, q);
+		bdp = fec_enet_get_prevdesc(bdp, &txq->bd);
 		bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP);
 		txq->dirty_tx = bdp;
 	}
@@ -880,7 +831,7 @@ static void fec_enet_enable_ring(struct net_device *ndev)
 
 	for (i = 0; i < fep->num_rx_queues; i++) {
 		rxq = fep->rx_queue[i];
-		writel(rxq->bd_dma, fep->hwp + FEC_R_DES_START(i));
+		writel(rxq->bd.dma, fep->hwp + FEC_R_DES_START(i));
 		writel(PKT_MAXBLR_SIZE, fep->hwp + FEC_R_BUFF_SIZE(i));
 
 		/* enable DMA1/2 */
@@ -891,7 +842,7 @@ static void fec_enet_enable_ring(struct net_device *ndev)
 
 	for (i = 0; i < fep->num_tx_queues; i++) {
 		txq = fep->tx_queue[i];
-		writel(txq->bd_dma, fep->hwp + FEC_X_DES_START(i));
+		writel(txq->bd.dma, fep->hwp + FEC_X_DES_START(i));
 
 		/* enable DMA1/2 */
 		if (i)
@@ -909,7 +860,7 @@ static void fec_enet_reset_skb(struct net_device *ndev)
 	for (i = 0; i < fep->num_tx_queues; i++) {
 		txq = fep->tx_queue[i];
 
-		for (j = 0; j < txq->tx_ring_size; j++) {
+		for (j = 0; j < txq->bd.ring_size; j++) {
 			if (txq->tx_skbuff[j]) {
 				dev_kfree_skb_any(txq->tx_skbuff[j]);
 				txq->tx_skbuff[j] = NULL;
@@ -1222,16 +1173,16 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 	bdp = txq->dirty_tx;
 
 	/* get next bdp of dirty_tx */
-	bdp = fec_enet_get_nextdesc(bdp, fep, queue_id);
+	bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 
-	while (bdp != READ_ONCE(txq->cur_tx)) {
-		/* Order the load of cur_tx and cbd_sc */
+	while (bdp != READ_ONCE(txq->bd.cur)) {
+		/* Order the load of bd.cur and cbd_sc */
 		rmb();
 		status = fec16_to_cpu(READ_ONCE(bdp->cbd_sc));
 		if (status & BD_ENET_TX_READY)
 			break;
 
-		index = fec_enet_get_bd_index(txq->tx_bd_base, bdp, fep);
+		index = fec_enet_get_bd_index(bdp, &txq->bd);
 
 		skb = txq->tx_skbuff[index];
 		txq->tx_skbuff[index] = NULL;
@@ -1242,7 +1193,7 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 					 DMA_TO_DEVICE);
 		bdp->cbd_bufaddr = cpu_to_fec32(0);
 		if (!skb) {
-			bdp = fec_enet_get_nextdesc(bdp, fep, queue_id);
+			bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 			continue;
 		}
 
@@ -1291,19 +1242,19 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 		txq->dirty_tx = bdp;
 
 		/* Update pointer to next buffer descriptor to be transmitted */
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue_id);
+		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 
 		/* Since we have freed up a buffer, the ring is no longer full
 		 */
 		if (netif_queue_stopped(ndev)) {
-			entries_free = fec_enet_get_free_txdesc_num(fep, txq);
+			entries_free = fec_enet_get_free_txdesc_num(txq);
 			if (entries_free >= txq->tx_wake_threshold)
 				netif_tx_wake_queue(nq);
 		}
 	}
 
 	/* ERR006538: Keep the transmitter going */
-	if (bdp != txq->cur_tx &&
+	if (bdp != txq->bd.cur &&
 	    readl(fep->hwp + FEC_X_DES_ACTIVE(queue_id)) == 0)
 		writel(0, fep->hwp + FEC_X_DES_ACTIVE(queue_id));
 }
@@ -1367,7 +1318,7 @@ static bool fec_enet_copybreak(struct net_device *ndev, struct sk_buff **skb,
 	return true;
 }
 
-/* During a receive, the cur_rx points to the current incoming buffer.
+/* During a receive, the bd_rx.cur points to the current incoming buffer.
  * When we update through the ring, if the next incoming buffer has
  * not been given to the system, we just set the empty indicator,
  * effectively tossing the packet.
@@ -1400,7 +1351,7 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 	/* First, grab all of the stats for the incoming packet.
 	 * These get messed up if we get called due to a busy condition.
 	 */
-	bdp = rxq->cur_rx;
+	bdp = rxq->bd.cur;
 
 	while (!((status = fec16_to_cpu(bdp->cbd_sc)) & BD_ENET_RX_EMPTY)) {
 
@@ -1441,7 +1392,7 @@ fec_enet_rx_queue(struct net_device *ndev, int budget, u16 queue_id)
 		pkt_len = fec16_to_cpu(bdp->cbd_datlen);
 		ndev->stats.rx_bytes += pkt_len;
 
-		index = fec_enet_get_bd_index(rxq->rx_bd_base, bdp, fep);
+		index = fec_enet_get_bd_index(bdp, &rxq->bd);
 		skb = rxq->rx_skbuff[index];
 
 		/* The packet length includes FCS, but we don't want to
@@ -1541,7 +1492,7 @@ rx_processing_done:
 		}
 
 		/* Update BD pointer to next entry */
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue_id);
+		bdp = fec_enet_get_nextdesc(bdp, &rxq->bd);
 
 		/* Doing this here will keep the FEC running while we process
 		 * incoming frames.  On a heavily loaded network, we should be
@@ -1549,7 +1500,7 @@ rx_processing_done:
 		 */
 		writel(0, fep->hwp + FEC_R_DES_ACTIVE(queue_id));
 	}
-	rxq->cur_rx = bdp;
+	rxq->bd.cur = bdp;
 	return pkt_received;
 }
 
@@ -2658,8 +2609,8 @@ static void fec_enet_free_buffers(struct net_device *ndev)
 
 	for (q = 0; q < fep->num_rx_queues; q++) {
 		rxq = fep->rx_queue[q];
-		bdp = rxq->rx_bd_base;
-		for (i = 0; i < rxq->rx_ring_size; i++) {
+		bdp = rxq->bd.base;
+		for (i = 0; i < rxq->bd.ring_size; i++) {
 			skb = rxq->rx_skbuff[i];
 			rxq->rx_skbuff[i] = NULL;
 			if (skb) {
@@ -2669,14 +2620,14 @@ static void fec_enet_free_buffers(struct net_device *ndev)
 						 DMA_FROM_DEVICE);
 				dev_kfree_skb(skb);
 			}
-			bdp = fec_enet_get_nextdesc(bdp, fep, q);
+			bdp = fec_enet_get_nextdesc(bdp, &rxq->bd);
 		}
 	}
 
 	for (q = 0; q < fep->num_tx_queues; q++) {
 		txq = fep->tx_queue[q];
-		bdp = txq->tx_bd_base;
-		for (i = 0; i < txq->tx_ring_size; i++) {
+		bdp = txq->bd.base;
+		for (i = 0; i < txq->bd.ring_size; i++) {
 			kfree(txq->tx_bounce[i]);
 			txq->tx_bounce[i] = NULL;
 			skb = txq->tx_skbuff[i];
@@ -2696,7 +2647,7 @@ static void fec_enet_free_queue(struct net_device *ndev)
 		if (fep->tx_queue[i] && fep->tx_queue[i]->tso_hdrs) {
 			txq = fep->tx_queue[i];
 			dma_free_coherent(NULL,
-					  txq->tx_ring_size * TSO_HEADER_SIZE,
+					  txq->bd.ring_size * TSO_HEADER_SIZE,
 					  txq->tso_hdrs,
 					  txq->tso_hdrs_dma);
 		}
@@ -2722,15 +2673,15 @@ static int fec_enet_alloc_queue(struct net_device *ndev)
 		}
 
 		fep->tx_queue[i] = txq;
-		txq->tx_ring_size = TX_RING_SIZE;
-		fep->total_tx_ring_size += fep->tx_queue[i]->tx_ring_size;
+		txq->bd.ring_size = TX_RING_SIZE;
+		fep->total_tx_ring_size += fep->tx_queue[i]->bd.ring_size;
 
 		txq->tx_stop_threshold = FEC_MAX_SKB_DESCS;
 		txq->tx_wake_threshold =
-				(txq->tx_ring_size - txq->tx_stop_threshold) / 2;
+			(txq->bd.ring_size - txq->tx_stop_threshold) / 2;
 
 		txq->tso_hdrs = dma_alloc_coherent(NULL,
-					txq->tx_ring_size * TSO_HEADER_SIZE,
+					txq->bd.ring_size * TSO_HEADER_SIZE,
 					&txq->tso_hdrs_dma,
 					GFP_KERNEL);
 		if (!txq->tso_hdrs) {
@@ -2747,8 +2698,8 @@ static int fec_enet_alloc_queue(struct net_device *ndev)
 			goto alloc_failed;
 		}
 
-		fep->rx_queue[i]->rx_ring_size = RX_RING_SIZE;
-		fep->total_rx_ring_size += fep->rx_queue[i]->rx_ring_size;
+		fep->rx_queue[i]->bd.ring_size = RX_RING_SIZE;
+		fep->total_rx_ring_size += fep->rx_queue[i]->bd.ring_size;
 	}
 	return ret;
 
@@ -2767,8 +2718,8 @@ fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue)
 	struct fec_enet_priv_rx_q *rxq;
 
 	rxq = fep->rx_queue[queue];
-	bdp = rxq->rx_bd_base;
-	for (i = 0; i < rxq->rx_ring_size; i++) {
+	bdp = rxq->bd.base;
+	for (i = 0; i < rxq->bd.ring_size; i++) {
 		skb = netdev_alloc_skb(ndev, FEC_ENET_RX_FRSIZE);
 		if (!skb)
 			goto err_alloc;
@@ -2786,11 +2737,11 @@ fec_enet_alloc_rxq_buffers(struct net_device *ndev, unsigned int queue)
 			ebdp->cbd_esc = cpu_to_fec32(BD_ENET_RX_INT);
 		}
 
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue);
+		bdp = fec_enet_get_nextdesc(bdp, &rxq->bd);
 	}
 
 	/* Set the last buffer to wrap. */
-	bdp = fec_enet_get_prevdesc(bdp, fep, queue);
+	bdp = fec_enet_get_prevdesc(bdp, &rxq->bd);
 	bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP);
 	return 0;
 
@@ -2808,8 +2759,8 @@ fec_enet_alloc_txq_buffers(struct net_device *ndev, unsigned int queue)
 	struct fec_enet_priv_tx_q *txq;
 
 	txq = fep->tx_queue[queue];
-	bdp = txq->tx_bd_base;
-	for (i = 0; i < txq->tx_ring_size; i++) {
+	bdp = txq->bd.base;
+	for (i = 0; i < txq->bd.ring_size; i++) {
 		txq->tx_bounce[i] = kmalloc(FEC_ENET_TX_FRSIZE, GFP_KERNEL);
 		if (!txq->tx_bounce[i])
 			goto err_alloc;
@@ -2822,11 +2773,11 @@ fec_enet_alloc_txq_buffers(struct net_device *ndev, unsigned int queue)
 			ebdp->cbd_esc = cpu_to_fec32(BD_ENET_TX_INT);
 		}
 
-		bdp = fec_enet_get_nextdesc(bdp, fep, queue);
+		bdp = fec_enet_get_nextdesc(bdp, &txq->bd);
 	}
 
 	/* Set the last buffer to wrap. */
-	bdp = fec_enet_get_prevdesc(bdp, fep, queue);
+	bdp = fec_enet_get_prevdesc(bdp, &txq->bd);
 	bdp->cbd_sc |= cpu_to_fec16(BD_SC_WRAP);
 
 	return 0;
@@ -3117,13 +3068,15 @@ static const struct net_device_ops fec_netdev_ops = {
 static int fec_enet_init(struct net_device *ndev)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
-	struct fec_enet_priv_tx_q *txq;
-	struct fec_enet_priv_rx_q *rxq;
 	struct bufdesc *cbd_base;
 	dma_addr_t bd_dma;
 	int bd_size;
 	unsigned int i;
+	unsigned dsize = fep->bufdesc_ex ? sizeof(struct bufdesc_ex) :
+			sizeof(struct bufdesc);
+	unsigned dsize_log2 = __fls(dsize);
 
+	WARN_ON(dsize != (1 << dsize_log2));
 #if defined(CONFIG_ARM)
 	fep->rx_align = 0xf;
 	fep->tx_align = 0xf;
@@ -3134,12 +3087,7 @@ static int fec_enet_init(struct net_device *ndev)
 
 	fec_enet_alloc_queue(ndev);
 
-	if (fep->bufdesc_ex)
-		fep->bufdesc_size = sizeof(struct bufdesc_ex);
-	else
-		fep->bufdesc_size = sizeof(struct bufdesc);
-	bd_size = (fep->total_tx_ring_size + fep->total_rx_ring_size) *
-			fep->bufdesc_size;
+	bd_size = (fep->total_tx_ring_size + fep->total_rx_ring_size) * dsize;
 
 	/* Allocate memory for buffer descriptors. */
 	cbd_base = dmam_alloc_coherent(&fep->pdev->dev, bd_size, &bd_dma,
@@ -3157,33 +3105,33 @@ static int fec_enet_init(struct net_device *ndev)
 
 	/* Set receive and transmit descriptor base. */
 	for (i = 0; i < fep->num_rx_queues; i++) {
-		rxq = fep->rx_queue[i];
-		rxq->index = i;
-		rxq->rx_bd_base = (struct bufdesc *)cbd_base;
-		rxq->bd_dma = bd_dma;
-		if (fep->bufdesc_ex) {
-			bd_dma += sizeof(struct bufdesc_ex) * rxq->rx_ring_size;
-			cbd_base = (struct bufdesc *)
-				(((struct bufdesc_ex *)cbd_base) + rxq->rx_ring_size);
-		} else {
-			bd_dma += sizeof(struct bufdesc) * rxq->rx_ring_size;
-			cbd_base += rxq->rx_ring_size;
-		}
+		struct fec_enet_priv_rx_q *rxq = fep->rx_queue[i];
+		unsigned size = dsize * rxq->bd.ring_size;
+
+		rxq->bd.qid = i;
+		rxq->bd.base = cbd_base;
+		rxq->bd.cur = cbd_base;
+		rxq->bd.dma = bd_dma;
+		rxq->bd.dsize = dsize;
+		rxq->bd.dsize_log2 = dsize_log2;
+		bd_dma += size;
+		cbd_base = (struct bufdesc *)(((void *)cbd_base) + size);
+		rxq->bd.last = (struct bufdesc *)(((void *)cbd_base) - dsize);
 	}
 
 	for (i = 0; i < fep->num_tx_queues; i++) {
-		txq = fep->tx_queue[i];
-		txq->index = i;
-		txq->tx_bd_base = (struct bufdesc *)cbd_base;
-		txq->bd_dma = bd_dma;
-		if (fep->bufdesc_ex) {
-			bd_dma += sizeof(struct bufdesc_ex) * txq->tx_ring_size;
-			cbd_base = (struct bufdesc *)
-			 (((struct bufdesc_ex *)cbd_base) + txq->tx_ring_size);
-		} else {
-			bd_dma += sizeof(struct bufdesc) * txq->tx_ring_size;
-			cbd_base += txq->tx_ring_size;
-		}
+		struct fec_enet_priv_tx_q *txq = fep->tx_queue[i];
+		unsigned size = dsize * txq->bd.ring_size;
+
+		txq->bd.qid = i;
+		txq->bd.base = cbd_base;
+		txq->bd.cur = cbd_base;
+		txq->bd.dma = bd_dma;
+		txq->bd.dsize = dsize;
+		txq->bd.dsize_log2 = dsize_log2;
+		bd_dma += size;
+		cbd_base = (struct bufdesc *)(((void *)cbd_base) + size);
+		txq->bd.last = (struct bufdesc *)(((void *)cbd_base) - dsize);
 	}
 
 
-- 
cgit v0.10.2


From 53bb20d1fabacc6601e68a6a69d9d1e3aff6bd69 Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:47 -0700
Subject: net: fec: add variable reg_desc_active to speed things up

There is no need for complex macros every time we need to activate
a queue. Also, no need to call skb_get_queue_mapping when we already
know which queue it is using.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index 53ec04f..bedd28a 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -310,12 +310,6 @@ struct bufdesc_ex {
 #define FEC_R_BUFF_SIZE(X)	(((X) == 1) ? FEC_R_BUFF_SIZE_1 : \
 				(((X) == 2) ? \
 					FEC_R_BUFF_SIZE_2 : FEC_R_BUFF_SIZE_0))
-#define FEC_R_DES_ACTIVE(X)	(((X) == 1) ? FEC_R_DES_ACTIVE_1 : \
-				(((X) == 2) ? \
-				   FEC_R_DES_ACTIVE_2 : FEC_R_DES_ACTIVE_0))
-#define FEC_X_DES_ACTIVE(X)	(((X) == 1) ? FEC_X_DES_ACTIVE_1 : \
-				(((X) == 2) ? \
-				   FEC_X_DES_ACTIVE_2 : FEC_X_DES_ACTIVE_0))
 
 #define FEC_DMA_CFG(X)		(((X) == 2) ? FEC_DMA_CFG_2 : FEC_DMA_CFG_1)
 
@@ -454,6 +448,7 @@ struct bufdesc_prop {
 	struct bufdesc	*base;
 	struct bufdesc	*last;
 	struct bufdesc	*cur;
+	void __iomem	*reg_desc_active;
 	dma_addr_t	dma;
 	unsigned short ring_size;
 	unsigned char dsize;
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index b039288..712e3bb 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -328,7 +328,6 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 	struct bufdesc *bdp = txq->bd.cur;
 	struct bufdesc_ex *ebdp;
 	int nr_frags = skb_shinfo(skb)->nr_frags;
-	unsigned short queue = skb_get_queue_mapping(skb);
 	int frag, frag_len;
 	unsigned short status;
 	unsigned int estatus = 0;
@@ -361,7 +360,7 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 
 		if (fep->bufdesc_ex) {
 			if (fep->quirks & FEC_QUIRK_HAS_AVB)
-				estatus |= FEC_TX_BD_FTYPE(queue);
+				estatus |= FEC_TX_BD_FTYPE(txq->bd.qid);
 			if (skb->ip_summed == CHECKSUM_PARTIAL)
 				estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
 			ebdp->cbd_bdu = 0;
@@ -415,7 +414,6 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	dma_addr_t addr;
 	unsigned short status;
 	unsigned short buflen;
-	unsigned short queue;
 	unsigned int estatus = 0;
 	unsigned int index;
 	int entries_free;
@@ -444,7 +442,6 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	bufaddr = skb->data;
 	buflen = skb_headlen(skb);
 
-	queue = skb_get_queue_mapping(skb);
 	index = fec_enet_get_bd_index(bdp, &txq->bd);
 	if (((unsigned long) bufaddr) & fep->tx_align ||
 		fep->quirks & FEC_QUIRK_SWAP_FRAME) {
@@ -487,7 +484,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 
 		if (fep->quirks & FEC_QUIRK_HAS_AVB)
-			estatus |= FEC_TX_BD_FTYPE(queue);
+			estatus |= FEC_TX_BD_FTYPE(txq->bd.qid);
 
 		if (skb->ip_summed == CHECKSUM_PARTIAL)
 			estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
@@ -521,7 +518,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	txq->bd.cur = bdp;
 
 	/* Trigger transmission start */
-	writel(0, fep->hwp + FEC_X_DES_ACTIVE(queue));
+	writel(0, txq->bd.reg_desc_active);
 
 	return 0;
 }
@@ -534,7 +531,6 @@ fec_enet_txq_put_data_tso(struct fec_enet_priv_tx_q *txq, struct sk_buff *skb,
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	struct bufdesc_ex *ebdp = container_of(bdp, struct bufdesc_ex, desc);
-	unsigned short queue = skb_get_queue_mapping(skb);
 	unsigned short status;
 	unsigned int estatus = 0;
 	dma_addr_t addr;
@@ -566,7 +562,7 @@ fec_enet_txq_put_data_tso(struct fec_enet_priv_tx_q *txq, struct sk_buff *skb,
 
 	if (fep->bufdesc_ex) {
 		if (fep->quirks & FEC_QUIRK_HAS_AVB)
-			estatus |= FEC_TX_BD_FTYPE(queue);
+			estatus |= FEC_TX_BD_FTYPE(txq->bd.qid);
 		if (skb->ip_summed == CHECKSUM_PARTIAL)
 			estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
 		ebdp->cbd_bdu = 0;
@@ -595,7 +591,6 @@ fec_enet_txq_put_hdr_tso(struct fec_enet_priv_tx_q *txq,
 	struct fec_enet_private *fep = netdev_priv(ndev);
 	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
 	struct bufdesc_ex *ebdp = container_of(bdp, struct bufdesc_ex, desc);
-	unsigned short queue = skb_get_queue_mapping(skb);
 	void *bufaddr;
 	unsigned long dmabuf;
 	unsigned short status;
@@ -630,7 +625,7 @@ fec_enet_txq_put_hdr_tso(struct fec_enet_priv_tx_q *txq,
 
 	if (fep->bufdesc_ex) {
 		if (fep->quirks & FEC_QUIRK_HAS_AVB)
-			estatus |= FEC_TX_BD_FTYPE(queue);
+			estatus |= FEC_TX_BD_FTYPE(txq->bd.qid);
 		if (skb->ip_summed == CHECKSUM_PARTIAL)
 			estatus |= BD_ENET_TX_PINS | BD_ENET_TX_IINS;
 		ebdp->cbd_bdu = 0;
@@ -650,7 +645,6 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 	int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
 	int total_len, data_left;
 	struct bufdesc *bdp = txq->bd.cur;
-	unsigned short queue = skb_get_queue_mapping(skb);
 	struct tso_t tso;
 	unsigned int index = 0;
 	int ret;
@@ -715,11 +709,11 @@ static int fec_enet_txq_submit_tso(struct fec_enet_priv_tx_q *txq,
 
 	/* Trigger transmission start */
 	if (!(fep->quirks & FEC_QUIRK_ERR007885) ||
-	    !readl(fep->hwp + FEC_X_DES_ACTIVE(queue)) ||
-	    !readl(fep->hwp + FEC_X_DES_ACTIVE(queue)) ||
-	    !readl(fep->hwp + FEC_X_DES_ACTIVE(queue)) ||
-	    !readl(fep->hwp + FEC_X_DES_ACTIVE(queue)))
-		writel(0, fep->hwp + FEC_X_DES_ACTIVE(queue));
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active))
+		writel(0, txq->bd.reg_desc_active);
 
 	return 0;
 
@@ -819,7 +813,7 @@ static void fec_enet_active_rxring(struct net_device *ndev)
 	int i;
 
 	for (i = 0; i < fep->num_rx_queues; i++)
-		writel(0, fep->hwp + FEC_R_DES_ACTIVE(i));
+		writel(0, fep->rx_queue[i]->bd.reg_desc_active);
 }
 
 static void fec_enet_enable_ring(struct net_device *ndev)
@@ -1255,8 +1249,8 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id)
 
 	/* ERR006538: Keep the transmitter going */
 	if (bdp != txq->bd.cur &&
-	    readl(fep->hwp + FEC_X_DES_ACTIVE(queue_id)) == 0)
-		writel(0, fep->hwp + FEC_X_DES_ACTIVE(queue_id));
+	    readl(txq->bd.reg_desc_active) == 0)
+		writel(0, txq->bd.reg_desc_active);
 }
 
 static void
@@ -1498,7 +1492,7 @@ rx_processing_done:
 		 * incoming frames.  On a heavily loaded network, we should be
 		 * able to keep up at the expense of system resources.
 		 */
-		writel(0, fep->hwp + FEC_R_DES_ACTIVE(queue_id));
+		writel(0, rxq->bd.reg_desc_active);
 	}
 	rxq->bd.cur = bdp;
 	return pkt_received;
@@ -3061,6 +3055,14 @@ static const struct net_device_ops fec_netdev_ops = {
 	.ndo_set_features	= fec_set_features,
 };
 
+static const unsigned short offset_des_active_rxq[] = {
+	FEC_R_DES_ACTIVE_0, FEC_R_DES_ACTIVE_1, FEC_R_DES_ACTIVE_2
+};
+
+static const unsigned short offset_des_active_txq[] = {
+	FEC_X_DES_ACTIVE_0, FEC_X_DES_ACTIVE_1, FEC_X_DES_ACTIVE_2
+};
+
  /*
   * XXX:  We need to clean up on failure exits here.
   *
@@ -3114,6 +3116,7 @@ static int fec_enet_init(struct net_device *ndev)
 		rxq->bd.dma = bd_dma;
 		rxq->bd.dsize = dsize;
 		rxq->bd.dsize_log2 = dsize_log2;
+		rxq->bd.reg_desc_active = fep->hwp + offset_des_active_rxq[i];
 		bd_dma += size;
 		cbd_base = (struct bufdesc *)(((void *)cbd_base) + size);
 		rxq->bd.last = (struct bufdesc *)(((void *)cbd_base) - dsize);
@@ -3129,6 +3132,7 @@ static int fec_enet_init(struct net_device *ndev)
 		txq->bd.dma = bd_dma;
 		txq->bd.dsize = dsize;
 		txq->bd.dsize_log2 = dsize_log2;
+		txq->bd.reg_desc_active = fep->hwp + offset_des_active_txq[i];
 		bd_dma += size;
 		cbd_base = (struct bufdesc *)(((void *)cbd_base) + size);
 		txq->bd.last = (struct bufdesc *)(((void *)cbd_base) - dsize);
-- 
cgit v0.10.2


From 80dc6a9f8e0dfb3f2c516273b2faa29c866974b2 Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:48 -0700
Subject: net: fec: don't disable FEC_ENET_TS_TIMER interrupt

Only the interrupt routine processes this condition.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index bedd28a..195122e 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -375,6 +375,7 @@ struct bufdesc_ex {
 #define FEC_ENET_TS_TIMER       ((uint)0x00008000)
 
 #define FEC_DEFAULT_IMASK (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII | FEC_ENET_TS_TIMER)
+#define FEC_NAPI_IMASK	(FEC_ENET_MII | FEC_ENET_TS_TIMER)
 #define FEC_RX_DISABLED_IMASK (FEC_DEFAULT_IMASK & (~FEC_ENET_RXF))
 
 /* ENET interrupt coalescing macro define */
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 712e3bb..ca2708d 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -1553,7 +1553,7 @@ fec_enet_interrupt(int irq, void *dev_id)
 
 		if (napi_schedule_prep(&fep->napi)) {
 			/* Disable the NAPI interrupts */
-			writel(FEC_ENET_MII, fep->hwp + FEC_IMASK);
+			writel(FEC_NAPI_IMASK, fep->hwp + FEC_IMASK);
 			__napi_schedule(&fep->napi);
 		}
 	}
-- 
cgit v0.10.2


From be293467b87c30cfdb24f74dcb225b693e436121 Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:49 -0700
Subject: net: fec: don't transfer ownership until descriptor write is complete

If you don't own it, you shouldn't write to it.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index ca2708d..97ca72a 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -390,6 +390,10 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 
 		bdp->cbd_bufaddr = cpu_to_fec32(addr);
 		bdp->cbd_datlen = cpu_to_fec16(frag_len);
+		/* Make sure the updates to rest of the descriptor are
+		 * performed before transferring ownership.
+		 */
+		wmb();
 		bdp->cbd_sc = cpu_to_fec16(status);
 	}
 
@@ -499,6 +503,10 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 
 	bdp->cbd_datlen = cpu_to_fec16(buflen);
 	bdp->cbd_bufaddr = cpu_to_fec32(addr);
+	/* Make sure the updates to rest of the descriptor are performed before
+	 * transferring ownership.
+	 */
+	wmb();
 
 	/* Send it on its way.  Tell FEC it's ready, interrupt when done,
 	 * it's the last BD of the frame, and to put the CRC on the end.
@@ -1475,7 +1483,6 @@ rx_processing_done:
 
 		/* Mark the buffer empty */
 		status |= BD_ENET_RX_EMPTY;
-		bdp->cbd_sc = cpu_to_fec16(status);
 
 		if (fep->bufdesc_ex) {
 			struct bufdesc_ex *ebdp = (struct bufdesc_ex *)bdp;
@@ -1484,6 +1491,11 @@ rx_processing_done:
 			ebdp->cbd_prot = 0;
 			ebdp->cbd_bdu = 0;
 		}
+		/* Make sure the updates to rest of the descriptor are
+		 * performed before transferring ownership.
+		 */
+		wmb();
+		bdp->cbd_sc = cpu_to_fec16(status);
 
 		/* Update BD pointer to next entry */
 		bdp = fec_enet_get_nextdesc(bdp, &rxq->bd);
-- 
cgit v0.10.2


From fc75ba5159a63fad40442b9a9527a0f9b06de79b Mon Sep 17 00:00:00 2001
From: Troy Kisky <troy.kisky@boundarydevices.com>
Date: Fri, 5 Feb 2016 14:52:50 -0700
Subject: net: fec: improve error handling

Unmap initial buffer on error.
Don't free skb until it has been unmapped.
Move cbd_bufaddr assignment closer to the mapping function.

Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 97ca72a..ef18ca5 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -382,7 +382,6 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq,
 		addr = dma_map_single(&fep->pdev->dev, bufaddr, frag_len,
 				      DMA_TO_DEVICE);
 		if (dma_mapping_error(&fep->pdev->dev, addr)) {
-			dev_kfree_skb_any(skb);
 			if (net_ratelimit())
 				netdev_err(ndev, "Tx DMA memory map failed\n");
 			goto dma_mapping_error;
@@ -467,8 +466,12 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 
 	if (nr_frags) {
 		last_bdp = fec_enet_txq_submit_frag_skb(txq, skb, ndev);
-		if (IS_ERR(last_bdp))
+		if (IS_ERR(last_bdp)) {
+			dma_unmap_single(&fep->pdev->dev, addr,
+					 buflen, DMA_TO_DEVICE);
+			dev_kfree_skb_any(skb);
 			return NETDEV_TX_OK;
+		}
 	} else {
 		status |= (BD_ENET_TX_INTR | BD_ENET_TX_LAST);
 		if (fep->bufdesc_ex) {
@@ -478,6 +481,8 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 				estatus |= BD_ENET_TX_TS;
 		}
 	}
+	bdp->cbd_bufaddr = cpu_to_fec32(addr);
+	bdp->cbd_datlen = cpu_to_fec16(buflen);
 
 	if (fep->bufdesc_ex) {
 
@@ -501,8 +506,6 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	/* Save skb pointer */
 	txq->tx_skbuff[index] = skb;
 
-	bdp->cbd_datlen = cpu_to_fec16(buflen);
-	bdp->cbd_bufaddr = cpu_to_fec32(addr);
 	/* Make sure the updates to rest of the descriptor are performed before
 	 * transferring ownership.
 	 */
-- 
cgit v0.10.2


From 9d3053ef57599025b93aab76324ae42c60d0bac2 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Fri, 5 Feb 2016 11:43:28 +0530
Subject: iw_cxgb4: make queue allocation code more readable

Rename local mm* variables to more meaningful names

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index e99345e..dadf5f1 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -1621,7 +1621,8 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 	unsigned int sqsize, rqsize;
 	struct c4iw_ucontext *ucontext;
 	int ret;
-	struct c4iw_mm_entry *mm1, *mm2, *mm3, *mm4, *mm5 = NULL;
+	struct c4iw_mm_entry *sq_key_mm, *rq_key_mm = NULL, *sq_db_key_mm;
+	struct c4iw_mm_entry *rq_db_key_mm = NULL, *ma_sync_key_mm = NULL;
 
 	PDBG("%s ib_pd %p\n", __func__, pd);
 
@@ -1706,29 +1707,30 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 		goto err2;
 
 	if (udata) {
-		mm1 = kmalloc(sizeof *mm1, GFP_KERNEL);
-		if (!mm1) {
+		sq_key_mm = kmalloc(sizeof(*sq_key_mm), GFP_KERNEL);
+		if (!sq_key_mm) {
 			ret = -ENOMEM;
 			goto err3;
 		}
-		mm2 = kmalloc(sizeof *mm2, GFP_KERNEL);
-		if (!mm2) {
+		rq_key_mm = kmalloc(sizeof(*rq_key_mm), GFP_KERNEL);
+		if (!rq_key_mm) {
 			ret = -ENOMEM;
 			goto err4;
 		}
-		mm3 = kmalloc(sizeof *mm3, GFP_KERNEL);
-		if (!mm3) {
+		sq_db_key_mm = kmalloc(sizeof(*sq_db_key_mm), GFP_KERNEL);
+		if (!sq_db_key_mm) {
 			ret = -ENOMEM;
 			goto err5;
 		}
-		mm4 = kmalloc(sizeof *mm4, GFP_KERNEL);
-		if (!mm4) {
+		rq_db_key_mm = kmalloc(sizeof(*rq_db_key_mm), GFP_KERNEL);
+		if (!rq_db_key_mm) {
 			ret = -ENOMEM;
 			goto err6;
 		}
 		if (t4_sq_onchip(&qhp->wq.sq)) {
-			mm5 = kmalloc(sizeof *mm5, GFP_KERNEL);
-			if (!mm5) {
+			ma_sync_key_mm = kmalloc(sizeof(*ma_sync_key_mm),
+						 GFP_KERNEL);
+			if (!ma_sync_key_mm) {
 				ret = -ENOMEM;
 				goto err7;
 			}
@@ -1743,7 +1745,7 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 		uresp.rq_size = qhp->wq.rq.size;
 		uresp.rq_memsize = qhp->wq.rq.memsize;
 		spin_lock(&ucontext->mmap_lock);
-		if (mm5) {
+		if (ma_sync_key_mm) {
 			uresp.ma_sync_key = ucontext->key;
 			ucontext->key += PAGE_SIZE;
 		} else {
@@ -1761,28 +1763,29 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 		ret = ib_copy_to_udata(udata, &uresp, sizeof uresp);
 		if (ret)
 			goto err8;
-		mm1->key = uresp.sq_key;
-		mm1->addr = qhp->wq.sq.phys_addr;
-		mm1->len = PAGE_ALIGN(qhp->wq.sq.memsize);
-		insert_mmap(ucontext, mm1);
-		mm2->key = uresp.rq_key;
-		mm2->addr = virt_to_phys(qhp->wq.rq.queue);
-		mm2->len = PAGE_ALIGN(qhp->wq.rq.memsize);
-		insert_mmap(ucontext, mm2);
-		mm3->key = uresp.sq_db_gts_key;
-		mm3->addr = (__force unsigned long)qhp->wq.sq.bar2_pa;
-		mm3->len = PAGE_SIZE;
-		insert_mmap(ucontext, mm3);
-		mm4->key = uresp.rq_db_gts_key;
-		mm4->addr = (__force unsigned long)qhp->wq.rq.bar2_pa;
-		mm4->len = PAGE_SIZE;
-		insert_mmap(ucontext, mm4);
-		if (mm5) {
-			mm5->key = uresp.ma_sync_key;
-			mm5->addr = (pci_resource_start(rhp->rdev.lldi.pdev, 0)
-				    + PCIE_MA_SYNC_A) & PAGE_MASK;
-			mm5->len = PAGE_SIZE;
-			insert_mmap(ucontext, mm5);
+		sq_key_mm->key = uresp.sq_key;
+		sq_key_mm->addr = qhp->wq.sq.phys_addr;
+		sq_key_mm->len = PAGE_ALIGN(qhp->wq.sq.memsize);
+		insert_mmap(ucontext, sq_key_mm);
+		rq_key_mm->key = uresp.rq_key;
+		rq_key_mm->addr = virt_to_phys(qhp->wq.rq.queue);
+		rq_key_mm->len = PAGE_ALIGN(qhp->wq.rq.memsize);
+		insert_mmap(ucontext, rq_key_mm);
+		sq_db_key_mm->key = uresp.sq_db_gts_key;
+		sq_db_key_mm->addr = (u64)(unsigned long)qhp->wq.sq.bar2_pa;
+		sq_db_key_mm->len = PAGE_SIZE;
+		insert_mmap(ucontext, sq_db_key_mm);
+		rq_db_key_mm->key = uresp.rq_db_gts_key;
+		rq_db_key_mm->addr = (u64)(unsigned long)qhp->wq.rq.bar2_pa;
+		rq_db_key_mm->len = PAGE_SIZE;
+		insert_mmap(ucontext, rq_db_key_mm);
+		if (ma_sync_key_mm) {
+			ma_sync_key_mm->key = uresp.ma_sync_key;
+			ma_sync_key_mm->addr =
+				(pci_resource_start(rhp->rdev.lldi.pdev, 0) +
+				PCIE_MA_SYNC_A) & PAGE_MASK;
+			ma_sync_key_mm->len = PAGE_SIZE;
+			insert_mmap(ucontext, ma_sync_key_mm);
 		}
 	}
 	qhp->ibqp.qp_num = qhp->wq.sq.qid;
@@ -1795,15 +1798,15 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 	     qhp->wq.rq.memsize, attrs->cap.max_recv_wr);
 	return &qhp->ibqp;
 err8:
-	kfree(mm5);
+	kfree(ma_sync_key_mm);
 err7:
-	kfree(mm4);
+	kfree(rq_db_key_mm);
 err6:
-	kfree(mm3);
+	kfree(sq_db_key_mm);
 err5:
-	kfree(mm2);
+	kfree(rq_key_mm);
 err4:
-	kfree(mm1);
+	kfree(sq_key_mm);
 err3:
 	remove_handle(rhp, &rhp->qpidr, qhp->wq.sq.qid);
 err2:
-- 
cgit v0.10.2


From 6102c66eeb7497e4462c896074315635807758d4 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Fri, 5 Feb 2016 11:43:29 +0530
Subject: iw_cxgb4: remove false error log entry

Don't log errors if a listening endpoint is going away when procesing a
PASS_ACCEPT_REQ message.  This can happen.  Change the error printk to
a PDBG() debug log entry

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index cd2ff5f..0c2111b 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -2399,8 +2399,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 	}
 
 	if (state_read(&parent_ep->com) != LISTEN) {
-		printk(KERN_ERR "%s - listening ep not in LISTEN\n",
-		       __func__);
+		PDBG("%s - listening ep not in LISTEN\n", __func__);
 		goto reject;
 	}
 
-- 
cgit v0.10.2


From ba9cee6aa67dd67ad1bf121ee20efc00241f086f Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Fri, 5 Feb 2016 11:43:30 +0530
Subject: cxgb4/iw_cxgb4: TOS support

This series provides support for iWARP applications to specify a TOS
value and have that map to a VLAN Priority for iw_cxgb4 iWARP connections.

In iw_cxgb4, when allocating an L2T entry, pass the skb_priority based
on the tos value in the cm_id. Also pass the correct tos value during
connection setup so the passive side gets the client's desired tos.
When sending the FLOWC work request to FW, if the egress device is
in a vlan, then use the vlan priority bits as the scheduling class.
This allows associating RDMA connections with scheduling classes to
provide traffic shaping per flow.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 0c2111b..8cd7b5e 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -485,12 +485,19 @@ static void send_flowc(struct c4iw_ep *ep, struct sk_buff *skb)
 	unsigned int flowclen = 80;
 	struct fw_flowc_wr *flowc;
 	int i;
+	u16 vlan = ep->l2t->vlan;
+	int nparams;
+
+	if (vlan == CPL_L2T_VLAN_NONE)
+		nparams = 8;
+	else
+		nparams = 9;
 
 	skb = get_skb(skb, flowclen, GFP_KERNEL);
 	flowc = (struct fw_flowc_wr *)__skb_put(skb, flowclen);
 
 	flowc->op_to_nparams = cpu_to_be32(FW_WR_OP_V(FW_FLOWC_WR) |
-					   FW_FLOWC_WR_NPARAMS_V(8));
+					   FW_FLOWC_WR_NPARAMS_V(nparams));
 	flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16_V(DIV_ROUND_UP(flowclen,
 					  16)) | FW_WR_FLOWID_V(ep->hwtid));
 
@@ -511,9 +518,17 @@ static void send_flowc(struct c4iw_ep *ep, struct sk_buff *skb)
 	flowc->mnemval[6].val = cpu_to_be32(ep->snd_win);
 	flowc->mnemval[7].mnemonic = FW_FLOWC_MNEM_MSS;
 	flowc->mnemval[7].val = cpu_to_be32(ep->emss);
-	/* Pad WR to 16 byte boundary */
-	flowc->mnemval[8].mnemonic = 0;
-	flowc->mnemval[8].val = 0;
+	if (nparams == 9) {
+		u16 pri;
+
+		pri = (vlan & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
+		flowc->mnemval[8].mnemonic = FW_FLOWC_MNEM_SCHEDCLASS;
+		flowc->mnemval[8].val = cpu_to_be32(pri);
+	} else {
+		/* Pad WR to 16 byte boundary */
+		flowc->mnemval[8].mnemonic = 0;
+		flowc->mnemval[8].val = 0;
+	}
 	for (i = 0; i < 9; i++) {
 		flowc->mnemval[i].r4[0] = 0;
 		flowc->mnemval[i].r4[1] = 0;
@@ -710,7 +725,7 @@ static int send_connect(struct c4iw_ep *ep)
 	       L2T_IDX_V(ep->l2t->idx) |
 	       TX_CHAN_V(ep->tx_chan) |
 	       SMAC_SEL_V(ep->smac_idx) |
-	       DSCP_V(ep->tos) |
+	       DSCP_V(ep->tos >> 2) |
 	       ULP_MODE_V(ULP_MODE_TCPDDP) |
 	       RCV_BUFSIZ_V(win);
 	opt2 = RX_CHANNEL_V(0) |
@@ -1864,7 +1879,7 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 		L2T_IDX_V(ep->l2t->idx) |
 		TX_CHAN_V(ep->tx_chan) |
 		SMAC_SEL_V(ep->smac_idx) |
-		DSCP_V(ep->tos) |
+		DSCP_V(ep->tos >> 2) |
 		ULP_MODE_V(ULP_MODE_TCPDDP) |
 		RCV_BUFSIZ_V(win));
 	req->tcb.opt2 = (__force __be32) (PACE_V(1) |
@@ -1928,7 +1943,7 @@ static void set_tcp_window(struct c4iw_ep *ep, struct port_info *pi)
 
 static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip,
 		     struct dst_entry *dst, struct c4iw_dev *cdev,
-		     bool clear_mpa_v1, enum chip_type adapter_type)
+		     bool clear_mpa_v1, enum chip_type adapter_type, u8 tos)
 {
 	struct neighbour *n;
 	int err, step;
@@ -1958,7 +1973,7 @@ static int import_ep(struct c4iw_ep *ep, int iptype, __u8 *peer_ip,
 			goto out;
 		}
 		ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t,
-					n, pdev, 0);
+					n, pdev, rt_tos2priority(tos));
 		if (!ep->l2t)
 			goto out;
 		ep->mtu = pdev->mtu;
@@ -2041,7 +2056,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep)
 	if (ep->com.cm_id->local_addr.ss_family == AF_INET) {
 		ep->dst = find_route(ep->com.dev, laddr->sin_addr.s_addr,
 				     raddr->sin_addr.s_addr, laddr->sin_port,
-				     raddr->sin_port, 0);
+				     raddr->sin_port, ep->com.cm_id->tos);
 		iptype = 4;
 		ra = (__u8 *)&raddr->sin_addr;
 	} else {
@@ -2058,7 +2073,8 @@ static int c4iw_reconnect(struct c4iw_ep *ep)
 		goto fail3;
 	}
 	err = import_ep(ep, iptype, ra, ep->dst, ep->com.dev, false,
-			ep->com.dev->rdev.lldi.adapter_type);
+			ep->com.dev->rdev.lldi.adapter_type,
+			ep->com.cm_id->tos);
 	if (err) {
 		pr_err("%s - cannot alloc l2e.\n", __func__);
 		goto fail4;
@@ -2069,7 +2085,7 @@ static int c4iw_reconnect(struct c4iw_ep *ep)
 	     ep->l2t->idx);
 
 	state_set(&ep->com, CONNECTING);
-	ep->tos = 0;
+	ep->tos = ep->com.cm_id->tos;
 
 	/* send connect request to rnic */
 	err = send_connect(ep);
@@ -2391,6 +2407,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 	u16 peer_mss = ntohs(req->tcpopt.mss);
 	int iptype;
 	unsigned short hdrs;
+	u8 tos = PASS_OPEN_TOS_G(ntohl(req->tos_stid));
 
 	parent_ep = lookup_stid(t, stid);
 	if (!parent_ep) {
@@ -2414,7 +2431,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 		     ntohs(peer_port), peer_mss);
 		dst = find_route(dev, *(__be32 *)local_ip, *(__be32 *)peer_ip,
 				 local_port, peer_port,
-				 PASS_OPEN_TOS_G(ntohl(req->tos_stid)));
+				 tos);
 	} else {
 		PDBG("%s parent ep %p hwtid %u laddr %pI6 raddr %pI6 lport %d rport %d peer_mss %d\n"
 		     , __func__, parent_ep, hwtid,
@@ -2440,7 +2457,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 	}
 
 	err = import_ep(child_ep, iptype, peer_ip, dst, dev, false,
-			parent_ep->com.dev->rdev.lldi.adapter_type);
+			parent_ep->com.dev->rdev.lldi.adapter_type, tos);
 	if (err) {
 		printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n",
 		       __func__);
@@ -2508,7 +2525,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	c4iw_get_ep(&parent_ep->com);
 	child_ep->parent_ep = parent_ep;
-	child_ep->tos = PASS_OPEN_TOS_G(ntohl(req->tos_stid));
+	child_ep->tos = tos;
 	child_ep->dst = dst;
 	child_ep->hwtid = hwtid;
 
@@ -3202,7 +3219,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 		     ra, ntohs(raddr->sin_port));
 		ep->dst = find_route(dev, laddr->sin_addr.s_addr,
 				     raddr->sin_addr.s_addr, laddr->sin_port,
-				     raddr->sin_port, 0);
+				     raddr->sin_port, cm_id->tos);
 	} else {
 		iptype = 6;
 		ra = (__u8 *)&raddr6->sin6_addr;
@@ -3233,7 +3250,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	}
 
 	err = import_ep(ep, iptype, ra, ep->dst, ep->com.dev, true,
-			ep->com.dev->rdev.lldi.adapter_type);
+			ep->com.dev->rdev.lldi.adapter_type, cm_id->tos);
 	if (err) {
 		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
 		goto fail3;
@@ -3244,7 +3261,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 		ep->l2t->idx);
 
 	state_set(&ep->com, CONNECTING);
-	ep->tos = 0;
+	ep->tos = cm_id->tos;
 
 	/* send connect request to rnic */
 	err = send_connect(ep);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
index a072d34..1d2d1da 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
@@ -1021,6 +1021,8 @@ struct cpl_l2t_write_req {
 #define L2T_W_NOREPLY_V(x) ((x) << L2T_W_NOREPLY_S)
 #define L2T_W_NOREPLY_F    L2T_W_NOREPLY_V(1U)
 
+#define CPL_L2T_VLAN_NONE 0xfff
+
 struct cpl_l2t_write_rpl {
 	union opcode_tid ot;
 	u8 status;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
index a32de30..c8661c7 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h
@@ -561,6 +561,7 @@ enum fw_flowc_mnem {
 	FW_FLOWC_MNEM_SNDBUF,
 	FW_FLOWC_MNEM_MSS,
 	FW_FLOWC_MNEM_TXDATAPLEN_MAX,
+	FW_FLOWC_MNEM_SCHEDCLASS = 11,
 };
 
 struct fw_flowc_mnemval {
-- 
cgit v0.10.2


From 72bb68721f80a1441e871b6afc9ab0b3793d5031 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Fri, 5 Feb 2016 11:16:21 +0000
Subject: ethtool: add IPv6 to the NFC API

Signed-off-by: Edward Cree <ecree@solarflare.com>
Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index b2e1801..5e0940d 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -748,6 +748,56 @@ struct ethtool_usrip4_spec {
 	__u8    proto;
 };
 
+/**
+ * struct ethtool_tcpip6_spec - flow specification for TCP/IPv6 etc.
+ * @ip6src: Source host
+ * @ip6dst: Destination host
+ * @psrc: Source port
+ * @pdst: Destination port
+ * @tclass: Traffic Class
+ *
+ * This can be used to specify a TCP/IPv6, UDP/IPv6 or SCTP/IPv6 flow.
+ */
+struct ethtool_tcpip6_spec {
+	__be32	ip6src[4];
+	__be32	ip6dst[4];
+	__be16	psrc;
+	__be16	pdst;
+	__u8    tclass;
+};
+
+/**
+ * struct ethtool_ah_espip6_spec - flow specification for IPsec/IPv6
+ * @ip6src: Source host
+ * @ip6dst: Destination host
+ * @spi: Security parameters index
+ * @tclass: Traffic Class
+ *
+ * This can be used to specify an IPsec transport or tunnel over IPv6.
+ */
+struct ethtool_ah_espip6_spec {
+	__be32	ip6src[4];
+	__be32	ip6dst[4];
+	__be32	spi;
+	__u8    tclass;
+};
+
+/**
+ * struct ethtool_usrip6_spec - general flow specification for IPv6
+ * @ip6src: Source host
+ * @ip6dst: Destination host
+ * @l4_4_bytes: First 4 bytes of transport (layer 4) header
+ * @tclass: Traffic Class
+ * @l4_proto: Transport protocol number (nexthdr after any Extension Headers)
+ */
+struct ethtool_usrip6_spec {
+	__be32	ip6src[4];
+	__be32	ip6dst[4];
+	__be32	l4_4_bytes;
+	__u8    tclass;
+	__u8    l4_proto;
+};
+
 union ethtool_flow_union {
 	struct ethtool_tcpip4_spec		tcp_ip4_spec;
 	struct ethtool_tcpip4_spec		udp_ip4_spec;
@@ -755,6 +805,12 @@ union ethtool_flow_union {
 	struct ethtool_ah_espip4_spec		ah_ip4_spec;
 	struct ethtool_ah_espip4_spec		esp_ip4_spec;
 	struct ethtool_usrip4_spec		usr_ip4_spec;
+	struct ethtool_tcpip6_spec		tcp_ip6_spec;
+	struct ethtool_tcpip6_spec		udp_ip6_spec;
+	struct ethtool_tcpip6_spec		sctp_ip6_spec;
+	struct ethtool_ah_espip6_spec		ah_ip6_spec;
+	struct ethtool_ah_espip6_spec		esp_ip6_spec;
+	struct ethtool_usrip6_spec		usr_ip6_spec;
 	struct ethhdr				ether_spec;
 	__u8					hdata[52];
 };
@@ -1401,15 +1457,17 @@ static inline int ethtool_validate_duplex(__u8 duplex)
 #define	UDP_V4_FLOW	0x02	/* hash or spec (udp_ip4_spec) */
 #define	SCTP_V4_FLOW	0x03	/* hash or spec (sctp_ip4_spec) */
 #define	AH_ESP_V4_FLOW	0x04	/* hash only */
-#define	TCP_V6_FLOW	0x05	/* hash only */
-#define	UDP_V6_FLOW	0x06	/* hash only */
-#define	SCTP_V6_FLOW	0x07	/* hash only */
+#define	TCP_V6_FLOW	0x05	/* hash or spec (tcp_ip6_spec; nfc only) */
+#define	UDP_V6_FLOW	0x06	/* hash or spec (udp_ip6_spec; nfc only) */
+#define	SCTP_V6_FLOW	0x07	/* hash or spec (sctp_ip6_spec; nfc only) */
 #define	AH_ESP_V6_FLOW	0x08	/* hash only */
 #define	AH_V4_FLOW	0x09	/* hash or spec (ah_ip4_spec) */
 #define	ESP_V4_FLOW	0x0a	/* hash or spec (esp_ip4_spec) */
-#define	AH_V6_FLOW	0x0b	/* hash only */
-#define	ESP_V6_FLOW	0x0c	/* hash only */
-#define	IP_USER_FLOW	0x0d	/* spec only (usr_ip4_spec) */
+#define	AH_V6_FLOW	0x0b	/* hash or spec (ah_ip6_spec; nfc only) */
+#define	ESP_V6_FLOW	0x0c	/* hash or spec (esp_ip6_spec; nfc only) */
+#define	IPV4_USER_FLOW	0x0d	/* spec only (usr_ip4_spec) */
+#define	IP_USER_FLOW	IPV4_USER_FLOW
+#define	IPV6_USER_FLOW	0x0e	/* spec only (usr_ip6_spec; nfc only) */
 #define	IPV4_FLOW	0x10	/* hash only */
 #define	IPV6_FLOW	0x11	/* hash only */
 #define	ETHER_FLOW	0x12	/* spec only (ether_spec) */
-- 
cgit v0.10.2


From a7ad40d00af18bff49437605f2ec148b80b16768 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Fri, 5 Feb 2016 11:16:50 +0000
Subject: sfc: implement IPv6 NFC (and IPV4_USER_FLOW)

Signed-off-by: Edward Cree <ecree@solarflare.com>
Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 0347976..445ccdb 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -783,14 +783,26 @@ static int efx_ethtool_reset(struct net_device *net_dev, u32 *flags)
 static const u8 mac_addr_ig_mask[ETH_ALEN] __aligned(2) = {0x01, 0, 0, 0, 0, 0};
 
 #define IP4_ADDR_FULL_MASK	((__force __be32)~0)
+#define IP_PROTO_FULL_MASK	0xFF
 #define PORT_FULL_MASK		((__force __be16)~0)
 #define ETHER_TYPE_FULL_MASK	((__force __be16)~0)
 
+static inline void ip6_fill_mask(__be32 *mask)
+{
+	mask[0] = mask[1] = mask[2] = mask[3] = ~(__be32)0;
+}
+
 static int efx_ethtool_get_class_rule(struct efx_nic *efx,
 				      struct ethtool_rx_flow_spec *rule)
 {
 	struct ethtool_tcpip4_spec *ip_entry = &rule->h_u.tcp_ip4_spec;
 	struct ethtool_tcpip4_spec *ip_mask = &rule->m_u.tcp_ip4_spec;
+	struct ethtool_usrip4_spec *uip_entry = &rule->h_u.usr_ip4_spec;
+	struct ethtool_usrip4_spec *uip_mask = &rule->m_u.usr_ip4_spec;
+	struct ethtool_tcpip6_spec *ip6_entry = &rule->h_u.tcp_ip6_spec;
+	struct ethtool_tcpip6_spec *ip6_mask = &rule->m_u.tcp_ip6_spec;
+	struct ethtool_usrip6_spec *uip6_entry = &rule->h_u.usr_ip6_spec;
+	struct ethtool_usrip6_spec *uip6_mask = &rule->m_u.usr_ip6_spec;
 	struct ethhdr *mac_entry = &rule->h_u.ether_spec;
 	struct ethhdr *mac_mask = &rule->m_u.ether_spec;
 	struct efx_filter_spec spec;
@@ -833,6 +845,35 @@ static int efx_ethtool_get_class_rule(struct efx_nic *efx,
 			ip_entry->psrc = spec.rem_port;
 			ip_mask->psrc = PORT_FULL_MASK;
 		}
+	} else if ((spec.match_flags & EFX_FILTER_MATCH_ETHER_TYPE) &&
+	    spec.ether_type == htons(ETH_P_IPV6) &&
+	    (spec.match_flags & EFX_FILTER_MATCH_IP_PROTO) &&
+	    (spec.ip_proto == IPPROTO_TCP || spec.ip_proto == IPPROTO_UDP) &&
+	    !(spec.match_flags &
+	      ~(EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_OUTER_VID |
+		EFX_FILTER_MATCH_LOC_HOST | EFX_FILTER_MATCH_REM_HOST |
+		EFX_FILTER_MATCH_IP_PROTO |
+		EFX_FILTER_MATCH_LOC_PORT | EFX_FILTER_MATCH_REM_PORT))) {
+		rule->flow_type = ((spec.ip_proto == IPPROTO_TCP) ?
+				   TCP_V6_FLOW : UDP_V6_FLOW);
+		if (spec.match_flags & EFX_FILTER_MATCH_LOC_HOST) {
+			memcpy(ip6_entry->ip6dst, spec.loc_host,
+			       sizeof(ip6_entry->ip6dst));
+			ip6_fill_mask(ip6_mask->ip6dst);
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_REM_HOST) {
+			memcpy(ip6_entry->ip6src, spec.rem_host,
+			       sizeof(ip6_entry->ip6src));
+			ip6_fill_mask(ip6_mask->ip6src);
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_LOC_PORT) {
+			ip6_entry->pdst = spec.loc_port;
+			ip6_mask->pdst = PORT_FULL_MASK;
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_REM_PORT) {
+			ip6_entry->psrc = spec.rem_port;
+			ip6_mask->psrc = PORT_FULL_MASK;
+		}
 	} else if (!(spec.match_flags &
 		     ~(EFX_FILTER_MATCH_LOC_MAC | EFX_FILTER_MATCH_LOC_MAC_IG |
 		       EFX_FILTER_MATCH_REM_MAC | EFX_FILTER_MATCH_ETHER_TYPE |
@@ -855,6 +896,47 @@ static int efx_ethtool_get_class_rule(struct efx_nic *efx,
 			mac_entry->h_proto = spec.ether_type;
 			mac_mask->h_proto = ETHER_TYPE_FULL_MASK;
 		}
+	} else if (spec.match_flags & EFX_FILTER_MATCH_ETHER_TYPE &&
+		   spec.ether_type == htons(ETH_P_IP) &&
+		   !(spec.match_flags &
+		     ~(EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_OUTER_VID |
+		       EFX_FILTER_MATCH_LOC_HOST | EFX_FILTER_MATCH_REM_HOST |
+		       EFX_FILTER_MATCH_IP_PROTO))) {
+		rule->flow_type = IPV4_USER_FLOW;
+		uip_entry->ip_ver = ETH_RX_NFC_IP4;
+		if (spec.match_flags & EFX_FILTER_MATCH_IP_PROTO) {
+			uip_mask->proto = IP_PROTO_FULL_MASK;
+			uip_entry->proto = spec.ip_proto;
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_LOC_HOST) {
+			uip_entry->ip4dst = spec.loc_host[0];
+			uip_mask->ip4dst = IP4_ADDR_FULL_MASK;
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_REM_HOST) {
+			uip_entry->ip4src = spec.rem_host[0];
+			uip_mask->ip4src = IP4_ADDR_FULL_MASK;
+		}
+	} else if (spec.match_flags & EFX_FILTER_MATCH_ETHER_TYPE &&
+		   spec.ether_type == htons(ETH_P_IPV6) &&
+		   !(spec.match_flags &
+		     ~(EFX_FILTER_MATCH_ETHER_TYPE | EFX_FILTER_MATCH_OUTER_VID |
+		       EFX_FILTER_MATCH_LOC_HOST | EFX_FILTER_MATCH_REM_HOST |
+		       EFX_FILTER_MATCH_IP_PROTO))) {
+		rule->flow_type = IPV6_USER_FLOW;
+		if (spec.match_flags & EFX_FILTER_MATCH_IP_PROTO) {
+			uip6_mask->l4_proto = IP_PROTO_FULL_MASK;
+			uip6_entry->l4_proto = spec.ip_proto;
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_LOC_HOST) {
+			memcpy(uip6_entry->ip6dst, spec.loc_host,
+			       sizeof(uip6_entry->ip6dst));
+			ip6_fill_mask(uip6_mask->ip6dst);
+		}
+		if (spec.match_flags & EFX_FILTER_MATCH_REM_HOST) {
+			memcpy(uip6_entry->ip6src, spec.rem_host,
+			       sizeof(uip6_entry->ip6src));
+			ip6_fill_mask(uip6_mask->ip6src);
+		}
 	} else {
 		/* The above should handle all filters that we insert */
 		WARN_ON(1);
@@ -946,11 +1028,27 @@ efx_ethtool_get_rxnfc(struct net_device *net_dev,
 	}
 }
 
+static inline bool ip6_mask_is_full(__be32 mask[4])
+{
+	return !~(mask[0] & mask[1] & mask[2] & mask[3]);
+}
+
+static inline bool ip6_mask_is_empty(__be32 mask[4])
+{
+	return !(mask[0] | mask[1] | mask[2] | mask[3]);
+}
+
 static int efx_ethtool_set_class_rule(struct efx_nic *efx,
 				      struct ethtool_rx_flow_spec *rule)
 {
 	struct ethtool_tcpip4_spec *ip_entry = &rule->h_u.tcp_ip4_spec;
 	struct ethtool_tcpip4_spec *ip_mask = &rule->m_u.tcp_ip4_spec;
+	struct ethtool_usrip4_spec *uip_entry = &rule->h_u.usr_ip4_spec;
+	struct ethtool_usrip4_spec *uip_mask = &rule->m_u.usr_ip4_spec;
+	struct ethtool_tcpip6_spec *ip6_entry = &rule->h_u.tcp_ip6_spec;
+	struct ethtool_tcpip6_spec *ip6_mask = &rule->m_u.tcp_ip6_spec;
+	struct ethtool_usrip6_spec *uip6_entry = &rule->h_u.usr_ip6_spec;
+	struct ethtool_usrip6_spec *uip6_mask = &rule->m_u.usr_ip6_spec;
 	struct ethhdr *mac_entry = &rule->h_u.ether_spec;
 	struct ethhdr *mac_mask = &rule->m_u.ether_spec;
 	struct efx_filter_spec spec;
@@ -1012,6 +1110,92 @@ static int efx_ethtool_set_class_rule(struct efx_nic *efx,
 			return -EINVAL;
 		break;
 
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+		spec.match_flags = (EFX_FILTER_MATCH_ETHER_TYPE |
+				    EFX_FILTER_MATCH_IP_PROTO);
+		spec.ether_type = htons(ETH_P_IPV6);
+		spec.ip_proto = ((rule->flow_type & ~FLOW_EXT) == TCP_V6_FLOW ?
+				 IPPROTO_TCP : IPPROTO_UDP);
+		if (!ip6_mask_is_empty(ip6_mask->ip6dst)) {
+			if (!ip6_mask_is_full(ip6_mask->ip6dst))
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_LOC_HOST;
+			memcpy(spec.loc_host, ip6_entry->ip6dst, sizeof(spec.loc_host));
+		}
+		if (!ip6_mask_is_empty(ip6_mask->ip6src)) {
+			if (!ip6_mask_is_full(ip6_mask->ip6src))
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_REM_HOST;
+			memcpy(spec.rem_host, ip6_entry->ip6src, sizeof(spec.rem_host));
+		}
+		if (ip6_mask->pdst) {
+			if (ip6_mask->pdst != PORT_FULL_MASK)
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_LOC_PORT;
+			spec.loc_port = ip6_entry->pdst;
+		}
+		if (ip6_mask->psrc) {
+			if (ip6_mask->psrc != PORT_FULL_MASK)
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_REM_PORT;
+			spec.rem_port = ip6_entry->psrc;
+		}
+		if (ip6_mask->tclass)
+			return -EINVAL;
+		break;
+
+	case IPV4_USER_FLOW:
+		if (uip_mask->l4_4_bytes || uip_mask->tos || uip_mask->ip_ver ||
+		    uip_entry->ip_ver != ETH_RX_NFC_IP4)
+			return -EINVAL;
+		spec.match_flags = EFX_FILTER_MATCH_ETHER_TYPE;
+		spec.ether_type = htons(ETH_P_IP);
+		if (uip_mask->ip4dst) {
+			if (uip_mask->ip4dst != IP4_ADDR_FULL_MASK)
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_LOC_HOST;
+			spec.loc_host[0] = uip_entry->ip4dst;
+		}
+		if (uip_mask->ip4src) {
+			if (uip_mask->ip4src != IP4_ADDR_FULL_MASK)
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_REM_HOST;
+			spec.rem_host[0] = uip_entry->ip4src;
+		}
+		if (uip_mask->proto) {
+			if (uip_mask->proto != IP_PROTO_FULL_MASK)
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_IP_PROTO;
+			spec.ip_proto = uip_entry->proto;
+		}
+		break;
+
+	case IPV6_USER_FLOW:
+		if (uip6_mask->l4_4_bytes || uip6_mask->tclass)
+			return -EINVAL;
+		spec.match_flags = EFX_FILTER_MATCH_ETHER_TYPE;
+		spec.ether_type = htons(ETH_P_IPV6);
+		if (!ip6_mask_is_empty(uip6_mask->ip6dst)) {
+			if (!ip6_mask_is_full(uip6_mask->ip6dst))
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_LOC_HOST;
+			memcpy(spec.loc_host, uip6_entry->ip6dst, sizeof(spec.loc_host));
+		}
+		if (!ip6_mask_is_empty(uip6_mask->ip6src)) {
+			if (!ip6_mask_is_full(uip6_mask->ip6src))
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_REM_HOST;
+			memcpy(spec.rem_host, uip6_entry->ip6src, sizeof(spec.rem_host));
+		}
+		if (uip6_mask->l4_proto) {
+			if (uip6_mask->l4_proto != IP_PROTO_FULL_MASK)
+				return -EINVAL;
+			spec.match_flags |= EFX_FILTER_MATCH_IP_PROTO;
+			spec.ip_proto = uip6_entry->l4_proto;
+		}
+		break;
+
 	case ETHER_FLOW:
 		if (!is_zero_ether_addr(mac_mask->h_dest)) {
 			if (ether_addr_equal(mac_mask->h_dest,
-- 
cgit v0.10.2


From a060679c6b3da17dc9e95d0500f811de118ec901 Mon Sep 17 00:00:00 2001
From: "sixiao@microsoft.com" <sixiao@microsoft.com>
Date: Thu, 4 Feb 2016 15:49:34 -0800
Subject: hv_netvsc: cleanup netdev feature flags for netvsc

1. Adding NETIF_F_TSO6 feature flag;
2. Adding NETIF_F_HW_CSUM. NETIF_F_IPV6_CSUM and NETIF_F_IP_CSUM are
being deprecated;
3. Cleanup the coding style of flag assignment by using macro.

Signed-off-by: Simon Xiao <sixiao@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 1d3a665..c72e5b8 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -43,6 +43,11 @@
 
 #define RING_SIZE_MIN 64
 #define LINKCHANGE_INT (2 * HZ)
+#define NETVSC_HW_FEATURES	(NETIF_F_RXCSUM | \
+				 NETIF_F_SG | \
+				 NETIF_F_TSO | \
+				 NETIF_F_TSO6 | \
+				 NETIF_F_HW_CSUM)
 static int ring_size = 128;
 module_param(ring_size, int, S_IRUGO);
 MODULE_PARM_DESC(ring_size, "Ring buffer size (# of pages)");
@@ -1081,10 +1086,8 @@ static int netvsc_probe(struct hv_device *dev,
 
 	net->netdev_ops = &device_ops;
 
-	net->hw_features = NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_IP_CSUM |
-				NETIF_F_TSO;
-	net->features = NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_SG | NETIF_F_RXCSUM |
-			NETIF_F_IP_CSUM | NETIF_F_TSO;
+	net->hw_features = NETVSC_HW_FEATURES;
+	net->features = NETVSC_HW_FEATURES | NETIF_F_HW_VLAN_CTAG_TX;
 
 	net->ethtool_ops = &ethtool_ops;
 	SET_NETDEV_DEV(net, &dev->device);
-- 
cgit v0.10.2


From bef3c6c9374da40ec63698cd504474366a9de2ff Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:27:31 -0800
Subject: net: Drop unecessary enc_features variable from tunnel segmentation
 functions

The enc_features variable isn't necessary since features isn't used
anywhere after we create enc_features so instead just use a destructive AND
on features itself and save ourselves the variable declaration.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 5a8ee32..02cb1a4 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -19,7 +19,6 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 				       netdev_features_t features)
 {
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
-	netdev_features_t enc_features;
 	int ghl;
 	struct gre_base_hdr *greh;
 	u16 mac_offset = skb->mac_header;
@@ -68,9 +67,10 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	skb_set_network_header(skb, skb_inner_network_offset(skb));
 	skb->mac_len = skb_inner_network_offset(skb);
 
+	features &= skb->dev->hw_enc_features;
+
 	/* segment inner packet. */
-	enc_features = skb->dev->hw_enc_features & features;
-	segs = skb_mac_gso_segment(skb, enc_features);
+	segs = skb_mac_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
 		skb_gso_error_unwind(skb, protocol, ghl, mac_offset, mac_len);
 		goto out;
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 4c519c1..ce64c2b 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -37,7 +37,6 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	int mac_len = skb->mac_len;
 	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
 	__be16 protocol = skb->protocol;
-	netdev_features_t enc_features;
 	int udp_offset, outer_hlen;
 	unsigned int oldlen;
 	bool need_csum = !!(skb_shinfo(skb)->gso_type &
@@ -65,9 +64,10 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 			   (skb->dev->features & (is_ipv6 ?
 			    NETIF_F_IPV6_CSUM : NETIF_F_IP_CSUM))));
 
+	features &= skb->dev->hw_enc_features;
+
 	/* segment inner packet. */
-	enc_features = skb->dev->hw_enc_features & features;
-	segs = gso_inner_segment(skb, enc_features);
+	segs = gso_inner_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
 		skb_gso_error_unwind(skb, protocol, tnl_hlen, mac_offset,
 				     mac_len);
-- 
cgit v0.10.2


From 76443456227097179c14826425f88a95d81a892e Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:27:37 -0800
Subject: net: Move GSO csum into SKB_GSO_CB

This patch moves the checksum maintained by GSO out of skb->csum and into
the GSO context block in order to allow for us to work on outer checksums
while maintaining the inner checksum offsets in the case of the inner
checksum being offloaded, while the outer checksums will be computed.

While updating the code I also did a minor cleanu-up on gso_make_checksum.
The change is mostly to make it so that we store the values and compute the
checksum instead of computing the checksum and then storing the values we
needed to update.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 11f935c..acece7c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3549,6 +3549,7 @@ static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
 struct skb_gso_cb {
 	int	mac_offset;
 	int	encap_level;
+	__wsum	csum;
 	__u16	csum_start;
 };
 #define SKB_SGO_CB_OFFSET	32
@@ -3585,15 +3586,14 @@ static inline int gso_pskb_expand_head(struct sk_buff *skb, int extra)
  */
 static inline __sum16 gso_make_checksum(struct sk_buff *skb, __wsum res)
 {
-	int plen = SKB_GSO_CB(skb)->csum_start - skb_headroom(skb) -
-		   skb_transport_offset(skb);
-	__wsum partial;
+	unsigned char *csum_start = skb_transport_header(skb);
+	int plen = (skb->head + SKB_GSO_CB(skb)->csum_start) - csum_start;
+	__wsum partial = SKB_GSO_CB(skb)->csum;
 
-	partial = csum_partial(skb_transport_header(skb), plen, skb->csum);
-	skb->csum = res;
-	SKB_GSO_CB(skb)->csum_start -= plen;
+	SKB_GSO_CB(skb)->csum = res;
+	SKB_GSO_CB(skb)->csum_start = csum_start - skb->head;
 
-	return csum_fold(partial);
+	return csum_fold(csum_partial(csum_start, plen, partial));
 }
 
 static inline bool skb_is_gso(const struct sk_buff *skb)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index b2df375..02c638a6 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3100,11 +3100,12 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 
 		if (!sg && !nskb->remcsum_offload) {
 			nskb->ip_summed = CHECKSUM_NONE;
-			nskb->csum = skb_copy_and_csum_bits(head_skb, offset,
-							    skb_put(nskb, len),
-							    len, 0);
+			SKB_GSO_CB(nskb)->csum =
+				skb_copy_and_csum_bits(head_skb, offset,
+						       skb_put(nskb, len),
+						       len, 0);
 			SKB_GSO_CB(nskb)->csum_start =
-			    skb_headroom(nskb) + doffset;
+				skb_headroom(nskb) + doffset;
 			continue;
 		}
 
@@ -3171,11 +3172,12 @@ skip_fraglist:
 
 perform_csum_check:
 		if (!csum && !nskb->remcsum_offload) {
-			nskb->csum = skb_checksum(nskb, doffset,
-						  nskb->len - doffset, 0);
 			nskb->ip_summed = CHECKSUM_NONE;
+			SKB_GSO_CB(nskb)->csum =
+				skb_checksum(nskb, doffset,
+					     nskb->len - doffset, 0);
 			SKB_GSO_CB(nskb)->csum_start =
-			    skb_headroom(nskb) + doffset;
+				skb_headroom(nskb) + doffset;
 		}
 	} while ((offset += len) < head_skb->len);
 
-- 
cgit v0.10.2


From 7fbeffed77c130ecf64e8a2f7f9d6d63a9d60a19 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:27:43 -0800
Subject: net: Update remote checksum segmentation to support use of GSO
 checksum

This patch addresses two main issues.

First in the case of remote checksum offload we were avoiding dealing with
scatter-gather issues.  As a result it would be possible to assemble a
series of frames that used frags instead of being linearized as they should
have if remote checksum offload was enabled.

Second I have updated the code so that we now let GSO take care of doing
the checksum on the data itself and drop the special case that was added
for remote checksum offload.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 02c638a6..9c065ac 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3098,8 +3098,9 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 		if (nskb->len == len + doffset)
 			goto perform_csum_check;
 
-		if (!sg && !nskb->remcsum_offload) {
-			nskb->ip_summed = CHECKSUM_NONE;
+		if (!sg) {
+			if (!nskb->remcsum_offload)
+				nskb->ip_summed = CHECKSUM_NONE;
 			SKB_GSO_CB(nskb)->csum =
 				skb_copy_and_csum_bits(head_skb, offset,
 						       skb_put(nskb, len),
@@ -3171,8 +3172,9 @@ skip_fraglist:
 		nskb->truesize += nskb->data_len;
 
 perform_csum_check:
-		if (!csum && !nskb->remcsum_offload) {
-			nskb->ip_summed = CHECKSUM_NONE;
+		if (!csum) {
+			if (!nskb->remcsum_offload)
+				nskb->ip_summed = CHECKSUM_NONE;
 			SKB_GSO_CB(nskb)->csum =
 				skb_checksum(nskb, doffset,
 					     nskb->len - doffset, 0);
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index ce64c2b..86687f5 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -66,6 +66,16 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 
 	features &= skb->dev->hw_enc_features;
 
+	/* The only checksum offload we care about from here on out is the
+	 * outer one so strip the existing checksum feature flags and
+	 * instead set the flag based on our outer checksum offload value.
+	 */
+	if (remcsum) {
+		features &= ~NETIF_F_CSUM_MASK;
+		if (offload_csum)
+			features |= NETIF_F_HW_CSUM;
+	}
+
 	/* segment inner packet. */
 	segs = gso_inner_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
@@ -116,18 +126,6 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 			skb->ip_summed = CHECKSUM_PARTIAL;
 			skb->csum_start = skb_transport_header(skb) - skb->head;
 			skb->csum_offset = offsetof(struct udphdr, check);
-		} else if (remcsum) {
-			/* Need to calculate checksum from scratch,
-			 * inner checksums are never when doing
-			 * remote_checksum_offload.
-			 */
-
-			skb->csum = skb_checksum(skb, udp_offset,
-						 skb->len - udp_offset,
-						 0);
-			uh->check = csum_fold(skb->csum);
-			if (uh->check == 0)
-				uh->check = CSUM_MANGLED_0;
 		} else {
 			uh->check = gso_make_checksum(skb, ~uh->check);
 
-- 
cgit v0.10.2


From 08b64fcca942733413bc5ac2321d57021d3e8578 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:27:49 -0800
Subject: net: Store checksum result for offloaded GSO checksums

This patch makes it so that we can offload the checksums for a packet up
to a certain point and then begin computing the checksums via software.
Setting this up is fairly straight forward as all we need to do is reset
the values stored in csum and csum_start for the GSO context block.

One complication for this is remote checksum offload.  In order to allow
the inner checksums to be offloaded while computing the outer checksum
manually we needed to have some way of indicating that the offload wasn't
real.  In order to do that I replaced CHECKSUM_PARTIAL with
CHECKSUM_UNNECESSARY in the case of us computing checksums for the outer
header while skipping computing checksums for the inner headers.  We clean
up the ip_summed flag and set it to either CHECKSUM_PARTIAL or
CHECKSUM_NONE once we hand the packet off to the next lower level.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index acece7c..a8fc222 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2161,6 +2161,11 @@ static inline int skb_checksum_start_offset(const struct sk_buff *skb)
 	return skb->csum_start - skb_headroom(skb);
 }
 
+static inline unsigned char *skb_checksum_start(const struct sk_buff *skb)
+{
+	return skb->head + skb->csum_start;
+}
+
 static inline int skb_transport_offset(const struct sk_buff *skb)
 {
 	return skb_transport_header(skb) - skb->data;
@@ -3576,6 +3581,16 @@ static inline int gso_pskb_expand_head(struct sk_buff *skb, int extra)
 	return 0;
 }
 
+static inline void gso_reset_checksum(struct sk_buff *skb, __wsum res)
+{
+	/* Do not update partial checksums if remote checksum is enabled. */
+	if (skb->remcsum_offload)
+		return;
+
+	SKB_GSO_CB(skb)->csum = res;
+	SKB_GSO_CB(skb)->csum_start = skb_checksum_start(skb) - skb->head;
+}
+
 /* Compute the checksum for a gso segment. First compute the checksum value
  * from the start of transport header to SKB_GSO_CB(skb)->csum_start, and
  * then add in skb->csum (checksum from csum_start to end of packet).
diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
index 9864a2d..773083b 100644
--- a/net/ipv4/tcp_offload.c
+++ b/net/ipv4/tcp_offload.c
@@ -135,7 +135,9 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
 		th->fin = th->psh = 0;
 		th->check = newcheck;
 
-		if (skb->ip_summed != CHECKSUM_PARTIAL)
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
+			gso_reset_checksum(skb, ~th->check);
+		else
 			th->check = gso_make_checksum(skb, ~th->check);
 
 		seq += mss;
@@ -169,7 +171,9 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
 		      skb->data_len);
 	th->check = ~csum_fold((__force __wsum)((__force u32)th->check +
 				(__force u32)delta));
-	if (skb->ip_summed != CHECKSUM_PARTIAL)
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		gso_reset_checksum(skb, ~th->check);
+	else
 		th->check = gso_make_checksum(skb, ~th->check);
 out:
 	return segs;
-- 
cgit v0.10.2


From ddff00d420432d54eb420bb33034bb8e22dd2543 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:27:55 -0800
Subject: net: Move skb_has_shared_frag check out of GRE code and into
 segmentation

The call skb_has_shared_frag is used in the GRE path and skb_checksum_help
to verify that no frags can be modified by an external entity.  This check
really doesn't belong in the GRE path but in the skb_segment function
itself.  This way any protocol that might be segmented will be performing
this check before attempting to offload a checksum to software.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 9c065ac..88262c8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3173,6 +3173,11 @@ skip_fraglist:
 
 perform_csum_check:
 		if (!csum) {
+			if (skb_has_shared_frag(nskb)) {
+				err = __skb_linearize(nskb);
+				if (err)
+					goto err;
+			}
 			if (!nskb->remcsum_offload)
 				nskb->ip_summed = CHECKSUM_NONE;
 			SKB_GSO_CB(nskb)->csum =
diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 02cb1a4..35a8dd35 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -83,17 +83,6 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 		if (csum) {
 			__be32 *pcsum;
 
-			if (skb_has_shared_frag(skb)) {
-				int err;
-
-				err = __skb_linearize(skb);
-				if (err) {
-					kfree_skb_list(segs);
-					segs = ERR_PTR(err);
-					goto out;
-				}
-			}
-
 			skb_reset_transport_header(skb);
 
 			greh = (struct gre_base_hdr *)
-- 
cgit v0.10.2


From 2e598af7135d1434b0ebf2e9c7ff8e014f48d572 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:28:01 -0800
Subject: gre: Use GSO flags to determine csum need instead of GRE flags

This patch updates the gre checksum path to follow something much closer to
the UDP checksum path.  By doing this we can avoid needing to do as much
header inspection and can just make use of the fields we were already
reading in the sk_buff structure.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 35a8dd35..c15441b 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -18,14 +18,14 @@
 static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 				       netdev_features_t features)
 {
+	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
-	int ghl;
 	struct gre_base_hdr *greh;
 	u16 mac_offset = skb->mac_header;
-	int mac_len = skb->mac_len;
 	__be16 protocol = skb->protocol;
-	int tnl_hlen;
-	bool csum;
+	u16 mac_len = skb->mac_len;
+	int gre_offset, outer_hlen;
+	bool need_csum;
 
 	if (unlikely(skb_shinfo(skb)->gso_type &
 				~(SKB_GSO_TCPV4 |
@@ -42,64 +42,60 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	if (!skb->encapsulation)
 		goto out;
 
-	if (unlikely(!pskb_may_pull(skb, sizeof(*greh))))
+	if (unlikely(tnl_hlen < sizeof(struct gre_base_hdr)))
 		goto out;
 
-	greh = (struct gre_base_hdr *)skb_transport_header(skb);
-
-	ghl = skb_inner_mac_header(skb) - skb_transport_header(skb);
-	if (unlikely(ghl < sizeof(*greh)))
+	if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
 		goto out;
 
-	csum = !!(greh->flags & GRE_CSUM);
-	if (csum)
-		skb->encap_hdr_csum = 1;
+	greh = (struct gre_base_hdr *)skb_transport_header(skb);
 
 	/* setup inner skb. */
 	skb->protocol = greh->protocol;
 	skb->encapsulation = 0;
-
-	if (unlikely(!pskb_may_pull(skb, ghl)))
-		goto out;
-
-	__skb_pull(skb, ghl);
+	__skb_pull(skb, tnl_hlen);
 	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, skb_inner_network_offset(skb));
 	skb->mac_len = skb_inner_network_offset(skb);
 
+	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE_CSUM);
+	skb->encap_hdr_csum = need_csum;
+
 	features &= skb->dev->hw_enc_features;
 
 	/* segment inner packet. */
 	segs = skb_mac_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
-		skb_gso_error_unwind(skb, protocol, ghl, mac_offset, mac_len);
+		skb_gso_error_unwind(skb, protocol, tnl_hlen, mac_offset,
+				     mac_len);
 		goto out;
 	}
 
+	outer_hlen = skb_tnl_header_len(skb);
+	gre_offset = outer_hlen - tnl_hlen;
 	skb = segs;
-	tnl_hlen = skb_tnl_header_len(skb);
 	do {
-		__skb_push(skb, ghl);
-		if (csum) {
-			__be32 *pcsum;
-
-			skb_reset_transport_header(skb);
-
-			greh = (struct gre_base_hdr *)
-			    skb_transport_header(skb);
-			pcsum = (__be32 *)(greh + 1);
-			*pcsum = 0;
-			*(__sum16 *)pcsum = gso_make_checksum(skb, 0);
-		}
-		__skb_push(skb, tnl_hlen - ghl);
+		__be32 *pcsum;
 
 		skb_reset_inner_headers(skb);
 		skb->encapsulation = 1;
 
-		skb_reset_mac_header(skb);
-		skb_set_network_header(skb, mac_len);
 		skb->mac_len = mac_len;
 		skb->protocol = protocol;
+
+		__skb_push(skb, outer_hlen);
+		skb_reset_mac_header(skb);
+		skb_set_network_header(skb, mac_len);
+		skb_set_transport_header(skb, gre_offset);
+
+		if (!need_csum)
+			continue;
+
+		greh = (struct gre_base_hdr *)skb_transport_header(skb);
+		pcsum = (__be32 *)(greh + 1);
+
+		*pcsum = 0;
+		*(__sum16 *)pcsum = gso_make_checksum(skb, 0);
 	} while ((skb = skb->next));
 out:
 	return segs;
-- 
cgit v0.10.2


From 38720352412a4305a108ff37ae6099dab3f44a38 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:28:08 -0800
Subject: gre: Use inner_proto to obtain inner header protocol

Instead of parsing headers to determine the inner protocol we can just pull
the value from inner_proto.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index c15441b..003b0eb 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -20,7 +20,6 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 {
 	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
-	struct gre_base_hdr *greh;
 	u16 mac_offset = skb->mac_header;
 	__be16 protocol = skb->protocol;
 	u16 mac_len = skb->mac_len;
@@ -48,15 +47,13 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
 		goto out;
 
-	greh = (struct gre_base_hdr *)skb_transport_header(skb);
-
 	/* setup inner skb. */
-	skb->protocol = greh->protocol;
 	skb->encapsulation = 0;
 	__skb_pull(skb, tnl_hlen);
 	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, skb_inner_network_offset(skb));
 	skb->mac_len = skb_inner_network_offset(skb);
+	skb->protocol = skb->inner_protocol;
 
 	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE_CSUM);
 	skb->encap_hdr_csum = need_csum;
@@ -75,6 +72,7 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	gre_offset = outer_hlen - tnl_hlen;
 	skb = segs;
 	do {
+		struct gre_base_hdr *greh;
 		__be32 *pcsum;
 
 		skb_reset_inner_headers(skb);
-- 
cgit v0.10.2


From fdaefd62fd658b266a8f389cdf7991630b4bc7b4 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:28:14 -0800
Subject: udp: Clean up the use of flags in UDP segmentation offload

This patch goes though and cleans up the logic related to several of the
control flags used in UDP segmentation.  Specifically the use of dont_encap
isn't really needed as we can just check the skb for CHECKSUM_PARTIAL and
if it isn't set then we don't need to update the internal headers.  As such
we can just drop that value.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 86687f5..9e4816f 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -33,16 +33,13 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	__be16 new_protocol, bool is_ipv6)
 {
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
+	bool remcsum, need_csum, offload_csum;
 	u16 mac_offset = skb->mac_header;
 	int mac_len = skb->mac_len;
 	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
 	__be16 protocol = skb->protocol;
 	int udp_offset, outer_hlen;
 	unsigned int oldlen;
-	bool need_csum = !!(skb_shinfo(skb)->gso_type &
-			    SKB_GSO_UDP_TUNNEL_CSUM);
-	bool remcsum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_TUNNEL_REMCSUM);
-	bool offload_csum = false, dont_encap = (need_csum || remcsum);
 
 	oldlen = (u16)~skb->len;
 
@@ -55,14 +52,18 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	skb_set_network_header(skb, skb_inner_network_offset(skb));
 	skb->mac_len = skb_inner_network_offset(skb);
 	skb->protocol = new_protocol;
+
+	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM);
 	skb->encap_hdr_csum = need_csum;
+
+	remcsum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_TUNNEL_REMCSUM);
 	skb->remcsum_offload = remcsum;
 
 	/* Try to offload checksum if possible */
 	offload_csum = !!(need_csum &&
-			  ((skb->dev->features & NETIF_F_HW_CSUM) ||
-			   (skb->dev->features & (is_ipv6 ?
-			    NETIF_F_IPV6_CSUM : NETIF_F_IP_CSUM))));
+			  (skb->dev->features &
+			   (is_ipv6 ? (NETIF_F_HW_CSUM | NETIF_F_IPV6_CSUM) :
+				      (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM))));
 
 	features &= skb->dev->hw_enc_features;
 
@@ -92,13 +93,11 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 		int len;
 		__be32 delta;
 
-		if (dont_encap) {
-			skb->encapsulation = 0;
+		if (remcsum)
 			skb->ip_summed = CHECKSUM_NONE;
-		} else {
-			/* Only set up inner headers if we might be offloading
-			 * inner checksum.
-			 */
+
+		/* Set up inner headers if we are offloading inner checksum */
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
 			skb_reset_inner_headers(skb);
 			skb->encapsulation = 1;
 		}
@@ -122,15 +121,15 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 		uh->check = ~csum_fold((__force __wsum)
 				       ((__force u32)uh->check +
 					(__force u32)delta));
-		if (offload_csum) {
-			skb->ip_summed = CHECKSUM_PARTIAL;
-			skb->csum_start = skb_transport_header(skb) - skb->head;
-			skb->csum_offset = offsetof(struct udphdr, check);
-		} else {
-			uh->check = gso_make_checksum(skb, ~uh->check);
 
+		if (skb->encapsulation || !offload_csum) {
+			uh->check = gso_make_checksum(skb, ~uh->check);
 			if (uh->check == 0)
 				uh->check = CSUM_MANGLED_0;
+		} else {
+			skb->ip_summed = CHECKSUM_PARTIAL;
+			skb->csum_start = skb_transport_header(skb) - skb->head;
+			skb->csum_offset = offsetof(struct udphdr, check);
 		}
 	} while ((skb = skb->next));
 out:
-- 
cgit v0.10.2


From dbef491ebe7f3a4fb1b9111878b86a426fd540b7 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:28:20 -0800
Subject: udp: Use uh->len instead of skb->len to compute checksum in
 segmentation

The segmentation code was having to do a bunch of work to pull the
skb->len and strip the udp header offset before the value could be used to
adjust the checksum.  Instead of doing all this work we can just use the
value that goes into uh->len since that is the correct value with the
correct byte order that we need anyway.  By using this value we can save
ourselves a bunch of pain as there is no need to do multiple byte swaps.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 9e4816f..56c4c8b 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -32,20 +32,23 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 					     netdev_features_t features),
 	__be16 new_protocol, bool is_ipv6)
 {
+	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
 	bool remcsum, need_csum, offload_csum;
+	struct udphdr *uh = udp_hdr(skb);
 	u16 mac_offset = skb->mac_header;
-	int mac_len = skb->mac_len;
-	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
 	__be16 protocol = skb->protocol;
+	u16 mac_len = skb->mac_len;
 	int udp_offset, outer_hlen;
-	unsigned int oldlen;
-
-	oldlen = (u16)~skb->len;
+	u32 partial;
 
 	if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
 		goto out;
 
+	/* adjust partial header checksum to negate old length */
+	partial = (__force u32)uh->check + (__force u16)~uh->len;
+
+	/* setup inner skb. */
 	skb->encapsulation = 0;
 	__skb_pull(skb, tnl_hlen);
 	skb_reset_mac_header(skb);
@@ -89,9 +92,7 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	udp_offset = outer_hlen - tnl_hlen;
 	skb = segs;
 	do {
-		struct udphdr *uh;
-		int len;
-		__be32 delta;
+		__be16 len;
 
 		if (remcsum)
 			skb->ip_summed = CHECKSUM_NONE;
@@ -105,22 +106,19 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 		skb->mac_len = mac_len;
 		skb->protocol = protocol;
 
-		skb_push(skb, outer_hlen);
+		__skb_push(skb, outer_hlen);
 		skb_reset_mac_header(skb);
 		skb_set_network_header(skb, mac_len);
 		skb_set_transport_header(skb, udp_offset);
-		len = skb->len - udp_offset;
+		len = htons(skb->len - udp_offset);
 		uh = udp_hdr(skb);
-		uh->len = htons(len);
+		uh->len = len;
 
 		if (!need_csum)
 			continue;
 
-		delta = htonl(oldlen + len);
-
 		uh->check = ~csum_fold((__force __wsum)
-				       ((__force u32)uh->check +
-					(__force u32)delta));
+				       ((__force u32)len + partial));
 
 		if (skb->encapsulation || !offload_csum) {
 			uh->check = gso_make_checksum(skb, ~uh->check);
-- 
cgit v0.10.2


From f245d079c1d11dc6927e56f5a89dd566fef2a415 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 5 Feb 2016 15:28:26 -0800
Subject: net: Allow tunnels to use inner checksum offloads with outer
 checksums needed

This patch enables us to use inner checksum offloads if provided by
hardware with outer checksums computed by software.

It basically reduces encap_hdr_csum to an advisory flag for now, but based
on the fact that SCTP may be getting segmentation support before long I
thought we may want to keep it as it is possible we may need to support
CRC32c and 1's compliment checksum in the same packet at some point in the
future.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 88262c8..b0cce74 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3004,8 +3004,7 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 	if (unlikely(!proto))
 		return ERR_PTR(-EINVAL);
 
-	csum = !head_skb->encap_hdr_csum &&
-	    !!can_checksum_protocol(features, proto);
+	csum = !!can_checksum_protocol(features, proto);
 
 	headroom = skb_headroom(head_skb);
 	pos = skb_headlen(head_skb);
-- 
cgit v0.10.2


From 6d9b6f424d0806e2a7f3e7f0a74170ec3ad30443 Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Sun, 7 Feb 2016 10:56:25 +0530
Subject: bonding: Return correct error code

The return value of kzalloc on failure of allocation of memory should
be -ENOMEM and not -1.

Found using Coccinelle. A simplified version of the semantic patch
used is:

//<smpl>
@@
expression *e;
@@

e = kzalloc(...);
if (e == NULL) {
...
return
- -1
+ -ENOMEM
;
}
//</smpl>

The single call site only checks that the return value is not 0,
hence no change is required at the call site.

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index bb9e9fc..c5ac160 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -159,7 +159,7 @@ static int tlb_initialize(struct bonding *bond)
 
 	new_hashtbl = kzalloc(size, GFP_KERNEL);
 	if (!new_hashtbl)
-		return -1;
+		return -ENOMEM;
 
 	spin_lock_bh(&bond->mode_lock);
 
-- 
cgit v0.10.2


From f48e72318aa6f6ac792eaf74f3b77491a5c43891 Mon Sep 17 00:00:00 2001
From: stephen hemminger <stephen@networkplumber.org>
Date: Tue, 9 Feb 2016 22:04:47 -0800
Subject: rds: duplicate include net/tcp.h

Duplicate include detected.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index 9d6ddba..ad60299 100644
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -37,7 +37,6 @@
 #include <net/tcp.h>
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
-#include <net/tcp.h>
 
 #include "rds.h"
 #include "tcp.h"
-- 
cgit v0.10.2


From 40d29af05703555b03c689e5cecc34bdea28fa89 Mon Sep 17 00:00:00 2001
From: stephen hemminger <stephen@networkplumber.org>
Date: Tue, 9 Feb 2016 22:07:29 -0800
Subject: vxlan: udp_tunnel duplicate include net/udp_tunnel.h

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 65f5247..ebf57d9 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -42,7 +42,7 @@
 #include <net/netns/generic.h>
 #include <net/vxlan.h>
 #include <net/protocol.h>
-#include <net/udp_tunnel.h>
+
 #if IS_ENABLED(CONFIG_IPV6)
 #include <net/ipv6.h>
 #include <net/addrconf.h>
-- 
cgit v0.10.2


From 809dc75e9b82fcae9c3c48b4eec4708c005259f3 Mon Sep 17 00:00:00 2001
From: stephen hemminger <shemming@brocade.com>
Date: Tue, 9 Feb 2016 22:11:27 -0800
Subject: vrf: duplicate include of rtnetlink.h

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 76e1fc9..9ce088b 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -32,7 +32,6 @@
 #include <net/ip_fib.h>
 #include <net/ip6_fib.h>
 #include <net/ip6_route.h>
-#include <net/rtnetlink.h>
 #include <net/route.h>
 #include <net/addrconf.h>
 #include <net/l3mdev.h>
-- 
cgit v0.10.2


From 4456ed04ea44b800d691b18c14a68ec9894d2aca Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Sun, 7 Feb 2016 23:27:55 +0200
Subject: ethtool: future-proof interface for speed extensions

Many virtual and not quite virtual devices allow any speed to be set
through ethtool. In particular, this applies to the virtio-net devices.
Document this fact to make sure people don't assume the enum lists all
possible values.  Reserve values greater than INT_MAX for future
extension and to avoid conflict with SPEED_UNKNOWN.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 5e0940d..4345f80 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -31,7 +31,7 @@
  *	physical connectors and other link features that are
  *	advertised through autonegotiation or enabled for
  *	auto-detection.
- * @speed: Low bits of the speed
+ * @speed: Low bits of the speed, 1Mb units, 0 to INT_MAX or SPEED_UNKNOWN
  * @duplex: Duplex mode; one of %DUPLEX_*
  * @port: Physical connector type; one of %PORT_*
  * @phy_address: MDIO address of PHY (transceiver); 0 or 255 if not
@@ -47,7 +47,7 @@
  *	obsoleted by &struct ethtool_coalesce.  Read-only; deprecated.
  * @maxrxpkt: Historically used to report RX IRQ coalescing; now
  *	obsoleted by &struct ethtool_coalesce.  Read-only; deprecated.
- * @speed_hi: High bits of the speed
+ * @speed_hi: High bits of the speed, 1Mb units, 0 to INT_MAX or SPEED_UNKNOWN
  * @eth_tp_mdix: Ethernet twisted-pair MDI(-X) status; one of
  *	%ETH_TP_MDI_*.  If the status is unknown or not applicable, the
  *	value will be %ETH_TP_MDI_INVALID.  Read-only.
@@ -1359,7 +1359,7 @@ enum ethtool_sfeatures_retval_bits {
  * it was forced up into this mode or autonegotiated.
  */
 
-/* The forced speed, 10Mb, 100Mb, gigabit, [2.5|5|10|20|25|40|50|56|100]GbE. */
+/* The forced speed, in units of 1Mb. All values 0 to INT_MAX are legal. */
 #define SPEED_10		10
 #define SPEED_100		100
 #define SPEED_1000		1000
-- 
cgit v0.10.2


From 4a92602aa1cd5bbaeedbd9536ff992f7d26fe9d1 Mon Sep 17 00:00:00 2001
From: Tycho Andersen <tycho.andersen@canonical.com>
Date: Fri, 5 Feb 2016 09:20:52 -0700
Subject: openvswitch: allow management from inside user namespaces

Operations with the GENL_ADMIN_PERM flag fail permissions checks because
this flag means we call netlink_capable, which uses the init user ns.

Instead, let's introduce a new flag, GENL_UNS_ADMIN_PERM for operations
which should be allowed inside a user namespace.

The motivation for this is to be able to run openvswitch in unprivileged
containers. I've tested this and it seems to work, but I really have no
idea about the security consequences of this patch, so thoughts would be
much appreciated.

v2: use the GENL_UNS_ADMIN_PERM flag instead of a check in each function
v3: use separate ifs for UNS_ADMIN_PERM and ADMIN_PERM, instead of one
    massive one

Reported-by: James Page <james.page@canonical.com>
Signed-off-by: Tycho Andersen <tycho.andersen@canonical.com>
CC: Eric Biederman <ebiederm@xmission.com>
CC: Pravin Shelar <pshelar@ovn.org>
CC: Justin Pettit <jpettit@nicira.com>
CC: "David S. Miller" <davem@davemloft.net>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/genetlink.h b/include/uapi/linux/genetlink.h
index c3363ba..5512c90 100644
--- a/include/uapi/linux/genetlink.h
+++ b/include/uapi/linux/genetlink.h
@@ -21,6 +21,7 @@ struct genlmsghdr {
 #define GENL_CMD_CAP_DO		0x02
 #define GENL_CMD_CAP_DUMP	0x04
 #define GENL_CMD_CAP_HASPOL	0x08
+#define GENL_UNS_ADMIN_PERM	0x10
 
 /*
  * List of reserved static generic netlink identifiers:
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index f830326..0ffd721 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -580,6 +580,10 @@ static int genl_family_rcv_msg(struct genl_family *family,
 	    !netlink_capable(skb, CAP_NET_ADMIN))
 		return -EPERM;
 
+	if ((ops->flags & GENL_UNS_ADMIN_PERM) &&
+	    !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
+		return -EPERM;
+
 	if ((nlh->nlmsg_flags & NLM_F_DUMP) == NLM_F_DUMP) {
 		int rc;
 
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index deadfda..d6f7fe9 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -654,7 +654,7 @@ static const struct nla_policy packet_policy[OVS_PACKET_ATTR_MAX + 1] = {
 
 static const struct genl_ops dp_packet_genl_ops[] = {
 	{ .cmd = OVS_PACKET_CMD_EXECUTE,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = packet_policy,
 	  .doit = ovs_packet_cmd_execute
 	}
@@ -1391,12 +1391,12 @@ static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = {
 
 static const struct genl_ops dp_flow_genl_ops[] = {
 	{ .cmd = OVS_FLOW_CMD_NEW,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = flow_policy,
 	  .doit = ovs_flow_cmd_new
 	},
 	{ .cmd = OVS_FLOW_CMD_DEL,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = flow_policy,
 	  .doit = ovs_flow_cmd_del
 	},
@@ -1407,7 +1407,7 @@ static const struct genl_ops dp_flow_genl_ops[] = {
 	  .dumpit = ovs_flow_cmd_dump
 	},
 	{ .cmd = OVS_FLOW_CMD_SET,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = flow_policy,
 	  .doit = ovs_flow_cmd_set,
 	},
@@ -1777,12 +1777,12 @@ static const struct nla_policy datapath_policy[OVS_DP_ATTR_MAX + 1] = {
 
 static const struct genl_ops dp_datapath_genl_ops[] = {
 	{ .cmd = OVS_DP_CMD_NEW,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = datapath_policy,
 	  .doit = ovs_dp_cmd_new
 	},
 	{ .cmd = OVS_DP_CMD_DEL,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = datapath_policy,
 	  .doit = ovs_dp_cmd_del
 	},
@@ -1793,7 +1793,7 @@ static const struct genl_ops dp_datapath_genl_ops[] = {
 	  .dumpit = ovs_dp_cmd_dump
 	},
 	{ .cmd = OVS_DP_CMD_SET,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = datapath_policy,
 	  .doit = ovs_dp_cmd_set,
 	},
@@ -2158,12 +2158,12 @@ static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = {
 
 static const struct genl_ops dp_vport_genl_ops[] = {
 	{ .cmd = OVS_VPORT_CMD_NEW,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = vport_policy,
 	  .doit = ovs_vport_cmd_new
 	},
 	{ .cmd = OVS_VPORT_CMD_DEL,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = vport_policy,
 	  .doit = ovs_vport_cmd_del
 	},
@@ -2174,7 +2174,7 @@ static const struct genl_ops dp_vport_genl_ops[] = {
 	  .dumpit = ovs_vport_cmd_dump
 	},
 	{ .cmd = OVS_VPORT_CMD_SET,
-	  .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
+	  .flags = GENL_UNS_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
 	  .policy = vport_policy,
 	  .doit = ovs_vport_cmd_set,
 	},
-- 
cgit v0.10.2


From e6515203202a913816abb3fb808e71e3a2a0967e Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 10 Feb 2016 10:08:54 +0530
Subject: net: hamradio: baycom_ser_fdx: Replace timeval with timespec64

32 bit systems using 'struct timeval' will break in the year 2038, so
we replace the code appropriately. However, this driver is not broken
in 2038 since we are only using microseconds portion of the time.

This patch replaces 'struct timeval' with 'struct timespec64'. We only
need to find elapsed microseconds rather than absolute time, so it's
better to use monotonic time, so using ktime_get_ts64() makes the code
more efficient and more robust against concurrent settimeofday()
calls.

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Thomas Sailer <t.sailer@alumni.ethz.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/hamradio/baycom_ser_fdx.c b/drivers/net/hamradio/baycom_ser_fdx.c
index 636b65c..7b916d5 100644
--- a/drivers/net/hamradio/baycom_ser_fdx.c
+++ b/drivers/net/hamradio/baycom_ser_fdx.c
@@ -80,6 +80,7 @@
 #include <linux/hdlcdrv.h>
 #include <linux/baycom.h>
 #include <linux/jiffies.h>
+#include <linux/time64.h>
 
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -228,14 +229,15 @@ static inline unsigned int hweight8(unsigned int w)
 
 /* --------------------------------------------------------------------- */
 
-static __inline__ void ser12_rx(struct net_device *dev, struct baycom_state *bc, struct timeval *tv, unsigned char curs)
+static __inline__ void ser12_rx(struct net_device *dev, struct baycom_state *bc, struct timespec64 *ts, unsigned char curs)
 {
 	int timediff;
 	int bdus8 = bc->baud_us >> 3;
 	int bdus4 = bc->baud_us >> 2;
 	int bdus2 = bc->baud_us >> 1;
 
-	timediff = 1000000 + tv->tv_usec - bc->modem.ser12.pll_time;
+	timediff = 1000000 + ts->tv_nsec / NSEC_PER_USEC -
+					bc->modem.ser12.pll_time;
 	while (timediff >= 500000)
 		timediff -= 1000000;
 	while (timediff >= bdus2) {
@@ -287,7 +289,7 @@ static irqreturn_t ser12_interrupt(int irq, void *dev_id)
 {
 	struct net_device *dev = (struct net_device *)dev_id;
 	struct baycom_state *bc = netdev_priv(dev);
-	struct timeval tv;
+	struct timespec64 ts;
 	unsigned char iir, msr;
 	unsigned int txcount = 0;
 
@@ -297,7 +299,7 @@ static irqreturn_t ser12_interrupt(int irq, void *dev_id)
 	if ((iir = inb(IIR(dev->base_addr))) & 1) 	
 		return IRQ_NONE;
 	/* get current time */
-	do_gettimeofday(&tv);
+	ktime_get_ts64(&ts);
 	msr = inb(MSR(dev->base_addr));
 	/* delta DCD */
 	if ((msr & 8) && bc->opt_dcd)
@@ -340,7 +342,7 @@ static irqreturn_t ser12_interrupt(int irq, void *dev_id)
 		}
 		iir = inb(IIR(dev->base_addr));
 	} while (!(iir & 1));
-	ser12_rx(dev, bc, &tv, msr & 0x10); /* CTS */
+	ser12_rx(dev, bc, &ts, msr & 0x10); /* CTS */
 	if (bc->modem.ptt && txcount) {
 		if (bc->modem.ser12.txshreg <= 1) {
 			bc->modem.ser12.txshreg = 0x10000 | hdlcdrv_getbits(&bc->hdrv);
-- 
cgit v0.10.2


From 3e2a5e1539064329f5f6888274600841ad6f87bd Mon Sep 17 00:00:00 2001
From: Sergio Prado <sergio.prado@e-labworks.com>
Date: Tue, 9 Feb 2016 12:07:16 -0200
Subject: net: macb: add wake-on-lan support via magic packet

Tested on Acqua A5 SoM (http://www.acmesystems.it/acqua).

Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/macb.txt b/Documentation/devicetree/bindings/net/macb.txt
index d2e243b..c6b1cb5 100644
--- a/Documentation/devicetree/bindings/net/macb.txt
+++ b/Documentation/devicetree/bindings/net/macb.txt
@@ -25,6 +25,8 @@ Required properties:
 
 Optional properties for PHY child node:
 - reset-gpios : Should specify the gpio for phy reset
+- cdns,magic-packet : If present, indicates that the hardware supports waking
+  up via magic packet.
 
 Examples:
 
diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c
index 50c9410..69af049 100644
--- a/drivers/net/ethernet/cadence/macb.c
+++ b/drivers/net/ethernet/cadence/macb.c
@@ -58,6 +58,9 @@
 
 #define GEM_MTU_MIN_SIZE	68
 
+#define MACB_WOL_HAS_MAGIC_PACKET	(0x1 << 0)
+#define MACB_WOL_ENABLED		(0x1 << 1)
+
 /*
  * Graceful stop timeouts in us. We should allow up to
  * 1 frame time (10 Mbits/s, full-duplex, ignoring collisions)
@@ -2124,6 +2127,39 @@ static void macb_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 	}
 }
 
+static void macb_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	wol->supported = 0;
+	wol->wolopts = 0;
+
+	if (bp->wol & MACB_WOL_HAS_MAGIC_PACKET) {
+		wol->supported = WAKE_MAGIC;
+
+		if (bp->wol & MACB_WOL_ENABLED)
+			wol->wolopts |= WAKE_MAGIC;
+	}
+}
+
+static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct macb *bp = netdev_priv(netdev);
+
+	if (!(bp->wol & MACB_WOL_HAS_MAGIC_PACKET) ||
+	    (wol->wolopts & ~WAKE_MAGIC))
+		return -EOPNOTSUPP;
+
+	if (wol->wolopts & WAKE_MAGIC)
+		bp->wol |= MACB_WOL_ENABLED;
+	else
+		bp->wol &= ~MACB_WOL_ENABLED;
+
+	device_set_wakeup_enable(&bp->pdev->dev, bp->wol & MACB_WOL_ENABLED);
+
+	return 0;
+}
+
 static const struct ethtool_ops macb_ethtool_ops = {
 	.get_settings		= macb_get_settings,
 	.set_settings		= macb_set_settings,
@@ -2131,6 +2167,8 @@ static const struct ethtool_ops macb_ethtool_ops = {
 	.get_regs		= macb_get_regs,
 	.get_link		= ethtool_op_get_link,
 	.get_ts_info		= ethtool_op_get_ts_info,
+	.get_wol		= macb_get_wol,
+	.set_wol		= macb_set_wol,
 };
 
 static const struct ethtool_ops gem_ethtool_ops = {
@@ -2890,6 +2928,11 @@ static int macb_probe(struct platform_device *pdev)
 	if (macb_config)
 		bp->jumbo_max_len = macb_config->jumbo_max_len;
 
+	bp->wol = 0;
+	if (of_get_property(np, "cdns,magic-packet", NULL))
+		bp->wol |= MACB_WOL_HAS_MAGIC_PACKET;
+	device_init_wakeup(&pdev->dev, bp->wol & MACB_WOL_HAS_MAGIC_PACKET);
+
 	spin_lock_init(&bp->lock);
 
 	/* setup capabilities */
@@ -3006,9 +3049,15 @@ static int __maybe_unused macb_suspend(struct device *dev)
 	netif_carrier_off(netdev);
 	netif_device_detach(netdev);
 
-	clk_disable_unprepare(bp->tx_clk);
-	clk_disable_unprepare(bp->hclk);
-	clk_disable_unprepare(bp->pclk);
+	if (bp->wol & MACB_WOL_ENABLED) {
+		macb_writel(bp, IER, MACB_BIT(WOL));
+		macb_writel(bp, WOL, MACB_BIT(MAG));
+		enable_irq_wake(bp->queues[0].irq);
+	} else {
+		clk_disable_unprepare(bp->tx_clk);
+		clk_disable_unprepare(bp->hclk);
+		clk_disable_unprepare(bp->pclk);
+	}
 
 	return 0;
 }
@@ -3019,9 +3068,15 @@ static int __maybe_unused macb_resume(struct device *dev)
 	struct net_device *netdev = platform_get_drvdata(pdev);
 	struct macb *bp = netdev_priv(netdev);
 
-	clk_prepare_enable(bp->pclk);
-	clk_prepare_enable(bp->hclk);
-	clk_prepare_enable(bp->tx_clk);
+	if (bp->wol & MACB_WOL_ENABLED) {
+		macb_writel(bp, IDR, MACB_BIT(WOL));
+		macb_writel(bp, WOL, 0);
+		disable_irq_wake(bp->queues[0].irq);
+	} else {
+		clk_prepare_enable(bp->pclk);
+		clk_prepare_enable(bp->hclk);
+		clk_prepare_enable(bp->tx_clk);
+	}
 
 	netif_device_attach(netdev);
 
diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 0d4ecfc..9ba416d 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -312,6 +312,8 @@
 #define MACB_PFR_SIZE		1
 #define MACB_PTZ_OFFSET		13 /* Enable pause time zero interrupt */
 #define MACB_PTZ_SIZE		1
+#define MACB_WOL_OFFSET		14 /* Enable wake-on-lan interrupt */
+#define MACB_WOL_SIZE		1
 
 /* Bitfields in MAN */
 #define MACB_DATA_OFFSET	0 /* data */
@@ -842,6 +844,8 @@ struct macb {
 
 	unsigned int		rx_frm_len_mask;
 	unsigned int		jumbo_max_len;
+
+	u32			wol;
 };
 
 static inline bool macb_is_gem(struct macb *bp)
-- 
cgit v0.10.2


From 1e2a8868e0541b7d8870781b714d74ac7cad9a6a Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Tue, 9 Feb 2016 10:37:46 +0000
Subject: bonding: use return instead of goto

Replace 'goto' with 'return' to remove unnecessary check at label:
err_undo_flags.

The reason is that 'err_undo_flags' do two things for the first slave device:
1.revert bond mac address if it is set by the slave device.
2.revert bond device type if it's not ARPHRD_ETHER.

It's not necessary for the following three places, they changed neither bond
mac address nor type. It's straightforward to return directly.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 705cb01..45bdd87 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1379,8 +1379,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	if (slave_dev->flags & IFF_UP) {
 		netdev_err(bond_dev, "%s is up - this may be due to an out of date ifenslave\n",
 			   slave_dev->name);
-		res = -EPERM;
-		goto err_undo_flags;
+		return -EPERM;
 	}
 
 	/* set bonding device ether type by slave - bonding netdevices are
@@ -1400,8 +1399,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 			res = notifier_to_errno(res);
 			if (res) {
 				netdev_err(bond_dev, "refused to change device type\n");
-				res = -EBUSY;
-				goto err_undo_flags;
+				return -EBUSY;
 			}
 
 			/* Flush unicast and multicast addresses */
@@ -1421,8 +1419,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	} else if (bond_dev->type != slave_dev->type) {
 		netdev_err(bond_dev, "%s ether type (%d) is different from other slaves (%d), can not enslave it\n",
 			   slave_dev->name, slave_dev->type, bond_dev->type);
-		res = -EINVAL;
-		goto err_undo_flags;
+		return -EINVAL;
 	}
 
 	if (slave_ops->ndo_set_mac_address == NULL) {
-- 
cgit v0.10.2


From 815c52700746cdcc0874a33390bac334a4b90107 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 8 Feb 2016 23:29:21 +0200
Subject: igmp: Namespaceify igmp_max_memberships sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 9c9de11..57d6d06 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -38,7 +38,6 @@ static inline struct igmpv3_query *
 }
 
 extern int sysctl_igmp_llm_reports;
-extern int sysctl_igmp_max_memberships;
 extern int sysctl_igmp_max_msf;
 extern int sysctl_igmp_qrv;
 
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 4d6ec3f..759cf62 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -108,6 +108,8 @@ struct netns_ipv4 {
 	int sysctl_tcp_fin_timeout;
 	unsigned int sysctl_tcp_notsent_lowat;
 
+	int sysctl_igmp_max_memberships;
+
 	struct ping_group_range ping_group_range;
 
 	atomic_t dev_addr_genid;
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 05e4cba..5b86257 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -107,7 +107,6 @@
 #include <linux/seq_file.h>
 #endif
 
-#define IP_MAX_MEMBERSHIPS	20
 #define IP_MAX_MSF		10
 
 /* IGMP reports for link-local multicast groups are enabled by default */
@@ -1727,7 +1726,6 @@ static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr)
 /*
  *	Join a socket to a group
  */
-int sysctl_igmp_max_memberships __read_mostly = IP_MAX_MEMBERSHIPS;
 int sysctl_igmp_max_msf __read_mostly = IP_MAX_MSF;
 #ifdef CONFIG_IP_MULTICAST
 int sysctl_igmp_qrv __read_mostly = IGMP_QUERY_ROBUSTNESS_VARIABLE;
@@ -2074,7 +2072,7 @@ int ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr)
 		count++;
 	}
 	err = -ENOBUFS;
-	if (count >= sysctl_igmp_max_memberships)
+	if (count >= net->ipv4.sysctl_igmp_max_memberships)
 		goto done;
 	iml = sock_kmalloc(sk, sizeof(*iml), GFP_KERNEL);
 	if (!iml)
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 44bb598..6ea3dbb 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -368,13 +368,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "igmp_max_memberships",
-		.data		= &sysctl_igmp_max_memberships,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "igmp_max_msf",
 		.data		= &sysctl_igmp_max_msf,
 		.maxlen		= sizeof(int),
@@ -872,6 +865,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
+		.procname	= "igmp_max_memberships",
+		.data		= &init_net.ipv4.sysctl_igmp_max_memberships,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{
 		.procname	= "tcp_keepalive_time",
 		.data		= &init_net.ipv4.sysctl_tcp_keepalive_time,
 		.maxlen		= sizeof(int),
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3f872a6..4b20378 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2399,6 +2399,8 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT;
 	net->ipv4.sysctl_tcp_notsent_lowat = UINT_MAX;
 
+	net->ipv4.sysctl_igmp_max_memberships = 20;
+
 	return 0;
 fail:
 	tcp_sk_exit(net);
-- 
cgit v0.10.2


From 166b6b2d6f01be67a83b87ab5c91350a68b17115 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 8 Feb 2016 23:29:22 +0200
Subject: igmp: Namespaceify igmp_max_msf sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index 57d6d06..a91ec9f 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -38,7 +38,6 @@ static inline struct igmpv3_query *
 }
 
 extern int sysctl_igmp_llm_reports;
-extern int sysctl_igmp_max_msf;
 extern int sysctl_igmp_qrv;
 
 struct ip_sf_socklist {
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 759cf62..522a2cf 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -109,6 +109,7 @@ struct netns_ipv4 {
 	unsigned int sysctl_tcp_notsent_lowat;
 
 	int sysctl_igmp_max_memberships;
+	int sysctl_igmp_max_msf;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 5b86257..6da2e46 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -107,8 +107,6 @@
 #include <linux/seq_file.h>
 #endif
 
-#define IP_MAX_MSF		10
-
 /* IGMP reports for link-local multicast groups are enabled by default */
 int sysctl_igmp_llm_reports __read_mostly = 1;
 
@@ -1726,7 +1724,6 @@ static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr)
 /*
  *	Join a socket to a group
  */
-int sysctl_igmp_max_msf __read_mostly = IP_MAX_MSF;
 #ifdef CONFIG_IP_MULTICAST
 int sysctl_igmp_qrv __read_mostly = IGMP_QUERY_ROBUSTNESS_VARIABLE;
 #endif
@@ -2244,7 +2241,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct
 	}
 	/* else, add a new source to the filter */
 
-	if (psl && psl->sl_count >= sysctl_igmp_max_msf) {
+	if (psl && psl->sl_count >= net->ipv4.sysctl_igmp_max_msf) {
 		err = -ENOBUFS;
 		goto done;
 	}
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 5f73a7c..92808f1 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -571,6 +571,7 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 			    int optname, char __user *optval, unsigned int optlen)
 {
 	struct inet_sock *inet = inet_sk(sk);
+	struct net *net = sock_net(sk);
 	int val = 0, err;
 	bool needs_rtnl = setsockopt_needs_rtnl(optname);
 
@@ -910,7 +911,7 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 		}
 		/* numsrc >= (1G-4) overflow in 32 bits */
 		if (msf->imsf_numsrc >= 0x3ffffffcU ||
-		    msf->imsf_numsrc > sysctl_igmp_max_msf) {
+		    msf->imsf_numsrc > net->ipv4.sysctl_igmp_max_msf) {
 			kfree(msf);
 			err = -ENOBUFS;
 			break;
@@ -1065,7 +1066,7 @@ static int do_ip_setsockopt(struct sock *sk, int level,
 
 		/* numsrc >= (4G-140)/128 overflow in 32 bits */
 		if (gsf->gf_numsrc >= 0x1ffffff ||
-		    gsf->gf_numsrc > sysctl_igmp_max_msf) {
+		    gsf->gf_numsrc > net->ipv4.sysctl_igmp_max_msf) {
 			err = -ENOBUFS;
 			goto mc_msf_out;
 		}
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 6ea3dbb..225659a 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -367,13 +367,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
-	{
-		.procname	= "igmp_max_msf",
-		.data		= &sysctl_igmp_max_msf,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
 #ifdef CONFIG_IP_MULTICAST
 	{
 		.procname	= "igmp_qrv",
@@ -872,6 +865,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
+		.procname	= "igmp_max_msf",
+		.data		= &init_net.ipv4.sysctl_igmp_max_msf,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{
 		.procname	= "tcp_keepalive_time",
 		.data		= &init_net.ipv4.sysctl_tcp_keepalive_time,
 		.maxlen		= sizeof(int),
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 4b20378..055d8a9 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2400,6 +2400,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_notsent_lowat = UINT_MAX;
 
 	net->ipv4.sysctl_igmp_max_memberships = 20;
+	net->ipv4.sysctl_igmp_max_msf = 10;
 
 	return 0;
 fail:
-- 
cgit v0.10.2


From 87a8a2ae65b7721893c7922f963502be8fa01c94 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <n.borisov@siteground.com>
Date: Tue, 9 Feb 2016 00:13:50 +0200
Subject: igmp: Namespaceify igmp_llm_reports sysctl knob

This was initially introduced in df2cf4a78e488d26 ("IGMP: Inhibit
reports for local multicast groups") by defining the sysctl in the
ipv4_net_table array, however it was never implemented to be
namespace aware. Fix this by changing the code accordingly.

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index a91ec9f..c683f4b 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -37,7 +37,6 @@ static inline struct igmpv3_query *
 	return (struct igmpv3_query *)skb_transport_header(skb);
 }
 
-extern int sysctl_igmp_llm_reports;
 extern int sysctl_igmp_qrv;
 
 struct ip_sf_socklist {
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 522a2cf..cbbf811 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -110,6 +110,7 @@ struct netns_ipv4 {
 
 	int sysctl_igmp_max_memberships;
 	int sysctl_igmp_max_msf;
+	int sysctl_igmp_llm_reports;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 6da2e46..2e22ee0 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -107,9 +107,6 @@
 #include <linux/seq_file.h>
 #endif
 
-/* IGMP reports for link-local multicast groups are enabled by default */
-int sysctl_igmp_llm_reports __read_mostly = 1;
-
 #ifdef CONFIG_IP_MULTICAST
 /* Parameter names and values are taken from igmp-v2-06 draft */
 
@@ -430,6 +427,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc,
 	int type, int gdeleted, int sdeleted)
 {
 	struct net_device *dev = pmc->interface->dev;
+	struct net *net = dev_net(dev);
 	struct igmpv3_report *pih;
 	struct igmpv3_grec *pgr = NULL;
 	struct ip_sf_list *psf, *psf_next, *psf_prev, **psf_list;
@@ -437,7 +435,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc,
 
 	if (pmc->multiaddr == IGMP_ALL_HOSTS)
 		return skb;
-	if (ipv4_is_local_multicast(pmc->multiaddr) && !sysctl_igmp_llm_reports)
+	if (ipv4_is_local_multicast(pmc->multiaddr) && !net->ipv4.sysctl_igmp_llm_reports)
 		return skb;
 
 	isquery = type == IGMPV3_MODE_IS_INCLUDE ||
@@ -540,6 +538,7 @@ empty_source:
 static int igmpv3_send_report(struct in_device *in_dev, struct ip_mc_list *pmc)
 {
 	struct sk_buff *skb = NULL;
+	struct net *net = dev_net(in_dev->dev);
 	int type;
 
 	if (!pmc) {
@@ -548,7 +547,7 @@ static int igmpv3_send_report(struct in_device *in_dev, struct ip_mc_list *pmc)
 			if (pmc->multiaddr == IGMP_ALL_HOSTS)
 				continue;
 			if (ipv4_is_local_multicast(pmc->multiaddr) &&
-			     !sysctl_igmp_llm_reports)
+			     !net->ipv4.sysctl_igmp_llm_reports)
 				continue;
 			spin_lock_bh(&pmc->lock);
 			if (pmc->sfcount[MCAST_EXCLUDE])
@@ -684,7 +683,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
 	if (type == IGMPV3_HOST_MEMBERSHIP_REPORT)
 		return igmpv3_send_report(in_dev, pmc);
 
-	if (ipv4_is_local_multicast(group) && !sysctl_igmp_llm_reports)
+	if (ipv4_is_local_multicast(group) && !net->ipv4.sysctl_igmp_llm_reports)
 		return 0;
 
 	if (type == IGMP_HOST_LEAVE_MESSAGE)
@@ -855,12 +854,13 @@ static int igmp_marksources(struct ip_mc_list *pmc, int nsrcs, __be32 *srcs)
 static bool igmp_heard_report(struct in_device *in_dev, __be32 group)
 {
 	struct ip_mc_list *im;
+	struct net *net = dev_net(in_dev->dev);
 
 	/* Timers are only set for non-local groups */
 
 	if (group == IGMP_ALL_HOSTS)
 		return false;
-	if (ipv4_is_local_multicast(group) && !sysctl_igmp_llm_reports)
+	if (ipv4_is_local_multicast(group) && !net->ipv4.sysctl_igmp_llm_reports)
 		return false;
 
 	rcu_read_lock();
@@ -884,6 +884,7 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb,
 	__be32			group = ih->group;
 	int			max_delay;
 	int			mark = 0;
+	struct net		*net = dev_net(in_dev->dev);
 
 
 	if (len == 8) {
@@ -969,7 +970,7 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb,
 		if (im->multiaddr == IGMP_ALL_HOSTS)
 			continue;
 		if (ipv4_is_local_multicast(im->multiaddr) &&
-		    !sysctl_igmp_llm_reports)
+		    !net->ipv4.sysctl_igmp_llm_reports)
 			continue;
 		spin_lock_bh(&im->lock);
 		if (im->tm_running)
@@ -1184,6 +1185,7 @@ static void igmp_group_dropped(struct ip_mc_list *im)
 {
 	struct in_device *in_dev = im->interface;
 #ifdef CONFIG_IP_MULTICAST
+	struct net *net = dev_net(in_dev->dev);
 	int reporter;
 #endif
 
@@ -1195,7 +1197,7 @@ static void igmp_group_dropped(struct ip_mc_list *im)
 #ifdef CONFIG_IP_MULTICAST
 	if (im->multiaddr == IGMP_ALL_HOSTS)
 		return;
-	if (ipv4_is_local_multicast(im->multiaddr) && !sysctl_igmp_llm_reports)
+	if (ipv4_is_local_multicast(im->multiaddr) && !net->ipv4.sysctl_igmp_llm_reports)
 		return;
 
 	reporter = im->reporter;
@@ -1220,6 +1222,7 @@ static void igmp_group_dropped(struct ip_mc_list *im)
 static void igmp_group_added(struct ip_mc_list *im)
 {
 	struct in_device *in_dev = im->interface;
+	struct net *net = dev_net(in_dev->dev);
 
 	if (im->loaded == 0) {
 		im->loaded = 1;
@@ -1229,7 +1232,7 @@ static void igmp_group_added(struct ip_mc_list *im)
 #ifdef CONFIG_IP_MULTICAST
 	if (im->multiaddr == IGMP_ALL_HOSTS)
 		return;
-	if (ipv4_is_local_multicast(im->multiaddr) && !sysctl_igmp_llm_reports)
+	if (ipv4_is_local_multicast(im->multiaddr) && !net->ipv4.sysctl_igmp_llm_reports)
 		return;
 
 	if (in_dev->dead)
@@ -1530,6 +1533,7 @@ static void ip_mc_rejoin_groups(struct in_device *in_dev)
 #ifdef CONFIG_IP_MULTICAST
 	struct ip_mc_list *im;
 	int type;
+	struct net *net = dev_net(in_dev->dev);
 
 	ASSERT_RTNL();
 
@@ -1537,7 +1541,7 @@ static void ip_mc_rejoin_groups(struct in_device *in_dev)
 		if (im->multiaddr == IGMP_ALL_HOSTS)
 			continue;
 		if (ipv4_is_local_multicast(im->multiaddr) &&
-		    !sysctl_igmp_llm_reports)
+		    !net->ipv4.sysctl_igmp_llm_reports)
 			continue;
 
 		/* a failover is happening and switches
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 225659a..fc40fa1 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -852,7 +852,7 @@ static struct ctl_table ipv4_net_table[] = {
 	},
 	{
 		.procname	= "igmp_link_local_mcast_reports",
-		.data		= &sysctl_igmp_llm_reports,
+		.data		= &init_net.ipv4.sysctl_igmp_llm_reports,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 055d8a9..6c3c1d5 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2401,6 +2401,8 @@ static int __net_init tcp_sk_init(struct net *net)
 
 	net->ipv4.sysctl_igmp_max_memberships = 20;
 	net->ipv4.sysctl_igmp_max_msf = 10;
+	/* IGMP reports for link-local multicast groups are enabled by default */
+	net->ipv4.sysctl_igmp_llm_reports = 1;
 
 	return 0;
 fail:
-- 
cgit v0.10.2


From 165094afcee79e4d5b6e94032a5d3be157460b4a Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 8 Feb 2016 23:29:24 +0200
Subject: igmp: Namespacify igmp_qrv sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index c683f4b..12f6fba 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -37,8 +37,6 @@ static inline struct igmpv3_query *
 	return (struct igmpv3_query *)skb_transport_header(skb);
 }
 
-extern int sysctl_igmp_qrv;
-
 struct ip_sf_socklist {
 	unsigned int		sl_max;
 	unsigned int		sl_count;
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index cbbf811..848fe80 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -111,6 +111,7 @@ struct netns_ipv4 {
 	int sysctl_igmp_max_memberships;
 	int sysctl_igmp_max_msf;
 	int sysctl_igmp_llm_reports;
+	int sysctl_igmp_qrv;
 
 	struct ping_group_range ping_group_range;
 
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 2e22ee0..7c95335 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -762,9 +762,10 @@ static void igmp_ifc_timer_expire(unsigned long data)
 
 static void igmp_ifc_event(struct in_device *in_dev)
 {
+	struct net *net = dev_net(in_dev->dev);
 	if (IGMP_V1_SEEN(in_dev) || IGMP_V2_SEEN(in_dev))
 		return;
-	in_dev->mr_ifc_count = in_dev->mr_qrv ?: sysctl_igmp_qrv;
+	in_dev->mr_ifc_count = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
 	igmp_ifc_start_timer(in_dev, 1);
 }
 
@@ -1086,6 +1087,7 @@ static void ip_mc_filter_del(struct in_device *in_dev, __be32 addr)
 static void igmpv3_add_delrec(struct in_device *in_dev, struct ip_mc_list *im)
 {
 	struct ip_mc_list *pmc;
+	struct net *net = dev_net(in_dev->dev);
 
 	/* this is an "ip_mc_list" for convenience; only the fields below
 	 * are actually used. In particular, the refcnt and users are not
@@ -1100,7 +1102,7 @@ static void igmpv3_add_delrec(struct in_device *in_dev, struct ip_mc_list *im)
 	pmc->interface = im->interface;
 	in_dev_hold(in_dev);
 	pmc->multiaddr = im->multiaddr;
-	pmc->crcount = in_dev->mr_qrv ?: sysctl_igmp_qrv;
+	pmc->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
 	pmc->sfmode = im->sfmode;
 	if (pmc->sfmode == MCAST_INCLUDE) {
 		struct ip_sf_list *psf;
@@ -1245,7 +1247,7 @@ static void igmp_group_added(struct ip_mc_list *im)
 	}
 	/* else, v3 */
 
-	im->crcount = in_dev->mr_qrv ?: sysctl_igmp_qrv;
+	im->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
 	igmp_ifc_event(in_dev);
 #endif
 }
@@ -1314,6 +1316,7 @@ static void ip_mc_hash_remove(struct in_device *in_dev,
 void ip_mc_inc_group(struct in_device *in_dev, __be32 addr)
 {
 	struct ip_mc_list *im;
+	struct net *net = dev_net(in_dev->dev);
 
 	ASSERT_RTNL();
 
@@ -1340,7 +1343,7 @@ void ip_mc_inc_group(struct in_device *in_dev, __be32 addr)
 	spin_lock_init(&im->lock);
 #ifdef CONFIG_IP_MULTICAST
 	setup_timer(&im->timer, igmp_timer_expire, (unsigned long)im);
-	im->unsolicit_count = sysctl_igmp_qrv;
+	im->unsolicit_count = net->ipv4.sysctl_igmp_qrv;
 #endif
 
 	im->next_rcu = in_dev->mc_list;
@@ -1640,6 +1643,7 @@ void ip_mc_down(struct in_device *in_dev)
 
 void ip_mc_init_dev(struct in_device *in_dev)
 {
+	struct net *net = dev_net(in_dev->dev);
 	ASSERT_RTNL();
 
 #ifdef CONFIG_IP_MULTICAST
@@ -1647,7 +1651,7 @@ void ip_mc_init_dev(struct in_device *in_dev)
 			(unsigned long)in_dev);
 	setup_timer(&in_dev->mr_ifc_timer, igmp_ifc_timer_expire,
 			(unsigned long)in_dev);
-	in_dev->mr_qrv = sysctl_igmp_qrv;
+	in_dev->mr_qrv = net->ipv4.sysctl_igmp_qrv;
 #endif
 
 	spin_lock_init(&in_dev->mc_tomb_lock);
@@ -1658,11 +1662,12 @@ void ip_mc_init_dev(struct in_device *in_dev)
 void ip_mc_up(struct in_device *in_dev)
 {
 	struct ip_mc_list *pmc;
+	struct net *net = dev_net(in_dev->dev);
 
 	ASSERT_RTNL();
 
 #ifdef CONFIG_IP_MULTICAST
-	in_dev->mr_qrv = sysctl_igmp_qrv;
+	in_dev->mr_qrv = net->ipv4.sysctl_igmp_qrv;
 #endif
 	ip_mc_inc_group(in_dev, IGMP_ALL_HOSTS);
 
@@ -1728,9 +1733,6 @@ static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr)
 /*
  *	Join a socket to a group
  */
-#ifdef CONFIG_IP_MULTICAST
-int sysctl_igmp_qrv __read_mostly = IGMP_QUERY_ROBUSTNESS_VARIABLE;
-#endif
 
 static int ip_mc_del1_src(struct ip_mc_list *pmc, int sfmode,
 	__be32 *psfsrc)
@@ -1755,6 +1757,7 @@ static int ip_mc_del1_src(struct ip_mc_list *pmc, int sfmode,
 	if (!psf->sf_count[MCAST_INCLUDE] && !psf->sf_count[MCAST_EXCLUDE]) {
 #ifdef CONFIG_IP_MULTICAST
 		struct in_device *in_dev = pmc->interface;
+		struct net *net = dev_net(in_dev->dev);
 #endif
 
 		/* no more filters for this source */
@@ -1765,7 +1768,7 @@ static int ip_mc_del1_src(struct ip_mc_list *pmc, int sfmode,
 #ifdef CONFIG_IP_MULTICAST
 		if (psf->sf_oldin &&
 		    !IGMP_V1_SEEN(in_dev) && !IGMP_V2_SEEN(in_dev)) {
-			psf->sf_crcount = in_dev->mr_qrv ?: sysctl_igmp_qrv;
+			psf->sf_crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
 			psf->sf_next = pmc->tomb;
 			pmc->tomb = psf;
 			rv = 1;
@@ -1823,12 +1826,13 @@ static int ip_mc_del_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
 	    pmc->sfcount[MCAST_INCLUDE]) {
 #ifdef CONFIG_IP_MULTICAST
 		struct ip_sf_list *psf;
+		struct net *net = dev_net(in_dev->dev);
 #endif
 
 		/* filter mode change */
 		pmc->sfmode = MCAST_INCLUDE;
 #ifdef CONFIG_IP_MULTICAST
-		pmc->crcount = in_dev->mr_qrv ?: sysctl_igmp_qrv;
+		pmc->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
 		in_dev->mr_ifc_count = pmc->crcount;
 		for (psf = pmc->sources; psf; psf = psf->sf_next)
 			psf->sf_crcount = 0;
@@ -1995,6 +1999,7 @@ static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
 	} else if (isexclude != (pmc->sfcount[MCAST_EXCLUDE] != 0)) {
 #ifdef CONFIG_IP_MULTICAST
 		struct ip_sf_list *psf;
+		struct net *net = dev_net(pmc->interface->dev);
 		in_dev = pmc->interface;
 #endif
 
@@ -2006,7 +2011,7 @@ static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
 #ifdef CONFIG_IP_MULTICAST
 		/* else no filters; keep old mode for reports */
 
-		pmc->crcount = in_dev->mr_qrv ?: sysctl_igmp_qrv;
+		pmc->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
 		in_dev->mr_ifc_count = pmc->crcount;
 		for (psf = pmc->sources; psf; psf = psf->sf_next)
 			psf->sf_crcount = 0;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index fc40fa1..b537338 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -367,16 +367,6 @@ static struct ctl_table ipv4_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
-#ifdef CONFIG_IP_MULTICAST
-	{
-		.procname	= "igmp_qrv",
-		.data		= &sysctl_igmp_qrv,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &one
-	},
-#endif
 	{
 		.procname	= "inet_peer_threshold",
 		.data		= &inet_peer_threshold,
@@ -871,6 +861,16 @@ static struct ctl_table ipv4_net_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
+#ifdef CONFIG_IP_MULTICAST
+	{
+		.procname	= "igmp_qrv",
+		.data		= &init_net.ipv4.sysctl_igmp_qrv,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &one
+	},
+#endif
 	{
 		.procname	= "tcp_keepalive_time",
 		.data		= &init_net.ipv4.sysctl_tcp_keepalive_time,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 6c3c1d5..ba5d014 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2403,6 +2403,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_igmp_max_msf = 10;
 	/* IGMP reports for link-local multicast groups are enabled by default */
 	net->ipv4.sysctl_igmp_llm_reports = 1;
+	net->ipv4.sysctl_igmp_qrv = 2;
 
 	return 0;
 fail:
-- 
cgit v0.10.2


From a05d4845907a6f0296612d24956b189a51fb8df7 Mon Sep 17 00:00:00 2001
From: Thanneeru Srinivasulu <tsrinivasulu@caviumnetworks.com>
Date: Thu, 11 Feb 2016 21:50:21 +0530
Subject: net, thunderx: Add TX timeout and RX buffer alloc failure stats.

When system is low on atomic memory, too many error messages are logged.
Since this is not a total failure but a simple switch to non-atomic allocation
better to have a stat.

Also add a stat for reset, kicked due to transmit watchdog timeout.

Signed-off-by: Thanneeru Srinivasulu <tsrinivasulu@caviumnetworks.com>
Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h
index 6888288..8af363a 100644
--- a/drivers/net/ethernet/cavium/thunder/nic.h
+++ b/drivers/net/ethernet/cavium/thunder/nic.h
@@ -248,10 +248,13 @@ struct nicvf_drv_stats {
 	u64 rx_frames_jumbo;
 	u64 rx_drops;
 
+	u64 rcv_buffer_alloc_failures;
+
 	/* Tx */
 	u64 tx_frames_ok;
 	u64 tx_drops;
 	u64 tx_tso;
+	u64 tx_timeout;
 	u64 txq_stop;
 	u64 txq_wake;
 };
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
index a12b2e3..d2d8ef2 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
@@ -89,9 +89,11 @@ static const struct nicvf_stat nicvf_drv_stats[] = {
 	NICVF_DRV_STAT(rx_frames_1518),
 	NICVF_DRV_STAT(rx_frames_jumbo),
 	NICVF_DRV_STAT(rx_drops),
+	NICVF_DRV_STAT(rcv_buffer_alloc_failures),
 	NICVF_DRV_STAT(tx_frames_ok),
 	NICVF_DRV_STAT(tx_tso),
 	NICVF_DRV_STAT(tx_drops),
+	NICVF_DRV_STAT(tx_timeout),
 	NICVF_DRV_STAT(txq_stop),
 	NICVF_DRV_STAT(txq_wake),
 };
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index c24cb2a..95db6b7 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -1394,6 +1394,7 @@ static void nicvf_tx_timeout(struct net_device *dev)
 		netdev_warn(dev, "%s: Transmit timed out, resetting\n",
 			    dev->name);
 
+	nic->drv_stats.tx_timeout++;
 	schedule_work(&nic->reset_task);
 }
 
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
index d0d1b54..50ab6f4 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
@@ -96,8 +96,7 @@ static inline int nicvf_alloc_rcv_buffer(struct nicvf *nic, gfp_t gfp,
 		nic->rb_page = alloc_pages(gfp | __GFP_COMP | __GFP_NOWARN,
 					   order);
 		if (!nic->rb_page) {
-			netdev_err(nic->netdev,
-				   "Failed to allocate new rcv buffer\n");
+			nic->drv_stats.rcv_buffer_alloc_failures++;
 			return -ENOMEM;
 		}
 		nic->rb_page_offset = 0;
-- 
cgit v0.10.2


From ef0a4d8601760b346d9d0893f2a554c338861c4f Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@cavium.com>
Date: Thu, 11 Feb 2016 21:50:22 +0530
Subject: net: thunderx: Use napi_schedule_irqoff()

napi_schedule is being called from hard irq context, hence
switch to napi_schedule_irqoff which avoids unneeded call
to local_irq_save and local_irq_restore.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index 95db6b7..c6f146c 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -828,7 +828,7 @@ static irqreturn_t nicvf_intr_handler(int irq, void *cq_irq)
 	nicvf_disable_intr(nic, NICVF_INTR_CQ, qidx);
 
 	/* Schedule NAPI */
-	napi_schedule(&cq_poll->napi);
+	napi_schedule_irqoff(&cq_poll->napi);
 
 	/* Clear interrupt */
 	nicvf_clear_intr(nic, NICVF_INTR_CQ, qidx);
-- 
cgit v0.10.2


From fb4b7d98a0215fc3310c8415a86acfe726de395c Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@cavium.com>
Date: Thu, 11 Feb 2016 21:50:23 +0530
Subject: net: thunderx: Assign affinity hints to vf's interrupts

This affinity hint can be used by user space irqbalance tool to set
preferred CPU mask for irqs registered by this VF. Irqbalance needs
to be in 'exact' mode to set irq affinity same as indicated by
affinity hint.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h
index 8af363a..00cc915 100644
--- a/drivers/net/ethernet/cavium/thunder/nic.h
+++ b/drivers/net/ethernet/cavium/thunder/nic.h
@@ -309,6 +309,7 @@ struct nicvf {
 	struct msix_entry	msix_entries[NIC_VF_MSIX_VECTORS];
 	char			irq_name[NIC_VF_MSIX_VECTORS][20];
 	bool			irq_allocated[NIC_VF_MSIX_VECTORS];
+	cpumask_var_t		affinity_mask[NIC_VF_MSIX_VECTORS];
 
 	/* VF <-> PF mailbox communication */
 	bool			pf_acked;
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index c6f146c..90ce93e 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -899,6 +899,31 @@ static void nicvf_disable_msix(struct nicvf *nic)
 	}
 }
 
+static void nicvf_set_irq_affinity(struct nicvf *nic)
+{
+	int vec, cpu;
+	int irqnum;
+
+	for (vec = 0; vec < nic->num_vec; vec++) {
+		if (!nic->irq_allocated[vec])
+			continue;
+
+		if (!zalloc_cpumask_var(&nic->affinity_mask[vec], GFP_KERNEL))
+			return;
+		 /* CQ interrupts */
+		if (vec < NICVF_INTR_ID_SQ)
+			/* Leave CPU0 for RBDR and other interrupts */
+			cpu = nicvf_netdev_qidx(nic, vec) + 1;
+		else
+			cpu = 0;
+
+		cpumask_set_cpu(cpumask_local_spread(cpu, nic->node),
+				nic->affinity_mask[vec]);
+		irqnum = nic->msix_entries[vec].vector;
+		irq_set_affinity_hint(irqnum, nic->affinity_mask[vec]);
+	}
+}
+
 static int nicvf_register_interrupts(struct nicvf *nic)
 {
 	int irq, ret = 0;
@@ -944,8 +969,13 @@ static int nicvf_register_interrupts(struct nicvf *nic)
 	ret = request_irq(nic->msix_entries[irq].vector,
 			  nicvf_qs_err_intr_handler,
 			  0, nic->irq_name[irq], nic);
-	if (!ret)
-		nic->irq_allocated[irq] = true;
+	if (ret)
+		goto err;
+
+	nic->irq_allocated[irq] = true;
+
+	/* Set IRQ affinities */
+	nicvf_set_irq_affinity(nic);
 
 err:
 	if (ret)
@@ -963,6 +993,9 @@ static void nicvf_unregister_interrupts(struct nicvf *nic)
 		if (!nic->irq_allocated[irq])
 			continue;
 
+		irq_set_affinity_hint(nic->msix_entries[irq].vector, NULL);
+		free_cpumask_var(nic->affinity_mask[irq]);
+
 		if (irq < NICVF_INTR_ID_SQ)
 			free_irq(nic->msix_entries[irq].vector, nic->napi[irq]);
 		else
-- 
cgit v0.10.2


From eee326fd83348ed39a06c0db999ed513d10d9c39 Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Thu, 11 Feb 2016 21:50:24 +0530
Subject: net: thunderx: bgx: Use standard firmware node infrastructure.

In the case of OF device tree, the firmware information is attached to
the BGX device structure in the standard manner, so use the firmware
iterators and accessors where possible.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
index 9df26c2..111835b 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
@@ -968,26 +968,27 @@ static int bgx_init_acpi_phy(struct bgx *bgx)
 
 static int bgx_init_of_phy(struct bgx *bgx)
 {
-	struct device_node *np;
-	struct device_node *np_child;
+	struct fwnode_handle *fwn;
 	u8 lmac = 0;
-	char bgx_sel[5];
 	const char *mac;
 
-	/* Get BGX node from DT */
-	snprintf(bgx_sel, 5, "bgx%d", bgx->bgx_id);
-	np = of_find_node_by_name(NULL, bgx_sel);
-	if (!np)
-		return -ENODEV;
+	device_for_each_child_node(&bgx->pdev->dev, fwn) {
+		struct device_node *phy_np;
+		struct device_node *node = to_of_node(fwn);
+
+		/* If it is not an OF node we cannot handle it yet, so
+		 * exit the loop.
+		 */
+		if (!node)
+			break;
 
-	for_each_child_of_node(np, np_child) {
-		struct device_node *phy_np = of_parse_phandle(np_child,
-							      "phy-handle", 0);
+		phy_np = of_parse_phandle(node, "phy-handle", 0);
 		if (!phy_np)
 			continue;
+
 		bgx->lmac[lmac].phydev = of_phy_find_device(phy_np);
 
-		mac = of_get_mac_address(np_child);
+		mac = of_get_mac_address(node);
 		if (mac)
 			ether_addr_copy(bgx->lmac[lmac].mac, mac);
 
@@ -995,7 +996,7 @@ static int bgx_init_of_phy(struct bgx *bgx)
 		bgx->lmac[lmac].lmacid = lmac;
 		lmac++;
 		if (lmac == MAX_LMAC_PER_BGX) {
-			of_node_put(np_child);
+			of_node_put(node);
 			break;
 		}
 	}
-- 
cgit v0.10.2


From 1d82efaca87ecf53e97c696f9d0a9adefea0c7b5 Mon Sep 17 00:00:00 2001
From: Robert Richter <rrichter@cavium.com>
Date: Thu, 11 Feb 2016 21:50:25 +0530
Subject: net: thunderx: bgx: Add log message when setting mac address

Signed-off-by: Robert Richter <rrichter@cavium.com>
Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
index 111835b..f8abdff 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
@@ -886,7 +886,8 @@ static void bgx_get_qlm_mode(struct bgx *bgx)
 
 #ifdef CONFIG_ACPI
 
-static int acpi_get_mac_address(struct acpi_device *adev, u8 *dst)
+static int acpi_get_mac_address(struct device *dev, struct acpi_device *adev,
+				u8 *dst)
 {
 	u8 mac[ETH_ALEN];
 	int ret;
@@ -897,10 +898,13 @@ static int acpi_get_mac_address(struct acpi_device *adev, u8 *dst)
 		goto out;
 
 	if (!is_valid_ether_addr(mac)) {
+		dev_err(dev, "MAC address invalid: %pM\n", mac);
 		ret = -EINVAL;
 		goto out;
 	}
 
+	dev_info(dev, "MAC address set to: %pM\n", mac);
+
 	memcpy(dst, mac, ETH_ALEN);
 out:
 	return ret;
@@ -911,14 +915,15 @@ static acpi_status bgx_acpi_register_phy(acpi_handle handle,
 					 u32 lvl, void *context, void **rv)
 {
 	struct bgx *bgx = context;
+	struct device *dev = &bgx->pdev->dev;
 	struct acpi_device *adev;
 
 	if (acpi_bus_get_device(handle, &adev))
 		goto out;
 
-	acpi_get_mac_address(adev, bgx->lmac[bgx->lmac_count].mac);
+	acpi_get_mac_address(dev, adev, bgx->lmac[bgx->lmac_count].mac);
 
-	SET_NETDEV_DEV(&bgx->lmac[bgx->lmac_count].netdev, &bgx->pdev->dev);
+	SET_NETDEV_DEV(&bgx->lmac[bgx->lmac_count].netdev, dev);
 
 	bgx->lmac[bgx->lmac_count].lmacid = bgx->lmac_count;
 out:
-- 
cgit v0.10.2


From 6e4be8d6717cb63c58f6b404e63a881c76d8878c Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@cavium.com>
Date: Thu, 11 Feb 2016 21:50:26 +0530
Subject: net: thunderx: Alloc higher order pages when pagesize is small

Allocate higher order pages when pagesize is small, this will
reduce number of calls to page allocator and wastage of memory.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
index 50ab6f4..5adb208 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
@@ -78,7 +78,7 @@ static void nicvf_free_q_desc_mem(struct nicvf *nic, struct q_desc_mem *dmem)
 static inline int nicvf_alloc_rcv_buffer(struct nicvf *nic, gfp_t gfp,
 					 u32 buf_len, u64 **rbuf)
 {
-	int order = get_order(buf_len);
+	int order = (PAGE_SIZE <= 4096) ?  PAGE_ALLOC_COSTLY_ORDER : 0;
 
 	/* Check if request can be accomodated in previous allocated page */
 	if (nic->rb_page) {
-- 
cgit v0.10.2


From aa54c8da9a8e480861346d88b57086490bdd7b03 Mon Sep 17 00:00:00 2001
From: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Date: Tue, 9 Feb 2016 20:47:13 +0100
Subject: net: phy: spi_ks8995: introduce spi_device_id table

Refactor to use spi_device_id table to facilitate easy
extendability.

Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index c72c422..e7a215d 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -89,6 +89,28 @@
 
 #define KS8995_RESET_DELAY	10 /* usec */
 
+enum ks8995_chip_variant {
+	ks8995,
+	ksz8864,
+	max_variant
+};
+
+struct ks8995_chip_params {
+	char *name;
+	int regs_size;
+};
+
+static const struct ks8995_chip_params ks8995_chip[] = {
+	[ks8995] = {
+		.name = "KS8995MA",
+		.regs_size = KS8995_REGS_SIZE,
+	},
+	[ksz8864] = {
+		.name = "KSZ8864RMN",
+		.regs_size = KSZ8864_REGS_SIZE,
+	},
+};
+
 struct ks8995_pdata {
 	/* not yet implemented */
 };
@@ -98,8 +120,16 @@ struct ks8995_switch {
 	struct mutex		lock;
 	struct ks8995_pdata	*pdata;
 	struct bin_attribute	regs_attr;
+	const struct ks8995_chip_params	*chip;
 };
 
+static const struct spi_device_id ks8995_id[] = {
+	{"ks8995", ks8995},
+	{"ksz8864", ksz8864},
+	{ }
+};
+MODULE_DEVICE_TABLE(spi, ks8995_id);
+
 static inline u8 get_chip_id(u8 val)
 {
 	return (val >> ID1_CHIPID_S) & ID1_CHIPID_M;
@@ -244,17 +274,22 @@ static const struct bin_attribute ks8995_registers_attr = {
 };
 
 /* ------------------------------------------------------------------------ */
-
 static int ks8995_probe(struct spi_device *spi)
 {
 	struct ks8995_switch    *ks;
 	struct ks8995_pdata     *pdata;
 	u8      ids[2];
 	int     err;
+	int variant = spi_get_device_id(spi)->driver_data;
 
 	/* Chip description */
 	pdata = spi->dev.platform_data;
 
+	if (variant >= max_variant) {
+		dev_err(&spi->dev, "bad chip variant %d\n", variant);
+		return -ENODEV;
+	}
+
 	ks = devm_kzalloc(&spi->dev, sizeof(*ks), GFP_KERNEL);
 	if (!ks)
 		return -ENOMEM;
@@ -262,6 +297,8 @@ static int ks8995_probe(struct spi_device *spi)
 	mutex_init(&ks->lock);
 	ks->pdata = pdata;
 	ks->spi = spi_dev_get(spi);
+	ks->chip = &ks8995_chip[variant];
+
 	spi_set_drvdata(spi, ks);
 
 	spi->mode = SPI_MODE_0;
@@ -287,6 +324,7 @@ static int ks8995_probe(struct spi_device *spi)
 		return -ENODEV;
 	}
 
+	ks->regs_attr.size = ks->chip->regs_size;
 	memcpy(&ks->regs_attr, &ks8995_registers_attr, sizeof(ks->regs_attr));
 	if (get_chip_id(ids[1]) != CHIPID_M) {
 		u8 val;
@@ -303,7 +341,6 @@ static int ks8995_probe(struct spi_device *spi)
 			dev_err(&spi->dev, "unknown chip:%02x,0\n", ids[1]);
 			return err;
 		}
-		ks->regs_attr.size = KSZ8864_REGS_SIZE;
 	}
 
 	err = ks8995_reset(ks);
@@ -346,6 +383,7 @@ static struct spi_driver ks8995_driver = {
 	},
 	.probe	  = ks8995_probe,
 	.remove	  = ks8995_remove,
+	.id_table = ks8995_id,
 };
 
 module_spi_driver(ks8995_driver);
-- 
cgit v0.10.2


From 484e36ff18341ca1fca4258139100b2e90530f53 Mon Sep 17 00:00:00 2001
From: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Date: Tue, 9 Feb 2016 20:47:14 +0100
Subject: net: phy: spi_ks8995: verify chip and determine revision

Since the chip variant is now determined by spi_device_id, verify
family and chip id and determine the revision id.

Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index e7a215d..5e66d4b 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -83,6 +83,8 @@
 
 #define FAMILY_KS8995		0x95
 #define CHIPID_M		0
+#define KS8995_CHIP_ID		0x00
+#define KSZ8864_CHIP_ID		0x01
 
 #define KS8995_CMD_WRITE	0x02U
 #define KS8995_CMD_READ		0x03U
@@ -97,16 +99,22 @@ enum ks8995_chip_variant {
 
 struct ks8995_chip_params {
 	char *name;
+	int family_id;
+	int chip_id;
 	int regs_size;
 };
 
 static const struct ks8995_chip_params ks8995_chip[] = {
 	[ks8995] = {
 		.name = "KS8995MA",
+		.family_id = FAMILY_KS8995,
+		.chip_id = KS8995_CHIP_ID,
 		.regs_size = KS8995_REGS_SIZE,
 	},
 	[ksz8864] = {
 		.name = "KSZ8864RMN",
+		.family_id = FAMILY_KS8995,
+		.chip_id = KSZ8864_CHIP_ID,
 		.regs_size = KSZ8864_REGS_SIZE,
 	},
 };
@@ -121,6 +129,7 @@ struct ks8995_switch {
 	struct ks8995_pdata	*pdata;
 	struct bin_attribute	regs_attr;
 	const struct ks8995_chip_params	*chip;
+	int			revision_id;
 };
 
 static const struct spi_device_id ks8995_id[] = {
@@ -263,6 +272,73 @@ static ssize_t ks8995_registers_write(struct file *filp, struct kobject *kobj,
 	return ks8995_write(ks8995, buf, off, count);
 }
 
+/* ks8995_get_revision - get chip revision
+ * @ks: pointer to switch instance
+ *
+ * Verify chip family and id and get chip revision.
+ */
+static int ks8995_get_revision(struct ks8995_switch *ks)
+{
+	int err;
+	u8 id0, id1, ksz8864_id;
+
+	/* read family id */
+	err = ks8995_read_reg(ks, KS8995_REG_ID0, &id0);
+	if (err) {
+		err = -EIO;
+		goto err_out;
+	}
+
+	/* verify family id */
+	if (id0 != ks->chip->family_id) {
+		dev_err(&ks->spi->dev, "chip family id mismatch: expected 0x%02x but 0x%02x read\n",
+			ks->chip->family_id, id0);
+		err = -ENODEV;
+		goto err_out;
+	}
+
+	switch (ks->chip->family_id) {
+	case FAMILY_KS8995:
+		/* try reading chip id at CHIP ID1 */
+		err = ks8995_read_reg(ks, KS8995_REG_ID1, &id1);
+		if (err) {
+			err = -EIO;
+			goto err_out;
+		}
+
+		/* verify chip id */
+		if ((get_chip_id(id1) == CHIPID_M) &&
+		    (get_chip_id(id1) == ks->chip->chip_id)) {
+			/* KS8995MA */
+			ks->revision_id = get_chip_rev(id1);
+		} else if (get_chip_id(id1) != CHIPID_M) {
+			/* KSZ8864RMN */
+			err = ks8995_read_reg(ks, KS8995_REG_ID1, &ksz8864_id);
+			if (err) {
+				err = -EIO;
+				goto err_out;
+			}
+
+			if ((ksz8864_id & 0x80) &&
+			    (ks->chip->chip_id == KSZ8864_CHIP_ID)) {
+				ks->revision_id = get_chip_rev(id1);
+			}
+
+		} else {
+			dev_err(&ks->spi->dev, "unsupported chip id for KS8995 family: 0x%02x\n",
+				id1);
+			err = -ENODEV;
+		}
+		break;
+	default:
+		dev_err(&ks->spi->dev, "unsupported family id: 0x%02x\n", id0);
+		err = -ENODEV;
+		break;
+	}
+err_out:
+	return err;
+}
+
 static const struct bin_attribute ks8995_registers_attr = {
 	.attr = {
 		.name   = "registers",
@@ -278,7 +354,6 @@ static int ks8995_probe(struct spi_device *spi)
 {
 	struct ks8995_switch    *ks;
 	struct ks8995_pdata     *pdata;
-	u8      ids[2];
 	int     err;
 	int variant = spi_get_device_id(spi)->driver_data;
 
@@ -309,39 +384,12 @@ static int ks8995_probe(struct spi_device *spi)
 		return err;
 	}
 
-	err = ks8995_read(ks, ids, KS8995_REG_ID0, sizeof(ids));
-	if (err < 0) {
-		dev_err(&spi->dev, "unable to read id registers, err=%d\n",
-				err);
+	err = ks8995_get_revision(ks);
+	if (err)
 		return err;
-	}
-
-	switch (ids[0]) {
-	case FAMILY_KS8995:
-		break;
-	default:
-		dev_err(&spi->dev, "unknown family id:%02x\n", ids[0]);
-		return -ENODEV;
-	}
 
 	ks->regs_attr.size = ks->chip->regs_size;
 	memcpy(&ks->regs_attr, &ks8995_registers_attr, sizeof(ks->regs_attr));
-	if (get_chip_id(ids[1]) != CHIPID_M) {
-		u8 val;
-
-		/* Check if this is a KSZ8864RMN */
-		err = ks8995_read(ks, &val, KSZ8864_REG_ID1, sizeof(val));
-		if (err < 0) {
-			dev_err(&spi->dev,
-				"unable to read chip id register, err=%d\n",
-				err);
-			return err;
-		}
-		if ((val & 0x80) == 0) {
-			dev_err(&spi->dev, "unknown chip:%02x,0\n", ids[1]);
-			return err;
-		}
-	}
 
 	err = ks8995_reset(ks);
 	if (err)
@@ -354,14 +402,8 @@ static int ks8995_probe(struct spi_device *spi)
 		return err;
 	}
 
-	if (get_chip_id(ids[1]) == CHIPID_M) {
-		dev_info(&spi->dev,
-			 "KS8995 device found, Chip ID:%x, Revision:%x\n",
-			 get_chip_id(ids[1]), get_chip_rev(ids[1]));
-	} else {
-		dev_info(&spi->dev, "KSZ8864 device found, Revision:%x\n",
-			 get_chip_rev(ids[1]));
-	}
+	dev_info(&spi->dev, "%s device found, Chip ID:%x, Revision:%x\n",
+		 ks->chip->name, ks->chip->chip_id, ks->revision_id);
 
 	return 0;
 }
-- 
cgit v0.10.2


From cd6f288cbaab656cebd524c5ef2388c11378c827 Mon Sep 17 00:00:00 2001
From: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Date: Tue, 9 Feb 2016 20:47:15 +0100
Subject: net: phy: spi_ks8995: add support for resetting switch using GPIO

When using device tree it is no more possible to reset the PHY at board
level. Furthermore, doing in the driver allows to power down the switch
when it is not used any more.

The patch introduces a new optional property "reset-gpios" denoting an
appropriate GPIO handle, e.g.:

reset-gpios = <&gpio0 46 GPIO_ACTIVE_LOW>

Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index 5e66d4b..8258c16 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -18,6 +18,9 @@
 #include <linux/module.h>
 #include <linux/delay.h>
 #include <linux/device.h>
+#include <linux/of.h>
+#include <linux/gpio.h>
+#include <linux/of_gpio.h>
 
 #include <linux/spi/spi.h>
 
@@ -120,7 +123,8 @@ static const struct ks8995_chip_params ks8995_chip[] = {
 };
 
 struct ks8995_pdata {
-	/* not yet implemented */
+	int reset_gpio;
+	enum of_gpio_flags reset_gpio_flags;
 };
 
 struct ks8995_switch {
@@ -339,6 +343,24 @@ err_out:
 	return err;
 }
 
+/* ks8995_parse_dt - setup platform data from devicetree
+ * @ks: pointer to switch instance
+ *
+ * Parses supported DT properties and sets up platform data
+ * accordingly.
+ */
+static void ks8995_parse_dt(struct ks8995_switch *ks)
+{
+	struct device_node *np = ks->spi->dev.of_node;
+	struct ks8995_pdata *pdata = ks->pdata;
+
+	if (!np)
+		return;
+
+	pdata->reset_gpio = of_get_named_gpio_flags(np, "reset-gpios", 0,
+		&pdata->reset_gpio_flags);
+}
+
 static const struct bin_attribute ks8995_registers_attr = {
 	.attr = {
 		.name   = "registers",
@@ -352,14 +374,10 @@ static const struct bin_attribute ks8995_registers_attr = {
 /* ------------------------------------------------------------------------ */
 static int ks8995_probe(struct spi_device *spi)
 {
-	struct ks8995_switch    *ks;
-	struct ks8995_pdata     *pdata;
-	int     err;
+	struct ks8995_switch *ks;
+	int err;
 	int variant = spi_get_device_id(spi)->driver_data;
 
-	/* Chip description */
-	pdata = spi->dev.platform_data;
-
 	if (variant >= max_variant) {
 		dev_err(&spi->dev, "bad chip variant %d\n", variant);
 		return -ENODEV;
@@ -370,10 +388,42 @@ static int ks8995_probe(struct spi_device *spi)
 		return -ENOMEM;
 
 	mutex_init(&ks->lock);
-	ks->pdata = pdata;
 	ks->spi = spi_dev_get(spi);
 	ks->chip = &ks8995_chip[variant];
 
+	if (ks->spi->dev.of_node) {
+		ks->pdata = devm_kzalloc(&spi->dev, sizeof(*ks->pdata),
+					 GFP_KERNEL);
+		if (!ks->pdata)
+			return -ENOMEM;
+
+		ks->pdata->reset_gpio = -1;
+
+		ks8995_parse_dt(ks);
+	}
+
+	if (!ks->pdata)
+		ks->pdata = spi->dev.platform_data;
+
+	/* de-assert switch reset */
+	if (ks->pdata && gpio_is_valid(ks->pdata->reset_gpio)) {
+		unsigned long flags;
+
+		flags = (ks->pdata->reset_gpio_flags == OF_GPIO_ACTIVE_LOW ?
+			 GPIOF_ACTIVE_LOW : 0);
+
+		err = devm_gpio_request_one(&spi->dev,
+					    ks->pdata->reset_gpio,
+					    flags, "switch-reset");
+		if (err) {
+			dev_err(&spi->dev,
+				"failed to get reset-gpios: %d\n", err);
+			return -EIO;
+		}
+
+		gpiod_set_value(gpio_to_desc(ks->pdata->reset_gpio), 0);
+	}
+
 	spi_set_drvdata(spi, ks);
 
 	spi->mode = SPI_MODE_0;
@@ -414,11 +464,14 @@ static int ks8995_remove(struct spi_device *spi)
 
 	sysfs_remove_bin_file(&spi->dev.kobj, &ks->regs_attr);
 
+	/* assert reset */
+	if (ks->pdata && gpio_is_valid(ks->pdata->reset_gpio))
+		gpiod_set_value(gpio_to_desc(ks->pdata->reset_gpio), 1);
+
 	return 0;
 }
 
 /* ------------------------------------------------------------------------ */
-
 static struct spi_driver ks8995_driver = {
 	.driver = {
 		.name	    = "spi-ks8995",
-- 
cgit v0.10.2


From 6665e62387c64054493411d00c7b0a5a37af88a5 Mon Sep 17 00:00:00 2001
From: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Date: Tue, 9 Feb 2016 20:47:16 +0100
Subject: net: phy: spi_ks8995: generalize creation of SPI commands

Prepare creating SPI reads and writes for other switch families.
The KS8995 family uses the straight forward
	<8bit CMD><8bit ADDR>
sequence.
To be able to support KSZ8795 family, which uses
	<3bit CMD><12bit ADDR><1 bit TR>
make the SPI command creation chip variant dependent.

Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index 8258c16..9dcc5b4 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -105,6 +105,8 @@ struct ks8995_chip_params {
 	int family_id;
 	int chip_id;
 	int regs_size;
+	int addr_width;
+	int addr_shift;
 };
 
 static const struct ks8995_chip_params ks8995_chip[] = {
@@ -113,12 +115,16 @@ static const struct ks8995_chip_params ks8995_chip[] = {
 		.family_id = FAMILY_KS8995,
 		.chip_id = KS8995_CHIP_ID,
 		.regs_size = KS8995_REGS_SIZE,
+		.addr_width = 8,
+		.addr_shift = 0,
 	},
 	[ksz8864] = {
 		.name = "KSZ8864RMN",
 		.family_id = FAMILY_KS8995,
 		.chip_id = KSZ8864_CHIP_ID,
 		.regs_size = KSZ8864_REGS_SIZE,
+		.addr_width = 8,
+		.addr_shift = 0,
 	},
 };
 
@@ -153,20 +159,44 @@ static inline u8 get_chip_rev(u8 val)
 	return (val >> ID1_REVISION_S) & ID1_REVISION_M;
 }
 
+/* create_spi_cmd - create a chip specific SPI command header
+ * @ks: pointer to switch instance
+ * @cmd: SPI command for switch
+ * @address: register address for command
+ *
+ * Different chip families use different bit pattern to address the switches
+ * registers:
+ *
+ * KS8995: 8bit command + 8bit address
+ * KSZ8795: 3bit command + 12bit address + 1bit TR (?)
+ */
+static inline __be16 create_spi_cmd(struct ks8995_switch *ks, int cmd,
+				    unsigned address)
+{
+	u16 result = cmd;
+
+	/* make room for address (incl. address shift) */
+	result <<= ks->chip->addr_width + ks->chip->addr_shift;
+	/* add address */
+	result |= address << ks->chip->addr_shift;
+	/* SPI protocol needs big endian */
+	return cpu_to_be16(result);
+}
 /* ------------------------------------------------------------------------ */
 static int ks8995_read(struct ks8995_switch *ks, char *buf,
 		 unsigned offset, size_t count)
 {
-	u8 cmd[2];
+	__be16 cmd;
 	struct spi_transfer t[2];
 	struct spi_message m;
 	int err;
 
+	cmd = create_spi_cmd(ks, KS8995_CMD_READ, offset);
 	spi_message_init(&m);
 
 	memset(&t, 0, sizeof(t));
 
-	t[0].tx_buf = cmd;
+	t[0].tx_buf = &cmd;
 	t[0].len = sizeof(cmd);
 	spi_message_add_tail(&t[0], &m);
 
@@ -174,9 +204,6 @@ static int ks8995_read(struct ks8995_switch *ks, char *buf,
 	t[1].len = count;
 	spi_message_add_tail(&t[1], &m);
 
-	cmd[0] = KS8995_CMD_READ;
-	cmd[1] = offset;
-
 	mutex_lock(&ks->lock);
 	err = spi_sync(ks->spi, &m);
 	mutex_unlock(&ks->lock);
@@ -184,20 +211,20 @@ static int ks8995_read(struct ks8995_switch *ks, char *buf,
 	return err ? err : count;
 }
 
-
 static int ks8995_write(struct ks8995_switch *ks, char *buf,
 		 unsigned offset, size_t count)
 {
-	u8 cmd[2];
+	__be16 cmd;
 	struct spi_transfer t[2];
 	struct spi_message m;
 	int err;
 
+	cmd = create_spi_cmd(ks, KS8995_CMD_WRITE, offset);
 	spi_message_init(&m);
 
 	memset(&t, 0, sizeof(t));
 
-	t[0].tx_buf = cmd;
+	t[0].tx_buf = &cmd;
 	t[0].len = sizeof(cmd);
 	spi_message_add_tail(&t[0], &m);
 
@@ -205,9 +232,6 @@ static int ks8995_write(struct ks8995_switch *ks, char *buf,
 	t[1].len = count;
 	spi_message_add_tail(&t[1], &m);
 
-	cmd[0] = KS8995_CMD_WRITE;
-	cmd[1] = offset;
-
 	mutex_lock(&ks->lock);
 	err = spi_sync(ks->spi, &m);
 	mutex_unlock(&ks->lock);
-- 
cgit v0.10.2


From c0e6cb1fed5a7390ac84396f6f43e70fc2034ac6 Mon Sep 17 00:00:00 2001
From: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Date: Tue, 9 Feb 2016 20:47:17 +0100
Subject: net: phy: spi_ks8995: add support for MICREL KSZ8795CLX

Add support for MICREL KSZ8795CLX Integrated 5-Port, 10-/100-Managed
Ethernet Switch with Gigabit GMII/RGMII and MII/RMII interfaces.

Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index 9dcc5b4..5e7340f 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -77,6 +77,7 @@
 
 #define KS8995_REGS_SIZE	0x80
 #define KSZ8864_REGS_SIZE	0x100
+#define KSZ8795_REGS_SIZE	0x100
 
 #define ID1_CHIPID_M		0xf
 #define ID1_CHIPID_S		4
@@ -85,9 +86,11 @@
 #define ID1_START_SW		1	/* start the switch */
 
 #define FAMILY_KS8995		0x95
+#define FAMILY_KSZ8795		0x87
 #define CHIPID_M		0
 #define KS8995_CHIP_ID		0x00
 #define KSZ8864_CHIP_ID		0x01
+#define KSZ8795_CHIP_ID		0x09
 
 #define KS8995_CMD_WRITE	0x02U
 #define KS8995_CMD_READ		0x03U
@@ -97,6 +100,7 @@
 enum ks8995_chip_variant {
 	ks8995,
 	ksz8864,
+	ksz8795,
 	max_variant
 };
 
@@ -126,6 +130,14 @@ static const struct ks8995_chip_params ks8995_chip[] = {
 		.addr_width = 8,
 		.addr_shift = 0,
 	},
+	[ksz8795] = {
+		.name = "KSZ8795CLX",
+		.family_id = FAMILY_KSZ8795,
+		.chip_id = KSZ8795_CHIP_ID,
+		.regs_size = KSZ8795_REGS_SIZE,
+		.addr_width = 12,
+		.addr_shift = 1,
+	},
 };
 
 struct ks8995_pdata {
@@ -145,6 +157,7 @@ struct ks8995_switch {
 static const struct spi_device_id ks8995_id[] = {
 	{"ks8995", ks8995},
 	{"ksz8864", ksz8864},
+	{"ksz8795", ksz8795},
 	{ }
 };
 MODULE_DEVICE_TABLE(spi, ks8995_id);
@@ -358,6 +371,22 @@ static int ks8995_get_revision(struct ks8995_switch *ks)
 			err = -ENODEV;
 		}
 		break;
+	case FAMILY_KSZ8795:
+		/* try reading chip id at CHIP ID1 */
+		err = ks8995_read_reg(ks, KS8995_REG_ID1, &id1);
+		if (err) {
+			err = -EIO;
+			goto err_out;
+		}
+
+		if (get_chip_id(id1) == ks->chip->chip_id) {
+			ks->revision_id = get_chip_rev(id1);
+		} else {
+			dev_err(&ks->spi->dev, "unsupported chip id for KSZ8795 family: 0x%02x\n",
+				id1);
+			err = -ENODEV;
+		}
+		break;
 	default:
 		dev_err(&ks->spi->dev, "unsupported family id: 0x%02x\n", id0);
 		err = -ENODEV;
-- 
cgit v0.10.2


From 7e406d124c7935ee0238b957ea7e563dc1710f29 Mon Sep 17 00:00:00 2001
From: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Date: Tue, 9 Feb 2016 20:47:18 +0100
Subject: dt-bindings: net: ks8995: add bindings documentation for ks8995

Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/micrel-ks8995.txt b/Documentation/devicetree/bindings/net/micrel-ks8995.txt
new file mode 100644
index 0000000..281bc24
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/micrel-ks8995.txt
@@ -0,0 +1,20 @@
+Micrel KS8995 SPI controlled Ethernet Switch families
+
+Required properties (according to spi-bus.txt):
+- compatible: either "micrel,ks8995", "micrel,ksz8864" or "micrel,ksz8795"
+
+Optional properties:
+- reset-gpios : phandle of gpio that will be used to reset chip during probe
+
+Example:
+
+spi-master {
+	...
+	switch@0 {
+		compatible = "micrel,ksz8795";
+
+		reg = <0>;
+		spi-max-frequency = <50000000>;
+		reset-gpios = <&gpio0 46 GPIO_ACTIVE_LOW>;
+	};
+};
-- 
cgit v0.10.2


From 8431706b4d1e684efee4b9e91e655f1161e883a8 Mon Sep 17 00:00:00 2001
From: Ivan Vecera <ivecera@redhat.com>
Date: Thu, 11 Feb 2016 12:42:26 +0100
Subject: be2net: don't report EVB for older chipsets when SR-IOV is disabled

The EVB (virtual bridge) functionality should be disabled on older BE3
and Lancer chips if SR-IOV is disabled in the NIC's BIOS. This setting
is identified by the zero value of total VFs reported by the card.
The GET_HSW_CONFIG command cannot be used as it is not supported by
these older chipset's FW.

v2: added the comment

Cc: Sathya Perla <sathya.perla@broadcom.com>
Cc: Ajit Khaparde <ajit.khaparde@broadcom.com>
Cc: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Cc: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Cc: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Sathya Perla <sathya.perla@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 4ecd8c2..88f427c 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -4617,6 +4617,9 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
 
 	/* BE and Lancer chips support VEB mode only */
 	if (BEx_chip(adapter) || lancer_chip(adapter)) {
+		/* VEB is disabled in non-SR-IOV profiles on BE3/Lancer */
+		if (!pci_sriov_get_totalvfs(adapter->pdev))
+			return 0;
 		hsw_mode = PORT_FWD_TYPE_VEB;
 	} else {
 		status = be_cmd_get_hsw_config(adapter, NULL, 0,
-- 
cgit v0.10.2


From 2f67864b6d5b653ee4518ab0d20549283b49a527 Mon Sep 17 00:00:00 2001
From: "Andrew F. Davis" <afd@ti.com>
Date: Sun, 7 Feb 2016 11:47:17 -0600
Subject: net: phy: dp83848: Add macro for dp83848 compatible devices

Add a helper macro for defining dp83848 compatible phy devices.
Update copyright info.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c
index 5e14e62..4e78f54 100644
--- a/drivers/net/phy/dp83848.c
+++ b/drivers/net/phy/dp83848.c
@@ -1,7 +1,7 @@
 /*
  * Driver for the Texas Instruments DP83848 PHY
  *
- * Copyright (C) 2015 Texas Instruments Inc.
+ * Copyright (C) 2015-2016 Texas Instruments Incorporated - http://www.ti.com/
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -70,25 +70,28 @@ static struct mdio_device_id __maybe_unused dp83848_tbl[] = {
 };
 MODULE_DEVICE_TABLE(mdio, dp83848_tbl);
 
-static struct phy_driver dp83848_driver[] = {
-	{
-		.phy_id		= DP83848_PHY_ID,
-		.phy_id_mask	= 0xfffffff0,
-		.name		= "TI DP83848",
-		.features	= PHY_BASIC_FEATURES,
-		.flags		= PHY_HAS_INTERRUPT,
-
-		.soft_reset	= genphy_soft_reset,
-		.config_init	= genphy_config_init,
-		.suspend	= genphy_suspend,
-		.resume		= genphy_resume,
-		.config_aneg	= genphy_config_aneg,
-		.read_status	= genphy_read_status,
+#define DP83848_PHY_DRIVER(_id, _name)				\
+	{							\
+		.phy_id		= _id,				\
+		.phy_id_mask	= 0xfffffff0,			\
+		.name		= _name,			\
+		.features	= PHY_BASIC_FEATURES,		\
+		.flags		= PHY_HAS_INTERRUPT,		\
+								\
+		.soft_reset	= genphy_soft_reset,		\
+		.config_init	= genphy_config_init,		\
+		.suspend	= genphy_suspend,		\
+		.resume		= genphy_resume,		\
+		.config_aneg	= genphy_config_aneg,		\
+		.read_status	= genphy_read_status,		\
+								\
+		/* IRQ related */				\
+		.ack_interrupt	= dp83848_ack_interrupt,	\
+		.config_intr	= dp83848_config_intr,		\
+	}
 
-		/* IRQ related */
-		.ack_interrupt	= dp83848_ack_interrupt,
-		.config_intr	= dp83848_config_intr,
-	},
+static struct phy_driver dp83848_driver[] = {
+	DP83848_PHY_DRIVER(DP83848_PHY_ID, "TI DP83848 10/100 Mbps PHY"),
 };
 module_phy_driver(dp83848_driver);
 
-- 
cgit v0.10.2


From 68336293462c92ba420528dd293d54ee879b6e15 Mon Sep 17 00:00:00 2001
From: "Andrew F. Davis" <afd@ti.com>
Date: Sun, 7 Feb 2016 11:47:18 -0600
Subject: net: phy: dp83848: Add PHY ID for TI version of DP83848C

After acquiring National Semiconductor, TI appears to have
changed the Vendor Model Number for the DP83848C PHYs,
add this new ID to supported IDs.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c
index 4e78f54..d4686d5f 100644
--- a/drivers/net/phy/dp83848.c
+++ b/drivers/net/phy/dp83848.c
@@ -16,7 +16,8 @@
 #include <linux/module.h>
 #include <linux/phy.h>
 
-#define DP83848_PHY_ID			0x20005c90
+#define TI_DP83848C_PHY_ID		0x20005ca0
+#define NS_DP83848C_PHY_ID		0x20005c90
 
 /* Registers */
 #define DP83848_MICR			0x11
@@ -65,7 +66,8 @@ static int dp83848_config_intr(struct phy_device *phydev)
 }
 
 static struct mdio_device_id __maybe_unused dp83848_tbl[] = {
-	{ DP83848_PHY_ID, 0xfffffff0 },
+	{ TI_DP83848C_PHY_ID, 0xfffffff0 },
+	{ NS_DP83848C_PHY_ID, 0xfffffff0 },
 	{ }
 };
 MODULE_DEVICE_TABLE(mdio, dp83848_tbl);
@@ -91,7 +93,8 @@ MODULE_DEVICE_TABLE(mdio, dp83848_tbl);
 	}
 
 static struct phy_driver dp83848_driver[] = {
-	DP83848_PHY_DRIVER(DP83848_PHY_ID, "TI DP83848 10/100 Mbps PHY"),
+	DP83848_PHY_DRIVER(TI_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"),
+	DP83848_PHY_DRIVER(NS_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"),
 };
 module_phy_driver(dp83848_driver);
 
-- 
cgit v0.10.2


From cf13be5afd30363e5e6e1232e004bae99ee3c623 Mon Sep 17 00:00:00 2001
From: "Andrew F. Davis" <afd@ti.com>
Date: Sun, 7 Feb 2016 11:47:19 -0600
Subject: net: phy: dp83848: Reorganize code for readability and safety

Reorganize code by moving the desired interrupt mask definition
out of function. Also rearrange the enable/disable interrupt function
to prevent accidental over-writing of values in registers.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c
index d4686d5f..20d3b9d 100644
--- a/drivers/net/phy/dp83848.c
+++ b/drivers/net/phy/dp83848.c
@@ -37,6 +37,12 @@
 #define DP83848_MISR_ED_INT_EN		BIT(6) /* Energy detect */
 #define DP83848_MISR_LQM_INT_EN		BIT(7) /* Link Quality Monitor */
 
+#define DP83848_INT_EN_MASK		\
+	(DP83848_MISR_ANC_INT_EN |	\
+	 DP83848_MISR_DUP_INT_EN |	\
+	 DP83848_MISR_SPD_INT_EN |	\
+	 DP83848_MISR_LINK_INT_EN)
+
 static int dp83848_ack_interrupt(struct phy_device *phydev)
 {
 	int err = phy_read(phydev, DP83848_MISR);
@@ -46,23 +52,24 @@ static int dp83848_ack_interrupt(struct phy_device *phydev)
 
 static int dp83848_config_intr(struct phy_device *phydev)
 {
-	int err;
+	int control, ret;
+
+	control = phy_read(phydev, DP83848_MICR);
+	if (control < 0)
+		return control;
 
 	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
-		err = phy_write(phydev, DP83848_MICR,
-				DP83848_MICR_INT_OE |
-				DP83848_MICR_INTEN);
-		if (err < 0)
-			return err;
-
-		return phy_write(phydev, DP83848_MISR,
-				 DP83848_MISR_ANC_INT_EN |
-				 DP83848_MISR_DUP_INT_EN |
-				 DP83848_MISR_SPD_INT_EN |
-				 DP83848_MISR_LINK_INT_EN);
+		control |= DP83848_MICR_INT_OE;
+		control |= DP83848_MICR_INTEN;
+
+		ret = phy_write(phydev, DP83848_MISR, DP83848_INT_EN_MASK);
+		if (ret < 0)
+			return ret;
+	} else {
+		control &= ~DP83848_MICR_INTEN;
 	}
 
-	return phy_write(phydev, DP83848_MICR, 0x0);
+	return phy_write(phydev, DP83848_MICR, control);
 }
 
 static struct mdio_device_id __maybe_unused dp83848_tbl[] = {
-- 
cgit v0.10.2


From d1782f7b0cb853ec0a9b7b3e5f8a3252ed8a054e Mon Sep 17 00:00:00 2001
From: "Andrew F. Davis" <afd@ti.com>
Date: Sun, 7 Feb 2016 11:47:20 -0600
Subject: net: phy: dp83848: Add support for TI TLK10x Ethernet PHYs

The TI TLK10x Ethernet PHYs are similar in the interrupt relevant
registers and so are compatible with the DP83848x devices already
supported.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c
index 20d3b9d..f897989 100644
--- a/drivers/net/phy/dp83848.c
+++ b/drivers/net/phy/dp83848.c
@@ -18,6 +18,7 @@
 
 #define TI_DP83848C_PHY_ID		0x20005ca0
 #define NS_DP83848C_PHY_ID		0x20005c90
+#define TLK10X_PHY_ID			0x2000a210
 
 /* Registers */
 #define DP83848_MICR			0x11
@@ -75,6 +76,7 @@ static int dp83848_config_intr(struct phy_device *phydev)
 static struct mdio_device_id __maybe_unused dp83848_tbl[] = {
 	{ TI_DP83848C_PHY_ID, 0xfffffff0 },
 	{ NS_DP83848C_PHY_ID, 0xfffffff0 },
+	{ TLK10X_PHY_ID, 0xfffffff0 },
 	{ }
 };
 MODULE_DEVICE_TABLE(mdio, dp83848_tbl);
@@ -102,6 +104,7 @@ MODULE_DEVICE_TABLE(mdio, dp83848_tbl);
 static struct phy_driver dp83848_driver[] = {
 	DP83848_PHY_DRIVER(TI_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"),
 	DP83848_PHY_DRIVER(NS_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"),
+	DP83848_PHY_DRIVER(TLK10X_PHY_ID, "TI TLK10X 10/100 Mbps PHY"),
 };
 module_phy_driver(dp83848_driver);
 
-- 
cgit v0.10.2


From 5fed039351dd54b295cba7c9afe34ede99d9b2e9 Mon Sep 17 00:00:00 2001
From: "Andrew F. Davis" <afd@ti.com>
Date: Sun, 7 Feb 2016 11:47:21 -0600
Subject: net: phy: dp83848: Add comments for register definitions

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c
index f897989..556904f 100644
--- a/drivers/net/phy/dp83848.c
+++ b/drivers/net/phy/dp83848.c
@@ -21,8 +21,8 @@
 #define TLK10X_PHY_ID			0x2000a210
 
 /* Registers */
-#define DP83848_MICR			0x11
-#define DP83848_MISR			0x12
+#define DP83848_MICR			0x11 /* MII Interrupt Control Register */
+#define DP83848_MISR			0x12 /* MII Interrupt Status Register */
 
 /* MICR Register Fields */
 #define DP83848_MICR_INT_OE		BIT(0) /* Interrupt Output Enable */
-- 
cgit v0.10.2


From e02564ee334a7ae46b71fc18576391cb9455433e Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Sun, 7 Feb 2016 21:52:23 +0100
Subject: ethtool: make validate_speed accept all speeds between 0 and INT_MAX

Devices these days can have any speed and as was recently pointed out
any speed from 0 to INT_MAX is valid so adjust speed validation to
accept such values.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 4345f80..190aea0 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -1377,24 +1377,7 @@ enum ethtool_sfeatures_retval_bits {
 
 static inline int ethtool_validate_speed(__u32 speed)
 {
-	switch (speed) {
-	case SPEED_10:
-	case SPEED_100:
-	case SPEED_1000:
-	case SPEED_2500:
-	case SPEED_5000:
-	case SPEED_10000:
-	case SPEED_20000:
-	case SPEED_25000:
-	case SPEED_40000:
-	case SPEED_50000:
-	case SPEED_56000:
-	case SPEED_100000:
-	case SPEED_UNKNOWN:
-		return 1;
-	}
-
-	return 0;
+	return speed <= INT_MAX || speed == SPEED_UNKNOWN;
 }
 
 /* Duplex, half or full. */
-- 
cgit v0.10.2


From 0cf3ace9e7cb47e3173561a8fb2601a12d8f75d2 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Sun, 7 Feb 2016 21:52:24 +0100
Subject: virtio_net: validate ethtool port setting and explain the user
 validation

We should validate the port setting that we got from the user and check
if it's what we've set it to (PORT_OTHER), also add explanation that
ignoring advertising is good as long as we don't have autonegotiation.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c9fd52a..fb0eae4 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1386,11 +1386,13 @@ static bool virtnet_validate_ethtool_cmd(const struct ethtool_cmd *cmd)
 	struct ethtool_cmd diff1 = *cmd;
 	struct ethtool_cmd diff2 = {};
 
-	/* advertising and cmd are usually set, ignore port because we set it */
+	/* cmd is always set so we need to clear it, validate the port type
+	 * and also without autonegotiation we can ignore advertising
+	 */
 	ethtool_cmd_speed_set(&diff1, 0);
+	diff2.port = PORT_OTHER;
 	diff1.advertising = 0;
 	diff1.duplex = 0;
-	diff1.port = 0;
 	diff1.cmd = 0;
 
 	return !memcmp(&diff1, &diff2, sizeof(diff1));
-- 
cgit v0.10.2


From 795bb1c00dd338aa0d12f9a7f1f4776fb3160416 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Mon, 8 Feb 2016 13:14:59 +0100
Subject: net: bulk free infrastructure for NAPI context, use napi_consume_skb

Discovered that network stack were hitting the kmem_cache/SLUB
slowpath when freeing SKBs.  Doing bulk free with kmem_cache_free_bulk
can speedup this slowpath.

NAPI context is a bit special, lets take advantage of that for bulk
free'ing SKBs.

In NAPI context we are running in softirq, which gives us certain
protection.  A softirq can run on several CPUs at once.  BUT the
important part is a softirq will never preempt another softirq running
on the same CPU.  This gives us the opportunity to access per-cpu
variables in softirq context.

Extend napi_alloc_cache (before only contained page_frag_cache) to be
a struct with a small array based stack for holding SKBs.  Introduce a
SKB defer and flush API for accessing this.

Introduce napi_consume_skb() as replacement for e.g. dev_consume_skb_any()
when running in NAPI context.  A small trick to handle/detect if we
are called from netpoll is to see if budget is 0.  In that case, we
need to invoke dev_consume_skb_irq().

Joint work with Alexander Duyck.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index a8fc222..b56c010 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2404,6 +2404,9 @@ static inline struct sk_buff *napi_alloc_skb(struct napi_struct *napi,
 {
 	return __napi_alloc_skb(napi, length, GFP_ATOMIC);
 }
+void napi_consume_skb(struct sk_buff *skb, int budget);
+
+void __kfree_skb_flush(void);
 
 /**
  * __dev_alloc_pages - allocate page for network Rx
diff --git a/net/core/dev.c b/net/core/dev.c
index f128483..9b2c7a9 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5155,6 +5155,7 @@ static void net_rx_action(struct softirq_action *h)
 		}
 	}
 
+	__kfree_skb_flush();
 	local_irq_disable();
 
 	list_splice_tail_init(&sd->poll_list, &list);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index b0cce74..b64187b 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -347,8 +347,16 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size)
 }
 EXPORT_SYMBOL(build_skb);
 
+#define NAPI_SKB_CACHE_SIZE	64
+
+struct napi_alloc_cache {
+	struct page_frag_cache page;
+	size_t skb_count;
+	void *skb_cache[NAPI_SKB_CACHE_SIZE];
+};
+
 static DEFINE_PER_CPU(struct page_frag_cache, netdev_alloc_cache);
-static DEFINE_PER_CPU(struct page_frag_cache, napi_alloc_cache);
+static DEFINE_PER_CPU(struct napi_alloc_cache, napi_alloc_cache);
 
 static void *__netdev_alloc_frag(unsigned int fragsz, gfp_t gfp_mask)
 {
@@ -378,9 +386,9 @@ EXPORT_SYMBOL(netdev_alloc_frag);
 
 static void *__napi_alloc_frag(unsigned int fragsz, gfp_t gfp_mask)
 {
-	struct page_frag_cache *nc = this_cpu_ptr(&napi_alloc_cache);
+	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
 
-	return __alloc_page_frag(nc, fragsz, gfp_mask);
+	return __alloc_page_frag(&nc->page, fragsz, gfp_mask);
 }
 
 void *napi_alloc_frag(unsigned int fragsz)
@@ -474,7 +482,7 @@ EXPORT_SYMBOL(__netdev_alloc_skb);
 struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len,
 				 gfp_t gfp_mask)
 {
-	struct page_frag_cache *nc = this_cpu_ptr(&napi_alloc_cache);
+	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
 	struct sk_buff *skb;
 	void *data;
 
@@ -494,7 +502,7 @@ struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len,
 	if (sk_memalloc_socks())
 		gfp_mask |= __GFP_MEMALLOC;
 
-	data = __alloc_page_frag(nc, len, gfp_mask);
+	data = __alloc_page_frag(&nc->page, len, gfp_mask);
 	if (unlikely(!data))
 		return NULL;
 
@@ -505,7 +513,7 @@ struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len,
 	}
 
 	/* use OR instead of assignment to avoid clearing of bits in mask */
-	if (nc->pfmemalloc)
+	if (nc->page.pfmemalloc)
 		skb->pfmemalloc = 1;
 	skb->head_frag = 1;
 
@@ -747,6 +755,69 @@ void consume_skb(struct sk_buff *skb)
 }
 EXPORT_SYMBOL(consume_skb);
 
+void __kfree_skb_flush(void)
+{
+	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
+
+	/* flush skb_cache if containing objects */
+	if (nc->skb_count) {
+		kmem_cache_free_bulk(skbuff_head_cache, nc->skb_count,
+				     nc->skb_cache);
+		nc->skb_count = 0;
+	}
+}
+
+static void __kfree_skb_defer(struct sk_buff *skb)
+{
+	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
+
+	/* drop skb->head and call any destructors for packet */
+	skb_release_all(skb);
+
+	/* record skb to CPU local list */
+	nc->skb_cache[nc->skb_count++] = skb;
+
+#ifdef CONFIG_SLUB
+	/* SLUB writes into objects when freeing */
+	prefetchw(skb);
+#endif
+
+	/* flush skb_cache if it is filled */
+	if (unlikely(nc->skb_count == NAPI_SKB_CACHE_SIZE)) {
+		kmem_cache_free_bulk(skbuff_head_cache, NAPI_SKB_CACHE_SIZE,
+				     nc->skb_cache);
+		nc->skb_count = 0;
+	}
+}
+
+void napi_consume_skb(struct sk_buff *skb, int budget)
+{
+	if (unlikely(!skb))
+		return;
+
+	/* if budget is 0 assume netpoll w/ IRQs disabled */
+	if (unlikely(!budget)) {
+		dev_consume_skb_irq(skb);
+		return;
+	}
+
+	if (likely(atomic_read(&skb->users) == 1))
+		smp_rmb();
+	else if (likely(!atomic_dec_and_test(&skb->users)))
+		return;
+	/* if reaching here SKB is ready to free */
+	trace_consume_skb(skb);
+
+	/* if SKB is a clone, don't handle this case */
+	if (unlikely(skb->fclone != SKB_FCLONE_UNAVAILABLE)) {
+		__kfree_skb(skb);
+		return;
+	}
+
+	__kfree_skb_defer(skb);
+}
+EXPORT_SYMBOL(napi_consume_skb);
+
 /* Make sure a field is enclosed inside headers_start/headers_end section */
 #define CHECK_SKB_FIELD(field) \
 	BUILD_BUG_ON(offsetof(struct sk_buff, field) <		\
-- 
cgit v0.10.2


From 15fad714be86eab13e7568fecaf475b2a9730d3e Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Mon, 8 Feb 2016 13:15:04 +0100
Subject: net: bulk free SKBs that were delay free'ed due to IRQ context

The network stack defers SKBs free, in-case free happens in IRQ or
when IRQs are disabled. This happens in __dev_kfree_skb_irq() that
writes SKBs that were free'ed during IRQ to the softirq completion
queue (softnet_data.completion_queue).

These SKBs are naturally delayed, and cleaned up during NET_TX_SOFTIRQ
in function net_tx_action().  Take advantage of this a use the skb
defer and flush API, as we are already in softirq context.

For modern drivers this rarely happens. Although most drivers do call
dev_kfree_skb_any(), which detects the situation and calls
__dev_kfree_skb_irq() when needed.  This due to netpoll can call from
IRQ context.

Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index b56c010..6ec86f1 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2407,6 +2407,7 @@ static inline struct sk_buff *napi_alloc_skb(struct napi_struct *napi,
 void napi_consume_skb(struct sk_buff *skb, int budget);
 
 void __kfree_skb_flush(void);
+void __kfree_skb_defer(struct sk_buff *skb);
 
 /**
  * __dev_alloc_pages - allocate page for network Rx
diff --git a/net/core/dev.c b/net/core/dev.c
index 9b2c7a9..3f4071a 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3829,8 +3829,14 @@ static void net_tx_action(struct softirq_action *h)
 				trace_consume_skb(skb);
 			else
 				trace_kfree_skb(skb, net_tx_action);
-			__kfree_skb(skb);
+
+			if (skb->fclone != SKB_FCLONE_UNAVAILABLE)
+				__kfree_skb(skb);
+			else
+				__kfree_skb_defer(skb);
 		}
+
+		__kfree_skb_flush();
 	}
 
 	if (sd->output_queue) {
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index b64187b..a5bd067 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -767,7 +767,7 @@ void __kfree_skb_flush(void)
 	}
 }
 
-static void __kfree_skb_defer(struct sk_buff *skb)
+static inline void _kfree_skb_defer(struct sk_buff *skb)
 {
 	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
 
@@ -789,6 +789,10 @@ static void __kfree_skb_defer(struct sk_buff *skb)
 		nc->skb_count = 0;
 	}
 }
+void __kfree_skb_defer(struct sk_buff *skb)
+{
+	_kfree_skb_defer(skb);
+}
 
 void napi_consume_skb(struct sk_buff *skb, int budget)
 {
@@ -814,7 +818,7 @@ void napi_consume_skb(struct sk_buff *skb, int budget)
 		return;
 	}
 
-	__kfree_skb_defer(skb);
+	_kfree_skb_defer(skb);
 }
 EXPORT_SYMBOL(napi_consume_skb);
 
-- 
cgit v0.10.2


From a3a8749d34d8a5ac071c7ead792bd21ffe328aa0 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Mon, 8 Feb 2016 13:15:09 +0100
Subject: ixgbe: bulk free SKBs during TX completion cleanup cycle

There is an opportunity to bulk free SKBs during reclaiming of
resources after DMA transmit completes in ixgbe_clean_tx_irq.  Thus,
bulk freeing at this point does not introduce any added latency.

Simply use napi_consume_skb() which were recently introduced.  The
napi_budget parameter is needed by napi_consume_skb() to detect if it
is called from netpoll.

Benchmarking IPv4-forwarding, on CPU i7-4790K @4.2GHz (no turbo boost)
 Single CPU/flow numbers: before: 1982144 pps ->  after : 2064446 pps
 Improvement: +82302 pps, -20 nanosec, +4.1%
 (SLUB and GCC version 5.1.1 20150618 (Red Hat 5.1.1-4))

Joint work with Alexander Duyck.

Signed-off-by: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index c4003a8..0c701b8 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1089,7 +1089,7 @@ static void ixgbe_tx_timeout_reset(struct ixgbe_adapter *adapter)
  * @tx_ring: tx ring to clean
  **/
 static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
-			       struct ixgbe_ring *tx_ring)
+			       struct ixgbe_ring *tx_ring, int napi_budget)
 {
 	struct ixgbe_adapter *adapter = q_vector->adapter;
 	struct ixgbe_tx_buffer *tx_buffer;
@@ -1127,7 +1127,7 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 		total_packets += tx_buffer->gso_segs;
 
 		/* free the skb */
-		dev_consume_skb_any(tx_buffer->skb);
+		napi_consume_skb(tx_buffer->skb, napi_budget);
 
 		/* unmap skb header data */
 		dma_unmap_single(tx_ring->dev,
@@ -2784,7 +2784,7 @@ int ixgbe_poll(struct napi_struct *napi, int budget)
 #endif
 
 	ixgbe_for_each_ring(ring, q_vector->tx)
-		clean_complete &= !!ixgbe_clean_tx_irq(q_vector, ring);
+		clean_complete &= !!ixgbe_clean_tx_irq(q_vector, ring, budget);
 
 	/* Exit if we are called by netpoll or busy polling is active */
 	if ((budget <= 0) || !ixgbe_qv_lock_napi(q_vector))
-- 
cgit v0.10.2


From 1580ab63fc9a03593072cc5656167a75c4f1d173 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 11 Feb 2016 16:28:49 -0800
Subject: tcp/dccp: better use of ephemeral ports in connect()

In commit 07f4c90062f8 ("tcp/dccp: try to not exhaust ip_local_port_range
in connect()"), I added a very simple heuristic, so that we got better
chances to use even ports, and allow bind() users to have more available
slots.

It gave nice results, but with more than 200,000 TCP sessions on a typical
server, the ~30,000 ephemeral ports are still a rare resource.

I chose to go a step further, by looking at all even ports, and if none
was available, fallback to odd ports.

The companion patch does the same in bind(), but in opposite way.

I've seen exec times of up to 30ms on busy servers, so I no longer
disable BH for the whole traversal, but only for each hash bucket.
I also call cond_resched() to be gentle to other tasks.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index c0f9942..bc68ece 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -565,106 +565,106 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row,
 			struct sock *, __u16, struct inet_timewait_sock **))
 {
 	struct inet_hashinfo *hinfo = death_row->hashinfo;
-	const unsigned short snum = inet_sk(sk)->inet_num;
+	struct inet_timewait_sock *tw = NULL;
 	struct inet_bind_hashbucket *head;
-	struct inet_bind_bucket *tb;
-	int ret;
+	int port = inet_sk(sk)->inet_num;
 	struct net *net = sock_net(sk);
+	struct inet_bind_bucket *tb;
+	u32 remaining, offset;
+	int ret, i, low, high;
+	static u32 hint;
+
+	if (port) {
+		head = &hinfo->bhash[inet_bhashfn(net, port,
+						  hinfo->bhash_size)];
+		tb = inet_csk(sk)->icsk_bind_hash;
+		spin_lock_bh(&head->lock);
+		if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
+			inet_ehash_nolisten(sk, NULL);
+			spin_unlock_bh(&head->lock);
+			return 0;
+		}
+		spin_unlock(&head->lock);
+		/* No definite answer... Walk to established hash table */
+		ret = check_established(death_row, sk, port, NULL);
+		local_bh_enable();
+		return ret;
+	}
 
-	if (!snum) {
-		int i, remaining, low, high, port;
-		static u32 hint;
-		u32 offset = hint + port_offset;
-		struct inet_timewait_sock *tw = NULL;
+	inet_get_local_port_range(net, &low, &high);
+	high++; /* [32768, 60999] -> [32768, 61000[ */
+	remaining = high - low;
+	if (likely(remaining > 1))
+		remaining &= ~1U;
 
-		inet_get_local_port_range(net, &low, &high);
-		remaining = (high - low) + 1;
+	offset = (hint + port_offset) % remaining;
+	/* In first pass we try ports of @low parity.
+	 * inet_csk_get_port() does the opposite choice.
+	 */
+	offset &= ~1U;
+other_parity_scan:
+	port = low + offset;
+	for (i = 0; i < remaining; i += 2, port += 2) {
+		if (unlikely(port >= high))
+			port -= remaining;
+		if (inet_is_local_reserved_port(net, port))
+			continue;
+		head = &hinfo->bhash[inet_bhashfn(net, port,
+						  hinfo->bhash_size)];
+		spin_lock_bh(&head->lock);
 
-		/* By starting with offset being an even number,
-		 * we tend to leave about 50% of ports for other uses,
-		 * like bind(0).
+		/* Does not bother with rcv_saddr checks, because
+		 * the established check is already unique enough.
 		 */
-		offset &= ~1;
-
-		local_bh_disable();
-		for (i = 0; i < remaining; i++) {
-			port = low + (i + offset) % remaining;
-			if (inet_is_local_reserved_port(net, port))
-				continue;
-			head = &hinfo->bhash[inet_bhashfn(net, port,
-					hinfo->bhash_size)];
-			spin_lock(&head->lock);
-
-			/* Does not bother with rcv_saddr checks,
-			 * because the established check is already
-			 * unique enough.
-			 */
-			inet_bind_bucket_for_each(tb, &head->chain) {
-				if (net_eq(ib_net(tb), net) &&
-				    tb->port == port) {
-					if (tb->fastreuse >= 0 ||
-					    tb->fastreuseport >= 0)
-						goto next_port;
-					WARN_ON(hlist_empty(&tb->owners));
-					if (!check_established(death_row, sk,
-								port, &tw))
-						goto ok;
+		inet_bind_bucket_for_each(tb, &head->chain) {
+			if (net_eq(ib_net(tb), net) && tb->port == port) {
+				if (tb->fastreuse >= 0 ||
+				    tb->fastreuseport >= 0)
 					goto next_port;
-				}
+				WARN_ON(hlist_empty(&tb->owners));
+				if (!check_established(death_row, sk,
+						       port, &tw))
+					goto ok;
+				goto next_port;
 			}
-
-			tb = inet_bind_bucket_create(hinfo->bind_bucket_cachep,
-					net, head, port);
-			if (!tb) {
-				spin_unlock(&head->lock);
-				break;
-			}
-			tb->fastreuse = -1;
-			tb->fastreuseport = -1;
-			goto ok;
-
-		next_port:
-			spin_unlock(&head->lock);
 		}
-		local_bh_enable();
-
-		return -EADDRNOTAVAIL;
 
-ok:
-		hint += (i + 2) & ~1;
-
-		/* Head lock still held and bh's disabled */
-		inet_bind_hash(sk, tb, port);
-		if (sk_unhashed(sk)) {
-			inet_sk(sk)->inet_sport = htons(port);
-			inet_ehash_nolisten(sk, (struct sock *)tw);
+		tb = inet_bind_bucket_create(hinfo->bind_bucket_cachep,
+					     net, head, port);
+		if (!tb) {
+			spin_unlock_bh(&head->lock);
+			return -ENOMEM;
 		}
-		if (tw)
-			inet_twsk_bind_unhash(tw, hinfo);
-		spin_unlock(&head->lock);
+		tb->fastreuse = -1;
+		tb->fastreuseport = -1;
+		goto ok;
+next_port:
+		spin_unlock_bh(&head->lock);
+		cond_resched();
+	}
 
-		if (tw)
-			inet_twsk_deschedule_put(tw);
+	offset++;
+	if ((offset & 1) && remaining > 1)
+		goto other_parity_scan;
 
-		ret = 0;
-		goto out;
-	}
+	return -EADDRNOTAVAIL;
 
-	head = &hinfo->bhash[inet_bhashfn(net, snum, hinfo->bhash_size)];
-	tb  = inet_csk(sk)->icsk_bind_hash;
-	spin_lock_bh(&head->lock);
-	if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
-		inet_ehash_nolisten(sk, NULL);
-		spin_unlock_bh(&head->lock);
-		return 0;
-	} else {
-		spin_unlock(&head->lock);
-		/* No definite answer... Walk to established hash table */
-		ret = check_established(death_row, sk, snum, NULL);
-out:
-		local_bh_enable();
-		return ret;
+ok:
+	hint += i + 2;
+
+	/* Head lock still held and bh's disabled */
+	inet_bind_hash(sk, tb, port);
+	if (sk_unhashed(sk)) {
+		inet_sk(sk)->inet_sport = htons(port);
+		inet_ehash_nolisten(sk, (struct sock *)tw);
 	}
+	if (tw)
+		inet_twsk_bind_unhash(tw, hinfo);
+	spin_unlock(&head->lock);
+	if (tw)
+		inet_twsk_deschedule_put(tw);
+	local_bh_enable();
+	return 0;
 }
 
 /*
-- 
cgit v0.10.2


From ea8add2b190395408b22a9127bed2c0912aecbc8 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 11 Feb 2016 16:28:50 -0800
Subject: tcp/dccp: better use of ephemeral ports in bind()

Implement strategy used in __inet_hash_connect() in opposite way :

Try to find a candidate using odd ports, then fallback to even ports.

We no longer disable BH for whole traversal, but one bucket at a time.
We also use cond_resched() to yield cpu to other tasks if needed.

I removed one indentation level and tried to mirror the loop we have
in __inet_hash_connect() and variable names to ease code maintenance.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index c16a2e6..3d28c6d 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -91,165 +91,153 @@ EXPORT_SYMBOL_GPL(inet_csk_bind_conflict);
 
 /* Obtain a reference to a local port for the given sock,
  * if snum is zero it means select any available local port.
+ * We try to allocate an odd port (and leave even ports for connect())
  */
 int inet_csk_get_port(struct sock *sk, unsigned short snum)
 {
-	struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo;
+	bool reuse = sk->sk_reuse && sk->sk_state != TCP_LISTEN;
+	struct inet_hashinfo *hinfo = sk->sk_prot->h.hashinfo;
+	int ret = 1, attempts = 5, port = snum;
+	int smallest_size = -1, smallest_port;
 	struct inet_bind_hashbucket *head;
-	struct inet_bind_bucket *tb;
-	int ret, attempts = 5;
 	struct net *net = sock_net(sk);
-	int smallest_size = -1, smallest_rover;
+	int i, low, high, attempt_half;
+	struct inet_bind_bucket *tb;
 	kuid_t uid = sock_i_uid(sk);
-	int attempt_half = (sk->sk_reuse == SK_CAN_REUSE) ? 1 : 0;
+	u32 remaining, offset;
 
-	local_bh_disable();
-	if (!snum) {
-		int remaining, rover, low, high;
+	if (port) {
+have_port:
+		head = &hinfo->bhash[inet_bhashfn(net, port,
+						  hinfo->bhash_size)];
+		spin_lock_bh(&head->lock);
+		inet_bind_bucket_for_each(tb, &head->chain)
+			if (net_eq(ib_net(tb), net) && tb->port == port)
+				goto tb_found;
 
+		goto tb_not_found;
+	}
 again:
-		inet_get_local_port_range(net, &low, &high);
-		if (attempt_half) {
-			int half = low + ((high - low) >> 1);
-
-			if (attempt_half == 1)
-				high = half;
-			else
-				low = half;
-		}
-		remaining = (high - low) + 1;
-		smallest_rover = rover = prandom_u32() % remaining + low;
-
-		smallest_size = -1;
-		do {
-			if (inet_is_local_reserved_port(net, rover))
-				goto next_nolock;
-			head = &hashinfo->bhash[inet_bhashfn(net, rover,
-					hashinfo->bhash_size)];
-			spin_lock(&head->lock);
-			inet_bind_bucket_for_each(tb, &head->chain)
-				if (net_eq(ib_net(tb), net) && tb->port == rover) {
-					if (((tb->fastreuse > 0 &&
-					      sk->sk_reuse &&
-					      sk->sk_state != TCP_LISTEN) ||
-					     (tb->fastreuseport > 0 &&
-					      sk->sk_reuseport &&
-					      !rcu_access_pointer(sk->sk_reuseport_cb) &&
-					      uid_eq(tb->fastuid, uid))) &&
-					    (tb->num_owners < smallest_size || smallest_size == -1)) {
-						smallest_size = tb->num_owners;
-						smallest_rover = rover;
-					}
-					if (!inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, false)) {
-						snum = rover;
-						goto tb_found;
-					}
-					goto next;
+	attempt_half = (sk->sk_reuse == SK_CAN_REUSE) ? 1 : 0;
+other_half_scan:
+	inet_get_local_port_range(net, &low, &high);
+	high++; /* [32768, 60999] -> [32768, 61000[ */
+	if (high - low < 4)
+		attempt_half = 0;
+	if (attempt_half) {
+		int half = low + (((high - low) >> 2) << 1);
+
+		if (attempt_half == 1)
+			high = half;
+		else
+			low = half;
+	}
+	remaining = high - low;
+	if (likely(remaining > 1))
+		remaining &= ~1U;
+
+	offset = prandom_u32() % remaining;
+	/* __inet_hash_connect() favors ports having @low parity
+	 * We do the opposite to not pollute connect() users.
+	 */
+	offset |= 1U;
+	smallest_size = -1;
+	smallest_port = low; /* avoid compiler warning */
+
+other_parity_scan:
+	port = low + offset;
+	for (i = 0; i < remaining; i += 2, port += 2) {
+		if (unlikely(port >= high))
+			port -= remaining;
+		if (inet_is_local_reserved_port(net, port))
+			continue;
+		head = &hinfo->bhash[inet_bhashfn(net, port,
+						  hinfo->bhash_size)];
+		spin_lock_bh(&head->lock);
+		inet_bind_bucket_for_each(tb, &head->chain)
+			if (net_eq(ib_net(tb), net) && tb->port == port) {
+				if (((tb->fastreuse > 0 && reuse) ||
+				     (tb->fastreuseport > 0 &&
+				      sk->sk_reuseport &&
+				      !rcu_access_pointer(sk->sk_reuseport_cb) &&
+				      uid_eq(tb->fastuid, uid))) &&
+				    (tb->num_owners < smallest_size || smallest_size == -1)) {
+					smallest_size = tb->num_owners;
+					smallest_port = port;
 				}
-			break;
-		next:
-			spin_unlock(&head->lock);
-		next_nolock:
-			if (++rover > high)
-				rover = low;
-		} while (--remaining > 0);
-
-		/* Exhausted local port range during search?  It is not
-		 * possible for us to be holding one of the bind hash
-		 * locks if this test triggers, because if 'remaining'
-		 * drops to zero, we broke out of the do/while loop at
-		 * the top level, not from the 'break;' statement.
-		 */
-		ret = 1;
-		if (remaining <= 0) {
-			if (smallest_size != -1) {
-				snum = smallest_rover;
-				goto have_snum;
+				if (!inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, false))
+					goto tb_found;
+				goto next_port;
 			}
-			if (attempt_half == 1) {
-				/* OK we now try the upper half of the range */
-				attempt_half = 2;
-				goto again;
-			}
-			goto fail;
-		}
-		/* OK, here is the one we will use.  HEAD is
-		 * non-NULL and we hold it's mutex.
-		 */
-		snum = rover;
-	} else {
-have_snum:
-		head = &hashinfo->bhash[inet_bhashfn(net, snum,
-				hashinfo->bhash_size)];
-		spin_lock(&head->lock);
-		inet_bind_bucket_for_each(tb, &head->chain)
-			if (net_eq(ib_net(tb), net) && tb->port == snum)
-				goto tb_found;
+		goto tb_not_found;
+next_port:
+		spin_unlock_bh(&head->lock);
+		cond_resched();
 	}
-	tb = NULL;
-	goto tb_not_found;
+
+	if (smallest_size != -1) {
+		port = smallest_port;
+		goto have_port;
+	}
+	offset--;
+	if (!(offset & 1))
+		goto other_parity_scan;
+
+	if (attempt_half == 1) {
+		/* OK we now try the upper half of the range */
+		attempt_half = 2;
+		goto other_half_scan;
+	}
+	return ret;
+
+tb_not_found:
+	tb = inet_bind_bucket_create(hinfo->bind_bucket_cachep,
+				     net, head, port);
+	if (!tb)
+		goto fail_unlock;
 tb_found:
 	if (!hlist_empty(&tb->owners)) {
 		if (sk->sk_reuse == SK_FORCE_REUSE)
 			goto success;
 
-		if (((tb->fastreuse > 0 &&
-		      sk->sk_reuse && sk->sk_state != TCP_LISTEN) ||
+		if (((tb->fastreuse > 0 && reuse) ||
 		     (tb->fastreuseport > 0 &&
-		      sk->sk_reuseport &&
-		      !rcu_access_pointer(sk->sk_reuseport_cb) &&
-		      uid_eq(tb->fastuid, uid))) && smallest_size == -1) {
+		      sk->sk_reuseport && uid_eq(tb->fastuid, uid))) &&
+		    smallest_size == -1)
 			goto success;
-		} else {
-			ret = 1;
-			if (inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, true)) {
-				if (((sk->sk_reuse && sk->sk_state != TCP_LISTEN) ||
-				     (tb->fastreuseport > 0 &&
-				      sk->sk_reuseport &&
-				      !rcu_access_pointer(sk->sk_reuseport_cb) &&
-				      uid_eq(tb->fastuid, uid))) &&
-				    smallest_size != -1 && --attempts >= 0) {
-					spin_unlock(&head->lock);
-					goto again;
-				}
-
-				goto fail_unlock;
+		if (inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, true)) {
+			if ((reuse ||
+			     (tb->fastreuseport > 0 &&
+			      sk->sk_reuseport &&
+			      !rcu_access_pointer(sk->sk_reuseport_cb) &&
+			      uid_eq(tb->fastuid, uid))) &&
+			    smallest_size != -1 && --attempts >= 0) {
+				spin_unlock_bh(&head->lock);
+				goto again;
 			}
+			goto fail_unlock;
 		}
-	}
-tb_not_found:
-	ret = 1;
-	if (!tb && (tb = inet_bind_bucket_create(hashinfo->bind_bucket_cachep,
-					net, head, snum)) == NULL)
-		goto fail_unlock;
-	if (hlist_empty(&tb->owners)) {
-		if (sk->sk_reuse && sk->sk_state != TCP_LISTEN)
-			tb->fastreuse = 1;
-		else
+		if (!reuse)
 			tb->fastreuse = 0;
+		if (!sk->sk_reuseport || !uid_eq(tb->fastuid, uid))
+			tb->fastreuseport = 0;
+	} else {
+		tb->fastreuse = reuse;
 		if (sk->sk_reuseport) {
 			tb->fastreuseport = 1;
 			tb->fastuid = uid;
-		} else
-			tb->fastreuseport = 0;
-	} else {
-		if (tb->fastreuse &&
-		    (!sk->sk_reuse || sk->sk_state == TCP_LISTEN))
-			tb->fastreuse = 0;
-		if (tb->fastreuseport &&
-		    (!sk->sk_reuseport || !uid_eq(tb->fastuid, uid)))
+		} else {
 			tb->fastreuseport = 0;
+		}
 	}
 success:
 	if (!inet_csk(sk)->icsk_bind_hash)
-		inet_bind_hash(sk, tb, snum);
+		inet_bind_hash(sk, tb, port);
 	WARN_ON(inet_csk(sk)->icsk_bind_hash != tb);
 	ret = 0;
 
 fail_unlock:
-	spin_unlock(&head->lock);
-fail:
-	local_bh_enable();
+	spin_unlock_bh(&head->lock);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(inet_csk_get_port);
-- 
cgit v0.10.2


From 179bc67f69b6cb53ad68cfdec5a917c2a2248355 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 20:48:04 +0000
Subject: net: local checksum offload for encapsulation

The arithmetic properties of the ones-complement checksum mean that a
 correctly checksummed inner packet, including its checksum, has a ones
 complement sum depending only on whatever value was used to initialise
 the checksum field before checksumming (in the case of TCP and UDP,
 this is the ones complement sum of the pseudo header, complemented).
Consequently, if we are going to offload the inner checksum with
 CHECKSUM_PARTIAL, we can compute the outer checksum based only on the
 packed data not covered by the inner checksum, and the initial value of
 the inner checksum field.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 6ec86f1..cf906d1 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3702,5 +3702,29 @@ static inline unsigned int skb_gso_network_seglen(const struct sk_buff *skb)
 	return hdr_len + skb_gso_transport_seglen(skb);
 }
 
+/* Local Checksum Offload.
+ * Compute outer checksum based on the assumption that the
+ * inner checksum will be offloaded later.
+ * Fill in outer checksum adjustment (e.g. with sum of outer
+ * pseudo-header) before calling.
+ * Also ensure that inner checksum is in linear data area.
+ */
+static inline __wsum lco_csum(struct sk_buff *skb)
+{
+	char *inner_csum_field;
+	__wsum csum;
+
+	/* Start with complement of inner checksum adjustment */
+	inner_csum_field = skb->data + skb_checksum_start_offset(skb) +
+				skb->csum_offset;
+	csum = ~csum_unfold(*(__force __sum16 *)inner_csum_field);
+	/* Add in checksum of our headers (incl. outer checksum
+	 * adjustment filled in by caller)
+	 */
+	csum = skb_checksum(skb, 0, skb_checksum_start_offset(skb), csum);
+	/* The result is the checksum from skb->data to end of packet */
+	return csum;
+}
+
 #endif	/* __KERNEL__ */
 #endif	/* _LINUX_SKBUFF_H */
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index 859d415..d74ce93 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -166,20 +166,20 @@ struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb,
 		return skb;
 	}
 
-	/* If packet is not gso and we are resolving any partial checksum,
+	/* If packet is not gso and we are not offloading inner checksum,
 	 * clear encapsulation flag. This allows setting CHECKSUM_PARTIAL
 	 * on the outer header without confusing devices that implement
 	 * NETIF_F_IP_CSUM with encapsulation.
 	 */
-	if (csum_help)
-		skb->encapsulation = 0;
-
 	if (skb->ip_summed == CHECKSUM_PARTIAL && csum_help) {
+		skb->encapsulation = 0;
 		err = skb_checksum_help(skb);
 		if (unlikely(err))
 			goto error;
-	} else if (skb->ip_summed != CHECKSUM_PARTIAL)
+	} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
 		skb->ip_summed = CHECKSUM_NONE;
+		skb->encapsulation = 0;
+	}
 
 	return skb;
 error:
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index ac3cedb..a59341c 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -848,16 +848,18 @@ void udp_set_csum(bool nocheck, struct sk_buff *skb,
 {
 	struct udphdr *uh = udp_hdr(skb);
 
-	if (nocheck)
+	if (nocheck) {
 		uh->check = 0;
-	else if (skb_is_gso(skb))
+	} else if (skb_is_gso(skb)) {
 		uh->check = ~udp_v4_check(len, saddr, daddr, 0);
-	else if (skb_dst(skb) && skb_dst(skb)->dev &&
-		 (skb_dst(skb)->dev->features &
-		  (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) {
-
-		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
-
+	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		uh->check = 0;
+		uh->check = udp_v4_check(len, saddr, daddr, lco_csum(skb));
+		if (uh->check == 0)
+			uh->check = CSUM_MANGLED_0;
+	} else if (skb_dst(skb) && skb_dst(skb)->dev &&
+		   (skb_dst(skb)->dev->features &
+		    (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) {
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum_start = skb_transport_header(skb) - skb->head;
 		skb->csum_offset = offsetof(struct udphdr, check);
@@ -865,8 +867,6 @@ void udp_set_csum(bool nocheck, struct sk_buff *skb,
 	} else {
 		__wsum csum;
 
-		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
-
 		uh->check = 0;
 		csum = skb_checksum(skb, 0, len, 0);
 		uh->check = udp_v4_check(len, saddr, daddr, csum);
diff --git a/net/ipv6/ip6_checksum.c b/net/ipv6/ip6_checksum.c
index 9a4d732..4924bd7 100644
--- a/net/ipv6/ip6_checksum.c
+++ b/net/ipv6/ip6_checksum.c
@@ -98,11 +98,13 @@ void udp6_set_csum(bool nocheck, struct sk_buff *skb,
 		uh->check = 0;
 	else if (skb_is_gso(skb))
 		uh->check = ~udp_v6_check(len, saddr, daddr, 0);
-	else if (skb_dst(skb) && skb_dst(skb)->dev &&
-		 (skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) {
-
-		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
-
+	else if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		uh->check = 0;
+		uh->check = udp_v6_check(len, saddr, daddr, lco_csum(skb));
+		if (uh->check == 0)
+			uh->check = CSUM_MANGLED_0;
+	} else if (skb_dst(skb) && skb_dst(skb)->dev &&
+		   (skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) {
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum_start = skb_transport_header(skb) - skb->head;
 		skb->csum_offset = offsetof(struct udphdr, check);
@@ -110,8 +112,6 @@ void udp6_set_csum(bool nocheck, struct sk_buff *skb,
 	} else {
 		__wsum csum;
 
-		BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL);
-
 		uh->check = 0;
 		csum = skb_checksum(skb, 0, len, 0);
 		uh->check = udp_v6_check(len, saddr, daddr, csum);
-- 
cgit v0.10.2


From d75f1306d9464d535007f05e5da0afcf9e3916d9 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 20:49:40 +0000
Subject: net: udp: always set up for CHECKSUM_PARTIAL offload

If the dst device doesn't support it, it'll get fixed up later anyway
 by validate_xmit_skb().  Also, this allows us to take advantage of LCO
 to avoid summing the payload multiple times.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index a59341c..9fc4e9c 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -857,23 +857,11 @@ void udp_set_csum(bool nocheck, struct sk_buff *skb,
 		uh->check = udp_v4_check(len, saddr, daddr, lco_csum(skb));
 		if (uh->check == 0)
 			uh->check = CSUM_MANGLED_0;
-	} else if (skb_dst(skb) && skb_dst(skb)->dev &&
-		   (skb_dst(skb)->dev->features &
-		    (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) {
+	} else {
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum_start = skb_transport_header(skb) - skb->head;
 		skb->csum_offset = offsetof(struct udphdr, check);
 		uh->check = ~udp_v4_check(len, saddr, daddr, 0);
-	} else {
-		__wsum csum;
-
-		uh->check = 0;
-		csum = skb_checksum(skb, 0, len, 0);
-		uh->check = udp_v4_check(len, saddr, daddr, csum);
-		if (uh->check == 0)
-			uh->check = CSUM_MANGLED_0;
-
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 }
 EXPORT_SYMBOL(udp_set_csum);
diff --git a/net/ipv6/ip6_checksum.c b/net/ipv6/ip6_checksum.c
index 4924bd7..8f92058 100644
--- a/net/ipv6/ip6_checksum.c
+++ b/net/ipv6/ip6_checksum.c
@@ -103,22 +103,11 @@ void udp6_set_csum(bool nocheck, struct sk_buff *skb,
 		uh->check = udp_v6_check(len, saddr, daddr, lco_csum(skb));
 		if (uh->check == 0)
 			uh->check = CSUM_MANGLED_0;
-	} else if (skb_dst(skb) && skb_dst(skb)->dev &&
-		   (skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) {
+	} else {
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum_start = skb_transport_header(skb) - skb->head;
 		skb->csum_offset = offsetof(struct udphdr, check);
 		uh->check = ~udp_v6_check(len, saddr, daddr, 0);
-	} else {
-		__wsum csum;
-
-		uh->check = 0;
-		csum = skb_checksum(skb, 0, len, 0);
-		uh->check = udp_v6_check(len, saddr, daddr, csum);
-		if (uh->check == 0)
-			uh->check = CSUM_MANGLED_0;
-
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 }
 EXPORT_SYMBOL(udp6_set_csum);
-- 
cgit v0.10.2


From 21e2e7f9b5fefdbf94a107a9b24d74baa5148ef3 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 20:50:44 +0000
Subject: net: enable LCO for udp_tunnel_handle_offloads() users

The only protocol affected at present is Geneve.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h
index cca2ad3..734c156 100644
--- a/include/net/udp_tunnel.h
+++ b/include/net/udp_tunnel.h
@@ -103,7 +103,8 @@ static inline struct sk_buff *udp_tunnel_handle_offloads(struct sk_buff *skb,
 {
 	int type = udp_csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
 
-	return iptunnel_handle_offloads(skb, udp_csum, type);
+	/* As we're a UDP tunnel, we support LCO, so don't need csum_help */
+	return iptunnel_handle_offloads(skb, false, type);
 }
 
 static inline void udp_tunnel_gro_complete(struct sk_buff *skb, int nhoff)
-- 
cgit v0.10.2


From b57085019dcdd52d40ba153c6957f9e5a2f4b931 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 20:57:17 +0000
Subject: net: vxlan: enable local checksum offload

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index ebf57d9..9f52203 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1702,10 +1702,8 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 		if (csum_start <= VXLAN_MAX_REMCSUM_START &&
 		    !(csum_start & VXLAN_RCO_SHIFT_MASK) &&
 		    (skb->csum_offset == offsetof(struct udphdr, check) ||
-		     skb->csum_offset == offsetof(struct tcphdr, check))) {
-			udp_sum = false;
+		     skb->csum_offset == offsetof(struct tcphdr, check)))
 			type |= SKB_GSO_TUNNEL_REMCSUM;
-		}
 	}
 
 	min_headroom = LL_RESERVED_SPACE(dst->dev) + dst->header_len
@@ -1723,7 +1721,7 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 	if (WARN_ON(!skb))
 		return -ENOMEM;
 
-	skb = iptunnel_handle_offloads(skb, udp_sum, type);
+	skb = iptunnel_handle_offloads(skb, false, type);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-- 
cgit v0.10.2


From 06f622926d0cdf923e9b33ab9f0a4d4b45983e8a Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 21:00:16 +0000
Subject: fou: enable LCO in FOU and GUE

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c
index 976f0dc..dac1874 100644
--- a/net/ipv4/fou.c
+++ b/net/ipv4/fou.c
@@ -774,7 +774,6 @@ static void fou_build_udp(struct sk_buff *skb, struct ip_tunnel_encap *e,
 	uh->dest = e->dport;
 	uh->source = sport;
 	uh->len = htons(skb->len);
-	uh->check = 0;
 	udp_set_csum(!(e->flags & TUNNEL_ENCAP_FLAG_CSUM), skb,
 		     fl4->saddr, fl4->daddr, skb->len);
 
@@ -784,11 +783,11 @@ static void fou_build_udp(struct sk_buff *skb, struct ip_tunnel_encap *e,
 int fou_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 		     u8 *protocol, struct flowi4 *fl4)
 {
-	bool csum = !!(e->flags & TUNNEL_ENCAP_FLAG_CSUM);
-	int type = csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
+	int type = e->flags & TUNNEL_ENCAP_FLAG_CSUM ? SKB_GSO_UDP_TUNNEL_CSUM :
+						       SKB_GSO_UDP_TUNNEL;
 	__be16 sport;
 
-	skb = iptunnel_handle_offloads(skb, csum, type);
+	skb = iptunnel_handle_offloads(skb, false, type);
 
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
@@ -804,8 +803,8 @@ EXPORT_SYMBOL(fou_build_header);
 int gue_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 		     u8 *protocol, struct flowi4 *fl4)
 {
-	bool csum = !!(e->flags & TUNNEL_ENCAP_FLAG_CSUM);
-	int type = csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
+	int type = e->flags & TUNNEL_ENCAP_FLAG_CSUM ? SKB_GSO_UDP_TUNNEL_CSUM :
+						       SKB_GSO_UDP_TUNNEL;
 	struct guehdr *guehdr;
 	size_t hdrlen, optlen = 0;
 	__be16 sport;
@@ -814,7 +813,6 @@ int gue_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 
 	if ((e->flags & TUNNEL_ENCAP_FLAG_REMCSUM) &&
 	    skb->ip_summed == CHECKSUM_PARTIAL) {
-		csum = false;
 		optlen += GUE_PLEN_REMCSUM;
 		type |= SKB_GSO_TUNNEL_REMCSUM;
 		need_priv = true;
@@ -822,7 +820,7 @@ int gue_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 
 	optlen += need_priv ? GUE_LEN_PRIV : 0;
 
-	skb = iptunnel_handle_offloads(skb, csum, type);
+	skb = iptunnel_handle_offloads(skb, false, type);
 
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
-- 
cgit v0.10.2


From 53936107ba2462f714bae19a754f3ebf69a11e40 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 21:01:07 +0000
Subject: net: gre: Implement LCO for GRE over IPv4

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 7c51c4e..9b31532 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -440,6 +440,17 @@ drop:
 	return 0;
 }
 
+static __sum16 gre_checksum(struct sk_buff *skb)
+{
+	__wsum csum;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		csum = lco_csum(skb);
+	else
+		csum = skb_checksum(skb, 0, skb->len, 0);
+	return csum_fold(csum);
+}
+
 static void build_header(struct sk_buff *skb, int hdr_len, __be16 flags,
 			 __be16 proto, __be32 key, __be32 seq)
 {
@@ -467,8 +478,7 @@ static void build_header(struct sk_buff *skb, int hdr_len, __be16 flags,
 		    !(skb_shinfo(skb)->gso_type &
 		      (SKB_GSO_GRE | SKB_GSO_GRE_CSUM))) {
 			*ptr = 0;
-			*(__sum16 *)ptr = csum_fold(skb_checksum(skb, 0,
-								 skb->len, 0));
+			*(__sum16 *)ptr = gre_checksum(skb);
 		}
 	}
 }
@@ -493,7 +503,7 @@ static void __gre_xmit(struct sk_buff *skb, struct net_device *dev,
 static struct sk_buff *gre_handle_offloads(struct sk_buff *skb,
 					   bool csum)
 {
-	return iptunnel_handle_offloads(skb, csum,
+	return iptunnel_handle_offloads(skb, false,
 					csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
 }
 
-- 
cgit v0.10.2


From 6fa79666e24d32be1b709f5269af41ed9e829e7e Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 21:02:31 +0000
Subject: net: ip_tunnel: remove 'csum_help' argument to
 iptunnel_handle_offloads

All users now pass false, so we can remove it, and remove the code that
 was conditional upon it.

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 9f52203..0a23c64 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1721,7 +1721,7 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 	if (WARN_ON(!skb))
 		return -ENOMEM;
 
-	skb = iptunnel_handle_offloads(skb, false, type);
+	skb = iptunnel_handle_offloads(skb, type);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 6db96ea..bc439f3 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -279,8 +279,7 @@ void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
 struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md,
 					     gfp_t flags);
 
-struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, bool gre_csum,
-					 int gso_type_mask);
+struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, int gso_type_mask);
 
 static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len)
 {
diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h
index 734c156..97f5adb 100644
--- a/include/net/udp_tunnel.h
+++ b/include/net/udp_tunnel.h
@@ -103,8 +103,7 @@ static inline struct sk_buff *udp_tunnel_handle_offloads(struct sk_buff *skb,
 {
 	int type = udp_csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
 
-	/* As we're a UDP tunnel, we support LCO, so don't need csum_help */
-	return iptunnel_handle_offloads(skb, false, type);
+	return iptunnel_handle_offloads(skb, type);
 }
 
 static inline void udp_tunnel_gro_complete(struct sk_buff *skb, int nhoff)
diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c
index dac1874..88dab0c 100644
--- a/net/ipv4/fou.c
+++ b/net/ipv4/fou.c
@@ -787,7 +787,7 @@ int fou_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 						       SKB_GSO_UDP_TUNNEL;
 	__be16 sport;
 
-	skb = iptunnel_handle_offloads(skb, false, type);
+	skb = iptunnel_handle_offloads(skb, type);
 
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
@@ -820,7 +820,7 @@ int gue_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
 
 	optlen += need_priv ? GUE_LEN_PRIV : 0;
 
-	skb = iptunnel_handle_offloads(skb, false, type);
+	skb = iptunnel_handle_offloads(skb, type);
 
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 9b31532..65748db 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -503,8 +503,7 @@ static void __gre_xmit(struct sk_buff *skb, struct net_device *dev,
 static struct sk_buff *gre_handle_offloads(struct sk_buff *skb,
 					   bool csum)
 {
-	return iptunnel_handle_offloads(skb, false,
-					csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
+	return iptunnel_handle_offloads(skb, csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
 }
 
 static struct rtable *gre_get_rt(struct sk_buff *skb,
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index d74ce93..a6e58b6 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -148,7 +148,6 @@ struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md,
 EXPORT_SYMBOL_GPL(iptunnel_metadata_reply);
 
 struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb,
-					 bool csum_help,
 					 int gso_type_mask)
 {
 	int err;
@@ -166,18 +165,13 @@ struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb,
 		return skb;
 	}
 
-	/* If packet is not gso and we are not offloading inner checksum,
-	 * clear encapsulation flag. This allows setting CHECKSUM_PARTIAL
-	 * on the outer header without confusing devices that implement
-	 * NETIF_F_IP_CSUM with encapsulation.
-	 */
-	if (skb->ip_summed == CHECKSUM_PARTIAL && csum_help) {
-		skb->encapsulation = 0;
-		err = skb_checksum_help(skb);
-		if (unlikely(err))
-			goto error;
-	} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
+	if (skb->ip_summed != CHECKSUM_PARTIAL) {
 		skb->ip_summed = CHECKSUM_NONE;
+		/* We clear encapsulation here to prevent badly-written
+		 * drivers potentially deciding to offload an inner checksum
+		 * if we set CHECKSUM_PARTIAL on the outer header.
+		 * This should go away when the drivers are all fixed.
+		 */
 		skb->encapsulation = 0;
 	}
 
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 4044da6..6ec5b42 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -219,7 +219,7 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (unlikely(skb->protocol != htons(ETH_P_IP)))
 		goto tx_error;
 
-	skb = iptunnel_handle_offloads(skb, false, SKB_GSO_IPIP);
+	skb = iptunnel_handle_offloads(skb, SKB_GSO_IPIP);
 	if (IS_ERR(skb))
 		goto out;
 
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 2066d1c..9a6b407 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -911,7 +911,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb,
 		goto tx_error;
 	}
 
-	skb = iptunnel_handle_offloads(skb, false, SKB_GSO_SIT);
+	skb = iptunnel_handle_offloads(skb, SKB_GSO_SIT);
 	if (IS_ERR(skb)) {
 		ip_rt_put(rt);
 		goto out;
@@ -1000,7 +1000,7 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	const struct iphdr  *tiph = &tunnel->parms.iph;
 
-	skb = iptunnel_handle_offloads(skb, false, SKB_GSO_IPIP);
+	skb = iptunnel_handle_offloads(skb, SKB_GSO_IPIP);
 	if (IS_ERR(skb))
 		goto out;
 
diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
index 3264cb49..a3f5cd9 100644
--- a/net/netfilter/ipvs/ip_vs_xmit.c
+++ b/net/netfilter/ipvs/ip_vs_xmit.c
@@ -1019,8 +1019,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	if (IS_ERR(skb))
 		goto tx_error;
 
-	skb = iptunnel_handle_offloads(
-		skb, false, __tun_gso_type_mask(AF_INET, cp->af));
+	skb = iptunnel_handle_offloads(skb, __tun_gso_type_mask(AF_INET, cp->af));
 	if (IS_ERR(skb))
 		goto tx_error;
 
@@ -1112,8 +1111,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	if (IS_ERR(skb))
 		goto tx_error;
 
-	skb = iptunnel_handle_offloads(
-		skb, false, __tun_gso_type_mask(AF_INET6, cp->af));
+	skb = iptunnel_handle_offloads(skb, __tun_gso_type_mask(AF_INET6, cp->af));
 	if (IS_ERR(skb))
 		goto tx_error;
 
-- 
cgit v0.10.2


From e8ae7b000e64cf76283c72cae5e3ecd246618ef4 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree@solarflare.com>
Date: Thu, 11 Feb 2016 21:03:37 +0000
Subject: Documentation/networking: add checksum-offloads.txt to explain LCO

Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX
index df27a1a..415154a 100644
--- a/Documentation/networking/00-INDEX
+++ b/Documentation/networking/00-INDEX
@@ -44,6 +44,8 @@ can.txt
 	- documentation on CAN protocol family.
 cdc_mbim.txt
 	- 3G/LTE USB modem (Mobile Broadband Interface Model)
+checksum-offloads.txt
+	- Explanation of checksum offloads; LCO, RCO
 cops.txt
 	- info on the COPS LocalTalk Linux driver
 cs89x0.txt
diff --git a/Documentation/networking/checksum-offloads.txt b/Documentation/networking/checksum-offloads.txt
new file mode 100644
index 0000000..de2a327
--- /dev/null
+++ b/Documentation/networking/checksum-offloads.txt
@@ -0,0 +1,119 @@
+Checksum Offloads in the Linux Networking Stack
+
+
+Introduction
+============
+
+This document describes a set of techniques in the Linux networking stack
+ to take advantage of checksum offload capabilities of various NICs.
+
+The following technologies are described:
+ * TX Checksum Offload
+ * LCO: Local Checksum Offload
+ * RCO: Remote Checksum Offload
+
+Things that should be documented here but aren't yet:
+ * RX Checksum Offload
+ * CHECKSUM_UNNECESSARY conversion
+
+
+TX Checksum Offload
+===================
+
+The interface for offloading a transmit checksum to a device is explained
+ in detail in comments near the top of include/linux/skbuff.h.
+In brief, it allows to request the device fill in a single ones-complement
+ checksum defined by the sk_buff fields skb->csum_start and
+ skb->csum_offset.  The device should compute the 16-bit ones-complement
+ checksum (i.e. the 'IP-style' checksum) from csum_start to the end of the
+ packet, and fill in the result at (csum_start + csum_offset).
+Because csum_offset cannot be negative, this ensures that the previous
+ value of the checksum field is included in the checksum computation, thus
+ it can be used to supply any needed corrections to the checksum (such as
+ the sum of the pseudo-header for UDP or TCP).
+This interface only allows a single checksum to be offloaded.  Where
+ encapsulation is used, the packet may have multiple checksum fields in
+ different header layers, and the rest will have to be handled by another
+ mechanism such as LCO or RCO.
+No offloading of the IP header checksum is performed; it is always done in
+ software.  This is OK because when we build the IP header, we obviously
+ have it in cache, so summing it isn't expensive.  It's also rather short.
+The requirements for GSO are more complicated, because when segmenting an
+ encapsulated packet both the inner and outer checksums may need to be
+ edited or recomputed for each resulting segment.  See the skbuff.h comment
+ (section 'E') for more details.
+
+A driver declares its offload capabilities in netdev->hw_features; see
+ Documentation/networking/netdev-features for more.  Note that a device
+ which only advertises NETIF_F_IP[V6]_CSUM must still obey the csum_start
+ and csum_offset given in the SKB; if it tries to deduce these itself in
+ hardware (as some NICs do) the driver should check that the values in the
+ SKB match those which the hardware will deduce, and if not, fall back to
+ checksumming in software instead (with skb_checksum_help or one of the
+ skb_csum_off_chk* functions as mentioned in include/linux/skbuff.h).  This
+ is a pain, but that's what you get when hardware tries to be clever.
+
+The stack should, for the most part, assume that checksum offload is
+ supported by the underlying device.  The only place that should check is
+ validate_xmit_skb(), and the functions it calls directly or indirectly.
+ That function compares the offload features requested by the SKB (which
+ may include other offloads besides TX Checksum Offload) and, if they are
+ not supported or enabled on the device (determined by netdev->features),
+ performs the corresponding offload in software.  In the case of TX
+ Checksum Offload, that means calling skb_checksum_help(skb).
+
+
+LCO: Local Checksum Offload
+===========================
+
+LCO is a technique for efficiently computing the outer checksum of an
+ encapsulated datagram when the inner checksum is due to be offloaded.
+The ones-complement sum of a correctly checksummed TCP or UDP packet is
+ equal to the sum of the pseudo header, because everything else gets
+ 'cancelled out' by the checksum field.  This is because the sum was
+ complemented before being written to the checksum field.
+More generally, this holds in any case where the 'IP-style' ones complement
+ checksum is used, and thus any checksum that TX Checksum Offload supports.
+That is, if we have set up TX Checksum Offload with a start/offset pair, we
+ know that _after the device has filled in that checksum_, the ones
+ complement sum from csum_start to the end of the packet will be equal to
+ _whatever value we put in the checksum field beforehand_.  This allows us
+ to compute the outer checksum without looking at the payload: we simply
+ stop summing when we get to csum_start, then add the 16-bit word at
+ (csum_start + csum_offset).
+Then, when the true inner checksum is filled in (either by hardware or by
+ skb_checksum_help()), the outer checksum will become correct by virtue of
+ the arithmetic.
+
+LCO is performed by the stack when constructing an outer UDP header for an
+ encapsulation such as VXLAN or GENEVE, in udp_set_csum().  Similarly for
+ the IPv6 equivalents, in udp6_set_csum().
+It is also performed when constructing an IPv4 GRE header, in
+ net/ipv4/ip_gre.c:build_header().  It is *not* currently performed when
+ constructing an IPv6 GRE header; the GRE checksum is computed over the
+ whole packet in net/ipv6/ip6_gre.c:ip6gre_xmit2(), but it should be
+ possible to use LCO here as IPv6 GRE still uses an IP-style checksum.
+All of the LCO implementations use a helper function lco_csum(), in
+ include/linux/skbuff.h.
+
+LCO can safely be used for nested encapsulations; in this case, the outer
+ encapsulation layer will sum over both its own header and the 'middle'
+ header.  This does mean that the 'middle' header will get summed multiple
+ times, but there doesn't seem to be a way to avoid that without incurring
+ bigger costs (e.g. in SKB bloat).
+
+
+RCO: Remote Checksum Offload
+============================
+
+RCO is a technique for eliding the inner checksum of an encapsulated
+ datagram, allowing the outer checksum to be offloaded.  It does, however,
+ involve a change to the encapsulation protocols, which the receiver must
+ also support.  For this reason, it is disabled by default.
+RCO is detailed in the following Internet-Drafts:
+https://tools.ietf.org/html/draft-herbert-remotecsumoffload-00
+https://tools.ietf.org/html/draft-herbert-vxlan-rco-00
+In Linux, RCO is implemented individually in each encapsulation protocol,
+ and most tunnel types have flags controlling its use.  For instance, VXLAN
+ has the flag VXLAN_F_REMCSUM_TX (per struct vxlan_rdst) to indicate that
+ RCO should be used when transmitting to a given remote destination.
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index cf906d1..3920675 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3705,6 +3705,8 @@ static inline unsigned int skb_gso_network_seglen(const struct sk_buff *skb)
 /* Local Checksum Offload.
  * Compute outer checksum based on the assumption that the
  * inner checksum will be offloaded later.
+ * See Documentation/networking/checksum-offloads.txt for
+ * explanation of how this works.
  * Fill in outer checksum adjustment (e.g. with sum of outer
  * pseudo-header) before calling.
  * Also ensure that inner checksum is in linear data area.
-- 
cgit v0.10.2


From ceb27759987ec10ba22332bd7fdf1cfb35b86991 Mon Sep 17 00:00:00 2001
From: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp>
Date: Fri, 11 Dec 2015 18:43:59 +0900
Subject: igb: Remove unnecessary flag setting in igb_set_flag_queue_pairs()

If VFs are enabled (max_vfs >= 1), both max_rss_queues and
adapter->rss_queues are set to 2 in the case of e1000_82576.
In this case, IGB_FLAG_QUEUE_PAIRS is always set in the default block as a
result of fall-through, thus setting it in the e1000_82576 block is not
necessary.

Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 31e5f39..eb24b40 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2921,14 +2921,6 @@ void igb_set_flag_queue_pairs(struct igb_adapter *adapter,
 		/* Device supports enough interrupts without queue pairing. */
 		break;
 	case e1000_82576:
-		/* If VFs are going to be allocated with RSS queues then we
-		 * should pair the queues in order to conserve interrupts due
-		 * to limited supply.
-		 */
-		if ((adapter->rss_queues > 1) &&
-		    (adapter->vfs_allocated_count > 6))
-			adapter->flags |= IGB_FLAG_QUEUE_PAIRS;
-		/* fall through */
 	case e1000_82580:
 	case e1000_i350:
 	case e1000_i354:
-- 
cgit v0.10.2


From 37a5d163fb447b39f7960d0534de30e88ad395bb Mon Sep 17 00:00:00 2001
From: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp>
Date: Fri, 11 Dec 2015 18:44:00 +0900
Subject: igb: Unpair the queues when changing the number of queues

By the commit 72ddef0506da ("igb: Fix oops caused by missing queue
pairing"), the IGB_FLAG_QUEUE_PAIRS flag can now be set when changing the
number of queues by "ethtool -L", but it is never cleared unless the igb
driver is reloaded.
This patch clears it if queue pairing becomes unnecessary as a result of
"ethtool -L".

Signed-off-by: Shota Suzuki <suzuki_shota_t3@lab.ntt.co.jp>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index eb24b40..85c47aa 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2931,6 +2931,8 @@ void igb_set_flag_queue_pairs(struct igb_adapter *adapter,
 		 */
 		if (adapter->rss_queues > (max_rss_queues / 2))
 			adapter->flags |= IGB_FLAG_QUEUE_PAIRS;
+		else
+			adapter->flags &= ~IGB_FLAG_QUEUE_PAIRS;
 		break;
 	}
 }
-- 
cgit v0.10.2


From 9ce0e8d72678b5b60c99ce4c7af15ec127c761cb Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Fri, 11 Dec 2015 16:45:38 -0800
Subject: igb/igbvf: don't give up

The driver shouldn't just give up if it fails to get the hardware
mailbox lock. This can happen in a situation where the PF-VF
communication channel is heavily loaded and causes complete
communications failure between the PF and VF drivers.

Add a counter and a delay. The driver will now retry ten times, waiting
one millisecond between retries.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_mbx.c b/drivers/net/ethernet/intel/igb/e1000_mbx.c
index 162cc49..10f5c9e 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mbx.c
+++ b/drivers/net/ethernet/intel/igb/e1000_mbx.c
@@ -322,14 +322,20 @@ static s32 igb_obtain_mbx_lock_pf(struct e1000_hw *hw, u16 vf_number)
 {
 	s32 ret_val = -E1000_ERR_MBX;
 	u32 p2v_mailbox;
+	int count = 10;
 
-	/* Take ownership of the buffer */
-	wr32(E1000_P2VMAILBOX(vf_number), E1000_P2VMAILBOX_PFU);
+	do {
+		/* Take ownership of the buffer */
+		wr32(E1000_P2VMAILBOX(vf_number), E1000_P2VMAILBOX_PFU);
 
-	/* reserve mailbox for vf use */
-	p2v_mailbox = rd32(E1000_P2VMAILBOX(vf_number));
-	if (p2v_mailbox & E1000_P2VMAILBOX_PFU)
-		ret_val = 0;
+		/* reserve mailbox for vf use */
+		p2v_mailbox = rd32(E1000_P2VMAILBOX(vf_number));
+		if (p2v_mailbox & E1000_P2VMAILBOX_PFU) {
+			ret_val = 0;
+			break;
+		}
+		udelay(1000);
+	} while (count-- > 0);
 
 	return ret_val;
 }
diff --git a/drivers/net/ethernet/intel/igbvf/mbx.c b/drivers/net/ethernet/intel/igbvf/mbx.c
index 7b6cb4c..01752f4 100644
--- a/drivers/net/ethernet/intel/igbvf/mbx.c
+++ b/drivers/net/ethernet/intel/igbvf/mbx.c
@@ -234,13 +234,19 @@ static s32 e1000_check_for_rst_vf(struct e1000_hw *hw)
 static s32 e1000_obtain_mbx_lock_vf(struct e1000_hw *hw)
 {
 	s32 ret_val = -E1000_ERR_MBX;
-
-	/* Take ownership of the buffer */
-	ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_VFU);
-
-	/* reserve mailbox for VF use */
-	if (e1000_read_v2p_mailbox(hw) & E1000_V2PMAILBOX_VFU)
-		ret_val = E1000_SUCCESS;
+	int count = 10;
+
+	do {
+		/* Take ownership of the buffer */
+		ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_VFU);
+
+		/* reserve mailbox for VF use */
+		if (e1000_read_v2p_mailbox(hw) & E1000_V2PMAILBOX_VFU) {
+			ret_val = 0;
+			break;
+		}
+		udelay(1000);
+	} while (count-- > 0);
 
 	return ret_val;
 }
-- 
cgit v0.10.2


From c3278587e7d34cfbc1d38d3ae25923343af7752a Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:10:23 -0800
Subject: igb: clean up code for setting MAC address

Drop a bunch of hand written byte swapping code in favor of just doing the
byte swapping ourselves.  The registers are little endian registers storing
a big endian value so if we read the MAC address array as little endian
then we will get the CPU registers into the proper layout.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 85c47aa..02f19e4 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -7698,15 +7698,14 @@ static void igb_io_resume(struct pci_dev *pdev)
 static void igb_rar_set_qsel(struct igb_adapter *adapter, u8 *addr, u32 index,
 			     u8 qsel)
 {
-	u32 rar_low, rar_high;
 	struct e1000_hw *hw = &adapter->hw;
+	u32 rar_low, rar_high;
 
 	/* HW expects these in little endian so we reverse the byte order
-	 * from network order (big endian) to little endian
+	 * from network order (big endian) to CPU endian
 	 */
-	rar_low = ((u32) addr[0] | ((u32) addr[1] << 8) |
-		   ((u32) addr[2] << 16) | ((u32) addr[3] << 24));
-	rar_high = ((u32) addr[4] | ((u32) addr[5] << 8));
+	rar_low = le32_to_cpup((__be32 *)(addr));
+	rar_high = le16_to_cpup((__be16 *)(addr + 4));
 
 	/* Indicate to hardware the Address is Valid. */
 	rar_high |= E1000_RAH_AV;
-- 
cgit v0.10.2


From 832e821c51e381966464c8a0f30f12eb1514eba0 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:10:30 -0800
Subject: igb: Refactor VFTA configuration

This patch starts the clean-up process on the VFTA configuration.
Specifically in this patch I attempt to address and simplify several items
while also updating the code to bring it more inline with what is already
in ixgbe.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.c b/drivers/net/ethernet/intel/igb/e1000_82575.c
index adb33e2..fff5052 100644
--- a/drivers/net/ethernet/intel/igb/e1000_82575.c
+++ b/drivers/net/ethernet/intel/igb/e1000_82575.c
@@ -34,6 +34,7 @@
 #include "e1000_mac.h"
 #include "e1000_82575.h"
 #include "e1000_i210.h"
+#include "igb.h"
 
 static s32  igb_get_invariants_82575(struct e1000_hw *);
 static s32  igb_acquire_phy_82575(struct e1000_hw *);
@@ -71,6 +72,32 @@ static s32 igb_update_nvm_checksum_i350(struct e1000_hw *hw);
 static const u16 e1000_82580_rxpbs_table[] = {
 	36, 72, 144, 1, 2, 4, 8, 16, 35, 70, 140 };
 
+/* Due to a hw errata, if the host tries to  configure the VFTA register
+ * while performing queries from the BMC or DMA, then the VFTA in some
+ * cases won't be written.
+ */
+
+/**
+ *  igb_write_vfta_i350 - Write value to VLAN filter table
+ *  @hw: pointer to the HW structure
+ *  @offset: register offset in VLAN filter table
+ *  @value: register value written to VLAN filter table
+ *
+ *  Writes value at the given offset in the register array which stores
+ *  the VLAN filter table.
+ **/
+static void igb_write_vfta_i350(struct e1000_hw *hw, u32 offset, u32 value)
+{
+	struct igb_adapter *adapter = hw->back;
+	int i;
+
+	for (i = 10; i--;)
+		array_wr32(E1000_VFTA, offset, value);
+
+	wrfl();
+	adapter->shadow_vfta[offset] = value;
+}
+
 /**
  *  igb_sgmii_uses_mdio_82575 - Determine if I2C pins are for external MDIO
  *  @hw: pointer to the HW structure
@@ -429,6 +456,11 @@ static s32 igb_init_mac_params_82575(struct e1000_hw *hw)
 		mac->ops.release_swfw_sync = igb_release_swfw_sync_82575;
 	}
 
+	if ((hw->mac.type == e1000_i350) || (hw->mac.type == e1000_i354))
+		mac->ops.write_vfta = igb_write_vfta_i350;
+	else
+		mac->ops.write_vfta = igb_write_vfta;
+
 	/* Set if part includes ASF firmware */
 	mac->asf_firmware_present = true;
 	/* Set if manageability features are enabled. */
@@ -1517,10 +1549,7 @@ static s32 igb_init_hw_82575(struct e1000_hw *hw)
 
 	/* Disabling VLAN filtering */
 	hw_dbg("Initializing the IEEE VLAN\n");
-	if ((hw->mac.type == e1000_i350) || (hw->mac.type == e1000_i354))
-		igb_clear_vfta_i350(hw);
-	else
-		igb_clear_vfta(hw);
+	igb_clear_vfta(hw);
 
 	/* Setup the receive address */
 	igb_init_rx_addrs(hw, rar_count);
diff --git a/drivers/net/ethernet/intel/igb/e1000_hw.h b/drivers/net/ethernet/intel/igb/e1000_hw.h
index 4034207..f0c416e 100644
--- a/drivers/net/ethernet/intel/igb/e1000_hw.h
+++ b/drivers/net/ethernet/intel/igb/e1000_hw.h
@@ -325,7 +325,7 @@ struct e1000_mac_operations {
 	s32 (*get_thermal_sensor_data)(struct e1000_hw *);
 	s32 (*init_thermal_sensor_thresh)(struct e1000_hw *);
 #endif
-
+	void (*write_vfta)(struct e1000_hw *, u32, u32);
 };
 
 struct e1000_phy_operations {
diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c
index 2a88595..97f6fae 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.c
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.c
@@ -92,10 +92,8 @@ void igb_clear_vfta(struct e1000_hw *hw)
 {
 	u32 offset;
 
-	for (offset = 0; offset < E1000_VLAN_FILTER_TBL_SIZE; offset++) {
-		array_wr32(E1000_VFTA, offset, 0);
-		wrfl();
-	}
+	for (offset = E1000_VLAN_FILTER_TBL_SIZE; offset--;)
+		hw->mac.ops.write_vfta(hw, offset, 0);
 }
 
 /**
@@ -107,54 +105,14 @@ void igb_clear_vfta(struct e1000_hw *hw)
  *  Writes value at the given offset in the register array which stores
  *  the VLAN filter table.
  **/
-static void igb_write_vfta(struct e1000_hw *hw, u32 offset, u32 value)
+void igb_write_vfta(struct e1000_hw *hw, u32 offset, u32 value)
 {
+	struct igb_adapter *adapter = hw->back;
+
 	array_wr32(E1000_VFTA, offset, value);
 	wrfl();
-}
-
-/* Due to a hw errata, if the host tries to  configure the VFTA register
- * while performing queries from the BMC or DMA, then the VFTA in some
- * cases won't be written.
- */
-
-/**
- *  igb_clear_vfta_i350 - Clear VLAN filter table
- *  @hw: pointer to the HW structure
- *
- *  Clears the register array which contains the VLAN filter table by
- *  setting all the values to 0.
- **/
-void igb_clear_vfta_i350(struct e1000_hw *hw)
-{
-	u32 offset;
-	int i;
-
-	for (offset = 0; offset < E1000_VLAN_FILTER_TBL_SIZE; offset++) {
-		for (i = 0; i < 10; i++)
-			array_wr32(E1000_VFTA, offset, 0);
-
-		wrfl();
-	}
-}
-
-/**
- *  igb_write_vfta_i350 - Write value to VLAN filter table
- *  @hw: pointer to the HW structure
- *  @offset: register offset in VLAN filter table
- *  @value: register value written to VLAN filter table
- *
- *  Writes value at the given offset in the register array which stores
- *  the VLAN filter table.
- **/
-static void igb_write_vfta_i350(struct e1000_hw *hw, u32 offset, u32 value)
-{
-	int i;
 
-	for (i = 0; i < 10; i++)
-		array_wr32(E1000_VFTA, offset, value);
-
-	wrfl();
+	adapter->shadow_vfta[offset] = value;
 }
 
 /**
@@ -185,38 +143,42 @@ void igb_init_rx_addrs(struct e1000_hw *hw, u16 rar_count)
 /**
  *  igb_vfta_set - enable or disable vlan in VLAN filter table
  *  @hw: pointer to the HW structure
- *  @vid: VLAN id to add or remove
- *  @add: if true add filter, if false remove
+ *  @vlan: VLAN id to add or remove
+ *  @vlan_on: if true add filter, if false remove
  *
  *  Sets or clears a bit in the VLAN filter table array based on VLAN id
  *  and if we are adding or removing the filter
  **/
-s32 igb_vfta_set(struct e1000_hw *hw, u32 vid, bool add)
+s32 igb_vfta_set(struct e1000_hw *hw, u32 vlan, bool vlan_on)
 {
-	u32 index = (vid >> E1000_VFTA_ENTRY_SHIFT) & E1000_VFTA_ENTRY_MASK;
-	u32 mask = 1 << (vid & E1000_VFTA_ENTRY_BIT_SHIFT_MASK);
-	u32 vfta;
 	struct igb_adapter *adapter = hw->back;
-	s32 ret_val = 0;
+	u32 regidx, vfta_delta, vfta;
+
+	if (vlan > 4095)
+		return E1000_ERR_PARAM;
+
+	/* Part 1
+	 * The VFTA is a bitstring made up of 128 32-bit registers
+	 * that enable the particular VLAN id, much like the MTA:
+	 *    bits[11-5]: which register
+	 *    bits[4-0]:  which bit in the register
+	 */
+	regidx = vlan / 32;
+	vfta_delta = 1 << (vlan % 32);
+	vfta = adapter->shadow_vfta[regidx];
 
-	vfta = adapter->shadow_vfta[index];
+	/* vfta_delta represents the difference between the current value
+	 * of vfta and the value we want in the register.  Since the diff
+	 * is an XOR mask we can just update vfta using an XOR.
+	 */
+	vfta_delta &= vlan_on ? ~vfta : vfta;
+	vfta ^= vfta_delta;
 
 	/* bit was set/cleared before we started */
-	if ((!!(vfta & mask)) == add) {
-		ret_val = -E1000_ERR_CONFIG;
-	} else {
-		if (add)
-			vfta |= mask;
-		else
-			vfta &= ~mask;
-	}
-	if ((hw->mac.type == e1000_i350) || (hw->mac.type == e1000_i354))
-		igb_write_vfta_i350(hw, index, vfta);
-	else
-		igb_write_vfta(hw, index, vfta);
-	adapter->shadow_vfta[index] = vfta;
+	if (vfta_delta)
+		hw->mac.ops.write_vfta(hw, regidx, vfta);
 
-	return ret_val;
+	return 0;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.h b/drivers/net/ethernet/intel/igb/e1000_mac.h
index ea24961b..4fbb953 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.h
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.h
@@ -56,7 +56,7 @@ s32  igb_write_8bit_ctrl_reg(struct e1000_hw *hw, u32 reg,
 
 void igb_clear_hw_cntrs_base(struct e1000_hw *hw);
 void igb_clear_vfta(struct e1000_hw *hw);
-void igb_clear_vfta_i350(struct e1000_hw *hw);
+void igb_write_vfta(struct e1000_hw *hw, u32 offset, u32 value);
 s32  igb_vfta_set(struct e1000_hw *hw, u32 vid, bool add);
 void igb_config_collision_dist(struct e1000_hw *hw);
 void igb_init_rx_addrs(struct e1000_hw *hw, u16 rar_count);
-- 
cgit v0.10.2


From 45693bcb00cbd379c373ab22ccd9a9d4755cc7ed Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:10:39 -0800
Subject: igb: Allow asymmetric configuration of MTU versus Rx frame size

Since the igb driver is using page based receive there is no point in
limiting the Rx capabilities of the device.  The driver can receive 9K
jumbo frames at all times.  The only changes needed due to MTU changes are
updates for the FIFO sizes and flow-control watermarks.

Update the maximum frame size to reflect the 9.5K limitation of the
hardware, and replace all instances of max_frame_size with
MAX_JUMBO_FRAME_SIZE when referring to an Rx FIFO or frame.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_defines.h b/drivers/net/ethernet/intel/igb/e1000_defines.h
index c3c598c..e9f23ee 100644
--- a/drivers/net/ethernet/intel/igb/e1000_defines.h
+++ b/drivers/net/ethernet/intel/igb/e1000_defines.h
@@ -356,7 +356,8 @@
 /* Ethertype field values */
 #define ETHERNET_IEEE_VLAN_TYPE 0x8100  /* 802.3ac packet */
 
-#define MAX_JUMBO_FRAME_SIZE    0x3F00
+/* As per the EAS the maximum supported size is 9.5KB (9728 bytes) */
+#define MAX_JUMBO_FRAME_SIZE	0x2600
 
 /* PBA constants */
 #define E1000_PBA_34K 0x0022
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 02f19e4..b676881 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -1862,7 +1862,7 @@ void igb_reset(struct igb_adapter *adapter)
 	struct e1000_hw *hw = &adapter->hw;
 	struct e1000_mac_info *mac = &hw->mac;
 	struct e1000_fc_info *fc = &hw->fc;
-	u32 pba = 0, tx_space, min_tx_space, min_rx_space, hwm;
+	u32 pba, hwm;
 
 	/* Repartition Pba for greater than 9k mtu
 	 * To take effect CTRL.RST is required.
@@ -1886,9 +1886,10 @@ void igb_reset(struct igb_adapter *adapter)
 		break;
 	}
 
-	if ((adapter->max_frame_size > ETH_FRAME_LEN + ETH_FCS_LEN) &&
-	    (mac->type < e1000_82576)) {
-		/* adjust PBA for jumbo frames */
+	if (mac->type == e1000_82575) {
+		u32 min_rx_space, min_tx_space, needed_tx_space;
+
+		/* write Rx PBA so that hardware can report correct Tx PBA */
 		wr32(E1000_PBA, pba);
 
 		/* To maintain wire speed transmits, the Tx FIFO should be
@@ -1898,31 +1899,26 @@ void igb_reset(struct igb_adapter *adapter)
 		 * one full receive packet and is similarly rounded up and
 		 * expressed in KB.
 		 */
-		pba = rd32(E1000_PBA);
-		/* upper 16 bits has Tx packet buffer allocation size in KB */
-		tx_space = pba >> 16;
-		/* lower 16 bits has Rx packet buffer allocation size in KB */
-		pba &= 0xffff;
-		/* the Tx fifo also stores 16 bytes of information about the Tx
-		 * but don't include ethernet FCS because hardware appends it
+		min_rx_space = DIV_ROUND_UP(MAX_JUMBO_FRAME_SIZE, 1024);
+
+		/* The Tx FIFO also stores 16 bytes of information about the Tx
+		 * but don't include Ethernet FCS because hardware appends it.
+		 * We only need to round down to the nearest 512 byte block
+		 * count since the value we care about is 2 frames, not 1.
 		 */
-		min_tx_space = (adapter->max_frame_size +
-				sizeof(union e1000_adv_tx_desc) -
-				ETH_FCS_LEN) * 2;
-		min_tx_space = ALIGN(min_tx_space, 1024);
-		min_tx_space >>= 10;
-		/* software strips receive CRC, so leave room for it */
-		min_rx_space = adapter->max_frame_size;
-		min_rx_space = ALIGN(min_rx_space, 1024);
-		min_rx_space >>= 10;
+		min_tx_space = adapter->max_frame_size;
+		min_tx_space += sizeof(union e1000_adv_tx_desc) - ETH_FCS_LEN;
+		min_tx_space = DIV_ROUND_UP(min_tx_space, 512);
+
+		/* upper 16 bits has Tx packet buffer allocation size in KB */
+		needed_tx_space = min_tx_space - (rd32(E1000_PBA) >> 16);
 
 		/* If current Tx allocation is less than the min Tx FIFO size,
 		 * and the min Tx FIFO size is less than the current Rx FIFO
-		 * allocation, take space away from current Rx allocation
+		 * allocation, take space away from current Rx allocation.
 		 */
-		if (tx_space < min_tx_space &&
-		    ((min_tx_space - tx_space) < pba)) {
-			pba = pba - (min_tx_space - tx_space);
+		if (needed_tx_space < pba) {
+			pba -= needed_tx_space;
 
 			/* if short on Rx space, Rx wins and must trump Tx
 			 * adjustment
@@ -1930,18 +1926,20 @@ void igb_reset(struct igb_adapter *adapter)
 			if (pba < min_rx_space)
 				pba = min_rx_space;
 		}
+
+		/* adjust PBA for jumbo frames */
 		wr32(E1000_PBA, pba);
 	}
 
-	/* flow control settings */
-	/* The high water mark must be low enough to fit one full frame
-	 * (or the size used for early receive) above it in the Rx FIFO.
-	 * Set it to the lower of:
-	 * - 90% of the Rx FIFO size, or
-	 * - the full Rx FIFO size minus one full frame
+	/* flow control settings
+	 * The high water mark must be low enough to fit one full frame
+	 * after transmitting the pause frame.  As such we must have enough
+	 * space to allow for us to complete our current transmit and then
+	 * receive the frame that is in progress from the link partner.
+	 * Set it to:
+	 * - the full Rx FIFO size minus one full Tx plus one full Rx frame
 	 */
-	hwm = min(((pba << 10) * 9 / 10),
-			((pba << 10) - 2 * adapter->max_frame_size));
+	hwm = (pba << 10) - (adapter->max_frame_size + MAX_JUMBO_FRAME_SIZE);
 
 	fc->high_water = hwm & 0xFFFFFFF0;	/* 16-byte granularity */
 	fc->low_water = fc->high_water - 16;
@@ -3492,7 +3490,7 @@ void igb_setup_rctl(struct igb_adapter *adapter)
 	/* disable store bad packets and clear size bits. */
 	rctl &= ~(E1000_RCTL_SBP | E1000_RCTL_SZ_256);
 
-	/* enable LPE to prevent packets larger than max_frame_size */
+	/* enable LPE to allow for reception of jumbo frames */
 	rctl |= E1000_RCTL_LPE;
 
 	/* disable queue 0 to prevent tail write w/o re-config */
@@ -3548,32 +3546,6 @@ static inline int igb_set_vf_rlpml(struct igb_adapter *adapter, int size,
 	return 0;
 }
 
-/**
- *  igb_rlpml_set - set maximum receive packet size
- *  @adapter: board private structure
- *
- *  Configure maximum receivable packet size.
- **/
-static void igb_rlpml_set(struct igb_adapter *adapter)
-{
-	u32 max_frame_size = adapter->max_frame_size;
-	struct e1000_hw *hw = &adapter->hw;
-	u16 pf_id = adapter->vfs_allocated_count;
-
-	if (pf_id) {
-		igb_set_vf_rlpml(adapter, max_frame_size, pf_id);
-		/* If we're in VMDQ or SR-IOV mode, then set global RLPML
-		 * to our max jumbo frame size, in case we need to enable
-		 * jumbo frames on one of the rings later.
-		 * This will not pass over-length frames into the default
-		 * queue because it's gated by the VMOLR.RLPML.
-		 */
-		max_frame_size = MAX_JUMBO_FRAME_SIZE;
-	}
-
-	wr32(E1000_RLPML, max_frame_size);
-}
-
 static inline void igb_set_vmolr(struct igb_adapter *adapter,
 				 int vfn, bool aupe)
 {
@@ -4067,7 +4039,14 @@ static void igb_set_rx_mode(struct net_device *netdev)
 
 	vmolr |= rd32(E1000_VMOLR(vfn)) &
 		 ~(E1000_VMOLR_ROPE | E1000_VMOLR_MPME | E1000_VMOLR_ROMPE);
+
+	/* enable Rx jumbo frames, no need for restriction */
+	vmolr &= ~E1000_VMOLR_RLPML_MASK;
+	vmolr |= MAX_JUMBO_FRAME_SIZE | E1000_VMOLR_LPE;
+
 	wr32(E1000_VMOLR(vfn), vmolr);
+	wr32(E1000_RLPML, MAX_JUMBO_FRAME_SIZE);
+
 	igb_restore_vf_multicasts(adapter);
 }
 
@@ -7195,8 +7174,6 @@ static void igb_vlan_mode(struct net_device *netdev, netdev_features_t features)
 		ctrl &= ~E1000_CTRL_VME;
 		wr32(E1000_CTRL, ctrl);
 	}
-
-	igb_rlpml_set(adapter);
 }
 
 static int igb_vlan_rx_add_vid(struct net_device *netdev,
@@ -7952,9 +7929,7 @@ static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
 			 * than the Rx threshold. Set hwm to PBA - max frame
 			 * size in 16B units, capping it at PBA - 6KB.
 			 */
-			hwm = 64 * pba - adapter->max_frame_size / 16;
-			if (hwm < 64 * (pba - 6))
-				hwm = 64 * (pba - 6);
+			hwm = 64 * (pba - 6);
 			reg = rd32(E1000_FCRTC);
 			reg &= ~E1000_FCRTC_RTH_COAL_MASK;
 			reg |= ((hwm << E1000_FCRTC_RTH_COAL_SHIFT)
@@ -7964,9 +7939,7 @@ static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
 			/* Set the DMA Coalescing Rx threshold to PBA - 2 * max
 			 * frame size, capping it at PBA - 10KB.
 			 */
-			dmac_thr = pba - adapter->max_frame_size / 512;
-			if (dmac_thr < pba - 10)
-				dmac_thr = pba - 10;
+			dmac_thr = pba - 10;
 			reg = rd32(E1000_DMACR);
 			reg &= ~E1000_DMACR_DMACTHR_MASK;
 			reg |= ((dmac_thr << E1000_DMACR_DMACTHR_SHIFT)
-- 
cgit v0.10.2


From d3836f8e2517fb04328c673989fd780030926694 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:10:47 -0800
Subject: igb: Do not factor VLANs into RLPML calculation

The RLPML registers already take the size of VLAN headers into account when
determining the maximum packet length.  This is called out in EAS documents
for several parts including the 82576 and the i350.  As such we can drop
the addition of size to the value programmed into the RLPML registers.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index e3cb93b..d135261 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -95,7 +95,6 @@ struct vf_data_storage {
 	unsigned char vf_mac_addresses[ETH_ALEN];
 	u16 vf_mc_hashes[IGB_MAX_VF_MC_ENTRIES];
 	u16 num_vf_mc_hashes;
-	u16 vlans_enabled;
 	u32 flags;
 	unsigned long last_nack;
 	u16 pf_vlan; /* When set, guest VLAN config not allowed. */
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index b676881..9461480 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -3531,12 +3531,8 @@ static inline int igb_set_vf_rlpml(struct igb_adapter *adapter, int size,
 	struct e1000_hw *hw = &adapter->hw;
 	u32 vmolr;
 
-	/* if it isn't the PF check to see if VFs are enabled and
-	 * increase the size to support vlan tags
-	 */
-	if (vfn < adapter->vfs_allocated_count &&
-	    adapter->vf_data[vfn].vlans_enabled)
-		size += VLAN_TAG_SIZE;
+	if (size > MAX_JUMBO_FRAME_SIZE)
+		size = MAX_JUMBO_FRAME_SIZE;
 
 	vmolr = rd32(E1000_VMOLR(vfn));
 	vmolr &= ~E1000_VMOLR_RLPML_MASK;
@@ -5787,8 +5783,6 @@ static void igb_clear_vf_vfta(struct igb_adapter *adapter, u32 vf)
 
 		wr32(E1000_VLVF(i), reg);
 	}
-
-	adapter->vf_data[vf].vlans_enabled = 0;
 }
 
 static s32 igb_vlvf_set(struct igb_adapter *adapter, u32 vid, bool add, u32 vf)
@@ -5837,23 +5831,6 @@ static s32 igb_vlvf_set(struct igb_adapter *adapter, u32 vid, bool add, u32 vf)
 			reg &= ~E1000_VLVF_VLANID_MASK;
 			reg |= vid;
 			wr32(E1000_VLVF(i), reg);
-
-			/* do not modify RLPML for PF devices */
-			if (vf >= adapter->vfs_allocated_count)
-				return 0;
-
-			if (!adapter->vf_data[vf].vlans_enabled) {
-				u32 size;
-
-				reg = rd32(E1000_VMOLR(vf));
-				size = reg & E1000_VMOLR_RLPML_MASK;
-				size += 4;
-				reg &= ~E1000_VMOLR_RLPML_MASK;
-				reg |= size;
-				wr32(E1000_VMOLR(vf), reg);
-			}
-
-			adapter->vf_data[vf].vlans_enabled++;
 		}
 	} else {
 		if (i < E1000_VLVF_ARRAY_SIZE) {
@@ -5865,22 +5842,6 @@ static s32 igb_vlvf_set(struct igb_adapter *adapter, u32 vid, bool add, u32 vf)
 				igb_vfta_set(hw, vid, false);
 			}
 			wr32(E1000_VLVF(i), reg);
-
-			/* do not modify RLPML for PF devices */
-			if (vf >= adapter->vfs_allocated_count)
-				return 0;
-
-			adapter->vf_data[vf].vlans_enabled--;
-			if (!adapter->vf_data[vf].vlans_enabled) {
-				u32 size;
-
-				reg = rd32(E1000_VMOLR(vf));
-				size = reg & E1000_VMOLR_RLPML_MASK;
-				size -= 4;
-				reg &= ~E1000_VMOLR_RLPML_MASK;
-				reg |= size;
-				wr32(E1000_VMOLR(vf), reg);
-			}
 		}
 	}
 	return 0;
-- 
cgit v0.10.2


From 5982a5565a08abf3b9ff18941b3e3cc94f7c8286 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:10:54 -0800
Subject: igb: Always enable VLAN 0 even if 8021q is not loaded

This patch makes it so that we always add VLAN 0.  This is important as we
need to guarantee the PF can receive untagged frames in the case of SR-IOV
being enabled but VLAN filtering not being enabled in the kernel.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 9461480..f3e1738 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -7177,11 +7177,12 @@ static int igb_vlan_rx_kill_vid(struct net_device *netdev,
 
 static void igb_restore_vlan(struct igb_adapter *adapter)
 {
-	u16 vid;
+	u16 vid = 1;
 
 	igb_vlan_mode(adapter->netdev, adapter->netdev->features);
+	igb_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), 0);
 
-	for_each_set_bit(vid, adapter->active_vlans, VLAN_N_VID)
+	for_each_set_bit_from(vid, adapter->active_vlans, VLAN_N_VID)
 		igb_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), vid);
 }
 
-- 
cgit v0.10.2


From 8b77c6b20f32511175dfd00322ae82fb31949d55 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:11:04 -0800
Subject: igb: Merge VLVF configuration into igb_vfta_set

This change makes it so that we can merge the configuration of the VLVF
registers into the setting of the VFTA register.  By doing this we simplify
the logic and make use of similar functionality that we have already added
for ixgbe making it easier to maintain both drivers.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c
index 97f6fae..07cf4fe 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.c
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.c
@@ -141,21 +141,69 @@ void igb_init_rx_addrs(struct e1000_hw *hw, u16 rar_count)
 }
 
 /**
+ *  igb_find_vlvf_slot - find the VLAN id or the first empty slot
+ *  @hw: pointer to hardware structure
+ *  @vlan: VLAN id to write to VLAN filter
+ *  @vlvf_bypass: skip VLVF if no match is found
+ *
+ *  return the VLVF index where this VLAN id should be placed
+ *
+ **/
+static s32 igb_find_vlvf_slot(struct e1000_hw *hw, u32 vlan, bool vlvf_bypass)
+{
+	s32 regindex, first_empty_slot;
+	u32 bits;
+
+	/* short cut the special case */
+	if (vlan == 0)
+		return 0;
+
+	/* if vlvf_bypass is set we don't want to use an empty slot, we
+	 * will simply bypass the VLVF if there are no entries present in the
+	 * VLVF that contain our VLAN
+	 */
+	first_empty_slot = vlvf_bypass ? -E1000_ERR_NO_SPACE : 0;
+
+	/* Search for the VLAN id in the VLVF entries. Save off the first empty
+	 * slot found along the way.
+	 *
+	 * pre-decrement loop covering (IXGBE_VLVF_ENTRIES - 1) .. 1
+	 */
+	for (regindex = E1000_VLVF_ARRAY_SIZE; --regindex > 0;) {
+		bits = rd32(E1000_VLVF(regindex)) & E1000_VLVF_VLANID_MASK;
+		if (bits == vlan)
+			return regindex;
+		if (!first_empty_slot && !bits)
+			first_empty_slot = regindex;
+	}
+
+	return first_empty_slot ? : -E1000_ERR_NO_SPACE;
+}
+
+/**
  *  igb_vfta_set - enable or disable vlan in VLAN filter table
  *  @hw: pointer to the HW structure
  *  @vlan: VLAN id to add or remove
+ *  @vind: VMDq output index that maps queue to VLAN id
  *  @vlan_on: if true add filter, if false remove
  *
  *  Sets or clears a bit in the VLAN filter table array based on VLAN id
  *  and if we are adding or removing the filter
  **/
-s32 igb_vfta_set(struct e1000_hw *hw, u32 vlan, bool vlan_on)
+s32 igb_vfta_set(struct e1000_hw *hw, u32 vlan, u32 vind,
+		 bool vlan_on, bool vlvf_bypass)
 {
 	struct igb_adapter *adapter = hw->back;
-	u32 regidx, vfta_delta, vfta;
+	u32 regidx, vfta_delta, vfta, bits;
+	s32 vlvf_index;
 
-	if (vlan > 4095)
-		return E1000_ERR_PARAM;
+	if ((vlan > 4095) || (vind > 7))
+		return -E1000_ERR_PARAM;
+
+	/* this is a 2 part operation - first the VFTA, then the
+	 * VLVF and VLVFB if VT Mode is set
+	 * We don't write the VFTA until we know the VLVF part succeeded.
+	 */
 
 	/* Part 1
 	 * The VFTA is a bitstring made up of 128 32-bit registers
@@ -174,6 +222,69 @@ s32 igb_vfta_set(struct e1000_hw *hw, u32 vlan, bool vlan_on)
 	vfta_delta &= vlan_on ? ~vfta : vfta;
 	vfta ^= vfta_delta;
 
+	/* Part 2
+	 * If VT Mode is set
+	 *   Either vlan_on
+	 *     make sure the VLAN is in VLVF
+	 *     set the vind bit in the matching VLVFB
+	 *   Or !vlan_on
+	 *     clear the pool bit and possibly the vind
+	 */
+	if (!adapter->vfs_allocated_count)
+		goto vfta_update;
+
+	vlvf_index = igb_find_vlvf_slot(hw, vlan, vlvf_bypass);
+	if (vlvf_index < 0) {
+		if (vlvf_bypass)
+			goto vfta_update;
+		return vlvf_index;
+	}
+
+	bits = rd32(E1000_VLVF(vlvf_index));
+
+	/* set the pool bit */
+	bits |= 1 << (E1000_VLVF_POOLSEL_SHIFT + vind);
+	if (vlan_on)
+		goto vlvf_update;
+
+	/* clear the pool bit */
+	bits ^= 1 << (E1000_VLVF_POOLSEL_SHIFT + vind);
+
+	if (!(bits & E1000_VLVF_POOLSEL_MASK)) {
+		/* Clear VFTA first, then disable VLVF.  Otherwise
+		 * we run the risk of stray packets leaking into
+		 * the PF via the default pool
+		 */
+		if (vfta_delta)
+			hw->mac.ops.write_vfta(hw, regidx, vfta);
+
+		/* disable VLVF and clear remaining bit from pool */
+		wr32(E1000_VLVF(vlvf_index), 0);
+
+		return 0;
+	}
+
+	/* If there are still bits set in the VLVFB registers
+	 * for the VLAN ID indicated we need to see if the
+	 * caller is requesting that we clear the VFTA entry bit.
+	 * If the caller has requested that we clear the VFTA
+	 * entry bit but there are still pools/VFs using this VLAN
+	 * ID entry then ignore the request.  We're not worried
+	 * about the case where we're turning the VFTA VLAN ID
+	 * entry bit on, only when requested to turn it off as
+	 * there may be multiple pools and/or VFs using the
+	 * VLAN ID entry.  In that case we cannot clear the
+	 * VFTA bit until all pools/VFs using that VLAN ID have also
+	 * been cleared.  This will be indicated by "bits" being
+	 * zero.
+	 */
+	vfta_delta = 0;
+
+vlvf_update:
+	/* record pool change and enable VLAN ID if not already enabled */
+	wr32(E1000_VLVF(vlvf_index), bits | vlan | E1000_VLVF_VLANID_ENABLE);
+
+vfta_update:
 	/* bit was set/cleared before we started */
 	if (vfta_delta)
 		hw->mac.ops.write_vfta(hw, regidx, vfta);
diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.h b/drivers/net/ethernet/intel/igb/e1000_mac.h
index 4fbb953..90c8893 100644
--- a/drivers/net/ethernet/intel/igb/e1000_mac.h
+++ b/drivers/net/ethernet/intel/igb/e1000_mac.h
@@ -57,7 +57,8 @@ s32  igb_write_8bit_ctrl_reg(struct e1000_hw *hw, u32 reg,
 void igb_clear_hw_cntrs_base(struct e1000_hw *hw);
 void igb_clear_vfta(struct e1000_hw *hw);
 void igb_write_vfta(struct e1000_hw *hw, u32 offset, u32 value);
-s32  igb_vfta_set(struct e1000_hw *hw, u32 vid, bool add);
+s32  igb_vfta_set(struct e1000_hw *hw, u32 vid, u32 vind,
+		  bool vlan_on, bool vlvf_bypass);
 void igb_config_collision_dist(struct e1000_hw *hw);
 void igb_init_rx_addrs(struct e1000_hw *hw, u16 rar_count);
 void igb_mta_set(struct e1000_hw *hw, u32 hash_value);
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index f3e1738..e7c3a94 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -1534,12 +1534,13 @@ static void igb_irq_enable(struct igb_adapter *adapter)
 static void igb_update_mng_vlan(struct igb_adapter *adapter)
 {
 	struct e1000_hw *hw = &adapter->hw;
+	u16 pf_id = adapter->vfs_allocated_count;
 	u16 vid = adapter->hw.mng_cookie.vlan_id;
 	u16 old_vid = adapter->mng_vlan_id;
 
 	if (hw->mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN) {
 		/* add VID to filter table */
-		igb_vfta_set(hw, vid, true);
+		igb_vfta_set(hw, vid, pf_id, true, true);
 		adapter->mng_vlan_id = vid;
 	} else {
 		adapter->mng_vlan_id = IGB_MNG_VLAN_NONE;
@@ -1549,7 +1550,7 @@ static void igb_update_mng_vlan(struct igb_adapter *adapter)
 	    (vid != old_vid) &&
 	    !test_bit(old_vid, adapter->active_vlans)) {
 		/* remove VID from filter table */
-		igb_vfta_set(hw, old_vid, false);
+		igb_vfta_set(hw, vid, pf_id, false, true);
 	}
 }
 
@@ -5778,75 +5779,13 @@ static void igb_clear_vf_vfta(struct igb_adapter *adapter, u32 vf)
 		    (reg & E1000_VLVF_VLANID_ENABLE)) {
 			reg = 0;
 			vid = reg & E1000_VLVF_VLANID_MASK;
-			igb_vfta_set(hw, vid, false);
+			igb_vfta_set(hw, vid, vf, false, true);
 		}
 
 		wr32(E1000_VLVF(i), reg);
 	}
 }
 
-static s32 igb_vlvf_set(struct igb_adapter *adapter, u32 vid, bool add, u32 vf)
-{
-	struct e1000_hw *hw = &adapter->hw;
-	u32 reg, i;
-
-	/* The vlvf table only exists on 82576 hardware and newer */
-	if (hw->mac.type < e1000_82576)
-		return -1;
-
-	/* we only need to do this if VMDq is enabled */
-	if (!adapter->vfs_allocated_count)
-		return -1;
-
-	/* Find the vlan filter for this id */
-	for (i = 0; i < E1000_VLVF_ARRAY_SIZE; i++) {
-		reg = rd32(E1000_VLVF(i));
-		if ((reg & E1000_VLVF_VLANID_ENABLE) &&
-		    vid == (reg & E1000_VLVF_VLANID_MASK))
-			break;
-	}
-
-	if (add) {
-		if (i == E1000_VLVF_ARRAY_SIZE) {
-			/* Did not find a matching VLAN ID entry that was
-			 * enabled.  Search for a free filter entry, i.e.
-			 * one without the enable bit set
-			 */
-			for (i = 0; i < E1000_VLVF_ARRAY_SIZE; i++) {
-				reg = rd32(E1000_VLVF(i));
-				if (!(reg & E1000_VLVF_VLANID_ENABLE))
-					break;
-			}
-		}
-		if (i < E1000_VLVF_ARRAY_SIZE) {
-			/* Found an enabled/available entry */
-			reg |= 1 << (E1000_VLVF_POOLSEL_SHIFT + vf);
-
-			/* if !enabled we need to set this up in vfta */
-			if (!(reg & E1000_VLVF_VLANID_ENABLE)) {
-				/* add VID to filter table */
-				igb_vfta_set(hw, vid, true);
-				reg |= E1000_VLVF_VLANID_ENABLE;
-			}
-			reg &= ~E1000_VLVF_VLANID_MASK;
-			reg |= vid;
-			wr32(E1000_VLVF(i), reg);
-		}
-	} else {
-		if (i < E1000_VLVF_ARRAY_SIZE) {
-			/* remove vf from the pool */
-			reg &= ~(1 << (E1000_VLVF_POOLSEL_SHIFT + vf));
-			/* if pool is empty then remove entry from vfta */
-			if (!(reg & E1000_VLVF_POOLSEL_MASK)) {
-				reg = 0;
-				igb_vfta_set(hw, vid, false);
-			}
-			wr32(E1000_VLVF(i), reg);
-		}
-	}
-	return 0;
-}
-
 static void igb_set_vmvir(struct igb_adapter *adapter, u32 vid, u32 vf)
 {
 	struct e1000_hw *hw = &adapter->hw;
@@ -5860,13 +5799,14 @@ static void igb_set_vmvir(struct igb_adapter *adapter, u32 vid, u32 vf)
 static int igb_ndo_set_vf_vlan(struct net_device *netdev,
 			       int vf, u16 vlan, u8 qos)
 {
-	int err = 0;
 	struct igb_adapter *adapter = netdev_priv(netdev);
+	struct e1000_hw *hw = &adapter->hw;
+	int err = 0;
 
 	if ((vf >= adapter->vfs_allocated_count) || (vlan > 4095) || (qos > 7))
 		return -EINVAL;
 	if (vlan || qos) {
-		err = igb_vlvf_set(adapter, vlan, !!vlan, vf);
+		err = igb_vfta_set(hw, vlan, vf, !!vlan, false);
 		if (err)
 			goto out;
 		igb_set_vmvir(adapter, vlan | (qos << VLAN_PRIO_SHIFT), vf);
@@ -5882,8 +5822,8 @@ static int igb_ndo_set_vf_vlan(struct net_device *netdev,
 				 "Bring the PF device up before attempting to use the VF device.\n");
 		}
 	} else {
-		igb_vlvf_set(adapter, adapter->vf_data[vf].pf_vlan,
-			     false, vf);
+		igb_vfta_set(hw, adapter->vf_data[vf].pf_vlan, vf,
+			     false, false);
 		igb_set_vmvir(adapter, vlan, vf);
 		igb_set_vmolr(adapter, vf, true);
 		adapter->vf_data[vf].pf_vlan = 0;
@@ -5924,12 +5864,12 @@ static int igb_set_vf_vlan(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
 	 * the VLAN filter set.
 	 */
 	if (add && (adapter->netdev->flags & IFF_PROMISC))
-		err = igb_vlvf_set(adapter, vid, add,
-				   adapter->vfs_allocated_count);
+		err = igb_vfta_set(hw, vid, adapter->vfs_allocated_count,
+				   true, false);
 	if (err)
 		goto out;
 
-	err = igb_vlvf_set(adapter, vid, add, vf);
+	err = igb_vfta_set(hw, vid, vf, !!add, false);
 
 	if (err)
 		goto out;
@@ -5956,8 +5896,8 @@ static int igb_set_vf_vlan(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
 		if ((vlvf & VLAN_VID_MASK) == vid &&
 		    !test_bit(vid, adapter->active_vlans) &&
 		    !bits)
-			igb_vlvf_set(adapter, vid, add,
-				     adapter->vfs_allocated_count);
+			igb_vfta_set(hw, vid, adapter->vfs_allocated_count,
+				     false, false);
 	}
 
 out:
@@ -7144,12 +7084,8 @@ static int igb_vlan_rx_add_vid(struct net_device *netdev,
 	struct e1000_hw *hw = &adapter->hw;
 	int pf_id = adapter->vfs_allocated_count;
 
-	/* attempt to add filter to vlvf array */
-	igb_vlvf_set(adapter, vid, true, pf_id);
-
 	/* add the filter since PF can receive vlans w/o entry in vlvf */
-	igb_vfta_set(hw, vid, true);
-
+	igb_vfta_set(hw, vid, pf_id, true, true);
 	set_bit(vid, adapter->active_vlans);
 
 	return 0;
@@ -7159,16 +7095,11 @@ static int igb_vlan_rx_kill_vid(struct net_device *netdev,
 				__be16 proto, u16 vid)
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
-	struct e1000_hw *hw = &adapter->hw;
 	int pf_id = adapter->vfs_allocated_count;
-	s32 err;
-
-	/* remove vlan from VLVF table array */
-	err = igb_vlvf_set(adapter, vid, false, pf_id);
+	struct e1000_hw *hw = &adapter->hw;
 
-	/* if vid was not present in VLVF just remove it from table */
-	if (err)
-		igb_vfta_set(hw, vid, false);
+	/* remove VID from filter table */
+	igb_vfta_set(hw, vid, pf_id, false, true);
 
 	clear_bit(vid, adapter->active_vlans);
 
-- 
cgit v0.10.2


From a15d92598a5a741037b873fd4a43595a63048bbd Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:11:11 -0800
Subject: igb: Clean-up configuration of VF port VLANs

This patch is meant to clean-up the configuration of the VF port based VLAN
configuration.  The original logic was a bit muddled and had some
undesirable side effects such as VLANs being either completely stripped
from the port or VLANs being left when they shouldn't be.  The idea behind
this code is to avoid any events such as spurious spoof notifications when
we are removing one VLAN tag and replacing it with another.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index e7c3a94..6876ae5 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -5786,53 +5786,6 @@ static void igb_clear_vf_vfta(struct igb_adapter *adapter, u32 vf)
 	}
 }
 
-static void igb_set_vmvir(struct igb_adapter *adapter, u32 vid, u32 vf)
-{
-	struct e1000_hw *hw = &adapter->hw;
-
-	if (vid)
-		wr32(E1000_VMVIR(vf), (vid | E1000_VMVIR_VLANA_DEFAULT));
-	else
-		wr32(E1000_VMVIR(vf), 0);
-}
-
-static int igb_ndo_set_vf_vlan(struct net_device *netdev,
-			       int vf, u16 vlan, u8 qos)
-{
-	struct igb_adapter *adapter = netdev_priv(netdev);
-	struct e1000_hw *hw = &adapter->hw;
-	int err = 0;
-
-	if ((vf >= adapter->vfs_allocated_count) || (vlan > 4095) || (qos > 7))
-		return -EINVAL;
-	if (vlan || qos) {
-		err = igb_vfta_set(hw, vlan, vf, !!vlan, false);
-		if (err)
-			goto out;
-		igb_set_vmvir(adapter, vlan | (qos << VLAN_PRIO_SHIFT), vf);
-		igb_set_vmolr(adapter, vf, !vlan);
-		adapter->vf_data[vf].pf_vlan = vlan;
-		adapter->vf_data[vf].pf_qos = qos;
-		dev_info(&adapter->pdev->dev,
-			 "Setting VLAN %d, QOS 0x%x on VF %d\n", vlan, qos, vf);
-		if (test_bit(__IGB_DOWN, &adapter->state)) {
-			dev_warn(&adapter->pdev->dev,
-				 "The VF VLAN has been set, but the PF device is not up.\n");
-			dev_warn(&adapter->pdev->dev,
-				 "Bring the PF device up before attempting to use the VF device.\n");
-		}
-	} else {
-		igb_vfta_set(hw, adapter->vf_data[vf].pf_vlan, vf,
-			     false, false);
-		igb_set_vmvir(adapter, vlan, vf);
-		igb_set_vmolr(adapter, vf, true);
-		adapter->vf_data[vf].pf_vlan = 0;
-		adapter->vf_data[vf].pf_qos = 0;
-	}
-out:
-	return err;
-}
-
 static int igb_find_vlvf_entry(struct igb_adapter *adapter, int vid)
 {
 	struct e1000_hw *hw = &adapter->hw;
@@ -5853,23 +5806,25 @@ static int igb_find_vlvf_entry(struct igb_adapter *adapter, int vid)
 	return i;
 }
 
-static int igb_set_vf_vlan(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
+static s32 igb_set_vf_vlan(struct igb_adapter *adapter, u32 vid,
+			   bool add, u32 vf)
 {
+	int pf_id = adapter->vfs_allocated_count;
 	struct e1000_hw *hw = &adapter->hw;
-	int add = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >> E1000_VT_MSGINFO_SHIFT;
-	int vid = (msgbuf[1] & E1000_VLVF_VLANID_MASK);
-	int err = 0;
+	int err;
 
-	/* If in promiscuous mode we need to make sure the PF also has
-	 * the VLAN filter set.
+	/* If VLAN overlaps with one the PF is currently monitoring make
+	 * sure that we are able to allocate a VLVF entry.  This may be
+	 * redundant but it guarantees PF will maintain visibility to
+	 * the VLAN.
 	 */
-	if (add && (adapter->netdev->flags & IFF_PROMISC))
-		err = igb_vfta_set(hw, vid, adapter->vfs_allocated_count,
-				   true, false);
-	if (err)
-		goto out;
+	if (add && (adapter->netdev->flags & IFF_PROMISC)) {
+		err = igb_vfta_set(hw, vid, pf_id, true, false);
+		if (err)
+			return err;
+	}
 
-	err = igb_vfta_set(hw, vid, vf, !!add, false);
+	err = igb_vfta_set(hw, vid, vf, add, false);
 
 	if (err)
 		goto out;
@@ -5904,23 +5859,107 @@ out:
 	return err;
 }
 
-static inline void igb_vf_reset(struct igb_adapter *adapter, u32 vf)
+static void igb_set_vmvir(struct igb_adapter *adapter, u32 vid, u32 vf)
 {
-	/* clear flags - except flag that indicates PF has set the MAC */
-	adapter->vf_data[vf].flags &= IGB_VF_FLAG_PF_SET_MAC;
-	adapter->vf_data[vf].last_nack = jiffies;
+	struct e1000_hw *hw = &adapter->hw;
 
-	/* reset offloads to defaults */
+	if (vid)
+		wr32(E1000_VMVIR(vf), (vid | E1000_VMVIR_VLANA_DEFAULT));
+	else
+		wr32(E1000_VMVIR(vf), 0);
+}
+
+static int igb_enable_port_vlan(struct igb_adapter *adapter, int vf,
+				u16 vlan, u8 qos)
+{
+	int err;
+
+	err = igb_set_vf_vlan(adapter, vlan, true, vf);
+	if (err)
+		return err;
+
+	igb_set_vmvir(adapter, vlan | (qos << VLAN_PRIO_SHIFT), vf);
+	igb_set_vmolr(adapter, vf, !vlan);
+
+	/* revoke access to previous VLAN */
+	if (vlan != adapter->vf_data[vf].pf_vlan)
+		igb_set_vf_vlan(adapter, adapter->vf_data[vf].pf_vlan,
+				false, vf);
+
+	adapter->vf_data[vf].pf_vlan = vlan;
+	adapter->vf_data[vf].pf_qos = qos;
+	dev_info(&adapter->pdev->dev,
+		 "Setting VLAN %d, QOS 0x%x on VF %d\n", vlan, qos, vf);
+	if (test_bit(__IGB_DOWN, &adapter->state)) {
+		dev_warn(&adapter->pdev->dev,
+			 "The VF VLAN has been set, but the PF device is not up.\n");
+		dev_warn(&adapter->pdev->dev,
+			 "Bring the PF device up before attempting to use the VF device.\n");
+	}
+
+	return err;
+}
+
+static int igb_disable_port_vlan(struct igb_adapter *adapter, int vf)
+{
+	/* Restore tagless access via VLAN 0 */
+	igb_set_vf_vlan(adapter, 0, true, vf);
+
+	igb_set_vmvir(adapter, 0, vf);
 	igb_set_vmolr(adapter, vf, true);
 
+	/* Remove any PF assigned VLAN */
+	if (adapter->vf_data[vf].pf_vlan)
+		igb_set_vf_vlan(adapter, adapter->vf_data[vf].pf_vlan,
+				false, vf);
+
+	adapter->vf_data[vf].pf_vlan = 0;
+	adapter->vf_data[vf].pf_qos = 0;
+
+	return 0;
+}
+
+static int igb_ndo_set_vf_vlan(struct net_device *netdev,
+			       int vf, u16 vlan, u8 qos)
+{
+	struct igb_adapter *adapter = netdev_priv(netdev);
+
+	if ((vf >= adapter->vfs_allocated_count) || (vlan > 4095) || (qos > 7))
+		return -EINVAL;
+
+	return (vlan || qos) ? igb_enable_port_vlan(adapter, vf, vlan, qos) :
+			       igb_disable_port_vlan(adapter, vf);
+}
+
+static int igb_set_vf_vlan_msg(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
+{
+	int add = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >> E1000_VT_MSGINFO_SHIFT;
+	int vid = (msgbuf[1] & E1000_VLVF_VLANID_MASK);
+
+	if (adapter->vf_data[vf].pf_vlan)
+		return -1;
+
+	/* VLAN 0 is a special case, don't allow it to be removed */
+	if (!vid && !add)
+		return 0;
+
+	return igb_set_vf_vlan(adapter, vid, !!add, vf);
+}
+
+static inline void igb_vf_reset(struct igb_adapter *adapter, u32 vf)
+{
+	struct vf_data_storage *vf_data = &adapter->vf_data[vf];
+
+	/* clear flags - except flag that indicates PF has set the MAC */
+	vf_data->flags &= IGB_VF_FLAG_PF_SET_MAC;
+	vf_data->last_nack = jiffies;
+
 	/* reset vlans for device */
 	igb_clear_vf_vfta(adapter, vf);
-	if (adapter->vf_data[vf].pf_vlan)
-		igb_ndo_set_vf_vlan(adapter->netdev, vf,
-				    adapter->vf_data[vf].pf_vlan,
-				    adapter->vf_data[vf].pf_qos);
-	else
-		igb_clear_vf_vfta(adapter, vf);
+	igb_set_vf_vlan(adapter, vf_data->pf_vlan, true, vf);
+	igb_set_vmvir(adapter, vf_data->pf_vlan |
+			       (vf_data->pf_qos << VLAN_PRIO_SHIFT), vf);
+	igb_set_vmolr(adapter, vf, !vf_data->pf_vlan);
 
 	/* reset multicast table array for vf */
 	adapter->vf_data[vf].num_vf_mc_hashes = 0;
@@ -6065,7 +6104,7 @@ static void igb_rcv_msg_from_vf(struct igb_adapter *adapter, u32 vf)
 				 "VF %d attempted to override administratively set VLAN tag\nReload the VF driver to resume operations\n",
 				 vf);
 		else
-			retval = igb_set_vf_vlan(adapter, msgbuf, vf);
+			retval = igb_set_vf_vlan_msg(adapter, msgbuf, vf);
 		break;
 	default:
 		dev_err(&pdev->dev, "Unhandled Msg %08x\n", msgbuf[0]);
-- 
cgit v0.10.2


From 16903caa339961b9f8a68b64f4f313789de48599 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:11:18 -0800
Subject: igb: Add support for VLAN promiscuous with SR-IOV and NTUPLE

This change fixes things so that we can fully support SR-IOV or the
recently added NTUPLE filtering while allowing support for VLAN promiscuous
mode.  By making this change we are able to support possible scenarios such
as SR-IOV with the PF connected to a Linux bridge hosting other VMs.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index d135261..707ae5c 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -481,6 +481,7 @@ struct igb_adapter {
 #define IGB_FLAG_MAS_ENABLE		(1 << 12)
 #define IGB_FLAG_HAS_MSIX		(1 << 13)
 #define IGB_FLAG_EEE			(1 << 14)
+#define IGB_FLAG_VLAN_PROMISC		BIT(15)
 
 /* Media Auto Sense */
 #define IGB_MAS_ENABLE_0		0X0001
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 6876ae5..7366d4f 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -1819,6 +1819,10 @@ void igb_down(struct igb_adapter *adapter)
 
 	if (!pci_channel_offline(adapter->pdev))
 		igb_reset(adapter);
+
+	/* clear VLAN promisc flag so VFTA will be updated if necessary */
+	adapter->flags &= ~IGB_FLAG_VLAN_PROMISC;
+
 	igb_clean_all_tx_rings(adapter);
 	igb_clean_all_rx_rings(adapter);
 #ifdef CONFIG_IGB_DCA
@@ -2050,7 +2054,7 @@ static int igb_set_features(struct net_device *netdev,
 	if (changed & NETIF_F_HW_VLAN_CTAG_RX)
 		igb_vlan_mode(netdev, features);
 
-	if (!(changed & NETIF_F_RXALL))
+	if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE)))
 		return 0;
 
 	netdev->features = features;
@@ -3515,8 +3519,7 @@ void igb_setup_rctl(struct igb_adapter *adapter)
 			 E1000_RCTL_BAM | /* RX All Bcast Pkts */
 			 E1000_RCTL_PMCF); /* RX All MAC Ctrl Pkts */
 
-		rctl &= ~(E1000_RCTL_VFE | /* Disable VLAN filter */
-			  E1000_RCTL_DPF | /* Allow filtered pause */
+		rctl &= ~(E1000_RCTL_DPF | /* Allow filtered pause */
 			  E1000_RCTL_CFIEN); /* Dis VLAN CFIEN Filter */
 		/* Do not mess with E1000_CTRL_VME, it affects transmit as well,
 		 * and that breaks VLANs.
@@ -3967,6 +3970,130 @@ static int igb_write_uc_addr_list(struct net_device *netdev)
 	return count;
 }
 
+static int igb_vlan_promisc_enable(struct igb_adapter *adapter)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	u32 i, pf_id;
+
+	switch (hw->mac.type) {
+	case e1000_i210:
+	case e1000_i211:
+	case e1000_i350:
+		/* VLAN filtering needed for VLAN prio filter */
+		if (adapter->netdev->features & NETIF_F_NTUPLE)
+			break;
+		/* fall through */
+	case e1000_82576:
+	case e1000_82580:
+	case e1000_i354:
+		/* VLAN filtering needed for pool filtering */
+		if (adapter->vfs_allocated_count)
+			break;
+		/* fall through */
+	default:
+		return 1;
+	}
+
+	/* We are already in VLAN promisc, nothing to do */
+	if (adapter->flags & IGB_FLAG_VLAN_PROMISC)
+		return 0;
+
+	if (!adapter->vfs_allocated_count)
+		goto set_vfta;
+
+	/* Add PF to all active pools */
+	pf_id = adapter->vfs_allocated_count + E1000_VLVF_POOLSEL_SHIFT;
+
+	for (i = E1000_VLVF_ARRAY_SIZE; --i;) {
+		u32 vlvf = rd32(E1000_VLVF(i));
+
+		vlvf |= 1 << pf_id;
+		wr32(E1000_VLVF(i), vlvf);
+	}
+
+set_vfta:
+	/* Set all bits in the VLAN filter table array */
+	for (i = E1000_VLAN_FILTER_TBL_SIZE; i--;)
+		hw->mac.ops.write_vfta(hw, i, ~0U);
+
+	/* Set flag so we don't redo unnecessary work */
+	adapter->flags |= IGB_FLAG_VLAN_PROMISC;
+
+	return 0;
+}
+
+#define VFTA_BLOCK_SIZE 8
+static void igb_scrub_vfta(struct igb_adapter *adapter, u32 vfta_offset)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	u32 vfta[VFTA_BLOCK_SIZE] = { 0 };
+	u32 vid_start = vfta_offset * 32;
+	u32 vid_end = vid_start + (VFTA_BLOCK_SIZE * 32);
+	u32 i, vid, word, bits, pf_id;
+
+	/* guarantee that we don't scrub out management VLAN */
+	vid = adapter->mng_vlan_id;
+	if (vid >= vid_start && vid < vid_end)
+		vfta[(vid - vid_start) / 32] |= 1 << (vid % 32);
+
+	if (!adapter->vfs_allocated_count)
+		goto set_vfta;
+
+	pf_id = adapter->vfs_allocated_count + E1000_VLVF_POOLSEL_SHIFT;
+
+	for (i = E1000_VLVF_ARRAY_SIZE; --i;) {
+		u32 vlvf = rd32(E1000_VLVF(i));
+
+		/* pull VLAN ID from VLVF */
+		vid = vlvf & VLAN_VID_MASK;
+
+		/* only concern ourselves with a certain range */
+		if (vid < vid_start || vid >= vid_end)
+			continue;
+
+		if (vlvf & E1000_VLVF_VLANID_ENABLE) {
+			/* record VLAN ID in VFTA */
+			vfta[(vid - vid_start) / 32] |= 1 << (vid % 32);
+
+			/* if PF is part of this then continue */
+			if (test_bit(vid, adapter->active_vlans))
+				continue;
+		}
+
+		/* remove PF from the pool */
+		bits = ~(1 << pf_id);
+		bits &= rd32(E1000_VLVF(i));
+		wr32(E1000_VLVF(i), bits);
+	}
+
+set_vfta:
+	/* extract values from active_vlans and write back to VFTA */
+	for (i = VFTA_BLOCK_SIZE; i--;) {
+		vid = (vfta_offset + i) * 32;
+		word = vid / BITS_PER_LONG;
+		bits = vid % BITS_PER_LONG;
+
+		vfta[i] |= adapter->active_vlans[word] >> bits;
+
+		hw->mac.ops.write_vfta(hw, vfta_offset + i, vfta[i]);
+	}
+}
+
+static void igb_vlan_promisc_disable(struct igb_adapter *adapter)
+{
+	u32 i;
+
+	/* We are not in VLAN promisc, nothing to do */
+	if (!(adapter->flags & IGB_FLAG_VLAN_PROMISC))
+		return;
+
+	/* Set flag so we don't redo unnecessary work */
+	adapter->flags &= ~IGB_FLAG_VLAN_PROMISC;
+
+	for (i = 0; i < E1000_VLAN_FILTER_TBL_SIZE; i += VFTA_BLOCK_SIZE)
+		igb_scrub_vfta(adapter, i);
+}
+
 /**
  *  igb_set_rx_mode - Secondary Unicast, Multicast and Promiscuous mode set
  *  @netdev: network interface device structure
@@ -3981,21 +4108,13 @@ static void igb_set_rx_mode(struct net_device *netdev)
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
 	unsigned int vfn = adapter->vfs_allocated_count;
-	u32 rctl, vmolr = 0;
+	u32 rctl = 0, vmolr = 0;
 	int count;
 
 	/* Check for Promiscuous and All Multicast modes */
-	rctl = rd32(E1000_RCTL);
-
-	/* clear the effected bits */
-	rctl &= ~(E1000_RCTL_UPE | E1000_RCTL_MPE | E1000_RCTL_VFE);
-
 	if (netdev->flags & IFF_PROMISC) {
-		/* retain VLAN HW filtering if in VT mode */
-		if (adapter->vfs_allocated_count)
-			rctl |= E1000_RCTL_VFE;
-		rctl |= (E1000_RCTL_UPE | E1000_RCTL_MPE);
-		vmolr |= (E1000_VMOLR_ROPE | E1000_VMOLR_MPME);
+		rctl |= E1000_RCTL_UPE | E1000_RCTL_MPE;
+		vmolr |= E1000_VMOLR_ROPE | E1000_VMOLR_MPME;
 	} else {
 		if (netdev->flags & IFF_ALLMULTI) {
 			rctl |= E1000_RCTL_MPE;
@@ -4022,8 +4141,24 @@ static void igb_set_rx_mode(struct net_device *netdev)
 			rctl |= E1000_RCTL_UPE;
 			vmolr |= E1000_VMOLR_ROPE;
 		}
-		rctl |= E1000_RCTL_VFE;
 	}
+
+	/* enable VLAN filtering by default */
+	rctl |= E1000_RCTL_VFE;
+
+	/* disable VLAN filtering for modes that require it */
+	if ((netdev->flags & IFF_PROMISC) ||
+	    (netdev->features & NETIF_F_RXALL)) {
+		/* if we fail to set all rules then just clear VFE */
+		if (igb_vlan_promisc_enable(adapter))
+			rctl &= ~E1000_RCTL_VFE;
+	} else {
+		igb_vlan_promisc_disable(adapter);
+	}
+
+	/* update state of unicast, multicast, and VLAN filtering modes */
+	rctl |= rd32(E1000_RCTL) & ~(E1000_RCTL_UPE | E1000_RCTL_MPE |
+				     E1000_RCTL_VFE);
 	wr32(E1000_RCTL, rctl);
 
 	/* In order to support SR-IOV and eventually VMDq it is necessary to set
@@ -5762,48 +5897,98 @@ static void igb_restore_vf_multicasts(struct igb_adapter *adapter)
 static void igb_clear_vf_vfta(struct igb_adapter *adapter, u32 vf)
 {
 	struct e1000_hw *hw = &adapter->hw;
-	u32 pool_mask, reg, vid;
-	int i;
+	u32 pool_mask, vlvf_mask, i;
+
+	/* create mask for VF and other pools */
+	pool_mask = E1000_VLVF_POOLSEL_MASK;
+	vlvf_mask = 1 << (E1000_VLVF_POOLSEL_SHIFT + vf);
 
-	pool_mask = 1 << (E1000_VLVF_POOLSEL_SHIFT + vf);
+	/* drop PF from pool bits */
+	pool_mask &= ~(1 << (E1000_VLVF_POOLSEL_SHIFT +
+			     adapter->vfs_allocated_count));
 
 	/* Find the vlan filter for this id */
-	for (i = 0; i < E1000_VLVF_ARRAY_SIZE; i++) {
-		reg = rd32(E1000_VLVF(i));
+	for (i = E1000_VLVF_ARRAY_SIZE; i--;) {
+		u32 vlvf = rd32(E1000_VLVF(i));
+		u32 vfta_mask, vid, vfta;
 
 		/* remove the vf from the pool */
-		reg &= ~pool_mask;
-
-		/* if pool is empty then remove entry from vfta */
-		if (!(reg & E1000_VLVF_POOLSEL_MASK) &&
-		    (reg & E1000_VLVF_VLANID_ENABLE)) {
-			reg = 0;
-			vid = reg & E1000_VLVF_VLANID_MASK;
-			igb_vfta_set(hw, vid, vf, false, true);
-		}
+		if (!(vlvf & vlvf_mask))
+			continue;
+
+		/* clear out bit from VLVF */
+		vlvf ^= vlvf_mask;
+
+		/* if other pools are present, just remove ourselves */
+		if (vlvf & pool_mask)
+			goto update_vlvfb;
+
+		/* if PF is present, leave VFTA */
+		if (vlvf & E1000_VLVF_POOLSEL_MASK)
+			goto update_vlvf;
+
+		vid = vlvf & E1000_VLVF_VLANID_MASK;
+		vfta_mask = 1 << (vid % 32);
 
-		wr32(E1000_VLVF(i), reg);
+		/* clear bit from VFTA */
+		vfta = adapter->shadow_vfta[vid / 32];
+		if (vfta & vfta_mask)
+			hw->mac.ops.write_vfta(hw, vid / 32, vfta ^ vfta_mask);
+update_vlvf:
+		/* clear pool selection enable */
+		if (adapter->flags & IGB_FLAG_VLAN_PROMISC)
+			vlvf &= E1000_VLVF_POOLSEL_MASK;
+		else
+			vlvf = 0;
+update_vlvfb:
+		/* clear pool bits */
+		wr32(E1000_VLVF(i), vlvf);
 	}
 }
 
-static int igb_find_vlvf_entry(struct igb_adapter *adapter, int vid)
+static int igb_find_vlvf_entry(struct e1000_hw *hw, u32 vlan)
 {
-	struct e1000_hw *hw = &adapter->hw;
-	int i;
-	u32 reg;
+	u32 vlvf;
+	int idx;
 
-	/* Find the vlan filter for this id */
-	for (i = 0; i < E1000_VLVF_ARRAY_SIZE; i++) {
-		reg = rd32(E1000_VLVF(i));
-		if ((reg & E1000_VLVF_VLANID_ENABLE) &&
-		    vid == (reg & E1000_VLVF_VLANID_MASK))
+	/* short cut the special case */
+	if (vlan == 0)
+		return 0;
+
+	/* Search for the VLAN id in the VLVF entries */
+	for (idx = E1000_VLVF_ARRAY_SIZE; --idx;) {
+		vlvf = rd32(E1000_VLVF(idx));
+		if ((vlvf & VLAN_VID_MASK) == vlan)
 			break;
 	}
 
-	if (i >= E1000_VLVF_ARRAY_SIZE)
-		i = -1;
+	return idx;
+}
+
+void igb_update_pf_vlvf(struct igb_adapter *adapter, u32 vid)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	u32 bits, pf_id;
+	int idx;
+
+	idx = igb_find_vlvf_entry(hw, vid);
+	if (!idx)
+		return;
 
-	return i;
+	/* See if any other pools are set for this VLAN filter
+	 * entry other than the PF.
+	 */
+	pf_id = adapter->vfs_allocated_count + E1000_VLVF_POOLSEL_SHIFT;
+	bits = ~(1 << pf_id) & E1000_VLVF_POOLSEL_MASK;
+	bits &= rd32(E1000_VLVF(idx));
+
+	/* Disable the filter so this falls into the default pool. */
+	if (!bits) {
+		if (adapter->flags & IGB_FLAG_VLAN_PROMISC)
+			wr32(E1000_VLVF(idx), 1 << pf_id);
+		else
+			wr32(E1000_VLVF(idx), 0);
+	}
 }
 
 static s32 igb_set_vf_vlan(struct igb_adapter *adapter, u32 vid,
@@ -5818,7 +6003,7 @@ static s32 igb_set_vf_vlan(struct igb_adapter *adapter, u32 vid,
 	 * redundant but it guarantees PF will maintain visibility to
 	 * the VLAN.
 	 */
-	if (add && (adapter->netdev->flags & IFF_PROMISC)) {
+	if (add && test_bit(vid, adapter->active_vlans)) {
 		err = igb_vfta_set(hw, vid, pf_id, true, false);
 		if (err)
 			return err;
@@ -5826,36 +6011,17 @@ static s32 igb_set_vf_vlan(struct igb_adapter *adapter, u32 vid,
 
 	err = igb_vfta_set(hw, vid, vf, add, false);
 
-	if (err)
-		goto out;
+	if (add && !err)
+		return err;
 
-	/* Go through all the checks to see if the VLAN filter should
-	 * be wiped completely.
+	/* If we failed to add the VF VLAN or we are removing the VF VLAN
+	 * we may need to drop the PF pool bit in order to allow us to free
+	 * up the VLVF resources.
 	 */
-	if (!add && (adapter->netdev->flags & IFF_PROMISC)) {
-		u32 vlvf, bits;
-		int regndx = igb_find_vlvf_entry(adapter, vid);
-
-		if (regndx < 0)
-			goto out;
-		/* See if any other pools are set for this VLAN filter
-		 * entry other than the PF.
-		 */
-		vlvf = bits = rd32(E1000_VLVF(regndx));
-		bits &= 1 << (E1000_VLVF_POOLSEL_SHIFT +
-			      adapter->vfs_allocated_count);
-		/* If the filter was removed then ensure PF pool bit
-		 * is cleared if the PF only added itself to the pool
-		 * because the PF is in promiscuous mode.
-		 */
-		if ((vlvf & VLAN_VID_MASK) == vid &&
-		    !test_bit(vid, adapter->active_vlans) &&
-		    !bits)
-			igb_vfta_set(hw, vid, adapter->vfs_allocated_count,
-				     false, false);
-	}
+	if (test_bit(vid, adapter->active_vlans) ||
+	    (adapter->flags & IGB_FLAG_VLAN_PROMISC))
+		igb_update_pf_vlvf(adapter, vid);
 
-out:
 	return err;
 }
 
@@ -7124,7 +7290,9 @@ static int igb_vlan_rx_add_vid(struct net_device *netdev,
 	int pf_id = adapter->vfs_allocated_count;
 
 	/* add the filter since PF can receive vlans w/o entry in vlvf */
-	igb_vfta_set(hw, vid, pf_id, true, true);
+	if (!vid || !(adapter->flags & IGB_FLAG_VLAN_PROMISC))
+		igb_vfta_set(hw, vid, pf_id, true, !!vid);
+
 	set_bit(vid, adapter->active_vlans);
 
 	return 0;
@@ -7138,7 +7306,8 @@ static int igb_vlan_rx_kill_vid(struct net_device *netdev,
 	struct e1000_hw *hw = &adapter->hw;
 
 	/* remove VID from filter table */
-	igb_vfta_set(hw, vid, pf_id, false, true);
+	if (vid && !(adapter->flags & IGB_FLAG_VLAN_PROMISC))
+		igb_vfta_set(hw, vid, pf_id, false, true);
 
 	clear_bit(vid, adapter->active_vlans);
 
-- 
cgit v0.10.2


From 9c2f186e45faa34d5f6ff52aa84c361d4be72288 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:11:26 -0800
Subject: igb: Drop unnecessary checks in transmit path

This patch drops several checks that we dropped from ixgbe some ago.  It
should not be possible for us to be called with either of the conditional
statements returning true so we can just drop them from the hot-path.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 7366d4f..bb5be40 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -5193,16 +5193,6 @@ static netdev_tx_t igb_xmit_frame(struct sk_buff *skb,
 {
 	struct igb_adapter *adapter = netdev_priv(netdev);
 
-	if (test_bit(__IGB_DOWN, &adapter->state)) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
-	if (skb->len <= 0) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
-
 	/* The minimum packet size with TCTL.PSP set is 17 so pad the skb
 	 * in order to meet this minimum size requirement.
 	 */
-- 
cgit v0.10.2


From 268f9d33a9319bb2d4d999e264aef9c00081bba0 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:11:34 -0800
Subject: igb: Enable use of "bridge fdb add" to set unicast table entries

This change makes it so that we can use the bridge utility to add a FDB
entry for the PF to an igb port.  By doing this we can enable the VFs to
talk to virtual ports residing on top of the PF.

In addition this should also address issues with MACVLANs trying to reside
on top of the PF as well as they would have had similar issues when added
to the PF with SR-IOV enabled.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index bb5be40..e9bdad7 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2067,6 +2067,25 @@ static int igb_set_features(struct net_device *netdev,
 	return 0;
 }
 
+static int igb_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
+			   struct net_device *dev,
+			   const unsigned char *addr, u16 vid,
+			   u16 flags)
+{
+	/* guarantee we can provide a unique filter for the unicast address */
+	if (is_unicast_ether_addr(addr) || is_link_local_ether_addr(addr)) {
+		struct igb_adapter *adapter = netdev_priv(dev);
+		struct e1000_hw *hw = &adapter->hw;
+		int vfn = adapter->vfs_allocated_count;
+		int rar_entries = hw->mac.rar_entry_count - (vfn + 1);
+
+		if (netdev_uc_count(dev) >= rar_entries)
+			return -ENOMEM;
+	}
+
+	return ndo_dflt_fdb_add(ndm, tb, dev, addr, vid, flags);
+}
+
 static const struct net_device_ops igb_netdev_ops = {
 	.ndo_open		= igb_open,
 	.ndo_stop		= igb_close,
@@ -2090,6 +2109,7 @@ static const struct net_device_ops igb_netdev_ops = {
 #endif
 	.ndo_fix_features	= igb_fix_features,
 	.ndo_set_features	= igb_set_features,
+	.ndo_fdb_add		= igb_ndo_fdb_add,
 	.ndo_features_check	= passthru_features_check,
 };
 
@@ -4132,15 +4152,16 @@ static void igb_set_rx_mode(struct net_device *netdev)
 				vmolr |= E1000_VMOLR_ROMPE;
 			}
 		}
-		/* Write addresses to available RAR registers, if there is not
-		 * sufficient space to store all the addresses then enable
-		 * unicast promiscuous mode
-		 */
-		count = igb_write_uc_addr_list(netdev);
-		if (count < 0) {
-			rctl |= E1000_RCTL_UPE;
-			vmolr |= E1000_VMOLR_ROPE;
-		}
+	}
+
+	/* Write addresses to available RAR registers, if there is not
+	 * sufficient space to store all the addresses then enable
+	 * unicast promiscuous mode
+	 */
+	count = igb_write_uc_addr_list(netdev);
+	if (count < 0) {
+		rctl |= E1000_RCTL_UPE;
+		vmolr |= E1000_VMOLR_ROPE;
 	}
 
 	/* enable VLAN filtering by default */
-- 
cgit v0.10.2


From bf456abb9b82d5376e7189cca00b528dd86d1559 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 6 Jan 2016 23:11:43 -0800
Subject: igb: Add workaround for VLAN tag stripping on 82576

There was a workaround partially implemented for the 82576 that is needed
in order for VLAN tag stripping to function correctly.  The original code
had side effects that would make it so the workaround was active on all
MACs.  I have updated the code so that the workaround is enabled, but
limited to the 82576, or activated if we exceed the available unicast
addresses.

The workaround has a side effect of mirroring all of the traffic outgoing
from the VFs back to the PF.  As such it is not recommended to use the
82576 in promiscuous mode as it will take a performance hit, though this is
now consistent with the performance as seen on the out-of-tree igb driver.

I also limited the scope of the UTA bits all being set to only when the
VMOLR register is enabled.  This should limit the effects of the UTA
register so that we don't pick up any excess traffic unless promiscuous
mode has been enabled on the PF, whereas before the PF would have ended up
in something equivalent to unicast promiscuous mode with VLAN filtering
otherwise.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.c b/drivers/net/ethernet/intel/igb/e1000_82575.c
index fff5052..9a1a9c7 100644
--- a/drivers/net/ethernet/intel/igb/e1000_82575.c
+++ b/drivers/net/ethernet/intel/igb/e1000_82575.c
@@ -425,6 +425,8 @@ static s32 igb_init_mac_params_82575(struct e1000_hw *hw)
 
 	/* Set mta register count */
 	mac->mta_reg_count = 128;
+	/* Set uta register count */
+	mac->uta_reg_count = (hw->mac.type == e1000_82575) ? 0 : 128;
 	/* Set rar entry count */
 	switch (mac->type) {
 	case e1000_82576:
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index e9bdad7..af46fcf 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -140,7 +140,7 @@ static struct rtnl_link_stats64 *igb_get_stats64(struct net_device *dev,
 					  struct rtnl_link_stats64 *stats);
 static int igb_change_mtu(struct net_device *, int);
 static int igb_set_mac(struct net_device *, void *);
-static void igb_set_uta(struct igb_adapter *adapter);
+static void igb_set_uta(struct igb_adapter *adapter, bool set);
 static irqreturn_t igb_intr(int irq, void *);
 static irqreturn_t igb_intr_msi(int irq, void *);
 static irqreturn_t igb_msix_other(int irq, void *);
@@ -3670,9 +3670,6 @@ static void igb_configure_rx(struct igb_adapter *adapter)
 {
 	int i;
 
-	/* set UTA to appropriate mode */
-	igb_set_uta(adapter);
-
 	/* set the correct pool for the PF default MAC address in entry 0 */
 	igb_rar_set_qsel(adapter, adapter->hw.mac.addr, 0,
 			 adapter->vfs_allocated_count);
@@ -4134,7 +4131,11 @@ static void igb_set_rx_mode(struct net_device *netdev)
 	/* Check for Promiscuous and All Multicast modes */
 	if (netdev->flags & IFF_PROMISC) {
 		rctl |= E1000_RCTL_UPE | E1000_RCTL_MPE;
-		vmolr |= E1000_VMOLR_ROPE | E1000_VMOLR_MPME;
+		vmolr |= E1000_VMOLR_MPME;
+
+		/* enable use of UTA filter to force packets to default pool */
+		if (hw->mac.type == e1000_82576)
+			vmolr |= E1000_VMOLR_ROPE;
 	} else {
 		if (netdev->flags & IFF_ALLMULTI) {
 			rctl |= E1000_RCTL_MPE;
@@ -4190,6 +4191,9 @@ static void igb_set_rx_mode(struct net_device *netdev)
 	if ((hw->mac.type < e1000_82576) || (hw->mac.type > e1000_i350))
 		return;
 
+	/* set UTA to appropriate mode */
+	igb_set_uta(adapter, !!(vmolr & E1000_VMOLR_ROPE));
+
 	vmolr |= rd32(E1000_VMOLR(vfn)) &
 		 ~(E1000_VMOLR_ROPE | E1000_VMOLR_MPME | E1000_VMOLR_ROMPE);
 
@@ -6323,6 +6327,7 @@ static void igb_msg_task(struct igb_adapter *adapter)
 /**
  *  igb_set_uta - Set unicast filter table address
  *  @adapter: board private structure
+ *  @set: boolean indicating if we are setting or clearing bits
  *
  *  The unicast table address is a register array of 32-bit registers.
  *  The table is meant to be used in a way similar to how the MTA is used
@@ -6330,21 +6335,18 @@ static void igb_msg_task(struct igb_adapter *adapter)
  *  set all the hash bits to 1 and use the VMOLR ROPE bit as a promiscuous
  *  enable bit to allow vlan tag stripping when promiscuous mode is enabled
  **/
-static void igb_set_uta(struct igb_adapter *adapter)
+static void igb_set_uta(struct igb_adapter *adapter, bool set)
 {
 	struct e1000_hw *hw = &adapter->hw;
+	u32 uta = set ? ~0 : 0;
 	int i;
 
-	/* The UTA table only exists on 82576 hardware and newer */
-	if (hw->mac.type < e1000_82576)
-		return;
-
 	/* we only need to do this if VMDq is enabled */
 	if (!adapter->vfs_allocated_count)
 		return;
 
-	for (i = 0; i < hw->mac.uta_reg_count; i++)
-		array_wr32(E1000_UTA, i, ~0);
+	for (i = hw->mac.uta_reg_count; i--;)
+		array_wr32(E1000_UTA, i, uta);
 }
 
 /**
-- 
cgit v0.10.2


From 3ccc30552e626934dcbaea7bfa82a4d9eb84ba6d Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Sat, 6 Feb 2016 13:09:36 -0800
Subject: net: phy: bcm7xxx: Make MII_BCM7XX_64CLK_MDIO naming consistent

The driver is BCM7xxx, we were missing an additional X in the constant naming,
fix that to be consistent.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c
index bf241a3..6215185 100644
--- a/drivers/net/phy/bcm7xxx.c
+++ b/drivers/net/phy/bcm7xxx.c
@@ -24,7 +24,7 @@
 #define MII_BCM7XXX_100TX_FALSE_CAR	0x13
 #define MII_BCM7XXX_100TX_DISC		0x14
 #define MII_BCM7XXX_AUX_MODE		0x1d
-#define  MII_BCM7XX_64CLK_MDIO		BIT(12)
+#define  MII_BCM7XXX_64CLK_MDIO		BIT(12)
 #define MII_BCM7XXX_TEST		0x1f
 #define  MII_BCM7XXX_SHD_MODE_2		BIT(2)
 
@@ -247,7 +247,7 @@ static int bcm7xxx_config_init(struct phy_device *phydev)
 	int ret;
 
 	/* Enable 64 clock MDIO */
-	phy_write(phydev, MII_BCM7XXX_AUX_MODE, MII_BCM7XX_64CLK_MDIO);
+	phy_write(phydev, MII_BCM7XXX_AUX_MODE, MII_BCM7XXX_64CLK_MDIO);
 	phy_read(phydev, MII_BCM7XXX_AUX_MODE);
 
 	/* Workaround only required for 100Mbits/sec capable PHYs */
-- 
cgit v0.10.2


From 3125c081a5937c548a6673cd7309d688c8c47b7c Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Sat, 6 Feb 2016 13:09:37 -0800
Subject: net: phy: bcm7xxx: Reduce boilerplate code for 40nm EPHY

Introduce a macro which helps adding new 40NM EPHY entries and reduces the
amount of boilerplate code.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c
index 6215185..9b31104 100644
--- a/drivers/net/phy/bcm7xxx.c
+++ b/drivers/net/phy/bcm7xxx.c
@@ -326,6 +326,21 @@ static int bcm7xxx_dummy_config_init(struct phy_device *phydev)
 	.resume		= bcm7xxx_28nm_resume,				\
 }
 
+#define BCM7XXX_40NM_EPHY(_oui, _name)					\
+{									\
+	.phy_id         = (_oui),					\
+	.phy_id_mask    = 0xfffffff0,					\
+	.name           = _name,					\
+	.features       = PHY_BASIC_FEATURES |				\
+			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,	\
+	.flags          = PHY_IS_INTERNAL,				\
+	.config_init    = bcm7xxx_config_init,				\
+	.config_aneg    = genphy_config_aneg,				\
+	.read_status    = genphy_read_status,				\
+	.suspend        = bcm7xxx_suspend,				\
+	.resume         = bcm7xxx_config_init,				\
+}
+
 static struct phy_driver bcm7xxx_driver[] = {
 	BCM7XXX_28NM_GPHY(PHY_ID_BCM7250, "Broadcom BCM7250"),
 	BCM7XXX_28NM_GPHY(PHY_ID_BCM7364, "Broadcom BCM7364"),
@@ -333,43 +348,10 @@ static struct phy_driver bcm7xxx_driver[] = {
 	BCM7XXX_28NM_GPHY(PHY_ID_BCM7439, "Broadcom BCM7439"),
 	BCM7XXX_28NM_GPHY(PHY_ID_BCM7439_2, "Broadcom BCM7439 (2)"),
 	BCM7XXX_28NM_GPHY(PHY_ID_BCM7445, "Broadcom BCM7445"),
+	BCM7XXX_40NM_EPHY(PHY_ID_BCM7425, "Broadcom BCM7425"),
+	BCM7XXX_40NM_EPHY(PHY_ID_BCM7429, "Broadcom BCM7429"),
+	BCM7XXX_40NM_EPHY(PHY_ID_BCM7435, "Broadcom BCM7435"),
 {
-	.phy_id         = PHY_ID_BCM7425,
-	.phy_id_mask    = 0xfffffff0,
-	.name           = "Broadcom BCM7425",
-	.features       = PHY_GBIT_FEATURES |
-			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
-	.flags          = PHY_IS_INTERNAL,
-	.config_init    = bcm7xxx_config_init,
-	.config_aneg    = genphy_config_aneg,
-	.read_status    = genphy_read_status,
-	.suspend        = bcm7xxx_suspend,
-	.resume         = bcm7xxx_config_init,
-}, {
-	.phy_id         = PHY_ID_BCM7429,
-	.phy_id_mask    = 0xfffffff0,
-	.name           = "Broadcom BCM7429",
-	.features       = PHY_GBIT_FEATURES |
-			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
-	.flags          = PHY_IS_INTERNAL,
-	.config_init    = bcm7xxx_config_init,
-	.config_aneg    = genphy_config_aneg,
-	.read_status    = genphy_read_status,
-	.suspend        = bcm7xxx_suspend,
-	.resume         = bcm7xxx_config_init,
-}, {
-	.phy_id         = PHY_ID_BCM7435,
-	.phy_id_mask    = 0xfffffff0,
-	.name           = "Broadcom BCM7435",
-	.features       = PHY_GBIT_FEATURES |
-			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
-	.flags          = PHY_IS_INTERNAL,
-	.config_init    = bcm7xxx_config_init,
-	.config_aneg    = genphy_config_aneg,
-	.read_status    = genphy_read_status,
-	.suspend        = bcm7xxx_suspend,
-	.resume         = bcm7xxx_config_init,
-}, {
 	.phy_id		= PHY_BCM_OUI_4,
 	.phy_id_mask	= 0xffff0000,
 	.name		= "Broadcom BCM7XXX 40nm",
-- 
cgit v0.10.2


From 64f10f6ebf5a6d3aef106af43cf697682d022f48 Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bernhard@bwalle.de>
Date: Mon, 8 Feb 2016 21:21:13 +0100
Subject: net: fec: Add "phy-reset-active-low" property to DT

We need that for a custom hardware that needs the reverse reset
sequence.

Signed-off-by: Bernhard Walle <bernhard@bwalle.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/fsl-fec.txt b/Documentation/devicetree/bindings/net/fsl-fec.txt
index a9eb611..a4799ff 100644
--- a/Documentation/devicetree/bindings/net/fsl-fec.txt
+++ b/Documentation/devicetree/bindings/net/fsl-fec.txt
@@ -12,6 +12,9 @@ Optional properties:
   only if property "phy-reset-gpios" is available.  Missing the property
   will have the duration be 1 millisecond.  Numbers greater than 1000 are
   invalid and 1 millisecond will be used instead.
+- phy-reset-active-low : If present then the reset sequence using the GPIO
+  specified in the "phy-reset-gpios" property is reversed (H=reset state,
+  L=operation state).
 - phy-supply : regulator that powers the Ethernet PHY.
 - phy-handle : phandle to the PHY device connected to this device.
 - fixed-link : Assume a fixed link. See fixed-link.txt in the same directory.
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index ef18ca5..bad0ba2 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -3191,6 +3191,7 @@ static int fec_enet_init(struct net_device *ndev)
 static void fec_reset_phy(struct platform_device *pdev)
 {
 	int err, phy_reset;
+	bool active_low = false;
 	int msec = 1;
 	struct device_node *np = pdev->dev.of_node;
 
@@ -3206,14 +3207,17 @@ static void fec_reset_phy(struct platform_device *pdev)
 	if (!gpio_is_valid(phy_reset))
 		return;
 
+	active_low = of_property_read_bool(np, "phy-reset-active-low");
+
 	err = devm_gpio_request_one(&pdev->dev, phy_reset,
-				    GPIOF_OUT_INIT_LOW, "phy-reset");
+			active_low ? GPIOF_OUT_INIT_HIGH : GPIOF_OUT_INIT_LOW,
+			"phy-reset");
 	if (err) {
 		dev_err(&pdev->dev, "failed to get phy-reset-gpios: %d\n", err);
 		return;
 	}
 	msleep(msec);
-	gpio_set_value_cansleep(phy_reset, 1);
+	gpio_set_value_cansleep(phy_reset, !active_low);
 }
 #else /* CONFIG_OF */
 static void fec_reset_phy(struct platform_device *pdev)
-- 
cgit v0.10.2


From d4ab4286276fcd6c155bafdf4422b712068d2516 Mon Sep 17 00:00:00 2001
From: "Keller, Jacob E" <jacob.e.keller@intel.com>
Date: Mon, 8 Feb 2016 16:05:03 -0800
Subject: ethtool: correctly ensure {GS}CHANNELS doesn't conflict with GS{RXFH}

Ethernet drivers implementing both {GS}RXFH and {GS}CHANNELS ethtool ops
incorrectly allow SCHANNELS when it would conflict with the settings
from SRXFH. This occurs because it is not possible for drivers to
understand whether their Rx flow indirection table has been configured
or is in the default state. In addition, drivers currently behave in
various ways when increasing the number of Rx channels.

Some drivers will always destroy the Rx flow indirection table when this
occurs, whether it has been set by the user or not. Other drivers will
attempt to preserve the table even if the user has never modified it
from the default driver settings. Neither of these situation is
desirable because it leads to unexpected behavior or loss of user
configuration.

The correct behavior is to simply return -EINVAL when SCHANNELS would
conflict with the current Rx flow table settings. However, it should
only do so if the current settings were modified by the user. If we
required that the new settings never conflict with the current (default)
Rx flow settings, we would force users to first reduce their Rx flow
settings and then reduce the number of Rx channels.

This patch proposes a solution implemented in net/core/ethtool.c which
ensures that all drivers behave correctly. It checks whether the RXFH
table has been configured to non-default settings, and stores this
information in a private netdev flag. When the number of channels is
requested to change, it first ensures that the current Rx flow table is
not going to assign flows to now disabled channels.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 219f53c..0499569 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1291,6 +1291,7 @@ struct net_device_ops {
  * @IFF_OPENVSWITCH: device is a Open vSwitch master
  * @IFF_L3MDEV_SLAVE: device is enslaved to an L3 master device
  * @IFF_TEAM: device is a team device
+ * @IFF_RXFH_CONFIGURED: device has had Rx Flow indirection table configured
  */
 enum netdev_priv_flags {
 	IFF_802_1Q_VLAN			= 1<<0,
@@ -1318,6 +1319,7 @@ enum netdev_priv_flags {
 	IFF_OPENVSWITCH			= 1<<22,
 	IFF_L3MDEV_SLAVE		= 1<<23,
 	IFF_TEAM			= 1<<24,
+	IFF_RXFH_CONFIGURED		= 1<<25,
 };
 
 #define IFF_802_1Q_VLAN			IFF_802_1Q_VLAN
@@ -1345,6 +1347,7 @@ enum netdev_priv_flags {
 #define IFF_OPENVSWITCH			IFF_OPENVSWITCH
 #define IFF_L3MDEV_SLAVE		IFF_L3MDEV_SLAVE
 #define IFF_TEAM			IFF_TEAM
+#define IFF_RXFH_CONFIGURED		IFF_RXFH_CONFIGURED
 
 /**
  *	struct net_device - The DEVICE structure.
@@ -4048,6 +4051,11 @@ static inline bool netif_is_lag_port(const struct net_device *dev)
 	return netif_is_bond_slave(dev) || netif_is_team_port(dev);
 }
 
+static inline bool netif_is_rxfh_configured(const struct net_device *dev)
+{
+	return dev->priv_flags & IFF_RXFH_CONFIGURED;
+}
+
 /* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */
 static inline void netif_keep_dst(struct net_device *dev)
 {
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 453c803..379bdc5 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -642,6 +642,37 @@ void netdev_rss_key_fill(void *buffer, size_t len)
 }
 EXPORT_SYMBOL(netdev_rss_key_fill);
 
+static int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max)
+{
+	u32 dev_size, current_max = 0;
+	u32 *indir;
+	int ret;
+
+	if (!dev->ethtool_ops->get_rxfh_indir_size ||
+	    !dev->ethtool_ops->get_rxfh)
+		return -EOPNOTSUPP;
+	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
+	if (dev_size == 0)
+		return -EOPNOTSUPP;
+
+	indir = kcalloc(dev_size, sizeof(indir[0]), GFP_USER);
+	if (!indir)
+		return -ENOMEM;
+
+	ret = dev->ethtool_ops->get_rxfh(dev, indir, NULL, NULL);
+	if (ret)
+		goto out;
+
+	while (dev_size--)
+		current_max = max(current_max, indir[dev_size]);
+
+	*max = current_max;
+
+out:
+	kfree(indir);
+	return ret;
+}
+
 static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev,
 						     void __user *useraddr)
 {
@@ -738,6 +769,14 @@ static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev,
 	}
 
 	ret = ops->set_rxfh(dev, indir, NULL, ETH_RSS_HASH_NO_CHANGE);
+	if (ret)
+		goto out;
+
+	/* indicate whether rxfh was set to default */
+	if (user_size == 0)
+		dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
+	else
+		dev->priv_flags |= IFF_RXFH_CONFIGURED;
 
 out:
 	kfree(indir);
@@ -897,6 +936,14 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	}
 
 	ret = ops->set_rxfh(dev, indir, hkey, rxfh.hfunc);
+	if (ret)
+		goto out;
+
+	/* indicate whether rxfh was set to default */
+	if (rxfh.indir_size == 0)
+		dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
+	else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
+		dev->priv_flags |= IFF_RXFH_CONFIGURED;
 
 out:
 	kfree(rss_config);
@@ -1228,6 +1275,7 @@ static noinline_for_stack int ethtool_set_channels(struct net_device *dev,
 						   void __user *useraddr)
 {
 	struct ethtool_channels channels;
+	u32 max_rx_in_use = 0;
 
 	if (!dev->ethtool_ops->set_channels)
 		return -EOPNOTSUPP;
@@ -1235,6 +1283,13 @@ static noinline_for_stack int ethtool_set_channels(struct net_device *dev,
 	if (copy_from_user(&channels, useraddr, sizeof(channels)))
 		return -EFAULT;
 
+	/* ensure the new Rx count fits within the configured Rx flow
+	 * indirection table settings */
+	if (netif_is_rxfh_configured(dev) &&
+	    !ethtool_get_max_rxfh_channel(dev, &max_rx_in_use) &&
+	    (channels.combined_count + channels.rx_count) <= max_rx_in_use)
+	    return -EINVAL;
+
 	return dev->ethtool_ops->set_channels(dev, &channels);
 }
 
-- 
cgit v0.10.2


From 8bf3686204861d39803797ebbd1e264442421907 Mon Sep 17 00:00:00 2001
From: "Keller, Jacob E" <jacob.e.keller@intel.com>
Date: Mon, 8 Feb 2016 16:05:04 -0800
Subject: ethtool: ensure channel counts are within bounds during SCHANNELS

Add a sanity check to ensure that all requested channel sizes are within
bounds, which should reduce errors in driver implementation.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 379bdc5..65f907a 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -1274,15 +1274,24 @@ static noinline_for_stack int ethtool_get_channels(struct net_device *dev,
 static noinline_for_stack int ethtool_set_channels(struct net_device *dev,
 						   void __user *useraddr)
 {
-	struct ethtool_channels channels;
+	struct ethtool_channels channels, max;
 	u32 max_rx_in_use = 0;
 
-	if (!dev->ethtool_ops->set_channels)
+	if (!dev->ethtool_ops->set_channels || !dev->ethtool_ops->get_channels)
 		return -EOPNOTSUPP;
 
 	if (copy_from_user(&channels, useraddr, sizeof(channels)))
 		return -EFAULT;
 
+	dev->ethtool_ops->get_channels(dev, &max);
+
+	/* ensure new counts are within the maximums */
+	if ((channels.rx_count > max.max_rx) ||
+	    (channels.tx_count > max.max_tx) ||
+	    (channels.combined_count > max.max_combined) ||
+	    (channels.other_count > max.max_other))
+		return -EINVAL;
+
 	/* ensure the new Rx count fits within the configured Rx flow
 	 * indirection table settings */
 	if (netif_is_rxfh_configured(dev) &&
-- 
cgit v0.10.2


From 1012014ef523510321728daac8e07bd70650b793 Mon Sep 17 00:00:00 2001
From: "Keller, Jacob E" <jacob.e.keller@intel.com>
Date: Mon, 8 Feb 2016 16:05:05 -0800
Subject: fm10k: don't reinitialize RSS flow table when RXFH configured

Also print an error message incase we do have to reconfigure as this
should no longer happen anymore due to ethtool changes. If it somehow
does occur, user should be made aware of it.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
index b243c3c..38f558e 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -1937,8 +1937,10 @@ static void fm10k_init_reta(struct fm10k_intfc *interface)
 	u16 i, rss_i = interface->ring_feature[RING_F_RSS].indices;
 	u32 reta, base;
 
-	/* If the netdev is initialized we have to maintain table if possible */
-	if (interface->netdev->reg_state != NETREG_UNINITIALIZED) {
+	/* If the Rx flow indirection table has been configured manually, we
+	 * need to maintain it when possible.
+	 */
+	if (netif_is_rxfh_configured(interface->netdev)) {
 		for (i = FM10K_RETA_SIZE; i--;) {
 			reta = interface->reta[i];
 			if ((((reta << 24) >> 24) < rss_i) &&
@@ -1946,6 +1948,10 @@ static void fm10k_init_reta(struct fm10k_intfc *interface)
 			    (((reta <<  8) >> 24) < rss_i) &&
 			    (((reta)       >> 24) < rss_i))
 				continue;
+
+			/* this should never happen */
+			dev_err(&interface->pdev->dev,
+				"RSS indirection table assigned flows out of queue bounds. Reconfiguring.\n");
 			goto repopulate_reta;
 		}
 
-- 
cgit v0.10.2


From 37ace20a3c99c54ebffb4b13671a01adb20926ca Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 10 Feb 2016 00:12:36 +0530
Subject: dmascc: Return correct error codes

This change has been made with the goal that kernel functions should
return something more descriptive than -1 on failure.

A variable `err` has been introduced for storing error codes.

The return value of kzalloc on failure should return a -1 and not a
-ENOMEM. This was found using Coccinelle. A simplified version of
the semantic patch used is:

//<smpl>
@@
expression *e;
identifier l1;
@@

e = kzalloc(...);
if (e == NULL) {
...
goto l1;
}
l1:
...
return -1
+ -ENOMEM
;
//</smpl

Furthermore, set `err` to -ENOMEM on failure of alloc_netdev(), and to
-ENODEV on failure of register_netdev() and probe_irq_off().

The single call site only checks that the return value is not 0,
hence no change is required at the call site.

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/hamradio/dmascc.c b/drivers/net/hamradio/dmascc.c
index c3d3777..e4137c1 100644
--- a/drivers/net/hamradio/dmascc.c
+++ b/drivers/net/hamradio/dmascc.c
@@ -451,7 +451,7 @@ static const struct net_device_ops scc_netdev_ops = {
 
 static int __init setup_adapter(int card_base, int type, int n)
 {
-	int i, irq, chip;
+	int i, irq, chip, err;
 	struct scc_info *info;
 	struct net_device *dev;
 	struct scc_priv *priv;
@@ -463,14 +463,17 @@ static int __init setup_adapter(int card_base, int type, int n)
 
 	/* Initialize what is necessary for write_scc and write_scc_data */
 	info = kzalloc(sizeof(struct scc_info), GFP_KERNEL | GFP_DMA);
-	if (!info)
+	if (!info) {
+		err = -ENOMEM;
 		goto out;
+	}
 
 	info->dev[0] = alloc_netdev(0, "", NET_NAME_UNKNOWN, dev_setup);
 	if (!info->dev[0]) {
 		printk(KERN_ERR "dmascc: "
 		       "could not allocate memory for %s at %#3x\n",
 		       hw[type].name, card_base);
+		err = -ENOMEM;
 		goto out1;
 	}
 
@@ -479,6 +482,7 @@ static int __init setup_adapter(int card_base, int type, int n)
 		printk(KERN_ERR "dmascc: "
 		       "could not allocate memory for %s at %#3x\n",
 		       hw[type].name, card_base);
+		err = -ENOMEM;
 		goto out2;
 	}
 	spin_lock_init(&info->register_lock);
@@ -549,6 +553,7 @@ static int __init setup_adapter(int card_base, int type, int n)
 		printk(KERN_ERR
 		       "dmascc: could not find irq of %s at %#3x (irq=%d)\n",
 		       hw[type].name, card_base, irq);
+		err = -ENODEV;
 		goto out3;
 	}
 
@@ -585,11 +590,13 @@ static int __init setup_adapter(int card_base, int type, int n)
 	if (register_netdev(info->dev[0])) {
 		printk(KERN_ERR "dmascc: could not register %s\n",
 		       info->dev[0]->name);
+		err = -ENODEV;
 		goto out3;
 	}
 	if (register_netdev(info->dev[1])) {
 		printk(KERN_ERR "dmascc: could not register %s\n",
 		       info->dev[1]->name);
+		err = -ENODEV;
 		goto out4;
 	}
 
@@ -612,7 +619,7 @@ static int __init setup_adapter(int card_base, int type, int n)
       out1:
 	kfree(info);
       out:
-	return -1;
+	return err;
 }
 
 
-- 
cgit v0.10.2


From 4952cd3e7b47dfe8f7d6c69973b13eb487eb2bd0 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Thu, 11 Feb 2016 10:43:15 +0100
Subject: tipc: refactor node xmit and fix memory leaks

Refactor tipc_node_xmit() to fail fast and fail early. Fix several
potential memory leaks in unexpected error paths.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/link.c b/net/tipc/link.c
index 6f4a6d9..3e513da 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -903,8 +903,10 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
 		if (unlikely(l->backlog[i].len >= l->backlog[i].limit))
 			return link_schedule_user(l, list);
 	}
-	if (unlikely(msg_size(hdr) > mtu))
+	if (unlikely(msg_size(hdr) > mtu)) {
+		skb_queue_purge(list);
 		return -EMSGSIZE;
+	}
 
 	/* Prepare each packet for sending, and add to relevant queue: */
 	while (skb_queue_len(list)) {
@@ -916,8 +918,10 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
 
 		if (likely(skb_queue_len(transmq) < maxwin)) {
 			_skb = skb_clone(skb, GFP_ATOMIC);
-			if (!_skb)
+			if (!_skb) {
+				skb_queue_purge(list);
 				return -ENOBUFS;
+			}
 			__skb_dequeue(list);
 			__skb_queue_tail(transmq, skb);
 			__skb_queue_tail(xmitq, _skb);
diff --git a/net/tipc/node.c b/net/tipc/node.c
index f8a8255..10a1e87 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -1166,7 +1166,7 @@ msg_full:
  * @dnode: address of destination node
  * @selector: a number used for deterministic link selection
  * Consumes the buffer chain, except when returning -ELINKCONG
- * Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE
+ * Returns 0 if success, otherwise: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE,-ENOBUF
  */
 int tipc_node_xmit(struct net *net, struct sk_buff_head *list,
 		   u32 dnode, int selector)
@@ -1174,33 +1174,43 @@ int tipc_node_xmit(struct net *net, struct sk_buff_head *list,
 	struct tipc_link_entry *le = NULL;
 	struct tipc_node *n;
 	struct sk_buff_head xmitq;
-	int bearer_id = -1;
-	int rc = -EHOSTUNREACH;
+	int bearer_id;
+	int rc;
+
+	if (in_own_node(net, dnode)) {
+		tipc_sk_rcv(net, list);
+		return 0;
+	}
 
-	__skb_queue_head_init(&xmitq);
 	n = tipc_node_find(net, dnode);
-	if (likely(n)) {
-		tipc_node_read_lock(n);
-		bearer_id = n->active_links[selector & 1];
-		if (bearer_id >= 0) {
-			le = &n->links[bearer_id];
-			spin_lock_bh(&le->lock);
-			rc = tipc_link_xmit(le->link, list, &xmitq);
-			spin_unlock_bh(&le->lock);
-		}
+	if (unlikely(!n)) {
+		skb_queue_purge(list);
+		return -EHOSTUNREACH;
+	}
+
+	tipc_node_read_lock(n);
+	bearer_id = n->active_links[selector & 1];
+	if (unlikely(bearer_id == INVALID_BEARER_ID)) {
 		tipc_node_read_unlock(n);
-		if (likely(!rc))
-			tipc_bearer_xmit(net, bearer_id, &xmitq, &le->maddr);
-		else if (rc == -ENOBUFS)
-			tipc_node_link_down(n, bearer_id, false);
 		tipc_node_put(n);
-		return rc;
+		skb_queue_purge(list);
+		return -EHOSTUNREACH;
 	}
 
-	if (likely(in_own_node(net, dnode))) {
-		tipc_sk_rcv(net, list);
-		return 0;
-	}
+	__skb_queue_head_init(&xmitq);
+	le = &n->links[bearer_id];
+	spin_lock_bh(&le->lock);
+	rc = tipc_link_xmit(le->link, list, &xmitq);
+	spin_unlock_bh(&le->lock);
+	tipc_node_read_unlock(n);
+
+	if (likely(rc == 0))
+		tipc_bearer_xmit(net, bearer_id, &xmitq, &le->maddr);
+	else if (rc == -ENOBUFS)
+		tipc_node_link_down(n, bearer_id, false);
+
+	tipc_node_put(n);
+
 	return rc;
 }
 
-- 
cgit v0.10.2


From 906807793092f7d659a8933ddd7b9fa97cd6cf39 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:41 -0800
Subject: i40e: AQ Shared resource flags

Add flags to MAC allocation requests to signify that the MAC VLAN filters
should come from the shared resource pool rather than the dedicated PF
resource pools.

Change-ID: I4c2da64c01856edcb0982bc4aab75c5a91047a7a
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 0e608d2..6e0834e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -976,6 +976,7 @@ struct i40e_aqc_add_macvlan_element_data {
 #define I40E_AQC_MACVLAN_ADD_HASH_MATCH		0x0002
 #define I40E_AQC_MACVLAN_ADD_IGNORE_VLAN	0x0004
 #define I40E_AQC_MACVLAN_ADD_TO_QUEUE		0x0008
+#define I40E_AQC_MACVLAN_ADD_USE_SHARED_MAC	0x0010
 	__le16	queue_number;
 #define I40E_AQC_MACVLAN_CMD_QUEUE_SHIFT	0
 #define I40E_AQC_MACVLAN_CMD_QUEUE_MASK		(0x7FF << \
@@ -1265,6 +1266,10 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
 
+#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_MAC	0x2000
+#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_INNER_MAC	0x4000
+#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_IP	0x8000
+
 	__le32	tenant_id;
 	u8	reserved[4];
 	__le16	queue_number;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 578b178..220b972 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -973,6 +973,7 @@ struct i40e_aqc_add_macvlan_element_data {
 #define I40E_AQC_MACVLAN_ADD_HASH_MATCH		0x0002
 #define I40E_AQC_MACVLAN_ADD_IGNORE_VLAN	0x0004
 #define I40E_AQC_MACVLAN_ADD_TO_QUEUE		0x0008
+#define I40E_AQC_MACVLAN_ADD_USE_SHARED_MAC	0x0010
 	__le16	queue_number;
 #define I40E_AQC_MACVLAN_CMD_QUEUE_SHIFT	0
 #define I40E_AQC_MACVLAN_CMD_QUEUE_MASK		(0x7FF << \
@@ -1262,6 +1263,10 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
 
+#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_MAC	0x2000
+#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_INNER_MAC	0x4000
+#define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_IP	0x8000
+
 	__le32	tenant_id;
 	u8	reserved[4];
 	__le16	queue_number;
-- 
cgit v0.10.2


From fa5623a6e6807baa7f3b8e65896199e23e3f939e Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:42 -0800
Subject: i40e: AQ Add set_switch_config

Add the new Set Switch Config AdminQ command, and mark the L2 Filter
bit as deprecated in the Add VEB command.

Change-ID: I5b24790f14c56f0ddf3f70df1e486844146b039f
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 6e0834e..7e96615 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -145,6 +145,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_remove_statistics		= 0x0202,
 	i40e_aqc_opc_set_port_parameters	= 0x0203,
 	i40e_aqc_opc_get_switch_resource_alloc	= 0x0204,
+	i40e_aqc_opc_set_switch_config		= 0x0205,
 
 	i40e_aqc_opc_add_vsi			= 0x0210,
 	i40e_aqc_opc_update_vsi_parameters	= 0x0211,
@@ -683,6 +684,17 @@ struct i40e_aqc_switch_resource_alloc_element_resp {
 
 I40E_CHECK_STRUCT_LEN(0x10, i40e_aqc_switch_resource_alloc_element_resp);
 
+/* Set Switch Configuration (direct 0x0205) */
+struct i40e_aqc_set_switch_config {
+	__le16	flags;
+#define I40E_AQ_SET_SWITCH_CFG_PROMISC		0x0001
+#define I40E_AQ_SET_SWITCH_CFG_L2_FILTER	0x0002
+	__le16	valid_flags;
+	u8	reserved[12];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config);
+
 /* Add VSI (indirect 0x0210)
  *    this indirect command uses struct i40e_aqc_vsi_properties_data
  *    as the indirect buffer (128 bytes)
@@ -909,7 +921,8 @@ struct i40e_aqc_add_veb {
 					I40E_AQC_ADD_VEB_PORT_TYPE_SHIFT)
 #define I40E_AQC_ADD_VEB_PORT_TYPE_DEFAULT	0x2
 #define I40E_AQC_ADD_VEB_PORT_TYPE_DATA		0x4
-#define I40E_AQC_ADD_VEB_ENABLE_L2_FILTER	0x8
+#define I40E_AQC_ADD_VEB_ENABLE_L2_FILTER	0x8     /* deprecated */
+#define I40E_AQC_ADD_VEB_ENABLE_DISABLE_STATS	0x10
 	u8	enable_tcs;
 	u8	reserved[9];
 };
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 220b972..2c8f3ed 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -145,6 +145,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_remove_statistics		= 0x0202,
 	i40e_aqc_opc_set_port_parameters	= 0x0203,
 	i40e_aqc_opc_get_switch_resource_alloc	= 0x0204,
+	i40e_aqc_opc_set_switch_config		= 0x0205,
 
 	i40e_aqc_opc_add_vsi			= 0x0210,
 	i40e_aqc_opc_update_vsi_parameters	= 0x0211,
@@ -680,6 +681,17 @@ struct i40e_aqc_switch_resource_alloc_element_resp {
 
 I40E_CHECK_STRUCT_LEN(0x10, i40e_aqc_switch_resource_alloc_element_resp);
 
+/* Set Switch Configuration (direct 0x0205) */
+struct i40e_aqc_set_switch_config {
+	__le16	flags;
+#define I40E_AQ_SET_SWITCH_CFG_PROMISC		0x0001
+#define I40E_AQ_SET_SWITCH_CFG_L2_FILTER	0x0002
+	__le16	valid_flags;
+	u8	reserved[12];
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config);
+
 /* Add VSI (indirect 0x0210)
  *    this indirect command uses struct i40e_aqc_vsi_properties_data
  *    as the indirect buffer (128 bytes)
@@ -906,7 +918,8 @@ struct i40e_aqc_add_veb {
 					I40E_AQC_ADD_VEB_PORT_TYPE_SHIFT)
 #define I40E_AQC_ADD_VEB_PORT_TYPE_DEFAULT	0x2
 #define I40E_AQC_ADD_VEB_PORT_TYPE_DATA		0x4
-#define I40E_AQC_ADD_VEB_ENABLE_L2_FILTER	0x8
+#define I40E_AQC_ADD_VEB_ENABLE_L2_FILTER	0x8     /* deprecated */
+#define I40E_AQC_ADD_VEB_ENABLE_DISABLE_STATS	0x10
 	u8	enable_tcs;
 	u8	reserved[9];
 };
-- 
cgit v0.10.2


From 445204644ba6e5565d7aba85a4323b52960179db Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Tue, 16 Feb 2016 18:07:57 +0200
Subject: bnx2x: Remove unneccessary EXPORT_SYMBOL

bnx2x_schedule_sp_rtnl is exported by bnx2x, although no other module
uses it.

Reported-by: Benjamin Poirier <bpoirier@suse.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 9695a4c..9e42bca 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -5086,4 +5086,3 @@ void bnx2x_schedule_sp_rtnl(struct bnx2x *bp, enum sp_rtnl_flag flag,
 	   flag);
 	schedule_delayed_work(&bp->sp_rtnl_task, 0);
 }
-EXPORT_SYMBOL(bnx2x_schedule_sp_rtnl);
-- 
cgit v0.10.2


From 883ce97d25b019ce8437ba6f49e38302ca5ec23f Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Tue, 16 Feb 2016 18:07:58 +0200
Subject: bnx2x: Add Geneve inner-RSS support

This adds the ability to perform RSS hashing based on encapsulated
headers for a geneve-encapsulated packet.

This also changes the Vxlan implementation in bnx2x to be uniform
for both vxlan and geneve [from configuration perspective].

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index cae0956..7dd7490 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -1277,8 +1277,7 @@ enum sp_rtnl_flag {
 	BNX2X_SP_RTNL_HYPERVISOR_VLAN,
 	BNX2X_SP_RTNL_TX_STOP,
 	BNX2X_SP_RTNL_GET_DRV_VERSION,
-	BNX2X_SP_RTNL_ADD_VXLAN_PORT,
-	BNX2X_SP_RTNL_DEL_VXLAN_PORT,
+	BNX2X_SP_RTNL_CHANGE_UDP_PORT,
 };
 
 enum bnx2x_iov_flag {
@@ -1327,6 +1326,17 @@ struct bnx2x_vlan_entry {
 	bool hw;
 };
 
+enum bnx2x_udp_port_type {
+	BNX2X_UDP_PORT_VXLAN,
+	BNX2X_UDP_PORT_GENEVE,
+	BNX2X_UDP_PORT_MAX,
+};
+
+struct bnx2x_udp_tunnel {
+	u16 dst_port;
+	u8 count;
+};
+
 struct bnx2x {
 	/* Fields used in the tx and intr/napi performance paths
 	 * are grouped together in the beginning of the structure
@@ -1830,9 +1840,10 @@ struct bnx2x {
 	struct list_head vlan_reg;
 	u16 vlan_cnt;
 	u16 vlan_credit;
-	u16 vxlan_dst_port;
-	u8 vxlan_dst_port_count;
 	bool accept_any_vlan;
+
+	/* Vxlan/Geneve related information */
+	struct bnx2x_udp_tunnel udp_tunnel_ports[BNX2X_UDP_PORT_MAX];
 };
 
 /* Tx queues may be less or equal to Rx queues */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 4cbb03f8..3736986 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -923,6 +923,7 @@ static inline int bnx2x_func_start(struct bnx2x *bp)
 	struct bnx2x_func_state_params func_params = {NULL};
 	struct bnx2x_func_start_params *start_params =
 		&func_params.params.start;
+	u16 port;
 
 	/* Prepare parameters for function state transitions */
 	__set_bit(RAMROD_COMP_WAIT, &func_params.ramrod_flags);
@@ -959,8 +960,14 @@ static inline int bnx2x_func_start(struct bnx2x *bp)
 		start_params->network_cos_mode = STATIC_COS;
 	else /* CHIP_IS_E1X */
 		start_params->network_cos_mode = FW_WRR;
-
-	start_params->vxlan_dst_port = bp->vxlan_dst_port;
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count) {
+		port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].dst_port;
+		start_params->vxlan_dst_port = port;
+	}
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) {
+		port = bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].dst_port;
+		start_params->geneve_dst_port = port;
+	}
 
 	start_params->inner_rss = 1;
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 6c4e3a6..5d6b2d9 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -59,7 +59,9 @@
 #include <linux/semaphore.h>
 #include <linux/stringify.h>
 #include <linux/vmalloc.h>
-
+#if IS_ENABLED(CONFIG_GENEVE)
+#include <net/geneve.h>
+#endif
 #include "bnx2x.h"
 #include "bnx2x_init.h"
 #include "bnx2x_init_ops.h"
@@ -10076,11 +10078,13 @@ static void bnx2x_parity_recover(struct bnx2x *bp)
 	}
 }
 
-#ifdef CONFIG_BNX2X_VXLAN
-static int bnx2x_vxlan_port_update(struct bnx2x *bp, u16 port)
+#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_GENEVE)
+static int bnx2x_udp_port_update(struct bnx2x *bp)
 {
 	struct bnx2x_func_switch_update_params *switch_update_params;
 	struct bnx2x_func_state_params func_params = {NULL};
+	struct bnx2x_udp_tunnel *udp_tunnel;
+	u16 vxlan_port = 0, geneve_port = 0;
 	int rc;
 
 	switch_update_params = &func_params.params.switch_update;
@@ -10095,69 +10099,125 @@ static int bnx2x_vxlan_port_update(struct bnx2x *bp, u16 port)
 	/* Function parameters */
 	__set_bit(BNX2X_F_UPDATE_TUNNEL_CFG_CHNG,
 		  &switch_update_params->changes);
-	switch_update_params->vxlan_dst_port = port;
+
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) {
+		udp_tunnel = &bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE];
+		geneve_port = udp_tunnel->dst_port;
+		switch_update_params->geneve_dst_port = geneve_port;
+	}
+
+	if (bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count) {
+		udp_tunnel = &bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN];
+		vxlan_port = udp_tunnel->dst_port;
+		switch_update_params->vxlan_dst_port = vxlan_port;
+	}
+
+	/* Re-enable inner-rss for the offloaded UDP tunnels */
+	__set_bit(BNX2X_F_UPDATE_TUNNEL_INNER_RSS,
+		  &switch_update_params->changes);
+
 	rc = bnx2x_func_state_change(bp, &func_params);
 	if (rc)
-		BNX2X_ERR("failed to change vxlan dst port to %d (rc = 0x%x)\n",
-			  port, rc);
+		BNX2X_ERR("failed to set UDP dst port to %04x %04x (rc = 0x%x)\n",
+			  vxlan_port, geneve_port, rc);
+	else
+		DP(BNX2X_MSG_SP,
+		   "Configured UDP ports: Vxlan [%04x] Geneve [%04x]\n",
+		   vxlan_port, geneve_port);
+
 	return rc;
 }
 
-static void __bnx2x_add_vxlan_port(struct bnx2x *bp, u16 port)
+static void __bnx2x_add_udp_port(struct bnx2x *bp, u16 port,
+				 enum bnx2x_udp_port_type type)
 {
-	if (!netif_running(bp->dev))
+	struct bnx2x_udp_tunnel *udp_port = &bp->udp_tunnel_ports[type];
+
+	if (!netif_running(bp->dev) || !IS_PF(bp))
+		return;
+
+	if (udp_port->count && udp_port->dst_port == port) {
+		udp_port->count++;
 		return;
+	}
 
-	if (bp->vxlan_dst_port_count && bp->vxlan_dst_port == port) {
-		bp->vxlan_dst_port_count++;
+	if (udp_port->count) {
+		DP(BNX2X_MSG_SP,
+		   "UDP tunnel [%d] -  destination port limit reached\n",
+		   type);
 		return;
 	}
 
-	if (bp->vxlan_dst_port_count || !IS_PF(bp)) {
-		DP(BNX2X_MSG_SP, "Vxlan destination port limit reached\n");
+	udp_port->dst_port = port;
+	udp_port->count = 1;
+	bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_CHANGE_UDP_PORT, 0);
+}
+
+static void __bnx2x_del_udp_port(struct bnx2x *bp, u16 port,
+				 enum bnx2x_udp_port_type type)
+{
+	struct bnx2x_udp_tunnel *udp_port = &bp->udp_tunnel_ports[type];
+
+	if (!IS_PF(bp))
+		return;
+
+	if (!udp_port->count || udp_port->dst_port != port) {
+		DP(BNX2X_MSG_SP, "Invalid UDP tunnel [%d] port\n",
+		   type);
 		return;
 	}
 
-	bp->vxlan_dst_port = port;
-	bp->vxlan_dst_port_count = 1;
-	bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_ADD_VXLAN_PORT, 0);
+	/* Remove reference, and make certain it's no longer in use */
+	udp_port->count--;
+	if (udp_port->count)
+		return;
+	udp_port->dst_port = 0;
+
+	if (netif_running(bp->dev))
+		bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_CHANGE_UDP_PORT, 0);
+	else
+		DP(BNX2X_MSG_SP, "Deleted UDP tunnel [%d] port %d\n",
+		   type, port);
 }
+#endif
 
+#ifdef CONFIG_BNX2X_VXLAN
 static void bnx2x_add_vxlan_port(struct net_device *netdev,
 				 sa_family_t sa_family, __be16 port)
 {
 	struct bnx2x *bp = netdev_priv(netdev);
 	u16 t_port = ntohs(port);
 
-	__bnx2x_add_vxlan_port(bp, t_port);
+	__bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN);
 }
 
-static void __bnx2x_del_vxlan_port(struct bnx2x *bp, u16 port)
+static void bnx2x_del_vxlan_port(struct net_device *netdev,
+				 sa_family_t sa_family, __be16 port)
 {
-	if (!bp->vxlan_dst_port_count || bp->vxlan_dst_port != port ||
-	    !IS_PF(bp)) {
-		DP(BNX2X_MSG_SP, "Invalid vxlan port\n");
-		return;
-	}
-	bp->vxlan_dst_port_count--;
-	if (bp->vxlan_dst_port_count)
-		return;
+	struct bnx2x *bp = netdev_priv(netdev);
+	u16 t_port = ntohs(port);
 
-	if (netif_running(bp->dev)) {
-		bnx2x_schedule_sp_rtnl(bp, BNX2X_SP_RTNL_DEL_VXLAN_PORT, 0);
-	} else {
-		bp->vxlan_dst_port = 0;
-		netdev_info(bp->dev, "Deleted vxlan dest port %d", port);
-	}
+	__bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN);
+}
+#endif
+
+#if IS_ENABLED(CONFIG_GENEVE)
+static void bnx2x_add_geneve_port(struct net_device *netdev,
+				  sa_family_t sa_family, __be16 port)
+{
+	struct bnx2x *bp = netdev_priv(netdev);
+	u16 t_port = ntohs(port);
+
+	__bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE);
 }
 
-static void bnx2x_del_vxlan_port(struct net_device *netdev,
-				 sa_family_t sa_family, __be16 port)
+static void bnx2x_del_geneve_port(struct net_device *netdev,
+				  sa_family_t sa_family, __be16 port)
 {
 	struct bnx2x *bp = netdev_priv(netdev);
 	u16 t_port = ntohs(port);
 
-	__bnx2x_del_vxlan_port(bp, t_port);
+	__bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE);
 }
 #endif
 
@@ -10169,9 +10229,6 @@ static int bnx2x_close(struct net_device *dev);
 static void bnx2x_sp_rtnl_task(struct work_struct *work)
 {
 	struct bnx2x *bp = container_of(work, struct bnx2x, sp_rtnl_task.work);
-#ifdef CONFIG_BNX2X_VXLAN
-	u16 port;
-#endif
 
 	rtnl_lock();
 
@@ -10270,23 +10327,27 @@ sp_rtnl_not_reset:
 			       &bp->sp_rtnl_state))
 		bnx2x_update_mng_version(bp);
 
-#ifdef CONFIG_BNX2X_VXLAN
-	port = bp->vxlan_dst_port;
-	if (test_and_clear_bit(BNX2X_SP_RTNL_ADD_VXLAN_PORT,
-			       &bp->sp_rtnl_state)) {
-		if (!bnx2x_vxlan_port_update(bp, port))
-			netdev_info(bp->dev, "Added vxlan dest port %d", port);
-		else
-			bp->vxlan_dst_port = 0;
-	}
-
-	if (test_and_clear_bit(BNX2X_SP_RTNL_DEL_VXLAN_PORT,
+#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_GENEVE)
+	if (test_and_clear_bit(BNX2X_SP_RTNL_CHANGE_UDP_PORT,
 			       &bp->sp_rtnl_state)) {
-		if (!bnx2x_vxlan_port_update(bp, 0)) {
-			netdev_info(bp->dev,
-				    "Deleted vxlan dest port %d", port);
-			bp->vxlan_dst_port = 0;
-			vxlan_get_rx_port(bp->dev);
+		if (bnx2x_udp_port_update(bp)) {
+			/* On error, forget configuration */
+			memset(bp->udp_tunnel_ports, 0,
+			       sizeof(struct bnx2x_udp_tunnel) *
+			       BNX2X_UDP_PORT_MAX);
+		} else {
+			/* Since we don't store additional port information,
+			 * if no port is configured for any feature ask for
+			 * information about currently configured ports.
+			 */
+#ifdef CONFIG_BNX2X_VXLAN
+			if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count)
+				vxlan_get_rx_port(bp->dev);
+#endif
+#if IS_ENABLED(CONFIG_GENEVE)
+			if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count)
+				geneve_get_rx_port(bp->dev);
+#endif
 		}
 	}
 #endif
@@ -12494,6 +12555,10 @@ static int bnx2x_open(struct net_device *dev)
 	if (IS_PF(bp))
 		vxlan_get_rx_port(dev);
 #endif
+#if IS_ENABLED(CONFIG_GENEVE)
+	if (IS_PF(bp))
+		geneve_get_rx_port(dev);
+#endif
 
 	return 0;
 }
@@ -13011,6 +13076,10 @@ static const struct net_device_ops bnx2x_netdev_ops = {
 	.ndo_add_vxlan_port	= bnx2x_add_vxlan_port,
 	.ndo_del_vxlan_port	= bnx2x_del_vxlan_port,
 #endif
+#if IS_ENABLED(CONFIG_GENEVE)
+	.ndo_add_geneve_port	= bnx2x_add_geneve_port,
+	.ndo_del_geneve_port	= bnx2x_del_geneve_port,
+#endif
 };
 
 static int bnx2x_set_coherency_mask(struct bnx2x *bp)
-- 
cgit v0.10.2


From 9c73267d2ea26b0148932f335a7deb5480a80425 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Tue, 16 Feb 2016 18:07:59 +0200
Subject: bnx2x: Add support for single-port DCBx

Driver is currently looking at shared information for determining whether
DCBx can be supported for a given port.
On 4-port devices, up-to-date management firmware can support DCBx on
each port of a given engine independently - but that would cause bnx2x to
misinterpert the support and assume DCBx is supported on both.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 5d6b2d9..c584525 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12429,8 +12429,10 @@ static int bnx2x_init_bp(struct bnx2x *bp)
 
 	if (SHMEM2_HAS(bp, dcbx_lldp_params_offset) &&
 	    SHMEM2_HAS(bp, dcbx_lldp_dcbx_stat_offset) &&
+	    SHMEM2_HAS(bp, dcbx_en) &&
 	    SHMEM2_RD(bp, dcbx_lldp_params_offset) &&
-	    SHMEM2_RD(bp, dcbx_lldp_dcbx_stat_offset)) {
+	    SHMEM2_RD(bp, dcbx_lldp_dcbx_stat_offset) &&
+	    SHMEM2_RD(bp, dcbx_en[BP_PORT(bp)])) {
 		bnx2x_dcbx_set_state(bp, true, BNX2X_DCBX_ENABLED_ON_NEG_ON);
 		bnx2x_dcbx_init_params(bp);
 	} else {
-- 
cgit v0.10.2


From e5d3a51cefbb08c8b8c521f627aef1f998834530 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Tue, 16 Feb 2016 18:08:00 +0200
Subject: bnx2x: extend DCBx support

This adds support for default application priority.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
index 7ccf668..2c6ba04 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c
@@ -195,6 +195,7 @@ static void bnx2x_dcbx_get_ap_feature(struct bnx2x *bp,
 				   u32 error) {
 	u8 index;
 	u32 *ttp = bp->dcbx_port_params.app.traffic_type_priority;
+	u8 iscsi_pri_found = 0, fcoe_pri_found = 0;
 
 	if (GET_FLAGS(error, DCBX_LOCAL_APP_ERROR))
 		DP(BNX2X_MSG_DCB, "DCBX_LOCAL_APP_ERROR\n");
@@ -210,29 +211,57 @@ static void bnx2x_dcbx_get_ap_feature(struct bnx2x *bp,
 
 		bp->dcbx_port_params.app.enabled = true;
 
+		/* Use 0 as the default application priority for all. */
 		for (index = 0 ; index < LLFC_DRIVER_TRAFFIC_TYPE_MAX; index++)
 			ttp[index] = 0;
 
-		if (app->default_pri < MAX_PFC_PRIORITIES)
-			ttp[LLFC_TRAFFIC_TYPE_NW] = app->default_pri;
-
 		for (index = 0 ; index < DCBX_MAX_APP_PROTOCOL; index++) {
 			struct dcbx_app_priority_entry *entry =
 							app->app_pri_tbl;
+			enum traffic_type type = MAX_TRAFFIC_TYPE;
 
 			if (GET_FLAGS(entry[index].appBitfield,
-				     DCBX_APP_SF_ETH_TYPE) &&
-			   ETH_TYPE_FCOE == entry[index].app_id)
-				bnx2x_dcbx_get_ap_priority(bp,
-						entry[index].pri_bitmap,
-						LLFC_TRAFFIC_TYPE_FCOE);
+				      DCBX_APP_SF_DEFAULT) &&
+			    GET_FLAGS(entry[index].appBitfield,
+				      DCBX_APP_SF_ETH_TYPE)) {
+				type = LLFC_TRAFFIC_TYPE_NW;
+			} else if (GET_FLAGS(entry[index].appBitfield,
+					     DCBX_APP_SF_PORT) &&
+				   TCP_PORT_ISCSI == entry[index].app_id) {
+				type = LLFC_TRAFFIC_TYPE_ISCSI;
+				iscsi_pri_found = 1;
+			} else if (GET_FLAGS(entry[index].appBitfield,
+					     DCBX_APP_SF_ETH_TYPE) &&
+				   ETH_TYPE_FCOE == entry[index].app_id) {
+				type = LLFC_TRAFFIC_TYPE_FCOE;
+				fcoe_pri_found = 1;
+			}
 
-			if (GET_FLAGS(entry[index].appBitfield,
-				     DCBX_APP_SF_PORT) &&
-			   TCP_PORT_ISCSI == entry[index].app_id)
-				bnx2x_dcbx_get_ap_priority(bp,
-						entry[index].pri_bitmap,
-						LLFC_TRAFFIC_TYPE_ISCSI);
+			if (type == MAX_TRAFFIC_TYPE)
+				continue;
+
+			bnx2x_dcbx_get_ap_priority(bp,
+						   entry[index].pri_bitmap,
+						   type);
+		}
+
+		/* If we have received a non-zero default application
+		 * priority, then use that for applications which are
+		 * not configured with any priority.
+		 */
+		if (ttp[LLFC_TRAFFIC_TYPE_NW] != 0) {
+			if (!iscsi_pri_found) {
+				ttp[LLFC_TRAFFIC_TYPE_ISCSI] =
+					ttp[LLFC_TRAFFIC_TYPE_NW];
+				DP(BNX2X_MSG_DCB,
+				   "ISCSI is using default priority.\n");
+			}
+			if (!fcoe_pri_found) {
+				ttp[LLFC_TRAFFIC_TYPE_FCOE] =
+					ttp[LLFC_TRAFFIC_TYPE_NW];
+				DP(BNX2X_MSG_DCB,
+				   "FCoE is using default priority.\n");
+			}
 		}
 	} else {
 		DP(BNX2X_MSG_DCB, "DCBX_LOCAL_APP_DISABLED\n");
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
index 27aa080..dd9d6e6 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
@@ -1831,10 +1831,13 @@ struct dcbx_app_priority_entry {
 #elif defined(__LITTLE_ENDIAN)
 	u8 appBitfield;
 	#define DCBX_APP_ENTRY_VALID         0x01
-	#define DCBX_APP_ENTRY_SF_MASK       0x30
+	#define DCBX_APP_ENTRY_SF_MASK       0xF0
 	#define DCBX_APP_ENTRY_SF_SHIFT      4
+	#define DCBX_APP_ENTRY_VALID         0x01
 	#define DCBX_APP_SF_ETH_TYPE         0x10
 	#define DCBX_APP_SF_PORT             0x20
+	#define DCBX_APP_SF_UDP              0x40
+	#define DCBX_APP_SF_DEFAULT          0x80
 	u8  pri_bitmap;
 	u16  app_id;
 #endif
-- 
cgit v0.10.2


From e56270f63529718bed5b9017c98fb355a25e9954 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Tue, 16 Feb 2016 18:08:01 +0200
Subject: bnx2x: Warn about grc timeouts in register dump

There are several scenarios where taking a register dump from a device
might log benign GRC timeout attentions to system logs.
Most common of those is when taking the dump from a 2-port device.

Sadly, there's no easy way to mask the problematic attentions during the
flow - Changing this behvaior would require a firmware update.
For now, simply warn users to ignore the warnings.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index 820b7e0..85a7800 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -981,6 +981,11 @@ static void bnx2x_get_regs(struct net_device *dev,
 	memcpy(p, &dump_hdr, sizeof(struct dump_header));
 	p += dump_hdr.header_size + 1;
 
+	/* This isn't really an error, but since attention handling is going
+	 * to print the GRC timeouts using this macro, we use the same.
+	 */
+	BNX2X_ERR("Generating register dump. Might trigger harmless GRC timeouts\n");
+
 	/* Actually read the registers */
 	__bnx2x_get_regs(bp, p);
 
-- 
cgit v0.10.2


From 911362c70df5b766c243dc297fadeaced786ffd8 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:53 +0100
Subject: net: add dst_cache support

This patch add a generic, lockless dst cache implementation.
The need for lock is avoided updating the dst cache fields
only in per cpu scope, and requiring that the cache manipulation
functions are invoked with the local bh disabled.

The refresh_ts and reset_ts fields are used to ensure the cache
consistency in case of cuncurrent cache update (dst_cache_set*) and
reset operation (dst_cache_reset).

Consider the following scenario:

CPU1:                                   	CPU2:
  <cache lookup with emtpy cache: it fails>
  <get dst via uncached route lookup>
						<related configuration changes>
                                        	dst_cache_reset()
  dst_cache_set()

The dst entry set passed to dst_cache_set() should not be used
for later dst cache lookup, because it's obtained using old
configuration values.

Since the refresh_ts is updated only on dst_cache lookup, the
cached value in the above scenario will be discarded on the next
lookup.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/dst_cache.h b/include/net/dst_cache.h
new file mode 100644
index 0000000..151accae
--- /dev/null
+++ b/include/net/dst_cache.h
@@ -0,0 +1,97 @@
+#ifndef _NET_DST_CACHE_H
+#define _NET_DST_CACHE_H
+
+#include <linux/jiffies.h>
+#include <net/dst.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ip6_fib.h>
+#endif
+
+struct dst_cache {
+	struct dst_cache_pcpu __percpu *cache;
+	unsigned long reset_ts;
+};
+
+/**
+ *	dst_cache_get - perform cache lookup
+ *	@dst_cache: the cache
+ *
+ *	The caller should use dst_cache_get_ip4() if it need to retrieve the
+ *	source address to be used when xmitting to the cached dst.
+ *	local BH must be disabled.
+ */
+struct dst_entry *dst_cache_get(struct dst_cache *dst_cache);
+
+/**
+ *	dst_cache_get_ip4 - perform cache lookup and fetch ipv4 source address
+ *	@dst_cache: the cache
+ *	@saddr: return value for the retrieved source address
+ *
+ *	local BH must be disabled.
+ */
+struct rtable *dst_cache_get_ip4(struct dst_cache *dst_cache, __be32 *saddr);
+
+/**
+ *	dst_cache_set_ip4 - store the ipv4 dst into the cache
+ *	@dst_cache: the cache
+ *	@dst: the entry to be cached
+ *	@saddr: the source address to be stored inside the cache
+ *
+ *	local BH must be disabled.
+ */
+void dst_cache_set_ip4(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       __be32 saddr);
+
+#if IS_ENABLED(CONFIG_IPV6)
+
+/**
+ *	dst_cache_set_ip6 - store the ipv6 dst into the cache
+ *	@dst_cache: the cache
+ *	@dst: the entry to be cached
+ *	@saddr: the source address to be stored inside the cache
+ *
+ *	local BH must be disabled.
+ */
+void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       const struct in6_addr *addr);
+
+/**
+ *	dst_cache_get_ip6 - perform cache lookup and fetch ipv6 source address
+ *	@dst_cache: the cache
+ *	@saddr: return value for the retrieved source address
+ *
+ *	local BH must be disabled.
+ */
+struct dst_entry *dst_cache_get_ip6(struct dst_cache *dst_cache,
+				    struct in6_addr *saddr);
+#endif
+
+/**
+ *	dst_cache_reset - invalidate the cache contents
+ *	@dst_cache: the cache
+ *
+ *	This do not free the cached dst to avoid races and contentions.
+ *	the dst will be freed on later cache lookup.
+ */
+static inline void dst_cache_reset(struct dst_cache *dst_cache)
+{
+	dst_cache->reset_ts = jiffies;
+}
+
+/**
+ *	dst_cache_init - initialize the cache, allocating the required storage
+ *	@dst_cache: the cache
+ *	@gfp: allocation flags
+ */
+int dst_cache_init(struct dst_cache *dst_cache, gfp_t gfp);
+
+/**
+ *	dst_cache_destroy - empty the cache and free the allocated storage
+ *	@dst_cache: the cache
+ *
+ *	No synchronization is enforced: it must be called only when the cache
+ *	is unsed.
+ */
+void dst_cache_destroy(struct dst_cache *dst_cache);
+
+#endif
diff --git a/net/Kconfig b/net/Kconfig
index 1743546..b80efec 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -392,6 +392,10 @@ config LWTUNNEL
 	  weight tunnel endpoint. Tunnel encapsulation parameters are stored
 	  with light weight tunnel state associated with fib routes.
 
+config DST_CACHE
+	bool "dst cache"
+	default n
+
 endif   # if NET
 
 # Used by archs to tell that they support BPF_JIT
diff --git a/net/core/Makefile b/net/core/Makefile
index 0b835de..7a8fb8a 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -24,3 +24,4 @@ obj-$(CONFIG_NET_PTP_CLASSIFY) += ptp_classifier.o
 obj-$(CONFIG_CGROUP_NET_PRIO) += netprio_cgroup.o
 obj-$(CONFIG_CGROUP_NET_CLASSID) += netclassid_cgroup.o
 obj-$(CONFIG_LWTUNNEL) += lwtunnel.o
+obj-$(CONFIG_DST_CACHE) += dst_cache.o
diff --git a/net/core/dst_cache.c b/net/core/dst_cache.c
new file mode 100644
index 0000000..3938f3f
--- /dev/null
+++ b/net/core/dst_cache.c
@@ -0,0 +1,168 @@
+/*
+ * net/core/dst_cache.c - dst entry cache
+ *
+ * Copyright (c) 2016 Paolo Abeni <pabeni@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <net/dst_cache.h>
+#include <net/route.h>
+#if IS_ENABLED(CONFIG_IPV6)
+#include <net/ip6_fib.h>
+#endif
+#include <uapi/linux/in.h>
+
+struct dst_cache_pcpu {
+	unsigned long refresh_ts;
+	struct dst_entry *dst;
+	u32 cookie;
+	union {
+		struct in_addr in_saddr;
+		struct in6_addr in6_saddr;
+	};
+};
+
+void dst_cache_per_cpu_dst_set(struct dst_cache_pcpu *dst_cache,
+			       struct dst_entry *dst, u32 cookie)
+{
+	dst_release(dst_cache->dst);
+	if (dst)
+		dst_hold(dst);
+
+	dst_cache->cookie = cookie;
+	dst_cache->dst = dst;
+}
+
+struct dst_entry *dst_cache_per_cpu_get(struct dst_cache *dst_cache,
+					struct dst_cache_pcpu *idst)
+{
+	struct dst_entry *dst;
+
+	dst = idst->dst;
+	if (!dst)
+		goto fail;
+
+	/* the cache already hold a dst reference; it can't go away */
+	dst_hold(dst);
+
+	if (unlikely(!time_after(idst->refresh_ts, dst_cache->reset_ts) ||
+		     (dst->obsolete && !dst->ops->check(dst, idst->cookie)))) {
+		dst_cache_per_cpu_dst_set(idst, NULL, 0);
+		dst_release(dst);
+		goto fail;
+	}
+	return dst;
+
+fail:
+	idst->refresh_ts = jiffies;
+	return NULL;
+}
+
+struct dst_entry *dst_cache_get(struct dst_cache *dst_cache)
+{
+	if (!dst_cache->cache)
+		return NULL;
+
+	return dst_cache_per_cpu_get(dst_cache, this_cpu_ptr(dst_cache->cache));
+}
+EXPORT_SYMBOL_GPL(dst_cache_get);
+
+struct rtable *dst_cache_get_ip4(struct dst_cache *dst_cache, __be32 *saddr)
+{
+	struct dst_cache_pcpu *idst;
+	struct dst_entry *dst;
+
+	if (!dst_cache->cache)
+		return NULL;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst = dst_cache_per_cpu_get(dst_cache, idst);
+	if (!dst)
+		return NULL;
+
+	*saddr = idst->in_saddr.s_addr;
+	return container_of(dst, struct rtable, dst);
+}
+EXPORT_SYMBOL_GPL(dst_cache_get_ip4);
+
+void dst_cache_set_ip4(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       __be32 saddr)
+{
+	struct dst_cache_pcpu *idst;
+
+	if (!dst_cache->cache)
+		return;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst_cache_per_cpu_dst_set(idst, dst, 0);
+	idst->in_saddr.s_addr = saddr;
+}
+EXPORT_SYMBOL_GPL(dst_cache_set_ip4);
+
+#if IS_ENABLED(CONFIG_IPV6)
+void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst,
+		       const struct in6_addr *addr)
+{
+	struct dst_cache_pcpu *idst;
+
+	if (!dst_cache->cache)
+		return;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst_cache_per_cpu_dst_set(this_cpu_ptr(dst_cache->cache), dst,
+				  rt6_get_cookie((struct rt6_info *)dst));
+	idst->in6_saddr = *addr;
+}
+EXPORT_SYMBOL_GPL(dst_cache_set_ip6);
+
+struct dst_entry *dst_cache_get_ip6(struct dst_cache *dst_cache,
+				    struct in6_addr *saddr)
+{
+	struct dst_cache_pcpu *idst;
+	struct dst_entry *dst;
+
+	if (!dst_cache->cache)
+		return NULL;
+
+	idst = this_cpu_ptr(dst_cache->cache);
+	dst = dst_cache_per_cpu_get(dst_cache, idst);
+	if (!dst)
+		return NULL;
+
+	*saddr = idst->in6_saddr;
+	return dst;
+}
+EXPORT_SYMBOL_GPL(dst_cache_get_ip6);
+#endif
+
+int dst_cache_init(struct dst_cache *dst_cache, gfp_t gfp)
+{
+	dst_cache->cache = alloc_percpu_gfp(struct dst_cache_pcpu,
+					    gfp | __GFP_ZERO);
+	if (!dst_cache->cache)
+		return -ENOMEM;
+
+	dst_cache_reset(dst_cache);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(dst_cache_init);
+
+void dst_cache_destroy(struct dst_cache *dst_cache)
+{
+	int i;
+
+	if (!dst_cache->cache)
+		return;
+
+	for_each_possible_cpu(i)
+		dst_release(per_cpu_ptr(dst_cache->cache, i)->dst);
+
+	free_percpu(dst_cache->cache);
+}
+EXPORT_SYMBOL_GPL(dst_cache_destroy);
-- 
cgit v0.10.2


From 607f725f6f7d5ec3759fbc16224afb60e2152a5b Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:54 +0100
Subject: net: replace dst_cache ip6_tunnel implementation with the generic one

This also fix a potential race into the existing tunnel code, which
could lead to the wrong dst to be permanenty cached:

CPU1:					CPU2:
  <xmit on ip6_tunnel>
  <cache lookup fails>
  dst = ip6_route_output(...)
					<tunnel params are changed via nl>
					dst_cache_reset() // no effect,
							// the cache is empty
  dst_cache_set() // the wrong dst
	// is permanenty stored
	// into the cache

With the new dst implementation the above race is not possible
since the first cache lookup after dst_cache_reset will fail due
to the timestamp check

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h
index 0d0ce0b..499a707 100644
--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -6,6 +6,7 @@
 #include <linux/if_tunnel.h>
 #include <linux/ip6_tunnel.h>
 #include <net/ip_tunnels.h>
+#include <net/dst_cache.h>
 
 #define IP6TUNNEL_ERR_TIMEO (30*HZ)
 
@@ -33,12 +34,6 @@ struct __ip6_tnl_parm {
 	__be32			o_key;
 };
 
-struct ip6_tnl_dst {
-	seqlock_t lock;
-	struct dst_entry __rcu *dst;
-	u32 cookie;
-};
-
 /* IPv6 tunnel */
 struct ip6_tnl {
 	struct ip6_tnl __rcu *next;	/* next tunnel in list */
@@ -46,7 +41,7 @@ struct ip6_tnl {
 	struct net *net;	/* netns for packet i/o */
 	struct __ip6_tnl_parm parms;	/* tunnel configuration parameters */
 	struct flowi fl;	/* flowi template for xmit */
-	struct ip6_tnl_dst __percpu *dst_cache;	/* cached dst */
+	struct dst_cache dst_cache;	/* cached dst */
 
 	int err_count;
 	unsigned long err_time;
@@ -66,11 +61,6 @@ struct ipv6_tlv_tnl_enc_lim {
 	__u8 encap_limit;	/* tunnel encapsulation limit   */
 } __packed;
 
-struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t);
-int ip6_tnl_dst_init(struct ip6_tnl *t);
-void ip6_tnl_dst_destroy(struct ip6_tnl *t);
-void ip6_tnl_dst_reset(struct ip6_tnl *t);
-void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst);
 int ip6_tnl_rcv_ctl(struct ip6_tnl *t, const struct in6_addr *laddr,
 		const struct in6_addr *raddr);
 int ip6_tnl_xmit_ctl(struct ip6_tnl *t, const struct in6_addr *laddr,
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 40c8975..11e875f 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -207,6 +207,7 @@ config IPV6_NDISC_NODETYPE
 config IPV6_TUNNEL
 	tristate "IPv6: IP-in-IPv6 tunnel (RFC2473)"
 	select INET6_TUNNEL
+	select DST_CACHE
 	---help---
 	  Support for IPv6-in-IPv6 and IPv4-in-IPv6 tunnels described in
 	  RFC 2473.
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index f37f18b..a94e506 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -360,7 +360,7 @@ static void ip6gre_tunnel_uninit(struct net_device *dev)
 	struct ip6gre_net *ign = net_generic(t->net, ip6gre_net_id);
 
 	ip6gre_tunnel_unlink(ign, t);
-	ip6_tnl_dst_reset(t);
+	dst_cache_reset(&t->dst_cache);
 	dev_put(dev);
 }
 
@@ -633,7 +633,7 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb,
 	}
 
 	if (!fl6->flowi6_mark)
-		dst = ip6_tnl_dst_get(tunnel);
+		dst = dst_cache_get(&tunnel->dst_cache);
 
 	if (!dst) {
 		dst = ip6_route_output(net, NULL, fl6);
@@ -702,7 +702,7 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb,
 	}
 
 	if (!fl6->flowi6_mark && ndst)
-		ip6_tnl_dst_set(tunnel, ndst);
+		dst_cache_set_ip6(&tunnel->dst_cache, ndst, &fl6->saddr);
 	skb_dst_set(skb, dst);
 
 	proto = NEXTHDR_GRE;
@@ -1009,7 +1009,7 @@ static int ip6gre_tnl_change(struct ip6_tnl *t,
 	t->parms.o_key = p->o_key;
 	t->parms.i_flags = p->i_flags;
 	t->parms.o_flags = p->o_flags;
-	ip6_tnl_dst_reset(t);
+	dst_cache_reset(&t->dst_cache);
 	ip6gre_tnl_link_config(t, set_mtu);
 	return 0;
 }
@@ -1219,7 +1219,7 @@ static void ip6gre_dev_free(struct net_device *dev)
 {
 	struct ip6_tnl *t = netdev_priv(dev);
 
-	ip6_tnl_dst_destroy(t);
+	dst_cache_destroy(&t->dst_cache);
 	free_percpu(dev->tstats);
 	free_netdev(dev);
 }
@@ -1257,7 +1257,7 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
-	ret = ip6_tnl_dst_init(tunnel);
+	ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
 	if (ret) {
 		free_percpu(dev->tstats);
 		dev->tstats = NULL;
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 137fca4..3f3aabd 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -122,97 +122,6 @@ static struct net_device_stats *ip6_get_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
-/*
- * Locking : hash tables are protected by RCU and RTNL
- */
-
-static void ip6_tnl_per_cpu_dst_set(struct ip6_tnl_dst *idst,
-				    struct dst_entry *dst)
-{
-	write_seqlock_bh(&idst->lock);
-	dst_release(rcu_dereference_protected(
-			    idst->dst,
-			    lockdep_is_held(&idst->lock.lock)));
-	if (dst) {
-		dst_hold(dst);
-		idst->cookie = rt6_get_cookie((struct rt6_info *)dst);
-	} else {
-		idst->cookie = 0;
-	}
-	rcu_assign_pointer(idst->dst, dst);
-	write_sequnlock_bh(&idst->lock);
-}
-
-struct dst_entry *ip6_tnl_dst_get(struct ip6_tnl *t)
-{
-	struct ip6_tnl_dst *idst;
-	struct dst_entry *dst;
-	unsigned int seq;
-	u32 cookie;
-
-	idst = raw_cpu_ptr(t->dst_cache);
-
-	rcu_read_lock();
-	do {
-		seq = read_seqbegin(&idst->lock);
-		dst = rcu_dereference(idst->dst);
-		cookie = idst->cookie;
-	} while (read_seqretry(&idst->lock, seq));
-
-	if (dst && !atomic_inc_not_zero(&dst->__refcnt))
-		dst = NULL;
-	rcu_read_unlock();
-
-	if (dst && dst->obsolete && !dst->ops->check(dst, cookie)) {
-		ip6_tnl_per_cpu_dst_set(idst, NULL);
-		dst_release(dst);
-		dst = NULL;
-	}
-	return dst;
-}
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_get);
-
-void ip6_tnl_dst_reset(struct ip6_tnl *t)
-{
-	int i;
-
-	for_each_possible_cpu(i)
-		ip6_tnl_per_cpu_dst_set(per_cpu_ptr(t->dst_cache, i), NULL);
-}
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_reset);
-
-void ip6_tnl_dst_set(struct ip6_tnl *t, struct dst_entry *dst)
-{
-	ip6_tnl_per_cpu_dst_set(raw_cpu_ptr(t->dst_cache), dst);
-
-}
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_set);
-
-void ip6_tnl_dst_destroy(struct ip6_tnl *t)
-{
-	if (!t->dst_cache)
-		return;
-
-	ip6_tnl_dst_reset(t);
-	free_percpu(t->dst_cache);
-}
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_destroy);
-
-int ip6_tnl_dst_init(struct ip6_tnl *t)
-{
-	int i;
-
-	t->dst_cache = alloc_percpu(struct ip6_tnl_dst);
-	if (!t->dst_cache)
-		return -ENOMEM;
-
-	for_each_possible_cpu(i)
-		seqlock_init(&per_cpu_ptr(t->dst_cache, i)->lock);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ip6_tnl_dst_init);
-
 /**
  * ip6_tnl_lookup - fetch tunnel matching the end-point addresses
  *   @remote: the address of the tunnel exit-point
@@ -329,7 +238,7 @@ static void ip6_dev_free(struct net_device *dev)
 {
 	struct ip6_tnl *t = netdev_priv(dev);
 
-	ip6_tnl_dst_destroy(t);
+	dst_cache_destroy(&t->dst_cache);
 	free_percpu(dev->tstats);
 	free_netdev(dev);
 }
@@ -462,7 +371,7 @@ ip6_tnl_dev_uninit(struct net_device *dev)
 		RCU_INIT_POINTER(ip6n->tnls_wc[0], NULL);
 	else
 		ip6_tnl_unlink(ip6n, t);
-	ip6_tnl_dst_reset(t);
+	dst_cache_reset(&t->dst_cache);
 	dev_put(dev);
 }
 
@@ -1069,7 +978,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 		memcpy(&fl6->daddr, addr6, sizeof(fl6->daddr));
 		neigh_release(neigh);
 	} else if (!fl6->flowi6_mark)
-		dst = ip6_tnl_dst_get(t);
+		dst = dst_cache_get(&t->dst_cache);
 
 	if (!ip6_tnl_xmit_ctl(t, &fl6->saddr, &fl6->daddr))
 		goto tx_err_link_failure;
@@ -1133,7 +1042,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb,
 	}
 
 	if (!fl6->flowi6_mark && ndst)
-		ip6_tnl_dst_set(t, ndst);
+		dst_cache_set_ip6(&t->dst_cache, ndst, &fl6->saddr);
 	skb_dst_set(skb, dst);
 
 	skb->transport_header = skb->network_header;
@@ -1366,7 +1275,7 @@ ip6_tnl_change(struct ip6_tnl *t, const struct __ip6_tnl_parm *p)
 	t->parms.flowinfo = p->flowinfo;
 	t->parms.link = p->link;
 	t->parms.proto = p->proto;
-	ip6_tnl_dst_reset(t);
+	dst_cache_reset(&t->dst_cache);
 	ip6_tnl_link_config(t);
 	return 0;
 }
@@ -1637,7 +1546,7 @@ ip6_tnl_dev_init_gen(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
-	ret = ip6_tnl_dst_init(t);
+	ret = dst_cache_init(&t->dst_cache, GFP_KERNEL);
 	if (ret) {
 		free_percpu(dev->tstats);
 		dev->tstats = NULL;
diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c
index 0a8610b..d90a11f 100644
--- a/net/ipv6/ip6_vti.c
+++ b/net/ipv6/ip6_vti.c
@@ -640,7 +640,7 @@ vti6_tnl_change(struct ip6_tnl *t, const struct __ip6_tnl_parm *p)
 	t->parms.i_key = p->i_key;
 	t->parms.o_key = p->o_key;
 	t->parms.proto = p->proto;
-	ip6_tnl_dst_reset(t);
+	dst_cache_reset(&t->dst_cache);
 	vti6_link_config(t);
 	return 0;
 }
-- 
cgit v0.10.2


From e09acddf873bf775b208b452a4c3a3fd26fa9427 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:55 +0100
Subject: ip_tunnel: replace dst_cache with generic implementation

The current ip_tunnel cache implementation is prone to a race
that will cause the wrong dst to be cached on cuncurrent dst cache
miss and ip tunnel update via netlink.

Replacing with the generic implementation fix the issue.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index bc439f3..fd36936 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -13,6 +13,7 @@
 #include <net/netns/generic.h>
 #include <net/rtnetlink.h>
 #include <net/lwtunnel.h>
+#include <net/dst_cache.h>
 
 #if IS_ENABLED(CONFIG_IPV6)
 #include <net/ipv6.h>
@@ -85,11 +86,6 @@ struct ip_tunnel_prl_entry {
 	struct rcu_head			rcu_head;
 };
 
-struct ip_tunnel_dst {
-	struct dst_entry __rcu 		*dst;
-	__be32				 saddr;
-};
-
 struct metadata_dst;
 
 struct ip_tunnel {
@@ -108,7 +104,7 @@ struct ip_tunnel {
 	int		tun_hlen;	/* Precalculated header length */
 	int		mlink;
 
-	struct ip_tunnel_dst __percpu *dst_cache;
+	struct dst_cache dst_cache;
 
 	struct ip_tunnel_parm parms;
 
@@ -247,7 +243,6 @@ int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[],
 int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
 		      struct ip_tunnel_parm *p);
 void ip_tunnel_setup(struct net_device *dev, int net_id);
-void ip_tunnel_dst_reset_all(struct ip_tunnel *t);
 int ip_tunnel_encap_setup(struct ip_tunnel *t,
 			  struct ip_tunnel_encap *ipencap);
 
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 7758247..395d827 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -186,6 +186,7 @@ config NET_IPGRE_DEMUX
 
 config NET_IP_TUNNEL
 	tristate
+	select DST_CACHE
 	default n
 
 config NET_IPGRE
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index c7bd72e..4569da7 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -68,61 +68,6 @@ static unsigned int ip_tunnel_hash(__be32 key, __be32 remote)
 			 IP_TNL_HASH_BITS);
 }
 
-static void __tunnel_dst_set(struct ip_tunnel_dst *idst,
-			     struct dst_entry *dst, __be32 saddr)
-{
-	struct dst_entry *old_dst;
-
-	dst_clone(dst);
-	old_dst = xchg((__force struct dst_entry **)&idst->dst, dst);
-	dst_release(old_dst);
-	idst->saddr = saddr;
-}
-
-static noinline void tunnel_dst_set(struct ip_tunnel *t,
-			   struct dst_entry *dst, __be32 saddr)
-{
-	__tunnel_dst_set(raw_cpu_ptr(t->dst_cache), dst, saddr);
-}
-
-static void tunnel_dst_reset(struct ip_tunnel *t)
-{
-	tunnel_dst_set(t, NULL, 0);
-}
-
-void ip_tunnel_dst_reset_all(struct ip_tunnel *t)
-{
-	int i;
-
-	for_each_possible_cpu(i)
-		__tunnel_dst_set(per_cpu_ptr(t->dst_cache, i), NULL, 0);
-}
-EXPORT_SYMBOL(ip_tunnel_dst_reset_all);
-
-static struct rtable *tunnel_rtable_get(struct ip_tunnel *t,
-					u32 cookie, __be32 *saddr)
-{
-	struct ip_tunnel_dst *idst;
-	struct dst_entry *dst;
-
-	rcu_read_lock();
-	idst = raw_cpu_ptr(t->dst_cache);
-	dst = rcu_dereference(idst->dst);
-	if (dst && !atomic_inc_not_zero(&dst->__refcnt))
-		dst = NULL;
-	if (dst) {
-		if (!dst->obsolete || dst->ops->check(dst, cookie)) {
-			*saddr = idst->saddr;
-		} else {
-			tunnel_dst_reset(t);
-			dst_release(dst);
-			dst = NULL;
-		}
-	}
-	rcu_read_unlock();
-	return (struct rtable *)dst;
-}
-
 static bool ip_tunnel_key_match(const struct ip_tunnel_parm *p,
 				__be16 flags, __be32 key)
 {
@@ -381,7 +326,8 @@ static int ip_tunnel_bind_dev(struct net_device *dev)
 
 		if (!IS_ERR(rt)) {
 			tdev = rt->dst.dev;
-			tunnel_dst_set(tunnel, &rt->dst, fl4.saddr);
+			dst_cache_set_ip4(&tunnel->dst_cache, &rt->dst,
+					  fl4.saddr);
 			ip_rt_put(rt);
 		}
 		if (dev->type != ARPHRD_ETHER)
@@ -729,7 +675,8 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 	if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0)
 		goto tx_error;
 
-	rt = connected ? tunnel_rtable_get(tunnel, 0, &fl4.saddr) : NULL;
+	rt = connected ? dst_cache_get_ip4(&tunnel->dst_cache, &fl4.saddr) :
+			 NULL;
 
 	if (!rt) {
 		rt = ip_route_output_key(tunnel->net, &fl4);
@@ -739,7 +686,8 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 			goto tx_error;
 		}
 		if (connected)
-			tunnel_dst_set(tunnel, &rt->dst, fl4.saddr);
+			dst_cache_set_ip4(&tunnel->dst_cache, &rt->dst,
+					  fl4.saddr);
 	}
 
 	if (rt->dst.dev == dev) {
@@ -836,7 +784,7 @@ static void ip_tunnel_update(struct ip_tunnel_net *itn,
 		if (set_mtu)
 			dev->mtu = mtu;
 	}
-	ip_tunnel_dst_reset_all(t);
+	dst_cache_reset(&t->dst_cache);
 	netdev_state_change(dev);
 }
 
@@ -961,7 +909,7 @@ static void ip_tunnel_dev_free(struct net_device *dev)
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 
 	gro_cells_destroy(&tunnel->gro_cells);
-	free_percpu(tunnel->dst_cache);
+	dst_cache_destroy(&tunnel->dst_cache);
 	free_percpu(dev->tstats);
 	free_netdev(dev);
 }
@@ -1155,15 +1103,15 @@ int ip_tunnel_init(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
-	tunnel->dst_cache = alloc_percpu(struct ip_tunnel_dst);
-	if (!tunnel->dst_cache) {
+	err = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
+	if (err) {
 		free_percpu(dev->tstats);
-		return -ENOMEM;
+		return err;
 	}
 
 	err = gro_cells_init(&tunnel->gro_cells, dev);
 	if (err) {
-		free_percpu(tunnel->dst_cache);
+		dst_cache_destroy(&tunnel->dst_cache);
 		free_percpu(dev->tstats);
 		return err;
 	}
@@ -1193,7 +1141,7 @@ void ip_tunnel_uninit(struct net_device *dev)
 	if (itn->fb_tunnel_dev != dev)
 		ip_tunnel_del(itn, netdev_priv(dev));
 
-	ip_tunnel_dst_reset_all(tunnel);
+	dst_cache_reset(&tunnel->dst_cache);
 }
 EXPORT_SYMBOL_GPL(ip_tunnel_uninit);
 
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 9a6b407..0625ac6 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -475,7 +475,7 @@ static void ipip6_tunnel_uninit(struct net_device *dev)
 		ipip6_tunnel_unlink(sitn, tunnel);
 		ipip6_tunnel_del_prl(tunnel, NULL);
 	}
-	ip_tunnel_dst_reset_all(tunnel);
+	dst_cache_reset(&tunnel->dst_cache);
 	dev_put(dev);
 }
 
@@ -1093,7 +1093,7 @@ static void ipip6_tunnel_update(struct ip_tunnel *t, struct ip_tunnel_parm *p)
 		t->parms.link = p->link;
 		ipip6_tunnel_bind_dev(t->dev);
 	}
-	ip_tunnel_dst_reset_all(t);
+	dst_cache_reset(&t->dst_cache);
 	netdev_state_change(t->dev);
 }
 
@@ -1124,7 +1124,7 @@ static int ipip6_tunnel_update_6rd(struct ip_tunnel *t,
 	t->ip6rd.relay_prefix = relay_prefix;
 	t->ip6rd.prefixlen = ip6rd->prefixlen;
 	t->ip6rd.relay_prefixlen = ip6rd->relay_prefixlen;
-	ip_tunnel_dst_reset_all(t);
+	dst_cache_reset(&t->dst_cache);
 	netdev_state_change(t->dev);
 	return 0;
 }
@@ -1278,7 +1278,7 @@ ipip6_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 			err = ipip6_tunnel_add_prl(t, &prl, cmd == SIOCCHGPRL);
 			break;
 		}
-		ip_tunnel_dst_reset_all(t);
+		dst_cache_reset(&t->dst_cache);
 		netdev_state_change(dev);
 		break;
 
@@ -1339,7 +1339,7 @@ static void ipip6_dev_free(struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 
-	free_percpu(tunnel->dst_cache);
+	dst_cache_destroy(&tunnel->dst_cache);
 	free_percpu(dev->tstats);
 	free_netdev(dev);
 }
@@ -1372,6 +1372,7 @@ static void ipip6_tunnel_setup(struct net_device *dev)
 static int ipip6_tunnel_init(struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
+	int err;
 
 	tunnel->dev = dev;
 	tunnel->net = dev_net(dev);
@@ -1382,10 +1383,10 @@ static int ipip6_tunnel_init(struct net_device *dev)
 	if (!dev->tstats)
 		return -ENOMEM;
 
-	tunnel->dst_cache = alloc_percpu(struct ip_tunnel_dst);
-	if (!tunnel->dst_cache) {
+	err = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
+	if (err) {
 		free_percpu(dev->tstats);
-		return -ENOMEM;
+		return err;
 	}
 
 	return 0;
-- 
cgit v0.10.2


From 0c1d70af924b966cc71e9e48920b2b635441aa50 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:56 +0100
Subject: net: use dst_cache for vxlan device

In case of UDP traffic with datagram length
below MTU this give about 3% performance increase
when tunneling over ipv4 and about 70% when
tunneling over ipv6.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 0a23c64..ad67303 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -480,6 +480,8 @@ static int vxlan_fdb_replace(struct vxlan_fdb *f,
 	rd = list_first_entry_or_null(&f->remotes, struct vxlan_rdst, list);
 	if (!rd)
 		return 0;
+
+	dst_cache_reset(&rd->dst_cache);
 	rd->remote_ip = *ip;
 	rd->remote_port = port;
 	rd->remote_vni = vni;
@@ -501,6 +503,12 @@ static int vxlan_fdb_append(struct vxlan_fdb *f,
 	rd = kmalloc(sizeof(*rd), GFP_ATOMIC);
 	if (rd == NULL)
 		return -ENOBUFS;
+
+	if (dst_cache_init(&rd->dst_cache, GFP_ATOMIC)) {
+		kfree(rd);
+		return -ENOBUFS;
+	}
+
 	rd->remote_ip = *ip;
 	rd->remote_port = port;
 	rd->remote_vni = vni;
@@ -749,8 +757,10 @@ static void vxlan_fdb_free(struct rcu_head *head)
 	struct vxlan_fdb *f = container_of(head, struct vxlan_fdb, rcu);
 	struct vxlan_rdst *rd, *nd;
 
-	list_for_each_entry_safe(rd, nd, &f->remotes, list)
+	list_for_each_entry_safe(rd, nd, &f->remotes, list) {
+		dst_cache_destroy(&rd->dst_cache);
 		kfree(rd);
+	}
 	kfree(f);
 }
 
@@ -1754,11 +1764,24 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 
 static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 				      struct sk_buff *skb, int oif, u8 tos,
-				      __be32 daddr, __be32 *saddr)
+				      __be32 daddr, __be32 *saddr,
+				      struct dst_cache *dst_cache,
+				      struct ip_tunnel_info *info)
 {
 	struct rtable *rt = NULL;
+	bool use_cache = false;
 	struct flowi4 fl4;
 
+	/* when the ip_tunnel_info is availble, the tos used for lookup is
+	 * packet independent, so we can use the cache
+	 */
+	if (dst_cache && !skb->mark && (!tos || info)) {
+		use_cache = true;
+		rt = dst_cache_get_ip4(dst_cache, saddr);
+		if (rt)
+			return rt;
+	}
+
 	memset(&fl4, 0, sizeof(fl4));
 	fl4.flowi4_oif = oif;
 	fl4.flowi4_tos = RT_TOS(tos);
@@ -1768,8 +1791,11 @@ static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 	fl4.saddr = vxlan->cfg.saddr.sin.sin_addr.s_addr;
 
 	rt = ip_route_output_key(vxlan->net, &fl4);
-	if (!IS_ERR(rt))
+	if (!IS_ERR(rt)) {
 		*saddr = fl4.saddr;
+		if (use_cache)
+			dst_cache_set_ip4(dst_cache, &rt->dst, fl4.saddr);
+	}
 	return rt;
 }
 
@@ -1777,12 +1803,21 @@ static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 					  struct sk_buff *skb, int oif,
 					  const struct in6_addr *daddr,
-					  struct in6_addr *saddr)
+					  struct in6_addr *saddr,
+					  struct dst_cache *dst_cache)
 {
+	bool use_cache = false;
 	struct dst_entry *ndst;
 	struct flowi6 fl6;
 	int err;
 
+	if (dst_cache && !skb->mark) {
+		use_cache = true;
+		ndst = dst_cache_get_ip6(dst_cache, saddr);
+		if (ndst)
+			return ndst;
+	}
+
 	memset(&fl6, 0, sizeof(fl6));
 	fl6.flowi6_oif = oif;
 	fl6.daddr = *daddr;
@@ -1797,6 +1832,8 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 		return ERR_PTR(err);
 
 	*saddr = fl6.saddr;
+	if (use_cache)
+		dst_cache_set_ip6(dst_cache, ndst, saddr);
 	return ndst;
 }
 #endif
@@ -1938,7 +1975,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 
 		rt = vxlan_get_route(vxlan, skb,
 				     rdst ? rdst->remote_ifindex : 0, tos,
-				     dst->sin.sin_addr.s_addr, &saddr);
+				     dst->sin.sin_addr.s_addr, &saddr,
+				     rdst ? &rdst->dst_cache : NULL, info);
 		if (IS_ERR(rt)) {
 			netdev_dbg(dev, "no route to %pI4\n",
 				   &dst->sin.sin_addr.s_addr);
@@ -1990,7 +2028,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 
 		ndst = vxlan6_get_route(vxlan, skb,
 					rdst ? rdst->remote_ifindex : 0,
-					&dst->sin6.sin6_addr, &saddr);
+					&dst->sin6.sin6_addr, &saddr,
+					rdst ? &rdst->dst_cache : NULL);
 		if (IS_ERR(ndst)) {
 			netdev_dbg(dev, "no route to %pI6\n",
 				   &dst->sin6.sin6_addr);
@@ -2331,7 +2370,7 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 			return -EINVAL;
 		rt = vxlan_get_route(vxlan, skb, 0, info->key.tos,
 				     info->key.u.ipv4.dst,
-				     &info->key.u.ipv4.src);
+				     &info->key.u.ipv4.src, NULL, info);
 		if (IS_ERR(rt))
 			return PTR_ERR(rt);
 		ip_rt_put(rt);
@@ -2343,7 +2382,7 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 			return -EINVAL;
 		ndst = vxlan6_get_route(vxlan, skb, 0,
 					&info->key.u.ipv6.dst,
-					&info->key.u.ipv6.src);
+					&info->key.u.ipv6.src, NULL);
 		if (IS_ERR(ndst))
 			return PTR_ERR(ndst);
 		dst_release(ndst);
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 25bd919..b314e4a 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -148,6 +148,7 @@ struct vxlan_rdst {
 	u32			 remote_ifindex;
 	struct list_head	 list;
 	struct rcu_head		 rcu;
+	struct dst_cache	 dst_cache;
 };
 
 struct vxlan_config {
-- 
cgit v0.10.2


From d71785ffc7e7cae3fbdc4ea8a9d05b7a1c59f7b8 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:57 +0100
Subject: net: add dst_cache to ovs vxlan lwtunnel

In case of UDP traffic with datagram length
below MTU this give about 2% performance increase
when tunneling over ipv4 and about 60% when tunneling
over ipv6

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index ad67303..ee1206d 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1775,7 +1775,7 @@ static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 	/* when the ip_tunnel_info is availble, the tos used for lookup is
 	 * packet independent, so we can use the cache
 	 */
-	if (dst_cache && !skb->mark && (!tos || info)) {
+	if (!skb->mark && (!tos || info)) {
 		use_cache = true;
 		rt = dst_cache_get_ip4(dst_cache, saddr);
 		if (rt)
@@ -1806,13 +1806,11 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 					  struct in6_addr *saddr,
 					  struct dst_cache *dst_cache)
 {
-	bool use_cache = false;
 	struct dst_entry *ndst;
 	struct flowi6 fl6;
 	int err;
 
-	if (dst_cache && !skb->mark) {
-		use_cache = true;
+	if (!skb->mark) {
 		ndst = dst_cache_get_ip6(dst_cache, saddr);
 		if (ndst)
 			return ndst;
@@ -1832,7 +1830,7 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 		return ERR_PTR(err);
 
 	*saddr = fl6.saddr;
-	if (use_cache)
+	if (!skb->mark)
 		dst_cache_set_ip6(dst_cache, ndst, saddr);
 	return ndst;
 }
@@ -1886,6 +1884,7 @@ static void vxlan_encap_bypass(struct sk_buff *skb, struct vxlan_dev *src_vxlan,
 static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			   struct vxlan_rdst *rdst, bool did_rsc)
 {
+	struct dst_cache *dst_cache;
 	struct ip_tunnel_info *info;
 	struct vxlan_dev *vxlan = netdev_priv(dev);
 	struct sock *sk;
@@ -1910,6 +1909,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		dst_port = rdst->remote_port ? rdst->remote_port : vxlan->cfg.dst_port;
 		vni = rdst->remote_vni;
 		dst = &rdst->remote_ip;
+		dst_cache = &rdst->dst_cache;
 	} else {
 		if (!info) {
 			WARN_ONCE(1, "%s: Missing encapsulation instructions\n",
@@ -1924,6 +1924,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		else
 			remote_ip.sin6.sin6_addr = info->key.u.ipv6.dst;
 		dst = &remote_ip;
+		dst_cache = &info->dst_cache;
 	}
 
 	if (vxlan_addr_any(dst)) {
@@ -1976,7 +1977,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		rt = vxlan_get_route(vxlan, skb,
 				     rdst ? rdst->remote_ifindex : 0, tos,
 				     dst->sin.sin_addr.s_addr, &saddr,
-				     rdst ? &rdst->dst_cache : NULL, info);
+				     dst_cache, info);
 		if (IS_ERR(rt)) {
 			netdev_dbg(dev, "no route to %pI4\n",
 				   &dst->sin.sin_addr.s_addr);
@@ -2029,7 +2030,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		ndst = vxlan6_get_route(vxlan, skb,
 					rdst ? rdst->remote_ifindex : 0,
 					&dst->sin6.sin6_addr, &saddr,
-					rdst ? &rdst->dst_cache : NULL);
+					dst_cache);
 		if (IS_ERR(ndst)) {
 			netdev_dbg(dev, "no route to %pI6\n",
 				   &dst->sin6.sin6_addr);
diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h
index 30a56ab..84b833a 100644
--- a/include/net/dst_metadata.h
+++ b/include/net/dst_metadata.h
@@ -62,6 +62,7 @@ static inline int skb_metadata_dst_cmp(const struct sk_buff *skb_a,
 		      sizeof(a->u.tun_info) + a->u.tun_info.options_len);
 }
 
+void metadata_dst_free(struct metadata_dst *);
 struct metadata_dst *metadata_dst_alloc(u8 optslen, gfp_t flags);
 struct metadata_dst __percpu *metadata_dst_alloc_percpu(u8 optslen, gfp_t flags);
 
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index fd36936..87408ab 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -58,6 +58,9 @@ struct ip_tunnel_key {
 
 struct ip_tunnel_info {
 	struct ip_tunnel_key	key;
+#ifdef CONFIG_DST_CACHE
+	struct dst_cache	dst_cache;
+#endif
 	u8			options_len;
 	u8			mode;
 };
diff --git a/net/core/dst.c b/net/core/dst.c
index a1656e3..b5cbbe0 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -265,7 +265,7 @@ again:
 	lwtstate_put(dst->lwtstate);
 
 	if (dst->flags & DST_METADATA)
-		kfree(dst);
+		metadata_dst_free((struct metadata_dst *)dst);
 	else
 		kmem_cache_free(dst->ops->kmem_cachep, dst);
 
@@ -395,6 +395,14 @@ struct metadata_dst *metadata_dst_alloc(u8 optslen, gfp_t flags)
 }
 EXPORT_SYMBOL_GPL(metadata_dst_alloc);
 
+void metadata_dst_free(struct metadata_dst *md_dst)
+{
+#ifdef CONFIG_DST_CACHE
+	dst_cache_destroy(&md_dst->u.tun_info.dst_cache);
+#endif
+	kfree(md_dst);
+}
+
 struct metadata_dst __percpu *metadata_dst_alloc_percpu(u8 optslen, gfp_t flags)
 {
 	int cpu;
diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig
index d143aa9..cd5fd9d 100644
--- a/net/openvswitch/Kconfig
+++ b/net/openvswitch/Kconfig
@@ -10,6 +10,7 @@ config OPENVSWITCH
 	select LIBCRC32C
 	select MPLS
 	select NET_MPLS_GSO
+	select DST_CACHE
 	---help---
 	  Open vSwitch is a multilayer Ethernet switch targeted at virtualized
 	  environments.  In addition to supporting a variety of features
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index d1bd4a4..58b8efc 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -1959,6 +1959,12 @@ static int validate_and_copy_set_tun(const struct nlattr *attr,
 	if (!tun_dst)
 		return -ENOMEM;
 
+	err = dst_cache_init(&tun_dst->u.tun_info.dst_cache, GFP_KERNEL);
+	if (err) {
+		dst_release((struct dst_entry *)tun_dst);
+		return err;
+	}
+
 	a = __add_action(sfa, OVS_KEY_ATTR_TUNNEL_INFO, NULL,
 			 sizeof(*ovs_tun), log);
 	if (IS_ERR(a)) {
-- 
cgit v0.10.2


From 468dfffcd762cbb2777ec5a76bc21e3748ebf47e Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:58 +0100
Subject: geneve: add dst caching support

use generic dst implementation for both plain geneve devices and
lwtunnels.

In case of UDP traffic with datagram length below MTU this give
about 2% performance increase for plain geneve tunnel over ipv4,
about 65% performance increase for ipv6 tunnel.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 0b14ac3..6f20813 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -72,6 +72,7 @@ struct geneve_dev {
 	bool		   collect_md;
 	struct gro_cells   gro_cells;
 	u32		   flags;
+	struct dst_cache   dst_cache;
 };
 
 /* Geneve device flags */
@@ -297,6 +298,13 @@ static int geneve_init(struct net_device *dev)
 		return err;
 	}
 
+	err = dst_cache_init(&geneve->dst_cache, GFP_KERNEL);
+	if (err) {
+		free_percpu(dev->tstats);
+		gro_cells_destroy(&geneve->gro_cells);
+		return err;
+	}
+
 	return 0;
 }
 
@@ -304,6 +312,7 @@ static void geneve_uninit(struct net_device *dev)
 {
 	struct geneve_dev *geneve = netdev_priv(dev);
 
+	dst_cache_destroy(&geneve->dst_cache);
 	gro_cells_destroy(&geneve->gro_cells);
 	free_percpu(dev->tstats);
 }
@@ -753,7 +762,9 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb,
 				       struct ip_tunnel_info *info)
 {
 	struct geneve_dev *geneve = netdev_priv(dev);
+	struct dst_cache *dst_cache;
 	struct rtable *rt = NULL;
+	bool use_cache = true;
 	__u8 tos;
 
 	memset(fl4, 0, sizeof(*fl4));
@@ -764,16 +775,26 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb,
 		fl4->daddr = info->key.u.ipv4.dst;
 		fl4->saddr = info->key.u.ipv4.src;
 		fl4->flowi4_tos = RT_TOS(info->key.tos);
+		dst_cache = &info->dst_cache;
 	} else {
 		tos = geneve->tos;
 		if (tos == 1) {
 			const struct iphdr *iip = ip_hdr(skb);
 
 			tos = ip_tunnel_get_dsfield(iip, skb);
+			use_cache = false;
 		}
 
 		fl4->flowi4_tos = RT_TOS(tos);
 		fl4->daddr = geneve->remote.sin.sin_addr.s_addr;
+		dst_cache = &geneve->dst_cache;
+	}
+
+	use_cache = use_cache && !skb->mark;
+	if (use_cache) {
+		rt = dst_cache_get_ip4(dst_cache, &fl4->saddr);
+		if (rt)
+			return rt;
 	}
 
 	rt = ip_route_output_key(geneve->net, fl4);
@@ -786,6 +807,8 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb,
 		ip_rt_put(rt);
 		return ERR_PTR(-ELOOP);
 	}
+	if (use_cache)
+		dst_cache_set_ip4(dst_cache, &rt->dst, fl4->saddr);
 	return rt;
 }
 
@@ -798,6 +821,8 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 	struct geneve_dev *geneve = netdev_priv(dev);
 	struct geneve_sock *gs6 = geneve->sock6;
 	struct dst_entry *dst = NULL;
+	struct dst_cache *dst_cache;
+	bool use_cache = true;
 	__u8 prio;
 
 	memset(fl6, 0, sizeof(*fl6));
@@ -808,16 +833,26 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 		fl6->daddr = info->key.u.ipv6.dst;
 		fl6->saddr = info->key.u.ipv6.src;
 		fl6->flowi6_tos = RT_TOS(info->key.tos);
+		dst_cache = &info->dst_cache;
 	} else {
 		prio = geneve->tos;
 		if (prio == 1) {
 			const struct iphdr *iip = ip_hdr(skb);
 
 			prio = ip_tunnel_get_dsfield(iip, skb);
+			use_cache = false;
 		}
 
 		fl6->flowi6_tos = RT_TOS(prio);
 		fl6->daddr = geneve->remote.sin6.sin6_addr;
+		dst_cache = &geneve->dst_cache;
+	}
+
+	use_cache = use_cache && !skb->mark;
+	if (use_cache) {
+		dst = dst_cache_get_ip6(dst_cache, &fl6->saddr);
+		if (dst)
+			return dst;
 	}
 
 	if (ipv6_stub->ipv6_dst_lookup(geneve->net, gs6->sock->sk, &dst, fl6)) {
@@ -830,6 +865,8 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 		return ERR_PTR(-ELOOP);
 	}
 
+	if (use_cache)
+		dst_cache_set_ip6(dst_cache, dst, &fl6->saddr);
 	return dst;
 }
 #endif
@@ -1272,6 +1309,8 @@ static int geneve_configure(struct net *net, struct net_device *dev,
 			return -EPERM;
 	}
 
+	dst_cache_reset(&geneve->dst_cache);
+
 	err = register_netdevice(dev);
 	if (err)
 		return err;
-- 
cgit v0.10.2


From 3c1cb4d2604c03779a6c9485204e2a80be6c28f0 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 12 Feb 2016 15:43:59 +0100
Subject: net/ipv4: add dst cache support for gre lwtunnels

In case of UDP traffic with datagram length below MTU this
gives about 4% performance increase

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Suggested-and-Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 65748db..917c2c1 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -540,9 +540,16 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto err_free_skb;
 
 	key = &tun_info->key;
-	rt = gre_get_rt(skb, dev, &fl, key);
-	if (IS_ERR(rt))
-		goto err_free_skb;
+	rt = !skb->mark ? dst_cache_get_ip4(&tun_info->dst_cache, &fl.saddr) :
+			 NULL;
+	if (!rt) {
+		rt = gre_get_rt(skb, dev, &fl, key);
+		if (IS_ERR(rt))
+				goto err_free_skb;
+		if (!skb->mark)
+			dst_cache_set_ip4(&tun_info->dst_cache, &rt->dst,
+					  fl.saddr);
+	}
 
 	tunnel_hlen = ip_gre_calc_hlen(key->tun_flags);
 
-- 
cgit v0.10.2


From cd9b266095f422267bddbec88f9098b48ea548fc Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 11 Feb 2016 22:02:53 -0800
Subject: tcp: add tcpi_min_rtt and tcpi_notsent_bytes to tcp_info

tcpi_min_rtt reports the minimal rtt observed by TCP stack for the flow,
in usec unit. Might be ~0U if not yet known.

tcpi_notsent_bytes reports the amount of bytes in the write queue that
were not yet sent.

This is done in a single patch to not add a temporary 32bit padding hole
in tcp_info.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index 65a77b0..fe95446 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -196,6 +196,9 @@ struct tcp_info {
 	__u64	tcpi_bytes_received; /* RFC4898 tcpEStatsAppHCThruOctetsReceived */
 	__u32	tcpi_segs_out;	     /* RFC4898 tcpEStatsPerfSegsOut */
 	__u32	tcpi_segs_in;	     /* RFC4898 tcpEStatsPerfSegsIn */
+
+	__u32	tcpi_notsent_bytes;
+	__u32	tcpi_min_rtt;
 };
 
 /* for TCP_MD5SIG socket option */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 014f18e..f93150d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2642,6 +2642,7 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info)
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	u32 now = tcp_time_stamp;
 	unsigned int start;
+	int notsent_bytes;
 	u64 rate64;
 	u32 rate;
 
@@ -2722,6 +2723,11 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info)
 	} while (u64_stats_fetch_retry_irq(&tp->syncp, start));
 	info->tcpi_segs_out = tp->segs_out;
 	info->tcpi_segs_in = tp->segs_in;
+
+	notsent_bytes = READ_ONCE(tp->write_seq) - READ_ONCE(tp->snd_nxt);
+	info->tcpi_notsent_bytes = max(0, notsent_bytes);
+
+	info->tcpi_min_rtt = tcp_min_rtt(tp);
 }
 EXPORT_SYMBOL_GPL(tcp_get_info);
 
-- 
cgit v0.10.2


From 2fcc1a401eb06ff41cd551672ea588c6efd8c4e6 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:43 -0800
Subject: i40e: AQ Add VXLAN-GPE tunnel type

Add the new Cisco VXLAN-GPE cloud tunnel type for the Add Cloud Filter
and UDP tunnel AQ commands.

Change-ID: I2c093c7d79726c7fca08a36e5c63581a905da3d2
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 7e96615..5effb91 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -1278,6 +1278,8 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_NVGRE_OMAC		1
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_RESERVED		4
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_VXLAN_GPE		5
 
 #define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_MAC	0x2000
 #define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_INNER_MAC	0x4000
@@ -2229,6 +2231,7 @@ struct i40e_aqc_add_udp_tunnel {
 #define I40E_AQC_TUNNEL_TYPE_VXLAN	0x00
 #define I40E_AQC_TUNNEL_TYPE_NGE	0x01
 #define I40E_AQC_TUNNEL_TYPE_TEREDO	0x10
+#define I40E_AQC_TUNNEL_TYPE_VXLAN_GPE	0x11
 	u8	reserved1[10];
 };
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 2c8f3ed..bdb18f6 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -1275,6 +1275,8 @@ struct i40e_aqc_add_remove_cloud_filters_element_data {
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_NVGRE_OMAC		1
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_GENEVE		2
 #define I40E_AQC_ADD_CLOUD_TNL_TYPE_IP			3
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_RESERVED		4
+#define I40E_AQC_ADD_CLOUD_TNL_TYPE_VXLAN_GPE		5
 
 #define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_OUTER_MAC	0x2000
 #define I40E_AQC_ADD_CLOUD_FLAGS_SHARED_INNER_MAC	0x4000
@@ -2121,6 +2123,7 @@ struct i40e_aqc_add_udp_tunnel {
 #define I40E_AQC_TUNNEL_TYPE_VXLAN	0x00
 #define I40E_AQC_TUNNEL_TYPE_NGE	0x01
 #define I40E_AQC_TUNNEL_TYPE_TEREDO	0x10
+#define I40E_AQC_TUNNEL_TYPE_VXLAN_GPE	0x11
 	u8	reserved1[10];
 };
 
-- 
cgit v0.10.2


From 27090cbdc345cd5fd6f6e57a8986c504c27e4d23 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 12 Feb 2016 11:42:34 +0100
Subject: net: phy: spi_ks8995: include linux/gpio/consumer.h

The ks8995 phy driver just started using gpiod_* functions, which are
declared in linux/gpio/consumer.h, not linux/gpio.h, resulting in a
build failure in randconfig builds that do not have CONFIG_GPIOLIB
enabled:

drivers/net/phy/spi_ks8995.c: In function 'ks8995_probe':
drivers/net/phy/spi_ks8995.c:477:3: error: implicit declaration of function 'gpiod_set_value' [-Werror=implicit-function-declaration]

This changes the header inclusion so it builds in all configurations.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: cd6f288cbaab ("net: phy: spi_ks8995: add support for resetting switch using GPIO")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/spi_ks8995.c b/drivers/net/phy/spi_ks8995.c
index 5e7340f..b5d50d4 100644
--- a/drivers/net/phy/spi_ks8995.c
+++ b/drivers/net/phy/spi_ks8995.c
@@ -18,8 +18,8 @@
 #include <linux/module.h>
 #include <linux/delay.h>
 #include <linux/device.h>
+#include <linux/gpio/consumer.h>
 #include <linux/of.h>
-#include <linux/gpio.h>
 #include <linux/of_gpio.h>
 
 #include <linux/spi/spi.h>
-- 
cgit v0.10.2


From fa50d974d104113630d68b7d03233a6686230d0c Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 15 Feb 2016 12:11:27 +0200
Subject: ipv4: Namespaceify ip_default_ttl sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 848fe80..bc8f7f9 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -80,6 +80,7 @@ struct netns_ipv4 {
 	int sysctl_tcp_ecn;
 	int sysctl_tcp_ecn_fallback;
 
+	int sysctl_ip_default_ttl;
 	int sysctl_ip_no_pmtu_disc;
 	int sysctl_ip_fwd_use_pmtu;
 	int sysctl_ip_nonlocal_bind;
diff --git a/include/net/route.h b/include/net/route.h
index a3b9ef7..9b0a523 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -329,14 +329,13 @@ static inline int inet_iif(const struct sk_buff *skb)
 	return skb->skb_iif;
 }
 
-extern int sysctl_ip_default_ttl;
-
 static inline int ip4_dst_hoplimit(const struct dst_entry *dst)
 {
 	int hoplimit = dst_metric_raw(dst, RTAX_HOPLIMIT);
+	struct net *net = dev_net(dst->dev);
 
 	if (hoplimit == 0)
-		hoplimit = sysctl_ip_default_ttl;
+		hoplimit = net->ipv4.sysctl_ip_default_ttl;
 	return hoplimit;
 }
 
diff --git a/net/bridge/netfilter/nft_reject_bridge.c b/net/bridge/netfilter/nft_reject_bridge.c
index fdba3d9..adc8d72 100644
--- a/net/bridge/netfilter/nft_reject_bridge.c
+++ b/net/bridge/netfilter/nft_reject_bridge.c
@@ -48,6 +48,7 @@ static void nft_reject_br_send_v4_tcp_reset(struct sk_buff *oldskb,
 	struct iphdr *niph;
 	const struct tcphdr *oth;
 	struct tcphdr _oth;
+	struct net *net = sock_net(oldskb->sk);
 
 	if (!nft_bridge_iphdr_validate(oldskb))
 		return;
@@ -63,9 +64,9 @@ static void nft_reject_br_send_v4_tcp_reset(struct sk_buff *oldskb,
 
 	skb_reserve(nskb, LL_MAX_HEADER);
 	niph = nf_reject_iphdr_put(nskb, oldskb, IPPROTO_TCP,
-				   sysctl_ip_default_ttl);
+				   net->ipv4.sysctl_ip_default_ttl);
 	nf_reject_ip_tcphdr_put(nskb, oldskb, oth);
-	niph->ttl	= sysctl_ip_default_ttl;
+	niph->ttl	= net->ipv4.sysctl_ip_default_ttl;
 	niph->tot_len	= htons(nskb->len);
 	ip_send_check(niph);
 
@@ -85,6 +86,7 @@ static void nft_reject_br_send_v4_unreach(struct sk_buff *oldskb,
 	void *payload;
 	__wsum csum;
 	u8 proto;
+	struct net *net = sock_net(oldskb->sk);
 
 	if (oldskb->csum_bad || !nft_bridge_iphdr_validate(oldskb))
 		return;
@@ -119,7 +121,7 @@ static void nft_reject_br_send_v4_unreach(struct sk_buff *oldskb,
 
 	skb_reserve(nskb, LL_MAX_HEADER);
 	niph = nf_reject_iphdr_put(nskb, oldskb, IPPROTO_ICMP,
-				   sysctl_ip_default_ttl);
+				   net->ipv4.sysctl_ip_default_ttl);
 
 	skb_reset_transport_header(nskb);
 	icmph = (struct icmphdr *)skb_put(nskb, sizeof(struct icmphdr));
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 64878ef..f734c42 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -79,9 +79,6 @@
 #include <linux/netlink.h>
 #include <linux/tcp.h>
 
-int sysctl_ip_default_ttl __read_mostly = IPDEFTTL;
-EXPORT_SYMBOL(sysctl_ip_default_ttl);
-
 static int
 ip_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 	    unsigned int mtu,
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 92808f1..3f1befc 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -1341,10 +1341,13 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname,
 		val = inet->tos;
 		break;
 	case IP_TTL:
+	{
+		struct net *net = sock_net(sk);
 		val = (inet->uc_ttl == -1 ?
-		       sysctl_ip_default_ttl :
+		       net->ipv4.sysctl_ip_default_ttl :
 		       inet->uc_ttl);
 		break;
+	}
 	case IP_HDRINCL:
 		val = inet->hdrincl;
 		break;
diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c
index 5fdc556..7b8fbb35 100644
--- a/net/ipv4/netfilter/ipt_SYNPROXY.c
+++ b/net/ipv4/netfilter/ipt_SYNPROXY.c
@@ -21,6 +21,7 @@ static struct iphdr *
 synproxy_build_ip(struct sk_buff *skb, __be32 saddr, __be32 daddr)
 {
 	struct iphdr *iph;
+	struct net *net = sock_net(skb->sk);
 
 	skb_reset_network_header(skb);
 	iph = (struct iphdr *)skb_put(skb, sizeof(*iph));
@@ -29,7 +30,7 @@ synproxy_build_ip(struct sk_buff *skb, __be32 saddr, __be32 daddr)
 	iph->tos	= 0;
 	iph->id		= 0;
 	iph->frag_off	= htons(IP_DF);
-	iph->ttl	= sysctl_ip_default_ttl;
+	iph->ttl	= net->ipv4.sysctl_ip_default_ttl;
 	iph->protocol	= IPPROTO_TCP;
 	iph->check	= 0;
 	iph->saddr	= saddr;
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 3abd9d7..9f665b6 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -390,7 +390,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 
 	seq_printf(seq, "\nIp: %d %d",
 		   IPV4_DEVCONF_ALL(net, FORWARDING) ? 1 : 2,
-		   sysctl_ip_default_ttl);
+		   net->ipv4.sysctl_ip_default_ttl);
 
 	BUILD_BUG_ON(offsetof(struct ipstats_mib, mibs) != 0);
 	for (i = 0; snmp4_ipstats_list[i].name != NULL; i++)
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index b537338..a833a9f 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -283,15 +283,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "ip_default_ttl",
-		.data		= &sysctl_ip_default_ttl,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &ip_ttl_min,
-		.extra2		= &ip_ttl_max,
-	},
-	{
 		.procname	= "tcp_max_orphans",
 		.data		= &sysctl_tcp_max_orphans,
 		.maxlen		= sizeof(int),
@@ -753,6 +744,15 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
+		.procname	= "ip_default_ttl",
+		.data		= &init_net.ipv4.sysctl_ip_default_ttl,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &ip_ttl_min,
+		.extra2		= &ip_ttl_max,
+	},
+	{
 		.procname	= "ip_local_port_range",
 		.maxlen		= sizeof(init_net.ipv4.ip_local_ports.range),
 		.data		= &init_net.ipv4.ip_local_ports.range,
@@ -988,6 +988,8 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
 	if (!net->ipv4.sysctl_local_reserved_ports)
 		goto err_ports;
 
+	net->ipv4.sysctl_ip_default_ttl = IPDEFTTL;
+
 	return 0;
 
 err_ports:
-- 
cgit v0.10.2


From dcd87999d415d39cf2ae510bfed6b8206d778e1c Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 15 Feb 2016 12:11:28 +0200
Subject: igmp: net: Move igmp namespace init to correct file

When igmp related sysctl were namespacified their initializatin was
erroneously put into the tcp socket namespace constructor. This
patch moves the relevant code into the igmp namespace constructor to
keep things consistent.

Also sprinkle some #ifdefs to silence warnings

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 7c95335..2aea9f1 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1224,7 +1224,9 @@ static void igmp_group_dropped(struct ip_mc_list *im)
 static void igmp_group_added(struct ip_mc_list *im)
 {
 	struct in_device *in_dev = im->interface;
+#ifdef CONFIG_IP_MULTICAST
 	struct net *net = dev_net(in_dev->dev);
+#endif
 
 	if (im->loaded == 0) {
 		im->loaded = 1;
@@ -1316,7 +1318,9 @@ static void ip_mc_hash_remove(struct in_device *in_dev,
 void ip_mc_inc_group(struct in_device *in_dev, __be32 addr)
 {
 	struct ip_mc_list *im;
+#ifdef CONFIG_IP_MULTICAST
 	struct net *net = dev_net(in_dev->dev);
+#endif
 
 	ASSERT_RTNL();
 
@@ -1643,7 +1647,9 @@ void ip_mc_down(struct in_device *in_dev)
 
 void ip_mc_init_dev(struct in_device *in_dev)
 {
+#ifdef CONFIG_IP_MULTICAST
 	struct net *net = dev_net(in_dev->dev);
+#endif
 	ASSERT_RTNL();
 
 #ifdef CONFIG_IP_MULTICAST
@@ -1662,7 +1668,9 @@ void ip_mc_init_dev(struct in_device *in_dev)
 void ip_mc_up(struct in_device *in_dev)
 {
 	struct ip_mc_list *pmc;
+#ifdef CONFIG_IP_MULTICAST
 	struct net *net = dev_net(in_dev->dev);
+#endif
 
 	ASSERT_RTNL();
 
@@ -2923,6 +2931,12 @@ static int __net_init igmp_net_init(struct net *net)
 		goto out_sock;
 	}
 
+	/* Sysctl initialization */
+	net->ipv4.sysctl_igmp_max_memberships = 20;
+	net->ipv4.sysctl_igmp_max_msf = 10;
+	/* IGMP reports for link-local multicast groups are enabled by default */
+	net->ipv4.sysctl_igmp_llm_reports = 1;
+	net->ipv4.sysctl_igmp_qrv = 2;
 	return 0;
 
 out_sock:
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index ba5d014..3f872a6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2399,12 +2399,6 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT;
 	net->ipv4.sysctl_tcp_notsent_lowat = UINT_MAX;
 
-	net->ipv4.sysctl_igmp_max_memberships = 20;
-	net->ipv4.sysctl_igmp_max_msf = 10;
-	/* IGMP reports for link-local multicast groups are enabled by default */
-	net->ipv4.sysctl_igmp_llm_reports = 1;
-	net->ipv4.sysctl_igmp_qrv = 2;
-
 	return 0;
 fail:
 	tcp_sk_exit(net);
-- 
cgit v0.10.2


From 287b7f38fd6842e534db1783cead3843f7677b79 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 15 Feb 2016 12:11:29 +0200
Subject: ipv4: Namespacify ip_dynaddr sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip.h b/include/net/ip.h
index 1a98f1c..e3fb25d 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -248,9 +248,6 @@ extern int inet_peer_maxttl;
 /* From ip_input.c */
 extern int sysctl_ip_early_demux;
 
-/* From ip_output.c */
-extern int sysctl_ip_dynaddr;
-
 void ipfrag_init(void);
 
 void ip_static_sysctl_init(void);
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index bc8f7f9..b7e3fb2 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -84,6 +84,8 @@ struct netns_ipv4 {
 	int sysctl_ip_no_pmtu_disc;
 	int sysctl_ip_fwd_use_pmtu;
 	int sysctl_ip_nonlocal_bind;
+	/* Shall we try to damage output packets if routing dev changes? */
+	int sysctl_ip_dynaddr;
 
 	int sysctl_fwmark_reflect;
 	int sysctl_tcp_fwmark_accept;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index eade66d..209d1ed 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1095,12 +1095,6 @@ void inet_unregister_protosw(struct inet_protosw *p)
 }
 EXPORT_SYMBOL(inet_unregister_protosw);
 
-/*
- *      Shall we try to damage output packets if routing dev changes?
- */
-
-int sysctl_ip_dynaddr __read_mostly;
-
 static int inet_sk_reselect_saddr(struct sock *sk)
 {
 	struct inet_sock *inet = inet_sk(sk);
@@ -1131,7 +1125,7 @@ static int inet_sk_reselect_saddr(struct sock *sk)
 	if (new_saddr == old_saddr)
 		return 0;
 
-	if (sysctl_ip_dynaddr > 1) {
+	if (sock_net(sk)->ipv4.sysctl_ip_dynaddr > 1) {
 		pr_info("%s(): shifting inet->saddr from %pI4 to %pI4\n",
 			__func__, &old_saddr, &new_saddr);
 	}
@@ -1186,7 +1180,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 		 * Other protocols have to map its equivalent state to TCP_SYN_SENT.
 		 * DCCP maps its DCCP_REQUESTING state to TCP_SYN_SENT. -acme
 		 */
-		if (!sysctl_ip_dynaddr ||
+		if (!sock_net(sk)->ipv4.sysctl_ip_dynaddr ||
 		    sk->sk_state != TCP_SYN_SENT ||
 		    (sk->sk_userlocks & SOCK_BINDADDR_LOCK) ||
 		    (err = inet_sk_reselect_saddr(sk)) != 0)
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index a833a9f..04ac5b7 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -304,13 +304,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "ip_dynaddr",
-		.data		= &sysctl_ip_dynaddr,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "tcp_fastopen",
 		.data		= &sysctl_tcp_fastopen,
 		.maxlen		= sizeof(int),
@@ -744,6 +737,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
+		.procname	= "ip_dynaddr",
+		.data		= &init_net.ipv4.sysctl_ip_dynaddr,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{
 		.procname	= "ip_default_ttl",
 		.data		= &init_net.ipv4.sysctl_ip_default_ttl,
 		.maxlen		= sizeof(int),
@@ -989,6 +989,7 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
 		goto err_ports;
 
 	net->ipv4.sysctl_ip_default_ttl = IPDEFTTL;
+	net->ipv4.sysctl_ip_dynaddr = 0;
 
 	return 0;
 
-- 
cgit v0.10.2


From e21145a9871aa5ae07e01926105bb8e523d64095 Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 15 Feb 2016 12:11:30 +0200
Subject: ipv4: namespacify ip_early_demux sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip.h b/include/net/ip.h
index e3fb25d..cbb134b 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -245,9 +245,6 @@ extern int inet_peer_threshold;
 extern int inet_peer_minttl;
 extern int inet_peer_maxttl;
 
-/* From ip_input.c */
-extern int sysctl_ip_early_demux;
-
 void ipfrag_init(void);
 
 void ip_static_sysctl_init(void);
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index b7e3fb2..a69cde3 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -86,6 +86,7 @@ struct netns_ipv4 {
 	int sysctl_ip_nonlocal_bind;
 	/* Shall we try to damage output packets if routing dev changes? */
 	int sysctl_ip_dynaddr;
+	int sysctl_ip_early_demux;
 
 	int sysctl_fwmark_reflect;
 	int sysctl_tcp_fwmark_accept;
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 852002f..e3d7827 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -308,15 +308,12 @@ drop:
 	return true;
 }
 
-int sysctl_ip_early_demux __read_mostly = 1;
-EXPORT_SYMBOL(sysctl_ip_early_demux);
-
 static int ip_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
 	const struct iphdr *iph = ip_hdr(skb);
 	struct rtable *rt;
 
-	if (sysctl_ip_early_demux &&
+	if (net->ipv4.sysctl_ip_early_demux &&
 	    !skb_dst(skb) &&
 	    !skb->sk &&
 	    !ip_is_fragment(iph)) {
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 04ac5b7..1e1fe60 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -297,13 +297,6 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
-		.procname	= "ip_early_demux",
-		.data		= &sysctl_ip_early_demux,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec
-	},
-	{
 		.procname	= "tcp_fastopen",
 		.data		= &sysctl_tcp_fastopen,
 		.maxlen		= sizeof(int),
@@ -744,6 +737,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dointvec
 	},
 	{
+		.procname	= "ip_early_demux",
+		.data		= &init_net.ipv4.sysctl_ip_early_demux,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{
 		.procname	= "ip_default_ttl",
 		.data		= &init_net.ipv4.sysctl_ip_default_ttl,
 		.maxlen		= sizeof(int),
@@ -990,6 +990,7 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
 
 	net->ipv4.sysctl_ip_default_ttl = IPDEFTTL;
 	net->ipv4.sysctl_ip_dynaddr = 0;
+	net->ipv4.sysctl_ip_early_demux = 1;
 
 	return 0;
 
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 31ac3c5..c05c425 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -49,7 +49,7 @@
 
 int ip6_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
-	if (sysctl_ip_early_demux && !skb_dst(skb) && skb->sk == NULL) {
+	if (net->ipv4.sysctl_ip_early_demux && !skb_dst(skb) && skb->sk == NULL) {
 		const struct inet6_protocol *ipprot;
 
 		ipprot = rcu_dereference(inet6_protos[ipv6_hdr(skb)->nexthdr]);
-- 
cgit v0.10.2


From 0fbf4cb27e061204c8cee8e7eb2870416bdf30fd Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 15 Feb 2016 12:11:31 +0200
Subject: ipv4: namespacify ip fragment max dist sysctl knob

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 12aac0f..909972a 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -13,6 +13,7 @@ struct netns_frags {
 	int			timeout;
 	int			high_thresh;
 	int			low_thresh;
+	int			max_dist;
 };
 
 /**
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 187c6fc..9571614 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -54,8 +54,6 @@
  * code now. If you change something here, _PLEASE_ update ipv6/reassembly.c
  * as well. Or notify me, at least. --ANK
  */
-
-static int sysctl_ipfrag_max_dist __read_mostly = 64;
 static const char ip_frag_cache_name[] = "ip4-frags";
 
 struct ipfrag_skb_cb
@@ -150,7 +148,7 @@ static void ip4_frag_init(struct inet_frag_queue *q, const void *a)
 	qp->daddr = arg->iph->daddr;
 	qp->vif = arg->vif;
 	qp->user = arg->user;
-	qp->peer = sysctl_ipfrag_max_dist ?
+	qp->peer = q->net->max_dist ?
 		inet_getpeer_v4(net->ipv4.peers, arg->iph->saddr, arg->vif, 1) :
 		NULL;
 }
@@ -275,7 +273,7 @@ static struct ipq *ip_find(struct net *net, struct iphdr *iph,
 static int ip_frag_too_far(struct ipq *qp)
 {
 	struct inet_peer *peer = qp->peer;
-	unsigned int max = sysctl_ipfrag_max_dist;
+	unsigned int max = qp->q.net->max_dist;
 	unsigned int start, end;
 
 	int rc;
@@ -749,6 +747,14 @@ static struct ctl_table ip4_frags_ns_ctl_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
+	{
+		.procname	= "ipfrag_max_dist",
+		.data		= &init_net.ipv4.frags.max_dist,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &zero
+	},
 	{ }
 };
 
@@ -762,14 +768,6 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
-	{
-		.procname	= "ipfrag_max_dist",
-		.data		= &sysctl_ipfrag_max_dist,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &zero
-	},
 	{ }
 };
 
@@ -790,6 +788,7 @@ static int __net_init ip4_frags_ns_ctl_register(struct net *net)
 		table[1].data = &net->ipv4.frags.low_thresh;
 		table[1].extra2 = &net->ipv4.frags.high_thresh;
 		table[2].data = &net->ipv4.frags.timeout;
+		table[3].data = &net->ipv4.frags.max_dist;
 
 		/* Don't export sysctls to unprivileged users */
 		if (net->user_ns != &init_user_ns)
@@ -865,6 +864,8 @@ static int __net_init ipv4_frags_init_net(struct net *net)
 	 */
 	net->ipv4.frags.timeout = IP_FRAG_TIME;
 
+	net->ipv4.frags.max_dist = 64;
+
 	res = inet_frags_init_net(&net->ipv4.frags);
 	if (res)
 		return res;
-- 
cgit v0.10.2


From 52a773d645e95515016b9d896ba8e323e0aaa57b Mon Sep 17 00:00:00 2001
From: Nikolay Borisov <kernel@kyup.com>
Date: Mon, 15 Feb 2016 12:11:32 +0200
Subject: net: Export ip fragment sysctl to unprivileged users

Now that all the ip fragmentation related sysctls are namespaceified
there is no reason to hide them anymore from "root" users inside
containers.

Signed-off-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 9571614..efbd47d 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -789,10 +789,6 @@ static int __net_init ip4_frags_ns_ctl_register(struct net *net)
 		table[1].extra2 = &net->ipv4.frags.high_thresh;
 		table[2].data = &net->ipv4.frags.timeout;
 		table[3].data = &net->ipv4.frags.max_dist;
-
-		/* Don't export sysctls to unprivileged users */
-		if (net->user_ns != &init_user_ns)
-			table[0].procname = NULL;
 	}
 
 	hdr = register_net_sysctl(net, "net/ipv4", table);
-- 
cgit v0.10.2


From 6774faf96437e8192e4ee2d16f1399ec9f842b80 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:44 -0800
Subject: i40e: AQ thermal sensor control struct

Add the new AQ command and struct for managing a thermal sensor.

Change-ID: I6f5631839a0f3dca352a6c222f1269a960e2310a
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 5effb91..9684572 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -230,6 +230,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_nvm_config_read		= 0x0704,
 	i40e_aqc_opc_nvm_config_write		= 0x0705,
 	i40e_aqc_opc_oem_post_update		= 0x0720,
+	i40e_aqc_opc_thermal_sensor		= 0x0721,
 
 	/* virtualization commands */
 	i40e_aqc_opc_send_msg_to_pf		= 0x0801,
@@ -1952,6 +1953,22 @@ struct i40e_aqc_nvm_oem_post_update_buffer {
 
 I40E_CHECK_STRUCT_LEN(0x28, i40e_aqc_nvm_oem_post_update_buffer);
 
+/* Thermal Sensor (indirect 0x0721)
+ *     read or set thermal sensor configs and values
+ *     takes a sensor and command specific data buffer, not detailed here
+ */
+struct i40e_aqc_thermal_sensor {
+	u8 sensor_action;
+#define I40E_AQ_THERMAL_SENSOR_READ_CONFIG	0
+#define I40E_AQ_THERMAL_SENSOR_SET_CONFIG	1
+#define I40E_AQ_THERMAL_SENSOR_READ_TEMP	2
+	u8 reserved[7];
+	__le32	addr_high;
+	__le32	addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_thermal_sensor);
+
 /* Send to PF command (indirect 0x0801) id is only used by PF
  * Send to VF command (indirect 0x0802) id is only used by PF
  * Send to Peer PF command (indirect 0x0803)
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index bdb18f6..6468537 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -230,6 +230,7 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_nvm_config_read		= 0x0704,
 	i40e_aqc_opc_nvm_config_write		= 0x0705,
 	i40e_aqc_opc_oem_post_update		= 0x0720,
+	i40e_aqc_opc_thermal_sensor		= 0x0721,
 
 	/* virtualization commands */
 	i40e_aqc_opc_send_msg_to_pf		= 0x0801,
@@ -1949,6 +1950,22 @@ struct i40e_aqc_nvm_oem_post_update_buffer {
 
 I40E_CHECK_STRUCT_LEN(0x28, i40e_aqc_nvm_oem_post_update_buffer);
 
+/* Thermal Sensor (indirect 0x0721)
+ *     read or set thermal sensor configs and values
+ *     takes a sensor and command specific data buffer, not detailed here
+ */
+struct i40e_aqc_thermal_sensor {
+	u8 sensor_action;
+#define I40E_AQ_THERMAL_SENSOR_READ_CONFIG	0
+#define I40E_AQ_THERMAL_SENSOR_SET_CONFIG	1
+#define I40E_AQ_THERMAL_SENSOR_READ_TEMP	2
+	u8 reserved[7];
+	__le32	addr_high;
+	__le32	addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_thermal_sensor);
+
 /* Send to PF command (indirect 0x0801) id is only used by PF
  * Send to VF command (indirect 0x0802) id is only used by PF
  * Send to Peer PF command (indirect 0x0803)
-- 
cgit v0.10.2


From 97b884fecd6186ac1bcc932c453dce305d81040b Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:45 -0800
Subject: i40e: Bump AQ minor version to 1.5 for new FW features

Bump AQ minor version to 1.5 for new FW features.

Change-ID: I5a790f7f519a2a8921aaa1c5663727dd1897ffec
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 9684572..5699299 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -34,7 +34,7 @@
  */
 
 #define I40E_FW_API_VERSION_MAJOR	0x0001
-#define I40E_FW_API_VERSION_MINOR	0x0004
+#define I40E_FW_API_VERSION_MINOR	0x0005
 
 struct i40e_aq_desc {
 	__le16 flags;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 6468537..ed963bc 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -34,7 +34,7 @@
  */
 
 #define I40E_FW_API_VERSION_MAJOR	0x0001
-#define I40E_FW_API_VERSION_MINOR	0x0004
+#define I40E_FW_API_VERSION_MINOR	0x0005
 
 struct i40e_aq_desc {
 	__le16 flags;
-- 
cgit v0.10.2


From 071c859b878a05c3f603669bfda2e5b4332dd695 Mon Sep 17 00:00:00 2001
From: Pandi Maharajan <pandi.maharajan@intel.com>
Date: Wed, 23 Dec 2015 12:05:46 -0800
Subject: i40e: Store lan_vsi_idx and lan_vsi_id in the right size

lan_vsi_idx and lan_vsi_id are assigned to u16 data sized variables but
declared in u8. This patch fixes the width of the datatype.

Change-ID: If4bcbcc7d32f2b287c51cb33d17879691258dce2
Signed-off-by: Pandi Kumar Maharajan <pandi.maharajan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
index da44995..e74642a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
@@ -91,8 +91,8 @@ struct i40e_vf {
 	 * When assigned, these will be non-zero, because VSI 0 is always
 	 * the main LAN VSI for the PF.
 	 */
-	u8 lan_vsi_idx;	        /* index into PF struct */
-	u8 lan_vsi_id;		/* ID as used by firmware */
+	u16 lan_vsi_idx;	/* index into PF struct */
+	u16 lan_vsi_id;		/* ID as used by firmware */
 
 	u8 num_queue_pairs;	/* num of qps assigned to VF vsis */
 	u64 num_mdd_events;	/* num of mdd events detected */
-- 
cgit v0.10.2


From a3d772a3925d85721ad8518db14603fb1cd99295 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Wed, 23 Dec 2015 12:05:47 -0800
Subject: i40e: fix write-back-on-itr to work with legacy itr

We were not doing write-back on interrupt throttle for Legacy case in X722.
This patch fixes that, so we do WB_ON_ITR for Legacy as well. Plus the issue
that we should still be setting NO_ITR if we are touching the DYN_CTLN register
since we do not want to change ITR setting here.

Change-ID: I5db8491ee1544118a389db839cecc93e1bbc480e
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 47bd8b3..6234136 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -789,12 +789,20 @@ void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
 		if (q_vector->arm_wb_state)
 			return;
 
-		val = I40E_PFINT_DYN_CTLN_WB_ON_ITR_MASK;
+		if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
+			val = I40E_PFINT_DYN_CTLN_WB_ON_ITR_MASK |
+			      I40E_PFINT_DYN_CTLN_ITR_INDX_MASK; /* set noitr */
+
+			wr32(&vsi->back->hw,
+			     I40E_PFINT_DYN_CTLN(q_vector->v_idx +
+						 vsi->base_vector - 1),
+			     val);
+		} else {
+			val = I40E_PFINT_DYN_CTL0_WB_ON_ITR_MASK |
+			      I40E_PFINT_DYN_CTL0_ITR_INDX_MASK; /* set noitr */
 
-		wr32(&vsi->back->hw,
-		     I40E_PFINT_DYN_CTLN(q_vector->v_idx +
-					 vsi->base_vector - 1),
-		     val);
+			wr32(&vsi->back->hw, I40E_PFINT_DYN_CTL0, val);
+		}
 		q_vector->arm_wb_state = true;
 	} else if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
 		u32 val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 7d663fb..3e69235 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -307,7 +307,8 @@ static void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector
 		if (q_vector->arm_wb_state)
 			return;
 
-		val = I40E_VFINT_DYN_CTLN1_WB_ON_ITR_MASK;
+		val = I40E_VFINT_DYN_CTLN1_WB_ON_ITR_MASK |
+		      I40E_VFINT_DYN_CTLN1_ITR_INDX_MASK; /* set noitr */
 
 		wr32(&vsi->back->hw,
 		     I40E_VFINT_DYN_CTLN1(q_vector->v_idx +
-- 
cgit v0.10.2


From 1d0a4ada8401c73b360fc8492064a1107fca019b Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 23 Dec 2015 12:05:48 -0800
Subject: i40e: add counter for arq overflows

Sometimes, ARQ overflows are a big deal and tell us that the
firmware/hardware/driver/something is having problems. But normally
they're no big deal. To assist in assessing this, add a counter to
our Ethtool stats. A handful of ARQ overflows during VF init is no
problem. A large, ever-growing number indicates that Something Bad is
happening.

Change-ID: Ie5348bfbc8a54a890559cb00279c28d976a55096
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 53ed3bd..944dee9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -393,6 +393,7 @@ struct i40e_pf {
 	struct i40e_vf *vf;
 	int num_alloc_vfs;	/* actual number of VFs allocated */
 	u32 vf_aq_requests;
+	u32 arq_overflows;	/* Not fatal, possibly indicative of problems */
 
 	/* DCBx/DCBNL capability for PF that indicates
 	 * whether DCBx is managed by firmware or host
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 4549591..812c954 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -143,6 +143,7 @@ static struct i40e_stats i40e_gstrings_stats[] = {
 	I40E_PF_STAT("rx_oversize", stats.rx_oversize),
 	I40E_PF_STAT("rx_jabber", stats.rx_jabber),
 	I40E_PF_STAT("VF_admin_queue_requests", vf_aq_requests),
+	I40E_PF_STAT("arq_overflows", arq_overflows),
 	I40E_PF_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared),
 	I40E_PF_STAT("fdir_flush_cnt", fd_flush_cnt),
 	I40E_PF_STAT("fdir_atr_match", stats.fd_atr_match),
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 320b049..7323e32 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -6248,6 +6248,7 @@ static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
 		if (hw->debug_mask & I40E_DEBUG_AQ)
 			dev_info(&pf->pdev->dev, "ARQ Overflow Error detected\n");
 		val &= ~I40E_PF_ARQLEN_ARQOVFL_MASK;
+		pf->arq_overflows++;
 	}
 	if (val & I40E_PF_ARQLEN_ARQCRIT_MASK) {
 		if (hw->debug_mask & I40E_DEBUG_AQ)
-- 
cgit v0.10.2


From 07f169c3e9df898afe951d74734a0e54ac89ee9d Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 23 Dec 2015 12:05:49 -0800
Subject: i40e: add 20G speed for Tx bandwidth calculations

When calculating TX bandwidth for VFs, we need to know the link speed to
make sure we don't allocate more bandwidth than is available. Add 20G
link speed to the switch statement so we can support devices that link
at that speed.

Change-ID: I5409f6139d549e5832777db9c22ca0664e0c5f8b
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 659d782..1635c7a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -2306,6 +2306,9 @@ int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate,
 	case I40E_LINK_SPEED_40GB:
 		speed = 40000;
 		break;
+	case I40E_LINK_SPEED_20GB:
+		speed = 20000;
+		break;
 	case I40E_LINK_SPEED_10GB:
 		speed = 10000;
 		break;
-- 
cgit v0.10.2


From dd54a1ada9cea9275c53c89ad869214be6e9731a Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 23 Dec 2015 12:05:50 -0800
Subject: i40e: refactor DCB function

This is a simple refactor suggested by the community
to change a multi-level if statement into a switch.

Change-ID: I831cf3c40426022220aa9b43990022d22dfd50db
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb.c b/drivers/net/ethernet/intel/i40e/i40e_dcb.c
index 582daa7..0fab3a9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb.c
@@ -380,17 +380,20 @@ static void i40e_parse_cee_app_tlv(struct i40e_cee_feat_tlv *tlv,
 {
 	u16 length, typelength, offset = 0;
 	struct i40e_cee_app_prio *app;
-	u8 i, up, selector;
+	u8 i;
 
 	typelength = ntohs(tlv->hdr.typelen);
 	length = (u16)((typelength & I40E_LLDP_TLV_LEN_MASK) >>
 		       I40E_LLDP_TLV_LEN_SHIFT);
 
 	dcbcfg->numapps = length / sizeof(*app);
+
 	if (!dcbcfg->numapps)
 		return;
 
 	for (i = 0; i < dcbcfg->numapps; i++) {
+		u8 up, selector;
+
 		app = (struct i40e_cee_app_prio *)(tlv->tlvinfo + offset);
 		for (up = 0; up < I40E_MAX_USER_PRIORITY; up++) {
 			if (app->prio_map & BIT(up))
@@ -400,13 +403,17 @@ static void i40e_parse_cee_app_tlv(struct i40e_cee_feat_tlv *tlv,
 
 		/* Get Selector from lower 2 bits, and convert to IEEE */
 		selector = (app->upper_oui_sel & I40E_CEE_APP_SELECTOR_MASK);
-		if (selector == I40E_CEE_APP_SEL_ETHTYPE)
+		switch (selector) {
+		case I40E_CEE_APP_SEL_ETHTYPE:
 			dcbcfg->app[i].selector = I40E_APP_SEL_ETHTYPE;
-		else if (selector == I40E_CEE_APP_SEL_TCPIP)
+			break;
+		case I40E_CEE_APP_SEL_TCPIP:
 			dcbcfg->app[i].selector = I40E_APP_SEL_TCPIP;
-		else
+			break;
+		default:
 			/* Keep selector as it is for unknown types */
 			dcbcfg->app[i].selector = selector;
+		}
 
 		dcbcfg->app[i].protocolid = ntohs(app->protocol);
 		/* Move to next app */
-- 
cgit v0.10.2


From 1d73b2db4b9f4c12d42451efd09887ff7b37b02d Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:51 -0800
Subject: i40e: add a little more to an NVM update debug message

Add a little more detail to an NVM update debug message in order to
see the full ethtool request data.

Change-ID: Iab10437cb32d6fddc67ee347e7c0b42511e152cd
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_nvm.c b/drivers/net/ethernet/intel/i40e/i40e_nvm.c
index 6100cdd..5730f80 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_nvm.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_nvm.c
@@ -693,10 +693,11 @@ i40e_status i40e_nvmupd_command(struct i40e_hw *hw,
 	/* early check for status command and debug msgs */
 	upd_cmd = i40e_nvmupd_validate_command(hw, cmd, perrno);
 
-	i40e_debug(hw, I40E_DEBUG_NVM, "%s state %d nvm_release_on_hold %d\n",
+	i40e_debug(hw, I40E_DEBUG_NVM, "%s state %d nvm_release_on_hold %d cmd 0x%08x config 0x%08x offset 0x%08x data_size 0x%08x\n",
 		   i40e_nvm_update_state_str[upd_cmd],
 		   hw->nvmupd_state,
-		   hw->aq.nvm_release_on_done);
+		   hw->aq.nvm_release_on_done,
+		   cmd->command, cmd->config, cmd->offset, cmd->data_size);
 
 	if (upd_cmd == I40E_NVMUPD_INVALID) {
 		*perrno = -EFAULT;
-- 
cgit v0.10.2


From 509a447ae80c2e48b991fa7556a1b0d2a7ee8a51 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 23 Dec 2015 12:05:52 -0800
Subject: i40evf: enable bus master after reset

If the VF is reset via VFLR, the device will be knocked out of bus
master mode, and the driver will fail to recover from the reset. Fix
this by enabling bus mastering after every reset. In a non-VFLR case,
the bus master bit will not be disabled, and this call will have no effect.

Change-ID: Id515859ac7a691db478222228add6d149e96801a
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 66964eb..933f1fc 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1837,6 +1837,7 @@ static void i40evf_reset_task(struct work_struct *work)
 			break;
 		msleep(I40EVF_RESET_WAIT_MS);
 	}
+	pci_set_master(adapter->pdev);
 	/* extra wait to make sure minimum wait is met */
 	msleep(I40EVF_RESET_WAIT_MS);
 	if (i == I40EVF_RESET_WAIT_COUNT) {
-- 
cgit v0.10.2


From de1017f76a9ba9c354e238ad94c88725a3688dce Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 23 Dec 2015 12:05:53 -0800
Subject: i40e: add netdev info to VSI dump

Add a few more bits of netdev data into the debugfs output for dump VSI.
For now, we'll add the features, hw_features, vlan_features, and flags
bitflags and the state. More could be added later if needed.

Also, tweak a couple nearby output lines for output readability.

Change-ID: I9fb5a9da75c9ad7679498ce9ac3ba24d065ddd2e
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Reviewed-by: Brandeburg, Jesse <jesse.brandeburg@intel.com>
Reviewed-by: Wyborny, Carolyn <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 10744a6..7aae056 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -379,19 +379,27 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 		return;
 	}
 	dev_info(&pf->pdev->dev, "vsi seid %d\n", seid);
-	if (vsi->netdev)
-		dev_info(&pf->pdev->dev,
-			 "    netdev: name = %s\n",
-			 vsi->netdev->name);
+	if (vsi->netdev) {
+		struct net_device *nd = vsi->netdev;
+
+		dev_info(&pf->pdev->dev, "    netdev: name = %s, state = %lu, flags = 0x%08x\n",
+			 nd->name, nd->state, nd->flags);
+		dev_info(&pf->pdev->dev, "        features      = 0x%08lx\n",
+			 (unsigned long int)nd->features);
+		dev_info(&pf->pdev->dev, "        hw_features   = 0x%08lx\n",
+			 (unsigned long int)nd->hw_features);
+		dev_info(&pf->pdev->dev, "        vlan_features = 0x%08lx\n",
+			 (unsigned long int)nd->vlan_features);
+	}
 	if (vsi->active_vlans)
 		dev_info(&pf->pdev->dev,
 			 "    vlgrp: & = %p\n", vsi->active_vlans);
 	dev_info(&pf->pdev->dev,
-		 "    netdev_registered = %i, current_netdev_flags = 0x%04x, state = %li flags = 0x%08lx\n",
-		 vsi->netdev_registered,
-		 vsi->current_netdev_flags, vsi->state, vsi->flags);
+		 "    state = %li flags = 0x%08lx, netdev_registered = %i, current_netdev_flags = 0x%04x\n",
+		 vsi->state, vsi->flags,
+		 vsi->netdev_registered, vsi->current_netdev_flags);
 	if (vsi == pf->vsi[pf->lan_vsi])
-		dev_info(&pf->pdev->dev, "MAC address: %pM SAN MAC: %pM Port MAC: %pM\n",
+		dev_info(&pf->pdev->dev, "    MAC address: %pM SAN MAC: %pM Port MAC: %pM\n",
 			 pf->hw.mac.addr,
 			 pf->hw.mac.san_addr,
 			 pf->hw.mac.port_addr);
-- 
cgit v0.10.2


From 2eccf1d611c09edb80cf5805041a34805a3e3b43 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 23 Dec 2015 12:05:54 -0800
Subject: i40e: remove VF device IDs from PF

The PF doesn't need to know about the VF's device IDs, so remove them.

Change-ID: I62cf0e0fffa1ace586e58e00bc271b10ae440f05
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 3b03a31..9018f28 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -62,14 +62,6 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_10G_BASE_T_X722:
 			hw->mac.type = I40E_MAC_X722;
 			break;
-		case I40E_DEV_ID_X722_VF:
-		case I40E_DEV_ID_X722_VF_HV:
-			hw->mac.type = I40E_MAC_X722_VF;
-			break;
-		case I40E_DEV_ID_VF:
-		case I40E_DEV_ID_VF_HV:
-			hw->mac.type = I40E_MAC_VF;
-			break;
 		default:
 			hw->mac.type = I40E_MAC_GENERIC;
 			break;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h
index f7ce5c7..99257fc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_devids.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h
@@ -39,15 +39,11 @@
 #define I40E_DEV_ID_20G_KR2		0x1587
 #define I40E_DEV_ID_20G_KR2_A		0x1588
 #define I40E_DEV_ID_10G_BASE_T4		0x1589
-#define I40E_DEV_ID_VF			0x154C
-#define I40E_DEV_ID_VF_HV		0x1571
 #define I40E_DEV_ID_KX_X722		0x37CE
 #define I40E_DEV_ID_QSFP_X722		0x37CF
 #define I40E_DEV_ID_SFP_X722		0x37D0
 #define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
 #define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
-#define I40E_DEV_ID_X722_VF		0x37CD
-#define I40E_DEV_ID_X722_VF_HV		0x37D9
 
 #define i40e_is_40G_device(d)		((d) == I40E_DEV_ID_QSFP_A  || \
 					 (d) == I40E_DEV_ID_QSFP_B  || \
-- 
cgit v0.10.2


From d17038d6872b02df984d3be453b732c833d282df Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 23 Dec 2015 12:05:55 -0800
Subject: i40e: trivial: remove unnecessary local var

Probe routine already has too many locals, just convert one
used for kzalloc into a kcalloc, eliminating the local.

Change-ID: I349049872b71f858cbeb91ad7836e6767fc7b7d1
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: Anjali Singhai <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 7323e32..a2643fe 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -10637,7 +10637,6 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	u16 wol_nvm_bits;
 	u16 link_status;
 	int err;
-	u32 len;
 	u32 val;
 	u32 i;
 	u8 set_fc_aq_fail;
@@ -10896,8 +10895,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		pf->num_alloc_vsi = pf->hw.func_caps.num_vsis;
 
 	/* Set up the *vsi struct and our local tracking of the MAIN PF vsi. */
-	len = sizeof(struct i40e_vsi *) * pf->num_alloc_vsi;
-	pf->vsi = kzalloc(len, GFP_KERNEL);
+	pf->vsi = kcalloc(pf->num_alloc_vsi, sizeof(struct i40e_vsi *),
+			  GFP_KERNEL);
 	if (!pf->vsi) {
 		err = -ENOMEM;
 		goto err_switch_setup;
-- 
cgit v0.10.2


From 82f399c9350dd2531acaf2fb25a701f58a49f936 Mon Sep 17 00:00:00 2001
From: Catherine Sullivan <catherine.sullivan@intel.com>
Date: Wed, 23 Dec 2015 12:05:56 -0800
Subject: i40e/i40evf: Bump i40e to 1.4.11 and i40evf to 1.4.7

Bump.

Change-ID: I21aa520a3c8c5f4f562a98019bf8b76b3706c480
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index a2643fe..9978c39 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -51,7 +51,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 10
+#define DRV_VERSION_BUILD 11
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 933f1fc..ae6c6e1 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -38,7 +38,7 @@ static const char i40evf_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 4
+#define DRV_VERSION_BUILD 7
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD) \
-- 
cgit v0.10.2


From e4c6734eaab90695db0ea8456307790cb0c1ccb5 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:16:15 -0800
Subject: net: rework ndo tc op to consume additional qdisc handle parameter

The ndo_setup_tc() op was added to support drivers offloading tx
qdiscs however only support for mqprio was ever added. So we
only ever added support for passing the number of traffic classes
to the driver.

This patch generalizes the ndo_setup_tc op so that a handle can
be provided to indicate if the offload is for ingress or egress
or potentially even child qdiscs.

CC: Murali Karicheri <m-karicheri2@ti.com>
CC: Shradha Shah <sshah@solarflare.com>
CC: Or Gerlitz <ogerlitz@mellanox.com>
CC: Ariel Elior <ariel.elior@qlogic.com>
CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
CC: Bruce Allan <bruce.w.allan@intel.com>
CC: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 8a9b493..9955cae 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -1626,12 +1626,15 @@ static void xgbe_poll_controller(struct net_device *netdev)
 }
 #endif /* End CONFIG_NET_POLL_CONTROLLER */
 
-static int xgbe_setup_tc(struct net_device *netdev, u8 tc)
+static int xgbe_setup_tc(struct net_device *netdev, u32 handle, u8 tc)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	unsigned int offset, queue;
 	u8 i;
 
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
 	if (tc && (tc != pdata->hw_feat.tc_cnt))
 		return -EINVAL;
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 9e42bca..b262cba 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -4272,6 +4272,13 @@ int bnx2x_setup_tc(struct net_device *dev, u8 num_tc)
 	return 0;
 }
 
+int __bnx2x_setup_tc(struct net_device *dev, u32 handle, u8 num_tc)
+{
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+	return bnx2x_setup_tc(dev, num_tc);
+}
+
 /* called with rtnl_lock */
 int bnx2x_change_mac_addr(struct net_device *dev, void *p)
 {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 3736986..60a4109 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -486,6 +486,7 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev);
 
 /* setup_tc callback */
 int bnx2x_setup_tc(struct net_device *dev, u8 num_tc);
+int __bnx2x_setup_tc(struct net_device *dev, u32 handle, u8 num_tc);
 
 int bnx2x_get_vf_config(struct net_device *dev, int vf,
 			struct ifla_vf_info *ivi);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index c584525..81fc51c 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -13061,7 +13061,7 @@ static const struct net_device_ops bnx2x_netdev_ops = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= poll_bnx2x,
 #endif
-	.ndo_setup_tc		= bnx2x_setup_tc,
+	.ndo_setup_tc		= __bnx2x_setup_tc,
 #ifdef CONFIG_BNX2X_SRIOV
 	.ndo_set_vf_mac		= bnx2x_set_vf_mac,
 	.ndo_set_vf_vlan	= bnx2x_set_vf_vlan,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 5dc89e5..ff08faf 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -5370,10 +5370,13 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
-static int bnxt_setup_tc(struct net_device *dev, u8 tc)
+static int bnxt_setup_tc(struct net_device *dev, u32 handle, u8 tc)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
 	if (tc > bp->max_tc) {
 		netdev_err(dev, "too many traffic classes requested: %d Max supported is %d\n",
 			   tc, bp->max_tc);
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index 662569d..12701a4 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -1204,6 +1204,14 @@ err_queueing_scheme:
 	return err;
 }
 
+static int __fm10k_setup_tc(struct net_device *dev, u32 handle, u8 tc)
+{
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
+	return fm10k_setup_tc(dev, tc);
+}
+
 static int fm10k_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 {
 	switch (cmd) {
@@ -1386,7 +1394,7 @@ static const struct net_device_ops fm10k_netdev_ops = {
 	.ndo_vlan_rx_kill_vid	= fm10k_vlan_rx_kill_vid,
 	.ndo_set_rx_mode	= fm10k_set_rx_mode,
 	.ndo_get_stats64	= fm10k_get_stats64,
-	.ndo_setup_tc		= fm10k_setup_tc,
+	.ndo_setup_tc		= __fm10k_setup_tc,
 	.ndo_set_vf_mac		= fm10k_ndo_set_vf_mac,
 	.ndo_set_vf_vlan	= fm10k_ndo_set_vf_vlan,
 	.ndo_set_vf_rate	= fm10k_ndo_set_vf_bw,
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 53ed3bd..ef9ca07 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -788,7 +788,7 @@ struct i40e_mac_filter *i40e_find_mac(struct i40e_vsi *vsi, u8 *macaddr,
 				      bool is_vf, bool is_netdev);
 #ifdef I40E_FCOE
 int i40e_close(struct net_device *netdev);
-int i40e_setup_tc(struct net_device *netdev, u8 tc);
+int __i40e_setup_tc(struct net_device *netdev, u32 handle, u8 tc);
 void i40e_netpoll(struct net_device *netdev);
 int i40e_fcoe_enable(struct net_device *netdev);
 int i40e_fcoe_disable(struct net_device *netdev);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
index 579a46c..7c66ce41 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
@@ -1457,7 +1457,7 @@ static const struct net_device_ops i40e_fcoe_netdev_ops = {
 	.ndo_tx_timeout		= i40e_tx_timeout,
 	.ndo_vlan_rx_add_vid	= i40e_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid	= i40e_vlan_rx_kill_vid,
-	.ndo_setup_tc		= i40e_setup_tc,
+	.ndo_setup_tc		= __i40e_setup_tc,
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= i40e_netpoll,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 320b049..abcb6c1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5253,11 +5253,7 @@ void i40e_down(struct i40e_vsi *vsi)
  * @netdev: net device to configure
  * @tc: number of traffic classes to enable
  **/
-#ifdef I40E_FCOE
-int i40e_setup_tc(struct net_device *netdev, u8 tc)
-#else
 static int i40e_setup_tc(struct net_device *netdev, u8 tc)
-#endif
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
@@ -5310,6 +5306,17 @@ exit:
 	return ret;
 }
 
+#ifdef I40E_FCOE
+int __i40e_setup_tc(struct net_device *netdev, u32 handle, u8 tc)
+#else
+static int __i40e_setup_tc(struct net_device *netdev, u32 handle, u8 tc)
+#endif
+{
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+	return i40e_setup_tc(netdev, tc);
+}
+
 /**
  * i40e_open - Called when a network interface is made active
  * @netdev: network interface device structure
@@ -8951,7 +8958,7 @@ static const struct net_device_ops i40e_netdev_ops = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= i40e_netpoll,
 #endif
-	.ndo_setup_tc		= i40e_setup_tc,
+	.ndo_setup_tc		= __i40e_setup_tc,
 #ifdef I40E_FCOE
 	.ndo_fcoe_enable	= i40e_fcoe_enable,
 	.ndo_fcoe_disable	= i40e_fcoe_disable,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 0c701b8..1ba714e 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8200,6 +8200,15 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 	return 0;
 }
 
+int __ixgbe_setup_tc(struct net_device *dev, u32 handle, u8 tc)
+{
+	/* Only support egress tc setup for now */
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
+	return ixgbe_setup_tc(dev, tc);
+}
+
 #ifdef CONFIG_PCI_IOV
 void ixgbe_sriov_reinit(struct ixgbe_adapter *adapter)
 {
@@ -8658,7 +8667,7 @@ static const struct net_device_ops ixgbe_netdev_ops = {
 	.ndo_get_vf_config	= ixgbe_ndo_get_vf_config,
 	.ndo_get_stats64	= ixgbe_get_stats64,
 #ifdef CONFIG_IXGBE_DCB
-	.ndo_setup_tc		= ixgbe_setup_tc,
+	.ndo_setup_tc		= __ixgbe_setup_tc,
 #endif
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= ixgbe_netpoll,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 0c7e3f6..d5c6c16 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -69,6 +69,14 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up)
 	return 0;
 }
 
+static int __mlx4_en_setup_tc(struct net_device *dev, u32 handle, u8 up)
+{
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
+	return mlx4_en_setup_tc(dev, up);
+}
+
 #ifdef CONFIG_RFS_ACCEL
 
 struct mlx4_en_filter {
@@ -2466,7 +2474,7 @@ static const struct net_device_ops mlx4_netdev_ops = {
 #endif
 	.ndo_set_features	= mlx4_en_set_features,
 	.ndo_fix_features	= mlx4_en_fix_features,
-	.ndo_setup_tc		= mlx4_en_setup_tc,
+	.ndo_setup_tc		= __mlx4_en_setup_tc,
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= mlx4_en_filter_rfs,
 #endif
@@ -2504,7 +2512,7 @@ static const struct net_device_ops mlx4_netdev_ops_master = {
 #endif
 	.ndo_set_features	= mlx4_en_set_features,
 	.ndo_fix_features	= mlx4_en_fix_features,
-	.ndo_setup_tc		= mlx4_en_setup_tc,
+	.ndo_setup_tc		= __mlx4_en_setup_tc,
 #ifdef CONFIG_RFS_ACCEL
 	.ndo_rx_flow_steer	= mlx4_en_filter_rfs,
 #endif
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 1082747..7815fa0 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -32,7 +32,7 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb,
 				struct net_device *net_dev);
 netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb);
 void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index);
-int efx_setup_tc(struct net_device *net_dev, u8 num_tc);
+int efx_setup_tc(struct net_device *net_dev, u32 handle, u8 num_tc);
 unsigned int efx_tx_max_skb_descs(struct efx_nic *efx);
 extern unsigned int efx_piobuf_size;
 extern bool efx_separate_tx_channels;
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index f7a0ec1..8f1d53e 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -562,7 +562,7 @@ void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue)
 				     efx->n_tx_channels : 0));
 }
 
-int efx_setup_tc(struct net_device *net_dev, u8 num_tc)
+int efx_setup_tc(struct net_device *net_dev, u32 handle, u8 num_tc)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
 	struct efx_channel *channel;
@@ -570,6 +570,9 @@ int efx_setup_tc(struct net_device *net_dev, u8 num_tc)
 	unsigned tc;
 	int rc;
 
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
 	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0 || num_tc > EFX_MAX_TX_TC)
 		return -EINVAL;
 
diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index c61d66d..40cde81 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -1835,13 +1835,16 @@ static u16 netcp_select_queue(struct net_device *dev, struct sk_buff *skb,
 	return 0;
 }
 
-static int netcp_setup_tc(struct net_device *dev, u8 num_tc)
+static int netcp_setup_tc(struct net_device *dev, u32 handle, u8 num_tc)
 {
 	int i;
 
 	/* setup tc must be called under rtnl lock */
 	ASSERT_RTNL();
 
+	if (handle != TC_H_ROOT)
+		return -EINVAL;
+
 	/* Sanity-check the number of traffic classes requested */
 	if ((dev->real_num_tx_queues <= 1) ||
 	    (dev->real_num_tx_queues < num_tc))
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 0499569..48928b6 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -51,6 +51,7 @@
 #include <linux/neighbour.h>
 #include <uapi/linux/netdevice.h>
 #include <uapi/linux/if_bonding.h>
+#include <uapi/linux/pkt_cls.h>
 
 struct netpoll_info;
 struct device;
@@ -1150,7 +1151,7 @@ struct net_device_ops {
 	int			(*ndo_set_vf_rss_query_en)(
 						   struct net_device *dev,
 						   int vf, bool setting);
-	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
+	int			(*ndo_setup_tc)(struct net_device *dev, u32 handle, u8 tc);
 #if IS_ENABLED(CONFIG_FCOE)
 	int			(*ndo_fcoe_enable)(struct net_device *dev);
 	int			(*ndo_fcoe_disable)(struct net_device *dev);
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index ad70ecf..f5a0e8a 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -39,7 +39,7 @@ static void mqprio_destroy(struct Qdisc *sch)
 	}
 
 	if (priv->hw_owned && dev->netdev_ops->ndo_setup_tc)
-		dev->netdev_ops->ndo_setup_tc(dev, 0);
+		dev->netdev_ops->ndo_setup_tc(dev, sch->handle, 0);
 	else
 		netdev_set_num_tc(dev, 0);
 }
@@ -141,7 +141,8 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt)
 	 */
 	if (qopt->hw) {
 		priv->hw_owned = 1;
-		err = dev->netdev_ops->ndo_setup_tc(dev, qopt->num_tc);
+		err = dev->netdev_ops->ndo_setup_tc(dev, sch->handle,
+						    qopt->num_tc);
 		if (err)
 			goto err;
 	} else {
-- 
cgit v0.10.2


From 16e5cc647173a97e33b3e3ba81f73eb455561794 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:16:43 -0800
Subject: net: rework setup_tc ndo op to consume general tc operand

This patch updates setup_tc so we can pass additional parameters into
the ndo op in a generic way. To do this we provide structured union
and type flag.

This lets each classifier and qdisc provide its own set of attributes
without having to add new ndo ops or grow the signature of the
callback.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 9955cae..cfd3f7e 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -1626,15 +1626,18 @@ static void xgbe_poll_controller(struct net_device *netdev)
 }
 #endif /* End CONFIG_NET_POLL_CONTROLLER */
 
-static int xgbe_setup_tc(struct net_device *netdev, u32 handle, u8 tc)
+static int xgbe_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
+			 struct tc_to_netdev *tc_to_netdev)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	unsigned int offset, queue;
-	u8 i;
+	u8 i, tc;
 
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc_to_netdev->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
+	tc = tc_to_netdev->tc;
+
 	if (tc && (tc != pdata->hw_feat.tc_cnt))
 		return -EINVAL;
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index b262cba..45843d1 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -4272,11 +4272,12 @@ int bnx2x_setup_tc(struct net_device *dev, u8 num_tc)
 	return 0;
 }
 
-int __bnx2x_setup_tc(struct net_device *dev, u32 handle, u8 num_tc)
+int __bnx2x_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+		     struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
-	return bnx2x_setup_tc(dev, num_tc);
+	return bnx2x_setup_tc(dev, tc->tc);
 }
 
 /* called with rtnl_lock */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 60a4109..0e68fad 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -486,7 +486,8 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev);
 
 /* setup_tc callback */
 int bnx2x_setup_tc(struct net_device *dev, u8 num_tc);
-int __bnx2x_setup_tc(struct net_device *dev, u32 handle, u8 num_tc);
+int __bnx2x_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+		     struct tc_to_netdev *tc);
 
 int bnx2x_get_vf_config(struct net_device *dev, int vf,
 			struct ifla_vf_info *ivi);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index ff08faf..169920a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -5370,13 +5370,17 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
-static int bnxt_setup_tc(struct net_device *dev, u32 handle, u8 tc)
+static int bnxt_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+			 struct tc_to_netdev *ntc)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	u8 tc;
 
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || ntc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
+	tc = ntc->tc;
+
 	if (tc > bp->max_tc) {
 		netdev_err(dev, "too many traffic classes requested: %d Max supported is %d\n",
 			   tc, bp->max_tc);
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index 12701a4..dc1a821 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -1204,12 +1204,13 @@ err_queueing_scheme:
 	return err;
 }
 
-static int __fm10k_setup_tc(struct net_device *dev, u32 handle, u8 tc)
+static int __fm10k_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+			    struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
-	return fm10k_setup_tc(dev, tc);
+	return fm10k_setup_tc(dev, tc->tc);
 }
 
 static int fm10k_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index ef9ca07..933c4b3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -788,7 +788,8 @@ struct i40e_mac_filter *i40e_find_mac(struct i40e_vsi *vsi, u8 *macaddr,
 				      bool is_vf, bool is_netdev);
 #ifdef I40E_FCOE
 int i40e_close(struct net_device *netdev);
-int __i40e_setup_tc(struct net_device *netdev, u32 handle, u8 tc);
+int __i40e_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
+		    struct tc_to_netdev *tc);
 void i40e_netpoll(struct net_device *netdev);
 int i40e_fcoe_enable(struct net_device *netdev);
 int i40e_fcoe_disable(struct net_device *netdev);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index abcb6c1..257d162 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5307,14 +5307,16 @@ exit:
 }
 
 #ifdef I40E_FCOE
-int __i40e_setup_tc(struct net_device *netdev, u32 handle, u8 tc)
+int __i40e_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
+		    struct tc_to_netdev *tc)
 #else
-static int __i40e_setup_tc(struct net_device *netdev, u32 handle, u8 tc)
+static int __i40e_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
+			   struct tc_to_netdev *tc)
 #endif
 {
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
-	return i40e_setup_tc(netdev, tc);
+	return i40e_setup_tc(netdev, tc->tc);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 1ba714e..dca2298 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8200,13 +8200,14 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 	return 0;
 }
 
-int __ixgbe_setup_tc(struct net_device *dev, u32 handle, u8 tc)
+int __ixgbe_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+		     struct tc_to_netdev *tc)
 {
 	/* Only support egress tc setup for now */
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
-	return ixgbe_setup_tc(dev, tc);
+	return ixgbe_setup_tc(dev, tc->tc);
 }
 
 #ifdef CONFIG_PCI_IOV
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index d5c6c16..01d6a96 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -69,12 +69,13 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up)
 	return 0;
 }
 
-static int __mlx4_en_setup_tc(struct net_device *dev, u32 handle, u8 up)
+static int __mlx4_en_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+			      struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
-	return mlx4_en_setup_tc(dev, up);
+	return mlx4_en_setup_tc(dev, tc->tc);
 }
 
 #ifdef CONFIG_RFS_ACCEL
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 7815fa0..5e3f93f 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -32,7 +32,8 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb,
 				struct net_device *net_dev);
 netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb);
 void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index);
-int efx_setup_tc(struct net_device *net_dev, u32 handle, u8 num_tc);
+int efx_setup_tc(struct net_device *net_dev, u32 handle, __be16 proto,
+		 struct tc_to_netdev *tc);
 unsigned int efx_tx_max_skb_descs(struct efx_nic *efx);
 extern unsigned int efx_piobuf_size;
 extern bool efx_separate_tx_channels;
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index 8f1d53e..2cdb571 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -562,17 +562,20 @@ void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue)
 				     efx->n_tx_channels : 0));
 }
 
-int efx_setup_tc(struct net_device *net_dev, u32 handle, u8 num_tc)
+int efx_setup_tc(struct net_device *net_dev, u32 handle, __be16 proto,
+		 struct tc_to_netdev *ntc)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
 	struct efx_channel *channel;
 	struct efx_tx_queue *tx_queue;
-	unsigned tc;
+	unsigned tc, num_tc;
 	int rc;
 
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || ntc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
+	num_tc = ntc->tc;
+
 	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0 || num_tc > EFX_MAX_TX_TC)
 		return -EINVAL;
 
diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index 40cde81..8586a20 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -1835,25 +1835,26 @@ static u16 netcp_select_queue(struct net_device *dev, struct sk_buff *skb,
 	return 0;
 }
 
-static int netcp_setup_tc(struct net_device *dev, u32 handle, u8 num_tc)
+static int netcp_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
+			  struct tc_to_netdev tc)
 {
 	int i;
 
 	/* setup tc must be called under rtnl lock */
 	ASSERT_RTNL();
 
-	if (handle != TC_H_ROOT)
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	/* Sanity-check the number of traffic classes requested */
 	if ((dev->real_num_tx_queues <= 1) ||
-	    (dev->real_num_tx_queues < num_tc))
+	    (dev->real_num_tx_queues < tc->tc))
 		return -EINVAL;
 
 	/* Configure traffic class to queue mappings */
-	if (num_tc) {
-		netdev_set_num_tc(dev, num_tc);
-		for (i = 0; i < num_tc; i++)
+	if (tc->tc) {
+		netdev_set_num_tc(dev, tc->tc);
+		for (i = 0; i < tc->tc; i++)
 			netdev_set_tc_queue(dev, i, 1, i);
 	} else {
 		netdev_reset_tc(dev);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 48928b6..e396060 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -779,6 +779,21 @@ static inline bool netdev_phys_item_id_same(struct netdev_phys_item_id *a,
 typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
 				       struct sk_buff *skb);
 
+/* This structure holds attributes of qdisc and classifiers
+ * that are being passed to the netdevice through the setup_tc op.
+ */
+enum {
+	TC_SETUP_MQPRIO,
+};
+
+struct tc_to_netdev {
+	unsigned int type;
+	union {
+		u8 tc;
+	};
+};
+
+
 /*
  * This structure defines the management hooks for network devices.
  * The following hooks can be defined; unless noted otherwise, they are
@@ -1151,7 +1166,10 @@ struct net_device_ops {
 	int			(*ndo_set_vf_rss_query_en)(
 						   struct net_device *dev,
 						   int vf, bool setting);
-	int			(*ndo_setup_tc)(struct net_device *dev, u32 handle, u8 tc);
+	int			(*ndo_setup_tc)(struct net_device *dev,
+						u32 handle,
+						__be16 protocol,
+						struct tc_to_netdev *tc);
 #if IS_ENABLED(CONFIG_FCOE)
 	int			(*ndo_fcoe_enable)(struct net_device *dev);
 	int			(*ndo_fcoe_disable)(struct net_device *dev);
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index f5a0e8a..f9947d1 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -28,6 +28,7 @@ static void mqprio_destroy(struct Qdisc *sch)
 {
 	struct net_device *dev = qdisc_dev(sch);
 	struct mqprio_sched *priv = qdisc_priv(sch);
+	struct tc_to_netdev tc = {.type = TC_SETUP_MQPRIO};
 	unsigned int ntx;
 
 	if (priv->qdiscs) {
@@ -39,7 +40,7 @@ static void mqprio_destroy(struct Qdisc *sch)
 	}
 
 	if (priv->hw_owned && dev->netdev_ops->ndo_setup_tc)
-		dev->netdev_ops->ndo_setup_tc(dev, sch->handle, 0);
+		dev->netdev_ops->ndo_setup_tc(dev, sch->handle, 0, &tc);
 	else
 		netdev_set_num_tc(dev, 0);
 }
@@ -140,9 +141,11 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt)
 	 * supplied and verified mapping
 	 */
 	if (qopt->hw) {
+		struct tc_to_netdev tc = {.type = TC_SETUP_MQPRIO,
+					  .tc = qopt->num_tc};
+
 		priv->hw_owned = 1;
-		err = dev->netdev_ops->ndo_setup_tc(dev, sch->handle,
-						    qopt->num_tc);
+		err = dev->netdev_ops->ndo_setup_tc(dev, sch->handle, 0, &tc);
 		if (err)
 			goto err;
 	} else {
-- 
cgit v0.10.2


From a1b7c5fd7fe98f51fbbc393ee1fc4c1cdb2f0119 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:17:09 -0800
Subject: net: sched: add cls_u32 offload hooks for netdevs

This patch allows netdev drivers to consume cls_u32 offloads via
the ndo_setup_tc ndo op.

This works aligns with how network drivers have been doing qdisc
offloads for mqprio.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e396060..47671ce0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -779,17 +779,21 @@ static inline bool netdev_phys_item_id_same(struct netdev_phys_item_id *a,
 typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
 				       struct sk_buff *skb);
 
-/* This structure holds attributes of qdisc and classifiers
+/* These structures hold the attributes of qdisc and classifiers
  * that are being passed to the netdevice through the setup_tc op.
  */
 enum {
 	TC_SETUP_MQPRIO,
+	TC_SETUP_CLSU32,
 };
 
+struct tc_cls_u32_offload;
+
 struct tc_to_netdev {
 	unsigned int type;
 	union {
 		u8 tc;
+		struct tc_cls_u32_offload *cls_u32;
 	};
 };
 
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index bc49967..59789ca 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -358,4 +358,38 @@ tcf_match_indev(struct sk_buff *skb, int ifindex)
 }
 #endif /* CONFIG_NET_CLS_IND */
 
+struct tc_cls_u32_knode {
+	struct tcf_exts *exts;
+	u8 fshift;
+	u32 handle;
+	u32 val;
+	u32 mask;
+	u32 link_handle;
+	struct tc_u32_sel *sel;
+};
+
+struct tc_cls_u32_hnode {
+	u32 handle;
+	u32 prio;
+	unsigned int divisor;
+};
+
+enum tc_clsu32_command {
+	TC_CLSU32_NEW_KNODE,
+	TC_CLSU32_REPLACE_KNODE,
+	TC_CLSU32_DELETE_KNODE,
+	TC_CLSU32_NEW_HNODE,
+	TC_CLSU32_REPLACE_HNODE,
+	TC_CLSU32_DELETE_HNODE,
+};
+
+struct tc_cls_u32_offload {
+	/* knode values */
+	enum tc_clsu32_command command;
+	union {
+		struct tc_cls_u32_knode knode;
+		struct tc_cls_u32_hnode hnode;
+	};
+};
+
 #endif
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 4fbb674..d54bc94 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -43,6 +43,7 @@
 #include <net/netlink.h>
 #include <net/act_api.h>
 #include <net/pkt_cls.h>
+#include <linux/netdevice.h>
 
 struct tc_u_knode {
 	struct tc_u_knode __rcu	*next;
@@ -424,6 +425,93 @@ static int u32_delete_key(struct tcf_proto *tp, struct tc_u_knode *key)
 	return 0;
 }
 
+static void u32_remove_hw_knode(struct tcf_proto *tp, u32 handle)
+{
+	struct net_device *dev = tp->q->dev_queue->dev;
+	struct tc_cls_u32_offload u32_offload = {0};
+	struct tc_to_netdev offload;
+
+	offload.type = TC_SETUP_CLSU32;
+	offload.cls_u32 = &u32_offload;
+
+	if (dev->netdev_ops->ndo_setup_tc) {
+		offload.cls_u32->command = TC_CLSU32_DELETE_KNODE;
+		offload.cls_u32->knode.handle = handle;
+		dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle,
+					      tp->protocol, &offload);
+	}
+}
+
+static void u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
+{
+	struct net_device *dev = tp->q->dev_queue->dev;
+	struct tc_cls_u32_offload u32_offload = {0};
+	struct tc_to_netdev offload;
+
+	offload.type = TC_SETUP_CLSU32;
+	offload.cls_u32 = &u32_offload;
+
+	if (dev->netdev_ops->ndo_setup_tc) {
+		offload.cls_u32->command = TC_CLSU32_NEW_HNODE;
+		offload.cls_u32->hnode.divisor = h->divisor;
+		offload.cls_u32->hnode.handle = h->handle;
+		offload.cls_u32->hnode.prio = h->prio;
+
+		dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle,
+					      tp->protocol, &offload);
+	}
+}
+
+static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
+{
+	struct net_device *dev = tp->q->dev_queue->dev;
+	struct tc_cls_u32_offload u32_offload = {0};
+	struct tc_to_netdev offload;
+
+	offload.type = TC_SETUP_CLSU32;
+	offload.cls_u32 = &u32_offload;
+
+	if (dev->netdev_ops->ndo_setup_tc) {
+		offload.cls_u32->command = TC_CLSU32_DELETE_HNODE;
+		offload.cls_u32->hnode.divisor = h->divisor;
+		offload.cls_u32->hnode.handle = h->handle;
+		offload.cls_u32->hnode.prio = h->prio;
+
+		dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle,
+					      tp->protocol, &offload);
+	}
+}
+
+static void u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n)
+{
+	struct net_device *dev = tp->q->dev_queue->dev;
+	struct tc_cls_u32_offload u32_offload = {0};
+	struct tc_to_netdev offload;
+
+	offload.type = TC_SETUP_CLSU32;
+	offload.cls_u32 = &u32_offload;
+
+	if (dev->netdev_ops->ndo_setup_tc) {
+		offload.cls_u32->command = TC_CLSU32_REPLACE_KNODE;
+		offload.cls_u32->knode.handle = n->handle;
+		offload.cls_u32->knode.fshift = n->fshift;
+#ifdef CONFIG_CLS_U32_MARK
+		offload.cls_u32->knode.val = n->val;
+		offload.cls_u32->knode.mask = n->mask;
+#else
+		offload.cls_u32->knode.val = 0;
+		offload.cls_u32->knode.mask = 0;
+#endif
+		offload.cls_u32->knode.sel = &n->sel;
+		offload.cls_u32->knode.exts = &n->exts;
+		if (n->ht_down)
+			offload.cls_u32->knode.link_handle = n->ht_down->handle;
+
+		dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle,
+					      tp->protocol, &offload);
+	}
+}
+
 static void u32_clear_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht)
 {
 	struct tc_u_knode *n;
@@ -434,6 +522,7 @@ static void u32_clear_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht)
 			RCU_INIT_POINTER(ht->ht[h],
 					 rtnl_dereference(n->next));
 			tcf_unbind_filter(tp, &n->res);
+			u32_remove_hw_knode(tp, n->handle);
 			call_rcu(&n->rcu, u32_delete_key_freepf_rcu);
 		}
 	}
@@ -454,6 +543,7 @@ static int u32_destroy_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht)
 	     phn;
 	     hn = &phn->next, phn = rtnl_dereference(*hn)) {
 		if (phn == ht) {
+			u32_clear_hw_hnode(tp, ht);
 			RCU_INIT_POINTER(*hn, ht->next);
 			kfree_rcu(ht, rcu);
 			return 0;
@@ -540,8 +630,10 @@ static int u32_delete(struct tcf_proto *tp, unsigned long arg)
 	if (ht == NULL)
 		return 0;
 
-	if (TC_U32_KEY(ht->handle))
+	if (TC_U32_KEY(ht->handle)) {
+		u32_remove_hw_knode(tp, ht->handle);
 		return u32_delete_key(tp, (struct tc_u_knode *)ht);
+	}
 
 	if (root_ht == ht)
 		return -EINVAL;
@@ -769,6 +861,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 		u32_replace_knode(tp, tp_c, new);
 		tcf_unbind_filter(tp, &n->res);
 		call_rcu(&n->rcu, u32_delete_key_rcu);
+		u32_replace_hw_knode(tp, new);
 		return 0;
 	}
 
@@ -795,6 +888,8 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 		RCU_INIT_POINTER(ht->next, tp_c->hlist);
 		rcu_assign_pointer(tp_c->hlist, ht);
 		*arg = (unsigned long)ht;
+
+		u32_replace_hw_hnode(tp, ht);
 		return 0;
 	}
 
@@ -877,7 +972,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 
 		RCU_INIT_POINTER(n->next, pins);
 		rcu_assign_pointer(*ins, n);
-
+		u32_replace_hw_knode(tp, n);
 		*arg = (unsigned long)n;
 		return 0;
 	}
-- 
cgit v0.10.2


From 1c78c64e9c6f43a490427d55cd2d213b7c6795c1 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:17:37 -0800
Subject: net: add tc offload feature flag

Its useful to turn off the qdisc offload feature at a per device
level. This gives us a big hammer to enable/disable offloading.
More fine grained control (i.e. per rule) may be supported later.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index d9654f0e..a734bf4 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -67,6 +67,8 @@ enum {
 	NETIF_F_HW_L2FW_DOFFLOAD_BIT,	/* Allow L2 Forwarding in Hardware */
 	NETIF_F_BUSY_POLL_BIT,		/* Busy poll */
 
+	NETIF_F_HW_TC_BIT,		/* Offload TC infrastructure */
+
 	/*
 	 * Add your fresh new feature above and remember to update
 	 * netdev_features_strings[] in net/core/ethtool.c and maybe
@@ -124,6 +126,7 @@ enum {
 #define NETIF_F_HW_VLAN_STAG_TX	__NETIF_F(HW_VLAN_STAG_TX)
 #define NETIF_F_HW_L2FW_DOFFLOAD	__NETIF_F(HW_L2FW_DOFFLOAD)
 #define NETIF_F_BUSY_POLL	__NETIF_F(BUSY_POLL)
+#define NETIF_F_HW_TC		__NETIF_F(HW_TC)
 
 #define for_each_netdev_feature(mask_addr, bit)	\
 	for_each_set_bit(bit, (unsigned long *)mask_addr, NETDEV_FEATURE_COUNT)
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 65f907a..c2d3118 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -98,6 +98,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN]
 	[NETIF_F_RXALL_BIT] =            "rx-all",
 	[NETIF_F_HW_L2FW_DOFFLOAD_BIT] = "l2-fwd-offload",
 	[NETIF_F_BUSY_POLL_BIT] =        "busy-poll",
+	[NETIF_F_HW_TC_BIT] =		 "hw-tc-offload",
 };
 
 static const char
-- 
cgit v0.10.2


From 3b01cf56daf96acf9b155d6201d94bc8b4de218e Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:18:03 -0800
Subject: net: tc: helper functions to query action types

This is a helper function drivers can use to learn if the
action type is a drop action.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h
index 592a6bc..04a3183 100644
--- a/include/net/tc_act/tc_gact.h
+++ b/include/net/tc_act/tc_gact.h
@@ -2,6 +2,7 @@
 #define __NET_TC_GACT_H
 
 #include <net/act_api.h>
+#include <linux/tc_act/tc_gact.h>
 
 struct tcf_gact {
 	struct tcf_common	common;
@@ -15,4 +16,19 @@ struct tcf_gact {
 #define to_gact(a) \
 	container_of(a->priv, struct tcf_gact, common)
 
+#ifdef CONFIG_NET_CLS_ACT
+static inline bool is_tcf_gact_shot(const struct tc_action *a)
+{
+	struct tcf_gact *gact;
+
+	if (a->ops && a->ops->type != TCA_ACT_GACT)
+		return false;
+
+	gact = a->priv;
+	if (gact->tcf_action == TC_ACT_SHOT)
+		return true;
+
+	return false;
+}
+#endif
 #endif /* __NET_TC_GACT_H */
-- 
cgit v0.10.2


From 9d35cf062e05be8b8b2b7dbc943cd95352cd90cb Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:18:28 -0800
Subject: net: ixgbe: add minimal parser details for ixgbe

This adds an ixgbe data structure that is used to determine what
headers:fields can be matched and in what order they are supported.

For hardware devices this can be a bit tricky because typically
only pre-programmed (firmware, ucode, rtl) parse graphs will be
supported and we don't yet have an interface to change these from
the OS. So its sort of a you get whatever your friendly vendor
provides affair at the moment.

In the future we can add the get routines and set routines to
update this data structure. One interesting thing to note here
is the data structure here identifies ethernet, ip, and tcp
fields without having to hardcode them as enumerations or use
other identifiers.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
new file mode 100644
index 0000000..43ebec4
--- /dev/null
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
@@ -0,0 +1,112 @@
+/*******************************************************************************
+ *
+ * Intel 10 Gigabit PCI Express Linux drive
+ * Copyright(c) 2013 - 2015 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ *
+ ******************************************************************************/
+
+#ifndef _IXGBE_MODEL_H_
+#define _IXGBE_MODEL_H_
+
+#include "ixgbe.h"
+#include "ixgbe_type.h"
+
+struct ixgbe_mat_field {
+	unsigned int off;
+	unsigned int mask;
+	int (*val)(struct ixgbe_fdir_filter *input,
+		   union ixgbe_atr_input *mask,
+		   __u32 val, __u32 m);
+	unsigned int type;
+};
+
+static inline int ixgbe_mat_prgm_sip(struct ixgbe_fdir_filter *input,
+				     union ixgbe_atr_input *mask,
+				     __u32 val, __u32 m)
+{
+	input->filter.formatted.src_ip[0] = val;
+	mask->formatted.src_ip[0] = m;
+	return 0;
+}
+
+static inline int ixgbe_mat_prgm_dip(struct ixgbe_fdir_filter *input,
+				     union ixgbe_atr_input *mask,
+				     __u32 val, __u32 m)
+{
+	input->filter.formatted.dst_ip[0] = val;
+	mask->formatted.dst_ip[0] = m;
+	return 0;
+}
+
+static struct ixgbe_mat_field ixgbe_ipv4_fields[] = {
+	{ .off = 12, .mask = -1, .val = ixgbe_mat_prgm_sip,
+	  .type = IXGBE_ATR_FLOW_TYPE_IPV4},
+	{ .off = 16, .mask = -1, .val = ixgbe_mat_prgm_dip,
+	  .type = IXGBE_ATR_FLOW_TYPE_IPV4},
+	{ .val = NULL } /* terminal node */
+};
+
+static inline int ixgbe_mat_prgm_sport(struct ixgbe_fdir_filter *input,
+				       union ixgbe_atr_input *mask,
+				       __u32 val, __u32 m)
+{
+	input->filter.formatted.src_port = val & 0xffff;
+	mask->formatted.src_port = m & 0xffff;
+	return 0;
+};
+
+static inline int ixgbe_mat_prgm_dport(struct ixgbe_fdir_filter *input,
+				       union ixgbe_atr_input *mask,
+				       __u32 val, __u32 m)
+{
+	input->filter.formatted.dst_port = val & 0xffff;
+	mask->formatted.dst_port = m & 0xffff;
+	return 0;
+};
+
+static struct ixgbe_mat_field ixgbe_tcp_fields[] = {
+	{.off = 0, .mask = 0xffff, .val = ixgbe_mat_prgm_sport,
+	 .type = IXGBE_ATR_FLOW_TYPE_TCPV4},
+	{.off = 2, .mask = 0xffff, .val = ixgbe_mat_prgm_dport,
+	 .type = IXGBE_ATR_FLOW_TYPE_TCPV4},
+	{ .val = NULL } /* terminal node */
+};
+
+struct ixgbe_nexthdr {
+	/* offset, shift, and mask of position to next header */
+	unsigned int o;
+	__u32 s;
+	__u32 m;
+	/* match criteria to make this jump*/
+	unsigned int off;
+	__u32 val;
+	__u32 mask;
+	/* location of jump to make */
+	struct ixgbe_mat_field *jump;
+};
+
+static struct ixgbe_nexthdr ixgbe_ipv4_jumps[] = {
+	{ .o = 0, .s = 6, .m = 0xf,
+	  .off = 8, .val = 0x600, .mask = 0xff00, .jump = ixgbe_tcp_fields},
+	{ .jump = NULL } /* terminal node */
+};
+#endif /* _IXGBE_MODEL_H_ */
-- 
cgit v0.10.2


From b82b17d929a692df1122fedc0ff4ddcef9cb6ad4 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:18:53 -0800
Subject: net: ixgbe: add support for tc_u32 offload

This adds initial support for offloading the u32 tc classifier. This
initial implementation only implements a few base matches and actions
to illustrate the use of the infrastructure patches.

However it is an interesting subset because it handles the u32 next
hdr logic to correctly map tcp packets from ip headers using the ihl
and protocol fields. After this is accepted we can extend the match
and action fields easily by updating the model header file.

Also only the drop action is supported initially.

Here is a short test script,

 #tc qdisc add dev eth4 ingress
 #tc filter add dev eth4 parent ffff: protocol ip \
	u32 ht 800: order 1 \
	match ip dst 15.0.0.1/32 match ip src 15.0.0.2/32 action drop

<-- hardware has dst/src ip match rule installed -->

 #tc filter del dev eth4 parent ffff: prio 49152
 #tc filter add dev eth4 parent ffff: protocol ip prio 99 \
	handle 1: u32 divisor 1
 #tc filter add dev eth4 protocol ip parent ffff: prio 99 \
	u32 ht 800: order 1 link 1: \
	offset at 0 mask 0f00 shift 6 plus 0 eat match ip protocol 6 ff
 #tc filter add dev eth4 parent ffff: protocol ip \
	u32 ht 1: order 3 match tcp src 23 ffff action drop

<-- hardware has tcp src port rule installed -->

 #tc qdisc del dev eth4 parent ffff:

<-- hardware cleaned up -->

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 4b9156c..fc877c7 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -796,6 +796,9 @@ struct ixgbe_adapter {
 	u8 default_up;
 	unsigned long fwd_bitmask; /* Bitmask indicating in use pools */
 
+#define IXGBE_MAX_LINK_HANDLE 10
+	struct ixgbe_mat_field *jump_tables[IXGBE_MAX_LINK_HANDLE];
+
 /* maximum number of RETA entries among all devices supported by ixgbe
  * driver: currently it's x550 device in non-SRIOV mode
  */
@@ -925,6 +928,9 @@ s32 ixgbe_fdir_erase_perfect_filter_82599(struct ixgbe_hw *hw,
 					  u16 soft_id);
 void ixgbe_atr_compute_perfect_hash_82599(union ixgbe_atr_input *input,
 					  union ixgbe_atr_input *mask);
+int ixgbe_update_ethtool_fdir_entry(struct ixgbe_adapter *adapter,
+				    struct ixgbe_fdir_filter *input,
+				    u16 sw_idx);
 void ixgbe_set_rx_mode(struct net_device *netdev);
 #ifdef CONFIG_IXGBE_DCB
 void ixgbe_set_rx_drop_en(struct ixgbe_adapter *adapter);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index bea96b3..726e0ee 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -2520,9 +2520,9 @@ static int ixgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 	return ret;
 }
 
-static int ixgbe_update_ethtool_fdir_entry(struct ixgbe_adapter *adapter,
-					   struct ixgbe_fdir_filter *input,
-					   u16 sw_idx)
+int ixgbe_update_ethtool_fdir_entry(struct ixgbe_adapter *adapter,
+				    struct ixgbe_fdir_filter *input,
+				    u16 sw_idx)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	struct hlist_node *node2;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dca2298..abdfaea 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -51,6 +51,8 @@
 #include <linux/prefetch.h>
 #include <scsi/fc/fc_fcoe.h>
 #include <net/vxlan.h>
+#include <net/pkt_cls.h>
+#include <net/tc_act/tc_gact.h>
 
 #ifdef CONFIG_OF
 #include <linux/of_net.h>
@@ -65,6 +67,7 @@
 #include "ixgbe_common.h"
 #include "ixgbe_dcb_82599.h"
 #include "ixgbe_sriov.h"
+#include "ixgbe_model.h"
 
 char ixgbe_driver_name[] = "ixgbe";
 static const char ixgbe_driver_string[] =
@@ -5545,6 +5548,9 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter)
 #endif /* CONFIG_IXGBE_DCB */
 #endif /* IXGBE_FCOE */
 
+	/* initialize static ixgbe jump table entries */
+	adapter->jump_tables[0] = ixgbe_ipv4_fields;
+
 	adapter->mac_table = kzalloc(sizeof(struct ixgbe_mac_addr) *
 				     hw->mac.num_rar_entries,
 				     GFP_ATOMIC);
@@ -8200,10 +8206,191 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 	return 0;
 }
 
+static int ixgbe_delete_clsu32(struct ixgbe_adapter *adapter,
+			       struct tc_cls_u32_offload *cls)
+{
+	int err;
+
+	spin_lock(&adapter->fdir_perfect_lock);
+	err = ixgbe_update_ethtool_fdir_entry(adapter, NULL, cls->knode.handle);
+	spin_unlock(&adapter->fdir_perfect_lock);
+	return err;
+}
+
+static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter,
+				  __be16 protocol,
+				  struct tc_cls_u32_offload *cls)
+{
+	u32 loc = cls->knode.handle & 0xfffff;
+	struct ixgbe_hw *hw = &adapter->hw;
+	struct ixgbe_mat_field *field_ptr;
+	struct ixgbe_fdir_filter *input;
+	union ixgbe_atr_input mask;
+#ifdef CONFIG_NET_CLS_ACT
+	const struct tc_action *a;
+#endif
+	int i, err = 0;
+	u8 queue;
+	u32 handle;
+
+	memset(&mask, 0, sizeof(union ixgbe_atr_input));
+	handle = cls->knode.handle;
+
+	/* At the moment cls_u32 jumps to transport layer and skips past
+	 * L2 headers. The canonical method to match L2 frames is to use
+	 * negative values. However this is error prone at best but really
+	 * just broken because there is no way to "know" what sort of hdr
+	 * is in front of the transport layer. Fix cls_u32 to support L2
+	 * headers when needed.
+	 */
+	if (protocol != htons(ETH_P_IP))
+		return -EINVAL;
+
+	if (cls->knode.link_handle ||
+	    cls->knode.link_handle >= IXGBE_MAX_LINK_HANDLE) {
+		struct ixgbe_nexthdr *nexthdr = ixgbe_ipv4_jumps;
+		u32 uhtid = TC_U32_USERHTID(cls->knode.link_handle);
+
+		for (i = 0; nexthdr[i].jump; i++) {
+			if (nexthdr->o != cls->knode.sel->offoff ||
+			    nexthdr->s != cls->knode.sel->offshift ||
+			    nexthdr->m != cls->knode.sel->offmask ||
+			    /* do not support multiple key jumps its just mad */
+			    cls->knode.sel->nkeys > 1)
+				return -EINVAL;
+
+			if (nexthdr->off != cls->knode.sel->keys[0].off ||
+			    nexthdr->val != cls->knode.sel->keys[0].val ||
+			    nexthdr->mask != cls->knode.sel->keys[0].mask)
+				return -EINVAL;
+
+			if (uhtid >= IXGBE_MAX_LINK_HANDLE)
+				return -EINVAL;
+
+			adapter->jump_tables[uhtid] = nexthdr->jump;
+		}
+		return 0;
+	}
+
+	if (loc >= ((1024 << adapter->fdir_pballoc) - 2)) {
+		e_err(drv, "Location out of range\n");
+		return -EINVAL;
+	}
+
+	/* cls u32 is a graph starting at root node 0x800. The driver tracks
+	 * links and also the fields used to advance the parser across each
+	 * link (e.g. nexthdr/eat parameters from 'tc'). This way we can map
+	 * the u32 graph onto the hardware parse graph denoted in ixgbe_model.h
+	 * To add support for new nodes update ixgbe_model.h parse structures
+	 * this function _should_ be generic try not to hardcode values here.
+	 */
+	if (TC_U32_USERHTID(handle) == 0x800) {
+		field_ptr = adapter->jump_tables[0];
+	} else {
+		if (TC_U32_USERHTID(handle) >= ARRAY_SIZE(adapter->jump_tables))
+			return -EINVAL;
+
+		field_ptr = adapter->jump_tables[TC_U32_USERHTID(handle)];
+	}
+
+	if (!field_ptr)
+		return -EINVAL;
+
+	input = kzalloc(sizeof(*input), GFP_KERNEL);
+	if (!input)
+		return -ENOMEM;
+
+	for (i = 0; i < cls->knode.sel->nkeys; i++) {
+		int off = cls->knode.sel->keys[i].off;
+		__be32 val = cls->knode.sel->keys[i].val;
+		__be32 m = cls->knode.sel->keys[i].mask;
+		bool found_entry = false;
+		int j;
+
+		for (j = 0; field_ptr[j].val; j++) {
+			if (field_ptr[j].off == off &&
+			    field_ptr[j].mask == m) {
+				field_ptr[j].val(input, &mask, val, m);
+				input->filter.formatted.flow_type |=
+					field_ptr[j].type;
+				found_entry = true;
+				break;
+			}
+		}
+
+		if (!found_entry)
+			goto err_out;
+	}
+
+	mask.formatted.flow_type = IXGBE_ATR_L4TYPE_IPV6_MASK |
+				   IXGBE_ATR_L4TYPE_MASK;
+
+	if (input->filter.formatted.flow_type == IXGBE_ATR_FLOW_TYPE_IPV4)
+		mask.formatted.flow_type &= IXGBE_ATR_L4TYPE_IPV6_MASK;
+
+#ifdef CONFIG_NET_CLS_ACT
+	if (list_empty(&cls->knode.exts->actions))
+		goto err_out;
+
+	list_for_each_entry(a, &cls->knode.exts->actions, list) {
+		if (!is_tcf_gact_shot(a))
+			goto err_out;
+	}
+#endif
+
+	input->action = IXGBE_FDIR_DROP_QUEUE;
+	queue = IXGBE_FDIR_DROP_QUEUE;
+	input->sw_idx = loc;
+
+	spin_lock(&adapter->fdir_perfect_lock);
+
+	if (hlist_empty(&adapter->fdir_filter_list)) {
+		memcpy(&adapter->fdir_mask, &mask, sizeof(mask));
+		err = ixgbe_fdir_set_input_mask_82599(hw, &mask);
+		if (err)
+			goto err_out_w_lock;
+	} else if (memcmp(&adapter->fdir_mask, &mask, sizeof(mask))) {
+		err = -EINVAL;
+		goto err_out_w_lock;
+	}
+
+	ixgbe_atr_compute_perfect_hash_82599(&input->filter, &mask);
+	err = ixgbe_fdir_write_perfect_filter_82599(hw, &input->filter,
+						    input->sw_idx, queue);
+	if (!err)
+		ixgbe_update_ethtool_fdir_entry(adapter, input, input->sw_idx);
+	spin_unlock(&adapter->fdir_perfect_lock);
+
+	return err;
+err_out_w_lock:
+	spin_unlock(&adapter->fdir_perfect_lock);
+err_out:
+	kfree(input);
+	return -EINVAL;
+}
+
 int __ixgbe_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 		     struct tc_to_netdev *tc)
 {
-	/* Only support egress tc setup for now */
+	struct ixgbe_adapter *adapter = netdev_priv(dev);
+
+	if (TC_H_MAJ(handle) == TC_H_MAJ(TC_H_INGRESS) &&
+	    tc->type == TC_SETUP_CLSU32) {
+		if (!(dev->features & NETIF_F_HW_TC))
+			return -EINVAL;
+
+		switch (tc->cls_u32->command) {
+		case TC_CLSU32_NEW_KNODE:
+		case TC_CLSU32_REPLACE_KNODE:
+			return ixgbe_configure_clsu32(adapter,
+						      proto, tc->cls_u32);
+		case TC_CLSU32_DELETE_KNODE:
+			return ixgbe_delete_clsu32(adapter, tc->cls_u32);
+		default:
+			return -EINVAL;
+		}
+	}
+
 	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
@@ -8272,19 +8459,17 @@ static int ixgbe_set_features(struct net_device *netdev,
 	}
 
 	/*
-	 * Check if Flow Director n-tuple support was enabled or disabled.  If
-	 * the state changed, we need to reset.
+	 * Check if Flow Director n-tuple support or hw_tc support was
+	 * enabled or disabled.  If the state changed, we need to reset.
 	 */
-	switch (features & NETIF_F_NTUPLE) {
-	case NETIF_F_NTUPLE:
+	if ((features & NETIF_F_NTUPLE) || (features & NETIF_F_HW_TC)) {
 		/* turn off ATR, enable perfect filters and reset */
 		if (!(adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE))
 			need_reset = true;
 
 		adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
 		adapter->flags |= IXGBE_FLAG_FDIR_PERFECT_CAPABLE;
-		break;
-	default:
+	} else {
 		/* turn off perfect filters, enable ATR and reset */
 		if (adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE)
 			need_reset = true;
@@ -8292,23 +8477,16 @@ static int ixgbe_set_features(struct net_device *netdev,
 		adapter->flags &= ~IXGBE_FLAG_FDIR_PERFECT_CAPABLE;
 
 		/* We cannot enable ATR if SR-IOV is enabled */
-		if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
-			break;
-
-		/* We cannot enable ATR if we have 2 or more traffic classes */
-		if (netdev_get_num_tc(netdev) > 1)
-			break;
-
-		/* We cannot enable ATR if RSS is disabled */
-		if (adapter->ring_feature[RING_F_RSS].limit <= 1)
-			break;
-
-		/* A sample rate of 0 indicates ATR disabled */
-		if (!adapter->atr_sample_rate)
-			break;
-
-		adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
-		break;
+		if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED ||
+		    /* We cannot enable ATR if we have 2 or more tcs */
+		    (netdev_get_num_tc(netdev) > 1) ||
+		    /* We cannot enable ATR if RSS is disabled */
+		    (adapter->ring_feature[RING_F_RSS].limit <= 1) ||
+		    /* A sample rate of 0 indicates ATR disabled */
+		    (!adapter->atr_sample_rate))
+			; /* do nothing not supported */
+		else /* otherwise supported and set the flag */
+			adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
 	}
 
 	if (features & NETIF_F_HW_VLAN_CTAG_RX)
@@ -8667,9 +8845,7 @@ static const struct net_device_ops ixgbe_netdev_ops = {
 	.ndo_set_vf_trust	= ixgbe_ndo_set_vf_trust,
 	.ndo_get_vf_config	= ixgbe_ndo_get_vf_config,
 	.ndo_get_stats64	= ixgbe_get_stats64,
-#ifdef CONFIG_IXGBE_DCB
 	.ndo_setup_tc		= __ixgbe_setup_tc,
-#endif
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= ixgbe_netpoll,
 #endif
@@ -9040,7 +9216,8 @@ skip_sriov:
 	case ixgbe_mac_X550EM_x:
 		netdev->features |= NETIF_F_SCTP_CRC;
 		netdev->hw_features |= NETIF_F_SCTP_CRC |
-				       NETIF_F_NTUPLE;
+				       NETIF_F_NTUPLE |
+				       NETIF_F_HW_TC;
 		break;
 	default:
 		break;
-- 
cgit v0.10.2


From db956ae882f4e7aa99c9c242a91ae942d08b6939 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Tue, 16 Feb 2016 21:19:19 -0800
Subject: net: ixgbe: abort with cls u32 divisor groups greater than 1

This patch ensures ixgbe will not try to offload hash tables from the
u32 module. The device class does not currently support this so until
it is enabled just abort on these tables.

Interestingly the more flexible your hardware is the less code you
need to implement to guard against these cases.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index fc877c7..84fa28c 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -798,6 +798,7 @@ struct ixgbe_adapter {
 
 #define IXGBE_MAX_LINK_HANDLE 10
 	struct ixgbe_mat_field *jump_tables[IXGBE_MAX_LINK_HANDLE];
+	unsigned long tables;
 
 /* maximum number of RETA entries among all devices supported by ixgbe
  * driver: currently it's x550 device in non-SRIOV mode
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index abdfaea..cf4b729 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8217,6 +8217,27 @@ static int ixgbe_delete_clsu32(struct ixgbe_adapter *adapter,
 	return err;
 }
 
+static int ixgbe_configure_clsu32_add_hnode(struct ixgbe_adapter *adapter,
+					    __be16 protocol,
+					    struct tc_cls_u32_offload *cls)
+{
+	/* This ixgbe devices do not support hash tables at the moment
+	 * so abort when given hash tables.
+	 */
+	if (cls->hnode.divisor > 0)
+		return -EINVAL;
+
+	set_bit(TC_U32_USERHTID(cls->hnode.handle), &adapter->tables);
+	return 0;
+}
+
+static int ixgbe_configure_clsu32_del_hnode(struct ixgbe_adapter *adapter,
+					    struct tc_cls_u32_offload *cls)
+{
+	clear_bit(TC_U32_USERHTID(cls->hnode.handle), &adapter->tables);
+	return 0;
+}
+
 static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter,
 				  __be16 protocol,
 				  struct tc_cls_u32_offload *cls)
@@ -8251,6 +8272,9 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter,
 		struct ixgbe_nexthdr *nexthdr = ixgbe_ipv4_jumps;
 		u32 uhtid = TC_U32_USERHTID(cls->knode.link_handle);
 
+		if (!test_bit(uhtid, &adapter->tables))
+			return -EINVAL;
+
 		for (i = 0; nexthdr[i].jump; i++) {
 			if (nexthdr->o != cls->knode.sel->offoff ||
 			    nexthdr->s != cls->knode.sel->offshift ||
@@ -8386,6 +8410,13 @@ int __ixgbe_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 						      proto, tc->cls_u32);
 		case TC_CLSU32_DELETE_KNODE:
 			return ixgbe_delete_clsu32(adapter, tc->cls_u32);
+		case TC_CLSU32_NEW_HNODE:
+		case TC_CLSU32_REPLACE_HNODE:
+			return ixgbe_configure_clsu32_add_hnode(adapter, proto,
+								tc->cls_u32);
+		case TC_CLSU32_DELETE_HNODE:
+			return ixgbe_configure_clsu32_del_hnode(adapter,
+								tc->cls_u32);
 		default:
 			return -EINVAL;
 		}
-- 
cgit v0.10.2


From f3699b3c5752216ec1529d6bd7d13d9b1a6d1b9d Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:01 -0800
Subject: i40e: trivial: drop duplicate definition

The probe routine already had a u32 val declared, no need
to do it again.  Found by W=2 compile.

Change-ID: Id7b65f6d0ef6bb71067d0557f5be0202b6d8741e
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 6b158c4..90ad7d7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -11008,8 +11008,6 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if ((pf->flags & I40E_FLAG_SRIOV_ENABLED) &&
 	    (pf->flags & I40E_FLAG_MSIX_ENABLED) &&
 	    !test_bit(__I40E_BAD_EEPROM, &pf->state)) {
-		u32 val;
-
 		/* disable link interrupts for VFs */
 		val = rd32(hw, I40E_PFGEN_PORTMDIO_NUM);
 		val &= ~I40E_PFGEN_PORTMDIO_NUM_VFLINK_STAT_ENA_MASK;
-- 
cgit v0.10.2


From d89d967f7188438cf9a3ffb9ce74e9ecc2795619 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:02 -0800
Subject: i40e: trivial: fix missing space

Missing space in comment, fixed.

Change-ID: I8cdf3ce5994b4a97dcc3eeb33422533918546667
Reported-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 90ad7d7..c4ad0f1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3754,7 +3754,7 @@ static int i40e_vsi_request_irq(struct i40e_vsi *vsi, char *basename)
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
 /**
- * i40e_netpoll - A Polling 'interrupt'handler
+ * i40e_netpoll - A Polling 'interrupt' handler
  * @netdev: network interface device structure
  *
  * This is used by netconsole to send skbs without having to re-enable
-- 
cgit v0.10.2


From 3578fa0a8cd6d9fb03092c673f42aad0e7c850df Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:03 -0800
Subject: i40e: fix bug in dma sync

Driver was using an offset based off a DMA handle while mapping and
unmapping using sync_single_range_for[cpu|device], where it should
be using DMA handle (returned from alloc_coherent) and the offset of the
memory to be sync'd.

Change-ID: I208256565b1595ff0e9171ab852de06b997917c6
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: Nelson, Shannon <shannon.nelson@intel.com>
Reviewed-by: Williams, Mitch A <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 6234136..f2e6afc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1226,8 +1226,8 @@ void i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 		}
 
 		dma_sync_single_range_for_device(rx_ring->dev,
-						 bi->dma,
-						 0,
+						 rx_ring->rx_bi[0].dma,
+						 i * rx_ring->rx_hdr_len,
 						 rx_ring->rx_hdr_len,
 						 DMA_FROM_DEVICE);
 		/* Refresh the desc even if buffer_addrs didn't change
@@ -1542,8 +1542,8 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 			skb_record_rx_queue(skb, rx_ring->queue_index);
 			/* we are reusing so sync this buffer for CPU use */
 			dma_sync_single_range_for_cpu(rx_ring->dev,
-						      rx_bi->dma,
-						      0,
+						      rx_ring->rx_bi[0].dma,
+						      i * rx_ring->rx_hdr_len,
 						      rx_ring->rx_hdr_len,
 						      DMA_FROM_DEVICE);
 		}
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 3e69235..69fb85f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -700,8 +700,8 @@ void i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 		}
 
 		dma_sync_single_range_for_device(rx_ring->dev,
-						 bi->dma,
-						 0,
+						 rx_ring->rx_bi[0].dma,
+						 i * rx_ring->rx_hdr_len,
 						 rx_ring->rx_hdr_len,
 						 DMA_FROM_DEVICE);
 		/* Refresh the desc even if buffer_addrs didn't change
@@ -1007,8 +1007,8 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 			skb_record_rx_queue(skb, rx_ring->queue_index);
 			/* we are reusing so sync this buffer for CPU use */
 			dma_sync_single_range_for_cpu(rx_ring->dev,
-						      rx_bi->dma,
-						      0,
+						      rx_ring->rx_bi[0].dma,
+						      i * rx_ring->rx_hdr_len,
 						      rx_ring->rx_hdr_len,
 						      DMA_FROM_DEVICE);
 		}
-- 
cgit v0.10.2


From e9f6563d7bb3914711c6601cdf5d9f9b6e1b461f Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Mon, 4 Jan 2016 10:33:04 -0800
Subject: i40e: do TSO only if CHECKSUM_PARTIAL is set

Don't bother trying to set up a TSO if the skb->ip_summed is not
set to CHECKSUM_PARTIAL.

Change-ID: I6495b3568e404907a2965b48cf3e2effa7c9ab55
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index f2e6afc..15130b5d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2221,6 +2221,9 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	u32 l4len;
 	int err;
 
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
 	if (!skb_is_gso(skb))
 		return 0;
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 69fb85f..4205aef 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1471,6 +1471,9 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	u32 l4len;
 	int err;
 
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
 	if (!skb_is_gso(skb))
 		return 0;
 
-- 
cgit v0.10.2


From 38c3cec73c7be6789a3e3ed06ff74ce84f5b1528 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:05 -0800
Subject: i40e: allocate memory safer

The sync_vsi_filter function was allocating memory in such
a way that it could sleep (GFP_KERNEL) which was causing a problem
when called by the team driver under rcu_read_lock(), which cannot
be held while sleeping.  Found with lockdep.

Change-ID: I4e59053cb5eedcf3d0ca151715be3dc42a94bdd5
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index c4ad0f1..fe67b90 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1937,7 +1937,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 			    sizeof(struct i40e_aqc_remove_macvlan_element_data);
 		del_list_size = filter_list_len *
 			    sizeof(struct i40e_aqc_remove_macvlan_element_data);
-		del_list = kzalloc(del_list_size, GFP_KERNEL);
+		del_list = kzalloc(del_list_size, GFP_ATOMIC);
 		if (!del_list) {
 			i40e_cleanup_add_list(&tmp_add_list);
 
@@ -2015,7 +2015,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 			       sizeof(struct i40e_aqc_add_macvlan_element_data),
 		add_list_size = filter_list_len *
 			       sizeof(struct i40e_aqc_add_macvlan_element_data);
-		add_list = kzalloc(add_list_size, GFP_KERNEL);
+		add_list = kzalloc(add_list_size, GFP_ATOMIC);
 		if (!add_list) {
 			/* Purge element from temporary lists */
 			i40e_cleanup_add_list(&tmp_add_list);
-- 
cgit v0.10.2


From c53934c6d1b11c676dafed04294cd906fc9ff5d3 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:06 -0800
Subject: i40e: fix: do not sleep in netdev_ops

The driver was being called by VLAN, bonding, teaming operations
that expected to be able to hold locks like rcu_read_lock().

This causes the driver to be held to the requirement to not sleep,
and was found by the kernel debug options for checking sleep
inside critical section, and the locking validator.

Change-ID: Ibc68c835f5ffa8ffe0638ffe910a66fc5649a7f7
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: Nelson, Shannon <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index fe67b90..ac89739 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1542,7 +1542,11 @@ static int i40e_set_mac(struct net_device *netdev, void *p)
 
 	ether_addr_copy(netdev->dev_addr, addr->sa_data);
 
-	return i40e_sync_vsi_filters(vsi);
+	/* schedule our worker thread which will take care of
+	 * applying the new filter changes
+	 */
+	i40e_service_event_schedule(vsi->back);
+	return 0;
 }
 
 /**
@@ -1766,6 +1770,11 @@ bottom_of_search_loop:
 		vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
 		vsi->back->flags |= I40E_FLAG_FILTER_SYNC;
 	}
+
+	/* schedule our worker thread which will take care of
+	 * applying the new filter changes
+	 */
+	i40e_service_event_schedule(vsi->back);
 }
 
 /**
-- 
cgit v0.10.2


From 7bd6875bef7055170b2c871e6d44a0b5e826643f Mon Sep 17 00:00:00 2001
From: Kiran Patil <kiran.patil@intel.com>
Date: Mon, 4 Jan 2016 10:33:07 -0800
Subject: i40e: APIs to Add/remove port mirroring rules

This patch implements necessary functions related to port
mirroring features such as add/delete mirror rule, function
to set promiscuous VLAN mode for VSI if mirror rule_type is
"VLAN Mirroring".

Change-ID: Iaf513fd5f188f99dcb977b48f99e73185dfddc40
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 9018f28..976b03f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -2033,6 +2033,37 @@ i40e_status i40e_aq_set_vsi_broadcast(struct i40e_hw *hw,
 }
 
 /**
+ * i40e_aq_set_vsi_vlan_promisc - control the VLAN promiscuous setting
+ * @hw: pointer to the hw struct
+ * @seid: vsi number
+ * @enable: set MAC L2 layer unicast promiscuous enable/disable for a given VLAN
+ * @cmd_details: pointer to command details structure or NULL
+ **/
+i40e_status i40e_aq_set_vsi_vlan_promisc(struct i40e_hw *hw,
+				       u16 seid, bool enable,
+				       struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
+		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
+	i40e_status status;
+	u16 flags = 0;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					i40e_aqc_opc_set_vsi_promiscuous_modes);
+	if (enable)
+		flags |= I40E_AQC_SET_VSI_PROMISC_VLAN;
+
+	cmd->promiscuous_flags = cpu_to_le16(flags);
+	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_VLAN);
+	cmd->seid = cpu_to_le16(seid);
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
  * i40e_get_vsi_params - get VSI configuration info
  * @hw: pointer to the hw struct
  * @vsi_ctx: pointer to a vsi context struct
@@ -2470,6 +2501,137 @@ i40e_status i40e_aq_remove_macvlan(struct i40e_hw *hw, u16 seid,
 }
 
 /**
+ * i40e_mirrorrule_op - Internal helper function to add/delete mirror rule
+ * @hw: pointer to the hw struct
+ * @opcode: AQ opcode for add or delete mirror rule
+ * @sw_seid: Switch SEID (to which rule refers)
+ * @rule_type: Rule Type (ingress/egress/VLAN)
+ * @id: Destination VSI SEID or Rule ID
+ * @count: length of the list
+ * @mr_list: list of mirrored VSI SEIDs or VLAN IDs
+ * @cmd_details: pointer to command details structure or NULL
+ * @rule_id: Rule ID returned from FW
+ * @rule_used: Number of rules used in internal switch
+ * @rule_free: Number of rules free in internal switch
+ *
+ * Add/Delete a mirror rule to a specific switch. Mirror rules are supported for
+ * VEBs/VEPA elements only
+ **/
+static i40e_status i40e_mirrorrule_op(struct i40e_hw *hw,
+				u16 opcode, u16 sw_seid, u16 rule_type, u16 id,
+				u16 count, __le16 *mr_list,
+				struct i40e_asq_cmd_details *cmd_details,
+				u16 *rule_id, u16 *rules_used, u16 *rules_free)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_add_delete_mirror_rule *cmd =
+		(struct i40e_aqc_add_delete_mirror_rule *)&desc.params.raw;
+	struct i40e_aqc_add_delete_mirror_rule_completion *resp =
+	(struct i40e_aqc_add_delete_mirror_rule_completion *)&desc.params.raw;
+	i40e_status status;
+	u16 buf_size;
+
+	buf_size = count * sizeof(*mr_list);
+
+	/* prep the rest of the request */
+	i40e_fill_default_direct_cmd_desc(&desc, opcode);
+	cmd->seid = cpu_to_le16(sw_seid);
+	cmd->rule_type = cpu_to_le16(rule_type &
+				     I40E_AQC_MIRROR_RULE_TYPE_MASK);
+	cmd->num_entries = cpu_to_le16(count);
+	/* Dest VSI for add, rule_id for delete */
+	cmd->destination = cpu_to_le16(id);
+	if (mr_list) {
+		desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF |
+						I40E_AQ_FLAG_RD));
+		if (buf_size > I40E_AQ_LARGE_BUF)
+			desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
+	}
+
+	status = i40e_asq_send_command(hw, &desc, mr_list, buf_size,
+				       cmd_details);
+	if (!status ||
+	    hw->aq.asq_last_status == I40E_AQ_RC_ENOSPC) {
+		if (rule_id)
+			*rule_id = le16_to_cpu(resp->rule_id);
+		if (rules_used)
+			*rules_used = le16_to_cpu(resp->mirror_rules_used);
+		if (rules_free)
+			*rules_free = le16_to_cpu(resp->mirror_rules_free);
+	}
+	return status;
+}
+
+/**
+ * i40e_aq_add_mirrorrule - add a mirror rule
+ * @hw: pointer to the hw struct
+ * @sw_seid: Switch SEID (to which rule refers)
+ * @rule_type: Rule Type (ingress/egress/VLAN)
+ * @dest_vsi: SEID of VSI to which packets will be mirrored
+ * @count: length of the list
+ * @mr_list: list of mirrored VSI SEIDs or VLAN IDs
+ * @cmd_details: pointer to command details structure or NULL
+ * @rule_id: Rule ID returned from FW
+ * @rule_used: Number of rules used in internal switch
+ * @rule_free: Number of rules free in internal switch
+ *
+ * Add mirror rule. Mirror rules are supported for VEBs or VEPA elements only
+ **/
+i40e_status i40e_aq_add_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
+			u16 rule_type, u16 dest_vsi, u16 count, __le16 *mr_list,
+			struct i40e_asq_cmd_details *cmd_details,
+			u16 *rule_id, u16 *rules_used, u16 *rules_free)
+{
+	if (!(rule_type == I40E_AQC_MIRROR_RULE_TYPE_ALL_INGRESS ||
+	    rule_type == I40E_AQC_MIRROR_RULE_TYPE_ALL_EGRESS)) {
+		if (count == 0 || !mr_list)
+			return I40E_ERR_PARAM;
+	}
+
+	return i40e_mirrorrule_op(hw, i40e_aqc_opc_add_mirror_rule, sw_seid,
+				  rule_type, dest_vsi, count, mr_list,
+				  cmd_details, rule_id, rules_used, rules_free);
+}
+
+/**
+ * i40e_aq_delete_mirrorrule - delete a mirror rule
+ * @hw: pointer to the hw struct
+ * @sw_seid: Switch SEID (to which rule refers)
+ * @rule_type: Rule Type (ingress/egress/VLAN)
+ * @count: length of the list
+ * @rule_id: Rule ID that is returned in the receive desc as part of
+ *		add_mirrorrule.
+ * @mr_list: list of mirrored VLAN IDs to be removed
+ * @cmd_details: pointer to command details structure or NULL
+ * @rule_used: Number of rules used in internal switch
+ * @rule_free: Number of rules free in internal switch
+ *
+ * Delete a mirror rule. Mirror rules are supported for VEBs/VEPA elements only
+ **/
+i40e_status i40e_aq_delete_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
+			u16 rule_type, u16 rule_id, u16 count, __le16 *mr_list,
+			struct i40e_asq_cmd_details *cmd_details,
+			u16 *rules_used, u16 *rules_free)
+{
+	/* Rule ID has to be valid except rule_type: INGRESS VLAN mirroring */
+	if (rule_type != I40E_AQC_MIRROR_RULE_TYPE_VLAN) {
+		if (!rule_id)
+			return I40E_ERR_PARAM;
+	} else {
+		/* count and mr_list shall be valid for rule_type INGRESS VLAN
+		 * mirroring. For other rule_type, count and rule_type should
+		 * not matter.
+		 */
+		if (count == 0 || !mr_list)
+			return I40E_ERR_PARAM;
+	}
+
+	return i40e_mirrorrule_op(hw, i40e_aqc_opc_delete_mirror_rule, sw_seid,
+				  rule_type, rule_id, count, mr_list,
+				  cmd_details, NULL, rules_used, rules_free);
+}
+
+/**
  * i40e_aq_send_msg_to_vf
  * @hw: pointer to the hardware structure
  * @vfid: VF id to send msg
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index bb9d583..45af29b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -127,6 +127,9 @@ i40e_status i40e_aq_set_vsi_unicast_promiscuous(struct i40e_hw *hw,
 		u16 vsi_id, bool set, struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_set_vsi_multicast_promiscuous(struct i40e_hw *hw,
 		u16 vsi_id, bool set, struct i40e_asq_cmd_details *cmd_details);
+i40e_status i40e_aq_set_vsi_vlan_promisc(struct i40e_hw *hw,
+				u16 seid, bool enable,
+				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_get_vsi_params(struct i40e_hw *hw,
 				struct i40e_vsi_context *vsi_ctx,
 				struct i40e_asq_cmd_details *cmd_details);
@@ -149,6 +152,15 @@ i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 vsi_id,
 i40e_status i40e_aq_remove_macvlan(struct i40e_hw *hw, u16 vsi_id,
 			struct i40e_aqc_remove_macvlan_element_data *mv_list,
 			u16 count, struct i40e_asq_cmd_details *cmd_details);
+i40e_status i40e_aq_add_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
+			u16 rule_type, u16 dest_vsi, u16 count, __le16 *mr_list,
+			struct i40e_asq_cmd_details *cmd_details,
+			u16 *rule_id, u16 *rules_used, u16 *rules_free);
+i40e_status i40e_aq_delete_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
+			u16 rule_type, u16 rule_id, u16 count, __le16 *mr_list,
+			struct i40e_asq_cmd_details *cmd_details,
+			u16 *rules_used, u16 *rules_free);
+
 i40e_status i40e_aq_send_msg_to_vf(struct i40e_hw *hw, u16 vfid,
 				u32 v_opcode, u32 v_retval, u8 *msg, u16 msglen,
 				struct i40e_asq_cmd_details *cmd_details);
-- 
cgit v0.10.2


From 2f0aff4151e13fff669b1fab4b93b7613b3b4be5 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Mon, 4 Jan 2016 10:33:08 -0800
Subject: i40e: negate PHY int mask bits

The PHY interrupt mask bits mask out the events we don't want,
so we need to negate the bitmask of events we want.

Change-ID: I273244da5a8d285b6abc84fd68a90f1e6fa0393e
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index ac89739..e147db2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -6826,12 +6826,12 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit)
 	if (ret)
 		goto end_core_reset;
 
-	/* driver is only interested in link up/down and module qualification
-	 * reports from firmware
+	/* The driver only wants link up/down and module qualification
+	 * reports from firmware.  Note the negative logic.
 	 */
 	ret = i40e_aq_set_phy_int_mask(&pf->hw,
-				       I40E_AQ_EVENT_LINK_UPDOWN |
-				       I40E_AQ_EVENT_MODULE_QUAL_FAIL, NULL);
+				       ~(I40E_AQ_EVENT_LINK_UPDOWN |
+					 I40E_AQ_EVENT_MODULE_QUAL_FAIL), NULL);
 	if (ret)
 		dev_info(&pf->pdev->dev, "set phy mask fail, err %s aq_err %s\n",
 			 i40e_stat_str(&pf->hw, ret),
@@ -10961,12 +10961,12 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		}
 	}
 
-	/* driver is only interested in link up/down and module qualification
-	 * reports from firmware
+	/* The driver only wants link up/down and module qualification
+	 * reports from firmware.  Note the negative logic.
 	 */
 	err = i40e_aq_set_phy_int_mask(&pf->hw,
-				       I40E_AQ_EVENT_LINK_UPDOWN |
-				       I40E_AQ_EVENT_MODULE_QUAL_FAIL, NULL);
+				       ~(I40E_AQ_EVENT_LINK_UPDOWN |
+					 I40E_AQ_EVENT_MODULE_QUAL_FAIL), NULL);
 	if (err)
 		dev_info(&pf->pdev->dev, "set phy mask fail, err %s aq_err %s\n",
 			 i40e_stat_str(&pf->hw, err),
-- 
cgit v0.10.2


From 426bda0fe47353c16893b689e01b9c8a4d668c2d Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:09 -0800
Subject: i40e: drop unused function

Delete the unused irq_dynamic_disable function.

Change-ID: Ia46071066babd121c7c90f141b6210b00078de3f
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: Anjali Singhai <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 175bec3..65cf344 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -757,7 +757,6 @@ static inline void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector)
 	/* skip the flush */
 }
 
-void i40e_irq_dynamic_disable(struct i40e_vsi *vsi, int vector);
 void i40e_irq_dynamic_disable_icr0(struct i40e_pf *pf);
 void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf);
 #ifdef I40E_FCOE
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index e147db2..38fd942 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3273,22 +3273,6 @@ void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf)
 }
 
 /**
- * i40e_irq_dynamic_disable - Disable default interrupt generation settings
- * @vsi: pointer to a vsi
- * @vector: disable a particular Hw Interrupt vector
- **/
-void i40e_irq_dynamic_disable(struct i40e_vsi *vsi, int vector)
-{
-	struct i40e_pf *pf = vsi->back;
-	struct i40e_hw *hw = &pf->hw;
-	u32 val;
-
-	val = I40E_ITR_NONE << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT;
-	wr32(hw, I40E_PFINT_DYN_CTLN(vector - 1), val);
-	i40e_flush(hw);
-}
-
-/**
  * i40e_msix_clean_rings - MSIX mode Interrupt Handler
  * @irq: interrupt number
  * @data: pointer to a q_vector
-- 
cgit v0.10.2


From c40918c3ad80e550ec68339d6816059e11648012 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:10 -0800
Subject: i40e: count allocation errors

Driver already counted allocation errors, so print
them as part of the ethtool -S output.  Useful for debugging
if your system is having trouble making memory available for
the driver.

Change-ID: I83839fa86e81e6d80f03b917c88dd3ef9a64dde0
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 812c954..9dcdf34 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -89,6 +89,8 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = {
 	I40E_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol),
 	I40E_VSI_STAT("tx_linearize", tx_linearize),
 	I40E_VSI_STAT("tx_force_wb", tx_force_wb),
+	I40E_VSI_STAT("rx_alloc_fail", rx_buf_failed),
+	I40E_VSI_STAT("rx_pg_alloc_fail", rx_page_failed),
 };
 
 /* These PF_STATs might look like duplicates of some NETDEV_STATs,
-- 
cgit v0.10.2


From 45d043597d2d0e780e768866c6fbfe8dbee5f2cf Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Mon, 4 Jan 2016 10:33:11 -0800
Subject: i40e: avoid large memcpy by assigning struct

Assign the i40e_pf structure directly instead of using a large memcpy,
which avoids a sparse warning and lets the compiler optimize the copy
since it knows the size of the structure in advance.

Change-ID: I17604e23be2616521eb760290befcb767b52b3f7
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 7aae056..3948587 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -185,9 +185,11 @@ static ssize_t i40e_dbg_dump_write(struct file *filp,
 		if (i40e_dbg_prep_dump_buf(pf, buflen)) {
 			p = i40e_dbg_dump_buf;
 
-			len = sizeof(struct i40e_pf);
-			memcpy(p, pf, len);
-			p += len;
+			/* avoid use of memcpy here due to sparse warning
+			 * about copy size.
+			 */
+			*((struct i40e_pf *)p) = *pf;
+			p += sizeof(struct i40e_pf);
 
 			len = (sizeof(struct i40e_aq_desc)
 					* pf->hw.aq.num_asq_entries);
-- 
cgit v0.10.2


From 4580de0de484a33814ccf0f4ae9c91dcd828a140 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 4 Jan 2016 10:33:12 -0800
Subject: i40e/i40evf: bump version to 1.4.12/1.4.8

Bump driver versions to i40e-1.4.12 and i40evf-1.4.8

Change-ID: I0ad82668c4ded04250391fda396ce191a42ab754
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 38fd942..6e12626 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -51,7 +51,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 11
+#define DRV_VERSION_BUILD 12
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index ae6c6e1..045cc7f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -38,7 +38,7 @@ static const char i40evf_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 7
+#define DRV_VERSION_BUILD 8
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD) \
-- 
cgit v0.10.2


From a340c7895a83501df68eb403496cec74bc363157 Mon Sep 17 00:00:00 2001
From: Anjali Singhai <anjali.singhai@intel.com>
Date: Wed, 6 Jan 2016 11:49:28 -0800
Subject: i40e: Enable Geneve offload for FW API ver > 1.4 for XL710/X710
 devices

This patch makes sure we check the GENEVE offload capable flag before
we attempt offload.
It also enables the Capability for XL710/X710 devices with FW API
version higher than 1.4

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 6e12626..99d7a5c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5354,7 +5354,8 @@ int i40e_open(struct net_device *netdev)
 	vxlan_get_rx_port(netdev);
 #endif
 #ifdef CONFIG_I40E_GENEVE
-	geneve_get_rx_port(netdev);
+	if (pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)
+		geneve_get_rx_port(netdev);
 #endif
 
 	return 0;
@@ -8458,7 +8459,13 @@ static int i40e_sw_init(struct i40e_pf *pf)
 			     I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE |
 			     I40E_FLAG_100M_SGMII_CAPABLE |
 			     I40E_FLAG_GENEVE_OFFLOAD_CAPABLE;
+	} else if ((pf->hw.aq.api_maj_ver > 1) ||
+		   ((pf->hw.aq.api_maj_ver == 1) &&
+		    (pf->hw.aq.api_min_ver > 4))) {
+		/* Supported in FW API version higher than 1.4 */
+		pf->flags |= I40E_FLAG_GENEVE_OFFLOAD_CAPABLE;
 	}
+
 	pf->eeprom_version = 0xDEAD;
 	pf->lan_veb = I40E_NO_VEB;
 	pf->lan_vsi = I40E_NO_VSI;
@@ -8672,6 +8679,9 @@ static void i40e_add_geneve_port(struct net_device *netdev,
 	u8 next_idx;
 	u8 idx;
 
+	if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE))
+		return;
+
 	if (sa_family == AF_INET6)
 		return;
 
@@ -8719,6 +8729,9 @@ static void i40e_del_geneve_port(struct net_device *netdev,
 	if (sa_family == AF_INET6)
 		return;
 
+	if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE))
+		return;
+
 	idx = i40e_get_udp_port_idx(pf, port);
 
 	/* Check if port already exists */
-- 
cgit v0.10.2


From 72b748698076b4609186e9554a5a890cd435b172 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Fri, 8 Jan 2016 17:50:21 -0800
Subject: i40e: add priv flag for automatic rule eviction

The X722 can support automatic rule eviction for automatically added
flow director rules.  Feature is (should be) disabled by default.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 65cf344..05af33e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -104,6 +104,7 @@
 #define I40E_PRIV_FLAGS_FD_ATR		BIT(2)
 #define I40E_PRIV_FLAGS_VEB_STATS	BIT(3)
 #define I40E_PRIV_FLAGS_PS		BIT(4)
+#define I40E_PRIV_FLAGS_HW_ATR_EVICT	BIT(5)
 
 #define I40E_NVM_VERSION_LO_SHIFT  0
 #define I40E_NVM_VERSION_LO_MASK   (0xff << I40E_NVM_VERSION_LO_SHIFT)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 9dcdf34..89ad2f7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -235,6 +235,7 @@ static const char i40e_priv_flags_strings[][ETH_GSTRING_LEN] = {
 	"flow-director-atr",
 	"veb-stats",
 	"packet-split",
+	"hw-atr-eviction",
 };
 
 #define I40E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40e_priv_flags_strings)
@@ -2731,6 +2732,8 @@ static u32 i40e_get_priv_flags(struct net_device *dev)
 		I40E_PRIV_FLAGS_VEB_STATS : 0;
 	ret_flags |= pf->flags & I40E_FLAG_RX_PS_ENABLED ?
 		I40E_PRIV_FLAGS_PS : 0;
+	ret_flags |= pf->auto_disable_flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE ?
+		0 : I40E_PRIV_FLAGS_HW_ATR_EVICT;
 
 	return ret_flags;
 }
@@ -2787,6 +2790,12 @@ static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
 	else
 		pf->flags &= ~I40E_FLAG_VEB_STATS_ENABLED;
 
+	if ((flags & I40E_PRIV_FLAGS_HW_ATR_EVICT) &&
+	    (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE))
+		pf->auto_disable_flags &= ~I40E_FLAG_HW_ATR_EVICT_CAPABLE;
+	else
+		pf->auto_disable_flags |= I40E_FLAG_HW_ATR_EVICT_CAPABLE;
+
 	/* if needed, issue reset to cause things to take effect */
 	if (reset_required)
 		i40e_do_reset(pf, BIT(__I40E_PF_RESET_REQUESTED));
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 99d7a5c..f6d61cc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8464,6 +8464,9 @@ static int i40e_sw_init(struct i40e_pf *pf)
 		    (pf->hw.aq.api_min_ver > 4))) {
 		/* Supported in FW API version higher than 1.4 */
 		pf->flags |= I40E_FLAG_GENEVE_OFFLOAD_CAPABLE;
+		pf->auto_disable_flags = I40E_FLAG_HW_ATR_EVICT_CAPABLE;
+	} else {
+		pf->auto_disable_flags = I40E_FLAG_HW_ATR_EVICT_CAPABLE;
 	}
 
 	pf->eeprom_version = 0xDEAD;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 15130b5d..6d1dd60 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2047,7 +2047,8 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	/* Due to lack of space, no more new filters can be programmed */
 	if (th->syn && (pf->auto_disable_flags & I40E_FLAG_FD_ATR_ENABLED))
 		return;
-	if (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE) {
+	if ((pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE) &&
+	    (!(pf->auto_disable_flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE))) {
 		/* HW ATR eviction will take care of removing filters on FIN
 		 * and RST packets.
 		 */
@@ -2109,7 +2110,8 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 			I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) &
 			I40E_TXD_FLTR_QW1_CNTINDEX_MASK;
 
-	if (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE)
+	if ((pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE) &&
+	    (!(pf->auto_disable_flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE)))
 		dtype_cmd |= I40E_TXD_FLTR_QW1_ATR_MASK;
 
 	fdir_desc->qindex_flex_ptype_vsi = cpu_to_le32(flex_ptype);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index dd2da35..b59a021 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -1098,6 +1098,10 @@ enum i40e_filter_program_desc_pcmd {
 					 I40E_TXD_FLTR_QW1_CMD_SHIFT)
 #define I40E_TXD_FLTR_QW1_ATR_MASK	BIT_ULL(I40E_TXD_FLTR_QW1_ATR_SHIFT)
 
+#define I40E_TXD_FLTR_QW1_ATR_SHIFT	(0xEULL + \
+					 I40E_TXD_FLTR_QW1_CMD_SHIFT)
+#define I40E_TXD_FLTR_QW1_ATR_MASK	BIT_ULL(I40E_TXD_FLTR_QW1_ATR_SHIFT)
+
 #define I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT 20
 #define I40E_TXD_FLTR_QW1_CNTINDEX_MASK	(0x1FFUL << \
 					 I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT)
-- 
cgit v0.10.2


From ba94272d08a7e7e716f17e491dbc9068f88eb241 Mon Sep 17 00:00:00 2001
From: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Date: Tue, 12 Jan 2016 19:32:31 -0800
Subject: i40e: use eth_platform_get_mac_address()

This commit converts commit b499ffb0a22c ("i40e: Look up MAC address in
Open Firmware or IDPROM") to use eth_platform_get_mac_address()
added by commit c7f5d105495a ("net: Add eth_platform_get_mac_address()
helper.")

Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index f6d61cc..04417e6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -28,11 +28,6 @@
 #include <linux/of_net.h>
 #include <linux/pci.h>
 
-#ifdef CONFIG_SPARC
-#include <asm/idprom.h>
-#include <asm/prom.h>
-#endif
-
 /* Local includes */
 #include "i40e.h"
 #include "i40e_diag.h"
@@ -10618,21 +10613,9 @@ static void i40e_print_features(struct i40e_pf *pf)
  **/
 static void i40e_get_platform_mac_addr(struct pci_dev *pdev, struct i40e_pf *pf)
 {
-	struct device_node *dp = pci_device_to_OF_node(pdev);
-	const unsigned char *addr;
-	u8 *mac_addr = pf->hw.mac.addr;
-
 	pf->flags &= ~I40E_FLAG_PF_MAC;
-	addr = of_get_mac_address(dp);
-	if (addr) {
-		ether_addr_copy(mac_addr, addr);
+	if (!eth_platform_get_mac_address(&pdev->dev, pf->hw.mac.addr))
 		pf->flags |= I40E_FLAG_PF_MAC;
-#ifdef CONFIG_SPARC
-	} else {
-		ether_addr_copy(mac_addr, idprom->id_ethaddr);
-		pf->flags |= I40E_FLAG_PF_MAC;
-#endif /* CONFIG_SPARC */
-	}
 }
 
 /**
-- 
cgit v0.10.2


From 7009212b157d3558bacb23bb037bdba75700e3fd Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Sun, 14 Feb 2016 14:10:39 +0000
Subject: macvlan: convert to use IFF_NO_QUEUE

Use IFF_NO_QUEUE to indicate that a device can run without a qdisc.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 94e6888..426a2cc 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1069,7 +1069,7 @@ EXPORT_SYMBOL_GPL(macvlan_common_setup);
 static void macvlan_setup(struct net_device *dev)
 {
 	macvlan_common_setup(dev);
-	dev->tx_queue_len	= 0;
+	dev->priv_flags |= IFF_NO_QUEUE;
 }
 
 static int macvlan_port_create(struct net_device *dev)
-- 
cgit v0.10.2


From b4eee84fcd8ecae88a7d6ab437f88284f9693d9c Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:48:08 -0600
Subject: amd-xgbe: Disable VLAN filtering when in promiscuous mode

When the hardware is placed in promiscuous mode it will still perform
VLAN filtering and therefore may not pass all packets to the driver.

Disable all VLAN filtering when entering promiscuous mode and restore
VLAN filtering upon exit from promiscuous mode. In order to avoid adding
forward declarations, move the VLAN related functions earlier in the
file.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index f6a7161..931e374 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -702,6 +702,113 @@ static int xgbe_set_xgmii_speed(struct xgbe_prv_data *pdata)
 	return 0;
 }
 
+static int xgbe_enable_rx_vlan_stripping(struct xgbe_prv_data *pdata)
+{
+	/* Put the VLAN tag in the Rx descriptor */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLRXS, 1);
+
+	/* Don't check the VLAN type */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, DOVLTC, 1);
+
+	/* Check only C-TAG (0x8100) packets */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ERSVLM, 0);
+
+	/* Don't consider an S-TAG (0x88A8) packet as a VLAN packet */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ESVL, 0);
+
+	/* Enable VLAN tag stripping */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLS, 0x3);
+
+	return 0;
+}
+
+static int xgbe_disable_rx_vlan_stripping(struct xgbe_prv_data *pdata)
+{
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLS, 0);
+
+	return 0;
+}
+
+static int xgbe_enable_rx_vlan_filtering(struct xgbe_prv_data *pdata)
+{
+	/* Enable VLAN filtering */
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, VTFE, 1);
+
+	/* Enable VLAN Hash Table filtering */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, VTHM, 1);
+
+	/* Disable VLAN tag inverse matching */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, VTIM, 0);
+
+	/* Only filter on the lower 12-bits of the VLAN tag */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ETV, 1);
+
+	/* In order for the VLAN Hash Table filtering to be effective,
+	 * the VLAN tag identifier in the VLAN Tag Register must not
+	 * be zero.  Set the VLAN tag identifier to "1" to enable the
+	 * VLAN Hash Table filtering.  This implies that a VLAN tag of
+	 * 1 will always pass filtering.
+	 */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, VL, 1);
+
+	return 0;
+}
+
+static int xgbe_disable_rx_vlan_filtering(struct xgbe_prv_data *pdata)
+{
+	/* Disable VLAN filtering */
+	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, VTFE, 0);
+
+	return 0;
+}
+
+static u32 xgbe_vid_crc32_le(__le16 vid_le)
+{
+	u32 poly = 0xedb88320;	/* CRCPOLY_LE */
+	u32 crc = ~0;
+	u32 temp = 0;
+	unsigned char *data = (unsigned char *)&vid_le;
+	unsigned char data_byte = 0;
+	int i, bits;
+
+	bits = get_bitmask_order(VLAN_VID_MASK);
+	for (i = 0; i < bits; i++) {
+		if ((i % 8) == 0)
+			data_byte = data[i / 8];
+
+		temp = ((crc & 1) ^ data_byte) & 1;
+		crc >>= 1;
+		data_byte >>= 1;
+
+		if (temp)
+			crc ^= poly;
+	}
+
+	return crc;
+}
+
+static int xgbe_update_vlan_hash_table(struct xgbe_prv_data *pdata)
+{
+	u32 crc;
+	u16 vid;
+	__le16 vid_le;
+	u16 vlan_hash_table = 0;
+
+	/* Generate the VLAN Hash Table value */
+	for_each_set_bit(vid, pdata->active_vlans, VLAN_N_VID) {
+		/* Get the CRC32 value of the VLAN ID */
+		vid_le = cpu_to_le16(vid);
+		crc = bitrev32(~xgbe_vid_crc32_le(vid_le)) >> 28;
+
+		vlan_hash_table |= (1 << crc);
+	}
+
+	/* Set the VLAN Hash Table filtering register */
+	XGMAC_IOWRITE_BITS(pdata, MAC_VLANHTR, VLHT, vlan_hash_table);
+
+	return 0;
+}
+
 static int xgbe_set_promiscuous_mode(struct xgbe_prv_data *pdata,
 				     unsigned int enable)
 {
@@ -714,6 +821,14 @@ static int xgbe_set_promiscuous_mode(struct xgbe_prv_data *pdata,
 		  enable ? "entering" : "leaving");
 	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, PR, val);
 
+	/* Hardware will still perform VLAN filtering in promiscuous mode */
+	if (enable) {
+		xgbe_disable_rx_vlan_filtering(pdata);
+	} else {
+		if (pdata->netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER)
+			xgbe_enable_rx_vlan_filtering(pdata);
+	}
+
 	return 0;
 }
 
@@ -944,116 +1059,6 @@ static int xgbe_enable_rx_csum(struct xgbe_prv_data *pdata)
 	return 0;
 }
 
-static int xgbe_enable_rx_vlan_stripping(struct xgbe_prv_data *pdata)
-{
-	/* Put the VLAN tag in the Rx descriptor */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLRXS, 1);
-
-	/* Don't check the VLAN type */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, DOVLTC, 1);
-
-	/* Check only C-TAG (0x8100) packets */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ERSVLM, 0);
-
-	/* Don't consider an S-TAG (0x88A8) packet as a VLAN packet */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ESVL, 0);
-
-	/* Enable VLAN tag stripping */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLS, 0x3);
-
-	return 0;
-}
-
-static int xgbe_disable_rx_vlan_stripping(struct xgbe_prv_data *pdata)
-{
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, EVLS, 0);
-
-	return 0;
-}
-
-static int xgbe_enable_rx_vlan_filtering(struct xgbe_prv_data *pdata)
-{
-	/* Enable VLAN filtering */
-	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, VTFE, 1);
-
-	/* Enable VLAN Hash Table filtering */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, VTHM, 1);
-
-	/* Disable VLAN tag inverse matching */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, VTIM, 0);
-
-	/* Only filter on the lower 12-bits of the VLAN tag */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, ETV, 1);
-
-	/* In order for the VLAN Hash Table filtering to be effective,
-	 * the VLAN tag identifier in the VLAN Tag Register must not
-	 * be zero.  Set the VLAN tag identifier to "1" to enable the
-	 * VLAN Hash Table filtering.  This implies that a VLAN tag of
-	 * 1 will always pass filtering.
-	 */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANTR, VL, 1);
-
-	return 0;
-}
-
-static int xgbe_disable_rx_vlan_filtering(struct xgbe_prv_data *pdata)
-{
-	/* Disable VLAN filtering */
-	XGMAC_IOWRITE_BITS(pdata, MAC_PFR, VTFE, 0);
-
-	return 0;
-}
-
-#ifndef CRCPOLY_LE
-#define CRCPOLY_LE 0xedb88320
-#endif
-static u32 xgbe_vid_crc32_le(__le16 vid_le)
-{
-	u32 poly = CRCPOLY_LE;
-	u32 crc = ~0;
-	u32 temp = 0;
-	unsigned char *data = (unsigned char *)&vid_le;
-	unsigned char data_byte = 0;
-	int i, bits;
-
-	bits = get_bitmask_order(VLAN_VID_MASK);
-	for (i = 0; i < bits; i++) {
-		if ((i % 8) == 0)
-			data_byte = data[i / 8];
-
-		temp = ((crc & 1) ^ data_byte) & 1;
-		crc >>= 1;
-		data_byte >>= 1;
-
-		if (temp)
-			crc ^= poly;
-	}
-
-	return crc;
-}
-
-static int xgbe_update_vlan_hash_table(struct xgbe_prv_data *pdata)
-{
-	u32 crc;
-	u16 vid;
-	__le16 vid_le;
-	u16 vlan_hash_table = 0;
-
-	/* Generate the VLAN Hash Table value */
-	for_each_set_bit(vid, pdata->active_vlans, VLAN_N_VID) {
-		/* Get the CRC32 value of the VLAN ID */
-		vid_le = cpu_to_le16(vid);
-		crc = bitrev32(~xgbe_vid_crc32_le(vid_le)) >> 28;
-
-		vlan_hash_table |= (1 << crc);
-	}
-
-	/* Set the VLAN Hash Table filtering register */
-	XGMAC_IOWRITE_BITS(pdata, MAC_VLANHTR, VLHT, vlan_hash_table);
-
-	return 0;
-}
-
 static void xgbe_tx_desc_reset(struct xgbe_ring_data *rdata)
 {
 	struct xgbe_ring_desc *rdesc = rdata->rdesc;
-- 
cgit v0.10.2


From 491aefb38a31bb163e1e4393ba42abf68ce4c6eb Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:48:19 -0600
Subject: amd-xgbe: Change from napi_complete to napi_complete_done

Change from using napi_complete to napi_complete_done to allow for the
use of gro_flush_timeout in tuning network processing.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index cfd3f7e..5a4d9ca 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -2068,7 +2068,7 @@ static int xgbe_one_poll(struct napi_struct *napi, int budget)
 	/* If we processed everything, we are done */
 	if (processed < budget) {
 		/* Turn off polling */
-		napi_complete(napi);
+		napi_complete_done(napi, processed);
 
 		/* Enable Tx and Rx interrupts */
 		enable_irq(channel->dma_irq);
@@ -2110,7 +2110,7 @@ static int xgbe_all_poll(struct napi_struct *napi, int budget)
 	/* If we processed everything, we are done */
 	if (processed < budget) {
 		/* Turn off polling */
-		napi_complete(napi);
+		napi_complete_done(napi, processed);
 
 		/* Enable Tx and Rx interrupts */
 		xgbe_enable_rx_tx_ints(pdata);
-- 
cgit v0.10.2


From 793494228fdf15acb252063f9f6dcccd3d1d497e Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:48:29 -0600
Subject: amd-xgbe: Use __napi_schedule_irqoff

Change from calling __napi_schedule to __napi_schedule_irqoff when running
in interrupt context or when called by netpoll with interrupts already
disabled. The Tx timer function will continue to use __napi_schedule.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 5a4d9ca..80ef404 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -356,7 +356,7 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 				xgbe_disable_rx_tx_ints(pdata);
 
 				/* Turn on polling */
-				__napi_schedule(&pdata->napi);
+				__napi_schedule_irqoff(&pdata->napi);
 			}
 		}
 
@@ -409,7 +409,7 @@ static irqreturn_t xgbe_dma_isr(int irq, void *data)
 		disable_irq_nosync(channel->dma_irq);
 
 		/* Turn on polling */
-		__napi_schedule(&channel->napi);
+		__napi_schedule_irqoff(&channel->napi);
 	}
 
 	return IRQ_HANDLED;
-- 
cgit v0.10.2


From 757e6aa34ca9529786877d3045684e7c9f686075 Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:48:39 -0600
Subject: amd-xgbe: Verify forced speed matches the active speedset

When using ethtool to set the speed for the device, verify that the
specified speed is valid within the active speedset.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
index 6040293..11d9f0c 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -318,8 +318,20 @@ static int xgbe_set_settings(struct net_device *netdev,
 	if (cmd->autoneg == AUTONEG_DISABLE) {
 		switch (speed) {
 		case SPEED_10000:
+			break;
 		case SPEED_2500:
+			if (pdata->speed_set != XGBE_SPEEDSET_2500_10000) {
+				netdev_err(netdev, "unsupported speed %u\n",
+					   speed);
+				return -EINVAL;
+			}
+			break;
 		case SPEED_1000:
+			if (pdata->speed_set != XGBE_SPEEDSET_1000_10000) {
+				netdev_err(netdev, "unsupported speed %u\n",
+					   speed);
+				return -EINVAL;
+			}
 			break;
 		default:
 			netdev_err(netdev, "unsupported speed %u\n", speed);
-- 
cgit v0.10.2


From 8dba2a2a88397dec6bdcae8bf7ceeefd62fd39fc Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:48:48 -0600
Subject: amd-xgbe: Enable/disable PFC per traffic class

Currently the PFC flow control is enabled on all traffic classes if
one or more traffic classes request it. The PFC enable setting of the
traffic class should be used to determine whether to enable or disable
flow control for the traffic class.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index 931e374..6b98a99 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -518,13 +518,45 @@ static int xgbe_disable_tx_flow_control(struct xgbe_prv_data *pdata)
 
 static int xgbe_enable_tx_flow_control(struct xgbe_prv_data *pdata)
 {
+	struct ieee_pfc *pfc = pdata->pfc;
+	struct ieee_ets *ets = pdata->ets;
 	unsigned int max_q_count, q_count;
 	unsigned int reg, reg_val;
 	unsigned int i;
 
 	/* Set MTL flow control */
-	for (i = 0; i < pdata->rx_q_count; i++)
-		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, EHFC, 1);
+	for (i = 0; i < pdata->rx_q_count; i++) {
+		unsigned int ehfc = 0;
+
+		if (pfc && ets) {
+			unsigned int prio;
+
+			for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) {
+				unsigned int tc;
+
+				/* Does this queue handle the priority? */
+				if (pdata->prio2q_map[prio] != i)
+					continue;
+
+				/* Get the Traffic Class for this priority */
+				tc = ets->prio_tc[prio];
+
+				/* Check if flow control should be enabled */
+				if (pfc->pfc_en & (1 << tc)) {
+					ehfc = 1;
+					break;
+				}
+			}
+		} else {
+			ehfc = 1;
+		}
+
+		XGMAC_MTL_IOWRITE_BITS(pdata, i, MTL_Q_RQOMR, EHFC, ehfc);
+
+		netif_dbg(pdata, drv, pdata->netdev,
+			  "flow control %s for RXq%u\n",
+			  ehfc ? "enabled" : "disabled", i);
+	}
 
 	/* Set MAC flow control */
 	max_q_count = XGMAC_MAX_FLOW_CONTROL_QUEUES;
-- 
cgit v0.10.2


From 6a49ee4e1b9eb8747ff79c5438a42c0726aaa03b Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:48:57 -0600
Subject: amd-xgbe: Fix the mapping of priorities to traffic classes

The driver is checking the pfc_en field of the ieee_pfc structure to
determine whether to associate a priority with a traffic class. This is
incorrect since the pfc_en field is for determining if PFC is enabled
for a traffic class.

The association of priority to traffic class does not depend on whether
the traffic class is enabled for PFC, so remove that check. Also, the
mapping of priorities to traffic classes should be done when configuring
the traffic classes and not the PFC support so move the priority to traffic
class association from xgbe_config_dcb_pfc to xgbe_config_dcb_tc.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index 6b98a99..67d234e 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -1329,7 +1329,8 @@ static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata)
 {
 	struct ieee_ets *ets = pdata->ets;
 	unsigned int total_weight, min_weight, weight;
-	unsigned int i;
+	unsigned int mask, reg, reg_val;
+	unsigned int i, prio;
 
 	if (!ets)
 		return;
@@ -1346,6 +1347,25 @@ static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata)
 		min_weight = 1;
 
 	for (i = 0; i < pdata->hw_feat.tc_cnt; i++) {
+		/* Map the priorities to the traffic class */
+		mask = 0;
+		for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) {
+			if (ets->prio_tc[prio] == i)
+				mask |= (1 << prio);
+		}
+		mask &= 0xff;
+
+		netif_dbg(pdata, drv, pdata->netdev, "TC%u PRIO mask=%#x\n",
+			  i, mask);
+		reg = MTL_TCPM0R + (MTL_TCPM_INC * (i / MTL_TCPM_TC_PER_REG));
+		reg_val = XGMAC_IOREAD(pdata, reg);
+
+		reg_val &= ~(0xff << ((i % MTL_TCPM_TC_PER_REG) << 3));
+		reg_val |= (mask << ((i % MTL_TCPM_TC_PER_REG) << 3));
+
+		XGMAC_IOWRITE(pdata, reg, reg_val);
+
+		/* Set the traffic class algorithm */
 		switch (ets->tc_tsa[i]) {
 		case IEEE_8021QAZ_TSA_STRICT:
 			netif_dbg(pdata, drv, pdata->netdev,
@@ -1370,34 +1390,6 @@ static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata)
 
 static void xgbe_config_dcb_pfc(struct xgbe_prv_data *pdata)
 {
-	struct ieee_pfc *pfc = pdata->pfc;
-	struct ieee_ets *ets = pdata->ets;
-	unsigned int mask, reg, reg_val;
-	unsigned int tc, prio;
-
-	if (!pfc || !ets)
-		return;
-
-	for (tc = 0; tc < pdata->hw_feat.tc_cnt; tc++) {
-		mask = 0;
-		for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) {
-			if ((pfc->pfc_en & (1 << prio)) &&
-			    (ets->prio_tc[prio] == tc))
-				mask |= (1 << prio);
-		}
-		mask &= 0xff;
-
-		netif_dbg(pdata, drv, pdata->netdev, "TC%u PFC mask=%#x\n",
-			  tc, mask);
-		reg = MTL_TCPM0R + (MTL_TCPM_INC * (tc / MTL_TCPM_TC_PER_REG));
-		reg_val = XGMAC_IOREAD(pdata, reg);
-
-		reg_val &= ~(0xff << ((tc % MTL_TCPM_TC_PER_REG) << 3));
-		reg_val |= (mask << ((tc % MTL_TCPM_TC_PER_REG) << 3));
-
-		XGMAC_IOWRITE(pdata, reg, reg_val);
-	}
-
 	xgbe_config_flow_control(pdata);
 }
 
-- 
cgit v0.10.2


From b3b715974bfe69f626d6a633b8c96590de1b7338 Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:49:08 -0600
Subject: amd-xgbe: Do traffic class setup when called through dcbnl

Currently the netdev traffic class setup is only performed when invoked
through the ndo_setup_tc interface. However, the same setup should be
performed when the dcbnl interface (ieee_setets) is invoked. Rework the
netdev traffic class setup to be invokable through either interface and
also provide the priority to traffic class mapping if available.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c b/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c
index a6b9899..895d356 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dcb.c
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -146,6 +146,7 @@ static int xgbe_dcb_ieee_setets(struct net_device *netdev,
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	unsigned int i, tc_ets, tc_ets_weight;
+	u8 max_tc = 0;
 
 	tc_ets = 0;
 	tc_ets_weight = 0;
@@ -157,12 +158,9 @@ static int xgbe_dcb_ieee_setets(struct net_device *netdev,
 		netif_dbg(pdata, drv, netdev, "PRIO%u: TC=%hhu\n", i,
 			  ets->prio_tc[i]);
 
-		if ((ets->tc_tx_bw[i] || ets->tc_tsa[i]) &&
-		    (i >= pdata->hw_feat.tc_cnt))
-				return -EINVAL;
-
-		if (ets->prio_tc[i] >= pdata->hw_feat.tc_cnt)
-			return -EINVAL;
+		max_tc = max_t(u8, max_tc, ets->prio_tc[i]);
+		if ((ets->tc_tx_bw[i] || ets->tc_tsa[i]))
+			max_tc = max_t(u8, max_tc, i);
 
 		switch (ets->tc_tsa[i]) {
 		case IEEE_8021QAZ_TSA_STRICT:
@@ -171,15 +169,28 @@ static int xgbe_dcb_ieee_setets(struct net_device *netdev,
 			tc_ets = 1;
 			tc_ets_weight += ets->tc_tx_bw[i];
 			break;
-
 		default:
+			netif_err(pdata, drv, netdev,
+				  "unsupported TSA algorithm (%hhu)\n",
+				  ets->tc_tsa[i]);
 			return -EINVAL;
 		}
 	}
 
+	/* Check maximum traffic class requested */
+	if (max_tc >= pdata->hw_feat.tc_cnt) {
+		netif_err(pdata, drv, netdev,
+			  "exceeded number of supported traffic classes\n");
+		return -EINVAL;
+	}
+
 	/* Weights must add up to 100% */
-	if (tc_ets && (tc_ets_weight != 100))
+	if (tc_ets && (tc_ets_weight != 100)) {
+		netif_err(pdata, drv, netdev,
+			  "sum of ETS algorithm weights is not 100 (%u)\n",
+			  tc_ets_weight);
 		return -EINVAL;
+	}
 
 	if (!pdata->ets) {
 		pdata->ets = devm_kzalloc(pdata->dev, sizeof(*pdata->ets),
@@ -188,6 +199,7 @@ static int xgbe_dcb_ieee_setets(struct net_device *netdev,
 			return -ENOMEM;
 	}
 
+	pdata->num_tcs = max_tc + 1;
 	memcpy(pdata->ets, ets, sizeof(*pdata->ets));
 
 	pdata->hw_if.config_dcb_tc(pdata);
@@ -221,6 +233,13 @@ static int xgbe_dcb_ieee_setpfc(struct net_device *netdev,
 		  "cap=%hhu, en=%#hhx, mbc=%hhu, delay=%hhu\n",
 		  pfc->pfc_cap, pfc->pfc_en, pfc->mbc, pfc->delay);
 
+	/* Check PFC for supported number of traffic classes */
+	if (pfc->pfc_en & ~((1 << pdata->hw_feat.tc_cnt) - 1)) {
+		netif_err(pdata, drv, netdev,
+			  "PFC requested for unsupported traffic class\n");
+		return -EINVAL;
+	}
+
 	if (!pdata->pfc) {
 		pdata->pfc = devm_kzalloc(pdata->dev, sizeof(*pdata->pfc),
 					  GFP_KERNEL);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index 67d234e..43273c98 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -1325,6 +1325,36 @@ static int xgbe_config_tstamp(struct xgbe_prv_data *pdata,
 	return 0;
 }
 
+static void xgbe_config_tc(struct xgbe_prv_data *pdata)
+{
+	unsigned int offset, queue, prio;
+	u8 i;
+
+	netdev_reset_tc(pdata->netdev);
+	if (!pdata->num_tcs)
+		return;
+
+	netdev_set_num_tc(pdata->netdev, pdata->num_tcs);
+
+	for (i = 0, queue = 0, offset = 0; i < pdata->num_tcs; i++) {
+		while ((queue < pdata->tx_q_count) &&
+		       (pdata->q2tc_map[queue] == i))
+			queue++;
+
+		netif_dbg(pdata, drv, pdata->netdev, "TC%u using TXq%u-%u\n",
+			  i, offset, queue - 1);
+		netdev_set_tc_queue(pdata->netdev, i, queue - offset, offset);
+		offset = queue;
+	}
+
+	if (!pdata->ets)
+		return;
+
+	for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++)
+		netdev_set_prio_tc_map(pdata->netdev, prio,
+				       pdata->ets->prio_tc[prio]);
+}
+
 static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata)
 {
 	struct ieee_ets *ets = pdata->ets;
@@ -1386,6 +1416,8 @@ static void xgbe_config_dcb_tc(struct xgbe_prv_data *pdata)
 			break;
 		}
 	}
+
+	xgbe_config_tc(pdata);
 }
 
 static void xgbe_config_dcb_pfc(struct xgbe_prv_data *pdata)
@@ -2910,6 +2942,7 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->get_tx_tstamp = xgbe_get_tx_tstamp;
 
 	/* For Data Center Bridging config */
+	hw_if->config_tc = xgbe_config_tc;
 	hw_if->config_dcb_tc = xgbe_config_dcb_tc;
 	hw_if->config_dcb_pfc = xgbe_config_dcb_pfc;
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 80ef404..3360684 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -1630,32 +1630,18 @@ static int xgbe_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
 			 struct tc_to_netdev *tc_to_netdev)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
-	unsigned int offset, queue;
-	u8 i, tc;
+	u8 tc;
 
 	if (handle != TC_H_ROOT || tc_to_netdev->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	tc = tc_to_netdev->tc;
 
-	if (tc && (tc != pdata->hw_feat.tc_cnt))
+	if (tc > pdata->hw_feat.tc_cnt)
 		return -EINVAL;
 
-	if (tc) {
-		netdev_set_num_tc(netdev, tc);
-		for (i = 0, queue = 0, offset = 0; i < tc; i++) {
-			while ((queue < pdata->tx_q_count) &&
-			       (pdata->q2tc_map[queue] == i))
-				queue++;
-
-			netif_dbg(pdata, drv, netdev, "TC%u using TXq%u-%u\n",
-				  i, offset, queue - 1);
-			netdev_set_tc_queue(netdev, i, queue - offset, offset);
-			offset = queue;
-		}
-	} else {
-		netdev_reset_tc(netdev);
-	}
+	pdata->num_tcs = tc;
+	pdata->hw_if.config_tc(pdata);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h
index e234b99..ca28354 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe.h
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -673,6 +673,7 @@ struct xgbe_hw_if {
 	u64 (*get_tx_tstamp)(struct xgbe_prv_data *);
 
 	/* For Data Center Bridging config */
+	void (*config_tc)(struct xgbe_prv_data *);
 	void (*config_dcb_tc)(struct xgbe_prv_data *);
 	void (*config_dcb_pfc)(struct xgbe_prv_data *);
 
@@ -880,6 +881,7 @@ struct xgbe_prv_data {
 	struct ieee_pfc *pfc;
 	unsigned int q2tc_map[XGBE_MAX_QUEUES];
 	unsigned int prio2q_map[IEEE_8021QAZ_MAX_TCS];
+	u8 num_tcs;
 
 	/* Hardware features of the device */
 	struct xgbe_hw_features hw_feat;
-- 
cgit v0.10.2


From c3727d61ea73a1df60bacd0b416e01282a94240b Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:49:16 -0600
Subject: amd-xgbe: Check Rx queue fifos before stopping Rx DMA

Check to be sure that the Rx queue fifos are empty before stopping the
Rx DMA channels.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
index b6fa891..bbef959 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -768,12 +768,16 @@
 #define MTL_Q_TQDR			0x08
 #define MTL_Q_RQOMR			0x40
 #define MTL_Q_RQMPOCR			0x44
-#define MTL_Q_RQDR			0x4c
+#define MTL_Q_RQDR			0x48
 #define MTL_Q_RQFCR			0x50
 #define MTL_Q_IER			0x70
 #define MTL_Q_ISR			0x74
 
 /* MTL queue register entry bit positions and sizes */
+#define MTL_Q_RQDR_PRXQ_INDEX		16
+#define MTL_Q_RQDR_PRXQ_WIDTH		14
+#define MTL_Q_RQDR_RXQSTS_INDEX		4
+#define MTL_Q_RQDR_RXQSTS_WIDTH		2
 #define MTL_Q_RQFCR_RFA_INDEX		1
 #define MTL_Q_RQFCR_RFA_WIDTH		6
 #define MTL_Q_RQFCR_RFD_INDEX		17
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index 43273c98..b48c6ec 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -2656,6 +2656,32 @@ static void xgbe_disable_tx(struct xgbe_prv_data *pdata)
 	}
 }
 
+static void xgbe_prepare_rx_stop(struct xgbe_prv_data *pdata,
+				 unsigned int queue)
+{
+	unsigned int rx_status;
+	unsigned long rx_timeout;
+
+	/* The Rx engine cannot be stopped if it is actively processing
+	 * packets. Wait for the Rx queue to empty the Rx fifo.  Don't
+	 * wait forever though...
+	 */
+	rx_timeout = jiffies + (XGBE_DMA_STOP_TIMEOUT * HZ);
+	while (time_before(jiffies, rx_timeout)) {
+		rx_status = XGMAC_MTL_IOREAD(pdata, queue, MTL_Q_RQDR);
+		if ((XGMAC_GET_BITS(rx_status, MTL_Q_RQDR, PRXQ) == 0) &&
+		    (XGMAC_GET_BITS(rx_status, MTL_Q_RQDR, RXQSTS) == 0))
+			break;
+
+		usleep_range(500, 1000);
+	}
+
+	if (!time_before(jiffies, rx_timeout))
+		netdev_info(pdata->netdev,
+			    "timed out waiting for Rx queue %u to empty\n",
+			    queue);
+}
+
 static void xgbe_enable_rx(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_channel *channel;
@@ -2694,6 +2720,10 @@ static void xgbe_disable_rx(struct xgbe_prv_data *pdata)
 	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, ACS, 0);
 	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, RE, 0);
 
+	/* Prepare for Rx DMA channel stop */
+	for (i = 0; i < pdata->rx_q_count; i++)
+		xgbe_prepare_rx_stop(pdata, i);
+
 	/* Disable each Rx queue */
 	XGMAC_IOWRITE(pdata, MAC_RQC0R, 0);
 
-- 
cgit v0.10.2


From ced3fcae693b563b20ee8d2dba966760e6b771d4 Mon Sep 17 00:00:00 2001
From: "Lendacky, Thomas" <Thomas.Lendacky@amd.com>
Date: Wed, 17 Feb 2016 11:49:28 -0600
Subject: amd-xgbe: Mask auto-negotiation interrupts in ISR

Currently the auto-negotiation interrupt handling disables the irq
instead of masking off the interrupts. This was done because the phy
library was originally used to read and write the PCS registers, which
could not be performed in interrupt context. Now that the phy library is
no longer used to read and write the PCS registers the interrupts can be
masked off in the interrupt service routine eliminating the need to call
disable_irq/enable_irq. This also requires changing the protection mutex
to a spinlock.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index b48c6ec..1babcc1 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -1022,6 +1022,7 @@ static int xgbe_config_rx_mode(struct xgbe_prv_data *pdata)
 static int xgbe_read_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
 			      int mmd_reg)
 {
+	unsigned long flags;
 	unsigned int mmd_address;
 	int mmd_data;
 
@@ -1039,10 +1040,10 @@ static int xgbe_read_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
 	 * register offsets must therefore be adjusted by left shifting the
 	 * offset 2 bits and reading 32 bits of data.
 	 */
-	mutex_lock(&pdata->xpcs_mutex);
+	spin_lock_irqsave(&pdata->xpcs_lock, flags);
 	XPCS_IOWRITE(pdata, PCS_MMD_SELECT << 2, mmd_address >> 8);
 	mmd_data = XPCS_IOREAD(pdata, (mmd_address & 0xff) << 2);
-	mutex_unlock(&pdata->xpcs_mutex);
+	spin_unlock_irqrestore(&pdata->xpcs_lock, flags);
 
 	return mmd_data;
 }
@@ -1051,6 +1052,7 @@ static void xgbe_write_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
 				int mmd_reg, int mmd_data)
 {
 	unsigned int mmd_address;
+	unsigned long flags;
 
 	if (mmd_reg & MII_ADDR_C45)
 		mmd_address = mmd_reg & ~MII_ADDR_C45;
@@ -1066,10 +1068,10 @@ static void xgbe_write_mmd_regs(struct xgbe_prv_data *pdata, int prtad,
 	 * register offsets must therefore be adjusted by left shifting the
 	 * offset 2 bits and reading 32 bits of data.
 	 */
-	mutex_lock(&pdata->xpcs_mutex);
+	spin_lock_irqsave(&pdata->xpcs_lock, flags);
 	XPCS_IOWRITE(pdata, PCS_MMD_SELECT << 2, mmd_address >> 8);
 	XPCS_IOWRITE(pdata, (mmd_address & 0xff) << 2, mmd_data);
-	mutex_unlock(&pdata->xpcs_mutex);
+	spin_unlock_irqrestore(&pdata->xpcs_lock, flags);
 }
 
 static int xgbe_tx_complete(struct xgbe_ring_desc *rdesc)
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-main.c b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
index 618d952..3eee320 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-main.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -363,7 +363,7 @@ static int xgbe_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, netdev);
 
 	spin_lock_init(&pdata->lock);
-	mutex_init(&pdata->xpcs_mutex);
+	spin_lock_init(&pdata->xpcs_lock);
 	mutex_init(&pdata->rss_mutex);
 	spin_lock_init(&pdata->tstamp_lock);
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
index 4460580..84c5d29 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c
@@ -6,7 +6,7 @@
  *
  * License 1: GPLv2
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  *
  * This file is free software; you may copy, redistribute and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -56,7 +56,7 @@
  *
  * License 2: Modified BSD
  *
- * Copyright (c) 2014 Advanced Micro Devices, Inc.
+ * Copyright (c) 2014-2016 Advanced Micro Devices, Inc.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -626,10 +626,22 @@ static irqreturn_t xgbe_an_isr(int irq, void *data)
 
 	netif_dbg(pdata, intr, pdata->netdev, "AN interrupt received\n");
 
-	/* Interrupt reason must be read and cleared outside of IRQ context */
-	disable_irq_nosync(pdata->an_irq);
+	/* Disable AN interrupts */
+	XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INTMASK, 0);
+
+	/* Save the interrupt(s) that fired */
+	pdata->an_int = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_INT);
 
-	queue_work(pdata->an_workqueue, &pdata->an_irq_work);
+	if (pdata->an_int) {
+		/* Clear the interrupt(s) that fired and process them */
+		XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, ~pdata->an_int);
+
+		queue_work(pdata->an_workqueue, &pdata->an_irq_work);
+	} else {
+		/* Enable AN interrupts */
+		XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INTMASK,
+			    XGBE_AN_INT_MASK);
+	}
 
 	return IRQ_HANDLED;
 }
@@ -673,34 +685,26 @@ static void xgbe_an_state_machine(struct work_struct *work)
 						   struct xgbe_prv_data,
 						   an_work);
 	enum xgbe_an cur_state = pdata->an_state;
-	unsigned int int_reg, int_mask;
 
 	mutex_lock(&pdata->an_mutex);
 
-	/* Read the interrupt */
-	int_reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_AN_INT);
-	if (!int_reg)
+	if (!pdata->an_int)
 		goto out;
 
 next_int:
-	if (int_reg & XGBE_AN_PG_RCV) {
+	if (pdata->an_int & XGBE_AN_PG_RCV) {
 		pdata->an_state = XGBE_AN_PAGE_RECEIVED;
-		int_mask = XGBE_AN_PG_RCV;
-	} else if (int_reg & XGBE_AN_INC_LINK) {
+		pdata->an_int &= ~XGBE_AN_PG_RCV;
+	} else if (pdata->an_int & XGBE_AN_INC_LINK) {
 		pdata->an_state = XGBE_AN_INCOMPAT_LINK;
-		int_mask = XGBE_AN_INC_LINK;
-	} else if (int_reg & XGBE_AN_INT_CMPLT) {
+		pdata->an_int &= ~XGBE_AN_INC_LINK;
+	} else if (pdata->an_int & XGBE_AN_INT_CMPLT) {
 		pdata->an_state = XGBE_AN_COMPLETE;
-		int_mask = XGBE_AN_INT_CMPLT;
+		pdata->an_int &= ~XGBE_AN_INT_CMPLT;
 	} else {
 		pdata->an_state = XGBE_AN_ERROR;
-		int_mask = 0;
 	}
 
-	/* Clear the interrupt to be processed */
-	int_reg &= ~int_mask;
-	XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, int_reg);
-
 	pdata->an_result = pdata->an_state;
 
 again:
@@ -740,14 +744,14 @@ again:
 	}
 
 	if (pdata->an_state == XGBE_AN_NO_LINK) {
-		int_reg = 0;
+		pdata->an_int = 0;
 		XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, 0);
 	} else if (pdata->an_state == XGBE_AN_ERROR) {
 		netdev_err(pdata->netdev,
 			   "error during auto-negotiation, state=%u\n",
 			   cur_state);
 
-		int_reg = 0;
+		pdata->an_int = 0;
 		XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INT, 0);
 	}
 
@@ -765,11 +769,12 @@ again:
 	if (cur_state != pdata->an_state)
 		goto again;
 
-	if (int_reg)
+	if (pdata->an_int)
 		goto next_int;
 
 out:
-	enable_irq(pdata->an_irq);
+	/* Enable AN interrupts on the way out */
+	XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_AN_INTMASK, XGBE_AN_INT_MASK);
 
 	mutex_unlock(&pdata->an_mutex);
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h
index ca28354..98d9d63 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe.h
@@ -774,8 +774,8 @@ struct xgbe_prv_data {
 	/* Overall device lock */
 	spinlock_t lock;
 
-	/* XPCS indirect addressing mutex */
-	struct mutex xpcs_mutex;
+	/* XPCS indirect addressing lock */
+	spinlock_t xpcs_lock;
 
 	/* RSS addressing mutex */
 	struct mutex rss_mutex;
@@ -927,6 +927,7 @@ struct xgbe_prv_data {
 	u32 serdes_dfe_tap_ena[XGBE_SPEEDS];
 
 	/* Auto-negotiation state machine support */
+	unsigned int an_int;
 	struct mutex an_mutex;
 	enum xgbe_an an_result;
 	enum xgbe_an an_state;
-- 
cgit v0.10.2


From bb64b035f8a8474c79c3554b921c136558bc18c5 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 14 Feb 2016 22:56:03 +0300
Subject: ravb: kill useless *switch* defaults

The  driver has the *default* case doing nothing in the *switch* statement
with an integer expression -- remove it.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index c936682..331c596 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -92,8 +92,6 @@ static void ravb_set_rate(struct net_device *ndev)
 	case 1000:		/* 1000BASE */
 		ravb_write(ndev, GECMR_SPEED_1000, GECMR);
 		break;
-	default:
-		break;
 	}
 }
 
-- 
cgit v0.10.2


From 82c133093c56c2b11b7a4cf9c18f494b7b7c0191 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 14 Feb 2016 22:56:33 +0300
Subject: sh_eth: kill useless *switch* defaults

The driver often has the *default* cases doing nothing in the *switch*
statements with  the integer expressions -- remove them.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 0a150b2..a276733 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -500,8 +500,6 @@ static void sh_eth_set_rate_gether(struct net_device *ndev)
 	case 1000: /* 1000BASE */
 		sh_eth_write(ndev, GECMR_1000, GECMR);
 		break;
-	default:
-		break;
 	}
 }
 
@@ -592,8 +590,6 @@ static void sh_eth_set_rate_r8a777x(struct net_device *ndev)
 	case 100:/* 100BASE */
 		sh_eth_modify(ndev, ECMR, ECMR_ELB, ECMR_ELB);
 		break;
-	default:
-		break;
 	}
 }
 
@@ -658,8 +654,6 @@ static void sh_eth_set_rate_sh7724(struct net_device *ndev)
 	case 100:/* 100BASE */
 		sh_eth_modify(ndev, ECMR, ECMR_RTM, ECMR_RTM);
 		break;
-	default:
-		break;
 	}
 }
 
@@ -698,8 +692,6 @@ static void sh_eth_set_rate_sh7757(struct net_device *ndev)
 	case 100:/* 100BASE */
 		sh_eth_write(ndev, 1, RTRATE);
 		break;
-	default:
-		break;
 	}
 }
 
@@ -767,8 +759,6 @@ static void sh_eth_set_rate_giga(struct net_device *ndev)
 	case 1000: /* 1000BASE */
 		sh_eth_write(ndev, 0x00000020, GECMR);
 		break;
-	default:
-		break;
 	}
 }
 
@@ -2917,8 +2907,6 @@ static const u16 *sh_eth_get_register_offset(int register_type)
 	case SH_ETH_REG_FAST_SH3_SH2:
 		reg_offset = sh_eth_offset_fast_sh3_sh2;
 		break;
-	default:
-		break;
 	}
 
 	return reg_offset;
-- 
cgit v0.10.2


From bd4508e8507cbeee3c0fd96a1f96327b07c490b9 Mon Sep 17 00:00:00 2001
From: "Rosen, Rami" <rami.rosen@intel.com>
Date: Mon, 15 Feb 2016 02:39:43 +0200
Subject: core: remove unneded headers for net cgroup controllers.

commit 3ed80a6 (cgroup: drop module support) made including
module.h redundant in the net cgroup controllers,
netclassid_cgroup.c and netprio_cgroup.c. This patch
removes them.

Signed-off-by: Rami Rosen <rami.rosen@intel.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c
index 0260c84..11fce17 100644
--- a/net/core/netclassid_cgroup.c
+++ b/net/core/netclassid_cgroup.c
@@ -9,7 +9,6 @@
  * Authors:	Thomas Graf <tgraf@suug.ch>
  */
 
-#include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/cgroup.h>
 #include <linux/fdtable.h>
diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c
index f1efbc3..2ec86fc 100644
--- a/net/core/netprio_cgroup.c
+++ b/net/core/netprio_cgroup.c
@@ -11,7 +11,6 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/types.h>
 #include <linux/string.h>
-- 
cgit v0.10.2


From f46c7011b062726b798fa3326b5b0c2711ead2a0 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 15 Feb 2016 14:28:03 +0800
Subject: sctp: move rcu_read_lock from __sctp_lookup_association to
 sctp_lookup_association

__sctp_lookup_association() is only invoked by sctp_v4_err() and
sctp_rcv(), both which run on the rx BH, and it has been protected
by rcu_read_lock [see ip_local_deliver_finish() / ipv6_rcv()].

So we can move it to sctp_lookup_association, only let
sctp_lookup_association use rcu_read_lock.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sctp/input.c b/net/sctp/input.c
index 49d2cc7..21a2d6b 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -937,7 +937,6 @@ static struct sctp_association *__sctp_lookup_association(
 	struct sctp_transport *t;
 	struct sctp_association *asoc = NULL;
 
-	rcu_read_lock();
 	t = sctp_addrs_lookup_transport(net, local, peer);
 	if (!t || !sctp_transport_hold(t))
 		goto out;
@@ -949,7 +948,6 @@ static struct sctp_association *__sctp_lookup_association(
 	sctp_transport_put(t);
 
 out:
-	rcu_read_unlock();
 	return asoc;
 }
 
@@ -962,7 +960,9 @@ struct sctp_association *sctp_lookup_association(struct net *net,
 {
 	struct sctp_association *asoc;
 
+	rcu_read_lock();
 	asoc = __sctp_lookup_association(net, laddr, paddr, transportp);
+	rcu_read_unlock();
 
 	return asoc;
 }
-- 
cgit v0.10.2


From ac1efde8022acf705c4b8393d5ee4b070594ba88 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 15 Feb 2016 14:28:04 +0800
Subject: sctp: remove rcu_read_lock in sctp_seq_dump_remote_addrs()

sctp_seq_dump_remote_addrs is only called by sctp_assocs_seq_show()
and it has been protected by rcu_read_lock that is from
rhashtable_walk_start().

So we will remove this one.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sctp/proc.c b/net/sctp/proc.c
index ded7d93..cfc3c71 100644
--- a/net/sctp/proc.c
+++ b/net/sctp/proc.c
@@ -161,7 +161,6 @@ static void sctp_seq_dump_remote_addrs(struct seq_file *seq, struct sctp_associa
 	struct sctp_af *af;
 
 	primary = &assoc->peer.primary_addr;
-	rcu_read_lock();
 	list_for_each_entry_rcu(transport, &assoc->peer.transport_addr_list,
 			transports) {
 		addr = &transport->ipaddr;
@@ -172,7 +171,6 @@ static void sctp_seq_dump_remote_addrs(struct seq_file *seq, struct sctp_associa
 		}
 		af->seq_dump_addr(seq, addr);
 	}
-	rcu_read_unlock();
 }
 
 static void *sctp_eps_seq_start(struct seq_file *seq, loff_t *pos)
-- 
cgit v0.10.2


From 1cd4d5c4326a7ed3bb0e346bd7d20f5057a80ae6 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 15 Feb 2016 14:28:05 +0800
Subject: sctp: remove the unused sctp_datamsg_free()

Since commit 8b570dc9f7b6 ("sctp: only drop the reference on the datamsg
after sending a msg") used sctp_datamsg_put in sctp_sendmsg, instead of
sctp_datamsg_free, this function has no use in sctp.

So we will remove it.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 205630b..d05b566 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -535,7 +535,6 @@ struct sctp_datamsg {
 struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *,
 					    struct sctp_sndrcvinfo *,
 					    struct iov_iter *);
-void sctp_datamsg_free(struct sctp_datamsg *);
 void sctp_datamsg_put(struct sctp_datamsg *);
 void sctp_chunk_fail(struct sctp_chunk *, int error);
 int sctp_chunk_abandoned(struct sctp_chunk *);
diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c
index a338091..3aa4307 100644
--- a/net/sctp/chunk.c
+++ b/net/sctp/chunk.c
@@ -70,19 +70,6 @@ static struct sctp_datamsg *sctp_datamsg_new(gfp_t gfp)
 	return msg;
 }
 
-void sctp_datamsg_free(struct sctp_datamsg *msg)
-{
-	struct sctp_chunk *chunk;
-
-	/* This doesn't have to be a _safe vairant because
-	 * sctp_chunk_free() only drops the refs.
-	 */
-	list_for_each_entry(chunk, &msg->chunks, frag_list)
-		sctp_chunk_free(chunk);
-
-	sctp_datamsg_put(msg);
-}
-
 /* Final destructruction of datamsg memory. */
 static void sctp_datamsg_destroy(struct sctp_datamsg *msg)
 {
-- 
cgit v0.10.2


From fc48b7a6148af974b49db145812a8b060324a503 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Mon, 15 Feb 2016 13:22:35 -0500
Subject: qed/qede: use 8.7.3.0 FW.

This patch moves the qed* driver into utilizing the 8.7.3.0 FW.
This new FW is required for a lot of new SW features, including:
  - Vlan filtering offload
  - Encapsulation offload support
  - HW ingress aggregations
As well as paving the way for the possibility of adding storage protocols
in the future.

V2:
 - Fix kbuild test robot error/warnings.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Sudarsana Reddy Kalluru <Sudarsana.Kalluru@qlogic.com>
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index 1292c36..d34da63 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -70,8 +70,8 @@ struct qed_sb_sp_info;
 struct qed_mcp_info;
 
 struct qed_rt_data {
-	u32 init_val;
-	bool b_valid;
+	u32	*init_val;
+	bool	*b_valid;
 };
 
 /* The PCI personality is not quite synonymous to protocol ID:
@@ -120,6 +120,10 @@ enum QED_PORT_MODE {
 	QED_PORT_MODE_DE_1X25G
 };
 
+enum qed_dev_cap {
+	QED_DEV_CAP_ETH,
+};
+
 struct qed_hw_info {
 	/* PCI personality */
 	enum qed_pci_personality	personality;
@@ -151,6 +155,7 @@ struct qed_hw_info {
 
 	u32				port_mode;
 	u32				hw_mode;
+	unsigned long		device_capabilities;
 };
 
 struct qed_hw_cid_data {
@@ -267,7 +272,7 @@ struct qed_hwfn {
 	struct qed_hw_info		hw_info;
 
 	/* rt_array (for init-tool) */
-	struct qed_rt_data		*rt_data;
+	struct qed_rt_data		rt_data;
 
 	/* SPQ */
 	struct qed_spq			*p_spq;
@@ -350,9 +355,20 @@ struct qed_dev {
 	char	name[NAME_SIZE];
 
 	u8	type;
-#define QED_DEV_TYPE_BB_A0      (0 << 0)
-#define QED_DEV_TYPE_MASK       (0x3)
-#define QED_DEV_TYPE_SHIFT      (0)
+#define QED_DEV_TYPE_BB (0 << 0)
+#define QED_DEV_TYPE_AH BIT(0)
+/* Translate type/revision combo into the proper conditions */
+#define QED_IS_BB(dev)  ((dev)->type == QED_DEV_TYPE_BB)
+#define QED_IS_BB_A0(dev)       (QED_IS_BB(dev) && \
+				 CHIP_REV_IS_A0(dev))
+#define QED_IS_BB_B0(dev)       (QED_IS_BB(dev) && \
+				 CHIP_REV_IS_B0(dev))
+
+#define QED_GET_TYPE(dev)       (QED_IS_BB_A0(dev) ? CHIP_BB_A0 : \
+				 QED_IS_BB_B0(dev) ? CHIP_BB_B0 : CHIP_K2)
+
+	u16	vendor_id;
+	u16	device_id;
 
 	u16	chip_num;
 #define CHIP_NUM_MASK                   0xffff
@@ -361,6 +377,8 @@ struct qed_dev {
 	u16	chip_rev;
 #define CHIP_REV_MASK                   0xf
 #define CHIP_REV_SHIFT                  12
+#define CHIP_REV_IS_A0(_cdev)   (!(_cdev)->chip_rev)
+#define CHIP_REV_IS_B0(_cdev)   ((_cdev)->chip_rev == 1)
 
 	u16				chip_metal;
 #define CHIP_METAL_MASK                 0xff
@@ -375,10 +393,10 @@ struct qed_dev {
 	u8				num_funcs_in_port;
 
 	u8				path_id;
-	enum mf_mode			mf_mode;
-#define IS_MF(_p_hwfn)          (((_p_hwfn)->cdev)->mf_mode != SF)
-#define IS_MF_SI(_p_hwfn)       (((_p_hwfn)->cdev)->mf_mode == MF_NPAR)
-#define IS_MF_SD(_p_hwfn)       (((_p_hwfn)->cdev)->mf_mode == MF_OVLAN)
+	enum qed_mf_mode		mf_mode;
+#define IS_MF_DEFAULT(_p_hwfn)  (((_p_hwfn)->cdev)->mf_mode == QED_MF_DEFAULT)
+#define IS_MF_SI(_p_hwfn)       (((_p_hwfn)->cdev)->mf_mode == QED_MF_NPAR)
+#define IS_MF_SD(_p_hwfn)       (((_p_hwfn)->cdev)->mf_mode == QED_MF_OVLAN)
 
 	int				pcie_width;
 	int				pcie_speed;
@@ -441,11 +459,6 @@ struct qed_dev {
 	const struct firmware		*firmware;
 };
 
-#define QED_GET_TYPE(dev)       (((dev)->type & QED_DEV_TYPE_MASK) >> \
-				 QED_DEV_TYPE_SHIFT)
-#define QED_IS_BB_A0(dev)       (QED_GET_TYPE(dev) == QED_DEV_TYPE_BB_A0)
-#define QED_IS_BB(dev)  (QED_IS_BB_A0(dev))
-
 #define NUM_OF_SBS(dev)         MAX_SB_PER_PATH_BB
 #define NUM_OF_ENG_PFS(dev)     MAX_NUM_PFS_BB
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
index 7ccdb46..d3f7a02 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
@@ -581,7 +581,8 @@ void qed_qm_init_pf(struct qed_hwfn *p_hwfn)
 	params.num_pf_cids = iids.cids;
 	params.start_pq = qm_info->start_pq;
 	params.num_pf_pqs = qm_info->num_pqs;
-	params.start_vport = qm_info->num_vports;
+	params.start_vport = qm_info->start_vport;
+	params.num_vports = qm_info->num_vports;
 	params.pf_wfq = qm_info->pf_wfq;
 	params.pf_rl = qm_info->pf_rl;
 	params.pq_params = qm_info->qm_pq_params;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 817bbd5..bc17ed2 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -341,11 +341,6 @@ void qed_resc_setup(struct qed_dev *cdev)
 	}
 }
 
-#define FINAL_CLEANUP_CMD_OFFSET        (0)
-#define FINAL_CLEANUP_CMD (0x1)
-#define FINAL_CLEANUP_VALID_OFFSET      (6)
-#define FINAL_CLEANUP_VFPF_ID_SHIFT     (7)
-#define FINAL_CLEANUP_COMP (0x2)
 #define FINAL_CLEANUP_POLL_CNT          (100)
 #define FINAL_CLEANUP_POLL_TIME         (10)
 int qed_final_cleanup(struct qed_hwfn *p_hwfn,
@@ -355,12 +350,14 @@ int qed_final_cleanup(struct qed_hwfn *p_hwfn,
 	u32 command = 0, addr, count = FINAL_CLEANUP_POLL_CNT;
 	int rc = -EBUSY;
 
-	addr = GTT_BAR0_MAP_REG_USDM_RAM + USTORM_FLR_FINAL_ACK_OFFSET;
+	addr = GTT_BAR0_MAP_REG_USDM_RAM +
+		USTORM_FLR_FINAL_ACK_OFFSET(p_hwfn->rel_pf_id);
 
-	command |= FINAL_CLEANUP_CMD << FINAL_CLEANUP_CMD_OFFSET;
-	command |= 1 << FINAL_CLEANUP_VALID_OFFSET;
-	command |= id << FINAL_CLEANUP_VFPF_ID_SHIFT;
-	command |= FINAL_CLEANUP_COMP << SDM_OP_GEN_COMP_TYPE_SHIFT;
+	command |= X_FINAL_CLEANUP_AGG_INT <<
+		SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_SHIFT;
+	command |= 1 << SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_SHIFT;
+	command |= id << SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_SHIFT;
+	command |= SDM_COMP_TYPE_AGG_INT << SDM_OP_GEN_COMP_TYPE_SHIFT;
 
 	/* Make sure notification is not set before initiating final cleanup */
 	if (REG_RD(p_hwfn, addr)) {
@@ -415,18 +412,16 @@ static void qed_calc_hw_mode(struct qed_hwfn *p_hwfn)
 	}
 
 	switch (p_hwfn->cdev->mf_mode) {
-	case SF:
-		hw_mode |= 1 << MODE_SF;
+	case QED_MF_DEFAULT:
+	case QED_MF_NPAR:
+		hw_mode |= 1 << MODE_MF_SI;
 		break;
-	case MF_OVLAN:
+	case QED_MF_OVLAN:
 		hw_mode |= 1 << MODE_MF_SD;
 		break;
-	case MF_NPAR:
-		hw_mode |= 1 << MODE_MF_SI;
-		break;
 	default:
-		DP_NOTICE(p_hwfn, "Unsupported MF mode, init as SF\n");
-		hw_mode |= 1 << MODE_SF;
+		DP_NOTICE(p_hwfn, "Unsupported MF mode, init as DEFAULT\n");
+		hw_mode |= 1 << MODE_MF_SI;
 	}
 
 	hw_mode |= 1 << MODE_ASIC;
@@ -1018,8 +1013,7 @@ static void qed_hw_get_resc(struct qed_hwfn *p_hwfn)
 	u32 *resc_num = p_hwfn->hw_info.resc_num;
 	int num_funcs, i;
 
-	num_funcs = IS_MF(p_hwfn) ? MAX_NUM_PFS_BB
-				  : p_hwfn->cdev->num_ports_in_engines;
+	num_funcs = MAX_NUM_PFS_BB;
 
 	resc_num[QED_SB] = min_t(u32,
 				 (MAX_SB_PER_PATH_BB / num_funcs),
@@ -1071,7 +1065,7 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn,
 			       struct qed_ptt *p_ptt)
 {
 	u32 nvm_cfg1_offset, mf_mode, addr, generic_cont0, core_cfg;
-	u32 port_cfg_addr, link_temp, val, nvm_cfg_addr;
+	u32 port_cfg_addr, link_temp, nvm_cfg_addr, device_capabilities;
 	struct qed_mcp_link_params *link;
 
 	/* Read global nvm_cfg address */
@@ -1134,21 +1128,6 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn,
 		break;
 	}
 
-	addr = MCP_REG_SCRATCH + nvm_cfg1_offset +
-	       offsetof(struct nvm_cfg1, func[MCP_PF_ID(p_hwfn)]) +
-	       offsetof(struct nvm_cfg1_func, device_id);
-	val = qed_rd(p_hwfn, p_ptt, addr);
-
-	if (IS_MF(p_hwfn)) {
-		p_hwfn->hw_info.device_id =
-			(val & NVM_CFG1_FUNC_MF_VENDOR_DEVICE_ID_MASK) >>
-			NVM_CFG1_FUNC_MF_VENDOR_DEVICE_ID_OFFSET;
-	} else {
-		p_hwfn->hw_info.device_id =
-			(val & NVM_CFG1_FUNC_VENDOR_DEVICE_ID_MASK) >>
-			NVM_CFG1_FUNC_VENDOR_DEVICE_ID_OFFSET;
-	}
-
 	/* Read default link configuration */
 	link = &p_hwfn->mcp_info->link_input;
 	port_cfg_addr = MCP_REG_SCRATCH + nvm_cfg1_offset +
@@ -1220,18 +1199,28 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn,
 
 	switch (mf_mode) {
 	case NVM_CFG1_GLOB_MF_MODE_MF_ALLOWED:
-		p_hwfn->cdev->mf_mode = MF_OVLAN;
+		p_hwfn->cdev->mf_mode = QED_MF_OVLAN;
 		break;
 	case NVM_CFG1_GLOB_MF_MODE_NPAR1_0:
-		p_hwfn->cdev->mf_mode = MF_NPAR;
+		p_hwfn->cdev->mf_mode = QED_MF_NPAR;
 		break;
-	case NVM_CFG1_GLOB_MF_MODE_FORCED_SF:
-		p_hwfn->cdev->mf_mode = SF;
+	case NVM_CFG1_GLOB_MF_MODE_DEFAULT:
+		p_hwfn->cdev->mf_mode = QED_MF_DEFAULT;
 		break;
 	}
 	DP_INFO(p_hwfn, "Multi function mode is %08x\n",
 		p_hwfn->cdev->mf_mode);
 
+	/* Read Multi-function information from shmem */
+	addr = MCP_REG_SCRATCH + nvm_cfg1_offset +
+		offsetof(struct nvm_cfg1, glob) +
+		offsetof(struct nvm_cfg1_glob, device_capabilities);
+
+	device_capabilities = qed_rd(p_hwfn, p_ptt, addr);
+	if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET)
+		__set_bit(QED_DEV_CAP_ETH,
+			  &p_hwfn->hw_info.device_capabilities);
+
 	return qed_mcp_fill_shmem_func_info(p_hwfn, p_ptt);
 }
 
@@ -1293,29 +1282,36 @@ qed_get_hw_info(struct qed_hwfn *p_hwfn,
 
 static void qed_get_dev_info(struct qed_dev *cdev)
 {
+	struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev);
 	u32 tmp;
 
-	cdev->chip_num = (u16)qed_rd(cdev->hwfns, cdev->hwfns[0].p_main_ptt,
+	/* Read Vendor Id / Device Id */
+	pci_read_config_word(cdev->pdev, PCI_VENDOR_ID,
+			     &cdev->vendor_id);
+	pci_read_config_word(cdev->pdev, PCI_DEVICE_ID,
+			     &cdev->device_id);
+	cdev->chip_num = (u16)qed_rd(p_hwfn, p_hwfn->p_main_ptt,
 				     MISCS_REG_CHIP_NUM);
-	cdev->chip_rev = (u16)qed_rd(cdev->hwfns, cdev->hwfns[0].p_main_ptt,
+	cdev->chip_rev = (u16)qed_rd(p_hwfn, p_hwfn->p_main_ptt,
 				     MISCS_REG_CHIP_REV);
 	MASK_FIELD(CHIP_REV, cdev->chip_rev);
 
+	cdev->type = QED_DEV_TYPE_BB;
 	/* Learn number of HW-functions */
-	tmp = qed_rd(cdev->hwfns, cdev->hwfns[0].p_main_ptt,
+	tmp = qed_rd(p_hwfn, p_hwfn->p_main_ptt,
 		     MISCS_REG_CMT_ENABLED_FOR_PAIR);
 
-	if (tmp & (1 << cdev->hwfns[0].rel_pf_id)) {
+	if (tmp & (1 << p_hwfn->rel_pf_id)) {
 		DP_NOTICE(cdev->hwfns, "device in CMT mode\n");
 		cdev->num_hwfns = 2;
 	} else {
 		cdev->num_hwfns = 1;
 	}
 
-	cdev->chip_bond_id = qed_rd(cdev->hwfns, cdev->hwfns[0].p_main_ptt,
+	cdev->chip_bond_id = qed_rd(p_hwfn, p_hwfn->p_main_ptt,
 				    MISCS_REG_CHIP_TEST_REG) >> 4;
 	MASK_FIELD(CHIP_BOND_ID, cdev->chip_bond_id);
-	cdev->chip_metal = (u16)qed_rd(cdev->hwfns, cdev->hwfns[0].p_main_ptt,
+	cdev->chip_metal = (u16)qed_rd(p_hwfn, p_hwfn->p_main_ptt,
 				       MISCS_REG_CHIP_METAL);
 	MASK_FIELD(CHIP_METAL, cdev->chip_metal);
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 264e954..49bbf69 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -34,6 +34,8 @@ enum common_event_opcode {
 	COMMON_EVENT_RESERVED3,
 	COMMON_EVENT_RESERVED4,
 	COMMON_EVENT_RESERVED5,
+	COMMON_EVENT_RESERVED6,
+	COMMON_EVENT_EMPTY,
 	MAX_COMMON_EVENT_OPCODE
 };
 
@@ -45,6 +47,7 @@ enum common_ramrod_cmd_id {
 	COMMON_RAMROD_RESERVED,
 	COMMON_RAMROD_RESERVED2,
 	COMMON_RAMROD_RESERVED3,
+	COMMON_RAMROD_EMPTY,
 	MAX_COMMON_RAMROD_CMD_ID
 };
 
@@ -331,6 +334,179 @@ struct xstorm_core_conn_ag_ctx {
 	__le16	word15 /* word15 */;
 };
 
+struct tstorm_core_conn_ag_ctx {
+	u8	byte0 /* cdu_validation */;
+	u8	byte1 /* state */;
+	u8	flags0;
+#define TSTORM_CORE_CONN_AG_CTX_BIT0_MASK     0x1       /* exist_in_qm0 */
+#define TSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT    0
+#define TSTORM_CORE_CONN_AG_CTX_BIT1_MASK     0x1       /* exist_in_qm1 */
+#define TSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT    1
+#define TSTORM_CORE_CONN_AG_CTX_BIT2_MASK     0x1       /* bit2 */
+#define TSTORM_CORE_CONN_AG_CTX_BIT2_SHIFT    2
+#define TSTORM_CORE_CONN_AG_CTX_BIT3_MASK     0x1       /* bit3 */
+#define TSTORM_CORE_CONN_AG_CTX_BIT3_SHIFT    3
+#define TSTORM_CORE_CONN_AG_CTX_BIT4_MASK     0x1       /* bit4 */
+#define TSTORM_CORE_CONN_AG_CTX_BIT4_SHIFT    4
+#define TSTORM_CORE_CONN_AG_CTX_BIT5_MASK     0x1       /* bit5 */
+#define TSTORM_CORE_CONN_AG_CTX_BIT5_SHIFT    5
+#define TSTORM_CORE_CONN_AG_CTX_CF0_MASK      0x3       /* timer0cf */
+#define TSTORM_CORE_CONN_AG_CTX_CF0_SHIFT     6
+	u8 flags1;
+#define TSTORM_CORE_CONN_AG_CTX_CF1_MASK      0x3       /* timer1cf */
+#define TSTORM_CORE_CONN_AG_CTX_CF1_SHIFT     0
+#define TSTORM_CORE_CONN_AG_CTX_CF2_MASK      0x3       /* timer2cf */
+#define TSTORM_CORE_CONN_AG_CTX_CF2_SHIFT     2
+#define TSTORM_CORE_CONN_AG_CTX_CF3_MASK      0x3       /* timer_stop_all */
+#define TSTORM_CORE_CONN_AG_CTX_CF3_SHIFT     4
+#define TSTORM_CORE_CONN_AG_CTX_CF4_MASK      0x3       /* cf4 */
+#define TSTORM_CORE_CONN_AG_CTX_CF4_SHIFT     6
+	u8 flags2;
+#define TSTORM_CORE_CONN_AG_CTX_CF5_MASK      0x3       /* cf5 */
+#define TSTORM_CORE_CONN_AG_CTX_CF5_SHIFT     0
+#define TSTORM_CORE_CONN_AG_CTX_CF6_MASK      0x3       /* cf6 */
+#define TSTORM_CORE_CONN_AG_CTX_CF6_SHIFT     2
+#define TSTORM_CORE_CONN_AG_CTX_CF7_MASK      0x3       /* cf7 */
+#define TSTORM_CORE_CONN_AG_CTX_CF7_SHIFT     4
+#define TSTORM_CORE_CONN_AG_CTX_CF8_MASK      0x3       /* cf8 */
+#define TSTORM_CORE_CONN_AG_CTX_CF8_SHIFT     6
+	u8 flags3;
+#define TSTORM_CORE_CONN_AG_CTX_CF9_MASK      0x3       /* cf9 */
+#define TSTORM_CORE_CONN_AG_CTX_CF9_SHIFT     0
+#define TSTORM_CORE_CONN_AG_CTX_CF10_MASK     0x3       /* cf10 */
+#define TSTORM_CORE_CONN_AG_CTX_CF10_SHIFT    2
+#define TSTORM_CORE_CONN_AG_CTX_CF0EN_MASK    0x1       /* cf0en */
+#define TSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT   4
+#define TSTORM_CORE_CONN_AG_CTX_CF1EN_MASK    0x1       /* cf1en */
+#define TSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT   5
+#define TSTORM_CORE_CONN_AG_CTX_CF2EN_MASK    0x1       /* cf2en */
+#define TSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT   6
+#define TSTORM_CORE_CONN_AG_CTX_CF3EN_MASK    0x1       /* cf3en */
+#define TSTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT   7
+	u8 flags4;
+#define TSTORM_CORE_CONN_AG_CTX_CF4EN_MASK    0x1       /* cf4en */
+#define TSTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT   0
+#define TSTORM_CORE_CONN_AG_CTX_CF5EN_MASK    0x1       /* cf5en */
+#define TSTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT   1
+#define TSTORM_CORE_CONN_AG_CTX_CF6EN_MASK    0x1       /* cf6en */
+#define TSTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT   2
+#define TSTORM_CORE_CONN_AG_CTX_CF7EN_MASK    0x1       /* cf7en */
+#define TSTORM_CORE_CONN_AG_CTX_CF7EN_SHIFT   3
+#define TSTORM_CORE_CONN_AG_CTX_CF8EN_MASK    0x1       /* cf8en */
+#define TSTORM_CORE_CONN_AG_CTX_CF8EN_SHIFT   4
+#define TSTORM_CORE_CONN_AG_CTX_CF9EN_MASK    0x1       /* cf9en */
+#define TSTORM_CORE_CONN_AG_CTX_CF9EN_SHIFT   5
+#define TSTORM_CORE_CONN_AG_CTX_CF10EN_MASK   0x1       /* cf10en */
+#define TSTORM_CORE_CONN_AG_CTX_CF10EN_SHIFT  6
+#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK  0x1       /* rule0en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7
+	u8 flags5;
+#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK  0x1       /* rule1en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0
+#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK  0x1       /* rule2en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1
+#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK  0x1       /* rule3en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2
+#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK  0x1       /* rule4en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3
+#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_MASK  0x1       /* rule5en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4
+#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_MASK  0x1       /* rule6en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5
+#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_MASK  0x1       /* rule7en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6
+#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_MASK  0x1       /* rule8en */
+#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7
+	__le32	reg0 /* reg0 */;
+	__le32	reg1 /* reg1 */;
+	__le32	reg2 /* reg2 */;
+	__le32	reg3 /* reg3 */;
+	__le32	reg4 /* reg4 */;
+	__le32	reg5 /* reg5 */;
+	__le32	reg6 /* reg6 */;
+	__le32	reg7 /* reg7 */;
+	__le32	reg8 /* reg8 */;
+	u8	byte2 /* byte2 */;
+	u8	byte3 /* byte3 */;
+	__le16	word0 /* word0 */;
+	u8	byte4 /* byte4 */;
+	u8	byte5 /* byte5 */;
+	__le16	word1 /* word1 */;
+	__le16	word2 /* conn_dpi */;
+	__le16	word3 /* word3 */;
+	__le32	reg9 /* reg9 */;
+	__le32	reg10 /* reg10 */;
+};
+
+struct ustorm_core_conn_ag_ctx {
+	u8	reserved /* cdu_validation */;
+	u8	byte1 /* state */;
+	u8	flags0;
+#define USTORM_CORE_CONN_AG_CTX_BIT0_MASK     0x1       /* exist_in_qm0 */
+#define USTORM_CORE_CONN_AG_CTX_BIT0_SHIFT    0
+#define USTORM_CORE_CONN_AG_CTX_BIT1_MASK     0x1       /* exist_in_qm1 */
+#define USTORM_CORE_CONN_AG_CTX_BIT1_SHIFT    1
+#define USTORM_CORE_CONN_AG_CTX_CF0_MASK      0x3       /* timer0cf */
+#define USTORM_CORE_CONN_AG_CTX_CF0_SHIFT     2
+#define USTORM_CORE_CONN_AG_CTX_CF1_MASK      0x3       /* timer1cf */
+#define USTORM_CORE_CONN_AG_CTX_CF1_SHIFT     4
+#define USTORM_CORE_CONN_AG_CTX_CF2_MASK      0x3       /* timer2cf */
+#define USTORM_CORE_CONN_AG_CTX_CF2_SHIFT     6
+	u8 flags1;
+#define USTORM_CORE_CONN_AG_CTX_CF3_MASK      0x3       /* timer_stop_all */
+#define USTORM_CORE_CONN_AG_CTX_CF3_SHIFT     0
+#define USTORM_CORE_CONN_AG_CTX_CF4_MASK      0x3       /* cf4 */
+#define USTORM_CORE_CONN_AG_CTX_CF4_SHIFT     2
+#define USTORM_CORE_CONN_AG_CTX_CF5_MASK      0x3       /* cf5 */
+#define USTORM_CORE_CONN_AG_CTX_CF5_SHIFT     4
+#define USTORM_CORE_CONN_AG_CTX_CF6_MASK      0x3       /* cf6 */
+#define USTORM_CORE_CONN_AG_CTX_CF6_SHIFT     6
+	u8 flags2;
+#define USTORM_CORE_CONN_AG_CTX_CF0EN_MASK    0x1       /* cf0en */
+#define USTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT   0
+#define USTORM_CORE_CONN_AG_CTX_CF1EN_MASK    0x1       /* cf1en */
+#define USTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT   1
+#define USTORM_CORE_CONN_AG_CTX_CF2EN_MASK    0x1       /* cf2en */
+#define USTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT   2
+#define USTORM_CORE_CONN_AG_CTX_CF3EN_MASK    0x1       /* cf3en */
+#define USTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT   3
+#define USTORM_CORE_CONN_AG_CTX_CF4EN_MASK    0x1       /* cf4en */
+#define USTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT   4
+#define USTORM_CORE_CONN_AG_CTX_CF5EN_MASK    0x1       /* cf5en */
+#define USTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT   5
+#define USTORM_CORE_CONN_AG_CTX_CF6EN_MASK    0x1       /* cf6en */
+#define USTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT   6
+#define USTORM_CORE_CONN_AG_CTX_RULE0EN_MASK  0x1       /* rule0en */
+#define USTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7
+	u8 flags3;
+#define USTORM_CORE_CONN_AG_CTX_RULE1EN_MASK  0x1       /* rule1en */
+#define USTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0
+#define USTORM_CORE_CONN_AG_CTX_RULE2EN_MASK  0x1       /* rule2en */
+#define USTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1
+#define USTORM_CORE_CONN_AG_CTX_RULE3EN_MASK  0x1       /* rule3en */
+#define USTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2
+#define USTORM_CORE_CONN_AG_CTX_RULE4EN_MASK  0x1       /* rule4en */
+#define USTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3
+#define USTORM_CORE_CONN_AG_CTX_RULE5EN_MASK  0x1       /* rule5en */
+#define USTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4
+#define USTORM_CORE_CONN_AG_CTX_RULE6EN_MASK  0x1       /* rule6en */
+#define USTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5
+#define USTORM_CORE_CONN_AG_CTX_RULE7EN_MASK  0x1       /* rule7en */
+#define USTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6
+#define USTORM_CORE_CONN_AG_CTX_RULE8EN_MASK  0x1       /* rule8en */
+#define USTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7
+	u8	byte2 /* byte2 */;
+	u8	byte3 /* byte3 */;
+	__le16	word0 /* conn_dpi */;
+	__le16	word1 /* word1 */;
+	__le32	rx_producers /* reg0 */;
+	__le32	reg1 /* reg1 */;
+	__le32	reg2 /* reg2 */;
+	__le32	reg3 /* reg3 */;
+	__le16	word2 /* word2 */;
+	__le16	word3 /* word3 */;
+};
+
 /* The core storm context for the Mstorm */
 struct mstorm_core_conn_st_ctx {
 	__le32 reserved[24];
@@ -349,8 +525,9 @@ struct core_conn_context {
 	struct regpair			pstorm_st_padding[2];
 	struct xstorm_core_conn_st_ctx	xstorm_st_context;
 	struct xstorm_core_conn_ag_ctx	xstorm_ag_context;
+	struct tstorm_core_conn_ag_ctx	tstorm_ag_context;
+	struct ustorm_core_conn_ag_ctx	ustorm_ag_context;
 	struct mstorm_core_conn_st_ctx	mstorm_st_context;
-	struct regpair			mstorm_st_padding[2];
 	struct ustorm_core_conn_st_ctx	ustorm_st_context;
 	struct regpair			ustorm_st_padding[2] /* padding */;
 };
@@ -397,10 +574,12 @@ union event_ring_element {
 };
 
 enum personality_type {
+	BAD_PERSONALITY_TYP,
 	PERSONALITY_RESERVED,
 	PERSONALITY_RESERVED2,
 	PERSONALITY_RDMA_AND_ETH /* Roce or Iwarp */,
 	PERSONALITY_RESERVED3,
+	PERSONALITY_CORE,
 	PERSONALITY_ETH /* Ethernet */,
 	PERSONALITY_RESERVED4,
 	MAX_PERSONALITY_TYPE
@@ -570,7 +749,7 @@ enum block_addr {
 	GRCBASE_NWM		= 0x800000,
 	GRCBASE_NWS		= 0x700000,
 	GRCBASE_MS		= 0x6a0000,
-	GRCBASE_PHY_PCIE	= 0x618000,
+	GRCBASE_PHY_PCIE	= 0x620000,
 	GRCBASE_MISC_AEU	= 0x8000,
 	GRCBASE_BAR0_MAP	= 0x1c00000,
 	MAX_BLOCK_ADDR
@@ -795,13 +974,13 @@ enum init_modes {
 	MODE_RESERVED3,
 	MODE_RESERVED4,
 	MODE_RESERVED5,
+	MODE_RESERVED6,
 	MODE_SF,
 	MODE_MF_SD,
 	MODE_MF_SI,
 	MODE_PORTS_PER_ENG_1,
 	MODE_PORTS_PER_ENG_2,
 	MODE_PORTS_PER_ENG_4,
-	MODE_40G,
 	MODE_100G,
 	MODE_EAGLE_ENG1_WORKAROUND,
 	MAX_INIT_MODES
@@ -816,43 +995,6 @@ enum init_phases {
 	MAX_INIT_PHASES
 };
 
-struct mstorm_core_conn_ag_ctx {
-	u8	byte0 /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define MSTORM_CORE_CONN_AG_CTX_BIT0_MASK     0x1       /* exist_in_qm0 */
-#define MSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT    0
-#define MSTORM_CORE_CONN_AG_CTX_BIT1_MASK     0x1       /* exist_in_qm1 */
-#define MSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT    1
-#define MSTORM_CORE_CONN_AG_CTX_CF0_MASK      0x3       /* cf0 */
-#define MSTORM_CORE_CONN_AG_CTX_CF0_SHIFT     2
-#define MSTORM_CORE_CONN_AG_CTX_CF1_MASK      0x3       /* cf1 */
-#define MSTORM_CORE_CONN_AG_CTX_CF1_SHIFT     4
-#define MSTORM_CORE_CONN_AG_CTX_CF2_MASK      0x3       /* cf2 */
-#define MSTORM_CORE_CONN_AG_CTX_CF2_SHIFT     6
-	u8 flags1;
-#define MSTORM_CORE_CONN_AG_CTX_CF0EN_MASK    0x1       /* cf0en */
-#define MSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT   0
-#define MSTORM_CORE_CONN_AG_CTX_CF1EN_MASK    0x1       /* cf1en */
-#define MSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT   1
-#define MSTORM_CORE_CONN_AG_CTX_CF2EN_MASK    0x1       /* cf2en */
-#define MSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT   2
-#define MSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK  0x1       /* rule0en */
-#define MSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 3
-#define MSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK  0x1       /* rule1en */
-#define MSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 4
-#define MSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK  0x1       /* rule2en */
-#define MSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 5
-#define MSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK  0x1       /* rule3en */
-#define MSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 6
-#define MSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK  0x1       /* rule4en */
-#define MSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 7
-	__le16	word0 /* word0 */;
-	__le16	word1 /* word1 */;
-	__le32	reg0 /* reg0 */;
-	__le32	reg1 /* reg1 */;
-};
-
 /* per encapsulation type enabling flags */
 struct prs_reg_encapsulation_type_en {
 	u8 flags;
@@ -927,250 +1069,44 @@ struct qm_rf_opportunistic_mask {
 };
 
 /* QM hardware structure of QM map memory */
-struct qm_rf_pq_map {
-	u32 reg;
-#define QM_RF_PQ_MAP_PQ_VALID_MASK          0x1         /* PQ active */
-#define QM_RF_PQ_MAP_PQ_VALID_SHIFT         0
-#define QM_RF_PQ_MAP_RL_ID_MASK             0xFF        /* RL ID */
-#define QM_RF_PQ_MAP_RL_ID_SHIFT            1
-#define QM_RF_PQ_MAP_VP_PQ_ID_MASK          0x1FF
-#define QM_RF_PQ_MAP_VP_PQ_ID_SHIFT         9
-#define QM_RF_PQ_MAP_VOQ_MASK               0x1F        /* VOQ */
-#define QM_RF_PQ_MAP_VOQ_SHIFT              18
-#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_MASK  0x3         /* WRR weight */
-#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_SHIFT 23
-#define QM_RF_PQ_MAP_RL_VALID_MASK          0x1         /* RL active */
-#define QM_RF_PQ_MAP_RL_VALID_SHIFT         25
-#define QM_RF_PQ_MAP_RESERVED_MASK          0x3F
-#define QM_RF_PQ_MAP_RESERVED_SHIFT         26
-};
-
-/* SDM operation gen command (generate aggregative interrupt) */
-struct sdm_op_gen {
-	__le32 command;
-#define SDM_OP_GEN_COMP_PARAM_MASK  0xFFFF      /* completion parameters 0-15 */
-#define SDM_OP_GEN_COMP_PARAM_SHIFT 0
-#define SDM_OP_GEN_COMP_TYPE_MASK   0xF         /* completion type 16-19 */
-#define SDM_OP_GEN_COMP_TYPE_SHIFT  16
-#define SDM_OP_GEN_RESERVED_MASK    0xFFF       /* reserved 20-31 */
-#define SDM_OP_GEN_RESERVED_SHIFT   20
-};
-
-struct tstorm_core_conn_ag_ctx {
-	u8	byte0 /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define TSTORM_CORE_CONN_AG_CTX_BIT0_MASK     0x1       /* exist_in_qm0 */
-#define TSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT    0
-#define TSTORM_CORE_CONN_AG_CTX_BIT1_MASK     0x1       /* exist_in_qm1 */
-#define TSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT    1
-#define TSTORM_CORE_CONN_AG_CTX_BIT2_MASK     0x1       /* bit2 */
-#define TSTORM_CORE_CONN_AG_CTX_BIT2_SHIFT    2
-#define TSTORM_CORE_CONN_AG_CTX_BIT3_MASK     0x1       /* bit3 */
-#define TSTORM_CORE_CONN_AG_CTX_BIT3_SHIFT    3
-#define TSTORM_CORE_CONN_AG_CTX_BIT4_MASK     0x1       /* bit4 */
-#define TSTORM_CORE_CONN_AG_CTX_BIT4_SHIFT    4
-#define TSTORM_CORE_CONN_AG_CTX_BIT5_MASK     0x1       /* bit5 */
-#define TSTORM_CORE_CONN_AG_CTX_BIT5_SHIFT    5
-#define TSTORM_CORE_CONN_AG_CTX_CF0_MASK      0x3       /* timer0cf */
-#define TSTORM_CORE_CONN_AG_CTX_CF0_SHIFT     6
-	u8 flags1;
-#define TSTORM_CORE_CONN_AG_CTX_CF1_MASK      0x3       /* timer1cf */
-#define TSTORM_CORE_CONN_AG_CTX_CF1_SHIFT     0
-#define TSTORM_CORE_CONN_AG_CTX_CF2_MASK      0x3       /* timer2cf */
-#define TSTORM_CORE_CONN_AG_CTX_CF2_SHIFT     2
-#define TSTORM_CORE_CONN_AG_CTX_CF3_MASK      0x3       /* timer_stop_all */
-#define TSTORM_CORE_CONN_AG_CTX_CF3_SHIFT     4
-#define TSTORM_CORE_CONN_AG_CTX_CF4_MASK      0x3       /* cf4 */
-#define TSTORM_CORE_CONN_AG_CTX_CF4_SHIFT     6
-	u8 flags2;
-#define TSTORM_CORE_CONN_AG_CTX_CF5_MASK      0x3       /* cf5 */
-#define TSTORM_CORE_CONN_AG_CTX_CF5_SHIFT     0
-#define TSTORM_CORE_CONN_AG_CTX_CF6_MASK      0x3       /* cf6 */
-#define TSTORM_CORE_CONN_AG_CTX_CF6_SHIFT     2
-#define TSTORM_CORE_CONN_AG_CTX_CF7_MASK      0x3       /* cf7 */
-#define TSTORM_CORE_CONN_AG_CTX_CF7_SHIFT     4
-#define TSTORM_CORE_CONN_AG_CTX_CF8_MASK      0x3       /* cf8 */
-#define TSTORM_CORE_CONN_AG_CTX_CF8_SHIFT     6
-	u8 flags3;
-#define TSTORM_CORE_CONN_AG_CTX_CF9_MASK      0x3       /* cf9 */
-#define TSTORM_CORE_CONN_AG_CTX_CF9_SHIFT     0
-#define TSTORM_CORE_CONN_AG_CTX_CF10_MASK     0x3       /* cf10 */
-#define TSTORM_CORE_CONN_AG_CTX_CF10_SHIFT    2
-#define TSTORM_CORE_CONN_AG_CTX_CF0EN_MASK    0x1       /* cf0en */
-#define TSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT   4
-#define TSTORM_CORE_CONN_AG_CTX_CF1EN_MASK    0x1       /* cf1en */
-#define TSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT   5
-#define TSTORM_CORE_CONN_AG_CTX_CF2EN_MASK    0x1       /* cf2en */
-#define TSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT   6
-#define TSTORM_CORE_CONN_AG_CTX_CF3EN_MASK    0x1       /* cf3en */
-#define TSTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT   7
-	u8 flags4;
-#define TSTORM_CORE_CONN_AG_CTX_CF4EN_MASK    0x1       /* cf4en */
-#define TSTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT   0
-#define TSTORM_CORE_CONN_AG_CTX_CF5EN_MASK    0x1       /* cf5en */
-#define TSTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT   1
-#define TSTORM_CORE_CONN_AG_CTX_CF6EN_MASK    0x1       /* cf6en */
-#define TSTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT   2
-#define TSTORM_CORE_CONN_AG_CTX_CF7EN_MASK    0x1       /* cf7en */
-#define TSTORM_CORE_CONN_AG_CTX_CF7EN_SHIFT   3
-#define TSTORM_CORE_CONN_AG_CTX_CF8EN_MASK    0x1       /* cf8en */
-#define TSTORM_CORE_CONN_AG_CTX_CF8EN_SHIFT   4
-#define TSTORM_CORE_CONN_AG_CTX_CF9EN_MASK    0x1       /* cf9en */
-#define TSTORM_CORE_CONN_AG_CTX_CF9EN_SHIFT   5
-#define TSTORM_CORE_CONN_AG_CTX_CF10EN_MASK   0x1       /* cf10en */
-#define TSTORM_CORE_CONN_AG_CTX_CF10EN_SHIFT  6
-#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK  0x1       /* rule0en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7
-	u8 flags5;
-#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK  0x1       /* rule1en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0
-#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK  0x1       /* rule2en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1
-#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK  0x1       /* rule3en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2
-#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK  0x1       /* rule4en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3
-#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_MASK  0x1       /* rule5en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4
-#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_MASK  0x1       /* rule6en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5
-#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_MASK  0x1       /* rule7en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6
-#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_MASK  0x1       /* rule8en */
-#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7
-	__le32	reg0 /* reg0 */;
-	__le32	reg1 /* reg1 */;
-	__le32	reg2 /* reg2 */;
-	__le32	reg3 /* reg3 */;
-	__le32	reg4 /* reg4 */;
-	__le32	reg5 /* reg5 */;
-	__le32	reg6 /* reg6 */;
-	__le32	reg7 /* reg7 */;
-	__le32	reg8 /* reg8 */;
-	u8	byte2 /* byte2 */;
-	u8	byte3 /* byte3 */;
-	__le16	word0 /* word0 */;
-	u8	byte4 /* byte4 */;
-	u8	byte5 /* byte5 */;
-	__le16	word1 /* word1 */;
-	__le16	word2 /* conn_dpi */;
-	__le16	word3 /* word3 */;
-	__le32	reg9 /* reg9 */;
-	__le32	reg10 /* reg10 */;
+struct qm_rf_pq_map {
+	u32 reg;
+#define QM_RF_PQ_MAP_PQ_VALID_MASK          0x1         /* PQ active */
+#define QM_RF_PQ_MAP_PQ_VALID_SHIFT         0
+#define QM_RF_PQ_MAP_RL_ID_MASK             0xFF        /* RL ID */
+#define QM_RF_PQ_MAP_RL_ID_SHIFT            1
+#define QM_RF_PQ_MAP_VP_PQ_ID_MASK          0x1FF
+#define QM_RF_PQ_MAP_VP_PQ_ID_SHIFT         9
+#define QM_RF_PQ_MAP_VOQ_MASK               0x1F        /* VOQ */
+#define QM_RF_PQ_MAP_VOQ_SHIFT              18
+#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_MASK  0x3         /* WRR weight */
+#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_SHIFT 23
+#define QM_RF_PQ_MAP_RL_VALID_MASK          0x1         /* RL active */
+#define QM_RF_PQ_MAP_RL_VALID_SHIFT         25
+#define QM_RF_PQ_MAP_RESERVED_MASK          0x3F
+#define QM_RF_PQ_MAP_RESERVED_SHIFT         26
 };
 
-struct ustorm_core_conn_ag_ctx {
-	u8	reserved /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define USTORM_CORE_CONN_AG_CTX_BIT0_MASK     0x1       /* exist_in_qm0 */
-#define USTORM_CORE_CONN_AG_CTX_BIT0_SHIFT    0
-#define USTORM_CORE_CONN_AG_CTX_BIT1_MASK     0x1       /* exist_in_qm1 */
-#define USTORM_CORE_CONN_AG_CTX_BIT1_SHIFT    1
-#define USTORM_CORE_CONN_AG_CTX_CF0_MASK      0x3       /* timer0cf */
-#define USTORM_CORE_CONN_AG_CTX_CF0_SHIFT     2
-#define USTORM_CORE_CONN_AG_CTX_CF1_MASK      0x3       /* timer1cf */
-#define USTORM_CORE_CONN_AG_CTX_CF1_SHIFT     4
-#define USTORM_CORE_CONN_AG_CTX_CF2_MASK      0x3       /* timer2cf */
-#define USTORM_CORE_CONN_AG_CTX_CF2_SHIFT     6
-	u8 flags1;
-#define USTORM_CORE_CONN_AG_CTX_CF3_MASK      0x3       /* timer_stop_all */
-#define USTORM_CORE_CONN_AG_CTX_CF3_SHIFT     0
-#define USTORM_CORE_CONN_AG_CTX_CF4_MASK      0x3       /* cf4 */
-#define USTORM_CORE_CONN_AG_CTX_CF4_SHIFT     2
-#define USTORM_CORE_CONN_AG_CTX_CF5_MASK      0x3       /* cf5 */
-#define USTORM_CORE_CONN_AG_CTX_CF5_SHIFT     4
-#define USTORM_CORE_CONN_AG_CTX_CF6_MASK      0x3       /* cf6 */
-#define USTORM_CORE_CONN_AG_CTX_CF6_SHIFT     6
-	u8 flags2;
-#define USTORM_CORE_CONN_AG_CTX_CF0EN_MASK    0x1       /* cf0en */
-#define USTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT   0
-#define USTORM_CORE_CONN_AG_CTX_CF1EN_MASK    0x1       /* cf1en */
-#define USTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT   1
-#define USTORM_CORE_CONN_AG_CTX_CF2EN_MASK    0x1       /* cf2en */
-#define USTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT   2
-#define USTORM_CORE_CONN_AG_CTX_CF3EN_MASK    0x1       /* cf3en */
-#define USTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT   3
-#define USTORM_CORE_CONN_AG_CTX_CF4EN_MASK    0x1       /* cf4en */
-#define USTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT   4
-#define USTORM_CORE_CONN_AG_CTX_CF5EN_MASK    0x1       /* cf5en */
-#define USTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT   5
-#define USTORM_CORE_CONN_AG_CTX_CF6EN_MASK    0x1       /* cf6en */
-#define USTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT   6
-#define USTORM_CORE_CONN_AG_CTX_RULE0EN_MASK  0x1       /* rule0en */
-#define USTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7
-	u8 flags3;
-#define USTORM_CORE_CONN_AG_CTX_RULE1EN_MASK  0x1       /* rule1en */
-#define USTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0
-#define USTORM_CORE_CONN_AG_CTX_RULE2EN_MASK  0x1       /* rule2en */
-#define USTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1
-#define USTORM_CORE_CONN_AG_CTX_RULE3EN_MASK  0x1       /* rule3en */
-#define USTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2
-#define USTORM_CORE_CONN_AG_CTX_RULE4EN_MASK  0x1       /* rule4en */
-#define USTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3
-#define USTORM_CORE_CONN_AG_CTX_RULE5EN_MASK  0x1       /* rule5en */
-#define USTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4
-#define USTORM_CORE_CONN_AG_CTX_RULE6EN_MASK  0x1       /* rule6en */
-#define USTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5
-#define USTORM_CORE_CONN_AG_CTX_RULE7EN_MASK  0x1       /* rule7en */
-#define USTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6
-#define USTORM_CORE_CONN_AG_CTX_RULE8EN_MASK  0x1       /* rule8en */
-#define USTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7
-	u8	byte2 /* byte2 */;
-	u8	byte3 /* byte3 */;
-	__le16	word0 /* conn_dpi */;
-	__le16	word1 /* word1 */;
-	__le32	rx_producers /* reg0 */;
-	__le32	reg1 /* reg1 */;
-	__le32	reg2 /* reg2 */;
-	__le32	reg3 /* reg3 */;
-	__le16	word2 /* word2 */;
-	__le16	word3 /* word3 */;
+/* Completion params for aggregated interrupt completion */
+struct sdm_agg_int_comp_params {
+	__le16 params;
+#define SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_MASK      0x3F
+#define SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_SHIFT     0
+#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_MASK  0x1
+#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_SHIFT 6
+#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_MASK     0x1FF
+#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_SHIFT    7
 };
 
-struct ystorm_core_conn_ag_ctx {
-	u8	byte0 /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define YSTORM_CORE_CONN_AG_CTX_BIT0_MASK     0x1       /* exist_in_qm0 */
-#define YSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT    0
-#define YSTORM_CORE_CONN_AG_CTX_BIT1_MASK     0x1       /* exist_in_qm1 */
-#define YSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT    1
-#define YSTORM_CORE_CONN_AG_CTX_CF0_MASK      0x3       /* cf0 */
-#define YSTORM_CORE_CONN_AG_CTX_CF0_SHIFT     2
-#define YSTORM_CORE_CONN_AG_CTX_CF1_MASK      0x3       /* cf1 */
-#define YSTORM_CORE_CONN_AG_CTX_CF1_SHIFT     4
-#define YSTORM_CORE_CONN_AG_CTX_CF2_MASK      0x3       /* cf2 */
-#define YSTORM_CORE_CONN_AG_CTX_CF2_SHIFT     6
-	u8 flags1;
-#define YSTORM_CORE_CONN_AG_CTX_CF0EN_MASK    0x1       /* cf0en */
-#define YSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT   0
-#define YSTORM_CORE_CONN_AG_CTX_CF1EN_MASK    0x1       /* cf1en */
-#define YSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT   1
-#define YSTORM_CORE_CONN_AG_CTX_CF2EN_MASK    0x1       /* cf2en */
-#define YSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT   2
-#define YSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK  0x1       /* rule0en */
-#define YSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 3
-#define YSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK  0x1       /* rule1en */
-#define YSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 4
-#define YSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK  0x1       /* rule2en */
-#define YSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 5
-#define YSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK  0x1       /* rule3en */
-#define YSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 6
-#define YSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK  0x1       /* rule4en */
-#define YSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 7
-	u8	byte2 /* byte2 */;
-	u8	byte3 /* byte3 */;
-	__le16	word0 /* word0 */;
-	__le32	reg0 /* reg0 */;
-	__le32	reg1 /* reg1 */;
-	__le16	word1 /* word1 */;
-	__le16	word2 /* word2 */;
-	__le16	word3 /* word3 */;
-	__le16	word4 /* word4 */;
-	__le32	reg2 /* reg2 */;
-	__le32	reg3 /* reg3 */;
+/* SDM operation gen command (generate aggregative interrupt) */
+struct sdm_op_gen {
+	__le32 command;
+#define SDM_OP_GEN_COMP_PARAM_MASK  0xFFFF      /* completion parameters 0-15 */
+#define SDM_OP_GEN_COMP_PARAM_SHIFT 0
+#define SDM_OP_GEN_COMP_TYPE_MASK   0xF         /* completion type 16-19 */
+#define SDM_OP_GEN_COMP_TYPE_SHIFT  16
+#define SDM_OP_GEN_RESERVED_MASK    0xFFF       /* reserved 20-31 */
+#define SDM_OP_GEN_RESERVED_SHIFT   20
 };
 
 /*********************************** Init ************************************/
@@ -1274,13 +1210,6 @@ enum chip_ids {
 	MAX_CHIP_IDS
 };
 
-enum idle_chk_severity_types {
-	IDLE_CHK_SEVERITY_ERROR /* idle check failure should cause an error */,
-	IDLE_CHK_SEVERITY_ERROR_NO_TRAFFIC,
-	IDLE_CHK_SEVERITY_WARNING,
-	MAX_IDLE_CHK_SEVERITY_TYPES
-};
-
 struct init_array_raw_hdr {
 	__le32 data;
 #define INIT_ARRAY_RAW_HDR_TYPE_MASK    0xF
@@ -1340,14 +1269,6 @@ struct init_callback_op {
 	__le16	block_id /* Blocks ID */;
 };
 
-/* init comparison types */
-enum init_comparison_types {
-	INIT_COMPARISON_EQ /* init value is included in the init command */,
-	INIT_COMPARISON_OR /* init value is all zeros */,
-	INIT_COMPARISON_AND /* init value is an array of values */,
-	MAX_INIT_COMPARISON_TYPES
-};
-
 /* init operation: delay */
 struct init_delay_op {
 	__le32	op_data;
@@ -1444,12 +1365,10 @@ struct init_read_op {
 	__le32 op_data;
 #define INIT_READ_OP_OP_MASK         0xF
 #define INIT_READ_OP_OP_SHIFT        0
-#define INIT_READ_OP_POLL_COMP_MASK  0x7
-#define INIT_READ_OP_POLL_COMP_SHIFT 4
+#define INIT_READ_OP_POLL_TYPE_MASK  0xF
+#define INIT_READ_OP_POLL_TYPE_SHIFT 4
 #define INIT_READ_OP_RESERVED_MASK   0x1
-#define INIT_READ_OP_RESERVED_SHIFT  7
-#define INIT_READ_OP_POLL_MASK       0x1
-#define INIT_READ_OP_POLL_SHIFT      8
+#define INIT_READ_OP_RESERVED_SHIFT  8
 #define INIT_READ_OP_ADDRESS_MASK    0x7FFFFF
 #define INIT_READ_OP_ADDRESS_SHIFT   9
 	__le32 expected_val;
@@ -1477,6 +1396,14 @@ enum init_op_types {
 	MAX_INIT_OP_TYPES
 };
 
+enum init_poll_types {
+	INIT_POLL_NONE /* No polling */,
+	INIT_POLL_EQ /* init value is included in the init command */,
+	INIT_POLL_OR /* init value is all zeros */,
+	INIT_POLL_AND /* init value is an array of values */,
+	MAX_INIT_POLL_TYPES
+};
+
 /* init source types */
 enum init_source_types {
 	INIT_SRC_INLINE /* init value is included in the init command */,
@@ -1677,175 +1604,213 @@ bool qed_send_qm_stop_cmd(struct qed_hwfn	*p_hwfn,
 			  u16			num_pqs);
 
 /* Ystorm flow control mode. Use enum fw_flow_ctrl_mode */
-#define YSTORM_FLOW_CONTROL_MODE_OFFSET			(IRO[0].base)
-#define YSTORM_FLOW_CONTROL_MODE_SIZE			(IRO[0].size)
+#define YSTORM_FLOW_CONTROL_MODE_OFFSET  (IRO[0].base)
+#define YSTORM_FLOW_CONTROL_MODE_SIZE    (IRO[0].size)
 /* Tstorm port statistics */
-#define TSTORM_PORT_STAT_OFFSET(port_id)		(IRO[1].base + \
-							 ((port_id) * \
-							  IRO[1].m1))
-#define TSTORM_PORT_STAT_SIZE				(IRO[1].size)
+#define TSTORM_PORT_STAT_OFFSET(port_id) (IRO[1].base + ((port_id) * IRO[1].m1))
+#define TSTORM_PORT_STAT_SIZE            (IRO[1].size)
+/* Tstorm ll2 port statistics */
+#define TSTORM_LL2_PORT_STAT_OFFSET(port_id) \
+				(IRO[2].base + ((port_id) * IRO[2].m1))
+#define TSTORM_LL2_PORT_STAT_SIZE            (IRO[2].size)
 /* Ustorm VF-PF Channel ready flag */
-#define USTORM_VF_PF_CHANNEL_READY_OFFSET(vf_id)	(IRO[2].base +	\
-							 ((vf_id) *	\
-							  IRO[2].m1))
-#define USTORM_VF_PF_CHANNEL_READY_SIZE			(IRO[2].size)
+#define USTORM_VF_PF_CHANNEL_READY_OFFSET(vf_id) \
+				(IRO[3].base +	((vf_id) * IRO[3].m1))
+#define USTORM_VF_PF_CHANNEL_READY_SIZE          (IRO[3].size)
 /* Ustorm Final flr cleanup ack */
-#define USTORM_FLR_FINAL_ACK_OFFSET			(IRO[3].base)
-#define USTORM_FLR_FINAL_ACK_SIZE			(IRO[3].size)
+#define USTORM_FLR_FINAL_ACK_OFFSET(pf_id) (IRO[4].base + ((pf_id) * IRO[4].m1))
+#define USTORM_FLR_FINAL_ACK_SIZE          (IRO[4].size)
 /* Ustorm Event ring consumer */
-#define USTORM_EQE_CONS_OFFSET(pf_id)			(IRO[4].base +	\
-							 ((pf_id) *	\
-							  IRO[4].m1))
-#define USTORM_EQE_CONS_SIZE				(IRO[4].size)
-/* Ustorm Completion ring consumer */
-#define USTORM_CQ_CONS_OFFSET(global_queue_id)		(IRO[5].base +	\
-							 ((global_queue_id) * \
-							  IRO[5].m1))
-#define USTORM_CQ_CONS_SIZE				(IRO[5].size)
+#define USTORM_EQE_CONS_OFFSET(pf_id)    (IRO[5].base +	((pf_id) * IRO[5].m1))
+#define USTORM_EQE_CONS_SIZE             (IRO[5].size)
+/* Ustorm Common Queue ring consumer */
+#define USTORM_COMMON_QUEUE_CONS_OFFSET(global_queue_id) \
+			(IRO[6].base + ((global_queue_id) * IRO[6].m1))
+#define USTORM_COMMON_QUEUE_CONS_SIZE    (IRO[6].size)
 /* Xstorm Integration Test Data */
-#define XSTORM_INTEG_TEST_DATA_OFFSET			(IRO[6].base)
-#define XSTORM_INTEG_TEST_DATA_SIZE			(IRO[6].size)
+#define XSTORM_INTEG_TEST_DATA_OFFSET    (IRO[7].base)
+#define XSTORM_INTEG_TEST_DATA_SIZE      (IRO[7].size)
 /* Ystorm Integration Test Data */
-#define YSTORM_INTEG_TEST_DATA_OFFSET			(IRO[7].base)
-#define YSTORM_INTEG_TEST_DATA_SIZE			(IRO[7].size)
+#define YSTORM_INTEG_TEST_DATA_OFFSET    (IRO[8].base)
+#define YSTORM_INTEG_TEST_DATA_SIZE      (IRO[8].size)
 /* Pstorm Integration Test Data */
-#define PSTORM_INTEG_TEST_DATA_OFFSET			(IRO[8].base)
-#define PSTORM_INTEG_TEST_DATA_SIZE			(IRO[8].size)
+#define PSTORM_INTEG_TEST_DATA_OFFSET    (IRO[9].base)
+#define PSTORM_INTEG_TEST_DATA_SIZE      (IRO[9].size)
 /* Tstorm Integration Test Data */
-#define TSTORM_INTEG_TEST_DATA_OFFSET			(IRO[9].base)
-#define TSTORM_INTEG_TEST_DATA_SIZE			(IRO[9].size)
+#define TSTORM_INTEG_TEST_DATA_OFFSET    (IRO[10].base)
+#define TSTORM_INTEG_TEST_DATA_SIZE      (IRO[10].size)
 /* Mstorm Integration Test Data */
-#define MSTORM_INTEG_TEST_DATA_OFFSET			(IRO[10].base)
-#define MSTORM_INTEG_TEST_DATA_SIZE			(IRO[10].size)
+#define MSTORM_INTEG_TEST_DATA_OFFSET    (IRO[11].base)
+#define MSTORM_INTEG_TEST_DATA_SIZE      (IRO[11].size)
 /* Ustorm Integration Test Data */
-#define USTORM_INTEG_TEST_DATA_OFFSET			(IRO[11].base)
-#define USTORM_INTEG_TEST_DATA_SIZE			(IRO[11].size)
+#define USTORM_INTEG_TEST_DATA_OFFSET    (IRO[12].base)
+#define USTORM_INTEG_TEST_DATA_SIZE      (IRO[12].size)
 /* Tstorm producers */
-#define TSTORM_LL2_RX_PRODS_OFFSET(core_rx_queue_id)	(IRO[12].base +	\
-							 ((core_rx_queue_id) * \
-							  IRO[12].m1))
-#define TSTORM_LL2_RX_PRODS_SIZE			(IRO[12].size)
-/* Tstorm LiteL2 queue statistics */
-#define CORE_LL2_TSTORM_PER_QUEUE_STAT_OFFSET(core_rx_q_id) (IRO[13].base + \
-							     ((core_rx_q_id) * \
-							      IRO[13].m1))
-#define CORE_LL2_TSTORM_PER_QUEUE_STAT_SIZE		(IRO[13].size)
+#define TSTORM_LL2_RX_PRODS_OFFSET(core_rx_queue_id) \
+			(IRO[13].base + ((core_rx_queue_id) * IRO[13].m1))
+#define TSTORM_LL2_RX_PRODS_SIZE         (IRO[13].size)
+/* Tstorm LightL2 queue statistics */
+#define CORE_LL2_TSTORM_PER_QUEUE_STAT_OFFSET(core_rx_queue_id) \
+			(IRO[14].base + ((core_rx_queue_id) * IRO[14].m1))
+#define CORE_LL2_TSTORM_PER_QUEUE_STAT_SIZE    (IRO[14].size)
 /* Ustorm LiteL2 queue statistics */
-#define CORE_LL2_USTORM_PER_QUEUE_STAT_OFFSET(core_rx_q_id) (IRO[14].base + \
-							     ((core_rx_q_id) * \
-							      IRO[14].m1))
-#define CORE_LL2_USTORM_PER_QUEUE_STAT_SIZE		(IRO[14].size)
+#define CORE_LL2_USTORM_PER_QUEUE_STAT_OFFSET(core_rx_queue_id) \
+			(IRO[15].base +	((core_rx_queue_id) * IRO[15].m1))
+#define CORE_LL2_USTORM_PER_QUEUE_STAT_SIZE    (IRO[15].size)
 /* Pstorm LiteL2 queue statistics */
-#define CORE_LL2_PSTORM_PER_QUEUE_STAT_OFFSET(core_txst_id) (IRO[15].base + \
-							     ((core_txst_id) * \
-							      IRO[15].m1))
-#define CORE_LL2_PSTORM_PER_QUEUE_STAT_SIZE		(IRO[15].size)
+#define CORE_LL2_PSTORM_PER_QUEUE_STAT_OFFSET(core_tx_stats_id) \
+			(IRO[16].base +	((core_tx_stats_id) * IRO[16].m1))
+#define CORE_LL2_PSTORM_PER_QUEUE_STAT_SIZE    (IRO[16].size)
 /* Mstorm queue statistics */
-#define MSTORM_QUEUE_STAT_OFFSET(stat_counter_id) (IRO[16].base + \
-						   ((stat_counter_id) *	\
-						    IRO[16].m1))
-#define MSTORM_QUEUE_STAT_SIZE				(IRO[16].size)
+#define MSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \
+			(IRO[17].base + ((stat_counter_id) * IRO[17].m1))
+#define MSTORM_QUEUE_STAT_SIZE                 (IRO[17].size)
 /* Mstorm producers */
-#define MSTORM_PRODS_OFFSET(queue_id)			(IRO[17].base +	\
-							 ((queue_id) *	\
-							  IRO[17].m1))
-#define MSTORM_PRODS_SIZE				(IRO[17].size)
+#define MSTORM_PRODS_OFFSET(queue_id) (IRO[18].base + ((queue_id) * IRO[18].m1))
+#define MSTORM_PRODS_SIZE             (IRO[18].size)
 /* TPA agregation timeout in us resolution (on ASIC) */
-#define MSTORM_TPA_TIMEOUT_US_OFFSET			(IRO[18].base)
-#define MSTORM_TPA_TIMEOUT_US_SIZE			(IRO[18].size)
+#define MSTORM_TPA_TIMEOUT_US_OFFSET  (IRO[19].base)
+#define MSTORM_TPA_TIMEOUT_US_SIZE    (IRO[19].size)
 /* Ustorm queue statistics */
-#define USTORM_QUEUE_STAT_OFFSET(stat_counter_id)	(IRO[19].base +	\
-							((stat_counter_id) * \
-							 IRO[19].m1))
-#define USTORM_QUEUE_STAT_SIZE				(IRO[19].size)
+#define USTORM_QUEUE_STAT_OFFSET(stat_counter_id) \
+			(IRO[20].base + ((stat_counter_id) * IRO[20].m1))
+#define USTORM_QUEUE_STAT_SIZE        (IRO[20].size)
 /* Ustorm queue zone */
-#define USTORM_ETH_QUEUE_ZONE_OFFSET(queue_id)		(IRO[20].base +	\
-							 ((queue_id) *	\
-							  IRO[20].m1))
-#define USTORM_ETH_QUEUE_ZONE_SIZE			(IRO[20].size)
+#define USTORM_ETH_QUEUE_ZONE_OFFSET(queue_id) \
+			(IRO[21].base +	((queue_id) * IRO[21].m1))
+#define USTORM_ETH_QUEUE_ZONE_SIZE    (IRO[21].size)
 /* Pstorm queue statistics */
-#define PSTORM_QUEUE_STAT_OFFSET(stat_counter_id)	(IRO[21].base +	\
-							 ((stat_counter_id) * \
-							  IRO[21].m1))
-#define PSTORM_QUEUE_STAT_SIZE				(IRO[21].size)
+#define PSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \
+		(IRO[22].base + ((stat_counter_id) * IRO[22].m1))
+#define PSTORM_QUEUE_STAT_SIZE        (IRO[22].size)
 /* Tstorm last parser message */
-#define TSTORM_ETH_PRS_INPUT_OFFSET(pf_id)		(IRO[22].base +	\
-							 ((pf_id) *	\
-							  IRO[22].m1))
-#define TSTORM_ETH_PRS_INPUT_SIZE			(IRO[22].size)
+#define TSTORM_ETH_PRS_INPUT_OFFSET  (IRO[23].base)
+#define TSTORM_ETH_PRS_INPUT_SIZE    (IRO[23].size)
+/* Tstorm Eth limit Rx rate */
+#define ETH_RX_RATE_LIMIT_OFFSET(pf_id) (IRO[24].base +	((pf_id) * IRO[24].m1))
+#define ETH_RX_RATE_LIMIT_SIZE       (IRO[24].size)
 /* Ystorm queue zone */
-#define YSTORM_ETH_QUEUE_ZONE_OFFSET(queue_id)		(IRO[23].base +	\
-							 ((queue_id) *	\
-							  IRO[23].m1))
-#define YSTORM_ETH_QUEUE_ZONE_SIZE			(IRO[23].size)
+#define YSTORM_ETH_QUEUE_ZONE_OFFSET(queue_id) \
+			(IRO[25].base +	((queue_id) * IRO[25].m1))
+#define YSTORM_ETH_QUEUE_ZONE_SIZE   (IRO[25].size)
 /* Ystorm cqe producer */
-#define YSTORM_TOE_CQ_PROD_OFFSET(rss_id)		(IRO[24].base +	\
-							 ((rss_id) *	\
-							  IRO[24].m1))
-#define YSTORM_TOE_CQ_PROD_SIZE				(IRO[24].size)
+#define YSTORM_TOE_CQ_PROD_OFFSET(rss_id) \
+			(IRO[26].base + ((rss_id) * IRO[26].m1))
+#define YSTORM_TOE_CQ_PROD_SIZE      (IRO[26].size)
 /* Ustorm cqe producer */
-#define USTORM_TOE_CQ_PROD_OFFSET(rss_id)		(IRO[25].base +	\
-							 ((rss_id) *	\
-							  IRO[25].m1))
-#define USTORM_TOE_CQ_PROD_SIZE				(IRO[25].size)
+#define USTORM_TOE_CQ_PROD_OFFSET(rss_id) \
+			(IRO[27].base + ((rss_id) * IRO[27].m1))
+#define USTORM_TOE_CQ_PROD_SIZE      (IRO[27].size)
 /* Ustorm grq producer */
-#define USTORM_TOE_GRQ_PROD_OFFSET(pf_id)		(IRO[26].base +	\
-							 ((pf_id) *	\
-							  IRO[26].m1))
-#define USTORM_TOE_GRQ_PROD_SIZE			(IRO[26].size)
+#define USTORM_TOE_GRQ_PROD_OFFSET(pf_id) \
+			(IRO[28].base + ((pf_id) * IRO[28].m1))
+#define USTORM_TOE_GRQ_PROD_SIZE     (IRO[28].size)
 /* Tstorm cmdq-cons of given command queue-id */
-#define TSTORM_SCSI_CMDQ_CONS_OFFSET(cmdq_queue_id)	(IRO[27].base +	\
-							 ((cmdq_queue_id) * \
-							  IRO[27].m1))
-#define TSTORM_SCSI_CMDQ_CONS_SIZE			(IRO[27].size)
+#define TSTORM_SCSI_CMDQ_CONS_OFFSET(cmdq_queue_id) \
+			(IRO[29].base + ((cmdq_queue_id) * IRO[29].m1))
+#define TSTORM_SCSI_CMDQ_CONS_SIZE   (IRO[29].size)
 /* Mstorm rq-cons of given queue-id */
-#define MSTORM_SCSI_RQ_CONS_OFFSET(rq_queue_id)		(IRO[28].base +	\
-							 ((rq_queue_id) * \
-							  IRO[28].m1))
-#define MSTORM_SCSI_RQ_CONS_SIZE			(IRO[28].size)
+#define MSTORM_SCSI_RQ_CONS_OFFSET(rq_queue_id) \
+		(IRO[30].base + ((rq_queue_id) * IRO[30].m1))
+#define MSTORM_SCSI_RQ_CONS_SIZE     (IRO[30].size)
+/* Mstorm bdq-external-producer of given BDQ function ID, BDqueue-id */
+#define MSTORM_SCSI_BDQ_EXT_PROD_OFFSET(func_id, bdq_id) \
+	(IRO[31].base + ((func_id) * IRO[31].m1) + ((bdq_id) * IRO[31].m2))
+#define MSTORM_SCSI_BDQ_EXT_PROD_SIZE (IRO[31].size)
+/* Tstorm (reflects M-Storm) bdq-external-producer of given fn ID, BDqueue-id */
+#define TSTORM_SCSI_BDQ_EXT_PROD_OFFSET(func_id, bdq_id) \
+	(IRO[32].base + ((func_id) * IRO[32].m1) + ((bdq_id) * IRO[32].m2))
+#define TSTORM_SCSI_BDQ_EXT_PROD_SIZE (IRO[32].size)
+/* Tstorm iSCSI RX stats */
+#define TSTORM_ISCSI_RX_STATS_OFFSET(pf_id) \
+				(IRO[33].base + ((pf_id) * IRO[33].m1))
+#define TSTORM_ISCSI_RX_STATS_SIZE    (IRO[33].size)
+/* Mstorm iSCSI RX stats */
+#define MSTORM_ISCSI_RX_STATS_OFFSET(pf_id) \
+				(IRO[34].base + ((pf_id) * IRO[34].m1))
+#define MSTORM_ISCSI_RX_STATS_SIZE    (IRO[34].size)
+/* Ustorm iSCSI RX stats */
+#define USTORM_ISCSI_RX_STATS_OFFSET(pf_id) \
+				(IRO[35].base +	((pf_id) * IRO[35].m1))
+#define USTORM_ISCSI_RX_STATS_SIZE    (IRO[35].size)
+/* Xstorm iSCSI TX stats */
+#define XSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \
+				(IRO[36].base +	((pf_id) * IRO[36].m1))
+#define XSTORM_ISCSI_TX_STATS_SIZE    (IRO[36].size)
+/* Ystorm iSCSI TX stats */
+#define YSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \
+				(IRO[37].base +	((pf_id) * IRO[37].m1))
+#define YSTORM_ISCSI_TX_STATS_SIZE    (IRO[37].size)
+/* Pstorm iSCSI TX stats */
+#define PSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \
+				(IRO[38].base +	((pf_id) * IRO[38].m1))
+#define PSTORM_ISCSI_TX_STATS_SIZE    (IRO[38].size)
+/* Tstorm FCoE RX stats */
+#define TSTORM_FCOE_RX_STATS_OFFSET(pf_id) \
+				(IRO[39].base +	((pf_id) * IRO[39].m1))
+#define TSTORM_FCOE_RX_STATS_SIZE      (IRO[39].size)
+/* Mstorm FCoE RX stats */
+#define MSTORM_FCOE_RX_STATS_OFFSET(pf_id) \
+				(IRO[40].base +	((pf_id) * IRO[40].m1))
+#define MSTORM_FCOE_RX_STATS_SIZE      (IRO[40].size)
+/* Pstorm FCoE TX stats */
+#define PSTORM_FCOE_TX_STATS_OFFSET(pf_id) \
+				(IRO[41].base +	((pf_id) * IRO[41].m1))
+#define PSTORM_FCOE_TX_STATS_SIZE      (IRO[41].size)
 /* Pstorm RoCE statistics */
-#define PSTORM_ROCE_STAT_OFFSET(stat_counter_id)	(IRO[29].base +	\
-							 ((stat_counter_id) * \
-							  IRO[29].m1))
-#define PSTORM_ROCE_STAT_SIZE				(IRO[29].size)
+#define PSTORM_ROCE_STAT_OFFSET(stat_counter_id) \
+			(IRO[42].base + ((stat_counter_id) * IRO[42].m1))
+#define PSTORM_ROCE_STAT_SIZE          (IRO[42].size)
 /* Tstorm RoCE statistics */
-#define TSTORM_ROCE_STAT_OFFSET(stat_counter_id)	(IRO[30].base +	\
-							 ((stat_counter_id) * \
-							  IRO[30].m1))
-#define TSTORM_ROCE_STAT_SIZE				(IRO[30].size)
-
-static const struct iro iro_arr[31] = {
-	{ 0x10,	  0x0,	 0x0,	0x0,   0x8     },
-	{ 0x4448, 0x60,	 0x0,	0x0,   0x60    },
-	{ 0x498,  0x8,	 0x0,	0x0,   0x4     },
-	{ 0x494,  0x0,	 0x0,	0x0,   0x4     },
-	{ 0x10,	  0x8,	 0x0,	0x0,   0x2     },
-	{ 0x90,	  0x8,	 0x0,	0x0,   0x2     },
-	{ 0x4540, 0x0,	 0x0,	0x0,   0xf8    },
-	{ 0x39e0, 0x0,	 0x0,	0x0,   0xf8    },
-	{ 0x2598, 0x0,	 0x0,	0x0,   0xf8    },
-	{ 0x4350, 0x0,	 0x0,	0x0,   0xf8    },
-	{ 0x52d0, 0x0,	 0x0,	0x0,   0xf8    },
-	{ 0x7a48, 0x0,	 0x0,	0x0,   0xf8    },
-	{ 0x100,  0x8,	 0x0,	0x0,   0x8     },
-	{ 0x5808, 0x10,	 0x0,	0x0,   0x10    },
-	{ 0xb100, 0x30,	 0x0,	0x0,   0x30    },
-	{ 0x95c0, 0x30,	 0x0,	0x0,   0x30    },
-	{ 0x54f8, 0x40,	 0x0,	0x0,   0x40    },
-	{ 0x200,  0x10,	 0x0,	0x0,   0x8     },
-	{ 0x9e70, 0x0,	 0x0,	0x0,   0x4     },
-	{ 0x7ca0, 0x40,	 0x0,	0x0,   0x30    },
-	{ 0xd00,  0x8,	 0x0,	0x0,   0x8     },
-	{ 0x2790, 0x80,	 0x0,	0x0,   0x38    },
-	{ 0xa520, 0xf0,	 0x0,	0x0,   0xf0    },
-	{ 0x80,	  0x8,	 0x0,	0x0,   0x8     },
-	{ 0xac0,  0x8,	 0x0,	0x0,   0x8     },
-	{ 0x2580, 0x8,	 0x0,	0x0,   0x8     },
-	{ 0x2500, 0x8,	 0x0,	0x0,   0x8     },
-	{ 0x440,  0x8,	 0x0,	0x0,   0x2     },
-	{ 0x1800, 0x8,	 0x0,	0x0,   0x2     },
-	{ 0x27c8, 0x80,	 0x0,	0x0,   0x10    },
-	{ 0x4710, 0x10,	 0x0,	0x0,   0x10    },
+#define TSTORM_ROCE_STAT_OFFSET(stat_counter_id) \
+			(IRO[43].base + ((stat_counter_id) * IRO[43].m1))
+#define TSTORM_ROCE_STAT_SIZE          (IRO[43].size)
+
+static const struct iro iro_arr[44] = {
+	{ 0x10,	   0x0,	   0x0,	   0x0,	   0x8	    },
+	{ 0x47c8,  0x60,   0x0,	   0x0,	   0x60	    },
+	{ 0x5e30,  0x20,   0x0,	   0x0,	   0x20	    },
+	{ 0x510,   0x8,	   0x0,	   0x0,	   0x4	    },
+	{ 0x490,   0x8,	   0x0,	   0x0,	   0x4	    },
+	{ 0x10,	   0x8,	   0x0,	   0x0,	   0x2	    },
+	{ 0x90,	   0x8,	   0x0,	   0x0,	   0x2	    },
+	{ 0x4940,  0x0,	   0x0,	   0x0,	   0x78	    },
+	{ 0x3de0,  0x0,	   0x0,	   0x0,	   0x78	    },
+	{ 0x2998,  0x0,	   0x0,	   0x0,	   0x78	    },
+	{ 0x4750,  0x0,	   0x0,	   0x0,	   0x78	    },
+	{ 0x56d0,  0x0,	   0x0,	   0x0,	   0x78	    },
+	{ 0x7e50,  0x0,	   0x0,	   0x0,	   0x78	    },
+	{ 0x100,   0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0x5c10,  0x10,   0x0,	   0x0,	   0x10	    },
+	{ 0xb508,  0x30,   0x0,	   0x0,	   0x30	    },
+	{ 0x95c0,  0x30,   0x0,	   0x0,	   0x30	    },
+	{ 0x58a0,  0x40,   0x0,	   0x0,	   0x40	    },
+	{ 0x200,   0x10,   0x0,	   0x0,	   0x8	    },
+	{ 0xa230,  0x0,	   0x0,	   0x0,	   0x4	    },
+	{ 0x8058,  0x40,   0x0,	   0x0,	   0x30	    },
+	{ 0xd00,   0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0x2b30,  0x80,   0x0,	   0x0,	   0x38	    },
+	{ 0xa808,  0x0,	   0x0,	   0x0,	   0xf0	    },
+	{ 0xa8f8,  0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0x80,	   0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0xac0,   0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0x2580,  0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0x2500,  0x8,	   0x0,	   0x0,	   0x8	    },
+	{ 0x440,   0x8,	   0x0,	   0x0,	   0x2	    },
+	{ 0x1800,  0x8,	   0x0,	   0x0,	   0x2	    },
+	{ 0x1a00,  0x10,   0x8,	   0x0,	   0x2	    },
+	{ 0x640,   0x10,   0x8,	   0x0,	   0x2	    },
+	{ 0xd9b8,  0x38,   0x0,	   0x0,	   0x24	    },
+	{ 0x11048, 0x10,   0x0,	   0x0,	   0x8	    },
+	{ 0x11678, 0x38,   0x0,	   0x0,	   0x18	    },
+	{ 0xaec0,  0x30,   0x0,	   0x0,	   0x10	    },
+	{ 0x8700,  0x28,   0x0,	   0x0,	   0x18	    },
+	{ 0xec00,  0x10,   0x0,	   0x0,	   0x10	    },
+	{ 0xde38,  0x40,   0x0,	   0x0,	   0x30	    },
+	{ 0x121a8, 0x38,   0x0,	   0x0,	   0x8	    },
+	{ 0xf068,  0x20,   0x0,	   0x0,	   0x20	    },
+	{ 0x2b68,  0x80,   0x0,	   0x0,	   0x10	    },
+	{ 0x4ab8,  0x10,   0x0,	   0x0,	   0x10	    },
 };
 
 /* Runtime array offsets */
@@ -1866,426 +1831,427 @@ static const struct iro iro_arr[31] = {
 #define DORQ_REG_VF_MAX_ICID_6_RT_OFFSET                                14
 #define DORQ_REG_VF_MAX_ICID_7_RT_OFFSET                                15
 #define DORQ_REG_PF_WAKE_ALL_RT_OFFSET                                  16
-#define IGU_REG_PF_CONFIGURATION_RT_OFFSET                              17
-#define IGU_REG_VF_CONFIGURATION_RT_OFFSET                              18
-#define IGU_REG_ATTN_MSG_ADDR_L_RT_OFFSET                               19
-#define IGU_REG_ATTN_MSG_ADDR_H_RT_OFFSET                               20
-#define IGU_REG_LEADING_EDGE_LATCH_RT_OFFSET                            21
-#define IGU_REG_TRAILING_EDGE_LATCH_RT_OFFSET                           22
-#define CAU_REG_CQE_AGG_UNIT_SIZE_RT_OFFSET                             23
-#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET                                 760
+#define DORQ_REG_TAG1_ETHERTYPE_RT_OFFSET                               17
+#define IGU_REG_PF_CONFIGURATION_RT_OFFSET                              18
+#define IGU_REG_VF_CONFIGURATION_RT_OFFSET                              19
+#define IGU_REG_ATTN_MSG_ADDR_L_RT_OFFSET                               20
+#define IGU_REG_ATTN_MSG_ADDR_H_RT_OFFSET                               21
+#define IGU_REG_LEADING_EDGE_LATCH_RT_OFFSET                            22
+#define IGU_REG_TRAILING_EDGE_LATCH_RT_OFFSET                           23
+#define CAU_REG_CQE_AGG_UNIT_SIZE_RT_OFFSET                             24
+#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET                                 761
 #define CAU_REG_SB_VAR_MEMORY_RT_SIZE                                   736
-#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET                                 760
+#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET                                 761
 #define CAU_REG_SB_VAR_MEMORY_RT_SIZE                                   736
-#define CAU_REG_SB_ADDR_MEMORY_RT_OFFSET                                1496
+#define CAU_REG_SB_ADDR_MEMORY_RT_OFFSET                                1497
 #define CAU_REG_SB_ADDR_MEMORY_RT_SIZE                                  736
-#define CAU_REG_PI_MEMORY_RT_OFFSET                                     2232
+#define CAU_REG_PI_MEMORY_RT_OFFSET                                     2233
 #define CAU_REG_PI_MEMORY_RT_SIZE                                       4416
-#define PRS_REG_SEARCH_RESP_INITIATOR_TYPE_RT_OFFSET                    6648
-#define PRS_REG_TASK_ID_MAX_INITIATOR_PF_RT_OFFSET                      6649
-#define PRS_REG_TASK_ID_MAX_INITIATOR_VF_RT_OFFSET                      6650
-#define PRS_REG_TASK_ID_MAX_TARGET_PF_RT_OFFSET                         6651
-#define PRS_REG_TASK_ID_MAX_TARGET_VF_RT_OFFSET                         6652
-#define PRS_REG_SEARCH_TCP_RT_OFFSET                                    6653
-#define PRS_REG_SEARCH_FCOE_RT_OFFSET                                   6654
-#define PRS_REG_SEARCH_ROCE_RT_OFFSET                                   6655
-#define PRS_REG_ROCE_DEST_QP_MAX_VF_RT_OFFSET                           6656
-#define PRS_REG_ROCE_DEST_QP_MAX_PF_RT_OFFSET                           6657
-#define PRS_REG_SEARCH_OPENFLOW_RT_OFFSET                               6658
-#define PRS_REG_SEARCH_NON_IP_AS_OPENFLOW_RT_OFFSET                     6659
-#define PRS_REG_OPENFLOW_SUPPORT_ONLY_KNOWN_OVER_IP_RT_OFFSET           6660
-#define PRS_REG_OPENFLOW_SEARCH_KEY_MASK_RT_OFFSET                      6661
-#define PRS_REG_LIGHT_L2_ETHERTYPE_EN_RT_OFFSET                         6662
-#define SRC_REG_FIRSTFREE_RT_OFFSET                                     6663
+#define PRS_REG_SEARCH_RESP_INITIATOR_TYPE_RT_OFFSET                    6649
+#define PRS_REG_TASK_ID_MAX_INITIATOR_PF_RT_OFFSET                      6650
+#define PRS_REG_TASK_ID_MAX_INITIATOR_VF_RT_OFFSET                      6651
+#define PRS_REG_TASK_ID_MAX_TARGET_PF_RT_OFFSET                         6652
+#define PRS_REG_TASK_ID_MAX_TARGET_VF_RT_OFFSET                         6653
+#define PRS_REG_SEARCH_TCP_RT_OFFSET                                    6654
+#define PRS_REG_SEARCH_FCOE_RT_OFFSET                                   6655
+#define PRS_REG_SEARCH_ROCE_RT_OFFSET                                   6656
+#define PRS_REG_ROCE_DEST_QP_MAX_VF_RT_OFFSET                           6657
+#define PRS_REG_ROCE_DEST_QP_MAX_PF_RT_OFFSET                           6658
+#define PRS_REG_SEARCH_OPENFLOW_RT_OFFSET                               6659
+#define PRS_REG_SEARCH_NON_IP_AS_OPENFLOW_RT_OFFSET                     6660
+#define PRS_REG_OPENFLOW_SUPPORT_ONLY_KNOWN_OVER_IP_RT_OFFSET           6661
+#define PRS_REG_OPENFLOW_SEARCH_KEY_MASK_RT_OFFSET                      6662
+#define PRS_REG_TAG_ETHERTYPE_0_RT_OFFSET                               6663
+#define PRS_REG_LIGHT_L2_ETHERTYPE_EN_RT_OFFSET                         6664
+#define SRC_REG_FIRSTFREE_RT_OFFSET                                     6665
 #define SRC_REG_FIRSTFREE_RT_SIZE                                       2
-#define SRC_REG_LASTFREE_RT_OFFSET                                      6665
+#define SRC_REG_LASTFREE_RT_OFFSET                                      6667
 #define SRC_REG_LASTFREE_RT_SIZE                                        2
-#define SRC_REG_COUNTFREE_RT_OFFSET                                     6667
-#define SRC_REG_NUMBER_HASH_BITS_RT_OFFSET                              6668
-#define PSWRQ2_REG_CDUT_P_SIZE_RT_OFFSET                                6669
-#define PSWRQ2_REG_CDUC_P_SIZE_RT_OFFSET                                6670
-#define PSWRQ2_REG_TM_P_SIZE_RT_OFFSET                                  6671
-#define PSWRQ2_REG_QM_P_SIZE_RT_OFFSET                                  6672
-#define PSWRQ2_REG_SRC_P_SIZE_RT_OFFSET                                 6673
-#define PSWRQ2_REG_TM_FIRST_ILT_RT_OFFSET                               6674
-#define PSWRQ2_REG_TM_LAST_ILT_RT_OFFSET                                6675
-#define PSWRQ2_REG_QM_FIRST_ILT_RT_OFFSET                               6676
-#define PSWRQ2_REG_QM_LAST_ILT_RT_OFFSET                                6677
-#define PSWRQ2_REG_SRC_FIRST_ILT_RT_OFFSET                              6678
-#define PSWRQ2_REG_SRC_LAST_ILT_RT_OFFSET                               6679
-#define PSWRQ2_REG_CDUC_FIRST_ILT_RT_OFFSET                             6680
-#define PSWRQ2_REG_CDUC_LAST_ILT_RT_OFFSET                              6681
-#define PSWRQ2_REG_CDUT_FIRST_ILT_RT_OFFSET                             6682
-#define PSWRQ2_REG_CDUT_LAST_ILT_RT_OFFSET                              6683
-#define PSWRQ2_REG_TSDM_FIRST_ILT_RT_OFFSET                             6684
-#define PSWRQ2_REG_TSDM_LAST_ILT_RT_OFFSET                              6685
-#define PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET                     6686
-#define PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET                   6687
-#define PSWRQ2_REG_CDUC_NUMBER_OF_PF_BLOCKS_RT_OFFSET                   6688
-#define PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET                               6689
-#define PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET                             6690
-#define PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET                             6691
-#define PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET                           6692
-#define PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET                         6693
-#define PSWRQ2_REG_CDUC_BLOCKS_FACTOR_RT_OFFSET                         6694
-#define PSWRQ2_REG_VF_BASE_RT_OFFSET                                    6695
-#define PSWRQ2_REG_VF_LAST_ILT_RT_OFFSET                                6696
-#define PSWRQ2_REG_WR_MBS0_RT_OFFSET                                    6697
-#define PSWRQ2_REG_RD_MBS0_RT_OFFSET                                    6698
-#define PSWRQ2_REG_DRAM_ALIGN_WR_RT_OFFSET                              6699
-#define PSWRQ2_REG_DRAM_ALIGN_RD_RT_OFFSET                              6700
-#define PSWRQ2_REG_ILT_MEMORY_RT_OFFSET                                 6701
+#define SRC_REG_COUNTFREE_RT_OFFSET                                     6669
+#define SRC_REG_NUMBER_HASH_BITS_RT_OFFSET                              6670
+#define PSWRQ2_REG_CDUT_P_SIZE_RT_OFFSET                                6671
+#define PSWRQ2_REG_CDUC_P_SIZE_RT_OFFSET                                6672
+#define PSWRQ2_REG_TM_P_SIZE_RT_OFFSET                                  6673
+#define PSWRQ2_REG_QM_P_SIZE_RT_OFFSET                                  6674
+#define PSWRQ2_REG_SRC_P_SIZE_RT_OFFSET                                 6675
+#define PSWRQ2_REG_TM_FIRST_ILT_RT_OFFSET                               6676
+#define PSWRQ2_REG_TM_LAST_ILT_RT_OFFSET                                6677
+#define PSWRQ2_REG_QM_FIRST_ILT_RT_OFFSET                               6678
+#define PSWRQ2_REG_QM_LAST_ILT_RT_OFFSET                                6679
+#define PSWRQ2_REG_SRC_FIRST_ILT_RT_OFFSET                              6680
+#define PSWRQ2_REG_SRC_LAST_ILT_RT_OFFSET                               6681
+#define PSWRQ2_REG_CDUC_FIRST_ILT_RT_OFFSET                             6682
+#define PSWRQ2_REG_CDUC_LAST_ILT_RT_OFFSET                              6683
+#define PSWRQ2_REG_CDUT_FIRST_ILT_RT_OFFSET                             6684
+#define PSWRQ2_REG_CDUT_LAST_ILT_RT_OFFSET                              6685
+#define PSWRQ2_REG_TSDM_FIRST_ILT_RT_OFFSET                             6686
+#define PSWRQ2_REG_TSDM_LAST_ILT_RT_OFFSET                              6687
+#define PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET                     6688
+#define PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET                   6689
+#define PSWRQ2_REG_CDUC_NUMBER_OF_PF_BLOCKS_RT_OFFSET                   6690
+#define PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET                               6691
+#define PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET                             6692
+#define PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET                             6693
+#define PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET                           6694
+#define PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET                         6695
+#define PSWRQ2_REG_CDUC_BLOCKS_FACTOR_RT_OFFSET                         6696
+#define PSWRQ2_REG_VF_BASE_RT_OFFSET                                    6697
+#define PSWRQ2_REG_VF_LAST_ILT_RT_OFFSET                                6698
+#define PSWRQ2_REG_WR_MBS0_RT_OFFSET                                    6699
+#define PSWRQ2_REG_RD_MBS0_RT_OFFSET                                    6700
+#define PSWRQ2_REG_DRAM_ALIGN_WR_RT_OFFSET                              6701
+#define PSWRQ2_REG_DRAM_ALIGN_RD_RT_OFFSET                              6702
+#define PSWRQ2_REG_ILT_MEMORY_RT_OFFSET                                 6703
 #define PSWRQ2_REG_ILT_MEMORY_RT_SIZE                                   22000
-#define PGLUE_REG_B_VF_BASE_RT_OFFSET                                   28701
-#define PGLUE_REG_B_CACHE_LINE_SIZE_RT_OFFSET                           28702
-#define PGLUE_REG_B_PF_BAR0_SIZE_RT_OFFSET                              28703
-#define PGLUE_REG_B_PF_BAR1_SIZE_RT_OFFSET                              28704
-#define PGLUE_REG_B_VF_BAR1_SIZE_RT_OFFSET                              28705
-#define TM_REG_VF_ENABLE_CONN_RT_OFFSET                                 28706
-#define TM_REG_PF_ENABLE_CONN_RT_OFFSET                                 28707
-#define TM_REG_PF_ENABLE_TASK_RT_OFFSET                                 28708
-#define TM_REG_GROUP_SIZE_RESOLUTION_CONN_RT_OFFSET                     28709
-#define TM_REG_GROUP_SIZE_RESOLUTION_TASK_RT_OFFSET                     28710
-#define TM_REG_CONFIG_CONN_MEM_RT_OFFSET                                28711
+#define PGLUE_REG_B_VF_BASE_RT_OFFSET                                   28703
+#define PGLUE_REG_B_CACHE_LINE_SIZE_RT_OFFSET                           28704
+#define PGLUE_REG_B_PF_BAR0_SIZE_RT_OFFSET                              28705
+#define PGLUE_REG_B_PF_BAR1_SIZE_RT_OFFSET                              28706
+#define PGLUE_REG_B_VF_BAR1_SIZE_RT_OFFSET                              28707
+#define TM_REG_VF_ENABLE_CONN_RT_OFFSET                                 28708
+#define TM_REG_PF_ENABLE_CONN_RT_OFFSET                                 28709
+#define TM_REG_PF_ENABLE_TASK_RT_OFFSET                                 28710
+#define TM_REG_GROUP_SIZE_RESOLUTION_CONN_RT_OFFSET                     28711
+#define TM_REG_GROUP_SIZE_RESOLUTION_TASK_RT_OFFSET                     28712
+#define TM_REG_CONFIG_CONN_MEM_RT_OFFSET                                28713
 #define TM_REG_CONFIG_CONN_MEM_RT_SIZE                                  416
-#define TM_REG_CONFIG_TASK_MEM_RT_OFFSET                                29127
+#define TM_REG_CONFIG_TASK_MEM_RT_OFFSET                                29129
 #define TM_REG_CONFIG_TASK_MEM_RT_SIZE                                  512
-#define QM_REG_MAXPQSIZE_0_RT_OFFSET                                    29639
-#define QM_REG_MAXPQSIZE_1_RT_OFFSET                                    29640
-#define QM_REG_MAXPQSIZE_2_RT_OFFSET                                    29641
-#define QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET                               29642
-#define QM_REG_MAXPQSIZETXSEL_1_RT_OFFSET                               29643
-#define QM_REG_MAXPQSIZETXSEL_2_RT_OFFSET                               29644
-#define QM_REG_MAXPQSIZETXSEL_3_RT_OFFSET                               29645
-#define QM_REG_MAXPQSIZETXSEL_4_RT_OFFSET                               29646
-#define QM_REG_MAXPQSIZETXSEL_5_RT_OFFSET                               29647
-#define QM_REG_MAXPQSIZETXSEL_6_RT_OFFSET                               29648
-#define QM_REG_MAXPQSIZETXSEL_7_RT_OFFSET                               29649
-#define QM_REG_MAXPQSIZETXSEL_8_RT_OFFSET                               29650
-#define QM_REG_MAXPQSIZETXSEL_9_RT_OFFSET                               29651
-#define QM_REG_MAXPQSIZETXSEL_10_RT_OFFSET                              29652
-#define QM_REG_MAXPQSIZETXSEL_11_RT_OFFSET                              29653
-#define QM_REG_MAXPQSIZETXSEL_12_RT_OFFSET                              29654
-#define QM_REG_MAXPQSIZETXSEL_13_RT_OFFSET                              29655
-#define QM_REG_MAXPQSIZETXSEL_14_RT_OFFSET                              29656
-#define QM_REG_MAXPQSIZETXSEL_15_RT_OFFSET                              29657
-#define QM_REG_MAXPQSIZETXSEL_16_RT_OFFSET                              29658
-#define QM_REG_MAXPQSIZETXSEL_17_RT_OFFSET                              29659
-#define QM_REG_MAXPQSIZETXSEL_18_RT_OFFSET                              29660
-#define QM_REG_MAXPQSIZETXSEL_19_RT_OFFSET                              29661
-#define QM_REG_MAXPQSIZETXSEL_20_RT_OFFSET                              29662
-#define QM_REG_MAXPQSIZETXSEL_21_RT_OFFSET                              29663
-#define QM_REG_MAXPQSIZETXSEL_22_RT_OFFSET                              29664
-#define QM_REG_MAXPQSIZETXSEL_23_RT_OFFSET                              29665
-#define QM_REG_MAXPQSIZETXSEL_24_RT_OFFSET                              29666
-#define QM_REG_MAXPQSIZETXSEL_25_RT_OFFSET                              29667
-#define QM_REG_MAXPQSIZETXSEL_26_RT_OFFSET                              29668
-#define QM_REG_MAXPQSIZETXSEL_27_RT_OFFSET                              29669
-#define QM_REG_MAXPQSIZETXSEL_28_RT_OFFSET                              29670
-#define QM_REG_MAXPQSIZETXSEL_29_RT_OFFSET                              29671
-#define QM_REG_MAXPQSIZETXSEL_30_RT_OFFSET                              29672
-#define QM_REG_MAXPQSIZETXSEL_31_RT_OFFSET                              29673
-#define QM_REG_MAXPQSIZETXSEL_32_RT_OFFSET                              29674
-#define QM_REG_MAXPQSIZETXSEL_33_RT_OFFSET                              29675
-#define QM_REG_MAXPQSIZETXSEL_34_RT_OFFSET                              29676
-#define QM_REG_MAXPQSIZETXSEL_35_RT_OFFSET                              29677
-#define QM_REG_MAXPQSIZETXSEL_36_RT_OFFSET                              29678
-#define QM_REG_MAXPQSIZETXSEL_37_RT_OFFSET                              29679
-#define QM_REG_MAXPQSIZETXSEL_38_RT_OFFSET                              29680
-#define QM_REG_MAXPQSIZETXSEL_39_RT_OFFSET                              29681
-#define QM_REG_MAXPQSIZETXSEL_40_RT_OFFSET                              29682
-#define QM_REG_MAXPQSIZETXSEL_41_RT_OFFSET                              29683
-#define QM_REG_MAXPQSIZETXSEL_42_RT_OFFSET                              29684
-#define QM_REG_MAXPQSIZETXSEL_43_RT_OFFSET                              29685
-#define QM_REG_MAXPQSIZETXSEL_44_RT_OFFSET                              29686
-#define QM_REG_MAXPQSIZETXSEL_45_RT_OFFSET                              29687
-#define QM_REG_MAXPQSIZETXSEL_46_RT_OFFSET                              29688
-#define QM_REG_MAXPQSIZETXSEL_47_RT_OFFSET                              29689
-#define QM_REG_MAXPQSIZETXSEL_48_RT_OFFSET                              29690
-#define QM_REG_MAXPQSIZETXSEL_49_RT_OFFSET                              29691
-#define QM_REG_MAXPQSIZETXSEL_50_RT_OFFSET                              29692
-#define QM_REG_MAXPQSIZETXSEL_51_RT_OFFSET                              29693
-#define QM_REG_MAXPQSIZETXSEL_52_RT_OFFSET                              29694
-#define QM_REG_MAXPQSIZETXSEL_53_RT_OFFSET                              29695
-#define QM_REG_MAXPQSIZETXSEL_54_RT_OFFSET                              29696
-#define QM_REG_MAXPQSIZETXSEL_55_RT_OFFSET                              29697
-#define QM_REG_MAXPQSIZETXSEL_56_RT_OFFSET                              29698
-#define QM_REG_MAXPQSIZETXSEL_57_RT_OFFSET                              29699
-#define QM_REG_MAXPQSIZETXSEL_58_RT_OFFSET                              29700
-#define QM_REG_MAXPQSIZETXSEL_59_RT_OFFSET                              29701
-#define QM_REG_MAXPQSIZETXSEL_60_RT_OFFSET                              29702
-#define QM_REG_MAXPQSIZETXSEL_61_RT_OFFSET                              29703
-#define QM_REG_MAXPQSIZETXSEL_62_RT_OFFSET                              29704
-#define QM_REG_MAXPQSIZETXSEL_63_RT_OFFSET                              29705
-#define QM_REG_BASEADDROTHERPQ_RT_OFFSET                                29706
+#define QM_REG_MAXPQSIZE_0_RT_OFFSET                                    29641
+#define QM_REG_MAXPQSIZE_1_RT_OFFSET                                    29642
+#define QM_REG_MAXPQSIZE_2_RT_OFFSET                                    29643
+#define QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET                               29644
+#define QM_REG_MAXPQSIZETXSEL_1_RT_OFFSET                               29645
+#define QM_REG_MAXPQSIZETXSEL_2_RT_OFFSET                               29646
+#define QM_REG_MAXPQSIZETXSEL_3_RT_OFFSET                               29647
+#define QM_REG_MAXPQSIZETXSEL_4_RT_OFFSET                               29648
+#define QM_REG_MAXPQSIZETXSEL_5_RT_OFFSET                               29649
+#define QM_REG_MAXPQSIZETXSEL_6_RT_OFFSET                               29650
+#define QM_REG_MAXPQSIZETXSEL_7_RT_OFFSET                               29651
+#define QM_REG_MAXPQSIZETXSEL_8_RT_OFFSET                               29652
+#define QM_REG_MAXPQSIZETXSEL_9_RT_OFFSET                               29653
+#define QM_REG_MAXPQSIZETXSEL_10_RT_OFFSET                              29654
+#define QM_REG_MAXPQSIZETXSEL_11_RT_OFFSET                              29655
+#define QM_REG_MAXPQSIZETXSEL_12_RT_OFFSET                              29656
+#define QM_REG_MAXPQSIZETXSEL_13_RT_OFFSET                              29657
+#define QM_REG_MAXPQSIZETXSEL_14_RT_OFFSET                              29658
+#define QM_REG_MAXPQSIZETXSEL_15_RT_OFFSET                              29659
+#define QM_REG_MAXPQSIZETXSEL_16_RT_OFFSET                              29660
+#define QM_REG_MAXPQSIZETXSEL_17_RT_OFFSET                              29661
+#define QM_REG_MAXPQSIZETXSEL_18_RT_OFFSET                              29662
+#define QM_REG_MAXPQSIZETXSEL_19_RT_OFFSET                              29663
+#define QM_REG_MAXPQSIZETXSEL_20_RT_OFFSET                              29664
+#define QM_REG_MAXPQSIZETXSEL_21_RT_OFFSET                              29665
+#define QM_REG_MAXPQSIZETXSEL_22_RT_OFFSET                              29666
+#define QM_REG_MAXPQSIZETXSEL_23_RT_OFFSET                              29667
+#define QM_REG_MAXPQSIZETXSEL_24_RT_OFFSET                              29668
+#define QM_REG_MAXPQSIZETXSEL_25_RT_OFFSET                              29669
+#define QM_REG_MAXPQSIZETXSEL_26_RT_OFFSET                              29670
+#define QM_REG_MAXPQSIZETXSEL_27_RT_OFFSET                              29671
+#define QM_REG_MAXPQSIZETXSEL_28_RT_OFFSET                              29672
+#define QM_REG_MAXPQSIZETXSEL_29_RT_OFFSET                              29673
+#define QM_REG_MAXPQSIZETXSEL_30_RT_OFFSET                              29674
+#define QM_REG_MAXPQSIZETXSEL_31_RT_OFFSET                              29675
+#define QM_REG_MAXPQSIZETXSEL_32_RT_OFFSET                              29676
+#define QM_REG_MAXPQSIZETXSEL_33_RT_OFFSET                              29677
+#define QM_REG_MAXPQSIZETXSEL_34_RT_OFFSET                              29678
+#define QM_REG_MAXPQSIZETXSEL_35_RT_OFFSET                              29679
+#define QM_REG_MAXPQSIZETXSEL_36_RT_OFFSET                              29680
+#define QM_REG_MAXPQSIZETXSEL_37_RT_OFFSET                              29681
+#define QM_REG_MAXPQSIZETXSEL_38_RT_OFFSET                              29682
+#define QM_REG_MAXPQSIZETXSEL_39_RT_OFFSET                              29683
+#define QM_REG_MAXPQSIZETXSEL_40_RT_OFFSET                              29684
+#define QM_REG_MAXPQSIZETXSEL_41_RT_OFFSET                              29685
+#define QM_REG_MAXPQSIZETXSEL_42_RT_OFFSET                              29686
+#define QM_REG_MAXPQSIZETXSEL_43_RT_OFFSET                              29687
+#define QM_REG_MAXPQSIZETXSEL_44_RT_OFFSET                              29688
+#define QM_REG_MAXPQSIZETXSEL_45_RT_OFFSET                              29689
+#define QM_REG_MAXPQSIZETXSEL_46_RT_OFFSET                              29690
+#define QM_REG_MAXPQSIZETXSEL_47_RT_OFFSET                              29691
+#define QM_REG_MAXPQSIZETXSEL_48_RT_OFFSET                              29692
+#define QM_REG_MAXPQSIZETXSEL_49_RT_OFFSET                              29693
+#define QM_REG_MAXPQSIZETXSEL_50_RT_OFFSET                              29694
+#define QM_REG_MAXPQSIZETXSEL_51_RT_OFFSET                              29695
+#define QM_REG_MAXPQSIZETXSEL_52_RT_OFFSET                              29696
+#define QM_REG_MAXPQSIZETXSEL_53_RT_OFFSET                              29697
+#define QM_REG_MAXPQSIZETXSEL_54_RT_OFFSET                              29698
+#define QM_REG_MAXPQSIZETXSEL_55_RT_OFFSET                              29699
+#define QM_REG_MAXPQSIZETXSEL_56_RT_OFFSET                              29700
+#define QM_REG_MAXPQSIZETXSEL_57_RT_OFFSET                              29701
+#define QM_REG_MAXPQSIZETXSEL_58_RT_OFFSET                              29702
+#define QM_REG_MAXPQSIZETXSEL_59_RT_OFFSET                              29703
+#define QM_REG_MAXPQSIZETXSEL_60_RT_OFFSET                              29704
+#define QM_REG_MAXPQSIZETXSEL_61_RT_OFFSET                              29705
+#define QM_REG_MAXPQSIZETXSEL_62_RT_OFFSET                              29706
+#define QM_REG_MAXPQSIZETXSEL_63_RT_OFFSET                              29707
+#define QM_REG_BASEADDROTHERPQ_RT_OFFSET                                29708
 #define QM_REG_BASEADDROTHERPQ_RT_SIZE                                  128
-#define QM_REG_VOQCRDLINE_RT_OFFSET                                     29834
+#define QM_REG_VOQCRDLINE_RT_OFFSET                                     29836
 #define QM_REG_VOQCRDLINE_RT_SIZE                                       20
-#define QM_REG_VOQINITCRDLINE_RT_OFFSET                                 29854
+#define QM_REG_VOQINITCRDLINE_RT_OFFSET                                 29856
 #define QM_REG_VOQINITCRDLINE_RT_SIZE                                   20
-#define QM_REG_AFULLQMBYPTHRPFWFQ_RT_OFFSET                             29874
-#define QM_REG_AFULLQMBYPTHRVPWFQ_RT_OFFSET                             29875
-#define QM_REG_AFULLQMBYPTHRPFRL_RT_OFFSET                              29876
-#define QM_REG_AFULLQMBYPTHRGLBLRL_RT_OFFSET                            29877
-#define QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET                           29878
-#define QM_REG_WRROTHERPQGRP_0_RT_OFFSET                                29879
-#define QM_REG_WRROTHERPQGRP_1_RT_OFFSET                                29880
-#define QM_REG_WRROTHERPQGRP_2_RT_OFFSET                                29881
-#define QM_REG_WRROTHERPQGRP_3_RT_OFFSET                                29882
-#define QM_REG_WRROTHERPQGRP_4_RT_OFFSET                                29883
-#define QM_REG_WRROTHERPQGRP_5_RT_OFFSET                                29884
-#define QM_REG_WRROTHERPQGRP_6_RT_OFFSET                                29885
-#define QM_REG_WRROTHERPQGRP_7_RT_OFFSET                                29886
-#define QM_REG_WRROTHERPQGRP_8_RT_OFFSET                                29887
-#define QM_REG_WRROTHERPQGRP_9_RT_OFFSET                                29888
-#define QM_REG_WRROTHERPQGRP_10_RT_OFFSET                               29889
-#define QM_REG_WRROTHERPQGRP_11_RT_OFFSET                               29890
-#define QM_REG_WRROTHERPQGRP_12_RT_OFFSET                               29891
-#define QM_REG_WRROTHERPQGRP_13_RT_OFFSET                               29892
-#define QM_REG_WRROTHERPQGRP_14_RT_OFFSET                               29893
-#define QM_REG_WRROTHERPQGRP_15_RT_OFFSET                               29894
-#define QM_REG_WRROTHERGRPWEIGHT_0_RT_OFFSET                            29895
-#define QM_REG_WRROTHERGRPWEIGHT_1_RT_OFFSET                            29896
-#define QM_REG_WRROTHERGRPWEIGHT_2_RT_OFFSET                            29897
-#define QM_REG_WRROTHERGRPWEIGHT_3_RT_OFFSET                            29898
-#define QM_REG_WRRTXGRPWEIGHT_0_RT_OFFSET                               29899
-#define QM_REG_WRRTXGRPWEIGHT_1_RT_OFFSET                               29900
-#define QM_REG_PQTX2PF_0_RT_OFFSET                                      29901
-#define QM_REG_PQTX2PF_1_RT_OFFSET                                      29902
-#define QM_REG_PQTX2PF_2_RT_OFFSET                                      29903
-#define QM_REG_PQTX2PF_3_RT_OFFSET                                      29904
-#define QM_REG_PQTX2PF_4_RT_OFFSET                                      29905
-#define QM_REG_PQTX2PF_5_RT_OFFSET                                      29906
-#define QM_REG_PQTX2PF_6_RT_OFFSET                                      29907
-#define QM_REG_PQTX2PF_7_RT_OFFSET                                      29908
-#define QM_REG_PQTX2PF_8_RT_OFFSET                                      29909
-#define QM_REG_PQTX2PF_9_RT_OFFSET                                      29910
-#define QM_REG_PQTX2PF_10_RT_OFFSET                                     29911
-#define QM_REG_PQTX2PF_11_RT_OFFSET                                     29912
-#define QM_REG_PQTX2PF_12_RT_OFFSET                                     29913
-#define QM_REG_PQTX2PF_13_RT_OFFSET                                     29914
-#define QM_REG_PQTX2PF_14_RT_OFFSET                                     29915
-#define QM_REG_PQTX2PF_15_RT_OFFSET                                     29916
-#define QM_REG_PQTX2PF_16_RT_OFFSET                                     29917
-#define QM_REG_PQTX2PF_17_RT_OFFSET                                     29918
-#define QM_REG_PQTX2PF_18_RT_OFFSET                                     29919
-#define QM_REG_PQTX2PF_19_RT_OFFSET                                     29920
-#define QM_REG_PQTX2PF_20_RT_OFFSET                                     29921
-#define QM_REG_PQTX2PF_21_RT_OFFSET                                     29922
-#define QM_REG_PQTX2PF_22_RT_OFFSET                                     29923
-#define QM_REG_PQTX2PF_23_RT_OFFSET                                     29924
-#define QM_REG_PQTX2PF_24_RT_OFFSET                                     29925
-#define QM_REG_PQTX2PF_25_RT_OFFSET                                     29926
-#define QM_REG_PQTX2PF_26_RT_OFFSET                                     29927
-#define QM_REG_PQTX2PF_27_RT_OFFSET                                     29928
-#define QM_REG_PQTX2PF_28_RT_OFFSET                                     29929
-#define QM_REG_PQTX2PF_29_RT_OFFSET                                     29930
-#define QM_REG_PQTX2PF_30_RT_OFFSET                                     29931
-#define QM_REG_PQTX2PF_31_RT_OFFSET                                     29932
-#define QM_REG_PQTX2PF_32_RT_OFFSET                                     29933
-#define QM_REG_PQTX2PF_33_RT_OFFSET                                     29934
-#define QM_REG_PQTX2PF_34_RT_OFFSET                                     29935
-#define QM_REG_PQTX2PF_35_RT_OFFSET                                     29936
-#define QM_REG_PQTX2PF_36_RT_OFFSET                                     29937
-#define QM_REG_PQTX2PF_37_RT_OFFSET                                     29938
-#define QM_REG_PQTX2PF_38_RT_OFFSET                                     29939
-#define QM_REG_PQTX2PF_39_RT_OFFSET                                     29940
-#define QM_REG_PQTX2PF_40_RT_OFFSET                                     29941
-#define QM_REG_PQTX2PF_41_RT_OFFSET                                     29942
-#define QM_REG_PQTX2PF_42_RT_OFFSET                                     29943
-#define QM_REG_PQTX2PF_43_RT_OFFSET                                     29944
-#define QM_REG_PQTX2PF_44_RT_OFFSET                                     29945
-#define QM_REG_PQTX2PF_45_RT_OFFSET                                     29946
-#define QM_REG_PQTX2PF_46_RT_OFFSET                                     29947
-#define QM_REG_PQTX2PF_47_RT_OFFSET                                     29948
-#define QM_REG_PQTX2PF_48_RT_OFFSET                                     29949
-#define QM_REG_PQTX2PF_49_RT_OFFSET                                     29950
-#define QM_REG_PQTX2PF_50_RT_OFFSET                                     29951
-#define QM_REG_PQTX2PF_51_RT_OFFSET                                     29952
-#define QM_REG_PQTX2PF_52_RT_OFFSET                                     29953
-#define QM_REG_PQTX2PF_53_RT_OFFSET                                     29954
-#define QM_REG_PQTX2PF_54_RT_OFFSET                                     29955
-#define QM_REG_PQTX2PF_55_RT_OFFSET                                     29956
-#define QM_REG_PQTX2PF_56_RT_OFFSET                                     29957
-#define QM_REG_PQTX2PF_57_RT_OFFSET                                     29958
-#define QM_REG_PQTX2PF_58_RT_OFFSET                                     29959
-#define QM_REG_PQTX2PF_59_RT_OFFSET                                     29960
-#define QM_REG_PQTX2PF_60_RT_OFFSET                                     29961
-#define QM_REG_PQTX2PF_61_RT_OFFSET                                     29962
-#define QM_REG_PQTX2PF_62_RT_OFFSET                                     29963
-#define QM_REG_PQTX2PF_63_RT_OFFSET                                     29964
-#define QM_REG_PQOTHER2PF_0_RT_OFFSET                                   29965
-#define QM_REG_PQOTHER2PF_1_RT_OFFSET                                   29966
-#define QM_REG_PQOTHER2PF_2_RT_OFFSET                                   29967
-#define QM_REG_PQOTHER2PF_3_RT_OFFSET                                   29968
-#define QM_REG_PQOTHER2PF_4_RT_OFFSET                                   29969
-#define QM_REG_PQOTHER2PF_5_RT_OFFSET                                   29970
-#define QM_REG_PQOTHER2PF_6_RT_OFFSET                                   29971
-#define QM_REG_PQOTHER2PF_7_RT_OFFSET                                   29972
-#define QM_REG_PQOTHER2PF_8_RT_OFFSET                                   29973
-#define QM_REG_PQOTHER2PF_9_RT_OFFSET                                   29974
-#define QM_REG_PQOTHER2PF_10_RT_OFFSET                                  29975
-#define QM_REG_PQOTHER2PF_11_RT_OFFSET                                  29976
-#define QM_REG_PQOTHER2PF_12_RT_OFFSET                                  29977
-#define QM_REG_PQOTHER2PF_13_RT_OFFSET                                  29978
-#define QM_REG_PQOTHER2PF_14_RT_OFFSET                                  29979
-#define QM_REG_PQOTHER2PF_15_RT_OFFSET                                  29980
-#define QM_REG_RLGLBLPERIOD_0_RT_OFFSET                                 29981
-#define QM_REG_RLGLBLPERIOD_1_RT_OFFSET                                 29982
-#define QM_REG_RLGLBLPERIODTIMER_0_RT_OFFSET                            29983
-#define QM_REG_RLGLBLPERIODTIMER_1_RT_OFFSET                            29984
-#define QM_REG_RLGLBLPERIODSEL_0_RT_OFFSET                              29985
-#define QM_REG_RLGLBLPERIODSEL_1_RT_OFFSET                              29986
-#define QM_REG_RLGLBLPERIODSEL_2_RT_OFFSET                              29987
-#define QM_REG_RLGLBLPERIODSEL_3_RT_OFFSET                              29988
-#define QM_REG_RLGLBLPERIODSEL_4_RT_OFFSET                              29989
-#define QM_REG_RLGLBLPERIODSEL_5_RT_OFFSET                              29990
-#define QM_REG_RLGLBLPERIODSEL_6_RT_OFFSET                              29991
-#define QM_REG_RLGLBLPERIODSEL_7_RT_OFFSET                              29992
-#define QM_REG_RLGLBLINCVAL_RT_OFFSET                                   29993
+#define QM_REG_AFULLQMBYPTHRPFWFQ_RT_OFFSET                             29876
+#define QM_REG_AFULLQMBYPTHRVPWFQ_RT_OFFSET                             29877
+#define QM_REG_AFULLQMBYPTHRPFRL_RT_OFFSET                              29878
+#define QM_REG_AFULLQMBYPTHRGLBLRL_RT_OFFSET                            29879
+#define QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET                           29880
+#define QM_REG_WRROTHERPQGRP_0_RT_OFFSET                                29881
+#define QM_REG_WRROTHERPQGRP_1_RT_OFFSET                                29882
+#define QM_REG_WRROTHERPQGRP_2_RT_OFFSET                                29883
+#define QM_REG_WRROTHERPQGRP_3_RT_OFFSET                                29884
+#define QM_REG_WRROTHERPQGRP_4_RT_OFFSET                                29885
+#define QM_REG_WRROTHERPQGRP_5_RT_OFFSET                                29886
+#define QM_REG_WRROTHERPQGRP_6_RT_OFFSET                                29887
+#define QM_REG_WRROTHERPQGRP_7_RT_OFFSET                                29888
+#define QM_REG_WRROTHERPQGRP_8_RT_OFFSET                                29889
+#define QM_REG_WRROTHERPQGRP_9_RT_OFFSET                                29890
+#define QM_REG_WRROTHERPQGRP_10_RT_OFFSET                               29891
+#define QM_REG_WRROTHERPQGRP_11_RT_OFFSET                               29892
+#define QM_REG_WRROTHERPQGRP_12_RT_OFFSET                               29893
+#define QM_REG_WRROTHERPQGRP_13_RT_OFFSET                               29894
+#define QM_REG_WRROTHERPQGRP_14_RT_OFFSET                               29895
+#define QM_REG_WRROTHERPQGRP_15_RT_OFFSET                               29896
+#define QM_REG_WRROTHERGRPWEIGHT_0_RT_OFFSET                            29897
+#define QM_REG_WRROTHERGRPWEIGHT_1_RT_OFFSET                            29898
+#define QM_REG_WRROTHERGRPWEIGHT_2_RT_OFFSET                            29899
+#define QM_REG_WRROTHERGRPWEIGHT_3_RT_OFFSET                            29900
+#define QM_REG_WRRTXGRPWEIGHT_0_RT_OFFSET                               29901
+#define QM_REG_WRRTXGRPWEIGHT_1_RT_OFFSET                               29902
+#define QM_REG_PQTX2PF_0_RT_OFFSET                                      29903
+#define QM_REG_PQTX2PF_1_RT_OFFSET                                      29904
+#define QM_REG_PQTX2PF_2_RT_OFFSET                                      29905
+#define QM_REG_PQTX2PF_3_RT_OFFSET                                      29906
+#define QM_REG_PQTX2PF_4_RT_OFFSET                                      29907
+#define QM_REG_PQTX2PF_5_RT_OFFSET                                      29908
+#define QM_REG_PQTX2PF_6_RT_OFFSET                                      29909
+#define QM_REG_PQTX2PF_7_RT_OFFSET                                      29910
+#define QM_REG_PQTX2PF_8_RT_OFFSET                                      29911
+#define QM_REG_PQTX2PF_9_RT_OFFSET                                      29912
+#define QM_REG_PQTX2PF_10_RT_OFFSET                                     29913
+#define QM_REG_PQTX2PF_11_RT_OFFSET                                     29914
+#define QM_REG_PQTX2PF_12_RT_OFFSET                                     29915
+#define QM_REG_PQTX2PF_13_RT_OFFSET                                     29916
+#define QM_REG_PQTX2PF_14_RT_OFFSET                                     29917
+#define QM_REG_PQTX2PF_15_RT_OFFSET                                     29918
+#define QM_REG_PQTX2PF_16_RT_OFFSET                                     29919
+#define QM_REG_PQTX2PF_17_RT_OFFSET                                     29920
+#define QM_REG_PQTX2PF_18_RT_OFFSET                                     29921
+#define QM_REG_PQTX2PF_19_RT_OFFSET                                     29922
+#define QM_REG_PQTX2PF_20_RT_OFFSET                                     29923
+#define QM_REG_PQTX2PF_21_RT_OFFSET                                     29924
+#define QM_REG_PQTX2PF_22_RT_OFFSET                                     29925
+#define QM_REG_PQTX2PF_23_RT_OFFSET                                     29926
+#define QM_REG_PQTX2PF_24_RT_OFFSET                                     29927
+#define QM_REG_PQTX2PF_25_RT_OFFSET                                     29928
+#define QM_REG_PQTX2PF_26_RT_OFFSET                                     29929
+#define QM_REG_PQTX2PF_27_RT_OFFSET                                     29930
+#define QM_REG_PQTX2PF_28_RT_OFFSET                                     29931
+#define QM_REG_PQTX2PF_29_RT_OFFSET                                     29932
+#define QM_REG_PQTX2PF_30_RT_OFFSET                                     29933
+#define QM_REG_PQTX2PF_31_RT_OFFSET                                     29934
+#define QM_REG_PQTX2PF_32_RT_OFFSET                                     29935
+#define QM_REG_PQTX2PF_33_RT_OFFSET                                     29936
+#define QM_REG_PQTX2PF_34_RT_OFFSET                                     29937
+#define QM_REG_PQTX2PF_35_RT_OFFSET                                     29938
+#define QM_REG_PQTX2PF_36_RT_OFFSET                                     29939
+#define QM_REG_PQTX2PF_37_RT_OFFSET                                     29940
+#define QM_REG_PQTX2PF_38_RT_OFFSET                                     29941
+#define QM_REG_PQTX2PF_39_RT_OFFSET                                     29942
+#define QM_REG_PQTX2PF_40_RT_OFFSET                                     29943
+#define QM_REG_PQTX2PF_41_RT_OFFSET                                     29944
+#define QM_REG_PQTX2PF_42_RT_OFFSET                                     29945
+#define QM_REG_PQTX2PF_43_RT_OFFSET                                     29946
+#define QM_REG_PQTX2PF_44_RT_OFFSET                                     29947
+#define QM_REG_PQTX2PF_45_RT_OFFSET                                     29948
+#define QM_REG_PQTX2PF_46_RT_OFFSET                                     29949
+#define QM_REG_PQTX2PF_47_RT_OFFSET                                     29950
+#define QM_REG_PQTX2PF_48_RT_OFFSET                                     29951
+#define QM_REG_PQTX2PF_49_RT_OFFSET                                     29952
+#define QM_REG_PQTX2PF_50_RT_OFFSET                                     29953
+#define QM_REG_PQTX2PF_51_RT_OFFSET                                     29954
+#define QM_REG_PQTX2PF_52_RT_OFFSET                                     29955
+#define QM_REG_PQTX2PF_53_RT_OFFSET                                     29956
+#define QM_REG_PQTX2PF_54_RT_OFFSET                                     29957
+#define QM_REG_PQTX2PF_55_RT_OFFSET                                     29958
+#define QM_REG_PQTX2PF_56_RT_OFFSET                                     29959
+#define QM_REG_PQTX2PF_57_RT_OFFSET                                     29960
+#define QM_REG_PQTX2PF_58_RT_OFFSET                                     29961
+#define QM_REG_PQTX2PF_59_RT_OFFSET                                     29962
+#define QM_REG_PQTX2PF_60_RT_OFFSET                                     29963
+#define QM_REG_PQTX2PF_61_RT_OFFSET                                     29964
+#define QM_REG_PQTX2PF_62_RT_OFFSET                                     29965
+#define QM_REG_PQTX2PF_63_RT_OFFSET                                     29966
+#define QM_REG_PQOTHER2PF_0_RT_OFFSET                                   29967
+#define QM_REG_PQOTHER2PF_1_RT_OFFSET                                   29968
+#define QM_REG_PQOTHER2PF_2_RT_OFFSET                                   29969
+#define QM_REG_PQOTHER2PF_3_RT_OFFSET                                   29970
+#define QM_REG_PQOTHER2PF_4_RT_OFFSET                                   29971
+#define QM_REG_PQOTHER2PF_5_RT_OFFSET                                   29972
+#define QM_REG_PQOTHER2PF_6_RT_OFFSET                                   29973
+#define QM_REG_PQOTHER2PF_7_RT_OFFSET                                   29974
+#define QM_REG_PQOTHER2PF_8_RT_OFFSET                                   29975
+#define QM_REG_PQOTHER2PF_9_RT_OFFSET                                   29976
+#define QM_REG_PQOTHER2PF_10_RT_OFFSET                                  29977
+#define QM_REG_PQOTHER2PF_11_RT_OFFSET                                  29978
+#define QM_REG_PQOTHER2PF_12_RT_OFFSET                                  29979
+#define QM_REG_PQOTHER2PF_13_RT_OFFSET                                  29980
+#define QM_REG_PQOTHER2PF_14_RT_OFFSET                                  29981
+#define QM_REG_PQOTHER2PF_15_RT_OFFSET                                  29982
+#define QM_REG_RLGLBLPERIOD_0_RT_OFFSET                                 29983
+#define QM_REG_RLGLBLPERIOD_1_RT_OFFSET                                 29984
+#define QM_REG_RLGLBLPERIODTIMER_0_RT_OFFSET                            29985
+#define QM_REG_RLGLBLPERIODTIMER_1_RT_OFFSET                            29986
+#define QM_REG_RLGLBLPERIODSEL_0_RT_OFFSET                              29987
+#define QM_REG_RLGLBLPERIODSEL_1_RT_OFFSET                              29988
+#define QM_REG_RLGLBLPERIODSEL_2_RT_OFFSET                              29989
+#define QM_REG_RLGLBLPERIODSEL_3_RT_OFFSET                              29990
+#define QM_REG_RLGLBLPERIODSEL_4_RT_OFFSET                              29991
+#define QM_REG_RLGLBLPERIODSEL_5_RT_OFFSET                              29992
+#define QM_REG_RLGLBLPERIODSEL_6_RT_OFFSET                              29993
+#define QM_REG_RLGLBLPERIODSEL_7_RT_OFFSET                              29994
+#define QM_REG_RLGLBLINCVAL_RT_OFFSET                                   29995
 #define QM_REG_RLGLBLINCVAL_RT_SIZE                                     256
-#define QM_REG_RLGLBLUPPERBOUND_RT_OFFSET                               30249
+#define QM_REG_RLGLBLUPPERBOUND_RT_OFFSET                               30251
 #define QM_REG_RLGLBLUPPERBOUND_RT_SIZE                                 256
-#define QM_REG_RLGLBLCRD_RT_OFFSET                                      30505
+#define QM_REG_RLGLBLCRD_RT_OFFSET                                      30507
 #define QM_REG_RLGLBLCRD_RT_SIZE                                        256
-#define QM_REG_RLGLBLENABLE_RT_OFFSET                                   30761
-#define QM_REG_RLPFPERIOD_RT_OFFSET                                     30762
-#define QM_REG_RLPFPERIODTIMER_RT_OFFSET                                30763
-#define QM_REG_RLPFINCVAL_RT_OFFSET                                     30764
+#define QM_REG_RLGLBLENABLE_RT_OFFSET                                   30763
+#define QM_REG_RLPFPERIOD_RT_OFFSET                                     30764
+#define QM_REG_RLPFPERIODTIMER_RT_OFFSET                                30765
+#define QM_REG_RLPFINCVAL_RT_OFFSET                                     30766
 #define QM_REG_RLPFINCVAL_RT_SIZE                                       16
-#define QM_REG_RLPFUPPERBOUND_RT_OFFSET                                 30780
+#define QM_REG_RLPFUPPERBOUND_RT_OFFSET                                 30782
 #define QM_REG_RLPFUPPERBOUND_RT_SIZE                                   16
-#define QM_REG_RLPFCRD_RT_OFFSET                                        30796
+#define QM_REG_RLPFCRD_RT_OFFSET                                        30798
 #define QM_REG_RLPFCRD_RT_SIZE                                          16
-#define QM_REG_RLPFENABLE_RT_OFFSET                                     30812
-#define QM_REG_RLPFVOQENABLE_RT_OFFSET                                  30813
-#define QM_REG_WFQPFWEIGHT_RT_OFFSET                                    30814
+#define QM_REG_RLPFENABLE_RT_OFFSET                                     30814
+#define QM_REG_RLPFVOQENABLE_RT_OFFSET                                  30815
+#define QM_REG_WFQPFWEIGHT_RT_OFFSET                                    30816
 #define QM_REG_WFQPFWEIGHT_RT_SIZE                                      16
-#define QM_REG_WFQPFUPPERBOUND_RT_OFFSET                                30830
+#define QM_REG_WFQPFUPPERBOUND_RT_OFFSET                                30832
 #define QM_REG_WFQPFUPPERBOUND_RT_SIZE                                  16
-#define QM_REG_WFQPFCRD_RT_OFFSET                                       30846
+#define QM_REG_WFQPFCRD_RT_OFFSET                                       30848
 #define QM_REG_WFQPFCRD_RT_SIZE                                         160
-#define QM_REG_WFQPFENABLE_RT_OFFSET                                    31006
-#define QM_REG_WFQVPENABLE_RT_OFFSET                                    31007
-#define QM_REG_BASEADDRTXPQ_RT_OFFSET                                   31008
+#define QM_REG_WFQPFENABLE_RT_OFFSET                                    31008
+#define QM_REG_WFQVPENABLE_RT_OFFSET                                    31009
+#define QM_REG_BASEADDRTXPQ_RT_OFFSET                                   31010
 #define QM_REG_BASEADDRTXPQ_RT_SIZE                                     512
-#define QM_REG_TXPQMAP_RT_OFFSET                                        31520
+#define QM_REG_TXPQMAP_RT_OFFSET                                        31522
 #define QM_REG_TXPQMAP_RT_SIZE                                          512
-#define QM_REG_WFQVPWEIGHT_RT_OFFSET                                    32032
+#define QM_REG_WFQVPWEIGHT_RT_OFFSET                                    32034
 #define QM_REG_WFQVPWEIGHT_RT_SIZE                                      512
-#define QM_REG_WFQVPUPPERBOUND_RT_OFFSET                                32544
-#define QM_REG_WFQVPUPPERBOUND_RT_SIZE                                  512
-#define QM_REG_WFQVPCRD_RT_OFFSET                                       33056
+#define QM_REG_WFQVPCRD_RT_OFFSET                                       32546
 #define QM_REG_WFQVPCRD_RT_SIZE                                         512
-#define QM_REG_WFQVPMAP_RT_OFFSET                                       33568
+#define QM_REG_WFQVPMAP_RT_OFFSET                                       33058
 #define QM_REG_WFQVPMAP_RT_SIZE                                         512
-#define QM_REG_WFQPFCRD_MSB_RT_OFFSET                                   34080
+#define QM_REG_WFQPFCRD_MSB_RT_OFFSET                                   33570
 #define QM_REG_WFQPFCRD_MSB_RT_SIZE                                     160
-#define NIG_REG_LLH_CLS_TYPE_DUALMODE_RT_OFFSET                         34240
-#define NIG_REG_OUTER_TAG_VALUE_LIST0_RT_OFFSET                         34241
-#define NIG_REG_OUTER_TAG_VALUE_LIST1_RT_OFFSET                         34242
-#define NIG_REG_OUTER_TAG_VALUE_LIST2_RT_OFFSET                         34243
-#define NIG_REG_OUTER_TAG_VALUE_LIST3_RT_OFFSET                         34244
-#define NIG_REG_OUTER_TAG_VALUE_MASK_RT_OFFSET                          34245
-#define NIG_REG_LLH_FUNC_TAGMAC_CLS_TYPE_RT_OFFSET                      34246
-#define NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET                               34247
+#define NIG_REG_TAG_ETHERTYPE_0_RT_OFFSET                               33730
+#define NIG_REG_OUTER_TAG_VALUE_LIST0_RT_OFFSET                         33731
+#define NIG_REG_OUTER_TAG_VALUE_LIST1_RT_OFFSET                         33732
+#define NIG_REG_OUTER_TAG_VALUE_LIST2_RT_OFFSET                         33733
+#define NIG_REG_OUTER_TAG_VALUE_LIST3_RT_OFFSET                         33734
+#define NIG_REG_OUTER_TAG_VALUE_MASK_RT_OFFSET                          33735
+#define NIG_REG_LLH_FUNC_TAGMAC_CLS_TYPE_RT_OFFSET                      33736
+#define NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET                               33737
 #define NIG_REG_LLH_FUNC_TAG_EN_RT_SIZE                                 4
-#define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_OFFSET                          34251
+#define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_OFFSET                          33741
 #define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_SIZE                            4
-#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET                            34255
+#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET                            33745
 #define NIG_REG_LLH_FUNC_TAG_VALUE_RT_SIZE                              4
-#define NIG_REG_LLH_FUNC_NO_TAG_RT_OFFSET                               34259
-#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_OFFSET                         34260
+#define NIG_REG_LLH_FUNC_NO_TAG_RT_OFFSET                               33749
+#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_OFFSET                         33750
 #define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_SIZE                           32
-#define NIG_REG_LLH_FUNC_FILTER_EN_RT_OFFSET                            34292
+#define NIG_REG_LLH_FUNC_FILTER_EN_RT_OFFSET                            33782
 #define NIG_REG_LLH_FUNC_FILTER_EN_RT_SIZE                              16
-#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_OFFSET                          34308
+#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_OFFSET                          33798
 #define NIG_REG_LLH_FUNC_FILTER_MODE_RT_SIZE                            16
-#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET                 34324
+#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET                 33814
 #define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_SIZE                   16
-#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_OFFSET                       34340
+#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_OFFSET                       33830
 #define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_SIZE                         16
-#define NIG_REG_TX_EDPM_CTRL_RT_OFFSET                                  34356
-#define CDU_REG_CID_ADDR_PARAMS_RT_OFFSET                               34357
-#define CDU_REG_SEGMENT0_PARAMS_RT_OFFSET                               34358
-#define CDU_REG_SEGMENT1_PARAMS_RT_OFFSET                               34359
-#define CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET                           34360
-#define CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET                           34361
-#define CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET                           34362
-#define CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET                           34363
-#define CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET                        34364
-#define CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET                        34365
-#define CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET                        34366
-#define CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET                        34367
-#define CDU_REG_VF_SEG_TYPE_OFFSET_RT_OFFSET                            34368
-#define CDU_REG_VF_FL_SEG_TYPE_OFFSET_RT_OFFSET                         34369
-#define PBF_REG_BTB_SHARED_AREA_SIZE_RT_OFFSET                          34370
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET                        34371
-#define PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET                           34372
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ0_RT_OFFSET                    34373
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET                        34374
-#define PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET                           34375
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ1_RT_OFFSET                    34376
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ2_RT_OFFSET                        34377
-#define PBF_REG_BTB_GUARANTEED_VOQ2_RT_OFFSET                           34378
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ2_RT_OFFSET                    34379
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ3_RT_OFFSET                        34380
-#define PBF_REG_BTB_GUARANTEED_VOQ3_RT_OFFSET                           34381
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ3_RT_OFFSET                    34382
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ4_RT_OFFSET                        34383
-#define PBF_REG_BTB_GUARANTEED_VOQ4_RT_OFFSET                           34384
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ4_RT_OFFSET                    34385
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ5_RT_OFFSET                        34386
-#define PBF_REG_BTB_GUARANTEED_VOQ5_RT_OFFSET                           34387
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ5_RT_OFFSET                    34388
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ6_RT_OFFSET                        34389
-#define PBF_REG_BTB_GUARANTEED_VOQ6_RT_OFFSET                           34390
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ6_RT_OFFSET                    34391
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ7_RT_OFFSET                        34392
-#define PBF_REG_BTB_GUARANTEED_VOQ7_RT_OFFSET                           34393
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ7_RT_OFFSET                    34394
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ8_RT_OFFSET                        34395
-#define PBF_REG_BTB_GUARANTEED_VOQ8_RT_OFFSET                           34396
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ8_RT_OFFSET                    34397
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ9_RT_OFFSET                        34398
-#define PBF_REG_BTB_GUARANTEED_VOQ9_RT_OFFSET                           34399
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ9_RT_OFFSET                    34400
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ10_RT_OFFSET                       34401
-#define PBF_REG_BTB_GUARANTEED_VOQ10_RT_OFFSET                          34402
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ10_RT_OFFSET                   34403
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ11_RT_OFFSET                       34404
-#define PBF_REG_BTB_GUARANTEED_VOQ11_RT_OFFSET                          34405
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ11_RT_OFFSET                   34406
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ12_RT_OFFSET                       34407
-#define PBF_REG_BTB_GUARANTEED_VOQ12_RT_OFFSET                          34408
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ12_RT_OFFSET                   34409
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ13_RT_OFFSET                       34410
-#define PBF_REG_BTB_GUARANTEED_VOQ13_RT_OFFSET                          34411
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ13_RT_OFFSET                   34412
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ14_RT_OFFSET                       34413
-#define PBF_REG_BTB_GUARANTEED_VOQ14_RT_OFFSET                          34414
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ14_RT_OFFSET                   34415
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ15_RT_OFFSET                       34416
-#define PBF_REG_BTB_GUARANTEED_VOQ15_RT_OFFSET                          34417
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ15_RT_OFFSET                   34418
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ16_RT_OFFSET                       34419
-#define PBF_REG_BTB_GUARANTEED_VOQ16_RT_OFFSET                          34420
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ16_RT_OFFSET                   34421
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ17_RT_OFFSET                       34422
-#define PBF_REG_BTB_GUARANTEED_VOQ17_RT_OFFSET                          34423
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ17_RT_OFFSET                   34424
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ18_RT_OFFSET                       34425
-#define PBF_REG_BTB_GUARANTEED_VOQ18_RT_OFFSET                          34426
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ18_RT_OFFSET                   34427
-#define PBF_REG_YCMD_QS_NUM_LINES_VOQ19_RT_OFFSET                       34428
-#define PBF_REG_BTB_GUARANTEED_VOQ19_RT_OFFSET                          34429
-#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ19_RT_OFFSET                   34430
-#define XCM_REG_CON_PHY_Q3_RT_OFFSET                                    34431
-
-#define RUNTIME_ARRAY_SIZE 34432
+#define NIG_REG_TX_EDPM_CTRL_RT_OFFSET                                  33846
+#define CDU_REG_CID_ADDR_PARAMS_RT_OFFSET                               33847
+#define CDU_REG_SEGMENT0_PARAMS_RT_OFFSET                               33848
+#define CDU_REG_SEGMENT1_PARAMS_RT_OFFSET                               33849
+#define CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET                           33850
+#define CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET                           33851
+#define CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET                           33852
+#define CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET                           33853
+#define CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET                        33854
+#define CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET                        33855
+#define CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET                        33856
+#define CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET                        33857
+#define CDU_REG_VF_SEG_TYPE_OFFSET_RT_OFFSET                            33858
+#define CDU_REG_VF_FL_SEG_TYPE_OFFSET_RT_OFFSET                         33859
+#define PBF_REG_TAG_ETHERTYPE_0_RT_OFFSET                               33860
+#define PBF_REG_BTB_SHARED_AREA_SIZE_RT_OFFSET                          33861
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET                        33862
+#define PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET                           33863
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ0_RT_OFFSET                    33864
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET                        33865
+#define PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET                           33866
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ1_RT_OFFSET                    33867
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ2_RT_OFFSET                        33868
+#define PBF_REG_BTB_GUARANTEED_VOQ2_RT_OFFSET                           33869
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ2_RT_OFFSET                    33870
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ3_RT_OFFSET                        33871
+#define PBF_REG_BTB_GUARANTEED_VOQ3_RT_OFFSET                           33872
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ3_RT_OFFSET                    33873
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ4_RT_OFFSET                        33874
+#define PBF_REG_BTB_GUARANTEED_VOQ4_RT_OFFSET                           33875
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ4_RT_OFFSET                    33876
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ5_RT_OFFSET                        33877
+#define PBF_REG_BTB_GUARANTEED_VOQ5_RT_OFFSET                           33878
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ5_RT_OFFSET                    33879
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ6_RT_OFFSET                        33880
+#define PBF_REG_BTB_GUARANTEED_VOQ6_RT_OFFSET                           33881
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ6_RT_OFFSET                    33882
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ7_RT_OFFSET                        33883
+#define PBF_REG_BTB_GUARANTEED_VOQ7_RT_OFFSET                           33884
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ7_RT_OFFSET                    33885
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ8_RT_OFFSET                        33886
+#define PBF_REG_BTB_GUARANTEED_VOQ8_RT_OFFSET                           33887
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ8_RT_OFFSET                    33888
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ9_RT_OFFSET                        33889
+#define PBF_REG_BTB_GUARANTEED_VOQ9_RT_OFFSET                           33890
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ9_RT_OFFSET                    33891
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ10_RT_OFFSET                       33892
+#define PBF_REG_BTB_GUARANTEED_VOQ10_RT_OFFSET                          33893
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ10_RT_OFFSET                   33894
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ11_RT_OFFSET                       33895
+#define PBF_REG_BTB_GUARANTEED_VOQ11_RT_OFFSET                          33896
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ11_RT_OFFSET                   33897
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ12_RT_OFFSET                       33898
+#define PBF_REG_BTB_GUARANTEED_VOQ12_RT_OFFSET                          33899
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ12_RT_OFFSET                   33900
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ13_RT_OFFSET                       33901
+#define PBF_REG_BTB_GUARANTEED_VOQ13_RT_OFFSET                          33902
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ13_RT_OFFSET                   33903
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ14_RT_OFFSET                       33904
+#define PBF_REG_BTB_GUARANTEED_VOQ14_RT_OFFSET                          33905
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ14_RT_OFFSET                   33906
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ15_RT_OFFSET                       33907
+#define PBF_REG_BTB_GUARANTEED_VOQ15_RT_OFFSET                          33908
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ15_RT_OFFSET                   33909
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ16_RT_OFFSET                       33910
+#define PBF_REG_BTB_GUARANTEED_VOQ16_RT_OFFSET                          33911
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ16_RT_OFFSET                   33912
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ17_RT_OFFSET                       33913
+#define PBF_REG_BTB_GUARANTEED_VOQ17_RT_OFFSET                          33914
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ17_RT_OFFSET                   33915
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ18_RT_OFFSET                       33916
+#define PBF_REG_BTB_GUARANTEED_VOQ18_RT_OFFSET                          33917
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ18_RT_OFFSET                   33918
+#define PBF_REG_YCMD_QS_NUM_LINES_VOQ19_RT_OFFSET                       33919
+#define PBF_REG_BTB_GUARANTEED_VOQ19_RT_OFFSET                          33920
+#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ19_RT_OFFSET                   33921
+#define XCM_REG_CON_PHY_Q3_RT_OFFSET                                    33922
+
+#define RUNTIME_ARRAY_SIZE 33923
 
-/* The eth storm context for the Ystorm */
-struct ystorm_eth_conn_st_ctx {
+/* The eth storm context for the Tstorm */
+struct tstorm_eth_conn_st_ctx {
 	__le32 reserved[4];
 };
 
@@ -2535,41 +2501,253 @@ struct xstorm_eth_conn_ag_ctx {
 	__le32	reg7 /* reg7 */;
 	__le32	reg8 /* reg8 */;
 	__le32	reg9 /* reg9 */;
-	u8	byte7 /* byte7 */;
-	u8	byte8 /* byte8 */;
-	u8	byte9 /* byte9 */;
-	u8	byte10 /* byte10 */;
-	u8	byte11 /* byte11 */;
-	u8	byte12 /* byte12 */;
-	u8	byte13 /* byte13 */;
-	u8	byte14 /* byte14 */;
-	u8	byte15 /* byte15 */;
-	u8	byte16 /* byte16 */;
-	__le16	word11 /* word11 */;
+	u8	byte7 /* byte7 */;
+	u8	byte8 /* byte8 */;
+	u8	byte9 /* byte9 */;
+	u8	byte10 /* byte10 */;
+	u8	byte11 /* byte11 */;
+	u8	byte12 /* byte12 */;
+	u8	byte13 /* byte13 */;
+	u8	byte14 /* byte14 */;
+	u8	byte15 /* byte15 */;
+	u8	byte16 /* byte16 */;
+	__le16	word11 /* word11 */;
+	__le32	reg10 /* reg10 */;
+	__le32	reg11 /* reg11 */;
+	__le32	reg12 /* reg12 */;
+	__le32	reg13 /* reg13 */;
+	__le32	reg14 /* reg14 */;
+	__le32	reg15 /* reg15 */;
+	__le32	reg16 /* reg16 */;
+	__le32	reg17 /* reg17 */;
+	__le32	reg18 /* reg18 */;
+	__le32	reg19 /* reg19 */;
+	__le16	word12 /* word12 */;
+	__le16	word13 /* word13 */;
+	__le16	word14 /* word14 */;
+	__le16	word15 /* word15 */;
+};
+
+/* The eth storm context for the Ystorm */
+struct ystorm_eth_conn_st_ctx {
+	__le32 reserved[8];
+};
+
+struct ystorm_eth_conn_ag_ctx {
+	u8	byte0 /* cdu_validation */;
+	u8	byte1 /* state */;
+	u8	flags0;
+#define YSTORM_ETH_CONN_AG_CTX_BIT0_MASK                  0x1 /* exist_in_qm0 */
+#define YSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT                 0
+#define YSTORM_ETH_CONN_AG_CTX_BIT1_MASK                  0x1 /* exist_in_qm1 */
+#define YSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT                 1
+#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK     0x3   /* cf0 */
+#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT    2
+#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_MASK      0x3   /* cf1 */
+#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_SHIFT     4
+#define YSTORM_ETH_CONN_AG_CTX_CF2_MASK                   0x3   /* cf2 */
+#define YSTORM_ETH_CONN_AG_CTX_CF2_SHIFT                  6
+	u8 flags1;
+#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK  0x1   /* cf0en */
+#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT 0
+#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_EN_MASK   0x1   /* cf1en */
+#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_EN_SHIFT  1
+#define YSTORM_ETH_CONN_AG_CTX_CF2EN_MASK                 0x1   /* cf2en */
+#define YSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT                2
+#define YSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK               0x1   /* rule0en */
+#define YSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT              3
+#define YSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK               0x1   /* rule1en */
+#define YSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT              4
+#define YSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK               0x1   /* rule2en */
+#define YSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT              5
+#define YSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK               0x1   /* rule3en */
+#define YSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT              6
+#define YSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK               0x1   /* rule4en */
+#define YSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT              7
+	u8	byte2 /* byte2 */;
+	u8	byte3 /* byte3 */;
+	__le16	word0 /* word0 */;
+	__le32	terminate_spqe /* reg0 */;
+	__le32	reg1 /* reg1 */;
+	__le16	tx_bd_cons_upd /* word1 */;
+	__le16	word2 /* word2 */;
+	__le16	word3 /* word3 */;
+	__le16	word4 /* word4 */;
+	__le32	reg2 /* reg2 */;
+	__le32	reg3 /* reg3 */;
+};
+
+struct tstorm_eth_conn_ag_ctx {
+	u8	byte0 /* cdu_validation */;
+	u8	byte1 /* state */;
+	u8	flags0;
+#define TSTORM_ETH_CONN_AG_CTX_BIT0_MASK      0x1       /* exist_in_qm0 */
+#define TSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT     0
+#define TSTORM_ETH_CONN_AG_CTX_BIT1_MASK      0x1       /* exist_in_qm1 */
+#define TSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT     1
+#define TSTORM_ETH_CONN_AG_CTX_BIT2_MASK      0x1       /* bit2 */
+#define TSTORM_ETH_CONN_AG_CTX_BIT2_SHIFT     2
+#define TSTORM_ETH_CONN_AG_CTX_BIT3_MASK      0x1       /* bit3 */
+#define TSTORM_ETH_CONN_AG_CTX_BIT3_SHIFT     3
+#define TSTORM_ETH_CONN_AG_CTX_BIT4_MASK      0x1       /* bit4 */
+#define TSTORM_ETH_CONN_AG_CTX_BIT4_SHIFT     4
+#define TSTORM_ETH_CONN_AG_CTX_BIT5_MASK      0x1       /* bit5 */
+#define TSTORM_ETH_CONN_AG_CTX_BIT5_SHIFT     5
+#define TSTORM_ETH_CONN_AG_CTX_CF0_MASK       0x3       /* timer0cf */
+#define TSTORM_ETH_CONN_AG_CTX_CF0_SHIFT      6
+	u8 flags1;
+#define TSTORM_ETH_CONN_AG_CTX_CF1_MASK       0x3       /* timer1cf */
+#define TSTORM_ETH_CONN_AG_CTX_CF1_SHIFT      0
+#define TSTORM_ETH_CONN_AG_CTX_CF2_MASK       0x3       /* timer2cf */
+#define TSTORM_ETH_CONN_AG_CTX_CF2_SHIFT      2
+#define TSTORM_ETH_CONN_AG_CTX_CF3_MASK       0x3       /* timer_stop_all */
+#define TSTORM_ETH_CONN_AG_CTX_CF3_SHIFT      4
+#define TSTORM_ETH_CONN_AG_CTX_CF4_MASK       0x3       /* cf4 */
+#define TSTORM_ETH_CONN_AG_CTX_CF4_SHIFT      6
+	u8 flags2;
+#define TSTORM_ETH_CONN_AG_CTX_CF5_MASK       0x3       /* cf5 */
+#define TSTORM_ETH_CONN_AG_CTX_CF5_SHIFT      0
+#define TSTORM_ETH_CONN_AG_CTX_CF6_MASK       0x3       /* cf6 */
+#define TSTORM_ETH_CONN_AG_CTX_CF6_SHIFT      2
+#define TSTORM_ETH_CONN_AG_CTX_CF7_MASK       0x3       /* cf7 */
+#define TSTORM_ETH_CONN_AG_CTX_CF7_SHIFT      4
+#define TSTORM_ETH_CONN_AG_CTX_CF8_MASK       0x3       /* cf8 */
+#define TSTORM_ETH_CONN_AG_CTX_CF8_SHIFT      6
+	u8 flags3;
+#define TSTORM_ETH_CONN_AG_CTX_CF9_MASK       0x3       /* cf9 */
+#define TSTORM_ETH_CONN_AG_CTX_CF9_SHIFT      0
+#define TSTORM_ETH_CONN_AG_CTX_CF10_MASK      0x3       /* cf10 */
+#define TSTORM_ETH_CONN_AG_CTX_CF10_SHIFT     2
+#define TSTORM_ETH_CONN_AG_CTX_CF0EN_MASK     0x1       /* cf0en */
+#define TSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT    4
+#define TSTORM_ETH_CONN_AG_CTX_CF1EN_MASK     0x1       /* cf1en */
+#define TSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT    5
+#define TSTORM_ETH_CONN_AG_CTX_CF2EN_MASK     0x1       /* cf2en */
+#define TSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT    6
+#define TSTORM_ETH_CONN_AG_CTX_CF3EN_MASK     0x1       /* cf3en */
+#define TSTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT    7
+	u8 flags4;
+#define TSTORM_ETH_CONN_AG_CTX_CF4EN_MASK     0x1       /* cf4en */
+#define TSTORM_ETH_CONN_AG_CTX_CF4EN_SHIFT    0
+#define TSTORM_ETH_CONN_AG_CTX_CF5EN_MASK     0x1       /* cf5en */
+#define TSTORM_ETH_CONN_AG_CTX_CF5EN_SHIFT    1
+#define TSTORM_ETH_CONN_AG_CTX_CF6EN_MASK     0x1       /* cf6en */
+#define TSTORM_ETH_CONN_AG_CTX_CF6EN_SHIFT    2
+#define TSTORM_ETH_CONN_AG_CTX_CF7EN_MASK     0x1       /* cf7en */
+#define TSTORM_ETH_CONN_AG_CTX_CF7EN_SHIFT    3
+#define TSTORM_ETH_CONN_AG_CTX_CF8EN_MASK     0x1       /* cf8en */
+#define TSTORM_ETH_CONN_AG_CTX_CF8EN_SHIFT    4
+#define TSTORM_ETH_CONN_AG_CTX_CF9EN_MASK     0x1       /* cf9en */
+#define TSTORM_ETH_CONN_AG_CTX_CF9EN_SHIFT    5
+#define TSTORM_ETH_CONN_AG_CTX_CF10EN_MASK    0x1       /* cf10en */
+#define TSTORM_ETH_CONN_AG_CTX_CF10EN_SHIFT   6
+#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK   0x1       /* rule0en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT  7
+	u8 flags5;
+#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK   0x1       /* rule1en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT  0
+#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK   0x1       /* rule2en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT  1
+#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK   0x1       /* rule3en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT  2
+#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK   0x1       /* rule4en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT  3
+#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_MASK   0x1       /* rule5en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT  4
+#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_MASK  0x1       /* rule6en */
+#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_SHIFT 5
+#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_MASK   0x1       /* rule7en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT  6
+#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_MASK   0x1       /* rule8en */
+#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT  7
+	__le32	reg0 /* reg0 */;
+	__le32	reg1 /* reg1 */;
+	__le32	reg2 /* reg2 */;
+	__le32	reg3 /* reg3 */;
+	__le32	reg4 /* reg4 */;
+	__le32	reg5 /* reg5 */;
+	__le32	reg6 /* reg6 */;
+	__le32	reg7 /* reg7 */;
+	__le32	reg8 /* reg8 */;
+	u8	byte2 /* byte2 */;
+	u8	byte3 /* byte3 */;
+	__le16	rx_bd_cons /* word0 */;
+	u8	byte4 /* byte4 */;
+	u8	byte5 /* byte5 */;
+	__le16	rx_bd_prod /* word1 */;
+	__le16	word2 /* conn_dpi */;
+	__le16	word3 /* word3 */;
+	__le32	reg9 /* reg9 */;
 	__le32	reg10 /* reg10 */;
-	__le32	reg11 /* reg11 */;
-	__le32	reg12 /* reg12 */;
-	__le32	reg13 /* reg13 */;
-	__le32	reg14 /* reg14 */;
-	__le32	reg15 /* reg15 */;
-	__le32	reg16 /* reg16 */;
-	__le32	reg17 /* reg17 */;
-	__le32	reg18 /* reg18 */;
-	__le32	reg19 /* reg19 */;
-	__le16	word12 /* word12 */;
-	__le16	word13 /* word13 */;
-	__le16	word14 /* word14 */;
-	__le16	word15 /* word15 */;
-};
-
-/* The eth storm context for the Tstorm */
-struct tstorm_eth_conn_st_ctx {
-	__le32 reserved[4];
 };
 
-/* The eth storm context for the Mstorm */
-struct mstorm_eth_conn_st_ctx {
-	__le32 reserved[8];
+struct ustorm_eth_conn_ag_ctx {
+	u8	byte0 /* cdu_validation */;
+	u8	byte1 /* state */;
+	u8	flags0;
+#define USTORM_ETH_CONN_AG_CTX_BIT0_MASK                  0x1 /* exist_in_qm0 */
+#define USTORM_ETH_CONN_AG_CTX_BIT0_SHIFT                 0
+#define USTORM_ETH_CONN_AG_CTX_BIT1_MASK                  0x1 /* exist_in_qm1 */
+#define USTORM_ETH_CONN_AG_CTX_BIT1_SHIFT                 1
+#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_MASK   0x3 /* timer0cf */
+#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_SHIFT  2
+#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_MASK   0x3 /* timer1cf */
+#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_SHIFT  4
+#define USTORM_ETH_CONN_AG_CTX_CF2_MASK                   0x3 /* timer2cf */
+#define USTORM_ETH_CONN_AG_CTX_CF2_SHIFT                  6
+	u8 flags1;
+#define USTORM_ETH_CONN_AG_CTX_CF3_MASK                 0x3 /* timer_stop_all */
+#define USTORM_ETH_CONN_AG_CTX_CF3_SHIFT                0
+#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_MASK           0x3 /* cf4 */
+#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_SHIFT          2
+#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_MASK           0x3 /* cf5 */
+#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_SHIFT          4
+#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK   0x3 /* cf6 */
+#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT  6
+	u8 flags2;
+#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_EN_MASK  0x1 /* cf0en */
+#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_EN_SHIFT 0
+#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_EN_MASK  0x1 /* cf1en */
+#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_EN_SHIFT 1
+#define USTORM_ETH_CONN_AG_CTX_CF2EN_MASK                   0x1 /* cf2en */
+#define USTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT                  2
+#define USTORM_ETH_CONN_AG_CTX_CF3EN_MASK                   0x1 /* cf3en */
+#define USTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT                  3
+#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_MASK            0x1 /* cf4en */
+#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_SHIFT           4
+#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_MASK            0x1 /* cf5en */
+#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_SHIFT           5
+#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK    0x1 /* cf6en */
+#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT   6
+#define USTORM_ETH_CONN_AG_CTX_RULE0EN_MASK                 0x1 /* rule0en */
+#define USTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT                7
+	u8 flags3;
+#define USTORM_ETH_CONN_AG_CTX_RULE1EN_MASK                 0x1 /* rule1en */
+#define USTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT                0
+#define USTORM_ETH_CONN_AG_CTX_RULE2EN_MASK                 0x1 /* rule2en */
+#define USTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT                1
+#define USTORM_ETH_CONN_AG_CTX_RULE3EN_MASK                 0x1 /* rule3en */
+#define USTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT                2
+#define USTORM_ETH_CONN_AG_CTX_RULE4EN_MASK                 0x1 /* rule4en */
+#define USTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT                3
+#define USTORM_ETH_CONN_AG_CTX_RULE5EN_MASK                 0x1 /* rule5en */
+#define USTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT                4
+#define USTORM_ETH_CONN_AG_CTX_RULE6EN_MASK                 0x1 /* rule6en */
+#define USTORM_ETH_CONN_AG_CTX_RULE6EN_SHIFT                5
+#define USTORM_ETH_CONN_AG_CTX_RULE7EN_MASK                 0x1 /* rule7en */
+#define USTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT                6
+#define USTORM_ETH_CONN_AG_CTX_RULE8EN_MASK                 0x1 /* rule8en */
+#define USTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT                7
+	u8	byte2 /* byte2 */;
+	u8	byte3 /* byte3 */;
+	__le16	word0 /* conn_dpi */;
+	__le16	tx_bd_cons /* word1 */;
+	__le32	reg0 /* reg0 */;
+	__le32	reg1 /* reg1 */;
+	__le32	reg2 /* reg2 */;
+	__le32	tx_int_coallecing_timeset /* reg3 */;
+	__le16	tx_drv_bd_cons /* word2 */;
+	__le16	rx_drv_cqe_cons /* word3 */;
 };
 
 /* The eth storm context for the Ustorm */
@@ -2577,24 +2755,30 @@ struct ustorm_eth_conn_st_ctx {
 	__le32 reserved[40];
 };
 
+/* The eth storm context for the Mstorm */
+struct mstorm_eth_conn_st_ctx {
+	__le32 reserved[8];
+};
+
 /* eth connection context */
 struct eth_conn_context {
-	struct ystorm_eth_conn_st_ctx	ystorm_st_context;
-	struct regpair			ystorm_st_padding[2] /* padding */;
+	struct tstorm_eth_conn_st_ctx	tstorm_st_context;
+	struct regpair			tstorm_st_padding[2];
 	struct pstorm_eth_conn_st_ctx	pstorm_st_context;
-	struct regpair			pstorm_st_padding[2] /* padding */;
 	struct xstorm_eth_conn_st_ctx	xstorm_st_context;
 	struct xstorm_eth_conn_ag_ctx	xstorm_ag_context;
-	struct tstorm_eth_conn_st_ctx	tstorm_st_context;
-	struct regpair			tstorm_st_padding[2] /* padding */;
-	struct mstorm_eth_conn_st_ctx	mstorm_st_context;
+	struct ystorm_eth_conn_st_ctx	ystorm_st_context;
+	struct ystorm_eth_conn_ag_ctx	ystorm_ag_context;
+	struct tstorm_eth_conn_ag_ctx	tstorm_ag_context;
+	struct ustorm_eth_conn_ag_ctx	ustorm_ag_context;
 	struct ustorm_eth_conn_st_ctx	ustorm_st_context;
+	struct mstorm_eth_conn_st_ctx	mstorm_st_context;
 };
 
 enum eth_filter_action {
 	ETH_FILTER_ACTION_REMOVE,
 	ETH_FILTER_ACTION_ADD,
-	ETH_FILTER_ACTION_REPLACE,
+	ETH_FILTER_ACTION_REMOVE_ALL,
 	MAX_ETH_FILTER_ACTION
 };
 
@@ -2653,6 +2837,32 @@ enum eth_ramrod_cmd_id {
 	MAX_ETH_RAMROD_CMD_ID
 };
 
+enum eth_tx_err {
+	ETH_TX_ERR_DROP /* Drop erronous packet. */,
+	ETH_TX_ERR_ASSERT_MALICIOUS,
+	MAX_ETH_TX_ERR
+};
+
+struct eth_tx_err_vals {
+	__le16 values;
+#define ETH_TX_ERR_VALS_ILLEGAL_VLAN_MODE_MASK            0x1
+#define ETH_TX_ERR_VALS_ILLEGAL_VLAN_MODE_SHIFT           0
+#define ETH_TX_ERR_VALS_PACKET_TOO_SMALL_MASK             0x1
+#define ETH_TX_ERR_VALS_PACKET_TOO_SMALL_SHIFT            1
+#define ETH_TX_ERR_VALS_ANTI_SPOOFING_ERR_MASK            0x1
+#define ETH_TX_ERR_VALS_ANTI_SPOOFING_ERR_SHIFT           2
+#define ETH_TX_ERR_VALS_ILLEGAL_INBAND_TAGS_MASK          0x1
+#define ETH_TX_ERR_VALS_ILLEGAL_INBAND_TAGS_SHIFT         3
+#define ETH_TX_ERR_VALS_VLAN_INSERTION_W_INBAND_TAG_MASK  0x1
+#define ETH_TX_ERR_VALS_VLAN_INSERTION_W_INBAND_TAG_SHIFT 4
+#define ETH_TX_ERR_VALS_MTU_VIOLATION_MASK                0x1
+#define ETH_TX_ERR_VALS_MTU_VIOLATION_SHIFT               5
+#define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_MASK        0x1
+#define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_SHIFT       6
+#define ETH_TX_ERR_VALS_RESERVED_MASK                     0x1FF
+#define ETH_TX_ERR_VALS_RESERVED_SHIFT                    7
+};
+
 struct eth_vport_rss_config {
 	__le16 capabilities;
 #define ETH_VPORT_RSS_CONFIG_IPV4_CAPABILITY_MASK	0x1
@@ -2669,12 +2879,8 @@ struct eth_vport_rss_config {
 #define ETH_VPORT_RSS_CONFIG_IPV6_UDP_CAPABILITY_SHIFT   5
 #define ETH_VPORT_RSS_CONFIG_EN_5_TUPLE_CAPABILITY_MASK  0x1
 #define ETH_VPORT_RSS_CONFIG_EN_5_TUPLE_CAPABILITY_SHIFT 6
-#define ETH_VPORT_RSS_CONFIG_CALC_4TUP_TCP_FRAG_MASK     0x1
-#define ETH_VPORT_RSS_CONFIG_CALC_4TUP_TCP_FRAG_SHIFT    7
-#define ETH_VPORT_RSS_CONFIG_CALC_4TUP_UDP_FRAG_MASK     0x1
-#define ETH_VPORT_RSS_CONFIG_CALC_4TUP_UDP_FRAG_SHIFT    8
-#define ETH_VPORT_RSS_CONFIG_RESERVED0_MASK	      0x7F
-#define ETH_VPORT_RSS_CONFIG_RESERVED0_SHIFT	     9
+#define ETH_VPORT_RSS_CONFIG_RESERVED0_MASK	      0x1FF
+#define ETH_VPORT_RSS_CONFIG_RESERVED0_SHIFT	     7
 	u8      rss_id;
 	u8      rss_mode;
 	u8      update_rss_key;
@@ -2749,10 +2955,14 @@ struct rx_queue_start_ramrod_data {
 	u8	      pxp_tph_valid_pkt;
 	u8	      pxp_st_hint;
 	__le16	  pxp_st_index;
-	u8	      reserved[4];
-	struct regpair  cqe_pbl_addr;
-	struct regpair  bd_base;
-	struct regpair  sge_base;
+	u8		pmd_mode;
+	u8		notify_en;
+	u8		toggle_val;
+	u8		reserved[7];
+	__le16		reserved1;
+	struct regpair	cqe_pbl_addr;
+	struct regpair	bd_base;
+	struct regpair	reserved2;
 };
 
 struct rx_queue_stop_ramrod_data {
@@ -2764,23 +2974,24 @@ struct rx_queue_stop_ramrod_data {
 };
 
 struct rx_queue_update_ramrod_data {
-	__le16	  rx_queue_id;
-	u8	      complete_cqe_flg;
-	u8	      complete_event_flg;
-	u8	      init_sge_ring_flg;
-	u8	      vport_id;
-	u8	      pxp_tph_valid_sge;
-	u8	      pxp_st_hint;
-	__le16	  pxp_st_index;
-	u8	      reserved[6];
-	struct regpair  sge_base;
+	__le16	rx_queue_id;
+	u8	complete_cqe_flg;
+	u8	complete_event_flg;
+	u8	vport_id;
+	u8	reserved[4];
+	u8	reserved1;
+	u8	reserved2;
+	u8	reserved3;
+	__le16	reserved4;
+	__le16	reserved5;
+	struct regpair reserved6;
 };
 
 struct tx_queue_start_ramrod_data {
 	__le16  sb_id;
 	u8      sb_index;
 	u8      vport_id;
-	u8      tc;
+	u8      reserved0;
 	u8      stats_counter_id;
 	__le16  qm_pq_id;
 	u8      flags;
@@ -2790,18 +3001,25 @@ struct tx_queue_start_ramrod_data {
 #define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_SHIFT     1
 #define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_MASK      0x1
 #define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_SHIFT     2
-#define TX_QUEUE_START_RAMROD_DATA_RESERVED0_MASK	      0x1F
-#define TX_QUEUE_START_RAMROD_DATA_RESERVED0_SHIFT	     3
-	u8	      pin_context;
-	u8	      pxp_tph_valid_bd;
-	u8	      pxp_tph_valid_pkt;
-	__le16	  pxp_st_index;
-	u8	      pxp_st_hint;
-	u8	      reserved1[3];
-	__le16	  queue_zone_id;
-	__le16	  test_dup_count;
-	__le16	  pbl_size;
-	struct regpair  pbl_base_addr;
+#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_MASK               0x1
+#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_SHIFT              3
+#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_MASK              0x1
+#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_SHIFT             4
+#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_MASK            0x1
+#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_SHIFT           5
+#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_MASK              0x3
+#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_SHIFT             6
+	u8	pxp_st_hint;
+	u8	pxp_tph_valid_bd;
+	u8	pxp_tph_valid_pkt;
+	__le16	pxp_st_index;
+	__le16	comp_agg_size;
+	__le16	queue_zone_id;
+	__le16	test_dup_count;
+	__le16	pbl_size;
+	__le16	tx_queue_id;
+	struct regpair	pbl_base_addr;
+	struct regpair	bd_cons_address;
 };
 
 struct tx_queue_stop_ramrod_data {
@@ -2822,16 +3040,16 @@ struct vport_start_ramrod_data {
 	struct eth_vport_rx_mode	rx_mode;
 	struct eth_vport_tx_mode	tx_mode;
 	struct eth_vport_tpa_param      tpa_param;
-	__le16			  sge_buff_size;
-	u8			      max_sges_num;
-	u8			      tx_switching_en;
-	u8			      anti_spoofing_en;
-	u8			      default_vlan_en;
-	u8			      handle_ptp_pkts;
-	u8			      silent_vlan_removal_en;
-	__le16			  default_vlan;
-	u8			      untagged;
-	u8			      reserved[7];
+	__le16				default_vlan;
+	u8				tx_switching_en;
+	u8				anti_spoofing_en;
+	u8				default_vlan_en;
+	u8				handle_ptp_pkts;
+	u8				silent_vlan_removal_en;
+	u8				untagged;
+	struct eth_tx_err_vals		tx_err_behav;
+	u8				zero_placement_offset;
+	u8				reserved[7];
 };
 
 struct vport_stop_ramrod_data {
@@ -2840,36 +3058,35 @@ struct vport_stop_ramrod_data {
 };
 
 struct vport_update_ramrod_data_cmn {
-	u8      vport_id;
-	u8      update_rx_active_flg;
-	u8      rx_active_flg;
-	u8      update_tx_active_flg;
-	u8      tx_active_flg;
-	u8      update_rx_mode_flg;
-	u8      update_tx_mode_flg;
-	u8      update_approx_mcast_flg;
-	u8      update_rss_flg;
-	u8      update_inner_vlan_removal_en_flg;
-	u8      inner_vlan_removal_en;
-	u8      update_tpa_param_flg;
-	u8      update_tpa_en_flg;
-	u8      update_sge_param_flg;
-	__le16  sge_buff_size;
-	u8      max_sges_num;
-	u8      update_tx_switching_en_flg;
-	u8      tx_switching_en;
-	u8      update_anti_spoofing_en_flg;
-	u8      anti_spoofing_en;
-	u8      update_handle_ptp_pkts;
-	u8      handle_ptp_pkts;
-	u8      update_default_vlan_en_flg;
-	u8      default_vlan_en;
-	u8      update_default_vlan_flg;
-	__le16  default_vlan;
-	u8      update_accept_any_vlan_flg;
-	u8      accept_any_vlan;
-	u8      silent_vlan_removal_en;
-	u8      reserved;
+	u8	vport_id;
+	u8	update_rx_active_flg;
+	u8	rx_active_flg;
+	u8	update_tx_active_flg;
+	u8	tx_active_flg;
+	u8	update_rx_mode_flg;
+	u8	update_tx_mode_flg;
+	u8	update_approx_mcast_flg;
+	u8	update_rss_flg;
+	u8	update_inner_vlan_removal_en_flg;
+	u8	inner_vlan_removal_en;
+	u8	update_tpa_param_flg;
+	u8	update_tpa_en_flg;
+	u8	update_tx_switching_en_flg;
+	u8	tx_switching_en;
+	u8	update_anti_spoofing_en_flg;
+	u8	anti_spoofing_en;
+	u8	update_handle_ptp_pkts;
+	u8	handle_ptp_pkts;
+	u8	update_default_vlan_en_flg;
+	u8	default_vlan_en;
+	u8	update_default_vlan_flg;
+	__le16	default_vlan;
+	u8	update_accept_any_vlan_flg;
+	u8	accept_any_vlan;
+	u8	silent_vlan_removal_en;
+	u8	update_mtu_flg;
+	__le16	mtu;
+	u8	reserved[2];
 };
 
 struct vport_update_ramrod_mcast {
@@ -2885,436 +3102,6 @@ struct vport_update_ramrod_data {
 	struct eth_vport_rss_config	     rss_config;
 };
 
-struct mstorm_eth_conn_ag_ctx {
-	u8	byte0 /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define MSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM0_MASK  0x1   /* exist_in_qm0 */
-#define MSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0
-#define MSTORM_ETH_CONN_AG_CTX_BIT1_MASK          0x1   /* exist_in_qm1 */
-#define MSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT         1
-#define MSTORM_ETH_CONN_AG_CTX_CF0_MASK           0x3   /* cf0 */
-#define MSTORM_ETH_CONN_AG_CTX_CF0_SHIFT          2
-#define MSTORM_ETH_CONN_AG_CTX_CF1_MASK           0x3   /* cf1 */
-#define MSTORM_ETH_CONN_AG_CTX_CF1_SHIFT          4
-#define MSTORM_ETH_CONN_AG_CTX_CF2_MASK           0x3   /* cf2 */
-#define MSTORM_ETH_CONN_AG_CTX_CF2_SHIFT          6
-	u8 flags1;
-#define MSTORM_ETH_CONN_AG_CTX_CF0EN_MASK         0x1   /* cf0en */
-#define MSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT        0
-#define MSTORM_ETH_CONN_AG_CTX_CF1EN_MASK         0x1   /* cf1en */
-#define MSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT        1
-#define MSTORM_ETH_CONN_AG_CTX_CF2EN_MASK         0x1   /* cf2en */
-#define MSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT        2
-#define MSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK       0x1   /* rule0en */
-#define MSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT      3
-#define MSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK       0x1   /* rule1en */
-#define MSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT      4
-#define MSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK       0x1   /* rule2en */
-#define MSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT      5
-#define MSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK       0x1   /* rule3en */
-#define MSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT      6
-#define MSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK       0x1   /* rule4en */
-#define MSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT      7
-	__le16	word0 /* word0 */;
-	__le16	word1 /* word1 */;
-	__le32	reg0 /* reg0 */;
-	__le32	reg1 /* reg1 */;
-};
-
-struct tstorm_eth_conn_ag_ctx {
-	u8	byte0 /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define TSTORM_ETH_CONN_AG_CTX_BIT0_MASK      0x1       /* exist_in_qm0 */
-#define TSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT     0
-#define TSTORM_ETH_CONN_AG_CTX_BIT1_MASK      0x1       /* exist_in_qm1 */
-#define TSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT     1
-#define TSTORM_ETH_CONN_AG_CTX_BIT2_MASK      0x1       /* bit2 */
-#define TSTORM_ETH_CONN_AG_CTX_BIT2_SHIFT     2
-#define TSTORM_ETH_CONN_AG_CTX_BIT3_MASK      0x1       /* bit3 */
-#define TSTORM_ETH_CONN_AG_CTX_BIT3_SHIFT     3
-#define TSTORM_ETH_CONN_AG_CTX_BIT4_MASK      0x1       /* bit4 */
-#define TSTORM_ETH_CONN_AG_CTX_BIT4_SHIFT     4
-#define TSTORM_ETH_CONN_AG_CTX_BIT5_MASK      0x1       /* bit5 */
-#define TSTORM_ETH_CONN_AG_CTX_BIT5_SHIFT     5
-#define TSTORM_ETH_CONN_AG_CTX_CF0_MASK       0x3       /* timer0cf */
-#define TSTORM_ETH_CONN_AG_CTX_CF0_SHIFT      6
-	u8 flags1;
-#define TSTORM_ETH_CONN_AG_CTX_CF1_MASK       0x3       /* timer1cf */
-#define TSTORM_ETH_CONN_AG_CTX_CF1_SHIFT      0
-#define TSTORM_ETH_CONN_AG_CTX_CF2_MASK       0x3       /* timer2cf */
-#define TSTORM_ETH_CONN_AG_CTX_CF2_SHIFT      2
-#define TSTORM_ETH_CONN_AG_CTX_CF3_MASK       0x3       /* timer_stop_all */
-#define TSTORM_ETH_CONN_AG_CTX_CF3_SHIFT      4
-#define TSTORM_ETH_CONN_AG_CTX_CF4_MASK       0x3       /* cf4 */
-#define TSTORM_ETH_CONN_AG_CTX_CF4_SHIFT      6
-	u8 flags2;
-#define TSTORM_ETH_CONN_AG_CTX_CF5_MASK       0x3       /* cf5 */
-#define TSTORM_ETH_CONN_AG_CTX_CF5_SHIFT      0
-#define TSTORM_ETH_CONN_AG_CTX_CF6_MASK       0x3       /* cf6 */
-#define TSTORM_ETH_CONN_AG_CTX_CF6_SHIFT      2
-#define TSTORM_ETH_CONN_AG_CTX_CF7_MASK       0x3       /* cf7 */
-#define TSTORM_ETH_CONN_AG_CTX_CF7_SHIFT      4
-#define TSTORM_ETH_CONN_AG_CTX_CF8_MASK       0x3       /* cf8 */
-#define TSTORM_ETH_CONN_AG_CTX_CF8_SHIFT      6
-	u8 flags3;
-#define TSTORM_ETH_CONN_AG_CTX_CF9_MASK       0x3       /* cf9 */
-#define TSTORM_ETH_CONN_AG_CTX_CF9_SHIFT      0
-#define TSTORM_ETH_CONN_AG_CTX_CF10_MASK      0x3       /* cf10 */
-#define TSTORM_ETH_CONN_AG_CTX_CF10_SHIFT     2
-#define TSTORM_ETH_CONN_AG_CTX_CF0EN_MASK     0x1       /* cf0en */
-#define TSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT    4
-#define TSTORM_ETH_CONN_AG_CTX_CF1EN_MASK     0x1       /* cf1en */
-#define TSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT    5
-#define TSTORM_ETH_CONN_AG_CTX_CF2EN_MASK     0x1       /* cf2en */
-#define TSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT    6
-#define TSTORM_ETH_CONN_AG_CTX_CF3EN_MASK     0x1       /* cf3en */
-#define TSTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT    7
-	u8 flags4;
-#define TSTORM_ETH_CONN_AG_CTX_CF4EN_MASK     0x1       /* cf4en */
-#define TSTORM_ETH_CONN_AG_CTX_CF4EN_SHIFT    0
-#define TSTORM_ETH_CONN_AG_CTX_CF5EN_MASK     0x1       /* cf5en */
-#define TSTORM_ETH_CONN_AG_CTX_CF5EN_SHIFT    1
-#define TSTORM_ETH_CONN_AG_CTX_CF6EN_MASK     0x1       /* cf6en */
-#define TSTORM_ETH_CONN_AG_CTX_CF6EN_SHIFT    2
-#define TSTORM_ETH_CONN_AG_CTX_CF7EN_MASK     0x1       /* cf7en */
-#define TSTORM_ETH_CONN_AG_CTX_CF7EN_SHIFT    3
-#define TSTORM_ETH_CONN_AG_CTX_CF8EN_MASK     0x1       /* cf8en */
-#define TSTORM_ETH_CONN_AG_CTX_CF8EN_SHIFT    4
-#define TSTORM_ETH_CONN_AG_CTX_CF9EN_MASK     0x1       /* cf9en */
-#define TSTORM_ETH_CONN_AG_CTX_CF9EN_SHIFT    5
-#define TSTORM_ETH_CONN_AG_CTX_CF10EN_MASK    0x1       /* cf10en */
-#define TSTORM_ETH_CONN_AG_CTX_CF10EN_SHIFT   6
-#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK   0x1       /* rule0en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT  7
-	u8 flags5;
-#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK   0x1       /* rule1en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT  0
-#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK   0x1       /* rule2en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT  1
-#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK   0x1       /* rule3en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT  2
-#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK   0x1       /* rule4en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT  3
-#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_MASK   0x1       /* rule5en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT  4
-#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_MASK  0x1       /* rule6en */
-#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_SHIFT 5
-#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_MASK   0x1       /* rule7en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT  6
-#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_MASK   0x1       /* rule8en */
-#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT  7
-	__le32	reg0 /* reg0 */;
-	__le32	reg1 /* reg1 */;
-	__le32	reg2 /* reg2 */;
-	__le32	reg3 /* reg3 */;
-	__le32	reg4 /* reg4 */;
-	__le32	reg5 /* reg5 */;
-	__le32	reg6 /* reg6 */;
-	__le32	reg7 /* reg7 */;
-	__le32	reg8 /* reg8 */;
-	u8	byte2 /* byte2 */;
-	u8	byte3 /* byte3 */;
-	__le16	rx_bd_cons /* word0 */;
-	u8	byte4 /* byte4 */;
-	u8	byte5 /* byte5 */;
-	__le16	rx_bd_prod /* word1 */;
-	__le16	word2 /* conn_dpi */;
-	__le16	word3 /* word3 */;
-	__le32	reg9 /* reg9 */;
-	__le32	reg10 /* reg10 */;
-};
-
-struct ustorm_eth_conn_ag_ctx {
-	u8	byte0 /* cdu_validation */;
-	u8	byte1 /* state */;
-	u8	flags0;
-#define USTORM_ETH_CONN_AG_CTX_BIT0_MASK                  0x1
-#define USTORM_ETH_CONN_AG_CTX_BIT0_SHIFT                 0
-#define USTORM_ETH_CONN_AG_CTX_BIT1_MASK                  0x1
-#define USTORM_ETH_CONN_AG_CTX_BIT1_SHIFT                 1
-#define USTORM_ETH_CONN_AG_CTX_CF0_MASK                   0x3   /* timer0cf */
-#define USTORM_ETH_CONN_AG_CTX_CF0_SHIFT                  2
-#define USTORM_ETH_CONN_AG_CTX_CF1_MASK                   0x3   /* timer1cf */
-#define USTORM_ETH_CONN_AG_CTX_CF1_SHIFT                  4
-#define USTORM_ETH_CONN_AG_CTX_CF2_MASK                   0x3   /* timer2cf */
-#define USTORM_ETH_CONN_AG_CTX_CF2_SHIFT                  6
-	u8 flags1;
-#define USTORM_ETH_CONN_AG_CTX_CF3_MASK                   0x3
-#define USTORM_ETH_CONN_AG_CTX_CF3_SHIFT                  0
-#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_MASK             0x3   /* cf4 */
-#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_SHIFT            2
-#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_MASK             0x3   /* cf5 */
-#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_SHIFT            4
-#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK     0x3   /* cf6 */
-#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT    6
-	u8 flags2;
-#define USTORM_ETH_CONN_AG_CTX_CF0EN_MASK                 0x1   /* cf0en */
-#define USTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT                0
-#define USTORM_ETH_CONN_AG_CTX_CF1EN_MASK                 0x1   /* cf1en */
-#define USTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT                1
-#define USTORM_ETH_CONN_AG_CTX_CF2EN_MASK                 0x1   /* cf2en */
-#define USTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT                2
-#define USTORM_ETH_CONN_AG_CTX_CF3EN_MASK                 0x1   /* cf3en */
-#define USTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT                3
-#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_MASK          0x1   /* cf4en */
-#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_SHIFT         4
-#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_MASK          0x1   /* cf5en */
-#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_SHIFT         5
-#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK  0x1   /* cf6en */
-#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT 6
-#define USTORM_ETH_CONN_AG_CTX_RULE0EN_MASK               0x1   /* rule0en */
-#define USTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT              7
-	u8 flags3;
-#define USTORM_ETH_CONN_AG_CTX_RULE1EN_MASK               0x1   /* rule1en */
-#define USTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT              0
-#define USTORM_ETH_CONN_AG_CTX_RULE2EN_MASK               0x1   /* rule2en */
-#define USTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT              1
-#define USTORM_ETH_CONN_AG_CTX_RULE3EN_MASK               0x1   /* rule3en */
-#define USTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT              2
-#define USTORM_ETH_CONN_AG_CTX_RULE4EN_MASK               0x1   /* rule4en */
-#define USTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT              3
-#define USTORM_ETH_CONN_AG_CTX_RULE5EN_MASK               0x1   /* rule5en */
-#define USTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT              4
-#define USTORM_ETH_CONN_AG_CTX_RULE6EN_MASK               0x1   /* rule6en */
-#define USTORM_ETH_CONN_AG_CTX_RULE6EN_SHIFT              5
-#define USTORM_ETH_CONN_AG_CTX_RULE7EN_MASK               0x1   /* rule7en */
-#define USTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT              6
-#define USTORM_ETH_CONN_AG_CTX_RULE8EN_MASK               0x1   /* rule8en */
-#define USTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT              7
-	u8	byte2 /* byte2 */;
-	u8	byte3 /* byte3 */;
-	__le16	word0 /* conn_dpi */;
-	__le16	tx_bd_cons /* word1 */;
-	__le32	reg0 /* reg0 */;
-	__le32	reg1 /* reg1 */;
-	__le32	reg2 /* reg2 */;
-	__le32	reg3 /* reg3 */;
-	__le16	tx_drv_bd_cons /* word2 */;
-	__le16	rx_drv_cqe_cons /* word3 */;
-};
-
-struct xstorm_eth_hw_conn_ag_ctx {
-	u8	reserved0 /* cdu_validation */;
-	u8	eth_state /* state */;
-	u8	flags0;
-#define XSTORM_ETH_HW_CONN_AG_CTX_EXIST_IN_QM0_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EXIST_IN_QM0_SHIFT           0
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED1_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED1_SHIFT              1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED2_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED2_SHIFT              2
-#define XSTORM_ETH_HW_CONN_AG_CTX_EXIST_IN_QM3_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EXIST_IN_QM3_SHIFT           3
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED3_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED3_SHIFT              4
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED4_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED4_SHIFT              5
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED5_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED5_SHIFT              6
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED6_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED6_SHIFT              7
-	u8 flags1;
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED7_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED7_SHIFT              0
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED8_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED8_SHIFT              1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED9_MASK               0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED9_SHIFT              2
-#define XSTORM_ETH_HW_CONN_AG_CTX_BIT11_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_BIT11_SHIFT                  3
-#define XSTORM_ETH_HW_CONN_AG_CTX_BIT12_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_BIT12_SHIFT                  4
-#define XSTORM_ETH_HW_CONN_AG_CTX_BIT13_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_BIT13_SHIFT                  5
-#define XSTORM_ETH_HW_CONN_AG_CTX_TX_RULE_ACTIVE_MASK          0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_TX_RULE_ACTIVE_SHIFT         6
-#define XSTORM_ETH_HW_CONN_AG_CTX_DQ_CF_ACTIVE_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_DQ_CF_ACTIVE_SHIFT           7
-	u8 flags2;
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF0_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF0_SHIFT                    0
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF1_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF1_SHIFT                    2
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF2_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF2_SHIFT                    4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF3_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF3_SHIFT                    6
-	u8 flags3;
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF4_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF4_SHIFT                    0
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF5_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF5_SHIFT                    2
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF6_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF6_SHIFT                    4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF7_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF7_SHIFT                    6
-	u8 flags4;
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF8_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF8_SHIFT                    0
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF9_MASK                     0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF9_SHIFT                    2
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF10_MASK                    0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF10_SHIFT                   4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF11_MASK                    0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF11_SHIFT                   6
-	u8 flags5;
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF12_MASK                    0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF12_SHIFT                   0
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF13_MASK                    0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF13_SHIFT                   2
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF14_MASK                    0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF14_SHIFT                   4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF15_MASK                    0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF15_SHIFT                   6
-	u8 flags6;
-#define XSTORM_ETH_HW_CONN_AG_CTX_GO_TO_BD_CONS_CF_MASK        0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_GO_TO_BD_CONS_CF_SHIFT       0
-#define XSTORM_ETH_HW_CONN_AG_CTX_MULTI_UNICAST_CF_MASK        0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_MULTI_UNICAST_CF_SHIFT       2
-#define XSTORM_ETH_HW_CONN_AG_CTX_DQ_CF_MASK                   0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_DQ_CF_SHIFT                  4
-#define XSTORM_ETH_HW_CONN_AG_CTX_TERMINATE_CF_MASK            0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_TERMINATE_CF_SHIFT           6
-	u8 flags7;
-#define XSTORM_ETH_HW_CONN_AG_CTX_FLUSH_Q0_MASK                0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_FLUSH_Q0_SHIFT               0
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED10_MASK              0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED10_SHIFT             2
-#define XSTORM_ETH_HW_CONN_AG_CTX_SLOW_PATH_MASK               0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_SLOW_PATH_SHIFT              4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF0EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF0EN_SHIFT                  6
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF1EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF1EN_SHIFT                  7
-	u8 flags8;
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF2EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF2EN_SHIFT                  0
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF3EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF3EN_SHIFT                  1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF4EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF4EN_SHIFT                  2
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF5EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF5EN_SHIFT                  3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF6EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF6EN_SHIFT                  4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF7EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF7EN_SHIFT                  5
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF8EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF8EN_SHIFT                  6
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF9EN_MASK                   0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF9EN_SHIFT                  7
-	u8 flags9;
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF10EN_MASK                  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF10EN_SHIFT                 0
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF11EN_MASK                  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF11EN_SHIFT                 1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF12EN_MASK                  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF12EN_SHIFT                 2
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF13EN_MASK                  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF13EN_SHIFT                 3
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF14EN_MASK                  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF14EN_SHIFT                 4
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF15EN_MASK                  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_CF15EN_SHIFT                 5
-#define XSTORM_ETH_HW_CONN_AG_CTX_GO_TO_BD_CONS_CF_EN_MASK     0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_GO_TO_BD_CONS_CF_EN_SHIFT    6
-#define XSTORM_ETH_HW_CONN_AG_CTX_MULTI_UNICAST_CF_EN_MASK     0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_MULTI_UNICAST_CF_EN_SHIFT    7
-	u8 flags10;
-#define XSTORM_ETH_HW_CONN_AG_CTX_DQ_CF_EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_DQ_CF_EN_SHIFT               0
-#define XSTORM_ETH_HW_CONN_AG_CTX_TERMINATE_CF_EN_MASK         0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_TERMINATE_CF_EN_SHIFT        1
-#define XSTORM_ETH_HW_CONN_AG_CTX_FLUSH_Q0_EN_MASK             0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT            2
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED11_MASK              0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED11_SHIFT             3
-#define XSTORM_ETH_HW_CONN_AG_CTX_SLOW_PATH_EN_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_SLOW_PATH_EN_SHIFT           4
-#define XSTORM_ETH_HW_CONN_AG_CTX_TPH_ENABLE_EN_RESERVED_MASK  0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_TPH_ENABLE_EN_RESERVED_SHIFT 5
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED12_MASK              0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED12_SHIFT             6
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED13_MASK              0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED13_SHIFT             7
-	u8 flags11;
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED14_MASK              0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED14_SHIFT             0
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED15_MASK              0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RESERVED15_SHIFT             1
-#define XSTORM_ETH_HW_CONN_AG_CTX_TX_DEC_RULE_EN_MASK          0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_TX_DEC_RULE_EN_SHIFT         2
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE5EN_MASK                 0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE5EN_SHIFT                3
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE6EN_MASK                 0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE6EN_SHIFT                4
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE7EN_MASK                 0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE7EN_SHIFT                5
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED1_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED1_SHIFT           6
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE9EN_MASK                 0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE9EN_SHIFT                7
-	u8 flags12;
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE10EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE10EN_SHIFT               0
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE11EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE11EN_SHIFT               1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED2_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED2_SHIFT           2
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED3_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED3_SHIFT           3
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE14EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE14EN_SHIFT               4
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE15EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE15EN_SHIFT               5
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE16EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE16EN_SHIFT               6
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE17EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE17EN_SHIFT               7
-	u8 flags13;
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE18EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE18EN_SHIFT               0
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE19EN_MASK                0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_RULE19EN_SHIFT               1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED4_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED4_SHIFT           2
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED5_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED5_SHIFT           3
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED6_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED6_SHIFT           4
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED7_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED7_SHIFT           5
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED8_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED8_SHIFT           6
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED9_MASK            0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_A0_RESERVED9_SHIFT           7
-	u8 flags14;
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_USE_EXT_HDR_MASK        0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_USE_EXT_HDR_SHIFT       0
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_SEND_RAW_L3L4_MASK      0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_SEND_RAW_L3L4_SHIFT     1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_INBAND_PROP_HDR_MASK    0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_INBAND_PROP_HDR_SHIFT   2
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_SEND_EXT_TUNNEL_MASK    0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_EDPM_SEND_EXT_TUNNEL_SHIFT   3
-#define XSTORM_ETH_HW_CONN_AG_CTX_L2_EDPM_ENABLE_MASK          0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_L2_EDPM_ENABLE_SHIFT         4
-#define XSTORM_ETH_HW_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK        0x1
-#define XSTORM_ETH_HW_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT       5
-#define XSTORM_ETH_HW_CONN_AG_CTX_TPH_ENABLE_MASK              0x3
-#define XSTORM_ETH_HW_CONN_AG_CTX_TPH_ENABLE_SHIFT             6
-	u8	edpm_event_id /* byte2 */;
-	__le16	physical_q0 /* physical_q0 */;
-	__le16	word1 /* physical_q1 */;
-	__le16	edpm_num_bds /* physical_q2 */;
-	__le16	tx_bd_cons /* word3 */;
-	__le16	tx_bd_prod /* word4 */;
-	__le16	go_to_bd_cons /* word5 */;
-	__le16	conn_dpi /* conn_dpi */;
-};
-
 #define VF_MAX_STATIC 192       /* In case of K2 */
 
 #define MCP_GLOB_PATH_MAX       2
@@ -3818,6 +3605,10 @@ struct public_port {
 	struct dcbx_local_params		local_admin_dcbx_mib;
 	struct dcbx_mib				remote_dcbx_mib;
 	struct dcbx_mib				operational_dcbx_mib;
+
+	u32					fc_npiv_nvram_tbl_addr;
+	u32					fc_npiv_nvram_tbl_size;
+	u32					transceiver_data;
 };
 
 /**************************************/
@@ -3830,7 +3621,11 @@ struct public_func {
 	u32	iscsi_boot_signature;
 	u32	iscsi_boot_block_offset;
 
-	u32	reserved[8];
+	u32	mtu_size;
+	u32	c2s_pcp_map_lower;
+	u32	c2s_pcp_map_upper;
+	u32	c2s_pcp_map_default;
+	u32	reserved[4];
 
 	u32	config;
 
@@ -3894,10 +3689,10 @@ struct public_func {
 #define DRV_ID_MCP_HSI_VER_SHIFT        16
 #define DRV_ID_MCP_HSI_VER_CURRENT	BIT(DRV_ID_MCP_HSI_VER_SHIFT)
 
-#define DRV_ID_DRV_TYPE_MASK            0xff000000
+#define DRV_ID_DRV_TYPE_MASK            0x7f000000
 #define DRV_ID_DRV_TYPE_SHIFT           24
 #define DRV_ID_DRV_TYPE_UNKNOWN         (0 << DRV_ID_DRV_TYPE_SHIFT)
-#define DRV_ID_DRV_TYPE_LINUX		BIT(DRV_ID_DRV_TYPE_SHIFT)
+#define DRV_ID_DRV_TYPE_LINUX           (1 << DRV_ID_DRV_TYPE_SHIFT)
 #define DRV_ID_DRV_TYPE_WINDOWS         (2 << DRV_ID_DRV_TYPE_SHIFT)
 #define DRV_ID_DRV_TYPE_DIAG            (3 << DRV_ID_DRV_TYPE_SHIFT)
 #define DRV_ID_DRV_TYPE_PREBOOT         (4 << DRV_ID_DRV_TYPE_SHIFT)
@@ -3905,6 +3700,10 @@ struct public_func {
 #define DRV_ID_DRV_TYPE_VMWARE          (6 << DRV_ID_DRV_TYPE_SHIFT)
 #define DRV_ID_DRV_TYPE_FREEBSD         (7 << DRV_ID_DRV_TYPE_SHIFT)
 #define DRV_ID_DRV_TYPE_AIX             (8 << DRV_ID_DRV_TYPE_SHIFT)
+
+#define DRV_ID_DRV_INIT_HW_MASK         0x80000000
+#define DRV_ID_DRV_INIT_HW_SHIFT        31
+#define DRV_ID_DRV_INIT_HW_FLAG         BIT(DRV_ID_DRV_INIT_HW_SHIFT)
 };
 
 /**************************************/
@@ -3964,6 +3763,7 @@ struct public_drv_mb {
 #define DRV_MSG_CODE_MASK                       0xffff0000
 #define DRV_MSG_CODE_LOAD_REQ                   0x10000000
 #define DRV_MSG_CODE_LOAD_DONE                  0x11000000
+#define DRV_MSG_CODE_INIT_HW                    0x12000000
 #define DRV_MSG_CODE_UNLOAD_REQ                 0x20000000
 #define DRV_MSG_CODE_UNLOAD_DONE                0x21000000
 #define DRV_MSG_CODE_INIT_PHY                   0x22000000
@@ -4100,6 +3900,7 @@ struct public_drv_mb {
 #define FW_MSG_CODE_SET_SECURE_MODE_ERROR       0x00130000
 #define FW_MSG_CODE_SET_SECURE_MODE_OK          0x00140000
 #define FW_MSG_MODE_PHY_PRIVILEGE_ERROR         0x00150000
+#define FW_MSG_CODE_OK                          0x00160000
 
 #define FW_MSG_SEQ_NUMBER_MASK                  0x0000ffff
 
@@ -4212,7 +4013,7 @@ struct nvm_cfg1_glob {
 #define NVM_CFG1_GLOB_MF_MODE_MASK                              0x00000FF0
 #define NVM_CFG1_GLOB_MF_MODE_OFFSET                            4
 #define NVM_CFG1_GLOB_MF_MODE_MF_ALLOWED                        0x0
-#define NVM_CFG1_GLOB_MF_MODE_FORCED_SF                         0x1
+#define NVM_CFG1_GLOB_MF_MODE_DEFAULT                           0x1
 #define NVM_CFG1_GLOB_MF_MODE_SPIO4                             0x2
 #define NVM_CFG1_GLOB_MF_MODE_NPAR1_0                           0x3
 #define NVM_CFG1_GLOB_MF_MODE_NPAR1_5                           0x4
@@ -4643,8 +4444,12 @@ struct nvm_cfg1_glob {
 #define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO29                  0x1E
 #define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO30                  0x1F
 #define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO31                  0x20
-
-	u32 reserved[46];					/* 0x88 */
+	u32	device_capabilities;                            /* 0x88 */
+#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET              0x1
+	u32	power_dissipated;                               /* 0x8C */
+	u32 power_consumed;                                     /* 0x90 */
+	u32	efi_version;                                    /* 0x94 */
+	u32	reserved[42];                                   /* 0x98 */
 };
 
 struct nvm_cfg1_path {
@@ -4652,26 +4457,8 @@ struct nvm_cfg1_path {
 };
 
 struct nvm_cfg1_port {
-	u32 power_dissipated;					/* 0x0 */
-#define NVM_CFG1_PORT_POWER_DIS_D0_MASK                         0x000000FF
-#define NVM_CFG1_PORT_POWER_DIS_D0_OFFSET                       0
-#define NVM_CFG1_PORT_POWER_DIS_D1_MASK                         0x0000FF00
-#define NVM_CFG1_PORT_POWER_DIS_D1_OFFSET                       8
-#define NVM_CFG1_PORT_POWER_DIS_D2_MASK                         0x00FF0000
-#define NVM_CFG1_PORT_POWER_DIS_D2_OFFSET                       16
-#define NVM_CFG1_PORT_POWER_DIS_D3_MASK                         0xFF000000
-#define NVM_CFG1_PORT_POWER_DIS_D3_OFFSET                       24
-
-	u32 power_consumed;					/* 0x4 */
-#define NVM_CFG1_PORT_POWER_CONS_D0_MASK                        0x000000FF
-#define NVM_CFG1_PORT_POWER_CONS_D0_OFFSET                      0
-#define NVM_CFG1_PORT_POWER_CONS_D1_MASK                        0x0000FF00
-#define NVM_CFG1_PORT_POWER_CONS_D1_OFFSET                      8
-#define NVM_CFG1_PORT_POWER_CONS_D2_MASK                        0x00FF0000
-#define NVM_CFG1_PORT_POWER_CONS_D2_OFFSET                      16
-#define NVM_CFG1_PORT_POWER_CONS_D3_MASK                        0xFF000000
-#define NVM_CFG1_PORT_POWER_CONS_D3_OFFSET                      24
-
+	u32	reserved__m_relocated_to_option_123;           /* 0x0 */
+	u32	reserved__m_relocated_to_option_124;           /* 0x4 */
 	u32 generic_cont0;					/* 0x8 */
 #define NVM_CFG1_PORT_LED_MODE_MASK                             0x000000FF
 #define NVM_CFG1_PORT_LED_MODE_OFFSET                           0
@@ -4699,7 +4486,9 @@ struct nvm_cfg1_port {
 #define NVM_CFG1_PORT_DCBX_MODE_IEEE                            0x1
 #define NVM_CFG1_PORT_DCBX_MODE_CEE                             0x2
 #define NVM_CFG1_PORT_DCBX_MODE_DYNAMIC                         0x3
-
+#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_MASK            0x00F00000
+#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_OFFSET          20
+#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_ETHERNET        0x1
 	u32	pcie_cfg;					/* 0xC */
 #define NVM_CFG1_PORT_RESERVED15_MASK                           0x00000007
 #define NVM_CFG1_PORT_RESERVED15_OFFSET                         0
@@ -4784,10 +4573,11 @@ struct nvm_cfg1_port {
 #define NVM_CFG1_PORT_SERDES_NET_INTERFACE_SFI                  0x9
 #define NVM_CFG1_PORT_SERDES_NET_INTERFACE_1000X                0xB
 #define NVM_CFG1_PORT_SERDES_NET_INTERFACE_SGMII                0xC
-#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XLAUI                0xD
-#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_CAUI                 0xE
-#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XLPPI                0xF
-#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_CPPI                 0x10
+#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XLAUI                0x11
+#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XLPPI                0x12
+#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_CAUI                 0x21
+#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_CPPI                 0x22
+#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_25GAUI               0x31
 #define NVM_CFG1_PORT_AN_MODE_MASK                              0xFF000000
 #define NVM_CFG1_PORT_AN_MODE_OFFSET                            24
 #define NVM_CFG1_PORT_AN_MODE_NONE                              0x0
@@ -4801,9 +4591,6 @@ struct nvm_cfg1_port {
 	u32 mgmt_traffic;					/* 0x20 */
 #define NVM_CFG1_PORT_RESERVED61_MASK                           0x0000000F
 #define NVM_CFG1_PORT_RESERVED61_OFFSET                         0
-#define NVM_CFG1_PORT_RESERVED61_DISABLED                       0x0
-#define NVM_CFG1_PORT_RESERVED61_NCSI_OVER_RMII                 0x1
-#define NVM_CFG1_PORT_RESERVED61_NCSI_OVER_SMBUS                0x2
 
 	u32 ext_phy;						/* 0x24 */
 #define NVM_CFG1_PORT_EXTERNAL_PHY_TYPE_MASK                    0x000000FF
@@ -4814,16 +4601,12 @@ struct nvm_cfg1_port {
 #define NVM_CFG1_PORT_EXTERNAL_PHY_ADDRESS_OFFSET               8
 
 	u32 mba_cfg1;						/* 0x28 */
-#define NVM_CFG1_PORT_MBA_MASK                                  0x00000001
-#define NVM_CFG1_PORT_MBA_OFFSET                                0
-#define NVM_CFG1_PORT_MBA_DISABLED                              0x0
-#define NVM_CFG1_PORT_MBA_ENABLED                               0x1
-#define NVM_CFG1_PORT_MBA_BOOT_TYPE_MASK                        0x00000006
-#define NVM_CFG1_PORT_MBA_BOOT_TYPE_OFFSET                      1
-#define NVM_CFG1_PORT_MBA_BOOT_TYPE_AUTO                        0x0
-#define NVM_CFG1_PORT_MBA_BOOT_TYPE_BBS                         0x1
-#define NVM_CFG1_PORT_MBA_BOOT_TYPE_INT18H                      0x2
-#define NVM_CFG1_PORT_MBA_BOOT_TYPE_INT19H                      0x3
+#define NVM_CFG1_PORT_PREBOOT_OPROM_MASK                        0x00000001
+#define NVM_CFG1_PORT_PREBOOT_OPROM_OFFSET                      0
+#define NVM_CFG1_PORT_PREBOOT_OPROM_DISABLED                    0x0
+#define NVM_CFG1_PORT_PREBOOT_OPROM_ENABLED                     0x1
+#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_TYPE_MASK            0x00000006
+#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_TYPE_OFFSET          1
 #define NVM_CFG1_PORT_MBA_DELAY_TIME_MASK                       0x00000078
 #define NVM_CFG1_PORT_MBA_DELAY_TIME_OFFSET                     3
 #define NVM_CFG1_PORT_MBA_SETUP_HOT_KEY_MASK                    0x00000080
@@ -4836,61 +4619,30 @@ struct nvm_cfg1_port {
 #define NVM_CFG1_PORT_MBA_HIDE_SETUP_PROMPT_ENABLED             0x1
 #define NVM_CFG1_PORT_RESERVED5_MASK                            0x0001FE00
 #define NVM_CFG1_PORT_RESERVED5_OFFSET                          9
-#define NVM_CFG1_PORT_RESERVED5_DISABLED                        0x0
-#define NVM_CFG1_PORT_RESERVED5_2K                              0x1
-#define NVM_CFG1_PORT_RESERVED5_4K                              0x2
-#define NVM_CFG1_PORT_RESERVED5_8K                              0x3
-#define NVM_CFG1_PORT_RESERVED5_16K                             0x4
-#define NVM_CFG1_PORT_RESERVED5_32K                             0x5
-#define NVM_CFG1_PORT_RESERVED5_64K                             0x6
-#define NVM_CFG1_PORT_RESERVED5_128K                            0x7
-#define NVM_CFG1_PORT_RESERVED5_256K                            0x8
-#define NVM_CFG1_PORT_RESERVED5_512K                            0x9
-#define NVM_CFG1_PORT_RESERVED5_1M                              0xA
-#define NVM_CFG1_PORT_RESERVED5_2M                              0xB
-#define NVM_CFG1_PORT_RESERVED5_4M                              0xC
-#define NVM_CFG1_PORT_RESERVED5_8M                              0xD
-#define NVM_CFG1_PORT_RESERVED5_16M                             0xE
-#define NVM_CFG1_PORT_RESERVED5_32M                             0xF
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_MASK                       0x001E0000
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_OFFSET                     17
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_AUTONEG                    0x0
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_1G                         0x1
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_10G                        0x2
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_25G                        0x4
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_40G                        0x5
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_50G                        0x6
-#define NVM_CFG1_PORT_MBA_LINK_SPEED_100G                       0x7
-#define NVM_CFG1_PORT_MBA_BOOT_RETRY_COUNT_MASK                 0x00E00000
-#define NVM_CFG1_PORT_MBA_BOOT_RETRY_COUNT_OFFSET               21
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_MASK                   0x001E0000
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_OFFSET                 17
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_AUTONEG                0x0
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_1G                     0x1
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_10G                    0x2
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_25G                    0x4
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_40G                    0x5
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_50G                    0x6
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_100G                   0x7
+#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_SMARTLINQ              0x8
+#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_RETRY_COUNT_MASK     0x00E00000
+#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_RETRY_COUNT_OFFSET   21
 
 	u32	mba_cfg2;					/* 0x2C */
-#define NVM_CFG1_PORT_MBA_VLAN_VALUE_MASK                       0x0000FFFF
-#define NVM_CFG1_PORT_MBA_VLAN_VALUE_OFFSET                     0
-#define NVM_CFG1_PORT_MBA_VLAN_MASK                             0x00010000
-#define NVM_CFG1_PORT_MBA_VLAN_OFFSET                           16
+#define NVM_CFG1_PORT_RESERVED65_MASK                           0x0000FFFF
+#define NVM_CFG1_PORT_RESERVED65_OFFSET                         0
+#define NVM_CFG1_PORT_RESERVED66_MASK                           0x00010000
+#define NVM_CFG1_PORT_RESERVED66_OFFSET                         16
 
 	u32	vf_cfg;						/* 0x30 */
 #define NVM_CFG1_PORT_RESERVED8_MASK                            0x0000FFFF
 #define NVM_CFG1_PORT_RESERVED8_OFFSET                          0
 #define NVM_CFG1_PORT_RESERVED6_MASK                            0x000F0000
 #define NVM_CFG1_PORT_RESERVED6_OFFSET                          16
-#define NVM_CFG1_PORT_RESERVED6_DISABLED                        0x0
-#define NVM_CFG1_PORT_RESERVED6_4K                              0x1
-#define NVM_CFG1_PORT_RESERVED6_8K                              0x2
-#define NVM_CFG1_PORT_RESERVED6_16K                             0x3
-#define NVM_CFG1_PORT_RESERVED6_32K                             0x4
-#define NVM_CFG1_PORT_RESERVED6_64K                             0x5
-#define NVM_CFG1_PORT_RESERVED6_128K                            0x6
-#define NVM_CFG1_PORT_RESERVED6_256K                            0x7
-#define NVM_CFG1_PORT_RESERVED6_512K                            0x8
-#define NVM_CFG1_PORT_RESERVED6_1M                              0x9
-#define NVM_CFG1_PORT_RESERVED6_2M                              0xA
-#define NVM_CFG1_PORT_RESERVED6_4M                              0xB
-#define NVM_CFG1_PORT_RESERVED6_8M                              0xC
-#define NVM_CFG1_PORT_RESERVED6_16M                             0xD
-#define NVM_CFG1_PORT_RESERVED6_32M                             0xE
-#define NVM_CFG1_PORT_RESERVED6_64M                             0xF
 
 	struct nvm_cfg_mac_address	lldp_mac_address;	/* 0x34 */
 
@@ -4973,18 +4725,16 @@ struct nvm_cfg1_func {
 	u32				device_id;		/* 0x10 */
 #define NVM_CFG1_FUNC_MF_VENDOR_DEVICE_ID_MASK                  0x0000FFFF
 #define NVM_CFG1_FUNC_MF_VENDOR_DEVICE_ID_OFFSET                0
-#define NVM_CFG1_FUNC_VENDOR_DEVICE_ID_MASK                     0xFFFF0000
-#define NVM_CFG1_FUNC_VENDOR_DEVICE_ID_OFFSET                   16
+#define NVM_CFG1_FUNC_RESERVED77_MASK                           0xFFFF0000
+#define NVM_CFG1_FUNC_RESERVED77_OFFSET                         16
 
 	u32				cmn_cfg;		/* 0x14 */
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_MASK                    0x00000007
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_OFFSET                  0
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_PXE                     0x0
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_RPL                     0x1
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_BOOTP                   0x2
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_ISCSI_BOOT              0x3
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_FCOE_BOOT               0x4
-#define NVM_CFG1_FUNC_MBA_BOOT_PROTOCOL_NONE                    0x7
+#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_MASK                0x00000007
+#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_OFFSET              0
+#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_PXE                 0x0
+#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_ISCSI_BOOT          0x3
+#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_FCOE_BOOT           0x4
+#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_NONE                0x7
 #define NVM_CFG1_FUNC_VF_PCI_DEVICE_ID_MASK                     0x0007FFF8
 #define NVM_CFG1_FUNC_VF_PCI_DEVICE_ID_OFFSET                   3
 #define NVM_CFG1_FUNC_PERSONALITY_MASK                          0x00780000
@@ -5029,8 +4779,8 @@ struct nvm_cfg1_func {
 	struct nvm_cfg_mac_address	fcoe_node_wwn_mac_addr;	/* 0x1C */
 
 	struct nvm_cfg_mac_address	fcoe_port_wwn_mac_addr;	/* 0x24 */
-
-	u32				reserved[9];		/* 0x2C */
+	u32				preboot_generic_cfg;    /* 0x2C */
+	u32				reserved[8];            /* 0x30 */
 };
 
 struct nvm_cfg1 {
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
index 0b21a55..f55ebdc 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c
@@ -513,17 +513,14 @@ static int qed_pf_rl_rt_init(struct qed_hwfn *p_hwfn,
  * Return -1 on error.
  */
 static int qed_vp_wfq_rt_init(struct qed_hwfn *p_hwfn,
-			      u8 start_vport,
 			      u8 num_vports,
 			      struct init_qm_vport_params *vport_params)
 {
-	u8 tc, i, vport_id;
 	u32 inc_val;
+	u8 tc, i;
 
 	/* go over all PF VPORTs */
-	for (i = 0, vport_id = start_vport; i < num_vports; i++, vport_id++) {
-		u32 temp = QM_REG_WFQVPUPPERBOUND_RT_OFFSET;
-		u16 *pq_ids = &vport_params[i].first_tx_pq_id[0];
+	for (i = 0; i < num_vports; i++) {
 
 		if (!vport_params[i].vport_wfq)
 			continue;
@@ -539,20 +536,16 @@ static int qed_vp_wfq_rt_init(struct qed_hwfn *p_hwfn,
 		 * different TCs
 		 */
 		for (tc = 0; tc < NUM_OF_TCS; tc++) {
-			u16 vport_pq_id = pq_ids[tc];
+			u16 vport_pq_id = vport_params[i].first_tx_pq_id[tc];
 
 			if (vport_pq_id != QM_INVALID_PQ_ID) {
 				STORE_RT_REG(p_hwfn,
-					     QM_REG_WFQVPWEIGHT_RT_OFFSET +
-					     vport_pq_id, inc_val);
-				STORE_RT_REG(p_hwfn, temp + vport_pq_id,
-					     QM_WFQ_UPPER_BOUND |
-					     QM_WFQ_CRD_REG_SIGN_BIT);
-				STORE_RT_REG(p_hwfn,
 					     QM_REG_WFQVPCRD_RT_OFFSET +
 					     vport_pq_id,
-					     QM_WFQ_INIT_CRD(inc_val) |
 					     QM_WFQ_CRD_REG_SIGN_BIT);
+				STORE_RT_REG(p_hwfn,
+					     QM_REG_WFQVPWEIGHT_RT_OFFSET +
+					     vport_pq_id, inc_val);
 			}
 		}
 	}
@@ -709,8 +702,7 @@ int qed_qm_pf_rt_init(struct qed_hwfn *p_hwfn,
 	if (qed_pf_rl_rt_init(p_hwfn, p_params->pf_id, p_params->pf_rl))
 		return -1;
 
-	if (qed_vp_wfq_rt_init(p_hwfn, p_params->start_vport,
-			       p_params->num_vports, vport_params))
+	if (qed_vp_wfq_rt_init(p_hwfn, p_params->num_vports, vport_params))
 		return -1;
 
 	if (qed_vport_rl_rt_init(p_hwfn, p_params->start_vport,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
index 796f139..3269b36 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c
@@ -55,63 +55,98 @@ void qed_init_clear_rt_data(struct qed_hwfn *p_hwfn)
 	int i;
 
 	for (i = 0; i < RUNTIME_ARRAY_SIZE; i++)
-		p_hwfn->rt_data[i].b_valid = false;
+		p_hwfn->rt_data.b_valid[i] = false;
 }
 
 void qed_init_store_rt_reg(struct qed_hwfn *p_hwfn,
 			   u32 rt_offset,
 			   u32 val)
 {
-	p_hwfn->rt_data[rt_offset].init_val = val;
-	p_hwfn->rt_data[rt_offset].b_valid = true;
+	p_hwfn->rt_data.init_val[rt_offset] = val;
+	p_hwfn->rt_data.b_valid[rt_offset] = true;
 }
 
 void qed_init_store_rt_agg(struct qed_hwfn *p_hwfn,
-			   u32 rt_offset,
-			   u32 *val,
+			   u32 rt_offset, u32 *p_val,
 			   size_t size)
 {
 	size_t i;
 
 	for (i = 0; i < size / sizeof(u32); i++) {
-		p_hwfn->rt_data[rt_offset + i].init_val = val[i];
-		p_hwfn->rt_data[rt_offset + i].b_valid = true;
+		p_hwfn->rt_data.init_val[rt_offset + i] = p_val[i];
+		p_hwfn->rt_data.b_valid[rt_offset + i]	= true;
 	}
 }
 
-static void qed_init_rt(struct qed_hwfn *p_hwfn,
-			struct qed_ptt *p_ptt,
-			u32 addr,
-			u32 rt_offset,
-			u32 size)
+static int qed_init_rt(struct qed_hwfn	*p_hwfn,
+		       struct qed_ptt *p_ptt,
+		       u32 addr,
+		       u16 rt_offset,
+		       u16 size,
+		       bool b_must_dmae)
 {
-	struct qed_rt_data *rt_data = p_hwfn->rt_data + rt_offset;
-	u32 i;
+	u32 *p_init_val = &p_hwfn->rt_data.init_val[rt_offset];
+	bool *p_valid = &p_hwfn->rt_data.b_valid[rt_offset];
+	u16 i, segment;
+	int rc = 0;
 
+	/* Since not all RT entries are initialized, go over the RT and
+	 * for each segment of initialized values use DMA.
+	 */
 	for (i = 0; i < size; i++) {
-		if (!rt_data[i].b_valid)
+		if (!p_valid[i])
 			continue;
-		qed_wr(p_hwfn, p_ptt, addr + (i << 2), rt_data[i].init_val);
+
+		/* In case there isn't any wide-bus configuration here,
+		 * simply write the data instead of using dmae.
+		 */
+		if (!b_must_dmae) {
+			qed_wr(p_hwfn, p_ptt, addr + (i << 2),
+			       p_init_val[i]);
+			continue;
+		}
+
+		/* Start of a new segment */
+		for (segment = 1; i + segment < size; segment++)
+			if (!p_valid[i + segment])
+				break;
+
+		rc = qed_dmae_host2grc(p_hwfn, p_ptt,
+				       (uintptr_t)(p_init_val + i),
+				       addr + (i << 2), segment, 0);
+		if (rc != 0)
+			return rc;
+
+		/* Jump over the entire segment, including invalid entry */
+		i += segment;
 	}
+
+	return rc;
 }
 
 int qed_init_alloc(struct qed_hwfn *p_hwfn)
 {
-	struct qed_rt_data *rt_data;
+	struct qed_rt_data *rt_data = &p_hwfn->rt_data;
 
-	rt_data = kzalloc(sizeof(*rt_data) * RUNTIME_ARRAY_SIZE, GFP_ATOMIC);
-	if (!rt_data)
+	rt_data->b_valid = kzalloc(sizeof(bool) * RUNTIME_ARRAY_SIZE,
+				   GFP_KERNEL);
+	if (!rt_data->b_valid)
 		return -ENOMEM;
 
-	p_hwfn->rt_data = rt_data;
+	rt_data->init_val = kzalloc(sizeof(u32) * RUNTIME_ARRAY_SIZE,
+				    GFP_KERNEL);
+	if (!rt_data->init_val) {
+		kfree(rt_data->b_valid);
+		return -ENOMEM;
+	}
 
 	return 0;
 }
 
 void qed_init_free(struct qed_hwfn *p_hwfn)
 {
-	kfree(p_hwfn->rt_data);
-	p_hwfn->rt_data = NULL;
+	kfree(p_hwfn->rt_data.init_val);
+	kfree(p_hwfn->rt_data.b_valid);
 }
 
 static int qed_init_array_dmae(struct qed_hwfn *p_hwfn,
@@ -289,7 +324,8 @@ static int qed_init_cmd_wr(struct qed_hwfn *p_hwfn,
 	case INIT_SRC_RUNTIME:
 		qed_init_rt(p_hwfn, p_ptt, addr,
 			    le16_to_cpu(arg->runtime.offset),
-			    le16_to_cpu(arg->runtime.size));
+			    le16_to_cpu(arg->runtime.size),
+			    b_must_dmae);
 		break;
 	}
 
@@ -316,49 +352,50 @@ static void qed_init_cmd_rd(struct qed_hwfn *p_hwfn,
 			    struct qed_ptt *p_ptt,
 			    struct init_read_op *cmd)
 {
-	u32 data = le32_to_cpu(cmd->op_data);
-	u32 addr = GET_FIELD(data, INIT_READ_OP_ADDRESS) << 2;
+	bool (*comp_check)(u32 val, u32 expected_val);
+	u32 delay = QED_INIT_POLL_PERIOD_US, val;
+	u32 data, addr, poll;
+	int i;
+
+	data = le32_to_cpu(cmd->op_data);
+	addr = GET_FIELD(data, INIT_READ_OP_ADDRESS) << 2;
+	poll = GET_FIELD(data, INIT_READ_OP_POLL_TYPE);
 
-	bool	(*comp_check)(u32	val,
-			      u32	expected_val);
-	u32	delay = QED_INIT_POLL_PERIOD_US, val;
 
 	val = qed_rd(p_hwfn, p_ptt, addr);
 
-	data = le32_to_cpu(cmd->op_data);
-	if (GET_FIELD(data, INIT_READ_OP_POLL)) {
-		int i;
+	if (poll == INIT_POLL_NONE)
+		return;
 
-		switch (GET_FIELD(data, INIT_READ_OP_POLL_COMP)) {
-		case INIT_COMPARISON_EQ:
-			comp_check = comp_eq;
-			break;
-		case INIT_COMPARISON_OR:
-			comp_check = comp_or;
-			break;
-		case INIT_COMPARISON_AND:
-			comp_check = comp_and;
-			break;
-		default:
-			comp_check = NULL;
-			DP_ERR(p_hwfn, "Invalid poll comparison type %08x\n",
-			       data);
-			return;
-		}
+	switch (poll) {
+	case INIT_POLL_EQ:
+		comp_check = comp_eq;
+		break;
+	case INIT_POLL_OR:
+		comp_check = comp_or;
+		break;
+	case INIT_POLL_AND:
+		comp_check = comp_and;
+		break;
+	default:
+		DP_ERR(p_hwfn, "Invalid poll comparison type %08x\n",
+		       cmd->op_data);
+		return;
+	}
 
-		for (i = 0;
-		     i < QED_INIT_MAX_POLL_COUNT &&
-		     !comp_check(val, le32_to_cpu(cmd->expected_val));
-		     i++) {
-			udelay(delay);
-			val = qed_rd(p_hwfn, p_ptt, addr);
-		}
+	data = le32_to_cpu(cmd->expected_val);
+	for (i = 0;
+	     i < QED_INIT_MAX_POLL_COUNT && !comp_check(val, data);
+	     i++) {
+		udelay(delay);
+		val = qed_rd(p_hwfn, p_ptt, addr);
+	}
 
-		if (i == QED_INIT_MAX_POLL_COUNT)
-			DP_ERR(p_hwfn,
-			       "Timeout when polling reg: 0x%08x [ Waiting-for: %08x Got: %08x (comparsion %08x)]\n",
-			       addr, le32_to_cpu(cmd->expected_val),
-			       val, data);
+	if (i == QED_INIT_MAX_POLL_COUNT) {
+		DP_ERR(p_hwfn,
+		       "Timeout when polling reg: 0x%08x [ Waiting-for: %08x Got: %08x (comparsion %08x)]\n",
+		       addr, le32_to_cpu(cmd->expected_val),
+		       val, le32_to_cpu(cmd->op_data));
 	}
 }
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index f72036a..978d07a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -714,7 +714,6 @@ qed_sp_eth_txq_start_ramrod(struct qed_hwfn  *p_hwfn,
 	p_ramrod->sb_id			= cpu_to_le16(p_params->sb);
 	p_ramrod->sb_index		= p_params->sb_idx;
 	p_ramrod->stats_counter_id	= stats_id;
-	p_ramrod->tc			= p_pq_params->eth.tc;
 
 	p_ramrod->pbl_size		= cpu_to_le16(pbl_size);
 	p_ramrod->pbl_base_addr.hi	= DMA_HI_LE(pbl_addr);
@@ -821,9 +820,8 @@ qed_filter_action(enum qed_filter_opcode opcode)
 	case QED_FILTER_REMOVE:
 		action = ETH_FILTER_ACTION_REMOVE;
 		break;
-	case QED_FILTER_REPLACE:
 	case QED_FILTER_FLUSH:
-		action = ETH_FILTER_ACTION_REPLACE;
+		action = ETH_FILTER_ACTION_REMOVE_ALL;
 		break;
 	default:
 		action = MAX_ETH_FILTER_ACTION;
@@ -892,8 +890,7 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn,
 	p_ramrod->filter_cmd_hdr.tx = p_filter_cmd->is_tx_filter ? 1 : 0;
 
 	switch (p_filter_cmd->opcode) {
-	case QED_FILTER_FLUSH:
-		p_ramrod->filter_cmd_hdr.cmd_cnt = 0; break;
+	case QED_FILTER_REPLACE:
 	case QED_FILTER_MOVE:
 		p_ramrod->filter_cmd_hdr.cmd_cnt = 2; break;
 	default:
@@ -962,6 +959,12 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn,
 
 		p_second_filter->action		= ETH_FILTER_ACTION_ADD;
 		p_second_filter->vport_id	= vport_to_add_to;
+	} else if (p_filter_cmd->opcode == QED_FILTER_REPLACE) {
+		p_first_filter->vport_id = vport_to_add_to;
+		memcpy(p_second_filter, p_first_filter,
+		       sizeof(*p_second_filter));
+		p_first_filter->action	= ETH_FILTER_ACTION_REMOVE_ALL;
+		p_second_filter->action = ETH_FILTER_ACTION_ADD;
 	} else {
 		action = qed_filter_action(p_filter_cmd->opcode);
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index 9d76ce2..593f887 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -190,7 +190,7 @@ int qed_fill_dev_info(struct qed_dev *cdev,
 	dev_info->pci_mem_start = cdev->pci_params.mem_start;
 	dev_info->pci_mem_end = cdev->pci_params.mem_end;
 	dev_info->pci_irq = cdev->pci_params.irq;
-	dev_info->is_mf = IS_MF(&cdev->hwfns[0]);
+	dev_info->is_mf_default = IS_MF_DEFAULT(&cdev->hwfns[0]);
 	ether_addr_copy(dev_info->hw_mac, cdev->hwfns[0].hw_info.hw_mac_addr);
 
 	dev_info->fw_major = FW_MAJOR_VERSION;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index ba1b1f1..1457e30 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -720,26 +720,25 @@ int qed_mcp_fill_shmem_func_info(struct qed_hwfn *p_hwfn,
 		return -EINVAL;
 	}
 
-	if (p_hwfn->cdev->mf_mode != SF) {
-		info->bandwidth_min = (shmem_info.config &
-				       FUNC_MF_CFG_MIN_BW_MASK) >>
-				      FUNC_MF_CFG_MIN_BW_SHIFT;
-		if (info->bandwidth_min < 1 || info->bandwidth_min > 100) {
-			DP_INFO(p_hwfn,
-				"bandwidth minimum out of bounds [%02x]. Set to 1\n",
-				info->bandwidth_min);
-			info->bandwidth_min = 1;
-		}
 
-		info->bandwidth_max = (shmem_info.config &
-				       FUNC_MF_CFG_MAX_BW_MASK) >>
-				      FUNC_MF_CFG_MAX_BW_SHIFT;
-		if (info->bandwidth_max < 1 || info->bandwidth_max > 100) {
-			DP_INFO(p_hwfn,
-				"bandwidth maximum out of bounds [%02x]. Set to 100\n",
-				info->bandwidth_max);
-			info->bandwidth_max = 100;
-		}
+	info->bandwidth_min = (shmem_info.config &
+			       FUNC_MF_CFG_MIN_BW_MASK) >>
+			      FUNC_MF_CFG_MIN_BW_SHIFT;
+	if (info->bandwidth_min < 1 || info->bandwidth_min > 100) {
+		DP_INFO(p_hwfn,
+			"bandwidth minimum out of bounds [%02x]. Set to 1\n",
+			info->bandwidth_min);
+		info->bandwidth_min = 1;
+	}
+
+	info->bandwidth_max = (shmem_info.config &
+			       FUNC_MF_CFG_MAX_BW_MASK) >>
+			      FUNC_MF_CFG_MAX_BW_SHIFT;
+	if (info->bandwidth_max < 1 || info->bandwidth_max > 100) {
+		DP_INFO(p_hwfn,
+			"bandwidth maximum out of bounds [%02x]. Set to 100\n",
+			info->bandwidth_max);
+		info->bandwidth_max = 100;
 	}
 
 	if (shmem_info.mac_upper || shmem_info.mac_lower) {
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h
index 287fadf..8a83609 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h
@@ -343,7 +343,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
  */
 
 int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
-		    enum mf_mode mode);
+		    enum qed_mf_mode mode);
 
 /**
  * @brief qed_sp_pf_stop - PF Function Stop Ramrod
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
index 6f78791..33090f6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
@@ -90,7 +90,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 }
 
 int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
-		    enum mf_mode mode)
+		    enum qed_mf_mode mode)
 {
 	struct qed_sp_init_request_params params;
 	struct pf_start_ramrod_data *p_ramrod = NULL;
@@ -125,6 +125,18 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
 	p_ramrod->dont_log_ramrods	= 0;
 	p_ramrod->log_type_mask		= cpu_to_le16(0xf);
 	p_ramrod->mf_mode = mode;
+	switch (mode) {
+	case QED_MF_DEFAULT:
+	case QED_MF_NPAR:
+		p_ramrod->mf_mode = MF_NPAR;
+		break;
+	case QED_MF_OVLAN:
+		p_ramrod->mf_mode = MF_OVLAN;
+		break;
+	default:
+		DP_NOTICE(p_hwfn, "Unsupported MF mode, init as DEFAULT\n");
+		p_ramrod->mf_mode = MF_NPAR;
+	}
 	p_ramrod->outer_tag = p_hwfn->hw_info.ovlan;
 
 	/* Place EQ address in RAMROD */
@@ -142,9 +154,8 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
 	p_hwfn->hw_info.personality = PERSONALITY_ETH;
 
 	DP_VERBOSE(p_hwfn, QED_MSG_SPQ,
-		   "Setting event_ring_sb [id %04x index %02x], mf [%s] outer_tag [%d]\n",
+		   "Setting event_ring_sb [id %04x index %02x], outer_tag [%d]\n",
 		   sb, sb_index,
-		   (p_ramrod->mf_mode == SF) ? "SF" : "Multi-Pf",
 		   p_ramrod->outer_tag);
 
 	return qed_spq_post(p_hwfn, p_ent, NULL);
diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index 7c6caf7..f75d9e0 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -173,9 +173,9 @@ enum QEDE_STATE {
  * skb are built only after the frame was DMA-ed.
  */
 struct sw_rx_data {
-	u8 *data;
-
-	DEFINE_DMA_UNMAP_ADDR(mapping);
+	struct page *data;
+	dma_addr_t mapping;
+	unsigned int page_offset;
 };
 
 struct qede_rx_queue {
@@ -188,6 +188,7 @@ struct qede_rx_queue {
 	void __iomem		*hw_rxq_prod_addr;
 
 	int			rx_buf_size;
+	unsigned int		rx_buf_seg_size;
 
 	u16			num_rx_buffers;
 	u16			rxq_id;
@@ -281,6 +282,7 @@ void qede_fill_by_demand_stats(struct qede_dev *edev);
 #define NUM_TX_BDS_MIN		128
 #define NUM_TX_BDS_DEF		NUM_TX_BDS_MAX
 
+#define QEDE_RX_HDR_SIZE	256
 #define	for_each_rss(i) for (i = 0; i < edev->num_rss; i++)
 
 #endif /* _QEDE_H_ */
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
index e442b85..c49dc10 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
@@ -217,9 +217,9 @@ static int qede_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	struct qed_link_params params;
 	u32 speed;
 
-	if (edev->dev_info.common.is_mf) {
+	if (!edev->dev_info.common.is_mf_default) {
 		DP_INFO(edev,
-			"Link parameters can not be changed in MF mode\n");
+			"Link parameters can not be changed in non-default mode\n");
 		return -EOPNOTSUPP;
 	}
 
@@ -428,7 +428,7 @@ static int qede_set_pauseparam(struct net_device *dev,
 	struct qed_link_params params;
 	struct qed_link_output current_link;
 
-	if (!edev->dev_info.common.is_mf) {
+	if (!edev->dev_info.common.is_mf_default) {
 		DP_INFO(edev,
 			"Pause parameters can not be updated in non-default mode\n");
 		return -EOPNOTSUPP;
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 6237f10..f50e0bd 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -330,15 +330,15 @@ static void qede_set_params_for_ipv6_ext(struct sk_buff *skb,
 					 struct eth_tx_3rd_bd *third_bd)
 {
 	u8 l4_proto;
-	u16 bd2_bits = 0, bd2_bits2 = 0;
+	u16 bd2_bits1 = 0, bd2_bits2 = 0;
 
-	bd2_bits2 |= (1 << ETH_TX_DATA_2ND_BD_IPV6_EXT_SHIFT);
+	bd2_bits1 |= (1 << ETH_TX_DATA_2ND_BD_IPV6_EXT_SHIFT);
 
-	bd2_bits |= ((((u8 *)skb_transport_header(skb) - skb->data) >> 1) &
+	bd2_bits2 |= ((((u8 *)skb_transport_header(skb) - skb->data) >> 1) &
 		     ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_MASK)
 		    << ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_SHIFT;
 
-	bd2_bits2 |= (ETH_L4_PSEUDO_CSUM_CORRECT_LENGTH <<
+	bd2_bits1 |= (ETH_L4_PSEUDO_CSUM_CORRECT_LENGTH <<
 		      ETH_TX_DATA_2ND_BD_L4_PSEUDO_CSUM_MODE_SHIFT);
 
 	if (vlan_get_protocol(skb) == htons(ETH_P_IPV6))
@@ -347,16 +347,15 @@ static void qede_set_params_for_ipv6_ext(struct sk_buff *skb,
 		l4_proto = ip_hdr(skb)->protocol;
 
 	if (l4_proto == IPPROTO_UDP)
-		bd2_bits2 |= 1 << ETH_TX_DATA_2ND_BD_L4_UDP_SHIFT;
+		bd2_bits1 |= 1 << ETH_TX_DATA_2ND_BD_L4_UDP_SHIFT;
 
-	if (third_bd) {
+	if (third_bd)
 		third_bd->data.bitfields |=
-			((tcp_hdrlen(skb) / 4) &
-			 ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_MASK) <<
-			ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_SHIFT;
-	}
+			cpu_to_le16(((tcp_hdrlen(skb) / 4) &
+				ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_MASK) <<
+				ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_SHIFT);
 
-	second_bd->data.bitfields = cpu_to_le16(bd2_bits);
+	second_bd->data.bitfields1 = cpu_to_le16(bd2_bits1);
 	second_bd->data.bitfields2 = cpu_to_le16(bd2_bits2);
 }
 
@@ -464,12 +463,16 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb,
 
 	/* Fill the parsing flags & params according to the requested offload */
 	if (xmit_type & XMIT_L4_CSUM) {
+		u16 temp = 1 << ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_SHIFT;
+
 		/* We don't re-calculate IP checksum as it is already done by
 		 * the upper stack
 		 */
 		first_bd->data.bd_flags.bitfields |=
 			1 << ETH_TX_1ST_BD_FLAGS_L4_CSUM_SHIFT;
 
+		first_bd->data.bitfields |= cpu_to_le16(temp);
+
 		/* If the packet is IPv6 with extension header, indicate that
 		 * to FW and pass few params, since the device cracker doesn't
 		 * support parsing IPv6 with extension header/s.
@@ -491,7 +494,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb,
 
 		/* @@@TBD - if will not be removed need to check */
 		third_bd->data.bitfields |=
-			(1 << ETH_TX_DATA_3RD_BD_HDR_NBD_SHIFT);
+			cpu_to_le16((1 << ETH_TX_DATA_3RD_BD_HDR_NBD_SHIFT));
 
 		/* Make life easier for FW guys who can't deal with header and
 		 * data on same BD. If we need to split, use the second bd...
@@ -719,26 +722,52 @@ static bool qede_has_tx_work(struct qede_fastpath *fp)
 	return false;
 }
 
-/* This function copies the Rx buffer from the CONS position to the PROD
- * position, since we failed to allocate a new Rx buffer.
+/* This function reuses the buffer(from an offset) from
+ * consumer index to producer index in the bd ring
  */
-static void qede_reuse_rx_data(struct qede_rx_queue *rxq)
+static inline void qede_reuse_page(struct qede_dev *edev,
+				   struct qede_rx_queue *rxq,
+				   struct sw_rx_data *curr_cons)
 {
-	struct eth_rx_bd *rx_bd_cons = qed_chain_consume(&rxq->rx_bd_ring);
 	struct eth_rx_bd *rx_bd_prod = qed_chain_produce(&rxq->rx_bd_ring);
-	struct sw_rx_data *sw_rx_data_cons =
-		&rxq->sw_rx_ring[rxq->sw_rx_cons & NUM_RX_BDS_MAX];
-	struct sw_rx_data *sw_rx_data_prod =
-		&rxq->sw_rx_ring[rxq->sw_rx_prod & NUM_RX_BDS_MAX];
+	struct sw_rx_data *curr_prod;
+	dma_addr_t new_mapping;
 
-	dma_unmap_addr_set(sw_rx_data_prod, mapping,
-			   dma_unmap_addr(sw_rx_data_cons, mapping));
+	curr_prod = &rxq->sw_rx_ring[rxq->sw_rx_prod & NUM_RX_BDS_MAX];
+	*curr_prod = *curr_cons;
 
-	sw_rx_data_prod->data = sw_rx_data_cons->data;
-	memcpy(rx_bd_prod, rx_bd_cons, sizeof(struct eth_rx_bd));
+	new_mapping = curr_prod->mapping + curr_prod->page_offset;
+
+	rx_bd_prod->addr.hi = cpu_to_le32(upper_32_bits(new_mapping));
+	rx_bd_prod->addr.lo = cpu_to_le32(lower_32_bits(new_mapping));
 
-	rxq->sw_rx_cons++;
 	rxq->sw_rx_prod++;
+	curr_cons->data = NULL;
+}
+
+static inline int qede_realloc_rx_buffer(struct qede_dev *edev,
+					 struct qede_rx_queue *rxq,
+					 struct sw_rx_data *curr_cons)
+{
+	/* Move to the next segment in the page */
+	curr_cons->page_offset += rxq->rx_buf_seg_size;
+
+	if (curr_cons->page_offset == PAGE_SIZE) {
+		if (unlikely(qede_alloc_rx_buffer(edev, rxq)))
+			return -ENOMEM;
+
+		dma_unmap_page(&edev->pdev->dev, curr_cons->mapping,
+			       PAGE_SIZE, DMA_FROM_DEVICE);
+	} else {
+		/* Increment refcount of the page as we don't want
+		 * network stack to take the ownership of the page
+		 * which can be recycled multiple times by the driver.
+		 */
+		atomic_inc(&curr_cons->data->_count);
+		qede_reuse_page(edev, rxq, curr_cons);
+	}
+
+	return 0;
 }
 
 static inline void qede_update_rx_prod(struct qede_dev *edev,
@@ -857,9 +886,10 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget)
 		struct sw_rx_data *sw_rx_data;
 		union eth_rx_cqe *cqe;
 		struct sk_buff *skb;
+		struct page *data;
+		__le16 flags;
 		u16 len, pad;
 		u32 rx_hash;
-		u8 *data;
 
 		/* Get the CQE from the completion ring */
 		cqe = (union eth_rx_cqe *)
@@ -879,56 +909,110 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget)
 		data = sw_rx_data->data;
 
 		fp_cqe = &cqe->fast_path_regular;
-		len =  le16_to_cpu(fp_cqe->pkt_len);
+		len =  le16_to_cpu(fp_cqe->len_on_first_bd);
 		pad = fp_cqe->placement_offset;
+		flags = cqe->fast_path_regular.pars_flags.flags;
 
-		/* For every Rx BD consumed, we allocate a new BD so the BD ring
-		 * is always with a fixed size. If allocation fails, we take the
-		 * consumed BD and return it to the ring in the PROD position.
-		 * The packet that was received on that BD will be dropped (and
-		 * not passed to the upper stack).
-		 */
-		if (likely(qede_alloc_rx_buffer(edev, rxq) == 0)) {
-			dma_unmap_single(&edev->pdev->dev,
-					 dma_unmap_addr(sw_rx_data, mapping),
-					 rxq->rx_buf_size, DMA_FROM_DEVICE);
-
-			/* If this is an error packet then drop it */
-			parse_flag =
-			le16_to_cpu(cqe->fast_path_regular.pars_flags.flags);
-			csum_flag = qede_check_csum(parse_flag);
-			if (csum_flag == QEDE_CSUM_ERROR) {
-				DP_NOTICE(edev,
-					  "CQE in CONS = %u has error, flags = %x, dropping incoming packet\n",
-					  sw_comp_cons, parse_flag);
-				rxq->rx_hw_errors++;
-				kfree(data);
-				goto next_rx;
-			}
-
-			skb = build_skb(data, 0);
-
-			if (unlikely(!skb)) {
-				DP_NOTICE(edev,
-					  "Build_skb failed, dropping incoming packet\n");
-				kfree(data);
-				rxq->rx_alloc_errors++;
-				goto next_rx;
-			}
+		/* If this is an error packet then drop it */
+		parse_flag = le16_to_cpu(flags);
 
-			skb_reserve(skb, pad);
+		csum_flag = qede_check_csum(parse_flag);
+		if (unlikely(csum_flag == QEDE_CSUM_ERROR)) {
+			DP_NOTICE(edev,
+				  "CQE in CONS = %u has error, flags = %x, dropping incoming packet\n",
+				  sw_comp_cons, parse_flag);
+			rxq->rx_hw_errors++;
+			qede_reuse_page(edev, rxq, sw_rx_data);
+			goto next_rx;
+		}
 
-		} else {
+		skb = netdev_alloc_skb(edev->ndev, QEDE_RX_HDR_SIZE);
+		if (unlikely(!skb)) {
 			DP_NOTICE(edev,
-				  "New buffer allocation failed, dropping incoming packet and reusing its buffer\n");
-			qede_reuse_rx_data(rxq);
+				  "Build_skb failed, dropping incoming packet\n");
+			qede_reuse_page(edev, rxq, sw_rx_data);
 			rxq->rx_alloc_errors++;
-			goto next_cqe;
+			goto next_rx;
 		}
 
-		sw_rx_data->data = NULL;
+		/* Copy data into SKB */
+		if (len + pad <= QEDE_RX_HDR_SIZE) {
+			memcpy(skb_put(skb, len),
+			       page_address(data) + pad +
+				sw_rx_data->page_offset, len);
+			qede_reuse_page(edev, rxq, sw_rx_data);
+		} else {
+			struct skb_frag_struct *frag;
+			unsigned int pull_len;
+			unsigned char *va;
 
-		skb_put(skb, len);
+			frag = &skb_shinfo(skb)->frags[0];
+
+			skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, data,
+					pad + sw_rx_data->page_offset,
+					len, rxq->rx_buf_seg_size);
+
+			va = skb_frag_address(frag);
+			pull_len = eth_get_headlen(va, QEDE_RX_HDR_SIZE);
+
+			/* Align the pull_len to optimize memcpy */
+			memcpy(skb->data, va, ALIGN(pull_len, sizeof(long)));
+
+			skb_frag_size_sub(frag, pull_len);
+			frag->page_offset += pull_len;
+			skb->data_len -= pull_len;
+			skb->tail += pull_len;
+
+			if (unlikely(qede_realloc_rx_buffer(edev, rxq,
+							    sw_rx_data))) {
+				DP_ERR(edev, "Failed to allocate rx buffer\n");
+				rxq->rx_alloc_errors++;
+				goto next_cqe;
+			}
+		}
+
+		if (fp_cqe->bd_num != 1) {
+			u16 pkt_len = le16_to_cpu(fp_cqe->pkt_len);
+			u8 num_frags;
+
+			pkt_len -= len;
+
+			for (num_frags = fp_cqe->bd_num - 1; num_frags > 0;
+			     num_frags--) {
+				u16 cur_size = pkt_len > rxq->rx_buf_size ?
+						rxq->rx_buf_size : pkt_len;
+
+				WARN_ONCE(!cur_size,
+					  "Still got %d BDs for mapping jumbo, but length became 0\n",
+					  num_frags);
+
+				if (unlikely(qede_alloc_rx_buffer(edev, rxq)))
+					goto next_cqe;
+
+				rxq->sw_rx_cons++;
+				sw_rx_index = rxq->sw_rx_cons & NUM_RX_BDS_MAX;
+				sw_rx_data = &rxq->sw_rx_ring[sw_rx_index];
+				qed_chain_consume(&rxq->rx_bd_ring);
+				dma_unmap_page(&edev->pdev->dev,
+					       sw_rx_data->mapping,
+					       PAGE_SIZE, DMA_FROM_DEVICE);
+
+				skb_fill_page_desc(skb,
+						   skb_shinfo(skb)->nr_frags++,
+						   sw_rx_data->data, 0,
+						   cur_size);
+
+				skb->truesize += PAGE_SIZE;
+				skb->data_len += cur_size;
+				skb->len += cur_size;
+				pkt_len -= cur_size;
+			}
+
+			if (pkt_len)
+				DP_ERR(edev,
+				       "Mapped all BDs of jumbo, but still have %d bytes\n",
+				       pkt_len);
+		}
 
 		skb->protocol = eth_type_trans(skb, edev->ndev);
 
@@ -1566,17 +1650,17 @@ static void qede_free_rx_buffers(struct qede_dev *edev,
 
 	for (i = rxq->sw_rx_cons; i != rxq->sw_rx_prod; i++) {
 		struct sw_rx_data *rx_buf;
-		u8 *data;
+		struct page *data;
 
 		rx_buf = &rxq->sw_rx_ring[i & NUM_RX_BDS_MAX];
 		data = rx_buf->data;
 
-		dma_unmap_single(&edev->pdev->dev,
-				 dma_unmap_addr(rx_buf, mapping),
-				 rxq->rx_buf_size, DMA_FROM_DEVICE);
+		dma_unmap_page(&edev->pdev->dev,
+			       rx_buf->mapping,
+			       PAGE_SIZE, DMA_FROM_DEVICE);
 
 		rx_buf->data = NULL;
-		kfree(data);
+		__free_page(data);
 	}
 }
 
@@ -1600,29 +1684,32 @@ static int qede_alloc_rx_buffer(struct qede_dev *edev,
 	struct sw_rx_data *sw_rx_data;
 	struct eth_rx_bd *rx_bd;
 	dma_addr_t mapping;
+	struct page *data;
 	u16 rx_buf_size;
-	u8 *data;
 
 	rx_buf_size = rxq->rx_buf_size;
 
-	data = kmalloc(rx_buf_size, GFP_ATOMIC);
+	data = alloc_pages(GFP_ATOMIC, 0);
 	if (unlikely(!data)) {
-		DP_NOTICE(edev, "Failed to allocate Rx data\n");
+		DP_NOTICE(edev, "Failed to allocate Rx data [page]\n");
 		return -ENOMEM;
 	}
 
-	mapping = dma_map_single(&edev->pdev->dev, data,
-				 rx_buf_size, DMA_FROM_DEVICE);
+	/* Map the entire page as it would be used
+	 * for multiple RX buffer segment size mapping.
+	 */
+	mapping = dma_map_page(&edev->pdev->dev, data, 0,
+			       PAGE_SIZE, DMA_FROM_DEVICE);
 	if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) {
-		kfree(data);
+		__free_page(data);
 		DP_NOTICE(edev, "Failed to map Rx buffer\n");
 		return -ENOMEM;
 	}
 
 	sw_rx_data = &rxq->sw_rx_ring[rxq->sw_rx_prod & NUM_RX_BDS_MAX];
+	sw_rx_data->page_offset = 0;
 	sw_rx_data->data = data;
-
-	dma_unmap_addr_set(sw_rx_data, mapping, mapping);
+	sw_rx_data->mapping = mapping;
 
 	/* Advance PROD and get BD pointer */
 	rx_bd = (struct eth_rx_bd *)qed_chain_produce(&rxq->rx_bd_ring);
@@ -1643,13 +1730,16 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev,
 
 	rxq->num_rx_buffers = edev->q_num_rx_buffers;
 
-	rxq->rx_buf_size = NET_IP_ALIGN +
-			   ETH_OVERHEAD +
-			   edev->ndev->mtu +
-			   QEDE_FW_RX_ALIGN_END;
+	rxq->rx_buf_size = NET_IP_ALIGN + ETH_OVERHEAD +
+			   edev->ndev->mtu;
+	if (rxq->rx_buf_size > PAGE_SIZE)
+		rxq->rx_buf_size = PAGE_SIZE;
+
+	/* Segment size to spilt a page in multiple equal parts */
+	rxq->rx_buf_seg_size = roundup_pow_of_two(rxq->rx_buf_size);
 
 	/* Allocate the parallel driver ring for Rx buffers */
-	size = sizeof(*rxq->sw_rx_ring) * NUM_RX_BDS_MAX;
+	size = sizeof(*rxq->sw_rx_ring) * RX_RING_SIZE;
 	rxq->sw_rx_ring = kzalloc(size, GFP_KERNEL);
 	if (!rxq->sw_rx_ring) {
 		DP_ERR(edev, "Rx buffers ring allocation failed\n");
@@ -1660,7 +1750,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev,
 	rc = edev->ops->common->chain_alloc(edev->cdev,
 					    QED_CHAIN_USE_TO_CONSUME_PRODUCE,
 					    QED_CHAIN_MODE_NEXT_PTR,
-					    NUM_RX_BDS_MAX,
+					    RX_RING_SIZE,
 					    sizeof(struct eth_rx_bd),
 					    &rxq->rx_bd_ring);
 
@@ -1671,7 +1761,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev,
 	rc = edev->ops->common->chain_alloc(edev->cdev,
 					    QED_CHAIN_USE_TO_CONSUME,
 					    QED_CHAIN_MODE_PBL,
-					    NUM_RX_BDS_MAX,
+					    RX_RING_SIZE,
 					    sizeof(union eth_rx_cqe),
 					    &rxq->rx_comp_ring);
 	if (rc)
diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h
index 1d1ba2c..53ecb37 100644
--- a/include/linux/qed/common_hsi.h
+++ b/include/linux/qed/common_hsi.h
@@ -11,9 +11,11 @@
 
 #define CORE_SPQE_PAGE_SIZE_BYTES                       4096
 
+#define X_FINAL_CLEANUP_AGG_INT 1
+
 #define FW_MAJOR_VERSION	8
-#define FW_MINOR_VERSION	4
-#define FW_REVISION_VERSION	2
+#define FW_MINOR_VERSION	7
+#define FW_REVISION_VERSION	3
 #define FW_ENGINEERING_VERSION	0
 
 /***********************/
@@ -152,6 +154,9 @@
 /* number of queues in a PF queue group */
 #define QM_PF_QUEUE_GROUP_SIZE	8
 
+/* the size of a single queue element in bytes */
+#define QM_PQ_ELEMENT_SIZE                      4
+
 /* base number of Tx PQs in the CM PQ representation.
  * should be used when storing PQ IDs in CM PQ registers and context
  */
@@ -285,6 +290,16 @@
 #define PXP_NUM_ILT_RECORDS_K2 11000
 #define MAX_NUM_ILT_RECORDS MAX(PXP_NUM_ILT_RECORDS_BB, PXP_NUM_ILT_RECORDS_K2)
 
+#define SDM_COMP_TYPE_NONE              0
+#define SDM_COMP_TYPE_WAKE_THREAD       1
+#define SDM_COMP_TYPE_AGG_INT           2
+#define SDM_COMP_TYPE_CM                3
+#define SDM_COMP_TYPE_LOADER            4
+#define SDM_COMP_TYPE_PXP               5
+#define SDM_COMP_TYPE_INDICATE_ERROR    6
+#define SDM_COMP_TYPE_RELEASE_THREAD    7
+#define SDM_COMP_TYPE_RAM               8
+
 /******************/
 /* PBF CONSTANTS  */
 /******************/
@@ -335,7 +350,7 @@ struct event_ring_entry {
 
 /* Multi function mode */
 enum mf_mode {
-	SF,
+	ERROR_MODE /* Unsupported mode */,
 	MF_OVLAN,
 	MF_NPAR,
 	MAX_MF_MODE
@@ -606,4 +621,19 @@ struct status_block {
 #define STATUS_BLOCK_ZERO_PAD3_SHIFT  24
 };
 
+struct tunnel_parsing_flags {
+	u8 flags;
+#define TUNNEL_PARSING_FLAGS_TYPE_MASK              0x3
+#define TUNNEL_PARSING_FLAGS_TYPE_SHIFT             0
+#define TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_MASK  0x1
+#define TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_SHIFT 2
+#define TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_MASK     0x3
+#define TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_SHIFT    3
+#define TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_MASK   0x1
+#define TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_SHIFT  5
+#define TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK     0x1
+#define TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT    6
+#define TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_MASK      0x1
+#define TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_SHIFT     7
+};
 #endif /* __COMMON_HSI__ */
diff --git a/include/linux/qed/eth_common.h b/include/linux/qed/eth_common.h
index 320b337..092cb0c 100644
--- a/include/linux/qed/eth_common.h
+++ b/include/linux/qed/eth_common.h
@@ -17,10 +17,8 @@
 #define ETH_MAX_RAMROD_PER_CON                          8
 #define ETH_TX_BD_PAGE_SIZE_BYTES                       4096
 #define ETH_RX_BD_PAGE_SIZE_BYTES                       4096
-#define ETH_RX_SGE_PAGE_SIZE_BYTES                      4096
 #define ETH_RX_CQE_PAGE_SIZE_BYTES                      4096
 #define ETH_RX_NUM_NEXT_PAGE_BDS                        2
-#define ETH_RX_NUM_NEXT_PAGE_SGES                       2
 
 #define ETH_TX_MIN_BDS_PER_NON_LSO_PKT                          1
 #define ETH_TX_MAX_BDS_PER_NON_LSO_PACKET                       18
@@ -34,7 +32,8 @@
 
 #define ETH_NUM_STATISTIC_COUNTERS                      MAX_NUM_VPORTS
 
-#define ETH_REG_CQE_PBL_SIZE                3
+/* Maximum number of buffers, used for RX packet placement */
+#define ETH_RX_MAX_BUFF_PER_PKT             5
 
 /* num of MAC/VLAN filters */
 #define ETH_NUM_MAC_FILTERS                                     512
@@ -54,9 +53,9 @@
 
 /* TPA constants */
 #define ETH_TPA_MAX_AGGS_NUM              64
-#define ETH_TPA_CQE_START_SGL_SIZE        3
-#define ETH_TPA_CQE_CONT_SGL_SIZE         6
-#define ETH_TPA_CQE_END_SGL_SIZE          4
+#define ETH_TPA_CQE_START_LEN_LIST_SIZE   ETH_RX_MAX_BUFF_PER_PKT
+#define ETH_TPA_CQE_CONT_LEN_LIST_SIZE    6
+#define ETH_TPA_CQE_END_LEN_LIST_SIZE     4
 
 /* Queue Zone sizes */
 #define TSTORM_QZONE_SIZE    0
@@ -74,18 +73,18 @@ struct coalescing_timeset {
 
 struct eth_tx_1st_bd_flags {
 	u8 bitfields;
+#define ETH_TX_1ST_BD_FLAGS_START_BD_MASK         0x1
+#define ETH_TX_1ST_BD_FLAGS_START_BD_SHIFT        0
 #define ETH_TX_1ST_BD_FLAGS_FORCE_VLAN_MODE_MASK  0x1
-#define ETH_TX_1ST_BD_FLAGS_FORCE_VLAN_MODE_SHIFT 0
+#define ETH_TX_1ST_BD_FLAGS_FORCE_VLAN_MODE_SHIFT 1
 #define ETH_TX_1ST_BD_FLAGS_IP_CSUM_MASK          0x1
-#define ETH_TX_1ST_BD_FLAGS_IP_CSUM_SHIFT         1
+#define ETH_TX_1ST_BD_FLAGS_IP_CSUM_SHIFT         2
 #define ETH_TX_1ST_BD_FLAGS_L4_CSUM_MASK          0x1
-#define ETH_TX_1ST_BD_FLAGS_L4_CSUM_SHIFT         2
+#define ETH_TX_1ST_BD_FLAGS_L4_CSUM_SHIFT         3
 #define ETH_TX_1ST_BD_FLAGS_VLAN_INSERTION_MASK   0x1
-#define ETH_TX_1ST_BD_FLAGS_VLAN_INSERTION_SHIFT  3
+#define ETH_TX_1ST_BD_FLAGS_VLAN_INSERTION_SHIFT  4
 #define ETH_TX_1ST_BD_FLAGS_LSO_MASK              0x1
-#define ETH_TX_1ST_BD_FLAGS_LSO_SHIFT             4
-#define ETH_TX_1ST_BD_FLAGS_START_BD_MASK         0x1
-#define ETH_TX_1ST_BD_FLAGS_START_BD_SHIFT        5
+#define ETH_TX_1ST_BD_FLAGS_LSO_SHIFT             5
 #define ETH_TX_1ST_BD_FLAGS_TUNN_IP_CSUM_MASK     0x1
 #define ETH_TX_1ST_BD_FLAGS_TUNN_IP_CSUM_SHIFT    6
 #define ETH_TX_1ST_BD_FLAGS_TUNN_L4_CSUM_MASK     0x1
@@ -97,38 +96,44 @@ struct eth_tx_data_1st_bd {
 	__le16				vlan;
 	u8				nbds;
 	struct eth_tx_1st_bd_flags	bd_flags;
-	__le16				fw_use_only;
+	__le16				bitfields;
+#define ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_MASK  0x1
+#define ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_SHIFT 0
+#define ETH_TX_DATA_1ST_BD_RESERVED0_MASK          0x1
+#define ETH_TX_DATA_1ST_BD_RESERVED0_SHIFT         1
+#define ETH_TX_DATA_1ST_BD_FW_USE_ONLY_MASK        0x3FFF
+#define ETH_TX_DATA_1ST_BD_FW_USE_ONLY_SHIFT       2
 };
 
 /* The parsing information data for the second tx bd of a given packet. */
 struct eth_tx_data_2nd_bd {
 	__le16	tunn_ip_size;
-	__le16	bitfields;
-#define ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_MASK     0x1FFF
-#define ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_SHIFT    0
-#define ETH_TX_DATA_2ND_BD_RESERVED0_MASK                 0x7
-#define ETH_TX_DATA_2ND_BD_RESERVED0_SHIFT                13
-	__le16	bitfields2;
+	__le16	bitfields1;
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_L2_HDR_SIZE_W_MASK  0xF
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_L2_HDR_SIZE_W_SHIFT 0
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_ETH_TYPE_MASK       0x3
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_ETH_TYPE_SHIFT      4
 #define ETH_TX_DATA_2ND_BD_DEST_PORT_MODE_MASK            0x3
 #define ETH_TX_DATA_2ND_BD_DEST_PORT_MODE_SHIFT           6
+#define ETH_TX_DATA_2ND_BD_START_BD_MASK                  0x1
+#define ETH_TX_DATA_2ND_BD_START_BD_SHIFT                 8
 #define ETH_TX_DATA_2ND_BD_TUNN_TYPE_MASK                 0x3
-#define ETH_TX_DATA_2ND_BD_TUNN_TYPE_SHIFT                8
+#define ETH_TX_DATA_2ND_BD_TUNN_TYPE_SHIFT                9
 #define ETH_TX_DATA_2ND_BD_TUNN_INNER_IPV6_MASK           0x1
-#define ETH_TX_DATA_2ND_BD_TUNN_INNER_IPV6_SHIFT          10
+#define ETH_TX_DATA_2ND_BD_TUNN_INNER_IPV6_SHIFT          11
 #define ETH_TX_DATA_2ND_BD_IPV6_EXT_MASK                  0x1
-#define ETH_TX_DATA_2ND_BD_IPV6_EXT_SHIFT                 11
+#define ETH_TX_DATA_2ND_BD_IPV6_EXT_SHIFT                 12
 #define ETH_TX_DATA_2ND_BD_TUNN_IPV6_EXT_MASK             0x1
-#define ETH_TX_DATA_2ND_BD_TUNN_IPV6_EXT_SHIFT            12
+#define ETH_TX_DATA_2ND_BD_TUNN_IPV6_EXT_SHIFT            13
 #define ETH_TX_DATA_2ND_BD_L4_UDP_MASK                    0x1
-#define ETH_TX_DATA_2ND_BD_L4_UDP_SHIFT                   13
+#define ETH_TX_DATA_2ND_BD_L4_UDP_SHIFT                   14
 #define ETH_TX_DATA_2ND_BD_L4_PSEUDO_CSUM_MODE_MASK       0x1
-#define ETH_TX_DATA_2ND_BD_L4_PSEUDO_CSUM_MODE_SHIFT      14
-#define ETH_TX_DATA_2ND_BD_RESERVED1_MASK                 0x1
-#define ETH_TX_DATA_2ND_BD_RESERVED1_SHIFT                15
+#define ETH_TX_DATA_2ND_BD_L4_PSEUDO_CSUM_MODE_SHIFT      15
+	__le16 bitfields2;
+#define ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_MASK     0x1FFF
+#define ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_SHIFT    0
+#define ETH_TX_DATA_2ND_BD_RESERVED0_MASK                 0x7
+#define ETH_TX_DATA_2ND_BD_RESERVED0_SHIFT                13
 };
 
 /* Regular ETH Rx FP CQE. */
@@ -145,11 +150,68 @@ struct eth_fast_path_rx_reg_cqe {
 	struct parsing_and_err_flags	pars_flags;
 	__le16				vlan_tag;
 	__le32				rss_hash;
-	__le16				len_on_bd;
+	__le16				len_on_first_bd;
 	u8				placement_offset;
-	u8				reserved;
-	__le16				pbl[ETH_REG_CQE_PBL_SIZE];
-	u8				reserved1[10];
+	struct tunnel_parsing_flags	tunnel_pars_flags;
+	u8				bd_num;
+	u8				reserved[7];
+	u32				fw_debug;
+	u8				reserved1[3];
+	u8				flags;
+#define ETH_FAST_PATH_RX_REG_CQE_VALID_MASK          0x1
+#define ETH_FAST_PATH_RX_REG_CQE_VALID_SHIFT         0
+#define ETH_FAST_PATH_RX_REG_CQE_VALID_TOGGLE_MASK   0x1
+#define ETH_FAST_PATH_RX_REG_CQE_VALID_TOGGLE_SHIFT  1
+#define ETH_FAST_PATH_RX_REG_CQE_RESERVED2_MASK      0x3F
+#define ETH_FAST_PATH_RX_REG_CQE_RESERVED2_SHIFT     2
+};
+
+/* TPA-continue ETH Rx FP CQE. */
+struct eth_fast_path_rx_tpa_cont_cqe {
+	u8	type;
+	u8	tpa_agg_index;
+	__le16	len_list[ETH_TPA_CQE_CONT_LEN_LIST_SIZE];
+	u8	reserved[5];
+	u8	reserved1;
+	__le16	reserved2[ETH_TPA_CQE_CONT_LEN_LIST_SIZE];
+};
+
+/* TPA-end ETH Rx FP CQE. */
+struct eth_fast_path_rx_tpa_end_cqe {
+	u8	type;
+	u8	tpa_agg_index;
+	__le16	total_packet_len;
+	u8	num_of_bds;
+	u8	end_reason;
+	__le16	num_of_coalesced_segs;
+	__le32	ts_delta;
+	__le16	len_list[ETH_TPA_CQE_END_LEN_LIST_SIZE];
+	u8	reserved1[3];
+	u8	reserved2;
+	__le16	reserved3[ETH_TPA_CQE_END_LEN_LIST_SIZE];
+};
+
+/* TPA-start ETH Rx FP CQE. */
+struct eth_fast_path_rx_tpa_start_cqe {
+	u8	type;
+	u8	bitfields;
+#define ETH_FAST_PATH_RX_TPA_START_CQE_RSS_HASH_TYPE_MASK  0x7
+#define ETH_FAST_PATH_RX_TPA_START_CQE_RSS_HASH_TYPE_SHIFT 0
+#define ETH_FAST_PATH_RX_TPA_START_CQE_TC_MASK             0xF
+#define ETH_FAST_PATH_RX_TPA_START_CQE_TC_SHIFT            3
+#define ETH_FAST_PATH_RX_TPA_START_CQE_RESERVED0_MASK      0x1
+#define ETH_FAST_PATH_RX_TPA_START_CQE_RESERVED0_SHIFT     7
+	__le16	seg_len;
+	struct parsing_and_err_flags pars_flags;
+	__le16	vlan_tag;
+	__le32	rss_hash;
+	__le16	len_on_first_bd;
+	u8	placement_offset;
+	struct tunnel_parsing_flags tunnel_pars_flags;
+	u8	tpa_agg_index;
+	u8	header_len;
+	__le16	ext_bd_len_list[ETH_TPA_CQE_START_LEN_LIST_SIZE];
+	u32	fw_debug;
 };
 
 /* The L4 pseudo checksum mode for Ethernet */
@@ -168,13 +230,26 @@ struct eth_slow_path_rx_cqe {
 	u8	type;
 	u8	ramrod_cmd_id;
 	u8	error_flag;
-	u8	reserved[27];
+	u8	reserved[25];
 	__le16	echo;
+	u8	reserved1;
+	u8	flags;
+/* for PMD mode - valid indication */
+#define ETH_SLOW_PATH_RX_CQE_VALID_MASK         0x1
+#define ETH_SLOW_PATH_RX_CQE_VALID_SHIFT        0
+/* for PMD mode - valid toggle indication */
+#define ETH_SLOW_PATH_RX_CQE_VALID_TOGGLE_MASK  0x1
+#define ETH_SLOW_PATH_RX_CQE_VALID_TOGGLE_SHIFT 1
+#define ETH_SLOW_PATH_RX_CQE_RESERVED2_MASK     0x3F
+#define ETH_SLOW_PATH_RX_CQE_RESERVED2_SHIFT    2
 };
 
 /* union for all ETH Rx CQE types */
 union eth_rx_cqe {
 	struct eth_fast_path_rx_reg_cqe		fast_path_regular;
+	struct eth_fast_path_rx_tpa_start_cqe	fast_path_tpa_start;
+	struct eth_fast_path_rx_tpa_cont_cqe	fast_path_tpa_cont;
+	struct eth_fast_path_rx_tpa_end_cqe	fast_path_tpa_end;
 	struct eth_slow_path_rx_cqe		slow_path;
 };
 
@@ -183,15 +258,18 @@ enum eth_rx_cqe_type {
 	ETH_RX_CQE_TYPE_UNUSED,
 	ETH_RX_CQE_TYPE_REGULAR,
 	ETH_RX_CQE_TYPE_SLOW_PATH,
+	ETH_RX_CQE_TYPE_TPA_START,
+	ETH_RX_CQE_TYPE_TPA_CONT,
+	ETH_RX_CQE_TYPE_TPA_END,
 	MAX_ETH_RX_CQE_TYPE
 };
 
 /* ETH Rx producers data */
 struct eth_rx_prod_data {
 	__le16	bd_prod;
-	__le16	sge_prod;
 	__le16	cqe_prod;
 	__le16	reserved;
+	__le16	reserved1;
 };
 
 /* The first tx bd of a given packet */
@@ -211,12 +289,17 @@ struct eth_tx_2nd_bd {
 /* The parsing information data for the third tx bd of a given packet. */
 struct eth_tx_data_3rd_bd {
 	__le16	lso_mss;
-	u8	bitfields;
+	__le16	bitfields;
 #define ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_MASK  0xF
 #define ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_SHIFT 0
 #define ETH_TX_DATA_3RD_BD_HDR_NBD_MASK         0xF
 #define ETH_TX_DATA_3RD_BD_HDR_NBD_SHIFT        4
-	u8	resereved0[3];
+#define ETH_TX_DATA_3RD_BD_START_BD_MASK        0x1
+#define ETH_TX_DATA_3RD_BD_START_BD_SHIFT       8
+#define ETH_TX_DATA_3RD_BD_RESERVED0_MASK       0x7F
+#define ETH_TX_DATA_3RD_BD_RESERVED0_SHIFT      9
+	u8	tunn_l4_hdr_start_offset_w;
+	u8	tunn_hdr_size_w;
 };
 
 /* The third tx bd of a given packet */
@@ -226,12 +309,24 @@ struct eth_tx_3rd_bd {
 	struct eth_tx_data_3rd_bd	data;
 };
 
+/* Complementary information for the regular tx bd of a given packet. */
+struct eth_tx_data_bd {
+	__le16	reserved0;
+	__le16	bitfields;
+#define ETH_TX_DATA_BD_RESERVED1_MASK  0xFF
+#define ETH_TX_DATA_BD_RESERVED1_SHIFT 0
+#define ETH_TX_DATA_BD_START_BD_MASK   0x1
+#define ETH_TX_DATA_BD_START_BD_SHIFT  8
+#define ETH_TX_DATA_BD_RESERVED2_MASK  0x7F
+#define ETH_TX_DATA_BD_RESERVED2_SHIFT 9
+	__le16 reserved3;
+};
+
 /* The common non-special TX BD ring element */
 struct eth_tx_bd {
 	struct regpair	addr;
 	__le16		nbytes;
-	__le16		reserved0;
-	__le32		reserved1;
+	struct eth_tx_data_bd	data;
 };
 
 union eth_tx_bd_types {
diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h
index d4a32e8..3d43c1d 100644
--- a/include/linux/qed/qed_if.h
+++ b/include/linux/qed/qed_if.h
@@ -80,7 +80,7 @@ struct qed_dev_info {
 	u8		num_hwfns;
 
 	u8		hw_mac[ETH_ALEN];
-	bool		is_mf;
+	bool		is_mf_default;
 
 	/* FW version */
 	u16		fw_major;
@@ -360,6 +360,12 @@ enum DP_MODULE {
 	/* to be added...up to 0x8000000 */
 };
 
+enum qed_mf_mode {
+	QED_MF_DEFAULT,
+	QED_MF_OVLAN,
+	QED_MF_NPAR,
+};
+
 struct qed_eth_stats {
 	u64	no_buff_discards;
 	u64	packet_too_big_discard;
-- 
cgit v0.10.2


From 82aaf4fcbe0f26adb14e72c9df6a2c4ed3f8c34d Mon Sep 17 00:00:00 2001
From: One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk>
Date: Mon, 15 Feb 2016 19:41:51 +0000
Subject: af_llc: fix types on llc_ui_wait_for_conn

The timeout is a long, we return it truncated if it is huge. Basically
harmless as the only caller does a boolean check, but tidy it up anyway.

(64bit build tested this time. Thank you 0day)

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c
index 8dab4e5..b3c52e3 100644
--- a/net/llc/af_llc.c
+++ b/net/llc/af_llc.c
@@ -38,7 +38,7 @@ static u16 llc_ui_sap_link_no_max[256];
 static struct sockaddr_llc llc_ui_addrnull;
 static const struct proto_ops llc_ui_ops;
 
-static int llc_ui_wait_for_conn(struct sock *sk, long timeout);
+static long llc_ui_wait_for_conn(struct sock *sk, long timeout);
 static int llc_ui_wait_for_disc(struct sock *sk, long timeout);
 static int llc_ui_wait_for_busy_core(struct sock *sk, long timeout);
 
@@ -551,7 +551,7 @@ static int llc_ui_wait_for_disc(struct sock *sk, long timeout)
 	return rc;
 }
 
-static int llc_ui_wait_for_conn(struct sock *sk, long timeout)
+static long llc_ui_wait_for_conn(struct sock *sk, long timeout)
 {
 	DEFINE_WAIT(wait);
 
-- 
cgit v0.10.2


From a4f9cdb2b44d2dceede4eaad00c6b11206af0fea Mon Sep 17 00:00:00 2001
From: Ben Hutchings <ben@decadent.org.uk>
Date: Mon, 15 Feb 2016 21:25:11 +0000
Subject: pasemi_mac: Replace LRO with GRO

GRO is simpler to use than the old inet_lro library, and is compatible
with forwarding and bridging configurations.

Compile-tested only.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/pasemi/Kconfig b/drivers/net/ethernet/pasemi/Kconfig
index db19c6f..7c92e83 100644
--- a/drivers/net/ethernet/pasemi/Kconfig
+++ b/drivers/net/ethernet/pasemi/Kconfig
@@ -5,7 +5,7 @@
 config NET_VENDOR_PASEMI
 	bool "PA Semi devices"
 	default y
-	depends on PPC_PASEMI && PCI && INET
+	depends on PPC_PASEMI && PCI
 	---help---
 	  If you have a network (Ethernet) card belonging to this class, say Y.
 
@@ -18,9 +18,8 @@ if NET_VENDOR_PASEMI
 
 config PASEMI_MAC
 	tristate "PA Semi 1/10Gbit MAC"
-	depends on PPC_PASEMI && PCI && INET
+	depends on PPC_PASEMI && PCI
 	select PHYLIB
-	select INET_LRO
 	---help---
 	  This driver supports the on-chip 1/10Gbit Ethernet controller on
 	  PA Semi's PWRficient line of chips.
diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.c b/drivers/net/ethernet/pasemi/pasemi_mac.c
index 57a6e6c..af54df5 100644
--- a/drivers/net/ethernet/pasemi/pasemi_mac.c
+++ b/drivers/net/ethernet/pasemi/pasemi_mac.c
@@ -30,9 +30,7 @@
 #include <linux/skbuff.h>
 
 #include <linux/ip.h>
-#include <linux/tcp.h>
 #include <net/checksum.h>
-#include <linux/inet_lro.h>
 #include <linux/prefetch.h>
 
 #include <asm/irq.h>
@@ -52,12 +50,9 @@
  *
  * - Multicast support
  * - Large MTU support
- * - SW LRO
  * - Multiqueue RX/TX
  */
 
-#define LRO_MAX_AGGR 64
-
 #define PE_MIN_MTU	64
 #define PE_MAX_MTU	9000
 #define PE_DEF_MTU	ETH_DATA_LEN
@@ -257,37 +252,6 @@ static int pasemi_mac_set_mac_addr(struct net_device *dev, void *p)
 	return 0;
 }
 
-static int get_skb_hdr(struct sk_buff *skb, void **iphdr,
-		       void **tcph, u64 *hdr_flags, void *data)
-{
-	u64 macrx = (u64) data;
-	unsigned int ip_len;
-	struct iphdr *iph;
-
-	/* IPv4 header checksum failed */
-	if ((macrx & XCT_MACRX_HTY_M) != XCT_MACRX_HTY_IPV4_OK)
-		return -1;
-
-	/* non tcp packet */
-	skb_reset_network_header(skb);
-	iph = ip_hdr(skb);
-	if (iph->protocol != IPPROTO_TCP)
-		return -1;
-
-	ip_len = ip_hdrlen(skb);
-	skb_set_transport_header(skb, ip_len);
-	*tcph = tcp_hdr(skb);
-
-	/* check if ip header and tcp header are complete */
-	if (ntohs(iph->tot_len) < ip_len + tcp_hdrlen(skb))
-		return -1;
-
-	*hdr_flags = LRO_IPV4 | LRO_TCP;
-	*iphdr = iph;
-
-	return 0;
-}
-
 static int pasemi_mac_unmap_tx_skb(struct pasemi_mac *mac,
 				    const int nfrags,
 				    struct sk_buff *skb,
@@ -817,7 +781,7 @@ static int pasemi_mac_clean_rx(struct pasemi_mac_rxring *rx,
 		skb_put(skb, len-4);
 
 		skb->protocol = eth_type_trans(skb, mac->netdev);
-		lro_receive_skb(&mac->lro_mgr, skb, (void *)macrx);
+		napi_gro_receive(&mac->napi, skb);
 
 next:
 		RX_DESC(rx, n) = 0;
@@ -839,8 +803,6 @@ next:
 
 	rx_ring(mac)->next_to_clean = n;
 
-	lro_flush_all(&mac->lro_mgr);
-
 	/* Increase is in number of 16-byte entries, and since each descriptor
 	 * with an 8BRES takes up 3x8 bytes (padded to 4x8), increase with
 	 * count*2.
@@ -1754,16 +1716,6 @@ pasemi_mac_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->features = NETIF_F_IP_CSUM | NETIF_F_LLTX | NETIF_F_SG |
 			NETIF_F_HIGHDMA | NETIF_F_GSO;
 
-	mac->lro_mgr.max_aggr = LRO_MAX_AGGR;
-	mac->lro_mgr.max_desc = MAX_LRO_DESCRIPTORS;
-	mac->lro_mgr.lro_arr = mac->lro_desc;
-	mac->lro_mgr.get_skb_header = get_skb_hdr;
-	mac->lro_mgr.features = LRO_F_NAPI | LRO_F_EXTRACT_VLAN_ID;
-	mac->lro_mgr.dev = mac->netdev;
-	mac->lro_mgr.ip_summed = CHECKSUM_UNNECESSARY;
-	mac->lro_mgr.ip_summed_aggr = CHECKSUM_UNNECESSARY;
-
-
 	mac->dma_pdev = pci_get_device(PCI_VENDOR_ID_PASEMI, 0xa007, NULL);
 	if (!mac->dma_pdev) {
 		dev_err(&mac->pdev->dev, "Can't find DMA Controller\n");
diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.h b/drivers/net/ethernet/pasemi/pasemi_mac.h
index a580770..161c99a 100644
--- a/drivers/net/ethernet/pasemi/pasemi_mac.h
+++ b/drivers/net/ethernet/pasemi/pasemi_mac.h
@@ -31,7 +31,6 @@
 #define CS_RING_SIZE (TX_RING_SIZE*2)
 
 
-#define MAX_LRO_DESCRIPTORS 8
 #define MAX_CS	2
 
 struct pasemi_mac_txring {
@@ -84,10 +83,7 @@ struct pasemi_mac {
 
 	u8		mac_addr[ETH_ALEN];
 
-	struct net_lro_mgr	lro_mgr;
-	struct net_lro_desc	lro_desc[MAX_LRO_DESCRIPTORS];
 	struct timer_list	rxtimer;
-	unsigned int		lro_max_aggr;
 
 	struct pasemi_mac_txring *tx;
 	struct pasemi_mac_rxring *rx;
diff --git a/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c b/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c
index 25fae56..f046bfc 100644
--- a/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c
+++ b/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c
@@ -20,7 +20,6 @@
 #include <linux/netdevice.h>
 #include <linux/ethtool.h>
 #include <linux/pci.h>
-#include <linux/inet_lro.h>
 
 #include <asm/pasemi_dma.h>
 #include "pasemi_mac.h"
-- 
cgit v0.10.2


From 63e5f9535fcf238b0779e6bcaf092e3f627e6942 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <ben@decadent.org.uk>
Date: Mon, 15 Feb 2016 21:25:44 +0000
Subject: RDMA/nes: Replace LRO with GRO

GRO is simpler to use than the old inet_lro library, and is compatible
with forwarding and bridging configurations.

Compile-tested only.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/nes/Kconfig b/drivers/infiniband/hw/nes/Kconfig
index 846dc97..7964eba8 100644
--- a/drivers/infiniband/hw/nes/Kconfig
+++ b/drivers/infiniband/hw/nes/Kconfig
@@ -2,7 +2,6 @@ config INFINIBAND_NES
 	tristate "NetEffect RNIC Driver"
 	depends on PCI && INET && INFINIBAND
 	select LIBCRC32C
-	select INET_LRO
 	---help---
 	  This is the RDMA Network Interface Card (RNIC) driver for
 	  NetEffect Ethernet Cluster Server Adapters.
diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c
index 4713dd7..a1c6481 100644
--- a/drivers/infiniband/hw/nes/nes_hw.c
+++ b/drivers/infiniband/hw/nes/nes_hw.c
@@ -35,18 +35,11 @@
 #include <linux/moduleparam.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
-#include <linux/ip.h>
-#include <linux/tcp.h>
 #include <linux/if_vlan.h>
-#include <linux/inet_lro.h>
 #include <linux/slab.h>
 
 #include "nes.h"
 
-static unsigned int nes_lro_max_aggr = NES_LRO_MAX_AGGR;
-module_param(nes_lro_max_aggr, uint, 0444);
-MODULE_PARM_DESC(nes_lro_max_aggr, "NIC LRO max packet aggregation");
-
 static int wide_ppm_offset;
 module_param(wide_ppm_offset, int, 0644);
 MODULE_PARM_DESC(wide_ppm_offset, "Increase CX4 interface clock ppm offset, 0=100ppm (default), 1=300ppm");
@@ -1642,25 +1635,6 @@ static void nes_rq_wqes_timeout(unsigned long parm)
 }
 
 
-static int nes_lro_get_skb_hdr(struct sk_buff *skb, void **iphdr,
-			       void **tcph, u64 *hdr_flags, void *priv)
-{
-	unsigned int ip_len;
-	struct iphdr *iph;
-	skb_reset_network_header(skb);
-	iph = ip_hdr(skb);
-	if (iph->protocol != IPPROTO_TCP)
-		return -1;
-	ip_len = ip_hdrlen(skb);
-	skb_set_transport_header(skb, ip_len);
-	*tcph = tcp_hdr(skb);
-
-	*hdr_flags = LRO_IPV4 | LRO_TCP;
-	*iphdr = iph;
-	return 0;
-}
-
-
 /**
  * nes_init_nic_qp
  */
@@ -1895,14 +1869,6 @@ int nes_init_nic_qp(struct nes_device *nesdev, struct net_device *netdev)
 		return -ENOMEM;
 	}
 
-	nesvnic->lro_mgr.max_aggr       = nes_lro_max_aggr;
-	nesvnic->lro_mgr.max_desc       = NES_MAX_LRO_DESCRIPTORS;
-	nesvnic->lro_mgr.lro_arr        = nesvnic->lro_desc;
-	nesvnic->lro_mgr.get_skb_header = nes_lro_get_skb_hdr;
-	nesvnic->lro_mgr.features       = LRO_F_NAPI | LRO_F_EXTRACT_VLAN_ID;
-	nesvnic->lro_mgr.dev            = netdev;
-	nesvnic->lro_mgr.ip_summed      = CHECKSUM_UNNECESSARY;
-	nesvnic->lro_mgr.ip_summed_aggr = CHECKSUM_UNNECESSARY;
 	return 0;
 }
 
@@ -2809,13 +2775,10 @@ void nes_nic_ce_handler(struct nes_device *nesdev, struct nes_hw_nic_cq *cq)
 	u16 pkt_type;
 	u16 rqes_processed = 0;
 	u8 sq_cqes = 0;
-	u8 nes_use_lro = 0;
 
 	head = cq->cq_head;
 	cq_size = cq->cq_size;
 	cq->cqes_pending = 1;
-	if (nesvnic->netdev->features & NETIF_F_LRO)
-		nes_use_lro = 1;
 	do {
 		if (le32_to_cpu(cq->cq_vbase[head].cqe_words[NES_NIC_CQE_MISC_IDX]) &
 				NES_NIC_CQE_VALID) {
@@ -2950,10 +2913,7 @@ void nes_nic_ce_handler(struct nes_device *nesdev, struct nes_hw_nic_cq *cq)
 
 					__vlan_hwaccel_put_tag(rx_skb, htons(ETH_P_8021Q), vlan_tag);
 				}
-				if (nes_use_lro)
-					lro_receive_skb(&nesvnic->lro_mgr, rx_skb, NULL);
-				else
-					netif_receive_skb(rx_skb);
+				napi_gro_receive(&nesvnic->napi, rx_skb);
 
 skip_rx_indicate0:
 				;
@@ -2984,8 +2944,6 @@ skip_rx_indicate0:
 
 	} while (1);
 
-	if (nes_use_lro)
-		lro_flush_all(&nesvnic->lro_mgr);
 	if (sq_cqes) {
 		barrier();
 		/* restart the queue if it had been stopped */
diff --git a/drivers/infiniband/hw/nes/nes_hw.h b/drivers/infiniband/hw/nes/nes_hw.h
index c908020..1b66ef1 100644
--- a/drivers/infiniband/hw/nes/nes_hw.h
+++ b/drivers/infiniband/hw/nes/nes_hw.h
@@ -33,8 +33,6 @@
 #ifndef __NES_HW_H
 #define __NES_HW_H
 
-#include <linux/inet_lro.h>
-
 #define NES_PHY_TYPE_CX4       1
 #define NES_PHY_TYPE_1G        2
 #define NES_PHY_TYPE_ARGUS     4
@@ -1049,8 +1047,6 @@ struct nes_hw_tune_timer {
 #define NES_TIMER_ENABLE_LIMIT      4
 #define NES_MAX_LINK_INTERRUPTS     128
 #define NES_MAX_LINK_CHECK          200
-#define NES_MAX_LRO_DESCRIPTORS     32
-#define NES_LRO_MAX_AGGR            64
 
 struct nes_adapter {
 	u64              fw_ver;
@@ -1263,9 +1259,6 @@ struct nes_vnic {
 	u8  next_qp_nic_index;
 	u8  of_device_registered;
 	u8  rdma_enabled;
-	u32 lro_max_aggr;
-	struct net_lro_mgr lro_mgr;
-	struct net_lro_desc lro_desc[NES_MAX_LRO_DESCRIPTORS];
 	struct timer_list event_timer;
 	enum ib_event_type delayed_event;
 	enum ib_event_type last_dispatched_event;
diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c
index 6a0bdfa..3ea9e05 100644
--- a/drivers/infiniband/hw/nes/nes_nic.c
+++ b/drivers/infiniband/hw/nes/nes_nic.c
@@ -1085,9 +1085,6 @@ static const char nes_ethtool_stringset[][ETH_GSTRING_LEN] = {
 	"Free 4Kpbls",
 	"Free 256pbls",
 	"Timer Inits",
-	"LRO aggregated",
-	"LRO flushed",
-	"LRO no_desc",
 	"PAU CreateQPs",
 	"PAU DestroyQPs",
 };
@@ -1302,9 +1299,6 @@ static void nes_netdev_get_ethtool_stats(struct net_device *netdev,
 	target_stat_values[++index] = nesadapter->free_4kpbl;
 	target_stat_values[++index] = nesadapter->free_256pbl;
 	target_stat_values[++index] = int_mod_timer_init;
-	target_stat_values[++index] = nesvnic->lro_mgr.stats.aggregated;
-	target_stat_values[++index] = nesvnic->lro_mgr.stats.flushed;
-	target_stat_values[++index] = nesvnic->lro_mgr.stats.no_desc;
 	target_stat_values[++index] = atomic_read(&pau_qps_created);
 	target_stat_values[++index] = atomic_read(&pau_qps_destroyed);
 }
@@ -1709,7 +1703,6 @@ struct net_device *nes_netdev_init(struct nes_device *nesdev,
 		netdev->hw_features |= NETIF_F_TSO;
 
 	netdev->features = netdev->hw_features | NETIF_F_HIGHDMA | NETIF_F_HW_VLAN_CTAG_TX;
-	netdev->hw_features |= NETIF_F_LRO;
 
 	nes_debug(NES_DBG_INIT, "nesvnic = %p, reported features = 0x%lX, QPid = %d,"
 			" nic_index = %d, logical_port = %d, mac_index = %d.\n",
-- 
cgit v0.10.2


From 7bbf3cae65b6e438bf52033b63fdce4a86e89e17 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <ben@decadent.org.uk>
Date: Mon, 15 Feb 2016 21:25:57 +0000
Subject: ipv4: Remove inet_lro library

There are no longer any in-tree drivers that use it.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/inet_lro.h b/include/linux/inet_lro.h
deleted file mode 100644
index 9a715cf..0000000
--- a/include/linux/inet_lro.h
+++ /dev/null
@@ -1,142 +0,0 @@
-/*
- *  linux/include/linux/inet_lro.h
- *
- *  Large Receive Offload (ipv4 / tcp)
- *
- *  (C) Copyright IBM Corp. 2007
- *
- *  Authors:
- *       Jan-Bernd Themann <themann@de.ibm.com>
- *       Christoph Raisch <raisch@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#ifndef __INET_LRO_H_
-#define __INET_LRO_H_
-
-#include <net/ip.h>
-#include <net/tcp.h>
-
-/*
- * LRO statistics
- */
-
-struct net_lro_stats {
-	unsigned long aggregated;
-	unsigned long flushed;
-	unsigned long no_desc;
-};
-
-/*
- * LRO descriptor for a tcp session
- */
-struct net_lro_desc {
-	struct sk_buff *parent;
-	struct sk_buff *last_skb;
-	struct skb_frag_struct *next_frag;
-	struct iphdr *iph;
-	struct tcphdr *tcph;
-	__wsum  data_csum;
-	__be32 tcp_rcv_tsecr;
-	__be32 tcp_rcv_tsval;
-	__be32 tcp_ack;
-	u32 tcp_next_seq;
-	u32 skb_tot_frags_len;
-	u16 ip_tot_len;
-	u16 tcp_saw_tstamp; 		/* timestamps enabled */
-	__be16 tcp_window;
-	int pkt_aggr_cnt;		/* counts aggregated packets */
-	int vlan_packet;
-	int mss;
-	int active;
-};
-
-/*
- * Large Receive Offload (LRO) Manager
- *
- * Fields must be set by driver
- */
-
-struct net_lro_mgr {
-	struct net_device *dev;
-	struct net_lro_stats stats;
-
-	/* LRO features */
-	unsigned long features;
-#define LRO_F_NAPI            1  /* Pass packets to stack via NAPI */
-#define LRO_F_EXTRACT_VLAN_ID 2  /* Set flag if VLAN IDs are extracted
-				    from received packets and eth protocol
-				    is still ETH_P_8021Q */
-
-	/*
-	 * Set for generated SKBs that are not added to
-	 * the frag list in fragmented mode
-	 */
-	u32 ip_summed;
-	u32 ip_summed_aggr; /* Set in aggregated SKBs: CHECKSUM_UNNECESSARY
-			     * or CHECKSUM_NONE */
-
-	int max_desc; /* Max number of LRO descriptors  */
-	int max_aggr; /* Max number of LRO packets to be aggregated */
-
-	int frag_align_pad; /* Padding required to properly align layer 3
-			     * headers in generated skb when using frags */
-
-	struct net_lro_desc *lro_arr; /* Array of LRO descriptors */
-
-	/*
-	 * Optimized driver functions
-	 *
-	 * get_skb_header: returns tcp and ip header for packet in SKB
-	 */
-	int (*get_skb_header)(struct sk_buff *skb, void **ip_hdr,
-			      void **tcpudp_hdr, u64 *hdr_flags, void *priv);
-
-	/* hdr_flags: */
-#define LRO_IPV4 1 /* ip_hdr is IPv4 header */
-#define LRO_TCP  2 /* tcpudp_hdr is TCP header */
-
-	/*
-	 * get_frag_header: returns mac, tcp and ip header for packet in SKB
-	 *
-	 * @hdr_flags: Indicate what kind of LRO has to be done
-	 *             (IPv4/IPv6/TCP/UDP)
-	 */
-	int (*get_frag_header)(struct skb_frag_struct *frag, void **mac_hdr,
-			       void **ip_hdr, void **tcpudp_hdr, u64 *hdr_flags,
-			       void *priv);
-};
-
-/*
- * Processes a SKB
- *
- * @lro_mgr: LRO manager to use
- * @skb: SKB to aggregate
- * @priv: Private data that may be used by driver functions
- *        (for example get_tcp_ip_hdr)
- */
-
-void lro_receive_skb(struct net_lro_mgr *lro_mgr,
-		     struct sk_buff *skb,
-		     void *priv);
-/*
- * Forward all aggregated SKBs held by lro_mgr to network stack
- */
-
-void lro_flush_all(struct net_lro_mgr *lro_mgr);
-
-#endif
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 395d827..238225b 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -406,14 +406,6 @@ config INET_XFRM_MODE_BEET
 
 	  If unsure, say Y.
 
-config INET_LRO
-	tristate "Large Receive Offload (ipv4/tcp)"
-	default y
-	---help---
-	  Support for Large Receive Offload (ipv4/tcp).
-
-	  If unsure, say Y.
-
 config INET_DIAG
 	tristate "INET: socket monitoring interface"
 	default y
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 62c049b..bfa1336 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -32,7 +32,6 @@ obj-$(CONFIG_INET_ESP) += esp4.o
 obj-$(CONFIG_INET_IPCOMP) += ipcomp.o
 obj-$(CONFIG_INET_XFRM_TUNNEL) += xfrm4_tunnel.o
 obj-$(CONFIG_INET_XFRM_MODE_BEET) += xfrm4_mode_beet.o
-obj-$(CONFIG_INET_LRO) += inet_lro.o
 obj-$(CONFIG_INET_TUNNEL) += tunnel4.o
 obj-$(CONFIG_INET_XFRM_MODE_TRANSPORT) += xfrm4_mode_transport.o
 obj-$(CONFIG_INET_XFRM_MODE_TUNNEL) += xfrm4_mode_tunnel.o
diff --git a/net/ipv4/inet_lro.c b/net/ipv4/inet_lro.c
deleted file mode 100644
index f17ea49..0000000
--- a/net/ipv4/inet_lro.c
+++ /dev/null
@@ -1,374 +0,0 @@
-/*
- *  linux/net/ipv4/inet_lro.c
- *
- *  Large Receive Offload (ipv4 / tcp)
- *
- *  (C) Copyright IBM Corp. 2007
- *
- *  Authors:
- *       Jan-Bernd Themann <themann@de.ibm.com>
- *       Christoph Raisch <raisch@de.ibm.com>
- *
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-
-#include <linux/module.h>
-#include <linux/if_vlan.h>
-#include <linux/inet_lro.h>
-#include <net/checksum.h>
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Jan-Bernd Themann <themann@de.ibm.com>");
-MODULE_DESCRIPTION("Large Receive Offload (ipv4 / tcp)");
-
-#define TCP_HDR_LEN(tcph) (tcph->doff << 2)
-#define IP_HDR_LEN(iph) (iph->ihl << 2)
-#define TCP_PAYLOAD_LENGTH(iph, tcph) \
-	(ntohs(iph->tot_len) - IP_HDR_LEN(iph) - TCP_HDR_LEN(tcph))
-
-#define IPH_LEN_WO_OPTIONS 5
-#define TCPH_LEN_WO_OPTIONS 5
-#define TCPH_LEN_W_TIMESTAMP 8
-
-#define LRO_MAX_PG_HLEN 64
-
-#define LRO_INC_STATS(lro_mgr, attr) { lro_mgr->stats.attr++; }
-
-/*
- * Basic tcp checks whether packet is suitable for LRO
- */
-
-static int lro_tcp_ip_check(const struct iphdr *iph, const struct tcphdr *tcph,
-			    int len, const struct net_lro_desc *lro_desc)
-{
-        /* check ip header: don't aggregate padded frames */
-	if (ntohs(iph->tot_len) != len)
-		return -1;
-
-	if (TCP_PAYLOAD_LENGTH(iph, tcph) == 0)
-		return -1;
-
-	if (iph->ihl != IPH_LEN_WO_OPTIONS)
-		return -1;
-
-	if (tcph->cwr || tcph->ece || tcph->urg || !tcph->ack ||
-	    tcph->rst || tcph->syn || tcph->fin)
-		return -1;
-
-	if (INET_ECN_is_ce(ipv4_get_dsfield(iph)))
-		return -1;
-
-	if (tcph->doff != TCPH_LEN_WO_OPTIONS &&
-	    tcph->doff != TCPH_LEN_W_TIMESTAMP)
-		return -1;
-
-	/* check tcp options (only timestamp allowed) */
-	if (tcph->doff == TCPH_LEN_W_TIMESTAMP) {
-		__be32 *topt = (__be32 *)(tcph + 1);
-
-		if (*topt != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
-				   | (TCPOPT_TIMESTAMP << 8)
-				   | TCPOLEN_TIMESTAMP))
-			return -1;
-
-		/* timestamp should be in right order */
-		topt++;
-		if (lro_desc && after(ntohl(lro_desc->tcp_rcv_tsval),
-				      ntohl(*topt)))
-			return -1;
-
-		/* timestamp reply should not be zero */
-		topt++;
-		if (*topt == 0)
-			return -1;
-	}
-
-	return 0;
-}
-
-static void lro_update_tcp_ip_header(struct net_lro_desc *lro_desc)
-{
-	struct iphdr *iph = lro_desc->iph;
-	struct tcphdr *tcph = lro_desc->tcph;
-	__be32 *p;
-	__wsum tcp_hdr_csum;
-
-	tcph->ack_seq = lro_desc->tcp_ack;
-	tcph->window = lro_desc->tcp_window;
-
-	if (lro_desc->tcp_saw_tstamp) {
-		p = (__be32 *)(tcph + 1);
-		*(p+2) = lro_desc->tcp_rcv_tsecr;
-	}
-
-	csum_replace2(&iph->check, iph->tot_len, htons(lro_desc->ip_tot_len));
-	iph->tot_len = htons(lro_desc->ip_tot_len);
-
-	tcph->check = 0;
-	tcp_hdr_csum = csum_partial(tcph, TCP_HDR_LEN(tcph), 0);
-	lro_desc->data_csum = csum_add(lro_desc->data_csum, tcp_hdr_csum);
-	tcph->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
-					lro_desc->ip_tot_len -
-					IP_HDR_LEN(iph), IPPROTO_TCP,
-					lro_desc->data_csum);
-}
-
-static __wsum lro_tcp_data_csum(struct iphdr *iph, struct tcphdr *tcph, int len)
-{
-	__wsum tcp_csum;
-	__wsum tcp_hdr_csum;
-	__wsum tcp_ps_hdr_csum;
-
-	tcp_csum = ~csum_unfold(tcph->check);
-	tcp_hdr_csum = csum_partial(tcph, TCP_HDR_LEN(tcph), tcp_csum);
-
-	tcp_ps_hdr_csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
-					     len + TCP_HDR_LEN(tcph),
-					     IPPROTO_TCP, 0);
-
-	return csum_sub(csum_sub(tcp_csum, tcp_hdr_csum),
-			tcp_ps_hdr_csum);
-}
-
-static void lro_init_desc(struct net_lro_desc *lro_desc, struct sk_buff *skb,
-			  struct iphdr *iph, struct tcphdr *tcph)
-{
-	int nr_frags;
-	__be32 *ptr;
-	u32 tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
-
-	nr_frags = skb_shinfo(skb)->nr_frags;
-	lro_desc->parent = skb;
-	lro_desc->next_frag = &(skb_shinfo(skb)->frags[nr_frags]);
-	lro_desc->iph = iph;
-	lro_desc->tcph = tcph;
-	lro_desc->tcp_next_seq = ntohl(tcph->seq) + tcp_data_len;
-	lro_desc->tcp_ack = tcph->ack_seq;
-	lro_desc->tcp_window = tcph->window;
-
-	lro_desc->pkt_aggr_cnt = 1;
-	lro_desc->ip_tot_len = ntohs(iph->tot_len);
-
-	if (tcph->doff == 8) {
-		ptr = (__be32 *)(tcph+1);
-		lro_desc->tcp_saw_tstamp = 1;
-		lro_desc->tcp_rcv_tsval = *(ptr+1);
-		lro_desc->tcp_rcv_tsecr = *(ptr+2);
-	}
-
-	lro_desc->mss = tcp_data_len;
-	lro_desc->active = 1;
-
-	lro_desc->data_csum = lro_tcp_data_csum(iph, tcph,
-						tcp_data_len);
-}
-
-static inline void lro_clear_desc(struct net_lro_desc *lro_desc)
-{
-	memset(lro_desc, 0, sizeof(struct net_lro_desc));
-}
-
-static void lro_add_common(struct net_lro_desc *lro_desc, struct iphdr *iph,
-			   struct tcphdr *tcph, int tcp_data_len)
-{
-	struct sk_buff *parent = lro_desc->parent;
-	__be32 *topt;
-
-	lro_desc->pkt_aggr_cnt++;
-	lro_desc->ip_tot_len += tcp_data_len;
-	lro_desc->tcp_next_seq += tcp_data_len;
-	lro_desc->tcp_window = tcph->window;
-	lro_desc->tcp_ack = tcph->ack_seq;
-
-	/* don't update tcp_rcv_tsval, would not work with PAWS */
-	if (lro_desc->tcp_saw_tstamp) {
-		topt = (__be32 *) (tcph + 1);
-		lro_desc->tcp_rcv_tsecr = *(topt + 2);
-	}
-
-	lro_desc->data_csum = csum_block_add(lro_desc->data_csum,
-					     lro_tcp_data_csum(iph, tcph,
-							       tcp_data_len),
-					     parent->len);
-
-	parent->len += tcp_data_len;
-	parent->data_len += tcp_data_len;
-	if (tcp_data_len > lro_desc->mss)
-		lro_desc->mss = tcp_data_len;
-}
-
-static void lro_add_packet(struct net_lro_desc *lro_desc, struct sk_buff *skb,
-			   struct iphdr *iph, struct tcphdr *tcph)
-{
-	struct sk_buff *parent = lro_desc->parent;
-	int tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
-
-	lro_add_common(lro_desc, iph, tcph, tcp_data_len);
-
-	skb_pull(skb, (skb->len - tcp_data_len));
-	parent->truesize += skb->truesize;
-
-	if (lro_desc->last_skb)
-		lro_desc->last_skb->next = skb;
-	else
-		skb_shinfo(parent)->frag_list = skb;
-
-	lro_desc->last_skb = skb;
-}
-
-
-static int lro_check_tcp_conn(struct net_lro_desc *lro_desc,
-			      struct iphdr *iph,
-			      struct tcphdr *tcph)
-{
-	if ((lro_desc->iph->saddr != iph->saddr) ||
-	    (lro_desc->iph->daddr != iph->daddr) ||
-	    (lro_desc->tcph->source != tcph->source) ||
-	    (lro_desc->tcph->dest != tcph->dest))
-		return -1;
-	return 0;
-}
-
-static struct net_lro_desc *lro_get_desc(struct net_lro_mgr *lro_mgr,
-					 struct net_lro_desc *lro_arr,
-					 struct iphdr *iph,
-					 struct tcphdr *tcph)
-{
-	struct net_lro_desc *lro_desc = NULL;
-	struct net_lro_desc *tmp;
-	int max_desc = lro_mgr->max_desc;
-	int i;
-
-	for (i = 0; i < max_desc; i++) {
-		tmp = &lro_arr[i];
-		if (tmp->active)
-			if (!lro_check_tcp_conn(tmp, iph, tcph)) {
-				lro_desc = tmp;
-				goto out;
-			}
-	}
-
-	for (i = 0; i < max_desc; i++) {
-		if (!lro_arr[i].active) {
-			lro_desc = &lro_arr[i];
-			goto out;
-		}
-	}
-
-	LRO_INC_STATS(lro_mgr, no_desc);
-out:
-	return lro_desc;
-}
-
-static void lro_flush(struct net_lro_mgr *lro_mgr,
-		      struct net_lro_desc *lro_desc)
-{
-	if (lro_desc->pkt_aggr_cnt > 1)
-		lro_update_tcp_ip_header(lro_desc);
-
-	skb_shinfo(lro_desc->parent)->gso_size = lro_desc->mss;
-
-	if (lro_mgr->features & LRO_F_NAPI)
-		netif_receive_skb(lro_desc->parent);
-	else
-		netif_rx(lro_desc->parent);
-
-	LRO_INC_STATS(lro_mgr, flushed);
-	lro_clear_desc(lro_desc);
-}
-
-static int __lro_proc_skb(struct net_lro_mgr *lro_mgr, struct sk_buff *skb,
-			  void *priv)
-{
-	struct net_lro_desc *lro_desc;
-	struct iphdr *iph;
-	struct tcphdr *tcph;
-	u64 flags;
-	int vlan_hdr_len = 0;
-
-	if (!lro_mgr->get_skb_header ||
-	    lro_mgr->get_skb_header(skb, (void *)&iph, (void *)&tcph,
-				    &flags, priv))
-		goto out;
-
-	if (!(flags & LRO_IPV4) || !(flags & LRO_TCP))
-		goto out;
-
-	lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
-	if (!lro_desc)
-		goto out;
-
-	if ((skb->protocol == htons(ETH_P_8021Q)) &&
-	    !(lro_mgr->features & LRO_F_EXTRACT_VLAN_ID))
-		vlan_hdr_len = VLAN_HLEN;
-
-	if (!lro_desc->active) { /* start new lro session */
-		if (lro_tcp_ip_check(iph, tcph, skb->len - vlan_hdr_len, NULL))
-			goto out;
-
-		skb->ip_summed = lro_mgr->ip_summed_aggr;
-		lro_init_desc(lro_desc, skb, iph, tcph);
-		LRO_INC_STATS(lro_mgr, aggregated);
-		return 0;
-	}
-
-	if (lro_desc->tcp_next_seq != ntohl(tcph->seq))
-		goto out2;
-
-	if (lro_tcp_ip_check(iph, tcph, skb->len, lro_desc))
-		goto out2;
-
-	lro_add_packet(lro_desc, skb, iph, tcph);
-	LRO_INC_STATS(lro_mgr, aggregated);
-
-	if ((lro_desc->pkt_aggr_cnt >= lro_mgr->max_aggr) ||
-	    lro_desc->parent->len > (0xFFFF - lro_mgr->dev->mtu))
-		lro_flush(lro_mgr, lro_desc);
-
-	return 0;
-
-out2: /* send aggregated SKBs to stack */
-	lro_flush(lro_mgr, lro_desc);
-
-out:
-	return 1;
-}
-
-void lro_receive_skb(struct net_lro_mgr *lro_mgr,
-		     struct sk_buff *skb,
-		     void *priv)
-{
-	if (__lro_proc_skb(lro_mgr, skb, priv)) {
-		if (lro_mgr->features & LRO_F_NAPI)
-			netif_receive_skb(skb);
-		else
-			netif_rx(skb);
-	}
-}
-EXPORT_SYMBOL(lro_receive_skb);
-
-void lro_flush_all(struct net_lro_mgr *lro_mgr)
-{
-	int i;
-	struct net_lro_desc *lro_desc = lro_mgr->lro_arr;
-
-	for (i = 0; i < lro_mgr->max_desc; i++) {
-		if (lro_desc[i].active)
-			lro_flush(lro_mgr, &lro_desc[i]);
-	}
-}
-EXPORT_SYMBOL(lro_flush_all);
-- 
cgit v0.10.2


From 2818ccd95684d39cc3bdad579e02ae56c0d6de88 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 13 Jan 2016 16:51:38 -0800
Subject: i40e: move sync_vsi_filters up in service_task

The sync_vsi_filters function is moved up in the service_task because
it may need to request a reset, and we don't want to wait another round
of service task time.

NOTE: Filters will be replayed by sync_vsi_filters including broadcast
and promiscuous settings.

Also, added some error handling in this space in case any of these
fail the driver will retry correctly.

Also update copyright year.

Change-ID: I23f3d552100baecea69466339f738f27614efd47
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 04417e6..e974db3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -2168,6 +2168,10 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		}
 	}
 out:
+	/* if something went wrong then set the changed flag so we try again */
+	if (retval)
+		vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
+
 	clear_bit(__I40E_CONFIG_BUSY, &vsi->state);
 	return retval;
 }
@@ -7113,6 +7117,7 @@ static void i40e_service_task(struct work_struct *work)
 	}
 
 	i40e_detect_recover_hung(pf);
+	i40e_sync_filters_subtask(pf);
 	i40e_reset_subtask(pf);
 	i40e_handle_mdd_event(pf);
 	i40e_vc_process_vflr_event(pf);
-- 
cgit v0.10.2


From f1bbad339cdd74cea945e33d758200d42310be17 Mon Sep 17 00:00:00 2001
From: Neerav Parikh <neerav.parikh@intel.com>
Date: Wed, 13 Jan 2016 16:51:39 -0800
Subject: i40e: Make the DCB firmware checks for X710/XL710 only

Make the DCB firmware version related checks specific to
X710 and XL710 only. These checks are not required for
X722 family of devices.

Introduced an inline routine to help determine if the
MAC type is X710/XL710 or not.

Moved the firmware version related checks in i40e_sw_init()
and defined flags for different cases

Fix the version check to allow using "Set LLDP MIB" AQ
for beyond FVL4 FW releases.

Change-ID: Ib78288343de983aa0354fc28aa36e99b073662c0
Signed-off-by: Neerav Parikh <neerav.parikh@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 05af33e..7bfd062 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -138,6 +138,19 @@
 /* default to trying for four seconds */
 #define I40E_TRY_LINK_TIMEOUT (4 * HZ)
 
+/**
+ * i40e_is_mac_710 - Return true if MAC is X710/XL710
+ * @hw: ptr to the hardware info
+ **/
+static inline bool i40e_is_mac_710(struct i40e_hw *hw)
+{
+	if ((hw->mac.type == I40E_MAC_X710) ||
+	    (hw->mac.type == I40E_MAC_XL710))
+		return true;
+
+	return false;
+}
+
 /* driver state flags */
 enum i40e_state_t {
 	__I40E_TESTING,
@@ -342,6 +355,9 @@ struct i40e_pf {
 #define I40E_FLAG_NO_PCI_LINK_CHECK		BIT_ULL(42)
 #define I40E_FLAG_100M_SGMII_CAPABLE		BIT_ULL(43)
 #define I40E_FLAG_RESTART_AUTONEG		BIT_ULL(44)
+#define I40E_FLAG_NO_DCB_SUPPORT		BIT_ULL(45)
+#define I40E_FLAG_USE_SET_LLDP_MIB		BIT_ULL(46)
+#define I40E_FLAG_STOP_FW_LLDP			BIT_ULL(47)
 #define I40E_FLAG_PF_MAC			BIT_ULL(50)
 
 	/* tracks features that get auto disabled by errors */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index e974db3..81b7895 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5012,8 +5012,7 @@ static int i40e_init_pf_dcb(struct i40e_pf *pf)
 	int err = 0;
 
 	/* Do not enable DCB for SW1 and SW2 images even if the FW is capable */
-	if (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 33)) ||
-	    (pf->hw.aq.fw_maj_ver < 4))
+	if (pf->flags & I40E_FLAG_NO_DCB_SUPPORT)
 		goto out;
 
 	/* Get the initial DCB configuration */
@@ -8425,11 +8424,25 @@ static int i40e_sw_init(struct i40e_pf *pf)
 				 pf->hw.func_caps.fd_filters_best_effort;
 	}
 
-	if (((pf->hw.mac.type == I40E_MAC_X710) ||
-	     (pf->hw.mac.type == I40E_MAC_XL710)) &&
+	if (i40e_is_mac_710(&pf->hw) &&
 	    (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 33)) ||
-	    (pf->hw.aq.fw_maj_ver < 4)))
+	    (pf->hw.aq.fw_maj_ver < 4))) {
 		pf->flags |= I40E_FLAG_RESTART_AUTONEG;
+		/* No DCB support  for FW < v4.33 */
+		pf->flags |= I40E_FLAG_NO_DCB_SUPPORT;
+	}
+
+	/* Disable FW LLDP if FW < v4.3 */
+	if (i40e_is_mac_710(&pf->hw) &&
+	    (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 3)) ||
+	    (pf->hw.aq.fw_maj_ver < 4)))
+		pf->flags |= I40E_FLAG_STOP_FW_LLDP;
+
+	/* Use the FW Set LLDP MIB API if FW > v4.40 */
+	if (i40e_is_mac_710(&pf->hw) &&
+	    (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver >= 40)) ||
+	    (pf->hw.aq.fw_maj_ver >= 5)))
+		pf->flags |= I40E_FLAG_USE_SET_LLDP_MIB;
 
 	if (pf->hw.func_caps.vmdq) {
 		pf->num_vmdq_vsis = I40E_DEFAULT_NUM_VMDQ_VSI;
@@ -8458,6 +8471,7 @@ static int i40e_sw_init(struct i40e_pf *pf)
 			     I40E_FLAG_WB_ON_ITR_CAPABLE |
 			     I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE |
 			     I40E_FLAG_100M_SGMII_CAPABLE |
+			     I40E_FLAG_USE_SET_LLDP_MIB |
 			     I40E_FLAG_GENEVE_OFFLOAD_CAPABLE;
 	} else if ((pf->hw.aq.api_maj_ver > 1) ||
 		   ((pf->hw.aq.api_maj_ver == 1) &&
@@ -10825,8 +10839,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * Ignore error return codes because if it was already disabled via
 	 * hardware settings this will fail
 	 */
-	if (((pf->hw.aq.fw_maj_ver == 4) && (pf->hw.aq.fw_min_ver < 3)) ||
-	    (pf->hw.aq.fw_maj_ver < 4)) {
+	if (pf->flags & I40E_FLAG_STOP_FW_LLDP) {
 		dev_info(&pdev->dev, "Stopping firmware LLDP agent.\n");
 		i40e_aq_stop_lldp(hw, true, NULL);
 	}
-- 
cgit v0.10.2


From 67be6eb2473ac86e71f68c9470f8e00f39df7258 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 13 Jan 2016 16:51:40 -0800
Subject: i40e: set shared bit for multicast filters

Add the use of the new Shared MAC filter bit for multicast and broadcast
filters in order to make better use of the filters available from the
device.  The FW folks have assured me that setting this bit on older FW
will have no affect, so we don't need a version check.

Also fixed a stray indent problem nearby.

Also update copyright year.

Change-ID: I4c5826a32594382a7937a592a24d228588cee7aa
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 976b03f..edfea38 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -2435,6 +2435,7 @@ i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid,
 		(struct i40e_aqc_macvlan *)&desc.params.raw;
 	i40e_status status;
 	u16 buf_size;
+	int i;
 
 	if (count == 0 || !mv_list || !hw)
 		return I40E_ERR_PARAM;
@@ -2448,12 +2449,17 @@ i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid,
 	cmd->seid[1] = 0;
 	cmd->seid[2] = 0;
 
+	for (i = 0; i < count; i++)
+		if (is_multicast_ether_addr(mv_list[i].mac_addr))
+			mv_list[i].flags |=
+			       cpu_to_le16(I40E_AQC_MACVLAN_ADD_USE_SHARED_MAC);
+
 	desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD));
 	if (buf_size > I40E_AQ_LARGE_BUF)
 		desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB);
 
 	status = i40e_asq_send_command(hw, &desc, mv_list, buf_size,
-				    cmd_details);
+				       cmd_details);
 
 	return status;
 }
-- 
cgit v0.10.2


From 8a187f448ede55147dfa2fdcacfacd795747972d Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 13 Jan 2016 16:51:41 -0800
Subject: i40e: add VEB stat control and remove L2 cloud filter

With the latest firmware, statistics gathering can now be enabled and
disabled in the HW switch, so we need to add a parameter to allow the
driver to set it as desired.  At the same time, the L2 cloud filtering
parameter has been removed as it was never used.

Older drivers working with the newer firmware and newer drivers working
with older firmware will not run into problems with these bits as the
defaults are reasonable and there is no overlap in the bit definitions.
Also, newer drivers will be forced to update because of the change in
function call parameters, a reminder that the functionality exists.

Also update copyright year.

Change-ID: I9acb9160b892ca3146f2f11a88fdcd86be3cadcc
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index edfea38..354e36c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -2308,8 +2308,8 @@ i40e_status i40e_update_link_info(struct i40e_hw *hw)
  * @downlink_seid: the VSI SEID
  * @enabled_tc: bitmap of TCs to be enabled
  * @default_port: true for default port VSI, false for control port
- * @enable_l2_filtering: true to add L2 filter table rules to regular forwarding rules for cloud support
  * @veb_seid: pointer to where to put the resulting VEB SEID
+ * @enable_stats: true to turn on VEB stats
  * @cmd_details: pointer to command details structure or NULL
  *
  * This asks the FW to add a VEB between the uplink and downlink
@@ -2317,8 +2317,8 @@ i40e_status i40e_update_link_info(struct i40e_hw *hw)
  **/
 i40e_status i40e_aq_add_veb(struct i40e_hw *hw, u16 uplink_seid,
 				u16 downlink_seid, u8 enabled_tc,
-				bool default_port, bool enable_l2_filtering,
-				u16 *veb_seid,
+				bool default_port, u16 *veb_seid,
+				bool enable_stats,
 				struct i40e_asq_cmd_details *cmd_details)
 {
 	struct i40e_aq_desc desc;
@@ -2345,8 +2345,9 @@ i40e_status i40e_aq_add_veb(struct i40e_hw *hw, u16 uplink_seid,
 	else
 		veb_flags |= I40E_AQC_ADD_VEB_PORT_TYPE_DATA;
 
-	if (enable_l2_filtering)
-		veb_flags |= I40E_AQC_ADD_VEB_ENABLE_L2_FILTER;
+	/* reverse logic here: set the bitflag to disable the stats */
+	if (!enable_stats)
+		veb_flags |= I40E_AQC_ADD_VEB_ENABLE_DISABLE_STATS;
 
 	cmd->veb_flags = cpu_to_le16(veb_flags);
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 81b7895..95fb342 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -10075,7 +10075,7 @@ static int i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi)
 	/* get a VEB from the hardware */
 	ret = i40e_aq_add_veb(&pf->hw, veb->uplink_seid, vsi->seid,
 			      veb->enabled_tc, is_default,
-			      is_cloud, &veb->seid, NULL);
+			      &veb->seid, is_cloud, NULL);
 	if (ret) {
 		dev_info(&pf->pdev->dev,
 			 "couldn't add VEB, err %s aq_err %s\n",
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index 45af29b..e8deabd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -138,8 +138,8 @@ i40e_status i40e_aq_update_vsi_params(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_add_veb(struct i40e_hw *hw, u16 uplink_seid,
 				u16 downlink_seid, u8 enabled_tc,
-				bool default_port, bool enable_l2_filtering,
-				u16 *pveb_seid,
+				bool default_port, u16 *pveb_seid,
+				bool enable_stats,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_get_veb_parameters(struct i40e_hw *hw,
 				u16 veb_seid, u16 *switch_id, bool *floating,
-- 
cgit v0.10.2


From 66fc360a0e9dd17e2c6ea02c15b02590b583abd1 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 13 Jan 2016 16:51:42 -0800
Subject: i40e: use new add_veb calling with VEB stats control

The new parameters for add_veb allow us to enable and disable VEB stats,
so let's use them.

Update copyright year.

Change-ID: Ie6e68c68e2d1d459e42168eda661051b56bf0a65
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 89ad2f7..230fa40 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -2785,10 +2785,15 @@ static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
 		pf->auto_disable_flags |= I40E_FLAG_FD_ATR_ENABLED;
 	}
 
-	if (flags & I40E_PRIV_FLAGS_VEB_STATS)
+	if ((flags & I40E_PRIV_FLAGS_VEB_STATS) &&
+	    !(pf->flags & I40E_FLAG_VEB_STATS_ENABLED)) {
 		pf->flags |= I40E_FLAG_VEB_STATS_ENABLED;
-	else
+		reset_required = true;
+	} else if (!(flags & I40E_PRIV_FLAGS_VEB_STATS) &&
+		   (pf->flags & I40E_FLAG_VEB_STATS_ENABLED)) {
 		pf->flags &= ~I40E_FLAG_VEB_STATS_ENABLED;
+		reset_required = true;
+	}
 
 	if ((flags & I40E_PRIV_FLAGS_HW_ATR_EVICT) &&
 	    (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE))
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 95fb342..0acec51 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -10069,13 +10069,13 @@ static int i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi)
 {
 	struct i40e_pf *pf = veb->pf;
 	bool is_default = veb->pf->cur_promisc;
-	bool is_cloud = false;
+	bool enable_stats = !!(pf->flags & I40E_FLAG_VEB_STATS_ENABLED);
 	int ret;
 
 	/* get a VEB from the hardware */
 	ret = i40e_aq_add_veb(&pf->hw, veb->uplink_seid, vsi->seid,
 			      veb->enabled_tc, is_default,
-			      &veb->seid, is_cloud, NULL);
+			      &veb->seid, enable_stats, NULL);
 	if (ret) {
 		dev_info(&pf->pdev->dev,
 			 "couldn't add VEB, err %s aq_err %s\n",
-- 
cgit v0.10.2


From ecc6a239e8dd233951bd3337f34d9ba4c46720c9 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Wed, 13 Jan 2016 16:51:43 -0800
Subject: i40e: Refactor force_wb and WB_ON_ITR functionality code

Now that the Force-WriteBack functionality in X710/XL710 devices
has been moved out of the clean routine and into the service task,
we need to make sure WriteBack-On-ITR is separated out since it
is still called from clean.

In the X722 devices, Force-WriteBack implies WriteBack-On-ITR but
without the interrupt, which put the driver into a missed
interrupt scenario and a potential tx-timeout report.

With this patch, we break the two functions out, and call the
appropriate ones at the right place. This will avoid creating missed
interrupt like scenarios for X722 devices.

Also update copyright year in file headers.

Change-ID: Iacbde39f95f332f82be8736864675052c3583a40
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 6d1dd60..7dfd45e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -774,37 +774,48 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
 }
 
 /**
- * i40e_force_wb - Arm hardware to do a wb on noncache aligned descriptors
+ * i40e_enable_wb_on_itr - Arm hardware to do a wb, interrupts are not enabled
  * @vsi: the VSI we care about
- * @q_vector: the vector  on which to force writeback
+ * @q_vector: the vector on which to enable writeback
  *
  **/
-void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
+static void i40e_enable_wb_on_itr(struct i40e_vsi *vsi,
+				  struct i40e_q_vector *q_vector)
 {
 	u16 flags = q_vector->tx.ring[0].flags;
+	u32 val;
 
-	if (flags & I40E_TXR_FLAGS_WB_ON_ITR) {
-		u32 val;
+	if (!(flags & I40E_TXR_FLAGS_WB_ON_ITR))
+		return;
 
-		if (q_vector->arm_wb_state)
-			return;
+	if (q_vector->arm_wb_state)
+		return;
 
-		if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
-			val = I40E_PFINT_DYN_CTLN_WB_ON_ITR_MASK |
-			      I40E_PFINT_DYN_CTLN_ITR_INDX_MASK; /* set noitr */
+	if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
+		val = I40E_PFINT_DYN_CTLN_WB_ON_ITR_MASK |
+		      I40E_PFINT_DYN_CTLN_ITR_INDX_MASK; /* set noitr */
 
-			wr32(&vsi->back->hw,
-			     I40E_PFINT_DYN_CTLN(q_vector->v_idx +
-						 vsi->base_vector - 1),
-			     val);
-		} else {
-			val = I40E_PFINT_DYN_CTL0_WB_ON_ITR_MASK |
-			      I40E_PFINT_DYN_CTL0_ITR_INDX_MASK; /* set noitr */
+		wr32(&vsi->back->hw,
+		     I40E_PFINT_DYN_CTLN(q_vector->v_idx + vsi->base_vector - 1),
+		     val);
+	} else {
+		val = I40E_PFINT_DYN_CTL0_WB_ON_ITR_MASK |
+		      I40E_PFINT_DYN_CTL0_ITR_INDX_MASK; /* set noitr */
 
-			wr32(&vsi->back->hw, I40E_PFINT_DYN_CTL0, val);
-		}
-		q_vector->arm_wb_state = true;
-	} else if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
+		wr32(&vsi->back->hw, I40E_PFINT_DYN_CTL0, val);
+	}
+	q_vector->arm_wb_state = true;
+}
+
+/**
+ * i40e_force_wb - Issue SW Interrupt so HW does a wb
+ * @vsi: the VSI we care about
+ * @q_vector: the vector  on which to force writeback
+ *
+ **/
+void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
+{
+	if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
 		u32 val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
 			  I40E_PFINT_DYN_CTLN_ITR_INDX_MASK | /* set noitr */
 			  I40E_PFINT_DYN_CTLN_SWINT_TRIG_MASK |
@@ -1946,7 +1957,7 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 tx_only:
 		if (arm_wb) {
 			q_vector->tx.ring[0].tx_stats.tx_force_wb++;
-			i40e_force_wb(vsi, q_vector);
+			i40e_enable_wb_on_itr(vsi, q_vector);
 		}
 		return budget;
 	}
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 4205aef..1c62cf5 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -292,40 +292,49 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
 }
 
 /**
- * i40evf_force_wb -Arm hardware to do a wb on noncache aligned descriptors
+ * i40evf_enable_wb_on_itr - Arm hardware to do a wb, interrupts are not enabled
  * @vsi: the VSI we care about
- * @q_vector: the vector  on which to force writeback
+ * @q_vector: the vector on which to enable writeback
  *
  **/
-static void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
+static void i40e_enable_wb_on_itr(struct i40e_vsi *vsi,
+				  struct i40e_q_vector *q_vector)
 {
 	u16 flags = q_vector->tx.ring[0].flags;
+	u32 val;
 
-	if (flags & I40E_TXR_FLAGS_WB_ON_ITR) {
-		u32 val;
+	if (!(flags & I40E_TXR_FLAGS_WB_ON_ITR))
+		return;
 
-		if (q_vector->arm_wb_state)
-			return;
+	if (q_vector->arm_wb_state)
+		return;
 
-		val = I40E_VFINT_DYN_CTLN1_WB_ON_ITR_MASK |
-		      I40E_VFINT_DYN_CTLN1_ITR_INDX_MASK; /* set noitr */
+	val = I40E_VFINT_DYN_CTLN1_WB_ON_ITR_MASK |
+	      I40E_VFINT_DYN_CTLN1_ITR_INDX_MASK; /* set noitr */
 
-		wr32(&vsi->back->hw,
-		     I40E_VFINT_DYN_CTLN1(q_vector->v_idx +
-					  vsi->base_vector - 1),
-		     val);
-		q_vector->arm_wb_state = true;
-	} else {
-		u32 val = I40E_VFINT_DYN_CTLN1_INTENA_MASK |
-			  I40E_VFINT_DYN_CTLN1_ITR_INDX_MASK | /* set noitr */
-			  I40E_VFINT_DYN_CTLN1_SWINT_TRIG_MASK |
-			  I40E_VFINT_DYN_CTLN1_SW_ITR_INDX_ENA_MASK;
-			  /* allow 00 to be written to the index */
-
-		wr32(&vsi->back->hw,
-		     I40E_VFINT_DYN_CTLN1(q_vector->v_idx +
-					  vsi->base_vector - 1), val);
-	}
+	wr32(&vsi->back->hw,
+	     I40E_VFINT_DYN_CTLN1(q_vector->v_idx +
+				  vsi->base_vector - 1), val);
+	q_vector->arm_wb_state = true;
+}
+
+/**
+ * i40evf_force_wb - Issue SW Interrupt so HW does a wb
+ * @vsi: the VSI we care about
+ * @q_vector: the vector  on which to force writeback
+ *
+ **/
+void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
+{
+	u32 val = I40E_VFINT_DYN_CTLN1_INTENA_MASK |
+		  I40E_VFINT_DYN_CTLN1_ITR_INDX_MASK | /* set noitr */
+		  I40E_VFINT_DYN_CTLN1_SWINT_TRIG_MASK |
+		  I40E_VFINT_DYN_CTLN1_SW_ITR_INDX_ENA_MASK
+		  /* allow 00 to be written to the index */;
+
+	wr32(&vsi->back->hw,
+	     I40E_VFINT_DYN_CTLN1(q_vector->v_idx + vsi->base_vector - 1),
+	     val);
 }
 
 /**
@@ -1384,7 +1393,7 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget)
 tx_only:
 		if (arm_wb) {
 			q_vector->tx.ring[0].tx_stats.tx_force_wb++;
-			i40evf_force_wb(vsi, q_vector);
+			i40e_enable_wb_on_itr(vsi, q_vector);
 		}
 		return budget;
 	}
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index e29bb3e..da701c5 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -324,6 +324,7 @@ int i40evf_setup_rx_descriptors(struct i40e_ring *rx_ring);
 void i40evf_free_tx_resources(struct i40e_ring *tx_ring);
 void i40evf_free_rx_resources(struct i40e_ring *rx_ring);
 int i40evf_napi_poll(struct napi_struct *napi, int budget);
+void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
 u32 i40evf_get_tx_pending(struct i40e_ring *ring);
 
 /**
-- 
cgit v0.10.2


From eaab59e9d2908df1c31d94f2da1aa3ebeebd8381 Mon Sep 17 00:00:00 2001
From: Catherine Sullivan <catherine.sullivan@intel.com>
Date: Wed, 13 Jan 2016 16:51:44 -0800
Subject: i40evf: Change vf driver string to reflect all products i40evf
 supports

Change the driver string to 40-10 Gigabit instead of XL710/X710 for X722
and all future products.

Also update copyright year in file header.

Change-ID: I57fae656b36dc4eb682b2b7a054f8f48f3589149
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 045cc7f..faa1bca 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -32,7 +32,7 @@ static int i40evf_close(struct net_device *netdev);
 
 char i40evf_driver_name[] = "i40evf";
 static const char i40evf_driver_string[] =
-	"Intel(R) XL710/X710 Virtual Function Network Driver";
+	"Intel(R) 40-10 Gigabit Virtual Function Network Driver";
 
 #define DRV_KERN "-k"
 
-- 
cgit v0.10.2


From 40d72a50986271e5c30f111e08ebd6d846708bbc Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 13 Jan 2016 16:51:45 -0800
Subject: i40e/i40evf: don't lose interrupts

While re-enabling interrupts the driver would clear all pending
causes. This meant that if an interrupt was generated while the driver
was cleaning or polling with interrupts disabled, then that interrupt
was lost.  This could cause a queue to become dead, especially for
receive.  Refactored the enable_icr0 function in order to allow
it to be decided by the caller whether the CLEARPBA (clear pending
events) bit will be set while re-enabling the interrupt.

Also update copyright year in file headers.

Change-ID: Ic1db100a05e13c98919057696db147a258ca365a
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 7bfd062..5ea431d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -767,6 +767,9 @@ static inline void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector)
 	struct i40e_hw *hw = &pf->hw;
 	u32 val;
 
+	/* definitely clear the PBA here, as this function is meant to
+	 * clean out all previous interrupts AND enable the interrupt
+	 */
 	val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
 	      I40E_PFINT_DYN_CTLN_CLEARPBA_MASK |
 	      (I40E_ITR_NONE << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT);
@@ -775,7 +778,7 @@ static inline void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector)
 }
 
 void i40e_irq_dynamic_disable_icr0(struct i40e_pf *pf);
-void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf);
+void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf, bool clearpba);
 #ifdef I40E_FCOE
 struct rtnl_link_stats64 *i40e_get_netdev_stats_struct(
 					     struct net_device *netdev,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 0acec51..8bc848f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3257,14 +3257,15 @@ void i40e_irq_dynamic_disable_icr0(struct i40e_pf *pf)
 /**
  * i40e_irq_dynamic_enable_icr0 - Enable default interrupt generation for icr0
  * @pf: board private structure
+ * @clearpba: true when all pending interrupt events should be cleared
  **/
-void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf)
+void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf, bool clearpba)
 {
 	struct i40e_hw *hw = &pf->hw;
 	u32 val;
 
 	val = I40E_PFINT_DYN_CTL0_INTENA_MASK   |
-	      I40E_PFINT_DYN_CTL0_CLEARPBA_MASK |
+	      (clearpba ? I40E_PFINT_DYN_CTL0_CLEARPBA_MASK : 0) |
 	      (I40E_ITR_NONE << I40E_PFINT_DYN_CTL0_ITR_INDX_SHIFT);
 
 	wr32(hw, I40E_PFINT_DYN_CTL0, val);
@@ -3396,7 +3397,7 @@ static int i40e_vsi_enable_irq(struct i40e_vsi *vsi)
 		for (i = 0; i < vsi->num_q_vectors; i++)
 			i40e_irq_dynamic_enable(vsi, i);
 	} else {
-		i40e_irq_dynamic_enable_icr0(pf);
+		i40e_irq_dynamic_enable_icr0(pf, true);
 	}
 
 	i40e_flush(&pf->hw);
@@ -3542,7 +3543,7 @@ enable_intr:
 	wr32(hw, I40E_PFINT_ICR0_ENA, ena_mask);
 	if (!test_bit(__I40E_DOWN, &pf->state)) {
 		i40e_service_event_schedule(pf);
-		i40e_irq_dynamic_enable_icr0(pf);
+		i40e_irq_dynamic_enable_icr0(pf, false);
 	}
 
 	return ret;
@@ -7858,7 +7859,7 @@ static int i40e_setup_misc_vector(struct i40e_pf *pf)
 
 	i40e_flush(hw);
 
-	i40e_irq_dynamic_enable_icr0(pf);
+	i40e_irq_dynamic_enable_icr0(pf, true);
 
 	return err;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 7dfd45e..353e5a0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1810,7 +1810,9 @@ static u32 i40e_buildreg_itr(const int type, const u16 itr)
 	u32 val;
 
 	val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
-	      I40E_PFINT_DYN_CTLN_CLEARPBA_MASK |
+	      /* Don't clear PBA because that can cause lost interrupts that
+	       * came in while we were cleaning/polling
+	       */
 	      (type << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT) |
 	      (itr << I40E_PFINT_DYN_CTLN_INTERVAL_SHIFT);
 
@@ -1983,7 +1985,7 @@ tx_only:
 		qval = rd32(hw, I40E_QINT_TQCTL(0)) |
 		       I40E_QINT_TQCTL_CAUSE_ENA_MASK;
 		wr32(hw, I40E_QINT_TQCTL(0), qval);
-		i40e_irq_dynamic_enable_icr0(vsi->back);
+		i40e_irq_dynamic_enable_icr0(vsi->back, false);
 	}
 	return 0;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 1635c7a..3e0d87e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -980,7 +980,7 @@ err_alloc:
 		i40e_free_vfs(pf);
 err_iov:
 	/* Re-enable interrupt 0. */
-	i40e_irq_dynamic_enable_icr0(pf);
+	i40e_irq_dynamic_enable_icr0(pf, false);
 	return ret;
 }
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 1c62cf5..6210424 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1248,7 +1248,9 @@ static u32 i40e_buildreg_itr(const int type, const u16 itr)
 	u32 val;
 
 	val = I40E_VFINT_DYN_CTLN1_INTENA_MASK |
-	      I40E_VFINT_DYN_CTLN1_CLEARPBA_MASK |
+	      /* Don't clear PBA because that can cause lost interrupts that
+	       * came in while we were cleaning/polling
+	       */
 	      (type << I40E_VFINT_DYN_CTLN1_ITR_INDX_SHIFT) |
 	      (itr << I40E_VFINT_DYN_CTLN1_INTERVAL_SHIFT);
 
-- 
cgit v0.10.2


From c2e245ab1e9a61e66217aafea66c7dc6481f12f0 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 13 Jan 2016 16:51:46 -0800
Subject: i40e/i40evf: try again after failure

This is the "Don't Give Up" patch.  Previously the
driver could fail an allocation, and then possibly stall
a queue forever, by never coming back to continue receiving
or allocating buffers.

With this patch, the driver will keep polling trying to allocate
receive buffers until it succeeds.  This should keep all receive
queues running even in the face of memory pressure.

Also update copyright year in file header.

Change-ID: I2b103d1ce95b9831288a7222c3343ffa1988b81b
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 353e5a0..8049206 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1195,8 +1195,10 @@ static inline void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val)
  * i40e_alloc_rx_buffers_ps - Replace used receive buffers; packet split
  * @rx_ring: ring to place buffers on
  * @cleaned_count: number of buffers to replace
+ *
+ * Returns true if any errors on allocation
  **/
-void i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
+bool i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 {
 	u16 i = rx_ring->next_to_use;
 	union i40e_rx_desc *rx_desc;
@@ -1204,7 +1206,7 @@ void i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 
 	/* do nothing if no valid netdev defined */
 	if (!rx_ring->netdev || !cleaned_count)
-		return;
+		return false;
 
 	while (cleaned_count--) {
 		rx_desc = I40E_RX_DESC(rx_ring, i);
@@ -1251,17 +1253,29 @@ void i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 			i = 0;
 	}
 
+	if (rx_ring->next_to_use != i)
+		i40e_release_rx_desc(rx_ring, i);
+
+	return false;
+
 no_buffers:
 	if (rx_ring->next_to_use != i)
 		i40e_release_rx_desc(rx_ring, i);
+
+	/* make sure to come back via polling to try again after
+	 * allocation failure
+	 */
+	return true;
 }
 
 /**
  * i40e_alloc_rx_buffers_1buf - Replace used receive buffers; single buffer
  * @rx_ring: ring to place buffers on
  * @cleaned_count: number of buffers to replace
+ *
+ * Returns true if any errors on allocation
  **/
-void i40e_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
+bool i40e_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 {
 	u16 i = rx_ring->next_to_use;
 	union i40e_rx_desc *rx_desc;
@@ -1270,7 +1284,7 @@ void i40e_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 
 	/* do nothing if no valid netdev defined */
 	if (!rx_ring->netdev || !cleaned_count)
-		return;
+		return false;
 
 	while (cleaned_count--) {
 		rx_desc = I40E_RX_DESC(rx_ring, i);
@@ -1297,6 +1311,8 @@ void i40e_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 			if (dma_mapping_error(rx_ring->dev, bi->dma)) {
 				rx_ring->rx_stats.alloc_buff_failed++;
 				bi->dma = 0;
+				dev_kfree_skb(bi->skb);
+				bi->skb = NULL;
 				goto no_buffers;
 			}
 		}
@@ -1308,9 +1324,19 @@ void i40e_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 			i = 0;
 	}
 
+	if (rx_ring->next_to_use != i)
+		i40e_release_rx_desc(rx_ring, i);
+
+	return false;
+
 no_buffers:
 	if (rx_ring->next_to_use != i)
 		i40e_release_rx_desc(rx_ring, i);
+
+	/* make sure to come back via polling to try again after
+	 * allocation failure
+	 */
+	return true;
 }
 
 /**
@@ -1494,7 +1520,7 @@ static inline void i40e_rx_hash(struct i40e_ring *ring,
  *
  * Returns true if there's any budget left (e.g. the clean is finished)
  **/
-static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
+static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 {
 	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 	u16 rx_packet_len, rx_header_len, rx_sph, rx_hbo;
@@ -1504,6 +1530,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 	u16 i = rx_ring->next_to_clean;
 	union i40e_rx_desc *rx_desc;
 	u32 rx_error, rx_status;
+	bool failure = false;
 	u8 rx_ptype;
 	u64 qword;
 
@@ -1516,7 +1543,9 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 		u16 vlan_tag;
 		/* return some buffers to hardware, one at a time is too slow */
 		if (cleaned_count >= I40E_RX_BUFFER_WRITE) {
-			i40e_alloc_rx_buffers_ps(rx_ring, cleaned_count);
+			failure = failure ||
+				  i40e_alloc_rx_buffers_ps(rx_ring,
+							   cleaned_count);
 			cleaned_count = 0;
 		}
 
@@ -1546,6 +1575,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 							rx_ring->rx_hdr_len);
 			if (!skb) {
 				rx_ring->rx_stats.alloc_buff_failed++;
+				failure = true;
 				break;
 			}
 
@@ -1675,7 +1705,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 	rx_ring->q_vector->rx.total_packets += total_rx_packets;
 	rx_ring->q_vector->rx.total_bytes += total_rx_bytes;
 
-	return total_rx_packets;
+	return failure ? budget : total_rx_packets;
 }
 
 /**
@@ -1693,6 +1723,7 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 	union i40e_rx_desc *rx_desc;
 	u32 rx_error, rx_status;
 	u16 rx_packet_len;
+	bool failure = false;
 	u8 rx_ptype;
 	u64 qword;
 	u16 i;
@@ -1703,7 +1734,9 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 		u16 vlan_tag;
 		/* return some buffers to hardware, one at a time is too slow */
 		if (cleaned_count >= I40E_RX_BUFFER_WRITE) {
-			i40e_alloc_rx_buffers_1buf(rx_ring, cleaned_count);
+			failure = failure ||
+				  i40e_alloc_rx_buffers_1buf(rx_ring,
+							     cleaned_count);
 			cleaned_count = 0;
 		}
 
@@ -1802,7 +1835,7 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 	rx_ring->q_vector->rx.total_packets += total_rx_packets;
 	rx_ring->q_vector->rx.total_bytes += total_rx_bytes;
 
-	return total_rx_packets;
+	return failure ? budget : total_rx_packets;
 }
 
 static u32 i40e_buildreg_itr(const int type, const u16 itr)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 3f081e2..5c73f3d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -316,8 +316,8 @@ struct i40e_ring_container {
 #define i40e_for_each_ring(pos, head) \
 	for (pos = (head).ring; pos != NULL; pos = pos->next)
 
-void i40e_alloc_rx_buffers_ps(struct i40e_ring *rxr, u16 cleaned_count);
-void i40e_alloc_rx_buffers_1buf(struct i40e_ring *rxr, u16 cleaned_count);
+bool i40e_alloc_rx_buffers_ps(struct i40e_ring *rxr, u16 cleaned_count);
+bool i40e_alloc_rx_buffers_1buf(struct i40e_ring *rxr, u16 cleaned_count);
 void i40e_alloc_rx_headers(struct i40e_ring *rxr);
 netdev_tx_t i40e_lan_xmit_frame(struct sk_buff *skb, struct net_device *netdev);
 void i40e_clean_tx_ring(struct i40e_ring *tx_ring);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 6210424..616daae 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -667,8 +667,10 @@ static inline void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val)
  * i40evf_alloc_rx_buffers_ps - Replace used receive buffers; packet split
  * @rx_ring: ring to place buffers on
  * @cleaned_count: number of buffers to replace
+ *
+ * Returns true if any errors on allocation
  **/
-void i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
+bool i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 {
 	u16 i = rx_ring->next_to_use;
 	union i40e_rx_desc *rx_desc;
@@ -676,7 +678,7 @@ void i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 
 	/* do nothing if no valid netdev defined */
 	if (!rx_ring->netdev || !cleaned_count)
-		return;
+		return false;
 
 	while (cleaned_count--) {
 		rx_desc = I40E_RX_DESC(rx_ring, i);
@@ -723,17 +725,29 @@ void i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 			i = 0;
 	}
 
+	if (rx_ring->next_to_use != i)
+		i40e_release_rx_desc(rx_ring, i);
+
+	return false;
+
 no_buffers:
 	if (rx_ring->next_to_use != i)
 		i40e_release_rx_desc(rx_ring, i);
+
+	/* make sure to come back via polling to try again after
+	 * allocation failure
+	 */
+	return true;
 }
 
 /**
  * i40evf_alloc_rx_buffers_1buf - Replace used receive buffers; single buffer
  * @rx_ring: ring to place buffers on
  * @cleaned_count: number of buffers to replace
+ *
+ * Returns true if any errors on allocation
  **/
-void i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
+bool i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 {
 	u16 i = rx_ring->next_to_use;
 	union i40e_rx_desc *rx_desc;
@@ -742,7 +756,7 @@ void i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 
 	/* do nothing if no valid netdev defined */
 	if (!rx_ring->netdev || !cleaned_count)
-		return;
+		return false;
 
 	while (cleaned_count--) {
 		rx_desc = I40E_RX_DESC(rx_ring, i);
@@ -769,6 +783,8 @@ void i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 			if (dma_mapping_error(rx_ring->dev, bi->dma)) {
 				rx_ring->rx_stats.alloc_buff_failed++;
 				bi->dma = 0;
+				dev_kfree_skb(bi->skb);
+				bi->skb = NULL;
 				goto no_buffers;
 			}
 		}
@@ -780,9 +796,19 @@ void i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 			i = 0;
 	}
 
+	if (rx_ring->next_to_use != i)
+		i40e_release_rx_desc(rx_ring, i);
+
+	return false;
+
 no_buffers:
 	if (rx_ring->next_to_use != i)
 		i40e_release_rx_desc(rx_ring, i);
+
+	/* make sure to come back via polling to try again after
+	 * allocation failure
+	 */
+	return true;
 }
 
 /**
@@ -965,7 +991,7 @@ static inline void i40e_rx_hash(struct i40e_ring *ring,
  *
  * Returns true if there's any budget left (e.g. the clean is finished)
  **/
-static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
+static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 {
 	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 	u16 rx_packet_len, rx_header_len, rx_sph, rx_hbo;
@@ -975,6 +1001,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 	u16 i = rx_ring->next_to_clean;
 	union i40e_rx_desc *rx_desc;
 	u32 rx_error, rx_status;
+	bool failure = false;
 	u8 rx_ptype;
 	u64 qword;
 
@@ -984,7 +1011,9 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 		u16 vlan_tag;
 		/* return some buffers to hardware, one at a time is too slow */
 		if (cleaned_count >= I40E_RX_BUFFER_WRITE) {
-			i40evf_alloc_rx_buffers_ps(rx_ring, cleaned_count);
+			failure = failure ||
+				  i40evf_alloc_rx_buffers_ps(rx_ring,
+							     cleaned_count);
 			cleaned_count = 0;
 		}
 
@@ -1009,6 +1038,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 							rx_ring->rx_hdr_len);
 			if (!skb) {
 				rx_ring->rx_stats.alloc_buff_failed++;
+				failure = true;
 				break;
 			}
 
@@ -1131,7 +1161,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 	rx_ring->q_vector->rx.total_packets += total_rx_packets;
 	rx_ring->q_vector->rx.total_bytes += total_rx_bytes;
 
-	return total_rx_packets;
+	return failure ? budget : total_rx_packets;
 }
 
 /**
@@ -1149,6 +1179,7 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 	union i40e_rx_desc *rx_desc;
 	u32 rx_error, rx_status;
 	u16 rx_packet_len;
+	bool failure = false;
 	u8 rx_ptype;
 	u64 qword;
 	u16 i;
@@ -1159,7 +1190,9 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 		u16 vlan_tag;
 		/* return some buffers to hardware, one at a time is too slow */
 		if (cleaned_count >= I40E_RX_BUFFER_WRITE) {
-			i40evf_alloc_rx_buffers_1buf(rx_ring, cleaned_count);
+			failure = failure ||
+				  i40evf_alloc_rx_buffers_1buf(rx_ring,
+							       cleaned_count);
 			cleaned_count = 0;
 		}
 
@@ -1240,7 +1273,7 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 	rx_ring->q_vector->rx.total_packets += total_rx_packets;
 	rx_ring->q_vector->rx.total_bytes += total_rx_bytes;
 
-	return total_rx_packets;
+	return failure ? budget : total_rx_packets;
 }
 
 static u32 i40e_buildreg_itr(const int type, const u16 itr)
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index da701c5..d8071ba 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -313,8 +313,8 @@ struct i40e_ring_container {
 #define i40e_for_each_ring(pos, head) \
 	for (pos = (head).ring; pos != NULL; pos = pos->next)
 
-void i40evf_alloc_rx_buffers_ps(struct i40e_ring *rxr, u16 cleaned_count);
-void i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rxr, u16 cleaned_count);
+bool i40evf_alloc_rx_buffers_ps(struct i40e_ring *rxr, u16 cleaned_count);
+bool i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rxr, u16 cleaned_count);
 void i40evf_alloc_rx_headers(struct i40e_ring *rxr);
 netdev_tx_t i40evf_xmit_frame(struct sk_buff *skb, struct net_device *netdev);
 void i40evf_clean_tx_ring(struct i40e_ring *tx_ring);
-- 
cgit v0.10.2


From 13cb3e9d71d5823a3169b10b428ce717434c657e Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 13 Jan 2016 16:51:47 -0800
Subject: i40e: dump descriptor indexes in hex

The debugging helpers for showing descriptor rings were
dumping the indexes in decimal and the offsets in hex.

Put everything in hex and at least be consistent.

Also update copyright year in file header.

Change-ID: Ia35a21411a2ddb713772dffb4e8718889fcfc895
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 3948587..fcae3c8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -825,20 +825,20 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
 			if (!is_rx_ring) {
 				txd = I40E_TX_DESC(ring, i);
 				dev_info(&pf->pdev->dev,
-					 "   d[%03i] = 0x%016llx 0x%016llx\n",
+					 "   d[%03x] = 0x%016llx 0x%016llx\n",
 					 i, txd->buffer_addr,
 					 txd->cmd_type_offset_bsz);
 			} else if (sizeof(union i40e_rx_desc) ==
 				   sizeof(union i40e_16byte_rx_desc)) {
 				rxd = I40E_RX_DESC(ring, i);
 				dev_info(&pf->pdev->dev,
-					 "   d[%03i] = 0x%016llx 0x%016llx\n",
+					 "   d[%03x] = 0x%016llx 0x%016llx\n",
 					 i, rxd->read.pkt_addr,
 					 rxd->read.hdr_addr);
 			} else {
 				rxd = I40E_RX_DESC(ring, i);
 				dev_info(&pf->pdev->dev,
-					 "   d[%03i] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n",
+					 "   d[%03x] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n",
 					 i, rxd->read.pkt_addr,
 					 rxd->read.hdr_addr,
 					 rxd->read.rsvd1, rxd->read.rsvd2);
@@ -853,20 +853,20 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n,
 		if (!is_rx_ring) {
 			txd = I40E_TX_DESC(ring, desc_n);
 			dev_info(&pf->pdev->dev,
-				 "vsi = %02i tx ring = %02i d[%03i] = 0x%016llx 0x%016llx\n",
+				 "vsi = %02i tx ring = %02i d[%03x] = 0x%016llx 0x%016llx\n",
 				 vsi_seid, ring_id, desc_n,
 				 txd->buffer_addr, txd->cmd_type_offset_bsz);
 		} else if (sizeof(union i40e_rx_desc) ==
 			   sizeof(union i40e_16byte_rx_desc)) {
 			rxd = I40E_RX_DESC(ring, desc_n);
 			dev_info(&pf->pdev->dev,
-				 "vsi = %02i rx ring = %02i d[%03i] = 0x%016llx 0x%016llx\n",
+				 "vsi = %02i rx ring = %02i d[%03x] = 0x%016llx 0x%016llx\n",
 				 vsi_seid, ring_id, desc_n,
 				 rxd->read.pkt_addr, rxd->read.hdr_addr);
 		} else {
 			rxd = I40E_RX_DESC(ring, desc_n);
 			dev_info(&pf->pdev->dev,
-				 "vsi = %02i rx ring = %02i d[%03i] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n",
+				 "vsi = %02i rx ring = %02i d[%03x] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n",
 				 vsi_seid, ring_id, desc_n,
 				 rxd->read.pkt_addr, rxd->read.hdr_addr,
 				 rxd->read.rsvd1, rxd->read.rsvd2);
-- 
cgit v0.10.2


From dd1a5df872d7889b37b5ce3058a9f2e20d4d8b56 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 13 Jan 2016 16:51:48 -0800
Subject: i40e/i40evf: use __GFP_NOWARN

The i40e and i40evf drivers now cleanly handle allocation
failures and can avoid kernel log spew from the memory allocator
when allocations fail, so set __GFP_NOWARN on Rx buffer alloc.

Change-ID: Ic9e1b83c495e2a3ef6b069ba7fb6e52ce134cd23
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 8049206..baaf093 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1292,8 +1292,10 @@ bool i40e_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 		skb = bi->skb;
 
 		if (!skb) {
-			skb = netdev_alloc_skb_ip_align(rx_ring->netdev,
-							rx_ring->rx_buf_len);
+			skb = __netdev_alloc_skb_ip_align(rx_ring->netdev,
+							  rx_ring->rx_buf_len,
+							  GFP_ATOMIC |
+							  __GFP_NOWARN);
 			if (!skb) {
 				rx_ring->rx_stats.alloc_buff_failed++;
 				goto no_buffers;
@@ -1571,8 +1573,10 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 		rx_bi = &rx_ring->rx_bi[i];
 		skb = rx_bi->skb;
 		if (likely(!skb)) {
-			skb = netdev_alloc_skb_ip_align(rx_ring->netdev,
-							rx_ring->rx_hdr_len);
+			skb = __netdev_alloc_skb_ip_align(rx_ring->netdev,
+							  rx_ring->rx_hdr_len,
+							  GFP_ATOMIC |
+							  __GFP_NOWARN);
 			if (!skb) {
 				rx_ring->rx_stats.alloc_buff_failed++;
 				failure = true;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 616daae..1dbdcf8 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -764,8 +764,10 @@ bool i40evf_alloc_rx_buffers_1buf(struct i40e_ring *rx_ring, u16 cleaned_count)
 		skb = bi->skb;
 
 		if (!skb) {
-			skb = netdev_alloc_skb_ip_align(rx_ring->netdev,
-							rx_ring->rx_buf_len);
+			skb = __netdev_alloc_skb_ip_align(rx_ring->netdev,
+							  rx_ring->rx_buf_len,
+							  GFP_ATOMIC |
+							  __GFP_NOWARN);
 			if (!skb) {
 				rx_ring->rx_stats.alloc_buff_failed++;
 				goto no_buffers;
@@ -1034,8 +1036,10 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 		rx_bi = &rx_ring->rx_bi[i];
 		skb = rx_bi->skb;
 		if (likely(!skb)) {
-			skb = netdev_alloc_skb_ip_align(rx_ring->netdev,
-							rx_ring->rx_hdr_len);
+			skb = __netdev_alloc_skb_ip_align(rx_ring->netdev,
+							  rx_ring->rx_hdr_len,
+							  GFP_ATOMIC |
+							  __GFP_NOWARN);
 			if (!skb) {
 				rx_ring->rx_stats.alloc_buff_failed++;
 				failure = true;
-- 
cgit v0.10.2


From f16704e5e8aed1dfed4084c56dde17006c2e81f1 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 13 Jan 2016 16:51:49 -0800
Subject: i40e/i40evf: use pages correctly in Rx

Refactor the packet split Rx code to properly use half-pages for
receives. The previous code was doing way more mapping and unmapping
than it needed to, and wasn't properly using half-pages.

Increment the page use count each time we give a half-page to an skb,
knowing that the stack will probably process and release the page before
we need it again. Only free and reallocate pages if the count shows that
both half-pages are in use. Add counters to track reallocations and page
reuse.

Change-ID: I534b299196036b64be82b4861a0a4036310a8f22
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index fcae3c8..bdac691 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -536,6 +536,11 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 rx_ring->rx_stats.alloc_page_failed,
 			 rx_ring->rx_stats.alloc_buff_failed);
 		dev_info(&pf->pdev->dev,
+			 "    rx_rings[%i]: rx_stats: realloc_count = %lld, page_reuse_count = %lld\n",
+			 i,
+			 rx_ring->rx_stats.realloc_count,
+			 rx_ring->rx_stats.page_reuse_count);
+		dev_info(&pf->pdev->dev,
 			 "    rx_rings[%i]: size = %i, dma = 0x%08lx\n",
 			 i, rx_ring->size,
 			 (unsigned long int)rx_ring->dma);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index baaf093..1abef01 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1060,7 +1060,7 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring)
 			if (rx_bi->page_dma) {
 				dma_unmap_page(dev,
 					       rx_bi->page_dma,
-					       PAGE_SIZE / 2,
+					       PAGE_SIZE,
 					       DMA_FROM_DEVICE);
 				rx_bi->page_dma = 0;
 			}
@@ -1203,6 +1203,7 @@ bool i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 	u16 i = rx_ring->next_to_use;
 	union i40e_rx_desc *rx_desc;
 	struct i40e_rx_buffer *bi;
+	const int current_node = numa_node_id();
 
 	/* do nothing if no valid netdev defined */
 	if (!rx_ring->netdev || !cleaned_count)
@@ -1214,39 +1215,50 @@ bool i40e_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 
 		if (bi->skb) /* desc is in use */
 			goto no_buffers;
+
+	/* If we've been moved to a different NUMA node, release the
+	 * page so we can get a new one on the current node.
+	 */
+		if (bi->page &&  page_to_nid(bi->page) != current_node) {
+			dma_unmap_page(rx_ring->dev,
+				       bi->page_dma,
+				       PAGE_SIZE,
+				       DMA_FROM_DEVICE);
+			__free_page(bi->page);
+			bi->page = NULL;
+			bi->page_dma = 0;
+			rx_ring->rx_stats.realloc_count++;
+		} else if (bi->page) {
+			rx_ring->rx_stats.page_reuse_count++;
+		}
+
 		if (!bi->page) {
 			bi->page = alloc_page(GFP_ATOMIC);
 			if (!bi->page) {
 				rx_ring->rx_stats.alloc_page_failed++;
 				goto no_buffers;
 			}
-		}
-
-		if (!bi->page_dma) {
-			/* use a half page if we're re-using */
-			bi->page_offset ^= PAGE_SIZE / 2;
 			bi->page_dma = dma_map_page(rx_ring->dev,
 						    bi->page,
-						    bi->page_offset,
-						    PAGE_SIZE / 2,
+						    0,
+						    PAGE_SIZE,
 						    DMA_FROM_DEVICE);
-			if (dma_mapping_error(rx_ring->dev,
-					      bi->page_dma)) {
+			if (dma_mapping_error(rx_ring->dev, bi->page_dma)) {
 				rx_ring->rx_stats.alloc_page_failed++;
+				__free_page(bi->page);
+				bi->page = NULL;
 				bi->page_dma = 0;
+				bi->page_offset = 0;
 				goto no_buffers;
 			}
+			bi->page_offset = 0;
 		}
 
-		dma_sync_single_range_for_device(rx_ring->dev,
-						 rx_ring->rx_bi[0].dma,
-						 i * rx_ring->rx_hdr_len,
-						 rx_ring->rx_hdr_len,
-						 DMA_FROM_DEVICE);
 		/* Refresh the desc even if buffer_addrs didn't change
 		 * because each write-back erases this info.
 		 */
-		rx_desc->read.pkt_addr = cpu_to_le64(bi->page_dma);
+		rx_desc->read.pkt_addr =
+				cpu_to_le64(bi->page_dma + bi->page_offset);
 		rx_desc->read.hdr_addr = cpu_to_le64(bi->dma);
 		i++;
 		if (i == rx_ring->count)
@@ -1527,7 +1539,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 	u16 rx_packet_len, rx_header_len, rx_sph, rx_hbo;
 	u16 cleaned_count = I40E_DESC_UNUSED(rx_ring);
-	const int current_node = numa_mem_id();
 	struct i40e_vsi *vsi = rx_ring->vsi;
 	u16 i = rx_ring->next_to_clean;
 	union i40e_rx_desc *rx_desc;
@@ -1535,6 +1546,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 	bool failure = false;
 	u8 rx_ptype;
 	u64 qword;
+	u32 copysize;
 
 	if (budget <= 0)
 		return 0;
@@ -1565,6 +1577,12 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 		 * DD bit is set.
 		 */
 		dma_rmb();
+		/* sync header buffer for reading */
+		dma_sync_single_range_for_cpu(rx_ring->dev,
+					      rx_ring->rx_bi[0].dma,
+					      i * rx_ring->rx_hdr_len,
+					      rx_ring->rx_hdr_len,
+					      DMA_FROM_DEVICE);
 		if (i40e_rx_is_programming_status(qword)) {
 			i40e_clean_programming_status(rx_ring, rx_desc);
 			I40E_RX_INCREMENT(rx_ring, i);
@@ -1606,9 +1624,16 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 
 		rx_ptype = (qword & I40E_RXD_QW1_PTYPE_MASK) >>
 			   I40E_RXD_QW1_PTYPE_SHIFT;
-		prefetch(rx_bi->page);
+		/* sync half-page for reading */
+		dma_sync_single_range_for_cpu(rx_ring->dev,
+					      rx_bi->page_dma,
+					      rx_bi->page_offset,
+					      PAGE_SIZE / 2,
+					      DMA_FROM_DEVICE);
+		prefetch(page_address(rx_bi->page) + rx_bi->page_offset);
 		rx_bi->skb = NULL;
 		cleaned_count++;
+		copysize = 0;
 		if (rx_hbo || rx_sph) {
 			int len;
 
@@ -1619,38 +1644,45 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 			memcpy(__skb_put(skb, len), rx_bi->hdr_buf, len);
 		} else if (skb->len == 0) {
 			int len;
+			unsigned char *va = page_address(rx_bi->page) +
+					    rx_bi->page_offset;
 
-			len = (rx_packet_len > skb_headlen(skb) ?
-				skb_headlen(skb) : rx_packet_len);
-			memcpy(__skb_put(skb, len),
-			       rx_bi->page + rx_bi->page_offset,
-			       len);
-			rx_bi->page_offset += len;
+			len = min(rx_packet_len, rx_ring->rx_hdr_len);
+			memcpy(__skb_put(skb, len), va, len);
+			copysize = len;
 			rx_packet_len -= len;
 		}
-
 		/* Get the rest of the data if this was a header split */
 		if (rx_packet_len) {
-			skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags,
-					   rx_bi->page,
-					   rx_bi->page_offset,
-					   rx_packet_len);
-
-			skb->len += rx_packet_len;
-			skb->data_len += rx_packet_len;
-			skb->truesize += rx_packet_len;
-
-			if ((page_count(rx_bi->page) == 1) &&
-			    (page_to_nid(rx_bi->page) == current_node))
-				get_page(rx_bi->page);
-			else
+			skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+					rx_bi->page,
+					rx_bi->page_offset + copysize,
+					rx_packet_len, I40E_RXBUFFER_2048);
+
+			get_page(rx_bi->page);
+			/* switch to the other half-page here; the allocation
+			 * code programs the right addr into HW. If we haven't
+			 * used this half-page, the address won't be changed,
+			 * and HW can just use it next time through.
+			 */
+			rx_bi->page_offset ^= PAGE_SIZE / 2;
+			/* If the page count is more than 2, then both halves
+			 * of the page are used and we need to free it. Do it
+			 * here instead of in the alloc code. Otherwise one
+			 * of the half-pages might be released between now and
+			 * then, and we wouldn't know which one to use.
+			 */
+			if (page_count(rx_bi->page) > 2) {
+				dma_unmap_page(rx_ring->dev,
+					       rx_bi->page_dma,
+					       PAGE_SIZE,
+					       DMA_FROM_DEVICE);
+				__free_page(rx_bi->page);
 				rx_bi->page = NULL;
+				rx_bi->page_dma = 0;
+				rx_ring->rx_stats.realloc_count++;
+			}
 
-			dma_unmap_page(rx_ring->dev,
-				       rx_bi->page_dma,
-				       PAGE_SIZE / 2,
-				       DMA_FROM_DEVICE);
-			rx_bi->page_dma = 0;
 		}
 		I40E_RX_INCREMENT(rx_ring, i);
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 5c73f3d..3b8d147 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -209,6 +209,8 @@ struct i40e_rx_queue_stats {
 	u64 non_eop_descs;
 	u64 alloc_page_failed;
 	u64 alloc_buff_failed;
+	u64 page_reuse_count;
+	u64 realloc_count;
 };
 
 enum i40e_ring_state_t {
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 1dbdcf8..6f739a7 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -532,7 +532,7 @@ void i40evf_clean_rx_ring(struct i40e_ring *rx_ring)
 			if (rx_bi->page_dma) {
 				dma_unmap_page(dev,
 					       rx_bi->page_dma,
-					       PAGE_SIZE / 2,
+					       PAGE_SIZE,
 					       DMA_FROM_DEVICE);
 				rx_bi->page_dma = 0;
 			}
@@ -675,6 +675,7 @@ bool i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 	u16 i = rx_ring->next_to_use;
 	union i40e_rx_desc *rx_desc;
 	struct i40e_rx_buffer *bi;
+	const int current_node = numa_node_id();
 
 	/* do nothing if no valid netdev defined */
 	if (!rx_ring->netdev || !cleaned_count)
@@ -686,39 +687,50 @@ bool i40evf_alloc_rx_buffers_ps(struct i40e_ring *rx_ring, u16 cleaned_count)
 
 		if (bi->skb) /* desc is in use */
 			goto no_buffers;
+
+	/* If we've been moved to a different NUMA node, release the
+	 * page so we can get a new one on the current node.
+	 */
+		if (bi->page &&  page_to_nid(bi->page) != current_node) {
+			dma_unmap_page(rx_ring->dev,
+				       bi->page_dma,
+				       PAGE_SIZE,
+				       DMA_FROM_DEVICE);
+			__free_page(bi->page);
+			bi->page = NULL;
+			bi->page_dma = 0;
+			rx_ring->rx_stats.realloc_count++;
+		} else if (bi->page) {
+			rx_ring->rx_stats.page_reuse_count++;
+		}
+
 		if (!bi->page) {
 			bi->page = alloc_page(GFP_ATOMIC);
 			if (!bi->page) {
 				rx_ring->rx_stats.alloc_page_failed++;
 				goto no_buffers;
 			}
-		}
-
-		if (!bi->page_dma) {
-			/* use a half page if we're re-using */
-			bi->page_offset ^= PAGE_SIZE / 2;
 			bi->page_dma = dma_map_page(rx_ring->dev,
 						    bi->page,
-						    bi->page_offset,
-						    PAGE_SIZE / 2,
+						    0,
+						    PAGE_SIZE,
 						    DMA_FROM_DEVICE);
-			if (dma_mapping_error(rx_ring->dev,
-					      bi->page_dma)) {
+			if (dma_mapping_error(rx_ring->dev, bi->page_dma)) {
 				rx_ring->rx_stats.alloc_page_failed++;
+				__free_page(bi->page);
+				bi->page = NULL;
 				bi->page_dma = 0;
+				bi->page_offset = 0;
 				goto no_buffers;
 			}
+			bi->page_offset = 0;
 		}
 
-		dma_sync_single_range_for_device(rx_ring->dev,
-						 rx_ring->rx_bi[0].dma,
-						 i * rx_ring->rx_hdr_len,
-						 rx_ring->rx_hdr_len,
-						 DMA_FROM_DEVICE);
 		/* Refresh the desc even if buffer_addrs didn't change
 		 * because each write-back erases this info.
 		 */
-		rx_desc->read.pkt_addr = cpu_to_le64(bi->page_dma);
+		rx_desc->read.pkt_addr =
+				cpu_to_le64(bi->page_dma + bi->page_offset);
 		rx_desc->read.hdr_addr = cpu_to_le64(bi->dma);
 		i++;
 		if (i == rx_ring->count)
@@ -998,7 +1010,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 	u16 rx_packet_len, rx_header_len, rx_sph, rx_hbo;
 	u16 cleaned_count = I40E_DESC_UNUSED(rx_ring);
-	const int current_node = numa_mem_id();
 	struct i40e_vsi *vsi = rx_ring->vsi;
 	u16 i = rx_ring->next_to_clean;
 	union i40e_rx_desc *rx_desc;
@@ -1006,6 +1017,7 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 	bool failure = false;
 	u8 rx_ptype;
 	u64 qword;
+	u32 copysize;
 
 	do {
 		struct i40e_rx_buffer *rx_bi;
@@ -1033,6 +1045,12 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 		 * DD bit is set.
 		 */
 		dma_rmb();
+		/* sync header buffer for reading */
+		dma_sync_single_range_for_cpu(rx_ring->dev,
+					      rx_ring->rx_bi[0].dma,
+					      i * rx_ring->rx_hdr_len,
+					      rx_ring->rx_hdr_len,
+					      DMA_FROM_DEVICE);
 		rx_bi = &rx_ring->rx_bi[i];
 		skb = rx_bi->skb;
 		if (likely(!skb)) {
@@ -1069,9 +1087,16 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 
 		rx_ptype = (qword & I40E_RXD_QW1_PTYPE_MASK) >>
 			   I40E_RXD_QW1_PTYPE_SHIFT;
-		prefetch(rx_bi->page);
+		/* sync half-page for reading */
+		dma_sync_single_range_for_cpu(rx_ring->dev,
+					      rx_bi->page_dma,
+					      rx_bi->page_offset,
+					      PAGE_SIZE / 2,
+					      DMA_FROM_DEVICE);
+		prefetch(page_address(rx_bi->page) + rx_bi->page_offset);
 		rx_bi->skb = NULL;
 		cleaned_count++;
+		copysize = 0;
 		if (rx_hbo || rx_sph) {
 			int len;
 
@@ -1082,38 +1107,45 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 			memcpy(__skb_put(skb, len), rx_bi->hdr_buf, len);
 		} else if (skb->len == 0) {
 			int len;
+			unsigned char *va = page_address(rx_bi->page) +
+					    rx_bi->page_offset;
 
-			len = (rx_packet_len > skb_headlen(skb) ?
-				skb_headlen(skb) : rx_packet_len);
-			memcpy(__skb_put(skb, len),
-			       rx_bi->page + rx_bi->page_offset,
-			       len);
-			rx_bi->page_offset += len;
+			len = min(rx_packet_len, rx_ring->rx_hdr_len);
+			memcpy(__skb_put(skb, len), va, len);
+			copysize = len;
 			rx_packet_len -= len;
 		}
-
 		/* Get the rest of the data if this was a header split */
 		if (rx_packet_len) {
-			skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags,
-					   rx_bi->page,
-					   rx_bi->page_offset,
-					   rx_packet_len);
-
-			skb->len += rx_packet_len;
-			skb->data_len += rx_packet_len;
-			skb->truesize += rx_packet_len;
-
-			if ((page_count(rx_bi->page) == 1) &&
-			    (page_to_nid(rx_bi->page) == current_node))
-				get_page(rx_bi->page);
-			else
+			skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+					rx_bi->page,
+					rx_bi->page_offset + copysize,
+					rx_packet_len, I40E_RXBUFFER_2048);
+
+			get_page(rx_bi->page);
+			/* switch to the other half-page here; the allocation
+			 * code programs the right addr into HW. If we haven't
+			 * used this half-page, the address won't be changed,
+			 * and HW can just use it next time through.
+			 */
+			rx_bi->page_offset ^= PAGE_SIZE / 2;
+			/* If the page count is more than 2, then both halves
+			 * of the page are used and we need to free it. Do it
+			 * here instead of in the alloc code. Otherwise one
+			 * of the half-pages might be released between now and
+			 * then, and we wouldn't know which one to use.
+			 */
+			if (page_count(rx_bi->page) > 2) {
+				dma_unmap_page(rx_ring->dev,
+					       rx_bi->page_dma,
+					       PAGE_SIZE,
+					       DMA_FROM_DEVICE);
+				__free_page(rx_bi->page);
 				rx_bi->page = NULL;
+				rx_bi->page_dma = 0;
+				rx_ring->rx_stats.realloc_count++;
+			}
 
-			dma_unmap_page(rx_ring->dev,
-				       rx_bi->page_dma,
-				       PAGE_SIZE / 2,
-				       DMA_FROM_DEVICE);
-			rx_bi->page_dma = 0;
 		}
 		I40E_RX_INCREMENT(rx_ring, i);
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index d8071ba..5f03c44 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -208,6 +208,8 @@ struct i40e_rx_queue_stats {
 	u64 non_eop_descs;
 	u64 alloc_page_failed;
 	u64 alloc_buff_failed;
+	u64 page_reuse_count;
+	u64 realloc_count;
 };
 
 enum i40e_ring_state_t {
-- 
cgit v0.10.2


From 1a36d7fadd1477761dc8b4fc8910d1defc31fad5 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 13 Jan 2016 16:51:50 -0800
Subject: i40e/i40evf: use logical operators, not bitwise

Mr. Spock would certainly raise an eyebrow to see us using bitwise
operators, when we should clearly be relying on logic. Fascinating.

Change-ID: Ie338010c016f93e9faa2002c07c90b15134b7477
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 1abef01..0ffa9a8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1996,7 +1996,8 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 	 * budget and be more aggressive about cleaning up the Tx descriptors.
 	 */
 	i40e_for_each_ring(ring, q_vector->tx) {
-		clean_complete &= i40e_clean_tx_irq(ring, vsi->work_limit);
+		clean_complete = clean_complete &&
+				 i40e_clean_tx_irq(ring, vsi->work_limit);
 		arm_wb = arm_wb || ring->arm_wb;
 		ring->arm_wb = false;
 	}
@@ -2020,7 +2021,7 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 
 		work_done += cleaned;
 		/* if we didn't clean as many as budgeted, we must be done */
-		clean_complete &= (budget_per_ring != cleaned);
+		clean_complete = clean_complete && (budget_per_ring > cleaned);
 	}
 
 	/* If work not completed, return budget and polling will return */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 6f739a7..76bad75 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1432,7 +1432,8 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget)
 	 * budget and be more aggressive about cleaning up the Tx descriptors.
 	 */
 	i40e_for_each_ring(ring, q_vector->tx) {
-		clean_complete &= i40e_clean_tx_irq(ring, vsi->work_limit);
+		clean_complete = clean_complete &&
+				 i40e_clean_tx_irq(ring, vsi->work_limit);
 		arm_wb = arm_wb || ring->arm_wb;
 		ring->arm_wb = false;
 	}
@@ -1456,7 +1457,7 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget)
 
 		work_done += cleaned;
 		/* if we didn't clean as many as budgeted, we must be done */
-		clean_complete &= (budget_per_ring != cleaned);
+		clean_complete = clean_complete && (budget_per_ring > cleaned);
 	}
 
 	/* If work not completed, return budget and polling will return */
-- 
cgit v0.10.2


From 4668607aa30b3879312823a0ddbcd15077644f4e Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 13 Jan 2016 16:51:51 -0800
Subject: i40e: properly show packet split status in debugfs

Get rid of the unused hsplit field in the ring struct and use the
existing macro to detect packet split enablement. This allows debugfs
dumps of the VSI to properly show which Rx routine is in use.

Change-ID: Ic4e9589e6a788ab196ed0850703f704e30c03781
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index bdac691..34da53b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -521,7 +521,7 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 rx_ring->dtype);
 		dev_info(&pf->pdev->dev,
 			 "    rx_rings[%i]: hsplit = %d, next_to_use = %d, next_to_clean = %d, ring_active = %i\n",
-			 i, rx_ring->hsplit,
+			 i, ring_is_ps_enabled(rx_ring),
 			 rx_ring->next_to_use,
 			 rx_ring->next_to_clean,
 			 rx_ring->ring_active);
@@ -572,8 +572,8 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 "    tx_rings[%i]: dtype = %d\n",
 			 i, tx_ring->dtype);
 		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: hsplit = %d, next_to_use = %d, next_to_clean = %d, ring_active = %i\n",
-			 i, tx_ring->hsplit,
+			 "    tx_rings[%i]: next_to_use = %d, next_to_clean = %d, ring_active = %i\n",
+			 i,
 			 tx_ring->next_to_use,
 			 tx_ring->next_to_clean,
 			 tx_ring->ring_active);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 3b8d147..ae22c4e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -256,7 +256,6 @@ struct i40e_ring {
 #define I40E_RX_DTYPE_NO_SPLIT      0
 #define I40E_RX_DTYPE_HEADER_SPLIT  1
 #define I40E_RX_DTYPE_SPLIT_ALWAYS  2
-	u8  hsplit;
 #define I40E_RX_SPLIT_L2      0x1
 #define I40E_RX_SPLIT_IP      0x2
 #define I40E_RX_SPLIT_TCP_UDP 0x4
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 5f03c44..5467fcdf 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -255,7 +255,6 @@ struct i40e_ring {
 #define I40E_RX_DTYPE_NO_SPLIT      0
 #define I40E_RX_DTYPE_HEADER_SPLIT  1
 #define I40E_RX_DTYPE_SPLIT_ALWAYS  2
-	u8  hsplit;
 #define I40E_RX_SPLIT_L2      0x1
 #define I40E_RX_SPLIT_IP      0x2
 #define I40E_RX_SPLIT_TCP_UDP 0x4
-- 
cgit v0.10.2


From c24215c04fe7bf9f5678abe6c496d774a6b8663a Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 13 Jan 2016 16:51:52 -0800
Subject: i40e/i40evf: Bump version

Bump version to i40e-1.4.13 and i40evf-1.4.9

Change-ID: I9db37f9d4899141c3e5455dfb456d45465b8c035
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 8bc848f..8d41c6c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -46,7 +46,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 12
+#define DRV_VERSION_BUILD 13
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index faa1bca..1d81d57 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -38,7 +38,7 @@ static const char i40evf_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 8
+#define DRV_VERSION_BUILD 9
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD) \
-- 
cgit v0.10.2


From 2d9e9b0d05876c9e6b52cfec142e51da9b9b3f6d Mon Sep 17 00:00:00 2001
From: Yannick Brosseau <scientist@fb.com>
Date: Mon, 25 Jan 2016 17:12:25 -0800
Subject: netfilter: ipvs: Remove noisy debug print from ip_vs_del_service

This have been there for a long time, but does not seem to add value

Signed-off-by: Yannick Brosseau <scientist@fb.com>
Signed-off-by: Simon Horman <horms@verge.net.au>

diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index e7c1b05..daf4cb7 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -1376,8 +1376,6 @@ static void __ip_vs_del_service(struct ip_vs_service *svc, bool cleanup)
 	struct ip_vs_pe *old_pe;
 	struct netns_ipvs *ipvs = svc->ipvs;
 
-	pr_info("%s: enter\n", __func__);
-
 	/* Count only IPv4 services for old get/setsockopt interface */
 	if (svc->af == AF_INET)
 		ipvs->num_services--;
-- 
cgit v0.10.2


From f6ca9f46f6615c3a87529550058d1b468c0cad89 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 27 Jan 2016 14:52:01 +0100
Subject: netfilter: ipvs: avoid unused variable warnings

The proc_create() and remove_proc_entry() functions do not reference
their arguments when CONFIG_PROC_FS is disabled, so we get a couple
of warnings about unused variables in IPVS:

ipvs/ip_vs_app.c:608:14: warning: unused variable 'net' [-Wunused-variable]
ipvs/ip_vs_ctl.c:3950:14: warning: unused variable 'net' [-Wunused-variable]
ipvs/ip_vs_ctl.c:3994:14: warning: unused variable 'net' [-Wunused-variable]

This removes the local variables and instead looks them up separately
for each use, which obviously avoids the warning.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 4c50a8ce2b63 ("netfilter: ipvs: avoid unused variable warning")
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>

diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c
index 0328f72..299edc6 100644
--- a/net/netfilter/ipvs/ip_vs_app.c
+++ b/net/netfilter/ipvs/ip_vs_app.c
@@ -605,17 +605,13 @@ static const struct file_operations ip_vs_app_fops = {
 
 int __net_init ip_vs_app_net_init(struct netns_ipvs *ipvs)
 {
-	struct net *net = ipvs->net;
-
 	INIT_LIST_HEAD(&ipvs->app_list);
-	proc_create("ip_vs_app", 0, net->proc_net, &ip_vs_app_fops);
+	proc_create("ip_vs_app", 0, ipvs->net->proc_net, &ip_vs_app_fops);
 	return 0;
 }
 
 void __net_exit ip_vs_app_net_cleanup(struct netns_ipvs *ipvs)
 {
-	struct net *net = ipvs->net;
-
 	unregister_ip_vs_app(ipvs, NULL /* all */);
-	remove_proc_entry("ip_vs_app", net->proc_net);
+	remove_proc_entry("ip_vs_app", ipvs->net->proc_net);
 }
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index daf4cb7..404b2a4 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -3945,7 +3945,6 @@ static struct notifier_block ip_vs_dst_notifier = {
 
 int __net_init ip_vs_control_net_init(struct netns_ipvs *ipvs)
 {
-	struct net *net = ipvs->net;
 	int i, idx;
 
 	/* Initialize rs_table */
@@ -3972,9 +3971,9 @@ int __net_init ip_vs_control_net_init(struct netns_ipvs *ipvs)
 
 	spin_lock_init(&ipvs->tot_stats.lock);
 
-	proc_create("ip_vs", 0, net->proc_net, &ip_vs_info_fops);
-	proc_create("ip_vs_stats", 0, net->proc_net, &ip_vs_stats_fops);
-	proc_create("ip_vs_stats_percpu", 0, net->proc_net,
+	proc_create("ip_vs", 0, ipvs->net->proc_net, &ip_vs_info_fops);
+	proc_create("ip_vs_stats", 0, ipvs->net->proc_net, &ip_vs_stats_fops);
+	proc_create("ip_vs_stats_percpu", 0, ipvs->net->proc_net,
 		    &ip_vs_stats_percpu_fops);
 
 	if (ip_vs_control_net_init_sysctl(ipvs))
@@ -3989,13 +3988,11 @@ err:
 
 void __net_exit ip_vs_control_net_cleanup(struct netns_ipvs *ipvs)
 {
-	struct net *net = ipvs->net;
-
 	ip_vs_trash_cleanup(ipvs);
 	ip_vs_control_net_cleanup_sysctl(ipvs);
-	remove_proc_entry("ip_vs_stats_percpu", net->proc_net);
-	remove_proc_entry("ip_vs_stats", net->proc_net);
-	remove_proc_entry("ip_vs", net->proc_net);
+	remove_proc_entry("ip_vs_stats_percpu", ipvs->net->proc_net);
+	remove_proc_entry("ip_vs_stats", ipvs->net->proc_net);
+	remove_proc_entry("ip_vs", ipvs->net->proc_net);
 	free_percpu(ipvs->tot_stats.cpustats);
 }
 
-- 
cgit v0.10.2


From fa477f4cb3de7bdd3899029803ebfcf269ba8c85 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Wed, 17 Feb 2016 14:34:53 -0800
Subject: ixgbe: use u32 instead of __u32 in model header

I incorrectly used __u32 types where we should be using u32 types when
I added the ixgbe_model.h file.

Fixes: 9d35cf062e05 ("net: ixgbe: add minimal parser details for ixgbe")
Suggested-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
index 43ebec4..62ea2e7 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
@@ -35,13 +35,13 @@ struct ixgbe_mat_field {
 	unsigned int mask;
 	int (*val)(struct ixgbe_fdir_filter *input,
 		   union ixgbe_atr_input *mask,
-		   __u32 val, __u32 m);
+		   u32 val, u32 m);
 	unsigned int type;
 };
 
 static inline int ixgbe_mat_prgm_sip(struct ixgbe_fdir_filter *input,
 				     union ixgbe_atr_input *mask,
-				     __u32 val, __u32 m)
+				     u32 val, u32 m)
 {
 	input->filter.formatted.src_ip[0] = val;
 	mask->formatted.src_ip[0] = m;
@@ -50,7 +50,7 @@ static inline int ixgbe_mat_prgm_sip(struct ixgbe_fdir_filter *input,
 
 static inline int ixgbe_mat_prgm_dip(struct ixgbe_fdir_filter *input,
 				     union ixgbe_atr_input *mask,
-				     __u32 val, __u32 m)
+				     u32 val, u32 m)
 {
 	input->filter.formatted.dst_ip[0] = val;
 	mask->formatted.dst_ip[0] = m;
@@ -67,7 +67,7 @@ static struct ixgbe_mat_field ixgbe_ipv4_fields[] = {
 
 static inline int ixgbe_mat_prgm_sport(struct ixgbe_fdir_filter *input,
 				       union ixgbe_atr_input *mask,
-				       __u32 val, __u32 m)
+				       u32 val, u32 m)
 {
 	input->filter.formatted.src_port = val & 0xffff;
 	mask->formatted.src_port = m & 0xffff;
@@ -76,7 +76,7 @@ static inline int ixgbe_mat_prgm_sport(struct ixgbe_fdir_filter *input,
 
 static inline int ixgbe_mat_prgm_dport(struct ixgbe_fdir_filter *input,
 				       union ixgbe_atr_input *mask,
-				       __u32 val, __u32 m)
+				       u32 val, u32 m)
 {
 	input->filter.formatted.dst_port = val & 0xffff;
 	mask->formatted.dst_port = m & 0xffff;
@@ -94,12 +94,12 @@ static struct ixgbe_mat_field ixgbe_tcp_fields[] = {
 struct ixgbe_nexthdr {
 	/* offset, shift, and mask of position to next header */
 	unsigned int o;
-	__u32 s;
-	__u32 m;
+	u32 s;
+	u32 m;
 	/* match criteria to make this jump*/
 	unsigned int off;
-	__u32 val;
-	__u32 mask;
+	u32 val;
+	u32 mask;
 	/* location of jump to make */
 	struct ixgbe_mat_field *jump;
 };
-- 
cgit v0.10.2


From a92265ce1cea3832a47103ae16afa328a396e9af Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Wed, 17 Feb 2016 14:35:23 -0800
Subject: ixgbe: fix dates on header of ixgbe_model.h

Fixes: 9d35cf062e05 ("net: ixgbe: add minimal parser details for ixgbe")
Reported-by: Mark Rustad <mark.d.rustad@intel.com>
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
index 62ea2e7..ce48872 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel 10 Gigabit PCI Express Linux drive
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
-- 
cgit v0.10.2


From e014860e31e2a66b1a94088504360a6ebc023564 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Wed, 17 Feb 2016 14:59:30 -0800
Subject: net: pack tc_cls_u32_knode struct slighter better

By packing the structure we can remove a few holes as Jamal
suggests.

before:

struct tc_cls_u32_knode {
	struct tcf_exts *          exts;                 /*     0     8 */
	u8                         fshift;               /*     8     1 */

	/* XXX 3 bytes hole, try to pack */

	u32                        handle;               /*    12     4 */
	u32                        val;                  /*    16     4 */
	u32                        mask;                 /*    20     4 */
	u32                        link_handle;          /*    24     4 */

	/* XXX 4 bytes hole, try to pack */

	struct tc_u32_sel *        sel;                  /*    32     8 */

	/* size: 40, cachelines: 1, members: 7 */
	/* sum members: 33, holes: 2, sum holes: 7 */
	/* last cacheline: 40 bytes */
};

after:

struct tc_cls_u32_knode {
	struct tcf_exts *          exts;                 /*     0     8 */
	struct tc_u32_sel *        sel;                  /*     8     8 */
	u32                        handle;               /*    16     4 */
	u32                        val;                  /*    20     4 */
	u32                        mask;                 /*    24     4 */
	u32                        link_handle;          /*    28     4 */
	u8                         fshift;               /*    32     1 */

	/* size: 40, cachelines: 1, members: 7 */
	/* padding: 7 */
	/* last cacheline: 40 bytes */
};

Suggested-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 59789ca..2121df5 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -360,12 +360,12 @@ tcf_match_indev(struct sk_buff *skb, int ifindex)
 
 struct tc_cls_u32_knode {
 	struct tcf_exts *exts;
-	u8 fshift;
+	struct tc_u32_sel *sel;
 	u32 handle;
 	u32 val;
 	u32 mask;
 	u32 link_handle;
-	struct tc_u32_sel *sel;
+	u8 fshift;
 };
 
 struct tc_cls_u32_hnode {
-- 
cgit v0.10.2


From e4999f256cdde4f2cbb8fc26fd61819cf4d95cf7 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 18 Feb 2016 02:29:30 +0000
Subject: vlan: change return type of vlan_proc_rem_dev

Since function vlan_proc_rem_dev() will only return 0, it's better to
return void instead of int.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c
index ae63cf7..5f1446c 100644
--- a/net/8021q/vlanproc.c
+++ b/net/8021q/vlanproc.c
@@ -184,12 +184,11 @@ int vlan_proc_add_dev(struct net_device *vlandev)
 /*
  *	Delete directory entry for VLAN device.
  */
-int vlan_proc_rem_dev(struct net_device *vlandev)
+void vlan_proc_rem_dev(struct net_device *vlandev)
 {
 	/** NOTE:  This will consume the memory pointed to by dent, it seems. */
 	proc_remove(vlan_dev_priv(vlandev)->dent);
 	vlan_dev_priv(vlandev)->dent = NULL;
-	return 0;
 }
 
 /****** Proc filesystem entry points ****************************************/
diff --git a/net/8021q/vlanproc.h b/net/8021q/vlanproc.h
index 063f60a..8838a2e 100644
--- a/net/8021q/vlanproc.h
+++ b/net/8021q/vlanproc.h
@@ -5,7 +5,7 @@
 struct net;
 
 int vlan_proc_init(struct net *net);
-int vlan_proc_rem_dev(struct net_device *vlandev);
+void vlan_proc_rem_dev(struct net_device *vlandev);
 int vlan_proc_add_dev(struct net_device *vlandev);
 void vlan_proc_cleanup(struct net *net);
 
@@ -14,7 +14,7 @@ void vlan_proc_cleanup(struct net *net);
 #define vlan_proc_init(net)	(0)
 #define vlan_proc_cleanup(net)	do {} while (0)
 #define vlan_proc_add_dev(dev)	({(void)(dev), 0; })
-#define vlan_proc_rem_dev(dev)	({(void)(dev), 0; })
+#define vlan_proc_rem_dev(dev)	do {} while (0)
 #endif
 
 #endif /* !(__BEN_VLAN_PROC_INC__) */
-- 
cgit v0.10.2


From 76f94a9c77d2fe6f0f02235eea84dae19edd3161 Mon Sep 17 00:00:00 2001
From: Iyappan Subramanian <isubramanian@apm.com>
Date: Wed, 17 Feb 2016 15:00:39 -0800
Subject: drivers: net: xgene: Add support for Classifier engine

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Khuong Dinh <kdinh@apm.com>
Signed-off-by: Tanmay Inamdar <tinamdar@apm.com>
Tested-by: Toan Le <toanle@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/apm/xgene/Makefile b/drivers/net/ethernet/apm/xgene/Makefile
index 700b5ab..f46321f 100644
--- a/drivers/net/ethernet/apm/xgene/Makefile
+++ b/drivers/net/ethernet/apm/xgene/Makefile
@@ -3,5 +3,6 @@
 #
 
 xgene-enet-objs := xgene_enet_hw.o xgene_enet_sgmac.o xgene_enet_xgmac.o \
-		   xgene_enet_main.o xgene_enet_ring2.o xgene_enet_ethtool.o
+		   xgene_enet_main.o xgene_enet_ring2.o xgene_enet_ethtool.o \
+		   xgene_enet_cle.o
 obj-$(CONFIG_NET_XGENE) += xgene-enet.o
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
new file mode 100644
index 0000000..ff24ca9
--- /dev/null
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
@@ -0,0 +1,357 @@
+/* Applied Micro X-Gene SoC Ethernet Classifier structures
+ *
+ * Copyright (c) 2016, Applied Micro Circuits Corporation
+ * Authors: Khuong Dinh <kdinh@apm.com>
+ *          Tanmay Inamdar <tinamdar@apm.com>
+ *          Iyappan Subramanian <isubramanian@apm.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "xgene_enet_main.h"
+
+static void xgene_cle_dbptr_to_hw(struct xgene_enet_pdata *pdata,
+				  struct xgene_cle_dbptr *dbptr, u32 *buf)
+{
+	buf[4] = SET_VAL(CLE_FPSEL, dbptr->fpsel) |
+		 SET_VAL(CLE_DSTQIDL, dbptr->dstqid);
+
+	buf[5] = SET_VAL(CLE_DSTQIDH, (u32)dbptr->dstqid >> CLE_DSTQIDL_LEN) |
+		 SET_VAL(CLE_PRIORITY, dbptr->cle_priority);
+}
+
+static void xgene_cle_kn_to_hw(struct xgene_cle_ptree_kn *kn, u32 *buf)
+{
+	u32 i, j = 0;
+	u32 data;
+
+	buf[j++] = SET_VAL(CLE_TYPE, kn->node_type);
+	for (i = 0; i < kn->num_keys; i++) {
+		struct xgene_cle_ptree_key *key = &kn->key[i];
+
+		if (!(i % 2)) {
+			buf[j] = SET_VAL(CLE_KN_PRIO, key->priority) |
+				 SET_VAL(CLE_KN_RPTR, key->result_pointer);
+		} else {
+			data = SET_VAL(CLE_KN_PRIO, key->priority) |
+			       SET_VAL(CLE_KN_RPTR, key->result_pointer);
+			buf[j++] |= (data << 16);
+		}
+	}
+}
+
+static void xgene_cle_dn_to_hw(struct xgene_cle_ptree_ewdn *dn,
+			       u32 *buf, u32 jb)
+{
+	struct xgene_cle_ptree_branch *br;
+	u32 i, j = 0;
+	u32 npp;
+
+	buf[j++] = SET_VAL(CLE_DN_TYPE, dn->node_type) |
+		   SET_VAL(CLE_DN_LASTN, dn->last_node) |
+		   SET_VAL(CLE_DN_HLS, dn->hdr_len_store) |
+		   SET_VAL(CLE_DN_EXT, dn->hdr_extn) |
+		   SET_VAL(CLE_DN_BSTOR, dn->byte_store) |
+		   SET_VAL(CLE_DN_SBSTOR, dn->search_byte_store) |
+		   SET_VAL(CLE_DN_RPTR, dn->result_pointer);
+
+	for (i = 0; i < dn->num_branches; i++) {
+		br = &dn->branch[i];
+		npp = br->next_packet_pointer;
+
+		if ((br->jump_rel == JMP_ABS) && (npp < CLE_PKTRAM_SIZE))
+			npp += jb;
+
+		buf[j++] = SET_VAL(CLE_BR_VALID, br->valid) |
+			   SET_VAL(CLE_BR_NPPTR, npp) |
+			   SET_VAL(CLE_BR_JB, br->jump_bw) |
+			   SET_VAL(CLE_BR_JR, br->jump_rel) |
+			   SET_VAL(CLE_BR_OP, br->operation) |
+			   SET_VAL(CLE_BR_NNODE, br->next_node) |
+			   SET_VAL(CLE_BR_NBR, br->next_branch);
+
+		buf[j++] = SET_VAL(CLE_BR_DATA, br->data) |
+			   SET_VAL(CLE_BR_MASK, br->mask);
+	}
+}
+
+static int xgene_cle_poll_cmd_done(void __iomem *base,
+				   enum xgene_cle_cmd_type cmd)
+{
+	u32 status, loop = 10;
+	int ret = -EBUSY;
+
+	while (loop--) {
+		status = ioread32(base + INDCMD_STATUS);
+		if (status & cmd) {
+			ret = 0;
+			break;
+		}
+		usleep_range(1000, 2000);
+	}
+
+	return ret;
+}
+
+static int xgene_cle_dram_wr(struct xgene_enet_cle *cle, u32 *data, u8 nregs,
+			     u32 index, enum xgene_cle_dram_type type,
+			     enum xgene_cle_cmd_type cmd)
+{
+	enum xgene_cle_parser parser = cle->active_parser;
+	void __iomem *base = cle->base;
+	u32 i, j, ind_addr;
+	u8 port, nparsers;
+	int ret = 0;
+
+	/* PTREE_RAM onwards, DRAM regions are common for all parsers */
+	nparsers = (type >= PTREE_RAM) ? 1 : cle->parsers;
+
+	for (i = 0; i < nparsers; i++) {
+		port = i;
+		if ((type < PTREE_RAM) && (parser != PARSER_ALL))
+			port = parser;
+
+		ind_addr = XGENE_CLE_DRAM(type + (port * 4)) | index;
+		iowrite32(ind_addr, base + INDADDR);
+		for (j = 0; j < nregs; j++)
+			iowrite32(data[j], base + DATA_RAM0 + (j * 4));
+		iowrite32(cmd, base + INDCMD);
+
+		ret = xgene_cle_poll_cmd_done(base, cmd);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+static void xgene_cle_enable_ptree(struct xgene_enet_pdata *pdata,
+				   struct xgene_enet_cle *cle)
+{
+	struct xgene_cle_ptree *ptree = &cle->ptree;
+	void __iomem *addr, *base = cle->base;
+	u32 offset = CLE_PORT_OFFSET;
+	u32 i;
+
+	/* 1G port has to advance 4 bytes and 10G has to advance 8 bytes */
+	ptree->start_pkt += cle->jump_bytes;
+	for (i = 0; i < cle->parsers; i++) {
+		if (cle->active_parser != PARSER_ALL)
+			addr = base + cle->active_parser * offset;
+		else
+			addr = base + (i * offset);
+
+		iowrite32(ptree->start_node & 0x3fff, addr + SNPTR0);
+		iowrite32(ptree->start_pkt & 0x1ff, addr + SPPTR0);
+	}
+}
+
+static int xgene_cle_setup_dbptr(struct xgene_enet_pdata *pdata,
+				 struct xgene_enet_cle *cle)
+{
+	struct xgene_cle_ptree *ptree = &cle->ptree;
+	u32 buf[CLE_DRAM_REGS];
+	u32 i;
+	int ret;
+
+	memset(buf, 0, sizeof(buf));
+	for (i = 0; i < ptree->num_dbptr; i++) {
+		xgene_cle_dbptr_to_hw(pdata, &ptree->dbptr[i], buf);
+		ret = xgene_cle_dram_wr(cle, buf, 6, i + ptree->start_dbptr,
+					DB_RAM,	CLE_CMD_WR);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int xgene_cle_setup_node(struct xgene_enet_pdata *pdata,
+				struct xgene_enet_cle *cle)
+{
+	struct xgene_cle_ptree *ptree = &cle->ptree;
+	struct xgene_cle_ptree_ewdn *dn = ptree->dn;
+	struct xgene_cle_ptree_kn *kn = ptree->kn;
+	u32 buf[CLE_DRAM_REGS];
+	int i, j, ret;
+
+	memset(buf, 0, sizeof(buf));
+	for (i = 0; i < ptree->num_dn; i++) {
+		xgene_cle_dn_to_hw(&dn[i], buf, cle->jump_bytes);
+		ret = xgene_cle_dram_wr(cle, buf, 17, i + ptree->start_node,
+					PTREE_RAM, CLE_CMD_WR);
+		if (ret)
+			return ret;
+	}
+
+	/* continue node index for key node */
+	memset(buf, 0, sizeof(buf));
+	for (j = i; j < (ptree->num_kn + ptree->num_dn); j++) {
+		xgene_cle_kn_to_hw(&kn[j - ptree->num_dn], buf);
+		ret = xgene_cle_dram_wr(cle, buf, 17, j + ptree->start_node,
+					PTREE_RAM, CLE_CMD_WR);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int xgene_cle_setup_ptree(struct xgene_enet_pdata *pdata,
+				 struct xgene_enet_cle *cle)
+{
+	int ret;
+
+	ret = xgene_cle_setup_node(pdata, cle);
+	if (ret)
+		return ret;
+
+	ret = xgene_cle_setup_dbptr(pdata, cle);
+	if (ret)
+		return ret;
+
+	xgene_cle_enable_ptree(pdata, cle);
+
+	return 0;
+}
+
+static void xgene_cle_setup_def_dbptr(struct xgene_enet_pdata *pdata,
+				      struct xgene_enet_cle *enet_cle,
+				      struct xgene_cle_dbptr *dbptr,
+				      u32 index, u8 priority)
+{
+	void __iomem *base = enet_cle->base;
+	void __iomem *base_addr;
+	u32 buf[CLE_DRAM_REGS];
+	u32 def_cls, offset;
+	u32 i, j;
+
+	memset(buf, 0, sizeof(buf));
+	xgene_cle_dbptr_to_hw(pdata, dbptr, buf);
+
+	for (i = 0; i < enet_cle->parsers; i++) {
+		if (enet_cle->active_parser != PARSER_ALL) {
+			offset = enet_cle->active_parser *
+				CLE_PORT_OFFSET;
+		} else {
+			offset = i * CLE_PORT_OFFSET;
+		}
+
+		base_addr = base + DFCLSRESDB00 + offset;
+		for (j = 0; j < 6; j++)
+			iowrite32(buf[j], base_addr + (j * 4));
+
+		def_cls = ((priority & 0x7) << 10) | (index & 0x3ff);
+		iowrite32(def_cls, base + DFCLSRESDBPTR0 + offset);
+	}
+}
+
+static int xgene_enet_cle_init(struct xgene_enet_pdata *pdata)
+{
+	struct xgene_enet_cle *enet_cle = &pdata->cle;
+	struct xgene_cle_dbptr dbptr[DB_MAX_PTRS];
+	u32 def_qid, def_fpsel, pool_id;
+	struct xgene_cle_ptree *ptree;
+	struct xgene_cle_ptree_kn kn;
+	struct xgene_cle_ptree_ewdn ptree_dn[] = {
+		{
+			/* PKT_TYPE_NODE */
+			.node_type = EWDN,
+			.last_node = 0,
+			.hdr_len_store = 0,
+			.hdr_extn = NO_BYTE,
+			.byte_store = NO_BYTE,
+			.search_byte_store = NO_BYTE,
+			.result_pointer = DB_RES_DROP,
+			.num_branches = 1,
+			.branch = {
+				{
+					/* Allow all packet type */
+					.valid = 0,
+					.next_packet_pointer = 0,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = LAST_NODE,
+					.next_branch = 0,
+					.data = 0x0,
+					.mask = 0xffff
+				}
+			}
+		},
+		{
+			/* LAST NODE */
+			.node_type = EWDN,
+			.last_node = 1,
+			.hdr_len_store = 0,
+			.hdr_extn = NO_BYTE,
+			.byte_store = NO_BYTE,
+			.search_byte_store = NO_BYTE,
+			.result_pointer = DB_RES_DROP,
+			.num_branches = 1,
+			.branch = {
+				{
+					.valid = 0,
+					.next_packet_pointer = 0,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = MAX_NODES,
+					.next_branch = 0,
+					.data = 0,
+					.mask = 0xffff
+				}
+			}
+		}
+	};
+
+	ptree = &enet_cle->ptree;
+	ptree->start_pkt = 12; /* Ethertype */
+
+	def_qid = xgene_enet_dst_ring_num(pdata->rx_ring);
+	pool_id = pdata->rx_ring->buf_pool->id;
+	def_fpsel = xgene_enet_ring_bufnum(pool_id) - 0x20;
+
+	memset(dbptr, 0, sizeof(struct xgene_cle_dbptr) * DB_MAX_PTRS);
+	dbptr[DB_RES_ACCEPT].fpsel =  def_fpsel;
+	dbptr[DB_RES_ACCEPT].dstqid = def_qid;
+	dbptr[DB_RES_ACCEPT].cle_priority = 1;
+
+	dbptr[DB_RES_DEF].fpsel = def_fpsel;
+	dbptr[DB_RES_DEF].dstqid = def_qid;
+	dbptr[DB_RES_DEF].cle_priority = 7;
+	xgene_cle_setup_def_dbptr(pdata, enet_cle, &dbptr[DB_RES_DEF],
+				  DB_RES_ACCEPT, 7);
+
+	dbptr[DB_RES_DROP].drop = 1;
+
+	memset(&kn, 0, sizeof(kn));
+	kn.node_type = KN;
+	kn.num_keys = 1;
+	kn.key[0].priority = 0;
+	kn.key[0].result_pointer = DB_RES_ACCEPT;
+
+	ptree->dn = ptree_dn;
+	ptree->kn = &kn;
+	ptree->dbptr = dbptr;
+	ptree->num_dn = MAX_NODES;
+	ptree->num_kn = 1;
+	ptree->num_dbptr = DB_MAX_PTRS;
+
+	return xgene_cle_setup_ptree(pdata, enet_cle);
+}
+
+struct xgene_cle_ops xgene_cle3in_ops = {
+	.cle_init = xgene_enet_cle_init,
+};
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
new file mode 100644
index 0000000..1db2fd7
--- /dev/null
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
@@ -0,0 +1,254 @@
+/* Applied Micro X-Gene SoC Ethernet Classifier structures
+ *
+ * Copyright (c) 2016, Applied Micro Circuits Corporation
+ * Authors: Khuong Dinh <kdinh@apm.com>
+ *          Tanmay Inamdar <tinamdar@apm.com>
+ *          Iyappan Subramanian <isubramanian@apm.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __XGENE_ENET_CLE_H__
+#define __XGENE_ENET_CLE_H__
+
+#include <linux/io.h>
+#include <linux/random.h>
+
+/* Register offsets */
+#define INDADDR			0x04
+#define INDCMD			0x08
+#define INDCMD_STATUS		0x0c
+#define DATA_RAM0		0x10
+#define SNPTR0			0x0100
+#define SPPTR0			0x0104
+#define DFCLSRESDBPTR0		0x0108
+#define DFCLSRESDB00		0x010c
+
+#define CLE_CMD_TO		10	/* ms */
+#define CLE_PKTRAM_SIZE		256	/* bytes */
+#define CLE_PORT_OFFSET		0x200
+#define CLE_DRAM_REGS		17
+
+#define CLE_DN_TYPE_LEN		2
+#define CLE_DN_TYPE_POS		0
+#define CLE_DN_LASTN_LEN	1
+#define CLE_DN_LASTN_POS	2
+#define CLE_DN_HLS_LEN		1
+#define CLE_DN_HLS_POS		3
+#define CLE_DN_EXT_LEN		2
+#define	CLE_DN_EXT_POS		4
+#define CLE_DN_BSTOR_LEN	2
+#define CLE_DN_BSTOR_POS	6
+#define CLE_DN_SBSTOR_LEN	2
+#define CLE_DN_SBSTOR_POS	8
+#define CLE_DN_RPTR_LEN		12
+#define CLE_DN_RPTR_POS		12
+
+#define CLE_BR_VALID_LEN	1
+#define CLE_BR_VALID_POS	0
+#define CLE_BR_NPPTR_LEN	9
+#define CLE_BR_NPPTR_POS	1
+#define CLE_BR_JB_LEN		1
+#define CLE_BR_JB_POS		10
+#define CLE_BR_JR_LEN		1
+#define CLE_BR_JR_POS		11
+#define CLE_BR_OP_LEN		3
+#define CLE_BR_OP_POS		12
+#define CLE_BR_NNODE_LEN	9
+#define CLE_BR_NNODE_POS	15
+#define CLE_BR_NBR_LEN		5
+#define CLE_BR_NBR_POS		24
+
+#define CLE_BR_DATA_LEN		16
+#define CLE_BR_DATA_POS		0
+#define CLE_BR_MASK_LEN		16
+#define CLE_BR_MASK_POS		16
+
+#define CLE_KN_PRIO_POS		0
+#define CLE_KN_PRIO_LEN		3
+#define CLE_KN_RPTR_POS		3
+#define CLE_KN_RPTR_LEN		10
+#define CLE_TYPE_POS		0
+#define CLE_TYPE_LEN		2
+
+#define CLE_DSTQIDL_POS		25
+#define CLE_DSTQIDL_LEN		7
+#define CLE_DSTQIDH_POS		0
+#define CLE_DSTQIDH_LEN		5
+#define CLE_FPSEL_POS		21
+#define CLE_FPSEL_LEN		4
+#define CLE_PRIORITY_POS	5
+#define CLE_PRIORITY_LEN	3
+
+#define JMP_ABS			0
+#define JMP_REL			1
+#define JMP_FW			0
+#define JMP_BW			1
+
+enum xgene_cle_ptree_nodes {
+	PKT_TYPE_NODE,
+	LAST_NODE,
+	MAX_NODES
+};
+
+enum xgene_cle_byte_store {
+	NO_BYTE,
+	FIRST_BYTE,
+	SECOND_BYTE,
+	BOTH_BYTES
+};
+
+/* Preclassification operation types */
+enum xgene_cle_node_type {
+	INV,
+	KN,
+	EWDN,
+	RES_NODE
+};
+
+/* Preclassification operation types */
+enum xgene_cle_op_type {
+	EQT,
+	NEQT,
+	LTEQT,
+	GTEQT,
+	AND,
+	NAND
+};
+
+enum xgene_cle_parser {
+	PARSER0,
+	PARSER1,
+	PARSER2,
+	PARSER_ALL
+};
+
+#define XGENE_CLE_DRAM(type)	(((type) & 0xf) << 28)
+enum xgene_cle_dram_type {
+	PKT_RAM,
+	PTREE_RAM = 0xc,
+	AVL_RAM,
+	DB_RAM
+};
+
+enum xgene_cle_cmd_type {
+	CLE_CMD_WR = 1,
+	CLE_CMD_RD = 2,
+	CLE_CMD_AVL_ADD = 8,
+	CLE_CMD_AVL_DEL = 16,
+	CLE_CMD_AVL_SRCH = 32
+};
+
+enum xgene_cle_ptree_dbptrs {
+	DB_RES_DROP,
+	DB_RES_DEF,
+	DB_RES_ACCEPT,
+	DB_MAX_PTRS
+};
+
+struct xgene_cle_ptree_branch {
+	bool valid;
+	u16 next_packet_pointer;
+	bool jump_bw;
+	bool jump_rel;
+	u8 operation;
+	u16 next_node;
+	u8 next_branch;
+	u16 data;
+	u16 mask;
+};
+
+struct xgene_cle_ptree_ewdn {
+	u8 node_type;
+	bool last_node;
+	bool hdr_len_store;
+	u8 hdr_extn;
+	u8 byte_store;
+	u8 search_byte_store;
+	u16 result_pointer;
+	u8 num_branches;
+	struct xgene_cle_ptree_branch branch[6];
+};
+
+struct xgene_cle_ptree_key {
+	u8 priority;
+	u16 result_pointer;
+};
+
+struct xgene_cle_ptree_kn {
+	u8 node_type;
+	u8 num_keys;
+	struct xgene_cle_ptree_key key[32];
+};
+
+struct xgene_cle_dbptr {
+	u8 split_boundary;
+	u8 mirror_nxtfpsel;
+	u8 mirror_fpsel;
+	u16 mirror_dstqid;
+	u8 drop;
+	u8 mirror;
+	u8 hdr_data_split;
+	u64 hopinfomsbs;
+	u8 DR;
+	u8 HR;
+	u64 hopinfomlsbs;
+	u16 h0enq_num;
+	u8 h0fpsel;
+	u8 nxtfpsel;
+	u8 fpsel;
+	u16 dstqid;
+	u8 cle_priority;
+	u8 cle_flowgroup;
+	u8 cle_perflow;
+	u8 cle_insert_timestamp;
+	u8 stash;
+	u8 in;
+	u8 perprioen;
+	u8 perflowgroupen;
+	u8 perflowen;
+	u8 selhash;
+	u8 selhdrext;
+	u8 mirror_nxtfpsel_msb;
+	u8 mirror_fpsel_msb;
+	u8 hfpsel_msb;
+	u8 nxtfpsel_msb;
+	u8 fpsel_msb;
+};
+
+struct xgene_cle_ptree {
+	struct xgene_cle_ptree_ewdn *dn;
+	struct xgene_cle_ptree_kn *kn;
+	struct xgene_cle_dbptr *dbptr;
+	u32 num_dn;
+	u32 num_kn;
+	u32 num_dbptr;
+	u32 start_node;
+	u32 start_pkt;
+	u32 start_dbptr;
+};
+
+struct xgene_enet_cle {
+	void __iomem *base;
+	struct xgene_cle_ptree ptree;
+	enum xgene_cle_parser active_parser;
+	u32 parsers;
+	u32 max_nodes;
+	u32 max_dbptrs;
+	u32 jump_bytes;
+};
+
+extern struct xgene_cle_ops xgene_cle3in_ops;
+
+#endif /* __XGENE_ENET_CLE_H__ */
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
index 8a90910..45725ec 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
@@ -101,6 +101,7 @@ enum xgene_enet_rm {
 #define MAC_OFFSET			0x30
 
 #define BLOCK_ETH_CSR_OFFSET		0x2000
+#define BLOCK_ETH_CLE_CSR_OFFSET	0x6000
 #define BLOCK_ETH_RING_IF_OFFSET	0x9000
 #define BLOCK_ETH_CLKRST_CSR_OFFSET	0xc000
 #define BLOCK_ETH_DIAG_CSR_OFFSET	0xD000
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index 5eb9b20..0bf3924 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -93,13 +93,6 @@ static int xgene_enet_refill_bufpool(struct xgene_enet_desc_ring *buf_pool,
 	return 0;
 }
 
-static u16 xgene_enet_dst_ring_num(struct xgene_enet_desc_ring *ring)
-{
-	struct xgene_enet_pdata *pdata = netdev_priv(ring->ndev);
-
-	return ((u16)pdata->rm << 10) | ring->num;
-}
-
 static u8 xgene_enet_hdr_len(const void *data)
 {
 	const struct ethhdr *eth = data;
@@ -1278,6 +1271,7 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
 	else
 		base_addr = pdata->base_addr;
 	pdata->eth_csr_addr = base_addr + BLOCK_ETH_CSR_OFFSET;
+	pdata->cle.base = base_addr + BLOCK_ETH_CLE_CSR_OFFSET;
 	pdata->eth_ring_if_addr = base_addr + BLOCK_ETH_RING_IF_OFFSET;
 	pdata->eth_diag_csr_addr = base_addr + BLOCK_ETH_DIAG_CSR_OFFSET;
 	if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII ||
@@ -1298,6 +1292,7 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
 
 static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata)
 {
+	struct xgene_enet_cle *enet_cle = &pdata->cle;
 	struct net_device *ndev = pdata->ndev;
 	struct xgene_enet_desc_ring *buf_pool;
 	u16 dst_ring_num;
@@ -1323,7 +1318,24 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata)
 	}
 
 	dst_ring_num = xgene_enet_dst_ring_num(pdata->rx_ring);
-	pdata->port_ops->cle_bypass(pdata, dst_ring_num, buf_pool->id);
+	if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
+		/* Initialize and Enable  PreClassifier Tree */
+		enet_cle->max_nodes = 512;
+		enet_cle->max_dbptrs = 1024;
+		enet_cle->parsers = 3;
+		enet_cle->active_parser = PARSER_ALL;
+		enet_cle->ptree.start_node = 0;
+		enet_cle->ptree.start_dbptr = 0;
+		enet_cle->jump_bytes = 8;
+		ret = pdata->cle_ops->cle_init(pdata);
+		if (ret) {
+			netdev_err(ndev, "Preclass Tree init error\n");
+			return ret;
+		}
+	} else {
+		pdata->port_ops->cle_bypass(pdata, dst_ring_num, buf_pool->id);
+	}
+
 	pdata->mac_ops->init(pdata);
 
 	return ret;
@@ -1345,6 +1357,7 @@ static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata)
 	default:
 		pdata->mac_ops = &xgene_xgmac_ops;
 		pdata->port_ops = &xgene_xgport_ops;
+		pdata->cle_ops = &xgene_cle3in_ops;
 		pdata->rm = RM0;
 		break;
 	}
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
index 248dfc4..05365c1 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
@@ -36,6 +36,7 @@
 #include <linux/if_vlan.h>
 #include <linux/phy.h>
 #include "xgene_enet_hw.h"
+#include "xgene_enet_cle.h"
 #include "xgene_enet_ring2.h"
 
 #define XGENE_DRV_VERSION	"v1.0"
@@ -145,6 +146,10 @@ struct xgene_ring_ops {
 	u32 (*len)(struct xgene_enet_desc_ring *);
 };
 
+struct xgene_cle_ops {
+	int (*cle_init)(struct xgene_enet_pdata *pdata);
+};
+
 /* ethernet private data */
 struct xgene_enet_pdata {
 	struct net_device *ndev;
@@ -174,10 +179,12 @@ struct xgene_enet_pdata {
 	void __iomem *ring_cmd_addr;
 	int phy_mode;
 	enum xgene_enet_rm rm;
+	struct xgene_enet_cle cle;
 	struct rtnl_link_stats64 stats;
 	const struct xgene_mac_ops *mac_ops;
 	const struct xgene_port_ops *port_ops;
 	struct xgene_ring_ops *ring_ops;
+	struct xgene_cle_ops *cle_ops;
 	struct delayed_work link_work;
 	u32 port_id;
 	u8 cpu_bufnum;
@@ -229,6 +236,13 @@ static inline struct device *ndev_to_dev(struct net_device *ndev)
 	return ndev->dev.parent;
 }
 
+static inline u16 xgene_enet_dst_ring_num(struct xgene_enet_desc_ring *ring)
+{
+	struct xgene_enet_pdata *pdata = netdev_priv(ring->ndev);
+
+	return ((u16)pdata->rm << 10) | ring->num;
+}
+
 void xgene_enet_set_ethtool_ops(struct net_device *netdev);
 
 #endif /* __XGENE_ENET_MAIN_H__ */
-- 
cgit v0.10.2


From fc4262d2aa9aa26ec4a29bf274bcfca569656e73 Mon Sep 17 00:00:00 2001
From: Iyappan Subramanian <isubramanian@apm.com>
Date: Wed, 17 Feb 2016 15:00:40 -0800
Subject: drivers: net: xgene: Add support for RSS

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Khuong Dinh <kdinh@apm.com>
Signed-off-by: Tanmay Inamdar <tinamdar@apm.com>
Tested-by: Toan Le <toanle@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
index ff24ca9..c007497 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
@@ -21,6 +21,25 @@
 
 #include "xgene_enet_main.h"
 
+/* interfaces to convert structures to HW recognized bit formats */
+static void xgene_cle_sband_to_hw(u8 frag, enum xgene_cle_prot_version ver,
+				  enum xgene_cle_prot_type type, u32 len,
+				  u32 *reg)
+{
+	*reg =  SET_VAL(SB_IPFRAG, frag) |
+		SET_VAL(SB_IPPROT, type) |
+		SET_VAL(SB_IPVER, ver) |
+		SET_VAL(SB_HDRLEN, len);
+}
+
+static void xgene_cle_idt_to_hw(u32 dstqid, u32 fpsel,
+				u32 nfpsel, u32 *idt_reg)
+{
+	*idt_reg =  SET_VAL(IDT_DSTQID, dstqid) |
+		    SET_VAL(IDT_FPSEL, fpsel) |
+		    SET_VAL(IDT_NFPSEL, nfpsel);
+}
+
 static void xgene_cle_dbptr_to_hw(struct xgene_enet_pdata *pdata,
 				  struct xgene_cle_dbptr *dbptr, u32 *buf)
 {
@@ -257,29 +276,372 @@ static void xgene_cle_setup_def_dbptr(struct xgene_enet_pdata *pdata,
 	}
 }
 
+static int xgene_cle_set_rss_sband(struct xgene_enet_cle *cle)
+{
+	u32 idx = CLE_PKTRAM_SIZE / sizeof(u32);
+	u32 mac_hdr_len = ETH_HLEN;
+	u32 sband, reg = 0;
+	u32 ipv4_ihl = 5;
+	u32 hdr_len;
+	int ret;
+
+	/* Sideband: IPV4/TCP packets */
+	hdr_len = (mac_hdr_len << 5) | ipv4_ihl;
+	xgene_cle_sband_to_hw(0, XGENE_CLE_IPV4, XGENE_CLE_TCP, hdr_len, &reg);
+	sband = reg;
+
+	/* Sideband: IPv4/UDP packets */
+	hdr_len = (mac_hdr_len << 5) | ipv4_ihl;
+	xgene_cle_sband_to_hw(1, XGENE_CLE_IPV4, XGENE_CLE_UDP, hdr_len, &reg);
+	sband |= (reg << 16);
+
+	ret = xgene_cle_dram_wr(cle, &sband, 1, idx, PKT_RAM, CLE_CMD_WR);
+	if (ret)
+		return ret;
+
+	/* Sideband: IPv4/RAW packets */
+	hdr_len = (mac_hdr_len << 5) | ipv4_ihl;
+	xgene_cle_sband_to_hw(0, XGENE_CLE_IPV4, XGENE_CLE_OTHER,
+			      hdr_len, &reg);
+	sband = reg;
+
+	/* Sideband: Ethernet II/RAW packets */
+	hdr_len = (mac_hdr_len << 5);
+	xgene_cle_sband_to_hw(0, XGENE_CLE_IPV4, XGENE_CLE_OTHER,
+			      hdr_len, &reg);
+	sband |= (reg << 16);
+
+	ret = xgene_cle_dram_wr(cle, &sband, 1, idx + 1, PKT_RAM, CLE_CMD_WR);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int xgene_cle_set_rss_skeys(struct xgene_enet_cle *cle)
+{
+	u32 secret_key_ipv4[4];  /* 16 Bytes*/
+	int ret = 0;
+
+	get_random_bytes(secret_key_ipv4, 16);
+	ret = xgene_cle_dram_wr(cle, secret_key_ipv4, 4, 0,
+				RSS_IPV4_HASH_SKEY, CLE_CMD_WR);
+	return ret;
+}
+
+static int xgene_cle_set_rss_idt(struct xgene_enet_pdata *pdata)
+{
+	u32 fpsel, dstqid, nfpsel, idt_reg;
+	int i, ret = 0;
+	u16 pool_id;
+
+	for (i = 0; i < XGENE_CLE_IDT_ENTRIES; i++) {
+		pool_id = pdata->rx_ring->buf_pool->id;
+		fpsel = xgene_enet_ring_bufnum(pool_id) - 0x20;
+		dstqid = xgene_enet_dst_ring_num(pdata->rx_ring);
+		nfpsel = 0;
+		idt_reg = 0;
+
+		xgene_cle_idt_to_hw(dstqid, fpsel, nfpsel, &idt_reg);
+		ret = xgene_cle_dram_wr(&pdata->cle, &idt_reg, 1, i,
+					RSS_IDT, CLE_CMD_WR);
+		if (ret)
+			return ret;
+	}
+
+	ret = xgene_cle_set_rss_skeys(&pdata->cle);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int xgene_cle_setup_rss(struct xgene_enet_pdata *pdata)
+{
+	struct xgene_enet_cle *cle = &pdata->cle;
+	void __iomem *base = cle->base;
+	u32 offset, val = 0;
+	int i, ret = 0;
+
+	offset = CLE_PORT_OFFSET;
+	for (i = 0; i < cle->parsers; i++) {
+		if (cle->active_parser != PARSER_ALL)
+			offset = cle->active_parser * CLE_PORT_OFFSET;
+		else
+			offset = i * CLE_PORT_OFFSET;
+
+		/* enable RSS */
+		val = (RSS_IPV4_12B << 1) | 0x1;
+		writel(val, base + RSS_CTRL0 + offset);
+	}
+
+	/* setup sideband data */
+	ret = xgene_cle_set_rss_sband(cle);
+	if (ret)
+		return ret;
+
+	/* setup indirection table */
+	ret = xgene_cle_set_rss_idt(pdata);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
 static int xgene_enet_cle_init(struct xgene_enet_pdata *pdata)
 {
 	struct xgene_enet_cle *enet_cle = &pdata->cle;
 	struct xgene_cle_dbptr dbptr[DB_MAX_PTRS];
+	struct xgene_cle_ptree_branch *br;
 	u32 def_qid, def_fpsel, pool_id;
 	struct xgene_cle_ptree *ptree;
 	struct xgene_cle_ptree_kn kn;
+	int ret;
 	struct xgene_cle_ptree_ewdn ptree_dn[] = {
 		{
 			/* PKT_TYPE_NODE */
 			.node_type = EWDN,
 			.last_node = 0,
-			.hdr_len_store = 0,
+			.hdr_len_store = 1,
 			.hdr_extn = NO_BYTE,
 			.byte_store = NO_BYTE,
 			.search_byte_store = NO_BYTE,
 			.result_pointer = DB_RES_DROP,
-			.num_branches = 1,
+			.num_branches = 2,
 			.branch = {
 				{
-					/* Allow all packet type */
+					/* IPV4 */
 					.valid = 0,
-					.next_packet_pointer = 0,
+					.next_packet_pointer = 22,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = PKT_PROT_NODE,
+					.next_branch = 0,
+					.data = 0x8,
+					.mask = 0xffff
+				},
+				{
+					.valid = 0,
+					.next_packet_pointer = 262,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = LAST_NODE,
+					.next_branch = 0,
+					.data = 0x0,
+					.mask = 0xffff
+				}
+			},
+		},
+		{
+			/* PKT_PROT_NODE */
+			.node_type = EWDN,
+			.last_node = 0,
+			.hdr_len_store = 1,
+			.hdr_extn = NO_BYTE,
+			.byte_store = NO_BYTE,
+			.search_byte_store = NO_BYTE,
+			.result_pointer = DB_RES_DROP,
+			.num_branches = 3,
+			.branch = {
+				{
+					/* TCP */
+					.valid = 1,
+					.next_packet_pointer = 26,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_TCP_NODE,
+					.next_branch = 0,
+					.data = 0x0600,
+					.mask = 0xffff
+				},
+				{
+					/* UDP */
+					.valid = 1,
+					.next_packet_pointer = 26,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_UDP_NODE,
+					.next_branch = 0,
+					.data = 0x1100,
+					.mask = 0xffff
+				},
+				{
+					.valid = 0,
+					.next_packet_pointer = 260,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = LAST_NODE,
+					.next_branch = 0,
+					.data = 0x0,
+					.mask = 0xffff
+				}
+			}
+		},
+		{
+			/* RSS_IPV4_TCP_NODE */
+			.node_type = EWDN,
+			.last_node = 0,
+			.hdr_len_store = 1,
+			.hdr_extn = NO_BYTE,
+			.byte_store = NO_BYTE,
+			.search_byte_store = BOTH_BYTES,
+			.result_pointer = DB_RES_DROP,
+			.num_branches = 6,
+			.branch = {
+				{
+					/* SRC IPV4 B01 */
+					.valid = 0,
+					.next_packet_pointer = 28,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_TCP_NODE,
+					.next_branch = 1,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* SRC IPV4 B23 */
+					.valid = 0,
+					.next_packet_pointer = 30,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_TCP_NODE,
+					.next_branch = 2,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* DST IPV4 B01 */
+					.valid = 0,
+					.next_packet_pointer = 32,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_TCP_NODE,
+					.next_branch = 3,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* DST IPV4 B23 */
+					.valid = 0,
+					.next_packet_pointer = 34,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_TCP_NODE,
+					.next_branch = 4,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* TCP SRC Port */
+					.valid = 0,
+					.next_packet_pointer = 36,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_TCP_NODE,
+					.next_branch = 5,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* TCP DST Port */
+					.valid = 0,
+					.next_packet_pointer = 256,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = LAST_NODE,
+					.next_branch = 0,
+					.data = 0x0,
+					.mask = 0xffff
+				}
+			}
+		},
+		{
+			/* RSS_IPV4_UDP_NODE */
+			.node_type = EWDN,
+			.last_node = 0,
+			.hdr_len_store = 1,
+			.hdr_extn = NO_BYTE,
+			.byte_store = NO_BYTE,
+			.search_byte_store = BOTH_BYTES,
+			.result_pointer = DB_RES_DROP,
+			.num_branches = 6,
+			.branch = {
+				{
+					/* SRC IPV4 B01 */
+					.valid = 0,
+					.next_packet_pointer = 28,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_UDP_NODE,
+					.next_branch = 1,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* SRC IPV4 B23 */
+					.valid = 0,
+					.next_packet_pointer = 30,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_UDP_NODE,
+					.next_branch = 2,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* DST IPV4 B01 */
+					.valid = 0,
+					.next_packet_pointer = 32,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_UDP_NODE,
+					.next_branch = 3,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* DST IPV4 B23 */
+					.valid = 0,
+					.next_packet_pointer = 34,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_UDP_NODE,
+					.next_branch = 4,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* TCP SRC Port */
+					.valid = 0,
+					.next_packet_pointer = 36,
+					.jump_bw = JMP_FW,
+					.jump_rel = JMP_ABS,
+					.operation = EQT,
+					.next_node = RSS_IPV4_UDP_NODE,
+					.next_branch = 5,
+					.data = 0x0,
+					.mask = 0xffff
+				},
+				{
+					/* TCP DST Port */
+					.valid = 0,
+					.next_packet_pointer = 256,
 					.jump_bw = JMP_FW,
 					.jump_rel = JMP_ABS,
 					.operation = EQT,
@@ -294,7 +656,7 @@ static int xgene_enet_cle_init(struct xgene_enet_pdata *pdata)
 			/* LAST NODE */
 			.node_type = EWDN,
 			.last_node = 1,
-			.hdr_len_store = 0,
+			.hdr_len_store = 1,
 			.hdr_extn = NO_BYTE,
 			.byte_store = NO_BYTE,
 			.search_byte_store = NO_BYTE,
@@ -318,6 +680,20 @@ static int xgene_enet_cle_init(struct xgene_enet_pdata *pdata)
 
 	ptree = &enet_cle->ptree;
 	ptree->start_pkt = 12; /* Ethertype */
+	if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
+		ret = xgene_cle_setup_rss(pdata);
+		if (ret) {
+			netdev_err(pdata->ndev, "RSS initialization failed\n");
+			return ret;
+		}
+	} else {
+		br = &ptree_dn[PKT_PROT_NODE].branch[0];
+		br->valid = 0;
+		br->next_packet_pointer = 260;
+		br->next_node = LAST_NODE;
+		br->data = 0x0000;
+		br->mask = 0xffff;
+	}
 
 	def_qid = xgene_enet_dst_ring_num(pdata->rx_ring);
 	pool_id = pdata->rx_ring->buf_pool->id;
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
index 1db2fd7..29a17ab 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h
@@ -34,6 +34,7 @@
 #define SPPTR0			0x0104
 #define DFCLSRESDBPTR0		0x0108
 #define DFCLSRESDB00		0x010c
+#define RSS_CTRL0		0x0000013c
 
 #define CLE_CMD_TO		10	/* ms */
 #define CLE_PKTRAM_SIZE		256	/* bytes */
@@ -98,6 +99,9 @@
 
 enum xgene_cle_ptree_nodes {
 	PKT_TYPE_NODE,
+	PKT_PROT_NODE,
+	RSS_IPV4_TCP_NODE,
+	RSS_IPV4_UDP_NODE,
 	LAST_NODE,
 	MAX_NODES
 };
@@ -137,6 +141,8 @@ enum xgene_cle_parser {
 #define XGENE_CLE_DRAM(type)	(((type) & 0xf) << 28)
 enum xgene_cle_dram_type {
 	PKT_RAM,
+	RSS_IDT,
+	RSS_IPV4_HASH_SKEY,
 	PTREE_RAM = 0xc,
 	AVL_RAM,
 	DB_RAM
@@ -150,6 +156,22 @@ enum xgene_cle_cmd_type {
 	CLE_CMD_AVL_SRCH = 32
 };
 
+enum xgene_cle_ipv4_rss_hashtype {
+	RSS_IPV4_8B,
+	RSS_IPV4_12B,
+};
+
+enum xgene_cle_prot_type {
+	XGENE_CLE_TCP,
+	XGENE_CLE_UDP,
+	XGENE_CLE_ESP,
+	XGENE_CLE_OTHER
+};
+
+enum xgene_cle_prot_version {
+	XGENE_CLE_IPV4,
+};
+
 enum xgene_cle_ptree_dbptrs {
 	DB_RES_DROP,
 	DB_RES_DEF,
@@ -157,6 +179,25 @@ enum xgene_cle_ptree_dbptrs {
 	DB_MAX_PTRS
 };
 
+/* RSS sideband signal info */
+#define SB_IPFRAG_POS	0
+#define SB_IPFRAG_LEN	1
+#define SB_IPPROT_POS	1
+#define SB_IPPROT_LEN	2
+#define SB_IPVER_POS	3
+#define SB_IPVER_LEN	1
+#define SB_HDRLEN_POS	4
+#define SB_HDRLEN_LEN	12
+
+/* RSS indirection table */
+#define XGENE_CLE_IDT_ENTRIES	128
+#define IDT_DSTQID_POS		0
+#define IDT_DSTQID_LEN		12
+#define IDT_FPSEL_POS		12
+#define IDT_FPSEL_LEN		4
+#define IDT_NFPSEL_POS		16
+#define IDT_NFPSEL_LEN		4
+
 struct xgene_cle_ptree_branch {
 	bool valid;
 	u16 next_packet_pointer;
-- 
cgit v0.10.2


From 107dec2749fed1b4172f77c091961b15221419db Mon Sep 17 00:00:00 2001
From: Iyappan Subramanian <isubramanian@apm.com>
Date: Wed, 17 Feb 2016 15:00:41 -0800
Subject: drivers: net: xgene: Add support for multiple queues

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Khuong Dinh <kdinh@apm.com>
Signed-off-by: Tanmay Inamdar <tinamdar@apm.com>
Tested-by: Toan Le <toanle@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
index c007497..b2124886 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c
@@ -331,14 +331,15 @@ static int xgene_cle_set_rss_skeys(struct xgene_enet_cle *cle)
 
 static int xgene_cle_set_rss_idt(struct xgene_enet_pdata *pdata)
 {
-	u32 fpsel, dstqid, nfpsel, idt_reg;
+	u32 fpsel, dstqid, nfpsel, idt_reg, idx;
 	int i, ret = 0;
 	u16 pool_id;
 
 	for (i = 0; i < XGENE_CLE_IDT_ENTRIES; i++) {
-		pool_id = pdata->rx_ring->buf_pool->id;
+		idx = i % pdata->rxq_cnt;
+		pool_id = pdata->rx_ring[idx]->buf_pool->id;
 		fpsel = xgene_enet_ring_bufnum(pool_id) - 0x20;
-		dstqid = xgene_enet_dst_ring_num(pdata->rx_ring);
+		dstqid = xgene_enet_dst_ring_num(pdata->rx_ring[idx]);
 		nfpsel = 0;
 		idt_reg = 0;
 
@@ -695,8 +696,8 @@ static int xgene_enet_cle_init(struct xgene_enet_pdata *pdata)
 		br->mask = 0xffff;
 	}
 
-	def_qid = xgene_enet_dst_ring_num(pdata->rx_ring);
-	pool_id = pdata->rx_ring->buf_pool->id;
+	def_qid = xgene_enet_dst_ring_num(pdata->rx_ring[0]);
+	pool_id = pdata->rx_ring[0]->buf_pool->id;
 	def_fpsel = xgene_enet_ring_bufnum(pool_id) - 0x20;
 
 	memset(dbptr, 0, sizeof(struct xgene_cle_dbptr) * DB_MAX_PTRS);
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
index db55c9f..39e081a 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c
@@ -204,6 +204,17 @@ static u32 xgene_enet_ring_len(struct xgene_enet_desc_ring *ring)
 	return num_msgs;
 }
 
+static void xgene_enet_setup_coalescing(struct xgene_enet_desc_ring *ring)
+{
+	u32 data = 0x7777;
+
+	xgene_enet_ring_wr32(ring, CSR_PBM_COAL, 0x8e);
+	xgene_enet_ring_wr32(ring, CSR_PBM_CTICK1, data);
+	xgene_enet_ring_wr32(ring, CSR_PBM_CTICK2, data << 16);
+	xgene_enet_ring_wr32(ring, CSR_THRESHOLD0_SET1, 0x40);
+	xgene_enet_ring_wr32(ring, CSR_THRESHOLD1_SET1, 0x80);
+}
+
 void xgene_enet_parse_error(struct xgene_enet_desc_ring *ring,
 			    struct xgene_enet_pdata *pdata,
 			    enum xgene_enet_err_code status)
@@ -892,4 +903,5 @@ struct xgene_ring_ops xgene_ring1_ops = {
 	.clear = xgene_enet_clear_ring,
 	.wr_cmd = xgene_enet_wr_cmd,
 	.len = xgene_enet_ring_len,
+	.coalesce = xgene_enet_setup_coalescing,
 };
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
index 45725ec..ba7da98 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h
@@ -54,6 +54,11 @@ enum xgene_enet_rm {
 #define IS_BUFFER_POOL		BIT(20)
 #define PREFETCH_BUF_EN		BIT(21)
 #define CSR_RING_ID_BUF		0x000c
+#define CSR_PBM_COAL		0x0014
+#define CSR_PBM_CTICK1		0x001c
+#define CSR_PBM_CTICK2		0x0020
+#define CSR_THRESHOLD0_SET1	0x0030
+#define CSR_THRESHOLD1_SET1	0x0034
 #define CSR_RING_NE_INT_MODE	0x017c
 #define CSR_RING_CONFIG		0x006c
 #define CSR_RING_WR_BASE	0x0070
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index 0bf3924..8d4c1ad 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -182,7 +182,6 @@ static int xgene_enet_tx_completion(struct xgene_enet_desc_ring *cp_ring,
 static u64 xgene_enet_work_msg(struct sk_buff *skb)
 {
 	struct net_device *ndev = skb->dev;
-	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	struct iphdr *iph;
 	u8 l3hlen = 0, l4hlen = 0;
 	u8 ethhdr, proto = 0, csum_enable = 0;
@@ -228,10 +227,6 @@ static u64 xgene_enet_work_msg(struct sk_buff *skb)
 			if (!mss || ((skb->len - hdr_len) <= mss))
 				goto out;
 
-			if (mss != pdata->mss) {
-				pdata->mss = mss;
-				pdata->mac_ops->set_mss(pdata);
-			}
 			hopinfo |= SET_BIT(ET);
 		}
 	} else if (iph->protocol == IPPROTO_UDP) {
@@ -413,7 +408,7 @@ out:
 	raw_desc->m0 = cpu_to_le64(SET_VAL(LL, ll) | SET_VAL(NV, nv) |
 				   SET_VAL(USERINFO, tx_ring->tail));
 	tx_ring->cp_ring->cp_skb[tx_ring->tail] = skb;
-	pdata->tx_level += count;
+	pdata->tx_level[tx_ring->cp_ring->index] += count;
 	tx_ring->tail = tail;
 
 	return count;
@@ -423,15 +418,17 @@ static netdev_tx_t xgene_enet_start_xmit(struct sk_buff *skb,
 					 struct net_device *ndev)
 {
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
-	struct xgene_enet_desc_ring *tx_ring = pdata->tx_ring;
-	u32 tx_level = pdata->tx_level;
+	struct xgene_enet_desc_ring *tx_ring;
+	int index = skb->queue_mapping;
+	u32 tx_level = pdata->tx_level[index];
 	int count;
 
-	if (tx_level < pdata->txc_level)
-		tx_level += ((typeof(pdata->tx_level))~0U);
+	tx_ring = pdata->tx_ring[index];
+	if (tx_level < pdata->txc_level[index])
+		tx_level += ((typeof(pdata->tx_level[index]))~0U);
 
-	if ((tx_level - pdata->txc_level) > pdata->tx_qcnt_hi) {
-		netif_stop_queue(ndev);
+	if ((tx_level - pdata->txc_level[index]) > pdata->tx_qcnt_hi) {
+		netif_stop_subqueue(ndev, index);
 		return NETDEV_TX_BUSY;
 	}
 
@@ -529,7 +526,8 @@ static bool is_rx_desc(struct xgene_enet_raw_desc *raw_desc)
 static int xgene_enet_process_ring(struct xgene_enet_desc_ring *ring,
 				   int budget)
 {
-	struct xgene_enet_pdata *pdata = netdev_priv(ring->ndev);
+	struct net_device *ndev = ring->ndev;
+	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	struct xgene_enet_raw_desc *raw_desc, *exp_desc;
 	u16 head = ring->head;
 	u16 slots = ring->slots - 1;
@@ -573,7 +571,7 @@ static int xgene_enet_process_ring(struct xgene_enet_desc_ring *ring,
 		desc_count++;
 		processed++;
 		if (is_completion)
-			pdata->txc_level += desc_count;
+			pdata->txc_level[ring->index] += desc_count;
 
 		if (ret)
 			break;
@@ -583,8 +581,8 @@ static int xgene_enet_process_ring(struct xgene_enet_desc_ring *ring,
 		pdata->ring_ops->wr_cmd(ring, -count);
 		ring->head = head;
 
-		if (netif_queue_stopped(ring->ndev))
-			netif_start_queue(ring->ndev);
+		if (__netif_subqueue_stopped(ndev, ring->index))
+			netif_start_subqueue(ndev, ring->index);
 	}
 
 	return processed;
@@ -609,8 +607,16 @@ static int xgene_enet_napi(struct napi_struct *napi, const int budget)
 static void xgene_enet_timeout(struct net_device *ndev)
 {
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
+	struct netdev_queue *txq;
+	int i;
 
 	pdata->mac_ops->reset(pdata);
+
+	for (i = 0; i < pdata->txq_cnt; i++) {
+		txq = netdev_get_tx_queue(ndev, i);
+		txq->trans_start = jiffies;
+		netif_tx_start_queue(txq);
+	}
 }
 
 static int xgene_enet_register_irq(struct net_device *ndev)
@@ -618,17 +624,21 @@ static int xgene_enet_register_irq(struct net_device *ndev)
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	struct device *dev = ndev_to_dev(ndev);
 	struct xgene_enet_desc_ring *ring;
-	int ret;
+	int ret = 0, i;
 
-	ring = pdata->rx_ring;
-	irq_set_status_flags(ring->irq, IRQ_DISABLE_UNLAZY);
-	ret = devm_request_irq(dev, ring->irq, xgene_enet_rx_irq,
-			       IRQF_SHARED, ring->irq_name, ring);
-	if (ret)
-		netdev_err(ndev, "Failed to request irq %s\n", ring->irq_name);
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		ring = pdata->rx_ring[i];
+		irq_set_status_flags(ring->irq, IRQ_DISABLE_UNLAZY);
+		ret = devm_request_irq(dev, ring->irq, xgene_enet_rx_irq,
+				       IRQF_SHARED, ring->irq_name, ring);
+		if (ret) {
+			netdev_err(ndev, "Failed to request irq %s\n",
+				   ring->irq_name);
+		}
+	}
 
-	if (pdata->cq_cnt) {
-		ring = pdata->tx_ring->cp_ring;
+	for (i = 0; i < pdata->cq_cnt; i++) {
+		ring = pdata->tx_ring[i]->cp_ring;
 		irq_set_status_flags(ring->irq, IRQ_DISABLE_UNLAZY);
 		ret = devm_request_irq(dev, ring->irq, xgene_enet_rx_irq,
 				       IRQF_SHARED, ring->irq_name, ring);
@@ -646,15 +656,19 @@ static void xgene_enet_free_irq(struct net_device *ndev)
 	struct xgene_enet_pdata *pdata;
 	struct xgene_enet_desc_ring *ring;
 	struct device *dev;
+	int i;
 
 	pdata = netdev_priv(ndev);
 	dev = ndev_to_dev(ndev);
-	ring = pdata->rx_ring;
-	irq_clear_status_flags(ring->irq, IRQ_DISABLE_UNLAZY);
-	devm_free_irq(dev, ring->irq, ring);
 
-	if (pdata->cq_cnt) {
-		ring = pdata->tx_ring->cp_ring;
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		ring = pdata->rx_ring[i];
+		irq_clear_status_flags(ring->irq, IRQ_DISABLE_UNLAZY);
+		devm_free_irq(dev, ring->irq, ring);
+	}
+
+	for (i = 0; i < pdata->cq_cnt; i++) {
+		ring = pdata->tx_ring[i]->cp_ring;
 		irq_clear_status_flags(ring->irq, IRQ_DISABLE_UNLAZY);
 		devm_free_irq(dev, ring->irq, ring);
 	}
@@ -663,12 +677,15 @@ static void xgene_enet_free_irq(struct net_device *ndev)
 static void xgene_enet_napi_enable(struct xgene_enet_pdata *pdata)
 {
 	struct napi_struct *napi;
+	int i;
 
-	napi = &pdata->rx_ring->napi;
-	napi_enable(napi);
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		napi = &pdata->rx_ring[i]->napi;
+		napi_enable(napi);
+	}
 
-	if (pdata->cq_cnt) {
-		napi = &pdata->tx_ring->cp_ring->napi;
+	for (i = 0; i < pdata->cq_cnt; i++) {
+		napi = &pdata->tx_ring[i]->cp_ring->napi;
 		napi_enable(napi);
 	}
 }
@@ -676,12 +693,15 @@ static void xgene_enet_napi_enable(struct xgene_enet_pdata *pdata)
 static void xgene_enet_napi_disable(struct xgene_enet_pdata *pdata)
 {
 	struct napi_struct *napi;
+	int i;
 
-	napi = &pdata->rx_ring->napi;
-	napi_disable(napi);
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		napi = &pdata->rx_ring[i]->napi;
+		napi_disable(napi);
+	}
 
-	if (pdata->cq_cnt) {
-		napi = &pdata->tx_ring->cp_ring->napi;
+	for (i = 0; i < pdata->cq_cnt; i++) {
+		napi = &pdata->tx_ring[i]->cp_ring->napi;
 		napi_disable(napi);
 	}
 }
@@ -692,6 +712,14 @@ static int xgene_enet_open(struct net_device *ndev)
 	const struct xgene_mac_ops *mac_ops = pdata->mac_ops;
 	int ret;
 
+	ret = netif_set_real_num_tx_queues(ndev, pdata->txq_cnt);
+	if (ret)
+		return ret;
+
+	ret = netif_set_real_num_rx_queues(ndev, pdata->rxq_cnt);
+	if (ret)
+		return ret;
+
 	mac_ops->tx_enable(pdata);
 	mac_ops->rx_enable(pdata);
 
@@ -714,6 +742,7 @@ static int xgene_enet_close(struct net_device *ndev)
 {
 	struct xgene_enet_pdata *pdata = netdev_priv(ndev);
 	const struct xgene_mac_ops *mac_ops = pdata->mac_ops;
+	int i;
 
 	netif_stop_queue(ndev);
 
@@ -727,7 +756,8 @@ static int xgene_enet_close(struct net_device *ndev)
 
 	xgene_enet_free_irq(ndev);
 	xgene_enet_napi_disable(pdata);
-	xgene_enet_process_ring(pdata->rx_ring, -1);
+	for (i = 0; i < pdata->rxq_cnt; i++)
+		xgene_enet_process_ring(pdata->rx_ring[i], -1);
 
 	return 0;
 }
@@ -747,18 +777,26 @@ static void xgene_enet_delete_ring(struct xgene_enet_desc_ring *ring)
 static void xgene_enet_delete_desc_rings(struct xgene_enet_pdata *pdata)
 {
 	struct xgene_enet_desc_ring *buf_pool;
+	struct xgene_enet_desc_ring *ring;
+	int i;
 
-	if (pdata->tx_ring) {
-		xgene_enet_delete_ring(pdata->tx_ring);
-		pdata->tx_ring = NULL;
+	for (i = 0; i < pdata->txq_cnt; i++) {
+		ring = pdata->tx_ring[i];
+		if (ring) {
+			xgene_enet_delete_ring(ring);
+			pdata->tx_ring[i] = NULL;
+		}
 	}
 
-	if (pdata->rx_ring) {
-		buf_pool = pdata->rx_ring->buf_pool;
-		xgene_enet_delete_bufpool(buf_pool);
-		xgene_enet_delete_ring(buf_pool);
-		xgene_enet_delete_ring(pdata->rx_ring);
-		pdata->rx_ring = NULL;
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		ring = pdata->rx_ring[i];
+		if (ring) {
+			buf_pool = ring->buf_pool;
+			xgene_enet_delete_bufpool(buf_pool);
+			xgene_enet_delete_ring(buf_pool);
+			xgene_enet_delete_ring(ring);
+			pdata->rx_ring[i] = NULL;
+		}
 	}
 }
 
@@ -813,24 +851,29 @@ static void xgene_enet_free_desc_rings(struct xgene_enet_pdata *pdata)
 {
 	struct device *dev = &pdata->pdev->dev;
 	struct xgene_enet_desc_ring *ring;
+	int i;
 
-	ring = pdata->tx_ring;
-	if (ring) {
-		if (ring->cp_ring && ring->cp_ring->cp_skb)
-			devm_kfree(dev, ring->cp_ring->cp_skb);
-		if (ring->cp_ring && pdata->cq_cnt)
-			xgene_enet_free_desc_ring(ring->cp_ring);
-		xgene_enet_free_desc_ring(ring);
-	}
-
-	ring = pdata->rx_ring;
-	if (ring) {
-		if (ring->buf_pool) {
-			if (ring->buf_pool->rx_skb)
-				devm_kfree(dev, ring->buf_pool->rx_skb);
-			xgene_enet_free_desc_ring(ring->buf_pool);
+	for (i = 0; i < pdata->txq_cnt; i++) {
+		ring = pdata->tx_ring[i];
+		if (ring) {
+			if (ring->cp_ring && ring->cp_ring->cp_skb)
+				devm_kfree(dev, ring->cp_ring->cp_skb);
+			if (ring->cp_ring && pdata->cq_cnt)
+				xgene_enet_free_desc_ring(ring->cp_ring);
+			xgene_enet_free_desc_ring(ring);
+		}
+	}
+
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		ring = pdata->rx_ring[i];
+		if (ring) {
+			if (ring->buf_pool) {
+				if (ring->buf_pool->rx_skb)
+					devm_kfree(dev, ring->buf_pool->rx_skb);
+				xgene_enet_free_desc_ring(ring->buf_pool);
+			}
+			xgene_enet_free_desc_ring(ring);
 		}
-		xgene_enet_free_desc_ring(ring);
 	}
 }
 
@@ -943,104 +986,120 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev)
 	u8 bp_bufnum = pdata->bp_bufnum;
 	u16 ring_num = pdata->ring_num;
 	u16 ring_id;
-	int ret, size;
-
-	/* allocate rx descriptor ring */
-	owner = xgene_derive_ring_owner(pdata);
-	ring_id = xgene_enet_get_ring_id(RING_OWNER_CPU, cpu_bufnum++);
-	rx_ring = xgene_enet_create_desc_ring(ndev, ring_num++,
-					      RING_CFGSIZE_16KB, ring_id);
-	if (!rx_ring) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	/* allocate buffer pool for receiving packets */
-	owner = xgene_derive_ring_owner(pdata);
-	ring_id = xgene_enet_get_ring_id(owner, bp_bufnum++);
-	buf_pool = xgene_enet_create_desc_ring(ndev, ring_num++,
-					       RING_CFGSIZE_2KB, ring_id);
-	if (!buf_pool) {
-		ret = -ENOMEM;
-		goto err;
-	}
+	int i, ret, size;
 
-	rx_ring->nbufpool = NUM_BUFPOOL;
-	rx_ring->buf_pool = buf_pool;
-	rx_ring->irq = pdata->rx_irq;
-	if (!pdata->cq_cnt) {
-		snprintf(rx_ring->irq_name, IRQ_ID_SIZE, "%s-rx-txc",
-			 ndev->name);
-	} else {
-		snprintf(rx_ring->irq_name, IRQ_ID_SIZE, "%s-rx", ndev->name);
-	}
-	buf_pool->rx_skb = devm_kcalloc(dev, buf_pool->slots,
-					sizeof(struct sk_buff *), GFP_KERNEL);
-	if (!buf_pool->rx_skb) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	buf_pool->dst_ring_num = xgene_enet_dst_ring_num(buf_pool);
-	rx_ring->buf_pool = buf_pool;
-	pdata->rx_ring = rx_ring;
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		/* allocate rx descriptor ring */
+		owner = xgene_derive_ring_owner(pdata);
+		ring_id = xgene_enet_get_ring_id(RING_OWNER_CPU, cpu_bufnum++);
+		rx_ring = xgene_enet_create_desc_ring(ndev, ring_num++,
+						      RING_CFGSIZE_16KB,
+						      ring_id);
+		if (!rx_ring) {
+			ret = -ENOMEM;
+			goto err;
+		}
 
-	/* allocate tx descriptor ring */
-	owner = xgene_derive_ring_owner(pdata);
-	ring_id = xgene_enet_get_ring_id(owner, eth_bufnum++);
-	tx_ring = xgene_enet_create_desc_ring(ndev, ring_num++,
-					      RING_CFGSIZE_16KB, ring_id);
-	if (!tx_ring) {
-		ret = -ENOMEM;
-		goto err;
-	}
+		/* allocate buffer pool for receiving packets */
+		owner = xgene_derive_ring_owner(pdata);
+		ring_id = xgene_enet_get_ring_id(owner, bp_bufnum++);
+		buf_pool = xgene_enet_create_desc_ring(ndev, ring_num++,
+						       RING_CFGSIZE_2KB,
+						       ring_id);
+		if (!buf_pool) {
+			ret = -ENOMEM;
+			goto err;
+		}
 
-	size = (tx_ring->slots / 2) * sizeof(__le64) * MAX_EXP_BUFFS;
-	tx_ring->exp_bufs = dma_zalloc_coherent(dev, size, &dma_exp_bufs,
+		rx_ring->nbufpool = NUM_BUFPOOL;
+		rx_ring->buf_pool = buf_pool;
+		rx_ring->irq = pdata->irqs[i];
+		if (!pdata->cq_cnt) {
+			snprintf(rx_ring->irq_name, IRQ_ID_SIZE, "%s-rx-txc",
+				 ndev->name);
+		} else {
+			snprintf(rx_ring->irq_name, IRQ_ID_SIZE, "%s-rx%d",
+				 ndev->name, i);
+		}
+		buf_pool->rx_skb = devm_kcalloc(dev, buf_pool->slots,
+						sizeof(struct sk_buff *),
 						GFP_KERNEL);
-	if (!tx_ring->exp_bufs) {
-		ret = -ENOMEM;
-		goto err;
-	}
+		if (!buf_pool->rx_skb) {
+			ret = -ENOMEM;
+			goto err;
+		}
 
-	pdata->tx_ring = tx_ring;
+		buf_pool->dst_ring_num = xgene_enet_dst_ring_num(buf_pool);
+		rx_ring->buf_pool = buf_pool;
+		pdata->rx_ring[i] = rx_ring;
+	}
 
-	if (!pdata->cq_cnt) {
-		cp_ring = pdata->rx_ring;
-	} else {
-		/* allocate tx completion descriptor ring */
-		ring_id = xgene_enet_get_ring_id(RING_OWNER_CPU, cpu_bufnum++);
-		cp_ring = xgene_enet_create_desc_ring(ndev, ring_num++,
+	for (i = 0; i < pdata->txq_cnt; i++) {
+		/* allocate tx descriptor ring */
+		owner = xgene_derive_ring_owner(pdata);
+		ring_id = xgene_enet_get_ring_id(owner, eth_bufnum++);
+		tx_ring = xgene_enet_create_desc_ring(ndev, ring_num++,
 						      RING_CFGSIZE_16KB,
 						      ring_id);
-		if (!cp_ring) {
+		if (!tx_ring) {
 			ret = -ENOMEM;
 			goto err;
 		}
-		cp_ring->irq = pdata->txc_irq;
-		snprintf(cp_ring->irq_name, IRQ_ID_SIZE, "%s-txc", ndev->name);
-	}
 
-	cp_ring->cp_skb = devm_kcalloc(dev, tx_ring->slots,
-				       sizeof(struct sk_buff *), GFP_KERNEL);
-	if (!cp_ring->cp_skb) {
-		ret = -ENOMEM;
-		goto err;
-	}
+		size = (tx_ring->slots / 2) * sizeof(__le64) * MAX_EXP_BUFFS;
+		tx_ring->exp_bufs = dma_zalloc_coherent(dev, size,
+							&dma_exp_bufs,
+							GFP_KERNEL);
+		if (!tx_ring->exp_bufs) {
+			ret = -ENOMEM;
+			goto err;
+		}
 
-	size = sizeof(dma_addr_t) * MAX_SKB_FRAGS;
-	cp_ring->frag_dma_addr = devm_kcalloc(dev, tx_ring->slots,
-					      size, GFP_KERNEL);
-	if (!cp_ring->frag_dma_addr) {
-		devm_kfree(dev, cp_ring->cp_skb);
-		ret = -ENOMEM;
-		goto err;
-	}
+		pdata->tx_ring[i] = tx_ring;
 
-	pdata->tx_ring->cp_ring = cp_ring;
-	pdata->tx_ring->dst_ring_num = xgene_enet_dst_ring_num(cp_ring);
+		if (!pdata->cq_cnt) {
+			cp_ring = pdata->rx_ring[i];
+		} else {
+			/* allocate tx completion descriptor ring */
+			ring_id = xgene_enet_get_ring_id(RING_OWNER_CPU,
+							 cpu_bufnum++);
+			cp_ring = xgene_enet_create_desc_ring(ndev, ring_num++,
+							      RING_CFGSIZE_16KB,
+							      ring_id);
+			if (!cp_ring) {
+				ret = -ENOMEM;
+				goto err;
+			}
+
+			cp_ring->irq = pdata->irqs[pdata->rxq_cnt + i];
+			cp_ring->index = i;
+			snprintf(cp_ring->irq_name, IRQ_ID_SIZE, "%s-txc%d",
+				 ndev->name, i);
+		}
+
+		cp_ring->cp_skb = devm_kcalloc(dev, tx_ring->slots,
+					       sizeof(struct sk_buff *),
+					       GFP_KERNEL);
+		if (!cp_ring->cp_skb) {
+			ret = -ENOMEM;
+			goto err;
+		}
 
-	pdata->tx_qcnt_hi = pdata->tx_ring->slots - 128;
+		size = sizeof(dma_addr_t) * MAX_SKB_FRAGS;
+		cp_ring->frag_dma_addr = devm_kcalloc(dev, tx_ring->slots,
+						      size, GFP_KERNEL);
+		if (!cp_ring->frag_dma_addr) {
+			devm_kfree(dev, cp_ring->cp_skb);
+			ret = -ENOMEM;
+			goto err;
+		}
+
+		tx_ring->cp_ring = cp_ring;
+		tx_ring->dst_ring_num = xgene_enet_dst_ring_num(cp_ring);
+	}
+
+	pdata->ring_ops->coalesce(pdata->tx_ring[0]);
+	pdata->tx_qcnt_hi = pdata->tx_ring[0]->slots - 128;
 
 	return 0;
 
@@ -1159,6 +1218,32 @@ static int xgene_get_rx_delay(struct xgene_enet_pdata *pdata)
 	return 0;
 }
 
+static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata)
+{
+	struct platform_device *pdev = pdata->pdev;
+	struct device *dev = &pdev->dev;
+	int i, ret, max_irqs;
+
+	if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII)
+		max_irqs = 1;
+	else if (pdata->phy_mode == PHY_INTERFACE_MODE_SGMII)
+		max_irqs = 2;
+	else
+		max_irqs = XGENE_MAX_ENET_IRQ;
+
+	for (i = 0; i < max_irqs; i++) {
+		ret = platform_get_irq(pdev, i);
+		if (ret <= 0) {
+			dev_err(dev, "Unable to get ENET IRQ\n");
+			ret = ret ? : -ENXIO;
+			return ret;
+		}
+		pdata->irqs[i] = ret;
+	}
+
+	return 0;
+}
+
 static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
 {
 	struct platform_device *pdev;
@@ -1240,25 +1325,9 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata)
 	if (ret)
 		return ret;
 
-	ret = platform_get_irq(pdev, 0);
-	if (ret <= 0) {
-		dev_err(dev, "Unable to get ENET Rx IRQ\n");
-		ret = ret ? : -ENXIO;
+	ret = xgene_enet_get_irqs(pdata);
+	if (ret)
 		return ret;
-	}
-	pdata->rx_irq = ret;
-
-	if (pdata->phy_mode != PHY_INTERFACE_MODE_RGMII) {
-		ret = platform_get_irq(pdev, 1);
-		if (ret <= 0) {
-			pdata->cq_cnt = 0;
-			dev_info(dev, "Unable to get Tx completion IRQ,"
-				 "using Rx IRQ instead\n");
-		} else {
-			pdata->cq_cnt = XGENE_MAX_TXC_RINGS;
-			pdata->txc_irq = ret;
-		}
-	}
 
 	pdata->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(pdata->clk)) {
@@ -1296,7 +1365,7 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata)
 	struct net_device *ndev = pdata->ndev;
 	struct xgene_enet_desc_ring *buf_pool;
 	u16 dst_ring_num;
-	int ret;
+	int i, ret;
 
 	ret = pdata->port_ops->reset(pdata);
 	if (ret)
@@ -1309,15 +1378,18 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata)
 	}
 
 	/* setup buffer pool */
-	buf_pool = pdata->rx_ring->buf_pool;
-	xgene_enet_init_bufpool(buf_pool);
-	ret = xgene_enet_refill_bufpool(buf_pool, pdata->rx_buff_cnt);
-	if (ret) {
-		xgene_enet_delete_desc_rings(pdata);
-		return ret;
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		buf_pool = pdata->rx_ring[i]->buf_pool;
+		xgene_enet_init_bufpool(buf_pool);
+		ret = xgene_enet_refill_bufpool(buf_pool, pdata->rx_buff_cnt);
+		if (ret) {
+			xgene_enet_delete_desc_rings(pdata);
+			return ret;
+		}
 	}
 
-	dst_ring_num = xgene_enet_dst_ring_num(pdata->rx_ring);
+	dst_ring_num = xgene_enet_dst_ring_num(pdata->rx_ring[0]);
+	buf_pool = pdata->rx_ring[0]->buf_pool;
 	if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
 		/* Initialize and Enable  PreClassifier Tree */
 		enet_cle->max_nodes = 512;
@@ -1348,17 +1420,26 @@ static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata)
 		pdata->mac_ops = &xgene_gmac_ops;
 		pdata->port_ops = &xgene_gport_ops;
 		pdata->rm = RM3;
+		pdata->rxq_cnt = 1;
+		pdata->txq_cnt = 1;
+		pdata->cq_cnt = 0;
 		break;
 	case PHY_INTERFACE_MODE_SGMII:
 		pdata->mac_ops = &xgene_sgmac_ops;
 		pdata->port_ops = &xgene_sgport_ops;
 		pdata->rm = RM1;
+		pdata->rxq_cnt = 1;
+		pdata->txq_cnt = 1;
+		pdata->cq_cnt = 1;
 		break;
 	default:
 		pdata->mac_ops = &xgene_xgmac_ops;
 		pdata->port_ops = &xgene_xgport_ops;
 		pdata->cle_ops = &xgene_cle3in_ops;
 		pdata->rm = RM0;
+		pdata->rxq_cnt = XGENE_NUM_RX_RING;
+		pdata->txq_cnt = XGENE_NUM_TX_RING;
+		pdata->cq_cnt = XGENE_NUM_TXC_RING;
 		break;
 	}
 
@@ -1412,12 +1493,16 @@ static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata)
 static void xgene_enet_napi_add(struct xgene_enet_pdata *pdata)
 {
 	struct napi_struct *napi;
+	int i;
 
-	napi = &pdata->rx_ring->napi;
-	netif_napi_add(pdata->ndev, napi, xgene_enet_napi, NAPI_POLL_WEIGHT);
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		napi = &pdata->rx_ring[i]->napi;
+		netif_napi_add(pdata->ndev, napi, xgene_enet_napi,
+			       NAPI_POLL_WEIGHT);
+	}
 
-	if (pdata->cq_cnt) {
-		napi = &pdata->tx_ring->cp_ring->napi;
+	for (i = 0; i < pdata->cq_cnt; i++) {
+		napi = &pdata->tx_ring[i]->cp_ring->napi;
 		netif_napi_add(pdata->ndev, napi, xgene_enet_napi,
 			       NAPI_POLL_WEIGHT);
 	}
@@ -1426,12 +1511,15 @@ static void xgene_enet_napi_add(struct xgene_enet_pdata *pdata)
 static void xgene_enet_napi_del(struct xgene_enet_pdata *pdata)
 {
 	struct napi_struct *napi;
+	int i;
 
-	napi = &pdata->rx_ring->napi;
-	netif_napi_del(napi);
+	for (i = 0; i < pdata->rxq_cnt; i++) {
+		napi = &pdata->rx_ring[i]->napi;
+		netif_napi_del(napi);
+	}
 
-	if (pdata->cq_cnt) {
-		napi = &pdata->tx_ring->cp_ring->napi;
+	for (i = 0; i < pdata->cq_cnt; i++) {
+		napi = &pdata->tx_ring[i]->cp_ring->napi;
 		netif_napi_del(napi);
 	}
 }
@@ -1445,7 +1533,8 @@ static int xgene_enet_probe(struct platform_device *pdev)
 	const struct of_device_id *of_id;
 	int ret;
 
-	ndev = alloc_etherdev(sizeof(struct xgene_enet_pdata));
+	ndev = alloc_etherdev_mqs(sizeof(struct xgene_enet_pdata),
+				  XGENE_NUM_RX_RING, XGENE_NUM_TX_RING);
 	if (!ndev)
 		return -ENOMEM;
 
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
index 05365c1..175d188 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
@@ -49,6 +49,11 @@
 #define XGENE_ENET_MSS	1448
 #define XGENE_MIN_ENET_FRAME_SIZE	60
 
+#define XGENE_MAX_ENET_IRQ	8
+#define XGENE_NUM_RX_RING	4
+#define XGENE_NUM_TX_RING	4
+#define XGENE_NUM_TXC_RING	4
+
 #define START_CPU_BUFNUM_0	0
 #define START_ETH_BUFNUM_0	2
 #define START_BP_BUFNUM_0	0x22
@@ -73,7 +78,6 @@
 #define X2_START_RING_NUM_1	256
 
 #define IRQ_ID_SIZE		16
-#define XGENE_MAX_TXC_RINGS	1
 
 #define PHY_POLL_LINK_ON	(10 * HZ)
 #define PHY_POLL_LINK_OFF	(PHY_POLL_LINK_ON / 5)
@@ -103,6 +107,7 @@ struct xgene_enet_desc_ring {
 	void *irq_mbox_addr;
 	u16 dst_ring_num;
 	u8 nbufpool;
+	u8 index;
 	struct sk_buff *(*rx_skb);
 	struct sk_buff *(*cp_skb);
 	dma_addr_t *frag_dma_addr;
@@ -144,6 +149,7 @@ struct xgene_ring_ops {
 	void (*clear)(struct xgene_enet_desc_ring *);
 	void (*wr_cmd)(struct xgene_enet_desc_ring *, int);
 	u32 (*len)(struct xgene_enet_desc_ring *);
+	void (*coalesce)(struct xgene_enet_desc_ring *);
 };
 
 struct xgene_cle_ops {
@@ -159,15 +165,16 @@ struct xgene_enet_pdata {
 	struct clk *clk;
 	struct platform_device *pdev;
 	enum xgene_enet_id enet_id;
-	struct xgene_enet_desc_ring *tx_ring;
-	struct xgene_enet_desc_ring *rx_ring;
-	u16 tx_level;
-	u16 txc_level;
+	struct xgene_enet_desc_ring *tx_ring[XGENE_NUM_TX_RING];
+	struct xgene_enet_desc_ring *rx_ring[XGENE_NUM_RX_RING];
+	u16 tx_level[XGENE_NUM_TX_RING];
+	u16 txc_level[XGENE_NUM_TX_RING];
 	char *dev_name;
 	u32 rx_buff_cnt;
 	u32 tx_qcnt_hi;
-	u32 rx_irq;
-	u32 txc_irq;
+	u32 irqs[XGENE_MAX_ENET_IRQ];
+	u8 rxq_cnt;
+	u8 txq_cnt;
 	u8 cq_cnt;
 	void __iomem *eth_csr_addr;
 	void __iomem *eth_ring_if_addr;
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c
index 0b6896b..2b76732 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ring2.c
@@ -190,6 +190,17 @@ static u32 xgene_enet_ring_len(struct xgene_enet_desc_ring *ring)
 	return num_msgs;
 }
 
+static void xgene_enet_setup_coalescing(struct xgene_enet_desc_ring *ring)
+{
+	u32 data = 0x7777;
+
+	xgene_enet_ring_wr32(ring, CSR_PBM_COAL, 0x8e);
+	xgene_enet_ring_wr32(ring, CSR_PBM_CTICK1, data);
+	xgene_enet_ring_wr32(ring, CSR_PBM_CTICK2, data << 16);
+	xgene_enet_ring_wr32(ring, CSR_THRESHOLD0_SET1, 0x40);
+	xgene_enet_ring_wr32(ring, CSR_THRESHOLD1_SET1, 0x80);
+}
+
 struct xgene_ring_ops xgene_ring2_ops = {
 	.num_ring_config = X2_NUM_RING_CONFIG,
 	.num_ring_id_shift = 13,
@@ -197,4 +208,5 @@ struct xgene_ring_ops xgene_ring2_ops = {
 	.clear = xgene_enet_clear_ring,
 	.wr_cmd = xgene_enet_wr_cmd,
 	.len = xgene_enet_ring_len,
+	.coalesce = xgene_enet_setup_coalescing,
 };
-- 
cgit v0.10.2


From 0d2c2515b84530ed60f50c2b464d14b6ba7aa1a5 Mon Sep 17 00:00:00 2001
From: Iyappan Subramanian <isubramanian@apm.com>
Date: Wed, 17 Feb 2016 15:00:42 -0800
Subject: dtb: xgene: Add irqs to support multi queue

Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Signed-off-by: Khuong Dinh <kdinh@apm.com>
Signed-off-by: Tanmay Inamdar <tinamdar@apm.com>
Tested-by: Toan Le <toanle@apm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi b/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi
index 5d87a3d..278f106 100644
--- a/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi
+++ b/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi
@@ -621,7 +621,13 @@
 			      <0x0 0x1f600000 0x0 0Xd100>,
 			      <0x0 0x20000000 0x0 0X220000>;
 			interrupts = <0 108 4>,
-				     <0 109 4>;
+				     <0 109 4>,
+				     <0 110 4>,
+				     <0 111 4>,
+				     <0 112 4>,
+				     <0 113 4>,
+				     <0 114 4>,
+				     <0 115 4>;
 			port-id = <1>;
 			dma-coherent;
 			clocks = <&xge1clk 0>;
diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi
index fe30f76..cafb2c2 100644
--- a/arch/arm64/boot/dts/apm/apm-storm.dtsi
+++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi
@@ -958,7 +958,13 @@
 			      <0x0 0x18000000 0x0 0X200>;
 			reg-names = "enet_csr", "ring_csr", "ring_cmd";
 			interrupts = <0x0 0x60 0x4>,
-				     <0x0 0x61 0x4>;
+				     <0x0 0x61 0x4>,
+				     <0x0 0x62 0x4>,
+				     <0x0 0x63 0x4>,
+				     <0x0 0x64 0x4>,
+				     <0x0 0x65 0x4>,
+				     <0x0 0x66 0x4>,
+				     <0x0 0x67 0x4>;
 			dma-coherent;
 			clocks = <&xge0clk 0>;
 			/* mac address will be overwritten by the bootloader */
-- 
cgit v0.10.2


From b15edf852a2a92ccc0ad77340b9d032545953c43 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:39 +0100
Subject: rocker: remove unused rocker_port param from alloc funcs and shorten
 their names

No need to pass rocker_port around to alloc/free rocker functions,
since they now use switchdev_trans for memory management storage.
With the param removal, shorten the name of the functions since they now
has nothing to do with rocker port.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker.c b/drivers/net/ethernet/rocker/rocker.c
index 166a7fc..e91efd7 100644
--- a/drivers/net/ethernet/rocker/rocker.c
+++ b/drivers/net/ethernet/rocker/rocker.c
@@ -341,9 +341,8 @@ static bool rocker_port_is_ovsed(const struct rocker_port *rocker_port)
 #define ROCKER_OP_FLAG_LEARNED		BIT(2)
 #define ROCKER_OP_FLAG_REFRESH		BIT(3)
 
-static void *__rocker_port_mem_alloc(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     size_t size)
+static void *__rocker_mem_alloc(struct switchdev_trans *trans, int flags,
+				size_t size)
 {
 	struct switchdev_trans_item *elem = NULL;
 	gfp_t gfp_flags = (flags & ROCKER_OP_FLAG_NOWAIT) ?
@@ -372,21 +371,19 @@ static void *__rocker_port_mem_alloc(struct rocker_port *rocker_port,
 	return elem ? elem + 1 : NULL;
 }
 
-static void *rocker_port_kzalloc(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 size_t size)
+static void *rocker_kzalloc(struct switchdev_trans *trans, int flags,
+			    size_t size)
 {
-	return __rocker_port_mem_alloc(rocker_port, trans, flags, size);
+	return __rocker_mem_alloc(trans, flags, size);
 }
 
-static void *rocker_port_kcalloc(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 size_t n, size_t size)
+static void *rocker_kcalloc(struct switchdev_trans *trans, int flags,
+			    size_t n, size_t size)
 {
-	return __rocker_port_mem_alloc(rocker_port, trans, flags, n * size);
+	return __rocker_mem_alloc(trans, flags, n * size);
 }
 
-static void rocker_port_kfree(struct switchdev_trans *trans, const void *mem)
+static void rocker_kfree(struct switchdev_trans *trans, const void *mem)
 {
 	struct switchdev_trans_item *elem;
 
@@ -426,7 +423,7 @@ static struct rocker_wait *rocker_wait_create(struct rocker_port *rocker_port,
 {
 	struct rocker_wait *wait;
 
-	wait = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*wait));
+	wait = rocker_kzalloc(trans, flags, sizeof(*wait));
 	if (!wait)
 		return NULL;
 	rocker_wait_init(wait);
@@ -436,7 +433,7 @@ static struct rocker_wait *rocker_wait_create(struct rocker_port *rocker_port,
 static void rocker_wait_destroy(struct switchdev_trans *trans,
 				struct rocker_wait *wait)
 {
-	rocker_port_kfree(trans, wait);
+	rocker_kfree(trans, wait);
 }
 
 static bool rocker_wait_event_timeout(struct rocker_wait *wait,
@@ -2441,7 +2438,7 @@ static int rocker_flow_tbl_add(struct rocker_port *rocker_port,
 		match->cookie = found->cookie;
 		if (!switchdev_trans_ph_prepare(trans))
 			hash_del(&found->entry);
-		rocker_port_kfree(trans, found);
+		rocker_kfree(trans, found);
 		found = match;
 		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD;
 	} else {
@@ -2483,13 +2480,13 @@ static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
 
 	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
 
-	rocker_port_kfree(trans, match);
+	rocker_kfree(trans, match);
 
 	if (found) {
 		err = rocker_cmd_exec(rocker_port, trans, flags,
 				      rocker_cmd_flow_tbl_del,
 				      found, NULL, NULL);
-		rocker_port_kfree(trans, found);
+		rocker_kfree(trans, found);
 	}
 
 	return err;
@@ -2512,7 +2509,7 @@ static int rocker_flow_tbl_ig_port(struct rocker_port *rocker_port,
 {
 	struct rocker_flow_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2534,7 +2531,7 @@ static int rocker_flow_tbl_vlan(struct rocker_port *rocker_port,
 {
 	struct rocker_flow_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2561,7 +2558,7 @@ static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port,
 {
 	struct rocker_flow_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2601,7 +2598,7 @@ static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port,
 	bool dflt = !eth_dst || (eth_dst && eth_dst_mask);
 	bool wild = false;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2651,7 +2648,7 @@ static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port,
 {
 	struct rocker_flow_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2681,7 +2678,7 @@ static int rocker_flow_tbl_acl(struct rocker_port *rocker_port,
 	u32 priority;
 	struct rocker_flow_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2740,12 +2737,12 @@ static void rocker_group_tbl_entry_free(struct switchdev_trans *trans,
 	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
 	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
 	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
-		rocker_port_kfree(trans, entry->group_ids);
+		rocker_kfree(trans, entry->group_ids);
 		break;
 	default:
 		break;
 	}
-	rocker_port_kfree(trans, entry);
+	rocker_kfree(trans, entry);
 }
 
 static int rocker_group_tbl_add(struct rocker_port *rocker_port,
@@ -2830,7 +2827,7 @@ static int rocker_group_l2_interface(struct rocker_port *rocker_port,
 {
 	struct rocker_group_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2847,17 +2844,17 @@ static int rocker_group_l2_fan_out(struct rocker_port *rocker_port,
 {
 	struct rocker_group_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
 	entry->group_id = group_id;
 	entry->group_count = group_count;
 
-	entry->group_ids = rocker_port_kcalloc(rocker_port, trans, flags,
-					       group_count, sizeof(u32));
+	entry->group_ids = rocker_kcalloc(trans, flags,
+					  group_count, sizeof(u32));
 	if (!entry->group_ids) {
-		rocker_port_kfree(trans, entry);
+		rocker_kfree(trans, entry);
 		return -ENOMEM;
 	}
 	memcpy(entry->group_ids, group_ids, group_count * sizeof(u32));
@@ -2882,7 +2879,7 @@ static int rocker_group_l3_unicast(struct rocker_port *rocker_port,
 {
 	struct rocker_group_tbl_entry *entry;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -2931,7 +2928,7 @@ static void _rocker_neigh_del(struct switchdev_trans *trans,
 		return;
 	if (--entry->ref_count == 0) {
 		hash_del(&entry->entry);
-		rocker_port_kfree(trans, entry);
+		rocker_kfree(trans, entry);
 	}
 }
 
@@ -2965,7 +2962,7 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port,
 	bool removing;
 	int err = 0;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -3034,7 +3031,7 @@ static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port,
 
 err_out:
 	if (!adding)
-		rocker_port_kfree(trans, entry);
+		rocker_kfree(trans, entry);
 
 	return err;
 }
@@ -3082,7 +3079,7 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port,
 	bool resolved = true;
 	int err = 0;
 
-	entry = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*entry));
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
 	if (!entry)
 		return -ENOMEM;
 
@@ -3114,7 +3111,7 @@ static int rocker_port_ipv4_nh(struct rocker_port *rocker_port,
 	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
 
 	if (!adding)
-		rocker_port_kfree(trans, entry);
+		rocker_kfree(trans, entry);
 
 	if (err)
 		return err;
@@ -3139,8 +3136,8 @@ static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port,
 	int err = 0;
 	int i;
 
-	group_ids = rocker_port_kcalloc(rocker_port, trans, flags,
-					rocker->port_count, sizeof(u32));
+	group_ids = rocker_kcalloc(trans, flags,
+				   rocker->port_count, sizeof(u32));
 	if (!group_ids)
 		return -ENOMEM;
 
@@ -3172,7 +3169,7 @@ static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port,
 			   "Error (%d) port VLAN l2 flood group\n", err);
 
 no_ports_in_vlan:
-	rocker_port_kfree(trans, group_ids);
+	rocker_kfree(trans, group_ids);
 	return err;
 }
 
@@ -3540,7 +3537,7 @@ static void rocker_port_fdb_learn_work(struct work_struct *work)
 					 lw->rocker_port->dev, &info.info);
 	rtnl_unlock();
 
-	rocker_port_kfree(lw->trans, work);
+	rocker_kfree(lw->trans, work);
 }
 
 static int rocker_port_fdb_learn(struct rocker_port *rocker_port,
@@ -3574,7 +3571,7 @@ static int rocker_port_fdb_learn(struct rocker_port *rocker_port,
 	if (!rocker_port_is_bridged(rocker_port))
 		return 0;
 
-	lw = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*lw));
+	lw = rocker_kzalloc(trans, flags, sizeof(*lw));
 	if (!lw)
 		return -ENOMEM;
 
@@ -3587,7 +3584,7 @@ static int rocker_port_fdb_learn(struct rocker_port *rocker_port,
 	lw->vid = rocker_port_vlan_to_vid(rocker_port, vlan_id);
 
 	if (switchdev_trans_ph_prepare(trans))
-		rocker_port_kfree(trans, lw);
+		rocker_kfree(trans, lw);
 	else
 		schedule_work(&lw->work);
 
@@ -3618,7 +3615,7 @@ static int rocker_port_fdb(struct rocker_port *rocker_port,
 	bool removing = (flags & ROCKER_OP_FLAG_REMOVE);
 	unsigned long lock_flags;
 
-	fdb = rocker_port_kzalloc(rocker_port, trans, flags, sizeof(*fdb));
+	fdb = rocker_kzalloc(trans, flags, sizeof(*fdb));
 	if (!fdb)
 		return -ENOMEM;
 
@@ -3636,7 +3633,7 @@ static int rocker_port_fdb(struct rocker_port *rocker_port,
 	if (found) {
 		found->touched = jiffies;
 		if (removing) {
-			rocker_port_kfree(trans, fdb);
+			rocker_kfree(trans, fdb);
 			if (!switchdev_trans_ph_prepare(trans))
 				hash_del(&found->entry);
 		}
@@ -3650,7 +3647,7 @@ static int rocker_port_fdb(struct rocker_port *rocker_port,
 
 	/* Check if adding and already exists, or removing and can't find */
 	if (!found != !removing) {
-		rocker_port_kfree(trans, fdb);
+		rocker_kfree(trans, fdb);
 		if (!found && removing)
 			return 0;
 		/* Refreshing existing to update aging timers */
-- 
cgit v0.10.2


From 0fe685f6a2831c8868e0a4389c76521d7fd42026 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:40 +0100
Subject: rocker: rename rocker.h to rocker_hw.h

Since "rocker.h" file is going to be used for different purpose,
rename the hardware-specific header to "rocker_hw.h".

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker.c b/drivers/net/ethernet/rocker/rocker.c
index e91efd7..3f18498 100644
--- a/drivers/net/ethernet/rocker/rocker.c
+++ b/drivers/net/ethernet/rocker/rocker.c
@@ -39,7 +39,7 @@
 #include <linux/io-64-nonatomic-lo-hi.h>
 #include <generated/utsrelease.h>
 
-#include "rocker.h"
+#include "rocker_hw.h"
 
 static const char rocker_driver_name[] = "rocker";
 
diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
deleted file mode 100644
index 12490b2..0000000
--- a/drivers/net/ethernet/rocker/rocker.h
+++ /dev/null
@@ -1,467 +0,0 @@
-/*
- * drivers/net/ethernet/rocker/rocker.h - Rocker switch device driver
- * Copyright (c) 2014 Jiri Pirko <jiri@resnulli.us>
- * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#ifndef _ROCKER_H
-#define _ROCKER_H
-
-#include <linux/types.h>
-
-/* Return codes */
-enum {
-	ROCKER_OK = 0,
-	ROCKER_ENOENT = 2,
-	ROCKER_ENXIO = 6,
-	ROCKER_ENOMEM = 12,
-	ROCKER_EEXIST = 17,
-	ROCKER_EINVAL = 22,
-	ROCKER_EMSGSIZE = 90,
-	ROCKER_ENOTSUP = 95,
-	ROCKER_ENOBUFS = 105,
-};
-
-#define ROCKER_FP_PORTS_MAX 62
-
-#define PCI_VENDOR_ID_REDHAT		0x1b36
-#define PCI_DEVICE_ID_REDHAT_ROCKER	0x0006
-
-#define ROCKER_PCI_BAR0_SIZE		0x2000
-
-/* MSI-X vectors */
-enum {
-	ROCKER_MSIX_VEC_CMD,
-	ROCKER_MSIX_VEC_EVENT,
-	ROCKER_MSIX_VEC_TEST,
-	ROCKER_MSIX_VEC_RESERVED0,
-	__ROCKER_MSIX_VEC_TX,
-	__ROCKER_MSIX_VEC_RX,
-#define ROCKER_MSIX_VEC_TX(port) \
-	(__ROCKER_MSIX_VEC_TX + ((port) * 2))
-#define ROCKER_MSIX_VEC_RX(port) \
-	(__ROCKER_MSIX_VEC_RX + ((port) * 2))
-#define ROCKER_MSIX_VEC_COUNT(portcnt) \
-	(ROCKER_MSIX_VEC_RX((portcnt - 1)) + 1)
-};
-
-/* Rocker bogus registers */
-#define ROCKER_BOGUS_REG0		0x0000
-#define ROCKER_BOGUS_REG1		0x0004
-#define ROCKER_BOGUS_REG2		0x0008
-#define ROCKER_BOGUS_REG3		0x000c
-
-/* Rocker test registers */
-#define ROCKER_TEST_REG			0x0010
-#define ROCKER_TEST_REG64		0x0018  /* 8-byte */
-#define ROCKER_TEST_IRQ			0x0020
-#define ROCKER_TEST_DMA_ADDR		0x0028  /* 8-byte */
-#define ROCKER_TEST_DMA_SIZE		0x0030
-#define ROCKER_TEST_DMA_CTRL		0x0034
-
-/* Rocker test register ctrl */
-#define ROCKER_TEST_DMA_CTRL_CLEAR	BIT(0)
-#define ROCKER_TEST_DMA_CTRL_FILL	BIT(1)
-#define ROCKER_TEST_DMA_CTRL_INVERT	BIT(2)
-
-/* Rocker DMA ring register offsets */
-#define ROCKER_DMA_DESC_ADDR(x)		(0x1000 + (x) * 32)  /* 8-byte */
-#define ROCKER_DMA_DESC_SIZE(x)		(0x1008 + (x) * 32)
-#define ROCKER_DMA_DESC_HEAD(x)		(0x100c + (x) * 32)
-#define ROCKER_DMA_DESC_TAIL(x)		(0x1010 + (x) * 32)
-#define ROCKER_DMA_DESC_CTRL(x)		(0x1014 + (x) * 32)
-#define ROCKER_DMA_DESC_CREDITS(x)	(0x1018 + (x) * 32)
-#define ROCKER_DMA_DESC_RES1(x)		(0x101c + (x) * 32)
-
-/* Rocker dma ctrl register bits */
-#define ROCKER_DMA_DESC_CTRL_RESET	BIT(0)
-
-/* Rocker DMA ring types */
-enum rocker_dma_type {
-	ROCKER_DMA_CMD,
-	ROCKER_DMA_EVENT,
-	__ROCKER_DMA_TX,
-	__ROCKER_DMA_RX,
-#define ROCKER_DMA_TX(port) (__ROCKER_DMA_TX + (port) * 2)
-#define ROCKER_DMA_RX(port) (__ROCKER_DMA_RX + (port) * 2)
-};
-
-/* Rocker DMA ring size limits and default sizes */
-#define ROCKER_DMA_SIZE_MIN		2ul
-#define ROCKER_DMA_SIZE_MAX		65536ul
-#define ROCKER_DMA_CMD_DEFAULT_SIZE	32ul
-#define ROCKER_DMA_EVENT_DEFAULT_SIZE	32ul
-#define ROCKER_DMA_TX_DEFAULT_SIZE	64ul
-#define ROCKER_DMA_TX_DESC_SIZE		256
-#define ROCKER_DMA_RX_DEFAULT_SIZE	64ul
-#define ROCKER_DMA_RX_DESC_SIZE		256
-
-/* Rocker DMA descriptor struct */
-struct rocker_desc {
-	u64 buf_addr;
-	u64 cookie;
-	u16 buf_size;
-	u16 tlv_size;
-	u16 resv[5];
-	u16 comp_err;
-};
-
-#define ROCKER_DMA_DESC_COMP_ERR_GEN	BIT(15)
-
-/* Rocker DMA TLV struct */
-struct rocker_tlv {
-	u32 type;
-	u16 len;
-};
-
-/* TLVs */
-enum {
-	ROCKER_TLV_CMD_UNSPEC,
-	ROCKER_TLV_CMD_TYPE,	/* u16 */
-	ROCKER_TLV_CMD_INFO,	/* nest */
-
-	__ROCKER_TLV_CMD_MAX,
-	ROCKER_TLV_CMD_MAX = __ROCKER_TLV_CMD_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_CMD_TYPE_UNSPEC,
-	ROCKER_TLV_CMD_TYPE_GET_PORT_SETTINGS,
-	ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_GET_STATS,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL,
-	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_GET_STATS,
-
-	ROCKER_TLV_CMD_TYPE_CLEAR_PORT_STATS,
-	ROCKER_TLV_CMD_TYPE_GET_PORT_STATS,
-
-	__ROCKER_TLV_CMD_TYPE_MAX,
-	ROCKER_TLV_CMD_TYPE_MAX = __ROCKER_TLV_CMD_TYPE_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_CMD_PORT_SETTINGS_UNSPEC,
-	ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,		/* u32 */
-	ROCKER_TLV_CMD_PORT_SETTINGS_SPEED,		/* u32 */
-	ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX,		/* u8 */
-	ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG,		/* u8 */
-	ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR,		/* binary */
-	ROCKER_TLV_CMD_PORT_SETTINGS_MODE,		/* u8 */
-	ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,		/* u8 */
-	ROCKER_TLV_CMD_PORT_SETTINGS_PHYS_NAME,		/* binary */
-	ROCKER_TLV_CMD_PORT_SETTINGS_MTU,		/* u16 */
-
-	__ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
-	ROCKER_TLV_CMD_PORT_SETTINGS_MAX =
-			__ROCKER_TLV_CMD_PORT_SETTINGS_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_CMD_PORT_STATS_UNSPEC,
-	ROCKER_TLV_CMD_PORT_STATS_PPORT,            /* u32 */
-
-	ROCKER_TLV_CMD_PORT_STATS_RX_PKTS,          /* u64 */
-	ROCKER_TLV_CMD_PORT_STATS_RX_BYTES,         /* u64 */
-	ROCKER_TLV_CMD_PORT_STATS_RX_DROPPED,       /* u64 */
-	ROCKER_TLV_CMD_PORT_STATS_RX_ERRORS,        /* u64 */
-
-	ROCKER_TLV_CMD_PORT_STATS_TX_PKTS,          /* u64 */
-	ROCKER_TLV_CMD_PORT_STATS_TX_BYTES,         /* u64 */
-	ROCKER_TLV_CMD_PORT_STATS_TX_DROPPED,       /* u64 */
-	ROCKER_TLV_CMD_PORT_STATS_TX_ERRORS,        /* u64 */
-
-	__ROCKER_TLV_CMD_PORT_STATS_MAX,
-	ROCKER_TLV_CMD_PORT_STATS_MAX = __ROCKER_TLV_CMD_PORT_STATS_MAX - 1,
-};
-
-enum rocker_port_mode {
-	ROCKER_PORT_MODE_OF_DPA,
-};
-
-enum {
-	ROCKER_TLV_EVENT_UNSPEC,
-	ROCKER_TLV_EVENT_TYPE,	/* u16 */
-	ROCKER_TLV_EVENT_INFO,	/* nest */
-
-	__ROCKER_TLV_EVENT_MAX,
-	ROCKER_TLV_EVENT_MAX = __ROCKER_TLV_EVENT_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_EVENT_TYPE_UNSPEC,
-	ROCKER_TLV_EVENT_TYPE_LINK_CHANGED,
-	ROCKER_TLV_EVENT_TYPE_MAC_VLAN_SEEN,
-
-	__ROCKER_TLV_EVENT_TYPE_MAX,
-	ROCKER_TLV_EVENT_TYPE_MAX = __ROCKER_TLV_EVENT_TYPE_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_EVENT_LINK_CHANGED_UNSPEC,
-	ROCKER_TLV_EVENT_LINK_CHANGED_PPORT,	/* u32 */
-	ROCKER_TLV_EVENT_LINK_CHANGED_LINKUP,	/* u8 */
-
-	__ROCKER_TLV_EVENT_LINK_CHANGED_MAX,
-	ROCKER_TLV_EVENT_LINK_CHANGED_MAX =
-			__ROCKER_TLV_EVENT_LINK_CHANGED_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_EVENT_MAC_VLAN_UNSPEC,
-	ROCKER_TLV_EVENT_MAC_VLAN_PPORT,	/* u32 */
-	ROCKER_TLV_EVENT_MAC_VLAN_MAC,		/* binary */
-	ROCKER_TLV_EVENT_MAC_VLAN_VLAN_ID,	/* __be16 */
-
-	__ROCKER_TLV_EVENT_MAC_VLAN_MAX,
-	ROCKER_TLV_EVENT_MAC_VLAN_MAX = __ROCKER_TLV_EVENT_MAC_VLAN_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_RX_UNSPEC,
-	ROCKER_TLV_RX_FLAGS,		/* u16, see ROCKER_RX_FLAGS_ */
-	ROCKER_TLV_RX_CSUM,		/* u16 */
-	ROCKER_TLV_RX_FRAG_ADDR,	/* u64 */
-	ROCKER_TLV_RX_FRAG_MAX_LEN,	/* u16 */
-	ROCKER_TLV_RX_FRAG_LEN,		/* u16 */
-
-	__ROCKER_TLV_RX_MAX,
-	ROCKER_TLV_RX_MAX = __ROCKER_TLV_RX_MAX - 1,
-};
-
-#define ROCKER_RX_FLAGS_IPV4			BIT(0)
-#define ROCKER_RX_FLAGS_IPV6			BIT(1)
-#define ROCKER_RX_FLAGS_CSUM_CALC		BIT(2)
-#define ROCKER_RX_FLAGS_IPV4_CSUM_GOOD		BIT(3)
-#define ROCKER_RX_FLAGS_IP_FRAG			BIT(4)
-#define ROCKER_RX_FLAGS_TCP			BIT(5)
-#define ROCKER_RX_FLAGS_UDP			BIT(6)
-#define ROCKER_RX_FLAGS_TCP_UDP_CSUM_GOOD	BIT(7)
-#define ROCKER_RX_FLAGS_FWD_OFFLOAD		BIT(8)
-
-enum {
-	ROCKER_TLV_TX_UNSPEC,
-	ROCKER_TLV_TX_OFFLOAD,		/* u8, see ROCKER_TX_OFFLOAD_ */
-	ROCKER_TLV_TX_L3_CSUM_OFF,	/* u16 */
-	ROCKER_TLV_TX_TSO_MSS,		/* u16 */
-	ROCKER_TLV_TX_TSO_HDR_LEN,	/* u16 */
-	ROCKER_TLV_TX_FRAGS,		/* array */
-
-	__ROCKER_TLV_TX_MAX,
-	ROCKER_TLV_TX_MAX = __ROCKER_TLV_TX_MAX - 1,
-};
-
-#define ROCKER_TX_OFFLOAD_NONE		0
-#define ROCKER_TX_OFFLOAD_IP_CSUM	1
-#define ROCKER_TX_OFFLOAD_TCP_UDP_CSUM	2
-#define ROCKER_TX_OFFLOAD_L3_CSUM	3
-#define ROCKER_TX_OFFLOAD_TSO		4
-
-#define ROCKER_TX_FRAGS_MAX		16
-
-enum {
-	ROCKER_TLV_TX_FRAG_UNSPEC,
-	ROCKER_TLV_TX_FRAG,		/* nest */
-
-	__ROCKER_TLV_TX_FRAG_MAX,
-	ROCKER_TLV_TX_FRAG_MAX = __ROCKER_TLV_TX_FRAG_MAX - 1,
-};
-
-enum {
-	ROCKER_TLV_TX_FRAG_ATTR_UNSPEC,
-	ROCKER_TLV_TX_FRAG_ATTR_ADDR,	/* u64 */
-	ROCKER_TLV_TX_FRAG_ATTR_LEN,	/* u16 */
-
-	__ROCKER_TLV_TX_FRAG_ATTR_MAX,
-	ROCKER_TLV_TX_FRAG_ATTR_MAX = __ROCKER_TLV_TX_FRAG_ATTR_MAX - 1,
-};
-
-/* cmd info nested for OF-DPA msgs */
-enum {
-	ROCKER_TLV_OF_DPA_UNSPEC,
-	ROCKER_TLV_OF_DPA_TABLE_ID,		/* u16 */
-	ROCKER_TLV_OF_DPA_PRIORITY,		/* u32 */
-	ROCKER_TLV_OF_DPA_HARDTIME,		/* u32 */
-	ROCKER_TLV_OF_DPA_IDLETIME,		/* u32 */
-	ROCKER_TLV_OF_DPA_COOKIE,		/* u64 */
-	ROCKER_TLV_OF_DPA_IN_PPORT,		/* u32 */
-	ROCKER_TLV_OF_DPA_IN_PPORT_MASK,	/* u32 */
-	ROCKER_TLV_OF_DPA_OUT_PPORT,		/* u32 */
-	ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,	/* u16 */
-	ROCKER_TLV_OF_DPA_GROUP_ID,		/* u32 */
-	ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,	/* u32 */
-	ROCKER_TLV_OF_DPA_GROUP_COUNT,		/* u16 */
-	ROCKER_TLV_OF_DPA_GROUP_IDS,		/* u32 array */
-	ROCKER_TLV_OF_DPA_VLAN_ID,		/* __be16 */
-	ROCKER_TLV_OF_DPA_VLAN_ID_MASK,		/* __be16 */
-	ROCKER_TLV_OF_DPA_VLAN_PCP,		/* __be16 */
-	ROCKER_TLV_OF_DPA_VLAN_PCP_MASK,	/* __be16 */
-	ROCKER_TLV_OF_DPA_VLAN_PCP_ACTION,	/* u8 */
-	ROCKER_TLV_OF_DPA_NEW_VLAN_ID,		/* __be16 */
-	ROCKER_TLV_OF_DPA_NEW_VLAN_PCP,		/* u8 */
-	ROCKER_TLV_OF_DPA_TUNNEL_ID,		/* u32 */
-	ROCKER_TLV_OF_DPA_TUNNEL_LPORT,		/* u32 */
-	ROCKER_TLV_OF_DPA_ETHERTYPE,		/* __be16 */
-	ROCKER_TLV_OF_DPA_DST_MAC,		/* binary */
-	ROCKER_TLV_OF_DPA_DST_MAC_MASK,		/* binary */
-	ROCKER_TLV_OF_DPA_SRC_MAC,		/* binary */
-	ROCKER_TLV_OF_DPA_SRC_MAC_MASK,		/* binary */
-	ROCKER_TLV_OF_DPA_IP_PROTO,		/* u8 */
-	ROCKER_TLV_OF_DPA_IP_PROTO_MASK,	/* u8 */
-	ROCKER_TLV_OF_DPA_IP_DSCP,		/* u8 */
-	ROCKER_TLV_OF_DPA_IP_DSCP_MASK,		/* u8 */
-	ROCKER_TLV_OF_DPA_IP_DSCP_ACTION,	/* u8 */
-	ROCKER_TLV_OF_DPA_NEW_IP_DSCP,		/* u8 */
-	ROCKER_TLV_OF_DPA_IP_ECN,		/* u8 */
-	ROCKER_TLV_OF_DPA_IP_ECN_MASK,		/* u8 */
-	ROCKER_TLV_OF_DPA_DST_IP,		/* __be32 */
-	ROCKER_TLV_OF_DPA_DST_IP_MASK,		/* __be32 */
-	ROCKER_TLV_OF_DPA_SRC_IP,		/* __be32 */
-	ROCKER_TLV_OF_DPA_SRC_IP_MASK,		/* __be32 */
-	ROCKER_TLV_OF_DPA_DST_IPV6,		/* binary */
-	ROCKER_TLV_OF_DPA_DST_IPV6_MASK,	/* binary */
-	ROCKER_TLV_OF_DPA_SRC_IPV6,		/* binary */
-	ROCKER_TLV_OF_DPA_SRC_IPV6_MASK,	/* binary */
-	ROCKER_TLV_OF_DPA_SRC_ARP_IP,		/* __be32 */
-	ROCKER_TLV_OF_DPA_SRC_ARP_IP_MASK,	/* __be32 */
-	ROCKER_TLV_OF_DPA_L4_DST_PORT,		/* __be16 */
-	ROCKER_TLV_OF_DPA_L4_DST_PORT_MASK,	/* __be16 */
-	ROCKER_TLV_OF_DPA_L4_SRC_PORT,		/* __be16 */
-	ROCKER_TLV_OF_DPA_L4_SRC_PORT_MASK,	/* __be16 */
-	ROCKER_TLV_OF_DPA_ICMP_TYPE,		/* u8 */
-	ROCKER_TLV_OF_DPA_ICMP_TYPE_MASK,	/* u8 */
-	ROCKER_TLV_OF_DPA_ICMP_CODE,		/* u8 */
-	ROCKER_TLV_OF_DPA_ICMP_CODE_MASK,	/* u8 */
-	ROCKER_TLV_OF_DPA_IPV6_LABEL,		/* __be32 */
-	ROCKER_TLV_OF_DPA_IPV6_LABEL_MASK,	/* __be32 */
-	ROCKER_TLV_OF_DPA_QUEUE_ID_ACTION,	/* u8 */
-	ROCKER_TLV_OF_DPA_NEW_QUEUE_ID,		/* u8 */
-	ROCKER_TLV_OF_DPA_CLEAR_ACTIONS,	/* u32 */
-	ROCKER_TLV_OF_DPA_POP_VLAN,		/* u8 */
-	ROCKER_TLV_OF_DPA_TTL_CHECK,		/* u8 */
-	ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,	/* u8 */
-
-	__ROCKER_TLV_OF_DPA_MAX,
-	ROCKER_TLV_OF_DPA_MAX = __ROCKER_TLV_OF_DPA_MAX - 1,
-};
-
-/* OF-DPA table IDs */
-
-enum rocker_of_dpa_table_id {
-	ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT = 0,
-	ROCKER_OF_DPA_TABLE_ID_VLAN = 10,
-	ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC = 20,
-	ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING = 30,
-	ROCKER_OF_DPA_TABLE_ID_MULTICAST_ROUTING = 40,
-	ROCKER_OF_DPA_TABLE_ID_BRIDGING = 50,
-	ROCKER_OF_DPA_TABLE_ID_ACL_POLICY = 60,
-};
-
-/* OF-DPA flow stats */
-enum {
-	ROCKER_TLV_OF_DPA_FLOW_STAT_UNSPEC,
-	ROCKER_TLV_OF_DPA_FLOW_STAT_DURATION,	/* u32 */
-	ROCKER_TLV_OF_DPA_FLOW_STAT_RX_PKTS,	/* u64 */
-	ROCKER_TLV_OF_DPA_FLOW_STAT_TX_PKTS,	/* u64 */
-
-	__ROCKER_TLV_OF_DPA_FLOW_STAT_MAX,
-	ROCKER_TLV_OF_DPA_FLOW_STAT_MAX = __ROCKER_TLV_OF_DPA_FLOW_STAT_MAX - 1,
-};
-
-/* OF-DPA group types */
-enum rocker_of_dpa_group_type {
-	ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE = 0,
-	ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE,
-	ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST,
-	ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST,
-	ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD,
-	ROCKER_OF_DPA_GROUP_TYPE_L3_INTERFACE,
-	ROCKER_OF_DPA_GROUP_TYPE_L3_MCAST,
-	ROCKER_OF_DPA_GROUP_TYPE_L3_ECMP,
-	ROCKER_OF_DPA_GROUP_TYPE_L2_OVERLAY,
-};
-
-/* OF-DPA group L2 overlay types */
-enum rocker_of_dpa_overlay_type {
-	ROCKER_OF_DPA_OVERLAY_TYPE_FLOOD_UCAST = 0,
-	ROCKER_OF_DPA_OVERLAY_TYPE_FLOOD_MCAST,
-	ROCKER_OF_DPA_OVERLAY_TYPE_MCAST_UCAST,
-	ROCKER_OF_DPA_OVERLAY_TYPE_MCAST_MCAST,
-};
-
-/* OF-DPA group ID encoding */
-#define ROCKER_GROUP_TYPE_SHIFT 28
-#define ROCKER_GROUP_TYPE_MASK 0xf0000000
-#define ROCKER_GROUP_VLAN_SHIFT 16
-#define ROCKER_GROUP_VLAN_MASK 0x0fff0000
-#define ROCKER_GROUP_PORT_SHIFT 0
-#define ROCKER_GROUP_PORT_MASK 0x0000ffff
-#define ROCKER_GROUP_TUNNEL_ID_SHIFT 12
-#define ROCKER_GROUP_TUNNEL_ID_MASK 0x0ffff000
-#define ROCKER_GROUP_SUBTYPE_SHIFT 10
-#define ROCKER_GROUP_SUBTYPE_MASK 0x00000c00
-#define ROCKER_GROUP_INDEX_SHIFT 0
-#define ROCKER_GROUP_INDEX_MASK 0x0000ffff
-#define ROCKER_GROUP_INDEX_LONG_SHIFT 0
-#define ROCKER_GROUP_INDEX_LONG_MASK 0x0fffffff
-
-#define ROCKER_GROUP_TYPE_GET(group_id) \
-	(((group_id) & ROCKER_GROUP_TYPE_MASK) >> ROCKER_GROUP_TYPE_SHIFT)
-#define ROCKER_GROUP_TYPE_SET(type) \
-	(((type) << ROCKER_GROUP_TYPE_SHIFT) & ROCKER_GROUP_TYPE_MASK)
-#define ROCKER_GROUP_VLAN_GET(group_id) \
-	(((group_id) & ROCKER_GROUP_VLAN_ID_MASK) >> ROCKER_GROUP_VLAN_ID_SHIFT)
-#define ROCKER_GROUP_VLAN_SET(vlan_id) \
-	(((vlan_id) << ROCKER_GROUP_VLAN_SHIFT) & ROCKER_GROUP_VLAN_MASK)
-#define ROCKER_GROUP_PORT_GET(group_id) \
-	(((group_id) & ROCKER_GROUP_PORT_MASK) >> ROCKER_GROUP_PORT_SHIFT)
-#define ROCKER_GROUP_PORT_SET(port) \
-	(((port) << ROCKER_GROUP_PORT_SHIFT) & ROCKER_GROUP_PORT_MASK)
-#define ROCKER_GROUP_INDEX_GET(group_id) \
-	(((group_id) & ROCKER_GROUP_INDEX_MASK) >> ROCKER_GROUP_INDEX_SHIFT)
-#define ROCKER_GROUP_INDEX_SET(index) \
-	(((index) << ROCKER_GROUP_INDEX_SHIFT) & ROCKER_GROUP_INDEX_MASK)
-#define ROCKER_GROUP_INDEX_LONG_GET(group_id) \
-	(((group_id) & ROCKER_GROUP_INDEX_LONG_MASK) >> \
-	 ROCKER_GROUP_INDEX_LONG_SHIFT)
-#define ROCKER_GROUP_INDEX_LONG_SET(index) \
-	(((index) << ROCKER_GROUP_INDEX_LONG_SHIFT) & \
-	 ROCKER_GROUP_INDEX_LONG_MASK)
-
-#define ROCKER_GROUP_NONE 0
-#define ROCKER_GROUP_L2_INTERFACE(vlan_id, port) \
-	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE) |\
-	 ROCKER_GROUP_VLAN_SET(ntohs(vlan_id)) | ROCKER_GROUP_PORT_SET(port))
-#define ROCKER_GROUP_L2_REWRITE(index) \
-	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE) |\
-	 ROCKER_GROUP_INDEX_LONG_SET(index))
-#define ROCKER_GROUP_L2_MCAST(vlan_id, index) \
-	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST) |\
-	 ROCKER_GROUP_VLAN_SET(ntohs(vlan_id)) | ROCKER_GROUP_INDEX_SET(index))
-#define ROCKER_GROUP_L2_FLOOD(vlan_id, index) \
-	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD) |\
-	ROCKER_GROUP_VLAN_SET(ntohs(vlan_id)) | ROCKER_GROUP_INDEX_SET(index))
-#define ROCKER_GROUP_L3_UNICAST(index) \
-	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST) |\
-	 ROCKER_GROUP_INDEX_LONG_SET(index))
-
-/* Rocker general purpose registers */
-#define ROCKER_CONTROL			0x0300
-#define ROCKER_PORT_PHYS_COUNT		0x0304
-#define ROCKER_PORT_PHYS_LINK_STATUS	0x0310 /* 8-byte */
-#define ROCKER_PORT_PHYS_ENABLE		0x0318 /* 8-byte */
-#define ROCKER_SWITCH_ID		0x0320 /* 8-byte */
-
-/* Rocker control bits */
-#define ROCKER_CONTROL_RESET		BIT(0)
-
-#endif
diff --git a/drivers/net/ethernet/rocker/rocker_hw.h b/drivers/net/ethernet/rocker/rocker_hw.h
new file mode 100644
index 0000000..2adfe88
--- /dev/null
+++ b/drivers/net/ethernet/rocker/rocker_hw.h
@@ -0,0 +1,467 @@
+/*
+ * drivers/net/ethernet/rocker/rocker_hw.h - Rocker switch device driver
+ * Copyright (c) 2014-2016 Jiri Pirko <jiri@mellanox.com>
+ * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _ROCKER_HW_H
+#define _ROCKER_HW_H
+
+#include <linux/types.h>
+
+/* Return codes */
+enum {
+	ROCKER_OK = 0,
+	ROCKER_ENOENT = 2,
+	ROCKER_ENXIO = 6,
+	ROCKER_ENOMEM = 12,
+	ROCKER_EEXIST = 17,
+	ROCKER_EINVAL = 22,
+	ROCKER_EMSGSIZE = 90,
+	ROCKER_ENOTSUP = 95,
+	ROCKER_ENOBUFS = 105,
+};
+
+#define ROCKER_FP_PORTS_MAX 62
+
+#define PCI_VENDOR_ID_REDHAT		0x1b36
+#define PCI_DEVICE_ID_REDHAT_ROCKER	0x0006
+
+#define ROCKER_PCI_BAR0_SIZE		0x2000
+
+/* MSI-X vectors */
+enum {
+	ROCKER_MSIX_VEC_CMD,
+	ROCKER_MSIX_VEC_EVENT,
+	ROCKER_MSIX_VEC_TEST,
+	ROCKER_MSIX_VEC_RESERVED0,
+	__ROCKER_MSIX_VEC_TX,
+	__ROCKER_MSIX_VEC_RX,
+#define ROCKER_MSIX_VEC_TX(port) \
+	(__ROCKER_MSIX_VEC_TX + ((port) * 2))
+#define ROCKER_MSIX_VEC_RX(port) \
+	(__ROCKER_MSIX_VEC_RX + ((port) * 2))
+#define ROCKER_MSIX_VEC_COUNT(portcnt) \
+	(ROCKER_MSIX_VEC_RX((portcnt - 1)) + 1)
+};
+
+/* Rocker bogus registers */
+#define ROCKER_BOGUS_REG0		0x0000
+#define ROCKER_BOGUS_REG1		0x0004
+#define ROCKER_BOGUS_REG2		0x0008
+#define ROCKER_BOGUS_REG3		0x000c
+
+/* Rocker test registers */
+#define ROCKER_TEST_REG			0x0010
+#define ROCKER_TEST_REG64		0x0018  /* 8-byte */
+#define ROCKER_TEST_IRQ			0x0020
+#define ROCKER_TEST_DMA_ADDR		0x0028  /* 8-byte */
+#define ROCKER_TEST_DMA_SIZE		0x0030
+#define ROCKER_TEST_DMA_CTRL		0x0034
+
+/* Rocker test register ctrl */
+#define ROCKER_TEST_DMA_CTRL_CLEAR	BIT(0)
+#define ROCKER_TEST_DMA_CTRL_FILL	BIT(1)
+#define ROCKER_TEST_DMA_CTRL_INVERT	BIT(2)
+
+/* Rocker DMA ring register offsets */
+#define ROCKER_DMA_DESC_ADDR(x)		(0x1000 + (x) * 32)  /* 8-byte */
+#define ROCKER_DMA_DESC_SIZE(x)		(0x1008 + (x) * 32)
+#define ROCKER_DMA_DESC_HEAD(x)		(0x100c + (x) * 32)
+#define ROCKER_DMA_DESC_TAIL(x)		(0x1010 + (x) * 32)
+#define ROCKER_DMA_DESC_CTRL(x)		(0x1014 + (x) * 32)
+#define ROCKER_DMA_DESC_CREDITS(x)	(0x1018 + (x) * 32)
+#define ROCKER_DMA_DESC_RES1(x)		(0x101c + (x) * 32)
+
+/* Rocker dma ctrl register bits */
+#define ROCKER_DMA_DESC_CTRL_RESET	BIT(0)
+
+/* Rocker DMA ring types */
+enum rocker_dma_type {
+	ROCKER_DMA_CMD,
+	ROCKER_DMA_EVENT,
+	__ROCKER_DMA_TX,
+	__ROCKER_DMA_RX,
+#define ROCKER_DMA_TX(port) (__ROCKER_DMA_TX + (port) * 2)
+#define ROCKER_DMA_RX(port) (__ROCKER_DMA_RX + (port) * 2)
+};
+
+/* Rocker DMA ring size limits and default sizes */
+#define ROCKER_DMA_SIZE_MIN		2ul
+#define ROCKER_DMA_SIZE_MAX		65536ul
+#define ROCKER_DMA_CMD_DEFAULT_SIZE	32ul
+#define ROCKER_DMA_EVENT_DEFAULT_SIZE	32ul
+#define ROCKER_DMA_TX_DEFAULT_SIZE	64ul
+#define ROCKER_DMA_TX_DESC_SIZE		256
+#define ROCKER_DMA_RX_DEFAULT_SIZE	64ul
+#define ROCKER_DMA_RX_DESC_SIZE		256
+
+/* Rocker DMA descriptor struct */
+struct rocker_desc {
+	u64 buf_addr;
+	u64 cookie;
+	u16 buf_size;
+	u16 tlv_size;
+	u16 resv[5];
+	u16 comp_err;
+};
+
+#define ROCKER_DMA_DESC_COMP_ERR_GEN	BIT(15)
+
+/* Rocker DMA TLV struct */
+struct rocker_tlv {
+	u32 type;
+	u16 len;
+};
+
+/* TLVs */
+enum {
+	ROCKER_TLV_CMD_UNSPEC,
+	ROCKER_TLV_CMD_TYPE,	/* u16 */
+	ROCKER_TLV_CMD_INFO,	/* nest */
+
+	__ROCKER_TLV_CMD_MAX,
+	ROCKER_TLV_CMD_MAX = __ROCKER_TLV_CMD_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_CMD_TYPE_UNSPEC,
+	ROCKER_TLV_CMD_TYPE_GET_PORT_SETTINGS,
+	ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_GET_STATS,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL,
+	ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_GET_STATS,
+
+	ROCKER_TLV_CMD_TYPE_CLEAR_PORT_STATS,
+	ROCKER_TLV_CMD_TYPE_GET_PORT_STATS,
+
+	__ROCKER_TLV_CMD_TYPE_MAX,
+	ROCKER_TLV_CMD_TYPE_MAX = __ROCKER_TLV_CMD_TYPE_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_CMD_PORT_SETTINGS_UNSPEC,
+	ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,		/* u32 */
+	ROCKER_TLV_CMD_PORT_SETTINGS_SPEED,		/* u32 */
+	ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX,		/* u8 */
+	ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG,		/* u8 */
+	ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR,		/* binary */
+	ROCKER_TLV_CMD_PORT_SETTINGS_MODE,		/* u8 */
+	ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,		/* u8 */
+	ROCKER_TLV_CMD_PORT_SETTINGS_PHYS_NAME,		/* binary */
+	ROCKER_TLV_CMD_PORT_SETTINGS_MTU,		/* u16 */
+
+	__ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
+	ROCKER_TLV_CMD_PORT_SETTINGS_MAX =
+			__ROCKER_TLV_CMD_PORT_SETTINGS_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_CMD_PORT_STATS_UNSPEC,
+	ROCKER_TLV_CMD_PORT_STATS_PPORT,            /* u32 */
+
+	ROCKER_TLV_CMD_PORT_STATS_RX_PKTS,          /* u64 */
+	ROCKER_TLV_CMD_PORT_STATS_RX_BYTES,         /* u64 */
+	ROCKER_TLV_CMD_PORT_STATS_RX_DROPPED,       /* u64 */
+	ROCKER_TLV_CMD_PORT_STATS_RX_ERRORS,        /* u64 */
+
+	ROCKER_TLV_CMD_PORT_STATS_TX_PKTS,          /* u64 */
+	ROCKER_TLV_CMD_PORT_STATS_TX_BYTES,         /* u64 */
+	ROCKER_TLV_CMD_PORT_STATS_TX_DROPPED,       /* u64 */
+	ROCKER_TLV_CMD_PORT_STATS_TX_ERRORS,        /* u64 */
+
+	__ROCKER_TLV_CMD_PORT_STATS_MAX,
+	ROCKER_TLV_CMD_PORT_STATS_MAX = __ROCKER_TLV_CMD_PORT_STATS_MAX - 1,
+};
+
+enum rocker_port_mode {
+	ROCKER_PORT_MODE_OF_DPA,
+};
+
+enum {
+	ROCKER_TLV_EVENT_UNSPEC,
+	ROCKER_TLV_EVENT_TYPE,	/* u16 */
+	ROCKER_TLV_EVENT_INFO,	/* nest */
+
+	__ROCKER_TLV_EVENT_MAX,
+	ROCKER_TLV_EVENT_MAX = __ROCKER_TLV_EVENT_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_EVENT_TYPE_UNSPEC,
+	ROCKER_TLV_EVENT_TYPE_LINK_CHANGED,
+	ROCKER_TLV_EVENT_TYPE_MAC_VLAN_SEEN,
+
+	__ROCKER_TLV_EVENT_TYPE_MAX,
+	ROCKER_TLV_EVENT_TYPE_MAX = __ROCKER_TLV_EVENT_TYPE_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_EVENT_LINK_CHANGED_UNSPEC,
+	ROCKER_TLV_EVENT_LINK_CHANGED_PPORT,	/* u32 */
+	ROCKER_TLV_EVENT_LINK_CHANGED_LINKUP,	/* u8 */
+
+	__ROCKER_TLV_EVENT_LINK_CHANGED_MAX,
+	ROCKER_TLV_EVENT_LINK_CHANGED_MAX =
+			__ROCKER_TLV_EVENT_LINK_CHANGED_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_EVENT_MAC_VLAN_UNSPEC,
+	ROCKER_TLV_EVENT_MAC_VLAN_PPORT,	/* u32 */
+	ROCKER_TLV_EVENT_MAC_VLAN_MAC,		/* binary */
+	ROCKER_TLV_EVENT_MAC_VLAN_VLAN_ID,	/* __be16 */
+
+	__ROCKER_TLV_EVENT_MAC_VLAN_MAX,
+	ROCKER_TLV_EVENT_MAC_VLAN_MAX = __ROCKER_TLV_EVENT_MAC_VLAN_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_RX_UNSPEC,
+	ROCKER_TLV_RX_FLAGS,		/* u16, see ROCKER_RX_FLAGS_ */
+	ROCKER_TLV_RX_CSUM,		/* u16 */
+	ROCKER_TLV_RX_FRAG_ADDR,	/* u64 */
+	ROCKER_TLV_RX_FRAG_MAX_LEN,	/* u16 */
+	ROCKER_TLV_RX_FRAG_LEN,		/* u16 */
+
+	__ROCKER_TLV_RX_MAX,
+	ROCKER_TLV_RX_MAX = __ROCKER_TLV_RX_MAX - 1,
+};
+
+#define ROCKER_RX_FLAGS_IPV4			BIT(0)
+#define ROCKER_RX_FLAGS_IPV6			BIT(1)
+#define ROCKER_RX_FLAGS_CSUM_CALC		BIT(2)
+#define ROCKER_RX_FLAGS_IPV4_CSUM_GOOD		BIT(3)
+#define ROCKER_RX_FLAGS_IP_FRAG			BIT(4)
+#define ROCKER_RX_FLAGS_TCP			BIT(5)
+#define ROCKER_RX_FLAGS_UDP			BIT(6)
+#define ROCKER_RX_FLAGS_TCP_UDP_CSUM_GOOD	BIT(7)
+#define ROCKER_RX_FLAGS_FWD_OFFLOAD		BIT(8)
+
+enum {
+	ROCKER_TLV_TX_UNSPEC,
+	ROCKER_TLV_TX_OFFLOAD,		/* u8, see ROCKER_TX_OFFLOAD_ */
+	ROCKER_TLV_TX_L3_CSUM_OFF,	/* u16 */
+	ROCKER_TLV_TX_TSO_MSS,		/* u16 */
+	ROCKER_TLV_TX_TSO_HDR_LEN,	/* u16 */
+	ROCKER_TLV_TX_FRAGS,		/* array */
+
+	__ROCKER_TLV_TX_MAX,
+	ROCKER_TLV_TX_MAX = __ROCKER_TLV_TX_MAX - 1,
+};
+
+#define ROCKER_TX_OFFLOAD_NONE		0
+#define ROCKER_TX_OFFLOAD_IP_CSUM	1
+#define ROCKER_TX_OFFLOAD_TCP_UDP_CSUM	2
+#define ROCKER_TX_OFFLOAD_L3_CSUM	3
+#define ROCKER_TX_OFFLOAD_TSO		4
+
+#define ROCKER_TX_FRAGS_MAX		16
+
+enum {
+	ROCKER_TLV_TX_FRAG_UNSPEC,
+	ROCKER_TLV_TX_FRAG,		/* nest */
+
+	__ROCKER_TLV_TX_FRAG_MAX,
+	ROCKER_TLV_TX_FRAG_MAX = __ROCKER_TLV_TX_FRAG_MAX - 1,
+};
+
+enum {
+	ROCKER_TLV_TX_FRAG_ATTR_UNSPEC,
+	ROCKER_TLV_TX_FRAG_ATTR_ADDR,	/* u64 */
+	ROCKER_TLV_TX_FRAG_ATTR_LEN,	/* u16 */
+
+	__ROCKER_TLV_TX_FRAG_ATTR_MAX,
+	ROCKER_TLV_TX_FRAG_ATTR_MAX = __ROCKER_TLV_TX_FRAG_ATTR_MAX - 1,
+};
+
+/* cmd info nested for OF-DPA msgs */
+enum {
+	ROCKER_TLV_OF_DPA_UNSPEC,
+	ROCKER_TLV_OF_DPA_TABLE_ID,		/* u16 */
+	ROCKER_TLV_OF_DPA_PRIORITY,		/* u32 */
+	ROCKER_TLV_OF_DPA_HARDTIME,		/* u32 */
+	ROCKER_TLV_OF_DPA_IDLETIME,		/* u32 */
+	ROCKER_TLV_OF_DPA_COOKIE,		/* u64 */
+	ROCKER_TLV_OF_DPA_IN_PPORT,		/* u32 */
+	ROCKER_TLV_OF_DPA_IN_PPORT_MASK,	/* u32 */
+	ROCKER_TLV_OF_DPA_OUT_PPORT,		/* u32 */
+	ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,	/* u16 */
+	ROCKER_TLV_OF_DPA_GROUP_ID,		/* u32 */
+	ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,	/* u32 */
+	ROCKER_TLV_OF_DPA_GROUP_COUNT,		/* u16 */
+	ROCKER_TLV_OF_DPA_GROUP_IDS,		/* u32 array */
+	ROCKER_TLV_OF_DPA_VLAN_ID,		/* __be16 */
+	ROCKER_TLV_OF_DPA_VLAN_ID_MASK,		/* __be16 */
+	ROCKER_TLV_OF_DPA_VLAN_PCP,		/* __be16 */
+	ROCKER_TLV_OF_DPA_VLAN_PCP_MASK,	/* __be16 */
+	ROCKER_TLV_OF_DPA_VLAN_PCP_ACTION,	/* u8 */
+	ROCKER_TLV_OF_DPA_NEW_VLAN_ID,		/* __be16 */
+	ROCKER_TLV_OF_DPA_NEW_VLAN_PCP,		/* u8 */
+	ROCKER_TLV_OF_DPA_TUNNEL_ID,		/* u32 */
+	ROCKER_TLV_OF_DPA_TUNNEL_LPORT,		/* u32 */
+	ROCKER_TLV_OF_DPA_ETHERTYPE,		/* __be16 */
+	ROCKER_TLV_OF_DPA_DST_MAC,		/* binary */
+	ROCKER_TLV_OF_DPA_DST_MAC_MASK,		/* binary */
+	ROCKER_TLV_OF_DPA_SRC_MAC,		/* binary */
+	ROCKER_TLV_OF_DPA_SRC_MAC_MASK,		/* binary */
+	ROCKER_TLV_OF_DPA_IP_PROTO,		/* u8 */
+	ROCKER_TLV_OF_DPA_IP_PROTO_MASK,	/* u8 */
+	ROCKER_TLV_OF_DPA_IP_DSCP,		/* u8 */
+	ROCKER_TLV_OF_DPA_IP_DSCP_MASK,		/* u8 */
+	ROCKER_TLV_OF_DPA_IP_DSCP_ACTION,	/* u8 */
+	ROCKER_TLV_OF_DPA_NEW_IP_DSCP,		/* u8 */
+	ROCKER_TLV_OF_DPA_IP_ECN,		/* u8 */
+	ROCKER_TLV_OF_DPA_IP_ECN_MASK,		/* u8 */
+	ROCKER_TLV_OF_DPA_DST_IP,		/* __be32 */
+	ROCKER_TLV_OF_DPA_DST_IP_MASK,		/* __be32 */
+	ROCKER_TLV_OF_DPA_SRC_IP,		/* __be32 */
+	ROCKER_TLV_OF_DPA_SRC_IP_MASK,		/* __be32 */
+	ROCKER_TLV_OF_DPA_DST_IPV6,		/* binary */
+	ROCKER_TLV_OF_DPA_DST_IPV6_MASK,	/* binary */
+	ROCKER_TLV_OF_DPA_SRC_IPV6,		/* binary */
+	ROCKER_TLV_OF_DPA_SRC_IPV6_MASK,	/* binary */
+	ROCKER_TLV_OF_DPA_SRC_ARP_IP,		/* __be32 */
+	ROCKER_TLV_OF_DPA_SRC_ARP_IP_MASK,	/* __be32 */
+	ROCKER_TLV_OF_DPA_L4_DST_PORT,		/* __be16 */
+	ROCKER_TLV_OF_DPA_L4_DST_PORT_MASK,	/* __be16 */
+	ROCKER_TLV_OF_DPA_L4_SRC_PORT,		/* __be16 */
+	ROCKER_TLV_OF_DPA_L4_SRC_PORT_MASK,	/* __be16 */
+	ROCKER_TLV_OF_DPA_ICMP_TYPE,		/* u8 */
+	ROCKER_TLV_OF_DPA_ICMP_TYPE_MASK,	/* u8 */
+	ROCKER_TLV_OF_DPA_ICMP_CODE,		/* u8 */
+	ROCKER_TLV_OF_DPA_ICMP_CODE_MASK,	/* u8 */
+	ROCKER_TLV_OF_DPA_IPV6_LABEL,		/* __be32 */
+	ROCKER_TLV_OF_DPA_IPV6_LABEL_MASK,	/* __be32 */
+	ROCKER_TLV_OF_DPA_QUEUE_ID_ACTION,	/* u8 */
+	ROCKER_TLV_OF_DPA_NEW_QUEUE_ID,		/* u8 */
+	ROCKER_TLV_OF_DPA_CLEAR_ACTIONS,	/* u32 */
+	ROCKER_TLV_OF_DPA_POP_VLAN,		/* u8 */
+	ROCKER_TLV_OF_DPA_TTL_CHECK,		/* u8 */
+	ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,	/* u8 */
+
+	__ROCKER_TLV_OF_DPA_MAX,
+	ROCKER_TLV_OF_DPA_MAX = __ROCKER_TLV_OF_DPA_MAX - 1,
+};
+
+/* OF-DPA table IDs */
+
+enum rocker_of_dpa_table_id {
+	ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT = 0,
+	ROCKER_OF_DPA_TABLE_ID_VLAN = 10,
+	ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC = 20,
+	ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING = 30,
+	ROCKER_OF_DPA_TABLE_ID_MULTICAST_ROUTING = 40,
+	ROCKER_OF_DPA_TABLE_ID_BRIDGING = 50,
+	ROCKER_OF_DPA_TABLE_ID_ACL_POLICY = 60,
+};
+
+/* OF-DPA flow stats */
+enum {
+	ROCKER_TLV_OF_DPA_FLOW_STAT_UNSPEC,
+	ROCKER_TLV_OF_DPA_FLOW_STAT_DURATION,	/* u32 */
+	ROCKER_TLV_OF_DPA_FLOW_STAT_RX_PKTS,	/* u64 */
+	ROCKER_TLV_OF_DPA_FLOW_STAT_TX_PKTS,	/* u64 */
+
+	__ROCKER_TLV_OF_DPA_FLOW_STAT_MAX,
+	ROCKER_TLV_OF_DPA_FLOW_STAT_MAX = __ROCKER_TLV_OF_DPA_FLOW_STAT_MAX - 1,
+};
+
+/* OF-DPA group types */
+enum rocker_of_dpa_group_type {
+	ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE = 0,
+	ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE,
+	ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST,
+	ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST,
+	ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD,
+	ROCKER_OF_DPA_GROUP_TYPE_L3_INTERFACE,
+	ROCKER_OF_DPA_GROUP_TYPE_L3_MCAST,
+	ROCKER_OF_DPA_GROUP_TYPE_L3_ECMP,
+	ROCKER_OF_DPA_GROUP_TYPE_L2_OVERLAY,
+};
+
+/* OF-DPA group L2 overlay types */
+enum rocker_of_dpa_overlay_type {
+	ROCKER_OF_DPA_OVERLAY_TYPE_FLOOD_UCAST = 0,
+	ROCKER_OF_DPA_OVERLAY_TYPE_FLOOD_MCAST,
+	ROCKER_OF_DPA_OVERLAY_TYPE_MCAST_UCAST,
+	ROCKER_OF_DPA_OVERLAY_TYPE_MCAST_MCAST,
+};
+
+/* OF-DPA group ID encoding */
+#define ROCKER_GROUP_TYPE_SHIFT 28
+#define ROCKER_GROUP_TYPE_MASK 0xf0000000
+#define ROCKER_GROUP_VLAN_SHIFT 16
+#define ROCKER_GROUP_VLAN_MASK 0x0fff0000
+#define ROCKER_GROUP_PORT_SHIFT 0
+#define ROCKER_GROUP_PORT_MASK 0x0000ffff
+#define ROCKER_GROUP_TUNNEL_ID_SHIFT 12
+#define ROCKER_GROUP_TUNNEL_ID_MASK 0x0ffff000
+#define ROCKER_GROUP_SUBTYPE_SHIFT 10
+#define ROCKER_GROUP_SUBTYPE_MASK 0x00000c00
+#define ROCKER_GROUP_INDEX_SHIFT 0
+#define ROCKER_GROUP_INDEX_MASK 0x0000ffff
+#define ROCKER_GROUP_INDEX_LONG_SHIFT 0
+#define ROCKER_GROUP_INDEX_LONG_MASK 0x0fffffff
+
+#define ROCKER_GROUP_TYPE_GET(group_id) \
+	(((group_id) & ROCKER_GROUP_TYPE_MASK) >> ROCKER_GROUP_TYPE_SHIFT)
+#define ROCKER_GROUP_TYPE_SET(type) \
+	(((type) << ROCKER_GROUP_TYPE_SHIFT) & ROCKER_GROUP_TYPE_MASK)
+#define ROCKER_GROUP_VLAN_GET(group_id) \
+	(((group_id) & ROCKER_GROUP_VLAN_ID_MASK) >> ROCKER_GROUP_VLAN_ID_SHIFT)
+#define ROCKER_GROUP_VLAN_SET(vlan_id) \
+	(((vlan_id) << ROCKER_GROUP_VLAN_SHIFT) & ROCKER_GROUP_VLAN_MASK)
+#define ROCKER_GROUP_PORT_GET(group_id) \
+	(((group_id) & ROCKER_GROUP_PORT_MASK) >> ROCKER_GROUP_PORT_SHIFT)
+#define ROCKER_GROUP_PORT_SET(port) \
+	(((port) << ROCKER_GROUP_PORT_SHIFT) & ROCKER_GROUP_PORT_MASK)
+#define ROCKER_GROUP_INDEX_GET(group_id) \
+	(((group_id) & ROCKER_GROUP_INDEX_MASK) >> ROCKER_GROUP_INDEX_SHIFT)
+#define ROCKER_GROUP_INDEX_SET(index) \
+	(((index) << ROCKER_GROUP_INDEX_SHIFT) & ROCKER_GROUP_INDEX_MASK)
+#define ROCKER_GROUP_INDEX_LONG_GET(group_id) \
+	(((group_id) & ROCKER_GROUP_INDEX_LONG_MASK) >> \
+	 ROCKER_GROUP_INDEX_LONG_SHIFT)
+#define ROCKER_GROUP_INDEX_LONG_SET(index) \
+	(((index) << ROCKER_GROUP_INDEX_LONG_SHIFT) & \
+	 ROCKER_GROUP_INDEX_LONG_MASK)
+
+#define ROCKER_GROUP_NONE 0
+#define ROCKER_GROUP_L2_INTERFACE(vlan_id, port) \
+	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE) |\
+	 ROCKER_GROUP_VLAN_SET(ntohs(vlan_id)) | ROCKER_GROUP_PORT_SET(port))
+#define ROCKER_GROUP_L2_REWRITE(index) \
+	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE) |\
+	 ROCKER_GROUP_INDEX_LONG_SET(index))
+#define ROCKER_GROUP_L2_MCAST(vlan_id, index) \
+	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST) |\
+	 ROCKER_GROUP_VLAN_SET(ntohs(vlan_id)) | ROCKER_GROUP_INDEX_SET(index))
+#define ROCKER_GROUP_L2_FLOOD(vlan_id, index) \
+	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD) |\
+	ROCKER_GROUP_VLAN_SET(ntohs(vlan_id)) | ROCKER_GROUP_INDEX_SET(index))
+#define ROCKER_GROUP_L3_UNICAST(index) \
+	(ROCKER_GROUP_TYPE_SET(ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST) |\
+	 ROCKER_GROUP_INDEX_LONG_SET(index))
+
+/* Rocker general purpose registers */
+#define ROCKER_CONTROL			0x0300
+#define ROCKER_PORT_PHYS_COUNT		0x0304
+#define ROCKER_PORT_PHYS_LINK_STATUS	0x0310 /* 8-byte */
+#define ROCKER_PORT_PHYS_ENABLE		0x0318 /* 8-byte */
+#define ROCKER_SWITCH_ID		0x0320 /* 8-byte */
+
+/* Rocker control bits */
+#define ROCKER_CONTROL_RESET		BIT(0)
+
+#endif
-- 
cgit v0.10.2


From 11ce2ba3d05b3ac330022da1f83b589a4f78569c Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:41 +0100
Subject: rocker: rename rocker.c to rocker_main.c

Since "rocker.c" is going to be split into multiple files, start with
renaming original "rocker.c" file to "rocker_main.c". Multiple code
parts are going to be cut from "rocker_main.c" later on.

Fix couple of checkpatch issues on the way.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/Makefile b/drivers/net/ethernet/rocker/Makefile
index f85fb12f..2bed423 100644
--- a/drivers/net/ethernet/rocker/Makefile
+++ b/drivers/net/ethernet/rocker/Makefile
@@ -3,3 +3,4 @@
 #
 
 obj-$(CONFIG_ROCKER) += rocker.o
+rocker-y := rocker_main.o
diff --git a/drivers/net/ethernet/rocker/rocker.c b/drivers/net/ethernet/rocker/rocker.c
deleted file mode 100644
index 3f18498..0000000
--- a/drivers/net/ethernet/rocker/rocker.c
+++ /dev/null
@@ -1,5492 +0,0 @@
-/*
- * drivers/net/ethernet/rocker/rocker.c - Rocker switch device driver
- * Copyright (c) 2014 Jiri Pirko <jiri@resnulli.us>
- * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/pci.h>
-#include <linux/interrupt.h>
-#include <linux/sched.h>
-#include <linux/wait.h>
-#include <linux/spinlock.h>
-#include <linux/hashtable.h>
-#include <linux/crc32.h>
-#include <linux/sort.h>
-#include <linux/random.h>
-#include <linux/netdevice.h>
-#include <linux/inetdevice.h>
-#include <linux/skbuff.h>
-#include <linux/socket.h>
-#include <linux/etherdevice.h>
-#include <linux/ethtool.h>
-#include <linux/if_ether.h>
-#include <linux/if_vlan.h>
-#include <linux/if_bridge.h>
-#include <linux/bitops.h>
-#include <linux/ctype.h>
-#include <net/switchdev.h>
-#include <net/rtnetlink.h>
-#include <net/ip_fib.h>
-#include <net/netevent.h>
-#include <net/arp.h>
-#include <linux/io-64-nonatomic-lo-hi.h>
-#include <generated/utsrelease.h>
-
-#include "rocker_hw.h"
-
-static const char rocker_driver_name[] = "rocker";
-
-static const struct pci_device_id rocker_pci_id_table[] = {
-	{PCI_VDEVICE(REDHAT, PCI_DEVICE_ID_REDHAT_ROCKER), 0},
-	{0, }
-};
-
-struct rocker_flow_tbl_key {
-	u32 priority;
-	enum rocker_of_dpa_table_id tbl_id;
-	union {
-		struct {
-			u32 in_pport;
-			u32 in_pport_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-		} ig_port;
-		struct {
-			u32 in_pport;
-			__be16 vlan_id;
-			__be16 vlan_id_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-			bool untagged;
-			__be16 new_vlan_id;
-		} vlan;
-		struct {
-			u32 in_pport;
-			u32 in_pport_mask;
-			__be16 eth_type;
-			u8 eth_dst[ETH_ALEN];
-			u8 eth_dst_mask[ETH_ALEN];
-			__be16 vlan_id;
-			__be16 vlan_id_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-			bool copy_to_cpu;
-		} term_mac;
-		struct {
-			__be16 eth_type;
-			__be32 dst4;
-			__be32 dst4_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-			u32 group_id;
-		} ucast_routing;
-		struct {
-			u8 eth_dst[ETH_ALEN];
-			u8 eth_dst_mask[ETH_ALEN];
-			int has_eth_dst;
-			int has_eth_dst_mask;
-			__be16 vlan_id;
-			u32 tunnel_id;
-			enum rocker_of_dpa_table_id goto_tbl;
-			u32 group_id;
-			bool copy_to_cpu;
-		} bridge;
-		struct {
-			u32 in_pport;
-			u32 in_pport_mask;
-			u8 eth_src[ETH_ALEN];
-			u8 eth_src_mask[ETH_ALEN];
-			u8 eth_dst[ETH_ALEN];
-			u8 eth_dst_mask[ETH_ALEN];
-			__be16 eth_type;
-			__be16 vlan_id;
-			__be16 vlan_id_mask;
-			u8 ip_proto;
-			u8 ip_proto_mask;
-			u8 ip_tos;
-			u8 ip_tos_mask;
-			u32 group_id;
-		} acl;
-	};
-};
-
-struct rocker_flow_tbl_entry {
-	struct hlist_node entry;
-	u32 cmd;
-	u64 cookie;
-	struct rocker_flow_tbl_key key;
-	size_t key_len;
-	u32 key_crc32; /* key */
-};
-
-struct rocker_group_tbl_entry {
-	struct hlist_node entry;
-	u32 cmd;
-	u32 group_id; /* key */
-	u16 group_count;
-	u32 *group_ids;
-	union {
-		struct {
-			u8 pop_vlan;
-		} l2_interface;
-		struct {
-			u8 eth_src[ETH_ALEN];
-			u8 eth_dst[ETH_ALEN];
-			__be16 vlan_id;
-			u32 group_id;
-		} l2_rewrite;
-		struct {
-			u8 eth_src[ETH_ALEN];
-			u8 eth_dst[ETH_ALEN];
-			__be16 vlan_id;
-			bool ttl_check;
-			u32 group_id;
-		} l3_unicast;
-	};
-};
-
-struct rocker_fdb_tbl_entry {
-	struct hlist_node entry;
-	u32 key_crc32; /* key */
-	bool learned;
-	unsigned long touched;
-	struct rocker_fdb_tbl_key {
-		struct rocker_port *rocker_port;
-		u8 addr[ETH_ALEN];
-		__be16 vlan_id;
-	} key;
-};
-
-struct rocker_internal_vlan_tbl_entry {
-	struct hlist_node entry;
-	int ifindex; /* key */
-	u32 ref_count;
-	__be16 vlan_id;
-};
-
-struct rocker_neigh_tbl_entry {
-	struct hlist_node entry;
-	__be32 ip_addr; /* key */
-	struct net_device *dev;
-	u32 ref_count;
-	u32 index;
-	u8 eth_dst[ETH_ALEN];
-	bool ttl_check;
-};
-
-struct rocker_desc_info {
-	char *data; /* mapped */
-	size_t data_size;
-	size_t tlv_size;
-	struct rocker_desc *desc;
-	dma_addr_t mapaddr;
-};
-
-struct rocker_dma_ring_info {
-	size_t size;
-	u32 head;
-	u32 tail;
-	struct rocker_desc *desc; /* mapped */
-	dma_addr_t mapaddr;
-	struct rocker_desc_info *desc_info;
-	unsigned int type;
-};
-
-struct rocker;
-
-enum {
-	ROCKER_CTRL_LINK_LOCAL_MCAST,
-	ROCKER_CTRL_LOCAL_ARP,
-	ROCKER_CTRL_IPV4_MCAST,
-	ROCKER_CTRL_IPV6_MCAST,
-	ROCKER_CTRL_DFLT_BRIDGING,
-	ROCKER_CTRL_DFLT_OVS,
-	ROCKER_CTRL_MAX,
-};
-
-#define ROCKER_INTERNAL_VLAN_ID_BASE	0x0f00
-#define ROCKER_N_INTERNAL_VLANS		255
-#define ROCKER_VLAN_BITMAP_LEN		BITS_TO_LONGS(VLAN_N_VID)
-#define ROCKER_INTERNAL_VLAN_BITMAP_LEN	BITS_TO_LONGS(ROCKER_N_INTERNAL_VLANS)
-
-struct rocker_port {
-	struct net_device *dev;
-	struct net_device *bridge_dev;
-	struct rocker *rocker;
-	unsigned int port_number;
-	u32 pport;
-	__be16 internal_vlan_id;
-	int stp_state;
-	u32 brport_flags;
-	unsigned long ageing_time;
-	bool ctrls[ROCKER_CTRL_MAX];
-	unsigned long vlan_bitmap[ROCKER_VLAN_BITMAP_LEN];
-	struct napi_struct napi_tx;
-	struct napi_struct napi_rx;
-	struct rocker_dma_ring_info tx_ring;
-	struct rocker_dma_ring_info rx_ring;
-};
-
-struct rocker {
-	struct pci_dev *pdev;
-	u8 __iomem *hw_addr;
-	struct msix_entry *msix_entries;
-	unsigned int port_count;
-	struct rocker_port **ports;
-	struct {
-		u64 id;
-	} hw;
-	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
-	struct rocker_dma_ring_info cmd_ring;
-	struct rocker_dma_ring_info event_ring;
-	DECLARE_HASHTABLE(flow_tbl, 16);
-	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
-	u64 flow_tbl_next_cookie;
-	DECLARE_HASHTABLE(group_tbl, 16);
-	spinlock_t group_tbl_lock;		/* for group tbl accesses */
-	struct timer_list fdb_cleanup_timer;
-	DECLARE_HASHTABLE(fdb_tbl, 16);
-	spinlock_t fdb_tbl_lock;		/* for fdb tbl accesses */
-	unsigned long internal_vlan_bitmap[ROCKER_INTERNAL_VLAN_BITMAP_LEN];
-	DECLARE_HASHTABLE(internal_vlan_tbl, 8);
-	spinlock_t internal_vlan_tbl_lock;	/* for vlan tbl accesses */
-	DECLARE_HASHTABLE(neigh_tbl, 16);
-	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
-	u32 neigh_tbl_next_index;
-};
-
-static const u8 zero_mac[ETH_ALEN]   = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
-static const u8 ff_mac[ETH_ALEN]     = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
-static const u8 ll_mac[ETH_ALEN]     = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
-static const u8 ll_mask[ETH_ALEN]    = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xf0 };
-static const u8 mcast_mac[ETH_ALEN]  = { 0x01, 0x00, 0x00, 0x00, 0x00, 0x00 };
-static const u8 ipv4_mcast[ETH_ALEN] = { 0x01, 0x00, 0x5e, 0x00, 0x00, 0x00 };
-static const u8 ipv4_mask[ETH_ALEN]  = { 0xff, 0xff, 0xff, 0x80, 0x00, 0x00 };
-static const u8 ipv6_mcast[ETH_ALEN] = { 0x33, 0x33, 0x00, 0x00, 0x00, 0x00 };
-static const u8 ipv6_mask[ETH_ALEN]  = { 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 };
-
-/* Rocker priority levels for flow table entries.  Higher
- * priority match takes precedence over lower priority match.
- */
-
-enum {
-	ROCKER_PRIORITY_UNKNOWN = 0,
-	ROCKER_PRIORITY_IG_PORT = 1,
-	ROCKER_PRIORITY_VLAN = 1,
-	ROCKER_PRIORITY_TERM_MAC_UCAST = 0,
-	ROCKER_PRIORITY_TERM_MAC_MCAST = 1,
-	ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_EXACT = 1,
-	ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_WILD = 2,
-	ROCKER_PRIORITY_BRIDGING_VLAN = 3,
-	ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_EXACT = 1,
-	ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_WILD = 2,
-	ROCKER_PRIORITY_BRIDGING_TENANT = 3,
-	ROCKER_PRIORITY_ACL_CTRL = 3,
-	ROCKER_PRIORITY_ACL_NORMAL = 2,
-	ROCKER_PRIORITY_ACL_DFLT = 1,
-};
-
-static bool rocker_vlan_id_is_internal(__be16 vlan_id)
-{
-	u16 start = ROCKER_INTERNAL_VLAN_ID_BASE;
-	u16 end = 0xffe;
-	u16 _vlan_id = ntohs(vlan_id);
-
-	return (_vlan_id >= start && _vlan_id <= end);
-}
-
-static __be16 rocker_port_vid_to_vlan(const struct rocker_port *rocker_port,
-				      u16 vid, bool *pop_vlan)
-{
-	__be16 vlan_id;
-
-	if (pop_vlan)
-		*pop_vlan = false;
-	vlan_id = htons(vid);
-	if (!vlan_id) {
-		vlan_id = rocker_port->internal_vlan_id;
-		if (pop_vlan)
-			*pop_vlan = true;
-	}
-
-	return vlan_id;
-}
-
-static u16 rocker_port_vlan_to_vid(const struct rocker_port *rocker_port,
-				   __be16 vlan_id)
-{
-	if (rocker_vlan_id_is_internal(vlan_id))
-		return 0;
-
-	return ntohs(vlan_id);
-}
-
-static bool rocker_port_is_bridged(const struct rocker_port *rocker_port)
-{
-	return rocker_port->bridge_dev &&
-	       netif_is_bridge_master(rocker_port->bridge_dev);
-}
-
-static bool rocker_port_is_ovsed(const struct rocker_port *rocker_port)
-{
-	return rocker_port->bridge_dev &&
-	       netif_is_ovs_master(rocker_port->bridge_dev);
-}
-
-#define ROCKER_OP_FLAG_REMOVE		BIT(0)
-#define ROCKER_OP_FLAG_NOWAIT		BIT(1)
-#define ROCKER_OP_FLAG_LEARNED		BIT(2)
-#define ROCKER_OP_FLAG_REFRESH		BIT(3)
-
-static void *__rocker_mem_alloc(struct switchdev_trans *trans, int flags,
-				size_t size)
-{
-	struct switchdev_trans_item *elem = NULL;
-	gfp_t gfp_flags = (flags & ROCKER_OP_FLAG_NOWAIT) ?
-			  GFP_ATOMIC : GFP_KERNEL;
-
-	/* If in transaction prepare phase, allocate the memory
-	 * and enqueue it on a transaction.  If in transaction
-	 * commit phase, dequeue the memory from the transaction
-	 * rather than re-allocating the memory.  The idea is the
-	 * driver code paths for prepare and commit are identical
-	 * so the memory allocated in the prepare phase is the
-	 * memory used in the commit phase.
-	 */
-
-	if (!trans) {
-		elem = kzalloc(size + sizeof(*elem), gfp_flags);
-	} else if (switchdev_trans_ph_prepare(trans)) {
-		elem = kzalloc(size + sizeof(*elem), gfp_flags);
-		if (!elem)
-			return NULL;
-		switchdev_trans_item_enqueue(trans, elem, kfree, elem);
-	} else {
-		elem = switchdev_trans_item_dequeue(trans);
-	}
-
-	return elem ? elem + 1 : NULL;
-}
-
-static void *rocker_kzalloc(struct switchdev_trans *trans, int flags,
-			    size_t size)
-{
-	return __rocker_mem_alloc(trans, flags, size);
-}
-
-static void *rocker_kcalloc(struct switchdev_trans *trans, int flags,
-			    size_t n, size_t size)
-{
-	return __rocker_mem_alloc(trans, flags, n * size);
-}
-
-static void rocker_kfree(struct switchdev_trans *trans, const void *mem)
-{
-	struct switchdev_trans_item *elem;
-
-	/* Frees are ignored if in transaction prepare phase.  The
-	 * memory remains on the per-port list until freed in the
-	 * commit phase.
-	 */
-
-	if (switchdev_trans_ph_prepare(trans))
-		return;
-
-	elem = (struct switchdev_trans_item *) mem - 1;
-	kfree(elem);
-}
-
-struct rocker_wait {
-	wait_queue_head_t wait;
-	bool done;
-	bool nowait;
-};
-
-static void rocker_wait_reset(struct rocker_wait *wait)
-{
-	wait->done = false;
-	wait->nowait = false;
-}
-
-static void rocker_wait_init(struct rocker_wait *wait)
-{
-	init_waitqueue_head(&wait->wait);
-	rocker_wait_reset(wait);
-}
-
-static struct rocker_wait *rocker_wait_create(struct rocker_port *rocker_port,
-					      struct switchdev_trans *trans,
-					      int flags)
-{
-	struct rocker_wait *wait;
-
-	wait = rocker_kzalloc(trans, flags, sizeof(*wait));
-	if (!wait)
-		return NULL;
-	rocker_wait_init(wait);
-	return wait;
-}
-
-static void rocker_wait_destroy(struct switchdev_trans *trans,
-				struct rocker_wait *wait)
-{
-	rocker_kfree(trans, wait);
-}
-
-static bool rocker_wait_event_timeout(struct rocker_wait *wait,
-				      unsigned long timeout)
-{
-	wait_event_timeout(wait->wait, wait->done, HZ / 10);
-	if (!wait->done)
-		return false;
-	return true;
-}
-
-static void rocker_wait_wake_up(struct rocker_wait *wait)
-{
-	wait->done = true;
-	wake_up(&wait->wait);
-}
-
-static u32 rocker_msix_vector(const struct rocker *rocker, unsigned int vector)
-{
-	return rocker->msix_entries[vector].vector;
-}
-
-static u32 rocker_msix_tx_vector(const struct rocker_port *rocker_port)
-{
-	return rocker_msix_vector(rocker_port->rocker,
-				  ROCKER_MSIX_VEC_TX(rocker_port->port_number));
-}
-
-static u32 rocker_msix_rx_vector(const struct rocker_port *rocker_port)
-{
-	return rocker_msix_vector(rocker_port->rocker,
-				  ROCKER_MSIX_VEC_RX(rocker_port->port_number));
-}
-
-#define rocker_write32(rocker, reg, val)	\
-	writel((val), (rocker)->hw_addr + (ROCKER_ ## reg))
-#define rocker_read32(rocker, reg)	\
-	readl((rocker)->hw_addr + (ROCKER_ ## reg))
-#define rocker_write64(rocker, reg, val)	\
-	writeq((val), (rocker)->hw_addr + (ROCKER_ ## reg))
-#define rocker_read64(rocker, reg)	\
-	readq((rocker)->hw_addr + (ROCKER_ ## reg))
-
-/*****************************
- * HW basic testing functions
- *****************************/
-
-static int rocker_reg_test(const struct rocker *rocker)
-{
-	const struct pci_dev *pdev = rocker->pdev;
-	u64 test_reg;
-	u64 rnd;
-
-	rnd = prandom_u32();
-	rnd >>= 1;
-	rocker_write32(rocker, TEST_REG, rnd);
-	test_reg = rocker_read32(rocker, TEST_REG);
-	if (test_reg != rnd * 2) {
-		dev_err(&pdev->dev, "unexpected 32bit register value %08llx, expected %08llx\n",
-			test_reg, rnd * 2);
-		return -EIO;
-	}
-
-	rnd = prandom_u32();
-	rnd <<= 31;
-	rnd |= prandom_u32();
-	rocker_write64(rocker, TEST_REG64, rnd);
-	test_reg = rocker_read64(rocker, TEST_REG64);
-	if (test_reg != rnd * 2) {
-		dev_err(&pdev->dev, "unexpected 64bit register value %16llx, expected %16llx\n",
-			test_reg, rnd * 2);
-		return -EIO;
-	}
-
-	return 0;
-}
-
-static int rocker_dma_test_one(const struct rocker *rocker,
-			       struct rocker_wait *wait, u32 test_type,
-			       dma_addr_t dma_handle, const unsigned char *buf,
-			       const unsigned char *expect, size_t size)
-{
-	const struct pci_dev *pdev = rocker->pdev;
-	int i;
-
-	rocker_wait_reset(wait);
-	rocker_write32(rocker, TEST_DMA_CTRL, test_type);
-
-	if (!rocker_wait_event_timeout(wait, HZ / 10)) {
-		dev_err(&pdev->dev, "no interrupt received within a timeout\n");
-		return -EIO;
-	}
-
-	for (i = 0; i < size; i++) {
-		if (buf[i] != expect[i]) {
-			dev_err(&pdev->dev, "unexpected memory content %02x at byte %x\n, %02x expected",
-				buf[i], i, expect[i]);
-			return -EIO;
-		}
-	}
-	return 0;
-}
-
-#define ROCKER_TEST_DMA_BUF_SIZE (PAGE_SIZE * 4)
-#define ROCKER_TEST_DMA_FILL_PATTERN 0x96
-
-static int rocker_dma_test_offset(const struct rocker *rocker,
-				  struct rocker_wait *wait, int offset)
-{
-	struct pci_dev *pdev = rocker->pdev;
-	unsigned char *alloc;
-	unsigned char *buf;
-	unsigned char *expect;
-	dma_addr_t dma_handle;
-	int i;
-	int err;
-
-	alloc = kzalloc(ROCKER_TEST_DMA_BUF_SIZE * 2 + offset,
-			GFP_KERNEL | GFP_DMA);
-	if (!alloc)
-		return -ENOMEM;
-	buf = alloc + offset;
-	expect = buf + ROCKER_TEST_DMA_BUF_SIZE;
-
-	dma_handle = pci_map_single(pdev, buf, ROCKER_TEST_DMA_BUF_SIZE,
-				    PCI_DMA_BIDIRECTIONAL);
-	if (pci_dma_mapping_error(pdev, dma_handle)) {
-		err = -EIO;
-		goto free_alloc;
-	}
-
-	rocker_write64(rocker, TEST_DMA_ADDR, dma_handle);
-	rocker_write32(rocker, TEST_DMA_SIZE, ROCKER_TEST_DMA_BUF_SIZE);
-
-	memset(expect, ROCKER_TEST_DMA_FILL_PATTERN, ROCKER_TEST_DMA_BUF_SIZE);
-	err = rocker_dma_test_one(rocker, wait, ROCKER_TEST_DMA_CTRL_FILL,
-				  dma_handle, buf, expect,
-				  ROCKER_TEST_DMA_BUF_SIZE);
-	if (err)
-		goto unmap;
-
-	memset(expect, 0, ROCKER_TEST_DMA_BUF_SIZE);
-	err = rocker_dma_test_one(rocker, wait, ROCKER_TEST_DMA_CTRL_CLEAR,
-				  dma_handle, buf, expect,
-				  ROCKER_TEST_DMA_BUF_SIZE);
-	if (err)
-		goto unmap;
-
-	prandom_bytes(buf, ROCKER_TEST_DMA_BUF_SIZE);
-	for (i = 0; i < ROCKER_TEST_DMA_BUF_SIZE; i++)
-		expect[i] = ~buf[i];
-	err = rocker_dma_test_one(rocker, wait, ROCKER_TEST_DMA_CTRL_INVERT,
-				  dma_handle, buf, expect,
-				  ROCKER_TEST_DMA_BUF_SIZE);
-	if (err)
-		goto unmap;
-
-unmap:
-	pci_unmap_single(pdev, dma_handle, ROCKER_TEST_DMA_BUF_SIZE,
-			 PCI_DMA_BIDIRECTIONAL);
-free_alloc:
-	kfree(alloc);
-
-	return err;
-}
-
-static int rocker_dma_test(const struct rocker *rocker,
-			   struct rocker_wait *wait)
-{
-	int i;
-	int err;
-
-	for (i = 0; i < 8; i++) {
-		err = rocker_dma_test_offset(rocker, wait, i);
-		if (err)
-			return err;
-	}
-	return 0;
-}
-
-static irqreturn_t rocker_test_irq_handler(int irq, void *dev_id)
-{
-	struct rocker_wait *wait = dev_id;
-
-	rocker_wait_wake_up(wait);
-
-	return IRQ_HANDLED;
-}
-
-static int rocker_basic_hw_test(const struct rocker *rocker)
-{
-	const struct pci_dev *pdev = rocker->pdev;
-	struct rocker_wait wait;
-	int err;
-
-	err = rocker_reg_test(rocker);
-	if (err) {
-		dev_err(&pdev->dev, "reg test failed\n");
-		return err;
-	}
-
-	err = request_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_TEST),
-			  rocker_test_irq_handler, 0,
-			  rocker_driver_name, &wait);
-	if (err) {
-		dev_err(&pdev->dev, "cannot assign test irq\n");
-		return err;
-	}
-
-	rocker_wait_init(&wait);
-	rocker_write32(rocker, TEST_IRQ, ROCKER_MSIX_VEC_TEST);
-
-	if (!rocker_wait_event_timeout(&wait, HZ / 10)) {
-		dev_err(&pdev->dev, "no interrupt received within a timeout\n");
-		err = -EIO;
-		goto free_irq;
-	}
-
-	err = rocker_dma_test(rocker, &wait);
-	if (err)
-		dev_err(&pdev->dev, "dma test failed\n");
-
-free_irq:
-	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_TEST), &wait);
-	return err;
-}
-
-/******
- * TLV
- ******/
-
-#define ROCKER_TLV_ALIGNTO 8U
-#define ROCKER_TLV_ALIGN(len) \
-	(((len) + ROCKER_TLV_ALIGNTO - 1) & ~(ROCKER_TLV_ALIGNTO - 1))
-#define ROCKER_TLV_HDRLEN ROCKER_TLV_ALIGN(sizeof(struct rocker_tlv))
-
-/*  <------- ROCKER_TLV_HDRLEN -------> <--- ROCKER_TLV_ALIGN(payload) --->
- * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
- * |             Header          | Pad |           Payload           | Pad |
- * |      (struct rocker_tlv)    | ing |                             | ing |
- * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
- *  <--------------------------- tlv->len -------------------------->
- */
-
-static struct rocker_tlv *rocker_tlv_next(const struct rocker_tlv *tlv,
-					  int *remaining)
-{
-	int totlen = ROCKER_TLV_ALIGN(tlv->len);
-
-	*remaining -= totlen;
-	return (struct rocker_tlv *) ((char *) tlv + totlen);
-}
-
-static int rocker_tlv_ok(const struct rocker_tlv *tlv, int remaining)
-{
-	return remaining >= (int) ROCKER_TLV_HDRLEN &&
-	       tlv->len >= ROCKER_TLV_HDRLEN &&
-	       tlv->len <= remaining;
-}
-
-#define rocker_tlv_for_each(pos, head, len, rem)	\
-	for (pos = head, rem = len;			\
-	     rocker_tlv_ok(pos, rem);			\
-	     pos = rocker_tlv_next(pos, &(rem)))
-
-#define rocker_tlv_for_each_nested(pos, tlv, rem)	\
-	rocker_tlv_for_each(pos, rocker_tlv_data(tlv),	\
-			    rocker_tlv_len(tlv), rem)
-
-static int rocker_tlv_attr_size(int payload)
-{
-	return ROCKER_TLV_HDRLEN + payload;
-}
-
-static int rocker_tlv_total_size(int payload)
-{
-	return ROCKER_TLV_ALIGN(rocker_tlv_attr_size(payload));
-}
-
-static int rocker_tlv_padlen(int payload)
-{
-	return rocker_tlv_total_size(payload) - rocker_tlv_attr_size(payload);
-}
-
-static int rocker_tlv_type(const struct rocker_tlv *tlv)
-{
-	return tlv->type;
-}
-
-static void *rocker_tlv_data(const struct rocker_tlv *tlv)
-{
-	return (char *) tlv + ROCKER_TLV_HDRLEN;
-}
-
-static int rocker_tlv_len(const struct rocker_tlv *tlv)
-{
-	return tlv->len - ROCKER_TLV_HDRLEN;
-}
-
-static u8 rocker_tlv_get_u8(const struct rocker_tlv *tlv)
-{
-	return *(u8 *) rocker_tlv_data(tlv);
-}
-
-static u16 rocker_tlv_get_u16(const struct rocker_tlv *tlv)
-{
-	return *(u16 *) rocker_tlv_data(tlv);
-}
-
-static __be16 rocker_tlv_get_be16(const struct rocker_tlv *tlv)
-{
-	return *(__be16 *) rocker_tlv_data(tlv);
-}
-
-static u32 rocker_tlv_get_u32(const struct rocker_tlv *tlv)
-{
-	return *(u32 *) rocker_tlv_data(tlv);
-}
-
-static u64 rocker_tlv_get_u64(const struct rocker_tlv *tlv)
-{
-	return *(u64 *) rocker_tlv_data(tlv);
-}
-
-static void rocker_tlv_parse(const struct rocker_tlv **tb, int maxtype,
-			     const char *buf, int buf_len)
-{
-	const struct rocker_tlv *tlv;
-	const struct rocker_tlv *head = (const struct rocker_tlv *) buf;
-	int rem;
-
-	memset(tb, 0, sizeof(struct rocker_tlv *) * (maxtype + 1));
-
-	rocker_tlv_for_each(tlv, head, buf_len, rem) {
-		u32 type = rocker_tlv_type(tlv);
-
-		if (type > 0 && type <= maxtype)
-			tb[type] = tlv;
-	}
-}
-
-static void rocker_tlv_parse_nested(const struct rocker_tlv **tb, int maxtype,
-				    const struct rocker_tlv *tlv)
-{
-	rocker_tlv_parse(tb, maxtype, rocker_tlv_data(tlv),
-			 rocker_tlv_len(tlv));
-}
-
-static void rocker_tlv_parse_desc(const struct rocker_tlv **tb, int maxtype,
-				  const struct rocker_desc_info *desc_info)
-{
-	rocker_tlv_parse(tb, maxtype, desc_info->data,
-			 desc_info->desc->tlv_size);
-}
-
-static struct rocker_tlv *rocker_tlv_start(struct rocker_desc_info *desc_info)
-{
-	return (struct rocker_tlv *) ((char *) desc_info->data +
-					       desc_info->tlv_size);
-}
-
-static int rocker_tlv_put(struct rocker_desc_info *desc_info,
-			  int attrtype, int attrlen, const void *data)
-{
-	int tail_room = desc_info->data_size - desc_info->tlv_size;
-	int total_size = rocker_tlv_total_size(attrlen);
-	struct rocker_tlv *tlv;
-
-	if (unlikely(tail_room < total_size))
-		return -EMSGSIZE;
-
-	tlv = rocker_tlv_start(desc_info);
-	desc_info->tlv_size += total_size;
-	tlv->type = attrtype;
-	tlv->len = rocker_tlv_attr_size(attrlen);
-	memcpy(rocker_tlv_data(tlv), data, attrlen);
-	memset((char *) tlv + tlv->len, 0, rocker_tlv_padlen(attrlen));
-	return 0;
-}
-
-static int rocker_tlv_put_u8(struct rocker_desc_info *desc_info,
-			     int attrtype, u8 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u8), &value);
-}
-
-static int rocker_tlv_put_u16(struct rocker_desc_info *desc_info,
-			      int attrtype, u16 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u16), &value);
-}
-
-static int rocker_tlv_put_be16(struct rocker_desc_info *desc_info,
-			       int attrtype, __be16 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(__be16), &value);
-}
-
-static int rocker_tlv_put_u32(struct rocker_desc_info *desc_info,
-			      int attrtype, u32 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u32), &value);
-}
-
-static int rocker_tlv_put_be32(struct rocker_desc_info *desc_info,
-			       int attrtype, __be32 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(__be32), &value);
-}
-
-static int rocker_tlv_put_u64(struct rocker_desc_info *desc_info,
-			      int attrtype, u64 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u64), &value);
-}
-
-static struct rocker_tlv *
-rocker_tlv_nest_start(struct rocker_desc_info *desc_info, int attrtype)
-{
-	struct rocker_tlv *start = rocker_tlv_start(desc_info);
-
-	if (rocker_tlv_put(desc_info, attrtype, 0, NULL) < 0)
-		return NULL;
-
-	return start;
-}
-
-static void rocker_tlv_nest_end(struct rocker_desc_info *desc_info,
-				struct rocker_tlv *start)
-{
-	start->len = (char *) rocker_tlv_start(desc_info) - (char *) start;
-}
-
-static void rocker_tlv_nest_cancel(struct rocker_desc_info *desc_info,
-				   const struct rocker_tlv *start)
-{
-	desc_info->tlv_size = (const char *) start - desc_info->data;
-}
-
-/******************************************
- * DMA rings and descriptors manipulations
- ******************************************/
-
-static u32 __pos_inc(u32 pos, size_t limit)
-{
-	return ++pos == limit ? 0 : pos;
-}
-
-static int rocker_desc_err(const struct rocker_desc_info *desc_info)
-{
-	int err = desc_info->desc->comp_err & ~ROCKER_DMA_DESC_COMP_ERR_GEN;
-
-	switch (err) {
-	case ROCKER_OK:
-		return 0;
-	case -ROCKER_ENOENT:
-		return -ENOENT;
-	case -ROCKER_ENXIO:
-		return -ENXIO;
-	case -ROCKER_ENOMEM:
-		return -ENOMEM;
-	case -ROCKER_EEXIST:
-		return -EEXIST;
-	case -ROCKER_EINVAL:
-		return -EINVAL;
-	case -ROCKER_EMSGSIZE:
-		return -EMSGSIZE;
-	case -ROCKER_ENOTSUP:
-		return -EOPNOTSUPP;
-	case -ROCKER_ENOBUFS:
-		return -ENOBUFS;
-	}
-
-	return -EINVAL;
-}
-
-static void rocker_desc_gen_clear(const struct rocker_desc_info *desc_info)
-{
-	desc_info->desc->comp_err &= ~ROCKER_DMA_DESC_COMP_ERR_GEN;
-}
-
-static bool rocker_desc_gen(const struct rocker_desc_info *desc_info)
-{
-	u32 comp_err = desc_info->desc->comp_err;
-
-	return comp_err & ROCKER_DMA_DESC_COMP_ERR_GEN ? true : false;
-}
-
-static void *rocker_desc_cookie_ptr_get(const struct rocker_desc_info *desc_info)
-{
-	return (void *)(uintptr_t)desc_info->desc->cookie;
-}
-
-static void rocker_desc_cookie_ptr_set(const struct rocker_desc_info *desc_info,
-				       void *ptr)
-{
-	desc_info->desc->cookie = (uintptr_t) ptr;
-}
-
-static struct rocker_desc_info *
-rocker_desc_head_get(const struct rocker_dma_ring_info *info)
-{
-	static struct rocker_desc_info *desc_info;
-	u32 head = __pos_inc(info->head, info->size);
-
-	desc_info = &info->desc_info[info->head];
-	if (head == info->tail)
-		return NULL; /* ring full */
-	desc_info->tlv_size = 0;
-	return desc_info;
-}
-
-static void rocker_desc_commit(const struct rocker_desc_info *desc_info)
-{
-	desc_info->desc->buf_size = desc_info->data_size;
-	desc_info->desc->tlv_size = desc_info->tlv_size;
-}
-
-static void rocker_desc_head_set(const struct rocker *rocker,
-				 struct rocker_dma_ring_info *info,
-				 const struct rocker_desc_info *desc_info)
-{
-	u32 head = __pos_inc(info->head, info->size);
-
-	BUG_ON(head == info->tail);
-	rocker_desc_commit(desc_info);
-	info->head = head;
-	rocker_write32(rocker, DMA_DESC_HEAD(info->type), head);
-}
-
-static struct rocker_desc_info *
-rocker_desc_tail_get(struct rocker_dma_ring_info *info)
-{
-	static struct rocker_desc_info *desc_info;
-
-	if (info->tail == info->head)
-		return NULL; /* nothing to be done between head and tail */
-	desc_info = &info->desc_info[info->tail];
-	if (!rocker_desc_gen(desc_info))
-		return NULL; /* gen bit not set, desc is not ready yet */
-	info->tail = __pos_inc(info->tail, info->size);
-	desc_info->tlv_size = desc_info->desc->tlv_size;
-	return desc_info;
-}
-
-static void rocker_dma_ring_credits_set(const struct rocker *rocker,
-					const struct rocker_dma_ring_info *info,
-					u32 credits)
-{
-	if (credits)
-		rocker_write32(rocker, DMA_DESC_CREDITS(info->type), credits);
-}
-
-static unsigned long rocker_dma_ring_size_fix(size_t size)
-{
-	return max(ROCKER_DMA_SIZE_MIN,
-		   min(roundup_pow_of_two(size), ROCKER_DMA_SIZE_MAX));
-}
-
-static int rocker_dma_ring_create(const struct rocker *rocker,
-				  unsigned int type,
-				  size_t size,
-				  struct rocker_dma_ring_info *info)
-{
-	int i;
-
-	BUG_ON(size != rocker_dma_ring_size_fix(size));
-	info->size = size;
-	info->type = type;
-	info->head = 0;
-	info->tail = 0;
-	info->desc_info = kcalloc(info->size, sizeof(*info->desc_info),
-				  GFP_KERNEL);
-	if (!info->desc_info)
-		return -ENOMEM;
-
-	info->desc = pci_alloc_consistent(rocker->pdev,
-					  info->size * sizeof(*info->desc),
-					  &info->mapaddr);
-	if (!info->desc) {
-		kfree(info->desc_info);
-		return -ENOMEM;
-	}
-
-	for (i = 0; i < info->size; i++)
-		info->desc_info[i].desc = &info->desc[i];
-
-	rocker_write32(rocker, DMA_DESC_CTRL(info->type),
-		       ROCKER_DMA_DESC_CTRL_RESET);
-	rocker_write64(rocker, DMA_DESC_ADDR(info->type), info->mapaddr);
-	rocker_write32(rocker, DMA_DESC_SIZE(info->type), info->size);
-
-	return 0;
-}
-
-static void rocker_dma_ring_destroy(const struct rocker *rocker,
-				    const struct rocker_dma_ring_info *info)
-{
-	rocker_write64(rocker, DMA_DESC_ADDR(info->type), 0);
-
-	pci_free_consistent(rocker->pdev,
-			    info->size * sizeof(struct rocker_desc),
-			    info->desc, info->mapaddr);
-	kfree(info->desc_info);
-}
-
-static void rocker_dma_ring_pass_to_producer(const struct rocker *rocker,
-					     struct rocker_dma_ring_info *info)
-{
-	int i;
-
-	BUG_ON(info->head || info->tail);
-
-	/* When ring is consumer, we need to advance head for each desc.
-	 * That tells hw that the desc is ready to be used by it.
-	 */
-	for (i = 0; i < info->size - 1; i++)
-		rocker_desc_head_set(rocker, info, &info->desc_info[i]);
-	rocker_desc_commit(&info->desc_info[i]);
-}
-
-static int rocker_dma_ring_bufs_alloc(const struct rocker *rocker,
-				      const struct rocker_dma_ring_info *info,
-				      int direction, size_t buf_size)
-{
-	struct pci_dev *pdev = rocker->pdev;
-	int i;
-	int err;
-
-	for (i = 0; i < info->size; i++) {
-		struct rocker_desc_info *desc_info = &info->desc_info[i];
-		struct rocker_desc *desc = &info->desc[i];
-		dma_addr_t dma_handle;
-		char *buf;
-
-		buf = kzalloc(buf_size, GFP_KERNEL | GFP_DMA);
-		if (!buf) {
-			err = -ENOMEM;
-			goto rollback;
-		}
-
-		dma_handle = pci_map_single(pdev, buf, buf_size, direction);
-		if (pci_dma_mapping_error(pdev, dma_handle)) {
-			kfree(buf);
-			err = -EIO;
-			goto rollback;
-		}
-
-		desc_info->data = buf;
-		desc_info->data_size = buf_size;
-		dma_unmap_addr_set(desc_info, mapaddr, dma_handle);
-
-		desc->buf_addr = dma_handle;
-		desc->buf_size = buf_size;
-	}
-	return 0;
-
-rollback:
-	for (i--; i >= 0; i--) {
-		const struct rocker_desc_info *desc_info = &info->desc_info[i];
-
-		pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr),
-				 desc_info->data_size, direction);
-		kfree(desc_info->data);
-	}
-	return err;
-}
-
-static void rocker_dma_ring_bufs_free(const struct rocker *rocker,
-				      const struct rocker_dma_ring_info *info,
-				      int direction)
-{
-	struct pci_dev *pdev = rocker->pdev;
-	int i;
-
-	for (i = 0; i < info->size; i++) {
-		const struct rocker_desc_info *desc_info = &info->desc_info[i];
-		struct rocker_desc *desc = &info->desc[i];
-
-		desc->buf_addr = 0;
-		desc->buf_size = 0;
-		pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr),
-				 desc_info->data_size, direction);
-		kfree(desc_info->data);
-	}
-}
-
-static int rocker_dma_rings_init(struct rocker *rocker)
-{
-	const struct pci_dev *pdev = rocker->pdev;
-	int err;
-
-	err = rocker_dma_ring_create(rocker, ROCKER_DMA_CMD,
-				     ROCKER_DMA_CMD_DEFAULT_SIZE,
-				     &rocker->cmd_ring);
-	if (err) {
-		dev_err(&pdev->dev, "failed to create command dma ring\n");
-		return err;
-	}
-
-	spin_lock_init(&rocker->cmd_ring_lock);
-
-	err = rocker_dma_ring_bufs_alloc(rocker, &rocker->cmd_ring,
-					 PCI_DMA_BIDIRECTIONAL, PAGE_SIZE);
-	if (err) {
-		dev_err(&pdev->dev, "failed to alloc command dma ring buffers\n");
-		goto err_dma_cmd_ring_bufs_alloc;
-	}
-
-	err = rocker_dma_ring_create(rocker, ROCKER_DMA_EVENT,
-				     ROCKER_DMA_EVENT_DEFAULT_SIZE,
-				     &rocker->event_ring);
-	if (err) {
-		dev_err(&pdev->dev, "failed to create event dma ring\n");
-		goto err_dma_event_ring_create;
-	}
-
-	err = rocker_dma_ring_bufs_alloc(rocker, &rocker->event_ring,
-					 PCI_DMA_FROMDEVICE, PAGE_SIZE);
-	if (err) {
-		dev_err(&pdev->dev, "failed to alloc event dma ring buffers\n");
-		goto err_dma_event_ring_bufs_alloc;
-	}
-	rocker_dma_ring_pass_to_producer(rocker, &rocker->event_ring);
-	return 0;
-
-err_dma_event_ring_bufs_alloc:
-	rocker_dma_ring_destroy(rocker, &rocker->event_ring);
-err_dma_event_ring_create:
-	rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring,
-				  PCI_DMA_BIDIRECTIONAL);
-err_dma_cmd_ring_bufs_alloc:
-	rocker_dma_ring_destroy(rocker, &rocker->cmd_ring);
-	return err;
-}
-
-static void rocker_dma_rings_fini(struct rocker *rocker)
-{
-	rocker_dma_ring_bufs_free(rocker, &rocker->event_ring,
-				  PCI_DMA_BIDIRECTIONAL);
-	rocker_dma_ring_destroy(rocker, &rocker->event_ring);
-	rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring,
-				  PCI_DMA_BIDIRECTIONAL);
-	rocker_dma_ring_destroy(rocker, &rocker->cmd_ring);
-}
-
-static int rocker_dma_rx_ring_skb_map(const struct rocker_port *rocker_port,
-				      struct rocker_desc_info *desc_info,
-				      struct sk_buff *skb, size_t buf_len)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct pci_dev *pdev = rocker->pdev;
-	dma_addr_t dma_handle;
-
-	dma_handle = pci_map_single(pdev, skb->data, buf_len,
-				    PCI_DMA_FROMDEVICE);
-	if (pci_dma_mapping_error(pdev, dma_handle))
-		return -EIO;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_RX_FRAG_ADDR, dma_handle))
-		goto tlv_put_failure;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_RX_FRAG_MAX_LEN, buf_len))
-		goto tlv_put_failure;
-	return 0;
-
-tlv_put_failure:
-	pci_unmap_single(pdev, dma_handle, buf_len, PCI_DMA_FROMDEVICE);
-	desc_info->tlv_size = 0;
-	return -EMSGSIZE;
-}
-
-static size_t rocker_port_rx_buf_len(const struct rocker_port *rocker_port)
-{
-	return rocker_port->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
-}
-
-static int rocker_dma_rx_ring_skb_alloc(const struct rocker_port *rocker_port,
-					struct rocker_desc_info *desc_info)
-{
-	struct net_device *dev = rocker_port->dev;
-	struct sk_buff *skb;
-	size_t buf_len = rocker_port_rx_buf_len(rocker_port);
-	int err;
-
-	/* Ensure that hw will see tlv_size zero in case of an error.
-	 * That tells hw to use another descriptor.
-	 */
-	rocker_desc_cookie_ptr_set(desc_info, NULL);
-	desc_info->tlv_size = 0;
-
-	skb = netdev_alloc_skb_ip_align(dev, buf_len);
-	if (!skb)
-		return -ENOMEM;
-	err = rocker_dma_rx_ring_skb_map(rocker_port, desc_info, skb, buf_len);
-	if (err) {
-		dev_kfree_skb_any(skb);
-		return err;
-	}
-	rocker_desc_cookie_ptr_set(desc_info, skb);
-	return 0;
-}
-
-static void rocker_dma_rx_ring_skb_unmap(const struct rocker *rocker,
-					 const struct rocker_tlv **attrs)
-{
-	struct pci_dev *pdev = rocker->pdev;
-	dma_addr_t dma_handle;
-	size_t len;
-
-	if (!attrs[ROCKER_TLV_RX_FRAG_ADDR] ||
-	    !attrs[ROCKER_TLV_RX_FRAG_MAX_LEN])
-		return;
-	dma_handle = rocker_tlv_get_u64(attrs[ROCKER_TLV_RX_FRAG_ADDR]);
-	len = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FRAG_MAX_LEN]);
-	pci_unmap_single(pdev, dma_handle, len, PCI_DMA_FROMDEVICE);
-}
-
-static void rocker_dma_rx_ring_skb_free(const struct rocker *rocker,
-					const struct rocker_desc_info *desc_info)
-{
-	const struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1];
-	struct sk_buff *skb = rocker_desc_cookie_ptr_get(desc_info);
-
-	if (!skb)
-		return;
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_RX_MAX, desc_info);
-	rocker_dma_rx_ring_skb_unmap(rocker, attrs);
-	dev_kfree_skb_any(skb);
-}
-
-static int rocker_dma_rx_ring_skbs_alloc(const struct rocker_port *rocker_port)
-{
-	const struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring;
-	const struct rocker *rocker = rocker_port->rocker;
-	int i;
-	int err;
-
-	for (i = 0; i < rx_ring->size; i++) {
-		err = rocker_dma_rx_ring_skb_alloc(rocker_port,
-						   &rx_ring->desc_info[i]);
-		if (err)
-			goto rollback;
-	}
-	return 0;
-
-rollback:
-	for (i--; i >= 0; i--)
-		rocker_dma_rx_ring_skb_free(rocker, &rx_ring->desc_info[i]);
-	return err;
-}
-
-static void rocker_dma_rx_ring_skbs_free(const struct rocker_port *rocker_port)
-{
-	const struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring;
-	const struct rocker *rocker = rocker_port->rocker;
-	int i;
-
-	for (i = 0; i < rx_ring->size; i++)
-		rocker_dma_rx_ring_skb_free(rocker, &rx_ring->desc_info[i]);
-}
-
-static int rocker_port_dma_rings_init(struct rocker_port *rocker_port)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	int err;
-
-	err = rocker_dma_ring_create(rocker,
-				     ROCKER_DMA_TX(rocker_port->port_number),
-				     ROCKER_DMA_TX_DEFAULT_SIZE,
-				     &rocker_port->tx_ring);
-	if (err) {
-		netdev_err(rocker_port->dev, "failed to create tx dma ring\n");
-		return err;
-	}
-
-	err = rocker_dma_ring_bufs_alloc(rocker, &rocker_port->tx_ring,
-					 PCI_DMA_TODEVICE,
-					 ROCKER_DMA_TX_DESC_SIZE);
-	if (err) {
-		netdev_err(rocker_port->dev, "failed to alloc tx dma ring buffers\n");
-		goto err_dma_tx_ring_bufs_alloc;
-	}
-
-	err = rocker_dma_ring_create(rocker,
-				     ROCKER_DMA_RX(rocker_port->port_number),
-				     ROCKER_DMA_RX_DEFAULT_SIZE,
-				     &rocker_port->rx_ring);
-	if (err) {
-		netdev_err(rocker_port->dev, "failed to create rx dma ring\n");
-		goto err_dma_rx_ring_create;
-	}
-
-	err = rocker_dma_ring_bufs_alloc(rocker, &rocker_port->rx_ring,
-					 PCI_DMA_BIDIRECTIONAL,
-					 ROCKER_DMA_RX_DESC_SIZE);
-	if (err) {
-		netdev_err(rocker_port->dev, "failed to alloc rx dma ring buffers\n");
-		goto err_dma_rx_ring_bufs_alloc;
-	}
-
-	err = rocker_dma_rx_ring_skbs_alloc(rocker_port);
-	if (err) {
-		netdev_err(rocker_port->dev, "failed to alloc rx dma ring skbs\n");
-		goto err_dma_rx_ring_skbs_alloc;
-	}
-	rocker_dma_ring_pass_to_producer(rocker, &rocker_port->rx_ring);
-
-	return 0;
-
-err_dma_rx_ring_skbs_alloc:
-	rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring,
-				  PCI_DMA_BIDIRECTIONAL);
-err_dma_rx_ring_bufs_alloc:
-	rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring);
-err_dma_rx_ring_create:
-	rocker_dma_ring_bufs_free(rocker, &rocker_port->tx_ring,
-				  PCI_DMA_TODEVICE);
-err_dma_tx_ring_bufs_alloc:
-	rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring);
-	return err;
-}
-
-static void rocker_port_dma_rings_fini(struct rocker_port *rocker_port)
-{
-	struct rocker *rocker = rocker_port->rocker;
-
-	rocker_dma_rx_ring_skbs_free(rocker_port);
-	rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring,
-				  PCI_DMA_BIDIRECTIONAL);
-	rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring);
-	rocker_dma_ring_bufs_free(rocker, &rocker_port->tx_ring,
-				  PCI_DMA_TODEVICE);
-	rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring);
-}
-
-static void rocker_port_set_enable(const struct rocker_port *rocker_port,
-				   bool enable)
-{
-	u64 val = rocker_read64(rocker_port->rocker, PORT_PHYS_ENABLE);
-
-	if (enable)
-		val |= 1ULL << rocker_port->pport;
-	else
-		val &= ~(1ULL << rocker_port->pport);
-	rocker_write64(rocker_port->rocker, PORT_PHYS_ENABLE, val);
-}
-
-/********************************
- * Interrupt handler and helpers
- ********************************/
-
-static irqreturn_t rocker_cmd_irq_handler(int irq, void *dev_id)
-{
-	struct rocker *rocker = dev_id;
-	const struct rocker_desc_info *desc_info;
-	struct rocker_wait *wait;
-	u32 credits = 0;
-
-	spin_lock(&rocker->cmd_ring_lock);
-	while ((desc_info = rocker_desc_tail_get(&rocker->cmd_ring))) {
-		wait = rocker_desc_cookie_ptr_get(desc_info);
-		if (wait->nowait) {
-			rocker_desc_gen_clear(desc_info);
-			rocker_wait_destroy(NULL, wait);
-		} else {
-			rocker_wait_wake_up(wait);
-		}
-		credits++;
-	}
-	spin_unlock(&rocker->cmd_ring_lock);
-	rocker_dma_ring_credits_set(rocker, &rocker->cmd_ring, credits);
-
-	return IRQ_HANDLED;
-}
-
-static void rocker_port_link_up(const struct rocker_port *rocker_port)
-{
-	netif_carrier_on(rocker_port->dev);
-	netdev_info(rocker_port->dev, "Link is up\n");
-}
-
-static void rocker_port_link_down(const struct rocker_port *rocker_port)
-{
-	netif_carrier_off(rocker_port->dev);
-	netdev_info(rocker_port->dev, "Link is down\n");
-}
-
-static int rocker_event_link_change(const struct rocker *rocker,
-				    const struct rocker_tlv *info)
-{
-	const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_LINK_CHANGED_MAX + 1];
-	unsigned int port_number;
-	bool link_up;
-	struct rocker_port *rocker_port;
-
-	rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_LINK_CHANGED_MAX, info);
-	if (!attrs[ROCKER_TLV_EVENT_LINK_CHANGED_PPORT] ||
-	    !attrs[ROCKER_TLV_EVENT_LINK_CHANGED_LINKUP])
-		return -EIO;
-	port_number =
-		rocker_tlv_get_u32(attrs[ROCKER_TLV_EVENT_LINK_CHANGED_PPORT]) - 1;
-	link_up = rocker_tlv_get_u8(attrs[ROCKER_TLV_EVENT_LINK_CHANGED_LINKUP]);
-
-	if (port_number >= rocker->port_count)
-		return -EINVAL;
-
-	rocker_port = rocker->ports[port_number];
-	if (netif_carrier_ok(rocker_port->dev) != link_up) {
-		if (link_up)
-			rocker_port_link_up(rocker_port);
-		else
-			rocker_port_link_down(rocker_port);
-	}
-
-	return 0;
-}
-
-static int rocker_port_fdb(struct rocker_port *rocker_port,
-			   struct switchdev_trans *trans,
-			   const unsigned char *addr,
-			   __be16 vlan_id, int flags);
-
-static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
-				      const struct rocker_tlv *info)
-{
-	const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAX + 1];
-	unsigned int port_number;
-	struct rocker_port *rocker_port;
-	const unsigned char *addr;
-	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_LEARNED;
-	__be16 vlan_id;
-
-	rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_MAC_VLAN_MAX, info);
-	if (!attrs[ROCKER_TLV_EVENT_MAC_VLAN_PPORT] ||
-	    !attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAC] ||
-	    !attrs[ROCKER_TLV_EVENT_MAC_VLAN_VLAN_ID])
-		return -EIO;
-	port_number =
-		rocker_tlv_get_u32(attrs[ROCKER_TLV_EVENT_MAC_VLAN_PPORT]) - 1;
-	addr = rocker_tlv_data(attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAC]);
-	vlan_id = rocker_tlv_get_be16(attrs[ROCKER_TLV_EVENT_MAC_VLAN_VLAN_ID]);
-
-	if (port_number >= rocker->port_count)
-		return -EINVAL;
-
-	rocker_port = rocker->ports[port_number];
-
-	if (rocker_port->stp_state != BR_STATE_LEARNING &&
-	    rocker_port->stp_state != BR_STATE_FORWARDING)
-		return 0;
-
-	return rocker_port_fdb(rocker_port, NULL, addr, vlan_id, flags);
-}
-
-static int rocker_event_process(const struct rocker *rocker,
-				const struct rocker_desc_info *desc_info)
-{
-	const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAX + 1];
-	const struct rocker_tlv *info;
-	u16 type;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_EVENT_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_EVENT_TYPE] ||
-	    !attrs[ROCKER_TLV_EVENT_INFO])
-		return -EIO;
-
-	type = rocker_tlv_get_u16(attrs[ROCKER_TLV_EVENT_TYPE]);
-	info = attrs[ROCKER_TLV_EVENT_INFO];
-
-	switch (type) {
-	case ROCKER_TLV_EVENT_TYPE_LINK_CHANGED:
-		return rocker_event_link_change(rocker, info);
-	case ROCKER_TLV_EVENT_TYPE_MAC_VLAN_SEEN:
-		return rocker_event_mac_vlan_seen(rocker, info);
-	}
-
-	return -EOPNOTSUPP;
-}
-
-static irqreturn_t rocker_event_irq_handler(int irq, void *dev_id)
-{
-	struct rocker *rocker = dev_id;
-	const struct pci_dev *pdev = rocker->pdev;
-	const struct rocker_desc_info *desc_info;
-	u32 credits = 0;
-	int err;
-
-	while ((desc_info = rocker_desc_tail_get(&rocker->event_ring))) {
-		err = rocker_desc_err(desc_info);
-		if (err) {
-			dev_err(&pdev->dev, "event desc received with err %d\n",
-				err);
-		} else {
-			err = rocker_event_process(rocker, desc_info);
-			if (err)
-				dev_err(&pdev->dev, "event processing failed with err %d\n",
-					err);
-		}
-		rocker_desc_gen_clear(desc_info);
-		rocker_desc_head_set(rocker, &rocker->event_ring, desc_info);
-		credits++;
-	}
-	rocker_dma_ring_credits_set(rocker, &rocker->event_ring, credits);
-
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t rocker_tx_irq_handler(int irq, void *dev_id)
-{
-	struct rocker_port *rocker_port = dev_id;
-
-	napi_schedule(&rocker_port->napi_tx);
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t rocker_rx_irq_handler(int irq, void *dev_id)
-{
-	struct rocker_port *rocker_port = dev_id;
-
-	napi_schedule(&rocker_port->napi_rx);
-	return IRQ_HANDLED;
-}
-
-/********************
- * Command interface
- ********************/
-
-typedef int (*rocker_cmd_prep_cb_t)(const struct rocker_port *rocker_port,
-				    struct rocker_desc_info *desc_info,
-				    void *priv);
-
-typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port,
-				    const struct rocker_desc_info *desc_info,
-				    void *priv);
-
-static int rocker_cmd_exec(struct rocker_port *rocker_port,
-			   struct switchdev_trans *trans, int flags,
-			   rocker_cmd_prep_cb_t prepare, void *prepare_priv,
-			   rocker_cmd_proc_cb_t process, void *process_priv)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_desc_info *desc_info;
-	struct rocker_wait *wait;
-	bool nowait = !!(flags & ROCKER_OP_FLAG_NOWAIT);
-	unsigned long lock_flags;
-	int err;
-
-	wait = rocker_wait_create(rocker_port, trans, flags);
-	if (!wait)
-		return -ENOMEM;
-	wait->nowait = nowait;
-
-	spin_lock_irqsave(&rocker->cmd_ring_lock, lock_flags);
-
-	desc_info = rocker_desc_head_get(&rocker->cmd_ring);
-	if (!desc_info) {
-		spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
-		err = -EAGAIN;
-		goto out;
-	}
-
-	err = prepare(rocker_port, desc_info, prepare_priv);
-	if (err) {
-		spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
-		goto out;
-	}
-
-	rocker_desc_cookie_ptr_set(desc_info, wait);
-
-	if (!switchdev_trans_ph_prepare(trans))
-		rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info);
-
-	spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
-
-	if (nowait)
-		return 0;
-
-	if (!switchdev_trans_ph_prepare(trans))
-		if (!rocker_wait_event_timeout(wait, HZ / 10))
-			return -EIO;
-
-	err = rocker_desc_err(desc_info);
-	if (err)
-		return err;
-
-	if (process)
-		err = process(rocker_port, desc_info, process_priv);
-
-	rocker_desc_gen_clear(desc_info);
-out:
-	rocker_wait_destroy(trans, wait);
-	return err;
-}
-
-static int
-rocker_cmd_get_port_settings_prep(const struct rocker_port *rocker_port,
-				  struct rocker_desc_info *desc_info,
-				  void *priv)
-{
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
-			       ROCKER_TLV_CMD_TYPE_GET_PORT_SETTINGS))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
-			       rocker_port->pport))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-	return 0;
-}
-
-static int
-rocker_cmd_get_port_settings_ethtool_proc(const struct rocker_port *rocker_port,
-					  const struct rocker_desc_info *desc_info,
-					  void *priv)
-{
-	struct ethtool_cmd *ecmd = priv;
-	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
-	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
-	u32 speed;
-	u8 duplex;
-	u8 autoneg;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_CMD_INFO])
-		return -EIO;
-
-	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
-				attrs[ROCKER_TLV_CMD_INFO]);
-	if (!info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_SPEED] ||
-	    !info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX] ||
-	    !info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG])
-		return -EIO;
-
-	speed = rocker_tlv_get_u32(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_SPEED]);
-	duplex = rocker_tlv_get_u8(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX]);
-	autoneg = rocker_tlv_get_u8(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG]);
-
-	ecmd->transceiver = XCVR_INTERNAL;
-	ecmd->supported = SUPPORTED_TP;
-	ecmd->phy_address = 0xff;
-	ecmd->port = PORT_TP;
-	ethtool_cmd_speed_set(ecmd, speed);
-	ecmd->duplex = duplex ? DUPLEX_FULL : DUPLEX_HALF;
-	ecmd->autoneg = autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_get_port_settings_macaddr_proc(const struct rocker_port *rocker_port,
-					  const struct rocker_desc_info *desc_info,
-					  void *priv)
-{
-	unsigned char *macaddr = priv;
-	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
-	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
-	const struct rocker_tlv *attr;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_CMD_INFO])
-		return -EIO;
-
-	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
-				attrs[ROCKER_TLV_CMD_INFO]);
-	attr = info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR];
-	if (!attr)
-		return -EIO;
-
-	if (rocker_tlv_len(attr) != ETH_ALEN)
-		return -EINVAL;
-
-	ether_addr_copy(macaddr, rocker_tlv_data(attr));
-	return 0;
-}
-
-struct port_name {
-	char *buf;
-	size_t len;
-};
-
-static int
-rocker_cmd_get_port_settings_phys_name_proc(const struct rocker_port *rocker_port,
-					    const struct rocker_desc_info *desc_info,
-					    void *priv)
-{
-	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
-	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
-	struct port_name *name = priv;
-	const struct rocker_tlv *attr;
-	size_t i, j, len;
-	const char *str;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_CMD_INFO])
-		return -EIO;
-
-	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
-				attrs[ROCKER_TLV_CMD_INFO]);
-	attr = info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_PHYS_NAME];
-	if (!attr)
-		return -EIO;
-
-	len = min_t(size_t, rocker_tlv_len(attr), name->len);
-	str = rocker_tlv_data(attr);
-
-	/* make sure name only contains alphanumeric characters */
-	for (i = j = 0; i < len; ++i) {
-		if (isalnum(str[i])) {
-			name->buf[j] = str[i];
-			j++;
-		}
-	}
-
-	if (j == 0)
-		return -EIO;
-
-	name->buf[j] = '\0';
-
-	return 0;
-}
-
-static int
-rocker_cmd_set_port_settings_ethtool_prep(const struct rocker_port *rocker_port,
-					  struct rocker_desc_info *desc_info,
-					  void *priv)
-{
-	struct ethtool_cmd *ecmd = priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
-			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
-			       rocker_port->pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_SPEED,
-			       ethtool_cmd_speed(ecmd)))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX,
-			      ecmd->duplex))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG,
-			      ecmd->autoneg))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-	return 0;
-}
-
-static int
-rocker_cmd_set_port_settings_macaddr_prep(const struct rocker_port *rocker_port,
-					  struct rocker_desc_info *desc_info,
-					  void *priv)
-{
-	const unsigned char *macaddr = priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
-			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
-			       rocker_port->pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR,
-			   ETH_ALEN, macaddr))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-	return 0;
-}
-
-static int
-rocker_cmd_set_port_settings_mtu_prep(const struct rocker_port *rocker_port,
-				      struct rocker_desc_info *desc_info,
-				      void *priv)
-{
-	int mtu = *(int *)priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
-			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
-			       rocker_port->pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_MTU,
-			       mtu))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-	return 0;
-}
-
-static int
-rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port,
-				  struct rocker_desc_info *desc_info,
-				  void *priv)
-{
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
-			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
-			       rocker_port->pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,
-			      !!(rocker_port->brport_flags & BR_LEARNING)))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-	return 0;
-}
-
-static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port,
-						struct ethtool_cmd *ecmd)
-{
-	return rocker_cmd_exec(rocker_port, NULL, 0,
-			       rocker_cmd_get_port_settings_prep, NULL,
-			       rocker_cmd_get_port_settings_ethtool_proc,
-			       ecmd);
-}
-
-static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
-						unsigned char *macaddr)
-{
-	return rocker_cmd_exec(rocker_port, NULL, 0,
-			       rocker_cmd_get_port_settings_prep, NULL,
-			       rocker_cmd_get_port_settings_macaddr_proc,
-			       macaddr);
-}
-
-static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
-						struct ethtool_cmd *ecmd)
-{
-	return rocker_cmd_exec(rocker_port, NULL, 0,
-			       rocker_cmd_set_port_settings_ethtool_prep,
-			       ecmd, NULL, NULL);
-}
-
-static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port,
-						unsigned char *macaddr)
-{
-	return rocker_cmd_exec(rocker_port, NULL, 0,
-			       rocker_cmd_set_port_settings_macaddr_prep,
-			       macaddr, NULL, NULL);
-}
-
-static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
-					    int mtu)
-{
-	return rocker_cmd_exec(rocker_port, NULL, 0,
-			       rocker_cmd_set_port_settings_mtu_prep,
-			       &mtu, NULL, NULL);
-}
-
-static int rocker_port_set_learning(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans)
-{
-	return rocker_cmd_exec(rocker_port, trans, 0,
-			       rocker_cmd_set_port_learning_prep,
-			       NULL, NULL, NULL);
-}
-
-static int
-rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info,
-				const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.ig_port.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
-			       entry->key.ig_port.in_pport_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.ig_port.goto_tbl))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info,
-			     const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.vlan.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.vlan.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
-				entry->key.vlan.vlan_id_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.vlan.goto_tbl))
-		return -EMSGSIZE;
-	if (entry->key.vlan.untagged &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_NEW_VLAN_ID,
-				entry->key.vlan.new_vlan_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info,
-				 const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.term_mac.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
-			       entry->key.term_mac.in_pport_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
-				entry->key.term_mac.eth_type))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->key.term_mac.eth_dst))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
-			   ETH_ALEN, entry->key.term_mac.eth_dst_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.term_mac.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
-				entry->key.term_mac.vlan_id_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.term_mac.goto_tbl))
-		return -EMSGSIZE;
-	if (entry->key.term_mac.copy_to_cpu &&
-	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
-			      entry->key.term_mac.copy_to_cpu))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_flow_tbl_add_ucast_routing(struct rocker_desc_info *desc_info,
-				      const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
-				entry->key.ucast_routing.eth_type))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP,
-				entry->key.ucast_routing.dst4))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP_MASK,
-				entry->key.ucast_routing.dst4_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.ucast_routing.goto_tbl))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->key.ucast_routing.group_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info,
-			       const struct rocker_flow_tbl_entry *entry)
-{
-	if (entry->key.bridge.has_eth_dst &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->key.bridge.eth_dst))
-		return -EMSGSIZE;
-	if (entry->key.bridge.has_eth_dst_mask &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
-			   ETH_ALEN, entry->key.bridge.eth_dst_mask))
-		return -EMSGSIZE;
-	if (entry->key.bridge.vlan_id &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.bridge.vlan_id))
-		return -EMSGSIZE;
-	if (entry->key.bridge.tunnel_id &&
-	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_TUNNEL_ID,
-			       entry->key.bridge.tunnel_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.bridge.goto_tbl))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->key.bridge.group_id))
-		return -EMSGSIZE;
-	if (entry->key.bridge.copy_to_cpu &&
-	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
-			      entry->key.bridge.copy_to_cpu))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info,
-			    const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.acl.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
-			       entry->key.acl.in_pport_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
-			   ETH_ALEN, entry->key.acl.eth_src))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC_MASK,
-			   ETH_ALEN, entry->key.acl.eth_src_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->key.acl.eth_dst))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
-			   ETH_ALEN, entry->key.acl.eth_dst_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
-				entry->key.acl.eth_type))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.acl.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
-				entry->key.acl.vlan_id_mask))
-		return -EMSGSIZE;
-
-	switch (ntohs(entry->key.acl.eth_type)) {
-	case ETH_P_IP:
-	case ETH_P_IPV6:
-		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_PROTO,
-				      entry->key.acl.ip_proto))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info,
-				      ROCKER_TLV_OF_DPA_IP_PROTO_MASK,
-				      entry->key.acl.ip_proto_mask))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_DSCP,
-				      entry->key.acl.ip_tos & 0x3f))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info,
-				      ROCKER_TLV_OF_DPA_IP_DSCP_MASK,
-				      entry->key.acl.ip_tos_mask & 0x3f))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_ECN,
-				      (entry->key.acl.ip_tos & 0xc0) >> 6))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info,
-				      ROCKER_TLV_OF_DPA_IP_ECN_MASK,
-				      (entry->key.acl.ip_tos_mask & 0xc0) >> 6))
-			return -EMSGSIZE;
-		break;
-	}
-
-	if (entry->key.acl.group_id != ROCKER_GROUP_NONE &&
-	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->key.acl.group_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int rocker_cmd_flow_tbl_add(const struct rocker_port *rocker_port,
-				   struct rocker_desc_info *desc_info,
-				   void *priv)
-{
-	const struct rocker_flow_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-	int err = 0;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_TABLE_ID,
-			       entry->key.tbl_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_PRIORITY,
-			       entry->key.priority))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_HARDTIME, 0))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
-			       entry->cookie))
-		return -EMSGSIZE;
-
-	switch (entry->key.tbl_id) {
-	case ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT:
-		err = rocker_cmd_flow_tbl_add_ig_port(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_VLAN:
-		err = rocker_cmd_flow_tbl_add_vlan(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC:
-		err = rocker_cmd_flow_tbl_add_term_mac(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING:
-		err = rocker_cmd_flow_tbl_add_ucast_routing(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_BRIDGING:
-		err = rocker_cmd_flow_tbl_add_bridge(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_ACL_POLICY:
-		err = rocker_cmd_flow_tbl_add_acl(desc_info, entry);
-		break;
-	default:
-		err = -ENOTSUPP;
-		break;
-	}
-
-	if (err)
-		return err;
-
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-static int rocker_cmd_flow_tbl_del(const struct rocker_port *rocker_port,
-				   struct rocker_desc_info *desc_info,
-				   void *priv)
-{
-	const struct rocker_flow_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
-			       entry->cookie))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_l2_interface(struct rocker_desc_info *desc_info,
-				      struct rocker_group_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_OUT_PPORT,
-			       ROCKER_GROUP_PORT_GET(entry->group_id)))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_POP_VLAN,
-			      entry->l2_interface.pop_vlan))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_l2_rewrite(struct rocker_desc_info *desc_info,
-				    const struct rocker_group_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
-			       entry->l2_rewrite.group_id))
-		return -EMSGSIZE;
-	if (!is_zero_ether_addr(entry->l2_rewrite.eth_src) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
-			   ETH_ALEN, entry->l2_rewrite.eth_src))
-		return -EMSGSIZE;
-	if (!is_zero_ether_addr(entry->l2_rewrite.eth_dst) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->l2_rewrite.eth_dst))
-		return -EMSGSIZE;
-	if (entry->l2_rewrite.vlan_id &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->l2_rewrite.vlan_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_group_ids(struct rocker_desc_info *desc_info,
-				   const struct rocker_group_tbl_entry *entry)
-{
-	int i;
-	struct rocker_tlv *group_ids;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GROUP_COUNT,
-			       entry->group_count))
-		return -EMSGSIZE;
-
-	group_ids = rocker_tlv_nest_start(desc_info,
-					  ROCKER_TLV_OF_DPA_GROUP_IDS);
-	if (!group_ids)
-		return -EMSGSIZE;
-
-	for (i = 0; i < entry->group_count; i++)
-		/* Note TLV array is 1-based */
-		if (rocker_tlv_put_u32(desc_info, i + 1, entry->group_ids[i]))
-			return -EMSGSIZE;
-
-	rocker_tlv_nest_end(desc_info, group_ids);
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_l3_unicast(struct rocker_desc_info *desc_info,
-				    const struct rocker_group_tbl_entry *entry)
-{
-	if (!is_zero_ether_addr(entry->l3_unicast.eth_src) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
-			   ETH_ALEN, entry->l3_unicast.eth_src))
-		return -EMSGSIZE;
-	if (!is_zero_ether_addr(entry->l3_unicast.eth_dst) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->l3_unicast.eth_dst))
-		return -EMSGSIZE;
-	if (entry->l3_unicast.vlan_id &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->l3_unicast.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_TTL_CHECK,
-			      entry->l3_unicast.ttl_check))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
-			       entry->l3_unicast.group_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int rocker_cmd_group_tbl_add(const struct rocker_port *rocker_port,
-				    struct rocker_desc_info *desc_info,
-				    void *priv)
-{
-	struct rocker_group_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-	int err = 0;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->group_id))
-		return -EMSGSIZE;
-
-	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE:
-		err = rocker_cmd_group_tbl_add_l2_interface(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE:
-		err = rocker_cmd_group_tbl_add_l2_rewrite(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
-		err = rocker_cmd_group_tbl_add_group_ids(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST:
-		err = rocker_cmd_group_tbl_add_l3_unicast(desc_info, entry);
-		break;
-	default:
-		err = -ENOTSUPP;
-		break;
-	}
-
-	if (err)
-		return err;
-
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-static int rocker_cmd_group_tbl_del(const struct rocker_port *rocker_port,
-				    struct rocker_desc_info *desc_info,
-				    void *priv)
-{
-	const struct rocker_group_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->group_id))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-/***************************************************
- * Flow, group, FDB, internal VLAN and neigh tables
- ***************************************************/
-
-static int rocker_init_tbls(struct rocker *rocker)
-{
-	hash_init(rocker->flow_tbl);
-	spin_lock_init(&rocker->flow_tbl_lock);
-
-	hash_init(rocker->group_tbl);
-	spin_lock_init(&rocker->group_tbl_lock);
-
-	hash_init(rocker->fdb_tbl);
-	spin_lock_init(&rocker->fdb_tbl_lock);
-
-	hash_init(rocker->internal_vlan_tbl);
-	spin_lock_init(&rocker->internal_vlan_tbl_lock);
-
-	hash_init(rocker->neigh_tbl);
-	spin_lock_init(&rocker->neigh_tbl_lock);
-
-	return 0;
-}
-
-static void rocker_free_tbls(struct rocker *rocker)
-{
-	unsigned long flags;
-	struct rocker_flow_tbl_entry *flow_entry;
-	struct rocker_group_tbl_entry *group_entry;
-	struct rocker_fdb_tbl_entry *fdb_entry;
-	struct rocker_internal_vlan_tbl_entry *internal_vlan_entry;
-	struct rocker_neigh_tbl_entry *neigh_entry;
-	struct hlist_node *tmp;
-	int bkt;
-
-	spin_lock_irqsave(&rocker->flow_tbl_lock, flags);
-	hash_for_each_safe(rocker->flow_tbl, bkt, tmp, flow_entry, entry)
-		hash_del(&flow_entry->entry);
-	spin_unlock_irqrestore(&rocker->flow_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->group_tbl_lock, flags);
-	hash_for_each_safe(rocker->group_tbl, bkt, tmp, group_entry, entry)
-		hash_del(&group_entry->entry);
-	spin_unlock_irqrestore(&rocker->group_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, flags);
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, fdb_entry, entry)
-		hash_del(&fdb_entry->entry);
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, flags);
-	hash_for_each_safe(rocker->internal_vlan_tbl, bkt,
-			   tmp, internal_vlan_entry, entry)
-		hash_del(&internal_vlan_entry->entry);
-	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->neigh_tbl_lock, flags);
-	hash_for_each_safe(rocker->neigh_tbl, bkt, tmp, neigh_entry, entry)
-		hash_del(&neigh_entry->entry);
-	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, flags);
-}
-
-static struct rocker_flow_tbl_entry *
-rocker_flow_tbl_find(const struct rocker *rocker,
-		     const struct rocker_flow_tbl_entry *match)
-{
-	struct rocker_flow_tbl_entry *found;
-	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
-
-	hash_for_each_possible(rocker->flow_tbl, found,
-			       entry, match->key_crc32) {
-		if (memcmp(&found->key, &match->key, key_len) == 0)
-			return found;
-	}
-
-	return NULL;
-}
-
-static int rocker_flow_tbl_add(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       struct rocker_flow_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_flow_tbl_entry *found;
-	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
-	unsigned long lock_flags;
-
-	match->key_crc32 = crc32(~0, &match->key, key_len);
-
-	spin_lock_irqsave(&rocker->flow_tbl_lock, lock_flags);
-
-	found = rocker_flow_tbl_find(rocker, match);
-
-	if (found) {
-		match->cookie = found->cookie;
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		rocker_kfree(trans, found);
-		found = match;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD;
-	} else {
-		found = match;
-		found->cookie = rocker->flow_tbl_next_cookie++;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD;
-	}
-
-	if (!switchdev_trans_ph_prepare(trans))
-		hash_add(rocker->flow_tbl, &found->entry, found->key_crc32);
-
-	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
-
-	return rocker_cmd_exec(rocker_port, trans, flags,
-			       rocker_cmd_flow_tbl_add, found, NULL, NULL);
-}
-
-static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       struct rocker_flow_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_flow_tbl_entry *found;
-	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
-	unsigned long lock_flags;
-	int err = 0;
-
-	match->key_crc32 = crc32(~0, &match->key, key_len);
-
-	spin_lock_irqsave(&rocker->flow_tbl_lock, lock_flags);
-
-	found = rocker_flow_tbl_find(rocker, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL;
-	}
-
-	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
-
-	rocker_kfree(trans, match);
-
-	if (found) {
-		err = rocker_cmd_exec(rocker_port, trans, flags,
-				      rocker_cmd_flow_tbl_del,
-				      found, NULL, NULL);
-		rocker_kfree(trans, found);
-	}
-
-	return err;
-}
-
-static int rocker_flow_tbl_do(struct rocker_port *rocker_port,
-			      struct switchdev_trans *trans, int flags,
-			      struct rocker_flow_tbl_entry *entry)
-{
-	if (flags & ROCKER_OP_FLAG_REMOVE)
-		return rocker_flow_tbl_del(rocker_port, trans, flags, entry);
-	else
-		return rocker_flow_tbl_add(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_ig_port(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans, int flags,
-				   u32 in_pport, u32 in_pport_mask,
-				   enum rocker_of_dpa_table_id goto_tbl)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.priority = ROCKER_PRIORITY_IG_PORT;
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT;
-	entry->key.ig_port.in_pport = in_pport;
-	entry->key.ig_port.in_pport_mask = in_pport_mask;
-	entry->key.ig_port.goto_tbl = goto_tbl;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_vlan(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, int flags,
-				u32 in_pport, __be16 vlan_id,
-				__be16 vlan_id_mask,
-				enum rocker_of_dpa_table_id goto_tbl,
-				bool untagged, __be16 new_vlan_id)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.priority = ROCKER_PRIORITY_VLAN;
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_VLAN;
-	entry->key.vlan.in_pport = in_pport;
-	entry->key.vlan.vlan_id = vlan_id;
-	entry->key.vlan.vlan_id_mask = vlan_id_mask;
-	entry->key.vlan.goto_tbl = goto_tbl;
-
-	entry->key.vlan.untagged = untagged;
-	entry->key.vlan.new_vlan_id = new_vlan_id;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans,
-				    u32 in_pport, u32 in_pport_mask,
-				    __be16 eth_type, const u8 *eth_dst,
-				    const u8 *eth_dst_mask, __be16 vlan_id,
-				    __be16 vlan_id_mask, bool copy_to_cpu,
-				    int flags)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	if (is_multicast_ether_addr(eth_dst)) {
-		entry->key.priority = ROCKER_PRIORITY_TERM_MAC_MCAST;
-		entry->key.term_mac.goto_tbl =
-			 ROCKER_OF_DPA_TABLE_ID_MULTICAST_ROUTING;
-	} else {
-		entry->key.priority = ROCKER_PRIORITY_TERM_MAC_UCAST;
-		entry->key.term_mac.goto_tbl =
-			 ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
-	}
-
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
-	entry->key.term_mac.in_pport = in_pport;
-	entry->key.term_mac.in_pport_mask = in_pport_mask;
-	entry->key.term_mac.eth_type = eth_type;
-	ether_addr_copy(entry->key.term_mac.eth_dst, eth_dst);
-	ether_addr_copy(entry->key.term_mac.eth_dst_mask, eth_dst_mask);
-	entry->key.term_mac.vlan_id = vlan_id;
-	entry->key.term_mac.vlan_id_mask = vlan_id_mask;
-	entry->key.term_mac.copy_to_cpu = copy_to_cpu;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags,
-				  const u8 *eth_dst, const u8 *eth_dst_mask,
-				  __be16 vlan_id, u32 tunnel_id,
-				  enum rocker_of_dpa_table_id goto_tbl,
-				  u32 group_id, bool copy_to_cpu)
-{
-	struct rocker_flow_tbl_entry *entry;
-	u32 priority;
-	bool vlan_bridging = !!vlan_id;
-	bool dflt = !eth_dst || (eth_dst && eth_dst_mask);
-	bool wild = false;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_BRIDGING;
-
-	if (eth_dst) {
-		entry->key.bridge.has_eth_dst = 1;
-		ether_addr_copy(entry->key.bridge.eth_dst, eth_dst);
-	}
-	if (eth_dst_mask) {
-		entry->key.bridge.has_eth_dst_mask = 1;
-		ether_addr_copy(entry->key.bridge.eth_dst_mask, eth_dst_mask);
-		if (!ether_addr_equal(eth_dst_mask, ff_mac))
-			wild = true;
-	}
-
-	priority = ROCKER_PRIORITY_UNKNOWN;
-	if (vlan_bridging && dflt && wild)
-		priority = ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_WILD;
-	else if (vlan_bridging && dflt && !wild)
-		priority = ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_EXACT;
-	else if (vlan_bridging && !dflt)
-		priority = ROCKER_PRIORITY_BRIDGING_VLAN;
-	else if (!vlan_bridging && dflt && wild)
-		priority = ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_WILD;
-	else if (!vlan_bridging && dflt && !wild)
-		priority = ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_EXACT;
-	else if (!vlan_bridging && !dflt)
-		priority = ROCKER_PRIORITY_BRIDGING_TENANT;
-
-	entry->key.priority = priority;
-	entry->key.bridge.vlan_id = vlan_id;
-	entry->key.bridge.tunnel_id = tunnel_id;
-	entry->key.bridge.goto_tbl = goto_tbl;
-	entry->key.bridge.group_id = group_id;
-	entry->key.bridge.copy_to_cpu = copy_to_cpu;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port,
-					  struct switchdev_trans *trans,
-					  __be16 eth_type, __be32 dst,
-					  __be32 dst_mask, u32 priority,
-					  enum rocker_of_dpa_table_id goto_tbl,
-					  u32 group_id, int flags)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
-	entry->key.priority = priority;
-	entry->key.ucast_routing.eth_type = eth_type;
-	entry->key.ucast_routing.dst4 = dst;
-	entry->key.ucast_routing.dst4_mask = dst_mask;
-	entry->key.ucast_routing.goto_tbl = goto_tbl;
-	entry->key.ucast_routing.group_id = group_id;
-	entry->key_len = offsetof(struct rocker_flow_tbl_key,
-				  ucast_routing.group_id);
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_acl(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       u32 in_pport, u32 in_pport_mask,
-			       const u8 *eth_src, const u8 *eth_src_mask,
-			       const u8 *eth_dst, const u8 *eth_dst_mask,
-			       __be16 eth_type, __be16 vlan_id,
-			       __be16 vlan_id_mask, u8 ip_proto,
-			       u8 ip_proto_mask, u8 ip_tos, u8 ip_tos_mask,
-			       u32 group_id)
-{
-	u32 priority;
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	priority = ROCKER_PRIORITY_ACL_NORMAL;
-	if (eth_dst && eth_dst_mask) {
-		if (ether_addr_equal(eth_dst_mask, mcast_mac))
-			priority = ROCKER_PRIORITY_ACL_DFLT;
-		else if (is_link_local_ether_addr(eth_dst))
-			priority = ROCKER_PRIORITY_ACL_CTRL;
-	}
-
-	entry->key.priority = priority;
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	entry->key.acl.in_pport = in_pport;
-	entry->key.acl.in_pport_mask = in_pport_mask;
-
-	if (eth_src)
-		ether_addr_copy(entry->key.acl.eth_src, eth_src);
-	if (eth_src_mask)
-		ether_addr_copy(entry->key.acl.eth_src_mask, eth_src_mask);
-	if (eth_dst)
-		ether_addr_copy(entry->key.acl.eth_dst, eth_dst);
-	if (eth_dst_mask)
-		ether_addr_copy(entry->key.acl.eth_dst_mask, eth_dst_mask);
-
-	entry->key.acl.eth_type = eth_type;
-	entry->key.acl.vlan_id = vlan_id;
-	entry->key.acl.vlan_id_mask = vlan_id_mask;
-	entry->key.acl.ip_proto = ip_proto;
-	entry->key.acl.ip_proto_mask = ip_proto_mask;
-	entry->key.acl.ip_tos = ip_tos;
-	entry->key.acl.ip_tos_mask = ip_tos_mask;
-	entry->key.acl.group_id = group_id;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static struct rocker_group_tbl_entry *
-rocker_group_tbl_find(const struct rocker *rocker,
-		      const struct rocker_group_tbl_entry *match)
-{
-	struct rocker_group_tbl_entry *found;
-
-	hash_for_each_possible(rocker->group_tbl, found,
-			       entry, match->group_id) {
-		if (found->group_id == match->group_id)
-			return found;
-	}
-
-	return NULL;
-}
-
-static void rocker_group_tbl_entry_free(struct switchdev_trans *trans,
-					struct rocker_group_tbl_entry *entry)
-{
-	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
-		rocker_kfree(trans, entry->group_ids);
-		break;
-	default:
-		break;
-	}
-	rocker_kfree(trans, entry);
-}
-
-static int rocker_group_tbl_add(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, int flags,
-				struct rocker_group_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_group_tbl_entry *found;
-	unsigned long lock_flags;
-
-	spin_lock_irqsave(&rocker->group_tbl_lock, lock_flags);
-
-	found = rocker_group_tbl_find(rocker, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		rocker_group_tbl_entry_free(trans, found);
-		found = match;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD;
-	} else {
-		found = match;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD;
-	}
-
-	if (!switchdev_trans_ph_prepare(trans))
-		hash_add(rocker->group_tbl, &found->entry, found->group_id);
-
-	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
-
-	return rocker_cmd_exec(rocker_port, trans, flags,
-			       rocker_cmd_group_tbl_add, found, NULL, NULL);
-}
-
-static int rocker_group_tbl_del(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, int flags,
-				struct rocker_group_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_group_tbl_entry *found;
-	unsigned long lock_flags;
-	int err = 0;
-
-	spin_lock_irqsave(&rocker->group_tbl_lock, lock_flags);
-
-	found = rocker_group_tbl_find(rocker, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL;
-	}
-
-	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
-
-	rocker_group_tbl_entry_free(trans, match);
-
-	if (found) {
-		err = rocker_cmd_exec(rocker_port, trans, flags,
-				      rocker_cmd_group_tbl_del,
-				      found, NULL, NULL);
-		rocker_group_tbl_entry_free(trans, found);
-	}
-
-	return err;
-}
-
-static int rocker_group_tbl_do(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       struct rocker_group_tbl_entry *entry)
-{
-	if (flags & ROCKER_OP_FLAG_REMOVE)
-		return rocker_group_tbl_del(rocker_port, trans, flags, entry);
-	else
-		return rocker_group_tbl_add(rocker_port, trans, flags, entry);
-}
-
-static int rocker_group_l2_interface(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     __be16 vlan_id, u32 out_pport,
-				     int pop_vlan)
-{
-	struct rocker_group_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
-	entry->l2_interface.pop_vlan = pop_vlan;
-
-	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_group_l2_fan_out(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans,
-				   int flags, u8 group_count,
-				   const u32 *group_ids, u32 group_id)
-{
-	struct rocker_group_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->group_id = group_id;
-	entry->group_count = group_count;
-
-	entry->group_ids = rocker_kcalloc(trans, flags,
-					  group_count, sizeof(u32));
-	if (!entry->group_ids) {
-		rocker_kfree(trans, entry);
-		return -ENOMEM;
-	}
-	memcpy(entry->group_ids, group_ids, group_count * sizeof(u32));
-
-	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_group_l2_flood(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 __be16 vlan_id, u8 group_count,
-				 const u32 *group_ids, u32 group_id)
-{
-	return rocker_group_l2_fan_out(rocker_port, trans, flags,
-				       group_count, group_ids,
-				       group_id);
-}
-
-static int rocker_group_l3_unicast(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans, int flags,
-				   u32 index, const u8 *src_mac, const u8 *dst_mac,
-				   __be16 vlan_id, bool ttl_check, u32 pport)
-{
-	struct rocker_group_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->group_id = ROCKER_GROUP_L3_UNICAST(index);
-	if (src_mac)
-		ether_addr_copy(entry->l3_unicast.eth_src, src_mac);
-	if (dst_mac)
-		ether_addr_copy(entry->l3_unicast.eth_dst, dst_mac);
-	entry->l3_unicast.vlan_id = vlan_id;
-	entry->l3_unicast.ttl_check = ttl_check;
-	entry->l3_unicast.group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, pport);
-
-	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static struct rocker_neigh_tbl_entry *
-rocker_neigh_tbl_find(const struct rocker *rocker, __be32 ip_addr)
-{
-	struct rocker_neigh_tbl_entry *found;
-
-	hash_for_each_possible(rocker->neigh_tbl, found,
-			       entry, be32_to_cpu(ip_addr))
-		if (found->ip_addr == ip_addr)
-			return found;
-
-	return NULL;
-}
-
-static void _rocker_neigh_add(struct rocker *rocker,
-			      struct switchdev_trans *trans,
-			      struct rocker_neigh_tbl_entry *entry)
-{
-	if (!switchdev_trans_ph_commit(trans))
-		entry->index = rocker->neigh_tbl_next_index++;
-	if (switchdev_trans_ph_prepare(trans))
-		return;
-	entry->ref_count++;
-	hash_add(rocker->neigh_tbl, &entry->entry,
-		 be32_to_cpu(entry->ip_addr));
-}
-
-static void _rocker_neigh_del(struct switchdev_trans *trans,
-			      struct rocker_neigh_tbl_entry *entry)
-{
-	if (switchdev_trans_ph_prepare(trans))
-		return;
-	if (--entry->ref_count == 0) {
-		hash_del(&entry->entry);
-		rocker_kfree(trans, entry);
-	}
-}
-
-static void _rocker_neigh_update(struct rocker_neigh_tbl_entry *entry,
-				 struct switchdev_trans *trans,
-				 const u8 *eth_dst, bool ttl_check)
-{
-	if (eth_dst) {
-		ether_addr_copy(entry->eth_dst, eth_dst);
-		entry->ttl_check = ttl_check;
-	} else if (!switchdev_trans_ph_prepare(trans)) {
-		entry->ref_count++;
-	}
-}
-
-static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans,
-				  int flags, __be32 ip_addr, const u8 *eth_dst)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_neigh_tbl_entry *entry;
-	struct rocker_neigh_tbl_entry *found;
-	unsigned long lock_flags;
-	__be16 eth_type = htons(ETH_P_IP);
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 group_id;
-	u32 priority = 0;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	bool updating;
-	bool removing;
-	int err = 0;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	spin_lock_irqsave(&rocker->neigh_tbl_lock, lock_flags);
-
-	found = rocker_neigh_tbl_find(rocker, ip_addr);
-
-	updating = found && adding;
-	removing = found && !adding;
-	adding = !found && adding;
-
-	if (adding) {
-		entry->ip_addr = ip_addr;
-		entry->dev = rocker_port->dev;
-		ether_addr_copy(entry->eth_dst, eth_dst);
-		entry->ttl_check = true;
-		_rocker_neigh_add(rocker, trans, entry);
-	} else if (removing) {
-		memcpy(entry, found, sizeof(*entry));
-		_rocker_neigh_del(trans, found);
-	} else if (updating) {
-		_rocker_neigh_update(found, trans, eth_dst, true);
-		memcpy(entry, found, sizeof(*entry));
-	} else {
-		err = -ENOENT;
-	}
-
-	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
-
-	if (err)
-		goto err_out;
-
-	/* For each active neighbor, we have an L3 unicast group and
-	 * a /32 route to the neighbor, which uses the L3 unicast
-	 * group.  The L3 unicast group can also be referred to by
-	 * other routes' nexthops.
-	 */
-
-	err = rocker_group_l3_unicast(rocker_port, trans, flags,
-				      entry->index,
-				      rocker_port->dev->dev_addr,
-				      entry->eth_dst,
-				      rocker_port->internal_vlan_id,
-				      entry->ttl_check,
-				      rocker_port->pport);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) L3 unicast group index %d\n",
-			   err, entry->index);
-		goto err_out;
-	}
-
-	if (adding || removing) {
-		group_id = ROCKER_GROUP_L3_UNICAST(entry->index);
-		err = rocker_flow_tbl_ucast4_routing(rocker_port, trans,
-						     eth_type, ip_addr,
-						     inet_make_mask(32),
-						     priority, goto_tbl,
-						     group_id, flags);
-
-		if (err)
-			netdev_err(rocker_port->dev,
-				   "Error (%d) /32 unicast route %pI4 group 0x%08x\n",
-				   err, &entry->ip_addr, group_id);
-	}
-
-err_out:
-	if (!adding)
-		rocker_kfree(trans, entry);
-
-	return err;
-}
-
-static int rocker_port_ipv4_resolve(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans,
-				    __be32 ip_addr)
-{
-	struct net_device *dev = rocker_port->dev;
-	struct neighbour *n = __ipv4_neigh_lookup(dev, (__force u32)ip_addr);
-	int err = 0;
-
-	if (!n) {
-		n = neigh_create(&arp_tbl, &ip_addr, dev);
-		if (IS_ERR(n))
-			return IS_ERR(n);
-	}
-
-	/* If the neigh is already resolved, then go ahead and
-	 * install the entry, otherwise start the ARP process to
-	 * resolve the neigh.
-	 */
-
-	if (n->nud_state & NUD_VALID)
-		err = rocker_port_ipv4_neigh(rocker_port, trans, 0,
-					     ip_addr, n->ha);
-	else
-		neigh_event_send(n, NULL);
-
-	neigh_release(n);
-	return err;
-}
-
-static int rocker_port_ipv4_nh(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       __be32 ip_addr, u32 *index)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_neigh_tbl_entry *entry;
-	struct rocker_neigh_tbl_entry *found;
-	unsigned long lock_flags;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	bool updating;
-	bool removing;
-	bool resolved = true;
-	int err = 0;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	spin_lock_irqsave(&rocker->neigh_tbl_lock, lock_flags);
-
-	found = rocker_neigh_tbl_find(rocker, ip_addr);
-	if (found)
-		*index = found->index;
-
-	updating = found && adding;
-	removing = found && !adding;
-	adding = !found && adding;
-
-	if (adding) {
-		entry->ip_addr = ip_addr;
-		entry->dev = rocker_port->dev;
-		_rocker_neigh_add(rocker, trans, entry);
-		*index = entry->index;
-		resolved = false;
-	} else if (removing) {
-		_rocker_neigh_del(trans, found);
-	} else if (updating) {
-		_rocker_neigh_update(found, trans, NULL, false);
-		resolved = !is_zero_ether_addr(found->eth_dst);
-	} else {
-		err = -ENOENT;
-	}
-
-	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
-
-	if (!adding)
-		rocker_kfree(trans, entry);
-
-	if (err)
-		return err;
-
-	/* Resolved means neigh ip_addr is resolved to neigh mac. */
-
-	if (!resolved)
-		err = rocker_port_ipv4_resolve(rocker_port, trans, ip_addr);
-
-	return err;
-}
-
-static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port,
-					struct switchdev_trans *trans,
-					int flags, __be16 vlan_id)
-{
-	struct rocker_port *p;
-	const struct rocker *rocker = rocker_port->rocker;
-	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
-	u32 *group_ids;
-	u8 group_count = 0;
-	int err = 0;
-	int i;
-
-	group_ids = rocker_kcalloc(trans, flags,
-				   rocker->port_count, sizeof(u32));
-	if (!group_ids)
-		return -ENOMEM;
-
-	/* Adjust the flood group for this VLAN.  The flood group
-	 * references an L2 interface group for each port in this
-	 * VLAN.
-	 */
-
-	for (i = 0; i < rocker->port_count; i++) {
-		p = rocker->ports[i];
-		if (!p)
-			continue;
-		if (!rocker_port_is_bridged(p))
-			continue;
-		if (test_bit(ntohs(vlan_id), p->vlan_bitmap)) {
-			group_ids[group_count++] =
-				ROCKER_GROUP_L2_INTERFACE(vlan_id, p->pport);
-		}
-	}
-
-	/* If there are no bridged ports in this VLAN, we're done */
-	if (group_count == 0)
-		goto no_ports_in_vlan;
-
-	err = rocker_group_l2_flood(rocker_port, trans, flags, vlan_id,
-				    group_count, group_ids, group_id);
-	if (err)
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 flood group\n", err);
-
-no_ports_in_vlan:
-	rocker_kfree(trans, group_ids);
-	return err;
-}
-
-static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port,
-				      struct switchdev_trans *trans, int flags,
-				      __be16 vlan_id, bool pop_vlan)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct rocker_port *p;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	u32 out_pport;
-	int ref = 0;
-	int err;
-	int i;
-
-	/* An L2 interface group for this port in this VLAN, but
-	 * only when port STP state is LEARNING|FORWARDING.
-	 */
-
-	if (rocker_port->stp_state == BR_STATE_LEARNING ||
-	    rocker_port->stp_state == BR_STATE_FORWARDING) {
-		out_pport = rocker_port->pport;
-		err = rocker_group_l2_interface(rocker_port, trans, flags,
-						vlan_id, out_pport, pop_vlan);
-		if (err) {
-			netdev_err(rocker_port->dev,
-				   "Error (%d) port VLAN l2 group for pport %d\n",
-				   err, out_pport);
-			return err;
-		}
-	}
-
-	/* An L2 interface group for this VLAN to CPU port.
-	 * Add when first port joins this VLAN and destroy when
-	 * last port leaves this VLAN.
-	 */
-
-	for (i = 0; i < rocker->port_count; i++) {
-		p = rocker->ports[i];
-		if (p && test_bit(ntohs(vlan_id), p->vlan_bitmap))
-			ref++;
-	}
-
-	if ((!adding || ref != 1) && (adding || ref != 0))
-		return 0;
-
-	out_pport = 0;
-	err = rocker_group_l2_interface(rocker_port, trans, flags,
-					vlan_id, out_pport, pop_vlan);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 group for CPU port\n", err);
-		return err;
-	}
-
-	return 0;
-}
-
-static struct rocker_ctrl {
-	const u8 *eth_dst;
-	const u8 *eth_dst_mask;
-	__be16 eth_type;
-	bool acl;
-	bool bridge;
-	bool term;
-	bool copy_to_cpu;
-} rocker_ctrls[] = {
-	[ROCKER_CTRL_LINK_LOCAL_MCAST] = {
-		/* pass link local multicast pkts up to CPU for filtering */
-		.eth_dst = ll_mac,
-		.eth_dst_mask = ll_mask,
-		.acl = true,
-	},
-	[ROCKER_CTRL_LOCAL_ARP] = {
-		/* pass local ARP pkts up to CPU */
-		.eth_dst = zero_mac,
-		.eth_dst_mask = zero_mac,
-		.eth_type = htons(ETH_P_ARP),
-		.acl = true,
-	},
-	[ROCKER_CTRL_IPV4_MCAST] = {
-		/* pass IPv4 mcast pkts up to CPU, RFC 1112 */
-		.eth_dst = ipv4_mcast,
-		.eth_dst_mask = ipv4_mask,
-		.eth_type = htons(ETH_P_IP),
-		.term  = true,
-		.copy_to_cpu = true,
-	},
-	[ROCKER_CTRL_IPV6_MCAST] = {
-		/* pass IPv6 mcast pkts up to CPU, RFC 2464 */
-		.eth_dst = ipv6_mcast,
-		.eth_dst_mask = ipv6_mask,
-		.eth_type = htons(ETH_P_IPV6),
-		.term  = true,
-		.copy_to_cpu = true,
-	},
-	[ROCKER_CTRL_DFLT_BRIDGING] = {
-		/* flood any pkts on vlan */
-		.bridge = true,
-		.copy_to_cpu = true,
-	},
-	[ROCKER_CTRL_DFLT_OVS] = {
-		/* pass all pkts up to CPU */
-		.eth_dst = zero_mac,
-		.eth_dst_mask = zero_mac,
-		.acl = true,
-	},
-};
-
-static int rocker_port_ctrl_vlan_acl(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     const struct rocker_ctrl *ctrl, __be16 vlan_id)
-{
-	u32 in_pport = rocker_port->pport;
-	u32 in_pport_mask = 0xffffffff;
-	u32 out_pport = 0;
-	const u8 *eth_src = NULL;
-	const u8 *eth_src_mask = NULL;
-	__be16 vlan_id_mask = htons(0xffff);
-	u8 ip_proto = 0;
-	u8 ip_proto_mask = 0;
-	u8 ip_tos = 0;
-	u8 ip_tos_mask = 0;
-	u32 group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
-	int err;
-
-	err = rocker_flow_tbl_acl(rocker_port, trans, flags,
-				  in_pport, in_pport_mask,
-				  eth_src, eth_src_mask,
-				  ctrl->eth_dst, ctrl->eth_dst_mask,
-				  ctrl->eth_type,
-				  vlan_id, vlan_id_mask,
-				  ip_proto, ip_proto_mask,
-				  ip_tos, ip_tos_mask,
-				  group_id);
-
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) ctrl ACL\n", err);
-
-	return err;
-}
-
-static int rocker_port_ctrl_vlan_bridge(struct rocker_port *rocker_port,
-					struct switchdev_trans *trans,
-					int flags,
-					const struct rocker_ctrl *ctrl,
-					__be16 vlan_id)
-{
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
-	u32 tunnel_id = 0;
-	int err;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return 0;
-
-	err = rocker_flow_tbl_bridge(rocker_port, trans, flags,
-				     ctrl->eth_dst, ctrl->eth_dst_mask,
-				     vlan_id, tunnel_id,
-				     goto_tbl, group_id, ctrl->copy_to_cpu);
-
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) ctrl FLOOD\n", err);
-
-	return err;
-}
-
-static int rocker_port_ctrl_vlan_term(struct rocker_port *rocker_port,
-				      struct switchdev_trans *trans, int flags,
-				      const struct rocker_ctrl *ctrl, __be16 vlan_id)
-{
-	u32 in_pport_mask = 0xffffffff;
-	__be16 vlan_id_mask = htons(0xffff);
-	int err;
-
-	if (ntohs(vlan_id) == 0)
-		vlan_id = rocker_port->internal_vlan_id;
-
-	err = rocker_flow_tbl_term_mac(rocker_port, trans,
-				       rocker_port->pport, in_pport_mask,
-				       ctrl->eth_type, ctrl->eth_dst,
-				       ctrl->eth_dst_mask, vlan_id,
-				       vlan_id_mask, ctrl->copy_to_cpu,
-				       flags);
-
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) ctrl term\n", err);
-
-	return err;
-}
-
-static int rocker_port_ctrl_vlan(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 const struct rocker_ctrl *ctrl, __be16 vlan_id)
-{
-	if (ctrl->acl)
-		return rocker_port_ctrl_vlan_acl(rocker_port, trans, flags,
-						 ctrl, vlan_id);
-	if (ctrl->bridge)
-		return rocker_port_ctrl_vlan_bridge(rocker_port, trans, flags,
-						    ctrl, vlan_id);
-
-	if (ctrl->term)
-		return rocker_port_ctrl_vlan_term(rocker_port, trans, flags,
-						  ctrl, vlan_id);
-
-	return -EOPNOTSUPP;
-}
-
-static int rocker_port_ctrl_vlan_add(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     __be16 vlan_id)
-{
-	int err = 0;
-	int i;
-
-	for (i = 0; i < ROCKER_CTRL_MAX; i++) {
-		if (rocker_port->ctrls[i]) {
-			err = rocker_port_ctrl_vlan(rocker_port, trans, flags,
-						    &rocker_ctrls[i], vlan_id);
-			if (err)
-				return err;
-		}
-	}
-
-	return err;
-}
-
-static int rocker_port_ctrl(struct rocker_port *rocker_port,
-			    struct switchdev_trans *trans, int flags,
-			    const struct rocker_ctrl *ctrl)
-{
-	u16 vid;
-	int err = 0;
-
-	for (vid = 1; vid < VLAN_N_VID; vid++) {
-		if (!test_bit(vid, rocker_port->vlan_bitmap))
-			continue;
-		err = rocker_port_ctrl_vlan(rocker_port, trans, flags,
-					    ctrl, htons(vid));
-		if (err)
-			break;
-	}
-
-	return err;
-}
-
-static int rocker_port_vlan(struct rocker_port *rocker_port,
-			    struct switchdev_trans *trans, int flags, u16 vid)
-{
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
-	u32 in_pport = rocker_port->pport;
-	__be16 vlan_id = htons(vid);
-	__be16 vlan_id_mask = htons(0xffff);
-	__be16 internal_vlan_id;
-	bool untagged;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	int err;
-
-	internal_vlan_id = rocker_port_vid_to_vlan(rocker_port, vid, &untagged);
-
-	if (adding && test_bit(ntohs(internal_vlan_id),
-			       rocker_port->vlan_bitmap))
-			return 0; /* already added */
-	else if (!adding && !test_bit(ntohs(internal_vlan_id),
-				      rocker_port->vlan_bitmap))
-			return 0; /* already removed */
-
-	change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap);
-
-	if (adding) {
-		err = rocker_port_ctrl_vlan_add(rocker_port, trans, flags,
-						internal_vlan_id);
-		if (err) {
-			netdev_err(rocker_port->dev,
-				   "Error (%d) port ctrl vlan add\n", err);
-			goto err_out;
-		}
-	}
-
-	err = rocker_port_vlan_l2_groups(rocker_port, trans, flags,
-					 internal_vlan_id, untagged);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 groups\n", err);
-		goto err_out;
-	}
-
-	err = rocker_port_vlan_flood_group(rocker_port, trans, flags,
-					   internal_vlan_id);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 flood group\n", err);
-		goto err_out;
-	}
-
-	err = rocker_flow_tbl_vlan(rocker_port, trans, flags,
-				   in_pport, vlan_id, vlan_id_mask,
-				   goto_tbl, untagged, internal_vlan_id);
-	if (err)
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN table\n", err);
-
-err_out:
-	if (switchdev_trans_ph_prepare(trans))
-		change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap);
-
-	return err;
-}
-
-static int rocker_port_ig_tbl(struct rocker_port *rocker_port,
-			      struct switchdev_trans *trans, int flags)
-{
-	enum rocker_of_dpa_table_id goto_tbl;
-	u32 in_pport;
-	u32 in_pport_mask;
-	int err;
-
-	/* Normal Ethernet Frames.  Matches pkts from any local physical
-	 * ports.  Goto VLAN tbl.
-	 */
-
-	in_pport = 0;
-	in_pport_mask = 0xffff0000;
-	goto_tbl = ROCKER_OF_DPA_TABLE_ID_VLAN;
-
-	err = rocker_flow_tbl_ig_port(rocker_port, trans, flags,
-				      in_pport, in_pport_mask,
-				      goto_tbl);
-	if (err)
-		netdev_err(rocker_port->dev,
-			   "Error (%d) ingress port table entry\n", err);
-
-	return err;
-}
-
-struct rocker_fdb_learn_work {
-	struct work_struct work;
-	struct rocker_port *rocker_port;
-	struct switchdev_trans *trans;
-	int flags;
-	u8 addr[ETH_ALEN];
-	u16 vid;
-};
-
-static void rocker_port_fdb_learn_work(struct work_struct *work)
-{
-	const struct rocker_fdb_learn_work *lw =
-		container_of(work, struct rocker_fdb_learn_work, work);
-	bool removing = (lw->flags & ROCKER_OP_FLAG_REMOVE);
-	bool learned = (lw->flags & ROCKER_OP_FLAG_LEARNED);
-	struct switchdev_notifier_fdb_info info;
-
-	info.addr = lw->addr;
-	info.vid = lw->vid;
-
-	rtnl_lock();
-	if (learned && removing)
-		call_switchdev_notifiers(SWITCHDEV_FDB_DEL,
-					 lw->rocker_port->dev, &info.info);
-	else if (learned && !removing)
-		call_switchdev_notifiers(SWITCHDEV_FDB_ADD,
-					 lw->rocker_port->dev, &info.info);
-	rtnl_unlock();
-
-	rocker_kfree(lw->trans, work);
-}
-
-static int rocker_port_fdb_learn(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 const u8 *addr, __be16 vlan_id)
-{
-	struct rocker_fdb_learn_work *lw;
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 out_pport = rocker_port->pport;
-	u32 tunnel_id = 0;
-	u32 group_id = ROCKER_GROUP_NONE;
-	bool syncing = !!(rocker_port->brport_flags & BR_LEARNING_SYNC);
-	bool copy_to_cpu = false;
-	int err;
-
-	if (rocker_port_is_bridged(rocker_port))
-		group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
-
-	if (!(flags & ROCKER_OP_FLAG_REFRESH)) {
-		err = rocker_flow_tbl_bridge(rocker_port, trans, flags, addr,
-					     NULL, vlan_id, tunnel_id, goto_tbl,
-					     group_id, copy_to_cpu);
-		if (err)
-			return err;
-	}
-
-	if (!syncing)
-		return 0;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return 0;
-
-	lw = rocker_kzalloc(trans, flags, sizeof(*lw));
-	if (!lw)
-		return -ENOMEM;
-
-	INIT_WORK(&lw->work, rocker_port_fdb_learn_work);
-
-	lw->rocker_port = rocker_port;
-	lw->trans = trans;
-	lw->flags = flags;
-	ether_addr_copy(lw->addr, addr);
-	lw->vid = rocker_port_vlan_to_vid(rocker_port, vlan_id);
-
-	if (switchdev_trans_ph_prepare(trans))
-		rocker_kfree(trans, lw);
-	else
-		schedule_work(&lw->work);
-
-	return 0;
-}
-
-static struct rocker_fdb_tbl_entry *
-rocker_fdb_tbl_find(const struct rocker *rocker,
-		    const struct rocker_fdb_tbl_entry *match)
-{
-	struct rocker_fdb_tbl_entry *found;
-
-	hash_for_each_possible(rocker->fdb_tbl, found, entry, match->key_crc32)
-		if (memcmp(&found->key, &match->key, sizeof(found->key)) == 0)
-			return found;
-
-	return NULL;
-}
-
-static int rocker_port_fdb(struct rocker_port *rocker_port,
-			   struct switchdev_trans *trans,
-			   const unsigned char *addr,
-			   __be16 vlan_id, int flags)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_fdb_tbl_entry *fdb;
-	struct rocker_fdb_tbl_entry *found;
-	bool removing = (flags & ROCKER_OP_FLAG_REMOVE);
-	unsigned long lock_flags;
-
-	fdb = rocker_kzalloc(trans, flags, sizeof(*fdb));
-	if (!fdb)
-		return -ENOMEM;
-
-	fdb->learned = (flags & ROCKER_OP_FLAG_LEARNED);
-	fdb->touched = jiffies;
-	fdb->key.rocker_port = rocker_port;
-	ether_addr_copy(fdb->key.addr, addr);
-	fdb->key.vlan_id = vlan_id;
-	fdb->key_crc32 = crc32(~0, &fdb->key, sizeof(fdb->key));
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-
-	found = rocker_fdb_tbl_find(rocker, fdb);
-
-	if (found) {
-		found->touched = jiffies;
-		if (removing) {
-			rocker_kfree(trans, fdb);
-			if (!switchdev_trans_ph_prepare(trans))
-				hash_del(&found->entry);
-		}
-	} else if (!removing) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_add(rocker->fdb_tbl, &fdb->entry,
-				 fdb->key_crc32);
-	}
-
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	/* Check if adding and already exists, or removing and can't find */
-	if (!found != !removing) {
-		rocker_kfree(trans, fdb);
-		if (!found && removing)
-			return 0;
-		/* Refreshing existing to update aging timers */
-		flags |= ROCKER_OP_FLAG_REFRESH;
-	}
-
-	return rocker_port_fdb_learn(rocker_port, trans, flags, addr, vlan_id);
-}
-
-static int rocker_port_fdb_flush(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_fdb_tbl_entry *found;
-	unsigned long lock_flags;
-	struct hlist_node *tmp;
-	int bkt;
-	int err = 0;
-
-	if (rocker_port->stp_state == BR_STATE_LEARNING ||
-	    rocker_port->stp_state == BR_STATE_FORWARDING)
-		return 0;
-
-	flags |= ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE;
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) {
-		if (found->key.rocker_port != rocker_port)
-			continue;
-		if (!found->learned)
-			continue;
-		err = rocker_port_fdb_learn(rocker_port, trans, flags,
-					    found->key.addr,
-					    found->key.vlan_id);
-		if (err)
-			goto err_out;
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-	}
-
-err_out:
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	return err;
-}
-
-static void rocker_fdb_cleanup(unsigned long data)
-{
-	struct rocker *rocker = (struct rocker *)data;
-	struct rocker_port *rocker_port;
-	struct rocker_fdb_tbl_entry *entry;
-	struct hlist_node *tmp;
-	unsigned long next_timer = jiffies + BR_MIN_AGEING_TIME;
-	unsigned long expires;
-	unsigned long lock_flags;
-	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE |
-		    ROCKER_OP_FLAG_LEARNED;
-	int bkt;
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, entry, entry) {
-		if (!entry->learned)
-			continue;
-		rocker_port = entry->key.rocker_port;
-		expires = entry->touched + rocker_port->ageing_time;
-		if (time_before_eq(expires, jiffies)) {
-			rocker_port_fdb_learn(rocker_port, NULL,
-					      flags, entry->key.addr,
-					      entry->key.vlan_id);
-			hash_del(&entry->entry);
-		} else if (time_before(expires, next_timer)) {
-			next_timer = expires;
-		}
-	}
-
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	mod_timer(&rocker->fdb_cleanup_timer, round_jiffies_up(next_timer));
-}
-
-static int rocker_port_router_mac(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags,
-				  __be16 vlan_id)
-{
-	u32 in_pport_mask = 0xffffffff;
-	__be16 eth_type;
-	const u8 *dst_mac_mask = ff_mac;
-	__be16 vlan_id_mask = htons(0xffff);
-	bool copy_to_cpu = false;
-	int err;
-
-	if (ntohs(vlan_id) == 0)
-		vlan_id = rocker_port->internal_vlan_id;
-
-	eth_type = htons(ETH_P_IP);
-	err = rocker_flow_tbl_term_mac(rocker_port, trans,
-				       rocker_port->pport, in_pport_mask,
-				       eth_type, rocker_port->dev->dev_addr,
-				       dst_mac_mask, vlan_id, vlan_id_mask,
-				       copy_to_cpu, flags);
-	if (err)
-		return err;
-
-	eth_type = htons(ETH_P_IPV6);
-	err = rocker_flow_tbl_term_mac(rocker_port, trans,
-				       rocker_port->pport, in_pport_mask,
-				       eth_type, rocker_port->dev->dev_addr,
-				       dst_mac_mask, vlan_id, vlan_id_mask,
-				       copy_to_cpu, flags);
-
-	return err;
-}
-
-static int rocker_port_fwding(struct rocker_port *rocker_port,
-			      struct switchdev_trans *trans, int flags)
-{
-	bool pop_vlan;
-	u32 out_pport;
-	__be16 vlan_id;
-	u16 vid;
-	int err;
-
-	/* Port will be forwarding-enabled if its STP state is LEARNING
-	 * or FORWARDING.  Traffic from CPU can still egress, regardless of
-	 * port STP state.  Use L2 interface group on port VLANs as a way
-	 * to toggle port forwarding: if forwarding is disabled, L2
-	 * interface group will not exist.
-	 */
-
-	if (rocker_port->stp_state != BR_STATE_LEARNING &&
-	    rocker_port->stp_state != BR_STATE_FORWARDING)
-		flags |= ROCKER_OP_FLAG_REMOVE;
-
-	out_pport = rocker_port->pport;
-	for (vid = 1; vid < VLAN_N_VID; vid++) {
-		if (!test_bit(vid, rocker_port->vlan_bitmap))
-			continue;
-		vlan_id = htons(vid);
-		pop_vlan = rocker_vlan_id_is_internal(vlan_id);
-		err = rocker_group_l2_interface(rocker_port, trans, flags,
-						vlan_id, out_pport, pop_vlan);
-		if (err) {
-			netdev_err(rocker_port->dev,
-				   "Error (%d) port VLAN l2 group for pport %d\n",
-				   err, out_pport);
-			return err;
-		}
-	}
-
-	return 0;
-}
-
-static int rocker_port_stp_update(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags,
-				  u8 state)
-{
-	bool want[ROCKER_CTRL_MAX] = { 0, };
-	bool prev_ctrls[ROCKER_CTRL_MAX];
-	u8 uninitialized_var(prev_state);
-	int err;
-	int i;
-
-	if (switchdev_trans_ph_prepare(trans)) {
-		memcpy(prev_ctrls, rocker_port->ctrls, sizeof(prev_ctrls));
-		prev_state = rocker_port->stp_state;
-	}
-
-	if (rocker_port->stp_state == state)
-		return 0;
-
-	rocker_port->stp_state = state;
-
-	switch (state) {
-	case BR_STATE_DISABLED:
-		/* port is completely disabled */
-		break;
-	case BR_STATE_LISTENING:
-	case BR_STATE_BLOCKING:
-		want[ROCKER_CTRL_LINK_LOCAL_MCAST] = true;
-		break;
-	case BR_STATE_LEARNING:
-	case BR_STATE_FORWARDING:
-		if (!rocker_port_is_ovsed(rocker_port))
-			want[ROCKER_CTRL_LINK_LOCAL_MCAST] = true;
-		want[ROCKER_CTRL_IPV4_MCAST] = true;
-		want[ROCKER_CTRL_IPV6_MCAST] = true;
-		if (rocker_port_is_bridged(rocker_port))
-			want[ROCKER_CTRL_DFLT_BRIDGING] = true;
-		else if (rocker_port_is_ovsed(rocker_port))
-			want[ROCKER_CTRL_DFLT_OVS] = true;
-		else
-			want[ROCKER_CTRL_LOCAL_ARP] = true;
-		break;
-	}
-
-	for (i = 0; i < ROCKER_CTRL_MAX; i++) {
-		if (want[i] != rocker_port->ctrls[i]) {
-			int ctrl_flags = flags |
-					 (want[i] ? 0 : ROCKER_OP_FLAG_REMOVE);
-			err = rocker_port_ctrl(rocker_port, trans, ctrl_flags,
-					       &rocker_ctrls[i]);
-			if (err)
-				goto err_out;
-			rocker_port->ctrls[i] = want[i];
-		}
-	}
-
-	err = rocker_port_fdb_flush(rocker_port, trans, flags);
-	if (err)
-		goto err_out;
-
-	err = rocker_port_fwding(rocker_port, trans, flags);
-
-err_out:
-	if (switchdev_trans_ph_prepare(trans)) {
-		memcpy(rocker_port->ctrls, prev_ctrls, sizeof(prev_ctrls));
-		rocker_port->stp_state = prev_state;
-	}
-
-	return err;
-}
-
-static int rocker_port_fwd_enable(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags)
-{
-	if (rocker_port_is_bridged(rocker_port))
-		/* bridge STP will enable port */
-		return 0;
-
-	/* port is not bridged, so simulate going to FORWARDING state */
-	return rocker_port_stp_update(rocker_port, trans, flags,
-				      BR_STATE_FORWARDING);
-}
-
-static int rocker_port_fwd_disable(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans, int flags)
-{
-	if (rocker_port_is_bridged(rocker_port))
-		/* bridge STP will disable port */
-		return 0;
-
-	/* port is not bridged, so simulate going to DISABLED state */
-	return rocker_port_stp_update(rocker_port, trans, flags,
-				      BR_STATE_DISABLED);
-}
-
-static struct rocker_internal_vlan_tbl_entry *
-rocker_internal_vlan_tbl_find(const struct rocker *rocker, int ifindex)
-{
-	struct rocker_internal_vlan_tbl_entry *found;
-
-	hash_for_each_possible(rocker->internal_vlan_tbl, found,
-			       entry, ifindex) {
-		if (found->ifindex == ifindex)
-			return found;
-	}
-
-	return NULL;
-}
-
-static __be16 rocker_port_internal_vlan_id_get(struct rocker_port *rocker_port,
-					       int ifindex)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_internal_vlan_tbl_entry *entry;
-	struct rocker_internal_vlan_tbl_entry *found;
-	unsigned long lock_flags;
-	int i;
-
-	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
-	if (!entry)
-		return 0;
-
-	entry->ifindex = ifindex;
-
-	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, lock_flags);
-
-	found = rocker_internal_vlan_tbl_find(rocker, ifindex);
-	if (found) {
-		kfree(entry);
-		goto found;
-	}
-
-	found = entry;
-	hash_add(rocker->internal_vlan_tbl, &found->entry, found->ifindex);
-
-	for (i = 0; i < ROCKER_N_INTERNAL_VLANS; i++) {
-		if (test_and_set_bit(i, rocker->internal_vlan_bitmap))
-			continue;
-		found->vlan_id = htons(ROCKER_INTERNAL_VLAN_ID_BASE + i);
-		goto found;
-	}
-
-	netdev_err(rocker_port->dev, "Out of internal VLAN IDs\n");
-
-found:
-	found->ref_count++;
-	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags);
-
-	return found->vlan_id;
-}
-
-static void
-rocker_port_internal_vlan_id_put(const struct rocker_port *rocker_port,
-				 int ifindex)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_internal_vlan_tbl_entry *found;
-	unsigned long lock_flags;
-	unsigned long bit;
-
-	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, lock_flags);
-
-	found = rocker_internal_vlan_tbl_find(rocker, ifindex);
-	if (!found) {
-		netdev_err(rocker_port->dev,
-			   "ifindex (%d) not found in internal VLAN tbl\n",
-			   ifindex);
-		goto not_found;
-	}
-
-	if (--found->ref_count <= 0) {
-		bit = ntohs(found->vlan_id) - ROCKER_INTERNAL_VLAN_ID_BASE;
-		clear_bit(bit, rocker->internal_vlan_bitmap);
-		hash_del(&found->entry);
-		kfree(found);
-	}
-
-not_found:
-	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags);
-}
-
-static int rocker_port_fib_ipv4(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, __be32 dst,
-				int dst_len, const struct fib_info *fi,
-				u32 tb_id, int flags)
-{
-	const struct fib_nh *nh;
-	__be16 eth_type = htons(ETH_P_IP);
-	__be32 dst_mask = inet_make_mask(dst_len);
-	__be16 internal_vlan_id = rocker_port->internal_vlan_id;
-	u32 priority = fi->fib_priority;
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 group_id;
-	bool nh_on_port;
-	bool has_gw;
-	u32 index;
-	int err;
-
-	/* XXX support ECMP */
-
-	nh = fi->fib_nh;
-	nh_on_port = (fi->fib_dev == rocker_port->dev);
-	has_gw = !!nh->nh_gw;
-
-	if (has_gw && nh_on_port) {
-		err = rocker_port_ipv4_nh(rocker_port, trans, flags,
-					  nh->nh_gw, &index);
-		if (err)
-			return err;
-
-		group_id = ROCKER_GROUP_L3_UNICAST(index);
-	} else {
-		/* Send to CPU for processing */
-		group_id = ROCKER_GROUP_L2_INTERFACE(internal_vlan_id, 0);
-	}
-
-	err = rocker_flow_tbl_ucast4_routing(rocker_port, trans, eth_type, dst,
-					     dst_mask, priority, goto_tbl,
-					     group_id, flags);
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) IPv4 route %pI4\n",
-			   err, &dst);
-
-	return err;
-}
-
-/*****************
- * Net device ops
- *****************/
-
-static int rocker_port_open(struct net_device *dev)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int err;
-
-	err = rocker_port_dma_rings_init(rocker_port);
-	if (err)
-		return err;
-
-	err = request_irq(rocker_msix_tx_vector(rocker_port),
-			  rocker_tx_irq_handler, 0,
-			  rocker_driver_name, rocker_port);
-	if (err) {
-		netdev_err(rocker_port->dev, "cannot assign tx irq\n");
-		goto err_request_tx_irq;
-	}
-
-	err = request_irq(rocker_msix_rx_vector(rocker_port),
-			  rocker_rx_irq_handler, 0,
-			  rocker_driver_name, rocker_port);
-	if (err) {
-		netdev_err(rocker_port->dev, "cannot assign rx irq\n");
-		goto err_request_rx_irq;
-	}
-
-	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
-	if (err)
-		goto err_fwd_enable;
-
-	napi_enable(&rocker_port->napi_tx);
-	napi_enable(&rocker_port->napi_rx);
-	if (!dev->proto_down)
-		rocker_port_set_enable(rocker_port, true);
-	netif_start_queue(dev);
-	return 0;
-
-err_fwd_enable:
-	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
-err_request_rx_irq:
-	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
-err_request_tx_irq:
-	rocker_port_dma_rings_fini(rocker_port);
-	return err;
-}
-
-static int rocker_port_stop(struct net_device *dev)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-
-	netif_stop_queue(dev);
-	rocker_port_set_enable(rocker_port, false);
-	napi_disable(&rocker_port->napi_rx);
-	napi_disable(&rocker_port->napi_tx);
-	rocker_port_fwd_disable(rocker_port, NULL,
-				ROCKER_OP_FLAG_NOWAIT);
-	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
-	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
-	rocker_port_dma_rings_fini(rocker_port);
-
-	return 0;
-}
-
-static void rocker_tx_desc_frags_unmap(const struct rocker_port *rocker_port,
-				       const struct rocker_desc_info *desc_info)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct pci_dev *pdev = rocker->pdev;
-	const struct rocker_tlv *attrs[ROCKER_TLV_TX_MAX + 1];
-	struct rocker_tlv *attr;
-	int rem;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_TX_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_TX_FRAGS])
-		return;
-	rocker_tlv_for_each_nested(attr, attrs[ROCKER_TLV_TX_FRAGS], rem) {
-		const struct rocker_tlv *frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_MAX + 1];
-		dma_addr_t dma_handle;
-		size_t len;
-
-		if (rocker_tlv_type(attr) != ROCKER_TLV_TX_FRAG)
-			continue;
-		rocker_tlv_parse_nested(frag_attrs, ROCKER_TLV_TX_FRAG_ATTR_MAX,
-					attr);
-		if (!frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR] ||
-		    !frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN])
-			continue;
-		dma_handle = rocker_tlv_get_u64(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR]);
-		len = rocker_tlv_get_u16(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN]);
-		pci_unmap_single(pdev, dma_handle, len, DMA_TO_DEVICE);
-	}
-}
-
-static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port,
-				       struct rocker_desc_info *desc_info,
-				       char *buf, size_t buf_len)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct pci_dev *pdev = rocker->pdev;
-	dma_addr_t dma_handle;
-	struct rocker_tlv *frag;
-
-	dma_handle = pci_map_single(pdev, buf, buf_len, DMA_TO_DEVICE);
-	if (unlikely(pci_dma_mapping_error(pdev, dma_handle))) {
-		if (net_ratelimit())
-			netdev_err(rocker_port->dev, "failed to dma map tx frag\n");
-		return -EIO;
-	}
-	frag = rocker_tlv_nest_start(desc_info, ROCKER_TLV_TX_FRAG);
-	if (!frag)
-		goto unmap_frag;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_TX_FRAG_ATTR_ADDR,
-			       dma_handle))
-		goto nest_cancel;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_TX_FRAG_ATTR_LEN,
-			       buf_len))
-		goto nest_cancel;
-	rocker_tlv_nest_end(desc_info, frag);
-	return 0;
-
-nest_cancel:
-	rocker_tlv_nest_cancel(desc_info, frag);
-unmap_frag:
-	pci_unmap_single(pdev, dma_handle, buf_len, DMA_TO_DEVICE);
-	return -EMSGSIZE;
-}
-
-static netdev_tx_t rocker_port_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_desc_info *desc_info;
-	struct rocker_tlv *frags;
-	int i;
-	int err;
-
-	desc_info = rocker_desc_head_get(&rocker_port->tx_ring);
-	if (unlikely(!desc_info)) {
-		if (net_ratelimit())
-			netdev_err(dev, "tx ring full when queue awake\n");
-		return NETDEV_TX_BUSY;
-	}
-
-	rocker_desc_cookie_ptr_set(desc_info, skb);
-
-	frags = rocker_tlv_nest_start(desc_info, ROCKER_TLV_TX_FRAGS);
-	if (!frags)
-		goto out;
-	err = rocker_tx_desc_frag_map_put(rocker_port, desc_info,
-					  skb->data, skb_headlen(skb));
-	if (err)
-		goto nest_cancel;
-	if (skb_shinfo(skb)->nr_frags > ROCKER_TX_FRAGS_MAX) {
-		err = skb_linearize(skb);
-		if (err)
-			goto unmap_frags;
-	}
-
-	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-
-		err = rocker_tx_desc_frag_map_put(rocker_port, desc_info,
-						  skb_frag_address(frag),
-						  skb_frag_size(frag));
-		if (err)
-			goto unmap_frags;
-	}
-	rocker_tlv_nest_end(desc_info, frags);
-
-	rocker_desc_gen_clear(desc_info);
-	rocker_desc_head_set(rocker, &rocker_port->tx_ring, desc_info);
-
-	desc_info = rocker_desc_head_get(&rocker_port->tx_ring);
-	if (!desc_info)
-		netif_stop_queue(dev);
-
-	return NETDEV_TX_OK;
-
-unmap_frags:
-	rocker_tx_desc_frags_unmap(rocker_port, desc_info);
-nest_cancel:
-	rocker_tlv_nest_cancel(desc_info, frags);
-out:
-	dev_kfree_skb(skb);
-	dev->stats.tx_dropped++;
-
-	return NETDEV_TX_OK;
-}
-
-static int rocker_port_set_mac_address(struct net_device *dev, void *p)
-{
-	struct sockaddr *addr = p;
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int err;
-
-	if (!is_valid_ether_addr(addr->sa_data))
-		return -EADDRNOTAVAIL;
-
-	err = rocker_cmd_set_port_settings_macaddr(rocker_port, addr->sa_data);
-	if (err)
-		return err;
-	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
-	return 0;
-}
-
-static int rocker_port_change_mtu(struct net_device *dev, int new_mtu)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int running = netif_running(dev);
-	int err;
-
-#define ROCKER_PORT_MIN_MTU	68
-#define ROCKER_PORT_MAX_MTU	9000
-
-	if (new_mtu < ROCKER_PORT_MIN_MTU || new_mtu > ROCKER_PORT_MAX_MTU)
-		return -EINVAL;
-
-	if (running)
-		rocker_port_stop(dev);
-
-	netdev_info(dev, "MTU change from %d to %d\n", dev->mtu, new_mtu);
-	dev->mtu = new_mtu;
-
-	err = rocker_cmd_set_port_settings_mtu(rocker_port, new_mtu);
-	if (err)
-		return err;
-
-	if (running)
-		err = rocker_port_open(dev);
-
-	return err;
-}
-
-static int rocker_port_get_phys_port_name(struct net_device *dev,
-					  char *buf, size_t len)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	struct port_name name = { .buf = buf, .len = len };
-	int err;
-
-	err = rocker_cmd_exec(rocker_port, NULL, 0,
-			      rocker_cmd_get_port_settings_prep, NULL,
-			      rocker_cmd_get_port_settings_phys_name_proc,
-			      &name);
-
-	return err ? -EOPNOTSUPP : 0;
-}
-
-static int rocker_port_change_proto_down(struct net_device *dev,
-					 bool proto_down)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-
-	if (rocker_port->dev->flags & IFF_UP)
-		rocker_port_set_enable(rocker_port, !proto_down);
-	rocker_port->dev->proto_down = proto_down;
-	return 0;
-}
-
-static void rocker_port_neigh_destroy(struct neighbour *n)
-{
-	struct rocker_port *rocker_port = netdev_priv(n->dev);
-	int flags = ROCKER_OP_FLAG_REMOVE | ROCKER_OP_FLAG_NOWAIT;
-	__be32 ip_addr = *(__be32 *)n->primary_key;
-
-	rocker_port_ipv4_neigh(rocker_port, NULL,
-			       flags, ip_addr, n->ha);
-}
-
-static const struct net_device_ops rocker_port_netdev_ops = {
-	.ndo_open			= rocker_port_open,
-	.ndo_stop			= rocker_port_stop,
-	.ndo_start_xmit			= rocker_port_xmit,
-	.ndo_set_mac_address		= rocker_port_set_mac_address,
-	.ndo_change_mtu			= rocker_port_change_mtu,
-	.ndo_bridge_getlink		= switchdev_port_bridge_getlink,
-	.ndo_bridge_setlink		= switchdev_port_bridge_setlink,
-	.ndo_bridge_dellink		= switchdev_port_bridge_dellink,
-	.ndo_fdb_add			= switchdev_port_fdb_add,
-	.ndo_fdb_del			= switchdev_port_fdb_del,
-	.ndo_fdb_dump			= switchdev_port_fdb_dump,
-	.ndo_get_phys_port_name		= rocker_port_get_phys_port_name,
-	.ndo_change_proto_down		= rocker_port_change_proto_down,
-	.ndo_neigh_destroy		= rocker_port_neigh_destroy,
-};
-
-/********************
- * swdev interface
- ********************/
-
-static int rocker_port_attr_get(struct net_device *dev,
-				struct switchdev_attr *attr)
-{
-	const struct rocker_port *rocker_port = netdev_priv(dev);
-	const struct rocker *rocker = rocker_port->rocker;
-
-	switch (attr->id) {
-	case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
-		attr->u.ppid.id_len = sizeof(rocker->hw.id);
-		memcpy(&attr->u.ppid.id, &rocker->hw.id, attr->u.ppid.id_len);
-		break;
-	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
-		attr->u.brport_flags = rocker_port->brport_flags;
-		break;
-	default:
-		return -EOPNOTSUPP;
-	}
-
-	return 0;
-}
-
-static int rocker_port_brport_flags_set(struct rocker_port *rocker_port,
-					struct switchdev_trans *trans,
-					unsigned long brport_flags)
-{
-	unsigned long orig_flags;
-	int err = 0;
-
-	orig_flags = rocker_port->brport_flags;
-	rocker_port->brport_flags = brport_flags;
-	if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING)
-		err = rocker_port_set_learning(rocker_port, trans);
-
-	if (switchdev_trans_ph_prepare(trans))
-		rocker_port->brport_flags = orig_flags;
-
-	return err;
-}
-
-static int rocker_port_bridge_ageing_time(struct rocker_port *rocker_port,
-					  struct switchdev_trans *trans,
-					  u32 ageing_time)
-{
-	if (!switchdev_trans_ph_prepare(trans)) {
-		rocker_port->ageing_time = clock_t_to_jiffies(ageing_time);
-		mod_timer(&rocker_port->rocker->fdb_cleanup_timer, jiffies);
-	}
-
-	return 0;
-}
-
-static int rocker_port_attr_set(struct net_device *dev,
-				const struct switchdev_attr *attr,
-				struct switchdev_trans *trans)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int err = 0;
-
-	switch (attr->id) {
-	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
-		err = rocker_port_stp_update(rocker_port, trans, 0,
-					     attr->u.stp_state);
-		break;
-	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
-		err = rocker_port_brport_flags_set(rocker_port, trans,
-						   attr->u.brport_flags);
-		break;
-	case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME:
-		err = rocker_port_bridge_ageing_time(rocker_port, trans,
-						     attr->u.ageing_time);
-		break;
-	default:
-		err = -EOPNOTSUPP;
-		break;
-	}
-
-	return err;
-}
-
-static int rocker_port_vlan_add(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans,
-				u16 vid, u16 flags)
-{
-	int err;
-
-	/* XXX deal with flags for PVID and untagged */
-
-	err = rocker_port_vlan(rocker_port, trans, 0, vid);
-	if (err)
-		return err;
-
-	err = rocker_port_router_mac(rocker_port, trans, 0, htons(vid));
-	if (err)
-		rocker_port_vlan(rocker_port, trans,
-				 ROCKER_OP_FLAG_REMOVE, vid);
-
-	return err;
-}
-
-static int rocker_port_vlans_add(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans,
-				 const struct switchdev_obj_port_vlan *vlan)
-{
-	u16 vid;
-	int err;
-
-	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
-		err = rocker_port_vlan_add(rocker_port, trans,
-					   vid, vlan->flags);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
-static int rocker_port_fdb_add(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans,
-			       const struct switchdev_obj_port_fdb *fdb)
-{
-	__be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL);
-	int flags = 0;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return -EINVAL;
-
-	return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags);
-}
-
-static int rocker_port_obj_add(struct net_device *dev,
-			       const struct switchdev_obj *obj,
-			       struct switchdev_trans *trans)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	const struct switchdev_obj_ipv4_fib *fib4;
-	int err = 0;
-
-	switch (obj->id) {
-	case SWITCHDEV_OBJ_ID_PORT_VLAN:
-		err = rocker_port_vlans_add(rocker_port, trans,
-					    SWITCHDEV_OBJ_PORT_VLAN(obj));
-		break;
-	case SWITCHDEV_OBJ_ID_IPV4_FIB:
-		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
-		err = rocker_port_fib_ipv4(rocker_port, trans,
-					   htonl(fib4->dst), fib4->dst_len,
-					   &fib4->fi, fib4->tb_id, 0);
-		break;
-	case SWITCHDEV_OBJ_ID_PORT_FDB:
-		err = rocker_port_fdb_add(rocker_port, trans,
-					  SWITCHDEV_OBJ_PORT_FDB(obj));
-		break;
-	default:
-		err = -EOPNOTSUPP;
-		break;
-	}
-
-	return err;
-}
-
-static int rocker_port_vlan_del(struct rocker_port *rocker_port,
-				u16 vid, u16 flags)
-{
-	int err;
-
-	err = rocker_port_router_mac(rocker_port, NULL,
-				     ROCKER_OP_FLAG_REMOVE, htons(vid));
-	if (err)
-		return err;
-
-	return rocker_port_vlan(rocker_port, NULL,
-				ROCKER_OP_FLAG_REMOVE, vid);
-}
-
-static int rocker_port_vlans_del(struct rocker_port *rocker_port,
-				 const struct switchdev_obj_port_vlan *vlan)
-{
-	u16 vid;
-	int err;
-
-	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
-		err = rocker_port_vlan_del(rocker_port, vid, vlan->flags);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
-static int rocker_port_fdb_del(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans,
-			       const struct switchdev_obj_port_fdb *fdb)
-{
-	__be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL);
-	int flags = ROCKER_OP_FLAG_REMOVE;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return -EINVAL;
-
-	return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags);
-}
-
-static int rocker_port_obj_del(struct net_device *dev,
-			       const struct switchdev_obj *obj)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	const struct switchdev_obj_ipv4_fib *fib4;
-	int err = 0;
-
-	switch (obj->id) {
-	case SWITCHDEV_OBJ_ID_PORT_VLAN:
-		err = rocker_port_vlans_del(rocker_port,
-					    SWITCHDEV_OBJ_PORT_VLAN(obj));
-		break;
-	case SWITCHDEV_OBJ_ID_IPV4_FIB:
-		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
-		err = rocker_port_fib_ipv4(rocker_port, NULL,
-					   htonl(fib4->dst), fib4->dst_len,
-					   &fib4->fi, fib4->tb_id,
-					   ROCKER_OP_FLAG_REMOVE);
-		break;
-	case SWITCHDEV_OBJ_ID_PORT_FDB:
-		err = rocker_port_fdb_del(rocker_port, NULL,
-					  SWITCHDEV_OBJ_PORT_FDB(obj));
-		break;
-	default:
-		err = -EOPNOTSUPP;
-		break;
-	}
-
-	return err;
-}
-
-static int rocker_port_fdb_dump(const struct rocker_port *rocker_port,
-				struct switchdev_obj_port_fdb *fdb,
-				switchdev_obj_dump_cb_t *cb)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_fdb_tbl_entry *found;
-	struct hlist_node *tmp;
-	unsigned long lock_flags;
-	int bkt;
-	int err = 0;
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) {
-		if (found->key.rocker_port != rocker_port)
-			continue;
-		ether_addr_copy(fdb->addr, found->key.addr);
-		fdb->ndm_state = NUD_REACHABLE;
-		fdb->vid = rocker_port_vlan_to_vid(rocker_port,
-						   found->key.vlan_id);
-		err = cb(&fdb->obj);
-		if (err)
-			break;
-	}
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	return err;
-}
-
-static int rocker_port_vlan_dump(const struct rocker_port *rocker_port,
-				 struct switchdev_obj_port_vlan *vlan,
-				 switchdev_obj_dump_cb_t *cb)
-{
-	u16 vid;
-	int err = 0;
-
-	for (vid = 1; vid < VLAN_N_VID; vid++) {
-		if (!test_bit(vid, rocker_port->vlan_bitmap))
-			continue;
-		vlan->flags = 0;
-		if (rocker_vlan_id_is_internal(htons(vid)))
-			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
-		vlan->vid_begin = vlan->vid_end = vid;
-		err = cb(&vlan->obj);
-		if (err)
-			break;
-	}
-
-	return err;
-}
-
-static int rocker_port_obj_dump(struct net_device *dev,
-				struct switchdev_obj *obj,
-				switchdev_obj_dump_cb_t *cb)
-{
-	const struct rocker_port *rocker_port = netdev_priv(dev);
-	int err = 0;
-
-	switch (obj->id) {
-	case SWITCHDEV_OBJ_ID_PORT_FDB:
-		err = rocker_port_fdb_dump(rocker_port,
-					   SWITCHDEV_OBJ_PORT_FDB(obj), cb);
-		break;
-	case SWITCHDEV_OBJ_ID_PORT_VLAN:
-		err = rocker_port_vlan_dump(rocker_port,
-					    SWITCHDEV_OBJ_PORT_VLAN(obj), cb);
-		break;
-	default:
-		err = -EOPNOTSUPP;
-		break;
-	}
-
-	return err;
-}
-
-static const struct switchdev_ops rocker_port_switchdev_ops = {
-	.switchdev_port_attr_get	= rocker_port_attr_get,
-	.switchdev_port_attr_set	= rocker_port_attr_set,
-	.switchdev_port_obj_add		= rocker_port_obj_add,
-	.switchdev_port_obj_del		= rocker_port_obj_del,
-	.switchdev_port_obj_dump	= rocker_port_obj_dump,
-};
-
-/********************
- * ethtool interface
- ********************/
-
-static int rocker_port_get_settings(struct net_device *dev,
-				    struct ethtool_cmd *ecmd)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-
-	return rocker_cmd_get_port_settings_ethtool(rocker_port, ecmd);
-}
-
-static int rocker_port_set_settings(struct net_device *dev,
-				    struct ethtool_cmd *ecmd)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-
-	return rocker_cmd_set_port_settings_ethtool(rocker_port, ecmd);
-}
-
-static void rocker_port_get_drvinfo(struct net_device *dev,
-				    struct ethtool_drvinfo *drvinfo)
-{
-	strlcpy(drvinfo->driver, rocker_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version));
-}
-
-static struct rocker_port_stats {
-	char str[ETH_GSTRING_LEN];
-	int type;
-} rocker_port_stats[] = {
-	{ "rx_packets", ROCKER_TLV_CMD_PORT_STATS_RX_PKTS,    },
-	{ "rx_bytes",   ROCKER_TLV_CMD_PORT_STATS_RX_BYTES,   },
-	{ "rx_dropped", ROCKER_TLV_CMD_PORT_STATS_RX_DROPPED, },
-	{ "rx_errors",  ROCKER_TLV_CMD_PORT_STATS_RX_ERRORS,  },
-
-	{ "tx_packets", ROCKER_TLV_CMD_PORT_STATS_TX_PKTS,    },
-	{ "tx_bytes",   ROCKER_TLV_CMD_PORT_STATS_TX_BYTES,   },
-	{ "tx_dropped", ROCKER_TLV_CMD_PORT_STATS_TX_DROPPED, },
-	{ "tx_errors",  ROCKER_TLV_CMD_PORT_STATS_TX_ERRORS,  },
-};
-
-#define ROCKER_PORT_STATS_LEN  ARRAY_SIZE(rocker_port_stats)
-
-static void rocker_port_get_strings(struct net_device *netdev, u32 stringset,
-				    u8 *data)
-{
-	u8 *p = data;
-	int i;
-
-	switch (stringset) {
-	case ETH_SS_STATS:
-		for (i = 0; i < ARRAY_SIZE(rocker_port_stats); i++) {
-			memcpy(p, rocker_port_stats[i].str, ETH_GSTRING_LEN);
-			p += ETH_GSTRING_LEN;
-		}
-		break;
-	}
-}
-
-static int
-rocker_cmd_get_port_stats_prep(const struct rocker_port *rocker_port,
-			       struct rocker_desc_info *desc_info,
-			       void *priv)
-{
-	struct rocker_tlv *cmd_stats;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
-			       ROCKER_TLV_CMD_TYPE_GET_PORT_STATS))
-		return -EMSGSIZE;
-
-	cmd_stats = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_stats)
-		return -EMSGSIZE;
-
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_STATS_PPORT,
-			       rocker_port->pport))
-		return -EMSGSIZE;
-
-	rocker_tlv_nest_end(desc_info, cmd_stats);
-
-	return 0;
-}
-
-static int
-rocker_cmd_get_port_stats_ethtool_proc(const struct rocker_port *rocker_port,
-				       const struct rocker_desc_info *desc_info,
-				       void *priv)
-{
-	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
-	const struct rocker_tlv *stats_attrs[ROCKER_TLV_CMD_PORT_STATS_MAX + 1];
-	const struct rocker_tlv *pattr;
-	u32 pport;
-	u64 *data = priv;
-	int i;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
-
-	if (!attrs[ROCKER_TLV_CMD_INFO])
-		return -EIO;
-
-	rocker_tlv_parse_nested(stats_attrs, ROCKER_TLV_CMD_PORT_STATS_MAX,
-				attrs[ROCKER_TLV_CMD_INFO]);
-
-	if (!stats_attrs[ROCKER_TLV_CMD_PORT_STATS_PPORT])
-		return -EIO;
-
-	pport = rocker_tlv_get_u32(stats_attrs[ROCKER_TLV_CMD_PORT_STATS_PPORT]);
-	if (pport != rocker_port->pport)
-		return -EIO;
-
-	for (i = 0; i < ARRAY_SIZE(rocker_port_stats); i++) {
-		pattr = stats_attrs[rocker_port_stats[i].type];
-		if (!pattr)
-			continue;
-
-		data[i] = rocker_tlv_get_u64(pattr);
-	}
-
-	return 0;
-}
-
-static int rocker_cmd_get_port_stats_ethtool(struct rocker_port *rocker_port,
-					     void *priv)
-{
-	return rocker_cmd_exec(rocker_port, NULL, 0,
-			       rocker_cmd_get_port_stats_prep, NULL,
-			       rocker_cmd_get_port_stats_ethtool_proc,
-			       priv);
-}
-
-static void rocker_port_get_stats(struct net_device *dev,
-				  struct ethtool_stats *stats, u64 *data)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-
-	if (rocker_cmd_get_port_stats_ethtool(rocker_port, data) != 0) {
-		int i;
-
-		for (i = 0; i < ARRAY_SIZE(rocker_port_stats); ++i)
-			data[i] = 0;
-	}
-}
-
-static int rocker_port_get_sset_count(struct net_device *netdev, int sset)
-{
-	switch (sset) {
-	case ETH_SS_STATS:
-		return ROCKER_PORT_STATS_LEN;
-	default:
-		return -EOPNOTSUPP;
-	}
-}
-
-static const struct ethtool_ops rocker_port_ethtool_ops = {
-	.get_settings		= rocker_port_get_settings,
-	.set_settings		= rocker_port_set_settings,
-	.get_drvinfo		= rocker_port_get_drvinfo,
-	.get_link		= ethtool_op_get_link,
-	.get_strings		= rocker_port_get_strings,
-	.get_ethtool_stats	= rocker_port_get_stats,
-	.get_sset_count		= rocker_port_get_sset_count,
-};
-
-/*****************
- * NAPI interface
- *****************/
-
-static struct rocker_port *rocker_port_napi_tx_get(struct napi_struct *napi)
-{
-	return container_of(napi, struct rocker_port, napi_tx);
-}
-
-static int rocker_port_poll_tx(struct napi_struct *napi, int budget)
-{
-	struct rocker_port *rocker_port = rocker_port_napi_tx_get(napi);
-	const struct rocker *rocker = rocker_port->rocker;
-	const struct rocker_desc_info *desc_info;
-	u32 credits = 0;
-	int err;
-
-	/* Cleanup tx descriptors */
-	while ((desc_info = rocker_desc_tail_get(&rocker_port->tx_ring))) {
-		struct sk_buff *skb;
-
-		err = rocker_desc_err(desc_info);
-		if (err && net_ratelimit())
-			netdev_err(rocker_port->dev, "tx desc received with err %d\n",
-				   err);
-		rocker_tx_desc_frags_unmap(rocker_port, desc_info);
-
-		skb = rocker_desc_cookie_ptr_get(desc_info);
-		if (err == 0) {
-			rocker_port->dev->stats.tx_packets++;
-			rocker_port->dev->stats.tx_bytes += skb->len;
-		} else {
-			rocker_port->dev->stats.tx_errors++;
-		}
-
-		dev_kfree_skb_any(skb);
-		credits++;
-	}
-
-	if (credits && netif_queue_stopped(rocker_port->dev))
-		netif_wake_queue(rocker_port->dev);
-
-	napi_complete(napi);
-	rocker_dma_ring_credits_set(rocker, &rocker_port->tx_ring, credits);
-
-	return 0;
-}
-
-static int rocker_port_rx_proc(const struct rocker *rocker,
-			       const struct rocker_port *rocker_port,
-			       struct rocker_desc_info *desc_info)
-{
-	const struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1];
-	struct sk_buff *skb = rocker_desc_cookie_ptr_get(desc_info);
-	size_t rx_len;
-	u16 rx_flags = 0;
-
-	if (!skb)
-		return -ENOENT;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_RX_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_RX_FRAG_LEN])
-		return -EINVAL;
-	if (attrs[ROCKER_TLV_RX_FLAGS])
-		rx_flags = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FLAGS]);
-
-	rocker_dma_rx_ring_skb_unmap(rocker, attrs);
-
-	rx_len = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FRAG_LEN]);
-	skb_put(skb, rx_len);
-	skb->protocol = eth_type_trans(skb, rocker_port->dev);
-
-	if (rx_flags & ROCKER_RX_FLAGS_FWD_OFFLOAD)
-		skb->offload_fwd_mark = rocker_port->dev->offload_fwd_mark;
-
-	rocker_port->dev->stats.rx_packets++;
-	rocker_port->dev->stats.rx_bytes += skb->len;
-
-	netif_receive_skb(skb);
-
-	return rocker_dma_rx_ring_skb_alloc(rocker_port, desc_info);
-}
-
-static struct rocker_port *rocker_port_napi_rx_get(struct napi_struct *napi)
-{
-	return container_of(napi, struct rocker_port, napi_rx);
-}
-
-static int rocker_port_poll_rx(struct napi_struct *napi, int budget)
-{
-	struct rocker_port *rocker_port = rocker_port_napi_rx_get(napi);
-	const struct rocker *rocker = rocker_port->rocker;
-	struct rocker_desc_info *desc_info;
-	u32 credits = 0;
-	int err;
-
-	/* Process rx descriptors */
-	while (credits < budget &&
-	       (desc_info = rocker_desc_tail_get(&rocker_port->rx_ring))) {
-		err = rocker_desc_err(desc_info);
-		if (err) {
-			if (net_ratelimit())
-				netdev_err(rocker_port->dev, "rx desc received with err %d\n",
-					   err);
-		} else {
-			err = rocker_port_rx_proc(rocker, rocker_port,
-						  desc_info);
-			if (err && net_ratelimit())
-				netdev_err(rocker_port->dev, "rx processing failed with err %d\n",
-					   err);
-		}
-		if (err)
-			rocker_port->dev->stats.rx_errors++;
-
-		rocker_desc_gen_clear(desc_info);
-		rocker_desc_head_set(rocker, &rocker_port->rx_ring, desc_info);
-		credits++;
-	}
-
-	if (credits < budget)
-		napi_complete(napi);
-
-	rocker_dma_ring_credits_set(rocker, &rocker_port->rx_ring, credits);
-
-	return credits;
-}
-
-/*****************
- * PCI driver ops
- *****************/
-
-static void rocker_carrier_init(const struct rocker_port *rocker_port)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	u64 link_status = rocker_read64(rocker, PORT_PHYS_LINK_STATUS);
-	bool link_up;
-
-	link_up = link_status & (1 << rocker_port->pport);
-	if (link_up)
-		netif_carrier_on(rocker_port->dev);
-	else
-		netif_carrier_off(rocker_port->dev);
-}
-
-static void rocker_remove_ports(const struct rocker *rocker)
-{
-	struct rocker_port *rocker_port;
-	int i;
-
-	for (i = 0; i < rocker->port_count; i++) {
-		rocker_port = rocker->ports[i];
-		if (!rocker_port)
-			continue;
-		rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
-		unregister_netdev(rocker_port->dev);
-		free_netdev(rocker_port->dev);
-	}
-	kfree(rocker->ports);
-}
-
-static void rocker_port_dev_addr_init(struct rocker_port *rocker_port)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	const struct pci_dev *pdev = rocker->pdev;
-	int err;
-
-	err = rocker_cmd_get_port_settings_macaddr(rocker_port,
-						   rocker_port->dev->dev_addr);
-	if (err) {
-		dev_warn(&pdev->dev, "failed to get mac address, using random\n");
-		eth_hw_addr_random(rocker_port->dev);
-	}
-}
-
-static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
-{
-	const struct pci_dev *pdev = rocker->pdev;
-	struct rocker_port *rocker_port;
-	struct net_device *dev;
-	u16 untagged_vid = 0;
-	int err;
-
-	dev = alloc_etherdev(sizeof(struct rocker_port));
-	if (!dev)
-		return -ENOMEM;
-	rocker_port = netdev_priv(dev);
-	rocker_port->dev = dev;
-	rocker_port->rocker = rocker;
-	rocker_port->port_number = port_number;
-	rocker_port->pport = port_number + 1;
-	rocker_port->brport_flags = BR_LEARNING | BR_LEARNING_SYNC;
-	rocker_port->ageing_time = BR_DEFAULT_AGEING_TIME;
-
-	rocker_port_dev_addr_init(rocker_port);
-	dev->netdev_ops = &rocker_port_netdev_ops;
-	dev->ethtool_ops = &rocker_port_ethtool_ops;
-	dev->switchdev_ops = &rocker_port_switchdev_ops;
-	netif_tx_napi_add(dev, &rocker_port->napi_tx, rocker_port_poll_tx,
-		       NAPI_POLL_WEIGHT);
-	netif_napi_add(dev, &rocker_port->napi_rx, rocker_port_poll_rx,
-		       NAPI_POLL_WEIGHT);
-	rocker_carrier_init(rocker_port);
-
-	dev->features |= NETIF_F_NETNS_LOCAL | NETIF_F_SG;
-
-	err = register_netdev(dev);
-	if (err) {
-		dev_err(&pdev->dev, "register_netdev failed\n");
-		goto err_register_netdev;
-	}
-	rocker->ports[port_number] = rocker_port;
-
-	switchdev_port_fwd_mark_set(rocker_port->dev, NULL, false);
-
-	rocker_port_set_learning(rocker_port, NULL);
-
-	err = rocker_port_ig_tbl(rocker_port, NULL, 0);
-	if (err) {
-		netdev_err(rocker_port->dev, "install ig port table failed\n");
-		goto err_port_ig_tbl;
-	}
-
-	rocker_port->internal_vlan_id =
-		rocker_port_internal_vlan_id_get(rocker_port, dev->ifindex);
-
-	err = rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
-	if (err) {
-		netdev_err(rocker_port->dev, "install untagged VLAN failed\n");
-		goto err_untagged_vlan;
-	}
-
-	return 0;
-
-err_untagged_vlan:
-	rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
-err_port_ig_tbl:
-	rocker->ports[port_number] = NULL;
-	unregister_netdev(dev);
-err_register_netdev:
-	free_netdev(dev);
-	return err;
-}
-
-static int rocker_probe_ports(struct rocker *rocker)
-{
-	int i;
-	size_t alloc_size;
-	int err;
-
-	alloc_size = sizeof(struct rocker_port *) * rocker->port_count;
-	rocker->ports = kzalloc(alloc_size, GFP_KERNEL);
-	if (!rocker->ports)
-		return -ENOMEM;
-	for (i = 0; i < rocker->port_count; i++) {
-		err = rocker_probe_port(rocker, i);
-		if (err)
-			goto remove_ports;
-	}
-	return 0;
-
-remove_ports:
-	rocker_remove_ports(rocker);
-	return err;
-}
-
-static int rocker_msix_init(struct rocker *rocker)
-{
-	struct pci_dev *pdev = rocker->pdev;
-	int msix_entries;
-	int i;
-	int err;
-
-	msix_entries = pci_msix_vec_count(pdev);
-	if (msix_entries < 0)
-		return msix_entries;
-
-	if (msix_entries != ROCKER_MSIX_VEC_COUNT(rocker->port_count))
-		return -EINVAL;
-
-	rocker->msix_entries = kmalloc_array(msix_entries,
-					     sizeof(struct msix_entry),
-					     GFP_KERNEL);
-	if (!rocker->msix_entries)
-		return -ENOMEM;
-
-	for (i = 0; i < msix_entries; i++)
-		rocker->msix_entries[i].entry = i;
-
-	err = pci_enable_msix_exact(pdev, rocker->msix_entries, msix_entries);
-	if (err < 0)
-		goto err_enable_msix;
-
-	return 0;
-
-err_enable_msix:
-	kfree(rocker->msix_entries);
-	return err;
-}
-
-static void rocker_msix_fini(const struct rocker *rocker)
-{
-	pci_disable_msix(rocker->pdev);
-	kfree(rocker->msix_entries);
-}
-
-static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id)
-{
-	struct rocker *rocker;
-	int err;
-
-	rocker = kzalloc(sizeof(*rocker), GFP_KERNEL);
-	if (!rocker)
-		return -ENOMEM;
-
-	err = pci_enable_device(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "pci_enable_device failed\n");
-		goto err_pci_enable_device;
-	}
-
-	err = pci_request_regions(pdev, rocker_driver_name);
-	if (err) {
-		dev_err(&pdev->dev, "pci_request_regions failed\n");
-		goto err_pci_request_regions;
-	}
-
-	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
-	if (!err) {
-		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
-		if (err) {
-			dev_err(&pdev->dev, "pci_set_consistent_dma_mask failed\n");
-			goto err_pci_set_dma_mask;
-		}
-	} else {
-		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
-		if (err) {
-			dev_err(&pdev->dev, "pci_set_dma_mask failed\n");
-			goto err_pci_set_dma_mask;
-		}
-	}
-
-	if (pci_resource_len(pdev, 0) < ROCKER_PCI_BAR0_SIZE) {
-		dev_err(&pdev->dev, "invalid PCI region size\n");
-		err = -EINVAL;
-		goto err_pci_resource_len_check;
-	}
-
-	rocker->hw_addr = ioremap(pci_resource_start(pdev, 0),
-				  pci_resource_len(pdev, 0));
-	if (!rocker->hw_addr) {
-		dev_err(&pdev->dev, "ioremap failed\n");
-		err = -EIO;
-		goto err_ioremap;
-	}
-	pci_set_master(pdev);
-
-	rocker->pdev = pdev;
-	pci_set_drvdata(pdev, rocker);
-
-	rocker->port_count = rocker_read32(rocker, PORT_PHYS_COUNT);
-
-	err = rocker_msix_init(rocker);
-	if (err) {
-		dev_err(&pdev->dev, "MSI-X init failed\n");
-		goto err_msix_init;
-	}
-
-	err = rocker_basic_hw_test(rocker);
-	if (err) {
-		dev_err(&pdev->dev, "basic hw test failed\n");
-		goto err_basic_hw_test;
-	}
-
-	rocker_write32(rocker, CONTROL, ROCKER_CONTROL_RESET);
-
-	err = rocker_dma_rings_init(rocker);
-	if (err)
-		goto err_dma_rings_init;
-
-	err = request_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD),
-			  rocker_cmd_irq_handler, 0,
-			  rocker_driver_name, rocker);
-	if (err) {
-		dev_err(&pdev->dev, "cannot assign cmd irq\n");
-		goto err_request_cmd_irq;
-	}
-
-	err = request_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT),
-			  rocker_event_irq_handler, 0,
-			  rocker_driver_name, rocker);
-	if (err) {
-		dev_err(&pdev->dev, "cannot assign event irq\n");
-		goto err_request_event_irq;
-	}
-
-	rocker->hw.id = rocker_read64(rocker, SWITCH_ID);
-
-	err = rocker_init_tbls(rocker);
-	if (err) {
-		dev_err(&pdev->dev, "cannot init rocker tables\n");
-		goto err_init_tbls;
-	}
-
-	setup_timer(&rocker->fdb_cleanup_timer, rocker_fdb_cleanup,
-		    (unsigned long) rocker);
-	mod_timer(&rocker->fdb_cleanup_timer, jiffies);
-
-	err = rocker_probe_ports(rocker);
-	if (err) {
-		dev_err(&pdev->dev, "failed to probe ports\n");
-		goto err_probe_ports;
-	}
-
-	dev_info(&pdev->dev, "Rocker switch with id %*phN\n",
-		 (int)sizeof(rocker->hw.id), &rocker->hw.id);
-
-	return 0;
-
-err_probe_ports:
-	del_timer_sync(&rocker->fdb_cleanup_timer);
-	rocker_free_tbls(rocker);
-err_init_tbls:
-	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT), rocker);
-err_request_event_irq:
-	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD), rocker);
-err_request_cmd_irq:
-	rocker_dma_rings_fini(rocker);
-err_dma_rings_init:
-err_basic_hw_test:
-	rocker_msix_fini(rocker);
-err_msix_init:
-	iounmap(rocker->hw_addr);
-err_ioremap:
-err_pci_resource_len_check:
-err_pci_set_dma_mask:
-	pci_release_regions(pdev);
-err_pci_request_regions:
-	pci_disable_device(pdev);
-err_pci_enable_device:
-	kfree(rocker);
-	return err;
-}
-
-static void rocker_remove(struct pci_dev *pdev)
-{
-	struct rocker *rocker = pci_get_drvdata(pdev);
-
-	del_timer_sync(&rocker->fdb_cleanup_timer);
-	rocker_free_tbls(rocker);
-	rocker_write32(rocker, CONTROL, ROCKER_CONTROL_RESET);
-	rocker_remove_ports(rocker);
-	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT), rocker);
-	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD), rocker);
-	rocker_dma_rings_fini(rocker);
-	rocker_msix_fini(rocker);
-	iounmap(rocker->hw_addr);
-	pci_release_regions(rocker->pdev);
-	pci_disable_device(rocker->pdev);
-	kfree(rocker);
-}
-
-static struct pci_driver rocker_pci_driver = {
-	.name		= rocker_driver_name,
-	.id_table	= rocker_pci_id_table,
-	.probe		= rocker_probe,
-	.remove		= rocker_remove,
-};
-
-/************************************
- * Net device notifier event handler
- ************************************/
-
-static bool rocker_port_dev_check(const struct net_device *dev)
-{
-	return dev->netdev_ops == &rocker_port_netdev_ops;
-}
-
-static int rocker_port_bridge_join(struct rocker_port *rocker_port,
-				   struct net_device *bridge)
-{
-	u16 untagged_vid = 0;
-	int err;
-
-	/* Port is joining bridge, so the internal VLAN for the
-	 * port is going to change to the bridge internal VLAN.
-	 * Let's remove untagged VLAN (vid=0) from port and
-	 * re-add once internal VLAN has changed.
-	 */
-
-	err = rocker_port_vlan_del(rocker_port, untagged_vid, 0);
-	if (err)
-		return err;
-
-	rocker_port_internal_vlan_id_put(rocker_port,
-					 rocker_port->dev->ifindex);
-	rocker_port->internal_vlan_id =
-		rocker_port_internal_vlan_id_get(rocker_port, bridge->ifindex);
-
-	rocker_port->bridge_dev = bridge;
-	switchdev_port_fwd_mark_set(rocker_port->dev, bridge, true);
-
-	return rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
-}
-
-static int rocker_port_bridge_leave(struct rocker_port *rocker_port)
-{
-	u16 untagged_vid = 0;
-	int err;
-
-	err = rocker_port_vlan_del(rocker_port, untagged_vid, 0);
-	if (err)
-		return err;
-
-	rocker_port_internal_vlan_id_put(rocker_port,
-					 rocker_port->bridge_dev->ifindex);
-	rocker_port->internal_vlan_id =
-		rocker_port_internal_vlan_id_get(rocker_port,
-						 rocker_port->dev->ifindex);
-
-	switchdev_port_fwd_mark_set(rocker_port->dev, rocker_port->bridge_dev,
-				    false);
-	rocker_port->bridge_dev = NULL;
-
-	err = rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
-	if (err)
-		return err;
-
-	if (rocker_port->dev->flags & IFF_UP)
-		err = rocker_port_fwd_enable(rocker_port, NULL, 0);
-
-	return err;
-}
-
-
-static int rocker_port_ovs_changed(struct rocker_port *rocker_port,
-				   struct net_device *master)
-{
-	int err;
-
-	rocker_port->bridge_dev = master;
-
-	err = rocker_port_fwd_disable(rocker_port, NULL, 0);
-	if (err)
-		return err;
-	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
-
-	return err;
-}
-
-static int rocker_port_master_linked(struct rocker_port *rocker_port,
-				     struct net_device *master)
-{
-	int err = 0;
-
-	if (netif_is_bridge_master(master))
-		err = rocker_port_bridge_join(rocker_port, master);
-	else if (netif_is_ovs_master(master))
-		err = rocker_port_ovs_changed(rocker_port, master);
-	return err;
-}
-
-static int rocker_port_master_unlinked(struct rocker_port *rocker_port)
-{
-	int err = 0;
-
-	if (rocker_port_is_bridged(rocker_port))
-		err = rocker_port_bridge_leave(rocker_port);
-	else if (rocker_port_is_ovsed(rocker_port))
-		err = rocker_port_ovs_changed(rocker_port, NULL);
-	return err;
-}
-
-static int rocker_netdevice_event(struct notifier_block *unused,
-				  unsigned long event, void *ptr)
-{
-	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
-	struct netdev_notifier_changeupper_info *info;
-	struct rocker_port *rocker_port;
-	int err;
-
-	if (!rocker_port_dev_check(dev))
-		return NOTIFY_DONE;
-
-	switch (event) {
-	case NETDEV_CHANGEUPPER:
-		info = ptr;
-		if (!info->master)
-			goto out;
-		rocker_port = netdev_priv(dev);
-		if (info->linking) {
-			err = rocker_port_master_linked(rocker_port,
-							info->upper_dev);
-			if (err)
-				netdev_warn(dev, "failed to reflect master linked (err %d)\n",
-					    err);
-		} else {
-			err = rocker_port_master_unlinked(rocker_port);
-			if (err)
-				netdev_warn(dev, "failed to reflect master unlinked (err %d)\n",
-					    err);
-		}
-		break;
-	}
-out:
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block rocker_netdevice_nb __read_mostly = {
-	.notifier_call = rocker_netdevice_event,
-};
-
-/************************************
- * Net event notifier event handler
- ************************************/
-
-static int rocker_neigh_update(struct net_device *dev, struct neighbour *n)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int flags = (n->nud_state & NUD_VALID ? 0 : ROCKER_OP_FLAG_REMOVE) |
-		    ROCKER_OP_FLAG_NOWAIT;
-	__be32 ip_addr = *(__be32 *)n->primary_key;
-
-	return rocker_port_ipv4_neigh(rocker_port, NULL, flags, ip_addr, n->ha);
-}
-
-static int rocker_netevent_event(struct notifier_block *unused,
-				 unsigned long event, void *ptr)
-{
-	struct net_device *dev;
-	struct neighbour *n = ptr;
-	int err;
-
-	switch (event) {
-	case NETEVENT_NEIGH_UPDATE:
-		if (n->tbl != &arp_tbl)
-			return NOTIFY_DONE;
-		dev = n->dev;
-		if (!rocker_port_dev_check(dev))
-			return NOTIFY_DONE;
-		err = rocker_neigh_update(dev, n);
-		if (err)
-			netdev_warn(dev,
-				    "failed to handle neigh update (err %d)\n",
-				    err);
-		break;
-	}
-
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block rocker_netevent_nb __read_mostly = {
-	.notifier_call = rocker_netevent_event,
-};
-
-/***********************
- * Module init and exit
- ***********************/
-
-static int __init rocker_module_init(void)
-{
-	int err;
-
-	register_netdevice_notifier(&rocker_netdevice_nb);
-	register_netevent_notifier(&rocker_netevent_nb);
-	err = pci_register_driver(&rocker_pci_driver);
-	if (err)
-		goto err_pci_register_driver;
-	return 0;
-
-err_pci_register_driver:
-	unregister_netevent_notifier(&rocker_netevent_nb);
-	unregister_netdevice_notifier(&rocker_netdevice_nb);
-	return err;
-}
-
-static void __exit rocker_module_exit(void)
-{
-	unregister_netevent_notifier(&rocker_netevent_nb);
-	unregister_netdevice_notifier(&rocker_netdevice_nb);
-	pci_unregister_driver(&rocker_pci_driver);
-}
-
-module_init(rocker_module_init);
-module_exit(rocker_module_exit);
-
-MODULE_LICENSE("GPL v2");
-MODULE_AUTHOR("Jiri Pirko <jiri@resnulli.us>");
-MODULE_AUTHOR("Scott Feldman <sfeldma@gmail.com>");
-MODULE_DESCRIPTION("Rocker switch device driver");
-MODULE_DEVICE_TABLE(pci, rocker_pci_id_table);
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
new file mode 100644
index 0000000..44fa9ac
--- /dev/null
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -0,0 +1,5493 @@
+/*
+ * drivers/net/ethernet/rocker/rocker.c - Rocker switch device driver
+ * Copyright (c) 2014-2016 Jiri Pirko <jiri@mellanox.com>
+ * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <linux/wait.h>
+#include <linux/spinlock.h>
+#include <linux/hashtable.h>
+#include <linux/crc32.h>
+#include <linux/sort.h>
+#include <linux/random.h>
+#include <linux/netdevice.h>
+#include <linux/inetdevice.h>
+#include <linux/skbuff.h>
+#include <linux/socket.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/if_bridge.h>
+#include <linux/bitops.h>
+#include <linux/ctype.h>
+#include <net/switchdev.h>
+#include <net/rtnetlink.h>
+#include <net/ip_fib.h>
+#include <net/netevent.h>
+#include <net/arp.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+#include <generated/utsrelease.h>
+
+#include "rocker_hw.h"
+
+static const char rocker_driver_name[] = "rocker";
+
+static const struct pci_device_id rocker_pci_id_table[] = {
+	{PCI_VDEVICE(REDHAT, PCI_DEVICE_ID_REDHAT_ROCKER), 0},
+	{0, }
+};
+
+struct rocker_flow_tbl_key {
+	u32 priority;
+	enum rocker_of_dpa_table_id tbl_id;
+	union {
+		struct {
+			u32 in_pport;
+			u32 in_pport_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+		} ig_port;
+		struct {
+			u32 in_pport;
+			__be16 vlan_id;
+			__be16 vlan_id_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+			bool untagged;
+			__be16 new_vlan_id;
+		} vlan;
+		struct {
+			u32 in_pport;
+			u32 in_pport_mask;
+			__be16 eth_type;
+			u8 eth_dst[ETH_ALEN];
+			u8 eth_dst_mask[ETH_ALEN];
+			__be16 vlan_id;
+			__be16 vlan_id_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+			bool copy_to_cpu;
+		} term_mac;
+		struct {
+			__be16 eth_type;
+			__be32 dst4;
+			__be32 dst4_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+			u32 group_id;
+		} ucast_routing;
+		struct {
+			u8 eth_dst[ETH_ALEN];
+			u8 eth_dst_mask[ETH_ALEN];
+			int has_eth_dst;
+			int has_eth_dst_mask;
+			__be16 vlan_id;
+			u32 tunnel_id;
+			enum rocker_of_dpa_table_id goto_tbl;
+			u32 group_id;
+			bool copy_to_cpu;
+		} bridge;
+		struct {
+			u32 in_pport;
+			u32 in_pport_mask;
+			u8 eth_src[ETH_ALEN];
+			u8 eth_src_mask[ETH_ALEN];
+			u8 eth_dst[ETH_ALEN];
+			u8 eth_dst_mask[ETH_ALEN];
+			__be16 eth_type;
+			__be16 vlan_id;
+			__be16 vlan_id_mask;
+			u8 ip_proto;
+			u8 ip_proto_mask;
+			u8 ip_tos;
+			u8 ip_tos_mask;
+			u32 group_id;
+		} acl;
+	};
+};
+
+struct rocker_flow_tbl_entry {
+	struct hlist_node entry;
+	u32 cmd;
+	u64 cookie;
+	struct rocker_flow_tbl_key key;
+	size_t key_len;
+	u32 key_crc32; /* key */
+};
+
+struct rocker_group_tbl_entry {
+	struct hlist_node entry;
+	u32 cmd;
+	u32 group_id; /* key */
+	u16 group_count;
+	u32 *group_ids;
+	union {
+		struct {
+			u8 pop_vlan;
+		} l2_interface;
+		struct {
+			u8 eth_src[ETH_ALEN];
+			u8 eth_dst[ETH_ALEN];
+			__be16 vlan_id;
+			u32 group_id;
+		} l2_rewrite;
+		struct {
+			u8 eth_src[ETH_ALEN];
+			u8 eth_dst[ETH_ALEN];
+			__be16 vlan_id;
+			bool ttl_check;
+			u32 group_id;
+		} l3_unicast;
+	};
+};
+
+struct rocker_fdb_tbl_entry {
+	struct hlist_node entry;
+	u32 key_crc32; /* key */
+	bool learned;
+	unsigned long touched;
+	struct rocker_fdb_tbl_key {
+		struct rocker_port *rocker_port;
+		u8 addr[ETH_ALEN];
+		__be16 vlan_id;
+	} key;
+};
+
+struct rocker_internal_vlan_tbl_entry {
+	struct hlist_node entry;
+	int ifindex; /* key */
+	u32 ref_count;
+	__be16 vlan_id;
+};
+
+struct rocker_neigh_tbl_entry {
+	struct hlist_node entry;
+	__be32 ip_addr; /* key */
+	struct net_device *dev;
+	u32 ref_count;
+	u32 index;
+	u8 eth_dst[ETH_ALEN];
+	bool ttl_check;
+};
+
+struct rocker_desc_info {
+	char *data; /* mapped */
+	size_t data_size;
+	size_t tlv_size;
+	struct rocker_desc *desc;
+	dma_addr_t mapaddr;
+};
+
+struct rocker_dma_ring_info {
+	size_t size;
+	u32 head;
+	u32 tail;
+	struct rocker_desc *desc; /* mapped */
+	dma_addr_t mapaddr;
+	struct rocker_desc_info *desc_info;
+	unsigned int type;
+};
+
+struct rocker;
+
+enum {
+	ROCKER_CTRL_LINK_LOCAL_MCAST,
+	ROCKER_CTRL_LOCAL_ARP,
+	ROCKER_CTRL_IPV4_MCAST,
+	ROCKER_CTRL_IPV6_MCAST,
+	ROCKER_CTRL_DFLT_BRIDGING,
+	ROCKER_CTRL_DFLT_OVS,
+	ROCKER_CTRL_MAX,
+};
+
+#define ROCKER_INTERNAL_VLAN_ID_BASE	0x0f00
+#define ROCKER_N_INTERNAL_VLANS		255
+#define ROCKER_VLAN_BITMAP_LEN		BITS_TO_LONGS(VLAN_N_VID)
+#define ROCKER_INTERNAL_VLAN_BITMAP_LEN	BITS_TO_LONGS(ROCKER_N_INTERNAL_VLANS)
+
+struct rocker_port {
+	struct net_device *dev;
+	struct net_device *bridge_dev;
+	struct rocker *rocker;
+	unsigned int port_number;
+	u32 pport;
+	__be16 internal_vlan_id;
+	int stp_state;
+	u32 brport_flags;
+	unsigned long ageing_time;
+	bool ctrls[ROCKER_CTRL_MAX];
+	unsigned long vlan_bitmap[ROCKER_VLAN_BITMAP_LEN];
+	struct napi_struct napi_tx;
+	struct napi_struct napi_rx;
+	struct rocker_dma_ring_info tx_ring;
+	struct rocker_dma_ring_info rx_ring;
+};
+
+struct rocker {
+	struct pci_dev *pdev;
+	u8 __iomem *hw_addr;
+	struct msix_entry *msix_entries;
+	unsigned int port_count;
+	struct rocker_port **ports;
+	struct {
+		u64 id;
+	} hw;
+	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
+	struct rocker_dma_ring_info cmd_ring;
+	struct rocker_dma_ring_info event_ring;
+	DECLARE_HASHTABLE(flow_tbl, 16);
+	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
+	u64 flow_tbl_next_cookie;
+	DECLARE_HASHTABLE(group_tbl, 16);
+	spinlock_t group_tbl_lock;		/* for group tbl accesses */
+	struct timer_list fdb_cleanup_timer;
+	DECLARE_HASHTABLE(fdb_tbl, 16);
+	spinlock_t fdb_tbl_lock;		/* for fdb tbl accesses */
+	unsigned long internal_vlan_bitmap[ROCKER_INTERNAL_VLAN_BITMAP_LEN];
+	DECLARE_HASHTABLE(internal_vlan_tbl, 8);
+	spinlock_t internal_vlan_tbl_lock;	/* for vlan tbl accesses */
+	DECLARE_HASHTABLE(neigh_tbl, 16);
+	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
+	u32 neigh_tbl_next_index;
+};
+
+static const u8 zero_mac[ETH_ALEN]   = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
+static const u8 ff_mac[ETH_ALEN]     = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
+static const u8 ll_mac[ETH_ALEN]     = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
+static const u8 ll_mask[ETH_ALEN]    = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xf0 };
+static const u8 mcast_mac[ETH_ALEN]  = { 0x01, 0x00, 0x00, 0x00, 0x00, 0x00 };
+static const u8 ipv4_mcast[ETH_ALEN] = { 0x01, 0x00, 0x5e, 0x00, 0x00, 0x00 };
+static const u8 ipv4_mask[ETH_ALEN]  = { 0xff, 0xff, 0xff, 0x80, 0x00, 0x00 };
+static const u8 ipv6_mcast[ETH_ALEN] = { 0x33, 0x33, 0x00, 0x00, 0x00, 0x00 };
+static const u8 ipv6_mask[ETH_ALEN]  = { 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 };
+
+/* Rocker priority levels for flow table entries.  Higher
+ * priority match takes precedence over lower priority match.
+ */
+
+enum {
+	ROCKER_PRIORITY_UNKNOWN = 0,
+	ROCKER_PRIORITY_IG_PORT = 1,
+	ROCKER_PRIORITY_VLAN = 1,
+	ROCKER_PRIORITY_TERM_MAC_UCAST = 0,
+	ROCKER_PRIORITY_TERM_MAC_MCAST = 1,
+	ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_EXACT = 1,
+	ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_WILD = 2,
+	ROCKER_PRIORITY_BRIDGING_VLAN = 3,
+	ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_EXACT = 1,
+	ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_WILD = 2,
+	ROCKER_PRIORITY_BRIDGING_TENANT = 3,
+	ROCKER_PRIORITY_ACL_CTRL = 3,
+	ROCKER_PRIORITY_ACL_NORMAL = 2,
+	ROCKER_PRIORITY_ACL_DFLT = 1,
+};
+
+static bool rocker_vlan_id_is_internal(__be16 vlan_id)
+{
+	u16 start = ROCKER_INTERNAL_VLAN_ID_BASE;
+	u16 end = 0xffe;
+	u16 _vlan_id = ntohs(vlan_id);
+
+	return (_vlan_id >= start && _vlan_id <= end);
+}
+
+static __be16 rocker_port_vid_to_vlan(const struct rocker_port *rocker_port,
+				      u16 vid, bool *pop_vlan)
+{
+	__be16 vlan_id;
+
+	if (pop_vlan)
+		*pop_vlan = false;
+	vlan_id = htons(vid);
+	if (!vlan_id) {
+		vlan_id = rocker_port->internal_vlan_id;
+		if (pop_vlan)
+			*pop_vlan = true;
+	}
+
+	return vlan_id;
+}
+
+static u16 rocker_port_vlan_to_vid(const struct rocker_port *rocker_port,
+				   __be16 vlan_id)
+{
+	if (rocker_vlan_id_is_internal(vlan_id))
+		return 0;
+
+	return ntohs(vlan_id);
+}
+
+static bool rocker_port_is_bridged(const struct rocker_port *rocker_port)
+{
+	return rocker_port->bridge_dev &&
+	       netif_is_bridge_master(rocker_port->bridge_dev);
+}
+
+static bool rocker_port_is_ovsed(const struct rocker_port *rocker_port)
+{
+	return rocker_port->bridge_dev &&
+	       netif_is_ovs_master(rocker_port->bridge_dev);
+}
+
+#define ROCKER_OP_FLAG_REMOVE		BIT(0)
+#define ROCKER_OP_FLAG_NOWAIT		BIT(1)
+#define ROCKER_OP_FLAG_LEARNED		BIT(2)
+#define ROCKER_OP_FLAG_REFRESH		BIT(3)
+
+static void *__rocker_mem_alloc(struct switchdev_trans *trans, int flags,
+				size_t size)
+{
+	struct switchdev_trans_item *elem = NULL;
+	gfp_t gfp_flags = (flags & ROCKER_OP_FLAG_NOWAIT) ?
+			  GFP_ATOMIC : GFP_KERNEL;
+
+	/* If in transaction prepare phase, allocate the memory
+	 * and enqueue it on a transaction.  If in transaction
+	 * commit phase, dequeue the memory from the transaction
+	 * rather than re-allocating the memory.  The idea is the
+	 * driver code paths for prepare and commit are identical
+	 * so the memory allocated in the prepare phase is the
+	 * memory used in the commit phase.
+	 */
+
+	if (!trans) {
+		elem = kzalloc(size + sizeof(*elem), gfp_flags);
+	} else if (switchdev_trans_ph_prepare(trans)) {
+		elem = kzalloc(size + sizeof(*elem), gfp_flags);
+		if (!elem)
+			return NULL;
+		switchdev_trans_item_enqueue(trans, elem, kfree, elem);
+	} else {
+		elem = switchdev_trans_item_dequeue(trans);
+	}
+
+	return elem ? elem + 1 : NULL;
+}
+
+static void *rocker_kzalloc(struct switchdev_trans *trans, int flags,
+			    size_t size)
+{
+	return __rocker_mem_alloc(trans, flags, size);
+}
+
+static void *rocker_kcalloc(struct switchdev_trans *trans, int flags,
+			    size_t n, size_t size)
+{
+	return __rocker_mem_alloc(trans, flags, n * size);
+}
+
+static void rocker_kfree(struct switchdev_trans *trans, const void *mem)
+{
+	struct switchdev_trans_item *elem;
+
+	/* Frees are ignored if in transaction prepare phase.  The
+	 * memory remains on the per-port list until freed in the
+	 * commit phase.
+	 */
+
+	if (switchdev_trans_ph_prepare(trans))
+		return;
+
+	elem = (struct switchdev_trans_item *) mem - 1;
+	kfree(elem);
+}
+
+struct rocker_wait {
+	wait_queue_head_t wait;
+	bool done;
+	bool nowait;
+};
+
+static void rocker_wait_reset(struct rocker_wait *wait)
+{
+	wait->done = false;
+	wait->nowait = false;
+}
+
+static void rocker_wait_init(struct rocker_wait *wait)
+{
+	init_waitqueue_head(&wait->wait);
+	rocker_wait_reset(wait);
+}
+
+static struct rocker_wait *rocker_wait_create(struct rocker_port *rocker_port,
+					      struct switchdev_trans *trans,
+					      int flags)
+{
+	struct rocker_wait *wait;
+
+	wait = rocker_kzalloc(trans, flags, sizeof(*wait));
+	if (!wait)
+		return NULL;
+	rocker_wait_init(wait);
+	return wait;
+}
+
+static void rocker_wait_destroy(struct switchdev_trans *trans,
+				struct rocker_wait *wait)
+{
+	rocker_kfree(trans, wait);
+}
+
+static bool rocker_wait_event_timeout(struct rocker_wait *wait,
+				      unsigned long timeout)
+{
+	wait_event_timeout(wait->wait, wait->done, HZ / 10);
+	if (!wait->done)
+		return false;
+	return true;
+}
+
+static void rocker_wait_wake_up(struct rocker_wait *wait)
+{
+	wait->done = true;
+	wake_up(&wait->wait);
+}
+
+static u32 rocker_msix_vector(const struct rocker *rocker, unsigned int vector)
+{
+	return rocker->msix_entries[vector].vector;
+}
+
+static u32 rocker_msix_tx_vector(const struct rocker_port *rocker_port)
+{
+	return rocker_msix_vector(rocker_port->rocker,
+				  ROCKER_MSIX_VEC_TX(rocker_port->port_number));
+}
+
+static u32 rocker_msix_rx_vector(const struct rocker_port *rocker_port)
+{
+	return rocker_msix_vector(rocker_port->rocker,
+				  ROCKER_MSIX_VEC_RX(rocker_port->port_number));
+}
+
+#define rocker_write32(rocker, reg, val)	\
+	writel((val), (rocker)->hw_addr + (ROCKER_ ## reg))
+#define rocker_read32(rocker, reg)	\
+	readl((rocker)->hw_addr + (ROCKER_ ## reg))
+#define rocker_write64(rocker, reg, val)	\
+	writeq((val), (rocker)->hw_addr + (ROCKER_ ## reg))
+#define rocker_read64(rocker, reg)	\
+	readq((rocker)->hw_addr + (ROCKER_ ## reg))
+
+/*****************************
+ * HW basic testing functions
+ *****************************/
+
+static int rocker_reg_test(const struct rocker *rocker)
+{
+	const struct pci_dev *pdev = rocker->pdev;
+	u64 test_reg;
+	u64 rnd;
+
+	rnd = prandom_u32();
+	rnd >>= 1;
+	rocker_write32(rocker, TEST_REG, rnd);
+	test_reg = rocker_read32(rocker, TEST_REG);
+	if (test_reg != rnd * 2) {
+		dev_err(&pdev->dev, "unexpected 32bit register value %08llx, expected %08llx\n",
+			test_reg, rnd * 2);
+		return -EIO;
+	}
+
+	rnd = prandom_u32();
+	rnd <<= 31;
+	rnd |= prandom_u32();
+	rocker_write64(rocker, TEST_REG64, rnd);
+	test_reg = rocker_read64(rocker, TEST_REG64);
+	if (test_reg != rnd * 2) {
+		dev_err(&pdev->dev, "unexpected 64bit register value %16llx, expected %16llx\n",
+			test_reg, rnd * 2);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int rocker_dma_test_one(const struct rocker *rocker,
+			       struct rocker_wait *wait, u32 test_type,
+			       dma_addr_t dma_handle, const unsigned char *buf,
+			       const unsigned char *expect, size_t size)
+{
+	const struct pci_dev *pdev = rocker->pdev;
+	int i;
+
+	rocker_wait_reset(wait);
+	rocker_write32(rocker, TEST_DMA_CTRL, test_type);
+
+	if (!rocker_wait_event_timeout(wait, HZ / 10)) {
+		dev_err(&pdev->dev, "no interrupt received within a timeout\n");
+		return -EIO;
+	}
+
+	for (i = 0; i < size; i++) {
+		if (buf[i] != expect[i]) {
+			dev_err(&pdev->dev, "unexpected memory content %02x at byte %x\n, %02x expected",
+				buf[i], i, expect[i]);
+			return -EIO;
+		}
+	}
+	return 0;
+}
+
+#define ROCKER_TEST_DMA_BUF_SIZE (PAGE_SIZE * 4)
+#define ROCKER_TEST_DMA_FILL_PATTERN 0x96
+
+static int rocker_dma_test_offset(const struct rocker *rocker,
+				  struct rocker_wait *wait, int offset)
+{
+	struct pci_dev *pdev = rocker->pdev;
+	unsigned char *alloc;
+	unsigned char *buf;
+	unsigned char *expect;
+	dma_addr_t dma_handle;
+	int i;
+	int err;
+
+	alloc = kzalloc(ROCKER_TEST_DMA_BUF_SIZE * 2 + offset,
+			GFP_KERNEL | GFP_DMA);
+	if (!alloc)
+		return -ENOMEM;
+	buf = alloc + offset;
+	expect = buf + ROCKER_TEST_DMA_BUF_SIZE;
+
+	dma_handle = pci_map_single(pdev, buf, ROCKER_TEST_DMA_BUF_SIZE,
+				    PCI_DMA_BIDIRECTIONAL);
+	if (pci_dma_mapping_error(pdev, dma_handle)) {
+		err = -EIO;
+		goto free_alloc;
+	}
+
+	rocker_write64(rocker, TEST_DMA_ADDR, dma_handle);
+	rocker_write32(rocker, TEST_DMA_SIZE, ROCKER_TEST_DMA_BUF_SIZE);
+
+	memset(expect, ROCKER_TEST_DMA_FILL_PATTERN, ROCKER_TEST_DMA_BUF_SIZE);
+	err = rocker_dma_test_one(rocker, wait, ROCKER_TEST_DMA_CTRL_FILL,
+				  dma_handle, buf, expect,
+				  ROCKER_TEST_DMA_BUF_SIZE);
+	if (err)
+		goto unmap;
+
+	memset(expect, 0, ROCKER_TEST_DMA_BUF_SIZE);
+	err = rocker_dma_test_one(rocker, wait, ROCKER_TEST_DMA_CTRL_CLEAR,
+				  dma_handle, buf, expect,
+				  ROCKER_TEST_DMA_BUF_SIZE);
+	if (err)
+		goto unmap;
+
+	prandom_bytes(buf, ROCKER_TEST_DMA_BUF_SIZE);
+	for (i = 0; i < ROCKER_TEST_DMA_BUF_SIZE; i++)
+		expect[i] = ~buf[i];
+	err = rocker_dma_test_one(rocker, wait, ROCKER_TEST_DMA_CTRL_INVERT,
+				  dma_handle, buf, expect,
+				  ROCKER_TEST_DMA_BUF_SIZE);
+	if (err)
+		goto unmap;
+
+unmap:
+	pci_unmap_single(pdev, dma_handle, ROCKER_TEST_DMA_BUF_SIZE,
+			 PCI_DMA_BIDIRECTIONAL);
+free_alloc:
+	kfree(alloc);
+
+	return err;
+}
+
+static int rocker_dma_test(const struct rocker *rocker,
+			   struct rocker_wait *wait)
+{
+	int i;
+	int err;
+
+	for (i = 0; i < 8; i++) {
+		err = rocker_dma_test_offset(rocker, wait, i);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static irqreturn_t rocker_test_irq_handler(int irq, void *dev_id)
+{
+	struct rocker_wait *wait = dev_id;
+
+	rocker_wait_wake_up(wait);
+
+	return IRQ_HANDLED;
+}
+
+static int rocker_basic_hw_test(const struct rocker *rocker)
+{
+	const struct pci_dev *pdev = rocker->pdev;
+	struct rocker_wait wait;
+	int err;
+
+	err = rocker_reg_test(rocker);
+	if (err) {
+		dev_err(&pdev->dev, "reg test failed\n");
+		return err;
+	}
+
+	err = request_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_TEST),
+			  rocker_test_irq_handler, 0,
+			  rocker_driver_name, &wait);
+	if (err) {
+		dev_err(&pdev->dev, "cannot assign test irq\n");
+		return err;
+	}
+
+	rocker_wait_init(&wait);
+	rocker_write32(rocker, TEST_IRQ, ROCKER_MSIX_VEC_TEST);
+
+	if (!rocker_wait_event_timeout(&wait, HZ / 10)) {
+		dev_err(&pdev->dev, "no interrupt received within a timeout\n");
+		err = -EIO;
+		goto free_irq;
+	}
+
+	err = rocker_dma_test(rocker, &wait);
+	if (err)
+		dev_err(&pdev->dev, "dma test failed\n");
+
+free_irq:
+	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_TEST), &wait);
+	return err;
+}
+
+/******
+ * TLV
+ ******/
+
+#define ROCKER_TLV_ALIGNTO 8U
+#define ROCKER_TLV_ALIGN(len) \
+	(((len) + ROCKER_TLV_ALIGNTO - 1) & ~(ROCKER_TLV_ALIGNTO - 1))
+#define ROCKER_TLV_HDRLEN ROCKER_TLV_ALIGN(sizeof(struct rocker_tlv))
+
+/*  <------- ROCKER_TLV_HDRLEN -------> <--- ROCKER_TLV_ALIGN(payload) --->
+ * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
+ * |             Header          | Pad |           Payload           | Pad |
+ * |      (struct rocker_tlv)    | ing |                             | ing |
+ * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
+ *  <--------------------------- tlv->len -------------------------->
+ */
+
+static struct rocker_tlv *rocker_tlv_next(const struct rocker_tlv *tlv,
+					  int *remaining)
+{
+	int totlen = ROCKER_TLV_ALIGN(tlv->len);
+
+	*remaining -= totlen;
+	return (struct rocker_tlv *) ((char *) tlv + totlen);
+}
+
+static int rocker_tlv_ok(const struct rocker_tlv *tlv, int remaining)
+{
+	return remaining >= (int) ROCKER_TLV_HDRLEN &&
+	       tlv->len >= ROCKER_TLV_HDRLEN &&
+	       tlv->len <= remaining;
+}
+
+#define rocker_tlv_for_each(pos, head, len, rem)	\
+	for (pos = head, rem = len;			\
+	     rocker_tlv_ok(pos, rem);			\
+	     pos = rocker_tlv_next(pos, &(rem)))
+
+#define rocker_tlv_for_each_nested(pos, tlv, rem)	\
+	rocker_tlv_for_each(pos, rocker_tlv_data(tlv),	\
+			    rocker_tlv_len(tlv), rem)
+
+static int rocker_tlv_attr_size(int payload)
+{
+	return ROCKER_TLV_HDRLEN + payload;
+}
+
+static int rocker_tlv_total_size(int payload)
+{
+	return ROCKER_TLV_ALIGN(rocker_tlv_attr_size(payload));
+}
+
+static int rocker_tlv_padlen(int payload)
+{
+	return rocker_tlv_total_size(payload) - rocker_tlv_attr_size(payload);
+}
+
+static int rocker_tlv_type(const struct rocker_tlv *tlv)
+{
+	return tlv->type;
+}
+
+static void *rocker_tlv_data(const struct rocker_tlv *tlv)
+{
+	return (char *) tlv + ROCKER_TLV_HDRLEN;
+}
+
+static int rocker_tlv_len(const struct rocker_tlv *tlv)
+{
+	return tlv->len - ROCKER_TLV_HDRLEN;
+}
+
+static u8 rocker_tlv_get_u8(const struct rocker_tlv *tlv)
+{
+	return *(u8 *) rocker_tlv_data(tlv);
+}
+
+static u16 rocker_tlv_get_u16(const struct rocker_tlv *tlv)
+{
+	return *(u16 *) rocker_tlv_data(tlv);
+}
+
+static __be16 rocker_tlv_get_be16(const struct rocker_tlv *tlv)
+{
+	return *(__be16 *) rocker_tlv_data(tlv);
+}
+
+static u32 rocker_tlv_get_u32(const struct rocker_tlv *tlv)
+{
+	return *(u32 *) rocker_tlv_data(tlv);
+}
+
+static u64 rocker_tlv_get_u64(const struct rocker_tlv *tlv)
+{
+	return *(u64 *) rocker_tlv_data(tlv);
+}
+
+static void rocker_tlv_parse(const struct rocker_tlv **tb, int maxtype,
+			     const char *buf, int buf_len)
+{
+	const struct rocker_tlv *tlv;
+	const struct rocker_tlv *head = (const struct rocker_tlv *) buf;
+	int rem;
+
+	memset(tb, 0, sizeof(struct rocker_tlv *) * (maxtype + 1));
+
+	rocker_tlv_for_each(tlv, head, buf_len, rem) {
+		u32 type = rocker_tlv_type(tlv);
+
+		if (type > 0 && type <= maxtype)
+			tb[type] = tlv;
+	}
+}
+
+static void rocker_tlv_parse_nested(const struct rocker_tlv **tb, int maxtype,
+				    const struct rocker_tlv *tlv)
+{
+	rocker_tlv_parse(tb, maxtype, rocker_tlv_data(tlv),
+			 rocker_tlv_len(tlv));
+}
+
+static void rocker_tlv_parse_desc(const struct rocker_tlv **tb, int maxtype,
+				  const struct rocker_desc_info *desc_info)
+{
+	rocker_tlv_parse(tb, maxtype, desc_info->data,
+			 desc_info->desc->tlv_size);
+}
+
+static struct rocker_tlv *rocker_tlv_start(struct rocker_desc_info *desc_info)
+{
+	return (struct rocker_tlv *) ((char *) desc_info->data +
+					       desc_info->tlv_size);
+}
+
+static int rocker_tlv_put(struct rocker_desc_info *desc_info,
+			  int attrtype, int attrlen, const void *data)
+{
+	int tail_room = desc_info->data_size - desc_info->tlv_size;
+	int total_size = rocker_tlv_total_size(attrlen);
+	struct rocker_tlv *tlv;
+
+	if (unlikely(tail_room < total_size))
+		return -EMSGSIZE;
+
+	tlv = rocker_tlv_start(desc_info);
+	desc_info->tlv_size += total_size;
+	tlv->type = attrtype;
+	tlv->len = rocker_tlv_attr_size(attrlen);
+	memcpy(rocker_tlv_data(tlv), data, attrlen);
+	memset((char *) tlv + tlv->len, 0, rocker_tlv_padlen(attrlen));
+	return 0;
+}
+
+static int rocker_tlv_put_u8(struct rocker_desc_info *desc_info,
+			     int attrtype, u8 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u8), &value);
+}
+
+static int rocker_tlv_put_u16(struct rocker_desc_info *desc_info,
+			      int attrtype, u16 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u16), &value);
+}
+
+static int rocker_tlv_put_be16(struct rocker_desc_info *desc_info,
+			       int attrtype, __be16 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(__be16), &value);
+}
+
+static int rocker_tlv_put_u32(struct rocker_desc_info *desc_info,
+			      int attrtype, u32 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u32), &value);
+}
+
+static int rocker_tlv_put_be32(struct rocker_desc_info *desc_info,
+			       int attrtype, __be32 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(__be32), &value);
+}
+
+static int rocker_tlv_put_u64(struct rocker_desc_info *desc_info,
+			      int attrtype, u64 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u64), &value);
+}
+
+static struct rocker_tlv *
+rocker_tlv_nest_start(struct rocker_desc_info *desc_info, int attrtype)
+{
+	struct rocker_tlv *start = rocker_tlv_start(desc_info);
+
+	if (rocker_tlv_put(desc_info, attrtype, 0, NULL) < 0)
+		return NULL;
+
+	return start;
+}
+
+static void rocker_tlv_nest_end(struct rocker_desc_info *desc_info,
+				struct rocker_tlv *start)
+{
+	start->len = (char *) rocker_tlv_start(desc_info) - (char *) start;
+}
+
+static void rocker_tlv_nest_cancel(struct rocker_desc_info *desc_info,
+				   const struct rocker_tlv *start)
+{
+	desc_info->tlv_size = (const char *) start - desc_info->data;
+}
+
+/******************************************
+ * DMA rings and descriptors manipulations
+ ******************************************/
+
+static u32 __pos_inc(u32 pos, size_t limit)
+{
+	return ++pos == limit ? 0 : pos;
+}
+
+static int rocker_desc_err(const struct rocker_desc_info *desc_info)
+{
+	int err = desc_info->desc->comp_err & ~ROCKER_DMA_DESC_COMP_ERR_GEN;
+
+	switch (err) {
+	case ROCKER_OK:
+		return 0;
+	case -ROCKER_ENOENT:
+		return -ENOENT;
+	case -ROCKER_ENXIO:
+		return -ENXIO;
+	case -ROCKER_ENOMEM:
+		return -ENOMEM;
+	case -ROCKER_EEXIST:
+		return -EEXIST;
+	case -ROCKER_EINVAL:
+		return -EINVAL;
+	case -ROCKER_EMSGSIZE:
+		return -EMSGSIZE;
+	case -ROCKER_ENOTSUP:
+		return -EOPNOTSUPP;
+	case -ROCKER_ENOBUFS:
+		return -ENOBUFS;
+	}
+
+	return -EINVAL;
+}
+
+static void rocker_desc_gen_clear(const struct rocker_desc_info *desc_info)
+{
+	desc_info->desc->comp_err &= ~ROCKER_DMA_DESC_COMP_ERR_GEN;
+}
+
+static bool rocker_desc_gen(const struct rocker_desc_info *desc_info)
+{
+	u32 comp_err = desc_info->desc->comp_err;
+
+	return comp_err & ROCKER_DMA_DESC_COMP_ERR_GEN ? true : false;
+}
+
+static void *
+rocker_desc_cookie_ptr_get(const struct rocker_desc_info *desc_info)
+{
+	return (void *)(uintptr_t)desc_info->desc->cookie;
+}
+
+static void rocker_desc_cookie_ptr_set(const struct rocker_desc_info *desc_info,
+				       void *ptr)
+{
+	desc_info->desc->cookie = (uintptr_t) ptr;
+}
+
+static struct rocker_desc_info *
+rocker_desc_head_get(const struct rocker_dma_ring_info *info)
+{
+	static struct rocker_desc_info *desc_info;
+	u32 head = __pos_inc(info->head, info->size);
+
+	desc_info = &info->desc_info[info->head];
+	if (head == info->tail)
+		return NULL; /* ring full */
+	desc_info->tlv_size = 0;
+	return desc_info;
+}
+
+static void rocker_desc_commit(const struct rocker_desc_info *desc_info)
+{
+	desc_info->desc->buf_size = desc_info->data_size;
+	desc_info->desc->tlv_size = desc_info->tlv_size;
+}
+
+static void rocker_desc_head_set(const struct rocker *rocker,
+				 struct rocker_dma_ring_info *info,
+				 const struct rocker_desc_info *desc_info)
+{
+	u32 head = __pos_inc(info->head, info->size);
+
+	BUG_ON(head == info->tail);
+	rocker_desc_commit(desc_info);
+	info->head = head;
+	rocker_write32(rocker, DMA_DESC_HEAD(info->type), head);
+}
+
+static struct rocker_desc_info *
+rocker_desc_tail_get(struct rocker_dma_ring_info *info)
+{
+	static struct rocker_desc_info *desc_info;
+
+	if (info->tail == info->head)
+		return NULL; /* nothing to be done between head and tail */
+	desc_info = &info->desc_info[info->tail];
+	if (!rocker_desc_gen(desc_info))
+		return NULL; /* gen bit not set, desc is not ready yet */
+	info->tail = __pos_inc(info->tail, info->size);
+	desc_info->tlv_size = desc_info->desc->tlv_size;
+	return desc_info;
+}
+
+static void rocker_dma_ring_credits_set(const struct rocker *rocker,
+					const struct rocker_dma_ring_info *info,
+					u32 credits)
+{
+	if (credits)
+		rocker_write32(rocker, DMA_DESC_CREDITS(info->type), credits);
+}
+
+static unsigned long rocker_dma_ring_size_fix(size_t size)
+{
+	return max(ROCKER_DMA_SIZE_MIN,
+		   min(roundup_pow_of_two(size), ROCKER_DMA_SIZE_MAX));
+}
+
+static int rocker_dma_ring_create(const struct rocker *rocker,
+				  unsigned int type,
+				  size_t size,
+				  struct rocker_dma_ring_info *info)
+{
+	int i;
+
+	BUG_ON(size != rocker_dma_ring_size_fix(size));
+	info->size = size;
+	info->type = type;
+	info->head = 0;
+	info->tail = 0;
+	info->desc_info = kcalloc(info->size, sizeof(*info->desc_info),
+				  GFP_KERNEL);
+	if (!info->desc_info)
+		return -ENOMEM;
+
+	info->desc = pci_alloc_consistent(rocker->pdev,
+					  info->size * sizeof(*info->desc),
+					  &info->mapaddr);
+	if (!info->desc) {
+		kfree(info->desc_info);
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < info->size; i++)
+		info->desc_info[i].desc = &info->desc[i];
+
+	rocker_write32(rocker, DMA_DESC_CTRL(info->type),
+		       ROCKER_DMA_DESC_CTRL_RESET);
+	rocker_write64(rocker, DMA_DESC_ADDR(info->type), info->mapaddr);
+	rocker_write32(rocker, DMA_DESC_SIZE(info->type), info->size);
+
+	return 0;
+}
+
+static void rocker_dma_ring_destroy(const struct rocker *rocker,
+				    const struct rocker_dma_ring_info *info)
+{
+	rocker_write64(rocker, DMA_DESC_ADDR(info->type), 0);
+
+	pci_free_consistent(rocker->pdev,
+			    info->size * sizeof(struct rocker_desc),
+			    info->desc, info->mapaddr);
+	kfree(info->desc_info);
+}
+
+static void rocker_dma_ring_pass_to_producer(const struct rocker *rocker,
+					     struct rocker_dma_ring_info *info)
+{
+	int i;
+
+	BUG_ON(info->head || info->tail);
+
+	/* When ring is consumer, we need to advance head for each desc.
+	 * That tells hw that the desc is ready to be used by it.
+	 */
+	for (i = 0; i < info->size - 1; i++)
+		rocker_desc_head_set(rocker, info, &info->desc_info[i]);
+	rocker_desc_commit(&info->desc_info[i]);
+}
+
+static int rocker_dma_ring_bufs_alloc(const struct rocker *rocker,
+				      const struct rocker_dma_ring_info *info,
+				      int direction, size_t buf_size)
+{
+	struct pci_dev *pdev = rocker->pdev;
+	int i;
+	int err;
+
+	for (i = 0; i < info->size; i++) {
+		struct rocker_desc_info *desc_info = &info->desc_info[i];
+		struct rocker_desc *desc = &info->desc[i];
+		dma_addr_t dma_handle;
+		char *buf;
+
+		buf = kzalloc(buf_size, GFP_KERNEL | GFP_DMA);
+		if (!buf) {
+			err = -ENOMEM;
+			goto rollback;
+		}
+
+		dma_handle = pci_map_single(pdev, buf, buf_size, direction);
+		if (pci_dma_mapping_error(pdev, dma_handle)) {
+			kfree(buf);
+			err = -EIO;
+			goto rollback;
+		}
+
+		desc_info->data = buf;
+		desc_info->data_size = buf_size;
+		dma_unmap_addr_set(desc_info, mapaddr, dma_handle);
+
+		desc->buf_addr = dma_handle;
+		desc->buf_size = buf_size;
+	}
+	return 0;
+
+rollback:
+	for (i--; i >= 0; i--) {
+		const struct rocker_desc_info *desc_info = &info->desc_info[i];
+
+		pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr),
+				 desc_info->data_size, direction);
+		kfree(desc_info->data);
+	}
+	return err;
+}
+
+static void rocker_dma_ring_bufs_free(const struct rocker *rocker,
+				      const struct rocker_dma_ring_info *info,
+				      int direction)
+{
+	struct pci_dev *pdev = rocker->pdev;
+	int i;
+
+	for (i = 0; i < info->size; i++) {
+		const struct rocker_desc_info *desc_info = &info->desc_info[i];
+		struct rocker_desc *desc = &info->desc[i];
+
+		desc->buf_addr = 0;
+		desc->buf_size = 0;
+		pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr),
+				 desc_info->data_size, direction);
+		kfree(desc_info->data);
+	}
+}
+
+static int rocker_dma_rings_init(struct rocker *rocker)
+{
+	const struct pci_dev *pdev = rocker->pdev;
+	int err;
+
+	err = rocker_dma_ring_create(rocker, ROCKER_DMA_CMD,
+				     ROCKER_DMA_CMD_DEFAULT_SIZE,
+				     &rocker->cmd_ring);
+	if (err) {
+		dev_err(&pdev->dev, "failed to create command dma ring\n");
+		return err;
+	}
+
+	spin_lock_init(&rocker->cmd_ring_lock);
+
+	err = rocker_dma_ring_bufs_alloc(rocker, &rocker->cmd_ring,
+					 PCI_DMA_BIDIRECTIONAL, PAGE_SIZE);
+	if (err) {
+		dev_err(&pdev->dev, "failed to alloc command dma ring buffers\n");
+		goto err_dma_cmd_ring_bufs_alloc;
+	}
+
+	err = rocker_dma_ring_create(rocker, ROCKER_DMA_EVENT,
+				     ROCKER_DMA_EVENT_DEFAULT_SIZE,
+				     &rocker->event_ring);
+	if (err) {
+		dev_err(&pdev->dev, "failed to create event dma ring\n");
+		goto err_dma_event_ring_create;
+	}
+
+	err = rocker_dma_ring_bufs_alloc(rocker, &rocker->event_ring,
+					 PCI_DMA_FROMDEVICE, PAGE_SIZE);
+	if (err) {
+		dev_err(&pdev->dev, "failed to alloc event dma ring buffers\n");
+		goto err_dma_event_ring_bufs_alloc;
+	}
+	rocker_dma_ring_pass_to_producer(rocker, &rocker->event_ring);
+	return 0;
+
+err_dma_event_ring_bufs_alloc:
+	rocker_dma_ring_destroy(rocker, &rocker->event_ring);
+err_dma_event_ring_create:
+	rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring,
+				  PCI_DMA_BIDIRECTIONAL);
+err_dma_cmd_ring_bufs_alloc:
+	rocker_dma_ring_destroy(rocker, &rocker->cmd_ring);
+	return err;
+}
+
+static void rocker_dma_rings_fini(struct rocker *rocker)
+{
+	rocker_dma_ring_bufs_free(rocker, &rocker->event_ring,
+				  PCI_DMA_BIDIRECTIONAL);
+	rocker_dma_ring_destroy(rocker, &rocker->event_ring);
+	rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring,
+				  PCI_DMA_BIDIRECTIONAL);
+	rocker_dma_ring_destroy(rocker, &rocker->cmd_ring);
+}
+
+static int rocker_dma_rx_ring_skb_map(const struct rocker_port *rocker_port,
+				      struct rocker_desc_info *desc_info,
+				      struct sk_buff *skb, size_t buf_len)
+{
+	const struct rocker *rocker = rocker_port->rocker;
+	struct pci_dev *pdev = rocker->pdev;
+	dma_addr_t dma_handle;
+
+	dma_handle = pci_map_single(pdev, skb->data, buf_len,
+				    PCI_DMA_FROMDEVICE);
+	if (pci_dma_mapping_error(pdev, dma_handle))
+		return -EIO;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_RX_FRAG_ADDR, dma_handle))
+		goto tlv_put_failure;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_RX_FRAG_MAX_LEN, buf_len))
+		goto tlv_put_failure;
+	return 0;
+
+tlv_put_failure:
+	pci_unmap_single(pdev, dma_handle, buf_len, PCI_DMA_FROMDEVICE);
+	desc_info->tlv_size = 0;
+	return -EMSGSIZE;
+}
+
+static size_t rocker_port_rx_buf_len(const struct rocker_port *rocker_port)
+{
+	return rocker_port->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
+}
+
+static int rocker_dma_rx_ring_skb_alloc(const struct rocker_port *rocker_port,
+					struct rocker_desc_info *desc_info)
+{
+	struct net_device *dev = rocker_port->dev;
+	struct sk_buff *skb;
+	size_t buf_len = rocker_port_rx_buf_len(rocker_port);
+	int err;
+
+	/* Ensure that hw will see tlv_size zero in case of an error.
+	 * That tells hw to use another descriptor.
+	 */
+	rocker_desc_cookie_ptr_set(desc_info, NULL);
+	desc_info->tlv_size = 0;
+
+	skb = netdev_alloc_skb_ip_align(dev, buf_len);
+	if (!skb)
+		return -ENOMEM;
+	err = rocker_dma_rx_ring_skb_map(rocker_port, desc_info, skb, buf_len);
+	if (err) {
+		dev_kfree_skb_any(skb);
+		return err;
+	}
+	rocker_desc_cookie_ptr_set(desc_info, skb);
+	return 0;
+}
+
+static void rocker_dma_rx_ring_skb_unmap(const struct rocker *rocker,
+					 const struct rocker_tlv **attrs)
+{
+	struct pci_dev *pdev = rocker->pdev;
+	dma_addr_t dma_handle;
+	size_t len;
+
+	if (!attrs[ROCKER_TLV_RX_FRAG_ADDR] ||
+	    !attrs[ROCKER_TLV_RX_FRAG_MAX_LEN])
+		return;
+	dma_handle = rocker_tlv_get_u64(attrs[ROCKER_TLV_RX_FRAG_ADDR]);
+	len = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FRAG_MAX_LEN]);
+	pci_unmap_single(pdev, dma_handle, len, PCI_DMA_FROMDEVICE);
+}
+
+static void rocker_dma_rx_ring_skb_free(const struct rocker *rocker,
+					const struct rocker_desc_info *desc_info)
+{
+	const struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1];
+	struct sk_buff *skb = rocker_desc_cookie_ptr_get(desc_info);
+
+	if (!skb)
+		return;
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_RX_MAX, desc_info);
+	rocker_dma_rx_ring_skb_unmap(rocker, attrs);
+	dev_kfree_skb_any(skb);
+}
+
+static int rocker_dma_rx_ring_skbs_alloc(const struct rocker_port *rocker_port)
+{
+	const struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring;
+	const struct rocker *rocker = rocker_port->rocker;
+	int i;
+	int err;
+
+	for (i = 0; i < rx_ring->size; i++) {
+		err = rocker_dma_rx_ring_skb_alloc(rocker_port,
+						   &rx_ring->desc_info[i]);
+		if (err)
+			goto rollback;
+	}
+	return 0;
+
+rollback:
+	for (i--; i >= 0; i--)
+		rocker_dma_rx_ring_skb_free(rocker, &rx_ring->desc_info[i]);
+	return err;
+}
+
+static void rocker_dma_rx_ring_skbs_free(const struct rocker_port *rocker_port)
+{
+	const struct rocker_dma_ring_info *rx_ring = &rocker_port->rx_ring;
+	const struct rocker *rocker = rocker_port->rocker;
+	int i;
+
+	for (i = 0; i < rx_ring->size; i++)
+		rocker_dma_rx_ring_skb_free(rocker, &rx_ring->desc_info[i]);
+}
+
+static int rocker_port_dma_rings_init(struct rocker_port *rocker_port)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	int err;
+
+	err = rocker_dma_ring_create(rocker,
+				     ROCKER_DMA_TX(rocker_port->port_number),
+				     ROCKER_DMA_TX_DEFAULT_SIZE,
+				     &rocker_port->tx_ring);
+	if (err) {
+		netdev_err(rocker_port->dev, "failed to create tx dma ring\n");
+		return err;
+	}
+
+	err = rocker_dma_ring_bufs_alloc(rocker, &rocker_port->tx_ring,
+					 PCI_DMA_TODEVICE,
+					 ROCKER_DMA_TX_DESC_SIZE);
+	if (err) {
+		netdev_err(rocker_port->dev, "failed to alloc tx dma ring buffers\n");
+		goto err_dma_tx_ring_bufs_alloc;
+	}
+
+	err = rocker_dma_ring_create(rocker,
+				     ROCKER_DMA_RX(rocker_port->port_number),
+				     ROCKER_DMA_RX_DEFAULT_SIZE,
+				     &rocker_port->rx_ring);
+	if (err) {
+		netdev_err(rocker_port->dev, "failed to create rx dma ring\n");
+		goto err_dma_rx_ring_create;
+	}
+
+	err = rocker_dma_ring_bufs_alloc(rocker, &rocker_port->rx_ring,
+					 PCI_DMA_BIDIRECTIONAL,
+					 ROCKER_DMA_RX_DESC_SIZE);
+	if (err) {
+		netdev_err(rocker_port->dev, "failed to alloc rx dma ring buffers\n");
+		goto err_dma_rx_ring_bufs_alloc;
+	}
+
+	err = rocker_dma_rx_ring_skbs_alloc(rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "failed to alloc rx dma ring skbs\n");
+		goto err_dma_rx_ring_skbs_alloc;
+	}
+	rocker_dma_ring_pass_to_producer(rocker, &rocker_port->rx_ring);
+
+	return 0;
+
+err_dma_rx_ring_skbs_alloc:
+	rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring,
+				  PCI_DMA_BIDIRECTIONAL);
+err_dma_rx_ring_bufs_alloc:
+	rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring);
+err_dma_rx_ring_create:
+	rocker_dma_ring_bufs_free(rocker, &rocker_port->tx_ring,
+				  PCI_DMA_TODEVICE);
+err_dma_tx_ring_bufs_alloc:
+	rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring);
+	return err;
+}
+
+static void rocker_port_dma_rings_fini(struct rocker_port *rocker_port)
+{
+	struct rocker *rocker = rocker_port->rocker;
+
+	rocker_dma_rx_ring_skbs_free(rocker_port);
+	rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring,
+				  PCI_DMA_BIDIRECTIONAL);
+	rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring);
+	rocker_dma_ring_bufs_free(rocker, &rocker_port->tx_ring,
+				  PCI_DMA_TODEVICE);
+	rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring);
+}
+
+static void rocker_port_set_enable(const struct rocker_port *rocker_port,
+				   bool enable)
+{
+	u64 val = rocker_read64(rocker_port->rocker, PORT_PHYS_ENABLE);
+
+	if (enable)
+		val |= 1ULL << rocker_port->pport;
+	else
+		val &= ~(1ULL << rocker_port->pport);
+	rocker_write64(rocker_port->rocker, PORT_PHYS_ENABLE, val);
+}
+
+/********************************
+ * Interrupt handler and helpers
+ ********************************/
+
+static irqreturn_t rocker_cmd_irq_handler(int irq, void *dev_id)
+{
+	struct rocker *rocker = dev_id;
+	const struct rocker_desc_info *desc_info;
+	struct rocker_wait *wait;
+	u32 credits = 0;
+
+	spin_lock(&rocker->cmd_ring_lock);
+	while ((desc_info = rocker_desc_tail_get(&rocker->cmd_ring))) {
+		wait = rocker_desc_cookie_ptr_get(desc_info);
+		if (wait->nowait) {
+			rocker_desc_gen_clear(desc_info);
+			rocker_wait_destroy(NULL, wait);
+		} else {
+			rocker_wait_wake_up(wait);
+		}
+		credits++;
+	}
+	spin_unlock(&rocker->cmd_ring_lock);
+	rocker_dma_ring_credits_set(rocker, &rocker->cmd_ring, credits);
+
+	return IRQ_HANDLED;
+}
+
+static void rocker_port_link_up(const struct rocker_port *rocker_port)
+{
+	netif_carrier_on(rocker_port->dev);
+	netdev_info(rocker_port->dev, "Link is up\n");
+}
+
+static void rocker_port_link_down(const struct rocker_port *rocker_port)
+{
+	netif_carrier_off(rocker_port->dev);
+	netdev_info(rocker_port->dev, "Link is down\n");
+}
+
+static int rocker_event_link_change(const struct rocker *rocker,
+				    const struct rocker_tlv *info)
+{
+	const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_LINK_CHANGED_MAX + 1];
+	unsigned int port_number;
+	bool link_up;
+	struct rocker_port *rocker_port;
+
+	rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_LINK_CHANGED_MAX, info);
+	if (!attrs[ROCKER_TLV_EVENT_LINK_CHANGED_PPORT] ||
+	    !attrs[ROCKER_TLV_EVENT_LINK_CHANGED_LINKUP])
+		return -EIO;
+	port_number =
+		rocker_tlv_get_u32(attrs[ROCKER_TLV_EVENT_LINK_CHANGED_PPORT]) - 1;
+	link_up = rocker_tlv_get_u8(attrs[ROCKER_TLV_EVENT_LINK_CHANGED_LINKUP]);
+
+	if (port_number >= rocker->port_count)
+		return -EINVAL;
+
+	rocker_port = rocker->ports[port_number];
+	if (netif_carrier_ok(rocker_port->dev) != link_up) {
+		if (link_up)
+			rocker_port_link_up(rocker_port);
+		else
+			rocker_port_link_down(rocker_port);
+	}
+
+	return 0;
+}
+
+static int rocker_port_fdb(struct rocker_port *rocker_port,
+			   struct switchdev_trans *trans,
+			   const unsigned char *addr,
+			   __be16 vlan_id, int flags);
+
+static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
+				      const struct rocker_tlv *info)
+{
+	const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAX + 1];
+	unsigned int port_number;
+	struct rocker_port *rocker_port;
+	const unsigned char *addr;
+	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_LEARNED;
+	__be16 vlan_id;
+
+	rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_MAC_VLAN_MAX, info);
+	if (!attrs[ROCKER_TLV_EVENT_MAC_VLAN_PPORT] ||
+	    !attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAC] ||
+	    !attrs[ROCKER_TLV_EVENT_MAC_VLAN_VLAN_ID])
+		return -EIO;
+	port_number =
+		rocker_tlv_get_u32(attrs[ROCKER_TLV_EVENT_MAC_VLAN_PPORT]) - 1;
+	addr = rocker_tlv_data(attrs[ROCKER_TLV_EVENT_MAC_VLAN_MAC]);
+	vlan_id = rocker_tlv_get_be16(attrs[ROCKER_TLV_EVENT_MAC_VLAN_VLAN_ID]);
+
+	if (port_number >= rocker->port_count)
+		return -EINVAL;
+
+	rocker_port = rocker->ports[port_number];
+
+	if (rocker_port->stp_state != BR_STATE_LEARNING &&
+	    rocker_port->stp_state != BR_STATE_FORWARDING)
+		return 0;
+
+	return rocker_port_fdb(rocker_port, NULL, addr, vlan_id, flags);
+}
+
+static int rocker_event_process(const struct rocker *rocker,
+				const struct rocker_desc_info *desc_info)
+{
+	const struct rocker_tlv *attrs[ROCKER_TLV_EVENT_MAX + 1];
+	const struct rocker_tlv *info;
+	u16 type;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_EVENT_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_EVENT_TYPE] ||
+	    !attrs[ROCKER_TLV_EVENT_INFO])
+		return -EIO;
+
+	type = rocker_tlv_get_u16(attrs[ROCKER_TLV_EVENT_TYPE]);
+	info = attrs[ROCKER_TLV_EVENT_INFO];
+
+	switch (type) {
+	case ROCKER_TLV_EVENT_TYPE_LINK_CHANGED:
+		return rocker_event_link_change(rocker, info);
+	case ROCKER_TLV_EVENT_TYPE_MAC_VLAN_SEEN:
+		return rocker_event_mac_vlan_seen(rocker, info);
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static irqreturn_t rocker_event_irq_handler(int irq, void *dev_id)
+{
+	struct rocker *rocker = dev_id;
+	const struct pci_dev *pdev = rocker->pdev;
+	const struct rocker_desc_info *desc_info;
+	u32 credits = 0;
+	int err;
+
+	while ((desc_info = rocker_desc_tail_get(&rocker->event_ring))) {
+		err = rocker_desc_err(desc_info);
+		if (err) {
+			dev_err(&pdev->dev, "event desc received with err %d\n",
+				err);
+		} else {
+			err = rocker_event_process(rocker, desc_info);
+			if (err)
+				dev_err(&pdev->dev, "event processing failed with err %d\n",
+					err);
+		}
+		rocker_desc_gen_clear(desc_info);
+		rocker_desc_head_set(rocker, &rocker->event_ring, desc_info);
+		credits++;
+	}
+	rocker_dma_ring_credits_set(rocker, &rocker->event_ring, credits);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t rocker_tx_irq_handler(int irq, void *dev_id)
+{
+	struct rocker_port *rocker_port = dev_id;
+
+	napi_schedule(&rocker_port->napi_tx);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t rocker_rx_irq_handler(int irq, void *dev_id)
+{
+	struct rocker_port *rocker_port = dev_id;
+
+	napi_schedule(&rocker_port->napi_rx);
+	return IRQ_HANDLED;
+}
+
+/********************
+ * Command interface
+ ********************/
+
+typedef int (*rocker_cmd_prep_cb_t)(const struct rocker_port *rocker_port,
+				    struct rocker_desc_info *desc_info,
+				    void *priv);
+
+typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port,
+				    const struct rocker_desc_info *desc_info,
+				    void *priv);
+
+static int rocker_cmd_exec(struct rocker_port *rocker_port,
+			   struct switchdev_trans *trans, int flags,
+			   rocker_cmd_prep_cb_t prepare, void *prepare_priv,
+			   rocker_cmd_proc_cb_t process, void *process_priv)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_desc_info *desc_info;
+	struct rocker_wait *wait;
+	bool nowait = !!(flags & ROCKER_OP_FLAG_NOWAIT);
+	unsigned long lock_flags;
+	int err;
+
+	wait = rocker_wait_create(rocker_port, trans, flags);
+	if (!wait)
+		return -ENOMEM;
+	wait->nowait = nowait;
+
+	spin_lock_irqsave(&rocker->cmd_ring_lock, lock_flags);
+
+	desc_info = rocker_desc_head_get(&rocker->cmd_ring);
+	if (!desc_info) {
+		spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
+		err = -EAGAIN;
+		goto out;
+	}
+
+	err = prepare(rocker_port, desc_info, prepare_priv);
+	if (err) {
+		spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
+		goto out;
+	}
+
+	rocker_desc_cookie_ptr_set(desc_info, wait);
+
+	if (!switchdev_trans_ph_prepare(trans))
+		rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info);
+
+	spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
+
+	if (nowait)
+		return 0;
+
+	if (!switchdev_trans_ph_prepare(trans))
+		if (!rocker_wait_event_timeout(wait, HZ / 10))
+			return -EIO;
+
+	err = rocker_desc_err(desc_info);
+	if (err)
+		return err;
+
+	if (process)
+		err = process(rocker_port, desc_info, process_priv);
+
+	rocker_desc_gen_clear(desc_info);
+out:
+	rocker_wait_destroy(trans, wait);
+	return err;
+}
+
+static int
+rocker_cmd_get_port_settings_prep(const struct rocker_port *rocker_port,
+				  struct rocker_desc_info *desc_info,
+				  void *priv)
+{
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
+			       ROCKER_TLV_CMD_TYPE_GET_PORT_SETTINGS))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
+			       rocker_port->pport))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+	return 0;
+}
+
+static int
+rocker_cmd_get_port_settings_ethtool_proc(const struct rocker_port *rocker_port,
+					  const struct rocker_desc_info *desc_info,
+					  void *priv)
+{
+	struct ethtool_cmd *ecmd = priv;
+	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
+	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
+	u32 speed;
+	u8 duplex;
+	u8 autoneg;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_CMD_INFO])
+		return -EIO;
+
+	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
+				attrs[ROCKER_TLV_CMD_INFO]);
+	if (!info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_SPEED] ||
+	    !info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX] ||
+	    !info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG])
+		return -EIO;
+
+	speed = rocker_tlv_get_u32(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_SPEED]);
+	duplex = rocker_tlv_get_u8(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX]);
+	autoneg = rocker_tlv_get_u8(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG]);
+
+	ecmd->transceiver = XCVR_INTERNAL;
+	ecmd->supported = SUPPORTED_TP;
+	ecmd->phy_address = 0xff;
+	ecmd->port = PORT_TP;
+	ethtool_cmd_speed_set(ecmd, speed);
+	ecmd->duplex = duplex ? DUPLEX_FULL : DUPLEX_HALF;
+	ecmd->autoneg = autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_get_port_settings_macaddr_proc(const struct rocker_port *rocker_port,
+					  const struct rocker_desc_info *desc_info,
+					  void *priv)
+{
+	unsigned char *macaddr = priv;
+	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
+	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
+	const struct rocker_tlv *attr;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_CMD_INFO])
+		return -EIO;
+
+	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
+				attrs[ROCKER_TLV_CMD_INFO]);
+	attr = info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR];
+	if (!attr)
+		return -EIO;
+
+	if (rocker_tlv_len(attr) != ETH_ALEN)
+		return -EINVAL;
+
+	ether_addr_copy(macaddr, rocker_tlv_data(attr));
+	return 0;
+}
+
+struct port_name {
+	char *buf;
+	size_t len;
+};
+
+static int
+rocker_cmd_get_port_settings_phys_name_proc(const struct rocker_port *rocker_port,
+					    const struct rocker_desc_info *desc_info,
+					    void *priv)
+{
+	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
+	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
+	struct port_name *name = priv;
+	const struct rocker_tlv *attr;
+	size_t i, j, len;
+	const char *str;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_CMD_INFO])
+		return -EIO;
+
+	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
+				attrs[ROCKER_TLV_CMD_INFO]);
+	attr = info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_PHYS_NAME];
+	if (!attr)
+		return -EIO;
+
+	len = min_t(size_t, rocker_tlv_len(attr), name->len);
+	str = rocker_tlv_data(attr);
+
+	/* make sure name only contains alphanumeric characters */
+	for (i = j = 0; i < len; ++i) {
+		if (isalnum(str[i])) {
+			name->buf[j] = str[i];
+			j++;
+		}
+	}
+
+	if (j == 0)
+		return -EIO;
+
+	name->buf[j] = '\0';
+
+	return 0;
+}
+
+static int
+rocker_cmd_set_port_settings_ethtool_prep(const struct rocker_port *rocker_port,
+					  struct rocker_desc_info *desc_info,
+					  void *priv)
+{
+	struct ethtool_cmd *ecmd = priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
+			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
+			       rocker_port->pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_SPEED,
+			       ethtool_cmd_speed(ecmd)))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_DUPLEX,
+			      ecmd->duplex))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_AUTONEG,
+			      ecmd->autoneg))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+	return 0;
+}
+
+static int
+rocker_cmd_set_port_settings_macaddr_prep(const struct rocker_port *rocker_port,
+					  struct rocker_desc_info *desc_info,
+					  void *priv)
+{
+	const unsigned char *macaddr = priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
+			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
+			       rocker_port->pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR,
+			   ETH_ALEN, macaddr))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+	return 0;
+}
+
+static int
+rocker_cmd_set_port_settings_mtu_prep(const struct rocker_port *rocker_port,
+				      struct rocker_desc_info *desc_info,
+				      void *priv)
+{
+	int mtu = *(int *)priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
+			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
+			       rocker_port->pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_MTU,
+			       mtu))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+	return 0;
+}
+
+static int
+rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port,
+				  struct rocker_desc_info *desc_info,
+				  void *priv)
+{
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
+			       ROCKER_TLV_CMD_TYPE_SET_PORT_SETTINGS))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_PPORT,
+			       rocker_port->pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,
+			      !!(rocker_port->brport_flags & BR_LEARNING)))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+	return 0;
+}
+
+static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port,
+						struct ethtool_cmd *ecmd)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_get_port_settings_prep, NULL,
+			       rocker_cmd_get_port_settings_ethtool_proc,
+			       ecmd);
+}
+
+static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
+						unsigned char *macaddr)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_get_port_settings_prep, NULL,
+			       rocker_cmd_get_port_settings_macaddr_proc,
+			       macaddr);
+}
+
+static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
+						struct ethtool_cmd *ecmd)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_set_port_settings_ethtool_prep,
+			       ecmd, NULL, NULL);
+}
+
+static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port,
+						unsigned char *macaddr)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_set_port_settings_macaddr_prep,
+			       macaddr, NULL, NULL);
+}
+
+static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
+					    int mtu)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_set_port_settings_mtu_prep,
+			       &mtu, NULL, NULL);
+}
+
+static int rocker_port_set_learning(struct rocker_port *rocker_port,
+				    struct switchdev_trans *trans)
+{
+	return rocker_cmd_exec(rocker_port, trans, 0,
+			       rocker_cmd_set_port_learning_prep,
+			       NULL, NULL, NULL);
+}
+
+static int
+rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info,
+				const struct rocker_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.ig_port.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
+			       entry->key.ig_port.in_pport_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.ig_port.goto_tbl))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info,
+			     const struct rocker_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.vlan.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.vlan.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
+				entry->key.vlan.vlan_id_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.vlan.goto_tbl))
+		return -EMSGSIZE;
+	if (entry->key.vlan.untagged &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_NEW_VLAN_ID,
+				entry->key.vlan.new_vlan_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info,
+				 const struct rocker_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.term_mac.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
+			       entry->key.term_mac.in_pport_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
+				entry->key.term_mac.eth_type))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->key.term_mac.eth_dst))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
+			   ETH_ALEN, entry->key.term_mac.eth_dst_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.term_mac.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
+				entry->key.term_mac.vlan_id_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.term_mac.goto_tbl))
+		return -EMSGSIZE;
+	if (entry->key.term_mac.copy_to_cpu &&
+	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
+			      entry->key.term_mac.copy_to_cpu))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_flow_tbl_add_ucast_routing(struct rocker_desc_info *desc_info,
+				      const struct rocker_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
+				entry->key.ucast_routing.eth_type))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP,
+				entry->key.ucast_routing.dst4))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP_MASK,
+				entry->key.ucast_routing.dst4_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.ucast_routing.goto_tbl))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->key.ucast_routing.group_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info,
+			       const struct rocker_flow_tbl_entry *entry)
+{
+	if (entry->key.bridge.has_eth_dst &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->key.bridge.eth_dst))
+		return -EMSGSIZE;
+	if (entry->key.bridge.has_eth_dst_mask &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
+			   ETH_ALEN, entry->key.bridge.eth_dst_mask))
+		return -EMSGSIZE;
+	if (entry->key.bridge.vlan_id &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.bridge.vlan_id))
+		return -EMSGSIZE;
+	if (entry->key.bridge.tunnel_id &&
+	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_TUNNEL_ID,
+			       entry->key.bridge.tunnel_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.bridge.goto_tbl))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->key.bridge.group_id))
+		return -EMSGSIZE;
+	if (entry->key.bridge.copy_to_cpu &&
+	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
+			      entry->key.bridge.copy_to_cpu))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info,
+			    const struct rocker_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.acl.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
+			       entry->key.acl.in_pport_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
+			   ETH_ALEN, entry->key.acl.eth_src))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC_MASK,
+			   ETH_ALEN, entry->key.acl.eth_src_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->key.acl.eth_dst))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
+			   ETH_ALEN, entry->key.acl.eth_dst_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
+				entry->key.acl.eth_type))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.acl.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
+				entry->key.acl.vlan_id_mask))
+		return -EMSGSIZE;
+
+	switch (ntohs(entry->key.acl.eth_type)) {
+	case ETH_P_IP:
+	case ETH_P_IPV6:
+		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_PROTO,
+				      entry->key.acl.ip_proto))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info,
+				      ROCKER_TLV_OF_DPA_IP_PROTO_MASK,
+				      entry->key.acl.ip_proto_mask))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_DSCP,
+				      entry->key.acl.ip_tos & 0x3f))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info,
+				      ROCKER_TLV_OF_DPA_IP_DSCP_MASK,
+				      entry->key.acl.ip_tos_mask & 0x3f))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_ECN,
+				      (entry->key.acl.ip_tos & 0xc0) >> 6))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info,
+				      ROCKER_TLV_OF_DPA_IP_ECN_MASK,
+				      (entry->key.acl.ip_tos_mask & 0xc0) >> 6))
+			return -EMSGSIZE;
+		break;
+	}
+
+	if (entry->key.acl.group_id != ROCKER_GROUP_NONE &&
+	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->key.acl.group_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int rocker_cmd_flow_tbl_add(const struct rocker_port *rocker_port,
+				   struct rocker_desc_info *desc_info,
+				   void *priv)
+{
+	const struct rocker_flow_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+	int err = 0;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_TABLE_ID,
+			       entry->key.tbl_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_PRIORITY,
+			       entry->key.priority))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_HARDTIME, 0))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
+			       entry->cookie))
+		return -EMSGSIZE;
+
+	switch (entry->key.tbl_id) {
+	case ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT:
+		err = rocker_cmd_flow_tbl_add_ig_port(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_VLAN:
+		err = rocker_cmd_flow_tbl_add_vlan(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC:
+		err = rocker_cmd_flow_tbl_add_term_mac(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING:
+		err = rocker_cmd_flow_tbl_add_ucast_routing(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_BRIDGING:
+		err = rocker_cmd_flow_tbl_add_bridge(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_ACL_POLICY:
+		err = rocker_cmd_flow_tbl_add_acl(desc_info, entry);
+		break;
+	default:
+		err = -ENOTSUPP;
+		break;
+	}
+
+	if (err)
+		return err;
+
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+static int rocker_cmd_flow_tbl_del(const struct rocker_port *rocker_port,
+				   struct rocker_desc_info *desc_info,
+				   void *priv)
+{
+	const struct rocker_flow_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
+			       entry->cookie))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+static int
+rocker_cmd_group_tbl_add_l2_interface(struct rocker_desc_info *desc_info,
+				      struct rocker_group_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_OUT_PPORT,
+			       ROCKER_GROUP_PORT_GET(entry->group_id)))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_POP_VLAN,
+			      entry->l2_interface.pop_vlan))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_group_tbl_add_l2_rewrite(struct rocker_desc_info *desc_info,
+				    const struct rocker_group_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
+			       entry->l2_rewrite.group_id))
+		return -EMSGSIZE;
+	if (!is_zero_ether_addr(entry->l2_rewrite.eth_src) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
+			   ETH_ALEN, entry->l2_rewrite.eth_src))
+		return -EMSGSIZE;
+	if (!is_zero_ether_addr(entry->l2_rewrite.eth_dst) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->l2_rewrite.eth_dst))
+		return -EMSGSIZE;
+	if (entry->l2_rewrite.vlan_id &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->l2_rewrite.vlan_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+rocker_cmd_group_tbl_add_group_ids(struct rocker_desc_info *desc_info,
+				   const struct rocker_group_tbl_entry *entry)
+{
+	int i;
+	struct rocker_tlv *group_ids;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GROUP_COUNT,
+			       entry->group_count))
+		return -EMSGSIZE;
+
+	group_ids = rocker_tlv_nest_start(desc_info,
+					  ROCKER_TLV_OF_DPA_GROUP_IDS);
+	if (!group_ids)
+		return -EMSGSIZE;
+
+	for (i = 0; i < entry->group_count; i++)
+		/* Note TLV array is 1-based */
+		if (rocker_tlv_put_u32(desc_info, i + 1, entry->group_ids[i]))
+			return -EMSGSIZE;
+
+	rocker_tlv_nest_end(desc_info, group_ids);
+
+	return 0;
+}
+
+static int
+rocker_cmd_group_tbl_add_l3_unicast(struct rocker_desc_info *desc_info,
+				    const struct rocker_group_tbl_entry *entry)
+{
+	if (!is_zero_ether_addr(entry->l3_unicast.eth_src) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
+			   ETH_ALEN, entry->l3_unicast.eth_src))
+		return -EMSGSIZE;
+	if (!is_zero_ether_addr(entry->l3_unicast.eth_dst) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->l3_unicast.eth_dst))
+		return -EMSGSIZE;
+	if (entry->l3_unicast.vlan_id &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->l3_unicast.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_TTL_CHECK,
+			      entry->l3_unicast.ttl_check))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
+			       entry->l3_unicast.group_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int rocker_cmd_group_tbl_add(const struct rocker_port *rocker_port,
+				    struct rocker_desc_info *desc_info,
+				    void *priv)
+{
+	struct rocker_group_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+	int err = 0;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->group_id))
+		return -EMSGSIZE;
+
+	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE:
+		err = rocker_cmd_group_tbl_add_l2_interface(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE:
+		err = rocker_cmd_group_tbl_add_l2_rewrite(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
+		err = rocker_cmd_group_tbl_add_group_ids(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST:
+		err = rocker_cmd_group_tbl_add_l3_unicast(desc_info, entry);
+		break;
+	default:
+		err = -ENOTSUPP;
+		break;
+	}
+
+	if (err)
+		return err;
+
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+static int rocker_cmd_group_tbl_del(const struct rocker_port *rocker_port,
+				    struct rocker_desc_info *desc_info,
+				    void *priv)
+{
+	const struct rocker_group_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->group_id))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+/***************************************************
+ * Flow, group, FDB, internal VLAN and neigh tables
+ ***************************************************/
+
+static int rocker_init_tbls(struct rocker *rocker)
+{
+	hash_init(rocker->flow_tbl);
+	spin_lock_init(&rocker->flow_tbl_lock);
+
+	hash_init(rocker->group_tbl);
+	spin_lock_init(&rocker->group_tbl_lock);
+
+	hash_init(rocker->fdb_tbl);
+	spin_lock_init(&rocker->fdb_tbl_lock);
+
+	hash_init(rocker->internal_vlan_tbl);
+	spin_lock_init(&rocker->internal_vlan_tbl_lock);
+
+	hash_init(rocker->neigh_tbl);
+	spin_lock_init(&rocker->neigh_tbl_lock);
+
+	return 0;
+}
+
+static void rocker_free_tbls(struct rocker *rocker)
+{
+	unsigned long flags;
+	struct rocker_flow_tbl_entry *flow_entry;
+	struct rocker_group_tbl_entry *group_entry;
+	struct rocker_fdb_tbl_entry *fdb_entry;
+	struct rocker_internal_vlan_tbl_entry *internal_vlan_entry;
+	struct rocker_neigh_tbl_entry *neigh_entry;
+	struct hlist_node *tmp;
+	int bkt;
+
+	spin_lock_irqsave(&rocker->flow_tbl_lock, flags);
+	hash_for_each_safe(rocker->flow_tbl, bkt, tmp, flow_entry, entry)
+		hash_del(&flow_entry->entry);
+	spin_unlock_irqrestore(&rocker->flow_tbl_lock, flags);
+
+	spin_lock_irqsave(&rocker->group_tbl_lock, flags);
+	hash_for_each_safe(rocker->group_tbl, bkt, tmp, group_entry, entry)
+		hash_del(&group_entry->entry);
+	spin_unlock_irqrestore(&rocker->group_tbl_lock, flags);
+
+	spin_lock_irqsave(&rocker->fdb_tbl_lock, flags);
+	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, fdb_entry, entry)
+		hash_del(&fdb_entry->entry);
+	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, flags);
+
+	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, flags);
+	hash_for_each_safe(rocker->internal_vlan_tbl, bkt,
+			   tmp, internal_vlan_entry, entry)
+		hash_del(&internal_vlan_entry->entry);
+	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, flags);
+
+	spin_lock_irqsave(&rocker->neigh_tbl_lock, flags);
+	hash_for_each_safe(rocker->neigh_tbl, bkt, tmp, neigh_entry, entry)
+		hash_del(&neigh_entry->entry);
+	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, flags);
+}
+
+static struct rocker_flow_tbl_entry *
+rocker_flow_tbl_find(const struct rocker *rocker,
+		     const struct rocker_flow_tbl_entry *match)
+{
+	struct rocker_flow_tbl_entry *found;
+	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
+
+	hash_for_each_possible(rocker->flow_tbl, found,
+			       entry, match->key_crc32) {
+		if (memcmp(&found->key, &match->key, key_len) == 0)
+			return found;
+	}
+
+	return NULL;
+}
+
+static int rocker_flow_tbl_add(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans, int flags,
+			       struct rocker_flow_tbl_entry *match)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_flow_tbl_entry *found;
+	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
+	unsigned long lock_flags;
+
+	match->key_crc32 = crc32(~0, &match->key, key_len);
+
+	spin_lock_irqsave(&rocker->flow_tbl_lock, lock_flags);
+
+	found = rocker_flow_tbl_find(rocker, match);
+
+	if (found) {
+		match->cookie = found->cookie;
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		rocker_kfree(trans, found);
+		found = match;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD;
+	} else {
+		found = match;
+		found->cookie = rocker->flow_tbl_next_cookie++;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD;
+	}
+
+	if (!switchdev_trans_ph_prepare(trans))
+		hash_add(rocker->flow_tbl, &found->entry, found->key_crc32);
+
+	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
+
+	return rocker_cmd_exec(rocker_port, trans, flags,
+			       rocker_cmd_flow_tbl_add, found, NULL, NULL);
+}
+
+static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans, int flags,
+			       struct rocker_flow_tbl_entry *match)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_flow_tbl_entry *found;
+	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
+	unsigned long lock_flags;
+	int err = 0;
+
+	match->key_crc32 = crc32(~0, &match->key, key_len);
+
+	spin_lock_irqsave(&rocker->flow_tbl_lock, lock_flags);
+
+	found = rocker_flow_tbl_find(rocker, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL;
+	}
+
+	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
+
+	rocker_kfree(trans, match);
+
+	if (found) {
+		err = rocker_cmd_exec(rocker_port, trans, flags,
+				      rocker_cmd_flow_tbl_del,
+				      found, NULL, NULL);
+		rocker_kfree(trans, found);
+	}
+
+	return err;
+}
+
+static int rocker_flow_tbl_do(struct rocker_port *rocker_port,
+			      struct switchdev_trans *trans, int flags,
+			      struct rocker_flow_tbl_entry *entry)
+{
+	if (flags & ROCKER_OP_FLAG_REMOVE)
+		return rocker_flow_tbl_del(rocker_port, trans, flags, entry);
+	else
+		return rocker_flow_tbl_add(rocker_port, trans, flags, entry);
+}
+
+static int rocker_flow_tbl_ig_port(struct rocker_port *rocker_port,
+				   struct switchdev_trans *trans, int flags,
+				   u32 in_pport, u32 in_pport_mask,
+				   enum rocker_of_dpa_table_id goto_tbl)
+{
+	struct rocker_flow_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.priority = ROCKER_PRIORITY_IG_PORT;
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT;
+	entry->key.ig_port.in_pport = in_pport;
+	entry->key.ig_port.in_pport_mask = in_pport_mask;
+	entry->key.ig_port.goto_tbl = goto_tbl;
+
+	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_flow_tbl_vlan(struct rocker_port *rocker_port,
+				struct switchdev_trans *trans, int flags,
+				u32 in_pport, __be16 vlan_id,
+				__be16 vlan_id_mask,
+				enum rocker_of_dpa_table_id goto_tbl,
+				bool untagged, __be16 new_vlan_id)
+{
+	struct rocker_flow_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.priority = ROCKER_PRIORITY_VLAN;
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_VLAN;
+	entry->key.vlan.in_pport = in_pport;
+	entry->key.vlan.vlan_id = vlan_id;
+	entry->key.vlan.vlan_id_mask = vlan_id_mask;
+	entry->key.vlan.goto_tbl = goto_tbl;
+
+	entry->key.vlan.untagged = untagged;
+	entry->key.vlan.new_vlan_id = new_vlan_id;
+
+	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port,
+				    struct switchdev_trans *trans,
+				    u32 in_pport, u32 in_pport_mask,
+				    __be16 eth_type, const u8 *eth_dst,
+				    const u8 *eth_dst_mask, __be16 vlan_id,
+				    __be16 vlan_id_mask, bool copy_to_cpu,
+				    int flags)
+{
+	struct rocker_flow_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	if (is_multicast_ether_addr(eth_dst)) {
+		entry->key.priority = ROCKER_PRIORITY_TERM_MAC_MCAST;
+		entry->key.term_mac.goto_tbl =
+			 ROCKER_OF_DPA_TABLE_ID_MULTICAST_ROUTING;
+	} else {
+		entry->key.priority = ROCKER_PRIORITY_TERM_MAC_UCAST;
+		entry->key.term_mac.goto_tbl =
+			 ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
+	}
+
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
+	entry->key.term_mac.in_pport = in_pport;
+	entry->key.term_mac.in_pport_mask = in_pport_mask;
+	entry->key.term_mac.eth_type = eth_type;
+	ether_addr_copy(entry->key.term_mac.eth_dst, eth_dst);
+	ether_addr_copy(entry->key.term_mac.eth_dst_mask, eth_dst_mask);
+	entry->key.term_mac.vlan_id = vlan_id;
+	entry->key.term_mac.vlan_id_mask = vlan_id_mask;
+	entry->key.term_mac.copy_to_cpu = copy_to_cpu;
+
+	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port,
+				  struct switchdev_trans *trans, int flags,
+				  const u8 *eth_dst, const u8 *eth_dst_mask,
+				  __be16 vlan_id, u32 tunnel_id,
+				  enum rocker_of_dpa_table_id goto_tbl,
+				  u32 group_id, bool copy_to_cpu)
+{
+	struct rocker_flow_tbl_entry *entry;
+	u32 priority;
+	bool vlan_bridging = !!vlan_id;
+	bool dflt = !eth_dst || (eth_dst && eth_dst_mask);
+	bool wild = false;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_BRIDGING;
+
+	if (eth_dst) {
+		entry->key.bridge.has_eth_dst = 1;
+		ether_addr_copy(entry->key.bridge.eth_dst, eth_dst);
+	}
+	if (eth_dst_mask) {
+		entry->key.bridge.has_eth_dst_mask = 1;
+		ether_addr_copy(entry->key.bridge.eth_dst_mask, eth_dst_mask);
+		if (!ether_addr_equal(eth_dst_mask, ff_mac))
+			wild = true;
+	}
+
+	priority = ROCKER_PRIORITY_UNKNOWN;
+	if (vlan_bridging && dflt && wild)
+		priority = ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_WILD;
+	else if (vlan_bridging && dflt && !wild)
+		priority = ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_EXACT;
+	else if (vlan_bridging && !dflt)
+		priority = ROCKER_PRIORITY_BRIDGING_VLAN;
+	else if (!vlan_bridging && dflt && wild)
+		priority = ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_WILD;
+	else if (!vlan_bridging && dflt && !wild)
+		priority = ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_EXACT;
+	else if (!vlan_bridging && !dflt)
+		priority = ROCKER_PRIORITY_BRIDGING_TENANT;
+
+	entry->key.priority = priority;
+	entry->key.bridge.vlan_id = vlan_id;
+	entry->key.bridge.tunnel_id = tunnel_id;
+	entry->key.bridge.goto_tbl = goto_tbl;
+	entry->key.bridge.group_id = group_id;
+	entry->key.bridge.copy_to_cpu = copy_to_cpu;
+
+	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port,
+					  struct switchdev_trans *trans,
+					  __be16 eth_type, __be32 dst,
+					  __be32 dst_mask, u32 priority,
+					  enum rocker_of_dpa_table_id goto_tbl,
+					  u32 group_id, int flags)
+{
+	struct rocker_flow_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
+	entry->key.priority = priority;
+	entry->key.ucast_routing.eth_type = eth_type;
+	entry->key.ucast_routing.dst4 = dst;
+	entry->key.ucast_routing.dst4_mask = dst_mask;
+	entry->key.ucast_routing.goto_tbl = goto_tbl;
+	entry->key.ucast_routing.group_id = group_id;
+	entry->key_len = offsetof(struct rocker_flow_tbl_key,
+				  ucast_routing.group_id);
+
+	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_flow_tbl_acl(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans, int flags,
+			       u32 in_pport, u32 in_pport_mask,
+			       const u8 *eth_src, const u8 *eth_src_mask,
+			       const u8 *eth_dst, const u8 *eth_dst_mask,
+			       __be16 eth_type, __be16 vlan_id,
+			       __be16 vlan_id_mask, u8 ip_proto,
+			       u8 ip_proto_mask, u8 ip_tos, u8 ip_tos_mask,
+			       u32 group_id)
+{
+	u32 priority;
+	struct rocker_flow_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	priority = ROCKER_PRIORITY_ACL_NORMAL;
+	if (eth_dst && eth_dst_mask) {
+		if (ether_addr_equal(eth_dst_mask, mcast_mac))
+			priority = ROCKER_PRIORITY_ACL_DFLT;
+		else if (is_link_local_ether_addr(eth_dst))
+			priority = ROCKER_PRIORITY_ACL_CTRL;
+	}
+
+	entry->key.priority = priority;
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	entry->key.acl.in_pport = in_pport;
+	entry->key.acl.in_pport_mask = in_pport_mask;
+
+	if (eth_src)
+		ether_addr_copy(entry->key.acl.eth_src, eth_src);
+	if (eth_src_mask)
+		ether_addr_copy(entry->key.acl.eth_src_mask, eth_src_mask);
+	if (eth_dst)
+		ether_addr_copy(entry->key.acl.eth_dst, eth_dst);
+	if (eth_dst_mask)
+		ether_addr_copy(entry->key.acl.eth_dst_mask, eth_dst_mask);
+
+	entry->key.acl.eth_type = eth_type;
+	entry->key.acl.vlan_id = vlan_id;
+	entry->key.acl.vlan_id_mask = vlan_id_mask;
+	entry->key.acl.ip_proto = ip_proto;
+	entry->key.acl.ip_proto_mask = ip_proto_mask;
+	entry->key.acl.ip_tos = ip_tos;
+	entry->key.acl.ip_tos_mask = ip_tos_mask;
+	entry->key.acl.group_id = group_id;
+
+	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static struct rocker_group_tbl_entry *
+rocker_group_tbl_find(const struct rocker *rocker,
+		      const struct rocker_group_tbl_entry *match)
+{
+	struct rocker_group_tbl_entry *found;
+
+	hash_for_each_possible(rocker->group_tbl, found,
+			       entry, match->group_id) {
+		if (found->group_id == match->group_id)
+			return found;
+	}
+
+	return NULL;
+}
+
+static void rocker_group_tbl_entry_free(struct switchdev_trans *trans,
+					struct rocker_group_tbl_entry *entry)
+{
+	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
+		rocker_kfree(trans, entry->group_ids);
+		break;
+	default:
+		break;
+	}
+	rocker_kfree(trans, entry);
+}
+
+static int rocker_group_tbl_add(struct rocker_port *rocker_port,
+				struct switchdev_trans *trans, int flags,
+				struct rocker_group_tbl_entry *match)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_group_tbl_entry *found;
+	unsigned long lock_flags;
+
+	spin_lock_irqsave(&rocker->group_tbl_lock, lock_flags);
+
+	found = rocker_group_tbl_find(rocker, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		rocker_group_tbl_entry_free(trans, found);
+		found = match;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD;
+	} else {
+		found = match;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD;
+	}
+
+	if (!switchdev_trans_ph_prepare(trans))
+		hash_add(rocker->group_tbl, &found->entry, found->group_id);
+
+	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
+
+	return rocker_cmd_exec(rocker_port, trans, flags,
+			       rocker_cmd_group_tbl_add, found, NULL, NULL);
+}
+
+static int rocker_group_tbl_del(struct rocker_port *rocker_port,
+				struct switchdev_trans *trans, int flags,
+				struct rocker_group_tbl_entry *match)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_group_tbl_entry *found;
+	unsigned long lock_flags;
+	int err = 0;
+
+	spin_lock_irqsave(&rocker->group_tbl_lock, lock_flags);
+
+	found = rocker_group_tbl_find(rocker, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL;
+	}
+
+	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
+
+	rocker_group_tbl_entry_free(trans, match);
+
+	if (found) {
+		err = rocker_cmd_exec(rocker_port, trans, flags,
+				      rocker_cmd_group_tbl_del,
+				      found, NULL, NULL);
+		rocker_group_tbl_entry_free(trans, found);
+	}
+
+	return err;
+}
+
+static int rocker_group_tbl_do(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans, int flags,
+			       struct rocker_group_tbl_entry *entry)
+{
+	if (flags & ROCKER_OP_FLAG_REMOVE)
+		return rocker_group_tbl_del(rocker_port, trans, flags, entry);
+	else
+		return rocker_group_tbl_add(rocker_port, trans, flags, entry);
+}
+
+static int rocker_group_l2_interface(struct rocker_port *rocker_port,
+				     struct switchdev_trans *trans, int flags,
+				     __be16 vlan_id, u32 out_pport,
+				     int pop_vlan)
+{
+	struct rocker_group_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
+	entry->l2_interface.pop_vlan = pop_vlan;
+
+	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_group_l2_fan_out(struct rocker_port *rocker_port,
+				   struct switchdev_trans *trans,
+				   int flags, u8 group_count,
+				   const u32 *group_ids, u32 group_id)
+{
+	struct rocker_group_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->group_id = group_id;
+	entry->group_count = group_count;
+
+	entry->group_ids = rocker_kcalloc(trans, flags,
+					  group_count, sizeof(u32));
+	if (!entry->group_ids) {
+		rocker_kfree(trans, entry);
+		return -ENOMEM;
+	}
+	memcpy(entry->group_ids, group_ids, group_count * sizeof(u32));
+
+	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static int rocker_group_l2_flood(struct rocker_port *rocker_port,
+				 struct switchdev_trans *trans, int flags,
+				 __be16 vlan_id, u8 group_count,
+				 const u32 *group_ids, u32 group_id)
+{
+	return rocker_group_l2_fan_out(rocker_port, trans, flags,
+				       group_count, group_ids,
+				       group_id);
+}
+
+static int rocker_group_l3_unicast(struct rocker_port *rocker_port,
+				   struct switchdev_trans *trans, int flags,
+				   u32 index, const u8 *src_mac, const u8 *dst_mac,
+				   __be16 vlan_id, bool ttl_check, u32 pport)
+{
+	struct rocker_group_tbl_entry *entry;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->group_id = ROCKER_GROUP_L3_UNICAST(index);
+	if (src_mac)
+		ether_addr_copy(entry->l3_unicast.eth_src, src_mac);
+	if (dst_mac)
+		ether_addr_copy(entry->l3_unicast.eth_dst, dst_mac);
+	entry->l3_unicast.vlan_id = vlan_id;
+	entry->l3_unicast.ttl_check = ttl_check;
+	entry->l3_unicast.group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, pport);
+
+	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
+}
+
+static struct rocker_neigh_tbl_entry *
+rocker_neigh_tbl_find(const struct rocker *rocker, __be32 ip_addr)
+{
+	struct rocker_neigh_tbl_entry *found;
+
+	hash_for_each_possible(rocker->neigh_tbl, found,
+			       entry, be32_to_cpu(ip_addr))
+		if (found->ip_addr == ip_addr)
+			return found;
+
+	return NULL;
+}
+
+static void _rocker_neigh_add(struct rocker *rocker,
+			      struct switchdev_trans *trans,
+			      struct rocker_neigh_tbl_entry *entry)
+{
+	if (!switchdev_trans_ph_commit(trans))
+		entry->index = rocker->neigh_tbl_next_index++;
+	if (switchdev_trans_ph_prepare(trans))
+		return;
+	entry->ref_count++;
+	hash_add(rocker->neigh_tbl, &entry->entry,
+		 be32_to_cpu(entry->ip_addr));
+}
+
+static void _rocker_neigh_del(struct switchdev_trans *trans,
+			      struct rocker_neigh_tbl_entry *entry)
+{
+	if (switchdev_trans_ph_prepare(trans))
+		return;
+	if (--entry->ref_count == 0) {
+		hash_del(&entry->entry);
+		rocker_kfree(trans, entry);
+	}
+}
+
+static void _rocker_neigh_update(struct rocker_neigh_tbl_entry *entry,
+				 struct switchdev_trans *trans,
+				 const u8 *eth_dst, bool ttl_check)
+{
+	if (eth_dst) {
+		ether_addr_copy(entry->eth_dst, eth_dst);
+		entry->ttl_check = ttl_check;
+	} else if (!switchdev_trans_ph_prepare(trans)) {
+		entry->ref_count++;
+	}
+}
+
+static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port,
+				  struct switchdev_trans *trans,
+				  int flags, __be32 ip_addr, const u8 *eth_dst)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_neigh_tbl_entry *entry;
+	struct rocker_neigh_tbl_entry *found;
+	unsigned long lock_flags;
+	__be16 eth_type = htons(ETH_P_IP);
+	enum rocker_of_dpa_table_id goto_tbl =
+		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 group_id;
+	u32 priority = 0;
+	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
+	bool updating;
+	bool removing;
+	int err = 0;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&rocker->neigh_tbl_lock, lock_flags);
+
+	found = rocker_neigh_tbl_find(rocker, ip_addr);
+
+	updating = found && adding;
+	removing = found && !adding;
+	adding = !found && adding;
+
+	if (adding) {
+		entry->ip_addr = ip_addr;
+		entry->dev = rocker_port->dev;
+		ether_addr_copy(entry->eth_dst, eth_dst);
+		entry->ttl_check = true;
+		_rocker_neigh_add(rocker, trans, entry);
+	} else if (removing) {
+		memcpy(entry, found, sizeof(*entry));
+		_rocker_neigh_del(trans, found);
+	} else if (updating) {
+		_rocker_neigh_update(found, trans, eth_dst, true);
+		memcpy(entry, found, sizeof(*entry));
+	} else {
+		err = -ENOENT;
+	}
+
+	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
+
+	if (err)
+		goto err_out;
+
+	/* For each active neighbor, we have an L3 unicast group and
+	 * a /32 route to the neighbor, which uses the L3 unicast
+	 * group.  The L3 unicast group can also be referred to by
+	 * other routes' nexthops.
+	 */
+
+	err = rocker_group_l3_unicast(rocker_port, trans, flags,
+				      entry->index,
+				      rocker_port->dev->dev_addr,
+				      entry->eth_dst,
+				      rocker_port->internal_vlan_id,
+				      entry->ttl_check,
+				      rocker_port->pport);
+	if (err) {
+		netdev_err(rocker_port->dev,
+			   "Error (%d) L3 unicast group index %d\n",
+			   err, entry->index);
+		goto err_out;
+	}
+
+	if (adding || removing) {
+		group_id = ROCKER_GROUP_L3_UNICAST(entry->index);
+		err = rocker_flow_tbl_ucast4_routing(rocker_port, trans,
+						     eth_type, ip_addr,
+						     inet_make_mask(32),
+						     priority, goto_tbl,
+						     group_id, flags);
+
+		if (err)
+			netdev_err(rocker_port->dev,
+				   "Error (%d) /32 unicast route %pI4 group 0x%08x\n",
+				   err, &entry->ip_addr, group_id);
+	}
+
+err_out:
+	if (!adding)
+		rocker_kfree(trans, entry);
+
+	return err;
+}
+
+static int rocker_port_ipv4_resolve(struct rocker_port *rocker_port,
+				    struct switchdev_trans *trans,
+				    __be32 ip_addr)
+{
+	struct net_device *dev = rocker_port->dev;
+	struct neighbour *n = __ipv4_neigh_lookup(dev, (__force u32)ip_addr);
+	int err = 0;
+
+	if (!n) {
+		n = neigh_create(&arp_tbl, &ip_addr, dev);
+		if (IS_ERR(n))
+			return IS_ERR(n);
+	}
+
+	/* If the neigh is already resolved, then go ahead and
+	 * install the entry, otherwise start the ARP process to
+	 * resolve the neigh.
+	 */
+
+	if (n->nud_state & NUD_VALID)
+		err = rocker_port_ipv4_neigh(rocker_port, trans, 0,
+					     ip_addr, n->ha);
+	else
+		neigh_event_send(n, NULL);
+
+	neigh_release(n);
+	return err;
+}
+
+static int rocker_port_ipv4_nh(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans, int flags,
+			       __be32 ip_addr, u32 *index)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_neigh_tbl_entry *entry;
+	struct rocker_neigh_tbl_entry *found;
+	unsigned long lock_flags;
+	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
+	bool updating;
+	bool removing;
+	bool resolved = true;
+	int err = 0;
+
+	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&rocker->neigh_tbl_lock, lock_flags);
+
+	found = rocker_neigh_tbl_find(rocker, ip_addr);
+	if (found)
+		*index = found->index;
+
+	updating = found && adding;
+	removing = found && !adding;
+	adding = !found && adding;
+
+	if (adding) {
+		entry->ip_addr = ip_addr;
+		entry->dev = rocker_port->dev;
+		_rocker_neigh_add(rocker, trans, entry);
+		*index = entry->index;
+		resolved = false;
+	} else if (removing) {
+		_rocker_neigh_del(trans, found);
+	} else if (updating) {
+		_rocker_neigh_update(found, trans, NULL, false);
+		resolved = !is_zero_ether_addr(found->eth_dst);
+	} else {
+		err = -ENOENT;
+	}
+
+	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
+
+	if (!adding)
+		rocker_kfree(trans, entry);
+
+	if (err)
+		return err;
+
+	/* Resolved means neigh ip_addr is resolved to neigh mac. */
+
+	if (!resolved)
+		err = rocker_port_ipv4_resolve(rocker_port, trans, ip_addr);
+
+	return err;
+}
+
+static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port,
+					struct switchdev_trans *trans,
+					int flags, __be16 vlan_id)
+{
+	struct rocker_port *p;
+	const struct rocker *rocker = rocker_port->rocker;
+	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
+	u32 *group_ids;
+	u8 group_count = 0;
+	int err = 0;
+	int i;
+
+	group_ids = rocker_kcalloc(trans, flags,
+				   rocker->port_count, sizeof(u32));
+	if (!group_ids)
+		return -ENOMEM;
+
+	/* Adjust the flood group for this VLAN.  The flood group
+	 * references an L2 interface group for each port in this
+	 * VLAN.
+	 */
+
+	for (i = 0; i < rocker->port_count; i++) {
+		p = rocker->ports[i];
+		if (!p)
+			continue;
+		if (!rocker_port_is_bridged(p))
+			continue;
+		if (test_bit(ntohs(vlan_id), p->vlan_bitmap)) {
+			group_ids[group_count++] =
+				ROCKER_GROUP_L2_INTERFACE(vlan_id, p->pport);
+		}
+	}
+
+	/* If there are no bridged ports in this VLAN, we're done */
+	if (group_count == 0)
+		goto no_ports_in_vlan;
+
+	err = rocker_group_l2_flood(rocker_port, trans, flags, vlan_id,
+				    group_count, group_ids, group_id);
+	if (err)
+		netdev_err(rocker_port->dev,
+			   "Error (%d) port VLAN l2 flood group\n", err);
+
+no_ports_in_vlan:
+	rocker_kfree(trans, group_ids);
+	return err;
+}
+
+static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port,
+				      struct switchdev_trans *trans, int flags,
+				      __be16 vlan_id, bool pop_vlan)
+{
+	const struct rocker *rocker = rocker_port->rocker;
+	struct rocker_port *p;
+	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
+	u32 out_pport;
+	int ref = 0;
+	int err;
+	int i;
+
+	/* An L2 interface group for this port in this VLAN, but
+	 * only when port STP state is LEARNING|FORWARDING.
+	 */
+
+	if (rocker_port->stp_state == BR_STATE_LEARNING ||
+	    rocker_port->stp_state == BR_STATE_FORWARDING) {
+		out_pport = rocker_port->pport;
+		err = rocker_group_l2_interface(rocker_port, trans, flags,
+						vlan_id, out_pport, pop_vlan);
+		if (err) {
+			netdev_err(rocker_port->dev,
+				   "Error (%d) port VLAN l2 group for pport %d\n",
+				   err, out_pport);
+			return err;
+		}
+	}
+
+	/* An L2 interface group for this VLAN to CPU port.
+	 * Add when first port joins this VLAN and destroy when
+	 * last port leaves this VLAN.
+	 */
+
+	for (i = 0; i < rocker->port_count; i++) {
+		p = rocker->ports[i];
+		if (p && test_bit(ntohs(vlan_id), p->vlan_bitmap))
+			ref++;
+	}
+
+	if ((!adding || ref != 1) && (adding || ref != 0))
+		return 0;
+
+	out_pport = 0;
+	err = rocker_group_l2_interface(rocker_port, trans, flags,
+					vlan_id, out_pport, pop_vlan);
+	if (err) {
+		netdev_err(rocker_port->dev,
+			   "Error (%d) port VLAN l2 group for CPU port\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static struct rocker_ctrl {
+	const u8 *eth_dst;
+	const u8 *eth_dst_mask;
+	__be16 eth_type;
+	bool acl;
+	bool bridge;
+	bool term;
+	bool copy_to_cpu;
+} rocker_ctrls[] = {
+	[ROCKER_CTRL_LINK_LOCAL_MCAST] = {
+		/* pass link local multicast pkts up to CPU for filtering */
+		.eth_dst = ll_mac,
+		.eth_dst_mask = ll_mask,
+		.acl = true,
+	},
+	[ROCKER_CTRL_LOCAL_ARP] = {
+		/* pass local ARP pkts up to CPU */
+		.eth_dst = zero_mac,
+		.eth_dst_mask = zero_mac,
+		.eth_type = htons(ETH_P_ARP),
+		.acl = true,
+	},
+	[ROCKER_CTRL_IPV4_MCAST] = {
+		/* pass IPv4 mcast pkts up to CPU, RFC 1112 */
+		.eth_dst = ipv4_mcast,
+		.eth_dst_mask = ipv4_mask,
+		.eth_type = htons(ETH_P_IP),
+		.term  = true,
+		.copy_to_cpu = true,
+	},
+	[ROCKER_CTRL_IPV6_MCAST] = {
+		/* pass IPv6 mcast pkts up to CPU, RFC 2464 */
+		.eth_dst = ipv6_mcast,
+		.eth_dst_mask = ipv6_mask,
+		.eth_type = htons(ETH_P_IPV6),
+		.term  = true,
+		.copy_to_cpu = true,
+	},
+	[ROCKER_CTRL_DFLT_BRIDGING] = {
+		/* flood any pkts on vlan */
+		.bridge = true,
+		.copy_to_cpu = true,
+	},
+	[ROCKER_CTRL_DFLT_OVS] = {
+		/* pass all pkts up to CPU */
+		.eth_dst = zero_mac,
+		.eth_dst_mask = zero_mac,
+		.acl = true,
+	},
+};
+
+static int rocker_port_ctrl_vlan_acl(struct rocker_port *rocker_port,
+				     struct switchdev_trans *trans, int flags,
+				     const struct rocker_ctrl *ctrl, __be16 vlan_id)
+{
+	u32 in_pport = rocker_port->pport;
+	u32 in_pport_mask = 0xffffffff;
+	u32 out_pport = 0;
+	const u8 *eth_src = NULL;
+	const u8 *eth_src_mask = NULL;
+	__be16 vlan_id_mask = htons(0xffff);
+	u8 ip_proto = 0;
+	u8 ip_proto_mask = 0;
+	u8 ip_tos = 0;
+	u8 ip_tos_mask = 0;
+	u32 group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
+	int err;
+
+	err = rocker_flow_tbl_acl(rocker_port, trans, flags,
+				  in_pport, in_pport_mask,
+				  eth_src, eth_src_mask,
+				  ctrl->eth_dst, ctrl->eth_dst_mask,
+				  ctrl->eth_type,
+				  vlan_id, vlan_id_mask,
+				  ip_proto, ip_proto_mask,
+				  ip_tos, ip_tos_mask,
+				  group_id);
+
+	if (err)
+		netdev_err(rocker_port->dev, "Error (%d) ctrl ACL\n", err);
+
+	return err;
+}
+
+static int rocker_port_ctrl_vlan_bridge(struct rocker_port *rocker_port,
+					struct switchdev_trans *trans,
+					int flags,
+					const struct rocker_ctrl *ctrl,
+					__be16 vlan_id)
+{
+	enum rocker_of_dpa_table_id goto_tbl =
+		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
+	u32 tunnel_id = 0;
+	int err;
+
+	if (!rocker_port_is_bridged(rocker_port))
+		return 0;
+
+	err = rocker_flow_tbl_bridge(rocker_port, trans, flags,
+				     ctrl->eth_dst, ctrl->eth_dst_mask,
+				     vlan_id, tunnel_id,
+				     goto_tbl, group_id, ctrl->copy_to_cpu);
+
+	if (err)
+		netdev_err(rocker_port->dev, "Error (%d) ctrl FLOOD\n", err);
+
+	return err;
+}
+
+static int rocker_port_ctrl_vlan_term(struct rocker_port *rocker_port,
+				      struct switchdev_trans *trans, int flags,
+				      const struct rocker_ctrl *ctrl, __be16 vlan_id)
+{
+	u32 in_pport_mask = 0xffffffff;
+	__be16 vlan_id_mask = htons(0xffff);
+	int err;
+
+	if (ntohs(vlan_id) == 0)
+		vlan_id = rocker_port->internal_vlan_id;
+
+	err = rocker_flow_tbl_term_mac(rocker_port, trans,
+				       rocker_port->pport, in_pport_mask,
+				       ctrl->eth_type, ctrl->eth_dst,
+				       ctrl->eth_dst_mask, vlan_id,
+				       vlan_id_mask, ctrl->copy_to_cpu,
+				       flags);
+
+	if (err)
+		netdev_err(rocker_port->dev, "Error (%d) ctrl term\n", err);
+
+	return err;
+}
+
+static int rocker_port_ctrl_vlan(struct rocker_port *rocker_port,
+				 struct switchdev_trans *trans, int flags,
+				 const struct rocker_ctrl *ctrl, __be16 vlan_id)
+{
+	if (ctrl->acl)
+		return rocker_port_ctrl_vlan_acl(rocker_port, trans, flags,
+						 ctrl, vlan_id);
+	if (ctrl->bridge)
+		return rocker_port_ctrl_vlan_bridge(rocker_port, trans, flags,
+						    ctrl, vlan_id);
+
+	if (ctrl->term)
+		return rocker_port_ctrl_vlan_term(rocker_port, trans, flags,
+						  ctrl, vlan_id);
+
+	return -EOPNOTSUPP;
+}
+
+static int rocker_port_ctrl_vlan_add(struct rocker_port *rocker_port,
+				     struct switchdev_trans *trans, int flags,
+				     __be16 vlan_id)
+{
+	int err = 0;
+	int i;
+
+	for (i = 0; i < ROCKER_CTRL_MAX; i++) {
+		if (rocker_port->ctrls[i]) {
+			err = rocker_port_ctrl_vlan(rocker_port, trans, flags,
+						    &rocker_ctrls[i], vlan_id);
+			if (err)
+				return err;
+		}
+	}
+
+	return err;
+}
+
+static int rocker_port_ctrl(struct rocker_port *rocker_port,
+			    struct switchdev_trans *trans, int flags,
+			    const struct rocker_ctrl *ctrl)
+{
+	u16 vid;
+	int err = 0;
+
+	for (vid = 1; vid < VLAN_N_VID; vid++) {
+		if (!test_bit(vid, rocker_port->vlan_bitmap))
+			continue;
+		err = rocker_port_ctrl_vlan(rocker_port, trans, flags,
+					    ctrl, htons(vid));
+		if (err)
+			break;
+	}
+
+	return err;
+}
+
+static int rocker_port_vlan(struct rocker_port *rocker_port,
+			    struct switchdev_trans *trans, int flags, u16 vid)
+{
+	enum rocker_of_dpa_table_id goto_tbl =
+		ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
+	u32 in_pport = rocker_port->pport;
+	__be16 vlan_id = htons(vid);
+	__be16 vlan_id_mask = htons(0xffff);
+	__be16 internal_vlan_id;
+	bool untagged;
+	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
+	int err;
+
+	internal_vlan_id = rocker_port_vid_to_vlan(rocker_port, vid, &untagged);
+
+	if (adding && test_bit(ntohs(internal_vlan_id),
+			       rocker_port->vlan_bitmap))
+		return 0; /* already added */
+	else if (!adding && !test_bit(ntohs(internal_vlan_id),
+				      rocker_port->vlan_bitmap))
+		return 0; /* already removed */
+
+	change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap);
+
+	if (adding) {
+		err = rocker_port_ctrl_vlan_add(rocker_port, trans, flags,
+						internal_vlan_id);
+		if (err) {
+			netdev_err(rocker_port->dev,
+				   "Error (%d) port ctrl vlan add\n", err);
+			goto err_out;
+		}
+	}
+
+	err = rocker_port_vlan_l2_groups(rocker_port, trans, flags,
+					 internal_vlan_id, untagged);
+	if (err) {
+		netdev_err(rocker_port->dev,
+			   "Error (%d) port VLAN l2 groups\n", err);
+		goto err_out;
+	}
+
+	err = rocker_port_vlan_flood_group(rocker_port, trans, flags,
+					   internal_vlan_id);
+	if (err) {
+		netdev_err(rocker_port->dev,
+			   "Error (%d) port VLAN l2 flood group\n", err);
+		goto err_out;
+	}
+
+	err = rocker_flow_tbl_vlan(rocker_port, trans, flags,
+				   in_pport, vlan_id, vlan_id_mask,
+				   goto_tbl, untagged, internal_vlan_id);
+	if (err)
+		netdev_err(rocker_port->dev,
+			   "Error (%d) port VLAN table\n", err);
+
+err_out:
+	if (switchdev_trans_ph_prepare(trans))
+		change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap);
+
+	return err;
+}
+
+static int rocker_port_ig_tbl(struct rocker_port *rocker_port,
+			      struct switchdev_trans *trans, int flags)
+{
+	enum rocker_of_dpa_table_id goto_tbl;
+	u32 in_pport;
+	u32 in_pport_mask;
+	int err;
+
+	/* Normal Ethernet Frames.  Matches pkts from any local physical
+	 * ports.  Goto VLAN tbl.
+	 */
+
+	in_pport = 0;
+	in_pport_mask = 0xffff0000;
+	goto_tbl = ROCKER_OF_DPA_TABLE_ID_VLAN;
+
+	err = rocker_flow_tbl_ig_port(rocker_port, trans, flags,
+				      in_pport, in_pport_mask,
+				      goto_tbl);
+	if (err)
+		netdev_err(rocker_port->dev,
+			   "Error (%d) ingress port table entry\n", err);
+
+	return err;
+}
+
+struct rocker_fdb_learn_work {
+	struct work_struct work;
+	struct rocker_port *rocker_port;
+	struct switchdev_trans *trans;
+	int flags;
+	u8 addr[ETH_ALEN];
+	u16 vid;
+};
+
+static void rocker_port_fdb_learn_work(struct work_struct *work)
+{
+	const struct rocker_fdb_learn_work *lw =
+		container_of(work, struct rocker_fdb_learn_work, work);
+	bool removing = (lw->flags & ROCKER_OP_FLAG_REMOVE);
+	bool learned = (lw->flags & ROCKER_OP_FLAG_LEARNED);
+	struct switchdev_notifier_fdb_info info;
+
+	info.addr = lw->addr;
+	info.vid = lw->vid;
+
+	rtnl_lock();
+	if (learned && removing)
+		call_switchdev_notifiers(SWITCHDEV_FDB_DEL,
+					 lw->rocker_port->dev, &info.info);
+	else if (learned && !removing)
+		call_switchdev_notifiers(SWITCHDEV_FDB_ADD,
+					 lw->rocker_port->dev, &info.info);
+	rtnl_unlock();
+
+	rocker_kfree(lw->trans, work);
+}
+
+static int rocker_port_fdb_learn(struct rocker_port *rocker_port,
+				 struct switchdev_trans *trans, int flags,
+				 const u8 *addr, __be16 vlan_id)
+{
+	struct rocker_fdb_learn_work *lw;
+	enum rocker_of_dpa_table_id goto_tbl =
+		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 out_pport = rocker_port->pport;
+	u32 tunnel_id = 0;
+	u32 group_id = ROCKER_GROUP_NONE;
+	bool syncing = !!(rocker_port->brport_flags & BR_LEARNING_SYNC);
+	bool copy_to_cpu = false;
+	int err;
+
+	if (rocker_port_is_bridged(rocker_port))
+		group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
+
+	if (!(flags & ROCKER_OP_FLAG_REFRESH)) {
+		err = rocker_flow_tbl_bridge(rocker_port, trans, flags, addr,
+					     NULL, vlan_id, tunnel_id, goto_tbl,
+					     group_id, copy_to_cpu);
+		if (err)
+			return err;
+	}
+
+	if (!syncing)
+		return 0;
+
+	if (!rocker_port_is_bridged(rocker_port))
+		return 0;
+
+	lw = rocker_kzalloc(trans, flags, sizeof(*lw));
+	if (!lw)
+		return -ENOMEM;
+
+	INIT_WORK(&lw->work, rocker_port_fdb_learn_work);
+
+	lw->rocker_port = rocker_port;
+	lw->trans = trans;
+	lw->flags = flags;
+	ether_addr_copy(lw->addr, addr);
+	lw->vid = rocker_port_vlan_to_vid(rocker_port, vlan_id);
+
+	if (switchdev_trans_ph_prepare(trans))
+		rocker_kfree(trans, lw);
+	else
+		schedule_work(&lw->work);
+
+	return 0;
+}
+
+static struct rocker_fdb_tbl_entry *
+rocker_fdb_tbl_find(const struct rocker *rocker,
+		    const struct rocker_fdb_tbl_entry *match)
+{
+	struct rocker_fdb_tbl_entry *found;
+
+	hash_for_each_possible(rocker->fdb_tbl, found, entry, match->key_crc32)
+		if (memcmp(&found->key, &match->key, sizeof(found->key)) == 0)
+			return found;
+
+	return NULL;
+}
+
+static int rocker_port_fdb(struct rocker_port *rocker_port,
+			   struct switchdev_trans *trans,
+			   const unsigned char *addr,
+			   __be16 vlan_id, int flags)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_fdb_tbl_entry *fdb;
+	struct rocker_fdb_tbl_entry *found;
+	bool removing = (flags & ROCKER_OP_FLAG_REMOVE);
+	unsigned long lock_flags;
+
+	fdb = rocker_kzalloc(trans, flags, sizeof(*fdb));
+	if (!fdb)
+		return -ENOMEM;
+
+	fdb->learned = (flags & ROCKER_OP_FLAG_LEARNED);
+	fdb->touched = jiffies;
+	fdb->key.rocker_port = rocker_port;
+	ether_addr_copy(fdb->key.addr, addr);
+	fdb->key.vlan_id = vlan_id;
+	fdb->key_crc32 = crc32(~0, &fdb->key, sizeof(fdb->key));
+
+	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
+
+	found = rocker_fdb_tbl_find(rocker, fdb);
+
+	if (found) {
+		found->touched = jiffies;
+		if (removing) {
+			rocker_kfree(trans, fdb);
+			if (!switchdev_trans_ph_prepare(trans))
+				hash_del(&found->entry);
+		}
+	} else if (!removing) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_add(rocker->fdb_tbl, &fdb->entry,
+				 fdb->key_crc32);
+	}
+
+	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
+
+	/* Check if adding and already exists, or removing and can't find */
+	if (!found != !removing) {
+		rocker_kfree(trans, fdb);
+		if (!found && removing)
+			return 0;
+		/* Refreshing existing to update aging timers */
+		flags |= ROCKER_OP_FLAG_REFRESH;
+	}
+
+	return rocker_port_fdb_learn(rocker_port, trans, flags, addr, vlan_id);
+}
+
+static int rocker_port_fdb_flush(struct rocker_port *rocker_port,
+				 struct switchdev_trans *trans, int flags)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_fdb_tbl_entry *found;
+	unsigned long lock_flags;
+	struct hlist_node *tmp;
+	int bkt;
+	int err = 0;
+
+	if (rocker_port->stp_state == BR_STATE_LEARNING ||
+	    rocker_port->stp_state == BR_STATE_FORWARDING)
+		return 0;
+
+	flags |= ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE;
+
+	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
+
+	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) {
+		if (found->key.rocker_port != rocker_port)
+			continue;
+		if (!found->learned)
+			continue;
+		err = rocker_port_fdb_learn(rocker_port, trans, flags,
+					    found->key.addr,
+					    found->key.vlan_id);
+		if (err)
+			goto err_out;
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+	}
+
+err_out:
+	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
+
+	return err;
+}
+
+static void rocker_fdb_cleanup(unsigned long data)
+{
+	struct rocker *rocker = (struct rocker *)data;
+	struct rocker_port *rocker_port;
+	struct rocker_fdb_tbl_entry *entry;
+	struct hlist_node *tmp;
+	unsigned long next_timer = jiffies + BR_MIN_AGEING_TIME;
+	unsigned long expires;
+	unsigned long lock_flags;
+	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE |
+		    ROCKER_OP_FLAG_LEARNED;
+	int bkt;
+
+	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
+
+	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, entry, entry) {
+		if (!entry->learned)
+			continue;
+		rocker_port = entry->key.rocker_port;
+		expires = entry->touched + rocker_port->ageing_time;
+		if (time_before_eq(expires, jiffies)) {
+			rocker_port_fdb_learn(rocker_port, NULL,
+					      flags, entry->key.addr,
+					      entry->key.vlan_id);
+			hash_del(&entry->entry);
+		} else if (time_before(expires, next_timer)) {
+			next_timer = expires;
+		}
+	}
+
+	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
+
+	mod_timer(&rocker->fdb_cleanup_timer, round_jiffies_up(next_timer));
+}
+
+static int rocker_port_router_mac(struct rocker_port *rocker_port,
+				  struct switchdev_trans *trans, int flags,
+				  __be16 vlan_id)
+{
+	u32 in_pport_mask = 0xffffffff;
+	__be16 eth_type;
+	const u8 *dst_mac_mask = ff_mac;
+	__be16 vlan_id_mask = htons(0xffff);
+	bool copy_to_cpu = false;
+	int err;
+
+	if (ntohs(vlan_id) == 0)
+		vlan_id = rocker_port->internal_vlan_id;
+
+	eth_type = htons(ETH_P_IP);
+	err = rocker_flow_tbl_term_mac(rocker_port, trans,
+				       rocker_port->pport, in_pport_mask,
+				       eth_type, rocker_port->dev->dev_addr,
+				       dst_mac_mask, vlan_id, vlan_id_mask,
+				       copy_to_cpu, flags);
+	if (err)
+		return err;
+
+	eth_type = htons(ETH_P_IPV6);
+	err = rocker_flow_tbl_term_mac(rocker_port, trans,
+				       rocker_port->pport, in_pport_mask,
+				       eth_type, rocker_port->dev->dev_addr,
+				       dst_mac_mask, vlan_id, vlan_id_mask,
+				       copy_to_cpu, flags);
+
+	return err;
+}
+
+static int rocker_port_fwding(struct rocker_port *rocker_port,
+			      struct switchdev_trans *trans, int flags)
+{
+	bool pop_vlan;
+	u32 out_pport;
+	__be16 vlan_id;
+	u16 vid;
+	int err;
+
+	/* Port will be forwarding-enabled if its STP state is LEARNING
+	 * or FORWARDING.  Traffic from CPU can still egress, regardless of
+	 * port STP state.  Use L2 interface group on port VLANs as a way
+	 * to toggle port forwarding: if forwarding is disabled, L2
+	 * interface group will not exist.
+	 */
+
+	if (rocker_port->stp_state != BR_STATE_LEARNING &&
+	    rocker_port->stp_state != BR_STATE_FORWARDING)
+		flags |= ROCKER_OP_FLAG_REMOVE;
+
+	out_pport = rocker_port->pport;
+	for (vid = 1; vid < VLAN_N_VID; vid++) {
+		if (!test_bit(vid, rocker_port->vlan_bitmap))
+			continue;
+		vlan_id = htons(vid);
+		pop_vlan = rocker_vlan_id_is_internal(vlan_id);
+		err = rocker_group_l2_interface(rocker_port, trans, flags,
+						vlan_id, out_pport, pop_vlan);
+		if (err) {
+			netdev_err(rocker_port->dev,
+				   "Error (%d) port VLAN l2 group for pport %d\n",
+				   err, out_pport);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int rocker_port_stp_update(struct rocker_port *rocker_port,
+				  struct switchdev_trans *trans, int flags,
+				  u8 state)
+{
+	bool want[ROCKER_CTRL_MAX] = { 0, };
+	bool prev_ctrls[ROCKER_CTRL_MAX];
+	u8 uninitialized_var(prev_state);
+	int err;
+	int i;
+
+	if (switchdev_trans_ph_prepare(trans)) {
+		memcpy(prev_ctrls, rocker_port->ctrls, sizeof(prev_ctrls));
+		prev_state = rocker_port->stp_state;
+	}
+
+	if (rocker_port->stp_state == state)
+		return 0;
+
+	rocker_port->stp_state = state;
+
+	switch (state) {
+	case BR_STATE_DISABLED:
+		/* port is completely disabled */
+		break;
+	case BR_STATE_LISTENING:
+	case BR_STATE_BLOCKING:
+		want[ROCKER_CTRL_LINK_LOCAL_MCAST] = true;
+		break;
+	case BR_STATE_LEARNING:
+	case BR_STATE_FORWARDING:
+		if (!rocker_port_is_ovsed(rocker_port))
+			want[ROCKER_CTRL_LINK_LOCAL_MCAST] = true;
+		want[ROCKER_CTRL_IPV4_MCAST] = true;
+		want[ROCKER_CTRL_IPV6_MCAST] = true;
+		if (rocker_port_is_bridged(rocker_port))
+			want[ROCKER_CTRL_DFLT_BRIDGING] = true;
+		else if (rocker_port_is_ovsed(rocker_port))
+			want[ROCKER_CTRL_DFLT_OVS] = true;
+		else
+			want[ROCKER_CTRL_LOCAL_ARP] = true;
+		break;
+	}
+
+	for (i = 0; i < ROCKER_CTRL_MAX; i++) {
+		if (want[i] != rocker_port->ctrls[i]) {
+			int ctrl_flags = flags |
+					 (want[i] ? 0 : ROCKER_OP_FLAG_REMOVE);
+			err = rocker_port_ctrl(rocker_port, trans, ctrl_flags,
+					       &rocker_ctrls[i]);
+			if (err)
+				goto err_out;
+			rocker_port->ctrls[i] = want[i];
+		}
+	}
+
+	err = rocker_port_fdb_flush(rocker_port, trans, flags);
+	if (err)
+		goto err_out;
+
+	err = rocker_port_fwding(rocker_port, trans, flags);
+
+err_out:
+	if (switchdev_trans_ph_prepare(trans)) {
+		memcpy(rocker_port->ctrls, prev_ctrls, sizeof(prev_ctrls));
+		rocker_port->stp_state = prev_state;
+	}
+
+	return err;
+}
+
+static int rocker_port_fwd_enable(struct rocker_port *rocker_port,
+				  struct switchdev_trans *trans, int flags)
+{
+	if (rocker_port_is_bridged(rocker_port))
+		/* bridge STP will enable port */
+		return 0;
+
+	/* port is not bridged, so simulate going to FORWARDING state */
+	return rocker_port_stp_update(rocker_port, trans, flags,
+				      BR_STATE_FORWARDING);
+}
+
+static int rocker_port_fwd_disable(struct rocker_port *rocker_port,
+				   struct switchdev_trans *trans, int flags)
+{
+	if (rocker_port_is_bridged(rocker_port))
+		/* bridge STP will disable port */
+		return 0;
+
+	/* port is not bridged, so simulate going to DISABLED state */
+	return rocker_port_stp_update(rocker_port, trans, flags,
+				      BR_STATE_DISABLED);
+}
+
+static struct rocker_internal_vlan_tbl_entry *
+rocker_internal_vlan_tbl_find(const struct rocker *rocker, int ifindex)
+{
+	struct rocker_internal_vlan_tbl_entry *found;
+
+	hash_for_each_possible(rocker->internal_vlan_tbl, found,
+			       entry, ifindex) {
+		if (found->ifindex == ifindex)
+			return found;
+	}
+
+	return NULL;
+}
+
+static __be16 rocker_port_internal_vlan_id_get(struct rocker_port *rocker_port,
+					       int ifindex)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_internal_vlan_tbl_entry *entry;
+	struct rocker_internal_vlan_tbl_entry *found;
+	unsigned long lock_flags;
+	int i;
+
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		return 0;
+
+	entry->ifindex = ifindex;
+
+	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, lock_flags);
+
+	found = rocker_internal_vlan_tbl_find(rocker, ifindex);
+	if (found) {
+		kfree(entry);
+		goto found;
+	}
+
+	found = entry;
+	hash_add(rocker->internal_vlan_tbl, &found->entry, found->ifindex);
+
+	for (i = 0; i < ROCKER_N_INTERNAL_VLANS; i++) {
+		if (test_and_set_bit(i, rocker->internal_vlan_bitmap))
+			continue;
+		found->vlan_id = htons(ROCKER_INTERNAL_VLAN_ID_BASE + i);
+		goto found;
+	}
+
+	netdev_err(rocker_port->dev, "Out of internal VLAN IDs\n");
+
+found:
+	found->ref_count++;
+	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags);
+
+	return found->vlan_id;
+}
+
+static void
+rocker_port_internal_vlan_id_put(const struct rocker_port *rocker_port,
+				 int ifindex)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_internal_vlan_tbl_entry *found;
+	unsigned long lock_flags;
+	unsigned long bit;
+
+	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, lock_flags);
+
+	found = rocker_internal_vlan_tbl_find(rocker, ifindex);
+	if (!found) {
+		netdev_err(rocker_port->dev,
+			   "ifindex (%d) not found in internal VLAN tbl\n",
+			   ifindex);
+		goto not_found;
+	}
+
+	if (--found->ref_count <= 0) {
+		bit = ntohs(found->vlan_id) - ROCKER_INTERNAL_VLAN_ID_BASE;
+		clear_bit(bit, rocker->internal_vlan_bitmap);
+		hash_del(&found->entry);
+		kfree(found);
+	}
+
+not_found:
+	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags);
+}
+
+static int rocker_port_fib_ipv4(struct rocker_port *rocker_port,
+				struct switchdev_trans *trans, __be32 dst,
+				int dst_len, const struct fib_info *fi,
+				u32 tb_id, int flags)
+{
+	const struct fib_nh *nh;
+	__be16 eth_type = htons(ETH_P_IP);
+	__be32 dst_mask = inet_make_mask(dst_len);
+	__be16 internal_vlan_id = rocker_port->internal_vlan_id;
+	u32 priority = fi->fib_priority;
+	enum rocker_of_dpa_table_id goto_tbl =
+		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 group_id;
+	bool nh_on_port;
+	bool has_gw;
+	u32 index;
+	int err;
+
+	/* XXX support ECMP */
+
+	nh = fi->fib_nh;
+	nh_on_port = (fi->fib_dev == rocker_port->dev);
+	has_gw = !!nh->nh_gw;
+
+	if (has_gw && nh_on_port) {
+		err = rocker_port_ipv4_nh(rocker_port, trans, flags,
+					  nh->nh_gw, &index);
+		if (err)
+			return err;
+
+		group_id = ROCKER_GROUP_L3_UNICAST(index);
+	} else {
+		/* Send to CPU for processing */
+		group_id = ROCKER_GROUP_L2_INTERFACE(internal_vlan_id, 0);
+	}
+
+	err = rocker_flow_tbl_ucast4_routing(rocker_port, trans, eth_type, dst,
+					     dst_mask, priority, goto_tbl,
+					     group_id, flags);
+	if (err)
+		netdev_err(rocker_port->dev, "Error (%d) IPv4 route %pI4\n",
+			   err, &dst);
+
+	return err;
+}
+
+/*****************
+ * Net device ops
+ *****************/
+
+static int rocker_port_open(struct net_device *dev)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	int err;
+
+	err = rocker_port_dma_rings_init(rocker_port);
+	if (err)
+		return err;
+
+	err = request_irq(rocker_msix_tx_vector(rocker_port),
+			  rocker_tx_irq_handler, 0,
+			  rocker_driver_name, rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "cannot assign tx irq\n");
+		goto err_request_tx_irq;
+	}
+
+	err = request_irq(rocker_msix_rx_vector(rocker_port),
+			  rocker_rx_irq_handler, 0,
+			  rocker_driver_name, rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "cannot assign rx irq\n");
+		goto err_request_rx_irq;
+	}
+
+	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
+	if (err)
+		goto err_fwd_enable;
+
+	napi_enable(&rocker_port->napi_tx);
+	napi_enable(&rocker_port->napi_rx);
+	if (!dev->proto_down)
+		rocker_port_set_enable(rocker_port, true);
+	netif_start_queue(dev);
+	return 0;
+
+err_fwd_enable:
+	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
+err_request_rx_irq:
+	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
+err_request_tx_irq:
+	rocker_port_dma_rings_fini(rocker_port);
+	return err;
+}
+
+static int rocker_port_stop(struct net_device *dev)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+
+	netif_stop_queue(dev);
+	rocker_port_set_enable(rocker_port, false);
+	napi_disable(&rocker_port->napi_rx);
+	napi_disable(&rocker_port->napi_tx);
+	rocker_port_fwd_disable(rocker_port, NULL,
+				ROCKER_OP_FLAG_NOWAIT);
+	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
+	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
+	rocker_port_dma_rings_fini(rocker_port);
+
+	return 0;
+}
+
+static void rocker_tx_desc_frags_unmap(const struct rocker_port *rocker_port,
+				       const struct rocker_desc_info *desc_info)
+{
+	const struct rocker *rocker = rocker_port->rocker;
+	struct pci_dev *pdev = rocker->pdev;
+	const struct rocker_tlv *attrs[ROCKER_TLV_TX_MAX + 1];
+	struct rocker_tlv *attr;
+	int rem;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_TX_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_TX_FRAGS])
+		return;
+	rocker_tlv_for_each_nested(attr, attrs[ROCKER_TLV_TX_FRAGS], rem) {
+		const struct rocker_tlv *frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_MAX + 1];
+		dma_addr_t dma_handle;
+		size_t len;
+
+		if (rocker_tlv_type(attr) != ROCKER_TLV_TX_FRAG)
+			continue;
+		rocker_tlv_parse_nested(frag_attrs, ROCKER_TLV_TX_FRAG_ATTR_MAX,
+					attr);
+		if (!frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR] ||
+		    !frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN])
+			continue;
+		dma_handle = rocker_tlv_get_u64(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR]);
+		len = rocker_tlv_get_u16(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN]);
+		pci_unmap_single(pdev, dma_handle, len, DMA_TO_DEVICE);
+	}
+}
+
+static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port,
+				       struct rocker_desc_info *desc_info,
+				       char *buf, size_t buf_len)
+{
+	const struct rocker *rocker = rocker_port->rocker;
+	struct pci_dev *pdev = rocker->pdev;
+	dma_addr_t dma_handle;
+	struct rocker_tlv *frag;
+
+	dma_handle = pci_map_single(pdev, buf, buf_len, DMA_TO_DEVICE);
+	if (unlikely(pci_dma_mapping_error(pdev, dma_handle))) {
+		if (net_ratelimit())
+			netdev_err(rocker_port->dev, "failed to dma map tx frag\n");
+		return -EIO;
+	}
+	frag = rocker_tlv_nest_start(desc_info, ROCKER_TLV_TX_FRAG);
+	if (!frag)
+		goto unmap_frag;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_TX_FRAG_ATTR_ADDR,
+			       dma_handle))
+		goto nest_cancel;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_TX_FRAG_ATTR_LEN,
+			       buf_len))
+		goto nest_cancel;
+	rocker_tlv_nest_end(desc_info, frag);
+	return 0;
+
+nest_cancel:
+	rocker_tlv_nest_cancel(desc_info, frag);
+unmap_frag:
+	pci_unmap_single(pdev, dma_handle, buf_len, DMA_TO_DEVICE);
+	return -EMSGSIZE;
+}
+
+static netdev_tx_t rocker_port_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_desc_info *desc_info;
+	struct rocker_tlv *frags;
+	int i;
+	int err;
+
+	desc_info = rocker_desc_head_get(&rocker_port->tx_ring);
+	if (unlikely(!desc_info)) {
+		if (net_ratelimit())
+			netdev_err(dev, "tx ring full when queue awake\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	rocker_desc_cookie_ptr_set(desc_info, skb);
+
+	frags = rocker_tlv_nest_start(desc_info, ROCKER_TLV_TX_FRAGS);
+	if (!frags)
+		goto out;
+	err = rocker_tx_desc_frag_map_put(rocker_port, desc_info,
+					  skb->data, skb_headlen(skb));
+	if (err)
+		goto nest_cancel;
+	if (skb_shinfo(skb)->nr_frags > ROCKER_TX_FRAGS_MAX) {
+		err = skb_linearize(skb);
+		if (err)
+			goto unmap_frags;
+	}
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+		err = rocker_tx_desc_frag_map_put(rocker_port, desc_info,
+						  skb_frag_address(frag),
+						  skb_frag_size(frag));
+		if (err)
+			goto unmap_frags;
+	}
+	rocker_tlv_nest_end(desc_info, frags);
+
+	rocker_desc_gen_clear(desc_info);
+	rocker_desc_head_set(rocker, &rocker_port->tx_ring, desc_info);
+
+	desc_info = rocker_desc_head_get(&rocker_port->tx_ring);
+	if (!desc_info)
+		netif_stop_queue(dev);
+
+	return NETDEV_TX_OK;
+
+unmap_frags:
+	rocker_tx_desc_frags_unmap(rocker_port, desc_info);
+nest_cancel:
+	rocker_tlv_nest_cancel(desc_info, frags);
+out:
+	dev_kfree_skb(skb);
+	dev->stats.tx_dropped++;
+
+	return NETDEV_TX_OK;
+}
+
+static int rocker_port_set_mac_address(struct net_device *dev, void *p)
+{
+	struct sockaddr *addr = p;
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	int err;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	err = rocker_cmd_set_port_settings_macaddr(rocker_port, addr->sa_data);
+	if (err)
+		return err;
+	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+	return 0;
+}
+
+static int rocker_port_change_mtu(struct net_device *dev, int new_mtu)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	int running = netif_running(dev);
+	int err;
+
+#define ROCKER_PORT_MIN_MTU	68
+#define ROCKER_PORT_MAX_MTU	9000
+
+	if (new_mtu < ROCKER_PORT_MIN_MTU || new_mtu > ROCKER_PORT_MAX_MTU)
+		return -EINVAL;
+
+	if (running)
+		rocker_port_stop(dev);
+
+	netdev_info(dev, "MTU change from %d to %d\n", dev->mtu, new_mtu);
+	dev->mtu = new_mtu;
+
+	err = rocker_cmd_set_port_settings_mtu(rocker_port, new_mtu);
+	if (err)
+		return err;
+
+	if (running)
+		err = rocker_port_open(dev);
+
+	return err;
+}
+
+static int rocker_port_get_phys_port_name(struct net_device *dev,
+					  char *buf, size_t len)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	struct port_name name = { .buf = buf, .len = len };
+	int err;
+
+	err = rocker_cmd_exec(rocker_port, NULL, 0,
+			      rocker_cmd_get_port_settings_prep, NULL,
+			      rocker_cmd_get_port_settings_phys_name_proc,
+			      &name);
+
+	return err ? -EOPNOTSUPP : 0;
+}
+
+static int rocker_port_change_proto_down(struct net_device *dev,
+					 bool proto_down)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+
+	if (rocker_port->dev->flags & IFF_UP)
+		rocker_port_set_enable(rocker_port, !proto_down);
+	rocker_port->dev->proto_down = proto_down;
+	return 0;
+}
+
+static void rocker_port_neigh_destroy(struct neighbour *n)
+{
+	struct rocker_port *rocker_port = netdev_priv(n->dev);
+	int flags = ROCKER_OP_FLAG_REMOVE | ROCKER_OP_FLAG_NOWAIT;
+	__be32 ip_addr = *(__be32 *)n->primary_key;
+
+	rocker_port_ipv4_neigh(rocker_port, NULL,
+			       flags, ip_addr, n->ha);
+}
+
+static const struct net_device_ops rocker_port_netdev_ops = {
+	.ndo_open			= rocker_port_open,
+	.ndo_stop			= rocker_port_stop,
+	.ndo_start_xmit			= rocker_port_xmit,
+	.ndo_set_mac_address		= rocker_port_set_mac_address,
+	.ndo_change_mtu			= rocker_port_change_mtu,
+	.ndo_bridge_getlink		= switchdev_port_bridge_getlink,
+	.ndo_bridge_setlink		= switchdev_port_bridge_setlink,
+	.ndo_bridge_dellink		= switchdev_port_bridge_dellink,
+	.ndo_fdb_add			= switchdev_port_fdb_add,
+	.ndo_fdb_del			= switchdev_port_fdb_del,
+	.ndo_fdb_dump			= switchdev_port_fdb_dump,
+	.ndo_get_phys_port_name		= rocker_port_get_phys_port_name,
+	.ndo_change_proto_down		= rocker_port_change_proto_down,
+	.ndo_neigh_destroy		= rocker_port_neigh_destroy,
+};
+
+/********************
+ * swdev interface
+ ********************/
+
+static int rocker_port_attr_get(struct net_device *dev,
+				struct switchdev_attr *attr)
+{
+	const struct rocker_port *rocker_port = netdev_priv(dev);
+	const struct rocker *rocker = rocker_port->rocker;
+
+	switch (attr->id) {
+	case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
+		attr->u.ppid.id_len = sizeof(rocker->hw.id);
+		memcpy(&attr->u.ppid.id, &rocker->hw.id, attr->u.ppid.id_len);
+		break;
+	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
+		attr->u.brport_flags = rocker_port->brport_flags;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int rocker_port_brport_flags_set(struct rocker_port *rocker_port,
+					struct switchdev_trans *trans,
+					unsigned long brport_flags)
+{
+	unsigned long orig_flags;
+	int err = 0;
+
+	orig_flags = rocker_port->brport_flags;
+	rocker_port->brport_flags = brport_flags;
+	if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING)
+		err = rocker_port_set_learning(rocker_port, trans);
+
+	if (switchdev_trans_ph_prepare(trans))
+		rocker_port->brport_flags = orig_flags;
+
+	return err;
+}
+
+static int rocker_port_bridge_ageing_time(struct rocker_port *rocker_port,
+					  struct switchdev_trans *trans,
+					  u32 ageing_time)
+{
+	if (!switchdev_trans_ph_prepare(trans)) {
+		rocker_port->ageing_time = clock_t_to_jiffies(ageing_time);
+		mod_timer(&rocker_port->rocker->fdb_cleanup_timer, jiffies);
+	}
+
+	return 0;
+}
+
+static int rocker_port_attr_set(struct net_device *dev,
+				const struct switchdev_attr *attr,
+				struct switchdev_trans *trans)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	int err = 0;
+
+	switch (attr->id) {
+	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
+		err = rocker_port_stp_update(rocker_port, trans, 0,
+					     attr->u.stp_state);
+		break;
+	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
+		err = rocker_port_brport_flags_set(rocker_port, trans,
+						   attr->u.brport_flags);
+		break;
+	case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME:
+		err = rocker_port_bridge_ageing_time(rocker_port, trans,
+						     attr->u.ageing_time);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int rocker_port_vlan_add(struct rocker_port *rocker_port,
+				struct switchdev_trans *trans,
+				u16 vid, u16 flags)
+{
+	int err;
+
+	/* XXX deal with flags for PVID and untagged */
+
+	err = rocker_port_vlan(rocker_port, trans, 0, vid);
+	if (err)
+		return err;
+
+	err = rocker_port_router_mac(rocker_port, trans, 0, htons(vid));
+	if (err)
+		rocker_port_vlan(rocker_port, trans,
+				 ROCKER_OP_FLAG_REMOVE, vid);
+
+	return err;
+}
+
+static int rocker_port_vlans_add(struct rocker_port *rocker_port,
+				 struct switchdev_trans *trans,
+				 const struct switchdev_obj_port_vlan *vlan)
+{
+	u16 vid;
+	int err;
+
+	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
+		err = rocker_port_vlan_add(rocker_port, trans,
+					   vid, vlan->flags);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int rocker_port_fdb_add(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans,
+			       const struct switchdev_obj_port_fdb *fdb)
+{
+	__be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL);
+	int flags = 0;
+
+	if (!rocker_port_is_bridged(rocker_port))
+		return -EINVAL;
+
+	return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags);
+}
+
+static int rocker_port_obj_add(struct net_device *dev,
+			       const struct switchdev_obj *obj,
+			       struct switchdev_trans *trans)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	const struct switchdev_obj_ipv4_fib *fib4;
+	int err = 0;
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = rocker_port_vlans_add(rocker_port, trans,
+					    SWITCHDEV_OBJ_PORT_VLAN(obj));
+		break;
+	case SWITCHDEV_OBJ_ID_IPV4_FIB:
+		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
+		err = rocker_port_fib_ipv4(rocker_port, trans,
+					   htonl(fib4->dst), fib4->dst_len,
+					   &fib4->fi, fib4->tb_id, 0);
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_FDB:
+		err = rocker_port_fdb_add(rocker_port, trans,
+					  SWITCHDEV_OBJ_PORT_FDB(obj));
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int rocker_port_vlan_del(struct rocker_port *rocker_port,
+				u16 vid, u16 flags)
+{
+	int err;
+
+	err = rocker_port_router_mac(rocker_port, NULL,
+				     ROCKER_OP_FLAG_REMOVE, htons(vid));
+	if (err)
+		return err;
+
+	return rocker_port_vlan(rocker_port, NULL,
+				ROCKER_OP_FLAG_REMOVE, vid);
+}
+
+static int rocker_port_vlans_del(struct rocker_port *rocker_port,
+				 const struct switchdev_obj_port_vlan *vlan)
+{
+	u16 vid;
+	int err;
+
+	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
+		err = rocker_port_vlan_del(rocker_port, vid, vlan->flags);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int rocker_port_fdb_del(struct rocker_port *rocker_port,
+			       struct switchdev_trans *trans,
+			       const struct switchdev_obj_port_fdb *fdb)
+{
+	__be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL);
+	int flags = ROCKER_OP_FLAG_REMOVE;
+
+	if (!rocker_port_is_bridged(rocker_port))
+		return -EINVAL;
+
+	return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags);
+}
+
+static int rocker_port_obj_del(struct net_device *dev,
+			       const struct switchdev_obj *obj)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	const struct switchdev_obj_ipv4_fib *fib4;
+	int err = 0;
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = rocker_port_vlans_del(rocker_port,
+					    SWITCHDEV_OBJ_PORT_VLAN(obj));
+		break;
+	case SWITCHDEV_OBJ_ID_IPV4_FIB:
+		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
+		err = rocker_port_fib_ipv4(rocker_port, NULL,
+					   htonl(fib4->dst), fib4->dst_len,
+					   &fib4->fi, fib4->tb_id,
+					   ROCKER_OP_FLAG_REMOVE);
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_FDB:
+		err = rocker_port_fdb_del(rocker_port, NULL,
+					  SWITCHDEV_OBJ_PORT_FDB(obj));
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int rocker_port_fdb_dump(const struct rocker_port *rocker_port,
+				struct switchdev_obj_port_fdb *fdb,
+				switchdev_obj_dump_cb_t *cb)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_fdb_tbl_entry *found;
+	struct hlist_node *tmp;
+	unsigned long lock_flags;
+	int bkt;
+	int err = 0;
+
+	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
+	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) {
+		if (found->key.rocker_port != rocker_port)
+			continue;
+		ether_addr_copy(fdb->addr, found->key.addr);
+		fdb->ndm_state = NUD_REACHABLE;
+		fdb->vid = rocker_port_vlan_to_vid(rocker_port,
+						   found->key.vlan_id);
+		err = cb(&fdb->obj);
+		if (err)
+			break;
+	}
+	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
+
+	return err;
+}
+
+static int rocker_port_vlan_dump(const struct rocker_port *rocker_port,
+				 struct switchdev_obj_port_vlan *vlan,
+				 switchdev_obj_dump_cb_t *cb)
+{
+	u16 vid;
+	int err = 0;
+
+	for (vid = 1; vid < VLAN_N_VID; vid++) {
+		if (!test_bit(vid, rocker_port->vlan_bitmap))
+			continue;
+		vlan->flags = 0;
+		if (rocker_vlan_id_is_internal(htons(vid)))
+			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
+		vlan->vid_begin = vid;
+		vlan->vid_end = vid;
+		err = cb(&vlan->obj);
+		if (err)
+			break;
+	}
+
+	return err;
+}
+
+static int rocker_port_obj_dump(struct net_device *dev,
+				struct switchdev_obj *obj,
+				switchdev_obj_dump_cb_t *cb)
+{
+	const struct rocker_port *rocker_port = netdev_priv(dev);
+	int err = 0;
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_FDB:
+		err = rocker_port_fdb_dump(rocker_port,
+					   SWITCHDEV_OBJ_PORT_FDB(obj), cb);
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = rocker_port_vlan_dump(rocker_port,
+					    SWITCHDEV_OBJ_PORT_VLAN(obj), cb);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static const struct switchdev_ops rocker_port_switchdev_ops = {
+	.switchdev_port_attr_get	= rocker_port_attr_get,
+	.switchdev_port_attr_set	= rocker_port_attr_set,
+	.switchdev_port_obj_add		= rocker_port_obj_add,
+	.switchdev_port_obj_del		= rocker_port_obj_del,
+	.switchdev_port_obj_dump	= rocker_port_obj_dump,
+};
+
+/********************
+ * ethtool interface
+ ********************/
+
+static int rocker_port_get_settings(struct net_device *dev,
+				    struct ethtool_cmd *ecmd)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+
+	return rocker_cmd_get_port_settings_ethtool(rocker_port, ecmd);
+}
+
+static int rocker_port_set_settings(struct net_device *dev,
+				    struct ethtool_cmd *ecmd)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+
+	return rocker_cmd_set_port_settings_ethtool(rocker_port, ecmd);
+}
+
+static void rocker_port_get_drvinfo(struct net_device *dev,
+				    struct ethtool_drvinfo *drvinfo)
+{
+	strlcpy(drvinfo->driver, rocker_driver_name, sizeof(drvinfo->driver));
+	strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version));
+}
+
+static struct rocker_port_stats {
+	char str[ETH_GSTRING_LEN];
+	int type;
+} rocker_port_stats[] = {
+	{ "rx_packets", ROCKER_TLV_CMD_PORT_STATS_RX_PKTS,    },
+	{ "rx_bytes",   ROCKER_TLV_CMD_PORT_STATS_RX_BYTES,   },
+	{ "rx_dropped", ROCKER_TLV_CMD_PORT_STATS_RX_DROPPED, },
+	{ "rx_errors",  ROCKER_TLV_CMD_PORT_STATS_RX_ERRORS,  },
+
+	{ "tx_packets", ROCKER_TLV_CMD_PORT_STATS_TX_PKTS,    },
+	{ "tx_bytes",   ROCKER_TLV_CMD_PORT_STATS_TX_BYTES,   },
+	{ "tx_dropped", ROCKER_TLV_CMD_PORT_STATS_TX_DROPPED, },
+	{ "tx_errors",  ROCKER_TLV_CMD_PORT_STATS_TX_ERRORS,  },
+};
+
+#define ROCKER_PORT_STATS_LEN  ARRAY_SIZE(rocker_port_stats)
+
+static void rocker_port_get_strings(struct net_device *netdev, u32 stringset,
+				    u8 *data)
+{
+	u8 *p = data;
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(rocker_port_stats); i++) {
+			memcpy(p, rocker_port_stats[i].str, ETH_GSTRING_LEN);
+			p += ETH_GSTRING_LEN;
+		}
+		break;
+	}
+}
+
+static int
+rocker_cmd_get_port_stats_prep(const struct rocker_port *rocker_port,
+			       struct rocker_desc_info *desc_info,
+			       void *priv)
+{
+	struct rocker_tlv *cmd_stats;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
+			       ROCKER_TLV_CMD_TYPE_GET_PORT_STATS))
+		return -EMSGSIZE;
+
+	cmd_stats = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_stats)
+		return -EMSGSIZE;
+
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_CMD_PORT_STATS_PPORT,
+			       rocker_port->pport))
+		return -EMSGSIZE;
+
+	rocker_tlv_nest_end(desc_info, cmd_stats);
+
+	return 0;
+}
+
+static int
+rocker_cmd_get_port_stats_ethtool_proc(const struct rocker_port *rocker_port,
+				       const struct rocker_desc_info *desc_info,
+				       void *priv)
+{
+	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
+	const struct rocker_tlv *stats_attrs[ROCKER_TLV_CMD_PORT_STATS_MAX + 1];
+	const struct rocker_tlv *pattr;
+	u32 pport;
+	u64 *data = priv;
+	int i;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
+
+	if (!attrs[ROCKER_TLV_CMD_INFO])
+		return -EIO;
+
+	rocker_tlv_parse_nested(stats_attrs, ROCKER_TLV_CMD_PORT_STATS_MAX,
+				attrs[ROCKER_TLV_CMD_INFO]);
+
+	if (!stats_attrs[ROCKER_TLV_CMD_PORT_STATS_PPORT])
+		return -EIO;
+
+	pport = rocker_tlv_get_u32(stats_attrs[ROCKER_TLV_CMD_PORT_STATS_PPORT]);
+	if (pport != rocker_port->pport)
+		return -EIO;
+
+	for (i = 0; i < ARRAY_SIZE(rocker_port_stats); i++) {
+		pattr = stats_attrs[rocker_port_stats[i].type];
+		if (!pattr)
+			continue;
+
+		data[i] = rocker_tlv_get_u64(pattr);
+	}
+
+	return 0;
+}
+
+static int rocker_cmd_get_port_stats_ethtool(struct rocker_port *rocker_port,
+					     void *priv)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_get_port_stats_prep, NULL,
+			       rocker_cmd_get_port_stats_ethtool_proc,
+			       priv);
+}
+
+static void rocker_port_get_stats(struct net_device *dev,
+				  struct ethtool_stats *stats, u64 *data)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+
+	if (rocker_cmd_get_port_stats_ethtool(rocker_port, data) != 0) {
+		int i;
+
+		for (i = 0; i < ARRAY_SIZE(rocker_port_stats); ++i)
+			data[i] = 0;
+	}
+}
+
+static int rocker_port_get_sset_count(struct net_device *netdev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return ROCKER_PORT_STATS_LEN;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static const struct ethtool_ops rocker_port_ethtool_ops = {
+	.get_settings		= rocker_port_get_settings,
+	.set_settings		= rocker_port_set_settings,
+	.get_drvinfo		= rocker_port_get_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_strings		= rocker_port_get_strings,
+	.get_ethtool_stats	= rocker_port_get_stats,
+	.get_sset_count		= rocker_port_get_sset_count,
+};
+
+/*****************
+ * NAPI interface
+ *****************/
+
+static struct rocker_port *rocker_port_napi_tx_get(struct napi_struct *napi)
+{
+	return container_of(napi, struct rocker_port, napi_tx);
+}
+
+static int rocker_port_poll_tx(struct napi_struct *napi, int budget)
+{
+	struct rocker_port *rocker_port = rocker_port_napi_tx_get(napi);
+	const struct rocker *rocker = rocker_port->rocker;
+	const struct rocker_desc_info *desc_info;
+	u32 credits = 0;
+	int err;
+
+	/* Cleanup tx descriptors */
+	while ((desc_info = rocker_desc_tail_get(&rocker_port->tx_ring))) {
+		struct sk_buff *skb;
+
+		err = rocker_desc_err(desc_info);
+		if (err && net_ratelimit())
+			netdev_err(rocker_port->dev, "tx desc received with err %d\n",
+				   err);
+		rocker_tx_desc_frags_unmap(rocker_port, desc_info);
+
+		skb = rocker_desc_cookie_ptr_get(desc_info);
+		if (err == 0) {
+			rocker_port->dev->stats.tx_packets++;
+			rocker_port->dev->stats.tx_bytes += skb->len;
+		} else {
+			rocker_port->dev->stats.tx_errors++;
+		}
+
+		dev_kfree_skb_any(skb);
+		credits++;
+	}
+
+	if (credits && netif_queue_stopped(rocker_port->dev))
+		netif_wake_queue(rocker_port->dev);
+
+	napi_complete(napi);
+	rocker_dma_ring_credits_set(rocker, &rocker_port->tx_ring, credits);
+
+	return 0;
+}
+
+static int rocker_port_rx_proc(const struct rocker *rocker,
+			       const struct rocker_port *rocker_port,
+			       struct rocker_desc_info *desc_info)
+{
+	const struct rocker_tlv *attrs[ROCKER_TLV_RX_MAX + 1];
+	struct sk_buff *skb = rocker_desc_cookie_ptr_get(desc_info);
+	size_t rx_len;
+	u16 rx_flags = 0;
+
+	if (!skb)
+		return -ENOENT;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_RX_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_RX_FRAG_LEN])
+		return -EINVAL;
+	if (attrs[ROCKER_TLV_RX_FLAGS])
+		rx_flags = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FLAGS]);
+
+	rocker_dma_rx_ring_skb_unmap(rocker, attrs);
+
+	rx_len = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FRAG_LEN]);
+	skb_put(skb, rx_len);
+	skb->protocol = eth_type_trans(skb, rocker_port->dev);
+
+	if (rx_flags & ROCKER_RX_FLAGS_FWD_OFFLOAD)
+		skb->offload_fwd_mark = rocker_port->dev->offload_fwd_mark;
+
+	rocker_port->dev->stats.rx_packets++;
+	rocker_port->dev->stats.rx_bytes += skb->len;
+
+	netif_receive_skb(skb);
+
+	return rocker_dma_rx_ring_skb_alloc(rocker_port, desc_info);
+}
+
+static struct rocker_port *rocker_port_napi_rx_get(struct napi_struct *napi)
+{
+	return container_of(napi, struct rocker_port, napi_rx);
+}
+
+static int rocker_port_poll_rx(struct napi_struct *napi, int budget)
+{
+	struct rocker_port *rocker_port = rocker_port_napi_rx_get(napi);
+	const struct rocker *rocker = rocker_port->rocker;
+	struct rocker_desc_info *desc_info;
+	u32 credits = 0;
+	int err;
+
+	/* Process rx descriptors */
+	while (credits < budget &&
+	       (desc_info = rocker_desc_tail_get(&rocker_port->rx_ring))) {
+		err = rocker_desc_err(desc_info);
+		if (err) {
+			if (net_ratelimit())
+				netdev_err(rocker_port->dev, "rx desc received with err %d\n",
+					   err);
+		} else {
+			err = rocker_port_rx_proc(rocker, rocker_port,
+						  desc_info);
+			if (err && net_ratelimit())
+				netdev_err(rocker_port->dev, "rx processing failed with err %d\n",
+					   err);
+		}
+		if (err)
+			rocker_port->dev->stats.rx_errors++;
+
+		rocker_desc_gen_clear(desc_info);
+		rocker_desc_head_set(rocker, &rocker_port->rx_ring, desc_info);
+		credits++;
+	}
+
+	if (credits < budget)
+		napi_complete(napi);
+
+	rocker_dma_ring_credits_set(rocker, &rocker_port->rx_ring, credits);
+
+	return credits;
+}
+
+/*****************
+ * PCI driver ops
+ *****************/
+
+static void rocker_carrier_init(const struct rocker_port *rocker_port)
+{
+	const struct rocker *rocker = rocker_port->rocker;
+	u64 link_status = rocker_read64(rocker, PORT_PHYS_LINK_STATUS);
+	bool link_up;
+
+	link_up = link_status & (1 << rocker_port->pport);
+	if (link_up)
+		netif_carrier_on(rocker_port->dev);
+	else
+		netif_carrier_off(rocker_port->dev);
+}
+
+static void rocker_remove_ports(const struct rocker *rocker)
+{
+	struct rocker_port *rocker_port;
+	int i;
+
+	for (i = 0; i < rocker->port_count; i++) {
+		rocker_port = rocker->ports[i];
+		if (!rocker_port)
+			continue;
+		rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
+		unregister_netdev(rocker_port->dev);
+		free_netdev(rocker_port->dev);
+	}
+	kfree(rocker->ports);
+}
+
+static void rocker_port_dev_addr_init(struct rocker_port *rocker_port)
+{
+	const struct rocker *rocker = rocker_port->rocker;
+	const struct pci_dev *pdev = rocker->pdev;
+	int err;
+
+	err = rocker_cmd_get_port_settings_macaddr(rocker_port,
+						   rocker_port->dev->dev_addr);
+	if (err) {
+		dev_warn(&pdev->dev, "failed to get mac address, using random\n");
+		eth_hw_addr_random(rocker_port->dev);
+	}
+}
+
+static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
+{
+	const struct pci_dev *pdev = rocker->pdev;
+	struct rocker_port *rocker_port;
+	struct net_device *dev;
+	u16 untagged_vid = 0;
+	int err;
+
+	dev = alloc_etherdev(sizeof(struct rocker_port));
+	if (!dev)
+		return -ENOMEM;
+	rocker_port = netdev_priv(dev);
+	rocker_port->dev = dev;
+	rocker_port->rocker = rocker;
+	rocker_port->port_number = port_number;
+	rocker_port->pport = port_number + 1;
+	rocker_port->brport_flags = BR_LEARNING | BR_LEARNING_SYNC;
+	rocker_port->ageing_time = BR_DEFAULT_AGEING_TIME;
+
+	rocker_port_dev_addr_init(rocker_port);
+	dev->netdev_ops = &rocker_port_netdev_ops;
+	dev->ethtool_ops = &rocker_port_ethtool_ops;
+	dev->switchdev_ops = &rocker_port_switchdev_ops;
+	netif_tx_napi_add(dev, &rocker_port->napi_tx, rocker_port_poll_tx,
+			  NAPI_POLL_WEIGHT);
+	netif_napi_add(dev, &rocker_port->napi_rx, rocker_port_poll_rx,
+		       NAPI_POLL_WEIGHT);
+	rocker_carrier_init(rocker_port);
+
+	dev->features |= NETIF_F_NETNS_LOCAL | NETIF_F_SG;
+
+	err = register_netdev(dev);
+	if (err) {
+		dev_err(&pdev->dev, "register_netdev failed\n");
+		goto err_register_netdev;
+	}
+	rocker->ports[port_number] = rocker_port;
+
+	switchdev_port_fwd_mark_set(rocker_port->dev, NULL, false);
+
+	rocker_port_set_learning(rocker_port, NULL);
+
+	err = rocker_port_ig_tbl(rocker_port, NULL, 0);
+	if (err) {
+		netdev_err(rocker_port->dev, "install ig port table failed\n");
+		goto err_port_ig_tbl;
+	}
+
+	rocker_port->internal_vlan_id =
+		rocker_port_internal_vlan_id_get(rocker_port, dev->ifindex);
+
+	err = rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
+	if (err) {
+		netdev_err(rocker_port->dev, "install untagged VLAN failed\n");
+		goto err_untagged_vlan;
+	}
+
+	return 0;
+
+err_untagged_vlan:
+	rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
+err_port_ig_tbl:
+	rocker->ports[port_number] = NULL;
+	unregister_netdev(dev);
+err_register_netdev:
+	free_netdev(dev);
+	return err;
+}
+
+static int rocker_probe_ports(struct rocker *rocker)
+{
+	int i;
+	size_t alloc_size;
+	int err;
+
+	alloc_size = sizeof(struct rocker_port *) * rocker->port_count;
+	rocker->ports = kzalloc(alloc_size, GFP_KERNEL);
+	if (!rocker->ports)
+		return -ENOMEM;
+	for (i = 0; i < rocker->port_count; i++) {
+		err = rocker_probe_port(rocker, i);
+		if (err)
+			goto remove_ports;
+	}
+	return 0;
+
+remove_ports:
+	rocker_remove_ports(rocker);
+	return err;
+}
+
+static int rocker_msix_init(struct rocker *rocker)
+{
+	struct pci_dev *pdev = rocker->pdev;
+	int msix_entries;
+	int i;
+	int err;
+
+	msix_entries = pci_msix_vec_count(pdev);
+	if (msix_entries < 0)
+		return msix_entries;
+
+	if (msix_entries != ROCKER_MSIX_VEC_COUNT(rocker->port_count))
+		return -EINVAL;
+
+	rocker->msix_entries = kmalloc_array(msix_entries,
+					     sizeof(struct msix_entry),
+					     GFP_KERNEL);
+	if (!rocker->msix_entries)
+		return -ENOMEM;
+
+	for (i = 0; i < msix_entries; i++)
+		rocker->msix_entries[i].entry = i;
+
+	err = pci_enable_msix_exact(pdev, rocker->msix_entries, msix_entries);
+	if (err < 0)
+		goto err_enable_msix;
+
+	return 0;
+
+err_enable_msix:
+	kfree(rocker->msix_entries);
+	return err;
+}
+
+static void rocker_msix_fini(const struct rocker *rocker)
+{
+	pci_disable_msix(rocker->pdev);
+	kfree(rocker->msix_entries);
+}
+
+static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	struct rocker *rocker;
+	int err;
+
+	rocker = kzalloc(sizeof(*rocker), GFP_KERNEL);
+	if (!rocker)
+		return -ENOMEM;
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "pci_enable_device failed\n");
+		goto err_pci_enable_device;
+	}
+
+	err = pci_request_regions(pdev, rocker_driver_name);
+	if (err) {
+		dev_err(&pdev->dev, "pci_request_regions failed\n");
+		goto err_pci_request_regions;
+	}
+
+	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
+	if (!err) {
+		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
+		if (err) {
+			dev_err(&pdev->dev, "pci_set_consistent_dma_mask failed\n");
+			goto err_pci_set_dma_mask;
+		}
+	} else {
+		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+		if (err) {
+			dev_err(&pdev->dev, "pci_set_dma_mask failed\n");
+			goto err_pci_set_dma_mask;
+		}
+	}
+
+	if (pci_resource_len(pdev, 0) < ROCKER_PCI_BAR0_SIZE) {
+		dev_err(&pdev->dev, "invalid PCI region size\n");
+		err = -EINVAL;
+		goto err_pci_resource_len_check;
+	}
+
+	rocker->hw_addr = ioremap(pci_resource_start(pdev, 0),
+				  pci_resource_len(pdev, 0));
+	if (!rocker->hw_addr) {
+		dev_err(&pdev->dev, "ioremap failed\n");
+		err = -EIO;
+		goto err_ioremap;
+	}
+	pci_set_master(pdev);
+
+	rocker->pdev = pdev;
+	pci_set_drvdata(pdev, rocker);
+
+	rocker->port_count = rocker_read32(rocker, PORT_PHYS_COUNT);
+
+	err = rocker_msix_init(rocker);
+	if (err) {
+		dev_err(&pdev->dev, "MSI-X init failed\n");
+		goto err_msix_init;
+	}
+
+	err = rocker_basic_hw_test(rocker);
+	if (err) {
+		dev_err(&pdev->dev, "basic hw test failed\n");
+		goto err_basic_hw_test;
+	}
+
+	rocker_write32(rocker, CONTROL, ROCKER_CONTROL_RESET);
+
+	err = rocker_dma_rings_init(rocker);
+	if (err)
+		goto err_dma_rings_init;
+
+	err = request_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD),
+			  rocker_cmd_irq_handler, 0,
+			  rocker_driver_name, rocker);
+	if (err) {
+		dev_err(&pdev->dev, "cannot assign cmd irq\n");
+		goto err_request_cmd_irq;
+	}
+
+	err = request_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT),
+			  rocker_event_irq_handler, 0,
+			  rocker_driver_name, rocker);
+	if (err) {
+		dev_err(&pdev->dev, "cannot assign event irq\n");
+		goto err_request_event_irq;
+	}
+
+	rocker->hw.id = rocker_read64(rocker, SWITCH_ID);
+
+	err = rocker_init_tbls(rocker);
+	if (err) {
+		dev_err(&pdev->dev, "cannot init rocker tables\n");
+		goto err_init_tbls;
+	}
+
+	setup_timer(&rocker->fdb_cleanup_timer, rocker_fdb_cleanup,
+		    (unsigned long) rocker);
+	mod_timer(&rocker->fdb_cleanup_timer, jiffies);
+
+	err = rocker_probe_ports(rocker);
+	if (err) {
+		dev_err(&pdev->dev, "failed to probe ports\n");
+		goto err_probe_ports;
+	}
+
+	dev_info(&pdev->dev, "Rocker switch with id %*phN\n",
+		 (int)sizeof(rocker->hw.id), &rocker->hw.id);
+
+	return 0;
+
+err_probe_ports:
+	del_timer_sync(&rocker->fdb_cleanup_timer);
+	rocker_free_tbls(rocker);
+err_init_tbls:
+	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT), rocker);
+err_request_event_irq:
+	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD), rocker);
+err_request_cmd_irq:
+	rocker_dma_rings_fini(rocker);
+err_dma_rings_init:
+err_basic_hw_test:
+	rocker_msix_fini(rocker);
+err_msix_init:
+	iounmap(rocker->hw_addr);
+err_ioremap:
+err_pci_resource_len_check:
+err_pci_set_dma_mask:
+	pci_release_regions(pdev);
+err_pci_request_regions:
+	pci_disable_device(pdev);
+err_pci_enable_device:
+	kfree(rocker);
+	return err;
+}
+
+static void rocker_remove(struct pci_dev *pdev)
+{
+	struct rocker *rocker = pci_get_drvdata(pdev);
+
+	del_timer_sync(&rocker->fdb_cleanup_timer);
+	rocker_free_tbls(rocker);
+	rocker_write32(rocker, CONTROL, ROCKER_CONTROL_RESET);
+	rocker_remove_ports(rocker);
+	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT), rocker);
+	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD), rocker);
+	rocker_dma_rings_fini(rocker);
+	rocker_msix_fini(rocker);
+	iounmap(rocker->hw_addr);
+	pci_release_regions(rocker->pdev);
+	pci_disable_device(rocker->pdev);
+	kfree(rocker);
+}
+
+static struct pci_driver rocker_pci_driver = {
+	.name		= rocker_driver_name,
+	.id_table	= rocker_pci_id_table,
+	.probe		= rocker_probe,
+	.remove		= rocker_remove,
+};
+
+/************************************
+ * Net device notifier event handler
+ ************************************/
+
+static bool rocker_port_dev_check(const struct net_device *dev)
+{
+	return dev->netdev_ops == &rocker_port_netdev_ops;
+}
+
+static int rocker_port_bridge_join(struct rocker_port *rocker_port,
+				   struct net_device *bridge)
+{
+	u16 untagged_vid = 0;
+	int err;
+
+	/* Port is joining bridge, so the internal VLAN for the
+	 * port is going to change to the bridge internal VLAN.
+	 * Let's remove untagged VLAN (vid=0) from port and
+	 * re-add once internal VLAN has changed.
+	 */
+
+	err = rocker_port_vlan_del(rocker_port, untagged_vid, 0);
+	if (err)
+		return err;
+
+	rocker_port_internal_vlan_id_put(rocker_port,
+					 rocker_port->dev->ifindex);
+	rocker_port->internal_vlan_id =
+		rocker_port_internal_vlan_id_get(rocker_port, bridge->ifindex);
+
+	rocker_port->bridge_dev = bridge;
+	switchdev_port_fwd_mark_set(rocker_port->dev, bridge, true);
+
+	return rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
+}
+
+static int rocker_port_bridge_leave(struct rocker_port *rocker_port)
+{
+	u16 untagged_vid = 0;
+	int err;
+
+	err = rocker_port_vlan_del(rocker_port, untagged_vid, 0);
+	if (err)
+		return err;
+
+	rocker_port_internal_vlan_id_put(rocker_port,
+					 rocker_port->bridge_dev->ifindex);
+	rocker_port->internal_vlan_id =
+		rocker_port_internal_vlan_id_get(rocker_port,
+						 rocker_port->dev->ifindex);
+
+	switchdev_port_fwd_mark_set(rocker_port->dev, rocker_port->bridge_dev,
+				    false);
+	rocker_port->bridge_dev = NULL;
+
+	err = rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
+	if (err)
+		return err;
+
+	if (rocker_port->dev->flags & IFF_UP)
+		err = rocker_port_fwd_enable(rocker_port, NULL, 0);
+
+	return err;
+}
+
+static int rocker_port_ovs_changed(struct rocker_port *rocker_port,
+				   struct net_device *master)
+{
+	int err;
+
+	rocker_port->bridge_dev = master;
+
+	err = rocker_port_fwd_disable(rocker_port, NULL, 0);
+	if (err)
+		return err;
+	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
+
+	return err;
+}
+
+static int rocker_port_master_linked(struct rocker_port *rocker_port,
+				     struct net_device *master)
+{
+	int err = 0;
+
+	if (netif_is_bridge_master(master))
+		err = rocker_port_bridge_join(rocker_port, master);
+	else if (netif_is_ovs_master(master))
+		err = rocker_port_ovs_changed(rocker_port, master);
+	return err;
+}
+
+static int rocker_port_master_unlinked(struct rocker_port *rocker_port)
+{
+	int err = 0;
+
+	if (rocker_port_is_bridged(rocker_port))
+		err = rocker_port_bridge_leave(rocker_port);
+	else if (rocker_port_is_ovsed(rocker_port))
+		err = rocker_port_ovs_changed(rocker_port, NULL);
+	return err;
+}
+
+static int rocker_netdevice_event(struct notifier_block *unused,
+				  unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct netdev_notifier_changeupper_info *info;
+	struct rocker_port *rocker_port;
+	int err;
+
+	if (!rocker_port_dev_check(dev))
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_CHANGEUPPER:
+		info = ptr;
+		if (!info->master)
+			goto out;
+		rocker_port = netdev_priv(dev);
+		if (info->linking) {
+			err = rocker_port_master_linked(rocker_port,
+							info->upper_dev);
+			if (err)
+				netdev_warn(dev, "failed to reflect master linked (err %d)\n",
+					    err);
+		} else {
+			err = rocker_port_master_unlinked(rocker_port);
+			if (err)
+				netdev_warn(dev, "failed to reflect master unlinked (err %d)\n",
+					    err);
+		}
+		break;
+	}
+out:
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block rocker_netdevice_nb __read_mostly = {
+	.notifier_call = rocker_netdevice_event,
+};
+
+/************************************
+ * Net event notifier event handler
+ ************************************/
+
+static int rocker_neigh_update(struct net_device *dev, struct neighbour *n)
+{
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	int flags = (n->nud_state & NUD_VALID ? 0 : ROCKER_OP_FLAG_REMOVE) |
+		    ROCKER_OP_FLAG_NOWAIT;
+	__be32 ip_addr = *(__be32 *)n->primary_key;
+
+	return rocker_port_ipv4_neigh(rocker_port, NULL, flags, ip_addr, n->ha);
+}
+
+static int rocker_netevent_event(struct notifier_block *unused,
+				 unsigned long event, void *ptr)
+{
+	struct net_device *dev;
+	struct neighbour *n = ptr;
+	int err;
+
+	switch (event) {
+	case NETEVENT_NEIGH_UPDATE:
+		if (n->tbl != &arp_tbl)
+			return NOTIFY_DONE;
+		dev = n->dev;
+		if (!rocker_port_dev_check(dev))
+			return NOTIFY_DONE;
+		err = rocker_neigh_update(dev, n);
+		if (err)
+			netdev_warn(dev,
+				    "failed to handle neigh update (err %d)\n",
+				    err);
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block rocker_netevent_nb __read_mostly = {
+	.notifier_call = rocker_netevent_event,
+};
+
+/***********************
+ * Module init and exit
+ ***********************/
+
+static int __init rocker_module_init(void)
+{
+	int err;
+
+	register_netdevice_notifier(&rocker_netdevice_nb);
+	register_netevent_notifier(&rocker_netevent_nb);
+	err = pci_register_driver(&rocker_pci_driver);
+	if (err)
+		goto err_pci_register_driver;
+	return 0;
+
+err_pci_register_driver:
+	unregister_netevent_notifier(&rocker_netevent_nb);
+	unregister_netdevice_notifier(&rocker_netdevice_nb);
+	return err;
+}
+
+static void __exit rocker_module_exit(void)
+{
+	unregister_netevent_notifier(&rocker_netevent_nb);
+	unregister_netdevice_notifier(&rocker_netdevice_nb);
+	pci_unregister_driver(&rocker_pci_driver);
+}
+
+module_init(rocker_module_init);
+module_exit(rocker_module_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Jiri Pirko <jiri@resnulli.us>");
+MODULE_AUTHOR("Scott Feldman <sfeldma@gmail.com>");
+MODULE_DESCRIPTION("Rocker switch device driver");
+MODULE_DEVICE_TABLE(pci, rocker_pci_id_table);
-- 
cgit v0.10.2


From de1521923c0f2ee9c6328ad6b3e46e662d88a3cd Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:42 +0100
Subject: rocker: push tlv processing into separate files

Carve out TLV processing helpers into separate files.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/Makefile b/drivers/net/ethernet/rocker/Makefile
index 2bed423..47f3492 100644
--- a/drivers/net/ethernet/rocker/Makefile
+++ b/drivers/net/ethernet/rocker/Makefile
@@ -3,4 +3,4 @@
 #
 
 obj-$(CONFIG_ROCKER) += rocker.o
-rocker-y := rocker_main.o
+rocker-y := rocker_main.o rocker_tlv.o
diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
new file mode 100644
index 0000000..23b5e3d
--- /dev/null
+++ b/drivers/net/ethernet/rocker/rocker.h
@@ -0,0 +1,27 @@
+/*
+ * drivers/net/ethernet/rocker/rocker.h - Rocker switch device driver
+ * Copyright (c) 2014-2016 Jiri Pirko <jiri@mellanox.com>
+ * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _ROCKER_H
+#define _ROCKER_H
+
+#include <linux/types.h>
+
+#include "rocker_hw.h"
+
+struct rocker_desc_info {
+	char *data; /* mapped */
+	size_t data_size;
+	size_t tlv_size;
+	struct rocker_desc *desc;
+	dma_addr_t mapaddr;
+};
+
+#endif
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 44fa9ac..eecbbd5 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -40,6 +40,8 @@
 #include <generated/utsrelease.h>
 
 #include "rocker_hw.h"
+#include "rocker.h"
+#include "rocker_tlv.h"
 
 static const char rocker_driver_name[] = "rocker";
 
@@ -177,14 +179,6 @@ struct rocker_neigh_tbl_entry {
 	bool ttl_check;
 };
 
-struct rocker_desc_info {
-	char *data; /* mapped */
-	size_t data_size;
-	size_t tlv_size;
-	struct rocker_desc *desc;
-	dma_addr_t mapaddr;
-};
-
 struct rocker_dma_ring_info {
 	size_t size;
 	u32 head;
@@ -661,218 +655,6 @@ free_irq:
 	return err;
 }
 
-/******
- * TLV
- ******/
-
-#define ROCKER_TLV_ALIGNTO 8U
-#define ROCKER_TLV_ALIGN(len) \
-	(((len) + ROCKER_TLV_ALIGNTO - 1) & ~(ROCKER_TLV_ALIGNTO - 1))
-#define ROCKER_TLV_HDRLEN ROCKER_TLV_ALIGN(sizeof(struct rocker_tlv))
-
-/*  <------- ROCKER_TLV_HDRLEN -------> <--- ROCKER_TLV_ALIGN(payload) --->
- * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
- * |             Header          | Pad |           Payload           | Pad |
- * |      (struct rocker_tlv)    | ing |                             | ing |
- * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
- *  <--------------------------- tlv->len -------------------------->
- */
-
-static struct rocker_tlv *rocker_tlv_next(const struct rocker_tlv *tlv,
-					  int *remaining)
-{
-	int totlen = ROCKER_TLV_ALIGN(tlv->len);
-
-	*remaining -= totlen;
-	return (struct rocker_tlv *) ((char *) tlv + totlen);
-}
-
-static int rocker_tlv_ok(const struct rocker_tlv *tlv, int remaining)
-{
-	return remaining >= (int) ROCKER_TLV_HDRLEN &&
-	       tlv->len >= ROCKER_TLV_HDRLEN &&
-	       tlv->len <= remaining;
-}
-
-#define rocker_tlv_for_each(pos, head, len, rem)	\
-	for (pos = head, rem = len;			\
-	     rocker_tlv_ok(pos, rem);			\
-	     pos = rocker_tlv_next(pos, &(rem)))
-
-#define rocker_tlv_for_each_nested(pos, tlv, rem)	\
-	rocker_tlv_for_each(pos, rocker_tlv_data(tlv),	\
-			    rocker_tlv_len(tlv), rem)
-
-static int rocker_tlv_attr_size(int payload)
-{
-	return ROCKER_TLV_HDRLEN + payload;
-}
-
-static int rocker_tlv_total_size(int payload)
-{
-	return ROCKER_TLV_ALIGN(rocker_tlv_attr_size(payload));
-}
-
-static int rocker_tlv_padlen(int payload)
-{
-	return rocker_tlv_total_size(payload) - rocker_tlv_attr_size(payload);
-}
-
-static int rocker_tlv_type(const struct rocker_tlv *tlv)
-{
-	return tlv->type;
-}
-
-static void *rocker_tlv_data(const struct rocker_tlv *tlv)
-{
-	return (char *) tlv + ROCKER_TLV_HDRLEN;
-}
-
-static int rocker_tlv_len(const struct rocker_tlv *tlv)
-{
-	return tlv->len - ROCKER_TLV_HDRLEN;
-}
-
-static u8 rocker_tlv_get_u8(const struct rocker_tlv *tlv)
-{
-	return *(u8 *) rocker_tlv_data(tlv);
-}
-
-static u16 rocker_tlv_get_u16(const struct rocker_tlv *tlv)
-{
-	return *(u16 *) rocker_tlv_data(tlv);
-}
-
-static __be16 rocker_tlv_get_be16(const struct rocker_tlv *tlv)
-{
-	return *(__be16 *) rocker_tlv_data(tlv);
-}
-
-static u32 rocker_tlv_get_u32(const struct rocker_tlv *tlv)
-{
-	return *(u32 *) rocker_tlv_data(tlv);
-}
-
-static u64 rocker_tlv_get_u64(const struct rocker_tlv *tlv)
-{
-	return *(u64 *) rocker_tlv_data(tlv);
-}
-
-static void rocker_tlv_parse(const struct rocker_tlv **tb, int maxtype,
-			     const char *buf, int buf_len)
-{
-	const struct rocker_tlv *tlv;
-	const struct rocker_tlv *head = (const struct rocker_tlv *) buf;
-	int rem;
-
-	memset(tb, 0, sizeof(struct rocker_tlv *) * (maxtype + 1));
-
-	rocker_tlv_for_each(tlv, head, buf_len, rem) {
-		u32 type = rocker_tlv_type(tlv);
-
-		if (type > 0 && type <= maxtype)
-			tb[type] = tlv;
-	}
-}
-
-static void rocker_tlv_parse_nested(const struct rocker_tlv **tb, int maxtype,
-				    const struct rocker_tlv *tlv)
-{
-	rocker_tlv_parse(tb, maxtype, rocker_tlv_data(tlv),
-			 rocker_tlv_len(tlv));
-}
-
-static void rocker_tlv_parse_desc(const struct rocker_tlv **tb, int maxtype,
-				  const struct rocker_desc_info *desc_info)
-{
-	rocker_tlv_parse(tb, maxtype, desc_info->data,
-			 desc_info->desc->tlv_size);
-}
-
-static struct rocker_tlv *rocker_tlv_start(struct rocker_desc_info *desc_info)
-{
-	return (struct rocker_tlv *) ((char *) desc_info->data +
-					       desc_info->tlv_size);
-}
-
-static int rocker_tlv_put(struct rocker_desc_info *desc_info,
-			  int attrtype, int attrlen, const void *data)
-{
-	int tail_room = desc_info->data_size - desc_info->tlv_size;
-	int total_size = rocker_tlv_total_size(attrlen);
-	struct rocker_tlv *tlv;
-
-	if (unlikely(tail_room < total_size))
-		return -EMSGSIZE;
-
-	tlv = rocker_tlv_start(desc_info);
-	desc_info->tlv_size += total_size;
-	tlv->type = attrtype;
-	tlv->len = rocker_tlv_attr_size(attrlen);
-	memcpy(rocker_tlv_data(tlv), data, attrlen);
-	memset((char *) tlv + tlv->len, 0, rocker_tlv_padlen(attrlen));
-	return 0;
-}
-
-static int rocker_tlv_put_u8(struct rocker_desc_info *desc_info,
-			     int attrtype, u8 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u8), &value);
-}
-
-static int rocker_tlv_put_u16(struct rocker_desc_info *desc_info,
-			      int attrtype, u16 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u16), &value);
-}
-
-static int rocker_tlv_put_be16(struct rocker_desc_info *desc_info,
-			       int attrtype, __be16 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(__be16), &value);
-}
-
-static int rocker_tlv_put_u32(struct rocker_desc_info *desc_info,
-			      int attrtype, u32 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u32), &value);
-}
-
-static int rocker_tlv_put_be32(struct rocker_desc_info *desc_info,
-			       int attrtype, __be32 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(__be32), &value);
-}
-
-static int rocker_tlv_put_u64(struct rocker_desc_info *desc_info,
-			      int attrtype, u64 value)
-{
-	return rocker_tlv_put(desc_info, attrtype, sizeof(u64), &value);
-}
-
-static struct rocker_tlv *
-rocker_tlv_nest_start(struct rocker_desc_info *desc_info, int attrtype)
-{
-	struct rocker_tlv *start = rocker_tlv_start(desc_info);
-
-	if (rocker_tlv_put(desc_info, attrtype, 0, NULL) < 0)
-		return NULL;
-
-	return start;
-}
-
-static void rocker_tlv_nest_end(struct rocker_desc_info *desc_info,
-				struct rocker_tlv *start)
-{
-	start->len = (char *) rocker_tlv_start(desc_info) - (char *) start;
-}
-
-static void rocker_tlv_nest_cancel(struct rocker_desc_info *desc_info,
-				   const struct rocker_tlv *start)
-{
-	desc_info->tlv_size = (const char *) start - desc_info->data;
-}
-
 /******************************************
  * DMA rings and descriptors manipulations
  ******************************************/
diff --git a/drivers/net/ethernet/rocker/rocker_tlv.c b/drivers/net/ethernet/rocker/rocker_tlv.c
new file mode 100644
index 0000000..8185118
--- /dev/null
+++ b/drivers/net/ethernet/rocker/rocker_tlv.c
@@ -0,0 +1,53 @@
+/*
+ * drivers/net/ethernet/rocker/rocker_tlv.c - Rocker switch device driver
+ * Copyright (c) 2014-2016 Jiri Pirko <jiri@mellanox.com>
+ * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+
+#include "rocker_hw.h"
+#include "rocker_tlv.h"
+
+void rocker_tlv_parse(const struct rocker_tlv **tb, int maxtype,
+		      const char *buf, int buf_len)
+{
+	const struct rocker_tlv *tlv;
+	const struct rocker_tlv *head = (const struct rocker_tlv *) buf;
+	int rem;
+
+	memset(tb, 0, sizeof(struct rocker_tlv *) * (maxtype + 1));
+
+	rocker_tlv_for_each(tlv, head, buf_len, rem) {
+		u32 type = rocker_tlv_type(tlv);
+
+		if (type > 0 && type <= maxtype)
+			tb[type] = tlv;
+	}
+}
+
+int rocker_tlv_put(struct rocker_desc_info *desc_info,
+		   int attrtype, int attrlen, const void *data)
+{
+	int tail_room = desc_info->data_size - desc_info->tlv_size;
+	int total_size = rocker_tlv_total_size(attrlen);
+	struct rocker_tlv *tlv;
+
+	if (unlikely(tail_room < total_size))
+		return -EMSGSIZE;
+
+	tlv = rocker_tlv_start(desc_info);
+	desc_info->tlv_size += total_size;
+	tlv->type = attrtype;
+	tlv->len = rocker_tlv_attr_size(attrlen);
+	memcpy(rocker_tlv_data(tlv), data, attrlen);
+	memset((char *) tlv + tlv->len, 0, rocker_tlv_padlen(attrlen));
+	return 0;
+}
diff --git a/drivers/net/ethernet/rocker/rocker_tlv.h b/drivers/net/ethernet/rocker/rocker_tlv.h
new file mode 100644
index 0000000..a63ef82
--- /dev/null
+++ b/drivers/net/ethernet/rocker/rocker_tlv.h
@@ -0,0 +1,201 @@
+/*
+ * drivers/net/ethernet/rocker/rocker_tlv.h - Rocker switch device driver
+ * Copyright (c) 2014-2016 Jiri Pirko <jiri@mellanox.com>
+ * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _ROCKER_TLV_H
+#define _ROCKER_TLV_H
+
+#include <linux/types.h>
+
+#include "rocker_hw.h"
+#include "rocker.h"
+
+#define ROCKER_TLV_ALIGNTO 8U
+#define ROCKER_TLV_ALIGN(len) \
+	(((len) + ROCKER_TLV_ALIGNTO - 1) & ~(ROCKER_TLV_ALIGNTO - 1))
+#define ROCKER_TLV_HDRLEN ROCKER_TLV_ALIGN(sizeof(struct rocker_tlv))
+
+/*  <------- ROCKER_TLV_HDRLEN -------> <--- ROCKER_TLV_ALIGN(payload) --->
+ * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
+ * |             Header          | Pad |           Payload           | Pad |
+ * |      (struct rocker_tlv)    | ing |                             | ing |
+ * +-----------------------------+- - -+- - - - - - - - - - - - - - -+- - -+
+ *  <--------------------------- tlv->len -------------------------->
+ */
+
+static inline struct rocker_tlv *rocker_tlv_next(const struct rocker_tlv *tlv,
+						 int *remaining)
+{
+	int totlen = ROCKER_TLV_ALIGN(tlv->len);
+
+	*remaining -= totlen;
+	return (struct rocker_tlv *) ((char *) tlv + totlen);
+}
+
+static inline int rocker_tlv_ok(const struct rocker_tlv *tlv, int remaining)
+{
+	return remaining >= (int) ROCKER_TLV_HDRLEN &&
+	       tlv->len >= ROCKER_TLV_HDRLEN &&
+	       tlv->len <= remaining;
+}
+
+#define rocker_tlv_for_each(pos, head, len, rem)	\
+	for (pos = head, rem = len;			\
+	     rocker_tlv_ok(pos, rem);			\
+	     pos = rocker_tlv_next(pos, &(rem)))
+
+#define rocker_tlv_for_each_nested(pos, tlv, rem)	\
+	rocker_tlv_for_each(pos, rocker_tlv_data(tlv),	\
+			    rocker_tlv_len(tlv), rem)
+
+static inline int rocker_tlv_attr_size(int payload)
+{
+	return ROCKER_TLV_HDRLEN + payload;
+}
+
+static inline int rocker_tlv_total_size(int payload)
+{
+	return ROCKER_TLV_ALIGN(rocker_tlv_attr_size(payload));
+}
+
+static inline int rocker_tlv_padlen(int payload)
+{
+	return rocker_tlv_total_size(payload) - rocker_tlv_attr_size(payload);
+}
+
+static inline int rocker_tlv_type(const struct rocker_tlv *tlv)
+{
+	return tlv->type;
+}
+
+static inline void *rocker_tlv_data(const struct rocker_tlv *tlv)
+{
+	return (char *) tlv + ROCKER_TLV_HDRLEN;
+}
+
+static inline int rocker_tlv_len(const struct rocker_tlv *tlv)
+{
+	return tlv->len - ROCKER_TLV_HDRLEN;
+}
+
+static inline u8 rocker_tlv_get_u8(const struct rocker_tlv *tlv)
+{
+	return *(u8 *) rocker_tlv_data(tlv);
+}
+
+static inline u16 rocker_tlv_get_u16(const struct rocker_tlv *tlv)
+{
+	return *(u16 *) rocker_tlv_data(tlv);
+}
+
+static inline __be16 rocker_tlv_get_be16(const struct rocker_tlv *tlv)
+{
+	return *(__be16 *) rocker_tlv_data(tlv);
+}
+
+static inline u32 rocker_tlv_get_u32(const struct rocker_tlv *tlv)
+{
+	return *(u32 *) rocker_tlv_data(tlv);
+}
+
+static inline u64 rocker_tlv_get_u64(const struct rocker_tlv *tlv)
+{
+	return *(u64 *) rocker_tlv_data(tlv);
+}
+
+void rocker_tlv_parse(const struct rocker_tlv **tb, int maxtype,
+		      const char *buf, int buf_len);
+
+static inline void rocker_tlv_parse_nested(const struct rocker_tlv **tb,
+					   int maxtype,
+					   const struct rocker_tlv *tlv)
+{
+	rocker_tlv_parse(tb, maxtype, rocker_tlv_data(tlv),
+			 rocker_tlv_len(tlv));
+}
+
+static inline void
+rocker_tlv_parse_desc(const struct rocker_tlv **tb, int maxtype,
+		      const struct rocker_desc_info *desc_info)
+{
+	rocker_tlv_parse(tb, maxtype, desc_info->data,
+			 desc_info->desc->tlv_size);
+}
+
+static inline struct rocker_tlv *
+rocker_tlv_start(struct rocker_desc_info *desc_info)
+{
+	return (struct rocker_tlv *) ((char *) desc_info->data +
+					       desc_info->tlv_size);
+}
+
+int rocker_tlv_put(struct rocker_desc_info *desc_info,
+		   int attrtype, int attrlen, const void *data);
+
+static inline int rocker_tlv_put_u8(struct rocker_desc_info *desc_info,
+				    int attrtype, u8 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u8), &value);
+}
+
+static inline int rocker_tlv_put_u16(struct rocker_desc_info *desc_info,
+				     int attrtype, u16 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u16), &value);
+}
+
+static inline int rocker_tlv_put_be16(struct rocker_desc_info *desc_info,
+				      int attrtype, __be16 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(__be16), &value);
+}
+
+static inline int rocker_tlv_put_u32(struct rocker_desc_info *desc_info,
+				     int attrtype, u32 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u32), &value);
+}
+
+static inline int rocker_tlv_put_be32(struct rocker_desc_info *desc_info,
+				      int attrtype, __be32 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(__be32), &value);
+}
+
+static inline int rocker_tlv_put_u64(struct rocker_desc_info *desc_info,
+				     int attrtype, u64 value)
+{
+	return rocker_tlv_put(desc_info, attrtype, sizeof(u64), &value);
+}
+
+static inline struct rocker_tlv *
+rocker_tlv_nest_start(struct rocker_desc_info *desc_info, int attrtype)
+{
+	struct rocker_tlv *start = rocker_tlv_start(desc_info);
+
+	if (rocker_tlv_put(desc_info, attrtype, 0, NULL) < 0)
+		return NULL;
+
+	return start;
+}
+
+static inline void rocker_tlv_nest_end(struct rocker_desc_info *desc_info,
+				       struct rocker_tlv *start)
+{
+	start->len = (char *) rocker_tlv_start(desc_info) - (char *) start;
+}
+
+static inline void rocker_tlv_nest_cancel(struct rocker_desc_info *desc_info,
+					  const struct rocker_tlv *start)
+{
+	desc_info->tlv_size = (const char *) start - desc_info->data;
+}
+
+#endif
-- 
cgit v0.10.2


From e1ba3dee77c450fe3cea00a334f30174f5cc0aa1 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:43 +0100
Subject: rocker: implement get settings mode command

Introduce a helper to ask HW for the port mode (world).

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index eecbbd5..1a99b35 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -1495,6 +1495,30 @@ rocker_cmd_get_port_settings_macaddr_proc(const struct rocker_port *rocker_port,
 	return 0;
 }
 
+static int
+rocker_cmd_get_port_settings_mode_proc(const struct rocker_port *rocker_port,
+				       const struct rocker_desc_info *desc_info,
+				       void *priv)
+{
+	u8 *p_mode = priv;
+	const struct rocker_tlv *attrs[ROCKER_TLV_CMD_MAX + 1];
+	const struct rocker_tlv *info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MAX + 1];
+	const struct rocker_tlv *attr;
+
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_CMD_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_CMD_INFO])
+		return -EIO;
+
+	rocker_tlv_parse_nested(info_attrs, ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
+				attrs[ROCKER_TLV_CMD_INFO]);
+	attr = info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MODE];
+	if (!attr)
+		return -EIO;
+
+	*p_mode = rocker_tlv_get_u8(info_attrs[ROCKER_TLV_CMD_PORT_SETTINGS_MODE]);
+	return 0;
+}
+
 struct port_name {
 	char *buf;
 	size_t len;
@@ -1660,6 +1684,14 @@ static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
 			       macaddr);
 }
 
+static int rocker_cmd_get_port_settings_mode(struct rocker_port *rocker_port,
+					     u8 *p_mode)
+{
+	return rocker_cmd_exec(rocker_port, NULL, 0,
+			       rocker_cmd_get_port_settings_prep, NULL,
+			       rocker_cmd_get_port_settings_mode_proc, p_mode);
+}
+
 static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
 						struct ethtool_cmd *ecmd)
 {
-- 
cgit v0.10.2


From 0514c4e809a93544253ba5fb797379fa9cc45cce Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:44 +0100
Subject: rocker: move rocker and rocker_port structs into header

And take some other related thing along. They are going to be pushed
into of-dpa part anyway.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
index 23b5e3d..05c1e1a 100644
--- a/drivers/net/ethernet/rocker/rocker.h
+++ b/drivers/net/ethernet/rocker/rocker.h
@@ -13,6 +13,8 @@
 #define _ROCKER_H
 
 #include <linux/types.h>
+#include <linux/hashtable.h>
+#include <linux/if_vlan.h>
 
 #include "rocker_hw.h"
 
@@ -24,4 +26,77 @@ struct rocker_desc_info {
 	dma_addr_t mapaddr;
 };
 
+struct rocker_dma_ring_info {
+	size_t size;
+	u32 head;
+	u32 tail;
+	struct rocker_desc *desc; /* mapped */
+	dma_addr_t mapaddr;
+	struct rocker_desc_info *desc_info;
+	unsigned int type;
+};
+
+struct rocker;
+
+enum {
+	ROCKER_CTRL_LINK_LOCAL_MCAST,
+	ROCKER_CTRL_LOCAL_ARP,
+	ROCKER_CTRL_IPV4_MCAST,
+	ROCKER_CTRL_IPV6_MCAST,
+	ROCKER_CTRL_DFLT_BRIDGING,
+	ROCKER_CTRL_DFLT_OVS,
+	ROCKER_CTRL_MAX,
+};
+
+#define ROCKER_INTERNAL_VLAN_ID_BASE	0x0f00
+#define ROCKER_N_INTERNAL_VLANS		255
+#define ROCKER_VLAN_BITMAP_LEN		BITS_TO_LONGS(VLAN_N_VID)
+#define ROCKER_INTERNAL_VLAN_BITMAP_LEN	BITS_TO_LONGS(ROCKER_N_INTERNAL_VLANS)
+
+struct rocker_port {
+	struct net_device *dev;
+	struct net_device *bridge_dev;
+	struct rocker *rocker;
+	unsigned int port_number;
+	u32 pport;
+	__be16 internal_vlan_id;
+	int stp_state;
+	u32 brport_flags;
+	unsigned long ageing_time;
+	bool ctrls[ROCKER_CTRL_MAX];
+	unsigned long vlan_bitmap[ROCKER_VLAN_BITMAP_LEN];
+	struct napi_struct napi_tx;
+	struct napi_struct napi_rx;
+	struct rocker_dma_ring_info tx_ring;
+	struct rocker_dma_ring_info rx_ring;
+};
+
+struct rocker {
+	struct pci_dev *pdev;
+	u8 __iomem *hw_addr;
+	struct msix_entry *msix_entries;
+	unsigned int port_count;
+	struct rocker_port **ports;
+	struct {
+		u64 id;
+	} hw;
+	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
+	struct rocker_dma_ring_info cmd_ring;
+	struct rocker_dma_ring_info event_ring;
+	DECLARE_HASHTABLE(flow_tbl, 16);
+	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
+	u64 flow_tbl_next_cookie;
+	DECLARE_HASHTABLE(group_tbl, 16);
+	spinlock_t group_tbl_lock;		/* for group tbl accesses */
+	struct timer_list fdb_cleanup_timer;
+	DECLARE_HASHTABLE(fdb_tbl, 16);
+	spinlock_t fdb_tbl_lock;		/* for fdb tbl accesses */
+	unsigned long internal_vlan_bitmap[ROCKER_INTERNAL_VLAN_BITMAP_LEN];
+	DECLARE_HASHTABLE(internal_vlan_tbl, 8);
+	spinlock_t internal_vlan_tbl_lock;	/* for vlan tbl accesses */
+	DECLARE_HASHTABLE(neigh_tbl, 16);
+	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
+	u32 neigh_tbl_next_index;
+};
+
 #endif
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 1a99b35..a67a6c7 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -179,79 +179,6 @@ struct rocker_neigh_tbl_entry {
 	bool ttl_check;
 };
 
-struct rocker_dma_ring_info {
-	size_t size;
-	u32 head;
-	u32 tail;
-	struct rocker_desc *desc; /* mapped */
-	dma_addr_t mapaddr;
-	struct rocker_desc_info *desc_info;
-	unsigned int type;
-};
-
-struct rocker;
-
-enum {
-	ROCKER_CTRL_LINK_LOCAL_MCAST,
-	ROCKER_CTRL_LOCAL_ARP,
-	ROCKER_CTRL_IPV4_MCAST,
-	ROCKER_CTRL_IPV6_MCAST,
-	ROCKER_CTRL_DFLT_BRIDGING,
-	ROCKER_CTRL_DFLT_OVS,
-	ROCKER_CTRL_MAX,
-};
-
-#define ROCKER_INTERNAL_VLAN_ID_BASE	0x0f00
-#define ROCKER_N_INTERNAL_VLANS		255
-#define ROCKER_VLAN_BITMAP_LEN		BITS_TO_LONGS(VLAN_N_VID)
-#define ROCKER_INTERNAL_VLAN_BITMAP_LEN	BITS_TO_LONGS(ROCKER_N_INTERNAL_VLANS)
-
-struct rocker_port {
-	struct net_device *dev;
-	struct net_device *bridge_dev;
-	struct rocker *rocker;
-	unsigned int port_number;
-	u32 pport;
-	__be16 internal_vlan_id;
-	int stp_state;
-	u32 brport_flags;
-	unsigned long ageing_time;
-	bool ctrls[ROCKER_CTRL_MAX];
-	unsigned long vlan_bitmap[ROCKER_VLAN_BITMAP_LEN];
-	struct napi_struct napi_tx;
-	struct napi_struct napi_rx;
-	struct rocker_dma_ring_info tx_ring;
-	struct rocker_dma_ring_info rx_ring;
-};
-
-struct rocker {
-	struct pci_dev *pdev;
-	u8 __iomem *hw_addr;
-	struct msix_entry *msix_entries;
-	unsigned int port_count;
-	struct rocker_port **ports;
-	struct {
-		u64 id;
-	} hw;
-	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
-	struct rocker_dma_ring_info cmd_ring;
-	struct rocker_dma_ring_info event_ring;
-	DECLARE_HASHTABLE(flow_tbl, 16);
-	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
-	u64 flow_tbl_next_cookie;
-	DECLARE_HASHTABLE(group_tbl, 16);
-	spinlock_t group_tbl_lock;		/* for group tbl accesses */
-	struct timer_list fdb_cleanup_timer;
-	DECLARE_HASHTABLE(fdb_tbl, 16);
-	spinlock_t fdb_tbl_lock;		/* for fdb tbl accesses */
-	unsigned long internal_vlan_bitmap[ROCKER_INTERNAL_VLAN_BITMAP_LEN];
-	DECLARE_HASHTABLE(internal_vlan_tbl, 8);
-	spinlock_t internal_vlan_tbl_lock;	/* for vlan tbl accesses */
-	DECLARE_HASHTABLE(neigh_tbl, 16);
-	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
-	u32 neigh_tbl_next_index;
-};
-
 static const u8 zero_mac[ETH_ALEN]   = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
 static const u8 ff_mac[ETH_ALEN]     = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
 static const u8 ll_mac[ETH_ALEN]     = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
-- 
cgit v0.10.2


From e420114eef4a3a5025a243b89b0dc343101e3d3c Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:45 +0100
Subject: rocker: introduce worlds infrastructure

This is another step on the way to per-world clean cut. Introduce world
ops hooks which each world can implement in world-specific way.
Also introduce world infrastructure along with OF-DPA world stub.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/Makefile b/drivers/net/ethernet/rocker/Makefile
index 47f3492..faa36ac 100644
--- a/drivers/net/ethernet/rocker/Makefile
+++ b/drivers/net/ethernet/rocker/Makefile
@@ -3,4 +3,4 @@
 #
 
 obj-$(CONFIG_ROCKER) += rocker.o
-rocker-y := rocker_main.o rocker_tlv.o
+rocker-y := rocker_main.o rocker_tlv.o rocker_ofdpa.o
diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
index 05c1e1a..5fd0e35 100644
--- a/drivers/net/ethernet/rocker/rocker.h
+++ b/drivers/net/ethernet/rocker/rocker.h
@@ -12,9 +12,13 @@
 #ifndef _ROCKER_H
 #define _ROCKER_H
 
+#include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/hashtable.h>
 #include <linux/if_vlan.h>
+#include <linux/netdevice.h>
+#include <net/neighbour.h>
+#include <net/switchdev.h>
 
 #include "rocker_hw.h"
 
@@ -57,6 +61,7 @@ struct rocker_port {
 	struct net_device *dev;
 	struct net_device *bridge_dev;
 	struct rocker *rocker;
+	void *wpriv;
 	unsigned int port_number;
 	u32 pport;
 	__be16 internal_vlan_id;
@@ -71,6 +76,8 @@ struct rocker_port {
 	struct rocker_dma_ring_info rx_ring;
 };
 
+struct rocker_world_ops;
+
 struct rocker {
 	struct pci_dev *pdev;
 	u8 __iomem *hw_addr;
@@ -83,6 +90,8 @@ struct rocker {
 	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
 	struct rocker_dma_ring_info cmd_ring;
 	struct rocker_dma_ring_info event_ring;
+	struct rocker_world_ops *wops;
+	void *wpriv;
 	DECLARE_HASHTABLE(flow_tbl, 16);
 	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
 	u64 flow_tbl_next_cookie;
@@ -99,4 +108,64 @@ struct rocker {
 	u32 neigh_tbl_next_index;
 };
 
+struct rocker_world_ops {
+	const char *kind;
+	size_t priv_size;
+	size_t port_priv_size;
+	u8 mode;
+	int (*init)(struct rocker *rocker);
+	void (*fini)(struct rocker *rocker);
+	int (*port_pre_init)(struct rocker_port *rocker_port);
+	int (*port_init)(struct rocker_port *rocker_port);
+	void (*port_fini)(struct rocker_port *rocker_port);
+	void (*port_post_fini)(struct rocker_port *rocker_port);
+	int (*port_open)(struct rocker_port *rocker_port);
+	void (*port_stop)(struct rocker_port *rocker_port);
+	int (*port_attr_stp_state_set)(struct rocker_port *rocker_port,
+				       u8 state,
+				       struct switchdev_trans *trans);
+	int (*port_attr_bridge_flags_set)(struct rocker_port *rocker_port,
+					  unsigned long brport_flags,
+					  struct switchdev_trans *trans);
+	int (*port_attr_bridge_flags_get)(const struct rocker_port *rocker_port,
+					  unsigned long *p_brport_flags);
+	int (*port_attr_bridge_ageing_time_set)(struct rocker_port *rocker_port,
+						u32 ageing_time,
+						struct switchdev_trans *trans);
+	int (*port_obj_vlan_add)(struct rocker_port *rocker_port,
+				 const struct switchdev_obj_port_vlan *vlan,
+				 struct switchdev_trans *trans);
+	int (*port_obj_vlan_del)(struct rocker_port *rocker_port,
+				 const struct switchdev_obj_port_vlan *vlan);
+	int (*port_obj_vlan_dump)(const struct rocker_port *rocker_port,
+				  struct switchdev_obj_port_vlan *vlan,
+				  switchdev_obj_dump_cb_t *cb);
+	int (*port_obj_fib4_add)(struct rocker_port *rocker_port,
+				 const struct switchdev_obj_ipv4_fib *fib4,
+				 struct switchdev_trans *trans);
+	int (*port_obj_fib4_del)(struct rocker_port *rocker_port,
+				 const struct switchdev_obj_ipv4_fib *fib4);
+	int (*port_obj_fdb_add)(struct rocker_port *rocker_port,
+				const struct switchdev_obj_port_fdb *fdb,
+				struct switchdev_trans *trans);
+	int (*port_obj_fdb_del)(struct rocker_port *rocker_port,
+				const struct switchdev_obj_port_fdb *fdb);
+	int (*port_obj_fdb_dump)(const struct rocker_port *rocker_port,
+				 struct switchdev_obj_port_fdb *fdb,
+				 switchdev_obj_dump_cb_t *cb);
+	int (*port_master_linked)(struct rocker_port *rocker_port,
+				  struct net_device *master);
+	int (*port_master_unlinked)(struct rocker_port *rocker_port,
+				    struct net_device *master);
+	int (*port_neigh_update)(struct rocker_port *rocker_port,
+				 struct neighbour *n);
+	int (*port_neigh_destroy)(struct rocker_port *rocker_port,
+				  struct neighbour *n);
+	int (*port_ev_mac_vlan_seen)(struct rocker_port *rocker_port,
+				     const unsigned char *addr,
+				     __be16 vlan_id);
+};
+
+extern struct rocker_world_ops rocker_ofdpa_ops;
+
 #endif
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index a67a6c7..871ccbe 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -1164,6 +1164,9 @@ static int rocker_port_fdb(struct rocker_port *rocker_port,
 			   struct switchdev_trans *trans,
 			   const unsigned char *addr,
 			   __be16 vlan_id, int flags);
+static int rocker_world_port_ev_mac_vlan_seen(struct rocker_port *rocker_port,
+					      const unsigned char *addr,
+					      __be16 vlan_id);
 
 static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
 				      const struct rocker_tlv *info)
@@ -1174,6 +1177,7 @@ static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
 	const unsigned char *addr;
 	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_LEARNED;
 	__be16 vlan_id;
+	int err;
 
 	rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_MAC_VLAN_MAX, info);
 	if (!attrs[ROCKER_TLV_EVENT_MAC_VLAN_PPORT] ||
@@ -1190,6 +1194,10 @@ static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
 
 	rocker_port = rocker->ports[port_number];
 
+	err = rocker_world_port_ev_mac_vlan_seen(rocker_port, addr, vlan_id);
+	if (err)
+		return err;
+
 	if (rocker_port->stp_state != BR_STATE_LEARNING &&
 	    rocker_port->stp_state != BR_STATE_FORWARDING)
 		return 0;
@@ -1651,6 +1659,335 @@ static int rocker_port_set_learning(struct rocker_port *rocker_port,
 			       NULL, NULL, NULL);
 }
 
+/**********************
+ * Worlds manipulation
+ **********************/
+
+static struct rocker_world_ops *rocker_world_ops[] = {
+	&rocker_ofdpa_ops,
+};
+
+#define ROCKER_WORLD_OPS_LEN ARRAY_SIZE(rocker_world_ops)
+
+static struct rocker_world_ops *rocker_world_ops_find(u8 mode)
+{
+	int i;
+
+	for (i = 0; i < ROCKER_WORLD_OPS_LEN; i++)
+		if (rocker_world_ops[i]->mode == mode)
+			return rocker_world_ops[i];
+	return NULL;
+}
+
+static int rocker_world_init(struct rocker *rocker, u8 mode)
+{
+	struct rocker_world_ops *wops;
+	int err;
+
+	wops = rocker_world_ops_find(mode);
+	if (!wops) {
+		dev_err(&rocker->pdev->dev, "port mode \"%d\" is not supported\n",
+			mode);
+		return -EINVAL;
+	}
+	rocker->wops = wops;
+	rocker->wpriv = kzalloc(wops->priv_size, GFP_KERNEL);
+	if (!rocker->wpriv)
+		return -ENOMEM;
+	if (!wops->init)
+		return 0;
+	err = wops->init(rocker);
+	if (err)
+		kfree(rocker->wpriv);
+	return err;
+}
+
+static void rocker_world_fini(struct rocker *rocker)
+{
+	struct rocker_world_ops *wops = rocker->wops;
+
+	if (!wops || !wops->fini)
+		return;
+	wops->fini(rocker);
+	kfree(rocker->wpriv);
+}
+
+static int rocker_world_check_init(struct rocker_port *rocker_port)
+{
+	struct rocker *rocker = rocker_port->rocker;
+	u8 mode;
+	int err;
+
+	err = rocker_cmd_get_port_settings_mode(rocker_port, &mode);
+	if (err) {
+		dev_err(&rocker->pdev->dev, "failed to get port mode\n");
+		return err;
+	}
+	if (rocker->wops) {
+		if (rocker->wops->mode != mode) {
+			dev_err(&rocker->pdev->dev, "hardware has ports in different worlds, which is not supported\n");
+			return err;
+		}
+		return 0;
+	}
+	return rocker_world_init(rocker, mode);
+}
+
+static int rocker_world_port_pre_init(struct rocker_port *rocker_port)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+	int err;
+
+	rocker_port->wpriv = kzalloc(wops->port_priv_size, GFP_KERNEL);
+	if (!rocker_port->wpriv)
+		return -ENOMEM;
+	if (!wops->port_pre_init)
+		return 0;
+	err = wops->port_pre_init(rocker_port);
+	if (err)
+		kfree(rocker_port->wpriv);
+	return 0;
+}
+
+static int rocker_world_port_init(struct rocker_port *rocker_port)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_init)
+		return 0;
+	return wops->port_init(rocker_port);
+}
+
+static void rocker_world_port_fini(struct rocker_port *rocker_port)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_fini)
+		return;
+	wops->port_fini(rocker_port);
+}
+
+static void rocker_world_port_post_fini(struct rocker_port *rocker_port)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_post_fini)
+		return;
+	wops->port_post_fini(rocker_port);
+	kfree(rocker_port->wpriv);
+}
+
+static int rocker_world_port_open(struct rocker_port *rocker_port)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_open)
+		return 0;
+	return wops->port_open(rocker_port);
+}
+
+static void rocker_world_port_stop(struct rocker_port *rocker_port)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_stop)
+		return;
+	wops->port_stop(rocker_port);
+}
+
+static int rocker_world_port_attr_stp_state_set(struct rocker_port *rocker_port,
+						u8 state,
+						struct switchdev_trans *trans)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_attr_stp_state_set)
+		return 0;
+	return wops->port_attr_stp_state_set(rocker_port, state, trans);
+}
+
+static int
+rocker_world_port_attr_bridge_flags_set(struct rocker_port *rocker_port,
+					unsigned long brport_flags,
+					struct switchdev_trans *trans)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_attr_bridge_flags_set)
+		return 0;
+	return wops->port_attr_bridge_flags_set(rocker_port, brport_flags,
+						trans);
+}
+
+static int
+rocker_world_port_attr_bridge_flags_get(const struct rocker_port *rocker_port,
+					unsigned long *p_brport_flags)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_attr_bridge_flags_get)
+		return 0;
+	return wops->port_attr_bridge_flags_get(rocker_port, p_brport_flags);
+}
+
+static int
+rocker_world_port_attr_bridge_ageing_time_set(struct rocker_port *rocker_port,
+					      u32 ageing_time,
+					      struct switchdev_trans *trans)
+
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_attr_bridge_ageing_time_set)
+		return 0;
+	return wops->port_attr_bridge_ageing_time_set(rocker_port, ageing_time,
+						      trans);
+}
+
+static int
+rocker_world_port_obj_vlan_add(struct rocker_port *rocker_port,
+			       const struct switchdev_obj_port_vlan *vlan,
+			       struct switchdev_trans *trans)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_vlan_add)
+		return 0;
+	return wops->port_obj_vlan_add(rocker_port, vlan, trans);
+}
+
+static int
+rocker_world_port_obj_vlan_del(struct rocker_port *rocker_port,
+			       const struct switchdev_obj_port_vlan *vlan)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_vlan_del)
+		return 0;
+	return wops->port_obj_vlan_del(rocker_port, vlan);
+}
+
+static int
+rocker_world_port_obj_vlan_dump(const struct rocker_port *rocker_port,
+				struct switchdev_obj_port_vlan *vlan,
+				switchdev_obj_dump_cb_t *cb)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_vlan_dump)
+		return 0;
+	return wops->port_obj_vlan_dump(rocker_port, vlan, cb);
+}
+
+static int
+rocker_world_port_obj_fib4_add(struct rocker_port *rocker_port,
+			       const struct switchdev_obj_ipv4_fib *fib4,
+			       struct switchdev_trans *trans)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_fib4_add)
+		return 0;
+	return wops->port_obj_fib4_add(rocker_port, fib4, trans);
+}
+
+static int
+rocker_world_port_obj_fib4_del(struct rocker_port *rocker_port,
+			       const struct switchdev_obj_ipv4_fib *fib4)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_fib4_del)
+		return 0;
+	return wops->port_obj_fib4_del(rocker_port, fib4);
+}
+
+static int
+rocker_world_port_obj_fdb_add(struct rocker_port *rocker_port,
+			      const struct switchdev_obj_port_fdb *fdb,
+			      struct switchdev_trans *trans)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_fdb_add)
+		return 0;
+	return wops->port_obj_fdb_add(rocker_port, fdb, trans);
+}
+
+static int
+rocker_world_port_obj_fdb_del(struct rocker_port *rocker_port,
+			      const struct switchdev_obj_port_fdb *fdb)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_fdb_del)
+		return 0;
+	return wops->port_obj_fdb_del(rocker_port, fdb);
+}
+
+static int
+rocker_world_port_obj_fdb_dump(const struct rocker_port *rocker_port,
+			       struct switchdev_obj_port_fdb *fdb,
+			       switchdev_obj_dump_cb_t *cb)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_obj_fdb_dump)
+		return 0;
+	return wops->port_obj_fdb_dump(rocker_port, fdb, cb);
+}
+
+static int rocker_world_port_master_linked(struct rocker_port *rocker_port,
+					   struct net_device *master)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_master_linked)
+		return 0;
+	return wops->port_master_linked(rocker_port, master);
+}
+
+static int rocker_world_port_master_unlinked(struct rocker_port *rocker_port,
+					     struct net_device *master)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_master_unlinked)
+		return 0;
+	return wops->port_master_unlinked(rocker_port, master);
+}
+
+static int rocker_world_port_neigh_update(struct rocker_port *rocker_port,
+					  struct neighbour *n)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_neigh_update)
+		return 0;
+	return wops->port_neigh_update(rocker_port, n);
+}
+
+static int rocker_world_port_neigh_destroy(struct rocker_port *rocker_port,
+					   struct neighbour *n)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_neigh_destroy)
+		return 0;
+	return wops->port_neigh_destroy(rocker_port, n);
+}
+
+static int rocker_world_port_ev_mac_vlan_seen(struct rocker_port *rocker_port,
+					      const unsigned char *addr,
+					      __be16 vlan_id)
+{
+	struct rocker_world_ops *wops = rocker_port->rocker->wops;
+
+	if (!wops->port_ev_mac_vlan_seen)
+		return 0;
+	return wops->port_ev_mac_vlan_seen(rocker_port, addr, vlan_id);
+}
+
 static int
 rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info,
 				const struct rocker_flow_tbl_entry *entry)
@@ -3799,6 +4136,12 @@ static int rocker_port_open(struct net_device *dev)
 		goto err_request_rx_irq;
 	}
 
+	err = rocker_world_port_open(rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "cannot open port in world\n");
+		goto err_world_port_open;
+	}
+
 	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
 	if (err)
 		goto err_fwd_enable;
@@ -3811,6 +4154,7 @@ static int rocker_port_open(struct net_device *dev)
 	return 0;
 
 err_fwd_enable:
+err_world_port_open:
 	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
 err_request_rx_irq:
 	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
@@ -3827,6 +4171,7 @@ static int rocker_port_stop(struct net_device *dev)
 	rocker_port_set_enable(rocker_port, false);
 	napi_disable(&rocker_port->napi_rx);
 	napi_disable(&rocker_port->napi_tx);
+	rocker_world_port_stop(rocker_port);
 	rocker_port_fwd_disable(rocker_port, NULL,
 				ROCKER_OP_FLAG_NOWAIT);
 	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
@@ -4037,9 +4382,14 @@ static void rocker_port_neigh_destroy(struct neighbour *n)
 	struct rocker_port *rocker_port = netdev_priv(n->dev);
 	int flags = ROCKER_OP_FLAG_REMOVE | ROCKER_OP_FLAG_NOWAIT;
 	__be32 ip_addr = *(__be32 *)n->primary_key;
+	int err;
 
 	rocker_port_ipv4_neigh(rocker_port, NULL,
 			       flags, ip_addr, n->ha);
+	err = rocker_world_port_neigh_destroy(rocker_port, n);
+	if (err)
+		netdev_warn(rocker_port->dev, "failed to handle neigh destroy (err %d)\n",
+			    err);
 }
 
 static const struct net_device_ops rocker_port_netdev_ops = {
@@ -4068,6 +4418,7 @@ static int rocker_port_attr_get(struct net_device *dev,
 {
 	const struct rocker_port *rocker_port = netdev_priv(dev);
 	const struct rocker *rocker = rocker_port->rocker;
+	int err = 0;
 
 	switch (attr->id) {
 	case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
@@ -4076,12 +4427,14 @@ static int rocker_port_attr_get(struct net_device *dev,
 		break;
 	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
 		attr->u.brport_flags = rocker_port->brport_flags;
+		err = rocker_world_port_attr_bridge_flags_get(rocker_port,
+							      &attr->u.brport_flags);
 		break;
 	default:
 		return -EOPNOTSUPP;
 	}
 
-	return 0;
+	return err;
 }
 
 static int rocker_port_brport_flags_set(struct rocker_port *rocker_port,
@@ -4125,14 +4478,29 @@ static int rocker_port_attr_set(struct net_device *dev,
 	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
 		err = rocker_port_stp_update(rocker_port, trans, 0,
 					     attr->u.stp_state);
+		if (err)
+			break;
+		err = rocker_world_port_attr_stp_state_set(rocker_port,
+							   attr->u.stp_state,
+							   trans);
 		break;
 	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
 		err = rocker_port_brport_flags_set(rocker_port, trans,
 						   attr->u.brport_flags);
+		if (err)
+			break;
+		err = rocker_world_port_attr_bridge_flags_set(rocker_port,
+							      attr->u.brport_flags,
+							      trans);
 		break;
 	case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME:
 		err = rocker_port_bridge_ageing_time(rocker_port, trans,
 						     attr->u.ageing_time);
+		if (err)
+			break;
+		err = rocker_world_port_attr_bridge_ageing_time_set(rocker_port,
+								    attr->u.ageing_time,
+								    trans);
 		break;
 	default:
 		err = -EOPNOTSUPP;
@@ -4204,16 +4572,31 @@ static int rocker_port_obj_add(struct net_device *dev,
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
 		err = rocker_port_vlans_add(rocker_port, trans,
 					    SWITCHDEV_OBJ_PORT_VLAN(obj));
+		if (err)
+			break;
+		err = rocker_world_port_obj_vlan_add(rocker_port,
+						     SWITCHDEV_OBJ_PORT_VLAN(obj),
+						     trans);
 		break;
 	case SWITCHDEV_OBJ_ID_IPV4_FIB:
 		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
 		err = rocker_port_fib_ipv4(rocker_port, trans,
 					   htonl(fib4->dst), fib4->dst_len,
 					   &fib4->fi, fib4->tb_id, 0);
+		if (err)
+			break;
+		err = rocker_world_port_obj_fib4_add(rocker_port,
+						     SWITCHDEV_OBJ_IPV4_FIB(obj),
+						     trans);
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
 		err = rocker_port_fdb_add(rocker_port, trans,
 					  SWITCHDEV_OBJ_PORT_FDB(obj));
+		if (err)
+			break;
+		err = rocker_world_port_obj_fdb_add(rocker_port,
+						    SWITCHDEV_OBJ_PORT_FDB(obj),
+						    trans);
 		break;
 	default:
 		err = -EOPNOTSUPP;
@@ -4276,6 +4659,10 @@ static int rocker_port_obj_del(struct net_device *dev,
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
 		err = rocker_port_vlans_del(rocker_port,
 					    SWITCHDEV_OBJ_PORT_VLAN(obj));
+		if (err)
+			break;
+		err = rocker_world_port_obj_vlan_del(rocker_port,
+						     SWITCHDEV_OBJ_PORT_VLAN(obj));
 		break;
 	case SWITCHDEV_OBJ_ID_IPV4_FIB:
 		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
@@ -4283,10 +4670,18 @@ static int rocker_port_obj_del(struct net_device *dev,
 					   htonl(fib4->dst), fib4->dst_len,
 					   &fib4->fi, fib4->tb_id,
 					   ROCKER_OP_FLAG_REMOVE);
+		if (err)
+			break;
+		err = rocker_world_port_obj_fib4_del(rocker_port,
+						     SWITCHDEV_OBJ_IPV4_FIB(obj));
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
 		err = rocker_port_fdb_del(rocker_port, NULL,
 					  SWITCHDEV_OBJ_PORT_FDB(obj));
+		if (err)
+			break;
+		err = rocker_world_port_obj_fdb_del(rocker_port,
+						    SWITCHDEV_OBJ_PORT_FDB(obj));
 		break;
 	default:
 		err = -EOPNOTSUPP;
@@ -4358,10 +4753,20 @@ static int rocker_port_obj_dump(struct net_device *dev,
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
 		err = rocker_port_fdb_dump(rocker_port,
 					   SWITCHDEV_OBJ_PORT_FDB(obj), cb);
+		if (err)
+			break;
+		err = rocker_world_port_obj_fdb_dump(rocker_port,
+						     SWITCHDEV_OBJ_PORT_FDB(obj),
+						     cb);
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
 		err = rocker_port_vlan_dump(rocker_port,
 					    SWITCHDEV_OBJ_PORT_VLAN(obj), cb);
+		if (err)
+			break;
+		err = rocker_world_port_obj_vlan_dump(rocker_port,
+						      SWITCHDEV_OBJ_PORT_VLAN(obj),
+						      cb);
 		break;
 	default:
 		err = -EOPNOTSUPP;
@@ -4687,7 +5092,7 @@ static void rocker_carrier_init(const struct rocker_port *rocker_port)
 		netif_carrier_off(rocker_port->dev);
 }
 
-static void rocker_remove_ports(const struct rocker *rocker)
+static void rocker_remove_ports(struct rocker *rocker)
 {
 	struct rocker_port *rocker_port;
 	int i;
@@ -4697,9 +5102,12 @@ static void rocker_remove_ports(const struct rocker *rocker)
 		if (!rocker_port)
 			continue;
 		rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
+		rocker_world_port_fini(rocker_port);
 		unregister_netdev(rocker_port->dev);
+		rocker_world_port_post_fini(rocker_port);
 		free_netdev(rocker_port->dev);
 	}
+	rocker_world_fini(rocker);
 	kfree(rocker->ports);
 }
 
@@ -4736,6 +5144,12 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 	rocker_port->brport_flags = BR_LEARNING | BR_LEARNING_SYNC;
 	rocker_port->ageing_time = BR_DEFAULT_AGEING_TIME;
 
+	err = rocker_world_check_init(rocker_port);
+	if (err) {
+		dev_err(&pdev->dev, "world init failed\n");
+		goto err_world_check_init;
+	}
+
 	rocker_port_dev_addr_init(rocker_port);
 	dev->netdev_ops = &rocker_port_netdev_ops;
 	dev->ethtool_ops = &rocker_port_ethtool_ops;
@@ -4748,6 +5162,11 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 
 	dev->features |= NETIF_F_NETNS_LOCAL | NETIF_F_SG;
 
+	err = rocker_world_port_pre_init(rocker_port);
+	if (err) {
+		dev_err(&pdev->dev, "port world pre-init failed\n");
+		goto err_world_port_pre_init;
+	}
 	err = register_netdev(dev);
 	if (err) {
 		dev_err(&pdev->dev, "register_netdev failed\n");
@@ -4755,6 +5174,12 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 	}
 	rocker->ports[port_number] = rocker_port;
 
+	err = rocker_world_port_init(rocker_port);
+	if (err) {
+		dev_err(&pdev->dev, "port world init failed\n");
+		goto err_world_port_init;
+	}
+
 	switchdev_port_fwd_mark_set(rocker_port->dev, NULL, false);
 
 	rocker_port_set_learning(rocker_port, NULL);
@@ -4779,9 +5204,14 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 err_untagged_vlan:
 	rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
 err_port_ig_tbl:
+	rocker_world_port_fini(rocker_port);
+err_world_port_init:
 	rocker->ports[port_number] = NULL;
 	unregister_netdev(dev);
 err_register_netdev:
+	rocker_world_port_post_fini(rocker_port);
+err_world_port_pre_init:
+err_world_check_init:
 	free_netdev(dev);
 	return err;
 }
@@ -5132,12 +5562,22 @@ static int rocker_netdevice_event(struct notifier_block *unused,
 			goto out;
 		rocker_port = netdev_priv(dev);
 		if (info->linking) {
+			err = rocker_world_port_master_linked(rocker_port,
+							      info->upper_dev);
+			if (err)
+				netdev_warn(dev, "failed to reflect master linked (err %d)\n",
+					    err);
 			err = rocker_port_master_linked(rocker_port,
 							info->upper_dev);
 			if (err)
 				netdev_warn(dev, "failed to reflect master linked (err %d)\n",
 					    err);
 		} else {
+			err = rocker_world_port_master_unlinked(rocker_port,
+								info->upper_dev);
+			if (err)
+				netdev_warn(dev, "failed to reflect master unlinked (err %d)\n",
+					    err);
 			err = rocker_port_master_unlinked(rocker_port);
 			if (err)
 				netdev_warn(dev, "failed to reflect master unlinked (err %d)\n",
@@ -5170,6 +5610,7 @@ static int rocker_neigh_update(struct net_device *dev, struct neighbour *n)
 static int rocker_netevent_event(struct notifier_block *unused,
 				 unsigned long event, void *ptr)
 {
+	struct rocker_port *rocker_port;
 	struct net_device *dev;
 	struct neighbour *n = ptr;
 	int err;
@@ -5181,6 +5622,11 @@ static int rocker_netevent_event(struct notifier_block *unused,
 		dev = n->dev;
 		if (!rocker_port_dev_check(dev))
 			return NOTIFY_DONE;
+		rocker_port = netdev_priv(dev);
+		err = rocker_world_port_neigh_update(rocker_port, n);
+		if (err)
+			netdev_warn(dev, "failed to handle neigh update (err %d)\n",
+				    err);
 		err = rocker_neigh_update(dev, n);
 		if (err)
 			netdev_warn(dev,
diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
new file mode 100644
index 0000000..155dc53
--- /dev/null
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -0,0 +1,27 @@
+/*
+ * drivers/net/ethernet/rocker/rocker_ofdpa.c - Rocker switch OF-DPA-like
+ *					        implementation
+ * Copyright (c) 2014 Scott Feldman <sfeldma@gmail.com>
+ * Copyright (c) 2014-2016 Jiri Pirko <jiri@mellanox.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include "rocker.h"
+
+struct ofdpa {
+};
+
+struct ofdpa_port {
+};
+
+struct rocker_world_ops rocker_ofdpa_ops = {
+	.kind = "ofdpa",
+	.priv_size = sizeof(struct ofdpa),
+	.port_priv_size = sizeof(struct ofdpa_port),
+	.mode = ROCKER_PORT_MODE_OF_DPA,
+};
-- 
cgit v0.10.2


From c1fe922e15ee73b0dd9fe6a9b08bdc204a7d7428 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:46 +0100
Subject: rocker: pass "learning" value as a parameter to
 rocker_port_set_learning

Be consistent with the rest of the setting functions, and pass
"learning" as a bool function parameter.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 871ccbe..0015dcb 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -1583,6 +1583,7 @@ rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port,
 				  struct rocker_desc_info *desc_info,
 				  void *priv)
 {
+	bool learning = *(bool *)priv;
 	struct rocker_tlv *cmd_info;
 
 	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE,
@@ -1595,7 +1596,7 @@ rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port,
 			       rocker_port->pport))
 		return -EMSGSIZE;
 	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,
-			      !!(rocker_port->brport_flags & BR_LEARNING)))
+			      learning))
 		return -EMSGSIZE;
 	rocker_tlv_nest_end(desc_info, cmd_info);
 	return 0;
@@ -1652,11 +1653,12 @@ static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
 }
 
 static int rocker_port_set_learning(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans)
+				    struct switchdev_trans *trans,
+				    bool learning)
 {
 	return rocker_cmd_exec(rocker_port, trans, 0,
 			       rocker_cmd_set_port_learning_prep,
-			       NULL, NULL, NULL);
+			       &learning, NULL, NULL);
 }
 
 /**********************
@@ -4447,7 +4449,8 @@ static int rocker_port_brport_flags_set(struct rocker_port *rocker_port,
 	orig_flags = rocker_port->brport_flags;
 	rocker_port->brport_flags = brport_flags;
 	if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING)
-		err = rocker_port_set_learning(rocker_port, trans);
+		err = rocker_port_set_learning(rocker_port, trans,
+					       !!(rocker_port->brport_flags & BR_LEARNING));
 
 	if (switchdev_trans_ph_prepare(trans))
 		rocker_port->brport_flags = orig_flags;
@@ -5182,7 +5185,8 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 
 	switchdev_port_fwd_mark_set(rocker_port->dev, NULL, false);
 
-	rocker_port_set_learning(rocker_port, NULL);
+	rocker_port_set_learning(rocker_port, NULL,
+				 !!(rocker_port->brport_flags & BR_LEARNING));
 
 	err = rocker_port_ig_tbl(rocker_port, NULL, 0);
 	if (err) {
-- 
cgit v0.10.2


From ca0a5f2a39cbc3bcbbcc651e82758fa135ac1b9b Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:47 +0100
Subject: rocker: pre-allocate wait structures during cmd ring init

This avoids need to alloc/free wait structure for every command call.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 0015dcb..8585d98 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -338,23 +338,19 @@ static void rocker_wait_init(struct rocker_wait *wait)
 	rocker_wait_reset(wait);
 }
 
-static struct rocker_wait *rocker_wait_create(struct rocker_port *rocker_port,
-					      struct switchdev_trans *trans,
-					      int flags)
+static struct rocker_wait *rocker_wait_create(void)
 {
 	struct rocker_wait *wait;
 
-	wait = rocker_kzalloc(trans, flags, sizeof(*wait));
+	wait = kzalloc(sizeof(*wait), GFP_KERNEL);
 	if (!wait)
 		return NULL;
-	rocker_wait_init(wait);
 	return wait;
 }
 
-static void rocker_wait_destroy(struct switchdev_trans *trans,
-				struct rocker_wait *wait)
+static void rocker_wait_destroy(struct rocker_wait *wait)
 {
-	rocker_kfree(trans, wait);
+	kfree(wait);
 }
 
 static bool rocker_wait_event_timeout(struct rocker_wait *wait,
@@ -831,6 +827,53 @@ static void rocker_dma_ring_bufs_free(const struct rocker *rocker,
 	}
 }
 
+static int rocker_dma_cmd_ring_wait_alloc(struct rocker_desc_info *desc_info)
+{
+	struct rocker_wait *wait;
+
+	wait = rocker_wait_create();
+	if (!wait)
+		return -ENOMEM;
+	rocker_desc_cookie_ptr_set(desc_info, wait);
+	return 0;
+}
+
+static void
+rocker_dma_cmd_ring_wait_free(const struct rocker_desc_info *desc_info)
+{
+	struct rocker_wait *wait = rocker_desc_cookie_ptr_get(desc_info);
+
+	rocker_wait_destroy(wait);
+}
+
+static int rocker_dma_cmd_ring_waits_alloc(const struct rocker *rocker)
+{
+	const struct rocker_dma_ring_info *cmd_ring = &rocker->cmd_ring;
+	int i;
+	int err;
+
+	for (i = 0; i < cmd_ring->size; i++) {
+		err = rocker_dma_cmd_ring_wait_alloc(&cmd_ring->desc_info[i]);
+		if (err)
+			goto rollback;
+	}
+	return 0;
+
+rollback:
+	for (i--; i >= 0; i--)
+		rocker_dma_cmd_ring_wait_free(&cmd_ring->desc_info[i]);
+	return err;
+}
+
+static void rocker_dma_cmd_ring_waits_free(const struct rocker *rocker)
+{
+	const struct rocker_dma_ring_info *cmd_ring = &rocker->cmd_ring;
+	int i;
+
+	for (i = 0; i < cmd_ring->size; i++)
+		rocker_dma_cmd_ring_wait_free(&cmd_ring->desc_info[i]);
+}
+
 static int rocker_dma_rings_init(struct rocker *rocker)
 {
 	const struct pci_dev *pdev = rocker->pdev;
@@ -853,6 +896,12 @@ static int rocker_dma_rings_init(struct rocker *rocker)
 		goto err_dma_cmd_ring_bufs_alloc;
 	}
 
+	err = rocker_dma_cmd_ring_waits_alloc(rocker);
+	if (err) {
+		dev_err(&pdev->dev, "failed to alloc command dma ring waits\n");
+		goto err_dma_cmd_ring_waits_alloc;
+	}
+
 	err = rocker_dma_ring_create(rocker, ROCKER_DMA_EVENT,
 				     ROCKER_DMA_EVENT_DEFAULT_SIZE,
 				     &rocker->event_ring);
@@ -875,6 +924,8 @@ err_dma_event_ring_bufs_alloc:
 err_dma_event_ring_create:
 	rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring,
 				  PCI_DMA_BIDIRECTIONAL);
+err_dma_cmd_ring_waits_alloc:
+	rocker_dma_cmd_ring_waits_free(rocker);
 err_dma_cmd_ring_bufs_alloc:
 	rocker_dma_ring_destroy(rocker, &rocker->cmd_ring);
 	return err;
@@ -885,6 +936,7 @@ static void rocker_dma_rings_fini(struct rocker *rocker)
 	rocker_dma_ring_bufs_free(rocker, &rocker->event_ring,
 				  PCI_DMA_BIDIRECTIONAL);
 	rocker_dma_ring_destroy(rocker, &rocker->event_ring);
+	rocker_dma_cmd_ring_waits_free(rocker);
 	rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring,
 				  PCI_DMA_BIDIRECTIONAL);
 	rocker_dma_ring_destroy(rocker, &rocker->cmd_ring);
@@ -1106,7 +1158,6 @@ static irqreturn_t rocker_cmd_irq_handler(int irq, void *dev_id)
 		wait = rocker_desc_cookie_ptr_get(desc_info);
 		if (wait->nowait) {
 			rocker_desc_gen_clear(desc_info);
-			rocker_wait_destroy(NULL, wait);
 		} else {
 			rocker_wait_wake_up(wait);
 		}
@@ -1298,28 +1349,24 @@ static int rocker_cmd_exec(struct rocker_port *rocker_port,
 	unsigned long lock_flags;
 	int err;
 
-	wait = rocker_wait_create(rocker_port, trans, flags);
-	if (!wait)
-		return -ENOMEM;
-	wait->nowait = nowait;
-
 	spin_lock_irqsave(&rocker->cmd_ring_lock, lock_flags);
 
 	desc_info = rocker_desc_head_get(&rocker->cmd_ring);
 	if (!desc_info) {
 		spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
-		err = -EAGAIN;
-		goto out;
+		return -EAGAIN;
 	}
 
+	wait = rocker_desc_cookie_ptr_get(desc_info);
+	rocker_wait_init(wait);
+	wait->nowait = nowait;
+
 	err = prepare(rocker_port, desc_info, prepare_priv);
 	if (err) {
 		spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
-		goto out;
+		return err;
 	}
 
-	rocker_desc_cookie_ptr_set(desc_info, wait);
-
 	if (!switchdev_trans_ph_prepare(trans))
 		rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info);
 
@@ -1340,8 +1387,6 @@ static int rocker_cmd_exec(struct rocker_port *rocker_port,
 		err = process(rocker_port, desc_info, process_priv);
 
 	rocker_desc_gen_clear(desc_info);
-out:
-	rocker_wait_destroy(trans, wait);
 	return err;
 }
 
-- 
cgit v0.10.2


From ae3907ecedb91c10ab8fd8b42ef62431206eee83 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:48 +0100
Subject: rocker: remove trans parameter to rocker_cmd_exec function

The only purpose of passing this parameter is to check for
prepare phase. The only reason for a failure in that state is if
TLVs don't fit into descriptor. That is highly unlikely and if that
happens, it is a driver bug. So remove this parameter from
rocker_cmd_exec, and check for prepare phase in caller.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 8585d98..8113b15 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -1337,8 +1337,7 @@ typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port,
 				    const struct rocker_desc_info *desc_info,
 				    void *priv);
 
-static int rocker_cmd_exec(struct rocker_port *rocker_port,
-			   struct switchdev_trans *trans, int flags,
+static int rocker_cmd_exec(struct rocker_port *rocker_port, int flags,
 			   rocker_cmd_prep_cb_t prepare, void *prepare_priv,
 			   rocker_cmd_proc_cb_t process, void *process_priv)
 {
@@ -1367,17 +1366,15 @@ static int rocker_cmd_exec(struct rocker_port *rocker_port,
 		return err;
 	}
 
-	if (!switchdev_trans_ph_prepare(trans))
-		rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info);
+	rocker_desc_head_set(rocker, &rocker->cmd_ring, desc_info);
 
 	spin_unlock_irqrestore(&rocker->cmd_ring_lock, lock_flags);
 
 	if (nowait)
 		return 0;
 
-	if (!switchdev_trans_ph_prepare(trans))
-		if (!rocker_wait_event_timeout(wait, HZ / 10))
-			return -EIO;
+	if (!rocker_wait_event_timeout(wait, HZ / 10))
+		return -EIO;
 
 	err = rocker_desc_err(desc_info);
 	if (err)
@@ -1650,7 +1647,7 @@ rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port,
 						struct ethtool_cmd *ecmd)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_get_port_settings_prep, NULL,
 			       rocker_cmd_get_port_settings_ethtool_proc,
 			       ecmd);
@@ -1659,7 +1656,7 @@ static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
 						unsigned char *macaddr)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_get_port_settings_prep, NULL,
 			       rocker_cmd_get_port_settings_macaddr_proc,
 			       macaddr);
@@ -1668,7 +1665,7 @@ static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_settings_mode(struct rocker_port *rocker_port,
 					     u8 *p_mode)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_get_port_settings_prep, NULL,
 			       rocker_cmd_get_port_settings_mode_proc, p_mode);
 }
@@ -1676,7 +1673,7 @@ static int rocker_cmd_get_port_settings_mode(struct rocker_port *rocker_port,
 static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
 						struct ethtool_cmd *ecmd)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_set_port_settings_ethtool_prep,
 			       ecmd, NULL, NULL);
 }
@@ -1684,7 +1681,7 @@ static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
 static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port,
 						unsigned char *macaddr)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_set_port_settings_macaddr_prep,
 			       macaddr, NULL, NULL);
 }
@@ -1692,16 +1689,15 @@ static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port,
 static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
 					    int mtu)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_set_port_settings_mtu_prep,
 			       &mtu, NULL, NULL);
 }
 
 static int rocker_port_set_learning(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans,
 				    bool learning)
 {
-	return rocker_cmd_exec(rocker_port, trans, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_set_port_learning_prep,
 			       &learning, NULL, NULL);
 }
@@ -2578,8 +2574,11 @@ static int rocker_flow_tbl_add(struct rocker_port *rocker_port,
 
 	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
 
-	return rocker_cmd_exec(rocker_port, trans, flags,
-			       rocker_cmd_flow_tbl_add, found, NULL, NULL);
+	if (!switchdev_trans_ph_prepare(trans))
+		return rocker_cmd_exec(rocker_port, flags,
+				       rocker_cmd_flow_tbl_add,
+				       found, NULL, NULL);
+	return 0;
 }
 
 static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
@@ -2609,9 +2608,10 @@ static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
 	rocker_kfree(trans, match);
 
 	if (found) {
-		err = rocker_cmd_exec(rocker_port, trans, flags,
-				      rocker_cmd_flow_tbl_del,
-				      found, NULL, NULL);
+		if (!switchdev_trans_ph_prepare(trans))
+			err = rocker_cmd_exec(rocker_port, flags,
+					      rocker_cmd_flow_tbl_del,
+					      found, NULL, NULL);
 		rocker_kfree(trans, found);
 	}
 
@@ -2899,8 +2899,11 @@ static int rocker_group_tbl_add(struct rocker_port *rocker_port,
 
 	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
 
-	return rocker_cmd_exec(rocker_port, trans, flags,
-			       rocker_cmd_group_tbl_add, found, NULL, NULL);
+	if (!switchdev_trans_ph_prepare(trans))
+		return rocker_cmd_exec(rocker_port, flags,
+				       rocker_cmd_group_tbl_add,
+				       found, NULL, NULL);
+	return 0;
 }
 
 static int rocker_group_tbl_del(struct rocker_port *rocker_port,
@@ -2927,9 +2930,10 @@ static int rocker_group_tbl_del(struct rocker_port *rocker_port,
 	rocker_group_tbl_entry_free(trans, match);
 
 	if (found) {
-		err = rocker_cmd_exec(rocker_port, trans, flags,
-				      rocker_cmd_group_tbl_del,
-				      found, NULL, NULL);
+		if (!switchdev_trans_ph_prepare(trans))
+			err = rocker_cmd_exec(rocker_port, flags,
+					      rocker_cmd_group_tbl_del,
+					      found, NULL, NULL);
 		rocker_group_tbl_entry_free(trans, found);
 	}
 
@@ -4405,7 +4409,7 @@ static int rocker_port_get_phys_port_name(struct net_device *dev,
 	struct port_name name = { .buf = buf, .len = len };
 	int err;
 
-	err = rocker_cmd_exec(rocker_port, NULL, 0,
+	err = rocker_cmd_exec(rocker_port, 0,
 			      rocker_cmd_get_port_settings_prep, NULL,
 			      rocker_cmd_get_port_settings_phys_name_proc,
 			      &name);
@@ -4493,8 +4497,9 @@ static int rocker_port_brport_flags_set(struct rocker_port *rocker_port,
 
 	orig_flags = rocker_port->brport_flags;
 	rocker_port->brport_flags = brport_flags;
-	if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING)
-		err = rocker_port_set_learning(rocker_port, trans,
+	if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING &&
+	    !switchdev_trans_ph_prepare(trans))
+		err = rocker_port_set_learning(rocker_port,
 					       !!(rocker_port->brport_flags & BR_LEARNING));
 
 	if (switchdev_trans_ph_prepare(trans))
@@ -4957,7 +4962,7 @@ rocker_cmd_get_port_stats_ethtool_proc(const struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_stats_ethtool(struct rocker_port *rocker_port,
 					     void *priv)
 {
-	return rocker_cmd_exec(rocker_port, NULL, 0,
+	return rocker_cmd_exec(rocker_port, 0,
 			       rocker_cmd_get_port_stats_prep, NULL,
 			       rocker_cmd_get_port_stats_ethtool_proc,
 			       priv);
@@ -5230,7 +5235,7 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 
 	switchdev_port_fwd_mark_set(rocker_port->dev, NULL, false);
 
-	rocker_port_set_learning(rocker_port, NULL,
+	rocker_port_set_learning(rocker_port,
 				 !!(rocker_port->brport_flags & BR_LEARNING));
 
 	err = rocker_port_ig_tbl(rocker_port, NULL, 0);
-- 
cgit v0.10.2


From 53901cc03a361f57127ed1ae967c72ee9e137454 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:49 +0100
Subject: rocker: call rocker_cmd_exec function with "nowait" boolean instead
 of flags

No need to push down rocker flags just to check if this is nowait or
not. Let the caller handle that.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 8113b15..0260249 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -262,6 +262,11 @@ static bool rocker_port_is_ovsed(const struct rocker_port *rocker_port)
 #define ROCKER_OP_FLAG_LEARNED		BIT(2)
 #define ROCKER_OP_FLAG_REFRESH		BIT(3)
 
+static bool rocker_flags_nowait(int flags)
+{
+	return flags & ROCKER_OP_FLAG_NOWAIT;
+}
+
 static void *__rocker_mem_alloc(struct switchdev_trans *trans, int flags,
 				size_t size)
 {
@@ -1337,14 +1342,13 @@ typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port,
 				    const struct rocker_desc_info *desc_info,
 				    void *priv);
 
-static int rocker_cmd_exec(struct rocker_port *rocker_port, int flags,
+static int rocker_cmd_exec(struct rocker_port *rocker_port, bool nowait,
 			   rocker_cmd_prep_cb_t prepare, void *prepare_priv,
 			   rocker_cmd_proc_cb_t process, void *process_priv)
 {
 	struct rocker *rocker = rocker_port->rocker;
 	struct rocker_desc_info *desc_info;
 	struct rocker_wait *wait;
-	bool nowait = !!(flags & ROCKER_OP_FLAG_NOWAIT);
 	unsigned long lock_flags;
 	int err;
 
@@ -1647,7 +1651,7 @@ rocker_cmd_set_port_learning_prep(const struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port,
 						struct ethtool_cmd *ecmd)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_get_port_settings_prep, NULL,
 			       rocker_cmd_get_port_settings_ethtool_proc,
 			       ecmd);
@@ -1656,7 +1660,7 @@ static int rocker_cmd_get_port_settings_ethtool(struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
 						unsigned char *macaddr)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_get_port_settings_prep, NULL,
 			       rocker_cmd_get_port_settings_macaddr_proc,
 			       macaddr);
@@ -1665,7 +1669,7 @@ static int rocker_cmd_get_port_settings_macaddr(struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_settings_mode(struct rocker_port *rocker_port,
 					     u8 *p_mode)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_get_port_settings_prep, NULL,
 			       rocker_cmd_get_port_settings_mode_proc, p_mode);
 }
@@ -1673,7 +1677,7 @@ static int rocker_cmd_get_port_settings_mode(struct rocker_port *rocker_port,
 static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
 						struct ethtool_cmd *ecmd)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_set_port_settings_ethtool_prep,
 			       ecmd, NULL, NULL);
 }
@@ -1681,7 +1685,7 @@ static int rocker_cmd_set_port_settings_ethtool(struct rocker_port *rocker_port,
 static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port,
 						unsigned char *macaddr)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_set_port_settings_macaddr_prep,
 			       macaddr, NULL, NULL);
 }
@@ -1689,7 +1693,7 @@ static int rocker_cmd_set_port_settings_macaddr(struct rocker_port *rocker_port,
 static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
 					    int mtu)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_set_port_settings_mtu_prep,
 			       &mtu, NULL, NULL);
 }
@@ -1697,7 +1701,7 @@ static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
 static int rocker_port_set_learning(struct rocker_port *rocker_port,
 				    bool learning)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_set_port_learning_prep,
 			       &learning, NULL, NULL);
 }
@@ -2575,7 +2579,8 @@ static int rocker_flow_tbl_add(struct rocker_port *rocker_port,
 	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
 
 	if (!switchdev_trans_ph_prepare(trans))
-		return rocker_cmd_exec(rocker_port, flags,
+		return rocker_cmd_exec(rocker_port,
+				       rocker_flags_nowait(flags),
 				       rocker_cmd_flow_tbl_add,
 				       found, NULL, NULL);
 	return 0;
@@ -2609,7 +2614,8 @@ static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
 
 	if (found) {
 		if (!switchdev_trans_ph_prepare(trans))
-			err = rocker_cmd_exec(rocker_port, flags,
+			err = rocker_cmd_exec(rocker_port,
+					      rocker_flags_nowait(flags),
 					      rocker_cmd_flow_tbl_del,
 					      found, NULL, NULL);
 		rocker_kfree(trans, found);
@@ -2900,7 +2906,8 @@ static int rocker_group_tbl_add(struct rocker_port *rocker_port,
 	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
 
 	if (!switchdev_trans_ph_prepare(trans))
-		return rocker_cmd_exec(rocker_port, flags,
+		return rocker_cmd_exec(rocker_port,
+				       rocker_flags_nowait(flags),
 				       rocker_cmd_group_tbl_add,
 				       found, NULL, NULL);
 	return 0;
@@ -2931,7 +2938,8 @@ static int rocker_group_tbl_del(struct rocker_port *rocker_port,
 
 	if (found) {
 		if (!switchdev_trans_ph_prepare(trans))
-			err = rocker_cmd_exec(rocker_port, flags,
+			err = rocker_cmd_exec(rocker_port,
+					      rocker_flags_nowait(flags),
 					      rocker_cmd_group_tbl_del,
 					      found, NULL, NULL);
 		rocker_group_tbl_entry_free(trans, found);
@@ -4409,7 +4417,7 @@ static int rocker_port_get_phys_port_name(struct net_device *dev,
 	struct port_name name = { .buf = buf, .len = len };
 	int err;
 
-	err = rocker_cmd_exec(rocker_port, 0,
+	err = rocker_cmd_exec(rocker_port, false,
 			      rocker_cmd_get_port_settings_prep, NULL,
 			      rocker_cmd_get_port_settings_phys_name_proc,
 			      &name);
@@ -4962,7 +4970,7 @@ rocker_cmd_get_port_stats_ethtool_proc(const struct rocker_port *rocker_port,
 static int rocker_cmd_get_port_stats_ethtool(struct rocker_port *rocker_port,
 					     void *priv)
 {
-	return rocker_cmd_exec(rocker_port, 0,
+	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_get_port_stats_prep, NULL,
 			       rocker_cmd_get_port_stats_ethtool_proc,
 			       priv);
-- 
cgit v0.10.2


From 3fbcdbf3f1cbe86fc11856ec268470d7d7bd459b Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Tue, 16 Feb 2016 15:14:50 +0100
Subject: rocker: move OF-DPA stuff into separate file

Carve out OF-DPA would specific code from the common file to the world
file. This change required struct rocker and struct rocker_port split
into world specific struct ofdpa and struct ofdpa_port. Along with this
the world specific functions and defines were renamed from prefix
"rocker_" to "ofdpa_".

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
index 5fd0e35..1ab995f 100644
--- a/drivers/net/ethernet/rocker/rocker.h
+++ b/drivers/net/ethernet/rocker/rocker.h
@@ -14,8 +14,6 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <linux/hashtable.h>
-#include <linux/if_vlan.h>
 #include <linux/netdevice.h>
 #include <net/neighbour.h>
 #include <net/switchdev.h>
@@ -42,34 +40,12 @@ struct rocker_dma_ring_info {
 
 struct rocker;
 
-enum {
-	ROCKER_CTRL_LINK_LOCAL_MCAST,
-	ROCKER_CTRL_LOCAL_ARP,
-	ROCKER_CTRL_IPV4_MCAST,
-	ROCKER_CTRL_IPV6_MCAST,
-	ROCKER_CTRL_DFLT_BRIDGING,
-	ROCKER_CTRL_DFLT_OVS,
-	ROCKER_CTRL_MAX,
-};
-
-#define ROCKER_INTERNAL_VLAN_ID_BASE	0x0f00
-#define ROCKER_N_INTERNAL_VLANS		255
-#define ROCKER_VLAN_BITMAP_LEN		BITS_TO_LONGS(VLAN_N_VID)
-#define ROCKER_INTERNAL_VLAN_BITMAP_LEN	BITS_TO_LONGS(ROCKER_N_INTERNAL_VLANS)
-
 struct rocker_port {
 	struct net_device *dev;
-	struct net_device *bridge_dev;
 	struct rocker *rocker;
 	void *wpriv;
 	unsigned int port_number;
 	u32 pport;
-	__be16 internal_vlan_id;
-	int stp_state;
-	u32 brport_flags;
-	unsigned long ageing_time;
-	bool ctrls[ROCKER_CTRL_MAX];
-	unsigned long vlan_bitmap[ROCKER_VLAN_BITMAP_LEN];
 	struct napi_struct napi_tx;
 	struct napi_struct napi_rx;
 	struct rocker_dma_ring_info tx_ring;
@@ -92,22 +68,23 @@ struct rocker {
 	struct rocker_dma_ring_info event_ring;
 	struct rocker_world_ops *wops;
 	void *wpriv;
-	DECLARE_HASHTABLE(flow_tbl, 16);
-	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
-	u64 flow_tbl_next_cookie;
-	DECLARE_HASHTABLE(group_tbl, 16);
-	spinlock_t group_tbl_lock;		/* for group tbl accesses */
-	struct timer_list fdb_cleanup_timer;
-	DECLARE_HASHTABLE(fdb_tbl, 16);
-	spinlock_t fdb_tbl_lock;		/* for fdb tbl accesses */
-	unsigned long internal_vlan_bitmap[ROCKER_INTERNAL_VLAN_BITMAP_LEN];
-	DECLARE_HASHTABLE(internal_vlan_tbl, 8);
-	spinlock_t internal_vlan_tbl_lock;	/* for vlan tbl accesses */
-	DECLARE_HASHTABLE(neigh_tbl, 16);
-	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
-	u32 neigh_tbl_next_index;
 };
 
+typedef int (*rocker_cmd_prep_cb_t)(const struct rocker_port *rocker_port,
+				    struct rocker_desc_info *desc_info,
+				    void *priv);
+
+typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port,
+				    const struct rocker_desc_info *desc_info,
+				    void *priv);
+
+int rocker_cmd_exec(struct rocker_port *rocker_port, bool nowait,
+		    rocker_cmd_prep_cb_t prepare, void *prepare_priv,
+		    rocker_cmd_proc_cb_t process, void *process_priv);
+
+int rocker_port_set_learning(struct rocker_port *rocker_port,
+			     bool learning);
+
 struct rocker_world_ops {
 	const char *kind;
 	size_t priv_size;
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 0260249..1efd8b7 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -16,12 +16,9 @@
 #include <linux/sched.h>
 #include <linux/wait.h>
 #include <linux/spinlock.h>
-#include <linux/hashtable.h>
-#include <linux/crc32.h>
 #include <linux/sort.h>
 #include <linux/random.h>
 #include <linux/netdevice.h>
-#include <linux/inetdevice.h>
 #include <linux/skbuff.h>
 #include <linux/socket.h>
 #include <linux/etherdevice.h>
@@ -33,7 +30,6 @@
 #include <linux/ctype.h>
 #include <net/switchdev.h>
 #include <net/rtnetlink.h>
-#include <net/ip_fib.h>
 #include <net/netevent.h>
 #include <net/arp.h>
 #include <linux/io-64-nonatomic-lo-hi.h>
@@ -50,281 +46,6 @@ static const struct pci_device_id rocker_pci_id_table[] = {
 	{0, }
 };
 
-struct rocker_flow_tbl_key {
-	u32 priority;
-	enum rocker_of_dpa_table_id tbl_id;
-	union {
-		struct {
-			u32 in_pport;
-			u32 in_pport_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-		} ig_port;
-		struct {
-			u32 in_pport;
-			__be16 vlan_id;
-			__be16 vlan_id_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-			bool untagged;
-			__be16 new_vlan_id;
-		} vlan;
-		struct {
-			u32 in_pport;
-			u32 in_pport_mask;
-			__be16 eth_type;
-			u8 eth_dst[ETH_ALEN];
-			u8 eth_dst_mask[ETH_ALEN];
-			__be16 vlan_id;
-			__be16 vlan_id_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-			bool copy_to_cpu;
-		} term_mac;
-		struct {
-			__be16 eth_type;
-			__be32 dst4;
-			__be32 dst4_mask;
-			enum rocker_of_dpa_table_id goto_tbl;
-			u32 group_id;
-		} ucast_routing;
-		struct {
-			u8 eth_dst[ETH_ALEN];
-			u8 eth_dst_mask[ETH_ALEN];
-			int has_eth_dst;
-			int has_eth_dst_mask;
-			__be16 vlan_id;
-			u32 tunnel_id;
-			enum rocker_of_dpa_table_id goto_tbl;
-			u32 group_id;
-			bool copy_to_cpu;
-		} bridge;
-		struct {
-			u32 in_pport;
-			u32 in_pport_mask;
-			u8 eth_src[ETH_ALEN];
-			u8 eth_src_mask[ETH_ALEN];
-			u8 eth_dst[ETH_ALEN];
-			u8 eth_dst_mask[ETH_ALEN];
-			__be16 eth_type;
-			__be16 vlan_id;
-			__be16 vlan_id_mask;
-			u8 ip_proto;
-			u8 ip_proto_mask;
-			u8 ip_tos;
-			u8 ip_tos_mask;
-			u32 group_id;
-		} acl;
-	};
-};
-
-struct rocker_flow_tbl_entry {
-	struct hlist_node entry;
-	u32 cmd;
-	u64 cookie;
-	struct rocker_flow_tbl_key key;
-	size_t key_len;
-	u32 key_crc32; /* key */
-};
-
-struct rocker_group_tbl_entry {
-	struct hlist_node entry;
-	u32 cmd;
-	u32 group_id; /* key */
-	u16 group_count;
-	u32 *group_ids;
-	union {
-		struct {
-			u8 pop_vlan;
-		} l2_interface;
-		struct {
-			u8 eth_src[ETH_ALEN];
-			u8 eth_dst[ETH_ALEN];
-			__be16 vlan_id;
-			u32 group_id;
-		} l2_rewrite;
-		struct {
-			u8 eth_src[ETH_ALEN];
-			u8 eth_dst[ETH_ALEN];
-			__be16 vlan_id;
-			bool ttl_check;
-			u32 group_id;
-		} l3_unicast;
-	};
-};
-
-struct rocker_fdb_tbl_entry {
-	struct hlist_node entry;
-	u32 key_crc32; /* key */
-	bool learned;
-	unsigned long touched;
-	struct rocker_fdb_tbl_key {
-		struct rocker_port *rocker_port;
-		u8 addr[ETH_ALEN];
-		__be16 vlan_id;
-	} key;
-};
-
-struct rocker_internal_vlan_tbl_entry {
-	struct hlist_node entry;
-	int ifindex; /* key */
-	u32 ref_count;
-	__be16 vlan_id;
-};
-
-struct rocker_neigh_tbl_entry {
-	struct hlist_node entry;
-	__be32 ip_addr; /* key */
-	struct net_device *dev;
-	u32 ref_count;
-	u32 index;
-	u8 eth_dst[ETH_ALEN];
-	bool ttl_check;
-};
-
-static const u8 zero_mac[ETH_ALEN]   = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
-static const u8 ff_mac[ETH_ALEN]     = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
-static const u8 ll_mac[ETH_ALEN]     = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
-static const u8 ll_mask[ETH_ALEN]    = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xf0 };
-static const u8 mcast_mac[ETH_ALEN]  = { 0x01, 0x00, 0x00, 0x00, 0x00, 0x00 };
-static const u8 ipv4_mcast[ETH_ALEN] = { 0x01, 0x00, 0x5e, 0x00, 0x00, 0x00 };
-static const u8 ipv4_mask[ETH_ALEN]  = { 0xff, 0xff, 0xff, 0x80, 0x00, 0x00 };
-static const u8 ipv6_mcast[ETH_ALEN] = { 0x33, 0x33, 0x00, 0x00, 0x00, 0x00 };
-static const u8 ipv6_mask[ETH_ALEN]  = { 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 };
-
-/* Rocker priority levels for flow table entries.  Higher
- * priority match takes precedence over lower priority match.
- */
-
-enum {
-	ROCKER_PRIORITY_UNKNOWN = 0,
-	ROCKER_PRIORITY_IG_PORT = 1,
-	ROCKER_PRIORITY_VLAN = 1,
-	ROCKER_PRIORITY_TERM_MAC_UCAST = 0,
-	ROCKER_PRIORITY_TERM_MAC_MCAST = 1,
-	ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_EXACT = 1,
-	ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_WILD = 2,
-	ROCKER_PRIORITY_BRIDGING_VLAN = 3,
-	ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_EXACT = 1,
-	ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_WILD = 2,
-	ROCKER_PRIORITY_BRIDGING_TENANT = 3,
-	ROCKER_PRIORITY_ACL_CTRL = 3,
-	ROCKER_PRIORITY_ACL_NORMAL = 2,
-	ROCKER_PRIORITY_ACL_DFLT = 1,
-};
-
-static bool rocker_vlan_id_is_internal(__be16 vlan_id)
-{
-	u16 start = ROCKER_INTERNAL_VLAN_ID_BASE;
-	u16 end = 0xffe;
-	u16 _vlan_id = ntohs(vlan_id);
-
-	return (_vlan_id >= start && _vlan_id <= end);
-}
-
-static __be16 rocker_port_vid_to_vlan(const struct rocker_port *rocker_port,
-				      u16 vid, bool *pop_vlan)
-{
-	__be16 vlan_id;
-
-	if (pop_vlan)
-		*pop_vlan = false;
-	vlan_id = htons(vid);
-	if (!vlan_id) {
-		vlan_id = rocker_port->internal_vlan_id;
-		if (pop_vlan)
-			*pop_vlan = true;
-	}
-
-	return vlan_id;
-}
-
-static u16 rocker_port_vlan_to_vid(const struct rocker_port *rocker_port,
-				   __be16 vlan_id)
-{
-	if (rocker_vlan_id_is_internal(vlan_id))
-		return 0;
-
-	return ntohs(vlan_id);
-}
-
-static bool rocker_port_is_bridged(const struct rocker_port *rocker_port)
-{
-	return rocker_port->bridge_dev &&
-	       netif_is_bridge_master(rocker_port->bridge_dev);
-}
-
-static bool rocker_port_is_ovsed(const struct rocker_port *rocker_port)
-{
-	return rocker_port->bridge_dev &&
-	       netif_is_ovs_master(rocker_port->bridge_dev);
-}
-
-#define ROCKER_OP_FLAG_REMOVE		BIT(0)
-#define ROCKER_OP_FLAG_NOWAIT		BIT(1)
-#define ROCKER_OP_FLAG_LEARNED		BIT(2)
-#define ROCKER_OP_FLAG_REFRESH		BIT(3)
-
-static bool rocker_flags_nowait(int flags)
-{
-	return flags & ROCKER_OP_FLAG_NOWAIT;
-}
-
-static void *__rocker_mem_alloc(struct switchdev_trans *trans, int flags,
-				size_t size)
-{
-	struct switchdev_trans_item *elem = NULL;
-	gfp_t gfp_flags = (flags & ROCKER_OP_FLAG_NOWAIT) ?
-			  GFP_ATOMIC : GFP_KERNEL;
-
-	/* If in transaction prepare phase, allocate the memory
-	 * and enqueue it on a transaction.  If in transaction
-	 * commit phase, dequeue the memory from the transaction
-	 * rather than re-allocating the memory.  The idea is the
-	 * driver code paths for prepare and commit are identical
-	 * so the memory allocated in the prepare phase is the
-	 * memory used in the commit phase.
-	 */
-
-	if (!trans) {
-		elem = kzalloc(size + sizeof(*elem), gfp_flags);
-	} else if (switchdev_trans_ph_prepare(trans)) {
-		elem = kzalloc(size + sizeof(*elem), gfp_flags);
-		if (!elem)
-			return NULL;
-		switchdev_trans_item_enqueue(trans, elem, kfree, elem);
-	} else {
-		elem = switchdev_trans_item_dequeue(trans);
-	}
-
-	return elem ? elem + 1 : NULL;
-}
-
-static void *rocker_kzalloc(struct switchdev_trans *trans, int flags,
-			    size_t size)
-{
-	return __rocker_mem_alloc(trans, flags, size);
-}
-
-static void *rocker_kcalloc(struct switchdev_trans *trans, int flags,
-			    size_t n, size_t size)
-{
-	return __rocker_mem_alloc(trans, flags, n * size);
-}
-
-static void rocker_kfree(struct switchdev_trans *trans, const void *mem)
-{
-	struct switchdev_trans_item *elem;
-
-	/* Frees are ignored if in transaction prepare phase.  The
-	 * memory remains on the per-port list until freed in the
-	 * commit phase.
-	 */
-
-	if (switchdev_trans_ph_prepare(trans))
-		return;
-
-	elem = (struct switchdev_trans_item *) mem - 1;
-	kfree(elem);
-}
-
 struct rocker_wait {
 	wait_queue_head_t wait;
 	bool done;
@@ -1216,10 +937,6 @@ static int rocker_event_link_change(const struct rocker *rocker,
 	return 0;
 }
 
-static int rocker_port_fdb(struct rocker_port *rocker_port,
-			   struct switchdev_trans *trans,
-			   const unsigned char *addr,
-			   __be16 vlan_id, int flags);
 static int rocker_world_port_ev_mac_vlan_seen(struct rocker_port *rocker_port,
 					      const unsigned char *addr,
 					      __be16 vlan_id);
@@ -1231,9 +948,7 @@ static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
 	unsigned int port_number;
 	struct rocker_port *rocker_port;
 	const unsigned char *addr;
-	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_LEARNED;
 	__be16 vlan_id;
-	int err;
 
 	rocker_tlv_parse_nested(attrs, ROCKER_TLV_EVENT_MAC_VLAN_MAX, info);
 	if (!attrs[ROCKER_TLV_EVENT_MAC_VLAN_PPORT] ||
@@ -1249,16 +964,7 @@ static int rocker_event_mac_vlan_seen(const struct rocker *rocker,
 		return -EINVAL;
 
 	rocker_port = rocker->ports[port_number];
-
-	err = rocker_world_port_ev_mac_vlan_seen(rocker_port, addr, vlan_id);
-	if (err)
-		return err;
-
-	if (rocker_port->stp_state != BR_STATE_LEARNING &&
-	    rocker_port->stp_state != BR_STATE_FORWARDING)
-		return 0;
-
-	return rocker_port_fdb(rocker_port, NULL, addr, vlan_id, flags);
+	return rocker_world_port_ev_mac_vlan_seen(rocker_port, addr, vlan_id);
 }
 
 static int rocker_event_process(const struct rocker *rocker,
@@ -1334,17 +1040,9 @@ static irqreturn_t rocker_rx_irq_handler(int irq, void *dev_id)
  * Command interface
  ********************/
 
-typedef int (*rocker_cmd_prep_cb_t)(const struct rocker_port *rocker_port,
-				    struct rocker_desc_info *desc_info,
-				    void *priv);
-
-typedef int (*rocker_cmd_proc_cb_t)(const struct rocker_port *rocker_port,
-				    const struct rocker_desc_info *desc_info,
-				    void *priv);
-
-static int rocker_cmd_exec(struct rocker_port *rocker_port, bool nowait,
-			   rocker_cmd_prep_cb_t prepare, void *prepare_priv,
-			   rocker_cmd_proc_cb_t process, void *process_priv)
+int rocker_cmd_exec(struct rocker_port *rocker_port, bool nowait,
+		    rocker_cmd_prep_cb_t prepare, void *prepare_priv,
+		    rocker_cmd_proc_cb_t process, void *process_priv)
 {
 	struct rocker *rocker = rocker_port->rocker;
 	struct rocker_desc_info *desc_info;
@@ -1698,8 +1396,8 @@ static int rocker_cmd_set_port_settings_mtu(struct rocker_port *rocker_port,
 			       &mtu, NULL, NULL);
 }
 
-static int rocker_port_set_learning(struct rocker_port *rocker_port,
-				    bool learning)
+int rocker_port_set_learning(struct rocker_port *rocker_port,
+			     bool learning)
 {
 	return rocker_cmd_exec(rocker_port, false,
 			       rocker_cmd_set_port_learning_prep,
@@ -2035,2283 +1733,145 @@ static int rocker_world_port_ev_mac_vlan_seen(struct rocker_port *rocker_port,
 	return wops->port_ev_mac_vlan_seen(rocker_port, addr, vlan_id);
 }
 
-static int
-rocker_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info,
-				const struct rocker_flow_tbl_entry *entry)
+/*****************
+ * Net device ops
+ *****************/
+
+static int rocker_port_open(struct net_device *dev)
 {
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.ig_port.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
-			       entry->key.ig_port.in_pport_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.ig_port.goto_tbl))
-		return -EMSGSIZE;
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	int err;
 
-	return 0;
-}
+	err = rocker_port_dma_rings_init(rocker_port);
+	if (err)
+		return err;
 
-static int
-rocker_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info,
-			     const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.vlan.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.vlan.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
-				entry->key.vlan.vlan_id_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.vlan.goto_tbl))
-		return -EMSGSIZE;
-	if (entry->key.vlan.untagged &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_NEW_VLAN_ID,
-				entry->key.vlan.new_vlan_id))
-		return -EMSGSIZE;
+	err = request_irq(rocker_msix_tx_vector(rocker_port),
+			  rocker_tx_irq_handler, 0,
+			  rocker_driver_name, rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "cannot assign tx irq\n");
+		goto err_request_tx_irq;
+	}
 
-	return 0;
-}
+	err = request_irq(rocker_msix_rx_vector(rocker_port),
+			  rocker_rx_irq_handler, 0,
+			  rocker_driver_name, rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "cannot assign rx irq\n");
+		goto err_request_rx_irq;
+	}
 
-static int
-rocker_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info,
-				 const struct rocker_flow_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.term_mac.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
-			       entry->key.term_mac.in_pport_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
-				entry->key.term_mac.eth_type))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->key.term_mac.eth_dst))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
-			   ETH_ALEN, entry->key.term_mac.eth_dst_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.term_mac.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
-				entry->key.term_mac.vlan_id_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.term_mac.goto_tbl))
-		return -EMSGSIZE;
-	if (entry->key.term_mac.copy_to_cpu &&
-	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
-			      entry->key.term_mac.copy_to_cpu))
-		return -EMSGSIZE;
+	err = rocker_world_port_open(rocker_port);
+	if (err) {
+		netdev_err(rocker_port->dev, "cannot open port in world\n");
+		goto err_world_port_open;
+	}
 
+	napi_enable(&rocker_port->napi_tx);
+	napi_enable(&rocker_port->napi_rx);
+	if (!dev->proto_down)
+		rocker_port_set_enable(rocker_port, true);
+	netif_start_queue(dev);
 	return 0;
+
+err_world_port_open:
+	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
+err_request_rx_irq:
+	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
+err_request_tx_irq:
+	rocker_port_dma_rings_fini(rocker_port);
+	return err;
 }
 
-static int
-rocker_cmd_flow_tbl_add_ucast_routing(struct rocker_desc_info *desc_info,
-				      const struct rocker_flow_tbl_entry *entry)
+static int rocker_port_stop(struct net_device *dev)
 {
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
-				entry->key.ucast_routing.eth_type))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP,
-				entry->key.ucast_routing.dst4))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP_MASK,
-				entry->key.ucast_routing.dst4_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.ucast_routing.goto_tbl))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->key.ucast_routing.group_id))
-		return -EMSGSIZE;
+	struct rocker_port *rocker_port = netdev_priv(dev);
+
+	netif_stop_queue(dev);
+	rocker_port_set_enable(rocker_port, false);
+	napi_disable(&rocker_port->napi_rx);
+	napi_disable(&rocker_port->napi_tx);
+	rocker_world_port_stop(rocker_port);
+	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
+	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
+	rocker_port_dma_rings_fini(rocker_port);
 
 	return 0;
 }
 
-static int
-rocker_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info,
-			       const struct rocker_flow_tbl_entry *entry)
+static void rocker_tx_desc_frags_unmap(const struct rocker_port *rocker_port,
+				       const struct rocker_desc_info *desc_info)
 {
-	if (entry->key.bridge.has_eth_dst &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->key.bridge.eth_dst))
-		return -EMSGSIZE;
-	if (entry->key.bridge.has_eth_dst_mask &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
-			   ETH_ALEN, entry->key.bridge.eth_dst_mask))
-		return -EMSGSIZE;
-	if (entry->key.bridge.vlan_id &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.bridge.vlan_id))
-		return -EMSGSIZE;
-	if (entry->key.bridge.tunnel_id &&
-	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_TUNNEL_ID,
-			       entry->key.bridge.tunnel_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
-			       entry->key.bridge.goto_tbl))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->key.bridge.group_id))
-		return -EMSGSIZE;
-	if (entry->key.bridge.copy_to_cpu &&
-	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
-			      entry->key.bridge.copy_to_cpu))
-		return -EMSGSIZE;
+	const struct rocker *rocker = rocker_port->rocker;
+	struct pci_dev *pdev = rocker->pdev;
+	const struct rocker_tlv *attrs[ROCKER_TLV_TX_MAX + 1];
+	struct rocker_tlv *attr;
+	int rem;
 
-	return 0;
+	rocker_tlv_parse_desc(attrs, ROCKER_TLV_TX_MAX, desc_info);
+	if (!attrs[ROCKER_TLV_TX_FRAGS])
+		return;
+	rocker_tlv_for_each_nested(attr, attrs[ROCKER_TLV_TX_FRAGS], rem) {
+		const struct rocker_tlv *frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_MAX + 1];
+		dma_addr_t dma_handle;
+		size_t len;
+
+		if (rocker_tlv_type(attr) != ROCKER_TLV_TX_FRAG)
+			continue;
+		rocker_tlv_parse_nested(frag_attrs, ROCKER_TLV_TX_FRAG_ATTR_MAX,
+					attr);
+		if (!frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR] ||
+		    !frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN])
+			continue;
+		dma_handle = rocker_tlv_get_u64(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR]);
+		len = rocker_tlv_get_u16(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN]);
+		pci_unmap_single(pdev, dma_handle, len, DMA_TO_DEVICE);
+	}
 }
 
-static int
-rocker_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info,
-			    const struct rocker_flow_tbl_entry *entry)
+static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port,
+				       struct rocker_desc_info *desc_info,
+				       char *buf, size_t buf_len)
 {
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
-			       entry->key.acl.in_pport))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
-			       entry->key.acl.in_pport_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
-			   ETH_ALEN, entry->key.acl.eth_src))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC_MASK,
-			   ETH_ALEN, entry->key.acl.eth_src_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->key.acl.eth_dst))
-		return -EMSGSIZE;
-	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
-			   ETH_ALEN, entry->key.acl.eth_dst_mask))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
-				entry->key.acl.eth_type))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->key.acl.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
-				entry->key.acl.vlan_id_mask))
-		return -EMSGSIZE;
+	const struct rocker *rocker = rocker_port->rocker;
+	struct pci_dev *pdev = rocker->pdev;
+	dma_addr_t dma_handle;
+	struct rocker_tlv *frag;
 
-	switch (ntohs(entry->key.acl.eth_type)) {
-	case ETH_P_IP:
-	case ETH_P_IPV6:
-		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_PROTO,
-				      entry->key.acl.ip_proto))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info,
-				      ROCKER_TLV_OF_DPA_IP_PROTO_MASK,
-				      entry->key.acl.ip_proto_mask))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_DSCP,
-				      entry->key.acl.ip_tos & 0x3f))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info,
-				      ROCKER_TLV_OF_DPA_IP_DSCP_MASK,
-				      entry->key.acl.ip_tos_mask & 0x3f))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_ECN,
-				      (entry->key.acl.ip_tos & 0xc0) >> 6))
-			return -EMSGSIZE;
-		if (rocker_tlv_put_u8(desc_info,
-				      ROCKER_TLV_OF_DPA_IP_ECN_MASK,
-				      (entry->key.acl.ip_tos_mask & 0xc0) >> 6))
-			return -EMSGSIZE;
-		break;
+	dma_handle = pci_map_single(pdev, buf, buf_len, DMA_TO_DEVICE);
+	if (unlikely(pci_dma_mapping_error(pdev, dma_handle))) {
+		if (net_ratelimit())
+			netdev_err(rocker_port->dev, "failed to dma map tx frag\n");
+		return -EIO;
 	}
-
-	if (entry->key.acl.group_id != ROCKER_GROUP_NONE &&
-	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->key.acl.group_id))
-		return -EMSGSIZE;
-
+	frag = rocker_tlv_nest_start(desc_info, ROCKER_TLV_TX_FRAG);
+	if (!frag)
+		goto unmap_frag;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_TX_FRAG_ATTR_ADDR,
+			       dma_handle))
+		goto nest_cancel;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_TX_FRAG_ATTR_LEN,
+			       buf_len))
+		goto nest_cancel;
+	rocker_tlv_nest_end(desc_info, frag);
 	return 0;
+
+nest_cancel:
+	rocker_tlv_nest_cancel(desc_info, frag);
+unmap_frag:
+	pci_unmap_single(pdev, dma_handle, buf_len, DMA_TO_DEVICE);
+	return -EMSGSIZE;
 }
 
-static int rocker_cmd_flow_tbl_add(const struct rocker_port *rocker_port,
-				   struct rocker_desc_info *desc_info,
-				   void *priv)
+static netdev_tx_t rocker_port_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-	const struct rocker_flow_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-	int err = 0;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_TABLE_ID,
-			       entry->key.tbl_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_PRIORITY,
-			       entry->key.priority))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_HARDTIME, 0))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
-			       entry->cookie))
-		return -EMSGSIZE;
-
-	switch (entry->key.tbl_id) {
-	case ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT:
-		err = rocker_cmd_flow_tbl_add_ig_port(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_VLAN:
-		err = rocker_cmd_flow_tbl_add_vlan(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC:
-		err = rocker_cmd_flow_tbl_add_term_mac(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING:
-		err = rocker_cmd_flow_tbl_add_ucast_routing(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_BRIDGING:
-		err = rocker_cmd_flow_tbl_add_bridge(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_TABLE_ID_ACL_POLICY:
-		err = rocker_cmd_flow_tbl_add_acl(desc_info, entry);
-		break;
-	default:
-		err = -ENOTSUPP;
-		break;
-	}
-
-	if (err)
-		return err;
-
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-static int rocker_cmd_flow_tbl_del(const struct rocker_port *rocker_port,
-				   struct rocker_desc_info *desc_info,
-				   void *priv)
-{
-	const struct rocker_flow_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
-			       entry->cookie))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_l2_interface(struct rocker_desc_info *desc_info,
-				      struct rocker_group_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_OUT_PPORT,
-			       ROCKER_GROUP_PORT_GET(entry->group_id)))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_POP_VLAN,
-			      entry->l2_interface.pop_vlan))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_l2_rewrite(struct rocker_desc_info *desc_info,
-				    const struct rocker_group_tbl_entry *entry)
-{
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
-			       entry->l2_rewrite.group_id))
-		return -EMSGSIZE;
-	if (!is_zero_ether_addr(entry->l2_rewrite.eth_src) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
-			   ETH_ALEN, entry->l2_rewrite.eth_src))
-		return -EMSGSIZE;
-	if (!is_zero_ether_addr(entry->l2_rewrite.eth_dst) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->l2_rewrite.eth_dst))
-		return -EMSGSIZE;
-	if (entry->l2_rewrite.vlan_id &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->l2_rewrite.vlan_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_group_ids(struct rocker_desc_info *desc_info,
-				   const struct rocker_group_tbl_entry *entry)
-{
-	int i;
-	struct rocker_tlv *group_ids;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GROUP_COUNT,
-			       entry->group_count))
-		return -EMSGSIZE;
-
-	group_ids = rocker_tlv_nest_start(desc_info,
-					  ROCKER_TLV_OF_DPA_GROUP_IDS);
-	if (!group_ids)
-		return -EMSGSIZE;
-
-	for (i = 0; i < entry->group_count; i++)
-		/* Note TLV array is 1-based */
-		if (rocker_tlv_put_u32(desc_info, i + 1, entry->group_ids[i]))
-			return -EMSGSIZE;
-
-	rocker_tlv_nest_end(desc_info, group_ids);
-
-	return 0;
-}
-
-static int
-rocker_cmd_group_tbl_add_l3_unicast(struct rocker_desc_info *desc_info,
-				    const struct rocker_group_tbl_entry *entry)
-{
-	if (!is_zero_ether_addr(entry->l3_unicast.eth_src) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
-			   ETH_ALEN, entry->l3_unicast.eth_src))
-		return -EMSGSIZE;
-	if (!is_zero_ether_addr(entry->l3_unicast.eth_dst) &&
-	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
-			   ETH_ALEN, entry->l3_unicast.eth_dst))
-		return -EMSGSIZE;
-	if (entry->l3_unicast.vlan_id &&
-	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
-				entry->l3_unicast.vlan_id))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_TTL_CHECK,
-			      entry->l3_unicast.ttl_check))
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
-			       entry->l3_unicast.group_id))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
-static int rocker_cmd_group_tbl_add(const struct rocker_port *rocker_port,
-				    struct rocker_desc_info *desc_info,
-				    void *priv)
-{
-	struct rocker_group_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-	int err = 0;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->group_id))
-		return -EMSGSIZE;
-
-	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE:
-		err = rocker_cmd_group_tbl_add_l2_interface(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE:
-		err = rocker_cmd_group_tbl_add_l2_rewrite(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
-		err = rocker_cmd_group_tbl_add_group_ids(desc_info, entry);
-		break;
-	case ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST:
-		err = rocker_cmd_group_tbl_add_l3_unicast(desc_info, entry);
-		break;
-	default:
-		err = -ENOTSUPP;
-		break;
-	}
-
-	if (err)
-		return err;
-
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-static int rocker_cmd_group_tbl_del(const struct rocker_port *rocker_port,
-				    struct rocker_desc_info *desc_info,
-				    void *priv)
-{
-	const struct rocker_group_tbl_entry *entry = priv;
-	struct rocker_tlv *cmd_info;
-
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
-		return -EMSGSIZE;
-	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
-	if (!cmd_info)
-		return -EMSGSIZE;
-	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
-			       entry->group_id))
-		return -EMSGSIZE;
-	rocker_tlv_nest_end(desc_info, cmd_info);
-
-	return 0;
-}
-
-/***************************************************
- * Flow, group, FDB, internal VLAN and neigh tables
- ***************************************************/
-
-static int rocker_init_tbls(struct rocker *rocker)
-{
-	hash_init(rocker->flow_tbl);
-	spin_lock_init(&rocker->flow_tbl_lock);
-
-	hash_init(rocker->group_tbl);
-	spin_lock_init(&rocker->group_tbl_lock);
-
-	hash_init(rocker->fdb_tbl);
-	spin_lock_init(&rocker->fdb_tbl_lock);
-
-	hash_init(rocker->internal_vlan_tbl);
-	spin_lock_init(&rocker->internal_vlan_tbl_lock);
-
-	hash_init(rocker->neigh_tbl);
-	spin_lock_init(&rocker->neigh_tbl_lock);
-
-	return 0;
-}
-
-static void rocker_free_tbls(struct rocker *rocker)
-{
-	unsigned long flags;
-	struct rocker_flow_tbl_entry *flow_entry;
-	struct rocker_group_tbl_entry *group_entry;
-	struct rocker_fdb_tbl_entry *fdb_entry;
-	struct rocker_internal_vlan_tbl_entry *internal_vlan_entry;
-	struct rocker_neigh_tbl_entry *neigh_entry;
-	struct hlist_node *tmp;
-	int bkt;
-
-	spin_lock_irqsave(&rocker->flow_tbl_lock, flags);
-	hash_for_each_safe(rocker->flow_tbl, bkt, tmp, flow_entry, entry)
-		hash_del(&flow_entry->entry);
-	spin_unlock_irqrestore(&rocker->flow_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->group_tbl_lock, flags);
-	hash_for_each_safe(rocker->group_tbl, bkt, tmp, group_entry, entry)
-		hash_del(&group_entry->entry);
-	spin_unlock_irqrestore(&rocker->group_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, flags);
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, fdb_entry, entry)
-		hash_del(&fdb_entry->entry);
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, flags);
-	hash_for_each_safe(rocker->internal_vlan_tbl, bkt,
-			   tmp, internal_vlan_entry, entry)
-		hash_del(&internal_vlan_entry->entry);
-	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, flags);
-
-	spin_lock_irqsave(&rocker->neigh_tbl_lock, flags);
-	hash_for_each_safe(rocker->neigh_tbl, bkt, tmp, neigh_entry, entry)
-		hash_del(&neigh_entry->entry);
-	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, flags);
-}
-
-static struct rocker_flow_tbl_entry *
-rocker_flow_tbl_find(const struct rocker *rocker,
-		     const struct rocker_flow_tbl_entry *match)
-{
-	struct rocker_flow_tbl_entry *found;
-	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
-
-	hash_for_each_possible(rocker->flow_tbl, found,
-			       entry, match->key_crc32) {
-		if (memcmp(&found->key, &match->key, key_len) == 0)
-			return found;
-	}
-
-	return NULL;
-}
-
-static int rocker_flow_tbl_add(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       struct rocker_flow_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_flow_tbl_entry *found;
-	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
-	unsigned long lock_flags;
-
-	match->key_crc32 = crc32(~0, &match->key, key_len);
-
-	spin_lock_irqsave(&rocker->flow_tbl_lock, lock_flags);
-
-	found = rocker_flow_tbl_find(rocker, match);
-
-	if (found) {
-		match->cookie = found->cookie;
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		rocker_kfree(trans, found);
-		found = match;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD;
-	} else {
-		found = match;
-		found->cookie = rocker->flow_tbl_next_cookie++;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD;
-	}
-
-	if (!switchdev_trans_ph_prepare(trans))
-		hash_add(rocker->flow_tbl, &found->entry, found->key_crc32);
-
-	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
-
-	if (!switchdev_trans_ph_prepare(trans))
-		return rocker_cmd_exec(rocker_port,
-				       rocker_flags_nowait(flags),
-				       rocker_cmd_flow_tbl_add,
-				       found, NULL, NULL);
-	return 0;
-}
-
-static int rocker_flow_tbl_del(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       struct rocker_flow_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_flow_tbl_entry *found;
-	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
-	unsigned long lock_flags;
-	int err = 0;
-
-	match->key_crc32 = crc32(~0, &match->key, key_len);
-
-	spin_lock_irqsave(&rocker->flow_tbl_lock, lock_flags);
-
-	found = rocker_flow_tbl_find(rocker, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL;
-	}
-
-	spin_unlock_irqrestore(&rocker->flow_tbl_lock, lock_flags);
-
-	rocker_kfree(trans, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			err = rocker_cmd_exec(rocker_port,
-					      rocker_flags_nowait(flags),
-					      rocker_cmd_flow_tbl_del,
-					      found, NULL, NULL);
-		rocker_kfree(trans, found);
-	}
-
-	return err;
-}
-
-static int rocker_flow_tbl_do(struct rocker_port *rocker_port,
-			      struct switchdev_trans *trans, int flags,
-			      struct rocker_flow_tbl_entry *entry)
-{
-	if (flags & ROCKER_OP_FLAG_REMOVE)
-		return rocker_flow_tbl_del(rocker_port, trans, flags, entry);
-	else
-		return rocker_flow_tbl_add(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_ig_port(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans, int flags,
-				   u32 in_pport, u32 in_pport_mask,
-				   enum rocker_of_dpa_table_id goto_tbl)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.priority = ROCKER_PRIORITY_IG_PORT;
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT;
-	entry->key.ig_port.in_pport = in_pport;
-	entry->key.ig_port.in_pport_mask = in_pport_mask;
-	entry->key.ig_port.goto_tbl = goto_tbl;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_vlan(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, int flags,
-				u32 in_pport, __be16 vlan_id,
-				__be16 vlan_id_mask,
-				enum rocker_of_dpa_table_id goto_tbl,
-				bool untagged, __be16 new_vlan_id)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.priority = ROCKER_PRIORITY_VLAN;
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_VLAN;
-	entry->key.vlan.in_pport = in_pport;
-	entry->key.vlan.vlan_id = vlan_id;
-	entry->key.vlan.vlan_id_mask = vlan_id_mask;
-	entry->key.vlan.goto_tbl = goto_tbl;
-
-	entry->key.vlan.untagged = untagged;
-	entry->key.vlan.new_vlan_id = new_vlan_id;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_term_mac(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans,
-				    u32 in_pport, u32 in_pport_mask,
-				    __be16 eth_type, const u8 *eth_dst,
-				    const u8 *eth_dst_mask, __be16 vlan_id,
-				    __be16 vlan_id_mask, bool copy_to_cpu,
-				    int flags)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	if (is_multicast_ether_addr(eth_dst)) {
-		entry->key.priority = ROCKER_PRIORITY_TERM_MAC_MCAST;
-		entry->key.term_mac.goto_tbl =
-			 ROCKER_OF_DPA_TABLE_ID_MULTICAST_ROUTING;
-	} else {
-		entry->key.priority = ROCKER_PRIORITY_TERM_MAC_UCAST;
-		entry->key.term_mac.goto_tbl =
-			 ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
-	}
-
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
-	entry->key.term_mac.in_pport = in_pport;
-	entry->key.term_mac.in_pport_mask = in_pport_mask;
-	entry->key.term_mac.eth_type = eth_type;
-	ether_addr_copy(entry->key.term_mac.eth_dst, eth_dst);
-	ether_addr_copy(entry->key.term_mac.eth_dst_mask, eth_dst_mask);
-	entry->key.term_mac.vlan_id = vlan_id;
-	entry->key.term_mac.vlan_id_mask = vlan_id_mask;
-	entry->key.term_mac.copy_to_cpu = copy_to_cpu;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_bridge(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags,
-				  const u8 *eth_dst, const u8 *eth_dst_mask,
-				  __be16 vlan_id, u32 tunnel_id,
-				  enum rocker_of_dpa_table_id goto_tbl,
-				  u32 group_id, bool copy_to_cpu)
-{
-	struct rocker_flow_tbl_entry *entry;
-	u32 priority;
-	bool vlan_bridging = !!vlan_id;
-	bool dflt = !eth_dst || (eth_dst && eth_dst_mask);
-	bool wild = false;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_BRIDGING;
-
-	if (eth_dst) {
-		entry->key.bridge.has_eth_dst = 1;
-		ether_addr_copy(entry->key.bridge.eth_dst, eth_dst);
-	}
-	if (eth_dst_mask) {
-		entry->key.bridge.has_eth_dst_mask = 1;
-		ether_addr_copy(entry->key.bridge.eth_dst_mask, eth_dst_mask);
-		if (!ether_addr_equal(eth_dst_mask, ff_mac))
-			wild = true;
-	}
-
-	priority = ROCKER_PRIORITY_UNKNOWN;
-	if (vlan_bridging && dflt && wild)
-		priority = ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_WILD;
-	else if (vlan_bridging && dflt && !wild)
-		priority = ROCKER_PRIORITY_BRIDGING_VLAN_DFLT_EXACT;
-	else if (vlan_bridging && !dflt)
-		priority = ROCKER_PRIORITY_BRIDGING_VLAN;
-	else if (!vlan_bridging && dflt && wild)
-		priority = ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_WILD;
-	else if (!vlan_bridging && dflt && !wild)
-		priority = ROCKER_PRIORITY_BRIDGING_TENANT_DFLT_EXACT;
-	else if (!vlan_bridging && !dflt)
-		priority = ROCKER_PRIORITY_BRIDGING_TENANT;
-
-	entry->key.priority = priority;
-	entry->key.bridge.vlan_id = vlan_id;
-	entry->key.bridge.tunnel_id = tunnel_id;
-	entry->key.bridge.goto_tbl = goto_tbl;
-	entry->key.bridge.group_id = group_id;
-	entry->key.bridge.copy_to_cpu = copy_to_cpu;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_ucast4_routing(struct rocker_port *rocker_port,
-					  struct switchdev_trans *trans,
-					  __be16 eth_type, __be32 dst,
-					  __be32 dst_mask, u32 priority,
-					  enum rocker_of_dpa_table_id goto_tbl,
-					  u32 group_id, int flags)
-{
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
-	entry->key.priority = priority;
-	entry->key.ucast_routing.eth_type = eth_type;
-	entry->key.ucast_routing.dst4 = dst;
-	entry->key.ucast_routing.dst4_mask = dst_mask;
-	entry->key.ucast_routing.goto_tbl = goto_tbl;
-	entry->key.ucast_routing.group_id = group_id;
-	entry->key_len = offsetof(struct rocker_flow_tbl_key,
-				  ucast_routing.group_id);
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_flow_tbl_acl(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       u32 in_pport, u32 in_pport_mask,
-			       const u8 *eth_src, const u8 *eth_src_mask,
-			       const u8 *eth_dst, const u8 *eth_dst_mask,
-			       __be16 eth_type, __be16 vlan_id,
-			       __be16 vlan_id_mask, u8 ip_proto,
-			       u8 ip_proto_mask, u8 ip_tos, u8 ip_tos_mask,
-			       u32 group_id)
-{
-	u32 priority;
-	struct rocker_flow_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	priority = ROCKER_PRIORITY_ACL_NORMAL;
-	if (eth_dst && eth_dst_mask) {
-		if (ether_addr_equal(eth_dst_mask, mcast_mac))
-			priority = ROCKER_PRIORITY_ACL_DFLT;
-		else if (is_link_local_ether_addr(eth_dst))
-			priority = ROCKER_PRIORITY_ACL_CTRL;
-	}
-
-	entry->key.priority = priority;
-	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	entry->key.acl.in_pport = in_pport;
-	entry->key.acl.in_pport_mask = in_pport_mask;
-
-	if (eth_src)
-		ether_addr_copy(entry->key.acl.eth_src, eth_src);
-	if (eth_src_mask)
-		ether_addr_copy(entry->key.acl.eth_src_mask, eth_src_mask);
-	if (eth_dst)
-		ether_addr_copy(entry->key.acl.eth_dst, eth_dst);
-	if (eth_dst_mask)
-		ether_addr_copy(entry->key.acl.eth_dst_mask, eth_dst_mask);
-
-	entry->key.acl.eth_type = eth_type;
-	entry->key.acl.vlan_id = vlan_id;
-	entry->key.acl.vlan_id_mask = vlan_id_mask;
-	entry->key.acl.ip_proto = ip_proto;
-	entry->key.acl.ip_proto_mask = ip_proto_mask;
-	entry->key.acl.ip_tos = ip_tos;
-	entry->key.acl.ip_tos_mask = ip_tos_mask;
-	entry->key.acl.group_id = group_id;
-
-	return rocker_flow_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static struct rocker_group_tbl_entry *
-rocker_group_tbl_find(const struct rocker *rocker,
-		      const struct rocker_group_tbl_entry *match)
-{
-	struct rocker_group_tbl_entry *found;
-
-	hash_for_each_possible(rocker->group_tbl, found,
-			       entry, match->group_id) {
-		if (found->group_id == match->group_id)
-			return found;
-	}
-
-	return NULL;
-}
-
-static void rocker_group_tbl_entry_free(struct switchdev_trans *trans,
-					struct rocker_group_tbl_entry *entry)
-{
-	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
-	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
-		rocker_kfree(trans, entry->group_ids);
-		break;
-	default:
-		break;
-	}
-	rocker_kfree(trans, entry);
-}
-
-static int rocker_group_tbl_add(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, int flags,
-				struct rocker_group_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_group_tbl_entry *found;
-	unsigned long lock_flags;
-
-	spin_lock_irqsave(&rocker->group_tbl_lock, lock_flags);
-
-	found = rocker_group_tbl_find(rocker, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		rocker_group_tbl_entry_free(trans, found);
-		found = match;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD;
-	} else {
-		found = match;
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD;
-	}
-
-	if (!switchdev_trans_ph_prepare(trans))
-		hash_add(rocker->group_tbl, &found->entry, found->group_id);
-
-	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
-
-	if (!switchdev_trans_ph_prepare(trans))
-		return rocker_cmd_exec(rocker_port,
-				       rocker_flags_nowait(flags),
-				       rocker_cmd_group_tbl_add,
-				       found, NULL, NULL);
-	return 0;
-}
-
-static int rocker_group_tbl_del(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, int flags,
-				struct rocker_group_tbl_entry *match)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_group_tbl_entry *found;
-	unsigned long lock_flags;
-	int err = 0;
-
-	spin_lock_irqsave(&rocker->group_tbl_lock, lock_flags);
-
-	found = rocker_group_tbl_find(rocker, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL;
-	}
-
-	spin_unlock_irqrestore(&rocker->group_tbl_lock, lock_flags);
-
-	rocker_group_tbl_entry_free(trans, match);
-
-	if (found) {
-		if (!switchdev_trans_ph_prepare(trans))
-			err = rocker_cmd_exec(rocker_port,
-					      rocker_flags_nowait(flags),
-					      rocker_cmd_group_tbl_del,
-					      found, NULL, NULL);
-		rocker_group_tbl_entry_free(trans, found);
-	}
-
-	return err;
-}
-
-static int rocker_group_tbl_do(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       struct rocker_group_tbl_entry *entry)
-{
-	if (flags & ROCKER_OP_FLAG_REMOVE)
-		return rocker_group_tbl_del(rocker_port, trans, flags, entry);
-	else
-		return rocker_group_tbl_add(rocker_port, trans, flags, entry);
-}
-
-static int rocker_group_l2_interface(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     __be16 vlan_id, u32 out_pport,
-				     int pop_vlan)
-{
-	struct rocker_group_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
-	entry->l2_interface.pop_vlan = pop_vlan;
-
-	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_group_l2_fan_out(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans,
-				   int flags, u8 group_count,
-				   const u32 *group_ids, u32 group_id)
-{
-	struct rocker_group_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->group_id = group_id;
-	entry->group_count = group_count;
-
-	entry->group_ids = rocker_kcalloc(trans, flags,
-					  group_count, sizeof(u32));
-	if (!entry->group_ids) {
-		rocker_kfree(trans, entry);
-		return -ENOMEM;
-	}
-	memcpy(entry->group_ids, group_ids, group_count * sizeof(u32));
-
-	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static int rocker_group_l2_flood(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 __be16 vlan_id, u8 group_count,
-				 const u32 *group_ids, u32 group_id)
-{
-	return rocker_group_l2_fan_out(rocker_port, trans, flags,
-				       group_count, group_ids,
-				       group_id);
-}
-
-static int rocker_group_l3_unicast(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans, int flags,
-				   u32 index, const u8 *src_mac, const u8 *dst_mac,
-				   __be16 vlan_id, bool ttl_check, u32 pport)
-{
-	struct rocker_group_tbl_entry *entry;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	entry->group_id = ROCKER_GROUP_L3_UNICAST(index);
-	if (src_mac)
-		ether_addr_copy(entry->l3_unicast.eth_src, src_mac);
-	if (dst_mac)
-		ether_addr_copy(entry->l3_unicast.eth_dst, dst_mac);
-	entry->l3_unicast.vlan_id = vlan_id;
-	entry->l3_unicast.ttl_check = ttl_check;
-	entry->l3_unicast.group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, pport);
-
-	return rocker_group_tbl_do(rocker_port, trans, flags, entry);
-}
-
-static struct rocker_neigh_tbl_entry *
-rocker_neigh_tbl_find(const struct rocker *rocker, __be32 ip_addr)
-{
-	struct rocker_neigh_tbl_entry *found;
-
-	hash_for_each_possible(rocker->neigh_tbl, found,
-			       entry, be32_to_cpu(ip_addr))
-		if (found->ip_addr == ip_addr)
-			return found;
-
-	return NULL;
-}
-
-static void _rocker_neigh_add(struct rocker *rocker,
-			      struct switchdev_trans *trans,
-			      struct rocker_neigh_tbl_entry *entry)
-{
-	if (!switchdev_trans_ph_commit(trans))
-		entry->index = rocker->neigh_tbl_next_index++;
-	if (switchdev_trans_ph_prepare(trans))
-		return;
-	entry->ref_count++;
-	hash_add(rocker->neigh_tbl, &entry->entry,
-		 be32_to_cpu(entry->ip_addr));
-}
-
-static void _rocker_neigh_del(struct switchdev_trans *trans,
-			      struct rocker_neigh_tbl_entry *entry)
-{
-	if (switchdev_trans_ph_prepare(trans))
-		return;
-	if (--entry->ref_count == 0) {
-		hash_del(&entry->entry);
-		rocker_kfree(trans, entry);
-	}
-}
-
-static void _rocker_neigh_update(struct rocker_neigh_tbl_entry *entry,
-				 struct switchdev_trans *trans,
-				 const u8 *eth_dst, bool ttl_check)
-{
-	if (eth_dst) {
-		ether_addr_copy(entry->eth_dst, eth_dst);
-		entry->ttl_check = ttl_check;
-	} else if (!switchdev_trans_ph_prepare(trans)) {
-		entry->ref_count++;
-	}
-}
-
-static int rocker_port_ipv4_neigh(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans,
-				  int flags, __be32 ip_addr, const u8 *eth_dst)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_neigh_tbl_entry *entry;
-	struct rocker_neigh_tbl_entry *found;
-	unsigned long lock_flags;
-	__be16 eth_type = htons(ETH_P_IP);
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 group_id;
-	u32 priority = 0;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	bool updating;
-	bool removing;
-	int err = 0;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	spin_lock_irqsave(&rocker->neigh_tbl_lock, lock_flags);
-
-	found = rocker_neigh_tbl_find(rocker, ip_addr);
-
-	updating = found && adding;
-	removing = found && !adding;
-	adding = !found && adding;
-
-	if (adding) {
-		entry->ip_addr = ip_addr;
-		entry->dev = rocker_port->dev;
-		ether_addr_copy(entry->eth_dst, eth_dst);
-		entry->ttl_check = true;
-		_rocker_neigh_add(rocker, trans, entry);
-	} else if (removing) {
-		memcpy(entry, found, sizeof(*entry));
-		_rocker_neigh_del(trans, found);
-	} else if (updating) {
-		_rocker_neigh_update(found, trans, eth_dst, true);
-		memcpy(entry, found, sizeof(*entry));
-	} else {
-		err = -ENOENT;
-	}
-
-	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
-
-	if (err)
-		goto err_out;
-
-	/* For each active neighbor, we have an L3 unicast group and
-	 * a /32 route to the neighbor, which uses the L3 unicast
-	 * group.  The L3 unicast group can also be referred to by
-	 * other routes' nexthops.
-	 */
-
-	err = rocker_group_l3_unicast(rocker_port, trans, flags,
-				      entry->index,
-				      rocker_port->dev->dev_addr,
-				      entry->eth_dst,
-				      rocker_port->internal_vlan_id,
-				      entry->ttl_check,
-				      rocker_port->pport);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) L3 unicast group index %d\n",
-			   err, entry->index);
-		goto err_out;
-	}
-
-	if (adding || removing) {
-		group_id = ROCKER_GROUP_L3_UNICAST(entry->index);
-		err = rocker_flow_tbl_ucast4_routing(rocker_port, trans,
-						     eth_type, ip_addr,
-						     inet_make_mask(32),
-						     priority, goto_tbl,
-						     group_id, flags);
-
-		if (err)
-			netdev_err(rocker_port->dev,
-				   "Error (%d) /32 unicast route %pI4 group 0x%08x\n",
-				   err, &entry->ip_addr, group_id);
-	}
-
-err_out:
-	if (!adding)
-		rocker_kfree(trans, entry);
-
-	return err;
-}
-
-static int rocker_port_ipv4_resolve(struct rocker_port *rocker_port,
-				    struct switchdev_trans *trans,
-				    __be32 ip_addr)
-{
-	struct net_device *dev = rocker_port->dev;
-	struct neighbour *n = __ipv4_neigh_lookup(dev, (__force u32)ip_addr);
-	int err = 0;
-
-	if (!n) {
-		n = neigh_create(&arp_tbl, &ip_addr, dev);
-		if (IS_ERR(n))
-			return IS_ERR(n);
-	}
-
-	/* If the neigh is already resolved, then go ahead and
-	 * install the entry, otherwise start the ARP process to
-	 * resolve the neigh.
-	 */
-
-	if (n->nud_state & NUD_VALID)
-		err = rocker_port_ipv4_neigh(rocker_port, trans, 0,
-					     ip_addr, n->ha);
-	else
-		neigh_event_send(n, NULL);
-
-	neigh_release(n);
-	return err;
-}
-
-static int rocker_port_ipv4_nh(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans, int flags,
-			       __be32 ip_addr, u32 *index)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_neigh_tbl_entry *entry;
-	struct rocker_neigh_tbl_entry *found;
-	unsigned long lock_flags;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	bool updating;
-	bool removing;
-	bool resolved = true;
-	int err = 0;
-
-	entry = rocker_kzalloc(trans, flags, sizeof(*entry));
-	if (!entry)
-		return -ENOMEM;
-
-	spin_lock_irqsave(&rocker->neigh_tbl_lock, lock_flags);
-
-	found = rocker_neigh_tbl_find(rocker, ip_addr);
-	if (found)
-		*index = found->index;
-
-	updating = found && adding;
-	removing = found && !adding;
-	adding = !found && adding;
-
-	if (adding) {
-		entry->ip_addr = ip_addr;
-		entry->dev = rocker_port->dev;
-		_rocker_neigh_add(rocker, trans, entry);
-		*index = entry->index;
-		resolved = false;
-	} else if (removing) {
-		_rocker_neigh_del(trans, found);
-	} else if (updating) {
-		_rocker_neigh_update(found, trans, NULL, false);
-		resolved = !is_zero_ether_addr(found->eth_dst);
-	} else {
-		err = -ENOENT;
-	}
-
-	spin_unlock_irqrestore(&rocker->neigh_tbl_lock, lock_flags);
-
-	if (!adding)
-		rocker_kfree(trans, entry);
-
-	if (err)
-		return err;
-
-	/* Resolved means neigh ip_addr is resolved to neigh mac. */
-
-	if (!resolved)
-		err = rocker_port_ipv4_resolve(rocker_port, trans, ip_addr);
-
-	return err;
-}
-
-static int rocker_port_vlan_flood_group(struct rocker_port *rocker_port,
-					struct switchdev_trans *trans,
-					int flags, __be16 vlan_id)
-{
-	struct rocker_port *p;
-	const struct rocker *rocker = rocker_port->rocker;
-	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
-	u32 *group_ids;
-	u8 group_count = 0;
-	int err = 0;
-	int i;
-
-	group_ids = rocker_kcalloc(trans, flags,
-				   rocker->port_count, sizeof(u32));
-	if (!group_ids)
-		return -ENOMEM;
-
-	/* Adjust the flood group for this VLAN.  The flood group
-	 * references an L2 interface group for each port in this
-	 * VLAN.
-	 */
-
-	for (i = 0; i < rocker->port_count; i++) {
-		p = rocker->ports[i];
-		if (!p)
-			continue;
-		if (!rocker_port_is_bridged(p))
-			continue;
-		if (test_bit(ntohs(vlan_id), p->vlan_bitmap)) {
-			group_ids[group_count++] =
-				ROCKER_GROUP_L2_INTERFACE(vlan_id, p->pport);
-		}
-	}
-
-	/* If there are no bridged ports in this VLAN, we're done */
-	if (group_count == 0)
-		goto no_ports_in_vlan;
-
-	err = rocker_group_l2_flood(rocker_port, trans, flags, vlan_id,
-				    group_count, group_ids, group_id);
-	if (err)
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 flood group\n", err);
-
-no_ports_in_vlan:
-	rocker_kfree(trans, group_ids);
-	return err;
-}
-
-static int rocker_port_vlan_l2_groups(struct rocker_port *rocker_port,
-				      struct switchdev_trans *trans, int flags,
-				      __be16 vlan_id, bool pop_vlan)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct rocker_port *p;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	u32 out_pport;
-	int ref = 0;
-	int err;
-	int i;
-
-	/* An L2 interface group for this port in this VLAN, but
-	 * only when port STP state is LEARNING|FORWARDING.
-	 */
-
-	if (rocker_port->stp_state == BR_STATE_LEARNING ||
-	    rocker_port->stp_state == BR_STATE_FORWARDING) {
-		out_pport = rocker_port->pport;
-		err = rocker_group_l2_interface(rocker_port, trans, flags,
-						vlan_id, out_pport, pop_vlan);
-		if (err) {
-			netdev_err(rocker_port->dev,
-				   "Error (%d) port VLAN l2 group for pport %d\n",
-				   err, out_pport);
-			return err;
-		}
-	}
-
-	/* An L2 interface group for this VLAN to CPU port.
-	 * Add when first port joins this VLAN and destroy when
-	 * last port leaves this VLAN.
-	 */
-
-	for (i = 0; i < rocker->port_count; i++) {
-		p = rocker->ports[i];
-		if (p && test_bit(ntohs(vlan_id), p->vlan_bitmap))
-			ref++;
-	}
-
-	if ((!adding || ref != 1) && (adding || ref != 0))
-		return 0;
-
-	out_pport = 0;
-	err = rocker_group_l2_interface(rocker_port, trans, flags,
-					vlan_id, out_pport, pop_vlan);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 group for CPU port\n", err);
-		return err;
-	}
-
-	return 0;
-}
-
-static struct rocker_ctrl {
-	const u8 *eth_dst;
-	const u8 *eth_dst_mask;
-	__be16 eth_type;
-	bool acl;
-	bool bridge;
-	bool term;
-	bool copy_to_cpu;
-} rocker_ctrls[] = {
-	[ROCKER_CTRL_LINK_LOCAL_MCAST] = {
-		/* pass link local multicast pkts up to CPU for filtering */
-		.eth_dst = ll_mac,
-		.eth_dst_mask = ll_mask,
-		.acl = true,
-	},
-	[ROCKER_CTRL_LOCAL_ARP] = {
-		/* pass local ARP pkts up to CPU */
-		.eth_dst = zero_mac,
-		.eth_dst_mask = zero_mac,
-		.eth_type = htons(ETH_P_ARP),
-		.acl = true,
-	},
-	[ROCKER_CTRL_IPV4_MCAST] = {
-		/* pass IPv4 mcast pkts up to CPU, RFC 1112 */
-		.eth_dst = ipv4_mcast,
-		.eth_dst_mask = ipv4_mask,
-		.eth_type = htons(ETH_P_IP),
-		.term  = true,
-		.copy_to_cpu = true,
-	},
-	[ROCKER_CTRL_IPV6_MCAST] = {
-		/* pass IPv6 mcast pkts up to CPU, RFC 2464 */
-		.eth_dst = ipv6_mcast,
-		.eth_dst_mask = ipv6_mask,
-		.eth_type = htons(ETH_P_IPV6),
-		.term  = true,
-		.copy_to_cpu = true,
-	},
-	[ROCKER_CTRL_DFLT_BRIDGING] = {
-		/* flood any pkts on vlan */
-		.bridge = true,
-		.copy_to_cpu = true,
-	},
-	[ROCKER_CTRL_DFLT_OVS] = {
-		/* pass all pkts up to CPU */
-		.eth_dst = zero_mac,
-		.eth_dst_mask = zero_mac,
-		.acl = true,
-	},
-};
-
-static int rocker_port_ctrl_vlan_acl(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     const struct rocker_ctrl *ctrl, __be16 vlan_id)
-{
-	u32 in_pport = rocker_port->pport;
-	u32 in_pport_mask = 0xffffffff;
-	u32 out_pport = 0;
-	const u8 *eth_src = NULL;
-	const u8 *eth_src_mask = NULL;
-	__be16 vlan_id_mask = htons(0xffff);
-	u8 ip_proto = 0;
-	u8 ip_proto_mask = 0;
-	u8 ip_tos = 0;
-	u8 ip_tos_mask = 0;
-	u32 group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
-	int err;
-
-	err = rocker_flow_tbl_acl(rocker_port, trans, flags,
-				  in_pport, in_pport_mask,
-				  eth_src, eth_src_mask,
-				  ctrl->eth_dst, ctrl->eth_dst_mask,
-				  ctrl->eth_type,
-				  vlan_id, vlan_id_mask,
-				  ip_proto, ip_proto_mask,
-				  ip_tos, ip_tos_mask,
-				  group_id);
-
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) ctrl ACL\n", err);
-
-	return err;
-}
-
-static int rocker_port_ctrl_vlan_bridge(struct rocker_port *rocker_port,
-					struct switchdev_trans *trans,
-					int flags,
-					const struct rocker_ctrl *ctrl,
-					__be16 vlan_id)
-{
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
-	u32 tunnel_id = 0;
-	int err;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return 0;
-
-	err = rocker_flow_tbl_bridge(rocker_port, trans, flags,
-				     ctrl->eth_dst, ctrl->eth_dst_mask,
-				     vlan_id, tunnel_id,
-				     goto_tbl, group_id, ctrl->copy_to_cpu);
-
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) ctrl FLOOD\n", err);
-
-	return err;
-}
-
-static int rocker_port_ctrl_vlan_term(struct rocker_port *rocker_port,
-				      struct switchdev_trans *trans, int flags,
-				      const struct rocker_ctrl *ctrl, __be16 vlan_id)
-{
-	u32 in_pport_mask = 0xffffffff;
-	__be16 vlan_id_mask = htons(0xffff);
-	int err;
-
-	if (ntohs(vlan_id) == 0)
-		vlan_id = rocker_port->internal_vlan_id;
-
-	err = rocker_flow_tbl_term_mac(rocker_port, trans,
-				       rocker_port->pport, in_pport_mask,
-				       ctrl->eth_type, ctrl->eth_dst,
-				       ctrl->eth_dst_mask, vlan_id,
-				       vlan_id_mask, ctrl->copy_to_cpu,
-				       flags);
-
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) ctrl term\n", err);
-
-	return err;
-}
-
-static int rocker_port_ctrl_vlan(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 const struct rocker_ctrl *ctrl, __be16 vlan_id)
-{
-	if (ctrl->acl)
-		return rocker_port_ctrl_vlan_acl(rocker_port, trans, flags,
-						 ctrl, vlan_id);
-	if (ctrl->bridge)
-		return rocker_port_ctrl_vlan_bridge(rocker_port, trans, flags,
-						    ctrl, vlan_id);
-
-	if (ctrl->term)
-		return rocker_port_ctrl_vlan_term(rocker_port, trans, flags,
-						  ctrl, vlan_id);
-
-	return -EOPNOTSUPP;
-}
-
-static int rocker_port_ctrl_vlan_add(struct rocker_port *rocker_port,
-				     struct switchdev_trans *trans, int flags,
-				     __be16 vlan_id)
-{
-	int err = 0;
-	int i;
-
-	for (i = 0; i < ROCKER_CTRL_MAX; i++) {
-		if (rocker_port->ctrls[i]) {
-			err = rocker_port_ctrl_vlan(rocker_port, trans, flags,
-						    &rocker_ctrls[i], vlan_id);
-			if (err)
-				return err;
-		}
-	}
-
-	return err;
-}
-
-static int rocker_port_ctrl(struct rocker_port *rocker_port,
-			    struct switchdev_trans *trans, int flags,
-			    const struct rocker_ctrl *ctrl)
-{
-	u16 vid;
-	int err = 0;
-
-	for (vid = 1; vid < VLAN_N_VID; vid++) {
-		if (!test_bit(vid, rocker_port->vlan_bitmap))
-			continue;
-		err = rocker_port_ctrl_vlan(rocker_port, trans, flags,
-					    ctrl, htons(vid));
-		if (err)
-			break;
-	}
-
-	return err;
-}
-
-static int rocker_port_vlan(struct rocker_port *rocker_port,
-			    struct switchdev_trans *trans, int flags, u16 vid)
-{
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
-	u32 in_pport = rocker_port->pport;
-	__be16 vlan_id = htons(vid);
-	__be16 vlan_id_mask = htons(0xffff);
-	__be16 internal_vlan_id;
-	bool untagged;
-	bool adding = !(flags & ROCKER_OP_FLAG_REMOVE);
-	int err;
-
-	internal_vlan_id = rocker_port_vid_to_vlan(rocker_port, vid, &untagged);
-
-	if (adding && test_bit(ntohs(internal_vlan_id),
-			       rocker_port->vlan_bitmap))
-		return 0; /* already added */
-	else if (!adding && !test_bit(ntohs(internal_vlan_id),
-				      rocker_port->vlan_bitmap))
-		return 0; /* already removed */
-
-	change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap);
-
-	if (adding) {
-		err = rocker_port_ctrl_vlan_add(rocker_port, trans, flags,
-						internal_vlan_id);
-		if (err) {
-			netdev_err(rocker_port->dev,
-				   "Error (%d) port ctrl vlan add\n", err);
-			goto err_out;
-		}
-	}
-
-	err = rocker_port_vlan_l2_groups(rocker_port, trans, flags,
-					 internal_vlan_id, untagged);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 groups\n", err);
-		goto err_out;
-	}
-
-	err = rocker_port_vlan_flood_group(rocker_port, trans, flags,
-					   internal_vlan_id);
-	if (err) {
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN l2 flood group\n", err);
-		goto err_out;
-	}
-
-	err = rocker_flow_tbl_vlan(rocker_port, trans, flags,
-				   in_pport, vlan_id, vlan_id_mask,
-				   goto_tbl, untagged, internal_vlan_id);
-	if (err)
-		netdev_err(rocker_port->dev,
-			   "Error (%d) port VLAN table\n", err);
-
-err_out:
-	if (switchdev_trans_ph_prepare(trans))
-		change_bit(ntohs(internal_vlan_id), rocker_port->vlan_bitmap);
-
-	return err;
-}
-
-static int rocker_port_ig_tbl(struct rocker_port *rocker_port,
-			      struct switchdev_trans *trans, int flags)
-{
-	enum rocker_of_dpa_table_id goto_tbl;
-	u32 in_pport;
-	u32 in_pport_mask;
-	int err;
-
-	/* Normal Ethernet Frames.  Matches pkts from any local physical
-	 * ports.  Goto VLAN tbl.
-	 */
-
-	in_pport = 0;
-	in_pport_mask = 0xffff0000;
-	goto_tbl = ROCKER_OF_DPA_TABLE_ID_VLAN;
-
-	err = rocker_flow_tbl_ig_port(rocker_port, trans, flags,
-				      in_pport, in_pport_mask,
-				      goto_tbl);
-	if (err)
-		netdev_err(rocker_port->dev,
-			   "Error (%d) ingress port table entry\n", err);
-
-	return err;
-}
-
-struct rocker_fdb_learn_work {
-	struct work_struct work;
-	struct rocker_port *rocker_port;
-	struct switchdev_trans *trans;
-	int flags;
-	u8 addr[ETH_ALEN];
-	u16 vid;
-};
-
-static void rocker_port_fdb_learn_work(struct work_struct *work)
-{
-	const struct rocker_fdb_learn_work *lw =
-		container_of(work, struct rocker_fdb_learn_work, work);
-	bool removing = (lw->flags & ROCKER_OP_FLAG_REMOVE);
-	bool learned = (lw->flags & ROCKER_OP_FLAG_LEARNED);
-	struct switchdev_notifier_fdb_info info;
-
-	info.addr = lw->addr;
-	info.vid = lw->vid;
-
-	rtnl_lock();
-	if (learned && removing)
-		call_switchdev_notifiers(SWITCHDEV_FDB_DEL,
-					 lw->rocker_port->dev, &info.info);
-	else if (learned && !removing)
-		call_switchdev_notifiers(SWITCHDEV_FDB_ADD,
-					 lw->rocker_port->dev, &info.info);
-	rtnl_unlock();
-
-	rocker_kfree(lw->trans, work);
-}
-
-static int rocker_port_fdb_learn(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags,
-				 const u8 *addr, __be16 vlan_id)
-{
-	struct rocker_fdb_learn_work *lw;
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 out_pport = rocker_port->pport;
-	u32 tunnel_id = 0;
-	u32 group_id = ROCKER_GROUP_NONE;
-	bool syncing = !!(rocker_port->brport_flags & BR_LEARNING_SYNC);
-	bool copy_to_cpu = false;
-	int err;
-
-	if (rocker_port_is_bridged(rocker_port))
-		group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
-
-	if (!(flags & ROCKER_OP_FLAG_REFRESH)) {
-		err = rocker_flow_tbl_bridge(rocker_port, trans, flags, addr,
-					     NULL, vlan_id, tunnel_id, goto_tbl,
-					     group_id, copy_to_cpu);
-		if (err)
-			return err;
-	}
-
-	if (!syncing)
-		return 0;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return 0;
-
-	lw = rocker_kzalloc(trans, flags, sizeof(*lw));
-	if (!lw)
-		return -ENOMEM;
-
-	INIT_WORK(&lw->work, rocker_port_fdb_learn_work);
-
-	lw->rocker_port = rocker_port;
-	lw->trans = trans;
-	lw->flags = flags;
-	ether_addr_copy(lw->addr, addr);
-	lw->vid = rocker_port_vlan_to_vid(rocker_port, vlan_id);
-
-	if (switchdev_trans_ph_prepare(trans))
-		rocker_kfree(trans, lw);
-	else
-		schedule_work(&lw->work);
-
-	return 0;
-}
-
-static struct rocker_fdb_tbl_entry *
-rocker_fdb_tbl_find(const struct rocker *rocker,
-		    const struct rocker_fdb_tbl_entry *match)
-{
-	struct rocker_fdb_tbl_entry *found;
-
-	hash_for_each_possible(rocker->fdb_tbl, found, entry, match->key_crc32)
-		if (memcmp(&found->key, &match->key, sizeof(found->key)) == 0)
-			return found;
-
-	return NULL;
-}
-
-static int rocker_port_fdb(struct rocker_port *rocker_port,
-			   struct switchdev_trans *trans,
-			   const unsigned char *addr,
-			   __be16 vlan_id, int flags)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_fdb_tbl_entry *fdb;
-	struct rocker_fdb_tbl_entry *found;
-	bool removing = (flags & ROCKER_OP_FLAG_REMOVE);
-	unsigned long lock_flags;
-
-	fdb = rocker_kzalloc(trans, flags, sizeof(*fdb));
-	if (!fdb)
-		return -ENOMEM;
-
-	fdb->learned = (flags & ROCKER_OP_FLAG_LEARNED);
-	fdb->touched = jiffies;
-	fdb->key.rocker_port = rocker_port;
-	ether_addr_copy(fdb->key.addr, addr);
-	fdb->key.vlan_id = vlan_id;
-	fdb->key_crc32 = crc32(~0, &fdb->key, sizeof(fdb->key));
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-
-	found = rocker_fdb_tbl_find(rocker, fdb);
-
-	if (found) {
-		found->touched = jiffies;
-		if (removing) {
-			rocker_kfree(trans, fdb);
-			if (!switchdev_trans_ph_prepare(trans))
-				hash_del(&found->entry);
-		}
-	} else if (!removing) {
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_add(rocker->fdb_tbl, &fdb->entry,
-				 fdb->key_crc32);
-	}
-
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	/* Check if adding and already exists, or removing and can't find */
-	if (!found != !removing) {
-		rocker_kfree(trans, fdb);
-		if (!found && removing)
-			return 0;
-		/* Refreshing existing to update aging timers */
-		flags |= ROCKER_OP_FLAG_REFRESH;
-	}
-
-	return rocker_port_fdb_learn(rocker_port, trans, flags, addr, vlan_id);
-}
-
-static int rocker_port_fdb_flush(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans, int flags)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_fdb_tbl_entry *found;
-	unsigned long lock_flags;
-	struct hlist_node *tmp;
-	int bkt;
-	int err = 0;
-
-	if (rocker_port->stp_state == BR_STATE_LEARNING ||
-	    rocker_port->stp_state == BR_STATE_FORWARDING)
-		return 0;
-
-	flags |= ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE;
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) {
-		if (found->key.rocker_port != rocker_port)
-			continue;
-		if (!found->learned)
-			continue;
-		err = rocker_port_fdb_learn(rocker_port, trans, flags,
-					    found->key.addr,
-					    found->key.vlan_id);
-		if (err)
-			goto err_out;
-		if (!switchdev_trans_ph_prepare(trans))
-			hash_del(&found->entry);
-	}
-
-err_out:
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	return err;
-}
-
-static void rocker_fdb_cleanup(unsigned long data)
-{
-	struct rocker *rocker = (struct rocker *)data;
-	struct rocker_port *rocker_port;
-	struct rocker_fdb_tbl_entry *entry;
-	struct hlist_node *tmp;
-	unsigned long next_timer = jiffies + BR_MIN_AGEING_TIME;
-	unsigned long expires;
-	unsigned long lock_flags;
-	int flags = ROCKER_OP_FLAG_NOWAIT | ROCKER_OP_FLAG_REMOVE |
-		    ROCKER_OP_FLAG_LEARNED;
-	int bkt;
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, entry, entry) {
-		if (!entry->learned)
-			continue;
-		rocker_port = entry->key.rocker_port;
-		expires = entry->touched + rocker_port->ageing_time;
-		if (time_before_eq(expires, jiffies)) {
-			rocker_port_fdb_learn(rocker_port, NULL,
-					      flags, entry->key.addr,
-					      entry->key.vlan_id);
-			hash_del(&entry->entry);
-		} else if (time_before(expires, next_timer)) {
-			next_timer = expires;
-		}
-	}
-
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	mod_timer(&rocker->fdb_cleanup_timer, round_jiffies_up(next_timer));
-}
-
-static int rocker_port_router_mac(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags,
-				  __be16 vlan_id)
-{
-	u32 in_pport_mask = 0xffffffff;
-	__be16 eth_type;
-	const u8 *dst_mac_mask = ff_mac;
-	__be16 vlan_id_mask = htons(0xffff);
-	bool copy_to_cpu = false;
-	int err;
-
-	if (ntohs(vlan_id) == 0)
-		vlan_id = rocker_port->internal_vlan_id;
-
-	eth_type = htons(ETH_P_IP);
-	err = rocker_flow_tbl_term_mac(rocker_port, trans,
-				       rocker_port->pport, in_pport_mask,
-				       eth_type, rocker_port->dev->dev_addr,
-				       dst_mac_mask, vlan_id, vlan_id_mask,
-				       copy_to_cpu, flags);
-	if (err)
-		return err;
-
-	eth_type = htons(ETH_P_IPV6);
-	err = rocker_flow_tbl_term_mac(rocker_port, trans,
-				       rocker_port->pport, in_pport_mask,
-				       eth_type, rocker_port->dev->dev_addr,
-				       dst_mac_mask, vlan_id, vlan_id_mask,
-				       copy_to_cpu, flags);
-
-	return err;
-}
-
-static int rocker_port_fwding(struct rocker_port *rocker_port,
-			      struct switchdev_trans *trans, int flags)
-{
-	bool pop_vlan;
-	u32 out_pport;
-	__be16 vlan_id;
-	u16 vid;
-	int err;
-
-	/* Port will be forwarding-enabled if its STP state is LEARNING
-	 * or FORWARDING.  Traffic from CPU can still egress, regardless of
-	 * port STP state.  Use L2 interface group on port VLANs as a way
-	 * to toggle port forwarding: if forwarding is disabled, L2
-	 * interface group will not exist.
-	 */
-
-	if (rocker_port->stp_state != BR_STATE_LEARNING &&
-	    rocker_port->stp_state != BR_STATE_FORWARDING)
-		flags |= ROCKER_OP_FLAG_REMOVE;
-
-	out_pport = rocker_port->pport;
-	for (vid = 1; vid < VLAN_N_VID; vid++) {
-		if (!test_bit(vid, rocker_port->vlan_bitmap))
-			continue;
-		vlan_id = htons(vid);
-		pop_vlan = rocker_vlan_id_is_internal(vlan_id);
-		err = rocker_group_l2_interface(rocker_port, trans, flags,
-						vlan_id, out_pport, pop_vlan);
-		if (err) {
-			netdev_err(rocker_port->dev,
-				   "Error (%d) port VLAN l2 group for pport %d\n",
-				   err, out_pport);
-			return err;
-		}
-	}
-
-	return 0;
-}
-
-static int rocker_port_stp_update(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags,
-				  u8 state)
-{
-	bool want[ROCKER_CTRL_MAX] = { 0, };
-	bool prev_ctrls[ROCKER_CTRL_MAX];
-	u8 uninitialized_var(prev_state);
-	int err;
-	int i;
-
-	if (switchdev_trans_ph_prepare(trans)) {
-		memcpy(prev_ctrls, rocker_port->ctrls, sizeof(prev_ctrls));
-		prev_state = rocker_port->stp_state;
-	}
-
-	if (rocker_port->stp_state == state)
-		return 0;
-
-	rocker_port->stp_state = state;
-
-	switch (state) {
-	case BR_STATE_DISABLED:
-		/* port is completely disabled */
-		break;
-	case BR_STATE_LISTENING:
-	case BR_STATE_BLOCKING:
-		want[ROCKER_CTRL_LINK_LOCAL_MCAST] = true;
-		break;
-	case BR_STATE_LEARNING:
-	case BR_STATE_FORWARDING:
-		if (!rocker_port_is_ovsed(rocker_port))
-			want[ROCKER_CTRL_LINK_LOCAL_MCAST] = true;
-		want[ROCKER_CTRL_IPV4_MCAST] = true;
-		want[ROCKER_CTRL_IPV6_MCAST] = true;
-		if (rocker_port_is_bridged(rocker_port))
-			want[ROCKER_CTRL_DFLT_BRIDGING] = true;
-		else if (rocker_port_is_ovsed(rocker_port))
-			want[ROCKER_CTRL_DFLT_OVS] = true;
-		else
-			want[ROCKER_CTRL_LOCAL_ARP] = true;
-		break;
-	}
-
-	for (i = 0; i < ROCKER_CTRL_MAX; i++) {
-		if (want[i] != rocker_port->ctrls[i]) {
-			int ctrl_flags = flags |
-					 (want[i] ? 0 : ROCKER_OP_FLAG_REMOVE);
-			err = rocker_port_ctrl(rocker_port, trans, ctrl_flags,
-					       &rocker_ctrls[i]);
-			if (err)
-				goto err_out;
-			rocker_port->ctrls[i] = want[i];
-		}
-	}
-
-	err = rocker_port_fdb_flush(rocker_port, trans, flags);
-	if (err)
-		goto err_out;
-
-	err = rocker_port_fwding(rocker_port, trans, flags);
-
-err_out:
-	if (switchdev_trans_ph_prepare(trans)) {
-		memcpy(rocker_port->ctrls, prev_ctrls, sizeof(prev_ctrls));
-		rocker_port->stp_state = prev_state;
-	}
-
-	return err;
-}
-
-static int rocker_port_fwd_enable(struct rocker_port *rocker_port,
-				  struct switchdev_trans *trans, int flags)
-{
-	if (rocker_port_is_bridged(rocker_port))
-		/* bridge STP will enable port */
-		return 0;
-
-	/* port is not bridged, so simulate going to FORWARDING state */
-	return rocker_port_stp_update(rocker_port, trans, flags,
-				      BR_STATE_FORWARDING);
-}
-
-static int rocker_port_fwd_disable(struct rocker_port *rocker_port,
-				   struct switchdev_trans *trans, int flags)
-{
-	if (rocker_port_is_bridged(rocker_port))
-		/* bridge STP will disable port */
-		return 0;
-
-	/* port is not bridged, so simulate going to DISABLED state */
-	return rocker_port_stp_update(rocker_port, trans, flags,
-				      BR_STATE_DISABLED);
-}
-
-static struct rocker_internal_vlan_tbl_entry *
-rocker_internal_vlan_tbl_find(const struct rocker *rocker, int ifindex)
-{
-	struct rocker_internal_vlan_tbl_entry *found;
-
-	hash_for_each_possible(rocker->internal_vlan_tbl, found,
-			       entry, ifindex) {
-		if (found->ifindex == ifindex)
-			return found;
-	}
-
-	return NULL;
-}
-
-static __be16 rocker_port_internal_vlan_id_get(struct rocker_port *rocker_port,
-					       int ifindex)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_internal_vlan_tbl_entry *entry;
-	struct rocker_internal_vlan_tbl_entry *found;
-	unsigned long lock_flags;
-	int i;
-
-	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
-	if (!entry)
-		return 0;
-
-	entry->ifindex = ifindex;
-
-	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, lock_flags);
-
-	found = rocker_internal_vlan_tbl_find(rocker, ifindex);
-	if (found) {
-		kfree(entry);
-		goto found;
-	}
-
-	found = entry;
-	hash_add(rocker->internal_vlan_tbl, &found->entry, found->ifindex);
-
-	for (i = 0; i < ROCKER_N_INTERNAL_VLANS; i++) {
-		if (test_and_set_bit(i, rocker->internal_vlan_bitmap))
-			continue;
-		found->vlan_id = htons(ROCKER_INTERNAL_VLAN_ID_BASE + i);
-		goto found;
-	}
-
-	netdev_err(rocker_port->dev, "Out of internal VLAN IDs\n");
-
-found:
-	found->ref_count++;
-	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags);
-
-	return found->vlan_id;
-}
-
-static void
-rocker_port_internal_vlan_id_put(const struct rocker_port *rocker_port,
-				 int ifindex)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_internal_vlan_tbl_entry *found;
-	unsigned long lock_flags;
-	unsigned long bit;
-
-	spin_lock_irqsave(&rocker->internal_vlan_tbl_lock, lock_flags);
-
-	found = rocker_internal_vlan_tbl_find(rocker, ifindex);
-	if (!found) {
-		netdev_err(rocker_port->dev,
-			   "ifindex (%d) not found in internal VLAN tbl\n",
-			   ifindex);
-		goto not_found;
-	}
-
-	if (--found->ref_count <= 0) {
-		bit = ntohs(found->vlan_id) - ROCKER_INTERNAL_VLAN_ID_BASE;
-		clear_bit(bit, rocker->internal_vlan_bitmap);
-		hash_del(&found->entry);
-		kfree(found);
-	}
-
-not_found:
-	spin_unlock_irqrestore(&rocker->internal_vlan_tbl_lock, lock_flags);
-}
-
-static int rocker_port_fib_ipv4(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans, __be32 dst,
-				int dst_len, const struct fib_info *fi,
-				u32 tb_id, int flags)
-{
-	const struct fib_nh *nh;
-	__be16 eth_type = htons(ETH_P_IP);
-	__be32 dst_mask = inet_make_mask(dst_len);
-	__be16 internal_vlan_id = rocker_port->internal_vlan_id;
-	u32 priority = fi->fib_priority;
-	enum rocker_of_dpa_table_id goto_tbl =
-		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
-	u32 group_id;
-	bool nh_on_port;
-	bool has_gw;
-	u32 index;
-	int err;
-
-	/* XXX support ECMP */
-
-	nh = fi->fib_nh;
-	nh_on_port = (fi->fib_dev == rocker_port->dev);
-	has_gw = !!nh->nh_gw;
-
-	if (has_gw && nh_on_port) {
-		err = rocker_port_ipv4_nh(rocker_port, trans, flags,
-					  nh->nh_gw, &index);
-		if (err)
-			return err;
-
-		group_id = ROCKER_GROUP_L3_UNICAST(index);
-	} else {
-		/* Send to CPU for processing */
-		group_id = ROCKER_GROUP_L2_INTERFACE(internal_vlan_id, 0);
-	}
-
-	err = rocker_flow_tbl_ucast4_routing(rocker_port, trans, eth_type, dst,
-					     dst_mask, priority, goto_tbl,
-					     group_id, flags);
-	if (err)
-		netdev_err(rocker_port->dev, "Error (%d) IPv4 route %pI4\n",
-			   err, &dst);
-
-	return err;
-}
-
-/*****************
- * Net device ops
- *****************/
-
-static int rocker_port_open(struct net_device *dev)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int err;
-
-	err = rocker_port_dma_rings_init(rocker_port);
-	if (err)
-		return err;
-
-	err = request_irq(rocker_msix_tx_vector(rocker_port),
-			  rocker_tx_irq_handler, 0,
-			  rocker_driver_name, rocker_port);
-	if (err) {
-		netdev_err(rocker_port->dev, "cannot assign tx irq\n");
-		goto err_request_tx_irq;
-	}
-
-	err = request_irq(rocker_msix_rx_vector(rocker_port),
-			  rocker_rx_irq_handler, 0,
-			  rocker_driver_name, rocker_port);
-	if (err) {
-		netdev_err(rocker_port->dev, "cannot assign rx irq\n");
-		goto err_request_rx_irq;
-	}
-
-	err = rocker_world_port_open(rocker_port);
-	if (err) {
-		netdev_err(rocker_port->dev, "cannot open port in world\n");
-		goto err_world_port_open;
-	}
-
-	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
-	if (err)
-		goto err_fwd_enable;
-
-	napi_enable(&rocker_port->napi_tx);
-	napi_enable(&rocker_port->napi_rx);
-	if (!dev->proto_down)
-		rocker_port_set_enable(rocker_port, true);
-	netif_start_queue(dev);
-	return 0;
-
-err_fwd_enable:
-err_world_port_open:
-	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
-err_request_rx_irq:
-	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
-err_request_tx_irq:
-	rocker_port_dma_rings_fini(rocker_port);
-	return err;
-}
-
-static int rocker_port_stop(struct net_device *dev)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-
-	netif_stop_queue(dev);
-	rocker_port_set_enable(rocker_port, false);
-	napi_disable(&rocker_port->napi_rx);
-	napi_disable(&rocker_port->napi_tx);
-	rocker_world_port_stop(rocker_port);
-	rocker_port_fwd_disable(rocker_port, NULL,
-				ROCKER_OP_FLAG_NOWAIT);
-	free_irq(rocker_msix_rx_vector(rocker_port), rocker_port);
-	free_irq(rocker_msix_tx_vector(rocker_port), rocker_port);
-	rocker_port_dma_rings_fini(rocker_port);
-
-	return 0;
-}
-
-static void rocker_tx_desc_frags_unmap(const struct rocker_port *rocker_port,
-				       const struct rocker_desc_info *desc_info)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct pci_dev *pdev = rocker->pdev;
-	const struct rocker_tlv *attrs[ROCKER_TLV_TX_MAX + 1];
-	struct rocker_tlv *attr;
-	int rem;
-
-	rocker_tlv_parse_desc(attrs, ROCKER_TLV_TX_MAX, desc_info);
-	if (!attrs[ROCKER_TLV_TX_FRAGS])
-		return;
-	rocker_tlv_for_each_nested(attr, attrs[ROCKER_TLV_TX_FRAGS], rem) {
-		const struct rocker_tlv *frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_MAX + 1];
-		dma_addr_t dma_handle;
-		size_t len;
-
-		if (rocker_tlv_type(attr) != ROCKER_TLV_TX_FRAG)
-			continue;
-		rocker_tlv_parse_nested(frag_attrs, ROCKER_TLV_TX_FRAG_ATTR_MAX,
-					attr);
-		if (!frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR] ||
-		    !frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN])
-			continue;
-		dma_handle = rocker_tlv_get_u64(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR]);
-		len = rocker_tlv_get_u16(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN]);
-		pci_unmap_single(pdev, dma_handle, len, DMA_TO_DEVICE);
-	}
-}
-
-static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port,
-				       struct rocker_desc_info *desc_info,
-				       char *buf, size_t buf_len)
-{
-	const struct rocker *rocker = rocker_port->rocker;
-	struct pci_dev *pdev = rocker->pdev;
-	dma_addr_t dma_handle;
-	struct rocker_tlv *frag;
-
-	dma_handle = pci_map_single(pdev, buf, buf_len, DMA_TO_DEVICE);
-	if (unlikely(pci_dma_mapping_error(pdev, dma_handle))) {
-		if (net_ratelimit())
-			netdev_err(rocker_port->dev, "failed to dma map tx frag\n");
-		return -EIO;
-	}
-	frag = rocker_tlv_nest_start(desc_info, ROCKER_TLV_TX_FRAG);
-	if (!frag)
-		goto unmap_frag;
-	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_TX_FRAG_ATTR_ADDR,
-			       dma_handle))
-		goto nest_cancel;
-	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_TX_FRAG_ATTR_LEN,
-			       buf_len))
-		goto nest_cancel;
-	rocker_tlv_nest_end(desc_info, frag);
-	return 0;
-
-nest_cancel:
-	rocker_tlv_nest_cancel(desc_info, frag);
-unmap_frag:
-	pci_unmap_single(pdev, dma_handle, buf_len, DMA_TO_DEVICE);
-	return -EMSGSIZE;
-}
-
-static netdev_tx_t rocker_port_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_desc_info *desc_info;
-	struct rocker_tlv *frags;
-	int i;
-	int err;
+	struct rocker_port *rocker_port = netdev_priv(dev);
+	struct rocker *rocker = rocker_port->rocker;
+	struct rocker_desc_info *desc_info;
+	struct rocker_tlv *frags;
+	int i;
+	int err;
 
 	desc_info = rocker_desc_head_get(&rocker_port->tx_ring);
 	if (unlikely(!desc_info)) {
@@ -4439,12 +1999,8 @@ static int rocker_port_change_proto_down(struct net_device *dev,
 static void rocker_port_neigh_destroy(struct neighbour *n)
 {
 	struct rocker_port *rocker_port = netdev_priv(n->dev);
-	int flags = ROCKER_OP_FLAG_REMOVE | ROCKER_OP_FLAG_NOWAIT;
-	__be32 ip_addr = *(__be32 *)n->primary_key;
 	int err;
 
-	rocker_port_ipv4_neigh(rocker_port, NULL,
-			       flags, ip_addr, n->ha);
 	err = rocker_world_port_neigh_destroy(rocker_port, n);
 	if (err)
 		netdev_warn(rocker_port->dev, "failed to handle neigh destroy (err %d)\n",
@@ -4485,7 +2041,6 @@ static int rocker_port_attr_get(struct net_device *dev,
 		memcpy(&attr->u.ppid.id, &rocker->hw.id, attr->u.ppid.id_len);
 		break;
 	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
-		attr->u.brport_flags = rocker_port->brport_flags;
 		err = rocker_world_port_attr_bridge_flags_get(rocker_port,
 							      &attr->u.brport_flags);
 		break;
@@ -4496,38 +2051,6 @@ static int rocker_port_attr_get(struct net_device *dev,
 	return err;
 }
 
-static int rocker_port_brport_flags_set(struct rocker_port *rocker_port,
-					struct switchdev_trans *trans,
-					unsigned long brport_flags)
-{
-	unsigned long orig_flags;
-	int err = 0;
-
-	orig_flags = rocker_port->brport_flags;
-	rocker_port->brport_flags = brport_flags;
-	if ((orig_flags ^ rocker_port->brport_flags) & BR_LEARNING &&
-	    !switchdev_trans_ph_prepare(trans))
-		err = rocker_port_set_learning(rocker_port,
-					       !!(rocker_port->brport_flags & BR_LEARNING));
-
-	if (switchdev_trans_ph_prepare(trans))
-		rocker_port->brport_flags = orig_flags;
-
-	return err;
-}
-
-static int rocker_port_bridge_ageing_time(struct rocker_port *rocker_port,
-					  struct switchdev_trans *trans,
-					  u32 ageing_time)
-{
-	if (!switchdev_trans_ph_prepare(trans)) {
-		rocker_port->ageing_time = clock_t_to_jiffies(ageing_time);
-		mod_timer(&rocker_port->rocker->fdb_cleanup_timer, jiffies);
-	}
-
-	return 0;
-}
-
 static int rocker_port_attr_set(struct net_device *dev,
 				const struct switchdev_attr *attr,
 				struct switchdev_trans *trans)
@@ -4537,28 +2060,16 @@ static int rocker_port_attr_set(struct net_device *dev,
 
 	switch (attr->id) {
 	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
-		err = rocker_port_stp_update(rocker_port, trans, 0,
-					     attr->u.stp_state);
-		if (err)
-			break;
 		err = rocker_world_port_attr_stp_state_set(rocker_port,
 							   attr->u.stp_state,
 							   trans);
 		break;
 	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
-		err = rocker_port_brport_flags_set(rocker_port, trans,
-						   attr->u.brport_flags);
-		if (err)
-			break;
 		err = rocker_world_port_attr_bridge_flags_set(rocker_port,
 							      attr->u.brport_flags,
 							      trans);
 		break;
 	case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME:
-		err = rocker_port_bridge_ageing_time(rocker_port, trans,
-						     attr->u.ageing_time);
-		if (err)
-			break;
 		err = rocker_world_port_attr_bridge_ageing_time_set(rocker_port,
 								    attr->u.ageing_time,
 								    trans);
@@ -4571,90 +2082,25 @@ static int rocker_port_attr_set(struct net_device *dev,
 	return err;
 }
 
-static int rocker_port_vlan_add(struct rocker_port *rocker_port,
-				struct switchdev_trans *trans,
-				u16 vid, u16 flags)
-{
-	int err;
-
-	/* XXX deal with flags for PVID and untagged */
-
-	err = rocker_port_vlan(rocker_port, trans, 0, vid);
-	if (err)
-		return err;
-
-	err = rocker_port_router_mac(rocker_port, trans, 0, htons(vid));
-	if (err)
-		rocker_port_vlan(rocker_port, trans,
-				 ROCKER_OP_FLAG_REMOVE, vid);
-
-	return err;
-}
-
-static int rocker_port_vlans_add(struct rocker_port *rocker_port,
-				 struct switchdev_trans *trans,
-				 const struct switchdev_obj_port_vlan *vlan)
-{
-	u16 vid;
-	int err;
-
-	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
-		err = rocker_port_vlan_add(rocker_port, trans,
-					   vid, vlan->flags);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
-static int rocker_port_fdb_add(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans,
-			       const struct switchdev_obj_port_fdb *fdb)
-{
-	__be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL);
-	int flags = 0;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return -EINVAL;
-
-	return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags);
-}
-
 static int rocker_port_obj_add(struct net_device *dev,
 			       const struct switchdev_obj *obj,
 			       struct switchdev_trans *trans)
 {
 	struct rocker_port *rocker_port = netdev_priv(dev);
-	const struct switchdev_obj_ipv4_fib *fib4;
 	int err = 0;
 
 	switch (obj->id) {
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
-		err = rocker_port_vlans_add(rocker_port, trans,
-					    SWITCHDEV_OBJ_PORT_VLAN(obj));
-		if (err)
-			break;
 		err = rocker_world_port_obj_vlan_add(rocker_port,
 						     SWITCHDEV_OBJ_PORT_VLAN(obj),
 						     trans);
 		break;
 	case SWITCHDEV_OBJ_ID_IPV4_FIB:
-		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
-		err = rocker_port_fib_ipv4(rocker_port, trans,
-					   htonl(fib4->dst), fib4->dst_len,
-					   &fib4->fi, fib4->tb_id, 0);
-		if (err)
-			break;
 		err = rocker_world_port_obj_fib4_add(rocker_port,
 						     SWITCHDEV_OBJ_IPV4_FIB(obj),
 						     trans);
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
-		err = rocker_port_fdb_add(rocker_port, trans,
-					  SWITCHDEV_OBJ_PORT_FDB(obj));
-		if (err)
-			break;
 		err = rocker_world_port_obj_fdb_add(rocker_port,
 						    SWITCHDEV_OBJ_PORT_FDB(obj),
 						    trans);
@@ -4667,80 +2113,22 @@ static int rocker_port_obj_add(struct net_device *dev,
 	return err;
 }
 
-static int rocker_port_vlan_del(struct rocker_port *rocker_port,
-				u16 vid, u16 flags)
-{
-	int err;
-
-	err = rocker_port_router_mac(rocker_port, NULL,
-				     ROCKER_OP_FLAG_REMOVE, htons(vid));
-	if (err)
-		return err;
-
-	return rocker_port_vlan(rocker_port, NULL,
-				ROCKER_OP_FLAG_REMOVE, vid);
-}
-
-static int rocker_port_vlans_del(struct rocker_port *rocker_port,
-				 const struct switchdev_obj_port_vlan *vlan)
-{
-	u16 vid;
-	int err;
-
-	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
-		err = rocker_port_vlan_del(rocker_port, vid, vlan->flags);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
-static int rocker_port_fdb_del(struct rocker_port *rocker_port,
-			       struct switchdev_trans *trans,
-			       const struct switchdev_obj_port_fdb *fdb)
-{
-	__be16 vlan_id = rocker_port_vid_to_vlan(rocker_port, fdb->vid, NULL);
-	int flags = ROCKER_OP_FLAG_REMOVE;
-
-	if (!rocker_port_is_bridged(rocker_port))
-		return -EINVAL;
-
-	return rocker_port_fdb(rocker_port, trans, fdb->addr, vlan_id, flags);
-}
-
 static int rocker_port_obj_del(struct net_device *dev,
 			       const struct switchdev_obj *obj)
 {
 	struct rocker_port *rocker_port = netdev_priv(dev);
-	const struct switchdev_obj_ipv4_fib *fib4;
 	int err = 0;
 
 	switch (obj->id) {
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
-		err = rocker_port_vlans_del(rocker_port,
-					    SWITCHDEV_OBJ_PORT_VLAN(obj));
-		if (err)
-			break;
 		err = rocker_world_port_obj_vlan_del(rocker_port,
 						     SWITCHDEV_OBJ_PORT_VLAN(obj));
 		break;
 	case SWITCHDEV_OBJ_ID_IPV4_FIB:
-		fib4 = SWITCHDEV_OBJ_IPV4_FIB(obj);
-		err = rocker_port_fib_ipv4(rocker_port, NULL,
-					   htonl(fib4->dst), fib4->dst_len,
-					   &fib4->fi, fib4->tb_id,
-					   ROCKER_OP_FLAG_REMOVE);
-		if (err)
-			break;
 		err = rocker_world_port_obj_fib4_del(rocker_port,
 						     SWITCHDEV_OBJ_IPV4_FIB(obj));
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
-		err = rocker_port_fdb_del(rocker_port, NULL,
-					  SWITCHDEV_OBJ_PORT_FDB(obj));
-		if (err)
-			break;
 		err = rocker_world_port_obj_fdb_del(rocker_port,
 						    SWITCHDEV_OBJ_PORT_FDB(obj));
 		break;
@@ -4752,57 +2140,6 @@ static int rocker_port_obj_del(struct net_device *dev,
 	return err;
 }
 
-static int rocker_port_fdb_dump(const struct rocker_port *rocker_port,
-				struct switchdev_obj_port_fdb *fdb,
-				switchdev_obj_dump_cb_t *cb)
-{
-	struct rocker *rocker = rocker_port->rocker;
-	struct rocker_fdb_tbl_entry *found;
-	struct hlist_node *tmp;
-	unsigned long lock_flags;
-	int bkt;
-	int err = 0;
-
-	spin_lock_irqsave(&rocker->fdb_tbl_lock, lock_flags);
-	hash_for_each_safe(rocker->fdb_tbl, bkt, tmp, found, entry) {
-		if (found->key.rocker_port != rocker_port)
-			continue;
-		ether_addr_copy(fdb->addr, found->key.addr);
-		fdb->ndm_state = NUD_REACHABLE;
-		fdb->vid = rocker_port_vlan_to_vid(rocker_port,
-						   found->key.vlan_id);
-		err = cb(&fdb->obj);
-		if (err)
-			break;
-	}
-	spin_unlock_irqrestore(&rocker->fdb_tbl_lock, lock_flags);
-
-	return err;
-}
-
-static int rocker_port_vlan_dump(const struct rocker_port *rocker_port,
-				 struct switchdev_obj_port_vlan *vlan,
-				 switchdev_obj_dump_cb_t *cb)
-{
-	u16 vid;
-	int err = 0;
-
-	for (vid = 1; vid < VLAN_N_VID; vid++) {
-		if (!test_bit(vid, rocker_port->vlan_bitmap))
-			continue;
-		vlan->flags = 0;
-		if (rocker_vlan_id_is_internal(htons(vid)))
-			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
-		vlan->vid_begin = vid;
-		vlan->vid_end = vid;
-		err = cb(&vlan->obj);
-		if (err)
-			break;
-	}
-
-	return err;
-}
-
 static int rocker_port_obj_dump(struct net_device *dev,
 				struct switchdev_obj *obj,
 				switchdev_obj_dump_cb_t *cb)
@@ -4812,19 +2149,11 @@ static int rocker_port_obj_dump(struct net_device *dev,
 
 	switch (obj->id) {
 	case SWITCHDEV_OBJ_ID_PORT_FDB:
-		err = rocker_port_fdb_dump(rocker_port,
-					   SWITCHDEV_OBJ_PORT_FDB(obj), cb);
-		if (err)
-			break;
 		err = rocker_world_port_obj_fdb_dump(rocker_port,
 						     SWITCHDEV_OBJ_PORT_FDB(obj),
 						     cb);
 		break;
 	case SWITCHDEV_OBJ_ID_PORT_VLAN:
-		err = rocker_port_vlan_dump(rocker_port,
-					    SWITCHDEV_OBJ_PORT_VLAN(obj), cb);
-		if (err)
-			break;
 		err = rocker_world_port_obj_vlan_dump(rocker_port,
 						      SWITCHDEV_OBJ_PORT_VLAN(obj),
 						      cb);
@@ -5162,7 +2491,6 @@ static void rocker_remove_ports(struct rocker *rocker)
 		rocker_port = rocker->ports[i];
 		if (!rocker_port)
 			continue;
-		rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
 		rocker_world_port_fini(rocker_port);
 		unregister_netdev(rocker_port->dev);
 		rocker_world_port_post_fini(rocker_port);
@@ -5191,7 +2519,6 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 	const struct pci_dev *pdev = rocker->pdev;
 	struct rocker_port *rocker_port;
 	struct net_device *dev;
-	u16 untagged_vid = 0;
 	int err;
 
 	dev = alloc_etherdev(sizeof(struct rocker_port));
@@ -5202,8 +2529,6 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 	rocker_port->rocker = rocker;
 	rocker_port->port_number = port_number;
 	rocker_port->pport = port_number + 1;
-	rocker_port->brport_flags = BR_LEARNING | BR_LEARNING_SYNC;
-	rocker_port->ageing_time = BR_DEFAULT_AGEING_TIME;
 
 	err = rocker_world_check_init(rocker_port);
 	if (err) {
@@ -5241,32 +2566,8 @@ static int rocker_probe_port(struct rocker *rocker, unsigned int port_number)
 		goto err_world_port_init;
 	}
 
-	switchdev_port_fwd_mark_set(rocker_port->dev, NULL, false);
-
-	rocker_port_set_learning(rocker_port,
-				 !!(rocker_port->brport_flags & BR_LEARNING));
-
-	err = rocker_port_ig_tbl(rocker_port, NULL, 0);
-	if (err) {
-		netdev_err(rocker_port->dev, "install ig port table failed\n");
-		goto err_port_ig_tbl;
-	}
-
-	rocker_port->internal_vlan_id =
-		rocker_port_internal_vlan_id_get(rocker_port, dev->ifindex);
-
-	err = rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
-	if (err) {
-		netdev_err(rocker_port->dev, "install untagged VLAN failed\n");
-		goto err_untagged_vlan;
-	}
-
 	return 0;
 
-err_untagged_vlan:
-	rocker_port_ig_tbl(rocker_port, NULL, ROCKER_OP_FLAG_REMOVE);
-err_port_ig_tbl:
-	rocker_world_port_fini(rocker_port);
 err_world_port_init:
 	rocker->ports[port_number] = NULL;
 	unregister_netdev(dev);
@@ -5432,16 +2733,6 @@ static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	rocker->hw.id = rocker_read64(rocker, SWITCH_ID);
 
-	err = rocker_init_tbls(rocker);
-	if (err) {
-		dev_err(&pdev->dev, "cannot init rocker tables\n");
-		goto err_init_tbls;
-	}
-
-	setup_timer(&rocker->fdb_cleanup_timer, rocker_fdb_cleanup,
-		    (unsigned long) rocker);
-	mod_timer(&rocker->fdb_cleanup_timer, jiffies);
-
 	err = rocker_probe_ports(rocker);
 	if (err) {
 		dev_err(&pdev->dev, "failed to probe ports\n");
@@ -5454,9 +2745,6 @@ static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	return 0;
 
 err_probe_ports:
-	del_timer_sync(&rocker->fdb_cleanup_timer);
-	rocker_free_tbls(rocker);
-err_init_tbls:
 	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT), rocker);
 err_request_event_irq:
 	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_CMD), rocker);
@@ -5482,8 +2770,6 @@ static void rocker_remove(struct pci_dev *pdev)
 {
 	struct rocker *rocker = pci_get_drvdata(pdev);
 
-	del_timer_sync(&rocker->fdb_cleanup_timer);
-	rocker_free_tbls(rocker);
 	rocker_write32(rocker, CONTROL, ROCKER_CONTROL_RESET);
 	rocker_remove_ports(rocker);
 	free_irq(rocker_msix_vector(rocker, ROCKER_MSIX_VEC_EVENT), rocker);
@@ -5512,100 +2798,6 @@ static bool rocker_port_dev_check(const struct net_device *dev)
 	return dev->netdev_ops == &rocker_port_netdev_ops;
 }
 
-static int rocker_port_bridge_join(struct rocker_port *rocker_port,
-				   struct net_device *bridge)
-{
-	u16 untagged_vid = 0;
-	int err;
-
-	/* Port is joining bridge, so the internal VLAN for the
-	 * port is going to change to the bridge internal VLAN.
-	 * Let's remove untagged VLAN (vid=0) from port and
-	 * re-add once internal VLAN has changed.
-	 */
-
-	err = rocker_port_vlan_del(rocker_port, untagged_vid, 0);
-	if (err)
-		return err;
-
-	rocker_port_internal_vlan_id_put(rocker_port,
-					 rocker_port->dev->ifindex);
-	rocker_port->internal_vlan_id =
-		rocker_port_internal_vlan_id_get(rocker_port, bridge->ifindex);
-
-	rocker_port->bridge_dev = bridge;
-	switchdev_port_fwd_mark_set(rocker_port->dev, bridge, true);
-
-	return rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
-}
-
-static int rocker_port_bridge_leave(struct rocker_port *rocker_port)
-{
-	u16 untagged_vid = 0;
-	int err;
-
-	err = rocker_port_vlan_del(rocker_port, untagged_vid, 0);
-	if (err)
-		return err;
-
-	rocker_port_internal_vlan_id_put(rocker_port,
-					 rocker_port->bridge_dev->ifindex);
-	rocker_port->internal_vlan_id =
-		rocker_port_internal_vlan_id_get(rocker_port,
-						 rocker_port->dev->ifindex);
-
-	switchdev_port_fwd_mark_set(rocker_port->dev, rocker_port->bridge_dev,
-				    false);
-	rocker_port->bridge_dev = NULL;
-
-	err = rocker_port_vlan_add(rocker_port, NULL, untagged_vid, 0);
-	if (err)
-		return err;
-
-	if (rocker_port->dev->flags & IFF_UP)
-		err = rocker_port_fwd_enable(rocker_port, NULL, 0);
-
-	return err;
-}
-
-static int rocker_port_ovs_changed(struct rocker_port *rocker_port,
-				   struct net_device *master)
-{
-	int err;
-
-	rocker_port->bridge_dev = master;
-
-	err = rocker_port_fwd_disable(rocker_port, NULL, 0);
-	if (err)
-		return err;
-	err = rocker_port_fwd_enable(rocker_port, NULL, 0);
-
-	return err;
-}
-
-static int rocker_port_master_linked(struct rocker_port *rocker_port,
-				     struct net_device *master)
-{
-	int err = 0;
-
-	if (netif_is_bridge_master(master))
-		err = rocker_port_bridge_join(rocker_port, master);
-	else if (netif_is_ovs_master(master))
-		err = rocker_port_ovs_changed(rocker_port, master);
-	return err;
-}
-
-static int rocker_port_master_unlinked(struct rocker_port *rocker_port)
-{
-	int err = 0;
-
-	if (rocker_port_is_bridged(rocker_port))
-		err = rocker_port_bridge_leave(rocker_port);
-	else if (rocker_port_is_ovsed(rocker_port))
-		err = rocker_port_ovs_changed(rocker_port, NULL);
-	return err;
-}
-
 static int rocker_netdevice_event(struct notifier_block *unused,
 				  unsigned long event, void *ptr)
 {
@@ -5629,23 +2821,13 @@ static int rocker_netdevice_event(struct notifier_block *unused,
 			if (err)
 				netdev_warn(dev, "failed to reflect master linked (err %d)\n",
 					    err);
-			err = rocker_port_master_linked(rocker_port,
-							info->upper_dev);
-			if (err)
-				netdev_warn(dev, "failed to reflect master linked (err %d)\n",
-					    err);
 		} else {
 			err = rocker_world_port_master_unlinked(rocker_port,
 								info->upper_dev);
 			if (err)
 				netdev_warn(dev, "failed to reflect master unlinked (err %d)\n",
 					    err);
-			err = rocker_port_master_unlinked(rocker_port);
-			if (err)
-				netdev_warn(dev, "failed to reflect master unlinked (err %d)\n",
-					    err);
 		}
-		break;
 	}
 out:
 	return NOTIFY_DONE;
@@ -5659,16 +2841,6 @@ static struct notifier_block rocker_netdevice_nb __read_mostly = {
  * Net event notifier event handler
  ************************************/
 
-static int rocker_neigh_update(struct net_device *dev, struct neighbour *n)
-{
-	struct rocker_port *rocker_port = netdev_priv(dev);
-	int flags = (n->nud_state & NUD_VALID ? 0 : ROCKER_OP_FLAG_REMOVE) |
-		    ROCKER_OP_FLAG_NOWAIT;
-	__be32 ip_addr = *(__be32 *)n->primary_key;
-
-	return rocker_port_ipv4_neigh(rocker_port, NULL, flags, ip_addr, n->ha);
-}
-
 static int rocker_netevent_event(struct notifier_block *unused,
 				 unsigned long event, void *ptr)
 {
@@ -5689,11 +2861,6 @@ static int rocker_netevent_event(struct notifier_block *unused,
 		if (err)
 			netdev_warn(dev, "failed to handle neigh update (err %d)\n",
 				    err);
-		err = rocker_neigh_update(dev, n);
-		if (err)
-			netdev_warn(dev,
-				    "failed to handle neigh update (err %d)\n",
-				    err);
 		break;
 	}
 
diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
index 155dc53..099008a 100644
--- a/drivers/net/ethernet/rocker/rocker_ofdpa.c
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -11,17 +11,2942 @@
  */
 
 #include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/spinlock.h>
+#include <linux/hashtable.h>
+#include <linux/crc32.h>
+#include <linux/netdevice.h>
+#include <linux/inetdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/if_bridge.h>
+#include <net/neighbour.h>
+#include <net/switchdev.h>
+#include <net/ip_fib.h>
+#include <net/arp.h>
+
 #include "rocker.h"
+#include "rocker_tlv.h"
+
+struct ofdpa_flow_tbl_key {
+	u32 priority;
+	enum rocker_of_dpa_table_id tbl_id;
+	union {
+		struct {
+			u32 in_pport;
+			u32 in_pport_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+		} ig_port;
+		struct {
+			u32 in_pport;
+			__be16 vlan_id;
+			__be16 vlan_id_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+			bool untagged;
+			__be16 new_vlan_id;
+		} vlan;
+		struct {
+			u32 in_pport;
+			u32 in_pport_mask;
+			__be16 eth_type;
+			u8 eth_dst[ETH_ALEN];
+			u8 eth_dst_mask[ETH_ALEN];
+			__be16 vlan_id;
+			__be16 vlan_id_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+			bool copy_to_cpu;
+		} term_mac;
+		struct {
+			__be16 eth_type;
+			__be32 dst4;
+			__be32 dst4_mask;
+			enum rocker_of_dpa_table_id goto_tbl;
+			u32 group_id;
+		} ucast_routing;
+		struct {
+			u8 eth_dst[ETH_ALEN];
+			u8 eth_dst_mask[ETH_ALEN];
+			int has_eth_dst;
+			int has_eth_dst_mask;
+			__be16 vlan_id;
+			u32 tunnel_id;
+			enum rocker_of_dpa_table_id goto_tbl;
+			u32 group_id;
+			bool copy_to_cpu;
+		} bridge;
+		struct {
+			u32 in_pport;
+			u32 in_pport_mask;
+			u8 eth_src[ETH_ALEN];
+			u8 eth_src_mask[ETH_ALEN];
+			u8 eth_dst[ETH_ALEN];
+			u8 eth_dst_mask[ETH_ALEN];
+			__be16 eth_type;
+			__be16 vlan_id;
+			__be16 vlan_id_mask;
+			u8 ip_proto;
+			u8 ip_proto_mask;
+			u8 ip_tos;
+			u8 ip_tos_mask;
+			u32 group_id;
+		} acl;
+	};
+};
+
+struct ofdpa_flow_tbl_entry {
+	struct hlist_node entry;
+	u32 cmd;
+	u64 cookie;
+	struct ofdpa_flow_tbl_key key;
+	size_t key_len;
+	u32 key_crc32; /* key */
+};
+
+struct ofdpa_group_tbl_entry {
+	struct hlist_node entry;
+	u32 cmd;
+	u32 group_id; /* key */
+	u16 group_count;
+	u32 *group_ids;
+	union {
+		struct {
+			u8 pop_vlan;
+		} l2_interface;
+		struct {
+			u8 eth_src[ETH_ALEN];
+			u8 eth_dst[ETH_ALEN];
+			__be16 vlan_id;
+			u32 group_id;
+		} l2_rewrite;
+		struct {
+			u8 eth_src[ETH_ALEN];
+			u8 eth_dst[ETH_ALEN];
+			__be16 vlan_id;
+			bool ttl_check;
+			u32 group_id;
+		} l3_unicast;
+	};
+};
+
+struct ofdpa_fdb_tbl_entry {
+	struct hlist_node entry;
+	u32 key_crc32; /* key */
+	bool learned;
+	unsigned long touched;
+	struct ofdpa_fdb_tbl_key {
+		struct ofdpa_port *ofdpa_port;
+		u8 addr[ETH_ALEN];
+		__be16 vlan_id;
+	} key;
+};
+
+struct ofdpa_internal_vlan_tbl_entry {
+	struct hlist_node entry;
+	int ifindex; /* key */
+	u32 ref_count;
+	__be16 vlan_id;
+};
+
+struct ofdpa_neigh_tbl_entry {
+	struct hlist_node entry;
+	__be32 ip_addr; /* key */
+	struct net_device *dev;
+	u32 ref_count;
+	u32 index;
+	u8 eth_dst[ETH_ALEN];
+	bool ttl_check;
+};
+
+enum {
+	OFDPA_CTRL_LINK_LOCAL_MCAST,
+	OFDPA_CTRL_LOCAL_ARP,
+	OFDPA_CTRL_IPV4_MCAST,
+	OFDPA_CTRL_IPV6_MCAST,
+	OFDPA_CTRL_DFLT_BRIDGING,
+	OFDPA_CTRL_DFLT_OVS,
+	OFDPA_CTRL_MAX,
+};
+
+#define OFDPA_INTERNAL_VLAN_ID_BASE	0x0f00
+#define OFDPA_N_INTERNAL_VLANS		255
+#define OFDPA_VLAN_BITMAP_LEN		BITS_TO_LONGS(VLAN_N_VID)
+#define OFDPA_INTERNAL_VLAN_BITMAP_LEN	BITS_TO_LONGS(OFDPA_N_INTERNAL_VLANS)
+#define OFDPA_UNTAGGED_VID 0
 
 struct ofdpa {
+	struct rocker *rocker;
+	DECLARE_HASHTABLE(flow_tbl, 16);
+	spinlock_t flow_tbl_lock;		/* for flow tbl accesses */
+	u64 flow_tbl_next_cookie;
+	DECLARE_HASHTABLE(group_tbl, 16);
+	spinlock_t group_tbl_lock;		/* for group tbl accesses */
+	struct timer_list fdb_cleanup_timer;
+	DECLARE_HASHTABLE(fdb_tbl, 16);
+	spinlock_t fdb_tbl_lock;		/* for fdb tbl accesses */
+	unsigned long internal_vlan_bitmap[OFDPA_INTERNAL_VLAN_BITMAP_LEN];
+	DECLARE_HASHTABLE(internal_vlan_tbl, 8);
+	spinlock_t internal_vlan_tbl_lock;	/* for vlan tbl accesses */
+	DECLARE_HASHTABLE(neigh_tbl, 16);
+	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
+	u32 neigh_tbl_next_index;
 };
 
 struct ofdpa_port {
+	struct ofdpa *ofdpa;
+	struct rocker_port *rocker_port;
+	struct net_device *dev;
+	u32 pport;
+	struct net_device *bridge_dev;
+	__be16 internal_vlan_id;
+	int stp_state;
+	u32 brport_flags;
+	unsigned long ageing_time;
+	bool ctrls[OFDPA_CTRL_MAX];
+	unsigned long vlan_bitmap[OFDPA_VLAN_BITMAP_LEN];
 };
 
+static const u8 zero_mac[ETH_ALEN]   = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 };
+static const u8 ff_mac[ETH_ALEN]     = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
+static const u8 ll_mac[ETH_ALEN]     = { 0x01, 0x80, 0xc2, 0x00, 0x00, 0x00 };
+static const u8 ll_mask[ETH_ALEN]    = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xf0 };
+static const u8 mcast_mac[ETH_ALEN]  = { 0x01, 0x00, 0x00, 0x00, 0x00, 0x00 };
+static const u8 ipv4_mcast[ETH_ALEN] = { 0x01, 0x00, 0x5e, 0x00, 0x00, 0x00 };
+static const u8 ipv4_mask[ETH_ALEN]  = { 0xff, 0xff, 0xff, 0x80, 0x00, 0x00 };
+static const u8 ipv6_mcast[ETH_ALEN] = { 0x33, 0x33, 0x00, 0x00, 0x00, 0x00 };
+static const u8 ipv6_mask[ETH_ALEN]  = { 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 };
+
+/* Rocker priority levels for flow table entries.  Higher
+ * priority match takes precedence over lower priority match.
+ */
+
+enum {
+	OFDPA_PRIORITY_UNKNOWN = 0,
+	OFDPA_PRIORITY_IG_PORT = 1,
+	OFDPA_PRIORITY_VLAN = 1,
+	OFDPA_PRIORITY_TERM_MAC_UCAST = 0,
+	OFDPA_PRIORITY_TERM_MAC_MCAST = 1,
+	OFDPA_PRIORITY_BRIDGING_VLAN_DFLT_EXACT = 1,
+	OFDPA_PRIORITY_BRIDGING_VLAN_DFLT_WILD = 2,
+	OFDPA_PRIORITY_BRIDGING_VLAN = 3,
+	OFDPA_PRIORITY_BRIDGING_TENANT_DFLT_EXACT = 1,
+	OFDPA_PRIORITY_BRIDGING_TENANT_DFLT_WILD = 2,
+	OFDPA_PRIORITY_BRIDGING_TENANT = 3,
+	OFDPA_PRIORITY_ACL_CTRL = 3,
+	OFDPA_PRIORITY_ACL_NORMAL = 2,
+	OFDPA_PRIORITY_ACL_DFLT = 1,
+};
+
+static bool ofdpa_vlan_id_is_internal(__be16 vlan_id)
+{
+	u16 start = OFDPA_INTERNAL_VLAN_ID_BASE;
+	u16 end = 0xffe;
+	u16 _vlan_id = ntohs(vlan_id);
+
+	return (_vlan_id >= start && _vlan_id <= end);
+}
+
+static __be16 ofdpa_port_vid_to_vlan(const struct ofdpa_port *ofdpa_port,
+				     u16 vid, bool *pop_vlan)
+{
+	__be16 vlan_id;
+
+	if (pop_vlan)
+		*pop_vlan = false;
+	vlan_id = htons(vid);
+	if (!vlan_id) {
+		vlan_id = ofdpa_port->internal_vlan_id;
+		if (pop_vlan)
+			*pop_vlan = true;
+	}
+
+	return vlan_id;
+}
+
+static u16 ofdpa_port_vlan_to_vid(const struct ofdpa_port *ofdpa_port,
+				  __be16 vlan_id)
+{
+	if (ofdpa_vlan_id_is_internal(vlan_id))
+		return 0;
+
+	return ntohs(vlan_id);
+}
+
+static bool ofdpa_port_is_slave(const struct ofdpa_port *ofdpa_port,
+				const char *kind)
+{
+	return ofdpa_port->bridge_dev &&
+		!strcmp(ofdpa_port->bridge_dev->rtnl_link_ops->kind, kind);
+}
+
+static bool ofdpa_port_is_bridged(const struct ofdpa_port *ofdpa_port)
+{
+	return ofdpa_port_is_slave(ofdpa_port, "bridge");
+}
+
+static bool ofdpa_port_is_ovsed(const struct ofdpa_port *ofdpa_port)
+{
+	return ofdpa_port_is_slave(ofdpa_port, "openvswitch");
+}
+
+#define OFDPA_OP_FLAG_REMOVE		BIT(0)
+#define OFDPA_OP_FLAG_NOWAIT		BIT(1)
+#define OFDPA_OP_FLAG_LEARNED		BIT(2)
+#define OFDPA_OP_FLAG_REFRESH		BIT(3)
+
+static bool ofdpa_flags_nowait(int flags)
+{
+	return flags & OFDPA_OP_FLAG_NOWAIT;
+}
+
+static void *__ofdpa_mem_alloc(struct switchdev_trans *trans, int flags,
+			       size_t size)
+{
+	struct switchdev_trans_item *elem = NULL;
+	gfp_t gfp_flags = (flags & OFDPA_OP_FLAG_NOWAIT) ?
+			  GFP_ATOMIC : GFP_KERNEL;
+
+	/* If in transaction prepare phase, allocate the memory
+	 * and enqueue it on a transaction.  If in transaction
+	 * commit phase, dequeue the memory from the transaction
+	 * rather than re-allocating the memory.  The idea is the
+	 * driver code paths for prepare and commit are identical
+	 * so the memory allocated in the prepare phase is the
+	 * memory used in the commit phase.
+	 */
+
+	if (!trans) {
+		elem = kzalloc(size + sizeof(*elem), gfp_flags);
+	} else if (switchdev_trans_ph_prepare(trans)) {
+		elem = kzalloc(size + sizeof(*elem), gfp_flags);
+		if (!elem)
+			return NULL;
+		switchdev_trans_item_enqueue(trans, elem, kfree, elem);
+	} else {
+		elem = switchdev_trans_item_dequeue(trans);
+	}
+
+	return elem ? elem + 1 : NULL;
+}
+
+static void *ofdpa_kzalloc(struct switchdev_trans *trans, int flags,
+			   size_t size)
+{
+	return __ofdpa_mem_alloc(trans, flags, size);
+}
+
+static void *ofdpa_kcalloc(struct switchdev_trans *trans, int flags,
+			   size_t n, size_t size)
+{
+	return __ofdpa_mem_alloc(trans, flags, n * size);
+}
+
+static void ofdpa_kfree(struct switchdev_trans *trans, const void *mem)
+{
+	struct switchdev_trans_item *elem;
+
+	/* Frees are ignored if in transaction prepare phase.  The
+	 * memory remains on the per-port list until freed in the
+	 * commit phase.
+	 */
+
+	if (switchdev_trans_ph_prepare(trans))
+		return;
+
+	elem = (struct switchdev_trans_item *) mem - 1;
+	kfree(elem);
+}
+
+/*************************************************************
+ * Flow, group, FDB, internal VLAN and neigh command prepares
+ *************************************************************/
+
+static int
+ofdpa_cmd_flow_tbl_add_ig_port(struct rocker_desc_info *desc_info,
+			       const struct ofdpa_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.ig_port.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
+			       entry->key.ig_port.in_pport_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.ig_port.goto_tbl))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_flow_tbl_add_vlan(struct rocker_desc_info *desc_info,
+			    const struct ofdpa_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.vlan.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.vlan.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
+				entry->key.vlan.vlan_id_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.vlan.goto_tbl))
+		return -EMSGSIZE;
+	if (entry->key.vlan.untagged &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_NEW_VLAN_ID,
+				entry->key.vlan.new_vlan_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_flow_tbl_add_term_mac(struct rocker_desc_info *desc_info,
+				const struct ofdpa_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.term_mac.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
+			       entry->key.term_mac.in_pport_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
+				entry->key.term_mac.eth_type))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->key.term_mac.eth_dst))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
+			   ETH_ALEN, entry->key.term_mac.eth_dst_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.term_mac.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
+				entry->key.term_mac.vlan_id_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.term_mac.goto_tbl))
+		return -EMSGSIZE;
+	if (entry->key.term_mac.copy_to_cpu &&
+	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
+			      entry->key.term_mac.copy_to_cpu))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_flow_tbl_add_ucast_routing(struct rocker_desc_info *desc_info,
+				     const struct ofdpa_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
+				entry->key.ucast_routing.eth_type))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP,
+				entry->key.ucast_routing.dst4))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be32(desc_info, ROCKER_TLV_OF_DPA_DST_IP_MASK,
+				entry->key.ucast_routing.dst4_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.ucast_routing.goto_tbl))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->key.ucast_routing.group_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_flow_tbl_add_bridge(struct rocker_desc_info *desc_info,
+			      const struct ofdpa_flow_tbl_entry *entry)
+{
+	if (entry->key.bridge.has_eth_dst &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->key.bridge.eth_dst))
+		return -EMSGSIZE;
+	if (entry->key.bridge.has_eth_dst_mask &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
+			   ETH_ALEN, entry->key.bridge.eth_dst_mask))
+		return -EMSGSIZE;
+	if (entry->key.bridge.vlan_id &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.bridge.vlan_id))
+		return -EMSGSIZE;
+	if (entry->key.bridge.tunnel_id &&
+	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_TUNNEL_ID,
+			       entry->key.bridge.tunnel_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GOTO_TABLE_ID,
+			       entry->key.bridge.goto_tbl))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->key.bridge.group_id))
+		return -EMSGSIZE;
+	if (entry->key.bridge.copy_to_cpu &&
+	    rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_COPY_CPU_ACTION,
+			      entry->key.bridge.copy_to_cpu))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_flow_tbl_add_acl(struct rocker_desc_info *desc_info,
+			   const struct ofdpa_flow_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT,
+			       entry->key.acl.in_pport))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_IN_PPORT_MASK,
+			       entry->key.acl.in_pport_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
+			   ETH_ALEN, entry->key.acl.eth_src))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC_MASK,
+			   ETH_ALEN, entry->key.acl.eth_src_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->key.acl.eth_dst))
+		return -EMSGSIZE;
+	if (rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC_MASK,
+			   ETH_ALEN, entry->key.acl.eth_dst_mask))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_ETHERTYPE,
+				entry->key.acl.eth_type))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->key.acl.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID_MASK,
+				entry->key.acl.vlan_id_mask))
+		return -EMSGSIZE;
+
+	switch (ntohs(entry->key.acl.eth_type)) {
+	case ETH_P_IP:
+	case ETH_P_IPV6:
+		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_PROTO,
+				      entry->key.acl.ip_proto))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info,
+				      ROCKER_TLV_OF_DPA_IP_PROTO_MASK,
+				      entry->key.acl.ip_proto_mask))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_DSCP,
+				      entry->key.acl.ip_tos & 0x3f))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info,
+				      ROCKER_TLV_OF_DPA_IP_DSCP_MASK,
+				      entry->key.acl.ip_tos_mask & 0x3f))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_IP_ECN,
+				      (entry->key.acl.ip_tos & 0xc0) >> 6))
+			return -EMSGSIZE;
+		if (rocker_tlv_put_u8(desc_info,
+				      ROCKER_TLV_OF_DPA_IP_ECN_MASK,
+				      (entry->key.acl.ip_tos_mask & 0xc0) >> 6))
+			return -EMSGSIZE;
+		break;
+	}
+
+	if (entry->key.acl.group_id != ROCKER_GROUP_NONE &&
+	    rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->key.acl.group_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int ofdpa_cmd_flow_tbl_add(const struct rocker_port *rocker_port,
+				  struct rocker_desc_info *desc_info,
+				  void *priv)
+{
+	const struct ofdpa_flow_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+	int err = 0;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_TABLE_ID,
+			       entry->key.tbl_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_PRIORITY,
+			       entry->key.priority))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_HARDTIME, 0))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
+			       entry->cookie))
+		return -EMSGSIZE;
+
+	switch (entry->key.tbl_id) {
+	case ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT:
+		err = ofdpa_cmd_flow_tbl_add_ig_port(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_VLAN:
+		err = ofdpa_cmd_flow_tbl_add_vlan(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC:
+		err = ofdpa_cmd_flow_tbl_add_term_mac(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING:
+		err = ofdpa_cmd_flow_tbl_add_ucast_routing(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_BRIDGING:
+		err = ofdpa_cmd_flow_tbl_add_bridge(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_TABLE_ID_ACL_POLICY:
+		err = ofdpa_cmd_flow_tbl_add_acl(desc_info, entry);
+		break;
+	default:
+		err = -ENOTSUPP;
+		break;
+	}
+
+	if (err)
+		return err;
+
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+static int ofdpa_cmd_flow_tbl_del(const struct rocker_port *rocker_port,
+				  struct rocker_desc_info *desc_info,
+				  void *priv)
+{
+	const struct ofdpa_flow_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_OF_DPA_COOKIE,
+			       entry->cookie))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_group_tbl_add_l2_interface(struct rocker_desc_info *desc_info,
+				     struct ofdpa_group_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_OUT_PPORT,
+			       ROCKER_GROUP_PORT_GET(entry->group_id)))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_POP_VLAN,
+			      entry->l2_interface.pop_vlan))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_group_tbl_add_l2_rewrite(struct rocker_desc_info *desc_info,
+				   const struct ofdpa_group_tbl_entry *entry)
+{
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
+			       entry->l2_rewrite.group_id))
+		return -EMSGSIZE;
+	if (!is_zero_ether_addr(entry->l2_rewrite.eth_src) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
+			   ETH_ALEN, entry->l2_rewrite.eth_src))
+		return -EMSGSIZE;
+	if (!is_zero_ether_addr(entry->l2_rewrite.eth_dst) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->l2_rewrite.eth_dst))
+		return -EMSGSIZE;
+	if (entry->l2_rewrite.vlan_id &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->l2_rewrite.vlan_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_group_tbl_add_group_ids(struct rocker_desc_info *desc_info,
+				  const struct ofdpa_group_tbl_entry *entry)
+{
+	int i;
+	struct rocker_tlv *group_ids;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_OF_DPA_GROUP_COUNT,
+			       entry->group_count))
+		return -EMSGSIZE;
+
+	group_ids = rocker_tlv_nest_start(desc_info,
+					  ROCKER_TLV_OF_DPA_GROUP_IDS);
+	if (!group_ids)
+		return -EMSGSIZE;
+
+	for (i = 0; i < entry->group_count; i++)
+		/* Note TLV array is 1-based */
+		if (rocker_tlv_put_u32(desc_info, i + 1, entry->group_ids[i]))
+			return -EMSGSIZE;
+
+	rocker_tlv_nest_end(desc_info, group_ids);
+
+	return 0;
+}
+
+static int
+ofdpa_cmd_group_tbl_add_l3_unicast(struct rocker_desc_info *desc_info,
+				   const struct ofdpa_group_tbl_entry *entry)
+{
+	if (!is_zero_ether_addr(entry->l3_unicast.eth_src) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_SRC_MAC,
+			   ETH_ALEN, entry->l3_unicast.eth_src))
+		return -EMSGSIZE;
+	if (!is_zero_ether_addr(entry->l3_unicast.eth_dst) &&
+	    rocker_tlv_put(desc_info, ROCKER_TLV_OF_DPA_DST_MAC,
+			   ETH_ALEN, entry->l3_unicast.eth_dst))
+		return -EMSGSIZE;
+	if (entry->l3_unicast.vlan_id &&
+	    rocker_tlv_put_be16(desc_info, ROCKER_TLV_OF_DPA_VLAN_ID,
+				entry->l3_unicast.vlan_id))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u8(desc_info, ROCKER_TLV_OF_DPA_TTL_CHECK,
+			      entry->l3_unicast.ttl_check))
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID_LOWER,
+			       entry->l3_unicast.group_id))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int ofdpa_cmd_group_tbl_add(const struct rocker_port *rocker_port,
+				   struct rocker_desc_info *desc_info,
+				   void *priv)
+{
+	struct ofdpa_group_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+	int err = 0;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->group_id))
+		return -EMSGSIZE;
+
+	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE:
+		err = ofdpa_cmd_group_tbl_add_l2_interface(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE:
+		err = ofdpa_cmd_group_tbl_add_l2_rewrite(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
+		err = ofdpa_cmd_group_tbl_add_group_ids(desc_info, entry);
+		break;
+	case ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST:
+		err = ofdpa_cmd_group_tbl_add_l3_unicast(desc_info, entry);
+		break;
+	default:
+		err = -ENOTSUPP;
+		break;
+	}
+
+	if (err)
+		return err;
+
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+static int ofdpa_cmd_group_tbl_del(const struct rocker_port *rocker_port,
+				   struct rocker_desc_info *desc_info,
+				   void *priv)
+{
+	const struct ofdpa_group_tbl_entry *entry = priv;
+	struct rocker_tlv *cmd_info;
+
+	if (rocker_tlv_put_u16(desc_info, ROCKER_TLV_CMD_TYPE, entry->cmd))
+		return -EMSGSIZE;
+	cmd_info = rocker_tlv_nest_start(desc_info, ROCKER_TLV_CMD_INFO);
+	if (!cmd_info)
+		return -EMSGSIZE;
+	if (rocker_tlv_put_u32(desc_info, ROCKER_TLV_OF_DPA_GROUP_ID,
+			       entry->group_id))
+		return -EMSGSIZE;
+	rocker_tlv_nest_end(desc_info, cmd_info);
+
+	return 0;
+}
+
+/***************************************************
+ * Flow, group, FDB, internal VLAN and neigh tables
+ ***************************************************/
+
+static struct ofdpa_flow_tbl_entry *
+ofdpa_flow_tbl_find(const struct ofdpa *ofdpa,
+		    const struct ofdpa_flow_tbl_entry *match)
+{
+	struct ofdpa_flow_tbl_entry *found;
+	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
+
+	hash_for_each_possible(ofdpa->flow_tbl, found,
+			       entry, match->key_crc32) {
+		if (memcmp(&found->key, &match->key, key_len) == 0)
+			return found;
+	}
+
+	return NULL;
+}
+
+static int ofdpa_flow_tbl_add(struct ofdpa_port *ofdpa_port,
+			      struct switchdev_trans *trans, int flags,
+			      struct ofdpa_flow_tbl_entry *match)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_flow_tbl_entry *found;
+	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
+	unsigned long lock_flags;
+
+	match->key_crc32 = crc32(~0, &match->key, key_len);
+
+	spin_lock_irqsave(&ofdpa->flow_tbl_lock, lock_flags);
+
+	found = ofdpa_flow_tbl_find(ofdpa, match);
+
+	if (found) {
+		match->cookie = found->cookie;
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		ofdpa_kfree(trans, found);
+		found = match;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_MOD;
+	} else {
+		found = match;
+		found->cookie = ofdpa->flow_tbl_next_cookie++;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_ADD;
+	}
+
+	if (!switchdev_trans_ph_prepare(trans))
+		hash_add(ofdpa->flow_tbl, &found->entry, found->key_crc32);
+
+	spin_unlock_irqrestore(&ofdpa->flow_tbl_lock, lock_flags);
+
+	if (!switchdev_trans_ph_prepare(trans))
+		return rocker_cmd_exec(ofdpa_port->rocker_port,
+				       ofdpa_flags_nowait(flags),
+				       ofdpa_cmd_flow_tbl_add,
+				       found, NULL, NULL);
+	return 0;
+}
+
+static int ofdpa_flow_tbl_del(struct ofdpa_port *ofdpa_port,
+			      struct switchdev_trans *trans, int flags,
+			      struct ofdpa_flow_tbl_entry *match)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_flow_tbl_entry *found;
+	size_t key_len = match->key_len ? match->key_len : sizeof(found->key);
+	unsigned long lock_flags;
+	int err = 0;
+
+	match->key_crc32 = crc32(~0, &match->key, key_len);
+
+	spin_lock_irqsave(&ofdpa->flow_tbl_lock, lock_flags);
+
+	found = ofdpa_flow_tbl_find(ofdpa, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_FLOW_DEL;
+	}
+
+	spin_unlock_irqrestore(&ofdpa->flow_tbl_lock, lock_flags);
+
+	ofdpa_kfree(trans, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			err = rocker_cmd_exec(ofdpa_port->rocker_port,
+					      ofdpa_flags_nowait(flags),
+					      ofdpa_cmd_flow_tbl_del,
+					      found, NULL, NULL);
+		ofdpa_kfree(trans, found);
+	}
+
+	return err;
+}
+
+static int ofdpa_flow_tbl_do(struct ofdpa_port *ofdpa_port,
+			     struct switchdev_trans *trans, int flags,
+			     struct ofdpa_flow_tbl_entry *entry)
+{
+	if (flags & OFDPA_OP_FLAG_REMOVE)
+		return ofdpa_flow_tbl_del(ofdpa_port, trans, flags, entry);
+	else
+		return ofdpa_flow_tbl_add(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_flow_tbl_ig_port(struct ofdpa_port *ofdpa_port,
+				  struct switchdev_trans *trans, int flags,
+				  u32 in_pport, u32 in_pport_mask,
+				  enum rocker_of_dpa_table_id goto_tbl)
+{
+	struct ofdpa_flow_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.priority = OFDPA_PRIORITY_IG_PORT;
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_INGRESS_PORT;
+	entry->key.ig_port.in_pport = in_pport;
+	entry->key.ig_port.in_pport_mask = in_pport_mask;
+	entry->key.ig_port.goto_tbl = goto_tbl;
+
+	return ofdpa_flow_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_flow_tbl_vlan(struct ofdpa_port *ofdpa_port,
+			       struct switchdev_trans *trans, int flags,
+			       u32 in_pport, __be16 vlan_id,
+			       __be16 vlan_id_mask,
+			       enum rocker_of_dpa_table_id goto_tbl,
+			       bool untagged, __be16 new_vlan_id)
+{
+	struct ofdpa_flow_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.priority = OFDPA_PRIORITY_VLAN;
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_VLAN;
+	entry->key.vlan.in_pport = in_pport;
+	entry->key.vlan.vlan_id = vlan_id;
+	entry->key.vlan.vlan_id_mask = vlan_id_mask;
+	entry->key.vlan.goto_tbl = goto_tbl;
+
+	entry->key.vlan.untagged = untagged;
+	entry->key.vlan.new_vlan_id = new_vlan_id;
+
+	return ofdpa_flow_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_flow_tbl_term_mac(struct ofdpa_port *ofdpa_port,
+				   struct switchdev_trans *trans,
+				   u32 in_pport, u32 in_pport_mask,
+				   __be16 eth_type, const u8 *eth_dst,
+				   const u8 *eth_dst_mask, __be16 vlan_id,
+				   __be16 vlan_id_mask, bool copy_to_cpu,
+				   int flags)
+{
+	struct ofdpa_flow_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	if (is_multicast_ether_addr(eth_dst)) {
+		entry->key.priority = OFDPA_PRIORITY_TERM_MAC_MCAST;
+		entry->key.term_mac.goto_tbl =
+			 ROCKER_OF_DPA_TABLE_ID_MULTICAST_ROUTING;
+	} else {
+		entry->key.priority = OFDPA_PRIORITY_TERM_MAC_UCAST;
+		entry->key.term_mac.goto_tbl =
+			 ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
+	}
+
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
+	entry->key.term_mac.in_pport = in_pport;
+	entry->key.term_mac.in_pport_mask = in_pport_mask;
+	entry->key.term_mac.eth_type = eth_type;
+	ether_addr_copy(entry->key.term_mac.eth_dst, eth_dst);
+	ether_addr_copy(entry->key.term_mac.eth_dst_mask, eth_dst_mask);
+	entry->key.term_mac.vlan_id = vlan_id;
+	entry->key.term_mac.vlan_id_mask = vlan_id_mask;
+	entry->key.term_mac.copy_to_cpu = copy_to_cpu;
+
+	return ofdpa_flow_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_flow_tbl_bridge(struct ofdpa_port *ofdpa_port,
+				 struct switchdev_trans *trans, int flags,
+				 const u8 *eth_dst, const u8 *eth_dst_mask,
+				 __be16 vlan_id, u32 tunnel_id,
+				 enum rocker_of_dpa_table_id goto_tbl,
+				 u32 group_id, bool copy_to_cpu)
+{
+	struct ofdpa_flow_tbl_entry *entry;
+	u32 priority;
+	bool vlan_bridging = !!vlan_id;
+	bool dflt = !eth_dst || (eth_dst && eth_dst_mask);
+	bool wild = false;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_BRIDGING;
+
+	if (eth_dst) {
+		entry->key.bridge.has_eth_dst = 1;
+		ether_addr_copy(entry->key.bridge.eth_dst, eth_dst);
+	}
+	if (eth_dst_mask) {
+		entry->key.bridge.has_eth_dst_mask = 1;
+		ether_addr_copy(entry->key.bridge.eth_dst_mask, eth_dst_mask);
+		if (!ether_addr_equal(eth_dst_mask, ff_mac))
+			wild = true;
+	}
+
+	priority = OFDPA_PRIORITY_UNKNOWN;
+	if (vlan_bridging && dflt && wild)
+		priority = OFDPA_PRIORITY_BRIDGING_VLAN_DFLT_WILD;
+	else if (vlan_bridging && dflt && !wild)
+		priority = OFDPA_PRIORITY_BRIDGING_VLAN_DFLT_EXACT;
+	else if (vlan_bridging && !dflt)
+		priority = OFDPA_PRIORITY_BRIDGING_VLAN;
+	else if (!vlan_bridging && dflt && wild)
+		priority = OFDPA_PRIORITY_BRIDGING_TENANT_DFLT_WILD;
+	else if (!vlan_bridging && dflt && !wild)
+		priority = OFDPA_PRIORITY_BRIDGING_TENANT_DFLT_EXACT;
+	else if (!vlan_bridging && !dflt)
+		priority = OFDPA_PRIORITY_BRIDGING_TENANT;
+
+	entry->key.priority = priority;
+	entry->key.bridge.vlan_id = vlan_id;
+	entry->key.bridge.tunnel_id = tunnel_id;
+	entry->key.bridge.goto_tbl = goto_tbl;
+	entry->key.bridge.group_id = group_id;
+	entry->key.bridge.copy_to_cpu = copy_to_cpu;
+
+	return ofdpa_flow_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_flow_tbl_ucast4_routing(struct ofdpa_port *ofdpa_port,
+					 struct switchdev_trans *trans,
+					 __be16 eth_type, __be32 dst,
+					 __be32 dst_mask, u32 priority,
+					 enum rocker_of_dpa_table_id goto_tbl,
+					 u32 group_id, int flags)
+{
+	struct ofdpa_flow_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING;
+	entry->key.priority = priority;
+	entry->key.ucast_routing.eth_type = eth_type;
+	entry->key.ucast_routing.dst4 = dst;
+	entry->key.ucast_routing.dst4_mask = dst_mask;
+	entry->key.ucast_routing.goto_tbl = goto_tbl;
+	entry->key.ucast_routing.group_id = group_id;
+	entry->key_len = offsetof(struct ofdpa_flow_tbl_key,
+				  ucast_routing.group_id);
+
+	return ofdpa_flow_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_flow_tbl_acl(struct ofdpa_port *ofdpa_port,
+			      struct switchdev_trans *trans, int flags,
+			      u32 in_pport, u32 in_pport_mask,
+			      const u8 *eth_src, const u8 *eth_src_mask,
+			      const u8 *eth_dst, const u8 *eth_dst_mask,
+			      __be16 eth_type, __be16 vlan_id,
+			      __be16 vlan_id_mask, u8 ip_proto,
+			      u8 ip_proto_mask, u8 ip_tos, u8 ip_tos_mask,
+			      u32 group_id)
+{
+	u32 priority;
+	struct ofdpa_flow_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	priority = OFDPA_PRIORITY_ACL_NORMAL;
+	if (eth_dst && eth_dst_mask) {
+		if (ether_addr_equal(eth_dst_mask, mcast_mac))
+			priority = OFDPA_PRIORITY_ACL_DFLT;
+		else if (is_link_local_ether_addr(eth_dst))
+			priority = OFDPA_PRIORITY_ACL_CTRL;
+	}
+
+	entry->key.priority = priority;
+	entry->key.tbl_id = ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	entry->key.acl.in_pport = in_pport;
+	entry->key.acl.in_pport_mask = in_pport_mask;
+
+	if (eth_src)
+		ether_addr_copy(entry->key.acl.eth_src, eth_src);
+	if (eth_src_mask)
+		ether_addr_copy(entry->key.acl.eth_src_mask, eth_src_mask);
+	if (eth_dst)
+		ether_addr_copy(entry->key.acl.eth_dst, eth_dst);
+	if (eth_dst_mask)
+		ether_addr_copy(entry->key.acl.eth_dst_mask, eth_dst_mask);
+
+	entry->key.acl.eth_type = eth_type;
+	entry->key.acl.vlan_id = vlan_id;
+	entry->key.acl.vlan_id_mask = vlan_id_mask;
+	entry->key.acl.ip_proto = ip_proto;
+	entry->key.acl.ip_proto_mask = ip_proto_mask;
+	entry->key.acl.ip_tos = ip_tos;
+	entry->key.acl.ip_tos_mask = ip_tos_mask;
+	entry->key.acl.group_id = group_id;
+
+	return ofdpa_flow_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static struct ofdpa_group_tbl_entry *
+ofdpa_group_tbl_find(const struct ofdpa *ofdpa,
+		     const struct ofdpa_group_tbl_entry *match)
+{
+	struct ofdpa_group_tbl_entry *found;
+
+	hash_for_each_possible(ofdpa->group_tbl, found,
+			       entry, match->group_id) {
+		if (found->group_id == match->group_id)
+			return found;
+	}
+
+	return NULL;
+}
+
+static void ofdpa_group_tbl_entry_free(struct switchdev_trans *trans,
+				       struct ofdpa_group_tbl_entry *entry)
+{
+	switch (ROCKER_GROUP_TYPE_GET(entry->group_id)) {
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
+	case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
+		ofdpa_kfree(trans, entry->group_ids);
+		break;
+	default:
+		break;
+	}
+	ofdpa_kfree(trans, entry);
+}
+
+static int ofdpa_group_tbl_add(struct ofdpa_port *ofdpa_port,
+			       struct switchdev_trans *trans, int flags,
+			       struct ofdpa_group_tbl_entry *match)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_group_tbl_entry *found;
+	unsigned long lock_flags;
+
+	spin_lock_irqsave(&ofdpa->group_tbl_lock, lock_flags);
+
+	found = ofdpa_group_tbl_find(ofdpa, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		ofdpa_group_tbl_entry_free(trans, found);
+		found = match;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_MOD;
+	} else {
+		found = match;
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_ADD;
+	}
+
+	if (!switchdev_trans_ph_prepare(trans))
+		hash_add(ofdpa->group_tbl, &found->entry, found->group_id);
+
+	spin_unlock_irqrestore(&ofdpa->group_tbl_lock, lock_flags);
+
+	if (!switchdev_trans_ph_prepare(trans))
+		return rocker_cmd_exec(ofdpa_port->rocker_port,
+				       ofdpa_flags_nowait(flags),
+				       ofdpa_cmd_group_tbl_add,
+				       found, NULL, NULL);
+	return 0;
+}
+
+static int ofdpa_group_tbl_del(struct ofdpa_port *ofdpa_port,
+			       struct switchdev_trans *trans, int flags,
+			       struct ofdpa_group_tbl_entry *match)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_group_tbl_entry *found;
+	unsigned long lock_flags;
+	int err = 0;
+
+	spin_lock_irqsave(&ofdpa->group_tbl_lock, lock_flags);
+
+	found = ofdpa_group_tbl_find(ofdpa, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+		found->cmd = ROCKER_TLV_CMD_TYPE_OF_DPA_GROUP_DEL;
+	}
+
+	spin_unlock_irqrestore(&ofdpa->group_tbl_lock, lock_flags);
+
+	ofdpa_group_tbl_entry_free(trans, match);
+
+	if (found) {
+		if (!switchdev_trans_ph_prepare(trans))
+			err = rocker_cmd_exec(ofdpa_port->rocker_port,
+					      ofdpa_flags_nowait(flags),
+					      ofdpa_cmd_group_tbl_del,
+					      found, NULL, NULL);
+		ofdpa_group_tbl_entry_free(trans, found);
+	}
+
+	return err;
+}
+
+static int ofdpa_group_tbl_do(struct ofdpa_port *ofdpa_port,
+			      struct switchdev_trans *trans, int flags,
+			      struct ofdpa_group_tbl_entry *entry)
+{
+	if (flags & OFDPA_OP_FLAG_REMOVE)
+		return ofdpa_group_tbl_del(ofdpa_port, trans, flags, entry);
+	else
+		return ofdpa_group_tbl_add(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_group_l2_interface(struct ofdpa_port *ofdpa_port,
+				    struct switchdev_trans *trans, int flags,
+				    __be16 vlan_id, u32 out_pport,
+				    int pop_vlan)
+{
+	struct ofdpa_group_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
+	entry->l2_interface.pop_vlan = pop_vlan;
+
+	return ofdpa_group_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_group_l2_fan_out(struct ofdpa_port *ofdpa_port,
+				  struct switchdev_trans *trans,
+				  int flags, u8 group_count,
+				  const u32 *group_ids, u32 group_id)
+{
+	struct ofdpa_group_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->group_id = group_id;
+	entry->group_count = group_count;
+
+	entry->group_ids = ofdpa_kcalloc(trans, flags,
+					 group_count, sizeof(u32));
+	if (!entry->group_ids) {
+		ofdpa_kfree(trans, entry);
+		return -ENOMEM;
+	}
+	memcpy(entry->group_ids, group_ids, group_count * sizeof(u32));
+
+	return ofdpa_group_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static int ofdpa_group_l2_flood(struct ofdpa_port *ofdpa_port,
+				struct switchdev_trans *trans, int flags,
+				__be16 vlan_id, u8 group_count,
+				const u32 *group_ids, u32 group_id)
+{
+	return ofdpa_group_l2_fan_out(ofdpa_port, trans, flags,
+				      group_count, group_ids,
+				      group_id);
+}
+
+static int ofdpa_group_l3_unicast(struct ofdpa_port *ofdpa_port,
+				  struct switchdev_trans *trans, int flags,
+				  u32 index, const u8 *src_mac, const u8 *dst_mac,
+				  __be16 vlan_id, bool ttl_check, u32 pport)
+{
+	struct ofdpa_group_tbl_entry *entry;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	entry->group_id = ROCKER_GROUP_L3_UNICAST(index);
+	if (src_mac)
+		ether_addr_copy(entry->l3_unicast.eth_src, src_mac);
+	if (dst_mac)
+		ether_addr_copy(entry->l3_unicast.eth_dst, dst_mac);
+	entry->l3_unicast.vlan_id = vlan_id;
+	entry->l3_unicast.ttl_check = ttl_check;
+	entry->l3_unicast.group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, pport);
+
+	return ofdpa_group_tbl_do(ofdpa_port, trans, flags, entry);
+}
+
+static struct ofdpa_neigh_tbl_entry *
+ofdpa_neigh_tbl_find(const struct ofdpa *ofdpa, __be32 ip_addr)
+{
+	struct ofdpa_neigh_tbl_entry *found;
+
+	hash_for_each_possible(ofdpa->neigh_tbl, found,
+			       entry, be32_to_cpu(ip_addr))
+		if (found->ip_addr == ip_addr)
+			return found;
+
+	return NULL;
+}
+
+static void ofdpa_neigh_add(struct ofdpa *ofdpa,
+			    struct switchdev_trans *trans,
+			    struct ofdpa_neigh_tbl_entry *entry)
+{
+	if (!switchdev_trans_ph_commit(trans))
+		entry->index = ofdpa->neigh_tbl_next_index++;
+	if (switchdev_trans_ph_prepare(trans))
+		return;
+	entry->ref_count++;
+	hash_add(ofdpa->neigh_tbl, &entry->entry,
+		 be32_to_cpu(entry->ip_addr));
+}
+
+static void ofdpa_neigh_del(struct switchdev_trans *trans,
+			    struct ofdpa_neigh_tbl_entry *entry)
+{
+	if (switchdev_trans_ph_prepare(trans))
+		return;
+	if (--entry->ref_count == 0) {
+		hash_del(&entry->entry);
+		ofdpa_kfree(trans, entry);
+	}
+}
+
+static void ofdpa_neigh_update(struct ofdpa_neigh_tbl_entry *entry,
+			       struct switchdev_trans *trans,
+			       const u8 *eth_dst, bool ttl_check)
+{
+	if (eth_dst) {
+		ether_addr_copy(entry->eth_dst, eth_dst);
+		entry->ttl_check = ttl_check;
+	} else if (!switchdev_trans_ph_prepare(trans)) {
+		entry->ref_count++;
+	}
+}
+
+static int ofdpa_port_ipv4_neigh(struct ofdpa_port *ofdpa_port,
+				 struct switchdev_trans *trans,
+				 int flags, __be32 ip_addr, const u8 *eth_dst)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_neigh_tbl_entry *entry;
+	struct ofdpa_neigh_tbl_entry *found;
+	unsigned long lock_flags;
+	__be16 eth_type = htons(ETH_P_IP);
+	enum rocker_of_dpa_table_id goto_tbl =
+			ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 group_id;
+	u32 priority = 0;
+	bool adding = !(flags & OFDPA_OP_FLAG_REMOVE);
+	bool updating;
+	bool removing;
+	int err = 0;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&ofdpa->neigh_tbl_lock, lock_flags);
+
+	found = ofdpa_neigh_tbl_find(ofdpa, ip_addr);
+
+	updating = found && adding;
+	removing = found && !adding;
+	adding = !found && adding;
+
+	if (adding) {
+		entry->ip_addr = ip_addr;
+		entry->dev = ofdpa_port->dev;
+		ether_addr_copy(entry->eth_dst, eth_dst);
+		entry->ttl_check = true;
+		ofdpa_neigh_add(ofdpa, trans, entry);
+	} else if (removing) {
+		memcpy(entry, found, sizeof(*entry));
+		ofdpa_neigh_del(trans, found);
+	} else if (updating) {
+		ofdpa_neigh_update(found, trans, eth_dst, true);
+		memcpy(entry, found, sizeof(*entry));
+	} else {
+		err = -ENOENT;
+	}
+
+	spin_unlock_irqrestore(&ofdpa->neigh_tbl_lock, lock_flags);
+
+	if (err)
+		goto err_out;
+
+	/* For each active neighbor, we have an L3 unicast group and
+	 * a /32 route to the neighbor, which uses the L3 unicast
+	 * group.  The L3 unicast group can also be referred to by
+	 * other routes' nexthops.
+	 */
+
+	err = ofdpa_group_l3_unicast(ofdpa_port, trans, flags,
+				     entry->index,
+				     ofdpa_port->dev->dev_addr,
+				     entry->eth_dst,
+				     ofdpa_port->internal_vlan_id,
+				     entry->ttl_check,
+				     ofdpa_port->pport);
+	if (err) {
+		netdev_err(ofdpa_port->dev, "Error (%d) L3 unicast group index %d\n",
+			   err, entry->index);
+		goto err_out;
+	}
+
+	if (adding || removing) {
+		group_id = ROCKER_GROUP_L3_UNICAST(entry->index);
+		err = ofdpa_flow_tbl_ucast4_routing(ofdpa_port, trans,
+						    eth_type, ip_addr,
+						    inet_make_mask(32),
+						    priority, goto_tbl,
+						    group_id, flags);
+
+		if (err)
+			netdev_err(ofdpa_port->dev, "Error (%d) /32 unicast route %pI4 group 0x%08x\n",
+				   err, &entry->ip_addr, group_id);
+	}
+
+err_out:
+	if (!adding)
+		ofdpa_kfree(trans, entry);
+
+	return err;
+}
+
+static int ofdpa_port_ipv4_resolve(struct ofdpa_port *ofdpa_port,
+				   struct switchdev_trans *trans,
+				   __be32 ip_addr)
+{
+	struct net_device *dev = ofdpa_port->dev;
+	struct neighbour *n = __ipv4_neigh_lookup(dev, (__force u32)ip_addr);
+	int err = 0;
+
+	if (!n) {
+		n = neigh_create(&arp_tbl, &ip_addr, dev);
+		if (IS_ERR(n))
+			return IS_ERR(n);
+	}
+
+	/* If the neigh is already resolved, then go ahead and
+	 * install the entry, otherwise start the ARP process to
+	 * resolve the neigh.
+	 */
+
+	if (n->nud_state & NUD_VALID)
+		err = ofdpa_port_ipv4_neigh(ofdpa_port, trans, 0,
+					    ip_addr, n->ha);
+	else
+		neigh_event_send(n, NULL);
+
+	neigh_release(n);
+	return err;
+}
+
+static int ofdpa_port_ipv4_nh(struct ofdpa_port *ofdpa_port,
+			      struct switchdev_trans *trans, int flags,
+			      __be32 ip_addr, u32 *index)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_neigh_tbl_entry *entry;
+	struct ofdpa_neigh_tbl_entry *found;
+	unsigned long lock_flags;
+	bool adding = !(flags & OFDPA_OP_FLAG_REMOVE);
+	bool updating;
+	bool removing;
+	bool resolved = true;
+	int err = 0;
+
+	entry = ofdpa_kzalloc(trans, flags, sizeof(*entry));
+	if (!entry)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&ofdpa->neigh_tbl_lock, lock_flags);
+
+	found = ofdpa_neigh_tbl_find(ofdpa, ip_addr);
+	if (found)
+		*index = found->index;
+
+	updating = found && adding;
+	removing = found && !adding;
+	adding = !found && adding;
+
+	if (adding) {
+		entry->ip_addr = ip_addr;
+		entry->dev = ofdpa_port->dev;
+		ofdpa_neigh_add(ofdpa, trans, entry);
+		*index = entry->index;
+		resolved = false;
+	} else if (removing) {
+		ofdpa_neigh_del(trans, found);
+	} else if (updating) {
+		ofdpa_neigh_update(found, trans, NULL, false);
+		resolved = !is_zero_ether_addr(found->eth_dst);
+	} else {
+		err = -ENOENT;
+	}
+
+	spin_unlock_irqrestore(&ofdpa->neigh_tbl_lock, lock_flags);
+
+	if (!adding)
+		ofdpa_kfree(trans, entry);
+
+	if (err)
+		return err;
+
+	/* Resolved means neigh ip_addr is resolved to neigh mac. */
+
+	if (!resolved)
+		err = ofdpa_port_ipv4_resolve(ofdpa_port, trans, ip_addr);
+
+	return err;
+}
+
+static struct ofdpa_port *ofdpa_port_get(const struct ofdpa *ofdpa,
+					 int port_index)
+{
+	struct rocker_port *rocker_port;
+
+	rocker_port = ofdpa->rocker->ports[port_index];
+	return rocker_port ? rocker_port->wpriv : NULL;
+}
+
+static int ofdpa_port_vlan_flood_group(struct ofdpa_port *ofdpa_port,
+				       struct switchdev_trans *trans,
+				       int flags, __be16 vlan_id)
+{
+	struct ofdpa_port *p;
+	const struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	unsigned int port_count = ofdpa->rocker->port_count;
+	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
+	u32 *group_ids;
+	u8 group_count = 0;
+	int err = 0;
+	int i;
+
+	group_ids = ofdpa_kcalloc(trans, flags, port_count, sizeof(u32));
+	if (!group_ids)
+		return -ENOMEM;
+
+	/* Adjust the flood group for this VLAN.  The flood group
+	 * references an L2 interface group for each port in this
+	 * VLAN.
+	 */
+
+	for (i = 0; i < port_count; i++) {
+		p = ofdpa_port_get(ofdpa, i);
+		if (!p)
+			continue;
+		if (!ofdpa_port_is_bridged(p))
+			continue;
+		if (test_bit(ntohs(vlan_id), p->vlan_bitmap)) {
+			group_ids[group_count++] =
+				ROCKER_GROUP_L2_INTERFACE(vlan_id, p->pport);
+		}
+	}
+
+	/* If there are no bridged ports in this VLAN, we're done */
+	if (group_count == 0)
+		goto no_ports_in_vlan;
+
+	err = ofdpa_group_l2_flood(ofdpa_port, trans, flags, vlan_id,
+				   group_count, group_ids, group_id);
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) port VLAN l2 flood group\n", err);
+
+no_ports_in_vlan:
+	ofdpa_kfree(trans, group_ids);
+	return err;
+}
+
+static int ofdpa_port_vlan_l2_groups(struct ofdpa_port *ofdpa_port,
+				     struct switchdev_trans *trans, int flags,
+				     __be16 vlan_id, bool pop_vlan)
+{
+	const struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	unsigned int port_count = ofdpa->rocker->port_count;
+	struct ofdpa_port *p;
+	bool adding = !(flags & OFDPA_OP_FLAG_REMOVE);
+	u32 out_pport;
+	int ref = 0;
+	int err;
+	int i;
+
+	/* An L2 interface group for this port in this VLAN, but
+	 * only when port STP state is LEARNING|FORWARDING.
+	 */
+
+	if (ofdpa_port->stp_state == BR_STATE_LEARNING ||
+	    ofdpa_port->stp_state == BR_STATE_FORWARDING) {
+		out_pport = ofdpa_port->pport;
+		err = ofdpa_group_l2_interface(ofdpa_port, trans, flags,
+					       vlan_id, out_pport, pop_vlan);
+		if (err) {
+			netdev_err(ofdpa_port->dev, "Error (%d) port VLAN l2 group for pport %d\n",
+				   err, out_pport);
+			return err;
+		}
+	}
+
+	/* An L2 interface group for this VLAN to CPU port.
+	 * Add when first port joins this VLAN and destroy when
+	 * last port leaves this VLAN.
+	 */
+
+	for (i = 0; i < port_count; i++) {
+		p = ofdpa_port_get(ofdpa, i);
+		if (p && test_bit(ntohs(vlan_id), p->vlan_bitmap))
+			ref++;
+	}
+
+	if ((!adding || ref != 1) && (adding || ref != 0))
+		return 0;
+
+	out_pport = 0;
+	err = ofdpa_group_l2_interface(ofdpa_port, trans, flags,
+				       vlan_id, out_pport, pop_vlan);
+	if (err) {
+		netdev_err(ofdpa_port->dev, "Error (%d) port VLAN l2 group for CPU port\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static struct ofdpa_ctrl {
+	const u8 *eth_dst;
+	const u8 *eth_dst_mask;
+	__be16 eth_type;
+	bool acl;
+	bool bridge;
+	bool term;
+	bool copy_to_cpu;
+} ofdpa_ctrls[] = {
+	[OFDPA_CTRL_LINK_LOCAL_MCAST] = {
+		/* pass link local multicast pkts up to CPU for filtering */
+		.eth_dst = ll_mac,
+		.eth_dst_mask = ll_mask,
+		.acl = true,
+	},
+	[OFDPA_CTRL_LOCAL_ARP] = {
+		/* pass local ARP pkts up to CPU */
+		.eth_dst = zero_mac,
+		.eth_dst_mask = zero_mac,
+		.eth_type = htons(ETH_P_ARP),
+		.acl = true,
+	},
+	[OFDPA_CTRL_IPV4_MCAST] = {
+		/* pass IPv4 mcast pkts up to CPU, RFC 1112 */
+		.eth_dst = ipv4_mcast,
+		.eth_dst_mask = ipv4_mask,
+		.eth_type = htons(ETH_P_IP),
+		.term  = true,
+		.copy_to_cpu = true,
+	},
+	[OFDPA_CTRL_IPV6_MCAST] = {
+		/* pass IPv6 mcast pkts up to CPU, RFC 2464 */
+		.eth_dst = ipv6_mcast,
+		.eth_dst_mask = ipv6_mask,
+		.eth_type = htons(ETH_P_IPV6),
+		.term  = true,
+		.copy_to_cpu = true,
+	},
+	[OFDPA_CTRL_DFLT_BRIDGING] = {
+		/* flood any pkts on vlan */
+		.bridge = true,
+		.copy_to_cpu = true,
+	},
+	[OFDPA_CTRL_DFLT_OVS] = {
+		/* pass all pkts up to CPU */
+		.eth_dst = zero_mac,
+		.eth_dst_mask = zero_mac,
+		.acl = true,
+	},
+};
+
+static int ofdpa_port_ctrl_vlan_acl(struct ofdpa_port *ofdpa_port,
+				    struct switchdev_trans *trans, int flags,
+				    const struct ofdpa_ctrl *ctrl, __be16 vlan_id)
+{
+	u32 in_pport = ofdpa_port->pport;
+	u32 in_pport_mask = 0xffffffff;
+	u32 out_pport = 0;
+	const u8 *eth_src = NULL;
+	const u8 *eth_src_mask = NULL;
+	__be16 vlan_id_mask = htons(0xffff);
+	u8 ip_proto = 0;
+	u8 ip_proto_mask = 0;
+	u8 ip_tos = 0;
+	u8 ip_tos_mask = 0;
+	u32 group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
+	int err;
+
+	err = ofdpa_flow_tbl_acl(ofdpa_port, trans, flags,
+				 in_pport, in_pport_mask,
+				 eth_src, eth_src_mask,
+				 ctrl->eth_dst, ctrl->eth_dst_mask,
+				 ctrl->eth_type,
+				 vlan_id, vlan_id_mask,
+				 ip_proto, ip_proto_mask,
+				 ip_tos, ip_tos_mask,
+				 group_id);
+
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) ctrl ACL\n", err);
+
+	return err;
+}
+
+static int ofdpa_port_ctrl_vlan_bridge(struct ofdpa_port *ofdpa_port,
+				       struct switchdev_trans *trans,
+				       int flags,
+				       const struct ofdpa_ctrl *ctrl,
+				       __be16 vlan_id)
+{
+	enum rocker_of_dpa_table_id goto_tbl =
+			ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 group_id = ROCKER_GROUP_L2_FLOOD(vlan_id, 0);
+	u32 tunnel_id = 0;
+	int err;
+
+	if (!ofdpa_port_is_bridged(ofdpa_port))
+		return 0;
+
+	err = ofdpa_flow_tbl_bridge(ofdpa_port, trans, flags,
+				    ctrl->eth_dst, ctrl->eth_dst_mask,
+				    vlan_id, tunnel_id,
+				    goto_tbl, group_id, ctrl->copy_to_cpu);
+
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) ctrl FLOOD\n", err);
+
+	return err;
+}
+
+static int ofdpa_port_ctrl_vlan_term(struct ofdpa_port *ofdpa_port,
+				     struct switchdev_trans *trans, int flags,
+				     const struct ofdpa_ctrl *ctrl, __be16 vlan_id)
+{
+	u32 in_pport_mask = 0xffffffff;
+	__be16 vlan_id_mask = htons(0xffff);
+	int err;
+
+	if (ntohs(vlan_id) == 0)
+		vlan_id = ofdpa_port->internal_vlan_id;
+
+	err = ofdpa_flow_tbl_term_mac(ofdpa_port, trans,
+				      ofdpa_port->pport, in_pport_mask,
+				      ctrl->eth_type, ctrl->eth_dst,
+				      ctrl->eth_dst_mask, vlan_id,
+				      vlan_id_mask, ctrl->copy_to_cpu,
+				      flags);
+
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) ctrl term\n", err);
+
+	return err;
+}
+
+static int ofdpa_port_ctrl_vlan(struct ofdpa_port *ofdpa_port,
+				struct switchdev_trans *trans, int flags,
+				const struct ofdpa_ctrl *ctrl, __be16 vlan_id)
+{
+	if (ctrl->acl)
+		return ofdpa_port_ctrl_vlan_acl(ofdpa_port, trans, flags,
+						ctrl, vlan_id);
+	if (ctrl->bridge)
+		return ofdpa_port_ctrl_vlan_bridge(ofdpa_port, trans, flags,
+						   ctrl, vlan_id);
+
+	if (ctrl->term)
+		return ofdpa_port_ctrl_vlan_term(ofdpa_port, trans, flags,
+						 ctrl, vlan_id);
+
+	return -EOPNOTSUPP;
+}
+
+static int ofdpa_port_ctrl_vlan_add(struct ofdpa_port *ofdpa_port,
+				    struct switchdev_trans *trans, int flags,
+				    __be16 vlan_id)
+{
+	int err = 0;
+	int i;
+
+	for (i = 0; i < OFDPA_CTRL_MAX; i++) {
+		if (ofdpa_port->ctrls[i]) {
+			err = ofdpa_port_ctrl_vlan(ofdpa_port, trans, flags,
+						   &ofdpa_ctrls[i], vlan_id);
+			if (err)
+				return err;
+		}
+	}
+
+	return err;
+}
+
+static int ofdpa_port_ctrl(struct ofdpa_port *ofdpa_port,
+			   struct switchdev_trans *trans, int flags,
+			   const struct ofdpa_ctrl *ctrl)
+{
+	u16 vid;
+	int err = 0;
+
+	for (vid = 1; vid < VLAN_N_VID; vid++) {
+		if (!test_bit(vid, ofdpa_port->vlan_bitmap))
+			continue;
+		err = ofdpa_port_ctrl_vlan(ofdpa_port, trans, flags,
+					   ctrl, htons(vid));
+		if (err)
+			break;
+	}
+
+	return err;
+}
+
+static int ofdpa_port_vlan(struct ofdpa_port *ofdpa_port,
+			   struct switchdev_trans *trans, int flags, u16 vid)
+{
+	enum rocker_of_dpa_table_id goto_tbl =
+			ROCKER_OF_DPA_TABLE_ID_TERMINATION_MAC;
+	u32 in_pport = ofdpa_port->pport;
+	__be16 vlan_id = htons(vid);
+	__be16 vlan_id_mask = htons(0xffff);
+	__be16 internal_vlan_id;
+	bool untagged;
+	bool adding = !(flags & OFDPA_OP_FLAG_REMOVE);
+	int err;
+
+	internal_vlan_id = ofdpa_port_vid_to_vlan(ofdpa_port, vid, &untagged);
+
+	if (adding &&
+	    test_bit(ntohs(internal_vlan_id), ofdpa_port->vlan_bitmap))
+		return 0; /* already added */
+	else if (!adding &&
+		 !test_bit(ntohs(internal_vlan_id), ofdpa_port->vlan_bitmap))
+		return 0; /* already removed */
+
+	change_bit(ntohs(internal_vlan_id), ofdpa_port->vlan_bitmap);
+
+	if (adding) {
+		err = ofdpa_port_ctrl_vlan_add(ofdpa_port, trans, flags,
+					       internal_vlan_id);
+		if (err) {
+			netdev_err(ofdpa_port->dev, "Error (%d) port ctrl vlan add\n", err);
+			goto err_out;
+		}
+	}
+
+	err = ofdpa_port_vlan_l2_groups(ofdpa_port, trans, flags,
+					internal_vlan_id, untagged);
+	if (err) {
+		netdev_err(ofdpa_port->dev, "Error (%d) port VLAN l2 groups\n", err);
+		goto err_out;
+	}
+
+	err = ofdpa_port_vlan_flood_group(ofdpa_port, trans, flags,
+					  internal_vlan_id);
+	if (err) {
+		netdev_err(ofdpa_port->dev, "Error (%d) port VLAN l2 flood group\n", err);
+		goto err_out;
+	}
+
+	err = ofdpa_flow_tbl_vlan(ofdpa_port, trans, flags,
+				  in_pport, vlan_id, vlan_id_mask,
+				  goto_tbl, untagged, internal_vlan_id);
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) port VLAN table\n", err);
+
+err_out:
+	if (switchdev_trans_ph_prepare(trans))
+		change_bit(ntohs(internal_vlan_id), ofdpa_port->vlan_bitmap);
+
+	return err;
+}
+
+static int ofdpa_port_ig_tbl(struct ofdpa_port *ofdpa_port,
+			     struct switchdev_trans *trans, int flags)
+{
+	enum rocker_of_dpa_table_id goto_tbl;
+	u32 in_pport;
+	u32 in_pport_mask;
+	int err;
+
+	/* Normal Ethernet Frames.  Matches pkts from any local physical
+	 * ports.  Goto VLAN tbl.
+	 */
+
+	in_pport = 0;
+	in_pport_mask = 0xffff0000;
+	goto_tbl = ROCKER_OF_DPA_TABLE_ID_VLAN;
+
+	err = ofdpa_flow_tbl_ig_port(ofdpa_port, trans, flags,
+				     in_pport, in_pport_mask,
+				     goto_tbl);
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) ingress port table entry\n", err);
+
+	return err;
+}
+
+struct ofdpa_fdb_learn_work {
+	struct work_struct work;
+	struct ofdpa_port *ofdpa_port;
+	struct switchdev_trans *trans;
+	int flags;
+	u8 addr[ETH_ALEN];
+	u16 vid;
+};
+
+static void ofdpa_port_fdb_learn_work(struct work_struct *work)
+{
+	const struct ofdpa_fdb_learn_work *lw =
+		container_of(work, struct ofdpa_fdb_learn_work, work);
+	bool removing = (lw->flags & OFDPA_OP_FLAG_REMOVE);
+	bool learned = (lw->flags & OFDPA_OP_FLAG_LEARNED);
+	struct switchdev_notifier_fdb_info info;
+
+	info.addr = lw->addr;
+	info.vid = lw->vid;
+
+	rtnl_lock();
+	if (learned && removing)
+		call_switchdev_notifiers(SWITCHDEV_FDB_DEL,
+					 lw->ofdpa_port->dev, &info.info);
+	else if (learned && !removing)
+		call_switchdev_notifiers(SWITCHDEV_FDB_ADD,
+					 lw->ofdpa_port->dev, &info.info);
+	rtnl_unlock();
+
+	ofdpa_kfree(lw->trans, work);
+}
+
+static int ofdpa_port_fdb_learn(struct ofdpa_port *ofdpa_port,
+				struct switchdev_trans *trans, int flags,
+				const u8 *addr, __be16 vlan_id)
+{
+	struct ofdpa_fdb_learn_work *lw;
+	enum rocker_of_dpa_table_id goto_tbl =
+			ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 out_pport = ofdpa_port->pport;
+	u32 tunnel_id = 0;
+	u32 group_id = ROCKER_GROUP_NONE;
+	bool syncing = !!(ofdpa_port->brport_flags & BR_LEARNING_SYNC);
+	bool copy_to_cpu = false;
+	int err;
+
+	if (ofdpa_port_is_bridged(ofdpa_port))
+		group_id = ROCKER_GROUP_L2_INTERFACE(vlan_id, out_pport);
+
+	if (!(flags & OFDPA_OP_FLAG_REFRESH)) {
+		err = ofdpa_flow_tbl_bridge(ofdpa_port, trans, flags, addr,
+					    NULL, vlan_id, tunnel_id, goto_tbl,
+					    group_id, copy_to_cpu);
+		if (err)
+			return err;
+	}
+
+	if (!syncing)
+		return 0;
+
+	if (!ofdpa_port_is_bridged(ofdpa_port))
+		return 0;
+
+	lw = ofdpa_kzalloc(trans, flags, sizeof(*lw));
+	if (!lw)
+		return -ENOMEM;
+
+	INIT_WORK(&lw->work, ofdpa_port_fdb_learn_work);
+
+	lw->ofdpa_port = ofdpa_port;
+	lw->trans = trans;
+	lw->flags = flags;
+	ether_addr_copy(lw->addr, addr);
+	lw->vid = ofdpa_port_vlan_to_vid(ofdpa_port, vlan_id);
+
+	if (switchdev_trans_ph_prepare(trans))
+		ofdpa_kfree(trans, lw);
+	else
+		schedule_work(&lw->work);
+
+	return 0;
+}
+
+static struct ofdpa_fdb_tbl_entry *
+ofdpa_fdb_tbl_find(const struct ofdpa *ofdpa,
+		   const struct ofdpa_fdb_tbl_entry *match)
+{
+	struct ofdpa_fdb_tbl_entry *found;
+
+	hash_for_each_possible(ofdpa->fdb_tbl, found, entry, match->key_crc32)
+		if (memcmp(&found->key, &match->key, sizeof(found->key)) == 0)
+			return found;
+
+	return NULL;
+}
+
+static int ofdpa_port_fdb(struct ofdpa_port *ofdpa_port,
+			  struct switchdev_trans *trans,
+			  const unsigned char *addr,
+			  __be16 vlan_id, int flags)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_fdb_tbl_entry *fdb;
+	struct ofdpa_fdb_tbl_entry *found;
+	bool removing = (flags & OFDPA_OP_FLAG_REMOVE);
+	unsigned long lock_flags;
+
+	fdb = ofdpa_kzalloc(trans, flags, sizeof(*fdb));
+	if (!fdb)
+		return -ENOMEM;
+
+	fdb->learned = (flags & OFDPA_OP_FLAG_LEARNED);
+	fdb->touched = jiffies;
+	fdb->key.ofdpa_port = ofdpa_port;
+	ether_addr_copy(fdb->key.addr, addr);
+	fdb->key.vlan_id = vlan_id;
+	fdb->key_crc32 = crc32(~0, &fdb->key, sizeof(fdb->key));
+
+	spin_lock_irqsave(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	found = ofdpa_fdb_tbl_find(ofdpa, fdb);
+
+	if (found) {
+		found->touched = jiffies;
+		if (removing) {
+			ofdpa_kfree(trans, fdb);
+			if (!switchdev_trans_ph_prepare(trans))
+				hash_del(&found->entry);
+		}
+	} else if (!removing) {
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_add(ofdpa->fdb_tbl, &fdb->entry,
+				 fdb->key_crc32);
+	}
+
+	spin_unlock_irqrestore(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	/* Check if adding and already exists, or removing and can't find */
+	if (!found != !removing) {
+		ofdpa_kfree(trans, fdb);
+		if (!found && removing)
+			return 0;
+		/* Refreshing existing to update aging timers */
+		flags |= OFDPA_OP_FLAG_REFRESH;
+	}
+
+	return ofdpa_port_fdb_learn(ofdpa_port, trans, flags, addr, vlan_id);
+}
+
+static int ofdpa_port_fdb_flush(struct ofdpa_port *ofdpa_port,
+				struct switchdev_trans *trans, int flags)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_fdb_tbl_entry *found;
+	unsigned long lock_flags;
+	struct hlist_node *tmp;
+	int bkt;
+	int err = 0;
+
+	if (ofdpa_port->stp_state == BR_STATE_LEARNING ||
+	    ofdpa_port->stp_state == BR_STATE_FORWARDING)
+		return 0;
+
+	flags |= OFDPA_OP_FLAG_NOWAIT | OFDPA_OP_FLAG_REMOVE;
+
+	spin_lock_irqsave(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	hash_for_each_safe(ofdpa->fdb_tbl, bkt, tmp, found, entry) {
+		if (found->key.ofdpa_port != ofdpa_port)
+			continue;
+		if (!found->learned)
+			continue;
+		err = ofdpa_port_fdb_learn(ofdpa_port, trans, flags,
+					   found->key.addr,
+					   found->key.vlan_id);
+		if (err)
+			goto err_out;
+		if (!switchdev_trans_ph_prepare(trans))
+			hash_del(&found->entry);
+	}
+
+err_out:
+	spin_unlock_irqrestore(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	return err;
+}
+
+static void ofdpa_fdb_cleanup(unsigned long data)
+{
+	struct ofdpa *ofdpa = (struct ofdpa *)data;
+	struct ofdpa_port *ofdpa_port;
+	struct ofdpa_fdb_tbl_entry *entry;
+	struct hlist_node *tmp;
+	unsigned long next_timer = jiffies + BR_MIN_AGEING_TIME;
+	unsigned long expires;
+	unsigned long lock_flags;
+	int flags = OFDPA_OP_FLAG_NOWAIT | OFDPA_OP_FLAG_REMOVE |
+		    OFDPA_OP_FLAG_LEARNED;
+	int bkt;
+
+	spin_lock_irqsave(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	hash_for_each_safe(ofdpa->fdb_tbl, bkt, tmp, entry, entry) {
+		if (!entry->learned)
+			continue;
+		ofdpa_port = entry->key.ofdpa_port;
+		expires = entry->touched + ofdpa_port->ageing_time;
+		if (time_before_eq(expires, jiffies)) {
+			ofdpa_port_fdb_learn(ofdpa_port, NULL,
+					     flags, entry->key.addr,
+					     entry->key.vlan_id);
+			hash_del(&entry->entry);
+		} else if (time_before(expires, next_timer)) {
+			next_timer = expires;
+		}
+	}
+
+	spin_unlock_irqrestore(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	mod_timer(&ofdpa->fdb_cleanup_timer, round_jiffies_up(next_timer));
+}
+
+static int ofdpa_port_router_mac(struct ofdpa_port *ofdpa_port,
+				 struct switchdev_trans *trans, int flags,
+				 __be16 vlan_id)
+{
+	u32 in_pport_mask = 0xffffffff;
+	__be16 eth_type;
+	const u8 *dst_mac_mask = ff_mac;
+	__be16 vlan_id_mask = htons(0xffff);
+	bool copy_to_cpu = false;
+	int err;
+
+	if (ntohs(vlan_id) == 0)
+		vlan_id = ofdpa_port->internal_vlan_id;
+
+	eth_type = htons(ETH_P_IP);
+	err = ofdpa_flow_tbl_term_mac(ofdpa_port, trans,
+				      ofdpa_port->pport, in_pport_mask,
+				      eth_type, ofdpa_port->dev->dev_addr,
+				      dst_mac_mask, vlan_id, vlan_id_mask,
+				      copy_to_cpu, flags);
+	if (err)
+		return err;
+
+	eth_type = htons(ETH_P_IPV6);
+	err = ofdpa_flow_tbl_term_mac(ofdpa_port, trans,
+				      ofdpa_port->pport, in_pport_mask,
+				      eth_type, ofdpa_port->dev->dev_addr,
+				      dst_mac_mask, vlan_id, vlan_id_mask,
+				      copy_to_cpu, flags);
+
+	return err;
+}
+
+static int ofdpa_port_fwding(struct ofdpa_port *ofdpa_port,
+			     struct switchdev_trans *trans, int flags)
+{
+	bool pop_vlan;
+	u32 out_pport;
+	__be16 vlan_id;
+	u16 vid;
+	int err;
+
+	/* Port will be forwarding-enabled if its STP state is LEARNING
+	 * or FORWARDING.  Traffic from CPU can still egress, regardless of
+	 * port STP state.  Use L2 interface group on port VLANs as a way
+	 * to toggle port forwarding: if forwarding is disabled, L2
+	 * interface group will not exist.
+	 */
+
+	if (ofdpa_port->stp_state != BR_STATE_LEARNING &&
+	    ofdpa_port->stp_state != BR_STATE_FORWARDING)
+		flags |= OFDPA_OP_FLAG_REMOVE;
+
+	out_pport = ofdpa_port->pport;
+	for (vid = 1; vid < VLAN_N_VID; vid++) {
+		if (!test_bit(vid, ofdpa_port->vlan_bitmap))
+			continue;
+		vlan_id = htons(vid);
+		pop_vlan = ofdpa_vlan_id_is_internal(vlan_id);
+		err = ofdpa_group_l2_interface(ofdpa_port, trans, flags,
+					       vlan_id, out_pport, pop_vlan);
+		if (err) {
+			netdev_err(ofdpa_port->dev, "Error (%d) port VLAN l2 group for pport %d\n",
+				   err, out_pport);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int ofdpa_port_stp_update(struct ofdpa_port *ofdpa_port,
+				 struct switchdev_trans *trans,
+				 int flags, u8 state)
+{
+	bool want[OFDPA_CTRL_MAX] = { 0, };
+	bool prev_ctrls[OFDPA_CTRL_MAX];
+	u8 uninitialized_var(prev_state);
+	int err;
+	int i;
+
+	if (switchdev_trans_ph_prepare(trans)) {
+		memcpy(prev_ctrls, ofdpa_port->ctrls, sizeof(prev_ctrls));
+		prev_state = ofdpa_port->stp_state;
+	}
+
+	if (ofdpa_port->stp_state == state)
+		return 0;
+
+	ofdpa_port->stp_state = state;
+
+	switch (state) {
+	case BR_STATE_DISABLED:
+		/* port is completely disabled */
+		break;
+	case BR_STATE_LISTENING:
+	case BR_STATE_BLOCKING:
+		want[OFDPA_CTRL_LINK_LOCAL_MCAST] = true;
+		break;
+	case BR_STATE_LEARNING:
+	case BR_STATE_FORWARDING:
+		if (!ofdpa_port_is_ovsed(ofdpa_port))
+			want[OFDPA_CTRL_LINK_LOCAL_MCAST] = true;
+		want[OFDPA_CTRL_IPV4_MCAST] = true;
+		want[OFDPA_CTRL_IPV6_MCAST] = true;
+		if (ofdpa_port_is_bridged(ofdpa_port))
+			want[OFDPA_CTRL_DFLT_BRIDGING] = true;
+		else if (ofdpa_port_is_ovsed(ofdpa_port))
+			want[OFDPA_CTRL_DFLT_OVS] = true;
+		else
+			want[OFDPA_CTRL_LOCAL_ARP] = true;
+		break;
+	}
+
+	for (i = 0; i < OFDPA_CTRL_MAX; i++) {
+		if (want[i] != ofdpa_port->ctrls[i]) {
+			int ctrl_flags = flags |
+					 (want[i] ? 0 : OFDPA_OP_FLAG_REMOVE);
+			err = ofdpa_port_ctrl(ofdpa_port, trans, ctrl_flags,
+					      &ofdpa_ctrls[i]);
+			if (err)
+				goto err_out;
+			ofdpa_port->ctrls[i] = want[i];
+		}
+	}
+
+	err = ofdpa_port_fdb_flush(ofdpa_port, trans, flags);
+	if (err)
+		goto err_out;
+
+	err = ofdpa_port_fwding(ofdpa_port, trans, flags);
+
+err_out:
+	if (switchdev_trans_ph_prepare(trans)) {
+		memcpy(ofdpa_port->ctrls, prev_ctrls, sizeof(prev_ctrls));
+		ofdpa_port->stp_state = prev_state;
+	}
+
+	return err;
+}
+
+static int ofdpa_port_fwd_enable(struct ofdpa_port *ofdpa_port, int flags)
+{
+	if (ofdpa_port_is_bridged(ofdpa_port))
+		/* bridge STP will enable port */
+		return 0;
+
+	/* port is not bridged, so simulate going to FORWARDING state */
+	return ofdpa_port_stp_update(ofdpa_port, NULL, flags,
+				     BR_STATE_FORWARDING);
+}
+
+static int ofdpa_port_fwd_disable(struct ofdpa_port *ofdpa_port, int flags)
+{
+	if (ofdpa_port_is_bridged(ofdpa_port))
+		/* bridge STP will disable port */
+		return 0;
+
+	/* port is not bridged, so simulate going to DISABLED state */
+	return ofdpa_port_stp_update(ofdpa_port, NULL, flags,
+				     BR_STATE_DISABLED);
+}
+
+static int ofdpa_port_vlan_add(struct ofdpa_port *ofdpa_port,
+			       struct switchdev_trans *trans,
+			       u16 vid, u16 flags)
+{
+	int err;
+
+	/* XXX deal with flags for PVID and untagged */
+
+	err = ofdpa_port_vlan(ofdpa_port, trans, 0, vid);
+	if (err)
+		return err;
+
+	err = ofdpa_port_router_mac(ofdpa_port, trans, 0, htons(vid));
+	if (err)
+		ofdpa_port_vlan(ofdpa_port, trans,
+				OFDPA_OP_FLAG_REMOVE, vid);
+
+	return err;
+}
+
+static int ofdpa_port_vlan_del(struct ofdpa_port *ofdpa_port,
+			       u16 vid, u16 flags)
+{
+	int err;
+
+	err = ofdpa_port_router_mac(ofdpa_port, NULL,
+				    OFDPA_OP_FLAG_REMOVE, htons(vid));
+	if (err)
+		return err;
+
+	return ofdpa_port_vlan(ofdpa_port, NULL,
+			       OFDPA_OP_FLAG_REMOVE, vid);
+}
+
+static struct ofdpa_internal_vlan_tbl_entry *
+ofdpa_internal_vlan_tbl_find(const struct ofdpa *ofdpa, int ifindex)
+{
+	struct ofdpa_internal_vlan_tbl_entry *found;
+
+	hash_for_each_possible(ofdpa->internal_vlan_tbl, found,
+			       entry, ifindex) {
+		if (found->ifindex == ifindex)
+			return found;
+	}
+
+	return NULL;
+}
+
+static __be16 ofdpa_port_internal_vlan_id_get(struct ofdpa_port *ofdpa_port,
+					      int ifindex)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_internal_vlan_tbl_entry *entry;
+	struct ofdpa_internal_vlan_tbl_entry *found;
+	unsigned long lock_flags;
+	int i;
+
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		return 0;
+
+	entry->ifindex = ifindex;
+
+	spin_lock_irqsave(&ofdpa->internal_vlan_tbl_lock, lock_flags);
+
+	found = ofdpa_internal_vlan_tbl_find(ofdpa, ifindex);
+	if (found) {
+		kfree(entry);
+		goto found;
+	}
+
+	found = entry;
+	hash_add(ofdpa->internal_vlan_tbl, &found->entry, found->ifindex);
+
+	for (i = 0; i < OFDPA_N_INTERNAL_VLANS; i++) {
+		if (test_and_set_bit(i, ofdpa->internal_vlan_bitmap))
+			continue;
+		found->vlan_id = htons(OFDPA_INTERNAL_VLAN_ID_BASE + i);
+		goto found;
+	}
+
+	netdev_err(ofdpa_port->dev, "Out of internal VLAN IDs\n");
+
+found:
+	found->ref_count++;
+	spin_unlock_irqrestore(&ofdpa->internal_vlan_tbl_lock, lock_flags);
+
+	return found->vlan_id;
+}
+
+static int ofdpa_port_fib_ipv4(struct ofdpa_port *ofdpa_port,
+			       struct switchdev_trans *trans, __be32 dst,
+			       int dst_len, const struct fib_info *fi,
+			       u32 tb_id, int flags)
+{
+	const struct fib_nh *nh;
+	__be16 eth_type = htons(ETH_P_IP);
+	__be32 dst_mask = inet_make_mask(dst_len);
+	__be16 internal_vlan_id = ofdpa_port->internal_vlan_id;
+	u32 priority = fi->fib_priority;
+	enum rocker_of_dpa_table_id goto_tbl =
+		ROCKER_OF_DPA_TABLE_ID_ACL_POLICY;
+	u32 group_id;
+	bool nh_on_port;
+	bool has_gw;
+	u32 index;
+	int err;
+
+	/* XXX support ECMP */
+
+	nh = fi->fib_nh;
+	nh_on_port = (fi->fib_dev == ofdpa_port->dev);
+	has_gw = !!nh->nh_gw;
+
+	if (has_gw && nh_on_port) {
+		err = ofdpa_port_ipv4_nh(ofdpa_port, trans, flags,
+					 nh->nh_gw, &index);
+		if (err)
+			return err;
+
+		group_id = ROCKER_GROUP_L3_UNICAST(index);
+	} else {
+		/* Send to CPU for processing */
+		group_id = ROCKER_GROUP_L2_INTERFACE(internal_vlan_id, 0);
+	}
+
+	err = ofdpa_flow_tbl_ucast4_routing(ofdpa_port, trans, eth_type, dst,
+					    dst_mask, priority, goto_tbl,
+					    group_id, flags);
+	if (err)
+		netdev_err(ofdpa_port->dev, "Error (%d) IPv4 route %pI4\n",
+			   err, &dst);
+
+	return err;
+}
+
+static void
+ofdpa_port_internal_vlan_id_put(const struct ofdpa_port *ofdpa_port,
+				int ifindex)
+{
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_internal_vlan_tbl_entry *found;
+	unsigned long lock_flags;
+	unsigned long bit;
+
+	spin_lock_irqsave(&ofdpa->internal_vlan_tbl_lock, lock_flags);
+
+	found = ofdpa_internal_vlan_tbl_find(ofdpa, ifindex);
+	if (!found) {
+		netdev_err(ofdpa_port->dev,
+			   "ifindex (%d) not found in internal VLAN tbl\n",
+			   ifindex);
+		goto not_found;
+	}
+
+	if (--found->ref_count <= 0) {
+		bit = ntohs(found->vlan_id) - OFDPA_INTERNAL_VLAN_ID_BASE;
+		clear_bit(bit, ofdpa->internal_vlan_bitmap);
+		hash_del(&found->entry);
+		kfree(found);
+	}
+
+not_found:
+	spin_unlock_irqrestore(&ofdpa->internal_vlan_tbl_lock, lock_flags);
+}
+
+/**********************************
+ * Rocker world ops implementation
+ **********************************/
+
+static int ofdpa_init(struct rocker *rocker)
+{
+	struct ofdpa *ofdpa = rocker->wpriv;
+
+	ofdpa->rocker = rocker;
+
+	hash_init(ofdpa->flow_tbl);
+	spin_lock_init(&ofdpa->flow_tbl_lock);
+
+	hash_init(ofdpa->group_tbl);
+	spin_lock_init(&ofdpa->group_tbl_lock);
+
+	hash_init(ofdpa->fdb_tbl);
+	spin_lock_init(&ofdpa->fdb_tbl_lock);
+
+	hash_init(ofdpa->internal_vlan_tbl);
+	spin_lock_init(&ofdpa->internal_vlan_tbl_lock);
+
+	hash_init(ofdpa->neigh_tbl);
+	spin_lock_init(&ofdpa->neigh_tbl_lock);
+
+	setup_timer(&ofdpa->fdb_cleanup_timer, ofdpa_fdb_cleanup,
+		    (unsigned long) ofdpa);
+	mod_timer(&ofdpa->fdb_cleanup_timer, jiffies);
+
+	return 0;
+}
+
+static void ofdpa_fini(struct rocker *rocker)
+{
+	struct ofdpa *ofdpa = rocker->wpriv;
+
+	unsigned long flags;
+	struct ofdpa_flow_tbl_entry *flow_entry;
+	struct ofdpa_group_tbl_entry *group_entry;
+	struct ofdpa_fdb_tbl_entry *fdb_entry;
+	struct ofdpa_internal_vlan_tbl_entry *internal_vlan_entry;
+	struct ofdpa_neigh_tbl_entry *neigh_entry;
+	struct hlist_node *tmp;
+	int bkt;
+
+	del_timer_sync(&ofdpa->fdb_cleanup_timer);
+
+	spin_lock_irqsave(&ofdpa->flow_tbl_lock, flags);
+	hash_for_each_safe(ofdpa->flow_tbl, bkt, tmp, flow_entry, entry)
+		hash_del(&flow_entry->entry);
+	spin_unlock_irqrestore(&ofdpa->flow_tbl_lock, flags);
+
+	spin_lock_irqsave(&ofdpa->group_tbl_lock, flags);
+	hash_for_each_safe(ofdpa->group_tbl, bkt, tmp, group_entry, entry)
+		hash_del(&group_entry->entry);
+	spin_unlock_irqrestore(&ofdpa->group_tbl_lock, flags);
+
+	spin_lock_irqsave(&ofdpa->fdb_tbl_lock, flags);
+	hash_for_each_safe(ofdpa->fdb_tbl, bkt, tmp, fdb_entry, entry)
+		hash_del(&fdb_entry->entry);
+	spin_unlock_irqrestore(&ofdpa->fdb_tbl_lock, flags);
+
+	spin_lock_irqsave(&ofdpa->internal_vlan_tbl_lock, flags);
+	hash_for_each_safe(ofdpa->internal_vlan_tbl, bkt,
+			   tmp, internal_vlan_entry, entry)
+		hash_del(&internal_vlan_entry->entry);
+	spin_unlock_irqrestore(&ofdpa->internal_vlan_tbl_lock, flags);
+
+	spin_lock_irqsave(&ofdpa->neigh_tbl_lock, flags);
+	hash_for_each_safe(ofdpa->neigh_tbl, bkt, tmp, neigh_entry, entry)
+		hash_del(&neigh_entry->entry);
+	spin_unlock_irqrestore(&ofdpa->neigh_tbl_lock, flags);
+}
+
+static int ofdpa_port_pre_init(struct rocker_port *rocker_port)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	ofdpa_port->ofdpa = rocker_port->rocker->wpriv;
+	ofdpa_port->rocker_port = rocker_port;
+	ofdpa_port->dev = rocker_port->dev;
+	ofdpa_port->pport = rocker_port->pport;
+	ofdpa_port->brport_flags = BR_LEARNING | BR_LEARNING_SYNC;
+	ofdpa_port->ageing_time = BR_DEFAULT_AGEING_TIME;
+	return 0;
+}
+
+static int ofdpa_port_init(struct rocker_port *rocker_port)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	int err;
+
+	switchdev_port_fwd_mark_set(ofdpa_port->dev, NULL, false);
+	rocker_port_set_learning(rocker_port,
+				 !!(ofdpa_port->brport_flags & BR_LEARNING));
+
+	err = ofdpa_port_ig_tbl(ofdpa_port, NULL, 0);
+	if (err) {
+		netdev_err(ofdpa_port->dev, "install ig port table failed\n");
+		return err;
+	}
+
+	ofdpa_port->internal_vlan_id =
+		ofdpa_port_internal_vlan_id_get(ofdpa_port,
+						ofdpa_port->dev->ifindex);
+
+	err = ofdpa_port_vlan_add(ofdpa_port, NULL, OFDPA_UNTAGGED_VID, 0);
+	if (err) {
+		netdev_err(ofdpa_port->dev, "install untagged VLAN failed\n");
+		goto err_untagged_vlan;
+	}
+	return 0;
+
+err_untagged_vlan:
+	ofdpa_port_ig_tbl(ofdpa_port, NULL, OFDPA_OP_FLAG_REMOVE);
+	return err;
+}
+
+static void ofdpa_port_fini(struct rocker_port *rocker_port)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	ofdpa_port_ig_tbl(ofdpa_port, NULL, OFDPA_OP_FLAG_REMOVE);
+}
+
+static int ofdpa_port_open(struct rocker_port *rocker_port)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	return ofdpa_port_fwd_enable(ofdpa_port, 0);
+}
+
+static void ofdpa_port_stop(struct rocker_port *rocker_port)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	ofdpa_port_fwd_disable(ofdpa_port, OFDPA_OP_FLAG_NOWAIT);
+}
+
+static int ofdpa_port_attr_stp_state_set(struct rocker_port *rocker_port,
+					 u8 state,
+					 struct switchdev_trans *trans)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	return ofdpa_port_stp_update(ofdpa_port, trans, 0, state);
+}
+
+static int ofdpa_port_attr_bridge_flags_set(struct rocker_port *rocker_port,
+					    unsigned long brport_flags,
+					    struct switchdev_trans *trans)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	unsigned long orig_flags;
+	int err = 0;
+
+	orig_flags = ofdpa_port->brport_flags;
+	ofdpa_port->brport_flags = brport_flags;
+	if ((orig_flags ^ ofdpa_port->brport_flags) & BR_LEARNING &&
+	    !switchdev_trans_ph_prepare(trans))
+		err = rocker_port_set_learning(ofdpa_port->rocker_port,
+					       !!(ofdpa_port->brport_flags & BR_LEARNING));
+
+	if (switchdev_trans_ph_prepare(trans))
+		ofdpa_port->brport_flags = orig_flags;
+
+	return err;
+}
+
+static int
+ofdpa_port_attr_bridge_flags_get(const struct rocker_port *rocker_port,
+				 unsigned long *p_brport_flags)
+{
+	const struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	*p_brport_flags = ofdpa_port->brport_flags;
+	return 0;
+}
+
+static int
+ofdpa_port_attr_bridge_ageing_time_set(struct rocker_port *rocker_port,
+				       u32 ageing_time,
+				       struct switchdev_trans *trans)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	if (!switchdev_trans_ph_prepare(trans)) {
+		ofdpa_port->ageing_time = clock_t_to_jiffies(ageing_time);
+		mod_timer(&ofdpa_port->ofdpa->fdb_cleanup_timer, jiffies);
+	}
+
+	return 0;
+}
+
+static int ofdpa_port_obj_vlan_add(struct rocker_port *rocker_port,
+				   const struct switchdev_obj_port_vlan *vlan,
+				   struct switchdev_trans *trans)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	u16 vid;
+	int err;
+
+	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
+		err = ofdpa_port_vlan_add(ofdpa_port, trans, vid, vlan->flags);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int ofdpa_port_obj_vlan_del(struct rocker_port *rocker_port,
+				   const struct switchdev_obj_port_vlan *vlan)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	u16 vid;
+	int err;
+
+	for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) {
+		err = ofdpa_port_vlan_del(ofdpa_port, vid, vlan->flags);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int ofdpa_port_obj_vlan_dump(const struct rocker_port *rocker_port,
+				    struct switchdev_obj_port_vlan *vlan,
+				    switchdev_obj_dump_cb_t *cb)
+{
+	const struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	u16 vid;
+	int err = 0;
+
+	for (vid = 1; vid < VLAN_N_VID; vid++) {
+		if (!test_bit(vid, ofdpa_port->vlan_bitmap))
+			continue;
+		vlan->flags = 0;
+		if (ofdpa_vlan_id_is_internal(htons(vid)))
+			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
+		vlan->vid_begin = vlan->vid_end = vid;
+		err = cb(&vlan->obj);
+		if (err)
+			break;
+	}
+
+	return err;
+}
+
+static int ofdpa_port_obj_fib4_add(struct rocker_port *rocker_port,
+				   const struct switchdev_obj_ipv4_fib *fib4,
+				   struct switchdev_trans *trans)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	return ofdpa_port_fib_ipv4(ofdpa_port, trans,
+				   htonl(fib4->dst), fib4->dst_len,
+				   &fib4->fi, fib4->tb_id, 0);
+}
+
+static int ofdpa_port_obj_fib4_del(struct rocker_port *rocker_port,
+				   const struct switchdev_obj_ipv4_fib *fib4)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+
+	return ofdpa_port_fib_ipv4(ofdpa_port, NULL,
+				   htonl(fib4->dst), fib4->dst_len,
+				   &fib4->fi, fib4->tb_id,
+				   OFDPA_OP_FLAG_REMOVE);
+}
+
+static int ofdpa_port_obj_fdb_add(struct rocker_port *rocker_port,
+				  const struct switchdev_obj_port_fdb *fdb,
+				  struct switchdev_trans *trans)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	__be16 vlan_id = ofdpa_port_vid_to_vlan(ofdpa_port, fdb->vid, NULL);
+
+	if (!ofdpa_port_is_bridged(ofdpa_port))
+		return -EINVAL;
+
+	return ofdpa_port_fdb(ofdpa_port, trans, fdb->addr, vlan_id, 0);
+}
+
+static int ofdpa_port_obj_fdb_del(struct rocker_port *rocker_port,
+				  const struct switchdev_obj_port_fdb *fdb)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	__be16 vlan_id = ofdpa_port_vid_to_vlan(ofdpa_port, fdb->vid, NULL);
+	int flags = OFDPA_OP_FLAG_REMOVE;
+
+	if (!ofdpa_port_is_bridged(ofdpa_port))
+		return -EINVAL;
+
+	return ofdpa_port_fdb(ofdpa_port, NULL, fdb->addr, vlan_id, flags);
+}
+
+static int ofdpa_port_obj_fdb_dump(const struct rocker_port *rocker_port,
+				   struct switchdev_obj_port_fdb *fdb,
+				   switchdev_obj_dump_cb_t *cb)
+{
+	const struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
+	struct ofdpa_fdb_tbl_entry *found;
+	struct hlist_node *tmp;
+	unsigned long lock_flags;
+	int bkt;
+	int err = 0;
+
+	spin_lock_irqsave(&ofdpa->fdb_tbl_lock, lock_flags);
+	hash_for_each_safe(ofdpa->fdb_tbl, bkt, tmp, found, entry) {
+		if (found->key.ofdpa_port != ofdpa_port)
+			continue;
+		ether_addr_copy(fdb->addr, found->key.addr);
+		fdb->ndm_state = NUD_REACHABLE;
+		fdb->vid = ofdpa_port_vlan_to_vid(ofdpa_port,
+						  found->key.vlan_id);
+		err = cb(&fdb->obj);
+		if (err)
+			break;
+	}
+	spin_unlock_irqrestore(&ofdpa->fdb_tbl_lock, lock_flags);
+
+	return err;
+}
+
+static int ofdpa_port_bridge_join(struct ofdpa_port *ofdpa_port,
+				  struct net_device *bridge)
+{
+	int err;
+
+	/* Port is joining bridge, so the internal VLAN for the
+	 * port is going to change to the bridge internal VLAN.
+	 * Let's remove untagged VLAN (vid=0) from port and
+	 * re-add once internal VLAN has changed.
+	 */
+
+	err = ofdpa_port_vlan_del(ofdpa_port, OFDPA_UNTAGGED_VID, 0);
+	if (err)
+		return err;
+
+	ofdpa_port_internal_vlan_id_put(ofdpa_port,
+					ofdpa_port->dev->ifindex);
+	ofdpa_port->internal_vlan_id =
+		ofdpa_port_internal_vlan_id_get(ofdpa_port, bridge->ifindex);
+
+	ofdpa_port->bridge_dev = bridge;
+	switchdev_port_fwd_mark_set(ofdpa_port->dev, bridge, true);
+
+	return ofdpa_port_vlan_add(ofdpa_port, NULL, OFDPA_UNTAGGED_VID, 0);
+}
+
+static int ofdpa_port_bridge_leave(struct ofdpa_port *ofdpa_port)
+{
+	int err;
+
+	err = ofdpa_port_vlan_del(ofdpa_port, OFDPA_UNTAGGED_VID, 0);
+	if (err)
+		return err;
+
+	ofdpa_port_internal_vlan_id_put(ofdpa_port,
+					ofdpa_port->bridge_dev->ifindex);
+	ofdpa_port->internal_vlan_id =
+		ofdpa_port_internal_vlan_id_get(ofdpa_port,
+						ofdpa_port->dev->ifindex);
+
+	switchdev_port_fwd_mark_set(ofdpa_port->dev, ofdpa_port->bridge_dev,
+				    false);
+	ofdpa_port->bridge_dev = NULL;
+
+	err = ofdpa_port_vlan_add(ofdpa_port, NULL, OFDPA_UNTAGGED_VID, 0);
+	if (err)
+		return err;
+
+	if (ofdpa_port->dev->flags & IFF_UP)
+		err = ofdpa_port_fwd_enable(ofdpa_port, 0);
+
+	return err;
+}
+
+static int ofdpa_port_ovs_changed(struct ofdpa_port *ofdpa_port,
+				  struct net_device *master)
+{
+	int err;
+
+	ofdpa_port->bridge_dev = master;
+
+	err = ofdpa_port_fwd_disable(ofdpa_port, 0);
+	if (err)
+		return err;
+	err = ofdpa_port_fwd_enable(ofdpa_port, 0);
+
+	return err;
+}
+
+static int ofdpa_port_master_linked(struct rocker_port *rocker_port,
+				    struct net_device *master)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	int err = 0;
+
+	if (netif_is_bridge_master(master))
+		err = ofdpa_port_bridge_join(ofdpa_port, master);
+	else if (netif_is_ovs_master(master))
+		err = ofdpa_port_ovs_changed(ofdpa_port, master);
+	return err;
+}
+
+static int ofdpa_port_master_unlinked(struct rocker_port *rocker_port,
+				      struct net_device *master)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	int err = 0;
+
+	if (ofdpa_port_is_bridged(ofdpa_port))
+		err = ofdpa_port_bridge_leave(ofdpa_port);
+	else if (ofdpa_port_is_ovsed(ofdpa_port))
+		err = ofdpa_port_ovs_changed(ofdpa_port, NULL);
+	return err;
+}
+
+static int ofdpa_port_neigh_update(struct rocker_port *rocker_port,
+				   struct neighbour *n)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	int flags = (n->nud_state & NUD_VALID ? 0 : OFDPA_OP_FLAG_REMOVE) |
+						    OFDPA_OP_FLAG_NOWAIT;
+	__be32 ip_addr = *(__be32 *) n->primary_key;
+
+	return ofdpa_port_ipv4_neigh(ofdpa_port, NULL, flags, ip_addr, n->ha);
+}
+
+static int ofdpa_port_neigh_destroy(struct rocker_port *rocker_port,
+				    struct neighbour *n)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	int flags = OFDPA_OP_FLAG_REMOVE | OFDPA_OP_FLAG_NOWAIT;
+	__be32 ip_addr = *(__be32 *) n->primary_key;
+
+	return ofdpa_port_ipv4_neigh(ofdpa_port, NULL, flags, ip_addr, n->ha);
+}
+
+static int ofdpa_port_ev_mac_vlan_seen(struct rocker_port *rocker_port,
+				       const unsigned char *addr,
+				       __be16 vlan_id)
+{
+	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	int flags = OFDPA_OP_FLAG_NOWAIT | OFDPA_OP_FLAG_LEARNED;
+
+	if (ofdpa_port->stp_state != BR_STATE_LEARNING &&
+	    ofdpa_port->stp_state != BR_STATE_FORWARDING)
+		return 0;
+
+	return ofdpa_port_fdb(ofdpa_port, NULL, addr, vlan_id, flags);
+}
+
 struct rocker_world_ops rocker_ofdpa_ops = {
 	.kind = "ofdpa",
 	.priv_size = sizeof(struct ofdpa),
 	.port_priv_size = sizeof(struct ofdpa_port),
 	.mode = ROCKER_PORT_MODE_OF_DPA,
+	.init = ofdpa_init,
+	.fini = ofdpa_fini,
+	.port_pre_init = ofdpa_port_pre_init,
+	.port_init = ofdpa_port_init,
+	.port_fini = ofdpa_port_fini,
+	.port_open = ofdpa_port_open,
+	.port_stop = ofdpa_port_stop,
+	.port_attr_stp_state_set = ofdpa_port_attr_stp_state_set,
+	.port_attr_bridge_flags_set = ofdpa_port_attr_bridge_flags_set,
+	.port_attr_bridge_flags_get = ofdpa_port_attr_bridge_flags_get,
+	.port_attr_bridge_ageing_time_set = ofdpa_port_attr_bridge_ageing_time_set,
+	.port_obj_vlan_add = ofdpa_port_obj_vlan_add,
+	.port_obj_vlan_del = ofdpa_port_obj_vlan_del,
+	.port_obj_vlan_dump = ofdpa_port_obj_vlan_dump,
+	.port_obj_fib4_add = ofdpa_port_obj_fib4_add,
+	.port_obj_fib4_del = ofdpa_port_obj_fib4_del,
+	.port_obj_fdb_add = ofdpa_port_obj_fdb_add,
+	.port_obj_fdb_del = ofdpa_port_obj_fdb_del,
+	.port_obj_fdb_dump = ofdpa_port_obj_fdb_dump,
+	.port_master_linked = ofdpa_port_master_linked,
+	.port_master_unlinked = ofdpa_port_master_unlinked,
+	.port_neigh_update = ofdpa_port_neigh_update,
+	.port_neigh_destroy = ofdpa_port_neigh_destroy,
+	.port_ev_mac_vlan_seen = ofdpa_port_ev_mac_vlan_seen,
 };
-- 
cgit v0.10.2


From fccd84d44912a15bea4a2265f5437f18330e7ce1 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@resnulli.us>
Date: Tue, 16 Feb 2016 15:14:51 +0100
Subject: rocker: return -EOPNOTSUPP for undefined world ops

Suggested-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 1efd8b7..acafbf8 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -1547,7 +1547,7 @@ static int rocker_world_port_attr_stp_state_set(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_attr_stp_state_set)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_attr_stp_state_set(rocker_port, state, trans);
 }
 
@@ -1559,7 +1559,7 @@ rocker_world_port_attr_bridge_flags_set(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_attr_bridge_flags_set)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_attr_bridge_flags_set(rocker_port, brport_flags,
 						trans);
 }
@@ -1571,7 +1571,7 @@ rocker_world_port_attr_bridge_flags_get(const struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_attr_bridge_flags_get)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_attr_bridge_flags_get(rocker_port, p_brport_flags);
 }
 
@@ -1584,7 +1584,7 @@ rocker_world_port_attr_bridge_ageing_time_set(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_attr_bridge_ageing_time_set)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_attr_bridge_ageing_time_set(rocker_port, ageing_time,
 						      trans);
 }
@@ -1597,6 +1597,7 @@ rocker_world_port_obj_vlan_add(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_vlan_add)
+		return -EOPNOTSUPP;
 		return 0;
 	return wops->port_obj_vlan_add(rocker_port, vlan, trans);
 }
@@ -1608,7 +1609,7 @@ rocker_world_port_obj_vlan_del(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_vlan_del)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_vlan_del(rocker_port, vlan);
 }
 
@@ -1620,7 +1621,7 @@ rocker_world_port_obj_vlan_dump(const struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_vlan_dump)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_vlan_dump(rocker_port, vlan, cb);
 }
 
@@ -1632,7 +1633,7 @@ rocker_world_port_obj_fib4_add(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_fib4_add)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_fib4_add(rocker_port, fib4, trans);
 }
 
@@ -1643,7 +1644,7 @@ rocker_world_port_obj_fib4_del(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_fib4_del)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_fib4_del(rocker_port, fib4);
 }
 
@@ -1655,7 +1656,7 @@ rocker_world_port_obj_fdb_add(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_fdb_add)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_fdb_add(rocker_port, fdb, trans);
 }
 
@@ -1666,7 +1667,7 @@ rocker_world_port_obj_fdb_del(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_fdb_del)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_fdb_del(rocker_port, fdb);
 }
 
@@ -1678,7 +1679,7 @@ rocker_world_port_obj_fdb_dump(const struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_obj_fdb_dump)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_obj_fdb_dump(rocker_port, fdb, cb);
 }
 
@@ -1688,7 +1689,7 @@ static int rocker_world_port_master_linked(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_master_linked)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_master_linked(rocker_port, master);
 }
 
@@ -1698,7 +1699,7 @@ static int rocker_world_port_master_unlinked(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_master_unlinked)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_master_unlinked(rocker_port, master);
 }
 
@@ -1708,7 +1709,7 @@ static int rocker_world_port_neigh_update(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_neigh_update)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_neigh_update(rocker_port, n);
 }
 
@@ -1718,7 +1719,7 @@ static int rocker_world_port_neigh_destroy(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_neigh_destroy)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_neigh_destroy(rocker_port, n);
 }
 
@@ -1729,7 +1730,7 @@ static int rocker_world_port_ev_mac_vlan_seen(struct rocker_port *rocker_port,
 	struct rocker_world_ops *wops = rocker_port->rocker->wops;
 
 	if (!wops->port_ev_mac_vlan_seen)
-		return 0;
+		return -EOPNOTSUPP;
 	return wops->port_ev_mac_vlan_seen(rocker_port, addr, vlan_id);
 }
 
-- 
cgit v0.10.2


From d4ac05ff3697e036dcb0e2e284c5f7eb77cc0966 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:58:57 +0100
Subject: vxlan: introduce vxlan_hdr

Currently, pointer to the vxlan header is kept in a local variable. It has
to be reloaded whenever the pskb pull operations are performed which usually
happens somewhere deep in called functions.

Create a vxlan_hdr function and use it to reference the vxlan header
instead.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index ee1206d..524e3b1 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1257,7 +1257,6 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct metadata_dst *tun_dst = NULL;
 	struct vxlan_sock *vs;
-	struct vxlanhdr *vxh;
 	u32 flags, vni;
 	struct vxlan_metadata _md;
 	struct vxlan_metadata *md = &_md;
@@ -1266,9 +1265,8 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!pskb_may_pull(skb, VXLAN_HLEN))
 		goto error;
 
-	vxh = (struct vxlanhdr *)(udp_hdr(skb) + 1);
-	flags = ntohl(vxh->vx_flags);
-	vni = ntohl(vxh->vx_vni);
+	flags = ntohl(vxlan_hdr(skb)->vx_flags);
+	vni = ntohl(vxlan_hdr(skb)->vx_vni);
 
 	if (flags & VXLAN_HF_VNI) {
 		flags &= ~VXLAN_HF_VNI;
@@ -1279,16 +1277,14 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 
 	if (iptunnel_pull_header(skb, VXLAN_HLEN, htons(ETH_P_TEB)))
 		goto drop;
-	vxh = (struct vxlanhdr *)(udp_hdr(skb) + 1);
 
 	vs = rcu_dereference_sk_user_data(sk);
 	if (!vs)
 		goto drop;
 
 	if ((flags & VXLAN_HF_RCO) && (vs->flags & VXLAN_F_REMCSUM_RX)) {
-		vxh = vxlan_remcsum(skb, vxh, sizeof(struct vxlanhdr), vni,
-				    !!(vs->flags & VXLAN_F_REMCSUM_NOPARTIAL));
-		if (!vxh)
+		if (!vxlan_remcsum(skb, vxlan_hdr(skb), sizeof(struct vxlanhdr), vni,
+				   !!(vs->flags & VXLAN_F_REMCSUM_NOPARTIAL)))
 			goto drop;
 
 		flags &= ~VXLAN_HF_RCO;
@@ -1313,7 +1309,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if ((flags & VXLAN_HF_GBP) && (vs->flags & VXLAN_F_GBP)) {
 		struct vxlanhdr_gbp *gbp;
 
-		gbp = (struct vxlanhdr_gbp *)vxh;
+		gbp = (struct vxlanhdr_gbp *)vxlan_hdr(skb);
 		md->gbp = ntohs(gbp->policy_id);
 
 		if (tun_dst)
@@ -1351,7 +1347,8 @@ drop:
 
 bad_flags:
 	netdev_dbg(skb->dev, "invalid vxlan flags=%#x vni=%#x\n",
-		   ntohl(vxh->vx_flags), ntohl(vxh->vx_vni));
+		   ntohl(vxlan_hdr(skb)->vx_flags),
+		   ntohl(vxlan_hdr(skb)->vx_vni));
 
 error:
 	if (tun_dst)
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index b314e4a..3f38b40 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -262,6 +262,11 @@ static inline netdev_features_t vxlan_features_check(struct sk_buff *skb,
 /* IPv6 header + UDP + VXLAN + Ethernet header */
 #define VXLAN6_HEADROOM (40 + 8 + 8 + 14)
 
+static inline struct vxlanhdr *vxlan_hdr(struct sk_buff *skb)
+{
+	return (struct vxlanhdr *)(udp_hdr(skb) + 1);
+}
+
 #if IS_ENABLED(CONFIG_VXLAN)
 void vxlan_get_rx_port(struct net_device *netdev);
 #else
-- 
cgit v0.10.2


From 54bfd872bf16d40b61bd0cd9b769b2fef67dd272 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:58:58 +0100
Subject: vxlan: keep flags and vni in network byte order

Prevent repeated conversions from and to network order in the fast path.

To achieve this, define all flag constants in big endian order and store VNI
as __be32. To prevent confusion between the actual VNI value and the VNI
field from the header (which contains additional reserved byte), strictly
distinguish between "vni" and "vni_field".

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 524e3b1..4e3d3df 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -197,9 +197,9 @@ static int vxlan_nla_put_addr(struct sk_buff *skb, int attr,
 #endif
 
 /* Virtual Network hash table head */
-static inline struct hlist_head *vni_head(struct vxlan_sock *vs, u32 id)
+static inline struct hlist_head *vni_head(struct vxlan_sock *vs, __be32 vni)
 {
-	return &vs->vni_list[hash_32(id, VNI_HASH_BITS)];
+	return &vs->vni_list[hash_32((__force u32)vni, VNI_HASH_BITS)];
 }
 
 /* Socket hash table head */
@@ -242,12 +242,12 @@ static struct vxlan_sock *vxlan_find_sock(struct net *net, sa_family_t family,
 	return NULL;
 }
 
-static struct vxlan_dev *vxlan_vs_find_vni(struct vxlan_sock *vs, u32 id)
+static struct vxlan_dev *vxlan_vs_find_vni(struct vxlan_sock *vs, __be32 vni)
 {
 	struct vxlan_dev *vxlan;
 
-	hlist_for_each_entry_rcu(vxlan, vni_head(vs, id), hlist) {
-		if (vxlan->default_dst.remote_vni == id)
+	hlist_for_each_entry_rcu(vxlan, vni_head(vs, vni), hlist) {
+		if (vxlan->default_dst.remote_vni == vni)
 			return vxlan;
 	}
 
@@ -255,7 +255,7 @@ static struct vxlan_dev *vxlan_vs_find_vni(struct vxlan_sock *vs, u32 id)
 }
 
 /* Look up VNI in a per net namespace table */
-static struct vxlan_dev *vxlan_find_vni(struct net *net, u32 id,
+static struct vxlan_dev *vxlan_find_vni(struct net *net, __be32 vni,
 					sa_family_t family, __be16 port,
 					u32 flags)
 {
@@ -265,7 +265,7 @@ static struct vxlan_dev *vxlan_find_vni(struct net *net, u32 id,
 	if (!vs)
 		return NULL;
 
-	return vxlan_vs_find_vni(vs, id);
+	return vxlan_vs_find_vni(vs, vni);
 }
 
 /* Fill in neighbour message in skbuff. */
@@ -315,7 +315,7 @@ static int vxlan_fdb_info(struct sk_buff *skb, struct vxlan_dev *vxlan,
 	    nla_put_be16(skb, NDA_PORT, rdst->remote_port))
 		goto nla_put_failure;
 	if (rdst->remote_vni != vxlan->default_dst.remote_vni &&
-	    nla_put_u32(skb, NDA_VNI, rdst->remote_vni))
+	    nla_put_u32(skb, NDA_VNI, be32_to_cpu(rdst->remote_vni)))
 		goto nla_put_failure;
 	if (rdst->remote_ifindex &&
 	    nla_put_u32(skb, NDA_IFINDEX, rdst->remote_ifindex))
@@ -383,7 +383,7 @@ static void vxlan_ip_miss(struct net_device *dev, union vxlan_addr *ipa)
 	};
 	struct vxlan_rdst remote = {
 		.remote_ip = *ipa, /* goes to NDA_DST */
-		.remote_vni = VXLAN_N_VID,
+		.remote_vni = cpu_to_be32(VXLAN_N_VID),
 	};
 
 	vxlan_fdb_notify(vxlan, &f, &remote, RTM_GETNEIGH);
@@ -452,7 +452,7 @@ static struct vxlan_fdb *vxlan_find_mac(struct vxlan_dev *vxlan,
 /* caller should hold vxlan->hash_lock */
 static struct vxlan_rdst *vxlan_fdb_find_rdst(struct vxlan_fdb *f,
 					      union vxlan_addr *ip, __be16 port,
-					      __u32 vni, __u32 ifindex)
+					      __be32 vni, __u32 ifindex)
 {
 	struct vxlan_rdst *rd;
 
@@ -469,7 +469,8 @@ static struct vxlan_rdst *vxlan_fdb_find_rdst(struct vxlan_fdb *f,
 
 /* Replace destination of unicast mac */
 static int vxlan_fdb_replace(struct vxlan_fdb *f,
-			     union vxlan_addr *ip, __be16 port, __u32 vni, __u32 ifindex)
+			     union vxlan_addr *ip, __be16 port, __be32 vni,
+			     __u32 ifindex)
 {
 	struct vxlan_rdst *rd;
 
@@ -491,7 +492,7 @@ static int vxlan_fdb_replace(struct vxlan_fdb *f,
 
 /* Add/update destinations for multicast */
 static int vxlan_fdb_append(struct vxlan_fdb *f,
-			    union vxlan_addr *ip, __be16 port, __u32 vni,
+			    union vxlan_addr *ip, __be16 port, __be32 vni,
 			    __u32 ifindex, struct vxlan_rdst **rdp)
 {
 	struct vxlan_rdst *rd;
@@ -523,7 +524,8 @@ static int vxlan_fdb_append(struct vxlan_fdb *f,
 static struct vxlanhdr *vxlan_gro_remcsum(struct sk_buff *skb,
 					  unsigned int off,
 					  struct vxlanhdr *vh, size_t hdrlen,
-					  u32 data, struct gro_remcsum *grc,
+					  __be32 vni_field,
+					  struct gro_remcsum *grc,
 					  bool nopartial)
 {
 	size_t start, offset;
@@ -534,10 +536,8 @@ static struct vxlanhdr *vxlan_gro_remcsum(struct sk_buff *skb,
 	if (!NAPI_GRO_CB(skb)->csum_valid)
 		return NULL;
 
-	start = (data & VXLAN_RCO_MASK) << VXLAN_RCO_SHIFT;
-	offset = start + ((data & VXLAN_RCO_UDP) ?
-			  offsetof(struct udphdr, check) :
-			  offsetof(struct tcphdr, check));
+	start = vxlan_rco_start(vni_field);
+	offset = start + vxlan_rco_offset(vni_field);
 
 	vh = skb_gro_remcsum_process(skb, (void *)vh, off, hdrlen,
 				     start, offset, grc, nopartial);
@@ -557,7 +557,7 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head,
 	int flush = 1;
 	struct vxlan_sock *vs = container_of(uoff, struct vxlan_sock,
 					     udp_offloads);
-	u32 flags;
+	__be32 flags;
 	struct gro_remcsum grc;
 
 	skb_gro_remcsum_init(&grc);
@@ -573,11 +573,11 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head,
 
 	skb_gro_postpull_rcsum(skb, vh, sizeof(struct vxlanhdr));
 
-	flags = ntohl(vh->vx_flags);
+	flags = vh->vx_flags;
 
 	if ((flags & VXLAN_HF_RCO) && (vs->flags & VXLAN_F_REMCSUM_RX)) {
 		vh = vxlan_gro_remcsum(skb, off_vx, vh, sizeof(struct vxlanhdr),
-				       ntohl(vh->vx_vni), &grc,
+				       vh->vx_vni, &grc,
 				       !!(vs->flags &
 					  VXLAN_F_REMCSUM_NOPARTIAL));
 
@@ -668,7 +668,7 @@ static void vxlan_notify_del_rx_port(struct vxlan_sock *vs)
 static int vxlan_fdb_create(struct vxlan_dev *vxlan,
 			    const u8 *mac, union vxlan_addr *ip,
 			    __u16 state, __u16 flags,
-			    __be16 port, __u32 vni, __u32 ifindex,
+			    __be16 port, __be32 vni, __u32 ifindex,
 			    __u8 ndm_flags)
 {
 	struct vxlan_rdst *rd = NULL;
@@ -777,7 +777,8 @@ static void vxlan_fdb_destroy(struct vxlan_dev *vxlan, struct vxlan_fdb *f)
 }
 
 static int vxlan_fdb_parse(struct nlattr *tb[], struct vxlan_dev *vxlan,
-			   union vxlan_addr *ip, __be16 *port, u32 *vni, u32 *ifindex)
+			   union vxlan_addr *ip, __be16 *port, __be32 *vni,
+			   u32 *ifindex)
 {
 	struct net *net = dev_net(vxlan->dev);
 	int err;
@@ -810,7 +811,7 @@ static int vxlan_fdb_parse(struct nlattr *tb[], struct vxlan_dev *vxlan,
 	if (tb[NDA_VNI]) {
 		if (nla_len(tb[NDA_VNI]) != sizeof(u32))
 			return -EINVAL;
-		*vni = nla_get_u32(tb[NDA_VNI]);
+		*vni = cpu_to_be32(nla_get_u32(tb[NDA_VNI]));
 	} else {
 		*vni = vxlan->default_dst.remote_vni;
 	}
@@ -840,7 +841,8 @@ static int vxlan_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
 	/* struct net *net = dev_net(vxlan->dev); */
 	union vxlan_addr ip;
 	__be16 port;
-	u32 vni, ifindex;
+	__be32 vni;
+	u32 ifindex;
 	int err;
 
 	if (!(ndm->ndm_state & (NUD_PERMANENT|NUD_REACHABLE))) {
@@ -877,7 +879,8 @@ static int vxlan_fdb_delete(struct ndmsg *ndm, struct nlattr *tb[],
 	struct vxlan_rdst *rd = NULL;
 	union vxlan_addr ip;
 	__be16 port;
-	u32 vni, ifindex;
+	__be32 vni;
+	u32 ifindex;
 	int err;
 
 	err = vxlan_fdb_parse(tb, vxlan, &ip, &port, &vni, &ifindex);
@@ -1133,17 +1136,16 @@ static int vxlan_igmp_leave(struct vxlan_dev *vxlan)
 }
 
 static struct vxlanhdr *vxlan_remcsum(struct sk_buff *skb, struct vxlanhdr *vh,
-				      size_t hdrlen, u32 data, bool nopartial)
+				      size_t hdrlen, __be32 vni_field,
+				      bool nopartial)
 {
 	size_t start, offset, plen;
 
 	if (skb->remcsum_offload)
 		return vh;
 
-	start = (data & VXLAN_RCO_MASK) << VXLAN_RCO_SHIFT;
-	offset = start + ((data & VXLAN_RCO_UDP) ?
-			  offsetof(struct udphdr, check) :
-			  offsetof(struct tcphdr, check));
+	start = vxlan_rco_start(vni_field);
+	offset = start + vxlan_rco_offset(vni_field);
 
 	plen = hdrlen + offset + sizeof(u16);
 
@@ -1159,7 +1161,7 @@ static struct vxlanhdr *vxlan_remcsum(struct sk_buff *skb, struct vxlanhdr *vh,
 }
 
 static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
-		      struct vxlan_metadata *md, u32 vni,
+		      struct vxlan_metadata *md, __be32 vni,
 		      struct metadata_dst *tun_dst)
 {
 	struct iphdr *oip = NULL;
@@ -1257,7 +1259,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct metadata_dst *tun_dst = NULL;
 	struct vxlan_sock *vs;
-	u32 flags, vni;
+	__be32 flags, vni_field;
 	struct vxlan_metadata _md;
 	struct vxlan_metadata *md = &_md;
 
@@ -1265,8 +1267,8 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!pskb_may_pull(skb, VXLAN_HLEN))
 		goto error;
 
-	flags = ntohl(vxlan_hdr(skb)->vx_flags);
-	vni = ntohl(vxlan_hdr(skb)->vx_vni);
+	flags = vxlan_hdr(skb)->vx_flags;
+	vni_field = vxlan_hdr(skb)->vx_vni;
 
 	if (flags & VXLAN_HF_VNI) {
 		flags &= ~VXLAN_HF_VNI;
@@ -1283,17 +1285,18 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto drop;
 
 	if ((flags & VXLAN_HF_RCO) && (vs->flags & VXLAN_F_REMCSUM_RX)) {
-		if (!vxlan_remcsum(skb, vxlan_hdr(skb), sizeof(struct vxlanhdr), vni,
+		if (!vxlan_remcsum(skb, vxlan_hdr(skb), sizeof(struct vxlanhdr),
+				   vni_field,
 				   !!(vs->flags & VXLAN_F_REMCSUM_NOPARTIAL)))
 			goto drop;
 
 		flags &= ~VXLAN_HF_RCO;
-		vni &= VXLAN_VNI_MASK;
+		vni_field &= VXLAN_VNI_MASK;
 	}
 
 	if (vxlan_collect_metadata(vs)) {
 		tun_dst = udp_tun_rx_dst(skb, vxlan_get_sk_family(vs), TUNNEL_KEY,
-					 cpu_to_be64(vni >> 8), sizeof(*md));
+					 vxlan_vni(vni_field), sizeof(*md));
 
 		if (!tun_dst)
 			goto drop;
@@ -1324,7 +1327,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		flags &= ~VXLAN_GBP_USED_BITS;
 	}
 
-	if (flags || vni & ~VXLAN_VNI_MASK) {
+	if (flags || vni_field & ~VXLAN_VNI_MASK) {
 		/* If there are any unprocessed flags remaining treat
 		 * this as a malformed packet. This behavior diverges from
 		 * VXLAN RFC (RFC7348) which stipulates that bits in reserved
@@ -1337,7 +1340,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto bad_flags;
 	}
 
-	vxlan_rcv(vs, skb, md, vni >> 8, tun_dst);
+	vxlan_rcv(vs, skb, md, vxlan_vni(vni_field), tun_dst);
 	return 0;
 
 drop:
@@ -1680,7 +1683,7 @@ static void vxlan_build_gbp_hdr(struct vxlanhdr *vxh, u32 vxflags,
 		return;
 
 	gbp = (struct vxlanhdr_gbp *)vxh;
-	vxh->vx_flags |= htonl(VXLAN_HF_GBP);
+	vxh->vx_flags |= VXLAN_HF_GBP;
 
 	if (md->gbp & VXLAN_GBP_DONT_LEARN)
 		gbp->dont_learn = 1;
@@ -1700,7 +1703,6 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 	int min_headroom;
 	int err;
 	int type = udp_sum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
-	u16 hdrlen = sizeof(struct vxlanhdr);
 
 	if ((vxflags & VXLAN_F_REMCSUM_TX) &&
 	    skb->ip_summed == CHECKSUM_PARTIAL) {
@@ -1733,18 +1735,15 @@ static int vxlan_build_skb(struct sk_buff *skb, struct dst_entry *dst,
 		return PTR_ERR(skb);
 
 	vxh = (struct vxlanhdr *) __skb_push(skb, sizeof(*vxh));
-	vxh->vx_flags = htonl(VXLAN_HF_VNI);
-	vxh->vx_vni = vni;
+	vxh->vx_flags = VXLAN_HF_VNI;
+	vxh->vx_vni = vxlan_vni_field(vni);
 
 	if (type & SKB_GSO_TUNNEL_REMCSUM) {
-		u32 data = (skb_checksum_start_offset(skb) - hdrlen) >>
-			   VXLAN_RCO_SHIFT;
+		unsigned int start;
 
-		if (skb->csum_offset == offsetof(struct udphdr, check))
-			data |= VXLAN_RCO_UDP;
-
-		vxh->vx_vni |= htonl(data);
-		vxh->vx_flags |= htonl(VXLAN_HF_RCO);
+		start = skb_checksum_start_offset(skb) - sizeof(struct vxlanhdr);
+		vxh->vx_vni |= vxlan_compute_rco(start, skb->csum_offset);
+		vxh->vx_flags |= VXLAN_HF_RCO;
 
 		if (!skb_is_gso(skb)) {
 			skb->ip_summed = CHECKSUM_NONE;
@@ -1892,7 +1891,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	struct vxlan_metadata _md;
 	struct vxlan_metadata *md = &_md;
 	__be16 src_port = 0, dst_port;
-	u32 vni;
+	__be32 vni;
 	__be16 df = 0;
 	__u8 tos, ttl;
 	int err;
@@ -1914,7 +1913,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			goto drop;
 		}
 		dst_port = info->key.tp_dst ? : vxlan->cfg.dst_port;
-		vni = be64_to_cpu(info->key.tun_id);
+		vni = vxlan_tun_id_to_vni(info->key.tun_id);
 		remote_ip.sa.sa_family = ip_tunnel_info_af(info);
 		if (remote_ip.sa.sa_family == AF_INET)
 			remote_ip.sin.sin_addr.s_addr = info->key.u.ipv4.dst;
@@ -2007,7 +2006,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		tos = ip_tunnel_ecn_encap(tos, old_iph, skb);
 		ttl = ttl ? : ip4_dst_hoplimit(&rt->dst);
 		err = vxlan_build_skb(skb, &rt->dst, sizeof(struct iphdr),
-				      htonl(vni << 8), md, flags, udp_sum);
+				      vni, md, flags, udp_sum);
 		if (err < 0)
 			goto xmit_tx_error;
 
@@ -2065,7 +2064,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		ttl = ttl ? : ip6_dst_hoplimit(ndst);
 		skb_scrub_packet(skb, xnet);
 		err = vxlan_build_skb(skb, ndst, sizeof(struct ipv6hdr),
-				      htonl(vni << 8), md, flags, udp_sum);
+				      vni, md, flags, udp_sum);
 		if (err < 0) {
 			dst_release(ndst);
 			return;
@@ -2222,7 +2221,7 @@ static void vxlan_cleanup(unsigned long arg)
 static void vxlan_vs_add_dev(struct vxlan_sock *vs, struct vxlan_dev *vxlan)
 {
 	struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
-	__u32 vni = vxlan->default_dst.remote_vni;
+	__be32 vni = vxlan->default_dst.remote_vni;
 
 	spin_lock(&vn->sock_lock);
 	hlist_add_head_rcu(&vxlan->hlist, vni_head(vs, vni));
@@ -2837,7 +2836,7 @@ static int vxlan_newlink(struct net *src_net, struct net_device *dev,
 	memset(&conf, 0, sizeof(conf));
 
 	if (data[IFLA_VXLAN_ID])
-		conf.vni = nla_get_u32(data[IFLA_VXLAN_ID]);
+		conf.vni = cpu_to_be32(nla_get_u32(data[IFLA_VXLAN_ID]));
 
 	if (data[IFLA_VXLAN_GROUP]) {
 		conf.remote_ip.sin.sin_addr.s_addr = nla_get_in_addr(data[IFLA_VXLAN_GROUP]);
@@ -2941,7 +2940,7 @@ static int vxlan_newlink(struct net *src_net, struct net_device *dev,
 		break;
 
 	case -EEXIST:
-		pr_info("duplicate VNI %u\n", conf.vni);
+		pr_info("duplicate VNI %u\n", be32_to_cpu(conf.vni));
 		break;
 	}
 
@@ -2999,7 +2998,7 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
 		.high = htons(vxlan->cfg.port_max),
 	};
 
-	if (nla_put_u32(skb, IFLA_VXLAN_ID, dst->remote_vni))
+	if (nla_put_u32(skb, IFLA_VXLAN_ID, be32_to_cpu(dst->remote_vni)))
 		goto nla_put_failure;
 
 	if (!vxlan_addr_any(&dst->remote_ip)) {
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 3f38b40..1b85a3b 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -24,11 +24,11 @@ struct vxlanhdr {
 };
 
 /* VXLAN header flags. */
-#define VXLAN_HF_VNI BIT(27)
+#define VXLAN_HF_VNI	cpu_to_be32(BIT(27))
 
 #define VXLAN_N_VID     (1u << 24)
 #define VXLAN_VID_MASK  (VXLAN_N_VID - 1)
-#define VXLAN_VNI_MASK  (VXLAN_VID_MASK << 8)
+#define VXLAN_VNI_MASK	cpu_to_be32(VXLAN_VID_MASK << 8)
 #define VXLAN_HLEN (sizeof(struct udphdr) + sizeof(struct vxlanhdr))
 
 #define VNI_HASH_BITS	10
@@ -55,14 +55,14 @@ struct vxlanhdr {
  */
 
 /* VXLAN-RCO header flags. */
-#define VXLAN_HF_RCO BIT(21)
+#define VXLAN_HF_RCO	cpu_to_be32(BIT(21))
 
 /* Remote checksum offload header option */
-#define VXLAN_RCO_MASK  0x7f    /* Last byte of vni field */
-#define VXLAN_RCO_UDP   0x80    /* Indicate UDP RCO (TCP when not set *) */
-#define VXLAN_RCO_SHIFT 1       /* Left shift of start */
+#define VXLAN_RCO_MASK	cpu_to_be32(0x7f)  /* Last byte of vni field */
+#define VXLAN_RCO_UDP	cpu_to_be32(0x80)  /* Indicate UDP RCO (TCP when not set *) */
+#define VXLAN_RCO_SHIFT	1		   /* Left shift of start */
 #define VXLAN_RCO_SHIFT_MASK ((1 << VXLAN_RCO_SHIFT) - 1)
-#define VXLAN_MAX_REMCSUM_START (VXLAN_RCO_MASK << VXLAN_RCO_SHIFT)
+#define VXLAN_MAX_REMCSUM_START (0x7f << VXLAN_RCO_SHIFT)
 
 /*
  * VXLAN Group Based Policy Extension (VXLAN_F_GBP):
@@ -105,9 +105,9 @@ struct vxlanhdr_gbp {
 };
 
 /* VXLAN-GBP header flags. */
-#define VXLAN_HF_GBP BIT(31)
+#define VXLAN_HF_GBP	cpu_to_be32(BIT(31))
 
-#define VXLAN_GBP_USED_BITS (VXLAN_HF_GBP | 0xFFFFFF)
+#define VXLAN_GBP_USED_BITS (VXLAN_HF_GBP | cpu_to_be32(0xFFFFFF))
 
 /* skb->mark mapping
  *
@@ -144,7 +144,7 @@ union vxlan_addr {
 struct vxlan_rdst {
 	union vxlan_addr	 remote_ip;
 	__be16			 remote_port;
-	u32			 remote_vni;
+	__be32			 remote_vni;
 	u32			 remote_ifindex;
 	struct list_head	 list;
 	struct rcu_head		 rcu;
@@ -154,7 +154,7 @@ struct vxlan_rdst {
 struct vxlan_config {
 	union vxlan_addr	remote_ip;
 	union vxlan_addr	saddr;
-	u32			vni;
+	__be32			vni;
 	int			remote_ifindex;
 	int			mtu;
 	__be16			dst_port;
@@ -267,6 +267,54 @@ static inline struct vxlanhdr *vxlan_hdr(struct sk_buff *skb)
 	return (struct vxlanhdr *)(udp_hdr(skb) + 1);
 }
 
+static inline __be32 vxlan_vni(__be32 vni_field)
+{
+#if defined(__BIG_ENDIAN)
+	return vni_field >> 8;
+#else
+	return (vni_field & VXLAN_VNI_MASK) << 8;
+#endif
+}
+
+static inline __be32 vxlan_vni_field(__be32 vni)
+{
+#if defined(__BIG_ENDIAN)
+	return vni << 8;
+#else
+	return vni >> 8;
+#endif
+}
+
+static inline __be32 vxlan_tun_id_to_vni(__be64 tun_id)
+{
+#if defined(__BIG_ENDIAN)
+	return tun_id;
+#else
+	return tun_id >> 32;
+#endif
+}
+
+static inline size_t vxlan_rco_start(__be32 vni_field)
+{
+	return be32_to_cpu(vni_field & VXLAN_RCO_MASK) << VXLAN_RCO_SHIFT;
+}
+
+static inline size_t vxlan_rco_offset(__be32 vni_field)
+{
+	return (vni_field & VXLAN_RCO_UDP) ?
+		offsetof(struct udphdr, check) :
+		offsetof(struct tcphdr, check);
+}
+
+static inline __be32 vxlan_compute_rco(unsigned int start, unsigned int offset)
+{
+	__be32 vni_field = cpu_to_be32(start >> VXLAN_RCO_SHIFT);
+
+	if (offset == offsetof(struct udphdr, check))
+		vni_field |= VXLAN_RCO_UDP;
+	return vni_field;
+}
+
 #if IS_ENABLED(CONFIG_VXLAN)
 void vxlan_get_rx_port(struct net_device *netdev);
 #else
-- 
cgit v0.10.2


From be5cfeab8f95995d5590ab919b9f4dde19d50ea7 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:58:59 +0100
Subject: vxlan: simplify vxlan_remcsum

Part of the parameters is not needed. Simplify the caller of this function
in preparation of making vxlan rx more comprehensible.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 4e3d3df..161e39c 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1135,29 +1135,25 @@ static int vxlan_igmp_leave(struct vxlan_dev *vxlan)
 	return ret;
 }
 
-static struct vxlanhdr *vxlan_remcsum(struct sk_buff *skb, struct vxlanhdr *vh,
-				      size_t hdrlen, __be32 vni_field,
-				      bool nopartial)
+static bool vxlan_remcsum(struct sk_buff *skb, u32 vxflags, __be32 vni_field)
 {
 	size_t start, offset, plen;
 
 	if (skb->remcsum_offload)
-		return vh;
+		return true;
 
 	start = vxlan_rco_start(vni_field);
 	offset = start + vxlan_rco_offset(vni_field);
 
-	plen = hdrlen + offset + sizeof(u16);
+	plen = sizeof(struct vxlanhdr) + offset + sizeof(u16);
 
 	if (!pskb_may_pull(skb, plen))
-		return NULL;
-
-	vh = (struct vxlanhdr *)(udp_hdr(skb) + 1);
+		return false;
 
-	skb_remcsum_process(skb, (void *)vh + hdrlen, start, offset,
-			    nopartial);
+	skb_remcsum_process(skb, (void *)(vxlan_hdr(skb) + 1), start, offset,
+			    !!(vxflags & VXLAN_F_REMCSUM_NOPARTIAL));
 
-	return vh;
+	return true;
 }
 
 static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
@@ -1285,9 +1281,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto drop;
 
 	if ((flags & VXLAN_HF_RCO) && (vs->flags & VXLAN_F_REMCSUM_RX)) {
-		if (!vxlan_remcsum(skb, vxlan_hdr(skb), sizeof(struct vxlanhdr),
-				   vni_field,
-				   !!(vs->flags & VXLAN_F_REMCSUM_NOPARTIAL)))
+		if (!vxlan_remcsum(skb, vs->flags, vni_field))
 			goto drop;
 
 		flags &= ~VXLAN_HF_RCO;
-- 
cgit v0.10.2


From 3288af0892e322b019a4c98810959b2a1c6dae7b Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:59:00 +0100
Subject: vxlan: move GBP header parsing to a separate function

To make vxlan_udp_encap_recv shorter and more comprehensible.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 161e39c..f65a923 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1156,6 +1156,24 @@ static bool vxlan_remcsum(struct sk_buff *skb, u32 vxflags, __be32 vni_field)
 	return true;
 }
 
+static void vxlan_parse_gbp_hdr(struct sk_buff *skb, struct vxlan_metadata *md,
+				struct metadata_dst *tun_dst)
+{
+	struct vxlanhdr_gbp *gbp;
+
+	gbp = (struct vxlanhdr_gbp *)vxlan_hdr(skb);
+	md->gbp = ntohs(gbp->policy_id);
+
+	if (tun_dst)
+		tun_dst->u.tun_info.key.tun_flags |= TUNNEL_VXLAN_OPT;
+
+	if (gbp->dont_learn)
+		md->gbp |= VXLAN_GBP_DONT_LEARN;
+
+	if (gbp->policy_applied)
+		md->gbp |= VXLAN_GBP_POLICY_APPLIED;
+}
+
 static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
 		      struct vxlan_metadata *md, __be32 vni,
 		      struct metadata_dst *tun_dst)
@@ -1304,20 +1322,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	 * used by VXLAN extensions if explicitly requested.
 	 */
 	if ((flags & VXLAN_HF_GBP) && (vs->flags & VXLAN_F_GBP)) {
-		struct vxlanhdr_gbp *gbp;
-
-		gbp = (struct vxlanhdr_gbp *)vxlan_hdr(skb);
-		md->gbp = ntohs(gbp->policy_id);
-
-		if (tun_dst)
-			tun_dst->u.tun_info.key.tun_flags |= TUNNEL_VXLAN_OPT;
-
-		if (gbp->dont_learn)
-			md->gbp |= VXLAN_GBP_DONT_LEARN;
-
-		if (gbp->policy_applied)
-			md->gbp |= VXLAN_GBP_POLICY_APPLIED;
-
+		vxlan_parse_gbp_hdr(skb, md, tun_dst);
 		flags &= ~VXLAN_GBP_USED_BITS;
 	}
 
-- 
cgit v0.10.2


From f14ecebb3a4e83eb6233e0167aa4ba675c99e514 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:59:01 +0100
Subject: vxlan: clean up extension handling on rx

Bring the extension handling to a single place and move the actual handling
logic out of vxlan_udp_encap_recv as much as possible.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index f65a923..ac688dc 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1135,15 +1135,16 @@ static int vxlan_igmp_leave(struct vxlan_dev *vxlan)
 	return ret;
 }
 
-static bool vxlan_remcsum(struct sk_buff *skb, u32 vxflags, __be32 vni_field)
+static bool vxlan_remcsum(struct vxlanhdr *unparsed,
+			  struct sk_buff *skb, u32 vxflags)
 {
 	size_t start, offset, plen;
 
-	if (skb->remcsum_offload)
-		return true;
+	if (!(unparsed->vx_flags & VXLAN_HF_RCO) || skb->remcsum_offload)
+		goto out;
 
-	start = vxlan_rco_start(vni_field);
-	offset = start + vxlan_rco_offset(vni_field);
+	start = vxlan_rco_start(unparsed->vx_vni);
+	offset = start + vxlan_rco_offset(unparsed->vx_vni);
 
 	plen = sizeof(struct vxlanhdr) + offset + sizeof(u16);
 
@@ -1152,16 +1153,21 @@ static bool vxlan_remcsum(struct sk_buff *skb, u32 vxflags, __be32 vni_field)
 
 	skb_remcsum_process(skb, (void *)(vxlan_hdr(skb) + 1), start, offset,
 			    !!(vxflags & VXLAN_F_REMCSUM_NOPARTIAL));
-
+out:
+	unparsed->vx_flags &= ~VXLAN_HF_RCO;
+	unparsed->vx_vni &= VXLAN_VNI_MASK;
 	return true;
 }
 
-static void vxlan_parse_gbp_hdr(struct sk_buff *skb, struct vxlan_metadata *md,
+static void vxlan_parse_gbp_hdr(struct vxlanhdr *unparsed,
+				struct vxlan_metadata *md,
 				struct metadata_dst *tun_dst)
 {
-	struct vxlanhdr_gbp *gbp;
+	struct vxlanhdr_gbp *gbp = (struct vxlanhdr_gbp *)unparsed;
+
+	if (!(unparsed->vx_flags & VXLAN_HF_GBP))
+		goto out;
 
-	gbp = (struct vxlanhdr_gbp *)vxlan_hdr(skb);
 	md->gbp = ntohs(gbp->policy_id);
 
 	if (tun_dst)
@@ -1172,6 +1178,9 @@ static void vxlan_parse_gbp_hdr(struct sk_buff *skb, struct vxlan_metadata *md,
 
 	if (gbp->policy_applied)
 		md->gbp |= VXLAN_GBP_POLICY_APPLIED;
+
+out:
+	unparsed->vx_flags &= ~VXLAN_GBP_USED_BITS;
 }
 
 static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
@@ -1273,7 +1282,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct metadata_dst *tun_dst = NULL;
 	struct vxlan_sock *vs;
-	__be32 flags, vni_field;
+	struct vxlanhdr unparsed;
 	struct vxlan_metadata _md;
 	struct vxlan_metadata *md = &_md;
 
@@ -1281,11 +1290,10 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!pskb_may_pull(skb, VXLAN_HLEN))
 		goto error;
 
-	flags = vxlan_hdr(skb)->vx_flags;
-	vni_field = vxlan_hdr(skb)->vx_vni;
-
-	if (flags & VXLAN_HF_VNI) {
-		flags &= ~VXLAN_HF_VNI;
+	unparsed = *vxlan_hdr(skb);
+	if (unparsed.vx_flags & VXLAN_HF_VNI) {
+		unparsed.vx_flags &= ~VXLAN_HF_VNI;
+		unparsed.vx_vni &= ~VXLAN_VNI_MASK;
 	} else {
 		/* VNI flag always required to be set */
 		goto bad_flags;
@@ -1298,17 +1306,10 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!vs)
 		goto drop;
 
-	if ((flags & VXLAN_HF_RCO) && (vs->flags & VXLAN_F_REMCSUM_RX)) {
-		if (!vxlan_remcsum(skb, vs->flags, vni_field))
-			goto drop;
-
-		flags &= ~VXLAN_HF_RCO;
-		vni_field &= VXLAN_VNI_MASK;
-	}
-
 	if (vxlan_collect_metadata(vs)) {
 		tun_dst = udp_tun_rx_dst(skb, vxlan_get_sk_family(vs), TUNNEL_KEY,
-					 vxlan_vni(vni_field), sizeof(*md));
+					 vxlan_vni(vxlan_hdr(skb)->vx_vni),
+					 sizeof(*md));
 
 		if (!tun_dst)
 			goto drop;
@@ -1321,12 +1322,13 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	/* For backwards compatibility, only allow reserved fields to be
 	 * used by VXLAN extensions if explicitly requested.
 	 */
-	if ((flags & VXLAN_HF_GBP) && (vs->flags & VXLAN_F_GBP)) {
-		vxlan_parse_gbp_hdr(skb, md, tun_dst);
-		flags &= ~VXLAN_GBP_USED_BITS;
-	}
+	if (vs->flags & VXLAN_F_REMCSUM_RX)
+		if (!vxlan_remcsum(&unparsed, skb, vs->flags))
+			goto drop;
+	if (vs->flags & VXLAN_F_GBP)
+		vxlan_parse_gbp_hdr(&unparsed, md, tun_dst);
 
-	if (flags || vni_field & ~VXLAN_VNI_MASK) {
+	if (unparsed.vx_flags || unparsed.vx_vni) {
 		/* If there are any unprocessed flags remaining treat
 		 * this as a malformed packet. This behavior diverges from
 		 * VXLAN RFC (RFC7348) which stipulates that bits in reserved
@@ -1339,7 +1341,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto bad_flags;
 	}
 
-	vxlan_rcv(vs, skb, md, vxlan_vni(vni_field), tun_dst);
+	vxlan_rcv(vs, skb, md, vxlan_vni(vxlan_hdr(skb)->vx_vni), tun_dst);
 	return 0;
 
 drop:
-- 
cgit v0.10.2


From 288b01c8c46eda01cd3921c6e3aca6293e57cee8 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:59:02 +0100
Subject: vxlan: clean up rx error path

When there are unrecognized flags present in the vxlan header, it doesn't
make much sense to return the packet for further UDP processing, especially
considering that for other invalid flag combinations we drop the packet
because of previous checks.

This means we return positive value only at the beginning of the function
where tun_dst is not yet allocated. This allows us to get rid of the
bad_flags and error jump labels.

When we're dropping packet, we need to free tun_dst now.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index ac688dc..dfbb3cb 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1288,16 +1288,19 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 
 	/* Need Vxlan and inner Ethernet header to be present */
 	if (!pskb_may_pull(skb, VXLAN_HLEN))
-		goto error;
+		return 1;
 
 	unparsed = *vxlan_hdr(skb);
-	if (unparsed.vx_flags & VXLAN_HF_VNI) {
-		unparsed.vx_flags &= ~VXLAN_HF_VNI;
-		unparsed.vx_vni &= ~VXLAN_VNI_MASK;
-	} else {
-		/* VNI flag always required to be set */
-		goto bad_flags;
+	/* VNI flag always required to be set */
+	if (!(unparsed.vx_flags & VXLAN_HF_VNI)) {
+		netdev_dbg(skb->dev, "invalid vxlan flags=%#x vni=%#x\n",
+			   ntohl(vxlan_hdr(skb)->vx_flags),
+			   ntohl(vxlan_hdr(skb)->vx_vni));
+		/* Return non vxlan pkt */
+		return 1;
 	}
+	unparsed.vx_flags &= ~VXLAN_HF_VNI;
+	unparsed.vx_vni &= ~VXLAN_VNI_MASK;
 
 	if (iptunnel_pull_header(skb, VXLAN_HLEN, htons(ETH_P_TEB)))
 		goto drop;
@@ -1337,29 +1340,19 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		 * is more robust and provides a little more security in
 		 * adding extensions to VXLAN.
 		 */
-
-		goto bad_flags;
+		goto drop;
 	}
 
 	vxlan_rcv(vs, skb, md, vxlan_vni(vxlan_hdr(skb)->vx_vni), tun_dst);
 	return 0;
 
 drop:
-	/* Consume bad packet */
-	kfree_skb(skb);
-	return 0;
-
-bad_flags:
-	netdev_dbg(skb->dev, "invalid vxlan flags=%#x vni=%#x\n",
-		   ntohl(vxlan_hdr(skb)->vx_flags),
-		   ntohl(vxlan_hdr(skb)->vx_vni));
-
-error:
 	if (tun_dst)
 		dst_release((struct dst_entry *)tun_dst);
 
-	/* Return non vxlan pkt */
-	return 1;
+	/* Consume bad packet */
+	kfree_skb(skb);
+	return 0;
 }
 
 static int arp_reduce(struct net_device *dev, struct sk_buff *skb)
-- 
cgit v0.10.2


From b9167b2e775999d267874b5a1a3a79020f54753a Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 16 Feb 2016 21:59:03 +0100
Subject: vxlan: treat vni in metadata based tunnels consistently

For metadata based tunnels, VNI is ignored when doing vxlan device lookups
(because such tunnel receives all VNIs). However, this was not honored by
vxlan_xmit_one when doing encapsulation bypass. Move the check for metadata
based tunnel to the common place where it belongs.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index dfbb3cb..3a84680 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -246,6 +246,10 @@ static struct vxlan_dev *vxlan_vs_find_vni(struct vxlan_sock *vs, __be32 vni)
 {
 	struct vxlan_dev *vxlan;
 
+	/* For flow based devices, map all packets to VNI 0 */
+	if (vs->flags & VXLAN_F_COLLECT_METADATA)
+		vni = 0;
+
 	hlist_for_each_entry_rcu(vxlan, vni_head(vs, vni), hlist) {
 		if (vxlan->default_dst.remote_vni == vni)
 			return vxlan;
@@ -1194,10 +1198,6 @@ static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
 	union vxlan_addr saddr;
 	int err = 0;
 
-	/* For flow based devices, map all packets to VNI 0 */
-	if (vs->flags & VXLAN_F_COLLECT_METADATA)
-		vni = 0;
-
 	/* Is this VNI defined? */
 	vxlan = vxlan_vs_find_vni(vs, vni);
 	if (!vxlan)
-- 
cgit v0.10.2


From d6b3bca11947888bc79e343be931b502040ab6a9 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Fri, 15 Jan 2016 14:33:08 -0800
Subject: i40e: get rid of magic number

We have a define for this, use it. No functional change.

Change-ID: Ic0e3ea4f562e46de63b2a8de07f291ccc10205fd
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 3e0d87e..597f209 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -461,7 +461,7 @@ static int i40e_config_vsi_rx_queue(struct i40e_vf *vf, u16 vsi_id,
 		rx_ctx.hbuff = info->hdr_size >> I40E_RXQ_CTX_HBUFF_SHIFT;
 
 		/* set splitalways mode 10b */
-		rx_ctx.dtype = 0x2;
+		rx_ctx.dtype = I40E_RX_DTYPE_HEADER_SPLIT;
 	}
 
 	/* databuffer length validation */
-- 
cgit v0.10.2


From cb5c260e330054591294a9924c283ef1ddcd74a4 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Fri, 15 Jan 2016 14:33:09 -0800
Subject: i40e: drop unused debugfs file "dump"

There was a completely unused file "dump" in debugfs that
never panned out to be useful.

Change-ID: I12bb9e37b5a83299725dda815a8746157baf6562
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 34da53b..631c2d6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -72,248 +72,6 @@ static struct i40e_veb *i40e_dbg_find_veb(struct i40e_pf *pf, int seid)
 }
 
 /**************************************************************
- * dump
- * The dump entry in debugfs is for getting a data snapshow of
- * the driver's current configuration and runtime details.
- * When the filesystem entry is written, a snapshot is taken.
- * When the entry is read, the most recent snapshot data is dumped.
- **************************************************************/
-static char *i40e_dbg_dump_buf;
-static ssize_t i40e_dbg_dump_data_len;
-static ssize_t i40e_dbg_dump_buffer_len;
-
-/**
- * i40e_dbg_dump_read - read the dump data
- * @filp: the opened file
- * @buffer: where to write the data for the user to read
- * @count: the size of the user's buffer
- * @ppos: file position offset
- **/
-static ssize_t i40e_dbg_dump_read(struct file *filp, char __user *buffer,
-				  size_t count, loff_t *ppos)
-{
-	int bytes_not_copied;
-	int len;
-
-	/* is *ppos bigger than the available data? */
-	if (*ppos >= i40e_dbg_dump_data_len || !i40e_dbg_dump_buf)
-		return 0;
-
-	/* be sure to not read beyond the end of available data */
-	len = min_t(int, count, (i40e_dbg_dump_data_len - *ppos));
-
-	bytes_not_copied = copy_to_user(buffer, &i40e_dbg_dump_buf[*ppos], len);
-	if (bytes_not_copied)
-		return -EFAULT;
-
-	*ppos += len;
-	return len;
-}
-
-/**
- * i40e_dbg_prep_dump_buf
- * @pf: the PF we're working with
- * @buflen: the desired buffer length
- *
- * Return positive if success, 0 if failed
- **/
-static int i40e_dbg_prep_dump_buf(struct i40e_pf *pf, int buflen)
-{
-	/* if not already big enough, prep for re alloc */
-	if (i40e_dbg_dump_buffer_len && i40e_dbg_dump_buffer_len < buflen) {
-		kfree(i40e_dbg_dump_buf);
-		i40e_dbg_dump_buffer_len = 0;
-		i40e_dbg_dump_buf = NULL;
-	}
-
-	/* get a new buffer if needed */
-	if (!i40e_dbg_dump_buf) {
-		i40e_dbg_dump_buf = kzalloc(buflen, GFP_KERNEL);
-		if (i40e_dbg_dump_buf != NULL)
-			i40e_dbg_dump_buffer_len = buflen;
-	}
-
-	return i40e_dbg_dump_buffer_len;
-}
-
-/**
- * i40e_dbg_dump_write - trigger a datadump snapshot
- * @filp: the opened file
- * @buffer: where to find the user's data
- * @count: the length of the user's data
- * @ppos: file position offset
- *
- * Any write clears the stats
- **/
-static ssize_t i40e_dbg_dump_write(struct file *filp,
-				   const char __user *buffer,
-				   size_t count, loff_t *ppos)
-{
-	struct i40e_pf *pf = filp->private_data;
-	bool seid_found = false;
-	long seid = -1;
-	int buflen = 0;
-	int i, ret;
-	int len;
-	u8 *p;
-
-	/* don't allow partial writes */
-	if (*ppos != 0)
-		return 0;
-
-	/* decode the SEID given to be dumped */
-	ret = kstrtol_from_user(buffer, count, 0, &seid);
-
-	if (ret) {
-		dev_info(&pf->pdev->dev, "bad seid value\n");
-	} else if (seid == 0) {
-		seid_found = true;
-
-		kfree(i40e_dbg_dump_buf);
-		i40e_dbg_dump_buffer_len = 0;
-		i40e_dbg_dump_data_len = 0;
-		i40e_dbg_dump_buf = NULL;
-		dev_info(&pf->pdev->dev, "debug buffer freed\n");
-
-	} else if (seid == pf->pf_seid || seid == 1) {
-		seid_found = true;
-
-		buflen = sizeof(struct i40e_pf);
-		buflen += (sizeof(struct i40e_aq_desc)
-		     * (pf->hw.aq.num_arq_entries + pf->hw.aq.num_asq_entries));
-
-		if (i40e_dbg_prep_dump_buf(pf, buflen)) {
-			p = i40e_dbg_dump_buf;
-
-			/* avoid use of memcpy here due to sparse warning
-			 * about copy size.
-			 */
-			*((struct i40e_pf *)p) = *pf;
-			p += sizeof(struct i40e_pf);
-
-			len = (sizeof(struct i40e_aq_desc)
-					* pf->hw.aq.num_asq_entries);
-			memcpy(p, pf->hw.aq.asq.desc_buf.va, len);
-			p += len;
-
-			len = (sizeof(struct i40e_aq_desc)
-					* pf->hw.aq.num_arq_entries);
-			memcpy(p, pf->hw.aq.arq.desc_buf.va, len);
-			p += len;
-
-			i40e_dbg_dump_data_len = buflen;
-			dev_info(&pf->pdev->dev,
-				 "PF seid %ld dumped %d bytes\n",
-				 seid, (int)i40e_dbg_dump_data_len);
-		}
-	} else if (seid >= I40E_BASE_VSI_SEID) {
-		struct i40e_vsi *vsi = NULL;
-		struct i40e_mac_filter *f;
-		int filter_count = 0;
-
-		mutex_lock(&pf->switch_mutex);
-		vsi = i40e_dbg_find_vsi(pf, seid);
-		if (!vsi) {
-			mutex_unlock(&pf->switch_mutex);
-			goto write_exit;
-		}
-
-		buflen = sizeof(struct i40e_vsi);
-		buflen += sizeof(struct i40e_q_vector) * vsi->num_q_vectors;
-		buflen += sizeof(struct i40e_ring) * 2 * vsi->num_queue_pairs;
-		buflen += sizeof(struct i40e_tx_buffer) * vsi->num_queue_pairs;
-		buflen += sizeof(struct i40e_rx_buffer) * vsi->num_queue_pairs;
-		list_for_each_entry(f, &vsi->mac_filter_list, list)
-			filter_count++;
-		buflen += sizeof(struct i40e_mac_filter) * filter_count;
-
-		if (i40e_dbg_prep_dump_buf(pf, buflen)) {
-			p = i40e_dbg_dump_buf;
-			seid_found = true;
-
-			len = sizeof(struct i40e_vsi);
-			memcpy(p, vsi, len);
-			p += len;
-
-			if (vsi->num_q_vectors) {
-				len = (sizeof(struct i40e_q_vector)
-					* vsi->num_q_vectors);
-				memcpy(p, vsi->q_vectors, len);
-				p += len;
-			}
-
-			if (vsi->num_queue_pairs) {
-				len = (sizeof(struct i40e_ring) *
-				      vsi->num_queue_pairs);
-				memcpy(p, vsi->tx_rings, len);
-				p += len;
-				memcpy(p, vsi->rx_rings, len);
-				p += len;
-			}
-
-			if (vsi->tx_rings[0]) {
-				len = sizeof(struct i40e_tx_buffer);
-				for (i = 0; i < vsi->num_queue_pairs; i++) {
-					memcpy(p, vsi->tx_rings[i]->tx_bi, len);
-					p += len;
-				}
-				len = sizeof(struct i40e_rx_buffer);
-				for (i = 0; i < vsi->num_queue_pairs; i++) {
-					memcpy(p, vsi->rx_rings[i]->rx_bi, len);
-					p += len;
-				}
-			}
-
-			/* macvlan filter list */
-			len = sizeof(struct i40e_mac_filter);
-			list_for_each_entry(f, &vsi->mac_filter_list, list) {
-				memcpy(p, f, len);
-				p += len;
-			}
-
-			i40e_dbg_dump_data_len = buflen;
-			dev_info(&pf->pdev->dev,
-				 "VSI seid %ld dumped %d bytes\n",
-				 seid, (int)i40e_dbg_dump_data_len);
-		}
-		mutex_unlock(&pf->switch_mutex);
-	} else if (seid >= I40E_BASE_VEB_SEID) {
-		struct i40e_veb *veb = NULL;
-
-		mutex_lock(&pf->switch_mutex);
-		veb = i40e_dbg_find_veb(pf, seid);
-		if (!veb) {
-			mutex_unlock(&pf->switch_mutex);
-			goto write_exit;
-		}
-
-		buflen = sizeof(struct i40e_veb);
-		if (i40e_dbg_prep_dump_buf(pf, buflen)) {
-			seid_found = true;
-			memcpy(i40e_dbg_dump_buf, veb, buflen);
-			i40e_dbg_dump_data_len = buflen;
-			dev_info(&pf->pdev->dev,
-				 "VEB seid %ld dumped %d bytes\n",
-				 seid, (int)i40e_dbg_dump_data_len);
-		}
-		mutex_unlock(&pf->switch_mutex);
-	}
-
-write_exit:
-	if (!seid_found)
-		dev_info(&pf->pdev->dev, "unknown seid %ld\n", seid);
-
-	return count;
-}
-
-static const struct file_operations i40e_dbg_dump_fops = {
-	.owner = THIS_MODULE,
-	.open =  simple_open,
-	.read =  i40e_dbg_dump_read,
-	.write = i40e_dbg_dump_write,
-};
-
-/**************************************************************
  * command
  * The command entry in debugfs is for giving the driver commands
  * to be executed - these may be for changing the internal switch
@@ -2217,11 +1975,6 @@ void i40e_dbg_pf_init(struct i40e_pf *pf)
 	if (!pfile)
 		goto create_failed;
 
-	pfile = debugfs_create_file("dump", 0600, pf->i40e_dbg_pf, pf,
-				    &i40e_dbg_dump_fops);
-	if (!pfile)
-		goto create_failed;
-
 	pfile = debugfs_create_file("netdev_ops", 0600, pf->i40e_dbg_pf, pf,
 				    &i40e_dbg_netdev_ops_fops);
 	if (!pfile)
@@ -2242,9 +1995,6 @@ void i40e_dbg_pf_exit(struct i40e_pf *pf)
 {
 	debugfs_remove_recursive(pf->i40e_dbg_pf);
 	pf->i40e_dbg_pf = NULL;
-
-	kfree(i40e_dbg_dump_buf);
-	i40e_dbg_dump_buf = NULL;
 }
 
 /**
-- 
cgit v0.10.2


From 00e5ec4bcc4aa3d4663502b5377e17ea43e058d6 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Fri, 15 Jan 2016 14:33:10 -0800
Subject: i40evf: support packet split receive

Support packet split receive on VFs. This is off by default but can be
enabled using ethtool private flags. Because we need to trigger a reset
from outside of i40evf_main.c, create a new function to do so, and
export it.

Also update copyright year in file headers.

Change-ID: I721aa5d70113d3d6d94102e5f31526f6fc57cbbb
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index 9e15f68..e657ecc 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -274,6 +274,9 @@ struct i40evf_adapter {
 };
 
 
+/* Ethtool Private Flags */
+#define I40EVF_PRIV_FLAGS_PS		BIT(0)
+
 /* needed by i40evf_ethtool.c */
 extern char i40evf_driver_name[];
 extern const char i40evf_driver_version[];
@@ -281,6 +284,7 @@ extern const char i40evf_driver_version[];
 int i40evf_up(struct i40evf_adapter *adapter);
 void i40evf_down(struct i40evf_adapter *adapter);
 int i40evf_process_config(struct i40evf_adapter *adapter);
+void i40evf_schedule_reset(struct i40evf_adapter *adapter);
 void i40evf_reset(struct i40evf_adapter *adapter);
 void i40evf_set_ethtool_ops(struct net_device *netdev);
 void i40evf_update_stats(struct i40evf_adapter *adapter);
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
index bd1c272..dd4430a 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2015 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -63,6 +63,12 @@ static const struct i40evf_stats i40evf_gstrings_stats[] = {
 #define I40EVF_STATS_LEN(_dev) \
 	(I40EVF_GLOBAL_STATS_LEN + I40EVF_QUEUE_STATS_LEN(_dev))
 
+static const char i40evf_priv_flags_strings[][ETH_GSTRING_LEN] = {
+	"packet-split",
+};
+
+#define I40EVF_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40evf_priv_flags_strings)
+
 /**
  * i40evf_get_settings - Get Link Speed and Duplex settings
  * @netdev: network interface device structure
@@ -97,6 +103,8 @@ static int i40evf_get_sset_count(struct net_device *netdev, int sset)
 {
 	if (sset == ETH_SS_STATS)
 		return I40EVF_STATS_LEN(netdev);
+	else if (sset == ETH_SS_PRIV_FLAGS)
+		return I40EVF_PRIV_FLAGS_STR_LEN;
 	else
 		return -EINVAL;
 }
@@ -162,6 +170,12 @@ static void i40evf_get_strings(struct net_device *netdev, u32 sset, u8 *data)
 			snprintf(p, ETH_GSTRING_LEN, "rx-%u.bytes", i);
 			p += ETH_GSTRING_LEN;
 		}
+	} else if (sset == ETH_SS_PRIV_FLAGS) {
+		for (i = 0; i < I40EVF_PRIV_FLAGS_STR_LEN; i++) {
+			memcpy(data, i40evf_priv_flags_strings[i],
+			       ETH_GSTRING_LEN);
+			data += ETH_GSTRING_LEN;
+		}
 	}
 }
 
@@ -211,6 +225,7 @@ static void i40evf_get_drvinfo(struct net_device *netdev,
 	strlcpy(drvinfo->version, i40evf_driver_version, 32);
 	strlcpy(drvinfo->fw_version, "N/A", 4);
 	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), 32);
+	drvinfo->n_priv_flags = I40EVF_PRIV_FLAGS_STR_LEN;
 }
 
 /**
@@ -710,6 +725,54 @@ static int i40evf_set_rxfh(struct net_device *netdev, const u32 *indir,
 				 I40EVF_HLUT_ARRAY_SIZE);
 }
 
+/**
+ * i40evf_get_priv_flags - report device private flags
+ * @dev: network interface device structure
+ *
+ * The get string set count and the string set should be matched for each
+ * flag returned.  Add new strings for each flag to the i40e_priv_flags_strings
+ * array.
+ *
+ * Returns a u32 bitmap of flags.
+ **/
+static u32 i40evf_get_priv_flags(struct net_device *dev)
+{
+	struct i40evf_adapter *adapter = netdev_priv(dev);
+	u32 ret_flags = 0;
+
+	ret_flags |= adapter->flags & I40EVF_FLAG_RX_PS_ENABLED ?
+		I40EVF_PRIV_FLAGS_PS : 0;
+
+	return ret_flags;
+}
+
+/**
+ * i40evf_set_priv_flags - set private flags
+ * @dev: network interface device structure
+ * @flags: bit flags to be set
+ **/
+static int i40evf_set_priv_flags(struct net_device *dev, u32 flags)
+{
+	struct i40evf_adapter *adapter = netdev_priv(dev);
+	bool reset_required = false;
+
+	if ((flags & I40EVF_PRIV_FLAGS_PS) &&
+	    !(adapter->flags & I40EVF_FLAG_RX_PS_ENABLED)) {
+		adapter->flags |= I40EVF_FLAG_RX_PS_ENABLED;
+		reset_required = true;
+	} else if (!(flags & I40EVF_PRIV_FLAGS_PS) &&
+		   (adapter->flags & I40EVF_FLAG_RX_PS_ENABLED)) {
+		adapter->flags &= ~I40EVF_FLAG_RX_PS_ENABLED;
+		reset_required = true;
+	}
+
+	/* if needed, issue reset to cause things to take effect */
+	if (reset_required)
+		i40evf_schedule_reset(adapter);
+
+	return 0;
+}
+
 static const struct ethtool_ops i40evf_ethtool_ops = {
 	.get_settings		= i40evf_get_settings,
 	.get_drvinfo		= i40evf_get_drvinfo,
@@ -719,6 +782,8 @@ static const struct ethtool_ops i40evf_ethtool_ops = {
 	.get_strings		= i40evf_get_strings,
 	.get_ethtool_stats	= i40evf_get_ethtool_stats,
 	.get_sset_count		= i40evf_get_sset_count,
+	.get_priv_flags		= i40evf_get_priv_flags,
+	.set_priv_flags		= i40evf_set_priv_flags,
 	.get_msglevel		= i40evf_get_msglevel,
 	.set_msglevel		= i40evf_set_msglevel,
 	.get_coalesce		= i40evf_get_coalesce,
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 1d81d57..1176326 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -173,6 +173,19 @@ void i40evf_debug_d(void *hw, u32 mask, char *fmt_str, ...)
 }
 
 /**
+ * i40evf_schedule_reset - Set the flags and schedule a reset event
+ * @adapter: board private structure
+ **/
+void i40evf_schedule_reset(struct i40evf_adapter *adapter)
+{
+	if (!(adapter->flags &
+	      (I40EVF_FLAG_RESET_PENDING | I40EVF_FLAG_RESET_NEEDED))) {
+		adapter->flags |= I40EVF_FLAG_RESET_NEEDED;
+		schedule_work(&adapter->reset_task);
+	}
+}
+
+/**
  * i40evf_tx_timeout - Respond to a Tx Hang
  * @netdev: network interface device structure
  **/
@@ -181,11 +194,7 @@ static void i40evf_tx_timeout(struct net_device *netdev)
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
 
 	adapter->tx_timeout_count++;
-	if (!(adapter->flags & (I40EVF_FLAG_RESET_PENDING |
-				I40EVF_FLAG_RESET_NEEDED))) {
-		adapter->flags |= I40EVF_FLAG_RESET_NEEDED;
-		queue_work(i40evf_wq, &adapter->reset_task);
-	}
+	i40evf_schedule_reset(adapter);
 }
 
 /**
@@ -638,35 +647,22 @@ static void i40evf_configure_rx(struct i40evf_adapter *adapter)
 	int rx_buf_len;
 
 
-	adapter->flags &= ~I40EVF_FLAG_RX_PS_CAPABLE;
-	adapter->flags |= I40EVF_FLAG_RX_1BUF_CAPABLE;
-
-	/* Decide whether to use packet split mode or not */
-	if (netdev->mtu > ETH_DATA_LEN) {
-		if (adapter->flags & I40EVF_FLAG_RX_PS_CAPABLE)
-			adapter->flags |= I40EVF_FLAG_RX_PS_ENABLED;
-		else
-			adapter->flags &= ~I40EVF_FLAG_RX_PS_ENABLED;
-	} else {
-		if (adapter->flags & I40EVF_FLAG_RX_1BUF_CAPABLE)
-			adapter->flags &= ~I40EVF_FLAG_RX_PS_ENABLED;
-		else
-			adapter->flags |= I40EVF_FLAG_RX_PS_ENABLED;
-	}
-
 	/* Set the RX buffer length according to the mode */
-	if (adapter->flags & I40EVF_FLAG_RX_PS_ENABLED) {
-		rx_buf_len = I40E_RX_HDR_SIZE;
-	} else {
-		if (netdev->mtu <= ETH_DATA_LEN)
-			rx_buf_len = I40EVF_RXBUFFER_2048;
-		else
-			rx_buf_len = ALIGN(max_frame, 1024);
-	}
+	if (adapter->flags & I40EVF_FLAG_RX_PS_ENABLED ||
+	    netdev->mtu <= ETH_DATA_LEN)
+		rx_buf_len = I40EVF_RXBUFFER_2048;
+	else
+		rx_buf_len = ALIGN(max_frame, 1024);
 
 	for (i = 0; i < adapter->num_active_queues; i++) {
 		adapter->rx_rings[i].tail = hw->hw_addr + I40E_QRX_TAIL1(i);
 		adapter->rx_rings[i].rx_buf_len = rx_buf_len;
+		if (adapter->flags & I40EVF_FLAG_RX_PS_ENABLED) {
+			set_ring_ps_enabled(&adapter->rx_rings[i]);
+			adapter->rx_rings[i].rx_hdr_len = I40E_RX_HDR_SIZE;
+		} else {
+			clear_ring_ps_enabled(&adapter->rx_rings[i]);
+		}
 	}
 }
 
@@ -1003,7 +999,12 @@ static void i40evf_configure(struct i40evf_adapter *adapter)
 	for (i = 0; i < adapter->num_active_queues; i++) {
 		struct i40e_ring *ring = &adapter->rx_rings[i];
 
+	if (adapter->flags & I40EVF_FLAG_RX_PS_ENABLED) {
+		i40evf_alloc_rx_headers(ring);
+		i40evf_alloc_rx_buffers_ps(ring, ring->count);
+	} else {
 		i40evf_alloc_rx_buffers_1buf(ring, ring->count);
+	}
 		ring->next_to_use = ring->count - 1;
 		writel(ring->next_to_use, ring->tail);
 	}
@@ -2481,6 +2482,11 @@ static void i40evf_init_task(struct work_struct *work)
 	adapter->current_op = I40E_VIRTCHNL_OP_UNKNOWN;
 
 	adapter->flags |= I40EVF_FLAG_RX_CSUM_ENABLED;
+	adapter->flags |= I40EVF_FLAG_RX_1BUF_CAPABLE;
+	adapter->flags |= I40EVF_FLAG_RX_PS_CAPABLE;
+
+	/* Default to single buffer rx, can be changed through ethtool. */
+	adapter->flags &= ~I40EVF_FLAG_RX_PS_ENABLED;
 
 	netdev->netdev_ops = &i40evf_netdev_ops;
 	i40evf_set_ethtool_ops(netdev);
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index d3739cc..488e738 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -270,6 +270,10 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter)
 		vqpi->rxq.max_pkt_size = adapter->netdev->mtu
 					+ ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN;
 		vqpi->rxq.databuffer_size = adapter->rx_rings[i].rx_buf_len;
+		if (adapter->flags & I40EVF_FLAG_RX_PS_ENABLED) {
+			vqpi->rxq.splithdr_enabled = true;
+			vqpi->rxq.hdr_size = I40E_RX_HDR_SIZE;
+		}
 		vqpi++;
 	}
 
-- 
cgit v0.10.2


From f734dfff2f2bc2d92c392fbcfe2039893a8cdd08 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Fri, 15 Jan 2016 14:33:11 -0800
Subject: i40e: trivial: cleanup use of pf->hw

This patch makes use of a pointer called hw consistent
in the i40e_remove function.

Change-ID: Idacc7ff0a09a68289c57457a78618bf5497de077
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 8d41c6c..161603b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -11219,8 +11219,8 @@ static void i40e_remove(struct pci_dev *pdev)
 		i40e_vsi_release(pf->vsi[pf->lan_vsi]);
 
 	/* shutdown and destroy the HMC */
-	if (pf->hw.hmc.hmc_obj) {
-		ret_code = i40e_shutdown_lan_hmc(&pf->hw);
+	if (hw->hmc.hmc_obj) {
+		ret_code = i40e_shutdown_lan_hmc(hw);
 		if (ret_code)
 			dev_warn(&pdev->dev,
 				 "Failed to destroy the HMC resources: %d\n",
@@ -11228,7 +11228,7 @@ static void i40e_remove(struct pci_dev *pdev)
 	}
 
 	/* shutdown the adminq */
-	ret_code = i40e_shutdown_adminq(&pf->hw);
+	ret_code = i40e_shutdown_adminq(hw);
 	if (ret_code)
 		dev_warn(&pdev->dev,
 			 "Failed to destroy the Admin Queue resources: %d\n",
@@ -11256,7 +11256,7 @@ static void i40e_remove(struct pci_dev *pdev)
 	kfree(pf->qp_pile);
 	kfree(pf->vsi);
 
-	iounmap(pf->hw.hw_addr);
+	iounmap(hw->hw_addr);
 	kfree(pf);
 	pci_release_selected_regions(pdev,
 				     pci_select_bars(pdev, IORESOURCE_MEM));
-- 
cgit v0.10.2


From dd353109e41c1e92e0cea9954404a6f5a7d46218 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Fri, 15 Jan 2016 14:33:12 -0800
Subject: i40e: Add a SW workaround for lost interrupts

This patch adds a workaround for cases where we might have
interrupts that got lost but WB happened.
If that happens without this patch we will see a tx_timeout.
To work around it, this patch goes ahead and reschedules NAPI
in that situation, if NAPI is not already scheduled.
We also add a counter in ethtool to keep track of when
we detect a case of tx_lost_interrupt.

Note: napi_reschedule() can be safely called from process/service_task
context and is done in other drivers as well without an issue.

Change-ID: I00f98f1ce3774524d9421227652bef20fcbd0d20
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 5ea431d..5d8e159e1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -512,6 +512,7 @@ struct i40e_vsi {
 	u32 tx_busy;
 	u64 tx_linearize;
 	u64 tx_force_wb;
+	u64 tx_lost_interrupt;
 	u32 rx_buf_failed;
 	u32 rx_page_failed;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 230fa40..5778f9e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -89,6 +89,7 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = {
 	I40E_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol),
 	I40E_VSI_STAT("tx_linearize", tx_linearize),
 	I40E_VSI_STAT("tx_force_wb", tx_force_wb),
+	I40E_VSI_STAT("tx_lost_interrupt", tx_lost_interrupt),
 	I40E_VSI_STAT("rx_alloc_fail", rx_buf_failed),
 	I40E_VSI_STAT("rx_pg_alloc_fail", rx_page_failed),
 };
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 161603b..aa1c7ca 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -819,6 +819,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 	struct i40e_eth_stats *oes;
 	struct i40e_eth_stats *es;     /* device's eth stats */
 	u32 tx_restart, tx_busy;
+	u64 tx_lost_interrupt;
 	struct i40e_ring *p;
 	u32 rx_page, rx_buf;
 	u64 bytes, packets;
@@ -844,6 +845,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 	rx_b = rx_p = 0;
 	tx_b = tx_p = 0;
 	tx_restart = tx_busy = tx_linearize = tx_force_wb = 0;
+	tx_lost_interrupt = 0;
 	rx_page = 0;
 	rx_buf = 0;
 	rcu_read_lock();
@@ -862,6 +864,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 		tx_busy += p->tx_stats.tx_busy;
 		tx_linearize += p->tx_stats.tx_linearize;
 		tx_force_wb += p->tx_stats.tx_force_wb;
+		tx_lost_interrupt += p->tx_stats.tx_lost_interrupt;
 
 		/* Rx queue is part of the same block as Tx queue */
 		p = &p[1];
@@ -880,6 +883,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 	vsi->tx_busy = tx_busy;
 	vsi->tx_linearize = tx_linearize;
 	vsi->tx_force_wb = tx_force_wb;
+	vsi->tx_lost_interrupt = tx_lost_interrupt;
 	vsi->rx_page_failed = rx_page;
 	vsi->rx_buf_failed = rx_buf;
 
@@ -4349,7 +4353,7 @@ static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
 {
 	struct i40e_ring *tx_ring = NULL;
 	struct i40e_pf	*pf;
-	u32 head, val, tx_pending;
+	u32 head, val, tx_pending_hw;
 	int i;
 
 	pf = vsi->back;
@@ -4375,16 +4379,9 @@ static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
 	else
 		val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0);
 
-	/* Bail out if interrupts are disabled because napi_poll
-	 * execution in-progress or will get scheduled soon.
-	 * napi_poll cleans TX and RX queues and updates 'next_to_clean'.
-	 */
-	if (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK))
-		return;
-
 	head = i40e_get_head(tx_ring);
 
-	tx_pending = i40e_get_tx_pending(tx_ring);
+	tx_pending_hw = i40e_get_tx_pending(tx_ring, false);
 
 	/* HW is done executing descriptors, updated HEAD write back,
 	 * but SW hasn't processed those descriptors. If interrupt is
@@ -4392,12 +4389,12 @@ static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
 	 * dev_watchdog detecting timeout on those netdev_queue,
 	 * hence proactively trigger SW interrupt.
 	 */
-	if (tx_pending) {
+	if (tx_pending_hw && (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK))) {
 		/* NAPI Poll didn't run and clear since it was set */
 		if (test_and_clear_bit(I40E_Q_VECTOR_HUNG_DETECT,
 				       &tx_ring->q_vector->hung_detected)) {
-			netdev_info(vsi->netdev, "VSI_seid %d, Hung TX queue %d, tx_pending: %d, NTC:0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x\n",
-				    vsi->seid, q_idx, tx_pending,
+			netdev_info(vsi->netdev, "VSI_seid %d, Hung TX queue %d, tx_pending_hw: %d, NTC:0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x\n",
+				    vsi->seid, q_idx, tx_pending_hw,
 				    tx_ring->next_to_clean, head,
 				    tx_ring->next_to_use,
 				    readl(tx_ring->tail));
@@ -4410,6 +4407,17 @@ static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
 				&tx_ring->q_vector->hung_detected);
 		}
 	}
+
+	/* This is the case where we have interrupts missing,
+	 * so the tx_pending in HW will most likely be 0, but we
+	 * will have tx_pending in SW since the WB happened but the
+	 * interrupt got lost.
+	 */
+	if ((!tx_pending_hw) && i40e_get_tx_pending(tx_ring, true) &&
+	    (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK))) {
+		if (napi_reschedule(&tx_ring->q_vector->napi))
+			tx_ring->tx_stats.tx_lost_interrupt++;
+	}
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 0ffa9a8..5b43585 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -610,15 +610,19 @@ void i40e_free_tx_resources(struct i40e_ring *tx_ring)
 /**
  * i40e_get_tx_pending - how many tx descriptors not processed
  * @tx_ring: the ring of descriptors
+ * @in_sw: is tx_pending being checked in SW or HW
  *
  * Since there is no access to the ring head register
  * in XL710, we need to use our local copies
  **/
-u32 i40e_get_tx_pending(struct i40e_ring *ring)
+u32 i40e_get_tx_pending(struct i40e_ring *ring, bool in_sw)
 {
 	u32 head, tail;
 
-	head = i40e_get_head(ring);
+	if (!in_sw)
+		head = i40e_get_head(ring);
+	else
+		head = ring->next_to_clean;
 	tail = readl(ring->tail);
 
 	if (head != tail)
@@ -741,7 +745,7 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
 		 * them to be written back in case we stay in NAPI.
 		 * In this mode on X722 we do not enable Interrupt.
 		 */
-		j = i40e_get_tx_pending(tx_ring);
+		j = i40e_get_tx_pending(tx_ring, false);
 
 		if (budget &&
 		    ((j / (WB_STRIDE + 1)) == 0) && (j != 0) &&
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index ae22c4e..3acc924 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -203,6 +203,7 @@ struct i40e_tx_queue_stats {
 	u64 tx_done_old;
 	u64 tx_linearize;
 	u64 tx_force_wb;
+	u64 tx_lost_interrupt;
 };
 
 struct i40e_rx_queue_stats {
@@ -338,7 +339,7 @@ int i40e_tx_prepare_vlan_flags(struct sk_buff *skb,
 			       struct i40e_ring *tx_ring, u32 *flags);
 #endif
 void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
-u32 i40e_get_tx_pending(struct i40e_ring *ring);
+u32 i40e_get_tx_pending(struct i40e_ring *ring, bool in_sw);
 
 /**
  * i40e_get_head - Retrieve head from head writeback
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 76bad75..0f73a4f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -129,15 +129,19 @@ void i40evf_free_tx_resources(struct i40e_ring *tx_ring)
 /**
  * i40evf_get_tx_pending - how many Tx descriptors not processed
  * @tx_ring: the ring of descriptors
+ * @in_sw: is tx_pending being checked in SW or HW
  *
  * Since there is no access to the ring head register
  * in XL710, we need to use our local copies
  **/
-u32 i40evf_get_tx_pending(struct i40e_ring *ring)
+u32 i40evf_get_tx_pending(struct i40e_ring *ring, bool in_sw)
 {
 	u32 head, tail;
 
-	head = i40e_get_head(ring);
+	if (!in_sw)
+		head = i40e_get_head(ring);
+	else
+		head = ring->next_to_clean;
 	tail = readl(ring->tail);
 
 	if (head != tail)
@@ -259,7 +263,7 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
 		 * them to be written back in case we stay in NAPI.
 		 * In this mode on X722 we do not enable Interrupt.
 		 */
-		j = i40evf_get_tx_pending(tx_ring);
+		j = i40evf_get_tx_pending(tx_ring, false);
 
 		if (budget &&
 		    ((j / (WB_STRIDE + 1)) == 0) && (j > 0) &&
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 5467fcdf..81c9661 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -202,6 +202,7 @@ struct i40e_tx_queue_stats {
 	u64 tx_done_old;
 	u64 tx_linearize;
 	u64 tx_force_wb;
+	u64 tx_lost_interrupt;
 };
 
 struct i40e_rx_queue_stats {
@@ -326,7 +327,7 @@ void i40evf_free_tx_resources(struct i40e_ring *tx_ring);
 void i40evf_free_rx_resources(struct i40e_ring *rx_ring);
 int i40evf_napi_poll(struct napi_struct *napi, int budget);
 void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
-u32 i40evf_get_tx_pending(struct i40e_ring *ring);
+u32 i40evf_get_tx_pending(struct i40e_ring *ring, bool in_sw);
 
 /**
  * i40e_get_head - Retrieve head from head writeback
-- 
cgit v0.10.2


From 6784ed5a7385f15bc34903abffaca3db0f4e0438 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Fri, 15 Jan 2016 14:33:13 -0800
Subject: i40e: Fix PROMISC mode for Multi-function per port (MFP) devices

This patch falls back to enabling unicast, multicast and
broadcast promiscuous mode when the driver must disable it's use
of "default port" aka defport mode (which is normally used to
provide a promiscuous mode), due to internal incompatibility
with Multiple Function per Port (aka MFP).

The situation that requires this patch is when Physical
Function 0 is the device being used, and it can support SR-IOV
when MFP is enabled, via the driver creating a VEB on an MFP
enabled adapter.

Change-ID: Ie90b00d0d58782a5dfcf2c3c9725a2eb90bd63d8
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index aa1c7ca..41d6d55 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2122,7 +2122,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		cur_promisc = (!!(vsi->current_netdev_flags & IFF_PROMISC) ||
 			       test_bit(__I40E_FILTER_OVERFLOW_PROMISC,
 					&vsi->state));
-		if (vsi->type == I40E_VSI_MAIN && pf->lan_veb != I40E_NO_VEB) {
+		if ((vsi->type == I40E_VSI_MAIN) &&
+		    (pf->lan_veb != I40E_NO_VEB) &&
+		    !(pf->flags & I40E_FLAG_MFP_ENABLED)) {
 			/* set defport ON for Main VSI instead of true promisc
 			 * this way we will get all unicast/multicast and VLAN
 			 * promisc behavior but will not get VF or VMDq traffic
-- 
cgit v0.10.2


From 4147e2c5eec037dc51e4767b0ed42b32c6fecf9d Mon Sep 17 00:00:00 2001
From: Kiran Patil <kiran.patil@intel.com>
Date: Fri, 15 Jan 2016 14:33:14 -0800
Subject: i40e: Removal of code which relies on BASE VEB SEID

Fixed mapping of SEID is removed from specification. Hence
this patch removes code which was using hard coded base VEB SEID.

Changed FCoE code to use "hw->pf_id" to obtain correct "idx"
and verified.

Removed defines for BASE VSI/VEB SEID and BASE_PF_SEID since it
is not used anymore.

Change-ID: Id507cf4b1fae1c0145e3f08ae9ea5846ea5840de
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 5d8e159e1..e0758dd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -64,9 +64,6 @@
 #include "i40e_dcb.h"
 
 /* Useful i40e defaults */
-#define I40E_BASE_PF_SEID     16
-#define I40E_BASE_VSI_SEID    512
-#define I40E_BASE_VEB_SEID    288
 #define I40E_MAX_VEB          16
 
 #define I40E_MAX_NUM_DESCRIPTORS      4096
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 631c2d6..2a44f2e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -61,13 +61,9 @@ static struct i40e_veb *i40e_dbg_find_veb(struct i40e_pf *pf, int seid)
 {
 	int i;
 
-	if ((seid < I40E_BASE_VEB_SEID) ||
-	    (seid > (I40E_BASE_VEB_SEID + I40E_MAX_VEB)))
-		dev_info(&pf->pdev->dev, "%d: bad seid\n", seid);
-	else
-		for (i = 0; i < I40E_MAX_VEB; i++)
-			if (pf->veb[i] && pf->veb[i]->seid == seid)
-				return pf->veb[i];
+	for (i = 0; i < I40E_MAX_VEB; i++)
+		if (pf->veb[i] && pf->veb[i]->seid == seid)
+			return pf->veb[i];
 	return NULL;
 }
 
@@ -691,12 +687,6 @@ static void i40e_dbg_dump_veb_seid(struct i40e_pf *pf, int seid)
 {
 	struct i40e_veb *veb;
 
-	if ((seid < I40E_BASE_VEB_SEID) ||
-	    (seid >= (I40E_MAX_VEB + I40E_BASE_VEB_SEID))) {
-		dev_info(&pf->pdev->dev, "%d: bad seid\n", seid);
-		return;
-	}
-
 	veb = i40e_dbg_find_veb(pf, seid);
 	if (!veb) {
 		dev_info(&pf->pdev->dev, "can't find veb %d\n", seid);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 41d6d55..adb2a04 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -768,7 +768,7 @@ static void i40e_update_fcoe_stats(struct i40e_vsi *vsi)
 	if (vsi->type != I40E_VSI_FCOE)
 		return;
 
-	idx = (pf->pf_seid - I40E_BASE_PF_SEID) + I40E_FCOE_PF_STAT_OFFSET;
+	idx = hw->pf_id + I40E_FCOE_PF_STAT_OFFSET;
 	fs = &vsi->fcoe_stats;
 	ofs = &vsi->fcoe_stats_offsets;
 
-- 
cgit v0.10.2


From 16fd08b85962f6288fe1191856aa98cb0d40aa30 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Fri, 15 Jan 2016 14:33:15 -0800
Subject: i40e/i40evf: avoid atomics

In the case where we have a page fully used by receive data, we need to
release the page fully to the stack. Instead of calling get_page (which
increments the page count) followed by free_page (which decrements the
page count), just donate our reference to the stack. Although this
donation is not tax deductible, it does allow us to avoid two very
expensive atomic operations that reverse each other.

Change-ID: If70739792d5748995fc175ec92ac2171ed4ad8fc
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 5b43585..25e378c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1663,28 +1663,33 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 					rx_bi->page_offset + copysize,
 					rx_packet_len, I40E_RXBUFFER_2048);
 
-			get_page(rx_bi->page);
-			/* switch to the other half-page here; the allocation
-			 * code programs the right addr into HW. If we haven't
-			 * used this half-page, the address won't be changed,
-			 * and HW can just use it next time through.
-			 */
-			rx_bi->page_offset ^= PAGE_SIZE / 2;
 			/* If the page count is more than 2, then both halves
 			 * of the page are used and we need to free it. Do it
 			 * here instead of in the alloc code. Otherwise one
 			 * of the half-pages might be released between now and
 			 * then, and we wouldn't know which one to use.
+			 * Don't call get_page and free_page since those are
+			 * both expensive atomic operations that just change
+			 * the refcount in opposite directions. Just give the
+			 * page to the stack; he can have our refcount.
 			 */
 			if (page_count(rx_bi->page) > 2) {
 				dma_unmap_page(rx_ring->dev,
 					       rx_bi->page_dma,
 					       PAGE_SIZE,
 					       DMA_FROM_DEVICE);
-				__free_page(rx_bi->page);
 				rx_bi->page = NULL;
 				rx_bi->page_dma = 0;
 				rx_ring->rx_stats.realloc_count++;
+			} else {
+				get_page(rx_bi->page);
+				/* switch to the other half-page here; the
+				 * allocation code programs the right addr
+				 * into HW. If we haven't used this half-page,
+				 * the address won't be changed, and HW can
+				 * just use it next time through.
+				 */
+				rx_bi->page_offset ^= PAGE_SIZE / 2;
 			}
 
 		}
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 0f73a4f..fb6cd7e 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1126,28 +1126,33 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, const int budget)
 					rx_bi->page_offset + copysize,
 					rx_packet_len, I40E_RXBUFFER_2048);
 
-			get_page(rx_bi->page);
-			/* switch to the other half-page here; the allocation
-			 * code programs the right addr into HW. If we haven't
-			 * used this half-page, the address won't be changed,
-			 * and HW can just use it next time through.
-			 */
-			rx_bi->page_offset ^= PAGE_SIZE / 2;
 			/* If the page count is more than 2, then both halves
 			 * of the page are used and we need to free it. Do it
 			 * here instead of in the alloc code. Otherwise one
 			 * of the half-pages might be released between now and
 			 * then, and we wouldn't know which one to use.
+			 * Don't call get_page and free_page since those are
+			 * both expensive atomic operations that just change
+			 * the refcount in opposite directions. Just give the
+			 * page to the stack; he can have our refcount.
 			 */
 			if (page_count(rx_bi->page) > 2) {
 				dma_unmap_page(rx_ring->dev,
 					       rx_bi->page_dma,
 					       PAGE_SIZE,
 					       DMA_FROM_DEVICE);
-				__free_page(rx_bi->page);
 				rx_bi->page = NULL;
 				rx_bi->page_dma = 0;
 				rx_ring->rx_stats.realloc_count++;
+			} else {
+				get_page(rx_bi->page);
+				/* switch to the other half-page here; the
+				 * allocation code programs the right addr
+				 * into HW. If we haven't used this half-page,
+				 * the address won't be changed, and HW can
+				 * just use it next time through.
+				 */
+				rx_bi->page_offset ^= PAGE_SIZE / 2;
 			}
 
 		}
-- 
cgit v0.10.2


From a16ae2d59c290b0c545edb9bf41ac9cf88352ee0 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Fri, 15 Jan 2016 14:33:16 -0800
Subject: i40e: Do not disable queues in the Legacy/MSI Interrupt handler

The queues should never be enabled/disabled in the interrupt handler,
ICR0 interrupt enable should be the only thing that needs to be
dynamically changed in the handler.

This patch fixes that. Without this patch X722 platforms were
seeing weird ping timings when in Legacy mode since it takes
a whole lot of time for the HW/FW to re-enable queues.

Change-ID: If065afc45d81c5a19d4a94a00cd5b8f61cefc40c
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index adb2a04..7869d74 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3462,16 +3462,12 @@ static irqreturn_t i40e_intr(int irq, void *data)
 		struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
 		struct i40e_q_vector *q_vector = vsi->q_vectors[0];
 
-		/* temporarily disable queue cause for NAPI processing */
-		u32 qval = rd32(hw, I40E_QINT_RQCTL(0));
-
-		qval &= ~I40E_QINT_RQCTL_CAUSE_ENA_MASK;
-		wr32(hw, I40E_QINT_RQCTL(0), qval);
-
-		qval = rd32(hw, I40E_QINT_TQCTL(0));
-		qval &= ~I40E_QINT_TQCTL_CAUSE_ENA_MASK;
-		wr32(hw, I40E_QINT_TQCTL(0), qval);
-
+		/* We do not have a way to disarm Queue causes while leaving
+		 * interrupt enabled for all other causes, ideally
+		 * interrupt should be disabled while we are in NAPI but
+		 * this is not a performance path and napi_schedule()
+		 * can deal with rescheduling.
+		 */
 		if (!test_bit(__I40E_DOWN, &pf->state))
 			napi_schedule_irqoff(&q_vector->napi);
 	}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 25e378c..65f2fd8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2051,19 +2051,6 @@ tx_only:
 	if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
 		i40e_update_enable_itr(vsi, q_vector);
 	} else { /* Legacy mode */
-		struct i40e_hw *hw = &vsi->back->hw;
-		/* We re-enable the queue 0 cause, but
-		 * don't worry about dynamic_enable
-		 * because we left it on for the other
-		 * possible interrupts during napi
-		 */
-		u32 qval = rd32(hw, I40E_QINT_RQCTL(0)) |
-			   I40E_QINT_RQCTL_CAUSE_ENA_MASK;
-
-		wr32(hw, I40E_QINT_RQCTL(0), qval);
-		qval = rd32(hw, I40E_QINT_TQCTL(0)) |
-		       I40E_QINT_TQCTL_CAUSE_ENA_MASK;
-		wr32(hw, I40E_QINT_TQCTL(0), qval);
 		i40e_irq_dynamic_enable_icr0(vsi->back, false);
 	}
 	return 0;
-- 
cgit v0.10.2


From 0d79032781caf8573d141259bd5008516439d634 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Fri, 15 Jan 2016 14:33:17 -0800
Subject: i40e: expand comment

Explain why we cannot remove this code, even though it works differently
than any of our other interrupt cause handling code.

Change-ID: Ie66203bd037a466066036611c31d44f759ec5176
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 597f209..5dcd198 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -2037,7 +2037,11 @@ int i40e_vc_process_vflr_event(struct i40e_pf *pf)
 	if (!test_bit(__I40E_VFLR_EVENT_PENDING, &pf->state))
 		return 0;
 
-	/* re-enable vflr interrupt cause */
+	/* Re-enable the VFLR interrupt cause here, before looking for which
+	 * VF got reset. Otherwise, if another VF gets a reset while the
+	 * first one is being processed, that interrupt will be lost, and
+	 * that VF will be stuck in reset forever.
+	 */
 	reg = rd32(hw, I40E_PFINT_ICR0_ENA);
 	reg |= I40E_PFINT_ICR0_ENA_VFLR_MASK;
 	wr32(hw, I40E_PFINT_ICR0_ENA, reg);
-- 
cgit v0.10.2


From 6e93d0c90fb2e20fc5b9dbdf28b45e91a16b4353 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Fri, 15 Jan 2016 14:33:18 -0800
Subject: i40e: better error reporting for nvmupdate

Make sure we return EBUSY while finishing up a reset, and add a few bits
for better debug messages.

Change-ID: I23f6c28a8d96d7aa171abcc265737cec7826c292
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 5778f9e..9fc7546b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1005,16 +1005,19 @@ static int i40e_get_eeprom(struct net_device *netdev,
 	/* check for NVMUpdate access method */
 	magic = hw->vendor_id | (hw->device_id << 16);
 	if (eeprom->magic && eeprom->magic != magic) {
-		struct i40e_nvm_access *cmd;
-		int errno;
+		struct i40e_nvm_access *cmd = (struct i40e_nvm_access *)eeprom;
+		int errno = 0;
 
 		/* make sure it is the right magic for NVMUpdate */
 		if ((eeprom->magic >> 16) != hw->device_id)
-			return -EINVAL;
+			errno = -EINVAL;
+		else if (test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state) ||
+			 test_bit(__I40E_RESET_INTR_RECEIVED, &pf->state))
+			errno = -EBUSY;
+		else
+			ret_val = i40e_nvmupd_command(hw, cmd, bytes, &errno);
 
-		cmd = (struct i40e_nvm_access *)eeprom;
-		ret_val = i40e_nvmupd_command(hw, cmd, bytes, &errno);
-		if (ret_val && (hw->debug_mask & I40E_DEBUG_NVM))
+		if ((errno || ret_val) && (hw->debug_mask & I40E_DEBUG_NVM))
 			dev_info(&pf->pdev->dev,
 				 "NVMUpdate read failed err=%d status=0x%x errno=%d module=%d offset=0x%x size=%d\n",
 				 ret_val, hw->aq.asq_last_status, errno,
@@ -1098,27 +1101,25 @@ static int i40e_set_eeprom(struct net_device *netdev,
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_hw *hw = &np->vsi->back->hw;
 	struct i40e_pf *pf = np->vsi->back;
-	struct i40e_nvm_access *cmd;
+	struct i40e_nvm_access *cmd = (struct i40e_nvm_access *)eeprom;
 	int ret_val = 0;
-	int errno;
+	int errno = 0;
 	u32 magic;
 
 	/* normal ethtool set_eeprom is not supported */
 	magic = hw->vendor_id | (hw->device_id << 16);
 	if (eeprom->magic == magic)
-		return -EOPNOTSUPP;
-
+		errno = -EOPNOTSUPP;
 	/* check for NVMUpdate access method */
-	if (!eeprom->magic || (eeprom->magic >> 16) != hw->device_id)
-		return -EINVAL;
-
-	if (test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state) ||
-	    test_bit(__I40E_RESET_INTR_RECEIVED, &pf->state))
-		return -EBUSY;
+	else if (!eeprom->magic || (eeprom->magic >> 16) != hw->device_id)
+		errno = -EINVAL;
+	else if (test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state) ||
+		 test_bit(__I40E_RESET_INTR_RECEIVED, &pf->state))
+		errno = -EBUSY;
+	else
+		ret_val = i40e_nvmupd_command(hw, cmd, bytes, &errno);
 
-	cmd = (struct i40e_nvm_access *)eeprom;
-	ret_val = i40e_nvmupd_command(hw, cmd, bytes, &errno);
-	if (ret_val && (hw->debug_mask & I40E_DEBUG_NVM))
+	if ((errno || ret_val) && (hw->debug_mask & I40E_DEBUG_NVM))
 		dev_info(&pf->pdev->dev,
 			 "NVMUpdate write failed err=%d status=0x%x errno=%d module=%d offset=0x%x size=%d\n",
 			 ret_val, hw->aq.asq_last_status, errno,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 7869d74..384912b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3475,6 +3475,7 @@ static irqreturn_t i40e_intr(int irq, void *data)
 	if (icr0 & I40E_PFINT_ICR0_ADMINQ_MASK) {
 		ena_mask &= ~I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
 		set_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state);
+		i40e_debug(&pf->hw, I40E_DEBUG_NVM, "AdminQ event\n");
 	}
 
 	if (icr0 & I40E_PFINT_ICR0_MAL_DETECT_MASK) {
@@ -6332,7 +6333,9 @@ static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
 		case i40e_aqc_opc_nvm_erase:
 		case i40e_aqc_opc_nvm_update:
 		case i40e_aqc_opc_oem_post_update:
-			i40e_debug(&pf->hw, I40E_DEBUG_NVM, "ARQ NVM operation completed\n");
+			i40e_debug(&pf->hw, I40E_DEBUG_NVM,
+				   "ARQ NVM operation 0x%04x completed\n",
+				   opcode);
 			break;
 		default:
 			dev_info(&pf->pdev->dev,
-- 
cgit v0.10.2


From 9b9344f7ce19c5c481f203a0af3e120b965afcd9 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Fri, 15 Jan 2016 14:33:19 -0800
Subject: i40evf: set adapter state on reset failure

If a reset fails to complete, the driver gets its affairs in order and
awaits the cold solace of rmmod. Unfortunately, it was not properly
setting the adapter state, which would cause a panic on rmmod, instead
of the desired surcease.

Set the adapter state to DOWN in this case, and avoid a panic.

Change-ID: I6fdd9906da52e023f8dc744f7da44b5d95278ca9
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 1176326..cbb7507 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1883,6 +1883,7 @@ static void i40evf_reset_task(struct work_struct *work)
 		adapter->netdev->flags &= ~IFF_UP;
 		clear_bit(__I40EVF_IN_CRITICAL_TASK, &adapter->crit_section);
 		adapter->flags &= ~I40EVF_FLAG_RESET_PENDING;
+		adapter->state = __I40EVF_DOWN;
 		dev_info(&adapter->pdev->dev, "Reset task did not complete, VF disabled\n");
 		return; /* Do not attempt to reinit. It's dead, Jim. */
 	}
-- 
cgit v0.10.2


From 73b03f9848aa7556c8a10fd6ad4e64254e493445 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Fri, 15 Jan 2016 14:33:20 -0800
Subject: i40e: clean event descriptor before use

In one obscure corner case, it was possible to clear the NVM update wait
flag when no update_done message was actually received.  This patch
cleans the event descriptor before use, and moves the opcode check to
where it won't get done if there was no event to clean.

Also update copyright year in file headers.

Change-ID: I68bbc41965e93f4adf07cbe98b9dfd63d41509a4
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c
index 1fd5ea8..df8e2fd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -953,6 +953,9 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 	u16 flags;
 	u16 ntu;
 
+	/* pre-clean the event info */
+	memset(&e->desc, 0, sizeof(e->desc));
+
 	/* take the lock before we start messing with the ring */
 	mutex_lock(&hw->aq.arq_mutex);
 
@@ -1020,14 +1023,6 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw,
 	hw->aq.arq.next_to_clean = ntc;
 	hw->aq.arq.next_to_use = ntu;
 
-clean_arq_element_out:
-	/* Set pending if needed, unlock and return */
-	if (pending != NULL)
-		*pending = (ntc > ntu ? hw->aq.arq.count : 0) + (ntu - ntc);
-
-clean_arq_element_err:
-	mutex_unlock(&hw->aq.arq_mutex);
-
 	if (i40e_is_nvm_update_op(&e->desc)) {
 		if (hw->aq.nvm_release_on_done) {
 			i40e_release_nvm(hw);
@@ -1048,6 +1043,13 @@ clean_arq_element_err:
 		}
 	}
 
+clean_arq_element_out:
+	/* Set pending if needed, unlock and return */
+	if (pending)
+		*pending = (ntc > ntu ? hw->aq.arq.count : 0) + (ntu - ntc);
+clean_arq_element_err:
+	mutex_unlock(&hw->aq.arq_mutex);
+
 	return ret_code;
 }
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c
index 3f65e39..44f7ed7 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -887,6 +887,9 @@ i40e_status i40evf_clean_arq_element(struct i40e_hw *hw,
 	u16 flags;
 	u16 ntu;
 
+	/* pre-clean the event info */
+	memset(&e->desc, 0, sizeof(e->desc));
+
 	/* take the lock before we start messing with the ring */
 	mutex_lock(&hw->aq.arq_mutex);
 
-- 
cgit v0.10.2


From 3b1200891b7f41e5725d8659b6f87e1d2060272c Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Fri, 15 Jan 2016 14:33:21 -0800
Subject: i40e: When in promisc mode apply promisc mode to Tx Traffic as well

In MFP mode particularly when we were setting the PF VSI in limited
promiscuous, the HW switch was still mirroring the outgoing packets
from other VSIs (VF/VMdq) onto the PF VSI.

With this new bit set, the mirroring doesn't happen any more and so
we are in limited promiscuous on the PF VSI in MFP which is similar
to defport.

An API check is not required, since this bit is reserved for FW API
version < 1.5

Also update copyright year in file headers.

Change-ID: I9840cb95f11dde733d943cb03ce84f68b9611bc8
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 5699299..bb7ecbb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -1087,6 +1087,7 @@ struct i40e_aqc_set_vsi_promiscuous_modes {
 #define I40E_AQC_SET_VSI_PROMISC_BROADCAST	0x04
 #define I40E_AQC_SET_VSI_DEFAULT		0x08
 #define I40E_AQC_SET_VSI_PROMISC_VLAN		0x10
+#define I40E_AQC_SET_VSI_PROMISC_TX		0x8000
 	__le16	seid;
 #define I40E_AQC_VSI_PROM_CMD_SEID_MASK		0x3FF
 	__le16	vlan_tag;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 354e36c..f923933 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1952,12 +1952,19 @@ i40e_status i40e_aq_set_vsi_unicast_promiscuous(struct i40e_hw *hw,
 	i40e_fill_default_direct_cmd_desc(&desc,
 					i40e_aqc_opc_set_vsi_promiscuous_modes);
 
-	if (set)
+	if (set) {
 		flags |= I40E_AQC_SET_VSI_PROMISC_UNICAST;
+		if (((hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver >= 5)) ||
+		    (hw->aq.api_maj_ver > 1))
+			flags |= I40E_AQC_SET_VSI_PROMISC_TX;
+	}
 
 	cmd->promiscuous_flags = cpu_to_le16(flags);
 
 	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_UNICAST);
+	if (((hw->aq.api_maj_ver >= 1) && (hw->aq.api_min_ver >= 5)) ||
+	    (hw->aq.api_maj_ver > 1))
+		cmd->valid_flags |= cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_TX);
 
 	cmd->seid = cpu_to_le16(seid);
 	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index ed963bc..815e481 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -1,7 +1,7 @@
 /*******************************************************************************
  *
  * Intel Ethernet Controller XL710 Family Linux Virtual Function Driver
- * Copyright(c) 2013 - 2014 Intel Corporation.
+ * Copyright(c) 2013 - 2016 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -1084,6 +1084,7 @@ struct i40e_aqc_set_vsi_promiscuous_modes {
 #define I40E_AQC_SET_VSI_PROMISC_BROADCAST	0x04
 #define I40E_AQC_SET_VSI_DEFAULT		0x08
 #define I40E_AQC_SET_VSI_PROMISC_VLAN		0x10
+#define I40E_AQC_SET_VSI_PROMISC_TX		0x8000
 	__le16	seid;
 #define I40E_AQC_VSI_PROM_CMD_SEID_MASK		0x3FF
 	__le16	vlan_tag;
-- 
cgit v0.10.2


From 8888fd8843ed501a26d9b74de0975b1ace663231 Mon Sep 17 00:00:00 2001
From: Catherine Sullivan <catherine.sullivan@intel.com>
Date: Fri, 15 Jan 2016 14:33:22 -0800
Subject: i40e/i40evf: Bump i40e to 1.4.15 and i40evf to 1.4.11.

Bump.

Change-ID: Ie280dc67e37a1cf667c3469499a4fb90f4177b75
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 384912b..16e5e0b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -46,7 +46,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 13
+#define DRV_VERSION_BUILD 15
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index cbb7507..41369a3 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -38,7 +38,7 @@ static const char i40evf_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 9
+#define DRV_VERSION_BUILD 11
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD) \
-- 
cgit v0.10.2


From fbbef866fce2d81aa1791d0c762afb07dbc4e660 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Mon, 15 Feb 2016 22:54:47 +0000
Subject: net-sysfs: remove unused fmt_long_hex

Ever since commit 04ed3e741d0f133e02bed7fa5c98edba128f90e7
("net: change netdev->features to u32") the format string
fmt_long_hex has not been used, so we may as well remove it.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index da7dbc2..4ae17c3 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -29,7 +29,6 @@
 
 #ifdef CONFIG_SYSFS
 static const char fmt_hex[] = "%#x\n";
-static const char fmt_long_hex[] = "%#lx\n";
 static const char fmt_dec[] = "%d\n";
 static const char fmt_ulong[] = "%lu\n";
 static const char fmt_u64[] = "%llu\n";
-- 
cgit v0.10.2


From 096f6262c573dfd9ede478a2ad76e6b9b994c9a0 Mon Sep 17 00:00:00 2001
From: Alison Schofield <amsfield22@gmail.com>
Date: Wed, 17 Feb 2016 22:35:11 -0800
Subject: isdn: divamnt: use y2038-safe ktime_get_ts64() for trace data
 timestamps

divamnt stores a start_time at module init and uses it to calculate
elapsed time. The elapsed time, stored in secs and usecs, is part of
the trace data the driver maintains for the DIVA Server ISDN cards.
No change to the format of that time data is required.

To avoid overflow on 32-bit systems use ktime_get_ts64() to return
the elapsed monotonic time since system boot.

This is a change from real to monotonic time. Since the driver only
stores elapsed time, monotonic time is sufficient and more robust
against real time clock changes. These new monotonic values can be
more useful for debugging because they can be easily compared to
other monotonic timestamps.

Note elaspsed time values will now start at system boot time rather
than module load time, so they will differ slightly from previously
reported values.

Remove declaration and init of previously unused time constants:
start_sec, start_usec.

Signed-off-by: Alison Schofield <amsfield22@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/isdn/hardware/eicon/debug.c b/drivers/isdn/hardware/eicon/debug.c
index b5226af..576b7b4 100644
--- a/drivers/isdn/hardware/eicon/debug.c
+++ b/drivers/isdn/hardware/eicon/debug.c
@@ -192,8 +192,6 @@ static diva_os_spin_lock_t dbg_q_lock;
 static diva_os_spin_lock_t dbg_adapter_lock;
 static int                 dbg_q_busy;
 static volatile dword      dbg_sequence;
-static dword               start_sec;
-static dword               start_usec;
 
 /*
   INTERFACE:
@@ -215,8 +213,6 @@ int diva_maint_init(byte *base, unsigned long length, int do_init) {
 
 	dbg_base = base;
 
-	diva_os_get_time(&start_sec, &start_usec);
-
 	*(dword *)base  = (dword)DBG_MAGIC; /* Store Magic */
 	base   += sizeof(dword);
 	length -= sizeof(dword);
diff --git a/drivers/isdn/hardware/eicon/divamnt.c b/drivers/isdn/hardware/eicon/divamnt.c
index 48db08d..0de29b7b 100644
--- a/drivers/isdn/hardware/eicon/divamnt.c
+++ b/drivers/isdn/hardware/eicon/divamnt.c
@@ -45,7 +45,6 @@ char *DRIVERRELEASE_MNT = "2.0";
 
 static wait_queue_head_t msgwaitq;
 static unsigned long opened;
-static struct timeval start_time;
 
 extern int mntfunc_init(int *, void **, unsigned long);
 extern void mntfunc_finit(void);
@@ -88,28 +87,12 @@ int diva_os_copy_from_user(void *os_handle, void *dst, const void __user *src,
  */
 void diva_os_get_time(dword *sec, dword *usec)
 {
-	struct timeval tv;
-
-	do_gettimeofday(&tv);
-
-	if (tv.tv_sec > start_time.tv_sec) {
-		if (start_time.tv_usec > tv.tv_usec) {
-			tv.tv_sec--;
-			tv.tv_usec += 1000000;
-		}
-		*sec = (dword) (tv.tv_sec - start_time.tv_sec);
-		*usec = (dword) (tv.tv_usec - start_time.tv_usec);
-	} else if (tv.tv_sec == start_time.tv_sec) {
-		*sec = 0;
-		if (start_time.tv_usec < tv.tv_usec) {
-			*usec = (dword) (tv.tv_usec - start_time.tv_usec);
-		} else {
-			*usec = 0;
-		}
-	} else {
-		*sec = (dword) tv.tv_sec;
-		*usec = (dword) tv.tv_usec;
-	}
+	struct timespec64 time;
+
+	ktime_get_ts64(&time);
+
+	*sec = (dword) time.tv_sec;
+	*usec = (dword) (time.tv_nsec / NSEC_PER_USEC);
 }
 
 /*
@@ -213,7 +196,6 @@ static int __init maint_init(void)
 	int ret = 0;
 	void *buffer = NULL;
 
-	do_gettimeofday(&start_time);
 	init_waitqueue_head(&msgwaitq);
 
 	printk(KERN_INFO "%s\n", DRIVERNAME);
-- 
cgit v0.10.2


From 930b37ee8d84e1f3008c89e9ce09acbe1a6b5670 Mon Sep 17 00:00:00 2001
From: Stefan Roese <sr@denx.de>
Date: Thu, 18 Feb 2016 10:59:07 +0100
Subject: net: phy: Add SGMII support for Marvell 88E1510/1512/1514/1518

Add code to select SGMII-to-copper mode upon SGMII interface selection.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index e3eb964..1dcbd3f 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -133,6 +133,11 @@
 #define MII_88E3016_DISABLE_SCRAMBLER	0x0200
 #define MII_88E3016_AUTO_MDIX_CROSSOVER	0x0030
 
+#define MII_88E1510_GEN_CTRL_REG_1		0x14
+#define MII_88E1510_GEN_CTRL_REG_1_MODE_MASK	0x7
+#define MII_88E1510_GEN_CTRL_REG_1_MODE_SGMII	0x1	/* SGMII to copper */
+#define MII_88E1510_GEN_CTRL_REG_1_RESET	0x8000	/* Soft reset */
+
 MODULE_DESCRIPTION("Marvell PHY driver");
 MODULE_AUTHOR("Andy Fleming");
 MODULE_LICENSE("GPL");
@@ -438,6 +443,41 @@ static int m88e1318_config_aneg(struct phy_device *phydev)
 	return m88e1121_config_aneg(phydev);
 }
 
+static int m88e1510_config_init(struct phy_device *phydev)
+{
+	int err;
+	int temp;
+
+	/* SGMII-to-Copper mode initialization */
+	if (phydev->interface == PHY_INTERFACE_MODE_SGMII) {
+		/* Select page 18 */
+		err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 18);
+		if (err < 0)
+			return err;
+
+		/* In reg 20, write MODE[2:0] = 0x1 (SGMII to Copper) */
+		temp = phy_read(phydev, MII_88E1510_GEN_CTRL_REG_1);
+		temp &= ~MII_88E1510_GEN_CTRL_REG_1_MODE_MASK;
+		temp |= MII_88E1510_GEN_CTRL_REG_1_MODE_SGMII;
+		err = phy_write(phydev, MII_88E1510_GEN_CTRL_REG_1, temp);
+		if (err < 0)
+			return err;
+
+		/* PHY reset is necessary after changing MODE[2:0] */
+		temp |= MII_88E1510_GEN_CTRL_REG_1_RESET;
+		err = phy_write(phydev, MII_88E1510_GEN_CTRL_REG_1, temp);
+		if (err < 0)
+			return err;
+
+		/* Reset page selection */
+		err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0);
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+}
+
 static int m88e1510_config_aneg(struct phy_device *phydev)
 {
 	int err;
@@ -1259,6 +1299,7 @@ static struct phy_driver marvell_drivers[] = {
 		.features = PHY_GBIT_FEATURES,
 		.flags = PHY_HAS_INTERRUPT,
 		.probe = marvell_probe,
+		.config_init = &m88e1510_config_init,
 		.config_aneg = &m88e1510_config_aneg,
 		.read_status = &marvell_read_status,
 		.ack_interrupt = &marvell_ack_interrupt,
-- 
cgit v0.10.2


From 7fbac984f33a374a06fa7b1fc41127bdaf7d6c07 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Thu, 18 Feb 2016 14:01:46 +0100
Subject: bridge: switchdev: Offload VLAN flags to hardware bridge

When VLANs are created / destroyed on a VLAN filtering bridge (MASTER
flag set), the configuration is passed down to the hardware. However,
when only the flags (e.g. PVID) are toggled, the configuration is done
in the software bridge alone.

While it is possible to pass these flags to hardware when invoked with
the SELF flag set, this creates inconsistency with regards to the way
the VLANs are initially configured.

Pass the flags down to the hardware even when the VLAN already exists
and only the flags are toggled.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 85e43af..9309bb4 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -955,6 +955,13 @@ err_rhtbl:
  */
 int nbp_vlan_add(struct net_bridge_port *port, u16 vid, u16 flags)
 {
+	struct switchdev_obj_port_vlan v = {
+		.obj.orig_dev = port->dev,
+		.obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN,
+		.flags = flags,
+		.vid_begin = vid,
+		.vid_end = vid,
+	};
 	struct net_bridge_vlan *vlan;
 	int ret;
 
@@ -962,6 +969,10 @@ int nbp_vlan_add(struct net_bridge_port *port, u16 vid, u16 flags)
 
 	vlan = br_vlan_find(nbp_vlan_group(port), vid);
 	if (vlan) {
+		/* Pass the flags to the hardware bridge */
+		ret = switchdev_port_obj_add(port->dev, &v.obj);
+		if (ret && ret != -EOPNOTSUPP)
+			return ret;
 		__vlan_add_flags(vlan, flags);
 		return 0;
 	}
-- 
cgit v0.10.2


From 7e6e18fbc033e00a4d4af3d4ea7bad0db6b7ad1b Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <jhs@mojatatu.com>
Date: Thu, 18 Feb 2016 08:04:43 -0500
Subject: net_sched: Improve readability of filter processing

Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index b5c2cf2..c9673b5e 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1841,7 +1841,7 @@ reclassify:
 			return err;
 	}
 
-	return -1;
+	return TC_ACT_UNSPEC; /* signal: continue lookup */
 #ifdef CONFIG_NET_CLS_ACT
 reset:
 	if (unlikely(limit++ >= MAX_REC_LOOP)) {
-- 
cgit v0.10.2


From d1b4c689d4130bcfd3532680b64db562300716b6 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 18 Feb 2016 15:03:24 +0100
Subject: netlink: remove mmapped netlink support

mmapped netlink has a number of unresolved issues:

- TX zerocopy support had to be disabled more than a year ago via
  commit 4682a0358639b29cf ("netlink: Always copy on mmap TX.")
  because the content of the mmapped area can change after netlink
  attribute validation but before message processing.

- RX support was implemented mainly to speed up nfqueue dumping packet
  payload to userspace.  However, since commit ae08ce0021087a5d812d2
  ("netfilter: nfnetlink_queue: zero copy support") we avoid one copy
  with the socket-based interface too (via the skb_zerocopy helper).

The other problem is that skbs attached to mmaped netlink socket
behave different from normal skbs:

- they don't have a shinfo area, so all functions that use skb_shinfo()
(e.g. skb_clone) cannot be used.

- reserving headroom prevents userspace from seeing the content as
it expects message to start at skb->head.
See for instance
commit aa3a022094fa ("netlink: not trim skb for mmaped socket when dump").

- skbs handed e.g. to netlink_ack must have non-NULL skb->sk, else we
crash because it needs the sk to check if a tx ring is attached.

Also not obvious, leads to non-intuitive bug fixes such as 7c7bdf359
("netfilter: nfnetlink: use original skbuff when acking batches").

mmaped netlink also didn't play nicely with the skb_zerocopy helper
used by nfqueue and openvswitch.  Daniel Borkmann fixed this via
commit 6bb0fef489f6 ("netlink, mmap: fix edge-case leakages in nf queue
zero-copy")' but at the cost of also needing to provide remaining
length to the allocation function.

nfqueue also has problems when used with mmaped rx netlink:
- mmaped netlink doesn't allow use of nfqueue batch verdict messages.
  Problem is that in the mmap case, the allocation time also determines
  the ordering in which the frame will be seen by userspace (A
  allocating before B means that A is located in earlier ring slot,
  but this also means that B might get a lower sequence number then A
  since seqno is decided later.  To fix this we would need to extend the
  spinlocked region to also cover the allocation and message setup which
  isn't desirable.
- nfqueue can now be configured to queue large (GSO) skbs to userspace.
  Queing GSO packets is faster than having to force a software segmentation
  in the kernel, so this is a desirable option.  However, with a mmap based
  ring one has to use 64kb per ring slot element, else mmap has to fall back
  to the socket path (NL_MMAP_STATUS_COPY) for all large packets.

To use the mmap interface, userspace not only has to probe for mmap netlink
support, it also has to implement a recv/socket receive path in order to
handle messages that exceed the size of an rx ring element.

Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Ken-ichirou MATSUZAWA <chamaken@gmail.com>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Thomas Graf <tgraf@suug.ch>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/netlink_mmap.txt b/Documentation/networking/netlink_mmap.txt
deleted file mode 100644
index 54f1047..0000000
--- a/Documentation/networking/netlink_mmap.txt
+++ /dev/null
@@ -1,332 +0,0 @@
-This file documents how to use memory mapped I/O with netlink.
-
-Author: Patrick McHardy <kaber@trash.net>
-
-Overview
---------
-
-Memory mapped netlink I/O can be used to increase throughput and decrease
-overhead of unicast receive and transmit operations. Some netlink subsystems
-require high throughput, these are mainly the netfilter subsystems
-nfnetlink_queue and nfnetlink_log, but it can also help speed up large
-dump operations of f.i. the routing database.
-
-Memory mapped netlink I/O used two circular ring buffers for RX and TX which
-are mapped into the processes address space.
-
-The RX ring is used by the kernel to directly construct netlink messages into
-user-space memory without copying them as done with regular socket I/O,
-additionally as long as the ring contains messages no recvmsg() or poll()
-syscalls have to be issued by user-space to get more message.
-
-The TX ring is used to process messages directly from user-space memory, the
-kernel processes all messages contained in the ring using a single sendmsg()
-call.
-
-Usage overview
---------------
-
-In order to use memory mapped netlink I/O, user-space needs three main changes:
-
-- ring setup
-- conversion of the RX path to get messages from the ring instead of recvmsg()
-- conversion of the TX path to construct messages into the ring
-
-Ring setup is done using setsockopt() to provide the ring parameters to the
-kernel, then a call to mmap() to map the ring into the processes address space:
-
-- setsockopt(fd, SOL_NETLINK, NETLINK_RX_RING, &params, sizeof(params));
-- setsockopt(fd, SOL_NETLINK, NETLINK_TX_RING, &params, sizeof(params));
-- ring = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0)
-
-Usage of either ring is optional, but even if only the RX ring is used the
-mapping still needs to be writable in order to update the frame status after
-processing.
-
-Conversion of the reception path involves calling poll() on the file
-descriptor, once the socket is readable the frames from the ring are
-processed in order until no more messages are available, as indicated by
-a status word in the frame header.
-
-On kernel side, in order to make use of memory mapped I/O on receive, the
-originating netlink subsystem needs to support memory mapped I/O, otherwise
-it will use an allocated socket buffer as usual and the contents will be
- copied to the ring on transmission, nullifying most of the performance gains.
-Dumps of kernel databases automatically support memory mapped I/O.
-
-Conversion of the transmit path involves changing message construction to
-use memory from the TX ring instead of (usually) a buffer declared on the
-stack and setting up the frame header appropriately. Optionally poll() can
-be used to wait for free frames in the TX ring.
-
-Structured and definitions for using memory mapped I/O are contained in
-<linux/netlink.h>.
-
-RX and TX rings
-----------------
-
-Each ring contains a number of continuous memory blocks, containing frames of
-fixed size dependent on the parameters used for ring setup.
-
-Ring:	[ block 0 ]
-		[ frame 0 ]
-		[ frame 1 ]
-	[ block 1 ]
-		[ frame 2 ]
-		[ frame 3 ]
-	...
-	[ block n ]
-		[ frame 2 * n ]
-		[ frame 2 * n + 1 ]
-
-The blocks are only visible to the kernel, from the point of view of user-space
-the ring just contains the frames in a continuous memory zone.
-
-The ring parameters used for setting up the ring are defined as follows:
-
-struct nl_mmap_req {
-	unsigned int	nm_block_size;
-	unsigned int	nm_block_nr;
-	unsigned int	nm_frame_size;
-	unsigned int	nm_frame_nr;
-};
-
-Frames are grouped into blocks, where each block is a continuous region of memory
-and holds nm_block_size / nm_frame_size frames. The total number of frames in
-the ring is nm_frame_nr. The following invariants hold:
-
-- frames_per_block = nm_block_size / nm_frame_size
-
-- nm_frame_nr = frames_per_block * nm_block_nr
-
-Some parameters are constrained, specifically:
-
-- nm_block_size must be a multiple of the architectures memory page size.
-  The getpagesize() function can be used to get the page size.
-
-- nm_frame_size must be equal or larger to NL_MMAP_HDRLEN, IOW a frame must be
-  able to hold at least the frame header
-
-- nm_frame_size must be smaller or equal to nm_block_size
-
-- nm_frame_size must be a multiple of NL_MMAP_MSG_ALIGNMENT
-
-- nm_frame_nr must equal the actual number of frames as specified above.
-
-When the kernel can't allocate physically continuous memory for a ring block,
-it will fall back to use physically discontinuous memory. This might affect
-performance negatively, in order to avoid this the nm_frame_size parameter
-should be chosen to be as small as possible for the required frame size and
-the number of blocks should be increased instead.
-
-Ring frames
-------------
-
-Each frames contain a frame header, consisting of a synchronization word and some
-meta-data, and the message itself.
-
-Frame:	[ header message ]
-
-The frame header is defined as follows:
-
-struct nl_mmap_hdr {
-	unsigned int	nm_status;
-	unsigned int	nm_len;
-	__u32		nm_group;
-	/* credentials */
-	__u32		nm_pid;
-	__u32		nm_uid;
-	__u32		nm_gid;
-};
-
-- nm_status is used for synchronizing processing between the kernel and user-
-  space and specifies ownership of the frame as well as the operation to perform
-
-- nm_len contains the length of the message contained in the data area
-
-- nm_group specified the destination multicast group of message
-
-- nm_pid, nm_uid and nm_gid contain the netlink pid, UID and GID of the sending
-  process. These values correspond to the data available using SOCK_PASSCRED in
-  the SCM_CREDENTIALS cmsg.
-
-The possible values in the status word are:
-
-- NL_MMAP_STATUS_UNUSED:
-	RX ring:	frame belongs to the kernel and contains no message
-			for user-space. Approriate action is to invoke poll()
-			to wait for new messages.
-
-	TX ring:	frame belongs to user-space and can be used for
-			message construction.
-
-- NL_MMAP_STATUS_RESERVED:
-	RX ring only:	frame is currently used by the kernel for message
-			construction and contains no valid message yet.
-			Appropriate action is to invoke poll() to wait for
-			new messages.
-
-- NL_MMAP_STATUS_VALID:
-	RX ring:	frame contains a valid message. Approriate action is
-			to process the message and release the frame back to
-			the kernel by setting the status to
-			NL_MMAP_STATUS_UNUSED or queue the frame by setting the
-			status to NL_MMAP_STATUS_SKIP.
-
-	TX ring:	the frame contains a valid message from user-space to
-			be processed by the kernel. After completing processing
-			the kernel will release the frame back to user-space by
-			setting the status to NL_MMAP_STATUS_UNUSED.
-
-- NL_MMAP_STATUS_COPY:
-	RX ring only:	a message is ready to be processed but could not be
-			stored in the ring, either because it exceeded the
-			frame size or because the originating subsystem does
-			not support memory mapped I/O. Appropriate action is
-			to invoke recvmsg() to receive the message and release
-			the frame back to the kernel by setting the status to
-			NL_MMAP_STATUS_UNUSED.
-
-- NL_MMAP_STATUS_SKIP:
-	RX ring only:	user-space queued the message for later processing, but
-			processed some messages following it in the ring. The
-			kernel should skip this frame when looking for unused
-			frames.
-
-The data area of a frame begins at a offset of NL_MMAP_HDRLEN relative to the
-frame header.
-
-TX limitations
---------------
-
-As of Jan 2015 the message is always copied from the ring frame to an
-allocated buffer due to unresolved security concerns.
-See commit 4682a0358639b29cf ("netlink: Always copy on mmap TX.").
-
-Example
--------
-
-Ring setup:
-
-	unsigned int block_size = 16 * getpagesize();
-	struct nl_mmap_req req = {
-		.nm_block_size		= block_size,
-		.nm_block_nr		= 64,
-		.nm_frame_size		= 16384,
-		.nm_frame_nr		= 64 * block_size / 16384,
-	};
-	unsigned int ring_size;
-	void *rx_ring, *tx_ring;
-
-	/* Configure ring parameters */
-	if (setsockopt(fd, SOL_NETLINK, NETLINK_RX_RING, &req, sizeof(req)) < 0)
-		exit(1);
-	if (setsockopt(fd, SOL_NETLINK, NETLINK_TX_RING, &req, sizeof(req)) < 0)
-		exit(1)
-
-	/* Calculate size of each individual ring */
-	ring_size = req.nm_block_nr * req.nm_block_size;
-
-	/* Map RX/TX rings. The TX ring is located after the RX ring */
-	rx_ring = mmap(NULL, 2 * ring_size, PROT_READ | PROT_WRITE,
-		       MAP_SHARED, fd, 0);
-	if ((long)rx_ring == -1L)
-		exit(1);
-	tx_ring = rx_ring + ring_size:
-
-Message reception:
-
-This example assumes some ring parameters of the ring setup are available.
-
-	unsigned int frame_offset = 0;
-	struct nl_mmap_hdr *hdr;
-	struct nlmsghdr *nlh;
-	unsigned char buf[16384];
-	ssize_t len;
-
-	while (1) {
-		struct pollfd pfds[1];
-
-		pfds[0].fd	= fd;
-		pfds[0].events	= POLLIN | POLLERR;
-		pfds[0].revents	= 0;
-
-		if (poll(pfds, 1, -1) < 0 && errno != -EINTR)
-			exit(1);
-
-		/* Check for errors. Error handling omitted */
-		if (pfds[0].revents & POLLERR)
-			<handle error>
-
-		/* If no new messages, poll again */
-		if (!(pfds[0].revents & POLLIN))
-			continue;
-
-		/* Process all frames */
-		while (1) {
-			/* Get next frame header */
-			hdr = rx_ring + frame_offset;
-
-			if (hdr->nm_status == NL_MMAP_STATUS_VALID) {
-				/* Regular memory mapped frame */
-				nlh = (void *)hdr + NL_MMAP_HDRLEN;
-				len = hdr->nm_len;
-
-				/* Release empty message immediately. May happen
-				 * on error during message construction.
-				 */
-				if (len == 0)
-					goto release;
-			} else if (hdr->nm_status == NL_MMAP_STATUS_COPY) {
-				/* Frame queued to socket receive queue */
-				len = recv(fd, buf, sizeof(buf), MSG_DONTWAIT);
-				if (len <= 0)
-					break;
-				nlh = buf;
-			} else
-				/* No more messages to process, continue polling */
-				break;
-
-			process_msg(nlh);
-release:
-			/* Release frame back to the kernel */
-			hdr->nm_status = NL_MMAP_STATUS_UNUSED;
-
-			/* Advance frame offset to next frame */
-			frame_offset = (frame_offset + frame_size) % ring_size;
-		}
-	}
-
-Message transmission:
-
-This example assumes some ring parameters of the ring setup are available.
-A single message is constructed and transmitted, to send multiple messages
-at once they would be constructed in consecutive frames before a final call
-to sendto().
-
-	unsigned int frame_offset = 0;
-	struct nl_mmap_hdr *hdr;
-	struct nlmsghdr *nlh;
-	struct sockaddr_nl addr = {
-		.nl_family	= AF_NETLINK,
-	};
-
-	hdr = tx_ring + frame_offset;
-	if (hdr->nm_status != NL_MMAP_STATUS_UNUSED)
-		/* No frame available. Use poll() to avoid. */
-		exit(1);
-
-	nlh = (void *)hdr + NL_MMAP_HDRLEN;
-
-	/* Build message */
-	build_message(nlh);
-
-	/* Fill frame header: length and status need to be set */
-	hdr->nm_len	= nlh->nlmsg_len;
-	hdr->nm_status	= NL_MMAP_STATUS_VALID;
-
-	if (sendto(fd, NULL, 0, 0, &addr, sizeof(addr)) < 0)
-		exit(1);
-
-	/* Advance frame offset to next frame */
-	frame_offset = (frame_offset + frame_size) % ring_size;
diff --git a/include/uapi/linux/netlink.h b/include/uapi/linux/netlink.h
index f095155..0dba4e4 100644
--- a/include/uapi/linux/netlink.h
+++ b/include/uapi/linux/netlink.h
@@ -107,8 +107,10 @@ struct nlmsgerr {
 #define NETLINK_PKTINFO			3
 #define NETLINK_BROADCAST_ERROR		4
 #define NETLINK_NO_ENOBUFS		5
+#ifndef __KERNEL__
 #define NETLINK_RX_RING			6
 #define NETLINK_TX_RING			7
+#endif
 #define NETLINK_LISTEN_ALL_NSID		8
 #define NETLINK_LIST_MEMBERSHIPS	9
 #define NETLINK_CAP_ACK			10
@@ -134,6 +136,7 @@ struct nl_mmap_hdr {
 	__u32		nm_gid;
 };
 
+#ifndef __KERNEL__
 enum nl_mmap_status {
 	NL_MMAP_STATUS_UNUSED,
 	NL_MMAP_STATUS_RESERVED,
@@ -145,6 +148,7 @@ enum nl_mmap_status {
 #define NL_MMAP_MSG_ALIGNMENT		NLMSG_ALIGNTO
 #define NL_MMAP_MSG_ALIGN(sz)		__ALIGN_KERNEL(sz, NL_MMAP_MSG_ALIGNMENT)
 #define NL_MMAP_HDRLEN			NL_MMAP_MSG_ALIGN(sizeof(struct nl_mmap_hdr))
+#endif
 
 #define NET_MAJOR 36		/* Major 36 is reserved for networking 						*/
 
diff --git a/include/uapi/linux/netlink_diag.h b/include/uapi/linux/netlink_diag.h
index f2159d3..d793993 100644
--- a/include/uapi/linux/netlink_diag.h
+++ b/include/uapi/linux/netlink_diag.h
@@ -48,6 +48,8 @@ enum {
 
 #define NDIAG_SHOW_MEMINFO	0x00000001 /* show memory info of a socket */
 #define NDIAG_SHOW_GROUPS	0x00000002 /* show groups of a netlink socket */
+#ifndef __KERNEL__
 #define NDIAG_SHOW_RING_CFG	0x00000004 /* show ring configuration */
+#endif
 
 #endif
diff --git a/net/netlink/Kconfig b/net/netlink/Kconfig
index 2c5e95e..5d6e8c0 100644
--- a/net/netlink/Kconfig
+++ b/net/netlink/Kconfig
@@ -2,15 +2,6 @@
 # Netlink Sockets
 #
 
-config NETLINK_MMAP
-	bool "NETLINK: mmaped IO"
-	---help---
-	  This option enables support for memory mapped netlink IO. This
-	  reduces overhead by avoiding copying data between kernel- and
-	  userspace.
-
-	  If unsure, say N.
-
 config NETLINK_DIAG
 	tristate "NETLINK: socket monitoring interface"
 	default n
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index f1ffb34..85aa6ef 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -225,7 +225,7 @@ static int __netlink_deliver_tap_skb(struct sk_buff *skb,
 
 	dev_hold(dev);
 
-	if (netlink_skb_is_mmaped(skb) || is_vmalloc_addr(skb->head))
+	if (is_vmalloc_addr(skb->head))
 		nskb = netlink_to_full_skb(skb, GFP_ATOMIC);
 	else
 		nskb = skb_clone(skb, GFP_ATOMIC);
@@ -300,610 +300,8 @@ static void netlink_rcv_wake(struct sock *sk)
 		wake_up_interruptible(&nlk->wait);
 }
 
-#ifdef CONFIG_NETLINK_MMAP
-static bool netlink_rx_is_mmaped(struct sock *sk)
-{
-	return nlk_sk(sk)->rx_ring.pg_vec != NULL;
-}
-
-static bool netlink_tx_is_mmaped(struct sock *sk)
-{
-	return nlk_sk(sk)->tx_ring.pg_vec != NULL;
-}
-
-static __pure struct page *pgvec_to_page(const void *addr)
-{
-	if (is_vmalloc_addr(addr))
-		return vmalloc_to_page(addr);
-	else
-		return virt_to_page(addr);
-}
-
-static void free_pg_vec(void **pg_vec, unsigned int order, unsigned int len)
-{
-	unsigned int i;
-
-	for (i = 0; i < len; i++) {
-		if (pg_vec[i] != NULL) {
-			if (is_vmalloc_addr(pg_vec[i]))
-				vfree(pg_vec[i]);
-			else
-				free_pages((unsigned long)pg_vec[i], order);
-		}
-	}
-	kfree(pg_vec);
-}
-
-static void *alloc_one_pg_vec_page(unsigned long order)
-{
-	void *buffer;
-	gfp_t gfp_flags = GFP_KERNEL | __GFP_COMP | __GFP_ZERO |
-			  __GFP_NOWARN | __GFP_NORETRY;
-
-	buffer = (void *)__get_free_pages(gfp_flags, order);
-	if (buffer != NULL)
-		return buffer;
-
-	buffer = vzalloc((1 << order) * PAGE_SIZE);
-	if (buffer != NULL)
-		return buffer;
-
-	gfp_flags &= ~__GFP_NORETRY;
-	return (void *)__get_free_pages(gfp_flags, order);
-}
-
-static void **alloc_pg_vec(struct netlink_sock *nlk,
-			   struct nl_mmap_req *req, unsigned int order)
-{
-	unsigned int block_nr = req->nm_block_nr;
-	unsigned int i;
-	void **pg_vec;
-
-	pg_vec = kcalloc(block_nr, sizeof(void *), GFP_KERNEL);
-	if (pg_vec == NULL)
-		return NULL;
-
-	for (i = 0; i < block_nr; i++) {
-		pg_vec[i] = alloc_one_pg_vec_page(order);
-		if (pg_vec[i] == NULL)
-			goto err1;
-	}
-
-	return pg_vec;
-err1:
-	free_pg_vec(pg_vec, order, block_nr);
-	return NULL;
-}
-
-
-static void
-__netlink_set_ring(struct sock *sk, struct nl_mmap_req *req, bool tx_ring, void **pg_vec,
-		   unsigned int order)
-{
-	struct netlink_sock *nlk = nlk_sk(sk);
-	struct sk_buff_head *queue;
-	struct netlink_ring *ring;
-
-	queue = tx_ring ? &sk->sk_write_queue : &sk->sk_receive_queue;
-	ring  = tx_ring ? &nlk->tx_ring : &nlk->rx_ring;
-
-	spin_lock_bh(&queue->lock);
-
-	ring->frame_max		= req->nm_frame_nr - 1;
-	ring->head		= 0;
-	ring->frame_size	= req->nm_frame_size;
-	ring->pg_vec_pages	= req->nm_block_size / PAGE_SIZE;
-
-	swap(ring->pg_vec_len, req->nm_block_nr);
-	swap(ring->pg_vec_order, order);
-	swap(ring->pg_vec, pg_vec);
-
-	__skb_queue_purge(queue);
-	spin_unlock_bh(&queue->lock);
-
-	WARN_ON(atomic_read(&nlk->mapped));
-
-	if (pg_vec)
-		free_pg_vec(pg_vec, order, req->nm_block_nr);
-}
-
-static int netlink_set_ring(struct sock *sk, struct nl_mmap_req *req,
-			    bool tx_ring)
-{
-	struct netlink_sock *nlk = nlk_sk(sk);
-	struct netlink_ring *ring;
-	void **pg_vec = NULL;
-	unsigned int order = 0;
-
-	ring  = tx_ring ? &nlk->tx_ring : &nlk->rx_ring;
-
-	if (atomic_read(&nlk->mapped))
-		return -EBUSY;
-	if (atomic_read(&ring->pending))
-		return -EBUSY;
-
-	if (req->nm_block_nr) {
-		if (ring->pg_vec != NULL)
-			return -EBUSY;
-
-		if ((int)req->nm_block_size <= 0)
-			return -EINVAL;
-		if (!PAGE_ALIGNED(req->nm_block_size))
-			return -EINVAL;
-		if (req->nm_frame_size < NL_MMAP_HDRLEN)
-			return -EINVAL;
-		if (!IS_ALIGNED(req->nm_frame_size, NL_MMAP_MSG_ALIGNMENT))
-			return -EINVAL;
-
-		ring->frames_per_block = req->nm_block_size /
-					 req->nm_frame_size;
-		if (ring->frames_per_block == 0)
-			return -EINVAL;
-		if (ring->frames_per_block * req->nm_block_nr !=
-		    req->nm_frame_nr)
-			return -EINVAL;
-
-		order = get_order(req->nm_block_size);
-		pg_vec = alloc_pg_vec(nlk, req, order);
-		if (pg_vec == NULL)
-			return -ENOMEM;
-	} else {
-		if (req->nm_frame_nr)
-			return -EINVAL;
-	}
-
-	mutex_lock(&nlk->pg_vec_lock);
-	if (atomic_read(&nlk->mapped) == 0) {
-		__netlink_set_ring(sk, req, tx_ring, pg_vec, order);
-		mutex_unlock(&nlk->pg_vec_lock);
-		return 0;
-	}
-
-	mutex_unlock(&nlk->pg_vec_lock);
-
-	if (pg_vec)
-		free_pg_vec(pg_vec, order, req->nm_block_nr);
-
-	return -EBUSY;
-}
-
-static void netlink_mm_open(struct vm_area_struct *vma)
-{
-	struct file *file = vma->vm_file;
-	struct socket *sock = file->private_data;
-	struct sock *sk = sock->sk;
-
-	if (sk)
-		atomic_inc(&nlk_sk(sk)->mapped);
-}
-
-static void netlink_mm_close(struct vm_area_struct *vma)
-{
-	struct file *file = vma->vm_file;
-	struct socket *sock = file->private_data;
-	struct sock *sk = sock->sk;
-
-	if (sk)
-		atomic_dec(&nlk_sk(sk)->mapped);
-}
-
-static const struct vm_operations_struct netlink_mmap_ops = {
-	.open	= netlink_mm_open,
-	.close	= netlink_mm_close,
-};
-
-static int netlink_mmap(struct file *file, struct socket *sock,
-			struct vm_area_struct *vma)
-{
-	struct sock *sk = sock->sk;
-	struct netlink_sock *nlk = nlk_sk(sk);
-	struct netlink_ring *ring;
-	unsigned long start, size, expected;
-	unsigned int i;
-	int err = -EINVAL;
-
-	if (vma->vm_pgoff)
-		return -EINVAL;
-
-	mutex_lock(&nlk->pg_vec_lock);
-
-	expected = 0;
-	for (ring = &nlk->rx_ring; ring <= &nlk->tx_ring; ring++) {
-		if (ring->pg_vec == NULL)
-			continue;
-		expected += ring->pg_vec_len * ring->pg_vec_pages * PAGE_SIZE;
-	}
-
-	if (expected == 0)
-		goto out;
-
-	size = vma->vm_end - vma->vm_start;
-	if (size != expected)
-		goto out;
-
-	start = vma->vm_start;
-	for (ring = &nlk->rx_ring; ring <= &nlk->tx_ring; ring++) {
-		if (ring->pg_vec == NULL)
-			continue;
-
-		for (i = 0; i < ring->pg_vec_len; i++) {
-			struct page *page;
-			void *kaddr = ring->pg_vec[i];
-			unsigned int pg_num;
-
-			for (pg_num = 0; pg_num < ring->pg_vec_pages; pg_num++) {
-				page = pgvec_to_page(kaddr);
-				err = vm_insert_page(vma, start, page);
-				if (err < 0)
-					goto out;
-				start += PAGE_SIZE;
-				kaddr += PAGE_SIZE;
-			}
-		}
-	}
-
-	atomic_inc(&nlk->mapped);
-	vma->vm_ops = &netlink_mmap_ops;
-	err = 0;
-out:
-	mutex_unlock(&nlk->pg_vec_lock);
-	return err;
-}
-
-static void netlink_frame_flush_dcache(const struct nl_mmap_hdr *hdr, unsigned int nm_len)
-{
-#if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE == 1
-	struct page *p_start, *p_end;
-
-	/* First page is flushed through netlink_{get,set}_status */
-	p_start = pgvec_to_page(hdr + PAGE_SIZE);
-	p_end   = pgvec_to_page((void *)hdr + NL_MMAP_HDRLEN + nm_len - 1);
-	while (p_start <= p_end) {
-		flush_dcache_page(p_start);
-		p_start++;
-	}
-#endif
-}
-
-static enum nl_mmap_status netlink_get_status(const struct nl_mmap_hdr *hdr)
-{
-	smp_rmb();
-	flush_dcache_page(pgvec_to_page(hdr));
-	return hdr->nm_status;
-}
-
-static void netlink_set_status(struct nl_mmap_hdr *hdr,
-			       enum nl_mmap_status status)
-{
-	smp_mb();
-	hdr->nm_status = status;
-	flush_dcache_page(pgvec_to_page(hdr));
-}
-
-static struct nl_mmap_hdr *
-__netlink_lookup_frame(const struct netlink_ring *ring, unsigned int pos)
-{
-	unsigned int pg_vec_pos, frame_off;
-
-	pg_vec_pos = pos / ring->frames_per_block;
-	frame_off  = pos % ring->frames_per_block;
-
-	return ring->pg_vec[pg_vec_pos] + (frame_off * ring->frame_size);
-}
-
-static struct nl_mmap_hdr *
-netlink_lookup_frame(const struct netlink_ring *ring, unsigned int pos,
-		     enum nl_mmap_status status)
-{
-	struct nl_mmap_hdr *hdr;
-
-	hdr = __netlink_lookup_frame(ring, pos);
-	if (netlink_get_status(hdr) != status)
-		return NULL;
-
-	return hdr;
-}
-
-static struct nl_mmap_hdr *
-netlink_current_frame(const struct netlink_ring *ring,
-		      enum nl_mmap_status status)
-{
-	return netlink_lookup_frame(ring, ring->head, status);
-}
-
-static void netlink_increment_head(struct netlink_ring *ring)
-{
-	ring->head = ring->head != ring->frame_max ? ring->head + 1 : 0;
-}
-
-static void netlink_forward_ring(struct netlink_ring *ring)
-{
-	unsigned int head = ring->head;
-	const struct nl_mmap_hdr *hdr;
-
-	do {
-		hdr = __netlink_lookup_frame(ring, ring->head);
-		if (hdr->nm_status == NL_MMAP_STATUS_UNUSED)
-			break;
-		if (hdr->nm_status != NL_MMAP_STATUS_SKIP)
-			break;
-		netlink_increment_head(ring);
-	} while (ring->head != head);
-}
-
-static bool netlink_has_valid_frame(struct netlink_ring *ring)
-{
-	unsigned int head = ring->head, pos = head;
-	const struct nl_mmap_hdr *hdr;
-
-	do {
-		hdr = __netlink_lookup_frame(ring, pos);
-		if (hdr->nm_status == NL_MMAP_STATUS_VALID)
-			return true;
-		pos = pos != 0 ? pos - 1 : ring->frame_max;
-	} while (pos != head);
-
-	return false;
-}
-
-static bool netlink_dump_space(struct netlink_sock *nlk)
-{
-	struct netlink_ring *ring = &nlk->rx_ring;
-	struct nl_mmap_hdr *hdr;
-	unsigned int n;
-
-	hdr = netlink_current_frame(ring, NL_MMAP_STATUS_UNUSED);
-	if (hdr == NULL)
-		return false;
-
-	n = ring->head + ring->frame_max / 2;
-	if (n > ring->frame_max)
-		n -= ring->frame_max;
-
-	hdr = __netlink_lookup_frame(ring, n);
-
-	return hdr->nm_status == NL_MMAP_STATUS_UNUSED;
-}
-
-static unsigned int netlink_poll(struct file *file, struct socket *sock,
-				 poll_table *wait)
-{
-	struct sock *sk = sock->sk;
-	struct netlink_sock *nlk = nlk_sk(sk);
-	unsigned int mask;
-	int err;
-
-	if (nlk->rx_ring.pg_vec != NULL) {
-		/* Memory mapped sockets don't call recvmsg(), so flow control
-		 * for dumps is performed here. A dump is allowed to continue
-		 * if at least half the ring is unused.
-		 */
-		while (nlk->cb_running && netlink_dump_space(nlk)) {
-			err = netlink_dump(sk);
-			if (err < 0) {
-				sk->sk_err = -err;
-				sk->sk_error_report(sk);
-				break;
-			}
-		}
-		netlink_rcv_wake(sk);
-	}
-
-	mask = datagram_poll(file, sock, wait);
-
-	/* We could already have received frames in the normal receive
-	 * queue, that will show up as NL_MMAP_STATUS_COPY in the ring,
-	 * so if mask contains pollin/etc already, there's no point
-	 * walking the ring.
-	 */
-	if ((mask & (POLLIN | POLLRDNORM)) != (POLLIN | POLLRDNORM)) {
-		spin_lock_bh(&sk->sk_receive_queue.lock);
-		if (nlk->rx_ring.pg_vec) {
-			if (netlink_has_valid_frame(&nlk->rx_ring))
-				mask |= POLLIN | POLLRDNORM;
-		}
-		spin_unlock_bh(&sk->sk_receive_queue.lock);
-	}
-
-	spin_lock_bh(&sk->sk_write_queue.lock);
-	if (nlk->tx_ring.pg_vec) {
-		if (netlink_current_frame(&nlk->tx_ring, NL_MMAP_STATUS_UNUSED))
-			mask |= POLLOUT | POLLWRNORM;
-	}
-	spin_unlock_bh(&sk->sk_write_queue.lock);
-
-	return mask;
-}
-
-static struct nl_mmap_hdr *netlink_mmap_hdr(struct sk_buff *skb)
-{
-	return (struct nl_mmap_hdr *)(skb->head - NL_MMAP_HDRLEN);
-}
-
-static void netlink_ring_setup_skb(struct sk_buff *skb, struct sock *sk,
-				   struct netlink_ring *ring,
-				   struct nl_mmap_hdr *hdr)
-{
-	unsigned int size;
-	void *data;
-
-	size = ring->frame_size - NL_MMAP_HDRLEN;
-	data = (void *)hdr + NL_MMAP_HDRLEN;
-
-	skb->head	= data;
-	skb->data	= data;
-	skb_reset_tail_pointer(skb);
-	skb->end	= skb->tail + size;
-	skb->len	= 0;
-
-	skb->destructor	= netlink_skb_destructor;
-	NETLINK_CB(skb).flags |= NETLINK_SKB_MMAPED;
-	NETLINK_CB(skb).sk = sk;
-}
-
-static int netlink_mmap_sendmsg(struct sock *sk, struct msghdr *msg,
-				u32 dst_portid, u32 dst_group,
-				struct scm_cookie *scm)
-{
-	struct netlink_sock *nlk = nlk_sk(sk);
-	struct netlink_ring *ring;
-	struct nl_mmap_hdr *hdr;
-	struct sk_buff *skb;
-	unsigned int maxlen;
-	int err = 0, len = 0;
-
-	mutex_lock(&nlk->pg_vec_lock);
-
-	ring   = &nlk->tx_ring;
-	maxlen = ring->frame_size - NL_MMAP_HDRLEN;
-
-	do {
-		unsigned int nm_len;
-
-		hdr = netlink_current_frame(ring, NL_MMAP_STATUS_VALID);
-		if (hdr == NULL) {
-			if (!(msg->msg_flags & MSG_DONTWAIT) &&
-			    atomic_read(&nlk->tx_ring.pending))
-				schedule();
-			continue;
-		}
-
-		nm_len = ACCESS_ONCE(hdr->nm_len);
-		if (nm_len > maxlen) {
-			err = -EINVAL;
-			goto out;
-		}
-
-		netlink_frame_flush_dcache(hdr, nm_len);
-
-		skb = alloc_skb(nm_len, GFP_KERNEL);
-		if (skb == NULL) {
-			err = -ENOBUFS;
-			goto out;
-		}
-		__skb_put(skb, nm_len);
-		memcpy(skb->data, (void *)hdr + NL_MMAP_HDRLEN, nm_len);
-		netlink_set_status(hdr, NL_MMAP_STATUS_UNUSED);
-
-		netlink_increment_head(ring);
-
-		NETLINK_CB(skb).portid	  = nlk->portid;
-		NETLINK_CB(skb).dst_group = dst_group;
-		NETLINK_CB(skb).creds	  = scm->creds;
-
-		err = security_netlink_send(sk, skb);
-		if (err) {
-			kfree_skb(skb);
-			goto out;
-		}
-
-		if (unlikely(dst_group)) {
-			atomic_inc(&skb->users);
-			netlink_broadcast(sk, skb, dst_portid, dst_group,
-					  GFP_KERNEL);
-		}
-		err = netlink_unicast(sk, skb, dst_portid,
-				      msg->msg_flags & MSG_DONTWAIT);
-		if (err < 0)
-			goto out;
-		len += err;
-
-	} while (hdr != NULL ||
-		 (!(msg->msg_flags & MSG_DONTWAIT) &&
-		  atomic_read(&nlk->tx_ring.pending)));
-
-	if (len > 0)
-		err = len;
-out:
-	mutex_unlock(&nlk->pg_vec_lock);
-	return err;
-}
-
-static void netlink_queue_mmaped_skb(struct sock *sk, struct sk_buff *skb)
-{
-	struct nl_mmap_hdr *hdr;
-
-	hdr = netlink_mmap_hdr(skb);
-	hdr->nm_len	= skb->len;
-	hdr->nm_group	= NETLINK_CB(skb).dst_group;
-	hdr->nm_pid	= NETLINK_CB(skb).creds.pid;
-	hdr->nm_uid	= from_kuid(sk_user_ns(sk), NETLINK_CB(skb).creds.uid);
-	hdr->nm_gid	= from_kgid(sk_user_ns(sk), NETLINK_CB(skb).creds.gid);
-	netlink_frame_flush_dcache(hdr, hdr->nm_len);
-	netlink_set_status(hdr, NL_MMAP_STATUS_VALID);
-
-	NETLINK_CB(skb).flags |= NETLINK_SKB_DELIVERED;
-	kfree_skb(skb);
-}
-
-static void netlink_ring_set_copied(struct sock *sk, struct sk_buff *skb)
-{
-	struct netlink_sock *nlk = nlk_sk(sk);
-	struct netlink_ring *ring = &nlk->rx_ring;
-	struct nl_mmap_hdr *hdr;
-
-	spin_lock_bh(&sk->sk_receive_queue.lock);
-	hdr = netlink_current_frame(ring, NL_MMAP_STATUS_UNUSED);
-	if (hdr == NULL) {
-		spin_unlock_bh(&sk->sk_receive_queue.lock);
-		kfree_skb(skb);
-		netlink_overrun(sk);
-		return;
-	}
-	netlink_increment_head(ring);
-	__skb_queue_tail(&sk->sk_receive_queue, skb);
-	spin_unlock_bh(&sk->sk_receive_queue.lock);
-
-	hdr->nm_len	= skb->len;
-	hdr->nm_group	= NETLINK_CB(skb).dst_group;
-	hdr->nm_pid	= NETLINK_CB(skb).creds.pid;
-	hdr->nm_uid	= from_kuid(sk_user_ns(sk), NETLINK_CB(skb).creds.uid);
-	hdr->nm_gid	= from_kgid(sk_user_ns(sk), NETLINK_CB(skb).creds.gid);
-	netlink_set_status(hdr, NL_MMAP_STATUS_COPY);
-}
-
-#else /* CONFIG_NETLINK_MMAP */
-#define netlink_rx_is_mmaped(sk)	false
-#define netlink_tx_is_mmaped(sk)	false
-#define netlink_mmap			sock_no_mmap
-#define netlink_poll			datagram_poll
-#define netlink_mmap_sendmsg(sk, msg, dst_portid, dst_group, scm)	0
-#endif /* CONFIG_NETLINK_MMAP */
-
 static void netlink_skb_destructor(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETLINK_MMAP
-	struct nl_mmap_hdr *hdr;
-	struct netlink_ring *ring;
-	struct sock *sk;
-
-	/* If a packet from the kernel to userspace was freed because of an
-	 * error without being delivered to userspace, the kernel must reset
-	 * the status. In the direction userspace to kernel, the status is
-	 * always reset here after the packet was processed and freed.
-	 */
-	if (netlink_skb_is_mmaped(skb)) {
-		hdr = netlink_mmap_hdr(skb);
-		sk = NETLINK_CB(skb).sk;
-
-		if (NETLINK_CB(skb).flags & NETLINK_SKB_TX) {
-			netlink_set_status(hdr, NL_MMAP_STATUS_UNUSED);
-			ring = &nlk_sk(sk)->tx_ring;
-		} else {
-			if (!(NETLINK_CB(skb).flags & NETLINK_SKB_DELIVERED)) {
-				hdr->nm_len = 0;
-				netlink_set_status(hdr, NL_MMAP_STATUS_VALID);
-			}
-			ring = &nlk_sk(sk)->rx_ring;
-		}
-
-		WARN_ON(atomic_read(&ring->pending) == 0);
-		atomic_dec(&ring->pending);
-		sock_put(sk);
-
-		skb->head = NULL;
-	}
-#endif
 	if (is_vmalloc_addr(skb->head)) {
 		if (!skb->cloned ||
 		    !atomic_dec_return(&(skb_shinfo(skb)->dataref)))
@@ -937,18 +335,6 @@ static void netlink_sock_destruct(struct sock *sk)
 	}
 
 	skb_queue_purge(&sk->sk_receive_queue);
-#ifdef CONFIG_NETLINK_MMAP
-	if (1) {
-		struct nl_mmap_req req;
-
-		memset(&req, 0, sizeof(req));
-		if (nlk->rx_ring.pg_vec)
-			__netlink_set_ring(sk, &req, false, NULL, 0);
-		memset(&req, 0, sizeof(req));
-		if (nlk->tx_ring.pg_vec)
-			__netlink_set_ring(sk, &req, true, NULL, 0);
-	}
-#endif /* CONFIG_NETLINK_MMAP */
 
 	if (!sock_flag(sk, SOCK_DEAD)) {
 		printk(KERN_ERR "Freeing alive netlink socket %p\n", sk);
@@ -1194,9 +580,6 @@ static int __netlink_create(struct net *net, struct socket *sock,
 		mutex_init(nlk->cb_mutex);
 	}
 	init_waitqueue_head(&nlk->wait);
-#ifdef CONFIG_NETLINK_MMAP
-	mutex_init(&nlk->pg_vec_lock);
-#endif
 
 	sk->sk_destruct = netlink_sock_destruct;
 	sk->sk_protocol = protocol;
@@ -1728,8 +1111,7 @@ int netlink_attachskb(struct sock *sk, struct sk_buff *skb,
 	nlk = nlk_sk(sk);
 
 	if ((atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf ||
-	     test_bit(NETLINK_S_CONGESTED, &nlk->state)) &&
-	    !netlink_skb_is_mmaped(skb)) {
+	     test_bit(NETLINK_S_CONGESTED, &nlk->state))) {
 		DECLARE_WAITQUEUE(wait, current);
 		if (!*timeo) {
 			if (!ssk || netlink_is_kernel(ssk))
@@ -1767,14 +1149,7 @@ static int __netlink_sendskb(struct sock *sk, struct sk_buff *skb)
 
 	netlink_deliver_tap(skb);
 
-#ifdef CONFIG_NETLINK_MMAP
-	if (netlink_skb_is_mmaped(skb))
-		netlink_queue_mmaped_skb(sk, skb);
-	else if (netlink_rx_is_mmaped(sk))
-		netlink_ring_set_copied(sk, skb);
-	else
-#endif /* CONFIG_NETLINK_MMAP */
-		skb_queue_tail(&sk->sk_receive_queue, skb);
+	skb_queue_tail(&sk->sk_receive_queue, skb);
 	sk->sk_data_ready(sk);
 	return len;
 }
@@ -1798,9 +1173,6 @@ static struct sk_buff *netlink_trim(struct sk_buff *skb, gfp_t allocation)
 	int delta;
 
 	WARN_ON(skb->sk != NULL);
-	if (netlink_skb_is_mmaped(skb))
-		return skb;
-
 	delta = skb->end - skb->tail;
 	if (is_vmalloc_addr(skb->head) || delta * 2 < skb->truesize)
 		return skb;
@@ -1880,71 +1252,6 @@ struct sk_buff *__netlink_alloc_skb(struct sock *ssk, unsigned int size,
 				    unsigned int ldiff, u32 dst_portid,
 				    gfp_t gfp_mask)
 {
-#ifdef CONFIG_NETLINK_MMAP
-	unsigned int maxlen, linear_size;
-	struct sock *sk = NULL;
-	struct sk_buff *skb;
-	struct netlink_ring *ring;
-	struct nl_mmap_hdr *hdr;
-
-	sk = netlink_getsockbyportid(ssk, dst_portid);
-	if (IS_ERR(sk))
-		goto out;
-
-	ring = &nlk_sk(sk)->rx_ring;
-	/* fast-path without atomic ops for common case: non-mmaped receiver */
-	if (ring->pg_vec == NULL)
-		goto out_put;
-
-	/* We need to account the full linear size needed as a ring
-	 * slot cannot have non-linear parts.
-	 */
-	linear_size = size + ldiff;
-	if (ring->frame_size - NL_MMAP_HDRLEN < linear_size)
-		goto out_put;
-
-	skb = alloc_skb_head(gfp_mask);
-	if (skb == NULL)
-		goto err1;
-
-	spin_lock_bh(&sk->sk_receive_queue.lock);
-	/* check again under lock */
-	if (ring->pg_vec == NULL)
-		goto out_free;
-
-	/* check again under lock */
-	maxlen = ring->frame_size - NL_MMAP_HDRLEN;
-	if (maxlen < linear_size)
-		goto out_free;
-
-	netlink_forward_ring(ring);
-	hdr = netlink_current_frame(ring, NL_MMAP_STATUS_UNUSED);
-	if (hdr == NULL)
-		goto err2;
-
-	netlink_ring_setup_skb(skb, sk, ring, hdr);
-	netlink_set_status(hdr, NL_MMAP_STATUS_RESERVED);
-	atomic_inc(&ring->pending);
-	netlink_increment_head(ring);
-
-	spin_unlock_bh(&sk->sk_receive_queue.lock);
-	return skb;
-
-err2:
-	kfree_skb(skb);
-	spin_unlock_bh(&sk->sk_receive_queue.lock);
-	netlink_overrun(sk);
-err1:
-	sock_put(sk);
-	return NULL;
-
-out_free:
-	kfree_skb(skb);
-	spin_unlock_bh(&sk->sk_receive_queue.lock);
-out_put:
-	sock_put(sk);
-out:
-#endif
 	return alloc_skb(size, gfp_mask);
 }
 EXPORT_SYMBOL_GPL(__netlink_alloc_skb);
@@ -2225,8 +1532,7 @@ static int netlink_setsockopt(struct socket *sock, int level, int optname,
 	if (level != SOL_NETLINK)
 		return -ENOPROTOOPT;
 
-	if (optname != NETLINK_RX_RING && optname != NETLINK_TX_RING &&
-	    optlen >= sizeof(int) &&
+	if (optlen >= sizeof(int) &&
 	    get_user(val, (unsigned int __user *)optval))
 		return -EFAULT;
 
@@ -2279,25 +1585,6 @@ static int netlink_setsockopt(struct socket *sock, int level, int optname,
 		}
 		err = 0;
 		break;
-#ifdef CONFIG_NETLINK_MMAP
-	case NETLINK_RX_RING:
-	case NETLINK_TX_RING: {
-		struct nl_mmap_req req;
-
-		/* Rings might consume more memory than queue limits, require
-		 * CAP_NET_ADMIN.
-		 */
-		if (!capable(CAP_NET_ADMIN))
-			return -EPERM;
-		if (optlen < sizeof(req))
-			return -EINVAL;
-		if (copy_from_user(&req, optval, sizeof(req)))
-			return -EFAULT;
-		err = netlink_set_ring(sk, &req,
-				       optname == NETLINK_TX_RING);
-		break;
-	}
-#endif /* CONFIG_NETLINK_MMAP */
 	case NETLINK_LISTEN_ALL_NSID:
 		if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_BROADCAST))
 			return -EPERM;
@@ -2467,18 +1754,6 @@ static int netlink_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 		smp_rmb();
 	}
 
-	/* It's a really convoluted way for userland to ask for mmaped
-	 * sendmsg(), but that's what we've got...
-	 */
-	if (netlink_tx_is_mmaped(sk) &&
-	    iter_is_iovec(&msg->msg_iter) &&
-	    msg->msg_iter.nr_segs == 1 &&
-	    msg->msg_iter.iov->iov_base == NULL) {
-		err = netlink_mmap_sendmsg(sk, msg, dst_portid, dst_group,
-					   &scm);
-		goto out;
-	}
-
 	err = -EMSGSIZE;
 	if (len > sk->sk_sndbuf - 32)
 		goto out;
@@ -2794,8 +2069,7 @@ static int netlink_dump(struct sock *sk)
 		goto errout_skb;
 	}
 
-	if (!netlink_rx_is_mmaped(sk) &&
-	    atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
+	if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
 		goto errout_skb;
 
 	/* NLMSG_GOODSIZE is small to avoid high order allocations being
@@ -2831,8 +2105,7 @@ static int netlink_dump(struct sock *sk)
 	 * reasonable static buffer based on the expected largest dump of a
 	 * single netdev. The outcome is MSG_TRUNC error.
 	 */
-	if (!netlink_rx_is_mmaped(sk))
-		skb_reserve(skb, skb_tailroom(skb) - alloc_size);
+	skb_reserve(skb, skb_tailroom(skb) - alloc_size);
 	netlink_skb_set_owner_r(skb, sk);
 
 	len = cb->dump(skb, cb);
@@ -2884,16 +2157,7 @@ int __netlink_dump_start(struct sock *ssk, struct sk_buff *skb,
 	struct netlink_sock *nlk;
 	int ret;
 
-	/* Memory mapped dump requests need to be copied to avoid looping
-	 * on the pending state in netlink_mmap_sendmsg() while the CB hold
-	 * a reference to the skb.
-	 */
-	if (netlink_skb_is_mmaped(skb)) {
-		skb = skb_copy(skb, GFP_KERNEL);
-		if (skb == NULL)
-			return -ENOBUFS;
-	} else
-		atomic_inc(&skb->users);
+	atomic_inc(&skb->users);
 
 	sk = netlink_lookup(sock_net(ssk), ssk->sk_protocol, NETLINK_CB(skb).portid);
 	if (sk == NULL) {
@@ -3241,7 +2505,7 @@ static const struct proto_ops netlink_ops = {
 	.socketpair =	sock_no_socketpair,
 	.accept =	sock_no_accept,
 	.getname =	netlink_getname,
-	.poll =		netlink_poll,
+	.poll =		datagram_poll,
 	.ioctl =	sock_no_ioctl,
 	.listen =	sock_no_listen,
 	.shutdown =	sock_no_shutdown,
@@ -3249,7 +2513,7 @@ static const struct proto_ops netlink_ops = {
 	.getsockopt =	netlink_getsockopt,
 	.sendmsg =	netlink_sendmsg,
 	.recvmsg =	netlink_recvmsg,
-	.mmap =		netlink_mmap,
+	.mmap =		sock_no_mmap,
 	.sendpage =	sock_no_sendpage,
 };
 
diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h
index 14437d9..e68ef9c 100644
--- a/net/netlink/af_netlink.h
+++ b/net/netlink/af_netlink.h
@@ -44,12 +44,6 @@ struct netlink_sock {
 	int			(*netlink_bind)(struct net *net, int group);
 	void			(*netlink_unbind)(struct net *net, int group);
 	struct module		*module;
-#ifdef CONFIG_NETLINK_MMAP
-	struct mutex		pg_vec_lock;
-	struct netlink_ring	rx_ring;
-	struct netlink_ring	tx_ring;
-	atomic_t		mapped;
-#endif /* CONFIG_NETLINK_MMAP */
 
 	struct rhash_head	node;
 	struct rcu_head		rcu;
@@ -60,15 +54,6 @@ static inline struct netlink_sock *nlk_sk(struct sock *sk)
 	return container_of(sk, struct netlink_sock, sk);
 }
 
-static inline bool netlink_skb_is_mmaped(const struct sk_buff *skb)
-{
-#ifdef CONFIG_NETLINK_MMAP
-	return NETLINK_CB(skb).flags & NETLINK_SKB_MMAPED;
-#else
-	return false;
-#endif /* CONFIG_NETLINK_MMAP */
-}
-
 struct netlink_table {
 	struct rhashtable	hash;
 	struct hlist_head	mc_list;
diff --git a/net/netlink/diag.c b/net/netlink/diag.c
index 3ee63a3cf..8dd836a 100644
--- a/net/netlink/diag.c
+++ b/net/netlink/diag.c
@@ -8,41 +8,6 @@
 
 #include "af_netlink.h"
 
-#ifdef CONFIG_NETLINK_MMAP
-static int sk_diag_put_ring(struct netlink_ring *ring, int nl_type,
-			    struct sk_buff *nlskb)
-{
-	struct netlink_diag_ring ndr;
-
-	ndr.ndr_block_size = ring->pg_vec_pages << PAGE_SHIFT;
-	ndr.ndr_block_nr   = ring->pg_vec_len;
-	ndr.ndr_frame_size = ring->frame_size;
-	ndr.ndr_frame_nr   = ring->frame_max + 1;
-
-	return nla_put(nlskb, nl_type, sizeof(ndr), &ndr);
-}
-
-static int sk_diag_put_rings_cfg(struct sock *sk, struct sk_buff *nlskb)
-{
-	struct netlink_sock *nlk = nlk_sk(sk);
-	int ret;
-
-	mutex_lock(&nlk->pg_vec_lock);
-	ret = sk_diag_put_ring(&nlk->rx_ring, NETLINK_DIAG_RX_RING, nlskb);
-	if (!ret)
-		ret = sk_diag_put_ring(&nlk->tx_ring, NETLINK_DIAG_TX_RING,
-				       nlskb);
-	mutex_unlock(&nlk->pg_vec_lock);
-
-	return ret;
-}
-#else
-static int sk_diag_put_rings_cfg(struct sock *sk, struct sk_buff *nlskb)
-{
-	return 0;
-}
-#endif
-
 static int sk_diag_dump_groups(struct sock *sk, struct sk_buff *nlskb)
 {
 	struct netlink_sock *nlk = nlk_sk(sk);
@@ -87,10 +52,6 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb,
 	    sock_diag_put_meminfo(sk, skb, NETLINK_DIAG_MEMINFO))
 		goto out_nlmsg_trim;
 
-	if ((req->ndiag_show & NDIAG_SHOW_RING_CFG) &&
-	    sk_diag_put_rings_cfg(sk, skb))
-		goto out_nlmsg_trim;
-
 	nlmsg_end(skb, nlh);
 	return 0;
 
-- 
cgit v0.10.2


From 551ddc057e290cc07ff900050da242f35d615d3c Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 18 Feb 2016 15:03:25 +0100
Subject: openvswitch: Revert: "Enable memory mapped Netlink i/o"

revert commit 795449d8b846 ("openvswitch: Enable memory mapped Netlink i/o").
Following the mmaped netlink removal this code can be removed.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index d6f7fe9..35a2659 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -422,10 +422,6 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb,
 	struct sk_buff *nskb = NULL;
 	struct sk_buff *user_skb = NULL; /* to be queued to userspace */
 	struct nlattr *nla;
-	struct genl_info info = {
-		.dst_sk = ovs_dp_get_net(dp)->genl_sock,
-		.snd_portid = upcall_info->portid,
-	};
 	size_t len;
 	unsigned int hlen;
 	int err, dp_ifindex;
@@ -466,7 +462,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb,
 		hlen = skb->len;
 
 	len = upcall_msg_size(upcall_info, hlen);
-	user_skb = genlmsg_new_unicast(len, &info, GFP_ATOMIC);
+	user_skb = genlmsg_new(len, GFP_ATOMIC);
 	if (!user_skb) {
 		err = -ENOMEM;
 		goto out;
@@ -876,7 +872,7 @@ static struct sk_buff *ovs_flow_cmd_alloc_info(const struct sw_flow_actions *act
 		return NULL;
 
 	len = ovs_flow_cmd_msg_size(acts, sfid, ufid_flags);
-	skb = genlmsg_new_unicast(len, info, GFP_KERNEL);
+	skb = genlmsg_new(len, GFP_KERNEL);
 	if (!skb)
 		return ERR_PTR(-ENOMEM);
 
@@ -1483,7 +1479,7 @@ error:
 
 static struct sk_buff *ovs_dp_cmd_alloc_info(struct genl_info *info)
 {
-	return genlmsg_new_unicast(ovs_dp_cmd_msg_size(), info, GFP_KERNEL);
+	return genlmsg_new(ovs_dp_cmd_msg_size(), GFP_KERNEL);
 }
 
 /* Called with rcu_read_lock or ovs_mutex. */
-- 
cgit v0.10.2


From 263ea09084d172cac6e40459a690babe8de8e448 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 18 Feb 2016 15:03:26 +0100
Subject: Revert "genl: Add genlmsg_new_unicast() for unicast message
 allocation"

This reverts commit bb9b18fb55b0 ("genl: Add genlmsg_new_unicast() for
unicast message allocation")'.

Nothing wrong with it; its no longer needed since this was only for
mmapped netlink support.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/genetlink.h b/include/net/genetlink.h
index 43c0e77..8d4608c 100644
--- a/include/net/genetlink.h
+++ b/include/net/genetlink.h
@@ -83,7 +83,6 @@ struct genl_family {
  * @attrs: netlink attributes
  * @_net: network namespace
  * @user_ptr: user pointers
- * @dst_sk: destination socket
  */
 struct genl_info {
 	u32			snd_seq;
@@ -94,7 +93,6 @@ struct genl_info {
 	struct nlattr **	attrs;
 	possible_net_t		_net;
 	void *			user_ptr[2];
-	struct sock *		dst_sk;
 };
 
 static inline struct net *genl_info_net(struct genl_info *info)
@@ -188,8 +186,6 @@ int genl_unregister_family(struct genl_family *family);
 void genl_notify(struct genl_family *family, struct sk_buff *skb,
 		 struct genl_info *info, u32 group, gfp_t flags);
 
-struct sk_buff *genlmsg_new_unicast(size_t payload, struct genl_info *info,
-				    gfp_t flags);
 void *genlmsg_put(struct sk_buff *skb, u32 portid, u32 seq,
 		  struct genl_family *family, int flags, u8 cmd);
 
diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index 0ffd721..a09132a 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -463,26 +463,6 @@ int genl_unregister_family(struct genl_family *family)
 EXPORT_SYMBOL(genl_unregister_family);
 
 /**
- * genlmsg_new_unicast - Allocate generic netlink message for unicast
- * @payload: size of the message payload
- * @info: information on destination
- * @flags: the type of memory to allocate
- *
- * Allocates a new sk_buff large enough to cover the specified payload
- * plus required Netlink headers. Will check receiving socket for
- * memory mapped i/o capability and use it if enabled. Will fall back
- * to non-mapped skb if message size exceeds the frame size of the ring.
- */
-struct sk_buff *genlmsg_new_unicast(size_t payload, struct genl_info *info,
-				    gfp_t flags)
-{
-	size_t len = nlmsg_total_size(genlmsg_total_size(payload));
-
-	return netlink_alloc_skb(info->dst_sk, len, info->snd_portid, flags);
-}
-EXPORT_SYMBOL_GPL(genlmsg_new_unicast);
-
-/**
  * genlmsg_put - Add generic netlink header to netlink message
  * @skb: socket buffer holding the message
  * @portid: netlink portid the message is addressed to
@@ -642,7 +622,6 @@ static int genl_family_rcv_msg(struct genl_family *family,
 	info.genlhdr = nlmsg_data(nlh);
 	info.userhdr = nlmsg_data(nlh) + GENL_HDRLEN;
 	info.attrs = attrbuf;
-	info.dst_sk = skb->sk;
 	genl_info_net_set(&info, net);
 	memset(&info.user_ptr, 0, sizeof(info.user_ptr));
 
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 35a2659..c4e8455 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1477,7 +1477,7 @@ error:
 	return -EMSGSIZE;
 }
 
-static struct sk_buff *ovs_dp_cmd_alloc_info(struct genl_info *info)
+static struct sk_buff *ovs_dp_cmd_alloc_info(void)
 {
 	return genlmsg_new(ovs_dp_cmd_msg_size(), GFP_KERNEL);
 }
@@ -1532,7 +1532,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 	if (!a[OVS_DP_ATTR_NAME] || !a[OVS_DP_ATTR_UPCALL_PID])
 		goto err;
 
-	reply = ovs_dp_cmd_alloc_info(info);
+	reply = ovs_dp_cmd_alloc_info();
 	if (!reply)
 		return -ENOMEM;
 
@@ -1653,7 +1653,7 @@ static int ovs_dp_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	struct datapath *dp;
 	int err;
 
-	reply = ovs_dp_cmd_alloc_info(info);
+	reply = ovs_dp_cmd_alloc_info();
 	if (!reply)
 		return -ENOMEM;
 
@@ -1686,7 +1686,7 @@ static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info)
 	struct datapath *dp;
 	int err;
 
-	reply = ovs_dp_cmd_alloc_info(info);
+	reply = ovs_dp_cmd_alloc_info();
 	if (!reply)
 		return -ENOMEM;
 
@@ -1719,7 +1719,7 @@ static int ovs_dp_cmd_get(struct sk_buff *skb, struct genl_info *info)
 	struct datapath *dp;
 	int err;
 
-	reply = ovs_dp_cmd_alloc_info(info);
+	reply = ovs_dp_cmd_alloc_info();
 	if (!reply)
 		return -ENOMEM;
 
diff --git a/net/tipc/netlink_compat.c b/net/tipc/netlink_compat.c
index 2c016fd..de66d8f 100644
--- a/net/tipc/netlink_compat.c
+++ b/net/tipc/netlink_compat.c
@@ -1104,7 +1104,6 @@ static int tipc_nl_compat_recv(struct sk_buff *skb, struct genl_info *info)
 	req_nlh = (struct nlmsghdr *)skb->data;
 	msg.req = nlmsg_data(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN;
 	msg.cmd = req_userhdr->cmd;
-	msg.dst_sk = info->dst_sk;
 	msg.net = genl_info_net(info);
 
 	if ((msg.cmd & 0xC000) && (!netlink_net_capable(skb, CAP_NET_ADMIN))) {
-- 
cgit v0.10.2


From 905f0a739ad82c6371fb0cb0e71db14a750702ad Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 18 Feb 2016 15:03:27 +0100
Subject: nfnetlink: remove nfnetlink_alloc_skb

Following mmapped netlink removal this code can be simplified by
removing the alloc wrapper.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h
index ba0d978..1d82dd5 100644
--- a/include/linux/netfilter/nfnetlink.h
+++ b/include/linux/netfilter/nfnetlink.h
@@ -34,8 +34,6 @@ int nfnetlink_subsys_register(const struct nfnetlink_subsystem *n);
 int nfnetlink_subsys_unregister(const struct nfnetlink_subsystem *n);
 
 int nfnetlink_has_listeners(struct net *net, unsigned int group);
-struct sk_buff *nfnetlink_alloc_skb(struct net *net, unsigned int size,
-				    u32 dst_portid, gfp_t gfp_mask);
 int nfnetlink_send(struct sk_buff *skb, struct net *net, u32 portid,
 		   unsigned int group, int echo, gfp_t flags);
 int nfnetlink_set_err(struct net *net, u32 portid, u32 group, int error);
diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c
index a7ba233..9a99f68 100644
--- a/net/netfilter/nfnetlink.c
+++ b/net/netfilter/nfnetlink.c
@@ -127,13 +127,6 @@ int nfnetlink_has_listeners(struct net *net, unsigned int group)
 }
 EXPORT_SYMBOL_GPL(nfnetlink_has_listeners);
 
-struct sk_buff *nfnetlink_alloc_skb(struct net *net, unsigned int size,
-				    u32 dst_portid, gfp_t gfp_mask)
-{
-	return netlink_alloc_skb(net->nfnl, size, dst_portid, gfp_mask);
-}
-EXPORT_SYMBOL_GPL(nfnetlink_alloc_skb);
-
 int nfnetlink_send(struct sk_buff *skb, struct net *net, u32 portid,
 		   unsigned int group, int echo, gfp_t flags)
 {
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index 8ca9320..11f81c8 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -330,14 +330,13 @@ nfulnl_alloc_skb(struct net *net, u32 peer_portid, unsigned int inst_size,
 	 * message.  WARNING: has to be <= 128k due to slab restrictions */
 
 	n = max(inst_size, pkt_size);
-	skb = nfnetlink_alloc_skb(net, n, peer_portid, GFP_ATOMIC);
+	skb = alloc_skb(n, GFP_ATOMIC);
 	if (!skb) {
 		if (n > pkt_size) {
 			/* try to allocate only as much as we need for current
 			 * packet */
 
-			skb = nfnetlink_alloc_skb(net, pkt_size,
-						  peer_portid, GFP_ATOMIC);
+			skb = alloc_skb(pkt_size, GFP_ATOMIC);
 		}
 	}
 
-- 
cgit v0.10.2


From c5b0db3263b92526bc0c1b6380c0c99f91f069fc Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 18 Feb 2016 15:03:28 +0100
Subject: nfnetlink: Revert "nfnetlink: add support for memory mapped netlink"

reverts commit 3ab1f683bf8b ("nfnetlink: add support for memory mapped
netlink")'

Like previous commits in the series, remove wrappers that are not needed
after mmapped netlink removal.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 0b41959..da14ab6 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -69,16 +69,6 @@ extern void __netlink_clear_multicast_users(struct sock *sk, unsigned int group)
 extern void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err);
 extern int netlink_has_listeners(struct sock *sk, unsigned int group);
 
-extern struct sk_buff *__netlink_alloc_skb(struct sock *ssk, unsigned int size,
-					   unsigned int ldiff, u32 dst_portid,
-					   gfp_t gfp_mask);
-static inline struct sk_buff *
-netlink_alloc_skb(struct sock *ssk, unsigned int size, u32 dst_portid,
-		  gfp_t gfp_mask)
-{
-	return __netlink_alloc_skb(ssk, size, 0, dst_portid, gfp_mask);
-}
-
 extern int netlink_unicast(struct sock *ssk, struct sk_buff *skb, __u32 portid, int nonblock);
 extern int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, __u32 portid,
 			     __u32 group, gfp_t allocation);
diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c
index 1d39365..7542999 100644
--- a/net/netfilter/nfnetlink_queue.c
+++ b/net/netfilter/nfnetlink_queue.c
@@ -301,7 +301,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 			   __be32 **packet_id_ptr)
 {
 	size_t size;
-	size_t data_len = 0, cap_len = 0, rem_len = 0;
+	size_t data_len = 0, cap_len = 0;
 	unsigned int hlen = 0;
 	struct sk_buff *skb;
 	struct nlattr *nla;
@@ -361,7 +361,6 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 		hlen = min_t(unsigned int, hlen, data_len);
 		size += sizeof(struct nlattr) + hlen;
 		cap_len = entskb->len;
-		rem_len = data_len - hlen;
 		break;
 	}
 
@@ -386,8 +385,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 			size += nla_total_size(seclen);
 	}
 
-	skb = __netlink_alloc_skb(net->nfnl, size, rem_len, queue->peer_portid,
-				  GFP_ATOMIC);
+	skb = alloc_skb(size, GFP_ATOMIC);
 	if (!skb) {
 		skb_tx_error(entskb);
 		return NULL;
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 85aa6ef..c841679 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -1248,14 +1248,6 @@ retry:
 }
 EXPORT_SYMBOL(netlink_unicast);
 
-struct sk_buff *__netlink_alloc_skb(struct sock *ssk, unsigned int size,
-				    unsigned int ldiff, u32 dst_portid,
-				    gfp_t gfp_mask)
-{
-	return alloc_skb(size, gfp_mask);
-}
-EXPORT_SYMBOL_GPL(__netlink_alloc_skb);
-
 int netlink_has_listeners(struct sock *sk, unsigned int group)
 {
 	int res = 0;
@@ -2082,15 +2074,12 @@ static int netlink_dump(struct sock *sk)
 
 	if (alloc_min_size < nlk->max_recvmsg_len) {
 		alloc_size = nlk->max_recvmsg_len;
-		skb = netlink_alloc_skb(sk, alloc_size, nlk->portid,
-					GFP_KERNEL |
-					__GFP_NOWARN |
-					__GFP_NORETRY);
+		skb = alloc_skb(alloc_size, GFP_KERNEL |
+					    __GFP_NOWARN | __GFP_NORETRY);
 	}
 	if (!skb) {
 		alloc_size = alloc_min_size;
-		skb = netlink_alloc_skb(sk, alloc_size, nlk->portid,
-					GFP_KERNEL);
+		skb = alloc_skb(alloc_size, GFP_KERNEL);
 	}
 	if (!skb)
 		goto errout_skb;
@@ -2230,8 +2219,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err)
 	if (!(nlk->flags & NETLINK_F_CAP_ACK) && err)
 		payload += nlmsg_len(nlh);
 
-	skb = netlink_alloc_skb(in_skb->sk, nlmsg_total_size(payload),
-				NETLINK_CB(in_skb).portid, GFP_KERNEL);
+	skb = nlmsg_new(payload, GFP_KERNEL);
 	if (!skb) {
 		struct sock *sk;
 
-- 
cgit v0.10.2


From a9c9a81f5892eb984234223399ee624f7dbd15e8 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:13 -0800
Subject: i40e/i40evf: Drop outer checksum offload that was not requested

The i40e and i40evf drivers contained code for inserting an outer checksum
on UDP tunnels.  The issue however is that the upper levels of the stack
never requested such an offload and it results in possible errors.

In addition the same logic was being applied to the Rx side where it was
attempting to validate the outer checksum, but the logic there was
incorrect in that it was testing for the resultant sum to be equal to the
header checksum instead of being equal to 0.

Since this code is so massively flawed, and doing things that we didn't ask
for it to do I am just dropping it, and will bring it back later to use as
an offload for SKB_GSO_UDP_TUNNEL_CSUM which can make use of such a
feature.

As far as the Rx feature I am dropping it completely since it would need to
be massively expanded and applied to IPv4 and IPv6 checksums for all parts,
not just the one that supports Tx checksum offload for the outer.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 16e5e0b..0fa52ed 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -7474,8 +7474,6 @@ static int i40e_alloc_rings(struct i40e_vsi *vsi)
 		tx_ring->dcb_tc = 0;
 		if (vsi->back->flags & I40E_FLAG_WB_ON_ITR_CAPABLE)
 			tx_ring->flags = I40E_TXR_FLAGS_WB_ON_ITR;
-		if (vsi->back->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE)
-			tx_ring->flags |= I40E_TXR_FLAGS_OUTER_UDP_CSUM;
 		vsi->tx_rings[i] = tx_ring;
 
 		rx_ring = &tx_ring[1];
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 65f2fd8..d4364ec 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1391,9 +1391,6 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
 	bool ipv4 = false, ipv6 = false;
 	bool ipv4_tunnel, ipv6_tunnel;
-	__wsum rx_udp_csum;
-	struct iphdr *iph;
-	__sum16 csum;
 
 	ipv4_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
 		     (rx_ptype <= I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
@@ -1443,37 +1440,12 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	if (rx_error & BIT(I40E_RX_DESC_ERROR_PPRS_SHIFT))
 		return;
 
-	/* If VXLAN/GENEVE traffic has an outer UDPv4 checksum we need to check
-	 * it in the driver, hardware does not do it for us.
-	 * Since L3L4P bit was set we assume a valid IHL value (>=5)
-	 * so the total length of IPv4 header is IHL*4 bytes
-	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
+	/* The hardware supported by this driver does not validate outer
+	 * checksums for tunneled VXLAN or GENEVE frames.  I don't agree
+	 * with it but the specification states that you "MAY validate", it
+	 * doesn't make it a hard requirement so if we have validated the
+	 * inner checksum report CHECKSUM_UNNECESSARY.
 	 */
-	if (!(vsi->back->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE) &&
-	    (ipv4_tunnel)) {
-		skb->transport_header = skb->mac_header +
-					sizeof(struct ethhdr) +
-					(ip_hdr(skb)->ihl * 4);
-
-		/* Add 4 bytes for VLAN tagged packets */
-		skb->transport_header += (skb->protocol == htons(ETH_P_8021Q) ||
-					  skb->protocol == htons(ETH_P_8021AD))
-					  ? VLAN_HLEN : 0;
-
-		if ((ip_hdr(skb)->protocol == IPPROTO_UDP) &&
-		    (udp_hdr(skb)->check != 0)) {
-			rx_udp_csum = udp_csum(skb);
-			iph = ip_hdr(skb);
-			csum = csum_tcpudp_magic(
-					iph->saddr, iph->daddr,
-					(skb->len - skb_transport_offset(skb)),
-					IPPROTO_UDP, rx_udp_csum);
-
-			if (udp_hdr(skb)->check != csum)
-				goto checksum_fail;
-
-		} /* else its GRE and so no outer UDP header */
-	}
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->csum_level = ipv4_tunnel || ipv6_tunnel;
@@ -2453,15 +2425,6 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*tx_flags &= ~I40E_TX_FLAGS_IPV4;
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
 		}
-		if ((tx_ring->flags & I40E_TXR_FLAGS_OUTER_UDP_CSUM) &&
-		    (l4_tunnel == I40E_TXD_CTX_UDP_TUNNELING)        &&
-		    (*cd_tunneling & I40E_TXD_CTX_QW0_EXT_IP_MASK)) {
-			oudph->check = ~csum_tcpudp_magic(oiph->saddr,
-					oiph->daddr,
-					(skb->len - skb_transport_offset(skb)),
-					IPPROTO_UDP, 0);
-			*cd_tunneling |= I40E_TXD_CTX_QW0_L4T_CS_MASK;
-		}
 	} else {
 		network_hdr_len = skb_network_header_len(skb);
 		this_ip_hdr = ip_hdr(skb);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 3acc924..fb065d4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -277,7 +277,6 @@ struct i40e_ring {
 
 	u16 flags;
 #define I40E_TXR_FLAGS_WB_ON_ITR	BIT(0)
-#define I40E_TXR_FLAGS_OUTER_UDP_CSUM	BIT(1)
 #define I40E_TXR_FLAGS_LAST_XMIT_MORE_SET BIT(2)
 
 	/* stats structs */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index fb6cd7e..8b20ed3 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -863,9 +863,6 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
 	bool ipv4 = false, ipv6 = false;
 	bool ipv4_tunnel, ipv6_tunnel;
-	__wsum rx_udp_csum;
-	struct iphdr *iph;
-	__sum16 csum;
 
 	ipv4_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
 		     (rx_ptype <= I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
@@ -915,36 +912,12 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	if (rx_error & BIT(I40E_RX_DESC_ERROR_PPRS_SHIFT))
 		return;
 
-	/* If VXLAN traffic has an outer UDPv4 checksum we need to check
-	 * it in the driver, hardware does not do it for us.
-	 * Since L3L4P bit was set we assume a valid IHL value (>=5)
-	 * so the total length of IPv4 header is IHL*4 bytes
-	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
+	/* The hardware supported by this driver does not validate outer
+	 * checksums for tunneled VXLAN or GENEVE frames.  I don't agree
+	 * with it but the specification states that you "MAY validate", it
+	 * doesn't make it a hard requirement so if we have validated the
+	 * inner checksum report CHECKSUM_UNNECESSARY.
 	 */
-	if (ipv4_tunnel) {
-		skb->transport_header = skb->mac_header +
-					sizeof(struct ethhdr) +
-					(ip_hdr(skb)->ihl * 4);
-
-		/* Add 4 bytes for VLAN tagged packets */
-		skb->transport_header += (skb->protocol == htons(ETH_P_8021Q) ||
-					  skb->protocol == htons(ETH_P_8021AD))
-					  ? VLAN_HLEN : 0;
-
-		if ((ip_hdr(skb)->protocol == IPPROTO_UDP) &&
-		    (udp_hdr(skb)->check != 0)) {
-			rx_udp_csum = udp_csum(skb);
-			iph = ip_hdr(skb);
-			csum = csum_tcpudp_magic(iph->saddr, iph->daddr,
-						 (skb->len -
-						  skb_transport_offset(skb)),
-						 IPPROTO_UDP, rx_udp_csum);
-
-			if (udp_hdr(skb)->check != csum)
-				goto checksum_fail;
-
-		} /* else its GRE and so no outer UDP header */
-	}
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->csum_level = ipv4_tunnel || ipv6_tunnel;
@@ -1667,15 +1640,6 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
 		}
 
-		if ((tx_ring->flags & I40E_TXR_FLAGS_OUTER_UDP_CSUM) &&
-		    (l4_tunnel == I40E_TXD_CTX_UDP_TUNNELING)        &&
-		    (*cd_tunneling & I40E_TXD_CTX_QW0_EXT_IP_MASK)) {
-			oudph->check = ~csum_tcpudp_magic(oiph->saddr,
-					oiph->daddr,
-					(skb->len - skb_transport_offset(skb)),
-					IPPROTO_UDP, 0);
-			*cd_tunneling |= I40E_TXD_CTX_QW0_L4T_CS_MASK;
-		}
 	} else {
 		network_hdr_len = skb_network_header_len(skb);
 		this_ip_hdr = ip_hdr(skb);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 81c9661..043b955 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -275,7 +275,6 @@ struct i40e_ring {
 
 	u16 flags;
 #define I40E_TXR_FLAGS_WB_ON_ITR	BIT(0)
-#define I40E_TXR_FLAGS_OUTER_UDP_CSUM	BIT(1)
 
 	/* stats structs */
 	struct i40e_queue_stats	stats;
-- 
cgit v0.10.2


From 03f9d6a59f94f70ae775ca2aae04f2accc01a973 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:20 -0800
Subject: i40e/i40evf: Use u64 values instead of casting them in TSO function

Instead of casing u32 values to u64 it makes more sense to just start out
with u64 values in the first place.  This way we don't need to create a
mess with all of the casts needed to populate a 64b value.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index d4364ec..5e82589 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2267,7 +2267,7 @@ out:
 static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		    u8 *hdr_len, u64 *cd_type_cmd_tso_mss)
 {
-	u32 cd_cmd, cd_tso_len, cd_mss;
+	u64 cd_cmd, cd_tso_len, cd_mss;
 	struct ipv6hdr *ipv6h;
 	struct tcphdr *tcph;
 	struct iphdr *iph;
@@ -2309,10 +2309,9 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	cd_cmd = I40E_TX_CTX_DESC_TSO;
 	cd_tso_len = skb->len - *hdr_len;
 	cd_mss = skb_shinfo(skb)->gso_size;
-	*cd_type_cmd_tso_mss |= ((u64)cd_cmd << I40E_TXD_CTX_QW1_CMD_SHIFT) |
-				((u64)cd_tso_len <<
-				 I40E_TXD_CTX_QW1_TSO_LEN_SHIFT) |
-				((u64)cd_mss << I40E_TXD_CTX_QW1_MSS_SHIFT);
+	*cd_type_cmd_tso_mss |= (cd_cmd << I40E_TXD_CTX_QW1_CMD_SHIFT) |
+				(cd_tso_len << I40E_TXD_CTX_QW1_TSO_LEN_SHIFT) |
+				(cd_mss << I40E_TXD_CTX_QW1_MSS_SHIFT);
 	return 1;
 }
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 8b20ed3..c5f98cb 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1527,7 +1527,7 @@ out:
 static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		    u8 *hdr_len, u64 *cd_type_cmd_tso_mss)
 {
-	u32 cd_cmd, cd_tso_len, cd_mss;
+	u64 cd_cmd, cd_tso_len, cd_mss;
 	struct ipv6hdr *ipv6h;
 	struct tcphdr *tcph;
 	struct iphdr *iph;
@@ -1569,10 +1569,9 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	cd_cmd = I40E_TX_CTX_DESC_TSO;
 	cd_tso_len = skb->len - *hdr_len;
 	cd_mss = skb_shinfo(skb)->gso_size;
-	*cd_type_cmd_tso_mss |= ((u64)cd_cmd << I40E_TXD_CTX_QW1_CMD_SHIFT) |
-				((u64)cd_tso_len <<
-				 I40E_TXD_CTX_QW1_TSO_LEN_SHIFT) |
-				((u64)cd_mss << I40E_TXD_CTX_QW1_MSS_SHIFT);
+	*cd_type_cmd_tso_mss |= (cd_cmd << I40E_TXD_CTX_QW1_CMD_SHIFT) |
+				(cd_tso_len << I40E_TXD_CTX_QW1_TSO_LEN_SHIFT) |
+				(cd_mss << I40E_TXD_CTX_QW1_MSS_SHIFT);
 	return 1;
 }
 
-- 
cgit v0.10.2


From c49a7bc3308099a8d5f9e2e38adfc5ab969804aa Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:28 -0800
Subject: i40e/i40evf: Factor out L4 header and checksum from L3 bits in TSO
 path

This patch makes it so that the L4 header offsets and such can be ignored
when dealing with the L3 checksum and length update.  This is done making
use of two things.

First we can just use the offset from the L4 header to the start of the
packet to determine the L4 offset, and from that we can then make use of
the data offset to determine the full length of the headers.

As far as adjusting the checksum to remove the length we can simply add the
inverse of the length instead of having to recompute the entire
pseudo-header without the length.  In the case of an IPv6 header this
should be significantly cheaper since we can make use of a value we already
needed instead of having to read the source and destination address out of
the packet.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 5e82589..ce0234e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2269,9 +2269,12 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 {
 	u64 cd_cmd, cd_tso_len, cd_mss;
 	struct ipv6hdr *ipv6h;
-	struct tcphdr *tcph;
 	struct iphdr *iph;
-	u32 l4len;
+	union {
+		struct tcphdr *tcp;
+		unsigned char *hdr;
+	} l4;
+	u32 paylen, l4_offset;
 	int err;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
@@ -2286,24 +2289,26 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 
 	iph = skb->encapsulation ? inner_ip_hdr(skb) : ip_hdr(skb);
 	ipv6h = skb->encapsulation ? inner_ipv6_hdr(skb) : ipv6_hdr(skb);
+	l4.hdr = skb->encapsulation ? skb_inner_transport_header(skb) :
+				      skb_transport_header(skb);
 
 	if (iph->version == 4) {
-		tcph = skb->encapsulation ? inner_tcp_hdr(skb) : tcp_hdr(skb);
 		iph->tot_len = 0;
 		iph->check = 0;
-		tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
-						 0, IPPROTO_TCP, 0);
-	} else if (ipv6h->version == 6) {
-		tcph = skb->encapsulation ? inner_tcp_hdr(skb) : tcp_hdr(skb);
+	} else {
 		ipv6h->payload_len = 0;
-		tcph->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
-					       0, IPPROTO_TCP, 0);
 	}
 
-	l4len = skb->encapsulation ? inner_tcp_hdrlen(skb) : tcp_hdrlen(skb);
-	*hdr_len = (skb->encapsulation
-		    ? (skb_inner_transport_header(skb) - skb->data)
-		    : skb_transport_offset(skb)) + l4len;
+	/* determine offset of inner transport header */
+	l4_offset = l4.hdr - skb->data;
+
+	/* remove payload length from inner checksum */
+	paylen = (__force u16)l4.tcp->check;
+	paylen += ntohs(1) * (u16)~(skb->len - l4_offset);
+	l4.tcp->check = ~csum_fold((__force __wsum)paylen);
+
+	/* compute length of segmentation header */
+	*hdr_len = (l4.tcp->doff * 4) + l4_offset;
 
 	/* find the field values */
 	cd_cmd = I40E_TX_CTX_DESC_TSO;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index c5f98cb..881d0ad 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1529,9 +1529,12 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 {
 	u64 cd_cmd, cd_tso_len, cd_mss;
 	struct ipv6hdr *ipv6h;
-	struct tcphdr *tcph;
 	struct iphdr *iph;
-	u32 l4len;
+	union {
+		struct tcphdr *tcp;
+		unsigned char *hdr;
+	} l4;
+	u32 paylen, l4_offset;
 	int err;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
@@ -1546,24 +1549,26 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 
 	iph = skb->encapsulation ? inner_ip_hdr(skb) : ip_hdr(skb);
 	ipv6h = skb->encapsulation ? inner_ipv6_hdr(skb) : ipv6_hdr(skb);
+	l4.hdr = skb->encapsulation ? skb_inner_transport_header(skb) :
+				      skb_transport_header(skb);
 
 	if (iph->version == 4) {
-		tcph = skb->encapsulation ? inner_tcp_hdr(skb) : tcp_hdr(skb);
 		iph->tot_len = 0;
 		iph->check = 0;
-		tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
-						 0, IPPROTO_TCP, 0);
-	} else if (ipv6h->version == 6) {
-		tcph = skb->encapsulation ? inner_tcp_hdr(skb) : tcp_hdr(skb);
+	} else {
 		ipv6h->payload_len = 0;
-		tcph->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
-					       0, IPPROTO_TCP, 0);
 	}
 
-	l4len = skb->encapsulation ? inner_tcp_hdrlen(skb) : tcp_hdrlen(skb);
-	*hdr_len = (skb->encapsulation
-		    ? (skb_inner_transport_header(skb) - skb->data)
-		    : skb_transport_offset(skb)) + l4len;
+	/* determine offset of inner transport header */
+	l4_offset = l4.hdr - skb->data;
+
+	/* remove payload length from inner checksum */
+	paylen = (__force u16)l4.tcp->check;
+	paylen += ntohs(1) * (u16)~(skb->len - l4_offset);
+	l4.tcp->check = ~csum_fold((__force __wsum)paylen);
+
+	/* compute length of segmentation header */
+	*hdr_len = (l4.tcp->doff * 4) + l4_offset;
 
 	/* find the field values */
 	cd_cmd = I40E_TX_CTX_DESC_TSO;
-- 
cgit v0.10.2


From c777019af1dc7343be8dc44bb4d32f5e2ef072dd Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:35 -0800
Subject: i40e/i40evf: Consolidate all header changes into TSO function

This patch goes through and pulls all of the spots where we were updating
either the TCP or IP checksums in the TSO and checksum path into the TSO
function.  The general idea here is that we should only be updating the
header after we verify we have completed a skb_cow_head check to verify the
head is writable.

One other advantage to doing this is that it makes things much more
obvious.  For example, in the case of IPv6 there was one spot where the
offset of the IPv4 header checksum was being updated which is obviously
incorrect.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index ce0234e..f47f9cb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2268,8 +2268,11 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		    u8 *hdr_len, u64 *cd_type_cmd_tso_mss)
 {
 	u64 cd_cmd, cd_tso_len, cd_mss;
-	struct ipv6hdr *ipv6h;
-	struct iphdr *iph;
+	union {
+		struct iphdr *v4;
+		struct ipv6hdr *v6;
+		unsigned char *hdr;
+	} ip;
 	union {
 		struct tcphdr *tcp;
 		unsigned char *hdr;
@@ -2287,16 +2290,29 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	if (err < 0)
 		return err;
 
-	iph = skb->encapsulation ? inner_ip_hdr(skb) : ip_hdr(skb);
-	ipv6h = skb->encapsulation ? inner_ipv6_hdr(skb) : ipv6_hdr(skb);
-	l4.hdr = skb->encapsulation ? skb_inner_transport_header(skb) :
-				      skb_transport_header(skb);
+	ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
 
-	if (iph->version == 4) {
-		iph->tot_len = 0;
-		iph->check = 0;
+	/* initialize outer IP header fields */
+	if (ip.v4->version == 4) {
+		ip.v4->tot_len = 0;
+		ip.v4->check = 0;
 	} else {
-		ipv6h->payload_len = 0;
+		ip.v6->payload_len = 0;
+	}
+
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE)) {
+		/* reset pointers to inner headers */
+		ip.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
+
+		/* initialize inner IP header fields */
+		if (ip.v4->version == 4) {
+			ip.v4->tot_len = 0;
+			ip.v4->check = 0;
+		} else {
+			ip.v6->payload_len = 0;
+		}
 	}
 
 	/* determine offset of inner transport header */
@@ -2381,15 +2397,11 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	struct iphdr *this_ip_hdr;
 	u32 network_hdr_len;
 	u8 l4_hdr = 0;
-	struct udphdr *oudph = NULL;
-	struct iphdr *oiph = NULL;
 	u32 l4_tunnel = 0;
 
 	if (skb->encapsulation) {
 		switch (ip_hdr(skb)->protocol) {
 		case IPPROTO_UDP:
-			oudph = udp_hdr(skb);
-			oiph = ip_hdr(skb);
 			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
@@ -2407,15 +2419,12 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
 			if (*tx_flags & I40E_TX_FLAGS_TSO) {
 				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
-				ip_hdr(skb)->check = 0;
 			} else {
 				*cd_tunneling |=
 					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
 			}
 		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
 			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
-			if (*tx_flags & I40E_TX_FLAGS_TSO)
-				ip_hdr(skb)->check = 0;
 		}
 
 		/* Now set the ctx descriptor fields */
@@ -2444,7 +2453,6 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		 */
 		if (*tx_flags & I40E_TX_FLAGS_TSO) {
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4_CSUM;
-			this_ip_hdr->check = 0;
 		} else {
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
 		}
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 881d0ad..b5b2527 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1528,8 +1528,11 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		    u8 *hdr_len, u64 *cd_type_cmd_tso_mss)
 {
 	u64 cd_cmd, cd_tso_len, cd_mss;
-	struct ipv6hdr *ipv6h;
-	struct iphdr *iph;
+	union {
+		struct iphdr *v4;
+		struct ipv6hdr *v6;
+		unsigned char *hdr;
+	} ip;
 	union {
 		struct tcphdr *tcp;
 		unsigned char *hdr;
@@ -1547,16 +1550,29 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	if (err < 0)
 		return err;
 
-	iph = skb->encapsulation ? inner_ip_hdr(skb) : ip_hdr(skb);
-	ipv6h = skb->encapsulation ? inner_ipv6_hdr(skb) : ipv6_hdr(skb);
-	l4.hdr = skb->encapsulation ? skb_inner_transport_header(skb) :
-				      skb_transport_header(skb);
+	ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
 
-	if (iph->version == 4) {
-		iph->tot_len = 0;
-		iph->check = 0;
+	/* initialize outer IP header fields */
+	if (ip.v4->version == 4) {
+		ip.v4->tot_len = 0;
+		ip.v4->check = 0;
 	} else {
-		ipv6h->payload_len = 0;
+		ip.v6->payload_len = 0;
+	}
+
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE)) {
+		/* reset pointers to inner headers */
+		ip.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
+
+		/* initialize inner IP header fields */
+		if (ip.v4->version == 4) {
+			ip.v4->tot_len = 0;
+			ip.v4->check = 0;
+		} else {
+			ip.v6->payload_len = 0;
+		}
 	}
 
 	/* determine offset of inner transport header */
@@ -1598,15 +1614,11 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	struct iphdr *this_ip_hdr;
 	u32 network_hdr_len;
 	u8 l4_hdr = 0;
-	struct udphdr *oudph;
-	struct iphdr *oiph;
 	u32 l4_tunnel = 0;
 
 	if (skb->encapsulation) {
 		switch (ip_hdr(skb)->protocol) {
 		case IPPROTO_UDP:
-			oudph = udp_hdr(skb);
-			oiph = ip_hdr(skb);
 			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
@@ -1621,15 +1633,12 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
 			if (*tx_flags & I40E_TX_FLAGS_TSO) {
 				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
-				ip_hdr(skb)->check = 0;
 			} else {
 				*cd_tunneling |=
 					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
 			}
 		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
 			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
-			if (*tx_flags & I40E_TX_FLAGS_TSO)
-				ip_hdr(skb)->check = 0;
 		}
 
 		/* Now set the ctx descriptor fields */
@@ -1659,7 +1668,6 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		 */
 		if (*tx_flags & I40E_TX_FLAGS_TSO) {
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4_CSUM;
-			this_ip_hdr->check = 0;
 		} else {
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
 		}
-- 
cgit v0.10.2


From b96b78f2b789ab8398e7ec0111bb8b4588ed42bf Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:42 -0800
Subject: i40e/i40evf: Replace header pointers with unions of pointers in Tx
 checksum path

The Tx checksum path was maintaining a set of 3 pointers and two lengths in
order to prepare the packet for being checksummed.  The thing is we only
really needed 2 pointers, and the lengths that were being maintained can
easily be computed.

As such we can replace the IPv4 and IPv6 header pointers with one single
union that represents both, or a generic pointer to the start of the
network header.  For the L4 headers we can do the same with TCP and a
generic pointer to the start of the transport header.  The length of the
TCP header is obtained by simply multiplying doff by 4, and the network
header length can be obtained by subtracting the network header pointer
from the transport header pointer.

While I was at it I renamed l4_hdr to l4_proto to make it a bit more clear
and less likely to be confused with l4.hdr which is the transport header
pointer.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index f47f9cb..5cc7e71 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2392,12 +2392,21 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 				struct i40e_ring *tx_ring,
 				u32 *cd_tunneling)
 {
-	struct ipv6hdr *this_ipv6_hdr;
-	unsigned int this_tcp_hdrlen;
-	struct iphdr *this_ip_hdr;
-	u32 network_hdr_len;
-	u8 l4_hdr = 0;
+	union {
+		struct iphdr *v4;
+		struct ipv6hdr *v6;
+		unsigned char *hdr;
+	} ip;
+	union {
+		struct tcphdr *tcp;
+		struct udphdr *udp;
+		unsigned char *hdr;
+	} l4;
 	u32 l4_tunnel = 0;
+	u8 l4_proto = 0;
+
+	ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
 
 	if (skb->encapsulation) {
 		switch (ip_hdr(skb)->protocol) {
@@ -2411,10 +2420,10 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		default:
 			return;
 		}
-		network_hdr_len = skb_inner_network_header_len(skb);
-		this_ip_hdr = inner_ip_hdr(skb);
-		this_ipv6_hdr = inner_ipv6_hdr(skb);
-		this_tcp_hdrlen = inner_tcp_hdrlen(skb);
+
+		/* switch L4 header pointer from outer to inner */
+		ip.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
 
 		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
 			if (*tx_flags & I40E_TX_FLAGS_TSO) {
@@ -2434,20 +2443,15 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 				   ((skb_inner_network_offset(skb) -
 					skb_transport_offset(skb)) >> 1) <<
 				   I40E_TXD_CTX_QW0_NATLEN_SHIFT;
-		if (this_ip_hdr->version == 6) {
+		if (ip.v6->version == 6) {
 			*tx_flags &= ~I40E_TX_FLAGS_IPV4;
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
 		}
-	} else {
-		network_hdr_len = skb_network_header_len(skb);
-		this_ip_hdr = ip_hdr(skb);
-		this_ipv6_hdr = ipv6_hdr(skb);
-		this_tcp_hdrlen = tcp_hdrlen(skb);
 	}
 
 	/* Enable IP checksum offloads */
 	if (*tx_flags & I40E_TX_FLAGS_IPV4) {
-		l4_hdr = this_ip_hdr->protocol;
+		l4_proto = ip.v4->protocol;
 		/* the stack computes the IP header already, the only time we
 		 * need the hardware to recompute it is in the case of TSO.
 		 */
@@ -2456,26 +2460,23 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		} else {
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
 		}
-		/* Now set the td_offset for IP header length */
-		*td_offset = (network_hdr_len >> 2) <<
-			      I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 	} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-		l4_hdr = this_ipv6_hdr->nexthdr;
+		l4_proto = ip.v6->nexthdr;
 		*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
-		/* Now set the td_offset for IP header length */
-		*td_offset = (network_hdr_len >> 2) <<
-			      I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 	}
+
+	/* Now set the td_offset for IP header length */
+	*td_offset = ((l4.hdr - ip.hdr) / 4) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 	/* words in MACLEN + dwords in IPLEN + dwords in L4Len */
 	*td_offset |= (skb_network_offset(skb) >> 1) <<
 		       I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
 
 	/* Enable L4 checksum offloads */
-	switch (l4_hdr) {
+	switch (l4_proto) {
 	case IPPROTO_TCP:
 		/* enable checksum offloads */
 		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_TCP;
-		*td_offset |= (this_tcp_hdrlen >> 2) <<
+		*td_offset |= l4.tcp->doff <<
 			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	case IPPROTO_SCTP:
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index b5b2527..29af3c9f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1609,12 +1609,21 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 				struct i40e_ring *tx_ring,
 				u32 *cd_tunneling)
 {
-	struct ipv6hdr *this_ipv6_hdr;
-	unsigned int this_tcp_hdrlen;
-	struct iphdr *this_ip_hdr;
-	u32 network_hdr_len;
-	u8 l4_hdr = 0;
+	union {
+		struct iphdr *v4;
+		struct ipv6hdr *v6;
+		unsigned char *hdr;
+	} ip;
+	union {
+		struct tcphdr *tcp;
+		struct udphdr *udp;
+		unsigned char *hdr;
+	} l4;
 	u32 l4_tunnel = 0;
+	u8 l4_proto = 0;
+
+	ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
 
 	if (skb->encapsulation) {
 		switch (ip_hdr(skb)->protocol) {
@@ -1625,10 +1634,10 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		default:
 			return;
 		}
-		network_hdr_len = skb_inner_network_header_len(skb);
-		this_ip_hdr = inner_ip_hdr(skb);
-		this_ipv6_hdr = inner_ipv6_hdr(skb);
-		this_tcp_hdrlen = inner_tcp_hdrlen(skb);
+
+		/* switch L4 header pointer from outer to inner */
+		ip.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
 
 		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
 			if (*tx_flags & I40E_TX_FLAGS_TSO) {
@@ -1648,21 +1657,15 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 				   ((skb_inner_network_offset(skb) -
 					skb_transport_offset(skb)) >> 1) <<
 				   I40E_TXD_CTX_QW0_NATLEN_SHIFT;
-		if (this_ip_hdr->version == 6) {
+		if (ip.v6->version == 6) {
 			*tx_flags &= ~I40E_TX_FLAGS_IPV4;
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
 		}
-
-	} else {
-		network_hdr_len = skb_network_header_len(skb);
-		this_ip_hdr = ip_hdr(skb);
-		this_ipv6_hdr = ipv6_hdr(skb);
-		this_tcp_hdrlen = tcp_hdrlen(skb);
 	}
 
 	/* Enable IP checksum offloads */
 	if (*tx_flags & I40E_TX_FLAGS_IPV4) {
-		l4_hdr = this_ip_hdr->protocol;
+		l4_proto = ip.v4->protocol;
 		/* the stack computes the IP header already, the only time we
 		 * need the hardware to recompute it is in the case of TSO.
 		 */
@@ -1671,26 +1674,23 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		} else {
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
 		}
-		/* Now set the td_offset for IP header length */
-		*td_offset = (network_hdr_len >> 2) <<
-			      I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 	} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-		l4_hdr = this_ipv6_hdr->nexthdr;
+		l4_proto = ip.v6->nexthdr;
 		*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
-		/* Now set the td_offset for IP header length */
-		*td_offset = (network_hdr_len >> 2) <<
-			      I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 	}
+
+	/* Now set the td_offset for IP header length */
+	*td_offset = ((l4.hdr - ip.hdr) / 4) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 	/* words in MACLEN + dwords in IPLEN + dwords in L4Len */
 	*td_offset |= (skb_network_offset(skb) >> 1) <<
 		       I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
 
 	/* Enable L4 checksum offloads */
-	switch (l4_hdr) {
+	switch (l4_proto) {
 	case IPPROTO_TCP:
 		/* enable checksum offloads */
 		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_TCP;
-		*td_offset |= (this_tcp_hdrlen >> 2) <<
+		*td_offset |= l4.tcp->doff <<
 			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	case IPPROTO_SCTP:
-- 
cgit v0.10.2


From a0064728f8a34f7a5afd9df86d9cdd8210977c8d Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:48 -0800
Subject: i40e/i40evf: Add support for IPv4 encapsulated in IPv6

This patch fixes two issues.  First was the fact that iphdr(skb)->protocl
was being used to test for the outer transport protocol.  This completely
breaks IPv6 support.  Second was the fact that we cleared the flag for v4
going to v6, but we didn't take care of txflags going the other way.  As
such we would have the v6 flag still set even if the inner header was v4.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 5cc7e71..1404cae 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2409,13 +2409,28 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	l4.hdr = skb_transport_header(skb);
 
 	if (skb->encapsulation) {
-		switch (ip_hdr(skb)->protocol) {
+		/* define outer network header type */
+		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
+			if (*tx_flags & I40E_TX_FLAGS_TSO)
+				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
+			else
+				*cd_tunneling |=
+					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
+			l4_proto = ip.v4->protocol;
+		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
+			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
+			l4_proto = ip.v6->nexthdr;
+		}
+
+		/* define outer transport */
+		switch (l4_proto) {
 		case IPPROTO_UDP:
 			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
 		case IPPROTO_GRE:
 			l4_tunnel = I40E_TXD_CTX_GRE_TUNNELING;
+			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
 		default:
 			return;
@@ -2424,17 +2439,7 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		/* switch L4 header pointer from outer to inner */
 		ip.hdr = skb_inner_network_header(skb);
 		l4.hdr = skb_inner_transport_header(skb);
-
-		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
-			if (*tx_flags & I40E_TX_FLAGS_TSO) {
-				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
-			} else {
-				*cd_tunneling |=
-					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
-			}
-		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
-		}
+		l4_proto = 0;
 
 		/* Now set the ctx descriptor fields */
 		*cd_tunneling |= (skb_network_header_len(skb) >> 2) <<
@@ -2443,10 +2448,13 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 				   ((skb_inner_network_offset(skb) -
 					skb_transport_offset(skb)) >> 1) <<
 				   I40E_TXD_CTX_QW0_NATLEN_SHIFT;
-		if (ip.v6->version == 6) {
-			*tx_flags &= ~I40E_TX_FLAGS_IPV4;
+
+		/* reset type as we transition from outer to inner headers */
+		*tx_flags &= ~(I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6);
+		if (ip.v4->version == 4)
+			*tx_flags |= I40E_TX_FLAGS_IPV4;
+		if (ip.v6->version == 6)
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
-		}
 	}
 
 	/* Enable IP checksum offloads */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 29af3c9f..39d5f80 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1626,11 +1626,29 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	l4.hdr = skb_transport_header(skb);
 
 	if (skb->encapsulation) {
-		switch (ip_hdr(skb)->protocol) {
+		/* define outer network header type */
+		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
+			if (*tx_flags & I40E_TX_FLAGS_TSO)
+				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
+			else
+				*cd_tunneling |=
+					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
+			l4_proto = ip.v4->protocol;
+		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
+			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
+			l4_proto = ip.v6->nexthdr;
+		}
+
+		/* define outer transport */
+		switch (l4_proto) {
 		case IPPROTO_UDP:
 			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
+		case IPPROTO_GRE:
+			l4_tunnel = I40E_TXD_CTX_GRE_TUNNELING;
+			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
+			break;
 		default:
 			return;
 		}
@@ -1638,17 +1656,7 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		/* switch L4 header pointer from outer to inner */
 		ip.hdr = skb_inner_network_header(skb);
 		l4.hdr = skb_inner_transport_header(skb);
-
-		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
-			if (*tx_flags & I40E_TX_FLAGS_TSO) {
-				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
-			} else {
-				*cd_tunneling |=
-					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
-			}
-		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
-		}
+		l4_proto = 0;
 
 		/* Now set the ctx descriptor fields */
 		*cd_tunneling |= (skb_network_header_len(skb) >> 2) <<
@@ -1657,10 +1665,13 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 				   ((skb_inner_network_offset(skb) -
 					skb_transport_offset(skb)) >> 1) <<
 				   I40E_TXD_CTX_QW0_NATLEN_SHIFT;
-		if (ip.v6->version == 6) {
-			*tx_flags &= ~I40E_TX_FLAGS_IPV4;
+
+		/* reset type as we transition from outer to inner headers */
+		*tx_flags &= ~(I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6);
+		if (ip.v4->version == 4)
+			*tx_flags |= I40E_TX_FLAGS_IPV4;
+		if (ip.v6->version == 6)
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
-		}
 	}
 
 	/* Enable IP checksum offloads */
-- 
cgit v0.10.2


From a3fd9d8876a589f05725237aced606b995956860 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:16:54 -0800
Subject: i40e/i40evf: Handle IPv6 extension headers in checksum offload

This patch adds support for IPv6 extension headers in setting up the Tx
checksum.  Without this patch extension headers would cause IPv6 traffic to
fail as the transport protocol could not be identified.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 1404cae..e49fe8f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2402,7 +2402,9 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		struct udphdr *udp;
 		unsigned char *hdr;
 	} l4;
+	unsigned char *exthdr;
 	u32 l4_tunnel = 0;
+	__be16 frag_off;
 	u8 l4_proto = 0;
 
 	ip.hdr = skb_network_header(skb);
@@ -2419,7 +2421,12 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			l4_proto = ip.v4->protocol;
 		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
 			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
+
+			exthdr = ip.hdr + sizeof(*ip.v6);
 			l4_proto = ip.v6->nexthdr;
+			if (l4.hdr != exthdr)
+				ipv6_skip_exthdr(skb, exthdr - skb->data,
+						 &l4_proto, &frag_off);
 		}
 
 		/* define outer transport */
@@ -2469,8 +2476,13 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
 		}
 	} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-		l4_proto = ip.v6->nexthdr;
 		*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
+
+		exthdr = ip.hdr + sizeof(*ip.v6);
+		l4_proto = ip.v6->nexthdr;
+		if (l4.hdr != exthdr)
+			ipv6_skip_exthdr(skb, exthdr - skb->data,
+					 &l4_proto, &frag_off);
 	}
 
 	/* Now set the td_offset for IP header length */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 39d5f80..48ec763 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1619,7 +1619,9 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		struct udphdr *udp;
 		unsigned char *hdr;
 	} l4;
+	unsigned char *exthdr;
 	u32 l4_tunnel = 0;
+	__be16 frag_off;
 	u8 l4_proto = 0;
 
 	ip.hdr = skb_network_header(skb);
@@ -1636,7 +1638,12 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			l4_proto = ip.v4->protocol;
 		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
 			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
+
+			exthdr = ip.hdr + sizeof(*ip.v6);
 			l4_proto = ip.v6->nexthdr;
+			if (l4.hdr != exthdr)
+				ipv6_skip_exthdr(skb, exthdr - skb->data,
+						 &l4_proto, &frag_off);
 		}
 
 		/* define outer transport */
@@ -1686,8 +1693,13 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
 		}
 	} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-		l4_proto = ip.v6->nexthdr;
 		*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
+
+		exthdr = ip.hdr + sizeof(*ip.v6);
+		l4_proto = ip.v6->nexthdr;
+		if (l4.hdr != exthdr)
+			ipv6_skip_exthdr(skb, exthdr - skb->data,
+					 &l4_proto, &frag_off);
 	}
 
 	/* Now set the td_offset for IP header length */
-- 
cgit v0.10.2


From 07dabf20d9867710b90b91108b2adcd448773e25 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Thu, 18 Feb 2016 19:19:29 +0100
Subject: vxlan: tun_id is 64bit, not 32bit

The tun_id field in struct ip_tunnel_key is __be64, not __be32. We need to
convert the vni to tun_id correctly.

Fixes: 54bfd872bf16 ("vxlan: keep flags and vni in network byte order")
Reported-by: Paolo Abeni <pabeni@redhat.com>
Tested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Acked-by: Thadeu Lima de Souza Cascardo <cascardo@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 3a84680..75bccb3 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1310,9 +1310,10 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto drop;
 
 	if (vxlan_collect_metadata(vs)) {
+		__be32 vni = vxlan_vni(vxlan_hdr(skb)->vx_vni);
+
 		tun_dst = udp_tun_rx_dst(skb, vxlan_get_sk_family(vs), TUNNEL_KEY,
-					 vxlan_vni(vxlan_hdr(skb)->vx_vni),
-					 sizeof(*md));
+					 vxlan_vni_to_tun_id(vni), sizeof(*md));
 
 		if (!tun_dst)
 			goto drop;
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 1b85a3b..748083d 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -294,6 +294,15 @@ static inline __be32 vxlan_tun_id_to_vni(__be64 tun_id)
 #endif
 }
 
+static inline __be64 vxlan_vni_to_tun_id(__be32 vni)
+{
+#if defined(__BIG_ENDIAN)
+	return (__be64)vni;
+#else
+	return (__be64)vni << 32;
+#endif
+}
+
 static inline size_t vxlan_rco_start(__be32 vni_field)
 {
 	return be32_to_cpu(vni_field & VXLAN_RCO_MASK) << VXLAN_RCO_SHIFT;
-- 
cgit v0.10.2


From 475b4205aa52c16feef08d55c8fd76e815b6bee7 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:01 -0800
Subject: i40e/i40evf: Do not write to descriptor unless we complete

This patch defers writing to the Tx descriptor bits until we know we have
successfully completed a given operation.  So for example we defer updating
the tunnelling portion of the context descriptor until we have fully
identified the type.

The advantage to this approach is that we can assemble values as we go
instead of having to try and kludge everything together all at once.  As a
result we can significantly clean up the tunneling configuration for
instance as we can just do a pointer walk and do the math for the distance
between each set of points.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index e49fe8f..5b591b8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2403,24 +2403,26 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		unsigned char *hdr;
 	} l4;
 	unsigned char *exthdr;
-	u32 l4_tunnel = 0;
+	u32 offset, cmd = 0, tunnel = 0;
 	__be16 frag_off;
 	u8 l4_proto = 0;
 
 	ip.hdr = skb_network_header(skb);
 	l4.hdr = skb_transport_header(skb);
 
+	/* compute outer L2 header size */
+	offset = ((ip.hdr - skb->data) / 2) << I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
+
 	if (skb->encapsulation) {
 		/* define outer network header type */
 		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
-			if (*tx_flags & I40E_TX_FLAGS_TSO)
-				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
-			else
-				*cd_tunneling |=
-					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
+			tunnel |= (*tx_flags & I40E_TX_FLAGS_TSO) ?
+				  I40E_TX_CTX_EXT_IP_IPV4 :
+				  I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
+
 			l4_proto = ip.v4->protocol;
 		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
+			tunnel |= I40E_TX_CTX_EXT_IP_IPV6;
 
 			exthdr = ip.hdr + sizeof(*ip.v6);
 			l4_proto = ip.v6->nexthdr;
@@ -2429,33 +2431,38 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 						 &l4_proto, &frag_off);
 		}
 
+		/* compute outer L3 header size */
+		tunnel |= ((l4.hdr - ip.hdr) / 4) <<
+			  I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT;
+
+		/* switch IP header pointer from outer to inner header */
+		ip.hdr = skb_inner_network_header(skb);
+
 		/* define outer transport */
 		switch (l4_proto) {
 		case IPPROTO_UDP:
-			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
+			tunnel |= I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
 		case IPPROTO_GRE:
-			l4_tunnel = I40E_TXD_CTX_GRE_TUNNELING;
+			tunnel |= I40E_TXD_CTX_GRE_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
 		default:
 			return;
 		}
 
+		/* compute tunnel header size */
+		tunnel |= ((ip.hdr - l4.hdr) / 2) <<
+			  I40E_TXD_CTX_QW0_NATLEN_SHIFT;
+
+		/* record tunnel offload values */
+		*cd_tunneling |= tunnel;
+
 		/* switch L4 header pointer from outer to inner */
-		ip.hdr = skb_inner_network_header(skb);
 		l4.hdr = skb_inner_transport_header(skb);
 		l4_proto = 0;
 
-		/* Now set the ctx descriptor fields */
-		*cd_tunneling |= (skb_network_header_len(skb) >> 2) <<
-				   I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT      |
-				   l4_tunnel                             |
-				   ((skb_inner_network_offset(skb) -
-					skb_transport_offset(skb)) >> 1) <<
-				   I40E_TXD_CTX_QW0_NATLEN_SHIFT;
-
 		/* reset type as we transition from outer to inner headers */
 		*tx_flags &= ~(I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6);
 		if (ip.v4->version == 4)
@@ -2470,13 +2477,11 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		/* the stack computes the IP header already, the only time we
 		 * need the hardware to recompute it is in the case of TSO.
 		 */
-		if (*tx_flags & I40E_TX_FLAGS_TSO) {
-			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4_CSUM;
-		} else {
-			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
-		}
+		cmd |= (*tx_flags & I40E_TX_FLAGS_TSO) ?
+		       I40E_TX_DESC_CMD_IIPT_IPV4_CSUM :
+		       I40E_TX_DESC_CMD_IIPT_IPV4;
 	} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-		*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
+		cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
 
 		exthdr = ip.hdr + sizeof(*ip.v6);
 		l4_proto = ip.v6->nexthdr;
@@ -2485,35 +2490,34 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 					 &l4_proto, &frag_off);
 	}
 
-	/* Now set the td_offset for IP header length */
-	*td_offset = ((l4.hdr - ip.hdr) / 4) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
-	/* words in MACLEN + dwords in IPLEN + dwords in L4Len */
-	*td_offset |= (skb_network_offset(skb) >> 1) <<
-		       I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
+	/* compute inner L3 header size */
+	offset |= ((l4.hdr - ip.hdr) / 4) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 
 	/* Enable L4 checksum offloads */
 	switch (l4_proto) {
 	case IPPROTO_TCP:
 		/* enable checksum offloads */
-		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_TCP;
-		*td_offset |= l4.tcp->doff <<
-			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
+		cmd |= I40E_TX_DESC_CMD_L4T_EOFT_TCP;
+		offset |= l4.tcp->doff << I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	case IPPROTO_SCTP:
 		/* enable SCTP checksum offload */
-		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_SCTP;
-		*td_offset |= (sizeof(struct sctphdr) >> 2) <<
-			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
+		cmd |= I40E_TX_DESC_CMD_L4T_EOFT_SCTP;
+		offset |= (sizeof(struct sctphdr) >> 2) <<
+			  I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	case IPPROTO_UDP:
 		/* enable UDP checksum offload */
-		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_UDP;
-		*td_offset |= (sizeof(struct udphdr) >> 2) <<
-			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
+		cmd |= I40E_TX_DESC_CMD_L4T_EOFT_UDP;
+		offset |= (sizeof(struct udphdr) >> 2) <<
+			  I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	default:
 		break;
 	}
+
+	*td_cmd |= cmd;
+	*td_offset |= offset;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 48ec763..0ee13f6 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1620,24 +1620,26 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		unsigned char *hdr;
 	} l4;
 	unsigned char *exthdr;
-	u32 l4_tunnel = 0;
+	u32 offset, cmd = 0, tunnel = 0;
 	__be16 frag_off;
 	u8 l4_proto = 0;
 
 	ip.hdr = skb_network_header(skb);
 	l4.hdr = skb_transport_header(skb);
 
+	/* compute outer L2 header size */
+	offset = ((ip.hdr - skb->data) / 2) << I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
+
 	if (skb->encapsulation) {
 		/* define outer network header type */
 		if (*tx_flags & I40E_TX_FLAGS_IPV4) {
-			if (*tx_flags & I40E_TX_FLAGS_TSO)
-				*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV4;
-			else
-				*cd_tunneling |=
-					 I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
+			tunnel |= (*tx_flags & I40E_TX_FLAGS_TSO) ?
+				  I40E_TX_CTX_EXT_IP_IPV4 :
+				  I40E_TX_CTX_EXT_IP_IPV4_NO_CSUM;
+
 			l4_proto = ip.v4->protocol;
 		} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-			*cd_tunneling |= I40E_TX_CTX_EXT_IP_IPV6;
+			tunnel |= I40E_TX_CTX_EXT_IP_IPV6;
 
 			exthdr = ip.hdr + sizeof(*ip.v6);
 			l4_proto = ip.v6->nexthdr;
@@ -1646,33 +1648,38 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 						 &l4_proto, &frag_off);
 		}
 
+		/* compute outer L3 header size */
+		tunnel |= ((l4.hdr - ip.hdr) / 4) <<
+			  I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT;
+
+		/* switch IP header pointer from outer to inner header */
+		ip.hdr = skb_inner_network_header(skb);
+
 		/* define outer transport */
 		switch (l4_proto) {
 		case IPPROTO_UDP:
-			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
+			tunnel |= I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
 		case IPPROTO_GRE:
-			l4_tunnel = I40E_TXD_CTX_GRE_TUNNELING;
+			tunnel |= I40E_TXD_CTX_GRE_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
 		default:
 			return;
 		}
 
+		/* compute tunnel header size */
+		tunnel |= ((ip.hdr - l4.hdr) / 2) <<
+			  I40E_TXD_CTX_QW0_NATLEN_SHIFT;
+
+		/* record tunnel offload values */
+		*cd_tunneling |= tunnel;
+
 		/* switch L4 header pointer from outer to inner */
-		ip.hdr = skb_inner_network_header(skb);
 		l4.hdr = skb_inner_transport_header(skb);
 		l4_proto = 0;
 
-		/* Now set the ctx descriptor fields */
-		*cd_tunneling |= (skb_network_header_len(skb) >> 2) <<
-				   I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT      |
-				   l4_tunnel                             |
-				   ((skb_inner_network_offset(skb) -
-					skb_transport_offset(skb)) >> 1) <<
-				   I40E_TXD_CTX_QW0_NATLEN_SHIFT;
-
 		/* reset type as we transition from outer to inner headers */
 		*tx_flags &= ~(I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6);
 		if (ip.v4->version == 4)
@@ -1687,13 +1694,11 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		/* the stack computes the IP header already, the only time we
 		 * need the hardware to recompute it is in the case of TSO.
 		 */
-		if (*tx_flags & I40E_TX_FLAGS_TSO) {
-			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4_CSUM;
-		} else {
-			*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV4;
-		}
+		cmd |= (*tx_flags & I40E_TX_FLAGS_TSO) ?
+		       I40E_TX_DESC_CMD_IIPT_IPV4_CSUM :
+		       I40E_TX_DESC_CMD_IIPT_IPV4;
 	} else if (*tx_flags & I40E_TX_FLAGS_IPV6) {
-		*td_cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
+		cmd |= I40E_TX_DESC_CMD_IIPT_IPV6;
 
 		exthdr = ip.hdr + sizeof(*ip.v6);
 		l4_proto = ip.v6->nexthdr;
@@ -1702,35 +1707,34 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 					 &l4_proto, &frag_off);
 	}
 
-	/* Now set the td_offset for IP header length */
-	*td_offset = ((l4.hdr - ip.hdr) / 4) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
-	/* words in MACLEN + dwords in IPLEN + dwords in L4Len */
-	*td_offset |= (skb_network_offset(skb) >> 1) <<
-		       I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
+	/* compute inner L3 header size */
+	offset |= ((l4.hdr - ip.hdr) / 4) << I40E_TX_DESC_LENGTH_IPLEN_SHIFT;
 
 	/* Enable L4 checksum offloads */
 	switch (l4_proto) {
 	case IPPROTO_TCP:
 		/* enable checksum offloads */
-		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_TCP;
-		*td_offset |= l4.tcp->doff <<
-			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
+		cmd |= I40E_TX_DESC_CMD_L4T_EOFT_TCP;
+		offset |= l4.tcp->doff << I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	case IPPROTO_SCTP:
 		/* enable SCTP checksum offload */
-		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_SCTP;
-		*td_offset |= (sizeof(struct sctphdr) >> 2) <<
-			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
+		cmd |= I40E_TX_DESC_CMD_L4T_EOFT_SCTP;
+		offset |= (sizeof(struct sctphdr) >> 2) <<
+			  I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	case IPPROTO_UDP:
 		/* enable UDP checksum offload */
-		*td_cmd |= I40E_TX_DESC_CMD_L4T_EOFT_UDP;
-		*td_offset |= (sizeof(struct udphdr) >> 2) <<
-			       I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
+		cmd |= I40E_TX_DESC_CMD_L4T_EOFT_UDP;
+		offset |= (sizeof(struct udphdr) >> 2) <<
+			  I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	default:
 		break;
 	}
+
+	*td_cmd |= cmd;
+	*td_offset |= offset;
 }
 
 /**
-- 
cgit v0.10.2


From 529f1f652e3c3c6db6ab5a6e3a35469ddfd9575d Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:10 -0800
Subject: i40e/i40evf: Add exception handling for Tx checksum

Add exception handling to the Tx checksum path so that we can handle cases
of TSO where the frame is bad, or Tx checksum where we didn't recognize a
protocol

Drop I40E_TX_FLAGS_CSUM as it is unused, move the CHECKSUM_PARTIAL check
into the function itself so that we can decrease indent.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 5b591b8..6b08b0f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2387,10 +2387,10 @@ static int i40e_tsyn(struct i40e_ring *tx_ring, struct sk_buff *skb,
  * @tx_ring: Tx descriptor ring
  * @cd_tunneling: ptr to context desc bits
  **/
-static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
-				u32 *td_cmd, u32 *td_offset,
-				struct i40e_ring *tx_ring,
-				u32 *cd_tunneling)
+static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
+			       u32 *td_cmd, u32 *td_offset,
+			       struct i40e_ring *tx_ring,
+			       u32 *cd_tunneling)
 {
 	union {
 		struct iphdr *v4;
@@ -2407,6 +2407,9 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	__be16 frag_off;
 	u8 l4_proto = 0;
 
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
 	ip.hdr = skb_network_header(skb);
 	l4.hdr = skb_transport_header(skb);
 
@@ -2449,7 +2452,11 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
 		default:
-			return;
+			if (*tx_flags & I40E_TX_FLAGS_TSO)
+				return -1;
+
+			skb_checksum_help(skb);
+			return 0;
 		}
 
 		/* compute tunnel header size */
@@ -2513,11 +2520,16 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			  I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	default:
-		break;
+		if (*tx_flags & I40E_TX_FLAGS_TSO)
+			return -1;
+		skb_checksum_help(skb);
+		return 0;
 	}
 
 	*td_cmd |= cmd;
 	*td_offset |= offset;
+
+	return 1;
 }
 
 /**
@@ -2954,12 +2966,10 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	td_cmd |= I40E_TX_DESC_CMD_ICRC;
 
 	/* Always offload the checksum, since it's in the data descriptor */
-	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		tx_flags |= I40E_TX_FLAGS_CSUM;
-
-		i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
-				    tx_ring, &cd_tunneling);
-	}
+	tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
+				  tx_ring, &cd_tunneling);
+	if (tso < 0)
+		goto out_drop;
 
 	i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss,
 			   cd_tunneling, cd_l2tag2);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index fb065d4..fde5f42 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -153,7 +153,6 @@ enum i40e_dyn_idx_t {
 #define DESC_NEEDED (MAX_SKB_FRAGS + 4)
 #define I40E_MIN_DESC_PENDING	4
 
-#define I40E_TX_FLAGS_CSUM		BIT(0)
 #define I40E_TX_FLAGS_HW_VLAN		BIT(1)
 #define I40E_TX_FLAGS_SW_VLAN		BIT(2)
 #define I40E_TX_FLAGS_TSO		BIT(3)
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 0ee13f6..143c570 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1602,12 +1602,13 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
  * @tx_flags: pointer to Tx flags currently set
  * @td_cmd: Tx descriptor command bits to set
  * @td_offset: Tx descriptor header offsets to set
+ * @tx_ring: Tx descriptor ring
  * @cd_tunneling: ptr to context desc bits
  **/
-static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
-				u32 *td_cmd, u32 *td_offset,
-				struct i40e_ring *tx_ring,
-				u32 *cd_tunneling)
+static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
+			       u32 *td_cmd, u32 *td_offset,
+			       struct i40e_ring *tx_ring,
+			       u32 *cd_tunneling)
 {
 	union {
 		struct iphdr *v4;
@@ -1624,6 +1625,9 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	__be16 frag_off;
 	u8 l4_proto = 0;
 
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return 0;
+
 	ip.hdr = skb_network_header(skb);
 	l4.hdr = skb_transport_header(skb);
 
@@ -1666,7 +1670,11 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
 		default:
-			return;
+			if (*tx_flags & I40E_TX_FLAGS_TSO)
+				return -1;
+
+			skb_checksum_help(skb);
+			return 0;
 		}
 
 		/* compute tunnel header size */
@@ -1730,11 +1738,16 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			  I40E_TX_DESC_LENGTH_L4_FC_LEN_SHIFT;
 		break;
 	default:
-		break;
+		if (*tx_flags & I40E_TX_FLAGS_TSO)
+			return -1;
+		skb_checksum_help(skb);
+		return 0;
 	}
 
 	*td_cmd |= cmd;
 	*td_offset |= offset;
+
+	return 1;
 }
 
 /**
@@ -2150,12 +2163,10 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	td_cmd |= I40E_TX_DESC_CMD_ICRC;
 
 	/* Always offload the checksum, since it's in the data descriptor */
-	if (skb->ip_summed == CHECKSUM_PARTIAL) {
-		tx_flags |= I40E_TX_FLAGS_CSUM;
-
-		i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
-				    tx_ring, &cd_tunneling);
-	}
+	tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
+				  tx_ring, &cd_tunneling);
+	if (tso < 0)
+		goto out_drop;
 
 	i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss,
 			   cd_tunneling, cd_l2tag2);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 043b955..6ea8701 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -153,7 +153,6 @@ enum i40e_dyn_idx_t {
 #define DESC_NEEDED (MAX_SKB_FRAGS + 4)
 #define I40E_MIN_DESC_PENDING	4
 
-#define I40E_TX_FLAGS_CSUM		BIT(0)
 #define I40E_TX_FLAGS_HW_VLAN		BIT(1)
 #define I40E_TX_FLAGS_SW_VLAN		BIT(2)
 #define I40E_TX_FLAGS_TSO		BIT(3)
-- 
cgit v0.10.2


From fc08a01a6925a1a0d69bb9026f266606a6a96a20 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 16 Feb 2016 10:07:09 +0530
Subject: cxgb4: Use __dev_uc_sync/__dev_mc_sync to sync MAC address

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index ec6e849..1dac6c6 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -702,6 +702,11 @@ struct doorbell_stats {
 	u32 db_full;
 };
 
+struct hash_mac_addr {
+	struct list_head list;
+	u8 addr[ETH_ALEN];
+};
+
 struct adapter {
 	void __iomem *regs;
 	void __iomem *bar2;
@@ -740,6 +745,7 @@ struct adapter {
 	void *uld_handle[CXGB4_ULD_MAX];
 	struct list_head list_node;
 	struct list_head rcu_node;
+	struct list_head mac_hlist; /* list of MAC addresses in MPS Hash */
 
 	struct tid_info tids;
 	void **tid_release_head;
@@ -1207,6 +1213,24 @@ static inline int t4_wr_mbox_ns(struct adapter *adap, int mbox, const void *cmd,
 	return t4_wr_mbox_meat(adap, mbox, cmd, size, rpl, false);
 }
 
+/**
+ *	hash_mac_addr - return the hash value of a MAC address
+ *	@addr: the 48-bit Ethernet MAC address
+ *
+ *	Hashes a MAC address according to the hash function used by HW inexact
+ *	(hash) address matching.
+ */
+static inline int hash_mac_addr(const u8 *addr)
+{
+	u32 a = ((u32)addr[0] << 16) | ((u32)addr[1] << 8) | addr[2];
+	u32 b = ((u32)addr[3] << 16) | ((u32)addr[4] << 8) | addr[5];
+
+	a ^= b;
+	a ^= (a >> 12);
+	a ^= (a >> 6);
+	return a & 0x3f;
+}
+
 void t4_write_indirect(struct adapter *adap, unsigned int addr_reg,
 		       unsigned int data_reg, const u32 *vals,
 		       unsigned int nregs, unsigned int start_idx);
@@ -1389,6 +1413,9 @@ int t4_set_rxmode(struct adapter *adap, unsigned int mbox, unsigned int viid,
 int t4_alloc_mac_filt(struct adapter *adap, unsigned int mbox,
 		      unsigned int viid, bool free, unsigned int naddr,
 		      const u8 **addr, u16 *idx, u64 *hash, bool sleep_ok);
+int t4_free_mac_filt(struct adapter *adap, unsigned int mbox,
+		     unsigned int viid, unsigned int naddr,
+		     const u8 **addr, bool sleep_ok);
 int t4_change_mac(struct adapter *adap, unsigned int mbox, unsigned int viid,
 		  int idx, const u8 *addr, bool persist, bool add_smt);
 int t4_set_addr_hash(struct adapter *adap, unsigned int mbox, unsigned int viid,
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index b8a5fb0..adad73f 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -338,84 +338,108 @@ void t4_os_portmod_changed(const struct adapter *adap, int port_id)
 		netdev_info(dev, "%s module inserted\n", mod_str[pi->mod_type]);
 }
 
+int dbfifo_int_thresh = 10; /* 10 == 640 entry threshold */
+module_param(dbfifo_int_thresh, int, 0644);
+MODULE_PARM_DESC(dbfifo_int_thresh, "doorbell fifo interrupt threshold");
+
 /*
- * Configure the exact and hash address filters to handle a port's multicast
- * and secondary unicast MAC addresses.
+ * usecs to sleep while draining the dbfifo
  */
-static int set_addr_filters(const struct net_device *dev, bool sleep)
+static int dbfifo_drain_delay = 1000;
+module_param(dbfifo_drain_delay, int, 0644);
+MODULE_PARM_DESC(dbfifo_drain_delay,
+		 "usecs to sleep while draining the dbfifo");
+
+static inline int cxgb4_set_addr_hash(struct port_info *pi)
 {
+	struct adapter *adap = pi->adapter;
+	u64 vec = 0;
+	bool ucast = false;
+	struct hash_mac_addr *entry;
+
+	/* Calculate the hash vector for the updated list and program it */
+	list_for_each_entry(entry, &adap->mac_hlist, list) {
+		ucast |= is_unicast_ether_addr(entry->addr);
+		vec |= (1ULL << hash_mac_addr(entry->addr));
+	}
+	return t4_set_addr_hash(adap, adap->mbox, pi->viid, ucast,
+				vec, false);
+}
+
+static int cxgb4_mac_sync(struct net_device *netdev, const u8 *mac_addr)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adap = pi->adapter;
+	int ret;
 	u64 mhash = 0;
 	u64 uhash = 0;
-	bool free = true;
-	u16 filt_idx[7];
-	const u8 *addr[7];
-	int ret, naddr = 0;
-	const struct netdev_hw_addr *ha;
-	int uc_cnt = netdev_uc_count(dev);
-	int mc_cnt = netdev_mc_count(dev);
-	const struct port_info *pi = netdev_priv(dev);
-	unsigned int mb = pi->adapter->pf;
+	bool free = false;
+	bool ucast = is_unicast_ether_addr(mac_addr);
+	const u8 *maclist[1] = {mac_addr};
+	struct hash_mac_addr *new_entry;
 
-	/* first do the secondary unicast addresses */
-	netdev_for_each_uc_addr(ha, dev) {
-		addr[naddr++] = ha->addr;
-		if (--uc_cnt == 0 || naddr >= ARRAY_SIZE(addr)) {
-			ret = t4_alloc_mac_filt(pi->adapter, mb, pi->viid, free,
-					naddr, addr, filt_idx, &uhash, sleep);
-			if (ret < 0)
-				return ret;
-
-			free = false;
-			naddr = 0;
-		}
+	ret = t4_alloc_mac_filt(adap, adap->mbox, pi->viid, free, 1, maclist,
+				NULL, ucast ? &uhash : &mhash, false);
+	if (ret < 0)
+		goto out;
+	/* if hash != 0, then add the addr to hash addr list
+	 * so on the end we will calculate the hash for the
+	 * list and program it
+	 */
+	if (uhash || mhash) {
+		new_entry = kzalloc(sizeof(*new_entry), GFP_ATOMIC);
+		if (!new_entry)
+			return -ENOMEM;
+		ether_addr_copy(new_entry->addr, mac_addr);
+		list_add_tail(&new_entry->list, &adap->mac_hlist);
+		ret = cxgb4_set_addr_hash(pi);
 	}
+out:
+	return ret < 0 ? ret : 0;
+}
 
-	/* next set up the multicast addresses */
-	netdev_for_each_mc_addr(ha, dev) {
-		addr[naddr++] = ha->addr;
-		if (--mc_cnt == 0 || naddr >= ARRAY_SIZE(addr)) {
-			ret = t4_alloc_mac_filt(pi->adapter, mb, pi->viid, free,
-					naddr, addr, filt_idx, &mhash, sleep);
-			if (ret < 0)
-				return ret;
+static int cxgb4_mac_unsync(struct net_device *netdev, const u8 *mac_addr)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adap = pi->adapter;
+	int ret;
+	const u8 *maclist[1] = {mac_addr};
+	struct hash_mac_addr *entry, *tmp;
 
-			free = false;
-			naddr = 0;
+	/* If the MAC address to be removed is in the hash addr
+	 * list, delete it from the list and update hash vector
+	 */
+	list_for_each_entry_safe(entry, tmp, &adap->mac_hlist, list) {
+		if (ether_addr_equal(entry->addr, mac_addr)) {
+			list_del(&entry->list);
+			kfree(entry);
+			return cxgb4_set_addr_hash(pi);
 		}
 	}
 
-	return t4_set_addr_hash(pi->adapter, mb, pi->viid, uhash != 0,
-				uhash | mhash, sleep);
+	ret = t4_free_mac_filt(adap, adap->mbox, pi->viid, 1, maclist, false);
+	return ret < 0 ? -EINVAL : 0;
 }
 
-int dbfifo_int_thresh = 10; /* 10 == 640 entry threshold */
-module_param(dbfifo_int_thresh, int, 0644);
-MODULE_PARM_DESC(dbfifo_int_thresh, "doorbell fifo interrupt threshold");
-
-/*
- * usecs to sleep while draining the dbfifo
- */
-static int dbfifo_drain_delay = 1000;
-module_param(dbfifo_drain_delay, int, 0644);
-MODULE_PARM_DESC(dbfifo_drain_delay,
-		 "usecs to sleep while draining the dbfifo");
-
 /*
  * Set Rx properties of a port, such as promiscruity, address filters, and MTU.
  * If @mtu is -1 it is left unchanged.
  */
 static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok)
 {
-	int ret;
 	struct port_info *pi = netdev_priv(dev);
+	struct adapter *adapter = pi->adapter;
 
-	ret = set_addr_filters(dev, sleep_ok);
-	if (ret == 0)
-		ret = t4_set_rxmode(pi->adapter, pi->adapter->pf, pi->viid, mtu,
-				    (dev->flags & IFF_PROMISC) ? 1 : 0,
-				    (dev->flags & IFF_ALLMULTI) ? 1 : 0, 1, -1,
-				    sleep_ok);
-	return ret;
+	if (!(dev->flags & IFF_PROMISC)) {
+		__dev_uc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync);
+		if (!(dev->flags & IFF_ALLMULTI))
+			__dev_mc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync);
+	}
+
+	return t4_set_rxmode(adapter, adapter->mbox, pi->viid, mtu,
+			     (dev->flags & IFF_PROMISC) ? 1 : 0,
+			     (dev->flags & IFF_ALLMULTI) ? 1 : 0, 1, -1,
+			     sleep_ok);
 }
 
 /**
@@ -2677,6 +2701,8 @@ static int cxgb_up(struct adapter *adap)
 #if IS_ENABLED(CONFIG_IPV6)
 	update_clip(adap);
 #endif
+	/* Initialize hash mac addr list*/
+	INIT_LIST_HEAD(&adap->mac_hlist);
  out:
 	return err;
  irq_err:
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index 636b469..cc1736b 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -4433,23 +4433,6 @@ void t4_intr_disable(struct adapter *adapter)
 }
 
 /**
- *	hash_mac_addr - return the hash value of a MAC address
- *	@addr: the 48-bit Ethernet MAC address
- *
- *	Hashes a MAC address according to the hash function used by HW inexact
- *	(hash) address matching.
- */
-static int hash_mac_addr(const u8 *addr)
-{
-	u32 a = ((u32)addr[0] << 16) | ((u32)addr[1] << 8) | addr[2];
-	u32 b = ((u32)addr[3] << 16) | ((u32)addr[4] << 8) | addr[5];
-	a ^= b;
-	a ^= (a >> 12);
-	a ^= (a >> 6);
-	return a & 0x3f;
-}
-
-/**
  *	t4_config_rss_range - configure a portion of the RSS mapping table
  *	@adapter: the adapter
  *	@mbox: mbox to use for the FW command
@@ -6738,6 +6721,81 @@ int t4_alloc_mac_filt(struct adapter *adap, unsigned int mbox,
 }
 
 /**
+ *	t4_free_mac_filt - frees exact-match filters of given MAC addresses
+ *	@adap: the adapter
+ *	@mbox: mailbox to use for the FW command
+ *	@viid: the VI id
+ *	@naddr: the number of MAC addresses to allocate filters for (up to 7)
+ *	@addr: the MAC address(es)
+ *	@sleep_ok: call is allowed to sleep
+ *
+ *	Frees the exact-match filter for each of the supplied addresses
+ *
+ *	Returns a negative error number or the number of filters freed.
+ */
+int t4_free_mac_filt(struct adapter *adap, unsigned int mbox,
+		     unsigned int viid, unsigned int naddr,
+		     const u8 **addr, bool sleep_ok)
+{
+	int offset, ret = 0;
+	struct fw_vi_mac_cmd c;
+	unsigned int nfilters = 0;
+	unsigned int max_naddr = is_t4(adap->params.chip) ?
+				       NUM_MPS_CLS_SRAM_L_INSTANCES :
+				       NUM_MPS_T5_CLS_SRAM_L_INSTANCES;
+	unsigned int rem = naddr;
+
+	if (naddr > max_naddr)
+		return -EINVAL;
+
+	for (offset = 0; offset < (int)naddr ; /**/) {
+		unsigned int fw_naddr = (rem < ARRAY_SIZE(c.u.exact)
+					 ? rem
+					 : ARRAY_SIZE(c.u.exact));
+		size_t len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd,
+						     u.exact[fw_naddr]), 16);
+		struct fw_vi_mac_exact *p;
+		int i;
+
+		memset(&c, 0, sizeof(c));
+		c.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_CMD_EXEC_V(0) |
+				     FW_VI_MAC_CMD_VIID_V(viid));
+		c.freemacs_to_len16 =
+				cpu_to_be32(FW_VI_MAC_CMD_FREEMACS_V(0) |
+					    FW_CMD_LEN16_V(len16));
+
+		for (i = 0, p = c.u.exact; i < (int)fw_naddr; i++, p++) {
+			p->valid_to_idx = cpu_to_be16(
+				FW_VI_MAC_CMD_VALID_F |
+				FW_VI_MAC_CMD_IDX_V(FW_VI_MAC_MAC_BASED_FREE));
+			memcpy(p->macaddr, addr[offset+i], sizeof(p->macaddr));
+		}
+
+		ret = t4_wr_mbox_meat(adap, mbox, &c, sizeof(c), &c, sleep_ok);
+		if (ret)
+			break;
+
+		for (i = 0, p = c.u.exact; i < fw_naddr; i++, p++) {
+			u16 index = FW_VI_MAC_CMD_IDX_G(
+						be16_to_cpu(p->valid_to_idx));
+
+			if (index < max_naddr)
+				nfilters++;
+		}
+
+		offset += fw_naddr;
+		rem -= fw_naddr;
+	}
+
+	if (ret == 0)
+		ret = nfilters;
+	return ret;
+}
+
+/**
  *	t4_change_mac - modifies the exact-match filter for a MAC address
  *	@adap: the adapter
  *	@mbox: mailbox to use for the FW command
-- 
cgit v0.10.2


From fe5d2709b09a38ac171df1438b88769a0767b8b2 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 16 Feb 2016 10:07:10 +0530
Subject: cxgb4vf: Use __dev_uc_sync/__dev_mc_sync to sync MAC address

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
index 6049f70..4a707c3 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
@@ -348,6 +348,11 @@ struct sge {
 #define for_each_ethrxq(sge, iter) \
 	for (iter = 0; iter < (sge)->ethqsets; iter++)
 
+struct hash_mac_addr {
+	struct list_head list;
+	u8 addr[ETH_ALEN];
+};
+
 /*
  * Per-"adapter" (Virtual Function) information.
  */
@@ -381,6 +386,9 @@ struct adapter {
 
 	/* various locks */
 	spinlock_t stats_lock;
+
+	/* list of MAC addresses in MPS Hash */
+	struct list_head mac_hlist;
 };
 
 enum { /* adapter flags */
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 0cfa5d7..8337514 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -741,6 +741,9 @@ static int adapter_up(struct adapter *adapter)
 	 */
 	enable_rx(adapter);
 	t4vf_sge_start(adapter);
+
+	/* Initialize hash mac addr list*/
+	INIT_LIST_HEAD(&adapter->mac_hlist);
 	return 0;
 }
 
@@ -905,51 +908,74 @@ static inline unsigned int collect_netdev_mc_list_addrs(const struct net_device
 	return naddr;
 }
 
-/*
- * Configure the exact and hash address filters to handle a port's multicast
- * and secondary unicast MAC addresses.
- */
-static int set_addr_filters(const struct net_device *dev, bool sleep)
+static inline int cxgb4vf_set_addr_hash(struct port_info *pi)
 {
-	u64 mhash = 0;
-	u64 uhash = 0;
-	bool free = true;
-	unsigned int offset, naddr;
-	const u8 *addr[7];
-	int ret;
-	const struct port_info *pi = netdev_priv(dev);
+	struct adapter *adapter = pi->adapter;
+	u64 vec = 0;
+	bool ucast = false;
+	struct hash_mac_addr *entry;
 
-	/* first do the secondary unicast addresses */
-	for (offset = 0; ; offset += naddr) {
-		naddr = collect_netdev_uc_list_addrs(dev, addr, offset,
-						     ARRAY_SIZE(addr));
-		if (naddr == 0)
-			break;
+	/* Calculate the hash vector for the updated list and program it */
+	list_for_each_entry(entry, &adapter->mac_hlist, list) {
+		ucast |= is_unicast_ether_addr(entry->addr);
+		vec |= (1ULL << hash_mac_addr(entry->addr));
+	}
+	return t4vf_set_addr_hash(adapter, pi->viid, ucast, vec, false);
+}
 
-		ret = t4vf_alloc_mac_filt(pi->adapter, pi->viid, free,
-					  naddr, addr, NULL, &uhash, sleep);
-		if (ret < 0)
-			return ret;
+static int cxgb4vf_mac_sync(struct net_device *netdev, const u8 *mac_addr)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adapter = pi->adapter;
+	int ret;
+	u64 mhash = 0;
+	u64 uhash = 0;
+	bool free = false;
+	bool ucast = is_unicast_ether_addr(mac_addr);
+	const u8 *maclist[1] = {mac_addr};
+	struct hash_mac_addr *new_entry;
 
-		free = false;
+	ret = t4vf_alloc_mac_filt(adapter, pi->viid, free, 1, maclist,
+				  NULL, ucast ? &uhash : &mhash, false);
+	if (ret < 0)
+		goto out;
+	/* if hash != 0, then add the addr to hash addr list
+	 * so on the end we will calculate the hash for the
+	 * list and program it
+	 */
+	if (uhash || mhash) {
+		new_entry = kzalloc(sizeof(*new_entry), GFP_ATOMIC);
+		if (!new_entry)
+			return -ENOMEM;
+		ether_addr_copy(new_entry->addr, mac_addr);
+		list_add_tail(&new_entry->list, &adapter->mac_hlist);
+		ret = cxgb4vf_set_addr_hash(pi);
 	}
+out:
+	return ret < 0 ? ret : 0;
+}
 
-	/* next set up the multicast addresses */
-	for (offset = 0; ; offset += naddr) {
-		naddr = collect_netdev_mc_list_addrs(dev, addr, offset,
-						     ARRAY_SIZE(addr));
-		if (naddr == 0)
-			break;
+static int cxgb4vf_mac_unsync(struct net_device *netdev, const u8 *mac_addr)
+{
+	struct port_info *pi = netdev_priv(netdev);
+	struct adapter *adapter = pi->adapter;
+	int ret;
+	const u8 *maclist[1] = {mac_addr};
+	struct hash_mac_addr *entry, *tmp;
 
-		ret = t4vf_alloc_mac_filt(pi->adapter, pi->viid, free,
-					  naddr, addr, NULL, &mhash, sleep);
-		if (ret < 0)
-			return ret;
-		free = false;
+	/* If the MAC address to be removed is in the hash addr
+	 * list, delete it from the list and update hash vector
+	 */
+	list_for_each_entry_safe(entry, tmp, &adapter->mac_hlist, list) {
+		if (ether_addr_equal(entry->addr, mac_addr)) {
+			list_del(&entry->list);
+			kfree(entry);
+			return cxgb4vf_set_addr_hash(pi);
+		}
 	}
 
-	return t4vf_set_addr_hash(pi->adapter, pi->viid, uhash != 0,
-				  uhash | mhash, sleep);
+	ret = t4vf_free_mac_filt(adapter, pi->viid, 1, maclist, false);
+	return ret < 0 ? -EINVAL : 0;
 }
 
 /*
@@ -958,16 +984,18 @@ static int set_addr_filters(const struct net_device *dev, bool sleep)
  */
 static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok)
 {
-	int ret;
 	struct port_info *pi = netdev_priv(dev);
 
-	ret = set_addr_filters(dev, sleep_ok);
-	if (ret == 0)
-		ret = t4vf_set_rxmode(pi->adapter, pi->viid, -1,
-				      (dev->flags & IFF_PROMISC) != 0,
-				      (dev->flags & IFF_ALLMULTI) != 0,
-				      1, -1, sleep_ok);
-	return ret;
+	if (!(dev->flags & IFF_PROMISC)) {
+		__dev_uc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync);
+		if (!(dev->flags & IFF_ALLMULTI))
+			__dev_mc_sync(dev, cxgb4vf_mac_sync,
+				      cxgb4vf_mac_unsync);
+	}
+	return t4vf_set_rxmode(pi->adapter, pi->viid, -1,
+			       (dev->flags & IFF_PROMISC) != 0,
+			       (dev->flags & IFF_ALLMULTI) != 0,
+			       1, -1, sleep_ok);
 }
 
 /*
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
index 88b8981b..6ce302fe 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
@@ -285,6 +285,24 @@ static inline int is_t4(enum chip_type chip)
 	return CHELSIO_CHIP_VERSION(chip) == CHELSIO_T4;
 }
 
+/**
+ *	hash_mac_addr - return the hash value of a MAC address
+ *	@addr: the 48-bit Ethernet MAC address
+ *
+ *	Hashes a MAC address according to the hash function used by hardware
+ *	inexact (hash) address matching.
+ */
+static inline int hash_mac_addr(const u8 *addr)
+{
+	u32 a = ((u32)addr[0] << 16) | ((u32)addr[1] << 8) | addr[2];
+	u32 b = ((u32)addr[3] << 16) | ((u32)addr[4] << 8) | addr[5];
+
+	a ^= b;
+	a ^= (a >> 12);
+	a ^= (a >> 6);
+	return a & 0x3f;
+}
+
 int t4vf_wait_dev_ready(struct adapter *);
 int t4vf_port_init(struct adapter *, int);
 
@@ -320,6 +338,8 @@ int t4vf_set_rxmode(struct adapter *, unsigned int, int, int, int, int, int,
 		    bool);
 int t4vf_alloc_mac_filt(struct adapter *, unsigned int, bool, unsigned int,
 			const u8 **, u16 *, u64 *, bool);
+int t4vf_free_mac_filt(struct adapter *, unsigned int, unsigned int naddr,
+		       const u8 **, bool);
 int t4vf_change_mac(struct adapter *, unsigned int, int, const u8 *, bool);
 int t4vf_set_addr_hash(struct adapter *, unsigned int, bool, u64, bool);
 int t4vf_get_port_stats(struct adapter *, int, struct t4vf_port_stats *);
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
index b6fa74a..5422011 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
@@ -236,23 +236,6 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size,
 	return -ETIMEDOUT;
 }
 
-/**
- *	hash_mac_addr - return the hash value of a MAC address
- *	@addr: the 48-bit Ethernet MAC address
- *
- *	Hashes a MAC address according to the hash function used by hardware
- *	inexact (hash) address matching.
- */
-static int hash_mac_addr(const u8 *addr)
-{
-	u32 a = ((u32)addr[0] << 16) | ((u32)addr[1] << 8) | addr[2];
-	u32 b = ((u32)addr[3] << 16) | ((u32)addr[4] << 8) | addr[5];
-	a ^= b;
-	a ^= (a >> 12);
-	a ^= (a >> 6);
-	return a & 0x3f;
-}
-
 #define ADVERT_MASK (FW_PORT_CAP_SPEED_100M | FW_PORT_CAP_SPEED_1G |\
 		     FW_PORT_CAP_SPEED_10G | FW_PORT_CAP_SPEED_40G | \
 		     FW_PORT_CAP_SPEED_100G | FW_PORT_CAP_ANEG)
@@ -1266,6 +1249,77 @@ int t4vf_alloc_mac_filt(struct adapter *adapter, unsigned int viid, bool free,
 }
 
 /**
+ *	t4vf_free_mac_filt - frees exact-match filters of given MAC addresses
+ *	@adapter: the adapter
+ *	@viid: the VI id
+ *	@naddr: the number of MAC addresses to allocate filters for (up to 7)
+ *	@addr: the MAC address(es)
+ *	@sleep_ok: call is allowed to sleep
+ *
+ *	Frees the exact-match filter for each of the supplied addresses
+ *
+ *	Returns a negative error number or the number of filters freed.
+ */
+int t4vf_free_mac_filt(struct adapter *adapter, unsigned int viid,
+		       unsigned int naddr, const u8 **addr, bool sleep_ok)
+{
+	int offset, ret = 0;
+	struct fw_vi_mac_cmd cmd;
+	unsigned int nfilters = 0;
+	unsigned int max_naddr = adapter->params.arch.mps_tcam_size;
+	unsigned int rem = naddr;
+
+	if (naddr > max_naddr)
+		return -EINVAL;
+
+	for (offset = 0; offset < (int)naddr ; /**/) {
+		unsigned int fw_naddr = (rem < ARRAY_SIZE(cmd.u.exact) ?
+					 rem : ARRAY_SIZE(cmd.u.exact));
+		size_t len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd,
+						     u.exact[fw_naddr]), 16);
+		struct fw_vi_mac_exact *p;
+		int i;
+
+		memset(&cmd, 0, sizeof(cmd));
+		cmd.op_to_viid = cpu_to_be32(FW_CMD_OP_V(FW_VI_MAC_CMD) |
+				     FW_CMD_REQUEST_F |
+				     FW_CMD_WRITE_F |
+				     FW_CMD_EXEC_V(0) |
+				     FW_VI_MAC_CMD_VIID_V(viid));
+		cmd.freemacs_to_len16 =
+				cpu_to_be32(FW_VI_MAC_CMD_FREEMACS_V(0) |
+					    FW_CMD_LEN16_V(len16));
+
+		for (i = 0, p = cmd.u.exact; i < (int)fw_naddr; i++, p++) {
+			p->valid_to_idx = cpu_to_be16(
+				FW_VI_MAC_CMD_VALID_F |
+				FW_VI_MAC_CMD_IDX_V(FW_VI_MAC_MAC_BASED_FREE));
+			memcpy(p->macaddr, addr[offset+i], sizeof(p->macaddr));
+		}
+
+		ret = t4vf_wr_mbox_core(adapter, &cmd, sizeof(cmd), &cmd,
+					sleep_ok);
+		if (ret)
+			break;
+
+		for (i = 0, p = cmd.u.exact; i < fw_naddr; i++, p++) {
+			u16 index = FW_VI_MAC_CMD_IDX_G(
+						be16_to_cpu(p->valid_to_idx));
+
+			if (index < max_naddr)
+				nfilters++;
+		}
+
+		offset += fw_naddr;
+		rem -= fw_naddr;
+	}
+
+	if (ret == 0)
+		ret = nfilters;
+	return ret;
+}
+
+/**
  *	t4vf_change_mac - modifies the exact-match filter for a MAC address
  *	@adapter: the adapter
  *	@viid: the Virtual Interface ID
-- 
cgit v0.10.2


From 7c25b16dbbcf37ce8019bacf634ee47d852225d9 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Tue, 16 Feb 2016 10:09:51 -0500
Subject: net: bridge: log port STP state on change

Remove the shared br_log_state function and print the info directly in
br_set_state, where the net_bridge_port state is actually changed.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index 40197ff..e9c635e 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -598,7 +598,6 @@ static int br_set_port_state(struct net_bridge_port *p, u8 state)
 		return -ENETDOWN;
 
 	br_set_state(p, state);
-	br_log_state(p);
 	br_port_state_selection(p->br);
 	return 0;
 }
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 302ab0a..1b5d145 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -900,7 +900,6 @@ static inline void br_nf_core_fini(void) {}
 #endif
 
 /* br_stp.c */
-void br_log_state(const struct net_bridge_port *p);
 void br_set_state(struct net_bridge_port *p, unsigned int state);
 struct net_bridge_port *br_get_port(struct net_bridge *br, u16 port_no);
 void br_init_port(struct net_bridge_port *p);
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index b3cca12..c22816a 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -30,13 +30,6 @@ static const char *const br_port_state_names[] = {
 	[BR_STATE_BLOCKING] = "blocking",
 };
 
-void br_log_state(const struct net_bridge_port *p)
-{
-	br_info(p->br, "port %u(%s) entered %s state\n",
-		(unsigned int) p->port_no, p->dev->name,
-		br_port_state_names[p->state]);
-}
-
 void br_set_state(struct net_bridge_port *p, unsigned int state)
 {
 	struct switchdev_attr attr = {
@@ -52,6 +45,10 @@ void br_set_state(struct net_bridge_port *p, unsigned int state)
 	if (err && err != -EOPNOTSUPP)
 		br_warn(p->br, "error setting offload STP state on port %u(%s)\n",
 				(unsigned int) p->port_no, p->dev->name);
+	else
+		br_info(p->br, "port %u(%s) entered %s state\n",
+				(unsigned int) p->port_no, p->dev->name,
+				br_port_state_names[p->state]);
 }
 
 /* called under bridge lock */
@@ -126,7 +123,6 @@ static void br_root_port_block(const struct net_bridge *br,
 		  (unsigned int) p->port_no, p->dev->name);
 
 	br_set_state(p, BR_STATE_LISTENING);
-	br_log_state(p);
 	br_ifinfo_notify(RTM_NEWLINK, p);
 
 	if (br->forward_delay > 0)
@@ -407,7 +403,6 @@ static void br_make_blocking(struct net_bridge_port *p)
 			br_topology_change_detection(p->br);
 
 		br_set_state(p, BR_STATE_BLOCKING);
-		br_log_state(p);
 		br_ifinfo_notify(RTM_NEWLINK, p);
 
 		del_timer(&p->forward_delay_timer);
@@ -431,7 +426,6 @@ static void br_make_forwarding(struct net_bridge_port *p)
 	else
 		br_set_state(p, BR_STATE_LEARNING);
 
-	br_log_state(p);
 	br_ifinfo_notify(RTM_NEWLINK, p);
 
 	if (br->forward_delay != 0)
diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c
index a31ac6a..984d462 100644
--- a/net/bridge/br_stp_if.c
+++ b/net/bridge/br_stp_if.c
@@ -102,7 +102,6 @@ void br_stp_enable_port(struct net_bridge_port *p)
 {
 	br_init_port(p);
 	br_port_state_selection(p->br);
-	br_log_state(p);
 	br_ifinfo_notify(RTM_NEWLINK, p);
 }
 
@@ -118,7 +117,6 @@ void br_stp_disable_port(struct net_bridge_port *p)
 	p->topology_change_ack = 0;
 	p->config_pending = 0;
 
-	br_log_state(p);
 	br_ifinfo_notify(RTM_NEWLINK, p);
 
 	del_timer(&p->message_age_timer);
diff --git a/net/bridge/br_stp_timer.c b/net/bridge/br_stp_timer.c
index 5f0f5af..da058b8 100644
--- a/net/bridge/br_stp_timer.c
+++ b/net/bridge/br_stp_timer.c
@@ -98,7 +98,6 @@ static void br_forward_delay_timer_expired(unsigned long arg)
 			br_topology_change_detection(br);
 		netif_carrier_on(br->dev);
 	}
-	br_log_state(p);
 	rcu_read_lock();
 	br_ifinfo_notify(RTM_NEWLINK, p);
 	rcu_read_unlock();
-- 
cgit v0.10.2


From 1e9f12ec92ab7307ac7386924e343905f7f12205 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Thu, 18 Feb 2016 11:22:49 +0100
Subject: geneve: implement geneve_get_sk_family helper

Similarly to the existing vxlan_get_sk_family.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 6f20813..f09de1e 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -110,6 +110,11 @@ static __be64 vni_to_tunnel_id(const __u8 *vni)
 #endif
 }
 
+static sa_family_t geneve_get_sk_family(struct geneve_sock *gs)
+{
+	return gs->sock->sk->sk_family;
+}
+
 static struct geneve_dev *geneve_lookup(struct geneve_sock *gs,
 					__be32 addr, u8 vni[])
 {
@@ -165,16 +170,13 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb)
 	static u8 zero_vni[3];
 	u8 *vni;
 	int err = 0;
-	sa_family_t sa_family;
 #if IS_ENABLED(CONFIG_IPV6)
 	struct ipv6hdr *ip6h = NULL;
 	struct in6_addr addr6;
 	static struct in6_addr zero_addr6;
 #endif
 
-	sa_family = gs->sock->sk->sk_family;
-
-	if (sa_family == AF_INET) {
+	if (geneve_get_sk_family(gs) == AF_INET) {
 		iph = ip_hdr(skb); /* outer IP header... */
 
 		if (gs->collect_md) {
@@ -188,7 +190,7 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb)
 
 		geneve = geneve_lookup(gs, addr, vni);
 #if IS_ENABLED(CONFIG_IPV6)
-	} else if (sa_family == AF_INET6) {
+	} else if (geneve_get_sk_family(gs) == AF_INET6) {
 		ip6h = ipv6_hdr(skb); /* outer IPv6 header... */
 
 		if (gs->collect_md) {
@@ -213,7 +215,7 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb)
 			(gnvh->oam ? TUNNEL_OAM : 0) |
 			(gnvh->critical ? TUNNEL_CRIT_OPT : 0);
 
-		tun_dst = udp_tun_rx_dst(skb, sa_family, flags,
+		tun_dst = udp_tun_rx_dst(skb, geneve_get_sk_family(gs), flags,
 					 vni_to_tunnel_id(gnvh->vni),
 					 gnvh->opt_len * 4);
 		if (!tun_dst)
@@ -392,7 +394,7 @@ static void geneve_notify_add_rx_port(struct geneve_sock *gs)
 	struct net_device *dev;
 	struct sock *sk = gs->sock->sk;
 	struct net *net = sock_net(sk);
-	sa_family_t sa_family = sk->sk_family;
+	sa_family_t sa_family = geneve_get_sk_family(gs);
 	__be16 port = inet_sk(sk)->inet_sport;
 	int err;
 
@@ -553,7 +555,7 @@ static void geneve_notify_del_rx_port(struct geneve_sock *gs)
 	struct net_device *dev;
 	struct sock *sk = gs->sock->sk;
 	struct net *net = sock_net(sk);
-	sa_family_t sa_family = sk->sk_family;
+	sa_family_t sa_family = geneve_get_sk_family(gs);
 	__be16 port = inet_sk(sk)->inet_sport;
 
 	rcu_read_lock();
@@ -596,7 +598,7 @@ static struct geneve_sock *geneve_find_sock(struct geneve_net *gn,
 
 	list_for_each_entry(gs, &gn->sock_list, list) {
 		if (inet_sk(gs->sock->sk)->inet_sport == dst_port &&
-		    inet_sk(gs->sock->sk)->sk.sk_family == family) {
+		    geneve_get_sk_family(gs) == family) {
 			return gs;
 		}
 	}
-- 
cgit v0.10.2


From 9fc4754582bf46b0998a64fce74f570cea720e18 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Thu, 18 Feb 2016 11:22:50 +0100
Subject: geneve: move geneve device lookup before iptunnel_pull_header

This is in preparation for iptunnel_pull_header calling skb_scrub_packet.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index f09de1e..4ceccf8 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -158,55 +158,60 @@ static inline struct genevehdr *geneve_hdr(const struct sk_buff *skb)
 	return (struct genevehdr *)(udp_hdr(skb) + 1);
 }
 
-/* geneve receive/decap routine */
-static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb)
+static struct geneve_dev *geneve_lookup_skb(struct geneve_sock *gs,
+					    struct sk_buff *skb)
 {
-	struct genevehdr *gnvh = geneve_hdr(skb);
-	struct metadata_dst *tun_dst = NULL;
-	struct geneve_dev *geneve = NULL;
-	struct pcpu_sw_netstats *stats;
-	struct iphdr *iph = NULL;
+	u8 *vni;
 	__be32 addr;
 	static u8 zero_vni[3];
-	u8 *vni;
-	int err = 0;
 #if IS_ENABLED(CONFIG_IPV6)
-	struct ipv6hdr *ip6h = NULL;
-	struct in6_addr addr6;
 	static struct in6_addr zero_addr6;
 #endif
 
 	if (geneve_get_sk_family(gs) == AF_INET) {
+		struct iphdr *iph;
+
 		iph = ip_hdr(skb); /* outer IP header... */
 
 		if (gs->collect_md) {
 			vni = zero_vni;
 			addr = 0;
 		} else {
-			vni = gnvh->vni;
-
+			vni = geneve_hdr(skb)->vni;
 			addr = iph->saddr;
 		}
 
-		geneve = geneve_lookup(gs, addr, vni);
+		return geneve_lookup(gs, addr, vni);
 #if IS_ENABLED(CONFIG_IPV6)
 	} else if (geneve_get_sk_family(gs) == AF_INET6) {
+		struct ipv6hdr *ip6h;
+		struct in6_addr addr6;
+
 		ip6h = ipv6_hdr(skb); /* outer IPv6 header... */
 
 		if (gs->collect_md) {
 			vni = zero_vni;
 			addr6 = zero_addr6;
 		} else {
-			vni = gnvh->vni;
-
+			vni = geneve_hdr(skb)->vni;
 			addr6 = ip6h->saddr;
 		}
 
-		geneve = geneve6_lookup(gs, addr6, vni);
+		return geneve6_lookup(gs, addr6, vni);
 #endif
 	}
-	if (!geneve)
-		goto drop;
+	return NULL;
+}
+
+/* geneve receive/decap routine */
+static void geneve_rx(struct geneve_dev *geneve, struct geneve_sock *gs,
+		      struct sk_buff *skb)
+{
+	struct genevehdr *gnvh = geneve_hdr(skb);
+	struct metadata_dst *tun_dst = NULL;
+	struct pcpu_sw_netstats *stats;
+	int err = 0;
+	void *oiph;
 
 	if (ip_tunnel_collect_metadata() || gs->collect_md) {
 		__be16 flags;
@@ -243,25 +248,27 @@ static void geneve_rx(struct geneve_sock *gs, struct sk_buff *skb)
 	if (ether_addr_equal(eth_hdr(skb)->h_source, geneve->dev->dev_addr))
 		goto drop;
 
+	oiph = skb_network_header(skb);
 	skb_reset_network_header(skb);
 
-	if (iph)
-		err = IP_ECN_decapsulate(iph, skb);
+	if (geneve_get_sk_family(gs) == AF_INET)
+		err = IP_ECN_decapsulate(oiph, skb);
 #if IS_ENABLED(CONFIG_IPV6)
-	if (ip6h)
-		err = IP6_ECN_decapsulate(ip6h, skb);
+	else
+		err = IP6_ECN_decapsulate(oiph, skb);
 #endif
 
 	if (unlikely(err)) {
 		if (log_ecn_error) {
-			if (iph)
+			if (geneve_get_sk_family(gs) == AF_INET)
 				net_info_ratelimited("non-ECT from %pI4 "
 						     "with TOS=%#x\n",
-						     &iph->saddr, iph->tos);
+						     &((struct iphdr *)oiph)->saddr,
+						     ((struct iphdr *)oiph)->tos);
 #if IS_ENABLED(CONFIG_IPV6)
-			if (ip6h)
+			else
 				net_info_ratelimited("non-ECT from %pI6\n",
-						     &ip6h->saddr);
+						     &((struct ipv6hdr *)oiph)->saddr);
 #endif
 		}
 		if (err > 1) {
@@ -323,6 +330,7 @@ static void geneve_uninit(struct net_device *dev)
 static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct genevehdr *geneveh;
+	struct geneve_dev *geneve;
 	struct geneve_sock *gs;
 	int opts_len;
 
@@ -338,16 +346,20 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (unlikely(geneveh->proto_type != htons(ETH_P_TEB)))
 		goto error;
 
+	gs = rcu_dereference_sk_user_data(sk);
+	if (!gs)
+		goto drop;
+
+	geneve = geneve_lookup_skb(gs, skb);
+	if (!geneve)
+		goto drop;
+
 	opts_len = geneveh->opt_len * 4;
 	if (iptunnel_pull_header(skb, GENEVE_BASE_HLEN + opts_len,
 				 htons(ETH_P_TEB)))
 		goto drop;
 
-	gs = rcu_dereference_sk_user_data(sk);
-	if (!gs)
-		goto drop;
-
-	geneve_rx(gs, skb);
+	geneve_rx(geneve, gs, skb);
 	return 0;
 
 drop:
-- 
cgit v0.10.2


From c9e78efb6f668c42e0e0f47398e814b758b95336 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Thu, 18 Feb 2016 11:22:51 +0100
Subject: vxlan: move vxlan device lookup before iptunnel_pull_header

This is in preparation for iptunnel_pull_header calling skb_scrub_packet.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 75bccb3..16a176c 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1187,22 +1187,16 @@ out:
 	unparsed->vx_flags &= ~VXLAN_GBP_USED_BITS;
 }
 
-static void vxlan_rcv(struct vxlan_sock *vs, struct sk_buff *skb,
-		      struct vxlan_metadata *md, __be32 vni,
+static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
+		      struct sk_buff *skb, struct vxlan_metadata *md,
 		      struct metadata_dst *tun_dst)
 {
 	struct iphdr *oip = NULL;
 	struct ipv6hdr *oip6 = NULL;
-	struct vxlan_dev *vxlan;
 	struct pcpu_sw_netstats *stats;
 	union vxlan_addr saddr;
 	int err = 0;
 
-	/* Is this VNI defined? */
-	vxlan = vxlan_vs_find_vni(vs, vni);
-	if (!vxlan)
-		goto drop;
-
 	skb_reset_mac_header(skb);
 	skb_scrub_packet(skb, !net_eq(vxlan->net, dev_net(vxlan->dev)));
 	skb->protocol = eth_type_trans(skb, vxlan->dev);
@@ -1281,6 +1275,7 @@ drop:
 static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct metadata_dst *tun_dst = NULL;
+	struct vxlan_dev *vxlan;
 	struct vxlan_sock *vs;
 	struct vxlanhdr unparsed;
 	struct vxlan_metadata _md;
@@ -1302,13 +1297,17 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	unparsed.vx_flags &= ~VXLAN_HF_VNI;
 	unparsed.vx_vni &= ~VXLAN_VNI_MASK;
 
-	if (iptunnel_pull_header(skb, VXLAN_HLEN, htons(ETH_P_TEB)))
-		goto drop;
-
 	vs = rcu_dereference_sk_user_data(sk);
 	if (!vs)
 		goto drop;
 
+	vxlan = vxlan_vs_find_vni(vs, vxlan_vni(vxlan_hdr(skb)->vx_vni));
+	if (!vxlan)
+		goto drop;
+
+	if (iptunnel_pull_header(skb, VXLAN_HLEN, htons(ETH_P_TEB)))
+		goto drop;
+
 	if (vxlan_collect_metadata(vs)) {
 		__be32 vni = vxlan_vni(vxlan_hdr(skb)->vx_vni);
 
@@ -1344,7 +1343,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto drop;
 	}
 
-	vxlan_rcv(vs, skb, md, vxlan_vni(vxlan_hdr(skb)->vx_vni), tun_dst);
+	vxlan_rcv(vxlan, vs, skb, md, tun_dst);
 	return 0;
 
 drop:
-- 
cgit v0.10.2


From 7f290c94352e59b1d720055fce760a69a63bd0a1 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Thu, 18 Feb 2016 11:22:52 +0100
Subject: iptunnel: scrub packet in iptunnel_pull_header

Part of skb_scrub_packet was open coded in iptunnel_pull_header. Let it call
skb_scrub_packet directly instead.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 4ceccf8..dfbe3ca 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -237,7 +237,6 @@ static void geneve_rx(struct geneve_dev *geneve, struct geneve_sock *gs,
 	}
 
 	skb_reset_mac_header(skb);
-	skb_scrub_packet(skb, !net_eq(geneve->net, dev_net(geneve->dev)));
 	skb->protocol = eth_type_trans(skb, geneve->dev);
 	skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
 
@@ -356,7 +355,8 @@ static int geneve_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 
 	opts_len = geneveh->opt_len * 4;
 	if (iptunnel_pull_header(skb, GENEVE_BASE_HLEN + opts_len,
-				 htons(ETH_P_TEB)))
+				 htons(ETH_P_TEB),
+				 !net_eq(geneve->net, dev_net(geneve->dev))))
 		goto drop;
 
 	geneve_rx(geneve, gs, skb);
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 16a176c..c963897 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1198,7 +1198,6 @@ static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
 	int err = 0;
 
 	skb_reset_mac_header(skb);
-	skb_scrub_packet(skb, !net_eq(vxlan->net, dev_net(vxlan->dev)));
 	skb->protocol = eth_type_trans(skb, vxlan->dev);
 	skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
 
@@ -1305,7 +1304,8 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 	if (!vxlan)
 		goto drop;
 
-	if (iptunnel_pull_header(skb, VXLAN_HLEN, htons(ETH_P_TEB)))
+	if (iptunnel_pull_header(skb, VXLAN_HLEN, htons(ETH_P_TEB),
+				 !net_eq(vxlan->net, dev_net(vxlan->dev))))
 		goto drop;
 
 	if (vxlan_collect_metadata(vs)) {
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 87408ab..4dd6163 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -270,7 +270,8 @@ static inline u8 ip_tunnel_ecn_encap(u8 tos, const struct iphdr *iph,
 	return INET_ECN_encapsulate(tos, inner);
 }
 
-int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto);
+int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto,
+			 bool xnet);
 void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
 		   __be32 src, __be32 dst, u8 proto,
 		   u8 tos, u8 ttl, __be16 df, bool xnet);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 917c2c1..12071e2 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -238,7 +238,7 @@ static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi,
 				return -EINVAL;
 		}
 	}
-	return iptunnel_pull_header(skb, hdr_len, tpi->proto);
+	return iptunnel_pull_header(skb, hdr_len, tpi->proto, false);
 }
 
 static void ipgre_err(struct sk_buff *skb, u32 info,
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index a6e58b6..eaca244 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -86,7 +86,8 @@ void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb,
 }
 EXPORT_SYMBOL_GPL(iptunnel_xmit);
 
-int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto)
+int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto,
+			 bool xnet)
 {
 	if (unlikely(!pskb_may_pull(skb, hdr_len)))
 		return -ENOMEM;
@@ -109,13 +110,10 @@ int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto)
 		skb->protocol = inner_proto;
 	}
 
-	nf_reset(skb);
-	secpath_reset(skb);
 	skb_clear_hash_if_not_l4(skb);
-	skb_dst_drop(skb);
 	skb->vlan_tci = 0;
 	skb_set_queue_mapping(skb, 0);
-	skb->pkt_type = PACKET_HOST;
+	skb_scrub_packet(skb, xnet);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(iptunnel_pull_header);
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 6ec5b42..ec51d02 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -195,7 +195,7 @@ static int ipip_rcv(struct sk_buff *skb)
 	if (tunnel) {
 		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
 			goto drop;
-		if (iptunnel_pull_header(skb, 0, tpi.proto))
+		if (iptunnel_pull_header(skb, 0, tpi.proto, false))
 			goto drop;
 		return ip_tunnel_rcv(tunnel, skb, &tpi, NULL, log_ecn_error);
 	}
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 0625ac6..f45b8ff 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -740,7 +740,7 @@ static int ipip_rcv(struct sk_buff *skb)
 
 		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
 			goto drop;
-		if (iptunnel_pull_header(skb, 0, tpi.proto))
+		if (iptunnel_pull_header(skb, 0, tpi.proto, false))
 			goto drop;
 		return ip_tunnel_rcv(tunnel, skb, &tpi, NULL, log_ecn_error);
 	}
-- 
cgit v0.10.2


From e550785c30f639b3cc6ca70c489a6463ff298453 Mon Sep 17 00:00:00 2001
From: Benjamin Poirier <bpoirier@suse.com>
Date: Wed, 17 Feb 2016 16:20:33 -0800
Subject: ipv6: Annotate change of locking mechanism for np->opt

follows up commit 45f6fad84cc3 ("ipv6: add complete rcu protection around
np->opt") which added mixed rcu/refcount protection to np->opt.

Given the current implementation of rcu_pointer_handoff(), this has no
effect at runtime.

Signed-off-by: Benjamin Poirier <bpoirier@suse.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 6570f379..f3c9857 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -259,8 +259,12 @@ static inline struct ipv6_txoptions *txopt_get(const struct ipv6_pinfo *np)
 
 	rcu_read_lock();
 	opt = rcu_dereference(np->opt);
-	if (opt && !atomic_inc_not_zero(&opt->refcnt))
-		opt = NULL;
+	if (opt) {
+		if (!atomic_inc_not_zero(&opt->refcnt))
+			opt = NULL;
+		else
+			opt = rcu_pointer_handoff(opt);
+	}
 	rcu_read_unlock();
 	return opt;
 }
-- 
cgit v0.10.2


From 9e74a6dadbbf31ac18a2712048bf866c8e32aab2 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 17 Feb 2016 11:23:55 -0800
Subject: net: Optimize local checksum offload

This patch takes advantage of several assumptions we can make about the
headers of the frame in order to reduce overall processing overhead for
computing the outer header checksum.

First we can assume the entire header is in the region pointed to by
skb->head as this is what csum_start is based on.

Second, as a result of our first assumption, we can just call csum_partial
instead of making a call to skb_checksum which would end up having to
configure things so that we could walk through the frags list.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 3920675..89b5367 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3713,19 +3713,18 @@ static inline unsigned int skb_gso_network_seglen(const struct sk_buff *skb)
  */
 static inline __wsum lco_csum(struct sk_buff *skb)
 {
-	char *inner_csum_field;
-	__wsum csum;
+	unsigned char *csum_start = skb_checksum_start(skb);
+	unsigned char *l4_hdr = skb_transport_header(skb);
+	__wsum partial;
 
 	/* Start with complement of inner checksum adjustment */
-	inner_csum_field = skb->data + skb_checksum_start_offset(skb) +
-				skb->csum_offset;
-	csum = ~csum_unfold(*(__force __sum16 *)inner_csum_field);
+	partial = ~csum_unfold(*(__force __sum16 *)(csum_start +
+						    skb->csum_offset));
+
 	/* Add in checksum of our headers (incl. outer checksum
-	 * adjustment filled in by caller)
+	 * adjustment filled in by caller) and return result.
 	 */
-	csum = skb_checksum(skb, 0, skb_checksum_start_offset(skb), csum);
-	/* The result is the checksum from skb->data to end of packet */
-	return csum;
+	return csum_partial(l4_hdr, csum_start - l4_hdr, partial);
 }
 
 #endif	/* __KERNEL__ */
-- 
cgit v0.10.2


From e12a285c9beccbdbb0346cce847dcd779096be5f Mon Sep 17 00:00:00 2001
From: "Andrew F. Davis" <afd@ti.com>
Date: Wed, 17 Feb 2016 18:10:00 -0600
Subject: net: phy: dp83848: Fix sysfs naming collision warning

Files in sysfs are created using the name from the phy_driver struct,
when two names are the same we may get a duplicate filename warning,
fix this.

Reported-by: kernel test robot <ying.huang@linux.intel.com>
Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/dp83848.c b/drivers/net/phy/dp83848.c
index 556904f..03d54c4 100644
--- a/drivers/net/phy/dp83848.c
+++ b/drivers/net/phy/dp83848.c
@@ -103,7 +103,7 @@ MODULE_DEVICE_TABLE(mdio, dp83848_tbl);
 
 static struct phy_driver dp83848_driver[] = {
 	DP83848_PHY_DRIVER(TI_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"),
-	DP83848_PHY_DRIVER(NS_DP83848C_PHY_ID, "TI DP83848C 10/100 Mbps PHY"),
+	DP83848_PHY_DRIVER(NS_DP83848C_PHY_ID, "NS DP83848C 10/100 Mbps PHY"),
 	DP83848_PHY_DRIVER(TLK10X_PHY_ID, "TI TLK10X 10/100 Mbps PHY"),
 };
 module_phy_driver(dp83848_driver);
-- 
cgit v0.10.2


From 3f9b4a6972d50562613daa649ed064244e6bc7bb Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Thu, 18 Feb 2016 17:00:39 +0200
Subject: qed: Lay infrastructure for vlan filtering offload

Today, interfaces are working in vlan-promisc mode; But once
vlan filtering offloaded would be supported, we'll need a method to
control it directly [e.g., when setting device to PROMISC, or when
running out of vlan credits].

This adds the necessary API for L2 client to manually choose whether to
accept all vlans or only those for which filters were configured.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index 978d07a..73feaf7 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -124,6 +124,8 @@ struct qed_sp_vport_update_params {
 	u8				update_vport_active_tx_flg;
 	u8				vport_active_tx_flg;
 	u8				update_approx_mcast_flg;
+	u8				update_accept_any_vlan_flg;
+	u8				accept_any_vlan;
 	unsigned long			bins[8];
 	struct qed_rss_params		*rss_params;
 	struct qed_filter_accept_flags	accept_flags;
@@ -393,7 +395,9 @@ qed_sp_vport_update(struct qed_hwfn *p_hwfn,
 	p_cmn->update_rx_active_flg = p_params->update_vport_active_rx_flg;
 	p_cmn->tx_active_flg = p_params->vport_active_tx_flg;
 	p_cmn->update_tx_active_flg = p_params->update_vport_active_tx_flg;
-
+	p_cmn->accept_any_vlan = p_params->accept_any_vlan;
+	p_cmn->update_accept_any_vlan_flg =
+			p_params->update_accept_any_vlan_flg;
 	rc = qed_sp_vport_update_rss(p_hwfn, p_ramrod, p_rss_params);
 	if (rc) {
 		/* Return spq entry which is taken in qed_sp_init_request()*/
@@ -444,8 +448,10 @@ static int qed_sp_vport_stop(struct qed_hwfn *p_hwfn,
 static int qed_filter_accept_cmd(struct qed_dev *cdev,
 				 u8 vport,
 				 struct qed_filter_accept_flags accept_flags,
-				 enum spq_mode comp_mode,
-				 struct qed_spq_comp_cb *p_comp_data)
+				 u8 update_accept_any_vlan,
+				 u8 accept_any_vlan,
+				enum spq_mode comp_mode,
+				struct qed_spq_comp_cb *p_comp_data)
 {
 	struct qed_sp_vport_update_params vport_update_params;
 	int i, rc;
@@ -454,6 +460,8 @@ static int qed_filter_accept_cmd(struct qed_dev *cdev,
 	memset(&vport_update_params, 0, sizeof(vport_update_params));
 	vport_update_params.vport_id = vport;
 	vport_update_params.accept_flags = accept_flags;
+	vport_update_params.update_accept_any_vlan_flg = update_accept_any_vlan;
+	vport_update_params.accept_any_vlan = accept_any_vlan;
 
 	for_each_hwfn(cdev, i) {
 		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
@@ -471,6 +479,10 @@ static int qed_filter_accept_cmd(struct qed_dev *cdev,
 			   "Accept filter configured, flags = [Rx]%x [Tx]%x\n",
 			   accept_flags.rx_accept_filter,
 			   accept_flags.tx_accept_filter);
+		if (update_accept_any_vlan)
+			DP_VERBOSE(p_hwfn, QED_MSG_SP,
+				   "accept_any_vlan=%d configured\n",
+				   accept_any_vlan);
 	}
 
 	return 0;
@@ -1347,6 +1359,9 @@ static int qed_update_vport(struct qed_dev *cdev,
 		params->update_vport_active_flg;
 	sp_params.vport_active_rx_flg = params->vport_active_flg;
 	sp_params.vport_active_tx_flg = params->vport_active_flg;
+	sp_params.accept_any_vlan = params->accept_any_vlan;
+	sp_params.update_accept_any_vlan_flg =
+		params->update_accept_any_vlan_flg;
 
 	/* RSS - is a bit tricky, since upper-layer isn't familiar with hwfns.
 	 * We need to re-fix the rss values per engine for CMT.
@@ -1566,7 +1581,7 @@ static int qed_configure_filter_rx_mode(struct qed_dev *cdev,
 	else if (type == QED_FILTER_RX_MODE_TYPE_MULTI_PROMISC)
 		accept_flags.rx_accept_filter |= QED_ACCEPT_MCAST_UNMATCHED;
 
-	return qed_filter_accept_cmd(cdev, 0, accept_flags,
+	return qed_filter_accept_cmd(cdev, 0, accept_flags, false, false,
 				     QED_SPQ_MODE_CB, NULL);
 }
 
diff --git a/include/linux/qed/qed_eth_if.h b/include/linux/qed/qed_eth_if.h
index 81ab178..e53b0ca 100644
--- a/include/linux/qed/qed_eth_if.h
+++ b/include/linux/qed/qed_eth_if.h
@@ -33,6 +33,8 @@ struct qed_update_vport_params {
 	u8 vport_id;
 	u8 update_vport_active_flg;
 	u8 vport_active_flg;
+	u8 update_accept_any_vlan_flg;
+	u8 accept_any_vlan;
 	u8 update_rss_flg;
 	struct qed_update_vport_rss_params rss_params;
 };
-- 
cgit v0.10.2


From 7c1bfcad9f3c8a7400c2dfaf67edc7a46e271c51 Mon Sep 17 00:00:00 2001
From: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Date: Thu, 18 Feb 2016 17:00:40 +0200
Subject: qede: Add vlan filtering offload support

Device would start receiving only vlan-tagged traffic with tags matching
that of one of the configured vlan IDs, unless:
  - Device is expliicly placed in PROMISC mode.
  - Device exhausts its vlan filter credits.

Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index f75d9e0..15c5528 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -100,6 +100,12 @@ struct qede_stats {
 	u64 tx_mac_ctrl_frames;
 };
 
+struct qede_vlan {
+	struct list_head list;
+	u16 vid;
+	bool configured;
+};
+
 struct qede_dev {
 	struct qed_dev			*cdev;
 	struct net_device		*ndev;
@@ -154,6 +160,10 @@ struct qede_dev {
 	u16			q_num_rx_buffers; /* Must be a power of two */
 	u16			q_num_tx_buffers; /* Must be a power of two */
 
+	struct list_head vlan_list;
+	u16 configured_vlans;
+	u16 non_configured_vlans;
+	bool accept_any_vlan;
 	struct delayed_work		sp_task;
 	unsigned long			sp_flags;
 };
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index f50e0bd..5f15e23 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -1140,6 +1140,21 @@ static int qede_set_ucast_rx_mac(struct qede_dev *edev,
 	return edev->ops->filter_config(edev->cdev, &filter_cmd);
 }
 
+static int qede_set_ucast_rx_vlan(struct qede_dev *edev,
+				  enum qed_filter_xcast_params_type opcode,
+				  u16 vid)
+{
+	struct qed_filter_params filter_cmd;
+
+	memset(&filter_cmd, 0, sizeof(filter_cmd));
+	filter_cmd.type = QED_FILTER_TYPE_UCAST;
+	filter_cmd.filter.ucast.type = opcode;
+	filter_cmd.filter.ucast.vlan_valid = 1;
+	filter_cmd.filter.ucast.vlan = vid;
+
+	return edev->ops->filter_config(edev->cdev, &filter_cmd);
+}
+
 void qede_fill_by_demand_stats(struct qede_dev *edev)
 {
 	struct qed_eth_stats stats;
@@ -1252,6 +1267,247 @@ static struct rtnl_link_stats64 *qede_get_stats64(
 	return stats;
 }
 
+static void qede_config_accept_any_vlan(struct qede_dev *edev, bool action)
+{
+	struct qed_update_vport_params params;
+	int rc;
+
+	/* Proceed only if action actually needs to be performed */
+	if (edev->accept_any_vlan == action)
+		return;
+
+	memset(&params, 0, sizeof(params));
+
+	params.vport_id = 0;
+	params.accept_any_vlan = action;
+	params.update_accept_any_vlan_flg = 1;
+
+	rc = edev->ops->vport_update(edev->cdev, &params);
+	if (rc) {
+		DP_ERR(edev, "Failed to %s accept-any-vlan\n",
+		       action ? "enable" : "disable");
+	} else {
+		DP_INFO(edev, "%s accept-any-vlan\n",
+			action ? "enabled" : "disabled");
+		edev->accept_any_vlan = action;
+	}
+}
+
+static int qede_vlan_rx_add_vid(struct net_device *dev, __be16 proto, u16 vid)
+{
+	struct qede_dev *edev = netdev_priv(dev);
+	struct qede_vlan *vlan, *tmp;
+	int rc;
+
+	DP_VERBOSE(edev, NETIF_MSG_IFUP, "Adding vlan 0x%04x\n", vid);
+
+	vlan = kzalloc(sizeof(*vlan), GFP_KERNEL);
+	if (!vlan) {
+		DP_INFO(edev, "Failed to allocate struct for vlan\n");
+		return -ENOMEM;
+	}
+	INIT_LIST_HEAD(&vlan->list);
+	vlan->vid = vid;
+	vlan->configured = false;
+
+	/* Verify vlan isn't already configured */
+	list_for_each_entry(tmp, &edev->vlan_list, list) {
+		if (tmp->vid == vlan->vid) {
+			DP_VERBOSE(edev, (NETIF_MSG_IFUP | NETIF_MSG_IFDOWN),
+				   "vlan already configured\n");
+			kfree(vlan);
+			return -EEXIST;
+		}
+	}
+
+	/* If interface is down, cache this VLAN ID and return */
+	if (edev->state != QEDE_STATE_OPEN) {
+		DP_VERBOSE(edev, NETIF_MSG_IFDOWN,
+			   "Interface is down, VLAN %d will be configured when interface is up\n",
+			   vid);
+		if (vid != 0)
+			edev->non_configured_vlans++;
+		list_add(&vlan->list, &edev->vlan_list);
+
+		return 0;
+	}
+
+	/* Check for the filter limit.
+	 * Note - vlan0 has a reserved filter and can be added without
+	 * worrying about quota
+	 */
+	if ((edev->configured_vlans < edev->dev_info.num_vlan_filters) ||
+	    (vlan->vid == 0)) {
+		rc = qede_set_ucast_rx_vlan(edev,
+					    QED_FILTER_XCAST_TYPE_ADD,
+					    vlan->vid);
+		if (rc) {
+			DP_ERR(edev, "Failed to configure VLAN %d\n",
+			       vlan->vid);
+			kfree(vlan);
+			return -EINVAL;
+		}
+		vlan->configured = true;
+
+		/* vlan0 filter isn't consuming out of our quota */
+		if (vlan->vid != 0)
+			edev->configured_vlans++;
+	} else {
+		/* Out of quota; Activate accept-any-VLAN mode */
+		if (!edev->non_configured_vlans)
+			qede_config_accept_any_vlan(edev, true);
+
+		edev->non_configured_vlans++;
+	}
+
+	list_add(&vlan->list, &edev->vlan_list);
+
+	return 0;
+}
+
+static void qede_del_vlan_from_list(struct qede_dev *edev,
+				    struct qede_vlan *vlan)
+{
+	/* vlan0 filter isn't consuming out of our quota */
+	if (vlan->vid != 0) {
+		if (vlan->configured)
+			edev->configured_vlans--;
+		else
+			edev->non_configured_vlans--;
+	}
+
+	list_del(&vlan->list);
+	kfree(vlan);
+}
+
+static int qede_configure_vlan_filters(struct qede_dev *edev)
+{
+	int rc = 0, real_rc = 0, accept_any_vlan = 0;
+	struct qed_dev_eth_info *dev_info;
+	struct qede_vlan *vlan = NULL;
+
+	if (list_empty(&edev->vlan_list))
+		return 0;
+
+	dev_info = &edev->dev_info;
+
+	/* Configure non-configured vlans */
+	list_for_each_entry(vlan, &edev->vlan_list, list) {
+		if (vlan->configured)
+			continue;
+
+		/* We have used all our credits, now enable accept_any_vlan */
+		if ((vlan->vid != 0) &&
+		    (edev->configured_vlans == dev_info->num_vlan_filters)) {
+			accept_any_vlan = 1;
+			continue;
+		}
+
+		DP_VERBOSE(edev, NETIF_MSG_IFUP, "Adding vlan %d\n", vlan->vid);
+
+		rc = qede_set_ucast_rx_vlan(edev, QED_FILTER_XCAST_TYPE_ADD,
+					    vlan->vid);
+		if (rc) {
+			DP_ERR(edev, "Failed to configure VLAN %u\n",
+			       vlan->vid);
+			real_rc = rc;
+			continue;
+		}
+
+		vlan->configured = true;
+		/* vlan0 filter doesn't consume our VLAN filter's quota */
+		if (vlan->vid != 0) {
+			edev->non_configured_vlans--;
+			edev->configured_vlans++;
+		}
+	}
+
+	/* enable accept_any_vlan mode if we have more VLANs than credits,
+	 * or remove accept_any_vlan mode if we've actually removed
+	 * a non-configured vlan, and all remaining vlans are truly configured.
+	 */
+
+	if (accept_any_vlan)
+		qede_config_accept_any_vlan(edev, true);
+	else if (!edev->non_configured_vlans)
+		qede_config_accept_any_vlan(edev, false);
+
+	return real_rc;
+}
+
+static int qede_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid)
+{
+	struct qede_dev *edev = netdev_priv(dev);
+	struct qede_vlan *vlan = NULL;
+	int rc;
+
+	DP_VERBOSE(edev, NETIF_MSG_IFDOWN, "Removing vlan 0x%04x\n", vid);
+
+	/* Find whether entry exists */
+	list_for_each_entry(vlan, &edev->vlan_list, list)
+		if (vlan->vid == vid)
+			break;
+
+	if (!vlan || (vlan->vid != vid)) {
+		DP_VERBOSE(edev, (NETIF_MSG_IFUP | NETIF_MSG_IFDOWN),
+			   "Vlan isn't configured\n");
+		return 0;
+	}
+
+	if (edev->state != QEDE_STATE_OPEN) {
+		/* As interface is already down, we don't have a VPORT
+		 * instance to remove vlan filter. So just update vlan list
+		 */
+		DP_VERBOSE(edev, NETIF_MSG_IFDOWN,
+			   "Interface is down, removing VLAN from list only\n");
+		qede_del_vlan_from_list(edev, vlan);
+		return 0;
+	}
+
+	/* Remove vlan */
+	rc = qede_set_ucast_rx_vlan(edev, QED_FILTER_XCAST_TYPE_DEL, vid);
+	if (rc) {
+		DP_ERR(edev, "Failed to remove VLAN %d\n", vid);
+		return -EINVAL;
+	}
+
+	qede_del_vlan_from_list(edev, vlan);
+
+	/* We have removed a VLAN - try to see if we can
+	 * configure non-configured VLAN from the list.
+	 */
+	rc = qede_configure_vlan_filters(edev);
+
+	return rc;
+}
+
+static void qede_vlan_mark_nonconfigured(struct qede_dev *edev)
+{
+	struct qede_vlan *vlan = NULL;
+
+	if (list_empty(&edev->vlan_list))
+		return;
+
+	list_for_each_entry(vlan, &edev->vlan_list, list) {
+		if (!vlan->configured)
+			continue;
+
+		vlan->configured = false;
+
+		/* vlan0 filter isn't consuming out of our quota */
+		if (vlan->vid != 0) {
+			edev->non_configured_vlans++;
+			edev->configured_vlans--;
+		}
+
+		DP_VERBOSE(edev, NETIF_MSG_IFDOWN,
+			   "marked vlan %d as non-configured\n",
+			   vlan->vid);
+	}
+
+	edev->accept_any_vlan = false;
+}
+
 static const struct net_device_ops qede_netdev_ops = {
 	.ndo_open = qede_open,
 	.ndo_stop = qede_close,
@@ -1260,6 +1516,8 @@ static const struct net_device_ops qede_netdev_ops = {
 	.ndo_set_mac_address = qede_set_mac_addr,
 	.ndo_validate_addr = eth_validate_addr,
 	.ndo_change_mtu = qede_change_mtu,
+	.ndo_vlan_rx_add_vid = qede_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = qede_vlan_rx_kill_vid,
 	.ndo_get_stats64 = qede_get_stats64,
 };
 
@@ -1304,6 +1562,8 @@ static struct qede_dev *qede_alloc_etherdev(struct qed_dev *cdev,
 
 	edev->num_tc = edev->dev_info.num_tc;
 
+	INIT_LIST_HEAD(&edev->vlan_list);
+
 	return edev;
 }
 
@@ -1335,7 +1595,7 @@ static void qede_init_ndev(struct qede_dev *edev)
 			      NETIF_F_HIGHDMA;
 	ndev->features = hw_features | NETIF_F_RXHASH | NETIF_F_RXCSUM |
 			 NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HIGHDMA |
-			 NETIF_F_HW_VLAN_CTAG_TX;
+			 NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_CTAG_TX;
 
 	ndev->hw_features = hw_features;
 
@@ -2342,6 +2602,7 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode)
 
 	DP_INFO(edev, "Stopped Queues\n");
 
+	qede_vlan_mark_nonconfigured(edev);
 	edev->ops->fastpath_stop(edev->cdev);
 
 	/* Release the interrupts */
@@ -2410,6 +2671,9 @@ static int qede_load(struct qede_dev *edev, enum qede_load_mode mode)
 	edev->state = QEDE_STATE_OPEN;
 	mutex_unlock(&edev->qede_lock);
 
+	/* Program un-configured VLANs */
+	qede_configure_vlan_filters(edev);
+
 	/* Ask for link-up using current configuration */
 	memset(&link_params, 0, sizeof(link_params));
 	link_params.link_up = true;
@@ -2670,6 +2934,17 @@ static void qede_config_rx_mode(struct net_device *ndev)
 			goto out;
 	}
 
+	/* take care of VLAN mode */
+	if (ndev->flags & IFF_PROMISC) {
+		qede_config_accept_any_vlan(edev, true);
+	} else if (!edev->non_configured_vlans) {
+		/* It's possible that accept_any_vlan mode is set due to a
+		 * previous setting of IFF_PROMISC. If vlan credits are
+		 * sufficient, disable accept_any_vlan.
+		 */
+		qede_config_accept_any_vlan(edev, false);
+	}
+
 	rx_mode.filter.accept_flags = accept_flags;
 	edev->ops->filter_config(edev->cdev, &rx_mode);
 out:
-- 
cgit v0.10.2


From fad57330b6d0710fdf39dc1c2b28ccebb97ae8a1 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:22 -0800
Subject: i40e/i40evf: Clean-up Rx packet checksum handling

This is mostly a minor clean-up for the Rx checksum path in order to avoid
some of the unnecessary conditional checks that were being applied.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 6b08b0f..ded73c0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1389,13 +1389,7 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    u16 rx_ptype)
 {
 	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
-	bool ipv4 = false, ipv6 = false;
-	bool ipv4_tunnel, ipv6_tunnel;
-
-	ipv4_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
-		     (rx_ptype <= I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
-	ipv6_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT6_MAC_PAY3) &&
-		     (rx_ptype <= I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4);
+	bool ipv4, ipv6, ipv4_tunnel, ipv6_tunnel;
 
 	skb->ip_summed = CHECKSUM_NONE;
 
@@ -1411,12 +1405,10 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	if (!(decoded.known && decoded.outer_ip))
 		return;
 
-	if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
-	    decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4)
-		ipv4 = true;
-	else if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
-		 decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6)
-		ipv6 = true;
+	ipv4 = (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP) &&
+	       (decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4);
+	ipv6 = (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP) &&
+	       (decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6);
 
 	if (ipv4 &&
 	    (rx_error & (BIT(I40E_RX_DESC_ERROR_IPE_SHIFT) |
@@ -1447,6 +1439,11 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	 * inner checksum report CHECKSUM_UNNECESSARY.
 	 */
 
+	ipv4_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
+		     (rx_ptype <= I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
+	ipv6_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT6_MAC_PAY3) &&
+		     (rx_ptype <= I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4);
+
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->csum_level = ipv4_tunnel || ipv6_tunnel;
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 143c570..3f40e0e 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -861,13 +861,7 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    u16 rx_ptype)
 {
 	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
-	bool ipv4 = false, ipv6 = false;
-	bool ipv4_tunnel, ipv6_tunnel;
-
-	ipv4_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
-		     (rx_ptype <= I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
-	ipv6_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT6_MAC_PAY3) &&
-		     (rx_ptype <= I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4);
+	bool ipv4, ipv6, ipv4_tunnel, ipv6_tunnel;
 
 	skb->ip_summed = CHECKSUM_NONE;
 
@@ -883,12 +877,10 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	if (!(decoded.known && decoded.outer_ip))
 		return;
 
-	if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
-	    decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4)
-		ipv4 = true;
-	else if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
-		 decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6)
-		ipv6 = true;
+	ipv4 = (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP) &&
+	       (decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4);
+	ipv6 = (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP) &&
+	       (decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6);
 
 	if (ipv4 &&
 	    (rx_error & (BIT(I40E_RX_DESC_ERROR_IPE_SHIFT) |
@@ -919,6 +911,11 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	 * inner checksum report CHECKSUM_UNNECESSARY.
 	 */
 
+	ipv4_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
+		     (rx_ptype <= I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
+	ipv6_tunnel = (rx_ptype >= I40E_RX_PTYPE_GRENAT6_MAC_PAY3) &&
+		     (rx_ptype <= I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4);
+
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->csum_level = ipv4_tunnel || ipv6_tunnel;
 
-- 
cgit v0.10.2


From 5453205cd0975b845f6f8837f0c2b7c8cb80fcf8 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:29 -0800
Subject: i40e/i40evf: Enable support for SKB_GSO_UDP_TUNNEL_CSUM

The XL722 has support for providing the outer UDP tunnel checksum on
transmits.  Make use of this feature to support segmenting UDP tunnels with
outer checksums enabled.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index ded73c0..1955c84 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2272,6 +2272,7 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	} ip;
 	union {
 		struct tcphdr *tcp;
+		struct udphdr *udp;
 		unsigned char *hdr;
 	} l4;
 	u32 paylen, l4_offset;
@@ -2298,7 +2299,18 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		ip.v6->payload_len = 0;
 	}
 
-	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE)) {
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE |
+					 SKB_GSO_UDP_TUNNEL_CSUM)) {
+		if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM) {
+			/* determine offset of outer transport header */
+			l4_offset = l4.hdr - skb->data;
+
+			/* remove payload length from outer checksum */
+			paylen = (__force u16)l4.udp->check;
+			paylen += ntohs(1) * (u16)~(skb->len - l4_offset);
+			l4.udp->check = ~csum_fold((__force __wsum)paylen);
+		}
+
 		/* reset pointers to inner headers */
 		ip.hdr = skb_inner_network_header(skb);
 		l4.hdr = skb_inner_transport_header(skb);
@@ -2460,6 +2472,11 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		tunnel |= ((ip.hdr - l4.hdr) / 2) <<
 			  I40E_TXD_CTX_QW0_NATLEN_SHIFT;
 
+		/* indicate if we need to offload outer UDP header */
+		if ((*tx_flags & I40E_TX_FLAGS_TSO) &&
+		    (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM))
+			tunnel |= I40E_TXD_CTX_QW0_L4T_CS_MASK;
+
 		/* record tunnel offload values */
 		*cd_tunneling |= tunnel;
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 3f40e0e..6d66fcd 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1532,6 +1532,7 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	} ip;
 	union {
 		struct tcphdr *tcp;
+		struct udphdr *udp;
 		unsigned char *hdr;
 	} l4;
 	u32 paylen, l4_offset;
@@ -1558,7 +1559,18 @@ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		ip.v6->payload_len = 0;
 	}
 
-	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE)) {
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE |
+					 SKB_GSO_UDP_TUNNEL_CSUM)) {
+		if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM) {
+			/* determine offset of outer transport header */
+			l4_offset = l4.hdr - skb->data;
+
+			/* remove payload length from outer checksum */
+			paylen = (__force u16)l4.udp->check;
+			paylen += ntohs(1) * (u16)~(skb->len - l4_offset);
+			l4.udp->check = ~csum_fold((__force __wsum)paylen);
+		}
+
 		/* reset pointers to inner headers */
 		ip.hdr = skb_inner_network_header(skb);
 		l4.hdr = skb_inner_transport_header(skb);
@@ -1678,6 +1690,11 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		tunnel |= ((ip.hdr - l4.hdr) / 2) <<
 			  I40E_TXD_CTX_QW0_NATLEN_SHIFT;
 
+		/* indicate if we need to offload outer UDP header */
+		if ((*tx_flags & I40E_TX_FLAGS_TSO) &&
+		    (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM))
+			tunnel |= I40E_TXD_CTX_QW0_L4T_CS_MASK;
+
 		/* record tunnel offload values */
 		*cd_tunneling |= tunnel;
 
-- 
cgit v0.10.2


From 6b037cd465ff6e5f6b96524658f47d59d1acc554 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:36 -0800
Subject: i40e: Fix ATR in relation to tunnels

This patch contains a number of fixes to make certain that we are using
the correct protocols when parsing both the inner and outer headers of a
frame that is mixed between IPv4 and IPv6 for inner and outer.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 1955c84..159fb6e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2030,10 +2030,9 @@ tx_only:
  * @tx_ring:  ring to add programming descriptor to
  * @skb:      send buffer
  * @tx_flags: send tx flags
- * @protocol: wire protocol
  **/
 static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
-		     u32 tx_flags, __be16 protocol)
+		     u32 tx_flags)
 {
 	struct i40e_filter_program_desc *fdir_desc;
 	struct i40e_pf *pf = tx_ring->vsi->back;
@@ -2045,6 +2044,7 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	struct tcphdr *th;
 	unsigned int hlen;
 	u32 flex_ptype, dtype_cmd;
+	u8 l4_proto;
 	u16 i;
 
 	/* make sure ATR is enabled */
@@ -2058,6 +2058,7 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	if (!tx_ring->atr_sample_rate)
 		return;
 
+	/* Currently only IPv4/IPv6 with TCP is supported */
 	if (!(tx_flags & (I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6)))
 		return;
 
@@ -2065,29 +2066,22 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		/* snag network header to get L4 type and address */
 		hdr.network = skb_network_header(skb);
 
-		/* Currently only IPv4/IPv6 with TCP is supported
-		 * access ihl as u8 to avoid unaligned access on ia64
-		 */
+		/* access ihl as u8 to avoid unaligned access on ia64 */
 		if (tx_flags & I40E_TX_FLAGS_IPV4)
 			hlen = (hdr.network[0] & 0x0F) << 2;
-		else if (protocol == htons(ETH_P_IPV6))
-			hlen = sizeof(struct ipv6hdr);
 		else
-			return;
+			hlen = sizeof(struct ipv6hdr);
 	} else {
 		hdr.network = skb_inner_network_header(skb);
 		hlen = skb_inner_network_header_len(skb);
 	}
 
-	/* Currently only IPv4/IPv6 with TCP is supported
-	 * Note: tx_flags gets modified to reflect inner protocols in
+	/* Note: tx_flags gets modified to reflect inner protocols in
 	 * tx_enable_csum function if encap is enabled.
 	 */
-	if ((tx_flags & I40E_TX_FLAGS_IPV4) &&
-	    (hdr.ipv4->protocol != IPPROTO_TCP))
-		return;
-	else if ((tx_flags & I40E_TX_FLAGS_IPV6) &&
-		 (hdr.ipv6->nexthdr != IPPROTO_TCP))
+	l4_proto = (tx_flags & I40E_TX_FLAGS_IPV4) ? hdr.ipv4->protocol :
+						     hdr.ipv6->nexthdr;
+	if (l4_proto != IPPROTO_TCP)
 		return;
 
 	th = (struct tcphdr *)(hdr.network + hlen);
@@ -2124,7 +2118,7 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 
 	flex_ptype = (tx_ring->queue_index << I40E_TXD_FLTR_QW0_QINDEX_SHIFT) &
 		      I40E_TXD_FLTR_QW0_QINDEX_MASK;
-	flex_ptype |= (protocol == htons(ETH_P_IP)) ?
+	flex_ptype |= (tx_flags & I40E_TX_FLAGS_IPV4) ?
 		      (I40E_FILTER_PCTYPE_NONF_IPV4_TCP <<
 		       I40E_TXD_FLTR_QW0_PCTYPE_SHIFT) :
 		      (I40E_FILTER_PCTYPE_NONF_IPV6_TCP <<
@@ -2992,7 +2986,7 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	 *
 	 * NOTE: this must always be directly before the data descriptor.
 	 */
-	i40e_atr(tx_ring, skb, tx_flags, protocol);
+	i40e_atr(tx_ring, skb, tx_flags);
 
 	i40e_tx_map(tx_ring, skb, first, tx_flags, hdr_len,
 		    td_cmd, td_offset);
-- 
cgit v0.10.2


From 84d5946d49cf9552d0f1740ad62d0f126cb3b6a9 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:43 -0800
Subject: i40e: Do not drop support for IPv6 VXLAN or GENEVE tunnels

All of the documentation in the datasheets for the XL710 do not call out
any reason to exclude support for IPv6 based tunnels.  As such I am
dropping the code that was excluding these tunnel types from having their
port numbers recognized.  This way we can take advantage of things such as
checksum offload for inner headers over IPv6 based VXLAN or GENEVE
tunnels.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 0fa52ed..955dc71 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8626,9 +8626,6 @@ static void i40e_add_vxlan_port(struct net_device *netdev,
 	u8 next_idx;
 	u8 idx;
 
-	if (sa_family == AF_INET6)
-		return;
-
 	idx = i40e_get_udp_port_idx(pf, port);
 
 	/* Check if port already exists */
@@ -8668,9 +8665,6 @@ static void i40e_del_vxlan_port(struct net_device *netdev,
 	struct i40e_pf *pf = vsi->back;
 	u8 idx;
 
-	if (sa_family == AF_INET6)
-		return;
-
 	idx = i40e_get_udp_port_idx(pf, port);
 
 	/* Check if port already exists */
@@ -8707,9 +8701,6 @@ static void i40e_add_geneve_port(struct net_device *netdev,
 	if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE))
 		return;
 
-	if (sa_family == AF_INET6)
-		return;
-
 	idx = i40e_get_udp_port_idx(pf, port);
 
 	/* Check if port already exists */
@@ -8751,9 +8742,6 @@ static void i40e_del_geneve_port(struct net_device *netdev,
 	struct i40e_pf *pf = vsi->back;
 	u8 idx;
 
-	if (sa_family == AF_INET6)
-		return;
-
 	if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE))
 		return;
 
-- 
cgit v0.10.2


From bc5d252b363cca63b7ddc1e20dd8b8b242631006 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:50 -0800
Subject: i40e: Update feature flags to reflect newly enabled features

Recent changes should have enabled support for IPv6 based tunnels and
support for TSO with outer UDP checksums.  As such we can update the
feature flags to reflect that.

In addition we can clean-up the flags that aren't needed such as SCTP and
RXCSUM since having the bits there doesn't add any value.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 955dc71..2f2b2d7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -9032,10 +9032,14 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	np = netdev_priv(netdev);
 	np->vsi = vsi;
 
-	netdev->hw_enc_features |= NETIF_F_IP_CSUM	  |
-				   NETIF_F_GSO_UDP_TUNNEL |
-				   NETIF_F_GSO_GRE	  |
-				   NETIF_F_TSO		  |
+	netdev->hw_enc_features |= NETIF_F_IP_CSUM	       |
+				   NETIF_F_IPV6_CSUM	       |
+				   NETIF_F_TSO		       |
+				   NETIF_F_TSO6		       |
+				   NETIF_F_TSO_ECN	       |
+				   NETIF_F_GSO_GRE	       |
+				   NETIF_F_GSO_UDP_TUNNEL      |
+				   NETIF_F_GSO_UDP_TUNNEL_CSUM |
 				   0;
 
 	netdev->features = NETIF_F_SG		       |
@@ -9057,6 +9061,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 
 	if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
 		netdev->features |= NETIF_F_NTUPLE;
+	if (pf->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE)
+		netdev->features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
 
 	/* copy netdev features into list of user selectable features */
 	netdev->hw_features |= netdev->features;
-- 
cgit v0.10.2


From f608e6a60fc85e4f261daab5e7aac6225e2120d6 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Sun, 24 Jan 2016 21:17:57 -0800
Subject: i40evf: Update feature flags to reflect newly enabled features

Recent changes should have enabled support for IPv6 based tunnels and
support for TSO with outer UDP checksums.  As such we can update the
feature flags to reflect that.

In addition we can clean-up the flags that aren't needed such as SCTP and
RXCSUM since having the bits there doesn't add any value.

I also found one spot where we were setting the same flag twice.  It looks
like it was probably a git merge error that resulted in the line being
duplicated.  As such I have dropped it in this patch.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 41369a3..3396fe3 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2337,9 +2337,24 @@ int i40evf_process_config(struct i40evf_adapter *adapter)
 			    NETIF_F_IPV6_CSUM |
 			    NETIF_F_TSO |
 			    NETIF_F_TSO6 |
+			    NETIF_F_TSO_ECN |
+			    NETIF_F_GSO_GRE	       |
+			    NETIF_F_GSO_UDP_TUNNEL |
 			    NETIF_F_RXCSUM |
 			    NETIF_F_GRO;
 
+	netdev->hw_enc_features |= NETIF_F_IP_CSUM	       |
+				   NETIF_F_IPV6_CSUM	       |
+				   NETIF_F_TSO		       |
+				   NETIF_F_TSO6		       |
+				   NETIF_F_TSO_ECN	       |
+				   NETIF_F_GSO_GRE	       |
+				   NETIF_F_GSO_UDP_TUNNEL      |
+				   NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+	if (adapter->flags & I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE)
+		netdev->features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
 	/* copy netdev features into list of user selectable features */
 	netdev->hw_features |= netdev->features;
 	netdev->hw_features &= ~NETIF_F_RXCSUM;
@@ -2478,6 +2493,10 @@ static void i40evf_init_task(struct work_struct *work)
 	default:
 		goto err_alloc;
 	}
+
+	if (hw->mac.type == I40E_MAC_X722_VF)
+		adapter->flags |= I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE;
+
 	if (i40evf_process_config(adapter))
 		goto err_alloc;
 	adapter->current_op = I40E_VIRTCHNL_OP_UNKNOWN;
@@ -2519,10 +2538,6 @@ static void i40evf_init_task(struct work_struct *work)
 		goto err_sw_init;
 	i40evf_map_rings_to_vectors(adapter);
 	if (adapter->vf_res->vf_offload_flags &
-		    I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR)
-		adapter->flags |= I40EVF_FLAG_WB_ON_ITR_CAPABLE;
-
-	if (adapter->vf_res->vf_offload_flags &
 	    I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR)
 		adapter->flags |= I40EVF_FLAG_WB_ON_ITR_CAPABLE;
 
-- 
cgit v0.10.2


From ffcc55c0c2a85835a4ac080bc1053c3a277b88e2 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Mon, 25 Jan 2016 19:32:54 -0800
Subject: i40e: Add support for ATR w/ IPv6 extension headers

This patch updates the code for determining the L4 protocol and L3 header
length so that when IPv6 extension headers are being used we can determine
the offset and type of the L4 protocol.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 159fb6e..1d3afa7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2044,7 +2044,7 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	struct tcphdr *th;
 	unsigned int hlen;
 	u32 flex_ptype, dtype_cmd;
-	u8 l4_proto;
+	int l4_proto;
 	u16 i;
 
 	/* make sure ATR is enabled */
@@ -2062,25 +2062,23 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	if (!(tx_flags & (I40E_TX_FLAGS_IPV4 | I40E_TX_FLAGS_IPV6)))
 		return;
 
-	if (!(tx_flags & I40E_TX_FLAGS_UDP_TUNNEL)) {
-		/* snag network header to get L4 type and address */
-		hdr.network = skb_network_header(skb);
+	/* snag network header to get L4 type and address */
+	hdr.network = (tx_flags & I40E_TX_FLAGS_UDP_TUNNEL) ?
+		      skb_inner_network_header(skb) : skb_network_header(skb);
 
+	/* Note: tx_flags gets modified to reflect inner protocols in
+	 * tx_enable_csum function if encap is enabled.
+	 */
+	if (tx_flags & I40E_TX_FLAGS_IPV4) {
 		/* access ihl as u8 to avoid unaligned access on ia64 */
-		if (tx_flags & I40E_TX_FLAGS_IPV4)
-			hlen = (hdr.network[0] & 0x0F) << 2;
-		else
-			hlen = sizeof(struct ipv6hdr);
+		hlen = (hdr.network[0] & 0x0F) << 2;
+		l4_proto = hdr.ipv4->protocol;
 	} else {
-		hdr.network = skb_inner_network_header(skb);
-		hlen = skb_inner_network_header_len(skb);
+		hlen = hdr.network - skb->data;
+		l4_proto = ipv6_find_hdr(skb, &hlen, IPPROTO_TCP, NULL, NULL);
+		hlen -= hdr.network - skb->data;
 	}
 
-	/* Note: tx_flags gets modified to reflect inner protocols in
-	 * tx_enable_csum function if encap is enabled.
-	 */
-	l4_proto = (tx_flags & I40E_TX_FLAGS_IPV4) ? hdr.ipv4->protocol :
-						     hdr.ipv6->nexthdr;
 	if (l4_proto != IPPROTO_TCP)
 		return;
 
-- 
cgit v0.10.2


From 376471a7b6e93067cb8a0ce5e57e8bd6071eebdd Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 17 Feb 2016 13:15:14 +0200
Subject: bnx2x: Add missing HSI for big-endian machines

Commit e5d3a51cefbb ("bnx2x: extend DCBx support") was missing HSI
changes for big-endian machine, breaking compilation on such
platforms.

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
index dd9d6e6..04523d4 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
@@ -1824,10 +1824,12 @@ struct dcbx_app_priority_entry {
 	u8  pri_bitmap;
 	u8  appBitfield;
 	#define DCBX_APP_ENTRY_VALID         0x01
-	#define DCBX_APP_ENTRY_SF_MASK       0x30
+	#define DCBX_APP_ENTRY_SF_MASK       0xF0
 	#define DCBX_APP_ENTRY_SF_SHIFT      4
 	#define DCBX_APP_SF_ETH_TYPE         0x10
 	#define DCBX_APP_SF_PORT             0x20
+	#define DCBX_APP_SF_UDP              0x40
+	#define DCBX_APP_SF_DEFAULT          0x80
 #elif defined(__LITTLE_ENDIAN)
 	u8 appBitfield;
 	#define DCBX_APP_ENTRY_VALID         0x01
-- 
cgit v0.10.2


From 4ec441df25a686518fb369086e2b34a1cedaa6c9 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 17 Feb 2016 11:02:43 -0800
Subject: i40e/i40evf: Break up xmit_descriptor_count from maybe_stop_tx

In an upcoming patch I would like to have access to the descriptor count
used for the data portion of the frame.  For this reason I am splitting up
the descriptor count function from the function that stops the ring.

Also in order to try and reduce unnecessary duplication of code I am moving
the slow-path portions of the code out of being inline calls so that we can
just jump to them and process them instead of having to build them into
each function that calls them.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
index 7c66ce41..518d72e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
@@ -1359,16 +1359,26 @@ static netdev_tx_t i40e_fcoe_xmit_frame(struct sk_buff *skb,
 	struct i40e_ring *tx_ring = vsi->tx_rings[skb->queue_mapping];
 	struct i40e_tx_buffer *first;
 	u32 tx_flags = 0;
+	int fso, count;
 	u8 hdr_len = 0;
 	u8 sof = 0;
 	u8 eof = 0;
-	int fso;
 
 	if (i40e_fcoe_set_skb_header(skb))
 		goto out_drop;
 
-	if (!i40e_xmit_descriptor_count(skb, tx_ring))
+	count = i40e_xmit_descriptor_count(skb);
+
+	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
+	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
+	 *       + 4 desc gap to avoid the cache line where head is,
+	 *       + 1 desc for context descriptor,
+	 * otherwise try next time
+	 */
+	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
+		tx_ring->tx_stats.tx_busy++;
 		return NETDEV_TX_BUSY;
+	}
 
 	/* prepare the xmit flags */
 	if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags))
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 1d3afa7..f036570 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2576,7 +2576,7 @@ static void i40e_create_tx_ctx(struct i40e_ring *tx_ring,
  *
  * Returns -EBUSY if a stop is needed, else 0
  **/
-static inline int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
+int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 {
 	netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index);
 	/* Memory barrier before checking head and tail */
@@ -2593,24 +2593,6 @@ static inline int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 }
 
 /**
- * i40e_maybe_stop_tx - 1st level check for tx stop conditions
- * @tx_ring: the ring to be checked
- * @size:    the size buffer we want to assure is available
- *
- * Returns 0 if stop is not needed
- **/
-#ifdef I40E_FCOE
-inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
-#else
-static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
-#endif
-{
-	if (likely(I40E_DESC_UNUSED(tx_ring) >= size))
-		return 0;
-	return __i40e_maybe_stop_tx(tx_ring, size);
-}
-
-/**
  * i40e_chk_linearize - Check if there are more than 8 fragments per packet
  * @skb:      send buffer
  * @tx_flags: collected send information
@@ -2870,43 +2852,6 @@ dma_error:
 }
 
 /**
- * i40e_xmit_descriptor_count - calculate number of tx descriptors needed
- * @skb:     send buffer
- * @tx_ring: ring to send buffer on
- *
- * Returns number of data descriptors needed for this skb. Returns 0 to indicate
- * there is not enough descriptors available in this ring since we need at least
- * one descriptor.
- **/
-#ifdef I40E_FCOE
-inline int i40e_xmit_descriptor_count(struct sk_buff *skb,
-				      struct i40e_ring *tx_ring)
-#else
-static inline int i40e_xmit_descriptor_count(struct sk_buff *skb,
-					     struct i40e_ring *tx_ring)
-#endif
-{
-	unsigned int f;
-	int count = 0;
-
-	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
-	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
-	 *       + 4 desc gap to avoid the cache line where head is,
-	 *       + 1 desc for context descriptor,
-	 * otherwise try next time
-	 */
-	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
-		count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
-
-	count += TXD_USE_COUNT(skb_headlen(skb));
-	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
-		tx_ring->tx_stats.tx_busy++;
-		return 0;
-	}
-	return count;
-}
-
-/**
  * i40e_xmit_frame_ring - Sends buffer on Tx ring
  * @skb:     send buffer
  * @tx_ring: ring to send buffer on
@@ -2924,14 +2869,24 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	__be16 protocol;
 	u32 td_cmd = 0;
 	u8 hdr_len = 0;
+	int tso, count;
 	int tsyn;
-	int tso;
 
 	/* prefetch the data, we'll need it later */
 	prefetch(skb->data);
 
-	if (0 == i40e_xmit_descriptor_count(skb, tx_ring))
+	count = i40e_xmit_descriptor_count(skb);
+
+	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
+	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
+	 *       + 4 desc gap to avoid the cache line where head is,
+	 *       + 1 desc for context descriptor,
+	 * otherwise try next time
+	 */
+	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
+		tx_ring->tx_stats.tx_busy++;
 		return NETDEV_TX_BUSY;
+	}
 
 	/* prepare the xmit flags */
 	if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags))
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index fde5f42..5dbc958 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -331,13 +331,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget);
 void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		 struct i40e_tx_buffer *first, u32 tx_flags,
 		 const u8 hdr_len, u32 td_cmd, u32 td_offset);
-int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size);
-int i40e_xmit_descriptor_count(struct sk_buff *skb, struct i40e_ring *tx_ring);
 int i40e_tx_prepare_vlan_flags(struct sk_buff *skb,
 			       struct i40e_ring *tx_ring, u32 *flags);
 #endif
 void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
 u32 i40e_get_tx_pending(struct i40e_ring *ring, bool in_sw);
+int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size);
 
 /**
  * i40e_get_head - Retrieve head from head writeback
@@ -352,4 +351,45 @@ static inline u32 i40e_get_head(struct i40e_ring *tx_ring)
 
 	return le32_to_cpu(*(volatile __le32 *)head);
 }
+
+/**
+ * i40e_xmit_descriptor_count - calculate number of Tx descriptors needed
+ * @skb:     send buffer
+ * @tx_ring: ring to send buffer on
+ *
+ * Returns number of data descriptors needed for this skb. Returns 0 to indicate
+ * there is not enough descriptors available in this ring since we need at least
+ * one descriptor.
+ **/
+static inline int i40e_xmit_descriptor_count(struct sk_buff *skb)
+{
+	const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
+	unsigned int nr_frags = skb_shinfo(skb)->nr_frags;
+	int count = 0, size = skb_headlen(skb);
+
+	for (;;) {
+		count += TXD_USE_COUNT(size);
+
+		if (!nr_frags--)
+			break;
+
+		size = skb_frag_size(frag++);
+	}
+
+	return count;
+}
+
+/**
+ * i40e_maybe_stop_tx - 1st level check for Tx stop conditions
+ * @tx_ring: the ring to be checked
+ * @size:    the size buffer we want to assure is available
+ *
+ * Returns 0 if stop is not needed
+ **/
+static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
+{
+	if (likely(I40E_DESC_UNUSED(tx_ring) >= size))
+		return 0;
+	return __i40e_maybe_stop_tx(tx_ring, size);
+}
 #endif /* _I40E_TXRX_H_ */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 6d66fcd..9f38127 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1858,7 +1858,7 @@ linearize_chk_done:
  *
  * Returns -EBUSY if a stop is needed, else 0
  **/
-static inline int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
+int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 {
 	netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index);
 	/* Memory barrier before checking head and tail */
@@ -1875,20 +1875,6 @@ static inline int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 }
 
 /**
- * i40evf_maybe_stop_tx - 1st level check for tx stop conditions
- * @tx_ring: the ring to be checked
- * @size:    the size buffer we want to assure is available
- *
- * Returns 0 if stop is not needed
- **/
-static inline int i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
-{
-	if (likely(I40E_DESC_UNUSED(tx_ring) >= size))
-		return 0;
-	return __i40evf_maybe_stop_tx(tx_ring, size);
-}
-
-/**
  * i40evf_tx_map - Build the Tx descriptor
  * @tx_ring:  ring to send buffer on
  * @skb:      send buffer
@@ -2003,7 +1989,7 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	netdev_tx_sent_queue(netdev_get_tx_queue(tx_ring->netdev,
 						 tx_ring->queue_index),
 						 first->bytecount);
-	i40evf_maybe_stop_tx(tx_ring, DESC_NEEDED);
+	i40e_maybe_stop_tx(tx_ring, DESC_NEEDED);
 
 	/* Algorithm to optimize tail and RS bit setting:
 	 * if xmit_more is supported
@@ -2086,38 +2072,6 @@ dma_error:
 }
 
 /**
- * i40evf_xmit_descriptor_count - calculate number of tx descriptors needed
- * @skb:     send buffer
- * @tx_ring: ring to send buffer on
- *
- * Returns number of data descriptors needed for this skb. Returns 0 to indicate
- * there is not enough descriptors available in this ring since we need at least
- * one descriptor.
- **/
-static inline int i40evf_xmit_descriptor_count(struct sk_buff *skb,
-					       struct i40e_ring *tx_ring)
-{
-	unsigned int f;
-	int count = 0;
-
-	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
-	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
-	 *       + 4 desc gap to avoid the cache line where head is,
-	 *       + 1 desc for context descriptor,
-	 * otherwise try next time
-	 */
-	for (f = 0; f < skb_shinfo(skb)->nr_frags; f++)
-		count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size);
-
-	count += TXD_USE_COUNT(skb_headlen(skb));
-	if (i40evf_maybe_stop_tx(tx_ring, count + 4 + 1)) {
-		tx_ring->tx_stats.tx_busy++;
-		return 0;
-	}
-	return count;
-}
-
-/**
  * i40e_xmit_frame_ring - Sends buffer on Tx ring
  * @skb:     send buffer
  * @tx_ring: ring to send buffer on
@@ -2135,13 +2089,23 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	__be16 protocol;
 	u32 td_cmd = 0;
 	u8 hdr_len = 0;
-	int tso;
+	int tso, count;
 
 	/* prefetch the data, we'll need it later */
 	prefetch(skb->data);
 
-	if (0 == i40evf_xmit_descriptor_count(skb, tx_ring))
+	count = i40e_xmit_descriptor_count(skb);
+
+	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
+	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
+	 *       + 4 desc gap to avoid the cache line where head is,
+	 *       + 1 desc for context descriptor,
+	 * otherwise try next time
+	 */
+	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
+		tx_ring->tx_stats.tx_busy++;
 		return NETDEV_TX_BUSY;
+	}
 
 	/* prepare the xmit flags */
 	if (i40evf_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags))
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 6ea8701..1f719d0 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -326,6 +326,7 @@ void i40evf_free_rx_resources(struct i40e_ring *rx_ring);
 int i40evf_napi_poll(struct napi_struct *napi, int budget);
 void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
 u32 i40evf_get_tx_pending(struct i40e_ring *ring, bool in_sw);
+int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size);
 
 /**
  * i40e_get_head - Retrieve head from head writeback
@@ -340,4 +341,45 @@ static inline u32 i40e_get_head(struct i40e_ring *tx_ring)
 
 	return le32_to_cpu(*(volatile __le32 *)head);
 }
+
+/**
+ * i40e_xmit_descriptor_count - calculate number of Tx descriptors needed
+ * @skb:     send buffer
+ * @tx_ring: ring to send buffer on
+ *
+ * Returns number of data descriptors needed for this skb. Returns 0 to indicate
+ * there is not enough descriptors available in this ring since we need at least
+ * one descriptor.
+ **/
+static inline int i40e_xmit_descriptor_count(struct sk_buff *skb)
+{
+	const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
+	unsigned int nr_frags = skb_shinfo(skb)->nr_frags;
+	int count = 0, size = skb_headlen(skb);
+
+	for (;;) {
+		count += TXD_USE_COUNT(size);
+
+		if (!nr_frags--)
+			break;
+
+		size = skb_frag_size(frag++);
+	}
+
+	return count;
+}
+
+/**
+ * i40e_maybe_stop_tx - 1st level check for Tx stop conditions
+ * @tx_ring: the ring to be checked
+ * @size:    the size buffer we want to assure is available
+ *
+ * Returns 0 if stop is not needed
+ **/
+static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
+{
+	if (likely(I40E_DESC_UNUSED(tx_ring) >= size))
+		return 0;
+	return __i40evf_maybe_stop_tx(tx_ring, size);
+}
 #endif /* _I40E_TXRX_H_ */
-- 
cgit v0.10.2


From 2d37490b82afe1d1b745811e6ce0a4d16bc5e996 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 17 Feb 2016 11:02:50 -0800
Subject: i40e/i40evf: Rewrite logic for 8 descriptor per packet check

This patch is meant to rewrite the logic for how we determine if we can
transmit the frame or if it needs to be linearized.

The previous code for this function was using a mix of division and modulus
division as a part of computing if we need to take the slow path.  Instead
I have replaced this by simply working with a sliding window which will
tell us if the frame would be capable of causing a single packet to span
several descriptors.

The logic for the scan is fairly simple.  If any given group of 6 fragments
is less than gso_size - 1 then it is possible for us to have one byte
coming out of the first fragment, 6 fragments, and one or more bytes coming
out of the last fragment.  This gives us a total of 8 fragments
which exceeds what we can allow so we send such frames to be linearized.

Arguably the use of modulus might be more exact as the approach I propose
may generate some false positives.  However the likelihood of us taking much
of a hit for those false positives is fairly low, and I would rather not
add more overhead in the case where we are receiving a frame composed of 4K
pages.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
index 518d72e..052df93 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
@@ -1368,6 +1368,12 @@ static netdev_tx_t i40e_fcoe_xmit_frame(struct sk_buff *skb,
 		goto out_drop;
 
 	count = i40e_xmit_descriptor_count(skb);
+	if (i40e_chk_linearize(skb, count)) {
+		if (__skb_linearize(skb))
+			goto out_drop;
+		count = TXD_USE_COUNT(skb->len);
+		tx_ring->tx_stats.tx_linearize++;
+	}
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
 	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index f036570..5123646 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2593,59 +2593,71 @@ int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 }
 
 /**
- * i40e_chk_linearize - Check if there are more than 8 fragments per packet
+ * __i40e_chk_linearize - Check if there are more than 8 fragments per packet
  * @skb:      send buffer
- * @tx_flags: collected send information
  *
  * Note: Our HW can't scatter-gather more than 8 fragments to build
  * a packet on the wire and so we need to figure out the cases where we
  * need to linearize the skb.
  **/
-static bool i40e_chk_linearize(struct sk_buff *skb, u32 tx_flags)
+bool __i40e_chk_linearize(struct sk_buff *skb)
 {
-	struct skb_frag_struct *frag;
-	bool linearize = false;
-	unsigned int size = 0;
-	u16 num_frags;
-	u16 gso_segs;
+	const struct skb_frag_struct *frag, *stale;
+	int gso_size, nr_frags, sum;
 
-	num_frags = skb_shinfo(skb)->nr_frags;
-	gso_segs = skb_shinfo(skb)->gso_segs;
+	/* check to see if TSO is enabled, if so we may get a repreive */
+	gso_size = skb_shinfo(skb)->gso_size;
+	if (unlikely(!gso_size))
+		return true;
 
-	if (tx_flags & (I40E_TX_FLAGS_TSO | I40E_TX_FLAGS_FSO)) {
-		u16 j = 0;
+	/* no need to check if number of frags is less than 8 */
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	if (nr_frags < I40E_MAX_BUFFER_TXD)
+		return false;
 
-		if (num_frags < (I40E_MAX_BUFFER_TXD))
-			goto linearize_chk_done;
-		/* try the simple math, if we have too many frags per segment */
-		if (DIV_ROUND_UP((num_frags + gso_segs), gso_segs) >
-		    I40E_MAX_BUFFER_TXD) {
-			linearize = true;
-			goto linearize_chk_done;
-		}
-		frag = &skb_shinfo(skb)->frags[0];
-		/* we might still have more fragments per segment */
-		do {
-			size += skb_frag_size(frag);
-			frag++; j++;
-			if ((size >= skb_shinfo(skb)->gso_size) &&
-			    (j < I40E_MAX_BUFFER_TXD)) {
-				size = (size % skb_shinfo(skb)->gso_size);
-				j = (size) ? 1 : 0;
-			}
-			if (j == I40E_MAX_BUFFER_TXD) {
-				linearize = true;
-				break;
-			}
-			num_frags--;
-		} while (num_frags);
-	} else {
-		if (num_frags >= I40E_MAX_BUFFER_TXD)
-			linearize = true;
+	/* We need to walk through the list and validate that each group
+	 * of 6 fragments totals at least gso_size.  However we don't need
+	 * to perform such validation on the first or last 6 since the first
+	 * 6 cannot inherit any data from a descriptor before them, and the
+	 * last 6 cannot inherit any data from a descriptor after them.
+	 */
+	nr_frags -= I40E_MAX_BUFFER_TXD - 1;
+	frag = &skb_shinfo(skb)->frags[0];
+
+	/* Initialize size to the negative value of gso_size minus 1.  We
+	 * use this as the worst case scenerio in which the frag ahead
+	 * of us only provides one byte which is why we are limited to 6
+	 * descriptors for a single transmit as the header and previous
+	 * fragment are already consuming 2 descriptors.
+	 */
+	sum = 1 - gso_size;
+
+	/* Add size of frags 1 through 5 to create our initial sum */
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+
+	/* Walk through fragments adding latest fragment, testing it, and
+	 * then removing stale fragments from the sum.
+	 */
+	stale = &skb_shinfo(skb)->frags[0];
+	for (;;) {
+		sum += skb_frag_size(++frag);
+
+		/* if sum is negative we failed to make sufficient progress */
+		if (sum < 0)
+			return true;
+
+		/* use pre-decrement to avoid processing last fragment */
+		if (!--nr_frags)
+			break;
+
+		sum -= skb_frag_size(++stale);
 	}
 
-linearize_chk_done:
-	return linearize;
+	return false;
 }
 
 /**
@@ -2876,6 +2888,12 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	prefetch(skb->data);
 
 	count = i40e_xmit_descriptor_count(skb);
+	if (i40e_chk_linearize(skb, count)) {
+		if (__skb_linearize(skb))
+			goto out_drop;
+		count = TXD_USE_COUNT(skb->len);
+		tx_ring->tx_stats.tx_linearize++;
+	}
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
 	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
@@ -2916,11 +2934,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	if (tsyn)
 		tx_flags |= I40E_TX_FLAGS_TSYN;
 
-	if (i40e_chk_linearize(skb, tx_flags)) {
-		if (skb_linearize(skb))
-			goto out_drop;
-		tx_ring->tx_stats.tx_linearize++;
-	}
 	skb_tx_timestamp(skb);
 
 	/* always enable CRC insertion offload */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 5dbc958..5600970 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -337,6 +337,7 @@ int i40e_tx_prepare_vlan_flags(struct sk_buff *skb,
 void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
 u32 i40e_get_tx_pending(struct i40e_ring *ring, bool in_sw);
 int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size);
+bool __i40e_chk_linearize(struct sk_buff *skb);
 
 /**
  * i40e_get_head - Retrieve head from head writeback
@@ -392,4 +393,22 @@ static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 		return 0;
 	return __i40e_maybe_stop_tx(tx_ring, size);
 }
+
+/**
+ * i40e_chk_linearize - Check if there are more than 8 fragments per packet
+ * @skb:      send buffer
+ * @count:    number of buffers used
+ *
+ * Note: Our HW can't scatter-gather more than 8 fragments to build
+ * a packet on the wire and so we need to figure out the cases where we
+ * need to linearize the skb.
+ **/
+static inline bool i40e_chk_linearize(struct sk_buff *skb, int count)
+{
+	/* we can only support up to 8 data buffers for a single send */
+	if (likely(count <= I40E_MAX_BUFFER_TXD))
+		return false;
+
+	return __i40e_chk_linearize(skb);
+}
 #endif /* _I40E_TXRX_H_ */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 9f38127..2369db5 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1796,59 +1796,71 @@ static void i40e_create_tx_ctx(struct i40e_ring *tx_ring,
 }
 
 /**
- * i40e_chk_linearize - Check if there are more than 8 fragments per packet
+ * __i40evf_chk_linearize - Check if there are more than 8 fragments per packet
  * @skb:      send buffer
- * @tx_flags: collected send information
  *
  * Note: Our HW can't scatter-gather more than 8 fragments to build
  * a packet on the wire and so we need to figure out the cases where we
  * need to linearize the skb.
  **/
-static bool i40e_chk_linearize(struct sk_buff *skb, u32 tx_flags)
+bool __i40evf_chk_linearize(struct sk_buff *skb)
 {
-	struct skb_frag_struct *frag;
-	bool linearize = false;
-	unsigned int size = 0;
-	u16 num_frags;
-	u16 gso_segs;
+	const struct skb_frag_struct *frag, *stale;
+	int gso_size, nr_frags, sum;
 
-	num_frags = skb_shinfo(skb)->nr_frags;
-	gso_segs = skb_shinfo(skb)->gso_segs;
+	/* check to see if TSO is enabled, if so we may get a repreive */
+	gso_size = skb_shinfo(skb)->gso_size;
+	if (unlikely(!gso_size))
+		return true;
 
-	if (tx_flags & (I40E_TX_FLAGS_TSO | I40E_TX_FLAGS_FSO)) {
-		u16 j = 0;
+	/* no need to check if number of frags is less than 8 */
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	if (nr_frags < I40E_MAX_BUFFER_TXD)
+		return false;
 
-		if (num_frags < (I40E_MAX_BUFFER_TXD))
-			goto linearize_chk_done;
-		/* try the simple math, if we have too many frags per segment */
-		if (DIV_ROUND_UP((num_frags + gso_segs), gso_segs) >
-		    I40E_MAX_BUFFER_TXD) {
-			linearize = true;
-			goto linearize_chk_done;
-		}
-		frag = &skb_shinfo(skb)->frags[0];
-		/* we might still have more fragments per segment */
-		do {
-			size += skb_frag_size(frag);
-			frag++; j++;
-			if ((size >= skb_shinfo(skb)->gso_size) &&
-			    (j < I40E_MAX_BUFFER_TXD)) {
-				size = (size % skb_shinfo(skb)->gso_size);
-				j = (size) ? 1 : 0;
-			}
-			if (j == I40E_MAX_BUFFER_TXD) {
-				linearize = true;
-				break;
-			}
-			num_frags--;
-		} while (num_frags);
-	} else {
-		if (num_frags >= I40E_MAX_BUFFER_TXD)
-			linearize = true;
+	/* We need to walk through the list and validate that each group
+	 * of 6 fragments totals at least gso_size.  However we don't need
+	 * to perform such validation on the first or last 6 since the first
+	 * 6 cannot inherit any data from a descriptor before them, and the
+	 * last 6 cannot inherit any data from a descriptor after them.
+	 */
+	nr_frags -= I40E_MAX_BUFFER_TXD - 1;
+	frag = &skb_shinfo(skb)->frags[0];
+
+	/* Initialize size to the negative value of gso_size minus 1.  We
+	 * use this as the worst case scenerio in which the frag ahead
+	 * of us only provides one byte which is why we are limited to 6
+	 * descriptors for a single transmit as the header and previous
+	 * fragment are already consuming 2 descriptors.
+	 */
+	sum = 1 - gso_size;
+
+	/* Add size of frags 1 through 5 to create our initial sum */
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+	sum += skb_frag_size(++frag);
+
+	/* Walk through fragments adding latest fragment, testing it, and
+	 * then removing stale fragments from the sum.
+	 */
+	stale = &skb_shinfo(skb)->frags[0];
+	for (;;) {
+		sum += skb_frag_size(++frag);
+
+		/* if sum is negative we failed to make sufficient progress */
+		if (sum < 0)
+			return true;
+
+		/* use pre-decrement to avoid processing last fragment */
+		if (!--nr_frags)
+			break;
+
+		sum -= skb_frag_size(++stale);
 	}
 
-linearize_chk_done:
-	return linearize;
+	return false;
 }
 
 /**
@@ -2095,6 +2107,12 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	prefetch(skb->data);
 
 	count = i40e_xmit_descriptor_count(skb);
+	if (i40e_chk_linearize(skb, count)) {
+		if (__skb_linearize(skb))
+			goto out_drop;
+		count = TXD_USE_COUNT(skb->len);
+		tx_ring->tx_stats.tx_linearize++;
+	}
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
 	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
@@ -2130,11 +2148,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	else if (tso)
 		tx_flags |= I40E_TX_FLAGS_TSO;
 
-	if (i40e_chk_linearize(skb, tx_flags)) {
-		if (skb_linearize(skb))
-			goto out_drop;
-		tx_ring->tx_stats.tx_linearize++;
-	}
 	skb_tx_timestamp(skb);
 
 	/* always enable CRC insertion offload */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 1f719d0..c1dd8c5 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -327,6 +327,7 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget);
 void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector);
 u32 i40evf_get_tx_pending(struct i40e_ring *ring, bool in_sw);
 int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size);
+bool __i40evf_chk_linearize(struct sk_buff *skb);
 
 /**
  * i40e_get_head - Retrieve head from head writeback
@@ -382,4 +383,22 @@ static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 		return 0;
 	return __i40evf_maybe_stop_tx(tx_ring, size);
 }
+
+/**
+ * i40e_chk_linearize - Check if there are more than 8 fragments per packet
+ * @skb:      send buffer
+ * @count:    number of buffers used
+ *
+ * Note: Our HW can't scatter-gather more than 8 fragments to build
+ * a packet on the wire and so we need to figure out the cases where we
+ * need to linearize the skb.
+ **/
+static inline bool i40e_chk_linearize(struct sk_buff *skb, int count)
+{
+	/* we can only support up to 8 data buffers for a single send */
+	if (likely(count <= I40E_MAX_BUFFER_TXD))
+		return false;
+
+	return __i40evf_chk_linearize(skb);
+}
 #endif /* _I40E_TXRX_H_ */
-- 
cgit v0.10.2


From 3bc67973e81d5104908a4ba7c2aab98a4f7bd64e Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 17 Feb 2016 11:02:56 -0800
Subject: i40e/i40evf: Move Tx checksum closer to TSO

On all of the other Intel drivers we place checksum close to TSO as they
have a significant amount in common and it can help to reduce the decision
tree for how to handle the frame as the first check in TSO is to see if
checksumming is offloaded, and if it is not we can skip _BOTH_ TSO and Tx
checksum offload based on a single check.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 5123646..cb52f39 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2929,6 +2929,12 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	else if (tso)
 		tx_flags |= I40E_TX_FLAGS_TSO;
 
+	/* Always offload the checksum, since it's in the data descriptor */
+	tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
+				  tx_ring, &cd_tunneling);
+	if (tso < 0)
+		goto out_drop;
+
 	tsyn = i40e_tsyn(tx_ring, skb, tx_flags, &cd_type_cmd_tso_mss);
 
 	if (tsyn)
@@ -2939,12 +2945,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	/* always enable CRC insertion offload */
 	td_cmd |= I40E_TX_DESC_CMD_ICRC;
 
-	/* Always offload the checksum, since it's in the data descriptor */
-	tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
-				  tx_ring, &cd_tunneling);
-	if (tso < 0)
-		goto out_drop;
-
 	i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss,
 			   cd_tunneling, cd_l2tag2);
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 2369db5..ebcc25c 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -2148,17 +2148,17 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	else if (tso)
 		tx_flags |= I40E_TX_FLAGS_TSO;
 
-	skb_tx_timestamp(skb);
-
-	/* always enable CRC insertion offload */
-	td_cmd |= I40E_TX_DESC_CMD_ICRC;
-
 	/* Always offload the checksum, since it's in the data descriptor */
 	tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset,
 				  tx_ring, &cd_tunneling);
 	if (tso < 0)
 		goto out_drop;
 
+	skb_tx_timestamp(skb);
+
+	/* always enable CRC insertion offload */
+	td_cmd |= I40E_TX_DESC_CMD_ICRC;
+
 	i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss,
 			   cd_tunneling, cd_l2tag2);
 
-- 
cgit v0.10.2


From fd077cd3399b164548f538625f07f3e9f1d7ef00 Mon Sep 17 00:00:00 2001
From: Carolyn Wyborny <carolyn.wyborny@intel.com>
Date: Wed, 17 Feb 2016 16:12:11 -0800
Subject: i40e: Add functions to blink led on 10GBaseT PHY

This patch adds functions to blink led on devices using
10GBaseT PHY since MAC registers used in other designs
do not work in this device configuration.

Change-ID: Id4b88c93c649fd2b88073a00b42867a77c761ca3
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index f923933..447729f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -4214,3 +4214,332 @@ i40e_status i40e_aq_configure_partition_bw(struct i40e_hw *hw,
 
 	return status;
 }
+
+/**
+ * i40e_read_phy_register
+ * @hw: pointer to the HW structure
+ * @page: registers page number
+ * @reg: register address in the page
+ * @phy_adr: PHY address on MDIO interface
+ * @value: PHY register value
+ *
+ * Reads specified PHY register value
+ **/
+i40e_status i40e_read_phy_register(struct i40e_hw *hw,
+				   u8 page, u16 reg, u8 phy_addr,
+				   u16 *value)
+{
+	i40e_status status = I40E_ERR_TIMEOUT;
+	u32 command = 0;
+	u16 retry = 1000;
+	u8 port_num = hw->func_caps.mdio_port_num;
+
+	command = (reg << I40E_GLGEN_MSCA_MDIADD_SHIFT) |
+		  (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) |
+		  (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) |
+		  (I40E_MDIO_OPCODE_ADDRESS) |
+		  (I40E_MDIO_STCODE) |
+		  (I40E_GLGEN_MSCA_MDICMD_MASK) |
+		  (I40E_GLGEN_MSCA_MDIINPROGEN_MASK);
+	wr32(hw, I40E_GLGEN_MSCA(port_num), command);
+	do {
+		command = rd32(hw, I40E_GLGEN_MSCA(port_num));
+		if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) {
+			status = 0;
+			break;
+		}
+		usleep_range(10, 20);
+		retry--;
+	} while (retry);
+
+	if (status) {
+		i40e_debug(hw, I40E_DEBUG_PHY,
+			   "PHY: Can't write command to external PHY.\n");
+		goto phy_read_end;
+	}
+
+	command = (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) |
+		  (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) |
+		  (I40E_MDIO_OPCODE_READ) |
+		  (I40E_MDIO_STCODE) |
+		  (I40E_GLGEN_MSCA_MDICMD_MASK) |
+		  (I40E_GLGEN_MSCA_MDIINPROGEN_MASK);
+	status = I40E_ERR_TIMEOUT;
+	retry = 1000;
+	wr32(hw, I40E_GLGEN_MSCA(port_num), command);
+	do {
+		command = rd32(hw, I40E_GLGEN_MSCA(port_num));
+		if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) {
+			status = 0;
+			break;
+		}
+		usleep_range(10, 20);
+		retry--;
+	} while (retry);
+
+	if (!status) {
+		command = rd32(hw, I40E_GLGEN_MSRWD(port_num));
+		*value = (command & I40E_GLGEN_MSRWD_MDIRDDATA_MASK) >>
+			 I40E_GLGEN_MSRWD_MDIRDDATA_SHIFT;
+	} else {
+		i40e_debug(hw, I40E_DEBUG_PHY,
+			   "PHY: Can't read register value from external PHY.\n");
+	}
+
+phy_read_end:
+	return status;
+}
+
+/**
+ * i40e_write_phy_register
+ * @hw: pointer to the HW structure
+ * @page: registers page number
+ * @reg: register address in the page
+ * @phy_adr: PHY address on MDIO interface
+ * @value: PHY register value
+ *
+ * Writes value to specified PHY register
+ **/
+i40e_status i40e_write_phy_register(struct i40e_hw *hw,
+				    u8 page, u16 reg, u8 phy_addr,
+				    u16 value)
+{
+	i40e_status status = I40E_ERR_TIMEOUT;
+	u32 command = 0;
+	u16 retry = 1000;
+	u8 port_num = hw->func_caps.mdio_port_num;
+
+	command = (reg << I40E_GLGEN_MSCA_MDIADD_SHIFT) |
+		  (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) |
+		  (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) |
+		  (I40E_MDIO_OPCODE_ADDRESS) |
+		  (I40E_MDIO_STCODE) |
+		  (I40E_GLGEN_MSCA_MDICMD_MASK) |
+		  (I40E_GLGEN_MSCA_MDIINPROGEN_MASK);
+	wr32(hw, I40E_GLGEN_MSCA(port_num), command);
+	do {
+		command = rd32(hw, I40E_GLGEN_MSCA(port_num));
+		if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) {
+			status = 0;
+			break;
+		}
+		usleep_range(10, 20);
+		retry--;
+	} while (retry);
+	if (status) {
+		i40e_debug(hw, I40E_DEBUG_PHY,
+			   "PHY: Can't write command to external PHY.\n");
+		goto phy_write_end;
+	}
+
+	command = value << I40E_GLGEN_MSRWD_MDIWRDATA_SHIFT;
+	wr32(hw, I40E_GLGEN_MSRWD(port_num), command);
+
+	command = (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) |
+		  (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) |
+		  (I40E_MDIO_OPCODE_WRITE) |
+		  (I40E_MDIO_STCODE) |
+		  (I40E_GLGEN_MSCA_MDICMD_MASK) |
+		  (I40E_GLGEN_MSCA_MDIINPROGEN_MASK);
+	status = I40E_ERR_TIMEOUT;
+	retry = 1000;
+	wr32(hw, I40E_GLGEN_MSCA(port_num), command);
+	do {
+		command = rd32(hw, I40E_GLGEN_MSCA(port_num));
+		if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) {
+			status = 0;
+			break;
+		}
+		usleep_range(10, 20);
+		retry--;
+	} while (retry);
+
+phy_write_end:
+	return status;
+}
+
+/**
+ * i40e_get_phy_address
+ * @hw: pointer to the HW structure
+ * @dev_num: PHY port num that address we want
+ * @phy_addr: Returned PHY address
+ *
+ * Gets PHY address for current port
+ **/
+u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num)
+{
+	u8 port_num = hw->func_caps.mdio_port_num;
+	u32 reg_val = rd32(hw, I40E_GLGEN_MDIO_I2C_SEL(port_num));
+
+	return (u8)(reg_val >> ((dev_num + 1) * 5)) & 0x1f;
+}
+
+/**
+ * i40e_blink_phy_led
+ * @hw: pointer to the HW structure
+ * @time: time how long led will blinks in secs
+ * @interval: gap between LED on and off in msecs
+ *
+ * Blinks PHY link LED
+ **/
+i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw,
+				    u32 time, u32 interval)
+{
+	i40e_status status = 0;
+	u32 i;
+	u16 led_ctl;
+	u16 gpio_led_port;
+	u16 led_reg;
+	u16 led_addr = I40E_PHY_LED_PROV_REG_1;
+	u8 phy_addr = 0;
+	u8 port_num;
+
+	i = rd32(hw, I40E_PFGEN_PORTNUM);
+	port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
+	phy_addr = i40e_get_phy_address(hw, port_num);
+
+	for (gpio_led_port = 0; gpio_led_port < 3; gpio_led_port++,
+	     led_addr++) {
+		status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE,
+						led_addr, phy_addr, &led_reg);
+		if (status)
+			goto phy_blinking_end;
+		led_ctl = led_reg;
+		if (led_reg & I40E_PHY_LED_LINK_MODE_MASK) {
+			led_reg = 0;
+			status = i40e_write_phy_register(hw,
+							 I40E_PHY_COM_REG_PAGE,
+							 led_addr, phy_addr,
+							 led_reg);
+			if (status)
+				goto phy_blinking_end;
+			break;
+		}
+	}
+
+	if (time > 0 && interval > 0) {
+		for (i = 0; i < time * 1000; i += interval) {
+			status = i40e_read_phy_register(hw,
+							I40E_PHY_COM_REG_PAGE,
+							led_addr, phy_addr,
+							&led_reg);
+			if (status)
+				goto restore_config;
+			if (led_reg & I40E_PHY_LED_MANUAL_ON)
+				led_reg = 0;
+			else
+				led_reg = I40E_PHY_LED_MANUAL_ON;
+			status = i40e_write_phy_register(hw,
+							 I40E_PHY_COM_REG_PAGE,
+							 led_addr, phy_addr,
+							 led_reg);
+			if (status)
+				goto restore_config;
+			msleep(interval);
+		}
+	}
+
+restore_config:
+	status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE, led_addr,
+					 phy_addr, led_ctl);
+
+phy_blinking_end:
+	return status;
+}
+
+/**
+ * i40e_led_get_phy - return current on/off mode
+ * @hw: pointer to the hw struct
+ * @led_addr: address of led register to use
+ * @val: original value of register to use
+ *
+ **/
+i40e_status i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr,
+			     u16 *val)
+{
+	i40e_status status = 0;
+	u16 gpio_led_port;
+	u8 phy_addr = 0;
+	u16 reg_val;
+	u16 temp_addr;
+	u8 port_num;
+	u32 i;
+
+	temp_addr = I40E_PHY_LED_PROV_REG_1;
+	i = rd32(hw, I40E_PFGEN_PORTNUM);
+	port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
+	phy_addr = i40e_get_phy_address(hw, port_num);
+
+	for (gpio_led_port = 0; gpio_led_port < 3; gpio_led_port++,
+	     temp_addr++) {
+		status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE,
+						temp_addr, phy_addr, &reg_val);
+		if (status)
+			return status;
+		*val = reg_val;
+		if (reg_val & I40E_PHY_LED_LINK_MODE_MASK) {
+			*led_addr = temp_addr;
+			break;
+		}
+	}
+	return status;
+}
+
+/**
+ * i40e_led_set_phy
+ * @hw: pointer to the HW structure
+ * @on: true or false
+ * @mode: original val plus bit for set or ignore
+ * Set led's on or off when controlled by the PHY
+ *
+ **/
+i40e_status i40e_led_set_phy(struct i40e_hw *hw, bool on,
+			     u16 led_addr, u32 mode)
+{
+	i40e_status status = 0;
+	u16 led_ctl = 0;
+	u16 led_reg = 0;
+	u8 phy_addr = 0;
+	u8 port_num;
+	u32 i;
+
+	i = rd32(hw, I40E_PFGEN_PORTNUM);
+	port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK);
+	phy_addr = i40e_get_phy_address(hw, port_num);
+
+	status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE, led_addr,
+					phy_addr, &led_reg);
+	if (status)
+		return status;
+	led_ctl = led_reg;
+	if (led_reg & I40E_PHY_LED_LINK_MODE_MASK) {
+		led_reg = 0;
+		status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE,
+						 led_addr, phy_addr, led_reg);
+		if (status)
+			return status;
+	}
+	status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE,
+					led_addr, phy_addr, &led_reg);
+	if (status)
+		goto restore_config;
+	if (on)
+		led_reg = I40E_PHY_LED_MANUAL_ON;
+	else
+		led_reg = 0;
+	status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE,
+					 led_addr, phy_addr, led_reg);
+	if (status)
+		goto restore_config;
+	if (mode & I40E_PHY_LED_MODE_ORIG) {
+		led_ctl = (mode & I40E_PHY_LED_MODE_MASK);
+		status = i40e_write_phy_register(hw,
+						 I40E_PHY_COM_REG_PAGE,
+						 led_addr, phy_addr, led_ctl);
+	}
+	return status;
+restore_config:
+	status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE, led_addr,
+					 phy_addr, led_ctl);
+	return status;
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index e8deabd..ca2f7ac 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -74,6 +74,12 @@ i40e_status i40e_aq_set_rss_key(struct i40e_hw *hw,
 
 u32 i40e_led_get(struct i40e_hw *hw);
 void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink);
+i40e_status i40e_led_set_phy(struct i40e_hw *hw, bool on,
+			     u16 led_addr, u32 mode);
+i40e_status i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr,
+			     u16 *val);
+i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw,
+				    u32 time, u32 interval);
 
 /* admin send queue commands */
 
@@ -336,4 +342,11 @@ i40e_status i40e_aq_debug_dump(struct i40e_hw *hw, u8 cluster_id,
 			       struct i40e_asq_cmd_details *cmd_details);
 void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw,
 						    u16 vsi_seid);
+i40e_status i40e_read_phy_register(struct i40e_hw *hw, u8 page,
+				   u16 reg, u8 phy_addr, u16 *value);
+i40e_status i40e_write_phy_register(struct i40e_hw *hw, u8 page,
+				    u16 reg, u8 phy_addr, u16 value);
+u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num);
+i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw,
+				    u32 time, u32 interval);
 #endif /* _I40E_PROTOTYPE_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index b59a021..0a0baf7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -90,6 +90,22 @@ enum i40e_debug_mask {
 	I40E_DEBUG_ALL			= 0xFFFFFFFF
 };
 
+#define I40E_MDIO_STCODE                0
+#define I40E_MDIO_OPCODE_ADDRESS        0
+#define I40E_MDIO_OPCODE_WRITE          I40E_MASK(1, \
+						  I40E_GLGEN_MSCA_OPCODE_SHIFT)
+#define I40E_MDIO_OPCODE_READ_INC_ADDR  I40E_MASK(2, \
+						  I40E_GLGEN_MSCA_OPCODE_SHIFT)
+#define I40E_MDIO_OPCODE_READ           I40E_MASK(3, \
+						  I40E_GLGEN_MSCA_OPCODE_SHIFT)
+
+#define I40E_PHY_COM_REG_PAGE                   0x1E
+#define I40E_PHY_LED_LINK_MODE_MASK             0xF0
+#define I40E_PHY_LED_MANUAL_ON                  0x100
+#define I40E_PHY_LED_PROV_REG_1                 0xC430
+#define I40E_PHY_LED_MODE_MASK                  0xFFFF
+#define I40E_PHY_LED_MODE_ORIG                  0x80000000
+
 /* These are structs for managing the hardware information and the operations.
  * The structures of function pointers are filled out at init time when we
  * know for sure exactly which hardware we're working with.  This gives us the
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
index cbd9a1b..fa34d85 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
@@ -103,4 +103,11 @@ i40e_status i40e_aq_add_rem_control_packet_filter(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
 void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw,
 						    u16 vsi_seid);
+i40e_status i40e_read_phy_register(struct i40e_hw *hw, u8 page,
+				   u16 reg, u8 phy_addr, u16 *value);
+i40e_status i40e_write_phy_register(struct i40e_hw *hw, u8 page,
+				    u16 reg, u8 phy_addr, u16 value);
+u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num);
+i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw,
+				    u32 time, u32 interval);
 #endif /* _I40E_PROTOTYPE_H_ */
-- 
cgit v0.10.2


From 31b606d0c40a1435c54bff18e4d3d3c33af1c3cf Mon Sep 17 00:00:00 2001
From: Carolyn Wyborny <carolyn.wyborny@intel.com>
Date: Wed, 17 Feb 2016 16:12:12 -0800
Subject: i40e: Fix led blink capability for 10GBaseT PHY

This patch fixes a problem where the ethtool identify adapter
functionality did not work for some copper PHY's.  Without this
patch, the blink led functionality fails on some parts.  This
patch adds PHY write code to blink led's on parts where this
functionality is contained in the PHY rather than the MAC.

Change-ID: Iee7b3453f61d5ffd0b3d03f720ee4f17f919fcc2
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index e0758dd..e99be9f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -111,6 +111,7 @@
 #define I40E_OEM_VER_PATCH_MASK    0xff
 #define I40E_OEM_VER_BUILD_SHIFT   8
 #define I40E_OEM_VER_SHIFT         24
+#define I40E_PHY_DEBUG_PORT        BIT(4)
 
 /* The values in here are decimal coded as hex as is the case in the NVM map*/
 #define I40E_CURRENT_NVM_VERSION_HI 0x2
@@ -355,6 +356,7 @@ struct i40e_pf {
 #define I40E_FLAG_NO_DCB_SUPPORT		BIT_ULL(45)
 #define I40E_FLAG_USE_SET_LLDP_MIB		BIT_ULL(46)
 #define I40E_FLAG_STOP_FW_LLDP			BIT_ULL(47)
+#define I40E_FLAG_HAVE_10GBASET_PHY		BIT_ULL(48)
 #define I40E_FLAG_PF_MAC			BIT_ULL(50)
 
 	/* tracks features that get auto disabled by errors */
@@ -440,6 +442,7 @@ struct i40e_pf {
 
 	u32 ioremap_len;
 	u32 fd_inv;
+	u16 phy_led_val;
 };
 
 struct i40e_mac_filter {
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 447729f..d417193 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1888,6 +1888,32 @@ i40e_status i40e_aq_set_phy_int_mask(struct i40e_hw *hw,
 }
 
 /**
+ * i40e_aq_set_phy_debug
+ * @hw: pointer to the hw struct
+ * @cmd_flags: debug command flags
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Reset the external PHY.
+ **/
+enum i40e_status_code i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags,
+					struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_set_phy_debug *cmd =
+		(struct i40e_aqc_set_phy_debug *)&desc.params.raw;
+	enum i40e_status_code status;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_set_phy_debug);
+
+	cmd->command_flags = cmd_flags;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
  * i40e_aq_add_vsi
  * @hw: pointer to the hw struct
  * @vsi_ctx: pointer to a vsi context struct
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 9fc7546b..f1ad1016 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1826,28 +1826,52 @@ static int i40e_set_phys_id(struct net_device *netdev,
 			    enum ethtool_phys_id_state state)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	i40e_status ret = 0;
 	struct i40e_pf *pf = np->vsi->back;
 	struct i40e_hw *hw = &pf->hw;
 	int blink_freq = 2;
+	u16 temp_status;
 
 	switch (state) {
 	case ETHTOOL_ID_ACTIVE:
-		pf->led_status = i40e_led_get(hw);
+		if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY)) {
+			pf->led_status = i40e_led_get(hw);
+		} else {
+			i40e_aq_set_phy_debug(hw, I40E_PHY_DEBUG_PORT, NULL);
+			ret = i40e_led_get_phy(hw, &temp_status,
+					       &pf->phy_led_val);
+			pf->led_status = temp_status;
+		}
 		return blink_freq;
 	case ETHTOOL_ID_ON:
-		i40e_led_set(hw, 0xF, false);
+		if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY))
+			i40e_led_set(hw, 0xf, false);
+		else
+			ret = i40e_led_set_phy(hw, true, pf->led_status, 0);
 		break;
 	case ETHTOOL_ID_OFF:
-		i40e_led_set(hw, 0x0, false);
+		if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY))
+			i40e_led_set(hw, 0x0, false);
+		else
+			ret = i40e_led_set_phy(hw, false, pf->led_status, 0);
 		break;
 	case ETHTOOL_ID_INACTIVE:
-		i40e_led_set(hw, pf->led_status, false);
+		if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY)) {
+			i40e_led_set(hw, false, pf->led_status);
+		} else {
+			ret = i40e_led_set_phy(hw, false, pf->led_status,
+					       (pf->phy_led_val |
+					       I40E_PHY_LED_MODE_ORIG));
+			i40e_aq_set_phy_debug(hw, 0, NULL);
+		}
 		break;
 	default:
 		break;
 	}
-
-	return 0;
+		if (ret)
+			return -ENOENT;
+		else
+			return 0;
 }
 
 /* NOTE: i40e hardware uses a conversion factor of 2 for Interrupt
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 2f2b2d7..b29b131 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -11130,6 +11130,10 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	i40e_add_filter_to_drop_tx_flow_control_frames(&pf->hw,
 						       pf->main_vsi_seid);
 
+	if ((pf->hw.device_id == I40E_DEV_ID_10G_BASE_T) ||
+	    (pf->hw.device_id == I40E_DEV_ID_10G_BASE_T4))
+		pf->flags |= I40E_FLAG_HAVE_10GBASET_PHY;
+
 	/* print a string summarizing features */
 	i40e_print_features(pf);
 
-- 
cgit v0.10.2


From 4d7cec078de864b7ba5459aa688278c4e6f3ad42 Mon Sep 17 00:00:00 2001
From: Kevin Scott <kevin.c.scott@intel.com>
Date: Wed, 17 Feb 2016 16:12:13 -0800
Subject: i40e: Increase timeout when checking GLGEN_RSTAT_DEVSTATE bit

When linking with particular PHY types (ex: copper PHY), the amount of
time it takes for the GLGEN_RSTAT_DEVSTATE to be set increases greatly,
which can lead to a timeout and failure to load the driver.

Change-ID: If02be0dfcd7c57fdde2d5c81cd63651260cd2029
Signed-off-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index d417193..3a57e59 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1239,7 +1239,13 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw)
 	grst_del = (rd32(hw, I40E_GLGEN_RSTCTL) &
 		    I40E_GLGEN_RSTCTL_GRSTDEL_MASK) >>
 		    I40E_GLGEN_RSTCTL_GRSTDEL_SHIFT;
-	for (cnt = 0; cnt < grst_del + 10; cnt++) {
+
+	/* It can take upto 15 secs for GRST steady state.
+	 * Bump it to 16 secs max to be safe.
+	 */
+	grst_del = grst_del * 20;
+
+	for (cnt = 0; cnt < grst_del; cnt++) {
 		reg = rd32(hw, I40E_GLGEN_RSTAT);
 		if (!(reg & I40E_GLGEN_RSTAT_DEVSTATE_MASK))
 			break;
-- 
cgit v0.10.2


From 3fe06f415b31ad06d2c2923216292057e899eb0d Mon Sep 17 00:00:00 2001
From: Neerav Parikh <neerav.parikh@intel.com>
Date: Wed, 17 Feb 2016 16:12:15 -0800
Subject: i40e: Do not wait for Rx queue disable in DCB reconfig

Just like Tx queues don't wait for Rx queues to be disabled before
DCB has been reconfigured.
Check the queues are disabled only after the DCB configuration has
been applied to the VSI(s) managed by the PF driver.

In case of any timeout issue a PF reset to recover.

Change-ID: Ic51e94c25baf9a5480cee983f35d15575a88642c
Signed-off-by: Neerav Parikh <neerav.parikh@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index b29b131..53660f1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3929,6 +3929,9 @@ static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable)
 		else
 			rx_reg &= ~I40E_QRX_ENA_QENA_REQ_MASK;
 		wr32(hw, I40E_QRX_ENA(pf_q), rx_reg);
+		/* No waiting for the Tx queue to disable */
+		if (!enable && test_bit(__I40E_PORT_TX_SUSPENDED, &pf->state))
+			continue;
 
 		/* wait for the change to finish */
 		ret = i40e_pf_rxq_wait(pf, pf_q, enable);
@@ -4287,12 +4290,12 @@ static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf)
 
 #ifdef CONFIG_I40E_DCB
 /**
- * i40e_vsi_wait_txq_disabled - Wait for VSI's queues to be disabled
+ * i40e_vsi_wait_queues_disabled - Wait for VSI's queues to be disabled
  * @vsi: the VSI being configured
  *
- * This function waits for the given VSI's Tx queues to be disabled.
+ * This function waits for the given VSI's queues to be disabled.
  **/
-static int i40e_vsi_wait_txq_disabled(struct i40e_vsi *vsi)
+static int i40e_vsi_wait_queues_disabled(struct i40e_vsi *vsi)
 {
 	struct i40e_pf *pf = vsi->back;
 	int i, pf_q, ret;
@@ -4309,24 +4312,36 @@ static int i40e_vsi_wait_txq_disabled(struct i40e_vsi *vsi)
 		}
 	}
 
+	pf_q = vsi->base_queue;
+	for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
+		/* Check and wait for the disable status of the queue */
+		ret = i40e_pf_rxq_wait(pf, pf_q, false);
+		if (ret) {
+			dev_info(&pf->pdev->dev,
+				 "VSI seid %d Rx ring %d disable timeout\n",
+				 vsi->seid, pf_q);
+			return ret;
+		}
+	}
+
 	return 0;
 }
 
 /**
- * i40e_pf_wait_txq_disabled - Wait for all queues of PF VSIs to be disabled
+ * i40e_pf_wait_queues_disabled - Wait for all queues of PF VSIs to be disabled
  * @pf: the PF
  *
- * This function waits for the Tx queues to be in disabled state for all the
+ * This function waits for the queues to be in disabled state for all the
  * VSIs that are managed by this PF.
  **/
-static int i40e_pf_wait_txq_disabled(struct i40e_pf *pf)
+static int i40e_pf_wait_queues_disabled(struct i40e_pf *pf)
 {
 	int v, ret = 0;
 
 	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
 		/* No need to wait for FCoE VSI queues */
 		if (pf->vsi[v] && pf->vsi[v]->type != I40E_VSI_FCOE) {
-			ret = i40e_vsi_wait_txq_disabled(pf->vsi[v]);
+			ret = i40e_vsi_wait_queues_disabled(pf->vsi[v]);
 			if (ret)
 				break;
 		}
@@ -5726,8 +5741,8 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf,
 	if (ret)
 		goto exit;
 
-	/* Wait for the PF's Tx queues to be disabled */
-	ret = i40e_pf_wait_txq_disabled(pf);
+	/* Wait for the PF's queues to be disabled */
+	ret = i40e_pf_wait_queues_disabled(pf);
 	if (ret) {
 		/* Schedule PF reset to recover */
 		set_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
-- 
cgit v0.10.2


From 730a8f8777e55912f445c2c29234d51cceb1dfc2 Mon Sep 17 00:00:00 2001
From: Carolyn Wyborny <carolyn.wyborny@intel.com>
Date: Wed, 17 Feb 2016 16:12:16 -0800
Subject: i40e: Fix for unexpected messaging

This fixes an issue where a previously removed message
has returned.  Changing the message type to dev_dbg
leaves the info, if desired, but takes it out of normal
everyday usage. Also changed call to only provide port
data when its valid and not when its not (delete case).

Change-ID: Ief6f33b915f6364c24fa8e5789c2fc3168b5e2ed
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 53660f1..05def9f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -7109,12 +7109,13 @@ static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf)
 				ret = i40e_aq_del_udp_tunnel(hw, i, NULL);
 
 			if (ret) {
-				dev_info(&pf->pdev->dev,
-					 "%s vxlan port %d, index %d failed, err %s aq_err %s\n",
-					 port ? "add" : "delete",
-					 ntohs(port), i,
-					 i40e_stat_str(&pf->hw, ret),
-					 i40e_aq_str(&pf->hw,
+				dev_dbg(&pf->pdev->dev,
+					"%s %s port %d, index %d failed, err %s aq_err %s\n",
+					pf->udp_ports[i].type ? "vxlan" : "geneve",
+					port ? "add" : "delete",
+					ntohs(port), i,
+					i40e_stat_str(&pf->hw, ret),
+					i40e_aq_str(&pf->hw,
 						    pf->hw.aq.asq_last_status));
 				pf->udp_ports[i].index = 0;
 			}
-- 
cgit v0.10.2


From fe726082728da9f653d4e747baf0816d826fc626 Mon Sep 17 00:00:00 2001
From: Anjali Singhai Jain <anjali.singhai@intel.com>
Date: Wed, 17 Feb 2016 16:12:17 -0800
Subject: i40e: Expose some registers to program parser, FD and RSS logic

This patch adds 7 new register definitions for programming the
parser, flow director and RSS blocks in the HW.

Change-ID: I31e76673125275f3c69a14c646361919d04dc987
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_register.h b/drivers/net/ethernet/intel/i40e/i40e_register.h
index dc0402f..86ca27f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_register.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_register.h
@@ -2045,6 +2045,14 @@
 #define I40E_PRTPM_TLPIC 0x001E43C0 /* Reset: GLOBR */
 #define I40E_PRTPM_TLPIC_ETLPIC_SHIFT 0
 #define I40E_PRTPM_TLPIC_ETLPIC_MASK I40E_MASK(0xFFFFFFFF, I40E_PRTPM_TLPIC_ETLPIC_SHIFT)
+#define I40E_GL_PRS_FVBM(_i) (0x00269760 + ((_i) * 4)) /* _i=0...3 */ /* Reset: CORER */
+#define I40E_GL_PRS_FVBM_MAX_INDEX 3
+#define I40E_GL_PRS_FVBM_FV_BYTE_INDX_SHIFT 0
+#define I40E_GL_PRS_FVBM_FV_BYTE_INDX_MASK I40E_MASK(0x7F, I40E_GL_PRS_FVBM_FV_BYTE_INDX_SHIFT)
+#define I40E_GL_PRS_FVBM_RULE_BUS_INDX_SHIFT 8
+#define I40E_GL_PRS_FVBM_RULE_BUS_INDX_MASK I40E_MASK(0x3F, I40E_GL_PRS_FVBM_RULE_BUS_INDX_SHIFT)
+#define I40E_GL_PRS_FVBM_MSK_ENA_SHIFT 31
+#define I40E_GL_PRS_FVBM_MSK_ENA_MASK I40E_MASK(0x1, I40E_GL_PRS_FVBM_MSK_ENA_SHIFT)
 #define I40E_GLRPB_DPSS 0x000AC828 /* Reset: CORER */
 #define I40E_GLRPB_DPSS_DPS_TCN_SHIFT 0
 #define I40E_GLRPB_DPSS_DPS_TCN_MASK I40E_MASK(0xFFFFF, I40E_GLRPB_DPSS_DPS_TCN_SHIFT)
@@ -2216,6 +2224,14 @@
 #define I40E_PRTQF_FD_FLXINSET_MAX_INDEX 63
 #define I40E_PRTQF_FD_FLXINSET_INSET_SHIFT 0
 #define I40E_PRTQF_FD_FLXINSET_INSET_MASK I40E_MASK(0xFF, I40E_PRTQF_FD_FLXINSET_INSET_SHIFT)
+#define I40E_PRTQF_FD_INSET(_i, _j) (0x00250000 + ((_i) * 64 + (_j) * 32)) /* _i=0...63, _j=0...1 */ /* Reset: CORER */
+#define I40E_PRTQF_FD_INSET_MAX_INDEX 63
+#define I40E_PRTQF_FD_INSET_INSET_SHIFT 0
+#define I40E_PRTQF_FD_INSET_INSET_MASK I40E_MASK(0xFFFFFFFF, I40E_PRTQF_FD_INSET_INSET_SHIFT)
+#define I40E_PRTQF_FD_INSET(_i, _j) (0x00250000 + ((_i) * 64 + (_j) * 32)) /* _i=0...63, _j=0...1 */ /* Reset: CORER */
+#define I40E_PRTQF_FD_INSET_MAX_INDEX 63
+#define I40E_PRTQF_FD_INSET_INSET_SHIFT 0
+#define I40E_PRTQF_FD_INSET_INSET_MASK I40E_MASK(0xFFFFFFFF, I40E_PRTQF_FD_INSET_INSET_SHIFT)
 #define I40E_PRTQF_FD_MSK(_i, _j) (0x00252000 + ((_i) * 64 + (_j) * 32)) /* _i=0...63, _j=0...1 */ /* Reset: CORER */
 #define I40E_PRTQF_FD_MSK_MAX_INDEX 63
 #define I40E_PRTQF_FD_MSK_MASK_SHIFT 0
@@ -5155,6 +5171,38 @@
 #define I40E_GLQF_FD_PCTYPES_MAX_INDEX 63
 #define I40E_GLQF_FD_PCTYPES_FD_PCTYPE_SHIFT 0
 #define I40E_GLQF_FD_PCTYPES_FD_PCTYPE_MASK I40E_MASK(0x3F, I40E_GLQF_FD_PCTYPES_FD_PCTYPE_SHIFT)
+#define I40E_GLQF_FD_MSK(_i, _j) (0x00267200 + ((_i) * 4 + (_j) * 8)) /* _i=0...1, _j=0...63 */ /* Reset: CORER */
+#define I40E_GLQF_FD_MSK_MAX_INDEX 1
+#define I40E_GLQF_FD_MSK_MASK_SHIFT 0
+#define I40E_GLQF_FD_MSK_MASK_MASK I40E_MASK(0xFFFF, I40E_GLQF_FD_MSK_MASK_SHIFT)
+#define I40E_GLQF_FD_MSK_OFFSET_SHIFT 16
+#define I40E_GLQF_FD_MSK_OFFSET_MASK I40E_MASK(0x3F, I40E_GLQF_FD_MSK_OFFSET_SHIFT)
+#define I40E_GLQF_HASH_INSET(_i, _j) (0x00267600 + ((_i) * 4 + (_j) * 8)) /* _i=0...1, _j=0...63 */ /* Reset: CORER */
+#define I40E_GLQF_HASH_INSET_MAX_INDEX 1
+#define I40E_GLQF_HASH_INSET_INSET_SHIFT 0
+#define I40E_GLQF_HASH_INSET_INSET_MASK I40E_MASK(0xFFFFFFFF, I40E_GLQF_HASH_INSET_INSET_SHIFT)
+#define I40E_GLQF_HASH_MSK(_i, _j) (0x00267A00 + ((_i) * 4 + (_j) * 8)) /* _i=0...1, _j=0...63 */ /* Reset: CORER */
+#define I40E_GLQF_HASH_MSK_MAX_INDEX 1
+#define I40E_GLQF_HASH_MSK_MASK_SHIFT 0
+#define I40E_GLQF_HASH_MSK_MASK_MASK I40E_MASK(0xFFFF, I40E_GLQF_HASH_MSK_MASK_SHIFT)
+#define I40E_GLQF_HASH_MSK_OFFSET_SHIFT 16
+#define I40E_GLQF_HASH_MSK_OFFSET_MASK I40E_MASK(0x3F, I40E_GLQF_HASH_MSK_OFFSET_SHIFT)
+#define I40E_GLQF_ORT(_i) (0x00268900 + ((_i) * 4)) /* _i=0...63 */ /* Reset: CORER */
+#define I40E_GLQF_ORT_MAX_INDEX 63
+#define I40E_GLQF_ORT_PIT_INDX_SHIFT 0
+#define I40E_GLQF_ORT_PIT_INDX_MASK I40E_MASK(0x1F, I40E_GLQF_ORT_PIT_INDX_SHIFT)
+#define I40E_GLQF_ORT_FIELD_CNT_SHIFT 5
+#define I40E_GLQF_ORT_FIELD_CNT_MASK I40E_MASK(0x3, I40E_GLQF_ORT_FIELD_CNT_SHIFT)
+#define I40E_GLQF_ORT_FLX_PAYLOAD_SHIFT 7
+#define I40E_GLQF_ORT_FLX_PAYLOAD_MASK I40E_MASK(0x1, I40E_GLQF_ORT_FLX_PAYLOAD_SHIFT)
+#define I40E_GLQF_PIT(_i) (0x00268C80 + ((_i) * 4)) /* _i=0...23 */ /* Reset: CORER */
+#define I40E_GLQF_PIT_MAX_INDEX 23
+#define I40E_GLQF_PIT_SOURCE_OFF_SHIFT 0
+#define I40E_GLQF_PIT_SOURCE_OFF_MASK I40E_MASK(0x1F, I40E_GLQF_PIT_SOURCE_OFF_SHIFT)
+#define I40E_GLQF_PIT_FSIZE_SHIFT 5
+#define I40E_GLQF_PIT_FSIZE_MASK I40E_MASK(0x1F, I40E_GLQF_PIT_FSIZE_SHIFT)
+#define I40E_GLQF_PIT_DEST_OFF_SHIFT 10
+#define I40E_GLQF_PIT_DEST_OFF_MASK I40E_MASK(0x3F, I40E_GLQF_PIT_DEST_OFF_SHIFT)
 #define I40E_GLQF_FDEVICTENA(_i) (0x00270384 + ((_i) * 4)) /* _i=0...1 */ /* Reset: CORER */
 #define I40E_GLQF_FDEVICTENA_MAX_INDEX 1
 #define I40E_GLQF_FDEVICTENA_GLQF_FDEVICTENA_SHIFT 0
-- 
cgit v0.10.2


From f534039dd8ab39cb3259e5860d2be3b0e70aacbf Mon Sep 17 00:00:00 2001
From: John Underwood <johnx.underwood@intel.com>
Date: Thu, 18 Feb 2016 09:19:24 -0800
Subject: i40e: add check for null VSI

Return from i40e_vsi_reinit_setup() if vsi param is NULL.
This makes this code consistent with all the other code that
checks for NULL before using one of the VSI pointers accessed
with an indexed variable. (Indexed VSI pointers are
intentionally set to NULL in i40e_vsi_clear() and
i40e_remove().

Change-ID: I3bc8b909c70fd2439334eeae994d151f61480985
Signed-off-by: John Underwood <johnx.underwood@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 05def9f..3ff3e83 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -9583,10 +9583,15 @@ vector_setup_out:
  **/
 static struct i40e_vsi *i40e_vsi_reinit_setup(struct i40e_vsi *vsi)
 {
-	struct i40e_pf *pf = vsi->back;
+	struct i40e_pf *pf;
 	u8 enabled_tc;
 	int ret;
 
+	if (!vsi)
+		return NULL;
+
+	pf = vsi->back;
+
 	i40e_put_lump(pf->qp_pile, vsi->base_queue, vsi->idx);
 	i40e_vsi_clear_rings(vsi);
 
-- 
cgit v0.10.2


From 3336514381f9ef99c50e5337ae1bf36f8138679d Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 17 Feb 2016 16:12:19 -0800
Subject: i40e: add adminq commands for Rx CTL registers

Add the new opcodes and struct used for asking the firmware to update Rx
control registers that need extra care when being accessed while under
heavy traffic - e.g. sustained 64byte packets at line rate on all ports.
The firmware will take extra steps to be sure the register accesses
are successful.

The registers involved are:
	PFQF_CTL_0
	PFQF_HENA
	PFQF_FDALLOC
	PFQF_HREGION
	PFLAN_QALLOC
	VPQF_CTL
	VFQF_HENA
	VFQF_HREGION
	VSIQF_CTL
	VSILAN_QBASE
	VSILAN_QTABLE
	VSIQF_TCREGION
	PFQF_HKEY
	VFQF_HKEY
	PRTQF_CTL_0
	GLFCOE_RCTL
	GLFCOE_RSOF
	GLQF_CTL
	GLQF_SWAP
	GLQF_HASH_MSK
	GLQF_HASH_INSET
	GLQF_HSYM
	GLQF_FC_MSK
	GLQF_FC_INSET
	GLQF_FD_MSK
	PRTQF_FD_INSET
	PRTQF_FD_FLXINSET
	PRTQF_FD_MSK

Change-ID: I56c8144000da66ad99f68948d8a184b2ec2aeb3e
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index bb7ecbb..8d5c65a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -146,6 +146,8 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_set_port_parameters	= 0x0203,
 	i40e_aqc_opc_get_switch_resource_alloc	= 0x0204,
 	i40e_aqc_opc_set_switch_config		= 0x0205,
+	i40e_aqc_opc_rx_ctl_reg_read		= 0x0206,
+	i40e_aqc_opc_rx_ctl_reg_write		= 0x0207,
 
 	i40e_aqc_opc_add_vsi			= 0x0210,
 	i40e_aqc_opc_update_vsi_parameters	= 0x0211,
@@ -696,6 +698,20 @@ struct i40e_aqc_set_switch_config {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config);
 
+/* Read Receive control registers  (direct 0x0206)
+ * Write Receive control registers (direct 0x0207)
+ *     used for accessing Rx control registers that can be
+ *     slow and need special handling when under high Rx load
+ */
+struct i40e_aqc_rx_ctl_reg_read_write {
+	__le32 reserved1;
+	__le32 address;
+	__le32 reserved2;
+	__le32 value;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_rx_ctl_reg_read_write);
+
 /* Add VSI (indirect 0x0210)
  *    this indirect command uses struct i40e_aqc_vsi_properties_data
  *    as the indirect buffer (128 bytes)
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index 815e481..aad8d62 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -146,6 +146,8 @@ enum i40e_admin_queue_opc {
 	i40e_aqc_opc_set_port_parameters	= 0x0203,
 	i40e_aqc_opc_get_switch_resource_alloc	= 0x0204,
 	i40e_aqc_opc_set_switch_config		= 0x0205,
+	i40e_aqc_opc_rx_ctl_reg_read		= 0x0206,
+	i40e_aqc_opc_rx_ctl_reg_write		= 0x0207,
 
 	i40e_aqc_opc_add_vsi			= 0x0210,
 	i40e_aqc_opc_update_vsi_parameters	= 0x0211,
@@ -693,6 +695,20 @@ struct i40e_aqc_set_switch_config {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config);
 
+/* Read Receive control registers  (direct 0x0206)
+ * Write Receive control registers (direct 0x0207)
+ *     used for accessing Rx control registers that can be
+ *     slow and need special handling when under high Rx load
+ */
+struct i40e_aqc_rx_ctl_reg_read_write {
+	__le32 reserved1;
+	__le32 address;
+	__le32 reserved2;
+	__le32 value;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_rx_ctl_reg_read_write);
+
 /* Add VSI (indirect 0x0210)
  *    this indirect command uses struct i40e_aqc_vsi_properties_data
  *    as the indirect buffer (128 bytes)
-- 
cgit v0.10.2


From f658137cbb1fddbe40ec7f1a2cebaf9dc9484ea7 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 17 Feb 2016 16:12:20 -0800
Subject: i40e: implement and use Rx CTL helper functions

Use the new AdminQ functions for safely accessing the Rx control
registers that may be affected by heavy small packet traffic.

Change-ID: Ibb00983e8dcba71f4b760222a609a5fcaa726f18
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 3a57e59..74d9599 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1328,7 +1328,7 @@ void i40e_clear_hw(struct i40e_hw *hw)
 	num_vf_int = (val & I40E_GLPCI_CNF2_MSI_X_VF_N_MASK) >>
 		     I40E_GLPCI_CNF2_MSI_X_VF_N_SHIFT;
 
-	val = rd32(hw, I40E_PFLAN_QALLOC);
+	val = i40e_read_rx_ctl(hw, I40E_PFLAN_QALLOC);
 	base_queue = (val & I40E_PFLAN_QALLOC_FIRSTQ_MASK) >>
 		     I40E_PFLAN_QALLOC_FIRSTQ_SHIFT;
 	j = (val & I40E_PFLAN_QALLOC_LASTQ_MASK) >>
@@ -3882,7 +3882,7 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw,
 		return ret;
 
 	/* Read the PF Queue Filter control register */
-	val = rd32(hw, I40E_PFQF_CTL_0);
+	val = i40e_read_rx_ctl(hw, I40E_PFQF_CTL_0);
 
 	/* Program required PE hash buckets for the PF */
 	val &= ~I40E_PFQF_CTL_0_PEHSIZE_MASK;
@@ -3919,7 +3919,7 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw,
 	if (settings->enable_macvlan)
 		val |= I40E_PFQF_CTL_0_MACVLAN_ENA_MASK;
 
-	wr32(hw, I40E_PFQF_CTL_0, val);
+	i40e_write_rx_ctl(hw, I40E_PFQF_CTL_0, val);
 
 	return 0;
 }
@@ -4575,3 +4575,125 @@ restore_config:
 					 phy_addr, led_ctl);
 	return status;
 }
+
+/**
+ * i40e_aq_rx_ctl_read_register - use FW to read from an Rx control register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ * @reg_val: ptr to register value
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Use the firmware to read the Rx control register,
+ * especially useful if the Rx unit is under heavy pressure
+ **/
+i40e_status i40e_aq_rx_ctl_read_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 *reg_val,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_rx_ctl_reg_read_write *cmd_resp =
+		(struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw;
+	i40e_status status;
+
+	if (!reg_val)
+		return I40E_ERR_PARAM;
+
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_rx_ctl_reg_read);
+
+	cmd_resp->address = cpu_to_le32(reg_addr);
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	if (status == 0)
+		*reg_val = le32_to_cpu(cmd_resp->value);
+
+	return status;
+}
+
+/**
+ * i40e_read_rx_ctl - read from an Rx control register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ **/
+u32 i40e_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr)
+{
+	i40e_status status = 0;
+	bool use_register;
+	int retry = 5;
+	u32 val = 0;
+
+	use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5);
+	if (!use_register) {
+do_retry:
+		status = i40e_aq_rx_ctl_read_register(hw, reg_addr, &val, NULL);
+		if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) {
+			usleep_range(1000, 2000);
+			retry--;
+			goto do_retry;
+		}
+	}
+
+	/* if the AQ access failed, try the old-fashioned way */
+	if (status || use_register)
+		val = rd32(hw, reg_addr);
+
+	return val;
+}
+
+/**
+ * i40e_aq_rx_ctl_write_register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ * @reg_val: register value
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Use the firmware to write to an Rx control register,
+ * especially useful if the Rx unit is under heavy pressure
+ **/
+i40e_status i40e_aq_rx_ctl_write_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 reg_val,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_rx_ctl_reg_read_write *cmd =
+		(struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw;
+	i40e_status status;
+
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_rx_ctl_reg_write);
+
+	cmd->address = cpu_to_le32(reg_addr);
+	cmd->value = cpu_to_le32(reg_val);
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
+ * i40e_write_rx_ctl - write to an Rx control register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ * @reg_val: register value
+ **/
+void i40e_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val)
+{
+	i40e_status status = 0;
+	bool use_register;
+	int retry = 5;
+
+	use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5);
+	if (!use_register) {
+do_retry:
+		status = i40e_aq_rx_ctl_write_register(hw, reg_addr,
+						       reg_val, NULL);
+		if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) {
+			usleep_range(1000, 2000);
+			retry--;
+			goto do_retry;
+		}
+	}
+
+	/* if the AQ access failed, try the old-fashioned way */
+	if (status || use_register)
+		wr32(hw, reg_addr, reg_val);
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index ca2f7ac..d51eee5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -342,6 +342,14 @@ i40e_status i40e_aq_debug_dump(struct i40e_hw *hw, u8 cluster_id,
 			       struct i40e_asq_cmd_details *cmd_details);
 void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw,
 						    u16 vsi_seid);
+i40e_status i40e_aq_rx_ctl_read_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 *reg_val,
+				struct i40e_asq_cmd_details *cmd_details);
+u32 i40e_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr);
+i40e_status i40e_aq_rx_ctl_write_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 reg_val,
+				struct i40e_asq_cmd_details *cmd_details);
+void i40e_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val);
 i40e_status i40e_read_phy_register(struct i40e_hw *hw, u8 page,
 				   u16 reg, u8 phy_addr, u16 *value);
 i40e_status i40e_write_phy_register(struct i40e_hw *hw, u8 page,
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_common.c b/drivers/net/ethernet/intel/i40evf/i40e_common.c
index 938783e..771ac6a 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_common.c
@@ -904,6 +904,131 @@ struct i40e_rx_ptype_decoded i40evf_ptype_lookup[] = {
 };
 
 /**
+ * i40evf_aq_rx_ctl_read_register - use FW to read from an Rx control register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ * @reg_val: ptr to register value
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Use the firmware to read the Rx control register,
+ * especially useful if the Rx unit is under heavy pressure
+ **/
+i40e_status i40evf_aq_rx_ctl_read_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 *reg_val,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_rx_ctl_reg_read_write *cmd_resp =
+		(struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw;
+	i40e_status status;
+
+	if (!reg_val)
+		return I40E_ERR_PARAM;
+
+	i40evf_fill_default_direct_cmd_desc(&desc,
+					    i40e_aqc_opc_rx_ctl_reg_read);
+
+	cmd_resp->address = cpu_to_le32(reg_addr);
+
+	status = i40evf_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	if (status == 0)
+		*reg_val = le32_to_cpu(cmd_resp->value);
+
+	return status;
+}
+
+/**
+ * i40evf_read_rx_ctl - read from an Rx control register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ **/
+u32 i40evf_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr)
+{
+	i40e_status status = 0;
+	bool use_register;
+	int retry = 5;
+	u32 val = 0;
+
+	use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5);
+	if (!use_register) {
+do_retry:
+		status = i40evf_aq_rx_ctl_read_register(hw, reg_addr,
+							&val, NULL);
+		if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) {
+			usleep_range(1000, 2000);
+			retry--;
+			goto do_retry;
+		}
+	}
+
+	/* if the AQ access failed, try the old-fashioned way */
+	if (status || use_register)
+		val = rd32(hw, reg_addr);
+
+	return val;
+}
+
+/**
+ * i40evf_aq_rx_ctl_write_register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ * @reg_val: register value
+ * @cmd_details: pointer to command details structure or NULL
+ *
+ * Use the firmware to write to an Rx control register,
+ * especially useful if the Rx unit is under heavy pressure
+ **/
+i40e_status i40evf_aq_rx_ctl_write_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 reg_val,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_rx_ctl_reg_read_write *cmd =
+		(struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw;
+	i40e_status status;
+
+	i40evf_fill_default_direct_cmd_desc(&desc,
+					    i40e_aqc_opc_rx_ctl_reg_write);
+
+	cmd->address = cpu_to_le32(reg_addr);
+	cmd->value = cpu_to_le32(reg_val);
+
+	status = i40evf_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
+ * i40evf_write_rx_ctl - write to an Rx control register
+ * @hw: pointer to the hw struct
+ * @reg_addr: register address
+ * @reg_val: register value
+ **/
+void i40evf_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val)
+{
+	i40e_status status = 0;
+	bool use_register;
+	int retry = 5;
+
+	use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5);
+	if (!use_register) {
+do_retry:
+		status = i40evf_aq_rx_ctl_write_register(hw, reg_addr,
+							 reg_val, NULL);
+		if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) {
+			usleep_range(1000, 2000);
+			retry--;
+			goto do_retry;
+		}
+	}
+
+	/* if the AQ access failed, try the old-fashioned way */
+	if (status || use_register)
+		wr32(hw, reg_addr, reg_val);
+}
+
+/**
  * i40e_aq_send_msg_to_pf
  * @hw: pointer to the hardware structure
  * @v_opcode: opcodes for VF-PF communication
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
index fa34d85..d89d521 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h
@@ -103,6 +103,14 @@ i40e_status i40e_aq_add_rem_control_packet_filter(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
 void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw,
 						    u16 vsi_seid);
+i40e_status i40evf_aq_rx_ctl_read_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 *reg_val,
+				struct i40e_asq_cmd_details *cmd_details);
+u32 i40evf_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr);
+i40e_status i40evf_aq_rx_ctl_write_register(struct i40e_hw *hw,
+				u32 reg_addr, u32 reg_val,
+				struct i40e_asq_cmd_details *cmd_details);
+void i40evf_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val);
 i40e_status i40e_read_phy_register(struct i40e_hw *hw, u8 page,
 				   u16 reg, u8 phy_addr, u16 *value);
 i40e_status i40e_write_phy_register(struct i40e_hw *hw, u8 page,
-- 
cgit v0.10.2


From 272cdaf2472ab7713deebe060bb90319b0382a94 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@intel.com>
Date: Wed, 17 Feb 2016 16:12:21 -0800
Subject: i40e: Use the new rx ctl register helpers. Don't use AQ calls from
 clear_hw.

Use the new AdminQ functions for safely accessing the Rx control
registers that may be affected by heavy small packet traffic.

We can't use AdminQ calls in i40e_clear_hw() because the HW is being
initialized and the AdminQ is not alive.  We recently added an AQ
related replacement for reading PFLAN_QALLOC, and this patch puts
back the original register read.

Change-ID: Ib027168c954a5733299aa3a4ce5f8218c6bb5636
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 74d9599..4596294 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1328,7 +1328,7 @@ void i40e_clear_hw(struct i40e_hw *hw)
 	num_vf_int = (val & I40E_GLPCI_CNF2_MSI_X_VF_N_MASK) >>
 		     I40E_GLPCI_CNF2_MSI_X_VF_N_SHIFT;
 
-	val = i40e_read_rx_ctl(hw, I40E_PFLAN_QALLOC);
+	val = rd32(hw, I40E_PFLAN_QALLOC);
 	base_queue = (val & I40E_PFLAN_QALLOC_FIRSTQ_MASK) >>
 		     I40E_PFLAN_QALLOC_FIRSTQ_SHIFT;
 	j = (val & I40E_PFLAN_QALLOC_LASTQ_MASK) >>
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index f1ad1016..a85bc94 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2181,8 +2181,8 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
 static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 {
 	struct i40e_hw *hw = &pf->hw;
-	u64 hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) |
-		   ((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32);
+	u64 hena = (u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0)) |
+		   ((u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)) << 32);
 
 	/* RSS does not support anything other than hashing
 	 * to queues on src and dst IPs and ports
@@ -2291,8 +2291,8 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
 		return -EINVAL;
 	}
 
-	wr32(hw, I40E_PFQF_HENA(0), (u32)hena);
-	wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (u32)hena);
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
 	i40e_flush(hw);
 
 	/* Save setting for future output/update */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
index 052df93..8ad162c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c
@@ -295,11 +295,11 @@ void i40e_init_pf_fcoe(struct i40e_pf *pf)
 	}
 
 	/* enable FCoE hash filter */
-	val = rd32(hw, I40E_PFQF_HENA(1));
+	val = i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1));
 	val |= BIT(I40E_FILTER_PCTYPE_FCOE_OX - 32);
 	val |= BIT(I40E_FILTER_PCTYPE_FCOE_RX - 32);
 	val &= I40E_PFQF_HENA_PTYPE_ENA_MASK;
-	wr32(hw, I40E_PFQF_HENA(1), val);
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), val);
 
 	/* enable flag */
 	pf->flags |= I40E_FLAG_FCOE_ENABLED;
@@ -317,11 +317,11 @@ void i40e_init_pf_fcoe(struct i40e_pf *pf)
 	pf->filter_settings.fcoe_cntx_num = I40E_DMA_CNTX_SIZE_4K;
 
 	/* Setup max frame with FCoE_MTU plus L2 overheads */
-	val = rd32(hw, I40E_GLFCOE_RCTL);
+	val = i40e_read_rx_ctl(hw, I40E_GLFCOE_RCTL);
 	val &= ~I40E_GLFCOE_RCTL_MAX_SIZE_MASK;
 	val |= ((FCOE_MTU + ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN)
 		 << I40E_GLFCOE_RCTL_MAX_SIZE_SHIFT);
-	wr32(hw, I40E_GLFCOE_RCTL, val);
+	i40e_write_rx_ctl(hw, I40E_GLFCOE_RCTL, val);
 
 	dev_info(&pf->pdev->dev, "FCoE is supported.\n");
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 3ff3e83..87325db 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8032,7 +8032,7 @@ static int i40e_config_rss_reg(struct i40e_vsi *vsi, const u8 *seed,
 		u32 *seed_dw = (u32 *)seed;
 
 		for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++)
-			wr32(hw, I40E_PFQF_HKEY(i), seed_dw[i]);
+			i40e_write_rx_ctl(hw, I40E_PFQF_HKEY(i), seed_dw[i]);
 	}
 
 	if (lut) {
@@ -8069,7 +8069,7 @@ static int i40e_get_rss_reg(struct i40e_vsi *vsi, u8 *seed,
 		u32 *seed_dw = (u32 *)seed;
 
 		for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++)
-			seed_dw[i] = rd32(hw, I40E_PFQF_HKEY(i));
+			seed_dw[i] = i40e_read_rx_ctl(hw, I40E_PFQF_HKEY(i));
 	}
 	if (lut) {
 		u32 *lut_dw = (u32 *)lut;
@@ -8152,19 +8152,19 @@ static int i40e_pf_config_rss(struct i40e_pf *pf)
 	int ret;
 
 	/* By default we enable TCP/UDP with IPv4/IPv6 ptypes */
-	hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) |
-		((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32);
+	hena = (u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0)) |
+		((u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)) << 32);
 	hena |= i40e_pf_get_default_rss_hena(pf);
 
-	wr32(hw, I40E_PFQF_HENA(0), (u32)hena);
-	wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (u32)hena);
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
 
 	/* Determine the RSS table size based on the hardware capabilities */
-	reg_val = rd32(hw, I40E_PFQF_CTL_0);
+	reg_val = i40e_read_rx_ctl(hw, I40E_PFQF_CTL_0);
 	reg_val = (pf->rss_table_size == 512) ?
 			(reg_val | I40E_PFQF_CTL_0_HASHLUTSIZE_512) :
 			(reg_val & ~I40E_PFQF_CTL_0_HASHLUTSIZE_512);
-	wr32(hw, I40E_PFQF_CTL_0, reg_val);
+	i40e_write_rx_ctl(hw, I40E_PFQF_CTL_0, reg_val);
 
 	/* Determine the RSS size of the VSI */
 	if (!vsi->rss_size)
@@ -11211,8 +11211,8 @@ static void i40e_remove(struct pci_dev *pdev)
 	i40e_ptp_stop(pf);
 
 	/* Disable RSS in hw */
-	wr32(hw, I40E_PFQF_HENA(0), 0);
-	wr32(hw, I40E_PFQF_HENA(1), 0);
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), 0);
+	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), 0);
 
 	/* no more scheduling of any task */
 	set_bit(__I40E_DOWN, &pf->state);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 5dcd198..93d8d98 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -602,8 +602,8 @@ static void i40e_enable_vf_mappings(struct i40e_vf *vf)
 	 * that VF queues be mapped using this method, even when they are
 	 * contiguous in real life
 	 */
-	wr32(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id),
-	     I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK);
+	i40e_write_rx_ctl(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id),
+			  I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK);
 
 	/* enable VF vplan_qtable mappings */
 	reg = I40E_VPLAN_MAPENA_TXRX_ENA_MASK;
@@ -630,7 +630,8 @@ static void i40e_enable_vf_mappings(struct i40e_vf *vf)
 						      (j * 2) + 1);
 			reg |= qid << 16;
 		}
-		wr32(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id), reg);
+		i40e_write_rx_ctl(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id),
+				  reg);
 	}
 
 	i40e_flush(hw);
-- 
cgit v0.10.2


From a4618ec88de95a86f290d01c74c506552f1a5d95 Mon Sep 17 00:00:00 2001
From: Pandi Kumar Maharajan <pandi.maharajan@intel.com>
Date: Thu, 18 Feb 2016 09:19:25 -0800
Subject: i40e: suspend scheduling during driver unload

We need to suspend scheduling or any pending service task during driver
unload process, so that new task will not be scheduled. This patch sets
the suspend flag bit during reload which avoids service task execution.

Change-ID: I017c57b5d6656564556e3c5387da671369a572ac
Signed-off-by: Pandi Kumar Maharajan <pandi.maharajan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 87325db..9076b29 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -11215,6 +11215,7 @@ static void i40e_remove(struct pci_dev *pdev)
 	i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), 0);
 
 	/* no more scheduling of any task */
+	set_bit(__I40E_SUSPENDED, &pf->state);
 	set_bit(__I40E_DOWN, &pf->state);
 	del_timer_sync(&pf->service_timer);
 	cancel_work_sync(&pf->service_task);
-- 
cgit v0.10.2


From 35f3472a750b3549f7f914ed96f41f0c2ca284f3 Mon Sep 17 00:00:00 2001
From: Mitch Williams <mitch.a.williams@intel.com>
Date: Wed, 17 Feb 2016 16:12:23 -0800
Subject: i40e: let go of the past

If we reset a VF, its VSI goes away, and it gets a new one. So don't
hang on to the now-stale local VSI pointer. It just leads to suffering
and kernel panics.

Change-ID: Ia8823b4e85893e95e963acee284968022b29177a
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 93d8d98..acd2693 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -2203,6 +2203,8 @@ int i40e_ndo_set_vf_port_vlan(struct net_device *netdev,
 		 * and then reloading the VF driver.
 		 */
 		i40e_vc_disable_vf(pf, vf);
+		/* During reset the VF got a new VSI, so refresh the pointer. */
+		vsi = pf->vsi[vf->lan_vsi_idx];
 	}
 
 	/* Check for condition where there was already a port VLAN ID
-- 
cgit v0.10.2


From b8f1343a13c669aaa3d475ed8513a32154ae5ffd Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed, 17 Feb 2016 16:12:24 -0800
Subject: i40e/i40evf: Bump i40e to 1.4.25 and i40evf to 1.4.15

Bump.

Change-ID: Ifa19aadaa892ad103f1b96fe2361fa690912c6a3
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 9076b29..3c8d8c4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -46,7 +46,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 15
+#define DRV_VERSION_BUILD 25
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 3396fe3..4b70aae 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -38,7 +38,7 @@ static const char i40evf_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 11
+#define DRV_VERSION_BUILD 15
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD) \
-- 
cgit v0.10.2


From 6bbd9a05a1f9839873a9290b5b7c6fafde8447ba Mon Sep 17 00:00:00 2001
From: Sasha Levin <sasha.levin@oracle.com>
Date: Fri, 19 Feb 2016 13:53:10 -0500
Subject: bpf: grab rcu read lock for bpf_percpu_hash_update

bpf_percpu_hash_update() expects rcu lock to be held and warns if it's not,
which pointed out a missing rcu read lock.

Fixes: 15a07b338 ("bpf: add lookup/update support for per-cpu hash and array maps")
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index fd5db8f..a68e951 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -619,7 +619,13 @@ out:
 int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value,
 			   u64 map_flags)
 {
-	return __htab_percpu_map_update_elem(map, key, value, map_flags, true);
+	int ret;
+
+	rcu_read_lock();
+	ret = __htab_percpu_map_update_elem(map, key, value, map_flags, true);
+	rcu_read_unlock();
+
+	return ret;
 }
 
 static const struct bpf_map_ops htab_percpu_ops = {
-- 
cgit v0.10.2


From 76cc173d48d9ed2206a1affe13b70d0e8e134010 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 16 Feb 2016 12:46:53 +0100
Subject: bridge: mdb: reduce the indentation level in br_mdb_fill_info

Switch the port check and skip if it's null, this allows us to reduce one
indentation level.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index ac089286..e666191 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -88,25 +88,26 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 			for (pp = &mp->ports;
 			     (p = rcu_dereference(*pp)) != NULL;
 			      pp = &p->next) {
+				struct br_mdb_entry e;
+
 				port = p->port;
-				if (port) {
-					struct br_mdb_entry e;
-					memset(&e, 0, sizeof(e));
-					e.ifindex = port->dev->ifindex;
-					e.vid = p->addr.vid;
-					__mdb_entry_fill_flags(&e, p->flags);
-					if (p->addr.proto == htons(ETH_P_IP))
-						e.addr.u.ip4 = p->addr.u.ip4;
+				if (!port)
+					continue;
+				memset(&e, 0, sizeof(e));
+				e.ifindex = port->dev->ifindex;
+				e.vid = p->addr.vid;
+				__mdb_entry_fill_flags(&e, p->flags);
+				if (p->addr.proto == htons(ETH_P_IP))
+					e.addr.u.ip4 = p->addr.u.ip4;
 #if IS_ENABLED(CONFIG_IPV6)
-					if (p->addr.proto == htons(ETH_P_IPV6))
-						e.addr.u.ip6 = p->addr.u.ip6;
+				if (p->addr.proto == htons(ETH_P_IPV6))
+					e.addr.u.ip6 = p->addr.u.ip6;
 #endif
-					e.addr.proto = p->addr.proto;
-					if (nla_put(skb, MDBA_MDB_ENTRY_INFO, sizeof(e), &e)) {
-						nla_nest_cancel(skb, nest2);
-						err = -EMSGSIZE;
-						goto out;
-					}
+				e.addr.proto = p->addr.proto;
+				if (nla_put(skb, MDBA_MDB_ENTRY_INFO, sizeof(e), &e)) {
+					nla_nest_cancel(skb, nest2);
+					err = -EMSGSIZE;
+					goto out;
 				}
 			}
 			nla_nest_end(skb, nest2);
-- 
cgit v0.10.2


From 2125715635053d4207a756a35aa718f548824e58 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Tue, 16 Feb 2016 12:46:54 +0100
Subject: bridge: mdb: add support for more attributes and export timer

Currently mdb entries are exported directly as a structure inside
MDBA_MDB_ENTRY_INFO attribute, we can't really extend it without
breaking user-space. In order to export new mdb fields, I've converted
the MDBA_MDB_ENTRY_INFO into a nested attribute which starts like before
with struct br_mdb_entry (without header, as it's casted directly in
iproute2) and continues with MDBA_MDB_EATTR_ attributes. This way we
keep compatibility with older users and can export new data.
I've tested this with iproute2, both with and without support for the
added attribute and it works fine.
So basically we again have MDBA_MDB_ENTRY_INFO with struct br_mdb_entry
inside but it may contain also some additional MDBA_MDB_EATTR_ attributes
such as MDBA_MDB_EATTR_TIMER which can be parsed by user-space.

So the new structure is:
[MDBA_MDB] = {
     [MDBA_MDB_ENTRY] = {
         [MDBA_MDB_ENTRY_INFO]
         [MDBA_MDB_ENTRY_INFO] { <- Nested attribute
             struct br_mdb_entry <- nla_put_nohdr()
             [MDBA_MDB_ENTRY attributes] <- normal netlink attributes
         }
     }
}

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index ec35472..0890b21 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -137,7 +137,10 @@ struct bridge_vlan_info {
 /* Bridge multicast database attributes
  * [MDBA_MDB] = {
  *     [MDBA_MDB_ENTRY] = {
- *         [MDBA_MDB_ENTRY_INFO]
+ *         [MDBA_MDB_ENTRY_INFO] {
+ *		struct br_mdb_entry
+ *		[MDBA_MDB_EATTR attributes]
+ *         }
  *     }
  * }
  * [MDBA_ROUTER] = {
@@ -166,6 +169,14 @@ enum {
 };
 #define MDBA_MDB_ENTRY_MAX (__MDBA_MDB_ENTRY_MAX - 1)
 
+/* per mdb entry additional attributes */
+enum {
+	MDBA_MDB_EATTR_UNSPEC,
+	MDBA_MDB_EATTR_TIMER,
+	__MDBA_MDB_EATTR_MAX
+};
+#define MDBA_MDB_EATTR_MAX (__MDBA_MDB_EATTR_MAX - 1)
+
 enum {
 	MDBA_ROUTER_UNSPEC,
 	MDBA_ROUTER_PORT,
diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index e666191..cf51b7b 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -88,11 +88,13 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 			for (pp = &mp->ports;
 			     (p = rcu_dereference(*pp)) != NULL;
 			      pp = &p->next) {
+				struct nlattr *nest_ent;
 				struct br_mdb_entry e;
 
 				port = p->port;
 				if (!port)
 					continue;
+
 				memset(&e, 0, sizeof(e));
 				e.ifindex = port->dev->ifindex;
 				e.vid = p->addr.vid;
@@ -104,11 +106,23 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 					e.addr.u.ip6 = p->addr.u.ip6;
 #endif
 				e.addr.proto = p->addr.proto;
-				if (nla_put(skb, MDBA_MDB_ENTRY_INFO, sizeof(e), &e)) {
+				nest_ent = nla_nest_start(skb,
+							  MDBA_MDB_ENTRY_INFO);
+				if (!nest_ent) {
+					nla_nest_cancel(skb, nest2);
+					err = -EMSGSIZE;
+					goto out;
+				}
+				if (nla_put_nohdr(skb, sizeof(e), &e) ||
+				    nla_put_u32(skb,
+						MDBA_MDB_EATTR_TIMER,
+						br_timer_value(&p->timer))) {
+					nla_nest_cancel(skb, nest_ent);
 					nla_nest_cancel(skb, nest2);
 					err = -EMSGSIZE;
 					goto out;
 				}
+				nla_nest_end(skb, nest_ent);
 			}
 			nla_nest_end(skb, nest2);
 		skip:
-- 
cgit v0.10.2


From 7c4a1d0cfdc169b746e61423ebf49b68133bc50a Mon Sep 17 00:00:00 2001
From: Sergio Prado <sergio.prado@e-labworks.com>
Date: Tue, 16 Feb 2016 21:10:45 -0200
Subject: net: macb: make magic-packet property generic

As requested by Rob Herring on patch
https://patchwork.ozlabs.org/patch/580862/.

This is a new property that it's still in net-next and has never been
used in production, so we are not breaking anything with the
incompatible binding change.

Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/macb.txt b/Documentation/devicetree/bindings/net/macb.txt
index c6b1cb5..b5a42df 100644
--- a/Documentation/devicetree/bindings/net/macb.txt
+++ b/Documentation/devicetree/bindings/net/macb.txt
@@ -25,7 +25,7 @@ Required properties:
 
 Optional properties for PHY child node:
 - reset-gpios : Should specify the gpio for phy reset
-- cdns,magic-packet : If present, indicates that the hardware supports waking
+- magic-packet : If present, indicates that the hardware supports waking
   up via magic packet.
 
 Examples:
diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c
index 69af049..7ccf229 100644
--- a/drivers/net/ethernet/cadence/macb.c
+++ b/drivers/net/ethernet/cadence/macb.c
@@ -2929,7 +2929,7 @@ static int macb_probe(struct platform_device *pdev)
 		bp->jumbo_max_len = macb_config->jumbo_max_len;
 
 	bp->wol = 0;
-	if (of_get_property(np, "cdns,magic-packet", NULL))
+	if (of_get_property(np, "magic-packet", NULL))
 		bp->wol |= MACB_WOL_HAS_MAGIC_PACKET;
 	device_init_wakeup(&pdev->dev, bp->wol & MACB_WOL_HAS_MAGIC_PACKET);
 
-- 
cgit v0.10.2


From 68f227930b887ef9f18065f15a17aa9dc1ac2d31 Mon Sep 17 00:00:00 2001
From: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Date: Thu, 18 Feb 2016 03:09:34 +0530
Subject: be2net: Fix pcie error recovery in case of NIC+RoCE adapters

Interrupts registered by RoCE driver are not unregistered when
msix interrupts are disabled during error recovery causing a
crash. Detach the adapter instance from RoCE driver when error
is detected to complete the cleanup. Attach the driver again after
the adapter is recovered from error.

Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 88f427c..4624846 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -5456,6 +5456,8 @@ static pci_ers_result_t be_eeh_err_detected(struct pci_dev *pdev,
 
 	dev_err(&adapter->pdev->dev, "EEH error detected\n");
 
+	be_roce_dev_remove(adapter);
+
 	if (!be_check_error(adapter, BE_ERROR_EEH)) {
 		be_set_error(adapter, BE_ERROR_EEH);
 
@@ -5520,6 +5522,8 @@ static void be_eeh_resume(struct pci_dev *pdev)
 	if (status)
 		goto err;
 
+	be_roce_dev_add(adapter);
+
 	be_schedule_err_detection(adapter, ERR_DETECTION_DELAY);
 	return;
 err:
-- 
cgit v0.10.2


From e0d8c1b7381133240f0c69eaf740f64975b6058e Mon Sep 17 00:00:00 2001
From: Wei Wang <weiwan@google.com>
Date: Wed, 17 Feb 2016 13:58:22 -0800
Subject: ipv6: pass up EMSGSIZE msg for UDP socket in Ipv6

In ipv4,  when  the machine receives a ICMP_FRAG_NEEDED message,  the
connected UDP socket will get EMSGSIZE message on its next read from the
socket.
However, this is not the case for ipv6.
This fix modifies the udp err handler in Ipv6 for ICMP6_PKT_TOOBIG to
make it similar to ipv4 behavior. That is when the machine gets an
ICMP6_PKT_TOOBIG message, the connected UDP socket will get EMSGSIZE
message on its next read from the socket.

Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index ac4e7e0..0711f8f 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -548,6 +548,7 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	const struct in6_addr *daddr = &hdr->daddr;
 	struct udphdr *uh = (struct udphdr *)(skb->data+offset);
 	struct sock *sk;
+	int harderr;
 	int err;
 	struct net *net = dev_net(skb->dev);
 
@@ -559,26 +560,27 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		return;
 	}
 
+	harderr = icmpv6_err_convert(type, code, &err);
+	np = inet6_sk(sk);
+
 	if (type == ICMPV6_PKT_TOOBIG) {
 		if (!ip6_sk_accept_pmtu(sk))
 			goto out;
 		ip6_sk_update_pmtu(skb, sk, info);
+		if (np->pmtudisc != IPV6_PMTUDISC_DONT)
+			harderr = 1;
 	}
 	if (type == NDISC_REDIRECT) {
 		ip6_sk_redirect(skb, sk);
 		goto out;
 	}
 
-	np = inet6_sk(sk);
-
-	if (!icmpv6_err_convert(type, code, &err) && !np->recverr)
-		goto out;
-
-	if (sk->sk_state != TCP_ESTABLISHED && !np->recverr)
-		goto out;
-
-	if (np->recverr)
+	if (!np->recverr) {
+		if (!harderr || sk->sk_state != TCP_ESTABLISHED)
+			goto out;
+	} else {
 		ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1));
+	}
 
 	sk->sk_err = err;
 	sk->sk_error_report(sk);
-- 
cgit v0.10.2


From 76d13b568776fdef646c70878de406c90aef46a9 Mon Sep 17 00:00:00 2001
From: "sixiao@microsoft.com" <sixiao@microsoft.com>
Date: Wed, 17 Feb 2016 16:43:59 -0800
Subject: hv_netvsc: add software transmit timestamp support

Enable skb_tx_timestamp in hyperv netvsc.

Signed-off-by: Simon Xiao <sixiao@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index c72e5b8..202e2b1 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -550,6 +550,8 @@ do_send:
 	packet->page_buf_cnt = init_page_array(rndis_msg, rndis_msg_size,
 					       skb, packet, &pb);
 
+	/* timestamp packet in software */
+	skb_tx_timestamp(skb);
 	ret = netvsc_send(net_device_ctx->device_ctx, packet,
 			  rndis_msg, &pb, skb);
 
@@ -920,6 +922,7 @@ static const struct ethtool_ops ethtool_ops = {
 	.get_link	= ethtool_op_get_link,
 	.get_channels   = netvsc_get_channels,
 	.set_channels   = netvsc_set_channels,
+	.get_ts_info	= ethtool_op_get_ts_info,
 };
 
 static const struct net_device_ops device_ops = {
-- 
cgit v0.10.2


From e52bc7c28ac9f54db6f86b19ed65c599def18c98 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Fri, 19 Feb 2016 09:23:59 -0500
Subject: lib/bitmap.c: conversion routines to/from u32 array

Aimed at transferring bitmaps to/from user-space in a 32/64-bit agnostic
way.

Tested:
  unit tests (next patch) on qemu i386, x86_64, ppc, ppc64 BE and LE,
  ARM.

Signed-off-by: David Decotigny <decot@googlers.com>
Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h
index 9653fdb..e9b0b9a 100644
--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -59,6 +59,8 @@
  * bitmap_find_free_region(bitmap, bits, order)	Find and allocate bit region
  * bitmap_release_region(bitmap, pos, order)	Free specified bit region
  * bitmap_allocate_region(bitmap, pos, order)	Allocate specified bit region
+ * bitmap_from_u32array(dst, nbits, buf, nwords) *dst = *buf (nwords 32b words)
+ * bitmap_to_u32array(buf, nwords, src, nbits)	*buf = *dst (nwords 32b words)
  */
 
 /*
@@ -163,6 +165,14 @@ extern void bitmap_fold(unsigned long *dst, const unsigned long *orig,
 extern int bitmap_find_free_region(unsigned long *bitmap, unsigned int bits, int order);
 extern void bitmap_release_region(unsigned long *bitmap, unsigned int pos, int order);
 extern int bitmap_allocate_region(unsigned long *bitmap, unsigned int pos, int order);
+extern unsigned int bitmap_from_u32array(unsigned long *bitmap,
+					 unsigned int nbits,
+					 const u32 *buf,
+					 unsigned int nwords);
+extern unsigned int bitmap_to_u32array(u32 *buf,
+				       unsigned int nwords,
+				       const unsigned long *bitmap,
+				       unsigned int nbits);
 #ifdef __BIG_ENDIAN
 extern void bitmap_copy_le(unsigned long *dst, const unsigned long *src, unsigned int nbits);
 #else
diff --git a/lib/bitmap.c b/lib/bitmap.c
index 8148143..c66da50 100644
--- a/lib/bitmap.c
+++ b/lib/bitmap.c
@@ -12,6 +12,8 @@
 #include <linux/bitmap.h>
 #include <linux/bitops.h>
 #include <linux/bug.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
 
 #include <asm/page.h>
 #include <asm/uaccess.h>
@@ -1060,6 +1062,93 @@ int bitmap_allocate_region(unsigned long *bitmap, unsigned int pos, int order)
 EXPORT_SYMBOL(bitmap_allocate_region);
 
 /**
+ * bitmap_from_u32array - copy the contents of a u32 array of bits to bitmap
+ *	@bitmap: array of unsigned longs, the destination bitmap, non NULL
+ *	@nbits: number of bits in @bitmap
+ *	@buf: array of u32 (in host byte order), the source bitmap, non NULL
+ *	@nwords: number of u32 words in @buf
+ *
+ * copy min(nbits, 32*nwords) bits from @buf to @bitmap, remaining
+ * bits between nword and nbits in @bitmap (if any) are cleared. In
+ * last word of @bitmap, the bits beyond nbits (if any) are kept
+ * unchanged.
+ *
+ * Return the number of bits effectively copied.
+ */
+unsigned int
+bitmap_from_u32array(unsigned long *bitmap, unsigned int nbits,
+		     const u32 *buf, unsigned int nwords)
+{
+	unsigned int dst_idx, src_idx;
+
+	for (src_idx = dst_idx = 0; dst_idx < BITS_TO_LONGS(nbits); ++dst_idx) {
+		unsigned long part = 0;
+
+		if (src_idx < nwords)
+			part = buf[src_idx++];
+
+#if BITS_PER_LONG == 64
+		if (src_idx < nwords)
+			part |= ((unsigned long) buf[src_idx++]) << 32;
+#endif
+
+		if (dst_idx < nbits/BITS_PER_LONG)
+			bitmap[dst_idx] = part;
+		else {
+			unsigned long mask = BITMAP_LAST_WORD_MASK(nbits);
+
+			bitmap[dst_idx] = (bitmap[dst_idx] & ~mask)
+				| (part & mask);
+		}
+	}
+
+	return min_t(unsigned int, nbits, 32*nwords);
+}
+EXPORT_SYMBOL(bitmap_from_u32array);
+
+/**
+ * bitmap_to_u32array - copy the contents of bitmap to a u32 array of bits
+ *	@buf: array of u32 (in host byte order), the dest bitmap, non NULL
+ *	@nwords: number of u32 words in @buf
+ *	@bitmap: array of unsigned longs, the source bitmap, non NULL
+ *	@nbits: number of bits in @bitmap
+ *
+ * copy min(nbits, 32*nwords) bits from @bitmap to @buf. Remaining
+ * bits after nbits in @buf (if any) are cleared.
+ *
+ * Return the number of bits effectively copied.
+ */
+unsigned int
+bitmap_to_u32array(u32 *buf, unsigned int nwords,
+		   const unsigned long *bitmap, unsigned int nbits)
+{
+	unsigned int dst_idx = 0, src_idx = 0;
+
+	while (dst_idx < nwords) {
+		unsigned long part = 0;
+
+		if (src_idx < BITS_TO_LONGS(nbits)) {
+			part = bitmap[src_idx];
+			if (src_idx >= nbits/BITS_PER_LONG)
+				part &= BITMAP_LAST_WORD_MASK(nbits);
+			src_idx++;
+		}
+
+		buf[dst_idx++] = part & 0xffffffffUL;
+
+#if BITS_PER_LONG == 64
+		if (dst_idx < nwords) {
+			part >>= 32;
+			buf[dst_idx++] = part & 0xffffffffUL;
+		}
+#endif
+	}
+
+	return min_t(unsigned int, nbits, 32*nwords);
+}
+EXPORT_SYMBOL(bitmap_to_u32array);
+
+/**
  * bitmap_copy_le - copy a bitmap, putting the bits into little-endian order.
  * @dst:   destination buffer
  * @src:   bitmap to copy
-- 
cgit v0.10.2


From 5fd003f56c2c584b62a0486ad25bbd4be02b8b6c Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Fri, 19 Feb 2016 09:24:00 -0500
Subject: test_bitmap: unit tests for lib/bitmap.c

This is mainly testing bitmap construction and conversion to/from u32[]
for now.

Tested:
  qemu i386, x86_64, ppc, ppc64 BE and LE, ARM.

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index ecb9e75..f890ee5 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1738,6 +1738,14 @@ config TEST_KSTRTOX
 config TEST_PRINTF
 	tristate "Test printf() family of functions at runtime"
 
+config TEST_BITMAP
+	tristate "Test bitmap_*() family of functions at runtime"
+	default n
+	help
+	  Enable this option to test the bitmap functions at boot.
+
+	  If unsure, say N.
+
 config TEST_RHASHTABLE
 	tristate "Perform selftest on resizable hash table"
 	default n
diff --git a/lib/Makefile b/lib/Makefile
index a7c26a4..dda4039 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -43,6 +43,7 @@ obj-$(CONFIG_TEST_USER_COPY) += test_user_copy.o
 obj-$(CONFIG_TEST_STATIC_KEYS) += test_static_keys.o
 obj-$(CONFIG_TEST_STATIC_KEYS) += test_static_key_base.o
 obj-$(CONFIG_TEST_PRINTF) += test_printf.o
+obj-$(CONFIG_TEST_BITMAP) += test_bitmap.o
 
 ifeq ($(CONFIG_DEBUG_KOBJECT),y)
 CFLAGS_kobject.o += -DDEBUG
diff --git a/lib/test_bitmap.c b/lib/test_bitmap.c
new file mode 100644
index 0000000..e2cbd43
--- /dev/null
+++ b/lib/test_bitmap.c
@@ -0,0 +1,358 @@
+/*
+ * Test cases for printf facility.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/bitmap.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/printk.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+
+static unsigned total_tests __initdata;
+static unsigned failed_tests __initdata;
+
+static char pbl_buffer[PAGE_SIZE] __initdata;
+
+
+static bool __init
+__check_eq_uint(const char *srcfile, unsigned int line,
+		const unsigned int exp_uint, unsigned int x)
+{
+	if (exp_uint != x) {
+		pr_warn("[%s:%u] expected %u, got %u\n",
+			srcfile, line, exp_uint, x);
+		return false;
+	}
+	return true;
+}
+
+
+static bool __init
+__check_eq_bitmap(const char *srcfile, unsigned int line,
+		  const unsigned long *exp_bmap, unsigned int exp_nbits,
+		  const unsigned long *bmap, unsigned int nbits)
+{
+	if (exp_nbits != nbits) {
+		pr_warn("[%s:%u] bitmap length mismatch: expected %u, got %u\n",
+			srcfile, line, exp_nbits, nbits);
+		return false;
+	}
+
+	if (!bitmap_equal(exp_bmap, bmap, nbits)) {
+		pr_warn("[%s:%u] bitmaps contents differ: expected \"%*pbl\", got \"%*pbl\"\n",
+			srcfile, line,
+			exp_nbits, exp_bmap, nbits, bmap);
+		return false;
+	}
+	return true;
+}
+
+static bool __init
+__check_eq_pbl(const char *srcfile, unsigned int line,
+	       const char *expected_pbl,
+	       const unsigned long *bitmap, unsigned int nbits)
+{
+	snprintf(pbl_buffer, sizeof(pbl_buffer), "%*pbl", nbits, bitmap);
+	if (strcmp(expected_pbl, pbl_buffer)) {
+		pr_warn("[%s:%u] expected \"%s\", got \"%s\"\n",
+			srcfile, line,
+			expected_pbl, pbl_buffer);
+		return false;
+	}
+	return true;
+}
+
+static bool __init
+__check_eq_u32_array(const char *srcfile, unsigned int line,
+		     const u32 *exp_arr, unsigned int exp_len,
+		     const u32 *arr, unsigned int len)
+{
+	if (exp_len != len) {
+		pr_warn("[%s:%u] array length differ: expected %u, got %u\n",
+			srcfile, line,
+			exp_len, len);
+		return false;
+	}
+
+	if (memcmp(exp_arr, arr, len*sizeof(*arr))) {
+		pr_warn("[%s:%u] array contents differ\n", srcfile, line);
+		print_hex_dump(KERN_WARNING, "  exp:  ", DUMP_PREFIX_OFFSET,
+			       32, 4, exp_arr, exp_len*sizeof(*exp_arr), false);
+		print_hex_dump(KERN_WARNING, "  got:  ", DUMP_PREFIX_OFFSET,
+			       32, 4, arr, len*sizeof(*arr), false);
+		return false;
+	}
+
+	return true;
+}
+
+#define __expect_eq(suffix, ...)					\
+	({								\
+		int result = 0;						\
+		total_tests++;						\
+		if (!__check_eq_ ## suffix(__FILE__, __LINE__,		\
+					   ##__VA_ARGS__)) {		\
+			failed_tests++;					\
+			result = 1;					\
+		}							\
+		result;							\
+	})
+
+#define expect_eq_uint(...)		__expect_eq(uint, ##__VA_ARGS__)
+#define expect_eq_bitmap(...)		__expect_eq(bitmap, ##__VA_ARGS__)
+#define expect_eq_pbl(...)		__expect_eq(pbl, ##__VA_ARGS__)
+#define expect_eq_u32_array(...)	__expect_eq(u32_array, ##__VA_ARGS__)
+
+static void __init test_zero_fill_copy(void)
+{
+	DECLARE_BITMAP(bmap1, 1024);
+	DECLARE_BITMAP(bmap2, 1024);
+
+	bitmap_zero(bmap1, 1024);
+	bitmap_zero(bmap2, 1024);
+
+	/* single-word bitmaps */
+	expect_eq_pbl("", bmap1, 23);
+
+	bitmap_fill(bmap1, 19);
+	expect_eq_pbl("0-18", bmap1, 1024);
+
+	bitmap_copy(bmap2, bmap1, 23);
+	expect_eq_pbl("0-18", bmap2, 1024);
+
+	bitmap_fill(bmap2, 23);
+	expect_eq_pbl("0-22", bmap2, 1024);
+
+	bitmap_copy(bmap2, bmap1, 23);
+	expect_eq_pbl("0-18", bmap2, 1024);
+
+	bitmap_zero(bmap1, 23);
+	expect_eq_pbl("", bmap1, 1024);
+
+	/* multi-word bitmaps */
+	bitmap_zero(bmap1, 1024);
+	expect_eq_pbl("", bmap1, 1024);
+
+	bitmap_fill(bmap1, 109);
+	expect_eq_pbl("0-108", bmap1, 1024);
+
+	bitmap_copy(bmap2, bmap1, 1024);
+	expect_eq_pbl("0-108", bmap2, 1024);
+
+	bitmap_fill(bmap2, 1024);
+	expect_eq_pbl("0-1023", bmap2, 1024);
+
+	bitmap_copy(bmap2, bmap1, 1024);
+	expect_eq_pbl("0-108", bmap2, 1024);
+
+	/* the following tests assume a 32- or 64-bit arch (even 128b
+	 * if we care)
+	 */
+
+	bitmap_fill(bmap2, 1024);
+	bitmap_copy(bmap2, bmap1, 109);  /* ... but 0-padded til word length */
+	expect_eq_pbl("0-108,128-1023", bmap2, 1024);
+
+	bitmap_fill(bmap2, 1024);
+	bitmap_copy(bmap2, bmap1, 97);  /* ... but aligned on word length */
+	expect_eq_pbl("0-108,128-1023", bmap2, 1024);
+
+	bitmap_zero(bmap2, 97);  /* ... but 0-padded til word length */
+	expect_eq_pbl("128-1023", bmap2, 1024);
+}
+
+static void __init test_bitmap_u32_array_conversions(void)
+{
+	DECLARE_BITMAP(bmap1, 1024);
+	DECLARE_BITMAP(bmap2, 1024);
+	u32 exp_arr[32], arr[32];
+	unsigned nbits;
+
+	for (nbits = 0 ; nbits < 257 ; ++nbits) {
+		const unsigned int used_u32s = DIV_ROUND_UP(nbits, 32);
+		unsigned int i, rv;
+
+		bitmap_zero(bmap1, nbits);
+		bitmap_set(bmap1, nbits, 1024 - nbits);  /* garbage */
+
+		memset(arr, 0xff, sizeof(arr));
+		rv = bitmap_to_u32array(arr, used_u32s, bmap1, nbits);
+		expect_eq_uint(nbits, rv);
+
+		memset(exp_arr, 0xff, sizeof(exp_arr));
+		memset(exp_arr, 0, used_u32s*sizeof(*exp_arr));
+		expect_eq_u32_array(exp_arr, 32, arr, 32);
+
+		bitmap_fill(bmap2, 1024);
+		rv = bitmap_from_u32array(bmap2, nbits, arr, used_u32s);
+		expect_eq_uint(nbits, rv);
+		expect_eq_bitmap(bmap1, 1024, bmap2, 1024);
+
+		for (i = 0 ; i < nbits ; ++i) {
+			/*
+			 * test conversion bitmap -> u32[]
+			 */
+
+			bitmap_zero(bmap1, 1024);
+			__set_bit(i, bmap1);
+			bitmap_set(bmap1, nbits, 1024 - nbits);  /* garbage */
+
+			memset(arr, 0xff, sizeof(arr));
+			rv = bitmap_to_u32array(arr, used_u32s, bmap1, nbits);
+			expect_eq_uint(nbits, rv);
+
+			/* 1st used u32 words contain expected bit set, the
+			 * remaining words are left unchanged (0xff)
+			 */
+			memset(exp_arr, 0xff, sizeof(exp_arr));
+			memset(exp_arr, 0, used_u32s*sizeof(*exp_arr));
+			exp_arr[i/32] = (1U<<(i%32));
+			expect_eq_u32_array(exp_arr, 32, arr, 32);
+
+
+			/* same, with longer array to fill
+			 */
+			memset(arr, 0xff, sizeof(arr));
+			rv = bitmap_to_u32array(arr, 32, bmap1, nbits);
+			expect_eq_uint(nbits, rv);
+
+			/* 1st used u32 words contain expected bit set, the
+			 * remaining words are all 0s
+			 */
+			memset(exp_arr, 0, sizeof(exp_arr));
+			exp_arr[i/32] = (1U<<(i%32));
+			expect_eq_u32_array(exp_arr, 32, arr, 32);
+
+			/*
+			 * test conversion u32[] -> bitmap
+			 */
+
+			/* the 1st nbits of bmap2 are identical to
+			 * bmap1, the remaining bits of bmap2 are left
+			 * unchanged (all 1s)
+			 */
+			bitmap_fill(bmap2, 1024);
+			rv = bitmap_from_u32array(bmap2, nbits,
+						  exp_arr, used_u32s);
+			expect_eq_uint(nbits, rv);
+
+			expect_eq_bitmap(bmap1, 1024, bmap2, 1024);
+
+			/* same, with more bits to fill
+			 */
+			memset(arr, 0xff, sizeof(arr));  /* garbage */
+			memset(arr, 0, used_u32s*sizeof(u32));
+			arr[i/32] = (1U<<(i%32));
+
+			bitmap_fill(bmap2, 1024);
+			rv = bitmap_from_u32array(bmap2, 1024, arr, used_u32s);
+			expect_eq_uint(used_u32s*32, rv);
+
+			/* the 1st nbits of bmap2 are identical to
+			 * bmap1, the remaining bits of bmap2 are cleared
+			 */
+			bitmap_zero(bmap1, 1024);
+			__set_bit(i, bmap1);
+			expect_eq_bitmap(bmap1, 1024, bmap2, 1024);
+
+
+			/*
+			 * test short conversion bitmap -> u32[] (1
+			 * word too short)
+			 */
+			if (used_u32s > 1) {
+				bitmap_zero(bmap1, 1024);
+				__set_bit(i, bmap1);
+				bitmap_set(bmap1, nbits,
+					   1024 - nbits);  /* garbage */
+				memset(arr, 0xff, sizeof(arr));
+
+				rv = bitmap_to_u32array(arr, used_u32s - 1,
+							bmap1, nbits);
+				expect_eq_uint((used_u32s - 1)*32, rv);
+
+				/* 1st used u32 words contain expected
+				 * bit set, the remaining words are
+				 * left unchanged (0xff)
+				 */
+				memset(exp_arr, 0xff, sizeof(exp_arr));
+				memset(exp_arr, 0,
+				       (used_u32s-1)*sizeof(*exp_arr));
+				if ((i/32) < (used_u32s - 1))
+					exp_arr[i/32] = (1U<<(i%32));
+				expect_eq_u32_array(exp_arr, 32, arr, 32);
+			}
+
+			/*
+			 * test short conversion u32[] -> bitmap (3
+			 * bits too short)
+			 */
+			if (nbits > 3) {
+				memset(arr, 0xff, sizeof(arr));  /* garbage */
+				memset(arr, 0, used_u32s*sizeof(*arr));
+				arr[i/32] = (1U<<(i%32));
+
+				bitmap_zero(bmap1, 1024);
+				rv = bitmap_from_u32array(bmap1, nbits - 3,
+							  arr, used_u32s);
+				expect_eq_uint(nbits - 3, rv);
+
+				/* we are expecting the bit < nbits -
+				 * 3 (none otherwise), and the rest of
+				 * bmap1 unchanged (0-filled)
+				 */
+				bitmap_zero(bmap2, 1024);
+				if (i < nbits - 3)
+					__set_bit(i, bmap2);
+				expect_eq_bitmap(bmap2, 1024, bmap1, 1024);
+
+				/* do the same with bmap1 initially
+				 * 1-filled
+				 */
+
+				bitmap_fill(bmap1, 1024);
+				rv = bitmap_from_u32array(bmap1, nbits - 3,
+							 arr, used_u32s);
+				expect_eq_uint(nbits - 3, rv);
+
+				/* we are expecting the bit < nbits -
+				 * 3 (none otherwise), and the rest of
+				 * bmap1 unchanged (1-filled)
+				 */
+				bitmap_zero(bmap2, 1024);
+				if (i < nbits - 3)
+					__set_bit(i, bmap2);
+				bitmap_set(bmap2, nbits-3, 1024 - nbits + 3);
+				expect_eq_bitmap(bmap2, 1024, bmap1, 1024);
+			}
+		}
+	}
+}
+
+static int __init test_bitmap_init(void)
+{
+	test_zero_fill_copy();
+	test_bitmap_u32_array_conversions();
+
+	if (failed_tests == 0)
+		pr_info("all %u tests passed\n", total_tests);
+	else
+		pr_warn("failed %u out of %u tests\n",
+			failed_tests, total_tests);
+
+	return failed_tests ? -EINVAL : 0;
+}
+
+static void __exit test_bitmap_cleanup(void)
+{
+}
+
+module_init(test_bitmap_init);
+module_exit(test_bitmap_cleanup);
+
+MODULE_AUTHOR("david decotigny <david.decotigny@googlers.com>");
+MODULE_LICENSE("GPL");
diff --git a/tools/testing/selftests/lib/Makefile b/tools/testing/selftests/lib/Makefile
index 47147b9..0836006 100644
--- a/tools/testing/selftests/lib/Makefile
+++ b/tools/testing/selftests/lib/Makefile
@@ -3,6 +3,6 @@
 # No binaries, but make sure arg-less "make" doesn't trigger "run_tests"
 all:
 
-TEST_PROGS := printf.sh
+TEST_PROGS := printf.sh bitmap.sh
 
 include ../lib.mk
diff --git a/tools/testing/selftests/lib/bitmap.sh b/tools/testing/selftests/lib/bitmap.sh
new file mode 100755
index 0000000..2da187b
--- /dev/null
+++ b/tools/testing/selftests/lib/bitmap.sh
@@ -0,0 +1,10 @@
+#!/bin/sh
+# Runs bitmap infrastructure tests using test_bitmap kernel module
+
+if /sbin/modprobe -q test_bitmap; then
+	/sbin/modprobe -q -r test_bitmap
+	echo "bitmap: ok"
+else
+	echo "bitmap: [FAIL]"
+	exit 1
+fi
-- 
cgit v0.10.2


From ac2c7ad0e5d6030452c9af2fafd192e17fd04264 Mon Sep 17 00:00:00 2001
From: Kan Liang <kan.liang@intel.com>
Date: Fri, 19 Feb 2016 09:24:01 -0500
Subject: net/ethtool: introduce a new ioctl for per queue setting

Introduce a new ioctl ETHTOOL_PERQUEUE for per queue parameters setting.
The following patches will enable some SUB_COMMANDs for per queue
setting.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 190aea0..f15ae02 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -1202,6 +1202,21 @@ enum ethtool_sfeatures_retval_bits {
 #define ETHTOOL_F_WISH          (1 << ETHTOOL_F_WISH__BIT)
 #define ETHTOOL_F_COMPAT        (1 << ETHTOOL_F_COMPAT__BIT)
 
+#define MAX_NUM_QUEUE		4096
+
+/**
+ * struct ethtool_per_queue_op - apply sub command to the queues in mask.
+ * @cmd: ETHTOOL_PERQUEUE
+ * @sub_command: the sub command which apply to each queues
+ * @queue_mask: Bitmap of the queues which sub command apply to
+ * @data: A complete command structure following for each of the queues addressed
+ */
+struct ethtool_per_queue_op {
+	__u32	cmd;
+	__u32	sub_command;
+	__u32	queue_mask[DIV_ROUND_UP(MAX_NUM_QUEUE, 32)];
+	char	data[];
+};
 
 /* CMDs currently supported */
 #define ETHTOOL_GSET		0x00000001 /* Get settings. */
@@ -1285,6 +1300,8 @@ enum ethtool_sfeatures_retval_bits {
 #define ETHTOOL_STUNABLE	0x00000049 /* Set tunable configuration */
 #define ETHTOOL_GPHYSTATS	0x0000004a /* get PHY-specific statistics */
 
+#define ETHTOOL_PERQUEUE	0x0000004b /* Set per queue options */
+
 /* compatibility with older code */
 #define SPARC_ETH_GSET		ETHTOOL_GSET
 #define SPARC_ETH_SSET		ETHTOOL_SSET
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index c2d3118..d640ecf 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -1888,13 +1888,27 @@ out:
 	return ret;
 }
 
+static int ethtool_set_per_queue(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_per_queue_op per_queue_opt;
+
+	if (copy_from_user(&per_queue_opt, useraddr, sizeof(per_queue_opt)))
+		return -EFAULT;
+
+	switch (per_queue_opt.sub_command) {
+
+	default:
+		return -EOPNOTSUPP;
+	};
+}
+
 /* The main entry point in this file.  Called from net/core/dev_ioctl.c */
 
 int dev_ethtool(struct net *net, struct ifreq *ifr)
 {
 	struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name);
 	void __user *useraddr = ifr->ifr_data;
-	u32 ethcmd;
+	u32 ethcmd, sub_cmd;
 	int rc;
 	netdev_features_t old_features;
 
@@ -1904,8 +1918,14 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	if (copy_from_user(&ethcmd, useraddr, sizeof(ethcmd)))
 		return -EFAULT;
 
+	if (ethcmd == ETHTOOL_PERQUEUE) {
+		if (copy_from_user(&sub_cmd, useraddr + sizeof(ethcmd), sizeof(sub_cmd)))
+			return -EFAULT;
+	} else {
+		sub_cmd = ethcmd;
+	}
 	/* Allow some commands to be done by anyone */
-	switch (ethcmd) {
+	switch (sub_cmd) {
 	case ETHTOOL_GSET:
 	case ETHTOOL_GDRVINFO:
 	case ETHTOOL_GMSGLVL:
@@ -2135,6 +2155,9 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_GPHYSTATS:
 		rc = ethtool_get_phy_stats(dev, useraddr);
 		break;
+	case ETHTOOL_PERQUEUE:
+		rc = ethtool_set_per_queue(dev, useraddr);
+		break;
 	default:
 		rc = -EOPNOTSUPP;
 	}
-- 
cgit v0.10.2


From 421797b1aa363cb897f29f7d365e068dc9d9db81 Mon Sep 17 00:00:00 2001
From: Kan Liang <kan.liang@intel.com>
Date: Fri, 19 Feb 2016 09:24:02 -0500
Subject: net/ethtool: support get coalesce per queue

This patch implements sub command ETHTOOL_GCOALESCE for ioctl
ETHTOOL_PERQUEUE. It introduces an interface get_per_queue_coalesce to
get coalesce of each masked queue from device driver. Then the interrupt
coalescing parameters will be copied back to user space one by one.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 653dc9c..de56600 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -201,6 +201,11 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
  * @get_module_eeprom: Get the eeprom information from the plug-in module
  * @get_eee: Get Energy-Efficient (EEE) supported and status.
  * @set_eee: Set EEE status (enable/disable) as well as LPI timers.
+ * @get_per_queue_coalesce: Get interrupt coalescing parameters per queue.
+ *	It must check that the given queue number is valid. If neither a RX nor
+ *	a TX queue has this number, return -EINVAL. If only a RX queue or a TX
+ *	queue has this number, set the inapplicable fields to ~0 and return 0.
+ *	Returns a negative error code or zero.
  *
  * All operations are optional (i.e. the function pointer may be set
  * to %NULL) and callers must take this into account.  Callers must
@@ -279,7 +284,8 @@ struct ethtool_ops {
 			       const struct ethtool_tunable *, void *);
 	int	(*set_tunable)(struct net_device *,
 			       const struct ethtool_tunable *, const void *);
-
+	int	(*get_per_queue_coalesce)(struct net_device *, u32,
+					  struct ethtool_coalesce *);
 
 };
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index d640ecf..2a6c3a2 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -1888,6 +1888,38 @@ out:
 	return ret;
 }
 
+static int ethtool_get_per_queue_coalesce(struct net_device *dev,
+					  void __user *useraddr,
+					  struct ethtool_per_queue_op *per_queue_opt)
+{
+	u32 bit;
+	int ret;
+	DECLARE_BITMAP(queue_mask, MAX_NUM_QUEUE);
+
+	if (!dev->ethtool_ops->get_per_queue_coalesce)
+		return -EOPNOTSUPP;
+
+	useraddr += sizeof(*per_queue_opt);
+
+	bitmap_from_u32array(queue_mask,
+			     MAX_NUM_QUEUE,
+			     per_queue_opt->queue_mask,
+			     DIV_ROUND_UP(MAX_NUM_QUEUE, 32));
+
+	for_each_set_bit(bit, queue_mask, MAX_NUM_QUEUE) {
+		struct ethtool_coalesce coalesce = { .cmd = ETHTOOL_GCOALESCE };
+
+		ret = dev->ethtool_ops->get_per_queue_coalesce(dev, bit, &coalesce);
+		if (ret != 0)
+			return ret;
+		if (copy_to_user(useraddr, &coalesce, sizeof(coalesce)))
+			return -EFAULT;
+		useraddr += sizeof(coalesce);
+	}
+
+	return 0;
+}
+
 static int ethtool_set_per_queue(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_per_queue_op per_queue_opt;
@@ -1896,7 +1928,8 @@ static int ethtool_set_per_queue(struct net_device *dev, void __user *useraddr)
 		return -EFAULT;
 
 	switch (per_queue_opt.sub_command) {
-
+	case ETHTOOL_GCOALESCE:
+		return ethtool_get_per_queue_coalesce(dev, useraddr, &per_queue_opt);
 	default:
 		return -EOPNOTSUPP;
 	};
-- 
cgit v0.10.2


From f38d138a7da6510a1184e3bc5f425deb187c3265 Mon Sep 17 00:00:00 2001
From: Kan Liang <kan.liang@intel.com>
Date: Fri, 19 Feb 2016 09:24:03 -0500
Subject: net/ethtool: support set coalesce per queue

This patch implements sub command ETHTOOL_SCOALESCE for ioctl
ETHTOOL_PERQUEUE. It introduces an interface set_per_queue_coalesce to
set coalesce of each masked queue to device driver. The wanted coalesce
information are stored in "data" for each masked queue, which can copy
from userspace.
If it fails to set coalesce to device driver, the value which already
set to specific queue will be tried to rollback.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index de56600..472d7d7 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -206,6 +206,11 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
  *	a TX queue has this number, return -EINVAL. If only a RX queue or a TX
  *	queue has this number, set the inapplicable fields to ~0 and return 0.
  *	Returns a negative error code or zero.
+ * @set_per_queue_coalesce: Set interrupt coalescing parameters per queue.
+ *	It must check that the given queue number is valid. If neither a RX nor
+ *	a TX queue has this number, return -EINVAL. If only a RX queue or a TX
+ *	queue has this number, ignore the inapplicable fields.
+ *	Returns a negative error code or zero.
  *
  * All operations are optional (i.e. the function pointer may be set
  * to %NULL) and callers must take this into account.  Callers must
@@ -286,6 +291,8 @@ struct ethtool_ops {
 			       const struct ethtool_tunable *, const void *);
 	int	(*get_per_queue_coalesce)(struct net_device *, u32,
 					  struct ethtool_coalesce *);
+	int	(*set_per_queue_coalesce)(struct net_device *, u32,
+					  struct ethtool_coalesce *);
 
 };
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 2a6c3a2..2406101 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -1920,6 +1920,65 @@ static int ethtool_get_per_queue_coalesce(struct net_device *dev,
 	return 0;
 }
 
+static int ethtool_set_per_queue_coalesce(struct net_device *dev,
+					  void __user *useraddr,
+					  struct ethtool_per_queue_op *per_queue_opt)
+{
+	u32 bit;
+	int i, ret = 0;
+	int n_queue;
+	struct ethtool_coalesce *backup = NULL, *tmp = NULL;
+	DECLARE_BITMAP(queue_mask, MAX_NUM_QUEUE);
+
+	if ((!dev->ethtool_ops->set_per_queue_coalesce) ||
+	    (!dev->ethtool_ops->get_per_queue_coalesce))
+		return -EOPNOTSUPP;
+
+	useraddr += sizeof(*per_queue_opt);
+
+	bitmap_from_u32array(queue_mask,
+			     MAX_NUM_QUEUE,
+			     per_queue_opt->queue_mask,
+			     DIV_ROUND_UP(MAX_NUM_QUEUE, 32));
+	n_queue = bitmap_weight(queue_mask, MAX_NUM_QUEUE);
+	tmp = backup = kmalloc_array(n_queue, sizeof(*backup), GFP_KERNEL);
+	if (!backup)
+		return -ENOMEM;
+
+	for_each_set_bit(bit, queue_mask, MAX_NUM_QUEUE) {
+		struct ethtool_coalesce coalesce;
+
+		ret = dev->ethtool_ops->get_per_queue_coalesce(dev, bit, tmp);
+		if (ret != 0)
+			goto roll_back;
+
+		tmp++;
+
+		if (copy_from_user(&coalesce, useraddr, sizeof(coalesce))) {
+			ret = -EFAULT;
+			goto roll_back;
+		}
+
+		ret = dev->ethtool_ops->set_per_queue_coalesce(dev, bit, &coalesce);
+		if (ret != 0)
+			goto roll_back;
+
+		useraddr += sizeof(coalesce);
+	}
+
+roll_back:
+	if (ret != 0) {
+		tmp = backup;
+		for_each_set_bit(i, queue_mask, bit) {
+			dev->ethtool_ops->set_per_queue_coalesce(dev, i, tmp);
+			tmp++;
+		}
+	}
+	kfree(backup);
+
+	return ret;
+}
+
 static int ethtool_set_per_queue(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_per_queue_op per_queue_opt;
@@ -1930,6 +1989,8 @@ static int ethtool_set_per_queue(struct net_device *dev, void __user *useraddr)
 	switch (per_queue_opt.sub_command) {
 	case ETHTOOL_GCOALESCE:
 		return ethtool_get_per_queue_coalesce(dev, useraddr, &per_queue_opt);
+	case ETHTOOL_SCOALESCE:
+		return ethtool_set_per_queue_coalesce(dev, useraddr, &per_queue_opt);
 	default:
 		return -EOPNOTSUPP;
 	};
-- 
cgit v0.10.2


From a75e8005d506f374554b17383c39aa82db0ea860 Mon Sep 17 00:00:00 2001
From: Kan Liang <kan.liang@intel.com>
Date: Fri, 19 Feb 2016 09:24:04 -0500
Subject: i40e: queue-specific settings for interrupt moderation

For i40e driver, each vector has its own ITR register. However, there
are no concept of queue-specific settings in the driver proper. Only
global variable is used to store ITR values. That will cause problems
especially when resetting the vector. The specific ITR values could be
lost.
This patch move rx_itr_setting and tx_itr_setting to i40e_ring to store
specific ITR register for each queue.
i40e_get_coalesce and i40e_set_coalesce are also modified accordingly to
support queue-specific settings. To make it compatible with old ethtool,
if user doesn't specify the queue number, i40e_get_coalesce will return
queue 0's value. While i40e_set_coalesce will apply value to all queues.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index e99be9f..2f6210a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -521,13 +521,6 @@ struct i40e_vsi {
 	struct i40e_ring **tx_rings;
 
 	u16 work_limit;
-	/* high bit set means dynamic, use accessor routines to read/write.
-	 * hardware only supports 2us resolution for the ITR registers.
-	 * these values always store the USER setting, and must be converted
-	 * before programming to a register.
-	 */
-	u16 rx_itr_setting;
-	u16 tx_itr_setting;
 	u16 int_rate_limit;  /* value in usecs */
 
 	u16 rss_table_size; /* HW RSS table size */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 2a44f2e..0c97733 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -302,6 +302,10 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 "    rx_rings[%i]: vsi = %p, q_vector = %p\n",
 			 i, rx_ring->vsi,
 			 rx_ring->q_vector);
+		dev_info(&pf->pdev->dev,
+			 "    rx_rings[%i]: rx_itr_setting = %d (%s)\n",
+			 i, rx_ring->rx_itr_setting,
+			 ITR_IS_DYNAMIC(rx_ring->rx_itr_setting) ? "dynamic" : "fixed");
 	}
 	for (i = 0; i < vsi->num_queue_pairs; i++) {
 		struct i40e_ring *tx_ring = ACCESS_ONCE(vsi->tx_rings[i]);
@@ -352,14 +356,15 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 		dev_info(&pf->pdev->dev,
 			 "    tx_rings[%i]: DCB tc = %d\n",
 			 i, tx_ring->dcb_tc);
+		dev_info(&pf->pdev->dev,
+			 "    tx_rings[%i]: tx_itr_setting = %d (%s)\n",
+			 i, tx_ring->tx_itr_setting,
+			 ITR_IS_DYNAMIC(tx_ring->tx_itr_setting) ? "dynamic" : "fixed");
 	}
 	rcu_read_unlock();
 	dev_info(&pf->pdev->dev,
-		 "    work_limit = %d, rx_itr_setting = %d (%s), tx_itr_setting = %d (%s)\n",
-		 vsi->work_limit, vsi->rx_itr_setting,
-		 ITR_IS_DYNAMIC(vsi->rx_itr_setting) ? "dynamic" : "fixed",
-		 vsi->tx_itr_setting,
-		 ITR_IS_DYNAMIC(vsi->tx_itr_setting) ? "dynamic" : "fixed");
+		 "    work_limit = %d\n",
+		 vsi->work_limit);
 	dev_info(&pf->pdev->dev,
 		 "    max_frame = %d, rx_hdr_len = %d, rx_buf_len = %d dtype = %d\n",
 		 vsi->max_frame, vsi->rx_hdr_len, vsi->rx_buf_len, vsi->dtype);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index a85bc94..a470599 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1879,8 +1879,9 @@ static int i40e_set_phys_id(struct net_device *netdev,
  * 125us (8000 interrupts per second) == ITR(62)
  */
 
-static int i40e_get_coalesce(struct net_device *netdev,
-			     struct ethtool_coalesce *ec)
+static int __i40e_get_coalesce(struct net_device *netdev,
+			       struct ethtool_coalesce *ec,
+			       int queue)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
@@ -1888,14 +1889,24 @@ static int i40e_get_coalesce(struct net_device *netdev,
 	ec->tx_max_coalesced_frames_irq = vsi->work_limit;
 	ec->rx_max_coalesced_frames_irq = vsi->work_limit;
 
-	if (ITR_IS_DYNAMIC(vsi->rx_itr_setting))
+	/* rx and tx usecs has per queue value. If user doesn't specify the queue,
+	 * return queue 0's value to represent.
+	 */
+	if (queue < 0) {
+		queue = 0;
+	} else if (queue >= vsi->num_queue_pairs) {
+		return -EINVAL;
+	}
+
+	if (ITR_IS_DYNAMIC(vsi->rx_rings[queue]->rx_itr_setting))
 		ec->use_adaptive_rx_coalesce = 1;
 
-	if (ITR_IS_DYNAMIC(vsi->tx_itr_setting))
+	if (ITR_IS_DYNAMIC(vsi->tx_rings[queue]->tx_itr_setting))
 		ec->use_adaptive_tx_coalesce = 1;
 
-	ec->rx_coalesce_usecs = vsi->rx_itr_setting & ~I40E_ITR_DYNAMIC;
-	ec->tx_coalesce_usecs = vsi->tx_itr_setting & ~I40E_ITR_DYNAMIC;
+	ec->rx_coalesce_usecs = vsi->rx_rings[queue]->rx_itr_setting & ~I40E_ITR_DYNAMIC;
+	ec->tx_coalesce_usecs = vsi->tx_rings[queue]->tx_itr_setting & ~I40E_ITR_DYNAMIC;
+
 	/* we use the _usecs_high to store/set the interrupt rate limit
 	 * that the hardware supports, that almost but not quite
 	 * fits the original intent of the ethtool variable,
@@ -1908,15 +1919,57 @@ static int i40e_get_coalesce(struct net_device *netdev,
 	return 0;
 }
 
-static int i40e_set_coalesce(struct net_device *netdev,
+static int i40e_get_coalesce(struct net_device *netdev,
 			     struct ethtool_coalesce *ec)
 {
-	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	return __i40e_get_coalesce(netdev, ec, -1);
+}
+
+static void i40e_set_itr_per_queue(struct i40e_vsi *vsi,
+				   struct ethtool_coalesce *ec,
+				   int queue)
+{
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
 	struct i40e_q_vector *q_vector;
+	u16 vector, intrl;
+
+	intrl = INTRL_USEC_TO_REG(vsi->int_rate_limit);
+
+	vsi->rx_rings[queue]->rx_itr_setting = ec->rx_coalesce_usecs;
+	vsi->tx_rings[queue]->tx_itr_setting = ec->tx_coalesce_usecs;
+
+	if (ec->use_adaptive_rx_coalesce)
+		vsi->rx_rings[queue]->rx_itr_setting |= I40E_ITR_DYNAMIC;
+	else
+		vsi->rx_rings[queue]->rx_itr_setting &= ~I40E_ITR_DYNAMIC;
+
+	if (ec->use_adaptive_tx_coalesce)
+		vsi->tx_rings[queue]->tx_itr_setting |= I40E_ITR_DYNAMIC;
+	else
+		vsi->tx_rings[queue]->tx_itr_setting &= ~I40E_ITR_DYNAMIC;
+
+	q_vector = vsi->rx_rings[queue]->q_vector;
+	q_vector->rx.itr = ITR_TO_REG(vsi->rx_rings[queue]->rx_itr_setting);
+	vector = vsi->base_vector + q_vector->v_idx;
+	wr32(hw, I40E_PFINT_ITRN(I40E_RX_ITR, vector - 1), q_vector->rx.itr);
+
+	q_vector = vsi->tx_rings[queue]->q_vector;
+	q_vector->tx.itr = ITR_TO_REG(vsi->tx_rings[queue]->tx_itr_setting);
+	vector = vsi->base_vector + q_vector->v_idx;
+	wr32(hw, I40E_PFINT_ITRN(I40E_TX_ITR, vector - 1), q_vector->tx.itr);
+
+	wr32(hw, I40E_PFINT_RATEN(vector - 1), intrl);
+	i40e_flush(hw);
+}
+
+static int __i40e_set_coalesce(struct net_device *netdev,
+			       struct ethtool_coalesce *ec,
+			       int queue)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
-	struct i40e_hw *hw = &pf->hw;
-	u16 vector;
 	int i;
 
 	if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq)
@@ -1933,59 +1986,49 @@ static int i40e_set_coalesce(struct net_device *netdev,
 		return -EINVAL;
 	}
 
-	vector = vsi->base_vector;
-	if ((ec->rx_coalesce_usecs >= (I40E_MIN_ITR << 1)) &&
-	    (ec->rx_coalesce_usecs <= (I40E_MAX_ITR << 1))) {
-		vsi->rx_itr_setting = ec->rx_coalesce_usecs;
-	} else if (ec->rx_coalesce_usecs == 0) {
-		vsi->rx_itr_setting = ec->rx_coalesce_usecs;
+	if (ec->rx_coalesce_usecs == 0) {
 		if (ec->use_adaptive_rx_coalesce)
 			netif_info(pf, drv, netdev, "rx-usecs=0, need to disable adaptive-rx for a complete disable\n");
-	} else {
-		netif_info(pf, drv, netdev, "Invalid value, rx-usecs range is 0-8160\n");
-		return -EINVAL;
+	} else if ((ec->rx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
+		   (ec->rx_coalesce_usecs > (I40E_MAX_ITR << 1))) {
+			netif_info(pf, drv, netdev, "Invalid value, rx-usecs range is 0-8160\n");
+			return -EINVAL;
 	}
 
 	vsi->int_rate_limit = ec->rx_coalesce_usecs_high;
 
-	if ((ec->tx_coalesce_usecs >= (I40E_MIN_ITR << 1)) &&
-	    (ec->tx_coalesce_usecs <= (I40E_MAX_ITR << 1))) {
-		vsi->tx_itr_setting = ec->tx_coalesce_usecs;
-	} else if (ec->tx_coalesce_usecs == 0) {
-		vsi->tx_itr_setting = ec->tx_coalesce_usecs;
+	if (ec->tx_coalesce_usecs == 0) {
 		if (ec->use_adaptive_tx_coalesce)
 			netif_info(pf, drv, netdev, "tx-usecs=0, need to disable adaptive-tx for a complete disable\n");
-	} else {
-		netif_info(pf, drv, netdev,
-			   "Invalid value, tx-usecs range is 0-8160\n");
-		return -EINVAL;
+	} else if ((ec->tx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
+		   (ec->tx_coalesce_usecs > (I40E_MAX_ITR << 1))) {
+			netif_info(pf, drv, netdev, "Invalid value, tx-usecs range is 0-8160\n");
+			return -EINVAL;
 	}
 
-	if (ec->use_adaptive_rx_coalesce)
-		vsi->rx_itr_setting |= I40E_ITR_DYNAMIC;
-	else
-		vsi->rx_itr_setting &= ~I40E_ITR_DYNAMIC;
-
-	if (ec->use_adaptive_tx_coalesce)
-		vsi->tx_itr_setting |= I40E_ITR_DYNAMIC;
-	else
-		vsi->tx_itr_setting &= ~I40E_ITR_DYNAMIC;
-
-	for (i = 0; i < vsi->num_q_vectors; i++, vector++) {
-		u16 intrl = INTRL_USEC_TO_REG(vsi->int_rate_limit);
-
-		q_vector = vsi->q_vectors[i];
-		q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
-		wr32(hw, I40E_PFINT_ITRN(0, vector - 1), q_vector->rx.itr);
-		q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
-		wr32(hw, I40E_PFINT_ITRN(1, vector - 1), q_vector->tx.itr);
-		wr32(hw, I40E_PFINT_RATEN(vector - 1), intrl);
-		i40e_flush(hw);
+	/* rx and tx usecs has per queue value. If user doesn't specify the queue,
+	 * apply to all queues.
+	 */
+	if (queue < 0) {
+		for (i = 0; i < vsi->num_queue_pairs; i++)
+			i40e_set_itr_per_queue(vsi, ec, i);
+	} else if (queue < vsi->num_queue_pairs) {
+		i40e_set_itr_per_queue(vsi, ec, queue);
+	} else {
+		netif_info(pf, drv, netdev, "Invalid queue value, queue range is 0 - %d\n",
+			   vsi->num_queue_pairs - 1);
+		return -EINVAL;
 	}
 
 	return 0;
 }
 
+static int i40e_set_coalesce(struct net_device *netdev,
+			     struct ethtool_coalesce *ec)
+{
+	return __i40e_set_coalesce(netdev, ec, -1);
+}
+
 /**
  * i40e_get_rss_hash_opts - Get RSS hash Input Set for each flow type
  * @pf: pointer to the physical function struct
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 3c8d8c4..70d9605 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3124,11 +3124,11 @@ static void i40e_vsi_configure_msix(struct i40e_vsi *vsi)
 		struct i40e_q_vector *q_vector = vsi->q_vectors[i];
 
 		q_vector->itr_countdown = ITR_COUNTDOWN_START;
-		q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
+		q_vector->rx.itr = ITR_TO_REG(vsi->rx_rings[i]->rx_itr_setting);
 		q_vector->rx.latency_range = I40E_LOW_LATENCY;
 		wr32(hw, I40E_PFINT_ITRN(I40E_RX_ITR, vector - 1),
 		     q_vector->rx.itr);
-		q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
+		q_vector->tx.itr = ITR_TO_REG(vsi->tx_rings[i]->tx_itr_setting);
 		q_vector->tx.latency_range = I40E_LOW_LATENCY;
 		wr32(hw, I40E_PFINT_ITRN(I40E_TX_ITR, vector - 1),
 		     q_vector->tx.itr);
@@ -3220,10 +3220,10 @@ static void i40e_configure_msi_and_legacy(struct i40e_vsi *vsi)
 
 	/* set the ITR configuration */
 	q_vector->itr_countdown = ITR_COUNTDOWN_START;
-	q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
+	q_vector->rx.itr = ITR_TO_REG(vsi->rx_rings[0]->rx_itr_setting);
 	q_vector->rx.latency_range = I40E_LOW_LATENCY;
 	wr32(hw, I40E_PFINT_ITR0(I40E_RX_ITR), q_vector->rx.itr);
-	q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
+	q_vector->tx.itr = ITR_TO_REG(vsi->tx_rings[0]->tx_itr_setting);
 	q_vector->tx.latency_range = I40E_LOW_LATENCY;
 	wr32(hw, I40E_PFINT_ITR0(I40E_TX_ITR), q_vector->tx.itr);
 
@@ -7322,8 +7322,6 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type)
 	set_bit(__I40E_DOWN, &vsi->state);
 	vsi->flags = 0;
 	vsi->idx = vsi_idx;
-	vsi->rx_itr_setting = pf->rx_itr_default;
-	vsi->tx_itr_setting = pf->tx_itr_default;
 	vsi->int_rate_limit = 0;
 	vsi->rss_table_size = (vsi->type == I40E_VSI_MAIN) ?
 				pf->rss_table_size : 64;
@@ -7490,6 +7488,7 @@ static int i40e_alloc_rings(struct i40e_vsi *vsi)
 		tx_ring->dcb_tc = 0;
 		if (vsi->back->flags & I40E_FLAG_WB_ON_ITR_CAPABLE)
 			tx_ring->flags = I40E_TXR_FLAGS_WB_ON_ITR;
+		tx_ring->tx_itr_setting = pf->tx_itr_default;
 		vsi->tx_rings[i] = tx_ring;
 
 		rx_ring = &tx_ring[1];
@@ -7506,6 +7505,7 @@ static int i40e_alloc_rings(struct i40e_vsi *vsi)
 			set_ring_16byte_desc_enabled(rx_ring);
 		else
 			clear_ring_16byte_desc_enabled(rx_ring);
+		rx_ring->rx_itr_setting = pf->rx_itr_default;
 		vsi->rx_rings[i] = rx_ring;
 	}
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index cb52f39..084d0ab 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1882,6 +1882,7 @@ static inline void i40e_update_enable_itr(struct i40e_vsi *vsi,
 	bool rx = false, tx = false;
 	u32 rxval, txval;
 	int vector;
+	int idx = q_vector->v_idx;
 
 	vector = (q_vector->v_idx + vsi->base_vector);
 
@@ -1891,17 +1892,17 @@ static inline void i40e_update_enable_itr(struct i40e_vsi *vsi,
 	rxval = txval = i40e_buildreg_itr(I40E_ITR_NONE, 0);
 
 	if (q_vector->itr_countdown > 0 ||
-	    (!ITR_IS_DYNAMIC(vsi->rx_itr_setting) &&
-	     !ITR_IS_DYNAMIC(vsi->tx_itr_setting))) {
+	    (!ITR_IS_DYNAMIC(vsi->rx_rings[idx]->rx_itr_setting) &&
+	     !ITR_IS_DYNAMIC(vsi->tx_rings[idx]->tx_itr_setting))) {
 		goto enable_int;
 	}
 
-	if (ITR_IS_DYNAMIC(vsi->rx_itr_setting)) {
+	if (ITR_IS_DYNAMIC(vsi->rx_rings[idx]->rx_itr_setting)) {
 		rx = i40e_set_new_dynamic_itr(&q_vector->rx);
 		rxval = i40e_buildreg_itr(I40E_RX_ITR, q_vector->rx.itr);
 	}
 
-	if (ITR_IS_DYNAMIC(vsi->tx_itr_setting)) {
+	if (ITR_IS_DYNAMIC(vsi->tx_rings[idx]->tx_itr_setting)) {
 		tx = i40e_set_new_dynamic_itr(&q_vector->tx);
 		txval = i40e_buildreg_itr(I40E_TX_ITR, q_vector->tx.itr);
 	}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 5600970..cdd5dc0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -248,6 +248,14 @@ struct i40e_ring {
 	u8 dcb_tc;			/* Traffic class of ring */
 	u8 __iomem *tail;
 
+	/* high bit set means dynamic, use accessor routines to read/write.
+	 * hardware only supports 2us resolution for the ITR registers.
+	 * these values always store the USER setting, and must be converted
+	 * before programming to a register.
+	 */
+	u16 rx_itr_setting;
+	u16 tx_itr_setting;
+
 	u16 count;			/* Number of descriptors */
 	u16 reg_idx;			/* HW register index of the ring */
 	u16 rx_hdr_len;
-- 
cgit v0.10.2


From be280bad15fafc0f7e7b90bdbd99170366f5e9bf Mon Sep 17 00:00:00 2001
From: Kan Liang <kan.liang@intel.com>
Date: Fri, 19 Feb 2016 09:24:05 -0500
Subject: i40e/ethtool: support coalesce getting by queue

This patch implements get_per_queue_coalesce for i40e driver.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index a470599..dd572ab 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1925,6 +1925,12 @@ static int i40e_get_coalesce(struct net_device *netdev,
 	return __i40e_get_coalesce(netdev, ec, -1);
 }
 
+static int i40e_get_per_queue_coalesce(struct net_device *netdev, u32 queue,
+				       struct ethtool_coalesce *ec)
+{
+	return __i40e_get_coalesce(netdev, ec, queue);
+}
+
 static void i40e_set_itr_per_queue(struct i40e_vsi *vsi,
 				   struct ethtool_coalesce *ec,
 				   int queue)
@@ -2914,6 +2920,7 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 	.get_ts_info		= i40e_get_ts_info,
 	.get_priv_flags		= i40e_get_priv_flags,
 	.set_priv_flags		= i40e_set_priv_flags,
+	.get_per_queue_coalesce	= i40e_get_per_queue_coalesce,
 };
 
 void i40e_set_ethtool_ops(struct net_device *netdev)
-- 
cgit v0.10.2


From f3757a4d9e823c43ccfe4db02b6cda77414e25f7 Mon Sep 17 00:00:00 2001
From: Kan Liang <kan.liang@intel.com>
Date: Fri, 19 Feb 2016 09:24:06 -0500
Subject: i40e/ethtool: support coalesce setting by queue

This patch implements set_per_queue_coalesce for i40e driver.

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index dd572ab..784b165 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2035,6 +2035,12 @@ static int i40e_set_coalesce(struct net_device *netdev,
 	return __i40e_set_coalesce(netdev, ec, -1);
 }
 
+static int i40e_set_per_queue_coalesce(struct net_device *netdev, u32 queue,
+				       struct ethtool_coalesce *ec)
+{
+	return __i40e_set_coalesce(netdev, ec, queue);
+}
+
 /**
  * i40e_get_rss_hash_opts - Get RSS hash Input Set for each flow type
  * @pf: pointer to the physical function struct
@@ -2921,6 +2927,7 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 	.get_priv_flags		= i40e_get_priv_flags,
 	.set_priv_flags		= i40e_set_priv_flags,
 	.get_per_queue_coalesce	= i40e_get_per_queue_coalesce,
+	.set_per_queue_coalesce	= i40e_set_per_queue_coalesce,
 };
 
 void i40e_set_ethtool_ops(struct net_device *netdev)
-- 
cgit v0.10.2


From 321b4d4bd12f90e7497c9ab057aafcc2649aa902 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 20 Feb 2016 00:35:29 +0100
Subject: phy: marvell/micrel: Fix Unpossible condition

commit 2b2427d06426 ("phy: micrel: Add ethtool statistics counters")
from Dec 30, 2015, leads to the following static checker
warning:

        drivers/net/phy/micrel.c:609 kszphy_get_stat()
        warn: unsigned 'val' is never less than zero.

drivers/net/phy/micrel.c
   602  static u64 kszphy_get_stat(struct phy_device *phydev, int i)
   603  {
   604          struct kszphy_hw_stat stat = kszphy_hw_stats[i];
   605          struct kszphy_priv *priv = phydev->priv;
   606          u64 val;
   607
   608          val = phy_read(phydev, stat.reg);
   609          if (val < 0) {
                    ^^^^^^^
Unpossible!

   610                  val = UINT64_MAX;
   611          } else {
   612                  val = val & ((1 << stat.bits) - 1);
   613                  priv->stats[i] += val;
   614                  val = priv->stats[i];
   615          }
   616
   617          return val;
   618  }

The same problem exists in the Marvell driver. Fix both.

Fixes: 2b2427d06426 ("phy: micrel: Add ethtool statistics counters")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Reported-by: Julia.Lawall <julia.lawall@lip6.fr>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 1dcbd3f..d0168f1 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -1065,8 +1065,8 @@ static u64 marvell_get_stat(struct phy_device *phydev, int i)
 {
 	struct marvell_hw_stat stat = marvell_hw_stats[i];
 	struct marvell_priv *priv = phydev->priv;
-	int err, oldpage;
-	u64 val;
+	int err, oldpage, val;
+	u64 ret;
 
 	oldpage = phy_read(phydev, MII_MARVELL_PHY_PAGE);
 	err = phy_write(phydev, MII_MARVELL_PHY_PAGE,
@@ -1076,16 +1076,16 @@ static u64 marvell_get_stat(struct phy_device *phydev, int i)
 
 	val = phy_read(phydev, stat.reg);
 	if (val < 0) {
-		val = UINT64_MAX;
+		ret = UINT64_MAX;
 	} else {
 		val = val & ((1 << stat.bits) - 1);
 		priv->stats[i] += val;
-		val = priv->stats[i];
+		ret = priv->stats[i];
 	}
 
 	phy_write(phydev, MII_MARVELL_PHY_PAGE, oldpage);
 
-	return val;
+	return ret;
 }
 
 static void marvell_get_stats(struct phy_device *phydev,
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 03833db..48219c8 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -612,18 +612,19 @@ static u64 kszphy_get_stat(struct phy_device *phydev, int i)
 {
 	struct kszphy_hw_stat stat = kszphy_hw_stats[i];
 	struct kszphy_priv *priv = phydev->priv;
-	u64 val;
+	int val;
+	u64 ret;
 
 	val = phy_read(phydev, stat.reg);
 	if (val < 0) {
-		val = UINT64_MAX;
+		ret = UINT64_MAX;
 	} else {
 		val = val & ((1 << stat.bits) - 1);
 		priv->stats[i] += val;
-		val = priv->stats[i];
+		ret = priv->stats[i];
 	}
 
-	return val;
+	return ret;
 }
 
 static void kszphy_get_stats(struct phy_device *phydev,
-- 
cgit v0.10.2


From 6b83d28a55a891a9d70fc61ccb1c138e47dcbe74 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Sat, 20 Feb 2016 00:29:30 +0100
Subject: net: use skb_postpush_rcsum instead of own implementations

Replace individual implementations with the recently introduced
skb_postpush_rcsum() helper.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Tom Herbert <tom@herbertland.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index a5bd067..8bd4b79 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4496,9 +4496,7 @@ int skb_vlan_push(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci)
 		skb->mac_len += VLAN_HLEN;
 		__skb_pull(skb, offset);
 
-		if (skb->ip_summed == CHECKSUM_COMPLETE)
-			skb->csum = csum_add(skb->csum, csum_partial(skb->data
-					+ (2 * ETH_ALEN), VLAN_HLEN, 0));
+		skb_postpush_rcsum(skb, skb->data + (2 * ETH_ALEN), VLAN_HLEN);
 	}
 	__vlan_hwaccel_put_tag(skb, vlan_proto, vlan_tci);
 	return 0;
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 18f3498..e2ea311 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -496,10 +496,8 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
 	IP6CB(head)->flags |= IP6SKB_FRAGMENTED;
 
 	/* Yes, and fold redundant checksum back. 8) */
-	if (head->ip_summed == CHECKSUM_COMPLETE)
-		head->csum = csum_partial(skb_network_header(head),
-					  skb_network_header_len(head),
-					  head->csum);
+	skb_postpush_rcsum(head, skb_network_header(head),
+			   skb_network_header_len(head));
 
 	rcu_read_lock();
 	IP6_INC_STATS_BH(net, __in6_dev_get(dev), IPSTATS_MIB_REASMOKS);
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index 2d59df5..e9dd47b 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -158,9 +158,7 @@ static int push_mpls(struct sk_buff *skb, struct sw_flow_key *key,
 	new_mpls_lse = (__be32 *)skb_mpls_header(skb);
 	*new_mpls_lse = mpls->mpls_lse;
 
-	if (skb->ip_summed == CHECKSUM_COMPLETE)
-		skb->csum = csum_add(skb->csum, csum_partial(new_mpls_lse,
-							     MPLS_HLEN, 0));
+	skb_postpush_rcsum(skb, new_mpls_lse, MPLS_HLEN);
 
 	hdr = eth_hdr(skb);
 	hdr->h_proto = mpls->mpls_ethertype;
@@ -280,7 +278,7 @@ static int set_eth_addr(struct sk_buff *skb, struct sw_flow_key *flow_key,
 	ether_addr_copy_masked(eth_hdr(skb)->h_dest, key->eth_dst,
 			       mask->eth_dst);
 
-	ovs_skb_postpush_rcsum(skb, eth_hdr(skb), ETH_ALEN * 2);
+	skb_postpush_rcsum(skb, eth_hdr(skb), ETH_ALEN * 2);
 
 	ether_addr_copy(flow_key->eth.src, eth_hdr(skb)->h_source);
 	ether_addr_copy(flow_key->eth.dst, eth_hdr(skb)->h_dest);
@@ -639,7 +637,7 @@ static int ovs_vport_output(struct net *net, struct sock *sk, struct sk_buff *sk
 	/* Reconstruct the MAC header.  */
 	skb_push(skb, data->l2_len);
 	memcpy(skb->data, &data->l2_data, data->l2_len);
-	ovs_skb_postpush_rcsum(skb, skb->data, data->l2_len);
+	skb_postpush_rcsum(skb, skb->data, data->l2_len);
 	skb_reset_mac_header(skb);
 
 	ovs_vport_send(vport, skb);
diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c
index 6a6adf3..4e39723 100644
--- a/net/openvswitch/vport-netdev.c
+++ b/net/openvswitch/vport-netdev.c
@@ -58,7 +58,7 @@ static void netdev_port_receive(struct sk_buff *skb)
 		return;
 
 	skb_push(skb, ETH_HLEN);
-	ovs_skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
+	skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
 	ovs_vport_receive(vport, skb, skb_tunnel_info(skb));
 	return;
 error:
diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h
index c10899cb..f01f28a 100644
--- a/net/openvswitch/vport.h
+++ b/net/openvswitch/vport.h
@@ -185,13 +185,6 @@ static inline struct vport *vport_from_priv(void *priv)
 int ovs_vport_receive(struct vport *, struct sk_buff *,
 		      const struct ip_tunnel_info *);
 
-static inline void ovs_skb_postpush_rcsum(struct sk_buff *skb,
-				      const void *start, unsigned int len)
-{
-	if (skb->ip_summed == CHECKSUM_COMPLETE)
-		skb->csum = csum_add(skb->csum, csum_partial(start, len, 0));
-}
-
 static inline const char *ovs_vport_name(struct vport *vport)
 {
 	return vport->dev->name;
-- 
cgit v0.10.2


From 568b329a02f75ed3aaae5eb2cca384cb9e09cb29 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Wed, 17 Feb 2016 19:58:57 -0800
Subject: perf: generalize perf_callchain

. avoid walking the stack when there is no room left in the buffer
. generalize get_perf_callchain() to be called from bpf helper

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h
index 70bbe39..7c247e7 100644
--- a/arch/x86/include/asm/stacktrace.h
+++ b/arch/x86/include/asm/stacktrace.h
@@ -37,7 +37,7 @@ print_context_stack_bp(struct thread_info *tinfo,
 /* Generic stack tracer with callbacks */
 
 struct stacktrace_ops {
-	void (*address)(void *data, unsigned long address, int reliable);
+	int (*address)(void *data, unsigned long address, int reliable);
 	/* On negative return stop dumping */
 	int (*stack)(void *data, char *name);
 	walk_stack_t	walk_stack;
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 1b443db..d276b31 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -2180,11 +2180,11 @@ static int backtrace_stack(void *data, char *name)
 	return 0;
 }
 
-static void backtrace_address(void *data, unsigned long addr, int reliable)
+static int backtrace_address(void *data, unsigned long addr, int reliable)
 {
 	struct perf_callchain_entry *entry = data;
 
-	perf_callchain_store(entry, addr);
+	return perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops backtrace_ops = {
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index 9c30acf..0d1ff4b 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -135,7 +135,8 @@ print_context_stack_bp(struct thread_info *tinfo,
 		if (!__kernel_text_address(addr))
 			break;
 
-		ops->address(data, addr, 1);
+		if (ops->address(data, addr, 1))
+			break;
 		frame = frame->next_frame;
 		ret_addr = &frame->return_address;
 		print_ftrace_graph_addr(addr, data, ops, tinfo, graph);
@@ -154,10 +155,11 @@ static int print_trace_stack(void *data, char *name)
 /*
  * Print one address/symbol entries per line.
  */
-static void print_trace_address(void *data, unsigned long addr, int reliable)
+static int print_trace_address(void *data, unsigned long addr, int reliable)
 {
 	touch_nmi_watchdog();
 	printk_stack_address(addr, reliable, data);
+	return 0;
 }
 
 static const struct stacktrace_ops print_trace_ops = {
diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
index fdd0c64..9ee98ee 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -14,30 +14,34 @@ static int save_stack_stack(void *data, char *name)
 	return 0;
 }
 
-static void
+static int
 __save_stack_address(void *data, unsigned long addr, bool reliable, bool nosched)
 {
 	struct stack_trace *trace = data;
 #ifdef CONFIG_FRAME_POINTER
 	if (!reliable)
-		return;
+		return 0;
 #endif
 	if (nosched && in_sched_functions(addr))
-		return;
+		return 0;
 	if (trace->skip > 0) {
 		trace->skip--;
-		return;
+		return 0;
 	}
-	if (trace->nr_entries < trace->max_entries)
+	if (trace->nr_entries < trace->max_entries) {
 		trace->entries[trace->nr_entries++] = addr;
+		return 0;
+	} else {
+		return -1; /* no more room, stop walking the stack */
+	}
 }
 
-static void save_stack_address(void *data, unsigned long addr, int reliable)
+static int save_stack_address(void *data, unsigned long addr, int reliable)
 {
 	return __save_stack_address(data, addr, reliable, false);
 }
 
-static void
+static int
 save_stack_address_nosched(void *data, unsigned long addr, int reliable)
 {
 	return __save_stack_address(data, addr, reliable, true);
diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c
index 4e664bd..cb31a44 100644
--- a/arch/x86/oprofile/backtrace.c
+++ b/arch/x86/oprofile/backtrace.c
@@ -23,12 +23,13 @@ static int backtrace_stack(void *data, char *name)
 	return 0;
 }
 
-static void backtrace_address(void *data, unsigned long addr, int reliable)
+static int backtrace_address(void *data, unsigned long addr, int reliable)
 {
 	unsigned int *depth = data;
 
 	if ((*depth)--)
 		oprofile_add_trace(addr);
+	return 0;
 }
 
 static struct stacktrace_ops backtrace_ops = {
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index b35a61a..7da3c25 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -964,11 +964,20 @@ DECLARE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry);
 
 extern void perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs);
 extern void perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs);
+extern struct perf_callchain_entry *
+get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user,
+		   bool crosstask, bool add_mark);
+extern int get_callchain_buffers(void);
+extern void put_callchain_buffers(void);
 
-static inline void perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
+static inline int perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
 {
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
+	if (entry->nr < PERF_MAX_STACK_DEPTH) {
 		entry->ip[entry->nr++] = ip;
+		return 0;
+	} else {
+		return -1; /* no more room, stop walking the stack */
+	}
 }
 
 extern int sysctl_perf_event_paranoid;
diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c
index 9c41800..343c22f 100644
--- a/kernel/events/callchain.c
+++ b/kernel/events/callchain.c
@@ -159,15 +159,24 @@ put_callchain_entry(int rctx)
 struct perf_callchain_entry *
 perf_callchain(struct perf_event *event, struct pt_regs *regs)
 {
-	int rctx;
-	struct perf_callchain_entry *entry;
-
-	int kernel = !event->attr.exclude_callchain_kernel;
-	int user   = !event->attr.exclude_callchain_user;
+	bool kernel = !event->attr.exclude_callchain_kernel;
+	bool user   = !event->attr.exclude_callchain_user;
+	/* Disallow cross-task user callchains. */
+	bool crosstask = event->ctx->task && event->ctx->task != current;
 
 	if (!kernel && !user)
 		return NULL;
 
+	return get_perf_callchain(regs, 0, kernel, user, crosstask, true);
+}
+
+struct perf_callchain_entry *
+get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user,
+		   bool crosstask, bool add_mark)
+{
+	struct perf_callchain_entry *entry;
+	int rctx;
+
 	entry = get_callchain_entry(&rctx);
 	if (rctx == -1)
 		return NULL;
@@ -175,10 +184,11 @@ perf_callchain(struct perf_event *event, struct pt_regs *regs)
 	if (!entry)
 		goto exit_put;
 
-	entry->nr = 0;
+	entry->nr = init_nr;
 
 	if (kernel && !user_mode(regs)) {
-		perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+		if (add_mark)
+			perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 		perf_callchain_kernel(entry, regs);
 	}
 
@@ -191,13 +201,11 @@ perf_callchain(struct perf_event *event, struct pt_regs *regs)
 		}
 
 		if (regs) {
-			/*
-			 * Disallow cross-task user callchains.
-			 */
-			if (event->ctx->task && event->ctx->task != current)
+			if (crosstask)
 				goto exit_put;
 
-			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			if (add_mark)
+				perf_callchain_store(entry, PERF_CONTEXT_USER);
 			perf_callchain_user(entry, regs);
 		}
 	}
diff --git a/kernel/events/internal.h b/kernel/events/internal.h
index 2bbad9c..4199b6d 100644
--- a/kernel/events/internal.h
+++ b/kernel/events/internal.h
@@ -182,8 +182,6 @@ DEFINE_OUTPUT_COPY(__output_copy_user, arch_perf_out_copy_user)
 /* Callchain handling */
 extern struct perf_callchain_entry *
 perf_callchain(struct perf_event *event, struct pt_regs *regs);
-extern int get_callchain_buffers(void);
-extern void put_callchain_buffers(void);
 
 static inline int get_recursion_context(int *recursion)
 {
-- 
cgit v0.10.2


From d5a3b1f691865be576c2bffa708549b8cdccda19 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Wed, 17 Feb 2016 19:58:58 -0800
Subject: bpf: introduce BPF_MAP_TYPE_STACK_TRACE

add new map type to store stack traces and corresponding helper
bpf_get_stackid(ctx, map, flags) - walk user or kernel stack and return id
@ctx: struct pt_regs*
@map: pointer to stack_trace map
@flags: bits 0-7 - numer of stack frames to skip
        bit 8 - collect user stack instead of kernel
        bit 9 - compare stacks by hash only
        bit 10 - if two different stacks hash into the same stackid
                 discard old
        other bits - reserved
Return: >= 0 stackid on success or negative error

stackid is a 32-bit integer handle that can be further combined with
other data (including other stackid) and used as a key into maps.

Userspace will access stackmap using standard lookup/delete syscall commands to
retrieve full stack trace for given stackid.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 90ee6ab..0cadbb7 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -237,6 +237,7 @@ extern const struct bpf_func_proto bpf_get_current_uid_gid_proto;
 extern const struct bpf_func_proto bpf_get_current_comm_proto;
 extern const struct bpf_func_proto bpf_skb_vlan_push_proto;
 extern const struct bpf_func_proto bpf_skb_vlan_pop_proto;
+extern const struct bpf_func_proto bpf_get_stackid_proto;
 
 /* Shared helpers among cBPF and eBPF. */
 void bpf_user_rnd_init_once(void);
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 2ee0fde..d3e77da 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -83,6 +83,7 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_PERF_EVENT_ARRAY,
 	BPF_MAP_TYPE_PERCPU_HASH,
 	BPF_MAP_TYPE_PERCPU_ARRAY,
+	BPF_MAP_TYPE_STACK_TRACE,
 };
 
 enum bpf_prog_type {
@@ -272,6 +273,20 @@ enum bpf_func_id {
 	 */
 	BPF_FUNC_perf_event_output,
 	BPF_FUNC_skb_load_bytes,
+
+	/**
+	 * bpf_get_stackid(ctx, map, flags) - walk user or kernel stack and return id
+	 * @ctx: struct pt_regs*
+	 * @map: pointer to stack_trace map
+	 * @flags: bits 0-7 - numer of stack frames to skip
+	 *         bit 8 - collect user stack instead of kernel
+	 *         bit 9 - compare stacks by hash only
+	 *         bit 10 - if two different stacks hash into the same stackid
+	 *                  discard old
+	 *         other bits - reserved
+	 * Return: >= 0 stackid on success or negative error
+	 */
+	BPF_FUNC_get_stackid,
 	__BPF_FUNC_MAX_ID,
 };
 
@@ -294,6 +309,12 @@ enum bpf_func_id {
 /* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */
 #define BPF_F_TUNINFO_IPV6		(1ULL << 0)
 
+/* BPF_FUNC_get_stackid flags. */
+#define BPF_F_SKIP_FIELD_MASK		0xffULL
+#define BPF_F_USER_STACK		(1ULL << 8)
+#define BPF_F_FAST_STACK_CMP		(1ULL << 9)
+#define BPF_F_REUSE_STACKID		(1ULL << 10)
+
 /* user accessible mirror of in-kernel sk_buff.
  * new fields can only be added to the end of this structure
  */
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index 1327258..8a932d0 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -2,3 +2,6 @@ obj-y := core.o
 
 obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o
 obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o
+ifeq ($(CONFIG_PERF_EVENTS),y)
+obj-$(CONFIG_BPF_SYSCALL) += stackmap.o
+endif
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
new file mode 100644
index 0000000..8a60ee1
--- /dev/null
+++ b/kernel/bpf/stackmap.c
@@ -0,0 +1,237 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <linux/bpf.h>
+#include <linux/jhash.h>
+#include <linux/filter.h>
+#include <linux/vmalloc.h>
+#include <linux/stacktrace.h>
+#include <linux/perf_event.h>
+
+struct stack_map_bucket {
+	struct rcu_head rcu;
+	u32 hash;
+	u32 nr;
+	u64 ip[];
+};
+
+struct bpf_stack_map {
+	struct bpf_map map;
+	u32 n_buckets;
+	struct stack_map_bucket __rcu *buckets[];
+};
+
+/* Called from syscall */
+static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
+{
+	u32 value_size = attr->value_size;
+	struct bpf_stack_map *smap;
+	u64 cost, n_buckets;
+	int err;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return ERR_PTR(-EPERM);
+
+	/* check sanity of attributes */
+	if (attr->max_entries == 0 || attr->key_size != 4 ||
+	    value_size < 8 || value_size % 8 ||
+	    value_size / 8 > PERF_MAX_STACK_DEPTH)
+		return ERR_PTR(-EINVAL);
+
+	/* hash table size must be power of 2 */
+	n_buckets = roundup_pow_of_two(attr->max_entries);
+
+	cost = n_buckets * sizeof(struct stack_map_bucket *) + sizeof(*smap);
+	if (cost >= U32_MAX - PAGE_SIZE)
+		return ERR_PTR(-E2BIG);
+
+	smap = kzalloc(cost, GFP_USER | __GFP_NOWARN);
+	if (!smap) {
+		smap = vzalloc(cost);
+		if (!smap)
+			return ERR_PTR(-ENOMEM);
+	}
+
+	err = -E2BIG;
+	cost += n_buckets * (value_size + sizeof(struct stack_map_bucket));
+	if (cost >= U32_MAX - PAGE_SIZE)
+		goto free_smap;
+
+	smap->map.map_type = attr->map_type;
+	smap->map.key_size = attr->key_size;
+	smap->map.value_size = value_size;
+	smap->map.max_entries = attr->max_entries;
+	smap->n_buckets = n_buckets;
+	smap->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT;
+
+	err = get_callchain_buffers();
+	if (err)
+		goto free_smap;
+
+	return &smap->map;
+
+free_smap:
+	kvfree(smap);
+	return ERR_PTR(err);
+}
+
+static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5)
+{
+	struct pt_regs *regs = (struct pt_regs *) (long) r1;
+	struct bpf_map *map = (struct bpf_map *) (long) r2;
+	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
+	struct perf_callchain_entry *trace;
+	struct stack_map_bucket *bucket, *new_bucket, *old_bucket;
+	u32 max_depth = map->value_size / 8;
+	/* stack_map_alloc() checks that max_depth <= PERF_MAX_STACK_DEPTH */
+	u32 init_nr = PERF_MAX_STACK_DEPTH - max_depth;
+	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
+	u32 hash, id, trace_nr, trace_len;
+	bool user = flags & BPF_F_USER_STACK;
+	bool kernel = !user;
+	u64 *ips;
+
+	if (unlikely(flags & ~(BPF_F_SKIP_FIELD_MASK | BPF_F_USER_STACK |
+			       BPF_F_FAST_STACK_CMP | BPF_F_REUSE_STACKID)))
+		return -EINVAL;
+
+	trace = get_perf_callchain(regs, init_nr, kernel, user, false, false);
+
+	if (unlikely(!trace))
+		/* couldn't fetch the stack trace */
+		return -EFAULT;
+
+	/* get_perf_callchain() guarantees that trace->nr >= init_nr
+	 * and trace-nr <= PERF_MAX_STACK_DEPTH, so trace_nr <= max_depth
+	 */
+	trace_nr = trace->nr - init_nr;
+
+	if (trace_nr <= skip)
+		/* skipping more than usable stack trace */
+		return -EFAULT;
+
+	trace_nr -= skip;
+	trace_len = trace_nr * sizeof(u64);
+	ips = trace->ip + skip + init_nr;
+	hash = jhash2((u32 *)ips, trace_len / sizeof(u32), 0);
+	id = hash & (smap->n_buckets - 1);
+	bucket = rcu_dereference(smap->buckets[id]);
+
+	if (bucket && bucket->hash == hash) {
+		if (flags & BPF_F_FAST_STACK_CMP)
+			return id;
+		if (bucket->nr == trace_nr &&
+		    memcmp(bucket->ip, ips, trace_len) == 0)
+			return id;
+	}
+
+	/* this call stack is not in the map, try to add it */
+	if (bucket && !(flags & BPF_F_REUSE_STACKID))
+		return -EEXIST;
+
+	new_bucket = kmalloc(sizeof(struct stack_map_bucket) + map->value_size,
+			     GFP_ATOMIC | __GFP_NOWARN);
+	if (unlikely(!new_bucket))
+		return -ENOMEM;
+
+	memcpy(new_bucket->ip, ips, trace_len);
+	memset(new_bucket->ip + trace_len / 8, 0, map->value_size - trace_len);
+	new_bucket->hash = hash;
+	new_bucket->nr = trace_nr;
+
+	old_bucket = xchg(&smap->buckets[id], new_bucket);
+	if (old_bucket)
+		kfree_rcu(old_bucket, rcu);
+	return id;
+}
+
+const struct bpf_func_proto bpf_get_stackid_proto = {
+	.func		= bpf_get_stackid,
+	.gpl_only	= true,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_CONST_MAP_PTR,
+	.arg3_type	= ARG_ANYTHING,
+};
+
+/* Called from syscall or from eBPF program */
+static void *stack_map_lookup_elem(struct bpf_map *map, void *key)
+{
+	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
+	struct stack_map_bucket *bucket;
+	u32 id = *(u32 *)key;
+
+	if (unlikely(id >= smap->n_buckets))
+		return NULL;
+	bucket = rcu_dereference(smap->buckets[id]);
+	return bucket ? bucket->ip : NULL;
+}
+
+static int stack_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
+{
+	return -EINVAL;
+}
+
+static int stack_map_update_elem(struct bpf_map *map, void *key, void *value,
+				 u64 map_flags)
+{
+	return -EINVAL;
+}
+
+/* Called from syscall or from eBPF program */
+static int stack_map_delete_elem(struct bpf_map *map, void *key)
+{
+	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
+	struct stack_map_bucket *old_bucket;
+	u32 id = *(u32 *)key;
+
+	if (unlikely(id >= smap->n_buckets))
+		return -E2BIG;
+
+	old_bucket = xchg(&smap->buckets[id], NULL);
+	if (old_bucket) {
+		kfree_rcu(old_bucket, rcu);
+		return 0;
+	} else {
+		return -ENOENT;
+	}
+}
+
+/* Called when map->refcnt goes to zero, either from workqueue or from syscall */
+static void stack_map_free(struct bpf_map *map)
+{
+	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
+	int i;
+
+	synchronize_rcu();
+
+	for (i = 0; i < smap->n_buckets; i++)
+		if (smap->buckets[i])
+			kfree_rcu(smap->buckets[i], rcu);
+	kvfree(smap);
+	put_callchain_buffers();
+}
+
+static const struct bpf_map_ops stack_map_ops = {
+	.map_alloc = stack_map_alloc,
+	.map_free = stack_map_free,
+	.map_get_next_key = stack_map_get_next_key,
+	.map_lookup_elem = stack_map_lookup_elem,
+	.map_update_elem = stack_map_update_elem,
+	.map_delete_elem = stack_map_delete_elem,
+};
+
+static struct bpf_map_type_list stack_map_type __read_mostly = {
+	.ops = &stack_map_ops,
+	.type = BPF_MAP_TYPE_STACK_TRACE,
+};
+
+static int __init register_stack_map(void)
+{
+	bpf_register_map_type(&stack_map_type);
+	return 0;
+}
+late_initcall(register_stack_map);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d1d3e8f..42ba4cc 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -246,6 +246,7 @@ static const struct {
 	{BPF_MAP_TYPE_PROG_ARRAY, BPF_FUNC_tail_call},
 	{BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_read},
 	{BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_output},
+	{BPF_MAP_TYPE_STACK_TRACE, BPF_FUNC_get_stackid},
 };
 
 static void print_verifier_state(struct verifier_env *env)
@@ -911,8 +912,11 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id)
 		 * don't allow any other map type to be passed into
 		 * the special func;
 		 */
-		if (bool_func && bool_map != bool_func)
+		if (bool_func && bool_map != bool_func) {
+			verbose("cannot pass map_type %d into func %d\n",
+				map->map_type, func_id);
 			return -EINVAL;
+		}
 	}
 
 	return 0;
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 326a75e..4b8caa3 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -299,6 +299,8 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func
 		return &bpf_perf_event_read_proto;
 	case BPF_FUNC_perf_event_output:
 		return &bpf_perf_event_output_proto;
+	case BPF_FUNC_get_stackid:
+		return &bpf_get_stackid_proto;
 	default:
 		return NULL;
 	}
-- 
cgit v0.10.2


From a6ffe7b9df6228d11c5689914eceb488bc4e38df Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Wed, 17 Feb 2016 19:58:59 -0800
Subject: samples/bpf: offwaketime example

This is simplified version of Brendan Gregg's offwaketime:
This program shows kernel stack traces and task names that were blocked and
"off-CPU", along with the stack traces and task names for the threads that woke
them, and the total elapsed time from when they blocked to when they were woken
up. The combined stacks, task names, and total time is summarized in kernel
context for efficiency.

Example:
$ sudo ./offwaketime | flamegraph.pl > demo.svg
Open demo.svg in the browser as FlameGraph visualization.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index edd638b..c4f8ae0 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -16,6 +16,7 @@ hostprogs-y += tracex5
 hostprogs-y += tracex6
 hostprogs-y += trace_output
 hostprogs-y += lathist
+hostprogs-y += offwaketime
 
 test_verifier-objs := test_verifier.o libbpf.o
 test_maps-objs := test_maps.o libbpf.o
@@ -32,6 +33,7 @@ tracex5-objs := bpf_load.o libbpf.o tracex5_user.o
 tracex6-objs := bpf_load.o libbpf.o tracex6_user.o
 trace_output-objs := bpf_load.o libbpf.o trace_output_user.o
 lathist-objs := bpf_load.o libbpf.o lathist_user.o
+offwaketime-objs := bpf_load.o libbpf.o offwaketime_user.o
 
 # Tell kbuild to always build the programs
 always := $(hostprogs-y)
@@ -47,6 +49,7 @@ always += tracex6_kern.o
 always += trace_output_kern.o
 always += tcbpf1_kern.o
 always += lathist_kern.o
+always += offwaketime_kern.o
 
 HOSTCFLAGS += -I$(objtree)/usr/include
 
@@ -63,6 +66,7 @@ HOSTLOADLIBES_tracex5 += -lelf
 HOSTLOADLIBES_tracex6 += -lelf
 HOSTLOADLIBES_trace_output += -lelf -lrt
 HOSTLOADLIBES_lathist += -lelf
+HOSTLOADLIBES_offwaketime += -lelf
 
 # point this to your LLVM backend with bpf support
 LLC=$(srctree)/tools/bpf/llvm/bld/Debug+Asserts/bin/llc
diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h
index 7ad19e1..811bcca 100644
--- a/samples/bpf/bpf_helpers.h
+++ b/samples/bpf/bpf_helpers.h
@@ -39,6 +39,8 @@ static int (*bpf_redirect)(int ifindex, int flags) =
 	(void *) BPF_FUNC_redirect;
 static int (*bpf_perf_event_output)(void *ctx, void *map, int index, void *data, int size) =
 	(void *) BPF_FUNC_perf_event_output;
+static int (*bpf_get_stackid)(void *ctx, void *map, int flags) =
+	(void *) BPF_FUNC_get_stackid;
 
 /* llvm builtin functions that eBPF C program may use to
  * emit BPF_LD_ABS and BPF_LD_IND instructions
diff --git a/samples/bpf/offwaketime_kern.c b/samples/bpf/offwaketime_kern.c
new file mode 100644
index 0000000..c0aa5a9
--- /dev/null
+++ b/samples/bpf/offwaketime_kern.c
@@ -0,0 +1,131 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <uapi/linux/bpf.h>
+#include "bpf_helpers.h"
+#include <uapi/linux/ptrace.h>
+#include <uapi/linux/perf_event.h>
+#include <linux/version.h>
+#include <linux/sched.h>
+
+#define _(P) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
+
+#define MINBLOCK_US	1
+
+struct key_t {
+	char waker[TASK_COMM_LEN];
+	char target[TASK_COMM_LEN];
+	u32 wret;
+	u32 tret;
+};
+
+struct bpf_map_def SEC("maps") counts = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(struct key_t),
+	.value_size = sizeof(u64),
+	.max_entries = 10000,
+};
+
+struct bpf_map_def SEC("maps") start = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(u32),
+	.value_size = sizeof(u64),
+	.max_entries = 10000,
+};
+
+struct wokeby_t {
+	char name[TASK_COMM_LEN];
+	u32 ret;
+};
+
+struct bpf_map_def SEC("maps") wokeby = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(u32),
+	.value_size = sizeof(struct wokeby_t),
+	.max_entries = 10000,
+};
+
+struct bpf_map_def SEC("maps") stackmap = {
+	.type = BPF_MAP_TYPE_STACK_TRACE,
+	.key_size = sizeof(u32),
+	.value_size = PERF_MAX_STACK_DEPTH * sizeof(u64),
+	.max_entries = 10000,
+};
+
+#define STACKID_FLAGS (0 | BPF_F_FAST_STACK_CMP)
+
+SEC("kprobe/try_to_wake_up")
+int waker(struct pt_regs *ctx)
+{
+	struct task_struct *p = (void *) PT_REGS_PARM1(ctx);
+	struct wokeby_t woke = {};
+	u32 pid;
+
+	pid = _(p->pid);
+
+	bpf_get_current_comm(&woke.name, sizeof(woke.name));
+	woke.ret = bpf_get_stackid(ctx, &stackmap, STACKID_FLAGS);
+
+	bpf_map_update_elem(&wokeby, &pid, &woke, BPF_ANY);
+	return 0;
+}
+
+static inline int update_counts(struct pt_regs *ctx, u32 pid, u64 delta)
+{
+	struct key_t key = {};
+	struct wokeby_t *woke;
+	u64 zero = 0, *val;
+
+	bpf_get_current_comm(&key.target, sizeof(key.target));
+	key.tret = bpf_get_stackid(ctx, &stackmap, STACKID_FLAGS);
+
+	woke = bpf_map_lookup_elem(&wokeby, &pid);
+	if (woke) {
+		key.wret = woke->ret;
+		__builtin_memcpy(&key.waker, woke->name, TASK_COMM_LEN);
+		bpf_map_delete_elem(&wokeby, &pid);
+	}
+
+	val = bpf_map_lookup_elem(&counts, &key);
+	if (!val) {
+		bpf_map_update_elem(&counts, &key, &zero, BPF_NOEXIST);
+		val = bpf_map_lookup_elem(&counts, &key);
+		if (!val)
+			return 0;
+	}
+	(*val) += delta;
+	return 0;
+}
+
+SEC("kprobe/finish_task_switch")
+int oncpu(struct pt_regs *ctx)
+{
+	struct task_struct *p = (void *) PT_REGS_PARM1(ctx);
+	u64 delta, ts, *tsp;
+	u32 pid;
+
+	/* record previous thread sleep time */
+	pid = _(p->pid);
+	ts = bpf_ktime_get_ns();
+	bpf_map_update_elem(&start, &pid, &ts, BPF_ANY);
+
+	/* calculate current thread's delta time */
+	pid = bpf_get_current_pid_tgid();
+	tsp = bpf_map_lookup_elem(&start, &pid);
+	if (!tsp)
+		/* missed start or filtered */
+		return 0;
+
+	delta = bpf_ktime_get_ns() - *tsp;
+	bpf_map_delete_elem(&start, &pid);
+	delta = delta / 1000;
+	if (delta < MINBLOCK_US)
+		return 0;
+
+	return update_counts(ctx, pid, delta);
+}
+char _license[] SEC("license") = "GPL";
+u32 _version SEC("version") = LINUX_VERSION_CODE;
diff --git a/samples/bpf/offwaketime_user.c b/samples/bpf/offwaketime_user.c
new file mode 100644
index 0000000..17cf302
--- /dev/null
+++ b/samples/bpf/offwaketime_user.c
@@ -0,0 +1,185 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <linux/bpf.h>
+#include <string.h>
+#include <linux/perf_event.h>
+#include <errno.h>
+#include <assert.h>
+#include <stdbool.h>
+#include <sys/resource.h>
+#include "libbpf.h"
+#include "bpf_load.h"
+
+#define MAX_SYMS 300000
+#define PRINT_RAW_ADDR 0
+
+static struct ksym {
+	long addr;
+	char *name;
+} syms[MAX_SYMS];
+static int sym_cnt;
+
+static int ksym_cmp(const void *p1, const void *p2)
+{
+	return ((struct ksym *)p1)->addr - ((struct ksym *)p2)->addr;
+}
+
+static int load_kallsyms(void)
+{
+	FILE *f = fopen("/proc/kallsyms", "r");
+	char func[256], buf[256];
+	char symbol;
+	void *addr;
+	int i = 0;
+
+	if (!f)
+		return -ENOENT;
+
+	while (!feof(f)) {
+		if (!fgets(buf, sizeof(buf), f))
+			break;
+		if (sscanf(buf, "%p %c %s", &addr, &symbol, func) != 3)
+			break;
+		if (!addr)
+			continue;
+		syms[i].addr = (long) addr;
+		syms[i].name = strdup(func);
+		i++;
+	}
+	sym_cnt = i;
+	qsort(syms, sym_cnt, sizeof(struct ksym), ksym_cmp);
+	return 0;
+}
+
+static void *search(long key)
+{
+	int start = 0, end = sym_cnt;
+	int result;
+
+	while (start < end) {
+		size_t mid = start + (end - start) / 2;
+
+		result = key - syms[mid].addr;
+		if (result < 0)
+			end = mid;
+		else if (result > 0)
+			start = mid + 1;
+		else
+			return &syms[mid];
+	}
+
+	if (start >= 1 && syms[start - 1].addr < key &&
+	    key < syms[start].addr)
+		/* valid ksym */
+		return &syms[start - 1];
+
+	/* out of range. return _stext */
+	return &syms[0];
+}
+
+static void print_ksym(__u64 addr)
+{
+	struct ksym *sym;
+
+	if (!addr)
+		return;
+	sym = search(addr);
+	if (PRINT_RAW_ADDR)
+		printf("%s/%llx;", sym->name, addr);
+	else
+		printf("%s;", sym->name);
+}
+
+#define TASK_COMM_LEN 16
+
+struct key_t {
+	char waker[TASK_COMM_LEN];
+	char target[TASK_COMM_LEN];
+	__u32 wret;
+	__u32 tret;
+};
+
+static void print_stack(struct key_t *key, __u64 count)
+{
+	__u64 ip[PERF_MAX_STACK_DEPTH] = {};
+	static bool warned;
+	int i;
+
+	printf("%s;", key->target);
+	if (bpf_lookup_elem(map_fd[3], &key->tret, ip) != 0) {
+		printf("---;");
+	} else {
+		for (i = PERF_MAX_STACK_DEPTH - 1; i >= 0; i--)
+			print_ksym(ip[i]);
+	}
+	printf("-;");
+	if (bpf_lookup_elem(map_fd[3], &key->wret, ip) != 0) {
+		printf("---;");
+	} else {
+		for (i = 0; i < PERF_MAX_STACK_DEPTH; i++)
+			print_ksym(ip[i]);
+	}
+	printf(";%s %lld\n", key->waker, count);
+
+	if ((key->tret == -EEXIST || key->wret == -EEXIST) && !warned) {
+		printf("stackmap collisions seen. Consider increasing size\n");
+		warned = true;
+	} else if (((int)(key->tret) < 0 || (int)(key->wret) < 0)) {
+		printf("err stackid %d %d\n", key->tret, key->wret);
+	}
+}
+
+static void print_stacks(int fd)
+{
+	struct key_t key = {}, next_key;
+	__u64 value;
+
+	while (bpf_get_next_key(fd, &key, &next_key) == 0) {
+		bpf_lookup_elem(fd, &next_key, &value);
+		print_stack(&next_key, value);
+		key = next_key;
+	}
+}
+
+static void int_exit(int sig)
+{
+	print_stacks(map_fd[0]);
+	exit(0);
+}
+
+int main(int argc, char **argv)
+{
+	struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+	char filename[256];
+	int delay = 1;
+
+	snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+	setrlimit(RLIMIT_MEMLOCK, &r);
+
+	signal(SIGINT, int_exit);
+
+	if (load_kallsyms()) {
+		printf("failed to process /proc/kallsyms\n");
+		return 2;
+	}
+
+	if (load_bpf_file(filename)) {
+		printf("%s", bpf_log_buf);
+		return 1;
+	}
+
+	if (argc > 1)
+		delay = atoi(argv[1]);
+	sleep(delay);
+	print_stacks(map_fd[0]);
+
+	return 0;
+}
-- 
cgit v0.10.2


From f49cbe6b7988b51aa2b72c45d4332fabea62fba6 Mon Sep 17 00:00:00 2001
From: Damien Riegel <damien.riegel@savoirfairelinux.com>
Date: Tue, 12 Jan 2016 17:31:18 -0500
Subject: can: sja1000: of: add per-compatible init hook

This commit adds the capability to allocate and init private data
embedded in the sja1000_priv structure on a per-compatible basis. The
device node is passed as a parameter of the init callback to allow
parsing of custom device tree properties.

Signed-off-by: Damien Riegel <damien.riegel@savoirfairelinux.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/sja1000/sja1000_platform.c b/drivers/net/can/sja1000/sja1000_platform.c
index 0552ed4..777d312 100644
--- a/drivers/net/can/sja1000/sja1000_platform.c
+++ b/drivers/net/can/sja1000/sja1000_platform.c
@@ -27,6 +27,7 @@
 #include <linux/can/platform/sja1000.h>
 #include <linux/io.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 #include <linux/of_irq.h>
 
 #include "sja1000.h"
@@ -40,6 +41,11 @@ MODULE_DESCRIPTION("Socket-CAN driver for SJA1000 on the platform bus");
 MODULE_ALIAS("platform:" DRV_NAME);
 MODULE_LICENSE("GPL v2");
 
+struct sja1000_of_data {
+	size_t  priv_sz;
+	int     (*init)(struct sja1000_priv *priv, struct device_node *of);
+};
+
 static u8 sp_read_reg8(const struct sja1000_priv *priv, int reg)
 {
 	return ioread8(priv->reg_base + reg);
@@ -154,6 +160,12 @@ static void sp_populate_of(struct sja1000_priv *priv, struct device_node *of)
 		priv->cdr |= CDR_CBP; /* default */
 }
 
+static const struct of_device_id sp_of_table[] = {
+	{ .compatible = "nxp,sja1000", .data = NULL, },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, sp_of_table);
+
 static int sp_probe(struct platform_device *pdev)
 {
 	int err, irq = 0;
@@ -163,6 +175,9 @@ static int sp_probe(struct platform_device *pdev)
 	struct resource *res_mem, *res_irq = NULL;
 	struct sja1000_platform_data *pdata;
 	struct device_node *of = pdev->dev.of_node;
+	const struct of_device_id *of_id;
+	const struct sja1000_of_data *of_data = NULL;
+	size_t priv_sz = 0;
 
 	pdata = dev_get_platdata(&pdev->dev);
 	if (!pdata && !of) {
@@ -191,7 +206,13 @@ static int sp_probe(struct platform_device *pdev)
 	if (!irq && !res_irq)
 		return -ENODEV;
 
-	dev = alloc_sja1000dev(0);
+	of_id = of_match_device(sp_of_table, &pdev->dev);
+	if (of_id && of_id->data) {
+		of_data = of_id->data;
+		priv_sz = of_data->priv_sz;
+	}
+
+	dev = alloc_sja1000dev(priv_sz);
 	if (!dev)
 		return -ENOMEM;
 	priv = netdev_priv(dev);
@@ -208,10 +229,17 @@ static int sp_probe(struct platform_device *pdev)
 	dev->irq = irq;
 	priv->reg_base = addr;
 
-	if (of)
+	if (of) {
 		sp_populate_of(priv, of);
-	else
+
+		if (of_data && of_data->init) {
+			err = of_data->init(priv, of);
+			if (err)
+				goto exit_free;
+		}
+	} else {
 		sp_populate(priv, pdata, res_mem->flags);
+	}
 
 	platform_set_drvdata(pdev, dev);
 	SET_NETDEV_DEV(dev, &pdev->dev);
@@ -242,12 +270,6 @@ static int sp_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static const struct of_device_id sp_of_table[] = {
-	{.compatible = "nxp,sja1000"},
-	{},
-};
-MODULE_DEVICE_TABLE(of, sp_of_table);
-
 static struct platform_driver sp_driver = {
 	.probe = sp_probe,
 	.remove = sp_remove,
-- 
cgit v0.10.2


From 83c26850fc36014f5c13a1b2bf460bfa200af8c9 Mon Sep 17 00:00:00 2001
From: Damien Riegel <damien.riegel@savoirfairelinux.com>
Date: Tue, 12 Jan 2016 17:31:19 -0500
Subject: can: sja1000: add documentation for Technologic Systems version

This commit adds documentation for the Technologic Systems version of
SJA1000. The difference with the NXP version is in the way the registers
are accessed.

Signed-off-by: Damien Riegel <damien.riegel@savoirfairelinux.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/Documentation/devicetree/bindings/net/can/sja1000.txt b/Documentation/devicetree/bindings/net/can/sja1000.txt
index b4a6d53..ac3160e 100644
--- a/Documentation/devicetree/bindings/net/can/sja1000.txt
+++ b/Documentation/devicetree/bindings/net/can/sja1000.txt
@@ -2,7 +2,7 @@ Memory mapped SJA1000 CAN controller from NXP (formerly Philips)
 
 Required properties:
 
-- compatible : should be "nxp,sja1000".
+- compatible : should be one of "nxp,sja1000", "technologic,sja1000".
 
 - reg : should specify the chip select, address offset and size required
 	to map the registers of the SJA1000. The size is usually 0x80.
@@ -14,6 +14,7 @@ Optional properties:
 
 - reg-io-width : Specify the size (in bytes) of the IO accesses that
 	should be performed on the device.  Valid value is 1, 2 or 4.
+	This property is ignored for technologic version.
 	Default to 1 (8 bits).
 
 - nxp,external-clock-frequency : Frequency of the external oscillator
-- 
cgit v0.10.2


From dfb86c0d5a9d1c068954d776bf66e3d0b27beb68 Mon Sep 17 00:00:00 2001
From: Damien Riegel <damien.riegel@savoirfairelinux.com>
Date: Tue, 12 Jan 2016 17:31:20 -0500
Subject: can: sja1000: of: add compatibility with Technologic Systems version

Technologic Systems provides an IP compatible with the SJA1000,
instantiated in an FPGA. Because of some bus widths issue, access to
registers is made through a "window" that works like this:

    base + 0x0: address to read/write
    base + 0x2: 8-bit register value

This commit adds a new compatible device, "technologic,sja1000", with
read and write functions using the window mechanism.

Signed-off-by: Damien Riegel <damien.riegel@savoirfairelinux.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/sja1000/sja1000_platform.c b/drivers/net/can/sja1000/sja1000_platform.c
index 777d312..dc9c6db 100644
--- a/drivers/net/can/sja1000/sja1000_platform.c
+++ b/drivers/net/can/sja1000/sja1000_platform.c
@@ -46,6 +46,10 @@ struct sja1000_of_data {
 	int     (*init)(struct sja1000_priv *priv, struct device_node *of);
 };
 
+struct technologic_priv {
+	spinlock_t      io_lock;
+};
+
 static u8 sp_read_reg8(const struct sja1000_priv *priv, int reg)
 {
 	return ioread8(priv->reg_base + reg);
@@ -76,6 +80,43 @@ static void sp_write_reg32(const struct sja1000_priv *priv, int reg, u8 val)
 	iowrite8(val, priv->reg_base + reg * 4);
 }
 
+static u8 sp_technologic_read_reg16(const struct sja1000_priv *priv, int reg)
+{
+	struct technologic_priv *tp = priv->priv;
+	unsigned long flags;
+	u8 val;
+
+	spin_lock_irqsave(&tp->io_lock, flags);
+	iowrite16(reg, priv->reg_base + 0);
+	val = ioread16(priv->reg_base + 2);
+	spin_unlock_irqrestore(&tp->io_lock, flags);
+
+	return val;
+}
+
+static void sp_technologic_write_reg16(const struct sja1000_priv *priv,
+				       int reg, u8 val)
+{
+	struct technologic_priv *tp = priv->priv;
+	unsigned long flags;
+
+	spin_lock_irqsave(&tp->io_lock, flags);
+	iowrite16(reg, priv->reg_base + 0);
+	iowrite16(val, priv->reg_base + 2);
+	spin_unlock_irqrestore(&tp->io_lock, flags);
+}
+
+static int sp_technologic_init(struct sja1000_priv *priv, struct device_node *of)
+{
+	struct technologic_priv *tp = priv->priv;
+
+	priv->read_reg = sp_technologic_read_reg16;
+	priv->write_reg = sp_technologic_write_reg16;
+	spin_lock_init(&tp->io_lock);
+
+	return 0;
+}
+
 static void sp_populate(struct sja1000_priv *priv,
 			struct sja1000_platform_data *pdata,
 			unsigned long resource_mem_flags)
@@ -160,8 +201,14 @@ static void sp_populate_of(struct sja1000_priv *priv, struct device_node *of)
 		priv->cdr |= CDR_CBP; /* default */
 }
 
+static struct sja1000_of_data technologic_data = {
+	.priv_sz = sizeof(struct technologic_priv),
+	.init = sp_technologic_init,
+};
+
 static const struct of_device_id sp_of_table[] = {
 	{ .compatible = "nxp,sja1000", .data = NULL, },
+	{ .compatible = "technologic,sja1000", .data = &technologic_data, },
 	{ /* sentinel */ },
 };
 MODULE_DEVICE_TABLE(of, sp_of_table);
-- 
cgit v0.10.2


From 83407c7f42b3ad13ba55211779c7267cd4819ad8 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Mon, 11 Jan 2016 19:47:59 +0100
Subject: can: Kconfig: Sort the Kconfig includes

Sort the Kconfig includes, no functional change.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index 6d04183..6554404 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -147,22 +147,16 @@ config CAN_XILINXCAN
 	  Xilinx CAN driver. This driver supports both soft AXI CAN IP and
 	  Zynq CANPS IP.
 
-source "drivers/net/can/mscan/Kconfig"
-
-source "drivers/net/can/sja1000/Kconfig"
 
 source "drivers/net/can/c_can/Kconfig"
-
-source "drivers/net/can/m_can/Kconfig"
-
 source "drivers/net/can/cc770/Kconfig"
-
+source "drivers/net/can/m_can/Kconfig"
+source "drivers/net/can/mscan/Kconfig"
+source "drivers/net/can/sja1000/Kconfig"
+source "drivers/net/can/softing/Kconfig"
 source "drivers/net/can/spi/Kconfig"
-
 source "drivers/net/can/usb/Kconfig"
 
-source "drivers/net/can/softing/Kconfig"
-
 endif
 
 config CAN_DEBUG_DEVICES
-- 
cgit v0.10.2


From 26821162f958c6211427bc0a12fd1412a0d99e49 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Tue, 16 Feb 2016 09:34:38 +0100
Subject: can: Kconfig: sort drivers alphabetically

Sort the drivers that are directly listed in the Kconfig alphabetically, no
functional change.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index 6554404..b77ef32 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -70,13 +70,6 @@ config CAN_AT91
 	  This is a driver for the SoC CAN controller in Atmel's AT91SAM9263
 	  and AT91SAM9X5 processors.
 
-config CAN_TI_HECC
-	depends on ARM
-	tristate "TI High End CAN Controller"
-	---help---
-	  Driver for TI HECC (High End CAN Controller) module found on many
-	  TI devices. The device specifications are available from www.ti.com
-
 config CAN_BFIN
 	depends on BF534 || BF536 || BF537 || BF538 || BF539 || BF54x
 	tristate "Analog Devices Blackfin on-chip CAN"
@@ -86,30 +79,12 @@ config CAN_BFIN
 	  To compile this driver as a module, choose M here: the
 	  module will be called bfin_can.
 
-config CAN_JANZ_ICAN3
-	tristate "Janz VMOD-ICAN3 Intelligent CAN controller"
-	depends on MFD_JANZ_CMODIO
-	---help---
-	  Driver for Janz VMOD-ICAN3 Intelligent CAN controller module, which
-	  connects to a MODULbus carrier board.
-
-	  This driver can also be built as a module. If so, the module will be
-	  called janz-ican3.ko.
-
 config CAN_FLEXCAN
 	tristate "Support for Freescale FLEXCAN based chips"
 	depends on ARM || PPC
 	---help---
 	  Say Y here if you want to support for Freescale FlexCAN.
 
-config PCH_CAN
-	tristate "Intel EG20T PCH CAN controller"
-	depends on PCI && (X86_32 || COMPILE_TEST)
-	---help---
-	  This driver is for PCH CAN of Topcliff (Intel EG20T PCH) which
-	  is an IOH for x86 embedded processor (Intel Atom E6xx series).
-	  This driver can access CAN bus.
-
 config CAN_GRCAN
 	tristate "Aeroflex Gaisler GRCAN and GRHCAN CAN devices"
 	depends on OF && HAS_DMA
@@ -119,6 +94,16 @@ config CAN_GRCAN
 	  endian syntheses of the cores would need some modifications on
 	  the hardware level to work.
 
+config CAN_JANZ_ICAN3
+	tristate "Janz VMOD-ICAN3 Intelligent CAN controller"
+	depends on MFD_JANZ_CMODIO
+	---help---
+	  Driver for Janz VMOD-ICAN3 Intelligent CAN controller module, which
+	  connects to a MODULbus carrier board.
+
+	  This driver can also be built as a module. If so, the module will be
+	  called janz-ican3.ko.
+
 config CAN_RCAR
 	tristate "Renesas R-Car CAN controller"
 	depends on ARM
@@ -139,6 +124,13 @@ config CAN_SUN4I
 	  To compile this driver as a module, choose M here: the module will
 	  be called sun4i_can.
 
+config CAN_TI_HECC
+	depends on ARM
+	tristate "TI High End CAN Controller"
+	---help---
+	  Driver for TI HECC (High End CAN Controller) module found on many
+	  TI devices. The device specifications are available from www.ti.com
+
 config CAN_XILINXCAN
 	tristate "Xilinx CAN"
 	depends on ARCH_ZYNQ || ARM64 || MICROBLAZE || COMPILE_TEST
@@ -147,6 +139,13 @@ config CAN_XILINXCAN
 	  Xilinx CAN driver. This driver supports both soft AXI CAN IP and
 	  Zynq CANPS IP.
 
+config PCH_CAN
+	tristate "Intel EG20T PCH CAN controller"
+	depends on PCI && (X86_32 || COMPILE_TEST)
+	---help---
+	  This driver is for PCH CAN of Topcliff (Intel EG20T PCH) which
+	  is an IOH for x86 embedded processor (Intel Atom E6xx series).
+	  This driver can access CAN bus.
 
 source "drivers/net/can/c_can/Kconfig"
 source "drivers/net/can/cc770/Kconfig"
-- 
cgit v0.10.2


From 47338082066752f68562d51bbe2e36b501a1ce33 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Mon, 11 Jan 2016 19:48:00 +0100
Subject: can: Makefile: Sort the Makefile

Just sort the drivers in the Makefile, no functional change.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile
index 1f21cef..4f85c2b 100644
--- a/drivers/net/can/Makefile
+++ b/drivers/net/can/Makefile
@@ -14,21 +14,21 @@ obj-y				+= spi/
 obj-y				+= usb/
 obj-y				+= softing/
 
-obj-$(CONFIG_CAN_SJA1000)	+= sja1000/
-obj-$(CONFIG_CAN_MSCAN)		+= mscan/
-obj-$(CONFIG_CAN_C_CAN)		+= c_can/
-obj-$(CONFIG_CAN_M_CAN)		+= m_can/
-obj-$(CONFIG_CAN_CC770)		+= cc770/
 obj-$(CONFIG_CAN_AT91)		+= at91_can.o
-obj-$(CONFIG_CAN_TI_HECC)	+= ti_hecc.o
 obj-$(CONFIG_CAN_BFIN)		+= bfin_can.o
-obj-$(CONFIG_CAN_JANZ_ICAN3)	+= janz-ican3.o
+obj-$(CONFIG_CAN_CC770)		+= cc770/
+obj-$(CONFIG_CAN_C_CAN)		+= c_can/
 obj-$(CONFIG_CAN_FLEXCAN)	+= flexcan.o
-obj-$(CONFIG_PCH_CAN)		+= pch_can.o
 obj-$(CONFIG_CAN_GRCAN)		+= grcan.o
+obj-$(CONFIG_CAN_JANZ_ICAN3)	+= janz-ican3.o
+obj-$(CONFIG_CAN_MSCAN)		+= mscan/
+obj-$(CONFIG_CAN_M_CAN)		+= m_can/
 obj-$(CONFIG_CAN_RCAR)		+= rcar_can.o
+obj-$(CONFIG_CAN_SJA1000)	+= sja1000/
 obj-$(CONFIG_CAN_SUN4I)		+= sun4i_can.o
+obj-$(CONFIG_CAN_TI_HECC)	+= ti_hecc.o
 obj-$(CONFIG_CAN_XILINXCAN)	+= xilinx_can.o
+obj-$(CONFIG_PCH_CAN)		+= pch_can.o
 
 subdir-ccflags-y += -D__CHECK_ENDIAN__
 subdir-ccflags-$(CONFIG_CAN_DEBUG_DEVICES) += -DDEBUG
-- 
cgit v0.10.2


From 5afec0800a9ebbe822fe92fabfa7ae1291a0b042 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Mon, 11 Jan 2016 19:48:01 +0100
Subject: of: Add vendor prefix for I/F/I
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add vendor prefix for I/F/I, Ingenieurbüro Für IC-Technologie
http://www.ifi-pld.de/

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt
index 72e2c5a..c6134dc 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.txt
+++ b/Documentation/devicetree/bindings/vendor-prefixes.txt
@@ -111,6 +111,7 @@ hp	Hewlett Packard
 i2se	I2SE GmbH
 ibm	International Business Machines (IBM)
 idt	Integrated Device Technologies, Inc.
+ifi	Ingenieurburo Fur Ic-Technologie (I/F/I)
 iom	Iomega Corporation
 img	Imagination Technologies Ltd.
 ingenic	Ingenic Semiconductor
-- 
cgit v0.10.2


From 36840646bbc26b18a560895505f9dea642bc18df Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Mon, 11 Jan 2016 19:48:02 +0100
Subject: can: ifi: Add DT bindings for ifi,canfd

Add device tree bindings for the I/F/I CANFD controller IP core.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/Documentation/devicetree/bindings/net/can/ifi_canfd.txt b/Documentation/devicetree/bindings/net/can/ifi_canfd.txt
new file mode 100644
index 0000000..20ea5c7
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/can/ifi_canfd.txt
@@ -0,0 +1,15 @@
+IFI CANFD controller
+--------------------
+
+Required properties:
+  - compatible: Should be "ifi,canfd-1.0"
+  - reg: Should contain CAN controller registers location and length
+  - interrupts: Should contain IRQ line for the CAN controller
+
+Example:
+
+	canfd0: canfd@ff220000 {
+		compatible = "ifi,canfd-1.0";
+		reg = <0xff220000 0x00001000>;
+		interrupts = <0 43 0>;
+	};
-- 
cgit v0.10.2


From 0c4d9c94a1303008b32079991360cea7ab849ef3 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Wed, 20 Jan 2016 15:33:39 +0100
Subject: can: ifi: Add IFI CANFD IP support

The patch adds support for IFI CAN/FD controller [1]. This driver
currently supports sending and receiving both standard CAN and new
CAN/FD frames. Both ISO and BOSCH variant of CAN/FD is supported.

[1] http://www.ifi-pld.de/IP/CANFD/canfd.html

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index b77ef32..164ccde 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -149,6 +149,7 @@ config PCH_CAN
 
 source "drivers/net/can/c_can/Kconfig"
 source "drivers/net/can/cc770/Kconfig"
+source "drivers/net/can/ifi_canfd/Kconfig"
 source "drivers/net/can/m_can/Kconfig"
 source "drivers/net/can/mscan/Kconfig"
 source "drivers/net/can/sja1000/Kconfig"
diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile
index 4f85c2b..e3db0c8 100644
--- a/drivers/net/can/Makefile
+++ b/drivers/net/can/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_CAN_CC770)		+= cc770/
 obj-$(CONFIG_CAN_C_CAN)		+= c_can/
 obj-$(CONFIG_CAN_FLEXCAN)	+= flexcan.o
 obj-$(CONFIG_CAN_GRCAN)		+= grcan.o
+obj-$(CONFIG_CAN_IFI_CANFD)	+= ifi_canfd/
 obj-$(CONFIG_CAN_JANZ_ICAN3)	+= janz-ican3.o
 obj-$(CONFIG_CAN_MSCAN)		+= mscan/
 obj-$(CONFIG_CAN_M_CAN)		+= m_can/
diff --git a/drivers/net/can/ifi_canfd/Kconfig b/drivers/net/can/ifi_canfd/Kconfig
new file mode 100644
index 0000000..9e8934f
--- /dev/null
+++ b/drivers/net/can/ifi_canfd/Kconfig
@@ -0,0 +1,8 @@
+config CAN_IFI_CANFD
+	depends on HAS_IOMEM
+	tristate "IFI CAN_FD IP"
+	---help---
+	  This driver adds support for the I/F/I CAN_FD soft IP block
+	  connected to the "platform bus" (Linux abstraction for directly
+	  to the processor attached devices). The CAN_FD is most often
+	  synthesised into an FPGA or CPLD.
diff --git a/drivers/net/can/ifi_canfd/Makefile b/drivers/net/can/ifi_canfd/Makefile
new file mode 100644
index 0000000..b229960
--- /dev/null
+++ b/drivers/net/can/ifi_canfd/Makefile
@@ -0,0 +1,5 @@
+#
+#  Makefile for the IFI CANFD controller driver.
+#
+
+obj-$(CONFIG_CAN_IFI_CANFD) += ifi_canfd.o
diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c
new file mode 100644
index 0000000..0d1c164
--- /dev/null
+++ b/drivers/net/can/ifi_canfd/ifi_canfd.c
@@ -0,0 +1,917 @@
+/*
+ * CAN bus driver for IFI CANFD controller
+ *
+ * Copyright (C) 2016 Marek Vasut <marex@denx.de>
+ *
+ * Details about this controller can be found at
+ * http://www.ifi-pld.de/IP/CANFD/canfd.html
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+
+#include <linux/can/dev.h>
+
+#define IFI_CANFD_STCMD				0x0
+#define IFI_CANFD_STCMD_HARDRESET		0xDEADCAFD
+#define IFI_CANFD_STCMD_ENABLE			BIT(0)
+#define IFI_CANFD_STCMD_ERROR_ACTIVE		BIT(2)
+#define IFI_CANFD_STCMD_ERROR_PASSIVE		BIT(3)
+#define IFI_CANFD_STCMD_BUSOFF			BIT(4)
+#define IFI_CANFD_STCMD_BUSMONITOR		BIT(16)
+#define IFI_CANFD_STCMD_LOOPBACK		BIT(18)
+#define IFI_CANFD_STCMD_DISABLE_CANFD		BIT(24)
+#define IFI_CANFD_STCMD_ENABLE_ISO		BIT(25)
+#define IFI_CANFD_STCMD_NORMAL_MODE		((u32)BIT(31))
+
+#define IFI_CANFD_RXSTCMD			0x4
+#define IFI_CANFD_RXSTCMD_REMOVE_MSG		BIT(0)
+#define IFI_CANFD_RXSTCMD_RESET			BIT(7)
+#define IFI_CANFD_RXSTCMD_EMPTY			BIT(8)
+#define IFI_CANFD_RXSTCMD_OVERFLOW		BIT(13)
+
+#define IFI_CANFD_TXSTCMD			0x8
+#define IFI_CANFD_TXSTCMD_ADD_MSG		BIT(0)
+#define IFI_CANFD_TXSTCMD_HIGH_PRIO		BIT(1)
+#define IFI_CANFD_TXSTCMD_RESET			BIT(7)
+#define IFI_CANFD_TXSTCMD_EMPTY			BIT(8)
+#define IFI_CANFD_TXSTCMD_FULL			BIT(12)
+#define IFI_CANFD_TXSTCMD_OVERFLOW		BIT(13)
+
+#define IFI_CANFD_INTERRUPT			0xc
+#define IFI_CANFD_INTERRUPT_ERROR_WARNING	((u32)BIT(1))
+#define IFI_CANFD_INTERRUPT_TXFIFO_EMPTY	BIT(16)
+#define IFI_CANFD_INTERRUPT_TXFIFO_REMOVE	BIT(22)
+#define IFI_CANFD_INTERRUPT_RXFIFO_NEMPTY	BIT(24)
+#define IFI_CANFD_INTERRUPT_RXFIFO_NEMPTY_PER	BIT(25)
+#define IFI_CANFD_INTERRUPT_SET_IRQ		((u32)BIT(31))
+
+#define IFI_CANFD_IRQMASK			0x10
+#define IFI_CANFD_IRQMASK_SET_ERR		BIT(7)
+#define IFI_CANFD_IRQMASK_SET_TS		BIT(15)
+#define IFI_CANFD_IRQMASK_TXFIFO_EMPTY		BIT(16)
+#define IFI_CANFD_IRQMASK_SET_TX		BIT(23)
+#define IFI_CANFD_IRQMASK_RXFIFO_NEMPTY		BIT(24)
+#define IFI_CANFD_IRQMASK_SET_RX		((u32)BIT(31))
+
+#define IFI_CANFD_TIME				0x14
+#define IFI_CANFD_FTIME				0x18
+#define IFI_CANFD_TIME_TIMEB_OFF		0
+#define IFI_CANFD_TIME_TIMEA_OFF		8
+#define IFI_CANFD_TIME_PRESCALE_OFF		16
+#define IFI_CANFD_TIME_SJW_OFF_ISO		25
+#define IFI_CANFD_TIME_SJW_OFF_BOSCH		28
+#define IFI_CANFD_TIME_SET_SJW_BOSCH		BIT(6)
+#define IFI_CANFD_TIME_SET_TIMEB_BOSCH		BIT(7)
+#define IFI_CANFD_TIME_SET_PRESC_BOSCH		BIT(14)
+#define IFI_CANFD_TIME_SET_TIMEA_BOSCH		BIT(15)
+
+#define IFI_CANFD_TDELAY			0x1c
+
+#define IFI_CANFD_ERROR				0x20
+#define IFI_CANFD_ERROR_TX_OFFSET		0
+#define IFI_CANFD_ERROR_TX_MASK			0xff
+#define IFI_CANFD_ERROR_RX_OFFSET		16
+#define IFI_CANFD_ERROR_RX_MASK			0xff
+
+#define IFI_CANFD_ERRCNT			0x24
+
+#define IFI_CANFD_SUSPEND			0x28
+
+#define IFI_CANFD_REPEAT			0x2c
+
+#define IFI_CANFD_TRAFFIC			0x30
+
+#define IFI_CANFD_TSCONTROL			0x34
+
+#define IFI_CANFD_TSC				0x38
+
+#define IFI_CANFD_TST				0x3c
+
+#define IFI_CANFD_RES1				0x40
+
+#define IFI_CANFD_RES2				0x44
+
+#define IFI_CANFD_PAR				0x48
+
+#define IFI_CANFD_CANCLOCK			0x4c
+
+#define IFI_CANFD_SYSCLOCK			0x50
+
+#define IFI_CANFD_VER				0x54
+
+#define IFI_CANFD_IP_ID				0x58
+#define IFI_CANFD_IP_ID_VALUE			0xD073CAFD
+
+#define IFI_CANFD_TEST				0x5c
+
+#define IFI_CANFD_RXFIFO_TS_63_32		0x60
+
+#define IFI_CANFD_RXFIFO_TS_31_0		0x64
+
+#define IFI_CANFD_RXFIFO_DLC			0x68
+#define IFI_CANFD_RXFIFO_DLC_DLC_OFFSET		0
+#define IFI_CANFD_RXFIFO_DLC_DLC_MASK		0xf
+#define IFI_CANFD_RXFIFO_DLC_RTR		BIT(4)
+#define IFI_CANFD_RXFIFO_DLC_EDL		BIT(5)
+#define IFI_CANFD_RXFIFO_DLC_BRS		BIT(6)
+#define IFI_CANFD_RXFIFO_DLC_ESI		BIT(7)
+#define IFI_CANFD_RXFIFO_DLC_OBJ_OFFSET		8
+#define IFI_CANFD_RXFIFO_DLC_OBJ_MASK		0x1ff
+#define IFI_CANFD_RXFIFO_DLC_FNR_OFFSET		24
+#define IFI_CANFD_RXFIFO_DLC_FNR_MASK		0xff
+
+#define IFI_CANFD_RXFIFO_ID			0x6c
+#define IFI_CANFD_RXFIFO_ID_ID_OFFSET		0
+#define IFI_CANFD_RXFIFO_ID_ID_STD_MASK		0x3ff
+#define IFI_CANFD_RXFIFO_ID_ID_XTD_MASK		0x1fffffff
+#define IFI_CANFD_RXFIFO_ID_IDE			BIT(29)
+
+#define IFI_CANFD_RXFIFO_DATA			0x70	/* 0x70..0xac */
+
+#define IFI_CANFD_TXFIFO_SUSPEND_US		0xb0
+
+#define IFI_CANFD_TXFIFO_REPEATCOUNT		0xb4
+
+#define IFI_CANFD_TXFIFO_DLC			0xb8
+#define IFI_CANFD_TXFIFO_DLC_DLC_OFFSET		0
+#define IFI_CANFD_TXFIFO_DLC_DLC_MASK		0xf
+#define IFI_CANFD_TXFIFO_DLC_RTR		BIT(4)
+#define IFI_CANFD_TXFIFO_DLC_EDL		BIT(5)
+#define IFI_CANFD_TXFIFO_DLC_BRS		BIT(6)
+#define IFI_CANFD_TXFIFO_DLC_FNR_OFFSET		24
+#define IFI_CANFD_TXFIFO_DLC_FNR_MASK		0xff
+
+#define IFI_CANFD_TXFIFO_ID			0xbc
+#define IFI_CANFD_TXFIFO_ID_ID_OFFSET		0
+#define IFI_CANFD_TXFIFO_ID_ID_STD_MASK		0x3ff
+#define IFI_CANFD_TXFIFO_ID_ID_XTD_MASK		0x1fffffff
+#define IFI_CANFD_TXFIFO_ID_IDE			BIT(29)
+
+#define IFI_CANFD_TXFIFO_DATA			0xc0	/* 0xb0..0xfc */
+
+#define IFI_CANFD_FILTER_MASK(n)		(0x800 + ((n) * 8) + 0)
+#define IFI_CANFD_FILTER_MASK_EXT		BIT(29)
+#define IFI_CANFD_FILTER_MASK_EDL		BIT(30)
+#define IFI_CANFD_FILTER_MASK_VALID		((u32)BIT(31))
+
+#define IFI_CANFD_FILTER_IDENT(n)		(0x800 + ((n) * 8) + 4)
+#define IFI_CANFD_FILTER_IDENT_IDE		BIT(29)
+#define IFI_CANFD_FILTER_IDENT_CANFD		BIT(30)
+#define IFI_CANFD_FILTER_IDENT_VALID		((u32)BIT(31))
+
+/* IFI CANFD private data structure */
+struct ifi_canfd_priv {
+	struct can_priv		can;	/* must be the first member */
+	struct napi_struct	napi;
+	struct net_device	*ndev;
+	void __iomem		*base;
+};
+
+static void ifi_canfd_irq_enable(struct net_device *ndev, bool enable)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	u32 enirq = 0;
+
+	if (enable) {
+		enirq = IFI_CANFD_IRQMASK_TXFIFO_EMPTY |
+			IFI_CANFD_IRQMASK_RXFIFO_NEMPTY;
+	}
+
+	writel(IFI_CANFD_IRQMASK_SET_ERR |
+	       IFI_CANFD_IRQMASK_SET_TS |
+	       IFI_CANFD_IRQMASK_SET_TX |
+	       IFI_CANFD_IRQMASK_SET_RX | enirq,
+	       priv->base + IFI_CANFD_IRQMASK);
+}
+
+static void ifi_canfd_read_fifo(struct net_device *ndev)
+{
+	struct net_device_stats *stats = &ndev->stats;
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	struct canfd_frame *cf;
+	struct sk_buff *skb;
+	const u32 rx_irq_mask = IFI_CANFD_INTERRUPT_RXFIFO_NEMPTY |
+				IFI_CANFD_INTERRUPT_RXFIFO_NEMPTY_PER;
+	u32 rxdlc, rxid;
+	u32 dlc, id;
+	int i;
+
+	rxdlc = readl(priv->base + IFI_CANFD_RXFIFO_DLC);
+	if (rxdlc & IFI_CANFD_RXFIFO_DLC_EDL)
+		skb = alloc_canfd_skb(ndev, &cf);
+	else
+		skb = alloc_can_skb(ndev, (struct can_frame **)&cf);
+
+	if (!skb) {
+		stats->rx_dropped++;
+		return;
+	}
+
+	dlc = (rxdlc >> IFI_CANFD_RXFIFO_DLC_DLC_OFFSET) &
+	      IFI_CANFD_RXFIFO_DLC_DLC_MASK;
+	if (rxdlc & IFI_CANFD_RXFIFO_DLC_EDL)
+		cf->len = can_dlc2len(dlc);
+	else
+		cf->len = get_can_dlc(dlc);
+
+	rxid = readl(priv->base + IFI_CANFD_RXFIFO_ID);
+	id = (rxid >> IFI_CANFD_RXFIFO_ID_ID_OFFSET);
+	if (id & IFI_CANFD_RXFIFO_ID_IDE)
+		id &= IFI_CANFD_RXFIFO_ID_ID_XTD_MASK;
+	else
+		id &= IFI_CANFD_RXFIFO_ID_ID_STD_MASK;
+	cf->can_id = id;
+
+	if (rxdlc & IFI_CANFD_RXFIFO_DLC_ESI) {
+		cf->flags |= CANFD_ESI;
+		netdev_dbg(ndev, "ESI Error\n");
+	}
+
+	if (!(rxdlc & IFI_CANFD_RXFIFO_DLC_EDL) &&
+	    (rxdlc & IFI_CANFD_RXFIFO_DLC_RTR)) {
+		cf->can_id |= CAN_RTR_FLAG;
+	} else {
+		if (rxdlc & IFI_CANFD_RXFIFO_DLC_BRS)
+			cf->flags |= CANFD_BRS;
+
+		for (i = 0; i < cf->len; i += 4) {
+			*(u32 *)(cf->data + i) =
+				readl(priv->base + IFI_CANFD_RXFIFO_DATA + i);
+		}
+	}
+
+	/* Remove the packet from FIFO */
+	writel(IFI_CANFD_RXSTCMD_REMOVE_MSG, priv->base + IFI_CANFD_RXSTCMD);
+	writel(rx_irq_mask, priv->base + IFI_CANFD_INTERRUPT);
+
+	stats->rx_packets++;
+	stats->rx_bytes += cf->len;
+
+	netif_receive_skb(skb);
+}
+
+static int ifi_canfd_do_rx_poll(struct net_device *ndev, int quota)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	u32 pkts = 0;
+	u32 rxst;
+
+	rxst = readl(priv->base + IFI_CANFD_RXSTCMD);
+	if (rxst & IFI_CANFD_RXSTCMD_EMPTY) {
+		netdev_dbg(ndev, "No messages in RX FIFO\n");
+		return 0;
+	}
+
+	for (;;) {
+		if (rxst & IFI_CANFD_RXSTCMD_EMPTY)
+			break;
+		if (quota <= 0)
+			break;
+
+		ifi_canfd_read_fifo(ndev);
+		quota--;
+		pkts++;
+		rxst = readl(priv->base + IFI_CANFD_RXSTCMD);
+	}
+
+	if (pkts)
+		can_led_event(ndev, CAN_LED_EVENT_RX);
+
+	return pkts;
+}
+
+static int ifi_canfd_handle_lost_msg(struct net_device *ndev)
+{
+	struct net_device_stats *stats = &ndev->stats;
+	struct sk_buff *skb;
+	struct can_frame *frame;
+
+	netdev_err(ndev, "RX FIFO overflow, message(s) lost.\n");
+
+	stats->rx_errors++;
+	stats->rx_over_errors++;
+
+	skb = alloc_can_err_skb(ndev, &frame);
+	if (unlikely(!skb))
+		return 0;
+
+	frame->can_id |= CAN_ERR_CRTL;
+	frame->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
+
+	netif_receive_skb(skb);
+
+	return 1;
+}
+
+static int ifi_canfd_get_berr_counter(const struct net_device *ndev,
+				      struct can_berr_counter *bec)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	u32 err;
+
+	err = readl(priv->base + IFI_CANFD_ERROR);
+	bec->rxerr = (err >> IFI_CANFD_ERROR_RX_OFFSET) &
+		     IFI_CANFD_ERROR_RX_MASK;
+	bec->txerr = (err >> IFI_CANFD_ERROR_TX_OFFSET) &
+		     IFI_CANFD_ERROR_TX_MASK;
+
+	return 0;
+}
+
+static int ifi_canfd_handle_state_change(struct net_device *ndev,
+					 enum can_state new_state)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	struct can_frame *cf;
+	struct sk_buff *skb;
+	struct can_berr_counter bec;
+
+	switch (new_state) {
+	case CAN_STATE_ERROR_ACTIVE:
+		/* error warning state */
+		priv->can.can_stats.error_warning++;
+		priv->can.state = CAN_STATE_ERROR_WARNING;
+		break;
+	case CAN_STATE_ERROR_PASSIVE:
+		/* error passive state */
+		priv->can.can_stats.error_passive++;
+		priv->can.state = CAN_STATE_ERROR_PASSIVE;
+		break;
+	case CAN_STATE_BUS_OFF:
+		/* bus-off state */
+		priv->can.state = CAN_STATE_BUS_OFF;
+		ifi_canfd_irq_enable(ndev, 0);
+		priv->can.can_stats.bus_off++;
+		can_bus_off(ndev);
+		break;
+	default:
+		break;
+	}
+
+	/* propagate the error condition to the CAN stack */
+	skb = alloc_can_err_skb(ndev, &cf);
+	if (unlikely(!skb))
+		return 0;
+
+	ifi_canfd_get_berr_counter(ndev, &bec);
+
+	switch (new_state) {
+	case CAN_STATE_ERROR_ACTIVE:
+		/* error warning state */
+		cf->can_id |= CAN_ERR_CRTL;
+		cf->data[1] = (bec.txerr > bec.rxerr) ?
+			CAN_ERR_CRTL_TX_WARNING :
+			CAN_ERR_CRTL_RX_WARNING;
+		cf->data[6] = bec.txerr;
+		cf->data[7] = bec.rxerr;
+		break;
+	case CAN_STATE_ERROR_PASSIVE:
+		/* error passive state */
+		cf->can_id |= CAN_ERR_CRTL;
+		cf->data[1] |= CAN_ERR_CRTL_RX_PASSIVE;
+		if (bec.txerr > 127)
+			cf->data[1] |= CAN_ERR_CRTL_TX_PASSIVE;
+		cf->data[6] = bec.txerr;
+		cf->data[7] = bec.rxerr;
+		break;
+	case CAN_STATE_BUS_OFF:
+		/* bus-off state */
+		cf->can_id |= CAN_ERR_BUSOFF;
+		break;
+	default:
+		break;
+	}
+
+	stats->rx_packets++;
+	stats->rx_bytes += cf->can_dlc;
+	netif_receive_skb(skb);
+
+	return 1;
+}
+
+static int ifi_canfd_handle_state_errors(struct net_device *ndev, u32 stcmd)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	int work_done = 0;
+	u32 isr;
+
+	/*
+	 * The ErrWarn condition is a little special, since the bit is
+	 * located in the INTERRUPT register instead of STCMD register.
+	 */
+	isr = readl(priv->base + IFI_CANFD_INTERRUPT);
+	if ((isr & IFI_CANFD_INTERRUPT_ERROR_WARNING) &&
+	    (priv->can.state != CAN_STATE_ERROR_WARNING)) {
+		/* Clear the interrupt */
+		writel(IFI_CANFD_INTERRUPT_ERROR_WARNING,
+		       priv->base + IFI_CANFD_INTERRUPT);
+		netdev_dbg(ndev, "Error, entered warning state\n");
+		work_done += ifi_canfd_handle_state_change(ndev,
+						CAN_STATE_ERROR_WARNING);
+	}
+
+	if ((stcmd & IFI_CANFD_STCMD_ERROR_PASSIVE) &&
+	    (priv->can.state != CAN_STATE_ERROR_PASSIVE)) {
+		netdev_dbg(ndev, "Error, entered passive state\n");
+		work_done += ifi_canfd_handle_state_change(ndev,
+						CAN_STATE_ERROR_PASSIVE);
+	}
+
+	if ((stcmd & IFI_CANFD_STCMD_BUSOFF) &&
+	    (priv->can.state != CAN_STATE_BUS_OFF)) {
+		netdev_dbg(ndev, "Error, entered bus-off state\n");
+		work_done += ifi_canfd_handle_state_change(ndev,
+						CAN_STATE_BUS_OFF);
+	}
+
+	return work_done;
+}
+
+static int ifi_canfd_poll(struct napi_struct *napi, int quota)
+{
+	struct net_device *ndev = napi->dev;
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	const u32 stcmd_state_mask = IFI_CANFD_STCMD_ERROR_PASSIVE |
+				     IFI_CANFD_STCMD_BUSOFF;
+	int work_done = 0;
+
+	u32 stcmd = readl(priv->base + IFI_CANFD_STCMD);
+	u32 rxstcmd = readl(priv->base + IFI_CANFD_STCMD);
+
+	/* Handle bus state changes */
+	if ((stcmd & stcmd_state_mask) ||
+	    ((stcmd & IFI_CANFD_STCMD_ERROR_ACTIVE) == 0))
+		work_done += ifi_canfd_handle_state_errors(ndev, stcmd);
+
+	/* Handle lost messages on RX */
+	if (rxstcmd & IFI_CANFD_RXSTCMD_OVERFLOW)
+		work_done += ifi_canfd_handle_lost_msg(ndev);
+
+	/* Handle normal messages on RX */
+	if (!(rxstcmd & IFI_CANFD_RXSTCMD_EMPTY))
+		work_done += ifi_canfd_do_rx_poll(ndev, quota - work_done);
+
+	if (work_done < quota) {
+		napi_complete(napi);
+		ifi_canfd_irq_enable(ndev, 1);
+	}
+
+	return work_done;
+}
+
+static irqreturn_t ifi_canfd_isr(int irq, void *dev_id)
+{
+	struct net_device *ndev = (struct net_device *)dev_id;
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	struct net_device_stats *stats = &ndev->stats;
+	const u32 rx_irq_mask = IFI_CANFD_INTERRUPT_RXFIFO_NEMPTY |
+				IFI_CANFD_INTERRUPT_RXFIFO_NEMPTY_PER;
+	const u32 tx_irq_mask = IFI_CANFD_INTERRUPT_TXFIFO_EMPTY |
+				IFI_CANFD_INTERRUPT_TXFIFO_REMOVE;
+	const u32 clr_irq_mask = ~(IFI_CANFD_INTERRUPT_SET_IRQ |
+				   IFI_CANFD_INTERRUPT_ERROR_WARNING);
+	u32 isr;
+
+	isr = readl(priv->base + IFI_CANFD_INTERRUPT);
+
+	/* No interrupt */
+	if (isr == 0)
+		return IRQ_NONE;
+
+	/* Clear all pending interrupts but ErrWarn */
+	writel(clr_irq_mask, priv->base + IFI_CANFD_INTERRUPT);
+
+	/* RX IRQ, start NAPI */
+	if (isr & rx_irq_mask) {
+		ifi_canfd_irq_enable(ndev, 0);
+		napi_schedule(&priv->napi);
+	}
+
+	/* TX IRQ */
+	if (isr & tx_irq_mask) {
+		stats->tx_bytes += can_get_echo_skb(ndev, 0);
+		stats->tx_packets++;
+		can_led_event(ndev, CAN_LED_EVENT_TX);
+		netif_wake_queue(ndev);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static const struct can_bittiming_const ifi_canfd_bittiming_const = {
+	.name		= KBUILD_MODNAME,
+	.tseg1_min	= 2,	/* Time segment 1 = prop_seg + phase_seg1 */
+	.tseg1_max	= 64,
+	.tseg2_min	= 1,	/* Time segment 2 = phase_seg2 */
+	.tseg2_max	= 16,
+	.sjw_max	= 16,
+	.brp_min	= 1,
+	.brp_max	= 1024,
+	.brp_inc	= 1,
+};
+
+static const struct can_bittiming_const ifi_canfd_data_bittiming_const = {
+	.name		= KBUILD_MODNAME,
+	.tseg1_min	= 2,	/* Time segment 1 = prop_seg + phase_seg1 */
+	.tseg1_max	= 16,
+	.tseg2_min	= 1,	/* Time segment 2 = phase_seg2 */
+	.tseg2_max	= 8,
+	.sjw_max	= 4,
+	.brp_min	= 1,
+	.brp_max	= 32,
+	.brp_inc	= 1,
+};
+
+static void ifi_canfd_set_bittiming(struct net_device *ndev)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	const struct can_bittiming *bt = &priv->can.bittiming;
+	const struct can_bittiming *dbt = &priv->can.data_bittiming;
+	u16 brp, sjw, tseg1, tseg2;
+	u32 noniso_arg = 0;
+	u32 time_off;
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_FD_NON_ISO) {
+		noniso_arg = IFI_CANFD_TIME_SET_TIMEB_BOSCH |
+			     IFI_CANFD_TIME_SET_TIMEA_BOSCH |
+			     IFI_CANFD_TIME_SET_PRESC_BOSCH |
+			     IFI_CANFD_TIME_SET_SJW_BOSCH;
+		time_off = IFI_CANFD_TIME_SJW_OFF_BOSCH;
+	} else {
+		time_off = IFI_CANFD_TIME_SJW_OFF_ISO;
+	}
+
+	/* Configure bit timing */
+	brp = bt->brp - 1;
+	sjw = bt->sjw - 1;
+	tseg1 = bt->prop_seg + bt->phase_seg1 - 1;
+	tseg2 = bt->phase_seg2 - 1;
+	writel((tseg2 << IFI_CANFD_TIME_TIMEB_OFF) |
+	       (tseg1 << IFI_CANFD_TIME_TIMEA_OFF) |
+	       (brp << IFI_CANFD_TIME_PRESCALE_OFF) |
+	       (sjw << time_off),
+	       priv->base + IFI_CANFD_TIME);
+
+	/* Configure data bit timing */
+	brp = dbt->brp - 1;
+	sjw = dbt->sjw - 1;
+	tseg1 = dbt->prop_seg + dbt->phase_seg1 - 1;
+	tseg2 = dbt->phase_seg2 - 1;
+	writel((tseg2 << IFI_CANFD_TIME_TIMEB_OFF) |
+	       (tseg1 << IFI_CANFD_TIME_TIMEA_OFF) |
+	       (brp << IFI_CANFD_TIME_PRESCALE_OFF) |
+	       (sjw << time_off) |
+	       noniso_arg,
+	       priv->base + IFI_CANFD_FTIME);
+}
+
+static void ifi_canfd_set_filter(struct net_device *ndev, const u32 id,
+				 const u32 mask, const u32 ident)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+
+	writel(mask, priv->base + IFI_CANFD_FILTER_MASK(id));
+	writel(ident, priv->base + IFI_CANFD_FILTER_IDENT(id));
+}
+
+static void ifi_canfd_set_filters(struct net_device *ndev)
+{
+	/* Receive all CAN frames (standard ID) */
+	ifi_canfd_set_filter(ndev, 0,
+			     IFI_CANFD_FILTER_MASK_VALID |
+			     IFI_CANFD_FILTER_MASK_EXT,
+			     IFI_CANFD_FILTER_IDENT_VALID);
+
+	/* Receive all CAN frames (extended ID) */
+	ifi_canfd_set_filter(ndev, 1,
+			     IFI_CANFD_FILTER_MASK_VALID |
+			     IFI_CANFD_FILTER_MASK_EXT,
+			     IFI_CANFD_FILTER_IDENT_VALID |
+			     IFI_CANFD_FILTER_IDENT_IDE);
+
+	/* Receive all CANFD frames */
+	ifi_canfd_set_filter(ndev, 2,
+			     IFI_CANFD_FILTER_MASK_VALID |
+			     IFI_CANFD_FILTER_MASK_EDL |
+			     IFI_CANFD_FILTER_MASK_EXT,
+			     IFI_CANFD_FILTER_IDENT_VALID |
+			     IFI_CANFD_FILTER_IDENT_CANFD |
+			     IFI_CANFD_FILTER_IDENT_IDE);
+}
+
+static void ifi_canfd_start(struct net_device *ndev)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	u32 stcmd;
+
+	/* Reset the IP */
+	writel(IFI_CANFD_STCMD_HARDRESET, priv->base + IFI_CANFD_STCMD);
+	writel(0, priv->base + IFI_CANFD_STCMD);
+
+	ifi_canfd_set_bittiming(ndev);
+	ifi_canfd_set_filters(ndev);
+
+	/* Reset FIFOs */
+	writel(IFI_CANFD_RXSTCMD_RESET, priv->base + IFI_CANFD_RXSTCMD);
+	writel(0, priv->base + IFI_CANFD_RXSTCMD);
+	writel(IFI_CANFD_TXSTCMD_RESET, priv->base + IFI_CANFD_TXSTCMD);
+	writel(0, priv->base + IFI_CANFD_TXSTCMD);
+
+	/* Repeat transmission until successful */
+	writel(0, priv->base + IFI_CANFD_REPEAT);
+	writel(0, priv->base + IFI_CANFD_SUSPEND);
+
+	/* Clear all pending interrupts */
+	writel((u32)(~IFI_CANFD_INTERRUPT_SET_IRQ),
+	       priv->base + IFI_CANFD_INTERRUPT);
+
+	stcmd = IFI_CANFD_STCMD_ENABLE | IFI_CANFD_STCMD_NORMAL_MODE;
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY)
+		stcmd |= IFI_CANFD_STCMD_BUSMONITOR;
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_LOOPBACK)
+		stcmd |= IFI_CANFD_STCMD_LOOPBACK;
+
+	if (priv->can.ctrlmode & CAN_CTRLMODE_FD)
+		stcmd |= IFI_CANFD_STCMD_ENABLE_ISO;
+
+	if (!(priv->can.ctrlmode & (CAN_CTRLMODE_FD | CAN_CTRLMODE_FD_NON_ISO)))
+		stcmd |= IFI_CANFD_STCMD_DISABLE_CANFD;
+
+	priv->can.state = CAN_STATE_ERROR_ACTIVE;
+
+	ifi_canfd_irq_enable(ndev, 1);
+
+	/* Enable controller */
+	writel(stcmd, priv->base + IFI_CANFD_STCMD);
+}
+
+static void ifi_canfd_stop(struct net_device *ndev)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+
+	/* Reset the IP */
+	writel(IFI_CANFD_STCMD_HARDRESET, priv->base + IFI_CANFD_STCMD);
+
+	/* Mask all interrupts */
+	writel(~0, priv->base + IFI_CANFD_IRQMASK);
+
+	/* Clear all pending interrupts */
+	writel((u32)(~IFI_CANFD_INTERRUPT_SET_IRQ),
+	       priv->base + IFI_CANFD_INTERRUPT);
+
+	/* Set the state as STOPPED */
+	priv->can.state = CAN_STATE_STOPPED;
+}
+
+static int ifi_canfd_set_mode(struct net_device *ndev, enum can_mode mode)
+{
+	switch (mode) {
+	case CAN_MODE_START:
+		ifi_canfd_start(ndev);
+		netif_wake_queue(ndev);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int ifi_canfd_open(struct net_device *ndev)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	int ret;
+
+	ret = open_candev(ndev);
+	if (ret) {
+		netdev_err(ndev, "Failed to open CAN device\n");
+		return ret;
+	}
+
+	/* Register interrupt handler */
+	ret = request_irq(ndev->irq, ifi_canfd_isr, IRQF_SHARED,
+			  ndev->name, ndev);
+	if (ret < 0) {
+		netdev_err(ndev, "Failed to request interrupt\n");
+		goto err_irq;
+	}
+
+	ifi_canfd_start(ndev);
+
+	can_led_event(ndev, CAN_LED_EVENT_OPEN);
+	napi_enable(&priv->napi);
+	netif_start_queue(ndev);
+
+	return 0;
+err_irq:
+	close_candev(ndev);
+	return ret;
+}
+
+static int ifi_canfd_close(struct net_device *ndev)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+
+	netif_stop_queue(ndev);
+	napi_disable(&priv->napi);
+
+	ifi_canfd_stop(ndev);
+
+	free_irq(ndev->irq, ndev);
+
+	close_candev(ndev);
+
+	can_led_event(ndev, CAN_LED_EVENT_STOP);
+
+	return 0;
+}
+
+static netdev_tx_t ifi_canfd_start_xmit(struct sk_buff *skb,
+					struct net_device *ndev)
+{
+	struct ifi_canfd_priv *priv = netdev_priv(ndev);
+	struct canfd_frame *cf = (struct canfd_frame *)skb->data;
+	u32 txst, txid;
+	u32 txdlc = 0;
+	int i;
+
+	if (can_dropped_invalid_skb(ndev, skb))
+		return NETDEV_TX_OK;
+
+	/* Check if the TX buffer is full */
+	txst = readl(priv->base + IFI_CANFD_TXSTCMD);
+	if (txst & IFI_CANFD_TXSTCMD_FULL) {
+		netif_stop_queue(ndev);
+		netdev_err(ndev, "BUG! TX FIFO full when queue awake!\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	netif_stop_queue(ndev);
+
+	if (cf->can_id & CAN_EFF_FLAG) {
+		txid = cf->can_id & CAN_EFF_MASK;
+		txid |= IFI_CANFD_TXFIFO_ID_IDE;
+	} else {
+		txid = cf->can_id & CAN_SFF_MASK;
+	}
+
+	if (priv->can.ctrlmode & (CAN_CTRLMODE_FD | CAN_CTRLMODE_FD_NON_ISO)) {
+		if (can_is_canfd_skb(skb)) {
+			txdlc |= IFI_CANFD_TXFIFO_DLC_EDL;
+			if (cf->flags & CANFD_BRS)
+				txdlc |= IFI_CANFD_TXFIFO_DLC_BRS;
+		}
+	}
+
+	if (cf->can_id & CAN_RTR_FLAG)
+		txdlc |= IFI_CANFD_TXFIFO_DLC_RTR;
+
+	/* message ram configuration */
+	writel(txid, priv->base + IFI_CANFD_TXFIFO_ID);
+	writel(txdlc, priv->base + IFI_CANFD_TXFIFO_DLC);
+
+	for (i = 0; i < cf->len; i += 4) {
+		writel(*(u32 *)(cf->data + i),
+		       priv->base + IFI_CANFD_TXFIFO_DATA + i);
+	}
+
+	writel(0, priv->base + IFI_CANFD_TXFIFO_REPEATCOUNT);
+	writel(0, priv->base + IFI_CANFD_TXFIFO_SUSPEND_US);
+
+	can_put_echo_skb(skb, ndev, 0);
+
+	/* Start the transmission */
+	writel(IFI_CANFD_TXSTCMD_ADD_MSG, priv->base + IFI_CANFD_TXSTCMD);
+
+	return NETDEV_TX_OK;
+}
+
+static const struct net_device_ops ifi_canfd_netdev_ops = {
+	.ndo_open	= ifi_canfd_open,
+	.ndo_stop	= ifi_canfd_close,
+	.ndo_start_xmit	= ifi_canfd_start_xmit,
+	.ndo_change_mtu	= can_change_mtu,
+};
+
+static int ifi_canfd_plat_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct net_device *ndev;
+	struct ifi_canfd_priv *priv;
+	struct resource *res;
+	void __iomem *addr;
+	int irq, ret;
+	u32 id;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	addr = devm_ioremap_resource(dev, res);
+	irq = platform_get_irq(pdev, 0);
+	if (IS_ERR(addr) || irq < 0)
+		return -EINVAL;
+
+	id = readl(addr + IFI_CANFD_IP_ID);
+	if (id != IFI_CANFD_IP_ID_VALUE) {
+		dev_err(dev, "This block is not IFI CANFD, id=%08x\n", id);
+		return -EINVAL;
+	}
+
+	ndev = alloc_candev(sizeof(*priv), 1);
+	if (!ndev)
+		return -ENOMEM;
+
+	ndev->irq = irq;
+	ndev->flags |= IFF_ECHO;	/* we support local echo */
+	ndev->netdev_ops = &ifi_canfd_netdev_ops;
+
+	priv = netdev_priv(ndev);
+	priv->ndev = ndev;
+	priv->base = addr;
+
+	netif_napi_add(ndev, &priv->napi, ifi_canfd_poll, 64);
+
+	priv->can.state = CAN_STATE_STOPPED;
+
+	priv->can.clock.freq = readl(addr + IFI_CANFD_SYSCLOCK);
+
+	priv->can.bittiming_const	= &ifi_canfd_bittiming_const;
+	priv->can.data_bittiming_const	= &ifi_canfd_data_bittiming_const;
+	priv->can.do_set_mode		= ifi_canfd_set_mode;
+	priv->can.do_get_berr_counter	= ifi_canfd_get_berr_counter;
+
+	/* IFI CANFD can do both Bosch FD and ISO FD */
+	priv->can.ctrlmode = CAN_CTRLMODE_FD;
+
+	/* IFI CANFD can do both Bosch FD and ISO FD */
+	priv->can.ctrlmode_supported = CAN_CTRLMODE_LOOPBACK |
+				       CAN_CTRLMODE_LISTENONLY |
+				       CAN_CTRLMODE_FD |
+				       CAN_CTRLMODE_FD_NON_ISO;
+
+	platform_set_drvdata(pdev, ndev);
+	SET_NETDEV_DEV(ndev, dev);
+
+	ret = register_candev(ndev);
+	if (ret) {
+		dev_err(dev, "Failed to register (ret=%d)\n", ret);
+		goto err_reg;
+	}
+
+	devm_can_led_init(ndev);
+
+	dev_info(dev, "Driver registered: regs=%p, irq=%d, clock=%d\n",
+		 priv->base, ndev->irq, priv->can.clock.freq);
+
+	return 0;
+
+err_reg:
+	free_candev(ndev);
+	return ret;
+}
+
+static int ifi_canfd_plat_remove(struct platform_device *pdev)
+{
+	struct net_device *ndev = platform_get_drvdata(pdev);
+
+	unregister_candev(ndev);
+	platform_set_drvdata(pdev, NULL);
+	free_candev(ndev);
+
+	return 0;
+}
+
+static const struct of_device_id ifi_canfd_of_table[] = {
+	{ .compatible = "ifi,canfd-1.0", .data = NULL },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, ifi_canfd_of_table);
+
+static struct platform_driver ifi_canfd_plat_driver = {
+	.driver = {
+		.name		= KBUILD_MODNAME,
+		.of_match_table	= ifi_canfd_of_table,
+	},
+	.probe	= ifi_canfd_plat_probe,
+	.remove	= ifi_canfd_plat_remove,
+};
+
+module_platform_driver(ifi_canfd_plat_driver);
+
+MODULE_AUTHOR("Marek Vasut <marex@denx.de>");
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("CAN bus driver for IFI CANFD controller");
-- 
cgit v0.10.2


From e817af27e01f9294055a441d1034bbc353ee39d2 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 18 Feb 2016 10:00:11 +0000
Subject: vlan: turn on unicast filtering on vlan device

Currently vlan device inherits unicast filtering flag from underlying
device. If underlying device doesn't support unicast filter, this will
put vlan device into promiscuous mode when it's stacked.

Tun on IFF_UNICAST_FLT on the vlan device in any case so that it does
not go into promiscuous mode needlessly. If underlying device does not
support unicast filtering, that device will enter promiscuous mode.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index d2cd9de..69929c0 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -261,7 +261,6 @@ static int register_vlan_device(struct net_device *real_dev, u16 vlan_id)
 	 * hope the underlying device can handle it.
 	 */
 	new_dev->mtu = real_dev->mtu;
-	new_dev->priv_flags |= (real_dev->priv_flags & IFF_UNICAST_FLT);
 
 	vlan = vlan_dev_priv(new_dev);
 	vlan->vlan_proto = htons(ETH_P_8021Q);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index ad5e2fd..055f0e9 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -799,6 +799,7 @@ void vlan_setup(struct net_device *dev)
 	ether_setup(dev);
 
 	dev->priv_flags		|= IFF_802_1Q_VLAN | IFF_NO_QUEUE;
+	dev->priv_flags		|= IFF_UNICAST_FLT;
 	dev->priv_flags		&= ~IFF_TX_SKB_SHARING;
 	netif_keep_dst(dev);
 
-- 
cgit v0.10.2


From 745041e2aaf1d668f293aaab4b0f6ad7daa056a5 Mon Sep 17 00:00:00 2001
From: Robert Shearman <rshearma@brocade.com>
Date: Fri, 19 Feb 2016 09:43:16 +0000
Subject: lwtunnel: autoload of lwt modules

The lwt implementations using net devices can autoload using the
existing mechanism using IFLA_INFO_KIND. However, there's no mechanism
that lwt modules not using net devices can use.

Therefore, add the ability to autoload modules registering lwt
operations for lwt implementations not using a net device so that
users don't have to manually load the modules.

Only users with the CAP_NET_ADMIN capability can cause modules to be
loaded, which is ensured by rtnetlink_rcv_msg rejecting non-RTM_GETxxx
messages for users without this capability, and by
lwtunnel_build_state not being called in response to RTM_GETxxx
messages.

Signed-off-by: Robert Shearman <rshearma@brocade.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/lwtunnel.h b/include/net/lwtunnel.h
index 66350ce..e9f116e 100644
--- a/include/net/lwtunnel.h
+++ b/include/net/lwtunnel.h
@@ -170,6 +170,8 @@ static inline int lwtunnel_input(struct sk_buff *skb)
 	return -EOPNOTSUPP;
 }
 
-#endif
+#endif /* CONFIG_LWTUNNEL */
+
+#define MODULE_ALIAS_RTNL_LWT(encap_type) MODULE_ALIAS("rtnl-lwt-" __stringify(encap_type))
 
 #endif /* __NET_LWTUNNEL_H */
diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c
index 299cfc2..669ecc9 100644
--- a/net/core/lwtunnel.c
+++ b/net/core/lwtunnel.c
@@ -27,6 +27,31 @@
 #include <net/rtnetlink.h>
 #include <net/ip6_fib.h>
 
+#ifdef CONFIG_MODULES
+
+static const char *lwtunnel_encap_str(enum lwtunnel_encap_types encap_type)
+{
+	/* Only lwt encaps implemented without using an interface for
+	 * the encap need to return a string here.
+	 */
+	switch (encap_type) {
+	case LWTUNNEL_ENCAP_MPLS:
+		return "MPLS";
+	case LWTUNNEL_ENCAP_ILA:
+		return "ILA";
+	case LWTUNNEL_ENCAP_IP6:
+	case LWTUNNEL_ENCAP_IP:
+	case LWTUNNEL_ENCAP_NONE:
+	case __LWTUNNEL_ENCAP_MAX:
+		/* should not have got here */
+		WARN_ON(1);
+		break;
+	}
+	return NULL;
+}
+
+#endif /* CONFIG_MODULES */
+
 struct lwtunnel_state *lwtunnel_state_alloc(int encap_len)
 {
 	struct lwtunnel_state *lws;
@@ -85,6 +110,18 @@ int lwtunnel_build_state(struct net_device *dev, u16 encap_type,
 	ret = -EOPNOTSUPP;
 	rcu_read_lock();
 	ops = rcu_dereference(lwtun_encaps[encap_type]);
+#ifdef CONFIG_MODULES
+	if (!ops) {
+		const char *encap_type_str = lwtunnel_encap_str(encap_type);
+
+		if (encap_type_str) {
+			rcu_read_unlock();
+			request_module("rtnl-lwt-%s", encap_type_str);
+			rcu_read_lock();
+			ops = rcu_dereference(lwtun_encaps[encap_type]);
+		}
+	}
+#endif
 	if (likely(ops && ops->build_state))
 		ret = ops->build_state(dev, encap, family, cfg, lws);
 	rcu_read_unlock();
-- 
cgit v0.10.2


From b2b04edceb7cc3c3023fea7d4ccc7dd243e839a9 Mon Sep 17 00:00:00 2001
From: Robert Shearman <rshearma@brocade.com>
Date: Fri, 19 Feb 2016 09:43:17 +0000
Subject: mpls: autoload lwt module

Avoid users having to manually load the module by adding a module
alias allowing it to be autoloaded by the lwt infra.

Signed-off-by: Robert Shearman <rshearma@brocade.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c
index fb31aa8..644a8da 100644
--- a/net/mpls/mpls_iptunnel.c
+++ b/net/mpls/mpls_iptunnel.c
@@ -227,5 +227,6 @@ static void __exit mpls_iptunnel_exit(void)
 }
 module_exit(mpls_iptunnel_exit);
 
+MODULE_ALIAS_RTNL_LWT(MPLS);
 MODULE_DESCRIPTION("MultiProtocol Label Switching IP Tunnels");
 MODULE_LICENSE("GPL v2");
-- 
cgit v0.10.2


From 84a8cbe46af0945859a8f61bce6100ddc3d51a51 Mon Sep 17 00:00:00 2001
From: Robert Shearman <rshearma@brocade.com>
Date: Fri, 19 Feb 2016 09:43:18 +0000
Subject: ila: autoload module

Avoid users having to manually load the module by adding a module
alias allowing it to be autoloaded by the lwt infra.

Signed-off-by: Robert Shearman <rshearma@brocade.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv6/ila/ila_common.c b/net/ipv6/ila/ila_common.c
index 32dc9aa..3061305 100644
--- a/net/ipv6/ila/ila_common.c
+++ b/net/ipv6/ila/ila_common.c
@@ -99,5 +99,6 @@ static void __exit ila_fini(void)
 
 module_init(ila_init);
 module_exit(ila_fini);
+MODULE_ALIAS_RTNL_LWT(ILA);
 MODULE_AUTHOR("Tom Herbert <tom@herbertland.com>");
 MODULE_LICENSE("GPL");
-- 
cgit v0.10.2


From 14f1f724355206dea1cf3f23ee87993bfd47c70c Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 19 Feb 2016 11:26:24 -0800
Subject: GENEVE: Support outer IPv4 Tx checksums by default

This change makes it so that if UDP CSUM is not specified we will default
to enabling it.  The main motivation behind this is the fact that with the
use of outer checksum we can greatly improve the performance for GENEVE
tunnels on hardware that doesn't know how to parse them.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index dfbe3ca..8fa8388 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -76,7 +76,7 @@ struct geneve_dev {
 };
 
 /* Geneve device flags */
-#define GENEVE_F_UDP_CSUM		BIT(0)
+#define GENEVE_F_UDP_ZERO_CSUM_TX	BIT(0)
 #define GENEVE_F_UDP_ZERO_CSUM6_TX	BIT(1)
 #define GENEVE_F_UDP_ZERO_CSUM6_RX	BIT(2)
 
@@ -703,7 +703,7 @@ static int geneve_build_skb(struct rtable *rt, struct sk_buff *skb,
 	struct genevehdr *gnvh;
 	int min_headroom;
 	int err;
-	bool udp_sum = !!(flags & GENEVE_F_UDP_CSUM);
+	bool udp_sum = !(flags & GENEVE_F_UDP_ZERO_CSUM_TX);
 
 	skb_scrub_packet(skb, xnet);
 
@@ -944,9 +944,9 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 			opts = ip_tunnel_info_opts(info);
 
 		if (key->tun_flags & TUNNEL_CSUM)
-			flags |= GENEVE_F_UDP_CSUM;
+			flags &= ~GENEVE_F_UDP_ZERO_CSUM_TX;
 		else
-			flags &= ~GENEVE_F_UDP_CSUM;
+			flags |= GENEVE_F_UDP_ZERO_CSUM_TX;
 
 		err = geneve_build_skb(rt, skb, key->tun_flags, vni,
 				       info->options_len, opts, flags, xnet);
@@ -972,7 +972,7 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 	udp_tunnel_xmit_skb(rt, gs4->sock->sk, skb, fl4.saddr, fl4.daddr,
 			    tos, ttl, df, sport, geneve->dst_port,
 			    !net_eq(geneve->net, dev_net(geneve->dev)),
-			    !(flags & GENEVE_F_UDP_CSUM));
+			    !!(flags & GENEVE_F_UDP_ZERO_CSUM_TX));
 
 	return NETDEV_TX_OK;
 
@@ -1383,8 +1383,8 @@ static int geneve_newlink(struct net *net, struct net_device *dev,
 		metadata = true;
 
 	if (data[IFLA_GENEVE_UDP_CSUM] &&
-	    nla_get_u8(data[IFLA_GENEVE_UDP_CSUM]))
-		flags |= GENEVE_F_UDP_CSUM;
+	    !nla_get_u8(data[IFLA_GENEVE_UDP_CSUM]))
+		flags |= GENEVE_F_UDP_ZERO_CSUM_TX;
 
 	if (data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX] &&
 	    nla_get_u8(data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX]))
@@ -1454,7 +1454,7 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev)
 	}
 
 	if (nla_put_u8(skb, IFLA_GENEVE_UDP_CSUM,
-		       !!(geneve->flags & GENEVE_F_UDP_CSUM)) ||
+		       !(geneve->flags & GENEVE_F_UDP_ZERO_CSUM_TX)) ||
 	    nla_put_u8(skb, IFLA_GENEVE_UDP_ZERO_CSUM6_TX,
 		       !!(geneve->flags & GENEVE_F_UDP_ZERO_CSUM6_TX)) ||
 	    nla_put_u8(skb, IFLA_GENEVE_UDP_ZERO_CSUM6_RX,
-- 
cgit v0.10.2


From 6ceb31ca5f65acff299dbc3da5854d54e147b7d8 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 19 Feb 2016 11:26:31 -0800
Subject: VXLAN: Support outer IPv4 Tx checksums by default

This change makes it so that if UDP CSUM is not specified we will default
to enabling it.  The main motivation behind this is the fact that with the
use of outer checksum we can greatly improve the performance for VXLAN
tunnels on devices that don't know how to parse tunnel headers.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index c963897..2ddc642 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1957,13 +1957,6 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			goto drop;
 		sk = vxlan->vn4_sock->sock->sk;
 
-		if (info) {
-			if (info->key.tun_flags & TUNNEL_DONT_FRAGMENT)
-				df = htons(IP_DF);
-		} else {
-			udp_sum = !!(flags & VXLAN_F_UDP_CSUM);
-		}
-
 		rt = vxlan_get_route(vxlan, skb,
 				     rdst ? rdst->remote_ifindex : 0, tos,
 				     dst->sin.sin_addr.s_addr, &saddr,
@@ -1997,6 +1990,11 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			return;
 		}
 
+		if (!info)
+			udp_sum = !(flags & VXLAN_F_UDP_ZERO_CSUM_TX);
+		else if (info->key.tun_flags & TUNNEL_DONT_FRAGMENT)
+			df = htons(IP_DF);
+
 		tos = ip_tunnel_ecn_encap(tos, old_iph, skb);
 		ttl = ttl ? : ip4_dst_hoplimit(&rt->dst);
 		err = vxlan_build_skb(skb, &rt->dst, sizeof(struct iphdr),
@@ -2898,8 +2896,9 @@ static int vxlan_newlink(struct net *src_net, struct net_device *dev,
 	if (data[IFLA_VXLAN_PORT])
 		conf.dst_port = nla_get_be16(data[IFLA_VXLAN_PORT]);
 
-	if (data[IFLA_VXLAN_UDP_CSUM] && nla_get_u8(data[IFLA_VXLAN_UDP_CSUM]))
-		conf.flags |= VXLAN_F_UDP_CSUM;
+	if (data[IFLA_VXLAN_UDP_CSUM] &&
+	    !nla_get_u8(data[IFLA_VXLAN_UDP_CSUM]))
+		conf.flags |= VXLAN_F_UDP_ZERO_CSUM_TX;
 
 	if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX] &&
 	    nla_get_u8(data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX]))
@@ -3043,7 +3042,7 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
 	    nla_put_u32(skb, IFLA_VXLAN_LIMIT, vxlan->cfg.addrmax) ||
 	    nla_put_be16(skb, IFLA_VXLAN_PORT, vxlan->cfg.dst_port) ||
 	    nla_put_u8(skb, IFLA_VXLAN_UDP_CSUM,
-			!!(vxlan->flags & VXLAN_F_UDP_CSUM)) ||
+			!(vxlan->flags & VXLAN_F_UDP_ZERO_CSUM_TX)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_TX,
 			!!(vxlan->flags & VXLAN_F_UDP_ZERO_CSUM6_TX)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_RX,
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 748083d..6eda4ed 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -197,7 +197,7 @@ struct vxlan_dev {
 #define VXLAN_F_L2MISS			0x08
 #define VXLAN_F_L3MISS			0x10
 #define VXLAN_F_IPV6			0x20
-#define VXLAN_F_UDP_CSUM		0x40
+#define VXLAN_F_UDP_ZERO_CSUM_TX	0x40
 #define VXLAN_F_UDP_ZERO_CSUM6_TX	0x80
 #define VXLAN_F_UDP_ZERO_CSUM6_RX	0x100
 #define VXLAN_F_REMCSUM_TX		0x200
-- 
cgit v0.10.2


From 8e2fe1d9f1a20924f98ea46931a1d7fb092aa876 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 19 Feb 2016 23:05:22 +0100
Subject: bpf: add new arg_type that allows for 0 sized stack buffer

Currently, when we pass a buffer from the eBPF stack into a helper
function, the function proto indicates argument types as ARG_PTR_TO_STACK
and ARG_CONST_STACK_SIZE pair. If R<X> contains the former, then R<X+1>
must be of the latter type. Then, verifier checks whether the buffer
points into eBPF stack, is initialized, etc. The verifier also guarantees
that the constant value passed in R<X+1> is greater than 0, so helper
functions don't need to test for it and can always assume a non-NULL
initialized buffer as well as non-0 buffer size.

This patch adds a new argument types ARG_CONST_STACK_SIZE_OR_ZERO that
allows to also pass NULL as R<X> and 0 as R<X+1> into the helper function.
Such helper functions, of course, need to be able to handle these cases
internally then. Verifier guarantees that either R<X> == NULL && R<X+1> == 0
or R<X> != NULL && R<X+1> != 0 (like the case of ARG_CONST_STACK_SIZE), any
other combinations are not possible to load.

I went through various options of extending the verifier, and introducing
the type ARG_CONST_STACK_SIZE_OR_ZERO seems to have most minimal changes
needed to the verifier.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 0cadbb7..51e498e 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -65,6 +65,7 @@ enum bpf_arg_type {
 	 */
 	ARG_PTR_TO_STACK,	/* any pointer to eBPF program stack */
 	ARG_CONST_STACK_SIZE,	/* number of bytes accessed from stack */
+	ARG_CONST_STACK_SIZE_OR_ZERO, /* number of bytes accessed from stack or 0 */
 
 	ARG_PTR_TO_CTX,		/* pointer to context */
 	ARG_ANYTHING,		/* any (initialized) argument is ok */
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 42ba4cc..36dc497 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -779,15 +779,24 @@ static int check_xadd(struct verifier_env *env, struct bpf_insn *insn)
  * bytes from that pointer, make sure that it's within stack boundary
  * and all elements of stack are initialized
  */
-static int check_stack_boundary(struct verifier_env *env,
-				int regno, int access_size)
+static int check_stack_boundary(struct verifier_env *env, int regno,
+				int access_size, bool zero_size_allowed)
 {
 	struct verifier_state *state = &env->cur_state;
 	struct reg_state *regs = state->regs;
 	int off, i;
 
-	if (regs[regno].type != PTR_TO_STACK)
+	if (regs[regno].type != PTR_TO_STACK) {
+		if (zero_size_allowed && access_size == 0 &&
+		    regs[regno].type == CONST_IMM &&
+		    regs[regno].imm  == 0)
+			return 0;
+
+		verbose("R%d type=%s expected=%s\n", regno,
+			reg_type_str[regs[regno].type],
+			reg_type_str[PTR_TO_STACK]);
 		return -EACCES;
+	}
 
 	off = regs[regno].imm;
 	if (off >= 0 || off < -MAX_BPF_STACK || off + access_size > 0 ||
@@ -830,15 +839,24 @@ static int check_func_arg(struct verifier_env *env, u32 regno,
 		return 0;
 	}
 
-	if (arg_type == ARG_PTR_TO_STACK || arg_type == ARG_PTR_TO_MAP_KEY ||
+	if (arg_type == ARG_PTR_TO_MAP_KEY ||
 	    arg_type == ARG_PTR_TO_MAP_VALUE) {
 		expected_type = PTR_TO_STACK;
-	} else if (arg_type == ARG_CONST_STACK_SIZE) {
+	} else if (arg_type == ARG_CONST_STACK_SIZE ||
+		   arg_type == ARG_CONST_STACK_SIZE_OR_ZERO) {
 		expected_type = CONST_IMM;
 	} else if (arg_type == ARG_CONST_MAP_PTR) {
 		expected_type = CONST_PTR_TO_MAP;
 	} else if (arg_type == ARG_PTR_TO_CTX) {
 		expected_type = PTR_TO_CTX;
+	} else if (arg_type == ARG_PTR_TO_STACK) {
+		expected_type = PTR_TO_STACK;
+		/* One exception here. In case function allows for NULL to be
+		 * passed in as argument, it's a CONST_IMM type. Final test
+		 * happens during stack boundary checking.
+		 */
+		if (reg->type == CONST_IMM && reg->imm == 0)
+			expected_type = CONST_IMM;
 	} else {
 		verbose("unsupported arg_type %d\n", arg_type);
 		return -EFAULT;
@@ -868,8 +886,8 @@ static int check_func_arg(struct verifier_env *env, u32 regno,
 			verbose("invalid map_ptr to access map->key\n");
 			return -EACCES;
 		}
-		err = check_stack_boundary(env, regno, (*mapp)->key_size);
-
+		err = check_stack_boundary(env, regno, (*mapp)->key_size,
+					   false);
 	} else if (arg_type == ARG_PTR_TO_MAP_VALUE) {
 		/* bpf_map_xxx(..., map_ptr, ..., value) call:
 		 * check [value, value + map->value_size) validity
@@ -879,9 +897,12 @@ static int check_func_arg(struct verifier_env *env, u32 regno,
 			verbose("invalid map_ptr to access map->value\n");
 			return -EACCES;
 		}
-		err = check_stack_boundary(env, regno, (*mapp)->value_size);
+		err = check_stack_boundary(env, regno, (*mapp)->value_size,
+					   false);
+	} else if (arg_type == ARG_CONST_STACK_SIZE ||
+		   arg_type == ARG_CONST_STACK_SIZE_OR_ZERO) {
+		bool zero_size_allowed = (arg_type == ARG_CONST_STACK_SIZE_OR_ZERO);
 
-	} else if (arg_type == ARG_CONST_STACK_SIZE) {
 		/* bpf_xxx(..., buf, len) call will access 'len' bytes
 		 * from stack pointer 'buf'. Check it
 		 * note: regno == len, regno - 1 == buf
@@ -891,7 +912,8 @@ static int check_func_arg(struct verifier_env *env, u32 regno,
 			verbose("ARG_CONST_STACK_SIZE cannot be first argument\n");
 			return -EACCES;
 		}
-		err = check_stack_boundary(env, regno - 1, reg->imm);
+		err = check_stack_boundary(env, regno - 1, reg->imm,
+					   zero_size_allowed);
 	}
 
 	return err;
-- 
cgit v0.10.2


From 7d672345ed295b1356a5d9f7111da1d1d7d65867 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 19 Feb 2016 23:05:23 +0100
Subject: bpf: add generic bpf_csum_diff helper

For L4 checksums, we currently have bpf_l4_csum_replace() helper. It's
currently limited to handle 2 and 4 byte changes in a header and feeds the
from/to into inet_proto_csum_replace{2,4}() helpers of the kernel. When
working with IPv6, for example, this makes it rather cumbersome to deal
with, similarly when editing larger parts of a header.

Instead, extend the API in a more generic way: For bpf_l4_csum_replace(),
add a case for header field mask of 0 to change the checksum at a given
offset through inet_proto_csum_replace_by_diff(), and provide a helper
bpf_csum_diff() that can generically calculate a from/to diff for arbitrary
amounts of data.

This can be used in multiple ways: for the bpf_l4_csum_replace() only
part, this even provides us with the option to insert precalculated diffs
from user space f.e. from a map, or from bpf_csum_diff() during runtime.

bpf_csum_diff() has a optional from/to stack buffer input, so we can
calculate a diff by using a scratchbuffer for scenarios where we're
inserting (from is NULL), removing (to is NULL) or diffing (from/to buffers
don't need to be of equal size) data. Also, bpf_csum_diff() allows to
feed a previous csum into csum_partial(), so the function can also be
cascaded.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index d3e77da..48d0a6c 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -287,6 +287,17 @@ enum bpf_func_id {
 	 * Return: >= 0 stackid on success or negative error
 	 */
 	BPF_FUNC_get_stackid,
+
+	/**
+	 * bpf_csum_diff(from, from_size, to, to_size, seed) - calculate csum diff
+	 * @from: raw from buffer
+	 * @from_size: length of from buffer
+	 * @to: raw to buffer
+	 * @to_size: length of to buffer
+	 * @seed: optional seed
+	 * Return: csum result
+	 */
+	BPF_FUNC_csum_diff,
 	__BPF_FUNC_MAX_ID,
 };
 
diff --git a/net/core/filter.c b/net/core/filter.c
index 2a6e956..bf504f8 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1491,6 +1491,12 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 		return -EFAULT;
 
 	switch (flags & BPF_F_HDR_FIELD_MASK) {
+	case 0:
+		if (unlikely(from != 0))
+			return -EINVAL;
+
+		inet_proto_csum_replace_by_diff(ptr, skb, to, is_pseudo);
+		break;
 	case 2:
 		inet_proto_csum_replace2(ptr, skb, from, to, is_pseudo);
 		break;
@@ -1519,6 +1525,51 @@ const struct bpf_func_proto bpf_l4_csum_replace_proto = {
 	.arg5_type	= ARG_ANYTHING,
 };
 
+struct bpf_csum_scratchpad {
+	__be32 diff[128];
+};
+
+static DEFINE_PER_CPU(struct bpf_csum_scratchpad, bpf_csum_sp);
+
+static u64 bpf_csum_diff(u64 r1, u64 from_size, u64 r3, u64 to_size, u64 seed)
+{
+	struct bpf_csum_scratchpad *sp = this_cpu_ptr(&bpf_csum_sp);
+	u64 diff_size = from_size + to_size;
+	__be32 *from = (__be32 *) (long) r1;
+	__be32 *to   = (__be32 *) (long) r3;
+	int i, j = 0;
+
+	/* This is quite flexible, some examples:
+	 *
+	 * from_size == 0, to_size > 0,  seed := csum --> pushing data
+	 * from_size > 0,  to_size == 0, seed := csum --> pulling data
+	 * from_size > 0,  to_size > 0,  seed := 0    --> diffing data
+	 *
+	 * Even for diffing, from_size and to_size don't need to be equal.
+	 */
+	if (unlikely(((from_size | to_size) & (sizeof(__be32) - 1)) ||
+		     diff_size > sizeof(sp->diff)))
+		return -EINVAL;
+
+	for (i = 0; i < from_size / sizeof(__be32); i++, j++)
+		sp->diff[j] = ~from[i];
+	for (i = 0; i <   to_size / sizeof(__be32); i++, j++)
+		sp->diff[j] = to[i];
+
+	return csum_partial(sp->diff, diff_size, seed);
+}
+
+const struct bpf_func_proto bpf_csum_diff_proto = {
+	.func		= bpf_csum_diff,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_STACK,
+	.arg2_type	= ARG_CONST_STACK_SIZE_OR_ZERO,
+	.arg3_type	= ARG_PTR_TO_STACK,
+	.arg4_type	= ARG_CONST_STACK_SIZE_OR_ZERO,
+	.arg5_type	= ARG_ANYTHING,
+};
+
 static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5)
 {
 	struct sk_buff *skb = (struct sk_buff *) (long) r1, *skb2;
@@ -1849,6 +1900,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id)
 		return &bpf_skb_store_bytes_proto;
 	case BPF_FUNC_skb_load_bytes:
 		return &bpf_skb_load_bytes_proto;
+	case BPF_FUNC_csum_diff:
+		return &bpf_csum_diff_proto;
 	case BPF_FUNC_l3_csum_replace:
 		return &bpf_l3_csum_replace_proto;
 	case BPF_FUNC_l4_csum_replace:
-- 
cgit v0.10.2


From 21cafc1dc2da999dabc5ed7aa94230454471fcf0 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 19 Feb 2016 23:05:24 +0100
Subject: bpf: remove artificial bpf_skb_{load, store}_bytes buffer limitation

We currently limit bpf_skb_store_bytes() and bpf_skb_load_bytes()
helpers to only store or load a maximum buffer of 16 bytes. Thus,
loading, rewriting and storing headers require several bpf_skb_load_bytes()
and bpf_skb_store_bytes() calls.

Also here we can use a per-cpu scratch buffer instead in order to not
pressure stack space any further. I do suspect that this limit was mainly
set in place for this particular reason. So, ease program development
by removing this limitation and make the scratchpad generic, so it can
be reused.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/filter.c b/net/core/filter.c
index bf504f8..ea391e6 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1333,15 +1333,22 @@ int sk_reuseport_attach_bpf(u32 ufd, struct sock *sk)
 	return 0;
 }
 
-#define BPF_LDST_LEN 16U
+struct bpf_scratchpad {
+	union {
+		__be32 diff[MAX_BPF_STACK / sizeof(__be32)];
+		u8     buff[MAX_BPF_STACK];
+	};
+};
+
+static DEFINE_PER_CPU(struct bpf_scratchpad, bpf_sp);
 
 static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 {
+	struct bpf_scratchpad *sp = this_cpu_ptr(&bpf_sp);
 	struct sk_buff *skb = (struct sk_buff *) (long) r1;
 	int offset = (int) r2;
 	void *from = (void *) (long) r3;
 	unsigned int len = (unsigned int) r4;
-	char buf[BPF_LDST_LEN];
 	void *ptr;
 
 	if (unlikely(flags & ~(BPF_F_RECOMPUTE_CSUM)))
@@ -1355,14 +1362,14 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 	 *
 	 * so check for invalid 'offset' and too large 'len'
 	 */
-	if (unlikely((u32) offset > 0xffff || len > sizeof(buf)))
+	if (unlikely((u32) offset > 0xffff || len > sizeof(sp->buff)))
 		return -EFAULT;
 
 	if (unlikely(skb_cloned(skb) &&
 		     !skb_clone_writable(skb, offset + len)))
 		return -EFAULT;
 
-	ptr = skb_header_pointer(skb, offset, len, buf);
+	ptr = skb_header_pointer(skb, offset, len, sp->buff);
 	if (unlikely(!ptr))
 		return -EFAULT;
 
@@ -1371,7 +1378,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 
 	memcpy(ptr, from, len);
 
-	if (ptr == buf)
+	if (ptr == sp->buff)
 		/* skb_store_bits cannot return -EFAULT here */
 		skb_store_bits(skb, offset, ptr, len);
 
@@ -1400,7 +1407,7 @@ static u64 bpf_skb_load_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
 	unsigned int len = (unsigned int) r4;
 	void *ptr;
 
-	if (unlikely((u32) offset > 0xffff || len > BPF_LDST_LEN))
+	if (unlikely((u32) offset > 0xffff || len > MAX_BPF_STACK))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, len, to);
@@ -1525,15 +1532,9 @@ const struct bpf_func_proto bpf_l4_csum_replace_proto = {
 	.arg5_type	= ARG_ANYTHING,
 };
 
-struct bpf_csum_scratchpad {
-	__be32 diff[128];
-};
-
-static DEFINE_PER_CPU(struct bpf_csum_scratchpad, bpf_csum_sp);
-
 static u64 bpf_csum_diff(u64 r1, u64 from_size, u64 r3, u64 to_size, u64 seed)
 {
-	struct bpf_csum_scratchpad *sp = this_cpu_ptr(&bpf_csum_sp);
+	struct bpf_scratchpad *sp = this_cpu_ptr(&bpf_sp);
 	u64 diff_size = from_size + to_size;
 	__be32 *from = (__be32 *) (long) r1;
 	__be32 *to   = (__be32 *) (long) r3;
-- 
cgit v0.10.2


From 3697649ff29e0f647565eed04b27a7779c646a22 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 19 Feb 2016 23:05:25 +0100
Subject: bpf: try harder on clones when writing into skb

When we're dealing with clones and the area is not writeable, try
harder and get a copy via pskb_expand_head(). Replace also other
occurences in tc actions with the new skb_try_make_writable().

Reported-by: Ashhad Sheikh <ashhadsheikh394@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 89b5367..6a57757 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2630,6 +2630,13 @@ static inline int skb_clone_writable(const struct sk_buff *skb, unsigned int len
 	       skb_headroom(skb) + len <= skb->hdr_len;
 }
 
+static inline int skb_try_make_writable(struct sk_buff *skb,
+					unsigned int write_len)
+{
+	return skb_cloned(skb) && !skb_clone_writable(skb, write_len) &&
+	       pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
+}
+
 static inline int __skb_cow(struct sk_buff *skb, unsigned int headroom,
 			    int cloned)
 {
diff --git a/net/core/filter.c b/net/core/filter.c
index ea391e6..f031b82 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1364,9 +1364,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 	 */
 	if (unlikely((u32) offset > 0xffff || len > sizeof(sp->buff)))
 		return -EFAULT;
-
-	if (unlikely(skb_cloned(skb) &&
-		     !skb_clone_writable(skb, offset + len)))
+	if (unlikely(skb_try_make_writable(skb, offset + len)))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, len, sp->buff);
@@ -1439,9 +1437,7 @@ static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 		return -EINVAL;
 	if (unlikely((u32) offset > 0xffff))
 		return -EFAULT;
-
-	if (unlikely(skb_cloned(skb) &&
-		     !skb_clone_writable(skb, offset + sizeof(sum))))
+	if (unlikely(skb_try_make_writable(skb, offset + sizeof(sum))))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum);
@@ -1488,9 +1484,7 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 		return -EINVAL;
 	if (unlikely((u32) offset > 0xffff))
 		return -EFAULT;
-
-	if (unlikely(skb_cloned(skb) &&
-		     !skb_clone_writable(skb, offset + sizeof(sum))))
+	if (unlikely(skb_try_make_writable(skb, offset + sizeof(sum))))
 		return -EFAULT;
 
 	ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum);
@@ -1734,6 +1728,13 @@ bool bpf_helper_changes_skb_data(void *func)
 		return true;
 	if (func == bpf_skb_vlan_pop)
 		return true;
+	if (func == bpf_skb_store_bytes)
+		return true;
+	if (func == bpf_l3_csum_replace)
+		return true;
+	if (func == bpf_l4_csum_replace)
+		return true;
+
 	return false;
 }
 
diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c
index b07c535..eeb3eb3 100644
--- a/net/sched/act_csum.c
+++ b/net/sched/act_csum.c
@@ -105,9 +105,7 @@ static void *tcf_csum_skb_nextlayer(struct sk_buff *skb,
 	int hl = ihl + jhl;
 
 	if (!pskb_may_pull(skb, ipl + ntkoff) || (ipl < hl) ||
-	    (skb_cloned(skb) &&
-	     !skb_clone_writable(skb, hl + ntkoff) &&
-	     pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
+	    skb_try_make_writable(skb, hl + ntkoff))
 		return NULL;
 	else
 		return (void *)(skb_network_header(skb) + ihl);
@@ -365,9 +363,7 @@ static int tcf_csum_ipv4(struct sk_buff *skb, u32 update_flags)
 	}
 
 	if (update_flags & TCA_CSUM_UPDATE_FLAG_IPV4HDR) {
-		if (skb_cloned(skb) &&
-		    !skb_clone_writable(skb, sizeof(*iph) + ntkoff) &&
-		    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
+		if (skb_try_make_writable(skb, sizeof(*iph) + ntkoff))
 			goto fail;
 
 		ip_send_check(ip_hdr(skb));
diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c
index b7c4ead..27607b8 100644
--- a/net/sched/act_nat.c
+++ b/net/sched/act_nat.c
@@ -126,9 +126,7 @@ static int tcf_nat(struct sk_buff *skb, const struct tc_action *a,
 		addr = iph->daddr;
 
 	if (!((old_addr ^ addr) & mask)) {
-		if (skb_cloned(skb) &&
-		    !skb_clone_writable(skb, sizeof(*iph) + noff) &&
-		    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
+		if (skb_try_make_writable(skb, sizeof(*iph) + noff))
 			goto drop;
 
 		new_addr &= mask;
@@ -156,9 +154,7 @@ static int tcf_nat(struct sk_buff *skb, const struct tc_action *a,
 		struct tcphdr *tcph;
 
 		if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) ||
-		    (skb_cloned(skb) &&
-		     !skb_clone_writable(skb, ihl + sizeof(*tcph) + noff) &&
-		     pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
+		    skb_try_make_writable(skb, ihl + sizeof(*tcph) + noff))
 			goto drop;
 
 		tcph = (void *)(skb_network_header(skb) + ihl);
@@ -171,9 +167,7 @@ static int tcf_nat(struct sk_buff *skb, const struct tc_action *a,
 		struct udphdr *udph;
 
 		if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) ||
-		    (skb_cloned(skb) &&
-		     !skb_clone_writable(skb, ihl + sizeof(*udph) + noff) &&
-		     pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
+		    skb_try_make_writable(skb, ihl + sizeof(*udph) + noff))
 			goto drop;
 
 		udph = (void *)(skb_network_header(skb) + ihl);
@@ -213,10 +207,8 @@ static int tcf_nat(struct sk_buff *skb, const struct tc_action *a,
 		if ((old_addr ^ addr) & mask)
 			break;
 
-		if (skb_cloned(skb) &&
-		    !skb_clone_writable(skb, ihl + sizeof(*icmph) +
-					     sizeof(*iph) + noff) &&
-		    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
+		if (skb_try_make_writable(skb, ihl + sizeof(*icmph) +
+					  sizeof(*iph) + noff))
 			goto drop;
 
 		icmph = (void *)(skb_network_header(skb) + ihl);
-- 
cgit v0.10.2


From 2f72959a9c1260ade234f353ccca91118151af66 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 19 Feb 2016 23:05:26 +0100
Subject: bpf: fix csum update in bpf_l4_csum_replace helper for udp

When using this helper for updating UDP checksums, we need to extend
this in order to write CSUM_MANGLED_0 for csum computations that result
into 0 as sum. Reason we need this is because packets with a checksum
could otherwise become incorrectly marked as a packet without a checksum.
Likewise, if the user indicates BPF_F_MARK_MANGLED_0, then we should
not turn packets without a checksum into ones with a checksum.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 48d0a6c..6496f98 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -313,6 +313,7 @@ enum bpf_func_id {
 
 /* BPF_FUNC_l4_csum_replace flags. */
 #define BPF_F_PSEUDO_HDR		(1ULL << 4)
+#define BPF_F_MARK_MANGLED_0		(1ULL << 5)
 
 /* BPF_FUNC_clone_redirect and BPF_FUNC_redirect flags. */
 #define BPF_F_INGRESS			(1ULL << 0)
diff --git a/net/core/filter.c b/net/core/filter.c
index f031b82..8a0b8c3 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1477,10 +1477,12 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 {
 	struct sk_buff *skb = (struct sk_buff *) (long) r1;
 	bool is_pseudo = flags & BPF_F_PSEUDO_HDR;
+	bool is_mmzero = flags & BPF_F_MARK_MANGLED_0;
 	int offset = (int) r2;
 	__sum16 sum, *ptr;
 
-	if (unlikely(flags & ~(BPF_F_PSEUDO_HDR | BPF_F_HDR_FIELD_MASK)))
+	if (unlikely(flags & ~(BPF_F_MARK_MANGLED_0 | BPF_F_PSEUDO_HDR |
+			       BPF_F_HDR_FIELD_MASK)))
 		return -EINVAL;
 	if (unlikely((u32) offset > 0xffff))
 		return -EFAULT;
@@ -1490,6 +1492,8 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 	ptr = skb_header_pointer(skb, offset, sizeof(sum), &sum);
 	if (unlikely(!ptr))
 		return -EFAULT;
+	if (is_mmzero && !*ptr)
+		return 0;
 
 	switch (flags & BPF_F_HDR_FIELD_MASK) {
 	case 0:
@@ -1508,6 +1512,8 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 		return -EINVAL;
 	}
 
+	if (is_mmzero && !*ptr)
+		*ptr = CSUM_MANGLED_0;
 	if (ptr == &sum)
 		/* skb_store_bits guaranteed to not return -EFAULT here */
 		skb_store_bits(skb, offset, ptr, sizeof(sum));
-- 
cgit v0.10.2


From 6205b9cf200d1c3dda5491666ddc33e7b70fe469 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 19 Feb 2016 23:05:27 +0100
Subject: bpf: don't emit mov A,A on return

While debugging with bpf_jit_disasm I noticed emissions of 'mov %eax,%eax',
and found that this comes from BPF_RET | BPF_A translations from classic
BPF. Emitting this is unnecessary as BPF_REG_A is mapped into BPF_REG_0
already, therefore only emit a mov when immediates are used as return value.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/filter.c b/net/core/filter.c
index 8a0b8c3..a3aba15 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -530,12 +530,14 @@ do_pass:
 			*insn = BPF_MOV64_REG(BPF_REG_A, BPF_REG_TMP);
 			break;
 
-		/* RET_K, RET_A are remaped into 2 insns. */
+		/* RET_K is remaped into 2 insns. RET_A case doesn't need an
+		 * extra mov as BPF_REG_0 is already mapped into BPF_REG_A.
+		 */
 		case BPF_RET | BPF_A:
 		case BPF_RET | BPF_K:
-			*insn++ = BPF_MOV32_RAW(BPF_RVAL(fp->code) == BPF_K ?
-						BPF_K : BPF_X, BPF_REG_0,
-						BPF_REG_A, fp->k);
+			if (BPF_RVAL(fp->code) == BPF_K)
+				*insn++ = BPF_MOV32_RAW(BPF_K, BPF_REG_0,
+							0, fp->k);
 			*insn = BPF_EXIT_INSN();
 			break;
 
-- 
cgit v0.10.2


From c3aaa06d5a63609641b7ad62ee0956f3de86c1cd Mon Sep 17 00:00:00 2001
From: Mahesh Bandewar <maheshb@google.com>
Date: Sat, 20 Feb 2016 19:31:32 -0800
Subject: ipvlan: scrub skb before routing in L3 mode.

Scrub skb before hitting the iptable hooks to ensure packets hit
these hooks. Set the xnet param only when the packet is crossing the
ns boundry so if the IPvlan slave and master belong to the same ns,
the param will be set to false.

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
CC: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c
index 8c48bb2..4e60c6b 100644
--- a/drivers/net/ipvlan/ipvlan_core.c
+++ b/drivers/net/ipvlan/ipvlan_core.c
@@ -342,7 +342,7 @@ static struct ipvl_addr *ipvlan_addr_lookup(struct ipvl_port *port,
 	return addr;
 }
 
-static int ipvlan_process_v4_outbound(struct sk_buff *skb)
+static int ipvlan_process_v4_outbound(struct sk_buff *skb, bool xnet)
 {
 	const struct iphdr *ip4h = ip_hdr(skb);
 	struct net_device *dev = skb->dev;
@@ -365,7 +365,7 @@ static int ipvlan_process_v4_outbound(struct sk_buff *skb)
 		ip_rt_put(rt);
 		goto err;
 	}
-	skb_dst_drop(skb);
+	skb_scrub_packet(skb, xnet);
 	skb_dst_set(skb, &rt->dst);
 	err = ip_local_out(net, skb->sk, skb);
 	if (unlikely(net_xmit_eval(err)))
@@ -380,7 +380,7 @@ out:
 	return ret;
 }
 
-static int ipvlan_process_v6_outbound(struct sk_buff *skb)
+static int ipvlan_process_v6_outbound(struct sk_buff *skb, bool xnet)
 {
 	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	struct net_device *dev = skb->dev;
@@ -403,7 +403,7 @@ static int ipvlan_process_v6_outbound(struct sk_buff *skb)
 		dst_release(dst);
 		goto err;
 	}
-	skb_dst_drop(skb);
+	skb_scrub_packet(skb, xnet);
 	skb_dst_set(skb, dst);
 	err = ip6_local_out(net, skb->sk, skb);
 	if (unlikely(net_xmit_eval(err)))
@@ -418,8 +418,7 @@ out:
 	return ret;
 }
 
-static int ipvlan_process_outbound(struct sk_buff *skb,
-				   const struct ipvl_dev *ipvlan)
+static int ipvlan_process_outbound(struct sk_buff *skb, bool xnet)
 {
 	struct ethhdr *ethh = eth_hdr(skb);
 	int ret = NET_XMIT_DROP;
@@ -443,9 +442,9 @@ static int ipvlan_process_outbound(struct sk_buff *skb,
 	}
 
 	if (skb->protocol == htons(ETH_P_IPV6))
-		ret = ipvlan_process_v6_outbound(skb);
+		ret = ipvlan_process_v6_outbound(skb, xnet);
 	else if (skb->protocol == htons(ETH_P_IP))
-		ret = ipvlan_process_v4_outbound(skb);
+		ret = ipvlan_process_v4_outbound(skb, xnet);
 	else {
 		pr_warn_ratelimited("Dropped outbound packet type=%x\n",
 				    ntohs(skb->protocol));
@@ -481,6 +480,7 @@ static int ipvlan_xmit_mode_l3(struct sk_buff *skb, struct net_device *dev)
 	void *lyr3h;
 	struct ipvl_addr *addr;
 	int addr_type;
+	bool xnet;
 
 	lyr3h = ipvlan_get_L3_hdr(skb, &addr_type);
 	if (!lyr3h)
@@ -491,8 +491,9 @@ static int ipvlan_xmit_mode_l3(struct sk_buff *skb, struct net_device *dev)
 		return ipvlan_rcv_frame(addr, &skb, true);
 
 out:
+	xnet = !net_eq(dev_net(skb->dev), dev_net(ipvlan->phy_dev));
 	skb->dev = ipvlan->phy_dev;
-	return ipvlan_process_outbound(skb, ipvlan);
+	return ipvlan_process_outbound(skb, xnet);
 }
 
 static int ipvlan_xmit_mode_l2(struct sk_buff *skb, struct net_device *dev)
-- 
cgit v0.10.2


From e93fbc5a15ff25d4f9fd92a13c33cd37d99a2340 Mon Sep 17 00:00:00 2001
From: Mahesh Bandewar <maheshb@google.com>
Date: Sat, 20 Feb 2016 19:31:36 -0800
Subject: ipvlan: mode is u16

The mode argument was erronusly defined as u32 but it has always
been u16. Also use ipvlan_set_mode() helper to set the mode instead
of assigning directly. This should avoid future erronus assignments /
updates.

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h
index 9542b7b..817cab1 100644
--- a/drivers/net/ipvlan/ipvlan.h
+++ b/drivers/net/ipvlan/ipvlan.h
@@ -115,7 +115,6 @@ static inline struct ipvl_port *ipvlan_port_get_rtnl(const struct net_device *d)
 }
 
 void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev);
-void ipvlan_set_port_mode(struct ipvl_port *port, u32 nval);
 void ipvlan_init_secret(void);
 unsigned int ipvlan_mac_hash(const unsigned char *addr);
 rx_handler_result_t ipvlan_handle_frame(struct sk_buff **pskb);
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 7a3b414..5bcb852 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -14,7 +14,7 @@ void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev)
 	ipvlan->dev->mtu = dev->mtu - ipvlan->mtu_adj;
 }
 
-void ipvlan_set_port_mode(struct ipvl_port *port, u32 nval)
+static void ipvlan_set_port_mode(struct ipvl_port *port, u16 nval)
 {
 	struct ipvl_dev *ipvlan;
 
@@ -442,6 +442,7 @@ static int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 	struct ipvl_port *port;
 	struct net_device *phy_dev;
 	int err;
+	u16 mode = IPVLAN_MODE_L3;
 
 	if (!tb[IFLA_LINK])
 		return -EINVAL;
@@ -460,10 +461,10 @@ static int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 			return err;
 	}
 
-	port = ipvlan_port_get_rtnl(phy_dev);
 	if (data && data[IFLA_IPVLAN_MODE])
-		port->mode = nla_get_u16(data[IFLA_IPVLAN_MODE]);
+		mode = nla_get_u16(data[IFLA_IPVLAN_MODE]);
 
+	port = ipvlan_port_get_rtnl(phy_dev);
 	ipvlan->phy_dev = phy_dev;
 	ipvlan->dev = dev;
 	ipvlan->port = port;
@@ -489,6 +490,8 @@ static int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 		goto ipvlan_destroy_port;
 
 	list_add_tail_rcu(&ipvlan->pnode, &port->ipvlans);
+	ipvlan_set_port_mode(port, mode);
+
 	netif_stacked_transfer_operstate(phy_dev, dev);
 	return 0;
 
-- 
cgit v0.10.2


From ab5b7013db3cc637a8f19e00d71310e40db75bf6 Mon Sep 17 00:00:00 2001
From: Mahesh Bandewar <maheshb@google.com>
Date: Sat, 20 Feb 2016 19:31:41 -0800
Subject: ipvlan: misc changes

1. scope correction for few functions that are used in single file.
2. Adjust variables that are used in fast-path to fit into single cacheline
3. Update rcv_frame() to skip shared check for frames coming over wire

Signed-off-by: Mahesh Bandewar <maheshb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h
index 817cab1..695a5dc 100644
--- a/drivers/net/ipvlan/ipvlan.h
+++ b/drivers/net/ipvlan/ipvlan.h
@@ -84,19 +84,19 @@ struct ipvl_addr {
 #define ip4addr ipu.ip4
 	struct hlist_node	hlnode;  /* Hash-table linkage */
 	struct list_head	anode;   /* logical-interface linkage */
-	struct rcu_head		rcu;
 	ipvl_hdr_type		atype;
+	struct rcu_head		rcu;
 };
 
 struct ipvl_port {
 	struct net_device	*dev;
 	struct hlist_head	hlhead[IPVLAN_HASH_SIZE];
 	struct list_head	ipvlans;
-	struct rcu_head		rcu;
+	u16			mode;
 	struct work_struct	wq;
 	struct sk_buff_head	backlog;
 	int			count;
-	u16			mode;
+	struct rcu_head		rcu;
 };
 
 static inline struct ipvl_port *ipvlan_port_get_rcu(const struct net_device *d)
@@ -114,7 +114,6 @@ static inline struct ipvl_port *ipvlan_port_get_rtnl(const struct net_device *d)
 	return rtnl_dereference(d->rx_handler_data);
 }
 
-void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev);
 void ipvlan_init_secret(void);
 unsigned int ipvlan_mac_hash(const unsigned char *addr);
 rx_handler_result_t ipvlan_handle_frame(struct sk_buff **pskb);
@@ -124,7 +123,5 @@ void ipvlan_ht_addr_add(struct ipvl_dev *ipvlan, struct ipvl_addr *addr);
 struct ipvl_addr *ipvlan_find_addr(const struct ipvl_dev *ipvlan,
 				   const void *iaddr, bool is_v6);
 bool ipvlan_addr_busy(struct ipvl_port *port, void *iaddr, bool is_v6);
-struct ipvl_addr *ipvlan_ht_addr_lookup(const struct ipvl_port *port,
-					const void *iaddr, bool is_v6);
 void ipvlan_ht_addr_del(struct ipvl_addr *addr);
 #endif /* __IPVLAN_H */
diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c
index 4e60c6b..d6d0524 100644
--- a/drivers/net/ipvlan/ipvlan_core.c
+++ b/drivers/net/ipvlan/ipvlan_core.c
@@ -53,8 +53,8 @@ static u8 ipvlan_get_v4_hash(const void *iaddr)
 	       IPVLAN_HASH_MASK;
 }
 
-struct ipvl_addr *ipvlan_ht_addr_lookup(const struct ipvl_port *port,
-					const void *iaddr, bool is_v6)
+static struct ipvl_addr *ipvlan_ht_addr_lookup(const struct ipvl_port *port,
+					       const void *iaddr, bool is_v6)
 {
 	struct ipvl_addr *addr;
 	u8 hash;
@@ -265,20 +265,25 @@ static int ipvlan_rcv_frame(struct ipvl_addr *addr, struct sk_buff **pskb,
 	struct sk_buff *skb = *pskb;
 
 	len = skb->len + ETH_HLEN;
-	if (unlikely(!(dev->flags & IFF_UP))) {
-		kfree_skb(skb);
-		goto out;
-	}
+	/* Only packets exchanged between two local slaves need to have
+	 * device-up check as well as skb-share check.
+	 */
+	if (local) {
+		if (unlikely(!(dev->flags & IFF_UP))) {
+			kfree_skb(skb);
+			goto out;
+		}
 
-	skb = skb_share_check(skb, GFP_ATOMIC);
-	if (!skb)
-		goto out;
+		skb = skb_share_check(skb, GFP_ATOMIC);
+		if (!skb)
+			goto out;
 
-	*pskb = skb;
+		*pskb = skb;
+	}
 	skb->dev = dev;
-	skb->pkt_type = PACKET_HOST;
 
 	if (local) {
+		skb->pkt_type = PACKET_HOST;
 		if (dev_forward_skb(ipvlan->dev, skb) == NET_RX_SUCCESS)
 			success = true;
 	} else {
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 5bcb852..a7ca1c5 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -9,7 +9,7 @@
 
 #include "ipvlan.h"
 
-void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev)
+static void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev)
 {
 	ipvlan->dev->mtu = dev->mtu - ipvlan->mtu_adj;
 }
-- 
cgit v0.10.2


From 60fffb3b80e8669636ab4bc067869b7516f47147 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 21 Feb 2016 11:40:07 +0200
Subject: qed: Turn most GFP_ATOMIC into GFP_KERNEL

Initial driver submission used GFP_ATOMIC almost inclusively when
allocating memory. We now remedy this point, using GFP_KERNEL where
it's possible.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
index d3f7a02..fc767c0 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c
@@ -448,7 +448,7 @@ int qed_cxt_mngr_alloc(struct qed_hwfn *p_hwfn)
 	struct qed_cxt_mngr *p_mngr;
 	u32 i;
 
-	p_mngr = kzalloc(sizeof(*p_mngr), GFP_ATOMIC);
+	p_mngr = kzalloc(sizeof(*p_mngr), GFP_KERNEL);
 	if (!p_mngr) {
 		DP_NOTICE(p_hwfn, "Failed to allocate `struct qed_cxt_mngr'\n");
 		return -ENOMEM;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index bc17ed2..acfe7be 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -134,17 +134,17 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn)
 	/* PQs will be arranged as follows: First per-TC PQ then pure-LB quete.
 	 */
 	qm_info->qm_pq_params = kzalloc(sizeof(*qm_info->qm_pq_params) *
-					num_pqs, GFP_ATOMIC);
+					num_pqs, GFP_KERNEL);
 	if (!qm_info->qm_pq_params)
 		goto alloc_err;
 
 	qm_info->qm_vport_params = kzalloc(sizeof(*qm_info->qm_vport_params) *
-					   num_vports, GFP_ATOMIC);
+					   num_vports, GFP_KERNEL);
 	if (!qm_info->qm_vport_params)
 		goto alloc_err;
 
 	qm_info->qm_port_params = kzalloc(sizeof(*qm_info->qm_port_params) *
-					  MAX_NUM_PORTS, GFP_ATOMIC);
+					  MAX_NUM_PORTS, GFP_KERNEL);
 	if (!qm_info->qm_port_params)
 		goto alloc_err;
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c
index ffa9927..a95a3e4 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hw.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c
@@ -44,7 +44,7 @@ struct qed_ptt_pool {
 int qed_ptt_pool_alloc(struct qed_hwfn *p_hwfn)
 {
 	struct qed_ptt_pool *p_pool = kmalloc(sizeof(*p_pool),
-					      GFP_ATOMIC);
+					      GFP_KERNEL);
 	int i;
 
 	if (!p_pool)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index 9cc9d62..90ba1d7 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -399,7 +399,7 @@ static int qed_int_sb_attn_alloc(struct qed_hwfn *p_hwfn,
 	dma_addr_t p_phys = 0;
 
 	/* SB struct */
-	p_sb = kmalloc(sizeof(*p_sb), GFP_ATOMIC);
+	p_sb = kmalloc(sizeof(*p_sb), GFP_KERNEL);
 	if (!p_sb) {
 		DP_NOTICE(cdev, "Failed to allocate `struct qed_sb_attn_info'\n");
 		return -ENOMEM;
@@ -666,7 +666,7 @@ static int qed_int_sp_sb_alloc(struct qed_hwfn *p_hwfn,
 	void *p_virt;
 
 	/* SB struct */
-	p_sb = kmalloc(sizeof(*p_sb), GFP_ATOMIC);
+	p_sb = kmalloc(sizeof(*p_sb), GFP_KERNEL);
 	if (!p_sb) {
 		DP_NOTICE(p_hwfn, "Failed to allocate `struct qed_sb_info'\n");
 		return -ENOMEM;
@@ -946,7 +946,7 @@ int qed_int_igu_read_cam(struct qed_hwfn *p_hwfn,
 	u16 sb_id;
 	u16 prev_sb_id = 0xFF;
 
-	p_hwfn->hw_info.p_igu_info = kzalloc(sizeof(*p_igu_info), GFP_ATOMIC);
+	p_hwfn->hw_info.p_igu_info = kzalloc(sizeof(*p_igu_info), GFP_KERNEL);
 
 	if (!p_hwfn->hw_info.p_igu_info)
 		return -ENOMEM;
@@ -1072,7 +1072,7 @@ static void qed_int_sp_dpc_setup(struct qed_hwfn *p_hwfn)
 
 static int qed_int_sp_dpc_alloc(struct qed_hwfn *p_hwfn)
 {
-	p_hwfn->sp_dpc = kmalloc(sizeof(*p_hwfn->sp_dpc), GFP_ATOMIC);
+	p_hwfn->sp_dpc = kmalloc(sizeof(*p_hwfn->sp_dpc), GFP_KERNEL);
 	if (!p_hwfn->sp_dpc)
 		return -ENOMEM;
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 1457e30..f23ce73 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -147,7 +147,7 @@ int qed_mcp_cmd_init(struct qed_hwfn *p_hwfn,
 	u32 size;
 
 	/* Allocate mcp_info structure */
-	p_hwfn->mcp_info = kzalloc(sizeof(*p_hwfn->mcp_info), GFP_ATOMIC);
+	p_hwfn->mcp_info = kzalloc(sizeof(*p_hwfn->mcp_info), GFP_KERNEL);
 	if (!p_hwfn->mcp_info)
 		goto err;
 	p_info = p_hwfn->mcp_info;
@@ -161,10 +161,10 @@ int qed_mcp_cmd_init(struct qed_hwfn *p_hwfn,
 	}
 
 	size = MFW_DRV_MSG_MAX_DWORDS(p_info->mfw_mb_length) * sizeof(u32);
-	p_info->mfw_mb_cur = kzalloc(size, GFP_ATOMIC);
+	p_info->mfw_mb_cur = kzalloc(size, GFP_KERNEL);
 	p_info->mfw_mb_shadow =
 		kzalloc(sizeof(u32) * MFW_DRV_MSG_MAX_DWORDS(
-				p_info->mfw_mb_length), GFP_ATOMIC);
+				p_info->mfw_mb_length), GFP_KERNEL);
 	if (!p_info->mfw_mb_shadow || !p_info->mfw_mb_addr)
 		goto err;
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c
index 3dd548a..f6c6c21 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_spq.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c
@@ -327,7 +327,7 @@ struct qed_eq *qed_eq_alloc(struct qed_hwfn *p_hwfn,
 	struct qed_eq *p_eq;
 
 	/* Allocate EQ struct */
-	p_eq = kzalloc(sizeof(*p_eq), GFP_ATOMIC);
+	p_eq = kzalloc(sizeof(*p_eq), GFP_KERNEL);
 	if (!p_eq) {
 		DP_NOTICE(p_hwfn, "Failed to allocate `struct qed_eq'\n");
 		return NULL;
@@ -457,7 +457,7 @@ int qed_spq_alloc(struct qed_hwfn *p_hwfn)
 
 	/* SPQ struct */
 	p_spq =
-		kzalloc(sizeof(struct qed_spq), GFP_ATOMIC);
+		kzalloc(sizeof(struct qed_spq), GFP_KERNEL);
 	if (!p_spq) {
 		DP_NOTICE(p_hwfn, "Failed to allocate `struct qed_spq'\n");
 		return -ENOMEM;
@@ -853,7 +853,7 @@ struct qed_consq *qed_consq_alloc(struct qed_hwfn *p_hwfn)
 	struct qed_consq *p_consq;
 
 	/* Allocate ConsQ struct */
-	p_consq = kzalloc(sizeof(*p_consq), GFP_ATOMIC);
+	p_consq = kzalloc(sizeof(*p_consq), GFP_KERNEL);
 	if (!p_consq) {
 		DP_NOTICE(p_hwfn, "Failed to allocate `struct qed_consq'\n");
 		return NULL;
-- 
cgit v0.10.2


From 0a0c5d3b81529de2c1b65eb6582a4b6d0da1ada9 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 21 Feb 2016 11:40:08 +0200
Subject: qed: Handle possible race in SB config

Due to HW design, some of the memories are wide-bus and access to those
needs to be sequentialized on a per-HW-block level; Read/write to a
given HW-block might break other read/write to wide-bus memory done at
~same time.

Status blocks initialization in CAU is done into such a wide-bus memory.
This moves the initialization into using DMAE which is guaranteed to be
safe to use on such memories.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index 90ba1d7..fa73daa 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -473,20 +473,20 @@ void qed_int_cau_conf_sb(struct qed_hwfn *p_hwfn,
 			 u8 vf_valid)
 {
 	struct cau_sb_entry sb_entry;
-	u32 val;
 
 	qed_init_cau_sb_entry(p_hwfn, &sb_entry, p_hwfn->rel_pf_id,
 			      vf_number, vf_valid);
 
 	if (p_hwfn->hw_init_done) {
-		val = CAU_REG_SB_ADDR_MEMORY + igu_sb_id * sizeof(u64);
-		qed_wr(p_hwfn, p_ptt, val, lower_32_bits(sb_phys));
-		qed_wr(p_hwfn, p_ptt, val + sizeof(u32),
-		       upper_32_bits(sb_phys));
-
-		val = CAU_REG_SB_VAR_MEMORY + igu_sb_id * sizeof(u64);
-		qed_wr(p_hwfn, p_ptt, val, sb_entry.data);
-		qed_wr(p_hwfn, p_ptt, val + sizeof(u32), sb_entry.params);
+		/* Wide-bus, initialize via DMAE */
+		u64 phys_addr = (u64)sb_phys;
+
+		qed_dmae_host2grc(p_hwfn, p_ptt, (u64)(uintptr_t)&phys_addr,
+				  CAU_REG_SB_ADDR_MEMORY +
+				  igu_sb_id * sizeof(u64), 2, 0);
+		qed_dmae_host2grc(p_hwfn, p_ptt, (u64)(uintptr_t)&sb_entry,
+				  CAU_REG_SB_VAR_MEMORY +
+				  igu_sb_id * sizeof(u64), 2, 0);
 	} else {
 		/* Initialize Status Block Address */
 		STORE_RT_REG_AGG(p_hwfn,
-- 
cgit v0.10.2


From 06f56b8136b75a04ee6e142a9e345cfd4b357de5 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 21 Feb 2016 11:40:09 +0200
Subject: qed: Change metadata needed for SPQ entries

Each configuration element send via ramrod requires a Slow Path Queue
entry. This slightly changes the way such an entry is configured, but
contains mostly semantic changes [where more parameters are gathered
in a sub-struct instead of being directly passed].

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index 73feaf7..8d1bc7e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -142,9 +142,9 @@ static int qed_sp_vport_start(struct qed_hwfn *p_hwfn,
 			      u8 drop_ttl0_flg,
 			      u8 inner_vlan_removal_en_flg)
 {
-	struct qed_sp_init_request_params params;
 	struct vport_start_ramrod_data *p_ramrod = NULL;
 	struct qed_spq_entry *p_ent =  NULL;
+	struct qed_sp_init_data init_data;
 	int rc = -EINVAL;
 	u16 rx_mode = 0;
 	u8 abs_vport_id = 0;
@@ -153,16 +153,14 @@ static int qed_sp_vport_start(struct qed_hwfn *p_hwfn,
 	if (rc != 0)
 		return rc;
 
-	memset(&params, 0, sizeof(params));
-	params.ramrod_data_size = sizeof(*p_ramrod);
-	params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 qed_spq_get_cid(p_hwfn),
-				 opaque_fid,
 				 ETH_RAMROD_VPORT_START,
-				 PROTOCOLID_ETH,
-				 &params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -362,7 +360,7 @@ qed_sp_vport_update(struct qed_hwfn *p_hwfn,
 {
 	struct qed_rss_params *p_rss_params = p_params->rss_params;
 	struct vport_update_ramrod_data_cmn *p_cmn;
-	struct qed_sp_init_request_params sp_params;
+	struct qed_sp_init_data init_data;
 	struct vport_update_ramrod_data *p_ramrod = NULL;
 	struct qed_spq_entry *p_ent = NULL;
 	u8 abs_vport_id = 0;
@@ -372,17 +370,15 @@ qed_sp_vport_update(struct qed_hwfn *p_hwfn,
 	if (rc != 0)
 		return rc;
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(*p_ramrod);
-	sp_params.comp_mode = comp_mode;
-	sp_params.p_comp_data = p_comp_data;
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = p_params->opaque_fid;
+	init_data.comp_mode = comp_mode;
+	init_data.p_comp_data = p_comp_data;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 qed_spq_get_cid(p_hwfn),
-				 p_params->opaque_fid,
 				 ETH_RAMROD_VPORT_UPDATE,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -416,8 +412,8 @@ static int qed_sp_vport_stop(struct qed_hwfn *p_hwfn,
 			     u16 opaque_fid,
 			     u8 vport_id)
 {
-	struct qed_sp_init_request_params sp_params;
 	struct vport_stop_ramrod_data *p_ramrod;
+	struct qed_sp_init_data init_data;
 	struct qed_spq_entry *p_ent;
 	u8 abs_vport_id = 0;
 	int rc;
@@ -426,16 +422,14 @@ static int qed_sp_vport_stop(struct qed_hwfn *p_hwfn,
 	if (rc != 0)
 		return rc;
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(*p_ramrod);
-	sp_params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 qed_spq_get_cid(p_hwfn),
-				 opaque_fid,
 				 ETH_RAMROD_VPORT_STOP,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -514,8 +508,8 @@ qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn,
 			    u16 cqe_pbl_size)
 {
 	struct rx_queue_start_ramrod_data *p_ramrod = NULL;
-	struct qed_sp_init_request_params sp_params;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	struct qed_hw_cid_data *p_rx_cid;
 	u16 abs_rx_q_id = 0;
 	u8 abs_vport_id = 0;
@@ -540,15 +534,15 @@ qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn,
 		   opaque_fid, cid, params->queue_id, params->vport_id,
 		   params->sb);
 
-	memset(&sp_params, 0, sizeof(params));
-	sp_params.comp_mode = QED_SPQ_MODE_EBLOCK;
-	sp_params.ramrod_data_size = sizeof(*p_ramrod);
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = cid;
+	init_data.opaque_fid = opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 cid, opaque_fid,
 				 ETH_RAMROD_RX_QUEUE_START,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -640,21 +634,20 @@ static int qed_sp_eth_rx_queue_stop(struct qed_hwfn *p_hwfn,
 {
 	struct qed_hw_cid_data *p_rx_cid = &p_hwfn->p_rx_cids[rx_queue_id];
 	struct rx_queue_stop_ramrod_data *p_ramrod = NULL;
-	struct qed_sp_init_request_params sp_params;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	u16 abs_rx_q_id = 0;
 	int rc = -EINVAL;
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(*p_ramrod);
-	sp_params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = p_rx_cid->cid;
+	init_data.opaque_fid = p_rx_cid->opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 p_rx_cid->cid,
-				 p_rx_cid->opaque_fid,
 				 ETH_RAMROD_RX_QUEUE_STOP,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -692,8 +685,8 @@ qed_sp_eth_txq_start_ramrod(struct qed_hwfn  *p_hwfn,
 			    union qed_qm_pq_params *p_pq_params)
 {
 	struct tx_queue_start_ramrod_data *p_ramrod = NULL;
-	struct qed_sp_init_request_params sp_params;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	struct qed_hw_cid_data *p_tx_cid;
 	u8 abs_vport_id;
 	int rc = -EINVAL;
@@ -708,15 +701,15 @@ qed_sp_eth_txq_start_ramrod(struct qed_hwfn  *p_hwfn,
 	if (rc)
 		return rc;
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(*p_ramrod);
-	sp_params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = cid;
+	init_data.opaque_fid = opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
-	rc = qed_sp_init_request(p_hwfn, &p_ent, cid,
-				 opaque_fid,
+	rc = qed_sp_init_request(p_hwfn, &p_ent,
 				 ETH_RAMROD_TX_QUEUE_START,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -796,20 +789,19 @@ static int qed_sp_eth_tx_queue_stop(struct qed_hwfn *p_hwfn,
 				    u16 tx_queue_id)
 {
 	struct qed_hw_cid_data *p_tx_cid = &p_hwfn->p_tx_cids[tx_queue_id];
-	struct qed_sp_init_request_params sp_params;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	int rc = -EINVAL;
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(struct tx_queue_stop_ramrod_data);
-	sp_params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = p_tx_cid->cid;
+	init_data.opaque_fid = p_tx_cid->opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 p_tx_cid->cid,
-				 p_tx_cid->opaque_fid,
 				 ETH_RAMROD_TX_QUEUE_STOP,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -866,9 +858,9 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn,
 {
 	u8 vport_to_add_to = 0, vport_to_remove_from = 0;
 	struct vport_filter_update_ramrod_data *p_ramrod;
-	struct qed_sp_init_request_params sp_params;
 	struct eth_filter_cmd *p_first_filter;
 	struct eth_filter_cmd *p_second_filter;
+	struct qed_sp_init_data init_data;
 	enum eth_filter_action action;
 	int rc;
 
@@ -882,17 +874,16 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn,
 	if (rc)
 		return rc;
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(**pp_ramrod);
-	sp_params.comp_mode = comp_mode;
-	sp_params.p_comp_data = p_comp_data;
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = opaque_fid;
+	init_data.comp_mode = comp_mode;
+	init_data.p_comp_data = p_comp_data;
 
 	rc = qed_sp_init_request(p_hwfn, pp_ent,
-				 qed_spq_get_cid(p_hwfn),
-				 opaque_fid,
 				 ETH_RAMROD_FILTERS_UPDATE,
-				 PROTOCOLID_ETH,
-				 &sp_params);
+				 PROTOCOLID_ETH, &init_data);
 	if (rc)
 		return rc;
 
@@ -1116,8 +1107,8 @@ qed_sp_eth_filter_mcast(struct qed_hwfn *p_hwfn,
 {
 	unsigned long bins[ETH_MULTICAST_MAC_BINS_IN_REGS];
 	struct vport_update_ramrod_data *p_ramrod = NULL;
-	struct qed_sp_init_request_params sp_params;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	u8 abs_vport_id = 0;
 	int rc, i;
 
@@ -1133,18 +1124,16 @@ qed_sp_eth_filter_mcast(struct qed_hwfn *p_hwfn,
 			return rc;
 	}
 
-	memset(&sp_params, 0, sizeof(sp_params));
-	sp_params.ramrod_data_size = sizeof(*p_ramrod);
-	sp_params.comp_mode = comp_mode;
-	sp_params.p_comp_data = p_comp_data;
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = p_hwfn->hw_info.opaque_fid;
+	init_data.comp_mode = comp_mode;
+	init_data.p_comp_data = p_comp_data;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
-				 qed_spq_get_cid(p_hwfn),
-				 p_hwfn->hw_info.opaque_fid,
 				 ETH_RAMROD_VPORT_UPDATE,
-				 PROTOCOLID_ETH,
-				 &sp_params);
-
+				 PROTOCOLID_ETH, &init_data);
 	if (rc) {
 		DP_ERR(p_hwfn, "Multi-cast command failed %d\n", rc);
 		return rc;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h
index 8a83609..d39f914 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h
@@ -311,19 +311,20 @@ void qed_consq_free(struct qed_hwfn *p_hwfn,
 #define QED_SP_EQ_COMPLETION  0x01
 #define QED_SP_CQE_COMPLETION 0x02
 
-struct qed_sp_init_request_params {
-	size_t			ramrod_data_size;
+struct qed_sp_init_data {
+	u32			cid;
+	u16			opaque_fid;
+
+	/* Information regarding operation upon sending & completion */
 	enum spq_mode		comp_mode;
 	struct qed_spq_comp_cb *p_comp_data;
 };
 
 int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 			struct qed_spq_entry **pp_ent,
-			u32 cid,
-			u16 opaque_fid,
 			u8 cmd,
 			u8 protocol,
-			struct qed_sp_init_request_params *p_params);
+			struct qed_sp_init_data *p_data);
 
 /**
  * @brief qed_sp_pf_start - PF Function Start Ramrod
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
index 33090f6..e271ef9 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
@@ -23,15 +23,13 @@
 
 int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 			struct qed_spq_entry **pp_ent,
-			u32 cid,
-			u16 opaque_fid,
 			u8 cmd,
 			u8 protocol,
-			struct qed_sp_init_request_params *p_params)
+			struct qed_sp_init_data *p_data)
 {
-	int rc = -EINVAL;
+	u32 opaque_cid = p_data->opaque_fid << 16 | p_data->cid;
 	struct qed_spq_entry *p_ent = NULL;
-	u32 opaque_cid = opaque_fid << 16 | cid;
+	int rc;
 
 	if (!pp_ent)
 		return -ENOMEM;
@@ -48,7 +46,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 	p_ent->elem.hdr.protocol_id	= protocol;
 
 	p_ent->priority		= QED_SPQ_PRIORITY_NORMAL;
-	p_ent->comp_mode	= p_params->comp_mode;
+	p_ent->comp_mode	= p_data->comp_mode;
 	p_ent->comp_done.done	= 0;
 
 	switch (p_ent->comp_mode) {
@@ -57,17 +55,17 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 		break;
 
 	case QED_SPQ_MODE_BLOCK:
-		if (!p_params->p_comp_data)
+		if (!p_data->p_comp_data)
 			return -EINVAL;
 
-		p_ent->comp_cb.cookie = p_params->p_comp_data->cookie;
+		p_ent->comp_cb.cookie = p_data->p_comp_data->cookie;
 		break;
 
 	case QED_SPQ_MODE_CB:
-		if (!p_params->p_comp_data)
+		if (!p_data->p_comp_data)
 			p_ent->comp_cb.function = NULL;
 		else
-			p_ent->comp_cb = *p_params->p_comp_data;
+			p_ent->comp_cb = *p_data->p_comp_data;
 		break;
 
 	default:
@@ -83,8 +81,8 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 		   D_TRINE(p_ent->comp_mode, QED_SPQ_MODE_EBLOCK,
 			   QED_SPQ_MODE_BLOCK, "MODE_EBLOCK", "MODE_BLOCK",
 			   "MODE_CB"));
-	if (p_params->ramrod_data_size)
-		memset(&p_ent->ramrod, 0, p_params->ramrod_data_size);
+
+	memset(&p_ent->ramrod, 0, sizeof(p_ent->ramrod));
 
 	return 0;
 }
@@ -92,28 +90,26 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
 int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
 		    enum qed_mf_mode mode)
 {
-	struct qed_sp_init_request_params params;
 	struct pf_start_ramrod_data *p_ramrod = NULL;
 	u16 sb = qed_int_get_sp_sb_id(p_hwfn);
 	u8 sb_index = p_hwfn->p_eq->eq_sb_index;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	int rc = -EINVAL;
 
 	/* update initial eq producer */
 	qed_eq_prod_update(p_hwfn,
 			   qed_chain_get_prod_idx(&p_hwfn->p_eq->chain));
 
-	memset(&params, 0, sizeof(params));
-	params.ramrod_data_size = sizeof(*p_ramrod);
-	params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = p_hwfn->hw_info.opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
-	rc = qed_sp_init_request(p_hwfn,
-				 &p_ent,
-				 qed_spq_get_cid(p_hwfn),
-				 p_hwfn->hw_info.opaque_fid,
+	rc = qed_sp_init_request(p_hwfn, &p_ent,
 				 COMMON_RAMROD_PF_START,
 				 PROTOCOLID_COMMON,
-				 &params);
+				 &init_data);
 	if (rc)
 		return rc;
 
@@ -163,17 +159,19 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
 
 int qed_sp_pf_stop(struct qed_hwfn *p_hwfn)
 {
-	struct qed_sp_init_request_params params;
 	struct qed_spq_entry *p_ent = NULL;
+	struct qed_sp_init_data init_data;
 	int rc = -EINVAL;
 
-	memset(&params, 0, sizeof(params));
-	params.comp_mode = QED_SPQ_MODE_EBLOCK;
+	/* Get SPQ entry */
+	memset(&init_data, 0, sizeof(init_data));
+	init_data.cid = qed_spq_get_cid(p_hwfn);
+	init_data.opaque_fid = p_hwfn->hw_info.opaque_fid;
+	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
-	rc = qed_sp_init_request(p_hwfn, &p_ent, qed_spq_get_cid(p_hwfn),
-				 p_hwfn->hw_info.opaque_fid,
+	rc = qed_sp_init_request(p_hwfn, &p_ent,
 				 COMMON_RAMROD_PF_STOP, PROTOCOLID_COMMON,
-				 &params);
+				 &init_data);
 	if (rc)
 		return rc;
 
-- 
cgit v0.10.2


From 944945986f125bdbbeaa78dac0c0eadb963eb34a Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 21 Feb 2016 11:40:10 +0200
Subject: qed: Introduce DMA_REGPAIR_LE

FW hsi contains regpairs, mostly for 64-bit address representations.
Since same paradigm is applied each time a regpair is filled, this
introduces a new utility macro for setting such regpairs.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index 8d1bc7e..bba59c5 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -557,12 +557,10 @@ qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn,
 	p_ramrod->complete_event_flg	= 1;
 
 	p_ramrod->bd_max_bytes	= cpu_to_le16(bd_max_bytes);
-	p_ramrod->bd_base.hi	= DMA_HI_LE(bd_chain_phys_addr);
-	p_ramrod->bd_base.lo	= DMA_LO_LE(bd_chain_phys_addr);
+	DMA_REGPAIR_LE(p_ramrod->bd_base, bd_chain_phys_addr);
 
 	p_ramrod->num_of_pbl_pages	= cpu_to_le16(cqe_pbl_size);
-	p_ramrod->cqe_pbl_addr.hi	= DMA_HI_LE(cqe_pbl_addr);
-	p_ramrod->cqe_pbl_addr.lo	= DMA_LO_LE(cqe_pbl_addr);
+	DMA_REGPAIR_LE(p_ramrod->cqe_pbl_addr, cqe_pbl_addr);
 
 	rc = qed_spq_post(p_hwfn, p_ent, NULL);
 
@@ -721,8 +719,7 @@ qed_sp_eth_txq_start_ramrod(struct qed_hwfn  *p_hwfn,
 	p_ramrod->stats_counter_id	= stats_id;
 
 	p_ramrod->pbl_size		= cpu_to_le16(pbl_size);
-	p_ramrod->pbl_base_addr.hi	= DMA_HI_LE(pbl_addr);
-	p_ramrod->pbl_base_addr.lo	= DMA_LO_LE(pbl_addr);
+	DMA_REGPAIR_LE(p_ramrod->pbl_base_addr, pbl_addr);
 
 	pq_id			= qed_get_qm_pq(p_hwfn,
 						PROTOCOLID_ETH,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
index e271ef9..1c06c37 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
@@ -136,16 +136,12 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn,
 	p_ramrod->outer_tag = p_hwfn->hw_info.ovlan;
 
 	/* Place EQ address in RAMROD */
-	p_ramrod->event_ring_pbl_addr.hi =
-			DMA_HI_LE(p_hwfn->p_eq->chain.pbl.p_phys_table);
-	p_ramrod->event_ring_pbl_addr.lo =
-			DMA_LO_LE(p_hwfn->p_eq->chain.pbl.p_phys_table);
+	DMA_REGPAIR_LE(p_ramrod->event_ring_pbl_addr,
+		       p_hwfn->p_eq->chain.pbl.p_phys_table);
 	p_ramrod->event_ring_num_pages = (u8)p_hwfn->p_eq->chain.page_cnt;
 
-	p_ramrod->consolid_q_pbl_addr.hi =
-			DMA_HI_LE(p_hwfn->p_consq->chain.pbl.p_phys_table);
-	p_ramrod->consolid_q_pbl_addr.lo =
-			DMA_LO_LE(p_hwfn->p_consq->chain.pbl.p_phys_table);
+	DMA_REGPAIR_LE(p_ramrod->consolid_q_pbl_addr,
+		       p_hwfn->p_consq->chain.pbl.p_phys_table);
 
 	p_hwfn->hw_info.personality = PERSONALITY_ETH;
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c
index f6c6c21..89469d5 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_spq.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c
@@ -183,10 +183,8 @@ static void qed_spq_hw_initialize(struct qed_hwfn *p_hwfn,
 	p_cxt->xstorm_st_context.spq_base_hi =
 		DMA_HI_LE(p_spq->chain.p_phys_addr);
 
-	p_cxt->xstorm_st_context.consolid_base_addr.lo =
-		DMA_LO_LE(p_hwfn->p_consq->chain.p_phys_addr);
-	p_cxt->xstorm_st_context.consolid_base_addr.hi =
-		DMA_HI_LE(p_hwfn->p_consq->chain.p_phys_addr);
+	DMA_REGPAIR_LE(p_cxt->xstorm_st_context.consolid_base_addr,
+		       p_hwfn->p_consq->chain.p_phys_addr);
 }
 
 static int qed_spq_hw_post(struct qed_hwfn *p_hwfn,
@@ -423,8 +421,7 @@ void qed_spq_setup(struct qed_hwfn *p_hwfn)
 	p_virt	= p_spq->p_virt;
 
 	for (i = 0; i < p_spq->chain.capacity; i++) {
-		p_virt->elem.data_ptr.hi = DMA_HI_LE(p_phys);
-		p_virt->elem.data_ptr.lo = DMA_LO_LE(p_phys);
+		DMA_REGPAIR_LE(p_virt->elem.data_ptr, p_phys);
 
 		list_add_tail(&p_virt->list, &p_spq->free_pool);
 
diff --git a/include/linux/qed/qed_chain.h b/include/linux/qed/qed_chain.h
index 41b9049..5f8fcaa 100644
--- a/include/linux/qed/qed_chain.h
+++ b/include/linux/qed/qed_chain.h
@@ -19,6 +19,10 @@
 /* dma_addr_t manip */
 #define DMA_LO_LE(x)            cpu_to_le32(lower_32_bits(x))
 #define DMA_HI_LE(x)            cpu_to_le32(upper_32_bits(x))
+#define DMA_REGPAIR_LE(x, val)  do { \
+					(x).hi = DMA_HI_LE((val)); \
+					(x).lo = DMA_LO_LE((val)); \
+				} while (0)
 
 #define HILO_GEN(hi, lo, type)  ((((type)(hi)) << 32) + (lo))
 #define HILO_DMA(hi, lo)        HILO_GEN(hi, lo, dma_addr_t)
-- 
cgit v0.10.2


From d4ee52897b5b5cf9137545f6fd41dc7217565517 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 21 Feb 2016 11:40:11 +0200
Subject: qed,qede: Bump driver versions to 8.7.0.0

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index d34da63..66b021e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -26,7 +26,7 @@
 #include "qed_hsi.h"
 
 extern const struct qed_common_ops qed_common_ops_pass;
-#define DRV_MODULE_VERSION "8.4.0.0"
+#define DRV_MODULE_VERSION "8.7.0.0"
 
 #define MAX_HWFNS_PER_DEVICE    (4)
 #define NAME_SIZE 16
diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index 15c5528..02e17d3 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -24,7 +24,7 @@
 #include <linux/qed/qed_eth_if.h>
 
 #define QEDE_MAJOR_VERSION		8
-#define QEDE_MINOR_VERSION		4
+#define QEDE_MINOR_VERSION		7
 #define QEDE_REVISION_VERSION		0
 #define QEDE_ENGINEERING_VERSION	0
 #define DRV_MODULE_VERSION __stringify(QEDE_MAJOR_VERSION) "."	\
-- 
cgit v0.10.2


From 0c71de6634c5c1b6c9dcb80049680ad334205c23 Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Date: Mon, 22 Feb 2016 22:55:43 +0530
Subject: netcp: use pointer to fix build fail

While building keystone_defconfig of arm we are getting build failure
with the error:

drivers/net/ethernet/ti/netcp_core.c:1846:31: error: invalid type argument of '->' (have 'struct tc_to_netdev')
  if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
                               ^
drivers/net/ethernet/ti/netcp_core.c:1851:35: error: invalid type argument of '->' (have 'struct tc_to_netdev')
      (dev->real_num_tx_queues < tc->tc))
                                   ^
drivers/net/ethernet/ti/netcp_core.c:1855:8: error: invalid type argument of '->' (have 'struct tc_to_netdev')
  if (tc->tc) {
        ^
drivers/net/ethernet/ti/netcp_core.c:1856:28: error: invalid type argument of '->' (have 'struct tc_to_netdev')
   netdev_set_num_tc(dev, tc->tc);
                            ^
drivers/net/ethernet/ti/netcp_core.c:1857:21: error: invalid type argument of '->' (have 'struct tc_to_netdev')
   for (i = 0; i < tc->tc; i++)
                     ^
drivers/net/ethernet/ti/netcp_core.c: At top level:
drivers/net/ethernet/ti/netcp_core.c:1879:2: warning: initialization from incompatible pointer type
  .ndo_setup_tc  = netcp_setup_tc,
  ^

The callback of ndo_setup_tc should be:
int (*ndo_setup_tc)(struct net_device *dev, u32 handle, __be16 protocol,
                    struct tc_to_netdev *tc);

But we missed marking the last argument as a pointer.

Fixes: 16e5cc647173 ("net: rework setup_tc ndo op to consume general tc operand")
CC: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index 8586a20..06a0a73 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -1836,7 +1836,7 @@ static u16 netcp_select_queue(struct net_device *dev, struct sk_buff *skb,
 }
 
 static int netcp_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
-			  struct tc_to_netdev tc)
+			  struct tc_to_netdev *tc)
 {
 	int i;
 
-- 
cgit v0.10.2


From b1d95ae5c5bd3deba84d00c4f83d7d0836b5936f Mon Sep 17 00:00:00 2001
From: Ray Bellis <ray@isc.org>
Date: Mon, 22 Feb 2016 11:02:40 +0100
Subject: tools, bpf_asm: simplify parser rule for BPF extensions

We can already use yylval in the lexer for encoding the BPF extension
number, so that the parser rules can be further reduced to a single one
for each B/H/W case.

Signed-off-by: Ray Bellis <ray@isc.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/tools/net/bpf_exp.l b/tools/net/bpf_exp.l
index 7cc72a3..bd83149 100644
--- a/tools/net/bpf_exp.l
+++ b/tools/net/bpf_exp.l
@@ -23,6 +23,9 @@
 #include <stdio.h>
 #include <stdint.h>
 #include <stdlib.h>
+#include <string.h>
+
+#include <linux/filter.h>
 
 #include "bpf_exp.yacc.h"
 
@@ -79,22 +82,71 @@ extern void yyerror(const char *str);
 "txa"		{ return OP_TXA; }
 
 "#"?("len")	{ return K_PKT_LEN; }
-"#"?("proto")	{ return K_PROTO; }
-"#"?("type")	{ return K_TYPE; }
-"#"?("poff")	{ return K_POFF; }
-"#"?("ifidx")	{ return K_IFIDX; }
-"#"?("nla")	{ return K_NLATTR; }
-"#"?("nlan")	{ return K_NLATTR_NEST; }
-"#"?("mark")	{ return K_MARK; }
-"#"?("queue")	{ return K_QUEUE; }
-"#"?("hatype")	{ return K_HATYPE; }
-"#"?("rxhash")	{ return K_RXHASH; }
-"#"?("cpu")	{ return K_CPU; }
-"#"?("vlan_tci")	{ return K_VLAN_TCI; }
-"#"?("vlan_pr")		{ return K_VLAN_AVAIL; }
-"#"?("vlan_avail")	{ return K_VLAN_AVAIL; }
-"#"?("vlan_tpid")	{ return K_VLAN_TPID; }
-"#"?("rand")	{ return K_RAND; }
+
+"#"?("proto")	{
+		yylval.number = SKF_AD_PROTOCOL;
+		return extension;
+	}
+"#"?("type")	{
+		yylval.number = SKF_AD_PKTTYPE;
+		return extension;
+	}
+"#"?("poff")	{
+		yylval.number = SKF_AD_PAY_OFFSET;
+		return extension;
+	}
+"#"?("ifidx")	{
+		yylval.number = SKF_AD_IFINDEX;
+		return extension;
+	}
+"#"?("nla")	{
+		yylval.number = SKF_AD_NLATTR;
+		return extension;
+	}
+"#"?("nlan")	{
+		yylval.number = SKF_AD_NLATTR_NEST;
+		return extension;
+	}
+"#"?("mark")	{
+		yylval.number = SKF_AD_MARK;
+		return extension;
+	}
+"#"?("queue")	{
+		yylval.number = SKF_AD_QUEUE;
+		return extension;
+	}
+"#"?("hatype")	{
+		yylval.number = SKF_AD_HATYPE;
+		return extension;
+	}
+"#"?("rxhash")	{
+		yylval.number = SKF_AD_RXHASH;
+		return extension;
+	}
+"#"?("cpu")	{
+		yylval.number = SKF_AD_CPU;
+		return extension;
+	}
+"#"?("vlan_tci") {
+		yylval.number = SKF_AD_VLAN_TAG;
+		return extension;
+	}
+"#"?("vlan_pr")	{
+		yylval.number = SKF_AD_VLAN_TAG_PRESENT;
+		return extension;
+	}
+"#"?("vlan_avail") {
+		yylval.number = SKF_AD_VLAN_TAG_PRESENT;
+		return extension;
+	}
+"#"?("vlan_tpid") {
+		yylval.number = SKF_AD_VLAN_TPID;
+		return extension;
+	}
+"#"?("rand")	{
+		yylval.number = SKF_AD_RANDOM;
+		return extension;
+	}
 
 ":"		{ return ':'; }
 ","		{ return ','; }
diff --git a/tools/net/bpf_exp.y b/tools/net/bpf_exp.y
index e24eea1..56ba1de 100644
--- a/tools/net/bpf_exp.y
+++ b/tools/net/bpf_exp.y
@@ -35,6 +35,7 @@
 enum jmp_type { JTL, JFL, JKL };
 
 extern FILE *yyin;
+extern int yylineno;
 extern int yylex(void);
 extern void yyerror(const char *str);
 
@@ -55,14 +56,14 @@ static void bpf_set_jmp_label(char *label, enum jmp_type type);
 %token OP_RET OP_TAX OP_TXA OP_LDXB OP_MOD OP_NEG OP_JNEQ OP_JLT OP_JLE OP_LDI
 %token OP_LDXI
 
-%token K_PKT_LEN K_PROTO K_TYPE K_NLATTR K_NLATTR_NEST K_MARK K_QUEUE K_HATYPE
-%token K_RXHASH K_CPU K_IFIDX K_VLAN_TCI K_VLAN_AVAIL K_VLAN_TPID K_POFF K_RAND
+%token K_PKT_LEN
 
 %token ':' ',' '[' ']' '(' ')' 'x' 'a' '+' 'M' '*' '&' '#' '%'
 
-%token number label
+%token extension number label
 
 %type <label> label
+%type <number> extension
 %type <number> number
 
 %%
@@ -125,51 +126,9 @@ ldb
 		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_IND, 0, 0, $6); }
 	| OP_LDB '[' number ']' {
 		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0, $3); }
-	| OP_LDB K_PROTO {
+	| OP_LDB extension {
 		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PROTOCOL); }
-	| OP_LDB K_TYPE {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PKTTYPE); }
-	| OP_LDB K_IFIDX {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_IFINDEX); }
-	| OP_LDB K_NLATTR {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_NLATTR); }
-	| OP_LDB K_NLATTR_NEST {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_NLATTR_NEST); }
-	| OP_LDB K_MARK {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_MARK); }
-	| OP_LDB K_QUEUE {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_QUEUE); }
-	| OP_LDB K_HATYPE {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_HATYPE); }
-	| OP_LDB K_RXHASH {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_RXHASH); }
-	| OP_LDB K_CPU {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_CPU); }
-	| OP_LDB K_VLAN_TCI {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TAG); }
-	| OP_LDB K_VLAN_AVAIL {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TAG_PRESENT); }
-	| OP_LDB K_POFF {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PAY_OFFSET); }
-	| OP_LDB K_RAND {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_RANDOM); }
-	| OP_LDB K_VLAN_TPID {
-		bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TPID); }
+				   SKF_AD_OFF + $2); }
 	;
 
 ldh
@@ -179,51 +138,9 @@ ldh
 		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_IND, 0, 0, $6); }
 	| OP_LDH '[' number ']' {
 		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0, $3); }
-	| OP_LDH K_PROTO {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PROTOCOL); }
-	| OP_LDH K_TYPE {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PKTTYPE); }
-	| OP_LDH K_IFIDX {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_IFINDEX); }
-	| OP_LDH K_NLATTR {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_NLATTR); }
-	| OP_LDH K_NLATTR_NEST {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_NLATTR_NEST); }
-	| OP_LDH K_MARK {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_MARK); }
-	| OP_LDH K_QUEUE {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_QUEUE); }
-	| OP_LDH K_HATYPE {
+	| OP_LDH extension {
 		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_HATYPE); }
-	| OP_LDH K_RXHASH {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_RXHASH); }
-	| OP_LDH K_CPU {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_CPU); }
-	| OP_LDH K_VLAN_TCI {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TAG); }
-	| OP_LDH K_VLAN_AVAIL {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TAG_PRESENT); }
-	| OP_LDH K_POFF {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PAY_OFFSET); }
-	| OP_LDH K_RAND {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_RANDOM); }
-	| OP_LDH K_VLAN_TPID {
-		bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TPID); }
+				   SKF_AD_OFF + $2); }
 	;
 
 ldi
@@ -238,51 +155,9 @@ ld
 		bpf_set_curr_instr(BPF_LD | BPF_IMM, 0, 0, $3); }
 	| OP_LD K_PKT_LEN {
 		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_LEN, 0, 0, 0); }
-	| OP_LD K_PROTO {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PROTOCOL); }
-	| OP_LD K_TYPE {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PKTTYPE); }
-	| OP_LD K_IFIDX {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_IFINDEX); }
-	| OP_LD K_NLATTR {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_NLATTR); }
-	| OP_LD K_NLATTR_NEST {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_NLATTR_NEST); }
-	| OP_LD K_MARK {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_MARK); }
-	| OP_LD K_QUEUE {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_QUEUE); }
-	| OP_LD K_HATYPE {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_HATYPE); }
-	| OP_LD K_RXHASH {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_RXHASH); }
-	| OP_LD K_CPU {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_CPU); }
-	| OP_LD K_VLAN_TCI {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TAG); }
-	| OP_LD K_VLAN_AVAIL {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TAG_PRESENT); }
-	| OP_LD K_POFF {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_PAY_OFFSET); }
-	| OP_LD K_RAND {
-		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_RANDOM); }
-	| OP_LD K_VLAN_TPID {
+	| OP_LD extension {
 		bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0,
-				   SKF_AD_OFF + SKF_AD_VLAN_TPID); }
+				   SKF_AD_OFF + $2); }
 	| OP_LD 'M' '[' number ']' {
 		bpf_set_curr_instr(BPF_LD | BPF_MEM, 0, 0, $4); }
 	| OP_LD '[' 'x' '+' number ']' {
@@ -776,5 +651,6 @@ void bpf_asm_compile(FILE *fp, bool cstyle)
 
 void yyerror(const char *str)
 {
+	fprintf(stderr, "error: %s at line %d\n", str, yylineno);
 	exit(1);
 }
-- 
cgit v0.10.2


From e9ccd7e39cd87cb5ff87e2b293dcd467377d93d6 Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <a@unstable.cc>
Date: Tue, 12 Jan 2016 09:15:25 +0800
Subject: batman-adv: remove unused BATADV_BONDING_TQ_THRESHOLD constant

BATADV_BONDING_TQ_THRESHOLD is not used anymore since the implementation
of the bat_neigh_is_similar_or_better() API function.
Such function uses the more generic BATADV_TQ_SIMILARITY_THRESHOLD
constant.

Therefore, remove definition of the unused BATADV_BONDING_TQ_THRESHOLD
constant.

Signed-off-by: Antonio Quartulli <a@unstable.cc>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 32dfc9e..8c01f54 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -100,11 +100,6 @@
  */
 #define BATADV_TQ_SIMILARITY_THRESHOLD 50
 
-/* how much worse secondary interfaces may be to be considered as bonding
- * candidates
- */
-#define BATADV_BONDING_TQ_THRESHOLD	50
-
 /* should not be bigger than 512 bytes or change the size of
  * forw_packet->direct_link_flags
  */
-- 
cgit v0.10.2


From 5d9673109c7c7c6a22b7bedba22701e173e1ea0d Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:09 +0100
Subject: batman-adv: Rename batadv_orig_node *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index bf0e7d6..32df38b 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -287,8 +287,8 @@ batadv_iv_ogm_orig_get(struct batadv_priv *bat_priv, const u8 *addr)
 
 free_orig_node:
 	/* free twice, as batadv_orig_node_new sets refcount to 2 */
-	batadv_orig_node_free_ref(orig_node);
-	batadv_orig_node_free_ref(orig_node);
+	batadv_orig_node_put(orig_node);
+	batadv_orig_node_put(orig_node);
 
 	return NULL;
 }
@@ -1041,7 +1041,7 @@ batadv_iv_ogm_orig_update(struct batadv_priv *bat_priv,
 						     ethhdr->h_source,
 						     orig_node, orig_tmp);
 
-		batadv_orig_node_free_ref(orig_tmp);
+		batadv_orig_node_put(orig_tmp);
 		if (!neigh_node)
 			goto unlock;
 	} else {
@@ -1306,7 +1306,7 @@ batadv_iv_ogm_update_seqnos(const struct ethhdr *ethhdr,
 
 	orig_ifinfo = batadv_orig_ifinfo_new(orig_node, if_outgoing);
 	if (WARN_ON(!orig_ifinfo)) {
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 		return 0;
 	}
 
@@ -1367,7 +1367,7 @@ batadv_iv_ogm_update_seqnos(const struct ethhdr *ethhdr,
 
 out:
 	spin_unlock_bh(&orig_node->bat_iv.ogm_cnt_lock);
-	batadv_orig_node_free_ref(orig_node);
+	batadv_orig_node_put(orig_node);
 	batadv_orig_ifinfo_free_ref(orig_ifinfo);
 	return ret;
 }
@@ -1563,7 +1563,7 @@ batadv_iv_ogm_process_per_outif(const struct sk_buff *skb, int ogm_offset,
 
 out_neigh:
 	if ((orig_neigh_node) && (!is_single_hop_neigh))
-		batadv_orig_node_free_ref(orig_neigh_node);
+		batadv_orig_node_put(orig_neigh_node);
 out:
 	if (router_ifinfo)
 		batadv_neigh_ifinfo_free_ref(router_ifinfo);
@@ -1697,7 +1697,7 @@ static void batadv_iv_ogm_process(const struct sk_buff *skb, int ogm_offset,
 
 		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
 			   "Drop packet: originator packet from myself (via neighbor)\n");
-		batadv_orig_node_free_ref(orig_neigh_node);
+		batadv_orig_node_put(orig_neigh_node);
 		return;
 	}
 
@@ -1735,7 +1735,7 @@ static void batadv_iv_ogm_process(const struct sk_buff *skb, int ogm_offset,
 	}
 	rcu_read_unlock();
 
-	batadv_orig_node_free_ref(orig_node);
+	batadv_orig_node_put(orig_node);
 }
 
 static int batadv_iv_ogm_receive(struct sk_buff *skb,
diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 7781f39..49008d9 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -486,7 +486,7 @@ batadv_bla_get_backbone_gw(struct batadv_priv *bat_priv, u8 *orig,
 	if (orig_node) {
 		batadv_tt_global_del_orig(bat_priv, orig_node, vid,
 					  "became a backbone gateway");
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	}
 
 	if (own_backbone) {
@@ -965,7 +965,7 @@ static int batadv_check_claim_group(struct batadv_priv *bat_priv,
 		bla_dst_own->group = bla_dst->group;
 	}
 
-	batadv_orig_node_free_ref(orig_node);
+	batadv_orig_node_put(orig_node);
 
 	return 2;
 }
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index e326111..8f466e7 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -547,7 +547,7 @@ static void batadv_choose_next_candidate(struct batadv_priv *bat_priv,
 
 			max = tmp_max;
 			if (max_orig_node)
-				batadv_orig_node_free_ref(max_orig_node);
+				batadv_orig_node_put(max_orig_node);
 			max_orig_node = orig_node;
 		}
 		rcu_read_unlock();
@@ -676,7 +676,7 @@ static bool batadv_dat_send_data(struct batadv_priv *bat_priv,
 free_neigh:
 		batadv_neigh_node_free_ref(neigh_node);
 free_orig:
-		batadv_orig_node_free_ref(cand[i].orig_node);
+		batadv_orig_node_put(cand[i].orig_node);
 	}
 
 out:
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index 55656e8..67cb224 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -385,7 +385,7 @@ bool batadv_frag_skb_fwd(struct sk_buff *skb,
 
 out:
 	if (orig_node_dst)
-		batadv_orig_node_free_ref(orig_node_dst);
+		batadv_orig_node_put(orig_node_dst);
 	if (neigh_node)
 		batadv_neigh_node_free_ref(neigh_node);
 	return ret;
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 4b598bd..9cb098f 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -71,7 +71,7 @@ static void batadv_gw_node_release(struct kref *ref)
 
 	gw_node = container_of(ref, struct batadv_gw_node, refcount);
 
-	batadv_orig_node_free_ref(gw_node->orig_node);
+	batadv_orig_node_put(gw_node->orig_node);
 	kfree_rcu(gw_node, rcu);
 }
 
@@ -415,7 +415,7 @@ reselect:
 	batadv_gw_reselect(bat_priv);
 out:
 	if (curr_gw_orig)
-		batadv_orig_node_free_ref(curr_gw_orig);
+		batadv_orig_node_put(curr_gw_orig);
 	if (router_gw)
 		batadv_neigh_node_free_ref(router_gw);
 	if (router_orig)
@@ -446,7 +446,7 @@ static void batadv_gw_node_add(struct batadv_priv *bat_priv,
 
 	gw_node = kzalloc(sizeof(*gw_node), GFP_ATOMIC);
 	if (!gw_node) {
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 		return;
 	}
 
@@ -879,7 +879,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
 
 out:
 	if (orig_dst_node)
-		batadv_orig_node_free_ref(orig_dst_node);
+		batadv_orig_node_put(orig_dst_node);
 	if (curr_gw)
 		batadv_gw_node_free_ref(curr_gw);
 	if (gw_node)
diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c
index a69da37..7d944af 100644
--- a/net/batman-adv/icmp_socket.c
+++ b/net/batman-adv/icmp_socket.c
@@ -292,7 +292,7 @@ out:
 	if (neigh_node)
 		batadv_neigh_node_free_ref(neigh_node);
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return len;
 }
 
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 568c550..5b573d1 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -1212,7 +1212,7 @@ void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src,
 	if (batadv_send_skb_to_orig(skb, orig_node, NULL) == NET_XMIT_DROP)
 		kfree_skb(skb);
 out:
-	batadv_orig_node_free_ref(orig_node);
+	batadv_orig_node_put(orig_node);
 }
 
 /**
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index a4eb8ee..00f74d4 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -218,7 +218,7 @@ static void batadv_nc_node_release(struct kref *ref)
 
 	nc_node = container_of(ref, struct batadv_nc_node, refcount);
 
-	batadv_orig_node_free_ref(nc_node->orig_node);
+	batadv_orig_node_put(nc_node->orig_node);
 	kfree_rcu(nc_node, rcu);
 }
 
@@ -1372,7 +1372,7 @@ batadv_nc_skb_src_search(struct batadv_priv *bat_priv,
 	}
 	rcu_read_unlock();
 
-	batadv_orig_node_free_ref(orig_node);
+	batadv_orig_node_put(orig_node);
 	return nc_packet;
 }
 
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index eacd0e5..4df9be4 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -810,11 +810,11 @@ static void batadv_orig_node_release(struct kref *ref)
 }
 
 /**
- * batadv_orig_node_free_ref - decrement the orig node refcounter and possibly
+ * batadv_orig_node_put - decrement the orig node refcounter and possibly
  *  release it
  * @orig_node: the orig node to free
  */
-void batadv_orig_node_free_ref(struct batadv_orig_node *orig_node)
+void batadv_orig_node_put(struct batadv_orig_node *orig_node)
 {
 	kref_put(&orig_node->refcount, batadv_orig_node_release);
 }
@@ -843,7 +843,7 @@ void batadv_originator_free(struct batadv_priv *bat_priv)
 		hlist_for_each_entry_safe(orig_node, node_tmp,
 					  head, hash_entry) {
 			hlist_del_rcu(&orig_node->hash_entry);
-			batadv_orig_node_free_ref(orig_node);
+			batadv_orig_node_put(orig_node);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -1204,7 +1204,7 @@ static void _batadv_purge_orig(struct batadv_priv *bat_priv)
 				batadv_tt_global_del_orig(orig_node->bat_priv,
 							  orig_node, -1,
 							  "originator timed out");
-				batadv_orig_node_free_ref(orig_node);
+				batadv_orig_node_put(orig_node);
 				continue;
 			}
 
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 9950740..c151f6a 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -37,7 +37,7 @@ int batadv_compare_orig(const struct hlist_node *node, const void *data2);
 int batadv_originator_init(struct batadv_priv *bat_priv);
 void batadv_originator_free(struct batadv_priv *bat_priv);
 void batadv_purge_orig_ref(struct batadv_priv *bat_priv);
-void batadv_orig_node_free_ref(struct batadv_orig_node *orig_node);
+void batadv_orig_node_put(struct batadv_orig_node *orig_node);
 struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv,
 					      const u8 *addr);
 struct batadv_hardif_neigh_node *
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 205310b..d2c490d 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -271,7 +271,7 @@ out:
 	if (primary_if)
 		batadv_hardif_free_ref(primary_if);
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return ret;
 }
 
@@ -319,7 +319,7 @@ out:
 	if (primary_if)
 		batadv_hardif_free_ref(primary_if);
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return ret;
 }
 
@@ -403,7 +403,7 @@ int batadv_recv_icmp_packet(struct sk_buff *skb,
 
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return ret;
 }
 
@@ -649,7 +649,7 @@ static int batadv_route_unicast_packet(struct sk_buff *skb,
 
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return ret;
 }
 
@@ -704,7 +704,7 @@ out:
 	if (primary_if)
 		batadv_hardif_free_ref(primary_if);
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 
 	return ret;
 }
@@ -768,7 +768,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv,
 			return 0;
 
 		curr_ttvn = (u8)atomic_read(&orig_node->last_ttvn);
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	}
 
 	/* check if the TTVN contained in the packet is fresher than what the
@@ -908,7 +908,7 @@ int batadv_recv_unicast_packet(struct sk_buff *skb,
 
 rx_success:
 		if (orig_node)
-			batadv_orig_node_free_ref(orig_node);
+			batadv_orig_node_put(orig_node);
 
 		return NET_RX_SUCCESS;
 	}
@@ -1019,7 +1019,7 @@ int batadv_recv_frag_packet(struct sk_buff *skb,
 
 out:
 	if (orig_node_src)
-		batadv_orig_node_free_ref(orig_node_src);
+		batadv_orig_node_put(orig_node_src);
 
 	return ret;
 }
@@ -1124,6 +1124,6 @@ spin_unlock:
 	spin_unlock_bh(&orig_node->bcast_seqno_lock);
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return ret;
 }
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index d8b03fd..261309e 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -317,7 +317,7 @@ int batadv_send_skb_unicast(struct batadv_priv *bat_priv,
 
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	if (ret == NET_XMIT_DROP)
 		kfree_skb(skb);
 	return ret;
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 5c7fa02..107532c 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -392,7 +392,7 @@ static void batadv_tt_orig_list_entry_release(struct kref *ref)
 	orig_entry = container_of(ref, struct batadv_tt_orig_list_entry,
 				  refcount);
 
-	batadv_orig_node_free_ref(orig_entry->orig_node);
+	batadv_orig_node_put(orig_entry->orig_node);
 	kfree_rcu(orig_entry, rcu);
 }
 
@@ -2776,9 +2776,9 @@ unlock:
 
 out:
 	if (res_dst_orig_node)
-		batadv_orig_node_free_ref(res_dst_orig_node);
+		batadv_orig_node_put(res_dst_orig_node);
 	if (req_dst_orig_node)
-		batadv_orig_node_free_ref(req_dst_orig_node);
+		batadv_orig_node_put(req_dst_orig_node);
 	kfree(tvlv_tt_data);
 	return ret;
 }
@@ -2893,7 +2893,7 @@ unlock:
 out:
 	spin_unlock_bh(&bat_priv->tt.commit_lock);
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	if (primary_if)
 		batadv_hardif_free_ref(primary_if);
 	kfree(tvlv_tt_data);
@@ -2981,7 +2981,7 @@ static void batadv_tt_fill_gtable(struct batadv_priv *bat_priv,
 
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 }
 
 static void batadv_tt_update_changes(struct batadv_priv *bat_priv,
@@ -3087,7 +3087,7 @@ static void batadv_handle_tt_response(struct batadv_priv *bat_priv,
 	spin_unlock_bh(&bat_priv->tt.req_list_lock);
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 }
 
 static void batadv_tt_roam_list_free(struct batadv_priv *bat_priv)
@@ -3802,7 +3802,7 @@ static int batadv_roam_tvlv_unicast_handler_v1(struct batadv_priv *bat_priv,
 
 out:
 	if (orig_node)
-		batadv_orig_node_free_ref(orig_node);
+		batadv_orig_node_put(orig_node);
 	return NET_RX_SUCCESS;
 }
 
-- 
cgit v0.10.2


From 82047ad7fecdc2d58bbeeb3872811b1c528fdf10 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:10 +0100
Subject: batman-adv: Rename batadv_hardif *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 32df38b..f01ee23 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -515,7 +515,7 @@ static void batadv_iv_ogm_emit(struct batadv_forw_packet *forw_packet)
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 /**
@@ -617,7 +617,7 @@ batadv_iv_ogm_can_aggregate(const struct batadv_ogm_packet *new_bat_ogm_packet,
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return res;
 }
 
@@ -711,9 +711,9 @@ out_nomem:
 	if (!own_packet)
 		atomic_inc(&bat_priv->batman_queue_left);
 out_free_outgoing:
-	batadv_hardif_free_ref(if_outgoing);
+	batadv_hardif_put(if_outgoing);
 out_free_incoming:
-	batadv_hardif_free_ref(if_incoming);
+	batadv_hardif_put(if_incoming);
 }
 
 /* aggregate a new packet into the existing ogm packet */
@@ -958,7 +958,7 @@ static void batadv_iv_ogm_schedule(struct batadv_hard_iface *hard_iface)
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 /**
diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index 49008d9..b7167cf 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -424,7 +424,7 @@ static void batadv_bla_send_claim(struct batadv_priv *bat_priv, u8 *mac,
 	netif_rx(skb);
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 /**
@@ -1282,7 +1282,7 @@ void batadv_bla_status_update(struct net_device *net_dev)
 	 * so just call that one.
 	 */
 	batadv_bla_update_orig_address(bat_priv, primary_if, primary_if);
-	batadv_hardif_free_ref(primary_if);
+	batadv_hardif_put(primary_if);
 }
 
 /**
@@ -1356,7 +1356,7 @@ static void batadv_bla_periodic_work(struct work_struct *work)
 	}
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 
 	queue_delayed_work(batadv_event_workqueue, &bat_priv->bla.work,
 			   msecs_to_jiffies(BATADV_BLA_PERIOD_LENGTH));
@@ -1395,7 +1395,7 @@ int batadv_bla_init(struct batadv_priv *bat_priv)
 	if (primary_if) {
 		crc = crc16(0, primary_if->net_dev->dev_addr, ETH_ALEN);
 		bat_priv->bla.claim_dest.group = htons(crc);
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	} else {
 		bat_priv->bla.claim_dest.group = 0; /* will be set later */
 	}
@@ -1599,7 +1599,7 @@ void batadv_bla_free(struct batadv_priv *bat_priv)
 		bat_priv->bla.backbone_hash = NULL;
 	}
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 /**
@@ -1692,7 +1692,7 @@ handled:
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (claim)
 		batadv_claim_free_ref(claim);
 	return ret;
@@ -1781,7 +1781,7 @@ handled:
 	ret = 1;
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (claim)
 		batadv_claim_free_ref(claim);
 	return ret;
@@ -1839,7 +1839,7 @@ int batadv_bla_claim_table_seq_print_text(struct seq_file *seq, void *offset)
 	}
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
 
@@ -1904,6 +1904,6 @@ int batadv_bla_backbone_table_seq_print_text(struct seq_file *seq, void *offset)
 	}
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 8f466e7..00408d6 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -840,7 +840,7 @@ int batadv_dat_cache_seq_print_text(struct seq_file *seq, void *offset)
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
 
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index 67cb224..d13c5b6 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -512,7 +512,7 @@ bool batadv_frag_send_packet(struct sk_buff *skb,
 
 out_err:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 
 	return ret;
 }
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 9cb098f..7bf84c1 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -663,7 +663,7 @@ int batadv_gw_client_seq_print_text(struct seq_file *seq, void *offset)
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
 
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index e2aaa4c..b22b277 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -226,7 +226,7 @@ static void batadv_primary_if_update_addr(struct batadv_priv *bat_priv,
 	batadv_bla_update_orig_address(bat_priv, primary_if, oldif);
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 static void batadv_primary_if_select(struct batadv_priv *bat_priv,
@@ -250,7 +250,7 @@ static void batadv_primary_if_select(struct batadv_priv *bat_priv,
 
 out:
 	if (curr_hard_iface)
-		batadv_hardif_free_ref(curr_hard_iface);
+		batadv_hardif_put(curr_hard_iface);
 }
 
 static bool
@@ -409,7 +409,7 @@ batadv_hardif_activate_interface(struct batadv_hard_iface *hard_iface)
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 static void
@@ -562,7 +562,7 @@ err_dev:
 	hard_iface->soft_iface = NULL;
 	dev_put(soft_iface);
 err:
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 	return ret;
 }
 
@@ -593,7 +593,7 @@ void batadv_hardif_disable_interface(struct batadv_hard_iface *hard_iface,
 		batadv_primary_if_select(bat_priv, new_if);
 
 		if (new_if)
-			batadv_hardif_free_ref(new_if);
+			batadv_hardif_put(new_if);
 	}
 
 	bat_priv->bat_algo_ops->bat_iface_disable(hard_iface);
@@ -616,11 +616,11 @@ void batadv_hardif_disable_interface(struct batadv_hard_iface *hard_iface,
 	}
 
 	hard_iface->soft_iface = NULL;
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 /**
@@ -639,7 +639,7 @@ static void batadv_hardif_remove_interface_finish(struct work_struct *work)
 
 	batadv_debugfs_del_hardif(hard_iface);
 	batadv_sysfs_del_hardif(&hard_iface->hardif_obj);
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 }
 
 static struct batadv_hard_iface *
@@ -794,10 +794,10 @@ static int batadv_hard_if_event(struct notifier_block *this,
 	}
 
 hardif_put:
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return NOTIFY_DONE;
 }
 
diff --git a/net/batman-adv/hard-interface.h b/net/batman-adv/hard-interface.h
index 5cecc6b..d74f198 100644
--- a/net/batman-adv/hard-interface.h
+++ b/net/batman-adv/hard-interface.h
@@ -64,11 +64,11 @@ void batadv_update_min_mtu(struct net_device *soft_iface);
 void batadv_hardif_release(struct kref *ref);
 
 /**
- * batadv_hardif_free_ref - decrement the hard interface refcounter and possibly
+ * batadv_hardif_put - decrement the hard interface refcounter and possibly
  *  release it
  * @hard_iface: the hard interface to free
  */
-static inline void batadv_hardif_free_ref(struct batadv_hard_iface *hard_iface)
+static inline void batadv_hardif_put(struct batadv_hard_iface *hard_iface)
 {
 	kref_put(&hard_iface->refcount, batadv_hardif_release);
 }
diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c
index 7d944af..292a648 100644
--- a/net/batman-adv/icmp_socket.c
+++ b/net/batman-adv/icmp_socket.c
@@ -288,7 +288,7 @@ free_skb:
 	kfree_skb(skb);
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (neigh_node)
 		batadv_neigh_node_free_ref(neigh_node);
 	if (orig_node)
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 5b573d1..476554b 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -287,7 +287,7 @@ batadv_seq_print_text_primary_if_get(struct seq_file *seq)
 	seq_printf(seq,
 		   "BATMAN mesh %s disabled - primary interface not active\n",
 		   net_dev->name);
-	batadv_hardif_free_ref(primary_if);
+	batadv_hardif_put(primary_if);
 	primary_if = NULL;
 
 out:
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 00f74d4..3a42e26 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -1950,7 +1950,7 @@ int batadv_nc_nodes_seq_print_text(struct seq_file *seq, void *offset)
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
 
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 4df9be4..3e959f4 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -196,7 +196,7 @@ static void batadv_neigh_ifinfo_release(struct kref *ref)
 	neigh_ifinfo = container_of(ref, struct batadv_neigh_ifinfo, refcount);
 
 	if (neigh_ifinfo->if_outgoing != BATADV_IF_DEFAULT)
-		batadv_hardif_free_ref(neigh_ifinfo->if_outgoing);
+		batadv_hardif_put(neigh_ifinfo->if_outgoing);
 
 	kfree_rcu(neigh_ifinfo, rcu);
 }
@@ -227,7 +227,7 @@ static void batadv_hardif_neigh_release(struct kref *ref)
 	hlist_del_init_rcu(&hardif_neigh->list);
 	spin_unlock_bh(&hardif_neigh->if_incoming->neigh_list_lock);
 
-	batadv_hardif_free_ref(hardif_neigh->if_incoming);
+	batadv_hardif_put(hardif_neigh->if_incoming);
 	kfree_rcu(hardif_neigh, rcu);
 }
 
@@ -273,7 +273,7 @@ static void batadv_neigh_node_release(struct kref *ref)
 	if (bao->bat_neigh_free)
 		bao->bat_neigh_free(neigh_node);
 
-	batadv_hardif_free_ref(neigh_node->if_incoming);
+	batadv_hardif_put(neigh_node->if_incoming);
 
 	kfree_rcu(neigh_node, rcu);
 }
@@ -544,7 +544,7 @@ batadv_hardif_neigh_create(struct batadv_hard_iface *hard_iface,
 
 	hardif_neigh = kzalloc(sizeof(*hardif_neigh), GFP_ATOMIC);
 	if (!hardif_neigh) {
-		batadv_hardif_free_ref(hard_iface);
+		batadv_hardif_put(hard_iface);
 		goto out;
 	}
 
@@ -707,7 +707,7 @@ int batadv_hardif_neigh_seq_print_text(struct seq_file *seq, void *offset)
 		   primary_if->net_dev->dev_addr, net_dev->name,
 		   bat_priv->bat_algo_ops->name);
 
-	batadv_hardif_free_ref(primary_if);
+	batadv_hardif_put(primary_if);
 
 	if (!bat_priv->bat_algo_ops->bat_neigh_print) {
 		seq_puts(seq,
@@ -732,7 +732,7 @@ static void batadv_orig_ifinfo_release(struct kref *ref)
 	orig_ifinfo = container_of(ref, struct batadv_orig_ifinfo, refcount);
 
 	if (orig_ifinfo->if_outgoing != BATADV_IF_DEFAULT)
-		batadv_hardif_free_ref(orig_ifinfo->if_outgoing);
+		batadv_hardif_put(orig_ifinfo->if_outgoing);
 
 	/* this is the last reference to this object */
 	router = rcu_dereference_protected(orig_ifinfo->router, true);
@@ -1250,7 +1250,7 @@ int batadv_orig_seq_print_text(struct seq_file *seq, void *offset)
 		   primary_if->net_dev->dev_addr, net_dev->name,
 		   bat_priv->bat_algo_ops->name);
 
-	batadv_hardif_free_ref(primary_if);
+	batadv_hardif_put(primary_if);
 
 	if (!bat_priv->bat_algo_ops->bat_orig_print) {
 		seq_puts(seq,
@@ -1306,7 +1306,7 @@ int batadv_orig_hardif_seq_print_text(struct seq_file *seq, void *offset)
 
 out:
 	if (hard_iface)
-		batadv_hardif_free_ref(hard_iface);
+		batadv_hardif_put(hard_iface);
 	return 0;
 }
 
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index d2c490d..a14ca42 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -269,7 +269,7 @@ static int batadv_recv_my_icmp_packet(struct batadv_priv *bat_priv,
 	}
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (orig_node)
 		batadv_orig_node_put(orig_node);
 	return ret;
@@ -317,7 +317,7 @@ static int batadv_recv_icmp_ttl_exceeded(struct batadv_priv *bat_priv,
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (orig_node)
 		batadv_orig_node_put(orig_node);
 	return ret;
@@ -702,7 +702,7 @@ batadv_reroute_unicast_packet(struct batadv_priv *bat_priv,
 	ret = true;
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (orig_node)
 		batadv_orig_node_put(orig_node);
 
@@ -808,7 +808,7 @@ static int batadv_check_unicast_ttvn(struct batadv_priv *bat_priv,
 
 	ether_addr_copy(unicast_packet->dest, primary_if->net_dev->dev_addr);
 
-	batadv_hardif_free_ref(primary_if);
+	batadv_hardif_put(primary_if);
 
 	unicast_packet->ttvn = curr_ttvn;
 
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index 261309e..c9cafc0 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -246,7 +246,7 @@ bool batadv_send_skb_prepare_unicast_4addr(struct batadv_priv *bat_priv,
 	ret = true;
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return ret;
 }
 
@@ -409,9 +409,9 @@ static void batadv_forw_packet_free(struct batadv_forw_packet *forw_packet)
 {
 	kfree_skb(forw_packet->skb);
 	if (forw_packet->if_incoming)
-		batadv_hardif_free_ref(forw_packet->if_incoming);
+		batadv_hardif_put(forw_packet->if_incoming);
 	if (forw_packet->if_outgoing)
-		batadv_hardif_free_ref(forw_packet->if_outgoing);
+		batadv_hardif_put(forw_packet->if_outgoing);
 	kfree(forw_packet);
 }
 
@@ -497,7 +497,7 @@ out_and_inc:
 	atomic_inc(&bat_priv->bcast_queue_left);
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return NETDEV_TX_BUSY;
 }
 
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index d4490ff..ca8fa4e2 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -377,7 +377,7 @@ dropped_freed:
 	batadv_inc_counter(bat_priv, BATADV_CNT_TX_DROPPED);
 end:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return NETDEV_TX_OK;
 }
 
@@ -878,7 +878,7 @@ static int batadv_softif_slave_add(struct net_device *dev,
 
 out:
 	if (hard_iface)
-		batadv_hardif_free_ref(hard_iface);
+		batadv_hardif_put(hard_iface);
 	return ret;
 }
 
@@ -905,7 +905,7 @@ static int batadv_softif_slave_del(struct net_device *dev,
 
 out:
 	if (hard_iface)
-		batadv_hardif_free_ref(hard_iface);
+		batadv_hardif_put(hard_iface);
 	return ret;
 }
 
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index ab4382b..0db7591 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -771,7 +771,7 @@ static ssize_t batadv_show_mesh_iface(struct kobject *kobj,
 
 	length = sprintf(buff, "%s\n", ifname);
 
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 
 	return length;
 }
@@ -795,7 +795,7 @@ static ssize_t batadv_store_mesh_iface(struct kobject *kobj,
 	if (strlen(buff) >= IFNAMSIZ) {
 		pr_err("Invalid parameter for 'mesh_iface' setting received: interface name too long '%s'\n",
 		       buff);
-		batadv_hardif_free_ref(hard_iface);
+		batadv_hardif_put(hard_iface);
 		return -EINVAL;
 	}
 
@@ -829,7 +829,7 @@ static ssize_t batadv_store_mesh_iface(struct kobject *kobj,
 unlock:
 	rtnl_unlock();
 out:
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 	return ret;
 }
 
@@ -863,7 +863,7 @@ static ssize_t batadv_show_iface_status(struct kobject *kobj,
 		break;
 	}
 
-	batadv_hardif_free_ref(hard_iface);
+	batadv_hardif_put(hard_iface);
 
 	return length;
 }
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 107532c..c6c6338 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -1060,7 +1060,7 @@ int batadv_tt_local_seq_print_text(struct seq_file *seq, void *offset)
 	}
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
 
@@ -1725,7 +1725,7 @@ int batadv_tt_global_seq_print_text(struct seq_file *seq, void *offset)
 	}
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	return 0;
 }
 
@@ -2638,7 +2638,7 @@ static int batadv_send_tt_request(struct batadv_priv *bat_priv,
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	if (ret && tt_req_node) {
 		spin_lock_bh(&bat_priv->tt.req_list_lock);
 		/* hlist_del_init() verifies tt_req_node still is in the list */
@@ -2895,7 +2895,7 @@ out:
 	if (orig_node)
 		batadv_orig_node_put(orig_node);
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 	kfree(tvlv_tt_data);
 	/* The packet was for this host, so it doesn't need to be re-routed */
 	return true;
@@ -3218,7 +3218,7 @@ static void batadv_send_roam_adv(struct batadv_priv *bat_priv, u8 *client,
 
 out:
 	if (primary_if)
-		batadv_hardif_free_ref(primary_if);
+		batadv_hardif_put(primary_if);
 }
 
 static void batadv_tt_purge(struct work_struct *work)
-- 
cgit v0.10.2


From 25bb250996eea422171ede0ada8814188dda8937 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:11 +0100
Subject: batman-adv: Rename batadv_neigh_node *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index f01ee23..0f8742b 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1005,7 +1005,7 @@ batadv_iv_ogm_orig_update(struct batadv_priv *bat_priv,
 		    tmp_neigh_node->if_incoming == if_incoming &&
 		    kref_get_unless_zero(&tmp_neigh_node->refcount)) {
 			if (WARN(neigh_node, "too many matching neigh_nodes"))
-				batadv_neigh_node_free_ref(neigh_node);
+				batadv_neigh_node_put(neigh_node);
 			neigh_node = tmp_neigh_node;
 			continue;
 		}
@@ -1116,9 +1116,9 @@ unlock:
 	rcu_read_unlock();
 out:
 	if (neigh_node)
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 	if (router)
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 	if (neigh_ifinfo)
 		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
 	if (router_ifinfo)
@@ -1265,7 +1265,7 @@ static int batadv_iv_ogm_calc_tq(struct batadv_orig_node *orig_node,
 
 out:
 	if (neigh_node)
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 	return ret;
 }
 
@@ -1568,11 +1568,11 @@ out:
 	if (router_ifinfo)
 		batadv_neigh_ifinfo_free_ref(router_ifinfo);
 	if (router)
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 	if (router_router)
-		batadv_neigh_node_free_ref(router_router);
+		batadv_neigh_node_put(router_router);
 	if (orig_neigh_router)
-		batadv_neigh_node_free_ref(orig_neigh_router);
+		batadv_neigh_node_put(orig_neigh_router);
 	if (hardif_neigh)
 		batadv_hardif_neigh_free_ref(hardif_neigh);
 
@@ -1868,7 +1868,7 @@ static void batadv_iv_ogm_orig_print(struct batadv_priv *bat_priv,
 			batman_count++;
 
 next:
-			batadv_neigh_node_free_ref(neigh_node);
+			batadv_neigh_node_put(neigh_node);
 			if (n_ifinfo)
 				batadv_neigh_ifinfo_free_ref(n_ifinfo);
 		}
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 00408d6..919f344 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -674,7 +674,7 @@ static bool batadv_dat_send_data(struct batadv_priv *bat_priv,
 			ret = true;
 		}
 free_neigh:
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 free_orig:
 		batadv_orig_node_put(cand[i].orig_node);
 	}
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index d13c5b6..adb9c39 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -387,7 +387,7 @@ out:
 	if (orig_node_dst)
 		batadv_orig_node_put(orig_node_dst);
 	if (neigh_node)
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 	return ret;
 }
 
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 7bf84c1..ea03c05 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -235,7 +235,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 		batadv_gw_node_free_ref(gw_node);
 
 next:
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 		if (router_ifinfo)
 			batadv_neigh_ifinfo_free_ref(router_ifinfo);
 	}
@@ -352,7 +352,7 @@ out:
 	if (next_gw)
 		batadv_gw_node_free_ref(next_gw);
 	if (router)
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 	if (router_ifinfo)
 		batadv_neigh_ifinfo_free_ref(router_ifinfo);
 }
@@ -417,9 +417,9 @@ out:
 	if (curr_gw_orig)
 		batadv_orig_node_put(curr_gw_orig);
 	if (router_gw)
-		batadv_neigh_node_free_ref(router_gw);
+		batadv_neigh_node_put(router_gw);
 	if (router_orig)
-		batadv_neigh_node_free_ref(router_orig);
+		batadv_neigh_node_put(router_orig);
 	if (router_gw_tq)
 		batadv_neigh_ifinfo_free_ref(router_gw_tq);
 	if (router_orig_tq)
@@ -626,7 +626,7 @@ out:
 	if (router_ifinfo)
 		batadv_neigh_ifinfo_free_ref(router_ifinfo);
 	if (router)
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 	return ret;
 }
 
@@ -885,8 +885,8 @@ out:
 	if (gw_node)
 		batadv_gw_node_free_ref(gw_node);
 	if (neigh_old)
-		batadv_neigh_node_free_ref(neigh_old);
+		batadv_neigh_node_put(neigh_old);
 	if (neigh_curr)
-		batadv_neigh_node_free_ref(neigh_curr);
+		batadv_neigh_node_put(neigh_curr);
 	return out_of_range;
 }
diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c
index 292a648..6268f08 100644
--- a/net/batman-adv/icmp_socket.c
+++ b/net/batman-adv/icmp_socket.c
@@ -290,7 +290,7 @@ out:
 	if (primary_if)
 		batadv_hardif_put(primary_if);
 	if (neigh_node)
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 	if (orig_node)
 		batadv_orig_node_put(orig_node);
 	return len;
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 3a42e26..e9409ba 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -1228,9 +1228,9 @@ static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 	res = true;
 out:
 	if (router_neigh)
-		batadv_neigh_node_free_ref(router_neigh);
+		batadv_neigh_node_put(router_neigh);
 	if (router_coding)
-		batadv_neigh_node_free_ref(router_coding);
+		batadv_neigh_node_put(router_coding);
 	if (router_neigh_ifinfo)
 		batadv_neigh_ifinfo_free_ref(router_neigh_ifinfo);
 	if (router_coding_ifinfo)
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 3e959f4..060f3a6 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -279,11 +279,11 @@ static void batadv_neigh_node_release(struct kref *ref)
 }
 
 /**
- * batadv_neigh_node_free_ref - decrement the neighbors refcounter and possibly
+ * batadv_neigh_node_put - decrement the neighbors refcounter and possibly
  *  release it
  * @neigh_node: neigh neighbor to free
  */
-void batadv_neigh_node_free_ref(struct batadv_neigh_node *neigh_node)
+void batadv_neigh_node_put(struct batadv_neigh_node *neigh_node)
 {
 	kref_put(&neigh_node->refcount, batadv_neigh_node_release);
 }
@@ -737,7 +737,7 @@ static void batadv_orig_ifinfo_release(struct kref *ref)
 	/* this is the last reference to this object */
 	router = rcu_dereference_protected(orig_ifinfo->router, true);
 	if (router)
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 
 	kfree_rcu(orig_ifinfo, rcu);
 }
@@ -793,7 +793,7 @@ static void batadv_orig_node_release(struct kref *ref)
 	hlist_for_each_entry_safe(neigh_node, node_tmp,
 				  &orig_node->neigh_list, list) {
 		hlist_del_rcu(&neigh_node->list);
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 	}
 
 	hlist_for_each_entry_safe(orig_ifinfo, node_tmp,
@@ -1069,7 +1069,7 @@ batadv_purge_orig_neighbors(struct batadv_priv *bat_priv,
 			neigh_purged = true;
 
 			hlist_del_rcu(&neigh_node->list);
-			batadv_neigh_node_free_ref(neigh_node);
+			batadv_neigh_node_put(neigh_node);
 		} else {
 			/* only necessary if not the whole neighbor is to be
 			 * deleted, but some interface has been removed.
@@ -1108,7 +1108,7 @@ batadv_find_best_neighbor(struct batadv_priv *bat_priv,
 			continue;
 
 		if (best)
-			batadv_neigh_node_free_ref(best);
+			batadv_neigh_node_put(best);
 
 		best = neigh;
 	}
@@ -1154,7 +1154,7 @@ static bool batadv_purge_orig_node(struct batadv_priv *bat_priv,
 	batadv_update_route(bat_priv, orig_node, BATADV_IF_DEFAULT,
 			    best_neigh_node);
 	if (best_neigh_node)
-		batadv_neigh_node_free_ref(best_neigh_node);
+		batadv_neigh_node_put(best_neigh_node);
 
 	/* ... then for all other interfaces. */
 	rcu_read_lock();
@@ -1171,7 +1171,7 @@ static bool batadv_purge_orig_node(struct batadv_priv *bat_priv,
 		batadv_update_route(bat_priv, orig_node, hard_iface,
 				    best_neigh_node);
 		if (best_neigh_node)
-			batadv_neigh_node_free_ref(best_neigh_node);
+			batadv_neigh_node_put(best_neigh_node);
 	}
 	rcu_read_unlock();
 
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index c151f6a..78f409c 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -49,7 +49,7 @@ struct batadv_neigh_node *
 batadv_neigh_node_new(struct batadv_orig_node *orig_node,
 		      struct batadv_hard_iface *hard_iface,
 		      const u8 *neigh_addr);
-void batadv_neigh_node_free_ref(struct batadv_neigh_node *neigh_node);
+void batadv_neigh_node_put(struct batadv_neigh_node *neigh_node);
 struct batadv_neigh_node *
 batadv_orig_router_get(struct batadv_orig_node *orig_node,
 		       const struct batadv_hard_iface *if_outgoing);
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index a14ca42..9107f69 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -98,7 +98,7 @@ static void _batadv_update_route(struct batadv_priv *bat_priv,
 	}
 
 	if (curr_router)
-		batadv_neigh_node_free_ref(curr_router);
+		batadv_neigh_node_put(curr_router);
 
 	/* increase refcount of new best neighbor */
 	if (neigh_node && !kref_get_unless_zero(&neigh_node->refcount))
@@ -111,7 +111,7 @@ static void _batadv_update_route(struct batadv_priv *bat_priv,
 
 	/* decrease refcount of previous best neighbor */
 	if (curr_router)
-		batadv_neigh_node_free_ref(curr_router);
+		batadv_neigh_node_put(curr_router);
 }
 
 /**
@@ -138,7 +138,7 @@ void batadv_update_route(struct batadv_priv *bat_priv,
 
 out:
 	if (router)
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 }
 
 /**
@@ -545,7 +545,7 @@ batadv_find_router(struct batadv_priv *bat_priv,
 next:
 		/* free references */
 		if (cand_router) {
-			batadv_neigh_node_free_ref(cand_router);
+			batadv_neigh_node_put(cand_router);
 			cand_router = NULL;
 		}
 		batadv_orig_ifinfo_free_ref(cand);
@@ -562,17 +562,17 @@ next:
 	 * 3) there is no candidate at all, return the default router
 	 */
 	if (next_candidate) {
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 
 		/* remove references to first candidate, we don't need it. */
 		if (first_candidate) {
-			batadv_neigh_node_free_ref(first_candidate_router);
+			batadv_neigh_node_put(first_candidate_router);
 			batadv_orig_ifinfo_free_ref(first_candidate);
 		}
 		router = next_candidate_router;
 		orig_node->last_bonding_candidate = next_candidate;
 	} else if (first_candidate) {
-		batadv_neigh_node_free_ref(router);
+		batadv_neigh_node_put(router);
 
 		/* refcounting has already been done in the loop above. */
 		router = first_candidate_router;
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index c9cafc0..caff32c 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -153,7 +153,7 @@ int batadv_send_skb_to_orig(struct sk_buff *skb,
 
 out:
 	if (neigh_node)
-		batadv_neigh_node_free_ref(neigh_node);
+		batadv_neigh_node_put(neigh_node);
 
 	return ret;
 }
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index c6c6338..132635a 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -1586,20 +1586,20 @@ batadv_transtable_best_orig(struct batadv_priv *bat_priv,
 		if (best_router &&
 		    bao->bat_neigh_cmp(router, BATADV_IF_DEFAULT,
 				       best_router, BATADV_IF_DEFAULT) <= 0) {
-			batadv_neigh_node_free_ref(router);
+			batadv_neigh_node_put(router);
 			continue;
 		}
 
 		/* release the refcount for the "old" best */
 		if (best_router)
-			batadv_neigh_node_free_ref(best_router);
+			batadv_neigh_node_put(best_router);
 
 		best_entry = orig_entry;
 		best_router = router;
 	}
 
 	if (best_router)
-		batadv_neigh_node_free_ref(best_router);
+		batadv_neigh_node_put(best_router);
 
 	return best_entry;
 }
-- 
cgit v0.10.2


From 044fa3ae124118c62575566cfc88e6046a64ae0e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:12 +0100
Subject: batman-adv: Rename batadv_neigh_ifinfo *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 0f8742b..7566cf6 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1026,7 +1026,7 @@ batadv_iv_ogm_orig_update(struct batadv_priv *bat_priv,
 		neigh_ifinfo->bat_iv.tq_avg = tq_avg;
 		spin_unlock_bh(&tmp_neigh_node->ifinfo_lock);
 
-		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
 		neigh_ifinfo = NULL;
 	}
 
@@ -1120,9 +1120,9 @@ out:
 	if (router)
 		batadv_neigh_node_put(router);
 	if (neigh_ifinfo)
-		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
 	if (router_ifinfo)
-		batadv_neigh_ifinfo_free_ref(router_ifinfo);
+		batadv_neigh_ifinfo_put(router_ifinfo);
 }
 
 /**
@@ -1192,7 +1192,7 @@ static int batadv_iv_ogm_calc_tq(struct batadv_orig_node *orig_node,
 	neigh_ifinfo = batadv_neigh_ifinfo_new(neigh_node, if_outgoing);
 	if (neigh_ifinfo) {
 		neigh_rq_count = neigh_ifinfo->bat_iv.real_packet_count;
-		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
 	} else {
 		neigh_rq_count = 0;
 	}
@@ -1353,7 +1353,7 @@ batadv_iv_ogm_update_seqnos(const struct ethhdr *ethhdr,
 		packet_count = bitmap_weight(bitmap,
 					     BATADV_TQ_LOCAL_WINDOW_SIZE);
 		neigh_ifinfo->bat_iv.real_packet_count = packet_count;
-		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
 	}
 	rcu_read_unlock();
 
@@ -1566,7 +1566,7 @@ out_neigh:
 		batadv_orig_node_put(orig_neigh_node);
 out:
 	if (router_ifinfo)
-		batadv_neigh_ifinfo_free_ref(router_ifinfo);
+		batadv_neigh_ifinfo_put(router_ifinfo);
 	if (router)
 		batadv_neigh_node_put(router);
 	if (router_router)
@@ -1805,7 +1805,7 @@ batadv_iv_ogm_orig_print_neigh(struct batadv_orig_node *orig_node,
 			   neigh_node->addr,
 			   n_ifinfo->bat_iv.tq_avg);
 
-		batadv_neigh_ifinfo_free_ref(n_ifinfo);
+		batadv_neigh_ifinfo_put(n_ifinfo);
 	}
 }
 
@@ -1870,7 +1870,7 @@ static void batadv_iv_ogm_orig_print(struct batadv_priv *bat_priv,
 next:
 			batadv_neigh_node_put(neigh_node);
 			if (n_ifinfo)
-				batadv_neigh_ifinfo_free_ref(n_ifinfo);
+				batadv_neigh_ifinfo_put(n_ifinfo);
 		}
 		rcu_read_unlock();
 	}
@@ -1964,9 +1964,9 @@ static int batadv_iv_ogm_neigh_cmp(struct batadv_neigh_node *neigh1,
 
 out:
 	if (neigh1_ifinfo)
-		batadv_neigh_ifinfo_free_ref(neigh1_ifinfo);
+		batadv_neigh_ifinfo_put(neigh1_ifinfo);
 	if (neigh2_ifinfo)
-		batadv_neigh_ifinfo_free_ref(neigh2_ifinfo);
+		batadv_neigh_ifinfo_put(neigh2_ifinfo);
 
 	return diff;
 }
@@ -2007,9 +2007,9 @@ batadv_iv_ogm_neigh_is_sob(struct batadv_neigh_node *neigh1,
 
 out:
 	if (neigh1_ifinfo)
-		batadv_neigh_ifinfo_free_ref(neigh1_ifinfo);
+		batadv_neigh_ifinfo_put(neigh1_ifinfo);
 	if (neigh2_ifinfo)
-		batadv_neigh_ifinfo_free_ref(neigh2_ifinfo);
+		batadv_neigh_ifinfo_put(neigh2_ifinfo);
 
 	return ret;
 }
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index ea03c05..37972df 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -237,7 +237,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 next:
 		batadv_neigh_node_put(router);
 		if (router_ifinfo)
-			batadv_neigh_ifinfo_free_ref(router_ifinfo);
+			batadv_neigh_ifinfo_put(router_ifinfo);
 	}
 	rcu_read_unlock();
 
@@ -354,7 +354,7 @@ out:
 	if (router)
 		batadv_neigh_node_put(router);
 	if (router_ifinfo)
-		batadv_neigh_ifinfo_free_ref(router_ifinfo);
+		batadv_neigh_ifinfo_put(router_ifinfo);
 }
 
 void batadv_gw_check_election(struct batadv_priv *bat_priv,
@@ -421,9 +421,9 @@ out:
 	if (router_orig)
 		batadv_neigh_node_put(router_orig);
 	if (router_gw_tq)
-		batadv_neigh_ifinfo_free_ref(router_gw_tq);
+		batadv_neigh_ifinfo_put(router_gw_tq);
 	if (router_orig_tq)
-		batadv_neigh_ifinfo_free_ref(router_orig_tq);
+		batadv_neigh_ifinfo_put(router_orig_tq);
 }
 
 /**
@@ -624,7 +624,7 @@ static int batadv_write_buffer_text(struct batadv_priv *bat_priv,
 		batadv_gw_node_free_ref(curr_gw);
 out:
 	if (router_ifinfo)
-		batadv_neigh_ifinfo_free_ref(router_ifinfo);
+		batadv_neigh_ifinfo_put(router_ifinfo);
 	if (router)
 		batadv_neigh_node_put(router);
 	return ret;
@@ -857,7 +857,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
 			goto out;
 
 		curr_tq_avg = curr_ifinfo->bat_iv.tq_avg;
-		batadv_neigh_ifinfo_free_ref(curr_ifinfo);
+		batadv_neigh_ifinfo_put(curr_ifinfo);
 
 		break;
 	case BATADV_GW_MODE_OFF:
@@ -875,7 +875,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv,
 
 	if ((curr_tq_avg - old_ifinfo->bat_iv.tq_avg) > BATADV_GW_THRESHOLD)
 		out_of_range = true;
-	batadv_neigh_ifinfo_free_ref(old_ifinfo);
+	batadv_neigh_ifinfo_put(old_ifinfo);
 
 out:
 	if (orig_dst_node)
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index e9409ba..8b36765 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -1232,9 +1232,9 @@ out:
 	if (router_coding)
 		batadv_neigh_node_put(router_coding);
 	if (router_neigh_ifinfo)
-		batadv_neigh_ifinfo_free_ref(router_neigh_ifinfo);
+		batadv_neigh_ifinfo_put(router_neigh_ifinfo);
 	if (router_coding_ifinfo)
-		batadv_neigh_ifinfo_free_ref(router_coding_ifinfo);
+		batadv_neigh_ifinfo_put(router_coding_ifinfo);
 	return res;
 }
 
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 060f3a6..7965fe5 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -202,11 +202,11 @@ static void batadv_neigh_ifinfo_release(struct kref *ref)
 }
 
 /**
- * batadv_neigh_ifinfo_free_ref - decrement the refcounter and possibly release
+ * batadv_neigh_ifinfo_put - decrement the refcounter and possibly release
  *  the neigh_ifinfo
  * @neigh_ifinfo: the neigh_ifinfo object to release
  */
-void batadv_neigh_ifinfo_free_ref(struct batadv_neigh_ifinfo *neigh_ifinfo)
+void batadv_neigh_ifinfo_put(struct batadv_neigh_ifinfo *neigh_ifinfo)
 {
 	kref_put(&neigh_ifinfo->refcount, batadv_neigh_ifinfo_release);
 }
@@ -259,7 +259,7 @@ static void batadv_neigh_node_release(struct kref *ref)
 
 	hlist_for_each_entry_safe(neigh_ifinfo, node_tmp,
 				  &neigh_node->ifinfo_list, list) {
-		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
 	}
 
 	hardif_neigh = batadv_hardif_neigh_get(neigh_node->if_incoming,
@@ -966,7 +966,7 @@ batadv_purge_neigh_ifinfo(struct batadv_priv *bat_priv,
 			   neigh->addr, if_outgoing->net_dev->name);
 
 		hlist_del_rcu(&neigh_ifinfo->list);
-		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
 	}
 
 	spin_unlock_bh(&neigh->ifinfo_lock);
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 78f409c..b4fe828 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -59,7 +59,7 @@ batadv_neigh_ifinfo_new(struct batadv_neigh_node *neigh,
 struct batadv_neigh_ifinfo *
 batadv_neigh_ifinfo_get(struct batadv_neigh_node *neigh,
 			struct batadv_hard_iface *if_outgoing);
-void batadv_neigh_ifinfo_free_ref(struct batadv_neigh_ifinfo *neigh_ifinfo);
+void batadv_neigh_ifinfo_put(struct batadv_neigh_ifinfo *neigh_ifinfo);
 
 int batadv_hardif_neigh_seq_print_text(struct seq_file *seq, void *offset);
 
-- 
cgit v0.10.2


From 35f94779c96b4f79e2fd3f2e59b5713704171c3f Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:13 +0100
Subject: batman-adv: Rename batadv_orig_ifinfo *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 7566cf6..ba48fb2 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1368,7 +1368,7 @@ batadv_iv_ogm_update_seqnos(const struct ethhdr *ethhdr,
 out:
 	spin_unlock_bh(&orig_node->bat_iv.ogm_cnt_lock);
 	batadv_orig_node_put(orig_node);
-	batadv_orig_ifinfo_free_ref(orig_ifinfo);
+	batadv_orig_ifinfo_put(orig_ifinfo);
 	return ret;
 }
 
@@ -1514,7 +1514,7 @@ batadv_iv_ogm_process_per_outif(const struct sk_buff *skb, int ogm_offset,
 					  ogm_packet, if_incoming,
 					  if_outgoing, dup_status);
 	}
-	batadv_orig_ifinfo_free_ref(orig_ifinfo);
+	batadv_orig_ifinfo_put(orig_ifinfo);
 
 	/* only forward for specific interface, not for the default one. */
 	if (if_outgoing == BATADV_IF_DEFAULT)
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 8b36765..9180ff1 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -772,7 +772,7 @@ static bool batadv_can_nc_with_orig(struct batadv_priv *bat_priv,
 
 	last_ttl = orig_ifinfo->last_ttl;
 	last_real_seqno = orig_ifinfo->last_real_seqno;
-	batadv_orig_ifinfo_free_ref(orig_ifinfo);
+	batadv_orig_ifinfo_put(orig_ifinfo);
 
 	if (last_real_seqno != ntohl(ogm_packet->seqno))
 		return false;
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 7965fe5..6056415 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -743,11 +743,11 @@ static void batadv_orig_ifinfo_release(struct kref *ref)
 }
 
 /**
- * batadv_orig_ifinfo_free_ref - decrement the refcounter and possibly release
+ * batadv_orig_ifinfo_put - decrement the refcounter and possibly release
  *  the orig_ifinfo
  * @orig_ifinfo: the orig_ifinfo object to release
  */
-void batadv_orig_ifinfo_free_ref(struct batadv_orig_ifinfo *orig_ifinfo)
+void batadv_orig_ifinfo_put(struct batadv_orig_ifinfo *orig_ifinfo)
 {
 	kref_put(&orig_ifinfo->refcount, batadv_orig_ifinfo_release);
 }
@@ -799,7 +799,7 @@ static void batadv_orig_node_release(struct kref *ref)
 	hlist_for_each_entry_safe(orig_ifinfo, node_tmp,
 				  &orig_node->ifinfo_list, list) {
 		hlist_del_rcu(&orig_ifinfo->list);
-		batadv_orig_ifinfo_free_ref(orig_ifinfo);
+		batadv_orig_ifinfo_put(orig_ifinfo);
 	}
 	spin_unlock_bh(&orig_node->neigh_list_lock);
 
@@ -1012,10 +1012,10 @@ batadv_purge_orig_ifinfo(struct batadv_priv *bat_priv,
 		ifinfo_purged = true;
 
 		hlist_del_rcu(&orig_ifinfo->list);
-		batadv_orig_ifinfo_free_ref(orig_ifinfo);
+		batadv_orig_ifinfo_put(orig_ifinfo);
 		if (orig_node->last_bonding_candidate == orig_ifinfo) {
 			orig_node->last_bonding_candidate = NULL;
-			batadv_orig_ifinfo_free_ref(orig_ifinfo);
+			batadv_orig_ifinfo_put(orig_ifinfo);
 		}
 	}
 
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index b4fe828..adbe7aa 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -69,7 +69,7 @@ batadv_orig_ifinfo_get(struct batadv_orig_node *orig_node,
 struct batadv_orig_ifinfo *
 batadv_orig_ifinfo_new(struct batadv_orig_node *orig_node,
 		       struct batadv_hard_iface *if_outgoing);
-void batadv_orig_ifinfo_free_ref(struct batadv_orig_ifinfo *orig_ifinfo);
+void batadv_orig_ifinfo_put(struct batadv_orig_ifinfo *orig_ifinfo);
 
 int batadv_orig_seq_print_text(struct seq_file *seq, void *offset);
 int batadv_orig_hardif_seq_print_text(struct seq_file *seq, void *offset);
diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c
index 9107f69..4dd646a 100644
--- a/net/batman-adv/routing.c
+++ b/net/batman-adv/routing.c
@@ -107,7 +107,7 @@ static void _batadv_update_route(struct batadv_priv *bat_priv,
 	spin_lock_bh(&orig_node->neigh_list_lock);
 	rcu_assign_pointer(orig_ifinfo->router, neigh_node);
 	spin_unlock_bh(&orig_node->neigh_list_lock);
-	batadv_orig_ifinfo_free_ref(orig_ifinfo);
+	batadv_orig_ifinfo_put(orig_ifinfo);
 
 	/* decrease refcount of previous best neighbor */
 	if (curr_router)
@@ -548,13 +548,13 @@ next:
 			batadv_neigh_node_put(cand_router);
 			cand_router = NULL;
 		}
-		batadv_orig_ifinfo_free_ref(cand);
+		batadv_orig_ifinfo_put(cand);
 	}
 	rcu_read_unlock();
 
 	/* last_bonding_candidate is reset below, remove the old reference. */
 	if (orig_node->last_bonding_candidate)
-		batadv_orig_ifinfo_free_ref(orig_node->last_bonding_candidate);
+		batadv_orig_ifinfo_put(orig_node->last_bonding_candidate);
 
 	/* After finding candidates, handle the three cases:
 	 * 1) there is a next candidate, use that
@@ -567,7 +567,7 @@ next:
 		/* remove references to first candidate, we don't need it. */
 		if (first_candidate) {
 			batadv_neigh_node_put(first_candidate_router);
-			batadv_orig_ifinfo_free_ref(first_candidate);
+			batadv_orig_ifinfo_put(first_candidate);
 		}
 		router = next_candidate_router;
 		orig_node->last_bonding_candidate = next_candidate;
-- 
cgit v0.10.2


From accadc35a147a8faa14a55e176cb94986f28888c Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:14 +0100
Subject: batman-adv: Rename batadv_hardif_neigh *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index ba48fb2..5651e33 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1574,7 +1574,7 @@ out:
 	if (orig_neigh_router)
 		batadv_neigh_node_put(orig_neigh_router);
 	if (hardif_neigh)
-		batadv_hardif_neigh_free_ref(hardif_neigh);
+		batadv_hardif_neigh_put(hardif_neigh);
 
 	kfree_skb(skb_priv);
 }
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 6056415..0b1a77d 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -232,11 +232,11 @@ static void batadv_hardif_neigh_release(struct kref *ref)
 }
 
 /**
- * batadv_hardif_neigh_free_ref - decrement the hardif neighbors refcounter
+ * batadv_hardif_neigh_put - decrement the hardif neighbors refcounter
  *  and possibly release it
  * @hardif_neigh: hardif neigh neighbor to free
  */
-void batadv_hardif_neigh_free_ref(struct batadv_hardif_neigh_node *hardif_neigh)
+void batadv_hardif_neigh_put(struct batadv_hardif_neigh_node *hardif_neigh)
 {
 	kref_put(&hardif_neigh->refcount, batadv_hardif_neigh_release);
 }
@@ -266,8 +266,8 @@ static void batadv_neigh_node_release(struct kref *ref)
 					       neigh_node->addr);
 	if (hardif_neigh) {
 		/* batadv_hardif_neigh_get() increases refcount too */
-		batadv_hardif_neigh_free_ref(hardif_neigh);
-		batadv_hardif_neigh_free_ref(hardif_neigh);
+		batadv_hardif_neigh_put(hardif_neigh);
+		batadv_hardif_neigh_put(hardif_neigh);
 	}
 
 	if (bao->bat_neigh_free)
@@ -681,7 +681,7 @@ batadv_neigh_node_new(struct batadv_orig_node *orig_node,
 
 out:
 	if (hardif_neigh)
-		batadv_hardif_neigh_free_ref(hardif_neigh);
+		batadv_hardif_neigh_put(hardif_neigh);
 	return neigh_node;
 }
 
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index adbe7aa..97748e8 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -44,7 +44,7 @@ struct batadv_hardif_neigh_node *
 batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface,
 			const u8 *neigh_addr);
 void
-batadv_hardif_neigh_free_ref(struct batadv_hardif_neigh_node *hardif_neigh);
+batadv_hardif_neigh_put(struct batadv_hardif_neigh_node *hardif_neigh);
 struct batadv_neigh_node *
 batadv_neigh_node_new(struct batadv_orig_node *orig_node,
 		      struct batadv_hard_iface *hard_iface,
-- 
cgit v0.10.2


From c8b86c1241e39e6b86ef3a5083b149336942008e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:15 +0100
Subject: batman-adv: Rename batadv_backbone_gw *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index b7167cf..ce2d070 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -159,12 +159,11 @@ static void batadv_backbone_gw_release(struct kref *ref)
 }
 
 /**
- * batadv_backbone_gw_free_ref - decrement the backbone gw refcounter and
- *  possibly release it
+ * batadv_backbone_gw_put - decrement the backbone gw refcounter and possibly
+ *  release it
  * @backbone_gw: backbone gateway to be free'd
  */
-static void
-batadv_backbone_gw_free_ref(struct batadv_bla_backbone_gw *backbone_gw)
+static void batadv_backbone_gw_put(struct batadv_bla_backbone_gw *backbone_gw)
 {
 	kref_put(&backbone_gw->refcount, batadv_backbone_gw_release);
 }
@@ -180,7 +179,7 @@ static void batadv_claim_release(struct kref *ref)
 
 	claim = container_of(ref, struct batadv_bla_claim, refcount);
 
-	batadv_backbone_gw_free_ref(claim->backbone_gw);
+	batadv_backbone_gw_put(claim->backbone_gw);
 	kfree_rcu(claim, rcu);
 }
 
@@ -524,7 +523,7 @@ batadv_bla_update_own_backbone_gw(struct batadv_priv *bat_priv,
 		return;
 
 	backbone_gw->lasttime = jiffies;
-	batadv_backbone_gw_free_ref(backbone_gw);
+	batadv_backbone_gw_put(backbone_gw);
 }
 
 /**
@@ -573,7 +572,7 @@ static void batadv_bla_answer_request(struct batadv_priv *bat_priv,
 
 	/* finally, send an announcement frame */
 	batadv_bla_send_announce(bat_priv, backbone_gw);
-	batadv_backbone_gw_free_ref(backbone_gw);
+	batadv_backbone_gw_put(backbone_gw);
 }
 
 /**
@@ -682,7 +681,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,
 		spin_lock_bh(&claim->backbone_gw->crc_lock);
 		claim->backbone_gw->crc ^= crc16(0, claim->addr, ETH_ALEN);
 		spin_unlock_bh(&claim->backbone_gw->crc_lock);
-		batadv_backbone_gw_free_ref(claim->backbone_gw);
+		batadv_backbone_gw_put(claim->backbone_gw);
 	}
 	/* set (new) backbone gw */
 	kref_get(&backbone_gw->refcount);
@@ -783,7 +782,7 @@ static int batadv_handle_announce(struct batadv_priv *bat_priv, u8 *an_addr,
 		}
 	}
 
-	batadv_backbone_gw_free_ref(backbone_gw);
+	batadv_backbone_gw_put(backbone_gw);
 	return 1;
 }
 
@@ -854,7 +853,7 @@ static int batadv_handle_unclaim(struct batadv_priv *bat_priv,
 		   claim_addr, BATADV_PRINT_VID(vid), backbone_gw->orig);
 
 	batadv_bla_del_claim(bat_priv, claim_addr, vid);
-	batadv_backbone_gw_free_ref(backbone_gw);
+	batadv_backbone_gw_put(backbone_gw);
 	return 1;
 }
 
@@ -891,7 +890,7 @@ static int batadv_handle_claim(struct batadv_priv *bat_priv,
 
 	/* TODO: we could call something like tt_local_del() here. */
 
-	batadv_backbone_gw_free_ref(backbone_gw);
+	batadv_backbone_gw_put(backbone_gw);
 	return 1;
 }
 
@@ -1154,7 +1153,7 @@ purge_now:
 			batadv_bla_del_backbone_claims(backbone_gw);
 
 			hlist_del_rcu(&backbone_gw->hash_entry);
-			batadv_backbone_gw_free_ref(backbone_gw);
+			batadv_backbone_gw_put(backbone_gw);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -1571,7 +1570,7 @@ int batadv_bla_is_backbone_gw(struct sk_buff *skb,
 	if (!backbone_gw)
 		return 0;
 
-	batadv_backbone_gw_free_ref(backbone_gw);
+	batadv_backbone_gw_put(backbone_gw);
 	return 1;
 }
 
-- 
cgit v0.10.2


From 321e3e0884829427b9bcf2c112b0945af3160875 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:16 +0100
Subject: batman-adv: Rename batadv_claim *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c
index ce2d070..0a6c8b8 100644
--- a/net/batman-adv/bridge_loop_avoidance.c
+++ b/net/batman-adv/bridge_loop_avoidance.c
@@ -184,11 +184,11 @@ static void batadv_claim_release(struct kref *ref)
 }
 
 /**
- * batadv_claim_free_ref - decrement the claim refcounter and possibly
+ * batadv_claim_put - decrement the claim refcounter and possibly
  *  release it
  * @claim: claim to be free'd
  */
-static void batadv_claim_free_ref(struct batadv_bla_claim *claim)
+static void batadv_claim_put(struct batadv_bla_claim *claim)
 {
 	kref_put(&claim->refcount, batadv_claim_release);
 }
@@ -304,7 +304,7 @@ batadv_bla_del_backbone_claims(struct batadv_bla_backbone_gw *backbone_gw)
 			if (claim->backbone_gw != backbone_gw)
 				continue;
 
-			batadv_claim_free_ref(claim);
+			batadv_claim_put(claim);
 			hlist_del_rcu(&claim->hash_entry);
 		}
 		spin_unlock_bh(list_lock);
@@ -693,7 +693,7 @@ static void batadv_bla_add_claim(struct batadv_priv *bat_priv,
 	backbone_gw->lasttime = jiffies;
 
 claim_free_ref:
-	batadv_claim_free_ref(claim);
+	batadv_claim_put(claim);
 }
 
 /**
@@ -718,14 +718,14 @@ static void batadv_bla_del_claim(struct batadv_priv *bat_priv,
 
 	batadv_hash_remove(bat_priv->bla.claim_hash, batadv_compare_claim,
 			   batadv_choose_claim, claim);
-	batadv_claim_free_ref(claim); /* reference from the hash is gone */
+	batadv_claim_put(claim); /* reference from the hash is gone */
 
 	spin_lock_bh(&claim->backbone_gw->crc_lock);
 	claim->backbone_gw->crc ^= crc16(0, claim->addr, ETH_ALEN);
 	spin_unlock_bh(&claim->backbone_gw->crc_lock);
 
 	/* don't need the reference from hash_find() anymore */
-	batadv_claim_free_ref(claim);
+	batadv_claim_put(claim);
 }
 
 /**
@@ -1693,7 +1693,7 @@ out:
 	if (primary_if)
 		batadv_hardif_put(primary_if);
 	if (claim)
-		batadv_claim_free_ref(claim);
+		batadv_claim_put(claim);
 	return ret;
 }
 
@@ -1782,7 +1782,7 @@ out:
 	if (primary_if)
 		batadv_hardif_put(primary_if);
 	if (claim)
-		batadv_claim_free_ref(claim);
+		batadv_claim_put(claim);
 	return ret;
 }
 
-- 
cgit v0.10.2


From a6416f9ffcb80a84a3ac8d7eb76b277107cae67c Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:17 +0100
Subject: batman-adv: Rename batadv_dat_entry *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 919f344..4c9b69d 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -77,11 +77,11 @@ static void batadv_dat_entry_release(struct kref *ref)
 }
 
 /**
- * batadv_dat_entry_free_ref - decrement the dat_entry refcounter and possibly
+ * batadv_dat_entry_put - decrement the dat_entry refcounter and possibly
  *  release it
  * @dat_entry: dat_entry to be free'd
  */
-static void batadv_dat_entry_free_ref(struct batadv_dat_entry *dat_entry)
+static void batadv_dat_entry_put(struct batadv_dat_entry *dat_entry)
 {
 	kref_put(&dat_entry->refcount, batadv_dat_entry_release);
 }
@@ -135,7 +135,7 @@ static void __batadv_dat_purge(struct batadv_priv *bat_priv,
 				continue;
 
 			hlist_del_rcu(&dat_entry->hash_entry);
-			batadv_dat_entry_free_ref(dat_entry);
+			batadv_dat_entry_put(dat_entry);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -349,7 +349,7 @@ static void batadv_dat_entry_add(struct batadv_priv *bat_priv, __be32 ip,
 
 	if (unlikely(hash_added != 0)) {
 		/* remove the reference for the hash */
-		batadv_dat_entry_free_ref(dat_entry);
+		batadv_dat_entry_put(dat_entry);
 		goto out;
 	}
 
@@ -358,7 +358,7 @@ static void batadv_dat_entry_add(struct batadv_priv *bat_priv, __be32 ip,
 
 out:
 	if (dat_entry)
-		batadv_dat_entry_free_ref(dat_entry);
+		batadv_dat_entry_put(dat_entry);
 }
 
 #ifdef CONFIG_BATMAN_ADV_DEBUG
@@ -1029,7 +1029,7 @@ bool batadv_dat_snoop_outgoing_arp_request(struct batadv_priv *bat_priv,
 	}
 out:
 	if (dat_entry)
-		batadv_dat_entry_free_ref(dat_entry);
+		batadv_dat_entry_put(dat_entry);
 	return ret;
 }
 
@@ -1109,7 +1109,7 @@ bool batadv_dat_snoop_incoming_arp_request(struct batadv_priv *bat_priv,
 	}
 out:
 	if (dat_entry)
-		batadv_dat_entry_free_ref(dat_entry);
+		batadv_dat_entry_put(dat_entry);
 	if (ret)
 		kfree_skb(skb);
 	return ret;
@@ -1262,6 +1262,6 @@ bool batadv_dat_drop_broadcast_packet(struct batadv_priv *bat_priv,
 
 out:
 	if (dat_entry)
-		batadv_dat_entry_free_ref(dat_entry);
+		batadv_dat_entry_put(dat_entry);
 	return ret;
 }
-- 
cgit v0.10.2


From 3a01743ddf2e64558d87b4af616916e4729e2041 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:18 +0100
Subject: batman-adv: Rename batadv_gw_node *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index 37972df..c59aff5 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -76,11 +76,10 @@ static void batadv_gw_node_release(struct kref *ref)
 }
 
 /**
- * batadv_gw_node_free_ref - decrement the gw_node refcounter and possibly
- *  release it
+ * batadv_gw_node_put - decrement the gw_node refcounter and possibly release it
  * @gw_node: gateway node to free
  */
-static void batadv_gw_node_free_ref(struct batadv_gw_node *gw_node)
+static void batadv_gw_node_put(struct batadv_gw_node *gw_node)
 {
 	kref_put(&gw_node->refcount, batadv_gw_node_release);
 }
@@ -125,7 +124,7 @@ unlock:
 	rcu_read_unlock();
 out:
 	if (gw_node)
-		batadv_gw_node_free_ref(gw_node);
+		batadv_gw_node_put(gw_node);
 	return orig_node;
 }
 
@@ -143,7 +142,7 @@ static void batadv_gw_select(struct batadv_priv *bat_priv,
 	rcu_assign_pointer(bat_priv->gw.curr_gw, new_gw_node);
 
 	if (curr_gw_node)
-		batadv_gw_node_free_ref(curr_gw_node);
+		batadv_gw_node_put(curr_gw_node);
 
 	spin_unlock_bh(&bat_priv->gw.list_lock);
 }
@@ -204,7 +203,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 			    ((tmp_gw_factor == max_gw_factor) &&
 			     (tq_avg > max_tq))) {
 				if (curr_gw)
-					batadv_gw_node_free_ref(curr_gw);
+					batadv_gw_node_put(curr_gw);
 				curr_gw = gw_node;
 				kref_get(&curr_gw->refcount);
 			}
@@ -219,7 +218,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 			  */
 			if (tq_avg > max_tq) {
 				if (curr_gw)
-					batadv_gw_node_free_ref(curr_gw);
+					batadv_gw_node_put(curr_gw);
 				curr_gw = gw_node;
 				kref_get(&curr_gw->refcount);
 			}
@@ -232,7 +231,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv)
 		if (tmp_gw_factor > max_gw_factor)
 			max_gw_factor = tmp_gw_factor;
 
-		batadv_gw_node_free_ref(gw_node);
+		batadv_gw_node_put(gw_node);
 
 next:
 		batadv_neigh_node_put(router);
@@ -273,7 +272,7 @@ void batadv_gw_check_client_stop(struct batadv_priv *bat_priv)
 	 */
 	batadv_throw_uevent(bat_priv, BATADV_UEV_GW, BATADV_UEV_DEL, NULL);
 
-	batadv_gw_node_free_ref(curr_gw);
+	batadv_gw_node_put(curr_gw);
 }
 
 void batadv_gw_election(struct batadv_priv *bat_priv)
@@ -348,9 +347,9 @@ void batadv_gw_election(struct batadv_priv *bat_priv)
 
 out:
 	if (curr_gw)
-		batadv_gw_node_free_ref(curr_gw);
+		batadv_gw_node_put(curr_gw);
 	if (next_gw)
-		batadv_gw_node_free_ref(next_gw);
+		batadv_gw_node_put(next_gw);
 	if (router)
 		batadv_neigh_node_put(router);
 	if (router_ifinfo)
@@ -547,7 +546,7 @@ void batadv_gw_node_update(struct batadv_priv *bat_priv,
 		spin_lock_bh(&bat_priv->gw.list_lock);
 		if (!hlist_unhashed(&gw_node->list)) {
 			hlist_del_init_rcu(&gw_node->list);
-			batadv_gw_node_free_ref(gw_node);
+			batadv_gw_node_put(gw_node);
 		}
 		spin_unlock_bh(&bat_priv->gw.list_lock);
 
@@ -556,12 +555,12 @@ void batadv_gw_node_update(struct batadv_priv *bat_priv,
 			batadv_gw_reselect(bat_priv);
 
 		if (curr_gw)
-			batadv_gw_node_free_ref(curr_gw);
+			batadv_gw_node_put(curr_gw);
 	}
 
 out:
 	if (gw_node)
-		batadv_gw_node_free_ref(gw_node);
+		batadv_gw_node_put(gw_node);
 }
 
 void batadv_gw_node_delete(struct batadv_priv *bat_priv,
@@ -584,7 +583,7 @@ void batadv_gw_node_free(struct batadv_priv *bat_priv)
 	hlist_for_each_entry_safe(gw_node, node_tmp,
 				  &bat_priv->gw.list, list) {
 		hlist_del_init_rcu(&gw_node->list);
-		batadv_gw_node_free_ref(gw_node);
+		batadv_gw_node_put(gw_node);
 	}
 	spin_unlock_bh(&bat_priv->gw.list_lock);
 }
@@ -621,7 +620,7 @@ static int batadv_write_buffer_text(struct batadv_priv *bat_priv,
 	ret = seq_has_overflowed(seq) ? -1 : 0;
 
 	if (curr_gw)
-		batadv_gw_node_free_ref(curr_gw);
+		batadv_gw_node_put(curr_gw);
 out:
 	if (router_ifinfo)
 		batadv_neigh_ifinfo_put(router_ifinfo);
@@ -881,9 +880,9 @@ out:
 	if (orig_dst_node)
 		batadv_orig_node_put(orig_dst_node);
 	if (curr_gw)
-		batadv_gw_node_free_ref(curr_gw);
+		batadv_gw_node_put(curr_gw);
 	if (gw_node)
-		batadv_gw_node_free_ref(gw_node);
+		batadv_gw_node_put(gw_node);
 	if (neigh_old)
 		batadv_neigh_node_put(neigh_old);
 	if (neigh_curr)
-- 
cgit v0.10.2


From ba610043afbcd12a1c34ed57fc0689a29aa35b58 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:19 +0100
Subject: batman-adv: Rename batadv_tvlv_handler *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 476554b..95aaff3 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -638,12 +638,11 @@ static void batadv_tvlv_handler_release(struct kref *ref)
 }
 
 /**
- * batadv_tvlv_handler_free_ref - decrement the tvlv container refcounter and
+ * batadv_tvlv_handler_put - decrement the tvlv container refcounter and
  *  possibly release it
  * @tvlv_handler: the tvlv handler to free
  */
-static void
-batadv_tvlv_handler_free_ref(struct batadv_tvlv_handler *tvlv_handler)
+static void batadv_tvlv_handler_put(struct batadv_tvlv_handler *tvlv_handler)
 {
 	kref_put(&tvlv_handler->refcount, batadv_tvlv_handler_release);
 }
@@ -1031,7 +1030,7 @@ int batadv_tvlv_containers_process(struct batadv_priv *bat_priv,
 						src, dst, tvlv_value,
 						tvlv_value_cont_len);
 		if (tvlv_handler)
-			batadv_tvlv_handler_free_ref(tvlv_handler);
+			batadv_tvlv_handler_put(tvlv_handler);
 		tvlv_value = (u8 *)tvlv_value + tvlv_value_cont_len;
 		tvlv_value_len -= tvlv_value_cont_len;
 	}
@@ -1111,7 +1110,7 @@ void batadv_tvlv_handler_register(struct batadv_priv *bat_priv,
 
 	tvlv_handler = batadv_tvlv_handler_get(bat_priv, type, version);
 	if (tvlv_handler) {
-		batadv_tvlv_handler_free_ref(tvlv_handler);
+		batadv_tvlv_handler_put(tvlv_handler);
 		return;
 	}
 
@@ -1148,11 +1147,11 @@ void batadv_tvlv_handler_unregister(struct batadv_priv *bat_priv,
 	if (!tvlv_handler)
 		return;
 
-	batadv_tvlv_handler_free_ref(tvlv_handler);
+	batadv_tvlv_handler_put(tvlv_handler);
 	spin_lock_bh(&bat_priv->tvlv.handler_list_lock);
 	hlist_del_rcu(&tvlv_handler->list);
 	spin_unlock_bh(&bat_priv->tvlv.handler_list_lock);
-	batadv_tvlv_handler_free_ref(tvlv_handler);
+	batadv_tvlv_handler_put(tvlv_handler);
 }
 
 /**
-- 
cgit v0.10.2


From 4a13147cb5d2cb45984074c1becb164a6e9d465f Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:20 +0100
Subject: batman-adv: Rename batadv_tvlv_container *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 95aaff3..bf8a01b 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -694,11 +694,11 @@ static void batadv_tvlv_container_release(struct kref *ref)
 }
 
 /**
- * batadv_tvlv_container_free_ref - decrement the tvlv container refcounter and
+ * batadv_tvlv_container_put - decrement the tvlv container refcounter and
  *  possibly release it
  * @tvlv: the tvlv container to free
  */
-static void batadv_tvlv_container_free_ref(struct batadv_tvlv_container *tvlv)
+static void batadv_tvlv_container_put(struct batadv_tvlv_container *tvlv)
 {
 	kref_put(&tvlv->refcount, batadv_tvlv_container_release);
 }
@@ -784,8 +784,8 @@ static void batadv_tvlv_container_remove(struct batadv_priv *bat_priv,
 	hlist_del(&tvlv->list);
 
 	/* first call to decrement the counter, second call to free */
-	batadv_tvlv_container_free_ref(tvlv);
-	batadv_tvlv_container_free_ref(tvlv);
+	batadv_tvlv_container_put(tvlv);
+	batadv_tvlv_container_put(tvlv);
 }
 
 /**
-- 
cgit v0.10.2


From 9c3bf08189bd4c0675032f5d3ca9dcb42cbb3ec8 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:21 +0100
Subject: batman-adv: Rename batadv_softif_vlan *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index bf8a01b..e3d7051 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -1261,7 +1261,7 @@ bool batadv_vlan_ap_isola_get(struct batadv_priv *bat_priv, unsigned short vid)
 	vlan = batadv_softif_vlan_get(bat_priv, vid);
 	if (vlan) {
 		ap_isolation_enabled = atomic_read(&vlan->ap_isolation);
-		batadv_softif_vlan_free_ref(vlan);
+		batadv_softif_vlan_put(vlan);
 	}
 
 	return ap_isolation_enabled;
diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c
index ca8fa4e2..0710379 100644
--- a/net/batman-adv/soft-interface.c
+++ b/net/batman-adv/soft-interface.c
@@ -497,11 +497,11 @@ static void batadv_softif_vlan_release(struct kref *ref)
 }
 
 /**
- * batadv_softif_vlan_free_ref - decrease the vlan object refcounter and
+ * batadv_softif_vlan_put - decrease the vlan object refcounter and
  *  possibly release it
  * @vlan: the vlan object to release
  */
-void batadv_softif_vlan_free_ref(struct batadv_softif_vlan *vlan)
+void batadv_softif_vlan_put(struct batadv_softif_vlan *vlan)
 {
 	if (!vlan)
 		return;
@@ -552,7 +552,7 @@ int batadv_softif_create_vlan(struct batadv_priv *bat_priv, unsigned short vid)
 
 	vlan = batadv_softif_vlan_get(bat_priv, vid);
 	if (vlan) {
-		batadv_softif_vlan_free_ref(vlan);
+		batadv_softif_vlan_put(vlan);
 		return -EEXIST;
 	}
 
@@ -601,7 +601,7 @@ static void batadv_softif_destroy_vlan(struct batadv_priv *bat_priv,
 			       vlan->vid, "vlan interface destroyed", false);
 
 	batadv_sysfs_del_vlan(bat_priv, vlan);
-	batadv_softif_vlan_free_ref(vlan);
+	batadv_softif_vlan_put(vlan);
 }
 
 /**
@@ -646,7 +646,7 @@ static int batadv_interface_add_vid(struct net_device *dev, __be16 proto,
 	if (!vlan->kobj) {
 		ret = batadv_sysfs_add_vlan(bat_priv->soft_iface, vlan);
 		if (ret) {
-			batadv_softif_vlan_free_ref(vlan);
+			batadv_softif_vlan_put(vlan);
 			return ret;
 		}
 	}
@@ -693,7 +693,7 @@ static int batadv_interface_kill_vid(struct net_device *dev, __be16 proto,
 	batadv_softif_destroy_vlan(bat_priv, vlan);
 
 	/* finally free the vlan object */
-	batadv_softif_vlan_free_ref(vlan);
+	batadv_softif_vlan_put(vlan);
 
 	return 0;
 }
@@ -749,7 +749,7 @@ static void batadv_softif_destroy_finish(struct work_struct *work)
 	vlan = batadv_softif_vlan_get(bat_priv, BATADV_NO_FLAGS);
 	if (vlan) {
 		batadv_softif_destroy_vlan(bat_priv, vlan);
-		batadv_softif_vlan_free_ref(vlan);
+		batadv_softif_vlan_put(vlan);
 	}
 
 	batadv_sysfs_del_meshif(soft_iface);
diff --git a/net/batman-adv/soft-interface.h b/net/batman-adv/soft-interface.h
index d17cfba..9ae2657 100644
--- a/net/batman-adv/soft-interface.h
+++ b/net/batman-adv/soft-interface.h
@@ -34,7 +34,7 @@ void batadv_softif_destroy_sysfs(struct net_device *soft_iface);
 int batadv_softif_is_valid(const struct net_device *net_dev);
 extern struct rtnl_link_ops batadv_link_ops;
 int batadv_softif_create_vlan(struct batadv_priv *bat_priv, unsigned short vid);
-void batadv_softif_vlan_free_ref(struct batadv_softif_vlan *softif_vlan);
+void batadv_softif_vlan_put(struct batadv_softif_vlan *softif_vlan);
 struct batadv_softif_vlan *batadv_softif_vlan_get(struct batadv_priv *bat_priv,
 						  unsigned short vid);
 
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index 0db7591..4d70d44 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -216,7 +216,7 @@ ssize_t batadv_store_vlan_##_name(struct kobject *kobj,			\
 					      attr, &vlan->_name,	\
 					      bat_priv->soft_iface);	\
 									\
-	batadv_softif_vlan_free_ref(vlan);				\
+	batadv_softif_vlan_put(vlan);					\
 	return res;							\
 }
 
@@ -231,7 +231,7 @@ ssize_t batadv_show_vlan_##_name(struct kobject *kobj,			\
 			     atomic_read(&vlan->_name) == 0 ?		\
 			     "disabled" : "enabled");			\
 									\
-	batadv_softif_vlan_free_ref(vlan);				\
+	batadv_softif_vlan_put(vlan);					\
 	return res;							\
 }
 
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 132635a..b81f8ea 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -301,7 +301,7 @@ static void batadv_tt_local_size_mod(struct batadv_priv *bat_priv,
 
 	atomic_add(v, &vlan->tt.num_entries);
 
-	batadv_softif_vlan_free_ref(vlan);
+	batadv_softif_vlan_put(vlan);
 }
 
 /**
@@ -688,7 +688,7 @@ bool batadv_tt_local_add(struct net_device *soft_iface, const u8 *addr,
 	if (unlikely(hash_added != 0)) {
 		/* remove the reference for the hash */
 		batadv_tt_local_entry_free_ref(tt_local);
-		batadv_softif_vlan_free_ref(vlan);
+		batadv_softif_vlan_put(vlan);
 		goto out;
 	}
 
@@ -1054,7 +1054,7 @@ int batadv_tt_local_seq_print_text(struct seq_file *seq, void *offset)
 				   no_purge ? 0 : last_seen_msecs,
 				   vlan->tt.crc);
 
-			batadv_softif_vlan_free_ref(vlan);
+			batadv_softif_vlan_put(vlan);
 		}
 		rcu_read_unlock();
 	}
@@ -1144,8 +1144,8 @@ u16 batadv_tt_local_remove(struct batadv_priv *bat_priv, const u8 *addr,
 	if (!vlan)
 		goto out;
 
-	batadv_softif_vlan_free_ref(vlan);
-	batadv_softif_vlan_free_ref(vlan);
+	batadv_softif_vlan_put(vlan);
+	batadv_softif_vlan_put(vlan);
 
 out:
 	if (tt_local_entry)
@@ -1245,8 +1245,8 @@ static void batadv_tt_local_table_free(struct batadv_priv *bat_priv)
 			vlan = batadv_softif_vlan_get(bat_priv,
 						      tt_common_entry->vid);
 			if (vlan) {
-				batadv_softif_vlan_free_ref(vlan);
-				batadv_softif_vlan_free_ref(vlan);
+				batadv_softif_vlan_put(vlan);
+				batadv_softif_vlan_put(vlan);
 			}
 
 			batadv_tt_local_entry_free_ref(tt_local);
@@ -3342,8 +3342,8 @@ static void batadv_tt_local_purge_pending_clients(struct batadv_priv *bat_priv)
 			/* decrease the reference held for this vlan */
 			vlan = batadv_softif_vlan_get(bat_priv, tt_common->vid);
 			if (vlan) {
-				batadv_softif_vlan_free_ref(vlan);
-				batadv_softif_vlan_free_ref(vlan);
+				batadv_softif_vlan_put(vlan);
+				batadv_softif_vlan_put(vlan);
 			}
 
 			batadv_tt_local_entry_free_ref(tt_local);
@@ -3429,7 +3429,7 @@ bool batadv_is_ap_isolated(struct batadv_priv *bat_priv, u8 *src, u8 *dst,
 	ret = true;
 
 out:
-	batadv_softif_vlan_free_ref(vlan);
+	batadv_softif_vlan_put(vlan);
 	if (tt_global_entry)
 		batadv_tt_global_entry_free_ref(tt_global_entry);
 	if (tt_local_entry)
-- 
cgit v0.10.2


From 27ad7545bb4fc849f84382717a486438310c6ae4 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:22 +0100
Subject: batman-adv: Rename batadv_nc_node *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 9180ff1..6105cfb 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -223,11 +223,11 @@ static void batadv_nc_node_release(struct kref *ref)
 }
 
 /**
- * batadv_nc_node_free_ref - decrement the nc_node refcounter and possibly
+ * batadv_nc_node_put - decrement the nc_node refcounter and possibly
  *  release it
  * @nc_node: nc_node to be free'd
  */
-static void batadv_nc_node_free_ref(struct batadv_nc_node *nc_node)
+static void batadv_nc_node_put(struct batadv_nc_node *nc_node)
 {
 	kref_put(&nc_node->refcount, batadv_nc_node_release);
 }
@@ -356,7 +356,7 @@ batadv_nc_purge_orig_nc_nodes(struct batadv_priv *bat_priv,
 			   "Removing nc_node %pM -> %pM\n",
 			   nc_node->addr, nc_node->orig_node->orig);
 		list_del_rcu(&nc_node->list);
-		batadv_nc_node_free_ref(nc_node);
+		batadv_nc_node_put(nc_node);
 	}
 	spin_unlock_bh(lock);
 }
@@ -942,9 +942,9 @@ void batadv_nc_update_nc_node(struct batadv_priv *bat_priv,
 
 out:
 	if (in_nc_node)
-		batadv_nc_node_free_ref(in_nc_node);
+		batadv_nc_node_put(in_nc_node);
 	if (out_nc_node)
-		batadv_nc_node_free_ref(out_nc_node);
+		batadv_nc_node_put(out_nc_node);
 }
 
 /**
-- 
cgit v0.10.2


From 5fff28255fa6cadbdc46fa43873de2c7f98b850d Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:23 +0100
Subject: batman-adv: Rename batadv_nc_path *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index 6105cfb..d253bb2 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -247,11 +247,11 @@ static void batadv_nc_path_release(struct kref *ref)
 }
 
 /**
- * batadv_nc_path_free_ref - decrement the nc_path refcounter and possibly
+ * batadv_nc_path_put - decrement the nc_path refcounter and possibly
  *  release it
  * @nc_path: nc_path to be free'd
  */
-static void batadv_nc_path_free_ref(struct batadv_nc_path *nc_path)
+static void batadv_nc_path_put(struct batadv_nc_path *nc_path)
 {
 	kref_put(&nc_path->refcount, batadv_nc_path_release);
 }
@@ -263,7 +263,7 @@ static void batadv_nc_path_free_ref(struct batadv_nc_path *nc_path)
 static void batadv_nc_packet_free(struct batadv_nc_packet *nc_packet)
 {
 	kfree_skb(nc_packet->skb);
-	batadv_nc_path_free_ref(nc_packet->nc_path);
+	batadv_nc_path_put(nc_packet->nc_path);
 	kfree(nc_packet);
 }
 
@@ -467,7 +467,7 @@ static void batadv_nc_purge_paths(struct batadv_priv *bat_priv,
 				   "Remove nc_path %pM -> %pM\n",
 				   nc_path->prev_hop, nc_path->next_hop);
 			hlist_del_rcu(&nc_path->hash_entry);
-			batadv_nc_path_free_ref(nc_path);
+			batadv_nc_path_put(nc_path);
 		}
 		spin_unlock_bh(lock);
 	}
@@ -1555,7 +1555,7 @@ bool batadv_nc_skb_forward(struct sk_buff *skb,
 	return true;
 
 free_nc_path:
-	batadv_nc_path_free_ref(nc_path);
+	batadv_nc_path_put(nc_path);
 out:
 	/* Packet is not consumed */
 	return false;
@@ -1617,7 +1617,7 @@ void batadv_nc_skb_store_for_decoding(struct batadv_priv *bat_priv,
 free_skb:
 	kfree_skb(skb);
 free_nc_path:
-	batadv_nc_path_free_ref(nc_path);
+	batadv_nc_path_put(nc_path);
 out:
 	return;
 }
-- 
cgit v0.10.2


From 21754e2501c1faf9cf530fb69e97f463937f8083 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:24 +0100
Subject: batman-adv: Rename batadv_orig_node_vlan *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index 0b1a77d..e4cbb07 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -151,11 +151,11 @@ static void batadv_orig_node_vlan_release(struct kref *ref)
 }
 
 /**
- * batadv_orig_node_vlan_free_ref - decrement the refcounter and possibly
- *  release the originator-vlan object
+ * batadv_orig_node_vlan_put - decrement the refcounter and possibly release
+ *  the originator-vlan object
  * @orig_vlan: the originator-vlan object to release
  */
-void batadv_orig_node_vlan_free_ref(struct batadv_orig_node_vlan *orig_vlan)
+void batadv_orig_node_vlan_put(struct batadv_orig_node_vlan *orig_vlan)
 {
 	kref_put(&orig_vlan->refcount, batadv_orig_node_vlan_release);
 }
@@ -917,7 +917,7 @@ struct batadv_orig_node *batadv_orig_node_new(struct batadv_priv *bat_priv,
 	 * Immediately release vlan since it is not needed anymore in this
 	 * context
 	 */
-	batadv_orig_node_vlan_free_ref(vlan);
+	batadv_orig_node_vlan_put(vlan);
 
 	for (i = 0; i < BATADV_FRAG_BUFFER_COUNT; i++) {
 		INIT_HLIST_HEAD(&orig_node->fragments[i].head);
diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h
index 97748e8..4e8b67f 100644
--- a/net/batman-adv/originator.h
+++ b/net/batman-adv/originator.h
@@ -83,7 +83,7 @@ batadv_orig_node_vlan_new(struct batadv_orig_node *orig_node,
 struct batadv_orig_node_vlan *
 batadv_orig_node_vlan_get(struct batadv_orig_node *orig_node,
 			  unsigned short vid);
-void batadv_orig_node_vlan_free_ref(struct batadv_orig_node_vlan *orig_vlan);
+void batadv_orig_node_vlan_put(struct batadv_orig_node_vlan *orig_vlan);
 
 /* hashfunction to choose an entry in a hash table of given size
  * hash algorithm from http://en.wikipedia.org/wiki/Hash_table
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index b81f8ea..396b095 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -348,12 +348,12 @@ static void batadv_tt_global_size_mod(struct batadv_orig_node *orig_node,
 		spin_lock_bh(&orig_node->vlan_list_lock);
 		if (!hlist_unhashed(&vlan->list)) {
 			hlist_del_init_rcu(&vlan->list);
-			batadv_orig_node_vlan_free_ref(vlan);
+			batadv_orig_node_vlan_put(vlan);
 		}
 		spin_unlock_bh(&orig_node->vlan_list_lock);
 	}
 
-	batadv_orig_node_vlan_free_ref(vlan);
+	batadv_orig_node_vlan_put(vlan);
 }
 
 /**
@@ -1652,7 +1652,7 @@ batadv_tt_global_print_entry(struct batadv_priv *bat_priv,
 			   ((flags & BATADV_TT_CLIENT_ISOLA) ? 'I' : '.'),
 			   ((flags & BATADV_TT_CLIENT_TEMP) ? 'T' : '.'));
 
-		batadv_orig_node_vlan_free_ref(vlan);
+		batadv_orig_node_vlan_put(vlan);
 	}
 
 print_list:
@@ -1684,7 +1684,7 @@ print_list:
 			   ((flags & BATADV_TT_CLIENT_ISOLA) ? 'I' : '.'),
 			   ((flags & BATADV_TT_CLIENT_TEMP) ? 'T' : '.'));
 
-		batadv_orig_node_vlan_free_ref(vlan);
+		batadv_orig_node_vlan_put(vlan);
 	}
 }
 
@@ -2503,7 +2503,7 @@ static bool batadv_tt_global_check_crc(struct batadv_orig_node *orig_node,
 			return false;
 
 		crc = vlan->tt.crc;
-		batadv_orig_node_vlan_free_ref(vlan);
+		batadv_orig_node_vlan_put(vlan);
 
 		if (crc != ntohl(tt_vlan_tmp->crc))
 			return false;
-- 
cgit v0.10.2


From 95c0db90c791ebb73de89f960d12806f1708b31e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:25 +0100
Subject: batman-adv: Rename batadv_tt_local_entry *_free_ref function to *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 396b095..de3e731 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -219,12 +219,12 @@ static void batadv_tt_local_entry_release(struct kref *ref)
 }
 
 /**
- * batadv_tt_local_entry_free_ref - decrement the tt_local_entry refcounter and
+ * batadv_tt_local_entry_put - decrement the tt_local_entry refcounter and
  *  possibly release it
  * @tt_local_entry: tt_local_entry to be free'd
  */
 static void
-batadv_tt_local_entry_free_ref(struct batadv_tt_local_entry *tt_local_entry)
+batadv_tt_local_entry_put(struct batadv_tt_local_entry *tt_local_entry)
 {
 	kref_put(&tt_local_entry->common.refcount,
 		 batadv_tt_local_entry_release);
@@ -687,7 +687,7 @@ bool batadv_tt_local_add(struct net_device *soft_iface, const u8 *addr,
 
 	if (unlikely(hash_added != 0)) {
 		/* remove the reference for the hash */
-		batadv_tt_local_entry_free_ref(tt_local);
+		batadv_tt_local_entry_put(tt_local);
 		batadv_softif_vlan_put(vlan);
 		goto out;
 	}
@@ -754,7 +754,7 @@ out:
 	if (in_dev)
 		dev_put(in_dev);
 	if (tt_local)
-		batadv_tt_local_entry_free_ref(tt_local);
+		batadv_tt_local_entry_put(tt_local);
 	if (tt_global)
 		batadv_tt_global_entry_free_ref(tt_global);
 	return ret;
@@ -1137,7 +1137,7 @@ u16 batadv_tt_local_remove(struct batadv_priv *bat_priv, const u8 *addr,
 		goto out;
 
 	/* extra call to free the local tt entry */
-	batadv_tt_local_entry_free_ref(tt_local_entry);
+	batadv_tt_local_entry_put(tt_local_entry);
 
 	/* decrease the reference held for this vlan */
 	vlan = batadv_softif_vlan_get(bat_priv, vid);
@@ -1149,7 +1149,7 @@ u16 batadv_tt_local_remove(struct batadv_priv *bat_priv, const u8 *addr,
 
 out:
 	if (tt_local_entry)
-		batadv_tt_local_entry_free_ref(tt_local_entry);
+		batadv_tt_local_entry_put(tt_local_entry);
 
 	return curr_flags;
 }
@@ -1249,7 +1249,7 @@ static void batadv_tt_local_table_free(struct batadv_priv *bat_priv)
 				batadv_softif_vlan_put(vlan);
 			}
 
-			batadv_tt_local_entry_free_ref(tt_local);
+			batadv_tt_local_entry_put(tt_local);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -1555,7 +1555,7 @@ out:
 	if (tt_global_entry)
 		batadv_tt_global_entry_free_ref(tt_global_entry);
 	if (tt_local_entry)
-		batadv_tt_local_entry_free_ref(tt_local_entry);
+		batadv_tt_local_entry_put(tt_local_entry);
 	return ret;
 }
 
@@ -1911,7 +1911,7 @@ out:
 	if (tt_global_entry)
 		batadv_tt_global_entry_free_ref(tt_global_entry);
 	if (local_entry)
-		batadv_tt_local_entry_free_ref(local_entry);
+		batadv_tt_local_entry_put(local_entry);
 }
 
 /**
@@ -2143,7 +2143,7 @@ out:
 	if (tt_global_entry)
 		batadv_tt_global_entry_free_ref(tt_global_entry);
 	if (tt_local_entry)
-		batadv_tt_local_entry_free_ref(tt_local_entry);
+		batadv_tt_local_entry_put(tt_local_entry);
 
 	return orig_node;
 }
@@ -3023,7 +3023,7 @@ bool batadv_is_my_client(struct batadv_priv *bat_priv, const u8 *addr,
 	ret = true;
 out:
 	if (tt_local_entry)
-		batadv_tt_local_entry_free_ref(tt_local_entry);
+		batadv_tt_local_entry_put(tt_local_entry);
 	return ret;
 }
 
@@ -3346,7 +3346,7 @@ static void batadv_tt_local_purge_pending_clients(struct batadv_priv *bat_priv)
 				batadv_softif_vlan_put(vlan);
 			}
 
-			batadv_tt_local_entry_free_ref(tt_local);
+			batadv_tt_local_entry_put(tt_local);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -3433,7 +3433,7 @@ out:
 	if (tt_global_entry)
 		batadv_tt_global_entry_free_ref(tt_global_entry);
 	if (tt_local_entry)
-		batadv_tt_local_entry_free_ref(tt_local_entry);
+		batadv_tt_local_entry_put(tt_local_entry);
 	return ret;
 }
 
@@ -3569,7 +3569,7 @@ bool batadv_tt_local_client_is_roaming(struct batadv_priv *bat_priv,
 		goto out;
 
 	ret = tt_local_entry->common.flags & BATADV_TT_CLIENT_ROAM;
-	batadv_tt_local_entry_free_ref(tt_local_entry);
+	batadv_tt_local_entry_put(tt_local_entry);
 out:
 	return ret;
 }
-- 
cgit v0.10.2


From 5dafd8a6ccf4175ba29f80430915eae2c5558e1e Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:26 +0100
Subject: batman-adv: Rename batadv_tt_global_entry *_free_ref function to
 *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index de3e731..7f65a8f 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -247,12 +247,12 @@ static void batadv_tt_global_entry_release(struct kref *ref)
 }
 
 /**
- * batadv_tt_global_entry_free_ref - decrement the tt_global_entry refcounter
- *  and possibly release it
+ * batadv_tt_global_entry_put - decrement the tt_global_entry refcounter and
+ *  possibly release it
  * @tt_global_entry: tt_global_entry to be free'd
  */
 static void
-batadv_tt_global_entry_free_ref(struct batadv_tt_global_entry *tt_global_entry)
+batadv_tt_global_entry_put(struct batadv_tt_global_entry *tt_global_entry)
 {
 	kref_put(&tt_global_entry->common.refcount,
 		 batadv_tt_global_entry_release);
@@ -278,7 +278,7 @@ int batadv_tt_global_hash_count(struct batadv_priv *bat_priv,
 		return 0;
 
 	count = atomic_read(&tt_global_entry->orig_list_count);
-	batadv_tt_global_entry_free_ref(tt_global_entry);
+	batadv_tt_global_entry_put(tt_global_entry);
 
 	return count;
 }
@@ -561,7 +561,7 @@ static void batadv_tt_global_free(struct batadv_priv *bat_priv,
 
 	batadv_hash_remove(bat_priv->tt.global_hash, batadv_compare_tt,
 			   batadv_choose_tt, &tt_global->common);
-	batadv_tt_global_entry_free_ref(tt_global);
+	batadv_tt_global_entry_put(tt_global);
 }
 
 /**
@@ -756,7 +756,7 @@ out:
 	if (tt_local)
 		batadv_tt_local_entry_put(tt_local);
 	if (tt_global)
-		batadv_tt_global_entry_free_ref(tt_global);
+		batadv_tt_global_entry_put(tt_global);
 	return ret;
 }
 
@@ -1467,7 +1467,7 @@ static bool batadv_tt_global_add(struct batadv_priv *bat_priv,
 
 		if (unlikely(hash_added != 0)) {
 			/* remove the reference for the hash */
-			batadv_tt_global_entry_free_ref(tt_global_entry);
+			batadv_tt_global_entry_put(tt_global_entry);
 			goto out_remove;
 		}
 	} else {
@@ -1553,7 +1553,7 @@ out_remove:
 
 out:
 	if (tt_global_entry)
-		batadv_tt_global_entry_free_ref(tt_global_entry);
+		batadv_tt_global_entry_put(tt_global_entry);
 	if (tt_local_entry)
 		batadv_tt_local_entry_put(tt_local_entry);
 	return ret;
@@ -1909,7 +1909,7 @@ static void batadv_tt_global_del(struct batadv_priv *bat_priv,
 
 out:
 	if (tt_global_entry)
-		batadv_tt_global_entry_free_ref(tt_global_entry);
+		batadv_tt_global_entry_put(tt_global_entry);
 	if (local_entry)
 		batadv_tt_local_entry_put(local_entry);
 }
@@ -1965,7 +1965,7 @@ void batadv_tt_global_del_orig(struct batadv_priv *bat_priv,
 					   tt_global->common.addr,
 					   BATADV_PRINT_VID(vid), message);
 				hlist_del_rcu(&tt_common_entry->hash_entry);
-				batadv_tt_global_entry_free_ref(tt_global);
+				batadv_tt_global_entry_put(tt_global);
 			}
 		}
 		spin_unlock_bh(list_lock);
@@ -2028,7 +2028,7 @@ static void batadv_tt_global_purge(struct batadv_priv *bat_priv)
 
 			hlist_del_rcu(&tt_common->hash_entry);
 
-			batadv_tt_global_entry_free_ref(tt_global);
+			batadv_tt_global_entry_put(tt_global);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -2060,7 +2060,7 @@ static void batadv_tt_global_table_free(struct batadv_priv *bat_priv)
 			tt_global = container_of(tt_common_entry,
 						 struct batadv_tt_global_entry,
 						 common);
-			batadv_tt_global_entry_free_ref(tt_global);
+			batadv_tt_global_entry_put(tt_global);
 		}
 		spin_unlock_bh(list_lock);
 	}
@@ -2141,7 +2141,7 @@ struct batadv_orig_node *batadv_transtable_search(struct batadv_priv *bat_priv,
 
 out:
 	if (tt_global_entry)
-		batadv_tt_global_entry_free_ref(tt_global_entry);
+		batadv_tt_global_entry_put(tt_global_entry);
 	if (tt_local_entry)
 		batadv_tt_local_entry_put(tt_local_entry);
 
@@ -3431,7 +3431,7 @@ bool batadv_is_ap_isolated(struct batadv_priv *bat_priv, u8 *src, u8 *dst,
 out:
 	batadv_softif_vlan_put(vlan);
 	if (tt_global_entry)
-		batadv_tt_global_entry_free_ref(tt_global_entry);
+		batadv_tt_global_entry_put(tt_global_entry);
 	if (tt_local_entry)
 		batadv_tt_local_entry_put(tt_local_entry);
 	return ret;
@@ -3543,7 +3543,7 @@ bool batadv_tt_global_client_is_roaming(struct batadv_priv *bat_priv,
 		goto out;
 
 	ret = tt_global_entry->common.flags & BATADV_TT_CLIENT_ROAM;
-	batadv_tt_global_entry_free_ref(tt_global_entry);
+	batadv_tt_global_entry_put(tt_global_entry);
 out:
 	return ret;
 }
@@ -3863,7 +3863,7 @@ bool batadv_tt_global_is_isolated(struct batadv_priv *bat_priv,
 
 	ret = tt->common.flags & BATADV_TT_CLIENT_ISOLA;
 
-	batadv_tt_global_entry_free_ref(tt);
+	batadv_tt_global_entry_put(tt);
 
 	return ret;
 }
-- 
cgit v0.10.2


From 7e2366c6267d4ea257c4d0deb2fcf0744749de5f Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 17 Jan 2016 11:01:27 +0100
Subject: batman-adv: Rename batadv_tt_orig_list_entry *_free_ref function to
 *_put

The batman-adv source code is the only place in the kernel which uses the
*_free_ref naming scheme for the *_put functions. Changing it to *_put
makes it more consistent and makes it easier to understand the connection
to the *_get functions.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index 7f65a8f..0b43e86 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -397,12 +397,12 @@ static void batadv_tt_orig_list_entry_release(struct kref *ref)
 }
 
 /**
- * batadv_tt_orig_list_entry_free_ref - decrement the tt orig entry refcounter
- *  and possibly release it
+ * batadv_tt_orig_list_entry_put - decrement the tt orig entry refcounter and
+ *  possibly release it
  * @orig_entry: tt orig entry to be free'd
  */
 static void
-batadv_tt_orig_list_entry_free_ref(struct batadv_tt_orig_list_entry *orig_entry)
+batadv_tt_orig_list_entry_put(struct batadv_tt_orig_list_entry *orig_entry)
 {
 	kref_put(&orig_entry->refcount, batadv_tt_orig_list_entry_release);
 }
@@ -1345,7 +1345,7 @@ batadv_tt_global_entry_has_orig(const struct batadv_tt_global_entry *entry,
 	orig_entry = batadv_tt_global_orig_entry_find(entry, orig_node);
 	if (orig_entry) {
 		found = true;
-		batadv_tt_orig_list_entry_free_ref(orig_entry);
+		batadv_tt_orig_list_entry_put(orig_entry);
 	}
 
 	return found;
@@ -1386,7 +1386,7 @@ batadv_tt_global_orig_entry_add(struct batadv_tt_global_entry *tt_global,
 
 out:
 	if (orig_entry)
-		batadv_tt_orig_list_entry_free_ref(orig_entry);
+		batadv_tt_orig_list_entry_put(orig_entry);
 }
 
 /**
@@ -1753,7 +1753,7 @@ _batadv_tt_global_del_orig_entry(struct batadv_tt_global_entry *tt_global_entry,
 	 * being part of a list
 	 */
 	hlist_del_rcu(&orig_entry->list);
-	batadv_tt_orig_list_entry_free_ref(orig_entry);
+	batadv_tt_orig_list_entry_put(orig_entry);
 }
 
 /* deletes the orig list of a tt_global_entry */
-- 
cgit v0.10.2


From a30a9ea6e21b495372aff549f3dfd63198bd1f45 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 23 Feb 2016 12:37:52 +0300
Subject: rocker: fix rocker_world_port_obj_vlan_add()

We were changing return values and accidentally made
rocker_world_port_obj_vlan_add() into a no-op.

Fixes: fccd84d44912 ('rocker: return -EOPNOTSUPP for undefined world ops')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index acafbf8..28b775e 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -1598,7 +1598,6 @@ rocker_world_port_obj_vlan_add(struct rocker_port *rocker_port,
 
 	if (!wops->port_obj_vlan_add)
 		return -EOPNOTSUPP;
-		return 0;
 	return wops->port_obj_vlan_add(rocker_port, vlan, trans);
 }
 
-- 
cgit v0.10.2


From 6d5d2ee63cee7025badda3b74ae2ef7ab097acfa Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Fri, 8 Jan 2016 19:28:58 +0100
Subject: Bluetooth: add LED trigger for indicating HCI is powered up

Add support for LED triggers to the Bluetooth subsystem and add kernel
config symbol BT_LEDS for it.

For now one trigger for indicating "HCI is powered up" is supported.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index d4f82ed..dc71473 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -25,6 +25,7 @@
 #ifndef __HCI_CORE_H
 #define __HCI_CORE_H
 
+#include <linux/leds.h>
 #include <net/bluetooth/hci.h>
 #include <net/bluetooth/hci_sock.h>
 
@@ -396,6 +397,8 @@ struct hci_dev {
 	struct delayed_work	rpa_expired;
 	bdaddr_t		rpa;
 
+	struct led_trigger	*power_led;
+
 	int (*open)(struct hci_dev *hdev);
 	int (*close)(struct hci_dev *hdev);
 	int (*flush)(struct hci_dev *hdev);
diff --git a/net/bluetooth/Kconfig b/net/bluetooth/Kconfig
index 95d1a66..06c31b9 100644
--- a/net/bluetooth/Kconfig
+++ b/net/bluetooth/Kconfig
@@ -69,6 +69,15 @@ config BT_6LOWPAN
 	help
 	  IPv6 compression over Bluetooth Low Energy.
 
+config BT_LEDS
+	bool "Enable LED triggers"
+	depends on BT
+	depends on LEDS_CLASS
+	select LEDS_TRIGGERS
+	help
+	  This option selects a few LED triggers for different
+	  Bluetooth events.
+
 config BT_SELFTEST
 	bool "Bluetooth self testing support"
 	depends on BT && DEBUG_KERNEL
diff --git a/net/bluetooth/Makefile b/net/bluetooth/Makefile
index 2b15ae8..b3ff12e 100644
--- a/net/bluetooth/Makefile
+++ b/net/bluetooth/Makefile
@@ -17,6 +17,7 @@ bluetooth-y := af_bluetooth.o hci_core.o hci_conn.o hci_event.o mgmt.o \
 
 bluetooth-$(CONFIG_BT_BREDR) += sco.o
 bluetooth-$(CONFIG_BT_HS) += a2mp.o amp.o
+bluetooth-$(CONFIG_BT_LEDS) += leds.o
 bluetooth-$(CONFIG_BT_DEBUGFS) += hci_debugfs.o
 bluetooth-$(CONFIG_BT_SELFTEST) += selftest.o
 
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 883c821..88f1ef3 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -40,6 +40,7 @@
 #include "hci_request.h"
 #include "hci_debugfs.h"
 #include "smp.h"
+#include "leds.h"
 
 static void hci_rx_work(struct work_struct *work);
 static void hci_cmd_work(struct work_struct *work);
@@ -1395,6 +1396,7 @@ static int hci_dev_do_open(struct hci_dev *hdev)
 		hci_dev_set_flag(hdev, HCI_RPA_EXPIRED);
 		set_bit(HCI_UP, &hdev->flags);
 		hci_sock_dev_event(hdev, HCI_DEV_UP);
+		hci_leds_update_powered(hdev, true);
 		if (!hci_dev_test_flag(hdev, HCI_SETUP) &&
 		    !hci_dev_test_flag(hdev, HCI_CONFIG) &&
 		    !hci_dev_test_flag(hdev, HCI_UNCONFIGURED) &&
@@ -1532,6 +1534,8 @@ int hci_dev_do_close(struct hci_dev *hdev)
 		return 0;
 	}
 
+	hci_leds_update_powered(hdev, false);
+
 	/* Flush RX and TX works */
 	flush_work(&hdev->tx_work);
 	flush_work(&hdev->rx_work);
@@ -3067,6 +3071,8 @@ int hci_register_dev(struct hci_dev *hdev)
 	if (error < 0)
 		goto err_wqueue;
 
+	hci_leds_init(hdev);
+
 	hdev->rfkill = rfkill_alloc(hdev->name, &hdev->dev,
 				    RFKILL_TYPE_BLUETOOTH, &hci_rfkill_ops,
 				    hdev);
@@ -3128,6 +3134,8 @@ void hci_unregister_dev(struct hci_dev *hdev)
 
 	id = hdev->id;
 
+	hci_leds_exit(hdev);
+
 	write_lock(&hci_dev_list_lock);
 	list_del(&hdev->list);
 	write_unlock(&hci_dev_list_lock);
diff --git a/net/bluetooth/leds.c b/net/bluetooth/leds.c
new file mode 100644
index 0000000..ded7c88
--- /dev/null
+++ b/net/bluetooth/leds.c
@@ -0,0 +1,80 @@
+/*
+ * Copyright 2015, Heiner Kallweit <hkallweit1@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+#include "leds.h"
+
+struct hci_basic_led_trigger {
+	struct led_trigger	led_trigger;
+	struct hci_dev		*hdev;
+};
+
+#define to_hci_basic_led_trigger(arg) container_of(arg, \
+			struct hci_basic_led_trigger, led_trigger)
+
+void hci_leds_update_powered(struct hci_dev *hdev, bool enabled)
+{
+	if (hdev->power_led)
+		led_trigger_event(hdev->power_led,
+				  enabled ? LED_FULL : LED_OFF);
+}
+
+static void power_activate(struct led_classdev *led_cdev)
+{
+	struct hci_basic_led_trigger *htrig;
+	bool powered;
+
+	htrig = to_hci_basic_led_trigger(led_cdev->trigger);
+	powered = test_bit(HCI_UP, &htrig->hdev->flags);
+
+	led_trigger_event(led_cdev->trigger, powered ? LED_FULL : LED_OFF);
+}
+
+static struct led_trigger *led_allocate_basic(struct hci_dev *hdev,
+			void (*activate)(struct led_classdev *led_cdev),
+			const char *name)
+{
+	struct hci_basic_led_trigger *htrig;
+
+	htrig =	devm_kzalloc(&hdev->dev, sizeof(*htrig), GFP_KERNEL);
+	if (!htrig)
+		return NULL;
+
+	htrig->hdev = hdev;
+	htrig->led_trigger.activate = activate;
+	htrig->led_trigger.name = devm_kasprintf(&hdev->dev, GFP_KERNEL,
+						 "%s-%s", hdev->name,
+						 name);
+	if (!htrig->led_trigger.name)
+		goto err_alloc;
+
+	if (led_trigger_register(&htrig->led_trigger))
+		goto err_register;
+
+	return &htrig->led_trigger;
+
+err_register:
+	devm_kfree(&hdev->dev, (void *)htrig->led_trigger.name);
+err_alloc:
+	devm_kfree(&hdev->dev, htrig);
+	return NULL;
+}
+
+void hci_leds_init(struct hci_dev *hdev)
+{
+	/* initialize power_led */
+	hdev->power_led = led_allocate_basic(hdev, power_activate, "power");
+}
+
+void hci_leds_exit(struct hci_dev *hdev)
+{
+	if (hdev->power_led)
+		led_trigger_unregister(hdev->power_led);
+}
diff --git a/net/bluetooth/leds.h b/net/bluetooth/leds.h
new file mode 100644
index 0000000..068261a
--- /dev/null
+++ b/net/bluetooth/leds.h
@@ -0,0 +1,18 @@
+/*
+ * Copyright 2015, Heiner Kallweit <hkallweit1@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#if IS_ENABLED(CONFIG_BT_LEDS)
+void hci_leds_update_powered(struct hci_dev *hdev, bool enabled);
+void hci_leds_init(struct hci_dev *hdev);
+void hci_leds_exit(struct hci_dev *hdev);
+#else
+static inline void hci_leds_update_powered(struct hci_dev *hdev,
+					   bool enabled) {}
+static inline void hci_leds_init(struct hci_dev *hdev) {}
+static inline void hci_leds_exit(struct hci_dev *hdev) {}
+#endif
-- 
cgit v0.10.2


From d2ee9c2ec659cfab715568d2d9837f7ff67402fa Mon Sep 17 00:00:00 2001
From: Bhumika Goyal <bhumirks@gmail.com>
Date: Tue, 12 Jan 2016 22:10:15 +0530
Subject: Bluetooth: ath3k: Fixed a blank line after declaration issue

Fixed a coding style issue. Added a blank link after declaration.

Signed-off-by: Bhumika Goyal <bhumirks@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c
index fa893c3..ebd641b 100644
--- a/drivers/bluetooth/ath3k.c
+++ b/drivers/bluetooth/ath3k.c
@@ -497,6 +497,7 @@ static int ath3k_probe(struct usb_interface *intf,
 	/* match device ID in ath3k blacklist table */
 	if (!id->driver_info) {
 		const struct usb_device_id *match;
+
 		match = usb_match_id(intf, ath3k_blist_tbl);
 		if (match)
 			id = match;
-- 
cgit v0.10.2


From b6e402fc84a76c0238b64de497920aeeaa495026 Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 25 Jan 2016 20:46:21 +0100
Subject: Bluetooth: Use managed version of led_trigger_register in LED trigger

Recently a managed version of led_trigger_register was introduced.
Using devm_led_trigger_register allows to simplify the LED trigger code.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 88f1ef3..9c0a683 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -3134,8 +3134,6 @@ void hci_unregister_dev(struct hci_dev *hdev)
 
 	id = hdev->id;
 
-	hci_leds_exit(hdev);
-
 	write_lock(&hci_dev_list_lock);
 	list_del(&hdev->list);
 	write_unlock(&hci_dev_list_lock);
diff --git a/net/bluetooth/leds.c b/net/bluetooth/leds.c
index ded7c88..8319c84 100644
--- a/net/bluetooth/leds.c
+++ b/net/bluetooth/leds.c
@@ -55,7 +55,7 @@ static struct led_trigger *led_allocate_basic(struct hci_dev *hdev,
 	if (!htrig->led_trigger.name)
 		goto err_alloc;
 
-	if (led_trigger_register(&htrig->led_trigger))
+	if (devm_led_trigger_register(&hdev->dev, &htrig->led_trigger))
 		goto err_register;
 
 	return &htrig->led_trigger;
@@ -72,9 +72,3 @@ void hci_leds_init(struct hci_dev *hdev)
 	/* initialize power_led */
 	hdev->power_led = led_allocate_basic(hdev, power_activate, "power");
 }
-
-void hci_leds_exit(struct hci_dev *hdev)
-{
-	if (hdev->power_led)
-		led_trigger_unregister(hdev->power_led);
-}
diff --git a/net/bluetooth/leds.h b/net/bluetooth/leds.h
index 068261a..a9c4d6e 100644
--- a/net/bluetooth/leds.h
+++ b/net/bluetooth/leds.h
@@ -9,10 +9,8 @@
 #if IS_ENABLED(CONFIG_BT_LEDS)
 void hci_leds_update_powered(struct hci_dev *hdev, bool enabled);
 void hci_leds_init(struct hci_dev *hdev);
-void hci_leds_exit(struct hci_dev *hdev);
 #else
 static inline void hci_leds_update_powered(struct hci_dev *hdev,
 					   bool enabled) {}
 static inline void hci_leds_init(struct hci_dev *hdev) {}
-static inline void hci_leds_exit(struct hci_dev *hdev) {}
 #endif
-- 
cgit v0.10.2


From 2be1149ed40812ab75bc1af67a68f5d09e8be762 Mon Sep 17 00:00:00 2001
From: Anton Protopopov <a.s.protopopov@gmail.com>
Date: Wed, 10 Feb 2016 12:22:54 -0500
Subject: Bluetooth: hci_intel: Fix a wrong comparison

A return value of the intel_wait_booting() function compared with
a constant ETIMEDOUT instead of -ETIMEDOUT.

Signed-off-by: Anton Protopopov <a.s.protopopov@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/hci_intel.c b/drivers/bluetooth/hci_intel.c
index 3d63ea3..91d6051 100644
--- a/drivers/bluetooth/hci_intel.c
+++ b/drivers/bluetooth/hci_intel.c
@@ -488,7 +488,7 @@ static int intel_set_baudrate(struct hci_uart *hu, unsigned int speed)
 	clear_bit(STATE_BOOTING, &intel->flags);
 
 	/* In case of timeout, try to continue anyway */
-	if (err && err != ETIMEDOUT)
+	if (err && err != -ETIMEDOUT)
 		return err;
 
 	bt_dev_info(hdev, "Change controller speed to %d", speed);
@@ -581,7 +581,7 @@ static int intel_setup(struct hci_uart *hu)
 	clear_bit(STATE_BOOTING, &intel->flags);
 
 	/* In case of timeout, try to continue anyway */
-	if (err && err != ETIMEDOUT)
+	if (err && err != -ETIMEDOUT)
 		return err;
 
 	set_bit(STATE_BOOTLOADER, &intel->flags);
-- 
cgit v0.10.2


From 609574eb46335cfac1421a07c0505627cbbab1f0 Mon Sep 17 00:00:00 2001
From: Dmitry Tunin <hanipouspilot@gmail.com>
Date: Wed, 10 Feb 2016 15:33:17 +0300
Subject: Bluetooth: btusb: Add new AR3012 ID 13d3:3395

T: Bus=03 Lev=02 Prnt=02 Port=00 Cnt=01 Dev#= 3 Spd=12 MxCh= 0
D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=13d3 ProdID=3395 Rev=00.01
C: #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb

BugLink: https://bugs.launchpad.net/bugs/1542564

Reported-and-tested-by: Christopher Simerly <kilikopela29@gmail.com>
Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Cc: stable@vger.kernel.org

diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c
index ebd641b..3532a77 100644
--- a/drivers/bluetooth/ath3k.c
+++ b/drivers/bluetooth/ath3k.c
@@ -113,6 +113,7 @@ static const struct usb_device_id ath3k_table[] = {
 	{ USB_DEVICE(0x13d3, 0x3362) },
 	{ USB_DEVICE(0x13d3, 0x3375) },
 	{ USB_DEVICE(0x13d3, 0x3393) },
+	{ USB_DEVICE(0x13d3, 0x3395) },
 	{ USB_DEVICE(0x13d3, 0x3402) },
 	{ USB_DEVICE(0x13d3, 0x3408) },
 	{ USB_DEVICE(0x13d3, 0x3423) },
@@ -175,6 +176,7 @@ static const struct usb_device_id ath3k_blist_tbl[] = {
 	{ USB_DEVICE(0x13d3, 0x3362), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3375), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3393), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x13d3, 0x3395), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3402), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3408), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3423), .driver_info = BTUSB_ATH3012 },
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index a191e31..ab6328c 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -227,6 +227,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x13d3, 0x3362), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3375), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3393), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x13d3, 0x3395), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3402), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3408), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3423), .driver_info = BTUSB_ATH3012 },
-- 
cgit v0.10.2


From b013a636b8571fa62b0fd3abeb2b36f0fc7199dd Mon Sep 17 00:00:00 2001
From: "J.J. Meijer" <jjmeijer88@gmail.com>
Date: Mon, 1 Feb 2016 23:47:55 +0100
Subject: Bluetooth: hci_bcm: Add new ACPI ID for bcm43241

This ACPI ID is used at least by HP for their Omni 10 5600eg tablet.

Signed-off-by: J.J. Meijer <jjmeijer88@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index 5f3de18..7092dab 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -820,6 +820,7 @@ static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E3D", 0 },
 	{ "BCM2E3F", 0 },
 	{ "BCM2E40", 0 },
+	{ "BCM2E54", 0 },
 	{ "BCM2E64", 0 },
 	{ "BCM2E65", 0 },
 	{ "BCM2E67", 0 },
-- 
cgit v0.10.2


From 2791b44d6b4a2280568e0dc84be51992b7b367f1 Mon Sep 17 00:00:00 2001
From: Mika Westerberg <mika.westerberg@linux.intel.com>
Date: Fri, 29 Jan 2016 16:53:27 +0200
Subject: Bluetooth: hci_bcm: Add BCM2E7C ACPI ID

Recent macbooks (early 2015) with BCM43241 use this ACPI ID. Add it to the
list of supported devices.

Reported-by: Leif Liddy <leif.liddy@gmail.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index 7092dab..bb4c5a0 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -825,6 +825,7 @@ static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E65", 0 },
 	{ "BCM2E67", 0 },
 	{ "BCM2E7B", 0 },
+	{ "BCM2E7C", 0 },
 	{ },
 };
 MODULE_DEVICE_TABLE(acpi, bcm_acpi_match);
-- 
cgit v0.10.2


From ad750fa142747f3f42b66ae051a0f275bd2035df Mon Sep 17 00:00:00 2001
From: Petri Gynther <pgynther@google.com>
Date: Wed, 10 Feb 2016 12:00:46 -0800
Subject: Bluetooth: btbcm: Fix handling of firmware not found

If the call to request_firmware() fails in btbcm_setup_patchram(),
the BCM chip will be operating with its default firmware.

In this case, btbcm_setup_patchram() should not return immediately
but instead should skip to btbcm_check_bdaddr() and quirk setup.

Signed-off-by: Petri Gynther <pgynther@google.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c
index 0b69794..fdb4482 100644
--- a/drivers/bluetooth/btbcm.c
+++ b/drivers/bluetooth/btbcm.c
@@ -467,7 +467,7 @@ int btbcm_setup_patchram(struct hci_dev *hdev)
 	err = request_firmware(&fw, fw_name, &hdev->dev);
 	if (err < 0) {
 		BT_INFO("%s: BCM: Patch %s not found", hdev->name, fw_name);
-		return 0;
+		goto done;
 	}
 
 	btbcm_patchram(hdev, fw);
@@ -501,6 +501,7 @@ int btbcm_setup_patchram(struct hci_dev *hdev)
 	BT_INFO("%s: %s", hdev->name, (char *)(skb->data + 1));
 	kfree_skb(skb);
 
+done:
 	btbcm_check_bdaddr(hdev);
 
 	set_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks);
-- 
cgit v0.10.2


From 28c971d82fb58ef7cba22e5308be6d2d2590473d Mon Sep 17 00:00:00 2001
From: Dmitry Tunin <hanipouspilot@gmail.com>
Date: Wed, 10 Feb 2016 00:49:11 +0300
Subject: Bluetooth: Add new AR3012 ID 0489:e095

T: Bus=01 Lev=01 Prnt=01 Port=04 Cnt=02 Dev#= 3 Spd=12 MxCh= 0
D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=0489 ProdID=e095 Rev=00.01
C: #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb

This device requires ar3k/AthrBT_0x31010100.dfu and
ar3k/ramps_0x31010100_40.dfu firmware files that are not in
linux-firmware yet.

BugLink: https://bugs.launchpad.net/bugs/1542944

Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Cc: stable@vger.kernel.org

diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c
index 3532a77..e2ccf90 100644
--- a/drivers/bluetooth/ath3k.c
+++ b/drivers/bluetooth/ath3k.c
@@ -82,6 +82,7 @@ static const struct usb_device_id ath3k_table[] = {
 	{ USB_DEVICE(0x0489, 0xe05f) },
 	{ USB_DEVICE(0x0489, 0xe076) },
 	{ USB_DEVICE(0x0489, 0xe078) },
+	{ USB_DEVICE(0x0489, 0xe095) },
 	{ USB_DEVICE(0x04c5, 0x1330) },
 	{ USB_DEVICE(0x04CA, 0x3004) },
 	{ USB_DEVICE(0x04CA, 0x3005) },
@@ -145,6 +146,7 @@ static const struct usb_device_id ath3k_blist_tbl[] = {
 	{ USB_DEVICE(0x0489, 0xe05f), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0489, 0xe076), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0489, 0xe078), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x0489, 0xe095), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04c5, 0x1330), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3004), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3005), .driver_info = BTUSB_ATH3012 },
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index ab6328c..55fbdfc 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -196,6 +196,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x0489, 0xe05f), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0489, 0xe076), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0489, 0xe078), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x0489, 0xe095), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04c5, 0x1330), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3004), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3005), .driver_info = BTUSB_ATH3012 },
-- 
cgit v0.10.2


From d82142a8b1338e6a4339920863423379c27b0b16 Mon Sep 17 00:00:00 2001
From: Wei-Ning Huang <wnhuang@chromium.org>
Date: Mon, 15 Feb 2016 17:09:51 +0800
Subject: Bluetooth: hci_core: cancel power off delayed work properly

When the HCI_AUTO_OFF flag is cleared, the power_off delayed work need
to be cancel or HCI will be powered off even if it's managed.

Signed-off-by: Wei-Ning Huang <wnhuang@chromium.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 9c0a683..2713fc8 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -2021,6 +2021,7 @@ static void hci_power_on(struct work_struct *work)
 	if (test_bit(HCI_UP, &hdev->flags) &&
 	    hci_dev_test_flag(hdev, HCI_MGMT) &&
 	    hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF)) {
+		cancel_delayed_work(&hdev->power_off);
 		hci_req_sync_lock(hdev);
 		err = __hci_req_hci_power_on(hdev);
 		hci_req_sync_unlock(hdev);
-- 
cgit v0.10.2


From aff3eaa03dc30d7f1ea07ab5931cd61a657c5974 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 19 Feb 2016 09:59:10 +0100
Subject: MAINTAINERS: update 802.15.4 entries

This patch updates my e-mail address and other pending information for
802.15.4 subsystem which are not correct anymore.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/MAINTAINERS b/MAINTAINERS
index 27393cf..a317e28 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -151,7 +151,7 @@ S:	Maintained
 F:	drivers/scsi/53c700*
 
 6LOWPAN GENERIC (BTLE/IEEE 802.15.4)
-M:	Alexander Aring <alex.aring@gmail.com>
+M:	Alexander Aring <aar@pengutronix.de>
 M:	Jukka Rissanen <jukka.rissanen@linux.intel.com>
 L:	linux-bluetooth@vger.kernel.org
 L:	linux-wpan@vger.kernel.org
@@ -5415,10 +5415,11 @@ S:	Supported
 F:	drivers/idle/i7300_idle.c
 
 IEEE 802.15.4 SUBSYSTEM
-M:	Alexander Aring <alex.aring@gmail.com>
+M:	Alexander Aring <aar@pengutronix.de>
 L:	linux-wpan@vger.kernel.org
-W:	https://github.com/linux-wpan
-T:	git git://github.com/linux-wpan/linux-wpan-next.git
+W:	http://wpan.cakelab.org/
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git
 S:	Maintained
 F:	net/ieee802154/
 F:	net/mac802154/
-- 
cgit v0.10.2


From 07b0188adf7298bf80a9890d3e90f27e973623d3 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 19 Feb 2016 09:59:11 +0100
Subject: mac802154: fix mac header length check

I got report about that sometimes the WARN_ON occurs there which should
never happen. I came to the conclusion that the mac header is there but
inside the headroom of skb. The skb->len information doesn't contain the
information about the headroom length and skb->len is lesser than two.

We check now if the skb_mac_header pointer is set and the room between
mac header pointer and tail pointer.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/include/net/mac802154.h b/include/net/mac802154.h
index da574bb..2e3cdd20 100644
--- a/include/net/mac802154.h
+++ b/include/net/mac802154.h
@@ -247,8 +247,9 @@ struct ieee802154_ops {
  */
 static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb)
 {
-	/* return some invalid fc on failure */
-	if (unlikely(skb->len < 2)) {
+	/* check if we can fc at skb_mac_header of sk buffer */
+	if (unlikely(!skb_mac_header_was_set(skb) ||
+		     (skb_tail_pointer(skb) - skb_mac_header(skb)) < 2)) {
 		WARN_ON(1);
 		return cpu_to_le16(0);
 	}
-- 
cgit v0.10.2


From c231c5a47a0c697e7bc821af0b5cb28d129fe8e0 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 19 Feb 2016 09:59:12 +0100
Subject: at86rf230: fix race on error handling

The resource "ctx" can be still used by at86rf230_async_state_change, we
need to free it at the complete handler of the async state change to
avoid a use after free.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/net/ieee802154/at86rf230.c b/drivers/net/ieee802154/at86rf230.c
index 0fbbba7..bf3cfe4 100644
--- a/drivers/net/ieee802154/at86rf230.c
+++ b/drivers/net/ieee802154/at86rf230.c
@@ -343,16 +343,26 @@ static const struct regmap_config at86rf230_regmap_spi_config = {
 };
 
 static void
-at86rf230_async_error_recover(void *context)
+at86rf230_async_error_recover_complete(void *context)
 {
 	struct at86rf230_state_change *ctx = context;
 	struct at86rf230_local *lp = ctx->lp;
 
-	lp->is_tx = 0;
-	at86rf230_async_state_change(lp, ctx, STATE_RX_AACK_ON, NULL);
-	ieee802154_wake_queue(lp->hw);
 	if (ctx->free)
 		kfree(ctx);
+
+	ieee802154_wake_queue(lp->hw);
+}
+
+static void
+at86rf230_async_error_recover(void *context)
+{
+	struct at86rf230_state_change *ctx = context;
+	struct at86rf230_local *lp = ctx->lp;
+
+	lp->is_tx = 0;
+	at86rf230_async_state_change(lp, ctx, STATE_RX_AACK_ON,
+				     at86rf230_async_error_recover_complete);
 }
 
 static inline void
-- 
cgit v0.10.2


From d981b5b5fe8ed0c237b2925ab37a17d28405494f Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 19 Feb 2016 09:59:13 +0100
Subject: at86rf230: fix state change handling on error

This patch force always to set "is_tx_from_off", when calibration
timeout was not occurred. In case of error handling the is_tx_from_off
can be inside in an invalid state.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/net/ieee802154/at86rf230.c b/drivers/net/ieee802154/at86rf230.c
index bf3cfe4..cb9e9fe 100644
--- a/drivers/net/ieee802154/at86rf230.c
+++ b/drivers/net/ieee802154/at86rf230.c
@@ -902,14 +902,12 @@ at86rf230_xmit_start(void *context)
 	struct at86rf230_local *lp = ctx->lp;
 
 	/* check if we change from off state */
-	if (lp->is_tx_from_off) {
-		lp->is_tx_from_off = false;
+	if (lp->is_tx_from_off)
 		at86rf230_async_state_change(lp, ctx, STATE_TX_ARET_ON,
 					     at86rf230_write_frame);
-	} else {
+	else
 		at86rf230_async_state_change(lp, ctx, STATE_TX_ON,
 					     at86rf230_xmit_tx_on);
-	}
 }
 
 static int
@@ -933,6 +931,7 @@ at86rf230_xmit(struct ieee802154_hw *hw, struct sk_buff *skb)
 		at86rf230_async_state_change(lp, ctx, STATE_TRX_OFF,
 					     at86rf230_xmit_start);
 	} else {
+		lp->is_tx_from_off = false;
 		at86rf230_xmit_start(ctx);
 	}
 
-- 
cgit v0.10.2


From 6367551f462a3c1e2f38f5ea7335e49443379ab2 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 19 Feb 2016 09:59:14 +0100
Subject: mrf24j40: add writeable missing reg

This patch adds a missing reg for writeable stuff for regmap.

Cc: Alan Ott <alan@signal11.us>
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/net/ieee802154/mrf24j40.c b/drivers/net/ieee802154/mrf24j40.c
index 4cdf516..764a2bd 100644
--- a/drivers/net/ieee802154/mrf24j40.c
+++ b/drivers/net/ieee802154/mrf24j40.c
@@ -310,6 +310,7 @@ mrf24j40_short_reg_writeable(struct device *dev, unsigned int reg)
 	case REG_TRISGPIO:
 	case REG_GPIO:
 	case REG_RFCTL:
+	case REG_SECCR2:
 	case REG_SLPACK:
 	case REG_BBREG0:
 	case REG_BBREG1:
-- 
cgit v0.10.2


From aef00c15b8c503083a703900a755fdb1cf2436e0 Mon Sep 17 00:00:00 2001
From: Koen Zandberg <koen@bergzand.net>
Date: Wed, 10 Feb 2016 11:49:38 +0100
Subject: mac802154: Fixes kernel oops when unloading a radio driver

Destroying the workqueue before unregistering the net device caused a
kernel oops

Signed-off-by: Koen Zandberg <koen@bergzand.net>
Acked-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/mac802154/main.c b/net/mac802154/main.c
index e8cab5b..87da85a 100644
--- a/net/mac802154/main.c
+++ b/net/mac802154/main.c
@@ -218,7 +218,6 @@ void ieee802154_unregister_hw(struct ieee802154_hw *hw)
 
 	tasklet_kill(&local->tasklet);
 	flush_workqueue(local->workqueue);
-	destroy_workqueue(local->workqueue);
 
 	rtnl_lock();
 
@@ -226,6 +225,7 @@ void ieee802154_unregister_hw(struct ieee802154_hw *hw)
 
 	rtnl_unlock();
 
+	destroy_workqueue(local->workqueue);
 	wpan_phy_unregister(local->phy);
 }
 EXPORT_SYMBOL(ieee802154_unregister_hw);
-- 
cgit v0.10.2


From 5609c185f24dffca5f6a9c127106869da150be03 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Mon, 22 Feb 2016 09:13:54 +0100
Subject: 6lowpan: iphc: add support for stateful compression

This patch introduce support for IPHC stateful address compression. It
will offer the context table via one debugfs entry.
This debugfs has and directory for each cid entry for the context table.
Inside each cid directory there exists the following files:

 - "active": If the entry is added or deleted. The context table is
   original a list implementation, this flag will indicate if the
   context is part of list or not.
 - "prefix": The ipv6 prefix.
 - "prefix_length": The prefix length for the prefix.
 - "compression": The compression flag according RFC6775.

This part should be moved into sysfs after some testing time.

Also the debugfs entry contains a "show" file which is a pretty-printout
for the current context table information.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/include/net/6lowpan.h b/include/net/6lowpan.h
index 2f6a3f2..da3a77d 100644
--- a/include/net/6lowpan.h
+++ b/include/net/6lowpan.h
@@ -75,6 +75,8 @@
 #define LOWPAN_IPHC_MAX_HC_BUF_LEN	(sizeof(struct ipv6hdr) +	\
 					 LOWPAN_IPHC_MAX_HEADER_LEN +	\
 					 LOWPAN_NHC_MAX_HDR_LEN)
+/* SCI/DCI is 4 bit width, so we have maximum 16 entries */
+#define LOWPAN_IPHC_CTX_TABLE_SIZE	(1 << 4)
 
 #define LOWPAN_DISPATCH_IPV6		0x41 /* 01000001 = 65 */
 #define LOWPAN_DISPATCH_IPHC		0x60 /* 011xxxxx = ... */
@@ -98,9 +100,39 @@ enum lowpan_lltypes {
 	LOWPAN_LLTYPE_IEEE802154,
 };
 
+enum lowpan_iphc_ctx_flags {
+	LOWPAN_IPHC_CTX_FLAG_ACTIVE,
+	LOWPAN_IPHC_CTX_FLAG_COMPRESSION,
+};
+
+struct lowpan_iphc_ctx {
+	u8 id;
+	struct in6_addr pfx;
+	u8 plen;
+	unsigned long flags;
+};
+
+struct lowpan_iphc_ctx_table {
+	spinlock_t lock;
+	const struct lowpan_iphc_ctx_ops *ops;
+	struct lowpan_iphc_ctx table[LOWPAN_IPHC_CTX_TABLE_SIZE];
+};
+
+static inline bool lowpan_iphc_ctx_is_active(const struct lowpan_iphc_ctx *ctx)
+{
+	return test_bit(LOWPAN_IPHC_CTX_FLAG_ACTIVE, &ctx->flags);
+}
+
+static inline bool
+lowpan_iphc_ctx_is_compression(const struct lowpan_iphc_ctx *ctx)
+{
+	return test_bit(LOWPAN_IPHC_CTX_FLAG_COMPRESSION, &ctx->flags);
+}
+
 struct lowpan_priv {
 	enum lowpan_lltypes lltype;
 	struct dentry *iface_debugfs;
+	struct lowpan_iphc_ctx_table ctx;
 
 	/* must be last */
 	u8 priv[0] __aligned(sizeof(void *));
diff --git a/net/6lowpan/core.c b/net/6lowpan/core.c
index faf65ba..34e44c0 100644
--- a/net/6lowpan/core.c
+++ b/net/6lowpan/core.c
@@ -20,7 +20,7 @@
 int lowpan_register_netdevice(struct net_device *dev,
 			      enum lowpan_lltypes lltype)
 {
-	int ret;
+	int i, ret;
 
 	dev->addr_len = EUI64_ADDR_LEN;
 	dev->type = ARPHRD_6LOWPAN;
@@ -29,6 +29,10 @@ int lowpan_register_netdevice(struct net_device *dev,
 
 	lowpan_priv(dev)->lltype = lltype;
 
+	spin_lock_init(&lowpan_priv(dev)->ctx.lock);
+	for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++)
+		lowpan_priv(dev)->ctx.table[i].id = i;
+
 	ret = register_netdevice(dev);
 	if (ret < 0)
 		return ret;
@@ -68,6 +72,32 @@ void lowpan_unregister_netdev(struct net_device *dev)
 }
 EXPORT_SYMBOL(lowpan_unregister_netdev);
 
+static int lowpan_event(struct notifier_block *unused,
+			unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	int i;
+
+	if (dev->type != ARPHRD_6LOWPAN)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_DOWN:
+		for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++)
+			clear_bit(LOWPAN_IPHC_CTX_FLAG_ACTIVE,
+				  &lowpan_priv(dev)->ctx.table[i].flags);
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block lowpan_notifier = {
+	.notifier_call = lowpan_event,
+};
+
 static int __init lowpan_module_init(void)
 {
 	int ret;
@@ -76,6 +106,12 @@ static int __init lowpan_module_init(void)
 	if (ret < 0)
 		return ret;
 
+	ret = register_netdevice_notifier(&lowpan_notifier);
+	if (ret < 0) {
+		lowpan_debugfs_exit();
+		return ret;
+	}
+
 	request_module_nowait("ipv6");
 
 	request_module_nowait("nhc_dest");
@@ -92,6 +128,7 @@ static int __init lowpan_module_init(void)
 static void __exit lowpan_module_exit(void)
 {
 	lowpan_debugfs_exit();
+	unregister_netdevice_notifier(&lowpan_notifier);
 }
 
 module_init(lowpan_module_init);
diff --git a/net/6lowpan/debugfs.c b/net/6lowpan/debugfs.c
index 88eef84..aa49ff4 100644
--- a/net/6lowpan/debugfs.c
+++ b/net/6lowpan/debugfs.c
@@ -16,19 +16,266 @@
 
 #include "6lowpan_i.h"
 
+#define LOWPAN_DEBUGFS_CTX_PFX_NUM_ARGS	8
+
 static struct dentry *lowpan_debugfs;
 
+static int lowpan_ctx_flag_active_set(void *data, u64 val)
+{
+	struct lowpan_iphc_ctx *ctx = data;
+
+	if (val != 0 && val != 1)
+		return -EINVAL;
+
+	if (val)
+		set_bit(LOWPAN_IPHC_CTX_FLAG_ACTIVE, &ctx->flags);
+	else
+		clear_bit(LOWPAN_IPHC_CTX_FLAG_ACTIVE, &ctx->flags);
+
+	return 0;
+}
+
+static int lowpan_ctx_flag_active_get(void *data, u64 *val)
+{
+	*val = lowpan_iphc_ctx_is_active(data);
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(lowpan_ctx_flag_active_fops,
+			lowpan_ctx_flag_active_get,
+			lowpan_ctx_flag_active_set, "%llu\n");
+
+static int lowpan_ctx_flag_c_set(void *data, u64 val)
+{
+	struct lowpan_iphc_ctx *ctx = data;
+
+	if (val != 0 && val != 1)
+		return -EINVAL;
+
+	if (val)
+		set_bit(LOWPAN_IPHC_CTX_FLAG_COMPRESSION, &ctx->flags);
+	else
+		clear_bit(LOWPAN_IPHC_CTX_FLAG_COMPRESSION, &ctx->flags);
+
+	return 0;
+}
+
+static int lowpan_ctx_flag_c_get(void *data, u64 *val)
+{
+	*val = lowpan_iphc_ctx_is_compression(data);
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(lowpan_ctx_flag_c_fops, lowpan_ctx_flag_c_get,
+			lowpan_ctx_flag_c_set, "%llu\n");
+
+static int lowpan_ctx_plen_set(void *data, u64 val)
+{
+	struct lowpan_iphc_ctx *ctx = data;
+	struct lowpan_iphc_ctx_table *t =
+		container_of(ctx, struct lowpan_iphc_ctx_table, table[ctx->id]);
+
+	if (val > 128)
+		return -EINVAL;
+
+	spin_lock_bh(&t->lock);
+	ctx->plen = val;
+	spin_unlock_bh(&t->lock);
+
+	return 0;
+}
+
+static int lowpan_ctx_plen_get(void *data, u64 *val)
+{
+	struct lowpan_iphc_ctx *ctx = data;
+	struct lowpan_iphc_ctx_table *t =
+		container_of(ctx, struct lowpan_iphc_ctx_table, table[ctx->id]);
+
+	spin_lock_bh(&t->lock);
+	*val = ctx->plen;
+	spin_unlock_bh(&t->lock);
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(lowpan_ctx_plen_fops, lowpan_ctx_plen_get,
+			lowpan_ctx_plen_set, "%llu\n");
+
+static int lowpan_ctx_pfx_show(struct seq_file *file, void *offset)
+{
+	struct lowpan_iphc_ctx *ctx = file->private;
+	struct lowpan_iphc_ctx_table *t =
+		container_of(ctx, struct lowpan_iphc_ctx_table, table[ctx->id]);
+
+	spin_lock_bh(&t->lock);
+	seq_printf(file, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n",
+		   be16_to_cpu(ctx->pfx.s6_addr16[0]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[1]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[2]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[3]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[4]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[5]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[6]),
+		   be16_to_cpu(ctx->pfx.s6_addr16[7]));
+	spin_unlock_bh(&t->lock);
+
+	return 0;
+}
+
+static int lowpan_ctx_pfx_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, lowpan_ctx_pfx_show, inode->i_private);
+}
+
+static ssize_t lowpan_ctx_pfx_write(struct file *fp,
+				    const char __user *user_buf, size_t count,
+				    loff_t *ppos)
+{
+	char buf[128] = {};
+	struct seq_file *file = fp->private_data;
+	struct lowpan_iphc_ctx *ctx = file->private;
+	struct lowpan_iphc_ctx_table *t =
+		container_of(ctx, struct lowpan_iphc_ctx_table, table[ctx->id]);
+	int status = count, n, i;
+	unsigned int addr[8];
+
+	if (copy_from_user(&buf, user_buf, min_t(size_t, sizeof(buf) - 1,
+						 count))) {
+		status = -EFAULT;
+		goto out;
+	}
+
+	n = sscanf(buf, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x",
+		   &addr[0], &addr[1], &addr[2], &addr[3], &addr[4],
+		   &addr[5], &addr[6], &addr[7]);
+	if (n != LOWPAN_DEBUGFS_CTX_PFX_NUM_ARGS) {
+		status = -EINVAL;
+		goto out;
+	}
+
+	spin_lock_bh(&t->lock);
+	for (i = 0; i < 8; i++)
+		ctx->pfx.s6_addr16[i] = cpu_to_be16(addr[i] & 0xffff);
+	spin_unlock_bh(&t->lock);
+
+out:
+	return status;
+}
+
+const struct file_operations lowpan_ctx_pfx_fops = {
+	.open		= lowpan_ctx_pfx_open,
+	.read		= seq_read,
+	.write		= lowpan_ctx_pfx_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int lowpan_dev_debugfs_ctx_init(struct net_device *dev,
+				       struct dentry *ctx, u8 id)
+{
+	struct lowpan_priv *lpriv = lowpan_priv(dev);
+	struct dentry *dentry, *root;
+	char buf[32];
+
+	WARN_ON_ONCE(id > LOWPAN_IPHC_CTX_TABLE_SIZE);
+
+	sprintf(buf, "%d", id);
+
+	root = debugfs_create_dir(buf, ctx);
+	if (!root)
+		return -EINVAL;
+
+	dentry = debugfs_create_file("active", 0644, root,
+				     &lpriv->ctx.table[id],
+				     &lowpan_ctx_flag_active_fops);
+	if (!dentry)
+		return -EINVAL;
+
+	dentry = debugfs_create_file("compression", 0644, root,
+				     &lpriv->ctx.table[id],
+				     &lowpan_ctx_flag_c_fops);
+	if (!dentry)
+		return -EINVAL;
+
+	dentry = debugfs_create_file("prefix", 0644, root,
+				     &lpriv->ctx.table[id],
+				     &lowpan_ctx_pfx_fops);
+	if (!dentry)
+		return -EINVAL;
+
+	dentry = debugfs_create_file("prefix_len", 0644, root,
+				     &lpriv->ctx.table[id],
+				     &lowpan_ctx_plen_fops);
+	if (!dentry)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int lowpan_context_show(struct seq_file *file, void *offset)
+{
+	struct lowpan_iphc_ctx_table *t = file->private;
+	int i;
+
+	seq_printf(file, "%3s|%-43s|%c\n", "cid", "prefix", 'C');
+	seq_puts(file, "-------------------------------------------------\n");
+
+	spin_lock_bh(&t->lock);
+	for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++) {
+		if (!lowpan_iphc_ctx_is_active(&t->table[i]))
+			continue;
+
+		seq_printf(file, "%3d|%39pI6c/%-3d|%d\n", t->table[i].id,
+			   &t->table[i].pfx, t->table[i].plen,
+			   lowpan_iphc_ctx_is_compression(&t->table[i]));
+	}
+	spin_unlock_bh(&t->lock);
+
+	return 0;
+}
+
+static int lowpan_context_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, lowpan_context_show, inode->i_private);
+}
+
+const struct file_operations lowpan_context_fops = {
+	.open		= lowpan_context_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
 int lowpan_dev_debugfs_init(struct net_device *dev)
 {
 	struct lowpan_priv *lpriv = lowpan_priv(dev);
+	struct dentry *contexts, *dentry;
+	int ret, i;
 
 	/* creating the root */
 	lpriv->iface_debugfs = debugfs_create_dir(dev->name, lowpan_debugfs);
 	if (!lpriv->iface_debugfs)
 		goto fail;
 
+	contexts = debugfs_create_dir("contexts", lpriv->iface_debugfs);
+	if (!contexts)
+		goto remove_root;
+
+	dentry = debugfs_create_file("show", 0644, contexts,
+				     &lowpan_priv(dev)->ctx,
+				     &lowpan_context_fops);
+	if (!dentry)
+		goto remove_root;
+
+	for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++) {
+		ret = lowpan_dev_debugfs_ctx_init(dev, contexts, i);
+		if (ret < 0)
+			goto remove_root;
+	}
+
 	return 0;
 
+remove_root:
+	lowpan_dev_debugfs_exit(dev);
 fail:
 	return -EINVAL;
 }
diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c
index 346b5c1..d2a565c 100644
--- a/net/6lowpan/iphc.c
+++ b/net/6lowpan/iphc.c
@@ -56,6 +56,7 @@
 /* special link-layer handling */
 #include <net/mac802154.h>
 
+#include "6lowpan_i.h"
 #include "nhc.h"
 
 /* Values of fields within the IPHC encoding first byte */
@@ -147,6 +148,9 @@
 	 (((a)->s6_addr16[6]) == 0) &&		\
 	 (((a)->s6_addr[14]) == 0))
 
+#define LOWPAN_IPHC_CID_DCI(cid)	(cid & 0x0f)
+#define LOWPAN_IPHC_CID_SCI(cid)	((cid & 0xf0) >> 4)
+
 static inline void iphc_uncompress_eui64_lladdr(struct in6_addr *ipaddr,
 						const void *lladdr)
 {
@@ -195,6 +199,98 @@ static inline void iphc_uncompress_802154_lladdr(struct in6_addr *ipaddr,
 	}
 }
 
+static struct lowpan_iphc_ctx *
+lowpan_iphc_ctx_get_by_id(const struct net_device *dev, u8 id)
+{
+	struct lowpan_iphc_ctx *ret = &lowpan_priv(dev)->ctx.table[id];
+
+	if (!lowpan_iphc_ctx_is_active(ret))
+		return NULL;
+
+	return ret;
+}
+
+static struct lowpan_iphc_ctx *
+lowpan_iphc_ctx_get_by_addr(const struct net_device *dev,
+			    const struct in6_addr *addr)
+{
+	struct lowpan_iphc_ctx *table = lowpan_priv(dev)->ctx.table;
+	struct lowpan_iphc_ctx *ret = NULL;
+	struct in6_addr addr_pfx;
+	u8 addr_plen;
+	int i;
+
+	for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++) {
+		/* Check if context is valid. A context that is not valid
+		 * MUST NOT be used for compression.
+		 */
+		if (!lowpan_iphc_ctx_is_active(&table[i]) ||
+		    !lowpan_iphc_ctx_is_compression(&table[i]))
+			continue;
+
+		ipv6_addr_prefix(&addr_pfx, addr, table[i].plen);
+
+		/* if prefix len < 64, the remaining bits until 64th bit is
+		 * zero. Otherwise we use table[i]->plen.
+		 */
+		if (table[i].plen < 64)
+			addr_plen = 64;
+		else
+			addr_plen = table[i].plen;
+
+		if (ipv6_prefix_equal(&addr_pfx, &table[i].pfx, addr_plen)) {
+			/* remember first match */
+			if (!ret) {
+				ret = &table[i];
+				continue;
+			}
+
+			/* get the context with longest prefix len */
+			if (table[i].plen > ret->plen)
+				ret = &table[i];
+		}
+	}
+
+	return ret;
+}
+
+static struct lowpan_iphc_ctx *
+lowpan_iphc_ctx_get_by_mcast_addr(const struct net_device *dev,
+				  const struct in6_addr *addr)
+{
+	struct lowpan_iphc_ctx *table = lowpan_priv(dev)->ctx.table;
+	struct lowpan_iphc_ctx *ret = NULL;
+	struct in6_addr addr_mcast, network_pfx = {};
+	int i;
+
+	/* init mcast address with  */
+	memcpy(&addr_mcast, addr, sizeof(*addr));
+
+	for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++) {
+		/* Check if context is valid. A context that is not valid
+		 * MUST NOT be used for compression.
+		 */
+		if (!lowpan_iphc_ctx_is_active(&table[i]) ||
+		    !lowpan_iphc_ctx_is_compression(&table[i]))
+			continue;
+
+		/* setting plen */
+		addr_mcast.s6_addr[3] = table[i].plen;
+		/* get network prefix to copy into multicast address */
+		ipv6_addr_prefix(&network_pfx, &table[i].pfx,
+				 table[i].plen);
+		/* setting network prefix */
+		memcpy(&addr_mcast.s6_addr[4], &network_pfx, 8);
+
+		if (ipv6_addr_equal(addr, &addr_mcast)) {
+			ret = &table[i];
+			break;
+		}
+	}
+
+	return ret;
+}
+
 /* Uncompress address function for source and
  * destination address(non-multicast).
  *
@@ -259,30 +355,59 @@ static int uncompress_addr(struct sk_buff *skb, const struct net_device *dev,
 /* Uncompress address function for source context
  * based address(non-multicast).
  */
-static int uncompress_context_based_src_addr(struct sk_buff *skb,
-					     struct in6_addr *ipaddr,
-					     u8 address_mode)
+static int uncompress_ctx_addr(struct sk_buff *skb,
+			       const struct net_device *dev,
+			       const struct lowpan_iphc_ctx *ctx,
+			       struct in6_addr *ipaddr, u8 address_mode,
+			       const void *lladdr)
 {
+	bool fail;
+
 	switch (address_mode) {
-	case LOWPAN_IPHC_SAM_00:
-		/* unspec address ::
+	/* SAM and DAM are the same here */
+	case LOWPAN_IPHC_DAM_00:
+		fail = false;
+		/* SAM_00 -> unspec address ::
 		 * Do nothing, address is already ::
+		 *
+		 * DAM 00 -> reserved should never occur.
 		 */
 		break;
 	case LOWPAN_IPHC_SAM_01:
-		/* TODO */
+	case LOWPAN_IPHC_DAM_01:
+		fail = lowpan_fetch_skb(skb, &ipaddr->s6_addr[8], 8);
+		ipv6_addr_prefix_copy(ipaddr, &ctx->pfx, ctx->plen);
+		break;
 	case LOWPAN_IPHC_SAM_10:
-		/* TODO */
+	case LOWPAN_IPHC_DAM_10:
+		ipaddr->s6_addr[11] = 0xFF;
+		ipaddr->s6_addr[12] = 0xFE;
+		fail = lowpan_fetch_skb(skb, &ipaddr->s6_addr[14], 2);
+		ipv6_addr_prefix_copy(ipaddr, &ctx->pfx, ctx->plen);
+		break;
 	case LOWPAN_IPHC_SAM_11:
-		/* TODO */
-		netdev_warn(skb->dev, "SAM value 0x%x not supported\n",
-			    address_mode);
-		return -EINVAL;
+	case LOWPAN_IPHC_DAM_11:
+		fail = false;
+		switch (lowpan_priv(dev)->lltype) {
+		case LOWPAN_LLTYPE_IEEE802154:
+			iphc_uncompress_802154_lladdr(ipaddr, lladdr);
+			break;
+		default:
+			iphc_uncompress_eui64_lladdr(ipaddr, lladdr);
+			break;
+		}
+		ipv6_addr_prefix_copy(ipaddr, &ctx->pfx, ctx->plen);
+		break;
 	default:
 		pr_debug("Invalid sam value: 0x%x\n", address_mode);
 		return -EINVAL;
 	}
 
+	if (fail) {
+		pr_debug("Failed to fetch skb data\n");
+		return -EIO;
+	}
+
 	raw_dump_inline(NULL,
 			"Reconstructed context based ipv6 src addr is",
 			ipaddr->s6_addr, 16);
@@ -346,6 +471,30 @@ static int lowpan_uncompress_multicast_daddr(struct sk_buff *skb,
 	return 0;
 }
 
+static int lowpan_uncompress_multicast_ctx_daddr(struct sk_buff *skb,
+						 struct lowpan_iphc_ctx *ctx,
+						 struct in6_addr *ipaddr,
+						 u8 address_mode)
+{
+	struct in6_addr network_pfx = {};
+	bool fail;
+
+	ipaddr->s6_addr[0] = 0xFF;
+	fail = lowpan_fetch_skb(skb, &ipaddr->s6_addr[1], 2);
+	fail |= lowpan_fetch_skb(skb, &ipaddr->s6_addr[12], 4);
+	if (fail < 0)
+		return -EIO;
+
+	/* take prefix_len and network prefix from the context */
+	ipaddr->s6_addr[3] = ctx->plen;
+	/* get network prefix to copy into multicast address */
+	ipv6_addr_prefix(&network_pfx, &ctx->pfx, ctx->plen);
+	/* setting network prefix */
+	memcpy(&ipaddr->s6_addr[4], &network_pfx, 8);
+
+	return 0;
+}
+
 /* get the ecn values from iphc tf format and set it to ipv6hdr */
 static inline void lowpan_iphc_tf_set_ecn(struct ipv6hdr *hdr, const u8 *tf)
 {
@@ -459,7 +608,8 @@ int lowpan_header_decompress(struct sk_buff *skb, const struct net_device *dev,
 			     const void *daddr, const void *saddr)
 {
 	struct ipv6hdr hdr = {};
-	u8 iphc0, iphc1;
+	struct lowpan_iphc_ctx *ci;
+	u8 iphc0, iphc1, cid = 0;
 	int err;
 
 	raw_dump_table(__func__, "raw skb data dump uncompressed",
@@ -469,12 +619,14 @@ int lowpan_header_decompress(struct sk_buff *skb, const struct net_device *dev,
 	    lowpan_fetch_skb(skb, &iphc1, sizeof(iphc1)))
 		return -EINVAL;
 
-	/* another if the CID flag is set */
-	if (iphc1 & LOWPAN_IPHC_CID)
-		return -ENOTSUPP;
-
 	hdr.version = 6;
 
+	/* default CID = 0, another if the CID flag is set */
+	if (iphc1 & LOWPAN_IPHC_CID) {
+		if (lowpan_fetch_skb(skb, &cid, sizeof(cid)))
+			return -EINVAL;
+	}
+
 	err = lowpan_iphc_tf_decompress(skb, &hdr,
 					iphc0 & LOWPAN_IPHC_TF_MASK);
 	if (err < 0)
@@ -500,10 +652,17 @@ int lowpan_header_decompress(struct sk_buff *skb, const struct net_device *dev,
 	}
 
 	if (iphc1 & LOWPAN_IPHC_SAC) {
-		/* Source address context based uncompression */
+		spin_lock_bh(&lowpan_priv(dev)->ctx.lock);
+		ci = lowpan_iphc_ctx_get_by_id(dev, LOWPAN_IPHC_CID_SCI(cid));
+		if (!ci) {
+			spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+			return -EINVAL;
+		}
+
 		pr_debug("SAC bit is set. Handle context based source address.\n");
-		err = uncompress_context_based_src_addr(skb, &hdr.saddr,
-							iphc1 & LOWPAN_IPHC_SAM_MASK);
+		err = uncompress_ctx_addr(skb, dev, ci, &hdr.saddr,
+					  iphc1 & LOWPAN_IPHC_SAM_MASK, saddr);
+		spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
 	} else {
 		/* Source address uncompression */
 		pr_debug("source address stateless compression\n");
@@ -515,27 +674,52 @@ int lowpan_header_decompress(struct sk_buff *skb, const struct net_device *dev,
 	if (err)
 		return -EINVAL;
 
-	/* check for Multicast Compression */
-	if (iphc1 & LOWPAN_IPHC_M) {
-		if (iphc1 & LOWPAN_IPHC_DAC) {
-			pr_debug("dest: context-based mcast compression\n");
-			/* TODO: implement this */
-		} else {
-			err = lowpan_uncompress_multicast_daddr(skb, &hdr.daddr,
-								iphc1 & LOWPAN_IPHC_DAM_MASK);
+	switch (iphc1 & (LOWPAN_IPHC_M | LOWPAN_IPHC_DAC)) {
+	case LOWPAN_IPHC_M | LOWPAN_IPHC_DAC:
+		spin_lock_bh(&lowpan_priv(dev)->ctx.lock);
+		ci = lowpan_iphc_ctx_get_by_id(dev, LOWPAN_IPHC_CID_DCI(cid));
+		if (!ci) {
+			spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+			return -EINVAL;
+		}
 
-			if (err)
-				return -EINVAL;
+		/* multicast with context */
+		pr_debug("dest: context-based mcast compression\n");
+		err = lowpan_uncompress_multicast_ctx_daddr(skb, ci,
+							    &hdr.daddr,
+							    iphc1 & LOWPAN_IPHC_DAM_MASK);
+		spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+		break;
+	case LOWPAN_IPHC_M:
+		/* multicast */
+		err = lowpan_uncompress_multicast_daddr(skb, &hdr.daddr,
+							iphc1 & LOWPAN_IPHC_DAM_MASK);
+		break;
+	case LOWPAN_IPHC_DAC:
+		spin_lock_bh(&lowpan_priv(dev)->ctx.lock);
+		ci = lowpan_iphc_ctx_get_by_id(dev, LOWPAN_IPHC_CID_DCI(cid));
+		if (!ci) {
+			spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+			return -EINVAL;
 		}
-	} else {
+
+		/* Destination address context based uncompression */
+		pr_debug("DAC bit is set. Handle context based destination address.\n");
+		err = uncompress_ctx_addr(skb, dev, ci, &hdr.daddr,
+					  iphc1 & LOWPAN_IPHC_DAM_MASK, daddr);
+		spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+		break;
+	default:
 		err = uncompress_addr(skb, dev, &hdr.daddr,
 				      iphc1 & LOWPAN_IPHC_DAM_MASK, daddr);
 		pr_debug("dest: stateless compression mode %d dest %pI6c\n",
 			 iphc1 & LOWPAN_IPHC_DAM_MASK, &hdr.daddr);
-		if (err)
-			return -EINVAL;
+		break;
 	}
 
+	if (err)
+		return -EINVAL;
+
 	/* Next header data uncompression */
 	if (iphc0 & LOWPAN_IPHC_NH) {
 		err = lowpan_nhc_do_uncompression(skb, dev, &hdr);
@@ -585,6 +769,58 @@ static const u8 lowpan_iphc_dam_to_sam_value[] = {
 	[LOWPAN_IPHC_DAM_11] = LOWPAN_IPHC_SAM_11,
 };
 
+static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct in6_addr *ipaddr,
+				   const struct lowpan_iphc_ctx *ctx,
+				   const unsigned char *lladdr, bool sam)
+{
+	struct in6_addr tmp = {};
+	u8 dam;
+
+	/* check for SAM/DAM = 11 */
+	memcpy(&tmp.s6_addr[8], lladdr, 8);
+	/* second bit-flip (Universe/Local) is done according RFC2464 */
+	tmp.s6_addr[8] ^= 0x02;
+	/* context information are always used */
+	ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen);
+	if (ipv6_addr_equal(&tmp, ipaddr)) {
+		dam = LOWPAN_IPHC_DAM_11;
+		goto out;
+	}
+
+	memset(&tmp, 0, sizeof(tmp));
+	/* check for SAM/DAM = 01 */
+	tmp.s6_addr[11] = 0xFF;
+	tmp.s6_addr[12] = 0xFE;
+	memcpy(&tmp.s6_addr[14], &ipaddr->s6_addr[14], 2);
+	/* context information are always used */
+	ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen);
+	if (ipv6_addr_equal(&tmp, ipaddr)) {
+		lowpan_push_hc_data(hc_ptr, &ipaddr->s6_addr[14], 2);
+		dam = LOWPAN_IPHC_DAM_10;
+		goto out;
+	}
+
+	memset(&tmp, 0, sizeof(tmp));
+	/* check for SAM/DAM = 10, should always match */
+	memcpy(&tmp.s6_addr[8], &ipaddr->s6_addr[8], 8);
+	/* context information are always used */
+	ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen);
+	if (ipv6_addr_equal(&tmp, ipaddr)) {
+		lowpan_push_hc_data(hc_ptr, &ipaddr->s6_addr[8], 8);
+		dam = LOWPAN_IPHC_DAM_01;
+		goto out;
+	}
+
+	WARN_ON_ONCE("context found but no address mode matched\n");
+	return -EINVAL;
+out:
+
+	if (sam)
+		return lowpan_iphc_dam_to_sam_value[dam];
+	else
+		return dam;
+}
+
 static u8 lowpan_compress_addr_64(u8 **hc_ptr, const struct in6_addr *ipaddr,
 				  const unsigned char *lladdr, bool sam)
 {
@@ -708,6 +944,21 @@ static u8 lowpan_iphc_tf_compress(u8 **hc_ptr, const struct ipv6hdr *hdr)
 	return val;
 }
 
+static u8 lowpan_iphc_mcast_ctx_addr_compress(u8 **hc_ptr,
+					      const struct lowpan_iphc_ctx *ctx,
+					      const struct in6_addr *ipaddr)
+{
+	u8 data[6];
+
+	/* flags/scope, reserved (RIID) */
+	memcpy(data, &ipaddr->s6_addr[1], 2);
+	/* group ID */
+	memcpy(&data[1], &ipaddr->s6_addr[11], 4);
+	lowpan_push_hc_data(hc_ptr, data, 6);
+
+	return LOWPAN_IPHC_DAM_00;
+}
+
 static u8 lowpan_iphc_mcast_addr_compress(u8 **hc_ptr,
 					  const struct in6_addr *ipaddr)
 {
@@ -742,10 +993,11 @@ static u8 lowpan_iphc_mcast_addr_compress(u8 **hc_ptr,
 int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev,
 			   const void *daddr, const void *saddr)
 {
-	u8 iphc0, iphc1, *hc_ptr;
+	u8 iphc0, iphc1, *hc_ptr, cid = 0;
 	struct ipv6hdr *hdr;
 	u8 head[LOWPAN_IPHC_MAX_HC_BUF_LEN] = {};
-	int ret, addr_type;
+	struct lowpan_iphc_ctx *dci, *sci, dci_entry, sci_entry;
+	int ret, ipv6_daddr_type, ipv6_saddr_type;
 
 	if (skb->protocol != htons(ETH_P_IPV6))
 		return -EINVAL;
@@ -769,14 +1021,38 @@ int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev,
 	iphc0 = LOWPAN_DISPATCH_IPHC;
 	iphc1 = 0;
 
-	/* TODO: context lookup */
-
 	raw_dump_inline(__func__, "saddr", saddr, EUI64_ADDR_LEN);
 	raw_dump_inline(__func__, "daddr", daddr, EUI64_ADDR_LEN);
 
 	raw_dump_table(__func__, "sending raw skb network uncompressed packet",
 		       skb->data, skb->len);
 
+	ipv6_daddr_type = ipv6_addr_type(&hdr->daddr);
+	spin_lock_bh(&lowpan_priv(dev)->ctx.lock);
+	if (ipv6_daddr_type & IPV6_ADDR_MULTICAST)
+		dci = lowpan_iphc_ctx_get_by_mcast_addr(dev, &hdr->daddr);
+	else
+		dci = lowpan_iphc_ctx_get_by_addr(dev, &hdr->daddr);
+	if (dci) {
+		memcpy(&dci_entry, dci, sizeof(*dci));
+		cid |= dci->id;
+	}
+	spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+
+	spin_lock_bh(&lowpan_priv(dev)->ctx.lock);
+	sci = lowpan_iphc_ctx_get_by_addr(dev, &hdr->saddr);
+	if (sci) {
+		memcpy(&sci_entry, sci, sizeof(*sci));
+		cid |= (sci->id << 4);
+	}
+	spin_unlock_bh(&lowpan_priv(dev)->ctx.lock);
+
+	/* if cid is zero it will be compressed */
+	if (cid) {
+		iphc1 |= LOWPAN_IPHC_CID;
+		lowpan_push_hc_data(&hc_ptr, &cid, sizeof(cid));
+	}
+
 	/* Traffic Class, Flow Label compression */
 	iphc0 |= lowpan_iphc_tf_compress(&hc_ptr, hdr);
 
@@ -813,39 +1089,63 @@ int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev,
 				    sizeof(hdr->hop_limit));
 	}
 
-	addr_type = ipv6_addr_type(&hdr->saddr);
+	ipv6_saddr_type = ipv6_addr_type(&hdr->saddr);
 	/* source address compression */
-	if (addr_type == IPV6_ADDR_ANY) {
+	if (ipv6_saddr_type == IPV6_ADDR_ANY) {
 		pr_debug("source address is unspecified, setting SAC\n");
 		iphc1 |= LOWPAN_IPHC_SAC;
 	} else {
-		if (addr_type & IPV6_ADDR_LINKLOCAL) {
-			iphc1 |= lowpan_compress_addr_64(&hc_ptr, &hdr->saddr,
-							 saddr, true);
-			pr_debug("source address unicast link-local %pI6c iphc1 0x%02x\n",
-				 &hdr->saddr, iphc1);
+		if (sci) {
+			iphc1 |= lowpan_compress_ctx_addr(&hc_ptr, &hdr->saddr,
+							  &sci_entry, saddr,
+							  true);
+			iphc1 |= LOWPAN_IPHC_SAC;
 		} else {
-			pr_debug("send the full source address\n");
-			lowpan_push_hc_data(&hc_ptr, hdr->saddr.s6_addr, 16);
+			if (ipv6_saddr_type & IPV6_ADDR_LINKLOCAL) {
+				iphc1 |= lowpan_compress_addr_64(&hc_ptr,
+								 &hdr->saddr,
+								 saddr, true);
+				pr_debug("source address unicast link-local %pI6c iphc1 0x%02x\n",
+					 &hdr->saddr, iphc1);
+			} else {
+				pr_debug("send the full source address\n");
+				lowpan_push_hc_data(&hc_ptr,
+						    hdr->saddr.s6_addr, 16);
+			}
 		}
 	}
 
-	addr_type = ipv6_addr_type(&hdr->daddr);
 	/* destination address compression */
-	if (addr_type & IPV6_ADDR_MULTICAST) {
+	if (ipv6_daddr_type & IPV6_ADDR_MULTICAST) {
 		pr_debug("destination address is multicast: ");
-		iphc1 |= LOWPAN_IPHC_M;
-		iphc1 |= lowpan_iphc_mcast_addr_compress(&hc_ptr, &hdr->daddr);
+		if (dci) {
+			iphc1 |= lowpan_iphc_mcast_ctx_addr_compress(&hc_ptr,
+								     &dci_entry,
+								     &hdr->daddr);
+		} else {
+			iphc1 |= LOWPAN_IPHC_M;
+			iphc1 |= lowpan_iphc_mcast_addr_compress(&hc_ptr,
+								 &hdr->daddr);
+		}
 	} else {
-		if (addr_type & IPV6_ADDR_LINKLOCAL) {
-			/* TODO: context lookup */
-			iphc1 |= lowpan_compress_addr_64(&hc_ptr, &hdr->daddr,
-							 daddr, false);
-			pr_debug("dest address unicast link-local %pI6c "
-				 "iphc1 0x%02x\n", &hdr->daddr, iphc1);
+		if (dci) {
+			iphc1 |= lowpan_compress_ctx_addr(&hc_ptr, &hdr->daddr,
+							  &dci_entry, daddr,
+							  false);
+			iphc1 |= LOWPAN_IPHC_DAC;
 		} else {
-			pr_debug("dest address unicast %pI6c\n", &hdr->daddr);
-			lowpan_push_hc_data(&hc_ptr, hdr->daddr.s6_addr, 16);
+			if (ipv6_daddr_type & IPV6_ADDR_LINKLOCAL) {
+				iphc1 |= lowpan_compress_addr_64(&hc_ptr,
+								 &hdr->daddr,
+								 daddr, false);
+				pr_debug("dest address unicast link-local %pI6c iphc1 0x%02x\n",
+					 &hdr->daddr, iphc1);
+			} else {
+				pr_debug("dest address unicast %pI6c\n",
+					 &hdr->daddr);
+				lowpan_push_hc_data(&hc_ptr,
+						    hdr->daddr.s6_addr, 16);
+			}
 		}
 	}
 
-- 
cgit v0.10.2


From ebba380cc94689e372387aa7f63c063a3663e846 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Mon, 22 Feb 2016 09:13:55 +0100
Subject: ieee802154: 6lowpan: fix return of netdev notifier

This patch fixed the return value of netdev notifier. If the command is
a don't care a NOTIFY_DONE should be returned. If the command matched a
NOTIFY_OK should be returned.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c
index 737c87a..0023c90 100644
--- a/net/ieee802154/6lowpan/core.c
+++ b/net/ieee802154/6lowpan/core.c
@@ -207,7 +207,7 @@ static int lowpan_device_event(struct notifier_block *unused,
 	struct net_device *wdev = netdev_notifier_info_to_dev(ptr);
 
 	if (wdev->type != ARPHRD_IEEE802154)
-		goto out;
+		return NOTIFY_DONE;
 
 	switch (event) {
 	case NETDEV_UNREGISTER:
@@ -219,11 +219,10 @@ static int lowpan_device_event(struct notifier_block *unused,
 			lowpan_dellink(wdev->ieee802154_ptr->lowpan_dev, NULL);
 		break;
 	default:
-		break;
+		return NOTIFY_DONE;
 	}
 
-out:
-	return NOTIFY_DONE;
+	return NOTIFY_OK;
 }
 
 static struct notifier_block lowpan_dev_notifier = {
-- 
cgit v0.10.2


From d715fa6431a794e6a8cdb53d87acd3d03ed8a941 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 12 Feb 2016 12:09:38 -0500
Subject: net: dsa: mv88e6xxx: add port private structure

Add a per-port mv88e6xxx_priv_port structure to store per-port related
data, instead of adding several arrays of DSA_MAX_PORTS elements in the
mv88e6xxx_priv_state structure.

It currently only contains the port STP state.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 512c8c0..b0e00ed 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1131,7 +1131,7 @@ int mv88e6xxx_port_stp_update(struct dsa_switch *ds, int port, u8 state)
 	/* mv88e6xxx_port_stp_update may be called with softirqs disabled,
 	 * so we can not update the port state directly but need to schedule it.
 	 */
-	ps->port_state[port] = stp_state;
+	ps->ports[port].state = stp_state;
 	set_bit(port, &ps->port_state_update_mask);
 	schedule_work(&ps->bridge_work);
 
@@ -1925,7 +1925,7 @@ static void mv88e6xxx_bridge_work(struct work_struct *work)
 	while (ps->port_state_update_mask) {
 		port = __ffs(ps->port_state_update_mask);
 		clear_bit(port, &ps->port_state_update_mask);
-		mv88e6xxx_set_port_state(ds, port, ps->port_state[port]);
+		mv88e6xxx_set_port_state(ds, port, ps->ports[port].state);
 	}
 }
 
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index ca08f91..63a6f58 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -379,6 +379,10 @@ struct mv88e6xxx_vtu_stu_entry {
 	u8	data[DSA_MAX_PORTS];
 };
 
+struct mv88e6xxx_priv_port {
+	u8 state;
+};
+
 struct mv88e6xxx_priv_state {
 	/* When using multi-chip addressing, this mutex protects
 	 * access to the indirect access registers.  (In single-chip
@@ -415,8 +419,9 @@ struct mv88e6xxx_priv_state {
 	int		id; /* switch product id */
 	int		num_ports;	/* number of switch ports */
 
+	struct mv88e6xxx_priv_port	ports[DSA_MAX_PORTS];
+
 	unsigned long port_state_update_mask;
-	u8 port_state[DSA_MAX_PORTS];
 
 	struct work_struct bridge_work;
 };
-- 
cgit v0.10.2


From a6692754d61a6b3735803783f394880805675f99 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 12 Feb 2016 12:09:39 -0500
Subject: net: dsa: pass bridge down to drivers

Some DSA drivers may or may not support multiple software bridges on top
of an hardware switch.

It is more convenient for them to access the bridge's net_device for
finer configuration.

Removing the need to craft and access a bitmask also simplifies the
code.

This patch changes the signature of bridge related functions, update DSA
drivers, and removes dsa_slave_br_port_mask.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/dsa/dsa.txt b/Documentation/networking/dsa/dsa.txt
index aa9c1f9..ebf2153 100644
--- a/Documentation/networking/dsa/dsa.txt
+++ b/Documentation/networking/dsa/dsa.txt
@@ -524,17 +524,14 @@ Bridge layer
 - port_join_bridge: bridge layer function invoked when a given switch port is
   added to a bridge, this function should be doing the necessary at the switch
   level to permit the joining port from being added to the relevant logical
-  domain for it to ingress/egress traffic with other members of the bridge. DSA
-  does nothing but calculate a bitmask of switch ports currently members of the
-  specified bridge being requested the join
+  domain for it to ingress/egress traffic with other members of the bridge.
 
 - port_leave_bridge: bridge layer function invoked when a given switch port is
   removed from a bridge, this function should be doing the necessary at the
   switch level to deny the leaving port from ingress/egress traffic from the
   remaining bridge members. When the port leaves the bridge, it should be aged
   out at the switch hardware for the switch to (re) learn MAC addresses behind
-  this port. DSA calculates the bitmask of ports still members of the bridge
-  being left
+  this port.
 
 - port_stp_update: bridge layer function invoked when a given switch port STP
   state is computed by the bridge layer and should be propagated to switch
diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 6f946fe..3f62759 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -483,16 +483,17 @@ static int bcm_sf2_sw_fast_age_port(struct dsa_switch  *ds, int port)
 }
 
 static int bcm_sf2_sw_br_join(struct dsa_switch *ds, int port,
-			      u32 br_port_mask)
+			      struct net_device *bridge)
 {
 	struct bcm_sf2_priv *priv = ds_to_priv(ds);
 	unsigned int i;
 	u32 reg, p_ctl;
 
+	priv->port_sts[port].bridge_dev = bridge;
 	p_ctl = core_readl(priv, CORE_PORT_VLAN_CTL_PORT(port));
 
 	for (i = 0; i < priv->hw_params.num_ports; i++) {
-		if (!((1 << i) & br_port_mask))
+		if (priv->port_sts[i].bridge_dev != bridge)
 			continue;
 
 		/* Add this local port to the remote port VLAN control
@@ -515,10 +516,10 @@ static int bcm_sf2_sw_br_join(struct dsa_switch *ds, int port,
 	return 0;
 }
 
-static int bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port,
-			       u32 br_port_mask)
+static int bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port)
 {
 	struct bcm_sf2_priv *priv = ds_to_priv(ds);
+	struct net_device *bridge = priv->port_sts[port].bridge_dev;
 	unsigned int i;
 	u32 reg, p_ctl;
 
@@ -526,7 +527,7 @@ static int bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port,
 
 	for (i = 0; i < priv->hw_params.num_ports; i++) {
 		/* Don't touch the remaining ports */
-		if (!((1 << i) & br_port_mask))
+		if (priv->port_sts[i].bridge_dev != bridge)
 			continue;
 
 		reg = core_readl(priv, CORE_PORT_VLAN_CTL_PORT(i));
@@ -541,6 +542,7 @@ static int bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port,
 
 	core_writel(priv, p_ctl, CORE_PORT_VLAN_CTL_PORT(port));
 	priv->port_sts[port].vlan_ctl_mask = p_ctl;
+	priv->port_sts[port].bridge_dev = NULL;
 
 	return 0;
 }
diff --git a/drivers/net/dsa/bcm_sf2.h b/drivers/net/dsa/bcm_sf2.h
index 6bba1c9..200b1f5 100644
--- a/drivers/net/dsa/bcm_sf2.h
+++ b/drivers/net/dsa/bcm_sf2.h
@@ -50,6 +50,8 @@ struct bcm_sf2_port_status {
 	struct ethtool_eee eee;
 
 	u32 vlan_ctl_mask;
+
+	struct net_device *bridge_dev;
 };
 
 struct bcm_sf2_arl_entry {
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index b0e00ed..2e515e8 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1889,13 +1889,22 @@ unlock:
 	return err;
 }
 
-int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, u32 members)
+int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port,
+			       struct net_device *bridge)
 {
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+
+	ps->ports[port].bridge_dev = bridge;
+
 	return 0;
 }
 
-int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port, u32 members)
+int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port)
 {
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+
+	ps->ports[port].bridge_dev = NULL;
+
 	return 0;
 }
 
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 63a6f58..260b491 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -380,6 +380,7 @@ struct mv88e6xxx_vtu_stu_entry {
 };
 
 struct mv88e6xxx_priv_port {
+	struct net_device *bridge_dev;
 	u8 state;
 };
 
@@ -481,8 +482,9 @@ int mv88e6xxx_phy_write_indirect(struct dsa_switch *ds, int addr, int regnum,
 int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, struct ethtool_eee *e);
 int mv88e6xxx_set_eee(struct dsa_switch *ds, int port,
 		      struct phy_device *phydev, struct ethtool_eee *e);
-int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, u32 members);
-int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port, u32 members);
+int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port,
+			       struct net_device *bridge);
+int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port);
 int mv88e6xxx_port_stp_update(struct dsa_switch *ds, int port, u8 state);
 int mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port,
 				const struct switchdev_obj_port_vlan *vlan,
diff --git a/include/net/dsa.h b/include/net/dsa.h
index 26a0e86..1c845d7 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -297,9 +297,8 @@ struct dsa_switch_driver {
 	 * Bridge integration
 	 */
 	int	(*port_join_bridge)(struct dsa_switch *ds, int port,
-				    u32 br_port_mask);
-	int	(*port_leave_bridge)(struct dsa_switch *ds, int port,
-				     u32 br_port_mask);
+				    struct net_device *bridge);
+	int	(*port_leave_bridge)(struct dsa_switch *ds, int port);
 	int	(*port_stp_update)(struct dsa_switch *ds, int port,
 				   u8 state);
 
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index ab24521..ab515df 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -385,31 +385,6 @@ static int dsa_slave_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-/* Return a bitmask of all ports being currently bridged within a given bridge
- * device. Note that on leave, the mask will still return the bitmask of ports
- * currently bridged, prior to port removal, and this is exactly what we want.
- */
-static u32 dsa_slave_br_port_mask(struct dsa_switch *ds,
-				  struct net_device *bridge)
-{
-	struct dsa_slave_priv *p;
-	unsigned int port;
-	u32 mask = 0;
-
-	for (port = 0; port < DSA_MAX_PORTS; port++) {
-		if (!dsa_is_port_initialized(ds, port))
-			continue;
-
-		p = netdev_priv(ds->ports[port]);
-
-		if (ds->ports[port]->priv_flags & IFF_BRIDGE_PORT &&
-		    p->bridge_dev == bridge)
-			mask |= 1 << port;
-	}
-
-	return mask;
-}
-
 static int dsa_slave_stp_update(struct net_device *dev, u8 state)
 {
 	struct dsa_slave_priv *p = netdev_priv(dev);
@@ -533,8 +508,7 @@ static int dsa_slave_bridge_port_join(struct net_device *dev,
 	p->bridge_dev = br;
 
 	if (ds->drv->port_join_bridge)
-		ret = ds->drv->port_join_bridge(ds, p->port,
-						dsa_slave_br_port_mask(ds, br));
+		ret = ds->drv->port_join_bridge(ds, p->port, br);
 
 	return ret;
 }
@@ -547,8 +521,7 @@ static int dsa_slave_bridge_port_leave(struct net_device *dev)
 
 
 	if (ds->drv->port_leave_bridge)
-		ret = ds->drv->port_leave_bridge(ds, p->port,
-						 dsa_slave_br_port_mask(ds, p->bridge_dev));
+		ret = ds->drv->port_leave_bridge(ds, p->port);
 
 	p->bridge_dev = NULL;
 
-- 
cgit v0.10.2


From da9c359e19f0a72a386a0e83a098b6dae21aa3c3 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 12 Feb 2016 12:09:40 -0500
Subject: net: dsa: mv88e6xxx: check hardware VLAN in use

The DSA drivers now have access to the VLAN prepare phase and the bridge
net_device. It is easier to check for overlapping bridges from within
the driver. Thus add such check in mv88e6xxx_port_vlan_prepare.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 2e515e8..685dcb0 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1471,14 +1471,78 @@ static int _mv88e6xxx_vlan_init(struct dsa_switch *ds, u16 vid,
 	return 0;
 }
 
+static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port,
+					u16 vid_begin, u16 vid_end)
+{
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	struct mv88e6xxx_vtu_stu_entry vlan;
+	int i, err;
+
+	if (!vid_begin)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&ps->smi_mutex);
+
+	err = _mv88e6xxx_vtu_vid_write(ds, vid_begin - 1);
+	if (err)
+		goto unlock;
+
+	do {
+		err = _mv88e6xxx_vtu_getnext(ds, &vlan);
+		if (err)
+			goto unlock;
+
+		if (!vlan.valid)
+			break;
+
+		if (vlan.vid > vid_end)
+			break;
+
+		for (i = 0; i < ps->num_ports; ++i) {
+			if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i))
+				continue;
+
+			if (vlan.data[i] ==
+			    GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER)
+				continue;
+
+			if (ps->ports[i].bridge_dev ==
+			    ps->ports[port].bridge_dev)
+				break; /* same bridge, check next VLAN */
+
+			netdev_warn(ds->ports[port],
+				    "hardware VLAN %d already used by %s\n",
+				    vlan.vid,
+				    netdev_name(ps->ports[i].bridge_dev));
+			err = -EOPNOTSUPP;
+			goto unlock;
+		}
+	} while (vlan.vid < vid_end);
+
+unlock:
+	mutex_unlock(&ps->smi_mutex);
+
+	return err;
+}
+
 int mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port,
 				const struct switchdev_obj_port_vlan *vlan,
 				struct switchdev_trans *trans)
 {
+	int err;
+
 	/* We reserve a few VLANs to isolate unbridged ports */
 	if (vlan->vid_end >= 4000)
 		return -EOPNOTSUPP;
 
+	/* If the requested port doesn't belong to the same bridge as the VLAN
+	 * members, do not support it (yet) and fallback to software VLAN.
+	 */
+	err = mv88e6xxx_port_check_hw_vlan(ds, port, vlan->vid_begin,
+					   vlan->vid_end);
+	if (err)
+		return err;
+
 	/* We don't need any dynamic resource from the kernel (yet),
 	 * so skip the prepare phase.
 	 */
-- 
cgit v0.10.2


From 9d2dd736698e6f1dc176b9669b6390ddcd2063b1 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 12 Feb 2016 12:09:41 -0500
Subject: net: dsa: remove dsa_bridge_check_vlan_range

DSA drivers may support multiple bridge groups with the same hardware
VLAN. The mv88e6xxx driver which cannot yet, already has its own check
for overlapping bridges. Thus remove the check from the DSA layer.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index ab515df..14ca978 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -201,47 +201,6 @@ out:
 	return 0;
 }
 
-static int dsa_bridge_check_vlan_range(struct dsa_switch *ds,
-				       const struct net_device *bridge,
-				       u16 vid_begin, u16 vid_end)
-{
-	struct dsa_slave_priv *p;
-	struct net_device *dev, *vlan_br;
-	DECLARE_BITMAP(members, DSA_MAX_PORTS);
-	DECLARE_BITMAP(untagged, DSA_MAX_PORTS);
-	u16 vid;
-	int member, err;
-
-	if (!ds->drv->vlan_getnext || !vid_begin)
-		return -EOPNOTSUPP;
-
-	vid = vid_begin - 1;
-
-	do {
-		err = ds->drv->vlan_getnext(ds, &vid, members, untagged);
-		if (err)
-			break;
-
-		if (vid > vid_end)
-			break;
-
-		member = find_first_bit(members, DSA_MAX_PORTS);
-		if (member == DSA_MAX_PORTS)
-			continue;
-
-		dev = ds->ports[member];
-		p = netdev_priv(dev);
-		vlan_br = p->bridge_dev;
-		if (vlan_br == bridge)
-			continue;
-
-		netdev_dbg(vlan_br, "hardware VLAN %d already in use\n", vid);
-		return -EOPNOTSUPP;
-	} while (vid < vid_end);
-
-	return err == -ENOENT ? 0 : err;
-}
-
 static int dsa_slave_port_vlan_add(struct net_device *dev,
 				   const struct switchdev_obj_port_vlan *vlan,
 				   struct switchdev_trans *trans)
@@ -254,15 +213,6 @@ static int dsa_slave_port_vlan_add(struct net_device *dev,
 		if (!ds->drv->port_vlan_prepare || !ds->drv->port_vlan_add)
 			return -EOPNOTSUPP;
 
-		/* If the requested port doesn't belong to the same bridge as
-		 * the VLAN members, fallback to software VLAN (hopefully).
-		 */
-		err = dsa_bridge_check_vlan_range(ds, p->bridge_dev,
-						  vlan->vid_begin,
-						  vlan->vid_end);
-		if (err)
-			return err;
-
 		err = ds->drv->port_vlan_prepare(ds, p->port, vlan, trans);
 		if (err)
 			return err;
-- 
cgit v0.10.2


From 412a6d800c7380c1b87c11080c7da905c27cfea8 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 8 Dec 2015 19:09:05 +0200
Subject: mac80211: support hw managing reorder logic

Enable driver to manage the reordering logic itself.
This is needed for example for the iwlwifi driver that
will support hardware assisted reordering.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 6c9c559..ee6305a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1929,6 +1929,11 @@ struct ieee80211_txq {
  *	by just its MAC address; this prevents, for example, the same station
  *	from connecting to two virtual AP interfaces at the same time.
  *
+ * @IEEE80211_HW_SUPPORTS_REORDERING_BUFFER: Hardware (or driver) manages the
+ *	reordering buffer internally, guaranteeing mac80211 receives frames in
+ *	order and does not need to manage its own reorder buffer or BA session
+ *	timeout.
+ *
  * @NUM_IEEE80211_HW_FLAGS: number of hardware flags, used for sizing arrays
  */
 enum ieee80211_hw_flags {
@@ -1965,6 +1970,7 @@ enum ieee80211_hw_flags {
 	IEEE80211_HW_SUPPORTS_AMSDU_IN_AMPDU,
 	IEEE80211_HW_BEACON_TX_STATUS,
 	IEEE80211_HW_NEEDS_UNIQUE_STA_ADDR,
+	IEEE80211_HW_SUPPORTS_REORDERING_BUFFER,
 
 	/* keep last, obviously */
 	NUM_IEEE80211_HW_FLAGS
diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c
index ec80db7..2ab5479 100644
--- a/net/mac80211/agg-rx.c
+++ b/net/mac80211/agg-rx.c
@@ -76,10 +76,11 @@ void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid,
 	tid_rx = rcu_dereference_protected(sta->ampdu_mlme.tid_rx[tid],
 					lockdep_is_held(&sta->ampdu_mlme.mtx));
 
-	if (!tid_rx)
+	if (!test_bit(tid, sta->ampdu_mlme.agg_session_valid))
 		return;
 
 	RCU_INIT_POINTER(sta->ampdu_mlme.tid_rx[tid], NULL);
+	__clear_bit(tid, sta->ampdu_mlme.agg_session_valid);
 
 	ht_dbg(sta->sdata,
 	       "Rx BA session stop requested for %pM tid %u %s reason: %d\n",
@@ -97,6 +98,13 @@ void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid,
 		ieee80211_send_delba(sta->sdata, sta->sta.addr,
 				     tid, WLAN_BACK_RECIPIENT, reason);
 
+	/*
+	 * return here in case tid_rx is not assigned - which will happen if
+	 * IEEE80211_HW_SUPPORTS_REORDERING_BUFFER is set.
+	 */
+	if (!tid_rx)
+		return;
+
 	del_timer_sync(&tid_rx->session_timer);
 
 	/* make sure ieee80211_sta_reorder_release() doesn't re-arm the timer */
@@ -297,7 +305,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 	/* examine state machine */
 	mutex_lock(&sta->ampdu_mlme.mtx);
 
-	if (sta->ampdu_mlme.tid_rx[tid]) {
+	if (test_bit(tid, sta->ampdu_mlme.agg_session_valid)) {
 		ht_dbg_ratelimited(sta->sdata,
 				   "unexpected AddBA Req from %pM on tid %u\n",
 				   sta->sta.addr, tid);
@@ -308,6 +316,16 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 						false);
 	}
 
+	if (ieee80211_hw_check(&local->hw, SUPPORTS_REORDERING_BUFFER)) {
+		ret = drv_ampdu_action(local, sta->sdata, &params);
+		ht_dbg(sta->sdata,
+		       "Rx A-MPDU request on %pM tid %d result %d\n",
+		       sta->sta.addr, tid, ret);
+		if (!ret)
+			status = WLAN_STATUS_SUCCESS;
+		goto end;
+	}
+
 	/* prepare A-MPDU MLME for Rx aggregation */
 	tid_agg_rx = kmalloc(sizeof(struct tid_ampdu_rx), GFP_KERNEL);
 	if (!tid_agg_rx)
@@ -369,6 +387,8 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 	}
 
 end:
+	if (status == WLAN_STATUS_SUCCESS)
+		__set_bit(tid, sta->ampdu_mlme.agg_session_valid);
 	mutex_unlock(&sta->ampdu_mlme.mtx);
 
 end_no_lock:
diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c
index abbdff0..e433d0c 100644
--- a/net/mac80211/debugfs.c
+++ b/net/mac80211/debugfs.c
@@ -126,6 +126,7 @@ static const char *hw_flag_names[NUM_IEEE80211_HW_FLAGS + 1] = {
 	FLAG(SUPPORTS_AMSDU_IN_AMPDU),
 	FLAG(BEACON_TX_STATUS),
 	FLAG(NEEDS_UNIQUE_STA_ADDR),
+	FLAG(SUPPORTS_REORDERING_BUFFER),
 
 	/* keep last for the build bug below */
 	(void *)0x1
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index d605162..f4d3899 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -1,6 +1,7 @@
 /*
  * Copyright 2002-2005, Devicescape Software, Inc.
  * Copyright 2013-2014  Intel Mobile Communications GmbH
+ * Copyright(c) 2015 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -212,20 +213,21 @@ struct tid_ampdu_rx {
 /**
  * struct sta_ampdu_mlme - STA aggregation information.
  *
+ * @mtx: mutex to protect all TX data (except non-NULL assignments
+ *	to tid_tx[idx], which are protected by the sta spinlock)
+ *	tid_start_tx is also protected by sta->lock.
  * @tid_rx: aggregation info for Rx per TID -- RCU protected
- * @tid_tx: aggregation info for Tx per TID
- * @tid_start_tx: sessions where start was requested
- * @addba_req_num: number of times addBA request has been sent.
- * @last_addba_req_time: timestamp of the last addBA request.
- * @dialog_token_allocator: dialog token enumerator for each new session;
- * @work: work struct for starting/stopping aggregation
  * @tid_rx_timer_expired: bitmap indicating on which TIDs the
  *	RX timer expired until the work for it runs
  * @tid_rx_stop_requested:  bitmap indicating which BA sessions per TID the
  *	driver requested to close until the work for it runs
- * @mtx: mutex to protect all TX data (except non-NULL assignments
- *	to tid_tx[idx], which are protected by the sta spinlock)
- *	tid_start_tx is also protected by sta->lock.
+ * @agg_session_valid: bitmap indicating which TID has a rx BA session open on
+ * @work: work struct for starting/stopping aggregation
+ * @tid_tx: aggregation info for Tx per TID
+ * @tid_start_tx: sessions where start was requested
+ * @last_addba_req_time: timestamp of the last addBA request.
+ * @addba_req_num: number of times addBA request has been sent.
+ * @dialog_token_allocator: dialog token enumerator for each new session;
  */
 struct sta_ampdu_mlme {
 	struct mutex mtx;
@@ -233,6 +235,7 @@ struct sta_ampdu_mlme {
 	struct tid_ampdu_rx __rcu *tid_rx[IEEE80211_NUM_TIDS];
 	unsigned long tid_rx_timer_expired[BITS_TO_LONGS(IEEE80211_NUM_TIDS)];
 	unsigned long tid_rx_stop_requested[BITS_TO_LONGS(IEEE80211_NUM_TIDS)];
+	unsigned long agg_session_valid[BITS_TO_LONGS(IEEE80211_NUM_TIDS)];
 	/* tx */
 	struct work_struct work;
 	struct tid_ampdu_tx __rcu *tid_tx[IEEE80211_NUM_TIDS];
-- 
cgit v0.10.2


From 178830481eee5eea147a1c8fab67a96e09d80345 Mon Sep 17 00:00:00 2001
From: Grzegorz Bajorski <grzegorz.bajorski@tieto.com>
Date: Fri, 11 Dec 2015 14:39:46 +0100
Subject: mac80211: allow drivers to report (non-)monitor frames

Some drivers offload some frames internally (e.g.
AddBa). Reporting such frames to mac80211 would
only confuse MLME. However it would be useful to
be able to pass such frames to monitor interfaces
for sniffing purposes, e.g. when running AP +
monitor.

To do that allow drivers to tell mac80211 whether
a given frame should be:
 - processed but not delivered to any monitor vif
 - not processed but delievered to monitor vifs
   only

Signed-off-by: Grzegorz Bajorski <grzegorz.bajorski@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index ee6305a..5910085 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1031,6 +1031,14 @@ ieee80211_tx_info_clear_status(struct ieee80211_tx_info *info)
  * @RX_FLAG_AMPDU_DELIM_CRC_KNOWN: The delimiter CRC field is known (the CRC
  *	is stored in the @ampdu_delimiter_crc field)
  * @RX_FLAG_LDPC: LDPC was used
+ * @RX_FLAG_ONLY_MONITOR: Report frame only to monitor interfaces without
+ *	processing it in any regular way.
+ *	This is useful if drivers offload some frames but still want to report
+ *	them for sniffing purposes.
+ * @RX_FLAG_SKIP_MONITOR: Process and report frame to all interfaces except
+ *	monitor interfaces.
+ *	This is useful if drivers offload some frames but still want to report
+ *	them for sniffing purposes.
  * @RX_FLAG_STBC_MASK: STBC 2 bit bitmask. 1 - Nss=1, 2 - Nss=2, 3 - Nss=3
  * @RX_FLAG_10MHZ: 10 MHz (half channel) was used
  * @RX_FLAG_5MHZ: 5 MHz (quarter channel) was used
@@ -1071,6 +1079,8 @@ enum mac80211_rx_flags {
 	RX_FLAG_MACTIME_END		= BIT(21),
 	RX_FLAG_VHT			= BIT(22),
 	RX_FLAG_LDPC			= BIT(23),
+	RX_FLAG_ONLY_MONITOR		= BIT(24),
+	RX_FLAG_SKIP_MONITOR		= BIT(25),
 	RX_FLAG_STBC_MASK		= BIT(26) | BIT(27),
 	RX_FLAG_10MHZ			= BIT(28),
 	RX_FLAG_5MHZ			= BIT(29),
@@ -1089,6 +1099,7 @@ enum mac80211_rx_flags {
  * @RX_VHT_FLAG_160MHZ: 160 MHz was used
  * @RX_VHT_FLAG_BF: packet was beamformed
  */
+
 enum mac80211_rx_vht_flags {
 	RX_VHT_FLAG_80MHZ		= BIT(0),
 	RX_VHT_FLAG_160MHZ		= BIT(1),
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index fe675d7..ae993ed 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -122,7 +122,8 @@ static inline bool should_drop_frame(struct sk_buff *skb, int present_fcs_len,
 	hdr = (void *)(skb->data + rtap_vendor_space);
 
 	if (status->flag & (RX_FLAG_FAILED_FCS_CRC |
-			    RX_FLAG_FAILED_PLCP_CRC))
+			    RX_FLAG_FAILED_PLCP_CRC |
+			    RX_FLAG_ONLY_MONITOR))
 		return true;
 
 	if (unlikely(skb->len < 16 + present_fcs_len + rtap_vendor_space))
@@ -507,7 +508,7 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 		return NULL;
 	}
 
-	if (!local->monitors) {
+	if (!local->monitors || (status->flag & RX_FLAG_SKIP_MONITOR)) {
 		if (should_drop_frame(origskb, present_fcs_len,
 				      rtap_vendor_space)) {
 			dev_kfree_skb(origskb);
-- 
cgit v0.10.2


From a7201a6c5ea0157e96d17240d185a538bfdb7038 Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Sun, 13 Dec 2015 13:41:43 +0200
Subject: mac80211: Recalc min chandef when station is associated

The minimum chandef bandwidth calculation was done only in case
a new station was inserted (or when an existing station was removed).
However, it is possible that stations are inserted before they are
associated, e.g., when FULL_AP_CLIENT_STATE is supported and user
space adds stations unassociated.

Fix this by calling ieee80211_recalc_min_chandef() whenever
a station transitions in/out the associated state, and only
consider station marked as associated.

Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 1d1b9b7..2839811 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -231,7 +231,7 @@ ieee80211_get_max_required_bw(struct ieee80211_sub_if_data *sdata)
 		    !(sta->sdata->bss && sta->sdata->bss == sdata->bss))
 			continue;
 
-		if (!sta->uploaded)
+		if (!sta->uploaded || !test_sta_flag(sta, WLAN_STA_ASSOC))
 			continue;
 
 		max_bw = max(max_bw, ieee80211_get_sta_bw(&sta->sta));
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 4402ad5..7e007cf 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -530,7 +530,6 @@ static int sta_info_insert_finish(struct sta_info *sta) __acquires(RCU)
 	/* accept BA sessions now */
 	clear_sta_flag(sta, WLAN_STA_BLOCK_BA);
 
-	ieee80211_recalc_min_chandef(sdata);
 	ieee80211_sta_debugfs_add(sta);
 	rate_control_add_sta_debugfs(sta);
 
@@ -941,7 +940,6 @@ static void __sta_info_destroy_part2(struct sta_info *sta)
 
 	rate_control_remove_sta_debugfs(sta);
 	ieee80211_sta_debugfs_remove(sta);
-	ieee80211_recalc_min_chandef(sdata);
 
 	cleanup_single_sta(sta);
 }
@@ -1808,14 +1806,17 @@ int sta_info_move_state(struct sta_info *sta,
 			clear_bit(WLAN_STA_AUTH, &sta->_flags);
 		break;
 	case IEEE80211_STA_AUTH:
-		if (sta->sta_state == IEEE80211_STA_NONE)
+		if (sta->sta_state == IEEE80211_STA_NONE) {
 			set_bit(WLAN_STA_AUTH, &sta->_flags);
-		else if (sta->sta_state == IEEE80211_STA_ASSOC)
+		} else if (sta->sta_state == IEEE80211_STA_ASSOC) {
 			clear_bit(WLAN_STA_ASSOC, &sta->_flags);
+			ieee80211_recalc_min_chandef(sta->sdata);
+		}
 		break;
 	case IEEE80211_STA_ASSOC:
 		if (sta->sta_state == IEEE80211_STA_AUTH) {
 			set_bit(WLAN_STA_ASSOC, &sta->_flags);
+			ieee80211_recalc_min_chandef(sta->sdata);
 		} else if (sta->sta_state == IEEE80211_STA_AUTHORIZED) {
 			if (sta->sdata->vif.type == NL80211_IFTYPE_AP ||
 			    (sta->sdata->vif.type == NL80211_IFTYPE_AP_VLAN &&
-- 
cgit v0.10.2


From 506bcfa8abebdbcebdc17b03e96e38dc0b8ce765 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 13 Dec 2015 15:41:05 +0200
Subject: mac80211: limit the A-MSDU Tx based on peer's capabilities

In VHT, the specification allows to limit the number of
MSDUs in an A-MSDU in the Extended Capabilities IE. There
is also a limitation on the byte size in the VHT IE.
In HT, the only limitation is on the byte size.
Parse the capabilities from the peer and make them
available to the driver.

In HT, there is another limitation when a BA agreement
is active: the byte size can't be greater than 4095.
This is not enforced here.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index d9ddb89..3b1f6ce 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -163,6 +163,14 @@ static inline u16 ieee80211_sn_sub(u16 sn1, u16 sn2)
 /* 30 byte 4 addr hdr, 2 byte QoS, 2304 byte MSDU, 12 byte crypt, 4 byte FCS */
 #define IEEE80211_MAX_FRAME_LEN		2352
 
+/* Maximal size of an A-MSDU */
+#define IEEE80211_MAX_MPDU_LEN_HT_3839		3839
+#define IEEE80211_MAX_MPDU_LEN_HT_7935		7935
+
+#define IEEE80211_MAX_MPDU_LEN_VHT_3895		3895
+#define IEEE80211_MAX_MPDU_LEN_VHT_7991		7991
+#define IEEE80211_MAX_MPDU_LEN_VHT_11454	11454
+
 #define IEEE80211_MAX_SSID_LEN		32
 
 #define IEEE80211_MAX_MESH_ID_LEN	32
@@ -1505,6 +1513,7 @@ struct ieee80211_vht_operation {
 #define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895			0x00000000
 #define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_7991			0x00000001
 #define IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454			0x00000002
+#define IEEE80211_VHT_CAP_MAX_MPDU_MASK				0x00000003
 #define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ		0x00000004
 #define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ	0x00000008
 #define IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK			0x0000000C
@@ -2086,6 +2095,16 @@ enum ieee80211_tdls_actioncode {
 #define WLAN_EXT_CAPA8_TDLS_WIDE_BW_ENABLED	BIT(5)
 #define WLAN_EXT_CAPA8_OPMODE_NOTIF	BIT(6)
 
+/* Defines the maximal number of MSDUs in an A-MSDU. */
+#define WLAN_EXT_CAPA8_MAX_MSDU_IN_AMSDU_LSB	BIT(7)
+#define WLAN_EXT_CAPA9_MAX_MSDU_IN_AMSDU_MSB	BIT(0)
+
+/*
+ * Fine Timing Measurement Initiator - bit 71 of @WLAN_EID_EXT_CAPABILITY
+ * information element
+ */
+#define WLAN_EXT_CAPA9_FTM_INITIATOR	BIT(7)
+
 /* TDLS specific payload type in the LLC/SNAP header */
 #define WLAN_TDLS_SNAP_RFTYPE	0x2
 
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5910085..df5698e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1714,6 +1714,18 @@ struct ieee80211_sta_rates {
  * @tdls_initiator: indicates the STA is an initiator of the TDLS link. Only
  *	valid if the STA is a TDLS peer in the first place.
  * @mfp: indicates whether the STA uses management frame protection or not.
+ * @max_amsdu_subframes: indicates the maximal number of MSDUs in a single
+ *	A-MSDU. Taken from the Extended Capabilities element. 0 means
+ *	unlimited.
+ * @max_amsdu_len: indicates the maximal length of an A-MSDU in bytes. This
+ *	field is always valid for packets with a VHT preamble. For packets
+ *	with a HT preamble, additional limits apply:
+ *		+ If the skb is transmitted as part of a BA agreement, the
+ *		  A-MSDU maximal size is min(max_amsdu_len, 4065) bytes.
+ *		+ If the skb is not part of a BA aggreement, the A-MSDU maximal
+ *		  size is min(max_amsdu_len, 7935) bytes.
+ *	Both additional HT limits must be enforced by the low level driver.
+ *	This is defined by the spec (IEEE 802.11-2012 section 8.3.2.2 NOTE 2).
  * @txq: per-TID data TX queues (if driver uses the TXQ abstraction)
  */
 struct ieee80211_sta {
@@ -1732,6 +1744,8 @@ struct ieee80211_sta {
 	bool tdls;
 	bool tdls_initiator;
 	bool mfp;
+	u8 max_amsdu_subframes;
+	u16 max_amsdu_len;
 
 	struct ieee80211_txq *txq[IEEE80211_NUM_TIDS];
 
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 166a29f..66d22de 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1131,6 +1131,34 @@ static int sta_apply_parameters(struct ieee80211_local *local,
 		sta->sta.max_sp = params->max_sp;
 	}
 
+	/* The sender might not have sent the last bit, consider it to be 0 */
+	if (params->ext_capab_len >= 8) {
+		u8 val = (params->ext_capab[7] &
+			  WLAN_EXT_CAPA8_MAX_MSDU_IN_AMSDU_LSB) >> 7;
+
+		/* we did get all the bits, take the MSB as well */
+		if (params->ext_capab_len >= 9) {
+			u8 val_msb = params->ext_capab[8] &
+				WLAN_EXT_CAPA9_MAX_MSDU_IN_AMSDU_MSB;
+			val_msb <<= 1;
+			val |= val_msb;
+		}
+
+		switch (val) {
+		case 1:
+			sta->sta.max_amsdu_subframes = 32;
+			break;
+		case 2:
+			sta->sta.max_amsdu_subframes = 16;
+			break;
+		case 3:
+			sta->sta.max_amsdu_subframes = 8;
+			break;
+		default:
+			sta->sta.max_amsdu_subframes = 0;
+		}
+	}
+
 	/*
 	 * cfg80211 validates this (1-2007) and allows setting the AID
 	 * only when creating a new station entry
@@ -1160,6 +1188,7 @@ static int sta_apply_parameters(struct ieee80211_local *local,
 		ieee80211_ht_cap_ie_to_sta_ht_cap(sdata, sband,
 						  params->ht_capa, sta);
 
+	/* VHT can override some HT caps such as the A-MSDU max length */
 	if (params->vht_capa)
 		ieee80211_vht_cap_ie_to_sta_vht_cap(sdata, sband,
 						    params->vht_capa, sta);
diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c
index 7a76ce6..f4a5287 100644
--- a/net/mac80211/ht.c
+++ b/net/mac80211/ht.c
@@ -230,6 +230,11 @@ bool ieee80211_ht_cap_ie_to_sta_ht_cap(struct ieee80211_sub_if_data *sdata,
 	/* set Rx highest rate */
 	ht_cap.mcs.rx_highest = ht_cap_ie->mcs.rx_highest;
 
+	if (ht_cap.cap & IEEE80211_HT_CAP_MAX_AMSDU)
+		sta->sta.max_amsdu_len = IEEE80211_MAX_MPDU_LEN_HT_7935;
+	else
+		sta->sta.max_amsdu_len = IEEE80211_MAX_MPDU_LEN_HT_3839;
+
  apply:
 	changed = memcmp(&sta->sta.ht_cap, &ht_cap, sizeof(ht_cap));
 
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index 050de08..204cf9a 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -281,6 +281,23 @@ ieee80211_vht_cap_ie_to_sta_vht_cap(struct ieee80211_sub_if_data *sdata,
 	}
 
 	sta->sta.bandwidth = ieee80211_sta_cur_vht_bw(sta);
+
+	/* If HT IE reported 3839 bytes only, stay with that size. */
+	if (sta->sta.max_amsdu_len == IEEE80211_MAX_MPDU_LEN_HT_3839)
+		return;
+
+	switch (vht_cap->cap & IEEE80211_VHT_CAP_MAX_MPDU_MASK) {
+	case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454:
+		sta->sta.max_amsdu_len = IEEE80211_MAX_MPDU_LEN_VHT_11454;
+		break;
+	case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_7991:
+		sta->sta.max_amsdu_len = IEEE80211_MAX_MPDU_LEN_VHT_7991;
+		break;
+	case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895:
+	default:
+		sta->sta.max_amsdu_len = IEEE80211_MAX_MPDU_LEN_VHT_3895;
+		break;
+	}
 }
 
 enum ieee80211_sta_rx_bandwidth ieee80211_sta_cap_rx_bw(struct sta_info *sta)
-- 
cgit v0.10.2


From de3bb771f471ba7c745a913b95eee70d5fb199b1 Mon Sep 17 00:00:00 2001
From: Ola Olsson <ola1olsson@gmail.com>
Date: Wed, 16 Dec 2015 22:43:35 +0100
Subject: cfg80211: add more warnings for inconsistent ops

Print a warning whenever an expected callback function
lacks implementation.

Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/core.c b/net/wireless/core.c
index b091551..3a9c41b 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -352,6 +352,16 @@ struct wiphy *wiphy_new_nm(const struct cfg80211_ops *ops, int sizeof_priv,
 	WARN_ON(ops->add_station && !ops->del_station);
 	WARN_ON(ops->add_mpath && !ops->del_mpath);
 	WARN_ON(ops->join_mesh && !ops->leave_mesh);
+	WARN_ON(ops->start_p2p_device && !ops->stop_p2p_device);
+	WARN_ON(ops->start_ap && !ops->stop_ap);
+	WARN_ON(ops->join_ocb && !ops->leave_ocb);
+	WARN_ON(ops->suspend && !ops->resume);
+	WARN_ON(ops->sched_scan_start && !ops->sched_scan_stop);
+	WARN_ON(ops->remain_on_channel && !ops->cancel_remain_on_channel);
+	WARN_ON(ops->tdls_channel_switch && !ops->tdls_cancel_channel_switch);
+	WARN_ON(ops->add_tx_ts && !ops->del_tx_ts);
+	WARN_ON(ops->set_tx_power && !ops->get_tx_power);
+	WARN_ON(ops->set_antenna && !ops->get_antenna);
 
 	alloc_size = sizeof(*rdev) + sizeof_priv;
 
-- 
cgit v0.10.2


From 8ac3c70419176b0fbc4aeae30de661f690b2e3ae Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 18 Dec 2015 15:08:34 +0100
Subject: mac80211: refactor HT/VHT to chandef code

The station MLME and IBSS/mesh ones use entirely different
code for interpreting HT and VHT operation elements. Change
the code that interprets them a bit - it now modifies an
existing chandef - and use it also in the MLME code.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c
index f7fc0e0..e5e8a31 100644
--- a/net/mac80211/ibss.c
+++ b/net/mac80211/ibss.c
@@ -1050,9 +1050,8 @@ static void ieee80211_update_sta_info(struct ieee80211_sub_if_data *sdata,
 		struct cfg80211_chan_def chandef;
 		enum ieee80211_sta_rx_bandwidth bw = sta->sta.bandwidth;
 
-		ieee80211_ht_oper_to_chandef(channel,
-					     elems->ht_operation,
-					     &chandef);
+		cfg80211_chandef_create(&chandef, channel, NL80211_CHAN_NO_HT);
+		ieee80211_chandef_ht_oper(elems->ht_operation, &chandef);
 
 		memcpy(&htcap_ie, elems->ht_cap_elem, sizeof(htcap_ie));
 		rates_updated |= ieee80211_ht_cap_ie_to_sta_ht_cap(sdata, sband,
@@ -1066,9 +1065,8 @@ static void ieee80211_update_sta_info(struct ieee80211_sub_if_data *sdata,
 			struct ieee80211_vht_cap cap_ie;
 			struct ieee80211_sta_vht_cap cap = sta->sta.vht_cap;
 
-			ieee80211_vht_oper_to_chandef(channel,
-						      elems->vht_operation,
-						      &chandef);
+			ieee80211_chandef_vht_oper(elems->vht_operation,
+						   &chandef);
 			memcpy(&cap_ie, elems->vht_cap_elem, sizeof(cap_ie));
 			ieee80211_vht_cap_ie_to_sta_vht_cap(sdata, sband,
 							    &cap_ie, sta);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 747402d..9934447 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1988,12 +1988,10 @@ int ieee80211_add_ext_srates_ie(struct ieee80211_sub_if_data *sdata,
 u8 *ieee80211_add_wmm_info_ie(u8 *buf, u8 qosinfo);
 
 /* channel management */
-void ieee80211_ht_oper_to_chandef(struct ieee80211_channel *control_chan,
-				  const struct ieee80211_ht_operation *ht_oper,
-				  struct cfg80211_chan_def *chandef);
-void ieee80211_vht_oper_to_chandef(struct ieee80211_channel *control_chan,
-				   const struct ieee80211_vht_operation *oper,
-				   struct cfg80211_chan_def *chandef);
+bool ieee80211_chandef_ht_oper(const struct ieee80211_ht_operation *ht_oper,
+			       struct cfg80211_chan_def *chandef);
+bool ieee80211_chandef_vht_oper(const struct ieee80211_vht_operation *oper,
+				struct cfg80211_chan_def *chandef);
 u32 ieee80211_chandef_downgrade(struct cfg80211_chan_def *c);
 
 int __must_check
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index fa28500..9a8e7b5 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -91,11 +91,10 @@ bool mesh_matches_local(struct ieee80211_sub_if_data *sdata,
 	if (sdata->vif.bss_conf.basic_rates != basic_rates)
 		return false;
 
-	ieee80211_ht_oper_to_chandef(sdata->vif.bss_conf.chandef.chan,
-				     ie->ht_operation, &sta_chan_def);
-
-	ieee80211_vht_oper_to_chandef(sdata->vif.bss_conf.chandef.chan,
-				      ie->vht_operation, &sta_chan_def);
+	cfg80211_chandef_create(&sta_chan_def, sdata->vif.bss_conf.chandef.chan,
+				NL80211_CHAN_NO_HT);
+	ieee80211_chandef_ht_oper(ie->ht_operation, &sta_chan_def);
+	ieee80211_chandef_vht_oper(ie->vht_operation, &sta_chan_def);
 
 	if (!cfg80211_chandef_compatible(&sdata->vif.bss_conf.chandef,
 					 &sta_chan_def))
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 4af9b2b..e6d198b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -196,16 +196,7 @@ ieee80211_determine_chantype(struct ieee80211_sub_if_data *sdata,
 
 	/* check 40 MHz support, if we have it */
 	if (sta_ht_cap.cap & IEEE80211_HT_CAP_SUP_WIDTH_20_40) {
-		switch (ht_oper->ht_param & IEEE80211_HT_PARAM_CHA_SEC_OFFSET) {
-		case IEEE80211_HT_PARAM_CHA_SEC_ABOVE:
-			chandef->width = NL80211_CHAN_WIDTH_40;
-			chandef->center_freq1 += 10;
-			break;
-		case IEEE80211_HT_PARAM_CHA_SEC_BELOW:
-			chandef->width = NL80211_CHAN_WIDTH_40;
-			chandef->center_freq1 -= 10;
-			break;
-		}
+		ieee80211_chandef_ht_oper(ht_oper, chandef);
 	} else {
 		/* 40 MHz (and 80 MHz) must be supported for VHT */
 		ret = IEEE80211_STA_DISABLE_VHT;
@@ -219,35 +210,11 @@ ieee80211_determine_chantype(struct ieee80211_sub_if_data *sdata,
 		goto out;
 	}
 
-	vht_chandef.chan = channel;
-	vht_chandef.center_freq1 =
-		ieee80211_channel_to_frequency(vht_oper->center_freq_seg1_idx,
-					       channel->band);
-	vht_chandef.center_freq2 = 0;
-
-	switch (vht_oper->chan_width) {
-	case IEEE80211_VHT_CHANWIDTH_USE_HT:
-		vht_chandef.width = chandef->width;
-		vht_chandef.center_freq1 = chandef->center_freq1;
-		break;
-	case IEEE80211_VHT_CHANWIDTH_80MHZ:
-		vht_chandef.width = NL80211_CHAN_WIDTH_80;
-		break;
-	case IEEE80211_VHT_CHANWIDTH_160MHZ:
-		vht_chandef.width = NL80211_CHAN_WIDTH_160;
-		break;
-	case IEEE80211_VHT_CHANWIDTH_80P80MHZ:
-		vht_chandef.width = NL80211_CHAN_WIDTH_80P80;
-		vht_chandef.center_freq2 =
-			ieee80211_channel_to_frequency(
-				vht_oper->center_freq_seg2_idx,
-				channel->band);
-		break;
-	default:
+	vht_chandef = *chandef;
+	if (!ieee80211_chandef_vht_oper(vht_oper, &vht_chandef)) {
 		if (!(ifmgd->flags & IEEE80211_STA_DISABLE_VHT))
 			sdata_info(sdata,
-				   "AP VHT operation IE has invalid channel width (%d), disable VHT\n",
-				   vht_oper->chan_width);
+				   "AP VHT information is invalid, disable VHT\n");
 		ret = IEEE80211_STA_DISABLE_VHT;
 		goto out;
 	}
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index f4b2c04..7d0479e 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -2383,17 +2383,13 @@ u8 *ieee80211_ie_build_vht_oper(u8 *pos, struct ieee80211_sta_vht_cap *vht_cap,
 	return pos + sizeof(struct ieee80211_vht_operation);
 }
 
-void ieee80211_ht_oper_to_chandef(struct ieee80211_channel *control_chan,
-				  const struct ieee80211_ht_operation *ht_oper,
-				  struct cfg80211_chan_def *chandef)
+bool ieee80211_chandef_ht_oper(const struct ieee80211_ht_operation *ht_oper,
+			       struct cfg80211_chan_def *chandef)
 {
 	enum nl80211_channel_type channel_type;
 
-	if (!ht_oper) {
-		cfg80211_chandef_create(chandef, control_chan,
-					NL80211_CHAN_NO_HT);
-		return;
-	}
+	if (!ht_oper)
+		return false;
 
 	switch (ht_oper->ht_param & IEEE80211_HT_PARAM_CHA_SEC_OFFSET) {
 	case IEEE80211_HT_PARAM_CHA_SEC_NONE:
@@ -2407,42 +2403,52 @@ void ieee80211_ht_oper_to_chandef(struct ieee80211_channel *control_chan,
 		break;
 	default:
 		channel_type = NL80211_CHAN_NO_HT;
+		return false;
 	}
 
-	cfg80211_chandef_create(chandef, control_chan, channel_type);
+	cfg80211_chandef_create(chandef, chandef->chan, channel_type);
+	return true;
 }
 
-void ieee80211_vht_oper_to_chandef(struct ieee80211_channel *control_chan,
-				   const struct ieee80211_vht_operation *oper,
-				   struct cfg80211_chan_def *chandef)
+bool ieee80211_chandef_vht_oper(const struct ieee80211_vht_operation *oper,
+				struct cfg80211_chan_def *chandef)
 {
+	struct cfg80211_chan_def new = *chandef;
+	int cf1, cf2;
+
 	if (!oper)
-		return;
+		return false;
 
-	chandef->chan = control_chan;
+	cf1 = ieee80211_channel_to_frequency(oper->center_freq_seg1_idx,
+					     chandef->chan->band);
+	cf2 = ieee80211_channel_to_frequency(oper->center_freq_seg2_idx,
+					     chandef->chan->band);
 
 	switch (oper->chan_width) {
 	case IEEE80211_VHT_CHANWIDTH_USE_HT:
 		break;
 	case IEEE80211_VHT_CHANWIDTH_80MHZ:
-		chandef->width = NL80211_CHAN_WIDTH_80;
+		new.width = NL80211_CHAN_WIDTH_80;
+		new.center_freq1 = cf1;
 		break;
 	case IEEE80211_VHT_CHANWIDTH_160MHZ:
-		chandef->width = NL80211_CHAN_WIDTH_160;
+		new.width = NL80211_CHAN_WIDTH_160;
+		new.center_freq1 = cf1;
 		break;
 	case IEEE80211_VHT_CHANWIDTH_80P80MHZ:
-		chandef->width = NL80211_CHAN_WIDTH_80P80;
+		new.width = NL80211_CHAN_WIDTH_80P80;
+		new.center_freq1 = cf1;
+		new.center_freq2 = cf2;
 		break;
 	default:
-		break;
+		return false;
 	}
 
-	chandef->center_freq1 =
-		ieee80211_channel_to_frequency(oper->center_freq_seg1_idx,
-					       control_chan->band);
-	chandef->center_freq2 =
-		ieee80211_channel_to_frequency(oper->center_freq_seg2_idx,
-					       control_chan->band);
+	if (!cfg80211_chandef_valid(&new))
+		return false;
+
+	*chandef = new;
+	return true;
 }
 
 int ieee80211_parse_bitrates(struct cfg80211_chan_def *chandef,
-- 
cgit v0.10.2


From 538dc9045251d3d6b5c0216a5c61c32bd9cedac9 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn@kryo.se>
Date: Thu, 24 Dec 2015 00:33:26 -0800
Subject: mac80211: Make addr const in SET_IEEE80211_PERM_ADDR()

Make the addr parameter const in SET_IEEE80211_PERM_ADDR() to save
clients from having to cast away a const qualifier.

Signed-off-by: Bjorn Andersson <bjorn.andersson@sonymobile.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index df5698e..566df20 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -2217,7 +2217,7 @@ static inline void SET_IEEE80211_DEV(struct ieee80211_hw *hw, struct device *dev
  * @hw: the &struct ieee80211_hw to set the MAC address for
  * @addr: the address to set
  */
-static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr)
+static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, const u8 *addr)
 {
 	memcpy(hw->wiphy->perm_addr, addr, ETH_ALEN);
 }
-- 
cgit v0.10.2


From f06b7ab875037c3d99cd30a07c51caf34fbecb2c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 13 Jan 2016 14:21:01 +0100
Subject: mac80211_hwsim: remove shadowing variable

The function here already has a variable hdr that even
contains the right thing, so the inner scope's hdr variable
that's shadowing the outer one can just be removed.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index e31a94f..ee37af1 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -1333,10 +1333,8 @@ static void mac80211_hwsim_tx(struct ieee80211_hw *hw,
 	data->tx_bytes += skb->len;
 	ack = mac80211_hwsim_tx_frame_no_nl(hw, skb, channel);
 
-	if (ack && skb->len >= 16) {
-		struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
+	if (ack && skb->len >= 16)
 		mac80211_hwsim_monitor_ack(channel, hdr->addr2);
-	}
 
 	ieee80211_tx_info_clear_status(txi);
 
-- 
cgit v0.10.2


From dd21dfc645d5dce0657af78761b3fa11a3a95398 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 20 Jan 2016 10:39:23 +0100
Subject: rfkill: disentangle polling pause and suspend

When suspended while polling is paused, polling will erroneously
resume at resume time. Fix this by tracking pause and suspend in
separate state variable and adding the necessary checks.

Clarify the documentation on this as well.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index d901078..7af625f 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -104,7 +104,8 @@ int __must_check rfkill_register(struct rfkill *rfkill);
  *
  * Pause polling -- say transmitter is off for other reasons.
  * NOTE: not necessary for suspend/resume -- in that case the
- * core stops polling anyway
+ * core stops polling anyway (but will also correctly handle
+ * the case of polling having been paused before suspend.)
  */
 void rfkill_pause_polling(struct rfkill *rfkill);
 
diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index f53bf3b6..1664399 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -57,6 +57,8 @@ struct rfkill {
 
 	bool			registered;
 	bool			persistent;
+	bool			polling_paused;
+	bool			suspended;
 
 	const struct rfkill_ops	*ops;
 	void			*data;
@@ -786,6 +788,7 @@ void rfkill_pause_polling(struct rfkill *rfkill)
 	if (!rfkill->ops->poll)
 		return;
 
+	rfkill->polling_paused = true;
 	cancel_delayed_work_sync(&rfkill->poll_work);
 }
 EXPORT_SYMBOL(rfkill_pause_polling);
@@ -797,6 +800,11 @@ void rfkill_resume_polling(struct rfkill *rfkill)
 	if (!rfkill->ops->poll)
 		return;
 
+	rfkill->polling_paused = false;
+
+	if (rfkill->suspended)
+		return;
+
 	queue_delayed_work(system_power_efficient_wq,
 			   &rfkill->poll_work, 0);
 }
@@ -807,7 +815,8 @@ static int rfkill_suspend(struct device *dev)
 {
 	struct rfkill *rfkill = to_rfkill(dev);
 
-	rfkill_pause_polling(rfkill);
+	rfkill->suspended = true;
+	cancel_delayed_work_sync(&rfkill->poll_work);
 
 	return 0;
 }
@@ -817,12 +826,16 @@ static int rfkill_resume(struct device *dev)
 	struct rfkill *rfkill = to_rfkill(dev);
 	bool cur;
 
+	rfkill->suspended = false;
+
 	if (!rfkill->persistent) {
 		cur = !!(rfkill->state & RFKILL_BLOCK_SW);
 		rfkill_set_block(rfkill, cur);
 	}
 
-	rfkill_resume_polling(rfkill);
+	if (rfkill->ops->poll && !rfkill->polling_paused)
+		queue_delayed_work(system_power_efficient_wq,
+				   &rfkill->poll_work, 0);
 
 	return 0;
 }
-- 
cgit v0.10.2


From 573a2b51acb48e745094cf3cb942cf6a55d2bf30 Mon Sep 17 00:00:00 2001
From: Ola Olsson <ola1olsson@gmail.com>
Date: Sun, 10 Jan 2016 18:41:42 +0100
Subject: cfg80211: Fix some linguistics in Kconfig

Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/Kconfig b/net/wireless/Kconfig
index ec3bf30..6c60612 100644
--- a/net/wireless/Kconfig
+++ b/net/wireless/Kconfig
@@ -50,8 +50,8 @@ config CFG80211_DEVELOPER_WARNINGS
 	default n
 	help
 	  This option enables some additional warnings that help
-	  cfg80211 developers and driver developers, but that can
-	  trigger due to races with userspace.
+	  cfg80211 developers and driver developers, but beware that
+	  they can also trigger due to races with userspace.
 
 	  For example, when a driver reports that it was disconnected
 	  from the AP, but the user disconnects manually at the same
@@ -110,7 +110,7 @@ config CFG80211_REG_RELAX_NO_IR
 	 interface which associated to an AP which userspace assumes or confirms
 	 to be an authorized master, i.e., with radar detection support and DFS
 	 capabilities. However, note that in order to not create daisy chain
-	 scenarios, this relaxation is not allowed in cases that the BSS client
+	 scenarios, this relaxation is not allowed in cases where the BSS client
 	 is associated to P2P GO and in addition the P2P GO instantiated on
 	 a channel due to this relaxation should not allow connection from
 	 non P2P clients.
@@ -135,7 +135,7 @@ config CFG80211_DEBUGFS
 	depends on CFG80211
 	depends on DEBUG_FS
 	---help---
-	  You can enable this if you want to debugfs entries for cfg80211.
+	  You can enable this if you want debugfs entries for cfg80211.
 
 	  If unsure, say N.
 
@@ -146,7 +146,7 @@ config CFG80211_INTERNAL_REGDB
 	---help---
 	  This option generates an internal data structure representing
 	  the wireless regulatory rules described in net/wireless/db.txt
-	  and includes code to query that database.  This is an alternative
+	  and includes code to query that database. This is an alternative
 	  to using CRDA for defining regulatory rules for the kernel.
 
 	  Using this option requires some parsing of the db.txt at build time,
@@ -159,7 +159,7 @@ config CFG80211_INTERNAL_REGDB
 
 	  http://wireless.kernel.org/en/developers/Regulatory
 
-	  Most distributions have a CRDA package.  So if unsure, say N.
+	  Most distributions have a CRDA package. So if unsure, say N.
 
 config CFG80211_CRDA_SUPPORT
 	bool "support CRDA" if CFG80211_INTERNAL_REGDB
-- 
cgit v0.10.2


From f3e7fae248224b9441098e3c344b48b1cd1d9eca Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Tue, 19 Jan 2016 10:42:37 -0500
Subject: rfkill: use variable instead of duplicating the expression
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

RFKILL_BLOCK_SW value have just been saved to prev, no need to check it
again in the if expression. This makes code a little bit easier to read.

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 1664399..061ed37 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -287,7 +287,7 @@ static void rfkill_set_block(struct rfkill *rfkill, bool blocked)
 	spin_lock_irqsave(&rfkill->lock, flags);
 	prev = rfkill->state & RFKILL_BLOCK_SW;
 
-	if (rfkill->state & RFKILL_BLOCK_SW)
+	if (prev)
 		rfkill->state |= RFKILL_BLOCK_SW_PREV;
 	else
 		rfkill->state &= ~RFKILL_BLOCK_SW_PREV;
-- 
cgit v0.10.2


From 1926e260d8d377b95b57a49eefeb54ac42919f53 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Tue, 19 Jan 2016 10:42:38 -0500
Subject: rfkill: remove/inline __rfkill_set_hw_state
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

__rfkill_set_hw_state() is only one used in rfkill_set_hw_state(), and
none of them are long or complicated, so merging the two makes the code
easier to read.

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 061ed37..076590e 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -235,29 +235,6 @@ static void rfkill_event(struct rfkill *rfkill)
 	rfkill_send_events(rfkill, RFKILL_OP_CHANGE);
 }
 
-static bool __rfkill_set_hw_state(struct rfkill *rfkill,
-				  bool blocked, bool *change)
-{
-	unsigned long flags;
-	bool prev, any;
-
-	BUG_ON(!rfkill);
-
-	spin_lock_irqsave(&rfkill->lock, flags);
-	prev = !!(rfkill->state & RFKILL_BLOCK_HW);
-	if (blocked)
-		rfkill->state |= RFKILL_BLOCK_HW;
-	else
-		rfkill->state &= ~RFKILL_BLOCK_HW;
-	*change = prev != blocked;
-	any = !!(rfkill->state & RFKILL_BLOCK_ANY);
-	spin_unlock_irqrestore(&rfkill->lock, flags);
-
-	rfkill_led_trigger_event(rfkill);
-
-	return any;
-}
-
 /**
  * rfkill_set_block - wrapper for set_block method
  *
@@ -482,14 +459,26 @@ bool rfkill_get_global_sw_state(const enum rfkill_type type)
 
 bool rfkill_set_hw_state(struct rfkill *rfkill, bool blocked)
 {
-	bool ret, change;
+	unsigned long flags;
+	bool ret, prev;
+
+	BUG_ON(!rfkill);
+
+	spin_lock_irqsave(&rfkill->lock, flags);
+	prev = !!(rfkill->state & RFKILL_BLOCK_HW);
+	if (blocked)
+		rfkill->state |= RFKILL_BLOCK_HW;
+	else
+		rfkill->state &= ~RFKILL_BLOCK_HW;
+	ret = !!(rfkill->state & RFKILL_BLOCK_ANY);
+	spin_unlock_irqrestore(&rfkill->lock, flags);
 
-	ret = __rfkill_set_hw_state(rfkill, blocked, &change);
+	rfkill_led_trigger_event(rfkill);
 
 	if (!rfkill->registered)
 		return ret;
 
-	if (change)
+	if (prev != blocked)
 		schedule_work(&rfkill->uevent_work);
 
 	return ret;
-- 
cgit v0.10.2


From e2a35e89291d70d1c4668b3216f84ec740d36be3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Tue, 19 Jan 2016 10:42:39 -0500
Subject: rfkill: Remove obsolete "claim" sysfs interface
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This was scheduled to be removed in 2012 by:

 commit 69c86373c6ea1149aa559e6088362d58d8ec8835
 Author: florian@mickler.org <florian@mickler.org>
 Date:   Wed Feb 24 12:05:16 2010 +0100

     Document the rfkill sysfs ABI

     This moves sysfs ABI info from Documentation/rfkill.txt to the
     ABI subfolder and reformats it.

     This also schedules the deprecated sysfs parts to be removed in
     2012 (claim file) and 2014 (state file).

     Signed-off-by: Florian Mickler <florian@mickler.org>
     Signed-off-by: John W. Linville <linville@tuxdriver.com>

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/Documentation/ABI/obsolete/sysfs-class-rfkill b/Documentation/ABI/obsolete/sysfs-class-rfkill
index ff60ad9..e736d14 100644
--- a/Documentation/ABI/obsolete/sysfs-class-rfkill
+++ b/Documentation/ABI/obsolete/sysfs-class-rfkill
@@ -18,12 +18,3 @@ Values: 	A numeric value.
 		2: RFKILL_STATE_HARD_BLOCKED
 			transmitter is forced off by something outside of
 			the driver's control.
-
-What:		/sys/class/rfkill/rfkill[0-9]+/claim
-Date:		09-Jul-2007
-KernelVersion	v2.6.22
-Contact:	linux-wireless@vger.kernel.org
-Description:	This file is deprecated because there no longer is a way to
-		claim just control over a single rfkill instance.
-		This file is scheduled to be removed in 2012.
-Values: 	0: Kernel handles events
diff --git a/Documentation/ABI/removed/sysfs-class-rfkill b/Documentation/ABI/removed/sysfs-class-rfkill
new file mode 100644
index 0000000..3ce6231
--- /dev/null
+++ b/Documentation/ABI/removed/sysfs-class-rfkill
@@ -0,0 +1,13 @@
+rfkill - radio frequency (RF) connector kill switch support
+
+For details to this subsystem look at Documentation/rfkill.txt.
+
+What:		/sys/class/rfkill/rfkill[0-9]+/claim
+Date:		09-Jul-2007
+KernelVersion	v2.6.22
+Contact:	linux-wireless@vger.kernel.org
+Description:	This file was deprecated because there no longer was a way to
+		claim just control over a single rfkill instance.
+		This file was scheduled to be removed in 2012, and was removed
+		in 2016.
+Values: 	0: Kernel handles events
diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 076590e..a805831 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -311,8 +311,7 @@ static atomic_t rfkill_input_disabled = ATOMIC_INIT(0);
  * @blocked: the new state
  *
  * This function sets the state of all switches of given type,
- * unless a specific switch is claimed by userspace (in which case,
- * that switch is left alone) or suspended.
+ * unless a specific switch is suspended.
  *
  * Caller must have acquired rfkill_global_mutex.
  */
@@ -721,20 +720,12 @@ static ssize_t state_store(struct device *dev, struct device_attribute *attr,
 }
 static DEVICE_ATTR_RW(state);
 
-static ssize_t claim_show(struct device *dev, struct device_attribute *attr,
-			  char *buf)
-{
-	return sprintf(buf, "%d\n", 0);
-}
-static DEVICE_ATTR_RO(claim);
-
 static struct attribute *rfkill_dev_attrs[] = {
 	&dev_attr_name.attr,
 	&dev_attr_type.attr,
 	&dev_attr_index.attr,
 	&dev_attr_persistent.attr,
 	&dev_attr_state.attr,
-	&dev_attr_claim.attr,
 	&dev_attr_soft.attr,
 	&dev_attr_hard.attr,
 	NULL,
-- 
cgit v0.10.2


From d4634e8dea13ccc969dd3f33dab3873cfdf3bc51 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Tue, 19 Jan 2016 10:42:42 -0500
Subject: rfkill: Update userspace API documentation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a note to userspace on the effect of RFKILL_OP_CHANGE_ALL also
updating the default state for hotplugged devices.

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
[reword a bit]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/uapi/linux/rfkill.h b/include/uapi/linux/rfkill.h
index 058757f..2e00dce 100644
--- a/include/uapi/linux/rfkill.h
+++ b/include/uapi/linux/rfkill.h
@@ -59,6 +59,8 @@ enum rfkill_type {
  * @RFKILL_OP_DEL: a device was removed
  * @RFKILL_OP_CHANGE: a device's state changed -- userspace changes one device
  * @RFKILL_OP_CHANGE_ALL: userspace changes all devices (of a type, or all)
+ *	into a state, also updating the default state used for devices that
+ *	are hot-plugged later.
  */
 enum rfkill_operation {
 	RFKILL_OP_ADD = 0,
-- 
cgit v0.10.2


From b8631c003307bc7abf0180d950c42d9032ebda70 Mon Sep 17 00:00:00 2001
From: Sunil Shahu <shshahu@gmail.com>
Date: Thu, 21 Jan 2016 11:06:25 +0530
Subject: mac80211: mesh_plink: remove redundant sta_info check

Remove unnecessory "if" statement and club it with previos "if" block.

Signed-off-by: Sunil Shahu <shshahu@gmail.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c
index bd3d55e..a07e93c 100644
--- a/net/mac80211/mesh_plink.c
+++ b/net/mac80211/mesh_plink.c
@@ -976,6 +976,10 @@ mesh_plink_get_event(struct ieee80211_sub_if_data *sdata,
 			mpl_dbg(sdata, "Mesh plink error: no more free plinks\n");
 			goto out;
 		}
+
+		/* new matching peer */
+		event = OPN_ACPT;
+		goto out;
 	} else {
 		if (!test_sta_flag(sta, WLAN_STA_AUTH)) {
 			mpl_dbg(sdata, "Mesh plink: Action frame from non-authed peer\n");
@@ -985,12 +989,6 @@ mesh_plink_get_event(struct ieee80211_sub_if_data *sdata,
 			goto out;
 	}
 
-	/* new matching peer */
-	if (!sta) {
-		event = OPN_ACPT;
-		goto out;
-	}
-
 	switch (ftype) {
 	case WLAN_SP_MESH_PEERING_OPEN:
 		if (!matches_local)
-- 
cgit v0.10.2


From 2a58d42c1e018ad514d4e23fd33fb2ded95d3ee6 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Thu, 21 Jan 2016 14:23:07 +0100
Subject: mac80211: fix txq queue related crashes

The driver can access the queue simultanously
while mac80211 tears down the interface. Without
spinlock protection this could lead to corrupting
sk_buff_head and subsequently to an invalid
pointer dereference.

Fixes: ba8c3d6f16a1 ("mac80211: add an intermediate software queue implementation")
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 33ae3c8..0451f12 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -977,7 +977,10 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata,
 	if (sdata->vif.txq) {
 		struct txq_info *txqi = to_txq_info(sdata->vif.txq);
 
+		spin_lock_bh(&txqi->queue.lock);
 		ieee80211_purge_tx_queue(&local->hw, &txqi->queue);
+		spin_unlock_bh(&txqi->queue.lock);
+
 		atomic_set(&sdata->txqs_len[txqi->txq.ac], 0);
 	}
 
-- 
cgit v0.10.2


From cf44012810ccdd8fd947518e965cb04b7b8498be Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Mon, 25 Jan 2016 14:43:24 +0100
Subject: mac80211: fix unnecessary frame drops in mesh fwding

The ieee80211_queue_stopped() expects hw queue
number but it was given raw WMM AC number instead.

This could cause frame drops and problems with
traffic in some cases - most notably if driver
doesn't map AC numbers to queue numbers 1:1 and
uses ieee80211_stop_queues() and
ieee80211_wake_queue() only without ever calling
ieee80211_wake_queues().

On ath10k it was possible to hit this problem in
the following case:

  1. wlan0 uses queue 0
     (ath10k maps queues per vif)
  2. offchannel uses queue 15
  3. queues 1-14 are unused
  4. ieee80211_stop_queues()
  5. ieee80211_wake_queue(q=0)
  6. ieee80211_wake_queue(q=15)
     (other queues are not woken up because both
      driver and mac80211 know other queues are
      unused)
  7. ieee80211_rx_h_mesh_fwding()
  8. ieee80211_select_queue_80211() returns 2
  9. ieee80211_queue_stopped(q=2) returns true
 10. frame is dropped (oops!)

Fixes: d3c1597b8d1b ("mac80211: fix forwarded mesh frame queue mapping")
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index ae993ed..664e886 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2235,7 +2235,7 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx)
 	struct ieee80211_local *local = rx->local;
 	struct ieee80211_sub_if_data *sdata = rx->sdata;
 	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
-	u16 q, hdrlen;
+	u16 ac, q, hdrlen;
 
 	hdr = (struct ieee80211_hdr *) skb->data;
 	hdrlen = ieee80211_hdrlen(hdr->frame_control);
@@ -2304,7 +2304,8 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx)
 	    ether_addr_equal(sdata->vif.addr, hdr->addr3))
 		return RX_CONTINUE;
 
-	q = ieee80211_select_queue_80211(sdata, skb, hdr);
+	ac = ieee80211_select_queue_80211(sdata, skb, hdr);
+	q = sdata->vif.hw_queue[ac];
 	if (ieee80211_queue_stopped(&local->hw, q)) {
 		IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, dropped_frames_congestion);
 		return RX_DROP_MONITOR;
-- 
cgit v0.10.2


From f4a0f0c5264e72d9279fbf9cf48a061526e8f788 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 25 Jan 2016 15:46:34 +0200
Subject: mac80211: add RX_FLAG_MACTIME_PLCP_START

The timestamp given by iwlwifi is at the beginning of the
frame over the air, at (or during) the SYNC field. Allow
such timestamps to be given to mac80211, at least (for now)
for frames with non-HT/VHT preambles.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 566df20..31337f8 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1010,6 +1010,8 @@ ieee80211_tx_info_clear_status(struct ieee80211_tx_info *info)
  * @RX_FLAG_MACTIME_END: The timestamp passed in the RX status (@mactime
  *	field) is valid and contains the time the last symbol of the MPDU
  *	(including FCS) was received.
+ * @RX_FLAG_MACTIME_PLCP_START: The timestamp passed in the RX status (@mactime
+ *	field) is valid and contains the time the SYNC preamble was received.
  * @RX_FLAG_SHORTPRE: Short preamble was used for this frame
  * @RX_FLAG_HT: HT MCS was used and rate_idx is MCS index
  * @RX_FLAG_VHT: VHT MCS was used and rate_index is MCS index
@@ -1058,6 +1060,7 @@ ieee80211_tx_info_clear_status(struct ieee80211_tx_info *info)
 enum mac80211_rx_flags {
 	RX_FLAG_MMIC_ERROR		= BIT(0),
 	RX_FLAG_DECRYPTED		= BIT(1),
+	RX_FLAG_MACTIME_PLCP_START	= BIT(2),
 	RX_FLAG_MMIC_STRIPPED		= BIT(3),
 	RX_FLAG_IV_STRIPPED		= BIT(4),
 	RX_FLAG_FAILED_FCS_CRC		= BIT(5),
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 9934447..a29f61d 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1466,7 +1466,13 @@ ieee80211_have_rx_timestamp(struct ieee80211_rx_status *status)
 {
 	WARN_ON_ONCE(status->flag & RX_FLAG_MACTIME_START &&
 		     status->flag & RX_FLAG_MACTIME_END);
-	return status->flag & (RX_FLAG_MACTIME_START | RX_FLAG_MACTIME_END);
+	if (status->flag & (RX_FLAG_MACTIME_START | RX_FLAG_MACTIME_END))
+		return true;
+	/* can't handle HT/VHT preamble yet */
+	if (status->flag & RX_FLAG_MACTIME_PLCP_START &&
+	    !(status->flag & (RX_FLAG_HT | RX_FLAG_VHT)))
+		return true;
+	return false;
 }
 
 u64 ieee80211_calculate_rx_timestamp(struct ieee80211_local *local,
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 7d0479e..fb90d9c 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -4,7 +4,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007	Johannes Berg <johannes@sipsolutions.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
- * Copyright (C) 2015	Intel Deutschland GmbH
+ * Copyright (C) 2015-2016	Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -2671,6 +2671,18 @@ u64 ieee80211_calculate_rx_timestamp(struct ieee80211_local *local,
 		sband = local->hw.wiphy->bands[status->band];
 		bitrate = sband->bitrates[status->rate_idx].bitrate;
 		ri.legacy = DIV_ROUND_UP(bitrate, (1 << shift));
+
+		if (status->flag & RX_FLAG_MACTIME_PLCP_START) {
+			/* TODO: handle HT/VHT preambles */
+			if (status->band == IEEE80211_BAND_5GHZ) {
+				ts += 20 << shift;
+				mpdu_offset += 2;
+			} else if (status->flag & RX_FLAG_SHORTPRE) {
+				ts += 96;
+			} else {
+				ts += 192;
+			}
+		}
 	}
 
 	rate = cfg80211_calculate_bitrate(&ri);
-- 
cgit v0.10.2


From d321cd014e51baab475efbdec468255b9e0ec822 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 25 Jan 2016 15:46:35 +0200
Subject: mac80211: fix ibss scan parameters

When joining IBSS a full scan should be initiated in order to search
for existing cell, unless the fixed_channel parameter was set.
A default channel to create the IBSS on if no cell was found is
provided as well.
However - a scan is initiated only on the default channel provided
regardless of whether ifibss->fixed_channel is set or not, with the
obvious result of the cell not joining existing IBSS cell that is
on another channel.

Fixes: 76bed0f43b27 ("mac80211: IBSS fix scan request")
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c
index e5e8a31..9b98378 100644
--- a/net/mac80211/ibss.c
+++ b/net/mac80211/ibss.c
@@ -7,6 +7,7 @@
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2009, Johannes Berg <johannes@sipsolutions.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -1483,14 +1484,21 @@ static void ieee80211_sta_find_ibss(struct ieee80211_sub_if_data *sdata)
 
 		sdata_info(sdata, "Trigger new scan to find an IBSS to join\n");
 
-		num = ieee80211_ibss_setup_scan_channels(local->hw.wiphy,
-							 &ifibss->chandef,
-							 channels,
-							 ARRAY_SIZE(channels));
 		scan_width = cfg80211_chandef_to_scan_width(&ifibss->chandef);
-		ieee80211_request_ibss_scan(sdata, ifibss->ssid,
-					    ifibss->ssid_len, channels, num,
-					    scan_width);
+
+		if (ifibss->fixed_channel) {
+			num = ieee80211_ibss_setup_scan_channels(local->hw.wiphy,
+								 &ifibss->chandef,
+								 channels,
+								 ARRAY_SIZE(channels));
+			ieee80211_request_ibss_scan(sdata, ifibss->ssid,
+						    ifibss->ssid_len, channels,
+						    num, scan_width);
+		} else {
+			ieee80211_request_ibss_scan(sdata, ifibss->ssid,
+						    ifibss->ssid_len, NULL,
+						    0, scan_width);
+		}
 	} else {
 		int interval = IEEE80211_SCAN_INTERVAL;
 
-- 
cgit v0.10.2


From 89f774e6c41efdcede9d4b37867af5aabcc835d5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 25 Jan 2016 15:46:36 +0200
Subject: mac80211: always print a message when disconnecting

Make sure there's at least a debug message whenever the
connection to the AP is terminated.

Also change one message from wiphy_debug() to the common
mlme_dbg().

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e6d198b..857089d 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3544,6 +3544,9 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata,
 				elems.ht_cap_elem, elems.ht_operation,
 				elems.vht_operation, bssid, &changed)) {
 		mutex_unlock(&local->sta_mtx);
+		sdata_info(sdata,
+			   "failed to follow AP %pM bandwidth change, disconnect\n",
+			   bssid);
 		ieee80211_set_disassoc(sdata, IEEE80211_STYPE_DEAUTH,
 				       WLAN_REASON_DEAUTH_LEAVING,
 				       true, deauth_buf);
@@ -3919,11 +3922,9 @@ void ieee80211_sta_work(struct ieee80211_sub_if_data *sdata)
 			 * We actually lost the connection ... or did we?
 			 * Let's make sure!
 			 */
-			wiphy_debug(local->hw.wiphy,
-				    "%s: No probe response from AP %pM"
-				    " after %dms, disconnecting.\n",
-				    sdata->name,
-				    bssid, probe_wait_ms);
+			mlme_dbg(sdata,
+				 "No probe response from AP %pM after %dms, disconnecting.\n",
+				 bssid, probe_wait_ms);
 
 			ieee80211_sta_connection_lost(sdata, bssid,
 				WLAN_REASON_DISASSOC_DUE_TO_INACTIVITY, false);
@@ -4511,6 +4512,9 @@ int ieee80211_mgd_auth(struct ieee80211_sub_if_data *sdata,
 	if (ifmgd->associated) {
 		u8 frame_buf[IEEE80211_DEAUTH_FRAME_LEN];
 
+		sdata_info(sdata,
+			   "disconnect from AP %pM for new auth to %pM\n",
+			   ifmgd->associated->bssid, req->bss->bssid);
 		ieee80211_set_disassoc(sdata, IEEE80211_STYPE_DEAUTH,
 				       WLAN_REASON_UNSPECIFIED,
 				       false, frame_buf);
@@ -4579,6 +4583,9 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata,
 	if (ifmgd->associated) {
 		u8 frame_buf[IEEE80211_DEAUTH_FRAME_LEN];
 
+		sdata_info(sdata,
+			   "disconnect from AP %pM for new assoc to %pM\n",
+			   ifmgd->associated->bssid, req->bss->bssid);
 		ieee80211_set_disassoc(sdata, IEEE80211_STYPE_DEAUTH,
 				       WLAN_REASON_UNSPECIFIED,
 				       false, frame_buf);
-- 
cgit v0.10.2


From 0ef049dc1167fe834d0ad5d63f89eddc5c70f6e4 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 26 Jan 2016 23:05:31 +0100
Subject: mac80211: avoid excessive stack usage in sta_info

When CONFIG_OPTIMIZE_INLINING is set, the sta_info_insert_finish
function consumes more stack than normally, exceeding the
1024 byte limit on ARM:

net/mac80211/sta_info.c: In function 'sta_info_insert_finish':
net/mac80211/sta_info.c:561:1: error: the frame size of 1080 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]

It turns out that there are two functions that put a 'struct station_info'
on the stack: __sta_info_destroy_part2 and sta_info_insert_finish, and
this structure alone requires up to 792 bytes.

Hoping that both are called rarely enough, this replaces the
on-stack structure with a dynamic allocation, which unfortunately
requires some suboptimal error handling for out-of-memory.

The __sta_info_destroy_part2 function is actually affected by the
stack usage twice because it calls cfg80211_del_sta_sinfo(), which
has another instance of struct station_info on its stack.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 98b6218388e3 ("mac80211/cfg80211: add station events")
Fixes: 6f7a8d26e266 ("mac80211: send statistics with delete station event")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 7e007cf..b28e7a2 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -498,11 +498,17 @@ static int sta_info_insert_finish(struct sta_info *sta) __acquires(RCU)
 {
 	struct ieee80211_local *local = sta->local;
 	struct ieee80211_sub_if_data *sdata = sta->sdata;
-	struct station_info sinfo;
+	struct station_info *sinfo;
 	int err = 0;
 
 	lockdep_assert_held(&local->sta_mtx);
 
+	sinfo = kzalloc(sizeof(struct station_info), GFP_KERNEL);
+	if (!sinfo) {
+		err = -ENOMEM;
+		goto out_err;
+	}
+
 	/* check if STA exists already */
 	if (sta_info_get_bss(sdata, sta->sta.addr)) {
 		err = -EEXIST;
@@ -533,10 +539,9 @@ static int sta_info_insert_finish(struct sta_info *sta) __acquires(RCU)
 	ieee80211_sta_debugfs_add(sta);
 	rate_control_add_sta_debugfs(sta);
 
-	memset(&sinfo, 0, sizeof(sinfo));
-	sinfo.filled = 0;
-	sinfo.generation = local->sta_generation;
-	cfg80211_new_sta(sdata->dev, sta->sta.addr, &sinfo, GFP_KERNEL);
+	sinfo->generation = local->sta_generation;
+	cfg80211_new_sta(sdata->dev, sta->sta.addr, sinfo, GFP_KERNEL);
+	kfree(sinfo);
 
 	sta_dbg(sdata, "Inserted STA %pM\n", sta->sta.addr);
 
@@ -897,7 +902,7 @@ static void __sta_info_destroy_part2(struct sta_info *sta)
 {
 	struct ieee80211_local *local = sta->local;
 	struct ieee80211_sub_if_data *sdata = sta->sdata;
-	struct station_info sinfo = {};
+	struct station_info *sinfo;
 	int ret;
 
 	/*
@@ -935,8 +940,11 @@ static void __sta_info_destroy_part2(struct sta_info *sta)
 
 	sta_dbg(sdata, "Removed STA %pM\n", sta->sta.addr);
 
-	sta_set_sinfo(sta, &sinfo);
-	cfg80211_del_sta_sinfo(sdata->dev, sta->sta.addr, &sinfo, GFP_KERNEL);
+	sinfo = kzalloc(sizeof(*sinfo), GFP_KERNEL);
+	if (sinfo)
+		sta_set_sinfo(sta, sinfo);
+	cfg80211_del_sta_sinfo(sdata->dev, sta->sta.addr, sinfo, GFP_KERNEL);
+	kfree(sinfo);
 
 	rate_control_remove_sta_debugfs(sta);
 	ieee80211_sta_debugfs_remove(sta);
-- 
cgit v0.10.2


From 3f73fe9fd8703123faf663908923ca9e62cb984e Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Sat, 30 Jan 2016 13:17:03 -0500
Subject: mac80211: mesh: drop constant field mean_chain_len

The mean_chain_len field in struct mesh_table is copied whenever a
new mesh table is allocated, but only ever has the value 2 and is
never otherwise updated, so just remove it and use the related
define instead.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h
index a159634..d941f3a 100644
--- a/net/mac80211/mesh.h
+++ b/net/mac80211/mesh.h
@@ -137,8 +137,6 @@ struct mesh_path {
  * @copy_node: function to copy nodes of the table
  * @size_order: determines size of the table, there will be 2^size_order hash
  *	buckets
- * @mean_chain_len: maximum average length for the hash buckets' list, if it is
- *	reached, the table will grow
  * @known_gates: list of known mesh gates and their mpaths by the station. The
  * gate's mpath may or may not be resolved and active.
  *
@@ -154,7 +152,6 @@ struct mesh_table {
 	void (*free_node) (struct hlist_node *p, bool free_leafs);
 	int (*copy_node) (struct hlist_node *p, struct mesh_table *newtbl);
 	int size_order;
-	int mean_chain_len;
 	struct hlist_head *known_gates;
 	spinlock_t gates_lock;
 
diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c
index dadf8dc..c95bdb8 100644
--- a/net/mac80211/mesh_pathtbl.c
+++ b/net/mac80211/mesh_pathtbl.c
@@ -160,11 +160,10 @@ static int mesh_table_grow(struct mesh_table *oldtbl,
 	int i;
 
 	if (atomic_read(&oldtbl->entries)
-			< oldtbl->mean_chain_len * (oldtbl->hash_mask + 1))
+			< MEAN_CHAIN_LEN * (oldtbl->hash_mask + 1))
 		return -EAGAIN;
 
 	newtbl->free_node = oldtbl->free_node;
-	newtbl->mean_chain_len = oldtbl->mean_chain_len;
 	newtbl->copy_node = oldtbl->copy_node;
 	newtbl->known_gates = oldtbl->known_gates;
 	atomic_set(&newtbl->entries, atomic_read(&oldtbl->entries));
@@ -585,7 +584,7 @@ struct mesh_path *mesh_path_add(struct ieee80211_sub_if_data *sdata,
 
 	hlist_add_head_rcu(&new_node->list, bucket);
 	if (atomic_inc_return(&tbl->entries) >=
-	    tbl->mean_chain_len * (tbl->hash_mask + 1))
+	    MEAN_CHAIN_LEN * (tbl->hash_mask + 1))
 		grow = 1;
 
 	mesh_paths_generation++;
@@ -714,7 +713,7 @@ int mpp_path_add(struct ieee80211_sub_if_data *sdata,
 
 	hlist_add_head_rcu(&new_node->list, bucket);
 	if (atomic_inc_return(&tbl->entries) >=
-	    tbl->mean_chain_len * (tbl->hash_mask + 1))
+	    MEAN_CHAIN_LEN * (tbl->hash_mask + 1))
 		grow = 1;
 
 	spin_unlock(&tbl->hashwlock[hash_idx]);
@@ -1076,7 +1075,6 @@ int mesh_pathtbl_init(void)
 		return -ENOMEM;
 	tbl_path->free_node = &mesh_path_node_free;
 	tbl_path->copy_node = &mesh_path_node_copy;
-	tbl_path->mean_chain_len = MEAN_CHAIN_LEN;
 	tbl_path->known_gates = kzalloc(sizeof(struct hlist_head), GFP_ATOMIC);
 	if (!tbl_path->known_gates) {
 		ret = -ENOMEM;
@@ -1092,7 +1090,6 @@ int mesh_pathtbl_init(void)
 	}
 	tbl_mpp->free_node = &mesh_path_node_free;
 	tbl_mpp->copy_node = &mesh_path_node_copy;
-	tbl_mpp->mean_chain_len = MEAN_CHAIN_LEN;
 	tbl_mpp->known_gates = kzalloc(sizeof(struct hlist_head), GFP_ATOMIC);
 	if (!tbl_mpp->known_gates) {
 		ret = -ENOMEM;
-- 
cgit v0.10.2


From dfdfc2beb0dd7e3a067d2eeacb4623cb48e77658 Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Tue, 26 Jan 2016 17:11:13 +0100
Subject: mac80211: Parse legacy and HT rate in injected frames

Drivers/devices without their own rate control algorithm can get the
information what rates they should use from either the radiotap header of
injected frames or from the rate control algorithm. But the parsing of the
legacy rate information from the radiotap header was removed in commit
e6a9854b05c1 ("mac80211/drivers: rewrite the rate control API").

The removal of this feature heavily reduced the usefulness of frame
injection when wanting to simulate specific transmission behavior. Having
rate parsing together with MCS rates and retry support allows a fine
grained selection of the tx behavior of injected frames for these kind of
tests.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Cc: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/Documentation/networking/mac80211-injection.txt b/Documentation/networking/mac80211-injection.txt
index 3a93007..ec8f934 100644
--- a/Documentation/networking/mac80211-injection.txt
+++ b/Documentation/networking/mac80211-injection.txt
@@ -28,6 +28,23 @@ radiotap headers and used to control injection:
    IEEE80211_RADIOTAP_F_TX_NOACK: frame should be sent without waiting for
 				  an ACK even if it is a unicast frame
 
+ * IEEE80211_RADIOTAP_RATE
+
+   legacy rate for the transmission (only for devices without own rate control)
+
+ * IEEE80211_RADIOTAP_MCS
+
+   HT rate for the transmission (only for devices without own rate control).
+   Also some flags are parsed
+
+   IEEE80211_TX_RC_SHORT_GI: use short guard interval
+   IEEE80211_TX_RC_40_MHZ_WIDTH: send in HT40 mode
+
+ * IEEE80211_RADIOTAP_DATA_RETRIES
+
+   number of retries when either IEEE80211_RADIOTAP_RATE or
+   IEEE80211_RADIOTAP_MCS was used
+
 The injection code can also skip all other currently defined radiotap fields
 facilitating replay of captured radiotap headers directly.
 
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 31337f8..dbcd69a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -708,12 +708,14 @@ enum mac80211_tx_info_flags {
  *	protocol frame (e.g. EAP)
  * @IEEE80211_TX_CTRL_PS_RESPONSE: This frame is a response to a poll
  *	frame (PS-Poll or uAPSD).
+ * @IEEE80211_TX_CTRL_RATE_INJECT: This frame is injected with rate information
  *
  * These flags are used in tx_info->control.flags.
  */
 enum mac80211_tx_control_flags {
 	IEEE80211_TX_CTRL_PORT_CTRL_PROTO	= BIT(0),
 	IEEE80211_TX_CTRL_PS_RESPONSE		= BIT(1),
+	IEEE80211_TX_CTRL_RATE_INJECT		= BIT(2),
 };
 
 /*
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 3311ce0..723cd7a 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -710,6 +710,10 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
 
 	info->control.short_preamble = txrc.short_preamble;
 
+	/* don't ask rate control when rate already injected via radiotap */
+	if (info->control.flags & IEEE80211_TX_CTRL_RATE_INJECT)
+		return TX_CONTINUE;
+
 	if (tx->sta)
 		assoc = test_sta_flag(tx->sta, WLAN_STA_ASSOC);
 
@@ -1665,15 +1669,24 @@ void ieee80211_xmit(struct ieee80211_sub_if_data *sdata,
 	ieee80211_tx(sdata, sta, skb, false);
 }
 
-static bool ieee80211_parse_tx_radiotap(struct sk_buff *skb)
+static bool ieee80211_parse_tx_radiotap(struct ieee80211_local *local,
+					struct sk_buff *skb)
 {
 	struct ieee80211_radiotap_iterator iterator;
 	struct ieee80211_radiotap_header *rthdr =
 		(struct ieee80211_radiotap_header *) skb->data;
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_supported_band *sband =
+		local->hw.wiphy->bands[info->band];
 	int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len,
 						   NULL);
 	u16 txflags;
+	u16 rate = 0;
+	bool rate_found = false;
+	u8 rate_retries = 0;
+	u16 rate_flags = 0;
+	u8 mcs_known, mcs_flags;
+	int i;
 
 	info->flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT |
 		       IEEE80211_TX_CTL_DONTFRAG;
@@ -1724,6 +1737,35 @@ static bool ieee80211_parse_tx_radiotap(struct sk_buff *skb)
 				info->flags |= IEEE80211_TX_CTL_NO_ACK;
 			break;
 
+		case IEEE80211_RADIOTAP_RATE:
+			rate = *iterator.this_arg;
+			rate_flags = 0;
+			rate_found = true;
+			break;
+
+		case IEEE80211_RADIOTAP_DATA_RETRIES:
+			rate_retries = *iterator.this_arg;
+			break;
+
+		case IEEE80211_RADIOTAP_MCS:
+			mcs_known = iterator.this_arg[0];
+			mcs_flags = iterator.this_arg[1];
+			if (!(mcs_known & IEEE80211_RADIOTAP_MCS_HAVE_MCS))
+				break;
+
+			rate_found = true;
+			rate = iterator.this_arg[2];
+			rate_flags = IEEE80211_TX_RC_MCS;
+
+			if (mcs_known & IEEE80211_RADIOTAP_MCS_HAVE_GI &&
+			    mcs_flags & IEEE80211_RADIOTAP_MCS_SGI)
+				rate_flags |= IEEE80211_TX_RC_SHORT_GI;
+
+			if (mcs_known & IEEE80211_RADIOTAP_MCS_HAVE_BW &&
+			    mcs_flags & IEEE80211_RADIOTAP_MCS_BW_40)
+				rate_flags |= IEEE80211_TX_RC_40_MHZ_WIDTH;
+			break;
+
 		/*
 		 * Please update the file
 		 * Documentation/networking/mac80211-injection.txt
@@ -1738,6 +1780,32 @@ static bool ieee80211_parse_tx_radiotap(struct sk_buff *skb)
 	if (ret != -ENOENT) /* ie, if we didn't simply run out of fields */
 		return false;
 
+	if (rate_found) {
+		info->control.flags |= IEEE80211_TX_CTRL_RATE_INJECT;
+
+		for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) {
+			info->control.rates[i].idx = -1;
+			info->control.rates[i].flags = 0;
+			info->control.rates[i].count = 0;
+		}
+
+		if (rate_flags & IEEE80211_TX_RC_MCS) {
+			info->control.rates[0].idx = rate;
+		} else {
+			for (i = 0; i < sband->n_bitrates; i++) {
+				if (rate * 5 != sband->bitrates[i].bitrate)
+					continue;
+
+				info->control.rates[0].idx = i;
+				break;
+			}
+		}
+
+		info->control.rates[0].flags = rate_flags;
+		info->control.rates[0].count = min_t(u8, rate_retries + 1,
+						     local->hw.max_rate_tries);
+	}
+
 	/*
 	 * remove the radiotap header
 	 * iterator->_max_length was sanity-checked against
@@ -1819,7 +1887,7 @@ netdev_tx_t ieee80211_monitor_start_xmit(struct sk_buff *skb,
 		      IEEE80211_TX_CTL_INJECTED;
 
 	/* process and remove the injection radiotap header */
-	if (!ieee80211_parse_tx_radiotap(skb))
+	if (!ieee80211_parse_tx_radiotap(local, skb))
 		goto fail;
 
 	rcu_read_lock();
-- 
cgit v0.10.2


From f2ac7e301ae6397669ff3f79e691942a9b5d2f39 Mon Sep 17 00:00:00 2001
From: Michal Kazior <michal.kazior@tieto.com>
Date: Wed, 27 Jan 2016 15:26:12 +0100
Subject: mac80211: expose txq queue depth and size to drivers

This will allow drivers to make more educated
decisions whether to defer transmission or not.

Relying on wake_tx_queue() call count implicitly
was not possible because it could be called
without queued frame count actually changing on
software tx aggregation start/stop code paths.

It was also not possible to know how long
byte-wise queue was without dequeueing.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index dbcd69a..fd35fc4 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -5596,4 +5596,19 @@ void ieee80211_unreserve_tid(struct ieee80211_sta *sta, u8 tid);
  */
 struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw,
 				     struct ieee80211_txq *txq);
+
+/**
+ * ieee80211_txq_get_depth - get pending frame/byte count of given txq
+ *
+ * The values are not guaranteed to be coherent with regard to each other, i.e.
+ * txq state can change half-way of this function and the caller may end up
+ * with "new" frame_cnt and "old" byte_cnt or vice-versa.
+ *
+ * @txq: pointer obtained from station or virtual interface
+ * @frame_cnt: pointer to store frame count
+ * @byte_cnt: pointer to store byte count
+ */
+void ieee80211_txq_get_depth(struct ieee80211_txq *txq,
+			     unsigned long *frame_cnt,
+			     unsigned long *byte_cnt);
 #endif /* MAC80211_H */
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index a29f61d..a96f8c0 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -804,6 +804,7 @@ enum txq_info_flags {
 struct txq_info {
 	struct sk_buff_head queue;
 	unsigned long flags;
+	unsigned long byte_cnt;
 
 	/* keep last! */
 	struct ieee80211_txq txq;
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 0451f12..453b4e74 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -979,6 +979,7 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata,
 
 		spin_lock_bh(&txqi->queue.lock);
 		ieee80211_purge_tx_queue(&local->hw, &txqi->queue);
+		txqi->byte_cnt = 0;
 		spin_unlock_bh(&txqi->queue.lock);
 
 		atomic_set(&sdata->txqs_len[txqi->txq.ac], 0);
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index b28e7a2..5894c0a 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -116,6 +116,7 @@ static void __cleanup_single_sta(struct sta_info *sta)
 
 			ieee80211_purge_tx_queue(&local->hw, &txqi->queue);
 			atomic_sub(n, &sdata->txqs_len[txqi->txq.ac]);
+			txqi->byte_cnt = 0;
 		}
 	}
 
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 723cd7a..a5aa275 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -1270,7 +1270,11 @@ static void ieee80211_drv_tx(struct ieee80211_local *local,
 	if (atomic_read(&sdata->txqs_len[ac]) >= local->hw.txq_ac_max_pending)
 		netif_stop_subqueue(sdata->dev, ac);
 
-	skb_queue_tail(&txqi->queue, skb);
+	spin_lock_bh(&txqi->queue.lock);
+	txqi->byte_cnt += skb->len;
+	__skb_queue_tail(&txqi->queue, skb);
+	spin_unlock_bh(&txqi->queue.lock);
+
 	drv_wake_tx_queue(local, txqi);
 
 	return;
@@ -1298,6 +1302,8 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw,
 	if (!skb)
 		goto out;
 
+	txqi->byte_cnt -= skb->len;
+
 	atomic_dec(&sdata->txqs_len[ac]);
 	if (__netif_subqueue_stopped(sdata->dev, ac))
 		ieee80211_propagate_queue_wake(local, sdata->vif.hw_queue[ac]);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index fb90d9c..091f3dd 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3368,3 +3368,17 @@ void ieee80211_init_tx_queue(struct ieee80211_sub_if_data *sdata,
 		txqi->txq.ac = IEEE80211_AC_BE;
 	}
 }
+
+void ieee80211_txq_get_depth(struct ieee80211_txq *txq,
+			     unsigned long *frame_cnt,
+			     unsigned long *byte_cnt)
+{
+	struct txq_info *txqi = to_txq_info(txq);
+
+	if (frame_cnt)
+		*frame_cnt = txqi->queue.qlen;
+
+	if (byte_cnt)
+		*byte_cnt = txqi->byte_cnt;
+}
+EXPORT_SYMBOL(ieee80211_txq_get_depth);
-- 
cgit v0.10.2


From fb4ea054857ec719dba84c3b5a7c4ee50d42bd5b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 28 Jan 2016 16:19:24 +0200
Subject: mac80211: change ieee80211_rx_reorder_ready() arguments

Clean up ieee80211_rx_reorder_ready() callers by passing the RX
TID struct and the index, instead of the frames list. This will
make it more extensible as well.

While at it, move the inline to rx.c as it's only used there.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index a96f8c0..a49c103 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1838,20 +1838,6 @@ static inline void ieee802_11_parse_elems(const u8 *start, size_t len,
 	ieee802_11_parse_elems_crc(start, len, action, elems, 0, 0);
 }
 
-static inline bool ieee80211_rx_reorder_ready(struct sk_buff_head *frames)
-{
-	struct sk_buff *tail = skb_peek_tail(frames);
-	struct ieee80211_rx_status *status;
-
-	if (!tail)
-		return false;
-
-	status = IEEE80211_SKB_RXCB(tail);
-	if (status->flag & RX_FLAG_AMSDU_MORE)
-		return false;
-
-	return true;
-}
 
 extern const int ieee802_1d_to_ac[8];
 
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 664e886..1153871 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -4,6 +4,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007-2010	Johannes Berg <johannes@sipsolutions.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -798,6 +799,23 @@ static ieee80211_rx_result ieee80211_rx_mesh_check(struct ieee80211_rx_data *rx)
 	return RX_CONTINUE;
 }
 
+static inline bool ieee80211_rx_reorder_ready(struct tid_ampdu_rx *tid_agg_rx,
+					      int index)
+{
+	struct sk_buff_head *frames = &tid_agg_rx->reorder_buf[index];
+	struct sk_buff *tail = skb_peek_tail(frames);
+	struct ieee80211_rx_status *status;
+
+	if (!tail)
+		return false;
+
+	status = IEEE80211_SKB_RXCB(tail);
+	if (status->flag & RX_FLAG_AMSDU_MORE)
+		return false;
+
+	return true;
+}
+
 static void ieee80211_release_reorder_frame(struct ieee80211_sub_if_data *sdata,
 					    struct tid_ampdu_rx *tid_agg_rx,
 					    int index,
@@ -812,7 +830,7 @@ static void ieee80211_release_reorder_frame(struct ieee80211_sub_if_data *sdata,
 	if (skb_queue_empty(skb_list))
 		goto no_frame;
 
-	if (!ieee80211_rx_reorder_ready(skb_list)) {
+	if (!ieee80211_rx_reorder_ready(tid_agg_rx, index)) {
 		__skb_queue_purge(skb_list);
 		goto no_frame;
 	}
@@ -866,7 +884,7 @@ static void ieee80211_sta_reorder_release(struct ieee80211_sub_if_data *sdata,
 
 	/* release the buffer until next missing frame */
 	index = tid_agg_rx->head_seq_num % tid_agg_rx->buf_size;
-	if (!ieee80211_rx_reorder_ready(&tid_agg_rx->reorder_buf[index]) &&
+	if (!ieee80211_rx_reorder_ready(tid_agg_rx, index) &&
 	    tid_agg_rx->stored_mpdu_num) {
 		/*
 		 * No buffers ready to be released, but check whether any
@@ -875,8 +893,7 @@ static void ieee80211_sta_reorder_release(struct ieee80211_sub_if_data *sdata,
 		int skipped = 1;
 		for (j = (index + 1) % tid_agg_rx->buf_size; j != index;
 		     j = (j + 1) % tid_agg_rx->buf_size) {
-			if (!ieee80211_rx_reorder_ready(
-					&tid_agg_rx->reorder_buf[j])) {
+			if (!ieee80211_rx_reorder_ready(tid_agg_rx, j)) {
 				skipped++;
 				continue;
 			}
@@ -903,8 +920,7 @@ static void ieee80211_sta_reorder_release(struct ieee80211_sub_if_data *sdata,
 				 skipped) & IEEE80211_SN_MASK;
 			skipped = 0;
 		}
-	} else while (ieee80211_rx_reorder_ready(
-				&tid_agg_rx->reorder_buf[index])) {
+	} else while (ieee80211_rx_reorder_ready(tid_agg_rx, index)) {
 		ieee80211_release_reorder_frame(sdata, tid_agg_rx, index,
 						frames);
 		index =	tid_agg_rx->head_seq_num % tid_agg_rx->buf_size;
@@ -915,8 +931,7 @@ static void ieee80211_sta_reorder_release(struct ieee80211_sub_if_data *sdata,
 
 		for (; j != (index - 1) % tid_agg_rx->buf_size;
 		     j = (j + 1) % tid_agg_rx->buf_size) {
-			if (ieee80211_rx_reorder_ready(
-					&tid_agg_rx->reorder_buf[j]))
+			if (ieee80211_rx_reorder_ready(tid_agg_rx, j))
 				break;
 		}
 
@@ -987,7 +1002,7 @@ static bool ieee80211_sta_manage_reorder_buf(struct ieee80211_sub_if_data *sdata
 	index = mpdu_seq_num % tid_agg_rx->buf_size;
 
 	/* check if we already stored this frame */
-	if (ieee80211_rx_reorder_ready(&tid_agg_rx->reorder_buf[index])) {
+	if (ieee80211_rx_reorder_ready(tid_agg_rx, index)) {
 		dev_kfree_skb(skb);
 		goto out;
 	}
-- 
cgit v0.10.2


From 06470f7468c8b6c95e72ebda803a61a99f4ee446 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Thu, 28 Jan 2016 16:19:25 +0200
Subject: mac80211: add API to allow filtering frames in BA sessions

If any frames are dropped that are part of a BA session, the reorder
buffer will "indefinitely" (until the timeout) wait for them to come
in (or a BAR moving the window) and won't release frames after them.
This means it isn't possible to filter frames within a BA session in
firmware.

Introduce an API function that allows such filtering. Calling this
function will move the BA window forward to the new SSN, and allows
marking frames after the SSN as having been filtered, so any future
reordering activity will release frames while skipping the holes.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index fd35fc4..5714774 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -5,7 +5,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007-2010	Johannes Berg <johannes@sipsolutions.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
- * Copyright (C) 2015 Intel Deutschland GmbH
+ * Copyright (C) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -5194,6 +5194,24 @@ void ieee80211_stop_rx_ba_session(struct ieee80211_vif *vif, u16 ba_rx_bitmap,
 				  const u8 *addr);
 
 /**
+ * ieee80211_mark_rx_ba_filtered_frames - move RX BA window and mark filtered
+ * @pubsta: station struct
+ * @tid: the session's TID
+ * @ssn: starting sequence number of the bitmap, all frames before this are
+ *	assumed to be out of the window after the call
+ * @filtered: bitmap of filtered frames, BIT(0) is the @ssn entry etc.
+ * @received_mpdus: number of received mpdus in firmware
+ *
+ * This function moves the BA window and releases all frames before @ssn, and
+ * marks frames marked in the bitmap as having been filtered. Afterwards, it
+ * checks if any frames in the window starting from @ssn can now be released
+ * (in case they were only waiting for frames that were filtered.)
+ */
+void ieee80211_mark_rx_ba_filtered_frames(struct ieee80211_sta *pubsta, u8 tid,
+					  u16 ssn, u64 filtered,
+					  u16 received_mpdus);
+
+/**
  * ieee80211_send_bar - send a BlockAckReq frame
  *
  * can be used to flush pending frames from the peer's aggregation reorder
diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c
index 2ab5479..1b8a5ca 100644
--- a/net/mac80211/agg-rx.c
+++ b/net/mac80211/agg-rx.c
@@ -376,6 +376,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 	tid_agg_rx->timeout = timeout;
 	tid_agg_rx->stored_mpdu_num = 0;
 	tid_agg_rx->auto_seq = auto_seq;
+	tid_agg_rx->reorder_buf_filtered = 0;
 	status = WLAN_STATUS_SUCCESS;
 
 	/* activate it for RX */
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 1153871..9fb7074 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -19,6 +19,7 @@
 #include <linux/etherdevice.h>
 #include <linux/rcupdate.h>
 #include <linux/export.h>
+#include <linux/bitops.h>
 #include <net/mac80211.h>
 #include <net/ieee80211_radiotap.h>
 #include <asm/unaligned.h>
@@ -806,6 +807,9 @@ static inline bool ieee80211_rx_reorder_ready(struct tid_ampdu_rx *tid_agg_rx,
 	struct sk_buff *tail = skb_peek_tail(frames);
 	struct ieee80211_rx_status *status;
 
+	if (tid_agg_rx->reorder_buf_filtered & BIT_ULL(index))
+		return true;
+
 	if (!tail)
 		return false;
 
@@ -844,6 +848,7 @@ static void ieee80211_release_reorder_frame(struct ieee80211_sub_if_data *sdata,
 	}
 
 no_frame:
+	tid_agg_rx->reorder_buf_filtered &= ~BIT_ULL(index);
 	tid_agg_rx->head_seq_num = ieee80211_sn_inc(tid_agg_rx->head_seq_num);
 }
 
@@ -3300,6 +3305,85 @@ void ieee80211_release_reorder_timeout(struct sta_info *sta, int tid)
 	ieee80211_rx_handlers(&rx, &frames);
 }
 
+void ieee80211_mark_rx_ba_filtered_frames(struct ieee80211_sta *pubsta, u8 tid,
+					  u16 ssn, u64 filtered,
+					  u16 received_mpdus)
+{
+	struct sta_info *sta;
+	struct tid_ampdu_rx *tid_agg_rx;
+	struct sk_buff_head frames;
+	struct ieee80211_rx_data rx = {
+		/* This is OK -- must be QoS data frame */
+		.security_idx = tid,
+		.seqno_idx = tid,
+	};
+	int i, diff;
+
+	if (WARN_ON(!pubsta || tid >= IEEE80211_NUM_TIDS))
+		return;
+
+	__skb_queue_head_init(&frames);
+
+	sta = container_of(pubsta, struct sta_info, sta);
+
+	rx.sta = sta;
+	rx.sdata = sta->sdata;
+	rx.local = sta->local;
+
+	rcu_read_lock();
+	tid_agg_rx = rcu_dereference(sta->ampdu_mlme.tid_rx[tid]);
+	if (!tid_agg_rx)
+		goto out;
+
+	spin_lock_bh(&tid_agg_rx->reorder_lock);
+
+	if (received_mpdus >= IEEE80211_SN_MODULO >> 1) {
+		int release;
+
+		/* release all frames in the reorder buffer */
+		release = (tid_agg_rx->head_seq_num + tid_agg_rx->buf_size) %
+			   IEEE80211_SN_MODULO;
+		ieee80211_release_reorder_frames(sta->sdata, tid_agg_rx,
+						 release, &frames);
+		/* update ssn to match received ssn */
+		tid_agg_rx->head_seq_num = ssn;
+	} else {
+		ieee80211_release_reorder_frames(sta->sdata, tid_agg_rx, ssn,
+						 &frames);
+	}
+
+	/* handle the case that received ssn is behind the mac ssn.
+	 * it can be tid_agg_rx->buf_size behind and still be valid */
+	diff = (tid_agg_rx->head_seq_num - ssn) & IEEE80211_SN_MASK;
+	if (diff >= tid_agg_rx->buf_size) {
+		tid_agg_rx->reorder_buf_filtered = 0;
+		goto release;
+	}
+	filtered = filtered >> diff;
+	ssn += diff;
+
+	/* update bitmap */
+	for (i = 0; i < tid_agg_rx->buf_size; i++) {
+		int index = (ssn + i) % tid_agg_rx->buf_size;
+
+		tid_agg_rx->reorder_buf_filtered &= ~BIT_ULL(index);
+		if (filtered & BIT_ULL(i))
+			tid_agg_rx->reorder_buf_filtered |= BIT_ULL(index);
+	}
+
+	/* now process also frames that the filter marking released */
+	ieee80211_sta_reorder_release(sta->sdata, tid_agg_rx, &frames);
+
+release:
+	spin_unlock_bh(&tid_agg_rx->reorder_lock);
+
+	ieee80211_rx_handlers(&rx, &frames);
+
+ out:
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL(ieee80211_mark_rx_ba_filtered_frames);
+
 /* main receive path */
 
 static bool ieee80211_accept_frame(struct ieee80211_rx_data *rx)
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index f4d3899..053f5c4 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -168,6 +168,8 @@ struct tid_ampdu_tx {
  *
  * @reorder_buf: buffer to reorder incoming aggregated MPDUs. An MPDU may be an
  *	A-MSDU with individually reported subframes.
+ * @reorder_buf_filtered: bitmap indicating where there are filtered frames in
+ *	the reorder buffer that should be ignored when releasing frames
  * @reorder_time: jiffies when skb was added
  * @session_timer: check if peer keeps Tx-ing on the TID (by timeout value)
  * @reorder_timer: releases expired frames from the reorder buffer.
@@ -195,6 +197,7 @@ struct tid_ampdu_tx {
 struct tid_ampdu_rx {
 	struct rcu_head rcu_head;
 	spinlock_t reorder_lock;
+	u64 reorder_buf_filtered;
 	struct sk_buff_head *reorder_buf;
 	unsigned long *reorder_time;
 	struct timer_list session_timer;
-- 
cgit v0.10.2


From 23665aaf9170ae6328cc4f68250c529a628af2ab Mon Sep 17 00:00:00 2001
From: Jouni Malinen <jouni@qca.qualcomm.com>
Date: Mon, 1 Feb 2016 11:40:55 +0200
Subject: mac80211: Interoperability workaround for 80+80 and 160 MHz channels

Number of deployed 80 MHz capable VHT stations that do not support 80+80
and 160 MHz bandwidths seem to misbehave when trying to connect to an AP
that advertises 80+80 or 160 MHz channel bandwidth in the VHT Operation
element. To avoid such issues with deployed devices, modify the design
based on recently accepted IEEE 802.11 standard changes (*).

This allows poorly implemented VHT 80 MHz stations to connect with the
AP in 80 MHz mode. 80+80 and 160 MHz capable stations need to support
the new workaround mechanism to allow full bandwidth to be used.
However, there are more or less no impacted station with 80+80/160
capability deployed.

The rebased version of this patch is based on the updated version from
Johannes Berg to take the HT/VHT chandef refactoring into account.

(*) Changes in
https://mentor.ieee.org/802.11/dcn/15/11-15-1530-04-000m-vht160-operation-signaling-through-non-zero-ccfs1.docx
were accepted during the IEEE 802.11 January 2016 meeting.

Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 091f3dd..f1e5b76 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -2364,10 +2364,23 @@ u8 *ieee80211_ie_build_vht_oper(u8 *pos, struct ieee80211_sta_vht_cap *vht_cap,
 
 	switch (chandef->width) {
 	case NL80211_CHAN_WIDTH_160:
-		vht_oper->chan_width = IEEE80211_VHT_CHANWIDTH_160MHZ;
+		/*
+		 * Convert 160 MHz channel width to new style as interop
+		 * workaround.
+		 */
+		vht_oper->chan_width = IEEE80211_VHT_CHANWIDTH_80MHZ;
+		vht_oper->center_freq_seg2_idx = vht_oper->center_freq_seg1_idx;
+		if (chandef->chan->center_freq < chandef->center_freq1)
+			vht_oper->center_freq_seg1_idx -= 8;
+		else
+			vht_oper->center_freq_seg1_idx += 8;
 		break;
 	case NL80211_CHAN_WIDTH_80P80:
-		vht_oper->chan_width = IEEE80211_VHT_CHANWIDTH_80P80MHZ;
+		/*
+		 * Convert 80+80 MHz channel width to new style as interop
+		 * workaround.
+		 */
+		vht_oper->chan_width = IEEE80211_VHT_CHANWIDTH_80MHZ;
 		break;
 	case NL80211_CHAN_WIDTH_80:
 		vht_oper->chan_width = IEEE80211_VHT_CHANWIDTH_80MHZ;
@@ -2430,6 +2443,20 @@ bool ieee80211_chandef_vht_oper(const struct ieee80211_vht_operation *oper,
 	case IEEE80211_VHT_CHANWIDTH_80MHZ:
 		new.width = NL80211_CHAN_WIDTH_80;
 		new.center_freq1 = cf1;
+		/* If needed, adjust based on the newer interop workaround. */
+		if (oper->center_freq_seg2_idx) {
+			unsigned int diff;
+
+			diff = abs(oper->center_freq_seg2_idx -
+				   oper->center_freq_seg1_idx);
+			if (diff == 8) {
+				new.width = NL80211_CHAN_WIDTH_160;
+				new.center_freq1 = cf2;
+			} else if (diff > 8) {
+				new.width = NL80211_CHAN_WIDTH_80P80;
+				new.center_freq2 = cf2;
+			}
+		}
 		break;
 	case IEEE80211_VHT_CHANWIDTH_160MHZ:
 		new.width = NL80211_CHAN_WIDTH_160;
-- 
cgit v0.10.2


From ea32f065bd3e3e09f0bcb3042f1664caf6b3e233 Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudip@vectorindia.org>
Date: Tue, 2 Feb 2016 13:21:14 +0530
Subject: mac80211: fix memory leak

On error we jumped to the error label and returned the error code but we
missed releasing sinfo.

Fixes: 5fe74014172d ("mac80211: avoid excessive stack usage in sta_info")
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 5894c0a..87b7e7a 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -562,6 +562,7 @@ static int sta_info_insert_finish(struct sta_info *sta) __acquires(RCU)
 	__cleanup_single_sta(sta);
  out_err:
 	mutex_unlock(&local->sta_mtx);
+	kfree(sinfo);
 	rcu_read_lock();
 	return err;
 }
-- 
cgit v0.10.2


From 88665f5a797a832ff7926ad2287ee41738bc09b9 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Tue, 2 Feb 2016 14:39:08 +0100
Subject: mac80211: move A-MSDU skb_linearize call to ieee80211_amsdu_to_8023s

Prepararation for zero-copy A-MSDU support with page fragment SKBs

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 9fb7074..4419214 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2223,9 +2223,6 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx)
 	skb->dev = dev;
 	__skb_queue_head_init(&frame_list);
 
-	if (skb_linearize(skb))
-		return RX_DROP_UNUSABLE;
-
 	ieee80211_amsdu_to_8023s(skb, &frame_list, dev->dev_addr,
 				 rx->sdata->vif.type,
 				 rx->local->hw.extra_tx_headroom, true);
diff --git a/net/wireless/util.c b/net/wireless/util.c
index 9277042..6a40769 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -657,6 +657,9 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 	int remaining, err;
 	u8 dst[ETH_ALEN], src[ETH_ALEN];
 
+	if (skb_linearize(skb))
+		goto out;
+
 	if (has_80211_header) {
 		err = ieee80211_data_to_8023(skb, addr, iftype);
 		if (err)
-- 
cgit v0.10.2


From 2d1c304cb2d5cf28d8362496bd9ea6aadf4f01d2 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Tue, 2 Feb 2016 14:39:09 +0100
Subject: cfg80211: add function for 802.3 conversion with separate output
 buffer

Use skb_copy_bits in preparation for allowing fragmented skbs

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/util.c b/net/wireless/util.c
index 6a40769..f75f601 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -393,9 +393,9 @@ unsigned int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb)
 }
 EXPORT_SYMBOL(ieee80211_get_hdrlen_from_skb);
 
-unsigned int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr)
+static unsigned int __ieee80211_get_mesh_hdrlen(u8 flags)
 {
-	int ae = meshhdr->flags & MESH_FLAGS_AE;
+	int ae = flags & MESH_FLAGS_AE;
 	/* 802.11-2012, 8.2.4.7.3 */
 	switch (ae) {
 	default:
@@ -407,21 +407,31 @@ unsigned int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr)
 		return 18;
 	}
 }
+
+unsigned int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr)
+{
+	return __ieee80211_get_mesh_hdrlen(meshhdr->flags);
+}
 EXPORT_SYMBOL(ieee80211_get_mesh_hdrlen);
 
-int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
-			   enum nl80211_iftype iftype)
+static int __ieee80211_data_to_8023(struct sk_buff *skb, struct ethhdr *ehdr,
+				    const u8 *addr, enum nl80211_iftype iftype)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
-	u16 hdrlen, ethertype;
-	u8 *payload;
-	u8 dst[ETH_ALEN];
-	u8 src[ETH_ALEN] __aligned(2);
+	struct {
+		u8 hdr[ETH_ALEN] __aligned(2);
+		__be16 proto;
+	} payload;
+	struct ethhdr tmp;
+	u16 hdrlen;
+	u8 mesh_flags = 0;
 
 	if (unlikely(!ieee80211_is_data_present(hdr->frame_control)))
 		return -1;
 
 	hdrlen = ieee80211_hdrlen(hdr->frame_control);
+	if (skb->len < hdrlen + 8)
+		return -1;
 
 	/* convert IEEE 802.11 header + possible LLC headers into Ethernet
 	 * header
@@ -432,8 +442,11 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
 	 *   1     0   BSSID SA    DA    n/a
 	 *   1     1   RA    TA    DA    SA
 	 */
-	memcpy(dst, ieee80211_get_DA(hdr), ETH_ALEN);
-	memcpy(src, ieee80211_get_SA(hdr), ETH_ALEN);
+	memcpy(tmp.h_dest, ieee80211_get_DA(hdr), ETH_ALEN);
+	memcpy(tmp.h_source, ieee80211_get_SA(hdr), ETH_ALEN);
+
+	if (iftype == NL80211_IFTYPE_MESH_POINT)
+		skb_copy_bits(skb, hdrlen, &mesh_flags, 1);
 
 	switch (hdr->frame_control &
 		cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS)) {
@@ -450,44 +463,31 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
 			     iftype != NL80211_IFTYPE_STATION))
 			return -1;
 		if (iftype == NL80211_IFTYPE_MESH_POINT) {
-			struct ieee80211s_hdr *meshdr =
-				(struct ieee80211s_hdr *) (skb->data + hdrlen);
-			/* make sure meshdr->flags is on the linear part */
-			if (!pskb_may_pull(skb, hdrlen + 1))
-				return -1;
-			if (meshdr->flags & MESH_FLAGS_AE_A4)
+			if (mesh_flags & MESH_FLAGS_AE_A4)
 				return -1;
-			if (meshdr->flags & MESH_FLAGS_AE_A5_A6) {
+			if (mesh_flags & MESH_FLAGS_AE_A5_A6) {
 				skb_copy_bits(skb, hdrlen +
 					offsetof(struct ieee80211s_hdr, eaddr1),
-				       	dst, ETH_ALEN);
-				skb_copy_bits(skb, hdrlen +
-					offsetof(struct ieee80211s_hdr, eaddr2),
-				        src, ETH_ALEN);
+					tmp.h_dest, 2 * ETH_ALEN);
 			}
-			hdrlen += ieee80211_get_mesh_hdrlen(meshdr);
+			hdrlen += __ieee80211_get_mesh_hdrlen(mesh_flags);
 		}
 		break;
 	case cpu_to_le16(IEEE80211_FCTL_FROMDS):
 		if ((iftype != NL80211_IFTYPE_STATION &&
 		     iftype != NL80211_IFTYPE_P2P_CLIENT &&
 		     iftype != NL80211_IFTYPE_MESH_POINT) ||
-		    (is_multicast_ether_addr(dst) &&
-		     ether_addr_equal(src, addr)))
+		    (is_multicast_ether_addr(tmp.h_dest) &&
+		     ether_addr_equal(tmp.h_source, addr)))
 			return -1;
 		if (iftype == NL80211_IFTYPE_MESH_POINT) {
-			struct ieee80211s_hdr *meshdr =
-				(struct ieee80211s_hdr *) (skb->data + hdrlen);
-			/* make sure meshdr->flags is on the linear part */
-			if (!pskb_may_pull(skb, hdrlen + 1))
+			if (mesh_flags & MESH_FLAGS_AE_A5_A6)
 				return -1;
-			if (meshdr->flags & MESH_FLAGS_AE_A5_A6)
-				return -1;
-			if (meshdr->flags & MESH_FLAGS_AE_A4)
+			if (mesh_flags & MESH_FLAGS_AE_A4)
 				skb_copy_bits(skb, hdrlen +
 					offsetof(struct ieee80211s_hdr, eaddr1),
-					src, ETH_ALEN);
-			hdrlen += ieee80211_get_mesh_hdrlen(meshdr);
+					tmp.h_source, ETH_ALEN);
+			hdrlen += __ieee80211_get_mesh_hdrlen(mesh_flags);
 		}
 		break;
 	case cpu_to_le16(0):
@@ -498,33 +498,33 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
 		break;
 	}
 
-	if (!pskb_may_pull(skb, hdrlen + 8))
-		return -1;
-
-	payload = skb->data + hdrlen;
-	ethertype = (payload[6] << 8) | payload[7];
+	skb_copy_bits(skb, hdrlen, &payload, sizeof(payload));
+	tmp.h_proto = payload.proto;
 
-	if (likely((ether_addr_equal(payload, rfc1042_header) &&
-		    ethertype != ETH_P_AARP && ethertype != ETH_P_IPX) ||
-		   ether_addr_equal(payload, bridge_tunnel_header))) {
+	if (likely((ether_addr_equal(payload.hdr, rfc1042_header) &&
+		    tmp.h_proto != htons(ETH_P_AARP) &&
+		    tmp.h_proto != htons(ETH_P_IPX)) ||
+		   ether_addr_equal(payload.hdr, bridge_tunnel_header)))
 		/* remove RFC1042 or Bridge-Tunnel encapsulation and
 		 * replace EtherType */
-		skb_pull(skb, hdrlen + 6);
-		memcpy(skb_push(skb, ETH_ALEN), src, ETH_ALEN);
-		memcpy(skb_push(skb, ETH_ALEN), dst, ETH_ALEN);
-	} else {
-		struct ethhdr *ehdr;
-		__be16 len;
+		hdrlen += ETH_ALEN + 2;
+	else
+		tmp.h_proto = htons(skb->len);
+
+	pskb_pull(skb, hdrlen);
 
-		skb_pull(skb, hdrlen);
-		len = htons(skb->len);
+	if (!ehdr)
 		ehdr = (struct ethhdr *) skb_push(skb, sizeof(struct ethhdr));
-		memcpy(ehdr->h_dest, dst, ETH_ALEN);
-		memcpy(ehdr->h_source, src, ETH_ALEN);
-		ehdr->h_proto = len;
-	}
+	memcpy(ehdr, &tmp, sizeof(tmp));
+
 	return 0;
 }
+
+int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
+			   enum nl80211_iftype iftype)
+{
+	return __ieee80211_data_to_8023(skb, NULL, addr, iftype);
+}
 EXPORT_SYMBOL(ieee80211_data_to_8023);
 
 int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr,
-- 
cgit v0.10.2


From 230fd28a91542f0abfe390bda74702a3504c3df6 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Tue, 2 Feb 2016 14:39:10 +0100
Subject: cfg80211: add support for non-linear skbs in ieee80211_amsdu_to_8023s

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/util.c b/net/wireless/util.c
index f75f601..6e4eb35 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -644,73 +644,75 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr,
 }
 EXPORT_SYMBOL(ieee80211_data_from_8023);
 
+static struct sk_buff *
+__ieee80211_amsdu_copy(struct sk_buff *skb, unsigned int hlen,
+		       int offset, int len)
+{
+	struct sk_buff *frame;
+
+	if (skb->len - offset < len)
+		return NULL;
+
+	/*
+	 * Allocate and reserve two bytes more for payload
+	 * alignment since sizeof(struct ethhdr) is 14.
+	 */
+	frame = dev_alloc_skb(hlen + sizeof(struct ethhdr) + 2 + len);
+
+	skb_reserve(frame, hlen + sizeof(struct ethhdr) + 2);
+	skb_copy_bits(skb, offset, skb_put(frame, len), len);
+
+	return frame;
+}
 
 void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 			      const u8 *addr, enum nl80211_iftype iftype,
 			      const unsigned int extra_headroom,
 			      bool has_80211_header)
 {
+	unsigned int hlen = ALIGN(extra_headroom, 4);
 	struct sk_buff *frame = NULL;
 	u16 ethertype;
 	u8 *payload;
-	const struct ethhdr *eth;
-	int remaining, err;
-	u8 dst[ETH_ALEN], src[ETH_ALEN];
-
-	if (skb_linearize(skb))
-		goto out;
+	int offset = 0, remaining, err;
+	struct ethhdr eth;
+	bool reuse_skb = true;
+	bool last = false;
 
 	if (has_80211_header) {
-		err = ieee80211_data_to_8023(skb, addr, iftype);
+		err = __ieee80211_data_to_8023(skb, &eth, addr, iftype);
 		if (err)
 			goto out;
-
-		/* skip the wrapping header */
-		eth = (struct ethhdr *) skb_pull(skb, sizeof(struct ethhdr));
-		if (!eth)
-			goto out;
-	} else {
-		eth = (struct ethhdr *) skb->data;
 	}
 
-	while (skb != frame) {
+	while (!last) {
+		unsigned int subframe_len;
+		int len;
 		u8 padding;
-		__be16 len = eth->h_proto;
-		unsigned int subframe_len = sizeof(struct ethhdr) + ntohs(len);
-
-		remaining = skb->len;
-		memcpy(dst, eth->h_dest, ETH_ALEN);
-		memcpy(src, eth->h_source, ETH_ALEN);
 
+		skb_copy_bits(skb, offset, &eth, sizeof(eth));
+		len = ntohs(eth.h_proto);
+		subframe_len = sizeof(struct ethhdr) + len;
 		padding = (4 - subframe_len) & 0x3;
+
 		/* the last MSDU has no padding */
+		remaining = skb->len - offset;
 		if (subframe_len > remaining)
 			goto purge;
 
-		skb_pull(skb, sizeof(struct ethhdr));
+		offset += sizeof(struct ethhdr);
 		/* reuse skb for the last subframe */
-		if (remaining <= subframe_len + padding)
+		last = remaining <= subframe_len + padding;
+		if (!skb_is_nonlinear(skb) && last) {
+			skb_pull(skb, offset);
 			frame = skb;
-		else {
-			unsigned int hlen = ALIGN(extra_headroom, 4);
-			/*
-			 * Allocate and reserve two bytes more for payload
-			 * alignment since sizeof(struct ethhdr) is 14.
-			 */
-			frame = dev_alloc_skb(hlen + subframe_len + 2);
+			reuse_skb = true;
+		} else {
+			frame = __ieee80211_amsdu_copy(skb, hlen, offset, len);
 			if (!frame)
 				goto purge;
 
-			skb_reserve(frame, hlen + sizeof(struct ethhdr) + 2);
-			memcpy(skb_put(frame, ntohs(len)), skb->data,
-				ntohs(len));
-
-			eth = (struct ethhdr *)skb_pull(skb, ntohs(len) +
-							padding);
-			if (!eth) {
-				dev_kfree_skb(frame);
-				goto purge;
-			}
+			offset += len + padding;
 		}
 
 		skb_reset_network_header(frame);
@@ -719,24 +721,20 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 
 		payload = frame->data;
 		ethertype = (payload[6] << 8) | payload[7];
-
 		if (likely((ether_addr_equal(payload, rfc1042_header) &&
 			    ethertype != ETH_P_AARP && ethertype != ETH_P_IPX) ||
 			   ether_addr_equal(payload, bridge_tunnel_header))) {
-			/* remove RFC1042 or Bridge-Tunnel
-			 * encapsulation and replace EtherType */
-			skb_pull(frame, 6);
-			memcpy(skb_push(frame, ETH_ALEN), src, ETH_ALEN);
-			memcpy(skb_push(frame, ETH_ALEN), dst, ETH_ALEN);
-		} else {
-			memcpy(skb_push(frame, sizeof(__be16)), &len,
-				sizeof(__be16));
-			memcpy(skb_push(frame, ETH_ALEN), src, ETH_ALEN);
-			memcpy(skb_push(frame, ETH_ALEN), dst, ETH_ALEN);
+			eth.h_proto = htons(ethertype);
+			skb_pull(frame, ETH_ALEN + 2);
 		}
+
+		memcpy(skb_push(frame, sizeof(eth)), &eth, sizeof(eth));
 		__skb_queue_tail(list, frame);
 	}
 
+	if (!reuse_skb)
+		dev_kfree_skb(skb);
+
 	return;
 
  purge:
-- 
cgit v0.10.2


From 34d505193bd10668acf1caba02d2f66bddc23fea Mon Sep 17 00:00:00 2001
From: Lior David <liord@codeaurora.org>
Date: Thu, 28 Jan 2016 10:58:25 +0200
Subject: cfg80211: basic support for PBSS network type

PBSS (Personal Basic Service Set) is a new BSS type for DMG
networks. It is similar to infrastructure BSS, having an AP-like
entity called PCP (PBSS Control Point), but it has few differences.
PBSS support is mandatory for 11ad devices.

Add support for PBSS by introducing a new PBSS flag attribute.
The PBSS flag is used in the START_AP command to request starting
a PCP instead of an AP, and in the CONNECT command to request
connecting to a PCP instead of an AP.

Signed-off-by: Lior David <liord@codeaurora.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c
index 20d07ef..1f231cd 100644
--- a/drivers/net/wireless/ath/wil6210/cfg80211.c
+++ b/drivers/net/wireless/ath/wil6210/cfg80211.c
@@ -422,6 +422,11 @@ static int wil_cfg80211_connect(struct wiphy *wiphy,
 	if (sme->privacy && !rsn_eid)
 		wil_info(wil, "WSC connection\n");
 
+	if (sme->pbss) {
+		wil_err(wil, "connect - PBSS not yet supported\n");
+		return -EOPNOTSUPP;
+	}
+
 	bss = cfg80211_get_bss(wiphy, sme->channel, sme->bssid,
 			       sme->ssid, sme->ssid_len,
 			       IEEE80211_BSS_TYPE_ESS, IEEE80211_PRIVACY_ANY);
@@ -870,6 +875,11 @@ static int wil_cfg80211_start_ap(struct wiphy *wiphy,
 		return -EINVAL;
 	}
 
+	if (info->pbss) {
+		wil_err(wil, "AP: PBSS not yet supported\n");
+		return -EOPNOTSUPP;
+	}
+
 	switch (info->hidden_ssid) {
 	case NL80211_HIDDEN_SSID_NOT_IN_USE:
 		hidden_ssid = WMI_HIDDEN_SSID_DISABLED;
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 9bcaaf7..9e1b24c 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -712,6 +712,8 @@ struct cfg80211_acl_data {
  * @p2p_opp_ps: P2P opportunistic PS
  * @acl: ACL configuration used by the drivers which has support for
  *	MAC address based access control
+ * @pbss: If set, start as a PCP instead of AP. Relevant for DMG
+ *	networks.
  */
 struct cfg80211_ap_settings {
 	struct cfg80211_chan_def chandef;
@@ -730,6 +732,7 @@ struct cfg80211_ap_settings {
 	u8 p2p_ctwindow;
 	bool p2p_opp_ps;
 	const struct cfg80211_acl_data *acl;
+	bool pbss;
 };
 
 /**
@@ -1888,6 +1891,8 @@ struct cfg80211_ibss_params {
  * @ht_capa_mask:  The bits of ht_capa which are to be used.
  * @vht_capa:  VHT Capability overrides
  * @vht_capa_mask: The bits of vht_capa which are to be used.
+ * @pbss: if set, connect to a PCP instead of AP. Valid for DMG
+ *	networks.
  */
 struct cfg80211_connect_params {
 	struct ieee80211_channel *channel;
@@ -1910,6 +1915,7 @@ struct cfg80211_connect_params {
 	struct ieee80211_ht_cap ht_capa_mask;
 	struct ieee80211_vht_cap vht_capa;
 	struct ieee80211_vht_cap vht_capa_mask;
+	bool pbss;
 };
 
 /**
@@ -3489,6 +3495,7 @@ struct cfg80211_cached_keys;
  *	registered for unexpected class 3 frames (AP mode)
  * @conn: (private) cfg80211 software SME connection state machine data
  * @connect_keys: (private) keys to set after connection is established
+ * @conn_bss_type: connecting/connected BSS type
  * @ibss_fixed: (private) IBSS is using fixed BSSID
  * @ibss_dfs_possible: (private) IBSS may change to a DFS channel
  * @event_list: (private) list for internal event processing
@@ -3519,6 +3526,7 @@ struct wireless_dev {
 	u8 ssid_len, mesh_id_len, mesh_id_up_len;
 	struct cfg80211_conn *conn;
 	struct cfg80211_cached_keys *connect_keys;
+	enum ieee80211_bss_type conn_bss_type;
 
 	struct list_head event_list;
 	spinlock_t event_lock;
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index 5b7b5eb..7758969 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -1789,6 +1789,10 @@ enum nl80211_commands {
  *	thus it must not specify the number of iterations, only the interval
  *	between scans. The scan plans are executed sequentially.
  *	Each scan plan is a nested attribute of &enum nl80211_sched_scan_plan.
+ * @NL80211_ATTR_PBSS: flag attribute. If set it means operate
+ *	in a PBSS. Specified in %NL80211_CMD_CONNECT to request
+ *	connecting to a PCP, and in %NL80211_CMD_START_AP to start
+ *	a PCP instead of AP. Relevant for DMG networks only.
  *
  * @NUM_NL80211_ATTR: total number of nl80211_attrs available
  * @NL80211_ATTR_MAX: highest attribute number currently defined
@@ -2164,6 +2168,8 @@ enum nl80211_attrs {
 	NL80211_ATTR_MAX_SCAN_PLAN_ITERATIONS,
 	NL80211_ATTR_SCHED_SCAN_PLANS,
 
+	NL80211_ATTR_PBSS,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index d4786f2..268cb49 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -401,6 +401,7 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = {
 	[NL80211_ATTR_NETNS_FD] = { .type = NLA_U32 },
 	[NL80211_ATTR_SCHED_SCAN_DELAY] = { .type = NLA_U32 },
 	[NL80211_ATTR_REG_INDOOR] = { .type = NLA_FLAG },
+	[NL80211_ATTR_PBSS] = { .type = NLA_FLAG },
 };
 
 /* policy for the key attributes */
@@ -3461,6 +3462,10 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
 			return PTR_ERR(params.acl);
 	}
 
+	params.pbss = nla_get_flag(info->attrs[NL80211_ATTR_PBSS]);
+	if (params.pbss && !rdev->wiphy.bands[IEEE80211_BAND_60GHZ])
+		return -EOPNOTSUPP;
+
 	wdev_lock(wdev);
 	err = rdev_start_ap(rdev, dev, &params);
 	if (!err) {
@@ -7980,6 +7985,12 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info)
 		connect.flags |= ASSOC_REQ_USE_RRM;
 	}
 
+	connect.pbss = nla_get_flag(info->attrs[NL80211_ATTR_PBSS]);
+	if (connect.pbss && !rdev->wiphy.bands[IEEE80211_BAND_60GHZ]) {
+		kzfree(connkeys);
+		return -EOPNOTSUPP;
+	}
+
 	wdev_lock(dev->ieee80211_ptr);
 	err = cfg80211_connect(rdev, dev, &connect, connkeys, NULL);
 	wdev_unlock(dev->ieee80211_ptr);
diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index 8020b5b..79bd3a1 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -264,7 +264,7 @@ static struct cfg80211_bss *cfg80211_get_conn_bss(struct wireless_dev *wdev)
 			       wdev->conn->params.bssid,
 			       wdev->conn->params.ssid,
 			       wdev->conn->params.ssid_len,
-			       IEEE80211_BSS_TYPE_ESS,
+			       wdev->conn_bss_type,
 			       IEEE80211_PRIVACY(wdev->conn->params.privacy));
 	if (!bss)
 		return NULL;
@@ -687,7 +687,7 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid,
 		WARN_ON_ONCE(!wiphy_to_rdev(wdev->wiphy)->ops->connect);
 		bss = cfg80211_get_bss(wdev->wiphy, NULL, bssid,
 				       wdev->ssid, wdev->ssid_len,
-				       IEEE80211_BSS_TYPE_ESS,
+				       wdev->conn_bss_type,
 				       IEEE80211_PRIVACY_ANY);
 		if (bss)
 			cfg80211_hold_bss(bss_from_pub(bss));
@@ -846,7 +846,7 @@ void cfg80211_roamed(struct net_device *dev,
 
 	bss = cfg80211_get_bss(wdev->wiphy, channel, bssid, wdev->ssid,
 			       wdev->ssid_len,
-			       IEEE80211_BSS_TYPE_ESS, IEEE80211_PRIVACY_ANY);
+			       wdev->conn_bss_type, IEEE80211_PRIVACY_ANY);
 	if (WARN_ON(!bss))
 		return;
 
@@ -1017,6 +1017,9 @@ int cfg80211_connect(struct cfg80211_registered_device *rdev,
 	memcpy(wdev->ssid, connect->ssid, connect->ssid_len);
 	wdev->ssid_len = connect->ssid_len;
 
+	wdev->conn_bss_type = connect->pbss ? IEEE80211_BSS_TYPE_PBSS :
+					      IEEE80211_BSS_TYPE_ESS;
+
 	if (!rdev->ops->connect)
 		err = cfg80211_sme_connect(wdev, connect, prev_bssid);
 	else
-- 
cgit v0.10.2


From bf5a70e105820eb214d3f33376cf6a3011d2e20c Mon Sep 17 00:00:00 2001
From: Henning Rogge <hrogge@gmail.com>
Date: Wed, 3 Feb 2016 13:58:36 +0100
Subject: mac80211: Remove MPP table entries with MPath

Make the mesh_path_del() function remove all mpp table entries
that are proxied by the removed mesh path.

Acked-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c
index c95bdb8..40ed206 100644
--- a/net/mac80211/mesh_pathtbl.c
+++ b/net/mac80211/mesh_pathtbl.c
@@ -834,6 +834,29 @@ void mesh_path_flush_by_nexthop(struct sta_info *sta)
 	rcu_read_unlock();
 }
 
+static void mpp_flush_by_proxy(struct ieee80211_sub_if_data *sdata,
+			       const u8 *proxy)
+{
+	struct mesh_table *tbl;
+	struct mesh_path *mpp;
+	struct mpath_node *node;
+	int i;
+
+	rcu_read_lock();
+	read_lock_bh(&pathtbl_resize_lock);
+	tbl = resize_dereference_mpp_paths();
+	for_each_mesh_entry(tbl, node, i) {
+		mpp = node->mpath;
+		if (ether_addr_equal(mpp->mpp, proxy)) {
+			spin_lock(&tbl->hashwlock[i]);
+			__mesh_path_del(tbl, node);
+			spin_unlock(&tbl->hashwlock[i]);
+		}
+	}
+	read_unlock_bh(&pathtbl_resize_lock);
+	rcu_read_unlock();
+}
+
 static void table_flush_by_iface(struct mesh_table *tbl,
 				 struct ieee80211_sub_if_data *sdata)
 {
@@ -891,6 +914,9 @@ int mesh_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
 	int hash_idx;
 	int err = 0;
 
+	/* flush relevant mpp entries first */
+	mpp_flush_by_proxy(sdata, addr);
+
 	read_lock_bh(&pathtbl_resize_lock);
 	tbl = resize_dereference_mesh_paths();
 	hash_idx = mesh_table_hash(addr, sdata, tbl);
-- 
cgit v0.10.2


From ab1c7906b0bead549065a3a2def5aeda21818c99 Mon Sep 17 00:00:00 2001
From: Henning Rogge <hrogge@gmail.com>
Date: Wed, 3 Feb 2016 13:58:37 +0100
Subject: mac80211: let unused MPP table entries timeout

Remember the last time when a mpp table entry is used for
rx or tx and remove them after MESH_PATH_EXPIRE time.

Acked-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c
index 40ed206..29e5040 100644
--- a/net/mac80211/mesh_pathtbl.c
+++ b/net/mac80211/mesh_pathtbl.c
@@ -941,6 +941,46 @@ enddel:
 }
 
 /**
+ * mpp_path_del - delete a mesh proxy path from the table
+ *
+ * @addr: addr address (ETH_ALEN length)
+ * @sdata: local subif
+ *
+ * Returns: 0 if successful
+ */
+static int mpp_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
+{
+	struct mesh_table *tbl;
+	struct mesh_path *mpath;
+	struct mpath_node *node;
+	struct hlist_head *bucket;
+	int hash_idx;
+	int err = 0;
+
+	read_lock_bh(&pathtbl_resize_lock);
+	tbl = resize_dereference_mpp_paths();
+	hash_idx = mesh_table_hash(addr, sdata, tbl);
+	bucket = &tbl->hash_buckets[hash_idx];
+
+	spin_lock(&tbl->hashwlock[hash_idx]);
+	hlist_for_each_entry(node, bucket, list) {
+		mpath = node->mpath;
+		if (mpath->sdata == sdata &&
+		    ether_addr_equal(addr, mpath->dst)) {
+			__mesh_path_del(tbl, node);
+			goto enddel;
+		}
+	}
+
+	err = -ENXIO;
+enddel:
+	mesh_paths_generation++;
+	spin_unlock(&tbl->hashwlock[hash_idx]);
+	read_unlock_bh(&pathtbl_resize_lock);
+	return err;
+}
+
+/**
  * mesh_path_tx_pending - sends pending frames in a mesh path queue
  *
  * @mpath: mesh path to activate
@@ -1154,6 +1194,17 @@ void mesh_path_expire(struct ieee80211_sub_if_data *sdata)
 		     time_after(jiffies, mpath->exp_time + MESH_PATH_EXPIRE))
 			mesh_path_del(mpath->sdata, mpath->dst);
 	}
+
+	tbl = rcu_dereference(mpp_paths);
+	for_each_mesh_entry(tbl, node, i) {
+		if (node->mpath->sdata != sdata)
+			continue;
+		mpath = node->mpath;
+		if ((!(mpath->flags & MESH_PATH_FIXED)) &&
+		    time_after(jiffies, mpath->exp_time + MESH_PATH_EXPIRE))
+			mpp_path_del(mpath->sdata, mpath->dst);
+	}
+
 	rcu_read_unlock();
 }
 
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 4419214..9127957 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -2311,6 +2311,7 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx)
 			spin_lock_bh(&mppath->state_lock);
 			if (!ether_addr_equal(mppath->mpp, mpp_addr))
 				memcpy(mppath->mpp, mpp_addr, ETH_ALEN);
+			mppath->exp_time = jiffies;
 			spin_unlock_bh(&mppath->state_lock);
 		}
 		rcu_read_unlock();
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index a5aa275..7bb67fa 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -2173,8 +2173,11 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata,
 					mpp_lookup = true;
 			}
 
-			if (mpp_lookup)
+			if (mpp_lookup) {
 				mppath = mpp_path_lookup(sdata, skb->data);
+				if (mppath)
+					mppath->exp_time = jiffies;
+			}
 
 			if (mppath && mpath)
 				mesh_path_del(mpath->sdata, mpath->dst);
-- 
cgit v0.10.2


From 4cc955de7c99fe60b09688d7c2564f8fc7d00ca2 Mon Sep 17 00:00:00 2001
From: Henning Rogge <hrogge@gmail.com>
Date: Wed, 3 Feb 2016 13:58:38 +0100
Subject: mac80211: Unify mesh and mpp path removal function

mpp_path_del() and mesh_path_del() are mostly the same function.
Move common code into a new static function.

Acked-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c
index 29e5040..2ba7aa5 100644
--- a/net/mac80211/mesh_pathtbl.c
+++ b/net/mac80211/mesh_pathtbl.c
@@ -55,16 +55,21 @@ int mpp_paths_generation;
 static DEFINE_RWLOCK(pathtbl_resize_lock);
 
 
+static inline struct mesh_table *resize_dereference_paths(
+	struct mesh_table __rcu *table)
+{
+	return rcu_dereference_protected(table,
+					lockdep_is_held(&pathtbl_resize_lock));
+}
+
 static inline struct mesh_table *resize_dereference_mesh_paths(void)
 {
-	return rcu_dereference_protected(mesh_paths,
-		lockdep_is_held(&pathtbl_resize_lock));
+	return resize_dereference_paths(mesh_paths);
 }
 
 static inline struct mesh_table *resize_dereference_mpp_paths(void)
 {
-	return rcu_dereference_protected(mpp_paths,
-		lockdep_is_held(&pathtbl_resize_lock));
+	return resize_dereference_paths(mpp_paths);
 }
 
 /*
@@ -898,14 +903,17 @@ void mesh_path_flush_by_iface(struct ieee80211_sub_if_data *sdata)
 }
 
 /**
- * mesh_path_del - delete a mesh path from the table
+ * table_path_del - delete a path from the mesh or mpp table
  *
- * @addr: dst address (ETH_ALEN length)
+ * @tbl: mesh or mpp path table
  * @sdata: local subif
+ * @addr: dst address (ETH_ALEN length)
  *
  * Returns: 0 if successful
  */
-int mesh_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
+static int table_path_del(struct mesh_table __rcu *rcu_tbl,
+			  struct ieee80211_sub_if_data *sdata,
+			  const u8 *addr)
 {
 	struct mesh_table *tbl;
 	struct mesh_path *mpath;
@@ -914,11 +922,7 @@ int mesh_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
 	int hash_idx;
 	int err = 0;
 
-	/* flush relevant mpp entries first */
-	mpp_flush_by_proxy(sdata, addr);
-
-	read_lock_bh(&pathtbl_resize_lock);
-	tbl = resize_dereference_mesh_paths();
+	tbl = resize_dereference_paths(rcu_tbl);
 	hash_idx = mesh_table_hash(addr, sdata, tbl);
 	bucket = &tbl->hash_buckets[hash_idx];
 
@@ -934,9 +938,30 @@ int mesh_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
 
 	err = -ENXIO;
 enddel:
-	mesh_paths_generation++;
 	spin_unlock(&tbl->hashwlock[hash_idx]);
+	return err;
+}
+
+/**
+ * mesh_path_del - delete a mesh path from the table
+ *
+ * @addr: dst address (ETH_ALEN length)
+ * @sdata: local subif
+ *
+ * Returns: 0 if successful
+ */
+int mesh_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
+{
+	int err = 0;
+
+	/* flush relevant mpp entries first */
+	mpp_flush_by_proxy(sdata, addr);
+
+	read_lock_bh(&pathtbl_resize_lock);
+	err = table_path_del(mesh_paths, sdata, addr);
+	mesh_paths_generation++;
 	read_unlock_bh(&pathtbl_resize_lock);
+
 	return err;
 }
 
@@ -950,33 +975,13 @@ enddel:
  */
 static int mpp_path_del(struct ieee80211_sub_if_data *sdata, const u8 *addr)
 {
-	struct mesh_table *tbl;
-	struct mesh_path *mpath;
-	struct mpath_node *node;
-	struct hlist_head *bucket;
-	int hash_idx;
 	int err = 0;
 
 	read_lock_bh(&pathtbl_resize_lock);
-	tbl = resize_dereference_mpp_paths();
-	hash_idx = mesh_table_hash(addr, sdata, tbl);
-	bucket = &tbl->hash_buckets[hash_idx];
-
-	spin_lock(&tbl->hashwlock[hash_idx]);
-	hlist_for_each_entry(node, bucket, list) {
-		mpath = node->mpath;
-		if (mpath->sdata == sdata &&
-		    ether_addr_equal(addr, mpath->dst)) {
-			__mesh_path_del(tbl, node);
-			goto enddel;
-		}
-	}
-
-	err = -ENXIO;
-enddel:
-	mesh_paths_generation++;
-	spin_unlock(&tbl->hashwlock[hash_idx]);
+	err = table_path_del(mpp_paths, sdata, addr);
+	mpp_paths_generation++;
 	read_unlock_bh(&pathtbl_resize_lock);
+
 	return err;
 }
 
-- 
cgit v0.10.2


From 0241fa19ef4021c3db6f70b8f08d69f9e21cc043 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 3 Feb 2016 21:52:23 +0200
Subject: mac80211: fix VHT MU-MIMO frame processing

The source and destintation addresses in the memcpy arguments
are flipped. Fix that.

Fixes: 23a1f8d44c0b("mac80211: process and save VHT MU-MIMO group frame")
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index 204cf9a..341d192 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -459,10 +459,12 @@ void ieee80211_process_mu_groups(struct ieee80211_sub_if_data *sdata,
 		    bss_conf->mu_group.membership, WLAN_MEMBERSHIP_LEN))
 		return;
 
-	memcpy(mgmt->u.action.u.vht_group_notif.membership,
-	       bss_conf->mu_group.membership, WLAN_MEMBERSHIP_LEN);
-	memcpy(mgmt->u.action.u.vht_group_notif.position,
-	       bss_conf->mu_group.position, WLAN_USER_POSITION_LEN);
+	memcpy(bss_conf->mu_group.membership,
+	       mgmt->u.action.u.vht_group_notif.membership,
+	       WLAN_MEMBERSHIP_LEN);
+	memcpy(bss_conf->mu_group.position,
+	       mgmt->u.action.u.vht_group_notif.position,
+	       WLAN_USER_POSITION_LEN);
 
 	ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_MU_GROUPS);
 }
-- 
cgit v0.10.2


From 2bf0ccc7095e6cf665bbdb7c32c352d24f1ae033 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Mon, 8 Feb 2016 14:25:26 +0100
Subject: cfg80211: fix faulty variable initialization in
 ieee80211_amsdu_to_8023s

reuse_skb is set to true if the code decides to use the last segment.
Fixes a memory leak

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/util.c b/net/wireless/util.c
index 6e4eb35..9880c89 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -676,7 +676,7 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 	u8 *payload;
 	int offset = 0, remaining, err;
 	struct ethhdr eth;
-	bool reuse_skb = true;
+	bool reuse_skb = false;
 	bool last = false;
 
 	if (has_80211_header) {
-- 
cgit v0.10.2


From 2b67f944f88c29e71efdcfc4d0c21ab28fbf23ae Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Mon, 8 Feb 2016 14:34:42 +0100
Subject: cfg80211: reuse existing page fragments in A-MSDU rx

This massively reduces data copying and thus improves rx performance

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/util.c b/net/wireless/util.c
index 9880c89..c7f6820 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -644,23 +644,93 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr,
 }
 EXPORT_SYMBOL(ieee80211_data_from_8023);
 
+static void
+__frame_add_frag(struct sk_buff *skb, struct page *page,
+		 void *ptr, int len, int size)
+{
+	struct skb_shared_info *sh = skb_shinfo(skb);
+	int page_offset;
+
+	atomic_inc(&page->_count);
+	page_offset = ptr - page_address(page);
+	skb_add_rx_frag(skb, sh->nr_frags, page, page_offset, len, size);
+}
+
+static void
+__ieee80211_amsdu_copy_frag(struct sk_buff *skb, struct sk_buff *frame,
+			    int offset, int len)
+{
+	struct skb_shared_info *sh = skb_shinfo(skb);
+	const skb_frag_t *frag = &sh->frags[-1];
+	struct page *frag_page;
+	void *frag_ptr;
+	int frag_len, frag_size;
+	int head_size = skb->len - skb->data_len;
+	int cur_len;
+
+	frag_page = virt_to_head_page(skb->head);
+	frag_ptr = skb->data;
+	frag_size = head_size;
+
+	while (offset >= frag_size) {
+		offset -= frag_size;
+		frag++;
+		frag_page = skb_frag_page(frag);
+		frag_ptr = skb_frag_address(frag);
+		frag_size = skb_frag_size(frag);
+	}
+
+	frag_ptr += offset;
+	frag_len = frag_size - offset;
+
+	cur_len = min(len, frag_len);
+
+	__frame_add_frag(frame, frag_page, frag_ptr, cur_len, frag_size);
+	len -= cur_len;
+
+	while (len > 0) {
+		frag++;
+		frag_len = skb_frag_size(frag);
+		cur_len = min(len, frag_len);
+		__frame_add_frag(frame, skb_frag_page(frag),
+				 skb_frag_address(frag), cur_len, frag_len);
+		len -= cur_len;
+	}
+}
+
 static struct sk_buff *
 __ieee80211_amsdu_copy(struct sk_buff *skb, unsigned int hlen,
-		       int offset, int len)
+		       int offset, int len, bool reuse_frag)
 {
 	struct sk_buff *frame;
+	int cur_len = len;
 
 	if (skb->len - offset < len)
 		return NULL;
 
 	/*
+	 * When reusing framents, copy some data to the head to simplify
+	 * ethernet header handling and speed up protocol header processing
+	 * in the stack later.
+	 */
+	if (reuse_frag)
+		cur_len = min_t(int, len, 32);
+
+	/*
 	 * Allocate and reserve two bytes more for payload
 	 * alignment since sizeof(struct ethhdr) is 14.
 	 */
-	frame = dev_alloc_skb(hlen + sizeof(struct ethhdr) + 2 + len);
+	frame = dev_alloc_skb(hlen + sizeof(struct ethhdr) + 2 + cur_len);
 
 	skb_reserve(frame, hlen + sizeof(struct ethhdr) + 2);
-	skb_copy_bits(skb, offset, skb_put(frame, len), len);
+	skb_copy_bits(skb, offset, skb_put(frame, cur_len), cur_len);
+
+	len -= cur_len;
+	if (!len)
+		return frame;
+
+	offset += cur_len;
+	__ieee80211_amsdu_copy_frag(skb, frame, offset, len);
 
 	return frame;
 }
@@ -676,6 +746,7 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 	u8 *payload;
 	int offset = 0, remaining, err;
 	struct ethhdr eth;
+	bool reuse_frag = skb->head_frag && !skb_has_frag_list(skb);
 	bool reuse_skb = false;
 	bool last = false;
 
@@ -703,12 +774,13 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list,
 		offset += sizeof(struct ethhdr);
 		/* reuse skb for the last subframe */
 		last = remaining <= subframe_len + padding;
-		if (!skb_is_nonlinear(skb) && last) {
+		if (!skb_is_nonlinear(skb) && !reuse_frag && last) {
 			skb_pull(skb, offset);
 			frame = skb;
 			reuse_skb = true;
 		} else {
-			frame = __ieee80211_amsdu_copy(skb, hlen, offset, len);
+			frame = __ieee80211_amsdu_copy(skb, hlen, offset, len,
+						       reuse_frag);
 			if (!frame)
 				goto purge;
 
-- 
cgit v0.10.2


From 109843b07aed8663a3c1fe9ef02c540b6ff203b4 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com>
Date: Fri, 19 Feb 2016 12:18:01 +0100
Subject: mac80211: fix wiphy supported_band access

Fix wiphy supported_band access in tx radiotap parsing introduced
in commit 5ec3aed9ba4c ("mac80211: Parse legacy and HT rate in
injected frames"). In particular, info->band is always set to 0
(IEEE80211_BAND_2GHZ) since it has not assigned yet.
This cause a kernel crash on 5GHz only devices.
Move ieee80211_parse_tx_radiotap() after info->band assignment

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 7bb67fa..3a7475f 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -1892,10 +1892,6 @@ netdev_tx_t ieee80211_monitor_start_xmit(struct sk_buff *skb,
 	info->flags = IEEE80211_TX_CTL_REQ_TX_STATUS |
 		      IEEE80211_TX_CTL_INJECTED;
 
-	/* process and remove the injection radiotap header */
-	if (!ieee80211_parse_tx_radiotap(local, skb))
-		goto fail;
-
 	rcu_read_lock();
 
 	/*
@@ -1957,6 +1953,11 @@ netdev_tx_t ieee80211_monitor_start_xmit(struct sk_buff *skb,
 		goto fail_rcu;
 
 	info->band = chandef->chan->band;
+
+	/* process and remove the injection radiotap header */
+	if (!ieee80211_parse_tx_radiotap(local, skb))
+		goto fail_rcu;
+
 	ieee80211_xmit(sdata, NULL, skb);
 	rcu_read_unlock();
 
-- 
cgit v0.10.2


From f8079d43cf0f1f0171606e75fcef6fe17bb183f2 Mon Sep 17 00:00:00 2001
From: Eliad Peller <eliad@wizery.com>
Date: Sun, 14 Feb 2016 13:56:35 +0200
Subject: mac80211: move TKIP TX IVs to public part of key struct

Some drivers/devices might want to set the IVs by
themselves (and still let mac80211 generate MMIC).

Specifically, this is needed when the device does
offloading at certain times, and the driver has
to make sure that the IVs of new tx frames (from
the host) are synchronized with IVs that were
potentially used during the offloading.

Similarly to CCMP, move the TX IVs of TKIP keys to the
public part of the key struct, and export a function
to add the IV right into the crypto header.

The public tx_pn field is defined as atomic64, so define
TKIP_PN_TO_IV16/32 helper macros to convert it to iv16/32
when needed.

Since the iv32 used for the p1k cache is taken
directly from the frame, we can safely remove
iv16/32 from being protected by tkip.txlock.

Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5714774..15879b4 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1521,9 +1521,8 @@ enum ieee80211_key_flags {
  *	wants to be given when a frame is transmitted and needs to be
  *	encrypted in hardware.
  * @cipher: The key's cipher suite selector.
- * @tx_pn: PN used for TX on non-TKIP keys, may be used by the driver
- *	as well if it needs to do software PN assignment by itself
- *	(e.g. due to TSO)
+ * @tx_pn: PN used for TX keys, may be used by the driver as well if it
+ *	needs to do software PN assignment by itself (e.g. due to TSO)
  * @flags: key flags, see &enum ieee80211_key_flags.
  * @keyidx: the key index (0-3)
  * @keylen: key material length
@@ -1549,6 +1548,9 @@ struct ieee80211_key_conf {
 
 #define IEEE80211_MAX_PN_LEN	16
 
+#define TKIP_PN_TO_IV16(pn) ((u16)(pn & 0xffff))
+#define TKIP_PN_TO_IV32(pn) ((u32)((pn >> 16) & 0xffffffff))
+
 /**
  * struct ieee80211_key_seq - key sequence counter
  *
@@ -4447,6 +4449,21 @@ void ieee80211_get_tkip_p2k(struct ieee80211_key_conf *keyconf,
 			    struct sk_buff *skb, u8 *p2k);
 
 /**
+ * ieee80211_tkip_add_iv - write TKIP IV and Ext. IV to pos
+ *
+ * @pos: start of crypto header
+ * @keyconf: the parameter passed with the set key
+ * @pn: PN to add
+ *
+ * Returns: pointer to the octet following IVs (i.e. beginning of
+ * the packet payload)
+ *
+ * This function writes the tkip IV value to pos (which should
+ * point to the crypto header)
+ */
+u8 *ieee80211_tkip_add_iv(u8 *pos, struct ieee80211_key_conf *keyconf, u64 pn);
+
+/**
  * ieee80211_get_key_tx_seq - get key TX sequence counter
  *
  * @keyconf: the parameter passed with the set key
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 66d22de..fe1704c 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -339,8 +339,9 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev,
 
 	switch (key->conf.cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
-		iv32 = key->u.tkip.tx.iv32;
-		iv16 = key->u.tkip.tx.iv16;
+		pn64 = atomic64_read(&key->conf.tx_pn);
+		iv32 = TKIP_PN_TO_IV32(pn64);
+		iv16 = TKIP_PN_TO_IV16(pn64);
 
 		if (key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE &&
 		    !(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV)) {
diff --git a/net/mac80211/debugfs_key.c b/net/mac80211/debugfs_key.c
index 7961e7d..a2ef95f 100644
--- a/net/mac80211/debugfs_key.c
+++ b/net/mac80211/debugfs_key.c
@@ -132,9 +132,10 @@ static ssize_t key_tx_spec_read(struct file *file, char __user *userbuf,
 		len = scnprintf(buf, sizeof(buf), "\n");
 		break;
 	case WLAN_CIPHER_SUITE_TKIP:
+		pn = atomic64_read(&key->conf.tx_pn);
 		len = scnprintf(buf, sizeof(buf), "%08x %04x\n",
-				key->u.tkip.tx.iv32,
-				key->u.tkip.tx.iv16);
+				TKIP_PN_TO_IV32(pn),
+				TKIP_PN_TO_IV16(pn));
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 	case WLAN_CIPHER_SUITE_CCMP_256:
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index 5e5bc59..f9c4cb9 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -945,8 +945,9 @@ void ieee80211_get_key_tx_seq(struct ieee80211_key_conf *keyconf,
 
 	switch (key->conf.cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
-		seq->tkip.iv32 = key->u.tkip.tx.iv32;
-		seq->tkip.iv16 = key->u.tkip.tx.iv16;
+		pn64 = atomic64_read(&key->conf.tx_pn);
+		seq->tkip.iv32 = TKIP_PN_TO_IV32(pn64);
+		seq->tkip.iv16 = TKIP_PN_TO_IV16(pn64);
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 	case WLAN_CIPHER_SUITE_CCMP_256:
@@ -1039,8 +1040,8 @@ void ieee80211_set_key_tx_seq(struct ieee80211_key_conf *keyconf,
 
 	switch (key->conf.cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
-		key->u.tkip.tx.iv32 = seq->tkip.iv32;
-		key->u.tkip.tx.iv16 = seq->tkip.iv16;
+		pn64 = (u64)seq->tkip.iv16 | ((u64)seq->tkip.iv32 << 16);
+		atomic64_set(&key->conf.tx_pn, pn64);
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 	case WLAN_CIPHER_SUITE_CCMP_256:
diff --git a/net/mac80211/key.h b/net/mac80211/key.h
index 9951ef0..4aa20ce 100644
--- a/net/mac80211/key.h
+++ b/net/mac80211/key.h
@@ -44,13 +44,17 @@ enum ieee80211_internal_tkip_state {
 };
 
 struct tkip_ctx {
-	u32 iv32;	/* current iv32 */
-	u16 iv16;	/* current iv16 */
 	u16 p1k[5];	/* p1k cache */
 	u32 p1k_iv32;	/* iv32 for which p1k computed */
 	enum ieee80211_internal_tkip_state state;
 };
 
+struct tkip_ctx_rx {
+	struct tkip_ctx ctx;
+	u32 iv32;	/* current iv32 */
+	u16 iv16;	/* current iv16 */
+};
+
 struct ieee80211_key {
 	struct ieee80211_local *local;
 	struct ieee80211_sub_if_data *sdata;
@@ -71,7 +75,7 @@ struct ieee80211_key {
 			struct tkip_ctx tx;
 
 			/* last received RSC */
-			struct tkip_ctx rx[IEEE80211_NUM_TIDS];
+			struct tkip_ctx_rx rx[IEEE80211_NUM_TIDS];
 
 			/* number of mic failures */
 			u32 mic_failures;
diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c
index 0ae2077..b362282 100644
--- a/net/mac80211/tkip.c
+++ b/net/mac80211/tkip.c
@@ -1,6 +1,7 @@
 /*
  * Copyright 2002-2004, Instant802 Networks, Inc.
  * Copyright 2005, Devicescape Software, Inc.
+ * Copyright (C) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -142,15 +143,14 @@ static void tkip_mixing_phase2(const u8 *tk, struct tkip_ctx *ctx,
 /* Add TKIP IV and Ext. IV at @pos. @iv0, @iv1, and @iv2 are the first octets
  * of the IV. Returns pointer to the octet following IVs (i.e., beginning of
  * the packet payload). */
-u8 *ieee80211_tkip_add_iv(u8 *pos, struct ieee80211_key *key)
+u8 *ieee80211_tkip_add_iv(u8 *pos, struct ieee80211_key_conf *keyconf, u64 pn)
 {
-	lockdep_assert_held(&key->u.tkip.txlock);
-
-	pos = write_tkip_iv(pos, key->u.tkip.tx.iv16);
-	*pos++ = (key->conf.keyidx << 6) | (1 << 5) /* Ext IV */;
-	put_unaligned_le32(key->u.tkip.tx.iv32, pos);
+	pos = write_tkip_iv(pos, TKIP_PN_TO_IV16(pn));
+	*pos++ = (keyconf->keyidx << 6) | (1 << 5) /* Ext IV */;
+	put_unaligned_le32(TKIP_PN_TO_IV32(pn), pos);
 	return pos + 4;
 }
+EXPORT_SYMBOL_GPL(ieee80211_tkip_add_iv);
 
 static void ieee80211_compute_tkip_p1k(struct ieee80211_key *key, u32 iv32)
 {
@@ -250,6 +250,7 @@ int ieee80211_tkip_decrypt_data(struct crypto_cipher *tfm,
 	u8 rc4key[16], keyid, *pos = payload;
 	int res;
 	const u8 *tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
+	struct tkip_ctx_rx *rx_ctx = &key->u.tkip.rx[queue];
 
 	if (payload_len < 12)
 		return -1;
@@ -265,37 +266,36 @@ int ieee80211_tkip_decrypt_data(struct crypto_cipher *tfm,
 	if ((keyid >> 6) != key->conf.keyidx)
 		return TKIP_DECRYPT_INVALID_KEYIDX;
 
-	if (key->u.tkip.rx[queue].state != TKIP_STATE_NOT_INIT &&
-	    (iv32 < key->u.tkip.rx[queue].iv32 ||
-	     (iv32 == key->u.tkip.rx[queue].iv32 &&
-	      iv16 <= key->u.tkip.rx[queue].iv16)))
+	if (rx_ctx->ctx.state != TKIP_STATE_NOT_INIT &&
+	    (iv32 < rx_ctx->iv32 ||
+	     (iv32 == rx_ctx->iv32 && iv16 <= rx_ctx->iv16)))
 		return TKIP_DECRYPT_REPLAY;
 
 	if (only_iv) {
 		res = TKIP_DECRYPT_OK;
-		key->u.tkip.rx[queue].state = TKIP_STATE_PHASE1_HW_UPLOADED;
+		rx_ctx->ctx.state = TKIP_STATE_PHASE1_HW_UPLOADED;
 		goto done;
 	}
 
-	if (key->u.tkip.rx[queue].state == TKIP_STATE_NOT_INIT ||
-	    key->u.tkip.rx[queue].iv32 != iv32) {
+	if (rx_ctx->ctx.state == TKIP_STATE_NOT_INIT ||
+	    rx_ctx->iv32 != iv32) {
 		/* IV16 wrapped around - perform TKIP phase 1 */
-		tkip_mixing_phase1(tk, &key->u.tkip.rx[queue], ta, iv32);
+		tkip_mixing_phase1(tk, &rx_ctx->ctx, ta, iv32);
 	}
 	if (key->local->ops->update_tkip_key &&
 	    key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE &&
-	    key->u.tkip.rx[queue].state != TKIP_STATE_PHASE1_HW_UPLOADED) {
+	    rx_ctx->ctx.state != TKIP_STATE_PHASE1_HW_UPLOADED) {
 		struct ieee80211_sub_if_data *sdata = key->sdata;
 
 		if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
 			sdata = container_of(key->sdata->bss,
 					struct ieee80211_sub_if_data, u.ap);
 		drv_update_tkip_key(key->local, sdata, &key->conf, key->sta,
-				iv32, key->u.tkip.rx[queue].p1k);
-		key->u.tkip.rx[queue].state = TKIP_STATE_PHASE1_HW_UPLOADED;
+				iv32, rx_ctx->ctx.p1k);
+		rx_ctx->ctx.state = TKIP_STATE_PHASE1_HW_UPLOADED;
 	}
 
-	tkip_mixing_phase2(tk, &key->u.tkip.rx[queue], iv16, rc4key);
+	tkip_mixing_phase2(tk, &rx_ctx->ctx, iv16, rc4key);
 
 	res = ieee80211_wep_decrypt_data(tfm, rc4key, 16, pos, payload_len - 12);
  done:
diff --git a/net/mac80211/tkip.h b/net/mac80211/tkip.h
index e3ecb65..a1bcbfb 100644
--- a/net/mac80211/tkip.h
+++ b/net/mac80211/tkip.h
@@ -13,8 +13,6 @@
 #include <linux/crypto.h>
 #include "key.h"
 
-u8 *ieee80211_tkip_add_iv(u8 *pos, struct ieee80211_key *key);
-
 int ieee80211_tkip_encrypt_data(struct crypto_cipher *tfm,
 				struct ieee80211_key *key,
 				struct sk_buff *skb,
diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c
index d824c38..1884825 100644
--- a/net/mac80211/wpa.c
+++ b/net/mac80211/wpa.c
@@ -1,6 +1,7 @@
 /*
  * Copyright 2002-2004, Instant802 Networks, Inc.
  * Copyright 2008, Jouni Malinen <j@w1.fi>
+ * Copyright (C) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -183,7 +184,6 @@ mic_fail_no_key:
 	return RX_DROP_UNUSABLE;
 }
 
-
 static int tkip_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
@@ -191,6 +191,7 @@ static int tkip_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	unsigned int hdrlen;
 	int len, tail;
+	u64 pn;
 	u8 *pos;
 
 	if (info->control.hw_key &&
@@ -222,12 +223,8 @@ static int tkip_encrypt_skb(struct ieee80211_tx_data *tx, struct sk_buff *skb)
 		return 0;
 
 	/* Increase IV for the frame */
-	spin_lock(&key->u.tkip.txlock);
-	key->u.tkip.tx.iv16++;
-	if (key->u.tkip.tx.iv16 == 0)
-		key->u.tkip.tx.iv32++;
-	pos = ieee80211_tkip_add_iv(pos, key);
-	spin_unlock(&key->u.tkip.txlock);
+	pn = atomic64_inc_return(&key->conf.tx_pn);
+	pos = ieee80211_tkip_add_iv(pos, &key->conf, pn);
 
 	/* hwaccel - with software IV */
 	if (info->control.hw_key)
-- 
cgit v0.10.2


From 1ad4f639cc3e09d6c8402a0fcc592e2391683e31 Mon Sep 17 00:00:00 2001
From: Eliad Peller <eliad@wizery.com>
Date: Sun, 14 Feb 2016 13:56:36 +0200
Subject: iwlwifi: mvm: move TX PN assignment for TKIP to the driver

If protocol offloading is configured, the fw might generate some
frames (e.g. arp response) on its own during d3/d0i3.

On d3/d0i3 exit the driver queries the updated PN (if relevant),
and updates its keys (for the d0i3 case, this is done by
iwl_mvm_d0i3_exit_work(), which is scheduled on d0i3 exit)

While in d0i3, iwlmvm defers tx frames until d0i3 exit, and
then continues their processing.

This is problematic with TKIP, since the frame's PN has already
been set at this stage (in contrast to CCMP, where the PN is
being set only later on), so both the frame's PN and the upcoming
PN update (from d0i3 exit work) might be wrong.

Fix it by moving the TX PN assignment (for TKIP) to the driver,
similarly to CCMP.

Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index d3e21d9..93e4958 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -249,16 +249,19 @@ static void iwl_mvm_wowlan_program_keys(struct ieee80211_hw *hw,
 		return;
 	case WLAN_CIPHER_SUITE_TKIP:
 		if (sta) {
+			u64 pn64;
+
 			tkip_sc = data->rsc_tsc->all_tsc_rsc.tkip.unicast_rsc;
 			tkip_tx_sc = &data->rsc_tsc->all_tsc_rsc.tkip.tsc;
 
 			rx_p1ks = data->tkip->rx_uni;
 
-			ieee80211_get_key_tx_seq(key, &seq);
-			tkip_tx_sc->iv16 = cpu_to_le16(seq.tkip.iv16);
-			tkip_tx_sc->iv32 = cpu_to_le32(seq.tkip.iv32);
+			pn64 = atomic64_read(&key->tx_pn);
+			tkip_tx_sc->iv16 = cpu_to_le16(TKIP_PN_TO_IV16(pn64));
+			tkip_tx_sc->iv32 = cpu_to_le32(TKIP_PN_TO_IV32(pn64));
 
-			ieee80211_get_tkip_p1k_iv(key, seq.tkip.iv32, p1k);
+			ieee80211_get_tkip_p1k_iv(key, TKIP_PN_TO_IV32(pn64),
+						  p1k);
 			iwl_mvm_convert_p1k(p1k, data->tkip->tx.p1k);
 
 			memcpy(data->tkip->mic_keys.tx,
@@ -1601,7 +1604,9 @@ static void iwl_mvm_d3_update_keys(struct ieee80211_hw *hw,
 		case WLAN_CIPHER_SUITE_TKIP:
 			iwl_mvm_tkip_sc_to_seq(&sc->tkip.tsc, &seq);
 			iwl_mvm_set_tkip_rx_seq(sc->tkip.unicast_rsc, key);
-			ieee80211_set_key_tx_seq(key, &seq);
+			atomic64_set(&key->tx_pn,
+				     (u64)seq.tkip.iv16 |
+				     ((u64)seq.tkip.iv32 << 16));
 			break;
 		}
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 1bd3f0b..2b53292 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2585,7 +2585,7 @@ static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 	switch (key->cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
 		key->flags |= IEEE80211_KEY_FLAG_GENERATE_MMIC;
-		key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV;
+		key->flags |= IEEE80211_KEY_FLAG_PUT_IV_SPACE;
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 		key->flags |= IEEE80211_KEY_FLAG_PUT_IV_SPACE;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 8bf48a7..ca1e485 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -299,6 +299,8 @@ static void iwl_mvm_set_tx_cmd_crypto(struct iwl_mvm *mvm,
 
 	case WLAN_CIPHER_SUITE_TKIP:
 		tx_cmd->sec_ctl = TX_CMD_SEC_TKIP;
+		pn = atomic64_inc_return(&keyconf->tx_pn);
+		ieee80211_tkip_add_iv(crypto_hdr, keyconf, pn);
 		ieee80211_get_tkip_p2k(keyconf, skb_frag, tx_cmd->key);
 		break;
 
-- 
cgit v0.10.2


From ca48ebbc7ea7e82e3ae4b55aacead0cdb54ff008 Mon Sep 17 00:00:00 2001
From: Eliad Peller <eliad@wizery.com>
Date: Mon, 15 Feb 2016 12:34:10 +0200
Subject: mac80211: remove ieee80211_get_key_tx_seq/ieee80211_set_key_tx_seq

Since the PNs of all the tx keys are now tracked in the public
part of the key struct (with atomic counter), we no longer
need these functions.

dvm and vt665{5,6} are currently the only users of these functions,
so update them accordingly.

Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/lib.c b/drivers/net/wireless/intel/iwlwifi/dvm/lib.c
index 4841be2..1799469 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/lib.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/lib.c
@@ -943,14 +943,16 @@ static void iwlagn_wowlan_program_keys(struct ieee80211_hw *hw,
 	switch (key->cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
 		if (sta) {
+			u64 pn64;
+
 			tkip_sc = data->rsc_tsc->all_tsc_rsc.tkip.unicast_rsc;
 			tkip_tx_sc = &data->rsc_tsc->all_tsc_rsc.tkip.tsc;
 
 			rx_p1ks = data->tkip->rx_uni;
 
-			ieee80211_get_key_tx_seq(key, &seq);
-			tkip_tx_sc->iv16 = cpu_to_le16(seq.tkip.iv16);
-			tkip_tx_sc->iv32 = cpu_to_le32(seq.tkip.iv32);
+			pn64 = atomic64_read(&key->tx_pn);
+			tkip_tx_sc->iv16 = cpu_to_le16(TKIP_PN_TO_IV16(pn64));
+			tkip_tx_sc->iv32 = cpu_to_le32(TKIP_PN_TO_IV32(pn64));
 
 			ieee80211_get_tkip_p1k_iv(key, seq.tkip.iv32, p1k);
 			iwlagn_convert_p1k(p1k, data->tkip->tx.p1k);
@@ -996,19 +998,13 @@ static void iwlagn_wowlan_program_keys(struct ieee80211_hw *hw,
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
 		if (sta) {
-			u8 *pn = seq.ccmp.pn;
+			u64 pn64;
 
 			aes_sc = data->rsc_tsc->all_tsc_rsc.aes.unicast_rsc;
 			aes_tx_sc = &data->rsc_tsc->all_tsc_rsc.aes.tsc;
 
-			ieee80211_get_key_tx_seq(key, &seq);
-			aes_tx_sc->pn = cpu_to_le64(
-					(u64)pn[5] |
-					((u64)pn[4] << 8) |
-					((u64)pn[3] << 16) |
-					((u64)pn[2] << 24) |
-					((u64)pn[1] << 32) |
-					((u64)pn[0] << 40));
+			pn64 = atomic64_read(&key->tx_pn);
+			aes_tx_sc->pn = cpu_to_le64(pn64);
 		} else
 			aes_sc = data->rsc_tsc->all_tsc_rsc.aes.multicast_rsc;
 
diff --git a/drivers/staging/vt6655/rxtx.c b/drivers/staging/vt6655/rxtx.c
index b668db6..1a2dda0 100644
--- a/drivers/staging/vt6655/rxtx.c
+++ b/drivers/staging/vt6655/rxtx.c
@@ -1210,7 +1210,7 @@ static void vnt_fill_txkey(struct ieee80211_hdr *hdr, u8 *key_buffer,
 			   struct sk_buff *skb,	u16 payload_len,
 			   struct vnt_mic_hdr *mic_hdr)
 {
-	struct ieee80211_key_seq seq;
+	u64 pn64;
 	u8 *iv = ((u8 *)hdr + ieee80211_get_hdrlen_from_skb(skb));
 
 	/* strip header and icv len from payload */
@@ -1243,9 +1243,13 @@ static void vnt_fill_txkey(struct ieee80211_hdr *hdr, u8 *key_buffer,
 		mic_hdr->payload_len = cpu_to_be16(payload_len);
 		ether_addr_copy(mic_hdr->mic_addr2, hdr->addr2);
 
-		ieee80211_get_key_tx_seq(tx_key, &seq);
-
-		memcpy(mic_hdr->ccmp_pn, seq.ccmp.pn, IEEE80211_CCMP_PN_LEN);
+		pn64 = atomic64_read(&tx_key->tx_pn);
+		mic_hdr->ccmp_pn[5] = pn64;
+		mic_hdr->ccmp_pn[4] = pn64 >> 8;
+		mic_hdr->ccmp_pn[3] = pn64 >> 16;
+		mic_hdr->ccmp_pn[2] = pn64 >> 24;
+		mic_hdr->ccmp_pn[1] = pn64 >> 32;
+		mic_hdr->ccmp_pn[0] = pn64 >> 40;
 
 		if (ieee80211_has_a4(hdr->frame_control))
 			mic_hdr->hlen = cpu_to_be16(28);
diff --git a/drivers/staging/vt6656/rxtx.c b/drivers/staging/vt6656/rxtx.c
index efb54f5..76378d2 100644
--- a/drivers/staging/vt6656/rxtx.c
+++ b/drivers/staging/vt6656/rxtx.c
@@ -719,7 +719,7 @@ static void vnt_fill_txkey(struct vnt_usb_send_context *tx_context,
 	u16 payload_len, struct vnt_mic_hdr *mic_hdr)
 {
 	struct ieee80211_hdr *hdr = tx_context->hdr;
-	struct ieee80211_key_seq seq;
+	u64 pn64;
 	u8 *iv = ((u8 *)hdr + ieee80211_get_hdrlen_from_skb(skb));
 
 	/* strip header and icv len from payload */
@@ -752,9 +752,13 @@ static void vnt_fill_txkey(struct vnt_usb_send_context *tx_context,
 		mic_hdr->payload_len = cpu_to_be16(payload_len);
 		ether_addr_copy(mic_hdr->mic_addr2, hdr->addr2);
 
-		ieee80211_get_key_tx_seq(tx_key, &seq);
-
-		memcpy(mic_hdr->ccmp_pn, seq.ccmp.pn, IEEE80211_CCMP_PN_LEN);
+		pn64 = atomic64_read(&tx_key->tx_pn);
+		mic_hdr->ccmp_pn[5] = pn64;
+		mic_hdr->ccmp_pn[4] = pn64 >> 8;
+		mic_hdr->ccmp_pn[3] = pn64 >> 16;
+		mic_hdr->ccmp_pn[2] = pn64 >> 24;
+		mic_hdr->ccmp_pn[1] = pn64 >> 32;
+		mic_hdr->ccmp_pn[0] = pn64 >> 40;
 
 		if (ieee80211_has_a4(hdr->frame_control))
 			mic_hdr->hlen = cpu_to_be16(28);
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 15879b4..66155d3 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -4464,23 +4464,6 @@ void ieee80211_get_tkip_p2k(struct ieee80211_key_conf *keyconf,
 u8 *ieee80211_tkip_add_iv(u8 *pos, struct ieee80211_key_conf *keyconf, u64 pn);
 
 /**
- * ieee80211_get_key_tx_seq - get key TX sequence counter
- *
- * @keyconf: the parameter passed with the set key
- * @seq: buffer to receive the sequence data
- *
- * This function allows a driver to retrieve the current TX IV/PN
- * for the given key. It must not be called if IV generation is
- * offloaded to the device.
- *
- * Note that this function may only be called when no TX processing
- * can be done concurrently, for example when queues are stopped
- * and the stop has been synchronized.
- */
-void ieee80211_get_key_tx_seq(struct ieee80211_key_conf *keyconf,
-			      struct ieee80211_key_seq *seq);
-
-/**
  * ieee80211_get_key_rx_seq - get key RX sequence counter
  *
  * @keyconf: the parameter passed with the set key
@@ -4500,23 +4483,6 @@ void ieee80211_get_key_rx_seq(struct ieee80211_key_conf *keyconf,
 			      int tid, struct ieee80211_key_seq *seq);
 
 /**
- * ieee80211_set_key_tx_seq - set key TX sequence counter
- *
- * @keyconf: the parameter passed with the set key
- * @seq: new sequence data
- *
- * This function allows a driver to set the current TX IV/PNs for the
- * given key. This is useful when resuming from WoWLAN sleep and the
- * device may have transmitted frames using the PTK, e.g. replies to
- * ARP requests.
- *
- * Note that this function may only be called when no TX processing
- * can be done concurrently.
- */
-void ieee80211_set_key_tx_seq(struct ieee80211_key_conf *keyconf,
-			      struct ieee80211_key_seq *seq);
-
-/**
  * ieee80211_set_key_rx_seq - set key RX sequence counter
  *
  * @keyconf: the parameter passed with the set key
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index f9c4cb9..3df7b03 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -932,51 +932,6 @@ void ieee80211_gtk_rekey_notify(struct ieee80211_vif *vif, const u8 *bssid,
 }
 EXPORT_SYMBOL_GPL(ieee80211_gtk_rekey_notify);
 
-void ieee80211_get_key_tx_seq(struct ieee80211_key_conf *keyconf,
-			      struct ieee80211_key_seq *seq)
-{
-	struct ieee80211_key *key;
-	u64 pn64;
-
-	if (WARN_ON(!(keyconf->flags & IEEE80211_KEY_FLAG_GENERATE_IV)))
-		return;
-
-	key = container_of(keyconf, struct ieee80211_key, conf);
-
-	switch (key->conf.cipher) {
-	case WLAN_CIPHER_SUITE_TKIP:
-		pn64 = atomic64_read(&key->conf.tx_pn);
-		seq->tkip.iv32 = TKIP_PN_TO_IV32(pn64);
-		seq->tkip.iv16 = TKIP_PN_TO_IV16(pn64);
-		break;
-	case WLAN_CIPHER_SUITE_CCMP:
-	case WLAN_CIPHER_SUITE_CCMP_256:
-	case WLAN_CIPHER_SUITE_AES_CMAC:
-	case WLAN_CIPHER_SUITE_BIP_CMAC_256:
-		BUILD_BUG_ON(offsetof(typeof(*seq), ccmp) !=
-			     offsetof(typeof(*seq), aes_cmac));
-	case WLAN_CIPHER_SUITE_BIP_GMAC_128:
-	case WLAN_CIPHER_SUITE_BIP_GMAC_256:
-		BUILD_BUG_ON(offsetof(typeof(*seq), ccmp) !=
-			     offsetof(typeof(*seq), aes_gmac));
-	case WLAN_CIPHER_SUITE_GCMP:
-	case WLAN_CIPHER_SUITE_GCMP_256:
-		BUILD_BUG_ON(offsetof(typeof(*seq), ccmp) !=
-			     offsetof(typeof(*seq), gcmp));
-		pn64 = atomic64_read(&key->conf.tx_pn);
-		seq->ccmp.pn[5] = pn64;
-		seq->ccmp.pn[4] = pn64 >> 8;
-		seq->ccmp.pn[3] = pn64 >> 16;
-		seq->ccmp.pn[2] = pn64 >> 24;
-		seq->ccmp.pn[1] = pn64 >> 32;
-		seq->ccmp.pn[0] = pn64 >> 40;
-		break;
-	default:
-		WARN_ON(1);
-	}
-}
-EXPORT_SYMBOL(ieee80211_get_key_tx_seq);
-
 void ieee80211_get_key_rx_seq(struct ieee80211_key_conf *keyconf,
 			      int tid, struct ieee80211_key_seq *seq)
 {
@@ -1030,48 +985,6 @@ void ieee80211_get_key_rx_seq(struct ieee80211_key_conf *keyconf,
 }
 EXPORT_SYMBOL(ieee80211_get_key_rx_seq);
 
-void ieee80211_set_key_tx_seq(struct ieee80211_key_conf *keyconf,
-			      struct ieee80211_key_seq *seq)
-{
-	struct ieee80211_key *key;
-	u64 pn64;
-
-	key = container_of(keyconf, struct ieee80211_key, conf);
-
-	switch (key->conf.cipher) {
-	case WLAN_CIPHER_SUITE_TKIP:
-		pn64 = (u64)seq->tkip.iv16 | ((u64)seq->tkip.iv32 << 16);
-		atomic64_set(&key->conf.tx_pn, pn64);
-		break;
-	case WLAN_CIPHER_SUITE_CCMP:
-	case WLAN_CIPHER_SUITE_CCMP_256:
-	case WLAN_CIPHER_SUITE_AES_CMAC:
-	case WLAN_CIPHER_SUITE_BIP_CMAC_256:
-		BUILD_BUG_ON(offsetof(typeof(*seq), ccmp) !=
-			     offsetof(typeof(*seq), aes_cmac));
-	case WLAN_CIPHER_SUITE_BIP_GMAC_128:
-	case WLAN_CIPHER_SUITE_BIP_GMAC_256:
-		BUILD_BUG_ON(offsetof(typeof(*seq), ccmp) !=
-			     offsetof(typeof(*seq), aes_gmac));
-	case WLAN_CIPHER_SUITE_GCMP:
-	case WLAN_CIPHER_SUITE_GCMP_256:
-		BUILD_BUG_ON(offsetof(typeof(*seq), ccmp) !=
-			     offsetof(typeof(*seq), gcmp));
-		pn64 = (u64)seq->ccmp.pn[5] |
-		       ((u64)seq->ccmp.pn[4] << 8) |
-		       ((u64)seq->ccmp.pn[3] << 16) |
-		       ((u64)seq->ccmp.pn[2] << 24) |
-		       ((u64)seq->ccmp.pn[1] << 32) |
-		       ((u64)seq->ccmp.pn[0] << 40);
-		atomic64_set(&key->conf.tx_pn, pn64);
-		break;
-	default:
-		WARN_ON(1);
-		break;
-	}
-}
-EXPORT_SYMBOL_GPL(ieee80211_set_key_tx_seq);
-
 void ieee80211_set_key_rx_seq(struct ieee80211_key_conf *keyconf,
 			      int tid, struct ieee80211_key_seq *seq)
 {
-- 
cgit v0.10.2


From 65554d07adfc22bb9e14f6df8c609a646f869a74 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 16 Feb 2016 12:48:17 +0200
Subject: mac80211: provide interface to driver to set VHT MU-MIMO data

Provide an interface to the lower level driver to set the VHT
MU-MIMO data. This is needed for example when there is an update
of the group data during low power state, where the management
frame will not be passed to the host at all.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 66155d3..23f2a5e 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -5445,6 +5445,21 @@ ieee80211_vif_type_p2p(struct ieee80211_vif *vif)
 	return ieee80211_iftype_p2p(vif->type, vif->p2p);
 }
 
+/**
+ * ieee80211_update_mu_groups - set the VHT MU-MIMO groud data
+ *
+ * @vif: the specified virtual interface
+ * @membership: 64 bits array - a bit is set if station is member of the group
+ * @position: 2 bits per group id indicating the position in the group
+ *
+ * Note: This function assumes that the given vif is valid and the position and
+ * membership data is of the correct size and are in the same byte order as the
+ * matching GroupId management frame.
+ * Calls to this function need to be serialized with RX path.
+ */
+void ieee80211_update_mu_groups(struct ieee80211_vif *vif,
+				const u8 *membership, const u8 *position);
+
 void ieee80211_enable_rssi_reports(struct ieee80211_vif *vif,
 				   int rssi_min_thold,
 				   int rssi_max_thold);
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index 341d192..f8f1611 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -2,7 +2,7 @@
  * VHT handling
  *
  * Portions of this file
- * Copyright(c) 2015 Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -469,6 +469,20 @@ void ieee80211_process_mu_groups(struct ieee80211_sub_if_data *sdata,
 	ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_MU_GROUPS);
 }
 
+void ieee80211_update_mu_groups(struct ieee80211_vif *vif,
+				const u8 *membership, const u8 *position)
+{
+	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
+	struct ieee80211_bss_conf *bss_conf = &sdata->vif.bss_conf;
+
+	if (WARN_ON_ONCE(!(sdata->flags & IEEE80211_SDATA_MU_MIMO_OWNER)))
+		return;
+
+	memcpy(bss_conf->mu_group.membership, membership, WLAN_MEMBERSHIP_LEN);
+	memcpy(bss_conf->mu_group.position, position, WLAN_USER_POSITION_LEN);
+}
+EXPORT_SYMBOL_GPL(ieee80211_update_mu_groups);
+
 void ieee80211_vht_handle_opmode(struct ieee80211_sub_if_data *sdata,
 				 struct sta_info *sta, u8 opmode,
 				 enum ieee80211_band band)
-- 
cgit v0.10.2


From b5a33d52595f0cb153f09bf45a5dcd66a7418dbb Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 16 Feb 2016 12:48:18 +0200
Subject: mac80211: move MU_MIMO_OWNER flag to ieee80211_vif

Drivers may need to track which vif is using VHT MU-MIMO.
Move the flag indicationg the ownership of MU_MIMO to
ieee80211_vif.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 23f2a5e..0c09da3 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1382,6 +1382,7 @@ enum ieee80211_vif_flags {
  * @csa_active: marks whether a channel switch is going on. Internally it is
  *	write-protected by sdata_lock and local->mtx so holding either is fine
  *	for read access.
+ * @mu_mimo_owner: indicates interface owns MU-MIMO capability
  * @driver_flags: flags/capabilities the driver has for this interface,
  *	these need to be set (or cleared) when the interface is added
  *	or, if supported by the driver, the interface type is changed
@@ -1408,6 +1409,7 @@ struct ieee80211_vif {
 	u8 addr[ETH_ALEN];
 	bool p2p;
 	bool csa_active;
+	bool mu_mimo_owner;
 
 	u8 cab_queue;
 	u8 hw_queue[IEEE80211_NUM_ACS];
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index a49c103..1630975 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -716,7 +716,6 @@ struct ieee80211_if_mesh {
  *	back to wireless media and to the local net stack.
  * @IEEE80211_SDATA_DISCONNECT_RESUME: Disconnect after resume.
  * @IEEE80211_SDATA_IN_DRIVER: indicates interface was added to driver
- * @IEEE80211_SDATA_MU_MIMO_OWNER: indicates interface owns MU-MIMO capability
  */
 enum ieee80211_sub_if_data_flags {
 	IEEE80211_SDATA_ALLMULTI		= BIT(0),
@@ -724,7 +723,6 @@ enum ieee80211_sub_if_data_flags {
 	IEEE80211_SDATA_DONT_BRIDGE_PACKETS	= BIT(3),
 	IEEE80211_SDATA_DISCONNECT_RESUME	= BIT(4),
 	IEEE80211_SDATA_IN_DRIVER		= BIT(5),
-	IEEE80211_SDATA_MU_MIMO_OWNER		= BIT(6),
 };
 
 /**
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 857089d..f41625b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -6,7 +6,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
- * Copyright (C) 2015 Intel Deutschland GmbH
+ * Copyright (C) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -559,7 +559,7 @@ static void ieee80211_add_vht_ie(struct ieee80211_sub_if_data *sdata,
 		struct ieee80211_sub_if_data *other;
 
 		list_for_each_entry_rcu(other, &local->interfaces, list) {
-			if (other->flags & IEEE80211_SDATA_MU_MIMO_OWNER) {
+			if (other->vif.mu_mimo_owner) {
 				disable_mu_mimo = true;
 				break;
 			}
@@ -567,7 +567,7 @@ static void ieee80211_add_vht_ie(struct ieee80211_sub_if_data *sdata,
 		if (disable_mu_mimo)
 			cap &= ~IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE;
 		else
-			sdata->flags |= IEEE80211_SDATA_MU_MIMO_OWNER;
+			sdata->vif.mu_mimo_owner = true;
 	}
 
 	mask = IEEE80211_VHT_CAP_BEAMFORMEE_STS_MASK;
@@ -2052,7 +2052,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	memset(sdata->vif.bss_conf.mu_group.position, 0,
 	       sizeof(sdata->vif.bss_conf.mu_group.position));
 	changed |= BSS_CHANGED_MU_GROUPS;
-	sdata->flags &= ~IEEE80211_SDATA_MU_MIMO_OWNER;
+	sdata->vif.mu_mimo_owner = false;
 
 	sdata->ap_power_level = IEEE80211_UNSET_POWER_LEVEL;
 
@@ -2509,7 +2509,8 @@ static void ieee80211_destroy_assoc_data(struct ieee80211_sub_if_data *sdata,
 		eth_zero_addr(sdata->u.mgd.bssid);
 		ieee80211_bss_info_change_notify(sdata, BSS_CHANGED_BSSID);
 		sdata->u.mgd.flags = 0;
-		sdata->flags &= ~IEEE80211_SDATA_MU_MIMO_OWNER;
+		sdata->vif.mu_mimo_owner = false;
+
 		mutex_lock(&sdata->local->mtx);
 		ieee80211_vif_release_channel(sdata);
 		mutex_unlock(&sdata->local->mtx);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index f1e5b76..89f7179 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -1928,7 +1928,7 @@ int ieee80211_reconfig(struct ieee80211_local *local)
 			  BSS_CHANGED_IDLE |
 			  BSS_CHANGED_TXPOWER;
 
-		if (sdata->flags & IEEE80211_SDATA_MU_MIMO_OWNER)
+		if (sdata->vif.mu_mimo_owner)
 			changed |= BSS_CHANGED_MU_GROUPS;
 
 		switch (sdata->vif.type) {
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index f8f1611..89e04d5 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -450,7 +450,7 @@ void ieee80211_process_mu_groups(struct ieee80211_sub_if_data *sdata,
 {
 	struct ieee80211_bss_conf *bss_conf = &sdata->vif.bss_conf;
 
-	if (!(sdata->flags & IEEE80211_SDATA_MU_MIMO_OWNER))
+	if (!sdata->vif.mu_mimo_owner)
 		return;
 
 	if (!memcmp(mgmt->u.action.u.vht_group_notif.position,
@@ -472,10 +472,9 @@ void ieee80211_process_mu_groups(struct ieee80211_sub_if_data *sdata,
 void ieee80211_update_mu_groups(struct ieee80211_vif *vif,
 				const u8 *membership, const u8 *position)
 {
-	struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif);
-	struct ieee80211_bss_conf *bss_conf = &sdata->vif.bss_conf;
+	struct ieee80211_bss_conf *bss_conf = &vif->bss_conf;
 
-	if (WARN_ON_ONCE(!(sdata->flags & IEEE80211_SDATA_MU_MIMO_OWNER)))
+	if (WARN_ON_ONCE(!vif->mu_mimo_owner))
 		return;
 
 	memcpy(bss_conf->mu_group.membership, membership, WLAN_MEMBERSHIP_LEN);
-- 
cgit v0.10.2


From 0c9ca11b1ae8eb16c1b6bbae91991392d2321372 Mon Sep 17 00:00:00 2001
From: Beni Lev <beni.lev@intel.com>
Date: Wed, 17 Feb 2016 20:30:00 +0200
Subject: cfg80211: Add global RRM capability

Today, the supplicant will add the RRM capabilities
Information Element in the association request only if
Quiet period is supported (NL80211_FEATURE_QUIET).

Quiet is one of many RRM features, and there are other RRM
features that are not related to Quiet (e.g. neighbor
report). Therefore, requiring Quiet to enable RRM is too
restrictive.
Some of the features, like neighbor report, can be
supported by user space without any help from the kernel.
Hence adding the RRM capabilities IE to association request
should be the sole user space's decision.
Removing the RRM dependency on Quiet in the driver solves
this problem, but using an old driver with a user space
tool that would not require Quiet feature would be
problematic: the user space would add NL80211_ATTR_USE_RRM
in the association request even if the kernel doesn't
advertize NL80211_FEATURE_QUIET and the association would
be denied by the kernel.

This solution adds a global RRM capability, that tells user
space that it can request RRM capabilities IE publishment
without any specific feature support in the kernel.

Signed-off-by: Beni Lev <beni.lev@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index 7758969..5a30a75 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -1727,6 +1727,8 @@ enum nl80211_commands {
  *	underlying device supports these minimal RRM features:
  *		%NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES,
  *		%NL80211_FEATURE_QUIET,
+ *	Or, if global RRM is supported, see:
+ *		%NL80211_EXT_FEATURE_RRM
  *	If this flag is used, driver must add the Power Capabilities IE to the
  *	association request. In addition, it must also set the RRM capability
  *	flag in the association request's Capability Info field.
@@ -4402,12 +4404,18 @@ enum nl80211_feature_flags {
 /**
  * enum nl80211_ext_feature_index - bit index of extended features.
  * @NL80211_EXT_FEATURE_VHT_IBSS: This driver supports IBSS with VHT datarates.
+ * @NL80211_EXT_FEATURE_RRM: This driver supports RRM. When featured, user can
+ *	can request to use RRM (see %NL80211_ATTR_USE_RRM) with
+ *	%NL80211_CMD_ASSOCIATE and %NL80211_CMD_CONNECT requests, which will set
+ *	the ASSOC_REQ_USE_RRM flag in the association request even if
+ *	NL80211_FEATURE_QUIET is not advertized.
  *
  * @NUM_NL80211_EXT_FEATURES: number of extended features.
  * @MAX_NL80211_EXT_FEATURES: highest extended feature index.
  */
 enum nl80211_ext_feature_index {
 	NL80211_EXT_FEATURE_VHT_IBSS,
+	NL80211_EXT_FEATURE_RRM,
 
 	/* add new features before the definition below */
 	NUM_NL80211_EXT_FEATURES,
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 268cb49..90890f1 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -3,7 +3,7 @@
  *
  * Copyright 2006-2010	Johannes Berg <johannes@sipsolutions.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
- * Copyright 2015	Intel Deutschland GmbH
+ * Copyright 2015-2016	Intel Deutschland GmbH
  */
 
 #include <linux/if.h>
@@ -7286,9 +7286,11 @@ static int nl80211_associate(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	if (nla_get_flag(info->attrs[NL80211_ATTR_USE_RRM])) {
-		if (!(rdev->wiphy.features &
-		      NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES) ||
-		    !(rdev->wiphy.features & NL80211_FEATURE_QUIET))
+		if (!((rdev->wiphy.features &
+			NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES) &&
+		       (rdev->wiphy.features & NL80211_FEATURE_QUIET)) &&
+		    !wiphy_ext_feature_isset(&rdev->wiphy,
+					     NL80211_EXT_FEATURE_RRM))
 			return -EINVAL;
 		req.flags |= ASSOC_REQ_USE_RRM;
 	}
@@ -7976,9 +7978,11 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	if (nla_get_flag(info->attrs[NL80211_ATTR_USE_RRM])) {
-		if (!(rdev->wiphy.features &
-		      NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES) ||
-		    !(rdev->wiphy.features & NL80211_FEATURE_QUIET)) {
+		if (!((rdev->wiphy.features &
+			NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES) &&
+		       (rdev->wiphy.features & NL80211_FEATURE_QUIET)) &&
+		    !wiphy_ext_feature_isset(&rdev->wiphy,
+					     NL80211_EXT_FEATURE_RRM)) {
 			kzfree(connkeys);
 			return -EINVAL;
 		}
-- 
cgit v0.10.2


From 7837a7778268191dae5f6622f2b92b8b37cb8d7f Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com>
Date: Fri, 19 Feb 2016 11:43:04 +0100
Subject: cfg80211: add radiotap VHT info to rtap_namespace_sizes

Add IEEE80211_RADIOTAP_VHT entry to rtap_namespace_sizes array in order to
define alignment and size of VHT info in tx radiotap

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/wireless/radiotap.c b/net/wireless/radiotap.c
index 722da61..6582d15 100644
--- a/net/wireless/radiotap.c
+++ b/net/wireless/radiotap.c
@@ -43,6 +43,7 @@ static const struct radiotap_align_size rtap_namespace_sizes[] = {
 	[IEEE80211_RADIOTAP_DATA_RETRIES] = { .align = 1, .size = 1, },
 	[IEEE80211_RADIOTAP_MCS] = { .align = 1, .size = 3, },
 	[IEEE80211_RADIOTAP_AMPDU_STATUS] = { .align = 4, .size = 8, },
+	[IEEE80211_RADIOTAP_VHT] = { .align = 2, .size = 12, },
 	/*
 	 * add more here as they are defined in radiotap.h
 	 */
-- 
cgit v0.10.2


From 648b50dd6abf8e6e5b589bb8e6873a4596389dbe Mon Sep 17 00:00:00 2001
From: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Date: Mon, 25 Jan 2016 12:03:46 +0300
Subject: net: rfkill: add rfkill_find_type function

Helper for finding the type based on name. Useful if the
type needs to be determined based on device property.

Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
[modify rfkill_types array and BUILD_BUG_ON to not cause errors]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index 7af625f..e6a0031 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -213,6 +213,15 @@ void rfkill_set_states(struct rfkill *rfkill, bool sw, bool hw);
  * @rfkill: rfkill struct to query
  */
 bool rfkill_blocked(struct rfkill *rfkill);
+
+/**
+ * rfkill_find_type - Helpper for finding rfkill type by name
+ * @name: the name of the type
+ *
+ * Returns enum rfkill_type that conrresponds the name.
+ */
+enum rfkill_type rfkill_find_type(const char *name);
+
 #else /* !RFKILL */
 static inline struct rfkill * __must_check
 rfkill_alloc(const char *name,
@@ -269,6 +278,12 @@ static inline bool rfkill_blocked(struct rfkill *rfkill)
 {
 	return false;
 }
+
+static inline enum rfkill_type rfkill_find_type(const char *name)
+{
+	return RFKILL_TYPE_ALL;
+}
+
 #endif /* RFKILL || RFKILL_MODULE */
 
 
diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index a805831..2a23479 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -572,6 +572,34 @@ void rfkill_set_states(struct rfkill *rfkill, bool sw, bool hw)
 }
 EXPORT_SYMBOL(rfkill_set_states);
 
+static const char * const rfkill_types[] = {
+	NULL, /* RFKILL_TYPE_ALL */
+	"wlan",
+	"bluetooth",
+	"ultrawideband",
+	"wimax",
+	"wwan",
+	"gps",
+	"fm",
+	"nfc",
+};
+
+enum rfkill_type rfkill_find_type(const char *name)
+{
+	int i;
+
+	BUILD_BUG_ON(ARRAY_SIZE(rfkill_types) != NUM_RFKILL_TYPES);
+
+	if (!name)
+		return RFKILL_TYPE_ALL;
+
+	for (i = 1; i < NUM_RFKILL_TYPES; i++)
+		if (!strcmp(name, rfkill_types[i]))
+			return i;
+	return RFKILL_TYPE_ALL;
+}
+EXPORT_SYMBOL(rfkill_find_type);
+
 static ssize_t name_show(struct device *dev, struct device_attribute *attr,
 			 char *buf)
 {
@@ -581,38 +609,12 @@ static ssize_t name_show(struct device *dev, struct device_attribute *attr,
 }
 static DEVICE_ATTR_RO(name);
 
-static const char *rfkill_get_type_str(enum rfkill_type type)
-{
-	BUILD_BUG_ON(NUM_RFKILL_TYPES != RFKILL_TYPE_NFC + 1);
-
-	switch (type) {
-	case RFKILL_TYPE_WLAN:
-		return "wlan";
-	case RFKILL_TYPE_BLUETOOTH:
-		return "bluetooth";
-	case RFKILL_TYPE_UWB:
-		return "ultrawideband";
-	case RFKILL_TYPE_WIMAX:
-		return "wimax";
-	case RFKILL_TYPE_WWAN:
-		return "wwan";
-	case RFKILL_TYPE_GPS:
-		return "gps";
-	case RFKILL_TYPE_FM:
-		return "fm";
-	case RFKILL_TYPE_NFC:
-		return "nfc";
-	default:
-		BUG();
-	}
-}
-
 static ssize_t type_show(struct device *dev, struct device_attribute *attr,
 			 char *buf)
 {
 	struct rfkill *rfkill = to_rfkill(dev);
 
-	return sprintf(buf, "%s\n", rfkill_get_type_str(rfkill->type));
+	return sprintf(buf, "%s\n", rfkill_types[rfkill->type]);
 }
 static DEVICE_ATTR_RO(type);
 
@@ -750,7 +752,7 @@ static int rfkill_dev_uevent(struct device *dev, struct kobj_uevent_env *env)
 	if (error)
 		return error;
 	error = add_uevent_var(env, "RFKILL_TYPE=%s",
-			       rfkill_get_type_str(rfkill->type));
+			       rfkill_types[rfkill->type]);
 	if (error)
 		return error;
 	spin_lock_irqsave(&rfkill->lock, flags);
-- 
cgit v0.10.2


From 7d5e9737efda16535e5b54bd627ef4881d11d31f Mon Sep 17 00:00:00 2001
From: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Date: Mon, 25 Jan 2016 12:03:47 +0300
Subject: net: rfkill: gpio: get the name and type from device property

This prepares the driver for removal of platform data.

Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/rfkill/rfkill-gpio.c b/net/rfkill/rfkill-gpio.c
index 4b1e3f3..1a9c031 100644
--- a/net/rfkill/rfkill-gpio.c
+++ b/net/rfkill/rfkill-gpio.c
@@ -81,7 +81,6 @@ static int rfkill_gpio_acpi_probe(struct device *dev,
 	if (!id)
 		return -ENODEV;
 
-	rfkill->name = dev_name(dev);
 	rfkill->type = (unsigned)id->driver_data;
 
 	return acpi_dev_add_driver_gpios(ACPI_COMPANION(dev),
@@ -93,12 +92,21 @@ static int rfkill_gpio_probe(struct platform_device *pdev)
 	struct rfkill_gpio_platform_data *pdata = pdev->dev.platform_data;
 	struct rfkill_gpio_data *rfkill;
 	struct gpio_desc *gpio;
+	const char *type_name;
 	int ret;
 
 	rfkill = devm_kzalloc(&pdev->dev, sizeof(*rfkill), GFP_KERNEL);
 	if (!rfkill)
 		return -ENOMEM;
 
+	device_property_read_string(&pdev->dev, "name", &rfkill->name);
+	device_property_read_string(&pdev->dev, "type", &type_name);
+
+	if (!rfkill->name)
+		rfkill->name = dev_name(&pdev->dev);
+
+	rfkill->type = rfkill_find_type(type_name);
+
 	if (ACPI_HANDLE(&pdev->dev)) {
 		ret = rfkill_gpio_acpi_probe(&pdev->dev, rfkill);
 		if (ret)
@@ -124,10 +132,8 @@ static int rfkill_gpio_probe(struct platform_device *pdev)
 
 	rfkill->shutdown_gpio = gpio;
 
-	/* Make sure at-least one of the GPIO is defined and that
-	 * a name is specified for this instance
-	 */
-	if ((!rfkill->reset_gpio && !rfkill->shutdown_gpio) || !rfkill->name) {
+	/* Make sure at-least one GPIO is defined for this instance */
+	if (!rfkill->reset_gpio && !rfkill->shutdown_gpio) {
 		dev_err(&pdev->dev, "invalid platform data\n");
 		return -EINVAL;
 	}
-- 
cgit v0.10.2


From d54bbaf45af04982d91b424ae0d2e8fe2fb78bd8 Mon Sep 17 00:00:00 2001
From: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Date: Mon, 25 Jan 2016 12:03:48 +0300
Subject: ARM: tegra: use build-in device properties with rfkill_gpio

Pass the rfkill name and type to the device with properties
instead of driver specific platform data.

Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
CC: Alexandre Courbot <gnurou@gmail.com>
CC: Thierry Reding <thierry.reding@gmail.com>
CC: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Thierry Reding <treding@nvidia.com>
Tested-by: Marc Dietrich <marvin24@gmx.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/arch/arm/mach-tegra/board-paz00.c b/arch/arm/mach-tegra/board-paz00.c
index 49d1110..52db8bf 100644
--- a/arch/arm/mach-tegra/board-paz00.c
+++ b/arch/arm/mach-tegra/board-paz00.c
@@ -17,23 +17,25 @@
  *
  */
 
+#include <linux/property.h>
 #include <linux/gpio/machine.h>
 #include <linux/platform_device.h>
-#include <linux/rfkill-gpio.h>
 
 #include "board.h"
 
-static struct rfkill_gpio_platform_data wifi_rfkill_platform_data = {
-	.name	= "wifi_rfkill",
-	.type	= RFKILL_TYPE_WLAN,
+static struct property_entry __initdata wifi_rfkill_prop[] = {
+	PROPERTY_ENTRY_STRING("name", "wifi_rfkill"),
+	PROPERTY_ENTRY_STRING("type", "wlan"),
+	{ },
+};
+
+static struct property_set __initdata wifi_rfkill_pset = {
+	.properties = wifi_rfkill_prop,
 };
 
 static struct platform_device wifi_rfkill_device = {
 	.name	= "rfkill_gpio",
 	.id	= -1,
-	.dev	= {
-		.platform_data = &wifi_rfkill_platform_data,
-	},
 };
 
 static struct gpiod_lookup_table wifi_gpio_lookup = {
@@ -47,6 +49,7 @@ static struct gpiod_lookup_table wifi_gpio_lookup = {
 
 void __init tegra_paz00_wifikill_init(void)
 {
+	platform_device_add_properties(&wifi_rfkill_device, &wifi_rfkill_pset);
 	gpiod_add_lookup_table(&wifi_gpio_lookup);
 	platform_device_register(&wifi_rfkill_device);
 }
-- 
cgit v0.10.2


From fb2e6b7b7b02ab35a9d5355a69097a6f60c69d38 Mon Sep 17 00:00:00 2001
From: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Date: Mon, 25 Jan 2016 12:03:49 +0300
Subject: net: rfkill: gpio: remove rfkill_gpio_platform_data

No more users for it.

Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/include/linux/rfkill-gpio.h b/include/linux/rfkill-gpio.h
deleted file mode 100644
index 20bcb55..0000000
--- a/include/linux/rfkill-gpio.h
+++ /dev/null
@@ -1,37 +0,0 @@
-/*
- * Copyright (c) 2011, NVIDIA Corporation.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
- */
-
-
-#ifndef __RFKILL_GPIO_H
-#define __RFKILL_GPIO_H
-
-#include <linux/types.h>
-#include <linux/rfkill.h>
-
-/**
- * struct rfkill_gpio_platform_data - platform data for rfkill gpio device.
- * for unused gpio's, the expected value is -1.
- * @name:		name for the gpio rf kill instance
- */
-
-struct rfkill_gpio_platform_data {
-	char			*name;
-	enum rfkill_type	type;
-};
-
-#endif /* __RFKILL_GPIO_H */
diff --git a/net/rfkill/Kconfig b/net/rfkill/Kconfig
index 598d374..868f1ad 100644
--- a/net/rfkill/Kconfig
+++ b/net/rfkill/Kconfig
@@ -41,5 +41,4 @@ config RFKILL_GPIO
 	default n
 	help
 	  If you say yes here you get support of a generic gpio RFKILL
-	  driver. The platform should fill in the appropriate fields in the
-	  rfkill_gpio_platform_data structure and pass that to the driver.
+	  driver.
diff --git a/net/rfkill/rfkill-gpio.c b/net/rfkill/rfkill-gpio.c
index 1a9c031..76c01cb 100644
--- a/net/rfkill/rfkill-gpio.c
+++ b/net/rfkill/rfkill-gpio.c
@@ -27,8 +27,6 @@
 #include <linux/acpi.h>
 #include <linux/gpio/consumer.h>
 
-#include <linux/rfkill-gpio.h>
-
 struct rfkill_gpio_data {
 	const char		*name;
 	enum rfkill_type	type;
@@ -89,7 +87,6 @@ static int rfkill_gpio_acpi_probe(struct device *dev,
 
 static int rfkill_gpio_probe(struct platform_device *pdev)
 {
-	struct rfkill_gpio_platform_data *pdata = pdev->dev.platform_data;
 	struct rfkill_gpio_data *rfkill;
 	struct gpio_desc *gpio;
 	const char *type_name;
@@ -111,11 +108,6 @@ static int rfkill_gpio_probe(struct platform_device *pdev)
 		ret = rfkill_gpio_acpi_probe(&pdev->dev, rfkill);
 		if (ret)
 			return ret;
-	} else if (pdata) {
-		rfkill->name = pdata->name;
-		rfkill->type = pdata->type;
-	} else {
-		return -ENODEV;
 	}
 
 	rfkill->clk = devm_clk_get(&pdev->dev, NULL);
-- 
cgit v0.10.2


From 3ff707d66881d308ef71a0939aa7a92ae1290702 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Mon, 22 Feb 2016 11:36:32 -0500
Subject: rfkill: Improve documentation language
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 2a23479..8f6906e 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -282,8 +282,8 @@ static void rfkill_set_block(struct rfkill *rfkill, bool blocked)
 	spin_lock_irqsave(&rfkill->lock, flags);
 	if (err) {
 		/*
-		 * Failed -- reset status to _prev, this may be different
-		 * from what set set _PREV to earlier in this function
+		 * Failed -- reset status to _PREV, which may be different
+		 * from what we have set _PREV to earlier in this function
 		 * if rfkill_set_sw_state was invoked.
 		 */
 		if (rfkill->state & RFKILL_BLOCK_SW_PREV)
-- 
cgit v0.10.2


From 1a1078901b72e5e8046579cc54250e4f167269d5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Mon, 22 Feb 2016 11:36:33 -0500
Subject: rfkill: Remove extra blank line
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 8f6906e..f843eee 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -455,7 +455,6 @@ bool rfkill_get_global_sw_state(const enum rfkill_type type)
 }
 #endif
 
-
 bool rfkill_set_hw_state(struct rfkill *rfkill, bool blocked)
 {
 	unsigned long flags;
-- 
cgit v0.10.2


From 26b359c85d6ce0960ca122adfa7c1229796f55d7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Mon, 22 Feb 2016 11:36:34 -0500
Subject: rfkill: Point to the correct deprecated doc location
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The "claim" sysfs interface has been removed, so its documentation now
lives in the "removed" folder.

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/Documentation/ABI/stable/sysfs-class-rfkill b/Documentation/ABI/stable/sysfs-class-rfkill
index 097f522..e51571e 100644
--- a/Documentation/ABI/stable/sysfs-class-rfkill
+++ b/Documentation/ABI/stable/sysfs-class-rfkill
@@ -2,8 +2,10 @@ rfkill - radio frequency (RF) connector kill switch support
 
 For details to this subsystem look at Documentation/rfkill.txt.
 
-For the deprecated /sys/class/rfkill/*/state and
-/sys/class/rfkill/*/claim knobs of this interface look in
+For the deprecated /sys/class/rfkill/*/claim knobs of this interface look in
+Documentation/ABI/removed/sysfs-class-rfkill.
+
+For the deprecated /sys/class/rfkill/*/state knobs of this interface look in
 Documentation/ABI/obsolete/sysfs-class-rfkill.
 
 What: 		/sys/class/rfkill
-- 
cgit v0.10.2


From 4b636dc659ffbca8bc8069dcdf5ae5c2d088a163 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Mon, 22 Feb 2016 11:36:35 -0500
Subject: rfkill: Move "state" sysfs file back to stable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is still quite a bit of code using this interface, so we can't
just remove it. Hopefully it will be possible in the future, but since
its scheduled removal date is past 2 years already, we are better having
the documentation reflecting the current state of things.

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/Documentation/ABI/obsolete/sysfs-class-rfkill b/Documentation/ABI/obsolete/sysfs-class-rfkill
deleted file mode 100644
index e736d14..0000000
--- a/Documentation/ABI/obsolete/sysfs-class-rfkill
+++ /dev/null
@@ -1,20 +0,0 @@
-rfkill - radio frequency (RF) connector kill switch support
-
-For details to this subsystem look at Documentation/rfkill.txt.
-
-What:		/sys/class/rfkill/rfkill[0-9]+/state
-Date:		09-Jul-2007
-KernelVersion	v2.6.22
-Contact:	linux-wireless@vger.kernel.org
-Description: 	Current state of the transmitter.
-		This file is deprecated and scheduled to be removed in 2014,
-		because its not possible to express the 'soft and hard block'
-		state of the rfkill driver.
-Values: 	A numeric value.
-		0: RFKILL_STATE_SOFT_BLOCKED
-			transmitter is turned off by software
-		1: RFKILL_STATE_UNBLOCKED
-			transmitter is (potentially) active
-		2: RFKILL_STATE_HARD_BLOCKED
-			transmitter is forced off by something outside of
-			the driver's control.
diff --git a/Documentation/ABI/stable/sysfs-class-rfkill b/Documentation/ABI/stable/sysfs-class-rfkill
index e51571e..e1ba4a1 100644
--- a/Documentation/ABI/stable/sysfs-class-rfkill
+++ b/Documentation/ABI/stable/sysfs-class-rfkill
@@ -5,9 +5,6 @@ For details to this subsystem look at Documentation/rfkill.txt.
 For the deprecated /sys/class/rfkill/*/claim knobs of this interface look in
 Documentation/ABI/removed/sysfs-class-rfkill.
 
-For the deprecated /sys/class/rfkill/*/state knobs of this interface look in
-Documentation/ABI/obsolete/sysfs-class-rfkill.
-
 What: 		/sys/class/rfkill
 Date:		09-Jul-2007
 KernelVersion:	v2.6.22
@@ -44,6 +41,28 @@ Values: 	A numeric value.
 		1: true
 
 
+What:		/sys/class/rfkill/rfkill[0-9]+/state
+Date:		09-Jul-2007
+KernelVersion	v2.6.22
+Contact:	linux-wireless@vger.kernel.org
+Description: 	Current state of the transmitter.
+		This file was scheduled to be removed in 2014, but due to its
+		large number of users it will be sticking around for a bit
+		longer. Despite it being marked as stabe, the newer "hard" and
+		"soft" interfaces should be preffered, since it is not possible
+		to express the 'soft and hard block' state of the rfkill driver
+		through this interface. There will likely be another attempt to
+		remove it in the future.
+Values: 	A numeric value.
+		0: RFKILL_STATE_SOFT_BLOCKED
+			transmitter is turned off by software
+		1: RFKILL_STATE_UNBLOCKED
+			transmitter is (potentially) active
+		2: RFKILL_STATE_HARD_BLOCKED
+			transmitter is forced off by something outside of
+			the driver's control.
+
+
 What:		/sys/class/rfkill/rfkill[0-9]+/hard
 Date:		12-March-2010
 KernelVersion	v2.6.34
-- 
cgit v0.10.2


From 9487bd6b96a98ee07661a7b5e6f6afce67e2860b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Mon, 22 Feb 2016 11:36:36 -0500
Subject: rfkill: Factor rfkill_global_states[].cur assignments
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Factor all assignments to rfkill_global_states[].cur into a single
function rfkill_update_global_state().

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index f843eee..a8c05e1 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -302,6 +302,19 @@ static void rfkill_set_block(struct rfkill *rfkill, bool blocked)
 		rfkill_event(rfkill);
 }
 
+static void rfkill_update_global_state(enum rfkill_type type, bool blocked)
+{
+	int i;
+
+	if (type != RFKILL_TYPE_ALL) {
+		rfkill_global_states[type].cur = blocked;
+		return;
+	}
+
+	for (i = 0; i < NUM_RFKILL_TYPES; i++)
+		rfkill_global_states[i].cur = blocked;
+}
+
 #ifdef CONFIG_RFKILL_INPUT
 static atomic_t rfkill_input_disabled = ATOMIC_INIT(0);
 
@@ -319,15 +332,7 @@ static void __rfkill_switch_all(const enum rfkill_type type, bool blocked)
 {
 	struct rfkill *rfkill;
 
-	if (type == RFKILL_TYPE_ALL) {
-		int i;
-
-		for (i = 0; i < NUM_RFKILL_TYPES; i++)
-			rfkill_global_states[i].cur = blocked;
-	} else {
-		rfkill_global_states[type].cur = blocked;
-	}
-
+	rfkill_update_global_state(type, blocked);
 	list_for_each_entry(rfkill, &rfkill_list, node) {
 		if (rfkill->type != type && type != RFKILL_TYPE_ALL)
 			continue;
@@ -1166,15 +1171,8 @@ static ssize_t rfkill_fop_write(struct file *file, const char __user *buf,
 
 	mutex_lock(&rfkill_global_mutex);
 
-	if (ev.op == RFKILL_OP_CHANGE_ALL) {
-		if (ev.type == RFKILL_TYPE_ALL) {
-			enum rfkill_type i;
-			for (i = 0; i < NUM_RFKILL_TYPES; i++)
-				rfkill_global_states[i].cur = ev.soft;
-		} else {
-			rfkill_global_states[ev.type].cur = ev.soft;
-		}
-	}
+	if (ev.op == RFKILL_OP_CHANGE_ALL)
+		rfkill_update_global_state(ev.type, ev.soft);
 
 	list_for_each_entry(rfkill, &rfkill_list, node) {
 		if (rfkill->idx != ev.idx && ev.op != RFKILL_OP_CHANGE_ALL)
@@ -1263,10 +1261,8 @@ static struct miscdevice rfkill_miscdev = {
 static int __init rfkill_init(void)
 {
 	int error;
-	int i;
 
-	for (i = 0; i < NUM_RFKILL_TYPES; i++)
-		rfkill_global_states[i].cur = !rfkill_default_state;
+	rfkill_update_global_state(RFKILL_TYPE_ALL, !rfkill_default_state);
 
 	error = class_register(&rfkill_class);
 	if (error)
-- 
cgit v0.10.2


From 50ee738d7271fe825e4024cdfa5c5301a871e2c2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jo=C3=A3o=20Paulo=20Rechi=20Vita?= <jprvita@gmail.com>
Date: Mon, 22 Feb 2016 11:36:37 -0500
Subject: rfkill: Add documentation about LED triggers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: João Paulo Rechi Vita <jprvita@endlessm.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt
index 2ee6ef9..1f0c270 100644
--- a/Documentation/rfkill.txt
+++ b/Documentation/rfkill.txt
@@ -83,6 +83,8 @@ rfkill drivers that control devices that can be hard-blocked unless they also
 assign the poll_hw_block() callback (then the rfkill core will poll the
 device). Don't do this unless you cannot get the event in any other way.
 
+RFKill provides per-switch LED triggers, which can be used to drive LEDs
+according to the switch state (LED_FULL when blocked, LED_OFF otherwise).
 
 
 5. Userspace support
-- 
cgit v0.10.2


From 8790404de0e5f47ef721e2ed36ebb0f2314722ac Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Wed, 24 Feb 2016 12:01:03 +0100
Subject: 6lowpan: iphc: fix stateful multicast compression

In case of multicast address we need to set always the LOWPAN_IPHC_M bit
and if a destination context identifier was found for a multicast
address then we need to set the LOWPAN_IPHC_DAC as well.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c
index d2a565c..8dd8388 100644
--- a/net/6lowpan/iphc.c
+++ b/net/6lowpan/iphc.c
@@ -1118,12 +1118,13 @@ int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev,
 	/* destination address compression */
 	if (ipv6_daddr_type & IPV6_ADDR_MULTICAST) {
 		pr_debug("destination address is multicast: ");
+		iphc1 |= LOWPAN_IPHC_M;
 		if (dci) {
 			iphc1 |= lowpan_iphc_mcast_ctx_addr_compress(&hc_ptr,
 								     &dci_entry,
 								     &hdr->daddr);
+			iphc1 |= LOWPAN_IPHC_DAC;
 		} else {
-			iphc1 |= LOWPAN_IPHC_M;
 			iphc1 |= lowpan_iphc_mcast_addr_compress(&hc_ptr,
 								 &hdr->daddr);
 		}
-- 
cgit v0.10.2


From 4c23d8745731b1b54bb8eadfc42070d7f9893b46 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Wed, 24 Feb 2016 09:32:13 +0100
Subject: 6lowpan: fix error checking code

Bool variable 'fail' is always non-negative, it indicates an error if it
is true.

The problem has been detected using coccinelle script
scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Acked-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c
index 8dd8388..06287ac 100644
--- a/net/6lowpan/iphc.c
+++ b/net/6lowpan/iphc.c
@@ -482,7 +482,7 @@ static int lowpan_uncompress_multicast_ctx_daddr(struct sk_buff *skb,
 	ipaddr->s6_addr[0] = 0xFF;
 	fail = lowpan_fetch_skb(skb, &ipaddr->s6_addr[1], 2);
 	fail |= lowpan_fetch_skb(skb, &ipaddr->s6_addr[12], 4);
-	if (fail < 0)
+	if (fail)
 		return -EIO;
 
 	/* take prefix_len and network prefix from the context */
-- 
cgit v0.10.2


From 395174bb07c1dce58fbf2baa3a01bb69f5103c59 Mon Sep 17 00:00:00 2001
From: Loic Poulain <loic.poulain@intel.com>
Date: Mon, 22 Feb 2016 10:48:03 +0100
Subject: Bluetooth: hci_uart: Add Intel/AG6xx support

This driver implements support for iBT2.1 Bluetooth controller embedded
in the AG620 communication combo. The controller needs to be configured
with bddata and can be patched with a binary patch file (pbn).
These operations are performed in manufacturing mode.

Signed-off-by: Loic Poulain <loic.poulain@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index ec6af15..cf50fd2 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -169,6 +169,17 @@ config BT_HCIUART_QCA
 
 	  Say Y here to compile support for QCA protocol.
 
+config BT_HCIUART_AG6XX
+	bool "Intel AG6XX protocol support"
+	depends on BT_HCIUART
+	select BT_HCIUART_H4
+	select BT_INTEL
+	help
+	  The Intel/AG6XX protocol support enables Bluetooth HCI over serial
+	  port interface for Intel ibt 2.1 Bluetooth controllers.
+
+	  Say Y here to compile support for Intel AG6XX protocol.
+
 config BT_HCIBCM203X
 	tristate "HCI BCM203x USB driver"
 	depends on USB
diff --git a/drivers/bluetooth/Makefile b/drivers/bluetooth/Makefile
index 07c9cf3..9c18939 100644
--- a/drivers/bluetooth/Makefile
+++ b/drivers/bluetooth/Makefile
@@ -36,6 +36,7 @@ hci_uart-$(CONFIG_BT_HCIUART_3WIRE)	+= hci_h5.o
 hci_uart-$(CONFIG_BT_HCIUART_INTEL)	+= hci_intel.o
 hci_uart-$(CONFIG_BT_HCIUART_BCM)	+= hci_bcm.o
 hci_uart-$(CONFIG_BT_HCIUART_QCA)	+= hci_qca.o
+hci_uart-$(CONFIG_BT_HCIUART_AG6XX)	+= hci_ag6xx.o
 hci_uart-objs				:= $(hci_uart-y)
 
 ccflags-y += -D__CHECK_ENDIAN__
diff --git a/drivers/bluetooth/hci_ag6xx.c b/drivers/bluetooth/hci_ag6xx.c
new file mode 100644
index 0000000..ea65c2d
--- /dev/null
+++ b/drivers/bluetooth/hci_ag6xx.c
@@ -0,0 +1,326 @@
+/*
+ *
+ *  Bluetooth HCI UART driver for Intel/AG6xx devices
+ *
+ *  Copyright (C) 2016  Intel Corporation
+ *
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/skbuff.h>
+#include <linux/firmware.h>
+#include <linux/module.h>
+#include <linux/tty.h>
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+#include "hci_uart.h"
+#include "btintel.h"
+
+struct ag6xx_data {
+	struct sk_buff *rx_skb;
+	struct sk_buff_head txq;
+};
+
+struct pbn_entry {
+	__le32 addr;
+	__le32 plen;
+	__u8 data[0];
+} __packed;
+
+static int ag6xx_open(struct hci_uart *hu)
+{
+	struct ag6xx_data *ag6xx;
+
+	BT_DBG("hu %p", hu);
+
+	ag6xx = kzalloc(sizeof(*ag6xx), GFP_KERNEL);
+	if (!ag6xx)
+		return -ENOMEM;
+
+	skb_queue_head_init(&ag6xx->txq);
+
+	hu->priv = ag6xx;
+	return 0;
+}
+
+static int ag6xx_close(struct hci_uart *hu)
+{
+	struct ag6xx_data *ag6xx = hu->priv;
+
+	BT_DBG("hu %p", hu);
+
+	skb_queue_purge(&ag6xx->txq);
+	kfree_skb(ag6xx->rx_skb);
+	kfree(ag6xx);
+
+	hu->priv = NULL;
+	return 0;
+}
+
+static int ag6xx_flush(struct hci_uart *hu)
+{
+	struct ag6xx_data *ag6xx = hu->priv;
+
+	BT_DBG("hu %p", hu);
+
+	skb_queue_purge(&ag6xx->txq);
+	return 0;
+}
+
+static struct sk_buff *ag6xx_dequeue(struct hci_uart *hu)
+{
+	struct ag6xx_data *ag6xx = hu->priv;
+	struct sk_buff *skb;
+
+	skb = skb_dequeue(&ag6xx->txq);
+	if (!skb)
+		return skb;
+
+	/* Prepend skb with frame type */
+	memcpy(skb_push(skb, 1), &bt_cb(skb)->pkt_type, 1);
+	return skb;
+}
+
+static int ag6xx_enqueue(struct hci_uart *hu, struct sk_buff *skb)
+{
+	struct ag6xx_data *ag6xx = hu->priv;
+
+	skb_queue_tail(&ag6xx->txq, skb);
+	return 0;
+}
+
+static const struct h4_recv_pkt ag6xx_recv_pkts[] = {
+	{ H4_RECV_ACL,    .recv = hci_recv_frame   },
+	{ H4_RECV_SCO,    .recv = hci_recv_frame   },
+	{ H4_RECV_EVENT,  .recv = hci_recv_frame   },
+};
+
+static int ag6xx_recv(struct hci_uart *hu, const void *data, int count)
+{
+	struct ag6xx_data *ag6xx = hu->priv;
+
+	if (!test_bit(HCI_UART_REGISTERED, &hu->flags))
+		return -EUNATCH;
+
+	ag6xx->rx_skb = h4_recv_buf(hu->hdev, ag6xx->rx_skb, data, count,
+				    ag6xx_recv_pkts,
+				    ARRAY_SIZE(ag6xx_recv_pkts));
+	if (IS_ERR(ag6xx->rx_skb)) {
+		int err = PTR_ERR(ag6xx->rx_skb);
+		bt_dev_err(hu->hdev, "Frame reassembly failed (%d)", err);
+		ag6xx->rx_skb = NULL;
+		return err;
+	}
+
+	return count;
+}
+
+static int intel_mem_write(struct hci_dev *hdev, u32 addr, u32 plen,
+			   const void *data)
+{
+	/* Can write a maximum of 247 bytes per HCI command.
+	 * HCI cmd Header (3), Intel mem write header (6), data (247).
+	 */
+	while (plen > 0) {
+		struct sk_buff *skb;
+		u8 cmd_param[253], fragment_len = (plen > 247) ? 247 : plen;
+		__le32 leaddr = cpu_to_le32(addr);
+
+		memcpy(cmd_param, &leaddr, 4);
+		cmd_param[4] = 0;
+		cmd_param[5] = fragment_len;
+		memcpy(cmd_param + 6, data, fragment_len);
+
+		skb = __hci_cmd_sync(hdev, 0xfc8e, fragment_len + 6, cmd_param,
+				     HCI_INIT_TIMEOUT);
+		if (IS_ERR(skb))
+			return PTR_ERR(skb);
+		kfree_skb(skb);
+
+		plen -= fragment_len;
+		data += fragment_len;
+		addr += fragment_len;
+	}
+
+	return 0;
+}
+
+static int ag6xx_setup(struct hci_uart *hu)
+{
+	struct hci_dev *hdev = hu->hdev;
+	struct sk_buff *skb;
+	struct intel_version ver;
+	const struct firmware *fw;
+	const u8 *fw_ptr;
+	char fwname[64];
+	bool patched = false;
+	int err;
+
+	err = btintel_enter_mfg(hdev);
+	if (err)
+		return err;
+
+	err = btintel_read_version(hdev, &ver);
+	if (err)
+		return err;
+
+	btintel_version_info(hdev, &ver);
+
+	/* The hardware platform number has a fixed value of 0x37 and
+	 * for now only accept this single value.
+	 */
+	if (ver.hw_platform != 0x37) {
+		bt_dev_err(hdev, "Unsupported Intel hardware platform: 0x%X",
+			   ver.hw_platform);
+		return -EINVAL;
+	}
+
+	/* Only the hardware variant iBT 2.1 (AG6XX) is supported by this
+	 * firmware setup method.
+	 */
+	if (ver.hw_variant != 0x0a) {
+		bt_dev_err(hdev, "Unsupported Intel hardware variant: 0x%x",
+			   ver.hw_variant);
+		return -EINVAL;
+	}
+
+	snprintf(fwname, sizeof(fwname), "intel/ibt-hw-%x.%x.bddata",
+		 ver.hw_platform, ver.hw_variant);
+
+	err = request_firmware(&fw, fwname, &hdev->dev);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to open Intel bddata file: %s (%d)",
+			   fwname, err);
+		goto patch;
+	}
+	fw_ptr = fw->data;
+
+	bt_dev_info(hdev, "Applying bddata (%s)", fwname);
+
+	skb = __hci_cmd_sync_ev(hdev, 0xfc2f, fw->size, fw->data,
+				HCI_EV_CMD_STATUS, HCI_CMD_TIMEOUT);
+	if (IS_ERR(skb)) {
+		bt_dev_err(hdev, "Applying bddata failed (%ld)", PTR_ERR(skb));
+		release_firmware(fw);
+		return PTR_ERR(skb);
+	}
+	kfree_skb(skb);
+
+	release_firmware(fw);
+
+patch:
+	/* If there is no applied patch, fw_patch_num is always 0x00. In other
+	 * cases, current firmware is already patched. No need to patch it.
+	 */
+	if (ver.fw_patch_num) {
+		bt_dev_info(hdev, "Device is already patched. patch num: %02x",
+			    ver.fw_patch_num);
+		patched = true;
+		goto complete;
+	}
+
+	snprintf(fwname, sizeof(fwname),
+		 "intel/ibt-hw-%x.%x.%x-fw-%x.%x.%x.%x.%x.pbn",
+		 ver.hw_platform, ver.hw_variant, ver.hw_revision,
+		 ver.fw_variant,  ver.fw_revision, ver.fw_build_num,
+		 ver.fw_build_ww, ver.fw_build_yy);
+
+	err = request_firmware(&fw, fwname, &hdev->dev);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to open Intel patch file: %s(%d)",
+			   fwname, err);
+		goto complete;
+	}
+	fw_ptr = fw->data;
+
+	bt_dev_info(hdev, "Patching firmware file (%s)", fwname);
+
+	/* PBN patch file contains a list of binary patches to be applied on top
+	 * of the embedded firmware. Each patch entry header contains the target
+	 * address and patch size.
+	 *
+	 * Patch entry:
+	 * | addr(le) | patch_len(le) | patch_data |
+	 * | 4 Bytes  |    4 Bytes    |   n Bytes  |
+	 *
+	 * PBN file is terminated by a patch entry whose address is 0xffffffff.
+	 */
+	while (fw->size > fw_ptr - fw->data) {
+		struct pbn_entry *pbn = (void *)fw_ptr;
+		u32 addr, plen;
+
+		if (pbn->addr == 0xffffffff) {
+			bt_dev_info(hdev, "Patching complete");
+			patched = true;
+			break;
+		}
+
+		addr = le32_to_cpu(pbn->addr);
+		plen = le32_to_cpu(pbn->plen);
+
+		if (fw->data + fw->size <= pbn->data + plen) {
+			bt_dev_info(hdev, "Invalid patch len (%d)", plen);
+			break;
+		}
+
+		bt_dev_info(hdev, "Patching %td/%zu", (fw_ptr - fw->data),
+			    fw->size);
+
+		err = intel_mem_write(hdev, addr, plen, pbn->data);
+		if (err) {
+			bt_dev_err(hdev, "Patching failed");
+			break;
+		}
+
+		fw_ptr = pbn->data + plen;
+	}
+
+	release_firmware(fw);
+
+complete:
+	/* Exit manufacturing mode and reset */
+	err = btintel_exit_mfg(hdev, true, patched);
+
+	return err;
+}
+
+static const struct hci_uart_proto ag6xx_proto = {
+	.id		= HCI_UART_AG6XX,
+	.name		= "AG6XX",
+	.manufacturer	= 2,
+	.open		= ag6xx_open,
+	.close		= ag6xx_close,
+	.flush		= ag6xx_flush,
+	.setup		= ag6xx_setup,
+	.recv		= ag6xx_recv,
+	.enqueue	= ag6xx_enqueue,
+	.dequeue	= ag6xx_dequeue,
+};
+
+int __init ag6xx_init(void)
+{
+	return hci_uart_register_proto(&ag6xx_proto);
+}
+
+int __exit ag6xx_deinit(void)
+{
+	return hci_uart_unregister_proto(&ag6xx_proto);
+}
diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c
index 73202624..c00168a 100644
--- a/drivers/bluetooth/hci_ldisc.c
+++ b/drivers/bluetooth/hci_ldisc.c
@@ -804,6 +804,9 @@ static int __init hci_uart_init(void)
 #ifdef CONFIG_BT_HCIUART_QCA
 	qca_init();
 #endif
+#ifdef CONFIG_BT_HCIUART_AG6XX
+	ag6xx_init();
+#endif
 
 	return 0;
 }
@@ -836,6 +839,9 @@ static void __exit hci_uart_exit(void)
 #ifdef CONFIG_BT_HCIUART_QCA
 	qca_deinit();
 #endif
+#ifdef CONFIG_BT_HCIUART_AG6XX
+	ag6xx_deinit();
+#endif
 
 	/* Release tty registration of line discipline */
 	err = tty_unregister_ldisc(N_HCI);
diff --git a/drivers/bluetooth/hci_uart.h b/drivers/bluetooth/hci_uart.h
index 82c92f1..4814ff0 100644
--- a/drivers/bluetooth/hci_uart.h
+++ b/drivers/bluetooth/hci_uart.h
@@ -35,7 +35,7 @@
 #define HCIUARTGETFLAGS		_IOR('U', 204, int)
 
 /* UART protocols */
-#define HCI_UART_MAX_PROTO	9
+#define HCI_UART_MAX_PROTO	10
 
 #define HCI_UART_H4	0
 #define HCI_UART_BCSP	1
@@ -46,6 +46,7 @@
 #define HCI_UART_INTEL	6
 #define HCI_UART_BCM	7
 #define HCI_UART_QCA	8
+#define HCI_UART_AG6XX	9
 
 #define HCI_UART_RAW_DEVICE	0
 #define HCI_UART_RESET_ON_INIT	1
@@ -182,3 +183,8 @@ int bcm_deinit(void);
 int qca_init(void);
 int qca_deinit(void);
 #endif
+
+#ifdef CONFIG_BT_HCIUART_AG6XX
+int ag6xx_init(void);
+int ag6xx_deinit(void);
+#endif
-- 
cgit v0.10.2


From e5fbfc1c2d7657eafed645727cb5d74731f6d68c Mon Sep 17 00:00:00 2001
From: Craig Gallek <kraig@google.com>
Date: Mon, 22 Feb 2016 10:45:29 -0500
Subject: soreuseport: fix merge conflict in tcp bind

One of the validation checks for the new array-based TCP SO_REUSEPORT
validation was unintentionally dropped in ea8add2b1903.  This adds it back.

Lack of this check allows the user to allocate multiple sock_reuseport
structures (leaking all but the first).

Fixes: ea8add2b1903 ("tcp/dccp: better use of ephemeral ports in bind()")
Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index d768230..bc5196e 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -202,6 +202,7 @@ tb_found:
 
 		if (((tb->fastreuse > 0 && reuse) ||
 		     (tb->fastreuseport > 0 &&
+		      !rcu_access_pointer(sk->sk_reuseport_cb) &&
 		      sk->sk_reuseport && uid_eq(tb->fastuid, uid))) &&
 		    smallest_size == -1)
 			goto success;
-- 
cgit v0.10.2


From ada68c31ba9c02d7aabdd87db979fe670b499d54 Mon Sep 17 00:00:00 2001
From: Achiad Shochat <achiad@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:23 +0200
Subject: net/mlx5: Introduce a new header file for physical port functions

All the device physical port access functions are implemented in the
port.c file.
We just extract the exposure of these functions from driver.h into a
dedicated header file called port.h.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 03c418c..e1cea44 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -42,6 +42,7 @@
 #include <rdma/ib_user_verbs.h>
 #include <rdma/ib_addr.h>
 #include <rdma/ib_cache.h>
+#include <linux/mlx5/port.h>
 #include <linux/mlx5/vport.h>
 #include <rdma/ib_smi.h>
 #include <rdma/ib_umem.h>
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index aac071a..15f6cdb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -38,6 +38,7 @@
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/qp.h>
 #include <linux/mlx5/cq.h>
+#include <linux/mlx5/port.h>
 #include <linux/mlx5/vport.h>
 #include <linux/mlx5/transobj.h>
 #include "wq.h"
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index a87e773..1e86321 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -32,6 +32,7 @@
 
 #include <linux/module.h>
 #include <linux/mlx5/driver.h>
+#include <linux/mlx5/port.h>
 #include <linux/mlx5/cmd.h>
 #include "mlx5_core.h"
 
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 1e3006d..02adc67 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -794,37 +794,6 @@ int mlx5_core_access_reg(struct mlx5_core_dev *dev, void *data_in,
 			 int size_in, void *data_out, int size_out,
 			 u16 reg_num, int arg, int write);
 
-int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps);
-int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys,
-			 int ptys_size, int proto_mask, u8 local_port);
-int mlx5_query_port_proto_cap(struct mlx5_core_dev *dev,
-			      u32 *proto_cap, int proto_mask);
-int mlx5_query_port_proto_admin(struct mlx5_core_dev *dev,
-				u32 *proto_admin, int proto_mask);
-int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
-				    u8 *link_width_oper, u8 local_port);
-int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev,
-			       u8 *proto_oper, int proto_mask,
-			       u8 local_port);
-int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin,
-			int proto_mask);
-int mlx5_set_port_admin_status(struct mlx5_core_dev *dev,
-			       enum mlx5_port_status status);
-int mlx5_query_port_admin_status(struct mlx5_core_dev *dev,
-				 enum mlx5_port_status *status);
-
-int mlx5_set_port_mtu(struct mlx5_core_dev *dev, int mtu, u8 port);
-void mlx5_query_port_max_mtu(struct mlx5_core_dev *dev, int *max_mtu, u8 port);
-void mlx5_query_port_oper_mtu(struct mlx5_core_dev *dev, int *oper_mtu,
-			      u8 port);
-
-int mlx5_query_port_vl_hw_cap(struct mlx5_core_dev *dev,
-			      u8 *vl_hw_cap, u8 local_port);
-
-int mlx5_set_port_pause(struct mlx5_core_dev *dev, u32 rx_pause, u32 tx_pause);
-int mlx5_query_port_pause(struct mlx5_core_dev *dev,
-			  u32 *rx_pause, u32 *tx_pause);
-
 int mlx5_debug_eq_add(struct mlx5_core_dev *dev, struct mlx5_eq *eq);
 void mlx5_debug_eq_remove(struct mlx5_core_dev *dev, struct mlx5_eq *eq);
 int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq,
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
new file mode 100644
index 0000000..7accd4a
--- /dev/null
+++ b/include/linux/mlx5/port.h
@@ -0,0 +1,69 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __MLX5_PORT_H__
+#define __MLX5_PORT_H__
+
+#include <linux/mlx5/driver.h>
+
+int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps);
+int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys,
+			 int ptys_size, int proto_mask, u8 local_port);
+int mlx5_query_port_proto_cap(struct mlx5_core_dev *dev,
+			      u32 *proto_cap, int proto_mask);
+int mlx5_query_port_proto_admin(struct mlx5_core_dev *dev,
+				u32 *proto_admin, int proto_mask);
+int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
+				    u8 *link_width_oper, u8 local_port);
+int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev,
+			       u8 *proto_oper, int proto_mask,
+			       u8 local_port);
+int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin,
+			int proto_mask);
+int mlx5_set_port_admin_status(struct mlx5_core_dev *dev,
+			       enum mlx5_port_status status);
+int mlx5_query_port_admin_status(struct mlx5_core_dev *dev,
+				 enum mlx5_port_status *status);
+
+int mlx5_set_port_mtu(struct mlx5_core_dev *dev, int mtu, u8 port);
+void mlx5_query_port_max_mtu(struct mlx5_core_dev *dev, int *max_mtu, u8 port);
+void mlx5_query_port_oper_mtu(struct mlx5_core_dev *dev, int *oper_mtu,
+			      u8 port);
+
+int mlx5_query_port_vl_hw_cap(struct mlx5_core_dev *dev,
+			      u8 *vl_hw_cap, u8 local_port);
+
+int mlx5_set_port_pause(struct mlx5_core_dev *dev, u32 rx_pause, u32 tx_pause);
+int mlx5_query_port_pause(struct mlx5_core_dev *dev,
+			  u32 *rx_pause, u32 *tx_pause);
+
+#endif /* __MLX5_PORT_H__ */
-- 
cgit v0.10.2


From ad909eb064219a64fd10e9c7d9f39a3042760025 Mon Sep 17 00:00:00 2001
From: Achiad Shochat <achiad@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:24 +0200
Subject: net/mlx5: Introduce physical port PFC access functions

Add access functions to set and query a physical port PFC
(Priority Flow Control) parameters.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index 1e86321..dae7050 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -364,3 +364,44 @@ int mlx5_query_port_pause(struct mlx5_core_dev *dev,
 	return 0;
 }
 EXPORT_SYMBOL_GPL(mlx5_query_port_pause);
+
+int mlx5_set_port_pfc(struct mlx5_core_dev *dev, u8 pfc_en_tx, u8 pfc_en_rx)
+{
+	u32 in[MLX5_ST_SZ_DW(pfcc_reg)];
+	u32 out[MLX5_ST_SZ_DW(pfcc_reg)];
+
+	memset(in, 0, sizeof(in));
+	MLX5_SET(pfcc_reg, in, local_port, 1);
+	MLX5_SET(pfcc_reg, in, pfctx, pfc_en_tx);
+	MLX5_SET(pfcc_reg, in, pfcrx, pfc_en_rx);
+	MLX5_SET_TO_ONES(pfcc_reg, in, prio_mask_tx);
+	MLX5_SET_TO_ONES(pfcc_reg, in, prio_mask_rx);
+
+	return mlx5_core_access_reg(dev, in, sizeof(in), out,
+				    sizeof(out), MLX5_REG_PFCC, 0, 1);
+}
+EXPORT_SYMBOL_GPL(mlx5_set_port_pfc);
+
+int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx)
+{
+	u32 in[MLX5_ST_SZ_DW(pfcc_reg)];
+	u32 out[MLX5_ST_SZ_DW(pfcc_reg)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+	MLX5_SET(pfcc_reg, in, local_port, 1);
+
+	err = mlx5_core_access_reg(dev, in, sizeof(in), out,
+				   sizeof(out), MLX5_REG_PFCC, 0, 0);
+	if (err)
+		return err;
+
+	if (pfc_en_tx)
+		*pfc_en_tx = MLX5_GET(pfcc_reg, out, pfctx);
+
+	if (pfc_en_rx)
+		*pfc_en_rx = MLX5_GET(pfcc_reg, out, pfcrx);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_port_pfc);
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 7accd4a..4b3644c 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -66,4 +66,8 @@ int mlx5_set_port_pause(struct mlx5_core_dev *dev, u32 rx_pause, u32 tx_pause);
 int mlx5_query_port_pause(struct mlx5_core_dev *dev,
 			  u32 *rx_pause, u32 *tx_pause);
 
+int mlx5_set_port_pfc(struct mlx5_core_dev *dev, u8 pfc_en_tx, u8 pfc_en_rx);
+int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx,
+			u8 *pfc_en_rx);
+
 #endif /* __MLX5_PORT_H__ */
-- 
cgit v0.10.2


From 4f3961eeafe0aca8f6b0933899ef0d91f561352d Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:25 +0200
Subject: net/mlx5: Introduce physical port TC/prio access functions

Add access functions to set and query a physical port TC groups
and prio parameters.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index dae7050..569100d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -405,3 +405,79 @@ int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx)
 	return 0;
 }
 EXPORT_SYMBOL_GPL(mlx5_query_port_pfc);
+
+int mlx5_max_tc(struct mlx5_core_dev *mdev)
+{
+	u8 num_tc = MLX5_CAP_GEN(mdev, max_tc) ? : 8;
+
+	return num_tc - 1;
+}
+
+int mlx5_set_port_prio_tc(struct mlx5_core_dev *mdev, u8 *prio_tc)
+{
+	u32 in[MLX5_ST_SZ_DW(qtct_reg)];
+	u32 out[MLX5_ST_SZ_DW(qtct_reg)];
+	int err;
+	int i;
+
+	memset(in, 0, sizeof(in));
+	for (i = 0; i < 8; i++) {
+		if (prio_tc[i] > mlx5_max_tc(mdev))
+			return -EINVAL;
+
+		MLX5_SET(qtct_reg, in, prio, i);
+		MLX5_SET(qtct_reg, in, tclass, prio_tc[i]);
+
+		err = mlx5_core_access_reg(mdev, in, sizeof(in), out,
+					   sizeof(out), MLX5_REG_QTCT, 0, 1);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx5_set_port_prio_tc);
+
+static int mlx5_set_port_qetcr_reg(struct mlx5_core_dev *mdev, u32 *in,
+				   int inlen)
+{
+	u32 out[MLX5_ST_SZ_DW(qtct_reg)];
+
+	if (!MLX5_CAP_GEN(mdev, ets))
+		return -ENOTSUPP;
+
+	return mlx5_core_access_reg(mdev, in, inlen, out, sizeof(out),
+				    MLX5_REG_QETCR, 0, 1);
+}
+
+int mlx5_set_port_tc_group(struct mlx5_core_dev *mdev, u8 *tc_group)
+{
+	u32 in[MLX5_ST_SZ_DW(qetc_reg)];
+	int i;
+
+	memset(in, 0, sizeof(in));
+
+	for (i = 0; i <= mlx5_max_tc(mdev); i++) {
+		MLX5_SET(qetc_reg, in, tc_configuration[i].g, 1);
+		MLX5_SET(qetc_reg, in, tc_configuration[i].group, tc_group[i]);
+	}
+
+	return mlx5_set_port_qetcr_reg(mdev, in, sizeof(in));
+}
+EXPORT_SYMBOL_GPL(mlx5_set_port_tc_group);
+
+int mlx5_set_port_tc_bw_alloc(struct mlx5_core_dev *mdev, u8 *tc_bw)
+{
+	u32 in[MLX5_ST_SZ_DW(qetc_reg)];
+	int i;
+
+	memset(in, 0, sizeof(in));
+
+	for (i = 0; i <= mlx5_max_tc(mdev); i++) {
+		MLX5_SET(qetc_reg, in, tc_configuration[i].b, 1);
+		MLX5_SET(qetc_reg, in, tc_configuration[i].bw_allocation, tc_bw[i]);
+	}
+
+	return mlx5_set_port_qetcr_reg(mdev, in, sizeof(in));
+}
+EXPORT_SYMBOL_GPL(mlx5_set_port_tc_bw_alloc);
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 02adc67..a815da9 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -99,6 +99,8 @@ enum {
 };
 
 enum {
+	MLX5_REG_QETCR		 = 0x4005,
+	MLX5_REG_QTCT		 = 0x400a,
 	MLX5_REG_PCAP		 = 0x5001,
 	MLX5_REG_PMTU		 = 0x5003,
 	MLX5_REG_PTYS		 = 0x5004,
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 51f1e54..ec957e0 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -729,7 +729,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 
 	u8         reserved_at_1bf[0x3];
 	u8         log_max_msg[0x5];
-	u8         reserved_at_1c7[0x18];
+	u8         reserved_at_1c7[0x4];
+	u8         max_tc[0x4];
+	u8         reserved_at_1cf[0x10];
 
 	u8         stat_rate_support[0x10];
 	u8         reserved_at_1ef[0xc];
@@ -7061,4 +7063,49 @@ struct mlx5_ifc_modify_flow_table_in_bits {
 	u8         reserved_at_100[0x100];
 };
 
+struct mlx5_ifc_ets_tcn_config_reg_bits {
+	u8         g[0x1];
+	u8         b[0x1];
+	u8         r[0x1];
+	u8         reserved_at_3[0x9];
+	u8         group[0x4];
+	u8         reserved_at_10[0x9];
+	u8         bw_allocation[0x7];
+
+	u8         reserved_at_20[0xc];
+	u8         max_bw_units[0x4];
+	u8         reserved_at_30[0x8];
+	u8         max_bw_value[0x8];
+};
+
+struct mlx5_ifc_ets_global_config_reg_bits {
+	u8         reserved_at_0[0x2];
+	u8         r[0x1];
+	u8         reserved_at_3[0x1d];
+
+	u8         reserved_at_20[0xc];
+	u8         max_bw_units[0x4];
+	u8         reserved_at_30[0x8];
+	u8         max_bw_value[0x8];
+};
+
+struct mlx5_ifc_qetc_reg_bits {
+	u8                                         reserved_at_0[0x8];
+	u8                                         port_number[0x8];
+	u8                                         reserved_at_10[0x30];
+
+	struct mlx5_ifc_ets_tcn_config_reg_bits    tc_configuration[0x8];
+	struct mlx5_ifc_ets_global_config_reg_bits global_configuration;
+};
+
+struct mlx5_ifc_qtct_reg_bits {
+	u8         reserved_at_0[0x8];
+	u8         port_number[0x8];
+	u8         reserved_at_10[0xd];
+	u8         prio[0x3];
+
+	u8         reserved_at_20[0x1d];
+	u8         tclass[0x3];
+};
+
 #endif /* MLX5_IFC_H */
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 4b3644c..0c67e69 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -70,4 +70,10 @@ int mlx5_set_port_pfc(struct mlx5_core_dev *dev, u8 pfc_en_tx, u8 pfc_en_rx);
 int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx,
 			u8 *pfc_en_rx);
 
+int mlx5_max_tc(struct mlx5_core_dev *mdev);
+
+int mlx5_set_port_prio_tc(struct mlx5_core_dev *mdev, u8 *prio_tc);
+int mlx5_set_port_tc_group(struct mlx5_core_dev *mdev, u8 *tc_group);
+int mlx5_set_port_tc_bw_alloc(struct mlx5_core_dev *mdev, u8 *tc_bw);
+
 #endif /* __MLX5_PORT_H__ */
-- 
cgit v0.10.2


From 08fb1dacdd763431436d648fd9dc4246e3d5517e Mon Sep 17 00:00:00 2001
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:26 +0200
Subject: net/mlx5e: Support DCBNL IEEE ETS

Support the ndo_setup_tc callback and the needed methods
for multi TC/UP support, and removed the default_vlan_prio
from mlx5e_priv which is always 0, it was replaced with
hardcoded "0" in the new select queue method.

For that we now create MAX_NUM_TC num of TISs (one per prio)
on netdevice creation instead of priv->params.num_tc which
was always 1.

So far each channel had a single TXQ, Now each channel has a
TXQ per TC (Traffic Class).

Added en_dcbnl.c which implements the set/get DCBNL IEEE ETS,
set/get dcbx and registers the mlx5e dcbnl ops.

We still use the kernel's default TXQ selection method to select the
channel to transmit through but now we use our own method to select
the TXQ inside the channel based on VLAN priority.

In mlx5, as opposed to mlx4, tc group N gets lower priority than
tc group N+1.

CC: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
index c503ea0..1cf722e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -19,3 +19,15 @@ config MLX5_CORE_EN
 	  Ethernet support in Mellanox Technologies ConnectX-4 NIC.
 	  Ethernet and Infiniband support in ConnectX-4 are currently mutually
 	  exclusive.
+
+config MLX5_CORE_EN_DCB
+	bool "Data Center Bridging (DCB) Support"
+	default y
+	depends on MLX5_CORE_EN && DCB
+	---help---
+	  Say Y here if you want to use Data Center Bridging (DCB) in the
+	  driver.
+	  If set to N, will not be able to configure QoS and ratelimit attributes.
+	  This flag is depended on the kernel's DCB support.
+
+	  If unsure, set to Y
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 01c0256..1a82e23 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -3,6 +3,9 @@ obj-$(CONFIG_MLX5_CORE)		+= mlx5_core.o
 mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 		health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o   \
 		mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o
+
 mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \
 		en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \
 		en_txrx.o en_clock.o
+
+mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) +=  en_dcbnl.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 15f6cdb..dfbc4e5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -70,6 +70,11 @@
 
 #define MLX5E_NUM_MAIN_GROUPS 9
 
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+#define MLX5E_MAX_BW_ALLOC 100 /* Max percentage of BW allocation */
+#define MLX5E_MIN_BW_ALLOC 1   /* Min percentage of BW allocation */
+#endif
+
 static const char vport_strings[][ETH_GSTRING_LEN] = {
 	/* vport statistics */
 	"rx_packets",
@@ -273,7 +278,6 @@ struct mlx5e_params {
 	u8  log_sq_size;
 	u8  log_rq_size;
 	u16 num_channels;
-	u8  default_vlan_prio;
 	u8  num_tc;
 	u16 rx_cq_moderation_usec;
 	u16 rx_cq_moderation_pkts;
@@ -286,6 +290,9 @@ struct mlx5e_params {
 	u8  rss_hfunc;
 	u8  toeplitz_hash_key[40];
 	u32 indirection_rqt[MLX5E_INDIR_RQT_SIZE];
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+	struct ieee_ets ets;
+#endif
 };
 
 struct mlx5e_tstamp {
@@ -506,7 +513,6 @@ struct mlx5e_flow_tables {
 
 struct mlx5e_priv {
 	/* priv data path fields - start */
-	int                        default_vlan_prio;
 	struct mlx5e_sq            **txq_to_sq_map;
 	int channeltc_to_txq_map[MLX5E_MAX_NUM_CHANNELS][MLX5E_MAX_NUM_TC];
 	/* priv data path fields - end */
@@ -666,4 +672,9 @@ static inline int mlx5e_get_max_num_channels(struct mlx5_core_dev *mdev)
 }
 
 extern const struct ethtool_ops mlx5e_ethtool_ops;
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+extern const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops;
+int mlx5e_dcbnl_ieee_setets_core(struct mlx5e_priv *priv, struct ieee_ets *ets);
+#endif
+
 u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
new file mode 100644
index 0000000..4f097da
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
@@ -0,0 +1,190 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/device.h>
+#include <linux/netdevice.h>
+#include "en.h"
+
+#define MLX5E_MAX_PRIORITY 8
+
+static int mlx5e_dcbnl_ieee_getets(struct net_device *netdev,
+				   struct ieee_ets *ets)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+
+	if (!MLX5_CAP_GEN(priv->mdev, ets))
+		return -ENOTSUPP;
+
+	memcpy(ets, &priv->params.ets, sizeof(*ets));
+	return 0;
+}
+
+enum {
+	MLX5E_VENDOR_TC_GROUP_NUM = 7,
+	MLX5E_ETS_TC_GROUP_NUM    = 0,
+};
+
+static void mlx5e_build_tc_group(struct ieee_ets *ets, u8 *tc_group, int max_tc)
+{
+	bool any_tc_mapped_to_ets = false;
+	int strict_group;
+	int i;
+
+	for (i = 0; i <= max_tc; i++)
+		if (ets->tc_tsa[i] == IEEE_8021QAZ_TSA_ETS)
+			any_tc_mapped_to_ets = true;
+
+	strict_group = any_tc_mapped_to_ets ? 1 : 0;
+
+	for (i = 0; i <= max_tc; i++) {
+		switch (ets->tc_tsa[i]) {
+		case IEEE_8021QAZ_TSA_VENDOR:
+			tc_group[i] = MLX5E_VENDOR_TC_GROUP_NUM;
+			break;
+		case IEEE_8021QAZ_TSA_STRICT:
+			tc_group[i] = strict_group++;
+			break;
+		case IEEE_8021QAZ_TSA_ETS:
+			tc_group[i] = MLX5E_ETS_TC_GROUP_NUM;
+			break;
+		}
+	}
+}
+
+static void mlx5e_build_tc_tx_bw(struct ieee_ets *ets, u8 *tc_tx_bw,
+				 u8 *tc_group, int max_tc)
+{
+	int i;
+
+	for (i = 0; i <= max_tc; i++) {
+		switch (ets->tc_tsa[i]) {
+		case IEEE_8021QAZ_TSA_VENDOR:
+			tc_tx_bw[i] = MLX5E_MAX_BW_ALLOC;
+			break;
+		case IEEE_8021QAZ_TSA_STRICT:
+			tc_tx_bw[i] = MLX5E_MAX_BW_ALLOC;
+			break;
+		case IEEE_8021QAZ_TSA_ETS:
+			tc_tx_bw[i] = ets->tc_tx_bw[i] ?: MLX5E_MIN_BW_ALLOC;
+			break;
+		}
+	}
+}
+
+int mlx5e_dcbnl_ieee_setets_core(struct mlx5e_priv *priv, struct ieee_ets *ets)
+{
+	struct mlx5_core_dev *mdev = priv->mdev;
+	u8 tc_tx_bw[IEEE_8021QAZ_MAX_TCS];
+	u8 tc_group[IEEE_8021QAZ_MAX_TCS];
+	int max_tc = mlx5_max_tc(mdev);
+	int err;
+
+	if (!MLX5_CAP_GEN(mdev, ets))
+		return -ENOTSUPP;
+
+	mlx5e_build_tc_group(ets, tc_group, max_tc);
+	mlx5e_build_tc_tx_bw(ets, tc_tx_bw, tc_group, max_tc);
+
+	err = mlx5_set_port_prio_tc(mdev, ets->prio_tc);
+	if (err)
+		return err;
+
+	err = mlx5_set_port_tc_group(mdev, tc_group);
+	if (err)
+		return err;
+
+	return mlx5_set_port_tc_bw_alloc(mdev, tc_tx_bw);
+}
+
+static int mlx5e_dbcnl_validate_ets(struct ieee_ets *ets)
+{
+	int bw_sum = 0;
+	int i;
+
+	/* Validate Priority */
+	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
+		if (ets->prio_tc[i] >= MLX5E_MAX_PRIORITY)
+			return -EINVAL;
+	}
+
+	/* Validate Bandwidth Sum */
+	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
+		if (ets->tc_tsa[i] == IEEE_8021QAZ_TSA_ETS)
+			bw_sum += ets->tc_tx_bw[i];
+	}
+
+	if (bw_sum != 0 && bw_sum != 100)
+		return -EINVAL;
+	return 0;
+}
+
+static int mlx5e_dcbnl_ieee_setets(struct net_device *netdev,
+				   struct ieee_ets *ets)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+	int err;
+
+	err = mlx5e_dbcnl_validate_ets(ets);
+	if (err)
+		return err;
+
+	err = mlx5e_dcbnl_ieee_setets_core(priv, ets);
+	if (err)
+		return err;
+
+	memcpy(&priv->params.ets, ets, sizeof(*ets));
+	priv->params.ets.ets_cap = mlx5_max_tc(priv->mdev) + 1;
+
+	return 0;
+}
+
+static u8 mlx5e_dcbnl_getdcbx(struct net_device *dev)
+{
+	return DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE;
+}
+
+static u8 mlx5e_dcbnl_setdcbx(struct net_device *dev, u8 mode)
+{
+	if ((mode & DCB_CAP_DCBX_LLD_MANAGED) ||
+	    (mode & DCB_CAP_DCBX_VER_CEE) ||
+	    !(mode & DCB_CAP_DCBX_VER_IEEE) ||
+	    !(mode & DCB_CAP_DCBX_HOST))
+		return 1;
+
+	return 0;
+}
+
+const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops = {
+	.ieee_getets	= mlx5e_dcbnl_ieee_getets,
+	.ieee_setets	= mlx5e_dcbnl_ieee_setets,
+	.getdcbx	= mlx5e_dcbnl_getdcbx,
+	.setdcbx	= mlx5e_dcbnl_setdcbx,
+};
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index d4e1c30..704d75c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1400,6 +1400,24 @@ static int mlx5e_set_dev_port_mtu(struct net_device *netdev)
 	return 0;
 }
 
+static void mlx5e_netdev_set_tcs(struct net_device *netdev)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+	int nch = priv->params.num_channels;
+	int ntc = priv->params.num_tc;
+	int tc;
+
+	netdev_reset_tc(netdev);
+
+	if (ntc == 1)
+		return;
+
+	netdev_set_num_tc(netdev, ntc);
+
+	for (tc = 0; tc < ntc; tc++)
+		netdev_set_tc_queue(netdev, tc, nch, tc * nch);
+}
+
 int mlx5e_open_locked(struct net_device *netdev)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
@@ -1408,6 +1426,8 @@ int mlx5e_open_locked(struct net_device *netdev)
 
 	set_bit(MLX5E_STATE_OPENED, &priv->state);
 
+	mlx5e_netdev_set_tcs(netdev);
+
 	num_txqs = priv->params.num_channels * priv->params.num_tc;
 	netif_set_real_num_tx_queues(netdev, num_txqs);
 	netif_set_real_num_rx_queues(netdev, priv->params.num_channels);
@@ -1602,7 +1622,7 @@ static int mlx5e_create_tis(struct mlx5e_priv *priv, int tc)
 
 	memset(in, 0, sizeof(in));
 
-	MLX5_SET(tisc, tisc, prio,  tc);
+	MLX5_SET(tisc, tisc, prio, tc << 1);
 	MLX5_SET(tisc, tisc, transport_domain, priv->tdn);
 
 	return mlx5_core_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]);
@@ -1618,7 +1638,7 @@ static int mlx5e_create_tises(struct mlx5e_priv *priv)
 	int err;
 	int tc;
 
-	for (tc = 0; tc < priv->params.num_tc; tc++) {
+	for (tc = 0; tc < MLX5E_MAX_NUM_TC; tc++) {
 		err = mlx5e_create_tis(priv, tc);
 		if (err)
 			goto err_close_tises;
@@ -1637,7 +1657,7 @@ static void mlx5e_destroy_tises(struct mlx5e_priv *priv)
 {
 	int tc;
 
-	for (tc = 0; tc < priv->params.num_tc; tc++)
+	for (tc = 0; tc < MLX5E_MAX_NUM_TC; tc++)
 		mlx5e_destroy_tis(priv, tc);
 }
 
@@ -1824,6 +1844,40 @@ static void mlx5e_destroy_tirs(struct mlx5e_priv *priv)
 		mlx5e_destroy_tir(priv, i);
 }
 
+static int mlx5e_setup_tc(struct net_device *netdev, u8 tc)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+	bool was_opened;
+	int err = 0;
+
+	if (tc && tc != MLX5E_MAX_NUM_TC)
+		return -EINVAL;
+
+	mutex_lock(&priv->state_lock);
+
+	was_opened = test_bit(MLX5E_STATE_OPENED, &priv->state);
+	if (was_opened)
+		mlx5e_close_locked(priv->netdev);
+
+	priv->params.num_tc = tc ? tc : 1;
+
+	if (was_opened)
+		err = mlx5e_open_locked(priv->netdev);
+
+	mutex_unlock(&priv->state_lock);
+
+	return err;
+}
+
+static int mlx5e_ndo_setup_tc(struct net_device *dev, u32 handle,
+			      __be16 proto, struct tc_to_netdev *tc)
+{
+	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+		return -EINVAL;
+
+	return mlx5e_setup_tc(dev, tc->tc);
+}
+
 static struct rtnl_link_stats64 *
 mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
 {
@@ -2028,6 +2082,8 @@ static const struct net_device_ops mlx5e_netdev_ops_basic = {
 	.ndo_open                = mlx5e_open,
 	.ndo_stop                = mlx5e_close,
 	.ndo_start_xmit          = mlx5e_xmit,
+	.ndo_setup_tc            = mlx5e_ndo_setup_tc,
+	.ndo_select_queue        = mlx5e_select_queue,
 	.ndo_get_stats64         = mlx5e_get_stats,
 	.ndo_set_rx_mode         = mlx5e_set_rx_mode,
 	.ndo_set_mac_address     = mlx5e_set_mac,
@@ -2042,6 +2098,8 @@ static const struct net_device_ops mlx5e_netdev_ops_sriov = {
 	.ndo_open                = mlx5e_open,
 	.ndo_stop                = mlx5e_close,
 	.ndo_start_xmit          = mlx5e_xmit,
+	.ndo_setup_tc            = mlx5e_ndo_setup_tc,
+	.ndo_select_queue        = mlx5e_select_queue,
 	.ndo_get_stats64         = mlx5e_get_stats,
 	.ndo_set_rx_mode         = mlx5e_set_rx_mode,
 	.ndo_set_mac_address     = mlx5e_set_mac,
@@ -2089,6 +2147,24 @@ u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev)
 	       2 /*sizeof(mlx5e_tx_wqe.inline_hdr_start)*/;
 }
 
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+static void mlx5e_ets_init(struct mlx5e_priv *priv)
+{
+	int i;
+
+	priv->params.ets.ets_cap = mlx5_max_tc(priv->mdev) + 1;
+	for (i = 0; i < priv->params.ets.ets_cap; i++) {
+		priv->params.ets.tc_tx_bw[i] = MLX5E_MAX_BW_ALLOC;
+		priv->params.ets.tc_tsa[i] = IEEE_8021QAZ_TSA_VENDOR;
+		priv->params.ets.prio_tc[i] = i;
+	}
+
+	/* tclass[prio=0]=1, tclass[prio=1]=0, tclass[prio=i]=i (for i>1) */
+	priv->params.ets.prio_tc[0] = 1;
+	priv->params.ets.prio_tc[1] = 0;
+}
+#endif
+
 static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 				    struct net_device *netdev,
 				    int num_channels)
@@ -2112,7 +2188,6 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 	priv->params.min_rx_wqes           =
 		MLX5E_PARAMS_DEFAULT_MIN_RX_WQES;
 	priv->params.num_tc                = 1;
-	priv->params.default_vlan_prio     = 0;
 	priv->params.rss_hfunc             = ETH_RSS_HASH_XOR;
 
 	netdev_rss_key_fill(priv->params.toeplitz_hash_key,
@@ -2127,7 +2202,10 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 	priv->mdev                         = mdev;
 	priv->netdev                       = netdev;
 	priv->params.num_channels          = num_channels;
-	priv->default_vlan_prio            = priv->params.default_vlan_prio;
+
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+	mlx5e_ets_init(priv);
+#endif
 
 	spin_lock_init(&priv->async_events_spinlock);
 	mutex_init(&priv->state_lock);
@@ -2156,10 +2234,14 @@ static void mlx5e_build_netdev(struct net_device *netdev)
 
 	SET_NETDEV_DEV(netdev, &mdev->pdev->dev);
 
-	if (MLX5_CAP_GEN(mdev, vport_group_manager))
+	if (MLX5_CAP_GEN(mdev, vport_group_manager)) {
 		netdev->netdev_ops = &mlx5e_netdev_ops_sriov;
-	else
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+		netdev->dcbnl_ops = &mlx5e_dcbnl_ops;
+#endif
+	} else {
 		netdev->netdev_ops = &mlx5e_netdev_ops_basic;
+	}
 
 	netdev->watchdog_timeo    = 15 * HZ;
 
@@ -2228,7 +2310,9 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 	if (mlx5e_check_required_hca_cap(mdev))
 		return NULL;
 
-	netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), nch, nch);
+	netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv),
+				    nch * MLX5E_MAX_NUM_TC,
+				    nch);
 	if (!netdev) {
 		mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n");
 		return NULL;
@@ -2303,6 +2387,10 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 
 	mlx5e_init_eth_addr(priv);
 
+#ifdef CONFIG_MLX5_CORE_EN_DCB
+	mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets);
+#endif
+
 	err = register_netdev(netdev);
 	if (err) {
 		mlx5_core_err(mdev, "register_netdev failed, %d\n", err);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 2c3fba0..00d855a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -109,12 +109,10 @@ u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb,
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
 	int channel_ix = fallback(dev, skb);
-	int up = skb_vlan_tag_present(skb)        ?
-		 skb->vlan_tci >> VLAN_PRIO_SHIFT :
-		 priv->default_vlan_prio;
-	int tc = netdev_get_prio_tc_map(dev, up);
+	int up = (netdev_get_num_tc(dev) && skb_vlan_tag_present(skb)) ?
+		 skb->vlan_tci >> VLAN_PRIO_SHIFT : 0;
 
-	return priv->channeltc_to_txq_map[channel_ix][tc];
+	return priv->channeltc_to_txq_map[channel_ix][up];
 }
 
 static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq,
-- 
cgit v0.10.2


From ef9184335e4da720b374a66fe4f80a32982519ca Mon Sep 17 00:00:00 2001
From: Achiad Shochat <achiad@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:27 +0200
Subject: net/mlx5e: Support DCBNL IEEE PFC

Implement the set/get DCBNL IEEE PFC callbacks.

Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
index 4f097da..39d8069 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
@@ -166,6 +166,43 @@ static int mlx5e_dcbnl_ieee_setets(struct net_device *netdev,
 	return 0;
 }
 
+static int mlx5e_dcbnl_ieee_getpfc(struct net_device *dev,
+				   struct ieee_pfc *pfc)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+
+	pfc->pfc_cap = mlx5_max_tc(mdev) + 1;
+
+	return mlx5_query_port_pfc(mdev, &pfc->pfc_en, NULL);
+}
+
+static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev,
+				   struct ieee_pfc *pfc)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+	enum mlx5_port_status ps;
+	u8 curr_pfc_en;
+	int ret;
+
+	mlx5_query_port_pfc(mdev, &curr_pfc_en, NULL);
+
+	if (pfc->pfc_en == curr_pfc_en)
+		return 0;
+
+	mlx5_query_port_admin_status(mdev, &ps);
+	if (ps == MLX5_PORT_UP)
+		mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN);
+
+	ret = mlx5_set_port_pfc(mdev, pfc->pfc_en, pfc->pfc_en);
+
+	if (ps == MLX5_PORT_UP)
+		mlx5_set_port_admin_status(mdev, MLX5_PORT_UP);
+
+	return ret;
+}
+
 static u8 mlx5e_dcbnl_getdcbx(struct net_device *dev)
 {
 	return DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE;
@@ -185,6 +222,8 @@ static u8 mlx5e_dcbnl_setdcbx(struct net_device *dev, u8 mode)
 const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops = {
 	.ieee_getets	= mlx5e_dcbnl_ieee_getets,
 	.ieee_setets	= mlx5e_dcbnl_ieee_setets,
+	.ieee_getpfc	= mlx5e_dcbnl_ieee_getpfc,
+	.ieee_setpfc	= mlx5e_dcbnl_ieee_setpfc,
 	.getdcbx	= mlx5e_dcbnl_getdcbx,
 	.setdcbx	= mlx5e_dcbnl_setdcbx,
 };
-- 
cgit v0.10.2


From d8880795dabf2381ed1e98348f6d9c7ea6fab950 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:28 +0200
Subject: net/mlx5e: Implement DCBNL IEEE max rate

Add support for DCBNL IEEE get/set max rate.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
index 39d8069..3036f27 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c
@@ -35,6 +35,9 @@
 
 #define MLX5E_MAX_PRIORITY 8
 
+#define MLX5E_100MB (100000)
+#define MLX5E_1GB   (1000000)
+
 static int mlx5e_dcbnl_ieee_getets(struct net_device *netdev,
 				   struct ieee_ets *ets)
 {
@@ -219,9 +222,79 @@ static u8 mlx5e_dcbnl_setdcbx(struct net_device *dev, u8 mode)
 	return 0;
 }
 
+static int mlx5e_dcbnl_ieee_getmaxrate(struct net_device *netdev,
+				       struct ieee_maxrate *maxrate)
+{
+	struct mlx5e_priv *priv    = netdev_priv(netdev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+	u8 max_bw_value[IEEE_8021QAZ_MAX_TCS];
+	u8 max_bw_unit[IEEE_8021QAZ_MAX_TCS];
+	int err;
+	int i;
+
+	err = mlx5_query_port_ets_rate_limit(mdev, max_bw_value, max_bw_unit);
+	if (err)
+		return err;
+
+	memset(maxrate->tc_maxrate, 0, sizeof(maxrate->tc_maxrate));
+
+	for (i = 0; i <= mlx5_max_tc(mdev); i++) {
+		switch (max_bw_unit[i]) {
+		case MLX5_100_MBPS_UNIT:
+			maxrate->tc_maxrate[i] = max_bw_value[i] * MLX5E_100MB;
+			break;
+		case MLX5_GBPS_UNIT:
+			maxrate->tc_maxrate[i] = max_bw_value[i] * MLX5E_1GB;
+			break;
+		case MLX5_BW_NO_LIMIT:
+			break;
+		default:
+			WARN(true, "non-supported BW unit");
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static int mlx5e_dcbnl_ieee_setmaxrate(struct net_device *netdev,
+				       struct ieee_maxrate *maxrate)
+{
+	struct mlx5e_priv *priv    = netdev_priv(netdev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+	u8 max_bw_value[IEEE_8021QAZ_MAX_TCS];
+	u8 max_bw_unit[IEEE_8021QAZ_MAX_TCS];
+	__u64 upper_limit_mbps = roundup(255 * MLX5E_100MB, MLX5E_1GB);
+	int i;
+
+	memset(max_bw_value, 0, sizeof(max_bw_value));
+	memset(max_bw_unit, 0, sizeof(max_bw_unit));
+
+	for (i = 0; i <= mlx5_max_tc(mdev); i++) {
+		if (!maxrate->tc_maxrate[i]) {
+			max_bw_unit[i]  = MLX5_BW_NO_LIMIT;
+			continue;
+		}
+		if (maxrate->tc_maxrate[i] < upper_limit_mbps) {
+			max_bw_value[i] = div_u64(maxrate->tc_maxrate[i],
+						  MLX5E_100MB);
+			max_bw_value[i] = max_bw_value[i] ? max_bw_value[i] : 1;
+			max_bw_unit[i]  = MLX5_100_MBPS_UNIT;
+		} else {
+			max_bw_value[i] = div_u64(maxrate->tc_maxrate[i],
+						  MLX5E_1GB);
+			max_bw_unit[i]  = MLX5_GBPS_UNIT;
+		}
+	}
+
+	return mlx5_modify_port_ets_rate_limit(mdev, max_bw_value, max_bw_unit);
+}
+
 const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops = {
 	.ieee_getets	= mlx5e_dcbnl_ieee_getets,
 	.ieee_setets	= mlx5e_dcbnl_ieee_setets,
+	.ieee_getmaxrate = mlx5e_dcbnl_ieee_getmaxrate,
+	.ieee_setmaxrate = mlx5e_dcbnl_ieee_setmaxrate,
 	.ieee_getpfc	= mlx5e_dcbnl_ieee_getpfc,
 	.ieee_setpfc	= mlx5e_dcbnl_ieee_setpfc,
 	.getdcbx	= mlx5e_dcbnl_getdcbx,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index 569100d..d97605e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -450,6 +450,19 @@ static int mlx5_set_port_qetcr_reg(struct mlx5_core_dev *mdev, u32 *in,
 				    MLX5_REG_QETCR, 0, 1);
 }
 
+static int mlx5_query_port_qetcr_reg(struct mlx5_core_dev *mdev, u32 *out,
+				     int outlen)
+{
+	u32 in[MLX5_ST_SZ_DW(qtct_reg)];
+
+	if (!MLX5_CAP_GEN(mdev, ets))
+		return -ENOTSUPP;
+
+	memset(in, 0, sizeof(in));
+	return mlx5_core_access_reg(mdev, in, sizeof(in), out, outlen,
+				    MLX5_REG_QETCR, 0, 0);
+}
+
 int mlx5_set_port_tc_group(struct mlx5_core_dev *mdev, u8 *tc_group)
 {
 	u32 in[MLX5_ST_SZ_DW(qetc_reg)];
@@ -481,3 +494,55 @@ int mlx5_set_port_tc_bw_alloc(struct mlx5_core_dev *mdev, u8 *tc_bw)
 	return mlx5_set_port_qetcr_reg(mdev, in, sizeof(in));
 }
 EXPORT_SYMBOL_GPL(mlx5_set_port_tc_bw_alloc);
+
+int mlx5_modify_port_ets_rate_limit(struct mlx5_core_dev *mdev,
+				    u8 *max_bw_value,
+				    u8 *max_bw_units)
+{
+	u32 in[MLX5_ST_SZ_DW(qetc_reg)];
+	void *ets_tcn_conf;
+	int i;
+
+	memset(in, 0, sizeof(in));
+
+	MLX5_SET(qetc_reg, in, port_number, 1);
+
+	for (i = 0; i <= mlx5_max_tc(mdev); i++) {
+		ets_tcn_conf = MLX5_ADDR_OF(qetc_reg, in, tc_configuration[i]);
+
+		MLX5_SET(ets_tcn_config_reg, ets_tcn_conf, r, 1);
+		MLX5_SET(ets_tcn_config_reg, ets_tcn_conf, max_bw_units,
+			 max_bw_units[i]);
+		MLX5_SET(ets_tcn_config_reg, ets_tcn_conf, max_bw_value,
+			 max_bw_value[i]);
+	}
+
+	return mlx5_set_port_qetcr_reg(mdev, in, sizeof(in));
+}
+EXPORT_SYMBOL_GPL(mlx5_modify_port_ets_rate_limit);
+
+int mlx5_query_port_ets_rate_limit(struct mlx5_core_dev *mdev,
+				   u8 *max_bw_value,
+				   u8 *max_bw_units)
+{
+	u32 out[MLX5_ST_SZ_DW(qetc_reg)];
+	void *ets_tcn_conf;
+	int err;
+	int i;
+
+	err = mlx5_query_port_qetcr_reg(mdev, out, sizeof(out));
+	if (err)
+		return err;
+
+	for (i = 0; i <= mlx5_max_tc(mdev); i++) {
+		ets_tcn_conf = MLX5_ADDR_OF(qetc_reg, out, tc_configuration[i]);
+
+		max_bw_value[i] = MLX5_GET(ets_tcn_config_reg, ets_tcn_conf,
+					   max_bw_value);
+		max_bw_units[i] = MLX5_GET(ets_tcn_config_reg, ets_tcn_conf,
+					   max_bw_units);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_port_ets_rate_limit);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 987764a..bfc1ab055 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -351,6 +351,12 @@ enum {
 };
 
 enum {
+	MLX5_BW_NO_LIMIT   = 0,
+	MLX5_100_MBPS_UNIT = 3,
+	MLX5_GBPS_UNIT	   = 4,
+};
+
+enum {
 	MLX5_MAX_PAGE_SHIFT		= 31
 };
 
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 0c67e69..595c7b2 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -75,5 +75,11 @@ int mlx5_max_tc(struct mlx5_core_dev *mdev);
 int mlx5_set_port_prio_tc(struct mlx5_core_dev *mdev, u8 *prio_tc);
 int mlx5_set_port_tc_group(struct mlx5_core_dev *mdev, u8 *tc_group);
 int mlx5_set_port_tc_bw_alloc(struct mlx5_core_dev *mdev, u8 *tc_bw);
+int mlx5_modify_port_ets_rate_limit(struct mlx5_core_dev *mdev,
+				    u8 *max_bw_value,
+				    u8 *max_bw_unit);
+int mlx5_query_port_ets_rate_limit(struct mlx5_core_dev *mdev,
+				   u8 *max_bw_value,
+				   u8 *max_bw_unit);
 
 #endif /* __MLX5_PORT_H__ */
-- 
cgit v0.10.2


From 928cfe8745a62e60c1e8e06676a74724e7786024 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:29 +0200
Subject: net/mlx5e: Wake On LAN support

Implement set/get WOL by ethtool and added the needed
device commands and structures to mlx5_ifc.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 037fc4c..9ce87c6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -560,6 +560,12 @@ const char *mlx5_command_str(int command)
 	case MLX5_CMD_OP_ACCESS_REG:
 		return "MLX5_CMD_OP_ACCESS_REG";
 
+	case MLX5_CMD_OP_SET_WOL_ROL:
+		return "SET_WOL_ROL";
+
+	case MLX5_CMD_OP_QUERY_WOL_ROL:
+		return "QUERY_WOL_ROL";
+
 	default: return "unknown command opcode";
 	}
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 65624ac..e9760f8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -884,6 +884,129 @@ static int mlx5e_get_ts_info(struct net_device *dev,
 	return 0;
 }
 
+static __u32 mlx5e_get_wol_supported(struct mlx5_core_dev *mdev)
+{
+	__u32 ret = 0;
+
+	if (MLX5_CAP_GEN(mdev, wol_g))
+		ret |= WAKE_MAGIC;
+
+	if (MLX5_CAP_GEN(mdev, wol_s))
+		ret |= WAKE_MAGICSECURE;
+
+	if (MLX5_CAP_GEN(mdev, wol_a))
+		ret |= WAKE_ARP;
+
+	if (MLX5_CAP_GEN(mdev, wol_b))
+		ret |= WAKE_BCAST;
+
+	if (MLX5_CAP_GEN(mdev, wol_m))
+		ret |= WAKE_MCAST;
+
+	if (MLX5_CAP_GEN(mdev, wol_u))
+		ret |= WAKE_UCAST;
+
+	if (MLX5_CAP_GEN(mdev, wol_p))
+		ret |= WAKE_PHY;
+
+	return ret;
+}
+
+static __u32 mlx5e_refomrat_wol_mode_mlx5_to_linux(u8 mode)
+{
+	__u32 ret = 0;
+
+	if (mode & MLX5_WOL_MAGIC)
+		ret |= WAKE_MAGIC;
+
+	if (mode & MLX5_WOL_SECURED_MAGIC)
+		ret |= WAKE_MAGICSECURE;
+
+	if (mode & MLX5_WOL_ARP)
+		ret |= WAKE_ARP;
+
+	if (mode & MLX5_WOL_BROADCAST)
+		ret |= WAKE_BCAST;
+
+	if (mode & MLX5_WOL_MULTICAST)
+		ret |= WAKE_MCAST;
+
+	if (mode & MLX5_WOL_UNICAST)
+		ret |= WAKE_UCAST;
+
+	if (mode & MLX5_WOL_PHY_ACTIVITY)
+		ret |= WAKE_PHY;
+
+	return ret;
+}
+
+static u8 mlx5e_refomrat_wol_mode_linux_to_mlx5(__u32 mode)
+{
+	u8 ret = 0;
+
+	if (mode & WAKE_MAGIC)
+		ret |= MLX5_WOL_MAGIC;
+
+	if (mode & WAKE_MAGICSECURE)
+		ret |= MLX5_WOL_SECURED_MAGIC;
+
+	if (mode & WAKE_ARP)
+		ret |= MLX5_WOL_ARP;
+
+	if (mode & WAKE_BCAST)
+		ret |= MLX5_WOL_BROADCAST;
+
+	if (mode & WAKE_MCAST)
+		ret |= MLX5_WOL_MULTICAST;
+
+	if (mode & WAKE_UCAST)
+		ret |= MLX5_WOL_UNICAST;
+
+	if (mode & WAKE_PHY)
+		ret |= MLX5_WOL_PHY_ACTIVITY;
+
+	return ret;
+}
+
+static void mlx5e_get_wol(struct net_device *netdev,
+			  struct ethtool_wolinfo *wol)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+	u8 mlx5_wol_mode;
+	int err;
+
+	memset(wol, 0, sizeof(*wol));
+
+	wol->supported = mlx5e_get_wol_supported(mdev);
+	if (!wol->supported)
+		return;
+
+	err = mlx5_query_port_wol(mdev, &mlx5_wol_mode);
+	if (err)
+		return;
+
+	wol->wolopts = mlx5e_refomrat_wol_mode_mlx5_to_linux(mlx5_wol_mode);
+}
+
+static int mlx5e_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+	struct mlx5_core_dev *mdev = priv->mdev;
+	__u32 wol_supported = mlx5e_get_wol_supported(mdev);
+	u32 mlx5_wol_mode;
+
+	if (!wol_supported)
+		return -ENOTSUPP;
+
+	if (wol->wolopts & ~wol_supported)
+		return -EINVAL;
+
+	mlx5_wol_mode = mlx5e_refomrat_wol_mode_linux_to_mlx5(wol->wolopts);
+
+	return mlx5_set_port_wol(mdev, mlx5_wol_mode);
+}
+
 const struct ethtool_ops mlx5e_ethtool_ops = {
 	.get_drvinfo       = mlx5e_get_drvinfo,
 	.get_link          = ethtool_op_get_link,
@@ -908,4 +1031,6 @@ const struct ethtool_ops mlx5e_ethtool_ops = {
 	.get_pauseparam    = mlx5e_get_pauseparam,
 	.set_pauseparam    = mlx5e_set_pauseparam,
 	.get_ts_info       = mlx5e_get_ts_info,
+	.get_wol	   = mlx5e_get_wol,
+	.set_wol	   = mlx5e_set_wol,
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index d97605e..e1f2e10 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -546,3 +546,41 @@ int mlx5_query_port_ets_rate_limit(struct mlx5_core_dev *mdev,
 	return 0;
 }
 EXPORT_SYMBOL_GPL(mlx5_query_port_ets_rate_limit);
+
+int mlx5_set_port_wol(struct mlx5_core_dev *mdev, u8 wol_mode)
+{
+	u32 in[MLX5_ST_SZ_DW(set_wol_rol_in)];
+	u32 out[MLX5_ST_SZ_DW(set_wol_rol_out)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(set_wol_rol_in, in, opcode, MLX5_CMD_OP_SET_WOL_ROL);
+	MLX5_SET(set_wol_rol_in, in, wol_mode_valid, 1);
+	MLX5_SET(set_wol_rol_in, in, wol_mode, wol_mode);
+
+	return mlx5_cmd_exec_check_status(mdev, in, sizeof(in),
+					  out, sizeof(out));
+}
+EXPORT_SYMBOL_GPL(mlx5_set_port_wol);
+
+int mlx5_query_port_wol(struct mlx5_core_dev *mdev, u8 *wol_mode)
+{
+	u32 in[MLX5_ST_SZ_DW(query_wol_rol_in)];
+	u32 out[MLX5_ST_SZ_DW(query_wol_rol_out)];
+	int err;
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(query_wol_rol_in, in, opcode, MLX5_CMD_OP_QUERY_WOL_ROL);
+
+	err = mlx5_cmd_exec_check_status(mdev, in, sizeof(in),
+					 out, sizeof(out));
+
+	if (!err)
+		*wol_mode = MLX5_GET(query_wol_rol_out, out, wol_mode);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_query_port_wol);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index bfc1ab055..68a56bc 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -1183,6 +1183,17 @@ enum {
 	MLX5_RQC_RQ_TYPE_MEMORY_RQ_RPM    = 0x1,
 };
 
+enum mlx5_wol_mode {
+	MLX5_WOL_DISABLE        = 0,
+	MLX5_WOL_SECURED_MAGIC  = 1 << 1,
+	MLX5_WOL_MAGIC          = 1 << 2,
+	MLX5_WOL_ARP            = 1 << 3,
+	MLX5_WOL_BROADCAST      = 1 << 4,
+	MLX5_WOL_MULTICAST      = 1 << 5,
+	MLX5_WOL_UNICAST        = 1 << 6,
+	MLX5_WOL_PHY_ACTIVITY   = 1 << 7,
+};
+
 /* MLX5 DEV CAPs */
 
 /* TODO: EAT.ME */
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index ec957e0..03ffe95 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -166,6 +166,8 @@ enum {
 	MLX5_CMD_OP_SET_L2_TABLE_ENTRY            = 0x829,
 	MLX5_CMD_OP_QUERY_L2_TABLE_ENTRY          = 0x82a,
 	MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY         = 0x82b,
+	MLX5_CMD_OP_SET_WOL_ROL                   = 0x830,
+	MLX5_CMD_OP_QUERY_WOL_ROL                 = 0x831,
 	MLX5_CMD_OP_CREATE_TIR                    = 0x900,
 	MLX5_CMD_OP_MODIFY_TIR                    = 0x901,
 	MLX5_CMD_OP_DESTROY_TIR                   = 0x902,
@@ -731,7 +733,17 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         log_max_msg[0x5];
 	u8         reserved_at_1c7[0x4];
 	u8         max_tc[0x4];
-	u8         reserved_at_1cf[0x10];
+	u8         reserved_at_1cf[0x6];
+	u8         rol_s[0x1];
+	u8         rol_g[0x1];
+	u8         reserved_at_1d7[0x1];
+	u8         wol_s[0x1];
+	u8         wol_g[0x1];
+	u8         wol_a[0x1];
+	u8         wol_b[0x1];
+	u8         wol_m[0x1];
+	u8         wol_u[0x1];
+	u8         wol_p[0x1];
 
 	u8         stat_rate_support[0x10];
 	u8         reserved_at_1ef[0xc];
@@ -6873,6 +6885,54 @@ struct mlx5_ifc_mtt_bits {
 	u8         rd_en[0x1];
 };
 
+struct mlx5_ifc_query_wol_rol_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x10];
+	u8         rol_mode[0x8];
+	u8         wol_mode[0x8];
+
+	u8         reserved_at_60[0x20];
+};
+
+struct mlx5_ifc_query_wol_rol_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_set_wol_rol_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_set_wol_rol_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         rol_mode_valid[0x1];
+	u8         wol_mode_valid[0x1];
+	u8         reserved_at_42[0xe];
+	u8         rol_mode[0x8];
+	u8         wol_mode[0x8];
+
+	u8         reserved_at_60[0x20];
+};
+
 enum {
 	MLX5_INITIAL_SEG_NIC_INTERFACE_FULL_DRIVER  = 0x0,
 	MLX5_INITIAL_SEG_NIC_INTERFACE_DISABLED     = 0x1,
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 595c7b2..a1d145a 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -81,5 +81,7 @@ int mlx5_modify_port_ets_rate_limit(struct mlx5_core_dev *mdev,
 int mlx5_query_port_ets_rate_limit(struct mlx5_core_dev *mdev,
 				   u8 *max_bw_value,
 				   u8 *max_bw_unit);
+int mlx5_set_port_wol(struct mlx5_core_dev *mdev, u8 wol_mode);
+int mlx5_query_port_wol(struct mlx5_core_dev *mdev, u8 *wol_mode);
 
 #endif /* __MLX5_PORT_H__ */
-- 
cgit v0.10.2


From 5f6d12d10f0c835fcd821f883f05c1107127986e Mon Sep 17 00:00:00 2001
From: Matthew Finlay <matt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:30 +0200
Subject: net/mlx5e: Move to checksum complete

Use checksum complete for all IP packets, unless they are HW LRO,
in which case, use checksum unnecessary.

Signed-off-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index dd959d9..519a07f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -167,14 +167,15 @@ static inline bool is_first_ethertype_ip(struct sk_buff *skb)
 static inline void mlx5e_handle_csum(struct net_device *netdev,
 				     struct mlx5_cqe64 *cqe,
 				     struct mlx5e_rq *rq,
-				     struct sk_buff *skb)
+				     struct sk_buff *skb,
+				     bool   lro)
 {
 	if (unlikely(!(netdev->features & NETIF_F_RXCSUM)))
 		goto csum_none;
 
-	if (likely(cqe->hds_ip_ext & CQE_L4_OK)) {
+	if (lro) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
-	} else if (is_first_ethertype_ip(skb)) {
+	} else if (likely(is_first_ethertype_ip(skb))) {
 		skb->ip_summed = CHECKSUM_COMPLETE;
 		skb->csum = csum_unfold((__force __sum16)cqe->check_sum);
 		rq->stats.csum_sw++;
@@ -211,7 +212,7 @@ static inline void mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe,
 	if (unlikely(mlx5e_rx_hw_stamp(tstamp)))
 		mlx5e_fill_hwstamp(tstamp, get_cqe_ts(cqe), skb_hwtstamps(skb));
 
-	mlx5e_handle_csum(netdev, cqe, rq, skb);
+	mlx5e_handle_csum(netdev, cqe, rq, skb, !!lro_num_seg);
 
 	skb->protocol = eth_type_trans(skb, netdev);
 
-- 
cgit v0.10.2


From 1afff42c062fe1d1f8d99423dddfb8ca5f100574 Mon Sep 17 00:00:00 2001
From: Matthew Finlay <matt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:31 +0200
Subject: net/mlx5e: Protect en header file from redefinitions

add ifndef to en.h. needed for upcoming vxlan patchset.

Signed-off-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index dfbc4e5..786a247 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2015-2016, Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -29,6 +29,8 @@
  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
+#ifndef __MLX5_EN_H__
+#define __MLX5_EN_H__
 
 #include <linux/if_vlan.h>
 #include <linux/etherdevice.h>
@@ -678,3 +680,5 @@ int mlx5e_dcbnl_ieee_setets_core(struct mlx5e_priv *priv, struct ieee_ets *ets);
 #endif
 
 u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev);
+
+#endif /* __MLX5_EN_H__ */
-- 
cgit v0.10.2


From b3f63c3d5e2cbb9c800516ef47e32d8cb0cf237b Mon Sep 17 00:00:00 2001
From: Matthew Finlay <matt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:32 +0200
Subject: net/mlx5e: Add netdev support for VXLAN tunneling

If a VXLAN udp dport is added to device it will:

   - Configure the hardware to offload the port (up to the max
     supported).
   - Advertise NETIF_F_GSO_UDP_TUNNEL and supported hw_enc_features.

Signed-off-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 1a82e23..11b592d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -6,6 +6,6 @@ mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 
 mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \
 		en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \
-		en_txrx.o en_clock.o
+		en_txrx.o en_clock.o vxlan.o
 
 mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) +=  en_dcbnl.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index 9ce87c6..97f5114 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2013-2016, Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -566,6 +566,12 @@ const char *mlx5_command_str(int command)
 	case MLX5_CMD_OP_QUERY_WOL_ROL:
 		return "QUERY_WOL_ROL";
 
+	case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
+		return "ADD_VXLAN_UDP_DPORT";
+
+	case MLX5_CMD_OP_DELETE_VXLAN_UDP_DPORT:
+		return "DELETE_VXLAN_UDP_DPORT";
+
 	default: return "unknown command opcode";
 	}
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 786a247..a700c57 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -501,6 +501,11 @@ struct mlx5e_vlan_db {
 	bool          filter_disabled;
 };
 
+struct mlx5e_vxlan_db {
+	spinlock_t			lock; /* protect vxlan table */
+	struct radix_tree_root		tree;
+};
+
 struct mlx5e_flow_table {
 	int num_groups;
 	struct mlx5_flow_table		*t;
@@ -535,6 +540,7 @@ struct mlx5e_priv {
 	struct mlx5e_flow_tables   fts;
 	struct mlx5e_eth_addr_db   eth_addr;
 	struct mlx5e_vlan_db       vlan;
+	struct mlx5e_vxlan_db      vxlan;
 
 	struct mlx5e_params        params;
 	spinlock_t                 async_events_spinlock; /* sync hw events */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 704d75c..6f7eb3b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2015-2016, Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -31,8 +31,10 @@
  */
 
 #include <linux/mlx5/fs.h>
+#include <net/vxlan.h>
 #include "en.h"
 #include "eswitch.h"
+#include "vxlan.h"
 
 struct mlx5e_rq_param {
 	u32                        rqc[MLX5_ST_SZ_DW(rqc)];
@@ -2078,6 +2080,78 @@ static int mlx5e_get_vf_stats(struct net_device *dev,
 					    vf_stats);
 }
 
+static void mlx5e_add_vxlan_port(struct net_device *netdev,
+				 sa_family_t sa_family, __be16 port)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+
+	if (!mlx5e_vxlan_allowed(priv->mdev))
+		return;
+
+	mlx5e_vxlan_add_port(priv, be16_to_cpu(port));
+}
+
+static void mlx5e_del_vxlan_port(struct net_device *netdev,
+				 sa_family_t sa_family, __be16 port)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+
+	if (!mlx5e_vxlan_allowed(priv->mdev))
+		return;
+
+	mlx5e_vxlan_del_port(priv, be16_to_cpu(port));
+}
+
+static netdev_features_t mlx5e_vxlan_features_check(struct mlx5e_priv *priv,
+						    struct sk_buff *skb,
+						    netdev_features_t features)
+{
+	struct udphdr *udph;
+	u16 proto;
+	u16 port = 0;
+
+	switch (vlan_get_protocol(skb)) {
+	case htons(ETH_P_IP):
+		proto = ip_hdr(skb)->protocol;
+		break;
+	case htons(ETH_P_IPV6):
+		proto = ipv6_hdr(skb)->nexthdr;
+		break;
+	default:
+		goto out;
+	}
+
+	if (proto == IPPROTO_UDP) {
+		udph = udp_hdr(skb);
+		port = be16_to_cpu(udph->dest);
+	}
+
+	/* Verify if UDP port is being offloaded by HW */
+	if (port && mlx5e_vxlan_lookup_port(priv, port))
+		return features;
+
+out:
+	/* Disable CSUM and GSO if the udp dport is not offloaded by HW */
+	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+}
+
+static netdev_features_t mlx5e_features_check(struct sk_buff *skb,
+					      struct net_device *netdev,
+					      netdev_features_t features)
+{
+	struct mlx5e_priv *priv = netdev_priv(netdev);
+
+	features = vlan_features_check(skb, features);
+	features = vxlan_features_check(skb, features);
+
+	/* Validate if the tunneled packet is being offloaded by HW */
+	if (skb->encapsulation &&
+	    (features & NETIF_F_CSUM_MASK || features & NETIF_F_GSO_MASK))
+		return mlx5e_vxlan_features_check(priv, skb, features);
+
+	return features;
+}
+
 static const struct net_device_ops mlx5e_netdev_ops_basic = {
 	.ndo_open                = mlx5e_open,
 	.ndo_stop                = mlx5e_close,
@@ -2108,6 +2182,9 @@ static const struct net_device_ops mlx5e_netdev_ops_sriov = {
 	.ndo_set_features        = mlx5e_set_features,
 	.ndo_change_mtu          = mlx5e_change_mtu,
 	.ndo_do_ioctl            = mlx5e_ioctl,
+	.ndo_add_vxlan_port      = mlx5e_add_vxlan_port,
+	.ndo_del_vxlan_port      = mlx5e_del_vxlan_port,
+	.ndo_features_check      = mlx5e_features_check,
 	.ndo_set_vf_mac          = mlx5e_set_vf_mac,
 	.ndo_set_vf_vlan         = mlx5e_set_vf_vlan,
 	.ndo_get_vf_config       = mlx5e_get_vf_config,
@@ -2264,6 +2341,16 @@ static void mlx5e_build_netdev(struct net_device *netdev)
 	netdev->hw_features      |= NETIF_F_HW_VLAN_CTAG_RX;
 	netdev->hw_features      |= NETIF_F_HW_VLAN_CTAG_FILTER;
 
+	if (mlx5e_vxlan_allowed(mdev)) {
+		netdev->hw_features     |= NETIF_F_GSO_UDP_TUNNEL;
+		netdev->hw_enc_features |= NETIF_F_IP_CSUM;
+		netdev->hw_enc_features |= NETIF_F_RXCSUM;
+		netdev->hw_enc_features |= NETIF_F_TSO;
+		netdev->hw_enc_features |= NETIF_F_TSO6;
+		netdev->hw_enc_features |= NETIF_F_RXHASH;
+		netdev->hw_enc_features |= NETIF_F_GSO_UDP_TUNNEL;
+	}
+
 	netdev->features          = netdev->hw_features;
 	if (!priv->params.lro_en)
 		netdev->features  &= ~NETIF_F_LRO;
@@ -2387,6 +2474,8 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 
 	mlx5e_init_eth_addr(priv);
 
+	mlx5e_vxlan_init(priv);
+
 #ifdef CONFIG_MLX5_CORE_EN_DCB
 	mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets);
 #endif
@@ -2397,6 +2486,9 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 		goto err_destroy_flow_tables;
 	}
 
+	if (mlx5e_vxlan_allowed(mdev))
+		vxlan_get_rx_port(netdev);
+
 	mlx5e_enable_async_events(priv);
 	schedule_work(&priv->set_rx_mode_work);
 
@@ -2449,6 +2541,7 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv)
 	mlx5e_disable_async_events(priv);
 	flush_scheduled_work();
 	unregister_netdev(netdev);
+	mlx5e_vxlan_cleanup(priv);
 	mlx5e_destroy_flow_tables(priv);
 	mlx5e_destroy_tirs(priv);
 	mlx5e_destroy_rqt(priv, MLX5E_SINGLE_RQ_RQT);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c
new file mode 100644
index 0000000..9f10df2
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c
@@ -0,0 +1,170 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies, Ltd.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mlx5/driver.h>
+#include "mlx5_core.h"
+#include "vxlan.h"
+
+void mlx5e_vxlan_init(struct mlx5e_priv *priv)
+{
+	struct mlx5e_vxlan_db *vxlan_db = &priv->vxlan;
+
+	spin_lock_init(&vxlan_db->lock);
+	INIT_RADIX_TREE(&vxlan_db->tree, GFP_ATOMIC);
+}
+
+static int mlx5e_vxlan_core_add_port_cmd(struct mlx5_core_dev *mdev, u16 port)
+{
+	struct mlx5_outbox_hdr *hdr;
+	int err;
+
+	u32 in[MLX5_ST_SZ_DW(add_vxlan_udp_dport_in)];
+	u32 out[MLX5_ST_SZ_DW(add_vxlan_udp_dport_out)];
+
+	memset(in, 0, sizeof(in));
+	memset(out, 0, sizeof(out));
+
+	MLX5_SET(add_vxlan_udp_dport_in, in, opcode,
+		 MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT);
+	MLX5_SET(add_vxlan_udp_dport_in, in, vxlan_udp_port, port);
+
+	err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out));
+	if (err)
+		return err;
+
+	hdr = (struct mlx5_outbox_hdr *)out;
+	return hdr->status ? -ENOMEM : 0;
+}
+
+static int mlx5e_vxlan_core_del_port_cmd(struct mlx5_core_dev *mdev, u16 port)
+{
+	u32 in[MLX5_ST_SZ_DW(delete_vxlan_udp_dport_in)];
+	u32 out[MLX5_ST_SZ_DW(delete_vxlan_udp_dport_out)];
+
+	memset(&in, 0, sizeof(in));
+	memset(&out, 0, sizeof(out));
+
+	MLX5_SET(delete_vxlan_udp_dport_in, in, opcode,
+		 MLX5_CMD_OP_DELETE_VXLAN_UDP_DPORT);
+	MLX5_SET(delete_vxlan_udp_dport_in, in, vxlan_udp_port, port);
+
+	return mlx5_cmd_exec_check_status(mdev, in, sizeof(in), out,
+					  sizeof(out));
+}
+
+struct mlx5e_vxlan *mlx5e_vxlan_lookup_port(struct mlx5e_priv *priv, u16 port)
+{
+	struct mlx5e_vxlan_db *vxlan_db = &priv->vxlan;
+	struct mlx5e_vxlan *vxlan;
+
+	spin_lock(&vxlan_db->lock);
+	vxlan = radix_tree_lookup(&vxlan_db->tree, port);
+	spin_unlock(&vxlan_db->lock);
+
+	return vxlan;
+}
+
+int mlx5e_vxlan_add_port(struct mlx5e_priv *priv, u16 port)
+{
+	struct mlx5e_vxlan_db *vxlan_db = &priv->vxlan;
+	struct mlx5e_vxlan *vxlan;
+	int err;
+
+	err = mlx5e_vxlan_core_add_port_cmd(priv->mdev, port);
+	if (err)
+		return err;
+
+	vxlan = kzalloc(sizeof(*vxlan), GFP_KERNEL);
+	if (!vxlan) {
+		err = -ENOMEM;
+		goto err_delete_port;
+	}
+
+	vxlan->udp_port = port;
+
+	spin_lock_irq(&vxlan_db->lock);
+	err = radix_tree_insert(&vxlan_db->tree, vxlan->udp_port, vxlan);
+	spin_unlock_irq(&vxlan_db->lock);
+	if (err)
+		goto err_free;
+
+	return 0;
+
+err_free:
+	kfree(vxlan);
+err_delete_port:
+	mlx5e_vxlan_core_del_port_cmd(priv->mdev, port);
+	return err;
+}
+
+static void __mlx5e_vxlan_core_del_port(struct mlx5e_priv *priv, u16 port)
+{
+	struct mlx5e_vxlan_db *vxlan_db = &priv->vxlan;
+	struct mlx5e_vxlan *vxlan;
+
+	spin_lock_irq(&vxlan_db->lock);
+	vxlan = radix_tree_delete(&vxlan_db->tree, port);
+	spin_unlock_irq(&vxlan_db->lock);
+
+	if (!vxlan)
+		return;
+
+	mlx5e_vxlan_core_del_port_cmd(priv->mdev, vxlan->udp_port);
+
+	kfree(vxlan);
+}
+
+void mlx5e_vxlan_del_port(struct mlx5e_priv *priv, u16 port)
+{
+	if (!mlx5e_vxlan_lookup_port(priv, port))
+		return;
+
+	__mlx5e_vxlan_core_del_port(priv, port);
+}
+
+void mlx5e_vxlan_cleanup(struct mlx5e_priv *priv)
+{
+	struct mlx5e_vxlan_db *vxlan_db = &priv->vxlan;
+	struct mlx5e_vxlan *vxlan;
+	unsigned int port = 0;
+
+	spin_lock_irq(&vxlan_db->lock);
+	while (radix_tree_gang_lookup(&vxlan_db->tree, (void **)&vxlan, port, 1)) {
+		port = vxlan->udp_port;
+		spin_unlock_irq(&vxlan_db->lock);
+		__mlx5e_vxlan_core_del_port(priv, (u16)port);
+		spin_lock_irq(&vxlan_db->lock);
+	}
+	spin_unlock_irq(&vxlan_db->lock);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vxlan.h b/drivers/net/ethernet/mellanox/mlx5/core/vxlan.h
new file mode 100644
index 0000000..a016850
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vxlan.h
@@ -0,0 +1,54 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies, Ltd.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef __MLX5_VXLAN_H__
+#define __MLX5_VXLAN_H__
+
+#include <linux/mlx5/driver.h>
+#include "en.h"
+
+struct mlx5e_vxlan {
+	u16 udp_port;
+};
+
+static inline bool mlx5e_vxlan_allowed(struct mlx5_core_dev *mdev)
+{
+	return (MLX5_CAP_ETH(mdev, tunnel_stateless_vxlan) &&
+		mlx5_core_is_pf(mdev));
+}
+
+void mlx5e_vxlan_init(struct mlx5e_priv *priv);
+int  mlx5e_vxlan_add_port(struct mlx5e_priv *priv, u16 port);
+void mlx5e_vxlan_del_port(struct mlx5e_priv *priv, u16 port);
+struct mlx5e_vxlan *mlx5e_vxlan_lookup_port(struct mlx5e_priv *priv, u16 port);
+void mlx5e_vxlan_cleanup(struct mlx5e_priv *priv);
+
+#endif /* __MLX5_VXLAN_H__ */
-- 
cgit v0.10.2


From 9879515895ffe46a424d00ad5ce92e32fd61790b Mon Sep 17 00:00:00 2001
From: Matthew Finlay <matt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:33 +0200
Subject: net/mlx5e: Add TX stateless offloads for tunneling

Add support for TSO and TX checksum when using hw assisted,
tunneled offloads.

Signed-off-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 00d855a..6ce2884 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2015-2016, Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
@@ -185,9 +185,14 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 
 	memset(wqe, 0, sizeof(*wqe));
 
-	if (likely(skb->ip_summed == CHECKSUM_PARTIAL))
-		eseg->cs_flags	= MLX5_ETH_WQE_L3_CSUM | MLX5_ETH_WQE_L4_CSUM;
-	else
+	if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
+		eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM;
+		if (skb->encapsulation)
+			eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM |
+					  MLX5_ETH_WQE_L4_INNER_CSUM;
+		else
+			eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM;
+	} else
 		sq->stats.csum_offload_none++;
 
 	if (sq->cc != sq->prev_cc) {
@@ -200,8 +205,13 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 
 		eseg->mss    = cpu_to_be16(skb_shinfo(skb)->gso_size);
 		opcode       = MLX5_OPCODE_LSO;
-		ihs          = skb_transport_offset(skb) + tcp_hdrlen(skb);
-		payload_len  = skb->len - ihs;
+
+		if (skb->encapsulation)
+			ihs = skb_inner_transport_offset(skb) + inner_tcp_hdrlen(skb);
+		else
+			ihs = skb_transport_offset(skb) + tcp_hdrlen(skb);
+
+		payload_len   = skb->len - ihs;
 		wi->num_bytes = skb->len +
 				(skb_shinfo(skb)->gso_segs - 1) * ihs;
 		sq->stats.tso_packets++;
-- 
cgit v0.10.2


From 89db09eb5979b74f1f50ee2fb2ce519ee8414c48 Mon Sep 17 00:00:00 2001
From: Matthew Finlay <matt@mellanox.com>
Date: Mon, 22 Feb 2016 18:17:34 +0200
Subject: net/mlx5e: Add TX inner packet counters

Add TSO and TX checksum counters for tunneled, inner packets

Signed-off-by: Matthew Finlay <matt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index a700c57..1dca3dc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -103,12 +103,15 @@ static const char vport_strings[][ETH_GSTRING_LEN] = {
 	/* SW counters */
 	"tso_packets",
 	"tso_bytes",
+	"tso_inner_packets",
+	"tso_inner_bytes",
 	"lro_packets",
 	"lro_bytes",
 	"rx_csum_good",
 	"rx_csum_none",
 	"rx_csum_sw",
 	"tx_csum_offload",
+	"tx_csum_inner",
 	"tx_queue_stopped",
 	"tx_queue_wake",
 	"tx_queue_dropped",
@@ -141,18 +144,21 @@ struct mlx5e_vport_stats {
 	/* SW counters */
 	u64 tso_packets;
 	u64 tso_bytes;
+	u64 tso_inner_packets;
+	u64 tso_inner_bytes;
 	u64 lro_packets;
 	u64 lro_bytes;
 	u64 rx_csum_good;
 	u64 rx_csum_none;
 	u64 rx_csum_sw;
 	u64 tx_csum_offload;
+	u64 tx_csum_inner;
 	u64 tx_queue_stopped;
 	u64 tx_queue_wake;
 	u64 tx_queue_dropped;
 	u64 rx_wqe_err;
 
-#define NUM_VPORT_COUNTERS     32
+#define NUM_VPORT_COUNTERS     35
 };
 
 static const char pport_strings[][ETH_GSTRING_LEN] = {
@@ -252,7 +258,10 @@ static const char sq_stats_strings[][ETH_GSTRING_LEN] = {
 	"packets",
 	"tso_packets",
 	"tso_bytes",
+	"tso_inner_packets",
+	"tso_inner_bytes",
 	"csum_offload_none",
+	"csum_offload_inner",
 	"stopped",
 	"wake",
 	"dropped",
@@ -263,12 +272,15 @@ struct mlx5e_sq_stats {
 	u64 packets;
 	u64 tso_packets;
 	u64 tso_bytes;
+	u64 tso_inner_packets;
+	u64 tso_inner_bytes;
 	u64 csum_offload_none;
+	u64 csum_offload_inner;
 	u64 stopped;
 	u64 wake;
 	u64 dropped;
 	u64 nop;
-#define NUM_SQ_STATS 8
+#define NUM_SQ_STATS 11
 };
 
 struct mlx5e_stats {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 6f7eb3b..0d45f35 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -145,9 +145,12 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 	/* Collect firts the SW counters and then HW for consistency */
 	s->tso_packets		= 0;
 	s->tso_bytes		= 0;
+	s->tso_inner_packets	= 0;
+	s->tso_inner_bytes	= 0;
 	s->tx_queue_stopped	= 0;
 	s->tx_queue_wake	= 0;
 	s->tx_queue_dropped	= 0;
+	s->tx_csum_inner	= 0;
 	tx_offload_none		= 0;
 	s->lro_packets		= 0;
 	s->lro_bytes		= 0;
@@ -168,9 +171,12 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 
 			s->tso_packets		+= sq_stats->tso_packets;
 			s->tso_bytes		+= sq_stats->tso_bytes;
+			s->tso_inner_packets	+= sq_stats->tso_inner_packets;
+			s->tso_inner_bytes	+= sq_stats->tso_inner_bytes;
 			s->tx_queue_stopped	+= sq_stats->stopped;
 			s->tx_queue_wake	+= sq_stats->wake;
 			s->tx_queue_dropped	+= sq_stats->dropped;
+			s->tx_csum_inner	+= sq_stats->csum_offload_inner;
 			tx_offload_none		+= sq_stats->csum_offload_none;
 		}
 	}
@@ -245,7 +251,7 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 		s->tx_broadcast_bytes;
 
 	/* Update calculated offload counters */
-	s->tx_csum_offload = s->tx_packets - tx_offload_none;
+	s->tx_csum_offload = s->tx_packets - tx_offload_none - s->tx_csum_inner;
 	s->rx_csum_good    = s->rx_packets - s->rx_csum_none -
 			       s->rx_csum_sw;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 6ce2884..a05c070 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -187,11 +187,13 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 
 	if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
 		eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM;
-		if (skb->encapsulation)
+		if (skb->encapsulation) {
 			eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM |
 					  MLX5_ETH_WQE_L4_INNER_CSUM;
-		else
+			sq->stats.csum_offload_inner++;
+		} else {
 			eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM;
+		}
 	} else
 		sq->stats.csum_offload_none++;
 
@@ -201,21 +203,21 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	}
 
 	if (skb_is_gso(skb)) {
-		u32 payload_len;
-
 		eseg->mss    = cpu_to_be16(skb_shinfo(skb)->gso_size);
 		opcode       = MLX5_OPCODE_LSO;
 
-		if (skb->encapsulation)
+		if (skb->encapsulation) {
 			ihs = skb_inner_transport_offset(skb) + inner_tcp_hdrlen(skb);
-		else
+			sq->stats.tso_inner_packets++;
+			sq->stats.tso_inner_bytes += skb->len - ihs;
+		} else {
 			ihs = skb_transport_offset(skb) + tcp_hdrlen(skb);
+			sq->stats.tso_packets++;
+			sq->stats.tso_bytes += skb->len - ihs;
+		}
 
-		payload_len   = skb->len - ihs;
 		wi->num_bytes = skb->len +
 				(skb_shinfo(skb)->gso_segs - 1) * ihs;
-		sq->stats.tso_packets++;
-		sq->stats.tso_bytes += payload_len;
 	} else {
 		bf = sq->bf_budget &&
 		     !skb->xmit_more &&
-- 
cgit v0.10.2


From 918c023f29ab2dd8c63cfcc6a1239ee15933871a Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 24 Feb 2016 09:29:38 -0800
Subject: flow_dissector: Check for IP fragmentation even if not using IPv4
 address

This patch corrects the logic for the IPv4 parsing so that it is consistent
with how we handle IPv6.  Specifically if we do not have the flow key
indicating we want the addresses we still may need to take a look at the IP
fragmentation bits and to see if we should stop after we have recognized
the L3 header.

Fixes: 807e165dc44f ("flow_dissector: Add control/reporting of fragmentation")
Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 12e7003..1f88f82 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -178,15 +178,16 @@ ip:
 
 		ip_proto = iph->protocol;
 
-		if (!dissector_uses_key(flow_dissector,
-					FLOW_DISSECTOR_KEY_IPV4_ADDRS))
-			break;
+		if (dissector_uses_key(flow_dissector,
+				       FLOW_DISSECTOR_KEY_IPV4_ADDRS)) {
+			key_addrs = skb_flow_dissector_target(flow_dissector,
+							      FLOW_DISSECTOR_KEY_IPV4_ADDRS,
+							      target_container);
 
-		key_addrs = skb_flow_dissector_target(flow_dissector,
-			      FLOW_DISSECTOR_KEY_IPV4_ADDRS, target_container);
-		memcpy(&key_addrs->v4addrs, &iph->saddr,
-		       sizeof(key_addrs->v4addrs));
-		key_control->addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS;
+			memcpy(&key_addrs->v4addrs, &iph->saddr,
+			       sizeof(key_addrs->v4addrs));
+			key_control->addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS;
+		}
 
 		if (ip_is_fragment(iph)) {
 			key_control->flags |= FLOW_DIS_IS_FRAGMENT;
-- 
cgit v0.10.2


From 43d2ccb3c122a47524019d1831a54f07f7fcb978 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 24 Feb 2016 09:29:44 -0800
Subject: flow_dissector: Fix fragment handling for header length computation

It turns out that for IPv4 we were reporting the ip_proto of the fragment,
and for IPv6 we were not.  This patch updates that behavior so that we
always report the IP protocol of the fragment.  In addition it takes the
steps of updating the payload offset code so that we will determine the
start of the payload not including the L4 header for any fragment after the
first.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 1f88f82..8bd745f 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -448,13 +448,12 @@ ip_proto_again:
 		key_control->flags |= FLOW_DIS_IS_FRAGMENT;
 
 		nhoff += sizeof(_fh);
+		ip_proto = fh->nexthdr;
 
 		if (!(fh->frag_off & htons(IP6_OFFSET))) {
 			key_control->flags |= FLOW_DIS_FIRST_FRAG;
-			if (flags & FLOW_DISSECTOR_F_PARSE_1ST_FRAG) {
-				ip_proto = fh->nexthdr;
+			if (flags & FLOW_DISSECTOR_F_PARSE_1ST_FRAG)
 				goto ip_proto_again;
-			}
 		}
 		goto out_good;
 	}
@@ -741,6 +740,11 @@ u32 __skb_get_poff(const struct sk_buff *skb, void *data,
 {
 	u32 poff = keys->control.thoff;
 
+	/* skip L4 headers for fragments after the first */
+	if ((keys->control.flags & FLOW_DIS_IS_FRAGMENT) &&
+	    !(keys->control.flags & FLOW_DIS_FIRST_FRAG))
+		return poff;
+
 	switch (keys->basic.ip_proto) {
 	case IPPROTO_TCP: {
 		/* access doff as u8 to avoid unaligned access */
-- 
cgit v0.10.2


From 224516b3a798a0563346748744f8cd19feaf09be Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 24 Feb 2016 09:29:51 -0800
Subject: flow_dissector: Correctly handle parsing FCoE

The flow dissector bits handling FCoE didn't bother to actually validate
that the space there was enough for the FCoE header.  So we need to update
things so that if there is room we add the header and report a good result,
otherwise we do not add the header, and report the bad result.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 8bd745f..6288153 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -340,8 +340,11 @@ mpls:
 	}
 
 	case htons(ETH_P_FCOE):
-		key_control->thoff = (u16)(nhoff + FCOE_HEADER_LEN);
-		/* fall through */
+		if ((hlen - nhoff) < FCOE_HEADER_LEN)
+			goto out_bad;
+
+		nhoff += FCOE_HEADER_LEN;
+		goto out_good;
 	default:
 		goto out_bad;
 	}
-- 
cgit v0.10.2


From b3c3106ce3f4646a008cd238b16f899ae14fd2a7 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 24 Feb 2016 09:29:57 -0800
Subject: flow_dissector: Use same pointer for IPv4 and IPv6 addresses

The IPv6 parsing was using a local pointer when it could use the same
pointer as the IPv4 portion of the code since the key_addrs can support
both IPv4 and IPv6 as it is just a pointer.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 6288153..7c7b873 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -220,13 +220,12 @@ ipv6:
 
 		if (dissector_uses_key(flow_dissector,
 				       FLOW_DISSECTOR_KEY_IPV6_ADDRS)) {
-			struct flow_dissector_key_ipv6_addrs *key_ipv6_addrs;
-
-			key_ipv6_addrs = skb_flow_dissector_target(flow_dissector,
-								   FLOW_DISSECTOR_KEY_IPV6_ADDRS,
-								   target_container);
+			key_addrs = skb_flow_dissector_target(flow_dissector,
+							      FLOW_DISSECTOR_KEY_IPV6_ADDRS,
+							      target_container);
 
-			memcpy(key_ipv6_addrs, &iph->saddr, sizeof(*key_ipv6_addrs));
+			memcpy(&key_addrs->v6addrs, &iph->saddr,
+			       sizeof(key_addrs->v6addrs));
 			key_control->addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
 		}
 
-- 
cgit v0.10.2


From d975ddd69698d068a8ebd294c382f1c791b380da Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 24 Feb 2016 09:30:04 -0800
Subject: eth: Pull header from first fragment via eth_get_headlen

We want to try and pull the L4 header in if it is available in the first
fragment.  As such add the flag to indicate we want to pull the headers on
the first fragment in.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Acked-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 1038717..66dff5e 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -125,6 +125,7 @@ EXPORT_SYMBOL(eth_header);
  */
 u32 eth_get_headlen(void *data, unsigned int len)
 {
+	const unsigned int flags = FLOW_DISSECTOR_F_PARSE_1ST_FRAG;
 	const struct ethhdr *eth = (const struct ethhdr *)data;
 	struct flow_keys keys;
 
@@ -134,7 +135,7 @@ u32 eth_get_headlen(void *data, unsigned int len)
 
 	/* parse any remaining L2/L3 headers, check for L4 */
 	if (!skb_flow_dissect_flow_keys_buf(&keys, data, eth->h_proto,
-					    sizeof(*eth), len, 0))
+					    sizeof(*eth), len, flags))
 		return max_t(u32, keys.control.thoff, sizeof(*eth));
 
 	/* parse for any L4 headers */
-- 
cgit v0.10.2


From 45040978c8994d1401baf5cc5ac71c1495d4e120 Mon Sep 17 00:00:00 2001
From: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Date: Wed, 24 Feb 2016 20:32:21 +0100
Subject: netfilter: ipset: Fix set:list type crash when flush/dump set in
 parallel

Flushing/listing entries was not RCU safe, so parallel flush/dump
could lead to kernel crash. Bug reported by Deniz Eren.

Fixes netfilter bugzilla id #1050.

Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>

diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index 95db43f..7e6568c 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -985,6 +985,9 @@ static int ip_set_destroy(struct net *net, struct sock *ctnl,
 	if (unlikely(protocol_failed(attr)))
 		return -IPSET_ERR_PROTOCOL;
 
+	/* Must wait for flush to be really finished in list:set */
+	rcu_barrier();
+
 	/* Commands are serialized and references are
 	 * protected by the ip_set_ref_lock.
 	 * External systems (i.e. xt_set) must call
diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c
index bbede95..24c6c19 100644
--- a/net/netfilter/ipset/ip_set_list_set.c
+++ b/net/netfilter/ipset/ip_set_list_set.c
@@ -30,6 +30,7 @@ MODULE_ALIAS("ip_set_list:set");
 struct set_elem {
 	struct rcu_head rcu;
 	struct list_head list;
+	struct ip_set *set;	/* Sigh, in order to cleanup reference */
 	ip_set_id_t id;
 } __aligned(__alignof__(u64));
 
@@ -151,30 +152,29 @@ list_set_kadt(struct ip_set *set, const struct sk_buff *skb,
 /* Userspace interfaces: we are protected by the nfnl mutex */
 
 static void
-__list_set_del(struct ip_set *set, struct set_elem *e)
+__list_set_del_rcu(struct rcu_head * rcu)
 {
+	struct set_elem *e = container_of(rcu, struct set_elem, rcu);
+	struct ip_set *set = e->set;
 	struct list_set *map = set->data;
 
 	ip_set_put_byindex(map->net, e->id);
-	/* We may call it, because we don't have a to be destroyed
-	 * extension which is used by the kernel.
-	 */
 	ip_set_ext_destroy(set, e);
-	kfree_rcu(e, rcu);
+	kfree(e);
 }
 
 static inline void
 list_set_del(struct ip_set *set, struct set_elem *e)
 {
 	list_del_rcu(&e->list);
-	__list_set_del(set, e);
+	call_rcu(&e->rcu, __list_set_del_rcu);
 }
 
 static inline void
-list_set_replace(struct ip_set *set, struct set_elem *e, struct set_elem *old)
+list_set_replace(struct set_elem *e, struct set_elem *old)
 {
 	list_replace_rcu(&old->list, &e->list);
-	__list_set_del(set, old);
+	call_rcu(&old->rcu, __list_set_del_rcu);
 }
 
 static void
@@ -244,9 +244,6 @@ list_set_uadd(struct ip_set *set, void *value, const struct ip_set_ext *ext,
 	struct set_elem *e, *n, *prev, *next;
 	bool flag_exist = flags & IPSET_FLAG_EXIST;
 
-	if (SET_WITH_TIMEOUT(set))
-		set_cleanup_entries(set);
-
 	/* Find where to add the new entry */
 	n = prev = next = NULL;
 	list_for_each_entry(e, &map->members, list) {
@@ -301,10 +298,11 @@ list_set_uadd(struct ip_set *set, void *value, const struct ip_set_ext *ext,
 	if (!e)
 		return -ENOMEM;
 	e->id = d->id;
+	e->set = set;
 	INIT_LIST_HEAD(&e->list);
 	list_set_init_extensions(set, ext, e);
 	if (n)
-		list_set_replace(set, e, n);
+		list_set_replace(e, n);
 	else if (next)
 		list_add_tail_rcu(&e->list, &next->list);
 	else if (prev)
@@ -431,6 +429,7 @@ list_set_destroy(struct ip_set *set)
 
 	if (SET_WITH_TIMEOUT(set))
 		del_timer_sync(&map->gc);
+
 	list_for_each_entry_safe(e, n, &map->members, list) {
 		list_del(&e->list);
 		ip_set_put_byindex(map->net, e->id);
@@ -450,8 +449,10 @@ list_set_head(struct ip_set *set, struct sk_buff *skb)
 	struct set_elem *e;
 	u32 n = 0;
 
-	list_for_each_entry(e, &map->members, list)
+	rcu_read_lock();
+	list_for_each_entry_rcu(e, &map->members, list)
 		n++;
+	rcu_read_unlock();
 
 	nested = ipset_nest_start(skb, IPSET_ATTR_DATA);
 	if (!nested)
@@ -483,33 +484,25 @@ list_set_list(const struct ip_set *set,
 	atd = ipset_nest_start(skb, IPSET_ATTR_ADT);
 	if (!atd)
 		return -EMSGSIZE;
-	list_for_each_entry(e, &map->members, list) {
-		if (i == first)
-			break;
-		i++;
-	}
 
 	rcu_read_lock();
-	list_for_each_entry_from(e, &map->members, list) {
-		i++;
-		if (SET_WITH_TIMEOUT(set) &&
-		    ip_set_timeout_expired(ext_timeout(e, set)))
+	list_for_each_entry_rcu(e, &map->members, list) {
+		if (i < first ||
+		    (SET_WITH_TIMEOUT(set) &&
+		     ip_set_timeout_expired(ext_timeout(e, set)))) {
+			i++;
 			continue;
+		}
 		nested = ipset_nest_start(skb, IPSET_ATTR_DATA);
-		if (!nested) {
-			if (i == first) {
-				nla_nest_cancel(skb, atd);
-				ret = -EMSGSIZE;
-				goto out;
-			}
+		if (!nested)
 			goto nla_put_failure;
-		}
 		if (nla_put_string(skb, IPSET_ATTR_NAME,
 				   ip_set_name_byindex(map->net, e->id)))
 			goto nla_put_failure;
 		if (ip_set_put_extensions(skb, set, e, true))
 			goto nla_put_failure;
 		ipset_nest_end(skb, nested);
+		i++;
 	}
 
 	ipset_nest_end(skb, atd);
@@ -520,10 +513,12 @@ list_set_list(const struct ip_set *set,
 nla_put_failure:
 	nla_nest_cancel(skb, nested);
 	if (unlikely(i == first)) {
+		nla_nest_cancel(skb, atd);
 		cb->args[IPSET_CB_ARG0] = 0;
 		ret = -EMSGSIZE;
+	} else {
+		cb->args[IPSET_CB_ARG0] = i;
 	}
-	cb->args[IPSET_CB_ARG0] = i - 1;
 	ipset_nest_end(skb, atd);
 out:
 	rcu_read_unlock();
-- 
cgit v0.10.2


From b0fd2eb28bd47e0c906e31dbaccaf17883e238ff Mon Sep 17 00:00:00 2001
From: "ajit.khaparde@broadcom.com" <ajit.khaparde@broadcom.com>
Date: Tue, 23 Feb 2016 00:33:48 +0530
Subject: be2net: Declare some u16 fields as u32 to improve performance

When 16-bit integers are loaded on CPUs with high order native
register sizes, the CPU could use some extra ops before using them.
And currently some of the frequently used fields in the driver like
the producer and consumer indices of the queues are declared as u16.

This patch declares such fields as u32.  With this change we see the
64-byte packets per second numbers improve by about 4%.

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index ab24f84..756f25b 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -124,27 +124,27 @@ struct be_dma_mem {
 };
 
 struct be_queue_info {
+	u32 len;
+	u32 entry_size;	/* Size of an element in the queue */
+	u32 tail, head;
+	atomic_t used;	/* Number of valid elements in the queue */
+	u32 id;
 	struct be_dma_mem dma_mem;
-	u16 len;
-	u16 entry_size;	/* Size of an element in the queue */
-	u16 id;
-	u16 tail, head;
 	bool created;
-	atomic_t used;	/* Number of valid elements in the queue */
 };
 
-static inline u32 MODULO(u16 val, u16 limit)
+static inline u32 MODULO(u32 val, u32 limit)
 {
 	BUG_ON(limit & (limit - 1));
 	return val & (limit - 1);
 }
 
-static inline void index_adv(u16 *index, u16 val, u16 limit)
+static inline void index_adv(u32 *index, u32 val, u32 limit)
 {
 	*index = MODULO((*index + val), limit);
 }
 
-static inline void index_inc(u16 *index, u16 limit)
+static inline void index_inc(u32 *index, u32 limit)
 {
 	*index = MODULO((*index + 1), limit);
 }
@@ -169,7 +169,7 @@ static inline void queue_head_inc(struct be_queue_info *q)
 	index_inc(&q->head, q->len);
 }
 
-static inline void index_dec(u16 *index, u16 limit)
+static inline void index_dec(u32 *index, u32 limit)
 {
 	*index = MODULO((*index - 1), limit);
 }
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index 66fa214..22402db 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -596,7 +596,7 @@ static int be_mcc_notify_wait(struct be_adapter *adapter)
 	int status;
 	struct be_mcc_wrb *wrb;
 	struct be_mcc_obj *mcc_obj = &adapter->mcc_obj;
-	u16 index = mcc_obj->q.head;
+	u32 index = mcc_obj->q.head;
 	struct be_cmd_resp_hdr *resp;
 
 	index_dec(&index, mcc_obj->q.len);
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 4624846..0bd64f1 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -849,9 +849,9 @@ static void unmap_tx_frag(struct device *dev, struct be_eth_wrb *wrb,
 }
 
 /* Grab a WRB header for xmit */
-static u16 be_tx_get_wrb_hdr(struct be_tx_obj *txo)
+static u32 be_tx_get_wrb_hdr(struct be_tx_obj *txo)
 {
-	u16 head = txo->q.head;
+	u32 head = txo->q.head;
 
 	queue_head_inc(&txo->q);
 	return head;
@@ -895,7 +895,7 @@ static void be_tx_setup_wrb_frag(struct be_tx_obj *txo, dma_addr_t busaddr,
  * WRBs of the current packet are unmapped. Invoked to handle tx setup errors.
  */
 static void be_xmit_restore(struct be_adapter *adapter,
-			    struct be_tx_obj *txo, u16 head, bool map_single,
+			    struct be_tx_obj *txo, u32 head, bool map_single,
 			    u32 copied)
 {
 	struct device *dev;
@@ -930,7 +930,7 @@ static u32 be_xmit_enqueue(struct be_adapter *adapter, struct be_tx_obj *txo,
 	struct device *dev = &adapter->pdev->dev;
 	struct be_queue_info *txq = &txo->q;
 	bool map_single = false;
-	u16 head = txq->head;
+	u32 head = txq->head;
 	dma_addr_t busaddr;
 	int len;
 
@@ -1990,7 +1990,7 @@ static struct be_rx_page_info *get_rx_page_info(struct be_rx_obj *rxo)
 	struct be_adapter *adapter = rxo->adapter;
 	struct be_rx_page_info *rx_page_info;
 	struct be_queue_info *rxq = &rxo->q;
-	u16 frag_idx = rxq->tail;
+	u32 frag_idx = rxq->tail;
 
 	rx_page_info = &rxo->page_info_tbl[frag_idx];
 	BUG_ON(!rx_page_info->page);
@@ -2401,10 +2401,11 @@ static u16 be_tx_compl_process(struct be_adapter *adapter,
 {
 	struct sk_buff **sent_skbs = txo->sent_skb_list;
 	struct be_queue_info *txq = &txo->q;
-	u16 frag_index, num_wrbs = 0;
 	struct sk_buff *skb = NULL;
 	bool unmap_skb_hdr = false;
 	struct be_eth_wrb *wrb;
+	u16 num_wrbs = 0;
+	u32 frag_index;
 
 	do {
 		if (sent_skbs[txq->tail]) {
@@ -2516,10 +2517,11 @@ static void be_rx_cq_clean(struct be_rx_obj *rxo)
 
 static void be_tx_compl_clean(struct be_adapter *adapter)
 {
-	u16 end_idx, notified_idx, cmpl = 0, timeo = 0, num_wrbs = 0;
 	struct device *dev = &adapter->pdev->dev;
+	u16 cmpl = 0, timeo = 0, num_wrbs = 0;
 	struct be_tx_compl_info *txcp;
 	struct be_queue_info *txq;
+	u32 end_idx, notified_idx;
 	struct be_tx_obj *txo;
 	int i, pending_txqs;
 
-- 
cgit v0.10.2


From 127bfce54197a6cd43205b7ba950547918128053 Mon Sep 17 00:00:00 2001
From: "ajit.khaparde@broadcom.com" <ajit.khaparde@broadcom.com>
Date: Tue, 23 Feb 2016 00:35:01 +0530
Subject: be2net: Fix a UE caused by passing large frames to the ASIC

In QnQ configurations like Flex-10 where the VLANs are inserted by the
ASIC, on rare occasions the HW is encountering a scenario where the
final frame length ends to be greater than what the ASIC can support.

This is because when the TXULP pulls the TX WRB to check the length
of the frame to be transmitted it also adds the size of VLANs to be
inserted by the HW to the length of the frame indicated in the WRB,
which in some cases fails the range check.  This causes a UE.

Avoid this by trimming the skb length to accommodate the VLAN insertion.

Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index 756f25b..ee584c5 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -72,6 +72,9 @@
 #define BE_MAX_MTU              (BE_MAX_JUMBO_FRAME_SIZE -	\
 				 (ETH_HLEN + ETH_FCS_LEN))
 
+/* Accommodate for QnQ configurations where VLAN insertion is enabled in HW */
+#define BE_MAX_GSO_SIZE		(65535 - 2 * VLAN_HLEN)
+
 #define BE_NUM_VLANS_SUPPORTED	64
 #define BE_MAX_EQD		128u
 #define	BE_MAX_TX_FRAG_COUNT	30
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 0bd64f1..17422b2 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -1123,6 +1123,8 @@ static struct sk_buff *be_xmit_workarounds(struct be_adapter *adapter,
 					   struct sk_buff *skb,
 					   struct be_wrb_params *wrb_params)
 {
+	int err;
+
 	/* Lancer, SH and BE3 in SRIOV mode have a bug wherein
 	 * packets that are 32b or less may cause a transmit stall
 	 * on that port. The workaround is to pad such packets
@@ -1139,6 +1141,13 @@ static struct sk_buff *be_xmit_workarounds(struct be_adapter *adapter,
 			return NULL;
 	}
 
+	/* The stack can send us skbs with length greater than
+	 * what the HW can handle. Trim the extra bytes.
+	 */
+	WARN_ON_ONCE(skb->len > BE_MAX_GSO_SIZE);
+	err = pskb_trim(skb, BE_MAX_GSO_SIZE);
+	WARN_ON(err);
+
 	return skb;
 }
 
@@ -4850,7 +4859,7 @@ static void be_netdev_init(struct net_device *netdev)
 
 	netdev->flags |= IFF_MULTICAST;
 
-	netif_set_gso_max_size(netdev, 65535 - ETH_HLEN);
+	netif_set_gso_max_size(netdev, BE_MAX_GSO_SIZE - ETH_HLEN);
 
 	netdev->netdev_ops = &be_netdev_ops;
 
-- 
cgit v0.10.2


From 9e4e6206c67ae11d68fc96882256f37c237087d4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Mon, 22 Feb 2016 22:51:13 +0100
Subject: bgmac: support Ethernet device on BCM47094 SoC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It needs very similar workarounds to the one on BCM4707. It was tested
on D-Link DIR-885L home router.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index 230f8e6..99b30a9 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -30,6 +30,7 @@ static inline bool bgmac_is_bcm4707_family(struct bgmac *bgmac)
 {
 	switch (bgmac->core->bus->chipinfo.id) {
 	case BCMA_CHIP_ID_BCM4707:
+	case BCMA_CHIP_ID_BCM47094:
 	case BCMA_CHIP_ID_BCM53018:
 		return true;
 	default:
@@ -1052,8 +1053,9 @@ static void bgmac_chip_reset(struct bgmac *bgmac)
 	    (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg == BCMA_PKG_ID_BCM47188))
 		iost &= ~BGMAC_BCMA_IOST_ATTACHED;
 
-	/* 3GMAC: for BCM4707, only do core reset at bgmac_probe() */
-	if (ci->id != BCMA_CHIP_ID_BCM4707) {
+	/* 3GMAC: for BCM4707 & BCM47094, only do core reset at bgmac_probe() */
+	if (ci->id != BCMA_CHIP_ID_BCM4707 &&
+	    ci->id != BCMA_CHIP_ID_BCM47094) {
 		flags = 0;
 		if (iost & BGMAC_BCMA_IOST_ATTACHED) {
 			flags = BGMAC_BCMA_IOCTL_SW_CLKEN;
-- 
cgit v0.10.2


From e19d0839e75718a0b06db66f6f7ec2df515ae0c2 Mon Sep 17 00:00:00 2001
From: Claudiu Manoil <claudiu.manoil@freescale.com>
Date: Tue, 23 Feb 2016 11:48:37 +0200
Subject: gianfar: Map head TxBD first

Move the mapping of the head BD before the mapping of fragments.
The TOE (h/w offload) decision logic block can be also moved up
(as the TOE flag belongs to the head BD), resulting in more
localized code (TOE logic vs BD mapping code blocks).
Note that, for this h/w, the R (status) bit for the head BD of a S/G
frame needs to be written last for a reliable transmission.
For the fragmented skb case, a local variable is used to temporarily
store the status info of the first BD, replacing a BD status read.
A merge of 2 "if(do_tstamp)" blocks was also possible.

Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 2aa7b40..cb80dba8 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -2389,6 +2389,47 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	txbdp = txbdp_start = tx_queue->cur_tx;
 	lstatus = be32_to_cpu(txbdp->lstatus);
 
+	/* Add TxPAL between FCB and frame if required */
+	if (unlikely(do_tstamp)) {
+		skb_push(skb, GMAC_TXPAL_LEN);
+		memset(skb->data, 0, GMAC_TXPAL_LEN);
+	}
+
+	/* Add TxFCB if required */
+	if (fcb_len) {
+		fcb = gfar_add_fcb(skb);
+		lstatus |= BD_LFLAG(TXBD_TOE);
+	}
+
+	/* Set up checksumming */
+	if (do_csum) {
+		gfar_tx_checksum(skb, fcb, fcb_len);
+
+		if (unlikely(gfar_csum_errata_12(priv, (unsigned long)fcb)) ||
+		    unlikely(gfar_csum_errata_76(priv, skb->len))) {
+			__skb_pull(skb, GMAC_FCB_LEN);
+			skb_checksum_help(skb);
+			if (do_vlan || do_tstamp) {
+				/* put back a new fcb for vlan/tstamp TOE */
+				fcb = gfar_add_fcb(skb);
+			} else {
+				/* Tx TOE not used */
+				lstatus &= ~(BD_LFLAG(TXBD_TOE));
+				fcb = NULL;
+			}
+		}
+	}
+
+	if (do_vlan)
+		gfar_tx_vlan(skb, fcb);
+
+	bufaddr = dma_map_single(priv->dev, skb->data, skb_headlen(skb),
+				 DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(priv->dev, bufaddr)))
+		goto dma_map_err;
+
+	txbdp_start->bufPtr = cpu_to_be32(bufaddr);
+
 	/* Time stamp insertion requires one additional TxBD */
 	if (unlikely(do_tstamp))
 		txbdp_tstamp = txbdp = next_txbd(txbdp, base,
@@ -2404,6 +2445,8 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			lstatus |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT);
 		}
 	} else {
+		u32 lstatus_start = lstatus;
+
 		/* Place the fragment addresses and lengths into the TxBDs */
 		for (i = 0; i < nr_frags; i++) {
 			unsigned int frag_len;
@@ -2432,56 +2475,9 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			txbdp->lstatus = cpu_to_be32(lstatus);
 		}
 
-		lstatus = be32_to_cpu(txbdp_start->lstatus);
+		lstatus = lstatus_start;
 	}
 
-	/* Add TxPAL between FCB and frame if required */
-	if (unlikely(do_tstamp)) {
-		skb_push(skb, GMAC_TXPAL_LEN);
-		memset(skb->data, 0, GMAC_TXPAL_LEN);
-	}
-
-	/* Add TxFCB if required */
-	if (fcb_len) {
-		fcb = gfar_add_fcb(skb);
-		lstatus |= BD_LFLAG(TXBD_TOE);
-	}
-
-	/* Set up checksumming */
-	if (do_csum) {
-		gfar_tx_checksum(skb, fcb, fcb_len);
-
-		if (unlikely(gfar_csum_errata_12(priv, (unsigned long)fcb)) ||
-		    unlikely(gfar_csum_errata_76(priv, skb->len))) {
-			__skb_pull(skb, GMAC_FCB_LEN);
-			skb_checksum_help(skb);
-			if (do_vlan || do_tstamp) {
-				/* put back a new fcb for vlan/tstamp TOE */
-				fcb = gfar_add_fcb(skb);
-			} else {
-				/* Tx TOE not used */
-				lstatus &= ~(BD_LFLAG(TXBD_TOE));
-				fcb = NULL;
-			}
-		}
-	}
-
-	if (do_vlan)
-		gfar_tx_vlan(skb, fcb);
-
-	/* Setup tx hardware time stamping if requested */
-	if (unlikely(do_tstamp)) {
-		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-		fcb->ptp = 1;
-	}
-
-	bufaddr = dma_map_single(priv->dev, skb->data, skb_headlen(skb),
-				 DMA_TO_DEVICE);
-	if (unlikely(dma_mapping_error(priv->dev, bufaddr)))
-		goto dma_map_err;
-
-	txbdp_start->bufPtr = cpu_to_be32(bufaddr);
-
 	/* If time stamping is requested one additional TxBD must be set up. The
 	 * first TxBD points to the FCB and must have a data length of
 	 * GMAC_FCB_LEN. The second TxBD points to the actual frame data with
@@ -2498,6 +2494,10 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		txbdp_tstamp->bufPtr = cpu_to_be32(bufaddr);
 		txbdp_tstamp->lstatus = cpu_to_be32(lstatus_ts);
 		lstatus |= BD_LFLAG(TXBD_CRC | TXBD_READY) | GMAC_FCB_LEN;
+
+		/* Setup tx hardware time stamping */
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+		fcb->ptp = 1;
 	} else {
 		lstatus |= BD_LFLAG(TXBD_CRC | TXBD_READY) | skb_headlen(skb);
 	}
-- 
cgit v0.10.2


From 42f397adfc0a062a17119c08c82ed710b143a006 Mon Sep 17 00:00:00 2001
From: Claudiu Manoil <claudiu.manoil@freescale.com>
Date: Tue, 23 Feb 2016 11:48:38 +0200
Subject: gianfar: Use skb_frag_t pointers inside xmit()

Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index cb80dba8..7b16ce6 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -2322,6 +2322,7 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct txfcb *fcb = NULL;
 	struct txbd8 *txbdp, *txbdp_start, *base, *txbdp_tstamp = NULL;
 	u32 lstatus;
+	skb_frag_t *frag;
 	int i, rq = 0;
 	int do_tstamp, do_csum, do_vlan;
 	u32 bufaddr;
@@ -2448,25 +2449,24 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		u32 lstatus_start = lstatus;
 
 		/* Place the fragment addresses and lengths into the TxBDs */
-		for (i = 0; i < nr_frags; i++) {
-			unsigned int frag_len;
+		frag = &skb_shinfo(skb)->frags[0];
+		for (i = 0; i < nr_frags; i++, frag++) {
+			unsigned int size;
+
 			/* Point at the next BD, wrapping as needed */
 			txbdp = next_txbd(txbdp, base, tx_queue->tx_ring_size);
 
-			frag_len = skb_shinfo(skb)->frags[i].size;
+			size = skb_frag_size(frag);
 
-			lstatus = be32_to_cpu(txbdp->lstatus) | frag_len |
+			lstatus = be32_to_cpu(txbdp->lstatus) | size |
 				  BD_LFLAG(TXBD_READY);
 
 			/* Handle the last BD specially */
 			if (i == nr_frags - 1)
 				lstatus |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT);
 
-			bufaddr = skb_frag_dma_map(priv->dev,
-						   &skb_shinfo(skb)->frags[i],
-						   0,
-						   frag_len,
-						   DMA_TO_DEVICE);
+			bufaddr = skb_frag_dma_map(priv->dev, frag, 0,
+						   size, DMA_TO_DEVICE);
 			if (unlikely(dma_mapping_error(priv->dev, bufaddr)))
 				goto dma_map_err;
 
-- 
cgit v0.10.2


From 48963b4492e98fd3b3040755a16215d72552f1bb Mon Sep 17 00:00:00 2001
From: Claudiu Manoil <claudiu.manoil@freescale.com>
Date: Tue, 23 Feb 2016 11:48:39 +0200
Subject: gianfar: Remove redundant ops for do_tstamp from xmit()

Timestamp BD status updates that can be merged into the
same "do_tstamp" block, no need for extra save/restore
to the BD area.  The code is more readable too.

Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 7b16ce6..1e1157f 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -2436,15 +2436,8 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		txbdp_tstamp = txbdp = next_txbd(txbdp, base,
 						 tx_queue->tx_ring_size);
 
-	if (nr_frags == 0) {
-		if (unlikely(do_tstamp)) {
-			u32 lstatus_ts = be32_to_cpu(txbdp_tstamp->lstatus);
-
-			lstatus_ts |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT);
-			txbdp_tstamp->lstatus = cpu_to_be32(lstatus_ts);
-		} else {
-			lstatus |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT);
-		}
+	if (likely(!nr_frags)) {
+		lstatus |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT);
 	} else {
 		u32 lstatus_start = lstatus;
 
@@ -2488,8 +2481,11 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 		bufaddr = be32_to_cpu(txbdp_start->bufPtr);
 		bufaddr += fcb_len;
+
 		lstatus_ts |= BD_LFLAG(TXBD_READY) |
 			      (skb_headlen(skb) - fcb_len);
+		if (!nr_frags)
+			lstatus_ts |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT);
 
 		txbdp_tstamp->bufPtr = cpu_to_be32(bufaddr);
 		txbdp_tstamp->lstatus = cpu_to_be32(lstatus_ts);
-- 
cgit v0.10.2


From 6721e9d568741ced04b1fe6eed42f2ddf585eac4 Mon Sep 17 00:00:00 2001
From: Raanan Avargil <raanan.avargil@intel.com>
Date: Tue, 22 Dec 2015 15:35:01 +0200
Subject: e1000e: Increase ULP timer

Due to system level changes introduced in Skylake, ULP exit takes
significantly longer to occur.  Therefore, driver must wait longer for.

Signed-off-by: Raanan Avargil <raanan.avargil@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index a049e30..c731465 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -1252,9 +1252,9 @@ static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force)
 			ew32(H2ME, mac_reg);
 		}
 
-		/* Poll up to 100msec for ME to clear ULP_CFG_DONE */
+		/* Poll up to 300msec for ME to clear ULP_CFG_DONE. */
 		while (er32(FWSM) & E1000_FWSM_ULP_CFG_DONE) {
-			if (i++ == 10) {
+			if (i++ == 30) {
 				ret_val = -E1000_ERR_PHY;
 				goto out;
 			}
-- 
cgit v0.10.2


From 74f31299a41e729226d60426087592b6790f22b7 Mon Sep 17 00:00:00 2001
From: Raanan Avargil <raanan.avargil@intel.com>
Date: Tue, 22 Dec 2015 15:35:02 +0200
Subject: e1000e: Increase PHY PLL clock gate timing

Several packet loss issues were reported for which the root cause for
them was an incorrect configuration of internal HW PHY clock gating
mechanism by SW.
This patch provides the correct mechanism.

Signed-off-by: Raanan Avargil <raanan.avargil@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index c731465..786d214 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -1433,6 +1433,18 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 			emi_addr = I217_RX_CONFIG;
 		ret_val = e1000_write_emi_reg_locked(hw, emi_addr, emi_val);
 
+		if (hw->mac.type == e1000_pch_lpt ||
+		    hw->mac.type == e1000_pch_spt) {
+			u16 phy_reg;
+
+			e1e_rphy_locked(hw, I217_PLL_CLOCK_GATE_REG, &phy_reg);
+			phy_reg &= ~I217_PLL_CLOCK_GATE_MASK;
+			if (speed == SPEED_100 || speed == SPEED_10)
+				phy_reg |= 0x3E8;
+			else
+				phy_reg |= 0xFA;
+			e1e_wphy_locked(hw, I217_PLL_CLOCK_GATE_REG, phy_reg);
+		}
 		hw->phy.ops.release(hw);
 
 		if (ret_val)
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.h b/drivers/net/ethernet/intel/e1000e/ich8lan.h
index 34c551e..7d85f00 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.h
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.h
@@ -226,6 +226,9 @@
 #define HV_PM_CTRL_PLL_STOP_IN_K1_GIGA	0x100
 #define HV_PM_CTRL_K1_ENABLE		0x4000
 
+#define I217_PLL_CLOCK_GATE_REG	PHY_REG(772, 28)
+#define I217_PLL_CLOCK_GATE_MASK	0x07FF
+
 #define SW_FLAG_TIMEOUT		1000	/* SW Semaphore flag timeout in ms */
 
 /* Inband Control */
-- 
cgit v0.10.2


From c26f40daf4e32f970b8337a88b65a8d00332ae6f Mon Sep 17 00:00:00 2001
From: Raanan Avargil <raanan.avargil@intel.com>
Date: Tue, 22 Dec 2015 15:35:03 +0200
Subject: e1000e: Set HW FIFO minimum pointer gap for non-gig speeds

Based on feedback from HW team, the configured value of the internal PHY
HW FIFO pointer gap was incorrect for non-gig speeds.
This patch provides the correct configuration.

Signed-off-by: Raanan Avargil <raanan.avargil@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index 786d214..e7ccf5f 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -1479,6 +1479,18 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw)
 				hw->phy.ops.release(hw);
 				if (ret_val)
 					return ret_val;
+			} else {
+				ret_val = hw->phy.ops.acquire(hw);
+				if (ret_val)
+					return ret_val;
+
+				ret_val = e1e_wphy_locked(hw,
+							  PHY_REG(776, 20),
+							  0xC023);
+				hw->phy.ops.release(hw);
+				if (ret_val)
+					return ret_val;
+
 			}
 		}
 	}
-- 
cgit v0.10.2


From c5c6d07761a9ff64f0ffff2ca410a578fb7c4579 Mon Sep 17 00:00:00 2001
From: Raanan Avargil <raanan.avargil@intel.com>
Date: Tue, 22 Dec 2015 15:35:04 +0200
Subject: e1000e: Clear ULP configuration register on ULP exit

There have been bugs caused by HW ULP configuration settings not being
properly cleared after cable connect in V-Pro capable systems.
This caused HW to get out of sync occasionally.
The fix ensures that ULP settings are cleared in HW after
LAN cable re-connect.

Signed-off-by: Raanan Avargil <raanan.avargil@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index e7ccf5f..c0f4887 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -1328,6 +1328,8 @@ static s32 e1000_disable_ulp_lpt_lp(struct e1000_hw *hw, bool force)
 		     I218_ULP_CONFIG1_RESET_TO_SMBUS |
 		     I218_ULP_CONFIG1_WOL_HOST |
 		     I218_ULP_CONFIG1_INBAND_EXIT |
+		     I218_ULP_CONFIG1_EN_ULP_LANPHYPC |
+		     I218_ULP_CONFIG1_DIS_CLR_STICKY_ON_PERST |
 		     I218_ULP_CONFIG1_DISABLE_SMB_PERST);
 	e1000_write_phy_reg_hv_locked(hw, I218_ULP_CONFIG1, phy_reg);
 
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.h b/drivers/net/ethernet/intel/e1000e/ich8lan.h
index 7d85f00..2311f60 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.h
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.h
@@ -188,6 +188,10 @@
 #define I218_ULP_CONFIG1_INBAND_EXIT	0x0020	/* Inband on ULP exit */
 #define I218_ULP_CONFIG1_WOL_HOST	0x0040	/* WoL Host on ULP exit */
 #define I218_ULP_CONFIG1_RESET_TO_SMBUS	0x0100	/* Reset to SMBus mode */
+/* enable ULP even if when phy powered down via lanphypc */
+#define I218_ULP_CONFIG1_EN_ULP_LANPHYPC	0x0400
+/* disable clear of sticky ULP on PERST */
+#define I218_ULP_CONFIG1_DIS_CLR_STICKY_ON_PERST	0x0800
 #define I218_ULP_CONFIG1_DISABLE_SMB_PERST	0x1000	/* Disable on PERST# */
 
 /* SMBus Address Phy Register */
-- 
cgit v0.10.2


From 9cd34b3a1cfd47692cbef8cb0761475021883e18 Mon Sep 17 00:00:00 2001
From: Raanan Avargil <raanan.avargil@intel.com>
Date: Tue, 22 Dec 2015 15:35:05 +0200
Subject: e1000e: Initial support for KabeLake

i219 (4) and i219 (5) are the next LOM generations that will be
available on the next Intel platform (KabeLake).
This patch provides the initial support for the devices.

Signed-off-by: Raanan Avargil <raanan.avargil@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index b3949d5..4e733bf 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -92,6 +92,10 @@ struct e1000_hw;
 #define E1000_DEV_ID_PCH_SPT_I219_LM2		0x15B7	/* SPT-H PCH */
 #define E1000_DEV_ID_PCH_SPT_I219_V2		0x15B8	/* SPT-H PCH */
 #define E1000_DEV_ID_PCH_LBG_I219_LM3		0x15B9	/* LBG PCH */
+#define E1000_DEV_ID_PCH_SPT_I219_LM4		0x15D7
+#define E1000_DEV_ID_PCH_SPT_I219_V4		0x15D8
+#define E1000_DEV_ID_PCH_SPT_I219_LM5		0x15E3
+#define E1000_DEV_ID_PCH_SPT_I219_V5		0x15D6
 
 #define E1000_REVISION_4	4
 
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index c71ba1b..9b4ec13 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -7452,6 +7452,10 @@ static const struct pci_device_id e1000_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM2), board_pch_spt },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V2), board_pch_spt },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LBG_I219_LM3), board_pch_spt },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM4), board_pch_spt },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V4), board_pch_spt },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM5), board_pch_spt },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V5), board_pch_spt },
 
 	{ 0, 0, 0, 0, 0, 0, 0 }	/* terminate list */
 };
-- 
cgit v0.10.2


From b72f3f72005dfd649d787535bd04ada3b3f1b3ba Mon Sep 17 00:00:00 2001
From: Takuma Ueba <t.ueba11@gmail.com>
Date: Thu, 31 Dec 2015 14:58:14 +0900
Subject: igb: When GbE link up, wait for Remote receiver status condition

I210 device IPv6 autoconf test sometimes fails,
because DAD NS for link-local is not transmitted.
This packet is silently dropped.
This problem is seen only GbE environment.

igb_watchdog_task link up detection continues to the following process.
The following cases are observed:
1.PHY 1000BASE-T Status Register Remote receiver status bit is NG.
(NG status becomes OK after about 200 - 700ms)
2.In this case, the transfer packet is silently dropped.

1000BASE-T Status register
[Expected]: 0x3800 or 0x7800
[problem occurred]: 0x2800 or 0x6800
Frequency of occurrence: approx 1/10 - 1/40 observed

In order to avoid this problem,
wait until 1000BASE-T Status register "Remote receiver status OK"

After applying this patch, at least 400 runs succeed with no problems.

Signed-off-by: Takuma Ueba <t.ueba11@gmail.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index af46fcf..1aa0180 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -4357,6 +4357,7 @@ static void igb_watchdog_task(struct work_struct *work)
 	u32 link;
 	int i;
 	u32 connsw;
+	u16 phy_data, retry_count = 20;
 
 	link = igb_has_link(adapter);
 
@@ -4435,6 +4436,25 @@ static void igb_watchdog_task(struct work_struct *work)
 				break;
 			}
 
+			if (adapter->link_speed != SPEED_1000)
+				goto no_wait;
+
+			/* wait for Remote receiver status OK */
+retry_read_status:
+			if (!igb_read_phy_reg(hw, PHY_1000T_STATUS,
+					      &phy_data)) {
+				if (!(phy_data & SR_1000T_REMOTE_RX_STATUS) &&
+				    retry_count) {
+					msleep(100);
+					retry_count--;
+					goto retry_read_status;
+				} else if (!retry_count) {
+					dev_err(&adapter->pdev->dev, "exceed max 2 second\n");
+				}
+			} else {
+				dev_err(&adapter->pdev->dev, "read 1000Base-T Status Reg\n");
+			}
+no_wait:
 			netif_carrier_on(netdev);
 
 			igb_ping_all_vfs(adapter);
-- 
cgit v0.10.2


From 5b70e4a12a525b5f3d4a3e3f0567ed877195b187 Mon Sep 17 00:00:00 2001
From: Julia Lawall <julia.lawall@lip6.fr>
Date: Sun, 3 Jan 2016 07:44:56 +0100
Subject: igb: constify e1000_phy_operations structure

This e1000_phy_operations structure is never modified, so declare it as
const.  Other structures of this type are already const.

Done with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.c b/drivers/net/ethernet/intel/igb/e1000_82575.c
index 9a1a9c7..a23aa67 100644
--- a/drivers/net/ethernet/intel/igb/e1000_82575.c
+++ b/drivers/net/ethernet/intel/igb/e1000_82575.c
@@ -2920,7 +2920,7 @@ static struct e1000_mac_operations e1000_mac_ops_82575 = {
 #endif
 };
 
-static struct e1000_phy_operations e1000_phy_ops_82575 = {
+static const struct e1000_phy_operations e1000_phy_ops_82575 = {
 	.acquire              = igb_acquire_phy_82575,
 	.get_cfg_done         = igb_get_cfg_done_82575,
 	.release              = igb_release_phy_82575,
diff --git a/drivers/net/ethernet/intel/igb/e1000_hw.h b/drivers/net/ethernet/intel/igb/e1000_hw.h
index f0c416e..2fb2213 100644
--- a/drivers/net/ethernet/intel/igb/e1000_hw.h
+++ b/drivers/net/ethernet/intel/igb/e1000_hw.h
@@ -372,7 +372,7 @@ struct e1000_thermal_sensor_data {
 struct e1000_info {
 	s32 (*get_invariants)(struct e1000_hw *);
 	struct e1000_mac_operations *mac_ops;
-	struct e1000_phy_operations *phy_ops;
+	const struct e1000_phy_operations *phy_ops;
 	struct e1000_nvm_operations *nvm_ops;
 };
 
-- 
cgit v0.10.2


From 5e350b9260a2e94a9dd1b20fb720d855d5bf1034 Mon Sep 17 00:00:00 2001
From: Todd Fujinaka <todd.fujinaka@intel.com>
Date: Tue, 5 Jan 2016 10:08:28 -0800
Subject: igb: enable WoL for OEM devices regardless of EEPROM setting

Override EEPROM settings for specific OEM devices.

Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 1aa0180..4238f3e 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2538,6 +2538,26 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		adapter->wol = 0;
 	}
 
+	/* Some vendors want the ability to Use the EEPROM setting as
+	 * enable/disable only, and not for capability
+	 */
+	if (((hw->mac.type == e1000_i350) ||
+	     (hw->mac.type == e1000_i354)) &&
+	    (pdev->subsystem_vendor == PCI_VENDOR_ID_DELL)) {
+		adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
+		adapter->wol = 0;
+	}
+	if (hw->mac.type == e1000_i350) {
+		if (((pdev->subsystem_device == 0x5001) ||
+		     (pdev->subsystem_device == 0x5002)) &&
+				(hw->bus.func == 0)) {
+			adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
+			adapter->wol = 0;
+		}
+		if (pdev->subsystem_device == 0x1F52)
+			adapter->flags |= IGB_FLAG_WOL_SUPPORTED;
+	}
+
 	device_set_wakeup_enable(&adapter->pdev->dev,
 				 adapter->flags & IGB_FLAG_WOL_SUPPORTED);
 
-- 
cgit v0.10.2


From 569f3b3d4e9898ae30788fde128e3277d996710e Mon Sep 17 00:00:00 2001
From: Roland Hii <roland.king.guan.hii@intel.com>
Date: Mon, 11 Jan 2016 15:34:18 +0800
Subject: igb: add conditions for I210 to generate periodic clock output

In general case the maximum supported half cycle time of the synchronized
output clock is 70msec. Slower half cycle time than 70msec can be
programmed also as long as the output clock is synchronized to whole
seconds, useful specifically for generating a 1Hz clock.

Permitted values for the clock half cycle time are: 125,000,000 decimal,
250,000,000 decimal and 500,000,000 decimal (equals to 125msec, 250msec
and 500msec respectively).

Before this patch, only the half cycle time of less than or equal to 70msec
uses the I210 clock output function. This patch adds additional conditions
when half cycle time is equal to 125msec or 250msec or 500msec to use
clock output function.

Under other conditions, interrupt driven target time output events method
is still used to generate the desired clock output.

Signed-off-by: Roland Hii <roland.king.guan.hii@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index c44df87..22a8a29 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -525,7 +525,8 @@ static int igb_ptp_feature_enable_i210(struct ptp_clock_info *ptp,
 		ts.tv_nsec = rq->perout.period.nsec;
 		ns = timespec64_to_ns(&ts);
 		ns = ns >> 1;
-		if (on && ns <= 70000000LL) {
+		if (on && ((ns <= 70000000LL) || (ns == 125000000LL) ||
+			   (ns == 250000000LL) || (ns == 500000000LL))) {
 			if (ns < 8LL)
 				return -EINVAL;
 			use_freq = 1;
-- 
cgit v0.10.2


From c883de9fd787b6f49bf825f3de3601aeb78a7114 Mon Sep 17 00:00:00 2001
From: Todd Fujinaka <todd.fujinaka@intel.com>
Date: Mon, 11 Jan 2016 09:34:50 -0800
Subject: igb: rename igb define to be more generic

E1000_MRQC_ENABLE_RSS_4Q enables 4 and 8 queues depending on the part
so rename to be generic.

Similarly, E1000_MRQC_ENABLE_VMDQ_RSS_2Q has no numeric meaning so
rename to be more generic.

Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.h b/drivers/net/ethernet/intel/igb/e1000_82575.h
index 2154aea..de8805a 100644
--- a/drivers/net/ethernet/intel/igb/e1000_82575.h
+++ b/drivers/net/ethernet/intel/igb/e1000_82575.h
@@ -56,10 +56,10 @@ s32 igb_write_i2c_byte(struct e1000_hw *hw, u8 byte_offset, u8 dev_addr,
 #define E1000_SRRCTL_TIMESTAMP                          0x40000000
 
 
-#define E1000_MRQC_ENABLE_RSS_4Q            0x00000002
+#define E1000_MRQC_ENABLE_RSS_MQ            0x00000002
 #define E1000_MRQC_ENABLE_VMDQ              0x00000003
 #define E1000_MRQC_RSS_FIELD_IPV4_UDP       0x00400000
-#define E1000_MRQC_ENABLE_VMDQ_RSS_2Q       0x00000005
+#define E1000_MRQC_ENABLE_VMDQ_RSS_MQ       0x00000005
 #define E1000_MRQC_RSS_FIELD_IPV6_UDP       0x00800000
 #define E1000_MRQC_RSS_FIELD_IPV6_UDP_EX    0x01000000
 
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 4238f3e..99acd84 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -3497,12 +3497,12 @@ static void igb_setup_mrqc(struct igb_adapter *adapter)
 			wr32(E1000_VT_CTL, vtctl);
 		}
 		if (adapter->rss_queues > 1)
-			mrqc |= E1000_MRQC_ENABLE_VMDQ_RSS_2Q;
+			mrqc |= E1000_MRQC_ENABLE_VMDQ_RSS_MQ;
 		else
 			mrqc |= E1000_MRQC_ENABLE_VMDQ;
 	} else {
 		if (hw->mac.type != e1000_i211)
-			mrqc |= E1000_MRQC_ENABLE_RSS_4Q;
+			mrqc |= E1000_MRQC_ENABLE_RSS_MQ;
 	}
 	igb_vmm_control(adapter);
 
-- 
cgit v0.10.2


From 6e033700887bf29d4e59f6978a02d989787be620 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 13 Jan 2016 07:31:23 -0800
Subject: igb: Add support for generic Tx checksums

This patch adds support for generic Tx checksums to the igb driver.  It
turns out this is actually pretty easy after going over the datasheet as we
were doing a number of steps we didn't need to.

In order to perform a Tx checksum for an L4 header we need to fill in the
following fields in the Tx descriptor:
  MACLEN (maximum of 127), retrieved from:
		skb_network_offset()
  IPLEN  (maximum of 511), retrieved from:
		skb_checksum_start_offset() - skb_network_offset()
  TUCMD.L4T indicates offset and if checksum or crc32c, based on:
		skb->csum_offset

The added advantage to doing this is that we can support inner checksum
offloads for tunnels and MPLS while still being able to transparently
insert VLAN tags.

I also took the opportunity to clean-up many of the feature flag
configuration bits to make them a bit more consistent between drivers.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 99acd84..d90dcd1 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2372,27 +2372,35 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * assignment.
 	 */
 	netdev->features |= NETIF_F_SG |
-			    NETIF_F_IP_CSUM |
-			    NETIF_F_IPV6_CSUM |
 			    NETIF_F_TSO |
 			    NETIF_F_TSO6 |
 			    NETIF_F_RXHASH |
 			    NETIF_F_RXCSUM |
+			    NETIF_F_HW_CSUM |
 			    NETIF_F_HW_VLAN_CTAG_RX |
 			    NETIF_F_HW_VLAN_CTAG_TX;
 
+	if (hw->mac.type >= e1000_82576)
+		netdev->features |= NETIF_F_SCTP_CRC;
+
 	/* copy netdev features into list of user selectable features */
 	netdev->hw_features |= netdev->features;
 	netdev->hw_features |= NETIF_F_RXALL;
 
+	if (hw->mac.type >= e1000_i350)
+		netdev->hw_features |= NETIF_F_NTUPLE;
+
 	/* set this bit last since it cannot be part of hw_features */
 	netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
 
-	netdev->vlan_features |= NETIF_F_TSO |
+	netdev->vlan_features |= NETIF_F_SG |
+				 NETIF_F_TSO |
 				 NETIF_F_TSO6 |
-				 NETIF_F_IP_CSUM |
-				 NETIF_F_IPV6_CSUM |
-				 NETIF_F_SG;
+				 NETIF_F_HW_CSUM |
+				 NETIF_F_SCTP_CRC;
+
+	netdev->mpls_features |= NETIF_F_HW_CSUM;
+	netdev->hw_enc_features |= NETIF_F_HW_CSUM;
 
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
 
@@ -2401,11 +2409,6 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		netdev->vlan_features |= NETIF_F_HIGHDMA;
 	}
 
-	if (hw->mac.type >= e1000_82576) {
-		netdev->hw_features |= NETIF_F_SCTP_CRC;
-		netdev->features |= NETIF_F_SCTP_CRC;
-	}
-
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
 	adapter->en_mng_pt = igb_enable_mng_pass_thru(hw);
@@ -4883,70 +4886,57 @@ static int igb_tso(struct igb_ring *tx_ring,
 	return 1;
 }
 
+static inline bool igb_ipv6_csum_is_sctp(struct sk_buff *skb)
+{
+	unsigned int offset = 0;
+
+	ipv6_find_hdr(skb, &offset, IPPROTO_SCTP, NULL, NULL);
+
+	return offset == skb_checksum_start_offset(skb);
+}
+
 static void igb_tx_csum(struct igb_ring *tx_ring, struct igb_tx_buffer *first)
 {
 	struct sk_buff *skb = first->skb;
 	u32 vlan_macip_lens = 0;
-	u32 mss_l4len_idx = 0;
 	u32 type_tucmd = 0;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL) {
+csum_failed:
 		if (!(first->tx_flags & IGB_TX_FLAGS_VLAN))
 			return;
-	} else {
-		u8 l4_hdr = 0;
-
-		switch (first->protocol) {
-		case htons(ETH_P_IP):
-			vlan_macip_lens |= skb_network_header_len(skb);
-			type_tucmd |= E1000_ADVTXD_TUCMD_IPV4;
-			l4_hdr = ip_hdr(skb)->protocol;
-			break;
-		case htons(ETH_P_IPV6):
-			vlan_macip_lens |= skb_network_header_len(skb);
-			l4_hdr = ipv6_hdr(skb)->nexthdr;
-			break;
-		default:
-			if (unlikely(net_ratelimit())) {
-				dev_warn(tx_ring->dev,
-					 "partial checksum but proto=%x!\n",
-					 first->protocol);
-			}
-			break;
-		}
+		goto no_csum;
+	}
 
-		switch (l4_hdr) {
-		case IPPROTO_TCP:
-			type_tucmd |= E1000_ADVTXD_TUCMD_L4T_TCP;
-			mss_l4len_idx = tcp_hdrlen(skb) <<
-					E1000_ADVTXD_L4LEN_SHIFT;
-			break;
-		case IPPROTO_SCTP:
-			type_tucmd |= E1000_ADVTXD_TUCMD_L4T_SCTP;
-			mss_l4len_idx = sizeof(struct sctphdr) <<
-					E1000_ADVTXD_L4LEN_SHIFT;
-			break;
-		case IPPROTO_UDP:
-			mss_l4len_idx = sizeof(struct udphdr) <<
-					E1000_ADVTXD_L4LEN_SHIFT;
-			break;
-		default:
-			if (unlikely(net_ratelimit())) {
-				dev_warn(tx_ring->dev,
-					 "partial checksum but l4 proto=%x!\n",
-					 l4_hdr);
-			}
+	switch (skb->csum_offset) {
+	case offsetof(struct tcphdr, check):
+		type_tucmd = E1000_ADVTXD_TUCMD_L4T_TCP;
+		/* fall through */
+	case offsetof(struct udphdr, check):
+		break;
+	case offsetof(struct sctphdr, checksum):
+		/* validate that this is actually an SCTP request */
+		if (((first->protocol == htons(ETH_P_IP)) &&
+		     (ip_hdr(skb)->protocol == IPPROTO_SCTP)) ||
+		    ((first->protocol == htons(ETH_P_IPV6)) &&
+		     igb_ipv6_csum_is_sctp(skb))) {
+			type_tucmd = E1000_ADVTXD_TUCMD_L4T_SCTP;
 			break;
 		}
-
-		/* update TX checksum flag */
-		first->tx_flags |= IGB_TX_FLAGS_CSUM;
+	default:
+		skb_checksum_help(skb);
+		goto csum_failed;
 	}
 
+	/* update TX checksum flag */
+	first->tx_flags |= IGB_TX_FLAGS_CSUM;
+	vlan_macip_lens = skb_checksum_start_offset(skb) -
+			  skb_network_offset(skb);
+no_csum:
 	vlan_macip_lens |= skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT;
 	vlan_macip_lens |= first->tx_flags & IGB_TX_FLAGS_VLAN_MASK;
 
-	igb_tx_ctxtdesc(tx_ring, vlan_macip_lens, type_tucmd, mss_l4len_idx);
+	igb_tx_ctxtdesc(tx_ring, vlan_macip_lens, type_tucmd, 0);
 }
 
 #define IGB_SET_FLAG(_input, _flag, _result) \
-- 
cgit v0.10.2


From ea6ce6024f9397ff2667fe16447447e622bc4c31 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 13 Jan 2016 07:31:30 -0800
Subject: igbvf: Add support for generic Tx checksums

This patch adds support for generic Tx checksums to the igbvf driver.  It
turns out this is actually pretty easy after going over the datasheet as we
were doing a number of steps we didn't need to.

In order to perform a Tx checksum for an L4 header we need to fill in the
following fields in the Tx descriptor:
  MACLEN (maximum of 127), retrieved from:
		skb_network_offset()
  IPLEN  (maximum of 511), retrieved from:
		skb_checksum_start_offset() - skb_network_offset()
  TUCMD.L4T indicates offset and if checksum or crc32c, based on:
		skb->csum_offset

The added advantage to doing this is that we can support inner checksum
offloads for tunnels and MPLS while still being able to transparently
insert VLAN tags.

I also took the opportunity to clean-up many of the feature flag
configuration bits to make them a bit more consistent between drivers.  In
the case of the VF drivers this meant adding support for SCTP CRCs, and
inner checksum offloads for MPLS and various tunnel types.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index 297af80..aa34865 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -43,6 +43,7 @@
 #include <linux/ethtool.h>
 #include <linux/if_vlan.h>
 #include <linux/prefetch.h>
+#include <linux/sctp.h>
 
 #include "igbvf.h"
 
@@ -1908,6 +1909,31 @@ static void igbvf_watchdog_task(struct work_struct *work)
 #define IGBVF_TX_FLAGS_VLAN_MASK	0xffff0000
 #define IGBVF_TX_FLAGS_VLAN_SHIFT	16
 
+static void igbvf_tx_ctxtdesc(struct igbvf_ring *tx_ring, u32 vlan_macip_lens,
+			      u32 type_tucmd, u32 mss_l4len_idx)
+{
+	struct e1000_adv_tx_context_desc *context_desc;
+	struct igbvf_buffer *buffer_info;
+	u16 i = tx_ring->next_to_use;
+
+	context_desc = IGBVF_TX_CTXTDESC_ADV(*tx_ring, i);
+	buffer_info = &tx_ring->buffer_info[i];
+
+	i++;
+	tx_ring->next_to_use = (i < tx_ring->count) ? i : 0;
+
+	/* set bits to identify this as an advanced context descriptor */
+	type_tucmd |= E1000_TXD_CMD_DEXT | E1000_ADVTXD_DTYP_CTXT;
+
+	context_desc->vlan_macip_lens	= cpu_to_le32(vlan_macip_lens);
+	context_desc->seqnum_seed	= 0;
+	context_desc->type_tucmd_mlhl	= cpu_to_le32(type_tucmd);
+	context_desc->mss_l4len_idx	= cpu_to_le32(mss_l4len_idx);
+
+	buffer_info->time_stamp = jiffies;
+	buffer_info->dma = 0;
+}
+
 static int igbvf_tso(struct igbvf_adapter *adapter,
 		     struct igbvf_ring *tx_ring,
 		     struct sk_buff *skb, u32 tx_flags, u8 *hdr_len,
@@ -1987,65 +2013,56 @@ static int igbvf_tso(struct igbvf_adapter *adapter,
 	return true;
 }
 
-static inline bool igbvf_tx_csum(struct igbvf_adapter *adapter,
-				 struct igbvf_ring *tx_ring,
-				 struct sk_buff *skb, u32 tx_flags,
-				 __be16 protocol)
+static inline bool igbvf_ipv6_csum_is_sctp(struct sk_buff *skb)
 {
-	struct e1000_adv_tx_context_desc *context_desc;
-	unsigned int i;
-	struct igbvf_buffer *buffer_info;
-	u32 info = 0, tu_cmd = 0;
-
-	if ((skb->ip_summed == CHECKSUM_PARTIAL) ||
-	    (tx_flags & IGBVF_TX_FLAGS_VLAN)) {
-		i = tx_ring->next_to_use;
-		buffer_info = &tx_ring->buffer_info[i];
-		context_desc = IGBVF_TX_CTXTDESC_ADV(*tx_ring, i);
+	unsigned int offset = 0;
 
-		if (tx_flags & IGBVF_TX_FLAGS_VLAN)
-			info |= (tx_flags & IGBVF_TX_FLAGS_VLAN_MASK);
+	ipv6_find_hdr(skb, &offset, IPPROTO_SCTP, NULL, NULL);
 
-		info |= (skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT);
-		if (skb->ip_summed == CHECKSUM_PARTIAL)
-			info |= (skb_transport_header(skb) -
-				 skb_network_header(skb));
+	return offset == skb_checksum_start_offset(skb);
+}
 
-		context_desc->vlan_macip_lens = cpu_to_le32(info);
+static bool igbvf_tx_csum(struct igbvf_ring *tx_ring, struct sk_buff *skb,
+			  u32 tx_flags, __be16 protocol)
+{
+	u32 vlan_macip_lens = 0;
+	u32 type_tucmd = 0;
 
-		tu_cmd |= (E1000_TXD_CMD_DEXT | E1000_ADVTXD_DTYP_CTXT);
+	if (skb->ip_summed != CHECKSUM_PARTIAL) {
+csum_failed:
+		if (!(tx_flags & IGBVF_TX_FLAGS_VLAN))
+			return false;
+		goto no_csum;
+	}
 
-		if (skb->ip_summed == CHECKSUM_PARTIAL) {
-			switch (protocol) {
-			case htons(ETH_P_IP):
-				tu_cmd |= E1000_ADVTXD_TUCMD_IPV4;
-				if (ip_hdr(skb)->protocol == IPPROTO_TCP)
-					tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP;
-				break;
-			case htons(ETH_P_IPV6):
-				if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
-					tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP;
-				break;
-			default:
-				break;
-			}
+	switch (skb->csum_offset) {
+	case offsetof(struct tcphdr, check):
+		type_tucmd = E1000_ADVTXD_TUCMD_L4T_TCP;
+		/* fall through */
+	case offsetof(struct udphdr, check):
+		break;
+	case offsetof(struct sctphdr, checksum):
+		/* validate that this is actually an SCTP request */
+		if (((protocol == htons(ETH_P_IP)) &&
+		     (ip_hdr(skb)->protocol == IPPROTO_SCTP)) ||
+		    ((protocol == htons(ETH_P_IPV6)) &&
+		     igbvf_ipv6_csum_is_sctp(skb))) {
+			type_tucmd = E1000_ADVTXD_TUCMD_L4T_SCTP;
+			break;
 		}
-
-		context_desc->type_tucmd_mlhl = cpu_to_le32(tu_cmd);
-		context_desc->seqnum_seed = 0;
-		context_desc->mss_l4len_idx = 0;
-
-		buffer_info->time_stamp = jiffies;
-		buffer_info->dma = 0;
-		i++;
-		if (i == tx_ring->count)
-			i = 0;
-		tx_ring->next_to_use = i;
-
-		return true;
+	default:
+		skb_checksum_help(skb);
+		goto csum_failed;
 	}
 
-	return false;
+	vlan_macip_lens = skb_checksum_start_offset(skb) -
+			  skb_network_offset(skb);
+no_csum:
+	vlan_macip_lens |= skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT;
+	vlan_macip_lens |= tx_flags & IGBVF_TX_FLAGS_VLAN_MASK;
+
+	igbvf_tx_ctxtdesc(tx_ring, vlan_macip_lens, type_tucmd, 0);
+	return true;
 }
 
 static int igbvf_maybe_stop_tx(struct net_device *netdev, int size)
@@ -2264,7 +2281,7 @@ static netdev_tx_t igbvf_xmit_frame_ring_adv(struct sk_buff *skb,
 
 	if (tso)
 		tx_flags |= IGBVF_TX_FLAGS_TSO;
-	else if (igbvf_tx_csum(adapter, tx_ring, skb, tx_flags, protocol) &&
+	else if (igbvf_tx_csum(tx_ring, skb, tx_flags, protocol) &&
 		 (skb->ip_summed == CHECKSUM_PARTIAL))
 		tx_flags |= IGBVF_TX_FLAGS_CSUM;
 
@@ -2717,11 +2734,11 @@ static int igbvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	adapter->bd_number = cards_found++;
 
 	netdev->hw_features = NETIF_F_SG |
-			   NETIF_F_IP_CSUM |
-			   NETIF_F_IPV6_CSUM |
-			   NETIF_F_TSO |
-			   NETIF_F_TSO6 |
-			   NETIF_F_RXCSUM;
+			      NETIF_F_TSO |
+			      NETIF_F_TSO6 |
+			      NETIF_F_RXCSUM |
+			      NETIF_F_HW_CSUM |
+			      NETIF_F_SCTP_CRC;
 
 	netdev->features = netdev->hw_features |
 			   NETIF_F_HW_VLAN_CTAG_TX |
@@ -2731,11 +2748,14 @@ static int igbvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-	netdev->vlan_features |= NETIF_F_TSO;
-	netdev->vlan_features |= NETIF_F_TSO6;
-	netdev->vlan_features |= NETIF_F_IP_CSUM;
-	netdev->vlan_features |= NETIF_F_IPV6_CSUM;
-	netdev->vlan_features |= NETIF_F_SG;
+	netdev->vlan_features |= NETIF_F_SG |
+				 NETIF_F_TSO |
+				 NETIF_F_TSO6 |
+				 NETIF_F_HW_CSUM |
+				 NETIF_F_SCTP_CRC;
+
+	netdev->mpls_features |= NETIF_F_HW_CSUM;
+	netdev->hw_enc_features |= NETIF_F_HW_CSUM;
 
 	/*reset the controller to put the device in a known good state */
 	err = hw->mac.ops.reset_hw(hw);
diff --git a/drivers/net/ethernet/intel/igbvf/vf.h b/drivers/net/ethernet/intel/igbvf/vf.h
index 0f1eca6..f00a41d 100644
--- a/drivers/net/ethernet/intel/igbvf/vf.h
+++ b/drivers/net/ethernet/intel/igbvf/vf.h
@@ -126,6 +126,7 @@ struct e1000_adv_tx_context_desc {
 #define E1000_ADVTXD_MACLEN_SHIFT	9  /* Adv ctxt desc mac len shift */
 #define E1000_ADVTXD_TUCMD_IPV4		0x00000400 /* IP Packet Type: 1=IPv4 */
 #define E1000_ADVTXD_TUCMD_L4T_TCP	0x00000800 /* L4 Packet TYPE of TCP */
+#define E1000_ADVTXD_TUCMD_L4T_SCTP	0x00001000 /* L4 packet TYPE of SCTP */
 #define E1000_ADVTXD_L4LEN_SHIFT	8  /* Adv ctxt L4LEN shift */
 #define E1000_ADVTXD_MSS_SHIFT		16 /* Adv ctxt MSS shift */
 
-- 
cgit v0.10.2


From cc54a59ae6e528c70666033ed085d059f555a57d Mon Sep 17 00:00:00 2001
From: Jon Maxwell <jmaxwell37@gmail.com>
Date: Mon, 25 Jan 2016 10:22:52 +1100
Subject: igbvf: remove "link is Up" message when registering mcast address

A similar issue was addressed a few years ago in the following thread:

http://www.spinics.net/lists/netdev/msg245877.html

At that time there were concerns that removing this statement may cause other
side effects. However the submitter addressed those concerns. But the dialogue
went cold. We have a new case where a customers application is registering and
un-registering multicast addresses every few seconds. This is leading to many
"Link is Up" messages in the logs as a result of the
"netif_carrier_off(netdev)" statement called by igbvf_msix_other(). Also on
some kernels it is interfering with the bonding driver causing it to failover
and subsequently affecting connectivity.

The Sourgeforge driver does not make this call and is therefore not affected.
If there were any side effects I would expect that driver to also be affected.
I have tested re-loading the igbvf driver and downing the adapter with the PF
entity on the host where the VM has this patch. When I bring it back up again
connectivity is restored as expected. Therefore I request that this patch gets
submitted.

Signed-off-by: Jon Maxwell <jmaxwell37@gmail.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index aa34865..c124422 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -877,7 +877,6 @@ static irqreturn_t igbvf_msix_other(int irq, void *data)
 
 	adapter->int_counter1++;
 
-	netif_carrier_off(netdev);
 	hw->mac.get_link_status = 1;
 	if (!test_bit(__IGBVF_DOWN, &adapter->state))
 		mod_timer(&adapter->watchdog_timer, jiffies + 1);
-- 
cgit v0.10.2


From 030f9f52642a20cbd8c1334a237e92e3ef55e2b1 Mon Sep 17 00:00:00 2001
From: Corinna Vinschen <vinschen@redhat.com>
Date: Thu, 28 Jan 2016 13:53:23 +0100
Subject: igb: Fix VLAN tag stripping on Intel i350

Problem: When switching off VLAN offloading on an i350, the VLAN
interface gets unusable.  For testing, set up a VLAN on an i350
and some remote machine, e.g.:

  $ ip link add link eth0 name eth0.42 type vlan id 42
  $ ip addr add 192.168.42.1/24 dev eth0.42
  $ ip link set dev eth0.42 up

Offloading is switched on by default:

  $ ethtool -k eth0 | grep vlan-offload
  rx-vlan-offload: on
  tx-vlan-offload: on

  $ ping -c 3 -I eth0.42 192.168.42.2
  [...works as usual...]

Now switch off VLAN offloading and try again:

  $ ethtool -K eth0 rxvlan off
  Actual changes:
  rx-vlan-offload: off
  tx-vlan-offload: off [requested on]
  $ ping -c 3 -I eth0.42 192.168.42.2
  PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da
ta.

  --- 192.168.42.2 ping statistics ---
  3 packets transmitted, 0 received, 100% packet loss, time 1999ms

I can only reproduce it on an i350, the above works fine on a 82580.

While inspecting the igb source, I came across the code in igb_set_vmolr
which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and
for all, and in all of the igb code there's no other place where the
STRVLAN is set or cleared.  Thus, VLAN stripping is enabled in igb
unconditionally, independently of the offloading setting.

I compared that to the latest Intel igb-5.3.3.5 driver from
http://sourceforge.net/projects/e1000/ which in fact sets and clears the
STRVLAN flag independently from igb_set_vmolr in its own function
igb_set_vf_vlan_strip, depending on the vlan settings.

So I included the STRVLAN handling from the igb-5.3.3.5 driver into our
current igb driver and tested the above scenario again.  This time ping
still works after switching off VLAN offloading.

Tested on i350, with and without addtional VFs, as well as on 82580
successfully.

Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index d90dcd1..3b56f80 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -3589,6 +3589,28 @@ static inline int igb_set_vf_rlpml(struct igb_adapter *adapter, int size,
 	return 0;
 }
 
+static inline void igb_set_vf_vlan_strip(struct igb_adapter *adapter,
+					 int vfn, bool enable)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	u32 val, reg;
+
+	if (hw->mac.type < e1000_82576)
+		return;
+
+	if (hw->mac.type == e1000_i350)
+		reg = E1000_DVMOLR(vfn);
+	else
+		reg = E1000_VMOLR(vfn);
+
+	val = rd32(reg);
+	if (enable)
+		val |= E1000_VMOLR_STRVLAN;
+	else
+		val &= ~(E1000_VMOLR_STRVLAN);
+	wr32(reg, val);
+}
+
 static inline void igb_set_vmolr(struct igb_adapter *adapter,
 				 int vfn, bool aupe)
 {
@@ -3602,14 +3624,6 @@ static inline void igb_set_vmolr(struct igb_adapter *adapter,
 		return;
 
 	vmolr = rd32(E1000_VMOLR(vfn));
-	vmolr |= E1000_VMOLR_STRVLAN; /* Strip vlan tags */
-	if (hw->mac.type == e1000_i350) {
-		u32 dvmolr;
-
-		dvmolr = rd32(E1000_DVMOLR(vfn));
-		dvmolr |= E1000_DVMOLR_STRVLAN;
-		wr32(E1000_DVMOLR(vfn), dvmolr);
-	}
 	if (aupe)
 		vmolr |= E1000_VMOLR_AUPE; /* Accept untagged packets */
 	else
@@ -6099,6 +6113,7 @@ static int igb_enable_port_vlan(struct igb_adapter *adapter, int vf,
 
 	adapter->vf_data[vf].pf_vlan = vlan;
 	adapter->vf_data[vf].pf_qos = qos;
+	igb_set_vf_vlan_strip(adapter, vf, true);
 	dev_info(&adapter->pdev->dev,
 		 "Setting VLAN %d, QOS 0x%x on VF %d\n", vlan, qos, vf);
 	if (test_bit(__IGB_DOWN, &adapter->state)) {
@@ -6126,6 +6141,7 @@ static int igb_disable_port_vlan(struct igb_adapter *adapter, int vf)
 
 	adapter->vf_data[vf].pf_vlan = 0;
 	adapter->vf_data[vf].pf_qos = 0;
+	igb_set_vf_vlan_strip(adapter, vf, false);
 
 	return 0;
 }
@@ -6146,6 +6162,7 @@ static int igb_set_vf_vlan_msg(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
 {
 	int add = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >> E1000_VT_MSGINFO_SHIFT;
 	int vid = (msgbuf[1] & E1000_VLVF_VLANID_MASK);
+	int ret;
 
 	if (adapter->vf_data[vf].pf_vlan)
 		return -1;
@@ -6154,7 +6171,10 @@ static int igb_set_vf_vlan_msg(struct igb_adapter *adapter, u32 *msgbuf, u32 vf)
 	if (!vid && !add)
 		return 0;
 
-	return igb_set_vf_vlan(adapter, vid, !!add, vf);
+	ret = igb_set_vf_vlan(adapter, vid, !!add, vf);
+	if (!ret)
+		igb_set_vf_vlan_strip(adapter, vf, !!vid);
+	return ret;
 }
 
 static inline void igb_vf_reset(struct igb_adapter *adapter, u32 vf)
@@ -6171,6 +6191,7 @@ static inline void igb_vf_reset(struct igb_adapter *adapter, u32 vf)
 	igb_set_vmvir(adapter, vf_data->pf_vlan |
 			       (vf_data->pf_qos << VLAN_PRIO_SHIFT), vf);
 	igb_set_vmolr(adapter, vf, !vf_data->pf_vlan);
+	igb_set_vf_vlan_strip(adapter, vf, !!(vf_data->pf_vlan));
 
 	/* reset multicast table array for vf */
 	adapter->vf_data[vf].num_vf_mc_hashes = 0;
@@ -7323,6 +7344,8 @@ static void igb_vlan_mode(struct net_device *netdev, netdev_features_t features)
 		ctrl &= ~E1000_CTRL_VME;
 		wr32(E1000_CTRL, ctrl);
 	}
+
+	igb_set_vf_vlan_strip(adapter, adapter->vfs_allocated_count, enable);
 }
 
 static int igb_vlan_rx_add_vid(struct net_device *netdev,
-- 
cgit v0.10.2


From 46eafa59e18d034ba616fdcca688c388d0bbfd91 Mon Sep 17 00:00:00 2001
From: Stefan Assmann <sassmann@kpanic.de>
Date: Wed, 3 Feb 2016 09:20:50 +0100
Subject: igb: call ndo_stop() instead of dev_close() when running offline
 selftest

Calling dev_close() causes IFF_UP to be cleared which will remove the
interfaces routes and some addresses. That's probably not what the user
intended when running the offline selftest. Besides this does not happen
if the interface is brought down before the test, so the current
behaviour is inconsistent.
Instead call the net_device_ops ndo_stop function directly and avoid
touching IFF_UP at all.

Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index 707ae5c..9413fa6 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -510,6 +510,8 @@ enum igb_boards {
 extern char igb_driver_name[];
 extern char igb_driver_version[];
 
+int igb_open(struct net_device *netdev);
+int igb_close(struct net_device *netdev);
 int igb_up(struct igb_adapter *);
 void igb_down(struct igb_adapter *);
 void igb_reinit_locked(struct igb_adapter *);
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index 1d329f1..7982243 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -2017,7 +2017,7 @@ static void igb_diag_test(struct net_device *netdev,
 
 		if (if_running)
 			/* indicate we're in test mode */
-			dev_close(netdev);
+			igb_close(netdev);
 		else
 			igb_reset(adapter);
 
@@ -2050,7 +2050,7 @@ static void igb_diag_test(struct net_device *netdev,
 
 		clear_bit(__IGB_TESTING, &adapter->state);
 		if (if_running)
-			dev_open(netdev);
+			igb_open(netdev);
 	} else {
 		dev_info(&adapter->pdev->dev, "online testing starting\n");
 
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 3b56f80..834b1b6 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -122,8 +122,8 @@ static void igb_setup_mrqc(struct igb_adapter *);
 static int igb_probe(struct pci_dev *, const struct pci_device_id *);
 static void igb_remove(struct pci_dev *pdev);
 static int igb_sw_init(struct igb_adapter *);
-static int igb_open(struct net_device *);
-static int igb_close(struct net_device *);
+int igb_open(struct net_device *);
+int igb_close(struct net_device *);
 static void igb_configure(struct igb_adapter *);
 static void igb_configure_tx(struct igb_adapter *);
 static void igb_configure_rx(struct igb_adapter *);
@@ -3172,7 +3172,7 @@ err_setup_tx:
 	return err;
 }
 
-static int igb_open(struct net_device *netdev)
+int igb_open(struct net_device *netdev)
 {
 	return __igb_open(netdev, false);
 }
@@ -3209,7 +3209,7 @@ static int __igb_close(struct net_device *netdev, bool suspending)
 	return 0;
 }
 
-static int igb_close(struct net_device *netdev)
+int igb_close(struct net_device *netdev)
 {
 	return __igb_close(netdev, false);
 }
-- 
cgit v0.10.2


From 4fee7dab07bb2c7dfc3369e0f0e28e3fd4fc00c4 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 23 Feb 2016 11:36:02 +0100
Subject: bnx2x: add a separate GENEVE Kconfig symbol

When CONFIG_GENEVE is built as a loadable module, and bnx2x is built-in,
we get this link error:

drivers/net/built-in.o: In function `bnx2x_open':
:(.text+0x33322): undefined reference to `geneve_get_rx_port'
drivers/net/built-in.o: In function `bnx2x_sp_rtnl_task':
:(.text+0x3e632): undefined reference to `geneve_get_rx_port'

This avoids the problem by adding a separate Kconfig symbol named
CONFIG_BNX2X_GENEVE that is only enabled when the code is
reachable from the driver.

This is the same trick that BNX2X does for VXLAN support, and
is similar to how I40E handles both.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 883ce97d25b0 ("bnx2x: Add Geneve inner-RSS support")
Acked-By: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig
index 19f7cd0..18042c2 100644
--- a/drivers/net/ethernet/broadcom/Kconfig
+++ b/drivers/net/ethernet/broadcom/Kconfig
@@ -149,6 +149,16 @@ config BNX2X_VXLAN
 	  Say Y here if you want to enable hardware offload support for
 	  Virtual eXtensible Local Area Network (VXLAN) in the driver.
 
+config BNX2X_GENEVE
+	bool "Generic Network Virtualization Encapsulation (GENEVE) support"
+	depends on BNX2X && GENEVE && !(BNX2X=y && GENEVE=m)
+	---help---
+          This allows one to create GENEVE virtual interfaces that provide
+          Layer 2 Networks over Layer 3 Networks. GENEVE is often used
+          to tunnel virtual network infrastructure in virtualized environments.
+	  Say Y here if you want to enable hardware offload support for
+	  Generic Network Virtualization Encapsulation (GENEVE) in the driver.
+
 config BGMAC
 	tristate "BCMA bus GBit core support"
 	depends on BCMA && BCMA_HOST_SOC
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 81fc51c..5c95d0c 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -59,7 +59,7 @@
 #include <linux/semaphore.h>
 #include <linux/stringify.h>
 #include <linux/vmalloc.h>
-#if IS_ENABLED(CONFIG_GENEVE)
+#if IS_ENABLED(CONFIG_BNX2X_GENEVE)
 #include <net/geneve.h>
 #endif
 #include "bnx2x.h"
@@ -10078,7 +10078,7 @@ static void bnx2x_parity_recover(struct bnx2x *bp)
 	}
 }
 
-#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_GENEVE)
+#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_BNX2X_GENEVE)
 static int bnx2x_udp_port_update(struct bnx2x *bp)
 {
 	struct bnx2x_func_switch_update_params *switch_update_params;
@@ -10201,7 +10201,7 @@ static void bnx2x_del_vxlan_port(struct net_device *netdev,
 }
 #endif
 
-#if IS_ENABLED(CONFIG_GENEVE)
+#if IS_ENABLED(CONFIG_BNX2X_GENEVE)
 static void bnx2x_add_geneve_port(struct net_device *netdev,
 				  sa_family_t sa_family, __be16 port)
 {
@@ -10327,7 +10327,7 @@ sp_rtnl_not_reset:
 			       &bp->sp_rtnl_state))
 		bnx2x_update_mng_version(bp);
 
-#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_GENEVE)
+#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_BNX2X_GENEVE)
 	if (test_and_clear_bit(BNX2X_SP_RTNL_CHANGE_UDP_PORT,
 			       &bp->sp_rtnl_state)) {
 		if (bnx2x_udp_port_update(bp)) {
@@ -10344,7 +10344,7 @@ sp_rtnl_not_reset:
 			if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count)
 				vxlan_get_rx_port(bp->dev);
 #endif
-#if IS_ENABLED(CONFIG_GENEVE)
+#if IS_ENABLED(CONFIG_BNX2X_GENEVE)
 			if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count)
 				geneve_get_rx_port(bp->dev);
 #endif
@@ -12557,7 +12557,7 @@ static int bnx2x_open(struct net_device *dev)
 	if (IS_PF(bp))
 		vxlan_get_rx_port(dev);
 #endif
-#if IS_ENABLED(CONFIG_GENEVE)
+#if IS_ENABLED(CONFIG_BNX2X_GENEVE)
 	if (IS_PF(bp))
 		geneve_get_rx_port(dev);
 #endif
@@ -13078,7 +13078,7 @@ static const struct net_device_ops bnx2x_netdev_ops = {
 	.ndo_add_vxlan_port	= bnx2x_add_vxlan_port,
 	.ndo_del_vxlan_port	= bnx2x_del_vxlan_port,
 #endif
-#if IS_ENABLED(CONFIG_GENEVE)
+#if IS_ENABLED(CONFIG_BNX2X_GENEVE)
 	.ndo_add_geneve_port	= bnx2x_add_geneve_port,
 	.ndo_del_geneve_port	= bnx2x_del_geneve_port,
 #endif
-- 
cgit v0.10.2


From 555d5b70f1597906dc2e31085f5e70b49d03a536 Mon Sep 17 00:00:00 2001
From: Guillaume Nault <g.nault@alphalink.fr>
Date: Tue, 23 Feb 2016 13:59:43 +0100
Subject: ppp: clarify parsing of user supplied data in ppp_set_compress()

* Split big conditional statement.
  * Check (data.length <= CCP_MAX_OPTION_LENGTH) only once.
  * Don't read ccp_option[1] if not initialised.

Reading uninitialised ccp_option[1] was harmless, because this could
only happen when data.length was 0 or 1. So even then, we couldn't pass
the (ccp_option[1] < 2 || ccp_option[1] > data.length) test anyway.

Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index fc8ad00..04f4eb3 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -2429,13 +2429,15 @@ ppp_set_compress(struct ppp *ppp, unsigned long arg)
 	unsigned char ccp_option[CCP_MAX_OPTION_LENGTH];
 
 	err = -EFAULT;
-	if (copy_from_user(&data, (void __user *) arg, sizeof(data)) ||
-	    (data.length <= CCP_MAX_OPTION_LENGTH &&
-	     copy_from_user(ccp_option, (void __user *) data.ptr, data.length)))
+	if (copy_from_user(&data, (void __user *) arg, sizeof(data)))
 		goto out;
+	if (data.length > CCP_MAX_OPTION_LENGTH)
+		goto out;
+	if (copy_from_user(ccp_option, (void __user *) data.ptr, data.length))
+		goto out;
+
 	err = -EINVAL;
-	if (data.length > CCP_MAX_OPTION_LENGTH ||
-	    ccp_option[1] < 2 || ccp_option[1] > data.length)
+	if (data.length < 2 || ccp_option[1] < 2 || ccp_option[1] > data.length)
 		goto out;
 
 	cp = try_then_request_module(
-- 
cgit v0.10.2


From 81ca7835f2cb0c3ba4236e3bcf31d997c6f5d71a Mon Sep 17 00:00:00 2001
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Fri, 29 Jan 2016 11:24:24 -0800
Subject: NFC: Use GFP_USER for user-controlled kmalloc

These two functions are called in sendmsg path, and the
'len' is passed from user-space, so we should not allow
malicious users to OOM kernel on purpose.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

diff --git a/net/nfc/llcp_commands.c b/net/nfc/llcp_commands.c
index 3621a90..3425532 100644
--- a/net/nfc/llcp_commands.c
+++ b/net/nfc/llcp_commands.c
@@ -663,7 +663,7 @@ int nfc_llcp_send_i_frame(struct nfc_llcp_sock *sock,
 		return -ENOBUFS;
 	}
 
-	msg_data = kzalloc(len, GFP_KERNEL);
+	msg_data = kmalloc(len, GFP_USER | __GFP_NOWARN);
 	if (msg_data == NULL)
 		return -ENOMEM;
 
@@ -729,7 +729,7 @@ int nfc_llcp_send_ui_frame(struct nfc_llcp_sock *sock, u8 ssap, u8 dsap,
 	if (local == NULL)
 		return -ENODEV;
 
-	msg_data = kzalloc(len, GFP_KERNEL);
+	msg_data = kmalloc(len, GFP_USER | __GFP_NOWARN);
 	if (msg_data == NULL)
 		return -ENOMEM;
 
-- 
cgit v0.10.2


From 03c05355543149bf610f4375e8382ee4fc0aaade Mon Sep 17 00:00:00 2001
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Fri, 29 Jan 2016 11:37:40 -0800
Subject: NFC: Close a race condition in llcp_sock_getname()

llcp_sock_getname() checks llcp_sock->dev to make sure
llcp_sock is already connected or bound, however, we could
be in the middle of llcp_sock_bind() where llcp_sock->dev
is bound and llcp_sock->service_name_len is set,
but llcp_sock->service_name is not, in this case we would
lead to copy some bytes from a NULL pointer.

Just lock the sock since this is not a hot path anyway.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c
index ecf0a01..b9edf5f 100644
--- a/net/nfc/llcp_sock.c
+++ b/net/nfc/llcp_sock.c
@@ -509,6 +509,11 @@ static int llcp_sock_getname(struct socket *sock, struct sockaddr *uaddr,
 	memset(llcp_addr, 0, sizeof(*llcp_addr));
 	*len = sizeof(struct sockaddr_nfc_llcp);
 
+	lock_sock(sk);
+	if (!llcp_sock->dev) {
+		release_sock(sk);
+		return -EBADFD;
+	}
 	llcp_addr->sa_family = AF_NFC;
 	llcp_addr->dev_idx = llcp_sock->dev->idx;
 	llcp_addr->target_idx = llcp_sock->target_idx;
@@ -518,6 +523,7 @@ static int llcp_sock_getname(struct socket *sock, struct sockaddr *uaddr,
 	llcp_addr->service_name_len = llcp_sock->service_name_len;
 	memcpy(llcp_addr->service_name, llcp_sock->service_name,
 	       llcp_addr->service_name_len);
+	release_sock(sk);
 
 	return 0;
 }
-- 
cgit v0.10.2


From 43154f6fd46bb3bcd363843c9e48f0b8b3a03a08 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:35 -0500
Subject: rtl8xxxu: Minor cleanup to rtl8xxxu_download_firmware()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 6aed923..6b39e4d 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2100,7 +2100,8 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 
 	/* 8051 enable */
 	val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
-	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16 | SYS_FUNC_CPU_ENABLE);
+	val16 |= SYS_FUNC_CPU_ENABLE;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
 
 	/* MCU firmware download enable */
 	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
-- 
cgit v0.10.2


From ef1c0499316b892d700201edeeb013b60be96877 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:36 -0500
Subject: rtl8xxxu: rtl8xxxu_download_firmware(): Cosmetic cleanups

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 6b39e4d..b6a264b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2105,15 +2105,18 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 
 	/* MCU firmware download enable */
 	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
-	rtl8xxxu_write8(priv, REG_MCU_FW_DL, val8 | MCU_FW_DL_ENABLE);
+	val8 |= MCU_FW_DL_ENABLE;
+	rtl8xxxu_write8(priv, REG_MCU_FW_DL, val8);
 
 	/* 8051 reset */
 	val32 = rtl8xxxu_read32(priv, REG_MCU_FW_DL);
-	rtl8xxxu_write32(priv, REG_MCU_FW_DL, val32 & ~BIT(19));
+	val32 &= ~BIT(19);
+	rtl8xxxu_write32(priv, REG_MCU_FW_DL, val32);
 
 	/* Reset firmware download checksum */
 	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
-	rtl8xxxu_write8(priv, REG_MCU_FW_DL, val8 | MCU_FW_DL_CSUM_REPORT);
+	val8 |= MCU_FW_DL_CSUM_REPORT;
+	rtl8xxxu_write8(priv, REG_MCU_FW_DL, val8);
 
 	pages = priv->fw_size / RTL_FW_PAGE_SIZE;
 	remainder = priv->fw_size % RTL_FW_PAGE_SIZE;
@@ -2122,7 +2125,8 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 
 	for (i = 0; i < pages; i++) {
 		val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL + 2) & 0xF8;
-		rtl8xxxu_write8(priv, REG_MCU_FW_DL + 2, val8 | i);
+		val8 |= i;
+		rtl8xxxu_write8(priv, REG_MCU_FW_DL + 2, val8);
 
 		ret = rtl8xxxu_writeN(priv, REG_FW_START_ADDRESS,
 				      fwptr, RTL_FW_PAGE_SIZE);
@@ -2136,7 +2140,8 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 
 	if (remainder) {
 		val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL + 2) & 0xF8;
-		rtl8xxxu_write8(priv, REG_MCU_FW_DL + 2, val8 | i);
+		val8 |= i;
+		rtl8xxxu_write8(priv, REG_MCU_FW_DL + 2, val8);
 		ret = rtl8xxxu_writeN(priv, REG_FW_START_ADDRESS,
 				      fwptr, remainder);
 		if (ret != remainder) {
@@ -2149,8 +2154,8 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 fw_abort:
 	/* MCU firmware download disable */
 	val16 = rtl8xxxu_read16(priv, REG_MCU_FW_DL);
-	rtl8xxxu_write16(priv, REG_MCU_FW_DL,
-			 val16 & (~MCU_FW_DL_ENABLE & 0xff));
+	val16 &= ~MCU_FW_DL_ENABLE;
+	rtl8xxxu_write16(priv, REG_MCU_FW_DL, val16);
 
 	return ret;
 }
-- 
cgit v0.10.2


From 216202ae20ea93585630e6a88486ea23200199cc Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:37 -0500
Subject: rtl8xxxu: If fw running in RAM, reset the 8051 before trying to
 download a new one

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index b6a264b..e98e053 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2089,7 +2089,7 @@ exit:
 static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 {
 	int pages, remainder, i, ret;
-	u8 val8;
+	u8 val8, sys_func;
 	u16 val16;
 	u32 val32;
 	u8 *fwptr;
@@ -2103,6 +2103,23 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 	val16 |= SYS_FUNC_CPU_ENABLE;
 	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
 
+	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
+	if (val8 & MCU_FW_RAM_SEL) {
+		pr_info("do the RAM reset\n");
+		rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00);
+		val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
+		val8 &= ~BIT(3);
+		rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+		sys_func = rtl8xxxu_read8(priv, REG_SYS_FUNC + 1);
+		sys_func &= ~BIT(2);
+		rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, sys_func);
+		val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
+		val8 |= BIT(3);
+		rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+		sys_func |= BIT(2);
+		rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, sys_func);
+	}
+
 	/* MCU firmware download enable */
 	val8 = rtl8xxxu_read8(priv, REG_MCU_FW_DL);
 	val8 |= MCU_FW_DL_ENABLE;
-- 
cgit v0.10.2


From 0635f8cede159bb2205021583a483b0cd3246c74 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:38 -0500
Subject: rtl8xxxu: Add RQPN_[NE]PQ_SHIFT values

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 23208f7..8b27a1c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -352,6 +352,8 @@
 #define REG_TXDMA_OFFSET_CHK		0x020c
 #define REG_TXDMA_STATUS		0x0210
 #define REG_RQPN_NPQ			0x0214
+#define  RQPN_NPQ_SHIFT			0
+#define  RQPN_EPQ_SHIFT			16
 
 /* 0x0280 ~ 0x02FF	RXDMA Configuration */
 #define REG_RXDMA_AGG_PG_TH		0x0280
-- 
cgit v0.10.2


From f076977537d9b6ac5d067d416b549aedd591da4f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:39 -0500
Subject: rtl8xxxu: Define SYS_CFG_SW_OFFLOAD_EN

This is needed for chips with IOL support.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 8b27a1c..7c795ce 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -213,6 +213,7 @@
 #define  SYS_CFG_PCIRSTB		BIT(4)
 #define  SYS_CFG_V15_VLD		BIT(5)
 #define  SYS_CFG_TRP_B15V_EN		BIT(7)
+#define  SYS_CFG_SW_OFFLOAD_EN		BIT(7)	/* For chips with IOL support */
 #define  SYS_CFG_SIC_IDLE		BIT(8)
 #define  SYS_CFG_BD_MAC2		BIT(9)
 #define  SYS_CFG_BD_MAC1		BIT(10)
-- 
cgit v0.10.2


From cecfd3cb29f8f39b8859c3c189185df23b69e45c Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:40 -0500
Subject: rtl8xxxu: Add REG_TX_REPORT_* defines

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 7c795ce..89547b6 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -456,6 +456,8 @@
 #define REG_NEED_CPU_HANDLE		0x04e0
 #define REG_PKT_LOSE_RPT		0x04e1
 #define REG_PTCL_ERR_STATUS		0x04e2
+#define REG_TX_REPORT_CTRL		0x04ec
+#define REG_TX_REPORT_TIME		0x04f0
 #define REG_DUMMY			0x04fc
 
 /* 0x0500 ~ 0x05FF	EDCA Configuration */
-- 
cgit v0.10.2


From 22229fcb487a161561cfcb08e6a4028d6343b2e7 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:41 -0500
Subject: rtl8xxxu: Add more RCR bits

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 89547b6..f752602 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -568,9 +568,18 @@
 #define  RCR_ACCEPT_CTRL_FRAME		BIT(12)
 #define  RCR_ACCEPT_MGMT_FRAME		BIT(13)
 #define  RCR_HTC_LOC_CTRL		BIT(14) /* MFC<--HTC=1 MFC-->HTC=0 */
+#define  RCR_UC_DATA_PKT_INT_ENABLE	BIT(16) /* Enable unicast data packet
+						   interrupt */
+#define  RCR_BM_DATA_PKT_INT_ENABLE	BIT(17) /* Enable broadcast data packet
+						   interrupt */
+#define  RCR_TIM_PARSER_ENABLE		BIT(18) /* Enable RX beacon TIM parser*/
 #define  RCR_MFBEN			BIT(22)
-#define  RCR_LSIGEN			BIT(23)
+#define  RCR_LSIG_ENABLE		BIT(23) /* Enable LSIG TXOP Protection
+						   function. Search KEYCAM for
+						   each rx packet to check if
+						   LSIGEN bit is set. */
 #define  RCR_MULTI_BSSID_ENABLE		BIT(24) /* Enable Multiple BssId */
+#define  RCR_FORCE_ACK			BIT(26)
 #define  RCR_ACCEPT_BA_SSN		BIT(27) /* Accept BA SSN */
 #define  RCR_APPEND_PHYSTAT		BIT(28)
 #define  RCR_APPEND_ICV			BIT(29)
-- 
cgit v0.10.2


From c1edece3c22be23cee3b28f74aecd1736ee816a1 Mon Sep 17 00:00:00 2001
From: Tobias Klauser <tklauser@distanz.ch>
Date: Wed, 3 Feb 2016 13:39:42 -0500
Subject: rtl8xxxu: Fix typo in Kconfig help text

Fix spelling of "benefits".

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/Kconfig b/drivers/net/wireless/realtek/rtl8xxxu/Kconfig
index dd4d626..8f053c3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/Kconfig
+++ b/drivers/net/wireless/realtek/rtl8xxxu/Kconfig
@@ -13,7 +13,7 @@ config RTL8XXXU
 	  This driver is under development and has a limited feature
 	  set. In particular it does not yet support 40MHz channels
 	  and power management. However it should have a smaller
-	  memory footprint than the vendor drivers and benetifs
+	  memory footprint than the vendor drivers and benefits
 	  from the in kernel mac80211 stack.
 
 	  It can coexist with drivers from drivers/staging/rtl8723au,
-- 
cgit v0.10.2


From 98e27cbd9453cd99403ee5f929a11a1000a3090b Mon Sep 17 00:00:00 2001
From: Tobias Klauser <tklauser@distanz.ch>
Date: Wed, 3 Feb 2016 13:39:43 -0500
Subject: rtl8xxxu: Check return value of kmemdup()

In rtl8xxxu_load_firmware() check the return value of kmemdup() and
error out with -ENOMEM in case of NULL to prevent a NULL pointer
dereference.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e98e053..08d8e1f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2197,6 +2197,10 @@ static int rtl8xxxu_load_firmware(struct rtl8xxxu_priv *priv, char *fw_name)
 	}
 
 	priv->fw_data = kmemdup(fw->data, fw->size, GFP_KERNEL);
+	if (!priv->fw_data) {
+		ret = -ENOMEM;
+		goto exit;
+	}
 	priv->fw_size = fw->size - sizeof(struct rtl8xxxu_firmware_header);
 
 	signature = le16_to_cpu(priv->fw_data->signature);
-- 
cgit v0.10.2


From d48fe60e06f404acf1acb346a4bf2d7b0e1d4d86 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:44 -0500
Subject: rtl8xxxu: Implement rtl8xxxu_reset_8051()

We need to reset the 8051 in order for it to launch the fw on the
rtl8192eu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 08d8e1f..7a571ef 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2043,6 +2043,24 @@ exit:
 	return ret;
 }
 
+static void rtl8xxxu_reset_8051(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u16 sys_func;
+
+	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
+	val8 &= ~BIT(3);
+	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+	sys_func = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+	sys_func &= ~SYS_FUNC_CPU_ENABLE;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
+	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
+	val8 |= BIT(3);
+	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+	sys_func |= SYS_FUNC_CPU_ENABLE;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
+}
+
 static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 {
 	struct device *dev = &priv->udev->dev;
@@ -2067,6 +2085,12 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 	val32 &= ~MCU_WINT_INIT_READY;
 	rtl8xxxu_write32(priv, REG_MCU_FW_DL, val32);
 
+	/*
+	 * Reset the 8051 in order for the firmware to start running,
+	 * otherwise it won't come up on the 8192eu
+	 */
+	rtl8xxxu_reset_8051(priv);
+
 	/* Wait for firmware to become ready */
 	for (i = 0; i < RTL8XXXU_FIRMWARE_POLL_MAX; i++) {
 		val32 = rtl8xxxu_read32(priv, REG_MCU_FW_DL);
@@ -2089,7 +2113,7 @@ exit:
 static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 {
 	int pages, remainder, i, ret;
-	u8 val8, sys_func;
+	u8 val8;
 	u16 val16;
 	u32 val32;
 	u8 *fwptr;
@@ -2107,17 +2131,7 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 	if (val8 & MCU_FW_RAM_SEL) {
 		pr_info("do the RAM reset\n");
 		rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00);
-		val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
-		val8 &= ~BIT(3);
-		rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
-		sys_func = rtl8xxxu_read8(priv, REG_SYS_FUNC + 1);
-		sys_func &= ~BIT(2);
-		rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, sys_func);
-		val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
-		val8 |= BIT(3);
-		rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
-		sys_func |= BIT(2);
-		rtl8xxxu_write8(priv, REG_SYS_FUNC + 1, sys_func);
+		rtl8xxxu_reset_8051(priv);
 	}
 
 	/* MCU firmware download enable */
-- 
cgit v0.10.2


From af9e4d6d9d07857b5fb7711832e8cda745e336ae Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:45 -0500
Subject: rtl8xxxu: Add definitions for antenna selection registers for 8723BU

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index f752602..997a857 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -756,6 +756,10 @@
 #define REG_FPGA1_RF_MODE		0x0900
 
 #define REG_FPGA1_TX_INFO		0x090c
+#define REG_DPDT_CTRL			0x092c	/* 8723BU */
+#define REG_RFE_CTRL_ANTA_SRC		0x0930	/* 8723BU */
+#define REG_RFE_PATH_SELECT		0x0940	/* 8723BU */
+#define REG_RFE_BUFFER			0x0944	/* 8723BU */
 
 #define REG_CCK0_SYSTEM			0x0a00
 #define  CCK0_SIDEBAND			BIT(4)
-- 
cgit v0.10.2


From 29c2139cd04ad23226c690221450b93bedaceea4 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:46 -0500
Subject: rtl8xxxu: Add mbox extension register definitions for 8723B

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 997a857..991b366 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -341,6 +341,11 @@
 #define REG_BB_ACCEESS_CTRL		0x01e8
 #define REG_BB_ACCESS_DATA		0x01ec
 
+#define REG_HMBOX_EXT0_8723B		0x01f0
+#define REG_HMBOX_EXT1_8723B		0x01f4
+#define REG_HMBOX_EXT2_8723B		0x01f8
+#define REG_HMBOX_EXT3_8723B		0x01fc
+
 /* 0x0200 ~ 0x027F	TXDMA Configuration */
 #define REG_RQPN			0x0200
 #define  RQPN_HI_PQ_SHIFT		0
-- 
cgit v0.10.2


From ce6f2e3669547ad18ea67761aa9cc32cb4407bf2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:47 -0500
Subject: rtl8xxxu: Add REG_OFDM0_RX_D_SYNC_PATH definitions

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 991b366..d2995f8a3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -793,6 +793,9 @@
 
 #define REG_OFDM0_ENERGY_CCA_THRES	0x0c4c
 
+#define REG_OFDM0_RX_D_SYNC_PATH	0x0c40
+#define  OFDM0_SYNC_PATH_NOTCH_FILTER	BIT(1)
+
 #define REG_OFDM0_XA_AGC_CORE1		0x0c50
 #define REG_OFDM0_XA_AGC_CORE2		0x0c54
 #define REG_OFDM0_XB_AGC_CORE1		0x0c58
-- 
cgit v0.10.2


From e5c447ccf8e568134847bb491663e26910a21d8a Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:48 -0500
Subject: rtl8xxxu: Remove unused clutter for handling recursive calls to
 rtl8xxxu_init_device()

This was a leftover from the vendor driver that was never utilized.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 7a571ef..4a6939f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4289,17 +4289,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	rtl8xxxu_write16(priv, REG_FAST_EDCA_CTRL, 0);
 
-	/*
-	 * Not sure if we should get into this at all
-	 */
-	if (priv->iqk_initialized) {
-		rtl8xxxu_restore_regs(priv, rtl8723au_iqk_phy_iq_bb_reg,
-				      priv->bb_recovery_backup,
-				      RTL8XXXU_BB_REGS);
-	} else {
-		rtl8723a_phy_iq_calibrate(priv);
-		priv->iqk_initialized = true;
-	}
+	rtl8723a_phy_iq_calibrate(priv);
 
 	/*
 	 * This should enable thermal meter
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index f2a1bac..f71fb8d 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -652,7 +652,6 @@ struct rtl8xxxu_priv {
 	u32 bb_recovery_backup[RTL8XXXU_BB_REGS];
 	u32 rtlchip;
 	u8 pi_enabled:1;
-	u8 iqk_initialized:1;
 	u8 int_buf[USB_INTR_CONTENT_LENGTH];
 };
 
-- 
cgit v0.10.2


From 38451998fb8a5f9c933b2c462a790d5633605ac8 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jsitnicki@gmail.com>
Date: Wed, 3 Feb 2016 13:39:49 -0500
Subject: rtl8xxxu: Use REG_EFUSE_TEST register only on multifunctional devices

rtl8192cu driver doesn't read/write the REG_EFUSE_TEST register.
Neither does the rtl8188eu driver.  Do it only for multifunc devices
RTL8723AU/RTL8723BU.

Signed-off-by: Jakub Sitnicki <jsitnicki@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4a6939f..94d3d84 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1704,6 +1704,7 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 			priv->has_bluetooth = 1;
 		if (val32 & MULTI_GPS_FUNC_EN)
 			priv->has_gps = 1;
+		priv->is_multi_func = 1;
 	} else if (val32 & SYS_CFG_TYPE_ID) {
 		bonding = rtl8xxxu_read32(priv, REG_HPON_FSM);
 		bonding &= HPON_FSM_BONDING_MASK;
@@ -1938,9 +1939,11 @@ static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 	if (val16 & EEPROM_BOOT)
 		priv->boot_eeprom = 1;
 
-	val32 = rtl8xxxu_read32(priv, REG_EFUSE_TEST);
-	val32 = (val32 & ~EFUSE_SELECT_MASK) | EFUSE_WIFI_SELECT;
-	rtl8xxxu_write32(priv, REG_EFUSE_TEST, val32);
+	if (priv->is_multi_func) {
+		val32 = rtl8xxxu_read32(priv, REG_EFUSE_TEST);
+		val32 = (val32 & ~EFUSE_SELECT_MASK) | EFUSE_WIFI_SELECT;
+		rtl8xxxu_write32(priv, REG_EFUSE_TEST, val32);
+	}
 
 	dev_dbg(dev, "Booting from %s\n",
 		priv->boot_eeprom ? "EEPROM" : "EFUSE");
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index f71fb8d..bbd0f6b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -597,6 +597,7 @@ struct rtl8xxxu_priv {
 	struct rtl8723au_idx ht20_max_power_offset[3];
 	u32 chip_cut:4;
 	u32 rom_rev:4;
+	u32 is_multi_func:1;
 	u32 has_wifi:1;
 	u32 has_bluetooth:1;
 	u32 enable_bluetooth:1;
-- 
cgit v0.10.2


From 56e4374a5e08b6971069de4bb0824c2aa2149a8f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:50 -0500
Subject: rtl8xxxu: Fix cosmetics to quiet down checkpatch police

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 94d3d84..ac4c211 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4415,7 +4415,7 @@ static void rtl8xxxu_cam_write(struct rtl8xxxu_priv *priv,
 }
 
 static void rtl8xxxu_sw_scan_start(struct ieee80211_hw *hw,
-				   struct ieee80211_vif *vif, const u8* mac)
+				   struct ieee80211_vif *vif, const u8 *mac)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
 	u8 val8;
-- 
cgit v0.10.2


From 3bed4bfa51f2635ff21c73f83503e3e8ac233db2 Mon Sep 17 00:00:00 2001
From: Bruno Randolf <br1@einfach.org>
Date: Wed, 3 Feb 2016 13:39:51 -0500
Subject: rtl8xxxu: Enable monitor mode by handling filters

Monitor mode is enabled by handling the filter flags we get from mac80211 in
rtl8xxxu_configure_filter() and writing them to the RCR register.

By handling the filters, we can also stop setting the BSSID filters in the
association event.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ac4c211..b58a60f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4199,7 +4199,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	 * Configure initial WMAC settings
 	 */
 	val32 = RCR_ACCEPT_PHYS_MATCH | RCR_ACCEPT_MCAST | RCR_ACCEPT_BCAST |
-		/* RCR_CHECK_BSSID_MATCH | RCR_CHECK_BSSID_BEACON | */
 		RCR_ACCEPT_MGMT_FRAME | RCR_HTC_LOC_CTRL |
 		RCR_APPEND_PHYSTAT | RCR_APPEND_ICV | RCR_APPEND_MIC;
 	rtl8xxxu_write32(priv, REG_RCR, val32);
@@ -4522,10 +4521,6 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 			rtl8xxxu_update_rate_mask(priv, ramask, sgi);
 
-			val32 = rtl8xxxu_read32(priv, REG_RCR);
-			val32 |= RCR_CHECK_BSSID_MATCH | RCR_CHECK_BSSID_BEACON;
-			rtl8xxxu_write32(priv, REG_RCR, val32);
-
 			/* Enable RX of data frames */
 			rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0xffff);
 
@@ -4539,11 +4534,6 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 			h2c.joinbss.data = H2C_JOIN_BSS_CONNECT;
 		} else {
-			val32 = rtl8xxxu_read32(priv, REG_RCR);
-			val32 &= ~(RCR_CHECK_BSSID_MATCH |
-				   RCR_CHECK_BSSID_BEACON);
-			rtl8xxxu_write32(priv, REG_RCR, val32);
-
 			val8 = rtl8xxxu_read8(priv, REG_BEACON_CTRL);
 			val8 |= BEACON_DISABLE_TSF_UPDATE;
 			rtl8xxxu_write8(priv, REG_BEACON_CTRL, val8);
@@ -5318,11 +5308,55 @@ static void rtl8xxxu_configure_filter(struct ieee80211_hw *hw,
 				      unsigned int *total_flags, u64 multicast)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
+	u32 rcr = rtl8xxxu_read32(priv, REG_RCR);
 
 	dev_dbg(&priv->udev->dev, "%s: changed_flags %08x, total_flags %08x\n",
 		__func__, changed_flags, *total_flags);
 
-	*total_flags &= (FIF_ALLMULTI | FIF_CONTROL | FIF_BCN_PRBRESP_PROMISC);
+	/*
+	 * FIF_ALLMULTI ignored as all multicast frames are accepted (REG_MAR)
+	 */
+
+	if (*total_flags & FIF_FCSFAIL)
+		rcr |= RCR_ACCEPT_CRC32;
+	else
+		rcr &= ~RCR_ACCEPT_CRC32;
+
+	/*
+	 * FIF_PLCPFAIL not supported?
+	 */
+
+	if (*total_flags & FIF_BCN_PRBRESP_PROMISC)
+		rcr &= ~RCR_CHECK_BSSID_BEACON;
+	else
+		rcr |= RCR_CHECK_BSSID_BEACON;
+
+	if (*total_flags & FIF_CONTROL)
+		rcr |= RCR_ACCEPT_CTRL_FRAME;
+	else
+		rcr &= ~RCR_ACCEPT_CTRL_FRAME;
+
+	if (*total_flags & FIF_OTHER_BSS) {
+		rcr |= RCR_ACCEPT_AP;
+		rcr &= ~RCR_CHECK_BSSID_MATCH;
+	} else {
+		rcr &= ~RCR_ACCEPT_AP;
+		rcr |= RCR_CHECK_BSSID_MATCH;
+	}
+
+	if (*total_flags & FIF_PSPOLL)
+		rcr |= RCR_ACCEPT_PM;
+	else
+		rcr &= ~RCR_ACCEPT_PM;
+
+	/*
+	 * FIF_PROBE_REQ ignored as probe requests always seem to be accepted
+	 */
+
+	rtl8xxxu_write32(priv, REG_RCR, rcr);
+
+	*total_flags &= (FIF_ALLMULTI | FIF_FCSFAIL | FIF_BCN_PRBRESP_PROMISC | \
+			 FIF_CONTROL | FIF_OTHER_BSS | FIF_PSPOLL | FIF_PROBE_REQ);
 }
 
 static int rtl8xxxu_set_rts_threshold(struct ieee80211_hw *hw, u32 rts)
-- 
cgit v0.10.2


From b40027ba8db1b2fe022bf42505cbf64cb0114eba Mon Sep 17 00:00:00 2001
From: Bruno Randolf <br1@einfach.org>
Date: Wed, 3 Feb 2016 13:39:52 -0500
Subject: rtl8xxxu: Document REG_RXFLTMAP registers

Add comments describing how REG_RXFLTMAP0, REG_RXFLTMAP1 and REG_RXFLTMAP2
work.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index d2995f8a3..d089edd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -569,9 +569,12 @@
 						 (Rx beacon, probe rsp) */
 #define  RCR_ACCEPT_CRC32		BIT(8)  /* Accept CRC32 error packet */
 #define  RCR_ACCEPT_ICV			BIT(9)  /* Accept ICV error packet */
-#define  RCR_ACCEPT_DATA_FRAME		BIT(11)
-#define  RCR_ACCEPT_CTRL_FRAME		BIT(12)
-#define  RCR_ACCEPT_MGMT_FRAME		BIT(13)
+#define  RCR_ACCEPT_DATA_FRAME		BIT(11) /* Accept all data pkt or use
+						   REG_RXFLTMAP2 */
+#define  RCR_ACCEPT_CTRL_FRAME		BIT(12) /* Accept all control pkt or use
+						   REG_RXFLTMAP1 */
+#define  RCR_ACCEPT_MGMT_FRAME		BIT(13) /* Accept all mgmt pkt or use
+						   REG_RXFLTMAP0 */
 #define  RCR_HTC_LOC_CTRL		BIT(14) /* MFC<--HTC=1 MFC-->HTC=0 */
 #define  RCR_UC_DATA_PKT_INT_ENABLE	BIT(16) /* Enable unicast data packet
 						   interrupt */
@@ -651,9 +654,20 @@
 #define REG_LPNAV_CTRL			0x0694
 #define REG_WKFMCAM_CMD			0x0698
 #define REG_WKFMCAM_RWD			0x069c
-#define REG_RXFLTMAP0			0x06a0
-#define REG_RXFLTMAP1			0x06a2
-#define REG_RXFLTMAP2			0x06a4
+
+/*
+ * RX Filters: each bit corresponds to the numerical value of the subtype.
+ * If it is set the subtype frame type is passed. The filter is only used when
+ * the RCR_ACCEPT_DATA_FRAME, RCR_ACCEPT_CTRL_FRAME, RCR_ACCEPT_MGMT_FRAME bit
+ * in the RCR are low.
+ *
+ * Example: Beacon subtype is binary 1000 which is decimal 8 so we have to set
+ * bit 8 (0x100) in REG_RXFLTMAP0 to enable reception.
+ */
+#define REG_RXFLTMAP0			0x06a0	/* Management frames */
+#define REG_RXFLTMAP1			0x06a2	/* Control frames */
+#define REG_RXFLTMAP2			0x06a4	/* Data frames */
+
 #define REG_BCN_PSR_RPT			0x06a8
 #define REG_CALB32K_CTRL		0x06ac
 #define REG_PKT_MON_CTRL		0x06b4
-- 
cgit v0.10.2


From 124bc63c995d32b647560ca1951006f5954a53bb Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:53 -0500
Subject: rtl8xxxu: Remove duplicate USB ID

Reported-by: Xose Vazquez Perez <xose.vazquez@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index b58a60f..78683b5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5959,8 +5959,6 @@ static struct usb_device_id dev_table[] = {
 	.driver_info = (unsigned long)&rtl8192cu_fops},
 {USB_DEVICE_AND_INTERFACE_INFO(0xcdab, 0x8010, 0xff, 0xff, 0xff),
 	.driver_info = (unsigned long)&rtl8192cu_fops},
-{USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x317f, 0xff, 0xff, 0xff),
-	.driver_info = (unsigned long)&rtl8192cu_fops}, /* Netcore 8188RU */
 {USB_DEVICE_AND_INTERFACE_INFO(0x04f2, 0xaff7, 0xff, 0xff, 0xff),
 	.driver_info = (unsigned long)&rtl8192cu_fops},
 {USB_DEVICE_AND_INTERFACE_INFO(0x04f2, 0xaff9, 0xff, 0xff, 0xff),
-- 
cgit v0.10.2


From 755bda116e8a724f0e6b6514fbeb6d761b091b1e Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:54 -0500
Subject: rtl8xxxu: Fix 80 character per line issue

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 78683b5..9f89f62 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5355,8 +5355,9 @@ static void rtl8xxxu_configure_filter(struct ieee80211_hw *hw,
 
 	rtl8xxxu_write32(priv, REG_RCR, rcr);
 
-	*total_flags &= (FIF_ALLMULTI | FIF_FCSFAIL | FIF_BCN_PRBRESP_PROMISC | \
-			 FIF_CONTROL | FIF_OTHER_BSS | FIF_PSPOLL | FIF_PROBE_REQ);
+	*total_flags &= (FIF_ALLMULTI | FIF_FCSFAIL | FIF_BCN_PRBRESP_PROMISC |
+			 FIF_CONTROL | FIF_OTHER_BSS | FIF_PSPOLL |
+			 FIF_PROBE_REQ);
 }
 
 static int rtl8xxxu_set_rts_threshold(struct ieee80211_hw *hw, u32 rts)
-- 
cgit v0.10.2


From c85ea1156d787ef52bbb835bcf2d5117d16282e1 Mon Sep 17 00:00:00 2001
From: Bruno Randolf <br1@einfach.org>
Date: Wed, 3 Feb 2016 13:39:55 -0500
Subject: rtl8xxxu: Enable data frame reception in rtl8xxxu_start

mac80211 documentation says, the ieee80211_ops.start callback "must turn on
frame reception (for possibly enabled monitor interfaces.)". If not a single
monitor interface does not receive data frames.

Similarly we should not change the data reception based on the association
state.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 9f89f62..1b7ff74 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4521,9 +4521,6 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 			rtl8xxxu_update_rate_mask(priv, ramask, sgi);
 
-			/* Enable RX of data frames */
-			rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0xffff);
-
 			rtl8xxxu_write8(priv, REG_BCN_MAX_ERR, 0xff);
 
 			rtl8723a_stop_tx_beacon(priv);
@@ -4538,8 +4535,6 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			val8 |= BEACON_DISABLE_TSF_UPDATE;
 			rtl8xxxu_write8(priv, REG_BEACON_CTRL, val8);
 
-			/* Disable RX of data frames */
-			rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0x0000);
 			h2c.joinbss.data = H2C_JOIN_BSS_DISCONNECT;
 		}
 		h2c.joinbss.cmd = H2C_JOIN_BSS_REPORT;
@@ -5542,12 +5537,9 @@ static int rtl8xxxu_start(struct ieee80211_hw *hw)
 	}
 exit:
 	/*
-	 * Disable all data frames
-	 */
-	rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0x0000);
-	/*
-	 * Accept all mgmt frames
+	 * Accept all data and mgmt frames
 	 */
+	rtl8xxxu_write16(priv, REG_RXFLTMAP2, 0xffff);
 	rtl8xxxu_write16(priv, REG_RXFLTMAP0, 0xffff);
 
 	rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, 0x6954341e);
-- 
cgit v0.10.2


From a26703f39d2fe8f06ca2327d1a3f011d85b56c7f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:56 -0500
Subject: rtl8xxxu: rtl8xxxu_set_linktype(): Be consistent aboutregister size
 REG_MSR

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 1b7ff74..c8ba698 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1599,9 +1599,9 @@ rtl8723a_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 static void rtl8xxxu_set_linktype(struct rtl8xxxu_priv *priv,
 				  enum nl80211_iftype linktype)
 {
-	u16 val8;
+	u8 val8;
 
-	val8 = rtl8xxxu_read16(priv, REG_MSR);
+	val8 = rtl8xxxu_read8(priv, REG_MSR);
 	val8 &= ~MSR_LINKTYPE_MASK;
 
 	switch (linktype) {
-- 
cgit v0.10.2


From 53b381c4d72a58ca7df75074d75df9355b77fc99 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:57 -0500
Subject: rtl8xxxu: Set correct bit to reset MCU IO wrapper

All chips, except for 8812 require BIT(0) for resetting the MCU IO
wrapper. 8723b requires a more complicated reset sequence, so it will
need a custom reset function.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index c8ba698..460145c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2052,13 +2052,13 @@ static void rtl8xxxu_reset_8051(struct rtl8xxxu_priv *priv)
 	u16 sys_func;
 
 	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
-	val8 &= ~BIT(3);
+	val8 &= ~BIT(0);
 	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
 	sys_func = rtl8xxxu_read16(priv, REG_SYS_FUNC);
 	sys_func &= ~SYS_FUNC_CPU_ENABLE;
 	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
 	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
-	val8 |= BIT(3);
+	val8 |= BIT(0);
 	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
 	sys_func |= SYS_FUNC_CPU_ENABLE;
 	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
-- 
cgit v0.10.2


From 796c554257b1a8a61b1112ac03bbbf3221c5a65f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:58 -0500
Subject: rtl8xxxu: Remove unused mgmt variable from rtl8xxxu_rx_complete()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 460145c..1201836 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5031,7 +5031,6 @@ static void rtl8xxxu_rx_complete(struct urb *urb)
 	struct rtl8xxxu_rx_desc *rx_desc = (struct rtl8xxxu_rx_desc *)skb->data;
 	struct rtl8723au_phy_stats *phy_stats;
 	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
-	struct ieee80211_mgmt *mgmt;
 	struct device *dev = &priv->udev->dev;
 	__le32 *_rx_desc_le = (__le32 *)skb->data;
 	u32 *_rx_desc = (u32 *)skb->data;
@@ -5052,8 +5051,6 @@ static void rtl8xxxu_rx_complete(struct urb *urb)
 
 		skb_pull(skb, drvinfo_sz + desc_shift);
 
-		mgmt = (struct ieee80211_mgmt *)skb->data;
-
 		memset(rx_status, 0, sizeof(struct ieee80211_rx_status));
 
 		if (rx_desc->phy_stats)
-- 
cgit v0.10.2


From a9ffa615aa4c661ce905f73f1b2d43e17ee55194 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:39:59 -0500
Subject: rtl8xxxu: Remove unused len/cnt variables from rtl8xxxu_rx_complete()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 1201836..a6c8dab 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5034,13 +5034,11 @@ static void rtl8xxxu_rx_complete(struct urb *urb)
 	struct device *dev = &priv->udev->dev;
 	__le32 *_rx_desc_le = (__le32 *)skb->data;
 	u32 *_rx_desc = (u32 *)skb->data;
-	int cnt, len, drvinfo_sz, desc_shift, i;
+	int drvinfo_sz, desc_shift, i;
 
 	for (i = 0; i < (sizeof(struct rtl8xxxu_rx_desc) / sizeof(u32)); i++)
 		_rx_desc[i] = le32_to_cpu(_rx_desc_le[i]);
 
-	cnt = rx_desc->frag;
-	len = rx_desc->pktlen;
 	drvinfo_sz = rx_desc->drvinfo_sz * 8;
 	desc_shift = rx_desc->shift;
 	skb_put(skb, urb->actual_length);
-- 
cgit v0.10.2


From dc254ca41134d9bbc6227d133ef5e220f14529b5 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:40:00 -0500
Subject: rtl8xxxu: Do not mask RF registers to 6 bits

Newer chips have RF registers beyond the original 0x3f address.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index a6c8dab..af17ae7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2536,8 +2536,6 @@ static int rtl8xxxu_init_rf_regs(struct rtl8xxxu_priv *priv,
 			continue;
 		}
 
-		reg &= 0x3f;
-
 		ret = rtl8xxxu_write_rfreg(priv, path, reg, val);
 		if (ret) {
 			dev_warn(&priv->udev->dev,
-- 
cgit v0.10.2


From 4042e61735ce79c227a67a893d5738f297a96b9c Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:40:01 -0500
Subject: rtl8xxxu: Make device_init kludge 8723au only

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index af17ae7..fdeae3b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4343,14 +4343,17 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	val8 = ((30000 + NAV_UPPER_UNIT - 1) / NAV_UPPER_UNIT);
 	rtl8xxxu_write8(priv, REG_NAV_UPPER, val8);
 
-	/*
-	 * 2011/03/09 MH debug only, UMC-B cut pass 2500 S5 test,
-	 * but we need to fin root cause.
-	 */
-	val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
-	if ((val32 & 0xff000000) != 0x83000000) {
-		val32 |= FPGA_RF_MODE_CCK;
-		rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+	if (priv->rtlchip == 0x8723a) {
+		/*
+		 * 2011/03/09 MH debug only, UMC-B cut pass 2500 S5 test,
+		 * but we need to find root cause.
+		 * This is 8723au only.
+		 */
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
+		if ((val32 & 0xff000000) != 0x83000000) {
+			val32 |= FPGA_RF_MODE_CCK;
+			rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+		}
 	}
 
 	val32 = rtl8xxxu_read32(priv, REG_FWHW_TXQ_CTRL);
-- 
cgit v0.10.2


From 5b22a111aa15bb8a010a0bec1da6a4b6690dd561 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Wed, 3 Feb 2016 13:40:02 -0500
Subject: rtl8xxxu: Add missing bit define for REG_APS_FSMCO

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index d089edd..8f6c9c6 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -45,6 +45,7 @@
 #define  APS_FSMCO_ENABLE_POWERDOWN	BIT(4)
 #define  APS_FSMCO_MAC_ENABLE		BIT(8)
 #define  APS_FSMCO_MAC_OFF		BIT(9)
+#define  APS_FSMCO_SW_LPS		BIT(10)
 #define  APS_FSMCO_HW_SUSPEND		BIT(11)
 #define  APS_FSMCO_PCIE			BIT(12)
 #define  APS_FSMCO_HW_POWERDOWN		BIT(15)
-- 
cgit v0.10.2


From 19c8f421a61947116898c9f8a28823b9d988df74 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Mon, 8 Feb 2016 00:00:30 +0100
Subject: brcmfmac: Increase nr of supported flowrings.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

New generation devices have firmware which has more than 256 flowrings.
E.g. following debugging message comes from 14e4:4365 BCM4366:
[  194.606245] brcmfmac: brcmf_pcie_init_ringbuffers Nr of flowrings is 264

At various code places (related to flowrings) we were using u8 which
could lead to storing wrong number or infinite loops when indexing with
this type. This issue was quite easy to spot in brcmf_flowring_detach
where it led to infinite loop e.g. on failed initialization.

This patch switches code to proper types and increases the maximum
number of supported flowrings to 512.

Originally this change was sent in September 2015, but back it was
causing a regression on BCM43602 resulting in:
Unable to handle kernel NULL pointer dereference at virtual address ...

The reason for this regression was missing update (s/u8/u16) of struct
brcmf_flowring_ring. This problem was handled in 9f64df9 ("brcmfmac: Fix
bug in flowring management."). Starting with that it's safe to apply
this original patch as it doesn't cause a regression anymore.

This patch fixes an infinite loop on BCM4366 which is supported since
4.4 so it makes sense to apply it to stable 4.4+.

Cc: <stable@vger.kernel.org> # 4.4+
Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.c
index 2ca783f..7e269f9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.c
@@ -32,7 +32,7 @@
 #define BRCMF_FLOWRING_LOW		(BRCMF_FLOWRING_HIGH - 256)
 #define BRCMF_FLOWRING_INVALID_IFIDX	0xff
 
-#define BRCMF_FLOWRING_HASH_AP(da, fifo, ifidx) (da[5] + fifo + ifidx * 16)
+#define BRCMF_FLOWRING_HASH_AP(da, fifo, ifidx) (da[5] * 2 + fifo + ifidx * 16)
 #define BRCMF_FLOWRING_HASH_STA(fifo, ifidx) (fifo + ifidx * 16)
 
 static const u8 brcmf_flowring_prio2fifo[] = {
@@ -68,7 +68,7 @@ u32 brcmf_flowring_lookup(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 			  u8 prio, u8 ifidx)
 {
 	struct brcmf_flowring_hash *hash;
-	u8 hash_idx;
+	u16 hash_idx;
 	u32 i;
 	bool found;
 	bool sta;
@@ -88,6 +88,7 @@ u32 brcmf_flowring_lookup(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 	}
 	hash_idx =  sta ? BRCMF_FLOWRING_HASH_STA(fifo, ifidx) :
 			  BRCMF_FLOWRING_HASH_AP(mac, fifo, ifidx);
+	hash_idx &= (BRCMF_FLOWRING_HASHSIZE - 1);
 	found = false;
 	hash = flow->hash;
 	for (i = 0; i < BRCMF_FLOWRING_HASHSIZE; i++) {
@@ -98,6 +99,7 @@ u32 brcmf_flowring_lookup(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 			break;
 		}
 		hash_idx++;
+		hash_idx &= (BRCMF_FLOWRING_HASHSIZE - 1);
 	}
 	if (found)
 		return hash[hash_idx].flowid;
@@ -111,7 +113,7 @@ u32 brcmf_flowring_create(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 {
 	struct brcmf_flowring_ring *ring;
 	struct brcmf_flowring_hash *hash;
-	u8 hash_idx;
+	u16 hash_idx;
 	u32 i;
 	bool found;
 	u8 fifo;
@@ -131,6 +133,7 @@ u32 brcmf_flowring_create(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 	}
 	hash_idx =  sta ? BRCMF_FLOWRING_HASH_STA(fifo, ifidx) :
 			  BRCMF_FLOWRING_HASH_AP(mac, fifo, ifidx);
+	hash_idx &= (BRCMF_FLOWRING_HASHSIZE - 1);
 	found = false;
 	hash = flow->hash;
 	for (i = 0; i < BRCMF_FLOWRING_HASHSIZE; i++) {
@@ -140,6 +143,7 @@ u32 brcmf_flowring_create(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 			break;
 		}
 		hash_idx++;
+		hash_idx &= (BRCMF_FLOWRING_HASHSIZE - 1);
 	}
 	if (found) {
 		for (i = 0; i < flow->nrofrings; i++) {
@@ -169,7 +173,7 @@ u32 brcmf_flowring_create(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 }
 
 
-u8 brcmf_flowring_tid(struct brcmf_flowring *flow, u8 flowid)
+u8 brcmf_flowring_tid(struct brcmf_flowring *flow, u16 flowid)
 {
 	struct brcmf_flowring_ring *ring;
 
@@ -179,7 +183,7 @@ u8 brcmf_flowring_tid(struct brcmf_flowring *flow, u8 flowid)
 }
 
 
-static void brcmf_flowring_block(struct brcmf_flowring *flow, u8 flowid,
+static void brcmf_flowring_block(struct brcmf_flowring *flow, u16 flowid,
 				 bool blocked)
 {
 	struct brcmf_flowring_ring *ring;
@@ -228,10 +232,10 @@ static void brcmf_flowring_block(struct brcmf_flowring *flow, u8 flowid,
 }
 
 
-void brcmf_flowring_delete(struct brcmf_flowring *flow, u8 flowid)
+void brcmf_flowring_delete(struct brcmf_flowring *flow, u16 flowid)
 {
 	struct brcmf_flowring_ring *ring;
-	u8 hash_idx;
+	u16 hash_idx;
 	struct sk_buff *skb;
 
 	ring = flow->rings[flowid];
@@ -253,7 +257,7 @@ void brcmf_flowring_delete(struct brcmf_flowring *flow, u8 flowid)
 }
 
 
-u32 brcmf_flowring_enqueue(struct brcmf_flowring *flow, u8 flowid,
+u32 brcmf_flowring_enqueue(struct brcmf_flowring *flow, u16 flowid,
 			   struct sk_buff *skb)
 {
 	struct brcmf_flowring_ring *ring;
@@ -279,7 +283,7 @@ u32 brcmf_flowring_enqueue(struct brcmf_flowring *flow, u8 flowid,
 }
 
 
-struct sk_buff *brcmf_flowring_dequeue(struct brcmf_flowring *flow, u8 flowid)
+struct sk_buff *brcmf_flowring_dequeue(struct brcmf_flowring *flow, u16 flowid)
 {
 	struct brcmf_flowring_ring *ring;
 	struct sk_buff *skb;
@@ -300,7 +304,7 @@ struct sk_buff *brcmf_flowring_dequeue(struct brcmf_flowring *flow, u8 flowid)
 }
 
 
-void brcmf_flowring_reinsert(struct brcmf_flowring *flow, u8 flowid,
+void brcmf_flowring_reinsert(struct brcmf_flowring *flow, u16 flowid,
 			     struct sk_buff *skb)
 {
 	struct brcmf_flowring_ring *ring;
@@ -311,7 +315,7 @@ void brcmf_flowring_reinsert(struct brcmf_flowring *flow, u8 flowid,
 }
 
 
-u32 brcmf_flowring_qlen(struct brcmf_flowring *flow, u8 flowid)
+u32 brcmf_flowring_qlen(struct brcmf_flowring *flow, u16 flowid)
 {
 	struct brcmf_flowring_ring *ring;
 
@@ -326,7 +330,7 @@ u32 brcmf_flowring_qlen(struct brcmf_flowring *flow, u8 flowid)
 }
 
 
-void brcmf_flowring_open(struct brcmf_flowring *flow, u8 flowid)
+void brcmf_flowring_open(struct brcmf_flowring *flow, u16 flowid)
 {
 	struct brcmf_flowring_ring *ring;
 
@@ -340,10 +344,10 @@ void brcmf_flowring_open(struct brcmf_flowring *flow, u8 flowid)
 }
 
 
-u8 brcmf_flowring_ifidx_get(struct brcmf_flowring *flow, u8 flowid)
+u8 brcmf_flowring_ifidx_get(struct brcmf_flowring *flow, u16 flowid)
 {
 	struct brcmf_flowring_ring *ring;
-	u8 hash_idx;
+	u16 hash_idx;
 
 	ring = flow->rings[flowid];
 	hash_idx = ring->hash_id;
@@ -384,7 +388,7 @@ void brcmf_flowring_detach(struct brcmf_flowring *flow)
 	struct brcmf_pub *drvr = bus_if->drvr;
 	struct brcmf_flowring_tdls_entry *search;
 	struct brcmf_flowring_tdls_entry *remove;
-	u8 flowid;
+	u16 flowid;
 
 	for (flowid = 0; flowid < flow->nrofrings; flowid++) {
 		if (flow->rings[flowid])
@@ -408,7 +412,7 @@ void brcmf_flowring_configure_addr_mode(struct brcmf_flowring *flow, int ifidx,
 	struct brcmf_bus *bus_if = dev_get_drvdata(flow->dev);
 	struct brcmf_pub *drvr = bus_if->drvr;
 	u32 i;
-	u8 flowid;
+	u16 flowid;
 
 	if (flow->addr_mode[ifidx] != addr_mode) {
 		for (i = 0; i < ARRAY_SIZE(flow->hash); i++) {
@@ -434,7 +438,7 @@ void brcmf_flowring_delete_peer(struct brcmf_flowring *flow, int ifidx,
 	struct brcmf_flowring_tdls_entry *prev;
 	struct brcmf_flowring_tdls_entry *search;
 	u32 i;
-	u8 flowid;
+	u16 flowid;
 	bool sta;
 
 	sta = (flow->addr_mode[ifidx] == ADDR_INDIRECT);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.h
index 95fd1c9..068e68d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/flowring.h
@@ -16,7 +16,7 @@
 #define BRCMFMAC_FLOWRING_H
 
 
-#define BRCMF_FLOWRING_HASHSIZE		256
+#define BRCMF_FLOWRING_HASHSIZE		512		/* has to be 2^x */
 #define BRCMF_FLOWRING_INVALID_ID	0xFFFFFFFF
 
 
@@ -24,7 +24,7 @@ struct brcmf_flowring_hash {
 	u8 mac[ETH_ALEN];
 	u8 fifo;
 	u8 ifidx;
-	u8 flowid;
+	u16 flowid;
 };
 
 enum ring_status {
@@ -61,16 +61,16 @@ u32 brcmf_flowring_lookup(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 			  u8 prio, u8 ifidx);
 u32 brcmf_flowring_create(struct brcmf_flowring *flow, u8 da[ETH_ALEN],
 			  u8 prio, u8 ifidx);
-void brcmf_flowring_delete(struct brcmf_flowring *flow, u8 flowid);
-void brcmf_flowring_open(struct brcmf_flowring *flow, u8 flowid);
-u8 brcmf_flowring_tid(struct brcmf_flowring *flow, u8 flowid);
-u32 brcmf_flowring_enqueue(struct brcmf_flowring *flow, u8 flowid,
+void brcmf_flowring_delete(struct brcmf_flowring *flow, u16 flowid);
+void brcmf_flowring_open(struct brcmf_flowring *flow, u16 flowid);
+u8 brcmf_flowring_tid(struct brcmf_flowring *flow, u16 flowid);
+u32 brcmf_flowring_enqueue(struct brcmf_flowring *flow, u16 flowid,
 			   struct sk_buff *skb);
-struct sk_buff *brcmf_flowring_dequeue(struct brcmf_flowring *flow, u8 flowid);
-void brcmf_flowring_reinsert(struct brcmf_flowring *flow, u8 flowid,
+struct sk_buff *brcmf_flowring_dequeue(struct brcmf_flowring *flow, u16 flowid);
+void brcmf_flowring_reinsert(struct brcmf_flowring *flow, u16 flowid,
 			     struct sk_buff *skb);
-u32 brcmf_flowring_qlen(struct brcmf_flowring *flow, u8 flowid);
-u8 brcmf_flowring_ifidx_get(struct brcmf_flowring *flow, u8 flowid);
+u32 brcmf_flowring_qlen(struct brcmf_flowring *flow, u16 flowid);
+u8 brcmf_flowring_ifidx_get(struct brcmf_flowring *flow, u16 flowid);
 struct brcmf_flowring *brcmf_flowring_attach(struct device *dev, u16 nrofrings);
 void brcmf_flowring_detach(struct brcmf_flowring *flow);
 void brcmf_flowring_configure_addr_mode(struct brcmf_flowring *flow, int ifidx,
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c
index c2bdb91..9229667 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.c
@@ -677,7 +677,7 @@ static u32 brcmf_msgbuf_flowring_create(struct brcmf_msgbuf *msgbuf, int ifidx,
 }
 
 
-static void brcmf_msgbuf_txflow(struct brcmf_msgbuf *msgbuf, u8 flowid)
+static void brcmf_msgbuf_txflow(struct brcmf_msgbuf *msgbuf, u16 flowid)
 {
 	struct brcmf_flowring *flow = msgbuf->flow;
 	struct brcmf_commonring *commonring;
@@ -1310,7 +1310,7 @@ int brcmf_proto_msgbuf_rx_trigger(struct device *dev)
 }
 
 
-void brcmf_msgbuf_delete_flowring(struct brcmf_pub *drvr, u8 flowid)
+void brcmf_msgbuf_delete_flowring(struct brcmf_pub *drvr, u16 flowid)
 {
 	struct brcmf_msgbuf *msgbuf = (struct brcmf_msgbuf *)drvr->proto->pd;
 	struct msgbuf_tx_flowring_delete_req *delete;
@@ -1415,6 +1415,13 @@ int brcmf_proto_msgbuf_attach(struct brcmf_pub *drvr)
 	u32 count;
 
 	if_msgbuf = drvr->bus_if->msgbuf;
+
+	if (if_msgbuf->nrof_flowrings >= BRCMF_FLOWRING_HASHSIZE) {
+		brcmf_err("driver not configured for this many flowrings %d\n",
+			  if_msgbuf->nrof_flowrings);
+		if_msgbuf->nrof_flowrings = BRCMF_FLOWRING_HASHSIZE - 1;
+	}
+
 	msgbuf = kzalloc(sizeof(*msgbuf), GFP_KERNEL);
 	if (!msgbuf)
 		goto fail;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.h
index 3d513e4..ee6906a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/msgbuf.h
@@ -33,7 +33,7 @@
 
 
 int brcmf_proto_msgbuf_rx_trigger(struct device *dev);
-void brcmf_msgbuf_delete_flowring(struct brcmf_pub *drvr, u8 flowid);
+void brcmf_msgbuf_delete_flowring(struct brcmf_pub *drvr, u16 flowid);
 int brcmf_proto_msgbuf_attach(struct brcmf_pub *drvr);
 void brcmf_proto_msgbuf_detach(struct brcmf_pub *drvr);
 #else
-- 
cgit v0.10.2


From 10fbc7cf031d7253770331417a4adc4179c979c5 Mon Sep 17 00:00:00 2001
From: Lars-Peter Clausen <lars@metafoo.de>
Date: Mon, 8 Feb 2016 21:41:12 +0100
Subject: b43: Remove unnecessary synchronize_irq() before free_irq()

Calling synchronize_irq() right before free_irq() is quite useless. On one
hand the IRQ can easily fire again before free_irq() is entered, on the
other hand free_irq() itself calls synchronize_irq() internally (in a race
condition free way), before any state associated with the IRQ is freed.

Patch was generated using the following semantic patch:
// <smpl>
@@
expression irq;
@@
-synchronize_irq(irq);
 free_irq(irq, ...);
// </smpl>

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/b43/main.c b/drivers/net/wireless/broadcom/b43/main.c
index c279211..72380af 100644
--- a/drivers/net/wireless/broadcom/b43/main.c
+++ b/drivers/net/wireless/broadcom/b43/main.c
@@ -4375,12 +4375,10 @@ redo:
 	/* Synchronize and free the interrupt handlers. Unlock to avoid deadlocks. */
 	orig_dev = dev;
 	mutex_unlock(&wl->mutex);
-	if (b43_bus_host_is_sdio(dev->dev)) {
+	if (b43_bus_host_is_sdio(dev->dev))
 		b43_sdio_free_irq(dev);
-	} else {
-		synchronize_irq(dev->dev->irq);
+	else
 		free_irq(dev->dev->irq, dev);
-	}
 	mutex_lock(&wl->mutex);
 	dev = wl->current_dev;
 	if (!dev)
-- 
cgit v0.10.2


From 400b43c5826964598e06bd3373ea4523ce749c5e Mon Sep 17 00:00:00 2001
From: Lars-Peter Clausen <lars@metafoo.de>
Date: Mon, 8 Feb 2016 21:41:13 +0100
Subject: rtlwifi: Remove unnecessary synchronize_irq() before free_irq()

Calling synchronize_irq() right before free_irq() is quite useless. On one
hand the IRQ can easily fire again before free_irq() is entered, on the
other hand free_irq() itself calls synchronize_irq() internally (in a race
condition free way), before any state associated with the IRQ is freed.

Patch was generated using the following semantic patch:
// <smpl>
@@
expression irq;
@@
-synchronize_irq(irq);
 free_irq(irq, ...);
// </smpl>

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/pci.c b/drivers/net/wireless/realtek/rtlwifi/pci.c
index 7f471bf..140d254 100644
--- a/drivers/net/wireless/realtek/rtlwifi/pci.c
+++ b/drivers/net/wireless/realtek/rtlwifi/pci.c
@@ -2392,7 +2392,6 @@ void rtl_pci_disconnect(struct pci_dev *pdev)
 	rtlpriv->cfg->ops->deinit_sw_vars(hw);
 
 	if (rtlpci->irq_alloc) {
-		synchronize_irq(rtlpci->pdev->irq);
 		free_irq(rtlpci->pdev->irq, hw);
 		rtlpci->irq_alloc = 0;
 	}
-- 
cgit v0.10.2


From ccf5fa4f33ba6488579775fd3193739e70378b90 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 9 Feb 2016 14:15:36 +0300
Subject: mwifiex: fix an indenting mistake

We recently added an indent level here but missed this line.  It needs
another tab.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index cc07214..efb19e2 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -2452,7 +2452,7 @@ static void mwifiex_pcie_fw_dump(struct mwifiex_adapter *adapter)
 			for (i = 0; i < 4; i++) {
 				mwifiex_read_reg_byte(adapter, reg, &read_reg);
 				memory_size |= (read_reg << (i * 8));
-			reg++;
+				reg++;
 			}
 		} else {
 			memory_size = MWIFIEX_FW_DUMP_MAX_MEMSIZE;
-- 
cgit v0.10.2


From 9696a159c3989e34aa9a6e658dcb89c8fcda0a80 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 11 Feb 2016 10:53:09 -0600
Subject: rtlwifi: Prepare for reworking 5G channels

There are 3 drivers in this family that have 5G radios. Each of them
defines local copies of the available channels. This patch adds the
two arrays to the core driver.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c
index 4ae421e..02eba0e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/core.c
+++ b/drivers/net/wireless/realtek/rtlwifi/core.c
@@ -35,6 +35,22 @@
 #include <linux/export.h>
 #include <net/cfg80211.h>
 
+u8 channel5g[CHANNEL_MAX_NUMBER_5G] = {
+	36, 38, 40, 42, 44, 46, 48,		/* Band 1 */
+	52, 54, 56, 58, 60, 62, 64,		/* Band 2 */
+	100, 102, 104, 106, 108, 110, 112,	/* Band 3 */
+	116, 118, 120, 122, 124, 126, 128,	/* Band 3 */
+	132, 134, 136, 138, 140, 142, 144,	/* Band 3 */
+	149, 151, 153, 155, 157, 159, 161,	/* Band 4 */
+	165, 167, 169, 171, 173, 175, 177	/* Band 4 */
+};
+EXPORT_SYMBOL(channel5g);
+
+u8 channel5g_80m[CHANNEL_MAX_NUMBER_5G_80M] = {
+	42, 58, 106, 122, 138, 155, 171
+};
+EXPORT_SYMBOL(channel5g_80m);
+
 void rtl_addr_delay(u32 addr)
 {
 	if (addr == 0xfe)
diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index 4544752..e603819 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -122,11 +122,6 @@
 					    */
 #define CHANNEL_MAX_NUMBER_5G_80M	7
 #define CHANNEL_GROUP_MAX	(3 + 9)	/*  ch1~3, 4~9, 10~14 = three groups */
-#define CHANNEL_MAX_NUMBER_5G		54 /* Please refer to
-					    *"phy_GetChnlGroup8812A" and
-					    * "Hal_ReadTxPowerInfo8812A"
-					    */
-#define CHANNEL_MAX_NUMBER_5G_80M	7
 #define MAX_PG_GROUP			13
 #define	CHANNEL_GROUP_MAX_2G		3
 #define	CHANNEL_GROUP_IDX_5GL		3
@@ -2904,6 +2899,10 @@ value to host byte ordering.*/
 #define	STBC_VHT_TEST_TX_ENABLE			BIT(2)
 #define	STBC_VHT_CAP_TX				BIT(3)
 
+extern u8 channel5g[CHANNEL_MAX_NUMBER_5G];
+
+extern u8 channel5g_80m[CHANNEL_MAX_NUMBER_5G_80M];
+
 static inline u8 rtl_read_byte(struct rtl_priv *rtlpriv, u32 addr)
 {
 	return rtlpriv->io.read8_sync(rtlpriv, addr);
-- 
cgit v0.10.2


From bb6fa826ba308ea4db55629ae383e66ded01bfdd Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 11 Feb 2016 10:53:10 -0600
Subject: rtlwifi: rtl8192de: Convert driver to use common 5G channels

This driver defines its owh copy of the 5G channels. Change it to use
the common definitions.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c
index bb06fe8..7810fe8 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c
@@ -924,19 +924,11 @@ static void _rtl92d_ccxpower_index_check(struct ieee80211_hw *hw,
 
 static u8 _rtl92c_phy_get_rightchnlplace(u8 chnl)
 {
-	u8 channel_5g[59] = {
-		1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58,
-		60, 62, 64, 100, 102, 104, 106, 108, 110, 112,
-		114, 116, 118, 120, 122, 124, 126, 128,
-		130, 132, 134, 136, 138, 140, 149, 151,
-		153, 155, 157, 159, 161, 163, 165
-	};
 	u8 place = chnl;
 
 	if (chnl > 14) {
-		for (place = 14; place < sizeof(channel_5g); place++) {
-			if (channel_5g[place] == chnl) {
+		for (place = 14; place < sizeof(channel5g); place++) {
+			if (channel5g[place] == chnl) {
 				place++;
 				break;
 			}
@@ -2471,16 +2463,9 @@ static bool _rtl92d_is_legal_5g_channel(struct ieee80211_hw *hw, u8 channel)
 {
 
 	int i;
-	u8 channel_5g[45] = {
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58,
-		60, 62, 64, 100, 102, 104, 106, 108, 110, 112,
-		114, 116, 118, 120, 122, 124, 126, 128, 130, 132,
-		134, 136, 138, 140, 149, 151, 153, 155, 157, 159,
-		161, 163, 165
-	};
 
-	for (i = 0; i < sizeof(channel_5g); i++)
-		if (channel == channel_5g[i])
+	for (i = 0; i < sizeof(channel5g); i++)
+		if (channel == channel5g[i])
 			return true;
 	return false;
 }
-- 
cgit v0.10.2


From 2784b00aa316b430e9440087dd15c94557a27c4a Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 11 Feb 2016 10:53:11 -0600
Subject: rtlwifi: rtl8192ee: Convert driver to use new 5G channel tables

The driver defines its own set of channel tables for the 5G band. With
this change, it will use those of the core.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c
index 5f14308..9fd3f1b 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c
@@ -2018,18 +2018,6 @@ static void _rtl92ee_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 	struct rtl_efuse *efu = rtl_efuse(rtl_priv(hw));
 	struct txpower_info_2g pwr2g;
 	struct txpower_info_5g pwr5g;
-	u8 channel5g[CHANNEL_MAX_NUMBER_5G] = {
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54,
-		56, 58, 60, 62, 64, 100, 102, 104, 106,
-		108, 110, 112, 114, 116, 118, 120, 122,
-		124, 126, 128, 130, 132, 134, 136, 138,
-		140, 142, 144, 149, 151, 153, 155, 157,
-		159, 161, 163, 165, 167, 168, 169, 171,
-		173, 175, 177
-	};
-	u8 channel5g_80m[CHANNEL_MAX_NUMBER_5G_80M] = {
-		42, 58, 106, 122, 138, 155, 171
-	};
 	u8 rf, idx;
 	u8 i;
 
-- 
cgit v0.10.2


From 0a44b22009d52f93b10f14bb4765915e8bafb6a5 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 11 Feb 2016 10:53:12 -0600
Subject: rtlwifi: rtl8821ae: Convert driver to use common 5G channel table

There are several copies of the 5G channel tables in this driver. These
are removed so that the tables in the core will be used. This change
also removes a useless message of "Channel 163 in Group not found".

The number of possible 5G channels was reduced from 54 to a better
value of 49 during the conversion.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c
index bbb789f..5da9bd0 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c
@@ -2786,14 +2786,6 @@ static void _rtl8812ae_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 	struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw));
 	struct txpower_info_2g pwrinfo24g;
 	struct txpower_info_5g pwrinfo5g;
-	u8 channel5g[CHANNEL_MAX_NUMBER_5G] = {
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54,
-		56, 58, 60, 62, 64, 100, 102, 104, 106,
-		108, 110, 112, 114, 116, 118, 120, 122,
-		124, 126, 128, 130, 132, 134, 136, 138,
-		140, 142, 144, 149, 151, 153, 155, 157,
-		159, 161, 163, 165, 167, 168, 169, 171, 173, 175, 177};
-	u8 channel5g_80m[CHANNEL_MAX_NUMBER_5G_80M] = {42, 58, 106, 122, 138, 155, 171};
 	u8 rf_path, index;
 	u8 i;
 
@@ -2872,16 +2864,6 @@ static void _rtl8821ae_read_txpower_info_from_hwpg(struct ieee80211_hw *hw,
 	struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw));
 	struct txpower_info_2g pwrinfo24g;
 	struct txpower_info_5g pwrinfo5g;
-	u8 channel5g[CHANNEL_MAX_NUMBER_5G] = {
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54,
-		56, 58, 60, 62, 64, 100, 102, 104, 106,
-		108, 110, 112, 114, 116, 118, 120, 122,
-		124, 126, 128, 130, 132, 134, 136, 138,
-		140, 142, 144, 149, 151, 153, 155, 157,
-		159, 161, 163, 165, 167, 168, 169, 171,
-		173, 175, 177};
-	u8 channel5g_80m[CHANNEL_MAX_NUMBER_5G_80M] = {
-		42, 58, 106, 122, 138, 155, 171};
 	u8 rf_path, index;
 	u8 i;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c
index 9b4d8a6..74165b3 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c
@@ -1472,18 +1472,13 @@ static char _rtl8812ae_phy_get_chnl_idx_of_txpwr_lmt(struct ieee80211_hw *hw,
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	char channel_index = -1;
-	u8 channel_5g[CHANNEL_MAX_NUMBER_5G] = {
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58, 60, 62, 64,
-		100, 102, 104, 106, 108, 110, 112, 114, 116, 118, 120, 122,
-		124, 126, 128, 130, 132, 134, 136, 138, 140, 142, 144, 149,
-		151, 153, 155, 157, 159, 161, 163, 165, 167, 168, 169, 171,
-		173, 175, 177};
 	u8  i = 0;
+
 	if (band == BAND_ON_2_4G)
 		channel_index = channel - 1;
 	else if (band == BAND_ON_5G) {
-		for (i = 0; i < sizeof(channel_5g)/sizeof(u8); ++i) {
-			if (channel_5g[i] == channel)
+		for (i = 0; i < sizeof(channel5g)/sizeof(u8); ++i) {
+			if (channel5g[i] == channel)
 				channel_index = i;
 		}
 	} else
@@ -2240,13 +2235,6 @@ void rtl8821ae_phy_get_txpower_level(struct ieee80211_hw *hw, long *powerlevel)
 
 static bool _rtl8821ae_phy_get_chnl_index(u8 channel, u8 *chnl_index)
 {
-	u8 channel_5g[CHANNEL_MAX_NUMBER_5G] = {
-		36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58, 60, 62,
-		64, 100, 102, 104, 106, 108, 110, 112, 114, 116, 118,
-		120, 122, 124, 126, 128, 130, 132, 134, 136, 138, 140,
-		142, 144, 149, 151, 153, 155, 157, 159, 161, 163, 165,
-		167, 168, 169, 171, 173, 175, 177
-	};
 	u8 i = 0;
 	bool in_24g = true;
 
@@ -2257,7 +2245,7 @@ static bool _rtl8821ae_phy_get_chnl_index(u8 channel, u8 *chnl_index)
 		in_24g = false;
 
 		for (i = 0; i < CHANNEL_MAX_NUMBER_5G; ++i) {
-			if (channel_5g[i] == channel) {
+			if (channel5g[i] == channel) {
 				*chnl_index = i;
 				return in_24g;
 			}
@@ -2728,13 +2716,10 @@ static u8 _rtl8821ae_get_txpower_index(struct ieee80211_hw *hw, u8 path,
 			     rate <= DESC_RATEVHT2SS_MCS9))
 				txpower += rtlefuse->txpwr_5g_bw40diff[path][TX_2S];
 		} else if (bandwidth == HT_CHANNEL_WIDTH_80) {
-			u8 channel_5g_80m[CHANNEL_MAX_NUMBER_5G_80M] = {
-				42, 58, 106, 122, 138, 155, 171
-			};
 			u8 i;
 
-			for (i = 0; i < sizeof(channel_5g_80m) / sizeof(u8); ++i)
-				if (channel_5g_80m[i] == channel)
+			for (i = 0; i < sizeof(channel5g_80m) / sizeof(u8); ++i)
+				if (channel5g_80m[i] == channel)
 					index = i;
 
 			if ((DESC_RATEMCS0 <= rate && rate <= DESC_RATEMCS15) ||
diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index e603819..9e3cdd7 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -116,7 +116,7 @@
 
 #define CHANNEL_MAX_NUMBER	(14 + 24 + 21)	/* 14 is the max channel no */
 #define CHANNEL_MAX_NUMBER_2G		14
-#define CHANNEL_MAX_NUMBER_5G		54 /* Please refer to
+#define CHANNEL_MAX_NUMBER_5G		49 /* Please refer to
 					    *"phy_GetChnlGroup8812A" and
 					    * "Hal_ReadTxPowerInfo8812A"
 					    */
-- 
cgit v0.10.2


From 4fb37186f3b2b1f5d1a8f3e6b26b719bf2df6ee3 Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudip@vectorindia.org>
Date: Tue, 2 Feb 2016 12:44:40 +0530
Subject: ath10k: remove impossible code

len has been initialized with a value of 0 and buf_len with 4096. There
is no way that this condition (len > buf_len) can be true now.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 8d4148a..3dd75a2 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -2182,9 +2182,6 @@ static ssize_t ath10k_debug_fw_checksums_read(struct file *file,
 
 	mutex_lock(&ar->conf_mutex);
 
-	if (len > buf_len)
-		len = buf_len;
-
 	len += scnprintf(buf + len, buf_len - len,
 			 "firmware-N.bin\t\t%08x\n",
 			 crc32_le(0, ar->firmware->data, ar->firmware->size));
-- 
cgit v0.10.2


From 74135f599f9a2d0c54674640aa7d2423ab8ca98c Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 3 Feb 2016 21:07:42 +0530
Subject: ath10k: fix updating peer stats rx duration

We are not updating peer stats rx_duration periodically
unless the user one polls for fw_stats, this is because
we discard the update event since pdev list is empty. Fix
this by updating rx duration periodically irrepective of checks
for pdev list (irrespective of ping-pong response)

Fixes: 856e7c3 ("ath10k: add debugfs support for Per STA total rx duration")
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 3dd75a2..848a0dd 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -348,6 +348,9 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 	 */
 
 	peer_stats_svc = test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map);
+	if (peer_stats_svc)
+		ath10k_sta_update_rx_duration(ar, &stats.peers);
+
 	if (ar->debug.fw_stats_done && !peer_stats_svc) {
 		ath10k_warn(ar, "received unsolicited stats update event\n");
 		goto free;
@@ -384,9 +387,6 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 			goto free;
 		}
 
-		if (peer_stats_svc)
-			ath10k_sta_update_rx_duration(ar, &stats.peers);
-
 		list_splice_tail_init(&stats.peers, &ar->debug.fw_stats.peers);
 		list_splice_tail_init(&stats.vdevs, &ar->debug.fw_stats.vdevs);
 	}
-- 
cgit v0.10.2


From 1d4150c02c5709fdfd80f10368a31867de35e72e Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Mon, 22 Feb 2016 15:57:52 -0800
Subject: net_sched: prepare tcf_hashinfo_destroy() for netns support

We only release the memory of the hashtable itself, not its
entries inside. This is not a problem yet since we only call
it in module release path, and module is refcount'ed by
actions. This would be a problem after we move the per module
hinfo into per netns in the latter patch.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/act_api.h b/include/net/act_api.h
index 9d446f13..8c4e3ff 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -65,11 +65,6 @@ static inline int tcf_hashinfo_init(struct tcf_hashinfo *hf, unsigned int mask)
 	return 0;
 }
 
-static inline void tcf_hashinfo_destroy(struct tcf_hashinfo *hf)
-{
-	kfree(hf->htab);
-}
-
 /* Update lastuse only if needed, to avoid dirtying a cache line.
  * We use a temp variable to avoid fetching jiffies twice.
  */
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 06e7c4a..acafaf7 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -69,7 +69,7 @@ int __tcf_hash_release(struct tc_action *a, bool bind, bool strict)
 			if (a->ops->cleanup)
 				a->ops->cleanup(a, bind);
 			tcf_hash_destroy(a);
-			ret = 1;
+			ret = ACT_P_DELETED;
 		}
 	}
 
@@ -302,6 +302,32 @@ void tcf_hash_insert(struct tc_action *a)
 }
 EXPORT_SYMBOL(tcf_hash_insert);
 
+static void tcf_hashinfo_destroy(const struct tc_action_ops *ops)
+{
+	struct tcf_hashinfo *hinfo = ops->hinfo;
+	struct tc_action a = {
+		.ops = ops,
+	};
+	int i;
+
+	for (i = 0; i < hinfo->hmask + 1; i++) {
+		struct tcf_common *p;
+		struct hlist_node *n;
+
+		hlist_for_each_entry_safe(p, n, &hinfo->htab[i], tcfc_head) {
+			int ret;
+
+			a.priv = p;
+			ret = __tcf_hash_release(&a, false, true);
+			if (ret == ACT_P_DELETED)
+				module_put(ops->owner);
+			else if (ret < 0)
+				return;
+		}
+	}
+	kfree(hinfo->htab);
+}
+
 static LIST_HEAD(act_base);
 static DEFINE_RWLOCK(act_mod_lock);
 
@@ -333,7 +359,7 @@ int tcf_register_action(struct tc_action_ops *act, unsigned int mask)
 	list_for_each_entry(a, &act_base, head) {
 		if (act->type == a->type || (strcmp(act->kind, a->kind) == 0)) {
 			write_unlock(&act_mod_lock);
-			tcf_hashinfo_destroy(act->hinfo);
+			tcf_hashinfo_destroy(act);
 			kfree(act->hinfo);
 			return -EEXIST;
 		}
@@ -353,7 +379,7 @@ int tcf_unregister_action(struct tc_action_ops *act)
 	list_for_each_entry(a, &act_base, head) {
 		if (a == act) {
 			list_del(&act->head);
-			tcf_hashinfo_destroy(act->hinfo);
+			tcf_hashinfo_destroy(act);
 			kfree(act->hinfo);
 			err = 0;
 			break;
-- 
cgit v0.10.2


From ddf97ccdd7cb7e00daba465a5c947b8d941dc2a4 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Mon, 22 Feb 2016 15:57:53 -0800
Subject: net_sched: add network namespace support for tc actions

Currently tc actions are stored in a per-module hashtable,
therefore are visible to all network namespaces. This is
probably the last part of the tc subsystem which is not
aware of netns now. This patch makes them per-netns,
several tc action API's need to be adjusted for this.

The tc action API code is ugly due to historical reasons,
we need to refactor that code in the future.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/act_api.h b/include/net/act_api.h
index 8c4e3ff..342be6c 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -7,6 +7,8 @@
 
 #include <net/sch_generic.h>
 #include <net/pkt_sched.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
 
 struct tcf_common {
 	struct hlist_node		tcfc_head;
@@ -87,31 +89,65 @@ struct tc_action {
 	__u32			type; /* for backward compat(TCA_OLD_COMPAT) */
 	__u32			order;
 	struct list_head	list;
+	struct tcf_hashinfo	*hinfo;
 };
 
 struct tc_action_ops {
 	struct list_head head;
-	struct tcf_hashinfo *hinfo;
 	char    kind[IFNAMSIZ];
 	__u32   type; /* TBD to match kind */
 	struct module		*owner;
 	int     (*act)(struct sk_buff *, const struct tc_action *, struct tcf_result *);
 	int     (*dump)(struct sk_buff *, struct tc_action *, int, int);
 	void	(*cleanup)(struct tc_action *, int bind);
-	int     (*lookup)(struct tc_action *, u32);
+	int     (*lookup)(struct net *, struct tc_action *, u32);
 	int     (*init)(struct net *net, struct nlattr *nla,
 			struct nlattr *est, struct tc_action *act, int ovr,
 			int bind);
-	int     (*walk)(struct sk_buff *, struct netlink_callback *, int, struct tc_action *);
+	int     (*walk)(struct net *, struct sk_buff *,
+			struct netlink_callback *, int, struct tc_action *);
+};
+
+struct tc_action_net {
+	struct tcf_hashinfo *hinfo;
+	const struct tc_action_ops *ops;
 };
 
-int tcf_hash_search(struct tc_action *a, u32 index);
-u32 tcf_hash_new_index(struct tcf_hashinfo *hinfo);
-int tcf_hash_check(u32 index, struct tc_action *a, int bind);
-int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a,
-		    int size, int bind, bool cpustats);
+static inline
+int tc_action_net_init(struct tc_action_net *tn, const struct tc_action_ops *ops,
+		       unsigned int mask)
+{
+	int err = 0;
+
+	tn->hinfo = kmalloc(sizeof(*tn->hinfo), GFP_KERNEL);
+	if (!tn->hinfo)
+		return -ENOMEM;
+	tn->ops = ops;
+	err = tcf_hashinfo_init(tn->hinfo, mask);
+	if (err)
+		kfree(tn->hinfo);
+	return err;
+}
+
+void tcf_hashinfo_destroy(const struct tc_action_ops *ops,
+			  struct tcf_hashinfo *hinfo);
+
+static inline void tc_action_net_exit(struct tc_action_net *tn)
+{
+	tcf_hashinfo_destroy(tn->ops, tn->hinfo);
+}
+
+int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb,
+		       struct netlink_callback *cb, int type,
+		       struct tc_action *a);
+int tcf_hash_search(struct tc_action_net *tn, struct tc_action *a, u32 index);
+u32 tcf_hash_new_index(struct tc_action_net *tn);
+int tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a,
+		   int bind);
+int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est,
+		    struct tc_action *a, int size, int bind, bool cpustats);
 void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est);
-void tcf_hash_insert(struct tc_action *a);
+void tcf_hash_insert(struct tc_action_net *tn, struct tc_action *a);
 
 int __tcf_hash_release(struct tc_action *a, bool bind, bool strict);
 
@@ -120,8 +156,8 @@ static inline int tcf_hash_release(struct tc_action *a, bool bind)
 	return __tcf_hash_release(a, bind, false);
 }
 
-int tcf_register_action(struct tc_action_ops *a, unsigned int mask);
-int tcf_unregister_action(struct tc_action_ops *a);
+int tcf_register_action(struct tc_action_ops *a, struct pernet_operations *ops);
+int tcf_unregister_action(struct tc_action_ops *a, struct pernet_operations *ops);
 int tcf_action_destroy(struct list_head *actions, int bind);
 int tcf_action_exec(struct sk_buff *skb, const struct list_head *actions,
 		    struct tcf_result *res);
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index acafaf7..9606666 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -36,10 +36,9 @@ static void free_tcf(struct rcu_head *head)
 	kfree(p);
 }
 
-static void tcf_hash_destroy(struct tc_action *a)
+static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *a)
 {
 	struct tcf_common *p = a->priv;
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
 
 	spin_lock_bh(&hinfo->lock);
 	hlist_del(&p->tcfc_head);
@@ -68,7 +67,7 @@ int __tcf_hash_release(struct tc_action *a, bool bind, bool strict)
 		if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) {
 			if (a->ops->cleanup)
 				a->ops->cleanup(a, bind);
-			tcf_hash_destroy(a);
+			tcf_hash_destroy(a->hinfo, a);
 			ret = ACT_P_DELETED;
 		}
 	}
@@ -77,10 +76,9 @@ int __tcf_hash_release(struct tc_action *a, bool bind, bool strict)
 }
 EXPORT_SYMBOL(__tcf_hash_release);
 
-static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb,
-			   struct tc_action *a)
+static int tcf_dump_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb,
+			   struct netlink_callback *cb, struct tc_action *a)
 {
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
 	struct hlist_head *head;
 	struct tcf_common *p;
 	int err = 0, index = -1, i = 0, s_i = 0, n_i = 0;
@@ -126,9 +124,9 @@ nla_put_failure:
 	goto done;
 }
 
-static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a)
+static int tcf_del_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb,
+			  struct tc_action *a)
 {
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
 	struct hlist_head *head;
 	struct hlist_node *n;
 	struct tcf_common *p;
@@ -163,18 +161,24 @@ nla_put_failure:
 	return ret;
 }
 
-static int tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb,
-			      int type, struct tc_action *a)
+int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb,
+		       struct netlink_callback *cb, int type,
+		       struct tc_action *a)
 {
+	struct tcf_hashinfo *hinfo = tn->hinfo;
+
+	a->hinfo = hinfo;
+
 	if (type == RTM_DELACTION) {
-		return tcf_del_walker(skb, a);
+		return tcf_del_walker(hinfo, skb, a);
 	} else if (type == RTM_GETACTION) {
-		return tcf_dump_walker(skb, cb, a);
+		return tcf_dump_walker(hinfo, skb, cb, a);
 	} else {
 		WARN(1, "tcf_generic_walker: unknown action %d\n", type);
 		return -EINVAL;
 	}
 }
+EXPORT_SYMBOL(tcf_generic_walker);
 
 static struct tcf_common *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo)
 {
@@ -191,8 +195,9 @@ static struct tcf_common *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo)
 	return p;
 }
 
-u32 tcf_hash_new_index(struct tcf_hashinfo *hinfo)
+u32 tcf_hash_new_index(struct tc_action_net *tn)
 {
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	u32 val = hinfo->index;
 
 	do {
@@ -205,28 +210,31 @@ u32 tcf_hash_new_index(struct tcf_hashinfo *hinfo)
 }
 EXPORT_SYMBOL(tcf_hash_new_index);
 
-int tcf_hash_search(struct tc_action *a, u32 index)
+int tcf_hash_search(struct tc_action_net *tn, struct tc_action *a, u32 index)
 {
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	struct tcf_common *p = tcf_hash_lookup(index, hinfo);
 
 	if (p) {
 		a->priv = p;
+		a->hinfo = hinfo;
 		return 1;
 	}
 	return 0;
 }
 EXPORT_SYMBOL(tcf_hash_search);
 
-int tcf_hash_check(u32 index, struct tc_action *a, int bind)
+int tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a,
+		   int bind)
 {
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	struct tcf_common *p = NULL;
 	if (index && (p = tcf_hash_lookup(index, hinfo)) != NULL) {
 		if (bind)
 			p->tcfc_bindcnt++;
 		p->tcfc_refcnt++;
 		a->priv = p;
+		a->hinfo = hinfo;
 		return 1;
 	}
 	return 0;
@@ -243,11 +251,11 @@ void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est)
 }
 EXPORT_SYMBOL(tcf_hash_cleanup);
 
-int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a,
-		    int size, int bind, bool cpustats)
+int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est,
+		    struct tc_action *a, int size, int bind, bool cpustats)
 {
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
 	struct tcf_common *p = kzalloc(size, GFP_KERNEL);
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	int err = -ENOMEM;
 
 	if (unlikely(!p))
@@ -272,7 +280,7 @@ err2:
 	}
 	spin_lock_init(&p->tcfc_lock);
 	INIT_HLIST_NODE(&p->tcfc_head);
-	p->tcfc_index = index ? index : tcf_hash_new_index(hinfo);
+	p->tcfc_index = index ? index : tcf_hash_new_index(tn);
 	p->tcfc_tm.install = jiffies;
 	p->tcfc_tm.lastuse = jiffies;
 	if (est) {
@@ -286,14 +294,15 @@ err2:
 	}
 
 	a->priv = (void *) p;
+	a->hinfo = hinfo;
 	return 0;
 }
 EXPORT_SYMBOL(tcf_hash_create);
 
-void tcf_hash_insert(struct tc_action *a)
+void tcf_hash_insert(struct tc_action_net *tn, struct tc_action *a)
 {
 	struct tcf_common *p = a->priv;
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask);
 
 	spin_lock_bh(&hinfo->lock);
@@ -302,11 +311,12 @@ void tcf_hash_insert(struct tc_action *a)
 }
 EXPORT_SYMBOL(tcf_hash_insert);
 
-static void tcf_hashinfo_destroy(const struct tc_action_ops *ops)
+void tcf_hashinfo_destroy(const struct tc_action_ops *ops,
+			  struct tcf_hashinfo *hinfo)
 {
-	struct tcf_hashinfo *hinfo = ops->hinfo;
 	struct tc_action a = {
 		.ops = ops,
+		.hinfo = hinfo,
 	};
 	int i;
 
@@ -327,60 +337,52 @@ static void tcf_hashinfo_destroy(const struct tc_action_ops *ops)
 	}
 	kfree(hinfo->htab);
 }
+EXPORT_SYMBOL(tcf_hashinfo_destroy);
 
 static LIST_HEAD(act_base);
 static DEFINE_RWLOCK(act_mod_lock);
 
-int tcf_register_action(struct tc_action_ops *act, unsigned int mask)
+int tcf_register_action(struct tc_action_ops *act,
+			struct pernet_operations *ops)
 {
 	struct tc_action_ops *a;
-	int err;
+	int ret;
 
-	/* Must supply act, dump and init */
-	if (!act->act || !act->dump || !act->init)
+	if (!act->act || !act->dump || !act->init || !act->walk || !act->lookup)
 		return -EINVAL;
 
-	/* Supply defaults */
-	if (!act->lookup)
-		act->lookup = tcf_hash_search;
-	if (!act->walk)
-		act->walk = tcf_generic_walker;
-
-	act->hinfo = kmalloc(sizeof(struct tcf_hashinfo), GFP_KERNEL);
-	if (!act->hinfo)
-		return -ENOMEM;
-	err = tcf_hashinfo_init(act->hinfo, mask);
-	if (err) {
-		kfree(act->hinfo);
-		return err;
-	}
-
 	write_lock(&act_mod_lock);
 	list_for_each_entry(a, &act_base, head) {
 		if (act->type == a->type || (strcmp(act->kind, a->kind) == 0)) {
 			write_unlock(&act_mod_lock);
-			tcf_hashinfo_destroy(act);
-			kfree(act->hinfo);
 			return -EEXIST;
 		}
 	}
 	list_add_tail(&act->head, &act_base);
 	write_unlock(&act_mod_lock);
+
+	ret = register_pernet_subsys(ops);
+	if (ret) {
+		tcf_unregister_action(act, ops);
+		return ret;
+	}
+
 	return 0;
 }
 EXPORT_SYMBOL(tcf_register_action);
 
-int tcf_unregister_action(struct tc_action_ops *act)
+int tcf_unregister_action(struct tc_action_ops *act,
+			  struct pernet_operations *ops)
 {
 	struct tc_action_ops *a;
 	int err = -ENOENT;
 
+	unregister_pernet_subsys(ops);
+
 	write_lock(&act_mod_lock);
 	list_for_each_entry(a, &act_base, head) {
 		if (a == act) {
 			list_del(&act->head);
-			tcf_hashinfo_destroy(act);
-			kfree(act->hinfo);
 			err = 0;
 			break;
 		}
@@ -747,8 +749,8 @@ static struct tc_action *create_a(int i)
 	return act;
 }
 
-static struct tc_action *
-tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 portid)
+static struct tc_action *tcf_action_get_1(struct net *net, struct nlattr *nla,
+					  struct nlmsghdr *n, u32 portid)
 {
 	struct nlattr *tb[TCA_ACT_MAX + 1];
 	struct tc_action *a;
@@ -775,7 +777,7 @@ tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 portid)
 	if (a->ops == NULL) /* could happen in batch of actions */
 		goto err_free;
 	err = -ENOENT;
-	if (a->ops->lookup(a, index) == 0)
+	if (a->ops->lookup(net, a, index) == 0)
 		goto err_mod;
 
 	module_put(a->ops->owner);
@@ -845,7 +847,7 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,
 	if (nest == NULL)
 		goto out_module_put;
 
-	err = a.ops->walk(skb, &dcb, RTM_DELACTION, &a);
+	err = a.ops->walk(net, skb, &dcb, RTM_DELACTION, &a);
 	if (err < 0)
 		goto out_module_put;
 	if (err == 0)
@@ -923,7 +925,7 @@ tca_action_gd(struct net *net, struct nlattr *nla, struct nlmsghdr *n,
 	}
 
 	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) {
-		act = tcf_action_get_1(tb[i], n, portid);
+		act = tcf_action_get_1(net, tb[i], n, portid);
 		if (IS_ERR(act)) {
 			ret = PTR_ERR(act);
 			goto err;
@@ -1070,6 +1072,7 @@ find_dump_kind(const struct nlmsghdr *n)
 static int
 tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	struct net *net = sock_net(skb->sk);
 	struct nlmsghdr *nlh;
 	unsigned char *b = skb_tail_pointer(skb);
 	struct nlattr *nest;
@@ -1104,7 +1107,7 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)
 	if (nest == NULL)
 		goto out_module_put;
 
-	ret = a_o->walk(skb, cb, RTM_GETACTION, &a);
+	ret = a_o->walk(net, skb, cb, RTM_GETACTION, &a);
 	if (ret < 0)
 		goto out_module_put;
 
diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c
index 0bc6f91..8c9f1f0 100644
--- a/net/sched/act_bpf.c
+++ b/net/sched/act_bpf.c
@@ -33,6 +33,8 @@ struct tcf_bpf_cfg {
 	bool is_ebpf;
 };
 
+static int bpf_net_id;
+
 static int tcf_bpf(struct sk_buff *skb, const struct tc_action *act,
 		   struct tcf_result *res)
 {
@@ -275,6 +277,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla,
 			struct nlattr *est, struct tc_action *act,
 			int replace, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, bpf_net_id);
 	struct nlattr *tb[TCA_ACT_BPF_MAX + 1];
 	struct tcf_bpf_cfg cfg, old;
 	struct tc_act_bpf *parm;
@@ -294,8 +297,8 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla,
 
 	parm = nla_data(tb[TCA_ACT_BPF_PARMS]);
 
-	if (!tcf_hash_check(parm->index, act, bind)) {
-		ret = tcf_hash_create(parm->index, est, act,
+	if (!tcf_hash_check(tn, parm->index, act, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, act,
 				      sizeof(*prog), bind, true);
 		if (ret < 0)
 			return ret;
@@ -344,7 +347,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla,
 	rcu_assign_pointer(prog->filter, cfg.filter);
 
 	if (res == ACT_P_CREATED) {
-		tcf_hash_insert(act);
+		tcf_hash_insert(tn, act);
 	} else {
 		/* make sure the program being replaced is no longer executing */
 		synchronize_rcu();
@@ -367,6 +370,22 @@ static void tcf_bpf_cleanup(struct tc_action *act, int bind)
 	tcf_bpf_cfg_cleanup(&tmp);
 }
 
+static int tcf_bpf_walker(struct net *net, struct sk_buff *skb,
+			  struct netlink_callback *cb, int type,
+			  struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, bpf_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_bpf_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, bpf_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_bpf_ops __read_mostly = {
 	.kind		=	"bpf",
 	.type		=	TCA_ACT_BPF,
@@ -375,16 +394,39 @@ static struct tc_action_ops act_bpf_ops __read_mostly = {
 	.dump		=	tcf_bpf_dump,
 	.cleanup	=	tcf_bpf_cleanup,
 	.init		=	tcf_bpf_init,
+	.walk		=	tcf_bpf_walker,
+	.lookup		=	tcf_bpf_search,
+};
+
+static __net_init int bpf_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, bpf_net_id);
+
+	return tc_action_net_init(tn, &act_bpf_ops, BPF_TAB_MASK);
+}
+
+static void __net_exit bpf_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, bpf_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations bpf_net_ops = {
+	.init = bpf_init_net,
+	.exit = bpf_exit_net,
+	.id   = &bpf_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 static int __init bpf_init_module(void)
 {
-	return tcf_register_action(&act_bpf_ops, BPF_TAB_MASK);
+	return tcf_register_action(&act_bpf_ops, &bpf_net_ops);
 }
 
 static void __exit bpf_cleanup_module(void)
 {
-	tcf_unregister_action(&act_bpf_ops);
+	tcf_unregister_action(&act_bpf_ops, &bpf_net_ops);
 }
 
 module_init(bpf_init_module);
diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c
index bb41699..c0ed93c 100644
--- a/net/sched/act_connmark.c
+++ b/net/sched/act_connmark.c
@@ -30,6 +30,8 @@
 
 #define CONNMARK_TAB_MASK     3
 
+static int connmark_net_id;
+
 static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a,
 			struct tcf_result *res)
 {
@@ -97,6 +99,7 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla,
 			     struct nlattr *est, struct tc_action *a,
 			     int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, connmark_net_id);
 	struct nlattr *tb[TCA_CONNMARK_MAX + 1];
 	struct tcf_connmark_info *ci;
 	struct tc_connmark *parm;
@@ -111,9 +114,9 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla,
 
 	parm = nla_data(tb[TCA_CONNMARK_PARMS]);
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*ci),
-				      bind, false);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*ci), bind, false);
 		if (ret)
 			return ret;
 
@@ -122,7 +125,7 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla,
 		ci->net = net;
 		ci->zone = parm->zone;
 
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 		ret = ACT_P_CREATED;
 	} else {
 		ci = to_connmark(a);
@@ -169,6 +172,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_connmark_walker(struct net *net, struct sk_buff *skb,
+			       struct netlink_callback *cb, int type,
+			       struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, connmark_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_connmark_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, connmark_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_connmark_ops = {
 	.kind		=	"connmark",
 	.type		=	TCA_ACT_CONNMARK,
@@ -176,16 +195,39 @@ static struct tc_action_ops act_connmark_ops = {
 	.act		=	tcf_connmark,
 	.dump		=	tcf_connmark_dump,
 	.init		=	tcf_connmark_init,
+	.walk		=	tcf_connmark_walker,
+	.lookup		=	tcf_connmark_search,
+};
+
+static __net_init int connmark_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, connmark_net_id);
+
+	return tc_action_net_init(tn, &act_connmark_ops, CONNMARK_TAB_MASK);
+}
+
+static void __net_exit connmark_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, connmark_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations connmark_net_ops = {
+	.init = connmark_init_net,
+	.exit = connmark_exit_net,
+	.id   = &connmark_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 static int __init connmark_init_module(void)
 {
-	return tcf_register_action(&act_connmark_ops, CONNMARK_TAB_MASK);
+	return tcf_register_action(&act_connmark_ops, &connmark_net_ops);
 }
 
 static void __exit connmark_cleanup_module(void)
 {
-	tcf_unregister_action(&act_connmark_ops);
+	tcf_unregister_action(&act_connmark_ops, &connmark_net_ops);
 }
 
 module_init(connmark_init_module);
diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c
index eeb3eb3..d22426c 100644
--- a/net/sched/act_csum.c
+++ b/net/sched/act_csum.c
@@ -42,9 +42,13 @@ static const struct nla_policy csum_policy[TCA_CSUM_MAX + 1] = {
 	[TCA_CSUM_PARMS] = { .len = sizeof(struct tc_csum), },
 };
 
-static int tcf_csum_init(struct net *n, struct nlattr *nla, struct nlattr *est,
-			 struct tc_action *a, int ovr, int bind)
+static int csum_net_id;
+
+static int tcf_csum_init(struct net *net, struct nlattr *nla,
+			 struct nlattr *est, struct tc_action *a, int ovr,
+			 int bind)
 {
+	struct tc_action_net *tn = net_generic(net, csum_net_id);
 	struct nlattr *tb[TCA_CSUM_MAX + 1];
 	struct tc_csum *parm;
 	struct tcf_csum *p;
@@ -61,9 +65,9 @@ static int tcf_csum_init(struct net *n, struct nlattr *nla, struct nlattr *est,
 		return -EINVAL;
 	parm = nla_data(tb[TCA_CSUM_PARMS]);
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*p),
-				      bind, false);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*p), bind, false);
 		if (ret)
 			return ret;
 		ret = ACT_P_CREATED;
@@ -82,7 +86,7 @@ static int tcf_csum_init(struct net *n, struct nlattr *nla, struct nlattr *est,
 	spin_unlock_bh(&p->tcf_lock);
 
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 
 	return ret;
 }
@@ -555,6 +559,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_csum_walker(struct net *net, struct sk_buff *skb,
+			   struct netlink_callback *cb, int type,
+			   struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, csum_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_csum_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, csum_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_csum_ops = {
 	.kind		= "csum",
 	.type		= TCA_ACT_CSUM,
@@ -562,6 +582,29 @@ static struct tc_action_ops act_csum_ops = {
 	.act		= tcf_csum,
 	.dump		= tcf_csum_dump,
 	.init		= tcf_csum_init,
+	.walk		= tcf_csum_walker,
+	.lookup		= tcf_csum_search,
+};
+
+static __net_init int csum_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, csum_net_id);
+
+	return tc_action_net_init(tn, &act_csum_ops, CSUM_TAB_MASK);
+}
+
+static void __net_exit csum_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, csum_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations csum_net_ops = {
+	.init = csum_init_net,
+	.exit = csum_exit_net,
+	.id   = &csum_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_DESCRIPTION("Checksum updating actions");
@@ -569,12 +612,12 @@ MODULE_LICENSE("GPL");
 
 static int __init csum_init_module(void)
 {
-	return tcf_register_action(&act_csum_ops, CSUM_TAB_MASK);
+	return tcf_register_action(&act_csum_ops, &csum_net_ops);
 }
 
 static void __exit csum_cleanup_module(void)
 {
-	tcf_unregister_action(&act_csum_ops);
+	tcf_unregister_action(&act_csum_ops, &csum_net_ops);
 }
 
 module_init(csum_init_module);
diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c
index 5c1b051..887fc1f 100644
--- a/net/sched/act_gact.c
+++ b/net/sched/act_gact.c
@@ -25,6 +25,8 @@
 
 #define GACT_TAB_MASK	15
 
+static int gact_net_id;
+
 #ifdef CONFIG_GACT_PROB
 static int gact_net_rand(struct tcf_gact *gact)
 {
@@ -57,6 +59,7 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla,
 			 struct nlattr *est, struct tc_action *a,
 			 int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, gact_net_id);
 	struct nlattr *tb[TCA_GACT_MAX + 1];
 	struct tc_gact *parm;
 	struct tcf_gact *gact;
@@ -88,9 +91,9 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla,
 	}
 #endif
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*gact),
-				      bind, true);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*gact), bind, true);
 		if (ret)
 			return ret;
 		ret = ACT_P_CREATED;
@@ -118,7 +121,7 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla,
 	}
 #endif
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	return ret;
 }
 
@@ -183,6 +186,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_gact_walker(struct net *net, struct sk_buff *skb,
+			   struct netlink_callback *cb, int type,
+			   struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, gact_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_gact_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, gact_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_gact_ops = {
 	.kind		=	"gact",
 	.type		=	TCA_ACT_GACT,
@@ -190,6 +209,29 @@ static struct tc_action_ops act_gact_ops = {
 	.act		=	tcf_gact,
 	.dump		=	tcf_gact_dump,
 	.init		=	tcf_gact_init,
+	.walk		=	tcf_gact_walker,
+	.lookup		=	tcf_gact_search,
+};
+
+static __net_init int gact_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, gact_net_id);
+
+	return tc_action_net_init(tn, &act_gact_ops, GACT_TAB_MASK);
+}
+
+static void __net_exit gact_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, gact_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations gact_net_ops = {
+	.init = gact_init_net,
+	.exit = gact_exit_net,
+	.id   = &gact_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_AUTHOR("Jamal Hadi Salim(2002-4)");
@@ -203,12 +245,13 @@ static int __init gact_init_module(void)
 #else
 	pr_info("GACT probability NOT on\n");
 #endif
-	return tcf_register_action(&act_gact_ops, GACT_TAB_MASK);
+
+	return tcf_register_action(&act_gact_ops, &gact_net_ops);
 }
 
 static void __exit gact_cleanup_module(void)
 {
-	tcf_unregister_action(&act_gact_ops);
+	tcf_unregister_action(&act_gact_ops, &gact_net_ops);
 }
 
 module_init(gact_init_module);
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index d058696..89c41a1 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -30,6 +30,10 @@
 
 #define IPT_TAB_MASK     15
 
+static int ipt_net_id;
+
+static int xt_net_id;
+
 static int ipt_init_target(struct xt_entry_target *t, char *table, unsigned int hook)
 {
 	struct xt_tgchk_param par;
@@ -83,8 +87,9 @@ static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = {
 	[TCA_IPT_TARG]	= { .len = sizeof(struct xt_entry_target) },
 };
 
-static int tcf_ipt_init(struct net *net, struct nlattr *nla, struct nlattr *est,
-			struct tc_action *a, int ovr, int bind)
+static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla,
+			  struct nlattr *est, struct tc_action *a, int ovr,
+			  int bind)
 {
 	struct nlattr *tb[TCA_IPT_MAX + 1];
 	struct tcf_ipt *ipt;
@@ -113,8 +118,9 @@ static int tcf_ipt_init(struct net *net, struct nlattr *nla, struct nlattr *est,
 	if (tb[TCA_IPT_INDEX] != NULL)
 		index = nla_get_u32(tb[TCA_IPT_INDEX]);
 
-	if (!tcf_hash_check(index, a, bind) ) {
-		ret = tcf_hash_create(index, est, a, sizeof(*ipt), bind, false);
+	if (!tcf_hash_check(tn, index, a, bind)) {
+		ret = tcf_hash_create(tn, index, est, a, sizeof(*ipt), bind,
+				      false);
 		if (ret)
 			return ret;
 		ret = ACT_P_CREATED;
@@ -157,7 +163,7 @@ static int tcf_ipt_init(struct net *net, struct nlattr *nla, struct nlattr *est,
 	ipt->tcfi_hook  = hook;
 	spin_unlock_bh(&ipt->tcf_lock);
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	return ret;
 
 err3:
@@ -170,6 +176,24 @@ err1:
 	return err;
 }
 
+static int tcf_ipt_init(struct net *net, struct nlattr *nla,
+			struct nlattr *est, struct tc_action *a, int ovr,
+			int bind)
+{
+	struct tc_action_net *tn = net_generic(net, ipt_net_id);
+
+	return __tcf_ipt_init(tn, nla, est, a, ovr, bind);
+}
+
+static int tcf_xt_init(struct net *net, struct nlattr *nla,
+		       struct nlattr *est, struct tc_action *a, int ovr,
+		       int bind)
+{
+	struct tc_action_net *tn = net_generic(net, xt_net_id);
+
+	return __tcf_ipt_init(tn, nla, est, a, ovr, bind);
+}
+
 static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a,
 		   struct tcf_result *res)
 {
@@ -260,6 +284,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_ipt_walker(struct net *net, struct sk_buff *skb,
+			  struct netlink_callback *cb, int type,
+			  struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, ipt_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_ipt_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, ipt_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_ipt_ops = {
 	.kind		=	"ipt",
 	.type		=	TCA_ACT_IPT,
@@ -268,8 +308,47 @@ static struct tc_action_ops act_ipt_ops = {
 	.dump		=	tcf_ipt_dump,
 	.cleanup	=	tcf_ipt_release,
 	.init		=	tcf_ipt_init,
+	.walk		=	tcf_ipt_walker,
+	.lookup		=	tcf_ipt_search,
+};
+
+static __net_init int ipt_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, ipt_net_id);
+
+	return tc_action_net_init(tn, &act_ipt_ops, IPT_TAB_MASK);
+}
+
+static void __net_exit ipt_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, ipt_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations ipt_net_ops = {
+	.init = ipt_init_net,
+	.exit = ipt_exit_net,
+	.id   = &ipt_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
+static int tcf_xt_walker(struct net *net, struct sk_buff *skb,
+			 struct netlink_callback *cb, int type,
+			 struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, xt_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_xt_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, xt_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_xt_ops = {
 	.kind		=	"xt",
 	.type		=	TCA_ACT_XT,
@@ -277,7 +356,30 @@ static struct tc_action_ops act_xt_ops = {
 	.act		=	tcf_ipt,
 	.dump		=	tcf_ipt_dump,
 	.cleanup	=	tcf_ipt_release,
-	.init		=	tcf_ipt_init,
+	.init		=	tcf_xt_init,
+	.walk		=	tcf_xt_walker,
+	.lookup		=	tcf_xt_search,
+};
+
+static __net_init int xt_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, xt_net_id);
+
+	return tc_action_net_init(tn, &act_xt_ops, IPT_TAB_MASK);
+}
+
+static void __net_exit xt_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, xt_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations xt_net_ops = {
+	.init = xt_init_net,
+	.exit = xt_exit_net,
+	.id   = &xt_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_AUTHOR("Jamal Hadi Salim(2002-13)");
@@ -289,12 +391,13 @@ static int __init ipt_init_module(void)
 {
 	int ret1, ret2;
 
-	ret1 = tcf_register_action(&act_xt_ops, IPT_TAB_MASK);
+	ret1 = tcf_register_action(&act_xt_ops, &xt_net_ops);
 	if (ret1 < 0)
-		printk("Failed to load xt action\n");
-	ret2 = tcf_register_action(&act_ipt_ops, IPT_TAB_MASK);
+		pr_err("Failed to load xt action\n");
+
+	ret2 = tcf_register_action(&act_ipt_ops, &ipt_net_ops);
 	if (ret2 < 0)
-		printk("Failed to load ipt action\n");
+		pr_err("Failed to load ipt action\n");
 
 	if (ret1 < 0 && ret2 < 0) {
 		return ret1;
@@ -304,8 +407,8 @@ static int __init ipt_init_module(void)
 
 static void __exit ipt_cleanup_module(void)
 {
-	tcf_unregister_action(&act_xt_ops);
-	tcf_unregister_action(&act_ipt_ops);
+	tcf_unregister_action(&act_ipt_ops, &ipt_net_ops);
+	tcf_unregister_action(&act_xt_ops, &xt_net_ops);
 }
 
 module_init(ipt_init_module);
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index 32fcdec..6b284d9 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -50,10 +50,13 @@ static const struct nla_policy mirred_policy[TCA_MIRRED_MAX + 1] = {
 	[TCA_MIRRED_PARMS]	= { .len = sizeof(struct tc_mirred) },
 };
 
+static int mirred_net_id;
+
 static int tcf_mirred_init(struct net *net, struct nlattr *nla,
 			   struct nlattr *est, struct tc_action *a, int ovr,
 			   int bind)
 {
+	struct tc_action_net *tn = net_generic(net, mirred_net_id);
 	struct nlattr *tb[TCA_MIRRED_MAX + 1];
 	struct tc_mirred *parm;
 	struct tcf_mirred *m;
@@ -96,11 +99,11 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla,
 		dev = NULL;
 	}
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
 		if (dev == NULL)
 			return -EINVAL;
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*m),
-				      bind, true);
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*m), bind, true);
 		if (ret)
 			return ret;
 		ret = ACT_P_CREATED;
@@ -130,7 +133,7 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla,
 		spin_lock_bh(&mirred_list_lock);
 		list_add(&m->tcfm_list, &mirred_list);
 		spin_unlock_bh(&mirred_list_lock);
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	}
 
 	return ret;
@@ -221,6 +224,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_mirred_walker(struct net *net, struct sk_buff *skb,
+			     struct netlink_callback *cb, int type,
+			     struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, mirred_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_mirred_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, mirred_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static int mirred_device_event(struct notifier_block *unused,
 			       unsigned long event, void *ptr)
 {
@@ -257,6 +276,29 @@ static struct tc_action_ops act_mirred_ops = {
 	.dump		=	tcf_mirred_dump,
 	.cleanup	=	tcf_mirred_release,
 	.init		=	tcf_mirred_init,
+	.walk		=	tcf_mirred_walker,
+	.lookup		=	tcf_mirred_search,
+};
+
+static __net_init int mirred_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, mirred_net_id);
+
+	return tc_action_net_init(tn, &act_mirred_ops, MIRRED_TAB_MASK);
+}
+
+static void __net_exit mirred_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, mirred_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations mirred_net_ops = {
+	.init = mirred_init_net,
+	.exit = mirred_exit_net,
+	.id   = &mirred_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_AUTHOR("Jamal Hadi Salim(2002)");
@@ -270,12 +312,12 @@ static int __init mirred_init_module(void)
 		return err;
 
 	pr_info("Mirror/redirect action on\n");
-	return tcf_register_action(&act_mirred_ops, MIRRED_TAB_MASK);
+	return tcf_register_action(&act_mirred_ops, &mirred_net_ops);
 }
 
 static void __exit mirred_cleanup_module(void)
 {
-	tcf_unregister_action(&act_mirred_ops);
+	tcf_unregister_action(&act_mirred_ops, &mirred_net_ops);
 	unregister_netdevice_notifier(&mirred_device_notifier);
 }
 
diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c
index 27607b8..0f65cdf 100644
--- a/net/sched/act_nat.c
+++ b/net/sched/act_nat.c
@@ -31,6 +31,8 @@
 
 #define NAT_TAB_MASK	15
 
+static int nat_net_id;
+
 static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
 	[TCA_NAT_PARMS]	= { .len = sizeof(struct tc_nat) },
 };
@@ -38,6 +40,7 @@ static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
 static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est,
 			struct tc_action *a, int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, nat_net_id);
 	struct nlattr *tb[TCA_NAT_MAX + 1];
 	struct tc_nat *parm;
 	int ret = 0, err;
@@ -54,9 +57,9 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est,
 		return -EINVAL;
 	parm = nla_data(tb[TCA_NAT_PARMS]);
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*p),
-				      bind, false);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*p), bind, false);
 		if (ret)
 			return ret;
 		ret = ACT_P_CREATED;
@@ -79,7 +82,7 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est,
 	spin_unlock_bh(&p->tcf_lock);
 
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 
 	return ret;
 }
@@ -274,6 +277,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_nat_walker(struct net *net, struct sk_buff *skb,
+			  struct netlink_callback *cb, int type,
+			  struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, nat_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_nat_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, nat_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_nat_ops = {
 	.kind		=	"nat",
 	.type		=	TCA_ACT_NAT,
@@ -281,6 +300,29 @@ static struct tc_action_ops act_nat_ops = {
 	.act		=	tcf_nat,
 	.dump		=	tcf_nat_dump,
 	.init		=	tcf_nat_init,
+	.walk		=	tcf_nat_walker,
+	.lookup		=	tcf_nat_search,
+};
+
+static __net_init int nat_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, nat_net_id);
+
+	return tc_action_net_init(tn, &act_nat_ops, NAT_TAB_MASK);
+}
+
+static void __net_exit nat_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, nat_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations nat_net_ops = {
+	.init = nat_init_net,
+	.exit = nat_exit_net,
+	.id   = &nat_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_DESCRIPTION("Stateless NAT actions");
@@ -288,12 +330,12 @@ MODULE_LICENSE("GPL");
 
 static int __init nat_init_module(void)
 {
-	return tcf_register_action(&act_nat_ops, NAT_TAB_MASK);
+	return tcf_register_action(&act_nat_ops, &nat_net_ops);
 }
 
 static void __exit nat_cleanup_module(void)
 {
-	tcf_unregister_action(&act_nat_ops);
+	tcf_unregister_action(&act_nat_ops, &nat_net_ops);
 }
 
 module_init(nat_init_module);
diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c
index e38a770..429c3ab 100644
--- a/net/sched/act_pedit.c
+++ b/net/sched/act_pedit.c
@@ -25,6 +25,8 @@
 
 #define PEDIT_TAB_MASK	15
 
+static int pedit_net_id;
+
 static const struct nla_policy pedit_policy[TCA_PEDIT_MAX + 1] = {
 	[TCA_PEDIT_PARMS]	= { .len = sizeof(struct tc_pedit) },
 };
@@ -33,6 +35,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla,
 			  struct nlattr *est, struct tc_action *a,
 			  int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, pedit_net_id);
 	struct nlattr *tb[TCA_PEDIT_MAX + 1];
 	struct tc_pedit *parm;
 	int ret = 0, err;
@@ -54,11 +57,11 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla,
 	if (nla_len(tb[TCA_PEDIT_PARMS]) < sizeof(*parm) + ksize)
 		return -EINVAL;
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
 		if (!parm->nkeys)
 			return -EINVAL;
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*p),
-				      bind, false);
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*p), bind, false);
 		if (ret)
 			return ret;
 		p = to_pedit(a);
@@ -93,7 +96,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla,
 	memcpy(p->tcfp_keys, parm->keys, ksize);
 	spin_unlock_bh(&p->tcf_lock);
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	return ret;
 }
 
@@ -211,6 +214,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_pedit_walker(struct net *net, struct sk_buff *skb,
+			    struct netlink_callback *cb, int type,
+			    struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, pedit_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_pedit_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, pedit_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_pedit_ops = {
 	.kind		=	"pedit",
 	.type		=	TCA_ACT_PEDIT,
@@ -219,6 +238,29 @@ static struct tc_action_ops act_pedit_ops = {
 	.dump		=	tcf_pedit_dump,
 	.cleanup	=	tcf_pedit_cleanup,
 	.init		=	tcf_pedit_init,
+	.walk		=	tcf_pedit_walker,
+	.lookup		=	tcf_pedit_search,
+};
+
+static __net_init int pedit_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, pedit_net_id);
+
+	return tc_action_net_init(tn, &act_pedit_ops, PEDIT_TAB_MASK);
+}
+
+static void __net_exit pedit_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, pedit_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations pedit_net_ops = {
+	.init = pedit_init_net,
+	.exit = pedit_exit_net,
+	.id   = &pedit_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_AUTHOR("Jamal Hadi Salim(2002-4)");
@@ -227,12 +269,12 @@ MODULE_LICENSE("GPL");
 
 static int __init pedit_init_module(void)
 {
-	return tcf_register_action(&act_pedit_ops, PEDIT_TAB_MASK);
+	return tcf_register_action(&act_pedit_ops, &pedit_net_ops);
 }
 
 static void __exit pedit_cleanup_module(void)
 {
-	tcf_unregister_action(&act_pedit_ops);
+	tcf_unregister_action(&act_pedit_ops, &pedit_net_ops);
 }
 
 module_init(pedit_init_module);
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index 9a1c42a..330f14e 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -55,10 +55,14 @@ struct tc_police_compat {
 
 /* Each policer is serialized by its individual spinlock */
 
-static int tcf_act_police_walker(struct sk_buff *skb, struct netlink_callback *cb,
-			      int type, struct tc_action *a)
+static int police_net_id;
+
+static int tcf_act_police_walker(struct net *net, struct sk_buff *skb,
+				 struct netlink_callback *cb, int type,
+				 struct tc_action *a)
 {
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+	struct tc_action_net *tn = net_generic(net, police_net_id);
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	struct hlist_head *head;
 	struct tcf_common *p;
 	int err = 0, index = -1, i = 0, s_i = 0, n_i = 0;
@@ -121,7 +125,8 @@ static int tcf_act_police_locate(struct net *net, struct nlattr *nla,
 	struct tc_police *parm;
 	struct tcf_police *police;
 	struct qdisc_rate_table *R_tab = NULL, *P_tab = NULL;
-	struct tcf_hashinfo *hinfo = a->ops->hinfo;
+	struct tc_action_net *tn = net_generic(net, police_net_id);
+	struct tcf_hashinfo *hinfo = tn->hinfo;
 	int size;
 
 	if (nla == NULL)
@@ -139,7 +144,7 @@ static int tcf_act_police_locate(struct net *net, struct nlattr *nla,
 	parm = nla_data(tb[TCA_POLICE_TBF]);
 
 	if (parm->index) {
-		if (tcf_hash_search(a, parm->index)) {
+		if (tcf_hash_search(tn, a, parm->index)) {
 			police = to_police(a->priv);
 			if (bind) {
 				police->tcf_bindcnt += 1;
@@ -233,7 +238,7 @@ override:
 
 	police->tcfp_t_c = ktime_get_ns();
 	police->tcf_index = parm->index ? parm->index :
-		tcf_hash_new_index(hinfo);
+		tcf_hash_new_index(tn);
 	h = tcf_hash(police->tcf_index, POL_TAB_MASK);
 	spin_lock_bh(&hinfo->lock);
 	hlist_add_head(&police->tcf_head, &hinfo->htab[h]);
@@ -342,6 +347,13 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_police_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, police_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 MODULE_AUTHOR("Alexey Kuznetsov");
 MODULE_DESCRIPTION("Policing actions");
 MODULE_LICENSE("GPL");
@@ -353,19 +365,41 @@ static struct tc_action_ops act_police_ops = {
 	.act		=	tcf_act_police,
 	.dump		=	tcf_act_police_dump,
 	.init		=	tcf_act_police_locate,
-	.walk		=	tcf_act_police_walker
+	.walk		=	tcf_act_police_walker,
+	.lookup		=	tcf_police_search,
+};
+
+static __net_init int police_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, police_net_id);
+
+	return tc_action_net_init(tn, &act_police_ops, POL_TAB_MASK);
+}
+
+static void __net_exit police_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, police_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations police_net_ops = {
+	.init = police_init_net,
+	.exit = police_exit_net,
+	.id   = &police_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 static int __init
 police_init_module(void)
 {
-	return tcf_register_action(&act_police_ops, POL_TAB_MASK);
+	return tcf_register_action(&act_police_ops, &police_net_ops);
 }
 
 static void __exit
 police_cleanup_module(void)
 {
-	tcf_unregister_action(&act_police_ops);
+	tcf_unregister_action(&act_police_ops, &police_net_ops);
 }
 
 module_init(police_init_module);
diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c
index d6b708d..75b2be1 100644
--- a/net/sched/act_simple.c
+++ b/net/sched/act_simple.c
@@ -26,6 +26,8 @@
 
 #define SIMP_TAB_MASK     7
 
+static int simp_net_id;
+
 #define SIMP_MAX_DATA	32
 static int tcf_simp(struct sk_buff *skb, const struct tc_action *a,
 		    struct tcf_result *res)
@@ -80,6 +82,7 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla,
 			 struct nlattr *est, struct tc_action *a,
 			 int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, simp_net_id);
 	struct nlattr *tb[TCA_DEF_MAX + 1];
 	struct tc_defact *parm;
 	struct tcf_defact *d;
@@ -102,9 +105,9 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla,
 	parm = nla_data(tb[TCA_DEF_PARMS]);
 	defdata = nla_data(tb[TCA_DEF_DATA]);
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*d),
-				      bind, false);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*d), bind, false);
 		if (ret)
 			return ret;
 
@@ -129,7 +132,7 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla,
 	}
 
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	return ret;
 }
 
@@ -161,6 +164,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_simp_walker(struct net *net, struct sk_buff *skb,
+			   struct netlink_callback *cb, int type,
+			   struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, simp_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_simp_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, simp_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_simp_ops = {
 	.kind		=	"simple",
 	.type		=	TCA_ACT_SIMP,
@@ -169,6 +188,29 @@ static struct tc_action_ops act_simp_ops = {
 	.dump		=	tcf_simp_dump,
 	.cleanup	=	tcf_simp_release,
 	.init		=	tcf_simp_init,
+	.walk		=	tcf_simp_walker,
+	.lookup		=	tcf_simp_search,
+};
+
+static __net_init int simp_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, simp_net_id);
+
+	return tc_action_net_init(tn, &act_simp_ops, SIMP_TAB_MASK);
+}
+
+static void __net_exit simp_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, simp_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations simp_net_ops = {
+	.init = simp_init_net,
+	.exit = simp_exit_net,
+	.id   = &simp_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_AUTHOR("Jamal Hadi Salim(2005)");
@@ -177,8 +219,7 @@ MODULE_LICENSE("GPL");
 
 static int __init simp_init_module(void)
 {
-	int ret;
-	ret = tcf_register_action(&act_simp_ops, SIMP_TAB_MASK);
+	int ret = tcf_register_action(&act_simp_ops, &simp_net_ops);
 	if (!ret)
 		pr_info("Simple TC action Loaded\n");
 	return ret;
@@ -186,7 +227,7 @@ static int __init simp_init_module(void)
 
 static void __exit simp_cleanup_module(void)
 {
-	tcf_unregister_action(&act_simp_ops);
+	tcf_unregister_action(&act_simp_ops, &simp_net_ops);
 }
 
 module_init(simp_init_module);
diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c
index 6751b5f..cfcdbdc 100644
--- a/net/sched/act_skbedit.c
+++ b/net/sched/act_skbedit.c
@@ -29,6 +29,8 @@
 
 #define SKBEDIT_TAB_MASK     15
 
+static int skbedit_net_id;
+
 static int tcf_skbedit(struct sk_buff *skb, const struct tc_action *a,
 		       struct tcf_result *res)
 {
@@ -61,6 +63,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla,
 			    struct nlattr *est, struct tc_action *a,
 			    int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, skbedit_net_id);
 	struct nlattr *tb[TCA_SKBEDIT_MAX + 1];
 	struct tc_skbedit *parm;
 	struct tcf_skbedit *d;
@@ -98,9 +101,9 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla,
 
 	parm = nla_data(tb[TCA_SKBEDIT_PARMS]);
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*d),
-				      bind, false);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*d), bind, false);
 		if (ret)
 			return ret;
 
@@ -130,7 +133,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla,
 	spin_unlock_bh(&d->tcf_lock);
 
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	return ret;
 }
 
@@ -173,6 +176,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_skbedit_walker(struct net *net, struct sk_buff *skb,
+			      struct netlink_callback *cb, int type,
+			      struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, skbedit_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_skbedit_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, skbedit_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_skbedit_ops = {
 	.kind		=	"skbedit",
 	.type		=	TCA_ACT_SKBEDIT,
@@ -180,6 +199,29 @@ static struct tc_action_ops act_skbedit_ops = {
 	.act		=	tcf_skbedit,
 	.dump		=	tcf_skbedit_dump,
 	.init		=	tcf_skbedit_init,
+	.walk		=	tcf_skbedit_walker,
+	.lookup		=	tcf_skbedit_search,
+};
+
+static __net_init int skbedit_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, skbedit_net_id);
+
+	return tc_action_net_init(tn, &act_skbedit_ops, SKBEDIT_TAB_MASK);
+}
+
+static void __net_exit skbedit_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, skbedit_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations skbedit_net_ops = {
+	.init = skbedit_init_net,
+	.exit = skbedit_exit_net,
+	.id   = &skbedit_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 MODULE_AUTHOR("Alexander Duyck, <alexander.h.duyck@intel.com>");
@@ -188,12 +230,12 @@ MODULE_LICENSE("GPL");
 
 static int __init skbedit_init_module(void)
 {
-	return tcf_register_action(&act_skbedit_ops, SKBEDIT_TAB_MASK);
+	return tcf_register_action(&act_skbedit_ops, &skbedit_net_ops);
 }
 
 static void __exit skbedit_cleanup_module(void)
 {
-	tcf_unregister_action(&act_skbedit_ops);
+	tcf_unregister_action(&act_skbedit_ops, &skbedit_net_ops);
 }
 
 module_init(skbedit_init_module);
diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c
index 796785e..bab8ae0 100644
--- a/net/sched/act_vlan.c
+++ b/net/sched/act_vlan.c
@@ -21,6 +21,8 @@
 
 #define VLAN_TAB_MASK     15
 
+static int vlan_net_id;
+
 static int tcf_vlan(struct sk_buff *skb, const struct tc_action *a,
 		    struct tcf_result *res)
 {
@@ -68,6 +70,7 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla,
 			 struct nlattr *est, struct tc_action *a,
 			 int ovr, int bind)
 {
+	struct tc_action_net *tn = net_generic(net, vlan_net_id);
 	struct nlattr *tb[TCA_VLAN_MAX + 1];
 	struct tc_vlan *parm;
 	struct tcf_vlan *v;
@@ -115,9 +118,9 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla,
 	}
 	action = parm->v_action;
 
-	if (!tcf_hash_check(parm->index, a, bind)) {
-		ret = tcf_hash_create(parm->index, est, a, sizeof(*v),
-				      bind, false);
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a,
+				      sizeof(*v), bind, false);
 		if (ret)
 			return ret;
 
@@ -143,7 +146,7 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla,
 	spin_unlock_bh(&v->tcf_lock);
 
 	if (ret == ACT_P_CREATED)
-		tcf_hash_insert(a);
+		tcf_hash_insert(tn, a);
 	return ret;
 }
 
@@ -181,6 +184,22 @@ nla_put_failure:
 	return -1;
 }
 
+static int tcf_vlan_walker(struct net *net, struct sk_buff *skb,
+			   struct netlink_callback *cb, int type,
+			   struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, vlan_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_vlan_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, vlan_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
 static struct tc_action_ops act_vlan_ops = {
 	.kind		=	"vlan",
 	.type		=	TCA_ACT_VLAN,
@@ -188,16 +207,39 @@ static struct tc_action_ops act_vlan_ops = {
 	.act		=	tcf_vlan,
 	.dump		=	tcf_vlan_dump,
 	.init		=	tcf_vlan_init,
+	.walk		=	tcf_vlan_walker,
+	.lookup		=	tcf_vlan_search,
+};
+
+static __net_init int vlan_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, vlan_net_id);
+
+	return tc_action_net_init(tn, &act_vlan_ops, VLAN_TAB_MASK);
+}
+
+static void __net_exit vlan_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, vlan_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations vlan_net_ops = {
+	.init = vlan_init_net,
+	.exit = vlan_exit_net,
+	.id   = &vlan_net_id,
+	.size = sizeof(struct tc_action_net),
 };
 
 static int __init vlan_init_module(void)
 {
-	return tcf_register_action(&act_vlan_ops, VLAN_TAB_MASK);
+	return tcf_register_action(&act_vlan_ops, &vlan_net_ops);
 }
 
 static void __exit vlan_cleanup_module(void)
 {
-	tcf_unregister_action(&act_vlan_ops);
+	tcf_unregister_action(&act_vlan_ops, &vlan_net_ops);
 }
 
 module_init(vlan_init_module);
-- 
cgit v0.10.2


From 64f87d3616a01c53262c6e0e948d62df15923f1c Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 23 Feb 2016 18:02:55 +0100
Subject: vxlan: consolidate GBP handling even more

Now when the packet is scrubbed early, skb->mark can be set in the GBP
handling code.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 909f793..656a303 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1164,6 +1164,7 @@ out:
 }
 
 static void vxlan_parse_gbp_hdr(struct vxlanhdr *unparsed,
+				struct sk_buff *skb, u32 vxflags,
 				struct vxlan_metadata *md,
 				struct metadata_dst *tun_dst)
 {
@@ -1183,6 +1184,9 @@ static void vxlan_parse_gbp_hdr(struct vxlanhdr *unparsed,
 	if (gbp->policy_applied)
 		md->gbp |= VXLAN_GBP_POLICY_APPLIED;
 
+	/* In flow-based mode, GBP is carried in dst_metadata */
+	if (!(vxflags & VXLAN_F_COLLECT_METADATA))
+		skb->mark = md->gbp;
 out:
 	unparsed->vx_flags &= ~VXLAN_GBP_USED_BITS;
 }
@@ -1228,9 +1232,6 @@ static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
 		goto drop;
 
 	skb_reset_network_header(skb);
-	/* In flow-based mode, GBP is carried in dst_metadata */
-	if (!(vs->flags & VXLAN_F_COLLECT_METADATA))
-		skb->mark = md->gbp;
 
 	if (oip6)
 		err = IP6_ECN_decapsulate(oip6, skb);
@@ -1329,7 +1330,7 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		if (!vxlan_remcsum(&unparsed, skb, vs->flags))
 			goto drop;
 	if (vs->flags & VXLAN_F_GBP)
-		vxlan_parse_gbp_hdr(&unparsed, md, tun_dst);
+		vxlan_parse_gbp_hdr(&unparsed, skb, vs->flags, md, tun_dst);
 
 	if (unparsed.vx_flags || unparsed.vx_vni) {
 		/* If there are any unprocessed flags remaining treat
-- 
cgit v0.10.2


From 1ab016e237e7d0734fb7adbbfe6d4cacb3520421 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 23 Feb 2016 18:02:56 +0100
Subject: vxlan: move inner L2 header processing to a separate function

This code will be different for VXLAN-GPE, so move it to a separate
function. It will also make the rx path less spaghetti-like.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 656a303..68a8f9f 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1191,15 +1191,11 @@ out:
 	unparsed->vx_flags &= ~VXLAN_GBP_USED_BITS;
 }
 
-static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
-		      struct sk_buff *skb, struct vxlan_metadata *md,
-		      struct metadata_dst *tun_dst)
+static bool vxlan_set_mac(struct vxlan_dev *vxlan,
+			  struct vxlan_sock *vs,
+			  struct sk_buff *skb)
 {
-	struct iphdr *oip = NULL;
-	struct ipv6hdr *oip6 = NULL;
-	struct pcpu_sw_netstats *stats;
 	union vxlan_addr saddr;
-	int err = 0;
 
 	skb_reset_mac_header(skb);
 	skb->protocol = eth_type_trans(skb, vxlan->dev);
@@ -1207,30 +1203,51 @@ static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
 
 	/* Ignore packet loops (and multicast echo) */
 	if (ether_addr_equal(eth_hdr(skb)->h_source, vxlan->dev->dev_addr))
-		goto drop;
+		return false;
 
 	/* Get data from the outer IP header */
 	if (vxlan_get_sk_family(vs) == AF_INET) {
-		oip = ip_hdr(skb);
-		saddr.sin.sin_addr.s_addr = oip->saddr;
+		saddr.sin.sin_addr.s_addr = ip_hdr(skb)->saddr;
 		saddr.sa.sa_family = AF_INET;
 #if IS_ENABLED(CONFIG_IPV6)
 	} else {
-		oip6 = ipv6_hdr(skb);
-		saddr.sin6.sin6_addr = oip6->saddr;
+		saddr.sin6.sin6_addr = ipv6_hdr(skb)->saddr;
 		saddr.sa.sa_family = AF_INET6;
 #endif
 	}
 
+	if ((vxlan->flags & VXLAN_F_LEARN) &&
+	    vxlan_snoop(skb->dev, &saddr, eth_hdr(skb)->h_source))
+		return false;
+
+	return true;
+}
+
+static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
+		      struct sk_buff *skb, struct vxlan_metadata *md,
+		      struct metadata_dst *tun_dst)
+{
+	struct iphdr *oip = NULL;
+	struct ipv6hdr *oip6 = NULL;
+	struct pcpu_sw_netstats *stats;
+	int err = 0;
+
+	if (!vxlan_set_mac(vxlan, vs, skb))
+		goto drop;
+
+	/* Get data from the outer IP header */
+	if (vxlan_get_sk_family(vs) == AF_INET)
+		oip = ip_hdr(skb);
+#if IS_ENABLED(CONFIG_IPV6)
+	else
+		oip6 = ipv6_hdr(skb);
+#endif
+
 	if (tun_dst) {
 		skb_dst_set(skb, (struct dst_entry *)tun_dst);
 		tun_dst = NULL;
 	}
 
-	if ((vxlan->flags & VXLAN_F_LEARN) &&
-	    vxlan_snoop(skb->dev, &saddr, eth_hdr(skb)->h_source))
-		goto drop;
-
 	skb_reset_network_header(skb);
 
 	if (oip6)
-- 
cgit v0.10.2


From 760c68054e9ed1c6e43b06f916f6efc2c8a1adcc Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 23 Feb 2016 18:02:57 +0100
Subject: vxlan: move ECN decapsulation to a separate function

It simplifies the vxlan_rcv function.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 68a8f9f..382535b 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1205,7 +1205,7 @@ static bool vxlan_set_mac(struct vxlan_dev *vxlan,
 	if (ether_addr_equal(eth_hdr(skb)->h_source, vxlan->dev->dev_addr))
 		return false;
 
-	/* Get data from the outer IP header */
+	/* Get address from the outer IP header */
 	if (vxlan_get_sk_family(vs) == AF_INET) {
 		saddr.sin.sin_addr.s_addr = ip_hdr(skb)->saddr;
 		saddr.sa.sa_family = AF_INET;
@@ -1223,52 +1223,52 @@ static bool vxlan_set_mac(struct vxlan_dev *vxlan,
 	return true;
 }
 
+static bool vxlan_ecn_decapsulate(struct vxlan_sock *vs, void *oiph,
+				  struct sk_buff *skb)
+{
+	int err = 0;
+
+	if (vxlan_get_sk_family(vs) == AF_INET)
+		err = IP_ECN_decapsulate(oiph, skb);
+#if IS_ENABLED(CONFIG_IPV6)
+	else
+		err = IP6_ECN_decapsulate(oiph, skb);
+#endif
+
+	if (unlikely(err) && log_ecn_error) {
+		if (vxlan_get_sk_family(vs) == AF_INET)
+			net_info_ratelimited("non-ECT from %pI4 with TOS=%#x\n",
+					     &((struct iphdr *)oiph)->saddr,
+					     ((struct iphdr *)oiph)->tos);
+		else
+			net_info_ratelimited("non-ECT from %pI6\n",
+					     &((struct ipv6hdr *)oiph)->saddr);
+	}
+	return err <= 1;
+}
+
 static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
 		      struct sk_buff *skb, struct vxlan_metadata *md,
 		      struct metadata_dst *tun_dst)
 {
-	struct iphdr *oip = NULL;
-	struct ipv6hdr *oip6 = NULL;
 	struct pcpu_sw_netstats *stats;
-	int err = 0;
+	void *oiph;
 
 	if (!vxlan_set_mac(vxlan, vs, skb))
 		goto drop;
 
-	/* Get data from the outer IP header */
-	if (vxlan_get_sk_family(vs) == AF_INET)
-		oip = ip_hdr(skb);
-#if IS_ENABLED(CONFIG_IPV6)
-	else
-		oip6 = ipv6_hdr(skb);
-#endif
-
 	if (tun_dst) {
 		skb_dst_set(skb, (struct dst_entry *)tun_dst);
 		tun_dst = NULL;
 	}
 
+	oiph = skb_network_header(skb);
 	skb_reset_network_header(skb);
 
-	if (oip6)
-		err = IP6_ECN_decapsulate(oip6, skb);
-	if (oip)
-		err = IP_ECN_decapsulate(oip, skb);
-
-	if (unlikely(err)) {
-		if (log_ecn_error) {
-			if (oip6)
-				net_info_ratelimited("non-ECT from %pI6\n",
-						     &oip6->saddr);
-			if (oip)
-				net_info_ratelimited("non-ECT from %pI4 with TOS=%#x\n",
-						     &oip->saddr, oip->tos);
-		}
-		if (err > 1) {
-			++vxlan->dev->stats.rx_frame_errors;
-			++vxlan->dev->stats.rx_errors;
-			goto drop;
-		}
+	if (!vxlan_ecn_decapsulate(vs, oiph, skb)) {
+		++vxlan->dev->stats.rx_frame_errors;
+		++vxlan->dev->stats.rx_errors;
+		goto drop;
 	}
 
 	stats = this_cpu_ptr(vxlan->dev->tstats);
-- 
cgit v0.10.2


From f2d1968ec85e85def98fdea0cf325851433bb60a Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 23 Feb 2016 18:02:58 +0100
Subject: vxlan: consolidate rx handling to a single function

Now when both vxlan_udp_encap_recv and vxlan_rcv are much shorter, combine
them into a single function.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 382535b..cfd6deb 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1247,56 +1247,17 @@ static bool vxlan_ecn_decapsulate(struct vxlan_sock *vs, void *oiph,
 	return err <= 1;
 }
 
-static void vxlan_rcv(struct vxlan_dev *vxlan, struct vxlan_sock *vs,
-		      struct sk_buff *skb, struct vxlan_metadata *md,
-		      struct metadata_dst *tun_dst)
-{
-	struct pcpu_sw_netstats *stats;
-	void *oiph;
-
-	if (!vxlan_set_mac(vxlan, vs, skb))
-		goto drop;
-
-	if (tun_dst) {
-		skb_dst_set(skb, (struct dst_entry *)tun_dst);
-		tun_dst = NULL;
-	}
-
-	oiph = skb_network_header(skb);
-	skb_reset_network_header(skb);
-
-	if (!vxlan_ecn_decapsulate(vs, oiph, skb)) {
-		++vxlan->dev->stats.rx_frame_errors;
-		++vxlan->dev->stats.rx_errors;
-		goto drop;
-	}
-
-	stats = this_cpu_ptr(vxlan->dev->tstats);
-	u64_stats_update_begin(&stats->syncp);
-	stats->rx_packets++;
-	stats->rx_bytes += skb->len;
-	u64_stats_update_end(&stats->syncp);
-
-	gro_cells_receive(&vxlan->gro_cells, skb);
-
-	return;
-drop:
-	if (tun_dst)
-		dst_release((struct dst_entry *)tun_dst);
-
-	/* Consume bad packet */
-	kfree_skb(skb);
-}
-
 /* Callback from net/ipv4/udp.c to receive packets */
-static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
+static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 {
 	struct metadata_dst *tun_dst = NULL;
+	struct pcpu_sw_netstats *stats;
 	struct vxlan_dev *vxlan;
 	struct vxlan_sock *vs;
 	struct vxlanhdr unparsed;
 	struct vxlan_metadata _md;
 	struct vxlan_metadata *md = &_md;
+	void *oiph;
 
 	/* Need Vxlan and inner Ethernet header to be present */
 	if (!pskb_may_pull(skb, VXLAN_HLEN))
@@ -1361,7 +1322,30 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		goto drop;
 	}
 
-	vxlan_rcv(vxlan, vs, skb, md, tun_dst);
+	if (!vxlan_set_mac(vxlan, vs, skb))
+		goto drop;
+
+	if (tun_dst) {
+		skb_dst_set(skb, (struct dst_entry *)tun_dst);
+		tun_dst = NULL;
+	}
+
+	oiph = skb_network_header(skb);
+	skb_reset_network_header(skb);
+
+	if (!vxlan_ecn_decapsulate(vs, oiph, skb)) {
+		++vxlan->dev->stats.rx_frame_errors;
+		++vxlan->dev->stats.rx_errors;
+		goto drop;
+	}
+
+	stats = this_cpu_ptr(vxlan->dev->tstats);
+	u64_stats_update_begin(&stats->syncp);
+	stats->rx_packets++;
+	stats->rx_bytes += skb->len;
+	u64_stats_update_end(&stats->syncp);
+
+	gro_cells_receive(&vxlan->gro_cells, skb);
 	return 0;
 
 drop:
@@ -2666,7 +2650,7 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, bool ipv6,
 	/* Mark socket as an encapsulation socket. */
 	tunnel_cfg.sk_user_data = vs;
 	tunnel_cfg.encap_type = 1;
-	tunnel_cfg.encap_rcv = vxlan_udp_encap_recv;
+	tunnel_cfg.encap_rcv = vxlan_rcv;
 	tunnel_cfg.encap_destroy = NULL;
 
 	setup_udp_tunnel_sock(net, sock, &tunnel_cfg);
-- 
cgit v0.10.2


From 10a5af238cd29b7e43af0dc0690ae9baa0650c36 Mon Sep 17 00:00:00 2001
From: Jiri Benc <jbenc@redhat.com>
Date: Tue, 23 Feb 2016 18:02:59 +0100
Subject: vxlan: simplify metadata_dst usage in vxlan_rcv

Now when the packet is scrubbed early, the metadata_dst can be assigned to
the skb as soon as it is allocated. This simplifies the error cleanup path,
as the dst will be freed by kfree_skb. It is also not necessary to pass it
as a parameter to functions anymore.

Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index cfd6deb..775ddb4 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1165,16 +1165,17 @@ out:
 
 static void vxlan_parse_gbp_hdr(struct vxlanhdr *unparsed,
 				struct sk_buff *skb, u32 vxflags,
-				struct vxlan_metadata *md,
-				struct metadata_dst *tun_dst)
+				struct vxlan_metadata *md)
 {
 	struct vxlanhdr_gbp *gbp = (struct vxlanhdr_gbp *)unparsed;
+	struct metadata_dst *tun_dst;
 
 	if (!(unparsed->vx_flags & VXLAN_HF_GBP))
 		goto out;
 
 	md->gbp = ntohs(gbp->policy_id);
 
+	tun_dst = (struct metadata_dst *)skb_dst(skb);
 	if (tun_dst)
 		tun_dst->u.tun_info.key.tun_flags |= TUNNEL_VXLAN_OPT;
 
@@ -1250,7 +1251,6 @@ static bool vxlan_ecn_decapsulate(struct vxlan_sock *vs, void *oiph,
 /* Callback from net/ipv4/udp.c to receive packets */
 static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 {
-	struct metadata_dst *tun_dst = NULL;
 	struct pcpu_sw_netstats *stats;
 	struct vxlan_dev *vxlan;
 	struct vxlan_sock *vs;
@@ -1289,6 +1289,7 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 
 	if (vxlan_collect_metadata(vs)) {
 		__be32 vni = vxlan_vni(vxlan_hdr(skb)->vx_vni);
+		struct metadata_dst *tun_dst;
 
 		tun_dst = udp_tun_rx_dst(skb, vxlan_get_sk_family(vs), TUNNEL_KEY,
 					 vxlan_vni_to_tun_id(vni), sizeof(*md));
@@ -1297,6 +1298,8 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 			goto drop;
 
 		md = ip_tunnel_info_opts(&tun_dst->u.tun_info);
+
+		skb_dst_set(skb, (struct dst_entry *)tun_dst);
 	} else {
 		memset(md, 0, sizeof(*md));
 	}
@@ -1308,7 +1311,7 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 		if (!vxlan_remcsum(&unparsed, skb, vs->flags))
 			goto drop;
 	if (vs->flags & VXLAN_F_GBP)
-		vxlan_parse_gbp_hdr(&unparsed, skb, vs->flags, md, tun_dst);
+		vxlan_parse_gbp_hdr(&unparsed, skb, vs->flags, md);
 
 	if (unparsed.vx_flags || unparsed.vx_vni) {
 		/* If there are any unprocessed flags remaining treat
@@ -1325,11 +1328,6 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 	if (!vxlan_set_mac(vxlan, vs, skb))
 		goto drop;
 
-	if (tun_dst) {
-		skb_dst_set(skb, (struct dst_entry *)tun_dst);
-		tun_dst = NULL;
-	}
-
 	oiph = skb_network_header(skb);
 	skb_reset_network_header(skb);
 
@@ -1349,9 +1347,6 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb)
 	return 0;
 
 drop:
-	if (tun_dst)
-		dst_release((struct dst_entry *)tun_dst);
-
 	/* Consume bad packet */
 	kfree_skb(skb);
 	return 0;
-- 
cgit v0.10.2


From 65aebfc002abc1827ac7c8644a2bba0459ce3ce2 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Tue, 23 Feb 2016 12:13:54 -0500
Subject: net: dsa: add port_vlan_dump routine

Similar to port_fdb_dump, add a port_vlan_dump function to DSA drivers
which gets passed the switchdev VLAN object and callback.

This function, if implemented, takes precedence over the soon legacy
vlan_getnext/port_pvid_get approach.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/dsa/dsa.txt b/Documentation/networking/dsa/dsa.txt
index ebf2153..350a502 100644
--- a/Documentation/networking/dsa/dsa.txt
+++ b/Documentation/networking/dsa/dsa.txt
@@ -554,6 +554,10 @@ Bridge VLAN filtering
 - port_vlan_del: bridge layer function invoked when a VLAN is removed from the
   given switch port
 
+- port_vlan_dump: bridge layer function invoked with a switchdev callback
+  function that the driver has to call for each VLAN the given port is a member
+  of. A switchdev object is used to carry the VID and bridge flags.
+
 - vlan_getnext: bridge layer function invoked to query the next configured VLAN
   in the switch, i.e. returns the bitmaps of members and untagged ports
 
diff --git a/include/net/dsa.h b/include/net/dsa.h
index 1c845d7..ebc0d9e 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -313,6 +313,9 @@ struct dsa_switch_driver {
 				 struct switchdev_trans *trans);
 	int	(*port_vlan_del)(struct dsa_switch *ds, int port,
 				 const struct switchdev_obj_port_vlan *vlan);
+	int	(*port_vlan_dump)(struct dsa_switch *ds, int port,
+				  struct switchdev_obj_port_vlan *vlan,
+				  int (*cb)(struct switchdev_obj *obj));
 	int	(*port_pvid_get)(struct dsa_switch *ds, int port, u16 *pvid);
 	int	(*vlan_getnext)(struct dsa_switch *ds, u16 *vid,
 				unsigned long *ports, unsigned long *untagged);
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 14ca978..a9cbb72 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -248,6 +248,9 @@ static int dsa_slave_port_vlan_dump(struct net_device *dev,
 	u16 pvid, vid = 0;
 	int err;
 
+	if (ds->drv->port_vlan_dump)
+		return ds->drv->port_vlan_dump(ds, p->port, vlan, cb);
+
 	if (!ds->drv->vlan_getnext || !ds->drv->port_pvid_get)
 		return -EOPNOTSUPP;
 
-- 
cgit v0.10.2


From ceff5eff3a4c690764d11f2d759c27a0d2dea732 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Tue, 23 Feb 2016 12:13:55 -0500
Subject: net: dsa: mv88e6xxx: implement port_vlan_dump

Remove the port_pvid_get and vlan_getnext functions in favor of a
simpler mv88e6xxx_port_vlan_dump function.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6171.c b/drivers/net/dsa/mv88e6171.c
index 6e18213..dd1ebaf 100644
--- a/drivers/net/dsa/mv88e6171.c
+++ b/drivers/net/dsa/mv88e6171.c
@@ -106,11 +106,10 @@ struct dsa_switch_driver mv88e6171_switch_driver = {
 	.port_join_bridge	= mv88e6xxx_port_bridge_join,
 	.port_leave_bridge	= mv88e6xxx_port_bridge_leave,
 	.port_stp_update        = mv88e6xxx_port_stp_update,
-	.port_pvid_get		= mv88e6xxx_port_pvid_get,
 	.port_vlan_prepare	= mv88e6xxx_port_vlan_prepare,
 	.port_vlan_add		= mv88e6xxx_port_vlan_add,
 	.port_vlan_del		= mv88e6xxx_port_vlan_del,
-	.vlan_getnext		= mv88e6xxx_vlan_getnext,
+	.port_vlan_dump		= mv88e6xxx_port_vlan_dump,
 	.port_fdb_prepare	= mv88e6xxx_port_fdb_prepare,
 	.port_fdb_add		= mv88e6xxx_port_fdb_add,
 	.port_fdb_del		= mv88e6xxx_port_fdb_del,
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
index a47f52f..bbca36a 100644
--- a/drivers/net/dsa/mv88e6352.c
+++ b/drivers/net/dsa/mv88e6352.c
@@ -327,11 +327,10 @@ struct dsa_switch_driver mv88e6352_switch_driver = {
 	.port_join_bridge	= mv88e6xxx_port_bridge_join,
 	.port_leave_bridge	= mv88e6xxx_port_bridge_leave,
 	.port_stp_update	= mv88e6xxx_port_stp_update,
-	.port_pvid_get		= mv88e6xxx_port_pvid_get,
 	.port_vlan_prepare	= mv88e6xxx_port_vlan_prepare,
 	.port_vlan_add		= mv88e6xxx_port_vlan_add,
 	.port_vlan_del		= mv88e6xxx_port_vlan_del,
-	.vlan_getnext		= mv88e6xxx_vlan_getnext,
+	.port_vlan_dump		= mv88e6xxx_port_vlan_dump,
 	.port_fdb_prepare	= mv88e6xxx_port_fdb_prepare,
 	.port_fdb_add		= mv88e6xxx_port_fdb_add,
 	.port_fdb_del		= mv88e6xxx_port_fdb_del,
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 685dcb0..d98dc63 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1151,19 +1151,6 @@ static int _mv88e6xxx_port_pvid_get(struct dsa_switch *ds, int port, u16 *pvid)
 	return 0;
 }
 
-int mv88e6xxx_port_pvid_get(struct dsa_switch *ds, int port, u16 *pvid)
-{
-	int ret;
-
-	ret = mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_DEFAULT_VLAN);
-	if (ret < 0)
-		return ret;
-
-	*pvid = ret & PORT_DEFAULT_VLAN_MASK;
-
-	return 0;
-}
-
 static int _mv88e6xxx_port_pvid_set(struct dsa_switch *ds, int port, u16 pvid)
 {
 	return _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_DEFAULT_VLAN,
@@ -1306,6 +1293,57 @@ static int _mv88e6xxx_vtu_getnext(struct dsa_switch *ds,
 	return 0;
 }
 
+int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port,
+			     struct switchdev_obj_port_vlan *vlan,
+			     int (*cb)(struct switchdev_obj *obj))
+{
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	struct mv88e6xxx_vtu_stu_entry next;
+	u16 pvid;
+	int err;
+
+	mutex_lock(&ps->smi_mutex);
+
+	err = _mv88e6xxx_port_pvid_get(ds, port, &pvid);
+	if (err)
+		goto unlock;
+
+	err = _mv88e6xxx_vtu_vid_write(ds, GLOBAL_VTU_VID_MASK);
+	if (err)
+		goto unlock;
+
+	do {
+		err = _mv88e6xxx_vtu_getnext(ds, &next);
+		if (err)
+			break;
+
+		if (!next.valid)
+			break;
+
+		if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER)
+			continue;
+
+		/* reinit and dump this VLAN obj */
+		vlan->vid_begin = vlan->vid_end = next.vid;
+		vlan->flags = 0;
+
+		if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED)
+			vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED;
+
+		if (next.vid == pvid)
+			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
+
+		err = cb(&vlan->obj);
+		if (err)
+			break;
+	} while (next.vid < GLOBAL_VTU_VID_MASK);
+
+unlock:
+	mutex_unlock(&ps->smi_mutex);
+
+	return err;
+}
+
 static int _mv88e6xxx_vtu_loadpurge(struct dsa_switch *ds,
 				    struct mv88e6xxx_vtu_stu_entry *entry)
 {
@@ -1675,52 +1713,6 @@ unlock:
 	return err;
 }
 
-int mv88e6xxx_vlan_getnext(struct dsa_switch *ds, u16 *vid,
-			   unsigned long *ports, unsigned long *untagged)
-{
-	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
-	struct mv88e6xxx_vtu_stu_entry next;
-	int port;
-	int err;
-
-	if (*vid == 4095)
-		return -ENOENT;
-
-	mutex_lock(&ps->smi_mutex);
-	err = _mv88e6xxx_vtu_vid_write(ds, *vid);
-	if (err)
-		goto unlock;
-
-	err = _mv88e6xxx_vtu_getnext(ds, &next);
-unlock:
-	mutex_unlock(&ps->smi_mutex);
-
-	if (err)
-		return err;
-
-	if (!next.valid)
-		return -ENOENT;
-
-	*vid = next.vid;
-
-	for (port = 0; port < ps->num_ports; ++port) {
-		clear_bit(port, ports);
-		clear_bit(port, untagged);
-
-		if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port))
-			continue;
-
-		if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED ||
-		    next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED)
-			set_bit(port, ports);
-
-		if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED)
-			set_bit(port, untagged);
-	}
-
-	return 0;
-}
-
 static int _mv88e6xxx_atu_mac_write(struct dsa_switch *ds,
 				    const unsigned char *addr)
 {
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 260b491..6a30bda 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -494,9 +494,9 @@ int mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port,
 			    struct switchdev_trans *trans);
 int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port,
 			    const struct switchdev_obj_port_vlan *vlan);
-int mv88e6xxx_port_pvid_get(struct dsa_switch *ds, int port, u16 *vid);
-int mv88e6xxx_vlan_getnext(struct dsa_switch *ds, u16 *vid,
-			   unsigned long *ports, unsigned long *untagged);
+int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port,
+			     struct switchdev_obj_port_vlan *vlan,
+			     int (*cb)(struct switchdev_obj *obj));
 int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port,
 			       const struct switchdev_obj_port_fdb *fdb,
 			       struct switchdev_trans *trans);
-- 
cgit v0.10.2


From 477b184526a7f44164029eea720da0e0c888cac6 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Tue, 23 Feb 2016 12:13:56 -0500
Subject: net: dsa: drop vlan_getnext

The VLAN GetNext operation is specific to some switches, and thus can be
complicated to implement for some drivers.

Remove the support for the vlan_getnext/port_pvid_get approach in favor
of the generic and simpler port_vlan_dump function.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/dsa/dsa.txt b/Documentation/networking/dsa/dsa.txt
index 350a502..974e9c3 100644
--- a/Documentation/networking/dsa/dsa.txt
+++ b/Documentation/networking/dsa/dsa.txt
@@ -542,12 +542,6 @@ Bridge layer
 Bridge VLAN filtering
 ---------------------
 
-- port_pvid_get: bridge layer function invoked when a Port-based VLAN ID is
-  queried for the given switch port
-
-- port_pvid_set: bridge layer function invoked when a Port-based VLAN ID needs
-  to be configured on the given switch port
-
 - port_vlan_add: bridge layer function invoked when a VLAN is configured
   (tagged or untagged) for the given switch port
 
@@ -558,9 +552,6 @@ Bridge VLAN filtering
   function that the driver has to call for each VLAN the given port is a member
   of. A switchdev object is used to carry the VID and bridge flags.
 
-- vlan_getnext: bridge layer function invoked to query the next configured VLAN
-  in the switch, i.e. returns the bitmaps of members and untagged ports
-
 - port_fdb_add: bridge layer function invoked when the bridge wants to install a
   Forwarding Database entry, the switch hardware should be programmed with the
   specified address in the specified VLAN Id in the forwarding database
diff --git a/include/net/dsa.h b/include/net/dsa.h
index ebc0d9e..3dd5486 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -316,9 +316,6 @@ struct dsa_switch_driver {
 	int	(*port_vlan_dump)(struct dsa_switch *ds, int port,
 				  struct switchdev_obj_port_vlan *vlan,
 				  int (*cb)(struct switchdev_obj *obj));
-	int	(*port_pvid_get)(struct dsa_switch *ds, int port, u16 *pvid);
-	int	(*vlan_getnext)(struct dsa_switch *ds, u16 *vid,
-				unsigned long *ports, unsigned long *untagged);
 
 	/*
 	 * Forwarding database
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index a9cbb72..cde2923 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -243,44 +243,11 @@ static int dsa_slave_port_vlan_dump(struct net_device *dev,
 {
 	struct dsa_slave_priv *p = netdev_priv(dev);
 	struct dsa_switch *ds = p->parent;
-	DECLARE_BITMAP(members, DSA_MAX_PORTS);
-	DECLARE_BITMAP(untagged, DSA_MAX_PORTS);
-	u16 pvid, vid = 0;
-	int err;
 
 	if (ds->drv->port_vlan_dump)
 		return ds->drv->port_vlan_dump(ds, p->port, vlan, cb);
 
-	if (!ds->drv->vlan_getnext || !ds->drv->port_pvid_get)
-		return -EOPNOTSUPP;
-
-	err = ds->drv->port_pvid_get(ds, p->port, &pvid);
-	if (err)
-		return err;
-
-	for (;;) {
-		err = ds->drv->vlan_getnext(ds, &vid, members, untagged);
-		if (err)
-			break;
-
-		if (!test_bit(p->port, members))
-			continue;
-
-		memset(vlan, 0, sizeof(*vlan));
-		vlan->vid_begin = vlan->vid_end = vid;
-
-		if (vid == pvid)
-			vlan->flags |= BRIDGE_VLAN_INFO_PVID;
-
-		if (test_bit(p->port, untagged))
-			vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED;
-
-		err = cb(&vlan->obj);
-		if (err)
-			break;
-	}
-
-	return err == -ENOENT ? 0 : err;
+	return -EOPNOTSUPP;
 }
 
 static int dsa_slave_port_fdb_add(struct net_device *dev,
-- 
cgit v0.10.2


From 407353ec85ccc775ce342a6c5a1202f34ed827cc Mon Sep 17 00:00:00 2001
From: Clemens Gruber <clemens.gruber@pqgruber.com>
Date: Tue, 23 Feb 2016 20:16:58 +0100
Subject: phy: marvell: Fix 88E1510 initialization

A bug was introduced in the merge commit b633353115e3 ("Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net")
The generic marvell_config_init (and therefore marvell_of_reg_init) is
not called anymore for the Marvell 88E1510 (in net-next).

This patch calls marvell_config_init and moves the specific init
function for the 88E1510 below the marvell_config_init function to avoid
adding a function predeclaration.

Signed-off-by: Clemens Gruber <clemens.gruber@pqgruber.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 9fb9d80..280e879 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -443,41 +443,6 @@ static int m88e1318_config_aneg(struct phy_device *phydev)
 	return m88e1121_config_aneg(phydev);
 }
 
-static int m88e1510_config_init(struct phy_device *phydev)
-{
-	int err;
-	int temp;
-
-	/* SGMII-to-Copper mode initialization */
-	if (phydev->interface == PHY_INTERFACE_MODE_SGMII) {
-		/* Select page 18 */
-		err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 18);
-		if (err < 0)
-			return err;
-
-		/* In reg 20, write MODE[2:0] = 0x1 (SGMII to Copper) */
-		temp = phy_read(phydev, MII_88E1510_GEN_CTRL_REG_1);
-		temp &= ~MII_88E1510_GEN_CTRL_REG_1_MODE_MASK;
-		temp |= MII_88E1510_GEN_CTRL_REG_1_MODE_SGMII;
-		err = phy_write(phydev, MII_88E1510_GEN_CTRL_REG_1, temp);
-		if (err < 0)
-			return err;
-
-		/* PHY reset is necessary after changing MODE[2:0] */
-		temp |= MII_88E1510_GEN_CTRL_REG_1_RESET;
-		err = phy_write(phydev, MII_88E1510_GEN_CTRL_REG_1, temp);
-		if (err < 0)
-			return err;
-
-		/* Reset page selection */
-		err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0);
-		if (err < 0)
-			return err;
-	}
-
-	return 0;
-}
-
 static int m88e1510_config_aneg(struct phy_device *phydev)
 {
 	int err;
@@ -671,6 +636,41 @@ static int m88e1111_config_init(struct phy_device *phydev)
 	return phy_write(phydev, MII_BMCR, BMCR_RESET);
 }
 
+static int m88e1510_config_init(struct phy_device *phydev)
+{
+	int err;
+	int temp;
+
+	/* SGMII-to-Copper mode initialization */
+	if (phydev->interface == PHY_INTERFACE_MODE_SGMII) {
+		/* Select page 18 */
+		err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 18);
+		if (err < 0)
+			return err;
+
+		/* In reg 20, write MODE[2:0] = 0x1 (SGMII to Copper) */
+		temp = phy_read(phydev, MII_88E1510_GEN_CTRL_REG_1);
+		temp &= ~MII_88E1510_GEN_CTRL_REG_1_MODE_MASK;
+		temp |= MII_88E1510_GEN_CTRL_REG_1_MODE_SGMII;
+		err = phy_write(phydev, MII_88E1510_GEN_CTRL_REG_1, temp);
+		if (err < 0)
+			return err;
+
+		/* PHY reset is necessary after changing MODE[2:0] */
+		temp |= MII_88E1510_GEN_CTRL_REG_1_RESET;
+		err = phy_write(phydev, MII_88E1510_GEN_CTRL_REG_1, temp);
+		if (err < 0)
+			return err;
+
+		/* Reset page selection */
+		err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0);
+		if (err < 0)
+			return err;
+	}
+
+	return marvell_config_init(phydev);
+}
+
 static int m88e1118_config_aneg(struct phy_device *phydev)
 {
 	int err;
-- 
cgit v0.10.2


From 3db66fdc5fb2a4e11df63286ee7ea3a828725529 Mon Sep 17 00:00:00 2001
From: Yangbo Lu <yangbo.lu@nxp.com>
Date: Wed, 24 Feb 2016 17:26:54 +0800
Subject: ARM: dts: ls1021a: add 1588 timer node

Add the 1588 timer node for ls1021a platform to
support gianfar ptp driver.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/ls1021a.dtsi b/arch/arm/boot/dts/ls1021a.dtsi
index 2c84ca2..ecf12dc 100644
--- a/arch/arm/boot/dts/ls1021a.dtsi
+++ b/arch/arm/boot/dts/ls1021a.dtsi
@@ -457,6 +457,18 @@
 			reg = <0x0 0x2d24000 0x0 0x4000>;
 		};
 
+		ptp_clock@2d10e00 {
+			compatible = "fsl,etsec-ptp";
+			reg = <0x0 0x2d10e00 0x0 0xb0>;
+			interrupts = <GIC_SPI 173 IRQ_TYPE_LEVEL_HIGH>;
+			fsl,tclk-period = <5>;
+			fsl,tmr-prsc    = <2>;
+			fsl,tmr-add     = <0xaaaaaaab>;
+			fsl,tmr-fiper1  = <999999990>;
+			fsl,tmr-fiper2  = <99990>;
+			fsl,max-adj     = <499999999>;
+		};
+
 		enet0: ethernet@2d10000 {
 			compatible = "fsl,etsec2";
 			device_type = "network";
-- 
cgit v0.10.2


From c35ec779663b030ad3a03147be23146274026ae3 Mon Sep 17 00:00:00 2001
From: Yangbo Lu <yangbo.lu@nxp.com>
Date: Wed, 24 Feb 2016 17:26:55 +0800
Subject: gianfar_ptp: replace get_of_u32 with of_property_read_u32

Replace get_of_u32 with standard helper function of_property_read_u32
since the latter can process cpu endianness.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/gianfar_ptp.c b/drivers/net/ethernet/freescale/gianfar_ptp.c
index b40fba9..5779881 100644
--- a/drivers/net/ethernet/freescale/gianfar_ptp.c
+++ b/drivers/net/ethernet/freescale/gianfar_ptp.c
@@ -422,19 +422,6 @@ static struct ptp_clock_info ptp_gianfar_caps = {
 	.enable		= ptp_gianfar_enable,
 };
 
-/* OF device tree */
-
-static int get_of_u32(struct device_node *node, char *str, u32 *val)
-{
-	int plen;
-	const u32 *prop = of_get_property(node, str, &plen);
-
-	if (!prop || plen != sizeof(*prop))
-		return -1;
-	*val = *prop;
-	return 0;
-}
-
 static int gianfar_ptp_probe(struct platform_device *dev)
 {
 	struct device_node *node = dev->dev.of_node;
@@ -452,15 +439,21 @@ static int gianfar_ptp_probe(struct platform_device *dev)
 
 	etsects->caps = ptp_gianfar_caps;
 
-	if (get_of_u32(node, "fsl,cksel", &etsects->cksel))
+	if (of_property_read_u32(node, "fsl,cksel", &etsects->cksel))
 		etsects->cksel = DEFAULT_CKSEL;
 
-	if (get_of_u32(node, "fsl,tclk-period", &etsects->tclk_period) ||
-	    get_of_u32(node, "fsl,tmr-prsc", &etsects->tmr_prsc) ||
-	    get_of_u32(node, "fsl,tmr-add", &etsects->tmr_add) ||
-	    get_of_u32(node, "fsl,tmr-fiper1", &etsects->tmr_fiper1) ||
-	    get_of_u32(node, "fsl,tmr-fiper2", &etsects->tmr_fiper2) ||
-	    get_of_u32(node, "fsl,max-adj", &etsects->caps.max_adj)) {
+	if (of_property_read_u32(node,
+				 "fsl,tclk-period", &etsects->tclk_period) ||
+	    of_property_read_u32(node,
+				 "fsl,tmr-prsc", &etsects->tmr_prsc) ||
+	    of_property_read_u32(node,
+				 "fsl,tmr-add", &etsects->tmr_add) ||
+	    of_property_read_u32(node,
+				 "fsl,tmr-fiper1", &etsects->tmr_fiper1) ||
+	    of_property_read_u32(node,
+				 "fsl,tmr-fiper2", &etsects->tmr_fiper2) ||
+	    of_property_read_u32(node,
+				 "fsl,max-adj", &etsects->caps.max_adj)) {
 		pr_err("device tree node missing required elements\n");
 		goto no_node;
 	}
-- 
cgit v0.10.2


From f54af12f4348b23d78f5186854d23e47795d21cb Mon Sep 17 00:00:00 2001
From: Yangbo Lu <yangbo.lu@nxp.com>
Date: Wed, 24 Feb 2016 17:26:56 +0800
Subject: gianfar: fix endianness for hardware timestamp

Fix endianness for the 64-bit hardware timestamp value with
be64_to_cpu to support both PowerPC platforms and ARM platforms.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 1e1157f..a01e5a3 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -2706,7 +2706,7 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 					  ~0x7UL);
 
 			memset(&shhwtstamps, 0, sizeof(shhwtstamps));
-			shhwtstamps.hwtstamp = ns_to_ktime(*ns);
+			shhwtstamps.hwtstamp = ns_to_ktime(be64_to_cpu(*ns));
 			skb_pull(skb, GMAC_FCB_LEN + GMAC_TXPAL_LEN);
 			skb_tstamp_tx(skb, &shhwtstamps);
 			gfar_clear_txbd_status(bdp);
@@ -3035,7 +3035,7 @@ static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb)
 		u64 *ns = (u64 *) skb->data;
 
 		memset(shhwtstamps, 0, sizeof(*shhwtstamps));
-		shhwtstamps->hwtstamp = ns_to_ktime(*ns);
+		shhwtstamps->hwtstamp = ns_to_ktime(be64_to_cpu(*ns));
 	}
 
 	if (priv->padding)
-- 
cgit v0.10.2


From 5327ef9c95b09f536e886c61d1d768b3236748d5 Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 24 Feb 2016 19:27:49 +0530
Subject: net: tulip: Use setup_timer()

Convert a call to init_timer and accompanying intializations of
the timer's data and function fields to a call to setup_timer.

The Coccinelle semantic patch that fixes this problem is
as follows:

// <smpl>
@@
expression t,f,d;
@@

-init_timer(&t);
+setup_timer(&t,f,d);
-t.data = d;
-t.function = f;
// </smpl>

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c
index b553409..94d0eeb 100644
--- a/drivers/net/ethernet/dec/tulip/tulip_core.c
+++ b/drivers/net/ethernet/dec/tulip/tulip_core.c
@@ -505,9 +505,7 @@ media_picked:
 	tp->timer.expires = RUN_AT(next_tick);
 	add_timer(&tp->timer);
 #ifdef CONFIG_TULIP_NAPI
-	init_timer(&tp->oom_timer);
-        tp->oom_timer.data = (unsigned long)dev;
-        tp->oom_timer.function = oom_timer;
+	setup_timer(&tp->oom_timer, oom_timer, (unsigned long)dev);
 #endif
 }
 
@@ -782,9 +780,8 @@ static void tulip_down (struct net_device *dev)
 
 	spin_unlock_irqrestore (&tp->lock, flags);
 
-	init_timer(&tp->timer);
-	tp->timer.data = (unsigned long)dev;
-	tp->timer.function = tulip_tbl[tp->chip_id].media_timer;
+	setup_timer(&tp->timer, tulip_tbl[tp->chip_id].media_timer,
+		    (unsigned long)dev);
 
 	dev->if_port = tp->saved_if_port;
 
@@ -1475,9 +1472,8 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	tp->csr0 = csr0;
 	spin_lock_init(&tp->lock);
 	spin_lock_init(&tp->mii_lock);
-	init_timer(&tp->timer);
-	tp->timer.data = (unsigned long)dev;
-	tp->timer.function = tulip_tbl[tp->chip_id].media_timer;
+	setup_timer(&tp->timer, tulip_tbl[tp->chip_id].media_timer,
+		    (unsigned long)dev);
 
 	INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task);
 
-- 
cgit v0.10.2


From de55558dc4e6562197bf0ea0fe249cbd7ccebae5 Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 24 Feb 2016 19:28:01 +0530
Subject: forcedeth: Use setup_timer()

Convert a call to init_timer and accompanying intializations of
the timer's data and function fields to a call to setup_timer.

The Coccinelle semantic patch that fixes this problem is
as follows:

// <smpl>
@@
expression t,f,d;
@@

-init_timer(&t);
+setup_timer(&t,f,d);
-t.data = d;
-t.function = f;
// </smpl>

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index 75e88f4..9b0d7f4 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -5629,12 +5629,8 @@ static int nv_probe(struct pci_dev *pci_dev, const struct pci_device_id *id)
 	u64_stats_init(&np->swstats_rx_syncp);
 	u64_stats_init(&np->swstats_tx_syncp);
 
-	init_timer(&np->oom_kick);
-	np->oom_kick.data = (unsigned long) dev;
-	np->oom_kick.function = nv_do_rx_refill;	/* timer handler */
-	init_timer(&np->nic_poll);
-	np->nic_poll.data = (unsigned long) dev;
-	np->nic_poll.function = nv_do_nic_poll;	/* timer handler */
+	setup_timer(&np->oom_kick, nv_do_rx_refill, (unsigned long)dev);
+	setup_timer(&np->nic_poll, nv_do_nic_poll, (unsigned long)dev);
 	init_timer_deferrable(&np->stats_poll);
 	np->stats_poll.data = (unsigned long) dev;
 	np->stats_poll.function = nv_do_stats_poll;	/* timer handler */
-- 
cgit v0.10.2


From 5b6490def9168af6a1609df910a7800bcbc97977 Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 24 Feb 2016 19:28:19 +0530
Subject: 3c59x: Use setup_timer()

Convert a call to init_timer and accompanying intializations of
the timer's data and function fields to a call to setup_timer.

The Coccinelle semantic patch that fixes this problem is
as follows:

// <smpl>
@@
expression t,f,d;
@@

-init_timer(&t);
+setup_timer(&t,f,d);
 ...
-t.data = d;
-t.function = f;
// </smpl>

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index 79e1a02..c377607 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -1601,15 +1601,9 @@ vortex_up(struct net_device *dev)
 				dev->name, media_tbl[dev->if_port].name);
 	}
 
-	init_timer(&vp->timer);
+	setup_timer(&vp->timer, vortex_timer, (unsigned long)dev);
 	vp->timer.expires = RUN_AT(media_tbl[dev->if_port].wait);
-	vp->timer.data = (unsigned long)dev;
-	vp->timer.function = vortex_timer;		/* timer handler */
-	add_timer(&vp->timer);
-
-	init_timer(&vp->rx_oom_timer);
-	vp->rx_oom_timer.data = (unsigned long)dev;
-	vp->rx_oom_timer.function = rx_oom_timer;
+	setup_timer(&vp->rx_oom_timer, rx_oom_timer, (unsigned long)dev);
 
 	if (vortex_debug > 1)
 		pr_debug("%s: Initial media type %s.\n",
-- 
cgit v0.10.2


From 016c0bbae1d17d4509b74227dca618e01263e61a Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 24 Feb 2016 20:09:38 +0530
Subject: netxen: Use kobj_to_dev()

Introduce the use of kobj_to_dev() helper function instead of open
coding it with container_of()

The Coccinelle semantic patch used to make this change is as follows:

//<smpl>
@@
expression a;
symbol kobj;
@@
- container_of(a, struct device, kobj)
+ kobj_to_dev(a)
//</smpl>

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index 6409a06..fd362b6 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -2891,7 +2891,7 @@ netxen_sysfs_read_crb(struct file *filp, struct kobject *kobj,
 		struct bin_attribute *attr,
 		char *buf, loff_t offset, size_t size)
 {
-	struct device *dev = container_of(kobj, struct device, kobj);
+	struct device *dev = kobj_to_dev(kobj);
 	struct netxen_adapter *adapter = dev_get_drvdata(dev);
 	u32 data;
 	u64 qmdata;
@@ -2919,7 +2919,7 @@ netxen_sysfs_write_crb(struct file *filp, struct kobject *kobj,
 		struct bin_attribute *attr,
 		char *buf, loff_t offset, size_t size)
 {
-	struct device *dev = container_of(kobj, struct device, kobj);
+	struct device *dev = kobj_to_dev(kobj);
 	struct netxen_adapter *adapter = dev_get_drvdata(dev);
 	u32 data;
 	u64 qmdata;
@@ -2960,7 +2960,7 @@ netxen_sysfs_read_mem(struct file *filp, struct kobject *kobj,
 		struct bin_attribute *attr,
 		char *buf, loff_t offset, size_t size)
 {
-	struct device *dev = container_of(kobj, struct device, kobj);
+	struct device *dev = kobj_to_dev(kobj);
 	struct netxen_adapter *adapter = dev_get_drvdata(dev);
 	u64 data;
 	int ret;
@@ -2981,7 +2981,7 @@ static ssize_t netxen_sysfs_write_mem(struct file *filp, struct kobject *kobj,
 		struct bin_attribute *attr, char *buf,
 		loff_t offset, size_t size)
 {
-	struct device *dev = container_of(kobj, struct device, kobj);
+	struct device *dev = kobj_to_dev(kobj);
 	struct netxen_adapter *adapter = dev_get_drvdata(dev);
 	u64 data;
 	int ret;
@@ -3018,7 +3018,7 @@ netxen_sysfs_read_dimm(struct file *filp, struct kobject *kobj,
 		struct bin_attribute *attr,
 		char *buf, loff_t offset, size_t size)
 {
-	struct device *dev = container_of(kobj, struct device, kobj);
+	struct device *dev = kobj_to_dev(kobj);
 	struct netxen_adapter *adapter = dev_get_drvdata(dev);
 	struct net_device *netdev = adapter->netdev;
 	struct netxen_dimm_cfg dimm;
-- 
cgit v0.10.2


From 0e7441d73dafe5481f62029bad8ef046e22313b1 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 24 Feb 2016 16:52:45 +0200
Subject: qede: Change pci DID for 10g device

The device ID for the 10g module has changed. Populate the pci_ids table
accordingly.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 5f15e23..76fa5d7 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -53,7 +53,7 @@ MODULE_PARM_DESC(debug, " Default debug msglevel");
 static const struct qed_eth_ops *qed_ops;
 
 #define CHIP_NUM_57980S_40		0x1634
-#define CHIP_NUM_57980S_10		0x1635
+#define CHIP_NUM_57980S_10		0x1666
 #define CHIP_NUM_57980S_MF		0x1636
 #define CHIP_NUM_57980S_100		0x1644
 #define CHIP_NUM_57980S_50		0x1654
-- 
cgit v0.10.2


From b1199b100e67f6d296e915b501f3c84d892a99c2 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 24 Feb 2016 16:52:46 +0200
Subject: qede: Linearize SKBs when needed

There's a corner-case in HW where an SKB queued for transmission that
contains too many frags will cause FW to assert.
This patch solves this by linearizing the SKB if necessary.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 76fa5d7..f07b9a9 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -380,6 +380,28 @@ static int map_frag_to_bd(struct qede_dev *edev,
 	return 0;
 }
 
+/* +2 for 1st BD for headers and 2nd BD for headlen (if required) */
+#if ((MAX_SKB_FRAGS + 2) > ETH_TX_MAX_BDS_PER_NON_LSO_PACKET)
+static bool qede_pkt_req_lin(struct qede_dev *edev, struct sk_buff *skb,
+			     u8 xmit_type)
+{
+	int allowed_frags = ETH_TX_MAX_BDS_PER_NON_LSO_PACKET - 1;
+
+	if (xmit_type & XMIT_LSO) {
+		int hlen;
+
+		hlen = skb_transport_header(skb) +
+		       tcp_hdrlen(skb) - skb->data;
+
+		/* linear payload would require its own BD */
+		if (skb_headlen(skb) > hlen)
+			allowed_frags--;
+	}
+
+	return (skb_shinfo(skb)->nr_frags > allowed_frags);
+}
+#endif
+
 /* Main transmit function */
 static
 netdev_tx_t qede_start_xmit(struct sk_buff *skb,
@@ -407,16 +429,22 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb,
 	txq = QEDE_TX_QUEUE(edev, txq_index);
 	netdev_txq = netdev_get_tx_queue(ndev, txq_index);
 
-	/* Current code doesn't support SKB linearization, since the max number
-	 * of skb frags can be passed in the FW HSI.
-	 */
-	BUILD_BUG_ON(MAX_SKB_FRAGS > ETH_TX_MAX_BDS_PER_NON_LSO_PACKET);
-
 	WARN_ON(qed_chain_get_elem_left(&txq->tx_pbl) <
 			       (MAX_SKB_FRAGS + 1));
 
 	xmit_type = qede_xmit_type(edev, skb, &ipv6_ext);
 
+#if ((MAX_SKB_FRAGS + 2) > ETH_TX_MAX_BDS_PER_NON_LSO_PACKET)
+	if (qede_pkt_req_lin(edev, skb, xmit_type)) {
+		if (skb_linearize(skb)) {
+			DP_NOTICE(edev,
+				  "SKB linearization failed - silently dropping this SKB\n");
+			dev_kfree_skb_any(skb);
+			return NETDEV_TX_OK;
+		}
+	}
+#endif
+
 	/* Fill the entry in the SW ring and the BDs in the FW ring */
 	idx = txq->sw_tx_prod & NUM_TX_BDS_MAX;
 	txq->sw_tx_ring[idx].skb = skb;
-- 
cgit v0.10.2


From 8e025ae28ae866c6a01239c4df6cf465a7deb02f Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 24 Feb 2016 16:52:47 +0200
Subject: qede: Don't report link change needlessly

There are several corner cases where driver might get a 2nd notification
about the same link change. Don't log any additional changes if the
physical carrier is already reported as it should.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index f07b9a9..c0dd236 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -2780,13 +2780,17 @@ static void qede_link_update(void *dev, struct qed_link_output *link)
 	}
 
 	if (link->link_up) {
-		DP_NOTICE(edev, "Link is up\n");
-		netif_tx_start_all_queues(edev->ndev);
-		netif_carrier_on(edev->ndev);
+		if (!netif_carrier_ok(edev->ndev)) {
+			DP_NOTICE(edev, "Link is up\n");
+			netif_tx_start_all_queues(edev->ndev);
+			netif_carrier_on(edev->ndev);
+		}
 	} else {
-		DP_NOTICE(edev, "Link is down\n");
-		netif_tx_disable(edev->ndev);
-		netif_carrier_off(edev->ndev);
+		if (netif_carrier_ok(edev->ndev)) {
+			DP_NOTICE(edev, "Link is down\n");
+			netif_tx_disable(edev->ndev);
+			netif_carrier_off(edev->ndev);
+		}
 	}
 }
 
-- 
cgit v0.10.2


From d43d3f0f393b21ee14c0487d5757edae194c4848 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 24 Feb 2016 16:52:48 +0200
Subject: qed: add MODULE_FIRMWARE()

Module is using a binary firmware file and so should be marked as such.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index 593f887..08cd92d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -45,6 +45,8 @@ MODULE_VERSION(DRV_MODULE_VERSION);
 #define QED_FW_FILE_NAME	\
 	"qed/qed_init_values_zipped-" FW_FILE_VERSION ".bin"
 
+MODULE_FIRMWARE(QED_FW_FILE_NAME);
+
 static int __init qed_init(void)
 {
 	pr_notice("qed_init called\n");
-- 
cgit v0.10.2


From 0dfaba6d0b4755fb379bd90dd5451e077617003f Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 24 Feb 2016 16:52:49 +0200
Subject: qed: Prevent probe on previous error

Don't allow driver to probe on an adapter at a failed state;
Gracefully block the probe instead.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index 08cd92d..e4e6ca5 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -99,12 +99,15 @@ static void qed_free_pci(struct qed_dev *cdev)
 	pci_disable_device(pdev);
 }
 
+#define PCI_REVISION_ID_ERROR_VAL	0xff
+
 /* Performs PCI initializations as well as initializing PCI-related parameters
  * in the device structrue. Returns 0 in case of success.
  */
 static int qed_init_pci(struct qed_dev *cdev,
 			struct pci_dev *pdev)
 {
+	u8 rev_id;
 	int rc;
 
 	cdev->pdev = pdev;
@@ -138,6 +141,14 @@ static int qed_init_pci(struct qed_dev *cdev,
 		pci_save_state(pdev);
 	}
 
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &rev_id);
+	if (rev_id == PCI_REVISION_ID_ERROR_VAL) {
+		DP_NOTICE(cdev,
+			  "Detected PCI device error [rev_id 0x%x]. Probably due to prior indication. Aborting.\n",
+			  rev_id);
+		rc = -ENODEV;
+		goto err2;
+	}
 	if (!pci_is_pcie(pdev)) {
 		DP_NOTICE(cdev, "The bus is not PCI Express\n");
 		rc = -EIO;
-- 
cgit v0.10.2


From 5abd7e92f46527baf88eadfe6a36ee15f0d7e89f Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 24 Feb 2016 16:52:50 +0200
Subject: qed, qede: rebrand module description

Drop the `QL4xxx 40G/100G' and use `FastLinQ 4xxxx' instead.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index e4e6ca5..25d6e91 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -29,10 +29,10 @@
 #include "qed_mcp.h"
 #include "qed_hw.h"
 
-static const char version[] =
-	"QLogic QL4xxx 40G/100G Ethernet Driver qed " DRV_MODULE_VERSION "\n";
+static char version[] =
+	"QLogic FastLinQ 4xxxx Core Module qed " DRV_MODULE_VERSION "\n";
 
-MODULE_DESCRIPTION("QLogic 25G/40G/50G/100G Core Module");
+MODULE_DESCRIPTION("QLogic FastLinQ 4xxxx Core Module");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_MODULE_VERSION);
 
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index c0dd236..ddd9e4a 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -39,10 +39,10 @@
 
 #include "qede.h"
 
-static const char version[] = "QLogic QL4xxx 40G/100G Ethernet Driver qede "
-			      DRV_MODULE_VERSION "\n";
+static char version[] =
+	"QLogic FastLinQ 4xxxx Ethernet Driver qede " DRV_MODULE_VERSION "\n";
 
-MODULE_DESCRIPTION("QLogic 40G/100G Ethernet Driver");
+MODULE_DESCRIPTION("QLogic FastLinQ 4xxxx Ethernet Driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_MODULE_VERSION);
 
-- 
cgit v0.10.2


From b170997acedc6c11ed2ec07b8d415601e65bb452 Mon Sep 17 00:00:00 2001
From: Jon Paul Maloy <jon.maloy@ericsson.com>
Date: Wed, 24 Feb 2016 11:00:19 -0500
Subject: tipc: eliminate risk of finding to-be-deleted node instance

Although we have never seen it happen, we have identified the
following problematic scenario when nodes are stopped and deleted:

CPU0:                            CPU1:

tipc_node_xxx()                                   //ref == 1
   tipc_node_put()                                //ref -> 0
                                 tipc_node_find() // node still in table
       tipc_node_delete()
         list_del_rcu(n. list)
                                 tipc_node_get()  //ref -> 1, bad
         kfree_rcu()

                                 tipc_node_put() //ref to 0 again.
                                 kfree_rcu()     // BOOM!

We fix this by introducing use of the conditional kref_get_if_not_zero()
instead of kref_get() in the function tipc_node_find(). This eliminates
any risk of post-mortem access.

Reported-by: Zhijiang Hu <huzhijiang@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/node.c b/net/tipc/node.c
index 9fcc2fb..792bbcb 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -245,23 +245,23 @@ static void tipc_node_get(struct tipc_node *node)
  */
 static struct tipc_node *tipc_node_find(struct net *net, u32 addr)
 {
-	struct tipc_net *tn = net_generic(net, tipc_net_id);
+	struct tipc_net *tn = tipc_net(net);
 	struct tipc_node *node;
+	unsigned int thash = tipc_hashfn(addr);
 
 	if (unlikely(!in_own_cluster_exact(net, addr)))
 		return NULL;
 
 	rcu_read_lock();
-	hlist_for_each_entry_rcu(node, &tn->node_htable[tipc_hashfn(addr)],
-				 hash) {
-		if (node->addr == addr) {
-			tipc_node_get(node);
-			rcu_read_unlock();
-			return node;
-		}
+	hlist_for_each_entry_rcu(node, &tn->node_htable[thash], hash) {
+		if (node->addr != addr)
+			continue;
+		if (!kref_get_unless_zero(&node->kref))
+			node = NULL;
+		break;
 	}
 	rcu_read_unlock();
-	return NULL;
+	return node;
 }
 
 static void tipc_node_read_lock(struct tipc_node *n)
-- 
cgit v0.10.2


From d25a01257e422a4bdeb426f69529d57c73b235fe Mon Sep 17 00:00:00 2001
From: Jon Paul Maloy <jon.maloy@ericsson.com>
Date: Wed, 24 Feb 2016 11:10:48 -0500
Subject: tipc: fix crash during node removal

When the TIPC module is unloaded, we have identified a race condition
that allows a node reference counter to go to zero and the node instance
being freed before the node timer is finished with accessing it. This
leads to occasional crashes, especially in multi-namespace environments.

The scenario goes as follows:

CPU0:(node_stop)                       CPU1:(node_timeout)  // ref == 2

1:                                          if(!mod_timer())
2: if (del_timer())
3:   tipc_node_put()                                        // ref -> 1
4: tipc_node_put()                                          // ref -> 0
5:   kfree_rcu(node);
6:                                               tipc_node_get(node)
7:                                               // BOOM!

We now clean up this functionality as follows:

1) We remove the node pointer from the node lookup table before we
   attempt deactivating the timer. This way, we reduce the risk that
   tipc_node_find() may obtain a valid pointer to an instance marked
   for deletion; a harmless but undesirable situation.

2) We use del_timer_sync() instead of del_timer() to safely deactivate
   the node timer without any risk that it might be reactivated by the
   timeout handler. There is no risk of deadlock here, since the two
   functions never touch the same spinlocks.

3: We remove a pointless tipc_node_get() + tipc_node_put() from the
   timeout handler.

Reported-by: Zhijiang Hu <huzhijiang@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/node.c b/net/tipc/node.c
index 792bbcb..cdb7950 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -225,9 +225,10 @@ static unsigned int tipc_hashfn(u32 addr)
 
 static void tipc_node_kref_release(struct kref *kref)
 {
-	struct tipc_node *node = container_of(kref, struct tipc_node, kref);
+	struct tipc_node *n = container_of(kref, struct tipc_node, kref);
 
-	tipc_node_delete(node);
+	kfree(n->bc_entry.link);
+	kfree_rcu(n, rcu);
 }
 
 static void tipc_node_put(struct tipc_node *node)
@@ -395,21 +396,20 @@ static void tipc_node_delete(struct tipc_node *node)
 {
 	list_del_rcu(&node->list);
 	hlist_del_rcu(&node->hash);
-	kfree(node->bc_entry.link);
-	kfree_rcu(node, rcu);
+	tipc_node_put(node);
+
+	del_timer_sync(&node->timer);
+	tipc_node_put(node);
 }
 
 void tipc_node_stop(struct net *net)
 {
-	struct tipc_net *tn = net_generic(net, tipc_net_id);
+	struct tipc_net *tn = tipc_net(net);
 	struct tipc_node *node, *t_node;
 
 	spin_lock_bh(&tn->node_list_lock);
-	list_for_each_entry_safe(node, t_node, &tn->node_list, list) {
-		if (del_timer(&node->timer))
-			tipc_node_put(node);
-		tipc_node_put(node);
-	}
+	list_for_each_entry_safe(node, t_node, &tn->node_list, list)
+		tipc_node_delete(node);
 	spin_unlock_bh(&tn->node_list_lock);
 }
 
@@ -530,9 +530,7 @@ static void tipc_node_timeout(unsigned long data)
 		if (rc & TIPC_LINK_DOWN_EVT)
 			tipc_node_link_down(n, bearer_id, false);
 	}
-	if (!mod_timer(&n->timer, jiffies + n->keepalive_intv))
-		tipc_node_get(n);
-	tipc_node_put(n);
+	mod_timer(&n->timer, jiffies + n->keepalive_intv);
 }
 
 /**
-- 
cgit v0.10.2


From 619b17452a388d6be69a7b28dc92310ea3f5aa65 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 24 Feb 2016 17:20:17 +0100
Subject: tipc: fix null deref crash in compat config path

msg.dst_sk needs to be set up with a valid socket because some callbacks
later derive the netns from it.

Fixes: 263ea09084d172d ("Revert "genl: Add genlmsg_new_unicast() for unicast message allocation")
Reported-by: Jon Maloy <maloy@donjonn.com>
Bisected-by: Jon Maloy <maloy@donjonn.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/netlink_compat.c b/net/tipc/netlink_compat.c
index de66d8f..d7d050f 100644
--- a/net/tipc/netlink_compat.c
+++ b/net/tipc/netlink_compat.c
@@ -1105,6 +1105,7 @@ static int tipc_nl_compat_recv(struct sk_buff *skb, struct genl_info *info)
 	msg.req = nlmsg_data(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN;
 	msg.cmd = req_userhdr->cmd;
 	msg.net = genl_info_net(info);
+	msg.dst_sk = skb->sk;
 
 	if ((msg.cmd & 0xC000) && (!netlink_net_capable(skb, CAP_NET_ADMIN))) {
 		msg.rep = tipc_get_err_tlv(TIPC_CFG_NOT_NET_ADMIN);
-- 
cgit v0.10.2


From f1705ec197e705b79ea40fe7a2cc5acfa1d3bfac Mon Sep 17 00:00:00 2001
From: David Ahern <dsa@cumulusnetworks.com>
Date: Wed, 24 Feb 2016 09:25:37 -0800
Subject: net: ipv6: Make address flushing on ifdown optional

Currently, all ipv6 addresses are flushed when the interface is configured
down, including global, static addresses:

    $ ip -6 addr show dev eth1
    3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
        inet6 2100:1::2/120 scope global
           valid_lft forever preferred_lft forever
        inet6 fe80::e0:f9ff:fe79:34bd/64 scope link
           valid_lft forever preferred_lft forever
    $ ip link set dev eth1 down
    $ ip -6 addr show dev eth1
    << nothing; all addresses have been flushed>>

Add a new sysctl to make this behavior optional. The new setting defaults to
flush all addresses to maintain backwards compatibility. When the set global
addresses with no expire times are not flushed on an admin down. The sysctl
is per-interface or system-wide for all interfaces

    $ sysctl -w net.ipv6.conf.eth1.keep_addr_on_down=1
or
    $ sysctl -w net.ipv6.conf.all.keep_addr_on_down=1

Will keep addresses on eth1 on an admin down.

    $ ip -6 addr show dev eth1
    3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
        inet6 2100:1::2/120 scope global
           valid_lft forever preferred_lft forever
        inet6 fe80::e0:f9ff:fe79:34bd/64 scope link
           valid_lft forever preferred_lft forever
    $ ip link set dev eth1 down
    $ ip -6 addr show dev eth1
    3: eth1: <BROADCAST,MULTICAST> mtu 1500 state DOWN qlen 1000
        inet6 2100:1::2/120 scope global tentative
           valid_lft forever preferred_lft forever
        inet6 fe80::e0:f9ff:fe79:34bd/64 scope link tentative
           valid_lft forever preferred_lft forever

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 24ce97f..d5df40c 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1563,6 +1563,15 @@ temp_prefered_lft - INTEGER
 	Preferred lifetime (in seconds) for temporary addresses.
 	Default: 86400 (1 day)
 
+keep_addr_on_down - INTEGER
+	Keep all IPv6 addresses on an interface down event. If set static
+	global addresses with no expiration time are not flushed.
+	  >0 : enabled
+	   0 : system default
+	  <0 : disabled
+
+	Default: 0 (addresses are removed)
+
 max_desync_factor - INTEGER
 	Maximum value for DESYNC_FACTOR, which is a random value
 	that ensures that clients don't synchronize with each
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 4b2267e..7edc14f 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -62,6 +62,7 @@ struct ipv6_devconf {
 		struct in6_addr secret;
 	} stable_secret;
 	__s32		use_oif_addrs_only;
+	__s32		keep_addr_on_down;
 	void		*sysctl;
 };
 
diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h
index ec117b6..3958760 100644
--- a/include/uapi/linux/ipv6.h
+++ b/include/uapi/linux/ipv6.h
@@ -176,6 +176,7 @@ enum {
 	DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN,
 	DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
 	DEVCONF_DROP_UNSOLICITED_NA,
+	DEVCONF_KEEP_ADDR_ON_DOWN,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 4751f89..a2d6f6c 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -216,6 +216,7 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = {
 	},
 	.use_oif_addrs_only	= 0,
 	.ignore_routes_with_linkdown = 0,
+	.keep_addr_on_down	= 0,
 };
 
 static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
@@ -260,6 +261,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	},
 	.use_oif_addrs_only	= 0,
 	.ignore_routes_with_linkdown = 0,
+	.keep_addr_on_down	= 0,
 };
 
 /* Check if a valid qdisc is available */
@@ -3168,6 +3170,55 @@ static void addrconf_gre_config(struct net_device *dev)
 }
 #endif
 
+static int fixup_permanent_addr(struct inet6_dev *idev,
+				struct inet6_ifaddr *ifp)
+{
+	if (!ifp->rt) {
+		struct rt6_info *rt;
+
+		rt = addrconf_dst_alloc(idev, &ifp->addr, false);
+		if (unlikely(IS_ERR(rt)))
+			return PTR_ERR(rt);
+
+		ifp->rt = rt;
+	}
+
+	if (!(ifp->flags & IFA_F_NOPREFIXROUTE)) {
+		addrconf_prefix_route(&ifp->addr, ifp->prefix_len,
+				      idev->dev, 0, 0);
+	}
+
+	addrconf_dad_start(ifp);
+
+	return 0;
+}
+
+static void addrconf_permanent_addr(struct net_device *dev)
+{
+	struct inet6_ifaddr *ifp, *tmp;
+	struct inet6_dev *idev;
+
+	idev = __in6_dev_get(dev);
+	if (!idev)
+		return;
+
+	write_lock_bh(&idev->lock);
+
+	list_for_each_entry_safe(ifp, tmp, &idev->addr_list, if_list) {
+		if ((ifp->flags & IFA_F_PERMANENT) &&
+		    fixup_permanent_addr(idev, ifp) < 0) {
+			write_unlock_bh(&idev->lock);
+			ipv6_del_addr(ifp);
+			write_lock_bh(&idev->lock);
+
+			net_info_ratelimited("%s: Failed to add prefix route for address %pI6c; dropping\n",
+					     idev->dev->name, &ifp->addr);
+		}
+	}
+
+	write_unlock_bh(&idev->lock);
+}
+
 static int addrconf_notify(struct notifier_block *this, unsigned long event,
 			   void *ptr)
 {
@@ -3253,6 +3304,9 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event,
 			run_pending = 1;
 		}
 
+		/* restore routes for permanent addresses */
+		addrconf_permanent_addr(dev);
+
 		switch (dev->type) {
 #if IS_ENABLED(CONFIG_IPV6_SIT)
 		case ARPHRD_SIT:
@@ -3356,7 +3410,10 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 {
 	struct net *net = dev_net(dev);
 	struct inet6_dev *idev;
-	struct inet6_ifaddr *ifa;
+	struct inet6_ifaddr *ifa, *tmp;
+	struct list_head del_list;
+	int _keep_addr;
+	bool keep_addr;
 	int state, i;
 
 	ASSERT_RTNL();
@@ -3383,6 +3440,16 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 
 	}
 
+	/* aggregate the system setting and interface setting */
+	_keep_addr = net->ipv6.devconf_all->keep_addr_on_down;
+	if (!_keep_addr)
+		_keep_addr = idev->cnf.keep_addr_on_down;
+
+	/* combine the user config with event to determine if permanent
+	 * addresses are to be removed from address hash table
+	 */
+	keep_addr = !(how || _keep_addr <= 0);
+
 	/* Step 2: clear hash table */
 	for (i = 0; i < IN6_ADDR_HSIZE; i++) {
 		struct hlist_head *h = &inet6_addr_lst[i];
@@ -3391,9 +3458,15 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 restart:
 		hlist_for_each_entry_rcu(ifa, h, addr_lst) {
 			if (ifa->idev == idev) {
-				hlist_del_init_rcu(&ifa->addr_lst);
 				addrconf_del_dad_work(ifa);
-				goto restart;
+				/* combined flag + permanent flag decide if
+				 * address is retained on a down event
+				 */
+				if (!keep_addr ||
+				    !(ifa->flags & IFA_F_PERMANENT)) {
+					hlist_del_init_rcu(&ifa->addr_lst);
+					goto restart;
+				}
 			}
 		}
 		spin_unlock_bh(&addrconf_hash_lock);
@@ -3427,31 +3500,53 @@ restart:
 		write_lock_bh(&idev->lock);
 	}
 
-	while (!list_empty(&idev->addr_list)) {
-		ifa = list_first_entry(&idev->addr_list,
-				       struct inet6_ifaddr, if_list);
-		addrconf_del_dad_work(ifa);
+	/* re-combine the user config with event to determine if permanent
+	 * addresses are to be removed from the interface list
+	 */
+	keep_addr = (!how && _keep_addr > 0);
 
-		list_del(&ifa->if_list);
+	INIT_LIST_HEAD(&del_list);
+	list_for_each_entry_safe(ifa, tmp, &idev->addr_list, if_list) {
+		addrconf_del_dad_work(ifa);
 
 		write_unlock_bh(&idev->lock);
-
 		spin_lock_bh(&ifa->lock);
-		state = ifa->state;
-		ifa->state = INET6_IFADDR_STATE_DEAD;
+
+		if (keep_addr && (ifa->flags & IFA_F_PERMANENT)) {
+			/* set state to skip the notifier below */
+			state = INET6_IFADDR_STATE_DEAD;
+			ifa->state = 0;
+			if (!(ifa->flags & IFA_F_NODAD))
+				ifa->flags |= IFA_F_TENTATIVE;
+		} else {
+			state = ifa->state;
+			ifa->state = INET6_IFADDR_STATE_DEAD;
+
+			list_del(&ifa->if_list);
+			list_add(&ifa->if_list, &del_list);
+		}
+
 		spin_unlock_bh(&ifa->lock);
 
 		if (state != INET6_IFADDR_STATE_DEAD) {
 			__ipv6_ifa_notify(RTM_DELADDR, ifa);
 			inet6addr_notifier_call_chain(NETDEV_DOWN, ifa);
 		}
-		in6_ifa_put(ifa);
 
 		write_lock_bh(&idev->lock);
 	}
 
 	write_unlock_bh(&idev->lock);
 
+	/* now clean up addresses to be removed */
+	while (!list_empty(&del_list)) {
+		ifa = list_first_entry(&del_list,
+				       struct inet6_ifaddr, if_list);
+		list_del(&ifa->if_list);
+
+		in6_ifa_put(ifa);
+	}
+
 	/* Step 5: Discard anycast and multicast list */
 	if (how) {
 		ipv6_ac_destroy_dev(idev);
@@ -4716,6 +4811,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_USE_OIF_ADDRS_ONLY] = cnf->use_oif_addrs_only;
 	array[DEVCONF_DROP_UNICAST_IN_L2_MULTICAST] = cnf->drop_unicast_in_l2_multicast;
 	array[DEVCONF_DROP_UNSOLICITED_NA] = cnf->drop_unsolicited_na;
+	array[DEVCONF_KEEP_ADDR_ON_DOWN] = cnf->keep_addr_on_down;
 }
 
 static inline size_t inet6_ifla6_size(void)
@@ -5197,10 +5293,12 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 			if (rt)
 				ip6_del_rt(rt);
 		}
-		dst_hold(&ifp->rt->dst);
-
-		ip6_del_rt(ifp->rt);
+		if (ifp->rt) {
+			dst_hold(&ifp->rt->dst);
 
+			ip6_del_rt(ifp->rt);
+			ifp->rt = NULL;
+		}
 		rt_genid_bump_ipv6(net);
 		break;
 	}
@@ -5804,6 +5902,14 @@ static struct addrconf_sysctl_table
 			.proc_handler	= proc_dointvec,
 		},
 		{
+			.procname       = "keep_addr_on_down",
+			.data           = &ipv6_devconf.keep_addr_on_down,
+			.maxlen         = sizeof(int),
+			.mode           = 0644,
+			.proc_handler   = proc_dointvec,
+
+		},
+		{
 			/* sentinel */
 		}
 	},
-- 
cgit v0.10.2


From a87cb3e48ee86d29868d3f59cfb9ce1a8fa63314 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Wed, 24 Feb 2016 10:02:52 -0800
Subject: net: Facility to report route quality of connected sockets

This patch add the SO_CNX_ADVICE socket option (setsockopt only). The
purpose is to allow an application to give feedback to the kernel about
the quality of the network path for a connected socket. The value
argument indicates the type of quality report. For this initial patch
the only supported advice is a value of 1 which indicates "bad path,
please reroute"-- the action taken by the kernel is to call
dst_negative_advice which will attempt to choose a different ECMP route,
reset the TX hash for flow label and UDP source port in encapsulation,
etc.

This facility should be useful for connected UDP sockets where only the
application can provide any feedback about path quality. It could also
be useful for TCP applications that have additional knowledge about the
path outside of the normal TCP control loop.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/alpha/include/uapi/asm/socket.h b/arch/alpha/include/uapi/asm/socket.h
index c5fb9e6..9e46d6e 100644
--- a/arch/alpha/include/uapi/asm/socket.h
+++ b/arch/alpha/include/uapi/asm/socket.h
@@ -95,4 +95,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _UAPI_ASM_SOCKET_H */
diff --git a/arch/avr32/include/uapi/asm/socket.h b/arch/avr32/include/uapi/asm/socket.h
index 9de0796..1fd147f 100644
--- a/arch/avr32/include/uapi/asm/socket.h
+++ b/arch/avr32/include/uapi/asm/socket.h
@@ -88,4 +88,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _UAPI__ASM_AVR32_SOCKET_H */
diff --git a/arch/frv/include/uapi/asm/socket.h b/arch/frv/include/uapi/asm/socket.h
index f02e484..afbc98f0 100644
--- a/arch/frv/include/uapi/asm/socket.h
+++ b/arch/frv/include/uapi/asm/socket.h
@@ -88,5 +88,7 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _ASM_SOCKET_H */
 
diff --git a/arch/ia64/include/uapi/asm/socket.h b/arch/ia64/include/uapi/asm/socket.h
index bce2916..0018fad 100644
--- a/arch/ia64/include/uapi/asm/socket.h
+++ b/arch/ia64/include/uapi/asm/socket.h
@@ -97,4 +97,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _ASM_IA64_SOCKET_H */
diff --git a/arch/m32r/include/uapi/asm/socket.h b/arch/m32r/include/uapi/asm/socket.h
index 14aa4a6..5fe42fc 100644
--- a/arch/m32r/include/uapi/asm/socket.h
+++ b/arch/m32r/include/uapi/asm/socket.h
@@ -88,4 +88,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _ASM_M32R_SOCKET_H */
diff --git a/arch/mips/include/uapi/asm/socket.h b/arch/mips/include/uapi/asm/socket.h
index 5910fe2..2027240a 100644
--- a/arch/mips/include/uapi/asm/socket.h
+++ b/arch/mips/include/uapi/asm/socket.h
@@ -106,4 +106,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _UAPI_ASM_SOCKET_H */
diff --git a/arch/mn10300/include/uapi/asm/socket.h b/arch/mn10300/include/uapi/asm/socket.h
index 58b1aa0..5129f23 100644
--- a/arch/mn10300/include/uapi/asm/socket.h
+++ b/arch/mn10300/include/uapi/asm/socket.h
@@ -88,4 +88,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _ASM_SOCKET_H */
diff --git a/arch/parisc/include/uapi/asm/socket.h b/arch/parisc/include/uapi/asm/socket.h
index f9cf122..9c935d7 100644
--- a/arch/parisc/include/uapi/asm/socket.h
+++ b/arch/parisc/include/uapi/asm/socket.h
@@ -87,4 +87,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	0x402C
 #define SO_ATTACH_REUSEPORT_EBPF	0x402D
 
+#define SO_CNX_ADVICE		0x402E
+
 #endif /* _UAPI_ASM_SOCKET_H */
diff --git a/arch/powerpc/include/uapi/asm/socket.h b/arch/powerpc/include/uapi/asm/socket.h
index dd54f28..1672e33 100644
--- a/arch/powerpc/include/uapi/asm/socket.h
+++ b/arch/powerpc/include/uapi/asm/socket.h
@@ -95,4 +95,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif	/* _ASM_POWERPC_SOCKET_H */
diff --git a/arch/s390/include/uapi/asm/socket.h b/arch/s390/include/uapi/asm/socket.h
index d02e89d..41b51c2 100644
--- a/arch/s390/include/uapi/asm/socket.h
+++ b/arch/s390/include/uapi/asm/socket.h
@@ -94,4 +94,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* _ASM_SOCKET_H */
diff --git a/arch/sparc/include/uapi/asm/socket.h b/arch/sparc/include/uapi/asm/socket.h
index d270ee9..31aede3 100644
--- a/arch/sparc/include/uapi/asm/socket.h
+++ b/arch/sparc/include/uapi/asm/socket.h
@@ -84,6 +84,8 @@
 #define SO_ATTACH_REUSEPORT_CBPF	0x0035
 #define SO_ATTACH_REUSEPORT_EBPF	0x0036
 
+#define SO_CNX_ADVICE		0x0037
+
 /* Security levels - as per NRL IPv6 - don't actually do anything */
 #define SO_SECURITY_AUTHENTICATION		0x5001
 #define SO_SECURITY_ENCRYPTION_TRANSPORT	0x5002
diff --git a/arch/xtensa/include/uapi/asm/socket.h b/arch/xtensa/include/uapi/asm/socket.h
index fd3b96d..81435d9 100644
--- a/arch/xtensa/include/uapi/asm/socket.h
+++ b/arch/xtensa/include/uapi/asm/socket.h
@@ -99,4 +99,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif	/* _XTENSA_SOCKET_H */
diff --git a/include/uapi/asm-generic/socket.h b/include/uapi/asm-generic/socket.h
index fb8a416..67d632f 100644
--- a/include/uapi/asm-generic/socket.h
+++ b/include/uapi/asm-generic/socket.h
@@ -90,4 +90,6 @@
 #define SO_ATTACH_REUSEPORT_CBPF	51
 #define SO_ATTACH_REUSEPORT_EBPF	52
 
+#define SO_CNX_ADVICE		53
+
 #endif /* __ASM_GENERIC_SOCKET_H */
diff --git a/net/core/sock.c b/net/core/sock.c
index 46dc8ad..4493ff8 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -987,6 +987,10 @@ set_rcvbuf:
 		sk->sk_incoming_cpu = val;
 		break;
 
+	case SO_CNX_ADVICE:
+		if (val == 1)
+			dst_negative_advice(sk);
+		break;
 	default:
 		ret = -ENOPROTOOPT;
 		break;
-- 
cgit v0.10.2


From 4f03980ca4d7f56222c8397d7b91f15efafa2918 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:57:57 -0800
Subject: net: usnic: remove unused call to ethtool_ops::get_settings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
index 6cdb4d2..ea003ec 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
@@ -269,7 +269,6 @@ int usnic_ib_query_device(struct ib_device *ibdev,
 	struct usnic_ib_dev *us_ibdev = to_usdev(ibdev);
 	union ib_gid gid;
 	struct ethtool_drvinfo info;
-	struct ethtool_cmd cmd;
 	int qp_per_vf;
 
 	usnic_dbg("\n");
@@ -278,7 +277,6 @@ int usnic_ib_query_device(struct ib_device *ibdev,
 
 	mutex_lock(&us_ibdev->usdev_lock);
 	us_ibdev->netdev->ethtool_ops->get_drvinfo(us_ibdev->netdev, &info);
-	us_ibdev->netdev->ethtool_ops->get_settings(us_ibdev->netdev, &cmd);
 	memset(props, 0, sizeof(*props));
 	usnic_mac_ip_to_gid(us_ibdev->ufdev->mac, us_ibdev->ufdev->inaddr,
 			&gid.raw[0]);
-- 
cgit v0.10.2


From 48133335d7e0a661f40a5899d20945dabb600f0a Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:57:58 -0800
Subject: net: usnic: use __ethtool_get_settings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
index ea003ec..1cf19a3 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
@@ -329,7 +329,7 @@ int usnic_ib_query_port(struct ib_device *ibdev, u8 port,
 	usnic_dbg("\n");
 
 	mutex_lock(&us_ibdev->usdev_lock);
-	us_ibdev->netdev->ethtool_ops->get_settings(us_ibdev->netdev, &cmd);
+	__ethtool_get_settings(us_ibdev->netdev, &cmd);
 	memset(props, 0, sizeof(*props));
 
 	props->lid = 0;
-- 
cgit v0.10.2


From 3f1ac7a700d039c61d8d8b99f28d605d489a60cf Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:57:59 -0800
Subject: net: ethtool: add new ETHTOOL_xLINKSETTINGS API

This patch defines a new ETHTOOL_GLINKSETTINGS/SLINKSETTINGS API,
handled by the new get_link_ksettings/set_link_ksettings callbacks.
This API provides support for most legacy ethtool_cmd fields, adds
support for larger link mode masks (up to 4064 bits, variable length),
and removes ethtool_cmd deprecated
fields (transceiver/maxrxpkt/maxtxpkt).

This API is deprecating the legacy ETHTOOL_GSET/SSET API and provides
the following backward compatibility properties:
 - legacy ethtool with legacy drivers: no change, still using the
   get_settings/set_settings callbacks.
 - legacy ethtool with new get/set_link_ksettings drivers: the new
   driver callbacks are used, data internally converted to legacy
   ethtool_cmd. ETHTOOL_GSET will return only the 1st 32b of each link
   mode mask. ETHTOOL_SSET will fail if user tries to set the
   ethtool_cmd deprecated fields to
   non-0 (transceiver/maxrxpkt/maxtxpkt). A kernel warning is logged if
   driver sets higher bits.
 - future ethtool with legacy drivers: no change, still using the
   get_settings/set_settings callbacks, internally converted to new data
   structure. Deprecated fields (transceiver/maxrxpkt/maxtxpkt) will be
   ignored and seen as 0 from user space. Note that that "future"
   ethtool tool will not allow changes to these deprecated fields.
 - future ethtool with new drivers: direct call to the new callbacks.

By "future" ethtool, what is meant is:
 - query: first try ETHTOOL_GLINKSETTINGS, and revert to ETHTOOL_GSET if
   fails
 - set: query first and remember which of ETHTOOL_GLINKSETTINGS or
   ETHTOOL_GSET was successful
   + if ETHTOOL_GLINKSETTINGS was successful, then change config with
     ETHTOOL_SLINKSETTINGS. A failure there is final (do not try
     ETHTOOL_SSET).
   + otherwise ETHTOOL_GSET was successful, change config with
     ETHTOOL_SSET. A failure there is final (do not try
     ETHTOOL_SLINKSETTINGS).

The interaction user/kernel via the new API requires a small
ETHTOOL_GLINKSETTINGS handshake first to agree on the length of the link
mode bitmaps. If kernel doesn't agree with user, it returns the bitmap
length it is expecting from user as a negative length (and cmd field is
0). When kernel and user agree, kernel returns valid info in all
fields (ie. link mode length > 0 and cmd is ETHTOOL_GLINKSETTINGS).

Data structure crossing user/kernel boundary is 32/64-bit
agnostic. Converted internally to a legal kernel bitmap.

The internal __ethtool_get_settings kernel helper will gradually be
replaced by __ethtool_get_link_ksettings by the time the first
"link_settings" drivers start to appear. So this patch doesn't change
it, it will be removed before it needs to be changed.

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 472d7d7..8a400a5 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -12,6 +12,7 @@
 #ifndef _LINUX_ETHTOOL_H
 #define _LINUX_ETHTOOL_H
 
+#include <linux/bitmap.h>
 #include <linux/compat.h>
 #include <uapi/linux/ethtool.h>
 
@@ -40,9 +41,6 @@ struct compat_ethtool_rxnfc {
 
 #include <linux/rculist.h>
 
-extern int __ethtool_get_settings(struct net_device *dev,
-				  struct ethtool_cmd *cmd);
-
 /**
  * enum ethtool_phys_id_state - indicator state for physical identification
  * @ETHTOOL_ID_INACTIVE: Physical ID indicator should be deactivated
@@ -97,13 +95,74 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
 	return index % n_rx_rings;
 }
 
+/* number of link mode bits/ulongs handled internally by kernel */
+#define __ETHTOOL_LINK_MODE_MASK_NBITS			\
+	(__ETHTOOL_LINK_MODE_LAST + 1)
+
+/* declare a link mode bitmap */
+#define __ETHTOOL_DECLARE_LINK_MODE_MASK(name)		\
+	DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS)
+
+/* drivers must ignore base.cmd and base.link_mode_masks_nwords
+ * fields, but they are allowed to overwrite them (will be ignored).
+ */
+struct ethtool_link_ksettings {
+	struct ethtool_link_settings base;
+	struct {
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(supported);
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(lp_advertising);
+	} link_modes;
+};
+
+/**
+ * ethtool_link_ksettings_zero_link_mode - clear link_ksettings link mode mask
+ *   @ptr : pointer to struct ethtool_link_ksettings
+ *   @name : one of supported/advertising/lp_advertising
+ */
+#define ethtool_link_ksettings_zero_link_mode(ptr, name)		\
+	bitmap_zero((ptr)->link_modes.name, __ETHTOOL_LINK_MODE_MASK_NBITS)
+
+/**
+ * ethtool_link_ksettings_add_link_mode - set bit in link_ksettings
+ * link mode mask
+ *   @ptr : pointer to struct ethtool_link_ksettings
+ *   @name : one of supported/advertising/lp_advertising
+ *   @mode : one of the ETHTOOL_LINK_MODE_*_BIT
+ * (not atomic, no bound checking)
+ */
+#define ethtool_link_ksettings_add_link_mode(ptr, name, mode)		\
+	__set_bit(ETHTOOL_LINK_MODE_ ## mode ## _BIT, (ptr)->link_modes.name)
+
+/**
+ * ethtool_link_ksettings_test_link_mode - test bit in ksettings link mode mask
+ *   @ptr : pointer to struct ethtool_link_ksettings
+ *   @name : one of supported/advertising/lp_advertising
+ *   @mode : one of the ETHTOOL_LINK_MODE_*_BIT
+ * (not atomic, no bound checking)
+ *
+ * Returns true/false.
+ */
+#define ethtool_link_ksettings_test_link_mode(ptr, name, mode)		\
+	test_bit(ETHTOOL_LINK_MODE_ ## mode ## _BIT, (ptr)->link_modes.name)
+
+extern int
+__ethtool_get_link_ksettings(struct net_device *dev,
+			     struct ethtool_link_ksettings *link_ksettings);
+
+/* DEPRECATED, use __ethtool_get_link_ksettings */
+extern int __ethtool_get_settings(struct net_device *dev,
+				  struct ethtool_cmd *cmd);
+
 /**
  * struct ethtool_ops - optional netdev operations
- * @get_settings: Get various device settings including Ethernet link
+ * @get_settings: DEPRECATED, use %get_link_ksettings/%set_link_ksettings
+ *	API. Get various device settings including Ethernet link
  *	settings. The @cmd parameter is expected to have been cleared
- *	before get_settings is called. Returns a negative error code or
- *	zero.
- * @set_settings: Set various device settings including Ethernet link
+ *	before get_settings is called. Returns a negative error code
+ *	or zero.
+ * @set_settings: DEPRECATED, use %get_link_ksettings/%set_link_ksettings
+ *	API. Set various device settings including Ethernet link
  *	settings.  Returns a negative error code or zero.
  * @get_drvinfo: Report driver/device information.  Should only set the
  *	@driver, @version, @fw_version and @bus_info fields.  If not
@@ -211,6 +270,19 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
  *	a TX queue has this number, return -EINVAL. If only a RX queue or a TX
  *	queue has this number, ignore the inapplicable fields.
  *	Returns a negative error code or zero.
+ * @get_link_ksettings: When defined, takes precedence over the
+ *	%get_settings method. Get various device settings
+ *	including Ethernet link settings. The %cmd and
+ *	%link_mode_masks_nwords fields should be ignored (use
+ *	%__ETHTOOL_LINK_MODE_MASK_NBITS instead of the latter), any
+ *	change to them will be overwritten by kernel. Returns a
+ *	negative error code or zero.
+ * @set_link_ksettings: When defined, takes precedence over the
+ *	%set_settings method. Set various device settings including
+ *	Ethernet link settings. The %cmd and %link_mode_masks_nwords
+ *	fields should be ignored (use %__ETHTOOL_LINK_MODE_MASK_NBITS
+ *	instead of the latter), any change to them will be overwritten
+ *	by kernel. Returns a negative error code or zero.
  *
  * All operations are optional (i.e. the function pointer may be set
  * to %NULL) and callers must take this into account.  Callers must
@@ -293,6 +365,9 @@ struct ethtool_ops {
 					  struct ethtool_coalesce *);
 	int	(*set_per_queue_coalesce)(struct net_device *, u32,
 					  struct ethtool_coalesce *);
-
+	int	(*get_link_ksettings)(struct net_device *,
+				      struct ethtool_link_ksettings *);
+	int	(*set_link_ksettings)(struct net_device *,
+				      const struct ethtool_link_ksettings *);
 };
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index f15ae02..37fd6dc 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -21,7 +21,8 @@
  */
 
 /**
- * struct ethtool_cmd - link control and status
+ * struct ethtool_cmd - DEPRECATED, link control and status
+ * This structure is DEPRECATED, please use struct ethtool_link_settings.
  * @cmd: Command number = %ETHTOOL_GSET or %ETHTOOL_SSET
  * @supported: Bitmask of %SUPPORTED_* flags for the link modes,
  *	physical connectors and other link features for which the
@@ -1219,8 +1220,12 @@ struct ethtool_per_queue_op {
 };
 
 /* CMDs currently supported */
-#define ETHTOOL_GSET		0x00000001 /* Get settings. */
-#define ETHTOOL_SSET		0x00000002 /* Set settings. */
+#define ETHTOOL_GSET		0x00000001 /* DEPRECATED, Get settings.
+					    * Please use ETHTOOL_GLINKSETTINGS
+					    */
+#define ETHTOOL_SSET		0x00000002 /* DEPRECATED, Set settings.
+					    * Please use ETHTOOL_SLINKSETTINGS
+					    */
 #define ETHTOOL_GDRVINFO	0x00000003 /* Get driver info. */
 #define ETHTOOL_GREGS		0x00000004 /* Get NIC registers. */
 #define ETHTOOL_GWOL		0x00000005 /* Get wake-on-lan options. */
@@ -1302,73 +1307,139 @@ struct ethtool_per_queue_op {
 
 #define ETHTOOL_PERQUEUE	0x0000004b /* Set per queue options */
 
+#define ETHTOOL_GLINKSETTINGS	0x0000004c /* Get ethtool_link_settings */
+#define ETHTOOL_SLINKSETTINGS	0x0000004d /* Set ethtool_link_settings */
+
+
 /* compatibility with older code */
 #define SPARC_ETH_GSET		ETHTOOL_GSET
 #define SPARC_ETH_SSET		ETHTOOL_SSET
 
-#define SUPPORTED_10baseT_Half		(1 << 0)
-#define SUPPORTED_10baseT_Full		(1 << 1)
-#define SUPPORTED_100baseT_Half		(1 << 2)
-#define SUPPORTED_100baseT_Full		(1 << 3)
-#define SUPPORTED_1000baseT_Half	(1 << 4)
-#define SUPPORTED_1000baseT_Full	(1 << 5)
-#define SUPPORTED_Autoneg		(1 << 6)
-#define SUPPORTED_TP			(1 << 7)
-#define SUPPORTED_AUI			(1 << 8)
-#define SUPPORTED_MII			(1 << 9)
-#define SUPPORTED_FIBRE			(1 << 10)
-#define SUPPORTED_BNC			(1 << 11)
-#define SUPPORTED_10000baseT_Full	(1 << 12)
-#define SUPPORTED_Pause			(1 << 13)
-#define SUPPORTED_Asym_Pause		(1 << 14)
-#define SUPPORTED_2500baseX_Full	(1 << 15)
-#define SUPPORTED_Backplane		(1 << 16)
-#define SUPPORTED_1000baseKX_Full	(1 << 17)
-#define SUPPORTED_10000baseKX4_Full	(1 << 18)
-#define SUPPORTED_10000baseKR_Full	(1 << 19)
-#define SUPPORTED_10000baseR_FEC	(1 << 20)
-#define SUPPORTED_20000baseMLD2_Full	(1 << 21)
-#define SUPPORTED_20000baseKR2_Full	(1 << 22)
-#define SUPPORTED_40000baseKR4_Full	(1 << 23)
-#define SUPPORTED_40000baseCR4_Full	(1 << 24)
-#define SUPPORTED_40000baseSR4_Full	(1 << 25)
-#define SUPPORTED_40000baseLR4_Full	(1 << 26)
-#define SUPPORTED_56000baseKR4_Full	(1 << 27)
-#define SUPPORTED_56000baseCR4_Full	(1 << 28)
-#define SUPPORTED_56000baseSR4_Full	(1 << 29)
-#define SUPPORTED_56000baseLR4_Full	(1 << 30)
-
-#define ADVERTISED_10baseT_Half		(1 << 0)
-#define ADVERTISED_10baseT_Full		(1 << 1)
-#define ADVERTISED_100baseT_Half	(1 << 2)
-#define ADVERTISED_100baseT_Full	(1 << 3)
-#define ADVERTISED_1000baseT_Half	(1 << 4)
-#define ADVERTISED_1000baseT_Full	(1 << 5)
-#define ADVERTISED_Autoneg		(1 << 6)
-#define ADVERTISED_TP			(1 << 7)
-#define ADVERTISED_AUI			(1 << 8)
-#define ADVERTISED_MII			(1 << 9)
-#define ADVERTISED_FIBRE		(1 << 10)
-#define ADVERTISED_BNC			(1 << 11)
-#define ADVERTISED_10000baseT_Full	(1 << 12)
-#define ADVERTISED_Pause		(1 << 13)
-#define ADVERTISED_Asym_Pause		(1 << 14)
-#define ADVERTISED_2500baseX_Full	(1 << 15)
-#define ADVERTISED_Backplane		(1 << 16)
-#define ADVERTISED_1000baseKX_Full	(1 << 17)
-#define ADVERTISED_10000baseKX4_Full	(1 << 18)
-#define ADVERTISED_10000baseKR_Full	(1 << 19)
-#define ADVERTISED_10000baseR_FEC	(1 << 20)
-#define ADVERTISED_20000baseMLD2_Full	(1 << 21)
-#define ADVERTISED_20000baseKR2_Full	(1 << 22)
-#define ADVERTISED_40000baseKR4_Full	(1 << 23)
-#define ADVERTISED_40000baseCR4_Full	(1 << 24)
-#define ADVERTISED_40000baseSR4_Full	(1 << 25)
-#define ADVERTISED_40000baseLR4_Full	(1 << 26)
-#define ADVERTISED_56000baseKR4_Full	(1 << 27)
-#define ADVERTISED_56000baseCR4_Full	(1 << 28)
-#define ADVERTISED_56000baseSR4_Full	(1 << 29)
-#define ADVERTISED_56000baseLR4_Full	(1 << 30)
+/* Link mode bit indices */
+enum ethtool_link_mode_bit_indices {
+	ETHTOOL_LINK_MODE_10baseT_Half_BIT	= 0,
+	ETHTOOL_LINK_MODE_10baseT_Full_BIT	= 1,
+	ETHTOOL_LINK_MODE_100baseT_Half_BIT	= 2,
+	ETHTOOL_LINK_MODE_100baseT_Full_BIT	= 3,
+	ETHTOOL_LINK_MODE_1000baseT_Half_BIT	= 4,
+	ETHTOOL_LINK_MODE_1000baseT_Full_BIT	= 5,
+	ETHTOOL_LINK_MODE_Autoneg_BIT		= 6,
+	ETHTOOL_LINK_MODE_TP_BIT		= 7,
+	ETHTOOL_LINK_MODE_AUI_BIT		= 8,
+	ETHTOOL_LINK_MODE_MII_BIT		= 9,
+	ETHTOOL_LINK_MODE_FIBRE_BIT		= 10,
+	ETHTOOL_LINK_MODE_BNC_BIT		= 11,
+	ETHTOOL_LINK_MODE_10000baseT_Full_BIT	= 12,
+	ETHTOOL_LINK_MODE_Pause_BIT		= 13,
+	ETHTOOL_LINK_MODE_Asym_Pause_BIT	= 14,
+	ETHTOOL_LINK_MODE_2500baseX_Full_BIT	= 15,
+	ETHTOOL_LINK_MODE_Backplane_BIT		= 16,
+	ETHTOOL_LINK_MODE_1000baseKX_Full_BIT	= 17,
+	ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT	= 18,
+	ETHTOOL_LINK_MODE_10000baseKR_Full_BIT	= 19,
+	ETHTOOL_LINK_MODE_10000baseR_FEC_BIT	= 20,
+	ETHTOOL_LINK_MODE_20000baseMLD2_Full_BIT = 21,
+	ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT	= 22,
+	ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT	= 23,
+	ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT	= 24,
+	ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT	= 25,
+	ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT	= 26,
+	ETHTOOL_LINK_MODE_56000baseKR4_Full_BIT	= 27,
+	ETHTOOL_LINK_MODE_56000baseCR4_Full_BIT	= 28,
+	ETHTOOL_LINK_MODE_56000baseSR4_Full_BIT	= 29,
+	ETHTOOL_LINK_MODE_56000baseLR4_Full_BIT	= 30,
+
+	/* Last allowed bit for __ETHTOOL_LINK_MODE_LEGACY_MASK is bit
+	 * 31. Please do NOT define any SUPPORTED_* or ADVERTISED_*
+	 * macro for bits > 31. The only way to use indices > 31 is to
+	 * use the new ETHTOOL_GLINKSETTINGS/ETHTOOL_SLINKSETTINGS API.
+	 */
+
+	__ETHTOOL_LINK_MODE_LAST
+	  = ETHTOOL_LINK_MODE_56000baseLR4_Full_BIT,
+};
+
+#define __ETHTOOL_LINK_MODE_LEGACY_MASK(base_name)	\
+	(1UL << (ETHTOOL_LINK_MODE_ ## base_name ## _BIT))
+
+/* DEPRECATED macros. Please migrate to
+ * ETHTOOL_GLINKSETTINGS/ETHTOOL_SLINKSETTINGS API. Please do NOT
+ * define any new SUPPORTED_* macro for bits > 31.
+ */
+#define SUPPORTED_10baseT_Half		__ETHTOOL_LINK_MODE_LEGACY_MASK(10baseT_Half)
+#define SUPPORTED_10baseT_Full		__ETHTOOL_LINK_MODE_LEGACY_MASK(10baseT_Full)
+#define SUPPORTED_100baseT_Half		__ETHTOOL_LINK_MODE_LEGACY_MASK(100baseT_Half)
+#define SUPPORTED_100baseT_Full		__ETHTOOL_LINK_MODE_LEGACY_MASK(100baseT_Full)
+#define SUPPORTED_1000baseT_Half	__ETHTOOL_LINK_MODE_LEGACY_MASK(1000baseT_Half)
+#define SUPPORTED_1000baseT_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(1000baseT_Full)
+#define SUPPORTED_Autoneg		__ETHTOOL_LINK_MODE_LEGACY_MASK(Autoneg)
+#define SUPPORTED_TP			__ETHTOOL_LINK_MODE_LEGACY_MASK(TP)
+#define SUPPORTED_AUI			__ETHTOOL_LINK_MODE_LEGACY_MASK(AUI)
+#define SUPPORTED_MII			__ETHTOOL_LINK_MODE_LEGACY_MASK(MII)
+#define SUPPORTED_FIBRE			__ETHTOOL_LINK_MODE_LEGACY_MASK(FIBRE)
+#define SUPPORTED_BNC			__ETHTOOL_LINK_MODE_LEGACY_MASK(BNC)
+#define SUPPORTED_10000baseT_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseT_Full)
+#define SUPPORTED_Pause			__ETHTOOL_LINK_MODE_LEGACY_MASK(Pause)
+#define SUPPORTED_Asym_Pause		__ETHTOOL_LINK_MODE_LEGACY_MASK(Asym_Pause)
+#define SUPPORTED_2500baseX_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(2500baseX_Full)
+#define SUPPORTED_Backplane		__ETHTOOL_LINK_MODE_LEGACY_MASK(Backplane)
+#define SUPPORTED_1000baseKX_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(1000baseKX_Full)
+#define SUPPORTED_10000baseKX4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseKX4_Full)
+#define SUPPORTED_10000baseKR_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseKR_Full)
+#define SUPPORTED_10000baseR_FEC	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseR_FEC)
+#define SUPPORTED_20000baseMLD2_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(20000baseMLD2_Full)
+#define SUPPORTED_20000baseKR2_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(20000baseKR2_Full)
+#define SUPPORTED_40000baseKR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseKR4_Full)
+#define SUPPORTED_40000baseCR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseCR4_Full)
+#define SUPPORTED_40000baseSR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseSR4_Full)
+#define SUPPORTED_40000baseLR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseLR4_Full)
+#define SUPPORTED_56000baseKR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseKR4_Full)
+#define SUPPORTED_56000baseCR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseCR4_Full)
+#define SUPPORTED_56000baseSR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseSR4_Full)
+#define SUPPORTED_56000baseLR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseLR4_Full)
+/* Please do not define any new SUPPORTED_* macro for bits > 31, see
+ * notice above.
+ */
+
+/*
+ * DEPRECATED macros. Please migrate to
+ * ETHTOOL_GLINKSETTINGS/ETHTOOL_SLINKSETTINGS API. Please do NOT
+ * define any new ADERTISE_* macro for bits > 31.
+ */
+#define ADVERTISED_10baseT_Half		__ETHTOOL_LINK_MODE_LEGACY_MASK(10baseT_Half)
+#define ADVERTISED_10baseT_Full		__ETHTOOL_LINK_MODE_LEGACY_MASK(10baseT_Full)
+#define ADVERTISED_100baseT_Half	__ETHTOOL_LINK_MODE_LEGACY_MASK(100baseT_Half)
+#define ADVERTISED_100baseT_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(100baseT_Full)
+#define ADVERTISED_1000baseT_Half	__ETHTOOL_LINK_MODE_LEGACY_MASK(1000baseT_Half)
+#define ADVERTISED_1000baseT_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(1000baseT_Full)
+#define ADVERTISED_Autoneg		__ETHTOOL_LINK_MODE_LEGACY_MASK(Autoneg)
+#define ADVERTISED_TP			__ETHTOOL_LINK_MODE_LEGACY_MASK(TP)
+#define ADVERTISED_AUI			__ETHTOOL_LINK_MODE_LEGACY_MASK(AUI)
+#define ADVERTISED_MII			__ETHTOOL_LINK_MODE_LEGACY_MASK(MII)
+#define ADVERTISED_FIBRE		__ETHTOOL_LINK_MODE_LEGACY_MASK(FIBRE)
+#define ADVERTISED_BNC			__ETHTOOL_LINK_MODE_LEGACY_MASK(BNC)
+#define ADVERTISED_10000baseT_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseT_Full)
+#define ADVERTISED_Pause		__ETHTOOL_LINK_MODE_LEGACY_MASK(Pause)
+#define ADVERTISED_Asym_Pause		__ETHTOOL_LINK_MODE_LEGACY_MASK(Asym_Pause)
+#define ADVERTISED_2500baseX_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(2500baseX_Full)
+#define ADVERTISED_Backplane		__ETHTOOL_LINK_MODE_LEGACY_MASK(Backplane)
+#define ADVERTISED_1000baseKX_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(1000baseKX_Full)
+#define ADVERTISED_10000baseKX4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseKX4_Full)
+#define ADVERTISED_10000baseKR_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseKR_Full)
+#define ADVERTISED_10000baseR_FEC	__ETHTOOL_LINK_MODE_LEGACY_MASK(10000baseR_FEC)
+#define ADVERTISED_20000baseMLD2_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(20000baseMLD2_Full)
+#define ADVERTISED_20000baseKR2_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(20000baseKR2_Full)
+#define ADVERTISED_40000baseKR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseKR4_Full)
+#define ADVERTISED_40000baseCR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseCR4_Full)
+#define ADVERTISED_40000baseSR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseSR4_Full)
+#define ADVERTISED_40000baseLR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(40000baseLR4_Full)
+#define ADVERTISED_56000baseKR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseKR4_Full)
+#define ADVERTISED_56000baseCR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseCR4_Full)
+#define ADVERTISED_56000baseSR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseSR4_Full)
+#define ADVERTISED_56000baseLR4_Full	__ETHTOOL_LINK_MODE_LEGACY_MASK(56000baseLR4_Full)
+/* Please do not define any new ADVERTISED_* macro for bits > 31, see
+ * notice above.
+ */
 
 /* The following are all involved in forcing a particular link
  * mode for the device for setting things.  When getting the
@@ -1533,4 +1604,123 @@ enum ethtool_reset_flags {
 };
 #define ETH_RESET_SHARED_SHIFT	16
 
+
+/**
+ * struct ethtool_link_settings - link control and status
+ *
+ * IMPORTANT, Backward compatibility notice: When implementing new
+ *	user-space tools, please first try %ETHTOOL_GLINKSETTINGS, and
+ *	if it succeeds use %ETHTOOL_SLINKSETTINGS to change link
+ *	settings; do not use %ETHTOOL_SSET if %ETHTOOL_GLINKSETTINGS
+ *	succeeded: stick to %ETHTOOL_GLINKSETTINGS/%SLINKSETTINGS in
+ *	that case.  Conversely, if %ETHTOOL_GLINKSETTINGS fails, use
+ *	%ETHTOOL_GSET to query and %ETHTOOL_SSET to change link
+ *	settings; do not use %ETHTOOL_SLINKSETTINGS if
+ *	%ETHTOOL_GLINKSETTINGS failed: stick to
+ *	%ETHTOOL_GSET/%ETHTOOL_SSET in that case.
+ *
+ * @cmd: Command number = %ETHTOOL_GLINKSETTINGS or %ETHTOOL_SLINKSETTINGS
+ * @speed: Link speed (Mbps)
+ * @duplex: Duplex mode; one of %DUPLEX_*
+ * @port: Physical connector type; one of %PORT_*
+ * @phy_address: MDIO address of PHY (transceiver); 0 or 255 if not
+ *	applicable.  For clause 45 PHYs this is the PRTAD.
+ * @autoneg: Enable/disable autonegotiation and auto-detection;
+ *	either %AUTONEG_DISABLE or %AUTONEG_ENABLE
+ * @mdio_support: Bitmask of %ETH_MDIO_SUPPORTS_* flags for the MDIO
+ *	protocols supported by the interface; 0 if unknown.
+ *	Read-only.
+ * @eth_tp_mdix: Ethernet twisted-pair MDI(-X) status; one of
+ *	%ETH_TP_MDI_*.  If the status is unknown or not applicable, the
+ *	value will be %ETH_TP_MDI_INVALID.  Read-only.
+ * @eth_tp_mdix_ctrl: Ethernet twisted pair MDI(-X) control; one of
+ *	%ETH_TP_MDI_*.  If MDI(-X) control is not implemented, reads
+ *	yield %ETH_TP_MDI_INVALID and writes may be ignored or rejected.
+ *	When written successfully, the link should be renegotiated if
+ *	necessary.
+ * @link_mode_masks_nwords: Number of 32-bit words for each of the
+ *	supported, advertising, lp_advertising link mode bitmaps. For
+ *	%ETHTOOL_GLINKSETTINGS: on entry, number of words passed by user
+ *	(>= 0); on return, if handshake in progress, negative if
+ *	request size unsupported by kernel: absolute value indicates
+ *	kernel recommended size and cmd field is 0, as well as all the
+ *	other fields; otherwise (handshake completed), strictly
+ *	positive to indicate size used by kernel and cmd field is
+ *	%ETHTOOL_GLINKSETTINGS, all other fields populated by driver. For
+ *	%ETHTOOL_SLINKSETTINGS: must be valid on entry, ie. a positive
+ *	value returned previously by %ETHTOOL_GLINKSETTINGS, otherwise
+ *	refused. For drivers: ignore this field (use kernel's
+ *	__ETHTOOL_LINK_MODE_MASK_NBITS instead), any change to it will
+ *	be overwritten by kernel.
+ * @supported: Bitmap with each bit meaning given by
+ *	%ethtool_link_mode_bit_indices for the link modes, physical
+ *	connectors and other link features for which the interface
+ *	supports autonegotiation or auto-detection.  Read-only.
+ * @advertising: Bitmap with each bit meaning given by
+ *	%ethtool_link_mode_bit_indices for the link modes, physical
+ *	connectors and other link features that are advertised through
+ *	autonegotiation or enabled for auto-detection.
+ * @lp_advertising: Bitmap with each bit meaning given by
+ *	%ethtool_link_mode_bit_indices for the link modes, and other
+ *	link features that the link partner advertised through
+ *	autonegotiation; 0 if unknown or not applicable.  Read-only.
+ *
+ * If autonegotiation is disabled, the speed and @duplex represent the
+ * fixed link mode and are writable if the driver supports multiple
+ * link modes.  If it is enabled then they are read-only; if the link
+ * is up they represent the negotiated link mode; if the link is down,
+ * the speed is 0, %SPEED_UNKNOWN or the highest enabled speed and
+ * @duplex is %DUPLEX_UNKNOWN or the best enabled duplex mode.
+ *
+ * Some hardware interfaces may have multiple PHYs and/or physical
+ * connectors fitted or do not allow the driver to detect which are
+ * fitted.  For these interfaces @port and/or @phy_address may be
+ * writable, possibly dependent on @autoneg being %AUTONEG_DISABLE.
+ * Otherwise, attempts to write different values may be ignored or
+ * rejected.
+ *
+ * Deprecated %ethtool_cmd fields transceiver, maxtxpkt and maxrxpkt
+ * are not available in %ethtool_link_settings. Until all drivers are
+ * converted to ignore them or to the new %ethtool_link_settings API,
+ * for both queries and changes, users should always try
+ * %ETHTOOL_GLINKSETTINGS first, and if it fails with -ENOTSUPP stick
+ * only to %ETHTOOL_GSET and %ETHTOOL_SSET consistently. If it
+ * succeeds, then users should stick to %ETHTOOL_GLINKSETTINGS and
+ * %ETHTOOL_SLINKSETTINGS (which would support drivers implementing
+ * either %ethtool_cmd or %ethtool_link_settings).
+ *
+ * Users should assume that all fields not marked read-only are
+ * writable and subject to validation by the driver.  They should use
+ * %ETHTOOL_GLINKSETTINGS to get the current values before making specific
+ * changes and then applying them with %ETHTOOL_SLINKSETTINGS.
+ *
+ * Drivers that implement %get_link_ksettings and/or
+ * %set_link_ksettings should ignore the @cmd
+ * and @link_mode_masks_nwords fields (any change to them overwritten
+ * by kernel), and rely only on kernel's internal
+ * %__ETHTOOL_LINK_MODE_MASK_NBITS and
+ * %ethtool_link_mode_mask_t. Drivers that implement
+ * %set_link_ksettings() should validate all fields other than @cmd
+ * and @link_mode_masks_nwords that are not described as read-only or
+ * deprecated, and must ignore all fields described as read-only.
+ */
+struct ethtool_link_settings {
+	__u32	cmd;
+	__u32	speed;
+	__u8	duplex;
+	__u8	port;
+	__u8	phy_address;
+	__u8	autoneg;
+	__u8	mdio_support;
+	__u8	eth_tp_mdix;
+	__u8	eth_tp_mdix_ctrl;
+	__s8	link_mode_masks_nwords;
+	__u32	reserved[8];
+	__u32	link_mode_masks[0];
+	/* layout of link_mode_masks fields:
+	 * __u32 map_supported[link_mode_masks_nwords];
+	 * __u32 map_advertising[link_mode_masks_nwords];
+	 * __u32 map_lp_advertising[link_mode_masks_nwords];
+	 */
+};
 #endif /* _UAPI_LINUX_ETHTOOL_H */
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 2406101..edcec56 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -387,6 +387,359 @@ static int __ethtool_set_flags(struct net_device *dev, u32 data)
 	return 0;
 }
 
+static void convert_legacy_u32_to_link_mode(unsigned long *dst, u32 legacy_u32)
+{
+	bitmap_zero(dst, __ETHTOOL_LINK_MODE_MASK_NBITS);
+	dst[0] = legacy_u32;
+}
+
+/* return false if src had higher bits set. lower bits always updated. */
+static bool convert_link_mode_to_legacy_u32(u32 *legacy_u32,
+					    const unsigned long *src)
+{
+	bool retval = true;
+
+	/* TODO: following test will soon always be true */
+	if (__ETHTOOL_LINK_MODE_MASK_NBITS > 32) {
+		__ETHTOOL_DECLARE_LINK_MODE_MASK(ext);
+
+		bitmap_zero(ext, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		bitmap_fill(ext, 32);
+		bitmap_complement(ext, ext, __ETHTOOL_LINK_MODE_MASK_NBITS);
+		if (bitmap_intersects(ext, src,
+				      __ETHTOOL_LINK_MODE_MASK_NBITS)) {
+			/* src mask goes beyond bit 31 */
+			retval = false;
+		}
+	}
+	*legacy_u32 = src[0];
+	return retval;
+}
+
+/* return false if legacy contained non-0 deprecated fields
+ * transceiver/maxtxpkt/maxrxpkt. rest of ksettings always updated
+ */
+static bool
+convert_legacy_settings_to_link_ksettings(
+	struct ethtool_link_ksettings *link_ksettings,
+	const struct ethtool_cmd *legacy_settings)
+{
+	bool retval = true;
+
+	memset(link_ksettings, 0, sizeof(*link_ksettings));
+
+	/* This is used to tell users that driver is still using these
+	 * deprecated legacy fields, and they should not use
+	 * %ETHTOOL_GLINKSETTINGS/%ETHTOOL_SLINKSETTINGS
+	 */
+	if (legacy_settings->transceiver ||
+	    legacy_settings->maxtxpkt ||
+	    legacy_settings->maxrxpkt)
+		retval = false;
+
+	convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.supported,
+		legacy_settings->supported);
+	convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.advertising,
+		legacy_settings->advertising);
+	convert_legacy_u32_to_link_mode(
+		link_ksettings->link_modes.lp_advertising,
+		legacy_settings->lp_advertising);
+	link_ksettings->base.speed
+		= ethtool_cmd_speed(legacy_settings);
+	link_ksettings->base.duplex
+		= legacy_settings->duplex;
+	link_ksettings->base.port
+		= legacy_settings->port;
+	link_ksettings->base.phy_address
+		= legacy_settings->phy_address;
+	link_ksettings->base.autoneg
+		= legacy_settings->autoneg;
+	link_ksettings->base.mdio_support
+		= legacy_settings->mdio_support;
+	link_ksettings->base.eth_tp_mdix
+		= legacy_settings->eth_tp_mdix;
+	link_ksettings->base.eth_tp_mdix_ctrl
+		= legacy_settings->eth_tp_mdix_ctrl;
+	return retval;
+}
+
+/* return false if ksettings link modes had higher bits
+ * set. legacy_settings always updated (best effort)
+ */
+static bool
+convert_link_ksettings_to_legacy_settings(
+	struct ethtool_cmd *legacy_settings,
+	const struct ethtool_link_ksettings *link_ksettings)
+{
+	bool retval = true;
+
+	memset(legacy_settings, 0, sizeof(*legacy_settings));
+	/* this also clears the deprecated fields in legacy structure:
+	 * __u8		transceiver;
+	 * __u32	maxtxpkt;
+	 * __u32	maxrxpkt;
+	 */
+
+	retval &= convert_link_mode_to_legacy_u32(
+		&legacy_settings->supported,
+		link_ksettings->link_modes.supported);
+	retval &= convert_link_mode_to_legacy_u32(
+		&legacy_settings->advertising,
+		link_ksettings->link_modes.advertising);
+	retval &= convert_link_mode_to_legacy_u32(
+		&legacy_settings->lp_advertising,
+		link_ksettings->link_modes.lp_advertising);
+	ethtool_cmd_speed_set(legacy_settings, link_ksettings->base.speed);
+	legacy_settings->duplex
+		= link_ksettings->base.duplex;
+	legacy_settings->port
+		= link_ksettings->base.port;
+	legacy_settings->phy_address
+		= link_ksettings->base.phy_address;
+	legacy_settings->autoneg
+		= link_ksettings->base.autoneg;
+	legacy_settings->mdio_support
+		= link_ksettings->base.mdio_support;
+	legacy_settings->eth_tp_mdix
+		= link_ksettings->base.eth_tp_mdix;
+	legacy_settings->eth_tp_mdix_ctrl
+		= link_ksettings->base.eth_tp_mdix_ctrl;
+	return retval;
+}
+
+/* number of 32-bit words to store the user's link mode bitmaps */
+#define __ETHTOOL_LINK_MODE_MASK_NU32			\
+	DIV_ROUND_UP(__ETHTOOL_LINK_MODE_MASK_NBITS, 32)
+
+/* layout of the struct passed from/to userland */
+struct ethtool_link_usettings {
+	struct ethtool_link_settings base;
+	struct {
+		__u32 supported[__ETHTOOL_LINK_MODE_MASK_NU32];
+		__u32 advertising[__ETHTOOL_LINK_MODE_MASK_NU32];
+		__u32 lp_advertising[__ETHTOOL_LINK_MODE_MASK_NU32];
+	} link_modes;
+};
+
+/* Internal kernel helper to query a device ethtool_link_settings.
+ *
+ * Backward compatibility note: for compatibility with legacy drivers
+ * that implement only the ethtool_cmd API, this has to work with both
+ * drivers implementing get_link_ksettings API and drivers
+ * implementing get_settings API. When drivers implement get_settings
+ * and report ethtool_cmd deprecated fields
+ * (transceiver/maxrxpkt/maxtxpkt), these fields are silently ignored
+ * because the resulting struct ethtool_link_settings does not report them.
+ */
+int __ethtool_get_link_ksettings(struct net_device *dev,
+				 struct ethtool_link_ksettings *link_ksettings)
+{
+	int err;
+	struct ethtool_cmd cmd;
+
+	ASSERT_RTNL();
+
+	if (dev->ethtool_ops->get_link_ksettings) {
+		memset(link_ksettings, 0, sizeof(*link_ksettings));
+		return dev->ethtool_ops->get_link_ksettings(dev,
+							    link_ksettings);
+	}
+
+	/* driver doesn't support %ethtool_link_ksettings API. revert to
+	 * legacy %ethtool_cmd API, unless it's not supported either.
+	 * TODO: remove when ethtool_ops::get_settings disappears internally
+	 */
+	err = __ethtool_get_settings(dev, &cmd);
+	if (err < 0)
+		return err;
+
+	/* we ignore deprecated fields transceiver/maxrxpkt/maxtxpkt
+	 */
+	convert_legacy_settings_to_link_ksettings(link_ksettings, &cmd);
+	return err;
+}
+EXPORT_SYMBOL(__ethtool_get_link_ksettings);
+
+/* convert ethtool_link_usettings in user space to a kernel internal
+ * ethtool_link_ksettings. return 0 on success, errno on error.
+ */
+static int load_link_ksettings_from_user(struct ethtool_link_ksettings *to,
+					 const void __user *from)
+{
+	struct ethtool_link_usettings link_usettings;
+
+	if (copy_from_user(&link_usettings, from, sizeof(link_usettings)))
+		return -EFAULT;
+
+	memcpy(&to->base, &link_usettings.base, sizeof(to->base));
+	bitmap_from_u32array(to->link_modes.supported,
+			     __ETHTOOL_LINK_MODE_MASK_NBITS,
+			     link_usettings.link_modes.supported,
+			     __ETHTOOL_LINK_MODE_MASK_NU32);
+	bitmap_from_u32array(to->link_modes.advertising,
+			     __ETHTOOL_LINK_MODE_MASK_NBITS,
+			     link_usettings.link_modes.advertising,
+			     __ETHTOOL_LINK_MODE_MASK_NU32);
+	bitmap_from_u32array(to->link_modes.lp_advertising,
+			     __ETHTOOL_LINK_MODE_MASK_NBITS,
+			     link_usettings.link_modes.lp_advertising,
+			     __ETHTOOL_LINK_MODE_MASK_NU32);
+
+	return 0;
+}
+
+/* convert a kernel internal ethtool_link_ksettings to
+ * ethtool_link_usettings in user space. return 0 on success, errno on
+ * error.
+ */
+static int
+store_link_ksettings_for_user(void __user *to,
+			      const struct ethtool_link_ksettings *from)
+{
+	struct ethtool_link_usettings link_usettings;
+
+	memcpy(&link_usettings.base, &from->base, sizeof(link_usettings));
+	bitmap_to_u32array(link_usettings.link_modes.supported,
+			   __ETHTOOL_LINK_MODE_MASK_NU32,
+			   from->link_modes.supported,
+			   __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_to_u32array(link_usettings.link_modes.advertising,
+			   __ETHTOOL_LINK_MODE_MASK_NU32,
+			   from->link_modes.advertising,
+			   __ETHTOOL_LINK_MODE_MASK_NBITS);
+	bitmap_to_u32array(link_usettings.link_modes.lp_advertising,
+			   __ETHTOOL_LINK_MODE_MASK_NU32,
+			   from->link_modes.lp_advertising,
+			   __ETHTOOL_LINK_MODE_MASK_NBITS);
+
+	if (copy_to_user(to, &link_usettings, sizeof(link_usettings)))
+		return -EFAULT;
+
+	return 0;
+}
+
+/* Query device for its ethtool_link_settings.
+ *
+ * Backward compatibility note: this function must fail when driver
+ * does not implement ethtool::get_link_ksettings, even if legacy
+ * ethtool_ops::get_settings is implemented. This tells new versions
+ * of ethtool that they should use the legacy API %ETHTOOL_GSET for
+ * this driver, so that they can correctly access the ethtool_cmd
+ * deprecated fields (transceiver/maxrxpkt/maxtxpkt), until no driver
+ * implements ethtool_ops::get_settings anymore.
+ */
+static int ethtool_get_link_ksettings(struct net_device *dev,
+				      void __user *useraddr)
+{
+	int err = 0;
+	struct ethtool_link_ksettings link_ksettings;
+
+	ASSERT_RTNL();
+
+	if (!dev->ethtool_ops->get_link_ksettings)
+		return -EOPNOTSUPP;
+
+	/* handle bitmap nbits handshake */
+	if (copy_from_user(&link_ksettings.base, useraddr,
+			   sizeof(link_ksettings.base)))
+		return -EFAULT;
+
+	if (__ETHTOOL_LINK_MODE_MASK_NU32
+	    != link_ksettings.base.link_mode_masks_nwords) {
+		/* wrong link mode nbits requested */
+		memset(&link_ksettings, 0, sizeof(link_ksettings));
+		/* keep cmd field reset to 0 */
+		/* send back number of words required as negative val */
+		compiletime_assert(__ETHTOOL_LINK_MODE_MASK_NU32 <= S8_MAX,
+				   "need too many bits for link modes!");
+		link_ksettings.base.link_mode_masks_nwords
+			= -((s8)__ETHTOOL_LINK_MODE_MASK_NU32);
+
+		/* copy the base fields back to user, not the link
+		 * mode bitmaps
+		 */
+		if (copy_to_user(useraddr, &link_ksettings.base,
+				 sizeof(link_ksettings.base)))
+			return -EFAULT;
+
+		return 0;
+	}
+
+	/* handshake successful: user/kernel agree on
+	 * link_mode_masks_nwords
+	 */
+
+	memset(&link_ksettings, 0, sizeof(link_ksettings));
+	err = dev->ethtool_ops->get_link_ksettings(dev, &link_ksettings);
+	if (err < 0)
+		return err;
+
+	/* make sure we tell the right values to user */
+	link_ksettings.base.cmd = ETHTOOL_GLINKSETTINGS;
+	link_ksettings.base.link_mode_masks_nwords
+		= __ETHTOOL_LINK_MODE_MASK_NU32;
+
+	return store_link_ksettings_for_user(useraddr, &link_ksettings);
+}
+
+/* Update device ethtool_link_settings.
+ *
+ * Backward compatibility note: this function must fail when driver
+ * does not implement ethtool::set_link_ksettings, even if legacy
+ * ethtool_ops::set_settings is implemented. This tells new versions
+ * of ethtool that they should use the legacy API %ETHTOOL_SSET for
+ * this driver, so that they can correctly update the ethtool_cmd
+ * deprecated fields (transceiver/maxrxpkt/maxtxpkt), until no driver
+ * implements ethtool_ops::get_settings anymore.
+ */
+static int ethtool_set_link_ksettings(struct net_device *dev,
+				      void __user *useraddr)
+{
+	int err;
+	struct ethtool_link_ksettings link_ksettings;
+
+	ASSERT_RTNL();
+
+	if (!dev->ethtool_ops->set_link_ksettings)
+		return -EOPNOTSUPP;
+
+	/* make sure nbits field has expected value */
+	if (copy_from_user(&link_ksettings.base, useraddr,
+			   sizeof(link_ksettings.base)))
+		return -EFAULT;
+
+	if (__ETHTOOL_LINK_MODE_MASK_NU32
+	    != link_ksettings.base.link_mode_masks_nwords)
+		return -EINVAL;
+
+	/* copy the whole structure, now that we know it has expected
+	 * format
+	 */
+	err = load_link_ksettings_from_user(&link_ksettings, useraddr);
+	if (err)
+		return err;
+
+	/* re-check nwords field, just in case */
+	if (__ETHTOOL_LINK_MODE_MASK_NU32
+	    != link_ksettings.base.link_mode_masks_nwords)
+		return -EINVAL;
+
+	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
+}
+
+/* Internal kernel helper to query a device ethtool_cmd settings.
+ *
+ * Note about transition to ethtool_link_settings API: We do not need
+ * (or want) this function to support "dev" instances that implement
+ * the ethtool_link_settings API as we will update the drivers calling
+ * this function to call __ethtool_get_link_ksettings instead, before
+ * the first drivers implement ethtool_ops::get_link_ksettings.
+ *
+ * TODO 1: at least make this function static when no driver is using it
+ * TODO 2: remove when ethtool_ops::get_settings disappears internally
+ */
 int __ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 {
 	ASSERT_RTNL();
@@ -400,30 +753,112 @@ int __ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 }
 EXPORT_SYMBOL(__ethtool_get_settings);
 
+static void
+warn_incomplete_ethtool_legacy_settings_conversion(const char *details)
+{
+	char name[sizeof(current->comm)];
+
+	pr_info_once("warning: `%s' uses legacy ethtool link settings API, %s\n",
+		     get_task_comm(name, current), details);
+}
+
+/* Query device for its ethtool_cmd settings.
+ *
+ * Backward compatibility note: for compatibility with legacy ethtool,
+ * this has to work with both drivers implementing get_link_ksettings
+ * API and drivers implementing get_settings API. When drivers
+ * implement get_link_ksettings and report higher link mode bits, a
+ * kernel warning is logged once (with name of 1st driver/device) to
+ * recommend user to upgrade ethtool, but the command is successful
+ * (only the lower link mode bits reported back to user).
+ */
 static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
 {
-	int err;
 	struct ethtool_cmd cmd;
 
-	err = __ethtool_get_settings(dev, &cmd);
-	if (err < 0)
-		return err;
+	ASSERT_RTNL();
+
+	if (dev->ethtool_ops->get_link_ksettings) {
+		/* First, use link_ksettings API if it is supported */
+		int err;
+		struct ethtool_link_ksettings link_ksettings;
+
+		memset(&link_ksettings, 0, sizeof(link_ksettings));
+		err = dev->ethtool_ops->get_link_ksettings(dev,
+							   &link_ksettings);
+		if (err < 0)
+			return err;
+		if (!convert_link_ksettings_to_legacy_settings(&cmd,
+							       &link_ksettings))
+			warn_incomplete_ethtool_legacy_settings_conversion(
+				"link modes are only partially reported");
+
+		/* send a sensible cmd tag back to user */
+		cmd.cmd = ETHTOOL_GSET;
+	} else {
+		int err;
+		/* TODO: return -EOPNOTSUPP when
+		 * ethtool_ops::get_settings disappears internally
+		 */
+
+		/* driver doesn't support %ethtool_link_ksettings
+		 * API. revert to legacy %ethtool_cmd API, unless it's
+		 * not supported either.
+		 */
+		err = __ethtool_get_settings(dev, &cmd);
+		if (err < 0)
+			return err;
+	}
 
 	if (copy_to_user(useraddr, &cmd, sizeof(cmd)))
 		return -EFAULT;
+
 	return 0;
 }
 
+/* Update device link settings with given ethtool_cmd.
+ *
+ * Backward compatibility note: for compatibility with legacy ethtool,
+ * this has to work with both drivers implementing set_link_ksettings
+ * API and drivers implementing set_settings API. When drivers
+ * implement set_link_ksettings and user's request updates deprecated
+ * ethtool_cmd fields (transceiver/maxrxpkt/maxtxpkt), a kernel
+ * warning is logged once (with name of 1st driver/device) to
+ * recommend user to upgrade ethtool, and the request is rejected.
+ */
 static int ethtool_set_settings(struct net_device *dev, void __user *useraddr)
 {
 	struct ethtool_cmd cmd;
 
-	if (!dev->ethtool_ops->set_settings)
-		return -EOPNOTSUPP;
+	ASSERT_RTNL();
 
 	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
 		return -EFAULT;
 
+	/* first, try new %ethtool_link_ksettings API. */
+	if (dev->ethtool_ops->set_link_ksettings) {
+		struct ethtool_link_ksettings link_ksettings;
+
+		if (!convert_legacy_settings_to_link_ksettings(&link_ksettings,
+							       &cmd))
+			return -EINVAL;
+
+		link_ksettings.base.cmd = ETHTOOL_SLINKSETTINGS;
+		link_ksettings.base.link_mode_masks_nwords
+			= __ETHTOOL_LINK_MODE_MASK_NU32;
+		return dev->ethtool_ops->set_link_ksettings(dev,
+							    &link_ksettings);
+	}
+
+	/* legacy %ethtool_cmd API */
+
+	/* TODO: return -EOPNOTSUPP when ethtool_ops::get_settings
+	 * disappears internally
+	 */
+
+	if (!dev->ethtool_ops->set_settings)
+		return -EOPNOTSUPP;
+
 	return dev->ethtool_ops->set_settings(dev, &cmd);
 }
 
@@ -2252,6 +2687,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_PERQUEUE:
 		rc = ethtool_set_per_queue(dev, useraddr);
 		break;
+	case ETHTOOL_GLINKSETTINGS:
+		rc = ethtool_get_link_ksettings(dev, useraddr);
+		break;
+	case ETHTOOL_SLINKSETTINGS:
+		rc = ethtool_set_link_ksettings(dev, useraddr);
+		break;
 	default:
 		rc = -EOPNOTSUPP;
 	}
-- 
cgit v0.10.2


From 091a9277c69be8d9ae1debca4f3eb1fee82d366a Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:00 -0800
Subject: tx4939: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/mips/txx9/generic/setup_tx4939.c b/arch/mips/txx9/generic/setup_tx4939.c
index e3733cd..402ac2e 100644
--- a/arch/mips/txx9/generic/setup_tx4939.c
+++ b/arch/mips/txx9/generic/setup_tx4939.c
@@ -320,11 +320,12 @@ void __init tx4939_sio_init(unsigned int sclk, unsigned int cts_mask)
 #if IS_ENABLED(CONFIG_TC35815)
 static u32 tx4939_get_eth_speed(struct net_device *dev)
 {
-	struct ethtool_cmd cmd;
-	if (__ethtool_get_settings(dev, &cmd))
+	struct ethtool_link_ksettings cmd;
+
+	if (__ethtool_get_link_ksettings(dev, &cmd))
 		return 100;	/* default 100Mbps */
 
-	return ethtool_cmd_speed(&cmd);
+	return cmd.base.speed;
 }
 
 static int tx4939_netdev_event(struct notifier_block *this,
-- 
cgit v0.10.2


From 96a0c39633cd3a50f1129c5f542e3a653b9fb8c9 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:01 -0800
Subject: net: usnic: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
index 1cf19a3..a5bfbba 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_verbs.c
@@ -324,12 +324,12 @@ int usnic_ib_query_port(struct ib_device *ibdev, u8 port,
 				struct ib_port_attr *props)
 {
 	struct usnic_ib_dev *us_ibdev = to_usdev(ibdev);
-	struct ethtool_cmd cmd;
+	struct ethtool_link_ksettings cmd;
 
 	usnic_dbg("\n");
 
 	mutex_lock(&us_ibdev->usdev_lock);
-	__ethtool_get_settings(us_ibdev->netdev, &cmd);
+	__ethtool_get_link_ksettings(us_ibdev->netdev, &cmd);
 	memset(props, 0, sizeof(*props));
 
 	props->lid = 0;
@@ -353,8 +353,8 @@ int usnic_ib_query_port(struct ib_device *ibdev, u8 port,
 	props->pkey_tbl_len = 1;
 	props->bad_pkey_cntr = 0;
 	props->qkey_viol_cntr = 0;
-	eth_speed_to_ib_speed(cmd.speed, &props->active_speed,
-				&props->active_width);
+	eth_speed_to_ib_speed(cmd.base.speed, &props->active_speed,
+			      &props->active_width);
 	props->max_mtu = IB_MTU_4096;
 	props->active_mtu = iboe_get_mtu(us_ibdev->ufdev->mtu);
 	/* Userspace will adjust for hdrs */
-- 
cgit v0.10.2


From 9856909c2abb32a795d9782bff2ce987696ffb9b Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:02 -0800
Subject: net: bonding: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index a6527d5..b6236ff 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -376,22 +376,20 @@ down:
 static void bond_update_speed_duplex(struct slave *slave)
 {
 	struct net_device *slave_dev = slave->dev;
-	struct ethtool_cmd ecmd;
-	u32 slave_speed;
+	struct ethtool_link_ksettings ecmd;
 	int res;
 
 	slave->speed = SPEED_UNKNOWN;
 	slave->duplex = DUPLEX_UNKNOWN;
 
-	res = __ethtool_get_settings(slave_dev, &ecmd);
+	res = __ethtool_get_link_ksettings(slave_dev, &ecmd);
 	if (res < 0)
 		return;
 
-	slave_speed = ethtool_cmd_speed(&ecmd);
-	if (slave_speed == 0 || slave_speed == ((__u32) -1))
+	if (ecmd.base.speed == 0 || ecmd.base.speed == ((__u32)-1))
 		return;
 
-	switch (ecmd.duplex) {
+	switch (ecmd.base.duplex) {
 	case DUPLEX_FULL:
 	case DUPLEX_HALF:
 		break;
@@ -399,8 +397,8 @@ static void bond_update_speed_duplex(struct slave *slave)
 		return;
 	}
 
-	slave->speed = slave_speed;
-	slave->duplex = ecmd.duplex;
+	slave->speed = ecmd.base.speed;
+	slave->duplex = ecmd.base.duplex;
 
 	return;
 }
-- 
cgit v0.10.2


From 314d10d73b2580c0a036536f6444d704d96d6082 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:03 -0800
Subject: net: ipvlan: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index a7ca1c5..5802b90 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -346,12 +346,12 @@ static const struct header_ops ipvlan_header_ops = {
 	.cache_update	= eth_header_cache_update,
 };
 
-static int ipvlan_ethtool_get_settings(struct net_device *dev,
-				       struct ethtool_cmd *cmd)
+static int ipvlan_ethtool_get_link_ksettings(struct net_device *dev,
+					     struct ethtool_link_ksettings *cmd)
 {
 	const struct ipvl_dev *ipvlan = netdev_priv(dev);
 
-	return __ethtool_get_settings(ipvlan->phy_dev, cmd);
+	return __ethtool_get_link_ksettings(ipvlan->phy_dev, cmd);
 }
 
 static void ipvlan_ethtool_get_drvinfo(struct net_device *dev,
@@ -377,7 +377,7 @@ static void ipvlan_ethtool_set_msglevel(struct net_device *dev, u32 value)
 
 static const struct ethtool_ops ipvlan_ethtool_ops = {
 	.get_link	= ethtool_op_get_link,
-	.get_settings	= ipvlan_ethtool_get_settings,
+	.get_link_ksettings	= ipvlan_ethtool_get_link_ksettings,
 	.get_drvinfo	= ipvlan_ethtool_get_drvinfo,
 	.get_msglevel	= ipvlan_ethtool_get_msglevel,
 	.set_msglevel	= ipvlan_ethtool_set_msglevel,
-- 
cgit v0.10.2


From 85f9581975dc0315b1b47c5323a8ee497ebd6a8f Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:04 -0800
Subject: net: macvlan: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 426a2cc..6e953e3 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -940,12 +940,12 @@ static void macvlan_ethtool_get_drvinfo(struct net_device *dev,
 	strlcpy(drvinfo->version, "0.1", sizeof(drvinfo->version));
 }
 
-static int macvlan_ethtool_get_settings(struct net_device *dev,
-					struct ethtool_cmd *cmd)
+static int macvlan_ethtool_get_link_ksettings(struct net_device *dev,
+					      struct ethtool_link_ksettings *cmd)
 {
 	const struct macvlan_dev *vlan = netdev_priv(dev);
 
-	return __ethtool_get_settings(vlan->lowerdev, cmd);
+	return __ethtool_get_link_ksettings(vlan->lowerdev, cmd);
 }
 
 static netdev_features_t macvlan_fix_features(struct net_device *dev,
@@ -1020,7 +1020,7 @@ static int macvlan_dev_get_iflink(const struct net_device *dev)
 
 static const struct ethtool_ops macvlan_ethtool_ops = {
 	.get_link		= ethtool_op_get_link,
-	.get_settings		= macvlan_ethtool_get_settings,
+	.get_link_ksettings	= macvlan_ethtool_get_link_ksettings,
 	.get_drvinfo		= macvlan_ethtool_get_drvinfo,
 };
 
-- 
cgit v0.10.2


From 0ab6b544c185de913fa662c7565d78bdeccb54c4 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:05 -0800
Subject: net: team: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 00558e1..2769835 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -2813,12 +2813,12 @@ static void __team_port_change_send(struct team_port *port, bool linkup)
 	port->state.linkup = linkup;
 	team_refresh_port_linkup(port);
 	if (linkup) {
-		struct ethtool_cmd ecmd;
+		struct ethtool_link_ksettings ecmd;
 
-		err = __ethtool_get_settings(port->dev, &ecmd);
+		err = __ethtool_get_link_ksettings(port->dev, &ecmd);
 		if (!err) {
-			port->state.speed = ethtool_cmd_speed(&ecmd);
-			port->state.duplex = ecmd.duplex;
+			port->state.speed = ecmd.base.speed;
+			port->state.duplex = ecmd.base.duplex;
 			goto send_event;
 		}
 	}
-- 
cgit v0.10.2


From 008eb736932dcec25e99232dd280ed87e59635c8 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:06 -0800
Subject: net: fcoe: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/scsi/fcoe/fcoe_transport.c b/drivers/scsi/fcoe/fcoe_transport.c
index d7597c0..641c60e 100644
--- a/drivers/scsi/fcoe/fcoe_transport.c
+++ b/drivers/scsi/fcoe/fcoe_transport.c
@@ -93,36 +93,40 @@ static struct notifier_block libfcoe_notifier = {
 int fcoe_link_speed_update(struct fc_lport *lport)
 {
 	struct net_device *netdev = fcoe_get_netdev(lport);
-	struct ethtool_cmd ecmd;
+	struct ethtool_link_ksettings ecmd;
 
-	if (!__ethtool_get_settings(netdev, &ecmd)) {
+	if (!__ethtool_get_link_ksettings(netdev, &ecmd)) {
 		lport->link_supported_speeds &= ~(FC_PORTSPEED_1GBIT  |
 		                                  FC_PORTSPEED_10GBIT |
 		                                  FC_PORTSPEED_20GBIT |
 		                                  FC_PORTSPEED_40GBIT);
 
-		if (ecmd.supported & (SUPPORTED_1000baseT_Half |
-		                      SUPPORTED_1000baseT_Full |
-		                      SUPPORTED_1000baseKX_Full))
+		if (ecmd.link_modes.supported[0] & (
+			    SUPPORTED_1000baseT_Half |
+			    SUPPORTED_1000baseT_Full |
+			    SUPPORTED_1000baseKX_Full))
 			lport->link_supported_speeds |= FC_PORTSPEED_1GBIT;
 
-		if (ecmd.supported & (SUPPORTED_10000baseT_Full   |
-		                      SUPPORTED_10000baseKX4_Full |
-		                      SUPPORTED_10000baseKR_Full  |
-		                      SUPPORTED_10000baseR_FEC))
+		if (ecmd.link_modes.supported[0] & (
+			    SUPPORTED_10000baseT_Full   |
+			    SUPPORTED_10000baseKX4_Full |
+			    SUPPORTED_10000baseKR_Full  |
+			    SUPPORTED_10000baseR_FEC))
 			lport->link_supported_speeds |= FC_PORTSPEED_10GBIT;
 
-		if (ecmd.supported & (SUPPORTED_20000baseMLD2_Full |
-		                      SUPPORTED_20000baseKR2_Full))
+		if (ecmd.link_modes.supported[0] & (
+			    SUPPORTED_20000baseMLD2_Full |
+			    SUPPORTED_20000baseKR2_Full))
 			lport->link_supported_speeds |= FC_PORTSPEED_20GBIT;
 
-		if (ecmd.supported & (SUPPORTED_40000baseKR4_Full |
-		                      SUPPORTED_40000baseCR4_Full |
-		                      SUPPORTED_40000baseSR4_Full |
-		                      SUPPORTED_40000baseLR4_Full))
+		if (ecmd.link_modes.supported[0] & (
+			    SUPPORTED_40000baseKR4_Full |
+			    SUPPORTED_40000baseCR4_Full |
+			    SUPPORTED_40000baseSR4_Full |
+			    SUPPORTED_40000baseLR4_Full))
 			lport->link_supported_speeds |= FC_PORTSPEED_40GBIT;
 
-		switch (ethtool_cmd_speed(&ecmd)) {
+		switch (ecmd.base.speed) {
 		case SPEED_1000:
 			lport->link_speed = FC_PORTSPEED_1GBIT;
 			break;
-- 
cgit v0.10.2


From 17605b961f766757fddec20810453fc51b266e77 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:07 -0800
Subject: net: rdma: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/rdma/ib_addr.h b/include/rdma/ib_addr.h
index c34c900..931a47b 100644
--- a/include/rdma/ib_addr.h
+++ b/include/rdma/ib_addr.h
@@ -262,24 +262,22 @@ static inline enum ib_mtu iboe_get_mtu(int mtu)
 
 static inline int iboe_get_rate(struct net_device *dev)
 {
-	struct ethtool_cmd cmd;
-	u32 speed;
+	struct ethtool_link_ksettings cmd;
 	int err;
 
 	rtnl_lock();
-	err = __ethtool_get_settings(dev, &cmd);
+	err = __ethtool_get_link_ksettings(dev, &cmd);
 	rtnl_unlock();
 	if (err)
 		return IB_RATE_PORT_CURRENT;
 
-	speed = ethtool_cmd_speed(&cmd);
-	if (speed >= 40000)
+	if (cmd.base.speed >= 40000)
 		return IB_RATE_40_GBPS;
-	else if (speed >= 30000)
+	else if (cmd.base.speed >= 30000)
 		return IB_RATE_30_GBPS;
-	else if (speed >= 20000)
+	else if (cmd.base.speed >= 20000)
 		return IB_RATE_20_GBPS;
-	else if (speed >= 10000)
+	else if (cmd.base.speed >= 10000)
 		return IB_RATE_10_GBPS;
 	else
 		return IB_RATE_PORT_CURRENT;
-- 
cgit v0.10.2


From 577097981d72a034e4e2354881478cf202a9ba72 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:08 -0800
Subject: net: 8021q: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 055f0e9..e416a40 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -621,12 +621,12 @@ static netdev_features_t vlan_dev_fix_features(struct net_device *dev,
 	return features;
 }
 
-static int vlan_ethtool_get_settings(struct net_device *dev,
-				     struct ethtool_cmd *cmd)
+static int vlan_ethtool_get_link_ksettings(struct net_device *dev,
+					   struct ethtool_link_ksettings *cmd)
 {
 	const struct vlan_dev_priv *vlan = vlan_dev_priv(dev);
 
-	return __ethtool_get_settings(vlan->real_dev, cmd);
+	return __ethtool_get_link_ksettings(vlan->real_dev, cmd);
 }
 
 static void vlan_ethtool_get_drvinfo(struct net_device *dev,
@@ -741,7 +741,7 @@ static int vlan_dev_get_iflink(const struct net_device *dev)
 }
 
 static const struct ethtool_ops vlan_ethtool_ops = {
-	.get_settings	        = vlan_ethtool_get_settings,
+	.get_link_ksettings	= vlan_ethtool_get_link_ksettings,
 	.get_drvinfo	        = vlan_ethtool_get_drvinfo,
 	.get_link		= ethtool_op_get_link,
 	.get_ts_info		= vlan_ethtool_get_ts_info,
-- 
cgit v0.10.2


From 702b26a24d3d012b9286efb3efbf2b81b23271f5 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:09 -0800
Subject: net: bridge: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index c367b3e..b37a1cc 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -36,10 +36,10 @@
  */
 static int port_cost(struct net_device *dev)
 {
-	struct ethtool_cmd ecmd;
+	struct ethtool_link_ksettings ecmd;
 
-	if (!__ethtool_get_settings(dev, &ecmd)) {
-		switch (ethtool_cmd_speed(&ecmd)) {
+	if (!__ethtool_get_link_ksettings(dev, &ecmd)) {
+		switch (ecmd.base.speed) {
 		case SPEED_10000:
 			return 2;
 		case SPEED_1000:
-- 
cgit v0.10.2


From 7cad1bac96d381d953b47695f6602b4e2480aa5d Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:10 -0800
Subject: net: core: use __ethtool_get_ksettings

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 4ae17c3..2b3f76f 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -198,9 +198,10 @@ static ssize_t speed_show(struct device *dev,
 		return restart_syscall();
 
 	if (netif_running(netdev)) {
-		struct ethtool_cmd cmd;
-		if (!__ethtool_get_settings(netdev, &cmd))
-			ret = sprintf(buf, fmt_dec, ethtool_cmd_speed(&cmd));
+		struct ethtool_link_ksettings cmd;
+
+		if (!__ethtool_get_link_ksettings(netdev, &cmd))
+			ret = sprintf(buf, fmt_dec, cmd.base.speed);
 	}
 	rtnl_unlock();
 	return ret;
@@ -217,10 +218,12 @@ static ssize_t duplex_show(struct device *dev,
 		return restart_syscall();
 
 	if (netif_running(netdev)) {
-		struct ethtool_cmd cmd;
-		if (!__ethtool_get_settings(netdev, &cmd)) {
+		struct ethtool_link_ksettings cmd;
+
+		if (!__ethtool_get_link_ksettings(netdev, &cmd)) {
 			const char *duplex;
-			switch (cmd.duplex) {
+
+			switch (cmd.base.duplex) {
 			case DUPLEX_HALF:
 				duplex = "half";
 				break;
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index b7e7851..d41b107 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -557,9 +557,8 @@ static int prb_calc_retire_blk_tmo(struct packet_sock *po,
 {
 	struct net_device *dev;
 	unsigned int mbits = 0, msec = 0, div = 0, tmo = 0;
-	struct ethtool_cmd ecmd;
+	struct ethtool_link_ksettings ecmd;
 	int err;
-	u32 speed;
 
 	rtnl_lock();
 	dev = __dev_get_by_index(sock_net(&po->sk), po->ifindex);
@@ -567,19 +566,19 @@ static int prb_calc_retire_blk_tmo(struct packet_sock *po,
 		rtnl_unlock();
 		return DEFAULT_PRB_RETIRE_TOV;
 	}
-	err = __ethtool_get_settings(dev, &ecmd);
-	speed = ethtool_cmd_speed(&ecmd);
+	err = __ethtool_get_link_ksettings(dev, &ecmd);
 	rtnl_unlock();
 	if (!err) {
 		/*
 		 * If the link speed is so slow you don't really
 		 * need to worry about perf anyways
 		 */
-		if (speed < SPEED_1000 || speed == SPEED_UNKNOWN) {
+		if (ecmd.base.speed < SPEED_1000 ||
+		    ecmd.base.speed == SPEED_UNKNOWN) {
 			return DEFAULT_PRB_RETIRE_TOV;
 		} else {
 			msec = 1;
-			div = speed / 1000;
+			div = ecmd.base.speed / 1000;
 		}
 	}
 
-- 
cgit v0.10.2


From 3237fc63a3297d472a8cec33cb914f20570cfc23 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:11 -0800
Subject: net: ethtool: remove unused __ethtool_get_settings

replaced by __ethtool_get_link_ksettings.

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 8a400a5..e2b7bf2 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -150,10 +150,6 @@ extern int
 __ethtool_get_link_ksettings(struct net_device *dev,
 			     struct ethtool_link_ksettings *link_ksettings);
 
-/* DEPRECATED, use __ethtool_get_link_ksettings */
-extern int __ethtool_get_settings(struct net_device *dev,
-				  struct ethtool_cmd *cmd);
-
 /**
  * struct ethtool_ops - optional netdev operations
  * @get_settings: DEPRECATED, use %get_link_ksettings/%set_link_ksettings
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index edcec56..2966cd0 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -551,7 +551,12 @@ int __ethtool_get_link_ksettings(struct net_device *dev,
 	 * legacy %ethtool_cmd API, unless it's not supported either.
 	 * TODO: remove when ethtool_ops::get_settings disappears internally
 	 */
-	err = __ethtool_get_settings(dev, &cmd);
+	if (!dev->ethtool_ops->get_settings)
+		return -EOPNOTSUPP;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.cmd = ETHTOOL_GSET;
+	err = dev->ethtool_ops->get_settings(dev, &cmd);
 	if (err < 0)
 		return err;
 
@@ -729,30 +734,6 @@ static int ethtool_set_link_ksettings(struct net_device *dev,
 	return dev->ethtool_ops->set_link_ksettings(dev, &link_ksettings);
 }
 
-/* Internal kernel helper to query a device ethtool_cmd settings.
- *
- * Note about transition to ethtool_link_settings API: We do not need
- * (or want) this function to support "dev" instances that implement
- * the ethtool_link_settings API as we will update the drivers calling
- * this function to call __ethtool_get_link_ksettings instead, before
- * the first drivers implement ethtool_ops::get_link_ksettings.
- *
- * TODO 1: at least make this function static when no driver is using it
- * TODO 2: remove when ethtool_ops::get_settings disappears internally
- */
-int __ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
-{
-	ASSERT_RTNL();
-
-	if (!dev->ethtool_ops->get_settings)
-		return -EOPNOTSUPP;
-
-	memset(cmd, 0, sizeof(struct ethtool_cmd));
-	cmd->cmd = ETHTOOL_GSET;
-	return dev->ethtool_ops->get_settings(dev, cmd);
-}
-EXPORT_SYMBOL(__ethtool_get_settings);
-
 static void
 warn_incomplete_ethtool_legacy_settings_conversion(const char *details)
 {
@@ -796,16 +777,18 @@ static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
 		/* send a sensible cmd tag back to user */
 		cmd.cmd = ETHTOOL_GSET;
 	} else {
-		int err;
-		/* TODO: return -EOPNOTSUPP when
-		 * ethtool_ops::get_settings disappears internally
-		 */
-
 		/* driver doesn't support %ethtool_link_ksettings
 		 * API. revert to legacy %ethtool_cmd API, unless it's
 		 * not supported either.
 		 */
-		err = __ethtool_get_settings(dev, &cmd);
+		int err;
+
+		if (!dev->ethtool_ops->get_settings)
+			return -EOPNOTSUPP;
+
+		memset(&cmd, 0, sizeof(cmd));
+		cmd.cmd = ETHTOOL_GSET;
+		err = dev->ethtool_ops->get_settings(dev, &cmd);
 		if (err < 0)
 			return err;
 	}
-- 
cgit v0.10.2


From 3d8f7cc78d0eb07641fdcfb3961e8794778a6678 Mon Sep 17 00:00:00 2001
From: David Decotigny <decot@googlers.com>
Date: Wed, 24 Feb 2016 10:58:12 -0800
Subject: net: mlx4: use new ETHTOOL_G/SSETTINGS API

Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index dd84cab..f69584a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -501,34 +501,30 @@ static u32 mlx4_en_autoneg_get(struct net_device *dev)
 	return autoneg;
 }
 
-static u32 ptys_get_supported_port(struct mlx4_ptys_reg *ptys_reg)
+static void ptys2ethtool_update_supported_port(unsigned long *mask,
+					       struct mlx4_ptys_reg *ptys_reg)
 {
 	u32 eth_proto = be32_to_cpu(ptys_reg->eth_proto_cap);
 
 	if (eth_proto & (MLX4_PROT_MASK(MLX4_10GBASE_T)
 			 | MLX4_PROT_MASK(MLX4_1000BASE_T)
 			 | MLX4_PROT_MASK(MLX4_100BASE_TX))) {
-			return SUPPORTED_TP;
-	}
-
-	if (eth_proto & (MLX4_PROT_MASK(MLX4_10GBASE_CR)
+		__set_bit(ETHTOOL_LINK_MODE_TP_BIT, mask);
+	} else if (eth_proto & (MLX4_PROT_MASK(MLX4_10GBASE_CR)
 			 | MLX4_PROT_MASK(MLX4_10GBASE_SR)
 			 | MLX4_PROT_MASK(MLX4_56GBASE_SR4)
 			 | MLX4_PROT_MASK(MLX4_40GBASE_CR4)
 			 | MLX4_PROT_MASK(MLX4_40GBASE_SR4)
 			 | MLX4_PROT_MASK(MLX4_1000BASE_CX_SGMII))) {
-			return SUPPORTED_FIBRE;
-	}
-
-	if (eth_proto & (MLX4_PROT_MASK(MLX4_56GBASE_KR4)
+		__set_bit(ETHTOOL_LINK_MODE_FIBRE_BIT, mask);
+	} else if (eth_proto & (MLX4_PROT_MASK(MLX4_56GBASE_KR4)
 			 | MLX4_PROT_MASK(MLX4_40GBASE_KR4)
 			 | MLX4_PROT_MASK(MLX4_20GBASE_KR2)
 			 | MLX4_PROT_MASK(MLX4_10GBASE_KR)
 			 | MLX4_PROT_MASK(MLX4_10GBASE_KX4)
 			 | MLX4_PROT_MASK(MLX4_1000BASE_KX))) {
-			return SUPPORTED_Backplane;
+		__set_bit(ETHTOOL_LINK_MODE_Backplane_BIT, mask);
 	}
-	return 0;
 }
 
 static u32 ptys_get_active_port(struct mlx4_ptys_reg *ptys_reg)
@@ -574,122 +570,111 @@ static u32 ptys_get_active_port(struct mlx4_ptys_reg *ptys_reg)
 enum ethtool_report {
 	SUPPORTED = 0,
 	ADVERTISED = 1,
-	SPEED = 2
 };
 
+struct ptys2ethtool_config {
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(supported);
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(advertised);
+	u32 speed;
+};
+
+static unsigned long *ptys2ethtool_link_mode(struct ptys2ethtool_config *cfg,
+					     enum ethtool_report report)
+{
+	switch (report) {
+	case SUPPORTED:
+		return cfg->supported;
+	case ADVERTISED:
+		return cfg->advertised;
+	}
+	return NULL;
+}
+
+#define MLX4_BUILD_PTYS2ETHTOOL_CONFIG(reg_, speed_, ...)		\
+	({								\
+		struct ptys2ethtool_config *cfg;			\
+		const unsigned int modes[] = { __VA_ARGS__ };		\
+		unsigned int i;						\
+		cfg = &ptys2ethtool_map[reg_];				\
+		cfg->speed = speed_;					\
+		bitmap_zero(cfg->supported,				\
+			    __ETHTOOL_LINK_MODE_MASK_NBITS);		\
+		bitmap_zero(cfg->advertised,				\
+			    __ETHTOOL_LINK_MODE_MASK_NBITS);		\
+		for (i = 0 ; i < ARRAY_SIZE(modes) ; ++i) {		\
+			__set_bit(modes[i], cfg->supported);		\
+			__set_bit(modes[i], cfg->advertised);		\
+		}							\
+	})
+
 /* Translates mlx4 link mode to equivalent ethtool Link modes/speed */
-static u32 ptys2ethtool_map[MLX4_LINK_MODES_SZ][3] = {
-	[MLX4_100BASE_TX] = {
-		SUPPORTED_100baseT_Full,
-		ADVERTISED_100baseT_Full,
-		SPEED_100
-		},
-
-	[MLX4_1000BASE_T] = {
-		SUPPORTED_1000baseT_Full,
-		ADVERTISED_1000baseT_Full,
-		SPEED_1000
-		},
-	[MLX4_1000BASE_CX_SGMII] = {
-		SUPPORTED_1000baseKX_Full,
-		ADVERTISED_1000baseKX_Full,
-		SPEED_1000
-		},
-	[MLX4_1000BASE_KX] = {
-		SUPPORTED_1000baseKX_Full,
-		ADVERTISED_1000baseKX_Full,
-		SPEED_1000
-		},
-
-	[MLX4_10GBASE_T] = {
-		SUPPORTED_10000baseT_Full,
-		ADVERTISED_10000baseT_Full,
-		SPEED_10000
-		},
-	[MLX4_10GBASE_CX4] = {
-		SUPPORTED_10000baseKX4_Full,
-		ADVERTISED_10000baseKX4_Full,
-		SPEED_10000
-		},
-	[MLX4_10GBASE_KX4] = {
-		SUPPORTED_10000baseKX4_Full,
-		ADVERTISED_10000baseKX4_Full,
-		SPEED_10000
-		},
-	[MLX4_10GBASE_KR] = {
-		SUPPORTED_10000baseKR_Full,
-		ADVERTISED_10000baseKR_Full,
-		SPEED_10000
-		},
-	[MLX4_10GBASE_CR] = {
-		SUPPORTED_10000baseKR_Full,
-		ADVERTISED_10000baseKR_Full,
-		SPEED_10000
-		},
-	[MLX4_10GBASE_SR] = {
-		SUPPORTED_10000baseKR_Full,
-		ADVERTISED_10000baseKR_Full,
-		SPEED_10000
-		},
-
-	[MLX4_20GBASE_KR2] = {
-		SUPPORTED_20000baseMLD2_Full | SUPPORTED_20000baseKR2_Full,
-		ADVERTISED_20000baseMLD2_Full | ADVERTISED_20000baseKR2_Full,
-		SPEED_20000
-		},
-
-	[MLX4_40GBASE_CR4] = {
-		SUPPORTED_40000baseCR4_Full,
-		ADVERTISED_40000baseCR4_Full,
-		SPEED_40000
-		},
-	[MLX4_40GBASE_KR4] = {
-		SUPPORTED_40000baseKR4_Full,
-		ADVERTISED_40000baseKR4_Full,
-		SPEED_40000
-		},
-	[MLX4_40GBASE_SR4] = {
-		SUPPORTED_40000baseSR4_Full,
-		ADVERTISED_40000baseSR4_Full,
-		SPEED_40000
-		},
-
-	[MLX4_56GBASE_KR4] = {
-		SUPPORTED_56000baseKR4_Full,
-		ADVERTISED_56000baseKR4_Full,
-		SPEED_56000
-		},
-	[MLX4_56GBASE_CR4] = {
-		SUPPORTED_56000baseCR4_Full,
-		ADVERTISED_56000baseCR4_Full,
-		SPEED_56000
-		},
-	[MLX4_56GBASE_SR4] = {
-		SUPPORTED_56000baseSR4_Full,
-		ADVERTISED_56000baseSR4_Full,
-		SPEED_56000
-		},
+static struct ptys2ethtool_config ptys2ethtool_map[MLX4_LINK_MODES_SZ];
+
+void __init mlx4_en_init_ptys2ethtool_map(void)
+{
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_100BASE_TX, SPEED_100,
+				       ETHTOOL_LINK_MODE_100baseT_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_1000BASE_T, SPEED_1000,
+				       ETHTOOL_LINK_MODE_1000baseT_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_1000BASE_CX_SGMII, SPEED_1000,
+				       ETHTOOL_LINK_MODE_1000baseKX_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_1000BASE_KX, SPEED_1000,
+				       ETHTOOL_LINK_MODE_1000baseKX_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_10GBASE_T, SPEED_10000,
+				       ETHTOOL_LINK_MODE_10000baseT_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_10GBASE_CX4, SPEED_10000,
+				       ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_10GBASE_KX4, SPEED_10000,
+				       ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_10GBASE_KR, SPEED_10000,
+				       ETHTOOL_LINK_MODE_10000baseKR_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_10GBASE_CR, SPEED_10000,
+				       ETHTOOL_LINK_MODE_10000baseKR_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_10GBASE_SR, SPEED_10000,
+				       ETHTOOL_LINK_MODE_10000baseKR_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_20GBASE_KR2, SPEED_20000,
+				       ETHTOOL_LINK_MODE_20000baseMLD2_Full_BIT,
+				       ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_40GBASE_CR4, SPEED_40000,
+				       ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_40GBASE_KR4, SPEED_40000,
+				       ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_40GBASE_SR4, SPEED_40000,
+				       ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_56GBASE_KR4, SPEED_56000,
+				       ETHTOOL_LINK_MODE_56000baseKR4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_56GBASE_CR4, SPEED_56000,
+				       ETHTOOL_LINK_MODE_56000baseCR4_Full_BIT);
+	MLX4_BUILD_PTYS2ETHTOOL_CONFIG(MLX4_56GBASE_SR4, SPEED_56000,
+				       ETHTOOL_LINK_MODE_56000baseSR4_Full_BIT);
 };
 
-static u32 ptys2ethtool_link_modes(u32 eth_proto, enum ethtool_report report)
+static void ptys2ethtool_update_link_modes(unsigned long *link_modes,
+					   u32 eth_proto,
+					   enum ethtool_report report)
 {
 	int i;
-	u32 link_modes = 0;
-
 	for (i = 0; i < MLX4_LINK_MODES_SZ; i++) {
 		if (eth_proto & MLX4_PROT_MASK(i))
-			link_modes |= ptys2ethtool_map[i][report];
+			bitmap_or(link_modes, link_modes,
+				  ptys2ethtool_link_mode(&ptys2ethtool_map[i],
+							 report),
+				  __ETHTOOL_LINK_MODE_MASK_NBITS);
 	}
-	return link_modes;
 }
 
-static u32 ethtool2ptys_link_modes(u32 link_modes, enum ethtool_report report)
+static u32 ethtool2ptys_link_modes(const unsigned long *link_modes,
+				   enum ethtool_report report)
 {
 	int i;
 	u32 ptys_modes = 0;
 
 	for (i = 0; i < MLX4_LINK_MODES_SZ; i++) {
-		if (ptys2ethtool_map[i][report] & link_modes)
+		if (bitmap_intersects(
+			    ptys2ethtool_link_mode(&ptys2ethtool_map[i],
+						   report),
+			    link_modes,
+			    __ETHTOOL_LINK_MODE_MASK_NBITS))
 			ptys_modes |= 1 << i;
 	}
 	return ptys_modes;
@@ -702,14 +687,15 @@ static u32 speed2ptys_link_modes(u32 speed)
 	u32 ptys_modes = 0;
 
 	for (i = 0; i < MLX4_LINK_MODES_SZ; i++) {
-		if (ptys2ethtool_map[i][SPEED] == speed)
+		if (ptys2ethtool_map[i].speed == speed)
 			ptys_modes |= 1 << i;
 	}
 	return ptys_modes;
 }
 
-static int ethtool_get_ptys_settings(struct net_device *dev,
-				     struct ethtool_cmd *cmd)
+static int
+ethtool_get_ptys_link_ksettings(struct net_device *dev,
+				struct ethtool_link_ksettings *link_ksettings)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_ptys_reg ptys_reg;
@@ -737,79 +723,102 @@ static int ethtool_get_ptys_settings(struct net_device *dev,
 	en_dbg(DRV, priv, "ptys_reg.eth_proto_lp_adv %x\n",
 	       be32_to_cpu(ptys_reg.eth_proto_lp_adv));
 
-	cmd->supported = 0;
-	cmd->advertising = 0;
+	/* reset supported/advertising masks */
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
 
-	cmd->supported |= ptys_get_supported_port(&ptys_reg);
+	ptys2ethtool_update_supported_port(link_ksettings->link_modes.supported,
+					   &ptys_reg);
 
 	eth_proto = be32_to_cpu(ptys_reg.eth_proto_cap);
-	cmd->supported |= ptys2ethtool_link_modes(eth_proto, SUPPORTED);
+	ptys2ethtool_update_link_modes(link_ksettings->link_modes.supported,
+				       eth_proto, SUPPORTED);
 
 	eth_proto = be32_to_cpu(ptys_reg.eth_proto_admin);
-	cmd->advertising |= ptys2ethtool_link_modes(eth_proto, ADVERTISED);
+	ptys2ethtool_update_link_modes(link_ksettings->link_modes.advertising,
+				       eth_proto, ADVERTISED);
 
-	cmd->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
-	cmd->advertising |= (priv->prof->tx_pause) ? ADVERTISED_Pause : 0;
+	ethtool_link_ksettings_add_link_mode(link_ksettings, supported,
+					     Pause);
+	ethtool_link_ksettings_add_link_mode(link_ksettings, supported,
+					     Asym_Pause);
 
-	cmd->advertising |= (priv->prof->tx_pause ^ priv->prof->rx_pause) ?
-		ADVERTISED_Asym_Pause : 0;
+	if (priv->prof->tx_pause)
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     advertising, Pause);
+	if (priv->prof->tx_pause ^ priv->prof->rx_pause)
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     advertising, Asym_Pause);
 
-	cmd->port = ptys_get_active_port(&ptys_reg);
-	cmd->transceiver = (SUPPORTED_TP & cmd->supported) ?
-		XCVR_EXTERNAL : XCVR_INTERNAL;
+	link_ksettings->base.port = ptys_get_active_port(&ptys_reg);
 
 	if (mlx4_en_autoneg_get(dev)) {
-		cmd->supported |= SUPPORTED_Autoneg;
-		cmd->advertising |= ADVERTISED_Autoneg;
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     supported, Autoneg);
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     advertising, Autoneg);
 	}
 
-	cmd->autoneg = (priv->port_state.flags & MLX4_EN_PORT_ANC) ?
+	link_ksettings->base.autoneg
+		= (priv->port_state.flags & MLX4_EN_PORT_ANC) ?
 		AUTONEG_ENABLE : AUTONEG_DISABLE;
 
 	eth_proto = be32_to_cpu(ptys_reg.eth_proto_lp_adv);
-	cmd->lp_advertising = ptys2ethtool_link_modes(eth_proto, ADVERTISED);
 
-	cmd->lp_advertising |= (priv->port_state.flags & MLX4_EN_PORT_ANC) ?
-			ADVERTISED_Autoneg : 0;
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, lp_advertising);
+	ptys2ethtool_update_link_modes(
+		link_ksettings->link_modes.lp_advertising,
+		eth_proto, ADVERTISED);
+	if (priv->port_state.flags & MLX4_EN_PORT_ANC)
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     lp_advertising, Autoneg);
 
-	cmd->phy_address = 0;
-	cmd->mdio_support = 0;
-	cmd->maxtxpkt = 0;
-	cmd->maxrxpkt = 0;
-	cmd->eth_tp_mdix = ETH_TP_MDI_INVALID;
-	cmd->eth_tp_mdix_ctrl = ETH_TP_MDI_AUTO;
+	link_ksettings->base.phy_address = 0;
+	link_ksettings->base.mdio_support = 0;
+	link_ksettings->base.eth_tp_mdix = ETH_TP_MDI_INVALID;
+	link_ksettings->base.eth_tp_mdix_ctrl = ETH_TP_MDI_AUTO;
 
 	return ret;
 }
 
-static void ethtool_get_default_settings(struct net_device *dev,
-					 struct ethtool_cmd *cmd)
+static void
+ethtool_get_default_link_ksettings(
+	struct net_device *dev, struct ethtool_link_ksettings *link_ksettings)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	int trans_type;
 
-	cmd->autoneg = AUTONEG_DISABLE;
-	cmd->supported = SUPPORTED_10000baseT_Full;
-	cmd->advertising = ADVERTISED_10000baseT_Full;
-	trans_type = priv->port_state.transceiver;
+	link_ksettings->base.autoneg = AUTONEG_DISABLE;
+
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, supported);
+	ethtool_link_ksettings_add_link_mode(link_ksettings, supported,
+					     10000baseT_Full);
 
+	ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising);
+	ethtool_link_ksettings_add_link_mode(link_ksettings, advertising,
+					     10000baseT_Full);
+
+	trans_type = priv->port_state.transceiver;
 	if (trans_type > 0 && trans_type <= 0xC) {
-		cmd->port = PORT_FIBRE;
-		cmd->transceiver = XCVR_EXTERNAL;
-		cmd->supported |= SUPPORTED_FIBRE;
-		cmd->advertising |= ADVERTISED_FIBRE;
+		link_ksettings->base.port = PORT_FIBRE;
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     supported, FIBRE);
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     advertising, FIBRE);
 	} else if (trans_type == 0x80 || trans_type == 0) {
-		cmd->port = PORT_TP;
-		cmd->transceiver = XCVR_INTERNAL;
-		cmd->supported |= SUPPORTED_TP;
-		cmd->advertising |= ADVERTISED_TP;
+		link_ksettings->base.port = PORT_TP;
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     supported, TP);
+		ethtool_link_ksettings_add_link_mode(link_ksettings,
+						     advertising, TP);
 	} else  {
-		cmd->port = -1;
-		cmd->transceiver = -1;
+		link_ksettings->base.port = -1;
 	}
 }
 
-static int mlx4_en_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+static int
+mlx4_en_get_link_ksettings(struct net_device *dev,
+			   struct ethtool_link_ksettings *link_ksettings)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	int ret = -EINVAL;
@@ -822,16 +831,16 @@ static int mlx4_en_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	       priv->port_state.flags & MLX4_EN_PORT_ANE);
 
 	if (priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETH_PROT_CTRL)
-		ret = ethtool_get_ptys_settings(dev, cmd);
+		ret = ethtool_get_ptys_link_ksettings(dev, link_ksettings);
 	if (ret) /* ETH PROT CRTL is not supported or PTYS CMD failed */
-		ethtool_get_default_settings(dev, cmd);
+		ethtool_get_default_link_ksettings(dev, link_ksettings);
 
 	if (netif_carrier_ok(dev)) {
-		ethtool_cmd_speed_set(cmd, priv->port_state.link_speed);
-		cmd->duplex = DUPLEX_FULL;
+		link_ksettings->base.speed = priv->port_state.link_speed;
+		link_ksettings->base.duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN);
-		cmd->duplex = DUPLEX_UNKNOWN;
+		link_ksettings->base.speed = SPEED_UNKNOWN;
+		link_ksettings->base.duplex = DUPLEX_UNKNOWN;
 	}
 	return 0;
 }
@@ -855,21 +864,29 @@ static __be32 speed_set_ptys_admin(struct mlx4_en_priv *priv, u32 speed,
 	return proto_admin;
 }
 
-static int mlx4_en_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+static int
+mlx4_en_set_link_ksettings(struct net_device *dev,
+			   const struct ethtool_link_ksettings *link_ksettings)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_ptys_reg ptys_reg;
 	__be32 proto_admin;
 	int ret;
 
-	u32 ptys_adv = ethtool2ptys_link_modes(cmd->advertising, ADVERTISED);
-	int speed = ethtool_cmd_speed(cmd);
+	u32 ptys_adv = ethtool2ptys_link_modes(
+		link_ksettings->link_modes.advertising, ADVERTISED);
+	const int speed = link_ksettings->base.speed;
 
-	en_dbg(DRV, priv, "Set Speed=%d adv=0x%x autoneg=%d duplex=%d\n",
-	       speed, cmd->advertising, cmd->autoneg, cmd->duplex);
+	en_dbg(DRV, priv,
+	       "Set Speed=%d adv={%*pbl} autoneg=%d duplex=%d\n",
+	       speed, __ETHTOOL_LINK_MODE_MASK_NBITS,
+	       link_ksettings->link_modes.advertising,
+	       link_ksettings->base.autoneg,
+	       link_ksettings->base.duplex);
 
-	if (!(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETH_PROT_CTRL) ||
-	    (cmd->duplex == DUPLEX_HALF))
+	if (!(priv->mdev->dev->caps.flags2 &
+	      MLX4_DEV_CAP_FLAG2_ETH_PROT_CTRL) ||
+	    (link_ksettings->base.duplex == DUPLEX_HALF))
 		return -EINVAL;
 
 	memset(&ptys_reg, 0, sizeof(ptys_reg));
@@ -883,7 +900,7 @@ static int mlx4_en_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		return 0;
 	}
 
-	proto_admin = cmd->autoneg == AUTONEG_ENABLE ?
+	proto_admin = link_ksettings->base.autoneg == AUTONEG_ENABLE ?
 		cpu_to_be32(ptys_adv) :
 		speed_set_ptys_admin(priv, speed,
 				     ptys_reg.eth_proto_cap);
@@ -1982,8 +1999,8 @@ static int mlx4_en_set_phys_id(struct net_device *dev,
 
 const struct ethtool_ops mlx4_en_ethtool_ops = {
 	.get_drvinfo = mlx4_en_get_drvinfo,
-	.get_settings = mlx4_en_get_settings,
-	.set_settings = mlx4_en_set_settings,
+	.get_link_ksettings = mlx4_en_get_link_ksettings,
+	.set_link_ksettings = mlx4_en_set_link_ksettings,
 	.get_link = ethtool_op_get_link,
 	.get_strings = mlx4_en_get_strings,
 	.get_sset_count = mlx4_en_get_sset_count,
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c
index e0ec280..bf7628d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c
@@ -382,6 +382,7 @@ static void mlx4_en_verify_params(void)
 static int __init mlx4_en_init(void)
 {
 	mlx4_en_verify_params();
+	mlx4_en_init_ptys2ethtool_map();
 
 	return mlx4_register_interface(&mlx4_en_interface);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 35de7d2..d12ab6a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -607,6 +607,7 @@ static inline struct mlx4_cqe *mlx4_en_get_cqe(void *buf, int idx, int cqe_sz)
 
 #define MLX4_EN_WOL_DO_MODIFY (1ULL << 63)
 
+void mlx4_en_init_ptys2ethtool_map(void);
 void mlx4_en_update_loopback_state(struct net_device *dev,
 				   netdev_features_t features);
 
-- 
cgit v0.10.2


From 59097ac9b20a0318b2b99b3338c0bb066309583e Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Wed, 24 Feb 2016 09:15:43 +0100
Subject: can: ems_usb: fix coding style

This patch fixes the coding style issues introduced in commit:

    90cfde46586d can: ems_usb: Fix possible tx overflow

Reported-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c
index eb7192f..3400fd1 100644
--- a/drivers/net/can/usb/ems_usb.c
+++ b/drivers/net/can/usb/ems_usb.c
@@ -281,11 +281,9 @@ static void ems_usb_read_interrupt_callback(struct urb *urb)
 	switch (urb->status) {
 	case 0:
 		dev->free_slots = dev->intr_in_buffer[1];
-		if(dev->free_slots > CPC_TX_QUEUE_TRIGGER_HIGH){
-			if (netif_queue_stopped(netdev)){
-				netif_wake_queue(netdev);
-			}
-		}
+		if (dev->free_slots > CPC_TX_QUEUE_TRIGGER_HIGH &&
+		    netif_queue_stopped(netdev))
+			netif_wake_queue(netdev);
 		break;
 
 	case -ECONNRESET: /* unlink */
-- 
cgit v0.10.2


From 0dfa61bba38513957240660a9cad82dd408789ca Mon Sep 17 00:00:00 2001
From: Simon Horman <horms+renesas@verge.net.au>
Date: Wed, 24 Feb 2016 10:56:32 +0900
Subject: can: rcar: add gen[12] fallback compatibility strings

Add fallback compatibility string for R-Car Gen 1 and Gen2.

In the case of Renesas R-Car hardware we know that there are generations of
SoCs, e.g. Gen 1 and Gen 2. But beyond that its not clear what the
relationship between IP blocks might be. For example, I believe that
r8a7779 is older than r8a7778 but that doesn't imply that the latter is a
descendant of the former or vice versa.

We can, however, by examining the documentation and behaviour of the
hardware at run-time observe that the current driver implementation appears
to be compatible with the IP blocks on SoCs within a given generation.

For the above reasons and convenience when enabling new SoCs a
per-generation fallback compatibility string scheme being adopted for
drivers for Renesas SoCs.

Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Rob Herring <robh@kernel.org>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/Documentation/devicetree/bindings/net/can/rcar_can.txt b/Documentation/devicetree/bindings/net/can/rcar_can.txt
index 002d844..f2172fb 100644
--- a/Documentation/devicetree/bindings/net/can/rcar_can.txt
+++ b/Documentation/devicetree/bindings/net/can/rcar_can.txt
@@ -6,6 +6,12 @@ Required properties:
 	      "renesas,can-r8a7779" if CAN controller is a part of R8A7779 SoC.
 	      "renesas,can-r8a7790" if CAN controller is a part of R8A7790 SoC.
 	      "renesas,can-r8a7791" if CAN controller is a part of R8A7791 SoC.
+	      "renesas,rcar-gen1-can" for a generic R-Car Gen1 compatible device.
+	      "renesas,rcar-gen2-can" for a generic R-Car Gen2 compatible device.
+	      When compatible with the generic version, nodes must list the
+	      SoC-specific version corresponding to the platform first
+	      followed by the generic version.
+
 - reg: physical base address and size of the R-Car CAN register map.
 - interrupts: interrupt specifier for the sole interrupt.
 - clocks: phandles and clock specifiers for 3 CAN clock inputs.
@@ -25,7 +31,7 @@ Example
 SoC common .dtsi file:
 
 	can0: can@e6e80000 {
-		compatible = "renesas,can-r8a7791";
+		compatible = "renesas,can-r8a7791", "renesas,rcar-gen2-can";
 		reg = <0 0xe6e80000 0 0x1000>;
 		interrupts = <0 186 IRQ_TYPE_LEVEL_HIGH>;
 		clocks = <&mstp9_clks R8A7791_CLK_RCAN0>,
diff --git a/drivers/net/can/rcar_can.c b/drivers/net/can/rcar_can.c
index bc46be3..ad3d2e0 100644
--- a/drivers/net/can/rcar_can.c
+++ b/drivers/net/can/rcar_can.c
@@ -904,6 +904,8 @@ static const struct of_device_id rcar_can_of_table[] __maybe_unused = {
 	{ .compatible = "renesas,can-r8a7779" },
 	{ .compatible = "renesas,can-r8a7790" },
 	{ .compatible = "renesas,can-r8a7791" },
+	{ .compatible = "renesas,rcar-gen1-can" },
+	{ .compatible = "renesas,rcar-gen2-can" },
 	{ }
 };
 MODULE_DEVICE_TABLE(of, rcar_can_of_table);
-- 
cgit v0.10.2


From f71096dfd129e4ad1ae80cd10d5ac050e5730f8a Mon Sep 17 00:00:00 2001
From: Simon Horman <horms+renesas@verge.net.au>
Date: Wed, 24 Feb 2016 10:56:33 +0900
Subject: can: rcar: add device tree support for r8a779[234]

Simply document new compatibility string.
As a previous patch adds a generic R-Car Gen2 compatibility string
there appears to be no need for a driver updates.

By documenting these compat stings they may be used in DTSs shipped, for
example as part of ROMs. They must be used in conjunction with the Gen2
fallback compat string. At this time there are no known differences between
the r8a779[234] IP blocks and that implemented by the driver for the Gen2
fallback compat string. Thus there is no need to update the driver as the
use of the Gen2 fallback compat string will activate the correct code in
the current driver while leaving the option for r8a779[234]-specific driver
code to be activated in an updated driver should the need arise.

Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/Documentation/devicetree/bindings/net/can/rcar_can.txt b/Documentation/devicetree/bindings/net/can/rcar_can.txt
index f2172fb..65edc05 100644
--- a/Documentation/devicetree/bindings/net/can/rcar_can.txt
+++ b/Documentation/devicetree/bindings/net/can/rcar_can.txt
@@ -6,6 +6,9 @@ Required properties:
 	      "renesas,can-r8a7779" if CAN controller is a part of R8A7779 SoC.
 	      "renesas,can-r8a7790" if CAN controller is a part of R8A7790 SoC.
 	      "renesas,can-r8a7791" if CAN controller is a part of R8A7791 SoC.
+	      "renesas,can-r8a7792" if CAN controller is a part of R8A7792 SoC.
+	      "renesas,can-r8a7793" if CAN controller is a part of R8A7793 SoC.
+	      "renesas,can-r8a7794" if CAN controller is a part of R8A7794 SoC.
 	      "renesas,rcar-gen1-can" for a generic R-Car Gen1 compatible device.
 	      "renesas,rcar-gen2-can" for a generic R-Car Gen2 compatible device.
 	      When compatible with the generic version, nodes must list the
-- 
cgit v0.10.2


From 2306f65637079c8922aec15b4ced75ea457e757b Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 26 Feb 2016 09:06:07 +0100
Subject: 6lowpan: iphc: fix invalid case handling

This patch fixes the return value in a case which should never occur.
Instead returning "-EINVAL" we return LOWPAN_IPHC_DAM_00 which is
invalid on context based addresses. Also change the WARN_ON_ONCE to
WARN_ONCE which was suggested by Dan Carpenter.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c
index 06287ac..7217251 100644
--- a/net/6lowpan/iphc.c
+++ b/net/6lowpan/iphc.c
@@ -811,8 +811,8 @@ static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct in6_addr *ipaddr,
 		goto out;
 	}
 
-	WARN_ON_ONCE("context found but no address mode matched\n");
-	return -EINVAL;
+	WARN_ONCE(1, "context found but no address mode matched\n");
+	return LOWPAN_IPHC_DAM_00;
 out:
 
 	if (sam)
-- 
cgit v0.10.2


From 3f2fb9a834cb1fcddbae22deca7fde136944dc89 Mon Sep 17 00:00:00 2001
From: David Ahern <dsa@cumulusnetworks.com>
Date: Wed, 24 Feb 2016 11:47:02 -0800
Subject: net: l3mdev: address selection should only consider devices in L3
 domain

David Lamparter noted a use case where the source address selection fails
to pick an address from a VRF interface - unnumbered interfaces.

Relevant commands from his script:
    ip addr add 9.9.9.9/32 dev lo
    ip link set lo up

    ip link add name vrf0 type vrf table 101
    ip rule add oif vrf0 table 101
    ip rule add iif vrf0 table 101
    ip link set vrf0 up
    ip addr add 10.0.0.3/32 dev vrf0

    ip link add name dummy2 type dummy
    ip link set dummy2 master vrf0 up

    --> note dummy2 has no address - unnumbered device

    ip route add 10.2.2.2/32 dev dummy2 table 101
    ip neigh add 10.2.2.2 dev dummy2 lladdr 02:00:00:00:00:02

    tcpdump -ni dummy2 &

And using ping instead of his socat example:
    $ ping -I vrf0 -c1 10.2.2.2
    ping: Warning: source address might be selected on device other than vrf0.
    PING 10.2.2.2 (10.2.2.2) from 9.9.9.9 vrf0: 56(84) bytes of data.

>From tcpdump:
    12:57:29.449128 IP 9.9.9.9 > 10.2.2.2: ICMP echo request, id 2491, seq 1, length 64

Note the source address is from lo and is not a VRF local address. With
this patch:

    $ ping -I vrf0 -c1 10.2.2.2
    PING 10.2.2.2 (10.2.2.2) from 10.0.0.3 vrf0: 56(84) bytes of data.

>From tcpdump:
    12:59:25.096426 IP 10.0.0.3 > 10.2.2.2: ICMP echo request, id 2113, seq 1, length 64

Now the source address comes from vrf0.

The ipv4 function for selecting source address takes a const argument.
Removing the const requires touching a lot of places, so instead
l3mdev_master_ifindex_rcu is changed to take a const argument and then
do the typecast to non-const as required by netdev_master_upper_dev_get_rcu.
This is similar to what l3mdev_fib_table_rcu does.

IPv6 for unnumbered interfaces appears to be selecting the addresses
properly.

Cc: David Lamparter <david@opensourcerouting.org>
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h
index 5567d46..c43a9c7 100644
--- a/include/net/l3mdev.h
+++ b/include/net/l3mdev.h
@@ -39,7 +39,7 @@ struct l3mdev_ops {
 
 #ifdef CONFIG_NET_L3_MASTER_DEV
 
-int l3mdev_master_ifindex_rcu(struct net_device *dev);
+int l3mdev_master_ifindex_rcu(const struct net_device *dev);
 static inline int l3mdev_master_ifindex(struct net_device *dev)
 {
 	int ifindex;
@@ -179,7 +179,7 @@ struct dst_entry *l3mdev_rt6_dst_by_oif(struct net *net,
 
 #else
 
-static inline int l3mdev_master_ifindex_rcu(struct net_device *dev)
+static inline int l3mdev_master_ifindex_rcu(const struct net_device *dev)
 {
 	return 0;
 }
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 29b8d3a..18d510f 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1194,6 +1194,7 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope)
 	__be32 addr = 0;
 	struct in_device *in_dev;
 	struct net *net = dev_net(dev);
+	int master_idx;
 
 	rcu_read_lock();
 	in_dev = __in_dev_get_rcu(dev);
@@ -1214,12 +1215,16 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope)
 	if (addr)
 		goto out_unlock;
 no_in_dev:
+	master_idx = l3mdev_master_ifindex_rcu(dev);
 
 	/* Not loopback addresses on loopback should be preferred
 	   in this case. It is important that lo is the first interface
 	   in dev_base list.
 	 */
 	for_each_netdev_rcu(net, dev) {
+		if (l3mdev_master_ifindex_rcu(dev) != master_idx)
+			continue;
+
 		in_dev = __in_dev_get_rcu(dev);
 		if (!in_dev)
 			continue;
diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c
index 8e5ead3..e925037 100644
--- a/net/l3mdev/l3mdev.c
+++ b/net/l3mdev/l3mdev.c
@@ -17,7 +17,7 @@
  *	@dev: targeted interface
  */
 
-int l3mdev_master_ifindex_rcu(struct net_device *dev)
+int l3mdev_master_ifindex_rcu(const struct net_device *dev)
 {
 	int ifindex = 0;
 
@@ -28,8 +28,15 @@ int l3mdev_master_ifindex_rcu(struct net_device *dev)
 		ifindex = dev->ifindex;
 	} else if (netif_is_l3_slave(dev)) {
 		struct net_device *master;
+		struct net_device *_dev = (struct net_device *)dev;
 
-		master = netdev_master_upper_dev_get_rcu(dev);
+		/* netdev_master_upper_dev_get_rcu calls
+		 * list_first_or_null_rcu to walk the upper dev list.
+		 * list_first_or_null_rcu does not handle a const arg. We aren't
+		 * making changes, just want the master device from that list so
+		 * typecast to remove the const
+		 */
+		master = netdev_master_upper_dev_get_rcu(_dev);
 		if (master)
 			ifindex = master->ifindex;
 	}
-- 
cgit v0.10.2


From 17b693cdd87635ae813ad61ad0b8a8458f4c3960 Mon Sep 17 00:00:00 2001
From: David Lamparter <equinox@diac24.net>
Date: Wed, 24 Feb 2016 11:47:03 -0800
Subject: net: l3mdev: prefer VRF master for source address selection

When selecting an address in context of a VRF, the vrf master should be
preferred for address selection.  If it isn't, the user has a hard time
getting the system to select to their preference - the code will pick
the address off the first in-VRF interface it can find, which on a
router could well be a non-routable address.

Signed-off-by: David Lamparter <equinox@diac24.net>
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
[dsa: Fixed comment style and removed extra blank link ]
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 18d510f..8c3df2c 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1217,6 +1217,23 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope)
 no_in_dev:
 	master_idx = l3mdev_master_ifindex_rcu(dev);
 
+	/* For VRFs, the VRF device takes the place of the loopback device,
+	 * with addresses on it being preferred.  Note in such cases the
+	 * loopback device will be among the devices that fail the master_idx
+	 * equality check in the loop below.
+	 */
+	if (master_idx &&
+	    (dev = dev_get_by_index_rcu(net, master_idx)) &&
+	    (in_dev = __in_dev_get_rcu(dev))) {
+		for_primary_ifa(in_dev) {
+			if (ifa->ifa_scope != RT_SCOPE_LINK &&
+			    ifa->ifa_scope <= scope) {
+				addr = ifa->ifa_local;
+				goto out_unlock;
+			}
+		} endfor_ifa(in_dev);
+	}
+
 	/* Not loopback addresses on loopback should be preferred
 	   in this case. It is important that lo is the first interface
 	   in dev_base list.
-- 
cgit v0.10.2


From 224638766235ba82c53b4216e4dabc510701fbf2 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 24 Feb 2016 16:46:21 -0800
Subject: GSO: Provide software checksum of tunneled UDP fragmentation offload

On reviewing the code I realized that GRE and UDP tunnels could cause a
kernel panic if we used GSO to segment a large UDP frame that was sent
through the tunnel with an outer checksum and hardware offloads were not
available.

In order to correct this we need to update the feature flags that are
passed to the skb_segment function so that in the event of UDP
fragmentation being requested for the inner header the segmentation
function will correctly generate the checksum for the payload if we cannot
segment the outer header.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 003b0eb..47f4c54 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -24,7 +24,7 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	__be16 protocol = skb->protocol;
 	u16 mac_len = skb->mac_len;
 	int gre_offset, outer_hlen;
-	bool need_csum;
+	bool need_csum, ufo;
 
 	if (unlikely(skb_shinfo(skb)->gso_type &
 				~(SKB_GSO_TCPV4 |
@@ -58,8 +58,20 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE_CSUM);
 	skb->encap_hdr_csum = need_csum;
 
+	ufo = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP);
+
 	features &= skb->dev->hw_enc_features;
 
+	/* The only checksum offload we care about from here on out is the
+	 * outer one so strip the existing checksum feature flags based
+	 * on the fact that we will be computing our checksum in software.
+	 */
+	if (ufo) {
+		features &= ~NETIF_F_CSUM_MASK;
+		if (!need_csum)
+			features |= NETIF_F_HW_CSUM;
+	}
+
 	/* segment inner packet. */
 	segs = skb_mac_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
@@ -75,8 +87,11 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 		struct gre_base_hdr *greh;
 		__be32 *pcsum;
 
-		skb_reset_inner_headers(skb);
-		skb->encapsulation = 1;
+		/* Set up inner headers if we are offloading inner checksum */
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			skb_reset_inner_headers(skb);
+			skb->encapsulation = 1;
+		}
 
 		skb->mac_len = mac_len;
 		skb->protocol = protocol;
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 56c4c8b..f5abb1a 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -33,8 +33,8 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	__be16 new_protocol, bool is_ipv6)
 {
 	int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
+	bool remcsum, need_csum, offload_csum, ufo;
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
-	bool remcsum, need_csum, offload_csum;
 	struct udphdr *uh = udp_hdr(skb);
 	u16 mac_offset = skb->mac_header;
 	__be16 protocol = skb->protocol;
@@ -62,6 +62,8 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	remcsum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_TUNNEL_REMCSUM);
 	skb->remcsum_offload = remcsum;
 
+	ufo = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP);
+
 	/* Try to offload checksum if possible */
 	offload_csum = !!(need_csum &&
 			  (skb->dev->features &
@@ -74,9 +76,9 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	 * outer one so strip the existing checksum feature flags and
 	 * instead set the flag based on our outer checksum offload value.
 	 */
-	if (remcsum) {
+	if (remcsum || ufo) {
 		features &= ~NETIF_F_CSUM_MASK;
-		if (offload_csum)
+		if (!need_csum || offload_csum)
 			features |= NETIF_F_HW_CSUM;
 	}
 
@@ -230,6 +232,13 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb,
 
 	skb->ip_summed = CHECKSUM_NONE;
 
+	/* If there is no outer header we can fake a checksum offload
+	 * due to the fact that we have already done the checksum in
+	 * software prior to segmenting the frame.
+	 */
+	if (!skb->encap_hdr_csum)
+		features |= NETIF_F_HW_CSUM;
+
 	/* Fragment the skb. IP headers of the fragments are updated in
 	 * inet_gso_segment()
 	 */
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index 7441e1e..2b0fbe6 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -81,12 +81,18 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb,
 		csum = skb_checksum(skb, 0, skb->len, 0);
 		uh->check = udp_v6_check(skb->len, &ipv6h->saddr,
 					  &ipv6h->daddr, csum);
-
 		if (uh->check == 0)
 			uh->check = CSUM_MANGLED_0;
 
 		skb->ip_summed = CHECKSUM_NONE;
 
+		/* If there is no outer header we can fake a checksum offload
+		 * due to the fact that we have already done the checksum in
+		 * software prior to segmenting the frame.
+		 */
+		if (!skb->encap_hdr_csum)
+			features |= NETIF_F_HW_CSUM;
+
 		/* Check if there is enough headroom to insert fragment header. */
 		tnl_hlen = skb_tnl_header_len(skb);
 		if (skb->mac_header < (tnl_hlen + frag_hdr_sz)) {
-- 
cgit v0.10.2


From f73e0c24bf075cf934e73133f5e89052afac1d32 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Sat, 27 Feb 2016 14:31:43 +0300
Subject: rocker: fix an error code

We intended to return PTR_ERR() here instead of 1.

Fixes: 1f9993f6825f ('rocker: fix a neigh entry leak issue')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
index 099008a..07218c3 100644
--- a/drivers/net/ethernet/rocker/rocker_ofdpa.c
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -1449,7 +1449,7 @@ static int ofdpa_port_ipv4_resolve(struct ofdpa_port *ofdpa_port,
 	if (!n) {
 		n = neigh_create(&arp_tbl, &ip_addr, dev);
 		if (IS_ERR(n))
-			return IS_ERR(n);
+			return PTR_ERR(n);
 	}
 
 	/* If the neigh is already resolved, then go ahead and
-- 
cgit v0.10.2


From 0a3b7119000d706dfbc7e0c5b66e192a646d365f Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Wed, 16 Dec 2015 16:34:55 +0200
Subject: iwlwifi: mvm: add CT-KILL notification

Up to today the driver was notified of the temperature from the FW
and decided whether to enter CT-kill or not.
From now on, the FW will decide when to enter CT-kill and will notify
the driver.
Add support for this notification.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index e2dbc67..724a3ee 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -318,6 +318,9 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t;
  * @IWL_UCODE_TLV_CAPA_BEACON_STORING: firmware will store the latest beacon
  *	from AP and will send it upon d0i3 exit.
  * @IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2: support LAR API V2
+ * @IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW: firmware responsible for CT-kill
+ * @IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT: supports temperature
+ *	thresholds reporting
  *
  * @NUM_IWL_UCODE_TLV_CAPA: number of bits used
  */
@@ -351,6 +354,8 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION		= (__force iwl_ucode_tlv_capa_t)71,
 	IWL_UCODE_TLV_CAPA_BEACON_STORING		= (__force iwl_ucode_tlv_capa_t)72,
 	IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2		= (__force iwl_ucode_tlv_capa_t)73,
+	IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW		= (__force iwl_ucode_tlv_capa_t)74,
+	IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT	= (__force iwl_ucode_tlv_capa_t)75,
 
 	NUM_IWL_UCODE_TLV_CAPA
 #ifdef __CHECKER__
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index f332497..ecbf7cb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -279,6 +279,7 @@ enum {
  */
 enum iwl_phy_ops_subcmd_ids {
 	CMD_DTS_MEASUREMENT_TRIGGER_WIDE = 0x0,
+	CT_KILL_NOTIFICATION = 0xFE,
 	DTS_MEASUREMENT_NOTIF_WIDE = 0xFF,
 };
 
@@ -1685,6 +1686,17 @@ struct iwl_dts_measurement_notif {
 	__le32 voltage;
 } __packed; /* TEMPERATURE_MEASUREMENT_TRIGGER_NTFY_S */
 
+/**
+ * struct ct_kill_notif - CT-kill entry notification
+ *
+ * @temperature: the current temperature in celsius
+ * @reserved: reserved
+ */
+struct ct_kill_notif {
+	__le16 temperature;
+	__le16 reserved;
+} __packed; /* GRP_PHY_CT_KILL_NTF */
+
 /***********************************
  * TDLS API
  ***********************************/
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ebe37bb..200bbb7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -1028,6 +1030,18 @@ static inline bool iwl_mvm_has_new_rx_api(struct iwl_mvm *mvm)
 			   IWL_UCODE_TLV_CAPA_MULTI_QUEUE_RX_SUPPORT);
 }
 
+static inline bool iwl_mvm_is_tt_in_fw(struct iwl_mvm *mvm)
+{
+	/* these two TLV are redundant since the responsibility to CT-kill by
+	 * FW happens only after we send at least one command of
+	 * temperature THs report.
+	 */
+	return fw_has_capa(&mvm->fw->ucode_capa,
+			   IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW) &&
+	       fw_has_capa(&mvm->fw->ucode_capa,
+			   IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT);
+}
+
 extern const u8 iwl_mvm_ac_to_tx_fifo[];
 
 struct iwl_rate_info {
@@ -1502,6 +1516,7 @@ void iwl_mvm_tt_initialize(struct iwl_mvm *mvm, u32 min_backoff);
 void iwl_mvm_tt_exit(struct iwl_mvm *mvm);
 void iwl_mvm_set_hw_ctkill_state(struct iwl_mvm *mvm, bool state);
 int iwl_mvm_get_temp(struct iwl_mvm *mvm, s32 *temp);
+void iwl_mvm_ct_kill_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 
 /* Location Aware Regulatory */
 struct iwl_mcc_update_resp *
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 09a94a5..ecc371e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -263,6 +263,8 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 	RX_HANDLER(DTS_MEASUREMENT_NOTIFICATION, iwl_mvm_temp_notif, true),
 	RX_HANDLER_GRP(PHY_OPS_GROUP, DTS_MEASUREMENT_NOTIF_WIDE,
 		       iwl_mvm_temp_notif, true),
+	RX_HANDLER_GRP(PHY_OPS_GROUP, CT_KILL_NOTIFICATION,
+		       iwl_mvm_ct_kill_notif, false),
 
 	RX_HANDLER(TDLS_CHANNEL_SWITCH_NOTIFICATION, iwl_mvm_rx_tdls_notif,
 		   true),
@@ -387,6 +389,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
  */
 static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
 	HCMD_NAME(CMD_DTS_MEASUREMENT_TRIGGER_WIDE),
+	HCMD_NAME(CT_KILL_NOTIFICATION),
 	HCMD_NAME(DTS_MEASUREMENT_NOTIF_WIDE),
 };
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 758d05a..6ba3910 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2013 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,7 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
- * Copyright(c) 2015 Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -171,6 +172,24 @@ void iwl_mvm_temp_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
 	iwl_mvm_tt_temp_changed(mvm, temp);
 }
 
+void iwl_mvm_ct_kill_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct ct_kill_notif *notif;
+	int len = iwl_rx_packet_payload_len(pkt);
+
+	if (WARN_ON_ONCE(len != sizeof(*notif))) {
+		IWL_ERR(mvm, "Invalid CT_KILL_NOTIFICATION\n");
+		return;
+	}
+
+	notif = (struct ct_kill_notif *)pkt->data;
+	IWL_DEBUG_TEMP(mvm, "CT Kill notification temperature = %d\n",
+		       notif->temperature);
+
+	iwl_mvm_enter_ctkill(mvm);
+}
+
 static int iwl_mvm_get_temp_cmd(struct iwl_mvm *mvm)
 {
 	struct iwl_dts_measurement_cmd cmd = {
@@ -236,6 +255,12 @@ static void check_exit_ctkill(struct work_struct *work)
 	tt = container_of(work, struct iwl_mvm_tt_mgmt, ct_kill_exit.work);
 	mvm = container_of(tt, struct iwl_mvm, thermal_throttle);
 
+	if (iwl_mvm_is_tt_in_fw(mvm)) {
+		iwl_mvm_exit_ctkill(mvm);
+
+		return;
+	}
+
 	duration = tt->params.ct_kill_duration;
 
 	mutex_lock(&mvm->mutex);
-- 
cgit v0.10.2


From c221daf219b1cf38e7c4307f0f420ea826678af5 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Tue, 29 Dec 2015 09:54:49 +0200
Subject: iwlwifi: mvm: add registration to thermal zone

Register to thermal_zone interface and implement the
thermal ops.
The thermal handles the device throttling, and sets the
the temperature thresholds the Thermal Manager would be
notified of crossing.
The thermal interface adds a new thermal zone device sensor
under /sys/class/thermal/ folder.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index ecbf7cb..e692098 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -279,6 +279,7 @@ enum {
  */
 enum iwl_phy_ops_subcmd_ids {
 	CMD_DTS_MEASUREMENT_TRIGGER_WIDE = 0x0,
+	TEMP_REPORTING_THRESHOLDS_CMD = 0x04,
 	CT_KILL_NOTIFICATION = 0xFE,
 	DTS_MEASUREMENT_NOTIF_WIDE = 0xFF,
 };
@@ -1676,15 +1677,28 @@ struct iwl_ext_dts_measurement_cmd {
 } __packed; /* XVT_FW_DTS_CONTROL_MEASUREMENT_REQUEST_API_S */
 
 /**
- * iwl_dts_measurement_notif - notification received with the measurements
+ * struct iwl_dts_measurement_notif_v1 - measurements notification
  *
  * @temp: the measured temperature
  * @voltage: the measured voltage
  */
-struct iwl_dts_measurement_notif {
+struct iwl_dts_measurement_notif_v1 {
 	__le32 temp;
 	__le32 voltage;
-} __packed; /* TEMPERATURE_MEASUREMENT_TRIGGER_NTFY_S */
+} __packed; /* TEMPERATURE_MEASUREMENT_TRIGGER_NTFY_S_VER_1*/
+
+/**
+ * struct iwl_dts_measurement_notif_v2 - measurements notification
+ *
+ * @temp: the measured temperature
+ * @voltage: the measured voltage
+ * @threshold_idx: the trip index that was crossed
+ */
+struct iwl_dts_measurement_notif_v2 {
+	__le32 temp;
+	__le32 voltage;
+	__le32 threshold_idx;
+} __packed; /* TEMPERATURE_MEASUREMENT_TRIGGER_NTFY_S_VER_2 */
 
 /**
  * struct ct_kill_notif - CT-kill entry notification
@@ -1697,6 +1711,19 @@ struct ct_kill_notif {
 	__le16 reserved;
 } __packed; /* GRP_PHY_CT_KILL_NTF */
 
+#define IWL_MAX_DTS_TRIPS	8
+
+/**
+ * struct iwl_temp_report_ths_cmd - set temperature thresholds
+ *
+ * @num_temps: number of temperature thresholds passed
+ * @thresholds: array with the thresholds to be configured
+ */
+struct temp_report_ths_cmd {
+	__le32 num_temps;
+	__le16 thresholds[IWL_MAX_DTS_TRIPS];
+} __packed; /* GRP_PHY_TEMP_REPORTING_THRESHOLDS_CMD */
+
 /***********************************
  * TDLS API
  ***********************************/
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 070e2af..07f2cbd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -952,8 +952,21 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 			goto error;
 	}
 
+#ifdef CONFIG_THERMAL
+	if (iwl_mvm_is_tt_in_fw(mvm)) {
+		/* in order to give the responsibility of ct-kill and
+		 * TX backoff to FW we need to send empty temperature reporting
+		 * cmd during init time
+		 */
+		iwl_mvm_send_temp_report_ths_cmd(mvm);
+	} else {
+		/* Initialize tx backoffs to the minimal possible */
+		iwl_mvm_tt_tx_backoff(mvm, 0);
+	}
+#else
 	/* Initialize tx backoffs to the minimal possible */
 	iwl_mvm_tt_tx_backoff(mvm, 0);
+#endif
 
 	WARN_ON(iwl_mvm_config_ltr(mvm));
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 200bbb7..87d3e28 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -73,6 +73,10 @@
 #include <linux/leds.h>
 #include <linux/in6.h>
 
+#ifdef CONFIG_THERMAL
+#include <linux/thermal.h>
+#endif
+
 #include "iwl-op-mode.h"
 #include "iwl-trans.h"
 #include "iwl-notif-wait.h"
@@ -519,6 +523,20 @@ struct iwl_mvm_tt_mgmt {
 	bool throttle;
 };
 
+#ifdef CONFIG_THERMAL
+/**
+ *struct iwl_mvm_thermal_device - thermal zone related data
+ * @temp_trips: temperature thresholds for report
+ * @fw_trips_index: keep indexes to original array - temp_trips
+ * @tzone: thermal zone device data
+*/
+struct iwl_mvm_thermal_device {
+	s16 temp_trips[IWL_MAX_DTS_TRIPS];
+	u8 fw_trips_index[IWL_MAX_DTS_TRIPS];
+	struct thermal_zone_device *tzone;
+};
+#endif
+
 #define IWL_MVM_NUM_LAST_FRAMES_UCODE_RATES 8
 
 struct iwl_mvm_frame_stats {
@@ -799,6 +817,10 @@ struct iwl_mvm {
 
 	/* Thermal Throttling and CTkill */
 	struct iwl_mvm_tt_mgmt thermal_throttle;
+#ifdef CONFIG_THERMAL
+	struct iwl_mvm_thermal_device tz_device;
+#endif
+
 	s32 temperature;	/* Celsius */
 	/*
 	 * Debug option to set the NIC temperature. This option makes the
@@ -1032,6 +1054,7 @@ static inline bool iwl_mvm_has_new_rx_api(struct iwl_mvm *mvm)
 
 static inline bool iwl_mvm_is_tt_in_fw(struct iwl_mvm *mvm)
 {
+#ifdef CONFIG_THERMAL
 	/* these two TLV are redundant since the responsibility to CT-kill by
 	 * FW happens only after we send at least one command of
 	 * temperature THs report.
@@ -1040,6 +1063,9 @@ static inline bool iwl_mvm_is_tt_in_fw(struct iwl_mvm *mvm)
 			   IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW) &&
 	       fw_has_capa(&mvm->fw->ucode_capa,
 			   IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT);
+#else /* CONFIG_THERMAL */
+	return false;
+#endif /* CONFIG_THERMAL */
 }
 
 extern const u8 iwl_mvm_ac_to_tx_fifo[];
@@ -1512,11 +1538,12 @@ void iwl_mvm_tt_temp_changed(struct iwl_mvm *mvm, u32 temp);
 void iwl_mvm_temp_notif(struct iwl_mvm *mvm,
 			struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_tt_handler(struct iwl_mvm *mvm);
-void iwl_mvm_tt_initialize(struct iwl_mvm *mvm, u32 min_backoff);
-void iwl_mvm_tt_exit(struct iwl_mvm *mvm);
+void iwl_mvm_thermal_initialize(struct iwl_mvm *mvm, u32 min_backoff);
+void iwl_mvm_thermal_exit(struct iwl_mvm *mvm);
 void iwl_mvm_set_hw_ctkill_state(struct iwl_mvm *mvm, bool state);
 int iwl_mvm_get_temp(struct iwl_mvm *mvm, s32 *temp);
 void iwl_mvm_ct_kill_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
+int iwl_mvm_send_temp_report_ths_cmd(struct iwl_mvm *mvm);
 
 /* Location Aware Regulatory */
 struct iwl_mcc_update_resp *
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index ecc371e..a7acadd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -389,6 +389,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
  */
 static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
 	HCMD_NAME(CMD_DTS_MEASUREMENT_TRIGGER_WIDE),
+	HCMD_NAME(TEMP_REPORTING_THRESHOLDS_CMD),
 	HCMD_NAME(CT_KILL_NOTIFICATION),
 	HCMD_NAME(DTS_MEASUREMENT_NOTIF_WIDE),
 };
@@ -591,7 +592,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		 mvm->cfg->name, mvm->trans->hw_rev);
 
 	min_backoff = calc_min_backoff(trans, cfg);
-	iwl_mvm_tt_initialize(mvm, min_backoff);
+	iwl_mvm_thermal_initialize(mvm, min_backoff);
 
 	if (iwlwifi_mod_params.nvm_file)
 		mvm->nvm_file_name = iwlwifi_mod_params.nvm_file;
@@ -664,6 +665,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
  out_unregister:
 	ieee80211_unregister_hw(mvm->hw);
 	iwl_mvm_leds_exit(mvm);
+	iwl_mvm_thermal_exit(mvm);
  out_free:
 	flush_delayed_work(&mvm->fw_dump_wk);
 	iwl_phy_db_free(mvm->phy_db);
@@ -681,7 +683,7 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode)
 
 	iwl_mvm_leds_exit(mvm);
 
-	iwl_mvm_tt_exit(mvm);
+	iwl_mvm_thermal_exit(mvm);
 
 	ieee80211_unregister_hw(mvm->hw);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 6ba3910..466d169 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -65,6 +65,8 @@
  *
  *****************************************************************************/
 
+#include <linux/sort.h>
+
 #include "mvm.h"
 
 #define IWL_MVM_TEMP_NOTIF_WAIT_TIMEOUT	HZ
@@ -80,8 +82,10 @@ static void iwl_mvm_enter_ctkill(struct iwl_mvm *mvm)
 	IWL_ERR(mvm, "Enter CT Kill\n");
 	iwl_mvm_set_hw_ctkill_state(mvm, true);
 
-	tt->throttle = false;
-	tt->dynamic_smps = false;
+	if (!iwl_mvm_is_tt_in_fw(mvm)) {
+		tt->throttle = false;
+		tt->dynamic_smps = false;
+	}
 
 	/* Don't schedule an exit work if we're in test mode, since
 	 * the temperature will not change unless we manually set it
@@ -117,18 +121,21 @@ void iwl_mvm_tt_temp_changed(struct iwl_mvm *mvm, u32 temp)
 static int iwl_mvm_temp_notif_parse(struct iwl_mvm *mvm,
 				    struct iwl_rx_packet *pkt)
 {
-	struct iwl_dts_measurement_notif *notif;
+	struct iwl_dts_measurement_notif_v1 *notif_v1;
 	int len = iwl_rx_packet_payload_len(pkt);
 	int temp;
 
-	if (WARN_ON_ONCE(len < sizeof(*notif))) {
+	/* we can use notif_v1 only, because v2 only adds an additional
+	 * parameter, which is not used in this function.
+	*/
+	if (WARN_ON_ONCE(len < sizeof(*notif_v1))) {
 		IWL_ERR(mvm, "Invalid DTS_MEASUREMENT_NOTIFICATION\n");
 		return -EINVAL;
 	}
 
-	notif = (void *)pkt->data;
+	notif_v1 = (void *)pkt->data;
 
-	temp = le32_to_cpu(notif->temp);
+	temp = le32_to_cpu(notif_v1->temp);
 
 	/* shouldn't be negative, but since it's s32, make sure it isn't */
 	if (WARN_ON_ONCE(temp < 0))
@@ -159,17 +166,56 @@ static bool iwl_mvm_temp_notif_wait(struct iwl_notif_wait_data *notif_wait,
 void iwl_mvm_temp_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
 {
 	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_dts_measurement_notif_v2 *notif_v2;
+	int len = iwl_rx_packet_payload_len(pkt);
 	int temp;
+	u32 ths_crossed;
 
 	/* the notification is handled synchronously in ctkill, so skip here */
 	if (test_bit(IWL_MVM_STATUS_HW_CTKILL, &mvm->status))
 		return;
 
 	temp = iwl_mvm_temp_notif_parse(mvm, pkt);
-	if (temp < 0)
+
+	if (!iwl_mvm_is_tt_in_fw(mvm)) {
+		if (temp >= 0)
+			iwl_mvm_tt_temp_changed(mvm, temp);
+		return;
+	}
+
+	if (WARN_ON_ONCE(len < sizeof(*notif_v2))) {
+		IWL_ERR(mvm, "Invalid DTS_MEASUREMENT_NOTIFICATION\n");
+		return;
+	}
+
+	notif_v2 = (void *)pkt->data;
+	ths_crossed = le32_to_cpu(notif_v2->threshold_idx);
+
+	/* 0xFF in ths_crossed means the notification is not related
+	 * to a trip, so we can ignore it here.
+	 */
+	if (ths_crossed == 0xFF)
+		return;
+
+	IWL_DEBUG_TEMP(mvm, "Temp = %d Threshold crossed = %d\n",
+		       temp, ths_crossed);
+
+#ifdef CONFIG_THERMAL
+	if (WARN_ON(ths_crossed >= IWL_MAX_DTS_TRIPS))
 		return;
 
-	iwl_mvm_tt_temp_changed(mvm, temp);
+	/*
+	 * We are now handling a temperature notification from the firmware
+	 * in ASYNC and hold the mutex. thermal_notify_framework will call
+	 * us back through get_temp() which ought to send a SYNC command to
+	 * the firmware and hence to take the mutex.
+	 * Avoid the deadlock by unlocking the mutex here.
+	 */
+	mutex_unlock(&mvm->mutex);
+	thermal_notify_framework(mvm->tz_device.tzone,
+				 mvm->tz_device.fw_trips_index[ths_crossed]);
+	mutex_lock(&mvm->mutex);
+#endif /* CONFIG_THERMAL */
 }
 
 void iwl_mvm_ct_kill_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
@@ -460,7 +506,220 @@ static const struct iwl_tt_params iwl_mvm_default_tt_params = {
 	.support_tx_backoff = true,
 };
 
-void iwl_mvm_tt_initialize(struct iwl_mvm *mvm, u32 min_backoff)
+#ifdef CONFIG_THERMAL
+static int compare_temps(const void *a, const void *b)
+{
+	return ((s16)le16_to_cpu(*(__le16 *)a) -
+		(s16)le16_to_cpu(*(__le16 *)b));
+}
+
+int iwl_mvm_send_temp_report_ths_cmd(struct iwl_mvm *mvm)
+{
+	struct temp_report_ths_cmd cmd = {0};
+	int ret, i, j, idx = 0;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	/* The driver holds array of temperature trips that are unsorted
+	 * and uncompressed, the FW should get it compressed and sorted
+	 */
+
+	/* compress temp_trips to cmd array, remove uninitialized values*/
+	for (i = 0; i < IWL_MAX_DTS_TRIPS; i++)
+		if (mvm->tz_device.temp_trips[i] != S16_MIN) {
+			cmd.thresholds[idx++] =
+				cpu_to_le16(mvm->tz_device.temp_trips[i]);
+		}
+	cmd.num_temps = cpu_to_le32(idx);
+
+	if (!idx)
+		goto send;
+
+	/*sort cmd array*/
+	sort(cmd.thresholds, idx, sizeof(s16), compare_temps, NULL);
+
+	/* we should save the indexes of trips because we sort
+	 * and compress the orginal array
+	 */
+	for (i = 0; i < idx; i++) {
+		for (j = 0; j < IWL_MAX_DTS_TRIPS; j++) {
+			if (le16_to_cpu(cmd.thresholds[i]) ==
+				mvm->tz_device.temp_trips[j])
+				mvm->tz_device.fw_trips_index[i] = j;
+		}
+	}
+
+send:
+	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(PHY_OPS_GROUP,
+						TEMP_REPORTING_THRESHOLDS_CMD),
+				   0, sizeof(cmd), &cmd);
+	if (ret)
+		IWL_ERR(mvm, "TEMP_REPORT_THS_CMD command failed (err=%d)\n",
+			ret);
+
+	return ret;
+}
+
+static int iwl_mvm_tzone_get_temp(struct thermal_zone_device *device,
+				  int *temperature)
+{
+	struct iwl_mvm *mvm = (struct iwl_mvm *)device->devdata;
+	int ret;
+	int temp;
+
+	mutex_lock(&mvm->mutex);
+
+	if (!mvm->ucode_loaded || !(mvm->cur_ucode == IWL_UCODE_REGULAR)) {
+		ret = -EIO;
+		goto out;
+	}
+
+	ret = iwl_mvm_get_temp(mvm, &temp);
+	if (ret)
+		goto out;
+
+	*temperature = temp * 1000;
+
+out:
+	mutex_unlock(&mvm->mutex);
+	return ret;
+}
+
+static int iwl_mvm_tzone_get_trip_temp(struct thermal_zone_device *device,
+				       int trip, int *temp)
+{
+	struct iwl_mvm *mvm = (struct iwl_mvm *)device->devdata;
+
+	if (trip < 0 || trip >= IWL_MAX_DTS_TRIPS)
+		return -EINVAL;
+
+	*temp = mvm->tz_device.temp_trips[trip] * 1000;
+
+	return 0;
+}
+
+static int iwl_mvm_tzone_get_trip_type(struct thermal_zone_device *device,
+				       int trip, enum thermal_trip_type *type)
+{
+	if (trip < 0 || trip >= IWL_MAX_DTS_TRIPS)
+		return -EINVAL;
+
+	*type = THERMAL_TRIP_PASSIVE;
+
+	return 0;
+}
+
+static int iwl_mvm_tzone_set_trip_temp(struct thermal_zone_device *device,
+				       int trip, int temp)
+{
+	struct iwl_mvm *mvm = (struct iwl_mvm *)device->devdata;
+	struct iwl_mvm_thermal_device *tzone;
+	int i, ret;
+	s16 temperature;
+
+	mutex_lock(&mvm->mutex);
+
+	if (!mvm->ucode_loaded || !(mvm->cur_ucode == IWL_UCODE_REGULAR)) {
+		ret = -EIO;
+		goto out;
+	}
+
+	if (trip < 0 || trip >= IWL_MAX_DTS_TRIPS) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if ((temp / 1000) > S16_MAX) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	temperature = (s16)(temp / 1000);
+	tzone = &mvm->tz_device;
+
+	if (!tzone) {
+		ret = -EIO;
+		goto out;
+	}
+
+	/* no updates*/
+	if (tzone->temp_trips[trip] == temperature) {
+		ret = 0;
+		goto out;
+	}
+
+	/* already existing temperature */
+	for (i = 0; i < IWL_MAX_DTS_TRIPS; i++) {
+		if (tzone->temp_trips[i] == temperature) {
+			ret = -EINVAL;
+			goto out;
+		}
+	}
+
+	tzone->temp_trips[trip] = temperature;
+
+	ret = iwl_mvm_send_temp_report_ths_cmd(mvm);
+out:
+	mutex_unlock(&mvm->mutex);
+	return ret;
+}
+
+static  struct thermal_zone_device_ops tzone_ops = {
+	.get_temp = iwl_mvm_tzone_get_temp,
+	.get_trip_temp = iwl_mvm_tzone_get_trip_temp,
+	.get_trip_type = iwl_mvm_tzone_get_trip_type,
+	.set_trip_temp = iwl_mvm_tzone_set_trip_temp,
+};
+
+/* make all trips writable */
+#define IWL_WRITABLE_TRIPS_MSK (BIT(IWL_MAX_DTS_TRIPS) - 1)
+
+static void iwl_mvm_thermal_zone_register(struct iwl_mvm *mvm)
+{
+	int i;
+	char name[] = "iwlwifi";
+
+	if (!iwl_mvm_is_tt_in_fw(mvm)) {
+		mvm->tz_device.tzone = NULL;
+
+		return;
+	}
+
+	BUILD_BUG_ON(ARRAY_SIZE(name) >= THERMAL_NAME_LENGTH);
+
+	mvm->tz_device.tzone = thermal_zone_device_register(name,
+							IWL_MAX_DTS_TRIPS,
+							IWL_WRITABLE_TRIPS_MSK,
+							mvm, &tzone_ops,
+							NULL, 0, 0);
+	if (IS_ERR(mvm->tz_device.tzone)) {
+		IWL_DEBUG_TEMP(mvm,
+			       "Failed to register to thermal zone (err = %ld)\n",
+			       PTR_ERR(mvm->tz_device.tzone));
+		return;
+	}
+
+	/* 0 is a valid temperature,
+	 * so initialize the array with S16_MIN which invalid temperature
+	 */
+	for (i = 0 ; i < IWL_MAX_DTS_TRIPS; i++)
+		mvm->tz_device.temp_trips[i] = S16_MIN;
+}
+
+static void iwl_mvm_thermal_zone_unregister(struct iwl_mvm *mvm)
+{
+	if (!iwl_mvm_is_tt_in_fw(mvm))
+		return;
+
+	if (mvm->tz_device.tzone) {
+		IWL_DEBUG_TEMP(mvm, "Thermal zone device unregister\n");
+		thermal_zone_device_unregister(mvm->tz_device.tzone);
+		mvm->tz_device.tzone = NULL;
+	}
+}
+#endif /* CONFIG_THERMAL */
+
+void iwl_mvm_thermal_initialize(struct iwl_mvm *mvm, u32 min_backoff)
 {
 	struct iwl_mvm_tt_mgmt *tt = &mvm->thermal_throttle;
 
@@ -475,10 +734,18 @@ void iwl_mvm_tt_initialize(struct iwl_mvm *mvm, u32 min_backoff)
 	tt->dynamic_smps = false;
 	tt->min_backoff = min_backoff;
 	INIT_DELAYED_WORK(&tt->ct_kill_exit, check_exit_ctkill);
+
+#ifdef CONFIG_THERMAL
+	iwl_mvm_thermal_zone_register(mvm);
+#endif
 }
 
-void iwl_mvm_tt_exit(struct iwl_mvm *mvm)
+void iwl_mvm_thermal_exit(struct iwl_mvm *mvm)
 {
 	cancel_delayed_work_sync(&mvm->thermal_throttle.ct_kill_exit);
 	IWL_DEBUG_TEMP(mvm, "Exit Thermal Throttling\n");
+
+#ifdef CONFIG_THERMAL
+	iwl_mvm_thermal_zone_unregister(mvm);
+#endif
 }
-- 
cgit v0.10.2


From 5c89e7bc557e1dd5650946a0fe4940166bc62abf Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Tue, 5 Jan 2016 10:34:47 +0200
Subject: iwlwifi: mvm: add registration to cooling device

Register cooling device in order to have the Thermal
Manager handle the device's power budget according to the sent
notifications.
The interface adds a new thermal cooling device to
/sys/class/thermal/ folder.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 724a3ee..63dc109 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -321,6 +321,7 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t;
  * @IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW: firmware responsible for CT-kill
  * @IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT: supports temperature
  *	thresholds reporting
+ * @IWL_UCODE_TLV_CAPA_CTDP_SUPPORT: supports cTDP command
  *
  * @NUM_IWL_UCODE_TLV_CAPA: number of bits used
  */
@@ -356,6 +357,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2		= (__force iwl_ucode_tlv_capa_t)73,
 	IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW		= (__force iwl_ucode_tlv_capa_t)74,
 	IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT	= (__force iwl_ucode_tlv_capa_t)75,
+	IWL_UCODE_TLV_CAPA_CTDP_SUPPORT			= (__force iwl_ucode_tlv_capa_t)76,
 
 	NUM_IWL_UCODE_TLV_CAPA
 #ifdef __CHECKER__
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index e692098..d5f9037 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -279,6 +279,7 @@ enum {
  */
 enum iwl_phy_ops_subcmd_ids {
 	CMD_DTS_MEASUREMENT_TRIGGER_WIDE = 0x0,
+	CTDP_CONFIG_CMD = 0x03,
 	TEMP_REPORTING_THRESHOLDS_CMD = 0x04,
 	CT_KILL_NOTIFICATION = 0xFE,
 	DTS_MEASUREMENT_NOTIF_WIDE = 0xFF,
@@ -1711,6 +1712,31 @@ struct ct_kill_notif {
 	__le16 reserved;
 } __packed; /* GRP_PHY_CT_KILL_NTF */
 
+/**
+* enum ctdp_cmd_operation - CTDP command operations
+* @CTDP_CMD_OPERATION_START: update the current budget
+* @CTDP_CMD_OPERATION_STOP: stop ctdp
+* @CTDP_CMD_OPERATION_REPORT: get the avgerage budget
+*/
+enum iwl_mvm_ctdp_cmd_operation {
+	CTDP_CMD_OPERATION_START	= 0x1,
+	CTDP_CMD_OPERATION_STOP		= 0x2,
+	CTDP_CMD_OPERATION_REPORT	= 0x4,
+};/* CTDP_CMD_OPERATION_TYPE_E */
+
+/**
+ * struct iwl_mvm_ctdp_cmd - track and manage the FW power consumption budget
+ *
+ * @operation: see &enum iwl_mvm_ctdp_cmd_operation
+ * @budget: the budget in milliwatt
+ * @window_size: defined in API but not used
+ */
+struct iwl_mvm_ctdp_cmd {
+	__le32 operation;
+	__le32 budget;
+	__le32 window_size;
+} __packed;
+
 #define IWL_MAX_DTS_TRIPS	8
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 07f2cbd..3e596e4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -963,6 +963,11 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 		/* Initialize tx backoffs to the minimal possible */
 		iwl_mvm_tt_tx_backoff(mvm, 0);
 	}
+
+	/* TODO: read the budget from BIOS / Platform NVM */
+	if (iwl_mvm_is_ctdp_supported(mvm) && mvm->cooling_dev.cur_state > 0)
+		ret = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_START,
+					   mvm->cooling_dev.cur_state);
 #else
 	/* Initialize tx backoffs to the minimal possible */
 	iwl_mvm_tt_tx_backoff(mvm, 0);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 87d3e28..d867cb4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -535,6 +535,16 @@ struct iwl_mvm_thermal_device {
 	u8 fw_trips_index[IWL_MAX_DTS_TRIPS];
 	struct thermal_zone_device *tzone;
 };
+
+/*
+ * iwl_mvm_cooling_device
+ * @cur_state: current state in milliwatts
+ * @cdev: struct thermal cooling device
+ */
+struct iwl_mvm_cooling_device {
+	u32 cur_state;
+	struct thermal_cooling_device *cdev;
+};
 #endif
 
 #define IWL_MVM_NUM_LAST_FRAMES_UCODE_RATES 8
@@ -819,6 +829,7 @@ struct iwl_mvm {
 	struct iwl_mvm_tt_mgmt thermal_throttle;
 #ifdef CONFIG_THERMAL
 	struct iwl_mvm_thermal_device tz_device;
+	struct iwl_mvm_cooling_device cooling_dev;
 #endif
 
 	s32 temperature;	/* Celsius */
@@ -1068,6 +1079,12 @@ static inline bool iwl_mvm_is_tt_in_fw(struct iwl_mvm *mvm)
 #endif /* CONFIG_THERMAL */
 }
 
+static inline bool iwl_mvm_is_ctdp_supported(struct iwl_mvm *mvm)
+{
+	return fw_has_capa(&mvm->fw->ucode_capa,
+			   IWL_UCODE_TLV_CAPA_CTDP_SUPPORT);
+}
+
 extern const u8 iwl_mvm_ac_to_tx_fifo[];
 
 struct iwl_rate_info {
@@ -1544,6 +1561,8 @@ void iwl_mvm_set_hw_ctkill_state(struct iwl_mvm *mvm, bool state);
 int iwl_mvm_get_temp(struct iwl_mvm *mvm, s32 *temp);
 void iwl_mvm_ct_kill_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 int iwl_mvm_send_temp_report_ths_cmd(struct iwl_mvm *mvm);
+int iwl_mvm_cooling_device_register(struct iwl_mvm *mvm);
+int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget);
 
 /* Location Aware Regulatory */
 struct iwl_mcc_update_resp *
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index a7acadd..ec36046 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -389,6 +389,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
  */
 static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
 	HCMD_NAME(CMD_DTS_MEASUREMENT_TRIGGER_WIDE),
+	HCMD_NAME(CTDP_CONFIG_CMD),
 	HCMD_NAME(TEMP_REPORTING_THRESHOLDS_CMD),
 	HCMD_NAME(CT_KILL_NOTIFICATION),
 	HCMD_NAME(DTS_MEASUREMENT_NOTIF_WIDE),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 466d169..999bcb8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -706,6 +706,139 @@ static void iwl_mvm_thermal_zone_register(struct iwl_mvm *mvm)
 		mvm->tz_device.temp_trips[i] = S16_MIN;
 }
 
+static const u32 iwl_mvm_cdev_budgets[] = {
+	2000,	/* cooling state 0 */
+	1800,	/* cooling state 1 */
+	1600,	/* cooling state 2 */
+	1400,	/* cooling state 3 */
+	1200,	/* cooling state 4 */
+	1000,	/* cooling state 5 */
+	900,	/* cooling state 6 */
+	800,	/* cooling state 7 */
+	700,	/* cooling state 8 */
+	650,	/* cooling state 9 */
+	600,	/* cooling state 10 */
+	550,	/* cooling state 11 */
+	500,	/* cooling state 12 */
+	450,	/* cooling state 13 */
+	400,	/* cooling state 14 */
+	350,	/* cooling state 15 */
+	300,	/* cooling state 16 */
+	250,	/* cooling state 17 */
+	200,	/* cooling state 18 */
+	150,	/* cooling state 19 */
+};
+
+int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget)
+{
+	struct iwl_mvm_ctdp_cmd cmd = {
+		.operation = cpu_to_le32(op),
+		.budget = cpu_to_le32(budget),
+		.window_size = 0,
+	};
+	int ret;
+	u32 status;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, WIDE_ID(PHY_OPS_GROUP,
+						       CTDP_CONFIG_CMD),
+					  sizeof(cmd), &cmd, &status);
+
+	if (ret) {
+		IWL_ERR(mvm, "cTDP command failed (err=%d)\n", ret);
+		return ret;
+	}
+
+	if (op == CTDP_CMD_OPERATION_START)
+		mvm->cooling_dev.cur_state = budget;
+
+	else if (op == CTDP_CMD_OPERATION_REPORT)
+		IWL_DEBUG_TEMP(mvm, "cTDP avg energy in mWatt = %d\n", status);
+
+	return 0;
+}
+
+static int iwl_mvm_tcool_get_max_state(struct thermal_cooling_device *cdev,
+				       unsigned long *state)
+{
+	*state = ARRAY_SIZE(iwl_mvm_cdev_budgets) - 1;
+
+	return 0;
+}
+
+static int iwl_mvm_tcool_get_cur_state(struct thermal_cooling_device *cdev,
+				       unsigned long *state)
+{
+	struct iwl_mvm *mvm = (struct iwl_mvm *)(cdev->devdata);
+
+	if (test_bit(IWL_MVM_STATUS_IN_D0I3, &mvm->status))
+		return -EBUSY;
+
+	*state = mvm->cooling_dev.cur_state;
+	return 0;
+}
+
+static int iwl_mvm_tcool_set_cur_state(struct thermal_cooling_device *cdev,
+				       unsigned long new_state)
+{
+	struct iwl_mvm *mvm = (struct iwl_mvm *)(cdev->devdata);
+	int ret;
+
+	if (!mvm->ucode_loaded || !(mvm->cur_ucode == IWL_UCODE_REGULAR))
+		return -EIO;
+
+	if (test_bit(IWL_MVM_STATUS_IN_D0I3, &mvm->status))
+		return -EBUSY;
+
+	mutex_lock(&mvm->mutex);
+
+	if (new_state >= ARRAY_SIZE(iwl_mvm_cdev_budgets)) {
+		ret = -EINVAL;
+		goto unlock;
+	}
+
+	ret = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_START,
+				   iwl_mvm_cdev_budgets[new_state]);
+
+unlock:
+	mutex_unlock(&mvm->mutex);
+	return ret;
+}
+
+static struct thermal_cooling_device_ops tcooling_ops = {
+	.get_max_state = iwl_mvm_tcool_get_max_state,
+	.get_cur_state = iwl_mvm_tcool_get_cur_state,
+	.set_cur_state = iwl_mvm_tcool_set_cur_state,
+};
+
+int iwl_mvm_cooling_device_register(struct iwl_mvm *mvm)
+{
+	char name[] = "iwlwifi";
+
+	if (!iwl_mvm_is_ctdp_supported(mvm)) {
+		mvm->cooling_dev.cdev = NULL;
+
+		return 0;
+	}
+
+	BUILD_BUG_ON(ARRAY_SIZE(name) >= THERMAL_NAME_LENGTH);
+
+	mvm->cooling_dev.cdev =
+		thermal_cooling_device_register(name,
+						mvm,
+						&tcooling_ops);
+
+	if (IS_ERR(mvm->cooling_dev.cdev)) {
+		IWL_DEBUG_TEMP(mvm,
+			       "Failed to register to cooling device (err = %ld)\n",
+			       PTR_ERR(mvm->cooling_dev.cdev));
+		return PTR_ERR(mvm->cooling_dev.cdev);
+	}
+
+	return 0;
+}
+
 static void iwl_mvm_thermal_zone_unregister(struct iwl_mvm *mvm)
 {
 	if (!iwl_mvm_is_tt_in_fw(mvm))
@@ -717,6 +850,18 @@ static void iwl_mvm_thermal_zone_unregister(struct iwl_mvm *mvm)
 		mvm->tz_device.tzone = NULL;
 	}
 }
+
+static void iwl_mvm_cooling_device_unregister(struct iwl_mvm *mvm)
+{
+	if (!iwl_mvm_is_ctdp_supported(mvm))
+		return;
+
+	if (mvm->cooling_dev.cdev) {
+		IWL_DEBUG_TEMP(mvm, "Cooling device unregister\n");
+		thermal_cooling_device_unregister(mvm->cooling_dev.cdev);
+		mvm->cooling_dev.cdev = NULL;
+	}
+}
 #endif /* CONFIG_THERMAL */
 
 void iwl_mvm_thermal_initialize(struct iwl_mvm *mvm, u32 min_backoff)
@@ -736,6 +881,7 @@ void iwl_mvm_thermal_initialize(struct iwl_mvm *mvm, u32 min_backoff)
 	INIT_DELAYED_WORK(&tt->ct_kill_exit, check_exit_ctkill);
 
 #ifdef CONFIG_THERMAL
+	iwl_mvm_cooling_device_register(mvm);
 	iwl_mvm_thermal_zone_register(mvm);
 #endif
 }
@@ -746,6 +892,7 @@ void iwl_mvm_thermal_exit(struct iwl_mvm *mvm)
 	IWL_DEBUG_TEMP(mvm, "Exit Thermal Throttling\n");
 
 #ifdef CONFIG_THERMAL
+	iwl_mvm_cooling_device_unregister(mvm);
 	iwl_mvm_thermal_zone_unregister(mvm);
 #endif
 }
-- 
cgit v0.10.2


From 25c2b22caceedf015d5852e3abb46d2e9271a3bd Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Sun, 7 Feb 2016 13:09:59 +0200
Subject: iwlwifi: mvm: set the correct descriptor size for tracing

The 9000 series uses a diffrent sized descriptor. Update the
relevant tracing field.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index ec36046..3661a9d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -479,8 +479,11 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 
 	if (iwl_mvm_has_new_rx_api(mvm)) {
 		op_mode->ops = &iwl_mvm_ops_mq;
+		trans->rx_mpdu_cmd_hdr_size = sizeof(struct iwl_rx_mpdu_desc);
 	} else {
 		op_mode->ops = &iwl_mvm_ops;
+		trans->rx_mpdu_cmd_hdr_size =
+			sizeof(struct iwl_rx_mpdu_res_start);
 
 		if (WARN_ON(trans->num_rx_queues > 1))
 			goto out_free;
@@ -572,7 +575,6 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	iwl_trans_configure(mvm->trans, &trans_cfg);
 
 	trans->rx_mpdu_cmd = REPLY_RX_MPDU_CMD;
-	trans->rx_mpdu_cmd_hdr_size = sizeof(struct iwl_rx_mpdu_res_start);
 	trans->dbg_dest_tlv = mvm->fw->dbg_dest_tlv;
 	trans->dbg_dest_reg_num = mvm->fw->dbg_dest_reg_num;
 	memcpy(trans->dbg_conf_tlv, mvm->fw->dbg_conf_tlv,
-- 
cgit v0.10.2


From dd4d3161d0f229031914f403721ced8cb64d4d52 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Sun, 7 Feb 2016 12:50:35 +0200
Subject: iwlwifi: mvm: fix RSS key sizing

The initialization and copying of the RSS secret key
should not use ARRAY_SIZE as we need to initialize a
dword array, and not a byte array.
Fix also the hook maximum write size to allow writing
a longer table - up to full indirection table size.

Fixes: 43413a975d06("iwlwifi: mvm: support rss queues configuration command")
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index c529e53..54d0d7d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -976,7 +976,7 @@ static ssize_t iwl_dbgfs_indirection_tbl_write(struct iwl_mvm *mvm,
 	memcpy(&cmd.indirection_table[i * nbytes], cmd.indirection_table,
 	       ARRAY_SIZE(cmd.indirection_table) % nbytes);
 
-	memcpy(cmd.secret_key, mvm->secret_key, ARRAY_SIZE(cmd.secret_key));
+	memcpy(cmd.secret_key, mvm->secret_key, sizeof(cmd.secret_key));
 
 	mutex_lock(&mvm->mutex);
 	ret = iwl_mvm_send_cmd_pdu(mvm, RSS_CONFIG_CMD, 0, sizeof(cmd), &cmd);
@@ -1497,7 +1497,8 @@ MVM_DEBUGFS_READ_WRITE_FILE_OPS(d0i3_refs, 8);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(fw_dbg_conf, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_dbg_collect, 64);
 MVM_DEBUGFS_WRITE_FILE_OPS(cont_recording, 8);
-MVM_DEBUGFS_WRITE_FILE_OPS(indirection_tbl, 16);
+MVM_DEBUGFS_WRITE_FILE_OPS(indirection_tbl,
+			   (IWL_RSS_INDIRECTION_TABLE_SIZE * 2));
 
 #ifdef CONFIG_IWLWIFI_BCAST_FILTERING
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(bcast_filters, 256);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 3e596e4..05475a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -121,7 +121,7 @@ static int iwl_send_rss_cfg_cmd(struct iwl_mvm *mvm)
 
 	for (i = 0; i < ARRAY_SIZE(cmd.indirection_table); i++)
 		cmd.indirection_table[i] = i % mvm->trans->num_rx_queues;
-	memcpy(cmd.secret_key, mvm->secret_key, ARRAY_SIZE(cmd.secret_key));
+	memcpy(cmd.secret_key, mvm->secret_key, sizeof(cmd.secret_key));
 
 	return iwl_mvm_send_cmd_pdu(mvm, RSS_CONFIG_CMD, 0, sizeof(cmd), &cmd);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 3661a9d..c5b6e86 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -661,7 +661,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	iwl_mvm_tof_init(mvm);
 
 	/* init RSS hash key */
-	get_random_bytes(mvm->secret_key, ARRAY_SIZE(mvm->secret_key));
+	get_random_bytes(mvm->secret_key, sizeof(mvm->secret_key));
 
 	return op_mode;
 
-- 
cgit v0.10.2


From a339e918daf281a0eb78783abc954c1e97af16c4 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Tue, 2 Feb 2016 22:58:46 +0200
Subject: iwlwifi: mvm: handle pass all scan reporting

The firmware doesn't send match found notifications when no matchsets
are passed.  This makes sense because if there are no matchsets,
nothing can be matched.  But the nl80211 API should report when there
are results available, even if no matchsets were passed.

To handle this, we can use the firmware's ITERATION_COMPLETE
reporting, which will send us notifications every time it completed a
scheduled scan iteration.  Then we can set a flag when we received
beacons and use that to report that results are available.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index d867cb4..bb811a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -493,6 +493,12 @@ enum iwl_mvm_scan_type {
 	IWL_SCAN_TYPE_FRAGMENTED,
 };
 
+enum iwl_mvm_sched_scan_pass_all_states {
+	SCHED_SCAN_PASS_ALL_DISABLED,
+	SCHED_SCAN_PASS_ALL_ENABLED,
+	SCHED_SCAN_PASS_ALL_FOUND,
+};
+
 /**
  * struct iwl_nvm_section - describes an NVM section in memory.
  *
@@ -687,6 +693,7 @@ struct iwl_mvm {
 	void *scan_cmd;
 	struct iwl_mcast_filter_cmd *mcast_filter_cmd;
 	enum iwl_mvm_scan_type scan_type;
+	enum iwl_mvm_sched_scan_pass_all_states sched_scan_pass_all;
 
 	/* max number of simultaneous scans the FW supports */
 	unsigned int max_scans;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index 145ec68..055a8b0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -448,6 +448,12 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi,
 	iwl_mvm_update_frame_stats(mvm, rate_n_flags,
 				   rx_status->flag & RX_FLAG_AMPDU_DETAILS);
 #endif
+
+	if (unlikely((ieee80211_is_beacon(hdr->frame_control) ||
+		      ieee80211_is_probe_resp(hdr->frame_control)) &&
+		     mvm->sched_scan_pass_all == SCHED_SCAN_PASS_ALL_ENABLED))
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_FOUND;
+
 	iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, hdr, len, ampdu_status,
 					crypt_len, rxb);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index aa6d807..09eb72c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -297,6 +299,12 @@ void iwl_mvm_rx_lmac_scan_iter_complete_notif(struct iwl_mvm *mvm,
 		       iwl_mvm_dump_channel_list(notif->results,
 						 notif->scanned_channels, buf,
 						 sizeof(buf)));
+
+	if (mvm->sched_scan_pass_all == SCHED_SCAN_PASS_ALL_FOUND) {
+		IWL_DEBUG_SCAN(mvm, "Pass all scheduled scan results found\n");
+		ieee80211_sched_scan_results(mvm->hw);
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_ENABLED;
+	}
 }
 
 void iwl_mvm_rx_scan_match_found(struct iwl_mvm *mvm,
@@ -380,6 +388,7 @@ void iwl_mvm_rx_lmac_scan_complete_notif(struct iwl_mvm *mvm,
 
 		mvm->scan_status &= ~IWL_MVM_SCAN_SCHED;
 		ieee80211_sched_scan_stopped(mvm->hw);
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED;
 	} else if (mvm->scan_status & IWL_MVM_SCAN_REGULAR) {
 		IWL_DEBUG_SCAN(mvm, "Regular scan %s, EBS status %s (FW)\n",
 			       aborted ? "aborted" : "completed",
@@ -533,10 +542,13 @@ static bool iwl_mvm_scan_pass_all(struct iwl_mvm *mvm,
 		IWL_DEBUG_SCAN(mvm,
 			       "Sending scheduled scan with filtering, n_match_sets %d\n",
 			       req->n_match_sets);
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED;
 		return false;
 	}
 
 	IWL_DEBUG_SCAN(mvm, "Sending Scheduled scan without filtering\n");
+
+	mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_ENABLED;
 	return true;
 }
 
@@ -788,6 +800,9 @@ static int iwl_mvm_scan_lmac_flags(struct iwl_mvm *mvm,
 		flags |= IWL_MVM_LMAC_SCAN_FLAG_ITER_COMPLETE;
 #endif
 
+	if (mvm->sched_scan_pass_all == SCHED_SCAN_PASS_ALL_ENABLED)
+		flags |= IWL_MVM_LMAC_SCAN_FLAG_ITER_COMPLETE;
+
 	if (iwl_mvm_is_regular_scan(params) &&
 	    vif->type != NL80211_IFTYPE_P2P_DEVICE &&
 	    params->type != IWL_SCAN_TYPE_FRAGMENTED)
@@ -1074,6 +1089,9 @@ static u32 iwl_mvm_scan_umac_flags(struct iwl_mvm *mvm,
 		flags |= IWL_UMAC_SCAN_GEN_FLAGS_ITER_COMPLETE;
 #endif
 
+	if (mvm->sched_scan_pass_all == SCHED_SCAN_PASS_ALL_ENABLED)
+		flags |= IWL_UMAC_SCAN_GEN_FLAGS_ITER_COMPLETE;
+
 	if (iwl_mvm_is_regular_scan(params) &&
 	    vif->type != NL80211_IFTYPE_P2P_DEVICE &&
 	    params->type != IWL_SCAN_TYPE_FRAGMENTED)
@@ -1301,10 +1319,6 @@ int iwl_mvm_sched_scan_start(struct iwl_mvm *mvm,
 		return -EBUSY;
 	}
 
-	/* we don't support "match all" in the firmware */
-	if (!req->n_match_sets)
-		return -EOPNOTSUPP;
-
 	ret = iwl_mvm_check_running_scans(mvm, type);
 	if (ret)
 		return ret;
@@ -1400,6 +1414,7 @@ void iwl_mvm_rx_umac_scan_complete_notif(struct iwl_mvm *mvm,
 		iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN);
 	} else if (mvm->scan_uid_status[uid] == IWL_MVM_SCAN_SCHED) {
 		ieee80211_sched_scan_stopped(mvm->hw);
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED;
 	}
 
 	mvm->scan_status &= ~mvm->scan_uid_status[uid];
@@ -1434,6 +1449,12 @@ void iwl_mvm_rx_umac_scan_iter_complete_notif(struct iwl_mvm *mvm,
 		       iwl_mvm_dump_channel_list(notif->results,
 						 notif->scanned_channels, buf,
 						 sizeof(buf)));
+
+	if (mvm->sched_scan_pass_all == SCHED_SCAN_PASS_ALL_FOUND) {
+		IWL_DEBUG_SCAN(mvm, "Pass all scheduled scan results found\n");
+		ieee80211_sched_scan_results(mvm->hw);
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_ENABLED;
+	}
 }
 
 static int iwl_mvm_umac_scan_abort(struct iwl_mvm *mvm, int type)
@@ -1528,6 +1549,7 @@ void iwl_mvm_report_scan_aborted(struct iwl_mvm *mvm)
 		uid = iwl_mvm_scan_uid_by_status(mvm, IWL_MVM_SCAN_SCHED);
 		if (uid >= 0 && !mvm->restart_fw) {
 			ieee80211_sched_scan_stopped(mvm->hw);
+			mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED;
 			mvm->scan_uid_status[uid] = 0;
 		}
 
@@ -1549,8 +1571,11 @@ void iwl_mvm_report_scan_aborted(struct iwl_mvm *mvm)
 		 * restart_hw, so do not report if FW is about to be
 		 * restarted.
 		 */
-		if ((mvm->scan_status & IWL_MVM_SCAN_SCHED) && !mvm->restart_fw)
+		if ((mvm->scan_status & IWL_MVM_SCAN_SCHED) &&
+		    !mvm->restart_fw) {
 			ieee80211_sched_scan_stopped(mvm->hw);
+			mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED;
+		}
 	}
 }
 
@@ -1586,6 +1611,7 @@ out:
 			ieee80211_scan_completed(mvm->hw, true);
 	} else if (notify) {
 		ieee80211_sched_scan_stopped(mvm->hw);
+		mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED;
 	}
 
 	return ret;
-- 
cgit v0.10.2


From 4494541c0c918688dfbbaf760548846a9c1bcf3f Mon Sep 17 00:00:00 2001
From: Eyal Shapira <eyal@wizery.com>
Date: Mon, 1 Feb 2016 09:07:05 +0200
Subject: iwlwifi: mvm: rs: fix a theoretical access to uninitialized array
 elements

Klocwork is unhappy as ht_vht_rates might be accessed with
rate->index being set to values between 0 and 3 which will
lead to accessing uninitialized array elements. Effectively this
doesn't happen as in HT/VHT we're not using these rate indices.
Still fix this.

Signed-off-by: Eyal Shapira <eyalx.shapira@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 6e7e78a..3dc94d2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -556,6 +556,7 @@ static char *rs_pretty_rate(const struct rs_rate *rate)
 	if (is_type_legacy(rate->type) && (rate->index <= IWL_RATE_54M_INDEX))
 		rate_str = legacy_rates[rate->index];
 	else if ((is_type_ht(rate->type) || is_type_vht(rate->type)) &&
+		 (rate->index >= IWL_RATE_MCS_0_INDEX) &&
 		 (rate->index <= IWL_RATE_MCS_9_INDEX))
 		rate_str = ht_vht_rates[rate->index];
 	else
-- 
cgit v0.10.2


From 5a51c03fa8f61b4c7554bad97426c87a93e4606b Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 23 Feb 2016 21:53:12 +0200
Subject: iwlwifi: mvm: bump firmware API to 21

The driver is now ready to handle this firmware.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-7000.c b/drivers/net/wireless/intel/iwlwifi/iwl-7000.c
index fa41a5e..fc475ce 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-7000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-7000.c
@@ -73,8 +73,8 @@
 /* Highest firmware API version supported */
 #define IWL7260_UCODE_API_MAX	17
 #define IWL7265_UCODE_API_MAX	17
-#define IWL7265D_UCODE_API_MAX	20
-#define IWL3168_UCODE_API_MAX	20
+#define IWL7265D_UCODE_API_MAX	21
+#define IWL3168_UCODE_API_MAX	21
 
 /* Oldest version we won't warn about */
 #define IWL7260_UCODE_API_OK	13
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
index bce9b3420..ad8c588 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
@@ -70,8 +70,8 @@
 #include "iwl-agn-hw.h"
 
 /* Highest firmware API version supported */
-#define IWL8000_UCODE_API_MAX	20
-#define IWL8265_UCODE_API_MAX	20
+#define IWL8000_UCODE_API_MAX	21
+#define IWL8265_UCODE_API_MAX	21
 
 /* Oldest version we won't warn about */
 #define IWL8000_UCODE_API_OK	13
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
index 4b93404..0835906 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
@@ -55,7 +55,7 @@
 #include "iwl-agn-hw.h"
 
 /* Highest firmware API version supported */
-#define IWL9000_UCODE_API_MAX	20
+#define IWL9000_UCODE_API_MAX	21
 
 /* Oldest version we won't warn about */
 #define IWL9000_UCODE_API_OK	13
-- 
cgit v0.10.2


From bac842da5b39f8f9521a8cb8d0b11e533424e90e Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 31 Jan 2016 09:29:39 +0200
Subject: iwlwifi: pcie: aggregate Flow Handler configuration writes

Instead of waking up the device each time we write a
register, wake it up once, and writes the registers
at once.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 58591ca..cda678d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -616,38 +616,38 @@ static int iwl_pcie_load_firmware_chunk(struct iwl_trans *trans, u32 dst_addr,
 				   dma_addr_t phy_addr, u32 byte_cnt)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	unsigned long flags;
 	int ret;
 
 	trans_pcie->ucode_write_complete = false;
 
-	iwl_write_direct32(trans,
-			   FH_TCSR_CHNL_TX_CONFIG_REG(FH_SRVC_CHNL),
-			   FH_TCSR_TX_CONFIG_REG_VAL_DMA_CHNL_PAUSE);
-
-	iwl_write_direct32(trans,
-			   FH_SRVC_CHNL_SRAM_ADDR_REG(FH_SRVC_CHNL),
-			   dst_addr);
-
-	iwl_write_direct32(trans,
-			   FH_TFDIB_CTRL0_REG(FH_SRVC_CHNL),
-			   phy_addr & FH_MEM_TFDIB_DRAM_ADDR_LSB_MSK);
-
-	iwl_write_direct32(trans,
-			   FH_TFDIB_CTRL1_REG(FH_SRVC_CHNL),
-			   (iwl_get_dma_hi_addr(phy_addr)
-				<< FH_MEM_TFDIB_REG1_ADDR_BITSHIFT) | byte_cnt);
-
-	iwl_write_direct32(trans,
-			   FH_TCSR_CHNL_TX_BUF_STS_REG(FH_SRVC_CHNL),
-			   1 << FH_TCSR_CHNL_TX_BUF_STS_REG_POS_TB_NUM |
-			   1 << FH_TCSR_CHNL_TX_BUF_STS_REG_POS_TB_IDX |
-			   FH_TCSR_CHNL_TX_BUF_STS_REG_VAL_TFDB_VALID);
-
-	iwl_write_direct32(trans,
-			   FH_TCSR_CHNL_TX_CONFIG_REG(FH_SRVC_CHNL),
-			   FH_TCSR_TX_CONFIG_REG_VAL_DMA_CHNL_ENABLE	|
-			   FH_TCSR_TX_CONFIG_REG_VAL_DMA_CREDIT_DISABLE	|
-			   FH_TCSR_TX_CONFIG_REG_VAL_CIRQ_HOST_ENDTFD);
+	if (!iwl_trans_grab_nic_access(trans, &flags))
+		return -EIO;
+
+	iwl_write32(trans, FH_TCSR_CHNL_TX_CONFIG_REG(FH_SRVC_CHNL),
+		    FH_TCSR_TX_CONFIG_REG_VAL_DMA_CHNL_PAUSE);
+
+	iwl_write32(trans, FH_SRVC_CHNL_SRAM_ADDR_REG(FH_SRVC_CHNL),
+		    dst_addr);
+
+	iwl_write32(trans, FH_TFDIB_CTRL0_REG(FH_SRVC_CHNL),
+		    phy_addr & FH_MEM_TFDIB_DRAM_ADDR_LSB_MSK);
+
+	iwl_write32(trans, FH_TFDIB_CTRL1_REG(FH_SRVC_CHNL),
+		    (iwl_get_dma_hi_addr(phy_addr)
+			<< FH_MEM_TFDIB_REG1_ADDR_BITSHIFT) | byte_cnt);
+
+	iwl_write32(trans, FH_TCSR_CHNL_TX_BUF_STS_REG(FH_SRVC_CHNL),
+		    BIT(FH_TCSR_CHNL_TX_BUF_STS_REG_POS_TB_NUM) |
+		    BIT(FH_TCSR_CHNL_TX_BUF_STS_REG_POS_TB_IDX) |
+		    FH_TCSR_CHNL_TX_BUF_STS_REG_VAL_TFDB_VALID);
+
+	iwl_write32(trans, FH_TCSR_CHNL_TX_CONFIG_REG(FH_SRVC_CHNL),
+		    FH_TCSR_TX_CONFIG_REG_VAL_DMA_CHNL_ENABLE |
+		    FH_TCSR_TX_CONFIG_REG_VAL_DMA_CREDIT_DISABLE |
+		    FH_TCSR_TX_CONFIG_REG_VAL_CIRQ_HOST_ENDTFD);
+
+	iwl_trans_release_nic_access(trans, &flags);
 
 	ret = wait_event_timeout(trans_pcie->ucode_write_waitq,
 				 trans_pcie->ucode_write_complete, 5 * HZ);
-- 
cgit v0.10.2


From 2e5d4a8f61dcae12996bbe2bcc446d3b47625cac Mon Sep 17 00:00:00 2001
From: Haim Dreyfuss <haim.dreyfuss@intel.com>
Date: Thu, 17 Dec 2015 12:17:58 +0200
Subject: iwlwifi: pcie: Add new configuration to enable MSIX

Working with MSIX requires prior configuration.
This includes requesting interrupt vectors from the OS,
registering the vectors and mapping the optional causes to the
relevant interrupt. In addition add new interrupt handler
to handle MSIX interrupt.

Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index 163b21b..a79c4f6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -549,4 +550,52 @@ enum dtd_diode_reg {
 	DTS_DIODE_REG_FLAGS_PASS_ONCE		= 0x00000080, /* bits [7:7] */
 };
 
+/*****************************************************************************
+ *                        MSIX related registers                             *
+ *****************************************************************************/
+
+#define CSR_MSIX_BASE			(0x2000)
+#define CSR_MSIX_FH_INT_CAUSES_AD	(CSR_MSIX_BASE + 0x800)
+#define CSR_MSIX_FH_INT_MASK_AD		(CSR_MSIX_BASE + 0x804)
+#define CSR_MSIX_HW_INT_CAUSES_AD	(CSR_MSIX_BASE + 0x808)
+#define CSR_MSIX_HW_INT_MASK_AD		(CSR_MSIX_BASE + 0x80C)
+#define CSR_MSIX_AUTOMASK_ST_AD		(CSR_MSIX_BASE + 0x810)
+#define CSR_MSIX_RX_IVAR_AD_REG		(CSR_MSIX_BASE + 0x880)
+#define CSR_MSIX_IVAR_AD_REG		(CSR_MSIX_BASE + 0x890)
+#define CSR_MSIX_PENDING_PBA_AD		(CSR_MSIX_BASE + 0x1000)
+#define CSR_MSIX_RX_IVAR(cause)		(CSR_MSIX_RX_IVAR_AD_REG + (cause))
+#define CSR_MSIX_IVAR(cause)		(CSR_MSIX_IVAR_AD_REG + (cause))
+
+#define MSIX_FH_INT_CAUSES_Q(q)		(q)
+
+/*
+ * Causes for the FH register interrupts
+ */
+enum msix_fh_int_causes {
+	MSIX_FH_INT_CAUSES_D2S_CH0_NUM		= BIT(16),
+	MSIX_FH_INT_CAUSES_D2S_CH1_NUM		= BIT(17),
+	MSIX_FH_INT_CAUSES_S2D			= BIT(19),
+	MSIX_FH_INT_CAUSES_FH_ERR		= BIT(21),
+};
+
+/*
+ * Causes for the HW register interrupts
+ */
+enum msix_hw_int_causes {
+	MSIX_HW_INT_CAUSES_REG_ALIVE		= BIT(0),
+	MSIX_HW_INT_CAUSES_REG_WAKEUP		= BIT(1),
+	MSIX_HW_INT_CAUSES_REG_CT_KILL		= BIT(6),
+	MSIX_HW_INT_CAUSES_REG_RF_KILL		= BIT(7),
+	MSIX_HW_INT_CAUSES_REG_PERIODIC		= BIT(8),
+	MSIX_HW_INT_CAUSES_REG_SW_ERR		= BIT(25),
+	MSIX_HW_INT_CAUSES_REG_SCD		= BIT(26),
+	MSIX_HW_INT_CAUSES_REG_FH_TX		= BIT(27),
+	MSIX_HW_INT_CAUSES_REG_HW_ERR		= BIT(29),
+	MSIX_HW_INT_CAUSES_REG_HAP		= BIT(30),
+};
+
+#define MSIX_MIN_INTERRUPT_VECTORS		2
+#define MSIX_AUTO_CLEAR_CAUSE			0
+#define MSIX_NON_AUTO_CLEAR_CAUSE		BIT(7)
+
 #endif /* !__iwl_csr_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
index 5bde23a..c46e596 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
@@ -404,4 +404,6 @@ enum {
 	LMPM_PAGE_PASS_NOTIF_POS = BIT(20),
 };
 
+#define UREG_CHICK		(0xA05C00)
+#define UREG_CHICK_MSIX_ENABLE	BIT(25)
 #endif				/* __iwl_prph_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 542bbc5..6677f31 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -336,6 +336,14 @@ struct iwl_tso_hdr_page {
  * @fw_mon_phys: physical address of the buffer for the firmware monitor
  * @fw_mon_page: points to the first page of the buffer for the firmware monitor
  * @fw_mon_size: size of the buffer for the firmware monitor
+ * @msix_entries: array of MSI-X entries
+ * @msix_enabled: true if managed to enable MSI-X
+ * @allocated_vector: the number of interrupt vector allocated by the OS
+ * @default_irq_num: default irq for non rx interrupt
+ * @fh_init_mask: initial unmasked fh causes
+ * @hw_init_mask: initial unmasked hw causes
+ * @fh_mask: current unmasked fh causes
+ * @hw_mask: current unmasked hw causes
  */
 struct iwl_trans_pcie {
 	struct iwl_rxq *rxq;
@@ -402,6 +410,15 @@ struct iwl_trans_pcie {
 	dma_addr_t fw_mon_phys;
 	struct page *fw_mon_page;
 	u32 fw_mon_size;
+
+	struct msix_entry msix_entries[IWL_MAX_RX_HW_QUEUES];
+	bool msix_enabled;
+	u32 allocated_vector;
+	u32 default_irq_num;
+	u32 fh_init_mask;
+	u32 hw_init_mask;
+	u32 fh_mask;
+	u32 hw_mask;
 };
 
 static inline struct iwl_trans_pcie *
@@ -430,7 +447,10 @@ void iwl_trans_pcie_free(struct iwl_trans *trans);
 * RX
 ******************************************************/
 int iwl_pcie_rx_init(struct iwl_trans *trans);
+irqreturn_t iwl_pcie_msix_isr(int irq, void *data);
 irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id);
+irqreturn_t iwl_pcie_irq_msix_handler(int irq, void *dev_id);
+irqreturn_t iwl_pcie_irq_rx_msix_handler(int irq, void *dev_id);
 int iwl_pcie_rx_stop(struct iwl_trans *trans);
 void iwl_pcie_rx_free(struct iwl_trans *trans);
 
@@ -485,15 +505,24 @@ void iwl_pcie_dump_csr(struct iwl_trans *trans);
 ******************************************************/
 static inline void iwl_disable_interrupts(struct iwl_trans *trans)
 {
-	clear_bit(STATUS_INT_ENABLED, &trans->status);
-
-	/* disable interrupts from uCode/NIC to host */
-	iwl_write32(trans, CSR_INT_MASK, 0x00000000);
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
-	/* acknowledge/clear/reset any interrupts still pending
-	 * from uCode or flow handler (Rx/Tx DMA) */
-	iwl_write32(trans, CSR_INT, 0xffffffff);
-	iwl_write32(trans, CSR_FH_INT_STATUS, 0xffffffff);
+	clear_bit(STATUS_INT_ENABLED, &trans->status);
+	if (!trans_pcie->msix_enabled) {
+		/* disable interrupts from uCode/NIC to host */
+		iwl_write32(trans, CSR_INT_MASK, 0x00000000);
+
+		/* acknowledge/clear/reset any interrupts still pending
+		 * from uCode or flow handler (Rx/Tx DMA) */
+		iwl_write32(trans, CSR_INT, 0xffffffff);
+		iwl_write32(trans, CSR_FH_INT_STATUS, 0xffffffff);
+	} else {
+		/* disable all the interrupt we might use */
+		iwl_write32(trans, CSR_MSIX_FH_INT_MASK_AD,
+			    trans_pcie->fh_init_mask);
+		iwl_write32(trans, CSR_MSIX_HW_INT_MASK_AD,
+			    trans_pcie->hw_init_mask);
+	}
 	IWL_DEBUG_ISR(trans, "Disabled interrupts\n");
 }
 
@@ -503,8 +532,37 @@ static inline void iwl_enable_interrupts(struct iwl_trans *trans)
 
 	IWL_DEBUG_ISR(trans, "Enabling interrupts\n");
 	set_bit(STATUS_INT_ENABLED, &trans->status);
-	trans_pcie->inta_mask = CSR_INI_SET_MASK;
-	iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	if (!trans_pcie->msix_enabled) {
+		trans_pcie->inta_mask = CSR_INI_SET_MASK;
+		iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	} else {
+		/*
+		 * fh/hw_mask keeps all the unmasked causes.
+		 * Unlike msi, in msix cause is enabled when it is unset.
+		 */
+		trans_pcie->hw_mask = trans_pcie->hw_init_mask;
+		trans_pcie->fh_mask = trans_pcie->fh_init_mask;
+		iwl_write32(trans, CSR_MSIX_FH_INT_MASK_AD,
+			    ~trans_pcie->fh_mask);
+		iwl_write32(trans, CSR_MSIX_HW_INT_MASK_AD,
+			    ~trans_pcie->hw_mask);
+	}
+}
+
+static inline void iwl_enable_hw_int_msk_msix(struct iwl_trans *trans, u32 msk)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	iwl_write32(trans, CSR_MSIX_HW_INT_MASK_AD, ~msk);
+	trans_pcie->hw_mask = msk;
+}
+
+static inline void iwl_enable_fh_int_msk_msix(struct iwl_trans *trans, u32 msk)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	iwl_write32(trans, CSR_MSIX_FH_INT_MASK_AD, ~msk);
+	trans_pcie->fh_mask = msk;
 }
 
 static inline void iwl_enable_fw_load_int(struct iwl_trans *trans)
@@ -512,8 +570,15 @@ static inline void iwl_enable_fw_load_int(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
 	IWL_DEBUG_ISR(trans, "Enabling FW load interrupt\n");
-	trans_pcie->inta_mask = CSR_INT_BIT_FH_TX;
-	iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	if (!trans_pcie->msix_enabled) {
+		trans_pcie->inta_mask = CSR_INT_BIT_FH_TX;
+		iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	} else {
+		iwl_write32(trans, CSR_MSIX_HW_INT_MASK_AD,
+			    trans_pcie->hw_init_mask);
+		iwl_enable_fh_int_msk_msix(trans,
+					   MSIX_FH_INT_CAUSES_D2S_CH0_NUM);
+	}
 }
 
 static inline void iwl_enable_rfkill_int(struct iwl_trans *trans)
@@ -521,8 +586,15 @@ static inline void iwl_enable_rfkill_int(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
 	IWL_DEBUG_ISR(trans, "Enabling rfkill interrupt\n");
-	trans_pcie->inta_mask = CSR_INT_BIT_RF_KILL;
-	iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	if (!trans_pcie->msix_enabled) {
+		trans_pcie->inta_mask = CSR_INT_BIT_RF_KILL;
+		iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	} else {
+		iwl_write32(trans, CSR_MSIX_FH_INT_MASK_AD,
+			    trans_pcie->fh_init_mask);
+		iwl_enable_hw_int_msk_msix(trans,
+					   MSIX_HW_INT_CAUSES_REG_RF_KILL);
+	}
 }
 
 static inline void iwl_wake_queue(struct iwl_trans *trans,
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 07973ef..c0db681 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -1135,10 +1135,10 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 /*
  * iwl_pcie_rx_handle - Main entry function for receiving responses from fw
  */
-static void iwl_pcie_rx_handle(struct iwl_trans *trans)
+static void iwl_pcie_rx_handle(struct iwl_trans *trans, int queue)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_rxq *rxq = &trans_pcie->rxq[0];
+	struct iwl_rxq *rxq = &trans_pcie->rxq[queue];
 	u32 r, i, j, count = 0;
 	bool emergency = false;
 
@@ -1259,6 +1259,51 @@ restart:
 		napi_gro_flush(&rxq->napi, false);
 }
 
+static struct iwl_trans_pcie *iwl_pcie_get_trans_pcie(struct msix_entry *entry)
+{
+	u8 queue = entry->entry;
+	struct msix_entry *entries = entry - queue;
+
+	return container_of(entries, struct iwl_trans_pcie, msix_entries[0]);
+}
+
+static inline void iwl_pcie_clear_irq(struct iwl_trans *trans,
+				      struct msix_entry *entry)
+{
+	/*
+	 * Before sending the interrupt the HW disables it to prevent
+	 * a nested interrupt. This is done by writing 1 to the corresponding
+	 * bit in the mask register. After handling the interrupt, it should be
+	 * re-enabled by clearing this bit. This register is defined as
+	 * write 1 clear (W1C) register, meaning that it's being clear
+	 * by writing 1 to the bit.
+	 */
+	iwl_write_direct32(trans, CSR_MSIX_AUTOMASK_ST_AD, BIT(entry->entry));
+}
+
+/*
+ * iwl_pcie_rx_msix_handle - Main entry function for receiving responses from fw
+ * This interrupt handler should be used with RSS queue only.
+ */
+irqreturn_t iwl_pcie_irq_rx_msix_handler(int irq, void *dev_id)
+{
+	struct msix_entry *entry = dev_id;
+	struct iwl_trans_pcie *trans_pcie = iwl_pcie_get_trans_pcie(entry);
+	struct iwl_trans *trans = trans_pcie->trans;
+
+	lock_map_acquire(&trans->sync_cmd_lockdep_map);
+
+	local_bh_disable();
+	iwl_pcie_rx_handle(trans, entry->entry);
+	local_bh_enable();
+
+	iwl_pcie_clear_irq(trans, entry);
+
+	lock_map_release(&trans->sync_cmd_lockdep_map);
+
+	return IRQ_HANDLED;
+}
+
 /*
  * iwl_pcie_irq_handle_error - called for HW or SW error interrupt from card
  */
@@ -1589,7 +1634,7 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id)
 		isr_stats->rx++;
 
 		local_bh_disable();
-		iwl_pcie_rx_handle(trans);
+		iwl_pcie_rx_handle(trans, 0);
 		local_bh_enable();
 	}
 
@@ -1732,3 +1777,129 @@ irqreturn_t iwl_pcie_isr(int irq, void *data)
 
 	return IRQ_WAKE_THREAD;
 }
+
+irqreturn_t iwl_pcie_msix_isr(int irq, void *data)
+{
+	return IRQ_WAKE_THREAD;
+}
+
+irqreturn_t iwl_pcie_irq_msix_handler(int irq, void *dev_id)
+{
+	struct msix_entry *entry = dev_id;
+	struct iwl_trans_pcie *trans_pcie = iwl_pcie_get_trans_pcie(entry);
+	struct iwl_trans *trans = trans_pcie->trans;
+	struct isr_statistics *isr_stats = isr_stats = &trans_pcie->isr_stats;
+	u32 inta_fh, inta_hw;
+
+	lock_map_acquire(&trans->sync_cmd_lockdep_map);
+
+	spin_lock(&trans_pcie->irq_lock);
+	inta_fh = iwl_read_direct32(trans, CSR_MSIX_FH_INT_CAUSES_AD);
+	inta_hw = iwl_read_direct32(trans, CSR_MSIX_HW_INT_CAUSES_AD);
+	/*
+	 * Clear causes registers to avoid being handling the same cause.
+	 */
+	iwl_write_direct32(trans, CSR_MSIX_FH_INT_CAUSES_AD, inta_fh);
+	iwl_write_direct32(trans, CSR_MSIX_HW_INT_CAUSES_AD, inta_hw);
+	spin_unlock(&trans_pcie->irq_lock);
+
+	if (unlikely(!(inta_fh | inta_hw))) {
+		IWL_DEBUG_ISR(trans, "Ignore interrupt, inta == 0\n");
+		lock_map_release(&trans->sync_cmd_lockdep_map);
+		return IRQ_NONE;
+	}
+
+	if (iwl_have_debug_level(IWL_DL_ISR))
+		IWL_DEBUG_ISR(trans, "ISR inta_fh 0x%08x, enabled 0x%08x\n",
+			      inta_fh,
+			      iwl_read32(trans, CSR_MSIX_FH_INT_MASK_AD));
+
+	/* This "Tx" DMA channel is used only for loading uCode */
+	if (inta_fh & MSIX_FH_INT_CAUSES_D2S_CH0_NUM) {
+		IWL_DEBUG_ISR(trans, "uCode load interrupt\n");
+		isr_stats->tx++;
+		/*
+		 * Wake up uCode load routine,
+		 * now that load is complete
+		 */
+		trans_pcie->ucode_write_complete = true;
+		wake_up(&trans_pcie->ucode_write_waitq);
+	}
+
+	/* Error detected by uCode */
+	if ((inta_fh & MSIX_FH_INT_CAUSES_FH_ERR) ||
+	    (inta_hw & MSIX_HW_INT_CAUSES_REG_SW_ERR)) {
+		IWL_ERR(trans,
+			"Microcode SW error detected. Restarting 0x%X.\n",
+			inta_fh);
+		isr_stats->sw++;
+		iwl_pcie_irq_handle_error(trans);
+	}
+
+	/* After checking FH register check HW register */
+	if (iwl_have_debug_level(IWL_DL_ISR))
+		IWL_DEBUG_ISR(trans,
+			      "ISR inta_hw 0x%08x, enabled 0x%08x\n",
+			      inta_hw,
+			      iwl_read32(trans, CSR_MSIX_HW_INT_MASK_AD));
+
+	/* Alive notification via Rx interrupt will do the real work */
+	if (inta_hw & MSIX_HW_INT_CAUSES_REG_ALIVE) {
+		IWL_DEBUG_ISR(trans, "Alive interrupt\n");
+		isr_stats->alive++;
+	}
+
+	/* uCode wakes up after power-down sleep */
+	if (inta_hw & MSIX_HW_INT_CAUSES_REG_WAKEUP) {
+		IWL_DEBUG_ISR(trans, "Wakeup interrupt\n");
+		iwl_pcie_rxq_check_wrptr(trans);
+		iwl_pcie_txq_check_wrptrs(trans);
+
+		isr_stats->wakeup++;
+	}
+
+	/* Chip got too hot and stopped itself */
+	if (inta_hw & MSIX_HW_INT_CAUSES_REG_CT_KILL) {
+		IWL_ERR(trans, "Microcode CT kill error detected.\n");
+		isr_stats->ctkill++;
+	}
+
+	/* HW RF KILL switch toggled */
+	if (inta_hw & MSIX_HW_INT_CAUSES_REG_RF_KILL) {
+		bool hw_rfkill;
+
+		hw_rfkill = iwl_is_rfkill_set(trans);
+		IWL_WARN(trans, "RF_KILL bit toggled to %s.\n",
+			 hw_rfkill ? "disable radio" : "enable radio");
+
+		isr_stats->rfkill++;
+
+		mutex_lock(&trans_pcie->mutex);
+		iwl_trans_pcie_rf_kill(trans, hw_rfkill);
+		mutex_unlock(&trans_pcie->mutex);
+		if (hw_rfkill) {
+			set_bit(STATUS_RFKILL, &trans->status);
+			if (test_and_clear_bit(STATUS_SYNC_HCMD_ACTIVE,
+					       &trans->status))
+				IWL_DEBUG_RF_KILL(trans,
+						  "Rfkill while SYNC HCMD in flight\n");
+			wake_up(&trans_pcie->wait_command_queue);
+		} else {
+			clear_bit(STATUS_RFKILL, &trans->status);
+		}
+	}
+
+	if (inta_hw & MSIX_HW_INT_CAUSES_REG_HW_ERR) {
+		IWL_ERR(trans,
+			"Hardware error detected. Restarting.\n");
+
+		isr_stats->hw++;
+		iwl_pcie_irq_handle_error(trans);
+	}
+
+	iwl_pcie_clear_irq(trans, entry);
+
+	lock_map_release(&trans->sync_cmd_lockdep_map);
+
+	return IRQ_HANDLED;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index cda678d..102d175 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1123,6 +1123,20 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 	iwl_pcie_prepare_card_hw(trans);
 }
 
+static void iwl_pcie_synchronize_irqs(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	if (trans_pcie->msix_enabled) {
+		int i;
+
+		for (i = 0; i < trans_pcie->allocated_vector; i++)
+			synchronize_irq(trans_pcie->msix_entries[i].vector);
+	} else {
+		synchronize_irq(trans_pcie->pci_dev->irq);
+	}
+}
+
 static int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
 				   const struct fw_img *fw, bool run_in_rfkill)
 {
@@ -1149,7 +1163,7 @@ static int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
 	iwl_disable_interrupts(trans);
 
 	/* Make sure it finished running */
-	synchronize_irq(trans_pcie->pci_dev->irq);
+	iwl_pcie_synchronize_irqs(trans);
 
 	mutex_lock(&trans_pcie->mutex);
 
@@ -1252,8 +1266,6 @@ void iwl_trans_pcie_rf_kill(struct iwl_trans *trans, bool state)
 static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test,
 				      bool reset)
 {
-	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-
 	if (!reset) {
 		/* Enable persistence mode to avoid reset */
 		iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
@@ -1271,7 +1283,7 @@ static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test,
 
 	iwl_pcie_disable_ict(trans);
 
-	synchronize_irq(trans_pcie->pci_dev->irq);
+	iwl_pcie_synchronize_irqs(trans);
 
 	iwl_clear_bit(trans, CSR_GP_CNTRL,
 		      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
@@ -1350,6 +1362,153 @@ static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
 	return 0;
 }
 
+struct iwl_causes_list {
+	u32 cause_num;
+	u32 mask_reg;
+	u8 addr;
+};
+
+static struct iwl_causes_list causes_list[] = {
+	{MSIX_FH_INT_CAUSES_D2S_CH0_NUM,	CSR_MSIX_FH_INT_MASK_AD, 0},
+	{MSIX_FH_INT_CAUSES_D2S_CH1_NUM,	CSR_MSIX_FH_INT_MASK_AD, 0x1},
+	{MSIX_FH_INT_CAUSES_S2D,		CSR_MSIX_FH_INT_MASK_AD, 0x3},
+	{MSIX_FH_INT_CAUSES_FH_ERR,		CSR_MSIX_FH_INT_MASK_AD, 0x5},
+	{MSIX_HW_INT_CAUSES_REG_ALIVE,		CSR_MSIX_HW_INT_MASK_AD, 0x10},
+	{MSIX_HW_INT_CAUSES_REG_WAKEUP,		CSR_MSIX_HW_INT_MASK_AD, 0x11},
+	{MSIX_HW_INT_CAUSES_REG_CT_KILL,	CSR_MSIX_HW_INT_MASK_AD, 0x16},
+	{MSIX_HW_INT_CAUSES_REG_RF_KILL,	CSR_MSIX_HW_INT_MASK_AD, 0x17},
+	{MSIX_HW_INT_CAUSES_REG_PERIODIC,	CSR_MSIX_HW_INT_MASK_AD, 0x18},
+	{MSIX_HW_INT_CAUSES_REG_SW_ERR,		CSR_MSIX_HW_INT_MASK_AD, 0x29},
+	{MSIX_HW_INT_CAUSES_REG_SCD,		CSR_MSIX_HW_INT_MASK_AD, 0x2A},
+	{MSIX_HW_INT_CAUSES_REG_FH_TX,		CSR_MSIX_HW_INT_MASK_AD, 0x2B},
+	{MSIX_HW_INT_CAUSES_REG_HW_ERR,		CSR_MSIX_HW_INT_MASK_AD, 0x2D},
+	{MSIX_HW_INT_CAUSES_REG_HAP,		CSR_MSIX_HW_INT_MASK_AD, 0x2E},
+};
+
+static void iwl_pcie_init_msix(struct iwl_trans_pcie *trans_pcie)
+{
+	u32 val, max_rx_vector, i;
+	struct iwl_trans *trans = trans_pcie->trans;
+
+	max_rx_vector = trans_pcie->allocated_vector - 1;
+
+	if (!trans_pcie->msix_enabled)
+		return;
+
+	iwl_write_prph(trans, UREG_CHICK, UREG_CHICK_MSIX_ENABLE);
+
+	/*
+	 * Each cause from the list above and the RX causes is represented as
+	 * a byte in the IVAR table. We access the first (N - 1) bytes and map
+	 * them to the (N - 1) vectors so these vectors will be used as rx
+	 * vectors. Then access all non rx causes and map them to the
+	 * default queue (N'th queue).
+	 */
+	for (i = 0; i < max_rx_vector; i++) {
+		iwl_write8(trans, CSR_MSIX_RX_IVAR(i), MSIX_FH_INT_CAUSES_Q(i));
+		iwl_clear_bit(trans, CSR_MSIX_FH_INT_MASK_AD,
+			      BIT(MSIX_FH_INT_CAUSES_Q(i)));
+	}
+
+	for (i = 0; i < ARRAY_SIZE(causes_list); i++) {
+		val = trans_pcie->default_irq_num |
+			MSIX_NON_AUTO_CLEAR_CAUSE;
+		iwl_write8(trans, CSR_MSIX_IVAR(causes_list[i].addr), val);
+		iwl_clear_bit(trans, causes_list[i].mask_reg,
+			      causes_list[i].cause_num);
+	}
+	trans_pcie->fh_init_mask =
+		~iwl_read32(trans, CSR_MSIX_FH_INT_MASK_AD);
+	trans_pcie->fh_mask = trans_pcie->fh_init_mask;
+	trans_pcie->hw_init_mask =
+		~iwl_read32(trans, CSR_MSIX_HW_INT_MASK_AD);
+	trans_pcie->hw_mask = trans_pcie->hw_init_mask;
+}
+
+static void iwl_pcie_set_interrupt_capa(struct pci_dev *pdev,
+					struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	u16 pci_cmd;
+	int max_vector;
+	int ret, i;
+
+	if (trans->cfg->mq_rx_supported) {
+		max_vector = min_t(u32, (num_possible_cpus() + 1),
+				   IWL_MAX_RX_HW_QUEUES);
+		for (i = 0; i < max_vector; i++)
+			trans_pcie->msix_entries[i].entry = i;
+
+		ret = pci_enable_msix_range(pdev, trans_pcie->msix_entries,
+					    MSIX_MIN_INTERRUPT_VECTORS,
+					    max_vector);
+		if (ret > 1) {
+			IWL_DEBUG_INFO(trans,
+				       "Enable MSI-X allocate %d interrupt vector\n",
+				       ret);
+			trans_pcie->allocated_vector = ret;
+			trans_pcie->default_irq_num =
+				trans_pcie->allocated_vector - 1;
+			trans_pcie->trans->num_rx_queues =
+				trans_pcie->allocated_vector - 1;
+			trans_pcie->msix_enabled = true;
+
+			return;
+		}
+		IWL_DEBUG_INFO(trans,
+			       "ret = %d %s move to msi mode\n", ret,
+			       (ret == 1) ?
+			       "can't allocate more than 1 interrupt vector" :
+			       "failed to enable msi-x mode");
+		pci_disable_msix(pdev);
+	}
+
+	ret = pci_enable_msi(pdev);
+	if (ret) {
+		dev_err(&pdev->dev, "pci_enable_msi failed(0X%x)\n", ret);
+		/* enable rfkill interrupt: hw bug w/a */
+		pci_read_config_word(pdev, PCI_COMMAND, &pci_cmd);
+		if (pci_cmd & PCI_COMMAND_INTX_DISABLE) {
+			pci_cmd &= ~PCI_COMMAND_INTX_DISABLE;
+			pci_write_config_word(pdev, PCI_COMMAND, pci_cmd);
+		}
+	}
+}
+
+static int iwl_pcie_init_msix_handler(struct pci_dev *pdev,
+				      struct iwl_trans_pcie *trans_pcie)
+{
+	int i, last_vector;
+
+	last_vector = trans_pcie->trans->num_rx_queues;
+
+	for (i = 0; i < trans_pcie->allocated_vector; i++) {
+		int ret;
+
+		ret = request_threaded_irq(trans_pcie->msix_entries[i].vector,
+					   iwl_pcie_msix_isr,
+					   (i == last_vector) ?
+					   iwl_pcie_irq_msix_handler :
+					   iwl_pcie_irq_rx_msix_handler,
+					   IRQF_SHARED,
+					   DRV_NAME,
+					   &trans_pcie->msix_entries[i]);
+		if (ret) {
+			int j;
+
+			IWL_ERR(trans_pcie->trans,
+				"Error allocating IRQ %d\n", i);
+			for (j = 0; j < i; j++)
+				free_irq(trans_pcie->msix_entries[i].vector,
+					 &trans_pcie->msix_entries[i]);
+			pci_disable_msix(pdev);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
 static int _iwl_trans_pcie_start_hw(struct iwl_trans *trans, bool low_power)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
@@ -1371,6 +1530,7 @@ static int _iwl_trans_pcie_start_hw(struct iwl_trans *trans, bool low_power)
 
 	iwl_pcie_apm_init(trans);
 
+	iwl_pcie_init_msix(trans_pcie);
 	/* From now on, the op_mode will be kept updated about RF kill state */
 	iwl_enable_rfkill_int(trans);
 
@@ -1425,7 +1585,7 @@ static void iwl_trans_pcie_op_mode_leave(struct iwl_trans *trans)
 
 	mutex_unlock(&trans_pcie->mutex);
 
-	synchronize_irq(trans_pcie->pci_dev->irq);
+	iwl_pcie_synchronize_irqs(trans);
 }
 
 static void iwl_trans_pcie_write8(struct iwl_trans *trans, u32 ofs, u8 val)
@@ -1506,15 +1666,25 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	/* TODO: check if this is really needed */
 	pm_runtime_disable(trans->dev);
 
-	synchronize_irq(trans_pcie->pci_dev->irq);
+	iwl_pcie_synchronize_irqs(trans);
 
 	iwl_pcie_tx_free(trans);
 	iwl_pcie_rx_free(trans);
 
-	free_irq(trans_pcie->pci_dev->irq, trans);
-	iwl_pcie_free_ict(trans);
+	if (trans_pcie->msix_enabled) {
+		for (i = 0; i < trans_pcie->allocated_vector; i++)
+			free_irq(trans_pcie->msix_entries[i].vector,
+				 &trans_pcie->msix_entries[i]);
+
+		pci_disable_msix(trans_pcie->pci_dev);
+		trans_pcie->msix_enabled = false;
+	} else {
+		free_irq(trans_pcie->pci_dev->irq, trans);
 
-	pci_disable_msi(trans_pcie->pci_dev);
+		iwl_pcie_free_ict(trans);
+
+		pci_disable_msi(trans_pcie->pci_dev);
+	}
 	iounmap(trans_pcie->hw_base);
 	pci_release_regions(trans_pcie->pci_dev);
 	pci_disable_device(trans_pcie->pci_dev);
@@ -2615,7 +2785,6 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 {
 	struct iwl_trans_pcie *trans_pcie;
 	struct iwl_trans *trans;
-	u16 pci_cmd;
 	int ret, addr_size;
 
 	trans = iwl_trans_alloc(sizeof(struct iwl_trans_pcie),
@@ -2698,17 +2867,6 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 	trans_pcie->pci_dev = pdev;
 	iwl_disable_interrupts(trans);
 
-	ret = pci_enable_msi(pdev);
-	if (ret) {
-		dev_err(&pdev->dev, "pci_enable_msi failed(0X%x)\n", ret);
-		/* enable rfkill interrupt: hw bug w/a */
-		pci_read_config_word(pdev, PCI_COMMAND, &pci_cmd);
-		if (pci_cmd & PCI_COMMAND_INTX_DISABLE) {
-			pci_cmd &= ~PCI_COMMAND_INTX_DISABLE;
-			pci_write_config_word(pdev, PCI_COMMAND, pci_cmd);
-		}
-	}
-
 	trans->hw_rev = iwl_read32(trans, CSR_HW_REV);
 	/*
 	 * In the 8000 HW family the format of the 4 bytes of CSR_HW_REV have
@@ -2760,6 +2918,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 		}
 	}
 
+	iwl_pcie_set_interrupt_capa(pdev, trans);
 	trans->hw_id = (pdev->device << 16) + pdev->subsystem_device;
 	snprintf(trans->hw_id_str, sizeof(trans->hw_id_str),
 		 "PCI ID: 0x%04X:0x%04X", pdev->device, pdev->subsystem_device);
@@ -2769,19 +2928,23 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 
 	init_waitqueue_head(&trans_pcie->d0i3_waitq);
 
-	ret = iwl_pcie_alloc_ict(trans);
-	if (ret)
-		goto out_pci_disable_msi;
-
-	ret = request_threaded_irq(pdev->irq, iwl_pcie_isr,
-				   iwl_pcie_irq_handler,
-				   IRQF_SHARED, DRV_NAME, trans);
-	if (ret) {
-		IWL_ERR(trans, "Error allocating IRQ %d\n", pdev->irq);
-		goto out_free_ict;
-	}
+	if (trans_pcie->msix_enabled) {
+		if (iwl_pcie_init_msix_handler(pdev, trans_pcie))
+			goto out_pci_release_regions;
+	 } else {
+		ret = iwl_pcie_alloc_ict(trans);
+		if (ret)
+			goto out_pci_disable_msi;
 
-	trans_pcie->inta_mask = CSR_INI_SET_MASK;
+		ret = request_threaded_irq(pdev->irq, iwl_pcie_isr,
+					   iwl_pcie_irq_handler,
+					   IRQF_SHARED, DRV_NAME, trans);
+		if (ret) {
+			IWL_ERR(trans, "Error allocating IRQ %d\n", pdev->irq);
+			goto out_free_ict;
+		}
+		trans_pcie->inta_mask = CSR_INI_SET_MASK;
+	 }
 
 #ifdef CONFIG_IWLWIFI_PCIE_RTPM
 	trans->runtime_pm_mode = IWL_PLAT_PM_MODE_D0I3;
-- 
cgit v0.10.2


From 60c0a88f2ded5947bdfebbd6d5e9b3925b90984e Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 7 Feb 2016 10:28:13 +0200
Subject: iwlwifi: pcie: fix identation in trans.c

A curly brace was misplaced, fix this.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 102d175..e67957d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -2239,7 +2239,7 @@ static ssize_t iwl_dbgfs_rx_queue_read(struct file *file,
 		} else {
 			pos += scnprintf(buf + pos, bufsz - pos,
 					 "\tclosed_rb_num: Not Allocated\n");
-	}
+		}
 	}
 	ret = simple_read_from_buffer(user_buf, count, ppos, buf, pos);
 	kfree(buf);
-- 
cgit v0.10.2


From e48c947f030d1f5c46c684a6cc2db332bc4af33f Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 27 Jan 2016 18:59:48 +0200
Subject: iwlwifi: mvm: enable VHT MU-MIMO for supported hardware

Incoming hardware will support VHT MU-MIMO. Declare this
capability for relevant hardware.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
index ad8c588..97be104 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
@@ -217,6 +217,7 @@ const struct iwl_cfg iwl8265_2ac_cfg = {
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
 	.max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K,
+	.vht_mu_mimo_supported = true,
 };
 
 const struct iwl_cfg iwl4165_2ac_cfg = {
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
index 0835906..8e32a57 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
@@ -139,7 +139,8 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 	.smem_len = IWL9000_SMEM_LEN,					\
 	.thermal_params = &iwl9000_tt_params,				\
 	.apmg_not_supported = true,					\
-	.mq_rx_supported = true
+	.mq_rx_supported = true,					\
+	.vht_mu_mimo_supported = true
 
 const struct iwl_cfg iwl9260_2ac_cfg = {
 		.name = "Intel(R) Dual Band Wireless AC 9260",
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index dad5570..4f2b57e 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -312,6 +312,7 @@ struct iwl_pwr_tx_backoff {
  * @smem_offset: offset from which the SMEM begins
  * @smem_len: the length of SMEM
  * @mq_rx_supported: multi-queue rx support
+ * @vht_mu_mimo_supported: VHT MU-MIMO support
  *
  * We enable the driver to be backward compatible wrt. hardware features.
  * API differences in uCode shouldn't be handled here but through TLVs
@@ -364,6 +365,7 @@ struct iwl_cfg {
 	const struct iwl_tt_params *thermal_params;
 	bool apmg_not_supported;
 	bool mq_rx_supported;
+	bool vht_mu_mimo_supported;
 };
 
 /*
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 50f4cc6..e84cb8d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -366,6 +366,9 @@ static void iwl_init_vht_hw_capab(const struct iwl_cfg *cfg,
 		       max_ampdu_exponent <<
 		       IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT;
 
+	if (cfg->vht_mu_mimo_supported)
+		vht_cap->cap |= IEEE80211_VHT_CAP_MU_BEAMFORMEE_CAPABLE;
+
 	if (cfg->ht_params->ldpc)
 		vht_cap->cap |= IEEE80211_VHT_CAP_RXLDPC;
 
-- 
cgit v0.10.2


From a07a8f37023acbd7a1c8453ffe4e35a7fb173c28 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Sun, 15 Nov 2015 11:11:59 +0200
Subject: iwlwifi: mvm: update firmware of VHT MU-MIMO groups status on restart

The firmware handles the VHT MU-MIMO group data on its own.
However, on HW restart (and future sniffer mode) the driver
shall update the firmware on the VHT MU-MIMO group membership
status.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index d5f9037..ca7fec71 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -285,6 +285,10 @@ enum iwl_phy_ops_subcmd_ids {
 	DTS_MEASUREMENT_NOTIF_WIDE = 0xFF,
 };
 
+enum iwl_data_path_subcmd_ids {
+	UPDATE_MU_GROUPS_CMD = 0x1,
+};
+
 enum iwl_prot_offload_subcmd_ids {
 	STORED_BEACON_NTF = 0xFF,
 };
@@ -294,6 +298,7 @@ enum {
 	LEGACY_GROUP = 0x0,
 	LONG_GROUP = 0x1,
 	PHY_OPS_GROUP = 0x4,
+	DATA_PATH_GROUP = 0x5,
 	PROT_OFFLOAD_GROUP = 0xb,
 };
 
@@ -1923,6 +1928,19 @@ struct iwl_shared_mem_cfg {
 	__le32 page_buff_size;
 } __packed; /* SHARED_MEM_ALLOC_API_S_VER_1 */
 
+/**
+ * VHT MU-MIMO group configuration
+ *
+ * @membership_status: a bitmap of MU groups
+ * @user_position:the position of station in a group. If the station is in the
+ *	group then bits (group * 2) is the position -1
+ */
+struct iwl_mu_group_mgmt_cmd {
+	__le32 reserved;
+	__le32 membership_status[2];
+	__le32 user_position[4];
+} __packed; /* MU_GROUP_ID_MNG_TABLE_API_S_VER_1 */
+
 #define MAX_STORED_BEACON_SIZE 600
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 5315681..f1ed90b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1762,6 +1762,22 @@ static inline int iwl_mvm_configure_bcast_filter(struct iwl_mvm *mvm)
 }
 #endif
 
+static int iwl_mvm_update_mu_groups(struct iwl_mvm *mvm,
+				    struct ieee80211_vif *vif)
+{
+	struct iwl_mu_group_mgmt_cmd cmd = {};
+
+	memcpy(cmd.membership_status, vif->bss_conf.mu_group.membership,
+	       WLAN_MEMBERSHIP_LEN);
+	memcpy(cmd.user_position, vif->bss_conf.mu_group.position,
+	       WLAN_USER_POSITION_LEN);
+
+	return iwl_mvm_send_cmd_pdu(mvm,
+				    WIDE_ID(DATA_PATH_GROUP,
+					    UPDATE_MU_GROUPS_CMD),
+				    0, sizeof(cmd), &cmd);
+}
+
 static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 					     struct ieee80211_vif *vif,
 					     struct ieee80211_bss_conf *bss_conf,
@@ -1870,6 +1886,18 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 					vif->addr);
 		}
 
+		/*
+		 * The firmware tracks the MU-MIMO group on its own.
+		 * However, on HW restart we should restore this data
+		 */
+		if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
+		    changes & BSS_CHANGED_MU_GROUPS) {
+			ret = iwl_mvm_update_mu_groups(mvm, vif);
+			if (ret)
+				IWL_ERR(mvm,
+					"failed to update VHT MU_MIMO groups\n");
+		}
+
 		iwl_mvm_recalc_multicast(mvm);
 		iwl_mvm_configure_bcast_filter(mvm);
 
-- 
cgit v0.10.2


From a6d5e32f247cbd3e34c7f86effbf4b426a018c32 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Wed, 14 Oct 2015 16:28:52 +0300
Subject: iwlwifi: mvm: send large SKBs to the transport

Now that PCIe knows how to create A-MSDUs, use this
capability and prepare SKBs that are large enough to
build an A-MSDU.
Advertise TSO support towards the network stack and
segment the packet with gso_size set to be the maximal
A-MSDU length (after having taken the headers to be added
into account) to make sure that the skb that is passed
down to the transport are not longer than the maximal
A-MSDU allowed.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 4fbaadd..6f67de5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -65,6 +65,7 @@
 #include <linux/ieee80211.h>
 #include <linux/etherdevice.h>
 #include <linux/tcp.h>
+#include <net/ip.h>
 
 #include "iwl-trans.h"
 #include "iwl-eeprom-parse.h"
@@ -182,7 +183,8 @@ void iwl_mvm_set_tx_cmd(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	tx_cmd->tx_flags = cpu_to_le32(tx_flags);
 	/* Total # bytes to be transmitted */
-	tx_cmd->len = cpu_to_le16((u16)skb->len);
+	tx_cmd->len = cpu_to_le16((u16)skb->len +
+		(uintptr_t)info->driver_data[0]);
 	tx_cmd->next_frame_len = 0;
 	tx_cmd->life_time = cpu_to_le32(TX_CMD_LIFE_TIME_INFINITE);
 	tx_cmd->sta_id = sta_id;
@@ -372,6 +374,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 			  info->hw_queue != info->control.vif->cab_queue)))
 		return -1;
 
+	/* This holds the amsdu headers length */
+	info->driver_data[0] = (void *)(uintptr_t)0;
+
 	/*
 	 * IWL_MVM_OFFCHANNEL_QUEUE is used for ROC packets that can be used
 	 * in 2 different types of vifs, P2P & STATION. P2P uses the offchannel
@@ -428,33 +433,156 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 	return 0;
 }
 
-static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb_gso,
+#ifdef CONFIG_INET
+static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 			  struct ieee80211_sta *sta,
 			  struct sk_buff_head *mpdus_skb)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+	unsigned int mss = skb_shinfo(skb)->gso_size;
 	struct sk_buff *tmp, *next;
-	char cb[sizeof(skb_gso->cb)];
+	char cb[sizeof(skb->cb)];
+	unsigned int num_subframes, tcp_payload_len, subf_len;
+	bool ipv4 = (skb->protocol == htons(ETH_P_IP));
+	u16 ip_base_id = ipv4 ? ntohs(ip_hdr(skb)->id) : 0;
+	u16 amsdu_add, snap_ip_tcp, pad, i = 0;
+
+	snap_ip_tcp = 8 + skb_transport_header(skb) - skb_network_header(skb) +
+		tcp_hdrlen(skb);
+
+	if (!sta->max_amsdu_len ||
+	    !ieee80211_is_data_qos(hdr->frame_control)) {
+		num_subframes = 1;
+		pad = 0;
+		goto segment;
+	}
+
+	/* TODO: for now, disable A-MSDU inside AMPDU */
+	if (info->flags & IEEE80211_TX_CTL_AMPDU) {
+		num_subframes = 1;
+		pad = 0;
+		goto segment;
+	}
+
+	/* Sub frame header + SNAP + IP header + TCP header + MSS */
+	subf_len = sizeof(struct ethhdr) + snap_ip_tcp + mss;
+	pad = (4 - subf_len) & 0x3;
+
+	/*
+	 * If we have N subframes in the A-MSDU, then the A-MSDU's size is
+	 * N * subf_len + (N - 1) * pad.
+	 */
+	num_subframes = (sta->max_amsdu_len + pad) / (subf_len + pad);
+	if (num_subframes > 1) {
+		u8 *qc = ieee80211_get_qos_ctl((void *)skb->data);
+
+		*qc |= IEEE80211_QOS_CTL_A_MSDU_PRESENT;
+	}
+
+	tcp_payload_len = skb_tail_pointer(skb) - skb_transport_header(skb) -
+		tcp_hdrlen(skb) + skb->data_len;
+
+	/*
+	 * Make sure we have enough TBs for the A-MSDU:
+	 *	2 for each subframe
+	 *	1 more for each fragment
+	 *	1 more for the potential data in the header
+	 */
+	num_subframes =
+		min_t(unsigned int, num_subframes,
+		      (mvm->trans->max_skb_frags - 1 -
+		       skb_shinfo(skb)->nr_frags) / 2);
+
+	/* This skb fits in one single A-MSDU */
+	if (num_subframes * mss >= tcp_payload_len) {
+		/*
+		 * Compute the length of all the data added for the A-MSDU.
+		 * This will be used to compute the length to write in the TX
+		 * command. We have: SNAP + IP + TCP for n -1 subframes and
+		 * ETH header for n subframes. Note that the original skb
+		 * already had one set of SNAP / IP / TCP headers.
+		 */
+		num_subframes = DIV_ROUND_UP(tcp_payload_len, mss);
+		info = IEEE80211_SKB_CB(skb);
+		amsdu_add = num_subframes * sizeof(struct ethhdr) +
+			(num_subframes - 1) * (snap_ip_tcp + pad);
+		/* This holds the amsdu headers length */
+		info->driver_data[0] = (void *)(uintptr_t)amsdu_add;
+
+		__skb_queue_tail(mpdus_skb, skb);
+		return 0;
+	}
 
-	memcpy(cb, skb_gso->cb, sizeof(cb));
-	next = skb_gso_segment(skb_gso, 0);
-	if (IS_ERR(next))
+	/*
+	 * Trick the segmentation function to make it
+	 * create SKBs that can fit into one A-MSDU.
+	 */
+segment:
+	skb_shinfo(skb)->gso_size = num_subframes * mss;
+	memcpy(cb, skb->cb, sizeof(cb));
+
+	next = skb_gso_segment(skb, NETIF_F_CSUM_MASK | NETIF_F_SG);
+	skb_shinfo(skb)->gso_size = mss;
+	if (WARN_ON_ONCE(IS_ERR(next)))
 		return -EINVAL;
 	else if (next)
-		consume_skb(skb_gso);
+		consume_skb(skb);
 
 	while (next) {
 		tmp = next;
 		next = tmp->next;
+
 		memcpy(tmp->cb, cb, sizeof(tmp->cb));
+		/*
+		 * Compute the length of all the data added for the A-MSDU.
+		 * This will be used to compute the length to write in the TX
+		 * command. We have: SNAP + IP + TCP for n -1 subframes and
+		 * ETH header for n subframes.
+		 */
+		tcp_payload_len = skb_tail_pointer(tmp) -
+			skb_transport_header(tmp) -
+			tcp_hdrlen(tmp) + tmp->data_len;
+
+		if (ipv4)
+			ip_hdr(tmp)->id = htons(ip_base_id + i * num_subframes);
+
+		if (tcp_payload_len > mss) {
+			num_subframes = DIV_ROUND_UP(tcp_payload_len, mss);
+			info = IEEE80211_SKB_CB(tmp);
+			amsdu_add = num_subframes * sizeof(struct ethhdr) +
+				(num_subframes - 1) * (snap_ip_tcp + pad);
+			info->driver_data[0] = (void *)(uintptr_t)amsdu_add;
+			skb_shinfo(tmp)->gso_size = mss;
+		} else {
+			u8 *qc = ieee80211_get_qos_ctl((void *)tmp->data);
+
+			if (ipv4)
+				ip_send_check(ip_hdr(tmp));
+			*qc &= ~IEEE80211_QOS_CTL_A_MSDU_PRESENT;
+			skb_shinfo(tmp)->gso_size = 0;
+		}
 
 		tmp->prev = NULL;
 		tmp->next = NULL;
 
 		__skb_queue_tail(mpdus_skb, tmp);
+		i++;
 	}
 
 	return 0;
 }
+#else /* CONFIG_INET */
+static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
+			  struct ieee80211_sta *sta,
+			  struct sk_buff_head *mpdus_skb)
+{
+	/* Impossible to get TSO with CONFIG_INET */
+	WARN_ON(1);
+
+	return -1;
+}
+#endif
 
 /*
  * Sets the fields in the Tx cmd that are crypto related
@@ -560,6 +688,7 @@ int iwl_mvm_tx_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 		   struct ieee80211_sta *sta)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct sk_buff_head mpdus_skbs;
 	unsigned int payload_len;
 	int ret;
@@ -570,6 +699,9 @@ int iwl_mvm_tx_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (WARN_ON_ONCE(mvmsta->sta_id == IWL_MVM_STATION_COUNT))
 		return -1;
 
+	/* This holds the amsdu headers length */
+	info->driver_data[0] = (void *)(uintptr_t)0;
+
 	if (!skb_is_gso(skb))
 		return iwl_mvm_tx_mpdu(mvm, skb, sta);
 
@@ -589,7 +721,7 @@ int iwl_mvm_tx_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
 		return ret;
 
 	while (!skb_queue_empty(&mpdus_skbs)) {
-		struct sk_buff *skb = __skb_dequeue(&mpdus_skbs);
+		skb = __skb_dequeue(&mpdus_skbs);
 
 		ret = iwl_mvm_tx_mpdu(mvm, skb, sta);
 		if (ret) {
-- 
cgit v0.10.2


From bb81bb68f472bb0f9afbb483259d91d4efd86bfb Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Mon, 26 Oct 2015 16:00:29 +0200
Subject: iwlwifi: mvm: add Tx A-MSDU inside A-MPDU

If the peer allows, we can have A-MSDU inside A-MDPU.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index f1ed90b..1a4946f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -847,6 +847,7 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 	u16 tid = params->tid;
 	u16 *ssn = &params->ssn;
 	u8 buf_size = params->buf_size;
+	bool amsdu = params->amsdu;
 
 	IWL_DEBUG_HT(mvm, "A-MPDU action on addr %pM tid %d: action %d\n",
 		     sta->addr, tid, action);
@@ -907,7 +908,8 @@ static int iwl_mvm_mac_ampdu_action(struct ieee80211_hw *hw,
 		ret = iwl_mvm_sta_tx_agg_flush(mvm, vif, sta, tid);
 		break;
 	case IEEE80211_AMPDU_TX_OPERATIONAL:
-		ret = iwl_mvm_sta_tx_agg_oper(mvm, vif, sta, tid, buf_size);
+		ret = iwl_mvm_sta_tx_agg_oper(mvm, vif, sta, tid,
+					      buf_size, amsdu);
 		break;
 	default:
 		WARN_ON_ONCE(1);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 4854e79..b2123ce 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1031,7 +1031,8 @@ release_locks:
 }
 
 int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			    struct ieee80211_sta *sta, u16 tid, u8 buf_size)
+			    struct ieee80211_sta *sta, u16 tid, u8 buf_size,
+			    bool amsdu)
 {
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_tid_data *tid_data = &mvmsta->tid_data[tid];
@@ -1051,6 +1052,7 @@ int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	tid_data->state = IWL_AGG_ON;
 	mvmsta->agg_tids |= BIT(tid);
 	tid_data->ssn = 0xffff;
+	tid_data->amsdu_in_ampdu_allowed = amsdu;
 	spin_unlock_bh(&mvmsta->lock);
 
 	fifo = iwl_mvm_ac_to_tx_fifo[tid_to_mac80211_ac[tid]];
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index e3b9446..b01650a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -260,6 +260,7 @@ enum iwl_mvm_agg_state {
  *	Tx response (TX_CMD), and the block ack notification (COMPRESSED_BA).
  * @reduced_tpc: Reduced tx power. Holds the data between the
  *	Tx response (TX_CMD), and the block ack notification (COMPRESSED_BA).
+ * @amsdu_in_ampdu_allowed: true if A-MSDU in A-MPDU is allowed.
  * @state: state of the BA agreement establishment / tear down.
  * @txq_id: Tx queue used by the BA session
  * @ssn: the first packet to be sent in AGG HW queue in Tx AGG start flow, or
@@ -274,6 +275,7 @@ struct iwl_mvm_tid_data {
 	/* The rest is Tx AGG related */
 	u32 rate_n_flags;
 	u8 reduced_tpc;
+	bool amsdu_in_ampdu_allowed;
 	enum iwl_mvm_agg_state state;
 	u16 txq_id;
 	u16 ssn;
@@ -405,7 +407,8 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 int iwl_mvm_sta_tx_agg_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta, u16 tid, u16 *ssn);
 int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-			struct ieee80211_sta *sta, u16 tid, u8 buf_size);
+			    struct ieee80211_sta *sta, u16 tid, u8 buf_size,
+			    bool amsdu);
 int iwl_mvm_sta_tx_agg_stop(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			    struct ieee80211_sta *sta, u16 tid);
 int iwl_mvm_sta_tx_agg_flush(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 6f67de5..ff08b17 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -438,19 +438,26 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 			  struct ieee80211_sta *sta,
 			  struct sk_buff_head *mpdus_skb)
 {
+	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	struct sk_buff *tmp, *next;
 	char cb[sizeof(skb->cb)];
-	unsigned int num_subframes, tcp_payload_len, subf_len;
+	unsigned int num_subframes, tcp_payload_len, subf_len, max_amsdu_len;
 	bool ipv4 = (skb->protocol == htons(ETH_P_IP));
 	u16 ip_base_id = ipv4 ? ntohs(ip_hdr(skb)->id) : 0;
 	u16 amsdu_add, snap_ip_tcp, pad, i = 0;
+	u8 *qc, tid;
 
 	snap_ip_tcp = 8 + skb_transport_header(skb) - skb_network_header(skb) +
 		tcp_hdrlen(skb);
 
+	qc = ieee80211_get_qos_ctl(hdr);
+	tid = *qc & IEEE80211_QOS_CTL_TID_MASK;
+	if (WARN_ON_ONCE(tid >= IWL_MAX_TID_COUNT))
+		return -EINVAL;
+
 	if (!sta->max_amsdu_len ||
 	    !ieee80211_is_data_qos(hdr->frame_control)) {
 		num_subframes = 1;
@@ -458,13 +465,28 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 		goto segment;
 	}
 
-	/* TODO: for now, disable A-MSDU inside AMPDU */
-	if (info->flags & IEEE80211_TX_CTL_AMPDU) {
+	/*
+	 * No need to lock amsdu_in_ampdu_allowed since it can't be modified
+	 * during an BA session.
+	 */
+	if (info->flags & IEEE80211_TX_CTL_AMPDU &&
+	    !mvmsta->tid_data[tid].amsdu_in_ampdu_allowed) {
 		num_subframes = 1;
 		pad = 0;
 		goto segment;
 	}
 
+	max_amsdu_len = sta->max_amsdu_len;
+
+	/*
+	 * Limit A-MSDU in A-MPDU to 4095 bytes when VHT is not
+	 * supported. This is a spec requirement (IEEE 802.11-2015
+	 * section 8.7.3 NOTE 3).
+	 */
+	if (info->flags & IEEE80211_TX_CTL_AMPDU &&
+	    !sta->vht_cap.vht_supported)
+		max_amsdu_len = min_t(unsigned int, max_amsdu_len, 4095);
+
 	/* Sub frame header + SNAP + IP header + TCP header + MSS */
 	subf_len = sizeof(struct ethhdr) + snap_ip_tcp + mss;
 	pad = (4 - subf_len) & 0x3;
@@ -473,12 +495,9 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	 * If we have N subframes in the A-MSDU, then the A-MSDU's size is
 	 * N * subf_len + (N - 1) * pad.
 	 */
-	num_subframes = (sta->max_amsdu_len + pad) / (subf_len + pad);
-	if (num_subframes > 1) {
-		u8 *qc = ieee80211_get_qos_ctl((void *)skb->data);
-
+	num_subframes = (max_amsdu_len + pad) / (subf_len + pad);
+	if (num_subframes > 1)
 		*qc |= IEEE80211_QOS_CTL_A_MSDU_PRESENT;
-	}
 
 	tcp_payload_len = skb_tail_pointer(skb) - skb_transport_header(skb) -
 		tcp_hdrlen(skb) + skb->data_len;
@@ -555,7 +574,7 @@ segment:
 			info->driver_data[0] = (void *)(uintptr_t)amsdu_add;
 			skb_shinfo(tmp)->gso_size = mss;
 		} else {
-			u8 *qc = ieee80211_get_qos_ctl((void *)tmp->data);
+			qc = ieee80211_get_qos_ctl((void *)tmp->data);
 
 			if (ipv4)
 				ip_send_check(ip_hdr(tmp));
-- 
cgit v0.10.2


From 9e7dce28659531b2741939e5b6f438bca1422720 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Mon, 26 Oct 2015 16:14:06 +0200
Subject: iwlwifi: mvm: allow to limit the A-MSDU from debugfs

in order to be able to tune the size of the desired A-MSDU
based on link condition, add a knob to modify the length
of the A-MSDU.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 54d0d7d..56e6b0b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -64,6 +64,7 @@
  *
  *****************************************************************************/
 #include <linux/vmalloc.h>
+#include <linux/ieee80211.h>
 
 #include "mvm.h"
 #include "fw-dbg.h"
@@ -1080,6 +1081,22 @@ static ssize_t iwl_dbgfs_fw_dbg_collect_write(struct iwl_mvm *mvm,
 	return count;
 }
 
+static ssize_t iwl_dbgfs_max_amsdu_len_write(struct iwl_mvm *mvm,
+					     char *buf, size_t count,
+					     loff_t *ppos)
+{
+	unsigned int max_amsdu_len;
+	int ret;
+
+	ret = kstrtouint(buf, 0, &max_amsdu_len);
+
+	if (max_amsdu_len > IEEE80211_MAX_MPDU_LEN_VHT_11454)
+		return -EINVAL;
+	mvm->max_amsdu_len = max_amsdu_len;
+
+	return count;
+}
+
 #define ADD_TEXT(...) pos += scnprintf(buf + pos, bufsz - pos, __VA_ARGS__)
 #ifdef CONFIG_IWLWIFI_BCAST_FILTERING
 static ssize_t iwl_dbgfs_bcast_filters_read(struct file *file,
@@ -1497,6 +1514,7 @@ MVM_DEBUGFS_READ_WRITE_FILE_OPS(d0i3_refs, 8);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(fw_dbg_conf, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_dbg_collect, 64);
 MVM_DEBUGFS_WRITE_FILE_OPS(cont_recording, 8);
+MVM_DEBUGFS_WRITE_FILE_OPS(max_amsdu_len, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(indirection_tbl,
 			   (IWL_RSS_INDIRECTION_TABLE_SIZE * 2));
 
@@ -1541,6 +1559,7 @@ int iwl_mvm_dbgfs_register(struct iwl_mvm *mvm, struct dentry *dbgfs_dir)
 	MVM_DEBUGFS_ADD_FILE(d0i3_refs, mvm->debugfs_dir, S_IRUSR | S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(fw_dbg_conf, mvm->debugfs_dir, S_IRUSR | S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(fw_dbg_collect, mvm->debugfs_dir, S_IWUSR);
+	MVM_DEBUGFS_ADD_FILE(max_amsdu_len, mvm->debugfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(send_echo_cmd, mvm->debugfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(cont_recording, mvm->debugfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(indirection_tbl, mvm->debugfs_dir, S_IWUSR);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index bb811a2..416aedb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -861,6 +861,7 @@ struct iwl_mvm {
 
 	/* Indicate if device power save is allowed */
 	u8 ps_disabled; /* u8 instead of bool to ease debugfs_create_* usage */
+	unsigned int max_amsdu_len; /* used for debugfs only */
 
 	struct ieee80211_vif __rcu *csa_vif;
 	struct ieee80211_vif __rcu *csa_tx_blocked_vif;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index ff08b17..a5a55da1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -448,6 +448,7 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	bool ipv4 = (skb->protocol == htons(ETH_P_IP));
 	u16 ip_base_id = ipv4 ? ntohs(ip_hdr(skb)->id) : 0;
 	u16 amsdu_add, snap_ip_tcp, pad, i = 0;
+	unsigned int dbg_max_amsdu_len;
 	u8 *qc, tid;
 
 	snap_ip_tcp = 8 + skb_transport_header(skb) - skb_network_header(skb) +
@@ -477,6 +478,10 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	}
 
 	max_amsdu_len = sta->max_amsdu_len;
+	dbg_max_amsdu_len = ACCESS_ONCE(mvm->max_amsdu_len);
+	if (dbg_max_amsdu_len)
+		max_amsdu_len = min_t(unsigned int, max_amsdu_len,
+				      dbg_max_amsdu_len);
 
 	/*
 	 * Limit A-MSDU in A-MPDU to 4095 bytes when VHT is not
-- 
cgit v0.10.2


From 04e3a5da6b73d7f01661288ead426986fa5312a1 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Wed, 28 Oct 2015 09:47:41 +0200
Subject: iwlwifi: mvm: don't enable A-MSDU when the rates are too low

Allow A-MSDU only when we are not downscaling and the
initial MCS is at least 5.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
index 3dc94d2..61d0a8cd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c
@@ -1673,6 +1673,20 @@ static void rs_stay_in_table(struct iwl_lq_sta *lq_sta, bool force_search)
 	}
 }
 
+static void rs_set_amsdu_len(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
+			     struct iwl_scale_tbl_info *tbl,
+			     enum rs_action scale_action)
+{
+	struct iwl_mvm_sta *sta_priv = iwl_mvm_sta_from_mac80211(sta);
+
+	if ((!is_vht(&tbl->rate) && !is_ht(&tbl->rate)) ||
+	    tbl->rate.index < IWL_RATE_MCS_5_INDEX ||
+	    scale_action == RS_ACTION_DOWNSCALE)
+		sta_priv->tlc_amsdu = false;
+	else
+		sta_priv->tlc_amsdu = true;
+}
+
 /*
  * setup rate table in uCode
  */
@@ -2416,6 +2430,7 @@ lq_update:
 		tbl->rate.index = index;
 		if (IWL_MVM_RS_80_20_FAR_RANGE_TWEAK)
 			rs_tweak_rate_tbl(mvm, sta, lq_sta, tbl, scale_action);
+		rs_set_amsdu_len(mvm, sta, tbl, scale_action);
 		rs_update_rate_tbl(mvm, sta, lq_sta, tbl);
 	}
 
@@ -3099,6 +3114,7 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 	sband = hw->wiphy->bands[band];
 
 	lq_sta->lq.sta_id = sta_priv->sta_id;
+	sta_priv->tlc_amsdu = false;
 
 	for (j = 0; j < LQ_SIZE; j++)
 		rs_rate_scale_clear_tbl_windows(mvm, &lq_sta->lq_info[j]);
@@ -3658,10 +3674,13 @@ static ssize_t rs_sta_dbgfs_scale_table_read(struct file *file,
 	ssize_t ret;
 
 	struct iwl_lq_sta *lq_sta = file->private_data;
+	struct iwl_mvm_sta *mvmsta =
+		container_of(lq_sta, struct iwl_mvm_sta, lq_sta);
 	struct iwl_mvm *mvm;
 	struct iwl_scale_tbl_info *tbl = &(lq_sta->lq_info[lq_sta->active_tbl]);
 	struct rs_rate *rate = &tbl->rate;
 	u32 ss_params;
+
 	mvm = lq_sta->pers.drv;
 	buff = kmalloc(2048, GFP_KERNEL);
 	if (!buff)
@@ -3687,10 +3706,11 @@ static ssize_t rs_sta_dbgfs_scale_table_read(struct file *file,
 				(is_ht20(rate)) ? "20MHz" :
 				(is_ht40(rate)) ? "40MHz" :
 				(is_ht80(rate)) ? "80Mhz" : "BAD BW");
-		desc += sprintf(buff + desc, " %s %s %s\n",
+		desc += sprintf(buff + desc, " %s %s %s %s\n",
 				(rate->sgi) ? "SGI" : "NGI",
 				(rate->ldpc) ? "LDPC" : "BCC",
-				(lq_sta->is_agg) ? "AGG on" : "");
+				(lq_sta->is_agg) ? "AGG on" : "",
+				(mvmsta->tlc_amsdu) ? "AMSDU on" : "");
 	}
 	desc += sprintf(buff+desc, "last tx rate=0x%X\n",
 			lq_sta->last_rate_n_flags);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index b01650a..f95f603 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -313,6 +313,7 @@ struct iwl_mvm_key_pn {
  * @tx_protection: reference counter for controlling the Tx protection.
  * @tt_tx_protection: is thermal throttling enable Tx protection?
  * @disable_tx: is tx to this STA disabled?
+ * @tlc_amsdu: true if A-MSDU is allowed
  * @agg_tids: bitmap of tids whose status is operational aggregated (IWL_AGG_ON)
  * @sleep_tx_count: the number of frames that we told the firmware to let out
  *	even when that station is asleep. This is useful in case the queue
@@ -347,6 +348,7 @@ struct iwl_mvm_sta {
 	bool tt_tx_protection;
 
 	bool disable_tx;
+	bool tlc_amsdu;
 	u8 agg_tids;
 	u8 sleep_tx_count;
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index a5a55da1..a73ae01 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -460,7 +460,8 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 		return -EINVAL;
 
 	if (!sta->max_amsdu_len ||
-	    !ieee80211_is_data_qos(hdr->frame_control)) {
+	    !ieee80211_is_data_qos(hdr->frame_control) ||
+	    !mvmsta->tlc_amsdu) {
 		num_subframes = 1;
 		pad = 0;
 		goto segment;
-- 
cgit v0.10.2


From 50b0213fdb830e21ed64375230c3271b21bb584a Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Wed, 11 Nov 2015 11:37:02 +0200
Subject: iwlwifi: mvm: don't send an A-MSDU that is larger than the TXF

The A-MSDU must be smaller than the Transmit FIFO in the
device. Since the size of the TXF can change depending
on the device / firmware compilation mode, take the size
of the FIFO dynamically from the what the firmware tells us.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index a73ae01..dd616c1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -449,7 +449,7 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	u16 ip_base_id = ipv4 ? ntohs(ip_hdr(skb)->id) : 0;
 	u16 amsdu_add, snap_ip_tcp, pad, i = 0;
 	unsigned int dbg_max_amsdu_len;
-	u8 *qc, tid;
+	u8 *qc, tid, txf;
 
 	snap_ip_tcp = 8 + skb_transport_header(skb) - skb_network_header(skb) +
 		tcp_hdrlen(skb);
@@ -480,6 +480,19 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 
 	max_amsdu_len = sta->max_amsdu_len;
 	dbg_max_amsdu_len = ACCESS_ONCE(mvm->max_amsdu_len);
+
+	/* the Tx FIFO to which this A-MSDU will be routed */
+	txf = iwl_mvm_ac_to_tx_fifo[tid_to_mac80211_ac[tid]];
+
+	/*
+	 * Don't send an AMSDU that will be longer than the TXF.
+	 * Add a security margin of 256 for the TX command + headers.
+	 * We also want to have the start of the next packet inside the
+	 * fifo to be able to send bursts.
+	 */
+	max_amsdu_len = min_t(unsigned int, max_amsdu_len,
+			      mvm->shared_mem_cfg.txfifo_size[txf] - 256);
+
 	if (dbg_max_amsdu_len)
 		max_amsdu_len = min_t(unsigned int, max_amsdu_len,
 				      dbg_max_amsdu_len);
-- 
cgit v0.10.2


From e0d8fdecf3acf36eb738e42d3caf46371500881c Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 28 Dec 2015 22:37:08 +0200
Subject: iwlwifi: support tracing wide commands

Current iwlwifi_trace_dev_rx prints only the cmd without the
group, which might be misleading. Change it to print the wide
id. While at it add the DATA_PATH group and sub commands to the
trace of the command names, sine it is missing due to patches
submitted in parallel.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
index 22786d7..2a0703f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
@@ -73,12 +73,12 @@ TRACE_EVENT(iwlwifi_dev_rx,
 	TP_ARGS(dev, trans, pkt, len),
 	TP_STRUCT__entry(
 		DEV_ENTRY
-		__field(u8, cmd)
+		__field(u16, cmd)
 		__dynamic_array(u8, rxbuf, iwl_rx_trace_len(trans, pkt, len))
 	),
 	TP_fast_assign(
 		DEV_ASSIGN;
-		__entry->cmd = pkt->hdr.cmd;
+		__entry->cmd = WIDE_ID(pkt->hdr.group_id, pkt->hdr.cmd);
 		memcpy(__get_dynamic_array(rxbuf), pkt,
 		       iwl_rx_trace_len(trans, pkt, len));
 	),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index c5b6e86..bfa6da1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -398,6 +398,13 @@ static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
 /* Please keep this array *SORTED* by hex value.
  * Access is done through binary search
  */
+static const struct iwl_hcmd_names iwl_mvm_data_path_names[] = {
+	HCMD_NAME(UPDATE_MU_GROUPS_CMD),
+};
+
+/* Please keep this array *SORTED* by hex value.
+ * Access is done through binary search
+ */
 static const struct iwl_hcmd_names iwl_mvm_prot_offload_names[] = {
 	HCMD_NAME(STORED_BEACON_NTF),
 };
@@ -406,6 +413,7 @@ static const struct iwl_hcmd_arr iwl_mvm_groups[] = {
 	[LEGACY_GROUP] = HCMD_ARR(iwl_mvm_legacy_names),
 	[LONG_GROUP] = HCMD_ARR(iwl_mvm_legacy_names),
 	[PHY_OPS_GROUP] = HCMD_ARR(iwl_mvm_phy_names),
+	[DATA_PATH_GROUP] = HCMD_ARR(iwl_mvm_data_path_names),
 	[PROT_OFFLOAD_GROUP] = HCMD_ARR(iwl_mvm_prot_offload_names),
 };
 
-- 
cgit v0.10.2


From 77fe739554e13d44466e115dbaba3e7aa3aececd Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Sun, 10 Jan 2016 14:23:25 +0200
Subject: iwlwifi: mvm: update rx_status with mactime flag

When forming IBSS, mac80211 scans in order to find an already
existing cell to join.
In case the scan does not find any existing cell a new IBSS
cell is formed.
When receiving the beacons of another IBSS cell we should
merge if the other IBSS cell's TSF is higher than ours.
However, currently iwlmvm does not set any timestamp flag in
rx_status so there is no valid rx timestamp to compare the
beacon's TSF to.
The reason for that is that TSF as indicated by the firmware
is at INA time, but up till now mac80211 expected the TSF at
the beginning or end of the MPDU.
Set the flag to the newly added RX_FLAG_MACTIME_PLCP_START flag.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 535134d..e885db3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -1484,6 +1484,8 @@ void iwl_mvm_rx_stored_beacon_notif(struct iwl_mvm *mvm,
 	/* update rx_status according to the notification's metadata */
 	memset(&rx_status, 0, sizeof(rx_status));
 	rx_status.mactime = le64_to_cpu(sb->tsf);
+	/* TSF as indicated by the firmware  is at INA time */
+	rx_status.flag |= RX_FLAG_MACTIME_PLCP_START;
 	rx_status.device_timestamp = le32_to_cpu(sb->system_time);
 	rx_status.band =
 		(sb->phy_flags & cpu_to_le16(RX_RES_PHY_FLAGS_BAND_24)) ?
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index 055a8b0..ad625f0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -322,11 +322,9 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi,
 	rx_status->freq =
 		ieee80211_channel_to_frequency(le16_to_cpu(phy_info->channel),
 					       rx_status->band);
-	/*
-	 * TSF as indicated by the fw is at INA time, but mac80211 expects the
-	 * TSF at the beginning of the MPDU.
-	 */
-	/*rx_status->flag |= RX_FLAG_MACTIME_MPDU;*/
+
+	/* TSF as indicated by the firmware  is at INA time */
+	rx_status->flag |= RX_FLAG_MACTIME_PLCP_START;
 
 	iwl_mvm_get_signal_strength(mvm, phy_info, rx_status);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 615dea1..a9180b0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -332,6 +332,8 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 	rx_status->freq = ieee80211_channel_to_frequency(desc->channel,
 							 rx_status->band);
 	iwl_mvm_get_signal_strength(mvm, desc, rx_status);
+	/* TSF as indicated by the firmware is at INA time */
+	rx_status->flag |= RX_FLAG_MACTIME_PLCP_START;
 
 	rcu_read_lock();
 
-- 
cgit v0.10.2


From 3af512d6aac7eb6420086f124abb4426f5f4b369 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 22 Jul 2015 11:38:40 +0300
Subject: iwlwifi: mvm: support filtered frames notification

During d0i3 frames might be filtered by the FW and this may
cause reordering buffer a delay - as the frames will not be
received and reorder will time out.

Introduce an API function to receive notification of filtered
frames and pass the information to the mac80211.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index ca7fec71..e1e1194 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -119,6 +119,8 @@ enum {
 	SCAN_ABORT_UMAC = 0xe,
 	SCAN_COMPLETE_UMAC = 0xf,
 
+	BA_WINDOW_STATUS_NOTIFICATION_ID = 0x13,
+
 	/* station table */
 	ADD_STA_KEY = 0x17,
 	ADD_STA = 0x18,
@@ -1286,6 +1288,26 @@ struct iwl_fw_bcast_filter {
 	struct iwl_fw_bcast_filter_attr attrs[MAX_BCAST_FILTER_ATTRS];
 } __packed; /* BCAST_FILTER_S_VER_1 */
 
+#define BA_WINDOW_STREAMS_MAX		16
+#define BA_WINDOW_STATUS_TID_MSK	0x000F
+#define BA_WINDOW_STATUS_STA_ID_POS	4
+#define BA_WINDOW_STATUS_STA_ID_MSK	0x01F0
+#define BA_WINDOW_STATUS_VALID_MSK	BIT(9)
+
+/**
+ * struct iwl_ba_window_status_notif - reordering window's status notification
+ * @bitmap: bitmap of received frames [start_seq_num + 0]..[start_seq_num + 63]
+ * @ra_tid: bit 3:0 - TID, bit 8:4 - STA_ID, bit 9 - valid
+ * @start_seq_num: the start sequence number of the bitmap
+ * @mpdu_rx_count: the number of received MPDUs since entering D0i3
+ */
+struct iwl_ba_window_status_notif {
+	__le64 bitmap[BA_WINDOW_STREAMS_MAX];
+	__le16 ra_tid[BA_WINDOW_STREAMS_MAX];
+	__le32 start_seq_num[BA_WINDOW_STREAMS_MAX];
+	__le16 mpdu_rx_count[BA_WINDOW_STREAMS_MAX];
+} __packed; /* BA_WINDOW_STATUS_NTFY_API_S_VER_1 */
+
 /**
  * struct iwl_fw_bcast_mac - per-mac broadcast filtering configuration.
  * @default_discard: default action for this mac (discard (1) / pass (0)).
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 416aedb..fa987bd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1268,6 +1268,8 @@ void iwl_mvm_rx_missed_beacons_notif(struct iwl_mvm *mvm,
 				     struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_rx_stored_beacon_notif(struct iwl_mvm *mvm,
 				    struct iwl_rx_cmd_buffer *rxb);
+void iwl_mvm_window_status_notif(struct iwl_mvm *mvm,
+				 struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_mac_ctxt_recalc_tsf_id(struct iwl_mvm *mvm,
 				    struct ieee80211_vif *vif);
 unsigned long iwl_mvm_get_used_hw_queues(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index bfa6da1..52c73d0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -236,6 +236,9 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 	RX_HANDLER(ANTENNA_COUPLING_NOTIFICATION,
 		   iwl_mvm_rx_ant_coupling_notif, true),
 
+	RX_HANDLER(BA_WINDOW_STATUS_NOTIFICATION_ID,
+		   iwl_mvm_window_status_notif, false),
+
 	RX_HANDLER(TIME_EVENT_NOTIFICATION, iwl_mvm_rx_time_event_notif, false),
 	RX_HANDLER(MCC_CHUB_UPDATE_CMD, iwl_mvm_rx_chub_update_mcc, true),
 
@@ -294,6 +297,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
 	HCMD_NAME(SCAN_COMPLETE_UMAC),
 	HCMD_NAME(TOF_CMD),
 	HCMD_NAME(TOF_NOTIFICATION),
+	HCMD_NAME(BA_WINDOW_STATUS_NOTIFICATION_ID),
 	HCMD_NAME(ADD_STA_KEY),
 	HCMD_NAME(ADD_STA),
 	HCMD_NAME(REMOVE_STA),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index ad625f0..485cfc1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -626,3 +627,51 @@ void iwl_mvm_rx_statistics(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
 {
 	iwl_mvm_handle_rx_statistics(mvm, rxb_addr(rxb));
 }
+
+void iwl_mvm_window_status_notif(struct iwl_mvm *mvm,
+				 struct iwl_rx_cmd_buffer *rxb)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_ba_window_status_notif *notif = (void *)pkt->data;
+	int i;
+	u32 pkt_len = iwl_rx_packet_payload_len(pkt);
+
+	if (WARN_ONCE(pkt_len != sizeof(*notif),
+		      "Received window status notification of wrong size (%u)\n",
+		      pkt_len))
+		return;
+
+	rcu_read_lock();
+	for (i = 0; i < BA_WINDOW_STREAMS_MAX; i++) {
+		struct ieee80211_sta *sta;
+		u8 sta_id, tid;
+		u64 bitmap;
+		u32 ssn;
+		u16 ratid;
+		u16 received_mpdu;
+
+		ratid = le16_to_cpu(notif->ra_tid[i]);
+		/* check that this TID is valid */
+		if (!(ratid & BA_WINDOW_STATUS_VALID_MSK))
+			continue;
+
+		received_mpdu = le16_to_cpu(notif->mpdu_rx_count[i]);
+		if (received_mpdu == 0)
+			continue;
+
+		tid = ratid & BA_WINDOW_STATUS_TID_MSK;
+		/* get the station */
+		sta_id = (ratid & BA_WINDOW_STATUS_STA_ID_MSK)
+			 >> BA_WINDOW_STATUS_STA_ID_POS;
+		sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]);
+		if (IS_ERR_OR_NULL(sta))
+			continue;
+		bitmap = le64_to_cpu(notif->bitmap[i]);
+		ssn = le32_to_cpu(notif->start_seq_num[i]);
+
+		/* update mac80211 with the bitmap for the reordering buffer */
+		ieee80211_mark_rx_ba_filtered_frames(sta, tid, ssn, bitmap,
+						     received_mpdu);
+	}
+	rcu_read_unlock();
+}
-- 
cgit v0.10.2


From 88076015f80f7645b1eaada6291f4a23ab96bd8f Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 15 Feb 2016 17:26:48 +0200
Subject: iwlwifi: pcie: configure more RFH settings

Fine tune RFH registers further:
* Set default queue explicitly
* Set RFH to drop frames exceeding RB size
* Set the maximum rx transfer size to DRAM to 128 instead of 64

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
index 4ab6682..8af818b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
@@ -6,7 +6,7 @@
  * GPL LICENSE SUMMARY
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
- * Copyright(c) 2015 Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -32,7 +32,7 @@
  * BSD LICENSE
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
- * Copyright(c) 2015 Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -368,20 +368,24 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl)
 #define RFH_RXF_DMA_RBDCB_SIZE_512	(0x9 << RFH_RXF_DMA_RBDCB_SIZE_POS)
 #define RFH_RXF_DMA_RBDCB_SIZE_1024	(0xA << RFH_RXF_DMA_RBDCB_SIZE_POS)
 #define RFH_RXF_DMA_RBDCB_SIZE_2048	(0xB << RFH_RXF_DMA_RBDCB_SIZE_POS)
-#define RFH_RXF_DMA_MIN_RB_SIZE_MASK (0x03000000) /* bit 24-25 */
+#define RFH_RXF_DMA_MIN_RB_SIZE_MASK	(0x03000000) /* bit 24-25 */
 #define RFH_RXF_DMA_MIN_RB_SIZE_POS	24
-#define RFH_RXF_DMA_MIN_RB_4_8	(3 << RFH_RXF_DMA_MIN_RB_SIZE_POS)
-#define RFH_RXF_DMA_SINGLE_FRAME_MASK (0x20000000) /* bit 29 */
-#define RFH_DMA_EN_MASK (0xC0000000) /* bits 30-31*/
-#define RFH_DMA_EN_ENABLE_VAL BIT(31)
+#define RFH_RXF_DMA_MIN_RB_4_8		(3 << RFH_RXF_DMA_MIN_RB_SIZE_POS)
+#define RFH_RXF_DMA_DROP_TOO_LARGE_MASK	(0x04000000) /* bit 26 */
+#define RFH_RXF_DMA_SINGLE_FRAME_MASK	(0x20000000) /* bit 29 */
+#define RFH_DMA_EN_MASK			(0xC0000000) /* bits 30-31*/
+#define RFH_DMA_EN_ENABLE_VAL		BIT(31)
 
 #define RFH_RXF_RXQ_ACTIVE 0xA0980C
 
 #define RFH_GEN_CFG	0xA09800
+#define RFH_GEN_CFG_SERVICE_DMA_SNOOP	BIT(0)
+#define RFH_GEN_CFG_RFH_DMA_SNOOP	BIT(1)
+#define RFH_GEN_CFG_RB_CHUNK_SIZE	BIT(4) /* 0 - 64B, 1- 128B */
 #define RFH_GEN_CFG_DEFAULT_RXQ_NUM_MASK 0xF00
-#define RFH_GEN_CFG_SERVICE_DMA_SNOOP BIT(0)
-#define RFH_GEN_CFG_RFH_DMA_SNOOP BIT(1)
-#define DEFAULT_RXQ_NUM 8
+#define RFH_GEN_CFG_DEFAULT_RXQ_NUM_POS 8
+
+#define DEFAULT_RXQ_NUM			0
 
 /* end of 9000 rx series registers */
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index c0db681..398dd93 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -783,16 +783,26 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans)
 	 * Single frame mode
 	 * Rx buffer size 4 or 8k or 12k
 	 * Min RB size 4 or 8
+	 * Drop frames that exceed RB size
 	 * 512 RBDs
 	 */
 	iwl_write_prph(trans, RFH_RXF_DMA_CFG,
 		       RFH_DMA_EN_ENABLE_VAL |
 		       rb_size | RFH_RXF_DMA_SINGLE_FRAME_MASK |
 		       RFH_RXF_DMA_MIN_RB_4_8 |
+		       RFH_RXF_DMA_DROP_TOO_LARGE_MASK |
 		       RFH_RXF_DMA_RBDCB_SIZE_512);
 
+	/*
+	 * Activate DMA snooping.
+	 * Set RX DMA chunk size to 128 bit
+	 * Default queue is 0
+	 */
 	iwl_write_prph(trans, RFH_GEN_CFG, RFH_GEN_CFG_RFH_DMA_SNOOP |
-					  RFH_GEN_CFG_SERVICE_DMA_SNOOP);
+		       RFH_GEN_CFG_RB_CHUNK_SIZE |
+		       (DEFAULT_RXQ_NUM << RFH_GEN_CFG_DEFAULT_RXQ_NUM_POS) |
+		       RFH_GEN_CFG_SERVICE_DMA_SNOOP);
+	/* Enable the relevant rx queues */
 	iwl_write_prph(trans, RFH_RXF_RXQ_ACTIVE, enabled);
 
 	/* Set interrupt coalescing timer to default (2048 usecs) */
-- 
cgit v0.10.2


From 863eac305a7ba75e7c394b2a24cdee2a636b7dc4 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Tue, 1 Dec 2015 20:39:57 +0200
Subject: iwlwifi: pcie: add pm_prepare and pm_complete ops

With these ops, we can know when we are about to enter system suspend.
This allows us to exit D0i3 state before entering suspend.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 0ca0f13..91d74b3 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -836,6 +836,7 @@ struct iwl_trans {
 
 	enum iwl_plat_pm_mode system_pm_mode;
 	enum iwl_plat_pm_mode runtime_pm_mode;
+	bool suspending;
 
 	/* pointer to trans specific struct */
 	/*Ensure that this pointer will always be aligned to sizeof pointer */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 753ec67..d33b6ba 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -811,6 +811,45 @@ static int iwl_pci_runtime_resume(struct device *device)
 
 	return 0;
 }
+
+static int iwl_pci_system_prepare(struct device *device)
+{
+	struct pci_dev *pdev = to_pci_dev(device);
+	struct iwl_trans *trans = pci_get_drvdata(pdev);
+
+	IWL_DEBUG_RPM(trans, "preparing for system suspend\n");
+
+	/* This is called before entering system suspend and before
+	 * the runtime resume is called.  Set the suspending flag to
+	 * prevent the wakelock from being taken.
+	 */
+	trans->suspending = true;
+
+	/* Wake the device up from runtime suspend before going to
+	 * platform suspend.  This is needed because we don't know
+	 * whether wowlan any is set and, if it's not, mac80211 will
+	 * disconnect (in which case, we can't be in D0i3).
+	 */
+	pm_runtime_resume(device);
+
+	return 0;
+}
+
+static void iwl_pci_system_complete(struct device *device)
+{
+	struct pci_dev *pdev = to_pci_dev(device);
+	struct iwl_trans *trans = pci_get_drvdata(pdev);
+
+	IWL_DEBUG_RPM(trans, "completing system suspend\n");
+
+	/* This is called as a counterpart to the prepare op.  It is
+	 * called either when suspending fails or when suspend
+	 * completed successfully.  Now there's no risk of grabbing
+	 * the wakelock anymore, so we can release the suspending
+	 * flag.
+	 */
+	trans->suspending = false;
+}
 #endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 
 static const struct dev_pm_ops iwl_dev_pm_ops = {
@@ -820,6 +859,8 @@ static const struct dev_pm_ops iwl_dev_pm_ops = {
 	SET_RUNTIME_PM_OPS(iwl_pci_runtime_suspend,
 			   iwl_pci_runtime_resume,
 			   NULL)
+	.prepare = iwl_pci_system_prepare,
+	.complete = iwl_pci_system_complete,
 #endif /* CONFIG_IWLWIFI_PCIE_RTPM */
 };
 
-- 
cgit v0.10.2


From 685b346c3af35bcdc1316289ce8c76e28a5a7a7e Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 23 Feb 2016 11:34:17 +0200
Subject: iwlwifi: pcie: prevent skbs shadowing in iwl_trans_pcie_reclaim

The patch below introduced a variable shadowing. Fix that.

Fixes: 3955525d5d17 ("iwlwifi: pcie: buffer packets to avoid overflowing Tx queues")
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index 837a7d5..16ad820 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -1062,10 +1062,10 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 
 	if (iwl_queue_space(&txq->q) > txq->q.low_mark &&
 	    test_bit(txq_id, trans_pcie->queue_stopped)) {
-		struct sk_buff_head skbs;
+		struct sk_buff_head overflow_skbs;
 
-		__skb_queue_head_init(&skbs);
-		skb_queue_splice_init(&txq->overflow_q, &skbs);
+		__skb_queue_head_init(&overflow_skbs);
+		skb_queue_splice_init(&txq->overflow_q, &overflow_skbs);
 
 		/*
 		 * This is tricky: we are in reclaim path which is non
@@ -1076,8 +1076,8 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		 */
 		spin_unlock_bh(&txq->lock);
 
-		while (!skb_queue_empty(&skbs)) {
-			struct sk_buff *skb = __skb_dequeue(&skbs);
+		while (!skb_queue_empty(&overflow_skbs)) {
+			struct sk_buff *skb = __skb_dequeue(&overflow_skbs);
 			struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 			u8 dev_cmd_idx = IWL_TRANS_FIRST_DRIVER_DATA + 1;
 			struct iwl_device_cmd *dev_cmd =
-- 
cgit v0.10.2


From a571f5f635ef40ff3a5e85acc7f20daa58bcf06a Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 7 Dec 2015 12:50:58 +0200
Subject: iwlwifi: mvm: add duplicate packet detection per rx queue

Next hardware will direct TCP/UDP streams to different cores.
Packets belonging to the same stream will be directed to the same
core.
The result is that duplicates will be always directed to the same
rx queue were the first packet was received.
This enabled parallelizing the duplicate packet detection across
the different cores, without sharing data between the rx queues.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index a9180b0..d891942 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -285,6 +285,66 @@ static void iwl_mvm_rx_csum(struct ieee80211_sta *sta,
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 }
 
+/*
+ * returns true if a packet outside BA session is a duplicate and
+ * should be dropped
+ */
+static bool iwl_mvm_is_nonagg_dup(struct ieee80211_sta *sta, int queue,
+				  struct ieee80211_rx_status *rx_status,
+				  struct ieee80211_hdr *hdr,
+				  struct iwl_rx_mpdu_desc *desc)
+{
+	struct iwl_mvm_sta *mvm_sta;
+	struct iwl_mvm_rxq_dup_data *dup_data;
+	u8 baid, tid, sub_frame_idx;
+
+	if (WARN_ON(IS_ERR_OR_NULL(sta)))
+		return false;
+
+	baid = (le32_to_cpu(desc->reorder_data) &
+		IWL_RX_MPDU_REORDER_BAID_MASK) >>
+		IWL_RX_MPDU_REORDER_BAID_SHIFT;
+
+	if (baid != IWL_RX_REORDER_DATA_INVALID_BAID)
+		return false;
+
+	mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	dup_data = &mvm_sta->dup_data[queue];
+
+	/*
+	 * Drop duplicate 802.11 retransmissions
+	 * (IEEE 802.11-2012: 9.3.2.10 "Duplicate detection and recovery")
+	 */
+	if (ieee80211_is_ctl(hdr->frame_control) ||
+	    ieee80211_is_qos_nullfunc(hdr->frame_control) ||
+	    is_multicast_ether_addr(hdr->addr1)) {
+		rx_status->flag |= RX_FLAG_DUP_VALIDATED;
+		return false;
+	}
+
+	if (ieee80211_is_data_qos(hdr->frame_control))
+		/* frame has qos control */
+		tid = *ieee80211_get_qos_ctl(hdr) &
+			IEEE80211_QOS_CTL_TID_MASK;
+	else
+		tid = IWL_MAX_TID_COUNT;
+
+	/* If this wasn't a part of an A-MSDU the sub-frame index will be 0 */
+	sub_frame_idx = desc->amsdu_info & IWL_RX_MPDU_AMSDU_SUBFRAME_IDX_MASK;
+
+	if (unlikely(ieee80211_has_retry(hdr->frame_control) &&
+		     dup_data->last_seq[tid] == hdr->seq_ctrl &&
+		     dup_data->last_sub_frame[tid] >= sub_frame_idx))
+		return true;
+
+	dup_data->last_seq[tid] = hdr->seq_ctrl;
+	dup_data->last_sub_frame[tid] = sub_frame_idx;
+
+	rx_status->flag |= RX_FLAG_DUP_VALIDATED;
+
+	return false;
+}
+
 void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			struct iwl_rx_cmd_buffer *rxb, int queue)
 {
@@ -389,6 +449,12 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 
 		if (ieee80211_is_data(hdr->frame_control))
 			iwl_mvm_rx_csum(sta, skb, desc);
+
+		if (iwl_mvm_is_nonagg_dup(sta, queue, rx_status, hdr, desc)) {
+			kfree_skb(skb);
+			rcu_read_unlock();
+			return;
+		}
 	}
 
 	/*
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index b2123ce..4717b18 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -280,6 +280,7 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta);
+	struct iwl_mvm_rxq_dup_data *dup_data;
 	int i, ret, sta_id;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -327,6 +328,16 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm,
 	}
 	mvm_sta->agg_tids = 0;
 
+	if (iwl_mvm_has_new_rx_api(mvm) &&
+	    !test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) {
+		dup_data = kcalloc(mvm->trans->num_rx_queues,
+				   sizeof(*dup_data),
+				   GFP_KERNEL);
+		if (!dup_data)
+			return -ENOMEM;
+		mvm_sta->dup_data = dup_data;
+	}
+
 	ret = iwl_mvm_sta_send_to_fw(mvm, sta, false);
 	if (ret)
 		goto err;
@@ -508,6 +519,9 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm,
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (iwl_mvm_has_new_rx_api(mvm))
+		kfree(mvm_sta->dup_data);
+
 	if (vif->type == NL80211_IFTYPE_STATION &&
 	    mvmvif->ap_sta_id == mvm_sta->sta_id) {
 		ret = iwl_mvm_drain_sta(mvm, mvm_sta, true);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index f95f603..db701ca 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -296,6 +296,16 @@ struct iwl_mvm_key_pn {
 };
 
 /**
+ * struct iwl_mvm_rxq_dup_data - per station per rx queue data
+ * @last_seq: last sequence per tid for duplicate packet detection
+ * @last_sub_frame: last subframe packet
+ */
+struct iwl_mvm_rxq_dup_data {
+	__le16 last_seq[IWL_MAX_TID_COUNT + 1];
+	u8 last_sub_frame[IWL_MAX_TID_COUNT + 1];
+} ____cacheline_aligned_in_smp;
+
+/**
  * struct iwl_mvm_sta - representation of a station in the driver
  * @sta_id: the index of the station in the fw (will be replaced by id_n_color)
  * @tfd_queue_msk: the tfd queues used by the station
@@ -321,6 +331,7 @@ struct iwl_mvm_key_pn {
  *	we are sending frames from an AMPDU queue and there was a hole in
  *	the BA window. To be used for UAPSD only.
  * @ptk_pn: per-queue PTK PN data structures
+ * @dup_data: per queue duplicate packet detection data
  *
  * When mac80211 creates a station it reserves some space (hw->sta_data_size)
  * in the structure for use by driver. This structure is placed in that
@@ -340,8 +351,8 @@ struct iwl_mvm_sta {
 	struct iwl_mvm_tid_data tid_data[IWL_MAX_TID_COUNT];
 	struct iwl_lq_sta lq_sta;
 	struct ieee80211_vif *vif;
-
 	struct iwl_mvm_key_pn __rcu *ptk_pn[4];
+	struct iwl_mvm_rxq_dup_data *dup_data;
 
 	/* Temporary, until the new TLC will control the Tx protection */
 	s8 tx_protection;
-- 
cgit v0.10.2


From 94bb44813ebe07ff06f869c7612c0c73e04ac4b0 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 16 Dec 2015 18:48:28 +0200
Subject: iwlwifi: mvm: add RSS queues notification infrastructure

In multi rx queue HW, without execessive locking, there is no sync
between the ctrl path (default queue) and the rest of the rx queues.
This might cause issues on certain situations. For example, in case
a delBA was processed on a default queue but out of order packets
still wait for processing on the other queue.

The solution is to introduce internal messaging between the CTRL path
and the other rx queues.
The driver will send a message to the firmware, which will echo it to
all the requested queues. The message will be in order inside the queue.
This way we can avoid CTRL path and RSS queues races.

Add support for this messaging mechanism. As the firmware is agnostic to
the data sent, add internal representation of the data as well.
Although currently only delBA flow will use it, the internal representation
will enable generic use of this infrastructure for future uses.
Next patch will utilize this messaging mechanism for the reorder buffer
delBA flow.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
index df939f5..eb9b870 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
@@ -391,4 +391,56 @@ struct iwl_rss_config_cmd {
 	u8 indirection_table[IWL_RSS_INDIRECTION_TABLE_SIZE];
 } __packed; /* RSS_CONFIG_CMD_API_S_VER_1 */
 
+#define IWL_MULTI_QUEUE_SYNC_MSG_MAX_SIZE 128
+#define IWL_MULTI_QUEUE_SYNC_SENDER_POS 0
+#define IWL_MULTI_QUEUE_SYNC_SENDER_MSK 0xf
+
+/**
+ * struct iwl_rxq_sync_cmd - RXQ notification trigger
+ *
+ * @flags: flags of the notification. bit 0:3 are the sender queue
+ * @rxq_mask: rx queues to send the notification on
+ * @count: number of bytes in payload, should be DWORD aligned
+ * @payload: data to send to rx queues
+ */
+struct iwl_rxq_sync_cmd {
+	__le32 flags;
+	__le32 rxq_mask;
+	__le32 count;
+	u8 payload[];
+} __packed; /* MULTI_QUEUE_DRV_SYNC_HDR_CMD_API_S_VER_1 */
+
+/**
+ * struct iwl_rxq_sync_notification - Notification triggered by RXQ
+ * sync command
+ *
+ * @count: number of bytes in payload
+ * @payload: data to send to rx queues
+ */
+struct iwl_rxq_sync_notification {
+	__le32 count;
+	u8 payload[];
+} __packed; /* MULTI_QUEUE_DRV_SYNC_HDR_CMD_API_S_VER_1 */
+
+/**
+* Internal message identifier
+*
+* @IWL_MVM_RXQ_NOTIF_DEL_BA: notify RSS queues of delBA
+*/
+enum iwl_mvm_rxq_notif_type {
+	IWL_MVM_RXQ_NOTIF_DEL_BA,
+};
+
+/**
+* struct iwl_mvm_internal_rxq_notif - Internal representation of the data sent
+* in &iwl_rxq_sync_cmd. Should be DWORD aligned.
+*
+* @type: value from &iwl_mvm_rxq_notif_type
+* @data: payload
+*/
+struct iwl_mvm_internal_rxq_notif {
+	u32 type;
+	u8 data[];
+} __packed;
+
 #endif /* __fw_api_rx_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index e1e1194..f432ddd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -289,6 +289,8 @@ enum iwl_phy_ops_subcmd_ids {
 
 enum iwl_data_path_subcmd_ids {
 	UPDATE_MU_GROUPS_CMD = 0x1,
+	TRIGGER_RX_QUEUES_NOTIF_CMD = 0x2,
+	RX_QUEUES_NOTIFICATION = 0xFF,
 };
 
 enum iwl_prot_offload_subcmd_ids {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index fa987bd..f05d2a1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1225,6 +1225,10 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			struct iwl_rx_cmd_buffer *rxb, int queue);
 void iwl_mvm_rx_frame_release(struct iwl_mvm *mvm,
 			      struct iwl_rx_cmd_buffer *rxb, int queue);
+int iwl_mvm_notify_rx_queue(struct iwl_mvm *mvm, u32 rxq_mask,
+			    const u8 *data, u32 count);
+void iwl_mvm_rx_queue_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb,
+			    int queue);
 void iwl_mvm_rx_tx_cmd(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_rx_ba_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_rx_ant_coupling_notif(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 52c73d0..ac271ff 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -404,6 +404,8 @@ static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
  */
 static const struct iwl_hcmd_names iwl_mvm_data_path_names[] = {
 	HCMD_NAME(UPDATE_MU_GROUPS_CMD),
+	HCMD_NAME(TRIGGER_RX_QUEUES_NOTIF_CMD),
+	HCMD_NAME(RX_QUEUES_NOTIFICATION),
 };
 
 /* Please keep this array *SORTED* by hex value.
@@ -876,6 +878,9 @@ static void iwl_mvm_rx_mq(struct iwl_op_mode *op_mode,
 		iwl_mvm_rx_mpdu_mq(mvm, napi, rxb, 0);
 	else if (pkt->hdr.cmd == REPLY_RX_PHY_CMD)
 		iwl_mvm_rx_phy_cmd_mq(mvm, rxb);
+	else if (unlikely(pkt->hdr.group_id == DATA_PATH_GROUP &&
+			  pkt->hdr.cmd == RX_QUEUES_NOTIFICATION))
+		iwl_mvm_rx_queue_notif(mvm, rxb, 0);
 	else
 		iwl_mvm_rx_common(mvm, rxb, pkt);
 }
@@ -1548,6 +1553,9 @@ static void iwl_mvm_rx_mq_rss(struct iwl_op_mode *op_mode,
 
 	if (unlikely(pkt->hdr.cmd == FRAME_RELEASE))
 		iwl_mvm_rx_frame_release(mvm, rxb, queue);
+	else if (unlikely(pkt->hdr.cmd == RX_QUEUES_NOTIFICATION &&
+			  pkt->hdr.group_id == DATA_PATH_GROUP))
+		iwl_mvm_rx_queue_notif(mvm, rxb, queue);
 	else
 		iwl_mvm_rx_mpdu_mq(mvm, napi, rxb, queue);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index d891942..590fc6f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -345,6 +345,54 @@ static bool iwl_mvm_is_nonagg_dup(struct ieee80211_sta *sta, int queue,
 	return false;
 }
 
+int iwl_mvm_notify_rx_queue(struct iwl_mvm *mvm, u32 rxq_mask,
+			    const u8 *data, u32 count)
+{
+	struct iwl_rxq_sync_cmd *cmd;
+	u32 data_size = sizeof(*cmd) + count;
+	int ret;
+
+	/* should be DWORD aligned */
+	if (WARN_ON(count & 3 || count > IWL_MULTI_QUEUE_SYNC_MSG_MAX_SIZE))
+		return -EINVAL;
+
+	cmd = kzalloc(data_size, GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	cmd->rxq_mask = cpu_to_le32(rxq_mask);
+	cmd->count =  cpu_to_le32(count);
+	cmd->flags = 0;
+	memcpy(cmd->payload, data, count);
+
+	ret = iwl_mvm_send_cmd_pdu(mvm,
+				   WIDE_ID(DATA_PATH_GROUP,
+					   TRIGGER_RX_QUEUES_NOTIF_CMD),
+				   0, data_size, cmd);
+
+	kfree(cmd);
+	return ret;
+}
+
+void iwl_mvm_rx_queue_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb,
+			    int queue)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_rxq_sync_notification *notif;
+	struct iwl_mvm_internal_rxq_notif *internal_notif;
+
+	notif = (void *)pkt->data;
+	internal_notif = (void *)notif->payload;
+
+	switch (internal_notif->type) {
+	case IWL_MVM_RXQ_NOTIF_DEL_BA:
+		/* TODO */
+		break;
+	default:
+		WARN_ONCE(1, "Invalid identifier %d", internal_notif->type);
+	}
+}
+
 void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			struct iwl_rx_cmd_buffer *rxb, int queue)
 {
-- 
cgit v0.10.2


From 7ed3eec53023deecffb61abbcb7f8408ff6ab064 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 25 Feb 2016 10:12:36 +0200
Subject: iwlwifi: mvm: remove unused field in iwl_mvm_tid_data

Commit 69c7fda40921c125eeeef6a827f6270ac6aa1baa removed the
users of iwl_mvm_tid_data.reduced_tpc. Due to a conflict,
I forgot to commit the hunk that removed the field itself.
Do this know.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index db701ca..1a8f69a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -258,8 +258,6 @@ enum iwl_mvm_agg_state {
  *	This is basically (last acked packet++).
  * @rate_n_flags: Rate at which Tx was attempted. Holds the data between the
  *	Tx response (TX_CMD), and the block ack notification (COMPRESSED_BA).
- * @reduced_tpc: Reduced tx power. Holds the data between the
- *	Tx response (TX_CMD), and the block ack notification (COMPRESSED_BA).
  * @amsdu_in_ampdu_allowed: true if A-MSDU in A-MPDU is allowed.
  * @state: state of the BA agreement establishment / tear down.
  * @txq_id: Tx queue used by the BA session
@@ -274,7 +272,6 @@ struct iwl_mvm_tid_data {
 	u16 next_reclaimed;
 	/* The rest is Tx AGG related */
 	u32 rate_n_flags;
-	u8 reduced_tpc;
 	bool amsdu_in_ampdu_allowed;
 	enum iwl_mvm_agg_state state;
 	u16 txq_id;
-- 
cgit v0.10.2


From f92659a110436cdfdd8d7661188ba059b61404b7 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 3 Feb 2016 15:04:49 +0200
Subject: iwlwifi: mvm: support VHT MU-MIMO notification

When the device is in d0i3/d3 we will not receive the VHT
MU-MIMO group id management frame. Instead, firmware will
notify us upon exit on the current status and we can in turn
update mac80211. Support this notification.
While at it, also check as a precaution that the vif is indeed
the VHT MU-MIMO owner before updating the firmware.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
index f432ddd..4a0fc47 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h
@@ -290,6 +290,7 @@ enum iwl_phy_ops_subcmd_ids {
 enum iwl_data_path_subcmd_ids {
 	UPDATE_MU_GROUPS_CMD = 0x1,
 	TRIGGER_RX_QUEUES_NOTIF_CMD = 0x2,
+	MU_GROUP_MGMT_NOTIF = 0xFE,
 	RX_QUEUES_NOTIFICATION = 0xFF,
 };
 
@@ -1965,6 +1966,18 @@ struct iwl_mu_group_mgmt_cmd {
 	__le32 user_position[4];
 } __packed; /* MU_GROUP_ID_MNG_TABLE_API_S_VER_1 */
 
+/**
+ * struct iwl_mu_group_mgmt_notif - VHT MU-MIMO group id notification
+ *
+ * @membership_status: a bitmap of MU groups
+ * @user_position: the position of station in a group. If the station is in the
+ *	group then bits (group * 2) is the position -1
+ */
+struct iwl_mu_group_mgmt_notif {
+	__le32 membership_status[2];
+	__le32 user_position[4];
+} __packed; /* MU_GROUP_MNG_NTFY_API_S_VER_1 */
+
 #define MAX_STORED_BEACON_SIZE 600
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 1a4946f..b755161 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -1780,6 +1781,34 @@ static int iwl_mvm_update_mu_groups(struct iwl_mvm *mvm,
 				    0, sizeof(cmd), &cmd);
 }
 
+static void iwl_mvm_mu_mimo_iface_iterator(void *_data, u8 *mac,
+					   struct ieee80211_vif *vif)
+{
+	if (vif->mu_mimo_owner) {
+		struct iwl_mu_group_mgmt_notif *notif = _data;
+
+		/*
+		 * MU-MIMO Group Id action frame is little endian. We treat
+		 * the data received from firmware as if it came from the
+		 * action frame, so no conversion is needed.
+		 */
+		ieee80211_update_mu_groups(vif,
+					   (u8 *)&notif->membership_status,
+					   (u8 *)&notif->user_position);
+	}
+}
+
+void iwl_mvm_mu_mimo_grp_notif(struct iwl_mvm *mvm,
+			       struct iwl_rx_cmd_buffer *rxb)
+{
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_mu_group_mgmt_notif *notif = (void *)pkt->data;
+
+	ieee80211_iterate_active_interfaces_atomic(
+			mvm->hw, IEEE80211_IFACE_ITER_NORMAL,
+			iwl_mvm_mu_mimo_iface_iterator, notif);
+}
+
 static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 					     struct ieee80211_vif *vif,
 					     struct ieee80211_bss_conf *bss_conf,
@@ -1890,10 +1919,10 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 
 		/*
 		 * The firmware tracks the MU-MIMO group on its own.
-		 * However, on HW restart we should restore this data
+		 * However, on HW restart we should restore this data.
 		 */
 		if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status) &&
-		    changes & BSS_CHANGED_MU_GROUPS) {
+		    (changes & BSS_CHANGED_MU_GROUPS) && vif->mu_mimo_owner) {
 			ret = iwl_mvm_update_mu_groups(mvm, vif);
 			if (ret)
 				IWL_ERR(mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index f05d2a1..21986f6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1272,6 +1272,8 @@ void iwl_mvm_rx_missed_beacons_notif(struct iwl_mvm *mvm,
 				     struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_rx_stored_beacon_notif(struct iwl_mvm *mvm,
 				    struct iwl_rx_cmd_buffer *rxb);
+void iwl_mvm_mu_mimo_grp_notif(struct iwl_mvm *mvm,
+			       struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_window_status_notif(struct iwl_mvm *mvm,
 				 struct iwl_rx_cmd_buffer *rxb);
 void iwl_mvm_mac_ctxt_recalc_tsf_id(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index ac271ff..8bdaa7b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -275,6 +275,8 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 	RX_HANDLER(TOF_NOTIFICATION, iwl_mvm_tof_resp_handler, true),
 	RX_HANDLER_GRP(PROT_OFFLOAD_GROUP, STORED_BEACON_NTF,
 		       iwl_mvm_rx_stored_beacon_notif, false),
+	RX_HANDLER_GRP(DATA_PATH_GROUP, MU_GROUP_MGMT_NOTIF,
+		       iwl_mvm_mu_mimo_grp_notif, false),
 
 };
 #undef RX_HANDLER
@@ -405,6 +407,7 @@ static const struct iwl_hcmd_names iwl_mvm_phy_names[] = {
 static const struct iwl_hcmd_names iwl_mvm_data_path_names[] = {
 	HCMD_NAME(UPDATE_MU_GROUPS_CMD),
 	HCMD_NAME(TRIGGER_RX_QUEUES_NOTIF_CMD),
+	HCMD_NAME(MU_GROUP_MGMT_NOTIF),
 	HCMD_NAME(RX_QUEUES_NOTIFICATION),
 };
 
-- 
cgit v0.10.2


From f1704618987b2710760582c87b5f7f600e529a60 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 25 Feb 2016 10:19:59 +0200
Subject: iwlwifi: mvm: various trivial cleanups

* Remove uneeded includes:
	iwl-csr.h and devcoredump aren't used in mac80211.c.
* Remove uneeded empty line

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index b755161..85ea5a8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -70,7 +70,6 @@
 #include <linux/etherdevice.h>
 #include <linux/ip.h>
 #include <linux/if_arp.h>
-#include <linux/devcoredump.h>
 #include <linux/time.h>
 #include <net/mac80211.h>
 #include <net/ieee80211_radiotap.h>
@@ -86,7 +85,6 @@
 #include "testmode.h"
 #include "iwl-fw-error-dump.h"
 #include "iwl-prph.h"
-#include "iwl-csr.h"
 #include "iwl-nvm-parse.h"
 #include "fw-dbg.h"
 
@@ -2142,7 +2140,6 @@ iwl_mvm_bss_info_changed_ap_ibss(struct iwl_mvm *mvm,
 				bss_conf->txpower);
 		iwl_mvm_set_tx_power(mvm, vif, bss_conf->txpower);
 	}
-
 }
 
 static void iwl_mvm_bss_info_changed(struct ieee80211_hw *hw,
-- 
cgit v0.10.2


From 21daff96a90e8bc2dfa57499c8305dbb9430f661 Mon Sep 17 00:00:00 2001
From: Beni Lev <beni.lev@intel.com>
Date: Wed, 23 Dec 2015 17:39:57 +0300
Subject: iwlwifi: mvm: Set global RRM capability

Allow to publish RRM capabilities without the need to support a minimal
capability set. Since some RRM features(e.g. neighbor report) are fw
independent, set this capability unconditionally.

Signed-off-by: Beni Lev <beni.lev@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 85ea5a8..4c2449b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -610,6 +610,8 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 			IWL_UCODE_TLV_CAPA_WFA_TPC_REP_IE_SUPPORT))
 		hw->wiphy->features |= NL80211_FEATURE_WFA_TPC_IE_IN_PROBES;
 
+	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_RRM);
+
 	mvm->rts_threshold = IEEE80211_MAX_RTS_THRESHOLD;
 
 #ifdef CONFIG_PM_SLEEP
-- 
cgit v0.10.2


From cee5a882d0273aac0d622368885c9aac676cce33 Mon Sep 17 00:00:00 2001
From: Avri Altman <avri.altman@intel.com>
Date: Wed, 3 Feb 2016 16:28:25 +0200
Subject: iwlwifi: mvm: forbid U-APSD for P2P Client if the firmware doesn't
 support it

Older versions of the firmware don't support U-APSD for
P2P Client. Forbid U-APSD for P2P Client when an old
firmware is being used.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 4c2449b..86b4b9d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2334,6 +2334,11 @@ static void iwl_mvm_check_uapsd(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_UAPSD_SUPPORT))
 		return;
 
+	if (vif->p2p && !iwl_mvm_is_p2p_standalone_uapsd_supported(mvm)) {
+		vif->driver_flags &= ~IEEE80211_VIF_SUPPORTS_UAPSD;
+		return;
+	}
+
 	if (iwlwifi_mod_params.uapsd_disable) {
 		vif->driver_flags &= ~IEEE80211_VIF_SUPPORTS_UAPSD;
 		return;
-- 
cgit v0.10.2


From 283115fb1dcf6b2162e988db98e156b998a5ebb3 Mon Sep 17 00:00:00 2001
From: Avri Altman <avri.altman@intel.com>
Date: Sun, 24 Jan 2016 22:16:08 +0200
Subject: iwlwifi: mvm: Send power command on BSS_CHANGED_BEACON_INFO if needed

Beacon abort (ba) is set while sending power command, but only
after at least one beacon_filter command was successfully sent.

If we heard a beacon before starting association, this order
is maintained and ba is properly set.

However, if the first beacon is received after association,
we send the power command upon association, configure the
beacon filtering when the first beacon arrives, and in that case,
beacon abort is not set.

So identify this, and send a power command post the beacon_filter
command if needed.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 86b4b9d..927b055 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1955,7 +1955,12 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 		WARN_ON(iwl_mvm_enable_beacon_filter(mvm, vif, 0));
 	}
 
-	if (changes & (BSS_CHANGED_PS | BSS_CHANGED_P2P_PS | BSS_CHANGED_QOS)) {
+	if (changes & (BSS_CHANGED_PS | BSS_CHANGED_P2P_PS | BSS_CHANGED_QOS |
+		       /*
+			* Send power command on every beacon change,
+			* because we may have not enabled beacon abort yet.
+			*/
+		       BSS_CHANGED_BEACON_INFO)) {
 		ret = iwl_mvm_power_update_mac(mvm);
 		if (ret)
 			IWL_ERR(mvm, "failed to update power mode\n");
-- 
cgit v0.10.2


From e29cc6b9c99639d9502bde0693afd3e49ec1ffd7 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Thu, 28 Jan 2016 14:25:33 +0200
Subject: iwlwifi: mvm: take care of padded packets

To ensure that the SNAP/TCP/IP headers are DW aligned, the firmware
may add 2-byte pad at the end of the mac header - after the IV, before
the SNAP.
In that case the mpdu descriptor pad bit will be turned on.
Driver should take it into consideration, and remove the padding before
passing the packet to mac80211. Do that.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 590fc6f..cd6ca37 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -7,7 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
- * Copyright(c) 2015        Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -29,7 +29,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
- * Copyright(c) 2015        Intel Deutschland GmbH
+ * Copyright(c) 2015 - 2016 Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -156,7 +156,14 @@ static void iwl_mvm_create_skb(struct sk_buff *skb, struct ieee80211_hdr *hdr,
 			       u16 len, u8 crypt_len,
 			       struct iwl_rx_cmd_buffer *rxb)
 {
-	unsigned int hdrlen, fraglen;
+	struct iwl_rx_packet *pkt = rxb_addr(rxb);
+	struct iwl_rx_mpdu_desc *desc = (void *)pkt->data;
+	unsigned int headlen, fraglen, pad_len = 0;
+	unsigned int hdrlen = ieee80211_hdrlen(hdr->frame_control);
+
+	if (desc->mac_flags2 & IWL_RX_MPDU_MFLG2_PAD)
+		pad_len = 2;
+	len -= pad_len;
 
 	/* If frame is small enough to fit in skb->head, pull it completely.
 	 * If not, only pull ieee80211_hdr (including crypto if present, and
@@ -170,14 +177,23 @@ static void iwl_mvm_create_skb(struct sk_buff *skb, struct ieee80211_hdr *hdr,
 	 * If the latter changes (there are efforts in the standards group
 	 * to do so) we should revisit this and ieee80211_data_to_8023().
 	 */
-	hdrlen = (len <= skb_tailroom(skb)) ? len :
-					      sizeof(*hdr) + crypt_len + 8;
+	headlen = (len <= skb_tailroom(skb)) ? len :
+					       hdrlen + crypt_len + 8;
 
+	/* The firmware may align the packet to DWORD.
+	 * The padding is inserted after the IV.
+	 * After copying the header + IV skip the padding if
+	 * present before copying packet data.
+	 */
+	hdrlen += crypt_len;
 	memcpy(skb_put(skb, hdrlen), hdr, hdrlen);
-	fraglen = len - hdrlen;
+	memcpy(skb_put(skb, headlen - hdrlen), (u8 *)hdr + hdrlen + pad_len,
+	       headlen - hdrlen);
+
+	fraglen = len - headlen;
 
 	if (fraglen) {
-		int offset = (void *)hdr + hdrlen -
+		int offset = (void *)hdr + headlen + pad_len -
 			     rxb_addr(rxb) + rxb_offset(rxb);
 
 		skb_add_rx_frag(skb, 0, rxb_steal_page(rxb), offset,
-- 
cgit v0.10.2


From eea76c364261aa4338f530c09a38afa69af41797 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 21 Feb 2016 16:29:17 +0200
Subject: iwlwifi: mvm: kill iwl_mvm_enable_agg_txq

iwl_mvm_enable_agg_txq has only one user. Kill it and
adapt the call site.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 21986f6..b461b90 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1549,22 +1549,6 @@ void iwl_mvm_enable_ac_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue,
 	iwl_mvm_enable_txq(mvm, queue, mac80211_queue, ssn, &cfg, wdg_timeout);
 }
 
-static inline void iwl_mvm_enable_agg_txq(struct iwl_mvm *mvm, int queue,
-					  int mac80211_queue, int fifo,
-					  int sta_id, int tid, int frame_limit,
-					  u16 ssn, unsigned int wdg_timeout)
-{
-	struct iwl_trans_txq_scd_cfg cfg = {
-		.fifo = fifo,
-		.sta_id = sta_id,
-		.tid = tid,
-		.frame_limit = frame_limit,
-		.aggregate = true,
-	};
-
-	iwl_mvm_enable_txq(mvm, queue, mac80211_queue, ssn, &cfg, wdg_timeout);
-}
-
 /* Thermal management and CT-kill */
 void iwl_mvm_tt_tx_backoff(struct iwl_mvm *mvm, u32 backoff);
 void iwl_mvm_tt_temp_changed(struct iwl_mvm *mvm, u32 temp);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 4717b18..ef99942 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -1052,9 +1052,16 @@ int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	struct iwl_mvm_tid_data *tid_data = &mvmsta->tid_data[tid];
 	unsigned int wdg_timeout =
 		iwl_mvm_get_wd_timeout(mvm, vif, sta->tdls, false);
-	int queue, fifo, ret;
+	int queue, ret;
 	u16 ssn;
 
+	struct iwl_trans_txq_scd_cfg cfg = {
+		.sta_id = mvmsta->sta_id,
+		.tid = tid,
+		.frame_limit = buf_size,
+		.aggregate = true,
+	};
+
 	BUILD_BUG_ON((sizeof(mvmsta->agg_tids) * BITS_PER_BYTE)
 		     != IWL_MAX_TID_COUNT);
 
@@ -1069,11 +1076,10 @@ int iwl_mvm_sta_tx_agg_oper(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	tid_data->amsdu_in_ampdu_allowed = amsdu;
 	spin_unlock_bh(&mvmsta->lock);
 
-	fifo = iwl_mvm_ac_to_tx_fifo[tid_to_mac80211_ac[tid]];
+	cfg.fifo = iwl_mvm_ac_to_tx_fifo[tid_to_mac80211_ac[tid]];
 
-	iwl_mvm_enable_agg_txq(mvm, queue,
-			       vif->hw_queue[tid_to_mac80211_ac[tid]], fifo,
-			       mvmsta->sta_id, tid, buf_size, ssn, wdg_timeout);
+	iwl_mvm_enable_txq(mvm, queue, vif->hw_queue[tid_to_mac80211_ac[tid]],
+			   ssn, &cfg, wdg_timeout);
 
 	ret = iwl_mvm_sta_tx_agg(mvm, sta, tid, queue, true);
 	if (ret)
-- 
cgit v0.10.2


From 81d90442eac779938217c3444b240aa51fd3db47 Mon Sep 17 00:00:00 2001
From: Dmitry Tunin <hanipouspilot@gmail.com>
Date: Sun, 28 Feb 2016 11:04:06 +0300
Subject: Bluetooth: btusb: Add a new AR3012 ID 04ca:3014

T: Bus=01 Lev=01 Prnt=01 Port=04 Cnt=03 Dev#= 5 Spd=12 MxCh= 0
D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=04ca ProdID=3014 Rev=00.02
C: #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb

BugLink: https://bugs.launchpad.net/bugs/1546694

Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Cc: stable@vger.kernel.org

diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c
index e2ccf90..9374738 100644
--- a/drivers/bluetooth/ath3k.c
+++ b/drivers/bluetooth/ath3k.c
@@ -93,6 +93,7 @@ static const struct usb_device_id ath3k_table[] = {
 	{ USB_DEVICE(0x04CA, 0x300d) },
 	{ USB_DEVICE(0x04CA, 0x300f) },
 	{ USB_DEVICE(0x04CA, 0x3010) },
+	{ USB_DEVICE(0x04CA, 0x3014) },
 	{ USB_DEVICE(0x0930, 0x0219) },
 	{ USB_DEVICE(0x0930, 0x021c) },
 	{ USB_DEVICE(0x0930, 0x0220) },
@@ -157,6 +158,7 @@ static const struct usb_device_id ath3k_blist_tbl[] = {
 	{ USB_DEVICE(0x04ca, 0x300d), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x300f), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3010), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x04ca, 0x3014), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x0219), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x021c), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x0220), .driver_info = BTUSB_ATH3012 },
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 55fbdfc..97f3bba 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -207,6 +207,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x04ca, 0x300d), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x300f), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3010), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x04ca, 0x3014), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x0219), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x021c), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x0220), .driver_info = BTUSB_ATH3012 },
-- 
cgit v0.10.2


From 32795a885ac31ccadd45e7886cc2ae31b2d2ef06 Mon Sep 17 00:00:00 2001
From: Matti Gottlieb <matti.gottlieb@intel.com>
Date: Thu, 25 Feb 2016 10:18:34 +0200
Subject: iwlwifi: mvm: Disable beacon storing in  D3 when WOWLAN configured

Currently when entering D3 with WOWLAN configured, we enable in the
configuration flags beacon storing, and do not disable beacon
filtering, and do not wake up from a magic packet.

Having both enabled is wrong (should not have both enabled),
and causes problems in the RX queues in the FW, causing
the FW not to recognize the magic packet when it comes.

Disable beacon storing in wowlan configuration.

Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index 5214482..cf5e634 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -856,8 +856,7 @@ iwl_mvm_get_wowlan_config(struct iwl_mvm *mvm,
 	wowlan_config_cmd->is_11n_connection =
 					ap_sta->ht_cap.ht_supported;
 	wowlan_config_cmd->flags = ENABLE_L3_FILTERING |
-		ENABLE_NBNS_FILTERING | ENABLE_DHCP_FILTERING |
-		ENABLE_STORE_BEACON;
+		ENABLE_NBNS_FILTERING | ENABLE_DHCP_FILTERING;
 
 	/* Query the last used seqno and set it */
 	ret = iwl_mvm_get_last_nonqos_seq(mvm, vif);
-- 
cgit v0.10.2


From 3d2d442236b3f87cf9ff70c4252488ed7c6d5be8 Mon Sep 17 00:00:00 2001
From: Golan Ben-Ami <golan.ben.ami@intel.com>
Date: Tue, 23 Feb 2016 10:34:48 +0200
Subject: iwlwifi: support ucode with d0 unified image - regular and usniffer

Till today, the ucode consisted of two d0 images - regular,
in which the usniffer wasn't enabled, and usniffer, in which the
usniffer logs were enabled.
Lately, the two images were unified, so there is only one d0 image,
in which the usniffer logs are enabled.

Add new TLV capability for supporting the consolidated images
(set 2, bit 13).

Signed-off-by: Golan Ben-Ami <golan.ben.ami@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index ab4c2a0..ee4ffa5 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -1033,7 +1033,8 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 		}
 	}
 
-	if (usniffer_req && !*usniffer_images) {
+	if (!fw_has_capa(capa, IWL_UCODE_TLV_CAPA_USNIFFER_UNIFIED) &&
+	    usniffer_req && !*usniffer_images) {
 		IWL_ERR(drv,
 			"user selected to work with usniffer but usniffer image isn't available in ucode package\n");
 		return -EINVAL;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 63dc109..5f69bf5 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -322,6 +322,8 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t;
  * @IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT: supports temperature
  *	thresholds reporting
  * @IWL_UCODE_TLV_CAPA_CTDP_SUPPORT: supports cTDP command
+ * @IWL_UCODE_TLV_CAPA_USNIFFER_UNIFIED: supports usniffer enabled in
+ *	regular image.
  *
  * @NUM_IWL_UCODE_TLV_CAPA: number of bits used
  */
@@ -358,6 +360,7 @@ enum iwl_ucode_tlv_capa {
 	IWL_UCODE_TLV_CAPA_CT_KILL_BY_FW		= (__force iwl_ucode_tlv_capa_t)74,
 	IWL_UCODE_TLV_CAPA_TEMP_THS_REPORT_SUPPORT	= (__force iwl_ucode_tlv_capa_t)75,
 	IWL_UCODE_TLV_CAPA_CTDP_SUPPORT			= (__force iwl_ucode_tlv_capa_t)76,
+	IWL_UCODE_TLV_CAPA_USNIFFER_UNIFIED		= (__force iwl_ucode_tlv_capa_t)77,
 
 	NUM_IWL_UCODE_TLV_CAPA
 #ifdef __CHECKER__
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 05475a2..f075c36 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -537,7 +537,9 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 	struct iwl_sf_region st_fwrd_space;
 
 	if (ucode_type == IWL_UCODE_REGULAR &&
-	    iwl_fw_dbg_conf_usniffer(mvm->fw, FW_DBG_START_FROM_ALIVE))
+	    iwl_fw_dbg_conf_usniffer(mvm->fw, FW_DBG_START_FROM_ALIVE) &&
+	    !(fw_has_capa(&mvm->fw->ucode_capa,
+			  IWL_UCODE_TLV_CAPA_USNIFFER_UNIFIED)))
 		fw = iwl_get_ucode_image(mvm, IWL_UCODE_REGULAR_USNIFFER);
 	else
 		fw = iwl_get_ucode_image(mvm, ucode_type);
-- 
cgit v0.10.2


From f12d33f4d83c6837d176e1aef337914089c77957 Mon Sep 17 00:00:00 2001
From: Stafford Horne <shorne@gmail.com>
Date: Sun, 28 Feb 2016 16:49:29 +0900
Subject: 3c59x: Ensure to apply the expires time

In commit 5b6490def9168af6a ("3c59x: Use setup_timer()") Amitoj
removed add_timer which sets up the epires timer.  In this patch
the behavior is restore but it uses mod_timer which is a bit more
compact.

Signed-off-by: Stafford Horne <shorne@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index c377607..7b881ed 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -1602,7 +1602,7 @@ vortex_up(struct net_device *dev)
 	}
 
 	setup_timer(&vp->timer, vortex_timer, (unsigned long)dev);
-	vp->timer.expires = RUN_AT(media_tbl[dev->if_port].wait);
+	mod_timer(&vp->timer, RUN_AT(media_tbl[dev->if_port].wait));
 	setup_timer(&vp->rx_oom_timer, rx_oom_timer, (unsigned long)dev);
 
 	if (vortex_debug > 1)
-- 
cgit v0.10.2


From 0744ff8fa8fad7aae669b73fd78f3efe1d8e31ca Mon Sep 17 00:00:00 2001
From: Linus Luessing <linus.luessing@web.de>
Date: Sat, 16 Jan 2016 16:40:08 +0800
Subject: batman-adv: Add hard_iface specific sysfs wrapper macros for UINT

This allows us to easily add a sysfs parameter for an
unsigned int later, which is not for a batman mesh interface
(e.g. bat0), but for a common interface instead. It allows
reading and writing an atomic_t in hard_iface (instead of
bat_priv compared to the mesh variant).

Developed by Linus during a 6 months trainee study period in
Ascom (Switzerland) AG.

Signed-off-by: Linus Luessing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
[antonio@open-mesh.com: rename functions and move macros]
Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>

diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index 4d70d44..d3f69d5 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -242,6 +242,55 @@ ssize_t batadv_show_vlan_##_name(struct kobject *kobj,			\
 	static BATADV_ATTR_VLAN(_name, _mode, batadv_show_vlan_##_name,	\
 				batadv_store_vlan_##_name)
 
+#define BATADV_ATTR_HIF_STORE_UINT(_name, _var, _min, _max, _post_func)	\
+ssize_t batadv_store_##_name(struct kobject *kobj,			\
+			     struct attribute *attr, char *buff,	\
+			     size_t count)				\
+{									\
+	struct net_device *net_dev = batadv_kobj_to_netdev(kobj);	\
+	struct batadv_hard_iface *hard_iface;				\
+	ssize_t length;							\
+									\
+	hard_iface = batadv_hardif_get_by_netdev(net_dev);		\
+	if (!hard_iface)						\
+		return 0;						\
+									\
+	length = __batadv_store_uint_attr(buff, count, _min, _max,	\
+					  _post_func, attr,		\
+					  &hard_iface->_var, net_dev);	\
+									\
+	batadv_hardif_put(hard_iface);				\
+	return length;							\
+}
+
+#define BATADV_ATTR_HIF_SHOW_UINT(_name, _var)				\
+ssize_t batadv_show_##_name(struct kobject *kobj,			\
+			    struct attribute *attr, char *buff)		\
+{									\
+	struct net_device *net_dev = batadv_kobj_to_netdev(kobj);	\
+	struct batadv_hard_iface *hard_iface;				\
+	ssize_t length;							\
+									\
+	hard_iface = batadv_hardif_get_by_netdev(net_dev);		\
+	if (!hard_iface)						\
+		return 0;						\
+									\
+	length = sprintf(buff, "%i\n", atomic_read(&hard_iface->_var));	\
+									\
+	batadv_hardif_put(hard_iface);				\
+	return length;							\
+}
+
+/* Use this, if you are going to set [name] in hard_iface to an
+ * unsigned integer value
+ */
+#define BATADV_ATTR_HIF_UINT(_name, _var, _mode, _min, _max, _post_func)\
+	static BATADV_ATTR_HIF_STORE_UINT(_name, _var, _min,		\
+					  _max, _post_func)		\
+	static BATADV_ATTR_HIF_SHOW_UINT(_name, _var)			\
+	static BATADV_ATTR(_name, _mode, batadv_show_##_name,		\
+			   batadv_store_##_name)
+
 static int batadv_store_bool_attr(char *buff, size_t count,
 				  struct net_device *net_dev,
 				  const char *attr_name, atomic_t *attr,
-- 
cgit v0.10.2


From d6f94d91f766b4205e5b0aa4b11f96271c793f6d Mon Sep 17 00:00:00 2001
From: Linus Luessing <linus.luessing@web.de>
Date: Sat, 16 Jan 2016 16:40:09 +0800
Subject: batman-adv: ELP - adding basic infrastructure

The B.A.T.M.A.N. protocol originally only used a single
message type (called OGM) to determine the link qualities to
the direct neighbors and spreading these link quality
information through the whole mesh. This procedure is
summarized on the BATMAN concept page and explained in
details in the RFC draft published in 2008.

This approach was chosen for its simplicity during the
protocol design phase and the implementation. However, it
also bears some drawbacks:

 *  Wireless interfaces usually come with some packet loss,
    therefore a higher broadcast rate is desirable to allow
    a fast reaction on flaky connections.
    Other interfaces of the same host might be connected to
    Ethernet LANs / VPNs / etc which rarely exhibit packet
    loss would benefit from a lower broadcast rate to reduce
    overhead.
 *  It generally is more desirable to detect local link
    quality changes at a faster rate than propagating all
    these changes through the entire mesh (the far end of
    the mesh does not need to care about local link quality
    changes that much). Other optimizations strategies, like
    reducing overhead, might be possible if OGMs weren't
    used for all tasks in the mesh at the same time.

As a result detecting local link qualities shall be handled
by an independent message type, ELP, whereas the OGM message
type remains responsible for flooding the mesh with these
link quality information and determining the overall path
transmit qualities.

Developed by Linus during a 6 months trainee study period in
Ascom (Switzerland) AG.

Signed-off-by: Linus Luessing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 2dd40e5..5c148a8 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -15,6 +15,20 @@ config BATMAN_ADV
           https://www.open-mesh.org/ for more information and user space
           tools.
 
+config BATMAN_ADV_BATMAN_V
+	bool "B.A.T.M.A.N. V protocol (experimental)"
+	depends on BATMAN_ADV
+	default n
+	help
+	  This option enables the B.A.T.M.A.N. V protocol, the successor
+	  of the currently used B.A.T.M.A.N. IV protocol. The main
+	  changes include splitting of the OGM protocol into a neighbor
+	  discovery protocol (Echo Location Protocol, ELP) and a new OGM
+	  Protocol OGMv2 for flooding protocol information through the
+	  network, as well as a throughput based metric.
+	  B.A.T.M.A.N. V is currently considered experimental and not
+	  compatible to B.A.T.M.A.N. IV networks.
+
 config BATMAN_ADV_BLA
 	bool "Bridge Loop Avoidance"
 	depends on BATMAN_ADV && INET
diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index 207e2af..ca51686 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -18,6 +18,8 @@
 
 obj-$(CONFIG_BATMAN_ADV) += batman-adv.o
 batman-adv-y += bat_iv_ogm.o
+batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v.o
+batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v_elp.o
 batman-adv-y += bitarray.o
 batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o
 batman-adv-$(CONFIG_DEBUG_FS) += debugfs.o
diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h
index a7485d6..a4e994e 100644
--- a/net/batman-adv/bat_algo.h
+++ b/net/batman-adv/bat_algo.h
@@ -1,6 +1,6 @@
 /* Copyright (C) 2011-2016  B.A.T.M.A.N. contributors:
  *
- * Marek Lindner
+ * Marek Lindner, Linus Lüssing
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of version 2 of the GNU General Public
@@ -20,4 +20,17 @@
 
 int batadv_iv_init(void);
 
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+
+int batadv_v_init(void);
+
+#else
+
+static inline int batadv_v_init(void)
+{
+	return 0;
+}
+
+#endif /* CONFIG_BATMAN_ADV_BATMAN_V */
+
 #endif /* _NET_BATMAN_ADV_BAT_ALGO_H_ */
diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
new file mode 100644
index 0000000..01327f6
--- /dev/null
+++ b/net/batman-adv/bat_v.c
@@ -0,0 +1,74 @@
+/* Copyright (C) 2013-2016 B.A.T.M.A.N. contributors:
+ *
+ * Linus Lüssing, Marek Lindner
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "bat_algo.h"
+#include "main.h"
+
+#include <linux/cache.h>
+#include <linux/init.h>
+
+#include "bat_v_elp.h"
+
+static int batadv_v_iface_enable(struct batadv_hard_iface *hard_iface)
+{
+	return batadv_v_elp_iface_enable(hard_iface);
+}
+
+static void batadv_v_iface_disable(struct batadv_hard_iface *hard_iface)
+{
+	batadv_v_elp_iface_disable(hard_iface);
+}
+
+static void batadv_v_iface_update_mac(struct batadv_hard_iface *hard_iface)
+{
+}
+
+static void batadv_v_primary_iface_set(struct batadv_hard_iface *hard_iface)
+{
+	batadv_v_elp_primary_iface_set(hard_iface);
+}
+
+static void batadv_v_ogm_schedule(struct batadv_hard_iface *hard_iface)
+{
+}
+
+static void batadv_v_ogm_emit(struct batadv_forw_packet *forw_packet)
+{
+}
+
+static struct batadv_algo_ops batadv_batman_v __read_mostly = {
+	.name = "BATMAN_V",
+	.bat_iface_enable = batadv_v_iface_enable,
+	.bat_iface_disable = batadv_v_iface_disable,
+	.bat_iface_update_mac = batadv_v_iface_update_mac,
+	.bat_primary_iface_set = batadv_v_primary_iface_set,
+	.bat_ogm_emit = batadv_v_ogm_emit,
+	.bat_ogm_schedule = batadv_v_ogm_schedule,
+};
+
+/**
+ * batadv_v_init - B.A.T.M.A.N. V initialization function
+ *
+ * Description: Takes care of initializing all the subcomponents.
+ * It is invoked upon module load only.
+ *
+ * Return: 0 on success or a negative error code otherwise
+ */
+int __init batadv_v_init(void)
+{
+	return batadv_algo_register(&batadv_batman_v);
+}
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
new file mode 100644
index 0000000..bc6e046
--- /dev/null
+++ b/net/batman-adv/bat_v_elp.c
@@ -0,0 +1,193 @@
+/* Copyright (C) 2011-2016 B.A.T.M.A.N. contributors:
+ *
+ * Linus Lüssing, Marek Lindner
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "bat_v_elp.h"
+#include "main.h"
+
+#include <linux/atomic.h>
+#include <linux/byteorder/generic.h>
+#include <linux/errno.h>
+#include <linux/etherdevice.h>
+#include <linux/fs.h>
+#include <linux/if_ether.h>
+#include <linux/jiffies.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/random.h>
+#include <linux/rculist.h>
+#include <linux/rcupdate.h>
+#include <linux/skbuff.h>
+#include <linux/stddef.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+
+#include "bat_algo.h"
+#include "hard-interface.h"
+#include "packet.h"
+#include "send.h"
+
+/**
+ * batadv_v_elp_start_timer - restart timer for ELP periodic work
+ * @hard_iface: the interface for which the timer has to be reset
+ */
+static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface)
+{
+	unsigned int msecs;
+
+	msecs = atomic_read(&hard_iface->bat_v.elp_interval) - BATADV_JITTER;
+	msecs += prandom_u32() % (2 * BATADV_JITTER);
+
+	queue_delayed_work(batadv_event_workqueue, &hard_iface->bat_v.elp_wq,
+			   msecs_to_jiffies(msecs));
+}
+
+/**
+ * batadv_v_elp_periodic_work - ELP periodic task per interface
+ * @work: work queue item
+ *
+ * Emits broadcast ELP message in regular intervals.
+ */
+static void batadv_v_elp_periodic_work(struct work_struct *work)
+{
+	struct batadv_hard_iface *hard_iface;
+	struct batadv_hard_iface_bat_v *bat_v;
+	struct batadv_elp_packet *elp_packet;
+	struct batadv_priv *bat_priv;
+	struct sk_buff *skb;
+	u32 elp_interval;
+
+	bat_v = container_of(work, struct batadv_hard_iface_bat_v, elp_wq.work);
+	hard_iface = container_of(bat_v, struct batadv_hard_iface, bat_v);
+	bat_priv = netdev_priv(hard_iface->soft_iface);
+
+	if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_DEACTIVATING)
+		goto out;
+
+	/* we are in the process of shutting this interface down */
+	if ((hard_iface->if_status == BATADV_IF_NOT_IN_USE) ||
+	    (hard_iface->if_status == BATADV_IF_TO_BE_REMOVED))
+		goto out;
+
+	/* the interface was enabled but may not be ready yet */
+	if (hard_iface->if_status != BATADV_IF_ACTIVE)
+		goto restart_timer;
+
+	skb = skb_copy(hard_iface->bat_v.elp_skb, GFP_ATOMIC);
+	if (!skb)
+		goto restart_timer;
+
+	elp_packet = (struct batadv_elp_packet *)skb->data;
+	elp_packet->seqno = htonl(atomic_read(&hard_iface->bat_v.elp_seqno));
+	elp_interval = atomic_read(&hard_iface->bat_v.elp_interval);
+	elp_packet->elp_interval = htonl(elp_interval);
+
+	batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+		   "Sending broadcast ELP packet on interface %s, seqno %u\n",
+		   hard_iface->net_dev->name,
+		   atomic_read(&hard_iface->bat_v.elp_seqno));
+
+	batadv_send_skb_packet(skb, hard_iface, batadv_broadcast_addr);
+
+	atomic_inc(&hard_iface->bat_v.elp_seqno);
+
+restart_timer:
+	batadv_v_elp_start_timer(hard_iface);
+out:
+	return;
+}
+
+/**
+ * batadv_v_elp_iface_enable - setup the ELP interface private resources
+ * @hard_iface: interface for which the data has to be prepared
+ *
+ * Return: 0 on success or a -ENOMEM in case of failure.
+ */
+int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface)
+{
+	struct batadv_elp_packet *elp_packet;
+	unsigned char *elp_buff;
+	u32 random_seqno;
+	size_t size;
+	int res = -ENOMEM;
+
+	size = ETH_HLEN + NET_IP_ALIGN + BATADV_ELP_HLEN;
+	hard_iface->bat_v.elp_skb = dev_alloc_skb(size);
+	if (!hard_iface->bat_v.elp_skb)
+		goto out;
+
+	skb_reserve(hard_iface->bat_v.elp_skb, ETH_HLEN + NET_IP_ALIGN);
+	elp_buff = skb_push(hard_iface->bat_v.elp_skb, BATADV_ELP_HLEN);
+	elp_packet = (struct batadv_elp_packet *)elp_buff;
+	memset(elp_packet, 0, BATADV_ELP_HLEN);
+
+	elp_packet->packet_type = BATADV_ELP;
+	elp_packet->version = BATADV_COMPAT_VERSION;
+
+	/* randomize initial seqno to avoid collision */
+	get_random_bytes(&random_seqno, sizeof(random_seqno));
+	atomic_set(&hard_iface->bat_v.elp_seqno, random_seqno);
+	atomic_set(&hard_iface->bat_v.elp_interval, 500);
+
+	INIT_DELAYED_WORK(&hard_iface->bat_v.elp_wq,
+			  batadv_v_elp_periodic_work);
+	batadv_v_elp_start_timer(hard_iface);
+	res = 0;
+
+out:
+	return res;
+}
+
+/**
+ * batadv_v_elp_iface_disable - release ELP interface private resources
+ * @hard_iface: interface for which the resources have to be released
+ */
+void batadv_v_elp_iface_disable(struct batadv_hard_iface *hard_iface)
+{
+	cancel_delayed_work_sync(&hard_iface->bat_v.elp_wq);
+
+	dev_kfree_skb(hard_iface->bat_v.elp_skb);
+	hard_iface->bat_v.elp_skb = NULL;
+}
+
+/**
+ * batadv_v_elp_primary_iface_set - change internal data to reflect the new
+ *  primary interface
+ * @primary_iface: the new primary interface
+ */
+void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface)
+{
+	struct batadv_hard_iface *hard_iface;
+	struct batadv_elp_packet *elp_packet;
+	struct sk_buff *skb;
+
+	/* update orig field of every elp iface belonging to this mesh */
+	rcu_read_lock();
+	list_for_each_entry_rcu(hard_iface, &batadv_hardif_list, list) {
+		if (primary_iface->soft_iface != hard_iface->soft_iface)
+			continue;
+
+		if (!hard_iface->bat_v.elp_skb)
+			continue;
+
+		skb = hard_iface->bat_v.elp_skb;
+		elp_packet = (struct batadv_elp_packet *)skb->data;
+		ether_addr_copy(elp_packet->orig,
+				primary_iface->net_dev->dev_addr);
+	}
+	rcu_read_unlock();
+}
diff --git a/net/batman-adv/bat_v_elp.h b/net/batman-adv/bat_v_elp.h
new file mode 100644
index 0000000..7a584d1
--- /dev/null
+++ b/net/batman-adv/bat_v_elp.h
@@ -0,0 +1,27 @@
+/* Copyright (C) 2013-2016 B.A.T.M.A.N. contributors:
+ *
+ * Linus Lüssing, Marek Lindner
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "main.h"
+
+#ifndef _NET_BATMAN_ADV_BAT_V_ELP_H_
+#define _NET_BATMAN_ADV_BAT_V_ELP_H_
+
+int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface);
+void batadv_v_elp_iface_disable(struct batadv_hard_iface *hard_iface);
+void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface);
+
+#endif /* _NET_BATMAN_ADV_BAT_V_ELP_H_ */
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index e3d7051..2160572 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -87,6 +87,7 @@ static int __init batadv_init(void)
 
 	batadv_recv_handler_init();
 
+	batadv_v_init();
 	batadv_iv_init();
 	batadv_nc_init();
 
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index e7f9151..c698b64 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -26,6 +26,7 @@
  * @BATADV_IV_OGM: originator messages for B.A.T.M.A.N. IV
  * @BATADV_BCAST: broadcast packets carrying broadcast payload
  * @BATADV_CODED: network coded packets
+ * @BATADV_ELP: echo location packets for B.A.T.M.A.N. V
  *
  * @BATADV_UNICAST: unicast packets carrying unicast payload traffic
  * @BATADV_UNICAST_FRAG: unicast packets carrying a fragment of the original
@@ -40,6 +41,7 @@ enum batadv_packettype {
 	BATADV_IV_OGM           = 0x00,
 	BATADV_BCAST            = 0x01,
 	BATADV_CODED            = 0x02,
+	BATADV_ELP		= 0x03,
 	/* 0x40 - 0x7f: unicast */
 #define BATADV_UNICAST_MIN     0x40
 	BATADV_UNICAST          = 0x40,
@@ -235,6 +237,24 @@ struct batadv_ogm_packet {
 #define BATADV_OGM_HLEN sizeof(struct batadv_ogm_packet)
 
 /**
+ * struct batadv_elp_packet - elp (neighbor discovery) packet
+ * @packet_type: batman-adv packet type, part of the general header
+ * @version: batman-adv protocol version, part of the genereal header
+ * @orig: originator mac address
+ * @seqno: sequence number
+ * @elp_interval: currently used ELP sending interval in ms
+ */
+struct batadv_elp_packet {
+	u8     packet_type;
+	u8     version;
+	u8     orig[ETH_ALEN];
+	__be32 seqno;
+	__be32 elp_interval;
+};
+
+#define BATADV_ELP_HLEN sizeof(struct batadv_elp_packet)
+
+/**
  * struct batadv_icmp_header - common members among all the ICMP packets
  * @packet_type: batman-adv packet type, part of the general header
  * @version: batman-adv protocol version, part of the genereal header
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 612de23..992d5fd 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -86,6 +86,20 @@ struct batadv_hard_iface_bat_iv {
 };
 
 /**
+ * struct batadv_hard_iface_bat_v - per hard-interface B.A.T.M.A.N. V data
+ * @elp_interval: time interval between two ELP transmissions
+ * @elp_seqno: current ELP sequence number
+ * @elp_skb: base skb containing the ELP message to send
+ * @elp_wq: workqueue used to schedule ELP transmissions
+ */
+struct batadv_hard_iface_bat_v {
+	atomic_t elp_interval;
+	atomic_t elp_seqno;
+	struct sk_buff *elp_skb;
+	struct delayed_work elp_wq;
+};
+
+/**
  * struct batadv_hard_iface - network device known to batman-adv
  * @list: list node for batadv_hardif_list
  * @if_num: identificator of the interface
@@ -99,6 +113,7 @@ struct batadv_hard_iface_bat_iv {
  * @soft_iface: the batman-adv interface which uses this network interface
  * @rcu: struct used for freeing in an RCU-safe manner
  * @bat_iv: per hard-interface B.A.T.M.A.N. IV data
+ * @bat_v: per hard-interface B.A.T.M.A.N. V data
  * @cleanup_work: work queue callback item for hard-interface deinit
  * @debug_dir: dentry for nc subdir in batman-adv directory in debugfs
  * @neigh_list: list of unique single hop neighbors via this interface
@@ -116,6 +131,9 @@ struct batadv_hard_iface {
 	struct net_device *soft_iface;
 	struct rcu_head rcu;
 	struct batadv_hard_iface_bat_iv bat_iv;
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	struct batadv_hard_iface_bat_v bat_v;
+#endif
 	struct work_struct cleanup_work;
 	struct dentry *debug_dir;
 	struct hlist_head neigh_list;
-- 
cgit v0.10.2


From 162bd64c24aba8efe68948e95e61628403106cd7 Mon Sep 17 00:00:00 2001
From: Linus Luessing <linus.luessing@web.de>
Date: Sat, 16 Jan 2016 16:40:10 +0800
Subject: batman-adv: ELP - creating neighbor structures

Initially developed by Linus during a 6 months trainee study
period in Ascom (Switzerland) AG.

Signed-off-by: Linus Luessing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>

diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index 01327f6..0fea8948 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -22,6 +22,7 @@
 #include <linux/init.h>
 
 #include "bat_v_elp.h"
+#include "packet.h"
 
 static int batadv_v_iface_enable(struct batadv_hard_iface *hard_iface)
 {
@@ -42,6 +43,12 @@ static void batadv_v_primary_iface_set(struct batadv_hard_iface *hard_iface)
 	batadv_v_elp_primary_iface_set(hard_iface);
 }
 
+static void
+batadv_v_hardif_neigh_init(struct batadv_hardif_neigh_node *hardif_neigh)
+{
+	ewma_throughput_init(&hardif_neigh->bat_v.throughput);
+}
+
 static void batadv_v_ogm_schedule(struct batadv_hard_iface *hard_iface)
 {
 }
@@ -56,6 +63,7 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = {
 	.bat_iface_disable = batadv_v_iface_disable,
 	.bat_iface_update_mac = batadv_v_iface_update_mac,
 	.bat_primary_iface_set = batadv_v_primary_iface_set,
+	.bat_hardif_neigh_init = batadv_v_hardif_neigh_init,
 	.bat_ogm_emit = batadv_v_ogm_emit,
 	.bat_ogm_schedule = batadv_v_ogm_schedule,
 };
@@ -70,5 +78,18 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = {
  */
 int __init batadv_v_init(void)
 {
-	return batadv_algo_register(&batadv_batman_v);
+	int ret;
+
+	/* B.A.T.M.A.N. V echo location protocol packet  */
+	ret = batadv_recv_handler_register(BATADV_ELP,
+					   batadv_v_elp_packet_recv);
+	if (ret < 0)
+		return ret;
+
+	ret = batadv_algo_register(&batadv_batman_v);
+
+	if (ret < 0)
+		batadv_recv_handler_unregister(BATADV_ELP);
+
+	return ret;
 }
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index bc6e046..dac88fa 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -38,7 +38,10 @@
 
 #include "bat_algo.h"
 #include "hard-interface.h"
+#include "hash.h"
+#include "originator.h"
 #include "packet.h"
+#include "routing.h"
 #include "send.h"
 
 /**
@@ -191,3 +194,152 @@ void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface)
 	}
 	rcu_read_unlock();
 }
+
+/**
+ * batadv_v_ogm_orig_get - retrieve and possibly create an originator node
+ * @bat_priv: the bat priv with all the soft interface information
+ * @addr: the address of the originator
+ *
+ * Return: the orig_node corresponding to the specified address. If such object
+ * does not exist it is allocated here. In case of allocation failure returns
+ * NULL.
+ */
+static struct batadv_orig_node *
+batadv_v_ogm_orig_get(struct batadv_priv *bat_priv,
+		      const u8 *addr)
+{
+	struct batadv_orig_node *orig_node;
+	int hash_added;
+
+	orig_node = batadv_orig_hash_find(bat_priv, addr);
+	if (orig_node)
+		return orig_node;
+
+	orig_node = batadv_orig_node_new(bat_priv, addr);
+	if (!orig_node)
+		return NULL;
+
+	hash_added = batadv_hash_add(bat_priv->orig_hash, batadv_compare_orig,
+				     batadv_choose_orig, orig_node,
+				     &orig_node->hash_entry);
+	if (hash_added != 0) {
+		/* orig_node->refcounter is initialised to 2 by
+		 * batadv_orig_node_new()
+		 */
+		batadv_orig_node_put(orig_node);
+		batadv_orig_node_put(orig_node);
+		orig_node = NULL;
+	}
+
+	return orig_node;
+}
+
+/**
+ * batadv_v_elp_neigh_update - update an ELP neighbour node
+ * @bat_priv: the bat priv with all the soft interface information
+ * @neigh_addr: the neighbour interface address
+ * @if_incoming: the interface the packet was received through
+ * @elp_packet: the received ELP packet
+ *
+ * Updates the ELP neighbour node state with the data received within the new
+ * ELP packet.
+ */
+static void batadv_v_elp_neigh_update(struct batadv_priv *bat_priv,
+				      u8 *neigh_addr,
+				      struct batadv_hard_iface *if_incoming,
+				      struct batadv_elp_packet *elp_packet)
+
+{
+	struct batadv_neigh_node *neigh;
+	struct batadv_orig_node *orig_neigh;
+	struct batadv_hardif_neigh_node *hardif_neigh;
+	s32 seqno_diff;
+	s32 elp_latest_seqno;
+
+	orig_neigh = batadv_v_ogm_orig_get(bat_priv, elp_packet->orig);
+	if (!orig_neigh)
+		return;
+
+	neigh = batadv_neigh_node_new(orig_neigh, if_incoming, neigh_addr);
+	if (!neigh)
+		goto orig_free;
+
+	hardif_neigh = batadv_hardif_neigh_get(if_incoming, neigh_addr);
+	if (!hardif_neigh)
+		goto neigh_free;
+
+	elp_latest_seqno = hardif_neigh->bat_v.elp_latest_seqno;
+	seqno_diff = ntohl(elp_packet->seqno) - elp_latest_seqno;
+
+	/* known or older sequence numbers are ignored. However always adopt
+	 * if the router seems to have been restarted.
+	 */
+	if (seqno_diff < 1 && seqno_diff > -BATADV_ELP_MAX_AGE)
+		goto hardif_free;
+
+	neigh->last_seen = jiffies;
+	hardif_neigh->last_seen = jiffies;
+	hardif_neigh->bat_v.elp_latest_seqno = ntohl(elp_packet->seqno);
+	hardif_neigh->bat_v.elp_interval = ntohl(elp_packet->elp_interval);
+
+hardif_free:
+	if (hardif_neigh)
+		batadv_hardif_neigh_put(hardif_neigh);
+neigh_free:
+	if (neigh)
+		batadv_neigh_node_put(neigh);
+orig_free:
+	if (orig_neigh)
+		batadv_orig_node_put(orig_neigh);
+}
+
+/**
+ * batadv_v_elp_packet_recv - main ELP packet handler
+ * @skb: the received packet
+ * @if_incoming: the interface this packet was received through
+ *
+ * Return: NET_RX_SUCCESS and consumes the skb if the packet was peoperly
+ * processed or NET_RX_DROP in case of failure.
+ */
+int batadv_v_elp_packet_recv(struct sk_buff *skb,
+			     struct batadv_hard_iface *if_incoming)
+{
+	struct batadv_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
+	struct batadv_elp_packet *elp_packet;
+	struct batadv_hard_iface *primary_if;
+	struct ethhdr *ethhdr = (struct ethhdr *)skb_mac_header(skb);
+	bool ret;
+
+	ret = batadv_check_management_packet(skb, if_incoming, BATADV_ELP_HLEN);
+	if (!ret)
+		return NET_RX_DROP;
+
+	if (batadv_is_my_mac(bat_priv, ethhdr->h_source))
+		return NET_RX_DROP;
+
+	/* did we receive a B.A.T.M.A.N. V ELP packet on an interface
+	 * that does not have B.A.T.M.A.N. V ELP enabled ?
+	 */
+	if (strcmp(bat_priv->bat_algo_ops->name, "BATMAN_V") != 0)
+		return NET_RX_DROP;
+
+	elp_packet = (struct batadv_elp_packet *)skb->data;
+
+	batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+		   "Received ELP packet from %pM seqno %u ORIG: %pM\n",
+		   ethhdr->h_source, ntohl(elp_packet->seqno),
+		   elp_packet->orig);
+
+	primary_if = batadv_primary_if_get_selected(bat_priv);
+	if (!primary_if)
+		goto out;
+
+	batadv_v_elp_neigh_update(bat_priv, ethhdr->h_source, if_incoming,
+				  elp_packet);
+
+out:
+	if (primary_if)
+		batadv_hardif_put(primary_if);
+	consume_skb(skb);
+	return NET_RX_SUCCESS;
+}
diff --git a/net/batman-adv/bat_v_elp.h b/net/batman-adv/bat_v_elp.h
index 7a584d1..5a7bc39 100644
--- a/net/batman-adv/bat_v_elp.h
+++ b/net/batman-adv/bat_v_elp.h
@@ -20,8 +20,12 @@
 #ifndef _NET_BATMAN_ADV_BAT_V_ELP_H_
 #define _NET_BATMAN_ADV_BAT_V_ELP_H_
 
+struct sk_buff;
+
 int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface);
 void batadv_v_elp_iface_disable(struct batadv_hard_iface *hard_iface);
 void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface);
+int batadv_v_elp_packet_recv(struct sk_buff *skb,
+			     struct batadv_hard_iface *if_incoming);
 
 #endif /* _NET_BATMAN_ADV_BAT_V_ELP_H_ */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 8c01f54..d91bae2 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -60,6 +60,9 @@
 #define BATADV_TQ_LOCAL_BIDRECT_RECV_MINIMUM 1
 #define BATADV_TQ_TOTAL_BIDRECT_LIMIT 1
 
+/* B.A.T.M.A.N. V */
+#define BATADV_ELP_MAX_AGE 64
+
 /* number of OGMs sent with the last tt diff */
 #define BATADV_TT_OGM_APPEND_MAX 3
 
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 992d5fd..7fc82a5 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -22,6 +22,7 @@
 #error only "main.h" can be included directly
 #endif
 
+#include <linux/average.h>
 #include <linux/bitops.h>
 #include <linux/compiler.h>
 #include <linux/if_ether.h>
@@ -364,12 +365,28 @@ struct batadv_gw_node {
 	struct rcu_head rcu;
 };
 
+DECLARE_EWMA(throughput, 1024, 8)
+
+/**
+ * struct batadv_hardif_neigh_node_bat_v - B.A.T.M.A.N. V private neighbor
+ *  information
+ * @throughput: ewma link throughput towards this neighbor
+ * @elp_interval: time interval between two ELP transmissions
+ * @elp_latest_seqno: latest and best known ELP sequence number
+ */
+struct batadv_hardif_neigh_node_bat_v {
+	struct ewma_throughput throughput;
+	u32 elp_interval;
+	u32 elp_latest_seqno;
+};
+
 /**
  * struct batadv_hardif_neigh_node - unique neighbor per hard-interface
  * @list: list node for batadv_hard_iface::neigh_list
  * @addr: the MAC address of the neighboring interface
  * @if_incoming: pointer to incoming hard-interface
  * @last_seen: when last packet via this neighbor was received
+ * @bat_v: B.A.T.M.A.N. V private data
  * @refcount: number of contexts the object is used
  * @rcu: struct used for freeing in a RCU-safe manner
  */
@@ -378,6 +395,9 @@ struct batadv_hardif_neigh_node {
 	u8 addr[ETH_ALEN];
 	struct batadv_hard_iface *if_incoming;
 	unsigned long last_seen;
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	struct batadv_hardif_neigh_node_bat_v bat_v;
+#endif
 	struct kref refcount;
 	struct rcu_head rcu;
 };
@@ -425,10 +445,20 @@ struct batadv_neigh_ifinfo_bat_iv {
 };
 
 /**
+ * struct batadv_neigh_ifinfo_bat_v - neighbor information per outgoing
+ *  interface for B.A.T.M.A.N. V
+ * @throughput: last throughput metric received from originator via this neigh
+ */
+struct batadv_neigh_ifinfo_bat_v {
+	u32 throughput;
+};
+
+/**
  * struct batadv_neigh_ifinfo - neighbor information per outgoing interface
  * @list: list node for batadv_neigh_node::ifinfo_list
  * @if_outgoing: pointer to outgoing hard-interface
  * @bat_iv: B.A.T.M.A.N. IV private structure
+ * @bat_v: B.A.T.M.A.N. V private data
  * @last_ttl: last received ttl from this neigh node
  * @refcount: number of contexts the object is used
  * @rcu: struct used for freeing in a RCU-safe manner
@@ -437,6 +467,9 @@ struct batadv_neigh_ifinfo {
 	struct hlist_node list;
 	struct batadv_hard_iface *if_outgoing;
 	struct batadv_neigh_ifinfo_bat_iv bat_iv;
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	struct batadv_neigh_ifinfo_bat_v bat_v;
+#endif
 	u8 last_ttl;
 	struct kref refcount;
 	struct rcu_head rcu;
-- 
cgit v0.10.2


From 7f136cd491013285442ee1e7854fab1736f5757c Mon Sep 17 00:00:00 2001
From: Linus Luessing <linus.luessing@web.de>
Date: Sat, 16 Jan 2016 16:40:11 +0800
Subject: batman-adv: ELP - adding sysfs parameter for elp interval

This parameter can be set individually on each interface and
allows the configuration of the elp interval for the link
quality measurements during runtime. Usually it is desirable
to set it to a higher (= slower) value on interfaces which
have a more static characteristic (e.g. wired interfaces)
or very dense neighbourhoods to reduce overhead.

Developed by Linus during a 6 months trainee study period in
Ascom (Switzerland) AG.

Signed-off-by: Linus Luessing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
[antonio@open-mesh.com: respin on top of the latest master]
Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>

diff --git a/Documentation/ABI/testing/sysfs-class-net-batman-adv b/Documentation/ABI/testing/sysfs-class-net-batman-adv
index 7f34a95..aea78db 100644
--- a/Documentation/ABI/testing/sysfs-class-net-batman-adv
+++ b/Documentation/ABI/testing/sysfs-class-net-batman-adv
@@ -1,4 +1,11 @@
 
+What:           /sys/class/net/<iface>/batman-adv/elp_interval
+Date:           Feb 2014
+Contact:        Linus Lüssing <linus.luessing@web.de>
+Description:
+                Defines the interval in milliseconds in which batman
+                sends its probing packets for link quality measurements.
+
 What:           /sys/class/net/<iface>/batman-adv/iface_status
 Date:           May 2010
 Contact:        Marek Lindner <mareklindner@neomailbox.ch>
@@ -12,4 +19,3 @@ Description:
                 The /sys/class/net/<iface>/batman-adv/mesh_iface file
                 displays the batman mesh interface this <iface>
                 currently is associated with.
-
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index d3f69d5..e860143 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -920,10 +920,17 @@ static ssize_t batadv_show_iface_status(struct kobject *kobj,
 static BATADV_ATTR(mesh_iface, S_IRUGO | S_IWUSR, batadv_show_mesh_iface,
 		   batadv_store_mesh_iface);
 static BATADV_ATTR(iface_status, S_IRUGO, batadv_show_iface_status, NULL);
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+BATADV_ATTR_HIF_UINT(elp_interval, bat_v.elp_interval, S_IRUGO | S_IWUSR,
+		     2 * BATADV_JITTER, INT_MAX, NULL);
+#endif
 
 static struct batadv_attribute *batadv_batman_attrs[] = {
 	&batadv_attr_mesh_iface,
 	&batadv_attr_iface_status,
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	&batadv_attr_elp_interval,
+#endif
 	NULL,
 };
 
-- 
cgit v0.10.2


From 0da0035942d47766c32843143fb5dba7a29cb48c Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Sat, 16 Jan 2016 16:40:12 +0800
Subject: batman-adv: OGMv2 - add basic infrastructure

This is the initial implementation of the new OGM protocol
(version 2). It has been designed to work on top of the
newly added ELP.

In the previous version the OGM protocol was used to both
measure link qualities and flood the network with the metric
information. In this version the protocol is in charge of
the latter task only, leaving the former to ELP.

This means being able to decouple the interval used by the
neighbor discovery from the OGM broadcasting, which revealed
to be costly in dense networks and needed to be relaxed so
leading to a less responsive routing protocol.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile
index ca51686..797cf2f 100644
--- a/net/batman-adv/Makefile
+++ b/net/batman-adv/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_BATMAN_ADV) += batman-adv.o
 batman-adv-y += bat_iv_ogm.o
 batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v.o
 batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v_elp.o
+batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v_ogm.o
 batman-adv-y += bitarray.o
 batman-adv-$(CONFIG_BATMAN_ADV_BLA) += bridge_loop_avoidance.o
 batman-adv-$(CONFIG_DEBUG_FS) += debugfs.o
diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h
index a4e994e..03dafd3 100644
--- a/net/batman-adv/bat_algo.h
+++ b/net/batman-adv/bat_algo.h
@@ -18,11 +18,15 @@
 #ifndef _NET_BATMAN_ADV_BAT_ALGO_H_
 #define _NET_BATMAN_ADV_BAT_ALGO_H_
 
+struct batadv_priv;
+
 int batadv_iv_init(void);
 
 #ifdef CONFIG_BATMAN_ADV_BATMAN_V
 
 int batadv_v_init(void);
+int batadv_v_mesh_init(struct batadv_priv *bat_priv);
+void batadv_v_mesh_free(struct batadv_priv *bat_priv);
 
 #else
 
@@ -31,6 +35,15 @@ static inline int batadv_v_init(void)
 	return 0;
 }
 
+static inline int batadv_v_mesh_init(struct batadv_priv *bat_priv)
+{
+	return 0;
+}
+
+static inline void batadv_v_mesh_free(struct batadv_priv *bat_priv)
+{
+}
+
 #endif /* CONFIG_BATMAN_ADV_BATMAN_V */
 
 #endif /* _NET_BATMAN_ADV_BAT_ALGO_H_ */
diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index 0fea8948..b90a4df 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -22,11 +22,22 @@
 #include <linux/init.h>
 
 #include "bat_v_elp.h"
+#include "bat_v_ogm.h"
 #include "packet.h"
 
 static int batadv_v_iface_enable(struct batadv_hard_iface *hard_iface)
 {
-	return batadv_v_elp_iface_enable(hard_iface);
+	int ret;
+
+	ret = batadv_v_elp_iface_enable(hard_iface);
+	if (ret < 0)
+		return ret;
+
+	ret = batadv_v_ogm_iface_enable(hard_iface);
+	if (ret < 0)
+		batadv_v_elp_iface_disable(hard_iface);
+
+	return ret;
 }
 
 static void batadv_v_iface_disable(struct batadv_hard_iface *hard_iface)
@@ -41,6 +52,7 @@ static void batadv_v_iface_update_mac(struct batadv_hard_iface *hard_iface)
 static void batadv_v_primary_iface_set(struct batadv_hard_iface *hard_iface)
 {
 	batadv_v_elp_primary_iface_set(hard_iface);
+	batadv_v_ogm_primary_iface_set(hard_iface);
 }
 
 static void
@@ -69,6 +81,27 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = {
 };
 
 /**
+ * batadv_v_mesh_init - initialize the B.A.T.M.A.N. V private resources for a
+ *  mesh
+ * @bat_priv: the object representing the mesh interface to initialise
+ *
+ * Return: 0 on success or a negative error code otherwise
+ */
+int batadv_v_mesh_init(struct batadv_priv *bat_priv)
+{
+	return batadv_v_ogm_init(bat_priv);
+}
+
+/**
+ * batadv_v_mesh_free - free the B.A.T.M.A.N. V private resources for a mesh
+ * @bat_priv: the object representing the mesh interface to free
+ */
+void batadv_v_mesh_free(struct batadv_priv *bat_priv)
+{
+	batadv_v_ogm_free(bat_priv);
+}
+
+/**
  * batadv_v_init - B.A.T.M.A.N. V initialization function
  *
  * Description: Takes care of initializing all the subcomponents.
@@ -86,10 +119,22 @@ int __init batadv_v_init(void)
 	if (ret < 0)
 		return ret;
 
-	ret = batadv_algo_register(&batadv_batman_v);
+	ret = batadv_recv_handler_register(BATADV_OGM2,
+					   batadv_v_ogm_packet_recv);
+	if (ret < 0)
+		goto elp_unregister;
 
+	ret = batadv_algo_register(&batadv_batman_v);
 	if (ret < 0)
-		batadv_recv_handler_unregister(BATADV_ELP);
+		goto ogm_unregister;
+
+	return ret;
+
+ogm_unregister:
+	batadv_recv_handler_unregister(BATADV_OGM2);
+
+elp_unregister:
+	batadv_recv_handler_unregister(BATADV_ELP);
 
 	return ret;
 }
diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c
new file mode 100644
index 0000000..23932f9
--- /dev/null
+++ b/net/batman-adv/bat_v_ogm.c
@@ -0,0 +1,279 @@
+/* Copyright (C) 2013-2016 B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "bat_v_ogm.h"
+#include "main.h"
+
+#include <linux/atomic.h>
+#include <linux/byteorder/generic.h>
+#include <linux/errno.h>
+#include <linux/etherdevice.h>
+#include <linux/fs.h>
+#include <linux/if_ether.h>
+#include <linux/jiffies.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/random.h>
+#include <linux/rculist.h>
+#include <linux/rcupdate.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <linux/stddef.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+
+#include "hard-interface.h"
+#include "packet.h"
+#include "routing.h"
+#include "send.h"
+#include "translation-table.h"
+
+/**
+ * batadv_v_ogm_start_timer - restart the OGM sending timer
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+static void batadv_v_ogm_start_timer(struct batadv_priv *bat_priv)
+{
+	unsigned long msecs;
+	/* this function may be invoked in different contexts (ogm rescheduling
+	 * or hard_iface activation), but the work timer should not be reset
+	 */
+	if (delayed_work_pending(&bat_priv->bat_v.ogm_wq))
+		return;
+
+	msecs = atomic_read(&bat_priv->orig_interval) - BATADV_JITTER;
+	msecs += prandom_u32() % (2 * BATADV_JITTER);
+	queue_delayed_work(batadv_event_workqueue, &bat_priv->bat_v.ogm_wq,
+			   msecs_to_jiffies(msecs));
+}
+
+/**
+ * batadv_v_ogm_send_to_if - send a batman ogm using a given interface
+ * @skb: the OGM to send
+ * @hard_iface: the interface to use to send the OGM
+ */
+static void batadv_v_ogm_send_to_if(struct sk_buff *skb,
+				    struct batadv_hard_iface *hard_iface)
+{
+	struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
+
+	if (hard_iface->if_status != BATADV_IF_ACTIVE)
+		return;
+
+	batadv_inc_counter(bat_priv, BATADV_CNT_MGMT_TX);
+	batadv_add_counter(bat_priv, BATADV_CNT_MGMT_TX_BYTES,
+			   skb->len + ETH_HLEN);
+
+	batadv_send_skb_packet(skb, hard_iface, batadv_broadcast_addr);
+}
+
+/**
+ * batadv_v_ogm_send - periodic worker broadcasting the own OGM
+ * @work: work queue item
+ */
+static void batadv_v_ogm_send(struct work_struct *work)
+{
+	struct batadv_hard_iface *hard_iface;
+	struct batadv_priv_bat_v *bat_v;
+	struct batadv_priv *bat_priv;
+	struct batadv_ogm2_packet *ogm_packet;
+	struct sk_buff *skb, *skb_tmp;
+	unsigned char *ogm_buff, *pkt_buff;
+	int ogm_buff_len;
+	u16 tvlv_len = 0;
+
+	bat_v = container_of(work, struct batadv_priv_bat_v, ogm_wq.work);
+	bat_priv = container_of(bat_v, struct batadv_priv, bat_v);
+
+	if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_DEACTIVATING)
+		goto out;
+
+	ogm_buff = bat_priv->bat_v.ogm_buff;
+	ogm_buff_len = bat_priv->bat_v.ogm_buff_len;
+	/* tt changes have to be committed before the tvlv data is
+	 * appended as it may alter the tt tvlv container
+	 */
+	batadv_tt_local_commit_changes(bat_priv);
+	tvlv_len = batadv_tvlv_container_ogm_append(bat_priv, &ogm_buff,
+						    &ogm_buff_len,
+						    BATADV_OGM2_HLEN);
+
+	bat_priv->bat_v.ogm_buff = ogm_buff;
+	bat_priv->bat_v.ogm_buff_len = ogm_buff_len;
+
+	skb = netdev_alloc_skb_ip_align(NULL, ETH_HLEN + ogm_buff_len);
+	if (!skb)
+		goto reschedule;
+
+	skb_reserve(skb, ETH_HLEN);
+	pkt_buff = skb_put(skb, ogm_buff_len);
+	memcpy(pkt_buff, ogm_buff, ogm_buff_len);
+
+	ogm_packet = (struct batadv_ogm2_packet *)skb->data;
+	ogm_packet->seqno = htonl(atomic_read(&bat_priv->bat_v.ogm_seqno));
+	atomic_inc(&bat_priv->bat_v.ogm_seqno);
+	ogm_packet->tvlv_len = htons(tvlv_len);
+
+	/* broadcast on every interface */
+	rcu_read_lock();
+	list_for_each_entry_rcu(hard_iface, &batadv_hardif_list, list) {
+		if (hard_iface->soft_iface != bat_priv->soft_iface)
+			continue;
+
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Sending own OGM2 packet (originator %pM, seqno %u, throughput %u, TTL %d) on interface %s [%pM]\n",
+			   ogm_packet->orig, ntohl(ogm_packet->seqno),
+			   ntohl(ogm_packet->throughput), ogm_packet->ttl,
+			   hard_iface->net_dev->name,
+			   hard_iface->net_dev->dev_addr);
+
+		/* this skb gets consumed by batadv_v_ogm_send_to_if() */
+		skb_tmp = skb_clone(skb, GFP_ATOMIC);
+		if (!skb_tmp)
+			break;
+
+		batadv_v_ogm_send_to_if(skb_tmp, hard_iface);
+	}
+	rcu_read_unlock();
+
+	consume_skb(skb);
+
+reschedule:
+	batadv_v_ogm_start_timer(bat_priv);
+out:
+	return;
+}
+
+/**
+ * batadv_v_ogm_iface_enable - prepare an interface for B.A.T.M.A.N. V
+ * @hard_iface: the interface to prepare
+ *
+ * Takes care of scheduling own OGM sending routine for this interface.
+ *
+ * Return: 0 on success or a negative error code otherwise
+ */
+int batadv_v_ogm_iface_enable(struct batadv_hard_iface *hard_iface)
+{
+	struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
+
+	batadv_v_ogm_start_timer(bat_priv);
+
+	return 0;
+}
+
+/**
+ * batadv_v_ogm_primary_iface_set - set a new primary interface
+ * @primary_iface: the new primary interface
+ */
+void batadv_v_ogm_primary_iface_set(struct batadv_hard_iface *primary_iface)
+{
+	struct batadv_priv *bat_priv = netdev_priv(primary_iface->soft_iface);
+	struct batadv_ogm2_packet *ogm_packet;
+
+	if (!bat_priv->bat_v.ogm_buff)
+		return;
+
+	ogm_packet = (struct batadv_ogm2_packet *)bat_priv->bat_v.ogm_buff;
+	ether_addr_copy(ogm_packet->orig, primary_iface->net_dev->dev_addr);
+}
+
+/**
+ * batadv_v_ogm_packet_recv - OGM2 receiving handler
+ * @skb: the received OGM
+ * @if_incoming: the interface where this OGM has been received
+ *
+ * Return: NET_RX_SUCCESS and consume the skb on success or returns NET_RX_DROP
+ * (without freeing the skb) on failure
+ */
+int batadv_v_ogm_packet_recv(struct sk_buff *skb,
+			     struct batadv_hard_iface *if_incoming)
+{
+	struct batadv_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
+	struct batadv_ogm2_packet *ogm_packet;
+	struct ethhdr *ethhdr = eth_hdr(skb);
+
+	/* did we receive a OGM2 packet on an interface that does not have
+	 * B.A.T.M.A.N. V enabled ?
+	 */
+	if (strcmp(bat_priv->bat_algo_ops->name, "BATMAN_V") != 0)
+		return NET_RX_DROP;
+
+	if (!batadv_check_management_packet(skb, if_incoming, BATADV_OGM2_HLEN))
+		return NET_RX_DROP;
+
+	if (batadv_is_my_mac(bat_priv, ethhdr->h_source))
+		return NET_RX_DROP;
+
+	ogm_packet = (struct batadv_ogm2_packet *)skb->data;
+
+	if (batadv_is_my_mac(bat_priv, ogm_packet->orig))
+		return NET_RX_DROP;
+
+	batadv_inc_counter(bat_priv, BATADV_CNT_MGMT_RX);
+	batadv_add_counter(bat_priv, BATADV_CNT_MGMT_RX_BYTES,
+			   skb->len + ETH_HLEN);
+
+	consume_skb(skb);
+	return NET_RX_SUCCESS;
+}
+
+/**
+ * batadv_v_ogm_init - initialise the OGM2 engine
+ * @bat_priv: the bat priv with all the soft interface information
+ *
+ * Return: 0 on success or a negative error code in case of failure
+ */
+int batadv_v_ogm_init(struct batadv_priv *bat_priv)
+{
+	struct batadv_ogm2_packet *ogm_packet;
+	unsigned char *ogm_buff;
+	u32 random_seqno;
+
+	bat_priv->bat_v.ogm_buff_len = BATADV_OGM2_HLEN;
+	ogm_buff = kzalloc(bat_priv->bat_v.ogm_buff_len, GFP_ATOMIC);
+	if (!ogm_buff)
+		return -ENOMEM;
+
+	bat_priv->bat_v.ogm_buff = ogm_buff;
+	ogm_packet = (struct batadv_ogm2_packet *)ogm_buff;
+	ogm_packet->packet_type = BATADV_OGM2;
+	ogm_packet->version = BATADV_COMPAT_VERSION;
+	ogm_packet->ttl = BATADV_TTL;
+	ogm_packet->flags = BATADV_NO_FLAGS;
+	ogm_packet->throughput = htonl(BATADV_THROUGHPUT_MAX_VALUE);
+
+	/* randomize initial seqno to avoid collision */
+	get_random_bytes(&random_seqno, sizeof(random_seqno));
+	atomic_set(&bat_priv->bat_v.ogm_seqno, random_seqno);
+	INIT_DELAYED_WORK(&bat_priv->bat_v.ogm_wq, batadv_v_ogm_send);
+
+	return 0;
+}
+
+/**
+ * batadv_v_ogm_free - free OGM private resources
+ * @bat_priv: the bat priv with all the soft interface information
+ */
+void batadv_v_ogm_free(struct batadv_priv *bat_priv)
+{
+	cancel_delayed_work_sync(&bat_priv->bat_v.ogm_wq);
+
+	kfree(bat_priv->bat_v.ogm_buff);
+	bat_priv->bat_v.ogm_buff = NULL;
+	bat_priv->bat_v.ogm_buff_len = 0;
+}
diff --git a/net/batman-adv/bat_v_ogm.h b/net/batman-adv/bat_v_ogm.h
new file mode 100644
index 0000000..d9f178c
--- /dev/null
+++ b/net/batman-adv/bat_v_ogm.h
@@ -0,0 +1,32 @@
+/* Copyright (C) 2013-2016 B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef _BATMAN_ADV_BATADV_V_OGM_H_
+#define _BATMAN_ADV_BATADV_V_OGM_H_
+
+struct batadv_hard_iface;
+struct batadv_priv;
+struct sk_buff;
+
+int batadv_v_ogm_init(struct batadv_priv *bat_priv);
+void batadv_v_ogm_free(struct batadv_priv *bat_priv);
+int batadv_v_ogm_iface_enable(struct batadv_hard_iface *hard_iface);
+void batadv_v_ogm_primary_iface_set(struct batadv_hard_iface *primary_iface);
+int batadv_v_ogm_packet_recv(struct sk_buff *skb,
+			     struct batadv_hard_iface *if_incoming);
+
+#endif /* _BATMAN_ADV_BATADV_V_OGM_H_ */
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 2160572..d64ddb9 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -160,6 +160,10 @@ int batadv_mesh_init(struct net_device *soft_iface)
 	INIT_HLIST_HEAD(&bat_priv->tvlv.handler_list);
 	INIT_HLIST_HEAD(&bat_priv->softif_vlan_list);
 
+	ret = batadv_v_mesh_init(bat_priv);
+	if (ret < 0)
+		goto err;
+
 	ret = batadv_originator_init(bat_priv);
 	if (ret < 0)
 		goto err;
@@ -202,6 +206,8 @@ void batadv_mesh_free(struct net_device *soft_iface)
 	batadv_purge_outstanding_packets(bat_priv, NULL);
 
 	batadv_gw_node_free(bat_priv);
+
+	batadv_v_mesh_free(bat_priv);
 	batadv_nc_mesh_free(bat_priv);
 	batadv_dat_free(bat_priv);
 	batadv_bla_free(bat_priv);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index d91bae2..fa0b067 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -30,6 +30,7 @@
 /* B.A.T.M.A.N. parameters */
 
 #define BATADV_TQ_MAX_VALUE 255
+#define BATADV_THROUGHPUT_MAX_VALUE 0xFFFFFFFF
 #define BATADV_JITTER 20
 
 /* Time To Live of broadcast messages */
diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h
index c698b64..8a8d7ca 100644
--- a/net/batman-adv/packet.h
+++ b/net/batman-adv/packet.h
@@ -27,6 +27,7 @@
  * @BATADV_BCAST: broadcast packets carrying broadcast payload
  * @BATADV_CODED: network coded packets
  * @BATADV_ELP: echo location packets for B.A.T.M.A.N. V
+ * @BATADV_OGM2: originator messages for B.A.T.M.A.N. V
  *
  * @BATADV_UNICAST: unicast packets carrying unicast payload traffic
  * @BATADV_UNICAST_FRAG: unicast packets carrying a fragment of the original
@@ -42,6 +43,7 @@ enum batadv_packettype {
 	BATADV_BCAST            = 0x01,
 	BATADV_CODED            = 0x02,
 	BATADV_ELP		= 0x03,
+	BATADV_OGM2		= 0x04,
 	/* 0x40 - 0x7f: unicast */
 #define BATADV_UNICAST_MIN     0x40
 	BATADV_UNICAST          = 0x40,
@@ -237,6 +239,33 @@ struct batadv_ogm_packet {
 #define BATADV_OGM_HLEN sizeof(struct batadv_ogm_packet)
 
 /**
+ * struct batadv_ogm2_packet - ogm2 (routing protocol) packet
+ * @packet_type: batman-adv packet type, part of the general header
+ * @version: batman-adv protocol version, part of the general header
+ * @ttl: time to live for this packet, part of the general header
+ * @flags: reseved for routing relevant flags - currently always 0
+ * @seqno: sequence number
+ * @orig: originator mac address
+ * @tvlv_len: length of the appended tvlv buffer (in bytes)
+ * @throughput: the currently flooded path throughput
+ */
+struct batadv_ogm2_packet {
+	u8     packet_type;
+	u8     version;
+	u8     ttl;
+	u8     flags;
+	__be32 seqno;
+	u8     orig[ETH_ALEN];
+	__be16 tvlv_len;
+	__be32 throughput;
+	/* __packed is not needed as the struct size is divisible by 4,
+	 * and the largest data type in this struct has a size of 4.
+	 */
+};
+
+#define BATADV_OGM2_HLEN sizeof(struct batadv_ogm2_packet)
+
+/**
  * struct batadv_elp_packet - elp (neighbor discovery) packet
  * @packet_type: batman-adv packet type, part of the general header
  * @version: batman-adv protocol version, part of the genereal header
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 7fc82a5..e213ccd 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -802,6 +802,20 @@ struct batadv_softif_vlan {
 };
 
 /**
+ * struct batadv_priv_bat_v - B.A.T.M.A.N. V per soft-interface private data
+ * @ogm_buff: buffer holding the OGM packet
+ * @ogm_buff_len: length of the OGM packet buffer
+ * @ogm_seqno: OGM sequence number - used to identify each OGM
+ * @ogm_wq: workqueue used to schedule OGM transmissions
+ */
+struct batadv_priv_bat_v {
+	unsigned char *ogm_buff;
+	int ogm_buff_len;
+	atomic_t ogm_seqno;
+	struct delayed_work ogm_wq;
+};
+
+/**
  * struct batadv_priv - per mesh interface data
  * @mesh_state: current status of the mesh (inactive/active/deactivating)
  * @soft_iface: net device which holds this struct as private data
@@ -855,6 +869,7 @@ struct batadv_softif_vlan {
  * @mcast: multicast data
  * @network_coding: bool indicating whether network coding is enabled
  * @nc: network coding data
+ * @bat_v: B.A.T.M.A.N. V per soft-interface private data
  */
 struct batadv_priv {
 	atomic_t mesh_state;
@@ -920,6 +935,9 @@ struct batadv_priv {
 	atomic_t network_coding;
 	struct batadv_priv_nc nc;
 #endif /* CONFIG_BATMAN_ADV_NC */
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	struct batadv_priv_bat_v bat_v;
+#endif
 };
 
 /**
-- 
cgit v0.10.2


From 9323158ef9f49935f0c61509919acd31dda8f11b Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Sat, 16 Jan 2016 16:40:13 +0800
Subject: batman-adv: OGMv2 - implement originators logic

Add the support for recognising new originators in the
network and rebroadcast their OGMs.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index dac88fa..fac5aee 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -37,8 +37,8 @@
 #include <linux/workqueue.h>
 
 #include "bat_algo.h"
+#include "bat_v_ogm.h"
 #include "hard-interface.h"
-#include "hash.h"
 #include "originator.h"
 #include "packet.h"
 #include "routing.h"
@@ -196,45 +196,6 @@ void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface)
 }
 
 /**
- * batadv_v_ogm_orig_get - retrieve and possibly create an originator node
- * @bat_priv: the bat priv with all the soft interface information
- * @addr: the address of the originator
- *
- * Return: the orig_node corresponding to the specified address. If such object
- * does not exist it is allocated here. In case of allocation failure returns
- * NULL.
- */
-static struct batadv_orig_node *
-batadv_v_ogm_orig_get(struct batadv_priv *bat_priv,
-		      const u8 *addr)
-{
-	struct batadv_orig_node *orig_node;
-	int hash_added;
-
-	orig_node = batadv_orig_hash_find(bat_priv, addr);
-	if (orig_node)
-		return orig_node;
-
-	orig_node = batadv_orig_node_new(bat_priv, addr);
-	if (!orig_node)
-		return NULL;
-
-	hash_added = batadv_hash_add(bat_priv->orig_hash, batadv_compare_orig,
-				     batadv_choose_orig, orig_node,
-				     &orig_node->hash_entry);
-	if (hash_added != 0) {
-		/* orig_node->refcounter is initialised to 2 by
-		 * batadv_orig_node_new()
-		 */
-		batadv_orig_node_put(orig_node);
-		batadv_orig_node_put(orig_node);
-		orig_node = NULL;
-	}
-
-	return orig_node;
-}
-
-/**
  * batadv_v_elp_neigh_update - update an ELP neighbour node
  * @bat_priv: the bat priv with all the soft interface information
  * @neigh_addr: the neighbour interface address
diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c
index 23932f9..22dda0e 100644
--- a/net/batman-adv/bat_v_ogm.c
+++ b/net/batman-adv/bat_v_ogm.c
@@ -26,6 +26,7 @@
 #include <linux/if_ether.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/netdevice.h>
 #include <linux/random.h>
 #include <linux/rculist.h>
@@ -38,12 +39,52 @@
 #include <linux/workqueue.h>
 
 #include "hard-interface.h"
+#include "hash.h"
+#include "originator.h"
 #include "packet.h"
 #include "routing.h"
 #include "send.h"
 #include "translation-table.h"
 
 /**
+ * batadv_v_ogm_orig_get - retrieve and possibly create an originator node
+ * @bat_priv: the bat priv with all the soft interface information
+ * @addr: the address of the originator
+ *
+ * Return: the orig_node corresponding to the specified address. If such object
+ * does not exist it is allocated here. In case of allocation failure returns
+ * NULL.
+ */
+struct batadv_orig_node *batadv_v_ogm_orig_get(struct batadv_priv *bat_priv,
+					       const u8 *addr)
+{
+	struct batadv_orig_node *orig_node;
+	int hash_added;
+
+	orig_node = batadv_orig_hash_find(bat_priv, addr);
+	if (orig_node)
+		return orig_node;
+
+	orig_node = batadv_orig_node_new(bat_priv, addr);
+	if (!orig_node)
+		return NULL;
+
+	hash_added = batadv_hash_add(bat_priv->orig_hash, batadv_compare_orig,
+				     batadv_choose_orig, orig_node,
+				     &orig_node->hash_entry);
+	if (hash_added != 0) {
+		/* orig_node->refcounter is initialised to 2 by
+		 * batadv_orig_node_new()
+		 */
+		batadv_orig_node_put(orig_node);
+		batadv_orig_node_put(orig_node);
+		orig_node = NULL;
+	}
+
+	return orig_node;
+}
+
+/**
  * batadv_v_ogm_start_timer - restart the OGM sending timer
  * @bat_priv: the bat priv with all the soft interface information
  */
@@ -193,6 +234,500 @@ void batadv_v_ogm_primary_iface_set(struct batadv_hard_iface *primary_iface)
 }
 
 /**
+ * batadv_v_ogm_orig_update - update the originator status based on the received
+ *  OGM
+ * @bat_priv: the bat priv with all the soft interface information
+ * @orig_node: the originator to update
+ * @neigh_node: the neighbour the OGM has been received from (to update)
+ * @ogm2: the received OGM
+ * @if_outgoing: the interface where this OGM is going to be forwarded through
+ */
+static void
+batadv_v_ogm_orig_update(struct batadv_priv *bat_priv,
+			 struct batadv_orig_node *orig_node,
+			 struct batadv_neigh_node *neigh_node,
+			 const struct batadv_ogm2_packet *ogm2,
+			 struct batadv_hard_iface *if_outgoing)
+{
+	struct batadv_neigh_ifinfo *router_ifinfo = NULL, *neigh_ifinfo = NULL;
+	struct batadv_neigh_node *router = NULL;
+	s32 neigh_seq_diff;
+	u32 neigh_last_seqno;
+	u32 router_last_seqno;
+	u32 router_throughput, neigh_throughput;
+
+	batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+		   "Searching and updating originator entry of received packet\n");
+
+	/* if this neighbor already is our next hop there is nothing
+	 * to change
+	 */
+	router = batadv_orig_router_get(orig_node, if_outgoing);
+	if (router == neigh_node)
+		goto out;
+
+	/* don't consider neighbours with worse throughput.
+	 * also switch route if this seqno is BATADV_V_MAX_ORIGDIFF newer than
+	 * the last received seqno from our best next hop.
+	 */
+	if (router) {
+		router_ifinfo = batadv_neigh_ifinfo_get(router, if_outgoing);
+		neigh_ifinfo = batadv_neigh_ifinfo_get(neigh_node, if_outgoing);
+
+		/* if these are not allocated, something is wrong. */
+		if (!router_ifinfo || !neigh_ifinfo)
+			goto out;
+
+		neigh_last_seqno = neigh_ifinfo->bat_v.last_seqno;
+		router_last_seqno = router_ifinfo->bat_v.last_seqno;
+		neigh_seq_diff = neigh_last_seqno - router_last_seqno;
+		router_throughput = router_ifinfo->bat_v.throughput;
+		neigh_throughput = neigh_ifinfo->bat_v.throughput;
+
+		if ((neigh_seq_diff < BATADV_OGM_MAX_ORIGDIFF) &&
+		    (router_throughput >= neigh_throughput))
+			goto out;
+	}
+
+	batadv_update_route(bat_priv, orig_node, if_outgoing, neigh_node);
+
+out:
+	if (router_ifinfo)
+		batadv_neigh_ifinfo_put(router_ifinfo);
+	if (neigh_ifinfo)
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
+	if (router)
+		batadv_neigh_node_put(router);
+}
+
+/**
+ * batadv_v_forward_penalty - apply a penalty to the throughput metric forwarded
+ *  with B.A.T.M.A.N. V OGMs
+ * @bat_priv: the bat priv with all the soft interface information
+ * @if_incoming: the interface where the OGM has been received
+ * @if_outgoing: the interface where the OGM has to be forwarded to
+ * @throughput: the current throughput
+ *
+ * Apply a penalty on the current throughput metric value based on the
+ * characteristic of the interface where the OGM has been received. The return
+ * value is computed as follows:
+ * - throughput * 50%          if the incoming and outgoing interface are the
+ *                             same WiFi interface and the throughput is above
+ *                             1MBit/s
+ * - throughput                if the outgoing interface is the default
+ *                             interface (i.e. this OGM is processed for the
+ *                             internal table and not forwarded)
+ * - throughput * hop penalty  otherwise
+ *
+ * Return: the penalised throughput metric.
+ */
+static u32 batadv_v_forward_penalty(struct batadv_priv *bat_priv,
+				    struct batadv_hard_iface *if_incoming,
+				    struct batadv_hard_iface *if_outgoing,
+				    u32 throughput)
+{
+	int hop_penalty = atomic_read(&bat_priv->hop_penalty);
+	int hop_penalty_max = BATADV_TQ_MAX_VALUE;
+
+	/* Don't apply hop penalty in default originator table. */
+	if (if_outgoing == BATADV_IF_DEFAULT)
+		return throughput;
+
+	/* Forwarding on the same WiFi interface cuts the throughput in half
+	 * due to the store & forward characteristics of WIFI.
+	 * Very low throughput values are the exception.
+	 */
+	if ((throughput > 10) &&
+	    (if_incoming == if_outgoing) &&
+	    (batadv_is_wifi_netdev(if_incoming->net_dev)))
+		return throughput / 2;
+
+	/* hop penalty of 255 equals 100% */
+	return throughput * (hop_penalty_max - hop_penalty) / hop_penalty_max;
+}
+
+/**
+ * batadv_v_ogm_forward - forward an OGM to the given outgoing interface
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ogm_received: previously received OGM to be forwarded
+ * @throughput: throughput to announce, may vary per outgoing interface
+ * @if_incoming: the interface on which this OGM was received on
+ * @if_outgoing: the interface to which the OGM has to be forwarded to
+ *
+ * Forward an OGM to an interface after having altered the throughput metric and
+ * the TTL value contained in it. The original OGM isn't modified.
+ */
+static void batadv_v_ogm_forward(struct batadv_priv *bat_priv,
+				 const struct batadv_ogm2_packet *ogm_received,
+				 u32 throughput,
+				 struct batadv_hard_iface *if_incoming,
+				 struct batadv_hard_iface *if_outgoing)
+{
+	struct batadv_ogm2_packet *ogm_forward;
+	unsigned char *skb_buff;
+	struct sk_buff *skb;
+	size_t packet_len;
+	u16 tvlv_len;
+
+	if (ogm_received->ttl <= 1) {
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv, "ttl exceeded\n");
+		return;
+	}
+
+	tvlv_len = ntohs(ogm_received->tvlv_len);
+
+	packet_len = BATADV_OGM2_HLEN + tvlv_len;
+	skb = netdev_alloc_skb_ip_align(if_outgoing->net_dev,
+					ETH_HLEN + packet_len);
+	if (!skb)
+		return;
+
+	skb_reserve(skb, ETH_HLEN);
+	skb_buff = skb_put(skb, packet_len);
+	memcpy(skb_buff, ogm_received, packet_len);
+
+	/* apply forward penalty */
+	ogm_forward = (struct batadv_ogm2_packet *)skb_buff;
+	ogm_forward->throughput = htonl(throughput);
+	ogm_forward->ttl--;
+
+	batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+		   "Forwarding OGM2 packet on %s: throughput %u, ttl %u, received via %s\n",
+		   if_outgoing->net_dev->name, throughput, ogm_forward->ttl,
+		   if_incoming->net_dev->name);
+
+	batadv_v_ogm_send_to_if(skb, if_outgoing);
+}
+
+/**
+ * batadv_v_ogm_metric_update - update route metric based on OGM
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ogm2: OGM2 structure
+ * @orig_node: Originator structure for which the OGM has been received
+ * @neigh_node: the neigh_node through with the OGM has been received
+ * @if_incoming: the interface where this packet was received
+ * @if_outgoing: the interface for which the packet should be considered
+ *
+ * Return:
+ *  1  if the OGM is new,
+ *  0  if it is not new but valid,
+ *  <0 on error (e.g. old OGM)
+ */
+static int batadv_v_ogm_metric_update(struct batadv_priv *bat_priv,
+				      const struct batadv_ogm2_packet *ogm2,
+				      struct batadv_orig_node *orig_node,
+				      struct batadv_neigh_node *neigh_node,
+				      struct batadv_hard_iface *if_incoming,
+				      struct batadv_hard_iface *if_outgoing)
+{
+	struct batadv_orig_ifinfo *orig_ifinfo = NULL;
+	struct batadv_neigh_ifinfo *neigh_ifinfo = NULL;
+	bool protection_started = false;
+	int ret = -EINVAL;
+	u32 path_throughput;
+	s32 seq_diff;
+
+	orig_ifinfo = batadv_orig_ifinfo_new(orig_node, if_outgoing);
+	if (!orig_ifinfo)
+		goto out;
+
+	seq_diff = ntohl(ogm2->seqno) - orig_ifinfo->last_real_seqno;
+
+	if (!hlist_empty(&orig_node->neigh_list) &&
+	    batadv_window_protected(bat_priv, seq_diff,
+				    BATADV_OGM_MAX_AGE,
+				    &orig_ifinfo->batman_seqno_reset,
+				    &protection_started)) {
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Drop packet: packet within window protection time from %pM\n",
+			   ogm2->orig);
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Last reset: %ld, %ld\n",
+			   orig_ifinfo->batman_seqno_reset, jiffies);
+		goto out;
+	}
+
+	/* drop packets with old seqnos, however accept the first packet after
+	 * a host has been rebooted.
+	 */
+	if ((seq_diff < 0) && !protection_started)
+		goto out;
+
+	neigh_node->last_seen = jiffies;
+
+	orig_node->last_seen = jiffies;
+
+	orig_ifinfo->last_real_seqno = ntohl(ogm2->seqno);
+	orig_ifinfo->last_ttl = ogm2->ttl;
+
+	neigh_ifinfo = batadv_neigh_ifinfo_new(neigh_node, if_outgoing);
+	if (!neigh_ifinfo)
+		goto out;
+
+	path_throughput = batadv_v_forward_penalty(bat_priv, if_incoming,
+						   if_outgoing,
+						   ntohl(ogm2->throughput));
+	neigh_ifinfo->bat_v.throughput = path_throughput;
+	neigh_ifinfo->bat_v.last_seqno = ntohl(ogm2->seqno);
+	neigh_ifinfo->last_ttl = ogm2->ttl;
+
+	if (seq_diff > 0 || protection_started)
+		ret = 1;
+	else
+		ret = 0;
+out:
+	if (orig_ifinfo)
+		batadv_orig_ifinfo_put(orig_ifinfo);
+	if (neigh_ifinfo)
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
+
+	return ret;
+}
+
+/**
+ * batadv_v_ogm_route_update - update routes based on OGM
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: the Ethernet header of the OGM2
+ * @ogm2: OGM2 structure
+ * @orig_node: Originator structure for which the OGM has been received
+ * @neigh_node: the neigh_node through with the OGM has been received
+ * @if_incoming: the interface where this packet was received
+ * @if_outgoing: the interface for which the packet should be considered
+ */
+static void batadv_v_ogm_route_update(struct batadv_priv *bat_priv,
+				      const struct ethhdr *ethhdr,
+				      const struct batadv_ogm2_packet *ogm2,
+				      struct batadv_orig_node *orig_node,
+				      struct batadv_neigh_node *neigh_node,
+				      struct batadv_hard_iface *if_incoming,
+				      struct batadv_hard_iface *if_outgoing)
+{
+	struct batadv_neigh_node *router = NULL;
+	struct batadv_neigh_ifinfo *neigh_ifinfo = NULL;
+	struct batadv_orig_node *orig_neigh_node = NULL;
+	struct batadv_orig_ifinfo *orig_ifinfo = NULL;
+	struct batadv_neigh_node *orig_neigh_router = NULL;
+
+	neigh_ifinfo = batadv_neigh_ifinfo_get(neigh_node, if_outgoing);
+	if (!neigh_ifinfo)
+		goto out;
+
+	orig_neigh_node = batadv_v_ogm_orig_get(bat_priv, ethhdr->h_source);
+	if (!orig_neigh_node)
+		goto out;
+
+	orig_neigh_router = batadv_orig_router_get(orig_neigh_node,
+						   if_outgoing);
+
+	/* drop packet if sender is not a direct neighbor and if we
+	 * don't route towards it
+	 */
+	router = batadv_orig_router_get(orig_node, if_outgoing);
+	if (router && router->orig_node != orig_node && !orig_neigh_router) {
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Drop packet: OGM via unknown neighbor!\n");
+		goto out;
+	}
+
+	if (router)
+		batadv_neigh_node_put(router);
+
+	/* Update routes, and check if the OGM is from the best next hop */
+	batadv_v_ogm_orig_update(bat_priv, orig_node, neigh_node, ogm2,
+				 if_outgoing);
+
+	orig_ifinfo = batadv_orig_ifinfo_new(orig_node, if_outgoing);
+	if (!orig_ifinfo)
+		goto out;
+
+	/* don't forward the same seqno twice on one interface */
+	if (orig_ifinfo->last_seqno_forwarded == ntohl(ogm2->seqno))
+		goto out;
+
+	/* acquire possibly updated router */
+	router = batadv_orig_router_get(orig_node, if_outgoing);
+
+	/* strict rule: forward packets coming from the best next hop only */
+	if (neigh_node != router)
+		goto out;
+
+	/* only forward for specific interface, not for the default one. */
+	if (if_outgoing != BATADV_IF_DEFAULT) {
+		orig_ifinfo->last_seqno_forwarded = ntohl(ogm2->seqno);
+		batadv_v_ogm_forward(bat_priv, ogm2,
+				     neigh_ifinfo->bat_v.throughput,
+				     if_incoming, if_outgoing);
+	}
+
+out:
+	if (orig_ifinfo)
+		batadv_orig_ifinfo_put(orig_ifinfo);
+	if (router)
+		batadv_neigh_node_put(router);
+	if (orig_neigh_router)
+		batadv_neigh_node_put(orig_neigh_router);
+	if (orig_neigh_node)
+		batadv_orig_node_put(orig_neigh_node);
+	if (neigh_ifinfo)
+		batadv_neigh_ifinfo_put(neigh_ifinfo);
+}
+
+/**
+ * batadv_v_ogm_process_per_outif - process a batman v OGM for an outgoing if
+ * @bat_priv: the bat priv with all the soft interface information
+ * @ethhdr: the Ethernet header of the OGM2
+ * @ogm2: OGM2 structure
+ * @orig_node: Originator structure for which the OGM has been received
+ * @neigh_node: the neigh_node through with the OGM has been received
+ * @if_incoming: the interface where this packet was received
+ * @if_outgoing: the interface for which the packet should be considered
+ */
+static void
+batadv_v_ogm_process_per_outif(struct batadv_priv *bat_priv,
+			       const struct ethhdr *ethhdr,
+			       const struct batadv_ogm2_packet *ogm2,
+			       struct batadv_orig_node *orig_node,
+			       struct batadv_neigh_node *neigh_node,
+			       struct batadv_hard_iface *if_incoming,
+			       struct batadv_hard_iface *if_outgoing)
+{
+	int seqno_age;
+
+	/* first, update the metric with according sanity checks */
+	seqno_age = batadv_v_ogm_metric_update(bat_priv, ogm2, orig_node,
+					       neigh_node, if_incoming,
+					       if_outgoing);
+
+	/* outdated sequence numbers are to be discarded */
+	if (seqno_age < 0)
+		return;
+
+	/* only unknown & newer OGMs contain TVLVs we are interested in */
+	if ((seqno_age > 0) && (if_outgoing == BATADV_IF_DEFAULT))
+		batadv_tvlv_containers_process(bat_priv, true, orig_node,
+					       NULL, NULL,
+					       (unsigned char *)(ogm2 + 1),
+					       ntohs(ogm2->tvlv_len));
+
+	/* if the metric update went through, update routes if needed */
+	batadv_v_ogm_route_update(bat_priv, ethhdr, ogm2, orig_node,
+				  neigh_node, if_incoming, if_outgoing);
+}
+
+/**
+ * batadv_v_ogm_aggr_packet - checks if there is another OGM aggregated
+ * @buff_pos: current position in the skb
+ * @packet_len: total length of the skb
+ * @tvlv_len: tvlv length of the previously considered OGM
+ *
+ * Return: true if there is enough space for another OGM, false otherwise.
+ */
+static bool batadv_v_ogm_aggr_packet(int buff_pos, int packet_len,
+				     __be16 tvlv_len)
+{
+	int next_buff_pos = 0;
+
+	next_buff_pos += buff_pos + BATADV_OGM2_HLEN;
+	next_buff_pos += ntohs(tvlv_len);
+
+	return (next_buff_pos <= packet_len) &&
+	       (next_buff_pos <= BATADV_MAX_AGGREGATION_BYTES);
+}
+
+/**
+ * batadv_v_ogm_process - process an incoming batman v OGM
+ * @skb: the skb containing the OGM
+ * @ogm_offset: offset to the OGM which should be processed (for aggregates)
+ * @if_incoming: the interface where this packet was receved
+ */
+static void batadv_v_ogm_process(const struct sk_buff *skb, int ogm_offset,
+				 struct batadv_hard_iface *if_incoming)
+{
+	struct batadv_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
+	struct ethhdr *ethhdr;
+	struct batadv_orig_node *orig_node = NULL;
+	struct batadv_hardif_neigh_node *hardif_neigh = NULL;
+	struct batadv_neigh_node *neigh_node = NULL;
+	struct batadv_hard_iface *hard_iface;
+	struct batadv_ogm2_packet *ogm_packet;
+	u32 ogm_throughput, link_throughput, path_throughput;
+
+	ethhdr = eth_hdr(skb);
+	ogm_packet = (struct batadv_ogm2_packet *)(skb->data + ogm_offset);
+
+	ogm_throughput = ntohl(ogm_packet->throughput);
+
+	batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+		   "Received OGM2 packet via NB: %pM, IF: %s [%pM] (from OG: %pM, seqno %u, troughput %u, TTL %u, V %u, tvlv_len %u)\n",
+		   ethhdr->h_source, if_incoming->net_dev->name,
+		   if_incoming->net_dev->dev_addr, ogm_packet->orig,
+		   ntohl(ogm_packet->seqno), ogm_throughput, ogm_packet->ttl,
+		   ogm_packet->version, ntohs(ogm_packet->tvlv_len));
+
+	/* If the troughput metric is 0, immediately drop the packet. No need to
+	 * create orig_node / neigh_node for an unusable route.
+	 */
+	if (ogm_throughput == 0) {
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Drop packet: originator packet with troughput metric of 0\n");
+		return;
+	}
+
+	/* require ELP packets be to received from this neighbor first */
+	hardif_neigh = batadv_hardif_neigh_get(if_incoming, ethhdr->h_source);
+	if (!hardif_neigh) {
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Drop packet: OGM via unknown neighbor!\n");
+		goto out;
+	}
+
+	orig_node = batadv_v_ogm_orig_get(bat_priv, ogm_packet->orig);
+	if (!orig_node)
+		return;
+
+	neigh_node = batadv_neigh_node_new(orig_node, if_incoming,
+					   ethhdr->h_source);
+	if (!neigh_node)
+		goto out;
+
+	/* Update the received throughput metric to match the link
+	 * characteristic:
+	 *  - If this OGM traveled one hop so far (emitted by single hop
+	 *    neighbor) the path throughput metric equals the link throughput.
+	 *  - For OGMs traversing more than hop the path throughput metric is
+	 *    the smaller of the path throughput and the link throughput.
+	 */
+	link_throughput = ewma_throughput_read(&hardif_neigh->bat_v.throughput);
+	path_throughput = min_t(u32, link_throughput, ogm_throughput);
+	ogm_packet->throughput = htonl(path_throughput);
+
+	batadv_v_ogm_process_per_outif(bat_priv, ethhdr, ogm_packet, orig_node,
+				       neigh_node, if_incoming,
+				       BATADV_IF_DEFAULT);
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(hard_iface, &batadv_hardif_list, list) {
+		if (hard_iface->if_status != BATADV_IF_ACTIVE)
+			continue;
+
+		if (hard_iface->soft_iface != bat_priv->soft_iface)
+			continue;
+
+		batadv_v_ogm_process_per_outif(bat_priv, ethhdr, ogm_packet,
+					       orig_node, neigh_node,
+					       if_incoming, hard_iface);
+	}
+	rcu_read_unlock();
+out:
+	if (orig_node)
+		batadv_orig_node_put(orig_node);
+	if (neigh_node)
+		batadv_neigh_node_put(neigh_node);
+	if (hardif_neigh)
+		batadv_hardif_neigh_put(hardif_neigh);
+}
+
+/**
  * batadv_v_ogm_packet_recv - OGM2 receiving handler
  * @skb: the received OGM
  * @if_incoming: the interface where this OGM has been received
@@ -206,6 +741,9 @@ int batadv_v_ogm_packet_recv(struct sk_buff *skb,
 	struct batadv_priv *bat_priv = netdev_priv(if_incoming->soft_iface);
 	struct batadv_ogm2_packet *ogm_packet;
 	struct ethhdr *ethhdr = eth_hdr(skb);
+	int ogm_offset;
+	u8 *packet_pos;
+	int ret = NET_RX_DROP;
 
 	/* did we receive a OGM2 packet on an interface that does not have
 	 * B.A.T.M.A.N. V enabled ?
@@ -228,8 +766,24 @@ int batadv_v_ogm_packet_recv(struct sk_buff *skb,
 	batadv_add_counter(bat_priv, BATADV_CNT_MGMT_RX_BYTES,
 			   skb->len + ETH_HLEN);
 
+	ogm_offset = 0;
+	ogm_packet = (struct batadv_ogm2_packet *)skb->data;
+
+	while (batadv_v_ogm_aggr_packet(ogm_offset, skb_headlen(skb),
+					ogm_packet->tvlv_len)) {
+		batadv_v_ogm_process(skb, ogm_offset, if_incoming);
+
+		ogm_offset += BATADV_OGM2_HLEN;
+		ogm_offset += ntohs(ogm_packet->tvlv_len);
+
+		packet_pos = skb->data + ogm_offset;
+		ogm_packet = (struct batadv_ogm2_packet *)packet_pos;
+	}
+
+	ret = NET_RX_SUCCESS;
 	consume_skb(skb);
-	return NET_RX_SUCCESS;
+
+	return ret;
 }
 
 /**
diff --git a/net/batman-adv/bat_v_ogm.h b/net/batman-adv/bat_v_ogm.h
index d9f178c..d849c75 100644
--- a/net/batman-adv/bat_v_ogm.h
+++ b/net/batman-adv/bat_v_ogm.h
@@ -18,6 +18,8 @@
 #ifndef _BATMAN_ADV_BATADV_V_OGM_H_
 #define _BATMAN_ADV_BATADV_V_OGM_H_
 
+#include <linux/types.h>
+
 struct batadv_hard_iface;
 struct batadv_priv;
 struct sk_buff;
@@ -25,6 +27,8 @@ struct sk_buff;
 int batadv_v_ogm_init(struct batadv_priv *bat_priv);
 void batadv_v_ogm_free(struct batadv_priv *bat_priv);
 int batadv_v_ogm_iface_enable(struct batadv_hard_iface *hard_iface);
+struct batadv_orig_node *batadv_v_ogm_orig_get(struct batadv_priv *bat_priv,
+					       const u8 *addr);
 void batadv_v_ogm_primary_iface_set(struct batadv_hard_iface *primary_iface);
 int batadv_v_ogm_packet_recv(struct sk_buff *skb,
 			     struct batadv_hard_iface *if_incoming);
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index fa0b067..2e79544 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -63,6 +63,8 @@
 
 /* B.A.T.M.A.N. V */
 #define BATADV_ELP_MAX_AGE 64
+#define BATADV_OGM_MAX_ORIGDIFF 5
+#define BATADV_OGM_MAX_AGE 64
 
 /* number of OGMs sent with the last tt diff */
 #define BATADV_TT_OGM_APPEND_MAX 3
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index e213ccd..cf55cd0 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -149,6 +149,7 @@ struct batadv_hard_iface {
  * @router: router that should be used to reach this originator
  * @last_real_seqno: last and best known sequence number
  * @last_ttl: ttl of last received packet
+ * @last_seqno_forwarded: seqno of the OGM which was forwarded last
  * @batman_seqno_reset: time when the batman seqno window was reset
  * @refcount: number of contexts the object is used
  * @rcu: struct used for freeing in an RCU-safe manner
@@ -159,6 +160,7 @@ struct batadv_orig_ifinfo {
 	struct batadv_neigh_node __rcu *router; /* rcu protected pointer */
 	u32 last_real_seqno;
 	u8 last_ttl;
+	u32 last_seqno_forwarded;
 	unsigned long batman_seqno_reset;
 	struct kref refcount;
 	struct rcu_head rcu;
@@ -448,9 +450,11 @@ struct batadv_neigh_ifinfo_bat_iv {
  * struct batadv_neigh_ifinfo_bat_v - neighbor information per outgoing
  *  interface for B.A.T.M.A.N. V
  * @throughput: last throughput metric received from originator via this neigh
+ * @last_seqno: last sequence number known for this neighbor
  */
 struct batadv_neigh_ifinfo_bat_v {
 	u32 throughput;
+	u32 last_seqno;
 };
 
 /**
-- 
cgit v0.10.2


From 0b5ecc6811bd576ecc9813bbe069f2293cb1c6aa Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Sat, 16 Jan 2016 16:40:14 +0800
Subject: batman-adv: add throughput override attribute to hard_ifaces

This attribute is exported to user space to disable the link
throughput auto-detection by setting a fixed value.
The throughput override value is used when batman-adv is
computing the link throughput towards a neighbour.

If the value is set to 0 then batman-adv will try to detect
the throughput by itself.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/Documentation/ABI/testing/sysfs-class-net-batman-adv b/Documentation/ABI/testing/sysfs-class-net-batman-adv
index aea78db..518f6a1 100644
--- a/Documentation/ABI/testing/sysfs-class-net-batman-adv
+++ b/Documentation/ABI/testing/sysfs-class-net-batman-adv
@@ -1,4 +1,13 @@
 
+What:		/sys/class/net/<iface>/batman-adv/throughput_override
+Date:		Feb 2014
+Contact:	Antonio Quartulli <antonio@meshcoding.com>
+description:
+		Defines the throughput value to be used by B.A.T.M.A.N. V
+		when estimating the link throughput using this interface.
+		If the value is set to 0 then batman-adv will try to
+		estimate the throughput by itself.
+
 What:           /sys/class/net/<iface>/batman-adv/elp_interval
 Date:           Feb 2014
 Contact:        Linus Lüssing <linus.luessing@web.de>
diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index b90a4df..d9cb5c4 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -18,6 +18,7 @@
 #include "bat_algo.h"
 #include "main.h"
 
+#include <linux/atomic.h>
 #include <linux/cache.h>
 #include <linux/init.h>
 
@@ -37,6 +38,11 @@ static int batadv_v_iface_enable(struct batadv_hard_iface *hard_iface)
 	if (ret < 0)
 		batadv_v_elp_iface_disable(hard_iface);
 
+	/* enable link throughput auto-detection by setting the throughput
+	 * override to zero
+	 */
+	atomic_set(&hard_iface->bat_v.throughput_override, 0);
+
 	return ret;
 }
 
diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c
index 5ee04f7..4423047 100644
--- a/net/batman-adv/gateway_common.c
+++ b/net/batman-adv/gateway_common.c
@@ -40,8 +40,8 @@
  *
  * Return: false on parse error and true otherwise.
  */
-static bool batadv_parse_throughput(struct net_device *net_dev, char *buff,
-				    const char *description, u32 *throughput)
+bool batadv_parse_throughput(struct net_device *net_dev, char *buff,
+			     const char *description, u32 *throughput)
 {
 	enum batadv_bandwidth_units bw_unit_type = BATADV_BW_UNIT_KBIT;
 	u64 lthroughput;
diff --git a/net/batman-adv/gateway_common.h b/net/batman-adv/gateway_common.h
index b583463..8a5e1dd 100644
--- a/net/batman-adv/gateway_common.h
+++ b/net/batman-adv/gateway_common.h
@@ -49,5 +49,7 @@ ssize_t batadv_gw_bandwidth_set(struct net_device *net_dev, char *buff,
 void batadv_gw_tvlv_container_update(struct batadv_priv *bat_priv);
 void batadv_gw_init(struct batadv_priv *bat_priv);
 void batadv_gw_free(struct batadv_priv *bat_priv);
+bool batadv_parse_throughput(struct net_device *net_dev, char *buff,
+			     const char *description, u32 *throughput);
 
 #endif /* _NET_BATMAN_ADV_GATEWAY_COMMON_H_ */
diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c
index e860143..e7cf513 100644
--- a/net/batman-adv/sysfs.c
+++ b/net/batman-adv/sysfs.c
@@ -917,12 +917,85 @@ static ssize_t batadv_show_iface_status(struct kobject *kobj,
 	return length;
 }
 
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+
+/**
+ * batadv_store_throughput_override - parse and store throughput override
+ *  entered by the user
+ * @kobj: kobject representing the private mesh sysfs directory
+ * @attr: the batman-adv attribute the user is interacting with
+ * @buff: the buffer containing the user data
+ * @count: number of bytes in the buffer
+ *
+ * Return: 'count' on success or a negative error code in case of failure
+ */
+static ssize_t batadv_store_throughput_override(struct kobject *kobj,
+						struct attribute *attr,
+						char *buff, size_t count)
+{
+	struct net_device *net_dev = batadv_kobj_to_netdev(kobj);
+	struct batadv_hard_iface *hard_iface;
+	u32 tp_override;
+	u32 old_tp_override;
+	bool ret;
+
+	hard_iface = batadv_hardif_get_by_netdev(net_dev);
+	if (!hard_iface)
+		return -EINVAL;
+
+	if (buff[count - 1] == '\n')
+		buff[count - 1] = '\0';
+
+	ret = batadv_parse_throughput(net_dev, buff, "throughput_override",
+				      &tp_override);
+	if (!ret)
+		return count;
+
+	old_tp_override = atomic_read(&hard_iface->bat_v.throughput_override);
+	if (old_tp_override == tp_override)
+		goto out;
+
+	batadv_info(net_dev, "%s: Changing from: %u.%u MBit to: %u.%u MBit\n",
+		    "throughput_override",
+		    old_tp_override / 10, old_tp_override % 10,
+		    tp_override / 10, tp_override % 10);
+
+	atomic_set(&hard_iface->bat_v.throughput_override, tp_override);
+
+out:
+	batadv_hardif_put(hard_iface);
+	return count;
+}
+
+static ssize_t batadv_show_throughput_override(struct kobject *kobj,
+					       struct attribute *attr,
+					       char *buff)
+{
+	struct net_device *net_dev = batadv_kobj_to_netdev(kobj);
+	struct batadv_hard_iface *hard_iface;
+	u32 tp_override;
+
+	hard_iface = batadv_hardif_get_by_netdev(net_dev);
+	if (!hard_iface)
+		return -EINVAL;
+
+	tp_override = atomic_read(&hard_iface->bat_v.throughput_override);
+
+	return sprintf(buff, "%u.%u MBit\n", tp_override / 10,
+		       tp_override % 10);
+}
+
+#endif
+
 static BATADV_ATTR(mesh_iface, S_IRUGO | S_IWUSR, batadv_show_mesh_iface,
 		   batadv_store_mesh_iface);
 static BATADV_ATTR(iface_status, S_IRUGO, batadv_show_iface_status, NULL);
 #ifdef CONFIG_BATMAN_ADV_BATMAN_V
 BATADV_ATTR_HIF_UINT(elp_interval, bat_v.elp_interval, S_IRUGO | S_IWUSR,
 		     2 * BATADV_JITTER, INT_MAX, NULL);
+static BATADV_ATTR(throughput_override, S_IRUGO | S_IWUSR,
+		   batadv_show_throughput_override,
+		   batadv_store_throughput_override);
 #endif
 
 static struct batadv_attribute *batadv_batman_attrs[] = {
@@ -930,6 +1003,7 @@ static struct batadv_attribute *batadv_batman_attrs[] = {
 	&batadv_attr_iface_status,
 #ifdef CONFIG_BATMAN_ADV_BATMAN_V
 	&batadv_attr_elp_interval,
+	&batadv_attr_throughput_override,
 #endif
 	NULL,
 };
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index cf55cd0..a6f0952 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -92,12 +92,14 @@ struct batadv_hard_iface_bat_iv {
  * @elp_seqno: current ELP sequence number
  * @elp_skb: base skb containing the ELP message to send
  * @elp_wq: workqueue used to schedule ELP transmissions
+ * @throughput_override: throughput override to disable link auto-detection
  */
 struct batadv_hard_iface_bat_v {
 	atomic_t elp_interval;
 	atomic_t elp_seqno;
 	struct sk_buff *elp_skb;
 	struct delayed_work elp_wq;
+	atomic_t throughput_override;
 };
 
 /**
-- 
cgit v0.10.2


From 95d392784dd0a51e4216e075f04a68c922745985 Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Sat, 16 Jan 2016 16:40:15 +0800
Subject: batman-adv: keep track of when unicast packets are sent

To enable ELP to send probing packets over wireless links
only if needed, batman-adv must keep track of the last time
it sent a unicast packet towards every neighbour.

For this purpose a 2 main changes are introduced:
1) a new member of the elp_neigh_node structure stores the
   last time a unicast packet was sent towards this neighbour;
2) a wrapper function for sending unicast packets is
   implemented. This function will simply update the member
   describe din point 1) and then forward the packet to the
   real sending routine.

Point 2) implies that any code-path leading to a unicast
sending now has to use the new wrapper.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index 5651e33..cb2d1b9 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -478,7 +478,7 @@ static void batadv_iv_ogm_send_to_if(struct batadv_forw_packet *forw_packet,
 		batadv_inc_counter(bat_priv, BATADV_CNT_MGMT_TX);
 		batadv_add_counter(bat_priv, BATADV_CNT_MGMT_TX_BYTES,
 				   skb->len + ETH_HLEN);
-		batadv_send_skb_packet(skb, hard_iface, batadv_broadcast_addr);
+		batadv_send_broadcast_skb(skb, hard_iface);
 	}
 }
 
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index fac5aee..1e4d13c 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -104,7 +104,7 @@ static void batadv_v_elp_periodic_work(struct work_struct *work)
 		   hard_iface->net_dev->name,
 		   atomic_read(&hard_iface->bat_v.elp_seqno));
 
-	batadv_send_skb_packet(skb, hard_iface, batadv_broadcast_addr);
+	batadv_send_broadcast_skb(skb, hard_iface);
 
 	atomic_inc(&hard_iface->bat_v.elp_seqno);
 
diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c
index 22dda0e..c800096 100644
--- a/net/batman-adv/bat_v_ogm.c
+++ b/net/batman-adv/bat_v_ogm.c
@@ -120,7 +120,7 @@ static void batadv_v_ogm_send_to_if(struct sk_buff *skb,
 	batadv_add_counter(bat_priv, BATADV_CNT_MGMT_TX_BYTES,
 			   skb->len + ETH_HLEN);
 
-	batadv_send_skb_packet(skb, hard_iface, batadv_broadcast_addr);
+	batadv_send_broadcast_skb(skb, hard_iface);
 }
 
 /**
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index 4c9b69d..e96d7c7 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -654,9 +654,7 @@ static bool batadv_dat_send_data(struct batadv_priv *bat_priv,
 			goto free_neigh;
 		}
 
-		send_status = batadv_send_skb_packet(tmp_skb,
-						     neigh_node->if_incoming,
-						     neigh_node->addr);
+		send_status = batadv_send_unicast_skb(tmp_skb, neigh_node);
 		if (send_status == NET_XMIT_SUCCESS) {
 			/* count the sent packet */
 			switch (packet_subtype) {
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index adb9c39..e6956d0 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -378,8 +378,7 @@ bool batadv_frag_skb_fwd(struct sk_buff *skb,
 				   skb->len + ETH_HLEN);
 
 		packet->ttl--;
-		batadv_send_skb_packet(skb, neigh_node->if_incoming,
-				       neigh_node->addr);
+		batadv_send_unicast_skb(skb, neigh_node);
 		ret = true;
 	}
 
@@ -486,8 +485,7 @@ bool batadv_frag_send_packet(struct sk_buff *skb,
 		batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX);
 		batadv_add_counter(bat_priv, BATADV_CNT_FRAG_TX_BYTES,
 				   skb_fragment->len + ETH_HLEN);
-		batadv_send_skb_packet(skb_fragment, neigh_node->if_incoming,
-				       neigh_node->addr);
+		batadv_send_unicast_skb(skb_fragment, neigh_node);
 		frag_header.no++;
 
 		/* The initial check in this function should cover this case */
@@ -506,7 +504,7 @@ bool batadv_frag_send_packet(struct sk_buff *skb,
 	batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX);
 	batadv_add_counter(bat_priv, BATADV_CNT_FRAG_TX_BYTES,
 			   skb->len + ETH_HLEN);
-	batadv_send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
+	batadv_send_unicast_skb(skb, neigh_node);
 
 	ret = true;
 
diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c
index 6268f08..14d0013 100644
--- a/net/batman-adv/icmp_socket.c
+++ b/net/batman-adv/icmp_socket.c
@@ -278,7 +278,7 @@ static ssize_t batadv_socket_write(struct file *file, const char __user *buff,
 
 	ether_addr_copy(icmp_header->orig, primary_if->net_dev->dev_addr);
 
-	batadv_send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
+	batadv_send_unicast_skb(skb, neigh_node);
 	goto out;
 
 dst_unreach:
diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c
index d253bb2..b41719b 100644
--- a/net/batman-adv/network-coding.c
+++ b/net/batman-adv/network-coding.c
@@ -575,9 +575,7 @@ batadv_nc_hash_find(struct batadv_hashtable *hash,
  */
 static void batadv_nc_send_packet(struct batadv_nc_packet *nc_packet)
 {
-	batadv_send_skb_packet(nc_packet->skb,
-			       nc_packet->neigh_node->if_incoming,
-			       nc_packet->nc_path->next_hop);
+	batadv_send_unicast_skb(nc_packet->skb, nc_packet->neigh_node);
 	nc_packet->skb = NULL;
 	batadv_nc_packet_free(nc_packet);
 }
@@ -1067,11 +1065,11 @@ static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 	struct batadv_unicast_packet *packet1;
 	struct batadv_unicast_packet *packet2;
 	struct batadv_coded_packet *coded_packet;
-	struct batadv_neigh_node *neigh_tmp, *router_neigh;
-	struct batadv_neigh_node *router_coding = NULL;
+	struct batadv_neigh_node *neigh_tmp, *router_neigh, *first_dest;
+	struct batadv_neigh_node *router_coding = NULL, *second_dest;
 	struct batadv_neigh_ifinfo *router_neigh_ifinfo = NULL;
 	struct batadv_neigh_ifinfo *router_coding_ifinfo = NULL;
-	u8 *first_source, *first_dest, *second_source, *second_dest;
+	u8 *first_source, *second_source;
 	__be32 packet_id1, packet_id2;
 	size_t count;
 	bool res = false;
@@ -1114,9 +1112,9 @@ static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 	 */
 	if (tq_weighted_neigh >= tq_weighted_coding) {
 		/* Destination from nc_packet is selected for MAC-header */
-		first_dest = nc_packet->nc_path->next_hop;
+		first_dest = nc_packet->neigh_node;
 		first_source = nc_packet->nc_path->prev_hop;
-		second_dest = neigh_node->addr;
+		second_dest = neigh_node;
 		second_source = ethhdr->h_source;
 		packet1 = (struct batadv_unicast_packet *)nc_packet->skb->data;
 		packet2 = (struct batadv_unicast_packet *)skb->data;
@@ -1125,9 +1123,9 @@ static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 					      skb->data + sizeof(*packet2));
 	} else {
 		/* Destination for skb is selected for MAC-header */
-		first_dest = neigh_node->addr;
+		first_dest = neigh_node;
 		first_source = ethhdr->h_source;
-		second_dest = nc_packet->nc_path->next_hop;
+		second_dest = nc_packet->neigh_node;
 		second_source = nc_packet->nc_path->prev_hop;
 		packet1 = (struct batadv_unicast_packet *)skb->data;
 		packet2 = (struct batadv_unicast_packet *)nc_packet->skb->data;
@@ -1169,7 +1167,7 @@ static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 	coded_packet->first_ttvn = packet1->ttvn;
 
 	/* Info about second unicast packet */
-	ether_addr_copy(coded_packet->second_dest, second_dest);
+	ether_addr_copy(coded_packet->second_dest, second_dest->addr);
 	ether_addr_copy(coded_packet->second_source, second_source);
 	ether_addr_copy(coded_packet->second_orig_dest, packet2->dest);
 	coded_packet->second_crc = packet_id2;
@@ -1224,7 +1222,7 @@ static bool batadv_nc_code_packets(struct batadv_priv *bat_priv,
 	batadv_nc_packet_free(nc_packet);
 
 	/* Send the coded packet and return true */
-	batadv_send_skb_packet(skb_dest, neigh_node->if_incoming, first_dest);
+	batadv_send_unicast_skb(skb_dest, first_dest);
 	res = true;
 out:
 	if (router_neigh)
diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c
index caff32c..3ce06e0 100644
--- a/net/batman-adv/send.c
+++ b/net/batman-adv/send.c
@@ -49,16 +49,30 @@
 
 static void batadv_send_outstanding_bcast_packet(struct work_struct *work);
 
-/* send out an already prepared packet to the given address via the
- * specified batman interface
+/**
+ * batadv_send_skb_packet - send an already prepared packet
+ * @skb: the packet to send
+ * @hard_iface: the interface to use to send the broadcast packet
+ * @dst_addr: the payload destination
+ *
+ * Send out an already prepared packet to the given neighbor or broadcast it
+ * using the specified interface. Either hard_iface or neigh_node must be not
+ * NULL.
+ * If neigh_node is NULL, then the packet is broadcasted using hard_iface,
+ * otherwise it is sent as unicast to the given neighbor.
+ *
+ * Return: NET_TX_DROP in case of error or the result of dev_queue_xmit(skb)
+ * otherwise
  */
 int batadv_send_skb_packet(struct sk_buff *skb,
 			   struct batadv_hard_iface *hard_iface,
 			   const u8 *dst_addr)
 {
-	struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
+	struct batadv_priv *bat_priv;
 	struct ethhdr *ethhdr;
 
+	bat_priv = netdev_priv(hard_iface->soft_iface);
+
 	if (hard_iface->if_status != BATADV_IF_ACTIVE)
 		goto send_skb_err;
 
@@ -100,6 +114,35 @@ send_skb_err:
 	return NET_XMIT_DROP;
 }
 
+int batadv_send_broadcast_skb(struct sk_buff *skb,
+			      struct batadv_hard_iface *hard_iface)
+{
+	return batadv_send_skb_packet(skb, hard_iface, batadv_broadcast_addr);
+}
+
+int batadv_send_unicast_skb(struct sk_buff *skb,
+			    struct batadv_neigh_node *neigh)
+{
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	struct batadv_hardif_neigh_node *hardif_neigh;
+#endif
+	int ret;
+
+	ret = batadv_send_skb_packet(skb, neigh->if_incoming, neigh->addr);
+
+#ifdef CONFIG_BATMAN_ADV_BATMAN_V
+	hardif_neigh = batadv_hardif_neigh_get(neigh->if_incoming, neigh->addr);
+
+	if ((hardif_neigh) && (ret != NET_XMIT_DROP))
+		hardif_neigh->bat_v.last_unicast_tx = jiffies;
+
+	if (hardif_neigh)
+		batadv_hardif_neigh_put(hardif_neigh);
+#endif
+
+	return ret;
+}
+
 /**
  * batadv_send_skb_to_orig - Lookup next-hop and transmit skb.
  * @skb: Packet to be transmitted.
@@ -146,8 +189,7 @@ int batadv_send_skb_to_orig(struct sk_buff *skb,
 	if (recv_if && batadv_nc_skb_forward(skb, neigh_node)) {
 		ret = NET_XMIT_POLICED;
 	} else {
-		batadv_send_skb_packet(skb, neigh_node->if_incoming,
-				       neigh_node->addr);
+		batadv_send_unicast_skb(skb, neigh_node);
 		ret = NET_XMIT_SUCCESS;
 	}
 
@@ -538,8 +580,7 @@ static void batadv_send_outstanding_bcast_packet(struct work_struct *work)
 		/* send a copy of the saved skb */
 		skb1 = skb_clone(forw_packet->skb, GFP_ATOMIC);
 		if (skb1)
-			batadv_send_skb_packet(skb1, hard_iface,
-					       batadv_broadcast_addr);
+			batadv_send_broadcast_skb(skb1, hard_iface);
 	}
 	rcu_read_unlock();
 
diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h
index 7ff95ca..6fd7270 100644
--- a/net/batman-adv/send.h
+++ b/net/batman-adv/send.h
@@ -28,12 +28,16 @@
 struct sk_buff;
 struct work_struct;
 
-int batadv_send_skb_packet(struct sk_buff *skb,
-			   struct batadv_hard_iface *hard_iface,
-			   const u8 *dst_addr);
 int batadv_send_skb_to_orig(struct sk_buff *skb,
 			    struct batadv_orig_node *orig_node,
 			    struct batadv_hard_iface *recv_if);
+int batadv_send_skb_packet(struct sk_buff *skb,
+			   struct batadv_hard_iface *hard_iface,
+			   const u8 *dst_addr);
+int batadv_send_broadcast_skb(struct sk_buff *skb,
+			      struct batadv_hard_iface *hard_iface);
+int batadv_send_unicast_skb(struct sk_buff *skb,
+			    struct batadv_neigh_node *neigh_node);
 void batadv_schedule_bat_ogm(struct batadv_hard_iface *hard_iface);
 int batadv_add_bcast_packet_to_list(struct batadv_priv *bat_priv,
 				    const struct sk_buff *skb,
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index a6f0952..c56bb88 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -377,11 +377,13 @@ DECLARE_EWMA(throughput, 1024, 8)
  * @throughput: ewma link throughput towards this neighbor
  * @elp_interval: time interval between two ELP transmissions
  * @elp_latest_seqno: latest and best known ELP sequence number
+ * @last_unicast_tx: when the last unicast packet has been sent to this neighbor
  */
 struct batadv_hardif_neigh_node_bat_v {
 	struct ewma_throughput throughput;
 	u32 elp_interval;
 	u32 elp_latest_seqno;
+	unsigned long last_unicast_tx;
 };
 
 /**
-- 
cgit v0.10.2


From c833484e5f3872a38fe232c663586069d5ad9645 Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Tue, 10 Nov 2015 18:50:51 +0100
Subject: batman-adv: ELP - compute the metric based on the estimated
 throughput

In case of wireless interface retrieve the throughput by
querying cfg80211. To perform this call a separate work
must be scheduled because the function may sleep and this
is not allowed within an RCU protected context (RCU in this
case is used to iterate over all the neighbours).

Use ethtool to retrieve information about an Ethernet link
like HALF/FULL_DUPLEX and advertised bandwidth (e.g.
100/10Mbps).

The metric is updated each time a new ELP packet is sent,
this way it is possible to timely react to a metric
variation which can imply (for example) a neighbour
disconnection.

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 5c148a8..e651dc9 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -17,7 +17,7 @@ config BATMAN_ADV
 
 config BATMAN_ADV_BATMAN_V
 	bool "B.A.T.M.A.N. V protocol (experimental)"
-	depends on BATMAN_ADV
+	depends on BATMAN_ADV && CFG80211
 	default n
 	help
 	  This option enables the B.A.T.M.A.N. V protocol, the successor
diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index d9cb5c4..ff31f2a 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -21,6 +21,7 @@
 #include <linux/atomic.h>
 #include <linux/cache.h>
 #include <linux/init.h>
+#include <linux/workqueue.h>
 
 #include "bat_v_elp.h"
 #include "bat_v_ogm.h"
@@ -65,6 +66,8 @@ static void
 batadv_v_hardif_neigh_init(struct batadv_hardif_neigh_node *hardif_neigh)
 {
 	ewma_throughput_init(&hardif_neigh->bat_v.throughput);
+	INIT_WORK(&hardif_neigh->bat_v.metric_work,
+		  batadv_v_elp_throughput_metric_update);
 }
 
 static void batadv_v_ogm_schedule(struct batadv_hard_iface *hard_iface)
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index 1e4d13c..8376730 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -22,19 +22,23 @@
 #include <linux/byteorder/generic.h>
 #include <linux/errno.h>
 #include <linux/etherdevice.h>
+#include <linux/ethtool.h>
 #include <linux/fs.h>
 #include <linux/if_ether.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
+#include <linux/kref.h>
 #include <linux/netdevice.h>
 #include <linux/random.h>
 #include <linux/rculist.h>
 #include <linux/rcupdate.h>
+#include <linux/rtnetlink.h>
 #include <linux/skbuff.h>
 #include <linux/stddef.h>
 #include <linux/string.h>
 #include <linux/types.h>
 #include <linux/workqueue.h>
+#include <net/cfg80211.h>
 
 #include "bat_algo.h"
 #include "bat_v_ogm.h"
@@ -60,6 +64,107 @@ static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface)
 }
 
 /**
+ * batadv_v_elp_get_throughput - get the throughput towards a neighbour
+ * @neigh: the neighbour for which the throughput has to be obtained
+ *
+ * Return: The throughput towards the given neighbour in multiples of 100kpbs
+ *         (a value of '1' equals to 0.1Mbps, '10' equals 1Mbps, etc).
+ */
+static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh)
+{
+	struct batadv_hard_iface *hard_iface = neigh->if_incoming;
+	struct ethtool_link_ksettings link_settings;
+	struct station_info sinfo;
+	u32 throughput;
+	int ret;
+
+	/* if the user specified a customised value for this interface, then
+	 * return it directly
+	 */
+	throughput =  atomic_read(&hard_iface->bat_v.throughput_override);
+	if (throughput != 0)
+		return throughput;
+
+	/* if this is a wireless device, then ask its throughput through
+	 * cfg80211 API
+	 */
+	if (batadv_is_wifi_netdev(hard_iface->net_dev)) {
+		if (hard_iface->net_dev->ieee80211_ptr) {
+			ret = cfg80211_get_station(hard_iface->net_dev,
+						   neigh->addr, &sinfo);
+			if (ret == -ENOENT) {
+				/* Node is not associated anymore! It would be
+				 * possible to delete this neighbor. For now set
+				 * the throughput metric to 0.
+				 */
+				return 0;
+			}
+			if (!ret)
+				return sinfo.expected_throughput / 100;
+		}
+
+		/* unsupported WiFi driver version */
+		goto default_throughput;
+	}
+
+	/* if not a wifi interface, check if this device provides data via
+	 * ethtool (e.g. an Ethernet adapter)
+	 */
+	memset(&link_settings, 0, sizeof(link_settings));
+	rtnl_lock();
+	ret = __ethtool_get_link_ksettings(hard_iface->net_dev, &link_settings);
+	rtnl_unlock();
+	if (ret == 0) {
+		/* link characteristics might change over time */
+		if (link_settings.base.duplex == DUPLEX_FULL)
+			hard_iface->bat_v.flags |= BATADV_FULL_DUPLEX;
+		else
+			hard_iface->bat_v.flags &= ~BATADV_FULL_DUPLEX;
+
+		throughput = link_settings.base.speed;
+		if (throughput && (throughput != SPEED_UNKNOWN))
+			return throughput * 10;
+	}
+
+default_throughput:
+	if (!(hard_iface->bat_v.flags & BATADV_WARNING_DEFAULT)) {
+		batadv_info(hard_iface->soft_iface,
+			    "WiFi driver or ethtool info does not provide information about link speeds on interface %s, therefore defaulting to hardcoded throughput values of %u.%1u Mbps. Consider overriding the throughput manually or checking your driver.\n",
+			    hard_iface->net_dev->name,
+			    BATADV_THROUGHPUT_DEFAULT_VALUE / 10,
+			    BATADV_THROUGHPUT_DEFAULT_VALUE % 10);
+		hard_iface->bat_v.flags |= BATADV_WARNING_DEFAULT;
+	}
+
+	/* if none of the above cases apply, return the base_throughput */
+	return BATADV_THROUGHPUT_DEFAULT_VALUE;
+}
+
+/**
+ * batadv_v_elp_throughput_metric_update - worker updating the throughput metric
+ *  of a single hop neighbour
+ * @work: the work queue item
+ */
+void batadv_v_elp_throughput_metric_update(struct work_struct *work)
+{
+	struct batadv_hardif_neigh_node_bat_v *neigh_bat_v;
+	struct batadv_hardif_neigh_node *neigh;
+
+	neigh_bat_v = container_of(work, struct batadv_hardif_neigh_node_bat_v,
+				   metric_work);
+	neigh = container_of(neigh_bat_v, struct batadv_hardif_neigh_node,
+			     bat_v);
+
+	ewma_throughput_add(&neigh->bat_v.throughput,
+			    batadv_v_elp_get_throughput(neigh));
+
+	/* decrement refcounter to balance increment performed before scheduling
+	 * this task
+	 */
+	batadv_hardif_neigh_put(neigh);
+}
+
+/**
  * batadv_v_elp_periodic_work - ELP periodic task per interface
  * @work: work queue item
  *
@@ -67,6 +172,7 @@ static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface)
  */
 static void batadv_v_elp_periodic_work(struct work_struct *work)
 {
+	struct batadv_hardif_neigh_node *hardif_neigh;
 	struct batadv_hard_iface *hard_iface;
 	struct batadv_hard_iface_bat_v *bat_v;
 	struct batadv_elp_packet *elp_packet;
@@ -108,6 +214,31 @@ static void batadv_v_elp_periodic_work(struct work_struct *work)
 
 	atomic_inc(&hard_iface->bat_v.elp_seqno);
 
+	/* The throughput metric is updated on each sent packet. This way, if a
+	 * node is dead and no longer sends packets, batman-adv is still able to
+	 * react timely to its death.
+	 *
+	 * The throughput metric is updated by following these steps:
+	 * 1) if the hard_iface is wifi => send a number of unicast ELPs for
+	 *    probing/sampling to each neighbor
+	 * 2) update the throughput metric value of each neighbor (note that the
+	 *    value retrieved in this step might be 100ms old because the
+	 *    probing packets at point 1) could still be in the HW queue)
+	 */
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(hardif_neigh, &hard_iface->neigh_list, list) {
+		if (!kref_get_unless_zero(&hardif_neigh->refcount))
+			continue;
+
+		/* Reading the estimated throughput from cfg80211 is a task that
+		 * may sleep and that is not allowed in an rcu protected
+		 * context. Therefore schedule a task for that.
+		 */
+		queue_work(batadv_event_workqueue,
+			   &hardif_neigh->bat_v.metric_work);
+	}
+	rcu_read_unlock();
+
 restart_timer:
 	batadv_v_elp_start_timer(hard_iface);
 out:
@@ -146,6 +277,15 @@ int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface)
 	atomic_set(&hard_iface->bat_v.elp_seqno, random_seqno);
 	atomic_set(&hard_iface->bat_v.elp_interval, 500);
 
+	/* assume full-duplex by default */
+	hard_iface->bat_v.flags |= BATADV_FULL_DUPLEX;
+
+	/* warn the user (again) if there is no throughput data is available */
+	hard_iface->bat_v.flags &= ~BATADV_WARNING_DEFAULT;
+
+	if (batadv_is_wifi_netdev(hard_iface->net_dev))
+		hard_iface->bat_v.flags &= ~BATADV_FULL_DUPLEX;
+
 	INIT_DELAYED_WORK(&hard_iface->bat_v.elp_wq,
 			  batadv_v_elp_periodic_work);
 	batadv_v_elp_start_timer(hard_iface);
diff --git a/net/batman-adv/bat_v_elp.h b/net/batman-adv/bat_v_elp.h
index 5a7bc39..e95f1bc 100644
--- a/net/batman-adv/bat_v_elp.h
+++ b/net/batman-adv/bat_v_elp.h
@@ -21,11 +21,13 @@
 #define _NET_BATMAN_ADV_BAT_V_ELP_H_
 
 struct sk_buff;
+struct work_struct;
 
 int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface);
 void batadv_v_elp_iface_disable(struct batadv_hard_iface *hard_iface);
 void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface);
 int batadv_v_elp_packet_recv(struct sk_buff *skb,
 			     struct batadv_hard_iface *if_incoming);
+void batadv_v_elp_throughput_metric_update(struct work_struct *work);
 
 #endif /* _NET_BATMAN_ADV_BAT_V_ELP_H_ */
diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c
index c800096..d9bcbe6 100644
--- a/net/batman-adv/bat_v_ogm.c
+++ b/net/batman-adv/bat_v_ogm.c
@@ -339,7 +339,7 @@ static u32 batadv_v_forward_penalty(struct batadv_priv *bat_priv,
 	 */
 	if ((throughput > 10) &&
 	    (if_incoming == if_outgoing) &&
-	    (batadv_is_wifi_netdev(if_incoming->net_dev)))
+	    !(if_incoming->bat_v.flags & BATADV_FULL_DUPLEX))
 		return throughput / 2;
 
 	/* hop penalty of 255 equals 100% */
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 2e79544..5c68bf2 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -62,6 +62,7 @@
 #define BATADV_TQ_TOTAL_BIDRECT_LIMIT 1
 
 /* B.A.T.M.A.N. V */
+#define BATADV_THROUGHPUT_DEFAULT_VALUE 10 /* 1 Mbps */
 #define BATADV_ELP_MAX_AGE 64
 #define BATADV_OGM_MAX_ORIGDIFF 5
 #define BATADV_OGM_MAX_AGE 64
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index c56bb88..9abfb3e 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -87,12 +87,25 @@ struct batadv_hard_iface_bat_iv {
 };
 
 /**
+ * enum batadv_v_hard_iface_flags - interface flags useful to B.A.T.M.A.N. V
+ * @BATADV_FULL_DUPLEX: tells if the connection over this link is full-duplex
+ * @BATADV_WARNING_DEFAULT: tells whether we have warned the user that no
+ *  throughput data is available for this interface and that default values are
+ *  assumed.
+ */
+enum batadv_v_hard_iface_flags {
+	BATADV_FULL_DUPLEX	= BIT(0),
+	BATADV_WARNING_DEFAULT	= BIT(1),
+};
+
+/**
  * struct batadv_hard_iface_bat_v - per hard-interface B.A.T.M.A.N. V data
  * @elp_interval: time interval between two ELP transmissions
  * @elp_seqno: current ELP sequence number
  * @elp_skb: base skb containing the ELP message to send
  * @elp_wq: workqueue used to schedule ELP transmissions
  * @throughput_override: throughput override to disable link auto-detection
+ * @flags: interface specific flags
  */
 struct batadv_hard_iface_bat_v {
 	atomic_t elp_interval;
@@ -100,6 +113,7 @@ struct batadv_hard_iface_bat_v {
 	struct sk_buff *elp_skb;
 	struct delayed_work elp_wq;
 	atomic_t throughput_override;
+	u8 flags;
 };
 
 /**
@@ -378,12 +392,14 @@ DECLARE_EWMA(throughput, 1024, 8)
  * @elp_interval: time interval between two ELP transmissions
  * @elp_latest_seqno: latest and best known ELP sequence number
  * @last_unicast_tx: when the last unicast packet has been sent to this neighbor
+ * @metric_work: work queue callback item for metric update
  */
 struct batadv_hardif_neigh_node_bat_v {
 	struct ewma_throughput throughput;
 	u32 elp_interval;
 	u32 elp_latest_seqno;
 	unsigned long last_unicast_tx;
+	struct work_struct metric_work;
 };
 
 /**
-- 
cgit v0.10.2


From 8d2d499e08145d9851097e1241ef15aad8c9170a Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Tue, 10 Nov 2015 18:51:22 +0100
Subject: batman-adv: ELP - send unicast ELP packets for throughput sampling

In case of an unused wireless link, the mac80211 throughput estimation
won't get updated further. Consequently, the reported throughput metric
will become obsolete.

With this patch unicast sampling is introduced by periodically sending
unicast ELP packets to each neighbor on idle WiFi links. These sampling
packets will fill an entire frame, so that the measurement is as
reliable as possible

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index 8376730..3844e7e 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -165,6 +165,69 @@ void batadv_v_elp_throughput_metric_update(struct work_struct *work)
 }
 
 /**
+ * batadv_v_elp_wifi_neigh_probe - send link probing packets to a neighbour
+ * @neigh: the neighbour to probe
+ *
+ * Sends a predefined number of unicast wifi packets to a given neighbour in
+ * order to trigger the throughput estimation on this link by the RC algorithm.
+ * Packets are sent only if there there is not enough payload unicast traffic
+ * towards this neighbour..
+ *
+ * Return: True on success and false in case of error during skb preparation.
+ */
+static bool
+batadv_v_elp_wifi_neigh_probe(struct batadv_hardif_neigh_node *neigh)
+{
+	struct batadv_hard_iface *hard_iface = neigh->if_incoming;
+	struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface);
+	unsigned long last_tx_diff;
+	struct sk_buff *skb;
+	int probe_len, i;
+	int elp_skb_len;
+
+	/* this probing routine is for Wifi neighbours only */
+	if (!batadv_is_wifi_netdev(hard_iface->net_dev))
+		return true;
+
+	/* probe the neighbor only if no unicast packets have been sent
+	 * to it in the last 100 milliseconds: this is the rate control
+	 * algorithm sampling interval (minstrel). In this way, if not
+	 * enough traffic has been sent to the neighbor, batman-adv can
+	 * generate 2 probe packets and push the RC algorithm to perform
+	 * the sampling
+	 */
+	last_tx_diff = jiffies_to_msecs(jiffies - neigh->bat_v.last_unicast_tx);
+	if (last_tx_diff <= BATADV_ELP_PROBE_MAX_TX_DIFF)
+		return true;
+
+	probe_len = max_t(int, sizeof(struct batadv_elp_packet),
+			  BATADV_ELP_MIN_PROBE_SIZE);
+
+	for (i = 0; i < BATADV_ELP_PROBES_PER_NODE; i++) {
+		elp_skb_len = hard_iface->bat_v.elp_skb->len;
+		skb = skb_copy_expand(hard_iface->bat_v.elp_skb, 0,
+				      probe_len - elp_skb_len,
+				      GFP_ATOMIC);
+		if (!skb)
+			return false;
+
+		/* Tell the skb to get as big as the allocated space (we want
+		 * the packet to be exactly of that size to make the link
+		 * throughput estimation effective.
+		 */
+		skb_put(skb, probe_len - hard_iface->bat_v.elp_skb->len);
+
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "Sending unicast (probe) ELP packet on interface %s to %pM\n",
+			   hard_iface->net_dev->name, neigh->addr);
+
+		batadv_send_skb_packet(skb, hard_iface, neigh->addr);
+	}
+
+	return true;
+}
+
+/**
  * batadv_v_elp_periodic_work - ELP periodic task per interface
  * @work: work queue item
  *
@@ -227,6 +290,12 @@ static void batadv_v_elp_periodic_work(struct work_struct *work)
 	 */
 	rcu_read_lock();
 	hlist_for_each_entry_rcu(hardif_neigh, &hard_iface->neigh_list, list) {
+		if (!batadv_v_elp_wifi_neigh_probe(hardif_neigh))
+			/* if something goes wrong while probing, better to stop
+			 * sending packets immediately and reschedule the task
+			 */
+			break;
+
 		if (!kref_get_unless_zero(&hardif_neigh->refcount))
 			continue;
 
diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 5c68bf2..878c359 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -63,6 +63,9 @@
 
 /* B.A.T.M.A.N. V */
 #define BATADV_THROUGHPUT_DEFAULT_VALUE 10 /* 1 Mbps */
+#define BATADV_ELP_PROBES_PER_NODE 2
+#define BATADV_ELP_MIN_PROBE_SIZE 200 /* bytes */
+#define BATADV_ELP_PROBE_MAX_TX_DIFF 100 /* milliseconds */
 #define BATADV_ELP_MAX_AGE 64
 #define BATADV_OGM_MAX_ORIGDIFF 5
 #define BATADV_OGM_MAX_AGE 64
-- 
cgit v0.10.2


From 9786906022eba35763b17c54a35913ca65151a78 Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Sat, 16 Jan 2016 16:40:17 +0800
Subject: batman-adv: B.A.T.M.A.N. V - implement neighbor comparison API calls

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index ff31f2a..953c0d1 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -19,12 +19,15 @@
 #include "main.h"
 
 #include <linux/atomic.h>
+#include <linux/bug.h>
 #include <linux/cache.h>
 #include <linux/init.h>
+#include <linux/types.h>
 #include <linux/workqueue.h>
 
 #include "bat_v_elp.h"
 #include "bat_v_ogm.h"
+#include "originator.h"
 #include "packet.h"
 
 static int batadv_v_iface_enable(struct batadv_hard_iface *hard_iface)
@@ -78,6 +81,39 @@ static void batadv_v_ogm_emit(struct batadv_forw_packet *forw_packet)
 {
 }
 
+static int batadv_v_neigh_cmp(struct batadv_neigh_node *neigh1,
+			      struct batadv_hard_iface *if_outgoing1,
+			      struct batadv_neigh_node *neigh2,
+			      struct batadv_hard_iface *if_outgoing2)
+{
+	struct batadv_neigh_ifinfo *ifinfo1, *ifinfo2;
+
+	ifinfo1 = batadv_neigh_ifinfo_get(neigh1, if_outgoing1);
+	ifinfo2 = batadv_neigh_ifinfo_get(neigh2, if_outgoing2);
+
+	if (WARN_ON(!ifinfo1 || !ifinfo2))
+		return 0;
+
+	return ifinfo1->bat_v.throughput - ifinfo2->bat_v.throughput;
+}
+
+static bool batadv_v_neigh_is_sob(struct batadv_neigh_node *neigh1,
+				  struct batadv_hard_iface *if_outgoing1,
+				  struct batadv_neigh_node *neigh2,
+				  struct batadv_hard_iface *if_outgoing2)
+{
+	struct batadv_neigh_ifinfo *ifinfo1, *ifinfo2;
+	u32 threshold;
+
+	ifinfo1 = batadv_neigh_ifinfo_get(neigh1, if_outgoing1);
+	ifinfo2 = batadv_neigh_ifinfo_get(neigh2, if_outgoing2);
+
+	threshold = ifinfo1->bat_v.throughput / 4;
+	threshold = ifinfo1->bat_v.throughput - threshold;
+
+	return ifinfo2->bat_v.throughput > threshold;
+}
+
 static struct batadv_algo_ops batadv_batman_v __read_mostly = {
 	.name = "BATMAN_V",
 	.bat_iface_enable = batadv_v_iface_enable,
@@ -87,6 +123,8 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = {
 	.bat_hardif_neigh_init = batadv_v_hardif_neigh_init,
 	.bat_ogm_emit = batadv_v_ogm_emit,
 	.bat_ogm_schedule = batadv_v_ogm_schedule,
+	.bat_neigh_cmp = batadv_v_neigh_cmp,
+	.bat_neigh_is_similar_or_better = batadv_v_neigh_is_sob,
 };
 
 /**
-- 
cgit v0.10.2


From 261e264db636ae1f4c43e56b8c57d7343b166fc9 Mon Sep 17 00:00:00 2001
From: Antonio Quartulli <antonio@open-mesh.com>
Date: Sat, 16 Jan 2016 16:40:18 +0800
Subject: batman-adv: B.A.T.M.A.N. V - implement bat_orig_print API

Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>

diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index 953c0d1..a90117c 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -22,11 +22,17 @@
 #include <linux/bug.h>
 #include <linux/cache.h>
 #include <linux/init.h>
+#include <linux/jiffies.h>
+#include <linux/netdevice.h>
+#include <linux/rculist.h>
+#include <linux/rcupdate.h>
+#include <linux/seq_file.h>
 #include <linux/types.h>
 #include <linux/workqueue.h>
 
 #include "bat_v_elp.h"
 #include "bat_v_ogm.h"
+#include "hash.h"
 #include "originator.h"
 #include "packet.h"
 
@@ -81,6 +87,104 @@ static void batadv_v_ogm_emit(struct batadv_forw_packet *forw_packet)
 {
 }
 
+/**
+ * batadv_v_orig_print_neigh - print neighbors for the originator table
+ * @orig_node: the orig_node for which the neighbors are printed
+ * @if_outgoing: outgoing interface for these entries
+ * @seq: debugfs table seq_file struct
+ *
+ * Must be called while holding an rcu lock.
+ */
+static void
+batadv_v_orig_print_neigh(struct batadv_orig_node *orig_node,
+			  struct batadv_hard_iface *if_outgoing,
+			  struct seq_file *seq)
+{
+	struct batadv_neigh_node *neigh_node;
+	struct batadv_neigh_ifinfo *n_ifinfo;
+
+	hlist_for_each_entry_rcu(neigh_node, &orig_node->neigh_list, list) {
+		n_ifinfo = batadv_neigh_ifinfo_get(neigh_node, if_outgoing);
+		if (!n_ifinfo)
+			continue;
+
+		seq_printf(seq, " %pM (%9u.%1u)",
+			   neigh_node->addr,
+			   n_ifinfo->bat_v.throughput / 10,
+			   n_ifinfo->bat_v.throughput % 10);
+
+		batadv_neigh_ifinfo_put(n_ifinfo);
+	}
+}
+
+/**
+ * batadv_v_orig_print - print the originator table
+ * @bat_priv: the bat priv with all the soft interface information
+ * @seq: debugfs table seq_file struct
+ * @if_outgoing: the outgoing interface for which this should be printed
+ */
+static void batadv_v_orig_print(struct batadv_priv *bat_priv,
+				struct seq_file *seq,
+				struct batadv_hard_iface *if_outgoing)
+{
+	struct batadv_neigh_node *neigh_node;
+	struct batadv_hashtable *hash = bat_priv->orig_hash;
+	int last_seen_msecs, last_seen_secs;
+	struct batadv_orig_node *orig_node;
+	struct batadv_neigh_ifinfo *n_ifinfo;
+	unsigned long last_seen_jiffies;
+	struct hlist_head *head;
+	int batman_count = 0;
+	u32 i;
+
+	seq_printf(seq, "  %-15s %s (%11s) %17s [%10s]: %20s ...\n",
+		   "Originator", "last-seen", "throughput", "Nexthop",
+		   "outgoingIF", "Potential nexthops");
+
+	for (i = 0; i < hash->size; i++) {
+		head = &hash->table[i];
+
+		rcu_read_lock();
+		hlist_for_each_entry_rcu(orig_node, head, hash_entry) {
+			neigh_node = batadv_orig_router_get(orig_node,
+							    if_outgoing);
+			if (!neigh_node)
+				continue;
+
+			n_ifinfo = batadv_neigh_ifinfo_get(neigh_node,
+							   if_outgoing);
+			if (!n_ifinfo)
+				goto next;
+
+			last_seen_jiffies = jiffies - orig_node->last_seen;
+			last_seen_msecs = jiffies_to_msecs(last_seen_jiffies);
+			last_seen_secs = last_seen_msecs / 1000;
+			last_seen_msecs = last_seen_msecs % 1000;
+
+			seq_printf(seq, "%pM %4i.%03is (%9u.%1u) %pM [%10s]:",
+				   orig_node->orig, last_seen_secs,
+				   last_seen_msecs,
+				   n_ifinfo->bat_v.throughput / 10,
+				   n_ifinfo->bat_v.throughput % 10,
+				   neigh_node->addr,
+				   neigh_node->if_incoming->net_dev->name);
+
+			batadv_v_orig_print_neigh(orig_node, if_outgoing, seq);
+			seq_puts(seq, "\n");
+			batman_count++;
+
+next:
+			batadv_neigh_node_put(neigh_node);
+			if (n_ifinfo)
+				batadv_neigh_ifinfo_put(n_ifinfo);
+		}
+		rcu_read_unlock();
+	}
+
+	if (batman_count == 0)
+		seq_puts(seq, "No batman nodes in range ...\n");
+}
+
 static int batadv_v_neigh_cmp(struct batadv_neigh_node *neigh1,
 			      struct batadv_hard_iface *if_outgoing1,
 			      struct batadv_neigh_node *neigh2,
@@ -123,6 +227,7 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = {
 	.bat_hardif_neigh_init = batadv_v_hardif_neigh_init,
 	.bat_ogm_emit = batadv_v_ogm_emit,
 	.bat_ogm_schedule = batadv_v_ogm_schedule,
+	.bat_orig_print = batadv_v_orig_print,
 	.bat_neigh_cmp = batadv_v_neigh_cmp,
 	.bat_neigh_is_similar_or_better = batadv_v_neigh_is_sob,
 };
-- 
cgit v0.10.2


From 626d23e83c88df5ff535414c2fe29e16b95d6b7a Mon Sep 17 00:00:00 2001
From: Linus Luessing <linus.luessing@web.de>
Date: Sat, 16 Jan 2016 16:40:19 +0800
Subject: batman-adv: B.A.T.M.A.N. V - implement bat_neigh_print API

Lists all neighbours detected by the Echo Locating Protocol
(ELP) and their throughput metric.

Initially Developed by Linus during a 6 months trainee study
period in Ascom (Switzerland) AG.

Signed-off-by: Linus Luessing <linus.luessing@web.de>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <antonio@open-mesh.com>

diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c
index a90117c..3315b9a 100644
--- a/net/batman-adv/bat_v.c
+++ b/net/batman-adv/bat_v.c
@@ -118,6 +118,60 @@ batadv_v_orig_print_neigh(struct batadv_orig_node *orig_node,
 }
 
 /**
+ * batadv_v_hardif_neigh_print - print a single ELP neighbour node
+ * @seq: neighbour table seq_file struct
+ * @hardif_neigh: hardif neighbour information
+ */
+static void
+batadv_v_hardif_neigh_print(struct seq_file *seq,
+			    struct batadv_hardif_neigh_node *hardif_neigh)
+{
+	int last_secs, last_msecs;
+	u32 throughput;
+
+	last_secs = jiffies_to_msecs(jiffies - hardif_neigh->last_seen) / 1000;
+	last_msecs = jiffies_to_msecs(jiffies - hardif_neigh->last_seen) % 1000;
+	throughput = ewma_throughput_read(&hardif_neigh->bat_v.throughput);
+
+	seq_printf(seq, "%pM %4i.%03is (%9u.%1u) [%10s]\n",
+		   hardif_neigh->addr, last_secs, last_msecs, throughput / 10,
+		   throughput % 10, hardif_neigh->if_incoming->net_dev->name);
+}
+
+/**
+ * batadv_v_neigh_print - print the single hop neighbour list
+ * @bat_priv: the bat priv with all the soft interface information
+ * @seq: neighbour table seq_file struct
+ */
+static void batadv_v_neigh_print(struct batadv_priv *bat_priv,
+				 struct seq_file *seq)
+{
+	struct net_device *net_dev = (struct net_device *)seq->private;
+	struct batadv_hardif_neigh_node *hardif_neigh;
+	struct batadv_hard_iface *hard_iface;
+	int batman_count = 0;
+
+	seq_printf(seq, "  %-15s %s (%11s) [%10s]\n", "Neighbor",
+		   "last-seen", "throughput", "IF");
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(hard_iface, &batadv_hardif_list, list) {
+		if (hard_iface->soft_iface != net_dev)
+			continue;
+
+		hlist_for_each_entry_rcu(hardif_neigh,
+					 &hard_iface->neigh_list, list) {
+			batadv_v_hardif_neigh_print(seq, hardif_neigh);
+			batman_count++;
+		}
+	}
+	rcu_read_unlock();
+
+	if (batman_count == 0)
+		seq_puts(seq, "No batman nodes in range ...\n");
+}
+
+/**
  * batadv_v_orig_print - print the originator table
  * @bat_priv: the bat priv with all the soft interface information
  * @seq: debugfs table seq_file struct
@@ -230,6 +284,7 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = {
 	.bat_orig_print = batadv_v_orig_print,
 	.bat_neigh_cmp = batadv_v_neigh_cmp,
 	.bat_neigh_is_similar_or_better = batadv_v_neigh_is_sob,
+	.bat_neigh_print = batadv_v_neigh_print,
 };
 
 /**
-- 
cgit v0.10.2


From 97575407e9547416a23de03d141072e639f04623 Mon Sep 17 00:00:00 2001
From: Simon Wunderlich <sw@simonwunderlich.de>
Date: Tue, 19 Jan 2016 17:08:29 +0100
Subject: batman-adv: Start new development cycle

Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h
index 878c359..db45336 100644
--- a/net/batman-adv/main.h
+++ b/net/batman-adv/main.h
@@ -24,7 +24,7 @@
 #define BATADV_DRIVER_DEVICE "batman-adv"
 
 #ifndef BATADV_SOURCE_VERSION
-#define BATADV_SOURCE_VERSION "2016.0"
+#define BATADV_SOURCE_VERSION "2016.1"
 #endif
 
 /* B.A.T.M.A.N. parameters */
-- 
cgit v0.10.2


From 8b823170550e5dcc80512143088c9f9e24598f9a Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Sun, 21 Feb 2016 16:40:55 +0100
Subject: MAINTAINERS: Add patchwork URL for BATMAN ADVANCED

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Antonio Quartulli <a@unstable.cc>

diff --git a/MAINTAINERS b/MAINTAINERS
index 27393cf..4fe223a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2152,6 +2152,7 @@ M:	Simon Wunderlich <sw@simonwunderlich.de>
 M:	Antonio Quartulli <a@unstable.cc>
 L:	b.a.t.m.a.n@lists.open-mesh.org
 W:	https://www.open-mesh.org/
+Q:	https://patchwork.open-mesh.org/project/batman/list/
 S:	Maintained
 F:	net/batman-adv/
 
-- 
cgit v0.10.2


From 017b1b6d28c479f1ad9a7a41f775545a3e1cba35 Mon Sep 17 00:00:00 2001
From: Phil Turnbull <phil.turnbull@oracle.com>
Date: Wed, 24 Feb 2016 15:34:43 -0500
Subject: netfilter: nfnetlink_acct: validate NFACCT_FILTER parameters

nfacct_filter_alloc doesn't validate the NFACCT_FILTER_MASK and
NFACCT_FILTER_VALUE parameters which can trigger a NULL pointer
dereference. CAP_NET_ADMIN is required to trigger the bug.

Signed-off-by: Phil Turnbull <phil.turnbull@oracle.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c
index 5274b04..4c2b4c0 100644
--- a/net/netfilter/nfnetlink_acct.c
+++ b/net/netfilter/nfnetlink_acct.c
@@ -242,6 +242,9 @@ nfacct_filter_alloc(const struct nlattr * const attr)
 	if (err < 0)
 		return ERR_PTR(err);
 
+	if (!tb[NFACCT_FILTER_MASK] || !tb[NFACCT_FILTER_VALUE])
+		return ERR_PTR(-EINVAL);
+
 	filter = kzalloc(sizeof(struct nfacct_filter), GFP_KERNEL);
 	if (!filter)
 		return ERR_PTR(-ENOMEM);
-- 
cgit v0.10.2


From b07edbe1cf3dae9ba81f24888e2f2a9dbe778918 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 16 Feb 2016 17:24:08 +0100
Subject: netfilter: meta: add PRANDOM support

Can be used to randomly match packets e.g. for statistic traffic sampling.

See commit 3ad0040573b0c00f8848
("bpf: split state from prandom_u32() and consolidate {c, e}BPF prngs")
for more info why this doesn't use prandom_u32 directly.

Unlike bpf nft_meta can be built as a module, so add an EXPORT_SYMBOL
for prandom_seed_full_state too.

Cc: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index be41ffc..b19be0a 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -681,6 +681,7 @@ enum nft_exthdr_attributes {
  * @NFT_META_IIFGROUP: packet input interface group
  * @NFT_META_OIFGROUP: packet output interface group
  * @NFT_META_CGROUP: socket control group (skb->sk->sk_classid)
+ * @NFT_META_PRANDOM: a 32bit pseudo-random number
  */
 enum nft_meta_keys {
 	NFT_META_LEN,
@@ -707,6 +708,7 @@ enum nft_meta_keys {
 	NFT_META_IIFGROUP,
 	NFT_META_OIFGROUP,
 	NFT_META_CGROUP,
+	NFT_META_PRANDOM,
 };
 
 /**
diff --git a/lib/random32.c b/lib/random32.c
index 1211191..510d1ce 100644
--- a/lib/random32.c
+++ b/lib/random32.c
@@ -255,6 +255,7 @@ void prandom_seed_full_state(struct rnd_state __percpu *pcpu_state)
 		prandom_warmup(state);
 	}
 }
+EXPORT_SYMBOL(prandom_seed_full_state);
 
 /*
  *	Generate better values after random number generator
diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c
index fe885bf..16c50b0 100644
--- a/net/netfilter/nft_meta.c
+++ b/net/netfilter/nft_meta.c
@@ -28,6 +28,8 @@
 
 #include <uapi/linux/netfilter_bridge.h> /* NF_BR_PRE_ROUTING */
 
+static DEFINE_PER_CPU(struct rnd_state, nft_prandom_state);
+
 void nft_meta_get_eval(const struct nft_expr *expr,
 		       struct nft_regs *regs,
 		       const struct nft_pktinfo *pkt)
@@ -181,6 +183,11 @@ void nft_meta_get_eval(const struct nft_expr *expr,
 		*dest = sock_cgroup_classid(&sk->sk_cgrp_data);
 		break;
 #endif
+	case NFT_META_PRANDOM: {
+		struct rnd_state *state = this_cpu_ptr(&nft_prandom_state);
+		*dest = prandom_u32_state(state);
+		break;
+	}
 	default:
 		WARN_ON(1);
 		goto err;
@@ -277,6 +284,10 @@ int nft_meta_get_init(const struct nft_ctx *ctx,
 	case NFT_META_OIFNAME:
 		len = IFNAMSIZ;
 		break;
+	case NFT_META_PRANDOM:
+		prandom_init_once(&nft_prandom_state);
+		len = sizeof(u32);
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
-- 
cgit v0.10.2


From 8ee225e78531981e479e79b74354b63e188b4ee2 Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Date: Tue, 23 Feb 2016 13:40:10 +0530
Subject: netfilter: xt_osf: remove unused variable

While building with W=1 we got the warning:
net/netfilter/xt_osf.c:265:9: warning: variable 'loop_cont' set but not used

The local variable loop_cont was only initialized and then assigned a
value but was never used or checked after that.
While removing the variable, the case of OSFOPT_TS was not removed so
that it will serve as a reminder to us that we can do something in that
particular case.

Signed-off-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/netfilter/xt_osf.c b/net/netfilter/xt_osf.c
index 4e3c3af..2455b69 100644
--- a/net/netfilter/xt_osf.c
+++ b/net/netfilter/xt_osf.c
@@ -262,7 +262,6 @@ xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p)
 			if (f->opt[optnum].kind == (*optp)) {
 				__u32 len = f->opt[optnum].length;
 				const __u8 *optend = optp + len;
-				int loop_cont = 0;
 
 				fmatch = FMATCH_OK;
 
@@ -275,7 +274,6 @@ xt_osf_match_packet(const struct sk_buff *skb, struct xt_action_param *p)
 					mss = ntohs((__force __be16)mss);
 					break;
 				case OSFOPT_TS:
-					loop_cont = 1;
 					break;
 				}
 
-- 
cgit v0.10.2


From 34bf1912bfc06bd9200893916078eb0f16480a95 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Sun, 28 Feb 2016 21:25:19 +0100
Subject: Bluetooth: hci_uart: Add diag and address support for Intel/AG6xx

The AG6xx devices behave similar to Wilkens Peak and Stone Peak and with
that it is needed to check for Intel default address. In addition it is
possible to enable vendor events and diag support.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>

diff --git a/drivers/bluetooth/hci_ag6xx.c b/drivers/bluetooth/hci_ag6xx.c
index ea65c2d..6923d17 100644
--- a/drivers/bluetooth/hci_ag6xx.c
+++ b/drivers/bluetooth/hci_ag6xx.c
@@ -174,6 +174,9 @@ static int ag6xx_setup(struct hci_uart *hu)
 	bool patched = false;
 	int err;
 
+	hu->hdev->set_diag = btintel_set_diag;
+	hu->hdev->set_bdaddr = btintel_set_bdaddr;
+
 	err = btintel_enter_mfg(hdev);
 	if (err)
 		return err;
@@ -298,8 +301,16 @@ patch:
 complete:
 	/* Exit manufacturing mode and reset */
 	err = btintel_exit_mfg(hdev, true, patched);
+	if (err)
+		return err;
 
-	return err;
+	/* Set the event mask for Intel specific vendor events. This enables
+	 * a few extra events that are useful during general operation.
+	 */
+	btintel_set_event_mask_mfg(hdev, false);
+
+	btintel_check_bdaddr(hdev);
+	return 0;
 }
 
 static const struct hci_uart_proto ag6xx_proto = {
-- 
cgit v0.10.2


From 86a7996cc8a078793670d82ed97d5a99bb4e8496 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Thu, 25 Feb 2016 14:55:00 -0800
Subject: net_sched: introduce qdisc_replace() helper

Remove nearly duplicated code and prepare for the following patch.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 636a362..8fdad9f 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -707,6 +707,23 @@ static inline void qdisc_reset_queue(struct Qdisc *sch)
 	sch->qstats.backlog = 0;
 }
 
+static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new,
+					  struct Qdisc **pold)
+{
+	struct Qdisc *old;
+
+	sch_tree_lock(sch);
+	old = *pold;
+	*pold = new;
+	if (old != NULL) {
+		qdisc_tree_decrease_qlen(old, old->q.qlen);
+		qdisc_reset(old);
+	}
+	sch_tree_unlock(sch);
+
+	return old;
+}
+
 static inline unsigned int __qdisc_queue_drop(struct Qdisc *sch,
 					      struct sk_buff_head *list)
 {
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index c538d9e..7f8474c 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1624,13 +1624,8 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 			new->reshape_fail = cbq_reshape_fail;
 #endif
 	}
-	sch_tree_lock(sch);
-	*old = cl->q;
-	cl->q = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
 
+	*old = qdisc_replace(sch, new, &cl->q);
 	return 0;
 }
 
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index a1cd778..b96c9a8 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -226,11 +226,7 @@ static int drr_graft_class(struct Qdisc *sch, unsigned long arg,
 			new = &noop_qdisc;
 	}
 
-	sch_tree_lock(sch);
-	drr_purge_queue(cl);
-	*old = cl->qdisc;
-	cl->qdisc = new;
-	sch_tree_unlock(sch);
+	*old = qdisc_replace(sch, new, &cl->qdisc);
 	return 0;
 }
 
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index f357f34..cfddb1c 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -73,13 +73,7 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg,
 			new = &noop_qdisc;
 	}
 
-	sch_tree_lock(sch);
-	*old = p->q;
-	p->q = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
-
+	*old = qdisc_replace(sch, new, &p->q);
 	return 0;
 }
 
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index b7ebe2c..089f3b6 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1215,11 +1215,7 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 			new = &noop_qdisc;
 	}
 
-	sch_tree_lock(sch);
-	hfsc_purge_queue(sch, cl);
-	*old = cl->qdisc;
-	cl->qdisc = new;
-	sch_tree_unlock(sch);
+	*old = qdisc_replace(sch, new, &cl->qdisc);
 	return 0;
 }
 
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 15ccd7f..0efbcf3 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1163,14 +1163,7 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 				     cl->common.classid)) == NULL)
 		return -ENOBUFS;
 
-	sch_tree_lock(sch);
-	*old = cl->un.leaf.q;
-	cl->un.leaf.q = new;
-	if (*old != NULL) {
-		qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-		qdisc_reset(*old);
-	}
-	sch_tree_unlock(sch);
+	*old = qdisc_replace(sch, new, &cl->un.leaf.q);
 	return 0;
 }
 
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index 4e904ca..a0103a1 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -303,13 +303,7 @@ static int multiq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (new == NULL)
 		new = &noop_qdisc;
 
-	sch_tree_lock(sch);
-	*old = q->queues[band];
-	q->queues[band] = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
-
+	*old = qdisc_replace(sch, new, &q->queues[band]);
 	return 0;
 }
 
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 5abd1d9..0a6ddaf 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -1037,15 +1037,7 @@ static int netem_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 {
 	struct netem_sched_data *q = qdisc_priv(sch);
 
-	sch_tree_lock(sch);
-	*old = q->qdisc;
-	q->qdisc = new;
-	if (*old) {
-		qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-		qdisc_reset(*old);
-	}
-	sch_tree_unlock(sch);
-
+	*old = qdisc_replace(sch, new, &q->qdisc);
 	return 0;
 }
 
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index ba6487f..1b4aaec 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -268,13 +268,7 @@ static int prio_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (new == NULL)
 		new = &noop_qdisc;
 
-	sch_tree_lock(sch);
-	*old = q->queues[band];
-	q->queues[band] = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
-
+	*old = qdisc_replace(sch, new, &q->queues[band]);
 	return 0;
 }
 
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 3dc3a6e..b5c52ca 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -617,11 +617,7 @@ static int qfq_graft_class(struct Qdisc *sch, unsigned long arg,
 			new = &noop_qdisc;
 	}
 
-	sch_tree_lock(sch);
-	qfq_purge_queue(cl);
-	*old = cl->qdisc;
-	cl->qdisc = new;
-	sch_tree_unlock(sch);
+	*old = qdisc_replace(sch, new, &cl->qdisc);
 	return 0;
 }
 
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 6c0534c..d5abcee 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -313,12 +313,7 @@ static int red_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (new == NULL)
 		new = &noop_qdisc;
 
-	sch_tree_lock(sch);
-	*old = q->qdisc;
-	q->qdisc = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
+	*old = qdisc_replace(sch, new, &q->qdisc);
 	return 0;
 }
 
diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c
index 5bbb633..0e74e55 100644
--- a/net/sched/sch_sfb.c
+++ b/net/sched/sch_sfb.c
@@ -606,12 +606,7 @@ static int sfb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (new == NULL)
 		new = &noop_qdisc;
 
-	sch_tree_lock(sch);
-	*old = q->qdisc;
-	q->qdisc = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
+	*old = qdisc_replace(sch, new, &q->qdisc);
 	return 0;
 }
 
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index a4afde1..56a1aef 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -502,13 +502,7 @@ static int tbf_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (new == NULL)
 		new = &noop_qdisc;
 
-	sch_tree_lock(sch);
-	*old = q->qdisc;
-	q->qdisc = new;
-	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
-	qdisc_reset(*old);
-	sch_tree_unlock(sch);
-
+	*old = qdisc_replace(sch, new, &q->qdisc);
 	return 0;
 }
 
-- 
cgit v0.10.2


From 2ccccf5fb43ff62b2b96cc58d95fc0b3596516e4 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Thu, 25 Feb 2016 14:55:01 -0800
Subject: net_sched: update hierarchical backlog too

When the bottom qdisc decides to, for example, drop some packet,
it calls qdisc_tree_decrease_qlen() to update the queue length
for all its ancestors, we need to update the backlog too to
keep the stats on root qdisc accurate.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/codel.h b/include/net/codel.h
index 267e702..d168aca 100644
--- a/include/net/codel.h
+++ b/include/net/codel.h
@@ -162,12 +162,14 @@ struct codel_vars {
  * struct codel_stats - contains codel shared variables and stats
  * @maxpacket:	largest packet we've seen so far
  * @drop_count:	temp count of dropped packets in dequeue()
+ * @drop_len:	bytes of dropped packets in dequeue()
  * ecn_mark:	number of packets we ECN marked instead of dropping
  * ce_mark:	number of packets CE marked because sojourn time was above ce_threshold
  */
 struct codel_stats {
 	u32		maxpacket;
 	u32		drop_count;
+	u32		drop_len;
 	u32		ecn_mark;
 	u32		ce_mark;
 };
@@ -308,6 +310,7 @@ static struct sk_buff *codel_dequeue(struct Qdisc *sch,
 								  vars->rec_inv_sqrt);
 					goto end;
 				}
+				stats->drop_len += qdisc_pkt_len(skb);
 				qdisc_drop(skb, sch);
 				stats->drop_count++;
 				skb = dequeue_func(vars, sch);
@@ -330,6 +333,7 @@ static struct sk_buff *codel_dequeue(struct Qdisc *sch,
 		if (params->ecn && INET_ECN_set_ce(skb)) {
 			stats->ecn_mark++;
 		} else {
+			stats->drop_len += qdisc_pkt_len(skb);
 			qdisc_drop(skb, sch);
 			stats->drop_count++;
 
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8fdad9f..e5bba89 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -396,7 +396,8 @@ struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue,
 			      struct Qdisc *qdisc);
 void qdisc_reset(struct Qdisc *qdisc);
 void qdisc_destroy(struct Qdisc *qdisc);
-void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n);
+void qdisc_tree_reduce_backlog(struct Qdisc *qdisc, unsigned int n,
+			       unsigned int len);
 struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 			  const struct Qdisc_ops *ops);
 struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue,
@@ -716,7 +717,7 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new,
 	old = *pold;
 	*pold = new;
 	if (old != NULL) {
-		qdisc_tree_decrease_qlen(old, old->q.qlen);
+		qdisc_tree_reduce_backlog(old, old->q.qlen, old->qstats.backlog);
 		qdisc_reset(old);
 	}
 	sch_tree_unlock(sch);
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index de1e176..3b180ff 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -744,14 +744,15 @@ static u32 qdisc_alloc_handle(struct net_device *dev)
 	return 0;
 }
 
-void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
+void qdisc_tree_reduce_backlog(struct Qdisc *sch, unsigned int n,
+			       unsigned int len)
 {
 	const struct Qdisc_class_ops *cops;
 	unsigned long cl;
 	u32 parentid;
 	int drops;
 
-	if (n == 0)
+	if (n == 0 && len == 0)
 		return;
 	drops = max_t(int, n, 0);
 	rcu_read_lock();
@@ -774,11 +775,12 @@ void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
 			cops->put(sch, cl);
 		}
 		sch->q.qlen -= n;
+		sch->qstats.backlog -= len;
 		__qdisc_qstats_drop(sch, drops);
 	}
 	rcu_read_unlock();
 }
-EXPORT_SYMBOL(qdisc_tree_decrease_qlen);
+EXPORT_SYMBOL(qdisc_tree_reduce_backlog);
 
 static void notify_and_destroy(struct net *net, struct sk_buff *skb,
 			       struct nlmsghdr *n, u32 clid,
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 7f8474c..baafddf 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1909,7 +1909,7 @@ static int cbq_delete(struct Qdisc *sch, unsigned long arg)
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
 	struct cbq_class *cl = (struct cbq_class *)arg;
-	unsigned int qlen;
+	unsigned int qlen, backlog;
 
 	if (cl->filters || cl->children || cl == &q->link)
 		return -EBUSY;
@@ -1917,8 +1917,9 @@ static int cbq_delete(struct Qdisc *sch, unsigned long arg)
 	sch_tree_lock(sch);
 
 	qlen = cl->q->q.qlen;
+	backlog = cl->q->qstats.backlog;
 	qdisc_reset(cl->q);
-	qdisc_tree_decrease_qlen(cl->q, qlen);
+	qdisc_tree_reduce_backlog(cl->q, qlen, backlog);
 
 	if (cl->next_alive)
 		cbq_deactivate_class(cl);
diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c
index 5ffb8b8..0a08c86 100644
--- a/net/sched/sch_choke.c
+++ b/net/sched/sch_choke.c
@@ -128,8 +128,8 @@ static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx)
 		choke_zap_tail_holes(q);
 
 	qdisc_qstats_backlog_dec(sch, skb);
+	qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb));
 	qdisc_drop(skb, sch);
-	qdisc_tree_decrease_qlen(sch, 1);
 	--sch->q.qlen;
 }
 
@@ -456,6 +456,7 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt)
 		old = q->tab;
 		if (old) {
 			unsigned int oqlen = sch->q.qlen, tail = 0;
+			unsigned dropped = 0;
 
 			while (q->head != q->tail) {
 				struct sk_buff *skb = q->tab[q->head];
@@ -467,11 +468,12 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt)
 					ntab[tail++] = skb;
 					continue;
 				}
+				dropped += qdisc_pkt_len(skb);
 				qdisc_qstats_backlog_dec(sch, skb);
 				--sch->q.qlen;
 				qdisc_drop(skb, sch);
 			}
-			qdisc_tree_decrease_qlen(sch, oqlen - sch->q.qlen);
+			qdisc_tree_reduce_backlog(sch, oqlen - sch->q.qlen, dropped);
 			q->head = 0;
 			q->tail = tail;
 		}
diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c
index 535007d..9b7e298 100644
--- a/net/sched/sch_codel.c
+++ b/net/sched/sch_codel.c
@@ -79,12 +79,13 @@ static struct sk_buff *codel_qdisc_dequeue(struct Qdisc *sch)
 
 	skb = codel_dequeue(sch, &q->params, &q->vars, &q->stats, dequeue);
 
-	/* We cant call qdisc_tree_decrease_qlen() if our qlen is 0,
+	/* We cant call qdisc_tree_reduce_backlog() if our qlen is 0,
 	 * or HTB crashes. Defer it for next round.
 	 */
 	if (q->stats.drop_count && sch->q.qlen) {
-		qdisc_tree_decrease_qlen(sch, q->stats.drop_count);
+		qdisc_tree_reduce_backlog(sch, q->stats.drop_count, q->stats.drop_len);
 		q->stats.drop_count = 0;
+		q->stats.drop_len = 0;
 	}
 	if (skb)
 		qdisc_bstats_update(sch, skb);
@@ -116,7 +117,7 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct codel_sched_data *q = qdisc_priv(sch);
 	struct nlattr *tb[TCA_CODEL_MAX + 1];
-	unsigned int qlen;
+	unsigned int qlen, dropped = 0;
 	int err;
 
 	if (!opt)
@@ -156,10 +157,11 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt)
 	while (sch->q.qlen > sch->limit) {
 		struct sk_buff *skb = __skb_dequeue(&sch->q);
 
+		dropped += qdisc_pkt_len(skb);
 		qdisc_qstats_backlog_dec(sch, skb);
 		qdisc_drop(skb, sch);
 	}
-	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
+	qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped);
 
 	sch_tree_unlock(sch);
 	return 0;
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index b96c9a8..a63e879 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -53,9 +53,10 @@ static struct drr_class *drr_find_class(struct Qdisc *sch, u32 classid)
 static void drr_purge_queue(struct drr_class *cl)
 {
 	unsigned int len = cl->qdisc->q.qlen;
+	unsigned int backlog = cl->qdisc->qstats.backlog;
 
 	qdisc_reset(cl->qdisc);
-	qdisc_tree_decrease_qlen(cl->qdisc, len);
+	qdisc_tree_reduce_backlog(cl->qdisc, len, backlog);
 }
 
 static const struct nla_policy drr_policy[TCA_DRR_MAX + 1] = {
diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index 109b232..3c6a47d 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -662,6 +662,7 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt)
 	struct fq_sched_data *q = qdisc_priv(sch);
 	struct nlattr *tb[TCA_FQ_MAX + 1];
 	int err, drop_count = 0;
+	unsigned drop_len = 0;
 	u32 fq_log;
 
 	if (!opt)
@@ -736,10 +737,11 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt)
 
 		if (!skb)
 			break;
+		drop_len += qdisc_pkt_len(skb);
 		kfree_skb(skb);
 		drop_count++;
 	}
-	qdisc_tree_decrease_qlen(sch, drop_count);
+	qdisc_tree_reduce_backlog(sch, drop_count, drop_len);
 
 	sch_tree_unlock(sch);
 	return err;
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 4c834e9..d3fc8f9 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -175,7 +175,7 @@ static unsigned int fq_codel_qdisc_drop(struct Qdisc *sch)
 static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	struct fq_codel_sched_data *q = qdisc_priv(sch);
-	unsigned int idx;
+	unsigned int idx, prev_backlog;
 	struct fq_codel_flow *flow;
 	int uninitialized_var(ret);
 
@@ -203,6 +203,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	if (++sch->q.qlen <= sch->limit)
 		return NET_XMIT_SUCCESS;
 
+	prev_backlog = sch->qstats.backlog;
 	q->drop_overlimit++;
 	/* Return Congestion Notification only if we dropped a packet
 	 * from this flow.
@@ -211,7 +212,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return NET_XMIT_CN;
 
 	/* As we dropped a packet, better let upper stack know this */
-	qdisc_tree_decrease_qlen(sch, 1);
+	qdisc_tree_reduce_backlog(sch, 1, prev_backlog - sch->qstats.backlog);
 	return NET_XMIT_SUCCESS;
 }
 
@@ -241,6 +242,7 @@ static struct sk_buff *fq_codel_dequeue(struct Qdisc *sch)
 	struct fq_codel_flow *flow;
 	struct list_head *head;
 	u32 prev_drop_count, prev_ecn_mark;
+	unsigned int prev_backlog;
 
 begin:
 	head = &q->new_flows;
@@ -259,6 +261,7 @@ begin:
 
 	prev_drop_count = q->cstats.drop_count;
 	prev_ecn_mark = q->cstats.ecn_mark;
+	prev_backlog = sch->qstats.backlog;
 
 	skb = codel_dequeue(sch, &q->cparams, &flow->cvars, &q->cstats,
 			    dequeue);
@@ -276,12 +279,14 @@ begin:
 	}
 	qdisc_bstats_update(sch, skb);
 	flow->deficit -= qdisc_pkt_len(skb);
-	/* We cant call qdisc_tree_decrease_qlen() if our qlen is 0,
+	/* We cant call qdisc_tree_reduce_backlog() if our qlen is 0,
 	 * or HTB crashes. Defer it for next round.
 	 */
 	if (q->cstats.drop_count && sch->q.qlen) {
-		qdisc_tree_decrease_qlen(sch, q->cstats.drop_count);
+		qdisc_tree_reduce_backlog(sch, q->cstats.drop_count,
+					  q->cstats.drop_len);
 		q->cstats.drop_count = 0;
+		q->cstats.drop_len = 0;
 	}
 	return skb;
 }
@@ -372,11 +377,13 @@ static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt)
 	while (sch->q.qlen > sch->limit) {
 		struct sk_buff *skb = fq_codel_dequeue(sch);
 
+		q->cstats.drop_len += qdisc_pkt_len(skb);
 		kfree_skb(skb);
 		q->cstats.drop_count++;
 	}
-	qdisc_tree_decrease_qlen(sch, q->cstats.drop_count);
+	qdisc_tree_reduce_backlog(sch, q->cstats.drop_count, q->cstats.drop_len);
 	q->cstats.drop_count = 0;
+	q->cstats.drop_len = 0;
 
 	sch_tree_unlock(sch);
 	return 0;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 089f3b6..d783d7c 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -895,9 +895,10 @@ static void
 hfsc_purge_queue(struct Qdisc *sch, struct hfsc_class *cl)
 {
 	unsigned int len = cl->qdisc->q.qlen;
+	unsigned int backlog = cl->qdisc->qstats.backlog;
 
 	qdisc_reset(cl->qdisc);
-	qdisc_tree_decrease_qlen(cl->qdisc, len);
+	qdisc_tree_reduce_backlog(cl->qdisc, len, backlog);
 }
 
 static void
diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c
index 86b04e3..13d6f83 100644
--- a/net/sched/sch_hhf.c
+++ b/net/sched/sch_hhf.c
@@ -382,6 +382,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	struct hhf_sched_data *q = qdisc_priv(sch);
 	enum wdrr_bucket_idx idx;
 	struct wdrr_bucket *bucket;
+	unsigned int prev_backlog;
 
 	idx = hhf_classify(skb, sch);
 
@@ -409,6 +410,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	if (++sch->q.qlen <= sch->limit)
 		return NET_XMIT_SUCCESS;
 
+	prev_backlog = sch->qstats.backlog;
 	q->drop_overlimit++;
 	/* Return Congestion Notification only if we dropped a packet from this
 	 * bucket.
@@ -417,7 +419,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return NET_XMIT_CN;
 
 	/* As we dropped a packet, better let upper stack know this. */
-	qdisc_tree_decrease_qlen(sch, 1);
+	qdisc_tree_reduce_backlog(sch, 1, prev_backlog - sch->qstats.backlog);
 	return NET_XMIT_SUCCESS;
 }
 
@@ -527,7 +529,7 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct hhf_sched_data *q = qdisc_priv(sch);
 	struct nlattr *tb[TCA_HHF_MAX + 1];
-	unsigned int qlen;
+	unsigned int qlen, prev_backlog;
 	int err;
 	u64 non_hh_quantum;
 	u32 new_quantum = q->quantum;
@@ -577,12 +579,14 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt)
 	}
 
 	qlen = sch->q.qlen;
+	prev_backlog = sch->qstats.backlog;
 	while (sch->q.qlen > sch->limit) {
 		struct sk_buff *skb = hhf_dequeue(sch);
 
 		kfree_skb(skb);
 	}
-	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
+	qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen,
+				  prev_backlog - sch->qstats.backlog);
 
 	sch_tree_unlock(sch);
 	return 0;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 0efbcf3..846a7f9 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1265,7 +1265,6 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 {
 	struct htb_sched *q = qdisc_priv(sch);
 	struct htb_class *cl = (struct htb_class *)arg;
-	unsigned int qlen;
 	struct Qdisc *new_q = NULL;
 	int last_child = 0;
 
@@ -1285,9 +1284,11 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 	sch_tree_lock(sch);
 
 	if (!cl->level) {
-		qlen = cl->un.leaf.q->q.qlen;
+		unsigned int qlen = cl->un.leaf.q->q.qlen;
+		unsigned int backlog = cl->un.leaf.q->qstats.backlog;
+
 		qdisc_reset(cl->un.leaf.q);
-		qdisc_tree_decrease_qlen(cl->un.leaf.q, qlen);
+		qdisc_tree_reduce_backlog(cl->un.leaf.q, qlen, backlog);
 	}
 
 	/* delete from hash and active; remainder in destroy_class */
@@ -1421,10 +1422,11 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		sch_tree_lock(sch);
 		if (parent && !parent->level) {
 			unsigned int qlen = parent->un.leaf.q->q.qlen;
+			unsigned int backlog = parent->un.leaf.q->qstats.backlog;
 
 			/* turn parent into inner node */
 			qdisc_reset(parent->un.leaf.q);
-			qdisc_tree_decrease_qlen(parent->un.leaf.q, qlen);
+			qdisc_tree_reduce_backlog(parent->un.leaf.q, qlen, backlog);
 			qdisc_destroy(parent->un.leaf.q);
 			if (parent->prio_activity)
 				htb_deactivate(q, parent);
diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c
index a0103a1..bcdd54b 100644
--- a/net/sched/sch_multiq.c
+++ b/net/sched/sch_multiq.c
@@ -218,7 +218,8 @@ static int multiq_tune(struct Qdisc *sch, struct nlattr *opt)
 		if (q->queues[i] != &noop_qdisc) {
 			struct Qdisc *child = q->queues[i];
 			q->queues[i] = &noop_qdisc;
-			qdisc_tree_decrease_qlen(child, child->q.qlen);
+			qdisc_tree_reduce_backlog(child, child->q.qlen,
+						  child->qstats.backlog);
 			qdisc_destroy(child);
 		}
 	}
@@ -238,8 +239,9 @@ static int multiq_tune(struct Qdisc *sch, struct nlattr *opt)
 				q->queues[i] = child;
 
 				if (old != &noop_qdisc) {
-					qdisc_tree_decrease_qlen(old,
-								 old->q.qlen);
+					qdisc_tree_reduce_backlog(old,
+								  old->q.qlen,
+								  old->qstats.backlog);
 					qdisc_destroy(old);
 				}
 				sch_tree_unlock(sch);
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 0a6ddaf..9640bb3 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -598,7 +598,8 @@ deliver:
 				if (unlikely(err != NET_XMIT_SUCCESS)) {
 					if (net_xmit_drop_count(err)) {
 						qdisc_qstats_drop(sch);
-						qdisc_tree_decrease_qlen(sch, 1);
+						qdisc_tree_reduce_backlog(sch, 1,
+									  qdisc_pkt_len(skb));
 					}
 				}
 				goto tfifo_dequeue;
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index b783a44..71ae3b9 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -183,7 +183,7 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct pie_sched_data *q = qdisc_priv(sch);
 	struct nlattr *tb[TCA_PIE_MAX + 1];
-	unsigned int qlen;
+	unsigned int qlen, dropped = 0;
 	int err;
 
 	if (!opt)
@@ -232,10 +232,11 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt)
 	while (sch->q.qlen > sch->limit) {
 		struct sk_buff *skb = __skb_dequeue(&sch->q);
 
+		dropped += qdisc_pkt_len(skb);
 		qdisc_qstats_backlog_dec(sch, skb);
 		qdisc_drop(skb, sch);
 	}
-	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
+	qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped);
 
 	sch_tree_unlock(sch);
 	return 0;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 1b4aaec..fee1b15 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -191,7 +191,7 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 		struct Qdisc *child = q->queues[i];
 		q->queues[i] = &noop_qdisc;
 		if (child != &noop_qdisc) {
-			qdisc_tree_decrease_qlen(child, child->q.qlen);
+			qdisc_tree_reduce_backlog(child, child->q.qlen, child->qstats.backlog);
 			qdisc_destroy(child);
 		}
 	}
@@ -210,8 +210,9 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 				q->queues[i] = child;
 
 				if (old != &noop_qdisc) {
-					qdisc_tree_decrease_qlen(old,
-								 old->q.qlen);
+					qdisc_tree_reduce_backlog(old,
+								  old->q.qlen,
+								  old->qstats.backlog);
 					qdisc_destroy(old);
 				}
 				sch_tree_unlock(sch);
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index b5c52ca..8d2d8d9 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -220,9 +220,10 @@ static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid)
 static void qfq_purge_queue(struct qfq_class *cl)
 {
 	unsigned int len = cl->qdisc->q.qlen;
+	unsigned int backlog = cl->qdisc->qstats.backlog;
 
 	qdisc_reset(cl->qdisc);
-	qdisc_tree_decrease_qlen(cl->qdisc, len);
+	qdisc_tree_reduce_backlog(cl->qdisc, len, backlog);
 }
 
 static const struct nla_policy qfq_policy[TCA_QFQ_MAX + 1] = {
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index d5abcee..8c0508c 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -210,7 +210,8 @@ static int red_change(struct Qdisc *sch, struct nlattr *opt)
 	q->flags = ctl->flags;
 	q->limit = ctl->limit;
 	if (child) {
-		qdisc_tree_decrease_qlen(q->qdisc, q->qdisc->q.qlen);
+		qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen,
+					  q->qdisc->qstats.backlog);
 		qdisc_destroy(q->qdisc);
 		q->qdisc = child;
 	}
diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c
index 0e74e55..c696116 100644
--- a/net/sched/sch_sfb.c
+++ b/net/sched/sch_sfb.c
@@ -510,7 +510,8 @@ static int sfb_change(struct Qdisc *sch, struct nlattr *opt)
 
 	sch_tree_lock(sch);
 
-	qdisc_tree_decrease_qlen(q->qdisc, q->qdisc->q.qlen);
+	qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen,
+				  q->qdisc->qstats.backlog);
 	qdisc_destroy(q->qdisc);
 	q->qdisc = child;
 
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 3abab53..498f0a2 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -346,7 +346,7 @@ static int
 sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	struct sfq_sched_data *q = qdisc_priv(sch);
-	unsigned int hash;
+	unsigned int hash, dropped;
 	sfq_index x, qlen;
 	struct sfq_slot *slot;
 	int uninitialized_var(ret);
@@ -461,7 +461,7 @@ enqueue:
 		return NET_XMIT_SUCCESS;
 
 	qlen = slot->qlen;
-	sfq_drop(sch);
+	dropped = sfq_drop(sch);
 	/* Return Congestion Notification only if we dropped a packet
 	 * from this flow.
 	 */
@@ -469,7 +469,7 @@ enqueue:
 		return NET_XMIT_CN;
 
 	/* As we dropped a packet, better let upper stack know this */
-	qdisc_tree_decrease_qlen(sch, 1);
+	qdisc_tree_reduce_backlog(sch, 1, dropped);
 	return NET_XMIT_SUCCESS;
 }
 
@@ -537,6 +537,7 @@ static void sfq_rehash(struct Qdisc *sch)
 	struct sfq_slot *slot;
 	struct sk_buff_head list;
 	int dropped = 0;
+	unsigned int drop_len = 0;
 
 	__skb_queue_head_init(&list);
 
@@ -565,6 +566,7 @@ static void sfq_rehash(struct Qdisc *sch)
 			if (x >= SFQ_MAX_FLOWS) {
 drop:
 				qdisc_qstats_backlog_dec(sch, skb);
+				drop_len += qdisc_pkt_len(skb);
 				kfree_skb(skb);
 				dropped++;
 				continue;
@@ -594,7 +596,7 @@ drop:
 		}
 	}
 	sch->q.qlen -= dropped;
-	qdisc_tree_decrease_qlen(sch, dropped);
+	qdisc_tree_reduce_backlog(sch, dropped, drop_len);
 }
 
 static void sfq_perturbation(unsigned long arg)
@@ -618,7 +620,7 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
 	struct sfq_sched_data *q = qdisc_priv(sch);
 	struct tc_sfq_qopt *ctl = nla_data(opt);
 	struct tc_sfq_qopt_v1 *ctl_v1 = NULL;
-	unsigned int qlen;
+	unsigned int qlen, dropped = 0;
 	struct red_parms *p = NULL;
 
 	if (opt->nla_len < nla_attr_size(sizeof(*ctl)))
@@ -667,8 +669,8 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
 
 	qlen = sch->q.qlen;
 	while (sch->q.qlen > q->limit)
-		sfq_drop(sch);
-	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
+		dropped += sfq_drop(sch);
+	qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped);
 
 	del_timer(&q->perturb_timer);
 	if (q->perturb_period) {
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 56a1aef..c2fbde7 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -160,6 +160,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch)
 	struct tbf_sched_data *q = qdisc_priv(sch);
 	struct sk_buff *segs, *nskb;
 	netdev_features_t features = netif_skb_features(skb);
+	unsigned int len = 0, prev_len = qdisc_pkt_len(skb);
 	int ret, nb;
 
 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
@@ -172,6 +173,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch)
 		nskb = segs->next;
 		segs->next = NULL;
 		qdisc_skb_cb(segs)->pkt_len = segs->len;
+		len += segs->len;
 		ret = qdisc_enqueue(segs, q->qdisc);
 		if (ret != NET_XMIT_SUCCESS) {
 			if (net_xmit_drop_count(ret))
@@ -183,7 +185,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch)
 	}
 	sch->q.qlen += nb;
 	if (nb > 1)
-		qdisc_tree_decrease_qlen(sch, 1 - nb);
+		qdisc_tree_reduce_backlog(sch, 1 - nb, prev_len - len);
 	consume_skb(skb);
 	return nb > 0 ? NET_XMIT_SUCCESS : NET_XMIT_DROP;
 }
@@ -399,7 +401,8 @@ static int tbf_change(struct Qdisc *sch, struct nlattr *opt)
 
 	sch_tree_lock(sch);
 	if (child) {
-		qdisc_tree_decrease_qlen(q->qdisc, q->qdisc->q.qlen);
+		qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen,
+					  q->qdisc->qstats.backlog);
 		qdisc_destroy(q->qdisc);
 		q->qdisc = child;
 	}
-- 
cgit v0.10.2


From 431e3a8e36a05a37126f34b41aa3a5a6456af04e Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Thu, 25 Feb 2016 14:55:02 -0800
Subject: sch_htb: update backlog as well

We saw qlen!=0 but backlog==0 on our production machine:

qdisc htb 1: dev eth0 root refcnt 2 r2q 10 default 1 direct_packets_stat 0 ver 3.17
 Sent 172680457356 bytes 222469449 pkt (dropped 0, overlimits 123575834 requeues 0)
 backlog 0b 72p requeues 0

The problem is we only count qlen for HTB qdisc but not backlog.
We need to update backlog too when we update qlen, so that we
can at least know the average packet length.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 846a7f9..87b02ed3 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -600,6 +600,7 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		htb_activate(q, cl);
 	}
 
+	qdisc_qstats_backlog_inc(sch, skb);
 	sch->q.qlen++;
 	return NET_XMIT_SUCCESS;
 }
@@ -889,6 +890,7 @@ static struct sk_buff *htb_dequeue(struct Qdisc *sch)
 ok:
 		qdisc_bstats_update(sch, skb);
 		qdisc_unthrottled(sch);
+		qdisc_qstats_backlog_dec(sch, skb);
 		sch->q.qlen--;
 		return skb;
 	}
@@ -955,6 +957,7 @@ static unsigned int htb_drop(struct Qdisc *sch)
 			unsigned int len;
 			if (cl->un.leaf.q->ops->drop &&
 			    (len = cl->un.leaf.q->ops->drop(cl->un.leaf.q))) {
+				sch->qstats.backlog -= len;
 				sch->q.qlen--;
 				if (!cl->un.leaf.q->q.qlen)
 					htb_deactivate(q, cl);
@@ -984,12 +987,12 @@ static void htb_reset(struct Qdisc *sch)
 			}
 			cl->prio_activity = 0;
 			cl->cmode = HTB_CAN_SEND;
-
 		}
 	}
 	qdisc_watchdog_cancel(&q->watchdog);
 	__skb_queue_purge(&q->direct_queue);
 	sch->q.qlen = 0;
+	sch->qstats.backlog = 0;
 	memset(q->hlevel, 0, sizeof(q->hlevel));
 	memset(q->row_mask, 0, sizeof(q->row_mask));
 	for (i = 0; i < TC_HTB_NUMPRIO; i++)
-- 
cgit v0.10.2


From bdf17661f63a79c3cb4209b970b1cc39e34f7543 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Thu, 25 Feb 2016 14:55:03 -0800
Subject: sch_dsmark: update backlog as well

Similarly, we need to update backlog too when we update qlen.

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index cfddb1c..d0dff0c 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -258,6 +258,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return err;
 	}
 
+	qdisc_qstats_backlog_inc(sch, skb);
 	sch->q.qlen++;
 
 	return NET_XMIT_SUCCESS;
@@ -280,6 +281,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
 		return NULL;
 
 	qdisc_bstats_update(sch, skb);
+	qdisc_qstats_backlog_dec(sch, skb);
 	sch->q.qlen--;
 
 	index = skb->tc_index & (p->indices - 1);
@@ -395,6 +397,7 @@ static void dsmark_reset(struct Qdisc *sch)
 
 	pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p);
 	qdisc_reset(p->q);
+	sch->qstats.backlog = 0;
 	sch->q.qlen = 0;
 }
 
-- 
cgit v0.10.2


From 49eb93892d8dfcf5dc5088e640f486d08572d8bf Mon Sep 17 00:00:00 2001
From: "sixiao@microsoft.com" <sixiao@microsoft.com>
Date: Thu, 25 Feb 2016 15:24:08 -0800
Subject: hv_netvsc: add ethtool support for set and get of settings

This patch allows the user to set and retrieve speed and duplex of the
hv_netvsc device via ethtool.

Example:
$ ethtool eth0
Settings for eth0:
...
    Speed: Unknown!
    Duplex: Unknown! (255)
...
$ ethtool -s eth0 speed 1000 duplex full
$ ethtool eth0
Settings for eth0:
...
    Speed: 1000Mb/s
    Duplex: Full
...

This is based on patches by Roopa Prabhu and Nikolay Aleksandrov.

Signed-off-by: Simon Xiao <sixiao@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index fcb92c0..b4c6878 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -658,6 +658,10 @@ struct net_device_context {
 
 	struct netvsc_stats __percpu *tx_stats;
 	struct netvsc_stats __percpu *rx_stats;
+
+	/* Ethtool settings */
+	u8 duplex;
+	u32 speed;
 };
 
 /* Per netvsc device */
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 2b6595e..0860849 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -799,6 +799,58 @@ static int netvsc_set_channels(struct net_device *net,
 	goto do_set;
 }
 
+static bool netvsc_validate_ethtool_ss_cmd(const struct ethtool_cmd *cmd)
+{
+	struct ethtool_cmd diff1 = *cmd;
+	struct ethtool_cmd diff2 = {};
+
+	ethtool_cmd_speed_set(&diff1, 0);
+	diff1.duplex = 0;
+	/* advertising and cmd are usually set */
+	diff1.advertising = 0;
+	diff1.cmd = 0;
+	/* We set port to PORT_OTHER */
+	diff2.port = PORT_OTHER;
+
+	return !memcmp(&diff1, &diff2, sizeof(diff1));
+}
+
+static void netvsc_init_settings(struct net_device *dev)
+{
+	struct net_device_context *ndc = netdev_priv(dev);
+
+	ndc->speed = SPEED_UNKNOWN;
+	ndc->duplex = DUPLEX_UNKNOWN;
+}
+
+static int netvsc_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct net_device_context *ndc = netdev_priv(dev);
+
+	ethtool_cmd_speed_set(cmd, ndc->speed);
+	cmd->duplex = ndc->duplex;
+	cmd->port = PORT_OTHER;
+
+	return 0;
+}
+
+static int netvsc_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct net_device_context *ndc = netdev_priv(dev);
+	u32 speed;
+
+	speed = ethtool_cmd_speed(cmd);
+	if (!ethtool_validate_speed(speed) ||
+	    !ethtool_validate_duplex(cmd->duplex) ||
+	    !netvsc_validate_ethtool_ss_cmd(cmd))
+		return -EINVAL;
+
+	ndc->speed = speed;
+	ndc->duplex = cmd->duplex;
+
+	return 0;
+}
+
 static int netvsc_change_mtu(struct net_device *ndev, int mtu)
 {
 	struct net_device_context *ndevctx = netdev_priv(ndev);
@@ -923,6 +975,8 @@ static const struct ethtool_ops ethtool_ops = {
 	.get_channels   = netvsc_get_channels,
 	.set_channels   = netvsc_set_channels,
 	.get_ts_info	= ethtool_op_get_ts_info,
+	.get_settings	= netvsc_get_settings,
+	.set_settings	= netvsc_set_settings,
 };
 
 static const struct net_device_ops device_ops = {
@@ -1115,6 +1169,8 @@ static int netvsc_probe(struct hv_device *dev,
 	netif_set_real_num_tx_queues(net, nvdev->num_chn);
 	netif_set_real_num_rx_queues(net, nvdev->num_chn);
 
+	netvsc_init_settings(net);
+
 	ret = register_netdev(net);
 	if (ret != 0) {
 		pr_err("Unable to register netdev.\n");
-- 
cgit v0.10.2


From 87177ba6e47e4e3268b999843399ee1566f682ac Mon Sep 17 00:00:00 2001
From: "Woojung.Huh@microchip.com" <Woojung.Huh@microchip.com>
Date: Thu, 25 Feb 2016 23:33:07 +0000
Subject: lan78xx: replace devid to chipid & chiprev

Replace devid to chipid & chiprev for easy access.

Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 1c299b8..4ec25e8 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -278,7 +278,8 @@ struct lan78xx_net {
 	int			link_on;
 	u8			mdix_ctrl;
 
-	u32			devid;
+	u32			chipid;
+	u32			chiprev;
 	struct mii_bus		*mdiobus;
 };
 
@@ -471,7 +472,7 @@ static int lan78xx_read_raw_eeprom(struct lan78xx_net *dev, u32 offset,
 	 */
 	ret = lan78xx_read_reg(dev, HW_CFG, &val);
 	saved = val;
-	if ((dev->devid & ID_REV_CHIP_ID_MASK_) == 0x78000000) {
+	if (dev->chipid == ID_REV_CHIP_ID_7800_) {
 		val &= ~(HW_CFG_LED1_EN_ | HW_CFG_LED0_EN_);
 		ret = lan78xx_write_reg(dev, HW_CFG, val);
 	}
@@ -505,7 +506,7 @@ static int lan78xx_read_raw_eeprom(struct lan78xx_net *dev, u32 offset,
 
 	retval = 0;
 exit:
-	if ((dev->devid & ID_REV_CHIP_ID_MASK_) == 0x78000000)
+	if (dev->chipid == ID_REV_CHIP_ID_7800_)
 		ret = lan78xx_write_reg(dev, HW_CFG, saved);
 
 	return retval;
@@ -539,7 +540,7 @@ static int lan78xx_write_raw_eeprom(struct lan78xx_net *dev, u32 offset,
 	 */
 	ret = lan78xx_read_reg(dev, HW_CFG, &val);
 	saved = val;
-	if ((dev->devid & ID_REV_CHIP_ID_MASK_) == 0x78000000) {
+	if (dev->chipid == ID_REV_CHIP_ID_7800_) {
 		val &= ~(HW_CFG_LED1_EN_ | HW_CFG_LED0_EN_);
 		ret = lan78xx_write_reg(dev, HW_CFG, val);
 	}
@@ -587,7 +588,7 @@ static int lan78xx_write_raw_eeprom(struct lan78xx_net *dev, u32 offset,
 
 	retval = 0;
 exit:
-	if ((dev->devid & ID_REV_CHIP_ID_MASK_) == 0x78000000)
+	if (dev->chipid == ID_REV_CHIP_ID_7800_)
 		ret = lan78xx_write_reg(dev, HW_CFG, saved);
 
 	return retval;
@@ -1555,9 +1556,9 @@ static int lan78xx_mdio_init(struct lan78xx_net *dev)
 	snprintf(dev->mdiobus->id, MII_BUS_ID_SIZE, "usb-%03d:%03d",
 		 dev->udev->bus->busnum, dev->udev->devnum);
 
-	switch (dev->devid & ID_REV_CHIP_ID_MASK_) {
-	case 0x78000000:
-	case 0x78500000:
+	switch (dev->chipid) {
+	case ID_REV_CHIP_ID_7800_:
+	case ID_REV_CHIP_ID_7850_:
 		/* set to internal PHY id */
 		dev->mdiobus->phy_mask = ~(1 << 1);
 		break;
@@ -1918,7 +1919,8 @@ static int lan78xx_reset(struct lan78xx_net *dev)
 
 	/* save DEVID for later usage */
 	ret = lan78xx_read_reg(dev, ID_REV, &buf);
-	dev->devid = buf;
+	dev->chipid = (buf & ID_REV_CHIP_ID_MASK_) >> 16;
+	dev->chiprev = buf & ID_REV_CHIP_REV_MASK_;
 
 	/* Respond to the IN token with a NAK */
 	ret = lan78xx_read_reg(dev, USB_CFG0, &buf);
diff --git a/drivers/net/usb/lan78xx.h b/drivers/net/usb/lan78xx.h
index a93fb65..4092790 100644
--- a/drivers/net/usb/lan78xx.h
+++ b/drivers/net/usb/lan78xx.h
@@ -107,6 +107,7 @@
 #define ID_REV_CHIP_ID_MASK_		(0xFFFF0000)
 #define ID_REV_CHIP_REV_MASK_		(0x0000FFFF)
 #define ID_REV_CHIP_ID_7800_		(0x7800)
+#define ID_REV_CHIP_ID_7850_		(0x7850)
 
 #define FPGA_REV			(0x04)
 #define FPGA_REV_MINOR_MASK_		(0x0000FF00)
-- 
cgit v0.10.2


From e270b2dbd83caf091348f609235d3e28f1b754af Mon Sep 17 00:00:00 2001
From: "Woojung.Huh@microchip.com" <Woojung.Huh@microchip.com>
Date: Thu, 25 Feb 2016 23:33:09 +0000
Subject: lan78xx: remove unnecessary code

It is not required after commit cd772de358d6
("phy: keep pause flags in phy driver features")

Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 4ec25e8..d100421 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -1623,12 +1623,7 @@ static int lan78xx_phy_init(struct lan78xx_net *dev)
 
 	/* MAC doesn't support 1000T Half */
 	phydev->supported &= ~SUPPORTED_1000baseT_Half;
-	phydev->supported |= (SUPPORTED_10baseT_Half |
-			      SUPPORTED_10baseT_Full |
-			      SUPPORTED_100baseT_Half |
-			      SUPPORTED_100baseT_Full |
-			      SUPPORTED_1000baseT_Full |
-			      SUPPORTED_Pause | SUPPORTED_Asym_Pause);
+
 	genphy_config_aneg(phydev);
 
 	phy_start(phydev);
-- 
cgit v0.10.2


From 349e0c5e2237744d2acc1bdb52526a69c5fdcd52 Mon Sep 17 00:00:00 2001
From: "Woojung.Huh@microchip.com" <Woojung.Huh@microchip.com>
Date: Thu, 25 Feb 2016 23:33:14 +0000
Subject: lan78xx: add ethtool set & get pause functions

Add ethtool operations of set_pauseram and get_pauseparm.

Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index d100421..705c180 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -36,7 +36,7 @@
 #define DRIVER_AUTHOR	"WOOJUNG HUH <woojung.huh@microchip.com>"
 #define DRIVER_DESC	"LAN78XX USB 3.0 Gigabit Ethernet Devices"
 #define DRIVER_NAME	"lan78xx"
-#define DRIVER_VERSION	"1.0.2"
+#define DRIVER_VERSION	"1.0.3"
 
 #define TX_TIMEOUT_JIFFIES		(5 * HZ)
 #define THROTTLE_JIFFIES		(HZ / 8)
@@ -281,6 +281,9 @@ struct lan78xx_net {
 	u32			chipid;
 	u32			chiprev;
 	struct mii_bus		*mdiobus;
+
+	int			fc_autoneg;
+	u8			fc_request_control;
 };
 
 /* use ethtool to change the level for any given device */
@@ -902,11 +905,15 @@ static int lan78xx_update_flowcontrol(struct lan78xx_net *dev, u8 duplex,
 {
 	u32 flow = 0, fct_flow = 0;
 	int ret;
+	u8 cap;
 
-	u8 cap = mii_resolve_flowctrl_fdx(lcladv, rmtadv);
+	if (dev->fc_autoneg)
+		cap = mii_resolve_flowctrl_fdx(lcladv, rmtadv);
+	else
+		cap = dev->fc_request_control;
 
 	if (cap & FLOW_CTRL_TX)
-		flow = (FLOW_CR_TX_FCEN_ | 0xFFFF);
+		flow |= (FLOW_CR_TX_FCEN_ | 0xFFFF);
 
 	if (cap & FLOW_CTRL_RX)
 		flow |= FLOW_CR_RX_FCEN_;
@@ -1386,6 +1393,62 @@ static int lan78xx_set_settings(struct net_device *net, struct ethtool_cmd *cmd)
 	return ret;
 }
 
+static void lan78xx_get_pause(struct net_device *net,
+			      struct ethtool_pauseparam *pause)
+{
+	struct lan78xx_net *dev = netdev_priv(net);
+	struct phy_device *phydev = net->phydev;
+	struct ethtool_cmd ecmd = { .cmd = ETHTOOL_GSET };
+
+	phy_ethtool_gset(phydev, &ecmd);
+
+	pause->autoneg = dev->fc_autoneg;
+
+	if (dev->fc_request_control & FLOW_CTRL_TX)
+		pause->tx_pause = 1;
+
+	if (dev->fc_request_control & FLOW_CTRL_RX)
+		pause->rx_pause = 1;
+}
+
+static int lan78xx_set_pause(struct net_device *net,
+			     struct ethtool_pauseparam *pause)
+{
+	struct lan78xx_net *dev = netdev_priv(net);
+	struct phy_device *phydev = net->phydev;
+	struct ethtool_cmd ecmd = { .cmd = ETHTOOL_GSET };
+	int ret;
+
+	phy_ethtool_gset(phydev, &ecmd);
+
+	if (pause->autoneg && !ecmd.autoneg) {
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	dev->fc_request_control = 0;
+	if (pause->rx_pause)
+		dev->fc_request_control |= FLOW_CTRL_RX;
+
+	if (pause->tx_pause)
+		dev->fc_request_control |= FLOW_CTRL_TX;
+
+	if (ecmd.autoneg) {
+		u32 mii_adv;
+
+		ecmd.advertising &= ~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
+		mii_adv = (u32)mii_advertise_flowctrl(dev->fc_request_control);
+		ecmd.advertising |= mii_adv_to_ethtool_adv_t(mii_adv);
+		phy_ethtool_sset(phydev, &ecmd);
+	}
+
+	dev->fc_autoneg = pause->autoneg;
+
+	ret = 0;
+exit:
+	return ret;
+}
+
 static const struct ethtool_ops lan78xx_ethtool_ops = {
 	.get_link	= lan78xx_get_link,
 	.nway_reset	= lan78xx_nway_reset,
@@ -1404,6 +1467,8 @@ static const struct ethtool_ops lan78xx_ethtool_ops = {
 	.set_wol	= lan78xx_set_wol,
 	.get_eee	= lan78xx_get_eee,
 	.set_eee	= lan78xx_set_eee,
+	.get_pauseparam	= lan78xx_get_pause,
+	.set_pauseparam	= lan78xx_set_pause,
 };
 
 static int lan78xx_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd)
@@ -1591,6 +1656,7 @@ static void lan78xx_link_status_change(struct net_device *net)
 static int lan78xx_phy_init(struct lan78xx_net *dev)
 {
 	int ret;
+	u32 mii_adv;
 	struct phy_device *phydev = dev->net->phydev;
 
 	phydev = phy_find_first(dev->mdiobus);
@@ -1624,8 +1690,16 @@ static int lan78xx_phy_init(struct lan78xx_net *dev)
 	/* MAC doesn't support 1000T Half */
 	phydev->supported &= ~SUPPORTED_1000baseT_Half;
 
+	/* support both flow controls */
+	dev->fc_request_control = (FLOW_CTRL_RX | FLOW_CTRL_TX);
+	phydev->advertising &= ~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
+	mii_adv = (u32)mii_advertise_flowctrl(dev->fc_request_control);
+	phydev->advertising |= mii_adv_to_ethtool_adv_t(mii_adv);
+
 	genphy_config_aneg(phydev);
 
+	dev->fc_autoneg = phydev->autoneg;
+
 	phy_start(phydev);
 
 	netif_dbg(dev, ifup, dev->net, "phy initialised successfully");
-- 
cgit v0.10.2


From 146498ea06141351ebc9360b9025a5d475d7e02f Mon Sep 17 00:00:00 2001
From: "Woojung.Huh@microchip.com" <Woojung.Huh@microchip.com>
Date: Thu, 25 Feb 2016 23:33:16 +0000
Subject: MAINTAINERS: Add LAN78XX entry

Add maintainers for Microchip LAN78XX.
UNGLinuxDriver@microchip.com is alias email which goes to current
developers work for Microchip Network related products.

Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/MAINTAINERS b/MAINTAINERS
index 27393cf..12b764f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11332,6 +11332,13 @@ S:	Maintained
 F:	drivers/usb/host/isp116x*
 F:	include/linux/usb/isp116x.h
 
+USB LAN78XX ETHERNET DRIVER
+M:	Woojung Huh <woojung.huh@microchip.com>
+M:	Microchip Linux Driver Support <UNGLinuxDriver@microchip.com>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/net/usb/lan78xx.*
+
 USB MASS STORAGE DRIVER
 M:	Matthew Dharm <mdharm-usb@one-eyed-alien.net>
 L:	linux-usb@vger.kernel.org
-- 
cgit v0.10.2


From 3874d6a8b61966a77aa743b4160ba96bf3081ce5 Mon Sep 17 00:00:00 2001
From: Jeffrey Huang <huangjw@broadcom.com>
Date: Fri, 26 Feb 2016 03:59:59 -0500
Subject: bnxt_en: Improve bnxt_vf_update_mac().

Allow the VF to setup its own MAC address if the PF has not administratively
set it for the VF.  To do that, we should always store the MAC address
from the firmware.  There are 2 cases:

1. The MAC address is valid.  This MAC address is assigned by the PF and
it needs to override the current VF MAC address.

2. The MAC address is zero.  The VF will use a random MAC address by default.
By storing this 0 MAC address in the VF structure, it will allow the VF
user to change the MAC address later using ndo_set_mac_address() when
it sees that the stored MAC address is 0.

v2: Expanded descriptions and added more comments.

Signed-off-by: Jeffrey Huang <huangjw@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index c1cc83d..0b4ca35 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -809,13 +809,19 @@ void bnxt_update_vf_mac(struct bnxt *bp)
 	if (_hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT))
 		goto update_vf_mac_exit;
 
-	if (!is_valid_ether_addr(resp->perm_mac_address))
-		goto update_vf_mac_exit;
-
+	/* Store MAC address from the firmware.  There are 2 cases:
+	 * 1. MAC address is valid.  It is assigned from the PF and we
+	 *    need to override the current VF MAC address with it.
+	 * 2. MAC address is zero.  The VF will use a random MAC address by
+	 *    default but the stored zero MAC will allow the VF user to change
+	 *    the random MAC address using ndo_set_mac_address() if he wants.
+	 */
 	if (!ether_addr_equal(resp->perm_mac_address, bp->vf.mac_addr))
 		memcpy(bp->vf.mac_addr, resp->perm_mac_address, ETH_ALEN);
-	/* overwrite netdev dev_adr with admin VF MAC */
-	memcpy(bp->dev->dev_addr, bp->vf.mac_addr, ETH_ALEN);
+
+	/* overwrite netdev dev_addr with admin VF MAC */
+	if (is_valid_ether_addr(bp->vf.mac_addr))
+		memcpy(bp->dev->dev_addr, bp->vf.mac_addr, ETH_ALEN);
 update_vf_mac_exit:
 	mutex_unlock(&bp->hwrm_cmd_lock);
 }
-- 
cgit v0.10.2


From 19241368443ff976b1924019d29eef8e972158e7 Mon Sep 17 00:00:00 2001
From: Jeffrey Huang <huangjw@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:00 -0500
Subject: bnxt_en: Send PF driver unload notification to all VFs.

During remove_one() when SRIOV is enabled, the PF driver
should broadcast PF driver unload notification to all
VFs that are attached to VMs. Upon receiving the PF
driver unload notification, the VF driver should print
a warning message to message log.  Certain operations on the
VF may not succeed after the PF has unloaded.

Signed-off-by: Jeffrey Huang <huangjw@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index ff1507f..80c4412 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -1239,13 +1239,17 @@ static int bnxt_async_event_process(struct bnxt *bp,
 	switch (event_id) {
 	case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE:
 		set_bit(BNXT_LINK_CHNG_SP_EVENT, &bp->sp_event);
-		schedule_work(&bp->sp_task);
+		break;
+	case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD:
+		set_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event);
 		break;
 	default:
 		netdev_err(bp->dev, "unhandled ASYNC event (id 0x%x)\n",
 			   event_id);
-		break;
+		goto async_event_process_exit;
 	}
+	schedule_work(&bp->sp_task);
+async_event_process_exit:
 	return 0;
 }
 
@@ -5559,6 +5563,8 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp)
 			}
 		}
 	}
+	if (test_and_clear_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event))
+		netdev_info(bp->dev, "Receive PF driver unload event!");
 }
 
 #else
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 2be51b3..cc798d5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -986,6 +986,7 @@ struct bnxt {
 #define BNXT_VXLAN_DEL_PORT_SP_EVENT	5
 #define BNXT_RESET_TASK_SP_EVENT	6
 #define BNXT_RST_RING_SP_EVENT		7
+#define BNXT_HWRM_PF_UNLOAD_SP_EVENT	8
 
 	struct bnxt_pf_info	pf;
 #ifdef CONFIG_BNXT_SRIOV
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index 0b4ca35..9f6b4cc 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -522,6 +522,46 @@ err_out1:
 	return rc;
 }
 
+static int bnxt_hwrm_fwd_async_event_cmpl(struct bnxt *bp,
+					  struct bnxt_vf_info *vf,
+					  u16 event_id)
+{
+	int rc = 0;
+	struct hwrm_fwd_async_event_cmpl_input req = {0};
+	struct hwrm_fwd_async_event_cmpl_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_async_event_cmpl *async_cmpl;
+
+	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FWD_ASYNC_EVENT_CMPL, -1, -1);
+	if (vf)
+		req.encap_async_event_target_id = cpu_to_le16(vf->fw_fid);
+	else
+		/* broadcast this async event to all VFs */
+		req.encap_async_event_target_id = cpu_to_le16(0xffff);
+	async_cmpl = (struct hwrm_async_event_cmpl *)req.encap_async_event_cmpl;
+	async_cmpl->type =
+		cpu_to_le16(HWRM_ASYNC_EVENT_CMPL_TYPE_HWRM_ASYNC_EVENT);
+	async_cmpl->event_id = cpu_to_le16(event_id);
+
+	mutex_lock(&bp->hwrm_cmd_lock);
+	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+
+	if (rc) {
+		netdev_err(bp->dev, "hwrm_fwd_async_event_cmpl failed. rc:%d\n",
+			   rc);
+		goto fwd_async_event_cmpl_exit;
+	}
+
+	if (resp->error_code) {
+		netdev_err(bp->dev, "hwrm_fwd_async_event_cmpl error %d\n",
+			   resp->error_code);
+		rc = -1;
+	}
+
+fwd_async_event_cmpl_exit:
+	mutex_unlock(&bp->hwrm_cmd_lock);
+	return rc;
+}
+
 void bnxt_sriov_disable(struct bnxt *bp)
 {
 	u16 num_vfs = pci_num_vf(bp->pdev);
@@ -530,6 +570,9 @@ void bnxt_sriov_disable(struct bnxt *bp)
 		return;
 
 	if (pci_vfs_assigned(bp->pdev)) {
+		bnxt_hwrm_fwd_async_event_cmpl(
+			bp, NULL,
+			HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD);
 		netdev_warn(bp->dev, "Unable to free %d VFs because some are assigned to VMs.\n",
 			    num_vfs);
 	} else {
-- 
cgit v0.10.2


From dfb5b894f87cb78168e04283e8d15626dc3e6d5a Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:01 -0500
Subject: bnxt_en: Store irq coalescing timer values in micro seconds.

Don't convert these to internal hardware tick values before storing
them.  This avoids the confusion of ethtool -c returning slightly
different values than the ones set using ethtool -C when we convert
hardware tick values back to micro seconds.  Add better comments for
the hardware settings.

Also, rename the current set of coalescing fields with rx_ prefix.
The next patch will add support of tx coalescing values.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 80c4412..a7103b5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3532,20 +3532,25 @@ int bnxt_hwrm_set_coal(struct bnxt *bp)
 	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS,
 			       -1, -1);
 
-	/* Each rx completion (2 records) should be DMAed immediately */
-	max_buf = min_t(u16, bp->coal_bufs / 4, 2);
+	/* Each rx completion (2 records) should be DMAed immediately.
+	 * DMA 1/4 of the completion buffers at a time.
+	 */
+	max_buf = min_t(u16, bp->rx_coal_bufs / 4, 2);
 	/* max_buf must not be zero */
 	max_buf = clamp_t(u16, max_buf, 1, 63);
-	max_buf_irq = clamp_t(u16, bp->coal_bufs_irq, 1, 63);
-	buf_tmr = max_t(u16, bp->coal_ticks / 4, 1);
-	buf_tmr_irq = max_t(u16, bp->coal_ticks_irq, 1);
+	max_buf_irq = clamp_t(u16, bp->rx_coal_bufs_irq, 1, 63);
+	buf_tmr = BNXT_USEC_TO_COAL_TIMER(bp->rx_coal_ticks);
+	/* buf timer set to 1/4 of interrupt timer */
+	buf_tmr = max_t(u16, buf_tmr / 4, 1);
+	buf_tmr_irq = BNXT_USEC_TO_COAL_TIMER(bp->rx_coal_ticks_irq);
+	buf_tmr_irq = max_t(u16, buf_tmr_irq, 1);
 
 	flags = RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
 
 	/* RING_IDLE generates more IRQs for lower latency.  Enable it only
 	 * if coal_ticks is less than 25 us.
 	 */
-	if (BNXT_COAL_TIMER_TO_USEC(bp->coal_ticks) < 25)
+	if (bp->rx_coal_ticks < 25)
 		flags |= RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_RING_IDLE;
 
 	req.flags = cpu_to_le16(flags);
@@ -3553,9 +3558,10 @@ int bnxt_hwrm_set_coal(struct bnxt *bp)
 	req.num_cmpl_dma_aggr_during_int = cpu_to_le16(max_buf_irq);
 	req.cmpl_aggr_dma_tmr = cpu_to_le16(buf_tmr);
 	req.cmpl_aggr_dma_tmr_during_int = cpu_to_le16(buf_tmr_irq);
-	req.int_lat_tmr_min = cpu_to_le16(buf_tmr);
-	req.int_lat_tmr_max = cpu_to_le16(bp->coal_ticks);
-	req.num_cmpl_aggr_int = cpu_to_le16(bp->coal_bufs);
+	/* Minimum time between 2 interrupts set to buf_tmr x 2 */
+	req.int_lat_tmr_min = cpu_to_le16(buf_tmr * 2);
+	req.int_lat_tmr_max = cpu_to_le16(buf_tmr * 4);
+	req.num_cmpl_aggr_int = cpu_to_le16(max_buf * 4);
 
 	mutex_lock(&bp->hwrm_cmd_lock);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
@@ -5295,10 +5301,11 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 	bp->rx_ring_size = BNXT_DEFAULT_RX_RING_SIZE;
 	bp->tx_ring_size = BNXT_DEFAULT_TX_RING_SIZE;
 
-	bp->coal_ticks = BNXT_USEC_TO_COAL_TIMER(4);
-	bp->coal_bufs = 20;
-	bp->coal_ticks_irq = BNXT_USEC_TO_COAL_TIMER(1);
-	bp->coal_bufs_irq = 2;
+	/* tick values in micro seconds */
+	bp->rx_coal_ticks = 4;
+	bp->rx_coal_bufs = 20;
+	bp->rx_coal_ticks_irq = 1;
+	bp->rx_coal_bufs_irq = 2;
 
 	init_timer(&bp->timer);
 	bp->timer.data = (unsigned long)bp;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index cc798d5..6913307 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -968,13 +968,12 @@ struct bnxt {
 	__le16			vxlan_fw_dst_port_id;
 	u8			nge_port_cnt;
 	__le16			nge_fw_dst_port_id;
-	u16			coal_ticks;
-	u16			coal_ticks_irq;
-	u16			coal_bufs;
-	u16			coal_bufs_irq;
+	u16			rx_coal_ticks;
+	u16			rx_coal_ticks_irq;
+	u16			rx_coal_bufs;
+	u16			rx_coal_bufs_irq;
 
 #define BNXT_USEC_TO_COAL_TIMER(x)	((x) * 25 / 2)
-#define BNXT_COAL_TIMER_TO_USEC(x) ((x) * 2 / 25)
 
 	struct work_struct	sp_task;
 	unsigned long		sp_event;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 3238817..d18a977 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -41,12 +41,11 @@ static int bnxt_get_coalesce(struct net_device *dev,
 
 	memset(coal, 0, sizeof(*coal));
 
-	coal->rx_coalesce_usecs =
-		max_t(u16, BNXT_COAL_TIMER_TO_USEC(bp->coal_ticks), 1);
-	coal->rx_max_coalesced_frames = bp->coal_bufs / 2;
-	coal->rx_coalesce_usecs_irq =
-		max_t(u16, BNXT_COAL_TIMER_TO_USEC(bp->coal_ticks_irq), 1);
-	coal->rx_max_coalesced_frames_irq = bp->coal_bufs_irq / 2;
+	coal->rx_coalesce_usecs = bp->rx_coal_ticks;
+	/* 2 completion records per rx packet */
+	coal->rx_max_coalesced_frames = bp->rx_coal_bufs / 2;
+	coal->rx_coalesce_usecs_irq = bp->rx_coal_ticks_irq;
+	coal->rx_max_coalesced_frames_irq = bp->rx_coal_bufs_irq / 2;
 
 	return 0;
 }
@@ -57,11 +56,11 @@ static int bnxt_set_coalesce(struct net_device *dev,
 	struct bnxt *bp = netdev_priv(dev);
 	int rc = 0;
 
-	bp->coal_ticks = BNXT_USEC_TO_COAL_TIMER(coal->rx_coalesce_usecs);
-	bp->coal_bufs = coal->rx_max_coalesced_frames * 2;
-	bp->coal_ticks_irq =
-		BNXT_USEC_TO_COAL_TIMER(coal->rx_coalesce_usecs_irq);
-	bp->coal_bufs_irq = coal->rx_max_coalesced_frames_irq * 2;
+	bp->rx_coal_ticks = coal->rx_coalesce_usecs;
+	/* 2 completion records per rx packet */
+	bp->rx_coal_bufs = coal->rx_max_coalesced_frames * 2;
+	bp->rx_coal_ticks_irq = coal->rx_coalesce_usecs_irq;
+	bp->rx_coal_bufs_irq = coal->rx_max_coalesced_frames_irq * 2;
 
 	if (netif_running(dev))
 		rc = bnxt_hwrm_set_coal(bp);
-- 
cgit v0.10.2


From bb053f52a54d66a6057c2220458349f7d39ce0d2 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:02 -0500
Subject: bnxt_en: Refactor bnxt_hwrm_set_coal().

Add a function to set all the coalescing parameters.  The function can
be used later to set both rx and tx coalescing parameters.

v2: Fixed function parameters formatting requested by DaveM.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index a7103b5..a391c9c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3521,6 +3521,21 @@ static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
 	}
 }
 
+static void bnxt_hwrm_set_coal_params(struct bnxt *bp, u32 max_bufs,
+	u32 buf_tmrs, u16 flags,
+	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input *req)
+{
+	req->flags = cpu_to_le16(flags);
+	req->num_cmpl_dma_aggr = cpu_to_le16((u16)max_bufs);
+	req->num_cmpl_dma_aggr_during_int = cpu_to_le16(max_bufs >> 16);
+	req->cmpl_aggr_dma_tmr = cpu_to_le16((u16)buf_tmrs);
+	req->cmpl_aggr_dma_tmr_during_int = cpu_to_le16(buf_tmrs >> 16);
+	/* Minimum time between 2 interrupts set to buf_tmr x 2 */
+	req->int_lat_tmr_min = cpu_to_le16((u16)buf_tmrs * 2);
+	req->int_lat_tmr_max = cpu_to_le16((u16)buf_tmrs * 4);
+	req->num_cmpl_aggr_int = cpu_to_le16((u16)max_bufs * 4);
+}
+
 int bnxt_hwrm_set_coal(struct bnxt *bp)
 {
 	int i, rc = 0;
@@ -3553,15 +3568,8 @@ int bnxt_hwrm_set_coal(struct bnxt *bp)
 	if (bp->rx_coal_ticks < 25)
 		flags |= RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_RING_IDLE;
 
-	req.flags = cpu_to_le16(flags);
-	req.num_cmpl_dma_aggr = cpu_to_le16(max_buf);
-	req.num_cmpl_dma_aggr_during_int = cpu_to_le16(max_buf_irq);
-	req.cmpl_aggr_dma_tmr = cpu_to_le16(buf_tmr);
-	req.cmpl_aggr_dma_tmr_during_int = cpu_to_le16(buf_tmr_irq);
-	/* Minimum time between 2 interrupts set to buf_tmr x 2 */
-	req.int_lat_tmr_min = cpu_to_le16(buf_tmr * 2);
-	req.int_lat_tmr_max = cpu_to_le16(buf_tmr * 4);
-	req.num_cmpl_aggr_int = cpu_to_le16(max_buf * 4);
+	bnxt_hwrm_set_coal_params(bp, max_buf_irq << 16 | max_buf,
+				  buf_tmr_irq << 16 | buf_tmr, flags, &req);
 
 	mutex_lock(&bp->hwrm_cmd_lock);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
-- 
cgit v0.10.2


From dfc9c94a83909f4be80e5d0c67e79793830aa312 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:03 -0500
Subject: bnxt_en: Add coalescing support for tx rings.

When tx and rx rings don't share the same completion ring, tx coalescing
parameters can be set differently from the rx coalescing parameters.
Otherwise, use rx coalescing parameters on shared completion rings.

Adjust rx coalescing default values to lower interrupt rate.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index a391c9c..198c69d4 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3539,13 +3539,16 @@ static void bnxt_hwrm_set_coal_params(struct bnxt *bp, u32 max_bufs,
 int bnxt_hwrm_set_coal(struct bnxt *bp)
 {
 	int i, rc = 0;
-	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input req = {0};
+	struct hwrm_ring_cmpl_ring_cfg_aggint_params_input req_rx = {0},
+							   req_tx = {0}, *req;
 	u16 max_buf, max_buf_irq;
 	u16 buf_tmr, buf_tmr_irq;
 	u32 flags;
 
-	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS,
-			       -1, -1);
+	bnxt_hwrm_cmd_hdr_init(bp, &req_rx,
+			       HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS, -1, -1);
+	bnxt_hwrm_cmd_hdr_init(bp, &req_tx,
+			       HWRM_RING_CMPL_RING_CFG_AGGINT_PARAMS, -1, -1);
 
 	/* Each rx completion (2 records) should be DMAed immediately.
 	 * DMA 1/4 of the completion buffers at a time.
@@ -3569,13 +3572,31 @@ int bnxt_hwrm_set_coal(struct bnxt *bp)
 		flags |= RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_RING_IDLE;
 
 	bnxt_hwrm_set_coal_params(bp, max_buf_irq << 16 | max_buf,
-				  buf_tmr_irq << 16 | buf_tmr, flags, &req);
+				  buf_tmr_irq << 16 | buf_tmr, flags, &req_rx);
+
+	/* max_buf must not be zero */
+	max_buf = clamp_t(u16, bp->tx_coal_bufs, 1, 63);
+	max_buf_irq = clamp_t(u16, bp->tx_coal_bufs_irq, 1, 63);
+	buf_tmr = BNXT_USEC_TO_COAL_TIMER(bp->tx_coal_ticks);
+	/* buf timer set to 1/4 of interrupt timer */
+	buf_tmr = max_t(u16, buf_tmr / 4, 1);
+	buf_tmr_irq = BNXT_USEC_TO_COAL_TIMER(bp->tx_coal_ticks_irq);
+	buf_tmr_irq = max_t(u16, buf_tmr_irq, 1);
+
+	flags = RING_CMPL_RING_CFG_AGGINT_PARAMS_REQ_FLAGS_TIMER_RESET;
+	bnxt_hwrm_set_coal_params(bp, max_buf_irq << 16 | max_buf,
+				  buf_tmr_irq << 16 | buf_tmr, flags, &req_tx);
 
 	mutex_lock(&bp->hwrm_cmd_lock);
 	for (i = 0; i < bp->cp_nr_rings; i++) {
-		req.ring_id = cpu_to_le16(bp->grp_info[i].cp_fw_ring_id);
+		struct bnxt_napi *bnapi = bp->bnapi[i];
 
-		rc = _hwrm_send_message(bp, &req, sizeof(req),
+		req = &req_rx;
+		if (!bnapi->rx_ring)
+			req = &req_tx;
+		req->ring_id = cpu_to_le16(bp->grp_info[i].cp_fw_ring_id);
+
+		rc = _hwrm_send_message(bp, req, sizeof(*req),
 					HWRM_CMD_TIMEOUT);
 		if (rc)
 			break;
@@ -5310,11 +5331,16 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 	bp->tx_ring_size = BNXT_DEFAULT_TX_RING_SIZE;
 
 	/* tick values in micro seconds */
-	bp->rx_coal_ticks = 4;
-	bp->rx_coal_bufs = 20;
+	bp->rx_coal_ticks = 12;
+	bp->rx_coal_bufs = 30;
 	bp->rx_coal_ticks_irq = 1;
 	bp->rx_coal_bufs_irq = 2;
 
+	bp->tx_coal_ticks = 25;
+	bp->tx_coal_bufs = 30;
+	bp->tx_coal_ticks_irq = 2;
+	bp->tx_coal_bufs_irq = 2;
+
 	init_timer(&bp->timer);
 	bp->timer.data = (unsigned long)bp;
 	bp->timer.function = bnxt_timer;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 6913307..ba67c4a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -968,10 +968,15 @@ struct bnxt {
 	__le16			vxlan_fw_dst_port_id;
 	u8			nge_port_cnt;
 	__le16			nge_fw_dst_port_id;
+
 	u16			rx_coal_ticks;
 	u16			rx_coal_ticks_irq;
 	u16			rx_coal_bufs;
 	u16			rx_coal_bufs_irq;
+	u16			tx_coal_ticks;
+	u16			tx_coal_ticks_irq;
+	u16			tx_coal_bufs;
+	u16			tx_coal_bufs_irq;
 
 #define BNXT_USEC_TO_COAL_TIMER(x)	((x) * 25 / 2)
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index d18a977..a2d2549 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -47,6 +47,11 @@ static int bnxt_get_coalesce(struct net_device *dev,
 	coal->rx_coalesce_usecs_irq = bp->rx_coal_ticks_irq;
 	coal->rx_max_coalesced_frames_irq = bp->rx_coal_bufs_irq / 2;
 
+	coal->tx_coalesce_usecs = bp->tx_coal_ticks;
+	coal->tx_max_coalesced_frames = bp->tx_coal_bufs;
+	coal->tx_coalesce_usecs_irq = bp->tx_coal_ticks_irq;
+	coal->tx_max_coalesced_frames_irq = bp->tx_coal_bufs_irq;
+
 	return 0;
 }
 
@@ -62,6 +67,11 @@ static int bnxt_set_coalesce(struct net_device *dev,
 	bp->rx_coal_ticks_irq = coal->rx_coalesce_usecs_irq;
 	bp->rx_coal_bufs_irq = coal->rx_max_coalesced_frames_irq * 2;
 
+	bp->tx_coal_ticks = coal->tx_coalesce_usecs;
+	bp->tx_coal_bufs = coal->tx_max_coalesced_frames;
+	bp->tx_coal_ticks_irq = coal->tx_coalesce_usecs_irq;
+	bp->tx_coal_bufs_irq = coal->tx_max_coalesced_frames_irq;
+
 	if (netif_running(dev))
 		rc = bnxt_hwrm_set_coal(bp);
 
-- 
cgit v0.10.2


From ff4fe81d2d49e3cad3bb45c8c5b9a49ca90ee10b Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:04 -0500
Subject: bnxt_en: Use firmware provided message timeout value.

Before this patch, we used a hardcoded value of 500 msec as the default
value for firmware message response timeout.  For better portability with
future hardware or debug platforms, use the value provided by firmware in
the first response and store it for all susequent messages.  Redefine the
macro HWRM_CMD_TIMEOUT to the stored value.  Since we don't have the
value yet in the first message, use the 500 ms default if the stored value
is zero.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 198c69d4..05a93a7 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -2638,6 +2638,9 @@ int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 	/* Ring channel doorbell */
 	writel(1, bp->bar0 + 0x100);
 
+	if (!timeout)
+		timeout = DFLT_HWRM_CMD_TIMEOUT;
+
 	i = 0;
 	if (intr_process) {
 		/* Wait until hwrm response cmpl interrupt is processed */
@@ -3809,6 +3812,10 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
 		 resp->hwrm_fw_maj, resp->hwrm_fw_min, resp->hwrm_fw_bld,
 		 resp->hwrm_intf_maj, resp->hwrm_intf_min, resp->hwrm_intf_upd);
 
+	bp->hwrm_cmd_timeout = le16_to_cpu(resp->def_req_timeout);
+	if (!bp->hwrm_cmd_timeout)
+		bp->hwrm_cmd_timeout = DFLT_HWRM_CMD_TIMEOUT;
+
 hwrm_ver_get_exit:
 	mutex_unlock(&bp->hwrm_cmd_lock);
 	return rc;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index ba67c4a..a3d4940 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -477,7 +477,8 @@ struct rx_tpa_end_cmp_ext {
 #define RING_CMP(idx)		((idx) & bp->cp_ring_mask)
 #define NEXT_CMP(idx)		RING_CMP(ADV_RAW_CMP(idx, 1))
 
-#define HWRM_CMD_TIMEOUT		500
+#define DFLT_HWRM_CMD_TIMEOUT		500
+#define HWRM_CMD_TIMEOUT		(bp->hwrm_cmd_timeout)
 #define HWRM_RESET_TIMEOUT		((HWRM_CMD_TIMEOUT) * 4)
 #define HWRM_RESP_ERR_CODE_MASK		0xffff
 #define HWRM_RESP_LEN_MASK		0xffff0000
@@ -957,6 +958,7 @@ struct bnxt {
 	void			*hwrm_dbg_resp_addr;
 	dma_addr_t		hwrm_dbg_resp_dma_addr;
 #define HWRM_DBG_REG_BUF_SIZE	128
+	int			hwrm_cmd_timeout;
 	struct mutex		hwrm_cmd_lock;	/* serialize hwrm messages */
 	struct hwrm_ver_get_output	ver_resp;
 #define FW_VER_STR_LEN		32
-- 
cgit v0.10.2


From a8643e1604c1f39a675c6b10a7f84260fa13590c Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:05 -0500
Subject: bnxt_en: Fix dmesg log firmware error messages.

Use appropriate firmware request header structure to prepare the
firmware messages.  This avoids the unnecessary conversion of the
fields to 32-bit fields.  Add appropriate endian conversion when
printing out the message fields in dmesg so that they appear correct
in the log.

Reported-by: Rob Swindell <swindell@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 05a93a7..9ff498e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -2600,28 +2600,26 @@ alloc_mem_err:
 void bnxt_hwrm_cmd_hdr_init(struct bnxt *bp, void *request, u16 req_type,
 			    u16 cmpl_ring, u16 target_id)
 {
-	struct hwrm_cmd_req_hdr *req = request;
+	struct input *req = request;
 
-	req->cmpl_ring_req_type =
-		cpu_to_le32(req_type | (cmpl_ring << HWRM_CMPL_RING_SFT));
-	req->target_id_seq_id = cpu_to_le32(target_id << HWRM_TARGET_FID_SFT);
+	req->req_type = cpu_to_le16(req_type);
+	req->cmpl_ring = cpu_to_le16(cmpl_ring);
+	req->target_id = cpu_to_le16(target_id);
 	req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
 }
 
 int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 {
 	int i, intr_process, rc;
-	struct hwrm_cmd_req_hdr *req = msg;
+	struct input *req = msg;
 	u32 *data = msg;
 	__le32 *resp_len, *valid;
 	u16 cp_ring_id, len = 0;
 	struct hwrm_err_output *resp = bp->hwrm_cmd_resp_addr;
 
-	req->target_id_seq_id |= cpu_to_le32(bp->hwrm_cmd_seq++);
+	req->seq_id = cpu_to_le16(bp->hwrm_cmd_seq++);
 	memset(resp, 0, PAGE_SIZE);
-	cp_ring_id = (le32_to_cpu(req->cmpl_ring_req_type) &
-		      HWRM_CMPL_RING_MASK) >>
-		     HWRM_CMPL_RING_SFT;
+	cp_ring_id = le16_to_cpu(req->cmpl_ring);
 	intr_process = (cp_ring_id == INVALID_HW_RING_ID) ? 0 : 1;
 
 	/* Write request msg to hwrm channel */
@@ -2632,8 +2630,7 @@ int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 
 	/* currently supports only one outstanding message */
 	if (intr_process)
-		bp->hwrm_intr_seq_id = le32_to_cpu(req->target_id_seq_id) &
-				       HWRM_SEQ_ID_MASK;
+		bp->hwrm_intr_seq_id = le16_to_cpu(req->seq_id);
 
 	/* Ring channel doorbell */
 	writel(1, bp->bar0 + 0x100);
@@ -2651,7 +2648,7 @@ int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 
 		if (bp->hwrm_intr_seq_id != HWRM_SEQ_ID_INVALID) {
 			netdev_err(bp->dev, "Resp cmpl intr err msg: 0x%x\n",
-				   req->cmpl_ring_req_type);
+				   le16_to_cpu(req->req_type));
 			return -1;
 		}
 	} else {
@@ -2667,8 +2664,8 @@ int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 
 		if (i >= timeout) {
 			netdev_err(bp->dev, "Error (timeout: %d) msg {0x%x 0x%x} len:%d\n",
-				   timeout, req->cmpl_ring_req_type,
-				   req->target_id_seq_id, *resp_len);
+				   timeout, le16_to_cpu(req->req_type),
+				   le16_to_cpu(req->seq_id), *resp_len);
 			return -1;
 		}
 
@@ -2682,8 +2679,8 @@ int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 
 		if (i >= timeout) {
 			netdev_err(bp->dev, "Error (timeout: %d) msg {0x%x 0x%x} len:%d v:%d\n",
-				   timeout, req->cmpl_ring_req_type,
-				   req->target_id_seq_id, len, *valid);
+				   timeout, le16_to_cpu(req->req_type),
+				   le16_to_cpu(req->seq_id), len, *valid);
 			return -1;
 		}
 	}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index a3d4940..3936e96 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -481,9 +481,11 @@ struct rx_tpa_end_cmp_ext {
 #define HWRM_CMD_TIMEOUT		(bp->hwrm_cmd_timeout)
 #define HWRM_RESET_TIMEOUT		((HWRM_CMD_TIMEOUT) * 4)
 #define HWRM_RESP_ERR_CODE_MASK		0xffff
+#define HWRM_RESP_LEN_OFFSET		4
 #define HWRM_RESP_LEN_MASK		0xffff0000
 #define HWRM_RESP_LEN_SFT		16
 #define HWRM_RESP_VALID_MASK		0xff000000
+#define HWRM_SEQ_ID_INVALID		-1
 #define BNXT_HWRM_REQ_MAX_SIZE		128
 #define BNXT_HWRM_REQS_PER_PAGE		(BNXT_PAGE_SIZE /	\
 					 BNXT_HWRM_REQ_MAX_SIZE)
@@ -645,19 +647,6 @@ struct bnxt_irq {
 
 #define INVALID_STATS_CTX_ID	-1
 
-struct hwrm_cmd_req_hdr {
-#define HWRM_CMPL_RING_MASK	0xffff0000
-#define HWRM_CMPL_RING_SFT	16
-	__le32	cmpl_ring_req_type;
-#define HWRM_SEQ_ID_MASK	0xffff
-#define HWRM_SEQ_ID_INVALID -1
-#define HWRM_RESP_LEN_OFFSET	4
-#define HWRM_TARGET_FID_MASK	0xffff0000
-#define HWRM_TARGET_FID_SFT	16
-	__le32	target_id_seq_id;
-	__le64	resp_addr;
-};
-
 struct bnxt_ring_grp_info {
 	u16	fw_stats_ctx;
 	u16	fw_grp_id;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index 9f6b4cc..0c5f510 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -801,8 +801,8 @@ static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf)
 static int bnxt_vf_req_validate_snd(struct bnxt *bp, struct bnxt_vf_info *vf)
 {
 	int rc = 0;
-	struct hwrm_cmd_req_hdr *encap_req = vf->hwrm_cmd_req_addr;
-	u32 req_type = le32_to_cpu(encap_req->cmpl_ring_req_type) & 0xffff;
+	struct input *encap_req = vf->hwrm_cmd_req_addr;
+	u32 req_type = le16_to_cpu(encap_req->req_type);
 
 	switch (req_type) {
 	case HWRM_CFA_L2_FILTER_ALLOC:
-- 
cgit v0.10.2


From 3ebf6f0a09a284adef62111c7cfca29f56d6cce7 Mon Sep 17 00:00:00 2001
From: Rob Swindell <swindell@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:06 -0500
Subject: bnxt_en: Add installed-package firmware version reporting via Ethtool
 GDRVINFO

For everything to fit, we remove the PHY microcode version and replace it
with the firmware package version in the fw_version string.

Signed-off-by: Rob Swindell <swindell@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 9ff498e..6e45a00 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3805,7 +3805,7 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
 			    resp->hwrm_intf_upd);
 		netdev_warn(bp->dev, "Please update firmware with HWRM interface 1.0.0 or newer.\n");
 	}
-	snprintf(bp->fw_ver_str, BC_HWRM_STR_LEN, "bc %d.%d.%d rm %d.%d.%d",
+	snprintf(bp->fw_ver_str, BC_HWRM_STR_LEN, "%d.%d.%d/%d.%d.%d",
 		 resp->hwrm_fw_maj, resp->hwrm_fw_min, resp->hwrm_fw_bld,
 		 resp->hwrm_intf_maj, resp->hwrm_intf_min, resp->hwrm_intf_upd);
 
@@ -5725,7 +5725,6 @@ static int bnxt_probe_phy(struct bnxt *bp)
 {
 	int rc = 0;
 	struct bnxt_link_info *link_info = &bp->link_info;
-	char phy_ver[PHY_VER_STR_LEN];
 
 	rc = bnxt_update_link(bp, false);
 	if (rc) {
@@ -5745,11 +5744,6 @@ static int bnxt_probe_phy(struct bnxt *bp)
 		link_info->req_duplex = link_info->duplex_setting;
 		link_info->req_flow_ctrl = link_info->force_pause_setting;
 	}
-	snprintf(phy_ver, PHY_VER_STR_LEN, " ph %d.%d.%d",
-		 link_info->phy_ver[0],
-		 link_info->phy_ver[1],
-		 link_info->phy_ver[2]);
-	strcat(bp->fw_ver_str, phy_ver);
 	return rc;
 }
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index a2d2549..bfda92e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -7,6 +7,7 @@
  * the Free Software Foundation.
  */
 
+#include <linux/ctype.h>
 #include <linux/ethtool.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
@@ -20,6 +21,8 @@
 #include "bnxt_fw_hdr.h"	/* Firmware hdr constant and structure defs */
 #define FLASH_NVRAM_TIMEOUT	((HWRM_CMD_TIMEOUT) * 100)
 
+static char *bnxt_get_pkgver(struct net_device *dev, char *buf, size_t buflen);
+
 static u32 bnxt_get_msglevel(struct net_device *dev)
 {
 	struct bnxt *bp = netdev_priv(dev);
@@ -469,10 +472,20 @@ static void bnxt_get_drvinfo(struct net_device *dev,
 			     struct ethtool_drvinfo *info)
 {
 	struct bnxt *bp = netdev_priv(dev);
+	char *pkglog;
+	char *pkgver = NULL;
 
+	pkglog = kmalloc(BNX_PKG_LOG_MAX_LENGTH, GFP_KERNEL);
+	if (pkglog)
+		pkgver = bnxt_get_pkgver(dev, pkglog, BNX_PKG_LOG_MAX_LENGTH);
 	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
 	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
-	strlcpy(info->fw_version, bp->fw_ver_str, sizeof(info->fw_version));
+	if (pkgver && *pkgver != 0 && isdigit(*pkgver))
+		snprintf(info->fw_version, sizeof(info->fw_version) - 1,
+			 "%s pkg %s", bp->fw_ver_str, pkgver);
+	else
+		strlcpy(info->fw_version, bp->fw_ver_str,
+			sizeof(info->fw_version));
 	strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info));
 	info->n_stats = BNXT_NUM_STATS * bp->cp_nr_rings;
 	info->testinfo_len = BNXT_NUM_TESTS(bp);
@@ -480,6 +493,7 @@ static void bnxt_get_drvinfo(struct net_device *dev,
 	info->eedump_len = 0;
 	/* TODO CHIMP FW: reg dump details */
 	info->regdump_len = 0;
+	kfree(pkglog);
 }
 
 static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info)
@@ -1111,6 +1125,85 @@ static int bnxt_get_nvram_item(struct net_device *dev, u32 index, u32 offset,
 	return rc;
 }
 
+static int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
+				u16 ext, u16 *index, u32 *item_length,
+				u32 *data_length)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	int rc;
+	struct hwrm_nvm_find_dir_entry_input req = {0};
+	struct hwrm_nvm_find_dir_entry_output *output = bp->hwrm_cmd_resp_addr;
+
+	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_FIND_DIR_ENTRY, -1, -1);
+	req.enables = 0;
+	req.dir_idx = 0;
+	req.dir_type = cpu_to_le16(type);
+	req.dir_ordinal = cpu_to_le16(ordinal);
+	req.dir_ext = cpu_to_le16(ext);
+	req.opt_ordinal = NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_EQ;
+	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (rc == 0) {
+		if (index)
+			*index = le16_to_cpu(output->dir_idx);
+		if (item_length)
+			*item_length = le32_to_cpu(output->dir_item_length);
+		if (data_length)
+			*data_length = le32_to_cpu(output->dir_data_length);
+	}
+	return rc;
+}
+
+static char *bnxt_parse_pkglog(int desired_field, u8 *data, size_t datalen)
+{
+	char	*retval = NULL;
+	char	*p;
+	char	*value;
+	int	field = 0;
+
+	if (datalen < 1)
+		return NULL;
+	/* null-terminate the log data (removing last '\n'): */
+	data[datalen - 1] = 0;
+	for (p = data; *p != 0; p++) {
+		field = 0;
+		retval = NULL;
+		while (*p != 0 && *p != '\n') {
+			value = p;
+			while (*p != 0 && *p != '\t' && *p != '\n')
+				p++;
+			if (field == desired_field)
+				retval = value;
+			if (*p != '\t')
+				break;
+			*p = 0;
+			field++;
+			p++;
+		}
+		if (*p == 0)
+			break;
+		*p = 0;
+	}
+	return retval;
+}
+
+static char *bnxt_get_pkgver(struct net_device *dev, char *buf, size_t buflen)
+{
+	u16 index = 0;
+	u32 datalen;
+
+	if (bnxt_find_nvram_item(dev, BNX_DIR_TYPE_PKG_LOG,
+				 BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE,
+				 &index, NULL, &datalen) != 0)
+		return NULL;
+
+	memset(buf, 0, buflen);
+	if (bnxt_get_nvram_item(dev, index, 0, datalen, buf) != 0)
+		return NULL;
+
+	return bnxt_parse_pkglog(BNX_PKG_LOG_FIELD_IDX_PKG_VERSION, buf,
+		datalen);
+}
+
 static int bnxt_get_eeprom(struct net_device *dev,
 			   struct ethtool_eeprom *eeprom,
 			   u8 *data)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
index 3cf3e1b..43ef392 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
@@ -50,10 +50,24 @@ enum bnxt_nvm_directory_type {
 
 #define BNX_DIR_ORDINAL_FIRST			0
 
+#define BNX_DIR_EXT_NONE			0
 #define BNX_DIR_EXT_INACTIVE			(1 << 0)
 #define BNX_DIR_EXT_UPDATE			(1 << 1)
 
+#define BNX_DIR_ATTR_NONE			0
 #define BNX_DIR_ATTR_NO_CHKSUM			(1 << 0)
 #define BNX_DIR_ATTR_PROP_STREAM		(1 << 1)
 
+#define BNX_PKG_LOG_MAX_LENGTH			4096
+
+enum bnxnvm_pkglog_field_index {
+	BNX_PKG_LOG_FIELD_IDX_INSTALLED_TIMESTAMP	= 0,
+	BNX_PKG_LOG_FIELD_IDX_PKG_DESCRIPTION		= 1,
+	BNX_PKG_LOG_FIELD_IDX_PKG_VERSION		= 2,
+	BNX_PKG_LOG_FIELD_IDX_PKG_TIMESTAMP		= 3,
+	BNX_PKG_LOG_FIELD_IDX_PKG_CHECKSUM		= 4,
+	BNX_PKG_LOG_FIELD_IDX_INSTALLED_ITEMS		= 5,
+	BNX_PKG_LOG_FIELD_IDX_INSTALLED_MASK		= 6
+};
+
 #endif				/* Don't add anything after this line */
-- 
cgit v0.10.2


From fbfbc4851dd709cf1327afc283f9cca00235dcb3 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:07 -0500
Subject: bnxt_en: Refactor _hwrm_send_message().

Add a new function bnxt_do_send_msg() to do essentially the same thing
with an additional paramter to silence error response messages.  All
current callers will set silent to false.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 6e45a00..48a7bbc 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -2608,7 +2608,8 @@ void bnxt_hwrm_cmd_hdr_init(struct bnxt *bp, void *request, u16 req_type,
 	req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
 }
 
-int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
+static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len,
+				 int timeout, bool silent)
 {
 	int i, intr_process, rc;
 	struct input *req = msg;
@@ -2686,13 +2687,16 @@ int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 	}
 
 	rc = le16_to_cpu(resp->error_code);
-	if (rc) {
+	if (rc && !silent)
 		netdev_err(bp->dev, "hwrm req_type 0x%x seq id 0x%x error 0x%x\n",
 			   le16_to_cpu(resp->req_type),
 			   le16_to_cpu(resp->seq_id), rc);
-		return rc;
-	}
-	return 0;
+	return rc;
+}
+
+int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
+{
+	return bnxt_hwrm_do_send_msg(bp, msg, msg_len, timeout, false);
 }
 
 int hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
-- 
cgit v0.10.2


From 90e209213096110bce06ef580e1c73702fe4a288 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 26 Feb 2016 04:00:08 -0500
Subject: bnxt_en: Add hwrm_send_message_silent().

This is used to send NVM_FIND_DIR_ENTRY messages which can return error
if the entry is not found.  This is normal and the error message will
cause unnecessary alarm, so silence it.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 48a7bbc..aa6a318 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -2709,6 +2709,17 @@ int hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
 	return rc;
 }
 
+int hwrm_send_message_silent(struct bnxt *bp, void *msg, u32 msg_len,
+			     int timeout)
+{
+	int rc;
+
+	mutex_lock(&bp->hwrm_cmd_lock);
+	rc = bnxt_hwrm_do_send_msg(bp, msg, msg_len, timeout, true);
+	mutex_unlock(&bp->hwrm_cmd_lock);
+	return rc;
+}
+
 static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
 {
 	struct hwrm_func_drv_rgtr_input req = {0};
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 3936e96..9aa38f5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1095,6 +1095,7 @@ void bnxt_set_ring_params(struct bnxt *);
 void bnxt_hwrm_cmd_hdr_init(struct bnxt *, void *, u16, u16, u16);
 int _hwrm_send_message(struct bnxt *, void *, u32, int);
 int hwrm_send_message(struct bnxt *, void *, u32, int);
+int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
 int bnxt_hwrm_set_coal(struct bnxt *);
 int bnxt_hwrm_func_qcaps(struct bnxt *);
 int bnxt_hwrm_set_pause(struct bnxt *);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index bfda92e..84ea26d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1141,7 +1141,7 @@ static int bnxt_find_nvram_item(struct net_device *dev, u16 type, u16 ordinal,
 	req.dir_ordinal = cpu_to_le16(ordinal);
 	req.dir_ext = cpu_to_le16(ext);
 	req.opt_ordinal = NVM_FIND_DIR_ENTRY_REQ_OPT_ORDINAL_EQ;
-	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_send_message_silent(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
 	if (rc == 0) {
 		if (index)
 			*index = le16_to_cpu(output->dir_idx);
-- 
cgit v0.10.2


From 871b642adebe300be2e50aa5f65a418510f636ec Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 26 Feb 2016 10:45:37 +0100
Subject: netdev: introduce ndo_set_rx_headroom

This method allows the controlling device (i.e. the bridge) to specify
additional headroom to be allocated for skb head on frame reception.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e52077f..efe7cec 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1093,6 +1093,12 @@ struct tc_to_netdev {
  *	This function is used to get egress tunnel information for given skb.
  *	This is useful for retrieving outer tunnel header parameters while
  *	sampling packet.
+ * void (*ndo_set_rx_headroom)(struct net_device *dev, int needed_headroom);
+ *	This function is used to specify the headroom that the skb must
+ *	consider when allocation skb during packet reception. Setting
+ *	appropriate rx headroom value allows avoiding skb head copy on
+ *	forward. Setting a negative value reset the rx headroom to the
+ *	default value.
  *
  */
 struct net_device_ops {
@@ -1278,6 +1284,8 @@ struct net_device_ops {
 							 bool proto_down);
 	int			(*ndo_fill_metadata_dst)(struct net_device *dev,
 						       struct sk_buff *skb);
+	void			(*ndo_set_rx_headroom)(struct net_device *dev,
+						       int needed_headroom);
 };
 
 /**
@@ -1315,6 +1323,8 @@ struct net_device_ops {
  * @IFF_L3MDEV_SLAVE: device is enslaved to an L3 master device
  * @IFF_TEAM: device is a team device
  * @IFF_RXFH_CONFIGURED: device has had Rx Flow indirection table configured
+ * @IFF_PHONY_HEADROOM: the headroom value is controlled by an external
+ *	entity (i.e. the master device for bridged veth)
  */
 enum netdev_priv_flags {
 	IFF_802_1Q_VLAN			= 1<<0,
@@ -1343,6 +1353,7 @@ enum netdev_priv_flags {
 	IFF_L3MDEV_SLAVE		= 1<<23,
 	IFF_TEAM			= 1<<24,
 	IFF_RXFH_CONFIGURED		= 1<<25,
+	IFF_PHONY_HEADROOM		= 1<<26,
 };
 
 #define IFF_802_1Q_VLAN			IFF_802_1Q_VLAN
@@ -1937,6 +1948,26 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev,
 				    struct sk_buff *skb,
 				    void *accel_priv);
 
+/* returns the headroom that the master device needs to take in account
+ * when forwarding to this dev
+ */
+static inline unsigned netdev_get_fwd_headroom(struct net_device *dev)
+{
+	return dev->priv_flags & IFF_PHONY_HEADROOM ? 0 : dev->needed_headroom;
+}
+
+static inline void netdev_set_rx_headroom(struct net_device *dev, int new_hr)
+{
+	if (dev->netdev_ops->ndo_set_rx_headroom)
+		dev->netdev_ops->ndo_set_rx_headroom(dev, new_hr);
+}
+
+/* set the device rx headroom to the dev's default */
+static inline void netdev_reset_rx_headroom(struct net_device *dev)
+{
+	netdev_set_rx_headroom(dev, -1);
+}
+
 /*
  * Net namespace inlines
  */
-- 
cgit v0.10.2


From 45493d47c3db862263583400c1f721f13f2a0610 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 26 Feb 2016 10:45:38 +0100
Subject: bridge: notify enslaved devices of headroom changes

On bridge needed_headroom changes, the enslaved devices are
notified via the ndo_set_rx_headroom method

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index b37a1cc..a73df33 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -223,6 +223,31 @@ static void destroy_nbp_rcu(struct rcu_head *head)
 	destroy_nbp(p);
 }
 
+static unsigned get_max_headroom(struct net_bridge *br)
+{
+	unsigned max_headroom = 0;
+	struct net_bridge_port *p;
+
+	list_for_each_entry(p, &br->port_list, list) {
+		unsigned dev_headroom = netdev_get_fwd_headroom(p->dev);
+
+		if (dev_headroom > max_headroom)
+			max_headroom = dev_headroom;
+	}
+
+	return max_headroom;
+}
+
+static void update_headroom(struct net_bridge *br, int new_hr)
+{
+	struct net_bridge_port *p;
+
+	list_for_each_entry(p, &br->port_list, list)
+		netdev_set_rx_headroom(p->dev, new_hr);
+
+	br->dev->needed_headroom = new_hr;
+}
+
 /* Delete port(interface) from bridge is done in two steps.
  * via RCU. First step, marks device as down. That deletes
  * all the timers and stops new packets from flowing through.
@@ -248,6 +273,9 @@ static void del_nbp(struct net_bridge_port *p)
 	br_ifinfo_notify(RTM_DELLINK, p);
 
 	list_del_rcu(&p->list);
+	if (netdev_get_fwd_headroom(dev) == br->dev->needed_headroom)
+		update_headroom(br, get_max_headroom(br));
+	netdev_reset_rx_headroom(dev);
 
 	nbp_vlan_flush(p);
 	br_fdb_delete_by_port(br, p, 0, 1);
@@ -438,6 +466,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
 {
 	struct net_bridge_port *p;
 	int err = 0;
+	unsigned br_hr, dev_hr;
 	bool changed_addr;
 
 	/* Don't allow bridging non-ethernet like devices, or DSA-enabled
@@ -505,8 +534,12 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
 
 	netdev_update_features(br->dev);
 
-	if (br->dev->needed_headroom < dev->needed_headroom)
-		br->dev->needed_headroom = dev->needed_headroom;
+	br_hr = br->dev->needed_headroom;
+	dev_hr = netdev_get_fwd_headroom(dev);
+	if (br_hr < dev_hr)
+		update_headroom(br, dev_hr);
+	else
+		netdev_set_rx_headroom(dev, br_hr);
 
 	if (br_fdb_insert(br, p, dev->dev_addr, 0))
 		netdev_err(dev, "failed insert local address bridge forwarding table\n");
-- 
cgit v0.10.2


From 3a927bc7cf9d0fbe8f4a8189dd5f8440228f64e7 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 26 Feb 2016 10:45:39 +0100
Subject: ovs: propagate per dp max headroom to all vports

This patch implements bookkeeping support to compute the maximum
headroom for all the devices in each datapath. When said value
changes, the underlying devs are notified via the
ndo_set_rx_headroom method.

This also increases the internal vports xmit performance.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index c4e8455..e6a7d49 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1908,6 +1908,29 @@ static struct vport *lookup_vport(struct net *net,
 		return ERR_PTR(-EINVAL);
 }
 
+/* Called with ovs_mutex */
+static void update_headroom(struct datapath *dp)
+{
+	unsigned dev_headroom, max_headroom = 0;
+	struct net_device *dev;
+	struct vport *vport;
+	int i;
+
+	for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++) {
+		hlist_for_each_entry_rcu(vport, &dp->ports[i], dp_hash_node) {
+			dev = vport->dev;
+			dev_headroom = netdev_get_fwd_headroom(dev);
+			if (dev_headroom > max_headroom)
+				max_headroom = dev_headroom;
+		}
+	}
+
+	dp->max_headroom = max_headroom;
+	for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++)
+		hlist_for_each_entry_rcu(vport, &dp->ports[i], dp_hash_node)
+			netdev_set_rx_headroom(vport->dev, max_headroom);
+}
+
 static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr **a = info->attrs;
@@ -1973,6 +1996,12 @@ restart:
 
 	err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid,
 				      info->snd_seq, 0, OVS_VPORT_CMD_NEW);
+
+	if (netdev_get_fwd_headroom(vport->dev) > dp->max_headroom)
+		update_headroom(dp);
+	else
+		netdev_set_rx_headroom(vport->dev, dp->max_headroom);
+
 	BUG_ON(err < 0);
 	ovs_unlock();
 
@@ -2039,8 +2068,10 @@ exit_unlock_free:
 
 static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
 {
+	bool must_update_headroom = false;
 	struct nlattr **a = info->attrs;
 	struct sk_buff *reply;
+	struct datapath *dp;
 	struct vport *vport;
 	int err;
 
@@ -2062,7 +2093,16 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid,
 				      info->snd_seq, 0, OVS_VPORT_CMD_DEL);
 	BUG_ON(err < 0);
+
+	/* the vport deletion may trigger dp headroom update */
+	dp = vport->dp;
+	if (netdev_get_fwd_headroom(vport->dev) == dp->max_headroom)
+		must_update_headroom = true;
+	netdev_reset_rx_headroom(vport->dev);
 	ovs_dp_detach_port(vport);
+
+	if (must_update_headroom)
+		update_headroom(dp);
 	ovs_unlock();
 
 	ovs_notify(&dp_vport_genl_family, reply, info);
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index 67bdecd..427e39a 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -68,6 +68,8 @@ struct dp_stats_percpu {
  * ovs_mutex and RCU.
  * @stats_percpu: Per-CPU datapath statistics.
  * @net: Reference to net namespace.
+ * @max_headroom: the maximum headroom of all vports in this datapath; it will
+ * be used by all the internal vports in this dp.
  *
  * Context: See the comment on locking at the top of datapath.c for additional
  * locking information.
@@ -89,6 +91,8 @@ struct datapath {
 	possible_net_t net;
 
 	u32 user_features;
+
+	u32 max_headroom;
 };
 
 /**
diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index ec76398..83a5534 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -138,6 +138,11 @@ internal_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	return stats;
 }
 
+void internal_set_rx_headroom(struct net_device *dev, int new_hr)
+{
+	dev->needed_headroom = new_hr;
+}
+
 static const struct net_device_ops internal_dev_netdev_ops = {
 	.ndo_open = internal_dev_open,
 	.ndo_stop = internal_dev_stop,
@@ -145,6 +150,7 @@ static const struct net_device_ops internal_dev_netdev_ops = {
 	.ndo_set_mac_address = eth_mac_addr,
 	.ndo_change_mtu = internal_dev_change_mtu,
 	.ndo_get_stats64 = internal_get_stats,
+	.ndo_set_rx_headroom = internal_set_rx_headroom,
 };
 
 static struct rtnl_link_ops internal_dev_link_ops __read_mostly = {
@@ -158,7 +164,8 @@ static void do_setup(struct net_device *netdev)
 	netdev->netdev_ops = &internal_dev_netdev_ops;
 
 	netdev->priv_flags &= ~IFF_TX_SKB_SHARING;
-	netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_OPENVSWITCH;
+	netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_OPENVSWITCH |
+			      IFF_PHONY_HEADROOM;
 	netdev->destructor = internal_dev_destructor;
 	netdev->ethtool_ops = &internal_dev_ethtool_ops;
 	netdev->rtnl_link_ops = &internal_dev_link_ops;
@@ -199,6 +206,7 @@ static struct vport *internal_dev_create(const struct vport_parms *parms)
 		err = -ENOMEM;
 		goto error_free_netdev;
 	}
+	vport->dev->needed_headroom = vport->dp->max_headroom;
 
 	dev_net_set(vport->dev, ovs_dp_get_net(vport->dp));
 	internal_dev = internal_dev_priv(vport->dev);
-- 
cgit v0.10.2


From eaea34b23c46bf17b4a5638be69ab3561854f34b Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 26 Feb 2016 10:45:40 +0100
Subject: net/tun: implement ndo_set_rx_headroom

ndo_set_rx_headroom controls the align value used by tun devices to
allocate skbs on frame reception.
When the xmit device adds a large encapsulation, this avoids an skb
head reallocation on forwarding.

The measured improvement when forwarding towards a vxlan dev with
frame size below the egress device MTU is as follow:

vxlan over ipv6, bridged: +6%
vxlan over ipv6, ovs: +7%

In case of ipv4 tunnels there is no improvement, since the tun
device default alignment provides enough headroom to avoid the skb
head reallocation.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 88bb8cc..afdf950 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -187,6 +187,7 @@ struct tun_struct {
 #define TUN_USER_FEATURES (NETIF_F_HW_CSUM|NETIF_F_TSO_ECN|NETIF_F_TSO| \
 			  NETIF_F_TSO6|NETIF_F_UFO)
 
+	int			align;
 	int			vnet_hdr_sz;
 	int			sndbuf;
 	struct tap_filter	txflt;
@@ -934,6 +935,17 @@ static void tun_poll_controller(struct net_device *dev)
 	return;
 }
 #endif
+
+static void tun_set_headroom(struct net_device *dev, int new_hr)
+{
+	struct tun_struct *tun = netdev_priv(dev);
+
+	if (new_hr < NET_SKB_PAD)
+		new_hr = NET_SKB_PAD;
+
+	tun->align = new_hr;
+}
+
 static const struct net_device_ops tun_netdev_ops = {
 	.ndo_uninit		= tun_net_uninit,
 	.ndo_open		= tun_net_open,
@@ -945,6 +957,7 @@ static const struct net_device_ops tun_netdev_ops = {
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= tun_poll_controller,
 #endif
+	.ndo_set_rx_headroom	= tun_set_headroom,
 };
 
 static const struct net_device_ops tap_netdev_ops = {
@@ -962,6 +975,7 @@ static const struct net_device_ops tap_netdev_ops = {
 	.ndo_poll_controller	= tun_poll_controller,
 #endif
 	.ndo_features_check	= passthru_features_check,
+	.ndo_set_rx_headroom	= tun_set_headroom,
 };
 
 static void tun_flow_init(struct tun_struct *tun)
@@ -1086,7 +1100,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 	struct tun_pi pi = { 0, cpu_to_be16(ETH_P_IP) };
 	struct sk_buff *skb;
 	size_t total_len = iov_iter_count(from);
-	size_t len = total_len, align = NET_SKB_PAD, linear;
+	size_t len = total_len, align = tun->align, linear;
 	struct virtio_net_hdr gso = { 0 };
 	int good_linear;
 	int copylen;
@@ -1694,6 +1708,7 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 		tun->txflt.count = 0;
 		tun->vnet_hdr_sz = sizeof(struct virtio_net_hdr);
 
+		tun->align = NET_SKB_PAD;
 		tun->filter_attached = false;
 		tun->sndbuf = tfile->socket.sk->sk_sndbuf;
 
-- 
cgit v0.10.2


From 163e529200af7b7521fbde5dbcc653cf3ce597df Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 26 Feb 2016 10:45:41 +0100
Subject: veth: implement ndo_set_rx_headroom

The rx headroom for veth dev is the peer device needed_headroom.
Avoid ping-pong updates setting the private flag IFF_PHONY_HEADROOM.

This avoids skb head reallocation when forwarding from a veth dev
towards a device adding some kind of encapsulation.

When transmitting frames below the MTU size towards a vxlan device,
this gives about 10% performance speed-up when OVS is used to connect
the veth and the vxlan device and a little more when using a
plain Linux bridge.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index ba21d07..4f30a6a 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -35,6 +35,7 @@ struct pcpu_vstats {
 struct veth_priv {
 	struct net_device __rcu	*peer;
 	atomic64_t		dropped;
+	unsigned		requested_headroom;
 };
 
 /*
@@ -271,6 +272,29 @@ static int veth_get_iflink(const struct net_device *dev)
 	return iflink;
 }
 
+static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
+{
+	struct veth_priv *peer_priv, *priv = netdev_priv(dev);
+	struct net_device *peer;
+
+	if (new_hr < 0)
+		new_hr = 0;
+
+	rcu_read_lock();
+	peer = rcu_dereference(priv->peer);
+	if (unlikely(!peer))
+		goto out;
+
+	peer_priv = netdev_priv(peer);
+	priv->requested_headroom = new_hr;
+	new_hr = max(priv->requested_headroom, peer_priv->requested_headroom);
+	dev->needed_headroom = new_hr;
+	peer->needed_headroom = new_hr;
+
+out:
+	rcu_read_unlock();
+}
+
 static const struct net_device_ops veth_netdev_ops = {
 	.ndo_init            = veth_dev_init,
 	.ndo_open            = veth_open,
@@ -285,6 +309,7 @@ static const struct net_device_ops veth_netdev_ops = {
 #endif
 	.ndo_get_iflink		= veth_get_iflink,
 	.ndo_features_check	= passthru_features_check,
+	.ndo_set_rx_headroom	= veth_set_rx_headroom,
 };
 
 #define VETH_FEATURES (NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_ALL_TSO |    \
@@ -301,6 +326,7 @@ static void veth_setup(struct net_device *dev)
 	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
 	dev->priv_flags |= IFF_NO_QUEUE;
+	dev->priv_flags |= IFF_PHONY_HEADROOM;
 
 	dev->netdev_ops = &veth_netdev_ops;
 	dev->ethtool_ops = &veth_ethtool_ops;
-- 
cgit v0.10.2


From 6843e7a2abe7cac10c19702ffec90018df6f040d Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Fri, 26 Feb 2016 07:53:49 -0800
Subject: net: sched: consolidate offload decision in cls_u32

The offload decision was originally very basic and tied to if the dev
implemented the appropriate ndo op hook. The next step is to allow
the user to more flexibly define if any paticular rule should be
offloaded or not. In order to have this logic in one function lift
the current check into a helper routine tc_should_offload().

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 2121df5..e64d20b 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -392,4 +392,9 @@ struct tc_cls_u32_offload {
 	};
 };
 
+static inline bool tc_should_offload(struct net_device *dev)
+{
+	return dev->netdev_ops->ndo_setup_tc;
+}
+
 #endif
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index d54bc94..24e888b 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -434,7 +434,7 @@ static void u32_remove_hw_knode(struct tcf_proto *tp, u32 handle)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (dev->netdev_ops->ndo_setup_tc) {
+	if (tc_should_offload(dev)) {
 		offload.cls_u32->command = TC_CLSU32_DELETE_KNODE;
 		offload.cls_u32->knode.handle = handle;
 		dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle,
@@ -451,7 +451,7 @@ static void u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (dev->netdev_ops->ndo_setup_tc) {
+	if (tc_should_offload(dev)) {
 		offload.cls_u32->command = TC_CLSU32_NEW_HNODE;
 		offload.cls_u32->hnode.divisor = h->divisor;
 		offload.cls_u32->hnode.handle = h->handle;
@@ -471,7 +471,7 @@ static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (dev->netdev_ops->ndo_setup_tc) {
+	if (tc_should_offload(dev)) {
 		offload.cls_u32->command = TC_CLSU32_DELETE_HNODE;
 		offload.cls_u32->hnode.divisor = h->divisor;
 		offload.cls_u32->hnode.handle = h->handle;
@@ -491,7 +491,7 @@ static void u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (dev->netdev_ops->ndo_setup_tc) {
+	if (tc_should_offload(dev)) {
 		offload.cls_u32->command = TC_CLSU32_REPLACE_KNODE;
 		offload.cls_u32->knode.handle = n->handle;
 		offload.cls_u32->knode.fshift = n->fshift;
-- 
cgit v0.10.2


From 2b6ab0d3aae6bf1e08118060b0c5565778cd6b21 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Fri, 26 Feb 2016 07:54:13 -0800
Subject: net: cls_u32: move TC offload feature bit into cls_u32 offload logic

In the original series drivers would get offload requests for cls_u32
rules even if the feature bit is disabled. This meant the driver had
to do a boiler plate check on the feature bit before adding/deleting
the rule.

This patch lifts the check into the core code and removes it from the
driver specific case.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index cf4b729..b893ff8 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8400,9 +8400,6 @@ int __ixgbe_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 
 	if (TC_H_MAJ(handle) == TC_H_MAJ(TC_H_INGRESS) &&
 	    tc->type == TC_SETUP_CLSU32) {
-		if (!(dev->features & NETIF_F_HW_TC))
-			return -EINVAL;
-
 		switch (tc->cls_u32->command) {
 		case TC_CLSU32_NEW_KNODE:
 		case TC_CLSU32_REPLACE_KNODE:
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index e64d20b..6096e96 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -394,6 +394,9 @@ struct tc_cls_u32_offload {
 
 static inline bool tc_should_offload(struct net_device *dev)
 {
+	if (!(dev->features & NETIF_F_HW_TC))
+		return false;
+
 	return dev->netdev_ops->ndo_setup_tc;
 }
 
-- 
cgit v0.10.2


From 9e8ce79cd711d4dfe09d8bba6822cd9bb7db96bd Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Fri, 26 Feb 2016 07:54:39 -0800
Subject: net: sched: cls_u32 add bit to specify software only rules

In the initial implementation the only way to stop a rule from being
inserted into the hardware table was via the device feature flag.
However this doesn't work well when working on an end host system
where packets are expect to hit both the hardware and software
datapaths.

For example we can imagine a rule that will match an IP address and
increment a field. If we install this rule in both hardware and
software we may increment the field twice. To date we have only
added support for the drop action so we have been able to ignore
these cases. But as we extend the action support we will hit this
example plus more such cases. Arguably these are not even corner
cases in many working systems these cases will be common.

To avoid forcing the driver to always abort (i.e. the above example)
this patch adds a flag to add a rule in software only. A careful
user can use this flag to build software and hardware datapaths
that work together. One example we have found particularly useful
is to use hardware resources to set the skb->mark on the skb when
the match may be expensive to run in software but a mark lookup
in a hash table is cheap. The idea here is hardware can do in one
lookup what the u32 classifier may need to traverse multiple lists
and hash tables to compute. The flag is only passed down on inserts.
On deletion to avoid stale references in hardware we always try
to remove a rule if it exists.

The flags field is part of the classifier specific options. Although
it is tempting to lift this into the generic structure doing this
proves difficult do to how the tc netlink attributes are implemented
along with how the dump/change routines are called. There is also
precedence for putting seemingly generic pieces in the specific
classifier options such as TCA_U32_POLICE, TCA_U32_ACT, etc. So
although not ideal I've left FLAGS in the u32 options as well as it
simplifies the code greatly and user space has already learned how
to manage these bits ala 'tc' tool.

Another thing if trying to update a rule we require the flags to
be unchanged. This is to force user space, software u32 and
the hardware u32 to keep in sync. Thanks to Simon Horman for
catching this case.

Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 6096e96..bea14ee 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -392,12 +392,21 @@ struct tc_cls_u32_offload {
 	};
 };
 
-static inline bool tc_should_offload(struct net_device *dev)
+/* tca flags definitions */
+#define TCA_CLS_FLAGS_SKIP_HW 1
+
+static inline bool tc_should_offload(struct net_device *dev, u32 flags)
 {
 	if (!(dev->features & NETIF_F_HW_TC))
 		return false;
 
-	return dev->netdev_ops->ndo_setup_tc;
+	if (flags & TCA_CLS_FLAGS_SKIP_HW)
+		return false;
+
+	if (!dev->netdev_ops->ndo_setup_tc)
+		return false;
+
+	return true;
 }
 
 #endif
diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index 4398737..9874f568 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -172,6 +172,7 @@ enum {
 	TCA_U32_INDEV,
 	TCA_U32_PCNT,
 	TCA_U32_MARK,
+	TCA_U32_FLAGS,
 	__TCA_U32_MAX
 };
 
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 24e888b..563cdad 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -59,6 +59,7 @@ struct tc_u_knode {
 #ifdef CONFIG_CLS_U32_PERF
 	struct tc_u32_pcnt __percpu *pf;
 #endif
+	u32			flags;
 #ifdef CONFIG_CLS_U32_MARK
 	u32			val;
 	u32			mask;
@@ -434,7 +435,7 @@ static void u32_remove_hw_knode(struct tcf_proto *tp, u32 handle)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (tc_should_offload(dev)) {
+	if (tc_should_offload(dev, 0)) {
 		offload.cls_u32->command = TC_CLSU32_DELETE_KNODE;
 		offload.cls_u32->knode.handle = handle;
 		dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle,
@@ -442,7 +443,9 @@ static void u32_remove_hw_knode(struct tcf_proto *tp, u32 handle)
 	}
 }
 
-static void u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
+static void u32_replace_hw_hnode(struct tcf_proto *tp,
+				 struct tc_u_hnode *h,
+				 u32 flags)
 {
 	struct net_device *dev = tp->q->dev_queue->dev;
 	struct tc_cls_u32_offload u32_offload = {0};
@@ -451,7 +454,7 @@ static void u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (tc_should_offload(dev)) {
+	if (tc_should_offload(dev, flags)) {
 		offload.cls_u32->command = TC_CLSU32_NEW_HNODE;
 		offload.cls_u32->hnode.divisor = h->divisor;
 		offload.cls_u32->hnode.handle = h->handle;
@@ -471,7 +474,7 @@ static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (tc_should_offload(dev)) {
+	if (tc_should_offload(dev, 0)) {
 		offload.cls_u32->command = TC_CLSU32_DELETE_HNODE;
 		offload.cls_u32->hnode.divisor = h->divisor;
 		offload.cls_u32->hnode.handle = h->handle;
@@ -482,7 +485,9 @@ static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h)
 	}
 }
 
-static void u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n)
+static void u32_replace_hw_knode(struct tcf_proto *tp,
+				 struct tc_u_knode *n,
+				 u32 flags)
 {
 	struct net_device *dev = tp->q->dev_queue->dev;
 	struct tc_cls_u32_offload u32_offload = {0};
@@ -491,7 +496,7 @@ static void u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n)
 	offload.type = TC_SETUP_CLSU32;
 	offload.cls_u32 = &u32_offload;
 
-	if (tc_should_offload(dev)) {
+	if (tc_should_offload(dev, flags)) {
 		offload.cls_u32->command = TC_CLSU32_REPLACE_KNODE;
 		offload.cls_u32->knode.handle = n->handle;
 		offload.cls_u32->knode.fshift = n->fshift;
@@ -679,6 +684,7 @@ static const struct nla_policy u32_policy[TCA_U32_MAX + 1] = {
 	[TCA_U32_SEL]		= { .len = sizeof(struct tc_u32_sel) },
 	[TCA_U32_INDEV]		= { .type = NLA_STRING, .len = IFNAMSIZ },
 	[TCA_U32_MARK]		= { .len = sizeof(struct tc_u32_mark) },
+	[TCA_U32_FLAGS]		= { .type = NLA_U32 },
 };
 
 static int u32_set_parms(struct net *net, struct tcf_proto *tp,
@@ -786,6 +792,7 @@ static struct tc_u_knode *u32_init_knode(struct tcf_proto *tp,
 #endif
 	new->fshift = n->fshift;
 	new->res = n->res;
+	new->flags = n->flags;
 	RCU_INIT_POINTER(new->ht_down, n->ht_down);
 
 	/* bump reference count as long as we hold pointer to structure */
@@ -825,7 +832,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 	struct tc_u32_sel *s;
 	struct nlattr *opt = tca[TCA_OPTIONS];
 	struct nlattr *tb[TCA_U32_MAX + 1];
-	u32 htid;
+	u32 htid, flags = 0;
 	int err;
 #ifdef CONFIG_CLS_U32_PERF
 	size_t size;
@@ -838,6 +845,9 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 	if (err < 0)
 		return err;
 
+	if (tb[TCA_U32_FLAGS])
+		flags = nla_get_u32(tb[TCA_U32_FLAGS]);
+
 	n = (struct tc_u_knode *)*arg;
 	if (n) {
 		struct tc_u_knode *new;
@@ -845,6 +855,9 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 		if (TC_U32_KEY(n->handle) == 0)
 			return -EINVAL;
 
+		if (n->flags != flags)
+			return -EINVAL;
+
 		new = u32_init_knode(tp, n);
 		if (!new)
 			return -ENOMEM;
@@ -861,7 +874,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 		u32_replace_knode(tp, tp_c, new);
 		tcf_unbind_filter(tp, &n->res);
 		call_rcu(&n->rcu, u32_delete_key_rcu);
-		u32_replace_hw_knode(tp, new);
+		u32_replace_hw_knode(tp, new, flags);
 		return 0;
 	}
 
@@ -889,7 +902,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 		rcu_assign_pointer(tp_c->hlist, ht);
 		*arg = (unsigned long)ht;
 
-		u32_replace_hw_hnode(tp, ht);
+		u32_replace_hw_hnode(tp, ht, flags);
 		return 0;
 	}
 
@@ -940,6 +953,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 	RCU_INIT_POINTER(n->ht_up, ht);
 	n->handle = handle;
 	n->fshift = s->hmask ? ffs(ntohl(s->hmask)) - 1 : 0;
+	n->flags = flags;
 	tcf_exts_init(&n->exts, TCA_U32_ACT, TCA_U32_POLICE);
 	n->tp = tp;
 
@@ -972,7 +986,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb,
 
 		RCU_INIT_POINTER(n->next, pins);
 		rcu_assign_pointer(*ins, n);
-		u32_replace_hw_knode(tp, n);
+		u32_replace_hw_knode(tp, n, flags);
 		*arg = (unsigned long)n;
 		return 0;
 	}
@@ -1077,6 +1091,9 @@ static int u32_dump(struct net *net, struct tcf_proto *tp, unsigned long fh,
 		    nla_put_u32(skb, TCA_U32_LINK, ht_down->handle))
 			goto nla_put_failure;
 
+		if (n->flags && nla_put_u32(skb, TCA_U32_FLAGS, n->flags))
+			goto nla_put_failure;
+
 #ifdef CONFIG_CLS_U32_MARK
 		if ((n->val || n->mask)) {
 			struct tc_u32_mark mark = {.val = n->val,
-- 
cgit v0.10.2


From bfcd3a46617209454cfc0947ab093e37fd1e84ef Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:23 +0100
Subject: Introduce devlink infrastructure

Introduce devlink infrastructure for drivers to register and expose to
userspace via generic Netlink interface.

There are two basic objects defined:
devlink - one instance for every "parent device", for example switch ASIC
devlink port - one instance for every physical port of the device.

This initial portion implements basic get/dump of objects to userspace.
Also, port splitter and port type setting is implemented.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/MAINTAINERS b/MAINTAINERS
index 12b764f..e456827 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3499,6 +3499,14 @@ F:	include/linux/device-mapper.h
 F:	include/linux/dm-*.h
 F:	include/uapi/linux/dm-*.h
 
+DEVLINK
+M:	Jiri Pirko <jiri@mellanox.com>
+L:	netdev@vger.kernel.org
+S:	Supported
+F:	net/core/devlink.c
+F:	include/net/devlink.h
+F:	include/uapi/linux/devlink.h
+
 DIALOG SEMICONDUCTOR DRIVERS
 M:	Support Opensource <support.opensource@diasemi.com>
 W:	http://www.dialog-semiconductor.com/products
diff --git a/include/net/devlink.h b/include/net/devlink.h
new file mode 100644
index 0000000..c37d257
--- /dev/null
+++ b/include/net/devlink.h
@@ -0,0 +1,140 @@
+/*
+ * include/net/devlink.h - Network physical device Netlink interface
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#ifndef _NET_DEVLINK_H_
+#define _NET_DEVLINK_H_
+
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/gfp.h>
+#include <linux/list.h>
+#include <linux/netdevice.h>
+#include <net/net_namespace.h>
+#include <uapi/linux/devlink.h>
+
+struct devlink_ops;
+
+struct devlink {
+	struct list_head list;
+	struct list_head port_list;
+	const struct devlink_ops *ops;
+	struct device *dev;
+	possible_net_t _net;
+	char priv[0] __aligned(NETDEV_ALIGN);
+};
+
+struct devlink_port {
+	struct list_head list;
+	struct devlink *devlink;
+	unsigned index;
+	bool registered;
+	enum devlink_port_type type;
+	enum devlink_port_type desired_type;
+	void *type_dev;
+	bool split;
+	u32 split_group;
+};
+
+struct devlink_ops {
+	size_t priv_size;
+	int (*port_type_set)(struct devlink_port *devlink_port,
+			     enum devlink_port_type port_type);
+	int (*port_split)(struct devlink *devlink, unsigned int port_index,
+			  unsigned int count);
+	int (*port_unsplit)(struct devlink *devlink, unsigned int port_index);
+};
+
+static inline void *devlink_priv(struct devlink *devlink)
+{
+	BUG_ON(!devlink);
+	return &devlink->priv;
+}
+
+static inline struct devlink *priv_to_devlink(void *priv)
+{
+	BUG_ON(!priv);
+	return container_of(priv, struct devlink, priv);
+}
+
+struct ib_device;
+
+#if IS_ENABLED(CONFIG_NET_DEVLINK)
+
+struct devlink *devlink_alloc(const struct devlink_ops *ops, size_t priv_size);
+int devlink_register(struct devlink *devlink, struct device *dev);
+void devlink_unregister(struct devlink *devlink);
+void devlink_free(struct devlink *devlink);
+int devlink_port_register(struct devlink *devlink,
+			  struct devlink_port *devlink_port,
+			  unsigned int port_index);
+void devlink_port_unregister(struct devlink_port *devlink_port);
+void devlink_port_type_eth_set(struct devlink_port *devlink_port,
+			       struct net_device *netdev);
+void devlink_port_type_ib_set(struct devlink_port *devlink_port,
+			      struct ib_device *ibdev);
+void devlink_port_type_clear(struct devlink_port *devlink_port);
+void devlink_port_split_set(struct devlink_port *devlink_port,
+			    u32 split_group);
+
+#else
+
+static inline struct devlink *devlink_alloc(const struct devlink_ops *ops,
+					    size_t priv_size)
+{
+	return kzalloc(sizeof(struct devlink) + priv_size, GFP_KERNEL);
+}
+
+static inline int devlink_register(struct devlink *devlink, struct device *dev)
+{
+	return 0;
+}
+
+static inline void devlink_unregister(struct devlink *devlink)
+{
+}
+
+static inline void devlink_free(struct devlink *devlink)
+{
+	kfree(devlink);
+}
+
+static inline int devlink_port_register(struct devlink *devlink,
+					struct devlink_port *devlink_port,
+					unsigned int port_index)
+{
+	return 0;
+}
+
+static inline void devlink_port_unregister(struct devlink_port *devlink_port)
+{
+}
+
+static inline void devlink_port_type_eth_set(struct devlink_port *devlink_port,
+					     struct net_device *netdev)
+{
+}
+
+static inline void devlink_port_type_ib_set(struct devlink_port *devlink_port,
+					    struct ib_device *ibdev)
+{
+}
+
+static inline void devlink_port_type_clear(struct devlink_port *devlink_port)
+{
+}
+
+static inline void devlink_port_split_set(struct devlink_port *devlink_port,
+					  u32 split_group)
+{
+}
+
+#endif
+
+#endif /* _NET_DEVLINK_H_ */
diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h
new file mode 100644
index 0000000..c9fee57
--- /dev/null
+++ b/include/uapi/linux/devlink.h
@@ -0,0 +1,72 @@
+/*
+ * include/uapi/linux/devlink.h - Network physical device Netlink interface
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _UAPI_LINUX_DEVLINK_H_
+#define _UAPI_LINUX_DEVLINK_H_
+
+#define DEVLINK_GENL_NAME "devlink"
+#define DEVLINK_GENL_VERSION 0x1
+#define DEVLINK_GENL_MCGRP_CONFIG_NAME "config"
+
+enum devlink_command {
+	/* don't change the order or add anything between, this is ABI! */
+	DEVLINK_CMD_UNSPEC,
+
+	DEVLINK_CMD_GET,		/* can dump */
+	DEVLINK_CMD_SET,
+	DEVLINK_CMD_NEW,
+	DEVLINK_CMD_DEL,
+
+	DEVLINK_CMD_PORT_GET,		/* can dump */
+	DEVLINK_CMD_PORT_SET,
+	DEVLINK_CMD_PORT_NEW,
+	DEVLINK_CMD_PORT_DEL,
+
+	DEVLINK_CMD_PORT_SPLIT,
+	DEVLINK_CMD_PORT_UNSPLIT,
+
+	/* add new commands above here */
+
+	__DEVLINK_CMD_MAX,
+	DEVLINK_CMD_MAX = __DEVLINK_CMD_MAX - 1
+};
+
+enum devlink_port_type {
+	DEVLINK_PORT_TYPE_NOTSET,
+	DEVLINK_PORT_TYPE_AUTO,
+	DEVLINK_PORT_TYPE_ETH,
+	DEVLINK_PORT_TYPE_IB,
+};
+
+enum devlink_attr {
+	/* don't change the order or add anything between, this is ABI! */
+	DEVLINK_ATTR_UNSPEC,
+
+	/* bus name + dev name together are a handle for devlink entity */
+	DEVLINK_ATTR_BUS_NAME,			/* string */
+	DEVLINK_ATTR_DEV_NAME,			/* string */
+
+	DEVLINK_ATTR_PORT_INDEX,		/* u32 */
+	DEVLINK_ATTR_PORT_TYPE,			/* u16 */
+	DEVLINK_ATTR_PORT_DESIRED_TYPE,		/* u16 */
+	DEVLINK_ATTR_PORT_NETDEV_IFINDEX,	/* u32 */
+	DEVLINK_ATTR_PORT_NETDEV_NAME,		/* string */
+	DEVLINK_ATTR_PORT_IBDEV_NAME,		/* string */
+	DEVLINK_ATTR_PORT_SPLIT_COUNT,		/* u32 */
+	DEVLINK_ATTR_PORT_SPLIT_GROUP,		/* u32 */
+
+	/* add new attributes above here, update the policy in devlink.c */
+
+	__DEVLINK_ATTR_MAX,
+	DEVLINK_ATTR_MAX = __DEVLINK_ATTR_MAX - 1
+};
+
+#endif /* _UAPI_LINUX_DEVLINK_H_ */
diff --git a/net/Kconfig b/net/Kconfig
index b80efec..6c9cfb0 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -396,6 +396,13 @@ config DST_CACHE
 	bool "dst cache"
 	default n
 
+config NET_DEVLINK
+	tristate "Network physical/parent device Netlink interface"
+	help
+	  Network physical/parent device Netlink interface provides
+	  infrastructure to support access to physical chip-wide config and
+	  monitoring.
+
 endif   # if NET
 
 # Used by archs to tell that they support BPF_JIT
diff --git a/net/core/Makefile b/net/core/Makefile
index 7a8fb8a..014422e 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -25,3 +25,4 @@ obj-$(CONFIG_CGROUP_NET_PRIO) += netprio_cgroup.o
 obj-$(CONFIG_CGROUP_NET_CLASSID) += netclassid_cgroup.o
 obj-$(CONFIG_LWTUNNEL) += lwtunnel.o
 obj-$(CONFIG_DST_CACHE) += dst_cache.o
+obj-$(CONFIG_NET_DEVLINK) += devlink.o
diff --git a/net/core/devlink.c b/net/core/devlink.c
new file mode 100644
index 0000000..590fa56
--- /dev/null
+++ b/net/core/devlink.c
@@ -0,0 +1,738 @@
+/*
+ * net/core/devlink.c - Network physical/parent device Netlink interface
+ *
+ * Heavily inspired by net/wireless/
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/gfp.h>
+#include <linux/device.h>
+#include <linux/list.h>
+#include <linux/netdevice.h>
+#include <rdma/ib_verbs.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+#include <net/rtnetlink.h>
+#include <net/net_namespace.h>
+#include <net/sock.h>
+#include <net/devlink.h>
+
+static LIST_HEAD(devlink_list);
+
+/* devlink_mutex
+ *
+ * An overall lock guarding every operation coming from userspace.
+ * It also guards devlink devices list and it is taken when
+ * driver registers/unregisters it.
+ */
+static DEFINE_MUTEX(devlink_mutex);
+
+/* devlink_port_mutex
+ *
+ * Shared lock to guard lists of ports in all devlink devices.
+ */
+static DEFINE_MUTEX(devlink_port_mutex);
+
+static struct net *devlink_net(const struct devlink *devlink)
+{
+	return read_pnet(&devlink->_net);
+}
+
+static void devlink_net_set(struct devlink *devlink, struct net *net)
+{
+	write_pnet(&devlink->_net, net);
+}
+
+static struct devlink *devlink_get_from_attrs(struct net *net,
+					      struct nlattr **attrs)
+{
+	struct devlink *devlink;
+	char *busname;
+	char *devname;
+
+	if (!attrs[DEVLINK_ATTR_BUS_NAME] || !attrs[DEVLINK_ATTR_DEV_NAME])
+		return ERR_PTR(-EINVAL);
+
+	busname = nla_data(attrs[DEVLINK_ATTR_BUS_NAME]);
+	devname = nla_data(attrs[DEVLINK_ATTR_DEV_NAME]);
+
+	list_for_each_entry(devlink, &devlink_list, list) {
+		if (strcmp(devlink->dev->bus->name, busname) == 0 &&
+		    strcmp(dev_name(devlink->dev), devname) == 0 &&
+		    net_eq(devlink_net(devlink), net))
+			return devlink;
+	}
+
+	return ERR_PTR(-ENODEV);
+}
+
+static struct devlink *devlink_get_from_info(struct genl_info *info)
+{
+	return devlink_get_from_attrs(genl_info_net(info), info->attrs);
+}
+
+static struct devlink_port *devlink_port_get_by_index(struct devlink *devlink,
+						      int port_index)
+{
+	struct devlink_port *devlink_port;
+
+	list_for_each_entry(devlink_port, &devlink->port_list, list) {
+		if (devlink_port->index == port_index)
+			return devlink_port;
+	}
+	return NULL;
+}
+
+static bool devlink_port_index_exists(struct devlink *devlink, int port_index)
+{
+	return devlink_port_get_by_index(devlink, port_index);
+}
+
+static struct devlink_port *devlink_port_get_from_attrs(struct devlink *devlink,
+							struct nlattr **attrs)
+{
+	if (attrs[DEVLINK_ATTR_PORT_INDEX]) {
+		u32 port_index = nla_get_u32(attrs[DEVLINK_ATTR_PORT_INDEX]);
+		struct devlink_port *devlink_port;
+
+		devlink_port = devlink_port_get_by_index(devlink, port_index);
+		if (!devlink_port)
+			return ERR_PTR(-ENODEV);
+		return devlink_port;
+	}
+	return ERR_PTR(-EINVAL);
+}
+
+static struct devlink_port *devlink_port_get_from_info(struct devlink *devlink,
+						       struct genl_info *info)
+{
+	return devlink_port_get_from_attrs(devlink, info->attrs);
+}
+
+#define DEVLINK_NL_FLAG_NEED_PORT	BIT(0)
+
+static int devlink_nl_pre_doit(const struct genl_ops *ops,
+			       struct sk_buff *skb, struct genl_info *info)
+{
+	struct devlink *devlink;
+
+	mutex_lock(&devlink_mutex);
+	devlink = devlink_get_from_info(info);
+	if (IS_ERR(devlink)) {
+		mutex_unlock(&devlink_mutex);
+		return PTR_ERR(devlink);
+	}
+	info->user_ptr[0] = devlink;
+	if (ops->internal_flags & DEVLINK_NL_FLAG_NEED_PORT) {
+		struct devlink_port *devlink_port;
+
+		mutex_lock(&devlink_port_mutex);
+		devlink_port = devlink_port_get_from_info(devlink, info);
+		if (IS_ERR(devlink_port)) {
+			mutex_unlock(&devlink_port_mutex);
+			mutex_unlock(&devlink_mutex);
+			return PTR_ERR(devlink_port);
+		}
+		info->user_ptr[1] = devlink_port;
+	}
+	return 0;
+}
+
+static void devlink_nl_post_doit(const struct genl_ops *ops,
+				 struct sk_buff *skb, struct genl_info *info)
+{
+	if (ops->internal_flags & DEVLINK_NL_FLAG_NEED_PORT)
+		mutex_unlock(&devlink_port_mutex);
+	mutex_unlock(&devlink_mutex);
+}
+
+static struct genl_family devlink_nl_family = {
+	.id		= GENL_ID_GENERATE,
+	.name		= DEVLINK_GENL_NAME,
+	.version	= DEVLINK_GENL_VERSION,
+	.maxattr	= DEVLINK_ATTR_MAX,
+	.netnsok	= true,
+	.pre_doit	= devlink_nl_pre_doit,
+	.post_doit	= devlink_nl_post_doit,
+};
+
+enum devlink_multicast_groups {
+	DEVLINK_MCGRP_CONFIG,
+};
+
+static const struct genl_multicast_group devlink_nl_mcgrps[] = {
+	[DEVLINK_MCGRP_CONFIG] = { .name = DEVLINK_GENL_MCGRP_CONFIG_NAME },
+};
+
+static int devlink_nl_put_handle(struct sk_buff *msg, struct devlink *devlink)
+{
+	if (nla_put_string(msg, DEVLINK_ATTR_BUS_NAME, devlink->dev->bus->name))
+		return -EMSGSIZE;
+	if (nla_put_string(msg, DEVLINK_ATTR_DEV_NAME, dev_name(devlink->dev)))
+		return -EMSGSIZE;
+	return 0;
+}
+
+static int devlink_nl_fill(struct sk_buff *msg, struct devlink *devlink,
+			   enum devlink_command cmd, u32 portid,
+			   u32 seq, int flags)
+{
+	void *hdr;
+
+	hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	if (devlink_nl_put_handle(msg, devlink))
+		goto nla_put_failure;
+
+	genlmsg_end(msg, hdr);
+	return 0;
+
+nla_put_failure:
+	genlmsg_cancel(msg, hdr);
+	return -EMSGSIZE;
+}
+
+static void devlink_notify(struct devlink *devlink, enum devlink_command cmd)
+{
+	struct sk_buff *msg;
+	int err;
+
+	WARN_ON(cmd != DEVLINK_CMD_NEW && cmd != DEVLINK_CMD_DEL);
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return;
+
+	err = devlink_nl_fill(msg, devlink, cmd, 0, 0, 0);
+	if (err) {
+		nlmsg_free(msg);
+		return;
+	}
+
+	genlmsg_multicast_netns(&devlink_nl_family, devlink_net(devlink),
+				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
+}
+
+static int devlink_nl_port_fill(struct sk_buff *msg, struct devlink *devlink,
+				struct devlink_port *devlink_port,
+				enum devlink_command cmd, u32 portid,
+				u32 seq, int flags)
+{
+	void *hdr;
+
+	hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	if (devlink_nl_put_handle(msg, devlink))
+		goto nla_put_failure;
+	if (nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, devlink_port->index))
+		goto nla_put_failure;
+	if (nla_put_u16(msg, DEVLINK_ATTR_PORT_TYPE, devlink_port->type))
+		goto nla_put_failure;
+	if (devlink_port->desired_type != DEVLINK_PORT_TYPE_NOTSET &&
+	    nla_put_u16(msg, DEVLINK_ATTR_PORT_DESIRED_TYPE,
+			devlink_port->desired_type))
+		goto nla_put_failure;
+	if (devlink_port->type == DEVLINK_PORT_TYPE_ETH) {
+		struct net_device *netdev = devlink_port->type_dev;
+
+		if (netdev &&
+		    (nla_put_u32(msg, DEVLINK_ATTR_PORT_NETDEV_IFINDEX,
+				 netdev->ifindex) ||
+		     nla_put_string(msg, DEVLINK_ATTR_PORT_NETDEV_NAME,
+				    netdev->name)))
+			goto nla_put_failure;
+	}
+	if (devlink_port->type == DEVLINK_PORT_TYPE_IB) {
+		struct ib_device *ibdev = devlink_port->type_dev;
+
+		if (ibdev &&
+		    nla_put_string(msg, DEVLINK_ATTR_PORT_IBDEV_NAME,
+				   ibdev->name))
+			goto nla_put_failure;
+	}
+	if (devlink_port->split &&
+	    nla_put_u32(msg, DEVLINK_ATTR_PORT_SPLIT_GROUP,
+			devlink_port->split_group))
+		goto nla_put_failure;
+
+	genlmsg_end(msg, hdr);
+	return 0;
+
+nla_put_failure:
+	genlmsg_cancel(msg, hdr);
+	return -EMSGSIZE;
+}
+
+static void devlink_port_notify(struct devlink_port *devlink_port,
+				enum devlink_command cmd)
+{
+	struct devlink *devlink = devlink_port->devlink;
+	struct sk_buff *msg;
+	int err;
+
+	if (!devlink_port->registered)
+		return;
+
+	WARN_ON(cmd != DEVLINK_CMD_PORT_NEW && cmd != DEVLINK_CMD_PORT_DEL);
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return;
+
+	err = devlink_nl_port_fill(msg, devlink, devlink_port, cmd, 0, 0, 0);
+	if (err) {
+		nlmsg_free(msg);
+		return;
+	}
+
+	genlmsg_multicast_netns(&devlink_nl_family, devlink_net(devlink),
+				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
+}
+
+static int devlink_nl_cmd_get_doit(struct sk_buff *skb, struct genl_info *info)
+{
+	struct devlink *devlink = info->user_ptr[0];
+	struct sk_buff *msg;
+	int err;
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return -ENOMEM;
+
+	err = devlink_nl_fill(msg, devlink, DEVLINK_CMD_NEW,
+			      info->snd_portid, info->snd_seq, 0);
+	if (err) {
+		nlmsg_free(msg);
+		return err;
+	}
+
+	return genlmsg_reply(msg, info);
+}
+
+static int devlink_nl_cmd_get_dumpit(struct sk_buff *msg,
+				     struct netlink_callback *cb)
+{
+	struct devlink *devlink;
+	int start = cb->args[0];
+	int idx = 0;
+	int err;
+
+	mutex_lock(&devlink_mutex);
+	list_for_each_entry(devlink, &devlink_list, list) {
+		if (!net_eq(devlink_net(devlink), sock_net(msg->sk)))
+			continue;
+		if (idx < start) {
+			idx++;
+			continue;
+		}
+		err = devlink_nl_fill(msg, devlink, DEVLINK_CMD_NEW,
+				      NETLINK_CB(cb->skb).portid,
+				      cb->nlh->nlmsg_seq, NLM_F_MULTI);
+		if (err)
+			goto out;
+		idx++;
+	}
+out:
+	mutex_unlock(&devlink_mutex);
+
+	cb->args[0] = idx;
+	return msg->len;
+}
+
+static int devlink_nl_cmd_port_get_doit(struct sk_buff *skb,
+					struct genl_info *info)
+{
+	struct devlink *devlink = info->user_ptr[0];
+	struct devlink_port *devlink_port = info->user_ptr[1];
+	struct sk_buff *msg;
+	int err;
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return -ENOMEM;
+
+	err = devlink_nl_port_fill(msg, devlink, devlink_port,
+				   DEVLINK_CMD_PORT_NEW,
+				   info->snd_portid, info->snd_seq, 0);
+	if (err) {
+		nlmsg_free(msg);
+		return err;
+	}
+
+	return genlmsg_reply(msg, info);
+}
+
+static int devlink_nl_cmd_port_get_dumpit(struct sk_buff *msg,
+					  struct netlink_callback *cb)
+{
+	struct devlink *devlink;
+	struct devlink_port *devlink_port;
+	int start = cb->args[0];
+	int idx = 0;
+	int err;
+
+	mutex_lock(&devlink_mutex);
+	mutex_lock(&devlink_port_mutex);
+	list_for_each_entry(devlink, &devlink_list, list) {
+		if (!net_eq(devlink_net(devlink), sock_net(msg->sk)))
+			continue;
+		list_for_each_entry(devlink_port, &devlink->port_list, list) {
+			if (idx < start) {
+				idx++;
+				continue;
+			}
+			err = devlink_nl_port_fill(msg, devlink, devlink_port,
+						   DEVLINK_CMD_NEW,
+						   NETLINK_CB(cb->skb).portid,
+						   cb->nlh->nlmsg_seq,
+						   NLM_F_MULTI);
+			if (err)
+				goto out;
+			idx++;
+		}
+	}
+out:
+	mutex_unlock(&devlink_port_mutex);
+	mutex_unlock(&devlink_mutex);
+
+	cb->args[0] = idx;
+	return msg->len;
+}
+
+static int devlink_port_type_set(struct devlink *devlink,
+				 struct devlink_port *devlink_port,
+				 enum devlink_port_type port_type)
+
+{
+	int err;
+
+	if (devlink->ops && devlink->ops->port_type_set) {
+		if (port_type == DEVLINK_PORT_TYPE_NOTSET)
+			return -EINVAL;
+		err = devlink->ops->port_type_set(devlink_port, port_type);
+		if (err)
+			return err;
+		devlink_port->desired_type = port_type;
+		devlink_port_notify(devlink_port, DEVLINK_CMD_PORT_NEW);
+		return 0;
+	}
+	return -EOPNOTSUPP;
+}
+
+static int devlink_nl_cmd_port_set_doit(struct sk_buff *skb,
+					struct genl_info *info)
+{
+	struct devlink *devlink = info->user_ptr[0];
+	struct devlink_port *devlink_port = info->user_ptr[1];
+	int err;
+
+	if (info->attrs[DEVLINK_ATTR_PORT_TYPE]) {
+		enum devlink_port_type port_type;
+
+		port_type = nla_get_u16(info->attrs[DEVLINK_ATTR_PORT_TYPE]);
+		err = devlink_port_type_set(devlink, devlink_port, port_type);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static int devlink_port_split(struct devlink *devlink,
+			      u32 port_index, u32 count)
+
+{
+	if (devlink->ops && devlink->ops->port_split)
+		return devlink->ops->port_split(devlink, port_index, count);
+	return -EOPNOTSUPP;
+}
+
+static int devlink_nl_cmd_port_split_doit(struct sk_buff *skb,
+					  struct genl_info *info)
+{
+	struct devlink *devlink = info->user_ptr[0];
+	u32 port_index;
+	u32 count;
+
+	if (!info->attrs[DEVLINK_ATTR_PORT_INDEX] ||
+	    !info->attrs[DEVLINK_ATTR_PORT_SPLIT_COUNT])
+		return -EINVAL;
+
+	port_index = nla_get_u32(info->attrs[DEVLINK_ATTR_PORT_INDEX]);
+	count = nla_get_u32(info->attrs[DEVLINK_ATTR_PORT_SPLIT_COUNT]);
+	return devlink_port_split(devlink, port_index, count);
+}
+
+static int devlink_port_unsplit(struct devlink *devlink, u32 port_index)
+
+{
+	if (devlink->ops && devlink->ops->port_unsplit)
+		return devlink->ops->port_unsplit(devlink, port_index);
+	return -EOPNOTSUPP;
+}
+
+static int devlink_nl_cmd_port_unsplit_doit(struct sk_buff *skb,
+					    struct genl_info *info)
+{
+	struct devlink *devlink = info->user_ptr[0];
+	u32 port_index;
+
+	if (!info->attrs[DEVLINK_ATTR_PORT_INDEX])
+		return -EINVAL;
+
+	port_index = nla_get_u32(info->attrs[DEVLINK_ATTR_PORT_INDEX]);
+	return devlink_port_unsplit(devlink, port_index);
+}
+
+static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = {
+	[DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING },
+	[DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING },
+	[DEVLINK_ATTR_PORT_INDEX] = { .type = NLA_U32 },
+	[DEVLINK_ATTR_PORT_TYPE] = { .type = NLA_U16 },
+	[DEVLINK_ATTR_PORT_SPLIT_COUNT] = { .type = NLA_U32 },
+};
+
+static const struct genl_ops devlink_nl_ops[] = {
+	{
+		.cmd = DEVLINK_CMD_GET,
+		.doit = devlink_nl_cmd_get_doit,
+		.dumpit = devlink_nl_cmd_get_dumpit,
+		.policy = devlink_nl_policy,
+		/* can be retrieved by unprivileged users */
+	},
+	{
+		.cmd = DEVLINK_CMD_PORT_GET,
+		.doit = devlink_nl_cmd_port_get_doit,
+		.dumpit = devlink_nl_cmd_port_get_dumpit,
+		.policy = devlink_nl_policy,
+		.internal_flags = DEVLINK_NL_FLAG_NEED_PORT,
+		/* can be retrieved by unprivileged users */
+	},
+	{
+		.cmd = DEVLINK_CMD_PORT_SET,
+		.doit = devlink_nl_cmd_port_set_doit,
+		.policy = devlink_nl_policy,
+		.flags = GENL_ADMIN_PERM,
+		.internal_flags = DEVLINK_NL_FLAG_NEED_PORT,
+	},
+	{
+		.cmd = DEVLINK_CMD_PORT_SPLIT,
+		.doit = devlink_nl_cmd_port_split_doit,
+		.policy = devlink_nl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = DEVLINK_CMD_PORT_UNSPLIT,
+		.doit = devlink_nl_cmd_port_unsplit_doit,
+		.policy = devlink_nl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+};
+
+/**
+ *	devlink_alloc - Allocate new devlink instance resources
+ *
+ *	@ops: ops
+ *	@priv_size: size of user private data
+ *
+ *	Allocate new devlink instance resources, including devlink index
+ *	and name.
+ */
+struct devlink *devlink_alloc(const struct devlink_ops *ops, size_t priv_size)
+{
+	struct devlink *devlink;
+
+	devlink = kzalloc(sizeof(*devlink) + priv_size, GFP_KERNEL);
+	if (!devlink)
+		return NULL;
+	devlink->ops = ops;
+	devlink_net_set(devlink, &init_net);
+	INIT_LIST_HEAD(&devlink->port_list);
+	return devlink;
+}
+EXPORT_SYMBOL_GPL(devlink_alloc);
+
+/**
+ *	devlink_register - Register devlink instance
+ *
+ *	@devlink: devlink
+ */
+int devlink_register(struct devlink *devlink, struct device *dev)
+{
+	mutex_lock(&devlink_mutex);
+	devlink->dev = dev;
+	list_add_tail(&devlink->list, &devlink_list);
+	devlink_notify(devlink, DEVLINK_CMD_NEW);
+	mutex_unlock(&devlink_mutex);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(devlink_register);
+
+/**
+ *	devlink_unregister - Unregister devlink instance
+ *
+ *	@devlink: devlink
+ */
+void devlink_unregister(struct devlink *devlink)
+{
+	mutex_lock(&devlink_mutex);
+	devlink_notify(devlink, DEVLINK_CMD_DEL);
+	list_del(&devlink->list);
+	mutex_unlock(&devlink_mutex);
+}
+EXPORT_SYMBOL_GPL(devlink_unregister);
+
+/**
+ *	devlink_free - Free devlink instance resources
+ *
+ *	@devlink: devlink
+ */
+void devlink_free(struct devlink *devlink)
+{
+	kfree(devlink);
+}
+EXPORT_SYMBOL_GPL(devlink_free);
+
+/**
+ *	devlink_port_register - Register devlink port
+ *
+ *	@devlink: devlink
+ *	@devlink_port: devlink port
+ *	@port_index
+ *
+ *	Register devlink port with provided port index. User can use
+ *	any indexing, even hw-related one. devlink_port structure
+ *	is convenient to be embedded inside user driver private structure.
+ *	Note that the caller should take care of zeroing the devlink_port
+ *	structure.
+ */
+int devlink_port_register(struct devlink *devlink,
+			  struct devlink_port *devlink_port,
+			  unsigned int port_index)
+{
+	mutex_lock(&devlink_port_mutex);
+	if (devlink_port_index_exists(devlink, port_index)) {
+		mutex_unlock(&devlink_port_mutex);
+		return -EEXIST;
+	}
+	devlink_port->devlink = devlink;
+	devlink_port->index = port_index;
+	devlink_port->type = DEVLINK_PORT_TYPE_NOTSET;
+	devlink_port->registered = true;
+	list_add_tail(&devlink_port->list, &devlink->port_list);
+	mutex_unlock(&devlink_port_mutex);
+	devlink_port_notify(devlink_port, DEVLINK_CMD_PORT_NEW);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(devlink_port_register);
+
+/**
+ *	devlink_port_unregister - Unregister devlink port
+ *
+ *	@devlink_port: devlink port
+ */
+void devlink_port_unregister(struct devlink_port *devlink_port)
+{
+	devlink_port_notify(devlink_port, DEVLINK_CMD_PORT_DEL);
+	mutex_lock(&devlink_port_mutex);
+	list_del(&devlink_port->list);
+	mutex_unlock(&devlink_port_mutex);
+}
+EXPORT_SYMBOL_GPL(devlink_port_unregister);
+
+static void __devlink_port_type_set(struct devlink_port *devlink_port,
+				    enum devlink_port_type type,
+				    void *type_dev)
+{
+	devlink_port->type = type;
+	devlink_port->type_dev = type_dev;
+	devlink_port_notify(devlink_port, DEVLINK_CMD_PORT_NEW);
+}
+
+/**
+ *	devlink_port_type_eth_set - Set port type to Ethernet
+ *
+ *	@devlink_port: devlink port
+ *	@netdev: related netdevice
+ */
+void devlink_port_type_eth_set(struct devlink_port *devlink_port,
+			       struct net_device *netdev)
+{
+	return __devlink_port_type_set(devlink_port,
+				       DEVLINK_PORT_TYPE_ETH, netdev);
+}
+EXPORT_SYMBOL_GPL(devlink_port_type_eth_set);
+
+/**
+ *	devlink_port_type_ib_set - Set port type to InfiniBand
+ *
+ *	@devlink_port: devlink port
+ *	@ibdev: related IB device
+ */
+void devlink_port_type_ib_set(struct devlink_port *devlink_port,
+			      struct ib_device *ibdev)
+{
+	return __devlink_port_type_set(devlink_port,
+				       DEVLINK_PORT_TYPE_IB, ibdev);
+}
+EXPORT_SYMBOL_GPL(devlink_port_type_ib_set);
+
+/**
+ *	devlink_port_type_clear - Clear port type
+ *
+ *	@devlink_port: devlink port
+ */
+void devlink_port_type_clear(struct devlink_port *devlink_port)
+{
+	return __devlink_port_type_set(devlink_port,
+				       DEVLINK_PORT_TYPE_NOTSET, NULL);
+}
+EXPORT_SYMBOL_GPL(devlink_port_type_clear);
+
+/**
+ *	devlink_port_split_set - Set port is split
+ *
+ *	@devlink_port: devlink port
+ *	@split_group: split group - identifies group split port is part of
+ */
+void devlink_port_split_set(struct devlink_port *devlink_port,
+			    u32 split_group)
+{
+	devlink_port->split = true;
+	devlink_port->split_group = split_group;
+	devlink_port_notify(devlink_port, DEVLINK_CMD_PORT_NEW);
+}
+EXPORT_SYMBOL_GPL(devlink_port_split_set);
+
+static int __init devlink_module_init(void)
+{
+	return genl_register_family_with_ops_groups(&devlink_nl_family,
+						    devlink_nl_ops,
+						    devlink_nl_mcgrps);
+}
+
+static void __exit devlink_module_exit(void)
+{
+	genl_unregister_family(&devlink_nl_family);
+}
+
+module_init(devlink_module_init);
+module_exit(devlink_module_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Jiri Pirko <jiri@mellanox.com>");
+MODULE_DESCRIPTION("Network physical device Netlink interface");
+MODULE_ALIAS_GENL_FAMILY(DEVLINK_GENL_NAME);
-- 
cgit v0.10.2


From 09d4d087cd4869859fcc5dfc692f0830550a1b48 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:24 +0100
Subject: mlx4: Implement devlink interface

Implement newly introduced devlink interface. Add devlink port instances
for every port and set the port types accordingly.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
v2->v3:
-add dev param to devlink_register (api change)
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 1c7ab6c..a15a7b3 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -41,6 +41,7 @@
 #include <linux/if_vlan.h>
 #include <net/ipv6.h>
 #include <net/addrconf.h>
+#include <net/devlink.h>
 
 #include <rdma/ib_smi.h>
 #include <rdma/ib_user_verbs.h>
@@ -2519,6 +2520,9 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	}
 
 	ibdev->ib_active = true;
+	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB)
+		devlink_port_type_ib_set(mlx4_get_devlink_port(dev, i),
+					 &ibdev->ib_dev);
 
 	if (mlx4_is_mfunc(ibdev->dev))
 		init_pkeys(ibdev);
@@ -2643,7 +2647,10 @@ static void mlx4_ib_remove(struct mlx4_dev *dev, void *ibdev_ptr)
 {
 	struct mlx4_ib_dev *ibdev = ibdev_ptr;
 	int p;
+	int i;
 
+	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB)
+		devlink_port_type_clear(mlx4_get_devlink_port(dev, i));
 	ibdev->ib_active = false;
 	flush_workqueue(wq);
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 96d95cb..e26b110 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -40,6 +40,7 @@
 #include <net/ip.h>
 #include <net/busy_poll.h>
 #include <net/vxlan.h>
+#include <net/devlink.h>
 
 #include <linux/mlx4/driver.h>
 #include <linux/mlx4/device.h>
@@ -2033,8 +2034,11 @@ void mlx4_en_destroy_netdev(struct net_device *dev)
 	en_dbg(DRV, priv, "Destroying netdev on port:%d\n", priv->port);
 
 	/* Unregister device - this will close the port if it was up */
-	if (priv->registered)
+	if (priv->registered) {
+		devlink_port_type_clear(mlx4_get_devlink_port(mdev->dev,
+							      priv->port));
 		unregister_netdev(dev);
+	}
 
 	if (priv->allocated)
 		mlx4_free_hwq_res(mdev->dev, &priv->res, MLX4_EN_PAGE_SIZE);
@@ -3051,6 +3055,8 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	}
 
 	priv->registered = 1;
+	devlink_port_type_eth_set(mlx4_get_devlink_port(mdev->dev, priv->port),
+				  dev);
 
 	return 0;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c
index 0472941..dec77d6 100644
--- a/drivers/net/ethernet/mellanox/mlx4/intf.c
+++ b/drivers/net/ethernet/mellanox/mlx4/intf.c
@@ -34,6 +34,7 @@
 #include <linux/slab.h>
 #include <linux/export.h>
 #include <linux/errno.h>
+#include <net/devlink.h>
 
 #include "mlx4.h"
 
@@ -249,3 +250,11 @@ void *mlx4_get_protocol_dev(struct mlx4_dev *dev, enum mlx4_protocol proto, int
 	return result;
 }
 EXPORT_SYMBOL_GPL(mlx4_get_protocol_dev);
+
+struct devlink_port *mlx4_get_devlink_port(struct mlx4_dev *dev, int port)
+{
+	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+
+	return &info->devlink_port;
+}
+EXPORT_SYMBOL_GPL(mlx4_get_devlink_port);
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 2cc3c62..4f5cfe4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -42,6 +42,7 @@
 #include <linux/io-mapping.h>
 #include <linux/delay.h>
 #include <linux/kmod.h>
+#include <net/devlink.h>
 
 #include <linux/mlx4/device.h>
 #include <linux/mlx4/doorbell.h>
@@ -2881,8 +2882,13 @@ no_msi:
 
 static int mlx4_init_port_info(struct mlx4_dev *dev, int port)
 {
+	struct devlink *devlink = priv_to_devlink(mlx4_priv(dev));
 	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
-	int err = 0;
+	int err;
+
+	err = devlink_port_register(devlink, &info->devlink_port, port);
+	if (err)
+		return err;
 
 	info->dev = dev;
 	info->port = port;
@@ -2907,6 +2913,7 @@ static int mlx4_init_port_info(struct mlx4_dev *dev, int port)
 	err = device_create_file(&dev->persist->pdev->dev, &info->port_attr);
 	if (err) {
 		mlx4_err(dev, "Failed to create file for port %d\n", port);
+		devlink_port_unregister(&info->devlink_port);
 		info->port = -1;
 	}
 
@@ -3680,21 +3687,23 @@ err_disable_pdev:
 
 static int mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
+	struct devlink *devlink;
 	struct mlx4_priv *priv;
 	struct mlx4_dev *dev;
 	int ret;
 
 	printk_once(KERN_INFO "%s", mlx4_version);
 
-	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
-	if (!priv)
+	devlink = devlink_alloc(NULL, sizeof(*priv));
+	if (!devlink)
 		return -ENOMEM;
+	priv = devlink_priv(devlink);
 
 	dev       = &priv->dev;
 	dev->persist = kzalloc(sizeof(*dev->persist), GFP_KERNEL);
 	if (!dev->persist) {
-		kfree(priv);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_devlink_free;
 	}
 	dev->persist->pdev = pdev;
 	dev->persist->dev = dev;
@@ -3703,14 +3712,23 @@ static int mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	mutex_init(&dev->persist->device_state_mutex);
 	mutex_init(&dev->persist->interface_state_mutex);
 
+	ret = devlink_register(devlink, &pdev->dev);
+	if (ret)
+		goto err_persist_free;
+
 	ret =  __mlx4_init_one(pdev, id->driver_data, priv);
-	if (ret) {
-		kfree(dev->persist);
-		kfree(priv);
-	} else {
-		pci_save_state(pdev);
-	}
+	if (ret)
+		goto err_devlink_unregister;
 
+	pci_save_state(pdev);
+	return 0;
+
+err_devlink_unregister:
+	devlink_unregister(devlink);
+err_persist_free:
+	kfree(dev->persist);
+err_devlink_free:
+	devlink_free(devlink);
 	return ret;
 }
 
@@ -3811,6 +3829,7 @@ static void mlx4_remove_one(struct pci_dev *pdev)
 	struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
 	struct mlx4_dev  *dev  = persist->dev;
 	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct devlink *devlink = priv_to_devlink(priv);
 	int active_vfs = 0;
 
 	mutex_lock(&persist->interface_state_mutex);
@@ -3841,8 +3860,9 @@ static void mlx4_remove_one(struct pci_dev *pdev)
 
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
+	devlink_unregister(devlink);
 	kfree(dev->persist);
-	kfree(priv);
+	devlink_free(devlink);
 	pci_set_drvdata(pdev, NULL);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index 7baef52..ef96831 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -45,6 +45,7 @@
 #include <linux/workqueue.h>
 #include <linux/interrupt.h>
 #include <linux/spinlock.h>
+#include <net/devlink.h>
 
 #include <linux/mlx4/device.h>
 #include <linux/mlx4/driver.h>
@@ -828,6 +829,7 @@ struct mlx4_port_info {
 	struct mlx4_roce_gid_table gid_table;
 	int			base_qpn;
 	struct cpu_rmap		*rmap;
+	struct devlink_port	devlink_port;
 };
 
 struct mlx4_sense {
diff --git a/include/linux/mlx4/driver.h b/include/linux/mlx4/driver.h
index 2e8af00..bd0e707 100644
--- a/include/linux/mlx4/driver.h
+++ b/include/linux/mlx4/driver.h
@@ -33,6 +33,7 @@
 #ifndef MLX4_DRIVER_H
 #define MLX4_DRIVER_H
 
+#include <net/devlink.h>
 #include <linux/mlx4/device.h>
 
 struct mlx4_dev;
@@ -89,6 +90,8 @@ int mlx4_port_map_set(struct mlx4_dev *dev, struct mlx4_port_map *v2p);
 
 void *mlx4_get_protocol_dev(struct mlx4_dev *dev, enum mlx4_protocol proto, int port);
 
+struct devlink_port *mlx4_get_devlink_port(struct mlx4_dev *dev, int port);
+
 static inline u64 mlx4_mac_to_u64(u8 *addr)
 {
 	u64 mac = 0;
-- 
cgit v0.10.2


From b2facd95ab965465373fc5ded7218c8c4fa8da7d Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:25 +0100
Subject: mlx4: Implement port type setting via devlink interface

So far, there has been an mlx4-specific sysfs file allowing user to
change port type to either Ethernet of InfiniBand. This is very
inconvenient.

Allow to expose the same ability to set port type in a generic way
using devlink interface.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 4f5cfe4..b8a5151 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -1082,36 +1082,20 @@ static ssize_t show_port_type(struct device *dev,
 	return strlen(buf);
 }
 
-static ssize_t set_port_type(struct device *dev,
-			     struct device_attribute *attr,
-			     const char *buf, size_t count)
+static int __set_port_type(struct mlx4_port_info *info,
+			   enum mlx4_port_type port_type)
 {
-	struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
-						   port_attr);
 	struct mlx4_dev *mdev = info->dev;
 	struct mlx4_priv *priv = mlx4_priv(mdev);
 	enum mlx4_port_type types[MLX4_MAX_PORTS];
 	enum mlx4_port_type new_types[MLX4_MAX_PORTS];
-	static DEFINE_MUTEX(set_port_type_mutex);
 	int i;
 	int err = 0;
 
-	mutex_lock(&set_port_type_mutex);
-
-	if (!strcmp(buf, "ib\n"))
-		info->tmp_type = MLX4_PORT_TYPE_IB;
-	else if (!strcmp(buf, "eth\n"))
-		info->tmp_type = MLX4_PORT_TYPE_ETH;
-	else if (!strcmp(buf, "auto\n"))
-		info->tmp_type = MLX4_PORT_TYPE_AUTO;
-	else {
-		mlx4_err(mdev, "%s is not supported port type\n", buf);
-		err = -EINVAL;
-		goto err_out;
-	}
-
 	mlx4_stop_sense(mdev);
 	mutex_lock(&priv->port_mutex);
+	info->tmp_type = port_type;
+
 	/* Possible type is always the one that was delivered */
 	mdev->caps.possible_type[info->port] = info->tmp_type;
 
@@ -1153,6 +1137,37 @@ static ssize_t set_port_type(struct device *dev,
 out:
 	mlx4_start_sense(mdev);
 	mutex_unlock(&priv->port_mutex);
+
+	return err;
+}
+
+static ssize_t set_port_type(struct device *dev,
+			     struct device_attribute *attr,
+			     const char *buf, size_t count)
+{
+	struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
+						   port_attr);
+	struct mlx4_dev *mdev = info->dev;
+	enum mlx4_port_type port_type;
+	static DEFINE_MUTEX(set_port_type_mutex);
+	int err;
+
+	mutex_lock(&set_port_type_mutex);
+
+	if (!strcmp(buf, "ib\n")) {
+		port_type = MLX4_PORT_TYPE_IB;
+	} else if (!strcmp(buf, "eth\n")) {
+		port_type = MLX4_PORT_TYPE_ETH;
+	} else if (!strcmp(buf, "auto\n")) {
+		port_type = MLX4_PORT_TYPE_AUTO;
+	} else {
+		mlx4_err(mdev, "%s is not supported port type\n", buf);
+		err = -EINVAL;
+		goto err_out;
+	}
+
+	err = __set_port_type(info, port_type);
+
 err_out:
 	mutex_unlock(&set_port_type_mutex);
 
@@ -3685,6 +3700,35 @@ err_disable_pdev:
 	return err;
 }
 
+static int mlx4_devlink_port_type_set(struct devlink_port *devlink_port,
+				      enum devlink_port_type port_type)
+{
+	struct mlx4_port_info *info = container_of(devlink_port,
+						   struct mlx4_port_info,
+						   devlink_port);
+	enum mlx4_port_type mlx4_port_type;
+
+	switch (port_type) {
+	case DEVLINK_PORT_TYPE_AUTO:
+		mlx4_port_type = MLX4_PORT_TYPE_AUTO;
+		break;
+	case DEVLINK_PORT_TYPE_ETH:
+		mlx4_port_type = MLX4_PORT_TYPE_ETH;
+		break;
+	case DEVLINK_PORT_TYPE_IB:
+		mlx4_port_type = MLX4_PORT_TYPE_IB;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return __set_port_type(info, mlx4_port_type);
+}
+
+static const struct devlink_ops mlx4_devlink_ops = {
+	.port_type_set	= mlx4_devlink_port_type_set,
+};
+
 static int mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct devlink *devlink;
@@ -3694,7 +3738,7 @@ static int mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	printk_once(KERN_INFO "%s", mlx4_version);
 
-	devlink = devlink_alloc(NULL, sizeof(*priv));
+	devlink = devlink_alloc(&mlx4_devlink_ops, sizeof(*priv));
 	if (!devlink)
 		return -ENOMEM;
 	priv = devlink_priv(devlink);
-- 
cgit v0.10.2


From c4745500e988b393189703bade6c3ec65194e24c Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:26 +0100
Subject: mlxsw: Implement devlink interface

Implement newly introduced devlink interface. Add devlink port instances
for every port and set the port types accordingly.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c
index 22379eb..a4026a4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.c
@@ -56,6 +56,7 @@
 #include <linux/rcupdate.h>
 #include <linux/slab.h>
 #include <asm/byteorder.h>
+#include <net/devlink.h>
 
 #include "core.h"
 #include "item.h"
@@ -791,6 +792,7 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 	const char *device_kind = mlxsw_bus_info->device_kind;
 	struct mlxsw_core *mlxsw_core;
 	struct mlxsw_driver *mlxsw_driver;
+	struct devlink *devlink;
 	size_t alloc_size;
 	int err;
 
@@ -798,12 +800,13 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 	if (!mlxsw_driver)
 		return -EINVAL;
 	alloc_size = sizeof(*mlxsw_core) + mlxsw_driver->priv_size;
-	mlxsw_core = kzalloc(alloc_size, GFP_KERNEL);
-	if (!mlxsw_core) {
+	devlink = devlink_alloc(NULL, alloc_size);
+	if (!devlink) {
 		err = -ENOMEM;
-		goto err_core_alloc;
+		goto err_devlink_alloc;
 	}
 
+	mlxsw_core = devlink_priv(devlink);
 	INIT_LIST_HEAD(&mlxsw_core->rx_listener_list);
 	INIT_LIST_HEAD(&mlxsw_core->event_listener_list);
 	mlxsw_core->driver = mlxsw_driver;
@@ -841,6 +844,10 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 	if (err)
 		goto err_hwmon_init;
 
+	err = devlink_register(devlink, mlxsw_bus_info->dev);
+	if (err)
+		goto err_devlink_register;
+
 	err = mlxsw_driver->init(mlxsw_core->driver_priv, mlxsw_core,
 				 mlxsw_bus_info);
 	if (err)
@@ -855,6 +862,8 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 err_debugfs_init:
 	mlxsw_core->driver->fini(mlxsw_core->driver_priv);
 err_driver_init:
+	devlink_unregister(devlink);
+err_devlink_register:
 err_hwmon_init:
 	mlxsw_emad_fini(mlxsw_core);
 err_emad_init:
@@ -864,8 +873,8 @@ err_bus_init:
 err_alloc_lag_mapping:
 	free_percpu(mlxsw_core->pcpu_stats);
 err_alloc_stats:
-	kfree(mlxsw_core);
-err_core_alloc:
+	devlink_free(devlink);
+err_devlink_alloc:
 	mlxsw_core_driver_put(device_kind);
 	return err;
 }
@@ -874,14 +883,16 @@ EXPORT_SYMBOL(mlxsw_core_bus_device_register);
 void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core)
 {
 	const char *device_kind = mlxsw_core->bus_info->device_kind;
+	struct devlink *devlink = priv_to_devlink(mlxsw_core);
 
 	mlxsw_core_debugfs_fini(mlxsw_core);
 	mlxsw_core->driver->fini(mlxsw_core->driver_priv);
+	devlink_unregister(devlink);
 	mlxsw_emad_fini(mlxsw_core);
 	mlxsw_core->bus->fini(mlxsw_core->bus_priv);
 	kfree(mlxsw_core->lag.mapping);
 	free_percpu(mlxsw_core->pcpu_stats);
-	kfree(mlxsw_core);
+	devlink_free(devlink);
 	mlxsw_core_driver_put(device_kind);
 }
 EXPORT_SYMBOL(mlxsw_core_bus_device_unregister);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 09ce451..20e6783 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -49,6 +49,7 @@
 #include <linux/jiffies.h>
 #include <linux/bitops.h>
 #include <linux/list.h>
+#include <net/devlink.h>
 #include <net/switchdev.h>
 #include <generated/utsrelease.h>
 
@@ -1351,7 +1352,9 @@ static const struct ethtool_ops mlxsw_sp_port_ethtool_ops = {
 
 static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 {
+	struct devlink *devlink = priv_to_devlink(mlxsw_sp->core);
 	struct mlxsw_sp_port *mlxsw_sp_port;
+	struct devlink_port *devlink_port;
 	struct net_device *dev;
 	bool usable;
 	size_t bytes;
@@ -1417,6 +1420,14 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 		goto port_not_usable;
 	}
 
+	devlink_port = &mlxsw_sp_port->devlink_port;
+	err = devlink_port_register(devlink, devlink_port, local_port);
+	if (err) {
+		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to register devlink port\n",
+			mlxsw_sp_port->local_port);
+		goto err_devlink_port_register;
+	}
+
 	err = mlxsw_sp_port_system_port_mapping_set(mlxsw_sp_port);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set system port mapping\n",
@@ -1457,6 +1468,8 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 		goto err_register_netdev;
 	}
 
+	devlink_port_type_eth_set(devlink_port, dev);
+
 	err = mlxsw_sp_port_vlan_init(mlxsw_sp_port);
 	if (err)
 		goto err_port_vlan_init;
@@ -1472,6 +1485,8 @@ err_port_admin_status_set:
 err_port_mtu_set:
 err_port_swid_set:
 err_port_system_port_mapping_set:
+	devlink_port_unregister(&mlxsw_sp_port->devlink_port);
+err_devlink_port_register:
 port_not_usable:
 err_port_module_check:
 err_dev_addr_init:
@@ -1505,10 +1520,14 @@ static void mlxsw_sp_port_vports_fini(struct mlxsw_sp_port *mlxsw_sp_port)
 static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port];
+	struct devlink_port *devlink_port;
 
 	if (!mlxsw_sp_port)
 		return;
+	devlink_port = &mlxsw_sp_port->devlink_port;
+	devlink_port_type_clear(devlink_port);
 	unregister_netdev(mlxsw_sp_port->dev); /* This calls ndo_stop */
+	devlink_port_unregister(devlink_port);
 	mlxsw_sp_port_vports_fini(mlxsw_sp_port);
 	mlxsw_sp_port_switchdev_fini(mlxsw_sp_port);
 	free_percpu(mlxsw_sp_port->pcpu_stats);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 3b89ed2..6bf6dac 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -43,6 +43,7 @@
 #include <linux/if_vlan.h>
 #include <linux/list.h>
 #include <net/switchdev.h>
+#include <net/devlink.h>
 
 #include "port.h"
 #include "core.h"
@@ -162,6 +163,7 @@ struct mlxsw_sp_port {
 	unsigned long *untagged_vlans;
 	/* VLAN interfaces */
 	struct list_head vports_list;
+	struct devlink_port devlink_port;
 };
 
 static inline struct mlxsw_sp_port *
diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c
index d85960c..7a60a26 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c
@@ -43,6 +43,7 @@
 #include <linux/device.h>
 #include <linux/skbuff.h>
 #include <linux/if_vlan.h>
+#include <net/devlink.h>
 #include <net/switchdev.h>
 #include <generated/utsrelease.h>
 
@@ -78,6 +79,7 @@ struct mlxsw_sx_port {
 	struct mlxsw_sx_port_pcpu_stats __percpu *pcpu_stats;
 	struct mlxsw_sx *mlxsw_sx;
 	u8 local_port;
+	struct devlink_port devlink_port;
 };
 
 /* tx_hdr_version
@@ -953,7 +955,9 @@ mlxsw_sx_port_mac_learning_mode_set(struct mlxsw_sx_port *mlxsw_sx_port,
 
 static int mlxsw_sx_port_create(struct mlxsw_sx *mlxsw_sx, u8 local_port)
 {
+	struct devlink *devlink = priv_to_devlink(mlxsw_sx->core);
 	struct mlxsw_sx_port *mlxsw_sx_port;
+	struct devlink_port *devlink_port;
 	struct net_device *dev;
 	bool usable;
 	int err;
@@ -1007,6 +1011,14 @@ static int mlxsw_sx_port_create(struct mlxsw_sx *mlxsw_sx, u8 local_port)
 		goto port_not_usable;
 	}
 
+	devlink_port = &mlxsw_sx_port->devlink_port;
+	err = devlink_port_register(devlink, devlink_port, local_port);
+	if (err) {
+		dev_err(mlxsw_sx->bus_info->dev, "Port %d: Failed to register devlink port\n",
+			mlxsw_sx_port->local_port);
+		goto err_devlink_port_register;
+	}
+
 	err = mlxsw_sx_port_system_port_mapping_set(mlxsw_sx_port);
 	if (err) {
 		dev_err(mlxsw_sx->bus_info->dev, "Port %d: Failed to set system port mapping\n",
@@ -1064,6 +1076,8 @@ static int mlxsw_sx_port_create(struct mlxsw_sx *mlxsw_sx, u8 local_port)
 		goto err_register_netdev;
 	}
 
+	devlink_port_type_eth_set(devlink_port, dev);
+
 	mlxsw_sx->ports[local_port] = mlxsw_sx_port;
 	return 0;
 
@@ -1075,6 +1089,8 @@ err_port_mtu_set:
 err_port_speed_set:
 err_port_swid_set:
 err_port_system_port_mapping_set:
+	devlink_port_unregister(&mlxsw_sx_port->devlink_port);
+err_devlink_port_register:
 port_not_usable:
 err_port_module_check:
 err_dev_addr_get:
@@ -1087,11 +1103,15 @@ err_alloc_stats:
 static void mlxsw_sx_port_remove(struct mlxsw_sx *mlxsw_sx, u8 local_port)
 {
 	struct mlxsw_sx_port *mlxsw_sx_port = mlxsw_sx->ports[local_port];
+	struct devlink_port *devlink_port;
 
 	if (!mlxsw_sx_port)
 		return;
+	devlink_port = &mlxsw_sx_port->devlink_port;
+	devlink_port_type_clear(devlink_port);
 	unregister_netdev(mlxsw_sx_port->dev); /* This calls ndo_stop */
 	mlxsw_sx_port_swid_set(mlxsw_sx_port, MLXSW_PORT_SWID_DISABLED_PORT);
+	devlink_port_unregister(devlink_port);
 	free_percpu(mlxsw_sx_port->pcpu_stats);
 	free_netdev(mlxsw_sx_port->dev);
 }
-- 
cgit v0.10.2


From 284ef8035708334df99d038da1d3c7ddf77ec1bd Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:27 +0100
Subject: mlxsw: core: Add devlink port splitter callbacks

Add middle layer in mlxsw core code to forward port split/unsplit calls
into specific ASIC drivers.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c
index a4026a4..f69f628 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.c
@@ -785,6 +785,38 @@ static void mlxsw_core_debugfs_fini(struct mlxsw_core *mlxsw_core)
 	debugfs_remove_recursive(mlxsw_core->dbg_dir);
 }
 
+static int mlxsw_devlink_port_split(struct devlink *devlink,
+				    unsigned int port_index,
+				    unsigned int count)
+{
+	struct mlxsw_core *mlxsw_core = devlink_priv(devlink);
+
+	if (port_index >= MLXSW_PORT_MAX_PORTS)
+		return -EINVAL;
+	if (!mlxsw_core->driver->port_split)
+		return -EOPNOTSUPP;
+	return mlxsw_core->driver->port_split(mlxsw_core->driver_priv,
+					      port_index, count);
+}
+
+static int mlxsw_devlink_port_unsplit(struct devlink *devlink,
+				      unsigned int port_index)
+{
+	struct mlxsw_core *mlxsw_core = devlink_priv(devlink);
+
+	if (port_index >= MLXSW_PORT_MAX_PORTS)
+		return -EINVAL;
+	if (!mlxsw_core->driver->port_unsplit)
+		return -EOPNOTSUPP;
+	return mlxsw_core->driver->port_unsplit(mlxsw_core->driver_priv,
+						port_index);
+}
+
+static const struct devlink_ops mlxsw_devlink_ops = {
+	.port_split	= mlxsw_devlink_port_split,
+	.port_unsplit	= mlxsw_devlink_port_unsplit,
+};
+
 int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 				   const struct mlxsw_bus *mlxsw_bus,
 				   void *bus_priv)
@@ -800,7 +832,7 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
 	if (!mlxsw_driver)
 		return -EINVAL;
 	alloc_size = sizeof(*mlxsw_core) + mlxsw_driver->priv_size;
-	devlink = devlink_alloc(NULL, alloc_size);
+	devlink = devlink_alloc(&mlxsw_devlink_ops, alloc_size);
 	if (!devlink) {
 		err = -ENOMEM;
 		goto err_devlink_alloc;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h
index a017236..c73d1c0 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.h
@@ -186,6 +186,8 @@ struct mlxsw_driver {
 	int (*init)(void *driver_priv, struct mlxsw_core *mlxsw_core,
 		    const struct mlxsw_bus_info *mlxsw_bus_info);
 	void (*fini)(void *driver_priv);
+	int (*port_split)(void *driver_priv, u8 local_port, unsigned int count);
+	int (*port_unsplit)(void *driver_priv, u8 local_port);
 	void (*txhdr_construct)(struct sk_buff *skb,
 				const struct mlxsw_tx_info *tx_info);
 	u8 txhdr_len;
-- 
cgit v0.10.2


From 3e9b27b8fc8b00c9edaaa5fd64636e6f2b331f43 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:28 +0100
Subject: mlxsw: spectrum: Unmap local port from module during teardown

When splitting a port we replace it with 2 or 4 other ports. To be able
to do that we need to remove the original port netdev and unmap it from
its module. However, we first mark it as disabled, as active ports
cannot be unmapped.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 20e6783..0dd72f8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -320,6 +320,15 @@ static int mlxsw_sp_port_module_check(struct mlxsw_sp_port *mlxsw_sp_port,
 	return 0;
 }
 
+static int mlxsw_sp_port_module_unmap(struct mlxsw_sp *mlxsw_sp, u8 local_port)
+{
+	char pmlp_pl[MLXSW_REG_PMLP_LEN];
+
+	mlxsw_reg_pmlp_pack(pmlp_pl, local_port);
+	mlxsw_reg_pmlp_width_set(pmlp_pl, 0);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pmlp), pmlp_pl);
+}
+
 static int mlxsw_sp_port_open(struct net_device *dev)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
@@ -1530,6 +1539,8 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	devlink_port_unregister(devlink_port);
 	mlxsw_sp_port_vports_fini(mlxsw_sp_port);
 	mlxsw_sp_port_switchdev_fini(mlxsw_sp_port);
+	mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT);
+	mlxsw_sp_port_module_unmap(mlxsw_sp, mlxsw_sp_port->local_port);
 	free_percpu(mlxsw_sp_port->pcpu_stats);
 	kfree(mlxsw_sp_port->untagged_vlans);
 	kfree(mlxsw_sp_port->active_vlans);
-- 
cgit v0.10.2


From 558c2d5e52142fc503f79bec89413724cfc37bdd Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:29 +0100
Subject: mlxsw: spectrum: Store local port to module mapping during init

The port netdevs are each associated with a different local port number
in the device. These local ports are grouped into groups of 4 (e.g.
(1-4), (5-8)) called clusters. The cluster constitutes the one of two
possible modules they can be mapped to. This mapping is board-specific
and done by the device's firmware during init.

When splitting a port by 4, the device requires us to first unmap all
the ports in the cluster and then map each to a single lane in the module
associated with the port netdev used as the handle for the operation.
This means that two port netdevs will disappear, as only 100Gb/s (4
lanes) ports can be split and we are guaranteed to have two of these
((1, 3), (5, 7) etc.) in a cluster.

When unsplit occurs we need to reinstantiate the two original 100Gb/s
ports and map each to its origianl module. Therefore, during driver init
store the initial local port to module mapping, so it can be used later
during unsplitting.

Note that a by 2 split doesn't require us to store the mapping, as we
only need to reinstantiate one port whose module is known.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 0dd72f8..240881c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -305,18 +305,19 @@ mlxsw_sp_port_system_port_mapping_set(struct mlxsw_sp_port *mlxsw_sp_port)
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sspr), sspr_pl);
 }
 
-static int mlxsw_sp_port_module_check(struct mlxsw_sp_port *mlxsw_sp_port,
-				      bool *p_usable)
+static int mlxsw_sp_port_module_info_get(struct mlxsw_sp *mlxsw_sp,
+					 u8 local_port, u8 *p_module,
+					 u8 *p_width)
 {
-	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char pmlp_pl[MLXSW_REG_PMLP_LEN];
 	int err;
 
-	mlxsw_reg_pmlp_pack(pmlp_pl, mlxsw_sp_port->local_port);
+	mlxsw_reg_pmlp_pack(pmlp_pl, local_port);
 	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmlp), pmlp_pl);
 	if (err)
 		return err;
-	*p_usable = mlxsw_reg_pmlp_width_get(pmlp_pl) ? true : false;
+	*p_module = mlxsw_reg_pmlp_module_get(pmlp_pl, 0);
+	*p_width = mlxsw_reg_pmlp_width_get(pmlp_pl);
 	return 0;
 }
 
@@ -1365,7 +1366,6 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	struct mlxsw_sp_port *mlxsw_sp_port;
 	struct devlink_port *devlink_port;
 	struct net_device *dev;
-	bool usable;
 	size_t bytes;
 	int err;
 
@@ -1416,19 +1416,6 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	 */
 	dev->hard_header_len += MLXSW_TXHDR_LEN;
 
-	err = mlxsw_sp_port_module_check(mlxsw_sp_port, &usable);
-	if (err) {
-		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to check module\n",
-			mlxsw_sp_port->local_port);
-		goto err_port_module_check;
-	}
-
-	if (!usable) {
-		dev_dbg(mlxsw_sp->bus_info->dev, "Port %d: Not usable, skipping initialization\n",
-			mlxsw_sp_port->local_port);
-		goto port_not_usable;
-	}
-
 	devlink_port = &mlxsw_sp_port->devlink_port;
 	err = devlink_port_register(devlink, devlink_port, local_port);
 	if (err) {
@@ -1496,8 +1483,6 @@ err_port_swid_set:
 err_port_system_port_mapping_set:
 	devlink_port_unregister(&mlxsw_sp_port->devlink_port);
 err_devlink_port_register:
-port_not_usable:
-err_port_module_check:
 err_dev_addr_init:
 	free_percpu(mlxsw_sp_port->pcpu_stats);
 err_alloc_stats:
@@ -1559,6 +1544,7 @@ static void mlxsw_sp_ports_remove(struct mlxsw_sp *mlxsw_sp)
 static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp)
 {
 	size_t alloc_size;
+	u8 module, width;
 	int i;
 	int err;
 
@@ -1568,6 +1554,13 @@ static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp)
 		return -ENOMEM;
 
 	for (i = 1; i < MLXSW_PORT_MAX_PORTS; i++) {
+		err = mlxsw_sp_port_module_info_get(mlxsw_sp, i, &module,
+						    &width);
+		if (err)
+			goto err_port_module_info_get;
+		if (!width)
+			continue;
+		mlxsw_sp->port_to_module[i] = module;
 		err = mlxsw_sp_port_create(mlxsw_sp, i);
 		if (err)
 			goto err_port_create;
@@ -1575,6 +1568,7 @@ static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp)
 	return 0;
 
 err_port_create:
+err_port_module_info_get:
 	for (i--; i >= 1; i--)
 		mlxsw_sp_port_remove(mlxsw_sp, i);
 	kfree(mlxsw_sp->ports);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 6bf6dac..a7d86ac 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -123,6 +123,7 @@ struct mlxsw_sp {
 	u32 ageing_time;
 	struct mlxsw_sp_upper master_bridge;
 	struct mlxsw_sp_upper lags[MLXSW_SP_LAG_MAX];
+	u8 port_to_module[MLXSW_PORT_MAX_PORTS];
 };
 
 static inline struct mlxsw_sp_upper *
-- 
cgit v0.10.2


From a133318cde2000a3264032ea3b561c9054613486 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:30 +0100
Subject: mlxsw: spectrum: Mark unused ports using NULL

When splitting and unsplitting we'll destroy usable ports on the fly, so
mark them using a NULL pointer to indicate that their local port number
is free and can be re-used.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 240881c..926019e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -1518,6 +1518,7 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 
 	if (!mlxsw_sp_port)
 		return;
+	mlxsw_sp->ports[local_port] = NULL;
 	devlink_port = &mlxsw_sp_port->devlink_port;
 	devlink_port_type_clear(devlink_port);
 	unregister_netdev(mlxsw_sp_port->dev); /* This calls ndo_stop */
-- 
cgit v0.10.2


From 18f1e70c413713f28629ffe6863a2c43248ff7a3 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Fri, 26 Feb 2016 17:32:31 +0100
Subject: mlxsw: spectrum: Introduce port splitting

Allow a user to split or unsplit a port using the newly introduced
devlink ops.

Once split, the original netdev is destroyed and 2 or 4 others are
created, according to user configuration. The new ports are like any
other port, with the sole difference of supporting a lower maximum
speed. When unsplit, the reverse process takes place.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/port.h b/drivers/net/ethernet/mellanox/mlxsw/port.h
index ae65b99..f33b997 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/port.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/port.h
@@ -59,6 +59,8 @@
 
 #define MLXSW_PORT_DONT_CARE		(MLXSW_PORT_MAX_PORTS)
 
+#define MLXSW_PORT_MODULE_MAX_WIDTH	4
+
 enum mlxsw_port_admin_status {
 	MLXSW_PORT_ADMIN_STATUS_UP = 1,
 	MLXSW_PORT_ADMIN_STATUS_DOWN = 2,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 926019e..53487d3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -321,6 +321,22 @@ static int mlxsw_sp_port_module_info_get(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 }
 
+static int mlxsw_sp_port_module_map(struct mlxsw_sp *mlxsw_sp, u8 local_port,
+				    u8 module, u8 width, u8 lane)
+{
+	char pmlp_pl[MLXSW_REG_PMLP_LEN];
+	int i;
+
+	mlxsw_reg_pmlp_pack(pmlp_pl, local_port);
+	mlxsw_reg_pmlp_width_set(pmlp_pl, width);
+	for (i = 0; i < width; i++) {
+		mlxsw_reg_pmlp_module_set(pmlp_pl, i, module);
+		mlxsw_reg_pmlp_tx_lane_set(pmlp_pl, i, lane + i);  /* Rx & Tx */
+	}
+
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pmlp), pmlp_pl);
+}
+
 static int mlxsw_sp_port_module_unmap(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 {
 	char pmlp_pl[MLXSW_REG_PMLP_LEN];
@@ -1284,6 +1300,18 @@ static u32 mlxsw_sp_to_ptys_speed(u32 speed)
 	return ptys_proto;
 }
 
+static u32 mlxsw_sp_to_ptys_upper_speed(u32 upper_speed)
+{
+	u32 ptys_proto = 0;
+	int i;
+
+	for (i = 0; i < MLXSW_SP_PORT_LINK_MODE_LEN; i++) {
+		if (mlxsw_sp_port_link_mode[i].speed <= upper_speed)
+			ptys_proto |= mlxsw_sp_port_link_mode[i].mask;
+	}
+	return ptys_proto;
+}
+
 static int mlxsw_sp_port_set_settings(struct net_device *dev,
 				      struct ethtool_cmd *cmd)
 {
@@ -1360,7 +1388,22 @@ static const struct ethtool_ops mlxsw_sp_port_ethtool_ops = {
 	.set_settings		= mlxsw_sp_port_set_settings,
 };
 
-static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
+static int
+mlxsw_sp_port_speed_by_width_set(struct mlxsw_sp_port *mlxsw_sp_port, u8 width)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	u32 upper_speed = MLXSW_SP_PORT_BASE_SPEED * width;
+	char ptys_pl[MLXSW_REG_PTYS_LEN];
+	u32 eth_proto_admin;
+
+	eth_proto_admin = mlxsw_sp_to_ptys_upper_speed(upper_speed);
+	mlxsw_reg_ptys_pack(ptys_pl, mlxsw_sp_port->local_port,
+			    eth_proto_admin);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl);
+}
+
+static int __mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port,
+				  bool split, u8 module, u8 width)
 {
 	struct devlink *devlink = priv_to_devlink(mlxsw_sp->core);
 	struct mlxsw_sp_port *mlxsw_sp_port;
@@ -1376,6 +1419,7 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	mlxsw_sp_port->dev = dev;
 	mlxsw_sp_port->mlxsw_sp = mlxsw_sp;
 	mlxsw_sp_port->local_port = local_port;
+	mlxsw_sp_port->split = split;
 	bytes = DIV_ROUND_UP(VLAN_N_VID, BITS_PER_BYTE);
 	mlxsw_sp_port->active_vlans = kzalloc(bytes, GFP_KERNEL);
 	if (!mlxsw_sp_port->active_vlans) {
@@ -1417,6 +1461,8 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 	dev->hard_header_len += MLXSW_TXHDR_LEN;
 
 	devlink_port = &mlxsw_sp_port->devlink_port;
+	if (mlxsw_sp_port->split)
+		devlink_port_split_set(devlink_port, module);
 	err = devlink_port_register(devlink, devlink_port, local_port);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to register devlink port\n",
@@ -1438,6 +1484,13 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port)
 		goto err_port_swid_set;
 	}
 
+	err = mlxsw_sp_port_speed_by_width_set(mlxsw_sp_port, width);
+	if (err) {
+		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to enable speeds\n",
+			mlxsw_sp_port->local_port);
+		goto err_port_speed_by_width_set;
+	}
+
 	err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, ETH_DATA_LEN);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set MTU\n",
@@ -1479,6 +1532,7 @@ err_register_netdev:
 err_port_buffers_init:
 err_port_admin_status_set:
 err_port_mtu_set:
+err_port_speed_by_width_set:
 err_port_swid_set:
 err_port_system_port_mapping_set:
 	devlink_port_unregister(&mlxsw_sp_port->devlink_port);
@@ -1494,6 +1548,28 @@ err_port_active_vlans_alloc:
 	return err;
 }
 
+static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port,
+				bool split, u8 module, u8 width, u8 lane)
+{
+	int err;
+
+	err = mlxsw_sp_port_module_map(mlxsw_sp, local_port, module, width,
+				       lane);
+	if (err)
+		return err;
+
+	err = __mlxsw_sp_port_create(mlxsw_sp, local_port, split, module,
+				     width);
+	if (err)
+		goto err_port_create;
+
+	return 0;
+
+err_port_create:
+	mlxsw_sp_port_module_unmap(mlxsw_sp, local_port);
+	return err;
+}
+
 static void mlxsw_sp_port_vports_fini(struct mlxsw_sp_port *mlxsw_sp_port)
 {
 	struct net_device *dev = mlxsw_sp_port->dev;
@@ -1562,7 +1638,7 @@ static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp)
 		if (!width)
 			continue;
 		mlxsw_sp->port_to_module[i] = module;
-		err = mlxsw_sp_port_create(mlxsw_sp, i);
+		err = __mlxsw_sp_port_create(mlxsw_sp, i, false, module, width);
 		if (err)
 			goto err_port_create;
 	}
@@ -1576,6 +1652,137 @@ err_port_module_info_get:
 	return err;
 }
 
+static u8 mlxsw_sp_cluster_base_port_get(u8 local_port)
+{
+	u8 offset = (local_port - 1) % MLXSW_SP_PORTS_PER_CLUSTER_MAX;
+
+	return local_port - offset;
+}
+
+static int mlxsw_sp_port_split(void *priv, u8 local_port, unsigned int count)
+{
+	struct mlxsw_sp *mlxsw_sp = priv;
+	struct mlxsw_sp_port *mlxsw_sp_port;
+	u8 width = MLXSW_PORT_MODULE_MAX_WIDTH / count;
+	u8 module, cur_width, base_port;
+	int i;
+	int err;
+
+	mlxsw_sp_port = mlxsw_sp->ports[local_port];
+	if (!mlxsw_sp_port) {
+		dev_err(mlxsw_sp->bus_info->dev, "Port number \"%d\" does not exist\n",
+			local_port);
+		return -EINVAL;
+	}
+
+	if (count != 2 && count != 4) {
+		netdev_err(mlxsw_sp_port->dev, "Port can only be split into 2 or 4 ports\n");
+		return -EINVAL;
+	}
+
+	err = mlxsw_sp_port_module_info_get(mlxsw_sp, local_port, &module,
+					    &cur_width);
+	if (err) {
+		netdev_err(mlxsw_sp_port->dev, "Failed to get port's width\n");
+		return err;
+	}
+
+	if (cur_width != MLXSW_PORT_MODULE_MAX_WIDTH) {
+		netdev_err(mlxsw_sp_port->dev, "Port cannot be split further\n");
+		return -EINVAL;
+	}
+
+	/* Make sure we have enough slave (even) ports for the split. */
+	if (count == 2) {
+		base_port = local_port;
+		if (mlxsw_sp->ports[base_port + 1]) {
+			netdev_err(mlxsw_sp_port->dev, "Invalid split configuration\n");
+			return -EINVAL;
+		}
+	} else {
+		base_port = mlxsw_sp_cluster_base_port_get(local_port);
+		if (mlxsw_sp->ports[base_port + 1] ||
+		    mlxsw_sp->ports[base_port + 3]) {
+			netdev_err(mlxsw_sp_port->dev, "Invalid split configuration\n");
+			return -EINVAL;
+		}
+	}
+
+	for (i = 0; i < count; i++)
+		mlxsw_sp_port_remove(mlxsw_sp, base_port + i);
+
+	for (i = 0; i < count; i++) {
+		err = mlxsw_sp_port_create(mlxsw_sp, base_port + i, true,
+					   module, width, i * width);
+		if (err) {
+			dev_err(mlxsw_sp->bus_info->dev, "Failed to create split port\n");
+			goto err_port_create;
+		}
+	}
+
+	return 0;
+
+err_port_create:
+	for (i--; i >= 0; i--)
+		mlxsw_sp_port_remove(mlxsw_sp, base_port + i);
+	for (i = 0; i < count / 2; i++) {
+		module = mlxsw_sp->port_to_module[base_port + i * 2];
+		mlxsw_sp_port_create(mlxsw_sp, base_port + i * 2, false,
+				     module, MLXSW_PORT_MODULE_MAX_WIDTH, 0);
+	}
+	return err;
+}
+
+static int mlxsw_sp_port_unsplit(void *priv, u8 local_port)
+{
+	struct mlxsw_sp *mlxsw_sp = priv;
+	struct mlxsw_sp_port *mlxsw_sp_port;
+	u8 module, cur_width, base_port;
+	unsigned int count;
+	int i;
+	int err;
+
+	mlxsw_sp_port = mlxsw_sp->ports[local_port];
+	if (!mlxsw_sp_port) {
+		dev_err(mlxsw_sp->bus_info->dev, "Port number \"%d\" does not exist\n",
+			local_port);
+		return -EINVAL;
+	}
+
+	if (!mlxsw_sp_port->split) {
+		netdev_err(mlxsw_sp_port->dev, "Port wasn't split\n");
+		return -EINVAL;
+	}
+
+	err = mlxsw_sp_port_module_info_get(mlxsw_sp, local_port, &module,
+					    &cur_width);
+	if (err) {
+		netdev_err(mlxsw_sp_port->dev, "Failed to get port's width\n");
+		return err;
+	}
+	count = cur_width == 1 ? 4 : 2;
+
+	base_port = mlxsw_sp_cluster_base_port_get(local_port);
+
+	/* Determine which ports to remove. */
+	if (count == 2 && local_port >= base_port + 2)
+		base_port = base_port + 2;
+
+	for (i = 0; i < count; i++)
+		mlxsw_sp_port_remove(mlxsw_sp, base_port + i);
+
+	for (i = 0; i < count / 2; i++) {
+		module = mlxsw_sp->port_to_module[base_port + i * 2];
+		err = mlxsw_sp_port_create(mlxsw_sp, base_port + i * 2, false,
+					   module, MLXSW_PORT_MODULE_MAX_WIDTH,
+					   0);
+		if (err)
+			dev_err(mlxsw_sp->bus_info->dev, "Failed to reinstantiate port\n");
+	}
+
+	return 0;
+}
+
 static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg,
 				     char *pude_pl, void *priv)
 {
@@ -1999,6 +2206,8 @@ static struct mlxsw_driver mlxsw_sp_driver = {
 	.priv_size		= sizeof(struct mlxsw_sp),
 	.init			= mlxsw_sp_init,
 	.fini			= mlxsw_sp_fini,
+	.port_split		= mlxsw_sp_port_split,
+	.port_unsplit		= mlxsw_sp_port_unsplit,
 	.txhdr_construct	= mlxsw_sp_txhdr_construct,
 	.txhdr_len		= MLXSW_TXHDR_LEN,
 	.profile		= &mlxsw_sp_config_profile,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index a7d86ac..1b691d7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -58,6 +58,10 @@
 
 #define MLXSW_SP_MID_MAX 7000
 
+#define MLXSW_SP_PORTS_PER_CLUSTER_MAX 4
+
+#define MLXSW_SP_PORT_BASE_SPEED 25000	/* Mb/s */
+
 struct mlxsw_sp_port;
 
 struct mlxsw_sp_upper {
@@ -151,7 +155,8 @@ struct mlxsw_sp_port {
 	   learning_sync:1,
 	   uc_flood:1,
 	   bridged:1,
-	   lagged:1;
+	   lagged:1,
+	   split:1;
 	u16 pvid;
 	u16 lag_id;
 	struct {
-- 
cgit v0.10.2


From fb2dabad69f099fb9c03a44276778911da50ba29 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:00 -0500
Subject: net: dsa: support VLAN filtering switchdev attr

When a user explicitly requests VLAN filtering with something like:

    # echo 1 > /sys/class/net/<bridge>/bridge/vlan_filtering

Switchdev propagates a SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING port
attribute.

Add support for it in the DSA layer with a new port_vlan_filtering
function to let drivers toggle 802.1Q filtering on user demand.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/dsa.h b/include/net/dsa.h
index 3dd5486..26c0a3f 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -305,6 +305,8 @@ struct dsa_switch_driver {
 	/*
 	 * VLAN support
 	 */
+	int	(*port_vlan_filtering)(struct dsa_switch *ds, int port,
+				       bool vlan_filtering);
 	int	(*port_vlan_prepare)(struct dsa_switch *ds, int port,
 				     const struct switchdev_obj_port_vlan *vlan,
 				     struct switchdev_trans *trans);
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index cde2923..27bf03d 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -317,6 +317,24 @@ static int dsa_slave_stp_update(struct net_device *dev, u8 state)
 	return ret;
 }
 
+static int dsa_slave_vlan_filtering(struct net_device *dev,
+				    const struct switchdev_attr *attr,
+				    struct switchdev_trans *trans)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct dsa_switch *ds = p->parent;
+
+	/* bridge skips -EOPNOTSUPP, so skip the prepare phase */
+	if (switchdev_trans_ph_prepare(trans))
+		return 0;
+
+	if (ds->drv->port_vlan_filtering)
+		return ds->drv->port_vlan_filtering(ds, p->port,
+						    attr->u.vlan_filtering);
+
+	return 0;
+}
+
 static int dsa_slave_port_attr_set(struct net_device *dev,
 				   const struct switchdev_attr *attr,
 				   struct switchdev_trans *trans)
@@ -333,6 +351,9 @@ static int dsa_slave_port_attr_set(struct net_device *dev,
 			ret = ds->drv->port_stp_update(ds, p->port,
 						       attr->u.stp_state);
 		break;
+	case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING:
+		ret = dsa_slave_vlan_filtering(dev, attr, trans);
+		break;
 	default:
 		ret = -EOPNOTSUPP;
 		break;
-- 
cgit v0.10.2


From 2fb5ef09de7c8ab30f7aa315e700a46f51ac4b98 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:01 -0500
Subject: net: dsa: mv88e6xxx: extract single VLAN retrieval

Rename _mv88e6xxx_vlan_init in _mv88e6xxx_vtu_new, eventually called
from a new _mv88e6xxx_vtu_get function, which abstracts the VTU GetNext
VID-1 trick to retrieve a single entry.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index d98dc63..e9e9922 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1458,8 +1458,8 @@ loadpurge:
 	return _mv88e6xxx_vtu_cmd(ds, GLOBAL_VTU_OP_STU_LOAD_PURGE);
 }
 
-static int _mv88e6xxx_vlan_init(struct dsa_switch *ds, u16 vid,
-				struct mv88e6xxx_vtu_stu_entry *entry)
+static int _mv88e6xxx_vtu_new(struct dsa_switch *ds, u16 vid,
+			      struct mv88e6xxx_vtu_stu_entry *entry)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	struct mv88e6xxx_vtu_stu_entry vlan = {
@@ -1509,6 +1509,35 @@ static int _mv88e6xxx_vlan_init(struct dsa_switch *ds, u16 vid,
 	return 0;
 }
 
+static int _mv88e6xxx_vtu_get(struct dsa_switch *ds, u16 vid,
+			      struct mv88e6xxx_vtu_stu_entry *entry, bool creat)
+{
+	int err;
+
+	if (!vid)
+		return -EINVAL;
+
+	err = _mv88e6xxx_vtu_vid_write(ds, vid - 1);
+	if (err)
+		return err;
+
+	err = _mv88e6xxx_vtu_getnext(ds, entry);
+	if (err)
+		return err;
+
+	if (entry->vid != vid || !entry->valid) {
+		if (!creat)
+			return -EOPNOTSUPP;
+		/* -ENOENT would've been more appropriate, but switchdev expects
+		 * -EOPNOTSUPP to inform bridge about an eventual software VLAN.
+		 */
+
+		err = _mv88e6xxx_vtu_new(ds, vid, entry);
+	}
+
+	return err;
+}
+
 static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port,
 					u16 vid_begin, u16 vid_end)
 {
@@ -1593,20 +1622,10 @@ static int _mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, u16 vid,
 	struct mv88e6xxx_vtu_stu_entry vlan;
 	int err;
 
-	err = _mv88e6xxx_vtu_vid_write(ds, vid - 1);
-	if (err)
-		return err;
-
-	err = _mv88e6xxx_vtu_getnext(ds, &vlan);
+	err = _mv88e6xxx_vtu_get(ds, vid, &vlan, true);
 	if (err)
 		return err;
 
-	if (vlan.vid != vid || !vlan.valid) {
-		err = _mv88e6xxx_vlan_init(ds, vid, &vlan);
-		if (err)
-			return err;
-	}
-
 	vlan.data[port] = untagged ?
 		GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED :
 		GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED;
@@ -1647,16 +1666,12 @@ static int _mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, u16 vid)
 	struct mv88e6xxx_vtu_stu_entry vlan;
 	int i, err;
 
-	err = _mv88e6xxx_vtu_vid_write(ds, vid - 1);
-	if (err)
-		return err;
-
-	err = _mv88e6xxx_vtu_getnext(ds, &vlan);
+	err = _mv88e6xxx_vtu_get(ds, vid, &vlan, false);
 	if (err)
 		return err;
 
-	if (vlan.vid != vid || !vlan.valid ||
-	    vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER)
+	/* Tell switchdev if this VLAN is handled in software */
+	if (vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER)
 		return -EOPNOTSUPP;
 
 	vlan.data[port] = GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER;
-- 
cgit v0.10.2


From 74b6ba0d76acd3ce5edac54aa73c4eb7e3a93859 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:02 -0500
Subject: net: dsa: mv88e6xxx: extract single FDB dump

Move out the code which dumps a single FDB to its own function.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index e9e9922..6329516 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1895,6 +1895,47 @@ static int _mv88e6xxx_atu_getnext(struct dsa_switch *ds, u16 fid,
 	return 0;
 }
 
+static int _mv88e6xxx_port_fdb_dump_one(struct dsa_switch *ds, u16 fid, u16 vid,
+					int port,
+					struct switchdev_obj_port_fdb *fdb,
+					int (*cb)(struct switchdev_obj *obj))
+{
+	struct mv88e6xxx_atu_entry addr = {
+		.mac = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff },
+	};
+	int err;
+
+	err = _mv88e6xxx_atu_mac_write(ds, addr.mac);
+	if (err)
+		return err;
+
+	do {
+		err = _mv88e6xxx_atu_getnext(ds, fid, &addr);
+		if (err)
+			break;
+
+		if (addr.state == GLOBAL_ATU_DATA_STATE_UNUSED)
+			break;
+
+		if (!addr.trunk && addr.portv_trunkid & BIT(port)) {
+			bool is_static = addr.state ==
+				(is_multicast_ether_addr(addr.mac) ?
+				 GLOBAL_ATU_DATA_STATE_MC_STATIC :
+				 GLOBAL_ATU_DATA_STATE_UC_STATIC);
+
+			fdb->vid = vid;
+			ether_addr_copy(fdb->addr, addr.mac);
+			fdb->ndm_state = is_static ? NUD_NOARP : NUD_REACHABLE;
+
+			err = cb(&fdb->obj);
+			if (err)
+				break;
+		}
+	} while (!is_broadcast_ether_addr(addr.mac));
+
+	return err;
+}
+
 int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port,
 			    struct switchdev_obj_port_fdb *fdb,
 			    int (*cb)(struct switchdev_obj *obj))
@@ -1907,51 +1948,23 @@ int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port,
 
 	mutex_lock(&ps->smi_mutex);
 
+	/* Dump VLANs' Filtering Information Databases */
 	err = _mv88e6xxx_vtu_vid_write(ds, vlan.vid);
 	if (err)
 		goto unlock;
 
 	do {
-		struct mv88e6xxx_atu_entry addr = {
-			.mac = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff },
-		};
-
 		err = _mv88e6xxx_vtu_getnext(ds, &vlan);
 		if (err)
-			goto unlock;
+			break;
 
 		if (!vlan.valid)
 			break;
 
-		err = _mv88e6xxx_atu_mac_write(ds, addr.mac);
+		err = _mv88e6xxx_port_fdb_dump_one(ds, vlan.fid, vlan.vid, port,
+						   fdb, cb);
 		if (err)
-			goto unlock;
-
-		do {
-			err = _mv88e6xxx_atu_getnext(ds, vlan.fid, &addr);
-			if (err)
-				goto unlock;
-
-			if (addr.state == GLOBAL_ATU_DATA_STATE_UNUSED)
-				break;
-
-			if (!addr.trunk && addr.portv_trunkid & BIT(port)) {
-				bool is_static = addr.state ==
-					(is_multicast_ether_addr(addr.mac) ?
-					 GLOBAL_ATU_DATA_STATE_MC_STATIC :
-					 GLOBAL_ATU_DATA_STATE_UC_STATIC);
-
-				fdb->vid = vlan.vid;
-				ether_addr_copy(fdb->addr, addr.mac);
-				fdb->ndm_state = is_static ? NUD_NOARP :
-					NUD_REACHABLE;
-
-				err = cb(&fdb->obj);
-				if (err)
-					goto unlock;
-			}
-		} while (!is_broadcast_ether_addr(addr.mac));
-
+			break;
 	} while (vlan.vid < GLOBAL_VTU_VID_MASK);
 
 unlock:
-- 
cgit v0.10.2


From 3285f9e8695b26b1e15a58f21aff21087a7e7555 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:03 -0500
Subject: net: dsa: mv88e6xxx: assign dynamic FDB to VLANs

Add a _mv88e6xxx_fid_new function which gives and flushes the lowest FID
available. Call it when preparing a new VTU entry.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 6329516..b4b2f05 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1458,6 +1458,41 @@ loadpurge:
 	return _mv88e6xxx_vtu_cmd(ds, GLOBAL_VTU_OP_STU_LOAD_PURGE);
 }
 
+static int _mv88e6xxx_fid_new(struct dsa_switch *ds, u16 *fid)
+{
+	DECLARE_BITMAP(fid_bitmap, MV88E6XXX_N_FID);
+	struct mv88e6xxx_vtu_stu_entry vlan;
+	int err;
+
+	bitmap_zero(fid_bitmap, MV88E6XXX_N_FID);
+
+	/* Set every FID bit used by the VLAN entries */
+	err = _mv88e6xxx_vtu_vid_write(ds, GLOBAL_VTU_VID_MASK);
+	if (err)
+		return err;
+
+	do {
+		err = _mv88e6xxx_vtu_getnext(ds, &vlan);
+		if (err)
+			return err;
+
+		if (!vlan.valid)
+			break;
+
+		set_bit(vlan.fid, fid_bitmap);
+	} while (vlan.vid < GLOBAL_VTU_VID_MASK);
+
+	/* The reset value 0x000 is used to indicate that multiple address
+	 * databases are not needed. Return the next positive available.
+	 */
+	*fid = find_next_zero_bit(fid_bitmap, MV88E6XXX_N_FID, 1);
+	if (unlikely(*fid == MV88E6XXX_N_FID))
+		return -ENOSPC;
+
+	/* Clear the database */
+	return _mv88e6xxx_atu_flush(ds, *fid, true);
+}
+
 static int _mv88e6xxx_vtu_new(struct dsa_switch *ds, u16 vid,
 			      struct mv88e6xxx_vtu_stu_entry *entry)
 {
@@ -1465,9 +1500,12 @@ static int _mv88e6xxx_vtu_new(struct dsa_switch *ds, u16 vid,
 	struct mv88e6xxx_vtu_stu_entry vlan = {
 		.valid = true,
 		.vid = vid,
-		.fid = vid, /* We use one FID per VLAN */
 	};
-	int i;
+	int i, err;
+
+	err = _mv88e6xxx_fid_new(ds, &vlan.fid);
+	if (err)
+		return err;
 
 	/* exclude all ports except the CPU and DSA ports */
 	for (i = 0; i < ps->num_ports; ++i)
@@ -1478,7 +1516,6 @@ static int _mv88e6xxx_vtu_new(struct dsa_switch *ds, u16 vid,
 	if (mv88e6xxx_6097_family(ds) || mv88e6xxx_6165_family(ds) ||
 	    mv88e6xxx_6351_family(ds) || mv88e6xxx_6352_family(ds)) {
 		struct mv88e6xxx_vtu_stu_entry vstp;
-		int err;
 
 		/* Adding a VTU entry requires a valid STU entry. As VSTP is not
 		 * implemented, only one STU entry is needed to cover all VTU
@@ -1498,11 +1535,6 @@ static int _mv88e6xxx_vtu_new(struct dsa_switch *ds, u16 vid,
 			if (err)
 				return err;
 		}
-
-		/* Clear all MAC addresses from the new database */
-		err = _mv88e6xxx_atu_flush(ds, vlan.fid, true);
-		if (err)
-			return err;
 	}
 
 	*entry = vlan;
@@ -1789,8 +1821,14 @@ static int _mv88e6xxx_port_fdb_load(struct dsa_switch *ds, int port,
 				    u8 state)
 {
 	struct mv88e6xxx_atu_entry entry = { 0 };
+	struct mv88e6xxx_vtu_stu_entry vlan;
+	int err;
+
+	err = _mv88e6xxx_vtu_get(ds, vid, &vlan, false);
+	if (err)
+		return err;
 
-	entry.fid = vid; /* We use one FID per VLAN */
+	entry.fid = vlan.fid;
 	entry.state = state;
 	ether_addr_copy(entry.mac, addr);
 	if (state != GLOBAL_ATU_DATA_STATE_UNUSED) {
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 6a30bda..9df331e 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -355,6 +355,8 @@
 #define GLOBAL2_QOS_WEIGHT	0x1c
 #define GLOBAL2_MISC		0x1d
 
+#define MV88E6XXX_N_FID		4096
+
 struct mv88e6xxx_switch_id {
 	u16 id;
 	char *name;
-- 
cgit v0.10.2


From 2db9ce1fd9a34ea560ff120bf763007ddf99c7bb Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:04 -0500
Subject: net: dsa: mv88e6xxx: assign default FDB to ports

Restore per-port FDB. Assign them on setup, allow adding and deleting
addresses into them, and dump them.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index b4b2f05..0f06488 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1458,14 +1458,82 @@ loadpurge:
 	return _mv88e6xxx_vtu_cmd(ds, GLOBAL_VTU_OP_STU_LOAD_PURGE);
 }
 
+static int _mv88e6xxx_port_fid(struct dsa_switch *ds, int port, u16 *new,
+			       u16 *old)
+{
+	u16 fid;
+	int ret;
+
+	/* Port's default FID bits 3:0 are located in reg 0x06, offset 12 */
+	ret = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_BASE_VLAN);
+	if (ret < 0)
+		return ret;
+
+	fid = (ret & PORT_BASE_VLAN_FID_3_0_MASK) >> 12;
+
+	if (new) {
+		ret &= ~PORT_BASE_VLAN_FID_3_0_MASK;
+		ret |= (*new << 12) & PORT_BASE_VLAN_FID_3_0_MASK;
+
+		ret = _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_BASE_VLAN,
+					   ret);
+		if (ret < 0)
+			return ret;
+	}
+
+	/* Port's default FID bits 11:4 are located in reg 0x05, offset 0 */
+	ret = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_CONTROL_1);
+	if (ret < 0)
+		return ret;
+
+	fid |= (ret & PORT_CONTROL_1_FID_11_4_MASK) << 4;
+
+	if (new) {
+		ret &= ~PORT_CONTROL_1_FID_11_4_MASK;
+		ret |= (*new >> 4) & PORT_CONTROL_1_FID_11_4_MASK;
+
+		ret = _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_CONTROL_1,
+					   ret);
+		if (ret < 0)
+			return ret;
+
+		netdev_dbg(ds->ports[port], "FID %d (was %d)\n", *new, fid);
+	}
+
+	if (old)
+		*old = fid;
+
+	return 0;
+}
+
+static int _mv88e6xxx_port_fid_get(struct dsa_switch *ds, int port, u16 *fid)
+{
+	return _mv88e6xxx_port_fid(ds, port, NULL, fid);
+}
+
+static int _mv88e6xxx_port_fid_set(struct dsa_switch *ds, int port, u16 fid)
+{
+	return _mv88e6xxx_port_fid(ds, port, &fid, NULL);
+}
+
 static int _mv88e6xxx_fid_new(struct dsa_switch *ds, u16 *fid)
 {
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	DECLARE_BITMAP(fid_bitmap, MV88E6XXX_N_FID);
 	struct mv88e6xxx_vtu_stu_entry vlan;
-	int err;
+	int i, err;
 
 	bitmap_zero(fid_bitmap, MV88E6XXX_N_FID);
 
+	/* Set every FID bit used by the (un)bridged ports */
+	for (i = 0; i < ps->num_ports; ++i) {
+		err = _mv88e6xxx_port_fid_get(ds, i, fid);
+		if (err)
+			return err;
+
+		set_bit(*fid, fid_bitmap);
+	}
+
 	/* Set every FID bit used by the VLAN entries */
 	err = _mv88e6xxx_vtu_vid_write(ds, GLOBAL_VTU_VID_MASK);
 	if (err)
@@ -1824,7 +1892,11 @@ static int _mv88e6xxx_port_fdb_load(struct dsa_switch *ds, int port,
 	struct mv88e6xxx_vtu_stu_entry vlan;
 	int err;
 
-	err = _mv88e6xxx_vtu_get(ds, vid, &vlan, false);
+	/* Null VLAN ID corresponds to the port private database */
+	if (vid == 0)
+		err = _mv88e6xxx_port_fid_get(ds, port, &vlan.fid);
+	else
+		err = _mv88e6xxx_vtu_get(ds, vid, &vlan, false);
 	if (err)
 		return err;
 
@@ -1843,10 +1915,6 @@ int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port,
 			       const struct switchdev_obj_port_fdb *fdb,
 			       struct switchdev_trans *trans)
 {
-	/* We don't use per-port FDB */
-	if (fdb->vid == 0)
-		return -EOPNOTSUPP;
-
 	/* We don't need any dynamic resource from the kernel (yet),
 	 * so skip the prepare phase.
 	 */
@@ -1982,10 +2050,20 @@ int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port,
 	struct mv88e6xxx_vtu_stu_entry vlan = {
 		.vid = GLOBAL_VTU_VID_MASK, /* all ones */
 	};
+	u16 fid;
 	int err;
 
 	mutex_lock(&ps->smi_mutex);
 
+	/* Dump port's default Filtering Information Database (VLAN ID 0) */
+	err = _mv88e6xxx_port_fid_get(ds, port, &fid);
+	if (err)
+		goto unlock;
+
+	err = _mv88e6xxx_port_fdb_dump_one(ds, fid, 0, port, fdb, cb);
+	if (err)
+		goto unlock;
+
 	/* Dump VLANs' Filtering Information Databases */
 	err = _mv88e6xxx_vtu_vid_write(ds, vlan.vid);
 	if (err)
@@ -2286,9 +2364,13 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port)
 	if (ret)
 		goto abort;
 
-	/* Port based VLAN map: do not give each port its own address
+	/* Port based VLAN map: give each port its own address
 	 * database, and allow every port to egress frames on all other ports.
 	 */
+	ret = _mv88e6xxx_port_fid_set(ds, port, port + 1);
+	if (ret)
+		goto abort;
+
 	reg = BIT(ps->num_ports) - 1; /* all ports */
 	reg &= ~BIT(port); /* except itself */
 	ret = _mv88e6xxx_port_vlan_map_set(ds, port, reg);
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 9df331e..85a4166 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -133,7 +133,9 @@
 #define PORT_CONTROL_STATE_LEARNING	0x02
 #define PORT_CONTROL_STATE_FORWARDING	0x03
 #define PORT_CONTROL_1		0x05
+#define PORT_CONTROL_1_FID_11_4_MASK	(0xff << 0)
 #define PORT_BASE_VLAN		0x06
+#define PORT_BASE_VLAN_FID_3_0_MASK	(0xf << 12)
 #define PORT_DEFAULT_VLAN	0x07
 #define PORT_DEFAULT_VLAN_MASK	0xfff
 #define PORT_CONTROL_2		0x08
-- 
cgit v0.10.2


From 466dfa0770220cebad2e58e1905489328fc9daf7 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:05 -0500
Subject: net: dsa: mv88e6xxx: assign dynamic FDB to bridges

Give a new bridge a fresh FDB, assign it to its members, and restore a
fresh FDB to a port leaving a bridge.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 0f06488..0f16911 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -2093,19 +2093,56 @@ int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port,
 			       struct net_device *bridge)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	u16 fid;
+	int i, err;
+
+	mutex_lock(&ps->smi_mutex);
+
+	/* Get or create the bridge FID and assign it to the port */
+	for (i = 0; i < ps->num_ports; ++i)
+		if (ps->ports[i].bridge_dev == bridge)
+			break;
+
+	if (i < ps->num_ports)
+		err = _mv88e6xxx_port_fid_get(ds, i, &fid);
+	else
+		err = _mv88e6xxx_fid_new(ds, &fid);
+	if (err)
+		goto unlock;
+
+	err = _mv88e6xxx_port_fid_set(ds, port, fid);
+	if (err)
+		goto unlock;
 
 	ps->ports[port].bridge_dev = bridge;
+unlock:
+	mutex_unlock(&ps->smi_mutex);
 
-	return 0;
+	return err;
 }
 
 int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	u16 fid;
+	int err;
+
+	mutex_lock(&ps->smi_mutex);
+
+	/* Give the port a fresh Filtering Information Database */
+	err = _mv88e6xxx_fid_new(ds, &fid);
+	if (err)
+		goto unlock;
+
+	err = _mv88e6xxx_port_fid_set(ds, port, fid);
+	if (err)
+		goto unlock;
 
 	ps->ports[port].bridge_dev = NULL;
+unlock:
+	mutex_unlock(&ps->smi_mutex);
 
-	return 0;
+	return err;
 }
 
 static int mv88e6xxx_setup_port_default_vlan(struct dsa_switch *ds, int port)
-- 
cgit v0.10.2


From b7666efe46caef008eb92f11392f6f839b35d824 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:06 -0500
Subject: net: dsa: mv88e6xxx: restore VLANTable map control

The In Chip Port Based VLAN Table contains bits used to restrict which
output ports this input port can send frames to.

With the VLAN filtering enabled, these tables work in conjunction with
the VLAN Table Unit to allow egressing frames.

In order to remove the current dependency to BRIDGE_VLAN_FILTERING for
basic hardware bridging to work, it is necessary to restore a fine
control of each port's VLANTable, on setup and when a port joins or
leaves a bridge.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 0f16911..7f3036b 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1087,12 +1087,32 @@ abort:
 	return ret;
 }
 
-static int _mv88e6xxx_port_vlan_map_set(struct dsa_switch *ds, int port,
-					u16 output_ports)
+static int _mv88e6xxx_port_based_vlan_map(struct dsa_switch *ds, int port)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	struct net_device *bridge = ps->ports[port].bridge_dev;
 	const u16 mask = (1 << ps->num_ports) - 1;
+	u16 output_ports = 0;
 	int reg;
+	int i;
+
+	/* allow CPU port or DSA link(s) to send frames to every port */
+	if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) {
+		output_ports = mask;
+	} else {
+		for (i = 0; i < ps->num_ports; ++i) {
+			/* allow sending frames to every group member */
+			if (bridge && ps->ports[i].bridge_dev == bridge)
+				output_ports |= BIT(i);
+
+			/* allow sending frames to CPU port and DSA link(s) */
+			if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i))
+				output_ports |= BIT(i);
+		}
+	}
+
+	/* prevent frames from going back out of the port they came in on */
+	output_ports &= ~BIT(port);
 
 	reg = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_BASE_VLAN);
 	if (reg < 0)
@@ -2114,7 +2134,17 @@ int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port,
 	if (err)
 		goto unlock;
 
+	/* Assign the bridge and remap each port's VLANTable */
 	ps->ports[port].bridge_dev = bridge;
+
+	for (i = 0; i < ps->num_ports; ++i) {
+		if (ps->ports[i].bridge_dev == bridge) {
+			err = _mv88e6xxx_port_based_vlan_map(ds, i);
+			if (err)
+				break;
+		}
+	}
+
 unlock:
 	mutex_unlock(&ps->smi_mutex);
 
@@ -2124,8 +2154,9 @@ unlock:
 int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	struct net_device *bridge = ps->ports[port].bridge_dev;
 	u16 fid;
-	int err;
+	int i, err;
 
 	mutex_lock(&ps->smi_mutex);
 
@@ -2138,7 +2169,17 @@ int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port)
 	if (err)
 		goto unlock;
 
+	/* Unassign the bridge and remap each port's VLANTable */
 	ps->ports[port].bridge_dev = NULL;
+
+	for (i = 0; i < ps->num_ports; ++i) {
+		if (i == port || ps->ports[i].bridge_dev == bridge) {
+			err = _mv88e6xxx_port_based_vlan_map(ds, i);
+			if (err)
+				break;
+		}
+	}
+
 unlock:
 	mutex_unlock(&ps->smi_mutex);
 
@@ -2402,15 +2443,14 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port)
 		goto abort;
 
 	/* Port based VLAN map: give each port its own address
-	 * database, and allow every port to egress frames on all other ports.
+	 * database, and allow bidirectional communication between the
+	 * CPU and DSA port(s), and the other ports.
 	 */
 	ret = _mv88e6xxx_port_fid_set(ds, port, port + 1);
 	if (ret)
 		goto abort;
 
-	reg = BIT(ps->num_ports) - 1; /* all ports */
-	reg &= ~BIT(port); /* except itself */
-	ret = _mv88e6xxx_port_vlan_map_set(ds, port, reg);
+	ret = _mv88e6xxx_port_based_vlan_map(ds, port);
 	if (ret)
 		goto abort;
 
-- 
cgit v0.10.2


From 46fbe5e5af772e0533ba3ab3d9977a5b097b88b5 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:07 -0500
Subject: net: dsa: mv88e6xxx: remove reserved VLANs

Now that ports isolation is correctly configured when joining or leaving
a bridge, there is no need to rely on reserved VLANs to isolate
unbridged ports anymore. Thus remove them, and disable 802.1Q on setup.

This restores the expected behavior of hardware bridging for systems
without 802.1Q or VLAN filtering enabled.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 7f3036b..27a19dc 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1718,10 +1718,6 @@ int mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port,
 {
 	int err;
 
-	/* We reserve a few VLANs to isolate unbridged ports */
-	if (vlan->vid_end >= 4000)
-		return -EOPNOTSUPP;
-
 	/* If the requested port doesn't belong to the same bridge as the VLAN
 	 * members, do not support it (yet) and fallback to software VLAN.
 	 */
@@ -1819,7 +1815,6 @@ int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port,
 			    const struct switchdev_obj_port_vlan *vlan)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
-	const u16 defpvid = 4000 + ds->index * DSA_MAX_PORTS + port;
 	u16 pvid, vid;
 	int err = 0;
 
@@ -1835,8 +1830,7 @@ int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port,
 			goto unlock;
 
 		if (vid == pvid) {
-			/* restore reserved VLAN ID */
-			err = _mv88e6xxx_port_pvid_set(ds, port, defpvid);
+			err = _mv88e6xxx_port_pvid_set(ds, port, 0);
 			if (err)
 				goto unlock;
 		}
@@ -2186,20 +2180,6 @@ unlock:
 	return err;
 }
 
-static int mv88e6xxx_setup_port_default_vlan(struct dsa_switch *ds, int port)
-{
-	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
-	const u16 pvid = 4000 + ds->index * DSA_MAX_PORTS + port;
-	int err;
-
-	mutex_lock(&ps->smi_mutex);
-	err = _mv88e6xxx_port_vlan_add(ds, port, pvid, true);
-	if (!err)
-		err = _mv88e6xxx_port_pvid_set(ds, port, pvid);
-	mutex_unlock(&ps->smi_mutex);
-	return err;
-}
-
 static void mv88e6xxx_bridge_work(struct work_struct *work)
 {
 	struct mv88e6xxx_priv_state *ps;
@@ -2320,7 +2300,7 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port)
 	}
 
 	/* Port Control 2: don't force a good FCS, set the maximum frame size to
-	 * 10240 bytes, enable secure 802.1q tags, don't discard tagged or
+	 * 10240 bytes, disable 802.1q tags checking, don't discard tagged or
 	 * untagged frames on this port, do a destination address lookup on all
 	 * received packets as usual, disable ARP mirroring and don't send a
 	 * copy of all transmitted/received frames on this port to the CPU.
@@ -2345,7 +2325,7 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port)
 			reg |= PORT_CONTROL_2_FORWARD_UNKNOWN;
 	}
 
-	reg |= PORT_CONTROL_2_8021Q_SECURE;
+	reg |= PORT_CONTROL_2_8021Q_DISABLED;
 
 	if (reg) {
 		ret = _mv88e6xxx_reg_write(ds, REG_PORT(port),
@@ -2474,13 +2454,6 @@ int mv88e6xxx_setup_ports(struct dsa_switch *ds)
 		ret = mv88e6xxx_setup_port(ds, i);
 		if (ret < 0)
 			return ret;
-
-		if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i))
-			continue;
-
-		ret = mv88e6xxx_setup_port_default_vlan(ds, i);
-		if (ret < 0)
-			return ret;
 	}
 	return 0;
 }
-- 
cgit v0.10.2


From 214cdb998739428b09d80b4b152faa7d1e6ad156 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Fri, 26 Feb 2016 13:16:08 -0500
Subject: net: dsa: mv88e6xxx: support VLAN filtering

Implement port_vlan_filtering in the driver to toggle the related port
802.1Q mode between DISABLED and SECURE, on user request.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6171.c b/drivers/net/dsa/mv88e6171.c
index dd1ebaf..d72ccbd 100644
--- a/drivers/net/dsa/mv88e6171.c
+++ b/drivers/net/dsa/mv88e6171.c
@@ -106,6 +106,7 @@ struct dsa_switch_driver mv88e6171_switch_driver = {
 	.port_join_bridge	= mv88e6xxx_port_bridge_join,
 	.port_leave_bridge	= mv88e6xxx_port_bridge_leave,
 	.port_stp_update        = mv88e6xxx_port_stp_update,
+	.port_vlan_filtering	= mv88e6xxx_port_vlan_filtering,
 	.port_vlan_prepare	= mv88e6xxx_port_vlan_prepare,
 	.port_vlan_add		= mv88e6xxx_port_vlan_add,
 	.port_vlan_del		= mv88e6xxx_port_vlan_del,
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
index bbca36a..a41fa50 100644
--- a/drivers/net/dsa/mv88e6352.c
+++ b/drivers/net/dsa/mv88e6352.c
@@ -327,6 +327,7 @@ struct dsa_switch_driver mv88e6352_switch_driver = {
 	.port_join_bridge	= mv88e6xxx_port_bridge_join,
 	.port_leave_bridge	= mv88e6xxx_port_bridge_leave,
 	.port_stp_update	= mv88e6xxx_port_stp_update,
+	.port_vlan_filtering	= mv88e6xxx_port_vlan_filtering,
 	.port_vlan_prepare	= mv88e6xxx_port_vlan_prepare,
 	.port_vlan_add		= mv88e6xxx_port_vlan_add,
 	.port_vlan_del		= mv88e6xxx_port_vlan_del,
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 27a19dc..d11c9d5 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1712,6 +1712,45 @@ unlock:
 	return err;
 }
 
+static const char * const mv88e6xxx_port_8021q_mode_names[] = {
+	[PORT_CONTROL_2_8021Q_DISABLED] = "Disabled",
+	[PORT_CONTROL_2_8021Q_FALLBACK] = "Fallback",
+	[PORT_CONTROL_2_8021Q_CHECK] = "Check",
+	[PORT_CONTROL_2_8021Q_SECURE] = "Secure",
+};
+
+int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port,
+				  bool vlan_filtering)
+{
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	u16 old, new = vlan_filtering ? PORT_CONTROL_2_8021Q_SECURE :
+		PORT_CONTROL_2_8021Q_DISABLED;
+	int ret;
+
+	mutex_lock(&ps->smi_mutex);
+
+	ret = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_CONTROL_2);
+	if (ret < 0)
+		goto unlock;
+
+	old = ret & PORT_CONTROL_2_8021Q_MASK;
+
+	ret &= ~PORT_CONTROL_2_8021Q_MASK;
+	ret |= new & PORT_CONTROL_2_8021Q_MASK;
+
+	ret = _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_CONTROL_2, ret);
+	if (ret < 0)
+		goto unlock;
+
+	netdev_dbg(ds->ports[port], "802.1Q Mode: %s (was %s)\n",
+		   mv88e6xxx_port_8021q_mode_names[new],
+		   mv88e6xxx_port_8021q_mode_names[old]);
+unlock:
+	mutex_unlock(&ps->smi_mutex);
+
+	return ret;
+}
+
 int mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port,
 				const struct switchdev_obj_port_vlan *vlan,
 				struct switchdev_trans *trans)
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 85a4166..d7b088d 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -490,6 +490,8 @@ int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port,
 			       struct net_device *bridge);
 int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port);
 int mv88e6xxx_port_stp_update(struct dsa_switch *ds, int port, u8 state);
+int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port,
+				  bool vlan_filtering);
 int mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port,
 				const struct switchdev_obj_port_vlan *vlan,
 				struct switchdev_trans *trans);
-- 
cgit v0.10.2


From 7f0aec7a668419bdbff12de6e8016544f874e708 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 26 Feb 2016 21:20:01 +0100
Subject: bridge: mcast: use names for the different multicast_router types

Using raw values makes it difficult to extend and also understand the
code, give them names and do explicit per-option manipulation in
br_multicast_set_port_router.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 0890b21..e47f3bc 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -177,6 +177,13 @@ enum {
 };
 #define MDBA_MDB_EATTR_MAX (__MDBA_MDB_EATTR_MAX - 1)
 
+/* multicast router types */
+enum {
+	MDB_RTR_TYPE_DISABLED,
+	MDB_RTR_TYPE_TEMP_QUERY,
+	MDB_RTR_TYPE_PERM,
+};
+
 enum {
 	MDBA_ROUTER_UNSPEC,
 	MDBA_ROUTER_PORT,
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 8b6e424..71c109b 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -759,7 +759,7 @@ static void br_multicast_router_expired(unsigned long data)
 	struct net_bridge *br = port->br;
 
 	spin_lock(&br->multicast_lock);
-	if (port->multicast_router != 1 ||
+	if (port->multicast_router != MDB_RTR_TYPE_TEMP_QUERY ||
 	    timer_pending(&port->multicast_router_timer) ||
 	    hlist_unhashed(&port->rlist))
 		goto out;
@@ -912,7 +912,7 @@ static void br_ip6_multicast_port_query_expired(unsigned long data)
 
 void br_multicast_add_port(struct net_bridge_port *port)
 {
-	port->multicast_router = 1;
+	port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY;
 
 	setup_timer(&port->multicast_router_timer, br_multicast_router_expired,
 		    (unsigned long)port);
@@ -959,7 +959,8 @@ void br_multicast_enable_port(struct net_bridge_port *port)
 #if IS_ENABLED(CONFIG_IPV6)
 	br_multicast_enable(&port->ip6_own_query);
 #endif
-	if (port->multicast_router == 2 && hlist_unhashed(&port->rlist))
+	if (port->multicast_router == MDB_RTR_TYPE_PERM &&
+	    hlist_unhashed(&port->rlist))
 		br_multicast_add_router(br, port);
 
 out:
@@ -1227,13 +1228,13 @@ static void br_multicast_mark_router(struct net_bridge *br,
 	unsigned long now = jiffies;
 
 	if (!port) {
-		if (br->multicast_router == 1)
+		if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY)
 			mod_timer(&br->multicast_router_timer,
 				  now + br->multicast_querier_interval);
 		return;
 	}
 
-	if (port->multicast_router != 1)
+	if (port->multicast_router != MDB_RTR_TYPE_TEMP_QUERY)
 		return;
 
 	br_multicast_add_router(br, port);
@@ -1713,7 +1714,7 @@ void br_multicast_init(struct net_bridge *br)
 	br->hash_elasticity = 4;
 	br->hash_max = 512;
 
-	br->multicast_router = 1;
+	br->multicast_router = MDB_RTR_TYPE_TEMP_QUERY;
 	br->multicast_querier = 0;
 	br->multicast_query_use_ifaddr = 0;
 	br->multicast_last_member_count = 2;
@@ -1823,11 +1824,11 @@ int br_multicast_set_router(struct net_bridge *br, unsigned long val)
 	spin_lock_bh(&br->multicast_lock);
 
 	switch (val) {
-	case 0:
-	case 2:
+	case MDB_RTR_TYPE_DISABLED:
+	case MDB_RTR_TYPE_PERM:
 		del_timer(&br->multicast_router_timer);
 		/* fall through */
-	case 1:
+	case MDB_RTR_TYPE_TEMP_QUERY:
 		br->multicast_router = val;
 		err = 0;
 		break;
@@ -1838,6 +1839,14 @@ int br_multicast_set_router(struct net_bridge *br, unsigned long val)
 	return err;
 }
 
+static void __del_port_router(struct net_bridge_port *p)
+{
+	if (hlist_unhashed(&p->rlist))
+		return;
+	hlist_del_init_rcu(&p->rlist);
+	br_rtr_notify(p->br->dev, p, RTM_DELMDB);
+}
+
 int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val)
 {
 	struct net_bridge *br = p->br;
@@ -1846,29 +1855,25 @@ int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val)
 	spin_lock(&br->multicast_lock);
 
 	switch (val) {
-	case 0:
-	case 1:
-	case 2:
-		p->multicast_router = val;
-		err = 0;
-
-		if (val < 2 && !hlist_unhashed(&p->rlist)) {
-			hlist_del_init_rcu(&p->rlist);
-			br_rtr_notify(br->dev, p, RTM_DELMDB);
-		}
-
-		if (val == 1)
-			break;
-
+	case MDB_RTR_TYPE_DISABLED:
+		p->multicast_router = MDB_RTR_TYPE_DISABLED;
+		__del_port_router(p);
+		del_timer(&p->multicast_router_timer);
+		break;
+	case MDB_RTR_TYPE_TEMP_QUERY:
+		p->multicast_router = MDB_RTR_TYPE_TEMP_QUERY;
+		__del_port_router(p);
+		break;
+	case MDB_RTR_TYPE_PERM:
+		p->multicast_router = MDB_RTR_TYPE_PERM;
 		del_timer(&p->multicast_router_timer);
-
-		if (val == 0)
-			break;
-
 		br_multicast_add_router(br, p);
 		break;
+	default:
+		goto unlock;
 	}
-
+	err = 0;
+unlock:
 	spin_unlock(&br->multicast_lock);
 
 	return err;
-- 
cgit v0.10.2


From 4950cfd1e6a85b5e079a78f19792e6ec54199170 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 26 Feb 2016 21:20:02 +0100
Subject: bridge: mcast: do nothing if port's multicast_router is set to the
 same val

This is needed for the upcoming temporary port router. There's no point
to go through the logic if the value is the same.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 71c109b..f1140cf 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -1853,7 +1853,10 @@ int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val)
 	int err = -EINVAL;
 
 	spin_lock(&br->multicast_lock);
-
+	if (p->multicast_router == val) {
+		err = 0;
+		goto unlock;
+	}
 	switch (val) {
 	case MDB_RTR_TYPE_DISABLED:
 		p->multicast_router = MDB_RTR_TYPE_DISABLED;
-- 
cgit v0.10.2


From a55d8246abcc910346771175b521ee2bce5a69b3 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 26 Feb 2016 21:20:03 +0100
Subject: bridge: mcast: add support for temporary port router

Add support for a temporary router port which doesn't depend only on the
incoming query. It can be refreshed if set to the same value, which is
a no-op for the rest.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index e47f3bc..74ee03a 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -182,6 +182,7 @@ enum {
 	MDB_RTR_TYPE_DISABLED,
 	MDB_RTR_TYPE_TEMP_QUERY,
 	MDB_RTR_TYPE_PERM,
+	MDB_RTR_TYPE_TEMP
 };
 
 enum {
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index f1140cf..a4c15df 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -759,13 +759,17 @@ static void br_multicast_router_expired(unsigned long data)
 	struct net_bridge *br = port->br;
 
 	spin_lock(&br->multicast_lock);
-	if (port->multicast_router != MDB_RTR_TYPE_TEMP_QUERY ||
+	if (port->multicast_router == MDB_RTR_TYPE_DISABLED ||
+	    port->multicast_router == MDB_RTR_TYPE_PERM ||
 	    timer_pending(&port->multicast_router_timer) ||
 	    hlist_unhashed(&port->rlist))
 		goto out;
 
 	hlist_del_init_rcu(&port->rlist);
 	br_rtr_notify(br->dev, port, RTM_DELMDB);
+	/* Don't allow timer refresh if the router expired */
+	if (port->multicast_router == MDB_RTR_TYPE_TEMP)
+		port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY;
 
 out:
 	spin_unlock(&br->multicast_lock);
@@ -981,6 +985,9 @@ void br_multicast_disable_port(struct net_bridge_port *port)
 	if (!hlist_unhashed(&port->rlist)) {
 		hlist_del_init_rcu(&port->rlist);
 		br_rtr_notify(br->dev, port, RTM_DELMDB);
+		/* Don't allow timer refresh if disabling */
+		if (port->multicast_router == MDB_RTR_TYPE_TEMP)
+			port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY;
 	}
 	del_timer(&port->multicast_router_timer);
 	del_timer(&port->ip4_own_query.timer);
@@ -1234,7 +1241,8 @@ static void br_multicast_mark_router(struct net_bridge *br,
 		return;
 	}
 
-	if (port->multicast_router != MDB_RTR_TYPE_TEMP_QUERY)
+	if (port->multicast_router == MDB_RTR_TYPE_DISABLED ||
+	    port->multicast_router == MDB_RTR_TYPE_PERM)
 		return;
 
 	br_multicast_add_router(br, port);
@@ -1850,10 +1858,15 @@ static void __del_port_router(struct net_bridge_port *p)
 int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val)
 {
 	struct net_bridge *br = p->br;
+	unsigned long now = jiffies;
 	int err = -EINVAL;
 
 	spin_lock(&br->multicast_lock);
 	if (p->multicast_router == val) {
+		/* Refresh the temp router port timer */
+		if (p->multicast_router == MDB_RTR_TYPE_TEMP)
+			mod_timer(&p->multicast_router_timer,
+				  now + br->multicast_querier_interval);
 		err = 0;
 		goto unlock;
 	}
@@ -1872,6 +1885,10 @@ int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val)
 		del_timer(&p->multicast_router_timer);
 		br_multicast_add_router(br, p);
 		break;
+	case MDB_RTR_TYPE_TEMP:
+		p->multicast_router = MDB_RTR_TYPE_TEMP;
+		br_multicast_mark_router(br, p);
+		break;
 	default:
 		goto unlock;
 	}
-- 
cgit v0.10.2


From 59f78f9f6c2e80dcf0f520be85b660f856217b79 Mon Sep 17 00:00:00 2001
From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 26 Feb 2016 21:20:04 +0100
Subject: bridge: mcast: add support for more router port information dumping

Allow for more multicast router port information to be dumped such as
timer and type attributes. For that that purpose we need to extend the
MDBA_ROUTER_PORT attribute similar to how it was done for the mdb entries
recently. The new format is thus:
[MDBA_ROUTER_PORT] = { <- nested attribute
    u32 ifindex <- router port ifindex for user-space compatibility
    [MDBA_ROUTER_PATTR attributes]
}
This way it remains compatible with older users (they'll simply retrieve
the u32 in the beginning) and new users can parse the remaining
attributes. It would also allow to add future extensions to the router
port without breaking compatibility.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 74ee03a..0536eef 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -144,7 +144,10 @@ struct bridge_vlan_info {
  *     }
  * }
  * [MDBA_ROUTER] = {
- *    [MDBA_ROUTER_PORT]
+ *    [MDBA_ROUTER_PORT] = {
+ *        u32 ifindex
+ *        [MDBA_ROUTER_PATTR attributes]
+ *    }
  * }
  */
 enum {
@@ -192,6 +195,15 @@ enum {
 };
 #define MDBA_ROUTER_MAX (__MDBA_ROUTER_MAX - 1)
 
+/* router port attributes */
+enum {
+	MDBA_ROUTER_PATTR_UNSPEC,
+	MDBA_ROUTER_PATTR_TIMER,
+	MDBA_ROUTER_PATTR_TYPE,
+	__MDBA_ROUTER_PATTR_MAX
+};
+#define MDBA_ROUTER_PATTR_MAX (__MDBA_ROUTER_PATTR_MAX - 1)
+
 struct br_port_msg {
 	__u8  family;
 	__u32 ifindex;
diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 73786e2..253bc77 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -20,7 +20,7 @@ static int br_rports_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 {
 	struct net_bridge *br = netdev_priv(dev);
 	struct net_bridge_port *p;
-	struct nlattr *nest;
+	struct nlattr *nest, *port_nest;
 
 	if (!br->multicast_router || hlist_empty(&br->router_list))
 		return 0;
@@ -30,8 +30,20 @@ static int br_rports_fill_info(struct sk_buff *skb, struct netlink_callback *cb,
 		return -EMSGSIZE;
 
 	hlist_for_each_entry_rcu(p, &br->router_list, rlist) {
-		if (p && nla_put_u32(skb, MDBA_ROUTER_PORT, p->dev->ifindex))
+		if (!p)
+			continue;
+		port_nest = nla_nest_start(skb, MDBA_ROUTER_PORT);
+		if (!port_nest)
 			goto fail;
+		if (nla_put_nohdr(skb, sizeof(u32), &p->dev->ifindex) ||
+		    nla_put_u32(skb, MDBA_ROUTER_PATTR_TIMER,
+				br_timer_value(&p->multicast_router_timer)) ||
+		    nla_put_u8(skb, MDBA_ROUTER_PATTR_TYPE,
+			       p->multicast_router)) {
+			nla_nest_cancel(skb, port_nest);
+			goto fail;
+		}
+		nla_nest_end(skb, port_nest);
 	}
 
 	nla_nest_end(skb, nest);
-- 
cgit v0.10.2


From ef6980b6becb1afd9d82a4f043749a10ae81bf14 Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <jhs@mojatatu.com>
Date: Sat, 27 Feb 2016 08:08:54 -0500
Subject: introduce IFE action

This action allows for a sending side to encapsulate arbitrary metadata
which is decapsulated by the receiving end.
The sender runs in encoding mode and the receiver in decode mode.
Both sender and receiver must specify the same ethertype.
At some point we hope to have a registered ethertype and we'll
then provide a default so the user doesnt have to specify it.
For now we enforce the user specify it.

Lets show example usage where we encode icmp from a sender towards
a receiver with an skbmark of 17; both sender and receiver use
ethertype of 0xdead to interop.

YYYY: Lets start with Receiver-side policy config:
xxx: add an ingress qdisc
sudo tc qdisc add dev $ETH ingress

xxx: any packets with ethertype 0xdead will be subjected to ife decoding
xxx: we then restart the classification so we can match on icmp at prio 3
sudo $TC filter add dev $ETH parent ffff: prio 2 protocol 0xdead \
u32 match u32 0 0 flowid 1:1 \
action ife decode reclassify

xxx: on restarting the classification from above if it was an icmp
xxx: packet, then match it here and continue to the next rule at prio 4
xxx: which will match based on skb mark of 17
sudo tc filter add dev $ETH parent ffff: prio 3 protocol ip \
u32 match ip protocol 1 0xff flowid 1:1 \
action continue

xxx: match on skbmark of 0x11 (decimal 17) and accept
sudo tc filter add dev $ETH parent ffff: prio 4 protocol ip \
handle 0x11 fw flowid 1:1 \
action ok

xxx: Lets show the decoding policy
sudo tc -s filter ls dev $ETH parent ffff: protocol 0xdead
xxx:
filter pref 2 u32
filter pref 2 u32 fh 800: ht divisor 1
filter pref 2 u32 fh 800::800 order 2048 key ht 800 bkt 0 flowid 1:1  (rule hit 0 success 0)
  match 00000000/00000000 at 0 (success 0 )
        action order 1: ife decode action reclassify
         index 1 ref 1 bind 1 installed 14 sec used 14 sec
         type: 0x0
         Metadata: allow mark allow hash allow prio allow qmap
        Action statistics:
        Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
        backlog 0b 0p requeues 0
xxx:
Observe that above lists all metadatum it can decode. Typically these
submodules will already be compiled into a monolithic kernel or
loaded as modules

YYYY: Lets show the sender side now ..

xxx: Add an egress qdisc on the sender netdev
sudo tc qdisc add dev $ETH root handle 1: prio
xxx:
xxx: Match all icmp packets to 192.168.122.237/24, then
xxx: tag the packet with skb mark of decimal 17, then
xxx: Encode it with:
xxx:	ethertype 0xdead
xxx:	add skb->mark to whitelist of metadatum to send
xxx:	rewrite target dst MAC address to 02:15:15:15:15:15
xxx:
sudo $TC filter add dev $ETH parent 1: protocol ip prio 10  u32 \
match ip dst 192.168.122.237/24 \
match ip protocol 1 0xff \
flowid 1:2 \
action skbedit mark 17 \
action ife encode \
type 0xDEAD \
allow mark \
dst 02:15:15:15:15:15

xxx: Lets show the encoding policy
sudo tc -s filter ls dev $ETH parent 1: protocol ip
xxx:
filter pref 10 u32
filter pref 10 u32 fh 800: ht divisor 1
filter pref 10 u32 fh 800::800 order 2048 key ht 800 bkt 0 flowid 1:2  (rule hit 0 success 0)
  match c0a87aed/ffffffff at 16 (success 0 )
  match 00010000/00ff0000 at 8 (success 0 )

	action order 1:  skbedit mark 17
	 index 6 ref 1 bind 1
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0

	action order 2: ife encode action pipe
	 index 3 ref 1 bind 1
	 dst MAC: 02:15:15:15:15:15 type: 0xDEAD
 	 Metadata: allow mark
 	Action statistics:
	Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
	backlog 0b 0p requeues 0
xxx:

test by sending ping from sender to destination

Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/tc_act/tc_ife.h b/include/net/tc_act/tc_ife.h
new file mode 100644
index 0000000..dc9a09a
--- /dev/null
+++ b/include/net/tc_act/tc_ife.h
@@ -0,0 +1,61 @@
+#ifndef __NET_TC_IFE_H
+#define __NET_TC_IFE_H
+
+#include <net/act_api.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/module.h>
+
+#define IFE_METAHDRLEN 2
+struct tcf_ife_info {
+	struct tcf_common common;
+	u8 eth_dst[ETH_ALEN];
+	u8 eth_src[ETH_ALEN];
+	u16 eth_type;
+	u16 flags;
+	/* list of metaids allowed */
+	struct list_head metalist;
+};
+#define to_ife(a) \
+	container_of(a->priv, struct tcf_ife_info, common)
+
+struct tcf_meta_info {
+	const struct tcf_meta_ops *ops;
+	void *metaval;
+	u16 metaid;
+	struct list_head metalist;
+};
+
+struct tcf_meta_ops {
+	u16 metaid; /*Maintainer provided ID */
+	u16 metatype; /*netlink attribute type (look at net/netlink.h) */
+	const char *name;
+	const char *synopsis;
+	struct list_head list;
+	int	(*check_presence)(struct sk_buff *, struct tcf_meta_info *);
+	int	(*encode)(struct sk_buff *, void *, struct tcf_meta_info *);
+	int	(*decode)(struct sk_buff *, void *, u16 len);
+	int	(*get)(struct sk_buff *skb, struct tcf_meta_info *mi);
+	int	(*alloc)(struct tcf_meta_info *, void *);
+	void	(*release)(struct tcf_meta_info *);
+	int	(*validate)(void *val, int len);
+	struct module	*owner;
+};
+
+#define MODULE_ALIAS_IFE_META(metan)   MODULE_ALIAS("ifemeta" __stringify_1(metan))
+
+int ife_get_meta_u32(struct sk_buff *skb, struct tcf_meta_info *mi);
+int ife_get_meta_u16(struct sk_buff *skb, struct tcf_meta_info *mi);
+int ife_tlv_meta_encode(void *skbdata, u16 attrtype, u16 dlen,
+			const void *dval);
+int ife_alloc_meta_u32(struct tcf_meta_info *mi, void *metaval);
+int ife_alloc_meta_u16(struct tcf_meta_info *mi, void *metaval);
+int ife_check_meta_u32(u32 metaval, struct tcf_meta_info *mi);
+int ife_encode_meta_u32(u32 metaval, void *skbdata, struct tcf_meta_info *mi);
+int ife_validate_meta_u32(void *val, int len);
+int ife_validate_meta_u16(void *val, int len);
+void ife_release_meta_gen(struct tcf_meta_info *mi);
+int register_ife_op(struct tcf_meta_ops *mops);
+int unregister_ife_op(struct tcf_meta_ops *mops);
+
+#endif /* __NET_TC_IFE_H */
diff --git a/include/uapi/linux/tc_act/tc_ife.h b/include/uapi/linux/tc_act/tc_ife.h
new file mode 100644
index 0000000..d648ff6
--- /dev/null
+++ b/include/uapi/linux/tc_act/tc_ife.h
@@ -0,0 +1,38 @@
+#ifndef __UAPI_TC_IFE_H
+#define __UAPI_TC_IFE_H
+
+#include <linux/types.h>
+#include <linux/pkt_cls.h>
+
+#define TCA_ACT_IFE 25
+/* Flag bits for now just encoding/decoding; mutually exclusive */
+#define IFE_ENCODE 1
+#define IFE_DECODE 0
+
+struct tc_ife {
+	tc_gen;
+	__u16 flags;
+};
+
+/*XXX: We need to encode the total number of bytes consumed */
+enum {
+	TCA_IFE_UNSPEC,
+	TCA_IFE_PARMS,
+	TCA_IFE_TM,
+	TCA_IFE_DMAC,
+	TCA_IFE_SMAC,
+	TCA_IFE_TYPE,
+	TCA_IFE_METALST,
+	__TCA_IFE_MAX
+};
+#define TCA_IFE_MAX (__TCA_IFE_MAX - 1)
+
+#define IFE_META_SKBMARK 1
+#define IFE_META_HASHID 2
+#define	IFE_META_PRIO 3
+#define	IFE_META_QMAP 4
+/*Can be overridden at runtime by module option*/
+#define	__IFE_META_MAX 5
+#define IFE_META_MAX (__IFE_META_MAX - 1)
+
+#endif
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 8283082..4d48ef5 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -739,6 +739,18 @@ config NET_ACT_CONNMARK
 	  To compile this code as a module, choose M here: the
 	  module will be called act_connmark.
 
+config NET_ACT_IFE
+        tristate "Inter-FE action based on IETF ForCES InterFE LFB"
+        depends on NET_CLS_ACT
+        ---help---
+	  Say Y here to allow for sourcing and terminating metadata
+	  For details refer to netdev01 paper:
+	  "Distributing Linux Traffic Control Classifier-Action Subsystem"
+	   Authors: Jamal Hadi Salim and Damascene M. Joachimpillai
+
+	  To compile this code as a module, choose M here: the
+	  module will be called act_ife.
+
 config NET_CLS_IND
 	bool "Incoming device classification"
 	depends on NET_CLS_U32 || NET_CLS_FW
diff --git a/net/sched/Makefile b/net/sched/Makefile
index 690c1689..3d17667 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_NET_ACT_CSUM)	+= act_csum.o
 obj-$(CONFIG_NET_ACT_VLAN)	+= act_vlan.o
 obj-$(CONFIG_NET_ACT_BPF)	+= act_bpf.o
 obj-$(CONFIG_NET_ACT_CONNMARK)	+= act_connmark.o
+obj-$(CONFIG_NET_ACT_IFE)	+= act_ife.o
 obj-$(CONFIG_NET_SCH_FIFO)	+= sch_fifo.o
 obj-$(CONFIG_NET_SCH_CBQ)	+= sch_cbq.o
 obj-$(CONFIG_NET_SCH_HTB)	+= sch_htb.o
diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c
new file mode 100644
index 0000000..6e7ec25
--- /dev/null
+++ b/net/sched/act_ife.c
@@ -0,0 +1,870 @@
+/*
+ * net/sched/ife.c	Inter-FE action based on ForCES WG InterFE LFB
+ *
+ *		Refer to:
+ *		draft-ietf-forces-interfelfb-03
+ *		and
+ *		netdev01 paper:
+ *		"Distributing Linux Traffic Control Classifier-Action
+ *		Subsystem"
+ *		Authors: Jamal Hadi Salim and Damascene M. Joachimpillai
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * copyright Jamal Hadi Salim (2015)
+ *
+*/
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/skbuff.h>
+#include <linux/rtnetlink.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <net/net_namespace.h>
+#include <net/netlink.h>
+#include <net/pkt_sched.h>
+#include <uapi/linux/tc_act/tc_ife.h>
+#include <net/tc_act/tc_ife.h>
+#include <linux/etherdevice.h>
+
+#define IFE_TAB_MASK 15
+
+static int ife_net_id;
+static int max_metacnt = IFE_META_MAX + 1;
+
+static const struct nla_policy ife_policy[TCA_IFE_MAX + 1] = {
+	[TCA_IFE_PARMS] = { .len = sizeof(struct tc_ife)},
+	[TCA_IFE_DMAC] = { .len = ETH_ALEN},
+	[TCA_IFE_SMAC] = { .len = ETH_ALEN},
+	[TCA_IFE_TYPE] = { .type = NLA_U16},
+};
+
+/* Caller takes care of presenting data in network order
+*/
+int ife_tlv_meta_encode(void *skbdata, u16 attrtype, u16 dlen, const void *dval)
+{
+	u32 *tlv = (u32 *)(skbdata);
+	u16 totlen = nla_total_size(dlen);	/*alignment + hdr */
+	char *dptr = (char *)tlv + NLA_HDRLEN;
+	u32 htlv = attrtype << 16 | totlen;
+
+	*tlv = htonl(htlv);
+	memset(dptr, 0, totlen - NLA_HDRLEN);
+	memcpy(dptr, dval, dlen);
+
+	return totlen;
+}
+EXPORT_SYMBOL_GPL(ife_tlv_meta_encode);
+
+int ife_get_meta_u32(struct sk_buff *skb, struct tcf_meta_info *mi)
+{
+	if (mi->metaval)
+		return nla_put_u32(skb, mi->metaid, *(u32 *)mi->metaval);
+	else
+		return nla_put(skb, mi->metaid, 0, NULL);
+}
+EXPORT_SYMBOL_GPL(ife_get_meta_u32);
+
+int ife_check_meta_u32(u32 metaval, struct tcf_meta_info *mi)
+{
+	if (metaval || mi->metaval)
+		return 8; /* T+L+V == 2+2+4 */
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ife_check_meta_u32);
+
+int ife_encode_meta_u32(u32 metaval, void *skbdata, struct tcf_meta_info *mi)
+{
+	u32 edata = metaval;
+
+	if (mi->metaval)
+		edata = *(u32 *)mi->metaval;
+	else if (metaval)
+		edata = metaval;
+
+	if (!edata) /* will not encode */
+		return 0;
+
+	edata = htonl(edata);
+	return ife_tlv_meta_encode(skbdata, mi->metaid, 4, &edata);
+}
+EXPORT_SYMBOL_GPL(ife_encode_meta_u32);
+
+int ife_get_meta_u16(struct sk_buff *skb, struct tcf_meta_info *mi)
+{
+	if (mi->metaval)
+		return nla_put_u16(skb, mi->metaid, *(u16 *)mi->metaval);
+	else
+		return nla_put(skb, mi->metaid, 0, NULL);
+}
+EXPORT_SYMBOL_GPL(ife_get_meta_u16);
+
+int ife_alloc_meta_u32(struct tcf_meta_info *mi, void *metaval)
+{
+	mi->metaval = kmemdup(&metaval, sizeof(u32), GFP_KERNEL);
+	if (!mi->metaval)
+		return -ENOMEM;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ife_alloc_meta_u32);
+
+int ife_alloc_meta_u16(struct tcf_meta_info *mi, void *metaval)
+{
+	mi->metaval = kmemdup(&metaval, sizeof(u16), GFP_KERNEL);
+	if (!mi->metaval)
+		return -ENOMEM;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ife_alloc_meta_u16);
+
+void ife_release_meta_gen(struct tcf_meta_info *mi)
+{
+	kfree(mi->metaval);
+}
+EXPORT_SYMBOL_GPL(ife_release_meta_gen);
+
+int ife_validate_meta_u32(void *val, int len)
+{
+	if (len == 4)
+		return 0;
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(ife_validate_meta_u32);
+
+int ife_validate_meta_u16(void *val, int len)
+{
+	/* length will include padding */
+	if (len == NLA_ALIGN(2))
+		return 0;
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(ife_validate_meta_u16);
+
+static LIST_HEAD(ifeoplist);
+static DEFINE_RWLOCK(ife_mod_lock);
+
+static struct tcf_meta_ops *find_ife_oplist(u16 metaid)
+{
+	struct tcf_meta_ops *o;
+
+	read_lock(&ife_mod_lock);
+	list_for_each_entry(o, &ifeoplist, list) {
+		if (o->metaid == metaid) {
+			if (!try_module_get(o->owner))
+				o = NULL;
+			read_unlock(&ife_mod_lock);
+			return o;
+		}
+	}
+	read_unlock(&ife_mod_lock);
+
+	return NULL;
+}
+
+int register_ife_op(struct tcf_meta_ops *mops)
+{
+	struct tcf_meta_ops *m;
+
+	if (!mops->metaid || !mops->metatype || !mops->name ||
+	    !mops->check_presence || !mops->encode || !mops->decode ||
+	    !mops->get || !mops->alloc)
+		return -EINVAL;
+
+	write_lock(&ife_mod_lock);
+
+	list_for_each_entry(m, &ifeoplist, list) {
+		if (m->metaid == mops->metaid ||
+		    (strcmp(mops->name, m->name) == 0)) {
+			write_unlock(&ife_mod_lock);
+			return -EEXIST;
+		}
+	}
+
+	if (!mops->release)
+		mops->release = ife_release_meta_gen;
+
+	list_add_tail(&mops->list, &ifeoplist);
+	write_unlock(&ife_mod_lock);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(unregister_ife_op);
+
+int unregister_ife_op(struct tcf_meta_ops *mops)
+{
+	struct tcf_meta_ops *m;
+	int err = -ENOENT;
+
+	write_lock(&ife_mod_lock);
+	list_for_each_entry(m, &ifeoplist, list) {
+		if (m->metaid == mops->metaid) {
+			list_del(&mops->list);
+			err = 0;
+			break;
+		}
+	}
+	write_unlock(&ife_mod_lock);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(register_ife_op);
+
+static int ife_validate_metatype(struct tcf_meta_ops *ops, void *val, int len)
+{
+	int ret = 0;
+	/* XXX: unfortunately cant use nla_policy at this point
+	* because a length of 0 is valid in the case of
+	* "allow". "use" semantics do enforce for proper
+	* length and i couldve use nla_policy but it makes it hard
+	* to use it just for that..
+	*/
+	if (ops->validate)
+		return ops->validate(val, len);
+
+	if (ops->metatype == NLA_U32)
+		ret = ife_validate_meta_u32(val, len);
+	else if (ops->metatype == NLA_U16)
+		ret = ife_validate_meta_u16(val, len);
+
+	return ret;
+}
+
+/* called when adding new meta information
+ * under ife->tcf_lock
+*/
+static int load_metaops_and_vet(struct tcf_ife_info *ife, u32 metaid,
+				void *val, int len)
+{
+	struct tcf_meta_ops *ops = find_ife_oplist(metaid);
+	int ret = 0;
+
+	if (!ops) {
+		ret = -ENOENT;
+#ifdef CONFIG_MODULES
+		spin_unlock_bh(&ife->tcf_lock);
+		rtnl_unlock();
+		request_module("ifemeta%u", metaid);
+		rtnl_lock();
+		spin_lock_bh(&ife->tcf_lock);
+		ops = find_ife_oplist(metaid);
+#endif
+	}
+
+	if (ops) {
+		ret = 0;
+		if (len)
+			ret = ife_validate_metatype(ops, val, len);
+
+		module_put(ops->owner);
+	}
+
+	return ret;
+}
+
+/* called when adding new meta information
+ * under ife->tcf_lock
+*/
+static int add_metainfo(struct tcf_ife_info *ife, u32 metaid, void *metaval,
+			int len)
+{
+	struct tcf_meta_info *mi = NULL;
+	struct tcf_meta_ops *ops = find_ife_oplist(metaid);
+	int ret = 0;
+
+	if (!ops)
+		return -ENOENT;
+
+	mi = kzalloc(sizeof(*mi), GFP_KERNEL);
+	if (!mi) {
+		/*put back what find_ife_oplist took */
+		module_put(ops->owner);
+		return -ENOMEM;
+	}
+
+	mi->metaid = metaid;
+	mi->ops = ops;
+	if (len > 0) {
+		ret = ops->alloc(mi, metaval);
+		if (ret != 0) {
+			kfree(mi);
+			module_put(ops->owner);
+			return ret;
+		}
+	}
+
+	list_add_tail(&mi->metalist, &ife->metalist);
+
+	return ret;
+}
+
+static int use_all_metadata(struct tcf_ife_info *ife)
+{
+	struct tcf_meta_ops *o;
+	int rc = 0;
+	int installed = 0;
+
+	list_for_each_entry(o, &ifeoplist, list) {
+		rc = add_metainfo(ife, o->metaid, NULL, 0);
+		if (rc == 0)
+			installed += 1;
+	}
+
+	if (installed)
+		return 0;
+	else
+		return -EINVAL;
+}
+
+static int dump_metalist(struct sk_buff *skb, struct tcf_ife_info *ife)
+{
+	struct tcf_meta_info *e;
+	struct nlattr *nest;
+	unsigned char *b = skb_tail_pointer(skb);
+	int total_encoded = 0;
+
+	/*can only happen on decode */
+	if (list_empty(&ife->metalist))
+		return 0;
+
+	nest = nla_nest_start(skb, TCA_IFE_METALST);
+	if (!nest)
+		goto out_nlmsg_trim;
+
+	list_for_each_entry(e, &ife->metalist, metalist) {
+		if (!e->ops->get(skb, e))
+			total_encoded += 1;
+	}
+
+	if (!total_encoded)
+		goto out_nlmsg_trim;
+
+	nla_nest_end(skb, nest);
+
+	return 0;
+
+out_nlmsg_trim:
+	nlmsg_trim(skb, b);
+	return -1;
+}
+
+/* under ife->tcf_lock */
+static void _tcf_ife_cleanup(struct tc_action *a, int bind)
+{
+	struct tcf_ife_info *ife = a->priv;
+	struct tcf_meta_info *e, *n;
+
+	list_for_each_entry_safe(e, n, &ife->metalist, metalist) {
+		module_put(e->ops->owner);
+		list_del(&e->metalist);
+		if (e->metaval) {
+			if (e->ops->release)
+				e->ops->release(e);
+			else
+				kfree(e->metaval);
+		}
+		kfree(e);
+	}
+}
+
+static void tcf_ife_cleanup(struct tc_action *a, int bind)
+{
+	struct tcf_ife_info *ife = a->priv;
+
+	spin_lock_bh(&ife->tcf_lock);
+	_tcf_ife_cleanup(a, bind);
+	spin_unlock_bh(&ife->tcf_lock);
+}
+
+/* under ife->tcf_lock */
+static int populate_metalist(struct tcf_ife_info *ife, struct nlattr **tb)
+{
+	int len = 0;
+	int rc = 0;
+	int i = 0;
+	void *val;
+
+	for (i = 1; i < max_metacnt; i++) {
+		if (tb[i]) {
+			val = nla_data(tb[i]);
+			len = nla_len(tb[i]);
+
+			rc = load_metaops_and_vet(ife, i, val, len);
+			if (rc != 0)
+				return rc;
+
+			rc = add_metainfo(ife, i, val, len);
+			if (rc)
+				return rc;
+		}
+	}
+
+	return rc;
+}
+
+static int tcf_ife_init(struct net *net, struct nlattr *nla,
+			struct nlattr *est, struct tc_action *a,
+			int ovr, int bind)
+{
+	struct tc_action_net *tn = net_generic(net, ife_net_id);
+	struct nlattr *tb[TCA_IFE_MAX + 1];
+	struct nlattr *tb2[IFE_META_MAX + 1];
+	struct tcf_ife_info *ife;
+	struct tc_ife *parm;
+	u16 ife_type = 0;
+	u8 *daddr = NULL;
+	u8 *saddr = NULL;
+	int ret = 0;
+	int err;
+
+	err = nla_parse_nested(tb, TCA_IFE_MAX, nla, ife_policy);
+	if (err < 0)
+		return err;
+
+	if (!tb[TCA_IFE_PARMS])
+		return -EINVAL;
+
+	parm = nla_data(tb[TCA_IFE_PARMS]);
+
+	if (parm->flags & IFE_ENCODE) {
+		/* Until we get issued the ethertype, we cant have
+		 * a default..
+		**/
+		if (!tb[TCA_IFE_TYPE]) {
+			pr_info("You MUST pass etherype for encoding\n");
+			return -EINVAL;
+		}
+	}
+
+	if (!tcf_hash_check(tn, parm->index, a, bind)) {
+		ret = tcf_hash_create(tn, parm->index, est, a, sizeof(*ife),
+				      bind, false);
+		if (ret)
+			return ret;
+		ret = ACT_P_CREATED;
+	} else {
+		if (bind)	/* dont override defaults */
+			return 0;
+		tcf_hash_release(a, bind);
+		if (!ovr)
+			return -EEXIST;
+	}
+
+	ife = to_ife(a);
+	ife->flags = parm->flags;
+
+	if (parm->flags & IFE_ENCODE) {
+		ife_type = nla_get_u16(tb[TCA_IFE_TYPE]);
+		if (tb[TCA_IFE_DMAC])
+			daddr = nla_data(tb[TCA_IFE_DMAC]);
+		if (tb[TCA_IFE_SMAC])
+			saddr = nla_data(tb[TCA_IFE_SMAC]);
+	}
+
+	spin_lock_bh(&ife->tcf_lock);
+	ife->tcf_action = parm->action;
+
+	if (parm->flags & IFE_ENCODE) {
+		if (daddr)
+			ether_addr_copy(ife->eth_dst, daddr);
+		else
+			eth_zero_addr(ife->eth_dst);
+
+		if (saddr)
+			ether_addr_copy(ife->eth_src, saddr);
+		else
+			eth_zero_addr(ife->eth_src);
+
+		ife->eth_type = ife_type;
+	}
+
+	if (ret == ACT_P_CREATED)
+		INIT_LIST_HEAD(&ife->metalist);
+
+	if (tb[TCA_IFE_METALST]) {
+		err = nla_parse_nested(tb2, IFE_META_MAX, tb[TCA_IFE_METALST],
+				       NULL);
+		if (err) {
+metadata_parse_err:
+			if (ret == ACT_P_CREATED)
+				_tcf_ife_cleanup(a, bind);
+
+			spin_unlock_bh(&ife->tcf_lock);
+			return err;
+		}
+
+		err = populate_metalist(ife, tb2);
+		if (err)
+			goto metadata_parse_err;
+
+	} else {
+		/* if no passed metadata allow list or passed allow-all
+		 * then here we process by adding as many supported metadatum
+		 * as we can. You better have at least one else we are
+		 * going to bail out
+		 */
+		err = use_all_metadata(ife);
+		if (err) {
+			if (ret == ACT_P_CREATED)
+				_tcf_ife_cleanup(a, bind);
+
+			spin_unlock_bh(&ife->tcf_lock);
+			return err;
+		}
+	}
+
+	spin_unlock_bh(&ife->tcf_lock);
+
+	if (ret == ACT_P_CREATED)
+		tcf_hash_insert(tn, a);
+
+	return ret;
+}
+
+static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind,
+			int ref)
+{
+	unsigned char *b = skb_tail_pointer(skb);
+	struct tcf_ife_info *ife = a->priv;
+	struct tc_ife opt = {
+		.index = ife->tcf_index,
+		.refcnt = ife->tcf_refcnt - ref,
+		.bindcnt = ife->tcf_bindcnt - bind,
+		.action = ife->tcf_action,
+		.flags = ife->flags,
+	};
+	struct tcf_t t;
+
+	if (nla_put(skb, TCA_IFE_PARMS, sizeof(opt), &opt))
+		goto nla_put_failure;
+
+	t.install = jiffies_to_clock_t(jiffies - ife->tcf_tm.install);
+	t.lastuse = jiffies_to_clock_t(jiffies - ife->tcf_tm.lastuse);
+	t.expires = jiffies_to_clock_t(ife->tcf_tm.expires);
+	if (nla_put(skb, TCA_IFE_TM, sizeof(t), &t))
+		goto nla_put_failure;
+
+	if (!is_zero_ether_addr(ife->eth_dst)) {
+		if (nla_put(skb, TCA_IFE_DMAC, ETH_ALEN, ife->eth_dst))
+			goto nla_put_failure;
+	}
+
+	if (!is_zero_ether_addr(ife->eth_src)) {
+		if (nla_put(skb, TCA_IFE_SMAC, ETH_ALEN, ife->eth_src))
+			goto nla_put_failure;
+	}
+
+	if (nla_put(skb, TCA_IFE_TYPE, 2, &ife->eth_type))
+		goto nla_put_failure;
+
+	if (dump_metalist(skb, ife)) {
+		/*ignore failure to dump metalist */
+		pr_info("Failed to dump metalist\n");
+	}
+
+	return skb->len;
+
+nla_put_failure:
+	nlmsg_trim(skb, b);
+	return -1;
+}
+
+int find_decode_metaid(struct sk_buff *skb, struct tcf_ife_info *ife,
+		       u16 metaid, u16 mlen, void *mdata)
+{
+	struct tcf_meta_info *e;
+
+	/* XXX: use hash to speed up */
+	list_for_each_entry(e, &ife->metalist, metalist) {
+		if (metaid == e->metaid) {
+			if (e->ops) {
+				/* We check for decode presence already */
+				return e->ops->decode(skb, mdata, mlen);
+			}
+		}
+	}
+
+	return 0;
+}
+
+struct ifeheadr {
+	__be16 metalen;
+	u8 tlv_data[];
+};
+
+struct meta_tlvhdr {
+	__be16 type;
+	__be16 len;
+};
+
+static int tcf_ife_decode(struct sk_buff *skb, const struct tc_action *a,
+			  struct tcf_result *res)
+{
+	struct tcf_ife_info *ife = a->priv;
+	int action = ife->tcf_action;
+	struct ifeheadr *ifehdr = (struct ifeheadr *)skb->data;
+	u16 ifehdrln = ifehdr->metalen;
+	struct meta_tlvhdr *tlv = (struct meta_tlvhdr *)(ifehdr->tlv_data);
+
+	spin_lock(&ife->tcf_lock);
+	bstats_update(&ife->tcf_bstats, skb);
+	ife->tcf_tm.lastuse = jiffies;
+	spin_unlock(&ife->tcf_lock);
+
+	ifehdrln = ntohs(ifehdrln);
+	if (unlikely(!pskb_may_pull(skb, ifehdrln))) {
+		spin_lock(&ife->tcf_lock);
+		ife->tcf_qstats.drops++;
+		spin_unlock(&ife->tcf_lock);
+		return TC_ACT_SHOT;
+	}
+
+	skb_set_mac_header(skb, ifehdrln);
+	__skb_pull(skb, ifehdrln);
+	skb->protocol = eth_type_trans(skb, skb->dev);
+	ifehdrln -= IFE_METAHDRLEN;
+
+	while (ifehdrln > 0) {
+		u8 *tlvdata = (u8 *)tlv;
+		u16 mtype = tlv->type;
+		u16 mlen = tlv->len;
+
+		mtype = ntohs(mtype);
+		mlen = ntohs(mlen);
+
+		if (find_decode_metaid(skb, ife, mtype, (mlen - 4),
+				       (void *)(tlvdata + 4))) {
+			/* abuse overlimits to count when we receive metadata
+			 * but dont have an ops for it
+			 */
+			pr_info_ratelimited("Unknown metaid %d alnlen %d\n",
+					    mtype, mlen);
+			ife->tcf_qstats.overlimits++;
+		}
+
+		tlvdata += mlen;
+		ifehdrln -= mlen;
+		tlv = (struct meta_tlvhdr *)tlvdata;
+	}
+
+	skb_reset_network_header(skb);
+	return action;
+}
+
+/*XXX: check if we can do this at install time instead of current
+ * send data path
+**/
+static int ife_get_sz(struct sk_buff *skb, struct tcf_ife_info *ife)
+{
+	struct tcf_meta_info *e, *n;
+	int tot_run_sz = 0, run_sz = 0;
+
+	list_for_each_entry_safe(e, n, &ife->metalist, metalist) {
+		if (e->ops->check_presence) {
+			run_sz = e->ops->check_presence(skb, e);
+			tot_run_sz += run_sz;
+		}
+	}
+
+	return tot_run_sz;
+}
+
+static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a,
+			  struct tcf_result *res)
+{
+	struct tcf_ife_info *ife = a->priv;
+	int action = ife->tcf_action;
+	struct ethhdr *oethh;	/* outer ether header */
+	struct ethhdr *iethh;	/* inner eth header */
+	struct tcf_meta_info *e;
+	/*
+	   OUTERHDR:TOTMETALEN:{TLVHDR:Metadatum:TLVHDR..}:ORIGDATA
+	   where ORIGDATA = original ethernet header ...
+	 */
+	u16 metalen = ife_get_sz(skb, ife);
+	int hdrm = metalen + skb->dev->hard_header_len + IFE_METAHDRLEN;
+	unsigned int skboff = skb->dev->hard_header_len;
+	u32 at = G_TC_AT(skb->tc_verd);
+	int new_len = skb->len + hdrm;
+	bool exceed_mtu = false;
+	int err;
+
+	if (at & AT_EGRESS) {
+		if (new_len > skb->dev->mtu)
+			exceed_mtu = true;
+	}
+
+	spin_lock(&ife->tcf_lock);
+	bstats_update(&ife->tcf_bstats, skb);
+	ife->tcf_tm.lastuse = jiffies;
+
+	if (!metalen) {		/* no metadata to send */
+		/* abuse overlimits to count when we allow packet
+		 * with no metadata
+		 */
+		ife->tcf_qstats.overlimits++;
+		spin_unlock(&ife->tcf_lock);
+		return action;
+	}
+	/* could be stupid policy setup or mtu config
+	 * so lets be conservative.. */
+	if ((action == TC_ACT_SHOT) || exceed_mtu) {
+		ife->tcf_qstats.drops++;
+		spin_unlock(&ife->tcf_lock);
+		return TC_ACT_SHOT;
+	}
+
+	iethh = eth_hdr(skb);
+
+	err = skb_cow_head(skb, hdrm);
+	if (unlikely(err)) {
+		ife->tcf_qstats.drops++;
+		spin_unlock(&ife->tcf_lock);
+		return TC_ACT_SHOT;
+	}
+
+	if (!(at & AT_EGRESS))
+		skb_push(skb, skb->dev->hard_header_len);
+
+	__skb_push(skb, hdrm);
+	memcpy(skb->data, iethh, skb->mac_len);
+	skb_reset_mac_header(skb);
+	oethh = eth_hdr(skb);
+
+	/*total metadata length */
+	metalen += IFE_METAHDRLEN;
+	metalen = htons(metalen);
+	memcpy((skb->data + skboff), &metalen, IFE_METAHDRLEN);
+	skboff += IFE_METAHDRLEN;
+
+	/* XXX: we dont have a clever way of telling encode to
+	 * not repeat some of the computations that are done by
+	 * ops->presence_check...
+	 */
+	list_for_each_entry(e, &ife->metalist, metalist) {
+		if (e->ops->encode) {
+			err = e->ops->encode(skb, (void *)(skb->data + skboff),
+					     e);
+		}
+		if (err < 0) {
+			/* too corrupt to keep around if overwritten */
+			ife->tcf_qstats.drops++;
+			spin_unlock(&ife->tcf_lock);
+			return TC_ACT_SHOT;
+		}
+		skboff += err;
+	}
+
+	if (!is_zero_ether_addr(ife->eth_src))
+		ether_addr_copy(oethh->h_source, ife->eth_src);
+	else
+		ether_addr_copy(oethh->h_source, iethh->h_source);
+	if (!is_zero_ether_addr(ife->eth_dst))
+		ether_addr_copy(oethh->h_dest, ife->eth_dst);
+	else
+		ether_addr_copy(oethh->h_dest, iethh->h_dest);
+	oethh->h_proto = htons(ife->eth_type);
+
+	if (!(at & AT_EGRESS))
+		skb_pull(skb, skb->dev->hard_header_len);
+
+	spin_unlock(&ife->tcf_lock);
+
+	return action;
+}
+
+static int tcf_ife_act(struct sk_buff *skb, const struct tc_action *a,
+		       struct tcf_result *res)
+{
+	struct tcf_ife_info *ife = a->priv;
+
+	if (ife->flags & IFE_ENCODE)
+		return tcf_ife_encode(skb, a, res);
+
+	if (!(ife->flags & IFE_ENCODE))
+		return tcf_ife_decode(skb, a, res);
+
+	pr_info_ratelimited("unknown failure(policy neither de/encode\n");
+	spin_lock(&ife->tcf_lock);
+	bstats_update(&ife->tcf_bstats, skb);
+	ife->tcf_tm.lastuse = jiffies;
+	ife->tcf_qstats.drops++;
+	spin_unlock(&ife->tcf_lock);
+
+	return TC_ACT_SHOT;
+}
+
+static int tcf_ife_walker(struct net *net, struct sk_buff *skb,
+			  struct netlink_callback *cb, int type,
+			  struct tc_action *a)
+{
+	struct tc_action_net *tn = net_generic(net, ife_net_id);
+
+	return tcf_generic_walker(tn, skb, cb, type, a);
+}
+
+static int tcf_ife_search(struct net *net, struct tc_action *a, u32 index)
+{
+	struct tc_action_net *tn = net_generic(net, ife_net_id);
+
+	return tcf_hash_search(tn, a, index);
+}
+
+static struct tc_action_ops act_ife_ops = {
+	.kind = "ife",
+	.type = TCA_ACT_IFE,
+	.owner = THIS_MODULE,
+	.act = tcf_ife_act,
+	.dump = tcf_ife_dump,
+	.cleanup = tcf_ife_cleanup,
+	.init = tcf_ife_init,
+	.walk = tcf_ife_walker,
+	.lookup = tcf_ife_search,
+};
+
+static __net_init int ife_init_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, ife_net_id);
+
+	return tc_action_net_init(tn, &act_ife_ops, IFE_TAB_MASK);
+}
+
+static void __net_exit ife_exit_net(struct net *net)
+{
+	struct tc_action_net *tn = net_generic(net, ife_net_id);
+
+	tc_action_net_exit(tn);
+}
+
+static struct pernet_operations ife_net_ops = {
+	.init = ife_init_net,
+	.exit = ife_exit_net,
+	.id   = &ife_net_id,
+	.size = sizeof(struct tc_action_net),
+};
+
+static int __init ife_init_module(void)
+{
+	return tcf_register_action(&act_ife_ops, &ife_net_ops);
+}
+
+static void __exit ife_cleanup_module(void)
+{
+	tcf_unregister_action(&act_ife_ops, &ife_net_ops);
+}
+
+module_init(ife_init_module);
+module_exit(ife_cleanup_module);
+
+MODULE_AUTHOR("Jamal Hadi Salim(2015)");
+MODULE_DESCRIPTION("Inter-FE LFB action");
+MODULE_LICENSE("GPL");
-- 
cgit v0.10.2


From 084e2f6566d2a39c007ed6473f58b551a2eeefeb Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <jhs@mojatatu.com>
Date: Sat, 27 Feb 2016 08:08:55 -0500
Subject: Support to encoding decoding skb mark on IFE action

Example usage:
Set the skb using skbedit then allow it to be encoded

sudo tc qdisc add dev $ETH root handle 1: prio
sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \
u32 match ip protocol 1 0xff flowid 1:2 \
action skbedit mark 17 \
action ife encode \
allow mark \
dst 02:15:15:15:15:15

Note: You dont need the skbedit action if you are already encoding the
skb mark earlier. A zero skb mark, when seen, will not be encoded.

Alternative hard code static mark of 0x12 every time the filter matches

sudo $TC filter add dev $ETH parent 1: protocol ip prio 10 \
u32 match ip protocol 1 0xff flowid 1:2 \
action ife encode \
type 0xDEAD \
use mark 0x12 \
dst 02:15:15:15:15:15

Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 4d48ef5..85854c0 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -751,6 +751,11 @@ config NET_ACT_IFE
 	  To compile this code as a module, choose M here: the
 	  module will be called act_ife.
 
+config NET_IFE_SKBMARK
+        tristate "Support to encoding decoding skb mark on IFE action"
+        depends on NET_ACT_IFE
+        ---help---
+
 config NET_CLS_IND
 	bool "Incoming device classification"
 	depends on NET_CLS_U32 || NET_CLS_FW
diff --git a/net/sched/Makefile b/net/sched/Makefile
index 3d17667..3f7a182 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_NET_ACT_VLAN)	+= act_vlan.o
 obj-$(CONFIG_NET_ACT_BPF)	+= act_bpf.o
 obj-$(CONFIG_NET_ACT_CONNMARK)	+= act_connmark.o
 obj-$(CONFIG_NET_ACT_IFE)	+= act_ife.o
+obj-$(CONFIG_NET_IFE_SKBMARK)	+= act_meta_mark.o
 obj-$(CONFIG_NET_SCH_FIFO)	+= sch_fifo.o
 obj-$(CONFIG_NET_SCH_CBQ)	+= sch_cbq.o
 obj-$(CONFIG_NET_SCH_HTB)	+= sch_htb.o
diff --git a/net/sched/act_meta_mark.c b/net/sched/act_meta_mark.c
new file mode 100644
index 0000000..8289217
--- /dev/null
+++ b/net/sched/act_meta_mark.c
@@ -0,0 +1,79 @@
+/*
+ * net/sched/act_meta_mark.c IFE skb->mark metadata module
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * copyright Jamal Hadi Salim (2015)
+ *
+*/
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/skbuff.h>
+#include <linux/rtnetlink.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <net/netlink.h>
+#include <net/pkt_sched.h>
+#include <uapi/linux/tc_act/tc_ife.h>
+#include <net/tc_act/tc_ife.h>
+#include <linux/rtnetlink.h>
+
+static int skbmark_encode(struct sk_buff *skb, void *skbdata,
+			  struct tcf_meta_info *e)
+{
+	u32 ifemark = skb->mark;
+
+	return ife_encode_meta_u32(ifemark, skbdata, e);
+}
+
+static int skbmark_decode(struct sk_buff *skb, void *data, u16 len)
+{
+	u32 ifemark = *(u32 *)data;
+
+	skb->mark = ntohl(ifemark);
+	return 0;
+}
+
+static int skbmark_check(struct sk_buff *skb, struct tcf_meta_info *e)
+{
+	return ife_check_meta_u32(skb->mark, e);
+}
+
+static struct tcf_meta_ops ife_skbmark_ops = {
+	.metaid = IFE_META_SKBMARK,
+	.metatype = NLA_U32,
+	.name = "skbmark",
+	.synopsis = "skb mark 32 bit metadata",
+	.check_presence = skbmark_check,
+	.encode = skbmark_encode,
+	.decode = skbmark_decode,
+	.get = ife_get_meta_u32,
+	.alloc = ife_alloc_meta_u32,
+	.release = ife_release_meta_gen,
+	.validate = ife_validate_meta_u32,
+	.owner = THIS_MODULE,
+};
+
+static int __init ifemark_init_module(void)
+{
+	return register_ife_op(&ife_skbmark_ops);
+}
+
+static void __exit ifemark_cleanup_module(void)
+{
+	unregister_ife_op(&ife_skbmark_ops);
+}
+
+module_init(ifemark_init_module);
+module_exit(ifemark_cleanup_module);
+
+MODULE_AUTHOR("Jamal Hadi Salim(2015)");
+MODULE_DESCRIPTION("Inter-FE skb mark metadata module");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_IFE_META(IFE_META_SKBMARK);
-- 
cgit v0.10.2


From 200e10f46936d95a553414f249cefb765194b235 Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <jhs@mojatatu.com>
Date: Sat, 27 Feb 2016 08:08:56 -0500
Subject: Support to encoding decoding skb prio on IFE action

    Example usage:
    Set the skb priority using skbedit then allow it to be encoded

    sudo tc qdisc add dev $ETH root handle 1: prio
    sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \
    u32 match ip protocol 1 0xff flowid 1:2 \
    action skbedit prio 17 \
    action ife encode \
    allow prio \
    dst 02:15:15:15:15:15

    Note: You dont need the skbedit action if you are already encoding the
    skb priority earlier. A zero skb priority will not be sent

    Alternative hard code static priority of decimal 33 (unlike skbedit)
    then mark of 0x12 every time the filter matches

    sudo $TC filter add dev $ETH parent 1: protocol ip prio 10 \
    u32 match ip protocol 1 0xff flowid 1:2 \
    action ife encode \
    type 0xDEAD \
    use prio 33 \
    use mark 0x12 \
    dst 02:15:15:15:15:15

    Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 85854c0..b148302 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -756,6 +756,11 @@ config NET_IFE_SKBMARK
         depends on NET_ACT_IFE
         ---help---
 
+config NET_IFE_SKBPRIO
+        tristate "Support to encoding decoding skb prio on IFE action"
+        depends on NET_ACT_IFE
+        ---help---
+
 config NET_CLS_IND
 	bool "Incoming device classification"
 	depends on NET_CLS_U32 || NET_CLS_FW
diff --git a/net/sched/Makefile b/net/sched/Makefile
index 3f7a182..84bddb3 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -21,6 +21,7 @@ obj-$(CONFIG_NET_ACT_BPF)	+= act_bpf.o
 obj-$(CONFIG_NET_ACT_CONNMARK)	+= act_connmark.o
 obj-$(CONFIG_NET_ACT_IFE)	+= act_ife.o
 obj-$(CONFIG_NET_IFE_SKBMARK)	+= act_meta_mark.o
+obj-$(CONFIG_NET_IFE_SKBPRIO)	+= act_meta_skbprio.o
 obj-$(CONFIG_NET_SCH_FIFO)	+= sch_fifo.o
 obj-$(CONFIG_NET_SCH_CBQ)	+= sch_cbq.o
 obj-$(CONFIG_NET_SCH_HTB)	+= sch_htb.o
diff --git a/net/sched/act_meta_skbprio.c b/net/sched/act_meta_skbprio.c
new file mode 100644
index 0000000..26bf4d8
--- /dev/null
+++ b/net/sched/act_meta_skbprio.c
@@ -0,0 +1,76 @@
+/*
+ * net/sched/act_meta_prio.c IFE skb->priority metadata module
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * copyright Jamal Hadi Salim (2015)
+ *
+*/
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/skbuff.h>
+#include <linux/rtnetlink.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <net/netlink.h>
+#include <net/pkt_sched.h>
+#include <uapi/linux/tc_act/tc_ife.h>
+#include <net/tc_act/tc_ife.h>
+
+static int skbprio_check(struct sk_buff *skb, struct tcf_meta_info *e)
+{
+	return ife_check_meta_u32(skb->priority, e);
+}
+
+static int skbprio_encode(struct sk_buff *skb, void *skbdata,
+			  struct tcf_meta_info *e)
+{
+	u32 ifeprio = skb->priority; /* avoid having to cast skb->priority*/
+
+	return ife_encode_meta_u32(ifeprio, skbdata, e);
+}
+
+static int skbprio_decode(struct sk_buff *skb, void *data, u16 len)
+{
+	u32 ifeprio = *(u32 *)data;
+
+	skb->priority = ntohl(ifeprio);
+	return 0;
+}
+
+static struct tcf_meta_ops ife_prio_ops = {
+	.metaid = IFE_META_PRIO,
+	.metatype = NLA_U32,
+	.name = "skbprio",
+	.synopsis = "skb prio metadata",
+	.check_presence = skbprio_check,
+	.encode = skbprio_encode,
+	.decode = skbprio_decode,
+	.get = ife_get_meta_u32,
+	.alloc = ife_alloc_meta_u32,
+	.owner = THIS_MODULE,
+};
+
+static int __init ifeprio_init_module(void)
+{
+	return register_ife_op(&ife_prio_ops);
+}
+
+static void __exit ifeprio_cleanup_module(void)
+{
+	unregister_ife_op(&ife_prio_ops);
+}
+
+module_init(ifeprio_init_module);
+module_exit(ifeprio_cleanup_module);
+
+MODULE_AUTHOR("Jamal Hadi Salim(2015)");
+MODULE_DESCRIPTION("Inter-FE skb prio metadata action");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_IFE_META(IFE_META_PRIO);
-- 
cgit v0.10.2


From 822c868532cae2cc1c51f4f18ab61c194d98aaf6 Mon Sep 17 00:00:00 2001
From: Deepa Dinamani <deepa.kernel@gmail.com>
Date: Sat, 27 Feb 2016 00:32:15 -0800
Subject: net: ipv4: Convert IP network timestamps to be y2038 safe

ICMP timestamp messages and IP source route options require
timestamps to be in milliseconds modulo 24 hours from
midnight UT format.

Add inet_current_timestamp() function to support this. The function
returns the required timestamp in network byte order.

Timestamp calculation is also changed to call ktime_get_real_ts64()
which uses struct timespec64. struct timespec64 is y2038 safe.
Previously it called getnstimeofday() which uses struct timespec.
struct timespec is not y2038 safe.

Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: James Morris <jmorris@namei.org>
Cc: Patrick McHardy <kaber@trash.net>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip.h b/include/net/ip.h
index cbb134b..fad74d3 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -240,6 +240,8 @@ static inline int inet_is_local_reserved_port(struct net *net, int port)
 }
 #endif
 
+__be32 inet_current_timestamp(void);
+
 /* From inetpeer.c */
 extern int inet_peer_threshold;
 extern int inet_peer_minttl;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 209d1ed..0cc923f 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1380,6 +1380,32 @@ out:
 	return pp;
 }
 
+#define SECONDS_PER_DAY	86400
+
+/* inet_current_timestamp - Return IP network timestamp
+ *
+ * Return milliseconds since midnight in network byte order.
+ */
+__be32 inet_current_timestamp(void)
+{
+	u32 secs;
+	u32 msecs;
+	struct timespec64 ts;
+
+	ktime_get_real_ts64(&ts);
+
+	/* Get secs since midnight. */
+	(void)div_u64_rem(ts.tv_sec, SECONDS_PER_DAY, &secs);
+	/* Convert to msecs. */
+	msecs = secs * MSEC_PER_SEC;
+	/* Convert nsec to msec. */
+	msecs += (u32)ts.tv_nsec / NSEC_PER_MSEC;
+
+	/* Convert to network byte order. */
+	return htons(msecs);
+}
+EXPORT_SYMBOL(inet_current_timestamp);
+
 int inet_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len)
 {
 	if (sk->sk_family == AF_INET)
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 36e2697..6333489 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -931,7 +931,6 @@ static bool icmp_echo(struct sk_buff *skb)
  */
 static bool icmp_timestamp(struct sk_buff *skb)
 {
-	struct timespec tv;
 	struct icmp_bxm icmp_param;
 	/*
 	 *	Too short.
@@ -942,9 +941,7 @@ static bool icmp_timestamp(struct sk_buff *skb)
 	/*
 	 *	Fill in the current time as ms since midnight UT:
 	 */
-	getnstimeofday(&tv);
-	icmp_param.data.times[1] = htonl((tv.tv_sec % 86400) * MSEC_PER_SEC +
-					 tv.tv_nsec / NSEC_PER_MSEC);
+	icmp_param.data.times[1] = inet_current_timestamp();
 	icmp_param.data.times[2] = icmp_param.data.times[1];
 	if (skb_copy_bits(skb, 0, &icmp_param.data.times[0], 4))
 		BUG();
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index bd24679..4d158ff 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -58,10 +58,9 @@ void ip_options_build(struct sk_buff *skb, struct ip_options *opt,
 		if (opt->ts_needaddr)
 			ip_rt_get_source(iph+opt->ts+iph[opt->ts+2]-9, skb, rt);
 		if (opt->ts_needtime) {
-			struct timespec tv;
 			__be32 midtime;
-			getnstimeofday(&tv);
-			midtime = htonl((tv.tv_sec % 86400) * MSEC_PER_SEC + tv.tv_nsec / NSEC_PER_MSEC);
+
+			midtime = inet_current_timestamp();
 			memcpy(iph+opt->ts+iph[opt->ts+2]-5, &midtime, 4);
 		}
 		return;
@@ -415,11 +414,10 @@ int ip_options_compile(struct net *net,
 					break;
 				}
 				if (timeptr) {
-					struct timespec tv;
-					u32  midtime;
-					getnstimeofday(&tv);
-					midtime = (tv.tv_sec % 86400) * MSEC_PER_SEC + tv.tv_nsec / NSEC_PER_MSEC;
-					put_unaligned_be32(midtime, timeptr);
+					__be32 midtime;
+
+					midtime = inet_current_timestamp();
+					memcpy(timeptr, &midtime, 4);
 					opt->is_changed = 1;
 				}
 			} else if ((optptr[3]&0xF) != IPOPT_TS_PRESPEC) {
-- 
cgit v0.10.2


From b1b270d863c022f3dbf1f8786fd2956703ee10fc Mon Sep 17 00:00:00 2001
From: Deepa Dinamani <deepa.kernel@gmail.com>
Date: Sat, 27 Feb 2016 00:32:16 -0800
Subject: net: ipv4: tcp_probe: Replace timespec with timespec64

TCP probe log timestamps use struct timespec which is
not y2038 safe. Even though timespec might be good enough here
as it is used to represent delta time, the plan is to get rid
of all uses of timespec in the kernel.
Replace with struct timespec64 which is y2038 safe.

Prints still use unsigned long format and type.

Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp_probe.c
index ebf5ff5..f6c50af 100644
--- a/net/ipv4/tcp_probe.c
+++ b/net/ipv4/tcp_probe.c
@@ -187,13 +187,13 @@ static int tcpprobe_sprint(char *tbuf, int n)
 {
 	const struct tcp_log *p
 		= tcp_probe.log + tcp_probe.tail;
-	struct timespec tv
-		= ktime_to_timespec(ktime_sub(p->tstamp, tcp_probe.start));
+	struct timespec64 ts
+		= ktime_to_timespec64(ktime_sub(p->tstamp, tcp_probe.start));
 
 	return scnprintf(tbuf, n,
 			"%lu.%09lu %pISpc %pISpc %d %#x %#x %u %u %u %u %u\n",
-			(unsigned long)tv.tv_sec,
-			(unsigned long)tv.tv_nsec,
+			(unsigned long)ts.tv_sec,
+			(unsigned long)ts.tv_nsec,
 			&p->src, &p->dst, p->length, p->snd_nxt, p->snd_una,
 			p->snd_cwnd, p->ssthresh, p->snd_wnd, p->srtt, p->rcv_wnd);
 }
-- 
cgit v0.10.2


From 6497c7e6400105424a7d1dddc3f3aadcc8e3588a Mon Sep 17 00:00:00 2001
From: Deepa Dinamani <deepa.kernel@gmail.com>
Date: Sat, 27 Feb 2016 00:32:17 -0800
Subject: net: sctp: Convert log timestamps to be y2038 safe

SCTP probe log timestamps use struct timespec which is
not y2038 safe.
Use struct timespec64 which is 2038 safe instead.

Use monotonic time instead of real time as only time
differences are logged.

Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Cc: Vlad Yasevich <vyasevich@gmail.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: linux-sctp@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sctp/probe.c b/net/sctp/probe.c
index 5e68b94..6cc2152 100644
--- a/net/sctp/probe.c
+++ b/net/sctp/probe.c
@@ -65,7 +65,7 @@ static struct {
 	struct kfifo	  fifo;
 	spinlock_t	  lock;
 	wait_queue_head_t wait;
-	struct timespec	  tstart;
+	struct timespec64 tstart;
 } sctpw;
 
 static __printf(1, 2) void printl(const char *fmt, ...)
@@ -85,7 +85,7 @@ static __printf(1, 2) void printl(const char *fmt, ...)
 static int sctpprobe_open(struct inode *inode, struct file *file)
 {
 	kfifo_reset(&sctpw.fifo);
-	getnstimeofday(&sctpw.tstart);
+	ktime_get_ts64(&sctpw.tstart);
 
 	return 0;
 }
@@ -138,7 +138,7 @@ static sctp_disposition_t jsctp_sf_eat_sack(struct net *net,
 	struct sk_buff *skb = chunk->skb;
 	struct sctp_transport *sp;
 	static __u32 lcwnd = 0;
-	struct timespec now;
+	struct timespec64 now;
 
 	sp = asoc->peer.primary_path;
 
@@ -149,8 +149,8 @@ static sctp_disposition_t jsctp_sf_eat_sack(struct net *net,
 	    (full || sp->cwnd != lcwnd)) {
 		lcwnd = sp->cwnd;
 
-		getnstimeofday(&now);
-		now = timespec_sub(now, sctpw.tstart);
+		ktime_get_ts64(&now);
+		now = timespec64_sub(now, sctpw.tstart);
 
 		printl("%lu.%06lu ", (unsigned long) now.tv_sec,
 		       (unsigned long) now.tv_nsec / NSEC_PER_USEC);
-- 
cgit v0.10.2


From 4f25a1110cd4f5746b750a9567303c7ef6826b7c Mon Sep 17 00:00:00 2001
From: David Ahern <dsa@cumulusnetworks.com>
Date: Sat, 27 Feb 2016 09:59:17 -0800
Subject: net: ipv6/l3mdev: Move host route on saved address if necessary

Commit f1705ec197e70 allows IPv6 addresses to be retained on a link down.
The address can have a cached host route which can point to the wrong
FIB table if the L3 enslavement is changed (e.g., route can point to local
table instead of VRF table if device is added to an L3 domain).

On link up check the table of the cached host route against the FIB
table associated with the device and correct if needed.

Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index a2d6f6c..b1169d5 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3170,9 +3170,35 @@ static void addrconf_gre_config(struct net_device *dev)
 }
 #endif
 
+#if IS_ENABLED(CONFIG_NET_L3_MASTER_DEV)
+/* If the host route is cached on the addr struct make sure it is associated
+ * with the proper table. e.g., enslavement can change and if so the cached
+ * host route needs to move to the new table.
+ */
+static void l3mdev_check_host_rt(struct inet6_dev *idev,
+				  struct inet6_ifaddr *ifp)
+{
+	if (ifp->rt) {
+		u32 tb_id = l3mdev_fib_table(idev->dev) ? : RT6_TABLE_LOCAL;
+
+		if (tb_id != ifp->rt->rt6i_table->tb6_id) {
+			ip6_del_rt(ifp->rt);
+			ifp->rt = NULL;
+		}
+	}
+}
+#else
+static void l3mdev_check_host_rt(struct inet6_dev *idev,
+				  struct inet6_ifaddr *ifp)
+{
+}
+#endif
+
 static int fixup_permanent_addr(struct inet6_dev *idev,
 				struct inet6_ifaddr *ifp)
 {
+	l3mdev_check_host_rt(idev, ifp);
+
 	if (!ifp->rt) {
 		struct rt6_info *rt;
 
-- 
cgit v0.10.2


From daa21560a29673f63263e907045f513a8d66dc33 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:32 +0200
Subject: net/mlx5e: Replace async events spinlock with synchronize_irq()

We only need to flush the irq handler to make sure it does not
queue a work into the global work queue after we start to flush it.
So using synchronize_irq() is more appropriate than a spin lock.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 1dca3dc..4511984 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -555,7 +555,6 @@ struct mlx5e_priv {
 	struct mlx5e_vxlan_db      vxlan;
 
 	struct mlx5e_params        params;
-	spinlock_t                 async_events_spinlock; /* sync hw events */
 	struct work_struct         update_carrier_work;
 	struct work_struct         set_rx_mode_work;
 	struct delayed_work        update_stats_work;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 0d45f35..38944b8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -275,9 +275,14 @@ static void mlx5e_update_stats_work(struct work_struct *work)
 	mutex_unlock(&priv->state_lock);
 }
 
-static void __mlx5e_async_event(struct mlx5e_priv *priv,
-				enum mlx5_dev_event event)
+static void mlx5e_async_event(struct mlx5_core_dev *mdev, void *vpriv,
+			      enum mlx5_dev_event event, unsigned long param)
 {
+	struct mlx5e_priv *priv = vpriv;
+
+	if (!test_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state))
+		return;
+
 	switch (event) {
 	case MLX5_DEV_EVENT_PORT_UP:
 	case MLX5_DEV_EVENT_PORT_DOWN:
@@ -289,17 +294,6 @@ static void __mlx5e_async_event(struct mlx5e_priv *priv,
 	}
 }
 
-static void mlx5e_async_event(struct mlx5_core_dev *mdev, void *vpriv,
-			      enum mlx5_dev_event event, unsigned long param)
-{
-	struct mlx5e_priv *priv = vpriv;
-
-	spin_lock(&priv->async_events_spinlock);
-	if (test_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state))
-		__mlx5e_async_event(priv, event);
-	spin_unlock(&priv->async_events_spinlock);
-}
-
 static void mlx5e_enable_async_events(struct mlx5e_priv *priv)
 {
 	set_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state);
@@ -307,9 +301,8 @@ static void mlx5e_enable_async_events(struct mlx5e_priv *priv)
 
 static void mlx5e_disable_async_events(struct mlx5e_priv *priv)
 {
-	spin_lock_irq(&priv->async_events_spinlock);
 	clear_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state);
-	spin_unlock_irq(&priv->async_events_spinlock);
+	synchronize_irq(mlx5_get_msix_vec(priv->mdev, MLX5_EQ_VEC_ASYNC));
 }
 
 #define MLX5E_HW2SW_MTU(hwmtu) (hwmtu - (ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN))
@@ -2290,7 +2283,6 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 	mlx5e_ets_init(priv);
 #endif
 
-	spin_lock_init(&priv->async_events_spinlock);
 	mutex_init(&priv->state_lock);
 
 	INIT_WORK(&priv->update_carrier_work, mlx5e_update_carrier_work);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 647a3ca..18fccec 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -442,6 +442,11 @@ int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq)
 }
 EXPORT_SYMBOL_GPL(mlx5_destroy_unmap_eq);
 
+u32 mlx5_get_msix_vec(struct mlx5_core_dev *dev, int vecidx)
+{
+	return dev->priv.msix_arr[MLX5_EQ_VEC_ASYNC].vector;
+}
+
 int mlx5_eq_init(struct mlx5_core_dev *dev)
 {
 	int err;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 0336847..0b0b226 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -99,6 +99,7 @@ int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id);
 int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id);
 int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev);
 cycle_t mlx5_read_internal_timer(struct mlx5_core_dev *dev);
+u32 mlx5_get_msix_vec(struct mlx5_core_dev *dev, int vecidx);
 
 void mlx5e_init(void);
 void mlx5e_cleanup(void);
-- 
cgit v0.10.2


From ce89ef36d2544b4180fdb9ce9e2918196b0082f8 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:33 +0200
Subject: net/mlx5e: Placement changed for carrier state updates

More proper to declare carrier state UP only after the channels
are ready for traffic.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 38944b8..013be09 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1451,8 +1451,8 @@ int mlx5e_open_locked(struct net_device *netdev)
 		goto err_close_channels;
 	}
 
-	mlx5e_update_carrier(priv);
 	mlx5e_redirect_rqts(priv);
+	mlx5e_update_carrier(priv);
 	mlx5e_timestamp_init(priv);
 
 	schedule_delayed_work(&priv->update_stats_work, 0);
@@ -1491,8 +1491,8 @@ int mlx5e_close_locked(struct net_device *netdev)
 	clear_bit(MLX5E_STATE_OPENED, &priv->state);
 
 	mlx5e_timestamp_cleanup(priv);
-	mlx5e_redirect_rqts(priv);
 	netif_carrier_off(priv->netdev);
+	mlx5e_redirect_rqts(priv);
 	mlx5e_close_channels(priv);
 
 	return 0;
-- 
cgit v0.10.2


From 3b6195240c52b38a2d7ecc9cc8407596d9f24663 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:34 +0200
Subject: net/mlx5e: Changed naming convention of tx queues in ethtool stats

Instead of the pair (channel, tc), we now use a single number that
goes over all tx queues of a TC, for all TCs.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index e9760f8..577b4b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -211,13 +211,14 @@ static void mlx5e_get_strings(struct net_device *dev,
 				sprintf(data + (idx++) * ETH_GSTRING_LEN,
 					"rx%d_%s", i, rq_stats_strings[j]);
 
-		for (i = 0; i < priv->params.num_channels; i++)
-			for (tc = 0; tc < priv->params.num_tc; tc++)
+		for (tc = 0; tc < priv->params.num_tc; tc++)
+			for (i = 0; i < priv->params.num_channels; i++)
 				for (j = 0; j < NUM_SQ_STATS; j++)
 					sprintf(data +
-						(idx++) * ETH_GSTRING_LEN,
-						"tx%d_%d_%s", i, tc,
-						sq_stats_strings[j]);
+					      (idx++) * ETH_GSTRING_LEN,
+					      "tx%d_%s",
+					      priv->channeltc_to_txq_map[i][tc],
+					      sq_stats_strings[j]);
 		break;
 	}
 }
@@ -249,8 +250,8 @@ static void mlx5e_get_ethtool_stats(struct net_device *dev,
 						&priv->state) ? 0 :
 				       ((u64 *)&priv->channel[i]->rq.stats)[j];
 
-	for (i = 0; i < priv->params.num_channels; i++)
-		for (tc = 0; tc < priv->params.num_tc; tc++)
+	for (tc = 0; tc < priv->params.num_tc; tc++)
+		for (i = 0; i < priv->params.num_channels; i++)
 			for (j = 0; j < NUM_SQ_STATS; j++)
 				data[idx++] = !test_bit(MLX5E_STATE_OPENED,
 							&priv->state) ? 0 :
-- 
cgit v0.10.2


From c89fb18b657b07d6c98810182057f844e71d3eee Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:35 +0200
Subject: net/mlx5e: Move common case counters within sq_stats struct

For data cache locality considerations, we moved the nop and
csum_offload_inner within sq_stats struct as they are more
commonly accessed in xmit path.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 4511984..b289660 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -260,26 +260,28 @@ static const char sq_stats_strings[][ETH_GSTRING_LEN] = {
 	"tso_bytes",
 	"tso_inner_packets",
 	"tso_inner_bytes",
-	"csum_offload_none",
 	"csum_offload_inner",
+	"nop",
+	"csum_offload_none",
 	"stopped",
 	"wake",
 	"dropped",
-	"nop"
 };
 
 struct mlx5e_sq_stats {
+	/* commonly accessed in data path */
 	u64 packets;
 	u64 tso_packets;
 	u64 tso_bytes;
 	u64 tso_inner_packets;
 	u64 tso_inner_bytes;
-	u64 csum_offload_none;
 	u64 csum_offload_inner;
+	u64 nop;
+	/* less likely accessed in data path */
+	u64 csum_offload_none;
 	u64 stopped;
 	u64 wake;
 	u64 dropped;
-	u64 nop;
 #define NUM_SQ_STATS 11
 };
 
-- 
cgit v0.10.2


From 556dd1b9c3134501aba707fab66cf4d739e40061 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:36 +0200
Subject: net/mlx5e: Set drop RQ's necessary parameters only

By its role, there is no need to set all the other parameters
for the drop RQ.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 013be09..30fd971 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1064,6 +1064,15 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv,
 	param->wq.linear = 1;
 }
 
+static void mlx5e_build_drop_rq_param(struct mlx5e_rq_param *param)
+{
+	void *rqc = param->rqc;
+	void *wq = MLX5_ADDR_OF(rqc, rqc, wq);
+
+	MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST);
+	MLX5_SET(wq, wq, log_wq_stride,    ilog2(sizeof(struct mlx5e_rx_wqe)));
+}
+
 static void mlx5e_build_sq_param(struct mlx5e_priv *priv,
 				 struct mlx5e_sq_param *param)
 {
@@ -1574,8 +1583,7 @@ static int mlx5e_open_drop_rq(struct mlx5e_priv *priv)
 
 	memset(&cq_param, 0, sizeof(cq_param));
 	memset(&rq_param, 0, sizeof(rq_param));
-	mlx5e_build_rx_cq_param(priv, &cq_param);
-	mlx5e_build_rq_param(priv, &rq_param);
+	mlx5e_build_drop_rq_param(&rq_param);
 
 	err = mlx5e_create_drop_cq(priv, cq, &cq_param);
 	if (err)
-- 
cgit v0.10.2


From 7524a5d88b94afef8397a79f1e664af5b7052c22 Mon Sep 17 00:00:00 2001
From: Gal Pressman <galp@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:37 +0200
Subject: net/mlx5e: Don't try to modify CQ moderation if it is not supported

If CQ moderation is not supported by the device, print a warning on
netdevice load, and return error when trying to modify/query cq
moderation via ethtool.

Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4
Ethernet functionality')
Signed-off-by: Gal Pressman <galp@mellanox.com>

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 577b4b1..a1b3bb4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -400,6 +400,9 @@ static int mlx5e_get_coalesce(struct net_device *netdev,
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 
+	if (!MLX5_CAP_GEN(priv->mdev, cq_moderation))
+		return -ENOTSUPP;
+
 	coal->rx_coalesce_usecs       = priv->params.rx_cq_moderation_usec;
 	coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation_pkts;
 	coal->tx_coalesce_usecs       = priv->params.tx_cq_moderation_usec;
@@ -417,6 +420,9 @@ static int mlx5e_set_coalesce(struct net_device *netdev,
 	int tc;
 	int i;
 
+	if (!MLX5_CAP_GEN(mdev, cq_moderation))
+		return -ENOTSUPP;
+
 	priv->params.tx_cq_moderation_usec = coal->tx_coalesce_usecs;
 	priv->params.tx_cq_moderation_pkts = coal->tx_max_coalesced_frames;
 	priv->params.rx_cq_moderation_usec = coal->rx_coalesce_usecs;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 30fd971..b20a35b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -870,12 +870,10 @@ static int mlx5e_open_cq(struct mlx5e_channel *c,
 	if (err)
 		goto err_destroy_cq;
 
-	err = mlx5_core_modify_cq_moderation(mdev, &cq->mcq,
-					     moderation_usecs,
-					     moderation_frames);
-	if (err)
-		goto err_destroy_cq;
-
+	if (MLX5_CAP_GEN(mdev, cq_moderation))
+		mlx5_core_modify_cq_moderation(mdev, &cq->mcq,
+					       moderation_usecs,
+					       moderation_frames);
 	return 0;
 
 err_destroy_cq:
@@ -2218,6 +2216,8 @@ static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev)
 	}
 	if (!MLX5_CAP_ETH(mdev, self_lb_en_modifiable))
 		mlx5_core_warn(mdev, "Self loop back prevention is not supported\n");
+	if (!MLX5_CAP_GEN(mdev, cq_moderation))
+		mlx5_core_warn(mdev, "CQ modiration is not supported\n");
 
 	return 0;
 }
-- 
cgit v0.10.2


From 2fcb92fbd04eef26dfe7e67839da6262d83d6b65 Mon Sep 17 00:00:00 2001
From: Gal Pressman <galp@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:38 +0200
Subject: net/mlx5e: Don't modify CQ before it was created

Calling mlx5e_set_coalesce while the interface is down will result in
modifying CQs that don't exist.

Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4
Ethernet functionality')
Signed-off-by: Gal Pressman <galp@mellanox.com>

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index a1b3bb4..0959656 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -423,11 +423,15 @@ static int mlx5e_set_coalesce(struct net_device *netdev,
 	if (!MLX5_CAP_GEN(mdev, cq_moderation))
 		return -ENOTSUPP;
 
+	mutex_lock(&priv->state_lock);
 	priv->params.tx_cq_moderation_usec = coal->tx_coalesce_usecs;
 	priv->params.tx_cq_moderation_pkts = coal->tx_max_coalesced_frames;
 	priv->params.rx_cq_moderation_usec = coal->rx_coalesce_usecs;
 	priv->params.rx_cq_moderation_pkts = coal->rx_max_coalesced_frames;
 
+	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
+		goto out;
+
 	for (i = 0; i < priv->params.num_channels; ++i) {
 		c = priv->channel[i];
 
@@ -443,6 +447,8 @@ static int mlx5e_set_coalesce(struct net_device *netdev,
 					       coal->rx_max_coalesced_frames);
 	}
 
+out:
+	mutex_unlock(&priv->state_lock);
 	return 0;
 }
 
-- 
cgit v0.10.2


From 6b6c07bdcdc97ccac2596063bfc32a5faddfe884 Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:39 +0200
Subject: net/mlx5: Make command timeout way shorter

The command timeout is terribly long, whole two hours. Make it 60s so if
things do go wrong, the user gets feedback in relatively short time, so
they can take corrective actions and/or investigate using tools and such.

Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index a815da9..3388a43 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -54,7 +54,7 @@ enum {
 	/* one minute for the sake of bringup. Generally, commands must always
 	 * complete and we may need to increase this timeout value
 	 */
-	MLX5_CMD_TIMEOUT_MSEC	= 7200 * 1000,
+	MLX5_CMD_TIMEOUT_MSEC	= 60 * 1000,
 	MLX5_CMD_WQ_MAX_NAME	= 32,
 };
 
-- 
cgit v0.10.2


From 0ba422410bbf7081c3c7d7b2dcc10e9eb5cb46f7 Mon Sep 17 00:00:00 2001
From: Moshe Lazer <moshel@mellanox.com>
Date: Wed, 2 Mar 2016 00:13:40 +0200
Subject: net/mlx5: Fix global UAR mapping

Avoid double mapping of io mapped memory, Device page may be
mapped to non-cached(NC) or to write-combining(WC).
The code before this fix tries to map it both to WC and NC
contrary to what stated in Intel's software developer manual.

Here we remove the global WC mapping of all UARS
"dev->priv.bf_mapping", since UAR mapping should be decided
per UAR (e.g we want different mappings for EQs, CQs vs QPs).

Caller will now have to choose whether to map via
write-combining API or not.

mlx5e SQs will choose write-combining in order to perform
BlueFlame writes.

Fixes: 88a85f99e51f ('TX latency optimization to save DMA reads')
Signed-off-by: Moshe Lazer <moshel@mellanox.com>
Reviewed-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index b289660..9c0e80e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -388,6 +388,7 @@ struct mlx5e_sq_dma {
 
 enum {
 	MLX5E_SQ_STATE_WAKE_TXQ_ENABLE,
+	MLX5E_SQ_STATE_BF_ENABLE,
 };
 
 struct mlx5e_sq {
@@ -416,7 +417,6 @@ struct mlx5e_sq {
 	struct mlx5_wq_cyc         wq;
 	u32                        dma_fifo_mask;
 	void __iomem              *uar_map;
-	void __iomem              *uar_bf_map;
 	struct netdev_queue       *txq;
 	u32                        sqn;
 	u16                        bf_buf_size;
@@ -664,16 +664,12 @@ static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq,
 	 * doorbell
 	 */
 	wmb();
-
-	if (bf_sz) {
-		__iowrite64_copy(sq->uar_bf_map + ofst, &wqe->ctrl, bf_sz);
-
-		/* flush the write-combining mapped buffer */
-		wmb();
-
-	} else {
+	if (bf_sz)
+		__iowrite64_copy(sq->uar_map + ofst, &wqe->ctrl, bf_sz);
+	else
 		mlx5_write64((__be32 *)&wqe->ctrl, sq->uar_map + ofst, NULL);
-	}
+	/* flush the write-combining mapped buffer */
+	wmb();
 
 	sq->bf_offset ^= sq->bf_buf_size;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index b20a35b..5063c0e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -548,7 +548,7 @@ static int mlx5e_create_sq(struct mlx5e_channel *c,
 	int txq_ix;
 	int err;
 
-	err = mlx5_alloc_map_uar(mdev, &sq->uar);
+	err = mlx5_alloc_map_uar(mdev, &sq->uar, true);
 	if (err)
 		return err;
 
@@ -560,8 +560,12 @@ static int mlx5e_create_sq(struct mlx5e_channel *c,
 		goto err_unmap_free_uar;
 
 	sq->wq.db       = &sq->wq.db[MLX5_SND_DBR];
-	sq->uar_map     = sq->uar.map;
-	sq->uar_bf_map  = sq->uar.bf_map;
+	if (sq->uar.bf_map) {
+		set_bit(MLX5E_SQ_STATE_BF_ENABLE, &sq->state);
+		sq->uar_map = sq->uar.bf_map;
+	} else {
+		sq->uar_map = sq->uar.map;
+	}
 	sq->bf_buf_size = (1 << MLX5_CAP_GEN(mdev, log_bf_reg_size)) / 2;
 	sq->max_inline  = param->max_inline;
 
@@ -2418,7 +2422,7 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 
 	priv = netdev_priv(netdev);
 
-	err = mlx5_alloc_map_uar(mdev, &priv->cq_uar);
+	err = mlx5_alloc_map_uar(mdev, &priv->cq_uar, false);
 	if (err) {
 		mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err);
 		goto err_free_netdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index a05c070..c34f4f3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -303,7 +303,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	if (!skb->xmit_more || netif_xmit_stopped(sq->txq)) {
 		int bf_sz = 0;
 
-		if (bf && sq->uar_bf_map)
+		if (bf && test_bit(MLX5E_SQ_STATE_BF_ENABLE, &sq->state))
 			bf_sz = wi->num_wqebbs << 3;
 
 		cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 1545a94..8b7133d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -767,22 +767,6 @@ static int mlx5_core_set_issi(struct mlx5_core_dev *dev)
 	return -ENOTSUPP;
 }
 
-static int map_bf_area(struct mlx5_core_dev *dev)
-{
-	resource_size_t bf_start = pci_resource_start(dev->pdev, 0);
-	resource_size_t bf_len = pci_resource_len(dev->pdev, 0);
-
-	dev->priv.bf_mapping = io_mapping_create_wc(bf_start, bf_len);
-
-	return dev->priv.bf_mapping ? 0 : -ENOMEM;
-}
-
-static void unmap_bf_area(struct mlx5_core_dev *dev)
-{
-	if (dev->priv.bf_mapping)
-		io_mapping_free(dev->priv.bf_mapping);
-}
-
 static void mlx5_add_device(struct mlx5_interface *intf, struct mlx5_priv *priv)
 {
 	struct mlx5_device_context *dev_ctx;
@@ -1103,14 +1087,9 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 		goto err_stop_eqs;
 	}
 
-	if (map_bf_area(dev))
-		dev_err(&pdev->dev, "Failed to map blue flame area\n");
-
 	err = mlx5_irq_set_affinity_hints(dev);
-	if (err) {
+	if (err)
 		dev_err(&pdev->dev, "Failed to alloc affinity hint cpumask\n");
-		goto err_unmap_bf_area;
-	}
 
 	MLX5_INIT_DOORBELL_LOCK(&priv->cq_uar_lock);
 
@@ -1169,10 +1148,6 @@ err_fs:
 	mlx5_cleanup_qp_table(dev);
 	mlx5_cleanup_cq_table(dev);
 	mlx5_irq_clear_affinity_hints(dev);
-
-err_unmap_bf_area:
-	unmap_bf_area(dev);
-
 	free_comp_eqs(dev);
 
 err_stop_eqs:
@@ -1242,7 +1217,6 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 	mlx5_cleanup_qp_table(dev);
 	mlx5_cleanup_cq_table(dev);
 	mlx5_irq_clear_affinity_hints(dev);
-	unmap_bf_area(dev);
 	free_comp_eqs(dev);
 	mlx5_stop_eqs(dev);
 	mlx5_free_uuars(dev, &priv->uuari);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/uar.c b/drivers/net/ethernet/mellanox/mlx5/core/uar.c
index eb05c84..8ba080e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/uar.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/uar.c
@@ -226,7 +226,8 @@ int mlx5_free_uuars(struct mlx5_core_dev *dev, struct mlx5_uuar_info *uuari)
 	return 0;
 }
 
-int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar)
+int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar,
+		       bool map_wc)
 {
 	phys_addr_t pfn;
 	phys_addr_t uar_bar_start;
@@ -240,20 +241,26 @@ int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar)
 
 	uar_bar_start = pci_resource_start(mdev->pdev, 0);
 	pfn           = (uar_bar_start >> PAGE_SHIFT) + uar->index;
-	uar->map      = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE);
-	if (!uar->map) {
-		mlx5_core_warn(mdev, "ioremap() failed, %d\n", err);
-		err = -ENOMEM;
-		goto err_free_uar;
-	}
 
-	if (mdev->priv.bf_mapping)
-		uar->bf_map = io_mapping_map_wc(mdev->priv.bf_mapping,
-						uar->index << PAGE_SHIFT);
+	if (map_wc) {
+		uar->bf_map = ioremap_wc(pfn << PAGE_SHIFT, PAGE_SIZE);
+		if (!uar->bf_map) {
+			mlx5_core_warn(mdev, "ioremap_wc() failed\n");
+			uar->map = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE);
+			if (!uar->map)
+				goto err_free_uar;
+		}
+	} else {
+		uar->map = ioremap(pfn << PAGE_SHIFT, PAGE_SIZE);
+		if (!uar->map)
+			goto err_free_uar;
+	}
 
 	return 0;
 
 err_free_uar:
+	mlx5_core_warn(mdev, "ioremap() failed\n");
+	err = -ENOMEM;
 	mlx5_cmd_free_uar(mdev, uar->index);
 
 	return err;
@@ -262,8 +269,8 @@ EXPORT_SYMBOL(mlx5_alloc_map_uar);
 
 void mlx5_unmap_free_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar)
 {
-	io_mapping_unmap(uar->bf_map);
 	iounmap(uar->map);
+	iounmap(uar->bf_map);
 	mlx5_cmd_free_uar(mdev, uar->index);
 }
 EXPORT_SYMBOL(mlx5_unmap_free_uar);
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 3388a43..bb1a880 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -460,8 +460,6 @@ struct mlx5_priv {
 	struct mlx5_uuar_info	uuari;
 	MLX5_DECLARE_DOORBELL_LOCK(cq_uar_lock);
 
-	struct io_mapping	*bf_mapping;
-
 	/* pages stuff */
 	struct workqueue_struct *pg_wq;
 	struct rb_root		page_root;
@@ -719,7 +717,8 @@ int mlx5_cmd_alloc_uar(struct mlx5_core_dev *dev, u32 *uarn);
 int mlx5_cmd_free_uar(struct mlx5_core_dev *dev, u32 uarn);
 int mlx5_alloc_uuars(struct mlx5_core_dev *dev, struct mlx5_uuar_info *uuari);
 int mlx5_free_uuars(struct mlx5_core_dev *dev, struct mlx5_uuar_info *uuari);
-int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar);
+int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar,
+		       bool map_wc);
 void mlx5_unmap_free_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar);
 void mlx5_health_cleanup(struct mlx5_core_dev *dev);
 int mlx5_health_init(struct mlx5_core_dev *dev);
-- 
cgit v0.10.2


From 64d4e3431e686dc37ce388ba531c4c4e866fb141 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Sat, 27 Feb 2016 20:19:54 -0800
Subject: net: remove skb_sender_cpu_clear()

After commit 52bd2d62ce67 ("net: better skb->sender_cpu and skb->napi_id cohabitation")
skb_sender_cpu_clear() becomes empty and can be removed.

Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index eab4f8f..797cefb 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1161,10 +1161,6 @@ static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from)
 	to->l4_hash = from->l4_hash;
 };
 
-static inline void skb_sender_cpu_clear(struct sk_buff *skb)
-{
-}
-
 #ifdef NET_SKBUFF_DATA_USES_OFFSET
 static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
 {
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index fcdb86d..f47759f 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -44,7 +44,6 @@ int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb
 
 	skb_push(skb, ETH_HLEN);
 	br_drop_fake_rtable(skb);
-	skb_sender_cpu_clear(skb);
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL &&
 	    (skb->protocol == htons(ETH_P_8021Q) ||
diff --git a/net/core/filter.c b/net/core/filter.c
index a3aba15..5e2a3b5 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1597,7 +1597,6 @@ static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5)
 	}
 
 	skb2->dev = dev;
-	skb_sender_cpu_clear(skb2);
 	return dev_queue_xmit(skb2);
 }
 
@@ -1650,7 +1649,6 @@ int skb_do_redirect(struct sk_buff *skb)
 	}
 
 	skb->dev = dev;
-	skb_sender_cpu_clear(skb);
 	return dev_queue_xmit(skb);
 }
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 488566b..7af7ec6 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4302,7 +4302,6 @@ void skb_scrub_packet(struct sk_buff *skb, bool xnet)
 	skb->skb_iif = 0;
 	skb->ignore_df = 0;
 	skb_dst_drop(skb);
-	skb_sender_cpu_clear(skb);
 	secpath_reset(skb);
 	nf_reset(skb);
 	nf_reset_trace(skb);
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index da0d7ce..af18f1e 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -71,7 +71,6 @@ static int ip_forward_finish(struct net *net, struct sock *sk, struct sk_buff *s
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
 
-	skb_sender_cpu_clear(skb);
 	return dst_output(net, sk, skb);
 }
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index a163102..9428345 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -332,7 +332,6 @@ static int ip6_forward_proxy_check(struct sk_buff *skb)
 static inline int ip6_forward_finish(struct net *net, struct sock *sk,
 				     struct sk_buff *skb)
 {
-	skb_sender_cpu_clear(skb);
 	return dst_output(net, sk, skb);
 }
 
diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
index a3f5cd9..dc196a0 100644
--- a/net/netfilter/ipvs/ip_vs_xmit.c
+++ b/net/netfilter/ipvs/ip_vs_xmit.c
@@ -531,8 +531,6 @@ static inline int ip_vs_tunnel_xmit_prepare(struct sk_buff *skb,
 	if (ret == NF_ACCEPT) {
 		nf_reset(skb);
 		skb_forward_csum(skb);
-		if (!skb->sk)
-			skb_sender_cpu_clear(skb);
 	}
 	return ret;
 }
@@ -573,8 +571,6 @@ static inline int ip_vs_nat_send_or_cont(int pf, struct sk_buff *skb,
 
 	if (!local) {
 		skb_forward_csum(skb);
-		if (!skb->sk)
-			skb_sender_cpu_clear(skb);
 		NF_HOOK(pf, NF_INET_LOCAL_OUT, cp->ipvs->net, NULL, skb,
 			NULL, skb_dst(skb)->dev, dst_output);
 	} else
@@ -595,8 +591,6 @@ static inline int ip_vs_send_or_cont(int pf, struct sk_buff *skb,
 	if (!local) {
 		ip_vs_drop_early_demux_sk(skb);
 		skb_forward_csum(skb);
-		if (!skb->sk)
-			skb_sender_cpu_clear(skb);
 		NF_HOOK(pf, NF_INET_LOCAL_OUT, cp->ipvs->net, NULL, skb,
 			NULL, skb_dst(skb)->dev, dst_output);
 	} else
diff --git a/net/netfilter/nf_dup_netdev.c b/net/netfilter/nf_dup_netdev.c
index 8414ee1..7ec6972 100644
--- a/net/netfilter/nf_dup_netdev.c
+++ b/net/netfilter/nf_dup_netdev.c
@@ -31,7 +31,6 @@ void nf_dup_netdev_egress(const struct nft_pktinfo *pkt, int oif)
 		skb_push(skb, skb->mac_len);
 
 	skb->dev = dev;
-	skb_sender_cpu_clear(skb);
 	dev_queue_xmit(skb);
 }
 EXPORT_SYMBOL_GPL(nf_dup_netdev_egress);
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index 6b284d9..e8a760c 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -182,7 +182,6 @@ static int tcf_mirred(struct sk_buff *skb, const struct tc_action *a,
 
 	skb2->skb_iif = skb->dev->ifindex;
 	skb2->dev = dev;
-	skb_sender_cpu_clear(skb2);
 	err = dev_queue_xmit(skb2);
 
 	if (err) {
-- 
cgit v0.10.2


From 4ac801b77e6f06e6b12c069fd29216a4102065fb Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 28 Feb 2016 12:26:52 +0200
Subject: qed: Semantic refactoring of interrupt code

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index acfe7be..d9a5175 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -1011,13 +1011,17 @@ static void qed_hw_get_resc(struct qed_hwfn *p_hwfn)
 {
 	u32 *resc_start = p_hwfn->hw_info.resc_start;
 	u32 *resc_num = p_hwfn->hw_info.resc_num;
+	struct qed_sb_cnt_info sb_cnt_info;
 	int num_funcs, i;
 
 	num_funcs = MAX_NUM_PFS_BB;
 
+	memset(&sb_cnt_info, 0, sizeof(sb_cnt_info));
+	qed_int_get_num_sbs(p_hwfn, &sb_cnt_info);
+
 	resc_num[QED_SB] = min_t(u32,
 				 (MAX_SB_PER_PATH_BB / num_funcs),
-				 qed_int_get_num_sbs(p_hwfn, NULL));
+				 sb_cnt_info.sb_cnt);
 	resc_num[QED_L2_QUEUE] = MAX_NUM_L2_QUEUES_BB / num_funcs;
 	resc_num[QED_VPORT] = MAX_NUM_VPORTS_BB / num_funcs;
 	resc_num[QED_RSS_ENG] = ETH_RSS_ENGINE_NUM_BB / num_funcs;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index fa73daa..7fd1be6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -343,17 +343,17 @@ void qed_int_sp_dpc(unsigned long hwfn_cookie)
 
 static void qed_int_sb_attn_free(struct qed_hwfn *p_hwfn)
 {
-	struct qed_dev *cdev   = p_hwfn->cdev;
-	struct qed_sb_attn_info *p_sb   = p_hwfn->p_sb_attn;
-
-	if (p_sb) {
-		if (p_sb->sb_attn)
-			dma_free_coherent(&cdev->pdev->dev,
-					  SB_ATTN_ALIGNED_SIZE(p_hwfn),
-					  p_sb->sb_attn,
-					  p_sb->sb_phys);
-		kfree(p_sb);
-	}
+	struct qed_sb_attn_info *p_sb = p_hwfn->p_sb_attn;
+
+	if (!p_sb)
+		return;
+
+	if (p_sb->sb_attn)
+		dma_free_coherent(&p_hwfn->cdev->pdev->dev,
+				  SB_ATTN_ALIGNED_SIZE(p_hwfn),
+				  p_sb->sb_attn,
+				  p_sb->sb_phys);
+	kfree(p_sb);
 }
 
 static void qed_int_sb_attn_setup(struct qed_hwfn *p_hwfn,
@@ -433,6 +433,7 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn,
 			   u16 vf_number,
 			   u8 vf_valid)
 {
+	struct qed_dev *cdev = p_hwfn->cdev;
 	u32 cau_state;
 
 	memset(p_sb_entry, 0, sizeof(*p_sb_entry));
@@ -451,14 +452,12 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn,
 
 	cau_state = CAU_HC_DISABLE_STATE;
 
-	if (p_hwfn->cdev->int_coalescing_mode == QED_COAL_MODE_ENABLE) {
+	if (cdev->int_coalescing_mode == QED_COAL_MODE_ENABLE) {
 		cau_state = CAU_HC_ENABLE_STATE;
-		if (!p_hwfn->cdev->rx_coalesce_usecs)
-			p_hwfn->cdev->rx_coalesce_usecs =
-				QED_CAU_DEF_RX_USECS;
-		if (!p_hwfn->cdev->tx_coalesce_usecs)
-			p_hwfn->cdev->tx_coalesce_usecs =
-				QED_CAU_DEF_TX_USECS;
+		if (!cdev->rx_coalesce_usecs)
+			cdev->rx_coalesce_usecs = QED_CAU_DEF_RX_USECS;
+		if (!cdev->tx_coalesce_usecs)
+			cdev->tx_coalesce_usecs = QED_CAU_DEF_TX_USECS;
 	}
 
 	SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE0, cau_state);
@@ -638,8 +637,10 @@ int qed_int_sb_release(struct qed_hwfn *p_hwfn,
 	sb_info->sb_ack = 0;
 	memset(sb_info->sb_virt, 0, sizeof(*sb_info->sb_virt));
 
-	p_hwfn->sbs_info[sb_id] = NULL;
-	p_hwfn->num_sbs--;
+	if (p_hwfn->sbs_info[sb_id] != NULL) {
+		p_hwfn->sbs_info[sb_id] = NULL;
+		p_hwfn->num_sbs--;
+	}
 
 	return 0;
 }
@@ -648,14 +649,15 @@ static void qed_int_sp_sb_free(struct qed_hwfn *p_hwfn)
 {
 	struct qed_sb_sp_info *p_sb = p_hwfn->p_sp_sb;
 
-	if (p_sb) {
-		if (p_sb->sb_info.sb_virt)
-			dma_free_coherent(&p_hwfn->cdev->pdev->dev,
-					  SB_ALIGNED_SIZE(p_hwfn),
-					  p_sb->sb_info.sb_virt,
-					  p_sb->sb_info.sb_phys);
-		kfree(p_sb);
-	}
+	if (!p_sb)
+		return;
+
+	if (p_sb->sb_info.sb_virt)
+		dma_free_coherent(&p_hwfn->cdev->pdev->dev,
+				  SB_ALIGNED_SIZE(p_hwfn),
+				  p_sb->sb_info.sb_virt,
+				  p_sb->sb_info.sb_phys);
+	kfree(p_sb);
 }
 
 static int qed_int_sp_sb_alloc(struct qed_hwfn *p_hwfn,
@@ -718,36 +720,36 @@ int qed_int_register_cb(struct qed_hwfn *p_hwfn,
 			__le16 **p_fw_cons)
 {
 	struct qed_sb_sp_info *p_sp_sb = p_hwfn->p_sp_sb;
-	int qed_status = -ENOMEM;
+	int rc = -ENOMEM;
 	u8 pi;
 
 	/* Look for a free index */
 	for (pi = 0; pi < ARRAY_SIZE(p_sp_sb->pi_info_arr); pi++) {
-		if (!p_sp_sb->pi_info_arr[pi].comp_cb) {
-			p_sp_sb->pi_info_arr[pi].comp_cb = comp_cb;
-			p_sp_sb->pi_info_arr[pi].cookie = cookie;
-			*sb_idx = pi;
-			*p_fw_cons = &p_sp_sb->sb_info.sb_virt->pi_array[pi];
-			qed_status = 0;
-			break;
-		}
+		if (p_sp_sb->pi_info_arr[pi].comp_cb)
+			continue;
+
+		p_sp_sb->pi_info_arr[pi].comp_cb = comp_cb;
+		p_sp_sb->pi_info_arr[pi].cookie = cookie;
+		*sb_idx = pi;
+		*p_fw_cons = &p_sp_sb->sb_info.sb_virt->pi_array[pi];
+		rc = 0;
+		break;
 	}
 
-	return qed_status;
+	return rc;
 }
 
 int qed_int_unregister_cb(struct qed_hwfn *p_hwfn, u8 pi)
 {
 	struct qed_sb_sp_info *p_sp_sb = p_hwfn->p_sp_sb;
-	int qed_status = -ENOMEM;
 
-	if (p_sp_sb->pi_info_arr[pi].comp_cb) {
-		p_sp_sb->pi_info_arr[pi].comp_cb = NULL;
-		p_sp_sb->pi_info_arr[pi].cookie = NULL;
-		qed_status = 0;
-	}
+	if (p_sp_sb->pi_info_arr[pi].comp_cb == NULL)
+		return -ENOMEM;
 
-	return qed_status;
+	p_sp_sb->pi_info_arr[pi].comp_cb = NULL;
+	p_sp_sb->pi_info_arr[pi].cookie = NULL;
+
+	return 0;
 }
 
 u16 qed_int_get_sp_sb_id(struct qed_hwfn *p_hwfn)
@@ -937,6 +939,39 @@ void qed_int_igu_init_pure_rt(struct qed_hwfn *p_hwfn,
 	}
 }
 
+static u32 qed_int_igu_read_cam_block(struct qed_hwfn	*p_hwfn,
+				      struct qed_ptt	*p_ptt,
+				      u16		sb_id)
+{
+	u32 val = qed_rd(p_hwfn, p_ptt,
+			 IGU_REG_MAPPING_MEMORY +
+			 sizeof(u32) * sb_id);
+	struct qed_igu_block *p_block;
+
+	p_block = &p_hwfn->hw_info.p_igu_info->igu_map.igu_blocks[sb_id];
+
+	/* stop scanning when hit first invalid PF entry */
+	if (!GET_FIELD(val, IGU_MAPPING_LINE_VALID) &&
+	    GET_FIELD(val, IGU_MAPPING_LINE_PF_VALID))
+		goto out;
+
+	/* Fill the block information */
+	p_block->status		= QED_IGU_STATUS_VALID;
+	p_block->function_id	= GET_FIELD(val,
+					    IGU_MAPPING_LINE_FUNCTION_NUMBER);
+	p_block->is_pf		= GET_FIELD(val, IGU_MAPPING_LINE_PF_VALID);
+	p_block->vector_number	= GET_FIELD(val,
+					    IGU_MAPPING_LINE_VECTOR_NUMBER);
+
+	DP_VERBOSE(p_hwfn, NETIF_MSG_INTR,
+		   "IGU_BLOCK: [SB 0x%04x, Value in CAM 0x%08x] func_id = %d is_pf = %d vector_num = 0x%x\n",
+		   sb_id, val, p_block->function_id,
+		   p_block->is_pf, p_block->vector_number);
+
+out:
+	return val;
+}
+
 int qed_int_igu_read_cam(struct qed_hwfn *p_hwfn,
 			 struct qed_ptt *p_ptt)
 {
@@ -963,26 +998,13 @@ int qed_int_igu_read_cam(struct qed_hwfn *p_hwfn,
 	     sb_id++) {
 		blk = &p_igu_info->igu_map.igu_blocks[sb_id];
 
-		val = qed_rd(p_hwfn, p_ptt,
-			     IGU_REG_MAPPING_MEMORY + sizeof(u32) * sb_id);
+		val	= qed_int_igu_read_cam_block(p_hwfn, p_ptt, sb_id);
 
 		/* stop scanning when hit first invalid PF entry */
 		if (!GET_FIELD(val, IGU_MAPPING_LINE_VALID) &&
 		    GET_FIELD(val, IGU_MAPPING_LINE_PF_VALID))
 			break;
 
-		blk->status = QED_IGU_STATUS_VALID;
-		blk->function_id = GET_FIELD(val,
-					     IGU_MAPPING_LINE_FUNCTION_NUMBER);
-		blk->is_pf = GET_FIELD(val, IGU_MAPPING_LINE_PF_VALID);
-		blk->vector_number = GET_FIELD(val,
-					       IGU_MAPPING_LINE_VECTOR_NUMBER);
-
-		DP_VERBOSE(p_hwfn, NETIF_MSG_INTR,
-			   "IGU_BLOCK[sb_id]:%x:func_id = %d is_pf = %d vector_num = 0x%x\n",
-			   val, blk->function_id, blk->is_pf,
-			   blk->vector_number);
-
 		if (blk->is_pf) {
 			if (blk->function_id == p_hwfn->rel_pf_id) {
 				blk->status |= QED_IGU_STATUS_PF;
@@ -1121,18 +1143,17 @@ void qed_int_setup(struct qed_hwfn *p_hwfn,
 	qed_int_sp_dpc_setup(p_hwfn);
 }
 
-int qed_int_get_num_sbs(struct qed_hwfn *p_hwfn,
-			int *p_iov_blks)
+void qed_int_get_num_sbs(struct qed_hwfn	*p_hwfn,
+			 struct qed_sb_cnt_info *p_sb_cnt_info)
 {
 	struct qed_igu_info *info = p_hwfn->hw_info.p_igu_info;
 
-	if (!info)
-		return 0;
-
-	if (p_iov_blks)
-		*p_iov_blks = info->free_blks;
+	if (!info || !p_sb_cnt_info)
+		return;
 
-	return info->igu_sb_cnt;
+	p_sb_cnt_info->sb_cnt		= info->igu_sb_cnt;
+	p_sb_cnt_info->sb_iov_cnt	= info->igu_sb_cnt_iov;
+	p_sb_cnt_info->sb_free_blk	= info->free_blks;
 }
 
 void qed_int_disable_post_isr_release(struct qed_dev *cdev)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h
index 51e0b09..c57f2e6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.h
@@ -161,12 +161,12 @@ void qed_int_sp_dpc(unsigned long hwfn_cookie);
  *        blocks configured for this funciton in the igu.
  *
  * @param p_hwfn
- * @param p_iov_blks - configured free blks for vfs
+ * @param p_sb_cnt_info
  *
  * @return int - number of status blocks configured
  */
-int qed_int_get_num_sbs(struct qed_hwfn *p_hwfn,
-			int *p_iov_blks);
+void qed_int_get_num_sbs(struct qed_hwfn	*p_hwfn,
+			 struct qed_sb_cnt_info *p_sb_cnt_info);
 
 /**
  * @brief qed_int_disable_post_isr_release - performs the cleanup post ISR
diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index 25d6e91..caa689e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -634,15 +634,18 @@ static int qed_get_int_fp(struct qed_dev *cdev, struct qed_int_info *info)
 static int qed_slowpath_setup_int(struct qed_dev *cdev,
 				  enum qed_int_mode int_mode)
 {
-	int rc, i;
-	u8 num_vectors = 0;
-
+	struct qed_sb_cnt_info sb_cnt_info;
+	int rc;
+	int i;
 	memset(&cdev->int_params, 0, sizeof(struct qed_int_params));
 
 	cdev->int_params.in.int_mode = int_mode;
-	for_each_hwfn(cdev, i)
-		num_vectors +=  qed_int_get_num_sbs(&cdev->hwfns[i], NULL) + 1;
-	cdev->int_params.in.num_vectors = num_vectors;
+	for_each_hwfn(cdev, i) {
+		memset(&sb_cnt_info, 0, sizeof(sb_cnt_info));
+		qed_int_get_num_sbs(&cdev->hwfns[i], &sb_cnt_info);
+		cdev->int_params.in.num_vectors += sb_cnt_info.sb_cnt;
+		cdev->int_params.in.num_vectors++; /* slowpath */
+	}
 
 	/* We want a minimum of one slowpath and one fastpath vector per hwfn */
 	cdev->int_params.in.min_msix_cnt = cdev->num_hwfns * 2;
diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h
index 3d43c1d..1f7599c7 100644
--- a/include/linux/qed/qed_if.h
+++ b/include/linux/qed/qed_if.h
@@ -446,6 +446,12 @@ struct qed_eth_stats {
 #define RX_PI           0
 #define TX_PI(tc)       (RX_PI + 1 + tc)
 
+struct qed_sb_cnt_info {
+	int	sb_cnt;
+	int	sb_iov_cnt;
+	int	sb_free_blk;
+};
+
 static inline u16 qed_sb_update_sb_idx(struct qed_sb_info *sb_info)
 {
 	u32 prod = 0;
-- 
cgit v0.10.2


From 0d956e8a65d53e0d1a71d28975c821cf0f6ba676 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 28 Feb 2016 12:26:53 +0200
Subject: qed: Add support for HW attentions

HW is capable of generating attentnions for a multitude of reasons,
but current driver is enabling attention generation only for management
firmware [required for link notifications].

This patch enables almost all of the possible reasons for HW attentions,
logging the HW block generating the attention and preventing further
attentions from that source [to prevent possible attention flood].
It also lays the infrastructure for additional exploration of the various
attentions.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index 7fd1be6..c914ac5 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -42,21 +42,210 @@ struct qed_sb_sp_info {
 #define SB_ATTN_ALIGNED_SIZE(p_hwfn) \
 	ALIGNED_TYPE_SIZE(struct atten_status_block, p_hwfn)
 
-#define ATTN_STATE_BITS (0xfff)
+struct aeu_invert_reg_bit {
+	char bit_name[30];
+
+#define ATTENTION_PARITY                (1 << 0)
+
+#define ATTENTION_LENGTH_MASK           (0x00000ff0)
+#define ATTENTION_LENGTH_SHIFT          (4)
+#define ATTENTION_LENGTH(flags)         (((flags) & ATTENTION_LENGTH_MASK) >> \
+					 ATTENTION_LENGTH_SHIFT)
+#define ATTENTION_SINGLE                (1 << ATTENTION_LENGTH_SHIFT)
+#define ATTENTION_PAR                   (ATTENTION_SINGLE | ATTENTION_PARITY)
+#define ATTENTION_PAR_INT               ((2 << ATTENTION_LENGTH_SHIFT) | \
+					 ATTENTION_PARITY)
+
+/* Multiple bits start with this offset */
+#define ATTENTION_OFFSET_MASK           (0x000ff000)
+#define ATTENTION_OFFSET_SHIFT          (12)
+	unsigned int flags;
+};
+
+struct aeu_invert_reg {
+	struct aeu_invert_reg_bit bits[32];
+};
+
+#define MAX_ATTN_GRPS           (8)
+#define NUM_ATTN_REGS           (9)
+
+/* Notice aeu_invert_reg must be defined in the same order of bits as HW;  */
+static struct aeu_invert_reg aeu_descs[NUM_ATTN_REGS] = {
+	{
+		{       /* After Invert 1 */
+			{"GPIO0 function%d",
+			 (32 << ATTENTION_LENGTH_SHIFT)},
+		}
+	},
+
+	{
+		{       /* After Invert 2 */
+			{"PGLUE config_space", ATTENTION_SINGLE},
+			{"PGLUE misc_flr", ATTENTION_SINGLE},
+			{"PGLUE B RBC", ATTENTION_PAR_INT},
+			{"PGLUE misc_mctp", ATTENTION_SINGLE},
+			{"Flash event", ATTENTION_SINGLE},
+			{"SMB event", ATTENTION_SINGLE},
+			{"Main Power", ATTENTION_SINGLE},
+			{"SW timers #%d", (8 << ATTENTION_LENGTH_SHIFT) |
+					  (1 << ATTENTION_OFFSET_SHIFT)},
+			{"PCIE glue/PXP VPD %d",
+			 (16 << ATTENTION_LENGTH_SHIFT)},
+		}
+	},
+
+	{
+		{       /* After Invert 3 */
+			{"General Attention %d",
+			 (32 << ATTENTION_LENGTH_SHIFT)},
+		}
+	},
+
+	{
+		{       /* After Invert 4 */
+			{"General Attention 32", ATTENTION_SINGLE},
+			{"General Attention %d",
+			 (2 << ATTENTION_LENGTH_SHIFT) |
+			 (33 << ATTENTION_OFFSET_SHIFT)},
+			{"General Attention 35", ATTENTION_SINGLE},
+			{"CNIG port %d", (4 << ATTENTION_LENGTH_SHIFT)},
+			{"MCP CPU", ATTENTION_SINGLE},
+			{"MCP Watchdog timer", ATTENTION_SINGLE},
+			{"MCP M2P", ATTENTION_SINGLE},
+			{"AVS stop status ready", ATTENTION_SINGLE},
+			{"MSTAT", ATTENTION_PAR_INT},
+			{"MSTAT per-path", ATTENTION_PAR_INT},
+			{"Reserved %d", (6 << ATTENTION_LENGTH_SHIFT)},
+			{"NIG", ATTENTION_PAR_INT},
+			{"BMB/OPTE/MCP", ATTENTION_PAR_INT},
+			{"BTB",	ATTENTION_PAR_INT},
+			{"BRB",	ATTENTION_PAR_INT},
+			{"PRS",	ATTENTION_PAR_INT},
+		}
+	},
+
+	{
+		{       /* After Invert 5 */
+			{"SRC", ATTENTION_PAR_INT},
+			{"PB Client1", ATTENTION_PAR_INT},
+			{"PB Client2", ATTENTION_PAR_INT},
+			{"RPB", ATTENTION_PAR_INT},
+			{"PBF", ATTENTION_PAR_INT},
+			{"QM", ATTENTION_PAR_INT},
+			{"TM", ATTENTION_PAR_INT},
+			{"MCM",  ATTENTION_PAR_INT},
+			{"MSDM", ATTENTION_PAR_INT},
+			{"MSEM", ATTENTION_PAR_INT},
+			{"PCM", ATTENTION_PAR_INT},
+			{"PSDM", ATTENTION_PAR_INT},
+			{"PSEM", ATTENTION_PAR_INT},
+			{"TCM", ATTENTION_PAR_INT},
+			{"TSDM", ATTENTION_PAR_INT},
+			{"TSEM", ATTENTION_PAR_INT},
+		}
+	},
+
+	{
+		{       /* After Invert 6 */
+			{"UCM", ATTENTION_PAR_INT},
+			{"USDM", ATTENTION_PAR_INT},
+			{"USEM", ATTENTION_PAR_INT},
+			{"XCM",	ATTENTION_PAR_INT},
+			{"XSDM", ATTENTION_PAR_INT},
+			{"XSEM", ATTENTION_PAR_INT},
+			{"YCM",	ATTENTION_PAR_INT},
+			{"YSDM", ATTENTION_PAR_INT},
+			{"YSEM", ATTENTION_PAR_INT},
+			{"XYLD", ATTENTION_PAR_INT},
+			{"TMLD", ATTENTION_PAR_INT},
+			{"MYLD", ATTENTION_PAR_INT},
+			{"YULD", ATTENTION_PAR_INT},
+			{"DORQ", ATTENTION_PAR_INT},
+			{"DBG", ATTENTION_PAR_INT},
+			{"IPC",	ATTENTION_PAR_INT},
+		}
+	},
+
+	{
+		{       /* After Invert 7 */
+			{"CCFC", ATTENTION_PAR_INT},
+			{"CDU", ATTENTION_PAR_INT},
+			{"DMAE", ATTENTION_PAR_INT},
+			{"IGU", ATTENTION_PAR_INT},
+			{"ATC", ATTENTION_PAR_INT},
+			{"CAU", ATTENTION_PAR_INT},
+			{"PTU", ATTENTION_PAR_INT},
+			{"PRM", ATTENTION_PAR_INT},
+			{"TCFC", ATTENTION_PAR_INT},
+			{"RDIF", ATTENTION_PAR_INT},
+			{"TDIF", ATTENTION_PAR_INT},
+			{"RSS", ATTENTION_PAR_INT},
+			{"MISC", ATTENTION_PAR_INT},
+			{"MISCS", ATTENTION_PAR_INT},
+			{"PCIE", ATTENTION_PAR},
+			{"Vaux PCI core", ATTENTION_SINGLE},
+			{"PSWRQ", ATTENTION_PAR_INT},
+		}
+	},
+
+	{
+		{       /* After Invert 8 */
+			{"PSWRQ (pci_clk)", ATTENTION_PAR_INT},
+			{"PSWWR", ATTENTION_PAR_INT},
+			{"PSWWR (pci_clk)", ATTENTION_PAR_INT},
+			{"PSWRD", ATTENTION_PAR_INT},
+			{"PSWRD (pci_clk)", ATTENTION_PAR_INT},
+			{"PSWHST", ATTENTION_PAR_INT},
+			{"PSWHST (pci_clk)", ATTENTION_PAR_INT},
+			{"GRC",	ATTENTION_PAR_INT},
+			{"CPMU", ATTENTION_PAR_INT},
+			{"NCSI", ATTENTION_PAR_INT},
+			{"MSEM PRAM", ATTENTION_PAR},
+			{"PSEM PRAM", ATTENTION_PAR},
+			{"TSEM PRAM", ATTENTION_PAR},
+			{"USEM PRAM", ATTENTION_PAR},
+			{"XSEM PRAM", ATTENTION_PAR},
+			{"YSEM PRAM", ATTENTION_PAR},
+			{"pxp_misc_mps", ATTENTION_PAR},
+			{"PCIE glue/PXP Exp. ROM", ATTENTION_SINGLE},
+			{"PERST_B assertion", ATTENTION_SINGLE},
+			{"PERST_B deassertion", ATTENTION_SINGLE},
+			{"Reserved %d", (2 << ATTENTION_LENGTH_SHIFT)},
+		}
+	},
+
+	{
+		{       /* After Invert 9 */
+			{"MCP Latched memory", ATTENTION_PAR},
+			{"MCP Latched scratchpad cache", ATTENTION_SINGLE},
+			{"MCP Latched ump_tx", ATTENTION_PAR},
+			{"MCP Latched scratchpad", ATTENTION_PAR},
+			{"Reserved %d", (28 << ATTENTION_LENGTH_SHIFT)},
+		}
+	},
+};
+
+#define ATTN_STATE_BITS         (0xfff)
 #define ATTN_BITS_MASKABLE      (0x3ff)
 struct qed_sb_attn_info {
 	/* Virtual & Physical address of the SB */
 	struct atten_status_block       *sb_attn;
-	dma_addr_t		      sb_phys;
+	dma_addr_t			sb_phys;
 
 	/* Last seen running index */
-	u16			     index;
+	u16				index;
+
+	/* A mask of the AEU bits resulting in a parity error */
+	u32				parity_mask[NUM_ATTN_REGS];
+
+	/* A pointer to the attention description structure */
+	struct aeu_invert_reg		*p_aeu_desc;
 
 	/* Previously asserted attentions, which are still unasserted */
-	u16			     known_attn;
+	u16				known_attn;
 
 	/* Cleanup address for the link's general hw attention */
-	u32			     mfw_attn_addr;
+	u32				mfw_attn_addr;
 };
 
 static inline u16 qed_attn_update_idx(struct qed_hwfn *p_hwfn,
@@ -128,6 +317,39 @@ static int qed_int_assertion(struct qed_hwfn *p_hwfn,
 }
 
 /**
+ * @brief qed_int_deassertion_aeu_bit - handles the effects of a single
+ * cause of the attention
+ *
+ * @param p_hwfn
+ * @param p_aeu - descriptor of an AEU bit which caused the attention
+ * @param aeu_en_reg - register offset of the AEU enable reg. which configured
+ *  this bit to this group.
+ * @param bit_index - index of this bit in the aeu_en_reg
+ *
+ * @return int
+ */
+static int
+qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn,
+			    struct aeu_invert_reg_bit *p_aeu,
+			    u32 aeu_en_reg,
+			    u32 bitmask)
+{
+	int rc = -EINVAL;
+	u32 val, mask = ~bitmask;
+
+	DP_INFO(p_hwfn, "Deasserted attention `%s'[%08x]\n",
+		p_aeu->bit_name, bitmask);
+
+	/* Prevent this Attention from being asserted in the future */
+	val = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg);
+	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg, (val & mask));
+	DP_INFO(p_hwfn, "`%s' - Disabled future attentions\n",
+		p_aeu->bit_name);
+
+	return rc;
+}
+
+/**
  * @brief - handles deassertion of previously asserted attentions.
  *
  * @param p_hwfn
@@ -139,17 +361,110 @@ static int qed_int_deassertion(struct qed_hwfn  *p_hwfn,
 			       u16 deasserted_bits)
 {
 	struct qed_sb_attn_info *sb_attn_sw = p_hwfn->p_sb_attn;
-	u32 aeu_mask;
+	u32 aeu_inv_arr[NUM_ATTN_REGS], aeu_mask;
+	u8 i, j, k, bit_idx;
+	int rc = 0;
+
+	/* Read the attention registers in the AEU */
+	for (i = 0; i < NUM_ATTN_REGS; i++) {
+		aeu_inv_arr[i] = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+					MISC_REG_AEU_AFTER_INVERT_1_IGU +
+					i * 0x4);
+		DP_VERBOSE(p_hwfn, NETIF_MSG_INTR,
+			   "Deasserted bits [%d]: %08x\n",
+			   i, aeu_inv_arr[i]);
+	}
+
+	/* Find parity attentions first */
+	for (i = 0; i < NUM_ATTN_REGS; i++) {
+		struct aeu_invert_reg *p_aeu = &sb_attn_sw->p_aeu_desc[i];
+		u32 en = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				MISC_REG_AEU_ENABLE1_IGU_OUT_0 +
+				i * sizeof(u32));
+		u32 parities;
+
+		/* Skip register in which no parity bit is currently set */
+		parities = sb_attn_sw->parity_mask[i] & aeu_inv_arr[i] & en;
+		if (!parities)
+			continue;
 
-	if (deasserted_bits != 0x100)
-		DP_ERR(p_hwfn, "Unexpected - non-link deassertion\n");
+		for (j = 0, bit_idx = 0; bit_idx < 32; j++) {
+			struct aeu_invert_reg_bit *p_bit = &p_aeu->bits[j];
+
+			if ((p_bit->flags & ATTENTION_PARITY) &&
+			    !!(parities & (1 << bit_idx))) {
+				DP_INFO(p_hwfn,
+					"%s[%d] parity attention is set\n",
+					p_bit->bit_name, bit_idx);
+			}
+
+			bit_idx += ATTENTION_LENGTH(p_bit->flags);
+		}
+	}
+
+	/* Find non-parity cause for attention and act */
+	for (k = 0; k < MAX_ATTN_GRPS; k++) {
+		struct aeu_invert_reg_bit *p_aeu;
+
+		/* Handle only groups whose attention is currently deasserted */
+		if (!(deasserted_bits & (1 << k)))
+			continue;
+
+		for (i = 0; i < NUM_ATTN_REGS; i++) {
+			u32 aeu_en = MISC_REG_AEU_ENABLE1_IGU_OUT_0 +
+				     i * sizeof(u32) +
+				     k * sizeof(u32) * NUM_ATTN_REGS;
+			u32 en, bits;
+
+			en = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en);
+			bits = aeu_inv_arr[i] & en;
+
+			/* Skip if no bit from this group is currently set */
+			if (!bits)
+				continue;
+
+			/* Find all set bits from current register which belong
+			 * to current group, making them responsible for the
+			 * previous assertion.
+			 */
+			for (j = 0, bit_idx = 0; bit_idx < 32; j++) {
+				u8 bit, bit_len;
+				u32 bitmask;
+
+				p_aeu = &sb_attn_sw->p_aeu_desc[i].bits[j];
+
+				/* No need to handle parity-only bits */
+				if (p_aeu->flags == ATTENTION_PAR)
+					continue;
+
+				bit = bit_idx;
+				bit_len = ATTENTION_LENGTH(p_aeu->flags);
+				if (p_aeu->flags & ATTENTION_PAR_INT) {
+					/* Skip Parity */
+					bit++;
+					bit_len--;
+				}
+
+				bitmask = bits & (((1 << bit_len) - 1) << bit);
+				if (bitmask) {
+					/* Handle source of the attention */
+					qed_int_deassertion_aeu_bit(p_hwfn,
+								    p_aeu,
+								    aeu_en,
+								    bitmask);
+				}
+
+				bit_idx += ATTENTION_LENGTH(p_aeu->flags);
+			}
+		}
+	}
 
 	/* Clear IGU indication for the deasserted bits */
 	DIRECT_REG_WR((u8 __iomem *)p_hwfn->regview +
-		      GTT_BAR0_MAP_REG_IGU_CMD +
-		      ((IGU_CMD_ATTN_BIT_CLR_UPPER -
-			IGU_CMD_INT_ACK_BASE) << 3),
-		      ~((u32)deasserted_bits));
+				    GTT_BAR0_MAP_REG_IGU_CMD +
+				    ((IGU_CMD_ATTN_BIT_CLR_UPPER -
+				      IGU_CMD_INT_ACK_BASE) << 3),
+				    ~((u32)deasserted_bits));
 
 	/* Unmask deasserted attentions in IGU */
 	aeu_mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
@@ -160,7 +475,7 @@ static int qed_int_deassertion(struct qed_hwfn  *p_hwfn,
 	/* Clear deassertion from inner state */
 	sb_attn_sw->known_attn &= ~deasserted_bits;
 
-	return 0;
+	return rc;
 }
 
 static int qed_int_attentions(struct qed_hwfn *p_hwfn)
@@ -379,10 +694,31 @@ static void qed_int_sb_attn_init(struct qed_hwfn *p_hwfn,
 				 dma_addr_t sb_phy_addr)
 {
 	struct qed_sb_attn_info *sb_info = p_hwfn->p_sb_attn;
+	int i, j, k;
 
 	sb_info->sb_attn = sb_virt_addr;
 	sb_info->sb_phys = sb_phy_addr;
 
+	/* Set the pointer to the AEU descriptors */
+	sb_info->p_aeu_desc = aeu_descs;
+
+	/* Calculate Parity Masks */
+	memset(sb_info->parity_mask, 0, sizeof(u32) * NUM_ATTN_REGS);
+	for (i = 0; i < NUM_ATTN_REGS; i++) {
+		/* j is array index, k is bit index */
+		for (j = 0, k = 0; k < 32; j++) {
+			unsigned int flags = aeu_descs[i].bits[j].flags;
+
+			if (flags & ATTENTION_PARITY)
+				sb_info->parity_mask[i] |= 1 << k;
+
+			k += ATTENTION_LENGTH(flags);
+		}
+		DP_VERBOSE(p_hwfn, NETIF_MSG_INTR,
+			   "Attn Mask [Reg %d]: 0x%08x\n",
+			   i, sb_info->parity_mask[i]);
+	}
+
 	/* Set the address of cleanup for the mcp attention */
 	sb_info->mfw_attn_addr = (p_hwfn->rel_pf_id << 3) +
 				 MISC_REG_AEU_GENERAL_ATTN_0;
@@ -694,25 +1030,6 @@ static int qed_int_sp_sb_alloc(struct qed_hwfn *p_hwfn,
 	return 0;
 }
 
-static void qed_int_sp_sb_setup(struct qed_hwfn *p_hwfn,
-				struct qed_ptt *p_ptt)
-{
-	if (!p_hwfn)
-		return;
-
-	if (p_hwfn->p_sp_sb)
-		qed_int_sb_setup(p_hwfn, p_ptt, &p_hwfn->p_sp_sb->sb_info);
-	else
-		DP_NOTICE(p_hwfn->cdev,
-			  "Failed to setup Slow path status block - NULL pointer\n");
-
-	if (p_hwfn->p_sb_attn)
-		qed_int_sb_attn_setup(p_hwfn, p_ptt);
-	else
-		DP_NOTICE(p_hwfn->cdev,
-			  "Failed to setup attentions status block - NULL pointer\n");
-}
-
 int qed_int_register_cb(struct qed_hwfn *p_hwfn,
 			qed_int_comp_cb_t comp_cb,
 			void *cookie,
@@ -788,16 +1105,13 @@ void qed_int_igu_enable_int(struct qed_hwfn *p_hwfn,
 int qed_int_igu_enable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt,
 		       enum qed_int_mode int_mode)
 {
-	int rc, i;
-
-	/* Mask non-link attentions */
-	for (i = 0; i < 9; i++)
-		qed_wr(p_hwfn, p_ptt,
-		       MISC_REG_AEU_ENABLE1_IGU_OUT_0 + (i << 2), 0);
+	int rc;
 
-	/* Configure AEU signal change to produce attentions for link */
+	/* Configure AEU signal change to produce attentions */
+	qed_wr(p_hwfn, p_ptt, IGU_REG_ATTENTION_ENABLE, 0);
 	qed_wr(p_hwfn, p_ptt, IGU_REG_LEADING_EDGE_LATCH, 0xfff);
 	qed_wr(p_hwfn, p_ptt, IGU_REG_TRAILING_EDGE_LATCH, 0xfff);
+	qed_wr(p_hwfn, p_ptt, IGU_REG_ATTENTION_ENABLE, 0xfff);
 
 	/* Flush the writes to IGU */
 	mmiowb();
@@ -1139,7 +1453,8 @@ void qed_int_free(struct qed_hwfn *p_hwfn)
 void qed_int_setup(struct qed_hwfn *p_hwfn,
 		   struct qed_ptt *p_ptt)
 {
-	qed_int_sp_sb_setup(p_hwfn, p_ptt);
+	qed_int_sb_setup(p_hwfn, p_ptt, &p_hwfn->p_sp_sb->sb_info);
+	qed_int_sb_attn_setup(p_hwfn, p_ptt);
 	qed_int_sp_dpc_setup(p_hwfn);
 }
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
index e8df123..2286e75 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
@@ -333,6 +333,8 @@
 	0x180800UL
 #define  MISC_REG_AEU_ENABLE1_IGU_OUT_0 \
 	0x00849cUL
+#define MISC_REG_AEU_AFTER_INVERT_1_IGU	\
+	0x0087b4UL
 #define  MISC_REG_AEU_MASK_ATTN_IGU \
 	0x008494UL
 #define  IGU_REG_CLEANUP_STATUS_0 \
-- 
cgit v0.10.2


From ff38577aa9534fed1434d2ff8a0d1666a7f11fe4 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 28 Feb 2016 12:26:54 +0200
Subject: qed: Print HW attention reasons

Each HW block contains common information about attention reasons,
raising a bit for each one of the different sub-reasons that caused it
to raise an attention.

This patch extends the infrastructure by allowing logging of the various
reasons causing the HW blocks to generate an attention.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index c914ac5..c8bca77 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -39,6 +39,11 @@ struct qed_sb_sp_info {
 	struct qed_pi_info	pi_info_arr[PIS_PER_SB];
 };
 
+enum qed_attention_type {
+	QED_ATTN_TYPE_ATTN,
+	QED_ATTN_TYPE_PARITY,
+};
+
 #define SB_ATTN_ALIGNED_SIZE(p_hwfn) \
 	ALIGNED_TYPE_SIZE(struct atten_status_block, p_hwfn)
 
@@ -60,6 +65,8 @@ struct aeu_invert_reg_bit {
 #define ATTENTION_OFFSET_MASK           (0x000ff000)
 #define ATTENTION_OFFSET_SHIFT          (12)
 	unsigned int flags;
+
+	enum block_id block_index;
 };
 
 struct aeu_invert_reg {
@@ -69,158 +76,1379 @@ struct aeu_invert_reg {
 #define MAX_ATTN_GRPS           (8)
 #define NUM_ATTN_REGS           (9)
 
+/* HW Attention register */
+struct attn_hw_reg {
+	u16 reg_idx;             /* Index of this register in its block */
+	u16 num_of_bits;         /* number of valid attention bits */
+	u32 sts_addr;            /* Address of the STS register */
+	u32 sts_clr_addr;        /* Address of the STS_CLR register */
+	u32 sts_wr_addr;         /* Address of the STS_WR register */
+	u32 mask_addr;           /* Address of the MASK register */
+};
+
+/* HW block attention registers */
+struct attn_hw_regs {
+	u16 num_of_int_regs;            /* Number of interrupt regs */
+	u16 num_of_prty_regs;           /* Number of parity regs */
+	struct attn_hw_reg **int_regs;  /* interrupt regs */
+	struct attn_hw_reg **prty_regs; /* parity regs */
+};
+
+/* HW block attention registers */
+struct attn_hw_block {
+	const char *name;                 /* Block name */
+	struct attn_hw_regs chip_regs[1];
+};
+
+static struct attn_hw_reg grc_int0_bb_b0 = {
+	0, 4, 0x50180, 0x5018c, 0x50188, 0x50184};
+
+static struct attn_hw_reg *grc_int_bb_b0_regs[1] = {
+	&grc_int0_bb_b0};
+
+static struct attn_hw_reg grc_prty1_bb_b0 = {
+	0, 2, 0x50200, 0x5020c, 0x50208, 0x50204};
+
+static struct attn_hw_reg *grc_prty_bb_b0_regs[1] = {
+	&grc_prty1_bb_b0};
+
+static struct attn_hw_reg miscs_int0_bb_b0 = {
+	0, 3, 0x9180, 0x918c, 0x9188, 0x9184};
+
+static struct attn_hw_reg miscs_int1_bb_b0 = {
+	1, 11, 0x9190, 0x919c, 0x9198, 0x9194};
+
+static struct attn_hw_reg *miscs_int_bb_b0_regs[2] = {
+	&miscs_int0_bb_b0, &miscs_int1_bb_b0};
+
+static struct attn_hw_reg miscs_prty0_bb_b0 = {
+	0, 1, 0x91a0, 0x91ac, 0x91a8, 0x91a4};
+
+static struct attn_hw_reg *miscs_prty_bb_b0_regs[1] = {
+	&miscs_prty0_bb_b0};
+
+static struct attn_hw_reg misc_int0_bb_b0 = {
+	0, 1, 0x8180, 0x818c, 0x8188, 0x8184};
+
+static struct attn_hw_reg *misc_int_bb_b0_regs[1] = {
+	&misc_int0_bb_b0};
+
+static struct attn_hw_reg pglue_b_int0_bb_b0 = {
+	0, 23, 0x2a8180, 0x2a818c, 0x2a8188, 0x2a8184};
+
+static struct attn_hw_reg *pglue_b_int_bb_b0_regs[1] = {
+	&pglue_b_int0_bb_b0};
+
+static struct attn_hw_reg pglue_b_prty0_bb_b0 = {
+	0, 1, 0x2a8190, 0x2a819c, 0x2a8198, 0x2a8194};
+
+static struct attn_hw_reg pglue_b_prty1_bb_b0 = {
+	1, 22, 0x2a8200, 0x2a820c, 0x2a8208, 0x2a8204};
+
+static struct attn_hw_reg *pglue_b_prty_bb_b0_regs[2] = {
+	&pglue_b_prty0_bb_b0, &pglue_b_prty1_bb_b0};
+
+static struct attn_hw_reg cnig_int0_bb_b0 = {
+	0, 6, 0x2182e8, 0x2182f4, 0x2182f0, 0x2182ec};
+
+static struct attn_hw_reg *cnig_int_bb_b0_regs[1] = {
+	&cnig_int0_bb_b0};
+
+static struct attn_hw_reg cnig_prty0_bb_b0 = {
+	0, 2, 0x218348, 0x218354, 0x218350, 0x21834c};
+
+static struct attn_hw_reg *cnig_prty_bb_b0_regs[1] = {
+	&cnig_prty0_bb_b0};
+
+static struct attn_hw_reg cpmu_int0_bb_b0 = {
+	0, 1, 0x303e0, 0x303ec, 0x303e8, 0x303e4};
+
+static struct attn_hw_reg *cpmu_int_bb_b0_regs[1] = {
+	&cpmu_int0_bb_b0};
+
+static struct attn_hw_reg ncsi_int0_bb_b0 = {
+	0, 1, 0x404cc, 0x404d8, 0x404d4, 0x404d0};
+
+static struct attn_hw_reg *ncsi_int_bb_b0_regs[1] = {
+	&ncsi_int0_bb_b0};
+
+static struct attn_hw_reg ncsi_prty1_bb_b0 = {
+	0, 1, 0x40000, 0x4000c, 0x40008, 0x40004};
+
+static struct attn_hw_reg *ncsi_prty_bb_b0_regs[1] = {
+	&ncsi_prty1_bb_b0};
+
+static struct attn_hw_reg opte_prty1_bb_b0 = {
+	0, 11, 0x53000, 0x5300c, 0x53008, 0x53004};
+
+static struct attn_hw_reg opte_prty0_bb_b0 = {
+	1, 1, 0x53208, 0x53214, 0x53210, 0x5320c};
+
+static struct attn_hw_reg *opte_prty_bb_b0_regs[2] = {
+	&opte_prty1_bb_b0, &opte_prty0_bb_b0};
+
+static struct attn_hw_reg bmb_int0_bb_b0 = {
+	0, 16, 0x5400c0, 0x5400cc, 0x5400c8, 0x5400c4};
+
+static struct attn_hw_reg bmb_int1_bb_b0 = {
+	1, 28, 0x5400d8, 0x5400e4, 0x5400e0, 0x5400dc};
+
+static struct attn_hw_reg bmb_int2_bb_b0 = {
+	2, 26, 0x5400f0, 0x5400fc, 0x5400f8, 0x5400f4};
+
+static struct attn_hw_reg bmb_int3_bb_b0 = {
+	3, 31, 0x540108, 0x540114, 0x540110, 0x54010c};
+
+static struct attn_hw_reg bmb_int4_bb_b0 = {
+	4, 27, 0x540120, 0x54012c, 0x540128, 0x540124};
+
+static struct attn_hw_reg bmb_int5_bb_b0 = {
+	5, 29, 0x540138, 0x540144, 0x540140, 0x54013c};
+
+static struct attn_hw_reg bmb_int6_bb_b0 = {
+	6, 30, 0x540150, 0x54015c, 0x540158, 0x540154};
+
+static struct attn_hw_reg bmb_int7_bb_b0 = {
+	7, 32, 0x540168, 0x540174, 0x540170, 0x54016c};
+
+static struct attn_hw_reg bmb_int8_bb_b0 = {
+	8, 32, 0x540184, 0x540190, 0x54018c, 0x540188};
+
+static struct attn_hw_reg bmb_int9_bb_b0 = {
+	9, 32, 0x54019c, 0x5401a8, 0x5401a4, 0x5401a0};
+
+static struct attn_hw_reg bmb_int10_bb_b0 = {
+	10, 3, 0x5401b4, 0x5401c0, 0x5401bc, 0x5401b8};
+
+static struct attn_hw_reg bmb_int11_bb_b0 = {
+	11, 4, 0x5401cc, 0x5401d8, 0x5401d4, 0x5401d0};
+
+static struct attn_hw_reg *bmb_int_bb_b0_regs[12] = {
+	&bmb_int0_bb_b0, &bmb_int1_bb_b0, &bmb_int2_bb_b0, &bmb_int3_bb_b0,
+	&bmb_int4_bb_b0, &bmb_int5_bb_b0, &bmb_int6_bb_b0, &bmb_int7_bb_b0,
+	&bmb_int8_bb_b0, &bmb_int9_bb_b0, &bmb_int10_bb_b0, &bmb_int11_bb_b0};
+
+static struct attn_hw_reg bmb_prty0_bb_b0 = {
+	0, 5, 0x5401dc, 0x5401e8, 0x5401e4, 0x5401e0};
+
+static struct attn_hw_reg bmb_prty1_bb_b0 = {
+	1, 31, 0x540400, 0x54040c, 0x540408, 0x540404};
+
+static struct attn_hw_reg bmb_prty2_bb_b0 = {
+	2, 15, 0x540410, 0x54041c, 0x540418, 0x540414};
+
+static struct attn_hw_reg *bmb_prty_bb_b0_regs[3] = {
+	&bmb_prty0_bb_b0, &bmb_prty1_bb_b0, &bmb_prty2_bb_b0};
+
+static struct attn_hw_reg pcie_prty1_bb_b0 = {
+	0, 17, 0x54000, 0x5400c, 0x54008, 0x54004};
+
+static struct attn_hw_reg *pcie_prty_bb_b0_regs[1] = {
+	&pcie_prty1_bb_b0};
+
+static struct attn_hw_reg mcp2_prty0_bb_b0 = {
+	0, 1, 0x52040, 0x5204c, 0x52048, 0x52044};
+
+static struct attn_hw_reg mcp2_prty1_bb_b0 = {
+	1, 12, 0x52204, 0x52210, 0x5220c, 0x52208};
+
+static struct attn_hw_reg *mcp2_prty_bb_b0_regs[2] = {
+	&mcp2_prty0_bb_b0, &mcp2_prty1_bb_b0};
+
+static struct attn_hw_reg pswhst_int0_bb_b0 = {
+	0, 18, 0x2a0180, 0x2a018c, 0x2a0188, 0x2a0184};
+
+static struct attn_hw_reg *pswhst_int_bb_b0_regs[1] = {
+	&pswhst_int0_bb_b0};
+
+static struct attn_hw_reg pswhst_prty0_bb_b0 = {
+	0, 1, 0x2a0190, 0x2a019c, 0x2a0198, 0x2a0194};
+
+static struct attn_hw_reg pswhst_prty1_bb_b0 = {
+	1, 17, 0x2a0200, 0x2a020c, 0x2a0208, 0x2a0204};
+
+static struct attn_hw_reg *pswhst_prty_bb_b0_regs[2] = {
+	&pswhst_prty0_bb_b0, &pswhst_prty1_bb_b0};
+
+static struct attn_hw_reg pswhst2_int0_bb_b0 = {
+	0, 5, 0x29e180, 0x29e18c, 0x29e188, 0x29e184};
+
+static struct attn_hw_reg *pswhst2_int_bb_b0_regs[1] = {
+	&pswhst2_int0_bb_b0};
+
+static struct attn_hw_reg pswhst2_prty0_bb_b0 = {
+	0, 1, 0x29e190, 0x29e19c, 0x29e198, 0x29e194};
+
+static struct attn_hw_reg *pswhst2_prty_bb_b0_regs[1] = {
+	&pswhst2_prty0_bb_b0};
+
+static struct attn_hw_reg pswrd_int0_bb_b0 = {
+	0, 3, 0x29c180, 0x29c18c, 0x29c188, 0x29c184};
+
+static struct attn_hw_reg *pswrd_int_bb_b0_regs[1] = {
+	&pswrd_int0_bb_b0};
+
+static struct attn_hw_reg pswrd_prty0_bb_b0 = {
+	0, 1, 0x29c190, 0x29c19c, 0x29c198, 0x29c194};
+
+static struct attn_hw_reg *pswrd_prty_bb_b0_regs[1] = {
+	&pswrd_prty0_bb_b0};
+
+static struct attn_hw_reg pswrd2_int0_bb_b0 = {
+	0, 5, 0x29d180, 0x29d18c, 0x29d188, 0x29d184};
+
+static struct attn_hw_reg *pswrd2_int_bb_b0_regs[1] = {
+	&pswrd2_int0_bb_b0};
+
+static struct attn_hw_reg pswrd2_prty0_bb_b0 = {
+	0, 1, 0x29d190, 0x29d19c, 0x29d198, 0x29d194};
+
+static struct attn_hw_reg pswrd2_prty1_bb_b0 = {
+	1, 31, 0x29d200, 0x29d20c, 0x29d208, 0x29d204};
+
+static struct attn_hw_reg pswrd2_prty2_bb_b0 = {
+	2, 3, 0x29d210, 0x29d21c, 0x29d218, 0x29d214};
+
+static struct attn_hw_reg *pswrd2_prty_bb_b0_regs[3] = {
+	&pswrd2_prty0_bb_b0, &pswrd2_prty1_bb_b0, &pswrd2_prty2_bb_b0};
+
+static struct attn_hw_reg pswwr_int0_bb_b0 = {
+	0, 16, 0x29a180, 0x29a18c, 0x29a188, 0x29a184};
+
+static struct attn_hw_reg *pswwr_int_bb_b0_regs[1] = {
+	&pswwr_int0_bb_b0};
+
+static struct attn_hw_reg pswwr_prty0_bb_b0 = {
+	0, 1, 0x29a190, 0x29a19c, 0x29a198, 0x29a194};
+
+static struct attn_hw_reg *pswwr_prty_bb_b0_regs[1] = {
+	&pswwr_prty0_bb_b0};
+
+static struct attn_hw_reg pswwr2_int0_bb_b0 = {
+	0, 19, 0x29b180, 0x29b18c, 0x29b188, 0x29b184};
+
+static struct attn_hw_reg *pswwr2_int_bb_b0_regs[1] = {
+	&pswwr2_int0_bb_b0};
+
+static struct attn_hw_reg pswwr2_prty0_bb_b0 = {
+	0, 1, 0x29b190, 0x29b19c, 0x29b198, 0x29b194};
+
+static struct attn_hw_reg pswwr2_prty1_bb_b0 = {
+	1, 31, 0x29b200, 0x29b20c, 0x29b208, 0x29b204};
+
+static struct attn_hw_reg pswwr2_prty2_bb_b0 = {
+	2, 31, 0x29b210, 0x29b21c, 0x29b218, 0x29b214};
+
+static struct attn_hw_reg pswwr2_prty3_bb_b0 = {
+	3, 31, 0x29b220, 0x29b22c, 0x29b228, 0x29b224};
+
+static struct attn_hw_reg pswwr2_prty4_bb_b0 = {
+	4, 20, 0x29b230, 0x29b23c, 0x29b238, 0x29b234};
+
+static struct attn_hw_reg *pswwr2_prty_bb_b0_regs[5] = {
+	&pswwr2_prty0_bb_b0, &pswwr2_prty1_bb_b0, &pswwr2_prty2_bb_b0,
+	&pswwr2_prty3_bb_b0, &pswwr2_prty4_bb_b0};
+
+static struct attn_hw_reg pswrq_int0_bb_b0 = {
+	0, 21, 0x280180, 0x28018c, 0x280188, 0x280184};
+
+static struct attn_hw_reg *pswrq_int_bb_b0_regs[1] = {
+	&pswrq_int0_bb_b0};
+
+static struct attn_hw_reg pswrq_prty0_bb_b0 = {
+	0, 1, 0x280190, 0x28019c, 0x280198, 0x280194};
+
+static struct attn_hw_reg *pswrq_prty_bb_b0_regs[1] = {
+	&pswrq_prty0_bb_b0};
+
+static struct attn_hw_reg pswrq2_int0_bb_b0 = {
+	0, 15, 0x240180, 0x24018c, 0x240188, 0x240184};
+
+static struct attn_hw_reg *pswrq2_int_bb_b0_regs[1] = {
+	&pswrq2_int0_bb_b0};
+
+static struct attn_hw_reg pswrq2_prty1_bb_b0 = {
+	0, 9, 0x240200, 0x24020c, 0x240208, 0x240204};
+
+static struct attn_hw_reg *pswrq2_prty_bb_b0_regs[1] = {
+	&pswrq2_prty1_bb_b0};
+
+static struct attn_hw_reg pglcs_int0_bb_b0 = {
+	0, 1, 0x1d00, 0x1d0c, 0x1d08, 0x1d04};
+
+static struct attn_hw_reg *pglcs_int_bb_b0_regs[1] = {
+	&pglcs_int0_bb_b0};
+
+static struct attn_hw_reg dmae_int0_bb_b0 = {
+	0, 2, 0xc180, 0xc18c, 0xc188, 0xc184};
+
+static struct attn_hw_reg *dmae_int_bb_b0_regs[1] = {
+	&dmae_int0_bb_b0};
+
+static struct attn_hw_reg dmae_prty1_bb_b0 = {
+	0, 3, 0xc200, 0xc20c, 0xc208, 0xc204};
+
+static struct attn_hw_reg *dmae_prty_bb_b0_regs[1] = {
+	&dmae_prty1_bb_b0};
+
+static struct attn_hw_reg ptu_int0_bb_b0 = {
+	0, 8, 0x560180, 0x56018c, 0x560188, 0x560184};
+
+static struct attn_hw_reg *ptu_int_bb_b0_regs[1] = {
+	&ptu_int0_bb_b0};
+
+static struct attn_hw_reg ptu_prty1_bb_b0 = {
+	0, 18, 0x560200, 0x56020c, 0x560208, 0x560204};
+
+static struct attn_hw_reg *ptu_prty_bb_b0_regs[1] = {
+	&ptu_prty1_bb_b0};
+
+static struct attn_hw_reg tcm_int0_bb_b0 = {
+	0, 8, 0x1180180, 0x118018c, 0x1180188, 0x1180184};
+
+static struct attn_hw_reg tcm_int1_bb_b0 = {
+	1, 32, 0x1180190, 0x118019c, 0x1180198, 0x1180194};
+
+static struct attn_hw_reg tcm_int2_bb_b0 = {
+	2, 1, 0x11801a0, 0x11801ac, 0x11801a8, 0x11801a4};
+
+static struct attn_hw_reg *tcm_int_bb_b0_regs[3] = {
+	&tcm_int0_bb_b0, &tcm_int1_bb_b0, &tcm_int2_bb_b0};
+
+static struct attn_hw_reg tcm_prty1_bb_b0 = {
+	0, 31, 0x1180200, 0x118020c, 0x1180208, 0x1180204};
+
+static struct attn_hw_reg tcm_prty2_bb_b0 = {
+	1, 2, 0x1180210, 0x118021c, 0x1180218, 0x1180214};
+
+static struct attn_hw_reg *tcm_prty_bb_b0_regs[2] = {
+	&tcm_prty1_bb_b0, &tcm_prty2_bb_b0};
+
+static struct attn_hw_reg mcm_int0_bb_b0 = {
+	0, 14, 0x1200180, 0x120018c, 0x1200188, 0x1200184};
+
+static struct attn_hw_reg mcm_int1_bb_b0 = {
+	1, 26, 0x1200190, 0x120019c, 0x1200198, 0x1200194};
+
+static struct attn_hw_reg mcm_int2_bb_b0 = {
+	2, 1, 0x12001a0, 0x12001ac, 0x12001a8, 0x12001a4};
+
+static struct attn_hw_reg *mcm_int_bb_b0_regs[3] = {
+	&mcm_int0_bb_b0, &mcm_int1_bb_b0, &mcm_int2_bb_b0};
+
+static struct attn_hw_reg mcm_prty1_bb_b0 = {
+	0, 31, 0x1200200, 0x120020c, 0x1200208, 0x1200204};
+
+static struct attn_hw_reg mcm_prty2_bb_b0 = {
+	1, 4, 0x1200210, 0x120021c, 0x1200218, 0x1200214};
+
+static struct attn_hw_reg *mcm_prty_bb_b0_regs[2] = {
+	&mcm_prty1_bb_b0, &mcm_prty2_bb_b0};
+
+static struct attn_hw_reg ucm_int0_bb_b0 = {
+	0, 17, 0x1280180, 0x128018c, 0x1280188, 0x1280184};
+
+static struct attn_hw_reg ucm_int1_bb_b0 = {
+	1, 29, 0x1280190, 0x128019c, 0x1280198, 0x1280194};
+
+static struct attn_hw_reg ucm_int2_bb_b0 = {
+	2, 1, 0x12801a0, 0x12801ac, 0x12801a8, 0x12801a4};
+
+static struct attn_hw_reg *ucm_int_bb_b0_regs[3] = {
+	&ucm_int0_bb_b0, &ucm_int1_bb_b0, &ucm_int2_bb_b0};
+
+static struct attn_hw_reg ucm_prty1_bb_b0 = {
+	0, 31, 0x1280200, 0x128020c, 0x1280208, 0x1280204};
+
+static struct attn_hw_reg ucm_prty2_bb_b0 = {
+	1, 7, 0x1280210, 0x128021c, 0x1280218, 0x1280214};
+
+static struct attn_hw_reg *ucm_prty_bb_b0_regs[2] = {
+	&ucm_prty1_bb_b0, &ucm_prty2_bb_b0};
+
+static struct attn_hw_reg xcm_int0_bb_b0 = {
+	0, 16, 0x1000180, 0x100018c, 0x1000188, 0x1000184};
+
+static struct attn_hw_reg xcm_int1_bb_b0 = {
+	1, 25, 0x1000190, 0x100019c, 0x1000198, 0x1000194};
+
+static struct attn_hw_reg xcm_int2_bb_b0 = {
+	2, 8, 0x10001a0, 0x10001ac, 0x10001a8, 0x10001a4};
+
+static struct attn_hw_reg *xcm_int_bb_b0_regs[3] = {
+	&xcm_int0_bb_b0, &xcm_int1_bb_b0, &xcm_int2_bb_b0};
+
+static struct attn_hw_reg xcm_prty1_bb_b0 = {
+	0, 31, 0x1000200, 0x100020c, 0x1000208, 0x1000204};
+
+static struct attn_hw_reg xcm_prty2_bb_b0 = {
+	1, 11, 0x1000210, 0x100021c, 0x1000218, 0x1000214};
+
+static struct attn_hw_reg *xcm_prty_bb_b0_regs[2] = {
+	&xcm_prty1_bb_b0, &xcm_prty2_bb_b0};
+
+static struct attn_hw_reg ycm_int0_bb_b0 = {
+	0, 13, 0x1080180, 0x108018c, 0x1080188, 0x1080184};
+
+static struct attn_hw_reg ycm_int1_bb_b0 = {
+	1, 23, 0x1080190, 0x108019c, 0x1080198, 0x1080194};
+
+static struct attn_hw_reg ycm_int2_bb_b0 = {
+	2, 1, 0x10801a0, 0x10801ac, 0x10801a8, 0x10801a4};
+
+static struct attn_hw_reg *ycm_int_bb_b0_regs[3] = {
+	&ycm_int0_bb_b0, &ycm_int1_bb_b0, &ycm_int2_bb_b0};
+
+static struct attn_hw_reg ycm_prty1_bb_b0 = {
+	0, 31, 0x1080200, 0x108020c, 0x1080208, 0x1080204};
+
+static struct attn_hw_reg ycm_prty2_bb_b0 = {
+	1, 3, 0x1080210, 0x108021c, 0x1080218, 0x1080214};
+
+static struct attn_hw_reg *ycm_prty_bb_b0_regs[2] = {
+	&ycm_prty1_bb_b0, &ycm_prty2_bb_b0};
+
+static struct attn_hw_reg pcm_int0_bb_b0 = {
+	0, 5, 0x1100180, 0x110018c, 0x1100188, 0x1100184};
+
+static struct attn_hw_reg pcm_int1_bb_b0 = {
+	1, 14, 0x1100190, 0x110019c, 0x1100198, 0x1100194};
+
+static struct attn_hw_reg pcm_int2_bb_b0 = {
+	2, 1, 0x11001a0, 0x11001ac, 0x11001a8, 0x11001a4};
+
+static struct attn_hw_reg *pcm_int_bb_b0_regs[3] = {
+	&pcm_int0_bb_b0, &pcm_int1_bb_b0, &pcm_int2_bb_b0};
+
+static struct attn_hw_reg pcm_prty1_bb_b0 = {
+	0, 11, 0x1100200, 0x110020c, 0x1100208, 0x1100204};
+
+static struct attn_hw_reg *pcm_prty_bb_b0_regs[1] = {
+	&pcm_prty1_bb_b0};
+
+static struct attn_hw_reg qm_int0_bb_b0 = {
+	0, 22, 0x2f0180, 0x2f018c, 0x2f0188, 0x2f0184};
+
+static struct attn_hw_reg *qm_int_bb_b0_regs[1] = {
+	&qm_int0_bb_b0};
+
+static struct attn_hw_reg qm_prty0_bb_b0 = {
+	0, 11, 0x2f0190, 0x2f019c, 0x2f0198, 0x2f0194};
+
+static struct attn_hw_reg qm_prty1_bb_b0 = {
+	1, 31, 0x2f0200, 0x2f020c, 0x2f0208, 0x2f0204};
+
+static struct attn_hw_reg qm_prty2_bb_b0 = {
+	2, 31, 0x2f0210, 0x2f021c, 0x2f0218, 0x2f0214};
+
+static struct attn_hw_reg qm_prty3_bb_b0 = {
+	3, 11, 0x2f0220, 0x2f022c, 0x2f0228, 0x2f0224};
+
+static struct attn_hw_reg *qm_prty_bb_b0_regs[4] = {
+	&qm_prty0_bb_b0, &qm_prty1_bb_b0, &qm_prty2_bb_b0, &qm_prty3_bb_b0};
+
+static struct attn_hw_reg tm_int0_bb_b0 = {
+	0, 32, 0x2c0180, 0x2c018c, 0x2c0188, 0x2c0184};
+
+static struct attn_hw_reg tm_int1_bb_b0 = {
+	1, 11, 0x2c0190, 0x2c019c, 0x2c0198, 0x2c0194};
+
+static struct attn_hw_reg *tm_int_bb_b0_regs[2] = {
+	&tm_int0_bb_b0, &tm_int1_bb_b0};
+
+static struct attn_hw_reg tm_prty1_bb_b0 = {
+	0, 17, 0x2c0200, 0x2c020c, 0x2c0208, 0x2c0204};
+
+static struct attn_hw_reg *tm_prty_bb_b0_regs[1] = {
+	&tm_prty1_bb_b0};
+
+static struct attn_hw_reg dorq_int0_bb_b0 = {
+	0, 9, 0x100180, 0x10018c, 0x100188, 0x100184};
+
+static struct attn_hw_reg *dorq_int_bb_b0_regs[1] = {
+	&dorq_int0_bb_b0};
+
+static struct attn_hw_reg dorq_prty0_bb_b0 = {
+	0, 1, 0x100190, 0x10019c, 0x100198, 0x100194};
+
+static struct attn_hw_reg dorq_prty1_bb_b0 = {
+	1, 6, 0x100200, 0x10020c, 0x100208, 0x100204};
+
+static struct attn_hw_reg *dorq_prty_bb_b0_regs[2] = {
+	&dorq_prty0_bb_b0, &dorq_prty1_bb_b0};
+
+static struct attn_hw_reg brb_int0_bb_b0 = {
+	0, 32, 0x3400c0, 0x3400cc, 0x3400c8, 0x3400c4};
+
+static struct attn_hw_reg brb_int1_bb_b0 = {
+	1, 30, 0x3400d8, 0x3400e4, 0x3400e0, 0x3400dc};
+
+static struct attn_hw_reg brb_int2_bb_b0 = {
+	2, 28, 0x3400f0, 0x3400fc, 0x3400f8, 0x3400f4};
+
+static struct attn_hw_reg brb_int3_bb_b0 = {
+	3, 31, 0x340108, 0x340114, 0x340110, 0x34010c};
+
+static struct attn_hw_reg brb_int4_bb_b0 = {
+	4, 27, 0x340120, 0x34012c, 0x340128, 0x340124};
+
+static struct attn_hw_reg brb_int5_bb_b0 = {
+	5, 1, 0x340138, 0x340144, 0x340140, 0x34013c};
+
+static struct attn_hw_reg brb_int6_bb_b0 = {
+	6, 8, 0x340150, 0x34015c, 0x340158, 0x340154};
+
+static struct attn_hw_reg brb_int7_bb_b0 = {
+	7, 32, 0x340168, 0x340174, 0x340170, 0x34016c};
+
+static struct attn_hw_reg brb_int8_bb_b0 = {
+	8, 17, 0x340184, 0x340190, 0x34018c, 0x340188};
+
+static struct attn_hw_reg brb_int9_bb_b0 = {
+	9, 1, 0x34019c, 0x3401a8, 0x3401a4, 0x3401a0};
+
+static struct attn_hw_reg brb_int10_bb_b0 = {
+	10, 14, 0x3401b4, 0x3401c0, 0x3401bc, 0x3401b8};
+
+static struct attn_hw_reg brb_int11_bb_b0 = {
+	11, 8, 0x3401cc, 0x3401d8, 0x3401d4, 0x3401d0};
+
+static struct attn_hw_reg *brb_int_bb_b0_regs[12] = {
+	&brb_int0_bb_b0, &brb_int1_bb_b0, &brb_int2_bb_b0, &brb_int3_bb_b0,
+	&brb_int4_bb_b0, &brb_int5_bb_b0, &brb_int6_bb_b0, &brb_int7_bb_b0,
+	&brb_int8_bb_b0, &brb_int9_bb_b0, &brb_int10_bb_b0, &brb_int11_bb_b0};
+
+static struct attn_hw_reg brb_prty0_bb_b0 = {
+	0, 5, 0x3401dc, 0x3401e8, 0x3401e4, 0x3401e0};
+
+static struct attn_hw_reg brb_prty1_bb_b0 = {
+	1, 31, 0x340400, 0x34040c, 0x340408, 0x340404};
+
+static struct attn_hw_reg brb_prty2_bb_b0 = {
+	2, 14, 0x340410, 0x34041c, 0x340418, 0x340414};
+
+static struct attn_hw_reg *brb_prty_bb_b0_regs[3] = {
+	&brb_prty0_bb_b0, &brb_prty1_bb_b0, &brb_prty2_bb_b0};
+
+static struct attn_hw_reg src_int0_bb_b0 = {
+	0, 1, 0x2381d8, 0x2381dc, 0x2381e0, 0x2381e4};
+
+static struct attn_hw_reg *src_int_bb_b0_regs[1] = {
+	&src_int0_bb_b0};
+
+static struct attn_hw_reg prs_int0_bb_b0 = {
+	0, 2, 0x1f0040, 0x1f004c, 0x1f0048, 0x1f0044};
+
+static struct attn_hw_reg *prs_int_bb_b0_regs[1] = {
+	&prs_int0_bb_b0};
+
+static struct attn_hw_reg prs_prty0_bb_b0 = {
+	0, 2, 0x1f0050, 0x1f005c, 0x1f0058, 0x1f0054};
+
+static struct attn_hw_reg prs_prty1_bb_b0 = {
+	1, 31, 0x1f0204, 0x1f0210, 0x1f020c, 0x1f0208};
+
+static struct attn_hw_reg prs_prty2_bb_b0 = {
+	2, 5, 0x1f0214, 0x1f0220, 0x1f021c, 0x1f0218};
+
+static struct attn_hw_reg *prs_prty_bb_b0_regs[3] = {
+	&prs_prty0_bb_b0, &prs_prty1_bb_b0, &prs_prty2_bb_b0};
+
+static struct attn_hw_reg tsdm_int0_bb_b0 = {
+	0, 26, 0xfb0040, 0xfb004c, 0xfb0048, 0xfb0044};
+
+static struct attn_hw_reg *tsdm_int_bb_b0_regs[1] = {
+	&tsdm_int0_bb_b0};
+
+static struct attn_hw_reg tsdm_prty1_bb_b0 = {
+	0, 10, 0xfb0200, 0xfb020c, 0xfb0208, 0xfb0204};
+
+static struct attn_hw_reg *tsdm_prty_bb_b0_regs[1] = {
+	&tsdm_prty1_bb_b0};
+
+static struct attn_hw_reg msdm_int0_bb_b0 = {
+	0, 26, 0xfc0040, 0xfc004c, 0xfc0048, 0xfc0044};
+
+static struct attn_hw_reg *msdm_int_bb_b0_regs[1] = {
+	&msdm_int0_bb_b0};
+
+static struct attn_hw_reg msdm_prty1_bb_b0 = {
+	0, 11, 0xfc0200, 0xfc020c, 0xfc0208, 0xfc0204};
+
+static struct attn_hw_reg *msdm_prty_bb_b0_regs[1] = {
+	&msdm_prty1_bb_b0};
+
+static struct attn_hw_reg usdm_int0_bb_b0 = {
+	0, 26, 0xfd0040, 0xfd004c, 0xfd0048, 0xfd0044};
+
+static struct attn_hw_reg *usdm_int_bb_b0_regs[1] = {
+	&usdm_int0_bb_b0};
+
+static struct attn_hw_reg usdm_prty1_bb_b0 = {
+	0, 10, 0xfd0200, 0xfd020c, 0xfd0208, 0xfd0204};
+
+static struct attn_hw_reg *usdm_prty_bb_b0_regs[1] = {
+	&usdm_prty1_bb_b0};
+
+static struct attn_hw_reg xsdm_int0_bb_b0 = {
+	0, 26, 0xf80040, 0xf8004c, 0xf80048, 0xf80044};
+
+static struct attn_hw_reg *xsdm_int_bb_b0_regs[1] = {
+	&xsdm_int0_bb_b0};
+
+static struct attn_hw_reg xsdm_prty1_bb_b0 = {
+	0, 10, 0xf80200, 0xf8020c, 0xf80208, 0xf80204};
+
+static struct attn_hw_reg *xsdm_prty_bb_b0_regs[1] = {
+	&xsdm_prty1_bb_b0};
+
+static struct attn_hw_reg ysdm_int0_bb_b0 = {
+	0, 26, 0xf90040, 0xf9004c, 0xf90048, 0xf90044};
+
+static struct attn_hw_reg *ysdm_int_bb_b0_regs[1] = {
+	&ysdm_int0_bb_b0};
+
+static struct attn_hw_reg ysdm_prty1_bb_b0 = {
+	0, 9, 0xf90200, 0xf9020c, 0xf90208, 0xf90204};
+
+static struct attn_hw_reg *ysdm_prty_bb_b0_regs[1] = {
+	&ysdm_prty1_bb_b0};
+
+static struct attn_hw_reg psdm_int0_bb_b0 = {
+	0, 26, 0xfa0040, 0xfa004c, 0xfa0048, 0xfa0044};
+
+static struct attn_hw_reg *psdm_int_bb_b0_regs[1] = {
+	&psdm_int0_bb_b0};
+
+static struct attn_hw_reg psdm_prty1_bb_b0 = {
+	0, 9, 0xfa0200, 0xfa020c, 0xfa0208, 0xfa0204};
+
+static struct attn_hw_reg *psdm_prty_bb_b0_regs[1] = {
+	&psdm_prty1_bb_b0};
+
+static struct attn_hw_reg tsem_int0_bb_b0 = {
+	0, 32, 0x1700040, 0x170004c, 0x1700048, 0x1700044};
+
+static struct attn_hw_reg tsem_int1_bb_b0 = {
+	1, 13, 0x1700050, 0x170005c, 0x1700058, 0x1700054};
+
+static struct attn_hw_reg tsem_fast_memory_int0_bb_b0 = {
+	2, 1, 0x1740040, 0x174004c, 0x1740048, 0x1740044};
+
+static struct attn_hw_reg *tsem_int_bb_b0_regs[3] = {
+	&tsem_int0_bb_b0, &tsem_int1_bb_b0, &tsem_fast_memory_int0_bb_b0};
+
+static struct attn_hw_reg tsem_prty0_bb_b0 = {
+	0, 3, 0x17000c8, 0x17000d4, 0x17000d0, 0x17000cc};
+
+static struct attn_hw_reg tsem_prty1_bb_b0 = {
+	1, 6, 0x1700200, 0x170020c, 0x1700208, 0x1700204};
+
+static struct attn_hw_reg tsem_fast_memory_vfc_config_prty1_bb_b0 = {
+	2, 6, 0x174a200, 0x174a20c, 0x174a208, 0x174a204};
+
+static struct attn_hw_reg *tsem_prty_bb_b0_regs[3] = {
+	&tsem_prty0_bb_b0, &tsem_prty1_bb_b0,
+	&tsem_fast_memory_vfc_config_prty1_bb_b0};
+
+static struct attn_hw_reg msem_int0_bb_b0 = {
+	0, 32, 0x1800040, 0x180004c, 0x1800048, 0x1800044};
+
+static struct attn_hw_reg msem_int1_bb_b0 = {
+	1, 13, 0x1800050, 0x180005c, 0x1800058, 0x1800054};
+
+static struct attn_hw_reg msem_fast_memory_int0_bb_b0 = {
+	2, 1, 0x1840040, 0x184004c, 0x1840048, 0x1840044};
+
+static struct attn_hw_reg *msem_int_bb_b0_regs[3] = {
+	&msem_int0_bb_b0, &msem_int1_bb_b0, &msem_fast_memory_int0_bb_b0};
+
+static struct attn_hw_reg msem_prty0_bb_b0 = {
+	0, 3, 0x18000c8, 0x18000d4, 0x18000d0, 0x18000cc};
+
+static struct attn_hw_reg msem_prty1_bb_b0 = {
+	1, 6, 0x1800200, 0x180020c, 0x1800208, 0x1800204};
+
+static struct attn_hw_reg *msem_prty_bb_b0_regs[2] = {
+	&msem_prty0_bb_b0, &msem_prty1_bb_b0};
+
+static struct attn_hw_reg usem_int0_bb_b0 = {
+	0, 32, 0x1900040, 0x190004c, 0x1900048, 0x1900044};
+
+static struct attn_hw_reg usem_int1_bb_b0 = {
+	1, 13, 0x1900050, 0x190005c, 0x1900058, 0x1900054};
+
+static struct attn_hw_reg usem_fast_memory_int0_bb_b0 = {
+	2, 1, 0x1940040, 0x194004c, 0x1940048, 0x1940044};
+
+static struct attn_hw_reg *usem_int_bb_b0_regs[3] = {
+	&usem_int0_bb_b0, &usem_int1_bb_b0, &usem_fast_memory_int0_bb_b0};
+
+static struct attn_hw_reg usem_prty0_bb_b0 = {
+	0, 3, 0x19000c8, 0x19000d4, 0x19000d0, 0x19000cc};
+
+static struct attn_hw_reg usem_prty1_bb_b0 = {
+	1, 6, 0x1900200, 0x190020c, 0x1900208, 0x1900204};
+
+static struct attn_hw_reg *usem_prty_bb_b0_regs[2] = {
+	&usem_prty0_bb_b0, &usem_prty1_bb_b0};
+
+static struct attn_hw_reg xsem_int0_bb_b0 = {
+	0, 32, 0x1400040, 0x140004c, 0x1400048, 0x1400044};
+
+static struct attn_hw_reg xsem_int1_bb_b0 = {
+	1, 13, 0x1400050, 0x140005c, 0x1400058, 0x1400054};
+
+static struct attn_hw_reg xsem_fast_memory_int0_bb_b0 = {
+	2, 1, 0x1440040, 0x144004c, 0x1440048, 0x1440044};
+
+static struct attn_hw_reg *xsem_int_bb_b0_regs[3] = {
+	&xsem_int0_bb_b0, &xsem_int1_bb_b0, &xsem_fast_memory_int0_bb_b0};
+
+static struct attn_hw_reg xsem_prty0_bb_b0 = {
+	0, 3, 0x14000c8, 0x14000d4, 0x14000d0, 0x14000cc};
+
+static struct attn_hw_reg xsem_prty1_bb_b0 = {
+	1, 7, 0x1400200, 0x140020c, 0x1400208, 0x1400204};
+
+static struct attn_hw_reg *xsem_prty_bb_b0_regs[2] = {
+	&xsem_prty0_bb_b0, &xsem_prty1_bb_b0};
+
+static struct attn_hw_reg ysem_int0_bb_b0 = {
+	0, 32, 0x1500040, 0x150004c, 0x1500048, 0x1500044};
+
+static struct attn_hw_reg ysem_int1_bb_b0 = {
+	1, 13, 0x1500050, 0x150005c, 0x1500058, 0x1500054};
+
+static struct attn_hw_reg ysem_fast_memory_int0_bb_b0 = {
+	2, 1, 0x1540040, 0x154004c, 0x1540048, 0x1540044};
+
+static struct attn_hw_reg *ysem_int_bb_b0_regs[3] = {
+	&ysem_int0_bb_b0, &ysem_int1_bb_b0, &ysem_fast_memory_int0_bb_b0};
+
+static struct attn_hw_reg ysem_prty0_bb_b0 = {
+	0, 3, 0x15000c8, 0x15000d4, 0x15000d0, 0x15000cc};
+
+static struct attn_hw_reg ysem_prty1_bb_b0 = {
+	1, 7, 0x1500200, 0x150020c, 0x1500208, 0x1500204};
+
+static struct attn_hw_reg *ysem_prty_bb_b0_regs[2] = {
+	&ysem_prty0_bb_b0, &ysem_prty1_bb_b0};
+
+static struct attn_hw_reg psem_int0_bb_b0 = {
+	0, 32, 0x1600040, 0x160004c, 0x1600048, 0x1600044};
+
+static struct attn_hw_reg psem_int1_bb_b0 = {
+	1, 13, 0x1600050, 0x160005c, 0x1600058, 0x1600054};
+
+static struct attn_hw_reg psem_fast_memory_int0_bb_b0 = {
+	2, 1, 0x1640040, 0x164004c, 0x1640048, 0x1640044};
+
+static struct attn_hw_reg *psem_int_bb_b0_regs[3] = {
+	&psem_int0_bb_b0, &psem_int1_bb_b0, &psem_fast_memory_int0_bb_b0};
+
+static struct attn_hw_reg psem_prty0_bb_b0 = {
+	0, 3, 0x16000c8, 0x16000d4, 0x16000d0, 0x16000cc};
+
+static struct attn_hw_reg psem_prty1_bb_b0 = {
+	1, 6, 0x1600200, 0x160020c, 0x1600208, 0x1600204};
+
+static struct attn_hw_reg psem_fast_memory_vfc_config_prty1_bb_b0 = {
+	2, 6, 0x164a200, 0x164a20c, 0x164a208, 0x164a204};
+
+static struct attn_hw_reg *psem_prty_bb_b0_regs[3] = {
+	&psem_prty0_bb_b0, &psem_prty1_bb_b0,
+	&psem_fast_memory_vfc_config_prty1_bb_b0};
+
+static struct attn_hw_reg rss_int0_bb_b0 = {
+	0, 12, 0x238980, 0x23898c, 0x238988, 0x238984};
+
+static struct attn_hw_reg *rss_int_bb_b0_regs[1] = {
+	&rss_int0_bb_b0};
+
+static struct attn_hw_reg rss_prty1_bb_b0 = {
+	0, 4, 0x238a00, 0x238a0c, 0x238a08, 0x238a04};
+
+static struct attn_hw_reg *rss_prty_bb_b0_regs[1] = {
+	&rss_prty1_bb_b0};
+
+static struct attn_hw_reg tmld_int0_bb_b0 = {
+	0, 6, 0x4d0180, 0x4d018c, 0x4d0188, 0x4d0184};
+
+static struct attn_hw_reg *tmld_int_bb_b0_regs[1] = {
+	&tmld_int0_bb_b0};
+
+static struct attn_hw_reg tmld_prty1_bb_b0 = {
+	0, 8, 0x4d0200, 0x4d020c, 0x4d0208, 0x4d0204};
+
+static struct attn_hw_reg *tmld_prty_bb_b0_regs[1] = {
+	&tmld_prty1_bb_b0};
+
+static struct attn_hw_reg muld_int0_bb_b0 = {
+	0, 6, 0x4e0180, 0x4e018c, 0x4e0188, 0x4e0184};
+
+static struct attn_hw_reg *muld_int_bb_b0_regs[1] = {
+	&muld_int0_bb_b0};
+
+static struct attn_hw_reg muld_prty1_bb_b0 = {
+	0, 10, 0x4e0200, 0x4e020c, 0x4e0208, 0x4e0204};
+
+static struct attn_hw_reg *muld_prty_bb_b0_regs[1] = {
+	&muld_prty1_bb_b0};
+
+static struct attn_hw_reg yuld_int0_bb_b0 = {
+	0, 6, 0x4c8180, 0x4c818c, 0x4c8188, 0x4c8184};
+
+static struct attn_hw_reg *yuld_int_bb_b0_regs[1] = {
+	&yuld_int0_bb_b0};
+
+static struct attn_hw_reg yuld_prty1_bb_b0 = {
+	0, 6, 0x4c8200, 0x4c820c, 0x4c8208, 0x4c8204};
+
+static struct attn_hw_reg *yuld_prty_bb_b0_regs[1] = {
+	&yuld_prty1_bb_b0};
+
+static struct attn_hw_reg xyld_int0_bb_b0 = {
+	0, 6, 0x4c0180, 0x4c018c, 0x4c0188, 0x4c0184};
+
+static struct attn_hw_reg *xyld_int_bb_b0_regs[1] = {
+	&xyld_int0_bb_b0};
+
+static struct attn_hw_reg xyld_prty1_bb_b0 = {
+	0, 9, 0x4c0200, 0x4c020c, 0x4c0208, 0x4c0204};
+
+static struct attn_hw_reg *xyld_prty_bb_b0_regs[1] = {
+	&xyld_prty1_bb_b0};
+
+static struct attn_hw_reg prm_int0_bb_b0 = {
+	0, 11, 0x230040, 0x23004c, 0x230048, 0x230044};
+
+static struct attn_hw_reg *prm_int_bb_b0_regs[1] = {
+	&prm_int0_bb_b0};
+
+static struct attn_hw_reg prm_prty0_bb_b0 = {
+	0, 1, 0x230050, 0x23005c, 0x230058, 0x230054};
+
+static struct attn_hw_reg prm_prty1_bb_b0 = {
+	1, 24, 0x230200, 0x23020c, 0x230208, 0x230204};
+
+static struct attn_hw_reg *prm_prty_bb_b0_regs[2] = {
+	&prm_prty0_bb_b0, &prm_prty1_bb_b0};
+
+static struct attn_hw_reg pbf_pb1_int0_bb_b0 = {
+	0, 9, 0xda0040, 0xda004c, 0xda0048, 0xda0044};
+
+static struct attn_hw_reg *pbf_pb1_int_bb_b0_regs[1] = {
+	&pbf_pb1_int0_bb_b0};
+
+static struct attn_hw_reg pbf_pb1_prty0_bb_b0 = {
+	0, 1, 0xda0050, 0xda005c, 0xda0058, 0xda0054};
+
+static struct attn_hw_reg *pbf_pb1_prty_bb_b0_regs[1] = {
+	&pbf_pb1_prty0_bb_b0};
+
+static struct attn_hw_reg pbf_pb2_int0_bb_b0 = {
+	0, 9, 0xda4040, 0xda404c, 0xda4048, 0xda4044};
+
+static struct attn_hw_reg *pbf_pb2_int_bb_b0_regs[1] = {
+	&pbf_pb2_int0_bb_b0};
+
+static struct attn_hw_reg pbf_pb2_prty0_bb_b0 = {
+	0, 1, 0xda4050, 0xda405c, 0xda4058, 0xda4054};
+
+static struct attn_hw_reg *pbf_pb2_prty_bb_b0_regs[1] = {
+	&pbf_pb2_prty0_bb_b0};
+
+static struct attn_hw_reg rpb_int0_bb_b0 = {
+	0, 9, 0x23c040, 0x23c04c, 0x23c048, 0x23c044};
+
+static struct attn_hw_reg *rpb_int_bb_b0_regs[1] = {
+	&rpb_int0_bb_b0};
+
+static struct attn_hw_reg rpb_prty0_bb_b0 = {
+	0, 1, 0x23c050, 0x23c05c, 0x23c058, 0x23c054};
+
+static struct attn_hw_reg *rpb_prty_bb_b0_regs[1] = {
+	&rpb_prty0_bb_b0};
+
+static struct attn_hw_reg btb_int0_bb_b0 = {
+	0, 16, 0xdb00c0, 0xdb00cc, 0xdb00c8, 0xdb00c4};
+
+static struct attn_hw_reg btb_int1_bb_b0 = {
+	1, 16, 0xdb00d8, 0xdb00e4, 0xdb00e0, 0xdb00dc};
+
+static struct attn_hw_reg btb_int2_bb_b0 = {
+	2, 4, 0xdb00f0, 0xdb00fc, 0xdb00f8, 0xdb00f4};
+
+static struct attn_hw_reg btb_int3_bb_b0 = {
+	3, 32, 0xdb0108, 0xdb0114, 0xdb0110, 0xdb010c};
+
+static struct attn_hw_reg btb_int4_bb_b0 = {
+	4, 23, 0xdb0120, 0xdb012c, 0xdb0128, 0xdb0124};
+
+static struct attn_hw_reg btb_int5_bb_b0 = {
+	5, 32, 0xdb0138, 0xdb0144, 0xdb0140, 0xdb013c};
+
+static struct attn_hw_reg btb_int6_bb_b0 = {
+	6, 1, 0xdb0150, 0xdb015c, 0xdb0158, 0xdb0154};
+
+static struct attn_hw_reg btb_int8_bb_b0 = {
+	7, 1, 0xdb0184, 0xdb0190, 0xdb018c, 0xdb0188};
+
+static struct attn_hw_reg btb_int9_bb_b0 = {
+	8, 1, 0xdb019c, 0xdb01a8, 0xdb01a4, 0xdb01a0};
+
+static struct attn_hw_reg btb_int10_bb_b0 = {
+	9, 1, 0xdb01b4, 0xdb01c0, 0xdb01bc, 0xdb01b8};
+
+static struct attn_hw_reg btb_int11_bb_b0 = {
+	10, 2, 0xdb01cc, 0xdb01d8, 0xdb01d4, 0xdb01d0};
+
+static struct attn_hw_reg *btb_int_bb_b0_regs[11] = {
+	&btb_int0_bb_b0, &btb_int1_bb_b0, &btb_int2_bb_b0, &btb_int3_bb_b0,
+	&btb_int4_bb_b0, &btb_int5_bb_b0, &btb_int6_bb_b0, &btb_int8_bb_b0,
+	&btb_int9_bb_b0, &btb_int10_bb_b0, &btb_int11_bb_b0};
+
+static struct attn_hw_reg btb_prty0_bb_b0 = {
+	0, 5, 0xdb01dc, 0xdb01e8, 0xdb01e4, 0xdb01e0};
+
+static struct attn_hw_reg btb_prty1_bb_b0 = {
+	1, 23, 0xdb0400, 0xdb040c, 0xdb0408, 0xdb0404};
+
+static struct attn_hw_reg *btb_prty_bb_b0_regs[2] = {
+	&btb_prty0_bb_b0, &btb_prty1_bb_b0};
+
+static struct attn_hw_reg pbf_int0_bb_b0 = {
+	0, 1, 0xd80180, 0xd8018c, 0xd80188, 0xd80184};
+
+static struct attn_hw_reg *pbf_int_bb_b0_regs[1] = {
+	&pbf_int0_bb_b0};
+
+static struct attn_hw_reg pbf_prty0_bb_b0 = {
+	0, 1, 0xd80190, 0xd8019c, 0xd80198, 0xd80194};
+
+static struct attn_hw_reg pbf_prty1_bb_b0 = {
+	1, 31, 0xd80200, 0xd8020c, 0xd80208, 0xd80204};
+
+static struct attn_hw_reg pbf_prty2_bb_b0 = {
+	2, 27, 0xd80210, 0xd8021c, 0xd80218, 0xd80214};
+
+static struct attn_hw_reg *pbf_prty_bb_b0_regs[3] = {
+	&pbf_prty0_bb_b0, &pbf_prty1_bb_b0, &pbf_prty2_bb_b0};
+
+static struct attn_hw_reg rdif_int0_bb_b0 = {
+	0, 8, 0x300180, 0x30018c, 0x300188, 0x300184};
+
+static struct attn_hw_reg *rdif_int_bb_b0_regs[1] = {
+	&rdif_int0_bb_b0};
+
+static struct attn_hw_reg rdif_prty0_bb_b0 = {
+	0, 1, 0x300190, 0x30019c, 0x300198, 0x300194};
+
+static struct attn_hw_reg *rdif_prty_bb_b0_regs[1] = {
+	&rdif_prty0_bb_b0};
+
+static struct attn_hw_reg tdif_int0_bb_b0 = {
+	0, 8, 0x310180, 0x31018c, 0x310188, 0x310184};
+
+static struct attn_hw_reg *tdif_int_bb_b0_regs[1] = {
+	&tdif_int0_bb_b0};
+
+static struct attn_hw_reg tdif_prty0_bb_b0 = {
+	0, 1, 0x310190, 0x31019c, 0x310198, 0x310194};
+
+static struct attn_hw_reg tdif_prty1_bb_b0 = {
+	1, 11, 0x310200, 0x31020c, 0x310208, 0x310204};
+
+static struct attn_hw_reg *tdif_prty_bb_b0_regs[2] = {
+	&tdif_prty0_bb_b0, &tdif_prty1_bb_b0};
+
+static struct attn_hw_reg cdu_int0_bb_b0 = {
+	0, 8, 0x5801c0, 0x5801c4, 0x5801c8, 0x5801cc};
+
+static struct attn_hw_reg *cdu_int_bb_b0_regs[1] = {
+	&cdu_int0_bb_b0};
+
+static struct attn_hw_reg cdu_prty1_bb_b0 = {
+	0, 5, 0x580200, 0x58020c, 0x580208, 0x580204};
+
+static struct attn_hw_reg *cdu_prty_bb_b0_regs[1] = {
+	&cdu_prty1_bb_b0};
+
+static struct attn_hw_reg ccfc_int0_bb_b0 = {
+	0, 2, 0x2e0180, 0x2e018c, 0x2e0188, 0x2e0184};
+
+static struct attn_hw_reg *ccfc_int_bb_b0_regs[1] = {
+	&ccfc_int0_bb_b0};
+
+static struct attn_hw_reg ccfc_prty1_bb_b0 = {
+	0, 2, 0x2e0200, 0x2e020c, 0x2e0208, 0x2e0204};
+
+static struct attn_hw_reg ccfc_prty0_bb_b0 = {
+	1, 6, 0x2e05e4, 0x2e05f0, 0x2e05ec, 0x2e05e8};
+
+static struct attn_hw_reg *ccfc_prty_bb_b0_regs[2] = {
+	&ccfc_prty1_bb_b0, &ccfc_prty0_bb_b0};
+
+static struct attn_hw_reg tcfc_int0_bb_b0 = {
+	0, 2, 0x2d0180, 0x2d018c, 0x2d0188, 0x2d0184};
+
+static struct attn_hw_reg *tcfc_int_bb_b0_regs[1] = {
+	&tcfc_int0_bb_b0};
+
+static struct attn_hw_reg tcfc_prty1_bb_b0 = {
+	0, 2, 0x2d0200, 0x2d020c, 0x2d0208, 0x2d0204};
+
+static struct attn_hw_reg tcfc_prty0_bb_b0 = {
+	1, 6, 0x2d05e4, 0x2d05f0, 0x2d05ec, 0x2d05e8};
+
+static struct attn_hw_reg *tcfc_prty_bb_b0_regs[2] = {
+	&tcfc_prty1_bb_b0, &tcfc_prty0_bb_b0};
+
+static struct attn_hw_reg igu_int0_bb_b0 = {
+	0, 11, 0x180180, 0x18018c, 0x180188, 0x180184};
+
+static struct attn_hw_reg *igu_int_bb_b0_regs[1] = {
+	&igu_int0_bb_b0};
+
+static struct attn_hw_reg igu_prty0_bb_b0 = {
+	0, 1, 0x180190, 0x18019c, 0x180198, 0x180194};
+
+static struct attn_hw_reg igu_prty1_bb_b0 = {
+	1, 31, 0x180200, 0x18020c, 0x180208, 0x180204};
+
+static struct attn_hw_reg igu_prty2_bb_b0 = {
+	2, 1, 0x180210, 0x18021c, 0x180218, 0x180214};
+
+static struct attn_hw_reg *igu_prty_bb_b0_regs[3] = {
+	&igu_prty0_bb_b0, &igu_prty1_bb_b0, &igu_prty2_bb_b0};
+
+static struct attn_hw_reg cau_int0_bb_b0 = {
+	0, 11, 0x1c00d4, 0x1c00d8, 0x1c00dc, 0x1c00e0};
+
+static struct attn_hw_reg *cau_int_bb_b0_regs[1] = {
+	&cau_int0_bb_b0};
+
+static struct attn_hw_reg cau_prty1_bb_b0 = {
+	0, 13, 0x1c0200, 0x1c020c, 0x1c0208, 0x1c0204};
+
+static struct attn_hw_reg *cau_prty_bb_b0_regs[1] = {
+	&cau_prty1_bb_b0};
+
+static struct attn_hw_reg dbg_int0_bb_b0 = {
+	0, 1, 0x10180, 0x1018c, 0x10188, 0x10184};
+
+static struct attn_hw_reg *dbg_int_bb_b0_regs[1] = {
+	&dbg_int0_bb_b0};
+
+static struct attn_hw_reg dbg_prty1_bb_b0 = {
+	0, 1, 0x10200, 0x1020c, 0x10208, 0x10204};
+
+static struct attn_hw_reg *dbg_prty_bb_b0_regs[1] = {
+	&dbg_prty1_bb_b0};
+
+static struct attn_hw_reg nig_int0_bb_b0 = {
+	0, 12, 0x500040, 0x50004c, 0x500048, 0x500044};
+
+static struct attn_hw_reg nig_int1_bb_b0 = {
+	1, 32, 0x500050, 0x50005c, 0x500058, 0x500054};
+
+static struct attn_hw_reg nig_int2_bb_b0 = {
+	2, 20, 0x500060, 0x50006c, 0x500068, 0x500064};
+
+static struct attn_hw_reg nig_int3_bb_b0 = {
+	3, 18, 0x500070, 0x50007c, 0x500078, 0x500074};
+
+static struct attn_hw_reg nig_int4_bb_b0 = {
+	4, 20, 0x500080, 0x50008c, 0x500088, 0x500084};
+
+static struct attn_hw_reg nig_int5_bb_b0 = {
+	5, 18, 0x500090, 0x50009c, 0x500098, 0x500094};
+
+static struct attn_hw_reg *nig_int_bb_b0_regs[6] = {
+	&nig_int0_bb_b0, &nig_int1_bb_b0, &nig_int2_bb_b0, &nig_int3_bb_b0,
+	&nig_int4_bb_b0, &nig_int5_bb_b0};
+
+static struct attn_hw_reg nig_prty0_bb_b0 = {
+	0, 1, 0x5000a0, 0x5000ac, 0x5000a8, 0x5000a4};
+
+static struct attn_hw_reg nig_prty1_bb_b0 = {
+	1, 31, 0x500200, 0x50020c, 0x500208, 0x500204};
+
+static struct attn_hw_reg nig_prty2_bb_b0 = {
+	2, 31, 0x500210, 0x50021c, 0x500218, 0x500214};
+
+static struct attn_hw_reg nig_prty3_bb_b0 = {
+	3, 31, 0x500220, 0x50022c, 0x500228, 0x500224};
+
+static struct attn_hw_reg nig_prty4_bb_b0 = {
+	4, 17, 0x500230, 0x50023c, 0x500238, 0x500234};
+
+static struct attn_hw_reg *nig_prty_bb_b0_regs[5] = {
+	&nig_prty0_bb_b0, &nig_prty1_bb_b0, &nig_prty2_bb_b0,
+	&nig_prty3_bb_b0, &nig_prty4_bb_b0};
+
+static struct attn_hw_reg ipc_int0_bb_b0 = {
+	0, 13, 0x2050c, 0x20518, 0x20514, 0x20510};
+
+static struct attn_hw_reg *ipc_int_bb_b0_regs[1] = {
+	&ipc_int0_bb_b0};
+
+static struct attn_hw_reg ipc_prty0_bb_b0 = {
+	0, 1, 0x2051c, 0x20528, 0x20524, 0x20520};
+
+static struct attn_hw_reg *ipc_prty_bb_b0_regs[1] = {
+	&ipc_prty0_bb_b0};
+
+static struct attn_hw_block attn_blocks[] = {
+	{"grc", {{1, 1, grc_int_bb_b0_regs, grc_prty_bb_b0_regs} } },
+	{"miscs", {{2, 1, miscs_int_bb_b0_regs, miscs_prty_bb_b0_regs} } },
+	{"misc", {{1, 0, misc_int_bb_b0_regs, NULL} } },
+	{"dbu", {{0, 0, NULL, NULL} } },
+	{"pglue_b", {{1, 2, pglue_b_int_bb_b0_regs,
+		      pglue_b_prty_bb_b0_regs} } },
+	{"cnig", {{1, 1, cnig_int_bb_b0_regs, cnig_prty_bb_b0_regs} } },
+	{"cpmu", {{1, 0, cpmu_int_bb_b0_regs, NULL} } },
+	{"ncsi", {{1, 1, ncsi_int_bb_b0_regs, ncsi_prty_bb_b0_regs} } },
+	{"opte", {{0, 2, NULL, opte_prty_bb_b0_regs} } },
+	{"bmb", {{12, 3, bmb_int_bb_b0_regs, bmb_prty_bb_b0_regs} } },
+	{"pcie", {{0, 1, NULL, pcie_prty_bb_b0_regs} } },
+	{"mcp", {{0, 0, NULL, NULL} } },
+	{"mcp2", {{0, 2, NULL, mcp2_prty_bb_b0_regs} } },
+	{"pswhst", {{1, 2, pswhst_int_bb_b0_regs, pswhst_prty_bb_b0_regs} } },
+	{"pswhst2", {{1, 1, pswhst2_int_bb_b0_regs,
+		      pswhst2_prty_bb_b0_regs} } },
+	{"pswrd", {{1, 1, pswrd_int_bb_b0_regs, pswrd_prty_bb_b0_regs} } },
+	{"pswrd2", {{1, 3, pswrd2_int_bb_b0_regs, pswrd2_prty_bb_b0_regs} } },
+	{"pswwr", {{1, 1, pswwr_int_bb_b0_regs, pswwr_prty_bb_b0_regs} } },
+	{"pswwr2", {{1, 5, pswwr2_int_bb_b0_regs, pswwr2_prty_bb_b0_regs} } },
+	{"pswrq", {{1, 1, pswrq_int_bb_b0_regs, pswrq_prty_bb_b0_regs} } },
+	{"pswrq2", {{1, 1, pswrq2_int_bb_b0_regs, pswrq2_prty_bb_b0_regs} } },
+	{"pglcs", {{1, 0, pglcs_int_bb_b0_regs, NULL} } },
+	{"dmae", {{1, 1, dmae_int_bb_b0_regs, dmae_prty_bb_b0_regs} } },
+	{"ptu", {{1, 1, ptu_int_bb_b0_regs, ptu_prty_bb_b0_regs} } },
+	{"tcm", {{3, 2, tcm_int_bb_b0_regs, tcm_prty_bb_b0_regs} } },
+	{"mcm", {{3, 2, mcm_int_bb_b0_regs, mcm_prty_bb_b0_regs} } },
+	{"ucm", {{3, 2, ucm_int_bb_b0_regs, ucm_prty_bb_b0_regs} } },
+	{"xcm", {{3, 2, xcm_int_bb_b0_regs, xcm_prty_bb_b0_regs} } },
+	{"ycm", {{3, 2, ycm_int_bb_b0_regs, ycm_prty_bb_b0_regs} } },
+	{"pcm", {{3, 1, pcm_int_bb_b0_regs, pcm_prty_bb_b0_regs} } },
+	{"qm", {{1, 4, qm_int_bb_b0_regs, qm_prty_bb_b0_regs} } },
+	{"tm", {{2, 1, tm_int_bb_b0_regs, tm_prty_bb_b0_regs} } },
+	{"dorq", {{1, 2, dorq_int_bb_b0_regs, dorq_prty_bb_b0_regs} } },
+	{"brb", {{12, 3, brb_int_bb_b0_regs, brb_prty_bb_b0_regs} } },
+	{"src", {{1, 0, src_int_bb_b0_regs, NULL} } },
+	{"prs", {{1, 3, prs_int_bb_b0_regs, prs_prty_bb_b0_regs} } },
+	{"tsdm", {{1, 1, tsdm_int_bb_b0_regs, tsdm_prty_bb_b0_regs} } },
+	{"msdm", {{1, 1, msdm_int_bb_b0_regs, msdm_prty_bb_b0_regs} } },
+	{"usdm", {{1, 1, usdm_int_bb_b0_regs, usdm_prty_bb_b0_regs} } },
+	{"xsdm", {{1, 1, xsdm_int_bb_b0_regs, xsdm_prty_bb_b0_regs} } },
+	{"ysdm", {{1, 1, ysdm_int_bb_b0_regs, ysdm_prty_bb_b0_regs} } },
+	{"psdm", {{1, 1, psdm_int_bb_b0_regs, psdm_prty_bb_b0_regs} } },
+	{"tsem", {{3, 3, tsem_int_bb_b0_regs, tsem_prty_bb_b0_regs} } },
+	{"msem", {{3, 2, msem_int_bb_b0_regs, msem_prty_bb_b0_regs} } },
+	{"usem", {{3, 2, usem_int_bb_b0_regs, usem_prty_bb_b0_regs} } },
+	{"xsem", {{3, 2, xsem_int_bb_b0_regs, xsem_prty_bb_b0_regs} } },
+	{"ysem", {{3, 2, ysem_int_bb_b0_regs, ysem_prty_bb_b0_regs} } },
+	{"psem", {{3, 3, psem_int_bb_b0_regs, psem_prty_bb_b0_regs} } },
+	{"rss", {{1, 1, rss_int_bb_b0_regs, rss_prty_bb_b0_regs} } },
+	{"tmld", {{1, 1, tmld_int_bb_b0_regs, tmld_prty_bb_b0_regs} } },
+	{"muld", {{1, 1, muld_int_bb_b0_regs, muld_prty_bb_b0_regs} } },
+	{"yuld", {{1, 1, yuld_int_bb_b0_regs, yuld_prty_bb_b0_regs} } },
+	{"xyld", {{1, 1, xyld_int_bb_b0_regs, xyld_prty_bb_b0_regs} } },
+	{"prm", {{1, 2, prm_int_bb_b0_regs, prm_prty_bb_b0_regs} } },
+	{"pbf_pb1", {{1, 1, pbf_pb1_int_bb_b0_regs,
+		      pbf_pb1_prty_bb_b0_regs} } },
+	{"pbf_pb2", {{1, 1, pbf_pb2_int_bb_b0_regs,
+		      pbf_pb2_prty_bb_b0_regs} } },
+	{"rpb", { {1, 1, rpb_int_bb_b0_regs, rpb_prty_bb_b0_regs} } },
+	{"btb", { {11, 2, btb_int_bb_b0_regs, btb_prty_bb_b0_regs} } },
+	{"pbf", { {1, 3, pbf_int_bb_b0_regs, pbf_prty_bb_b0_regs} } },
+	{"rdif", { {1, 1, rdif_int_bb_b0_regs, rdif_prty_bb_b0_regs} } },
+	{"tdif", { {1, 2, tdif_int_bb_b0_regs, tdif_prty_bb_b0_regs} } },
+	{"cdu", { {1, 1, cdu_int_bb_b0_regs, cdu_prty_bb_b0_regs} } },
+	{"ccfc", { {1, 2, ccfc_int_bb_b0_regs, ccfc_prty_bb_b0_regs} } },
+	{"tcfc", { {1, 2, tcfc_int_bb_b0_regs, tcfc_prty_bb_b0_regs} } },
+	{"igu", { {1, 3, igu_int_bb_b0_regs, igu_prty_bb_b0_regs} } },
+	{"cau", { {1, 1, cau_int_bb_b0_regs, cau_prty_bb_b0_regs} } },
+	{"umac", { {0, 0, NULL, NULL} } },
+	{"xmac", { {0, 0, NULL, NULL} } },
+	{"dbg", { {1, 1, dbg_int_bb_b0_regs, dbg_prty_bb_b0_regs} } },
+	{"nig", { {6, 5, nig_int_bb_b0_regs, nig_prty_bb_b0_regs} } },
+	{"wol", { {0, 0, NULL, NULL} } },
+	{"bmbn", { {0, 0, NULL, NULL} } },
+	{"ipc", { {1, 1, ipc_int_bb_b0_regs, ipc_prty_bb_b0_regs} } },
+	{"nwm", { {0, 0, NULL, NULL} } },
+	{"nws", { {0, 0, NULL, NULL} } },
+	{"ms", { {0, 0, NULL, NULL} } },
+	{"phy_pcie", { {0, 0, NULL, NULL} } },
+	{"misc_aeu", { {0, 0, NULL, NULL} } },
+	{"bar0_map", { {0, 0, NULL, NULL} } },};
+
 /* Notice aeu_invert_reg must be defined in the same order of bits as HW;  */
 static struct aeu_invert_reg aeu_descs[NUM_ATTN_REGS] = {
 	{
 		{       /* After Invert 1 */
 			{"GPIO0 function%d",
-			 (32 << ATTENTION_LENGTH_SHIFT)},
+			 (32 << ATTENTION_LENGTH_SHIFT), MAX_BLOCK_ID},
 		}
 	},
 
 	{
 		{       /* After Invert 2 */
-			{"PGLUE config_space", ATTENTION_SINGLE},
-			{"PGLUE misc_flr", ATTENTION_SINGLE},
-			{"PGLUE B RBC", ATTENTION_PAR_INT},
-			{"PGLUE misc_mctp", ATTENTION_SINGLE},
-			{"Flash event", ATTENTION_SINGLE},
-			{"SMB event", ATTENTION_SINGLE},
-			{"Main Power", ATTENTION_SINGLE},
+			{"PGLUE config_space", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"PGLUE misc_flr", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"PGLUE B RBC", ATTENTION_PAR_INT, BLOCK_PGLUE_B},
+			{"PGLUE misc_mctp", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"Flash event", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"SMB event", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"Main Power", ATTENTION_SINGLE, MAX_BLOCK_ID},
 			{"SW timers #%d", (8 << ATTENTION_LENGTH_SHIFT) |
-					  (1 << ATTENTION_OFFSET_SHIFT)},
+					  (1 << ATTENTION_OFFSET_SHIFT),
+			 MAX_BLOCK_ID},
 			{"PCIE glue/PXP VPD %d",
-			 (16 << ATTENTION_LENGTH_SHIFT)},
+			 (16 << ATTENTION_LENGTH_SHIFT), BLOCK_PGLCS},
 		}
 	},
 
 	{
 		{       /* After Invert 3 */
 			{"General Attention %d",
-			 (32 << ATTENTION_LENGTH_SHIFT)},
+			 (32 << ATTENTION_LENGTH_SHIFT), MAX_BLOCK_ID},
 		}
 	},
 
 	{
 		{       /* After Invert 4 */
-			{"General Attention 32", ATTENTION_SINGLE},
+			{"General Attention 32", ATTENTION_SINGLE,
+			 MAX_BLOCK_ID},
 			{"General Attention %d",
 			 (2 << ATTENTION_LENGTH_SHIFT) |
-			 (33 << ATTENTION_OFFSET_SHIFT)},
-			{"General Attention 35", ATTENTION_SINGLE},
-			{"CNIG port %d", (4 << ATTENTION_LENGTH_SHIFT)},
-			{"MCP CPU", ATTENTION_SINGLE},
-			{"MCP Watchdog timer", ATTENTION_SINGLE},
-			{"MCP M2P", ATTENTION_SINGLE},
-			{"AVS stop status ready", ATTENTION_SINGLE},
-			{"MSTAT", ATTENTION_PAR_INT},
-			{"MSTAT per-path", ATTENTION_PAR_INT},
-			{"Reserved %d", (6 << ATTENTION_LENGTH_SHIFT)},
-			{"NIG", ATTENTION_PAR_INT},
-			{"BMB/OPTE/MCP", ATTENTION_PAR_INT},
-			{"BTB",	ATTENTION_PAR_INT},
-			{"BRB",	ATTENTION_PAR_INT},
-			{"PRS",	ATTENTION_PAR_INT},
+			 (33 << ATTENTION_OFFSET_SHIFT), MAX_BLOCK_ID},
+			{"General Attention 35", ATTENTION_SINGLE,
+			 MAX_BLOCK_ID},
+			{"CNIG port %d", (4 << ATTENTION_LENGTH_SHIFT),
+			 BLOCK_CNIG},
+			{"MCP CPU", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"MCP Watchdog timer", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"MCP M2P", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"AVS stop status ready", ATTENTION_SINGLE,
+			 MAX_BLOCK_ID},
+			{"MSTAT", ATTENTION_PAR_INT, MAX_BLOCK_ID},
+			{"MSTAT per-path", ATTENTION_PAR_INT, MAX_BLOCK_ID},
+			{"Reserved %d", (6 << ATTENTION_LENGTH_SHIFT),
+			 MAX_BLOCK_ID},
+			{"NIG", ATTENTION_PAR_INT, BLOCK_NIG},
+			{"BMB/OPTE/MCP", ATTENTION_PAR_INT, BLOCK_BMB},
+			{"BTB",	ATTENTION_PAR_INT, BLOCK_BTB},
+			{"BRB",	ATTENTION_PAR_INT, BLOCK_BRB},
+			{"PRS",	ATTENTION_PAR_INT, BLOCK_PRS},
 		}
 	},
 
 	{
 		{       /* After Invert 5 */
-			{"SRC", ATTENTION_PAR_INT},
-			{"PB Client1", ATTENTION_PAR_INT},
-			{"PB Client2", ATTENTION_PAR_INT},
-			{"RPB", ATTENTION_PAR_INT},
-			{"PBF", ATTENTION_PAR_INT},
-			{"QM", ATTENTION_PAR_INT},
-			{"TM", ATTENTION_PAR_INT},
-			{"MCM",  ATTENTION_PAR_INT},
-			{"MSDM", ATTENTION_PAR_INT},
-			{"MSEM", ATTENTION_PAR_INT},
-			{"PCM", ATTENTION_PAR_INT},
-			{"PSDM", ATTENTION_PAR_INT},
-			{"PSEM", ATTENTION_PAR_INT},
-			{"TCM", ATTENTION_PAR_INT},
-			{"TSDM", ATTENTION_PAR_INT},
-			{"TSEM", ATTENTION_PAR_INT},
+			{"SRC", ATTENTION_PAR_INT, BLOCK_SRC},
+			{"PB Client1", ATTENTION_PAR_INT, BLOCK_PBF_PB1},
+			{"PB Client2", ATTENTION_PAR_INT, BLOCK_PBF_PB2},
+			{"RPB", ATTENTION_PAR_INT, BLOCK_RPB},
+			{"PBF", ATTENTION_PAR_INT, BLOCK_PBF},
+			{"QM", ATTENTION_PAR_INT, BLOCK_QM},
+			{"TM", ATTENTION_PAR_INT, BLOCK_TM},
+			{"MCM",  ATTENTION_PAR_INT, BLOCK_MCM},
+			{"MSDM", ATTENTION_PAR_INT, BLOCK_MSDM},
+			{"MSEM", ATTENTION_PAR_INT, BLOCK_MSEM},
+			{"PCM", ATTENTION_PAR_INT, BLOCK_PCM},
+			{"PSDM", ATTENTION_PAR_INT, BLOCK_PSDM},
+			{"PSEM", ATTENTION_PAR_INT, BLOCK_PSEM},
+			{"TCM", ATTENTION_PAR_INT, BLOCK_TCM},
+			{"TSDM", ATTENTION_PAR_INT, BLOCK_TSDM},
+			{"TSEM", ATTENTION_PAR_INT, BLOCK_TSEM},
 		}
 	},
 
 	{
 		{       /* After Invert 6 */
-			{"UCM", ATTENTION_PAR_INT},
-			{"USDM", ATTENTION_PAR_INT},
-			{"USEM", ATTENTION_PAR_INT},
-			{"XCM",	ATTENTION_PAR_INT},
-			{"XSDM", ATTENTION_PAR_INT},
-			{"XSEM", ATTENTION_PAR_INT},
-			{"YCM",	ATTENTION_PAR_INT},
-			{"YSDM", ATTENTION_PAR_INT},
-			{"YSEM", ATTENTION_PAR_INT},
-			{"XYLD", ATTENTION_PAR_INT},
-			{"TMLD", ATTENTION_PAR_INT},
-			{"MYLD", ATTENTION_PAR_INT},
-			{"YULD", ATTENTION_PAR_INT},
-			{"DORQ", ATTENTION_PAR_INT},
-			{"DBG", ATTENTION_PAR_INT},
-			{"IPC",	ATTENTION_PAR_INT},
+			{"UCM", ATTENTION_PAR_INT, BLOCK_UCM},
+			{"USDM", ATTENTION_PAR_INT, BLOCK_USDM},
+			{"USEM", ATTENTION_PAR_INT, BLOCK_USEM},
+			{"XCM",	ATTENTION_PAR_INT, BLOCK_XCM},
+			{"XSDM", ATTENTION_PAR_INT, BLOCK_XSDM},
+			{"XSEM", ATTENTION_PAR_INT, BLOCK_XSEM},
+			{"YCM",	ATTENTION_PAR_INT, BLOCK_YCM},
+			{"YSDM", ATTENTION_PAR_INT, BLOCK_YSDM},
+			{"YSEM", ATTENTION_PAR_INT, BLOCK_YSEM},
+			{"XYLD", ATTENTION_PAR_INT, BLOCK_XYLD},
+			{"TMLD", ATTENTION_PAR_INT, BLOCK_TMLD},
+			{"MYLD", ATTENTION_PAR_INT, BLOCK_MULD},
+			{"YULD", ATTENTION_PAR_INT, BLOCK_YULD},
+			{"DORQ", ATTENTION_PAR_INT, BLOCK_DORQ},
+			{"DBG", ATTENTION_PAR_INT, BLOCK_DBG},
+			{"IPC",	ATTENTION_PAR_INT, BLOCK_IPC},
 		}
 	},
 
 	{
 		{       /* After Invert 7 */
-			{"CCFC", ATTENTION_PAR_INT},
-			{"CDU", ATTENTION_PAR_INT},
-			{"DMAE", ATTENTION_PAR_INT},
-			{"IGU", ATTENTION_PAR_INT},
-			{"ATC", ATTENTION_PAR_INT},
-			{"CAU", ATTENTION_PAR_INT},
-			{"PTU", ATTENTION_PAR_INT},
-			{"PRM", ATTENTION_PAR_INT},
-			{"TCFC", ATTENTION_PAR_INT},
-			{"RDIF", ATTENTION_PAR_INT},
-			{"TDIF", ATTENTION_PAR_INT},
-			{"RSS", ATTENTION_PAR_INT},
-			{"MISC", ATTENTION_PAR_INT},
-			{"MISCS", ATTENTION_PAR_INT},
-			{"PCIE", ATTENTION_PAR},
-			{"Vaux PCI core", ATTENTION_SINGLE},
-			{"PSWRQ", ATTENTION_PAR_INT},
+			{"CCFC", ATTENTION_PAR_INT, BLOCK_CCFC},
+			{"CDU", ATTENTION_PAR_INT, BLOCK_CDU},
+			{"DMAE", ATTENTION_PAR_INT, BLOCK_DMAE},
+			{"IGU", ATTENTION_PAR_INT, BLOCK_IGU},
+			{"ATC", ATTENTION_PAR_INT, MAX_BLOCK_ID},
+			{"CAU", ATTENTION_PAR_INT, BLOCK_CAU},
+			{"PTU", ATTENTION_PAR_INT, BLOCK_PTU},
+			{"PRM", ATTENTION_PAR_INT, BLOCK_PRM},
+			{"TCFC", ATTENTION_PAR_INT, BLOCK_TCFC},
+			{"RDIF", ATTENTION_PAR_INT, BLOCK_RDIF},
+			{"TDIF", ATTENTION_PAR_INT, BLOCK_TDIF},
+			{"RSS", ATTENTION_PAR_INT, BLOCK_RSS},
+			{"MISC", ATTENTION_PAR_INT, BLOCK_MISC},
+			{"MISCS", ATTENTION_PAR_INT, BLOCK_MISCS},
+			{"PCIE", ATTENTION_PAR, BLOCK_PCIE},
+			{"Vaux PCI core", ATTENTION_SINGLE, BLOCK_PGLCS},
+			{"PSWRQ", ATTENTION_PAR_INT, BLOCK_PSWRQ},
 		}
 	},
 
 	{
 		{       /* After Invert 8 */
-			{"PSWRQ (pci_clk)", ATTENTION_PAR_INT},
-			{"PSWWR", ATTENTION_PAR_INT},
-			{"PSWWR (pci_clk)", ATTENTION_PAR_INT},
-			{"PSWRD", ATTENTION_PAR_INT},
-			{"PSWRD (pci_clk)", ATTENTION_PAR_INT},
-			{"PSWHST", ATTENTION_PAR_INT},
-			{"PSWHST (pci_clk)", ATTENTION_PAR_INT},
-			{"GRC",	ATTENTION_PAR_INT},
-			{"CPMU", ATTENTION_PAR_INT},
-			{"NCSI", ATTENTION_PAR_INT},
-			{"MSEM PRAM", ATTENTION_PAR},
-			{"PSEM PRAM", ATTENTION_PAR},
-			{"TSEM PRAM", ATTENTION_PAR},
-			{"USEM PRAM", ATTENTION_PAR},
-			{"XSEM PRAM", ATTENTION_PAR},
-			{"YSEM PRAM", ATTENTION_PAR},
-			{"pxp_misc_mps", ATTENTION_PAR},
-			{"PCIE glue/PXP Exp. ROM", ATTENTION_SINGLE},
-			{"PERST_B assertion", ATTENTION_SINGLE},
-			{"PERST_B deassertion", ATTENTION_SINGLE},
-			{"Reserved %d", (2 << ATTENTION_LENGTH_SHIFT)},
+			{"PSWRQ (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWRQ2},
+			{"PSWWR", ATTENTION_PAR_INT, BLOCK_PSWWR},
+			{"PSWWR (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWWR2},
+			{"PSWRD", ATTENTION_PAR_INT, BLOCK_PSWRD},
+			{"PSWRD (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWRD2},
+			{"PSWHST", ATTENTION_PAR_INT, BLOCK_PSWHST},
+			{"PSWHST (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWHST2},
+			{"GRC",	ATTENTION_PAR_INT, BLOCK_GRC},
+			{"CPMU", ATTENTION_PAR_INT, BLOCK_CPMU},
+			{"NCSI", ATTENTION_PAR_INT, BLOCK_NCSI},
+			{"MSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"PSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"TSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"USEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"XSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"YSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"pxp_misc_mps", ATTENTION_PAR, BLOCK_PGLCS},
+			{"PCIE glue/PXP Exp. ROM", ATTENTION_SINGLE,
+			 BLOCK_PGLCS},
+			{"PERST_B assertion", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"PERST_B deassertion", ATTENTION_SINGLE,
+			 MAX_BLOCK_ID},
+			{"Reserved %d", (2 << ATTENTION_LENGTH_SHIFT),
+			 MAX_BLOCK_ID},
 		}
 	},
 
 	{
 		{       /* After Invert 9 */
-			{"MCP Latched memory", ATTENTION_PAR},
-			{"MCP Latched scratchpad cache", ATTENTION_SINGLE},
-			{"MCP Latched ump_tx", ATTENTION_PAR},
-			{"MCP Latched scratchpad", ATTENTION_PAR},
-			{"Reserved %d", (28 << ATTENTION_LENGTH_SHIFT)},
+			{"MCP Latched memory", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"MCP Latched scratchpad cache", ATTENTION_SINGLE,
+			 MAX_BLOCK_ID},
+			{"MCP Latched ump_tx", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"MCP Latched scratchpad", ATTENTION_PAR,
+			 MAX_BLOCK_ID},
+			{"Reserved %d", (28 << ATTENTION_LENGTH_SHIFT),
+			 MAX_BLOCK_ID},
 		}
 	},
 };
@@ -316,6 +1544,28 @@ static int qed_int_assertion(struct qed_hwfn *p_hwfn,
 	return 0;
 }
 
+static void qed_int_deassertion_print_bit(struct qed_hwfn *p_hwfn,
+					  struct attn_hw_reg *p_reg_desc,
+					  struct attn_hw_block *p_block,
+					  enum qed_attention_type type,
+					  u32 val, u32 mask)
+{
+	int j;
+
+	for (j = 0; j < p_reg_desc->num_of_bits; j++) {
+		if (!(val & (1 << j)))
+			continue;
+
+		DP_NOTICE(p_hwfn,
+			  "%s (%s): reg %d [0x%08x], bit %d [%s]\n",
+			  p_block->name,
+			  type == QED_ATTN_TYPE_ATTN ? "Interrupt" :
+						       "Parity",
+			  p_reg_desc->reg_idx, p_reg_desc->sts_addr,
+			  j, (mask & (1 << j)) ? " [MASKED]" : "");
+	}
+}
+
 /**
  * @brief qed_int_deassertion_aeu_bit - handles the effects of a single
  * cause of the attention
@@ -340,6 +1590,31 @@ qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn,
 	DP_INFO(p_hwfn, "Deasserted attention `%s'[%08x]\n",
 		p_aeu->bit_name, bitmask);
 
+	/* Handle HW block interrupt registers */
+	if (p_aeu->block_index != MAX_BLOCK_ID) {
+		struct attn_hw_block *p_block;
+		int i;
+
+		p_block = &attn_blocks[p_aeu->block_index];
+
+		/* Handle each interrupt register */
+		for (i = 0; i < p_block->chip_regs[0].num_of_int_regs; i++) {
+			struct attn_hw_reg *p_reg_desc;
+			u32 sts_addr;
+
+			p_reg_desc = p_block->chip_regs[0].int_regs[i];
+			sts_addr = p_reg_desc->sts_addr;
+
+			val = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, sts_addr);
+			mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				      p_reg_desc->mask_addr);
+			qed_int_deassertion_print_bit(p_hwfn, p_reg_desc,
+						      p_block,
+						      QED_ATTN_TYPE_ATTN,
+						      val, mask);
+		}
+	}
+
 	/* Prevent this Attention from being asserted in the future */
 	val = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg);
 	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg, (val & mask));
@@ -349,6 +1624,62 @@ qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn,
 	return rc;
 }
 
+static void qed_int_parity_print(struct qed_hwfn *p_hwfn,
+				 struct aeu_invert_reg_bit *p_aeu,
+				 struct attn_hw_block *p_block,
+				 u8 bit_index)
+{
+	int i;
+
+	for (i = 0; i < p_block->chip_regs[0].num_of_prty_regs; i++) {
+		struct attn_hw_reg *p_reg_desc;
+		u32 val, mask;
+
+		p_reg_desc = p_block->chip_regs[0].prty_regs[i];
+
+		val = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+			     p_reg_desc->sts_clr_addr);
+		mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+			      p_reg_desc->mask_addr);
+		qed_int_deassertion_print_bit(p_hwfn, p_reg_desc,
+					      p_block,
+					      QED_ATTN_TYPE_PARITY,
+					      val, mask);
+	}
+}
+
+/**
+ * @brief qed_int_deassertion_parity - handle a single parity AEU source
+ *
+ * @param p_hwfn
+ * @param p_aeu - descriptor of an AEU bit which caused the parity
+ * @param bit_index
+ */
+static void qed_int_deassertion_parity(struct qed_hwfn *p_hwfn,
+				       struct aeu_invert_reg_bit *p_aeu,
+				       u8 bit_index)
+{
+	u32 block_id = p_aeu->block_index;
+
+	DP_INFO(p_hwfn->cdev, "%s[%d] parity attention is set\n",
+		p_aeu->bit_name, bit_index);
+
+	if (block_id != MAX_BLOCK_ID) {
+		qed_int_parity_print(p_hwfn, p_aeu, &attn_blocks[block_id],
+				     bit_index);
+
+		/* In BB, there's a single parity bit for several blocks */
+		if (block_id == BLOCK_BTB) {
+			qed_int_parity_print(p_hwfn, p_aeu,
+					     &attn_blocks[BLOCK_OPTE],
+					     bit_index);
+			qed_int_parity_print(p_hwfn, p_aeu,
+					     &attn_blocks[BLOCK_MCP],
+					     bit_index);
+		}
+	}
+}
+
 /**
  * @brief - handles deassertion of previously asserted attentions.
  *
@@ -392,11 +1723,9 @@ static int qed_int_deassertion(struct qed_hwfn  *p_hwfn,
 			struct aeu_invert_reg_bit *p_bit = &p_aeu->bits[j];
 
 			if ((p_bit->flags & ATTENTION_PARITY) &&
-			    !!(parities & (1 << bit_idx))) {
-				DP_INFO(p_hwfn,
-					"%s[%d] parity attention is set\n",
-					p_bit->bit_name, bit_idx);
-			}
+			    !!(parities & (1 << bit_idx)))
+				qed_int_deassertion_parity(p_hwfn, p_bit,
+							   bit_idx);
 
 			bit_idx += ATTENTION_LENGTH(p_bit->flags);
 		}
-- 
cgit v0.10.2


From b4149dc7e11e8fe58773ecb8862c5217ed205ad9 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 28 Feb 2016 12:26:55 +0200
Subject: qed: Print additional HW attention info

This patch utilizes the attention infrastructure to log additional
information that relates only to specific HW blocks.
For some of those HW blocks, it also stops automatically disabling the
attention generation as the attention is considered benign and thus
should only be logged; No fear of it flooding the system.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c
index c8bca77..ffd0acc 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_int.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_int.c
@@ -66,6 +66,9 @@ struct aeu_invert_reg_bit {
 #define ATTENTION_OFFSET_SHIFT          (12)
 	unsigned int flags;
 
+	/* Callback to call if attention will be triggered */
+	int (*cb)(struct qed_hwfn *p_hwfn);
+
 	enum block_id block_index;
 };
 
@@ -1285,170 +1288,463 @@ static struct attn_hw_block attn_blocks[] = {
 	{"misc_aeu", { {0, 0, NULL, NULL} } },
 	{"bar0_map", { {0, 0, NULL, NULL} } },};
 
+/* Specific HW attention callbacks */
+static int qed_mcp_attn_cb(struct qed_hwfn *p_hwfn)
+{
+	u32 tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, MCP_REG_CPU_STATE);
+
+	/* This might occur on certain instances; Log it once then mask it */
+	DP_INFO(p_hwfn->cdev, "MCP_REG_CPU_STATE: %08x - Masking...\n",
+		tmp);
+	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, MCP_REG_CPU_EVENT_MASK,
+	       0xffffffff);
+
+	return 0;
+}
+
+#define QED_PSWHST_ATTENTION_INCORRECT_ACCESS		(0x1)
+#define ATTENTION_INCORRECT_ACCESS_WR_MASK		(0x1)
+#define ATTENTION_INCORRECT_ACCESS_WR_SHIFT		(0)
+#define ATTENTION_INCORRECT_ACCESS_CLIENT_MASK		(0xf)
+#define ATTENTION_INCORRECT_ACCESS_CLIENT_SHIFT		(1)
+#define ATTENTION_INCORRECT_ACCESS_VF_VALID_MASK	(0x1)
+#define ATTENTION_INCORRECT_ACCESS_VF_VALID_SHIFT	(5)
+#define ATTENTION_INCORRECT_ACCESS_VF_ID_MASK		(0xff)
+#define ATTENTION_INCORRECT_ACCESS_VF_ID_SHIFT		(6)
+#define ATTENTION_INCORRECT_ACCESS_PF_ID_MASK		(0xf)
+#define ATTENTION_INCORRECT_ACCESS_PF_ID_SHIFT		(14)
+#define ATTENTION_INCORRECT_ACCESS_BYTE_EN_MASK		(0xff)
+#define ATTENTION_INCORRECT_ACCESS_BYTE_EN_SHIFT	(18)
+static int qed_pswhst_attn_cb(struct qed_hwfn *p_hwfn)
+{
+	u32 tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+			 PSWHST_REG_INCORRECT_ACCESS_VALID);
+
+	if (tmp & QED_PSWHST_ATTENTION_INCORRECT_ACCESS) {
+		u32 addr, data, length;
+
+		addr = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+			      PSWHST_REG_INCORRECT_ACCESS_ADDRESS);
+		data = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+			      PSWHST_REG_INCORRECT_ACCESS_DATA);
+		length = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				PSWHST_REG_INCORRECT_ACCESS_LENGTH);
+
+		DP_INFO(p_hwfn->cdev,
+			"Incorrect access to %08x of length %08x - PF [%02x] VF [%04x] [valid %02x] client [%02x] write [%02x] Byte-Enable [%04x] [%08x]\n",
+			addr, length,
+			(u8) GET_FIELD(data, ATTENTION_INCORRECT_ACCESS_PF_ID),
+			(u8) GET_FIELD(data, ATTENTION_INCORRECT_ACCESS_VF_ID),
+			(u8) GET_FIELD(data,
+				       ATTENTION_INCORRECT_ACCESS_VF_VALID),
+			(u8) GET_FIELD(data,
+				       ATTENTION_INCORRECT_ACCESS_CLIENT),
+			(u8) GET_FIELD(data, ATTENTION_INCORRECT_ACCESS_WR),
+			(u8) GET_FIELD(data,
+				       ATTENTION_INCORRECT_ACCESS_BYTE_EN),
+			data);
+	}
+
+	return 0;
+}
+
+#define QED_GRC_ATTENTION_VALID_BIT	(1 << 0)
+#define QED_GRC_ATTENTION_ADDRESS_MASK	(0x7fffff)
+#define QED_GRC_ATTENTION_ADDRESS_SHIFT	(0)
+#define QED_GRC_ATTENTION_RDWR_BIT	(1 << 23)
+#define QED_GRC_ATTENTION_MASTER_MASK	(0xf)
+#define QED_GRC_ATTENTION_MASTER_SHIFT	(24)
+#define QED_GRC_ATTENTION_PF_MASK	(0xf)
+#define QED_GRC_ATTENTION_PF_SHIFT	(0)
+#define QED_GRC_ATTENTION_VF_MASK	(0xff)
+#define QED_GRC_ATTENTION_VF_SHIFT	(4)
+#define QED_GRC_ATTENTION_PRIV_MASK	(0x3)
+#define QED_GRC_ATTENTION_PRIV_SHIFT	(14)
+#define QED_GRC_ATTENTION_PRIV_VF	(0)
+static const char *attn_master_to_str(u8 master)
+{
+	switch (master) {
+	case 1: return "PXP";
+	case 2: return "MCP";
+	case 3: return "MSDM";
+	case 4: return "PSDM";
+	case 5: return "YSDM";
+	case 6: return "USDM";
+	case 7: return "TSDM";
+	case 8: return "XSDM";
+	case 9: return "DBU";
+	case 10: return "DMAE";
+	default:
+		return "Unkown";
+	}
+}
+
+static int qed_grc_attn_cb(struct qed_hwfn *p_hwfn)
+{
+	u32 tmp, tmp2;
+
+	/* We've already cleared the timeout interrupt register, so we learn
+	 * of interrupts via the validity register
+	 */
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     GRC_REG_TIMEOUT_ATTN_ACCESS_VALID);
+	if (!(tmp & QED_GRC_ATTENTION_VALID_BIT))
+		goto out;
+
+	/* Read the GRC timeout information */
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     GRC_REG_TIMEOUT_ATTN_ACCESS_DATA_0);
+	tmp2 = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		      GRC_REG_TIMEOUT_ATTN_ACCESS_DATA_1);
+
+	DP_INFO(p_hwfn->cdev,
+		"GRC timeout [%08x:%08x] - %s Address [%08x] [Master %s] [PF: %02x %s %02x]\n",
+		tmp2, tmp,
+		(tmp & QED_GRC_ATTENTION_RDWR_BIT) ? "Write to" : "Read from",
+		GET_FIELD(tmp, QED_GRC_ATTENTION_ADDRESS) << 2,
+		attn_master_to_str(GET_FIELD(tmp, QED_GRC_ATTENTION_MASTER)),
+		GET_FIELD(tmp2, QED_GRC_ATTENTION_PF),
+		(GET_FIELD(tmp2, QED_GRC_ATTENTION_PRIV) ==
+		 QED_GRC_ATTENTION_PRIV_VF) ? "VF" : "(Ireelevant)",
+		GET_FIELD(tmp2, QED_GRC_ATTENTION_VF));
+
+out:
+	/* Regardles of anything else, clean the validity bit */
+	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt,
+	       GRC_REG_TIMEOUT_ATTN_ACCESS_VALID, 0);
+	return 0;
+}
+
+#define PGLUE_ATTENTION_VALID			(1 << 29)
+#define PGLUE_ATTENTION_RD_VALID		(1 << 26)
+#define PGLUE_ATTENTION_DETAILS_PFID_MASK	(0xf)
+#define PGLUE_ATTENTION_DETAILS_PFID_SHIFT	(20)
+#define PGLUE_ATTENTION_DETAILS_VF_VALID_MASK	(0x1)
+#define PGLUE_ATTENTION_DETAILS_VF_VALID_SHIFT	(19)
+#define PGLUE_ATTENTION_DETAILS_VFID_MASK	(0xff)
+#define PGLUE_ATTENTION_DETAILS_VFID_SHIFT	(24)
+#define PGLUE_ATTENTION_DETAILS2_WAS_ERR_MASK	(0x1)
+#define PGLUE_ATTENTION_DETAILS2_WAS_ERR_SHIFT	(21)
+#define PGLUE_ATTENTION_DETAILS2_BME_MASK	(0x1)
+#define PGLUE_ATTENTION_DETAILS2_BME_SHIFT	(22)
+#define PGLUE_ATTENTION_DETAILS2_FID_EN_MASK	(0x1)
+#define PGLUE_ATTENTION_DETAILS2_FID_EN_SHIFT	(23)
+#define PGLUE_ATTENTION_ICPL_VALID		(1 << 23)
+#define PGLUE_ATTENTION_ZLR_VALID		(1 << 25)
+#define PGLUE_ATTENTION_ILT_VALID		(1 << 23)
+static int qed_pglub_rbc_attn_cb(struct qed_hwfn *p_hwfn)
+{
+	u32 tmp;
+
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     PGLUE_B_REG_TX_ERR_WR_DETAILS2);
+	if (tmp & PGLUE_ATTENTION_VALID) {
+		u32 addr_lo, addr_hi, details;
+
+		addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_TX_ERR_WR_ADD_31_0);
+		addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_TX_ERR_WR_ADD_63_32);
+		details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_TX_ERR_WR_DETAILS);
+
+		DP_INFO(p_hwfn,
+			"Illegal write by chip to [%08x:%08x] blocked.\n"
+			"Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n"
+			"Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n",
+			addr_hi, addr_lo, details,
+			(u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID),
+			(u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID),
+			GET_FIELD(details,
+				  PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0,
+			tmp,
+			GET_FIELD(tmp,
+				  PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0,
+			GET_FIELD(tmp,
+				  PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0,
+			GET_FIELD(tmp,
+				  PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0);
+	}
+
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     PGLUE_B_REG_TX_ERR_RD_DETAILS2);
+	if (tmp & PGLUE_ATTENTION_RD_VALID) {
+		u32 addr_lo, addr_hi, details;
+
+		addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_TX_ERR_RD_ADD_31_0);
+		addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_TX_ERR_RD_ADD_63_32);
+		details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_TX_ERR_RD_DETAILS);
+
+		DP_INFO(p_hwfn,
+			"Illegal read by chip from [%08x:%08x] blocked.\n"
+			" Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n"
+			" Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n",
+			addr_hi, addr_lo, details,
+			(u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID),
+			(u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID),
+			GET_FIELD(details,
+				  PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0,
+			tmp,
+			GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1
+									 : 0,
+			GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0,
+			GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1
+									: 0);
+	}
+
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     PGLUE_B_REG_TX_ERR_WR_DETAILS_ICPL);
+	if (tmp & PGLUE_ATTENTION_ICPL_VALID)
+		DP_INFO(p_hwfn, "ICPL eror - %08x\n", tmp);
+
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     PGLUE_B_REG_MASTER_ZLR_ERR_DETAILS);
+	if (tmp & PGLUE_ATTENTION_ZLR_VALID) {
+		u32 addr_hi, addr_lo;
+
+		addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_MASTER_ZLR_ERR_ADD_31_0);
+		addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_MASTER_ZLR_ERR_ADD_63_32);
+
+		DP_INFO(p_hwfn, "ZLR eror - %08x [Address %08x:%08x]\n",
+			tmp, addr_hi, addr_lo);
+	}
+
+	tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+		     PGLUE_B_REG_VF_ILT_ERR_DETAILS2);
+	if (tmp & PGLUE_ATTENTION_ILT_VALID) {
+		u32 addr_hi, addr_lo, details;
+
+		addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_VF_ILT_ERR_ADD_31_0);
+		addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_VF_ILT_ERR_ADD_63_32);
+		details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				 PGLUE_B_REG_VF_ILT_ERR_DETAILS);
+
+		DP_INFO(p_hwfn,
+			"ILT error - Details %08x Details2 %08x [Address %08x:%08x]\n",
+			details, tmp, addr_hi, addr_lo);
+	}
+
+	/* Clear the indications */
+	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt,
+	       PGLUE_B_REG_LATCHED_ERRORS_CLR, (1 << 2));
+
+	return 0;
+}
+
+#define QED_DORQ_ATTENTION_REASON_MASK	(0xfffff)
+#define QED_DORQ_ATTENTION_OPAQUE_MASK (0xffff)
+#define QED_DORQ_ATTENTION_SIZE_MASK	(0x7f)
+#define QED_DORQ_ATTENTION_SIZE_SHIFT	(16)
+static int qed_dorq_attn_cb(struct qed_hwfn *p_hwfn)
+{
+	u32 reason;
+
+	reason = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, DORQ_REG_DB_DROP_REASON) &
+			QED_DORQ_ATTENTION_REASON_MASK;
+	if (reason) {
+		u32 details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+				     DORQ_REG_DB_DROP_DETAILS);
+
+		DP_INFO(p_hwfn->cdev,
+			"DORQ db_drop: adress 0x%08x Opaque FID 0x%04x Size [bytes] 0x%08x Reason: 0x%08x\n",
+			qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
+			       DORQ_REG_DB_DROP_DETAILS_ADDRESS),
+			(u16)(details & QED_DORQ_ATTENTION_OPAQUE_MASK),
+			GET_FIELD(details, QED_DORQ_ATTENTION_SIZE) * 4,
+			reason);
+	}
+
+	return -EINVAL;
+}
+
 /* Notice aeu_invert_reg must be defined in the same order of bits as HW;  */
 static struct aeu_invert_reg aeu_descs[NUM_ATTN_REGS] = {
 	{
 		{       /* After Invert 1 */
 			{"GPIO0 function%d",
-			 (32 << ATTENTION_LENGTH_SHIFT), MAX_BLOCK_ID},
+			 (32 << ATTENTION_LENGTH_SHIFT), NULL, MAX_BLOCK_ID},
 		}
 	},
 
 	{
 		{       /* After Invert 2 */
-			{"PGLUE config_space", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"PGLUE misc_flr", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"PGLUE B RBC", ATTENTION_PAR_INT, BLOCK_PGLUE_B},
-			{"PGLUE misc_mctp", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"Flash event", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"SMB event", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"Main Power", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			{"PGLUE config_space", ATTENTION_SINGLE,
+			 NULL, MAX_BLOCK_ID},
+			{"PGLUE misc_flr", ATTENTION_SINGLE,
+			 NULL, MAX_BLOCK_ID},
+			{"PGLUE B RBC", ATTENTION_PAR_INT,
+			 qed_pglub_rbc_attn_cb, BLOCK_PGLUE_B},
+			{"PGLUE misc_mctp", ATTENTION_SINGLE,
+			 NULL, MAX_BLOCK_ID},
+			{"Flash event", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID},
+			{"SMB event", ATTENTION_SINGLE,	NULL, MAX_BLOCK_ID},
+			{"Main Power", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID},
 			{"SW timers #%d", (8 << ATTENTION_LENGTH_SHIFT) |
 					  (1 << ATTENTION_OFFSET_SHIFT),
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 			{"PCIE glue/PXP VPD %d",
-			 (16 << ATTENTION_LENGTH_SHIFT), BLOCK_PGLCS},
+			 (16 << ATTENTION_LENGTH_SHIFT), NULL, BLOCK_PGLCS},
 		}
 	},
 
 	{
 		{       /* After Invert 3 */
 			{"General Attention %d",
-			 (32 << ATTENTION_LENGTH_SHIFT), MAX_BLOCK_ID},
+			 (32 << ATTENTION_LENGTH_SHIFT), NULL, MAX_BLOCK_ID},
 		}
 	},
 
 	{
 		{       /* After Invert 4 */
 			{"General Attention 32", ATTENTION_SINGLE,
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 			{"General Attention %d",
 			 (2 << ATTENTION_LENGTH_SHIFT) |
-			 (33 << ATTENTION_OFFSET_SHIFT), MAX_BLOCK_ID},
+			 (33 << ATTENTION_OFFSET_SHIFT), NULL, MAX_BLOCK_ID},
 			{"General Attention 35", ATTENTION_SINGLE,
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 			{"CNIG port %d", (4 << ATTENTION_LENGTH_SHIFT),
-			 BLOCK_CNIG},
-			{"MCP CPU", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"MCP Watchdog timer", ATTENTION_SINGLE, MAX_BLOCK_ID},
-			{"MCP M2P", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			 NULL, BLOCK_CNIG},
+			{"MCP CPU", ATTENTION_SINGLE,
+			 qed_mcp_attn_cb, MAX_BLOCK_ID},
+			{"MCP Watchdog timer", ATTENTION_SINGLE,
+			 NULL, MAX_BLOCK_ID},
+			{"MCP M2P", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID},
 			{"AVS stop status ready", ATTENTION_SINGLE,
-			 MAX_BLOCK_ID},
-			{"MSTAT", ATTENTION_PAR_INT, MAX_BLOCK_ID},
-			{"MSTAT per-path", ATTENTION_PAR_INT, MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
+			{"MSTAT", ATTENTION_PAR_INT, NULL, MAX_BLOCK_ID},
+			{"MSTAT per-path", ATTENTION_PAR_INT,
+			 NULL, MAX_BLOCK_ID},
 			{"Reserved %d", (6 << ATTENTION_LENGTH_SHIFT),
-			 MAX_BLOCK_ID},
-			{"NIG", ATTENTION_PAR_INT, BLOCK_NIG},
-			{"BMB/OPTE/MCP", ATTENTION_PAR_INT, BLOCK_BMB},
-			{"BTB",	ATTENTION_PAR_INT, BLOCK_BTB},
-			{"BRB",	ATTENTION_PAR_INT, BLOCK_BRB},
-			{"PRS",	ATTENTION_PAR_INT, BLOCK_PRS},
+			 NULL, MAX_BLOCK_ID},
+			{"NIG", ATTENTION_PAR_INT, NULL, BLOCK_NIG},
+			{"BMB/OPTE/MCP", ATTENTION_PAR_INT, NULL, BLOCK_BMB},
+			{"BTB",	ATTENTION_PAR_INT, NULL, BLOCK_BTB},
+			{"BRB",	ATTENTION_PAR_INT, NULL, BLOCK_BRB},
+			{"PRS",	ATTENTION_PAR_INT, NULL, BLOCK_PRS},
 		}
 	},
 
 	{
 		{       /* After Invert 5 */
-			{"SRC", ATTENTION_PAR_INT, BLOCK_SRC},
-			{"PB Client1", ATTENTION_PAR_INT, BLOCK_PBF_PB1},
-			{"PB Client2", ATTENTION_PAR_INT, BLOCK_PBF_PB2},
-			{"RPB", ATTENTION_PAR_INT, BLOCK_RPB},
-			{"PBF", ATTENTION_PAR_INT, BLOCK_PBF},
-			{"QM", ATTENTION_PAR_INT, BLOCK_QM},
-			{"TM", ATTENTION_PAR_INT, BLOCK_TM},
-			{"MCM",  ATTENTION_PAR_INT, BLOCK_MCM},
-			{"MSDM", ATTENTION_PAR_INT, BLOCK_MSDM},
-			{"MSEM", ATTENTION_PAR_INT, BLOCK_MSEM},
-			{"PCM", ATTENTION_PAR_INT, BLOCK_PCM},
-			{"PSDM", ATTENTION_PAR_INT, BLOCK_PSDM},
-			{"PSEM", ATTENTION_PAR_INT, BLOCK_PSEM},
-			{"TCM", ATTENTION_PAR_INT, BLOCK_TCM},
-			{"TSDM", ATTENTION_PAR_INT, BLOCK_TSDM},
-			{"TSEM", ATTENTION_PAR_INT, BLOCK_TSEM},
+			{"SRC", ATTENTION_PAR_INT, NULL, BLOCK_SRC},
+			{"PB Client1", ATTENTION_PAR_INT, NULL, BLOCK_PBF_PB1},
+			{"PB Client2", ATTENTION_PAR_INT, NULL, BLOCK_PBF_PB2},
+			{"RPB", ATTENTION_PAR_INT, NULL, BLOCK_RPB},
+			{"PBF", ATTENTION_PAR_INT, NULL, BLOCK_PBF},
+			{"QM", ATTENTION_PAR_INT, NULL, BLOCK_QM},
+			{"TM", ATTENTION_PAR_INT, NULL, BLOCK_TM},
+			{"MCM",  ATTENTION_PAR_INT, NULL, BLOCK_MCM},
+			{"MSDM", ATTENTION_PAR_INT, NULL, BLOCK_MSDM},
+			{"MSEM", ATTENTION_PAR_INT, NULL, BLOCK_MSEM},
+			{"PCM", ATTENTION_PAR_INT, NULL, BLOCK_PCM},
+			{"PSDM", ATTENTION_PAR_INT, NULL, BLOCK_PSDM},
+			{"PSEM", ATTENTION_PAR_INT, NULL, BLOCK_PSEM},
+			{"TCM", ATTENTION_PAR_INT, NULL, BLOCK_TCM},
+			{"TSDM", ATTENTION_PAR_INT, NULL, BLOCK_TSDM},
+			{"TSEM", ATTENTION_PAR_INT, NULL, BLOCK_TSEM},
 		}
 	},
 
 	{
 		{       /* After Invert 6 */
-			{"UCM", ATTENTION_PAR_INT, BLOCK_UCM},
-			{"USDM", ATTENTION_PAR_INT, BLOCK_USDM},
-			{"USEM", ATTENTION_PAR_INT, BLOCK_USEM},
-			{"XCM",	ATTENTION_PAR_INT, BLOCK_XCM},
-			{"XSDM", ATTENTION_PAR_INT, BLOCK_XSDM},
-			{"XSEM", ATTENTION_PAR_INT, BLOCK_XSEM},
-			{"YCM",	ATTENTION_PAR_INT, BLOCK_YCM},
-			{"YSDM", ATTENTION_PAR_INT, BLOCK_YSDM},
-			{"YSEM", ATTENTION_PAR_INT, BLOCK_YSEM},
-			{"XYLD", ATTENTION_PAR_INT, BLOCK_XYLD},
-			{"TMLD", ATTENTION_PAR_INT, BLOCK_TMLD},
-			{"MYLD", ATTENTION_PAR_INT, BLOCK_MULD},
-			{"YULD", ATTENTION_PAR_INT, BLOCK_YULD},
-			{"DORQ", ATTENTION_PAR_INT, BLOCK_DORQ},
-			{"DBG", ATTENTION_PAR_INT, BLOCK_DBG},
-			{"IPC",	ATTENTION_PAR_INT, BLOCK_IPC},
+			{"UCM", ATTENTION_PAR_INT, NULL, BLOCK_UCM},
+			{"USDM", ATTENTION_PAR_INT, NULL, BLOCK_USDM},
+			{"USEM", ATTENTION_PAR_INT, NULL, BLOCK_USEM},
+			{"XCM",	ATTENTION_PAR_INT, NULL, BLOCK_XCM},
+			{"XSDM", ATTENTION_PAR_INT, NULL, BLOCK_XSDM},
+			{"XSEM", ATTENTION_PAR_INT, NULL, BLOCK_XSEM},
+			{"YCM",	ATTENTION_PAR_INT, NULL, BLOCK_YCM},
+			{"YSDM", ATTENTION_PAR_INT, NULL, BLOCK_YSDM},
+			{"YSEM", ATTENTION_PAR_INT, NULL, BLOCK_YSEM},
+			{"XYLD", ATTENTION_PAR_INT, NULL, BLOCK_XYLD},
+			{"TMLD", ATTENTION_PAR_INT, NULL, BLOCK_TMLD},
+			{"MYLD", ATTENTION_PAR_INT, NULL, BLOCK_MULD},
+			{"YULD", ATTENTION_PAR_INT, NULL, BLOCK_YULD},
+			{"DORQ", ATTENTION_PAR_INT,
+			 qed_dorq_attn_cb, BLOCK_DORQ},
+			{"DBG", ATTENTION_PAR_INT, NULL, BLOCK_DBG},
+			{"IPC",	ATTENTION_PAR_INT, NULL, BLOCK_IPC},
 		}
 	},
 
 	{
 		{       /* After Invert 7 */
-			{"CCFC", ATTENTION_PAR_INT, BLOCK_CCFC},
-			{"CDU", ATTENTION_PAR_INT, BLOCK_CDU},
-			{"DMAE", ATTENTION_PAR_INT, BLOCK_DMAE},
-			{"IGU", ATTENTION_PAR_INT, BLOCK_IGU},
-			{"ATC", ATTENTION_PAR_INT, MAX_BLOCK_ID},
-			{"CAU", ATTENTION_PAR_INT, BLOCK_CAU},
-			{"PTU", ATTENTION_PAR_INT, BLOCK_PTU},
-			{"PRM", ATTENTION_PAR_INT, BLOCK_PRM},
-			{"TCFC", ATTENTION_PAR_INT, BLOCK_TCFC},
-			{"RDIF", ATTENTION_PAR_INT, BLOCK_RDIF},
-			{"TDIF", ATTENTION_PAR_INT, BLOCK_TDIF},
-			{"RSS", ATTENTION_PAR_INT, BLOCK_RSS},
-			{"MISC", ATTENTION_PAR_INT, BLOCK_MISC},
-			{"MISCS", ATTENTION_PAR_INT, BLOCK_MISCS},
-			{"PCIE", ATTENTION_PAR, BLOCK_PCIE},
-			{"Vaux PCI core", ATTENTION_SINGLE, BLOCK_PGLCS},
-			{"PSWRQ", ATTENTION_PAR_INT, BLOCK_PSWRQ},
+			{"CCFC", ATTENTION_PAR_INT, NULL, BLOCK_CCFC},
+			{"CDU", ATTENTION_PAR_INT, NULL, BLOCK_CDU},
+			{"DMAE", ATTENTION_PAR_INT, NULL, BLOCK_DMAE},
+			{"IGU", ATTENTION_PAR_INT, NULL, BLOCK_IGU},
+			{"ATC", ATTENTION_PAR_INT, NULL, MAX_BLOCK_ID},
+			{"CAU", ATTENTION_PAR_INT, NULL, BLOCK_CAU},
+			{"PTU", ATTENTION_PAR_INT, NULL, BLOCK_PTU},
+			{"PRM", ATTENTION_PAR_INT, NULL, BLOCK_PRM},
+			{"TCFC", ATTENTION_PAR_INT, NULL, BLOCK_TCFC},
+			{"RDIF", ATTENTION_PAR_INT, NULL, BLOCK_RDIF},
+			{"TDIF", ATTENTION_PAR_INT, NULL, BLOCK_TDIF},
+			{"RSS", ATTENTION_PAR_INT, NULL, BLOCK_RSS},
+			{"MISC", ATTENTION_PAR_INT, NULL, BLOCK_MISC},
+			{"MISCS", ATTENTION_PAR_INT, NULL, BLOCK_MISCS},
+			{"PCIE", ATTENTION_PAR, NULL, BLOCK_PCIE},
+			{"Vaux PCI core", ATTENTION_SINGLE, NULL, BLOCK_PGLCS},
+			{"PSWRQ", ATTENTION_PAR_INT, NULL, BLOCK_PSWRQ},
 		}
 	},
 
 	{
 		{       /* After Invert 8 */
-			{"PSWRQ (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWRQ2},
-			{"PSWWR", ATTENTION_PAR_INT, BLOCK_PSWWR},
-			{"PSWWR (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWWR2},
-			{"PSWRD", ATTENTION_PAR_INT, BLOCK_PSWRD},
-			{"PSWRD (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWRD2},
-			{"PSWHST", ATTENTION_PAR_INT, BLOCK_PSWHST},
-			{"PSWHST (pci_clk)", ATTENTION_PAR_INT, BLOCK_PSWHST2},
-			{"GRC",	ATTENTION_PAR_INT, BLOCK_GRC},
-			{"CPMU", ATTENTION_PAR_INT, BLOCK_CPMU},
-			{"NCSI", ATTENTION_PAR_INT, BLOCK_NCSI},
-			{"MSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
-			{"PSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
-			{"TSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
-			{"USEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
-			{"XSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
-			{"YSEM PRAM", ATTENTION_PAR, MAX_BLOCK_ID},
-			{"pxp_misc_mps", ATTENTION_PAR, BLOCK_PGLCS},
+			{"PSWRQ (pci_clk)", ATTENTION_PAR_INT,
+			 NULL, BLOCK_PSWRQ2},
+			{"PSWWR", ATTENTION_PAR_INT, NULL, BLOCK_PSWWR},
+			{"PSWWR (pci_clk)", ATTENTION_PAR_INT,
+			 NULL, BLOCK_PSWWR2},
+			{"PSWRD", ATTENTION_PAR_INT, NULL, BLOCK_PSWRD},
+			{"PSWRD (pci_clk)", ATTENTION_PAR_INT,
+			 NULL, BLOCK_PSWRD2},
+			{"PSWHST", ATTENTION_PAR_INT,
+			 qed_pswhst_attn_cb, BLOCK_PSWHST},
+			{"PSWHST (pci_clk)", ATTENTION_PAR_INT,
+			 NULL, BLOCK_PSWHST2},
+			{"GRC",	ATTENTION_PAR_INT,
+			 qed_grc_attn_cb, BLOCK_GRC},
+			{"CPMU", ATTENTION_PAR_INT, NULL, BLOCK_CPMU},
+			{"NCSI", ATTENTION_PAR_INT, NULL, BLOCK_NCSI},
+			{"MSEM PRAM", ATTENTION_PAR, NULL, MAX_BLOCK_ID},
+			{"PSEM PRAM", ATTENTION_PAR, NULL, MAX_BLOCK_ID},
+			{"TSEM PRAM", ATTENTION_PAR, NULL, MAX_BLOCK_ID},
+			{"USEM PRAM", ATTENTION_PAR, NULL, MAX_BLOCK_ID},
+			{"XSEM PRAM", ATTENTION_PAR, NULL, MAX_BLOCK_ID},
+			{"YSEM PRAM", ATTENTION_PAR, NULL, MAX_BLOCK_ID},
+			{"pxp_misc_mps", ATTENTION_PAR, NULL, BLOCK_PGLCS},
 			{"PCIE glue/PXP Exp. ROM", ATTENTION_SINGLE,
-			 BLOCK_PGLCS},
-			{"PERST_B assertion", ATTENTION_SINGLE, MAX_BLOCK_ID},
+			 NULL, BLOCK_PGLCS},
+			{"PERST_B assertion", ATTENTION_SINGLE,
+			 NULL, MAX_BLOCK_ID},
 			{"PERST_B deassertion", ATTENTION_SINGLE,
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 			{"Reserved %d", (2 << ATTENTION_LENGTH_SHIFT),
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 		}
 	},
 
 	{
 		{       /* After Invert 9 */
-			{"MCP Latched memory", ATTENTION_PAR, MAX_BLOCK_ID},
+			{"MCP Latched memory", ATTENTION_PAR,
+			 NULL, MAX_BLOCK_ID},
 			{"MCP Latched scratchpad cache", ATTENTION_SINGLE,
-			 MAX_BLOCK_ID},
-			{"MCP Latched ump_tx", ATTENTION_PAR, MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
+			{"MCP Latched ump_tx", ATTENTION_PAR,
+			 NULL, MAX_BLOCK_ID},
 			{"MCP Latched scratchpad", ATTENTION_PAR,
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 			{"Reserved %d", (28 << ATTENTION_LENGTH_SHIFT),
-			 MAX_BLOCK_ID},
+			 NULL, MAX_BLOCK_ID},
 		}
 	},
 };
@@ -1585,14 +1881,22 @@ qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn,
 			    u32 bitmask)
 {
 	int rc = -EINVAL;
-	u32 val, mask = ~bitmask;
+	u32 val;
 
 	DP_INFO(p_hwfn, "Deasserted attention `%s'[%08x]\n",
 		p_aeu->bit_name, bitmask);
 
+	/* Call callback before clearing the interrupt status */
+	if (p_aeu->cb) {
+		DP_INFO(p_hwfn, "`%s (attention)': Calling Callback function\n",
+			p_aeu->bit_name);
+		rc = p_aeu->cb(p_hwfn);
+	}
+
 	/* Handle HW block interrupt registers */
 	if (p_aeu->block_index != MAX_BLOCK_ID) {
 		struct attn_hw_block *p_block;
+		u32 mask;
 		int i;
 
 		p_block = &attn_blocks[p_aeu->block_index];
@@ -1603,7 +1907,14 @@ qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn,
 			u32 sts_addr;
 
 			p_reg_desc = p_block->chip_regs[0].int_regs[i];
-			sts_addr = p_reg_desc->sts_addr;
+
+			/* In case of fatal attention, don't clear the status
+			 * so it would appear in following idle check.
+			 */
+			if (rc == 0)
+				sts_addr = p_reg_desc->sts_clr_addr;
+			else
+				sts_addr = p_reg_desc->sts_addr;
 
 			val = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, sts_addr);
 			mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt,
@@ -1615,12 +1926,17 @@ qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn,
 		}
 	}
 
+	/* If the attention is benign, no need to prevent it */
+	if (!rc)
+		goto out;
+
 	/* Prevent this Attention from being asserted in the future */
 	val = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg);
-	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg, (val & mask));
+	qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, aeu_en_reg, (val & ~bitmask));
 	DP_INFO(p_hwfn, "`%s' - Disabled future attentions\n",
 		p_aeu->bit_name);
 
+out:
 	return rc;
 }
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
index 2286e75..c15b162 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
@@ -127,8 +127,20 @@
 	0x00c000UL
 #define  DORQ_REG_IFEN \
 	0x100040UL
+#define DORQ_REG_DB_DROP_REASON \
+	0x100a2cUL
+#define DORQ_REG_DB_DROP_DETAILS \
+	0x100a24UL
+#define DORQ_REG_DB_DROP_DETAILS_ADDRESS \
+	0x100a1cUL
 #define  GRC_REG_TIMEOUT_EN \
 	0x050404UL
+#define GRC_REG_TIMEOUT_ATTN_ACCESS_VALID \
+	0x050054UL
+#define GRC_REG_TIMEOUT_ATTN_ACCESS_DATA_0 \
+	0x05004cUL
+#define GRC_REG_TIMEOUT_ATTN_ACCESS_DATA_1 \
+	0x050050UL
 #define  IGU_REG_BLOCK_CONFIGURATION \
 	0x180040UL
 #define  MCM_REG_INIT \
@@ -155,6 +167,40 @@
 	0x1100000UL
 #define  PGLUE_B_REG_ADMIN_PER_PF_REGION	\
 	0x2a9000UL
+#define PGLUE_B_REG_TX_ERR_WR_DETAILS2 \
+	0x2aa150UL
+#define PGLUE_B_REG_TX_ERR_WR_ADD_31_0 \
+	0x2aa144UL
+#define PGLUE_B_REG_TX_ERR_WR_ADD_63_32 \
+	0x2aa148UL
+#define PGLUE_B_REG_TX_ERR_WR_DETAILS \
+	0x2aa14cUL
+#define PGLUE_B_REG_TX_ERR_RD_ADD_31_0 \
+	0x2aa154UL
+#define PGLUE_B_REG_TX_ERR_RD_ADD_63_32 \
+	0x2aa158UL
+#define PGLUE_B_REG_TX_ERR_RD_DETAILS \
+	0x2aa15cUL
+#define PGLUE_B_REG_TX_ERR_RD_DETAILS2 \
+	0x2aa160UL
+#define PGLUE_B_REG_TX_ERR_WR_DETAILS_ICPL \
+	0x2aa164UL
+#define PGLUE_B_REG_MASTER_ZLR_ERR_DETAILS \
+	0x2aa54cUL
+#define PGLUE_B_REG_MASTER_ZLR_ERR_ADD_31_0 \
+	0x2aa544UL
+#define PGLUE_B_REG_MASTER_ZLR_ERR_ADD_63_32 \
+	0x2aa548UL
+#define PGLUE_B_REG_VF_ILT_ERR_ADD_31_0 \
+	0x2aae74UL
+#define PGLUE_B_REG_VF_ILT_ERR_ADD_63_32 \
+	0x2aae78UL
+#define PGLUE_B_REG_VF_ILT_ERR_DETAILS \
+	0x2aae7cUL
+#define PGLUE_B_REG_VF_ILT_ERR_DETAILS2 \
+	0x2aae80UL
+#define PGLUE_B_REG_LATCHED_ERRORS_CLR \
+	0x2aa3bcUL
 #define  PRM_REG_DISABLE_PRM \
 	0x230000UL
 #define  PRS_REG_SOFT_RST \
@@ -171,6 +217,14 @@
 	0x2a0040UL
 #define  PSWHST2_REG_DBGSYN_ALMOST_FULL_THR \
 	0x29e050UL
+#define PSWHST_REG_INCORRECT_ACCESS_VALID \
+	0x2a0070UL
+#define PSWHST_REG_INCORRECT_ACCESS_ADDRESS \
+	0x2a0074UL
+#define PSWHST_REG_INCORRECT_ACCESS_DATA \
+	0x2a0068UL
+#define PSWHST_REG_INCORRECT_ACCESS_LENGTH \
+	0x2a006cUL
 #define  PSWRD_REG_DBG_SELECT \
 	0x29c040UL
 #define  PSWRD2_REG_CONF11 \
@@ -365,6 +419,10 @@
 		0x7 << 0)
 #define  MCP_REG_NVM_CFG4_FLASH_SIZE_SHIFT \
 	0
+#define MCP_REG_CPU_STATE \
+	0xe05004UL
+#define MCP_REG_CPU_EVENT_MASK \
+	0xe05008UL
 #define PGLUE_B_REG_PF_BAR0_SIZE \
 	0x2aae60UL
 #define PGLUE_B_REG_PF_BAR1_SIZE \
-- 
cgit v0.10.2


From 2fca6d288d7cac2debf170da021780cd8265c645 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 28 Feb 2016 21:56:34 +0300
Subject: of_mdio: kill useless variable in of_mdiobus_register()

of_mdiobus_register()  declares the 'paddr' variable to hold the result of
the of_get_property()  but only uses it once after that while the function
can be called directly from the *if* statement. Remove that variable and
switch to calling of_find_property() instead since  we don't care about
the "reg" property's value anyway...

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 39c4be4..669739b 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -211,7 +211,6 @@ static bool of_mdiobus_child_is_phy(struct device_node *child)
 int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 {
 	struct device_node *child;
-	const __be32 *paddr;
 	bool scanphys = false;
 	int addr, rc;
 
@@ -246,8 +245,7 @@ int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 	/* auto scan for PHYs with empty reg property */
 	for_each_available_child_of_node(np, child) {
 		/* Skip PHYs with reg property set */
-		paddr = of_get_property(child, "reg", NULL);
-		if (paddr)
+		if (of_find_property(child, "reg", NULL))
 			continue;
 
 		for (addr = 0; addr < PHY_MAX_ADDR; addr++) {
-- 
cgit v0.10.2


From 241deec94425b5e962e2bf8a929d563afda045f7 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 1 Mar 2016 17:44:59 -0500
Subject: sch_mqprio: Fix build with older gcc.

  CC [M]  net/sched/sch_mqprio.o
net/sched/sch_mqprio.c: In function ?mqprio_init?:
net/sched/sch_mqprio.c:145: error: unknown field ?tc? specified in initializer
net/sched/sch_mqprio.c:145: warning: missing braces around initializer
net/sched/sch_mqprio.c:145: warning: (near initialization for ?tc.<anonymous>?)
make[2]: *** [net/sched/sch_mqprio.o] Error 1
make[1]: *** [net/sched] Error 2
make: *** [net] Error 2

Several people reported this, surround the unnamed union
member initialization with braces to fix.

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index f9947d1..02ffb3f 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -142,7 +142,7 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt)
 	 */
 	if (qopt->hw) {
 		struct tc_to_netdev tc = {.type = TC_SETUP_MQPRIO,
-					  .tc = qopt->num_tc};
+					  { .tc = qopt->num_tc }};
 
 		priv->hw_owned = 1;
 		err = dev->netdev_ops->ndo_setup_tc(dev, sch->handle, 0, &tc);
-- 
cgit v0.10.2


From c145aeb3ff899f38bd28ccbc40dbb647dd1b1fd3 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Mon, 29 Feb 2016 08:21:30 +0000
Subject: net: pktgen: use reset to set mac header

Since offset is zero, it's not necessary to use set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 1474cfd..20999aa 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2856,7 +2856,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 		*vlan_encapsulated_proto = htons(ETH_P_IP);
 	}
 
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, skb->len);
 	iph = (struct iphdr *) skb_put(skb, sizeof(struct iphdr));
 
@@ -2983,7 +2983,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 		*vlan_encapsulated_proto = htons(ETH_P_IPV6);
 	}
 
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, skb->len);
 	iph = (struct ipv6hdr *) skb_put(skb, sizeof(struct ipv6hdr));
 
-- 
cgit v0.10.2


From fcb6b92a682fe5032fdc31af7f8ed86f1dabb1e2 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Mon, 22 Feb 2016 10:21:41 +0200
Subject: iwlwifi: mvm: update ucode status before stopping device

Leaving ucode_loaded to true after stop_device() has been called
is a recipe for problems. Flows that are not sync'ed with the
driver life cycle (like debugfs hooks and thermal hooks) must
check that the firmware is loaded before they interact with it.
Therefore we need to keep this variable updated with the real
status of the firmware.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index cf5e634..c1a3131 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -816,8 +816,7 @@ static int iwl_mvm_switch_to_d3(struct iwl_mvm *mvm)
 {
 	iwl_mvm_scan_stop(mvm, IWL_MVM_SCAN_REGULAR, true);
 
-	iwl_trans_stop_device(mvm->trans);
-
+	iwl_mvm_stop_device(mvm);
 	/*
 	 * Set the HW restart bit -- this is mostly true as we're
 	 * going to load new firmware and reprogram that, though
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index f075c36..efec6ec 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1009,7 +1009,7 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	IWL_DEBUG_INFO(mvm, "RT uCode started.\n");
 	return 0;
  error:
-	iwl_trans_stop_device(mvm->trans);
+	iwl_mvm_stop_device(mvm);
 	return ret;
 }
 
@@ -1053,7 +1053,7 @@ int iwl_mvm_load_d3_fw(struct iwl_mvm *mvm)
 
 	return 0;
  error:
-	iwl_trans_stop_device(mvm->trans);
+	iwl_mvm_stop_device(mvm);
 	return ret;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 927b055..ec6b072 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -972,7 +972,7 @@ static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm)
 	 */
 	iwl_mvm_unref_all_except(mvm, IWL_MVM_REF_UCODE_DOWN);
 
-	iwl_trans_stop_device(mvm->trans);
+	iwl_mvm_stop_device(mvm);
 
 	mvm->scan_status = 0;
 	mvm->ps_disabled = false;
@@ -1141,7 +1141,7 @@ void __iwl_mvm_mac_stop(struct iwl_mvm *mvm)
 	 */
 	flush_work(&mvm->roc_done_wk);
 
-	iwl_trans_stop_device(mvm->trans);
+	iwl_mvm_stop_device(mvm);
 
 	iwl_mvm_async_handlers_purge(mvm);
 	/* async_handlers_list is empty and will stay empty: HW is stopped */
@@ -1172,8 +1172,6 @@ void __iwl_mvm_mac_stop(struct iwl_mvm *mvm)
 				mvm->scan_uid_status[i] = 0;
 		}
 	}
-
-	mvm->ucode_loaded = false;
 }
 
 static void iwl_mvm_mac_stop(struct ieee80211_hw *hw)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index b461b90..6d4e487 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1549,6 +1549,12 @@ void iwl_mvm_enable_ac_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue,
 	iwl_mvm_enable_txq(mvm, queue, mac80211_queue, ssn, &cfg, wdg_timeout);
 }
 
+static inline void iwl_mvm_stop_device(struct iwl_mvm *mvm)
+{
+	mvm->ucode_loaded = false;
+	iwl_trans_stop_device(mvm->trans);
+}
+
 /* Thermal management and CT-kill */
 void iwl_mvm_tt_tx_backoff(struct iwl_mvm *mvm, u32 backoff);
 void iwl_mvm_tt_temp_changed(struct iwl_mvm *mvm, u32 temp);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 8bdaa7b..4716176 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -643,7 +643,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		iwl_mvm_ref(mvm, IWL_MVM_REF_INIT_UCODE);
 		err = iwl_run_init_mvm_ucode(mvm, true);
 		if (!err || !iwlmvm_mod_params.init_dbg)
-			iwl_trans_stop_device(trans);
+			iwl_mvm_stop_device(mvm);
 		iwl_mvm_unref(mvm, IWL_MVM_REF_INIT_UCODE);
 		mutex_unlock(&mvm->mutex);
 		/* returns 0 if successful, 1 if success but in rfkill */
-- 
cgit v0.10.2


From 5eae443eb5e2b3777582ea37c6a002171ec134d5 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 24 Feb 2016 14:56:21 +0200
Subject: iwlwifi: pcie: detect and workaround invalid write ptr behavior

In 9000 series A0 step the closed_rb_num is not wrapping around
properly. The queue is wrapping around as it should, so we can
W/A it by wrapping the closed_rb_num in the driver.
While at it, extend RX logging and add error handling of other
cases HW values may cause us to access invalid memory locations.
Add also a proper masking of vid value read from HW - this should
not have actual affect, but better to be on the safe side.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 398dd93..489b07a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -1159,9 +1159,12 @@ restart:
 	r = le16_to_cpu(ACCESS_ONCE(rxq->rb_stts->closed_rb_num)) & 0x0FFF;
 	i = rxq->read;
 
+	/* W/A 9000 device step A0 wrap-around bug */
+	r &= (rxq->queue_size - 1);
+
 	/* Rx interrupt, but nothing sent from uCode */
 	if (i == r)
-		IWL_DEBUG_RX(trans, "HW = SW = %d\n", r);
+		IWL_DEBUG_RX(trans, "Q %d: HW = SW = %d\n", rxq->id, r);
 
 	while (i != r) {
 		struct iwl_rx_mem_buffer *rxb;
@@ -1174,15 +1177,18 @@ restart:
 			 * used_bd is a 32 bit but only 12 are used to retrieve
 			 * the vid
 			 */
-			u16 vid = (u16)le32_to_cpu(rxq->used_bd[i]);
+			u16 vid = le32_to_cpu(rxq->used_bd[i]) & 0x0FFF;
 
+			if (WARN(vid >= ARRAY_SIZE(trans_pcie->global_table),
+				 "Invalid rxb index from HW %u\n", (u32)vid))
+				goto out;
 			rxb = trans_pcie->global_table[vid];
 		} else {
 			rxb = rxq->queue[i];
 			rxq->queue[i] = NULL;
 		}
 
-		IWL_DEBUG_RX(trans, "rxbuf: HW = %d, SW = %d\n", r, i);
+		IWL_DEBUG_RX(trans, "Q %d: HW = %d, SW = %d\n", rxq->id, r, i);
 		iwl_pcie_rx_handle_rb(trans, rxq, rxb, emergency);
 
 		i = (i + 1) & (rxq->queue_size - 1);
@@ -1245,7 +1251,7 @@ restart:
 			goto restart;
 		}
 	}
-
+out:
 	/* Backtrack one entry */
 	rxq->read = i;
 	spin_unlock(&rxq->lock);
@@ -1301,6 +1307,9 @@ irqreturn_t iwl_pcie_irq_rx_msix_handler(int irq, void *dev_id)
 	struct iwl_trans_pcie *trans_pcie = iwl_pcie_get_trans_pcie(entry);
 	struct iwl_trans *trans = trans_pcie->trans;
 
+	if (WARN_ON(entry->entry >= trans->num_rx_queues))
+		return IRQ_NONE;
+
 	lock_map_acquire(&trans->sync_cmd_lockdep_map);
 
 	local_bh_disable();
-- 
cgit v0.10.2


From 280452c9056421364111930b6336e27d85240438 Mon Sep 17 00:00:00 2001
From: Liad Kaufman <liad.kaufman@intel.com>
Date: Tue, 23 Feb 2016 16:23:38 +0200
Subject: iwlwifi: mvm: disable DQA support

Do not allow entrance into DQA flows until feature is
completely ready and merged.

Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 6d4e487..7ea2baa 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -986,8 +986,9 @@ static inline bool iwl_mvm_is_d0i3_supported(struct iwl_mvm *mvm)
 
 static inline bool iwl_mvm_is_dqa_supported(struct iwl_mvm *mvm)
 {
-	return fw_has_capa(&mvm->fw->ucode_capa,
-			   IWL_UCODE_TLV_CAPA_DQA_SUPPORT);
+	/* Make sure DQA isn't allowed in driver until feature is complete */
+	return false && fw_has_capa(&mvm->fw->ucode_capa,
+				    IWL_UCODE_TLV_CAPA_DQA_SUPPORT);
 }
 
 static inline bool iwl_mvm_enter_d0i3_on_suspend(struct iwl_mvm *mvm)
-- 
cgit v0.10.2


From 33c85ead7115ad0a1c3c47d5d1d8f8474775d644 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Mon, 22 Feb 2016 15:44:13 +0200
Subject: iwlwifi: mvm: only release the trans ref if d0i3 is supported in fw

If d0i3 is not supported by the firmware (or if it's disabled via
module parameters) we shouldn't release the initial transport
reference, so that we won't enter runtime suspend.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 4716176..f3283f4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -672,8 +672,10 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 
 	memset(&mvm->rx_stats, 0, sizeof(struct mvm_statistics_rx));
 
-	/* rpm starts with a taken reference, we can release it now */
-	iwl_trans_unref(mvm->trans);
+	/* The transport always starts with a taken reference, we can
+	 * release it now if d0i3 is supported */
+	if (iwl_mvm_is_d0i3_supported(mvm))
+		iwl_trans_unref(mvm->trans);
 
 	iwl_mvm_tof_init(mvm);
 
-- 
cgit v0.10.2


From 0d0985adf6519952e0d51b5272abc3ef28d833bc Mon Sep 17 00:00:00 2001
From: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Date: Mon, 29 Feb 2016 13:25:48 +0200
Subject: iwlwifi: add disable_11ac module param

Add module parameter that disables VHT capabilities.

Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index ee4ffa5..184c0fe 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -1719,3 +1719,7 @@ MODULE_PARM_DESC(fw_monitor,
 module_param_named(d0i3_timeout, iwlwifi_mod_params.d0i3_entry_delay,
 		   uint, S_IRUGO);
 MODULE_PARM_DESC(d0i3_timeout, "Timeout to D0i3 entry when idle (ms)");
+
+module_param_named(disable_11ac, iwlwifi_mod_params.disable_11ac, bool,
+		   S_IRUGO);
+MODULE_PARM_DESC(disable_11ac, "Disable VHT capabilities");
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
index b88ecc7..d1a5dd1 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
@@ -115,6 +115,7 @@ enum iwl_amsdu_size {
  *	entering D0i3 (in msecs)
  * @lar_disable: disable LAR (regulatory), default = 0
  * @fw_monitor: allow to use firmware monitor
+ * @disable_11ac: disable VHT capabilities, default = false.
  */
 struct iwl_mod_params {
 	int sw_crypto;
@@ -135,6 +136,7 @@ struct iwl_mod_params {
 	unsigned int d0i3_entry_delay;
 	bool lar_disable;
 	bool fw_monitor;
+	bool disable_11ac;
 };
 
 #endif /* #__iwl_modparams_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index e84cb8d..3481357 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -452,7 +452,7 @@ static void iwl_init_sbands(struct device *dev, const struct iwl_cfg *cfg,
 					  IEEE80211_BAND_5GHZ);
 	iwl_init_ht_hw_capab(cfg, data, &sband->ht_cap, IEEE80211_BAND_5GHZ,
 			     tx_chains, rx_chains);
-	if (data->sku_cap_11ac_enable)
+	if (data->sku_cap_11ac_enable && !iwlwifi_mod_params.disable_11ac)
 		iwl_init_vht_hw_capab(cfg, data, &sband->vht_cap,
 				      tx_chains, rx_chains);
 
-- 
cgit v0.10.2


From e27deb4583642d6c5b3cf18060e8239db1be9e59 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Tue, 1 Mar 2016 10:30:48 +0200
Subject: iwlwifi: mvm: take the transport ref back when leaving

If d0i3 is supported, we have released the initial transport reference
in iwl_op_mode_mvm_start(), so we should take it back in
iwl_op_mode_mvm_stop() to keep it balanced.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index f3283f4..3760a09 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -703,6 +703,13 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode)
 	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
 	int i;
 
+	/* If d0i3 is supported, we have released the reference that
+	 * the transport started with, so we should take it back now
+	 * that we are leaving.
+	 */
+	if (iwl_mvm_is_d0i3_supported(mvm))
+		iwl_trans_ref(mvm->trans);
+
 	iwl_mvm_leds_exit(mvm);
 
 	iwl_mvm_thermal_exit(mvm);
-- 
cgit v0.10.2


From b4f7a9d168d9f8204abf06d244a7938dd62b87b0 Mon Sep 17 00:00:00 2001
From: Liad Kaufman <liad.kaufman@intel.com>
Date: Wed, 3 Feb 2016 11:05:41 +0200
Subject: iwlwifi: mvm: support sw queue start/stop from mvm

Add a wrapper function to allow stopping SW queues from MVM
as well.

Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 7ea2baa..f9019c4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1556,6 +1556,10 @@ static inline void iwl_mvm_stop_device(struct iwl_mvm *mvm)
 	iwl_trans_stop_device(mvm->trans);
 }
 
+/* Stop/start all mac queues in a given bitmap */
+void iwl_mvm_start_mac_queues(struct iwl_mvm *mvm, unsigned long mq);
+void iwl_mvm_stop_mac_queues(struct iwl_mvm *mvm, unsigned long mq);
+
 /* Thermal management and CT-kill */
 void iwl_mvm_tt_tx_backoff(struct iwl_mvm *mvm, u32 backoff);
 void iwl_mvm_tt_temp_changed(struct iwl_mvm *mvm, u32 temp);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 3760a09..e6d0adc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -897,24 +897,17 @@ static void iwl_mvm_rx_mq(struct iwl_op_mode *op_mode,
 		iwl_mvm_rx_common(mvm, rxb, pkt);
 }
 
-static void iwl_mvm_stop_sw_queue(struct iwl_op_mode *op_mode, int queue)
+void iwl_mvm_stop_mac_queues(struct iwl_mvm *mvm, unsigned long mq)
 {
-	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
-	unsigned long mq;
 	int q;
 
-	spin_lock_bh(&mvm->queue_info_lock);
-	mq = mvm->queue_info[queue].hw_queue_to_mac80211;
-	spin_unlock_bh(&mvm->queue_info_lock);
-
 	if (WARN_ON_ONCE(!mq))
 		return;
 
 	for_each_set_bit(q, &mq, IEEE80211_MAX_QUEUES) {
 		if (atomic_inc_return(&mvm->mac80211_queue_stop_count[q]) > 1) {
 			IWL_DEBUG_TX_QUEUES(mvm,
-					    "queue %d (mac80211 %d) already stopped\n",
-					    queue, q);
+					    "mac80211 %d already stopped\n", q);
 			continue;
 		}
 
@@ -934,24 +927,29 @@ static void iwl_mvm_async_cb(struct iwl_op_mode *op_mode,
 	iwl_trans_block_txq_ptrs(mvm->trans, false);
 }
 
-static void iwl_mvm_wake_sw_queue(struct iwl_op_mode *op_mode, int queue)
+static void iwl_mvm_stop_sw_queue(struct iwl_op_mode *op_mode, int hw_queue)
 {
 	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
 	unsigned long mq;
-	int q;
 
 	spin_lock_bh(&mvm->queue_info_lock);
-	mq = mvm->queue_info[queue].hw_queue_to_mac80211;
+	mq = mvm->queue_info[hw_queue].hw_queue_to_mac80211;
 	spin_unlock_bh(&mvm->queue_info_lock);
 
+	iwl_mvm_stop_mac_queues(mvm, mq);
+}
+
+void iwl_mvm_start_mac_queues(struct iwl_mvm *mvm, unsigned long mq)
+{
+	int q;
+
 	if (WARN_ON_ONCE(!mq))
 		return;
 
 	for_each_set_bit(q, &mq, IEEE80211_MAX_QUEUES) {
 		if (atomic_dec_return(&mvm->mac80211_queue_stop_count[q]) > 0) {
 			IWL_DEBUG_TX_QUEUES(mvm,
-					    "queue %d (mac80211 %d) still stopped\n",
-					    queue, q);
+					    "mac80211 %d still stopped\n", q);
 			continue;
 		}
 
@@ -959,6 +957,18 @@ static void iwl_mvm_wake_sw_queue(struct iwl_op_mode *op_mode, int queue)
 	}
 }
 
+static void iwl_mvm_wake_sw_queue(struct iwl_op_mode *op_mode, int hw_queue)
+{
+	struct iwl_mvm *mvm = IWL_OP_MODE_GET_MVM(op_mode);
+	unsigned long mq;
+
+	spin_lock_bh(&mvm->queue_info_lock);
+	mq = mvm->queue_info[hw_queue].hw_queue_to_mac80211;
+	spin_unlock_bh(&mvm->queue_info_lock);
+
+	iwl_mvm_start_mac_queues(mvm, mq);
+}
+
 void iwl_mvm_set_hw_ctkill_state(struct iwl_mvm *mvm, bool state)
 {
 	if (state)
-- 
cgit v0.10.2


From 7cb43bebab52d8a9aa5943fbd1babe633a255e2c Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Sat, 27 Feb 2016 22:34:16 +0530
Subject: wan: lmc: Switch to using managed resources

Use managed resource functions devm_kzalloc and pcim_enable_device
to simplify error handling. Subsequently, remove unnecessary
kfree, pci_disable_device and pci_release_regions.

To be compatible with the change, various gotos are replaced with
direct returns and unneeded labels are dropped.

Also, `sc` was only being freed in the probe function and not the
remove function before the change. By using devm_kzalloc this patch
also fixes this memory leak.

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/wan/lmc/lmc_main.c b/drivers/net/wan/lmc/lmc_main.c
index 317bc79..bb33b24 100644
--- a/drivers/net/wan/lmc/lmc_main.c
+++ b/drivers/net/wan/lmc/lmc_main.c
@@ -826,7 +826,7 @@ static int lmc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* lmc_trace(dev, "lmc_init_one in"); */
 
-	err = pci_enable_device(pdev);
+	err = pcim_enable_device(pdev);
 	if (err) {
 		printk(KERN_ERR "lmc: pci enable failed: %d\n", err);
 		return err;
@@ -835,23 +835,20 @@ static int lmc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	err = pci_request_regions(pdev, "lmc");
 	if (err) {
 		printk(KERN_ERR "lmc: pci_request_region failed\n");
-		goto err_req_io;
+		return err;
 	}
 
 	/*
 	 * Allocate our own device structure
 	 */
-	sc = kzalloc(sizeof(lmc_softc_t), GFP_KERNEL);
-	if (!sc) {
-		err = -ENOMEM;
-		goto err_kzalloc;
-	}
+	sc = devm_kzalloc(&pdev->dev, sizeof(lmc_softc_t), GFP_KERNEL);
+	if (!sc)
+		return -ENOMEM;
 
 	dev = alloc_hdlcdev(sc);
 	if (!dev) {
 		printk(KERN_ERR "lmc:alloc_netdev for device failed\n");
-		err = -ENOMEM;
-		goto err_hdlcdev;
+		return -ENOMEM;
 	}
 
 
@@ -888,7 +885,7 @@ static int lmc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err) {
 		printk(KERN_ERR "%s: register_netdev failed.\n", dev->name);
 		free_netdev(dev);
-		goto err_hdlcdev;
+		return err;
 	}
 
     sc->lmc_cardtype = LMC_CARDTYPE_UNKNOWN;
@@ -971,14 +968,6 @@ static int lmc_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
     lmc_trace(dev, "lmc_init_one out");
     return 0;
-
-err_hdlcdev:
-	kfree(sc);
-err_kzalloc:
-	pci_release_regions(pdev);
-err_req_io:
-	pci_disable_device(pdev);
-	return err;
 }
 
 /*
@@ -992,8 +981,6 @@ static void lmc_remove_one(struct pci_dev *pdev)
 		printk(KERN_DEBUG "%s: removing...\n", dev->name);
 		unregister_hdlc_device(dev);
 		free_netdev(dev);
-		pci_release_regions(pdev);
-		pci_disable_device(pdev);
 	}
 }
 
-- 
cgit v0.10.2


From fb653ebddcf871e8a83a63c1e4e73d41bc3cf599 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 2 Mar 2016 14:54:35 +0100
Subject: batman-adv: clarify CFG80211 dependency

The driver calls cfg80211_get_station, which may be part of a
module, so we must not enable BATMAN_ADV_BATMAN_V if
BATMAN_ADV=y and CFG80211=m:

net/built-in.o: In function `batadv_v_elp_get_throughput':
(text+0x5c62c): undefined reference to `cfg80211_get_station'

This clarifies the dependency to cover all combinations.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: c833484e5f38 ("batman-adv: ELP - compute the metric based on the estimated throughput")
Acked-by: Antonio Quartulli <a@unstable.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index e651dc9..f66930e 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -17,7 +17,7 @@ config BATMAN_ADV
 
 config BATMAN_ADV_BATMAN_V
 	bool "B.A.T.M.A.N. V protocol (experimental)"
-	depends on BATMAN_ADV && CFG80211
+	depends on BATMAN_ADV && CFG80211=y || (CFG80211=m && BATMAN_ADV=m)
 	default n
 	help
 	  This option enables the B.A.T.M.A.N. V protocol, the successor
-- 
cgit v0.10.2


From c2035eea3e9e1df768285fc8d72808469687aec0 Mon Sep 17 00:00:00 2001
From: Ram Amrani <Ram.Amrani@qlogic.com>
Date: Wed, 2 Mar 2016 20:26:00 +0200
Subject: qed: Correct BAR sizes for older MFW

Driver learns the inner bar sized from a register configured by management
firmware, but older versions are not setting this register.
But since we know which values were configured back then, use them instead.

Signed-off-by: Ram Amrani <Ram.Amrani@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index d9a5175..706c614 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -32,6 +32,33 @@
 #include "qed_sp.h"
 
 /* API common to all protocols */
+enum BAR_ID {
+	BAR_ID_0,       /* used for GRC */
+	BAR_ID_1        /* Used for doorbells */
+};
+
+static u32 qed_hw_bar_size(struct qed_hwfn	*p_hwfn,
+			   enum BAR_ID		bar_id)
+{
+	u32	bar_reg = (bar_id == BAR_ID_0 ?
+			   PGLUE_B_REG_PF_BAR0_SIZE : PGLUE_B_REG_PF_BAR1_SIZE);
+	u32	val = qed_rd(p_hwfn, p_hwfn->p_main_ptt, bar_reg);
+
+	if (val)
+		return 1 << (val + 15);
+
+	/* Old MFW initialized above registered only conditionally */
+	if (p_hwfn->cdev->num_hwfns > 1) {
+		DP_INFO(p_hwfn,
+			"BAR size not configured. Assuming BAR size of 256kB for GRC and 512kB for DB\n");
+			return BAR_ID_0 ? 256 * 1024 : 512 * 1024;
+	} else {
+		DP_INFO(p_hwfn,
+			"BAR size not configured. Assuming BAR size of 512kB for GRC and 512kB for DB\n");
+			return 512 * 1024;
+	}
+}
+
 void qed_init_dp(struct qed_dev *cdev,
 		 u32 dp_module, u8 dp_level)
 {
@@ -1385,17 +1412,6 @@ err0:
 	return rc;
 }
 
-static u32 qed_hw_bar_size(struct qed_hwfn	*p_hwfn,
-			   u8			bar_id)
-{
-	u32 bar_reg = (bar_id == 0 ? PGLUE_B_REG_PF_BAR0_SIZE
-		       : PGLUE_B_REG_PF_BAR1_SIZE);
-	u32 val = qed_rd(p_hwfn, p_hwfn->p_main_ptt, bar_reg);
-
-	/* Get the BAR size(in KB) from hardware given val */
-	return 1 << (val + 15);
-}
-
 int qed_hw_prepare(struct qed_dev *cdev,
 		   int personality)
 {
@@ -1420,11 +1436,11 @@ int qed_hw_prepare(struct qed_dev *cdev,
 		u8 __iomem *addr;
 
 		/* adjust bar offset for second engine */
-		addr = cdev->regview + qed_hw_bar_size(p_hwfn, 0) / 2;
+		addr = cdev->regview + qed_hw_bar_size(p_hwfn, BAR_ID_0) / 2;
 		p_regview = addr;
 
 		/* adjust doorbell bar offset for second engine */
-		addr = cdev->doorbells + qed_hw_bar_size(p_hwfn, 1) / 2;
+		addr = cdev->doorbells + qed_hw_bar_size(p_hwfn, BAR_ID_1) / 2;
 		p_doorbell = addr;
 
 		/* prepare second hw function */
-- 
cgit v0.10.2


From 12e09c694a68a3d71c7c655eeb689b7fe43b09f5 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 2 Mar 2016 20:26:01 +0200
Subject: qed: Support B0 instead of A0

BB_A0 is a development model that is will not reach actual clients.
In fact, future firmware would simply fail to initialize such chip.

This changes the configuration into B0 instead of A0, and adds a safeguard
against the slim chance someone would actually try this with an A0 adapter
in which case probe would gracefully fail.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 706c614..2e02e05 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -420,7 +420,7 @@ static void qed_calc_hw_mode(struct qed_hwfn *p_hwfn)
 {
 	int hw_mode = 0;
 
-	hw_mode = (1 << MODE_BB_A0);
+	hw_mode = (1 << MODE_BB_B0);
 
 	switch (p_hwfn->cdev->num_ports_in_engines) {
 	case 1:
@@ -976,18 +976,8 @@ static void qed_hw_hwfn_free(struct qed_hwfn *p_hwfn)
 }
 
 /* Setup bar access */
-static int qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn)
+static void qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn)
 {
-	int rc;
-
-	/* Allocate PTT pool */
-	rc = qed_ptt_pool_alloc(p_hwfn);
-	if (rc)
-		return rc;
-
-	/* Allocate the main PTT */
-	p_hwfn->p_main_ptt = qed_get_reserved_ptt(p_hwfn, RESERVED_PTT_MAIN);
-
 	/* clear indirect access */
 	qed_wr(p_hwfn, p_hwfn->p_main_ptt, PGLUE_B_REG_PGL_ADDR_88_F0, 0);
 	qed_wr(p_hwfn, p_hwfn->p_main_ptt, PGLUE_B_REG_PGL_ADDR_8C_F0, 0);
@@ -1002,8 +992,6 @@ static int qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn)
 	/* enable internal target-read */
 	qed_wr(p_hwfn, p_hwfn->p_main_ptt,
 	       PGLUE_B_REG_INTERNAL_PFID_ENABLE_TARGET_READ, 1);
-
-	return 0;
 }
 
 static void get_function_id(struct qed_hwfn *p_hwfn)
@@ -1311,7 +1299,7 @@ qed_get_hw_info(struct qed_hwfn *p_hwfn,
 	return rc;
 }
 
-static void qed_get_dev_info(struct qed_dev *cdev)
+static int qed_get_dev_info(struct qed_dev *cdev)
 {
 	struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev);
 	u32 tmp;
@@ -1350,6 +1338,14 @@ static void qed_get_dev_info(struct qed_dev *cdev)
 		"Chip details - Num: %04x Rev: %04x Bond id: %04x Metal: %04x\n",
 		cdev->chip_num, cdev->chip_rev,
 		cdev->chip_bond_id, cdev->chip_metal);
+
+	if (QED_IS_BB(cdev) && CHIP_REV_IS_A0(cdev)) {
+		DP_NOTICE(cdev->hwfns,
+			  "The chip type/rev (BB A0) is not supported!\n");
+		return -EINVAL;
+	}
+
+	return 0;
 }
 
 static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn,
@@ -1372,15 +1368,24 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn,
 
 	get_function_id(p_hwfn);
 
-	rc = qed_hw_hwfn_prepare(p_hwfn);
+	/* Allocate PTT pool */
+	rc = qed_ptt_pool_alloc(p_hwfn);
 	if (rc) {
 		DP_NOTICE(p_hwfn, "Failed to prepare hwfn's hw\n");
 		goto err0;
 	}
 
+	/* Allocate the main PTT */
+	p_hwfn->p_main_ptt = qed_get_reserved_ptt(p_hwfn, RESERVED_PTT_MAIN);
+
 	/* First hwfn learns basic information, e.g., number of hwfns */
-	if (!p_hwfn->my_id)
-		qed_get_dev_info(p_hwfn->cdev);
+	if (!p_hwfn->my_id) {
+		rc = qed_get_dev_info(p_hwfn->cdev);
+		if (rc != 0)
+			goto err1;
+	}
+
+	qed_hw_hwfn_prepare(p_hwfn);
 
 	/* Initialize MCP structure */
 	rc = qed_mcp_cmd_init(p_hwfn, p_hwfn->p_main_ptt);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 49bbf69..592e0e6 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -968,7 +968,7 @@ struct igu_msix_vector {
 
 enum init_modes {
 	MODE_BB_A0,
-	MODE_RESERVED,
+	MODE_BB_B0,
 	MODE_RESERVED2,
 	MODE_ASIC,
 	MODE_RESERVED3,
-- 
cgit v0.10.2


From 86622ee75312568e9a862b6f023e54b149380b6d Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 2 Mar 2016 20:26:02 +0200
Subject: qed: Move statistics to L2 code

Current statistics logic is meant for L2, not for all future protocols.
Move this content to the proper designated file.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index 2e02e05..d08078f 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -677,10 +677,8 @@ int qed_hw_init(struct qed_dev *cdev,
 		bool allow_npar_tx_switch,
 		const u8 *bin_fw_data)
 {
-	struct qed_storm_stats *p_stat;
-	u32 load_code, param, *p_address;
+	u32 load_code, param;
 	int rc, mfw_rc, i;
-	u8 fw_vport = 0;
 
 	rc = qed_init_fw_data(cdev, bin_fw_data);
 	if (rc != 0)
@@ -689,10 +687,6 @@ int qed_hw_init(struct qed_dev *cdev,
 	for_each_hwfn(cdev, i) {
 		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
 
-		rc = qed_fw_vport(p_hwfn, 0, &fw_vport);
-		if (rc != 0)
-			return rc;
-
 		/* Enable DMAE in PXP */
 		rc = qed_change_pci_hwfn(p_hwfn, p_hwfn->p_main_ptt, true);
 
@@ -756,25 +750,6 @@ int qed_hw_init(struct qed_dev *cdev,
 		}
 
 		p_hwfn->hw_init_done = true;
-
-		/* init PF stats */
-		p_stat = &p_hwfn->storm_stats;
-		p_stat->mstats.address = BAR0_MAP_REG_MSDM_RAM +
-					 MSTORM_QUEUE_STAT_OFFSET(fw_vport);
-		p_stat->mstats.len = sizeof(struct eth_mstorm_per_queue_stat);
-
-		p_stat->ustats.address = BAR0_MAP_REG_USDM_RAM +
-					 USTORM_QUEUE_STAT_OFFSET(fw_vport);
-		p_stat->ustats.len = sizeof(struct eth_ustorm_per_queue_stat);
-
-		p_stat->pstats.address = BAR0_MAP_REG_PSDM_RAM +
-					 PSTORM_QUEUE_STAT_OFFSET(fw_vport);
-		p_stat->pstats.len = sizeof(struct eth_pstorm_per_queue_stat);
-
-		p_address = &p_stat->tstats.address;
-		*p_address = BAR0_MAP_REG_TSDM_RAM +
-			     TSTORM_PORT_STAT_OFFSET(MFW_PORT(p_hwfn));
-		p_stat->tstats.len = sizeof(struct tstorm_per_port_stat);
 	}
 
 	return 0;
@@ -1557,223 +1532,6 @@ void qed_chain_free(struct qed_dev *cdev,
 			  p_chain->p_phys_addr);
 }
 
-static void __qed_get_vport_stats(struct qed_dev *cdev,
-				  struct qed_eth_stats  *stats)
-{
-	int i, j;
-
-	memset(stats, 0, sizeof(*stats));
-
-	for_each_hwfn(cdev, i) {
-		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
-		struct eth_mstorm_per_queue_stat mstats;
-		struct eth_ustorm_per_queue_stat ustats;
-		struct eth_pstorm_per_queue_stat pstats;
-		struct tstorm_per_port_stat tstats;
-		struct port_stats port_stats;
-		struct qed_ptt *p_ptt = qed_ptt_acquire(p_hwfn);
-
-		if (!p_ptt) {
-			DP_ERR(p_hwfn, "Failed to acquire ptt\n");
-			continue;
-		}
-
-		memset(&mstats, 0, sizeof(mstats));
-		qed_memcpy_from(p_hwfn, p_ptt, &mstats,
-				p_hwfn->storm_stats.mstats.address,
-				p_hwfn->storm_stats.mstats.len);
-
-		memset(&ustats, 0, sizeof(ustats));
-		qed_memcpy_from(p_hwfn, p_ptt, &ustats,
-				p_hwfn->storm_stats.ustats.address,
-				p_hwfn->storm_stats.ustats.len);
-
-		memset(&pstats, 0, sizeof(pstats));
-		qed_memcpy_from(p_hwfn, p_ptt, &pstats,
-				p_hwfn->storm_stats.pstats.address,
-				p_hwfn->storm_stats.pstats.len);
-
-		memset(&tstats, 0, sizeof(tstats));
-		qed_memcpy_from(p_hwfn, p_ptt, &tstats,
-				p_hwfn->storm_stats.tstats.address,
-				p_hwfn->storm_stats.tstats.len);
-
-		memset(&port_stats, 0, sizeof(port_stats));
-
-		if (p_hwfn->mcp_info)
-			qed_memcpy_from(p_hwfn, p_ptt, &port_stats,
-					p_hwfn->mcp_info->port_addr +
-					offsetof(struct public_port, stats),
-					sizeof(port_stats));
-		qed_ptt_release(p_hwfn, p_ptt);
-
-		stats->no_buff_discards +=
-			HILO_64_REGPAIR(mstats.no_buff_discard);
-		stats->packet_too_big_discard +=
-			HILO_64_REGPAIR(mstats.packet_too_big_discard);
-		stats->ttl0_discard +=
-			HILO_64_REGPAIR(mstats.ttl0_discard);
-		stats->tpa_coalesced_pkts +=
-			HILO_64_REGPAIR(mstats.tpa_coalesced_pkts);
-		stats->tpa_coalesced_events +=
-			HILO_64_REGPAIR(mstats.tpa_coalesced_events);
-		stats->tpa_aborts_num +=
-			HILO_64_REGPAIR(mstats.tpa_aborts_num);
-		stats->tpa_coalesced_bytes +=
-			HILO_64_REGPAIR(mstats.tpa_coalesced_bytes);
-
-		stats->rx_ucast_bytes +=
-			HILO_64_REGPAIR(ustats.rcv_ucast_bytes);
-		stats->rx_mcast_bytes +=
-			HILO_64_REGPAIR(ustats.rcv_mcast_bytes);
-		stats->rx_bcast_bytes +=
-			HILO_64_REGPAIR(ustats.rcv_bcast_bytes);
-		stats->rx_ucast_pkts +=
-			HILO_64_REGPAIR(ustats.rcv_ucast_pkts);
-		stats->rx_mcast_pkts +=
-			HILO_64_REGPAIR(ustats.rcv_mcast_pkts);
-		stats->rx_bcast_pkts +=
-			HILO_64_REGPAIR(ustats.rcv_bcast_pkts);
-
-		stats->mftag_filter_discards +=
-			HILO_64_REGPAIR(tstats.mftag_filter_discard);
-		stats->mac_filter_discards +=
-			HILO_64_REGPAIR(tstats.eth_mac_filter_discard);
-
-		stats->tx_ucast_bytes +=
-			HILO_64_REGPAIR(pstats.sent_ucast_bytes);
-		stats->tx_mcast_bytes +=
-			HILO_64_REGPAIR(pstats.sent_mcast_bytes);
-		stats->tx_bcast_bytes +=
-			HILO_64_REGPAIR(pstats.sent_bcast_bytes);
-		stats->tx_ucast_pkts +=
-			HILO_64_REGPAIR(pstats.sent_ucast_pkts);
-		stats->tx_mcast_pkts +=
-			HILO_64_REGPAIR(pstats.sent_mcast_pkts);
-		stats->tx_bcast_pkts +=
-			HILO_64_REGPAIR(pstats.sent_bcast_pkts);
-		stats->tx_err_drop_pkts +=
-			HILO_64_REGPAIR(pstats.error_drop_pkts);
-		stats->rx_64_byte_packets       += port_stats.pmm.r64;
-		stats->rx_127_byte_packets      += port_stats.pmm.r127;
-		stats->rx_255_byte_packets      += port_stats.pmm.r255;
-		stats->rx_511_byte_packets      += port_stats.pmm.r511;
-		stats->rx_1023_byte_packets     += port_stats.pmm.r1023;
-		stats->rx_1518_byte_packets     += port_stats.pmm.r1518;
-		stats->rx_1522_byte_packets     += port_stats.pmm.r1522;
-		stats->rx_2047_byte_packets     += port_stats.pmm.r2047;
-		stats->rx_4095_byte_packets     += port_stats.pmm.r4095;
-		stats->rx_9216_byte_packets     += port_stats.pmm.r9216;
-		stats->rx_16383_byte_packets    += port_stats.pmm.r16383;
-		stats->rx_crc_errors	    += port_stats.pmm.rfcs;
-		stats->rx_mac_crtl_frames       += port_stats.pmm.rxcf;
-		stats->rx_pause_frames	  += port_stats.pmm.rxpf;
-		stats->rx_pfc_frames	    += port_stats.pmm.rxpp;
-		stats->rx_align_errors	  += port_stats.pmm.raln;
-		stats->rx_carrier_errors	+= port_stats.pmm.rfcr;
-		stats->rx_oversize_packets      += port_stats.pmm.rovr;
-		stats->rx_jabbers	       += port_stats.pmm.rjbr;
-		stats->rx_undersize_packets     += port_stats.pmm.rund;
-		stats->rx_fragments	     += port_stats.pmm.rfrg;
-		stats->tx_64_byte_packets       += port_stats.pmm.t64;
-		stats->tx_65_to_127_byte_packets += port_stats.pmm.t127;
-		stats->tx_128_to_255_byte_packets += port_stats.pmm.t255;
-		stats->tx_256_to_511_byte_packets  += port_stats.pmm.t511;
-		stats->tx_512_to_1023_byte_packets += port_stats.pmm.t1023;
-		stats->tx_1024_to_1518_byte_packets += port_stats.pmm.t1518;
-		stats->tx_1519_to_2047_byte_packets += port_stats.pmm.t2047;
-		stats->tx_2048_to_4095_byte_packets += port_stats.pmm.t4095;
-		stats->tx_4096_to_9216_byte_packets += port_stats.pmm.t9216;
-		stats->tx_9217_to_16383_byte_packets += port_stats.pmm.t16383;
-		stats->tx_pause_frames	  += port_stats.pmm.txpf;
-		stats->tx_pfc_frames	    += port_stats.pmm.txpp;
-		stats->tx_lpi_entry_count       += port_stats.pmm.tlpiec;
-		stats->tx_total_collisions      += port_stats.pmm.tncl;
-		stats->rx_mac_bytes	     += port_stats.pmm.rbyte;
-		stats->rx_mac_uc_packets	+= port_stats.pmm.rxuca;
-		stats->rx_mac_mc_packets	+= port_stats.pmm.rxmca;
-		stats->rx_mac_bc_packets	+= port_stats.pmm.rxbca;
-		stats->rx_mac_frames_ok	 += port_stats.pmm.rxpok;
-		stats->tx_mac_bytes	     += port_stats.pmm.tbyte;
-		stats->tx_mac_uc_packets	+= port_stats.pmm.txuca;
-		stats->tx_mac_mc_packets	+= port_stats.pmm.txmca;
-		stats->tx_mac_bc_packets	+= port_stats.pmm.txbca;
-		stats->tx_mac_ctrl_frames       += port_stats.pmm.txcf;
-
-		for (j = 0; j < 8; j++) {
-			stats->brb_truncates += port_stats.brb.brb_truncate[j];
-			stats->brb_discards += port_stats.brb.brb_discard[j];
-		}
-	}
-}
-
-void qed_get_vport_stats(struct qed_dev *cdev,
-			 struct qed_eth_stats *stats)
-{
-	u32 i;
-
-	if (!cdev) {
-		memset(stats, 0, sizeof(*stats));
-		return;
-	}
-
-	__qed_get_vport_stats(cdev, stats);
-
-	if (!cdev->reset_stats)
-		return;
-
-	/* Reduce the statistics baseline */
-	for (i = 0; i < sizeof(struct qed_eth_stats) / sizeof(u64); i++)
-		((u64 *)stats)[i] -= ((u64 *)cdev->reset_stats)[i];
-}
-
-/* zeroes V-PORT specific portion of stats (Port stats remains untouched) */
-void qed_reset_vport_stats(struct qed_dev *cdev)
-{
-	int i;
-
-	for_each_hwfn(cdev, i) {
-		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
-		struct eth_mstorm_per_queue_stat mstats;
-		struct eth_ustorm_per_queue_stat ustats;
-		struct eth_pstorm_per_queue_stat pstats;
-		struct qed_ptt *p_ptt = qed_ptt_acquire(p_hwfn);
-
-		if (!p_ptt) {
-			DP_ERR(p_hwfn, "Failed to acquire ptt\n");
-			continue;
-		}
-
-		memset(&mstats, 0, sizeof(mstats));
-		qed_memcpy_to(p_hwfn, p_ptt,
-			      p_hwfn->storm_stats.mstats.address,
-			      &mstats,
-			      p_hwfn->storm_stats.mstats.len);
-
-		memset(&ustats, 0, sizeof(ustats));
-		qed_memcpy_to(p_hwfn, p_ptt,
-			      p_hwfn->storm_stats.ustats.address,
-			      &ustats,
-			      p_hwfn->storm_stats.ustats.len);
-
-		memset(&pstats, 0, sizeof(pstats));
-		qed_memcpy_to(p_hwfn, p_ptt,
-			      p_hwfn->storm_stats.pstats.address,
-			      &pstats,
-			      p_hwfn->storm_stats.pstats.len);
-
-		qed_ptt_release(p_hwfn, p_ptt);
-	}
-
-	/* PORT statistics are not necessarily reset, so we need to
-	 * read and create a baseline for future statistics.
-	 */
-	if (!cdev->reset_stats)
-		DP_INFO(cdev, "Reset stats not allocated\n");
-	else
-		__qed_get_vport_stats(cdev, cdev->reset_stats);
-}
-
 int qed_fw_l2_queue(struct qed_hwfn *p_hwfn,
 		    u16 src_id, u16 *dst_id)
 {
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
index e29a3ba..155f26b 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
@@ -156,8 +156,6 @@ struct qed_ptt *qed_ptt_acquire(struct qed_hwfn *p_hwfn);
  */
 void qed_ptt_release(struct qed_hwfn *p_hwfn,
 		     struct qed_ptt *p_ptt);
-void qed_get_vport_stats(struct qed_dev *cdev,
-			 struct qed_eth_stats   *stats);
 void qed_reset_vport_stats(struct qed_dev *cdev);
 
 enum qed_dmae_address_type_t {
diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index bba59c5..102ddc7 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -31,6 +31,7 @@
 #include "qed_hsi.h"
 #include "qed_hw.h"
 #include "qed_int.h"
+#include "qed_mcp.h"
 #include "qed_reg_addr.h"
 #include "qed_sp.h"
 
@@ -1231,6 +1232,328 @@ static int qed_filter_ucast_cmd(struct qed_dev *cdev,
 	return rc;
 }
 
+/* Statistics related code */
+static void __qed_get_vport_pstats_addrlen(struct qed_hwfn *p_hwfn,
+					   u32 *p_addr,
+					   u32 *p_len,
+					   u16 statistics_bin)
+{
+	*p_addr = BAR0_MAP_REG_PSDM_RAM +
+		  PSTORM_QUEUE_STAT_OFFSET(statistics_bin);
+	*p_len = sizeof(struct eth_pstorm_per_queue_stat);
+}
+
+static void __qed_get_vport_pstats(struct qed_hwfn *p_hwfn,
+				   struct qed_ptt *p_ptt,
+				   struct qed_eth_stats *p_stats,
+				   u16 statistics_bin)
+{
+	struct eth_pstorm_per_queue_stat pstats;
+	u32 pstats_addr = 0, pstats_len = 0;
+
+	__qed_get_vport_pstats_addrlen(p_hwfn, &pstats_addr, &pstats_len,
+				       statistics_bin);
+
+	memset(&pstats, 0, sizeof(pstats));
+	qed_memcpy_from(p_hwfn, p_ptt, &pstats,
+			pstats_addr, pstats_len);
+
+	p_stats->tx_ucast_bytes +=
+		HILO_64_REGPAIR(pstats.sent_ucast_bytes);
+	p_stats->tx_mcast_bytes +=
+		HILO_64_REGPAIR(pstats.sent_mcast_bytes);
+	p_stats->tx_bcast_bytes +=
+		HILO_64_REGPAIR(pstats.sent_bcast_bytes);
+	p_stats->tx_ucast_pkts +=
+		HILO_64_REGPAIR(pstats.sent_ucast_pkts);
+	p_stats->tx_mcast_pkts +=
+		HILO_64_REGPAIR(pstats.sent_mcast_pkts);
+	p_stats->tx_bcast_pkts +=
+		HILO_64_REGPAIR(pstats.sent_bcast_pkts);
+	p_stats->tx_err_drop_pkts +=
+		HILO_64_REGPAIR(pstats.error_drop_pkts);
+}
+
+static void __qed_get_vport_tstats_addrlen(struct qed_hwfn *p_hwfn,
+					   u32 *p_addr,
+					   u32 *p_len)
+{
+	*p_addr = BAR0_MAP_REG_TSDM_RAM +
+		  TSTORM_PORT_STAT_OFFSET(MFW_PORT(p_hwfn));
+	*p_len = sizeof(struct tstorm_per_port_stat);
+}
+
+static void __qed_get_vport_tstats(struct qed_hwfn *p_hwfn,
+				   struct qed_ptt *p_ptt,
+				   struct qed_eth_stats *p_stats,
+				   u16 statistics_bin)
+{
+	u32 tstats_addr = 0, tstats_len = 0;
+	struct tstorm_per_port_stat tstats;
+
+	__qed_get_vport_tstats_addrlen(p_hwfn, &tstats_addr, &tstats_len);
+
+	memset(&tstats, 0, sizeof(tstats));
+	qed_memcpy_from(p_hwfn, p_ptt, &tstats,
+			tstats_addr, tstats_len);
+
+	p_stats->mftag_filter_discards +=
+		HILO_64_REGPAIR(tstats.mftag_filter_discard);
+	p_stats->mac_filter_discards +=
+		HILO_64_REGPAIR(tstats.eth_mac_filter_discard);
+}
+
+static void __qed_get_vport_ustats_addrlen(struct qed_hwfn *p_hwfn,
+					   u32 *p_addr,
+					   u32 *p_len,
+					   u16 statistics_bin)
+{
+	*p_addr = BAR0_MAP_REG_USDM_RAM +
+		  USTORM_QUEUE_STAT_OFFSET(statistics_bin);
+	*p_len = sizeof(struct eth_ustorm_per_queue_stat);
+}
+
+static void __qed_get_vport_ustats(struct qed_hwfn *p_hwfn,
+				   struct qed_ptt *p_ptt,
+				   struct qed_eth_stats *p_stats,
+				   u16 statistics_bin)
+{
+	struct eth_ustorm_per_queue_stat ustats;
+	u32 ustats_addr = 0, ustats_len = 0;
+
+	__qed_get_vport_ustats_addrlen(p_hwfn, &ustats_addr, &ustats_len,
+				       statistics_bin);
+
+	memset(&ustats, 0, sizeof(ustats));
+	qed_memcpy_from(p_hwfn, p_ptt, &ustats,
+			ustats_addr, ustats_len);
+
+	p_stats->rx_ucast_bytes +=
+		HILO_64_REGPAIR(ustats.rcv_ucast_bytes);
+	p_stats->rx_mcast_bytes +=
+		HILO_64_REGPAIR(ustats.rcv_mcast_bytes);
+	p_stats->rx_bcast_bytes +=
+		HILO_64_REGPAIR(ustats.rcv_bcast_bytes);
+	p_stats->rx_ucast_pkts +=
+		HILO_64_REGPAIR(ustats.rcv_ucast_pkts);
+	p_stats->rx_mcast_pkts +=
+		HILO_64_REGPAIR(ustats.rcv_mcast_pkts);
+	p_stats->rx_bcast_pkts +=
+		HILO_64_REGPAIR(ustats.rcv_bcast_pkts);
+}
+
+static void __qed_get_vport_mstats_addrlen(struct qed_hwfn *p_hwfn,
+					   u32 *p_addr,
+					   u32 *p_len,
+					   u16 statistics_bin)
+{
+	*p_addr = BAR0_MAP_REG_MSDM_RAM +
+		  MSTORM_QUEUE_STAT_OFFSET(statistics_bin);
+	*p_len = sizeof(struct eth_mstorm_per_queue_stat);
+}
+
+static void __qed_get_vport_mstats(struct qed_hwfn *p_hwfn,
+				   struct qed_ptt *p_ptt,
+				   struct qed_eth_stats *p_stats,
+				   u16 statistics_bin)
+{
+	struct eth_mstorm_per_queue_stat mstats;
+	u32 mstats_addr = 0, mstats_len = 0;
+
+	__qed_get_vport_mstats_addrlen(p_hwfn, &mstats_addr, &mstats_len,
+				       statistics_bin);
+
+	memset(&mstats, 0, sizeof(mstats));
+	qed_memcpy_from(p_hwfn, p_ptt, &mstats,
+			mstats_addr, mstats_len);
+
+	p_stats->no_buff_discards +=
+		HILO_64_REGPAIR(mstats.no_buff_discard);
+	p_stats->packet_too_big_discard +=
+		HILO_64_REGPAIR(mstats.packet_too_big_discard);
+	p_stats->ttl0_discard +=
+		HILO_64_REGPAIR(mstats.ttl0_discard);
+	p_stats->tpa_coalesced_pkts +=
+		HILO_64_REGPAIR(mstats.tpa_coalesced_pkts);
+	p_stats->tpa_coalesced_events +=
+		HILO_64_REGPAIR(mstats.tpa_coalesced_events);
+	p_stats->tpa_aborts_num +=
+		HILO_64_REGPAIR(mstats.tpa_aborts_num);
+	p_stats->tpa_coalesced_bytes +=
+		HILO_64_REGPAIR(mstats.tpa_coalesced_bytes);
+}
+
+static void __qed_get_vport_port_stats(struct qed_hwfn *p_hwfn,
+				       struct qed_ptt *p_ptt,
+				       struct qed_eth_stats *p_stats)
+{
+	struct port_stats port_stats;
+	int j;
+
+	memset(&port_stats, 0, sizeof(port_stats));
+
+	qed_memcpy_from(p_hwfn, p_ptt, &port_stats,
+			p_hwfn->mcp_info->port_addr +
+			offsetof(struct public_port, stats),
+			sizeof(port_stats));
+
+	p_stats->rx_64_byte_packets		+= port_stats.pmm.r64;
+	p_stats->rx_127_byte_packets		+= port_stats.pmm.r127;
+	p_stats->rx_255_byte_packets		+= port_stats.pmm.r255;
+	p_stats->rx_511_byte_packets		+= port_stats.pmm.r511;
+	p_stats->rx_1023_byte_packets		+= port_stats.pmm.r1023;
+	p_stats->rx_1518_byte_packets		+= port_stats.pmm.r1518;
+	p_stats->rx_1522_byte_packets		+= port_stats.pmm.r1522;
+	p_stats->rx_2047_byte_packets		+= port_stats.pmm.r2047;
+	p_stats->rx_4095_byte_packets		+= port_stats.pmm.r4095;
+	p_stats->rx_9216_byte_packets		+= port_stats.pmm.r9216;
+	p_stats->rx_16383_byte_packets		+= port_stats.pmm.r16383;
+	p_stats->rx_crc_errors			+= port_stats.pmm.rfcs;
+	p_stats->rx_mac_crtl_frames		+= port_stats.pmm.rxcf;
+	p_stats->rx_pause_frames		+= port_stats.pmm.rxpf;
+	p_stats->rx_pfc_frames			+= port_stats.pmm.rxpp;
+	p_stats->rx_align_errors		+= port_stats.pmm.raln;
+	p_stats->rx_carrier_errors		+= port_stats.pmm.rfcr;
+	p_stats->rx_oversize_packets		+= port_stats.pmm.rovr;
+	p_stats->rx_jabbers			+= port_stats.pmm.rjbr;
+	p_stats->rx_undersize_packets		+= port_stats.pmm.rund;
+	p_stats->rx_fragments			+= port_stats.pmm.rfrg;
+	p_stats->tx_64_byte_packets		+= port_stats.pmm.t64;
+	p_stats->tx_65_to_127_byte_packets	+= port_stats.pmm.t127;
+	p_stats->tx_128_to_255_byte_packets	+= port_stats.pmm.t255;
+	p_stats->tx_256_to_511_byte_packets	+= port_stats.pmm.t511;
+	p_stats->tx_512_to_1023_byte_packets	+= port_stats.pmm.t1023;
+	p_stats->tx_1024_to_1518_byte_packets	+= port_stats.pmm.t1518;
+	p_stats->tx_1519_to_2047_byte_packets	+= port_stats.pmm.t2047;
+	p_stats->tx_2048_to_4095_byte_packets	+= port_stats.pmm.t4095;
+	p_stats->tx_4096_to_9216_byte_packets	+= port_stats.pmm.t9216;
+	p_stats->tx_9217_to_16383_byte_packets	+= port_stats.pmm.t16383;
+	p_stats->tx_pause_frames		+= port_stats.pmm.txpf;
+	p_stats->tx_pfc_frames			+= port_stats.pmm.txpp;
+	p_stats->tx_lpi_entry_count		+= port_stats.pmm.tlpiec;
+	p_stats->tx_total_collisions		+= port_stats.pmm.tncl;
+	p_stats->rx_mac_bytes			+= port_stats.pmm.rbyte;
+	p_stats->rx_mac_uc_packets		+= port_stats.pmm.rxuca;
+	p_stats->rx_mac_mc_packets		+= port_stats.pmm.rxmca;
+	p_stats->rx_mac_bc_packets		+= port_stats.pmm.rxbca;
+	p_stats->rx_mac_frames_ok		+= port_stats.pmm.rxpok;
+	p_stats->tx_mac_bytes			+= port_stats.pmm.tbyte;
+	p_stats->tx_mac_uc_packets		+= port_stats.pmm.txuca;
+	p_stats->tx_mac_mc_packets		+= port_stats.pmm.txmca;
+	p_stats->tx_mac_bc_packets		+= port_stats.pmm.txbca;
+	p_stats->tx_mac_ctrl_frames		+= port_stats.pmm.txcf;
+	for (j = 0; j < 8; j++) {
+		p_stats->brb_truncates	+= port_stats.brb.brb_truncate[j];
+		p_stats->brb_discards	+= port_stats.brb.brb_discard[j];
+	}
+}
+
+static void __qed_get_vport_stats(struct qed_hwfn *p_hwfn,
+				  struct qed_ptt *p_ptt,
+				  struct qed_eth_stats *stats,
+				  u16 statistics_bin)
+{
+	__qed_get_vport_mstats(p_hwfn, p_ptt, stats, statistics_bin);
+	__qed_get_vport_ustats(p_hwfn, p_ptt, stats, statistics_bin);
+	__qed_get_vport_tstats(p_hwfn, p_ptt, stats, statistics_bin);
+	__qed_get_vport_pstats(p_hwfn, p_ptt, stats, statistics_bin);
+
+	if (p_hwfn->mcp_info)
+		__qed_get_vport_port_stats(p_hwfn, p_ptt, stats);
+}
+
+static void _qed_get_vport_stats(struct qed_dev *cdev,
+				 struct qed_eth_stats *stats)
+{
+	u8	fw_vport = 0;
+	int	i;
+
+	memset(stats, 0, sizeof(*stats));
+
+	for_each_hwfn(cdev, i) {
+		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
+		struct qed_ptt *p_ptt;
+
+		/* The main vport index is relative first */
+		if (qed_fw_vport(p_hwfn, 0, &fw_vport)) {
+			DP_ERR(p_hwfn, "No vport available!\n");
+			continue;
+		}
+
+		p_ptt = qed_ptt_acquire(p_hwfn);
+		if (!p_ptt) {
+			DP_ERR(p_hwfn, "Failed to acquire ptt\n");
+			continue;
+		}
+
+		__qed_get_vport_stats(p_hwfn, p_ptt, stats, fw_vport);
+
+		qed_ptt_release(p_hwfn, p_ptt);
+	}
+}
+
+void qed_get_vport_stats(struct qed_dev *cdev,
+			 struct qed_eth_stats *stats)
+{
+	u32 i;
+
+	if (!cdev) {
+		memset(stats, 0, sizeof(*stats));
+		return;
+	}
+
+	_qed_get_vport_stats(cdev, stats);
+
+	if (!cdev->reset_stats)
+		return;
+
+	/* Reduce the statistics baseline */
+	for (i = 0; i < sizeof(struct qed_eth_stats) / sizeof(u64); i++)
+		((u64 *)stats)[i] -= ((u64 *)cdev->reset_stats)[i];
+}
+
+/* zeroes V-PORT specific portion of stats (Port stats remains untouched) */
+void qed_reset_vport_stats(struct qed_dev *cdev)
+{
+	int i;
+
+	for_each_hwfn(cdev, i) {
+		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
+		struct eth_mstorm_per_queue_stat mstats;
+		struct eth_ustorm_per_queue_stat ustats;
+		struct eth_pstorm_per_queue_stat pstats;
+		struct qed_ptt *p_ptt = qed_ptt_acquire(p_hwfn);
+		u32 addr = 0, len = 0;
+
+		if (!p_ptt) {
+			DP_ERR(p_hwfn, "Failed to acquire ptt\n");
+			continue;
+		}
+
+		memset(&mstats, 0, sizeof(mstats));
+		__qed_get_vport_mstats_addrlen(p_hwfn, &addr, &len, 0);
+		qed_memcpy_to(p_hwfn, p_ptt, addr, &mstats, len);
+
+		memset(&ustats, 0, sizeof(ustats));
+		__qed_get_vport_ustats_addrlen(p_hwfn, &addr, &len, 0);
+		qed_memcpy_to(p_hwfn, p_ptt, addr, &ustats, len);
+
+		memset(&pstats, 0, sizeof(pstats));
+		__qed_get_vport_pstats_addrlen(p_hwfn, &addr, &len, 0);
+		qed_memcpy_to(p_hwfn, p_ptt, addr, &pstats, len);
+
+		qed_ptt_release(p_hwfn, p_ptt);
+	}
+
+	/* PORT statistics are not necessarily reset, so we need to
+	 * read and create a baseline for future statistics.
+	 */
+	if (!cdev->reset_stats)
+		DP_INFO(cdev, "Reset stats not allocated\n");
+	else
+		_qed_get_vport_stats(cdev, cdev->reset_stats);
+}
+
 static int qed_fill_eth_dev_info(struct qed_dev *cdev,
 				 struct qed_dev_eth_info *info)
 {
-- 
cgit v0.10.2


From 8c925c443801721d2731e540055782bcdbbde03b Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 2 Mar 2016 20:26:03 +0200
Subject: qed: Fix error flow on slowpath start

In case of problems when initializing the chip, the error flows aren't
being properly done. Specifically, it's possible that the chip would be
left in a configuration allowing it [internally] to access the host
memory, causing fatal problems in the device that would require power
cycle to overcome.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index d08078f..c3f293d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -756,10 +756,54 @@ int qed_hw_init(struct qed_dev *cdev,
 }
 
 #define QED_HW_STOP_RETRY_LIMIT (10)
+static inline void qed_hw_timers_stop(struct qed_dev *cdev,
+				      struct qed_hwfn *p_hwfn,
+				      struct qed_ptt *p_ptt)
+{
+	int i;
+
+	/* close timers */
+	qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0);
+	qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0);
+
+	for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) {
+		if ((!qed_rd(p_hwfn, p_ptt,
+			     TM_REG_PF_SCAN_ACTIVE_CONN)) &&
+		    (!qed_rd(p_hwfn, p_ptt,
+			     TM_REG_PF_SCAN_ACTIVE_TASK)))
+			break;
+
+		/* Dependent on number of connection/tasks, possibly
+		 * 1ms sleep is required between polls
+		 */
+		usleep_range(1000, 2000);
+	}
+
+	if (i < QED_HW_STOP_RETRY_LIMIT)
+		return;
+
+	DP_NOTICE(p_hwfn,
+		  "Timers linear scans are not over [Connection %02x Tasks %02x]\n",
+		  (u8)qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_CONN),
+		  (u8)qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_TASK));
+}
+
+void qed_hw_timers_stop_all(struct qed_dev *cdev)
+{
+	int j;
+
+	for_each_hwfn(cdev, j) {
+		struct qed_hwfn *p_hwfn = &cdev->hwfns[j];
+		struct qed_ptt *p_ptt = p_hwfn->p_main_ptt;
+
+		qed_hw_timers_stop(cdev, p_hwfn, p_ptt);
+	}
+}
+
 int qed_hw_stop(struct qed_dev *cdev)
 {
 	int rc = 0, t_rc;
-	int i, j;
+	int j;
 
 	for_each_hwfn(cdev, j) {
 		struct qed_hwfn *p_hwfn = &cdev->hwfns[j];
@@ -772,7 +816,8 @@ int qed_hw_stop(struct qed_dev *cdev)
 
 		rc = qed_sp_pf_stop(p_hwfn);
 		if (rc)
-			return rc;
+			DP_NOTICE(p_hwfn,
+				  "Failed to close PF against FW. Continue to stop HW to prevent illegal host access by the device\n");
 
 		qed_wr(p_hwfn, p_ptt,
 		       NIG_REG_RX_LLH_BRB_GATE_DNTFWD_PERPF, 0x1);
@@ -783,24 +828,7 @@ int qed_hw_stop(struct qed_dev *cdev)
 		qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_ROCE, 0x0);
 		qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_OPENFLOW, 0x0);
 
-		qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0);
-		qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0);
-		for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) {
-			if ((!qed_rd(p_hwfn, p_ptt,
-				     TM_REG_PF_SCAN_ACTIVE_CONN)) &&
-			    (!qed_rd(p_hwfn, p_ptt,
-				     TM_REG_PF_SCAN_ACTIVE_TASK)))
-				break;
-
-			usleep_range(1000, 2000);
-		}
-		if (i == QED_HW_STOP_RETRY_LIMIT)
-			DP_NOTICE(p_hwfn,
-				  "Timers linear scans are not over [Connection %02x Tasks %02x]\n",
-				  (u8)qed_rd(p_hwfn, p_ptt,
-					     TM_REG_PF_SCAN_ACTIVE_CONN),
-				  (u8)qed_rd(p_hwfn, p_ptt,
-					     TM_REG_PF_SCAN_ACTIVE_TASK));
+		qed_hw_timers_stop(cdev, p_hwfn, p_ptt);
 
 		/* Disable Attention Generation */
 		qed_int_igu_disable_int(p_hwfn, p_ptt);
@@ -829,7 +857,7 @@ int qed_hw_stop(struct qed_dev *cdev)
 
 void qed_hw_stop_fastpath(struct qed_dev *cdev)
 {
-	int i, j;
+	int j;
 
 	for_each_hwfn(cdev, j) {
 		struct qed_hwfn *p_hwfn = &cdev->hwfns[j];
@@ -848,25 +876,6 @@ void qed_hw_stop_fastpath(struct qed_dev *cdev)
 		qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_ROCE, 0x0);
 		qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_OPENFLOW, 0x0);
 
-		qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0);
-		qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0);
-		for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) {
-			if ((!qed_rd(p_hwfn, p_ptt,
-				     TM_REG_PF_SCAN_ACTIVE_CONN)) &&
-			    (!qed_rd(p_hwfn, p_ptt,
-				     TM_REG_PF_SCAN_ACTIVE_TASK)))
-				break;
-
-			usleep_range(1000, 2000);
-		}
-		if (i == QED_HW_STOP_RETRY_LIMIT)
-			DP_NOTICE(p_hwfn,
-				  "Timers linear scans are not over [Connection %02x Tasks %02x]\n",
-				  (u8)qed_rd(p_hwfn, p_ptt,
-					     TM_REG_PF_SCAN_ACTIVE_CONN),
-				  (u8)qed_rd(p_hwfn, p_ptt,
-					     TM_REG_PF_SCAN_ACTIVE_TASK));
-
 		qed_int_igu_init_pure_rt(p_hwfn, p_ptt, false, false);
 
 		/* Need to wait 1ms to guarantee SBs are cleared */
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
index 155f26b..d6c7ddf 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h
@@ -78,6 +78,15 @@ int qed_hw_init(struct qed_dev *cdev,
 		const u8 *bin_fw_data);
 
 /**
+ * @brief qed_hw_timers_stop_all - stop the timers HW block
+ *
+ * @param cdev
+ *
+ * @return void
+ */
+void qed_hw_timers_stop_all(struct qed_dev *cdev);
+
+/**
  * @brief qed_hw_stop -
  *
  * @param cdev
diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c
index caa689e..26d40db 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_main.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_main.c
@@ -779,7 +779,7 @@ static int qed_slowpath_start(struct qed_dev *cdev,
 	rc = qed_hw_init(cdev, true, cdev->int_params.out.int_mode,
 			 true, data);
 	if (rc)
-		goto err3;
+		goto err2;
 
 	DP_INFO(cdev,
 		"HW initialization and function start completed successfully\n");
@@ -798,12 +798,14 @@ static int qed_slowpath_start(struct qed_dev *cdev,
 		return rc;
 	}
 
+	qed_reset_vport_stats(cdev);
+
 	return 0;
 
-err3:
-	qed_free_stream_mem(cdev);
-	qed_slowpath_irq_free(cdev);
 err2:
+	qed_hw_timers_stop_all(cdev);
+	qed_slowpath_irq_free(cdev);
+	qed_free_stream_mem(cdev);
 	qed_disable_msix(cdev);
 err1:
 	qed_resc_free(cdev);
-- 
cgit v0.10.2


From 619db46b259749b2ffe79aeea2247d7573861586 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 2 Mar 2016 20:26:04 +0200
Subject: qed: Remove unused NVM vendor ID

Remove 2 unused fields from driver code.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index 66b021e..e5604ee 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -146,9 +146,6 @@ struct qed_hw_info {
 	u16				ovlan;
 	u32				part_num[4];
 
-	u32				vendor_id;
-	u32				device_id;
-
 	unsigned char			hw_mac_addr[ETH_ALEN];
 
 	struct qed_igu_info		*p_igu_info;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c
index c3f293d..b7d100f 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_dev.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c
@@ -1083,13 +1083,6 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn,
 	/* Read nvm_cfg1  (Notice this is just offset, and not offsize (TBD) */
 	nvm_cfg1_offset = qed_rd(p_hwfn, p_ptt, nvm_cfg_addr + 4);
 
-	/* Read Vendor Id / Device Id */
-	addr = MCP_REG_SCRATCH + nvm_cfg1_offset +
-	       offsetof(struct nvm_cfg1, glob) +
-	       offsetof(struct nvm_cfg1_glob, pci_id);
-	p_hwfn->hw_info.vendor_id = qed_rd(p_hwfn, p_ptt, addr) &
-				    NVM_CFG1_GLOB_VENDOR_ID_MASK;
-
 	addr = MCP_REG_SCRATCH + nvm_cfg1_offset +
 	       offsetof(struct nvm_cfg1, glob) +
 	       offsetof(struct nvm_cfg1_glob, core_cfg);
-- 
cgit v0.10.2


From 5f547391f5f25bb71f27860ad25bd1e4715f0752 Mon Sep 17 00:00:00 2001
From: Joe Stringer <joe@ovn.org>
Date: Wed, 3 Feb 2016 10:00:10 -0800
Subject: netfilter: nf_defrag_ipv4: Drop redundant ip_send_check()

Since commit 0848f6428ba3 ("inet: frags: fix defragmented packet's IP
header for af_packet"), ip_send_check() would be called twice for
defragmentation that occurs from netfilter ipv4 defrag hooks. Remove the
extra call.

Signed-off-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c
index a04dee5..d88da36 100644
--- a/net/ipv4/netfilter/nf_defrag_ipv4.c
+++ b/net/ipv4/netfilter/nf_defrag_ipv4.c
@@ -31,10 +31,8 @@ static int nf_ct_ipv4_gather_frags(struct net *net, struct sk_buff *skb,
 	err = ip_defrag(net, skb, user);
 	local_bh_enable();
 
-	if (!err) {
-		ip_send_check(ip_hdr(skb));
+	if (!err)
 		skb->ignore_df = 1;
-	}
 
 	return err;
 }
-- 
cgit v0.10.2


From a67dd266adf42a24df31380e9da78390bb4d65ef Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 25 Feb 2016 10:08:35 +0100
Subject: netfilter: xtables: prepare for on-demand hook register

This change prepares for upcoming on-demand xtables hook registration.

We change the protoypes of the register/unregister functions.
A followup patch will then add nf_hook_register/unregister calls
to the iptables one.

Once a hook is registered packets will be picked up, so all assignments
of the form

net->ipv4.iptable_$table = new_table

have to be moved to ip(6)t_register_table, else we can see NULL
net->ipv4.iptable_$table later.

This patch doesn't change functionality; without this the actual change
simply gets too big.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 6f074db..029b95e 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -48,10 +48,11 @@ struct arpt_error {
 }
 
 extern void *arpt_alloc_initial_table(const struct xt_table *);
-extern struct xt_table *arpt_register_table(struct net *net,
-					    const struct xt_table *table,
-					    const struct arpt_replace *repl);
-extern void arpt_unregister_table(struct xt_table *table);
+int arpt_register_table(struct net *net, const struct xt_table *table,
+			const struct arpt_replace *repl,
+			const struct nf_hook_ops *ops, struct xt_table **res);
+void arpt_unregister_table(struct net *net, struct xt_table *table,
+			   const struct nf_hook_ops *ops);
 extern unsigned int arpt_do_table(struct sk_buff *skb,
 				  const struct nf_hook_state *state,
 				  struct xt_table *table);
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index aa598f9..7bfc589 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -24,10 +24,11 @@
 
 extern void ipt_init(void) __init;
 
-extern struct xt_table *ipt_register_table(struct net *net,
-					   const struct xt_table *table,
-					   const struct ipt_replace *repl);
-extern void ipt_unregister_table(struct net *net, struct xt_table *table);
+int ipt_register_table(struct net *net, const struct xt_table *table,
+		       const struct ipt_replace *repl,
+		       const struct nf_hook_ops *ops, struct xt_table **res);
+void ipt_unregister_table(struct net *net, struct xt_table *table,
+			  const struct nf_hook_ops *ops);
 
 /* Standard entry. */
 struct ipt_standard {
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index 0f76e5c..b21c392 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -25,10 +25,11 @@
 extern void ip6t_init(void) __init;
 
 extern void *ip6t_alloc_initial_table(const struct xt_table *);
-extern struct xt_table *ip6t_register_table(struct net *net,
-					    const struct xt_table *table,
-					    const struct ip6t_replace *repl);
-extern void ip6t_unregister_table(struct net *net, struct xt_table *table);
+int ip6t_register_table(struct net *net, const struct xt_table *table,
+			const struct ip6t_replace *repl,
+			const struct nf_hook_ops *ops, struct xt_table **res);
+void ip6t_unregister_table(struct net *net, struct xt_table *table,
+			   const struct nf_hook_ops *ops);
 extern unsigned int ip6t_do_table(struct sk_buff *skb,
 				  const struct nf_hook_state *state,
 				  struct xt_table *table);
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index b488cac..00eed08 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -1780,9 +1780,11 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len
 	return ret;
 }
 
-struct xt_table *arpt_register_table(struct net *net,
-				     const struct xt_table *table,
-				     const struct arpt_replace *repl)
+int arpt_register_table(struct net *net,
+			const struct xt_table *table,
+			const struct arpt_replace *repl,
+			const struct nf_hook_ops *ops,
+			struct xt_table **res)
 {
 	int ret;
 	struct xt_table_info *newinfo;
@@ -1791,10 +1793,8 @@ struct xt_table *arpt_register_table(struct net *net,
 	struct xt_table *new_table;
 
 	newinfo = xt_alloc_table_info(repl->size);
-	if (!newinfo) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!newinfo)
+		return -ENOMEM;
 
 	loc_cpu_entry = newinfo->entries;
 	memcpy(loc_cpu_entry, repl->entries, repl->size);
@@ -1809,15 +1809,18 @@ struct xt_table *arpt_register_table(struct net *net,
 		ret = PTR_ERR(new_table);
 		goto out_free;
 	}
-	return new_table;
+
+	WRITE_ONCE(*res, new_table);
+
+	return ret;
 
 out_free:
 	xt_free_table_info(newinfo);
-out:
-	return ERR_PTR(ret);
+	return ret;
 }
 
-void arpt_unregister_table(struct xt_table *table)
+void arpt_unregister_table(struct net *net, struct xt_table *table,
+			   const struct nf_hook_ops *ops)
 {
 	struct xt_table_info *private;
 	void *loc_cpu_entry;
diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c
index 1897ee1..4c02416 100644
--- a/net/ipv4/netfilter/arptable_filter.c
+++ b/net/ipv4/netfilter/arptable_filter.c
@@ -38,19 +38,20 @@ static struct nf_hook_ops *arpfilter_ops __read_mostly;
 static int __net_init arptable_filter_net_init(struct net *net)
 {
 	struct arpt_replace *repl;
-	
+	int err;
+
 	repl = arpt_alloc_initial_table(&packet_filter);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv4.arptable_filter =
-		arpt_register_table(net, &packet_filter, repl);
+	err = arpt_register_table(net, &packet_filter, repl, arpfilter_ops,
+				  &net->ipv4.arptable_filter);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv4.arptable_filter);
+	return err;
 }
 
 static void __net_exit arptable_filter_net_exit(struct net *net)
 {
-	arpt_unregister_table(net->ipv4.arptable_filter);
+	arpt_unregister_table(net, net->ipv4.arptable_filter, arpfilter_ops);
 }
 
 static struct pernet_operations arptable_filter_net_ops = {
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index b99affa..1eb4fe5 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -2062,9 +2062,9 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 	return ret;
 }
 
-struct xt_table *ipt_register_table(struct net *net,
-				    const struct xt_table *table,
-				    const struct ipt_replace *repl)
+int ipt_register_table(struct net *net, const struct xt_table *table,
+		       const struct ipt_replace *repl,
+		       const struct nf_hook_ops *ops, struct xt_table **res)
 {
 	int ret;
 	struct xt_table_info *newinfo;
@@ -2073,10 +2073,8 @@ struct xt_table *ipt_register_table(struct net *net,
 	struct xt_table *new_table;
 
 	newinfo = xt_alloc_table_info(repl->size);
-	if (!newinfo) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!newinfo)
+		return -ENOMEM;
 
 	loc_cpu_entry = newinfo->entries;
 	memcpy(loc_cpu_entry, repl->entries, repl->size);
@@ -2091,15 +2089,16 @@ struct xt_table *ipt_register_table(struct net *net,
 		goto out_free;
 	}
 
-	return new_table;
+	WRITE_ONCE(*res, new_table);
+	return ret;
 
 out_free:
 	xt_free_table_info(newinfo);
-out:
-	return ERR_PTR(ret);
+	return ret;
 }
 
-void ipt_unregister_table(struct net *net, struct xt_table *table)
+void ipt_unregister_table(struct net *net, struct xt_table *table,
+			  const struct nf_hook_ops *ops)
 {
 	struct xt_table_info *private;
 	void *loc_cpu_entry;
diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c
index 397ef2d..3fbe4ac 100644
--- a/net/ipv4/netfilter/iptable_filter.c
+++ b/net/ipv4/netfilter/iptable_filter.c
@@ -54,6 +54,7 @@ module_param(forward, bool, 0000);
 static int __net_init iptable_filter_net_init(struct net *net)
 {
 	struct ipt_replace *repl;
+	int err;
 
 	repl = ipt_alloc_initial_table(&packet_filter);
 	if (repl == NULL)
@@ -62,15 +63,15 @@ static int __net_init iptable_filter_net_init(struct net *net)
 	((struct ipt_standard *)repl->entries)[1].target.verdict =
 		forward ? -NF_ACCEPT - 1 : -NF_DROP - 1;
 
-	net->ipv4.iptable_filter =
-		ipt_register_table(net, &packet_filter, repl);
+	err = ipt_register_table(net, &packet_filter, repl, filter_ops,
+				 &net->ipv4.iptable_filter);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv4.iptable_filter);
+	return err;
 }
 
 static void __net_exit iptable_filter_net_exit(struct net *net)
 {
-	ipt_unregister_table(net, net->ipv4.iptable_filter);
+	ipt_unregister_table(net, net->ipv4.iptable_filter, filter_ops);
 }
 
 static struct pernet_operations iptable_filter_net_ops = {
diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c
index ba5d392..668e7916 100644
--- a/net/ipv4/netfilter/iptable_mangle.c
+++ b/net/ipv4/netfilter/iptable_mangle.c
@@ -96,19 +96,20 @@ static struct nf_hook_ops *mangle_ops __read_mostly;
 static int __net_init iptable_mangle_net_init(struct net *net)
 {
 	struct ipt_replace *repl;
+	int ret;
 
 	repl = ipt_alloc_initial_table(&packet_mangler);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv4.iptable_mangle =
-		ipt_register_table(net, &packet_mangler, repl);
+	ret = ipt_register_table(net, &packet_mangler, repl, mangle_ops,
+				 &net->ipv4.iptable_mangle);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv4.iptable_mangle);
+	return ret;
 }
 
 static void __net_exit iptable_mangle_net_exit(struct net *net)
 {
-	ipt_unregister_table(net, net->ipv4.iptable_mangle);
+	ipt_unregister_table(net, net->ipv4.iptable_mangle, mangle_ops);
 }
 
 static struct pernet_operations iptable_mangle_net_ops = {
diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c
index ae2cd27..e984f1d 100644
--- a/net/ipv4/netfilter/iptable_nat.c
+++ b/net/ipv4/netfilter/iptable_nat.c
@@ -98,18 +98,20 @@ static struct nf_hook_ops nf_nat_ipv4_ops[] __read_mostly = {
 static int __net_init iptable_nat_net_init(struct net *net)
 {
 	struct ipt_replace *repl;
+	int ret;
 
 	repl = ipt_alloc_initial_table(&nf_nat_ipv4_table);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv4.nat_table = ipt_register_table(net, &nf_nat_ipv4_table, repl);
+	ret = ipt_register_table(net, &nf_nat_ipv4_table, repl,
+				 nf_nat_ipv4_ops, &net->ipv4.nat_table);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv4.nat_table);
+	return ret;
 }
 
 static void __net_exit iptable_nat_net_exit(struct net *net)
 {
-	ipt_unregister_table(net, net->ipv4.nat_table);
+	ipt_unregister_table(net, net->ipv4.nat_table, nf_nat_ipv4_ops);
 }
 
 static struct pernet_operations iptable_nat_net_ops = {
diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c
index 1ba0281..9d78780 100644
--- a/net/ipv4/netfilter/iptable_raw.c
+++ b/net/ipv4/netfilter/iptable_raw.c
@@ -37,19 +37,20 @@ static struct nf_hook_ops *rawtable_ops __read_mostly;
 static int __net_init iptable_raw_net_init(struct net *net)
 {
 	struct ipt_replace *repl;
+	int ret;
 
 	repl = ipt_alloc_initial_table(&packet_raw);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv4.iptable_raw =
-		ipt_register_table(net, &packet_raw, repl);
+	ret = ipt_register_table(net, &packet_raw, repl, rawtable_ops,
+				 &net->ipv4.iptable_raw);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv4.iptable_raw);
+	return ret;
 }
 
 static void __net_exit iptable_raw_net_exit(struct net *net)
 {
-	ipt_unregister_table(net, net->ipv4.iptable_raw);
+	ipt_unregister_table(net, net->ipv4.iptable_raw, rawtable_ops);
 }
 
 static struct pernet_operations iptable_raw_net_ops = {
diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c
index c2e23d5..88bc52f 100644
--- a/net/ipv4/netfilter/iptable_security.c
+++ b/net/ipv4/netfilter/iptable_security.c
@@ -54,19 +54,20 @@ static struct nf_hook_ops *sectbl_ops __read_mostly;
 static int __net_init iptable_security_net_init(struct net *net)
 {
 	struct ipt_replace *repl;
+	int ret;
 
 	repl = ipt_alloc_initial_table(&security_table);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv4.iptable_security =
-		ipt_register_table(net, &security_table, repl);
+	ret = ipt_register_table(net, &security_table, repl, sectbl_ops,
+				 &net->ipv4.iptable_security);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv4.iptable_security);
+	return ret;
 }
 
 static void __net_exit iptable_security_net_exit(struct net *net)
 {
-	ipt_unregister_table(net, net->ipv4.iptable_security);
+	ipt_unregister_table(net, net->ipv4.iptable_security, sectbl_ops);
 }
 
 static struct pernet_operations iptable_security_net_ops = {
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 99425cf..052d744 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -2071,9 +2071,10 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 	return ret;
 }
 
-struct xt_table *ip6t_register_table(struct net *net,
-				     const struct xt_table *table,
-				     const struct ip6t_replace *repl)
+int ip6t_register_table(struct net *net, const struct xt_table *table,
+			const struct ip6t_replace *repl,
+			const struct nf_hook_ops *ops,
+			struct xt_table **res)
 {
 	int ret;
 	struct xt_table_info *newinfo;
@@ -2082,10 +2083,8 @@ struct xt_table *ip6t_register_table(struct net *net,
 	struct xt_table *new_table;
 
 	newinfo = xt_alloc_table_info(repl->size);
-	if (!newinfo) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!newinfo)
+		return -ENOMEM;
 
 	loc_cpu_entry = newinfo->entries;
 	memcpy(loc_cpu_entry, repl->entries, repl->size);
@@ -2099,15 +2098,17 @@ struct xt_table *ip6t_register_table(struct net *net,
 		ret = PTR_ERR(new_table);
 		goto out_free;
 	}
-	return new_table;
+
+	WRITE_ONCE(*res, new_table);
+	return ret;
 
 out_free:
 	xt_free_table_info(newinfo);
-out:
-	return ERR_PTR(ret);
+	return ret;
 }
 
-void ip6t_unregister_table(struct net *net, struct xt_table *table)
+void ip6t_unregister_table(struct net *net, struct xt_table *table,
+			   const struct nf_hook_ops *ops)
 {
 	struct xt_table_info *private;
 	void *loc_cpu_entry;
diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c
index 8b277b9..d191d54 100644
--- a/net/ipv6/netfilter/ip6table_filter.c
+++ b/net/ipv6/netfilter/ip6table_filter.c
@@ -47,6 +47,7 @@ module_param(forward, bool, 0000);
 static int __net_init ip6table_filter_net_init(struct net *net)
 {
 	struct ip6t_replace *repl;
+	int err;
 
 	repl = ip6t_alloc_initial_table(&packet_filter);
 	if (repl == NULL)
@@ -55,15 +56,15 @@ static int __net_init ip6table_filter_net_init(struct net *net)
 	((struct ip6t_standard *)repl->entries)[1].target.verdict =
 		forward ? -NF_ACCEPT - 1 : -NF_DROP - 1;
 
-	net->ipv6.ip6table_filter =
-		ip6t_register_table(net, &packet_filter, repl);
+	err = ip6t_register_table(net, &packet_filter, repl, filter_ops,
+				  &net->ipv6.ip6table_filter);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv6.ip6table_filter);
+	return err;
 }
 
 static void __net_exit ip6table_filter_net_exit(struct net *net)
 {
-	ip6t_unregister_table(net, net->ipv6.ip6table_filter);
+	ip6t_unregister_table(net, net->ipv6.ip6table_filter, filter_ops);
 }
 
 static struct pernet_operations ip6table_filter_net_ops = {
diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c
index abe278b..fe43d08 100644
--- a/net/ipv6/netfilter/ip6table_mangle.c
+++ b/net/ipv6/netfilter/ip6table_mangle.c
@@ -91,19 +91,20 @@ static struct nf_hook_ops *mangle_ops __read_mostly;
 static int __net_init ip6table_mangle_net_init(struct net *net)
 {
 	struct ip6t_replace *repl;
+	int ret;
 
 	repl = ip6t_alloc_initial_table(&packet_mangler);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv6.ip6table_mangle =
-		ip6t_register_table(net, &packet_mangler, repl);
+	ret = ip6t_register_table(net, &packet_mangler, repl, mangle_ops,
+				  &net->ipv6.ip6table_mangle);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv6.ip6table_mangle);
+	return ret;
 }
 
 static void __net_exit ip6table_mangle_net_exit(struct net *net)
 {
-	ip6t_unregister_table(net, net->ipv6.ip6table_mangle);
+	ip6t_unregister_table(net, net->ipv6.ip6table_mangle, mangle_ops);
 }
 
 static struct pernet_operations ip6table_mangle_net_ops = {
diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c
index de2a10a..7f9740e 100644
--- a/net/ipv6/netfilter/ip6table_nat.c
+++ b/net/ipv6/netfilter/ip6table_nat.c
@@ -100,18 +100,20 @@ static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
 static int __net_init ip6table_nat_net_init(struct net *net)
 {
 	struct ip6t_replace *repl;
+	int ret;
 
 	repl = ip6t_alloc_initial_table(&nf_nat_ipv6_table);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv6.ip6table_nat = ip6t_register_table(net, &nf_nat_ipv6_table, repl);
+	ret = ip6t_register_table(net, &nf_nat_ipv6_table, repl,
+				  nf_nat_ipv6_ops, &net->ipv6.ip6table_nat);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv6.ip6table_nat);
+	return ret;
 }
 
 static void __net_exit ip6table_nat_net_exit(struct net *net)
 {
-	ip6t_unregister_table(net, net->ipv6.ip6table_nat);
+	ip6t_unregister_table(net, net->ipv6.ip6table_nat, nf_nat_ipv6_ops);
 }
 
 static struct pernet_operations ip6table_nat_net_ops = {
diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c
index 9021963..5fac433 100644
--- a/net/ipv6/netfilter/ip6table_raw.c
+++ b/net/ipv6/netfilter/ip6table_raw.c
@@ -30,19 +30,20 @@ static struct nf_hook_ops *rawtable_ops __read_mostly;
 static int __net_init ip6table_raw_net_init(struct net *net)
 {
 	struct ip6t_replace *repl;
+	int ret;
 
 	repl = ip6t_alloc_initial_table(&packet_raw);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv6.ip6table_raw =
-		ip6t_register_table(net, &packet_raw, repl);
+	ret = ip6t_register_table(net, &packet_raw, repl, rawtable_ops,
+				  &net->ipv6.ip6table_raw);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv6.ip6table_raw);
+	return ret;
 }
 
 static void __net_exit ip6table_raw_net_exit(struct net *net)
 {
-	ip6t_unregister_table(net, net->ipv6.ip6table_raw);
+	ip6t_unregister_table(net, net->ipv6.ip6table_raw, rawtable_ops);
 }
 
 static struct pernet_operations ip6table_raw_net_ops = {
diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c
index 0d856fe..cf58745 100644
--- a/net/ipv6/netfilter/ip6table_security.c
+++ b/net/ipv6/netfilter/ip6table_security.c
@@ -47,19 +47,20 @@ static struct nf_hook_ops *sectbl_ops __read_mostly;
 static int __net_init ip6table_security_net_init(struct net *net)
 {
 	struct ip6t_replace *repl;
+	int ret;
 
 	repl = ip6t_alloc_initial_table(&security_table);
 	if (repl == NULL)
 		return -ENOMEM;
-	net->ipv6.ip6table_security =
-		ip6t_register_table(net, &security_table, repl);
+	ret = ip6t_register_table(net, &security_table, repl, sectbl_ops,
+				  &net->ipv6.ip6table_security);
 	kfree(repl);
-	return PTR_ERR_OR_ZERO(net->ipv6.ip6table_security);
+	return ret;
 }
 
 static void __net_exit ip6table_security_net_exit(struct net *net)
 {
-	ip6t_unregister_table(net, net->ipv6.ip6table_security);
+	ip6t_unregister_table(net, net->ipv6.ip6table_security, sectbl_ops);
 }
 
 static struct pernet_operations ip6table_security_net_ops = {
-- 
cgit v0.10.2


From b9e69e127397187b70c813a4397cce7afb5e8cb1 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 25 Feb 2016 10:08:36 +0100
Subject: netfilter: xtables: don't hook tables by default

delay hook registration until the table is being requested inside a
namespace.

Historically, a particular table (iptables mangle, ip6tables filter, etc)
was registered on module load.

When netns support was added to iptables only the ip/ip6tables ruleset was
made namespace aware, not the actual hook points.

This means f.e. that when ipt_filter table/module is loaded on a system,
then each namespace on that system has an (empty) iptables filter ruleset.

In other words, if a namespace sends a packet, such skb is 'caught' by
netfilter machinery and fed to hooking points for that table (i.e. INPUT,
FORWARD, etc).

Thanks to Eric Biederman, hooks are no longer global, but per namespace.

This means that we can avoid allocation of empty ruleset in a namespace and
defer hook registration until we need the functionality.

We register a tables hook entry points ONLY in the initial namespace.
When an iptables get/setockopt is issued inside a given namespace, we check
if the table is found in the per-namespace list.

If not, we attempt to find it in the initial namespace, and, if found,
create an empty default table in the requesting namespace and register the
needed hooks.

Hook points are destroyed only once namespace is deleted, there is no
'usage count' (it makes no sense since there is no 'remove table' operation
in xtables api).

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index c557741..80a305b 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -200,6 +200,9 @@ struct xt_table {
 	u_int8_t af;		/* address/protocol family */
 	int priority;		/* hook order */
 
+	/* called when table is needed in the given netns */
+	int (*table_init)(struct net *net);
+
 	/* A unique name... */
 	const char name[XT_TABLE_MAXNAMELEN];
 };
@@ -408,8 +411,7 @@ xt_get_per_cpu_counter(struct xt_counters *cnt, unsigned int cpu)
 	return cnt;
 }
 
-struct nf_hook_ops *xt_hook_link(const struct xt_table *, nf_hookfn *);
-void xt_hook_unlink(const struct xt_table *, struct nf_hook_ops *);
+struct nf_hook_ops *xt_hook_ops_alloc(const struct xt_table *, nf_hookfn *);
 
 #ifdef CONFIG_COMPAT
 #include <net/compat.h>
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index 00eed08..bf08192 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -1780,6 +1780,24 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len
 	return ret;
 }
 
+static void __arpt_unregister_table(struct xt_table *table)
+{
+	struct xt_table_info *private;
+	void *loc_cpu_entry;
+	struct module *table_owner = table->me;
+	struct arpt_entry *iter;
+
+	private = xt_unregister_table(table);
+
+	/* Decrease module usage counts and free resources */
+	loc_cpu_entry = private->entries;
+	xt_entry_foreach(iter, loc_cpu_entry, private->size)
+		cleanup_entry(iter);
+	if (private->number > private->initial_entries)
+		module_put(table_owner);
+	xt_free_table_info(private);
+}
+
 int arpt_register_table(struct net *net,
 			const struct xt_table *table,
 			const struct arpt_replace *repl,
@@ -1810,8 +1828,15 @@ int arpt_register_table(struct net *net,
 		goto out_free;
 	}
 
+	/* set res now, will see skbs right after nf_register_net_hooks */
 	WRITE_ONCE(*res, new_table);
 
+	ret = nf_register_net_hooks(net, ops, hweight32(table->valid_hooks));
+	if (ret != 0) {
+		__arpt_unregister_table(new_table);
+		*res = NULL;
+	}
+
 	return ret;
 
 out_free:
@@ -1822,20 +1847,8 @@ out_free:
 void arpt_unregister_table(struct net *net, struct xt_table *table,
 			   const struct nf_hook_ops *ops)
 {
-	struct xt_table_info *private;
-	void *loc_cpu_entry;
-	struct module *table_owner = table->me;
-	struct arpt_entry *iter;
-
-	private = xt_unregister_table(table);
-
-	/* Decrease module usage counts and free resources */
-	loc_cpu_entry = private->entries;
-	xt_entry_foreach(iter, loc_cpu_entry, private->size)
-		cleanup_entry(iter);
-	if (private->number > private->initial_entries)
-		module_put(table_owner);
-	xt_free_table_info(private);
+	nf_unregister_net_hooks(net, ops, hweight32(table->valid_hooks));
+	__arpt_unregister_table(table);
 }
 
 /* The built-in targets: standard (NULL) and error. */
diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c
index 4c02416..dd8c80d 100644
--- a/net/ipv4/netfilter/arptable_filter.c
+++ b/net/ipv4/netfilter/arptable_filter.c
@@ -17,12 +17,15 @@ MODULE_DESCRIPTION("arptables filter table");
 #define FILTER_VALID_HOOKS ((1 << NF_ARP_IN) | (1 << NF_ARP_OUT) | \
 			   (1 << NF_ARP_FORWARD))
 
+static int __net_init arptable_filter_table_init(struct net *net);
+
 static const struct xt_table packet_filter = {
 	.name		= "filter",
 	.valid_hooks	= FILTER_VALID_HOOKS,
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_ARP,
 	.priority	= NF_IP_PRI_FILTER,
+	.table_init	= arptable_filter_table_init,
 };
 
 /* The work comes in here from netfilter.c */
@@ -35,11 +38,14 @@ arptable_filter_hook(void *priv, struct sk_buff *skb,
 
 static struct nf_hook_ops *arpfilter_ops __read_mostly;
 
-static int __net_init arptable_filter_net_init(struct net *net)
+static int __net_init arptable_filter_table_init(struct net *net)
 {
 	struct arpt_replace *repl;
 	int err;
 
+	if (net->ipv4.arptable_filter)
+		return 0;
+
 	repl = arpt_alloc_initial_table(&packet_filter);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -51,11 +57,13 @@ static int __net_init arptable_filter_net_init(struct net *net)
 
 static void __net_exit arptable_filter_net_exit(struct net *net)
 {
+	if (!net->ipv4.arptable_filter)
+		return;
 	arpt_unregister_table(net, net->ipv4.arptable_filter, arpfilter_ops);
+	net->ipv4.arptable_filter = NULL;
 }
 
 static struct pernet_operations arptable_filter_net_ops = {
-	.init = arptable_filter_net_init,
 	.exit = arptable_filter_net_exit,
 };
 
@@ -63,26 +71,23 @@ static int __init arptable_filter_init(void)
 {
 	int ret;
 
+	arpfilter_ops = xt_hook_ops_alloc(&packet_filter, arptable_filter_hook);
+	if (IS_ERR(arpfilter_ops))
+		return PTR_ERR(arpfilter_ops);
+
 	ret = register_pernet_subsys(&arptable_filter_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(arpfilter_ops);
 		return ret;
-
-	arpfilter_ops = xt_hook_link(&packet_filter, arptable_filter_hook);
-	if (IS_ERR(arpfilter_ops)) {
-		ret = PTR_ERR(arpfilter_ops);
-		goto cleanup_table;
 	}
-	return ret;
 
-cleanup_table:
-	unregister_pernet_subsys(&arptable_filter_net_ops);
 	return ret;
 }
 
 static void __exit arptable_filter_fini(void)
 {
-	xt_hook_unlink(&packet_filter, arpfilter_ops);
 	unregister_pernet_subsys(&arptable_filter_net_ops);
+	kfree(arpfilter_ops);
 }
 
 module_init(arptable_filter_init);
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 1eb4fe5..e53f8d6 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -2062,6 +2062,24 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 	return ret;
 }
 
+static void __ipt_unregister_table(struct net *net, struct xt_table *table)
+{
+	struct xt_table_info *private;
+	void *loc_cpu_entry;
+	struct module *table_owner = table->me;
+	struct ipt_entry *iter;
+
+	private = xt_unregister_table(table);
+
+	/* Decrease module usage counts and free resources */
+	loc_cpu_entry = private->entries;
+	xt_entry_foreach(iter, loc_cpu_entry, private->size)
+		cleanup_entry(iter, net);
+	if (private->number > private->initial_entries)
+		module_put(table_owner);
+	xt_free_table_info(private);
+}
+
 int ipt_register_table(struct net *net, const struct xt_table *table,
 		       const struct ipt_replace *repl,
 		       const struct nf_hook_ops *ops, struct xt_table **res)
@@ -2089,7 +2107,15 @@ int ipt_register_table(struct net *net, const struct xt_table *table,
 		goto out_free;
 	}
 
+	/* set res now, will see skbs right after nf_register_net_hooks */
 	WRITE_ONCE(*res, new_table);
+
+	ret = nf_register_net_hooks(net, ops, hweight32(table->valid_hooks));
+	if (ret != 0) {
+		__ipt_unregister_table(net, new_table);
+		*res = NULL;
+	}
+
 	return ret;
 
 out_free:
@@ -2100,20 +2126,8 @@ out_free:
 void ipt_unregister_table(struct net *net, struct xt_table *table,
 			  const struct nf_hook_ops *ops)
 {
-	struct xt_table_info *private;
-	void *loc_cpu_entry;
-	struct module *table_owner = table->me;
-	struct ipt_entry *iter;
-
-	private = xt_unregister_table(table);
-
-	/* Decrease module usage counts and free resources */
-	loc_cpu_entry = private->entries;
-	xt_entry_foreach(iter, loc_cpu_entry, private->size)
-		cleanup_entry(iter, net);
-	if (private->number > private->initial_entries)
-		module_put(table_owner);
-	xt_free_table_info(private);
+	nf_unregister_net_hooks(net, ops, hweight32(table->valid_hooks));
+	__ipt_unregister_table(net, table);
 }
 
 /* Returns 1 if the type and code is matched by the range, 0 otherwise */
diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c
index 3fbe4ac..7667f22 100644
--- a/net/ipv4/netfilter/iptable_filter.c
+++ b/net/ipv4/netfilter/iptable_filter.c
@@ -23,6 +23,7 @@ MODULE_DESCRIPTION("iptables filter table");
 #define FILTER_VALID_HOOKS ((1 << NF_INET_LOCAL_IN) | \
 			    (1 << NF_INET_FORWARD) | \
 			    (1 << NF_INET_LOCAL_OUT))
+static int __net_init iptable_filter_table_init(struct net *net);
 
 static const struct xt_table packet_filter = {
 	.name		= "filter",
@@ -30,6 +31,7 @@ static const struct xt_table packet_filter = {
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV4,
 	.priority	= NF_IP_PRI_FILTER,
+	.table_init	= iptable_filter_table_init,
 };
 
 static unsigned int
@@ -48,14 +50,17 @@ iptable_filter_hook(void *priv, struct sk_buff *skb,
 static struct nf_hook_ops *filter_ops __read_mostly;
 
 /* Default to forward because I got too much mail already. */
-static bool forward = true;
+static bool forward __read_mostly = true;
 module_param(forward, bool, 0000);
 
-static int __net_init iptable_filter_net_init(struct net *net)
+static int __net_init iptable_filter_table_init(struct net *net)
 {
 	struct ipt_replace *repl;
 	int err;
 
+	if (net->ipv4.iptable_filter)
+		return 0;
+
 	repl = ipt_alloc_initial_table(&packet_filter);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -69,9 +74,20 @@ static int __net_init iptable_filter_net_init(struct net *net)
 	return err;
 }
 
+static int __net_init iptable_filter_net_init(struct net *net)
+{
+	if (net == &init_net || !forward)
+		return iptable_filter_table_init(net);
+
+	return 0;
+}
+
 static void __net_exit iptable_filter_net_exit(struct net *net)
 {
+	if (!net->ipv4.iptable_filter)
+		return;
 	ipt_unregister_table(net, net->ipv4.iptable_filter, filter_ops);
+	net->ipv4.iptable_filter = NULL;
 }
 
 static struct pernet_operations iptable_filter_net_ops = {
@@ -83,24 +99,21 @@ static int __init iptable_filter_init(void)
 {
 	int ret;
 
+	filter_ops = xt_hook_ops_alloc(&packet_filter, iptable_filter_hook);
+	if (IS_ERR(filter_ops))
+		return PTR_ERR(filter_ops);
+
 	ret = register_pernet_subsys(&iptable_filter_net_ops);
 	if (ret < 0)
-		return ret;
-
-	/* Register hooks */
-	filter_ops = xt_hook_link(&packet_filter, iptable_filter_hook);
-	if (IS_ERR(filter_ops)) {
-		ret = PTR_ERR(filter_ops);
-		unregister_pernet_subsys(&iptable_filter_net_ops);
-	}
+		kfree(filter_ops);
 
 	return ret;
 }
 
 static void __exit iptable_filter_fini(void)
 {
-	xt_hook_unlink(&packet_filter, filter_ops);
 	unregister_pernet_subsys(&iptable_filter_net_ops);
+	kfree(filter_ops);
 }
 
 module_init(iptable_filter_init);
diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c
index 668e7916..57fc97c 100644
--- a/net/ipv4/netfilter/iptable_mangle.c
+++ b/net/ipv4/netfilter/iptable_mangle.c
@@ -28,12 +28,15 @@ MODULE_DESCRIPTION("iptables mangle table");
 			    (1 << NF_INET_LOCAL_OUT) | \
 			    (1 << NF_INET_POST_ROUTING))
 
+static int __net_init iptable_mangle_table_init(struct net *net);
+
 static const struct xt_table packet_mangler = {
 	.name		= "mangle",
 	.valid_hooks	= MANGLE_VALID_HOOKS,
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV4,
 	.priority	= NF_IP_PRI_MANGLE,
+	.table_init	= iptable_mangle_table_init,
 };
 
 static unsigned int
@@ -92,12 +95,14 @@ iptable_mangle_hook(void *priv,
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
-
-static int __net_init iptable_mangle_net_init(struct net *net)
+static int __net_init iptable_mangle_table_init(struct net *net)
 {
 	struct ipt_replace *repl;
 	int ret;
 
+	if (net->ipv4.iptable_mangle)
+		return 0;
+
 	repl = ipt_alloc_initial_table(&packet_mangler);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -109,11 +114,13 @@ static int __net_init iptable_mangle_net_init(struct net *net)
 
 static void __net_exit iptable_mangle_net_exit(struct net *net)
 {
+	if (!net->ipv4.iptable_mangle)
+		return;
 	ipt_unregister_table(net, net->ipv4.iptable_mangle, mangle_ops);
+	net->ipv4.iptable_mangle = NULL;
 }
 
 static struct pernet_operations iptable_mangle_net_ops = {
-	.init = iptable_mangle_net_init,
 	.exit = iptable_mangle_net_exit,
 };
 
@@ -121,15 +128,22 @@ static int __init iptable_mangle_init(void)
 {
 	int ret;
 
+	mangle_ops = xt_hook_ops_alloc(&packet_mangler, iptable_mangle_hook);
+	if (IS_ERR(mangle_ops)) {
+		ret = PTR_ERR(mangle_ops);
+		return ret;
+	}
+
 	ret = register_pernet_subsys(&iptable_mangle_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(mangle_ops);
 		return ret;
+	}
 
-	/* Register hooks */
-	mangle_ops = xt_hook_link(&packet_mangler, iptable_mangle_hook);
-	if (IS_ERR(mangle_ops)) {
-		ret = PTR_ERR(mangle_ops);
+	ret = iptable_mangle_table_init(&init_net);
+	if (ret) {
 		unregister_pernet_subsys(&iptable_mangle_net_ops);
+		kfree(mangle_ops);
 	}
 
 	return ret;
@@ -137,8 +151,8 @@ static int __init iptable_mangle_init(void)
 
 static void __exit iptable_mangle_fini(void)
 {
-	xt_hook_unlink(&packet_mangler, mangle_ops);
 	unregister_pernet_subsys(&iptable_mangle_net_ops);
+	kfree(mangle_ops);
 }
 
 module_init(iptable_mangle_init);
diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c
index e984f1d..138a24b 100644
--- a/net/ipv4/netfilter/iptable_nat.c
+++ b/net/ipv4/netfilter/iptable_nat.c
@@ -18,6 +18,8 @@
 #include <net/netfilter/nf_nat_core.h>
 #include <net/netfilter/nf_nat_l3proto.h>
 
+static int __net_init iptable_nat_table_init(struct net *net);
+
 static const struct xt_table nf_nat_ipv4_table = {
 	.name		= "nat",
 	.valid_hooks	= (1 << NF_INET_PRE_ROUTING) |
@@ -26,6 +28,7 @@ static const struct xt_table nf_nat_ipv4_table = {
 			  (1 << NF_INET_LOCAL_IN),
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV4,
+	.table_init	= iptable_nat_table_init,
 };
 
 static unsigned int iptable_nat_do_chain(void *priv,
@@ -95,11 +98,14 @@ static struct nf_hook_ops nf_nat_ipv4_ops[] __read_mostly = {
 	},
 };
 
-static int __net_init iptable_nat_net_init(struct net *net)
+static int __net_init iptable_nat_table_init(struct net *net)
 {
 	struct ipt_replace *repl;
 	int ret;
 
+	if (net->ipv4.nat_table)
+		return 0;
+
 	repl = ipt_alloc_initial_table(&nf_nat_ipv4_table);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -111,36 +117,31 @@ static int __net_init iptable_nat_net_init(struct net *net)
 
 static void __net_exit iptable_nat_net_exit(struct net *net)
 {
+	if (!net->ipv4.nat_table)
+		return;
 	ipt_unregister_table(net, net->ipv4.nat_table, nf_nat_ipv4_ops);
+	net->ipv4.nat_table = NULL;
 }
 
 static struct pernet_operations iptable_nat_net_ops = {
-	.init	= iptable_nat_net_init,
 	.exit	= iptable_nat_net_exit,
 };
 
 static int __init iptable_nat_init(void)
 {
-	int err;
-
-	err = register_pernet_subsys(&iptable_nat_net_ops);
-	if (err < 0)
-		goto err1;
+	int ret = register_pernet_subsys(&iptable_nat_net_ops);
 
-	err = nf_register_hooks(nf_nat_ipv4_ops, ARRAY_SIZE(nf_nat_ipv4_ops));
-	if (err < 0)
-		goto err2;
-	return 0;
+	if (ret)
+		return ret;
 
-err2:
-	unregister_pernet_subsys(&iptable_nat_net_ops);
-err1:
-	return err;
+	ret = iptable_nat_table_init(&init_net);
+	if (ret)
+		unregister_pernet_subsys(&iptable_nat_net_ops);
+	return ret;
 }
 
 static void __exit iptable_nat_exit(void)
 {
-	nf_unregister_hooks(nf_nat_ipv4_ops, ARRAY_SIZE(nf_nat_ipv4_ops));
 	unregister_pernet_subsys(&iptable_nat_net_ops);
 }
 
diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c
index 9d78780..2642ecd 100644
--- a/net/ipv4/netfilter/iptable_raw.c
+++ b/net/ipv4/netfilter/iptable_raw.c
@@ -10,12 +10,15 @@
 
 #define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT))
 
+static int __net_init iptable_raw_table_init(struct net *net);
+
 static const struct xt_table packet_raw = {
 	.name = "raw",
 	.valid_hooks =  RAW_VALID_HOOKS,
 	.me = THIS_MODULE,
 	.af = NFPROTO_IPV4,
 	.priority = NF_IP_PRI_RAW,
+	.table_init = iptable_raw_table_init,
 };
 
 /* The work comes in here from netfilter.c. */
@@ -34,11 +37,14 @@ iptable_raw_hook(void *priv, struct sk_buff *skb,
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
 
-static int __net_init iptable_raw_net_init(struct net *net)
+static int __net_init iptable_raw_table_init(struct net *net)
 {
 	struct ipt_replace *repl;
 	int ret;
 
+	if (net->ipv4.iptable_raw)
+		return 0;
+
 	repl = ipt_alloc_initial_table(&packet_raw);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -50,11 +56,13 @@ static int __net_init iptable_raw_net_init(struct net *net)
 
 static void __net_exit iptable_raw_net_exit(struct net *net)
 {
+	if (!net->ipv4.iptable_raw)
+		return;
 	ipt_unregister_table(net, net->ipv4.iptable_raw, rawtable_ops);
+	net->ipv4.iptable_raw = NULL;
 }
 
 static struct pernet_operations iptable_raw_net_ops = {
-	.init = iptable_raw_net_init,
 	.exit = iptable_raw_net_exit,
 };
 
@@ -62,15 +70,20 @@ static int __init iptable_raw_init(void)
 {
 	int ret;
 
+	rawtable_ops = xt_hook_ops_alloc(&packet_raw, iptable_raw_hook);
+	if (IS_ERR(rawtable_ops))
+		return PTR_ERR(rawtable_ops);
+
 	ret = register_pernet_subsys(&iptable_raw_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(rawtable_ops);
 		return ret;
+	}
 
-	/* Register hooks */
-	rawtable_ops = xt_hook_link(&packet_raw, iptable_raw_hook);
-	if (IS_ERR(rawtable_ops)) {
-		ret = PTR_ERR(rawtable_ops);
+	ret = iptable_raw_table_init(&init_net);
+	if (ret) {
 		unregister_pernet_subsys(&iptable_raw_net_ops);
+		kfree(rawtable_ops);
 	}
 
 	return ret;
@@ -78,8 +91,8 @@ static int __init iptable_raw_init(void)
 
 static void __exit iptable_raw_fini(void)
 {
-	xt_hook_unlink(&packet_raw, rawtable_ops);
 	unregister_pernet_subsys(&iptable_raw_net_ops);
+	kfree(rawtable_ops);
 }
 
 module_init(iptable_raw_init);
diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c
index 88bc52f..ff22659 100644
--- a/net/ipv4/netfilter/iptable_security.c
+++ b/net/ipv4/netfilter/iptable_security.c
@@ -28,12 +28,15 @@ MODULE_DESCRIPTION("iptables security table, for MAC rules");
 				(1 << NF_INET_FORWARD) | \
 				(1 << NF_INET_LOCAL_OUT)
 
+static int __net_init iptable_security_table_init(struct net *net);
+
 static const struct xt_table security_table = {
 	.name		= "security",
 	.valid_hooks	= SECURITY_VALID_HOOKS,
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV4,
 	.priority	= NF_IP_PRI_SECURITY,
+	.table_init	= iptable_security_table_init,
 };
 
 static unsigned int
@@ -51,11 +54,14 @@ iptable_security_hook(void *priv, struct sk_buff *skb,
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
 
-static int __net_init iptable_security_net_init(struct net *net)
+static int __net_init iptable_security_table_init(struct net *net)
 {
 	struct ipt_replace *repl;
 	int ret;
 
+	if (net->ipv4.iptable_security)
+		return 0;
+
 	repl = ipt_alloc_initial_table(&security_table);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -67,11 +73,14 @@ static int __net_init iptable_security_net_init(struct net *net)
 
 static void __net_exit iptable_security_net_exit(struct net *net)
 {
+	if (!net->ipv4.iptable_security)
+		return;
+
 	ipt_unregister_table(net, net->ipv4.iptable_security, sectbl_ops);
+	net->ipv4.iptable_security = NULL;
 }
 
 static struct pernet_operations iptable_security_net_ops = {
-	.init = iptable_security_net_init,
 	.exit = iptable_security_net_exit,
 };
 
@@ -79,27 +88,29 @@ static int __init iptable_security_init(void)
 {
 	int ret;
 
+	sectbl_ops = xt_hook_ops_alloc(&security_table, iptable_security_hook);
+	if (IS_ERR(sectbl_ops))
+		return PTR_ERR(sectbl_ops);
+
 	ret = register_pernet_subsys(&iptable_security_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(sectbl_ops);
 		return ret;
-
-	sectbl_ops = xt_hook_link(&security_table, iptable_security_hook);
-	if (IS_ERR(sectbl_ops)) {
-		ret = PTR_ERR(sectbl_ops);
-		goto cleanup_table;
 	}
 
-	return ret;
+	ret = iptable_security_table_init(&init_net);
+	if (ret) {
+		unregister_pernet_subsys(&iptable_security_net_ops);
+		kfree(sectbl_ops);
+	}
 
-cleanup_table:
-	unregister_pernet_subsys(&iptable_security_net_ops);
 	return ret;
 }
 
 static void __exit iptable_security_fini(void)
 {
-	xt_hook_unlink(&security_table, sectbl_ops);
 	unregister_pernet_subsys(&iptable_security_net_ops);
+	kfree(sectbl_ops);
 }
 
 module_init(iptable_security_init);
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 052d744..84f9baf 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -2071,6 +2071,24 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 	return ret;
 }
 
+static void __ip6t_unregister_table(struct net *net, struct xt_table *table)
+{
+	struct xt_table_info *private;
+	void *loc_cpu_entry;
+	struct module *table_owner = table->me;
+	struct ip6t_entry *iter;
+
+	private = xt_unregister_table(table);
+
+	/* Decrease module usage counts and free resources */
+	loc_cpu_entry = private->entries;
+	xt_entry_foreach(iter, loc_cpu_entry, private->size)
+		cleanup_entry(iter, net);
+	if (private->number > private->initial_entries)
+		module_put(table_owner);
+	xt_free_table_info(private);
+}
+
 int ip6t_register_table(struct net *net, const struct xt_table *table,
 			const struct ip6t_replace *repl,
 			const struct nf_hook_ops *ops,
@@ -2099,7 +2117,15 @@ int ip6t_register_table(struct net *net, const struct xt_table *table,
 		goto out_free;
 	}
 
+	/* set res now, will see skbs right after nf_register_net_hooks */
 	WRITE_ONCE(*res, new_table);
+
+	ret = nf_register_net_hooks(net, ops, hweight32(table->valid_hooks));
+	if (ret != 0) {
+		__ip6t_unregister_table(net, new_table);
+		*res = NULL;
+	}
+
 	return ret;
 
 out_free:
@@ -2110,20 +2136,8 @@ out_free:
 void ip6t_unregister_table(struct net *net, struct xt_table *table,
 			   const struct nf_hook_ops *ops)
 {
-	struct xt_table_info *private;
-	void *loc_cpu_entry;
-	struct module *table_owner = table->me;
-	struct ip6t_entry *iter;
-
-	private = xt_unregister_table(table);
-
-	/* Decrease module usage counts and free resources */
-	loc_cpu_entry = private->entries;
-	xt_entry_foreach(iter, loc_cpu_entry, private->size)
-		cleanup_entry(iter, net);
-	if (private->number > private->initial_entries)
-		module_put(table_owner);
-	xt_free_table_info(private);
+	nf_unregister_net_hooks(net, ops, hweight32(table->valid_hooks));
+	__ip6t_unregister_table(net, table);
 }
 
 /* Returns 1 if the type and code is matched by the range, 0 otherwise */
diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c
index d191d54..1343077 100644
--- a/net/ipv6/netfilter/ip6table_filter.c
+++ b/net/ipv6/netfilter/ip6table_filter.c
@@ -22,12 +22,15 @@ MODULE_DESCRIPTION("ip6tables filter table");
 			    (1 << NF_INET_FORWARD) | \
 			    (1 << NF_INET_LOCAL_OUT))
 
+static int __net_init ip6table_filter_table_init(struct net *net);
+
 static const struct xt_table packet_filter = {
 	.name		= "filter",
 	.valid_hooks	= FILTER_VALID_HOOKS,
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV6,
 	.priority	= NF_IP6_PRI_FILTER,
+	.table_init	= ip6table_filter_table_init,
 };
 
 /* The work comes in here from netfilter.c. */
@@ -44,11 +47,14 @@ static struct nf_hook_ops *filter_ops __read_mostly;
 static bool forward = true;
 module_param(forward, bool, 0000);
 
-static int __net_init ip6table_filter_net_init(struct net *net)
+static int __net_init ip6table_filter_table_init(struct net *net)
 {
 	struct ip6t_replace *repl;
 	int err;
 
+	if (net->ipv6.ip6table_filter)
+		return 0;
+
 	repl = ip6t_alloc_initial_table(&packet_filter);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -62,9 +68,20 @@ static int __net_init ip6table_filter_net_init(struct net *net)
 	return err;
 }
 
+static int __net_init ip6table_filter_net_init(struct net *net)
+{
+	if (net == &init_net || !forward)
+		return ip6table_filter_table_init(net);
+
+	return 0;
+}
+
 static void __net_exit ip6table_filter_net_exit(struct net *net)
 {
+	if (!net->ipv6.ip6table_filter)
+		return;
 	ip6t_unregister_table(net, net->ipv6.ip6table_filter, filter_ops);
+	net->ipv6.ip6table_filter = NULL;
 }
 
 static struct pernet_operations ip6table_filter_net_ops = {
@@ -76,28 +93,21 @@ static int __init ip6table_filter_init(void)
 {
 	int ret;
 
+	filter_ops = xt_hook_ops_alloc(&packet_filter, ip6table_filter_hook);
+	if (IS_ERR(filter_ops))
+		return PTR_ERR(filter_ops);
+
 	ret = register_pernet_subsys(&ip6table_filter_net_ops);
 	if (ret < 0)
-		return ret;
-
-	/* Register hooks */
-	filter_ops = xt_hook_link(&packet_filter, ip6table_filter_hook);
-	if (IS_ERR(filter_ops)) {
-		ret = PTR_ERR(filter_ops);
-		goto cleanup_table;
-	}
+		kfree(filter_ops);
 
 	return ret;
-
- cleanup_table:
-	unregister_pernet_subsys(&ip6table_filter_net_ops);
-	return ret;
 }
 
 static void __exit ip6table_filter_fini(void)
 {
-	xt_hook_unlink(&packet_filter, filter_ops);
 	unregister_pernet_subsys(&ip6table_filter_net_ops);
+	kfree(filter_ops);
 }
 
 module_init(ip6table_filter_init);
diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c
index fe43d08..cb2b288 100644
--- a/net/ipv6/netfilter/ip6table_mangle.c
+++ b/net/ipv6/netfilter/ip6table_mangle.c
@@ -23,12 +23,15 @@ MODULE_DESCRIPTION("ip6tables mangle table");
 			    (1 << NF_INET_LOCAL_OUT) | \
 			    (1 << NF_INET_POST_ROUTING))
 
+static int __net_init ip6table_mangle_table_init(struct net *net);
+
 static const struct xt_table packet_mangler = {
 	.name		= "mangle",
 	.valid_hooks	= MANGLE_VALID_HOOKS,
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV6,
 	.priority	= NF_IP6_PRI_MANGLE,
+	.table_init	= ip6table_mangle_table_init,
 };
 
 static unsigned int
@@ -88,11 +91,14 @@ ip6table_mangle_hook(void *priv, struct sk_buff *skb,
 }
 
 static struct nf_hook_ops *mangle_ops __read_mostly;
-static int __net_init ip6table_mangle_net_init(struct net *net)
+static int __net_init ip6table_mangle_table_init(struct net *net)
 {
 	struct ip6t_replace *repl;
 	int ret;
 
+	if (net->ipv6.ip6table_mangle)
+		return 0;
+
 	repl = ip6t_alloc_initial_table(&packet_mangler);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -104,11 +110,14 @@ static int __net_init ip6table_mangle_net_init(struct net *net)
 
 static void __net_exit ip6table_mangle_net_exit(struct net *net)
 {
+	if (!net->ipv6.ip6table_mangle)
+		return;
+
 	ip6t_unregister_table(net, net->ipv6.ip6table_mangle, mangle_ops);
+	net->ipv6.ip6table_mangle = NULL;
 }
 
 static struct pernet_operations ip6table_mangle_net_ops = {
-	.init = ip6table_mangle_net_init,
 	.exit = ip6table_mangle_net_exit,
 };
 
@@ -116,28 +125,28 @@ static int __init ip6table_mangle_init(void)
 {
 	int ret;
 
+	mangle_ops = xt_hook_ops_alloc(&packet_mangler, ip6table_mangle_hook);
+	if (IS_ERR(mangle_ops))
+		return PTR_ERR(mangle_ops);
+
 	ret = register_pernet_subsys(&ip6table_mangle_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(mangle_ops);
 		return ret;
-
-	/* Register hooks */
-	mangle_ops = xt_hook_link(&packet_mangler, ip6table_mangle_hook);
-	if (IS_ERR(mangle_ops)) {
-		ret = PTR_ERR(mangle_ops);
-		goto cleanup_table;
 	}
 
-	return ret;
-
- cleanup_table:
-	unregister_pernet_subsys(&ip6table_mangle_net_ops);
+	ret = ip6table_mangle_table_init(&init_net);
+	if (ret) {
+		unregister_pernet_subsys(&ip6table_mangle_net_ops);
+		kfree(mangle_ops);
+	}
 	return ret;
 }
 
 static void __exit ip6table_mangle_fini(void)
 {
-	xt_hook_unlink(&packet_mangler, mangle_ops);
 	unregister_pernet_subsys(&ip6table_mangle_net_ops);
+	kfree(mangle_ops);
 }
 
 module_init(ip6table_mangle_init);
diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c
index 7f9740e..7d2bd94 100644
--- a/net/ipv6/netfilter/ip6table_nat.c
+++ b/net/ipv6/netfilter/ip6table_nat.c
@@ -20,6 +20,8 @@
 #include <net/netfilter/nf_nat_core.h>
 #include <net/netfilter/nf_nat_l3proto.h>
 
+static int __net_init ip6table_nat_table_init(struct net *net);
+
 static const struct xt_table nf_nat_ipv6_table = {
 	.name		= "nat",
 	.valid_hooks	= (1 << NF_INET_PRE_ROUTING) |
@@ -28,6 +30,7 @@ static const struct xt_table nf_nat_ipv6_table = {
 			  (1 << NF_INET_LOCAL_IN),
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV6,
+	.table_init	= ip6table_nat_table_init,
 };
 
 static unsigned int ip6table_nat_do_chain(void *priv,
@@ -97,11 +100,14 @@ static struct nf_hook_ops nf_nat_ipv6_ops[] __read_mostly = {
 	},
 };
 
-static int __net_init ip6table_nat_net_init(struct net *net)
+static int __net_init ip6table_nat_table_init(struct net *net)
 {
 	struct ip6t_replace *repl;
 	int ret;
 
+	if (net->ipv6.ip6table_nat)
+		return 0;
+
 	repl = ip6t_alloc_initial_table(&nf_nat_ipv6_table);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -113,36 +119,31 @@ static int __net_init ip6table_nat_net_init(struct net *net)
 
 static void __net_exit ip6table_nat_net_exit(struct net *net)
 {
+	if (!net->ipv6.ip6table_nat)
+		return;
 	ip6t_unregister_table(net, net->ipv6.ip6table_nat, nf_nat_ipv6_ops);
+	net->ipv6.ip6table_nat = NULL;
 }
 
 static struct pernet_operations ip6table_nat_net_ops = {
-	.init	= ip6table_nat_net_init,
 	.exit	= ip6table_nat_net_exit,
 };
 
 static int __init ip6table_nat_init(void)
 {
-	int err;
-
-	err = register_pernet_subsys(&ip6table_nat_net_ops);
-	if (err < 0)
-		goto err1;
+	int ret = register_pernet_subsys(&ip6table_nat_net_ops);
 
-	err = nf_register_hooks(nf_nat_ipv6_ops, ARRAY_SIZE(nf_nat_ipv6_ops));
-	if (err < 0)
-		goto err2;
-	return 0;
+	if (ret)
+		return ret;
 
-err2:
-	unregister_pernet_subsys(&ip6table_nat_net_ops);
-err1:
-	return err;
+	ret = ip6table_nat_table_init(&init_net);
+	if (ret)
+		unregister_pernet_subsys(&ip6table_nat_net_ops);
+	return ret;
 }
 
 static void __exit ip6table_nat_exit(void)
 {
-	nf_unregister_hooks(nf_nat_ipv6_ops, ARRAY_SIZE(nf_nat_ipv6_ops));
 	unregister_pernet_subsys(&ip6table_nat_net_ops);
 }
 
diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c
index 5fac433..d4bc564 100644
--- a/net/ipv6/netfilter/ip6table_raw.c
+++ b/net/ipv6/netfilter/ip6table_raw.c
@@ -9,12 +9,15 @@
 
 #define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT))
 
+static int __net_init ip6table_raw_table_init(struct net *net);
+
 static const struct xt_table packet_raw = {
 	.name = "raw",
 	.valid_hooks = RAW_VALID_HOOKS,
 	.me = THIS_MODULE,
 	.af = NFPROTO_IPV6,
 	.priority = NF_IP6_PRI_RAW,
+	.table_init = ip6table_raw_table_init,
 };
 
 /* The work comes in here from netfilter.c. */
@@ -27,11 +30,14 @@ ip6table_raw_hook(void *priv, struct sk_buff *skb,
 
 static struct nf_hook_ops *rawtable_ops __read_mostly;
 
-static int __net_init ip6table_raw_net_init(struct net *net)
+static int __net_init ip6table_raw_table_init(struct net *net)
 {
 	struct ip6t_replace *repl;
 	int ret;
 
+	if (net->ipv6.ip6table_raw)
+		return 0;
+
 	repl = ip6t_alloc_initial_table(&packet_raw);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -43,11 +49,13 @@ static int __net_init ip6table_raw_net_init(struct net *net)
 
 static void __net_exit ip6table_raw_net_exit(struct net *net)
 {
+	if (!net->ipv6.ip6table_raw)
+		return;
 	ip6t_unregister_table(net, net->ipv6.ip6table_raw, rawtable_ops);
+	net->ipv6.ip6table_raw = NULL;
 }
 
 static struct pernet_operations ip6table_raw_net_ops = {
-	.init = ip6table_raw_net_init,
 	.exit = ip6table_raw_net_exit,
 };
 
@@ -55,28 +63,29 @@ static int __init ip6table_raw_init(void)
 {
 	int ret;
 
+	/* Register hooks */
+	rawtable_ops = xt_hook_ops_alloc(&packet_raw, ip6table_raw_hook);
+	if (IS_ERR(rawtable_ops))
+		return PTR_ERR(rawtable_ops);
+
 	ret = register_pernet_subsys(&ip6table_raw_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(rawtable_ops);
 		return ret;
-
-	/* Register hooks */
-	rawtable_ops = xt_hook_link(&packet_raw, ip6table_raw_hook);
-	if (IS_ERR(rawtable_ops)) {
-		ret = PTR_ERR(rawtable_ops);
-		goto cleanup_table;
 	}
 
-	return ret;
-
- cleanup_table:
-	unregister_pernet_subsys(&ip6table_raw_net_ops);
+	ret = ip6table_raw_table_init(&init_net);
+	if (ret) {
+		unregister_pernet_subsys(&ip6table_raw_net_ops);
+		kfree(rawtable_ops);
+	}
 	return ret;
 }
 
 static void __exit ip6table_raw_fini(void)
 {
-	xt_hook_unlink(&packet_raw, rawtable_ops);
 	unregister_pernet_subsys(&ip6table_raw_net_ops);
+	kfree(rawtable_ops);
 }
 
 module_init(ip6table_raw_init);
diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c
index cf58745..cf26ccb 100644
--- a/net/ipv6/netfilter/ip6table_security.c
+++ b/net/ipv6/netfilter/ip6table_security.c
@@ -27,12 +27,15 @@ MODULE_DESCRIPTION("ip6tables security table, for MAC rules");
 				(1 << NF_INET_FORWARD) | \
 				(1 << NF_INET_LOCAL_OUT)
 
+static int __net_init ip6table_security_table_init(struct net *net);
+
 static const struct xt_table security_table = {
 	.name		= "security",
 	.valid_hooks	= SECURITY_VALID_HOOKS,
 	.me		= THIS_MODULE,
 	.af		= NFPROTO_IPV6,
 	.priority	= NF_IP6_PRI_SECURITY,
+	.table_init     = ip6table_security_table_init,
 };
 
 static unsigned int
@@ -44,11 +47,14 @@ ip6table_security_hook(void *priv, struct sk_buff *skb,
 
 static struct nf_hook_ops *sectbl_ops __read_mostly;
 
-static int __net_init ip6table_security_net_init(struct net *net)
+static int __net_init ip6table_security_table_init(struct net *net)
 {
 	struct ip6t_replace *repl;
 	int ret;
 
+	if (net->ipv6.ip6table_security)
+		return 0;
+
 	repl = ip6t_alloc_initial_table(&security_table);
 	if (repl == NULL)
 		return -ENOMEM;
@@ -60,11 +66,13 @@ static int __net_init ip6table_security_net_init(struct net *net)
 
 static void __net_exit ip6table_security_net_exit(struct net *net)
 {
+	if (!net->ipv6.ip6table_security)
+		return;
 	ip6t_unregister_table(net, net->ipv6.ip6table_security, sectbl_ops);
+	net->ipv6.ip6table_security = NULL;
 }
 
 static struct pernet_operations ip6table_security_net_ops = {
-	.init = ip6table_security_net_init,
 	.exit = ip6table_security_net_exit,
 };
 
@@ -72,27 +80,28 @@ static int __init ip6table_security_init(void)
 {
 	int ret;
 
+	sectbl_ops = xt_hook_ops_alloc(&security_table, ip6table_security_hook);
+	if (IS_ERR(sectbl_ops))
+		return PTR_ERR(sectbl_ops);
+
 	ret = register_pernet_subsys(&ip6table_security_net_ops);
-	if (ret < 0)
+	if (ret < 0) {
+		kfree(sectbl_ops);
 		return ret;
-
-	sectbl_ops = xt_hook_link(&security_table, ip6table_security_hook);
-	if (IS_ERR(sectbl_ops)) {
-		ret = PTR_ERR(sectbl_ops);
-		goto cleanup_table;
 	}
 
-	return ret;
-
-cleanup_table:
-	unregister_pernet_subsys(&ip6table_security_net_ops);
+	ret = ip6table_security_table_init(&init_net);
+	if (ret) {
+		unregister_pernet_subsys(&ip6table_security_net_ops);
+		kfree(sectbl_ops);
+	}
 	return ret;
 }
 
 static void __exit ip6table_security_fini(void)
 {
-	xt_hook_unlink(&security_table, sectbl_ops);
 	unregister_pernet_subsys(&ip6table_security_net_ops);
+	kfree(sectbl_ops);
 }
 
 module_init(ip6table_security_init);
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index c8a0b7d..d0cd2b9 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -694,12 +694,45 @@ EXPORT_SYMBOL(xt_free_table_info);
 struct xt_table *xt_find_table_lock(struct net *net, u_int8_t af,
 				    const char *name)
 {
-	struct xt_table *t;
+	struct xt_table *t, *found = NULL;
 
 	mutex_lock(&xt[af].mutex);
 	list_for_each_entry(t, &net->xt.tables[af], list)
 		if (strcmp(t->name, name) == 0 && try_module_get(t->me))
 			return t;
+
+	if (net == &init_net)
+		goto out;
+
+	/* Table doesn't exist in this netns, re-try init */
+	list_for_each_entry(t, &init_net.xt.tables[af], list) {
+		if (strcmp(t->name, name))
+			continue;
+		if (!try_module_get(t->me))
+			return NULL;
+
+		mutex_unlock(&xt[af].mutex);
+		if (t->table_init(net) != 0) {
+			module_put(t->me);
+			return NULL;
+		}
+
+		found = t;
+
+		mutex_lock(&xt[af].mutex);
+		break;
+	}
+
+	if (!found)
+		goto out;
+
+	/* and once again: */
+	list_for_each_entry(t, &net->xt.tables[af], list)
+		if (strcmp(t->name, name) == 0)
+			return t;
+
+	module_put(found->me);
+ out:
 	mutex_unlock(&xt[af].mutex);
 	return NULL;
 }
@@ -1170,20 +1203,20 @@ static const struct file_operations xt_target_ops = {
 #endif /* CONFIG_PROC_FS */
 
 /**
- * xt_hook_link - set up hooks for a new table
+ * xt_hook_ops_alloc - set up hooks for a new table
  * @table:	table with metadata needed to set up hooks
  * @fn:		Hook function
  *
- * This function will take care of creating and registering the necessary
- * Netfilter hooks for XT tables.
+ * This function will create the nf_hook_ops that the x_table needs
+ * to hand to xt_hook_link_net().
  */
-struct nf_hook_ops *xt_hook_link(const struct xt_table *table, nf_hookfn *fn)
+struct nf_hook_ops *
+xt_hook_ops_alloc(const struct xt_table *table, nf_hookfn *fn)
 {
 	unsigned int hook_mask = table->valid_hooks;
 	uint8_t i, num_hooks = hweight32(hook_mask);
 	uint8_t hooknum;
 	struct nf_hook_ops *ops;
-	int ret;
 
 	ops = kmalloc(sizeof(*ops) * num_hooks, GFP_KERNEL);
 	if (ops == NULL)
@@ -1200,27 +1233,9 @@ struct nf_hook_ops *xt_hook_link(const struct xt_table *table, nf_hookfn *fn)
 		++i;
 	}
 
-	ret = nf_register_hooks(ops, num_hooks);
-	if (ret < 0) {
-		kfree(ops);
-		return ERR_PTR(ret);
-	}
-
 	return ops;
 }
-EXPORT_SYMBOL_GPL(xt_hook_link);
-
-/**
- * xt_hook_unlink - remove hooks for a table
- * @ops:	nf_hook_ops array as returned by nf_hook_link
- * @hook_mask:	the very same mask that was passed to nf_hook_link
- */
-void xt_hook_unlink(const struct xt_table *table, struct nf_hook_ops *ops)
-{
-	nf_unregister_hooks(ops, hweight32(table->valid_hooks));
-	kfree(ops);
-}
-EXPORT_SYMBOL_GPL(xt_hook_unlink);
+EXPORT_SYMBOL_GPL(xt_hook_ops_alloc);
 
 int xt_proto_init(struct net *net, u_int8_t af)
 {
-- 
cgit v0.10.2


From 5f6c253ebe93b02dece01c6f58447f16b29f6dd3 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 25 Feb 2016 10:08:37 +0100
Subject: netfilter: bridge: register hooks only when bridge interface is added

This moves bridge hooks to a register-when-needed scheme.

We use a device notifier to register the 'call-iptables' netfilter hooks
only once a bridge gets added.

This means that if the initial namespace uses a bridge, newly created
network namespaces no longer get the PRE_ROUTING ipt_sabotage hook.

It will registered in that network namespace once a bridge is created
within that namespace.

A few modules still use global hooks:

- conntrack
- bridge PF_BRIDGE hooks
- IPVS
- CLUSTER match (deprecated)
- SYNPROXY

As long as these modules are not loaded/used, a new network namespace has
empty hook list and NF_HOOK() will boil down to single list_empty test even
if initial namespace does stateless packet filtering.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c
index 7ddbe7e..44114a9 100644
--- a/net/bridge/br_netfilter_hooks.c
+++ b/net/bridge/br_netfilter_hooks.c
@@ -37,6 +37,7 @@
 #include <net/addrconf.h>
 #include <net/route.h>
 #include <net/netfilter/br_netfilter.h>
+#include <net/netns/generic.h>
 
 #include <asm/uaccess.h>
 #include "br_private.h"
@@ -44,6 +45,12 @@
 #include <linux/sysctl.h>
 #endif
 
+static int brnf_net_id __read_mostly;
+
+struct brnf_net {
+	bool enabled;
+};
+
 #ifdef CONFIG_SYSCTL
 static struct ctl_table_header *brnf_sysctl_header;
 static int brnf_call_iptables __read_mostly = 1;
@@ -938,6 +945,53 @@ static struct nf_hook_ops br_nf_ops[] __read_mostly = {
 	},
 };
 
+static int brnf_device_event(struct notifier_block *unused, unsigned long event,
+			     void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct brnf_net *brnet;
+	struct net *net;
+	int ret;
+
+	if (event != NETDEV_REGISTER || !(dev->priv_flags & IFF_EBRIDGE))
+		return NOTIFY_DONE;
+
+	ASSERT_RTNL();
+
+	net = dev_net(dev);
+	brnet = net_generic(net, brnf_net_id);
+	if (brnet->enabled)
+		return NOTIFY_OK;
+
+	ret = nf_register_net_hooks(net, br_nf_ops, ARRAY_SIZE(br_nf_ops));
+	if (ret)
+		return NOTIFY_BAD;
+
+	brnet->enabled = true;
+	return NOTIFY_OK;
+}
+
+static void __net_exit brnf_exit_net(struct net *net)
+{
+	struct brnf_net *brnet = net_generic(net, brnf_net_id);
+
+	if (!brnet->enabled)
+		return;
+
+	nf_unregister_net_hooks(net, br_nf_ops, ARRAY_SIZE(br_nf_ops));
+	brnet->enabled = false;
+}
+
+static struct pernet_operations brnf_net_ops __read_mostly = {
+	.exit = brnf_exit_net,
+	.id   = &brnf_net_id,
+	.size = sizeof(struct brnf_net),
+};
+
+static struct notifier_block brnf_notifier __read_mostly = {
+	.notifier_call = brnf_device_event,
+};
+
 #ifdef CONFIG_SYSCTL
 static
 int brnf_sysctl_call_tables(struct ctl_table *ctl, int write,
@@ -1003,16 +1057,23 @@ static int __init br_netfilter_init(void)
 {
 	int ret;
 
-	ret = nf_register_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops));
+	ret = register_pernet_subsys(&brnf_net_ops);
 	if (ret < 0)
 		return ret;
 
+	ret = register_netdevice_notifier(&brnf_notifier);
+	if (ret < 0) {
+		unregister_pernet_subsys(&brnf_net_ops);
+		return ret;
+	}
+
 #ifdef CONFIG_SYSCTL
 	brnf_sysctl_header = register_net_sysctl(&init_net, "net/bridge", brnf_table);
 	if (brnf_sysctl_header == NULL) {
 		printk(KERN_WARNING
 		       "br_netfilter: can't register to sysctl.\n");
-		nf_unregister_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops));
+		unregister_netdevice_notifier(&brnf_notifier);
+		unregister_pernet_subsys(&brnf_net_ops);
 		return -ENOMEM;
 	}
 #endif
@@ -1024,7 +1085,8 @@ static int __init br_netfilter_init(void)
 static void __exit br_netfilter_fini(void)
 {
 	RCU_INIT_POINTER(nf_br_ops, NULL);
-	nf_unregister_hooks(br_nf_ops, ARRAY_SIZE(br_nf_ops));
+	unregister_netdevice_notifier(&brnf_notifier);
+	unregister_pernet_subsys(&brnf_net_ops);
 #ifdef CONFIG_SYSCTL
 	unregister_net_sysctl_table(brnf_sysctl_header);
 #endif
-- 
cgit v0.10.2


From af4610c39589d839551da104f7da342d86f23ea0 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 25 Feb 2016 10:08:38 +0100
Subject: netfilter: don't call hooks unless needed

With the previous patches in place, a netns nf_hook_list might be empty,
even if e.g. init_net performs filtering.

Thus change nf_hook_thresh to check the hook_list as well before
initializing hook_state and calling nf_hook_slow().

We still make use of static keys; if no netfilter modules are loaded
list is guaranteed to be empty.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 0ad5567..9230f9a 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -141,22 +141,6 @@ void nf_unregister_sockopt(struct nf_sockopt_ops *reg);
 
 #ifdef HAVE_JUMP_LABEL
 extern struct static_key nf_hooks_needed[NFPROTO_NUMPROTO][NF_MAX_HOOKS];
-
-static inline bool nf_hook_list_active(struct list_head *hook_list,
-				       u_int8_t pf, unsigned int hook)
-{
-	if (__builtin_constant_p(pf) &&
-	    __builtin_constant_p(hook))
-		return static_key_false(&nf_hooks_needed[pf][hook]);
-
-	return !list_empty(hook_list);
-}
-#else
-static inline bool nf_hook_list_active(struct list_head *hook_list,
-				       u_int8_t pf, unsigned int hook)
-{
-	return !list_empty(hook_list);
-}
 #endif
 
 int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state);
@@ -177,9 +161,18 @@ static inline int nf_hook_thresh(u_int8_t pf, unsigned int hook,
 				 int (*okfn)(struct net *, struct sock *, struct sk_buff *),
 				 int thresh)
 {
-	struct list_head *hook_list = &net->nf.hooks[pf][hook];
+	struct list_head *hook_list;
+
+#ifdef HAVE_JUMP_LABEL
+	if (__builtin_constant_p(pf) &&
+	    __builtin_constant_p(hook) &&
+	    !static_key_false(&nf_hooks_needed[pf][hook]))
+		return 1;
+#endif
+
+	hook_list = &net->nf.hooks[pf][hook];
 
-	if (nf_hook_list_active(hook_list, pf, hook)) {
+	if (!list_empty(hook_list)) {
 		struct nf_hook_state state;
 
 		nf_hook_state_init(&state, hook_list, hook, thresh,
-- 
cgit v0.10.2


From 8a6bf5da1aefdafd60b73d9122c7af9fd2d7bb9c Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 1 Mar 2016 19:55:14 +0100
Subject: netfilter: nft_masq: support port range

Complete masquerading support by allowing port range selection.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/net/netfilter/nft_masq.h b/include/net/netfilter/nft_masq.h
index e2a518b..a3f3c11 100644
--- a/include/net/netfilter/nft_masq.h
+++ b/include/net/netfilter/nft_masq.h
@@ -2,7 +2,9 @@
 #define _NFT_MASQ_H_
 
 struct nft_masq {
-	u32	flags;
+	u32			flags;
+	enum nft_registers      sreg_proto_min:8;
+	enum nft_registers      sreg_proto_max:8;
 };
 
 extern const struct nla_policy nft_masq_policy[];
diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index b19be0a..eeffde1 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -951,10 +951,14 @@ enum nft_nat_attributes {
  * enum nft_masq_attributes - nf_tables masquerade expression attributes
  *
  * @NFTA_MASQ_FLAGS: NAT flags (see NF_NAT_RANGE_* in linux/netfilter/nf_nat.h) (NLA_U32)
+ * @NFTA_MASQ_REG_PROTO_MIN: source register of proto range start (NLA_U32: nft_registers)
+ * @NFTA_MASQ_REG_PROTO_MAX: source register of proto range end (NLA_U32: nft_registers)
  */
 enum nft_masq_attributes {
 	NFTA_MASQ_UNSPEC,
 	NFTA_MASQ_FLAGS,
+	NFTA_MASQ_REG_PROTO_MIN,
+	NFTA_MASQ_REG_PROTO_MAX,
 	__NFTA_MASQ_MAX
 };
 #define NFTA_MASQ_MAX		(__NFTA_MASQ_MAX - 1)
diff --git a/net/ipv4/netfilter/nft_masq_ipv4.c b/net/ipv4/netfilter/nft_masq_ipv4.c
index b72ffc5..51ced81 100644
--- a/net/ipv4/netfilter/nft_masq_ipv4.c
+++ b/net/ipv4/netfilter/nft_masq_ipv4.c
@@ -25,7 +25,12 @@ static void nft_masq_ipv4_eval(const struct nft_expr *expr,
 
 	memset(&range, 0, sizeof(range));
 	range.flags = priv->flags;
-
+	if (priv->sreg_proto_min) {
+		range.min_proto.all =
+			*(__be16 *)&regs->data[priv->sreg_proto_min];
+		range.max_proto.all =
+			*(__be16 *)&regs->data[priv->sreg_proto_max];
+	}
 	regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, pkt->hook,
 						    &range, pkt->out);
 }
diff --git a/net/ipv6/netfilter/nft_masq_ipv6.c b/net/ipv6/netfilter/nft_masq_ipv6.c
index cd1ac16..9597ffb 100644
--- a/net/ipv6/netfilter/nft_masq_ipv6.c
+++ b/net/ipv6/netfilter/nft_masq_ipv6.c
@@ -26,7 +26,12 @@ static void nft_masq_ipv6_eval(const struct nft_expr *expr,
 
 	memset(&range, 0, sizeof(range));
 	range.flags = priv->flags;
-
+	if (priv->sreg_proto_min) {
+		range.min_proto.all =
+			*(__be16 *)&regs->data[priv->sreg_proto_min];
+		range.max_proto.all =
+			*(__be16 *)&regs->data[priv->sreg_proto_max];
+	}
 	regs->verdict.code = nf_nat_masquerade_ipv6(pkt->skb, &range, pkt->out);
 }
 
diff --git a/net/netfilter/nft_masq.c b/net/netfilter/nft_masq.c
index 9aea747..81b5ad6 100644
--- a/net/netfilter/nft_masq.c
+++ b/net/netfilter/nft_masq.c
@@ -17,7 +17,9 @@
 #include <net/netfilter/nft_masq.h>
 
 const struct nla_policy nft_masq_policy[NFTA_MASQ_MAX + 1] = {
-	[NFTA_MASQ_FLAGS]	= { .type = NLA_U32 },
+	[NFTA_MASQ_FLAGS]		= { .type = NLA_U32 },
+	[NFTA_MASQ_REG_PROTO_MIN]	= { .type = NLA_U32 },
+	[NFTA_MASQ_REG_PROTO_MAX]	= { .type = NLA_U32 },
 };
 EXPORT_SYMBOL_GPL(nft_masq_policy);
 
@@ -40,6 +42,7 @@ int nft_masq_init(const struct nft_ctx *ctx,
 		  const struct nft_expr *expr,
 		  const struct nlattr * const tb[])
 {
+	u32 plen = FIELD_SIZEOF(struct nf_nat_range, min_addr.all);
 	struct nft_masq *priv = nft_expr_priv(expr);
 	int err;
 
@@ -47,12 +50,32 @@ int nft_masq_init(const struct nft_ctx *ctx,
 	if (err)
 		return err;
 
-	if (tb[NFTA_MASQ_FLAGS] == NULL)
-		return 0;
-
-	priv->flags = ntohl(nla_get_be32(tb[NFTA_MASQ_FLAGS]));
-	if (priv->flags & ~NF_NAT_RANGE_MASK)
-		return -EINVAL;
+	if (tb[NFTA_MASQ_FLAGS]) {
+		priv->flags = ntohl(nla_get_be32(tb[NFTA_MASQ_FLAGS]));
+		if (priv->flags & ~NF_NAT_RANGE_MASK)
+			return -EINVAL;
+	}
+
+	if (tb[NFTA_MASQ_REG_PROTO_MIN]) {
+		priv->sreg_proto_min =
+			nft_parse_register(tb[NFTA_MASQ_REG_PROTO_MIN]);
+
+		err = nft_validate_register_load(priv->sreg_proto_min, plen);
+		if (err < 0)
+			return err;
+
+		if (tb[NFTA_MASQ_REG_PROTO_MAX]) {
+			priv->sreg_proto_max =
+				nft_parse_register(tb[NFTA_MASQ_REG_PROTO_MAX]);
+
+			err = nft_validate_register_load(priv->sreg_proto_max,
+							 plen);
+			if (err < 0)
+				return err;
+		} else {
+			priv->sreg_proto_max = priv->sreg_proto_min;
+		}
+	}
 
 	return 0;
 }
@@ -62,12 +85,18 @@ int nft_masq_dump(struct sk_buff *skb, const struct nft_expr *expr)
 {
 	const struct nft_masq *priv = nft_expr_priv(expr);
 
-	if (priv->flags == 0)
-		return 0;
-
-	if (nla_put_be32(skb, NFTA_MASQ_FLAGS, htonl(priv->flags)))
+	if (priv->flags != 0 &&
+	    nla_put_be32(skb, NFTA_MASQ_FLAGS, htonl(priv->flags)))
 		goto nla_put_failure;
 
+	if (priv->sreg_proto_min) {
+		if (nft_dump_register(skb, NFTA_MASQ_REG_PROTO_MIN,
+				      priv->sreg_proto_min) ||
+		    nft_dump_register(skb, NFTA_MASQ_REG_PROTO_MAX,
+				      priv->sreg_proto_max))
+			goto nla_put_failure;
+	}
+
 	return 0;
 
 nla_put_failure:
-- 
cgit v0.10.2


From dcdede0406d30e14a6ed727dce69dfac0518371d Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:42 -0800
Subject: RDS: Drop stale iWARP RDMA transport

RDS iWarp support code has become stale and non testable. As
indicated earlier, am dropping the support for it.

If new iWarp user(s) shows up in future, we can adapat the RDS IB
transprt for the special RDMA READ sink case. iWarp needs an MR
for the RDMA READ sink.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/rds.txt b/Documentation/networking/rds.txt
index e1a3d59..9d219d8 100644
--- a/Documentation/networking/rds.txt
+++ b/Documentation/networking/rds.txt
@@ -19,9 +19,7 @@ to N*N if you use a connection-oriented socket transport like TCP.
 
 RDS is not Infiniband-specific; it was designed to support different
 transports.  The current implementation used to support RDS over TCP as well
-as IB. Work is in progress to support RDS over iWARP, and using DCE to
-guarantee no dropped packets on Ethernet, it may be possible to use RDS over
-UDP in the future.
+as IB.
 
 The high-level semantics of RDS from the application's point of view are
 
diff --git a/net/rds/Kconfig b/net/rds/Kconfig
index f2c670b..bffde4b 100644
--- a/net/rds/Kconfig
+++ b/net/rds/Kconfig
@@ -4,14 +4,13 @@ config RDS
 	depends on INET
 	---help---
 	  The RDS (Reliable Datagram Sockets) protocol provides reliable,
-	  sequenced delivery of datagrams over Infiniband, iWARP,
-	  or TCP.
+	  sequenced delivery of datagrams over Infiniband or TCP.
 
 config RDS_RDMA
-	tristate "RDS over Infiniband and iWARP"
+	tristate "RDS over Infiniband"
 	depends on RDS && INFINIBAND && INFINIBAND_ADDR_TRANS
 	---help---
-	  Allow RDS to use Infiniband and iWARP as a transport.
+	  Allow RDS to use Infiniband as a transport.
 	  This transport supports RDMA operations.
 
 config RDS_TCP
diff --git a/net/rds/Makefile b/net/rds/Makefile
index 56d3f60..19e5485 100644
--- a/net/rds/Makefile
+++ b/net/rds/Makefile
@@ -6,9 +6,7 @@ rds-y :=	af_rds.o bind.o cong.o connection.o info.o message.o   \
 obj-$(CONFIG_RDS_RDMA) += rds_rdma.o
 rds_rdma-y :=	rdma_transport.o \
 			ib.o ib_cm.o ib_recv.o ib_ring.o ib_send.o ib_stats.o \
-			ib_sysctl.o ib_rdma.o \
-			iw.o iw_cm.o iw_recv.o iw_ring.o iw_send.o iw_stats.o \
-			iw_sysctl.o iw_rdma.o
+			ib_sysctl.o ib_rdma.o
 
 
 obj-$(CONFIG_RDS_TCP) += rds_tcp.o
diff --git a/net/rds/iw.c b/net/rds/iw.c
deleted file mode 100644
index f4a9fff..0000000
--- a/net/rds/iw.c
+++ /dev/null
@@ -1,312 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-#include <linux/in.h>
-#include <linux/if.h>
-#include <linux/netdevice.h>
-#include <linux/inetdevice.h>
-#include <linux/if_arp.h>
-#include <linux/delay.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-
-#include "rds.h"
-#include "iw.h"
-
-unsigned int fastreg_pool_size = RDS_FASTREG_POOL_SIZE;
-unsigned int fastreg_message_size = RDS_FASTREG_SIZE + 1; /* +1 allows for unaligned MRs */
-
-module_param(fastreg_pool_size, int, 0444);
-MODULE_PARM_DESC(fastreg_pool_size, " Max number of fastreg MRs per device");
-module_param(fastreg_message_size, int, 0444);
-MODULE_PARM_DESC(fastreg_message_size, " Max size of a RDMA transfer (fastreg MRs)");
-
-struct list_head rds_iw_devices;
-
-/* NOTE: if also grabbing iwdev lock, grab this first */
-DEFINE_SPINLOCK(iw_nodev_conns_lock);
-LIST_HEAD(iw_nodev_conns);
-
-static void rds_iw_add_one(struct ib_device *device)
-{
-	struct rds_iw_device *rds_iwdev;
-
-	/* Only handle iwarp devices */
-	if (device->node_type != RDMA_NODE_RNIC)
-		return;
-
-	rds_iwdev = kmalloc(sizeof *rds_iwdev, GFP_KERNEL);
-	if (!rds_iwdev)
-		return;
-
-	spin_lock_init(&rds_iwdev->spinlock);
-
-	rds_iwdev->dma_local_lkey = !!(device->attrs.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY);
-	rds_iwdev->max_wrs = device->attrs.max_qp_wr;
-	rds_iwdev->max_sge = min(device->attrs.max_sge, RDS_IW_MAX_SGE);
-
-	rds_iwdev->dev = device;
-	rds_iwdev->pd = ib_alloc_pd(device);
-	if (IS_ERR(rds_iwdev->pd))
-		goto free_dev;
-
-	if (!rds_iwdev->dma_local_lkey) {
-		rds_iwdev->mr = ib_get_dma_mr(rds_iwdev->pd,
-					IB_ACCESS_REMOTE_READ |
-					IB_ACCESS_REMOTE_WRITE |
-					IB_ACCESS_LOCAL_WRITE);
-		if (IS_ERR(rds_iwdev->mr))
-			goto err_pd;
-	} else
-		rds_iwdev->mr = NULL;
-
-	rds_iwdev->mr_pool = rds_iw_create_mr_pool(rds_iwdev);
-	if (IS_ERR(rds_iwdev->mr_pool)) {
-		rds_iwdev->mr_pool = NULL;
-		goto err_mr;
-	}
-
-	INIT_LIST_HEAD(&rds_iwdev->cm_id_list);
-	INIT_LIST_HEAD(&rds_iwdev->conn_list);
-	list_add_tail(&rds_iwdev->list, &rds_iw_devices);
-
-	ib_set_client_data(device, &rds_iw_client, rds_iwdev);
-	return;
-
-err_mr:
-	if (rds_iwdev->mr)
-		ib_dereg_mr(rds_iwdev->mr);
-err_pd:
-	ib_dealloc_pd(rds_iwdev->pd);
-free_dev:
-	kfree(rds_iwdev);
-}
-
-static void rds_iw_remove_one(struct ib_device *device, void *client_data)
-{
-	struct rds_iw_device *rds_iwdev = client_data;
-	struct rds_iw_cm_id *i_cm_id, *next;
-
-	if (!rds_iwdev)
-		return;
-
-	spin_lock_irq(&rds_iwdev->spinlock);
-	list_for_each_entry_safe(i_cm_id, next, &rds_iwdev->cm_id_list, list) {
-		list_del(&i_cm_id->list);
-		kfree(i_cm_id);
-	}
-	spin_unlock_irq(&rds_iwdev->spinlock);
-
-	rds_iw_destroy_conns(rds_iwdev);
-
-	if (rds_iwdev->mr_pool)
-		rds_iw_destroy_mr_pool(rds_iwdev->mr_pool);
-
-	if (rds_iwdev->mr)
-		ib_dereg_mr(rds_iwdev->mr);
-
-	ib_dealloc_pd(rds_iwdev->pd);
-
-	list_del(&rds_iwdev->list);
-	kfree(rds_iwdev);
-}
-
-struct ib_client rds_iw_client = {
-	.name   = "rds_iw",
-	.add    = rds_iw_add_one,
-	.remove = rds_iw_remove_one
-};
-
-static int rds_iw_conn_info_visitor(struct rds_connection *conn,
-				    void *buffer)
-{
-	struct rds_info_rdma_connection *iinfo = buffer;
-	struct rds_iw_connection *ic;
-
-	/* We will only ever look at IB transports */
-	if (conn->c_trans != &rds_iw_transport)
-		return 0;
-
-	iinfo->src_addr = conn->c_laddr;
-	iinfo->dst_addr = conn->c_faddr;
-
-	memset(&iinfo->src_gid, 0, sizeof(iinfo->src_gid));
-	memset(&iinfo->dst_gid, 0, sizeof(iinfo->dst_gid));
-	if (rds_conn_state(conn) == RDS_CONN_UP) {
-		struct rds_iw_device *rds_iwdev;
-		struct rdma_dev_addr *dev_addr;
-
-		ic = conn->c_transport_data;
-		dev_addr = &ic->i_cm_id->route.addr.dev_addr;
-
-		rdma_addr_get_sgid(dev_addr, (union ib_gid *) &iinfo->src_gid);
-		rdma_addr_get_dgid(dev_addr, (union ib_gid *) &iinfo->dst_gid);
-
-		rds_iwdev = ib_get_client_data(ic->i_cm_id->device, &rds_iw_client);
-		iinfo->max_send_wr = ic->i_send_ring.w_nr;
-		iinfo->max_recv_wr = ic->i_recv_ring.w_nr;
-		iinfo->max_send_sge = rds_iwdev->max_sge;
-		rds_iw_get_mr_info(rds_iwdev, iinfo);
-	}
-	return 1;
-}
-
-static void rds_iw_ic_info(struct socket *sock, unsigned int len,
-			   struct rds_info_iterator *iter,
-			   struct rds_info_lengths *lens)
-{
-	rds_for_each_conn_info(sock, len, iter, lens,
-				rds_iw_conn_info_visitor,
-				sizeof(struct rds_info_rdma_connection));
-}
-
-
-/*
- * Early RDS/IB was built to only bind to an address if there is an IPoIB
- * device with that address set.
- *
- * If it were me, I'd advocate for something more flexible.  Sending and
- * receiving should be device-agnostic.  Transports would try and maintain
- * connections between peers who have messages queued.  Userspace would be
- * allowed to influence which paths have priority.  We could call userspace
- * asserting this policy "routing".
- */
-static int rds_iw_laddr_check(struct net *net, __be32 addr)
-{
-	int ret;
-	struct rdma_cm_id *cm_id;
-	struct sockaddr_in sin;
-
-	/* Create a CMA ID and try to bind it. This catches both
-	 * IB and iWARP capable NICs.
-	 */
-	cm_id = rdma_create_id(&init_net, NULL, NULL, RDMA_PS_TCP, IB_QPT_RC);
-	if (IS_ERR(cm_id))
-		return PTR_ERR(cm_id);
-
-	memset(&sin, 0, sizeof(sin));
-	sin.sin_family = AF_INET;
-	sin.sin_addr.s_addr = addr;
-
-	/* rdma_bind_addr will only succeed for IB & iWARP devices */
-	ret = rdma_bind_addr(cm_id, (struct sockaddr *)&sin);
-	/* due to this, we will claim to support IB devices unless we
-	   check node_type. */
-	if (ret || !cm_id->device ||
-	    cm_id->device->node_type != RDMA_NODE_RNIC)
-		ret = -EADDRNOTAVAIL;
-
-	rdsdebug("addr %pI4 ret %d node type %d\n",
-		&addr, ret,
-		cm_id->device ? cm_id->device->node_type : -1);
-
-	rdma_destroy_id(cm_id);
-
-	return ret;
-}
-
-void rds_iw_exit(void)
-{
-	rds_info_deregister_func(RDS_INFO_IWARP_CONNECTIONS, rds_iw_ic_info);
-	rds_iw_destroy_nodev_conns();
-	ib_unregister_client(&rds_iw_client);
-	rds_iw_sysctl_exit();
-	rds_iw_recv_exit();
-	rds_trans_unregister(&rds_iw_transport);
-}
-
-struct rds_transport rds_iw_transport = {
-	.laddr_check		= rds_iw_laddr_check,
-	.xmit_complete		= rds_iw_xmit_complete,
-	.xmit			= rds_iw_xmit,
-	.xmit_rdma		= rds_iw_xmit_rdma,
-	.recv			= rds_iw_recv,
-	.conn_alloc		= rds_iw_conn_alloc,
-	.conn_free		= rds_iw_conn_free,
-	.conn_connect		= rds_iw_conn_connect,
-	.conn_shutdown		= rds_iw_conn_shutdown,
-	.inc_copy_to_user	= rds_iw_inc_copy_to_user,
-	.inc_free		= rds_iw_inc_free,
-	.cm_initiate_connect	= rds_iw_cm_initiate_connect,
-	.cm_handle_connect	= rds_iw_cm_handle_connect,
-	.cm_connect_complete	= rds_iw_cm_connect_complete,
-	.stats_info_copy	= rds_iw_stats_info_copy,
-	.exit			= rds_iw_exit,
-	.get_mr			= rds_iw_get_mr,
-	.sync_mr		= rds_iw_sync_mr,
-	.free_mr		= rds_iw_free_mr,
-	.flush_mrs		= rds_iw_flush_mrs,
-	.t_owner		= THIS_MODULE,
-	.t_name			= "iwarp",
-	.t_type			= RDS_TRANS_IWARP,
-	.t_prefer_loopback	= 1,
-};
-
-int rds_iw_init(void)
-{
-	int ret;
-
-	INIT_LIST_HEAD(&rds_iw_devices);
-
-	ret = ib_register_client(&rds_iw_client);
-	if (ret)
-		goto out;
-
-	ret = rds_iw_sysctl_init();
-	if (ret)
-		goto out_ibreg;
-
-	ret = rds_iw_recv_init();
-	if (ret)
-		goto out_sysctl;
-
-	ret = rds_trans_register(&rds_iw_transport);
-	if (ret)
-		goto out_recv;
-
-	rds_info_register_func(RDS_INFO_IWARP_CONNECTIONS, rds_iw_ic_info);
-
-	goto out;
-
-out_recv:
-	rds_iw_recv_exit();
-out_sysctl:
-	rds_iw_sysctl_exit();
-out_ibreg:
-	ib_unregister_client(&rds_iw_client);
-out:
-	return ret;
-}
-
-MODULE_LICENSE("GPL");
-
diff --git a/net/rds/iw.h b/net/rds/iw.h
deleted file mode 100644
index 5af01d1..0000000
--- a/net/rds/iw.h
+++ /dev/null
@@ -1,398 +0,0 @@
-#ifndef _RDS_IW_H
-#define _RDS_IW_H
-
-#include <linux/interrupt.h>
-#include <rdma/ib_verbs.h>
-#include <rdma/rdma_cm.h>
-#include "rds.h"
-#include "rdma_transport.h"
-
-#define RDS_FASTREG_SIZE		20
-#define RDS_FASTREG_POOL_SIZE		2048
-
-#define RDS_IW_MAX_SGE			8
-#define RDS_IW_RECV_SGE 		2
-
-#define RDS_IW_DEFAULT_RECV_WR		1024
-#define RDS_IW_DEFAULT_SEND_WR		256
-
-#define RDS_IW_SUPPORTED_PROTOCOLS	0x00000003	/* minor versions supported */
-
-extern struct list_head rds_iw_devices;
-
-/*
- * IB posts RDS_FRAG_SIZE fragments of pages to the receive queues to
- * try and minimize the amount of memory tied up both the device and
- * socket receive queues.
- */
-/* page offset of the final full frag that fits in the page */
-#define RDS_PAGE_LAST_OFF (((PAGE_SIZE  / RDS_FRAG_SIZE) - 1) * RDS_FRAG_SIZE)
-struct rds_page_frag {
-	struct list_head	f_item;
-	struct page		*f_page;
-	unsigned long		f_offset;
-	dma_addr_t 		f_mapped;
-};
-
-struct rds_iw_incoming {
-	struct list_head	ii_frags;
-	struct rds_incoming	ii_inc;
-};
-
-struct rds_iw_connect_private {
-	/* Add new fields at the end, and don't permute existing fields. */
-	__be32			dp_saddr;
-	__be32			dp_daddr;
-	u8			dp_protocol_major;
-	u8			dp_protocol_minor;
-	__be16			dp_protocol_minor_mask; /* bitmask */
-	__be32			dp_reserved1;
-	__be64			dp_ack_seq;
-	__be32			dp_credit;		/* non-zero enables flow ctl */
-};
-
-struct rds_iw_scatterlist {
-	struct scatterlist	*list;
-	unsigned int		len;
-	int			dma_len;
-	unsigned int		dma_npages;
-	unsigned int		bytes;
-};
-
-struct rds_iw_mapping {
-	spinlock_t		m_lock;	/* protect the mapping struct */
-	struct list_head	m_list;
-	struct rds_iw_mr	*m_mr;
-	uint32_t		m_rkey;
-	struct rds_iw_scatterlist m_sg;
-};
-
-struct rds_iw_send_work {
-	struct rds_message	*s_rm;
-
-	/* We should really put these into a union: */
-	struct rm_rdma_op	*s_op;
-	struct rds_iw_mapping	*s_mapping;
-	struct ib_mr		*s_mr;
-	unsigned char		s_remap_count;
-
-	union {
-		struct ib_send_wr	s_send_wr;
-		struct ib_rdma_wr	s_rdma_wr;
-		struct ib_reg_wr	s_reg_wr;
-	};
-	struct ib_sge		s_sge[RDS_IW_MAX_SGE];
-	unsigned long		s_queued;
-};
-
-struct rds_iw_recv_work {
-	struct rds_iw_incoming 	*r_iwinc;
-	struct rds_page_frag	*r_frag;
-	struct ib_recv_wr	r_wr;
-	struct ib_sge		r_sge[2];
-};
-
-struct rds_iw_work_ring {
-	u32		w_nr;
-	u32		w_alloc_ptr;
-	u32		w_alloc_ctr;
-	u32		w_free_ptr;
-	atomic_t	w_free_ctr;
-};
-
-struct rds_iw_device;
-
-struct rds_iw_connection {
-
-	struct list_head	iw_node;
-	struct rds_iw_device 	*rds_iwdev;
-	struct rds_connection	*conn;
-
-	/* alphabet soup, IBTA style */
-	struct rdma_cm_id	*i_cm_id;
-	struct ib_pd		*i_pd;
-	struct ib_mr		*i_mr;
-	struct ib_cq		*i_send_cq;
-	struct ib_cq		*i_recv_cq;
-
-	/* tx */
-	struct rds_iw_work_ring	i_send_ring;
-	struct rds_message	*i_rm;
-	struct rds_header	*i_send_hdrs;
-	u64			i_send_hdrs_dma;
-	struct rds_iw_send_work *i_sends;
-
-	/* rx */
-	struct tasklet_struct	i_recv_tasklet;
-	struct mutex		i_recv_mutex;
-	struct rds_iw_work_ring	i_recv_ring;
-	struct rds_iw_incoming	*i_iwinc;
-	u32			i_recv_data_rem;
-	struct rds_header	*i_recv_hdrs;
-	u64			i_recv_hdrs_dma;
-	struct rds_iw_recv_work *i_recvs;
-	struct rds_page_frag	i_frag;
-	u64			i_ack_recv;	/* last ACK received */
-
-	/* sending acks */
-	unsigned long		i_ack_flags;
-#ifdef KERNEL_HAS_ATOMIC64
-	atomic64_t		i_ack_next;	/* next ACK to send */
-#else
-	spinlock_t		i_ack_lock;	/* protect i_ack_next */
-	u64			i_ack_next;	/* next ACK to send */
-#endif
-	struct rds_header	*i_ack;
-	struct ib_send_wr	i_ack_wr;
-	struct ib_sge		i_ack_sge;
-	u64			i_ack_dma;
-	unsigned long		i_ack_queued;
-
-	/* Flow control related information
-	 *
-	 * Our algorithm uses a pair variables that we need to access
-	 * atomically - one for the send credits, and one posted
-	 * recv credits we need to transfer to remote.
-	 * Rather than protect them using a slow spinlock, we put both into
-	 * a single atomic_t and update it using cmpxchg
-	 */
-	atomic_t		i_credits;
-
-	/* Protocol version specific information */
-	unsigned int		i_flowctl:1;	/* enable/disable flow ctl */
-	unsigned int		i_dma_local_lkey:1;
-	unsigned int		i_fastreg_posted:1; /* fastreg posted on this connection */
-	/* Batched completions */
-	unsigned int		i_unsignaled_wrs;
-	long			i_unsignaled_bytes;
-};
-
-/* This assumes that atomic_t is at least 32 bits */
-#define IB_GET_SEND_CREDITS(v)	((v) & 0xffff)
-#define IB_GET_POST_CREDITS(v)	((v) >> 16)
-#define IB_SET_SEND_CREDITS(v)	((v) & 0xffff)
-#define IB_SET_POST_CREDITS(v)	((v) << 16)
-
-struct rds_iw_cm_id {
-	struct list_head	list;
-	struct rdma_cm_id	*cm_id;
-};
-
-struct rds_iw_device {
-	struct list_head	list;
-	struct list_head	cm_id_list;
-	struct list_head	conn_list;
-	struct ib_device	*dev;
-	struct ib_pd		*pd;
-	struct ib_mr		*mr;
-	struct rds_iw_mr_pool	*mr_pool;
-	int			max_sge;
-	unsigned int		max_wrs;
-	unsigned int		dma_local_lkey:1;
-	spinlock_t		spinlock;	/* protect the above */
-};
-
-/* bits for i_ack_flags */
-#define IB_ACK_IN_FLIGHT	0
-#define IB_ACK_REQUESTED	1
-
-/* Magic WR_ID for ACKs */
-#define RDS_IW_ACK_WR_ID	((u64)0xffffffffffffffffULL)
-#define RDS_IW_REG_WR_ID	((u64)0xefefefefefefefefULL)
-#define RDS_IW_LOCAL_INV_WR_ID	((u64)0xdfdfdfdfdfdfdfdfULL)
-
-struct rds_iw_statistics {
-	uint64_t	s_iw_connect_raced;
-	uint64_t	s_iw_listen_closed_stale;
-	uint64_t	s_iw_tx_cq_call;
-	uint64_t	s_iw_tx_cq_event;
-	uint64_t	s_iw_tx_ring_full;
-	uint64_t	s_iw_tx_throttle;
-	uint64_t	s_iw_tx_sg_mapping_failure;
-	uint64_t	s_iw_tx_stalled;
-	uint64_t	s_iw_tx_credit_updates;
-	uint64_t	s_iw_rx_cq_call;
-	uint64_t	s_iw_rx_cq_event;
-	uint64_t	s_iw_rx_ring_empty;
-	uint64_t	s_iw_rx_refill_from_cq;
-	uint64_t	s_iw_rx_refill_from_thread;
-	uint64_t	s_iw_rx_alloc_limit;
-	uint64_t	s_iw_rx_credit_updates;
-	uint64_t	s_iw_ack_sent;
-	uint64_t	s_iw_ack_send_failure;
-	uint64_t	s_iw_ack_send_delayed;
-	uint64_t	s_iw_ack_send_piggybacked;
-	uint64_t	s_iw_ack_received;
-	uint64_t	s_iw_rdma_mr_alloc;
-	uint64_t	s_iw_rdma_mr_free;
-	uint64_t	s_iw_rdma_mr_used;
-	uint64_t	s_iw_rdma_mr_pool_flush;
-	uint64_t	s_iw_rdma_mr_pool_wait;
-	uint64_t	s_iw_rdma_mr_pool_depleted;
-};
-
-extern struct workqueue_struct *rds_iw_wq;
-
-/*
- * Fake ib_dma_sync_sg_for_{cpu,device} as long as ib_verbs.h
- * doesn't define it.
- */
-static inline void rds_iw_dma_sync_sg_for_cpu(struct ib_device *dev,
-		struct scatterlist *sg, unsigned int sg_dma_len, int direction)
-{
-	unsigned int i;
-
-	for (i = 0; i < sg_dma_len; ++i) {
-		ib_dma_sync_single_for_cpu(dev,
-				ib_sg_dma_address(dev, &sg[i]),
-				ib_sg_dma_len(dev, &sg[i]),
-				direction);
-	}
-}
-#define ib_dma_sync_sg_for_cpu	rds_iw_dma_sync_sg_for_cpu
-
-static inline void rds_iw_dma_sync_sg_for_device(struct ib_device *dev,
-		struct scatterlist *sg, unsigned int sg_dma_len, int direction)
-{
-	unsigned int i;
-
-	for (i = 0; i < sg_dma_len; ++i) {
-		ib_dma_sync_single_for_device(dev,
-				ib_sg_dma_address(dev, &sg[i]),
-				ib_sg_dma_len(dev, &sg[i]),
-				direction);
-	}
-}
-#define ib_dma_sync_sg_for_device	rds_iw_dma_sync_sg_for_device
-
-static inline u32 rds_iw_local_dma_lkey(struct rds_iw_connection *ic)
-{
-	return ic->i_dma_local_lkey ? ic->i_cm_id->device->local_dma_lkey : ic->i_mr->lkey;
-}
-
-/* ib.c */
-extern struct rds_transport rds_iw_transport;
-extern struct ib_client rds_iw_client;
-
-extern unsigned int fastreg_pool_size;
-extern unsigned int fastreg_message_size;
-
-extern spinlock_t iw_nodev_conns_lock;
-extern struct list_head iw_nodev_conns;
-
-/* ib_cm.c */
-int rds_iw_conn_alloc(struct rds_connection *conn, gfp_t gfp);
-void rds_iw_conn_free(void *arg);
-int rds_iw_conn_connect(struct rds_connection *conn);
-void rds_iw_conn_shutdown(struct rds_connection *conn);
-void rds_iw_state_change(struct sock *sk);
-int rds_iw_listen_init(void);
-void rds_iw_listen_stop(void);
-void __rds_iw_conn_error(struct rds_connection *conn, const char *, ...);
-int rds_iw_cm_handle_connect(struct rdma_cm_id *cm_id,
-			     struct rdma_cm_event *event);
-int rds_iw_cm_initiate_connect(struct rdma_cm_id *cm_id);
-void rds_iw_cm_connect_complete(struct rds_connection *conn,
-				struct rdma_cm_event *event);
-
-
-#define rds_iw_conn_error(conn, fmt...) \
-	__rds_iw_conn_error(conn, KERN_WARNING "RDS/IW: " fmt)
-
-/* ib_rdma.c */
-int rds_iw_update_cm_id(struct rds_iw_device *rds_iwdev, struct rdma_cm_id *cm_id);
-void rds_iw_add_conn(struct rds_iw_device *rds_iwdev, struct rds_connection *conn);
-void rds_iw_remove_conn(struct rds_iw_device *rds_iwdev, struct rds_connection *conn);
-void __rds_iw_destroy_conns(struct list_head *list, spinlock_t *list_lock);
-static inline void rds_iw_destroy_nodev_conns(void)
-{
-	__rds_iw_destroy_conns(&iw_nodev_conns, &iw_nodev_conns_lock);
-}
-static inline void rds_iw_destroy_conns(struct rds_iw_device *rds_iwdev)
-{
-	__rds_iw_destroy_conns(&rds_iwdev->conn_list, &rds_iwdev->spinlock);
-}
-struct rds_iw_mr_pool *rds_iw_create_mr_pool(struct rds_iw_device *);
-void rds_iw_get_mr_info(struct rds_iw_device *rds_iwdev, struct rds_info_rdma_connection *iinfo);
-void rds_iw_destroy_mr_pool(struct rds_iw_mr_pool *);
-void *rds_iw_get_mr(struct scatterlist *sg, unsigned long nents,
-		    struct rds_sock *rs, u32 *key_ret);
-void rds_iw_sync_mr(void *trans_private, int dir);
-void rds_iw_free_mr(void *trans_private, int invalidate);
-void rds_iw_flush_mrs(void);
-
-/* ib_recv.c */
-int rds_iw_recv_init(void);
-void rds_iw_recv_exit(void);
-int rds_iw_recv(struct rds_connection *conn);
-int rds_iw_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp,
-		       gfp_t page_gfp, int prefill);
-void rds_iw_inc_free(struct rds_incoming *inc);
-int rds_iw_inc_copy_to_user(struct rds_incoming *inc, struct iov_iter *to);
-void rds_iw_recv_cq_comp_handler(struct ib_cq *cq, void *context);
-void rds_iw_recv_tasklet_fn(unsigned long data);
-void rds_iw_recv_init_ring(struct rds_iw_connection *ic);
-void rds_iw_recv_clear_ring(struct rds_iw_connection *ic);
-void rds_iw_recv_init_ack(struct rds_iw_connection *ic);
-void rds_iw_attempt_ack(struct rds_iw_connection *ic);
-void rds_iw_ack_send_complete(struct rds_iw_connection *ic);
-u64 rds_iw_piggyb_ack(struct rds_iw_connection *ic);
-
-/* ib_ring.c */
-void rds_iw_ring_init(struct rds_iw_work_ring *ring, u32 nr);
-void rds_iw_ring_resize(struct rds_iw_work_ring *ring, u32 nr);
-u32 rds_iw_ring_alloc(struct rds_iw_work_ring *ring, u32 val, u32 *pos);
-void rds_iw_ring_free(struct rds_iw_work_ring *ring, u32 val);
-void rds_iw_ring_unalloc(struct rds_iw_work_ring *ring, u32 val);
-int rds_iw_ring_empty(struct rds_iw_work_ring *ring);
-int rds_iw_ring_low(struct rds_iw_work_ring *ring);
-u32 rds_iw_ring_oldest(struct rds_iw_work_ring *ring);
-u32 rds_iw_ring_completed(struct rds_iw_work_ring *ring, u32 wr_id, u32 oldest);
-extern wait_queue_head_t rds_iw_ring_empty_wait;
-
-/* ib_send.c */
-void rds_iw_xmit_complete(struct rds_connection *conn);
-int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm,
-		unsigned int hdr_off, unsigned int sg, unsigned int off);
-void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context);
-void rds_iw_send_init_ring(struct rds_iw_connection *ic);
-void rds_iw_send_clear_ring(struct rds_iw_connection *ic);
-int rds_iw_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op);
-void rds_iw_send_add_credits(struct rds_connection *conn, unsigned int credits);
-void rds_iw_advertise_credits(struct rds_connection *conn, unsigned int posted);
-int rds_iw_send_grab_credits(struct rds_iw_connection *ic, u32 wanted,
-			     u32 *adv_credits, int need_posted, int max_posted);
-
-/* ib_stats.c */
-DECLARE_PER_CPU(struct rds_iw_statistics, rds_iw_stats);
-#define rds_iw_stats_inc(member) rds_stats_inc_which(rds_iw_stats, member)
-unsigned int rds_iw_stats_info_copy(struct rds_info_iterator *iter,
-				    unsigned int avail);
-
-/* ib_sysctl.c */
-int rds_iw_sysctl_init(void);
-void rds_iw_sysctl_exit(void);
-extern unsigned long rds_iw_sysctl_max_send_wr;
-extern unsigned long rds_iw_sysctl_max_recv_wr;
-extern unsigned long rds_iw_sysctl_max_unsig_wrs;
-extern unsigned long rds_iw_sysctl_max_unsig_bytes;
-extern unsigned long rds_iw_sysctl_max_recv_allocation;
-extern unsigned int rds_iw_sysctl_flow_control;
-
-/*
- * Helper functions for getting/setting the header and data SGEs in
- * RDS packets (not RDMA)
- */
-static inline struct ib_sge *
-rds_iw_header_sge(struct rds_iw_connection *ic, struct ib_sge *sge)
-{
-	return &sge[0];
-}
-
-static inline struct ib_sge *
-rds_iw_data_sge(struct rds_iw_connection *ic, struct ib_sge *sge)
-{
-	return &sge[1];
-}
-
-#endif
diff --git a/net/rds/iw_cm.c b/net/rds/iw_cm.c
deleted file mode 100644
index aea4c91..0000000
--- a/net/rds/iw_cm.c
+++ /dev/null
@@ -1,769 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-#include <linux/in.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <linux/ratelimit.h>
-
-#include "rds.h"
-#include "iw.h"
-
-/*
- * Set the selected protocol version
- */
-static void rds_iw_set_protocol(struct rds_connection *conn, unsigned int version)
-{
-	conn->c_version = version;
-}
-
-/*
- * Set up flow control
- */
-static void rds_iw_set_flow_control(struct rds_connection *conn, u32 credits)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	if (rds_iw_sysctl_flow_control && credits != 0) {
-		/* We're doing flow control */
-		ic->i_flowctl = 1;
-		rds_iw_send_add_credits(conn, credits);
-	} else {
-		ic->i_flowctl = 0;
-	}
-}
-
-/*
- * Connection established.
- * We get here for both outgoing and incoming connection.
- */
-void rds_iw_cm_connect_complete(struct rds_connection *conn, struct rdma_cm_event *event)
-{
-	const struct rds_iw_connect_private *dp = NULL;
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct rds_iw_device *rds_iwdev;
-	int err;
-
-	if (event->param.conn.private_data_len) {
-		dp = event->param.conn.private_data;
-
-		rds_iw_set_protocol(conn,
-				RDS_PROTOCOL(dp->dp_protocol_major,
-					dp->dp_protocol_minor));
-		rds_iw_set_flow_control(conn, be32_to_cpu(dp->dp_credit));
-	}
-
-	/* update ib_device with this local ipaddr & conn */
-	rds_iwdev = ib_get_client_data(ic->i_cm_id->device, &rds_iw_client);
-	err = rds_iw_update_cm_id(rds_iwdev, ic->i_cm_id);
-	if (err)
-		printk(KERN_ERR "rds_iw_update_ipaddr failed (%d)\n", err);
-	rds_iw_add_conn(rds_iwdev, conn);
-
-	/* If the peer gave us the last packet it saw, process this as if
-	 * we had received a regular ACK. */
-	if (dp && dp->dp_ack_seq)
-		rds_send_drop_acked(conn, be64_to_cpu(dp->dp_ack_seq), NULL);
-
-	printk(KERN_NOTICE "RDS/IW: connected to %pI4<->%pI4 version %u.%u%s\n",
-			&conn->c_laddr, &conn->c_faddr,
-			RDS_PROTOCOL_MAJOR(conn->c_version),
-			RDS_PROTOCOL_MINOR(conn->c_version),
-			ic->i_flowctl ? ", flow control" : "");
-
-	rds_connect_complete(conn);
-}
-
-static void rds_iw_cm_fill_conn_param(struct rds_connection *conn,
-			struct rdma_conn_param *conn_param,
-			struct rds_iw_connect_private *dp,
-			u32 protocol_version)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	memset(conn_param, 0, sizeof(struct rdma_conn_param));
-	/* XXX tune these? */
-	conn_param->responder_resources = 1;
-	conn_param->initiator_depth = 1;
-
-	if (dp) {
-		memset(dp, 0, sizeof(*dp));
-		dp->dp_saddr = conn->c_laddr;
-		dp->dp_daddr = conn->c_faddr;
-		dp->dp_protocol_major = RDS_PROTOCOL_MAJOR(protocol_version);
-		dp->dp_protocol_minor = RDS_PROTOCOL_MINOR(protocol_version);
-		dp->dp_protocol_minor_mask = cpu_to_be16(RDS_IW_SUPPORTED_PROTOCOLS);
-		dp->dp_ack_seq = rds_iw_piggyb_ack(ic);
-
-		/* Advertise flow control */
-		if (ic->i_flowctl) {
-			unsigned int credits;
-
-			credits = IB_GET_POST_CREDITS(atomic_read(&ic->i_credits));
-			dp->dp_credit = cpu_to_be32(credits);
-			atomic_sub(IB_SET_POST_CREDITS(credits), &ic->i_credits);
-		}
-
-		conn_param->private_data = dp;
-		conn_param->private_data_len = sizeof(*dp);
-	}
-}
-
-static void rds_iw_cq_event_handler(struct ib_event *event, void *data)
-{
-	rdsdebug("event %u data %p\n", event->event, data);
-}
-
-static void rds_iw_qp_event_handler(struct ib_event *event, void *data)
-{
-	struct rds_connection *conn = data;
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	rdsdebug("conn %p ic %p event %u\n", conn, ic, event->event);
-
-	switch (event->event) {
-	case IB_EVENT_COMM_EST:
-		rdma_notify(ic->i_cm_id, IB_EVENT_COMM_EST);
-		break;
-	case IB_EVENT_QP_REQ_ERR:
-	case IB_EVENT_QP_FATAL:
-	default:
-		rdsdebug("Fatal QP Event %u "
-			"- connection %pI4->%pI4, reconnecting\n",
-			event->event, &conn->c_laddr,
-			&conn->c_faddr);
-		rds_conn_drop(conn);
-		break;
-	}
-}
-
-/*
- * Create a QP
- */
-static int rds_iw_init_qp_attrs(struct ib_qp_init_attr *attr,
-		struct rds_iw_device *rds_iwdev,
-		struct rds_iw_work_ring *send_ring,
-		void (*send_cq_handler)(struct ib_cq *, void *),
-		struct rds_iw_work_ring *recv_ring,
-		void (*recv_cq_handler)(struct ib_cq *, void *),
-		void *context)
-{
-	struct ib_device *dev = rds_iwdev->dev;
-	struct ib_cq_init_attr cq_attr = {};
-	unsigned int send_size, recv_size;
-	int ret;
-
-	/* The offset of 1 is to accommodate the additional ACK WR. */
-	send_size = min_t(unsigned int, rds_iwdev->max_wrs, rds_iw_sysctl_max_send_wr + 1);
-	recv_size = min_t(unsigned int, rds_iwdev->max_wrs, rds_iw_sysctl_max_recv_wr + 1);
-	rds_iw_ring_resize(send_ring, send_size - 1);
-	rds_iw_ring_resize(recv_ring, recv_size - 1);
-
-	memset(attr, 0, sizeof(*attr));
-	attr->event_handler = rds_iw_qp_event_handler;
-	attr->qp_context = context;
-	attr->cap.max_send_wr = send_size;
-	attr->cap.max_recv_wr = recv_size;
-	attr->cap.max_send_sge = rds_iwdev->max_sge;
-	attr->cap.max_recv_sge = RDS_IW_RECV_SGE;
-	attr->sq_sig_type = IB_SIGNAL_REQ_WR;
-	attr->qp_type = IB_QPT_RC;
-
-	cq_attr.cqe = send_size;
-	attr->send_cq = ib_create_cq(dev, send_cq_handler,
-				     rds_iw_cq_event_handler,
-				     context, &cq_attr);
-	if (IS_ERR(attr->send_cq)) {
-		ret = PTR_ERR(attr->send_cq);
-		attr->send_cq = NULL;
-		rdsdebug("ib_create_cq send failed: %d\n", ret);
-		goto out;
-	}
-
-	cq_attr.cqe = recv_size;
-	attr->recv_cq = ib_create_cq(dev, recv_cq_handler,
-				     rds_iw_cq_event_handler,
-				     context, &cq_attr);
-	if (IS_ERR(attr->recv_cq)) {
-		ret = PTR_ERR(attr->recv_cq);
-		attr->recv_cq = NULL;
-		rdsdebug("ib_create_cq send failed: %d\n", ret);
-		goto out;
-	}
-
-	ret = ib_req_notify_cq(attr->send_cq, IB_CQ_NEXT_COMP);
-	if (ret) {
-		rdsdebug("ib_req_notify_cq send failed: %d\n", ret);
-		goto out;
-	}
-
-	ret = ib_req_notify_cq(attr->recv_cq, IB_CQ_SOLICITED);
-	if (ret) {
-		rdsdebug("ib_req_notify_cq recv failed: %d\n", ret);
-		goto out;
-	}
-
-out:
-	if (ret) {
-		if (attr->send_cq)
-			ib_destroy_cq(attr->send_cq);
-		if (attr->recv_cq)
-			ib_destroy_cq(attr->recv_cq);
-	}
-	return ret;
-}
-
-/*
- * This needs to be very careful to not leave IS_ERR pointers around for
- * cleanup to trip over.
- */
-static int rds_iw_setup_qp(struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct ib_device *dev = ic->i_cm_id->device;
-	struct ib_qp_init_attr attr;
-	struct rds_iw_device *rds_iwdev;
-	int ret;
-
-	/* rds_iw_add_one creates a rds_iw_device object per IB device,
-	 * and allocates a protection domain, memory range and MR pool
-	 * for each.  If that fails for any reason, it will not register
-	 * the rds_iwdev at all.
-	 */
-	rds_iwdev = ib_get_client_data(dev, &rds_iw_client);
-	if (!rds_iwdev) {
-		printk_ratelimited(KERN_NOTICE "RDS/IW: No client_data for device %s\n",
-					dev->name);
-		return -EOPNOTSUPP;
-	}
-
-	/* Protection domain and memory range */
-	ic->i_pd = rds_iwdev->pd;
-	ic->i_mr = rds_iwdev->mr;
-
-	ret = rds_iw_init_qp_attrs(&attr, rds_iwdev,
-			&ic->i_send_ring, rds_iw_send_cq_comp_handler,
-			&ic->i_recv_ring, rds_iw_recv_cq_comp_handler,
-			conn);
-	if (ret < 0)
-		goto out;
-
-	ic->i_send_cq = attr.send_cq;
-	ic->i_recv_cq = attr.recv_cq;
-
-	/*
-	 * XXX this can fail if max_*_wr is too large?  Are we supposed
-	 * to back off until we get a value that the hardware can support?
-	 */
-	ret = rdma_create_qp(ic->i_cm_id, ic->i_pd, &attr);
-	if (ret) {
-		rdsdebug("rdma_create_qp failed: %d\n", ret);
-		goto out;
-	}
-
-	ic->i_send_hdrs = ib_dma_alloc_coherent(dev,
-					   ic->i_send_ring.w_nr *
-						sizeof(struct rds_header),
-					   &ic->i_send_hdrs_dma, GFP_KERNEL);
-	if (!ic->i_send_hdrs) {
-		ret = -ENOMEM;
-		rdsdebug("ib_dma_alloc_coherent send failed\n");
-		goto out;
-	}
-
-	ic->i_recv_hdrs = ib_dma_alloc_coherent(dev,
-					   ic->i_recv_ring.w_nr *
-						sizeof(struct rds_header),
-					   &ic->i_recv_hdrs_dma, GFP_KERNEL);
-	if (!ic->i_recv_hdrs) {
-		ret = -ENOMEM;
-		rdsdebug("ib_dma_alloc_coherent recv failed\n");
-		goto out;
-	}
-
-	ic->i_ack = ib_dma_alloc_coherent(dev, sizeof(struct rds_header),
-				       &ic->i_ack_dma, GFP_KERNEL);
-	if (!ic->i_ack) {
-		ret = -ENOMEM;
-		rdsdebug("ib_dma_alloc_coherent ack failed\n");
-		goto out;
-	}
-
-	ic->i_sends = vmalloc(ic->i_send_ring.w_nr * sizeof(struct rds_iw_send_work));
-	if (!ic->i_sends) {
-		ret = -ENOMEM;
-		rdsdebug("send allocation failed\n");
-		goto out;
-	}
-	rds_iw_send_init_ring(ic);
-
-	ic->i_recvs = vmalloc(ic->i_recv_ring.w_nr * sizeof(struct rds_iw_recv_work));
-	if (!ic->i_recvs) {
-		ret = -ENOMEM;
-		rdsdebug("recv allocation failed\n");
-		goto out;
-	}
-
-	rds_iw_recv_init_ring(ic);
-	rds_iw_recv_init_ack(ic);
-
-	/* Post receive buffers - as a side effect, this will update
-	 * the posted credit count. */
-	rds_iw_recv_refill(conn, GFP_KERNEL, GFP_HIGHUSER, 1);
-
-	rdsdebug("conn %p pd %p mr %p cq %p %p\n", conn, ic->i_pd, ic->i_mr,
-		 ic->i_send_cq, ic->i_recv_cq);
-
-out:
-	return ret;
-}
-
-static u32 rds_iw_protocol_compatible(const struct rds_iw_connect_private *dp)
-{
-	u16 common;
-	u32 version = 0;
-
-	/* rdma_cm private data is odd - when there is any private data in the
-	 * request, we will be given a pretty large buffer without telling us the
-	 * original size. The only way to tell the difference is by looking at
-	 * the contents, which are initialized to zero.
-	 * If the protocol version fields aren't set, this is a connection attempt
-	 * from an older version. This could could be 3.0 or 2.0 - we can't tell.
-	 * We really should have changed this for OFED 1.3 :-( */
-	if (dp->dp_protocol_major == 0)
-		return RDS_PROTOCOL_3_0;
-
-	common = be16_to_cpu(dp->dp_protocol_minor_mask) & RDS_IW_SUPPORTED_PROTOCOLS;
-	if (dp->dp_protocol_major == 3 && common) {
-		version = RDS_PROTOCOL_3_0;
-		while ((common >>= 1) != 0)
-			version++;
-	}
-	printk_ratelimited(KERN_NOTICE "RDS: Connection from %pI4 using "
-			"incompatible protocol version %u.%u\n",
-			&dp->dp_saddr,
-			dp->dp_protocol_major,
-			dp->dp_protocol_minor);
-	return version;
-}
-
-int rds_iw_cm_handle_connect(struct rdma_cm_id *cm_id,
-				    struct rdma_cm_event *event)
-{
-	const struct rds_iw_connect_private *dp = event->param.conn.private_data;
-	struct rds_iw_connect_private dp_rep;
-	struct rds_connection *conn = NULL;
-	struct rds_iw_connection *ic = NULL;
-	struct rdma_conn_param conn_param;
-	struct rds_iw_device *rds_iwdev;
-	u32 version;
-	int err, destroy = 1;
-
-	/* Check whether the remote protocol version matches ours. */
-	version = rds_iw_protocol_compatible(dp);
-	if (!version)
-		goto out;
-
-	rdsdebug("saddr %pI4 daddr %pI4 RDSv%u.%u\n",
-		 &dp->dp_saddr, &dp->dp_daddr,
-		 RDS_PROTOCOL_MAJOR(version), RDS_PROTOCOL_MINOR(version));
-
-	/* RDS/IW is not currently netns aware, thus init_net */
-	conn = rds_conn_create(&init_net, dp->dp_daddr, dp->dp_saddr,
-			       &rds_iw_transport, GFP_KERNEL);
-	if (IS_ERR(conn)) {
-		rdsdebug("rds_conn_create failed (%ld)\n", PTR_ERR(conn));
-		conn = NULL;
-		goto out;
-	}
-
-	/*
-	 * The connection request may occur while the
-	 * previous connection exist, e.g. in case of failover.
-	 * But as connections may be initiated simultaneously
-	 * by both hosts, we have a random backoff mechanism -
-	 * see the comment above rds_queue_reconnect()
-	 */
-	mutex_lock(&conn->c_cm_lock);
-	if (!rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_CONNECTING)) {
-		if (rds_conn_state(conn) == RDS_CONN_UP) {
-			rdsdebug("incoming connect while connecting\n");
-			rds_conn_drop(conn);
-			rds_iw_stats_inc(s_iw_listen_closed_stale);
-		} else
-		if (rds_conn_state(conn) == RDS_CONN_CONNECTING) {
-			/* Wait and see - our connect may still be succeeding */
-			rds_iw_stats_inc(s_iw_connect_raced);
-		}
-		mutex_unlock(&conn->c_cm_lock);
-		goto out;
-	}
-
-	ic = conn->c_transport_data;
-
-	rds_iw_set_protocol(conn, version);
-	rds_iw_set_flow_control(conn, be32_to_cpu(dp->dp_credit));
-
-	/* If the peer gave us the last packet it saw, process this as if
-	 * we had received a regular ACK. */
-	if (dp->dp_ack_seq)
-		rds_send_drop_acked(conn, be64_to_cpu(dp->dp_ack_seq), NULL);
-
-	BUG_ON(cm_id->context);
-	BUG_ON(ic->i_cm_id);
-
-	ic->i_cm_id = cm_id;
-	cm_id->context = conn;
-
-	rds_iwdev = ib_get_client_data(cm_id->device, &rds_iw_client);
-	ic->i_dma_local_lkey = rds_iwdev->dma_local_lkey;
-
-	/* We got halfway through setting up the ib_connection, if we
-	 * fail now, we have to take the long route out of this mess. */
-	destroy = 0;
-
-	err = rds_iw_setup_qp(conn);
-	if (err) {
-		rds_iw_conn_error(conn, "rds_iw_setup_qp failed (%d)\n", err);
-		mutex_unlock(&conn->c_cm_lock);
-		goto out;
-	}
-
-	rds_iw_cm_fill_conn_param(conn, &conn_param, &dp_rep, version);
-
-	/* rdma_accept() calls rdma_reject() internally if it fails */
-	err = rdma_accept(cm_id, &conn_param);
-	mutex_unlock(&conn->c_cm_lock);
-	if (err) {
-		rds_iw_conn_error(conn, "rdma_accept failed (%d)\n", err);
-		goto out;
-	}
-
-	return 0;
-
-out:
-	rdma_reject(cm_id, NULL, 0);
-	return destroy;
-}
-
-
-int rds_iw_cm_initiate_connect(struct rdma_cm_id *cm_id)
-{
-	struct rds_connection *conn = cm_id->context;
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct rdma_conn_param conn_param;
-	struct rds_iw_connect_private dp;
-	int ret;
-
-	/* If the peer doesn't do protocol negotiation, we must
-	 * default to RDSv3.0 */
-	rds_iw_set_protocol(conn, RDS_PROTOCOL_3_0);
-	ic->i_flowctl = rds_iw_sysctl_flow_control;	/* advertise flow control */
-
-	ret = rds_iw_setup_qp(conn);
-	if (ret) {
-		rds_iw_conn_error(conn, "rds_iw_setup_qp failed (%d)\n", ret);
-		goto out;
-	}
-
-	rds_iw_cm_fill_conn_param(conn, &conn_param, &dp, RDS_PROTOCOL_VERSION);
-
-	ret = rdma_connect(cm_id, &conn_param);
-	if (ret)
-		rds_iw_conn_error(conn, "rdma_connect failed (%d)\n", ret);
-
-out:
-	/* Beware - returning non-zero tells the rdma_cm to destroy
-	 * the cm_id. We should certainly not do it as long as we still
-	 * "own" the cm_id. */
-	if (ret) {
-		struct rds_iw_connection *ic = conn->c_transport_data;
-
-		if (ic->i_cm_id == cm_id)
-			ret = 0;
-	}
-	return ret;
-}
-
-int rds_iw_conn_connect(struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct rds_iw_device *rds_iwdev;
-	struct sockaddr_in src, dest;
-	int ret;
-
-	/* XXX I wonder what affect the port space has */
-	/* delegate cm event handler to rdma_transport */
-	ic->i_cm_id = rdma_create_id(&init_net, rds_rdma_cm_event_handler, conn,
-				     RDMA_PS_TCP, IB_QPT_RC);
-	if (IS_ERR(ic->i_cm_id)) {
-		ret = PTR_ERR(ic->i_cm_id);
-		ic->i_cm_id = NULL;
-		rdsdebug("rdma_create_id() failed: %d\n", ret);
-		goto out;
-	}
-
-	rdsdebug("created cm id %p for conn %p\n", ic->i_cm_id, conn);
-
-	src.sin_family = AF_INET;
-	src.sin_addr.s_addr = (__force u32)conn->c_laddr;
-	src.sin_port = (__force u16)htons(0);
-
-	/* First, bind to the local address and device. */
-	ret = rdma_bind_addr(ic->i_cm_id, (struct sockaddr *) &src);
-	if (ret) {
-		rdsdebug("rdma_bind_addr(%pI4) failed: %d\n",
-				&conn->c_laddr, ret);
-		rdma_destroy_id(ic->i_cm_id);
-		ic->i_cm_id = NULL;
-		goto out;
-	}
-
-	rds_iwdev = ib_get_client_data(ic->i_cm_id->device, &rds_iw_client);
-	ic->i_dma_local_lkey = rds_iwdev->dma_local_lkey;
-
-	dest.sin_family = AF_INET;
-	dest.sin_addr.s_addr = (__force u32)conn->c_faddr;
-	dest.sin_port = (__force u16)htons(RDS_PORT);
-
-	ret = rdma_resolve_addr(ic->i_cm_id, (struct sockaddr *)&src,
-				(struct sockaddr *)&dest,
-				RDS_RDMA_RESOLVE_TIMEOUT_MS);
-	if (ret) {
-		rdsdebug("addr resolve failed for cm id %p: %d\n", ic->i_cm_id,
-			 ret);
-		rdma_destroy_id(ic->i_cm_id);
-		ic->i_cm_id = NULL;
-	}
-
-out:
-	return ret;
-}
-
-/*
- * This is so careful about only cleaning up resources that were built up
- * so that it can be called at any point during startup.  In fact it
- * can be called multiple times for a given connection.
- */
-void rds_iw_conn_shutdown(struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	int err = 0;
-	struct ib_qp_attr qp_attr;
-
-	rdsdebug("cm %p pd %p cq %p %p qp %p\n", ic->i_cm_id,
-		 ic->i_pd, ic->i_send_cq, ic->i_recv_cq,
-		 ic->i_cm_id ? ic->i_cm_id->qp : NULL);
-
-	if (ic->i_cm_id) {
-		struct ib_device *dev = ic->i_cm_id->device;
-
-		rdsdebug("disconnecting cm %p\n", ic->i_cm_id);
-		err = rdma_disconnect(ic->i_cm_id);
-		if (err) {
-			/* Actually this may happen quite frequently, when
-			 * an outgoing connect raced with an incoming connect.
-			 */
-			rdsdebug("failed to disconnect, cm: %p err %d\n",
-				 ic->i_cm_id, err);
-		}
-
-		if (ic->i_cm_id->qp) {
-			qp_attr.qp_state = IB_QPS_ERR;
-			ib_modify_qp(ic->i_cm_id->qp, &qp_attr, IB_QP_STATE);
-		}
-
-		wait_event(rds_iw_ring_empty_wait,
-			rds_iw_ring_empty(&ic->i_send_ring) &&
-			rds_iw_ring_empty(&ic->i_recv_ring));
-
-		if (ic->i_send_hdrs)
-			ib_dma_free_coherent(dev,
-					   ic->i_send_ring.w_nr *
-						sizeof(struct rds_header),
-					   ic->i_send_hdrs,
-					   ic->i_send_hdrs_dma);
-
-		if (ic->i_recv_hdrs)
-			ib_dma_free_coherent(dev,
-					   ic->i_recv_ring.w_nr *
-						sizeof(struct rds_header),
-					   ic->i_recv_hdrs,
-					   ic->i_recv_hdrs_dma);
-
-		if (ic->i_ack)
-			ib_dma_free_coherent(dev, sizeof(struct rds_header),
-					     ic->i_ack, ic->i_ack_dma);
-
-		if (ic->i_sends)
-			rds_iw_send_clear_ring(ic);
-		if (ic->i_recvs)
-			rds_iw_recv_clear_ring(ic);
-
-		if (ic->i_cm_id->qp)
-			rdma_destroy_qp(ic->i_cm_id);
-		if (ic->i_send_cq)
-			ib_destroy_cq(ic->i_send_cq);
-		if (ic->i_recv_cq)
-			ib_destroy_cq(ic->i_recv_cq);
-
-		/*
-		 * If associated with an rds_iw_device:
-		 * 	Move connection back to the nodev list.
-		 * 	Remove cm_id from the device cm_id list.
-		 */
-		if (ic->rds_iwdev)
-			rds_iw_remove_conn(ic->rds_iwdev, conn);
-
-		rdma_destroy_id(ic->i_cm_id);
-
-		ic->i_cm_id = NULL;
-		ic->i_pd = NULL;
-		ic->i_mr = NULL;
-		ic->i_send_cq = NULL;
-		ic->i_recv_cq = NULL;
-		ic->i_send_hdrs = NULL;
-		ic->i_recv_hdrs = NULL;
-		ic->i_ack = NULL;
-	}
-	BUG_ON(ic->rds_iwdev);
-
-	/* Clear pending transmit */
-	if (ic->i_rm) {
-		rds_message_put(ic->i_rm);
-		ic->i_rm = NULL;
-	}
-
-	/* Clear the ACK state */
-	clear_bit(IB_ACK_IN_FLIGHT, &ic->i_ack_flags);
-#ifdef KERNEL_HAS_ATOMIC64
-	atomic64_set(&ic->i_ack_next, 0);
-#else
-	ic->i_ack_next = 0;
-#endif
-	ic->i_ack_recv = 0;
-
-	/* Clear flow control state */
-	ic->i_flowctl = 0;
-	atomic_set(&ic->i_credits, 0);
-
-	rds_iw_ring_init(&ic->i_send_ring, rds_iw_sysctl_max_send_wr);
-	rds_iw_ring_init(&ic->i_recv_ring, rds_iw_sysctl_max_recv_wr);
-
-	if (ic->i_iwinc) {
-		rds_inc_put(&ic->i_iwinc->ii_inc);
-		ic->i_iwinc = NULL;
-	}
-
-	vfree(ic->i_sends);
-	ic->i_sends = NULL;
-	vfree(ic->i_recvs);
-	ic->i_recvs = NULL;
-	rdsdebug("shutdown complete\n");
-}
-
-int rds_iw_conn_alloc(struct rds_connection *conn, gfp_t gfp)
-{
-	struct rds_iw_connection *ic;
-	unsigned long flags;
-
-	/* XXX too lazy? */
-	ic = kzalloc(sizeof(struct rds_iw_connection), gfp);
-	if (!ic)
-		return -ENOMEM;
-
-	INIT_LIST_HEAD(&ic->iw_node);
-	tasklet_init(&ic->i_recv_tasklet, rds_iw_recv_tasklet_fn,
-		     (unsigned long) ic);
-	mutex_init(&ic->i_recv_mutex);
-#ifndef KERNEL_HAS_ATOMIC64
-	spin_lock_init(&ic->i_ack_lock);
-#endif
-
-	/*
-	 * rds_iw_conn_shutdown() waits for these to be emptied so they
-	 * must be initialized before it can be called.
-	 */
-	rds_iw_ring_init(&ic->i_send_ring, rds_iw_sysctl_max_send_wr);
-	rds_iw_ring_init(&ic->i_recv_ring, rds_iw_sysctl_max_recv_wr);
-
-	ic->conn = conn;
-	conn->c_transport_data = ic;
-
-	spin_lock_irqsave(&iw_nodev_conns_lock, flags);
-	list_add_tail(&ic->iw_node, &iw_nodev_conns);
-	spin_unlock_irqrestore(&iw_nodev_conns_lock, flags);
-
-
-	rdsdebug("conn %p conn ic %p\n", conn, conn->c_transport_data);
-	return 0;
-}
-
-/*
- * Free a connection. Connection must be shut down and not set for reconnect.
- */
-void rds_iw_conn_free(void *arg)
-{
-	struct rds_iw_connection *ic = arg;
-	spinlock_t	*lock_ptr;
-
-	rdsdebug("ic %p\n", ic);
-
-	/*
-	 * Conn is either on a dev's list or on the nodev list.
-	 * A race with shutdown() or connect() would cause problems
-	 * (since rds_iwdev would change) but that should never happen.
-	 */
-	lock_ptr = ic->rds_iwdev ? &ic->rds_iwdev->spinlock : &iw_nodev_conns_lock;
-
-	spin_lock_irq(lock_ptr);
-	list_del(&ic->iw_node);
-	spin_unlock_irq(lock_ptr);
-
-	kfree(ic);
-}
-
-/*
- * An error occurred on the connection
- */
-void
-__rds_iw_conn_error(struct rds_connection *conn, const char *fmt, ...)
-{
-	va_list ap;
-
-	rds_conn_drop(conn);
-
-	va_start(ap, fmt);
-	vprintk(fmt, ap);
-	va_end(ap);
-}
diff --git a/net/rds/iw_rdma.c b/net/rds/iw_rdma.c
deleted file mode 100644
index b09a40c..0000000
--- a/net/rds/iw_rdma.c
+++ /dev/null
@@ -1,837 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/ratelimit.h>
-
-#include "rds.h"
-#include "iw.h"
-
-
-/*
- * This is stored as mr->r_trans_private.
- */
-struct rds_iw_mr {
-	struct rds_iw_device	*device;
-	struct rds_iw_mr_pool	*pool;
-	struct rdma_cm_id	*cm_id;
-
-	struct ib_mr	*mr;
-
-	struct rds_iw_mapping	mapping;
-	unsigned char		remap_count;
-};
-
-/*
- * Our own little MR pool
- */
-struct rds_iw_mr_pool {
-	struct rds_iw_device	*device;		/* back ptr to the device that owns us */
-
-	struct mutex		flush_lock;		/* serialize fmr invalidate */
-	struct work_struct	flush_worker;		/* flush worker */
-
-	spinlock_t		list_lock;		/* protect variables below */
-	atomic_t		item_count;		/* total # of MRs */
-	atomic_t		dirty_count;		/* # dirty of MRs */
-	struct list_head	dirty_list;		/* dirty mappings */
-	struct list_head	clean_list;		/* unused & unamapped MRs */
-	atomic_t		free_pinned;		/* memory pinned by free MRs */
-	unsigned long		max_message_size;	/* in pages */
-	unsigned long		max_items;
-	unsigned long		max_items_soft;
-	unsigned long		max_free_pinned;
-	int			max_pages;
-};
-
-static void rds_iw_flush_mr_pool(struct rds_iw_mr_pool *pool, int free_all);
-static void rds_iw_mr_pool_flush_worker(struct work_struct *work);
-static int rds_iw_init_reg(struct rds_iw_mr_pool *pool, struct rds_iw_mr *ibmr);
-static int rds_iw_map_reg(struct rds_iw_mr_pool *pool,
-			  struct rds_iw_mr *ibmr,
-			  struct scatterlist *sg, unsigned int nents);
-static void rds_iw_free_fastreg(struct rds_iw_mr_pool *pool, struct rds_iw_mr *ibmr);
-static unsigned int rds_iw_unmap_fastreg_list(struct rds_iw_mr_pool *pool,
-			struct list_head *unmap_list,
-			struct list_head *kill_list,
-			int *unpinned);
-static void rds_iw_destroy_fastreg(struct rds_iw_mr_pool *pool, struct rds_iw_mr *ibmr);
-
-static int rds_iw_get_device(struct sockaddr_in *src, struct sockaddr_in *dst,
-			     struct rds_iw_device **rds_iwdev,
-			     struct rdma_cm_id **cm_id)
-{
-	struct rds_iw_device *iwdev;
-	struct rds_iw_cm_id *i_cm_id;
-
-	*rds_iwdev = NULL;
-	*cm_id = NULL;
-
-	list_for_each_entry(iwdev, &rds_iw_devices, list) {
-		spin_lock_irq(&iwdev->spinlock);
-		list_for_each_entry(i_cm_id, &iwdev->cm_id_list, list) {
-			struct sockaddr_in *src_addr, *dst_addr;
-
-			src_addr = (struct sockaddr_in *)&i_cm_id->cm_id->route.addr.src_addr;
-			dst_addr = (struct sockaddr_in *)&i_cm_id->cm_id->route.addr.dst_addr;
-
-			rdsdebug("local ipaddr = %x port %d, "
-				 "remote ipaddr = %x port %d"
-				 "..looking for %x port %d, "
-				 "remote ipaddr = %x port %d\n",
-				src_addr->sin_addr.s_addr,
-				src_addr->sin_port,
-				dst_addr->sin_addr.s_addr,
-				dst_addr->sin_port,
-				src->sin_addr.s_addr,
-				src->sin_port,
-				dst->sin_addr.s_addr,
-				dst->sin_port);
-#ifdef WORKING_TUPLE_DETECTION
-			if (src_addr->sin_addr.s_addr == src->sin_addr.s_addr &&
-			    src_addr->sin_port == src->sin_port &&
-			    dst_addr->sin_addr.s_addr == dst->sin_addr.s_addr &&
-			    dst_addr->sin_port == dst->sin_port) {
-#else
-			/* FIXME - needs to compare the local and remote
-			 * ipaddr/port tuple, but the ipaddr is the only
-			 * available information in the rds_sock (as the rest are
-			 * zero'ed.  It doesn't appear to be properly populated
-			 * during connection setup...
-			 */
-			if (src_addr->sin_addr.s_addr == src->sin_addr.s_addr) {
-#endif
-				spin_unlock_irq(&iwdev->spinlock);
-				*rds_iwdev = iwdev;
-				*cm_id = i_cm_id->cm_id;
-				return 0;
-			}
-		}
-		spin_unlock_irq(&iwdev->spinlock);
-	}
-
-	return 1;
-}
-
-static int rds_iw_add_cm_id(struct rds_iw_device *rds_iwdev, struct rdma_cm_id *cm_id)
-{
-	struct rds_iw_cm_id *i_cm_id;
-
-	i_cm_id = kmalloc(sizeof *i_cm_id, GFP_KERNEL);
-	if (!i_cm_id)
-		return -ENOMEM;
-
-	i_cm_id->cm_id = cm_id;
-
-	spin_lock_irq(&rds_iwdev->spinlock);
-	list_add_tail(&i_cm_id->list, &rds_iwdev->cm_id_list);
-	spin_unlock_irq(&rds_iwdev->spinlock);
-
-	return 0;
-}
-
-static void rds_iw_remove_cm_id(struct rds_iw_device *rds_iwdev,
-				struct rdma_cm_id *cm_id)
-{
-	struct rds_iw_cm_id *i_cm_id;
-
-	spin_lock_irq(&rds_iwdev->spinlock);
-	list_for_each_entry(i_cm_id, &rds_iwdev->cm_id_list, list) {
-		if (i_cm_id->cm_id == cm_id) {
-			list_del(&i_cm_id->list);
-			kfree(i_cm_id);
-			break;
-		}
-	}
-	spin_unlock_irq(&rds_iwdev->spinlock);
-}
-
-
-int rds_iw_update_cm_id(struct rds_iw_device *rds_iwdev, struct rdma_cm_id *cm_id)
-{
-	struct sockaddr_in *src_addr, *dst_addr;
-	struct rds_iw_device *rds_iwdev_old;
-	struct rdma_cm_id *pcm_id;
-	int rc;
-
-	src_addr = (struct sockaddr_in *)&cm_id->route.addr.src_addr;
-	dst_addr = (struct sockaddr_in *)&cm_id->route.addr.dst_addr;
-
-	rc = rds_iw_get_device(src_addr, dst_addr, &rds_iwdev_old, &pcm_id);
-	if (rc)
-		rds_iw_remove_cm_id(rds_iwdev, cm_id);
-
-	return rds_iw_add_cm_id(rds_iwdev, cm_id);
-}
-
-void rds_iw_add_conn(struct rds_iw_device *rds_iwdev, struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	/* conn was previously on the nodev_conns_list */
-	spin_lock_irq(&iw_nodev_conns_lock);
-	BUG_ON(list_empty(&iw_nodev_conns));
-	BUG_ON(list_empty(&ic->iw_node));
-	list_del(&ic->iw_node);
-
-	spin_lock(&rds_iwdev->spinlock);
-	list_add_tail(&ic->iw_node, &rds_iwdev->conn_list);
-	spin_unlock(&rds_iwdev->spinlock);
-	spin_unlock_irq(&iw_nodev_conns_lock);
-
-	ic->rds_iwdev = rds_iwdev;
-}
-
-void rds_iw_remove_conn(struct rds_iw_device *rds_iwdev, struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	/* place conn on nodev_conns_list */
-	spin_lock(&iw_nodev_conns_lock);
-
-	spin_lock_irq(&rds_iwdev->spinlock);
-	BUG_ON(list_empty(&ic->iw_node));
-	list_del(&ic->iw_node);
-	spin_unlock_irq(&rds_iwdev->spinlock);
-
-	list_add_tail(&ic->iw_node, &iw_nodev_conns);
-
-	spin_unlock(&iw_nodev_conns_lock);
-
-	rds_iw_remove_cm_id(ic->rds_iwdev, ic->i_cm_id);
-	ic->rds_iwdev = NULL;
-}
-
-void __rds_iw_destroy_conns(struct list_head *list, spinlock_t *list_lock)
-{
-	struct rds_iw_connection *ic, *_ic;
-	LIST_HEAD(tmp_list);
-
-	/* avoid calling conn_destroy with irqs off */
-	spin_lock_irq(list_lock);
-	list_splice(list, &tmp_list);
-	INIT_LIST_HEAD(list);
-	spin_unlock_irq(list_lock);
-
-	list_for_each_entry_safe(ic, _ic, &tmp_list, iw_node)
-		rds_conn_destroy(ic->conn);
-}
-
-static void rds_iw_set_scatterlist(struct rds_iw_scatterlist *sg,
-		struct scatterlist *list, unsigned int sg_len)
-{
-	sg->list = list;
-	sg->len = sg_len;
-	sg->dma_len = 0;
-	sg->dma_npages = 0;
-	sg->bytes = 0;
-}
-
-static int rds_iw_map_scatterlist(struct rds_iw_device *rds_iwdev,
-				  struct rds_iw_scatterlist *sg)
-{
-	struct ib_device *dev = rds_iwdev->dev;
-	int i, ret;
-
-	WARN_ON(sg->dma_len);
-
-	sg->dma_len = ib_dma_map_sg(dev, sg->list, sg->len, DMA_BIDIRECTIONAL);
-	if (unlikely(!sg->dma_len)) {
-		printk(KERN_WARNING "RDS/IW: dma_map_sg failed!\n");
-		return -EBUSY;
-	}
-
-	sg->bytes = 0;
-	sg->dma_npages = 0;
-
-	ret = -EINVAL;
-	for (i = 0; i < sg->dma_len; ++i) {
-		unsigned int dma_len = ib_sg_dma_len(dev, &sg->list[i]);
-		u64 dma_addr = ib_sg_dma_address(dev, &sg->list[i]);
-		u64 end_addr;
-
-		sg->bytes += dma_len;
-
-		end_addr = dma_addr + dma_len;
-		if (dma_addr & PAGE_MASK) {
-			if (i > 0)
-				goto out_unmap;
-			dma_addr &= ~PAGE_MASK;
-		}
-		if (end_addr & PAGE_MASK) {
-			if (i < sg->dma_len - 1)
-				goto out_unmap;
-			end_addr = (end_addr + PAGE_MASK) & ~PAGE_MASK;
-		}
-
-		sg->dma_npages += (end_addr - dma_addr) >> PAGE_SHIFT;
-	}
-
-	/* Now gather the dma addrs into one list */
-	if (sg->dma_npages > fastreg_message_size)
-		goto out_unmap;
-
-
-
-	return 0;
-
-out_unmap:
-	ib_dma_unmap_sg(rds_iwdev->dev, sg->list, sg->len, DMA_BIDIRECTIONAL);
-	sg->dma_len = 0;
-	return ret;
-}
-
-
-struct rds_iw_mr_pool *rds_iw_create_mr_pool(struct rds_iw_device *rds_iwdev)
-{
-	struct rds_iw_mr_pool *pool;
-
-	pool = kzalloc(sizeof(*pool), GFP_KERNEL);
-	if (!pool) {
-		printk(KERN_WARNING "RDS/IW: rds_iw_create_mr_pool alloc error\n");
-		return ERR_PTR(-ENOMEM);
-	}
-
-	pool->device = rds_iwdev;
-	INIT_LIST_HEAD(&pool->dirty_list);
-	INIT_LIST_HEAD(&pool->clean_list);
-	mutex_init(&pool->flush_lock);
-	spin_lock_init(&pool->list_lock);
-	INIT_WORK(&pool->flush_worker, rds_iw_mr_pool_flush_worker);
-
-	pool->max_message_size = fastreg_message_size;
-	pool->max_items = fastreg_pool_size;
-	pool->max_free_pinned = pool->max_items * pool->max_message_size / 4;
-	pool->max_pages = fastreg_message_size;
-
-	/* We never allow more than max_items MRs to be allocated.
-	 * When we exceed more than max_items_soft, we start freeing
-	 * items more aggressively.
-	 * Make sure that max_items > max_items_soft > max_items / 2
-	 */
-	pool->max_items_soft = pool->max_items * 3 / 4;
-
-	return pool;
-}
-
-void rds_iw_get_mr_info(struct rds_iw_device *rds_iwdev, struct rds_info_rdma_connection *iinfo)
-{
-	struct rds_iw_mr_pool *pool = rds_iwdev->mr_pool;
-
-	iinfo->rdma_mr_max = pool->max_items;
-	iinfo->rdma_mr_size = pool->max_pages;
-}
-
-void rds_iw_destroy_mr_pool(struct rds_iw_mr_pool *pool)
-{
-	flush_workqueue(rds_wq);
-	rds_iw_flush_mr_pool(pool, 1);
-	BUG_ON(atomic_read(&pool->item_count));
-	BUG_ON(atomic_read(&pool->free_pinned));
-	kfree(pool);
-}
-
-static inline struct rds_iw_mr *rds_iw_reuse_fmr(struct rds_iw_mr_pool *pool)
-{
-	struct rds_iw_mr *ibmr = NULL;
-	unsigned long flags;
-
-	spin_lock_irqsave(&pool->list_lock, flags);
-	if (!list_empty(&pool->clean_list)) {
-		ibmr = list_entry(pool->clean_list.next, struct rds_iw_mr, mapping.m_list);
-		list_del_init(&ibmr->mapping.m_list);
-	}
-	spin_unlock_irqrestore(&pool->list_lock, flags);
-
-	return ibmr;
-}
-
-static struct rds_iw_mr *rds_iw_alloc_mr(struct rds_iw_device *rds_iwdev)
-{
-	struct rds_iw_mr_pool *pool = rds_iwdev->mr_pool;
-	struct rds_iw_mr *ibmr = NULL;
-	int err = 0, iter = 0;
-
-	while (1) {
-		ibmr = rds_iw_reuse_fmr(pool);
-		if (ibmr)
-			return ibmr;
-
-		/* No clean MRs - now we have the choice of either
-		 * allocating a fresh MR up to the limit imposed by the
-		 * driver, or flush any dirty unused MRs.
-		 * We try to avoid stalling in the send path if possible,
-		 * so we allocate as long as we're allowed to.
-		 *
-		 * We're fussy with enforcing the FMR limit, though. If the driver
-		 * tells us we can't use more than N fmrs, we shouldn't start
-		 * arguing with it */
-		if (atomic_inc_return(&pool->item_count) <= pool->max_items)
-			break;
-
-		atomic_dec(&pool->item_count);
-
-		if (++iter > 2) {
-			rds_iw_stats_inc(s_iw_rdma_mr_pool_depleted);
-			return ERR_PTR(-EAGAIN);
-		}
-
-		/* We do have some empty MRs. Flush them out. */
-		rds_iw_stats_inc(s_iw_rdma_mr_pool_wait);
-		rds_iw_flush_mr_pool(pool, 0);
-	}
-
-	ibmr = kzalloc(sizeof(*ibmr), GFP_KERNEL);
-	if (!ibmr) {
-		err = -ENOMEM;
-		goto out_no_cigar;
-	}
-
-	spin_lock_init(&ibmr->mapping.m_lock);
-	INIT_LIST_HEAD(&ibmr->mapping.m_list);
-	ibmr->mapping.m_mr = ibmr;
-
-	err = rds_iw_init_reg(pool, ibmr);
-	if (err)
-		goto out_no_cigar;
-
-	rds_iw_stats_inc(s_iw_rdma_mr_alloc);
-	return ibmr;
-
-out_no_cigar:
-	if (ibmr) {
-		rds_iw_destroy_fastreg(pool, ibmr);
-		kfree(ibmr);
-	}
-	atomic_dec(&pool->item_count);
-	return ERR_PTR(err);
-}
-
-void rds_iw_sync_mr(void *trans_private, int direction)
-{
-	struct rds_iw_mr *ibmr = trans_private;
-	struct rds_iw_device *rds_iwdev = ibmr->device;
-
-	switch (direction) {
-	case DMA_FROM_DEVICE:
-		ib_dma_sync_sg_for_cpu(rds_iwdev->dev, ibmr->mapping.m_sg.list,
-			ibmr->mapping.m_sg.dma_len, DMA_BIDIRECTIONAL);
-		break;
-	case DMA_TO_DEVICE:
-		ib_dma_sync_sg_for_device(rds_iwdev->dev, ibmr->mapping.m_sg.list,
-			ibmr->mapping.m_sg.dma_len, DMA_BIDIRECTIONAL);
-		break;
-	}
-}
-
-/*
- * Flush our pool of MRs.
- * At a minimum, all currently unused MRs are unmapped.
- * If the number of MRs allocated exceeds the limit, we also try
- * to free as many MRs as needed to get back to this limit.
- */
-static void rds_iw_flush_mr_pool(struct rds_iw_mr_pool *pool, int free_all)
-{
-	struct rds_iw_mr *ibmr, *next;
-	LIST_HEAD(unmap_list);
-	LIST_HEAD(kill_list);
-	unsigned long flags;
-	unsigned int nfreed = 0, ncleaned = 0, unpinned = 0;
-
-	rds_iw_stats_inc(s_iw_rdma_mr_pool_flush);
-
-	mutex_lock(&pool->flush_lock);
-
-	spin_lock_irqsave(&pool->list_lock, flags);
-	/* Get the list of all mappings to be destroyed */
-	list_splice_init(&pool->dirty_list, &unmap_list);
-	if (free_all)
-		list_splice_init(&pool->clean_list, &kill_list);
-	spin_unlock_irqrestore(&pool->list_lock, flags);
-
-	/* Batched invalidate of dirty MRs.
-	 * For FMR based MRs, the mappings on the unmap list are
-	 * actually members of an ibmr (ibmr->mapping). They either
-	 * migrate to the kill_list, or have been cleaned and should be
-	 * moved to the clean_list.
-	 * For fastregs, they will be dynamically allocated, and
-	 * will be destroyed by the unmap function.
-	 */
-	if (!list_empty(&unmap_list)) {
-		ncleaned = rds_iw_unmap_fastreg_list(pool, &unmap_list,
-						     &kill_list, &unpinned);
-		/* If we've been asked to destroy all MRs, move those
-		 * that were simply cleaned to the kill list */
-		if (free_all)
-			list_splice_init(&unmap_list, &kill_list);
-	}
-
-	/* Destroy any MRs that are past their best before date */
-	list_for_each_entry_safe(ibmr, next, &kill_list, mapping.m_list) {
-		rds_iw_stats_inc(s_iw_rdma_mr_free);
-		list_del(&ibmr->mapping.m_list);
-		rds_iw_destroy_fastreg(pool, ibmr);
-		kfree(ibmr);
-		nfreed++;
-	}
-
-	/* Anything that remains are laundered ibmrs, which we can add
-	 * back to the clean list. */
-	if (!list_empty(&unmap_list)) {
-		spin_lock_irqsave(&pool->list_lock, flags);
-		list_splice(&unmap_list, &pool->clean_list);
-		spin_unlock_irqrestore(&pool->list_lock, flags);
-	}
-
-	atomic_sub(unpinned, &pool->free_pinned);
-	atomic_sub(ncleaned, &pool->dirty_count);
-	atomic_sub(nfreed, &pool->item_count);
-
-	mutex_unlock(&pool->flush_lock);
-}
-
-static void rds_iw_mr_pool_flush_worker(struct work_struct *work)
-{
-	struct rds_iw_mr_pool *pool = container_of(work, struct rds_iw_mr_pool, flush_worker);
-
-	rds_iw_flush_mr_pool(pool, 0);
-}
-
-void rds_iw_free_mr(void *trans_private, int invalidate)
-{
-	struct rds_iw_mr *ibmr = trans_private;
-	struct rds_iw_mr_pool *pool = ibmr->device->mr_pool;
-
-	rdsdebug("RDS/IW: free_mr nents %u\n", ibmr->mapping.m_sg.len);
-	if (!pool)
-		return;
-
-	/* Return it to the pool's free list */
-	rds_iw_free_fastreg(pool, ibmr);
-
-	/* If we've pinned too many pages, request a flush */
-	if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned ||
-	    atomic_read(&pool->dirty_count) >= pool->max_items / 10)
-		queue_work(rds_wq, &pool->flush_worker);
-
-	if (invalidate) {
-		if (likely(!in_interrupt())) {
-			rds_iw_flush_mr_pool(pool, 0);
-		} else {
-			/* We get here if the user created a MR marked
-			 * as use_once and invalidate at the same time. */
-			queue_work(rds_wq, &pool->flush_worker);
-		}
-	}
-}
-
-void rds_iw_flush_mrs(void)
-{
-	struct rds_iw_device *rds_iwdev;
-
-	list_for_each_entry(rds_iwdev, &rds_iw_devices, list) {
-		struct rds_iw_mr_pool *pool = rds_iwdev->mr_pool;
-
-		if (pool)
-			rds_iw_flush_mr_pool(pool, 0);
-	}
-}
-
-void *rds_iw_get_mr(struct scatterlist *sg, unsigned long nents,
-		    struct rds_sock *rs, u32 *key_ret)
-{
-	struct rds_iw_device *rds_iwdev;
-	struct rds_iw_mr *ibmr = NULL;
-	struct rdma_cm_id *cm_id;
-	struct sockaddr_in src = {
-		.sin_addr.s_addr = rs->rs_bound_addr,
-		.sin_port = rs->rs_bound_port,
-	};
-	struct sockaddr_in dst = {
-		.sin_addr.s_addr = rs->rs_conn_addr,
-		.sin_port = rs->rs_conn_port,
-	};
-	int ret;
-
-	ret = rds_iw_get_device(&src, &dst, &rds_iwdev, &cm_id);
-	if (ret || !cm_id) {
-		ret = -ENODEV;
-		goto out;
-	}
-
-	if (!rds_iwdev->mr_pool) {
-		ret = -ENODEV;
-		goto out;
-	}
-
-	ibmr = rds_iw_alloc_mr(rds_iwdev);
-	if (IS_ERR(ibmr))
-		return ibmr;
-
-	ibmr->cm_id = cm_id;
-	ibmr->device = rds_iwdev;
-
-	ret = rds_iw_map_reg(rds_iwdev->mr_pool, ibmr, sg, nents);
-	if (ret == 0)
-		*key_ret = ibmr->mr->rkey;
-	else
-		printk(KERN_WARNING "RDS/IW: failed to map mr (errno=%d)\n", ret);
-
-out:
-	if (ret) {
-		if (ibmr)
-			rds_iw_free_mr(ibmr, 0);
-		ibmr = ERR_PTR(ret);
-	}
-	return ibmr;
-}
-
-/*
- * iWARP reg handling
- *
- * The life cycle of a fastreg registration is a bit different from
- * FMRs.
- * The idea behind fastreg is to have one MR, to which we bind different
- * mappings over time. To avoid stalling on the expensive map and invalidate
- * operations, these operations are pipelined on the same send queue on
- * which we want to send the message containing the r_key.
- *
- * This creates a bit of a problem for us, as we do not have the destination
- * IP in GET_MR, so the connection must be setup prior to the GET_MR call for
- * RDMA to be correctly setup.  If a fastreg request is present, rds_iw_xmit
- * will try to queue a LOCAL_INV (if needed) and a REG_MR work request
- * before queuing the SEND. When completions for these arrive, they are
- * dispatched to the MR has a bit set showing that RDMa can be performed.
- *
- * There is another interesting aspect that's related to invalidation.
- * The application can request that a mapping is invalidated in FREE_MR.
- * The expectation there is that this invalidation step includes ALL
- * PREVIOUSLY FREED MRs.
- */
-static int rds_iw_init_reg(struct rds_iw_mr_pool *pool,
-			   struct rds_iw_mr *ibmr)
-{
-	struct rds_iw_device *rds_iwdev = pool->device;
-	struct ib_mr *mr;
-	int err;
-
-	mr = ib_alloc_mr(rds_iwdev->pd, IB_MR_TYPE_MEM_REG,
-			 pool->max_message_size);
-	if (IS_ERR(mr)) {
-		err = PTR_ERR(mr);
-
-		printk(KERN_WARNING "RDS/IW: ib_alloc_mr failed (err=%d)\n", err);
-		return err;
-	}
-
-	ibmr->mr = mr;
-	return 0;
-}
-
-static int rds_iw_rdma_reg_mr(struct rds_iw_mapping *mapping)
-{
-	struct rds_iw_mr *ibmr = mapping->m_mr;
-	struct rds_iw_scatterlist *m_sg = &mapping->m_sg;
-	struct ib_reg_wr reg_wr;
-	struct ib_send_wr *failed_wr;
-	int ret, n;
-
-	n = ib_map_mr_sg_zbva(ibmr->mr, m_sg->list, m_sg->len, PAGE_SIZE);
-	if (unlikely(n != m_sg->len))
-		return n < 0 ? n : -EINVAL;
-
-	reg_wr.wr.next = NULL;
-	reg_wr.wr.opcode = IB_WR_REG_MR;
-	reg_wr.wr.wr_id = RDS_IW_REG_WR_ID;
-	reg_wr.wr.num_sge = 0;
-	reg_wr.mr = ibmr->mr;
-	reg_wr.key = mapping->m_rkey;
-	reg_wr.access = IB_ACCESS_LOCAL_WRITE |
-			IB_ACCESS_REMOTE_READ |
-			IB_ACCESS_REMOTE_WRITE;
-
-	/*
-	 * Perform a WR for the reg_mr. Each individual page
-	 * in the sg list is added to the fast reg page list and placed
-	 * inside the reg_mr WR.  The key used is a rolling 8bit
-	 * counter, which should guarantee uniqueness.
-	 */
-	ib_update_fast_reg_key(ibmr->mr, ibmr->remap_count++);
-	mapping->m_rkey = ibmr->mr->rkey;
-
-	failed_wr = &reg_wr.wr;
-	ret = ib_post_send(ibmr->cm_id->qp, &reg_wr.wr, &failed_wr);
-	BUG_ON(failed_wr != &reg_wr.wr);
-	if (ret)
-		printk_ratelimited(KERN_WARNING "RDS/IW: %s:%d ib_post_send returned %d\n",
-			__func__, __LINE__, ret);
-	return ret;
-}
-
-static int rds_iw_rdma_fastreg_inv(struct rds_iw_mr *ibmr)
-{
-	struct ib_send_wr s_wr, *failed_wr;
-	int ret = 0;
-
-	if (!ibmr->cm_id->qp || !ibmr->mr)
-		goto out;
-
-	memset(&s_wr, 0, sizeof(s_wr));
-	s_wr.wr_id = RDS_IW_LOCAL_INV_WR_ID;
-	s_wr.opcode = IB_WR_LOCAL_INV;
-	s_wr.ex.invalidate_rkey = ibmr->mr->rkey;
-	s_wr.send_flags = IB_SEND_SIGNALED;
-
-	failed_wr = &s_wr;
-	ret = ib_post_send(ibmr->cm_id->qp, &s_wr, &failed_wr);
-	if (ret) {
-		printk_ratelimited(KERN_WARNING "RDS/IW: %s:%d ib_post_send returned %d\n",
-			__func__, __LINE__, ret);
-		goto out;
-	}
-out:
-	return ret;
-}
-
-static int rds_iw_map_reg(struct rds_iw_mr_pool *pool,
-			  struct rds_iw_mr *ibmr,
-			  struct scatterlist *sg,
-			  unsigned int sg_len)
-{
-	struct rds_iw_device *rds_iwdev = pool->device;
-	struct rds_iw_mapping *mapping = &ibmr->mapping;
-	u64 *dma_pages;
-	int ret = 0;
-
-	rds_iw_set_scatterlist(&mapping->m_sg, sg, sg_len);
-
-	ret = rds_iw_map_scatterlist(rds_iwdev, &mapping->m_sg);
-	if (ret) {
-		dma_pages = NULL;
-		goto out;
-	}
-
-	if (mapping->m_sg.dma_len > pool->max_message_size) {
-		ret = -EMSGSIZE;
-		goto out;
-	}
-
-	ret = rds_iw_rdma_reg_mr(mapping);
-	if (ret)
-		goto out;
-
-	rds_iw_stats_inc(s_iw_rdma_mr_used);
-
-out:
-	kfree(dma_pages);
-
-	return ret;
-}
-
-/*
- * "Free" a fastreg MR.
- */
-static void rds_iw_free_fastreg(struct rds_iw_mr_pool *pool,
-		struct rds_iw_mr *ibmr)
-{
-	unsigned long flags;
-	int ret;
-
-	if (!ibmr->mapping.m_sg.dma_len)
-		return;
-
-	ret = rds_iw_rdma_fastreg_inv(ibmr);
-	if (ret)
-		return;
-
-	/* Try to post the LOCAL_INV WR to the queue. */
-	spin_lock_irqsave(&pool->list_lock, flags);
-
-	list_add_tail(&ibmr->mapping.m_list, &pool->dirty_list);
-	atomic_add(ibmr->mapping.m_sg.len, &pool->free_pinned);
-	atomic_inc(&pool->dirty_count);
-
-	spin_unlock_irqrestore(&pool->list_lock, flags);
-}
-
-static unsigned int rds_iw_unmap_fastreg_list(struct rds_iw_mr_pool *pool,
-				struct list_head *unmap_list,
-				struct list_head *kill_list,
-				int *unpinned)
-{
-	struct rds_iw_mapping *mapping, *next;
-	unsigned int ncleaned = 0;
-	LIST_HEAD(laundered);
-
-	/* Batched invalidation of fastreg MRs.
-	 * Why do we do it this way, even though we could pipeline unmap
-	 * and remap? The reason is the application semantics - when the
-	 * application requests an invalidation of MRs, it expects all
-	 * previously released R_Keys to become invalid.
-	 *
-	 * If we implement MR reuse naively, we risk memory corruption
-	 * (this has actually been observed). So the default behavior
-	 * requires that a MR goes through an explicit unmap operation before
-	 * we can reuse it again.
-	 *
-	 * We could probably improve on this a little, by allowing immediate
-	 * reuse of a MR on the same socket (eg you could add small
-	 * cache of unused MRs to strct rds_socket - GET_MR could grab one
-	 * of these without requiring an explicit invalidate).
-	 */
-	while (!list_empty(unmap_list)) {
-		unsigned long flags;
-
-		spin_lock_irqsave(&pool->list_lock, flags);
-		list_for_each_entry_safe(mapping, next, unmap_list, m_list) {
-			*unpinned += mapping->m_sg.len;
-			list_move(&mapping->m_list, &laundered);
-			ncleaned++;
-		}
-		spin_unlock_irqrestore(&pool->list_lock, flags);
-	}
-
-	/* Move all laundered mappings back to the unmap list.
-	 * We do not kill any WRs right now - it doesn't seem the
-	 * fastreg API has a max_remap limit. */
-	list_splice_init(&laundered, unmap_list);
-
-	return ncleaned;
-}
-
-static void rds_iw_destroy_fastreg(struct rds_iw_mr_pool *pool,
-		struct rds_iw_mr *ibmr)
-{
-	if (ibmr->mr)
-		ib_dereg_mr(ibmr->mr);
-}
diff --git a/net/rds/iw_recv.c b/net/rds/iw_recv.c
deleted file mode 100644
index a66d179..0000000
--- a/net/rds/iw_recv.c
+++ /dev/null
@@ -1,904 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/pci.h>
-#include <linux/dma-mapping.h>
-#include <rdma/rdma_cm.h>
-
-#include "rds.h"
-#include "iw.h"
-
-static struct kmem_cache *rds_iw_incoming_slab;
-static struct kmem_cache *rds_iw_frag_slab;
-static atomic_t	rds_iw_allocation = ATOMIC_INIT(0);
-
-static void rds_iw_frag_drop_page(struct rds_page_frag *frag)
-{
-	rdsdebug("frag %p page %p\n", frag, frag->f_page);
-	__free_page(frag->f_page);
-	frag->f_page = NULL;
-}
-
-static void rds_iw_frag_free(struct rds_page_frag *frag)
-{
-	rdsdebug("frag %p page %p\n", frag, frag->f_page);
-	BUG_ON(frag->f_page);
-	kmem_cache_free(rds_iw_frag_slab, frag);
-}
-
-/*
- * We map a page at a time.  Its fragments are posted in order.  This
- * is called in fragment order as the fragments get send completion events.
- * Only the last frag in the page performs the unmapping.
- *
- * It's OK for ring cleanup to call this in whatever order it likes because
- * DMA is not in flight and so we can unmap while other ring entries still
- * hold page references in their frags.
- */
-static void rds_iw_recv_unmap_page(struct rds_iw_connection *ic,
-				   struct rds_iw_recv_work *recv)
-{
-	struct rds_page_frag *frag = recv->r_frag;
-
-	rdsdebug("recv %p frag %p page %p\n", recv, frag, frag->f_page);
-	if (frag->f_mapped)
-		ib_dma_unmap_page(ic->i_cm_id->device,
-			       frag->f_mapped,
-			       RDS_FRAG_SIZE, DMA_FROM_DEVICE);
-	frag->f_mapped = 0;
-}
-
-void rds_iw_recv_init_ring(struct rds_iw_connection *ic)
-{
-	struct rds_iw_recv_work *recv;
-	u32 i;
-
-	for (i = 0, recv = ic->i_recvs; i < ic->i_recv_ring.w_nr; i++, recv++) {
-		struct ib_sge *sge;
-
-		recv->r_iwinc = NULL;
-		recv->r_frag = NULL;
-
-		recv->r_wr.next = NULL;
-		recv->r_wr.wr_id = i;
-		recv->r_wr.sg_list = recv->r_sge;
-		recv->r_wr.num_sge = RDS_IW_RECV_SGE;
-
-		sge = rds_iw_data_sge(ic, recv->r_sge);
-		sge->addr = 0;
-		sge->length = RDS_FRAG_SIZE;
-		sge->lkey = 0;
-
-		sge = rds_iw_header_sge(ic, recv->r_sge);
-		sge->addr = ic->i_recv_hdrs_dma + (i * sizeof(struct rds_header));
-		sge->length = sizeof(struct rds_header);
-		sge->lkey = 0;
-	}
-}
-
-static void rds_iw_recv_clear_one(struct rds_iw_connection *ic,
-				  struct rds_iw_recv_work *recv)
-{
-	if (recv->r_iwinc) {
-		rds_inc_put(&recv->r_iwinc->ii_inc);
-		recv->r_iwinc = NULL;
-	}
-	if (recv->r_frag) {
-		rds_iw_recv_unmap_page(ic, recv);
-		if (recv->r_frag->f_page)
-			rds_iw_frag_drop_page(recv->r_frag);
-		rds_iw_frag_free(recv->r_frag);
-		recv->r_frag = NULL;
-	}
-}
-
-void rds_iw_recv_clear_ring(struct rds_iw_connection *ic)
-{
-	u32 i;
-
-	for (i = 0; i < ic->i_recv_ring.w_nr; i++)
-		rds_iw_recv_clear_one(ic, &ic->i_recvs[i]);
-
-	if (ic->i_frag.f_page)
-		rds_iw_frag_drop_page(&ic->i_frag);
-}
-
-static int rds_iw_recv_refill_one(struct rds_connection *conn,
-				  struct rds_iw_recv_work *recv,
-				  gfp_t kptr_gfp, gfp_t page_gfp)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	dma_addr_t dma_addr;
-	struct ib_sge *sge;
-	int ret = -ENOMEM;
-
-	if (!recv->r_iwinc) {
-		if (!atomic_add_unless(&rds_iw_allocation, 1, rds_iw_sysctl_max_recv_allocation)) {
-			rds_iw_stats_inc(s_iw_rx_alloc_limit);
-			goto out;
-		}
-		recv->r_iwinc = kmem_cache_alloc(rds_iw_incoming_slab,
-						 kptr_gfp);
-		if (!recv->r_iwinc) {
-			atomic_dec(&rds_iw_allocation);
-			goto out;
-		}
-		INIT_LIST_HEAD(&recv->r_iwinc->ii_frags);
-		rds_inc_init(&recv->r_iwinc->ii_inc, conn, conn->c_faddr);
-	}
-
-	if (!recv->r_frag) {
-		recv->r_frag = kmem_cache_alloc(rds_iw_frag_slab, kptr_gfp);
-		if (!recv->r_frag)
-			goto out;
-		INIT_LIST_HEAD(&recv->r_frag->f_item);
-		recv->r_frag->f_page = NULL;
-	}
-
-	if (!ic->i_frag.f_page) {
-		ic->i_frag.f_page = alloc_page(page_gfp);
-		if (!ic->i_frag.f_page)
-			goto out;
-		ic->i_frag.f_offset = 0;
-	}
-
-	dma_addr = ib_dma_map_page(ic->i_cm_id->device,
-				  ic->i_frag.f_page,
-				  ic->i_frag.f_offset,
-				  RDS_FRAG_SIZE,
-				  DMA_FROM_DEVICE);
-	if (ib_dma_mapping_error(ic->i_cm_id->device, dma_addr))
-		goto out;
-
-	/*
-	 * Once we get the RDS_PAGE_LAST_OFF frag then rds_iw_frag_unmap()
-	 * must be called on this recv.  This happens as completions hit
-	 * in order or on connection shutdown.
-	 */
-	recv->r_frag->f_page = ic->i_frag.f_page;
-	recv->r_frag->f_offset = ic->i_frag.f_offset;
-	recv->r_frag->f_mapped = dma_addr;
-
-	sge = rds_iw_data_sge(ic, recv->r_sge);
-	sge->addr = dma_addr;
-	sge->length = RDS_FRAG_SIZE;
-
-	sge = rds_iw_header_sge(ic, recv->r_sge);
-	sge->addr = ic->i_recv_hdrs_dma + (recv - ic->i_recvs) * sizeof(struct rds_header);
-	sge->length = sizeof(struct rds_header);
-
-	get_page(recv->r_frag->f_page);
-
-	if (ic->i_frag.f_offset < RDS_PAGE_LAST_OFF) {
-		ic->i_frag.f_offset += RDS_FRAG_SIZE;
-	} else {
-		put_page(ic->i_frag.f_page);
-		ic->i_frag.f_page = NULL;
-		ic->i_frag.f_offset = 0;
-	}
-
-	ret = 0;
-out:
-	return ret;
-}
-
-/*
- * This tries to allocate and post unused work requests after making sure that
- * they have all the allocations they need to queue received fragments into
- * sockets.  The i_recv_mutex is held here so that ring_alloc and _unalloc
- * pairs don't go unmatched.
- *
- * -1 is returned if posting fails due to temporary resource exhaustion.
- */
-int rds_iw_recv_refill(struct rds_connection *conn, gfp_t kptr_gfp,
-		       gfp_t page_gfp, int prefill)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct rds_iw_recv_work *recv;
-	struct ib_recv_wr *failed_wr;
-	unsigned int posted = 0;
-	int ret = 0;
-	u32 pos;
-
-	while ((prefill || rds_conn_up(conn)) &&
-	       rds_iw_ring_alloc(&ic->i_recv_ring, 1, &pos)) {
-		if (pos >= ic->i_recv_ring.w_nr) {
-			printk(KERN_NOTICE "Argh - ring alloc returned pos=%u\n",
-					pos);
-			ret = -EINVAL;
-			break;
-		}
-
-		recv = &ic->i_recvs[pos];
-		ret = rds_iw_recv_refill_one(conn, recv, kptr_gfp, page_gfp);
-		if (ret) {
-			ret = -1;
-			break;
-		}
-
-		/* XXX when can this fail? */
-		ret = ib_post_recv(ic->i_cm_id->qp, &recv->r_wr, &failed_wr);
-		rdsdebug("recv %p iwinc %p page %p addr %lu ret %d\n", recv,
-			 recv->r_iwinc, recv->r_frag->f_page,
-			 (long) recv->r_frag->f_mapped, ret);
-		if (ret) {
-			rds_iw_conn_error(conn, "recv post on "
-			       "%pI4 returned %d, disconnecting and "
-			       "reconnecting\n", &conn->c_faddr,
-			       ret);
-			ret = -1;
-			break;
-		}
-
-		posted++;
-	}
-
-	/* We're doing flow control - update the window. */
-	if (ic->i_flowctl && posted)
-		rds_iw_advertise_credits(conn, posted);
-
-	if (ret)
-		rds_iw_ring_unalloc(&ic->i_recv_ring, 1);
-	return ret;
-}
-
-static void rds_iw_inc_purge(struct rds_incoming *inc)
-{
-	struct rds_iw_incoming *iwinc;
-	struct rds_page_frag *frag;
-	struct rds_page_frag *pos;
-
-	iwinc = container_of(inc, struct rds_iw_incoming, ii_inc);
-	rdsdebug("purging iwinc %p inc %p\n", iwinc, inc);
-
-	list_for_each_entry_safe(frag, pos, &iwinc->ii_frags, f_item) {
-		list_del_init(&frag->f_item);
-		rds_iw_frag_drop_page(frag);
-		rds_iw_frag_free(frag);
-	}
-}
-
-void rds_iw_inc_free(struct rds_incoming *inc)
-{
-	struct rds_iw_incoming *iwinc;
-
-	iwinc = container_of(inc, struct rds_iw_incoming, ii_inc);
-
-	rds_iw_inc_purge(inc);
-	rdsdebug("freeing iwinc %p inc %p\n", iwinc, inc);
-	BUG_ON(!list_empty(&iwinc->ii_frags));
-	kmem_cache_free(rds_iw_incoming_slab, iwinc);
-	atomic_dec(&rds_iw_allocation);
-	BUG_ON(atomic_read(&rds_iw_allocation) < 0);
-}
-
-int rds_iw_inc_copy_to_user(struct rds_incoming *inc, struct iov_iter *to)
-{
-	struct rds_iw_incoming *iwinc;
-	struct rds_page_frag *frag;
-	unsigned long to_copy;
-	unsigned long frag_off = 0;
-	int copied = 0;
-	int ret;
-	u32 len;
-
-	iwinc = container_of(inc, struct rds_iw_incoming, ii_inc);
-	frag = list_entry(iwinc->ii_frags.next, struct rds_page_frag, f_item);
-	len = be32_to_cpu(inc->i_hdr.h_len);
-
-	while (iov_iter_count(to) && copied < len) {
-		if (frag_off == RDS_FRAG_SIZE) {
-			frag = list_entry(frag->f_item.next,
-					  struct rds_page_frag, f_item);
-			frag_off = 0;
-		}
-		to_copy = min_t(unsigned long, iov_iter_count(to),
-				RDS_FRAG_SIZE - frag_off);
-		to_copy = min_t(unsigned long, to_copy, len - copied);
-
-		/* XXX needs + offset for multiple recvs per page */
-		rds_stats_add(s_copy_to_user, to_copy);
-		ret = copy_page_to_iter(frag->f_page,
-					frag->f_offset + frag_off,
-					to_copy,
-					to);
-		if (ret != to_copy)
-			return -EFAULT;
-
-		frag_off += to_copy;
-		copied += to_copy;
-	}
-
-	return copied;
-}
-
-/* ic starts out kzalloc()ed */
-void rds_iw_recv_init_ack(struct rds_iw_connection *ic)
-{
-	struct ib_send_wr *wr = &ic->i_ack_wr;
-	struct ib_sge *sge = &ic->i_ack_sge;
-
-	sge->addr = ic->i_ack_dma;
-	sge->length = sizeof(struct rds_header);
-	sge->lkey = rds_iw_local_dma_lkey(ic);
-
-	wr->sg_list = sge;
-	wr->num_sge = 1;
-	wr->opcode = IB_WR_SEND;
-	wr->wr_id = RDS_IW_ACK_WR_ID;
-	wr->send_flags = IB_SEND_SIGNALED | IB_SEND_SOLICITED;
-}
-
-/*
- * You'd think that with reliable IB connections you wouldn't need to ack
- * messages that have been received.  The problem is that IB hardware generates
- * an ack message before it has DMAed the message into memory.  This creates a
- * potential message loss if the HCA is disabled for any reason between when it
- * sends the ack and before the message is DMAed and processed.  This is only a
- * potential issue if another HCA is available for fail-over.
- *
- * When the remote host receives our ack they'll free the sent message from
- * their send queue.  To decrease the latency of this we always send an ack
- * immediately after we've received messages.
- *
- * For simplicity, we only have one ack in flight at a time.  This puts
- * pressure on senders to have deep enough send queues to absorb the latency of
- * a single ack frame being in flight.  This might not be good enough.
- *
- * This is implemented by have a long-lived send_wr and sge which point to a
- * statically allocated ack frame.  This ack wr does not fall under the ring
- * accounting that the tx and rx wrs do.  The QP attribute specifically makes
- * room for it beyond the ring size.  Send completion notices its special
- * wr_id and avoids working with the ring in that case.
- */
-#ifndef KERNEL_HAS_ATOMIC64
-static void rds_iw_set_ack(struct rds_iw_connection *ic, u64 seq,
-				int ack_required)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&ic->i_ack_lock, flags);
-	ic->i_ack_next = seq;
-	if (ack_required)
-		set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-	spin_unlock_irqrestore(&ic->i_ack_lock, flags);
-}
-
-static u64 rds_iw_get_ack(struct rds_iw_connection *ic)
-{
-	unsigned long flags;
-	u64 seq;
-
-	clear_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-
-	spin_lock_irqsave(&ic->i_ack_lock, flags);
-	seq = ic->i_ack_next;
-	spin_unlock_irqrestore(&ic->i_ack_lock, flags);
-
-	return seq;
-}
-#else
-static void rds_iw_set_ack(struct rds_iw_connection *ic, u64 seq,
-				int ack_required)
-{
-	atomic64_set(&ic->i_ack_next, seq);
-	if (ack_required) {
-		smp_mb__before_atomic();
-		set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-	}
-}
-
-static u64 rds_iw_get_ack(struct rds_iw_connection *ic)
-{
-	clear_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-	smp_mb__after_atomic();
-
-	return atomic64_read(&ic->i_ack_next);
-}
-#endif
-
-
-static void rds_iw_send_ack(struct rds_iw_connection *ic, unsigned int adv_credits)
-{
-	struct rds_header *hdr = ic->i_ack;
-	struct ib_send_wr *failed_wr;
-	u64 seq;
-	int ret;
-
-	seq = rds_iw_get_ack(ic);
-
-	rdsdebug("send_ack: ic %p ack %llu\n", ic, (unsigned long long) seq);
-	rds_message_populate_header(hdr, 0, 0, 0);
-	hdr->h_ack = cpu_to_be64(seq);
-	hdr->h_credit = adv_credits;
-	rds_message_make_checksum(hdr);
-	ic->i_ack_queued = jiffies;
-
-	ret = ib_post_send(ic->i_cm_id->qp, &ic->i_ack_wr, &failed_wr);
-	if (unlikely(ret)) {
-		/* Failed to send. Release the WR, and
-		 * force another ACK.
-		 */
-		clear_bit(IB_ACK_IN_FLIGHT, &ic->i_ack_flags);
-		set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-
-		rds_iw_stats_inc(s_iw_ack_send_failure);
-
-		rds_iw_conn_error(ic->conn, "sending ack failed\n");
-	} else
-		rds_iw_stats_inc(s_iw_ack_sent);
-}
-
-/*
- * There are 3 ways of getting acknowledgements to the peer:
- *  1.	We call rds_iw_attempt_ack from the recv completion handler
- *	to send an ACK-only frame.
- *	However, there can be only one such frame in the send queue
- *	at any time, so we may have to postpone it.
- *  2.	When another (data) packet is transmitted while there's
- *	an ACK in the queue, we piggyback the ACK sequence number
- *	on the data packet.
- *  3.	If the ACK WR is done sending, we get called from the
- *	send queue completion handler, and check whether there's
- *	another ACK pending (postponed because the WR was on the
- *	queue). If so, we transmit it.
- *
- * We maintain 2 variables:
- *  -	i_ack_flags, which keeps track of whether the ACK WR
- *	is currently in the send queue or not (IB_ACK_IN_FLIGHT)
- *  -	i_ack_next, which is the last sequence number we received
- *
- * Potentially, send queue and receive queue handlers can run concurrently.
- * It would be nice to not have to use a spinlock to synchronize things,
- * but the one problem that rules this out is that 64bit updates are
- * not atomic on all platforms. Things would be a lot simpler if
- * we had atomic64 or maybe cmpxchg64 everywhere.
- *
- * Reconnecting complicates this picture just slightly. When we
- * reconnect, we may be seeing duplicate packets. The peer
- * is retransmitting them, because it hasn't seen an ACK for
- * them. It is important that we ACK these.
- *
- * ACK mitigation adds a header flag "ACK_REQUIRED"; any packet with
- * this flag set *MUST* be acknowledged immediately.
- */
-
-/*
- * When we get here, we're called from the recv queue handler.
- * Check whether we ought to transmit an ACK.
- */
-void rds_iw_attempt_ack(struct rds_iw_connection *ic)
-{
-	unsigned int adv_credits;
-
-	if (!test_bit(IB_ACK_REQUESTED, &ic->i_ack_flags))
-		return;
-
-	if (test_and_set_bit(IB_ACK_IN_FLIGHT, &ic->i_ack_flags)) {
-		rds_iw_stats_inc(s_iw_ack_send_delayed);
-		return;
-	}
-
-	/* Can we get a send credit? */
-	if (!rds_iw_send_grab_credits(ic, 1, &adv_credits, 0, RDS_MAX_ADV_CREDIT)) {
-		rds_iw_stats_inc(s_iw_tx_throttle);
-		clear_bit(IB_ACK_IN_FLIGHT, &ic->i_ack_flags);
-		return;
-	}
-
-	clear_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-	rds_iw_send_ack(ic, adv_credits);
-}
-
-/*
- * We get here from the send completion handler, when the
- * adapter tells us the ACK frame was sent.
- */
-void rds_iw_ack_send_complete(struct rds_iw_connection *ic)
-{
-	clear_bit(IB_ACK_IN_FLIGHT, &ic->i_ack_flags);
-	rds_iw_attempt_ack(ic);
-}
-
-/*
- * This is called by the regular xmit code when it wants to piggyback
- * an ACK on an outgoing frame.
- */
-u64 rds_iw_piggyb_ack(struct rds_iw_connection *ic)
-{
-	if (test_and_clear_bit(IB_ACK_REQUESTED, &ic->i_ack_flags))
-		rds_iw_stats_inc(s_iw_ack_send_piggybacked);
-	return rds_iw_get_ack(ic);
-}
-
-/*
- * It's kind of lame that we're copying from the posted receive pages into
- * long-lived bitmaps.  We could have posted the bitmaps and rdma written into
- * them.  But receiving new congestion bitmaps should be a *rare* event, so
- * hopefully we won't need to invest that complexity in making it more
- * efficient.  By copying we can share a simpler core with TCP which has to
- * copy.
- */
-static void rds_iw_cong_recv(struct rds_connection *conn,
-			      struct rds_iw_incoming *iwinc)
-{
-	struct rds_cong_map *map;
-	unsigned int map_off;
-	unsigned int map_page;
-	struct rds_page_frag *frag;
-	unsigned long frag_off;
-	unsigned long to_copy;
-	unsigned long copied;
-	uint64_t uncongested = 0;
-	void *addr;
-
-	/* catch completely corrupt packets */
-	if (be32_to_cpu(iwinc->ii_inc.i_hdr.h_len) != RDS_CONG_MAP_BYTES)
-		return;
-
-	map = conn->c_fcong;
-	map_page = 0;
-	map_off = 0;
-
-	frag = list_entry(iwinc->ii_frags.next, struct rds_page_frag, f_item);
-	frag_off = 0;
-
-	copied = 0;
-
-	while (copied < RDS_CONG_MAP_BYTES) {
-		uint64_t *src, *dst;
-		unsigned int k;
-
-		to_copy = min(RDS_FRAG_SIZE - frag_off, PAGE_SIZE - map_off);
-		BUG_ON(to_copy & 7); /* Must be 64bit aligned. */
-
-		addr = kmap_atomic(frag->f_page);
-
-		src = addr + frag_off;
-		dst = (void *)map->m_page_addrs[map_page] + map_off;
-		for (k = 0; k < to_copy; k += 8) {
-			/* Record ports that became uncongested, ie
-			 * bits that changed from 0 to 1. */
-			uncongested |= ~(*src) & *dst;
-			*dst++ = *src++;
-		}
-		kunmap_atomic(addr);
-
-		copied += to_copy;
-
-		map_off += to_copy;
-		if (map_off == PAGE_SIZE) {
-			map_off = 0;
-			map_page++;
-		}
-
-		frag_off += to_copy;
-		if (frag_off == RDS_FRAG_SIZE) {
-			frag = list_entry(frag->f_item.next,
-					  struct rds_page_frag, f_item);
-			frag_off = 0;
-		}
-	}
-
-	/* the congestion map is in little endian order */
-	uncongested = le64_to_cpu(uncongested);
-
-	rds_cong_map_updated(map, uncongested);
-}
-
-/*
- * Rings are posted with all the allocations they'll need to queue the
- * incoming message to the receiving socket so this can't fail.
- * All fragments start with a header, so we can make sure we're not receiving
- * garbage, and we can tell a small 8 byte fragment from an ACK frame.
- */
-struct rds_iw_ack_state {
-	u64		ack_next;
-	u64		ack_recv;
-	unsigned int	ack_required:1;
-	unsigned int	ack_next_valid:1;
-	unsigned int	ack_recv_valid:1;
-};
-
-static void rds_iw_process_recv(struct rds_connection *conn,
-				struct rds_iw_recv_work *recv, u32 byte_len,
-				struct rds_iw_ack_state *state)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct rds_iw_incoming *iwinc = ic->i_iwinc;
-	struct rds_header *ihdr, *hdr;
-
-	/* XXX shut down the connection if port 0,0 are seen? */
-
-	rdsdebug("ic %p iwinc %p recv %p byte len %u\n", ic, iwinc, recv,
-		 byte_len);
-
-	if (byte_len < sizeof(struct rds_header)) {
-		rds_iw_conn_error(conn, "incoming message "
-		       "from %pI4 didn't include a "
-		       "header, disconnecting and "
-		       "reconnecting\n",
-		       &conn->c_faddr);
-		return;
-	}
-	byte_len -= sizeof(struct rds_header);
-
-	ihdr = &ic->i_recv_hdrs[recv - ic->i_recvs];
-
-	/* Validate the checksum. */
-	if (!rds_message_verify_checksum(ihdr)) {
-		rds_iw_conn_error(conn, "incoming message "
-		       "from %pI4 has corrupted header - "
-		       "forcing a reconnect\n",
-		       &conn->c_faddr);
-		rds_stats_inc(s_recv_drop_bad_checksum);
-		return;
-	}
-
-	/* Process the ACK sequence which comes with every packet */
-	state->ack_recv = be64_to_cpu(ihdr->h_ack);
-	state->ack_recv_valid = 1;
-
-	/* Process the credits update if there was one */
-	if (ihdr->h_credit)
-		rds_iw_send_add_credits(conn, ihdr->h_credit);
-
-	if (ihdr->h_sport == 0 && ihdr->h_dport == 0 && byte_len == 0) {
-		/* This is an ACK-only packet. The fact that it gets
-		 * special treatment here is that historically, ACKs
-		 * were rather special beasts.
-		 */
-		rds_iw_stats_inc(s_iw_ack_received);
-
-		/*
-		 * Usually the frags make their way on to incs and are then freed as
-		 * the inc is freed.  We don't go that route, so we have to drop the
-		 * page ref ourselves.  We can't just leave the page on the recv
-		 * because that confuses the dma mapping of pages and each recv's use
-		 * of a partial page.  We can leave the frag, though, it will be
-		 * reused.
-		 *
-		 * FIXME: Fold this into the code path below.
-		 */
-		rds_iw_frag_drop_page(recv->r_frag);
-		return;
-	}
-
-	/*
-	 * If we don't already have an inc on the connection then this
-	 * fragment has a header and starts a message.. copy its header
-	 * into the inc and save the inc so we can hang upcoming fragments
-	 * off its list.
-	 */
-	if (!iwinc) {
-		iwinc = recv->r_iwinc;
-		recv->r_iwinc = NULL;
-		ic->i_iwinc = iwinc;
-
-		hdr = &iwinc->ii_inc.i_hdr;
-		memcpy(hdr, ihdr, sizeof(*hdr));
-		ic->i_recv_data_rem = be32_to_cpu(hdr->h_len);
-
-		rdsdebug("ic %p iwinc %p rem %u flag 0x%x\n", ic, iwinc,
-			 ic->i_recv_data_rem, hdr->h_flags);
-	} else {
-		hdr = &iwinc->ii_inc.i_hdr;
-		/* We can't just use memcmp here; fragments of a
-		 * single message may carry different ACKs */
-		if (hdr->h_sequence != ihdr->h_sequence ||
-		    hdr->h_len != ihdr->h_len ||
-		    hdr->h_sport != ihdr->h_sport ||
-		    hdr->h_dport != ihdr->h_dport) {
-			rds_iw_conn_error(conn,
-				"fragment header mismatch; forcing reconnect\n");
-			return;
-		}
-	}
-
-	list_add_tail(&recv->r_frag->f_item, &iwinc->ii_frags);
-	recv->r_frag = NULL;
-
-	if (ic->i_recv_data_rem > RDS_FRAG_SIZE)
-		ic->i_recv_data_rem -= RDS_FRAG_SIZE;
-	else {
-		ic->i_recv_data_rem = 0;
-		ic->i_iwinc = NULL;
-
-		if (iwinc->ii_inc.i_hdr.h_flags == RDS_FLAG_CONG_BITMAP)
-			rds_iw_cong_recv(conn, iwinc);
-		else {
-			rds_recv_incoming(conn, conn->c_faddr, conn->c_laddr,
-					  &iwinc->ii_inc, GFP_ATOMIC);
-			state->ack_next = be64_to_cpu(hdr->h_sequence);
-			state->ack_next_valid = 1;
-		}
-
-		/* Evaluate the ACK_REQUIRED flag *after* we received
-		 * the complete frame, and after bumping the next_rx
-		 * sequence. */
-		if (hdr->h_flags & RDS_FLAG_ACK_REQUIRED) {
-			rds_stats_inc(s_recv_ack_required);
-			state->ack_required = 1;
-		}
-
-		rds_inc_put(&iwinc->ii_inc);
-	}
-}
-
-/*
- * Plucking the oldest entry from the ring can be done concurrently with
- * the thread refilling the ring.  Each ring operation is protected by
- * spinlocks and the transient state of refilling doesn't change the
- * recording of which entry is oldest.
- *
- * This relies on IB only calling one cq comp_handler for each cq so that
- * there will only be one caller of rds_recv_incoming() per RDS connection.
- */
-void rds_iw_recv_cq_comp_handler(struct ib_cq *cq, void *context)
-{
-	struct rds_connection *conn = context;
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	rdsdebug("conn %p cq %p\n", conn, cq);
-
-	rds_iw_stats_inc(s_iw_rx_cq_call);
-
-	tasklet_schedule(&ic->i_recv_tasklet);
-}
-
-static inline void rds_poll_cq(struct rds_iw_connection *ic,
-			       struct rds_iw_ack_state *state)
-{
-	struct rds_connection *conn = ic->conn;
-	struct ib_wc wc;
-	struct rds_iw_recv_work *recv;
-
-	while (ib_poll_cq(ic->i_recv_cq, 1, &wc) > 0) {
-		rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n",
-			 (unsigned long long)wc.wr_id, wc.status, wc.byte_len,
-			 be32_to_cpu(wc.ex.imm_data));
-		rds_iw_stats_inc(s_iw_rx_cq_event);
-
-		recv = &ic->i_recvs[rds_iw_ring_oldest(&ic->i_recv_ring)];
-
-		rds_iw_recv_unmap_page(ic, recv);
-
-		/*
-		 * Also process recvs in connecting state because it is possible
-		 * to get a recv completion _before_ the rdmacm ESTABLISHED
-		 * event is processed.
-		 */
-		if (rds_conn_up(conn) || rds_conn_connecting(conn)) {
-			/* We expect errors as the qp is drained during shutdown */
-			if (wc.status == IB_WC_SUCCESS) {
-				rds_iw_process_recv(conn, recv, wc.byte_len, state);
-			} else {
-				rds_iw_conn_error(conn, "recv completion on "
-				       "%pI4 had status %u, disconnecting and "
-				       "reconnecting\n", &conn->c_faddr,
-				       wc.status);
-			}
-		}
-
-		rds_iw_ring_free(&ic->i_recv_ring, 1);
-	}
-}
-
-void rds_iw_recv_tasklet_fn(unsigned long data)
-{
-	struct rds_iw_connection *ic = (struct rds_iw_connection *) data;
-	struct rds_connection *conn = ic->conn;
-	struct rds_iw_ack_state state = { 0, };
-
-	rds_poll_cq(ic, &state);
-	ib_req_notify_cq(ic->i_recv_cq, IB_CQ_SOLICITED);
-	rds_poll_cq(ic, &state);
-
-	if (state.ack_next_valid)
-		rds_iw_set_ack(ic, state.ack_next, state.ack_required);
-	if (state.ack_recv_valid && state.ack_recv > ic->i_ack_recv) {
-		rds_send_drop_acked(conn, state.ack_recv, NULL);
-		ic->i_ack_recv = state.ack_recv;
-	}
-	if (rds_conn_up(conn))
-		rds_iw_attempt_ack(ic);
-
-	/* If we ever end up with a really empty receive ring, we're
-	 * in deep trouble, as the sender will definitely see RNR
-	 * timeouts. */
-	if (rds_iw_ring_empty(&ic->i_recv_ring))
-		rds_iw_stats_inc(s_iw_rx_ring_empty);
-
-	/*
-	 * If the ring is running low, then schedule the thread to refill.
-	 */
-	if (rds_iw_ring_low(&ic->i_recv_ring))
-		queue_delayed_work(rds_wq, &conn->c_recv_w, 0);
-}
-
-int rds_iw_recv(struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	int ret = 0;
-
-	rdsdebug("conn %p\n", conn);
-
-	/*
-	 * If we get a temporary posting failure in this context then
-	 * we're really low and we want the caller to back off for a bit.
-	 */
-	mutex_lock(&ic->i_recv_mutex);
-	if (rds_iw_recv_refill(conn, GFP_KERNEL, GFP_HIGHUSER, 0))
-		ret = -ENOMEM;
-	else
-		rds_iw_stats_inc(s_iw_rx_refill_from_thread);
-	mutex_unlock(&ic->i_recv_mutex);
-
-	if (rds_conn_up(conn))
-		rds_iw_attempt_ack(ic);
-
-	return ret;
-}
-
-int rds_iw_recv_init(void)
-{
-	struct sysinfo si;
-	int ret = -ENOMEM;
-
-	/* Default to 30% of all available RAM for recv memory */
-	si_meminfo(&si);
-	rds_iw_sysctl_max_recv_allocation = si.totalram / 3 * PAGE_SIZE / RDS_FRAG_SIZE;
-
-	rds_iw_incoming_slab = kmem_cache_create("rds_iw_incoming",
-					sizeof(struct rds_iw_incoming),
-					0, 0, NULL);
-	if (!rds_iw_incoming_slab)
-		goto out;
-
-	rds_iw_frag_slab = kmem_cache_create("rds_iw_frag",
-					sizeof(struct rds_page_frag),
-					0, 0, NULL);
-	if (!rds_iw_frag_slab)
-		kmem_cache_destroy(rds_iw_incoming_slab);
-	else
-		ret = 0;
-out:
-	return ret;
-}
-
-void rds_iw_recv_exit(void)
-{
-	kmem_cache_destroy(rds_iw_incoming_slab);
-	kmem_cache_destroy(rds_iw_frag_slab);
-}
diff --git a/net/rds/iw_ring.c b/net/rds/iw_ring.c
deleted file mode 100644
index da8e3b6..0000000
--- a/net/rds/iw_ring.c
+++ /dev/null
@@ -1,169 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-
-#include "rds.h"
-#include "iw.h"
-
-/*
- * Locking for IB rings.
- * We assume that allocation is always protected by a mutex
- * in the caller (this is a valid assumption for the current
- * implementation).
- *
- * Freeing always happens in an interrupt, and hence only
- * races with allocations, but not with other free()s.
- *
- * The interaction between allocation and freeing is that
- * the alloc code has to determine the number of free entries.
- * To this end, we maintain two counters; an allocation counter
- * and a free counter. Both are allowed to run freely, and wrap
- * around.
- * The number of used entries is always (alloc_ctr - free_ctr) % NR.
- *
- * The current implementation makes free_ctr atomic. When the
- * caller finds an allocation fails, it should set an "alloc fail"
- * bit and retry the allocation. The "alloc fail" bit essentially tells
- * the CQ completion handlers to wake it up after freeing some
- * more entries.
- */
-
-/*
- * This only happens on shutdown.
- */
-DECLARE_WAIT_QUEUE_HEAD(rds_iw_ring_empty_wait);
-
-void rds_iw_ring_init(struct rds_iw_work_ring *ring, u32 nr)
-{
-	memset(ring, 0, sizeof(*ring));
-	ring->w_nr = nr;
-	rdsdebug("ring %p nr %u\n", ring, ring->w_nr);
-}
-
-static inline u32 __rds_iw_ring_used(struct rds_iw_work_ring *ring)
-{
-	u32 diff;
-
-	/* This assumes that atomic_t has at least as many bits as u32 */
-	diff = ring->w_alloc_ctr - (u32) atomic_read(&ring->w_free_ctr);
-	BUG_ON(diff > ring->w_nr);
-
-	return diff;
-}
-
-void rds_iw_ring_resize(struct rds_iw_work_ring *ring, u32 nr)
-{
-	/* We only ever get called from the connection setup code,
-	 * prior to creating the QP. */
-	BUG_ON(__rds_iw_ring_used(ring));
-	ring->w_nr = nr;
-}
-
-static int __rds_iw_ring_empty(struct rds_iw_work_ring *ring)
-{
-	return __rds_iw_ring_used(ring) == 0;
-}
-
-u32 rds_iw_ring_alloc(struct rds_iw_work_ring *ring, u32 val, u32 *pos)
-{
-	u32 ret = 0, avail;
-
-	avail = ring->w_nr - __rds_iw_ring_used(ring);
-
-	rdsdebug("ring %p val %u next %u free %u\n", ring, val,
-		 ring->w_alloc_ptr, avail);
-
-	if (val && avail) {
-		ret = min(val, avail);
-		*pos = ring->w_alloc_ptr;
-
-		ring->w_alloc_ptr = (ring->w_alloc_ptr + ret) % ring->w_nr;
-		ring->w_alloc_ctr += ret;
-	}
-
-	return ret;
-}
-
-void rds_iw_ring_free(struct rds_iw_work_ring *ring, u32 val)
-{
-	ring->w_free_ptr = (ring->w_free_ptr + val) % ring->w_nr;
-	atomic_add(val, &ring->w_free_ctr);
-
-	if (__rds_iw_ring_empty(ring) &&
-	    waitqueue_active(&rds_iw_ring_empty_wait))
-		wake_up(&rds_iw_ring_empty_wait);
-}
-
-void rds_iw_ring_unalloc(struct rds_iw_work_ring *ring, u32 val)
-{
-	ring->w_alloc_ptr = (ring->w_alloc_ptr - val) % ring->w_nr;
-	ring->w_alloc_ctr -= val;
-}
-
-int rds_iw_ring_empty(struct rds_iw_work_ring *ring)
-{
-	return __rds_iw_ring_empty(ring);
-}
-
-int rds_iw_ring_low(struct rds_iw_work_ring *ring)
-{
-	return __rds_iw_ring_used(ring) <= (ring->w_nr >> 1);
-}
-
-
-/*
- * returns the oldest alloced ring entry.  This will be the next one
- * freed.  This can't be called if there are none allocated.
- */
-u32 rds_iw_ring_oldest(struct rds_iw_work_ring *ring)
-{
-	return ring->w_free_ptr;
-}
-
-/*
- * returns the number of completed work requests.
- */
-
-u32 rds_iw_ring_completed(struct rds_iw_work_ring *ring, u32 wr_id, u32 oldest)
-{
-	u32 ret;
-
-	if (oldest <= (unsigned long long)wr_id)
-		ret = (unsigned long long)wr_id - oldest + 1;
-	else
-		ret = ring->w_nr - oldest + (unsigned long long)wr_id + 1;
-
-	rdsdebug("ring %p ret %u wr_id %u oldest %u\n", ring, ret,
-		 wr_id, oldest);
-	return ret;
-}
diff --git a/net/rds/iw_send.c b/net/rds/iw_send.c
deleted file mode 100644
index e20bd50..0000000
--- a/net/rds/iw_send.c
+++ /dev/null
@@ -1,981 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-#include <linux/in.h>
-#include <linux/device.h>
-#include <linux/dmapool.h>
-#include <linux/ratelimit.h>
-
-#include "rds.h"
-#include "iw.h"
-
-static void rds_iw_send_rdma_complete(struct rds_message *rm,
-				      int wc_status)
-{
-	int notify_status;
-
-	switch (wc_status) {
-	case IB_WC_WR_FLUSH_ERR:
-		return;
-
-	case IB_WC_SUCCESS:
-		notify_status = RDS_RDMA_SUCCESS;
-		break;
-
-	case IB_WC_REM_ACCESS_ERR:
-		notify_status = RDS_RDMA_REMOTE_ERROR;
-		break;
-
-	default:
-		notify_status = RDS_RDMA_OTHER_ERROR;
-		break;
-	}
-	rds_rdma_send_complete(rm, notify_status);
-}
-
-static void rds_iw_send_unmap_rdma(struct rds_iw_connection *ic,
-				   struct rm_rdma_op *op)
-{
-	if (op->op_mapped) {
-		ib_dma_unmap_sg(ic->i_cm_id->device,
-			op->op_sg, op->op_nents,
-			op->op_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
-		op->op_mapped = 0;
-	}
-}
-
-static void rds_iw_send_unmap_rm(struct rds_iw_connection *ic,
-			  struct rds_iw_send_work *send,
-			  int wc_status)
-{
-	struct rds_message *rm = send->s_rm;
-
-	rdsdebug("ic %p send %p rm %p\n", ic, send, rm);
-
-	ib_dma_unmap_sg(ic->i_cm_id->device,
-		     rm->data.op_sg, rm->data.op_nents,
-		     DMA_TO_DEVICE);
-
-	if (rm->rdma.op_active) {
-		rds_iw_send_unmap_rdma(ic, &rm->rdma);
-
-		/* If the user asked for a completion notification on this
-		 * message, we can implement three different semantics:
-		 *  1.	Notify when we received the ACK on the RDS message
-		 *	that was queued with the RDMA. This provides reliable
-		 *	notification of RDMA status at the expense of a one-way
-		 *	packet delay.
-		 *  2.	Notify when the IB stack gives us the completion event for
-		 *	the RDMA operation.
-		 *  3.	Notify when the IB stack gives us the completion event for
-		 *	the accompanying RDS messages.
-		 * Here, we implement approach #3. To implement approach #2,
-		 * call rds_rdma_send_complete from the cq_handler. To implement #1,
-		 * don't call rds_rdma_send_complete at all, and fall back to the notify
-		 * handling in the ACK processing code.
-		 *
-		 * Note: There's no need to explicitly sync any RDMA buffers using
-		 * ib_dma_sync_sg_for_cpu - the completion for the RDMA
-		 * operation itself unmapped the RDMA buffers, which takes care
-		 * of synching.
-		 */
-		rds_iw_send_rdma_complete(rm, wc_status);
-
-		if (rm->rdma.op_write)
-			rds_stats_add(s_send_rdma_bytes, rm->rdma.op_bytes);
-		else
-			rds_stats_add(s_recv_rdma_bytes, rm->rdma.op_bytes);
-	}
-
-	/* If anyone waited for this message to get flushed out, wake
-	 * them up now */
-	rds_message_unmapped(rm);
-
-	rds_message_put(rm);
-	send->s_rm = NULL;
-}
-
-void rds_iw_send_init_ring(struct rds_iw_connection *ic)
-{
-	struct rds_iw_send_work *send;
-	u32 i;
-
-	for (i = 0, send = ic->i_sends; i < ic->i_send_ring.w_nr; i++, send++) {
-		struct ib_sge *sge;
-
-		send->s_rm = NULL;
-		send->s_op = NULL;
-		send->s_mapping = NULL;
-
-		send->s_send_wr.next = NULL;
-		send->s_send_wr.wr_id = i;
-		send->s_send_wr.sg_list = send->s_sge;
-		send->s_send_wr.num_sge = 1;
-		send->s_send_wr.opcode = IB_WR_SEND;
-		send->s_send_wr.send_flags = 0;
-		send->s_send_wr.ex.imm_data = 0;
-
-		sge = rds_iw_data_sge(ic, send->s_sge);
-		sge->lkey = 0;
-
-		sge = rds_iw_header_sge(ic, send->s_sge);
-		sge->addr = ic->i_send_hdrs_dma + (i * sizeof(struct rds_header));
-		sge->length = sizeof(struct rds_header);
-		sge->lkey = 0;
-
-		send->s_mr = ib_alloc_mr(ic->i_pd, IB_MR_TYPE_MEM_REG,
-					 fastreg_message_size);
-		if (IS_ERR(send->s_mr)) {
-			printk(KERN_WARNING "RDS/IW: ib_alloc_mr failed\n");
-			break;
-		}
-	}
-}
-
-void rds_iw_send_clear_ring(struct rds_iw_connection *ic)
-{
-	struct rds_iw_send_work *send;
-	u32 i;
-
-	for (i = 0, send = ic->i_sends; i < ic->i_send_ring.w_nr; i++, send++) {
-		BUG_ON(!send->s_mr);
-		ib_dereg_mr(send->s_mr);
-		if (send->s_send_wr.opcode == 0xdead)
-			continue;
-		if (send->s_rm)
-			rds_iw_send_unmap_rm(ic, send, IB_WC_WR_FLUSH_ERR);
-		if (send->s_op)
-			rds_iw_send_unmap_rdma(ic, send->s_op);
-	}
-}
-
-/*
- * The _oldest/_free ring operations here race cleanly with the alloc/unalloc
- * operations performed in the send path.  As the sender allocs and potentially
- * unallocs the next free entry in the ring it doesn't alter which is
- * the next to be freed, which is what this is concerned with.
- */
-void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context)
-{
-	struct rds_connection *conn = context;
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct ib_wc wc;
-	struct rds_iw_send_work *send;
-	u32 completed;
-	u32 oldest;
-	u32 i;
-	int ret;
-
-	rdsdebug("cq %p conn %p\n", cq, conn);
-	rds_iw_stats_inc(s_iw_tx_cq_call);
-	ret = ib_req_notify_cq(cq, IB_CQ_NEXT_COMP);
-	if (ret)
-		rdsdebug("ib_req_notify_cq send failed: %d\n", ret);
-
-	while (ib_poll_cq(cq, 1, &wc) > 0) {
-		rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n",
-			 (unsigned long long)wc.wr_id, wc.status, wc.byte_len,
-			 be32_to_cpu(wc.ex.imm_data));
-		rds_iw_stats_inc(s_iw_tx_cq_event);
-
-		if (wc.status != IB_WC_SUCCESS) {
-			printk(KERN_ERR "WC Error:  status = %d opcode = %d\n", wc.status, wc.opcode);
-			break;
-		}
-
-		if (wc.opcode == IB_WC_LOCAL_INV && wc.wr_id == RDS_IW_LOCAL_INV_WR_ID) {
-			ic->i_fastreg_posted = 0;
-			continue;
-		}
-
-		if (wc.opcode == IB_WC_REG_MR && wc.wr_id == RDS_IW_REG_WR_ID) {
-			ic->i_fastreg_posted = 1;
-			continue;
-		}
-
-		if (wc.wr_id == RDS_IW_ACK_WR_ID) {
-			if (time_after(jiffies, ic->i_ack_queued + HZ/2))
-				rds_iw_stats_inc(s_iw_tx_stalled);
-			rds_iw_ack_send_complete(ic);
-			continue;
-		}
-
-		oldest = rds_iw_ring_oldest(&ic->i_send_ring);
-
-		completed = rds_iw_ring_completed(&ic->i_send_ring, wc.wr_id, oldest);
-
-		for (i = 0; i < completed; i++) {
-			send = &ic->i_sends[oldest];
-
-			/* In the error case, wc.opcode sometimes contains garbage */
-			switch (send->s_send_wr.opcode) {
-			case IB_WR_SEND:
-				if (send->s_rm)
-					rds_iw_send_unmap_rm(ic, send, wc.status);
-				break;
-			case IB_WR_REG_MR:
-			case IB_WR_RDMA_WRITE:
-			case IB_WR_RDMA_READ:
-			case IB_WR_RDMA_READ_WITH_INV:
-				/* Nothing to be done - the SG list will be unmapped
-				 * when the SEND completes. */
-				break;
-			default:
-				printk_ratelimited(KERN_NOTICE
-						"RDS/IW: %s: unexpected opcode 0x%x in WR!\n",
-						__func__, send->s_send_wr.opcode);
-				break;
-			}
-
-			send->s_send_wr.opcode = 0xdead;
-			send->s_send_wr.num_sge = 1;
-			if (time_after(jiffies, send->s_queued + HZ/2))
-				rds_iw_stats_inc(s_iw_tx_stalled);
-
-			/* If a RDMA operation produced an error, signal this right
-			 * away. If we don't, the subsequent SEND that goes with this
-			 * RDMA will be canceled with ERR_WFLUSH, and the application
-			 * never learn that the RDMA failed. */
-			if (unlikely(wc.status == IB_WC_REM_ACCESS_ERR && send->s_op)) {
-				struct rds_message *rm;
-
-				rm = rds_send_get_message(conn, send->s_op);
-				if (rm)
-					rds_iw_send_rdma_complete(rm, wc.status);
-			}
-
-			oldest = (oldest + 1) % ic->i_send_ring.w_nr;
-		}
-
-		rds_iw_ring_free(&ic->i_send_ring, completed);
-
-		if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags) ||
-		    test_bit(0, &conn->c_map_queued))
-			queue_delayed_work(rds_wq, &conn->c_send_w, 0);
-
-		/* We expect errors as the qp is drained during shutdown */
-		if (wc.status != IB_WC_SUCCESS && rds_conn_up(conn)) {
-			rds_iw_conn_error(conn,
-				"send completion on %pI4 "
-				"had status %u, disconnecting and reconnecting\n",
-				&conn->c_faddr, wc.status);
-		}
-	}
-}
-
-/*
- * This is the main function for allocating credits when sending
- * messages.
- *
- * Conceptually, we have two counters:
- *  -	send credits: this tells us how many WRs we're allowed
- *	to submit without overruning the receiver's queue. For
- *	each SEND WR we post, we decrement this by one.
- *
- *  -	posted credits: this tells us how many WRs we recently
- *	posted to the receive queue. This value is transferred
- *	to the peer as a "credit update" in a RDS header field.
- *	Every time we transmit credits to the peer, we subtract
- *	the amount of transferred credits from this counter.
- *
- * It is essential that we avoid situations where both sides have
- * exhausted their send credits, and are unable to send new credits
- * to the peer. We achieve this by requiring that we send at least
- * one credit update to the peer before exhausting our credits.
- * When new credits arrive, we subtract one credit that is withheld
- * until we've posted new buffers and are ready to transmit these
- * credits (see rds_iw_send_add_credits below).
- *
- * The RDS send code is essentially single-threaded; rds_send_xmit
- * grabs c_send_lock to ensure exclusive access to the send ring.
- * However, the ACK sending code is independent and can race with
- * message SENDs.
- *
- * In the send path, we need to update the counters for send credits
- * and the counter of posted buffers atomically - when we use the
- * last available credit, we cannot allow another thread to race us
- * and grab the posted credits counter.  Hence, we have to use a
- * spinlock to protect the credit counter, or use atomics.
- *
- * Spinlocks shared between the send and the receive path are bad,
- * because they create unnecessary delays. An early implementation
- * using a spinlock showed a 5% degradation in throughput at some
- * loads.
- *
- * This implementation avoids spinlocks completely, putting both
- * counters into a single atomic, and updating that atomic using
- * atomic_add (in the receive path, when receiving fresh credits),
- * and using atomic_cmpxchg when updating the two counters.
- */
-int rds_iw_send_grab_credits(struct rds_iw_connection *ic,
-			     u32 wanted, u32 *adv_credits, int need_posted, int max_posted)
-{
-	unsigned int avail, posted, got = 0, advertise;
-	long oldval, newval;
-
-	*adv_credits = 0;
-	if (!ic->i_flowctl)
-		return wanted;
-
-try_again:
-	advertise = 0;
-	oldval = newval = atomic_read(&ic->i_credits);
-	posted = IB_GET_POST_CREDITS(oldval);
-	avail = IB_GET_SEND_CREDITS(oldval);
-
-	rdsdebug("wanted=%u credits=%u posted=%u\n",
-			wanted, avail, posted);
-
-	/* The last credit must be used to send a credit update. */
-	if (avail && !posted)
-		avail--;
-
-	if (avail < wanted) {
-		struct rds_connection *conn = ic->i_cm_id->context;
-
-		/* Oops, there aren't that many credits left! */
-		set_bit(RDS_LL_SEND_FULL, &conn->c_flags);
-		got = avail;
-	} else {
-		/* Sometimes you get what you want, lalala. */
-		got = wanted;
-	}
-	newval -= IB_SET_SEND_CREDITS(got);
-
-	/*
-	 * If need_posted is non-zero, then the caller wants
-	 * the posted regardless of whether any send credits are
-	 * available.
-	 */
-	if (posted && (got || need_posted)) {
-		advertise = min_t(unsigned int, posted, max_posted);
-		newval -= IB_SET_POST_CREDITS(advertise);
-	}
-
-	/* Finally bill everything */
-	if (atomic_cmpxchg(&ic->i_credits, oldval, newval) != oldval)
-		goto try_again;
-
-	*adv_credits = advertise;
-	return got;
-}
-
-void rds_iw_send_add_credits(struct rds_connection *conn, unsigned int credits)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	if (credits == 0)
-		return;
-
-	rdsdebug("credits=%u current=%u%s\n",
-			credits,
-			IB_GET_SEND_CREDITS(atomic_read(&ic->i_credits)),
-			test_bit(RDS_LL_SEND_FULL, &conn->c_flags) ? ", ll_send_full" : "");
-
-	atomic_add(IB_SET_SEND_CREDITS(credits), &ic->i_credits);
-	if (test_and_clear_bit(RDS_LL_SEND_FULL, &conn->c_flags))
-		queue_delayed_work(rds_wq, &conn->c_send_w, 0);
-
-	WARN_ON(IB_GET_SEND_CREDITS(credits) >= 16384);
-
-	rds_iw_stats_inc(s_iw_rx_credit_updates);
-}
-
-void rds_iw_advertise_credits(struct rds_connection *conn, unsigned int posted)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	if (posted == 0)
-		return;
-
-	atomic_add(IB_SET_POST_CREDITS(posted), &ic->i_credits);
-
-	/* Decide whether to send an update to the peer now.
-	 * If we would send a credit update for every single buffer we
-	 * post, we would end up with an ACK storm (ACK arrives,
-	 * consumes buffer, we refill the ring, send ACK to remote
-	 * advertising the newly posted buffer... ad inf)
-	 *
-	 * Performance pretty much depends on how often we send
-	 * credit updates - too frequent updates mean lots of ACKs.
-	 * Too infrequent updates, and the peer will run out of
-	 * credits and has to throttle.
-	 * For the time being, 16 seems to be a good compromise.
-	 */
-	if (IB_GET_POST_CREDITS(atomic_read(&ic->i_credits)) >= 16)
-		set_bit(IB_ACK_REQUESTED, &ic->i_ack_flags);
-}
-
-static inline void
-rds_iw_xmit_populate_wr(struct rds_iw_connection *ic,
-		struct rds_iw_send_work *send, unsigned int pos,
-		unsigned long buffer, unsigned int length,
-		int send_flags)
-{
-	struct ib_sge *sge;
-
-	WARN_ON(pos != send - ic->i_sends);
-
-	send->s_send_wr.send_flags = send_flags;
-	send->s_send_wr.opcode = IB_WR_SEND;
-	send->s_send_wr.num_sge = 2;
-	send->s_send_wr.next = NULL;
-	send->s_queued = jiffies;
-	send->s_op = NULL;
-
-	if (length != 0) {
-		sge = rds_iw_data_sge(ic, send->s_sge);
-		sge->addr = buffer;
-		sge->length = length;
-		sge->lkey = rds_iw_local_dma_lkey(ic);
-
-		sge = rds_iw_header_sge(ic, send->s_sge);
-	} else {
-		/* We're sending a packet with no payload. There is only
-		 * one SGE */
-		send->s_send_wr.num_sge = 1;
-		sge = &send->s_sge[0];
-	}
-
-	sge->addr = ic->i_send_hdrs_dma + (pos * sizeof(struct rds_header));
-	sge->length = sizeof(struct rds_header);
-	sge->lkey = rds_iw_local_dma_lkey(ic);
-}
-
-/*
- * This can be called multiple times for a given message.  The first time
- * we see a message we map its scatterlist into the IB device so that
- * we can provide that mapped address to the IB scatter gather entries
- * in the IB work requests.  We translate the scatterlist into a series
- * of work requests that fragment the message.  These work requests complete
- * in order so we pass ownership of the message to the completion handler
- * once we send the final fragment.
- *
- * The RDS core uses the c_send_lock to only enter this function once
- * per connection.  This makes sure that the tx ring alloc/unalloc pairs
- * don't get out of sync and confuse the ring.
- */
-int rds_iw_xmit(struct rds_connection *conn, struct rds_message *rm,
-		unsigned int hdr_off, unsigned int sg, unsigned int off)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct ib_device *dev = ic->i_cm_id->device;
-	struct rds_iw_send_work *send = NULL;
-	struct rds_iw_send_work *first;
-	struct rds_iw_send_work *prev;
-	struct ib_send_wr *failed_wr;
-	struct scatterlist *scat;
-	u32 pos;
-	u32 i;
-	u32 work_alloc;
-	u32 credit_alloc;
-	u32 posted;
-	u32 adv_credits = 0;
-	int send_flags = 0;
-	int sent;
-	int ret;
-	int flow_controlled = 0;
-
-	BUG_ON(off % RDS_FRAG_SIZE);
-	BUG_ON(hdr_off != 0 && hdr_off != sizeof(struct rds_header));
-
-	/* Fastreg support */
-	if (rds_rdma_cookie_key(rm->m_rdma_cookie) && !ic->i_fastreg_posted) {
-		ret = -EAGAIN;
-		goto out;
-	}
-
-	/* FIXME we may overallocate here */
-	if (be32_to_cpu(rm->m_inc.i_hdr.h_len) == 0)
-		i = 1;
-	else
-		i = ceil(be32_to_cpu(rm->m_inc.i_hdr.h_len), RDS_FRAG_SIZE);
-
-	work_alloc = rds_iw_ring_alloc(&ic->i_send_ring, i, &pos);
-	if (work_alloc == 0) {
-		set_bit(RDS_LL_SEND_FULL, &conn->c_flags);
-		rds_iw_stats_inc(s_iw_tx_ring_full);
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	credit_alloc = work_alloc;
-	if (ic->i_flowctl) {
-		credit_alloc = rds_iw_send_grab_credits(ic, work_alloc, &posted, 0, RDS_MAX_ADV_CREDIT);
-		adv_credits += posted;
-		if (credit_alloc < work_alloc) {
-			rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc - credit_alloc);
-			work_alloc = credit_alloc;
-			flow_controlled++;
-		}
-		if (work_alloc == 0) {
-			set_bit(RDS_LL_SEND_FULL, &conn->c_flags);
-			rds_iw_stats_inc(s_iw_tx_throttle);
-			ret = -ENOMEM;
-			goto out;
-		}
-	}
-
-	/* map the message the first time we see it */
-	if (!ic->i_rm) {
-		/*
-		printk(KERN_NOTICE "rds_iw_xmit prep msg dport=%u flags=0x%x len=%d\n",
-				be16_to_cpu(rm->m_inc.i_hdr.h_dport),
-				rm->m_inc.i_hdr.h_flags,
-				be32_to_cpu(rm->m_inc.i_hdr.h_len));
-		   */
-		if (rm->data.op_nents) {
-			rm->data.op_count = ib_dma_map_sg(dev,
-							  rm->data.op_sg,
-							  rm->data.op_nents,
-							  DMA_TO_DEVICE);
-			rdsdebug("ic %p mapping rm %p: %d\n", ic, rm, rm->data.op_count);
-			if (rm->data.op_count == 0) {
-				rds_iw_stats_inc(s_iw_tx_sg_mapping_failure);
-				rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc);
-				ret = -ENOMEM; /* XXX ? */
-				goto out;
-			}
-		} else {
-			rm->data.op_count = 0;
-		}
-
-		ic->i_unsignaled_wrs = rds_iw_sysctl_max_unsig_wrs;
-		ic->i_unsignaled_bytes = rds_iw_sysctl_max_unsig_bytes;
-		rds_message_addref(rm);
-		rm->data.op_dmasg = 0;
-		rm->data.op_dmaoff = 0;
-		ic->i_rm = rm;
-
-		/* Finalize the header */
-		if (test_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags))
-			rm->m_inc.i_hdr.h_flags |= RDS_FLAG_ACK_REQUIRED;
-		if (test_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags))
-			rm->m_inc.i_hdr.h_flags |= RDS_FLAG_RETRANSMITTED;
-
-		/* If it has a RDMA op, tell the peer we did it. This is
-		 * used by the peer to release use-once RDMA MRs. */
-		if (rm->rdma.op_active) {
-			struct rds_ext_header_rdma ext_hdr;
-
-			ext_hdr.h_rdma_rkey = cpu_to_be32(rm->rdma.op_rkey);
-			rds_message_add_extension(&rm->m_inc.i_hdr,
-					RDS_EXTHDR_RDMA, &ext_hdr, sizeof(ext_hdr));
-		}
-		if (rm->m_rdma_cookie) {
-			rds_message_add_rdma_dest_extension(&rm->m_inc.i_hdr,
-					rds_rdma_cookie_key(rm->m_rdma_cookie),
-					rds_rdma_cookie_offset(rm->m_rdma_cookie));
-		}
-
-		/* Note - rds_iw_piggyb_ack clears the ACK_REQUIRED bit, so
-		 * we should not do this unless we have a chance of at least
-		 * sticking the header into the send ring. Which is why we
-		 * should call rds_iw_ring_alloc first. */
-		rm->m_inc.i_hdr.h_ack = cpu_to_be64(rds_iw_piggyb_ack(ic));
-		rds_message_make_checksum(&rm->m_inc.i_hdr);
-
-		/*
-		 * Update adv_credits since we reset the ACK_REQUIRED bit.
-		 */
-		rds_iw_send_grab_credits(ic, 0, &posted, 1, RDS_MAX_ADV_CREDIT - adv_credits);
-		adv_credits += posted;
-		BUG_ON(adv_credits > 255);
-	}
-
-	send = &ic->i_sends[pos];
-	first = send;
-	prev = NULL;
-	scat = &rm->data.op_sg[rm->data.op_dmasg];
-	sent = 0;
-	i = 0;
-
-	/* Sometimes you want to put a fence between an RDMA
-	 * READ and the following SEND.
-	 * We could either do this all the time
-	 * or when requested by the user. Right now, we let
-	 * the application choose.
-	 */
-	if (rm->rdma.op_active && rm->rdma.op_fence)
-		send_flags = IB_SEND_FENCE;
-
-	/*
-	 * We could be copying the header into the unused tail of the page.
-	 * That would need to be changed in the future when those pages might
-	 * be mapped userspace pages or page cache pages.  So instead we always
-	 * use a second sge and our long-lived ring of mapped headers.  We send
-	 * the header after the data so that the data payload can be aligned on
-	 * the receiver.
-	 */
-
-	/* handle a 0-len message */
-	if (be32_to_cpu(rm->m_inc.i_hdr.h_len) == 0) {
-		rds_iw_xmit_populate_wr(ic, send, pos, 0, 0, send_flags);
-		goto add_header;
-	}
-
-	/* if there's data reference it with a chain of work reqs */
-	for (; i < work_alloc && scat != &rm->data.op_sg[rm->data.op_count]; i++) {
-		unsigned int len;
-
-		send = &ic->i_sends[pos];
-
-		len = min(RDS_FRAG_SIZE,
-			  ib_sg_dma_len(dev, scat) - rm->data.op_dmaoff);
-		rds_iw_xmit_populate_wr(ic, send, pos,
-			ib_sg_dma_address(dev, scat) + rm->data.op_dmaoff, len,
-			send_flags);
-
-		/*
-		 * We want to delay signaling completions just enough to get
-		 * the batching benefits but not so much that we create dead time
-		 * on the wire.
-		 */
-		if (ic->i_unsignaled_wrs-- == 0) {
-			ic->i_unsignaled_wrs = rds_iw_sysctl_max_unsig_wrs;
-			send->s_send_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED;
-		}
-
-		ic->i_unsignaled_bytes -= len;
-		if (ic->i_unsignaled_bytes <= 0) {
-			ic->i_unsignaled_bytes = rds_iw_sysctl_max_unsig_bytes;
-			send->s_send_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED;
-		}
-
-		/*
-		 * Always signal the last one if we're stopping due to flow control.
-		 */
-		if (flow_controlled && i == (work_alloc-1))
-			send->s_send_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED;
-
-		rdsdebug("send %p wr %p num_sge %u next %p\n", send,
-			 &send->s_send_wr, send->s_send_wr.num_sge, send->s_send_wr.next);
-
-		sent += len;
-		rm->data.op_dmaoff += len;
-		if (rm->data.op_dmaoff == ib_sg_dma_len(dev, scat)) {
-			scat++;
-			rm->data.op_dmaoff = 0;
-			rm->data.op_dmasg++;
-		}
-
-add_header:
-		/* Tack on the header after the data. The header SGE should already
-		 * have been set up to point to the right header buffer. */
-		memcpy(&ic->i_send_hdrs[pos], &rm->m_inc.i_hdr, sizeof(struct rds_header));
-
-		if (0) {
-			struct rds_header *hdr = &ic->i_send_hdrs[pos];
-
-			printk(KERN_NOTICE "send WR dport=%u flags=0x%x len=%d\n",
-				be16_to_cpu(hdr->h_dport),
-				hdr->h_flags,
-				be32_to_cpu(hdr->h_len));
-		}
-		if (adv_credits) {
-			struct rds_header *hdr = &ic->i_send_hdrs[pos];
-
-			/* add credit and redo the header checksum */
-			hdr->h_credit = adv_credits;
-			rds_message_make_checksum(hdr);
-			adv_credits = 0;
-			rds_iw_stats_inc(s_iw_tx_credit_updates);
-		}
-
-		if (prev)
-			prev->s_send_wr.next = &send->s_send_wr;
-		prev = send;
-
-		pos = (pos + 1) % ic->i_send_ring.w_nr;
-	}
-
-	/* Account the RDS header in the number of bytes we sent, but just once.
-	 * The caller has no concept of fragmentation. */
-	if (hdr_off == 0)
-		sent += sizeof(struct rds_header);
-
-	/* if we finished the message then send completion owns it */
-	if (scat == &rm->data.op_sg[rm->data.op_count]) {
-		prev->s_rm = ic->i_rm;
-		prev->s_send_wr.send_flags |= IB_SEND_SIGNALED | IB_SEND_SOLICITED;
-		ic->i_rm = NULL;
-	}
-
-	if (i < work_alloc) {
-		rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc - i);
-		work_alloc = i;
-	}
-	if (ic->i_flowctl && i < credit_alloc)
-		rds_iw_send_add_credits(conn, credit_alloc - i);
-
-	/* XXX need to worry about failed_wr and partial sends. */
-	failed_wr = &first->s_send_wr;
-	ret = ib_post_send(ic->i_cm_id->qp, &first->s_send_wr, &failed_wr);
-	rdsdebug("ic %p first %p (wr %p) ret %d wr %p\n", ic,
-		 first, &first->s_send_wr, ret, failed_wr);
-	BUG_ON(failed_wr != &first->s_send_wr);
-	if (ret) {
-		printk(KERN_WARNING "RDS/IW: ib_post_send to %pI4 "
-		       "returned %d\n", &conn->c_faddr, ret);
-		rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc);
-		if (prev->s_rm) {
-			ic->i_rm = prev->s_rm;
-			prev->s_rm = NULL;
-		}
-		goto out;
-	}
-
-	ret = sent;
-out:
-	BUG_ON(adv_credits);
-	return ret;
-}
-
-static int rds_iw_build_send_reg(struct rds_iw_send_work *send,
-				 struct scatterlist *sg,
-				 int sg_nents)
-{
-	int n;
-
-	n = ib_map_mr_sg(send->s_mr, sg, sg_nents, PAGE_SIZE);
-	if (unlikely(n != sg_nents))
-		return n < 0 ? n : -EINVAL;
-
-	send->s_reg_wr.wr.opcode = IB_WR_REG_MR;
-	send->s_reg_wr.wr.wr_id = 0;
-	send->s_reg_wr.wr.num_sge = 0;
-	send->s_reg_wr.mr = send->s_mr;
-	send->s_reg_wr.key = send->s_mr->rkey;
-	send->s_reg_wr.access = IB_ACCESS_REMOTE_WRITE;
-
-	ib_update_fast_reg_key(send->s_mr, send->s_remap_count++);
-
-	return 0;
-}
-
-int rds_iw_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-	struct rds_iw_send_work *send = NULL;
-	struct rds_iw_send_work *first;
-	struct rds_iw_send_work *prev;
-	struct ib_send_wr *failed_wr;
-	struct rds_iw_device *rds_iwdev;
-	struct scatterlist *scat;
-	unsigned long len;
-	u64 remote_addr = op->op_remote_addr;
-	u32 pos, fr_pos;
-	u32 work_alloc;
-	u32 i;
-	u32 j;
-	int sent;
-	int ret;
-	int num_sge;
-	int sg_nents;
-
-	rds_iwdev = ib_get_client_data(ic->i_cm_id->device, &rds_iw_client);
-
-	/* map the message the first time we see it */
-	if (!op->op_mapped) {
-		op->op_count = ib_dma_map_sg(ic->i_cm_id->device,
-					     op->op_sg, op->op_nents, (op->op_write) ?
-					     DMA_TO_DEVICE : DMA_FROM_DEVICE);
-		rdsdebug("ic %p mapping op %p: %d\n", ic, op, op->op_count);
-		if (op->op_count == 0) {
-			rds_iw_stats_inc(s_iw_tx_sg_mapping_failure);
-			ret = -ENOMEM; /* XXX ? */
-			goto out;
-		}
-
-		op->op_mapped = 1;
-	}
-
-	if (!op->op_write) {
-		/* Alloc space on the send queue for the fastreg */
-		work_alloc = rds_iw_ring_alloc(&ic->i_send_ring, 1, &fr_pos);
-		if (work_alloc != 1) {
-			rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc);
-			rds_iw_stats_inc(s_iw_tx_ring_full);
-			ret = -ENOMEM;
-			goto out;
-		}
-	}
-
-	/*
-	 * Instead of knowing how to return a partial rdma read/write we insist that there
-	 * be enough work requests to send the entire message.
-	 */
-	i = ceil(op->op_count, rds_iwdev->max_sge);
-
-	work_alloc = rds_iw_ring_alloc(&ic->i_send_ring, i, &pos);
-	if (work_alloc != i) {
-		rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc);
-		rds_iw_stats_inc(s_iw_tx_ring_full);
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	send = &ic->i_sends[pos];
-	if (!op->op_write) {
-		first = prev = &ic->i_sends[fr_pos];
-	} else {
-		first = send;
-		prev = NULL;
-	}
-	scat = &op->op_sg[0];
-	sent = 0;
-	num_sge = op->op_count;
-	sg_nents = 0;
-
-	for (i = 0; i < work_alloc && scat != &op->op_sg[op->op_count]; i++) {
-		send->s_rdma_wr.wr.send_flags = 0;
-		send->s_queued = jiffies;
-
-		/*
-		 * We want to delay signaling completions just enough to get
-		 * the batching benefits but not so much that we create dead time on the wire.
-		 */
-		if (ic->i_unsignaled_wrs-- == 0) {
-			ic->i_unsignaled_wrs = rds_iw_sysctl_max_unsig_wrs;
-			send->s_rdma_wr.wr.send_flags = IB_SEND_SIGNALED;
-		}
-
-		/* To avoid the need to have the plumbing to invalidate the fastreg_mr used
-		 * for local access after RDS is finished with it, using
-		 * IB_WR_RDMA_READ_WITH_INV will invalidate it after the read has completed.
-		 */
-		if (op->op_write)
-			send->s_rdma_wr.wr.opcode = IB_WR_RDMA_WRITE;
-		else
-			send->s_rdma_wr.wr.opcode = IB_WR_RDMA_READ_WITH_INV;
-
-		send->s_rdma_wr.remote_addr = remote_addr;
-		send->s_rdma_wr.rkey = op->op_rkey;
-		send->s_op = op;
-
-		if (num_sge > rds_iwdev->max_sge) {
-			send->s_rdma_wr.wr.num_sge = rds_iwdev->max_sge;
-			num_sge -= rds_iwdev->max_sge;
-		} else
-			send->s_rdma_wr.wr.num_sge = num_sge;
-
-		send->s_rdma_wr.wr.next = NULL;
-
-		if (prev)
-			prev->s_send_wr.next = &send->s_rdma_wr.wr;
-
-		for (j = 0; j < send->s_rdma_wr.wr.num_sge &&
-		     scat != &op->op_sg[op->op_count]; j++) {
-			len = ib_sg_dma_len(ic->i_cm_id->device, scat);
-
-			if (send->s_rdma_wr.wr.opcode == IB_WR_RDMA_READ_WITH_INV)
-				sg_nents++;
-			else {
-				send->s_sge[j].addr = ib_sg_dma_address(ic->i_cm_id->device, scat);
-				send->s_sge[j].length = len;
-				send->s_sge[j].lkey = rds_iw_local_dma_lkey(ic);
-			}
-
-			sent += len;
-			rdsdebug("ic %p sent %d remote_addr %llu\n", ic, sent, remote_addr);
-			remote_addr += len;
-
-			scat++;
-		}
-
-		if (send->s_rdma_wr.wr.opcode == IB_WR_RDMA_READ_WITH_INV) {
-			send->s_rdma_wr.wr.num_sge = 1;
-			send->s_sge[0].addr = conn->c_xmit_rm->m_rs->rs_user_addr;
-			send->s_sge[0].length = conn->c_xmit_rm->m_rs->rs_user_bytes;
-			send->s_sge[0].lkey = ic->i_sends[fr_pos].s_mr->lkey;
-		}
-
-		rdsdebug("send %p wr %p num_sge %u next %p\n", send,
-			&send->s_rdma_wr,
-			send->s_rdma_wr.wr.num_sge,
-			send->s_rdma_wr.wr.next);
-
-		prev = send;
-		if (++send == &ic->i_sends[ic->i_send_ring.w_nr])
-			send = ic->i_sends;
-	}
-
-	/* if we finished the message then send completion owns it */
-	if (scat == &op->op_sg[op->op_count])
-		first->s_rdma_wr.wr.send_flags = IB_SEND_SIGNALED;
-
-	if (i < work_alloc) {
-		rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc - i);
-		work_alloc = i;
-	}
-
-	/* On iWARP, local memory access by a remote system (ie, RDMA Read) is not
-	 * recommended.  Putting the lkey on the wire is a security hole, as it can
-	 * allow for memory access to all of memory on the remote system.  Some
-	 * adapters do not allow using the lkey for this at all.  To bypass this use a
-	 * fastreg_mr (or possibly a dma_mr)
-	 */
-	if (!op->op_write) {
-		ret = rds_iw_build_send_reg(&ic->i_sends[fr_pos],
-					    &op->op_sg[0], sg_nents);
-		if (ret) {
-			printk(KERN_WARNING "RDS/IW: failed to reg send mem\n");
-			goto out;
-		}
-		work_alloc++;
-	}
-
-	failed_wr = &first->s_rdma_wr.wr;
-	ret = ib_post_send(ic->i_cm_id->qp, &first->s_rdma_wr.wr, &failed_wr);
-	rdsdebug("ic %p first %p (wr %p) ret %d wr %p\n", ic,
-		 first, &first->s_rdma_wr, ret, failed_wr);
-	BUG_ON(failed_wr != &first->s_rdma_wr.wr);
-	if (ret) {
-		printk(KERN_WARNING "RDS/IW: rdma ib_post_send to %pI4 "
-		       "returned %d\n", &conn->c_faddr, ret);
-		rds_iw_ring_unalloc(&ic->i_send_ring, work_alloc);
-		goto out;
-	}
-
-out:
-	return ret;
-}
-
-void rds_iw_xmit_complete(struct rds_connection *conn)
-{
-	struct rds_iw_connection *ic = conn->c_transport_data;
-
-	/* We may have a pending ACK or window update we were unable
-	 * to send previously (due to flow control). Try again. */
-	rds_iw_attempt_ack(ic);
-}
diff --git a/net/rds/iw_stats.c b/net/rds/iw_stats.c
deleted file mode 100644
index 5fe67f6..0000000
--- a/net/rds/iw_stats.c
+++ /dev/null
@@ -1,95 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/percpu.h>
-#include <linux/seq_file.h>
-#include <linux/proc_fs.h>
-
-#include "rds.h"
-#include "iw.h"
-
-DEFINE_PER_CPU_SHARED_ALIGNED(struct rds_iw_statistics, rds_iw_stats);
-
-static const char *const rds_iw_stat_names[] = {
-	"iw_connect_raced",
-	"iw_listen_closed_stale",
-	"iw_tx_cq_call",
-	"iw_tx_cq_event",
-	"iw_tx_ring_full",
-	"iw_tx_throttle",
-	"iw_tx_sg_mapping_failure",
-	"iw_tx_stalled",
-	"iw_tx_credit_updates",
-	"iw_rx_cq_call",
-	"iw_rx_cq_event",
-	"iw_rx_ring_empty",
-	"iw_rx_refill_from_cq",
-	"iw_rx_refill_from_thread",
-	"iw_rx_alloc_limit",
-	"iw_rx_credit_updates",
-	"iw_ack_sent",
-	"iw_ack_send_failure",
-	"iw_ack_send_delayed",
-	"iw_ack_send_piggybacked",
-	"iw_ack_received",
-	"iw_rdma_mr_alloc",
-	"iw_rdma_mr_free",
-	"iw_rdma_mr_used",
-	"iw_rdma_mr_pool_flush",
-	"iw_rdma_mr_pool_wait",
-	"iw_rdma_mr_pool_depleted",
-};
-
-unsigned int rds_iw_stats_info_copy(struct rds_info_iterator *iter,
-				    unsigned int avail)
-{
-	struct rds_iw_statistics stats = {0, };
-	uint64_t *src;
-	uint64_t *sum;
-	size_t i;
-	int cpu;
-
-	if (avail < ARRAY_SIZE(rds_iw_stat_names))
-		goto out;
-
-	for_each_online_cpu(cpu) {
-		src = (uint64_t *)&(per_cpu(rds_iw_stats, cpu));
-		sum = (uint64_t *)&stats;
-		for (i = 0; i < sizeof(stats) / sizeof(uint64_t); i++)
-			*(sum++) += *(src++);
-	}
-
-	rds_stats_info_copy(iter, (uint64_t *)&stats, rds_iw_stat_names,
-			    ARRAY_SIZE(rds_iw_stat_names));
-out:
-	return ARRAY_SIZE(rds_iw_stat_names);
-}
diff --git a/net/rds/iw_sysctl.c b/net/rds/iw_sysctl.c
deleted file mode 100644
index 139239d..0000000
--- a/net/rds/iw_sysctl.c
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * Copyright (c) 2006 Oracle.  All rights reserved.
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the
- * OpenIB.org BSD license below:
- *
- *     Redistribution and use in source and binary forms, with or
- *     without modification, are permitted provided that the following
- *     conditions are met:
- *
- *      - Redistributions of source code must retain the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer.
- *
- *      - Redistributions in binary form must reproduce the above
- *        copyright notice, this list of conditions and the following
- *        disclaimer in the documentation and/or other materials
- *        provided with the distribution.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
- * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
- * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
- * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
- * SOFTWARE.
- *
- */
-#include <linux/kernel.h>
-#include <linux/sysctl.h>
-#include <linux/proc_fs.h>
-
-#include "iw.h"
-
-static struct ctl_table_header *rds_iw_sysctl_hdr;
-
-unsigned long rds_iw_sysctl_max_send_wr = RDS_IW_DEFAULT_SEND_WR;
-unsigned long rds_iw_sysctl_max_recv_wr = RDS_IW_DEFAULT_RECV_WR;
-unsigned long rds_iw_sysctl_max_recv_allocation = (128 * 1024 * 1024) / RDS_FRAG_SIZE;
-static unsigned long rds_iw_sysctl_max_wr_min = 1;
-/* hardware will fail CQ creation long before this */
-static unsigned long rds_iw_sysctl_max_wr_max = (u32)~0;
-
-unsigned long rds_iw_sysctl_max_unsig_wrs = 16;
-static unsigned long rds_iw_sysctl_max_unsig_wr_min = 1;
-static unsigned long rds_iw_sysctl_max_unsig_wr_max = 64;
-
-unsigned long rds_iw_sysctl_max_unsig_bytes = (16 << 20);
-static unsigned long rds_iw_sysctl_max_unsig_bytes_min = 1;
-static unsigned long rds_iw_sysctl_max_unsig_bytes_max = ~0UL;
-
-unsigned int rds_iw_sysctl_flow_control = 1;
-
-static struct ctl_table rds_iw_sysctl_table[] = {
-	{
-		.procname       = "max_send_wr",
-		.data		= &rds_iw_sysctl_max_send_wr,
-		.maxlen         = sizeof(unsigned long),
-		.mode           = 0644,
-		.proc_handler   = proc_doulongvec_minmax,
-		.extra1		= &rds_iw_sysctl_max_wr_min,
-		.extra2		= &rds_iw_sysctl_max_wr_max,
-	},
-	{
-		.procname       = "max_recv_wr",
-		.data		= &rds_iw_sysctl_max_recv_wr,
-		.maxlen         = sizeof(unsigned long),
-		.mode           = 0644,
-		.proc_handler   = proc_doulongvec_minmax,
-		.extra1		= &rds_iw_sysctl_max_wr_min,
-		.extra2		= &rds_iw_sysctl_max_wr_max,
-	},
-	{
-		.procname       = "max_unsignaled_wr",
-		.data		= &rds_iw_sysctl_max_unsig_wrs,
-		.maxlen         = sizeof(unsigned long),
-		.mode           = 0644,
-		.proc_handler   = proc_doulongvec_minmax,
-		.extra1		= &rds_iw_sysctl_max_unsig_wr_min,
-		.extra2		= &rds_iw_sysctl_max_unsig_wr_max,
-	},
-	{
-		.procname       = "max_unsignaled_bytes",
-		.data		= &rds_iw_sysctl_max_unsig_bytes,
-		.maxlen         = sizeof(unsigned long),
-		.mode           = 0644,
-		.proc_handler   = proc_doulongvec_minmax,
-		.extra1		= &rds_iw_sysctl_max_unsig_bytes_min,
-		.extra2		= &rds_iw_sysctl_max_unsig_bytes_max,
-	},
-	{
-		.procname       = "max_recv_allocation",
-		.data		= &rds_iw_sysctl_max_recv_allocation,
-		.maxlen         = sizeof(unsigned long),
-		.mode           = 0644,
-		.proc_handler   = proc_doulongvec_minmax,
-	},
-	{
-		.procname	= "flow_control",
-		.data		= &rds_iw_sysctl_flow_control,
-		.maxlen		= sizeof(rds_iw_sysctl_flow_control),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
-	},
-	{ }
-};
-
-void rds_iw_sysctl_exit(void)
-{
-	unregister_net_sysctl_table(rds_iw_sysctl_hdr);
-}
-
-int rds_iw_sysctl_init(void)
-{
-	rds_iw_sysctl_hdr = register_net_sysctl(&init_net, "net/rds/iw", rds_iw_sysctl_table);
-	if (!rds_iw_sysctl_hdr)
-		return -ENOMEM;
-	return 0;
-}
diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c
index 9c1fed8..4f4b3d8 100644
--- a/net/rds/rdma_transport.c
+++ b/net/rds/rdma_transport.c
@@ -49,9 +49,7 @@ int rds_rdma_cm_event_handler(struct rdma_cm_id *cm_id,
 	rdsdebug("conn %p id %p handling event %u (%s)\n", conn, cm_id,
 		 event->event, rdma_event_msg(event->event));
 
-	if (cm_id->device->node_type == RDMA_NODE_RNIC)
-		trans = &rds_iw_transport;
-	else
+	if (cm_id->device->node_type == RDMA_NODE_IB_CA)
 		trans = &rds_ib_transport;
 
 	/* Prevent shutdown from tearing down the connection
@@ -200,10 +198,6 @@ static int rds_rdma_init(void)
 	if (ret)
 		goto out;
 
-	ret = rds_iw_init();
-	if (ret)
-		goto err_iw_init;
-
 	ret = rds_ib_init();
 	if (ret)
 		goto err_ib_init;
@@ -211,8 +205,6 @@ static int rds_rdma_init(void)
 	goto out;
 
 err_ib_init:
-	rds_iw_exit();
-err_iw_init:
 	rds_rdma_listen_stop();
 out:
 	return ret;
@@ -224,11 +216,10 @@ static void rds_rdma_exit(void)
 	/* stop listening first to ensure no new connections are attempted */
 	rds_rdma_listen_stop();
 	rds_ib_exit();
-	rds_iw_exit();
 }
 module_exit(rds_rdma_exit);
 
 MODULE_AUTHOR("Oracle Corporation <rds-devel@oss.oracle.com>");
-MODULE_DESCRIPTION("RDS: IB/iWARP transport");
+MODULE_DESCRIPTION("RDS: IB transport");
 MODULE_LICENSE("Dual BSD/GPL");
 
diff --git a/net/rds/rdma_transport.h b/net/rds/rdma_transport.h
index faba4e3..ff2010e 100644
--- a/net/rds/rdma_transport.h
+++ b/net/rds/rdma_transport.h
@@ -16,9 +16,4 @@ extern struct rds_transport rds_ib_transport;
 int rds_ib_init(void);
 void rds_ib_exit(void);
 
-/* from iw.c */
-extern struct rds_transport rds_iw_transport;
-int rds_iw_init(void);
-void rds_iw_exit(void);
-
 #endif
-- 
cgit v0.10.2


From 5711f8b353176393ee2ab023808a0a8c4a9d39b5 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:43 -0800
Subject: RDS: Add support for SO_TIMESTAMP for incoming messages

The SO_TIMESTAMP generates time stamp for each incoming RDS messages
User app can enable it by using SO_TIMESTAMP setsocketopt() at
SOL_SOCKET level. CMSG data of cmsg type SO_TIMESTAMP contains the
time stamp in struct timeval format.

Reviewed-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/af_rds.c b/net/rds/af_rds.c
index b5476aeb..6beaeb1 100644
--- a/net/rds/af_rds.c
+++ b/net/rds/af_rds.c
@@ -277,6 +277,27 @@ static int rds_set_transport(struct rds_sock *rs, char __user *optval,
 	return rs->rs_transport ? 0 : -ENOPROTOOPT;
 }
 
+static int rds_enable_recvtstamp(struct sock *sk, char __user *optval,
+				 int optlen)
+{
+	int val, valbool;
+
+	if (optlen != sizeof(int))
+		return -EFAULT;
+
+	if (get_user(val, (int __user *)optval))
+		return -EFAULT;
+
+	valbool = val ? 1 : 0;
+
+	if (valbool)
+		sock_set_flag(sk, SOCK_RCVTSTAMP);
+	else
+		sock_reset_flag(sk, SOCK_RCVTSTAMP);
+
+	return 0;
+}
+
 static int rds_setsockopt(struct socket *sock, int level, int optname,
 			  char __user *optval, unsigned int optlen)
 {
@@ -312,6 +333,11 @@ static int rds_setsockopt(struct socket *sock, int level, int optname,
 		ret = rds_set_transport(rs, optval, optlen);
 		release_sock(sock->sk);
 		break;
+	case SO_TIMESTAMP:
+		lock_sock(sock->sk);
+		ret = rds_enable_recvtstamp(sock->sk, optval, optlen);
+		release_sock(sock->sk);
+		break;
 	default:
 		ret = -ENOPROTOOPT;
 	}
diff --git a/net/rds/rds.h b/net/rds/rds.h
index 0e2797b..80256b0 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -222,6 +222,7 @@ struct rds_incoming {
 	__be32			i_saddr;
 
 	rds_rdma_cookie_t	i_rdma_cookie;
+	struct timeval		i_rx_tstamp;
 };
 
 struct rds_mr {
diff --git a/net/rds/recv.c b/net/rds/recv.c
index a00462b..c0be1ec 100644
--- a/net/rds/recv.c
+++ b/net/rds/recv.c
@@ -35,6 +35,8 @@
 #include <net/sock.h>
 #include <linux/in.h>
 #include <linux/export.h>
+#include <linux/time.h>
+#include <linux/rds.h>
 
 #include "rds.h"
 
@@ -46,6 +48,8 @@ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn,
 	inc->i_conn = conn;
 	inc->i_saddr = saddr;
 	inc->i_rdma_cookie = 0;
+	inc->i_rx_tstamp.tv_sec = 0;
+	inc->i_rx_tstamp.tv_usec = 0;
 }
 EXPORT_SYMBOL_GPL(rds_inc_init);
 
@@ -228,6 +232,8 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr,
 		rds_recv_rcvbuf_delta(rs, sk, inc->i_conn->c_lcong,
 				      be32_to_cpu(inc->i_hdr.h_len),
 				      inc->i_hdr.h_dport);
+		if (sock_flag(sk, SOCK_RCVTSTAMP))
+			do_gettimeofday(&inc->i_rx_tstamp);
 		rds_inc_addref(inc);
 		list_add_tail(&inc->i_item, &rs->rs_recv_queue);
 		__rds_wake_sk_sleep(sk);
@@ -381,7 +387,8 @@ static int rds_notify_cong(struct rds_sock *rs, struct msghdr *msghdr)
 /*
  * Receive any control messages.
  */
-static int rds_cmsg_recv(struct rds_incoming *inc, struct msghdr *msg)
+static int rds_cmsg_recv(struct rds_incoming *inc, struct msghdr *msg,
+			 struct rds_sock *rs)
 {
 	int ret = 0;
 
@@ -392,6 +399,15 @@ static int rds_cmsg_recv(struct rds_incoming *inc, struct msghdr *msg)
 			return ret;
 	}
 
+	if ((inc->i_rx_tstamp.tv_sec != 0) &&
+	    sock_flag(rds_rs_to_sk(rs), SOCK_RCVTSTAMP)) {
+		ret = put_cmsg(msg, SOL_SOCKET, SCM_TIMESTAMP,
+			       sizeof(struct timeval),
+			       &inc->i_rx_tstamp);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -474,7 +490,7 @@ int rds_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
 			msg->msg_flags |= MSG_TRUNC;
 		}
 
-		if (rds_cmsg_recv(inc, msg)) {
+		if (rds_cmsg_recv(inc, msg, rs)) {
 			ret = -EFAULT;
 			goto out;
 		}
-- 
cgit v0.10.2


From 72f26eee51e89c9d13b3aa199262fad57386f9e5 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:44 -0800
Subject: MAINTAINERS: update RDS entry

Acked-by: Chien Yen <chien.yen@oracle.com>
Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/MAINTAINERS b/MAINTAINERS
index 0499f9e..9c4d157 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9076,10 +9076,14 @@ S:	Maintained
 F:	drivers/net/ethernet/rdc/r6040.c
 
 RDS - RELIABLE DATAGRAM SOCKETS
-M:	Chien Yen <chien.yen@oracle.com>
+M:	Santosh Shilimkar <santosh.shilimkar@oracle.com>
+L:	netdev@vger.kernel.org
+L:	linux-rdma@vger.kernel.org
 L:	rds-devel@oss.oracle.com (moderated for non-subscribers)
+W:	https://oss.oracle.com/projects/rds/
 S:	Supported
 F:	net/rds/
+F:	Documentation/networking/rds.txt
 
 READ-COPY UPDATE (RCU)
 M:	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
-- 
cgit v0.10.2


From dcfd041c8710320d59fce322fd901bddaf912ae8 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:45 -0800
Subject: RDS: IB: Remove the RDS_IB_SEND_OP dependency

This helps to combine asynchronous fastreg MR completion handler
with send completion handler.

No functional change.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib.h b/net/rds/ib.h
index b3fdebb..09cd8e3 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -28,7 +28,6 @@
 #define RDS_IB_RECYCLE_BATCH_COUNT	32
 
 #define RDS_IB_WC_MAX			32
-#define RDS_IB_SEND_OP			BIT_ULL(63)
 
 extern struct rw_semaphore rds_ib_devices_lock;
 extern struct list_head rds_ib_devices;
diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c
index da5a7fb..7f68abc 100644
--- a/net/rds/ib_cm.c
+++ b/net/rds/ib_cm.c
@@ -236,12 +236,10 @@ static void rds_ib_cq_comp_handler_recv(struct ib_cq *cq, void *context)
 	tasklet_schedule(&ic->i_recv_tasklet);
 }
 
-static void poll_cq(struct rds_ib_connection *ic, struct ib_cq *cq,
-		    struct ib_wc *wcs,
-		    struct rds_ib_ack_state *ack_state)
+static void poll_scq(struct rds_ib_connection *ic, struct ib_cq *cq,
+		     struct ib_wc *wcs)
 {
-	int nr;
-	int i;
+	int nr, i;
 	struct ib_wc *wc;
 
 	while ((nr = ib_poll_cq(cq, RDS_IB_WC_MAX, wcs)) > 0) {
@@ -251,10 +249,7 @@ static void poll_cq(struct rds_ib_connection *ic, struct ib_cq *cq,
 				 (unsigned long long)wc->wr_id, wc->status,
 				 wc->byte_len, be32_to_cpu(wc->ex.imm_data));
 
-			if (wc->wr_id & RDS_IB_SEND_OP)
-				rds_ib_send_cqe_handler(ic, wc);
-			else
-				rds_ib_recv_cqe_handler(ic, wc, ack_state);
+			rds_ib_send_cqe_handler(ic, wc);
 		}
 	}
 }
@@ -263,14 +258,12 @@ static void rds_ib_tasklet_fn_send(unsigned long data)
 {
 	struct rds_ib_connection *ic = (struct rds_ib_connection *)data;
 	struct rds_connection *conn = ic->conn;
-	struct rds_ib_ack_state state;
 
 	rds_ib_stats_inc(s_ib_tasklet_call);
 
-	memset(&state, 0, sizeof(state));
-	poll_cq(ic, ic->i_send_cq, ic->i_send_wc, &state);
+	poll_scq(ic, ic->i_send_cq, ic->i_send_wc);
 	ib_req_notify_cq(ic->i_send_cq, IB_CQ_NEXT_COMP);
-	poll_cq(ic, ic->i_send_cq, ic->i_send_wc, &state);
+	poll_scq(ic, ic->i_send_cq, ic->i_send_wc);
 
 	if (rds_conn_up(conn) &&
 	    (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags) ||
@@ -278,6 +271,25 @@ static void rds_ib_tasklet_fn_send(unsigned long data)
 		rds_send_xmit(ic->conn);
 }
 
+static void poll_rcq(struct rds_ib_connection *ic, struct ib_cq *cq,
+		     struct ib_wc *wcs,
+		     struct rds_ib_ack_state *ack_state)
+{
+	int nr, i;
+	struct ib_wc *wc;
+
+	while ((nr = ib_poll_cq(cq, RDS_IB_WC_MAX, wcs)) > 0) {
+		for (i = 0; i < nr; i++) {
+			wc = wcs + i;
+			rdsdebug("wc wr_id 0x%llx status %u byte_len %u imm_data %u\n",
+				 (unsigned long long)wc->wr_id, wc->status,
+				 wc->byte_len, be32_to_cpu(wc->ex.imm_data));
+
+			rds_ib_recv_cqe_handler(ic, wc, ack_state);
+		}
+	}
+}
+
 static void rds_ib_tasklet_fn_recv(unsigned long data)
 {
 	struct rds_ib_connection *ic = (struct rds_ib_connection *)data;
@@ -291,9 +303,9 @@ static void rds_ib_tasklet_fn_recv(unsigned long data)
 	rds_ib_stats_inc(s_ib_tasklet_call);
 
 	memset(&state, 0, sizeof(state));
-	poll_cq(ic, ic->i_recv_cq, ic->i_recv_wc, &state);
+	poll_rcq(ic, ic->i_recv_cq, ic->i_recv_wc, &state);
 	ib_req_notify_cq(ic->i_recv_cq, IB_CQ_SOLICITED);
-	poll_cq(ic, ic->i_recv_cq, ic->i_recv_wc, &state);
+	poll_rcq(ic, ic->i_recv_cq, ic->i_recv_wc, &state);
 
 	if (state.ack_next_valid)
 		rds_ib_set_ack(ic, state.ack_next, state.ack_required);
diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c
index eac30bf..f27d2c8 100644
--- a/net/rds/ib_send.c
+++ b/net/rds/ib_send.c
@@ -195,7 +195,7 @@ void rds_ib_send_init_ring(struct rds_ib_connection *ic)
 
 		send->s_op = NULL;
 
-		send->s_wr.wr_id = i | RDS_IB_SEND_OP;
+		send->s_wr.wr_id = i;
 		send->s_wr.sg_list = send->s_sge;
 		send->s_wr.ex.imm_data = 0;
 
@@ -263,9 +263,7 @@ void rds_ib_send_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc)
 
 	oldest = rds_ib_ring_oldest(&ic->i_send_ring);
 
-	completed = rds_ib_ring_completed(&ic->i_send_ring,
-					  (wc->wr_id & ~RDS_IB_SEND_OP),
-					  oldest);
+	completed = rds_ib_ring_completed(&ic->i_send_ring, wc->wr_id, oldest);
 
 	for (i = 0; i < completed; i++) {
 		send = &ic->i_sends[oldest];
-- 
cgit v0.10.2


From f6df683f32f09f8c0c440bdd3741a1e7dc7927fe Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:46 -0800
Subject: RDS: IB: Re-organise ibmr code

No functional changes. This is in preperation towards adding
fastreg memory resgitration support.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/Makefile b/net/rds/Makefile
index 19e5485..bcf5591 100644
--- a/net/rds/Makefile
+++ b/net/rds/Makefile
@@ -6,7 +6,7 @@ rds-y :=	af_rds.o bind.o cong.o connection.o info.o message.o   \
 obj-$(CONFIG_RDS_RDMA) += rds_rdma.o
 rds_rdma-y :=	rdma_transport.o \
 			ib.o ib_cm.o ib_recv.o ib_ring.o ib_send.o ib_stats.o \
-			ib_sysctl.o ib_rdma.o
+			ib_sysctl.o ib_rdma.o ib_fmr.o
 
 
 obj-$(CONFIG_RDS_TCP) += rds_tcp.o
diff --git a/net/rds/ib.c b/net/rds/ib.c
index 9481d55..bb32cb9 100644
--- a/net/rds/ib.c
+++ b/net/rds/ib.c
@@ -42,15 +42,16 @@
 
 #include "rds.h"
 #include "ib.h"
+#include "ib_mr.h"
 
-unsigned int rds_ib_fmr_1m_pool_size = RDS_FMR_1M_POOL_SIZE;
-unsigned int rds_ib_fmr_8k_pool_size = RDS_FMR_8K_POOL_SIZE;
+unsigned int rds_ib_mr_1m_pool_size = RDS_MR_1M_POOL_SIZE;
+unsigned int rds_ib_mr_8k_pool_size = RDS_MR_8K_POOL_SIZE;
 unsigned int rds_ib_retry_count = RDS_IB_DEFAULT_RETRY_COUNT;
 
-module_param(rds_ib_fmr_1m_pool_size, int, 0444);
-MODULE_PARM_DESC(rds_ib_fmr_1m_pool_size, " Max number of 1M fmr per HCA");
-module_param(rds_ib_fmr_8k_pool_size, int, 0444);
-MODULE_PARM_DESC(rds_ib_fmr_8k_pool_size, " Max number of 8K fmr per HCA");
+module_param(rds_ib_mr_1m_pool_size, int, 0444);
+MODULE_PARM_DESC(rds_ib_mr_1m_pool_size, " Max number of 1M mr per HCA");
+module_param(rds_ib_mr_8k_pool_size, int, 0444);
+MODULE_PARM_DESC(rds_ib_mr_8k_pool_size, " Max number of 8K mr per HCA");
 module_param(rds_ib_retry_count, int, 0444);
 MODULE_PARM_DESC(rds_ib_retry_count, " Number of hw retries before reporting an error");
 
@@ -140,13 +141,13 @@ static void rds_ib_add_one(struct ib_device *device)
 	rds_ibdev->max_sge = min(device->attrs.max_sge, RDS_IB_MAX_SGE);
 
 	rds_ibdev->fmr_max_remaps = device->attrs.max_map_per_fmr?: 32;
-	rds_ibdev->max_1m_fmrs = device->attrs.max_mr ?
+	rds_ibdev->max_1m_mrs = device->attrs.max_mr ?
 		min_t(unsigned int, (device->attrs.max_mr / 2),
-		      rds_ib_fmr_1m_pool_size) : rds_ib_fmr_1m_pool_size;
+		      rds_ib_mr_1m_pool_size) : rds_ib_mr_1m_pool_size;
 
-	rds_ibdev->max_8k_fmrs = device->attrs.max_mr ?
+	rds_ibdev->max_8k_mrs = device->attrs.max_mr ?
 		min_t(unsigned int, ((device->attrs.max_mr / 2) * RDS_MR_8K_SCALE),
-		      rds_ib_fmr_8k_pool_size) : rds_ib_fmr_8k_pool_size;
+		      rds_ib_mr_8k_pool_size) : rds_ib_mr_8k_pool_size;
 
 	rds_ibdev->max_initiator_depth = device->attrs.max_qp_init_rd_atom;
 	rds_ibdev->max_responder_resources = device->attrs.max_qp_rd_atom;
@@ -172,10 +173,10 @@ static void rds_ib_add_one(struct ib_device *device)
 		goto put_dev;
 	}
 
-	rdsdebug("RDS/IB: max_mr = %d, max_wrs = %d, max_sge = %d, fmr_max_remaps = %d, max_1m_fmrs = %d, max_8k_fmrs = %d\n",
+	rdsdebug("RDS/IB: max_mr = %d, max_wrs = %d, max_sge = %d, fmr_max_remaps = %d, max_1m_mrs = %d, max_8k_mrs = %d\n",
 		 device->attrs.max_fmr, rds_ibdev->max_wrs, rds_ibdev->max_sge,
-		 rds_ibdev->fmr_max_remaps, rds_ibdev->max_1m_fmrs,
-		 rds_ibdev->max_8k_fmrs);
+		 rds_ibdev->fmr_max_remaps, rds_ibdev->max_1m_mrs,
+		 rds_ibdev->max_8k_mrs);
 
 	INIT_LIST_HEAD(&rds_ibdev->ipaddr_list);
 	INIT_LIST_HEAD(&rds_ibdev->conn_list);
@@ -364,7 +365,7 @@ void rds_ib_exit(void)
 	rds_ib_sysctl_exit();
 	rds_ib_recv_exit();
 	rds_trans_unregister(&rds_ib_transport);
-	rds_ib_fmr_exit();
+	rds_ib_mr_exit();
 }
 
 struct rds_transport rds_ib_transport = {
@@ -400,13 +401,13 @@ int rds_ib_init(void)
 
 	INIT_LIST_HEAD(&rds_ib_devices);
 
-	ret = rds_ib_fmr_init();
+	ret = rds_ib_mr_init();
 	if (ret)
 		goto out;
 
 	ret = ib_register_client(&rds_ib_client);
 	if (ret)
-		goto out_fmr_exit;
+		goto out_mr_exit;
 
 	ret = rds_ib_sysctl_init();
 	if (ret)
@@ -430,8 +431,8 @@ out_sysctl:
 	rds_ib_sysctl_exit();
 out_ibreg:
 	rds_ib_unregister_client();
-out_fmr_exit:
-	rds_ib_fmr_exit();
+out_mr_exit:
+	rds_ib_mr_exit();
 out:
 	return ret;
 }
diff --git a/net/rds/ib.h b/net/rds/ib.h
index 09cd8e3..c88cb22 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -9,12 +9,6 @@
 #include "rds.h"
 #include "rdma_transport.h"
 
-#define RDS_FMR_1M_POOL_SIZE		(8192 / 2)
-#define RDS_FMR_1M_MSG_SIZE		256
-#define RDS_FMR_8K_MSG_SIZE		2
-#define RDS_MR_8K_SCALE			(256 / (RDS_FMR_8K_MSG_SIZE + 1))
-#define RDS_FMR_8K_POOL_SIZE		(RDS_MR_8K_SCALE * (8192 / 2))
-
 #define RDS_IB_MAX_SGE			8
 #define RDS_IB_RECV_SGE 		2
 
@@ -206,12 +200,12 @@ struct rds_ib_device {
 	struct list_head	conn_list;
 	struct ib_device	*dev;
 	struct ib_pd		*pd;
-	unsigned int		max_fmrs;
+	unsigned int		max_mrs;
 	struct rds_ib_mr_pool	*mr_1m_pool;
 	struct rds_ib_mr_pool   *mr_8k_pool;
 	unsigned int		fmr_max_remaps;
-	unsigned int		max_8k_fmrs;
-	unsigned int		max_1m_fmrs;
+	unsigned int		max_8k_mrs;
+	unsigned int		max_1m_mrs;
 	int			max_sge;
 	unsigned int		max_wrs;
 	unsigned int		max_initiator_depth;
@@ -316,8 +310,6 @@ struct rds_ib_device *rds_ib_get_client_data(struct ib_device *device);
 void rds_ib_dev_put(struct rds_ib_device *rds_ibdev);
 extern struct ib_client rds_ib_client;
 
-extern unsigned int rds_ib_fmr_1m_pool_size;
-extern unsigned int rds_ib_fmr_8k_pool_size;
 extern unsigned int rds_ib_retry_count;
 
 extern spinlock_t ib_nodev_conns_lock;
@@ -347,17 +339,6 @@ int rds_ib_update_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr);
 void rds_ib_add_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn);
 void rds_ib_remove_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn);
 void rds_ib_destroy_nodev_conns(void);
-struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_dev,
-					     int npages);
-void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_connection *iinfo);
-void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *);
-void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
-		    struct rds_sock *rs, u32 *key_ret);
-void rds_ib_sync_mr(void *trans_private, int dir);
-void rds_ib_free_mr(void *trans_private, int invalidate);
-void rds_ib_flush_mrs(void);
-int rds_ib_fmr_init(void);
-void rds_ib_fmr_exit(void);
 
 /* ib_recv.c */
 int rds_ib_recv_init(void);
diff --git a/net/rds/ib_fmr.c b/net/rds/ib_fmr.c
new file mode 100644
index 0000000..d4f200d
--- /dev/null
+++ b/net/rds/ib_fmr.c
@@ -0,0 +1,217 @@
+/*
+ * Copyright (c) 2016 Oracle.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "ib_mr.h"
+
+struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev, int npages)
+{
+	struct rds_ib_mr_pool *pool;
+	struct rds_ib_mr *ibmr = NULL;
+	int err = 0, iter = 0;
+
+	if (npages <= RDS_MR_8K_MSG_SIZE)
+		pool = rds_ibdev->mr_8k_pool;
+	else
+		pool = rds_ibdev->mr_1m_pool;
+
+	if (atomic_read(&pool->dirty_count) >= pool->max_items / 10)
+		queue_delayed_work(rds_ib_mr_wq, &pool->flush_worker, 10);
+
+	/* Switch pools if one of the pool is reaching upper limit */
+	if (atomic_read(&pool->dirty_count) >=  pool->max_items * 9 / 10) {
+		if (pool->pool_type == RDS_IB_MR_8K_POOL)
+			pool = rds_ibdev->mr_1m_pool;
+		else
+			pool = rds_ibdev->mr_8k_pool;
+	}
+
+	while (1) {
+		ibmr = rds_ib_reuse_mr(pool);
+		if (ibmr)
+			return ibmr;
+
+		/* No clean MRs - now we have the choice of either
+		 * allocating a fresh MR up to the limit imposed by the
+		 * driver, or flush any dirty unused MRs.
+		 * We try to avoid stalling in the send path if possible,
+		 * so we allocate as long as we're allowed to.
+		 *
+		 * We're fussy with enforcing the FMR limit, though. If the
+		 * driver tells us we can't use more than N fmrs, we shouldn't
+		 * start arguing with it
+		 */
+		if (atomic_inc_return(&pool->item_count) <= pool->max_items)
+			break;
+
+		atomic_dec(&pool->item_count);
+
+		if (++iter > 2) {
+			if (pool->pool_type == RDS_IB_MR_8K_POOL)
+				rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_depleted);
+			else
+				rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_depleted);
+			return ERR_PTR(-EAGAIN);
+		}
+
+		/* We do have some empty MRs. Flush them out. */
+		if (pool->pool_type == RDS_IB_MR_8K_POOL)
+			rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_wait);
+		else
+			rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_wait);
+		rds_ib_flush_mr_pool(pool, 0, &ibmr);
+		if (ibmr)
+			return ibmr;
+	}
+
+	ibmr = kzalloc_node(sizeof(*ibmr), GFP_KERNEL,
+			    rdsibdev_to_node(rds_ibdev));
+	if (!ibmr) {
+		err = -ENOMEM;
+		goto out_no_cigar;
+	}
+
+	ibmr->fmr = ib_alloc_fmr(rds_ibdev->pd,
+			(IB_ACCESS_LOCAL_WRITE |
+			 IB_ACCESS_REMOTE_READ |
+			 IB_ACCESS_REMOTE_WRITE |
+			 IB_ACCESS_REMOTE_ATOMIC),
+			&pool->fmr_attr);
+	if (IS_ERR(ibmr->fmr)) {
+		err = PTR_ERR(ibmr->fmr);
+		ibmr->fmr = NULL;
+		pr_warn("RDS/IB: %s failed (err=%d)\n", __func__, err);
+		goto out_no_cigar;
+	}
+
+	ibmr->pool = pool;
+	if (pool->pool_type == RDS_IB_MR_8K_POOL)
+		rds_ib_stats_inc(s_ib_rdma_mr_8k_alloc);
+	else
+		rds_ib_stats_inc(s_ib_rdma_mr_1m_alloc);
+
+	return ibmr;
+
+out_no_cigar:
+	if (ibmr) {
+		if (ibmr->fmr)
+			ib_dealloc_fmr(ibmr->fmr);
+		kfree(ibmr);
+	}
+	atomic_dec(&pool->item_count);
+	return ERR_PTR(err);
+}
+
+int rds_ib_map_fmr(struct rds_ib_device *rds_ibdev, struct rds_ib_mr *ibmr,
+		   struct scatterlist *sg, unsigned int nents)
+{
+	struct ib_device *dev = rds_ibdev->dev;
+	struct scatterlist *scat = sg;
+	u64 io_addr = 0;
+	u64 *dma_pages;
+	u32 len;
+	int page_cnt, sg_dma_len;
+	int i, j;
+	int ret;
+
+	sg_dma_len = ib_dma_map_sg(dev, sg, nents, DMA_BIDIRECTIONAL);
+	if (unlikely(!sg_dma_len)) {
+		pr_warn("RDS/IB: %s failed!\n", __func__);
+		return -EBUSY;
+	}
+
+	len = 0;
+	page_cnt = 0;
+
+	for (i = 0; i < sg_dma_len; ++i) {
+		unsigned int dma_len = ib_sg_dma_len(dev, &scat[i]);
+		u64 dma_addr = ib_sg_dma_address(dev, &scat[i]);
+
+		if (dma_addr & ~PAGE_MASK) {
+			if (i > 0)
+				return -EINVAL;
+			else
+				++page_cnt;
+		}
+		if ((dma_addr + dma_len) & ~PAGE_MASK) {
+			if (i < sg_dma_len - 1)
+				return -EINVAL;
+			else
+				++page_cnt;
+		}
+
+		len += dma_len;
+	}
+
+	page_cnt += len >> PAGE_SHIFT;
+	if (page_cnt > ibmr->pool->fmr_attr.max_pages)
+		return -EINVAL;
+
+	dma_pages = kmalloc_node(sizeof(u64) * page_cnt, GFP_ATOMIC,
+				 rdsibdev_to_node(rds_ibdev));
+	if (!dma_pages)
+		return -ENOMEM;
+
+	page_cnt = 0;
+	for (i = 0; i < sg_dma_len; ++i) {
+		unsigned int dma_len = ib_sg_dma_len(dev, &scat[i]);
+		u64 dma_addr = ib_sg_dma_address(dev, &scat[i]);
+
+		for (j = 0; j < dma_len; j += PAGE_SIZE)
+			dma_pages[page_cnt++] =
+				(dma_addr & PAGE_MASK) + j;
+	}
+
+	ret = ib_map_phys_fmr(ibmr->fmr, dma_pages, page_cnt, io_addr);
+	if (ret)
+		goto out;
+
+	/* Success - we successfully remapped the MR, so we can
+	 * safely tear down the old mapping.
+	 */
+	rds_ib_teardown_mr(ibmr);
+
+	ibmr->sg = scat;
+	ibmr->sg_len = nents;
+	ibmr->sg_dma_len = sg_dma_len;
+	ibmr->remap_count++;
+
+	if (ibmr->pool->pool_type == RDS_IB_MR_8K_POOL)
+		rds_ib_stats_inc(s_ib_rdma_mr_8k_used);
+	else
+		rds_ib_stats_inc(s_ib_rdma_mr_1m_used);
+	ret = 0;
+
+out:
+	kfree(dma_pages);
+
+	return ret;
+}
diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
new file mode 100644
index 0000000..d88724f
--- /dev/null
+++ b/net/rds/ib_mr.h
@@ -0,0 +1,109 @@
+/*
+ * Copyright (c) 2016 Oracle.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _RDS_IB_MR_H
+#define _RDS_IB_MR_H
+
+#include <linux/kernel.h>
+
+#include "rds.h"
+#include "ib.h"
+
+#define RDS_MR_1M_POOL_SIZE		(8192 / 2)
+#define RDS_MR_1M_MSG_SIZE		256
+#define RDS_MR_8K_MSG_SIZE		2
+#define RDS_MR_8K_SCALE			(256 / (RDS_MR_8K_MSG_SIZE + 1))
+#define RDS_MR_8K_POOL_SIZE		(RDS_MR_8K_SCALE * (8192 / 2))
+
+/* This is stored as mr->r_trans_private. */
+struct rds_ib_mr {
+	struct rds_ib_device	*device;
+	struct rds_ib_mr_pool	*pool;
+	struct ib_fmr		*fmr;
+
+	struct llist_node	llnode;
+
+	/* unmap_list is for freeing */
+	struct list_head	unmap_list;
+	unsigned int		remap_count;
+
+	struct scatterlist	*sg;
+	unsigned int		sg_len;
+	u64			*dma;
+	int			sg_dma_len;
+};
+
+/* Our own little MR pool */
+struct rds_ib_mr_pool {
+	unsigned int            pool_type;
+	struct mutex		flush_lock;	/* serialize fmr invalidate */
+	struct delayed_work	flush_worker;	/* flush worker */
+
+	atomic_t		item_count;	/* total # of MRs */
+	atomic_t		dirty_count;	/* # dirty of MRs */
+
+	struct llist_head	drop_list;	/* MRs not reached max_maps */
+	struct llist_head	free_list;	/* unused MRs */
+	struct llist_head	clean_list;	/* unused & unmapped MRs */
+	wait_queue_head_t	flush_wait;
+
+	atomic_t		free_pinned;	/* memory pinned by free MRs */
+	unsigned long		max_items;
+	unsigned long		max_items_soft;
+	unsigned long		max_free_pinned;
+	struct ib_fmr_attr	fmr_attr;
+};
+
+extern struct workqueue_struct *rds_ib_mr_wq;
+extern unsigned int rds_ib_mr_1m_pool_size;
+extern unsigned int rds_ib_mr_8k_pool_size;
+
+struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_dev,
+					     int npages);
+void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev,
+			struct rds_info_rdma_connection *iinfo);
+void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *);
+void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
+		    struct rds_sock *rs, u32 *key_ret);
+void rds_ib_sync_mr(void *trans_private, int dir);
+void rds_ib_free_mr(void *trans_private, int invalidate);
+void rds_ib_flush_mrs(void);
+int rds_ib_mr_init(void);
+void rds_ib_mr_exit(void);
+
+void __rds_ib_teardown_mr(struct rds_ib_mr *);
+void rds_ib_teardown_mr(struct rds_ib_mr *);
+struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *, int);
+int rds_ib_map_fmr(struct rds_ib_device *, struct rds_ib_mr *,
+		   struct scatterlist *, unsigned int);
+struct rds_ib_mr *rds_ib_reuse_mr(struct rds_ib_mr_pool *);
+int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *, int, struct rds_ib_mr **);
+#endif
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index a234074..c594519 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -35,78 +35,13 @@
 #include <linux/rculist.h>
 #include <linux/llist.h>
 
-#include "rds.h"
-#include "ib.h"
+#include "ib_mr.h"
+
+struct workqueue_struct *rds_ib_mr_wq;
 
 static DEFINE_PER_CPU(unsigned long, clean_list_grace);
 #define CLEAN_LIST_BUSY_BIT 0
 
-/*
- * This is stored as mr->r_trans_private.
- */
-struct rds_ib_mr {
-	struct rds_ib_device	*device;
-	struct rds_ib_mr_pool	*pool;
-	struct ib_fmr		*fmr;
-
-	struct llist_node	llnode;
-
-	/* unmap_list is for freeing */
-	struct list_head	unmap_list;
-	unsigned int		remap_count;
-
-	struct scatterlist	*sg;
-	unsigned int		sg_len;
-	u64			*dma;
-	int			sg_dma_len;
-};
-
-/*
- * Our own little FMR pool
- */
-struct rds_ib_mr_pool {
-	unsigned int            pool_type;
-	struct mutex		flush_lock;		/* serialize fmr invalidate */
-	struct delayed_work	flush_worker;		/* flush worker */
-
-	atomic_t		item_count;		/* total # of MRs */
-	atomic_t		dirty_count;		/* # dirty of MRs */
-
-	struct llist_head	drop_list;		/* MRs that have reached their max_maps limit */
-	struct llist_head	free_list;		/* unused MRs */
-	struct llist_head	clean_list;		/* global unused & unamapped MRs */
-	wait_queue_head_t	flush_wait;
-
-	atomic_t		free_pinned;		/* memory pinned by free MRs */
-	unsigned long		max_items;
-	unsigned long		max_items_soft;
-	unsigned long		max_free_pinned;
-	struct ib_fmr_attr	fmr_attr;
-};
-
-static struct workqueue_struct *rds_ib_fmr_wq;
-
-int rds_ib_fmr_init(void)
-{
-	rds_ib_fmr_wq = create_workqueue("rds_fmr_flushd");
-	if (!rds_ib_fmr_wq)
-		return -ENOMEM;
-	return 0;
-}
-
-/* By the time this is called all the IB devices should have been torn down and
- * had their pools freed.  As each pool is freed its work struct is waited on,
- * so the pool flushing work queue should be idle by the time we get here.
- */
-void rds_ib_fmr_exit(void)
-{
-	destroy_workqueue(rds_ib_fmr_wq);
-}
-
-static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool, int free_all, struct rds_ib_mr **);
-static void rds_ib_teardown_mr(struct rds_ib_mr *ibmr);
-static void rds_ib_mr_pool_flush_worker(struct work_struct *work);
-
 static struct rds_ib_device *rds_ib_get_device(__be32 ipaddr)
 {
 	struct rds_ib_device *rds_ibdev;
@@ -235,41 +170,6 @@ void rds_ib_destroy_nodev_conns(void)
 		rds_conn_destroy(ic->conn);
 }
 
-struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_ibdev,
-					     int pool_type)
-{
-	struct rds_ib_mr_pool *pool;
-
-	pool = kzalloc(sizeof(*pool), GFP_KERNEL);
-	if (!pool)
-		return ERR_PTR(-ENOMEM);
-
-	pool->pool_type = pool_type;
-	init_llist_head(&pool->free_list);
-	init_llist_head(&pool->drop_list);
-	init_llist_head(&pool->clean_list);
-	mutex_init(&pool->flush_lock);
-	init_waitqueue_head(&pool->flush_wait);
-	INIT_DELAYED_WORK(&pool->flush_worker, rds_ib_mr_pool_flush_worker);
-
-	if (pool_type == RDS_IB_MR_1M_POOL) {
-		/* +1 allows for unaligned MRs */
-		pool->fmr_attr.max_pages = RDS_FMR_1M_MSG_SIZE + 1;
-		pool->max_items = RDS_FMR_1M_POOL_SIZE;
-	} else {
-		/* pool_type == RDS_IB_MR_8K_POOL */
-		pool->fmr_attr.max_pages = RDS_FMR_8K_MSG_SIZE + 1;
-		pool->max_items = RDS_FMR_8K_POOL_SIZE;
-	}
-
-	pool->max_free_pinned = pool->max_items * pool->fmr_attr.max_pages / 4;
-	pool->fmr_attr.max_maps = rds_ibdev->fmr_max_remaps;
-	pool->fmr_attr.page_shift = PAGE_SHIFT;
-	pool->max_items_soft = rds_ibdev->max_fmrs * 3 / 4;
-
-	return pool;
-}
-
 void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_connection *iinfo)
 {
 	struct rds_ib_mr_pool *pool_1m = rds_ibdev->mr_1m_pool;
@@ -278,16 +178,7 @@ void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_co
 	iinfo->rdma_mr_size = pool_1m->fmr_attr.max_pages;
 }
 
-void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *pool)
-{
-	cancel_delayed_work_sync(&pool->flush_worker);
-	rds_ib_flush_mr_pool(pool, 1, NULL);
-	WARN_ON(atomic_read(&pool->item_count));
-	WARN_ON(atomic_read(&pool->free_pinned));
-	kfree(pool);
-}
-
-static inline struct rds_ib_mr *rds_ib_reuse_fmr(struct rds_ib_mr_pool *pool)
+struct rds_ib_mr *rds_ib_reuse_mr(struct rds_ib_mr_pool *pool)
 {
 	struct rds_ib_mr *ibmr = NULL;
 	struct llist_node *ret;
@@ -317,190 +208,6 @@ static inline void wait_clean_list_grace(void)
 	}
 }
 
-static struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev,
-					  int npages)
-{
-	struct rds_ib_mr_pool *pool;
-	struct rds_ib_mr *ibmr = NULL;
-	int err = 0, iter = 0;
-
-	if (npages <= RDS_FMR_8K_MSG_SIZE)
-		pool = rds_ibdev->mr_8k_pool;
-	else
-		pool = rds_ibdev->mr_1m_pool;
-
-	if (atomic_read(&pool->dirty_count) >= pool->max_items / 10)
-		queue_delayed_work(rds_ib_fmr_wq, &pool->flush_worker, 10);
-
-	/* Switch pools if one of the pool is reaching upper limit */
-	if (atomic_read(&pool->dirty_count) >=  pool->max_items * 9 / 10) {
-		if (pool->pool_type == RDS_IB_MR_8K_POOL)
-			pool = rds_ibdev->mr_1m_pool;
-		else
-			pool = rds_ibdev->mr_8k_pool;
-	}
-
-	while (1) {
-		ibmr = rds_ib_reuse_fmr(pool);
-		if (ibmr)
-			return ibmr;
-
-		/* No clean MRs - now we have the choice of either
-		 * allocating a fresh MR up to the limit imposed by the
-		 * driver, or flush any dirty unused MRs.
-		 * We try to avoid stalling in the send path if possible,
-		 * so we allocate as long as we're allowed to.
-		 *
-		 * We're fussy with enforcing the FMR limit, though. If the driver
-		 * tells us we can't use more than N fmrs, we shouldn't start
-		 * arguing with it */
-		if (atomic_inc_return(&pool->item_count) <= pool->max_items)
-			break;
-
-		atomic_dec(&pool->item_count);
-
-		if (++iter > 2) {
-			if (pool->pool_type == RDS_IB_MR_8K_POOL)
-				rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_depleted);
-			else
-				rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_depleted);
-			return ERR_PTR(-EAGAIN);
-		}
-
-		/* We do have some empty MRs. Flush them out. */
-		if (pool->pool_type == RDS_IB_MR_8K_POOL)
-			rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_wait);
-		else
-			rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_wait);
-		rds_ib_flush_mr_pool(pool, 0, &ibmr);
-		if (ibmr)
-			return ibmr;
-	}
-
-	ibmr = kzalloc_node(sizeof(*ibmr), GFP_KERNEL, rdsibdev_to_node(rds_ibdev));
-	if (!ibmr) {
-		err = -ENOMEM;
-		goto out_no_cigar;
-	}
-
-	ibmr->fmr = ib_alloc_fmr(rds_ibdev->pd,
-			(IB_ACCESS_LOCAL_WRITE |
-			 IB_ACCESS_REMOTE_READ |
-			 IB_ACCESS_REMOTE_WRITE|
-			 IB_ACCESS_REMOTE_ATOMIC),
-			&pool->fmr_attr);
-	if (IS_ERR(ibmr->fmr)) {
-		err = PTR_ERR(ibmr->fmr);
-		ibmr->fmr = NULL;
-		printk(KERN_WARNING "RDS/IB: ib_alloc_fmr failed (err=%d)\n", err);
-		goto out_no_cigar;
-	}
-
-	ibmr->pool = pool;
-	if (pool->pool_type == RDS_IB_MR_8K_POOL)
-		rds_ib_stats_inc(s_ib_rdma_mr_8k_alloc);
-	else
-		rds_ib_stats_inc(s_ib_rdma_mr_1m_alloc);
-
-	return ibmr;
-
-out_no_cigar:
-	if (ibmr) {
-		if (ibmr->fmr)
-			ib_dealloc_fmr(ibmr->fmr);
-		kfree(ibmr);
-	}
-	atomic_dec(&pool->item_count);
-	return ERR_PTR(err);
-}
-
-static int rds_ib_map_fmr(struct rds_ib_device *rds_ibdev, struct rds_ib_mr *ibmr,
-	       struct scatterlist *sg, unsigned int nents)
-{
-	struct ib_device *dev = rds_ibdev->dev;
-	struct scatterlist *scat = sg;
-	u64 io_addr = 0;
-	u64 *dma_pages;
-	u32 len;
-	int page_cnt, sg_dma_len;
-	int i, j;
-	int ret;
-
-	sg_dma_len = ib_dma_map_sg(dev, sg, nents,
-				 DMA_BIDIRECTIONAL);
-	if (unlikely(!sg_dma_len)) {
-		printk(KERN_WARNING "RDS/IB: dma_map_sg failed!\n");
-		return -EBUSY;
-	}
-
-	len = 0;
-	page_cnt = 0;
-
-	for (i = 0; i < sg_dma_len; ++i) {
-		unsigned int dma_len = ib_sg_dma_len(dev, &scat[i]);
-		u64 dma_addr = ib_sg_dma_address(dev, &scat[i]);
-
-		if (dma_addr & ~PAGE_MASK) {
-			if (i > 0)
-				return -EINVAL;
-			else
-				++page_cnt;
-		}
-		if ((dma_addr + dma_len) & ~PAGE_MASK) {
-			if (i < sg_dma_len - 1)
-				return -EINVAL;
-			else
-				++page_cnt;
-		}
-
-		len += dma_len;
-	}
-
-	page_cnt += len >> PAGE_SHIFT;
-	if (page_cnt > ibmr->pool->fmr_attr.max_pages)
-		return -EINVAL;
-
-	dma_pages = kmalloc_node(sizeof(u64) * page_cnt, GFP_ATOMIC,
-				 rdsibdev_to_node(rds_ibdev));
-	if (!dma_pages)
-		return -ENOMEM;
-
-	page_cnt = 0;
-	for (i = 0; i < sg_dma_len; ++i) {
-		unsigned int dma_len = ib_sg_dma_len(dev, &scat[i]);
-		u64 dma_addr = ib_sg_dma_address(dev, &scat[i]);
-
-		for (j = 0; j < dma_len; j += PAGE_SIZE)
-			dma_pages[page_cnt++] =
-				(dma_addr & PAGE_MASK) + j;
-	}
-
-	ret = ib_map_phys_fmr(ibmr->fmr,
-				   dma_pages, page_cnt, io_addr);
-	if (ret)
-		goto out;
-
-	/* Success - we successfully remapped the MR, so we can
-	 * safely tear down the old mapping. */
-	rds_ib_teardown_mr(ibmr);
-
-	ibmr->sg = scat;
-	ibmr->sg_len = nents;
-	ibmr->sg_dma_len = sg_dma_len;
-	ibmr->remap_count++;
-
-	if (ibmr->pool->pool_type == RDS_IB_MR_8K_POOL)
-		rds_ib_stats_inc(s_ib_rdma_mr_8k_used);
-	else
-		rds_ib_stats_inc(s_ib_rdma_mr_1m_used);
-	ret = 0;
-
-out:
-	kfree(dma_pages);
-
-	return ret;
-}
-
 void rds_ib_sync_mr(void *trans_private, int direction)
 {
 	struct rds_ib_mr *ibmr = trans_private;
@@ -518,7 +225,7 @@ void rds_ib_sync_mr(void *trans_private, int direction)
 	}
 }
 
-static void __rds_ib_teardown_mr(struct rds_ib_mr *ibmr)
+void __rds_ib_teardown_mr(struct rds_ib_mr *ibmr)
 {
 	struct rds_ib_device *rds_ibdev = ibmr->device;
 
@@ -549,7 +256,7 @@ static void __rds_ib_teardown_mr(struct rds_ib_mr *ibmr)
 	}
 }
 
-static void rds_ib_teardown_mr(struct rds_ib_mr *ibmr)
+void rds_ib_teardown_mr(struct rds_ib_mr *ibmr)
 {
 	unsigned int pinned = ibmr->sg_len;
 
@@ -623,8 +330,8 @@ static void list_to_llist_nodes(struct rds_ib_mr_pool *pool,
  * If the number of MRs allocated exceeds the limit, we also try
  * to free as many MRs as needed to get back to this limit.
  */
-static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
-				int free_all, struct rds_ib_mr **ibmr_ret)
+int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
+			 int free_all, struct rds_ib_mr **ibmr_ret)
 {
 	struct rds_ib_mr *ibmr, *next;
 	struct llist_node *clean_nodes;
@@ -643,7 +350,7 @@ static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 	if (ibmr_ret) {
 		DEFINE_WAIT(wait);
 		while (!mutex_trylock(&pool->flush_lock)) {
-			ibmr = rds_ib_reuse_fmr(pool);
+			ibmr = rds_ib_reuse_mr(pool);
 			if (ibmr) {
 				*ibmr_ret = ibmr;
 				finish_wait(&pool->flush_wait, &wait);
@@ -655,7 +362,7 @@ static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 			if (llist_empty(&pool->clean_list))
 				schedule();
 
-			ibmr = rds_ib_reuse_fmr(pool);
+			ibmr = rds_ib_reuse_mr(pool);
 			if (ibmr) {
 				*ibmr_ret = ibmr;
 				finish_wait(&pool->flush_wait, &wait);
@@ -667,7 +374,7 @@ static int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 		mutex_lock(&pool->flush_lock);
 
 	if (ibmr_ret) {
-		ibmr = rds_ib_reuse_fmr(pool);
+		ibmr = rds_ib_reuse_mr(pool);
 		if (ibmr) {
 			*ibmr_ret = ibmr;
 			goto out;
@@ -773,7 +480,7 @@ void rds_ib_free_mr(void *trans_private, int invalidate)
 	/* If we've pinned too many pages, request a flush */
 	if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned ||
 	    atomic_read(&pool->dirty_count) >= pool->max_items / 5)
-		queue_delayed_work(rds_ib_fmr_wq, &pool->flush_worker, 10);
+		queue_delayed_work(rds_ib_mr_wq, &pool->flush_worker, 10);
 
 	if (invalidate) {
 		if (likely(!in_interrupt())) {
@@ -782,7 +489,7 @@ void rds_ib_free_mr(void *trans_private, int invalidate)
 			/* We get here if the user created a MR marked
 			 * as use_once and invalidate at the same time.
 			 */
-			queue_delayed_work(rds_ib_fmr_wq,
+			queue_delayed_work(rds_ib_mr_wq,
 					   &pool->flush_worker, 10);
 		}
 	}
@@ -849,3 +556,63 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 	return ibmr;
 }
 
+void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *pool)
+{
+	cancel_delayed_work_sync(&pool->flush_worker);
+	rds_ib_flush_mr_pool(pool, 1, NULL);
+	WARN_ON(atomic_read(&pool->item_count));
+	WARN_ON(atomic_read(&pool->free_pinned));
+	kfree(pool);
+}
+
+struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_ibdev,
+					     int pool_type)
+{
+	struct rds_ib_mr_pool *pool;
+
+	pool = kzalloc(sizeof(*pool), GFP_KERNEL);
+	if (!pool)
+		return ERR_PTR(-ENOMEM);
+
+	pool->pool_type = pool_type;
+	init_llist_head(&pool->free_list);
+	init_llist_head(&pool->drop_list);
+	init_llist_head(&pool->clean_list);
+	mutex_init(&pool->flush_lock);
+	init_waitqueue_head(&pool->flush_wait);
+	INIT_DELAYED_WORK(&pool->flush_worker, rds_ib_mr_pool_flush_worker);
+
+	if (pool_type == RDS_IB_MR_1M_POOL) {
+		/* +1 allows for unaligned MRs */
+		pool->fmr_attr.max_pages = RDS_MR_1M_MSG_SIZE + 1;
+		pool->max_items = RDS_MR_1M_POOL_SIZE;
+	} else {
+		/* pool_type == RDS_IB_MR_8K_POOL */
+		pool->fmr_attr.max_pages = RDS_MR_8K_MSG_SIZE + 1;
+		pool->max_items = RDS_MR_8K_POOL_SIZE;
+	}
+
+	pool->max_free_pinned = pool->max_items * pool->fmr_attr.max_pages / 4;
+	pool->fmr_attr.max_maps = rds_ibdev->fmr_max_remaps;
+	pool->fmr_attr.page_shift = PAGE_SHIFT;
+	pool->max_items_soft = rds_ibdev->max_mrs * 3 / 4;
+
+	return pool;
+}
+
+int rds_ib_mr_init(void)
+{
+	rds_ib_mr_wq = create_workqueue("rds_mr_flushd");
+	if (!rds_ib_mr_wq)
+		return -ENOMEM;
+	return 0;
+}
+
+/* By the time this is called all the IB devices should have been torn down and
+ * had their pools freed.  As each pool is freed its work struct is waited on,
+ * so the pool flushing work queue should be idle by the time we get here.
+ */
+void rds_ib_mr_exit(void)
+{
+	destroy_workqueue(rds_ib_mr_wq);
+}
-- 
cgit v0.10.2


From a69365a39cd35d0ad11be7d4a896b2f76a8fde44 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:47 -0800
Subject: RDS: IB: create struct rds_ib_fmr

Keep fmr related filed in its own struct. Fastreg MR structure
will be added to the union.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib_fmr.c b/net/rds/ib_fmr.c
index d4f200d..74f2c21 100644
--- a/net/rds/ib_fmr.c
+++ b/net/rds/ib_fmr.c
@@ -36,6 +36,7 @@ struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev, int npages)
 {
 	struct rds_ib_mr_pool *pool;
 	struct rds_ib_mr *ibmr = NULL;
+	struct rds_ib_fmr *fmr;
 	int err = 0, iter = 0;
 
 	if (npages <= RDS_MR_8K_MSG_SIZE)
@@ -99,15 +100,16 @@ struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev, int npages)
 		goto out_no_cigar;
 	}
 
-	ibmr->fmr = ib_alloc_fmr(rds_ibdev->pd,
+	fmr = &ibmr->u.fmr;
+	fmr->fmr = ib_alloc_fmr(rds_ibdev->pd,
 			(IB_ACCESS_LOCAL_WRITE |
 			 IB_ACCESS_REMOTE_READ |
 			 IB_ACCESS_REMOTE_WRITE |
 			 IB_ACCESS_REMOTE_ATOMIC),
 			&pool->fmr_attr);
-	if (IS_ERR(ibmr->fmr)) {
-		err = PTR_ERR(ibmr->fmr);
-		ibmr->fmr = NULL;
+	if (IS_ERR(fmr->fmr)) {
+		err = PTR_ERR(fmr->fmr);
+		fmr->fmr = NULL;
 		pr_warn("RDS/IB: %s failed (err=%d)\n", __func__, err);
 		goto out_no_cigar;
 	}
@@ -122,8 +124,8 @@ struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev, int npages)
 
 out_no_cigar:
 	if (ibmr) {
-		if (ibmr->fmr)
-			ib_dealloc_fmr(ibmr->fmr);
+		if (fmr->fmr)
+			ib_dealloc_fmr(fmr->fmr);
 		kfree(ibmr);
 	}
 	atomic_dec(&pool->item_count);
@@ -134,6 +136,7 @@ int rds_ib_map_fmr(struct rds_ib_device *rds_ibdev, struct rds_ib_mr *ibmr,
 		   struct scatterlist *sg, unsigned int nents)
 {
 	struct ib_device *dev = rds_ibdev->dev;
+	struct rds_ib_fmr *fmr = &ibmr->u.fmr;
 	struct scatterlist *scat = sg;
 	u64 io_addr = 0;
 	u64 *dma_pages;
@@ -190,7 +193,7 @@ int rds_ib_map_fmr(struct rds_ib_device *rds_ibdev, struct rds_ib_mr *ibmr,
 				(dma_addr & PAGE_MASK) + j;
 	}
 
-	ret = ib_map_phys_fmr(ibmr->fmr, dma_pages, page_cnt, io_addr);
+	ret = ib_map_phys_fmr(fmr->fmr, dma_pages, page_cnt, io_addr);
 	if (ret)
 		goto out;
 
diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
index d88724f..309ad59 100644
--- a/net/rds/ib_mr.h
+++ b/net/rds/ib_mr.h
@@ -43,11 +43,15 @@
 #define RDS_MR_8K_SCALE			(256 / (RDS_MR_8K_MSG_SIZE + 1))
 #define RDS_MR_8K_POOL_SIZE		(RDS_MR_8K_SCALE * (8192 / 2))
 
+struct rds_ib_fmr {
+	struct ib_fmr		*fmr;
+	u64			*dma;
+};
+
 /* This is stored as mr->r_trans_private. */
 struct rds_ib_mr {
 	struct rds_ib_device	*device;
 	struct rds_ib_mr_pool	*pool;
-	struct ib_fmr		*fmr;
 
 	struct llist_node	llnode;
 
@@ -57,8 +61,11 @@ struct rds_ib_mr {
 
 	struct scatterlist	*sg;
 	unsigned int		sg_len;
-	u64			*dma;
 	int			sg_dma_len;
+
+	union {
+		struct rds_ib_fmr	fmr;
+	} u;
 };
 
 /* Our own little MR pool */
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index c594519..9e608d9 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -334,6 +334,7 @@ int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 			 int free_all, struct rds_ib_mr **ibmr_ret)
 {
 	struct rds_ib_mr *ibmr, *next;
+	struct rds_ib_fmr *fmr;
 	struct llist_node *clean_nodes;
 	struct llist_node *clean_tail;
 	LIST_HEAD(unmap_list);
@@ -395,8 +396,10 @@ int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 		goto out;
 
 	/* String all ib_mr's onto one list and hand them to ib_unmap_fmr */
-	list_for_each_entry(ibmr, &unmap_list, unmap_list)
-		list_add(&ibmr->fmr->list, &fmr_list);
+	list_for_each_entry(ibmr, &unmap_list, unmap_list) {
+		fmr = &ibmr->u.fmr;
+		list_add(&fmr->fmr->list, &fmr_list);
+	}
 
 	ret = ib_unmap_fmr(&fmr_list);
 	if (ret)
@@ -405,6 +408,7 @@ int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 	/* Now we can destroy the DMA mapping and unpin any pages */
 	list_for_each_entry_safe(ibmr, next, &unmap_list, unmap_list) {
 		unpinned += ibmr->sg_len;
+		fmr = &ibmr->u.fmr;
 		__rds_ib_teardown_mr(ibmr);
 		if (nfreed < free_goal ||
 		    ibmr->remap_count >= pool->fmr_attr.max_maps) {
@@ -413,7 +417,7 @@ int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 			else
 				rds_ib_stats_inc(s_ib_rdma_mr_1m_free);
 			list_del(&ibmr->unmap_list);
-			ib_dealloc_fmr(ibmr->fmr);
+			ib_dealloc_fmr(fmr->fmr);
 			kfree(ibmr);
 			nfreed++;
 		}
@@ -517,6 +521,7 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 {
 	struct rds_ib_device *rds_ibdev;
 	struct rds_ib_mr *ibmr = NULL;
+	struct rds_ib_fmr *fmr;
 	int ret;
 
 	rds_ibdev = rds_ib_get_device(rs->rs_bound_addr);
@@ -536,9 +541,10 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		return ibmr;
 	}
 
+	fmr = &ibmr->u.fmr;
 	ret = rds_ib_map_fmr(rds_ibdev, ibmr, sg, nents);
 	if (ret == 0)
-		*key_ret = ibmr->fmr->rkey;
+		*key_ret = fmr->fmr->rkey;
 	else
 		printk(KERN_WARNING "RDS/IB: map_fmr failed (errno=%d)\n", ret);
 
-- 
cgit v0.10.2


From 490ea5967b0d94b84747d3327903d31f287f30e5 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:48 -0800
Subject: RDS: IB: move FMR code to its own file

No functional change.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib_fmr.c b/net/rds/ib_fmr.c
index 74f2c21..4fe8f4f 100644
--- a/net/rds/ib_fmr.c
+++ b/net/rds/ib_fmr.c
@@ -37,61 +37,16 @@ struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev, int npages)
 	struct rds_ib_mr_pool *pool;
 	struct rds_ib_mr *ibmr = NULL;
 	struct rds_ib_fmr *fmr;
-	int err = 0, iter = 0;
+	int err = 0;
 
 	if (npages <= RDS_MR_8K_MSG_SIZE)
 		pool = rds_ibdev->mr_8k_pool;
 	else
 		pool = rds_ibdev->mr_1m_pool;
 
-	if (atomic_read(&pool->dirty_count) >= pool->max_items / 10)
-		queue_delayed_work(rds_ib_mr_wq, &pool->flush_worker, 10);
-
-	/* Switch pools if one of the pool is reaching upper limit */
-	if (atomic_read(&pool->dirty_count) >=  pool->max_items * 9 / 10) {
-		if (pool->pool_type == RDS_IB_MR_8K_POOL)
-			pool = rds_ibdev->mr_1m_pool;
-		else
-			pool = rds_ibdev->mr_8k_pool;
-	}
-
-	while (1) {
-		ibmr = rds_ib_reuse_mr(pool);
-		if (ibmr)
-			return ibmr;
-
-		/* No clean MRs - now we have the choice of either
-		 * allocating a fresh MR up to the limit imposed by the
-		 * driver, or flush any dirty unused MRs.
-		 * We try to avoid stalling in the send path if possible,
-		 * so we allocate as long as we're allowed to.
-		 *
-		 * We're fussy with enforcing the FMR limit, though. If the
-		 * driver tells us we can't use more than N fmrs, we shouldn't
-		 * start arguing with it
-		 */
-		if (atomic_inc_return(&pool->item_count) <= pool->max_items)
-			break;
-
-		atomic_dec(&pool->item_count);
-
-		if (++iter > 2) {
-			if (pool->pool_type == RDS_IB_MR_8K_POOL)
-				rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_depleted);
-			else
-				rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_depleted);
-			return ERR_PTR(-EAGAIN);
-		}
-
-		/* We do have some empty MRs. Flush them out. */
-		if (pool->pool_type == RDS_IB_MR_8K_POOL)
-			rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_wait);
-		else
-			rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_wait);
-		rds_ib_flush_mr_pool(pool, 0, &ibmr);
-		if (ibmr)
-			return ibmr;
-	}
+	ibmr = rds_ib_try_reuse_ibmr(pool);
+	if (ibmr)
+		return ibmr;
 
 	ibmr = kzalloc_node(sizeof(*ibmr), GFP_KERNEL,
 			    rdsibdev_to_node(rds_ibdev));
@@ -218,3 +173,76 @@ out:
 
 	return ret;
 }
+
+struct rds_ib_mr *rds_ib_reg_fmr(struct rds_ib_device *rds_ibdev,
+				 struct scatterlist *sg,
+				 unsigned long nents,
+				 u32 *key)
+{
+	struct rds_ib_mr *ibmr = NULL;
+	struct rds_ib_fmr *fmr;
+	int ret;
+
+	ibmr = rds_ib_alloc_fmr(rds_ibdev, nents);
+	if (IS_ERR(ibmr))
+		return ibmr;
+
+	ibmr->device = rds_ibdev;
+	fmr = &ibmr->u.fmr;
+	ret = rds_ib_map_fmr(rds_ibdev, ibmr, sg, nents);
+	if (ret == 0)
+		*key = fmr->fmr->rkey;
+	else
+		rds_ib_free_mr(ibmr, 0);
+
+	return ibmr;
+}
+
+void rds_ib_unreg_fmr(struct list_head *list, unsigned int *nfreed,
+		      unsigned long *unpinned, unsigned int goal)
+{
+	struct rds_ib_mr *ibmr, *next;
+	struct rds_ib_fmr *fmr;
+	LIST_HEAD(fmr_list);
+	int ret = 0;
+	unsigned int freed = *nfreed;
+
+	/* String all ib_mr's onto one list and hand them to  ib_unmap_fmr */
+	list_for_each_entry(ibmr, list, unmap_list) {
+		fmr = &ibmr->u.fmr;
+		list_add(&fmr->fmr->list, &fmr_list);
+	}
+
+	ret = ib_unmap_fmr(&fmr_list);
+	if (ret)
+		pr_warn("RDS/IB: FMR invalidation failed (err=%d)\n", ret);
+
+	/* Now we can destroy the DMA mapping and unpin any pages */
+	list_for_each_entry_safe(ibmr, next, list, unmap_list) {
+		fmr = &ibmr->u.fmr;
+		*unpinned += ibmr->sg_len;
+		__rds_ib_teardown_mr(ibmr);
+		if (freed < goal ||
+		    ibmr->remap_count >= ibmr->pool->fmr_attr.max_maps) {
+			if (ibmr->pool->pool_type == RDS_IB_MR_8K_POOL)
+				rds_ib_stats_inc(s_ib_rdma_mr_8k_free);
+			else
+				rds_ib_stats_inc(s_ib_rdma_mr_1m_free);
+			list_del(&ibmr->unmap_list);
+			ib_dealloc_fmr(fmr->fmr);
+			kfree(ibmr);
+			freed++;
+		}
+	}
+	*nfreed = freed;
+}
+
+void rds_ib_free_fmr_list(struct rds_ib_mr *ibmr)
+{
+	struct rds_ib_mr_pool *pool = ibmr->pool;
+
+	if (ibmr->remap_count >= pool->fmr_attr.max_maps)
+		llist_add(&ibmr->llnode, &pool->drop_list);
+	else
+		llist_add(&ibmr->llnode, &pool->free_list);
+}
diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
index 309ad59..f5c1fcb 100644
--- a/net/rds/ib_mr.h
+++ b/net/rds/ib_mr.h
@@ -113,4 +113,10 @@ int rds_ib_map_fmr(struct rds_ib_device *, struct rds_ib_mr *,
 		   struct scatterlist *, unsigned int);
 struct rds_ib_mr *rds_ib_reuse_mr(struct rds_ib_mr_pool *);
 int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *, int, struct rds_ib_mr **);
+struct rds_ib_mr *rds_ib_reg_fmr(struct rds_ib_device *, struct scatterlist *,
+				 unsigned long, u32 *);
+struct rds_ib_mr *rds_ib_try_reuse_ibmr(struct rds_ib_mr_pool *);
+void rds_ib_unreg_fmr(struct list_head *, unsigned int *,
+		      unsigned long *, unsigned int);
+void rds_ib_free_fmr_list(struct rds_ib_mr *);
 #endif
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index 9e608d9..0e84843 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -333,15 +333,12 @@ static void list_to_llist_nodes(struct rds_ib_mr_pool *pool,
 int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 			 int free_all, struct rds_ib_mr **ibmr_ret)
 {
-	struct rds_ib_mr *ibmr, *next;
-	struct rds_ib_fmr *fmr;
+	struct rds_ib_mr *ibmr;
 	struct llist_node *clean_nodes;
 	struct llist_node *clean_tail;
 	LIST_HEAD(unmap_list);
-	LIST_HEAD(fmr_list);
 	unsigned long unpinned = 0;
 	unsigned int nfreed = 0, dirty_to_clean = 0, free_goal;
-	int ret = 0;
 
 	if (pool->pool_type == RDS_IB_MR_8K_POOL)
 		rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_flush);
@@ -395,33 +392,7 @@ int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 	if (list_empty(&unmap_list))
 		goto out;
 
-	/* String all ib_mr's onto one list and hand them to ib_unmap_fmr */
-	list_for_each_entry(ibmr, &unmap_list, unmap_list) {
-		fmr = &ibmr->u.fmr;
-		list_add(&fmr->fmr->list, &fmr_list);
-	}
-
-	ret = ib_unmap_fmr(&fmr_list);
-	if (ret)
-		printk(KERN_WARNING "RDS/IB: ib_unmap_fmr failed (err=%d)\n", ret);
-
-	/* Now we can destroy the DMA mapping and unpin any pages */
-	list_for_each_entry_safe(ibmr, next, &unmap_list, unmap_list) {
-		unpinned += ibmr->sg_len;
-		fmr = &ibmr->u.fmr;
-		__rds_ib_teardown_mr(ibmr);
-		if (nfreed < free_goal ||
-		    ibmr->remap_count >= pool->fmr_attr.max_maps) {
-			if (ibmr->pool->pool_type == RDS_IB_MR_8K_POOL)
-				rds_ib_stats_inc(s_ib_rdma_mr_8k_free);
-			else
-				rds_ib_stats_inc(s_ib_rdma_mr_1m_free);
-			list_del(&ibmr->unmap_list);
-			ib_dealloc_fmr(fmr->fmr);
-			kfree(ibmr);
-			nfreed++;
-		}
-	}
+	rds_ib_unreg_fmr(&unmap_list, &nfreed, &unpinned, free_goal);
 
 	if (!list_empty(&unmap_list)) {
 		/* we have to make sure that none of the things we're about
@@ -454,7 +425,47 @@ out:
 	if (waitqueue_active(&pool->flush_wait))
 		wake_up(&pool->flush_wait);
 out_nolock:
-	return ret;
+	return 0;
+}
+
+struct rds_ib_mr *rds_ib_try_reuse_ibmr(struct rds_ib_mr_pool *pool)
+{
+	struct rds_ib_mr *ibmr = NULL;
+	int iter = 0;
+
+	if (atomic_read(&pool->dirty_count) >= pool->max_items_soft / 10)
+		queue_delayed_work(rds_ib_mr_wq, &pool->flush_worker, 10);
+
+	while (1) {
+		ibmr = rds_ib_reuse_mr(pool);
+		if (ibmr)
+			return ibmr;
+
+		if (atomic_inc_return(&pool->item_count) <= pool->max_items)
+			break;
+
+		atomic_dec(&pool->item_count);
+
+		if (++iter > 2) {
+			if (pool->pool_type == RDS_IB_MR_8K_POOL)
+				rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_depleted);
+			else
+				rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_depleted);
+			return ERR_PTR(-EAGAIN);
+		}
+
+		/* We do have some empty MRs. Flush them out. */
+		if (pool->pool_type == RDS_IB_MR_8K_POOL)
+			rds_ib_stats_inc(s_ib_rdma_mr_8k_pool_wait);
+		else
+			rds_ib_stats_inc(s_ib_rdma_mr_1m_pool_wait);
+
+		rds_ib_flush_mr_pool(pool, 0, &ibmr);
+		if (ibmr)
+			return ibmr;
+	}
+
+	return ibmr;
 }
 
 static void rds_ib_mr_pool_flush_worker(struct work_struct *work)
@@ -473,10 +484,7 @@ void rds_ib_free_mr(void *trans_private, int invalidate)
 	rdsdebug("RDS/IB: free_mr nents %u\n", ibmr->sg_len);
 
 	/* Return it to the pool's free list */
-	if (ibmr->remap_count >= pool->fmr_attr.max_maps)
-		llist_add(&ibmr->llnode, &pool->drop_list);
-	else
-		llist_add(&ibmr->llnode, &pool->free_list);
+	rds_ib_free_fmr_list(ibmr);
 
 	atomic_add(ibmr->sg_len, &pool->free_pinned);
 	atomic_inc(&pool->dirty_count);
@@ -521,7 +529,6 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 {
 	struct rds_ib_device *rds_ibdev;
 	struct rds_ib_mr *ibmr = NULL;
-	struct rds_ib_fmr *fmr;
 	int ret;
 
 	rds_ibdev = rds_ib_get_device(rs->rs_bound_addr);
@@ -535,30 +542,17 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		goto out;
 	}
 
-	ibmr = rds_ib_alloc_fmr(rds_ibdev, nents);
-	if (IS_ERR(ibmr)) {
-		rds_ib_dev_put(rds_ibdev);
-		return ibmr;
-	}
-
-	fmr = &ibmr->u.fmr;
-	ret = rds_ib_map_fmr(rds_ibdev, ibmr, sg, nents);
-	if (ret == 0)
-		*key_ret = fmr->fmr->rkey;
-	else
-		printk(KERN_WARNING "RDS/IB: map_fmr failed (errno=%d)\n", ret);
-
-	ibmr->device = rds_ibdev;
-	rds_ibdev = NULL;
+	ibmr = rds_ib_reg_fmr(rds_ibdev, sg, nents, key_ret);
+	if (ibmr)
+		rds_ibdev = NULL;
 
  out:
-	if (ret) {
-		if (ibmr)
-			rds_ib_free_mr(ibmr, 0);
-		ibmr = ERR_PTR(ret);
-	}
+	if (!ibmr)
+		pr_warn("RDS/IB: rds_ib_get_mr failed (errno=%d)\n", ret);
+
 	if (rds_ibdev)
 		rds_ib_dev_put(rds_ibdev);
+
 	return ibmr;
 }
 
-- 
cgit v0.10.2


From d4de76da5c39582aa14a94186c35b4aec6ef292c Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:49 -0800
Subject: RDS: IB: add connection info to ibmr

Preperatory patch for FRMR support. From connection info,
we can retrieve cm_id which contains qp handled needed for
work request posting.

We also need to drop the RDS connection on QP error states
where connection handle becomes useful.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
index f5c1fcb..add7725 100644
--- a/net/rds/ib_mr.h
+++ b/net/rds/ib_mr.h
@@ -50,18 +50,19 @@ struct rds_ib_fmr {
 
 /* This is stored as mr->r_trans_private. */
 struct rds_ib_mr {
-	struct rds_ib_device	*device;
-	struct rds_ib_mr_pool	*pool;
+	struct rds_ib_device		*device;
+	struct rds_ib_mr_pool		*pool;
+	struct rds_ib_connection	*ic;
 
-	struct llist_node	llnode;
+	struct llist_node		llnode;
 
 	/* unmap_list is for freeing */
-	struct list_head	unmap_list;
-	unsigned int		remap_count;
+	struct list_head		unmap_list;
+	unsigned int			remap_count;
 
-	struct scatterlist	*sg;
-	unsigned int		sg_len;
-	int			sg_dma_len;
+	struct scatterlist		*sg;
+	unsigned int			sg_len;
+	int				sg_dma_len;
 
 	union {
 		struct rds_ib_fmr	fmr;
-- 
cgit v0.10.2


From 37ea401e9ccc44453da6785514bb64ad6e9c181f Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:50 -0800
Subject: RDS: IB: handle the RDMA CM time wait event

Drop the RDS connection on RDMA_CM_EVENT_TIMEWAIT_EXIT so that
it can reconnect and resume.

While testing fastreg, this error happened in couple of tests but
was getting un-noticed.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c
index 4f4b3d8..7220beb 100644
--- a/net/rds/rdma_transport.c
+++ b/net/rds/rdma_transport.c
@@ -117,6 +117,14 @@ int rds_rdma_cm_event_handler(struct rdma_cm_id *cm_id,
 		rds_conn_drop(conn);
 		break;
 
+	case RDMA_CM_EVENT_TIMEWAIT_EXIT:
+		if (conn) {
+			pr_info("RDS: RDMA_CM_EVENT_TIMEWAIT_EXIT event: dropping connection %pI4->%pI4\n",
+				&conn->c_laddr, &conn->c_faddr);
+			rds_conn_drop(conn);
+		}
+		break;
+
 	default:
 		/* things like device disconnect? */
 		printk(KERN_ERR "RDS: unknown event %u (%s)!\n",
-- 
cgit v0.10.2


From db42753adb638b63572583162bb08ea193947309 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:51 -0800
Subject: RDS: IB: add mr reused stats

Add MR reuse statistics to RDS IB transport.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib.h b/net/rds/ib.h
index c88cb22..62fe7d5 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -259,6 +259,8 @@ struct rds_ib_statistics {
 	uint64_t	s_ib_rdma_mr_1m_pool_flush;
 	uint64_t	s_ib_rdma_mr_1m_pool_wait;
 	uint64_t	s_ib_rdma_mr_1m_pool_depleted;
+	uint64_t	s_ib_rdma_mr_8k_reused;
+	uint64_t	s_ib_rdma_mr_1m_reused;
 	uint64_t	s_ib_atomic_cswp;
 	uint64_t	s_ib_atomic_fadd;
 };
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index 0e84843..ec7ea32 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -188,8 +188,13 @@ struct rds_ib_mr *rds_ib_reuse_mr(struct rds_ib_mr_pool *pool)
 	flag = this_cpu_ptr(&clean_list_grace);
 	set_bit(CLEAN_LIST_BUSY_BIT, flag);
 	ret = llist_del_first(&pool->clean_list);
-	if (ret)
+	if (ret) {
 		ibmr = llist_entry(ret, struct rds_ib_mr, llnode);
+		if (pool->pool_type == RDS_IB_MR_8K_POOL)
+			rds_ib_stats_inc(s_ib_rdma_mr_8k_reused);
+		else
+			rds_ib_stats_inc(s_ib_rdma_mr_1m_reused);
+	}
 
 	clear_bit(CLEAN_LIST_BUSY_BIT, flag);
 	preempt_enable();
diff --git a/net/rds/ib_stats.c b/net/rds/ib_stats.c
index d77e044..7e78dca 100644
--- a/net/rds/ib_stats.c
+++ b/net/rds/ib_stats.c
@@ -73,6 +73,8 @@ static const char *const rds_ib_stat_names[] = {
 	"ib_rdma_mr_1m_pool_flush",
 	"ib_rdma_mr_1m_pool_wait",
 	"ib_rdma_mr_1m_pool_depleted",
+	"ib_rdma_mr_8k_reused",
+	"ib_rdma_mr_1m_reused",
 	"ib_atomic_cswp",
 	"ib_atomic_fadd",
 };
-- 
cgit v0.10.2


From 2cb2912d65633e751d3f8397377174501412aa47 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:52 -0800
Subject: RDS: IB: add Fastreg MR (FRMR) detection support

Discovere Fast Memmory Registration support using IB device
IB_DEVICE_MEM_MGT_EXTENSIONS. Certain HCA might support just FRMR
or FMR or both FMR and FRWR. In case both mr type are supported,
default FMR is used.

Default MR is still kept as FMR against what everyone else
is following. Default will be changed to FRMR once the
RDS performance with FRMR is comparable with FMR. The
work is in progress for the same.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib.c b/net/rds/ib.c
index bb32cb9..b5342fd 100644
--- a/net/rds/ib.c
+++ b/net/rds/ib.c
@@ -140,6 +140,12 @@ static void rds_ib_add_one(struct ib_device *device)
 	rds_ibdev->max_wrs = device->attrs.max_qp_wr;
 	rds_ibdev->max_sge = min(device->attrs.max_sge, RDS_IB_MAX_SGE);
 
+	rds_ibdev->has_fr = (device->attrs.device_cap_flags &
+				  IB_DEVICE_MEM_MGT_EXTENSIONS);
+	rds_ibdev->has_fmr = (device->alloc_fmr && device->dealloc_fmr &&
+			    device->map_phys_fmr && device->unmap_fmr);
+	rds_ibdev->use_fastreg = (rds_ibdev->has_fr && !rds_ibdev->has_fmr);
+
 	rds_ibdev->fmr_max_remaps = device->attrs.max_map_per_fmr?: 32;
 	rds_ibdev->max_1m_mrs = device->attrs.max_mr ?
 		min_t(unsigned int, (device->attrs.max_mr / 2),
@@ -178,6 +184,10 @@ static void rds_ib_add_one(struct ib_device *device)
 		 rds_ibdev->fmr_max_remaps, rds_ibdev->max_1m_mrs,
 		 rds_ibdev->max_8k_mrs);
 
+	pr_info("RDS/IB: %s: %s supported and preferred\n",
+		device->name,
+		rds_ibdev->use_fastreg ? "FRMR" : "FMR");
+
 	INIT_LIST_HEAD(&rds_ibdev->ipaddr_list);
 	INIT_LIST_HEAD(&rds_ibdev->conn_list);
 
diff --git a/net/rds/ib.h b/net/rds/ib.h
index 62fe7d5..c5eddc2 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -200,6 +200,10 @@ struct rds_ib_device {
 	struct list_head	conn_list;
 	struct ib_device	*dev;
 	struct ib_pd		*pd;
+	bool                    has_fmr;
+	bool                    has_fr;
+	bool                    use_fastreg;
+
 	unsigned int		max_mrs;
 	struct rds_ib_mr_pool	*mr_1m_pool;
 	struct rds_ib_mr_pool   *mr_8k_pool;
diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
index add7725..2f9b9c3 100644
--- a/net/rds/ib_mr.h
+++ b/net/rds/ib_mr.h
@@ -93,6 +93,7 @@ struct rds_ib_mr_pool {
 extern struct workqueue_struct *rds_ib_mr_wq;
 extern unsigned int rds_ib_mr_1m_pool_size;
 extern unsigned int rds_ib_mr_8k_pool_size;
+extern bool prefer_frmr;
 
 struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_dev,
 					     int npages);
-- 
cgit v0.10.2


From ad6832f950d35df8c70b577993a24b31b34d88e4 Mon Sep 17 00:00:00 2001
From: "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com>
Date: Tue, 1 Mar 2016 15:20:53 -0800
Subject: RDS: IB: allocate extra space on queues for FRMR support

Fastreg MR(FRMR) memory registration and invalidation makes use
of work request and completion queues for its operation. Patch
allocates extra queue space towards these operation(s).

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib.h b/net/rds/ib.h
index c5eddc2..eeb0d6c 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -14,6 +14,7 @@
 
 #define RDS_IB_DEFAULT_RECV_WR		1024
 #define RDS_IB_DEFAULT_SEND_WR		256
+#define RDS_IB_DEFAULT_FR_WR		512
 
 #define RDS_IB_DEFAULT_RETRY_COUNT	2
 
@@ -122,6 +123,9 @@ struct rds_ib_connection {
 	struct ib_wc		i_send_wc[RDS_IB_WC_MAX];
 	struct ib_wc		i_recv_wc[RDS_IB_WC_MAX];
 
+	/* To control the number of wrs from fastreg */
+	atomic_t		i_fastreg_wrs;
+
 	/* interrupt handling */
 	struct tasklet_struct	i_send_tasklet;
 	struct tasklet_struct	i_recv_tasklet;
diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c
index 7f68abc..83f4673 100644
--- a/net/rds/ib_cm.c
+++ b/net/rds/ib_cm.c
@@ -363,7 +363,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
 	struct ib_qp_init_attr attr;
 	struct ib_cq_init_attr cq_attr = {};
 	struct rds_ib_device *rds_ibdev;
-	int ret;
+	int ret, fr_queue_space;
 
 	/*
 	 * It's normal to see a null device if an incoming connection races
@@ -373,6 +373,12 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
 	if (!rds_ibdev)
 		return -EOPNOTSUPP;
 
+	/* The fr_queue_space is currently set to 512, to add extra space on
+	 * completion queue and send queue. This extra space is used for FRMR
+	 * registration and invalidation work requests
+	 */
+	fr_queue_space = (rds_ibdev->use_fastreg ? RDS_IB_DEFAULT_FR_WR : 0);
+
 	/* add the conn now so that connection establishment has the dev */
 	rds_ib_add_conn(rds_ibdev, conn);
 
@@ -384,7 +390,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
 	/* Protection domain and memory range */
 	ic->i_pd = rds_ibdev->pd;
 
-	cq_attr.cqe = ic->i_send_ring.w_nr + 1;
+	cq_attr.cqe = ic->i_send_ring.w_nr + fr_queue_space + 1;
 
 	ic->i_send_cq = ib_create_cq(dev, rds_ib_cq_comp_handler_send,
 				     rds_ib_cq_event_handler, conn,
@@ -424,7 +430,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
 	attr.event_handler = rds_ib_qp_event_handler;
 	attr.qp_context = conn;
 	/* + 1 to allow for the single ack message */
-	attr.cap.max_send_wr = ic->i_send_ring.w_nr + 1;
+	attr.cap.max_send_wr = ic->i_send_ring.w_nr + fr_queue_space + 1;
 	attr.cap.max_recv_wr = ic->i_recv_ring.w_nr + 1;
 	attr.cap.max_send_sge = rds_ibdev->max_sge;
 	attr.cap.max_recv_sge = RDS_IB_RECV_SGE;
@@ -432,6 +438,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
 	attr.qp_type = IB_QPT_RC;
 	attr.send_cq = ic->i_send_cq;
 	attr.recv_cq = ic->i_recv_cq;
+	atomic_set(&ic->i_fastreg_wrs, RDS_IB_DEFAULT_FR_WR);
 
 	/*
 	 * XXX this can fail if max_*_wr is too large?  Are we supposed
@@ -751,7 +758,8 @@ void rds_ib_conn_shutdown(struct rds_connection *conn)
 		 */
 		wait_event(rds_ib_ring_empty_wait,
 			   rds_ib_ring_empty(&ic->i_recv_ring) &&
-			   (atomic_read(&ic->i_signaled_sends) == 0));
+			   (atomic_read(&ic->i_signaled_sends) == 0) &&
+			   (atomic_read(&ic->i_fastreg_wrs) == RDS_IB_DEFAULT_FR_WR));
 		tasklet_kill(&ic->i_send_tasklet);
 		tasklet_kill(&ic->i_recv_tasklet);
 
-- 
cgit v0.10.2


From 1659185fb4d0025835eb2058a141f0746c5cab00 Mon Sep 17 00:00:00 2001
From: Avinash Repaka <avinash.repaka@oracle.com>
Date: Tue, 1 Mar 2016 15:20:54 -0800
Subject: RDS: IB: Support Fastreg MR (FRMR) memory registration mode

Fastreg MR(FRMR) is another method with which one can
register memory to HCA. Some of the newer HCAs supports only fastreg
mr mode, so we need to add support for it to have RDS functional
on them.

Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Avinash Repaka <avinash.repaka@oracle.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/Makefile b/net/rds/Makefile
index bcf5591..0e72bec 100644
--- a/net/rds/Makefile
+++ b/net/rds/Makefile
@@ -6,7 +6,7 @@ rds-y :=	af_rds.o bind.o cong.o connection.o info.o message.o   \
 obj-$(CONFIG_RDS_RDMA) += rds_rdma.o
 rds_rdma-y :=	rdma_transport.o \
 			ib.o ib_cm.o ib_recv.o ib_ring.o ib_send.o ib_stats.o \
-			ib_sysctl.o ib_rdma.o ib_fmr.o
+			ib_sysctl.o ib_rdma.o ib_fmr.o ib_frmr.o
 
 
 obj-$(CONFIG_RDS_TCP) += rds_tcp.o
diff --git a/net/rds/ib.h b/net/rds/ib.h
index eeb0d6c..627fb79 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -349,6 +349,7 @@ int rds_ib_update_ipaddr(struct rds_ib_device *rds_ibdev, __be32 ipaddr);
 void rds_ib_add_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn);
 void rds_ib_remove_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn);
 void rds_ib_destroy_nodev_conns(void);
+void rds_ib_mr_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc);
 
 /* ib_recv.c */
 int rds_ib_recv_init(void);
diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c
index 83f4673..8764970 100644
--- a/net/rds/ib_cm.c
+++ b/net/rds/ib_cm.c
@@ -249,7 +249,12 @@ static void poll_scq(struct rds_ib_connection *ic, struct ib_cq *cq,
 				 (unsigned long long)wc->wr_id, wc->status,
 				 wc->byte_len, be32_to_cpu(wc->ex.imm_data));
 
-			rds_ib_send_cqe_handler(ic, wc);
+			if (wc->wr_id <= ic->i_send_ring.w_nr ||
+			    wc->wr_id == RDS_IB_ACK_WR_ID)
+				rds_ib_send_cqe_handler(ic, wc);
+			else
+				rds_ib_mr_cqe_handler(ic, wc);
+
 		}
 	}
 }
diff --git a/net/rds/ib_frmr.c b/net/rds/ib_frmr.c
new file mode 100644
index 0000000..93ff038
--- /dev/null
+++ b/net/rds/ib_frmr.c
@@ -0,0 +1,376 @@
+/*
+ * Copyright (c) 2016 Oracle.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "ib_mr.h"
+
+static struct rds_ib_mr *rds_ib_alloc_frmr(struct rds_ib_device *rds_ibdev,
+					   int npages)
+{
+	struct rds_ib_mr_pool *pool;
+	struct rds_ib_mr *ibmr = NULL;
+	struct rds_ib_frmr *frmr;
+	int err = 0;
+
+	if (npages <= RDS_MR_8K_MSG_SIZE)
+		pool = rds_ibdev->mr_8k_pool;
+	else
+		pool = rds_ibdev->mr_1m_pool;
+
+	ibmr = rds_ib_try_reuse_ibmr(pool);
+	if (ibmr)
+		return ibmr;
+
+	ibmr = kzalloc_node(sizeof(*ibmr), GFP_KERNEL,
+			    rdsibdev_to_node(rds_ibdev));
+	if (!ibmr) {
+		err = -ENOMEM;
+		goto out_no_cigar;
+	}
+
+	frmr = &ibmr->u.frmr;
+	frmr->mr = ib_alloc_mr(rds_ibdev->pd, IB_MR_TYPE_MEM_REG,
+			 pool->fmr_attr.max_pages);
+	if (IS_ERR(frmr->mr)) {
+		pr_warn("RDS/IB: %s failed to allocate MR", __func__);
+		goto out_no_cigar;
+	}
+
+	ibmr->pool = pool;
+	if (pool->pool_type == RDS_IB_MR_8K_POOL)
+		rds_ib_stats_inc(s_ib_rdma_mr_8k_alloc);
+	else
+		rds_ib_stats_inc(s_ib_rdma_mr_1m_alloc);
+
+	if (atomic_read(&pool->item_count) > pool->max_items_soft)
+		pool->max_items_soft = pool->max_items;
+
+	frmr->fr_state = FRMR_IS_FREE;
+	return ibmr;
+
+out_no_cigar:
+	kfree(ibmr);
+	atomic_dec(&pool->item_count);
+	return ERR_PTR(err);
+}
+
+static void rds_ib_free_frmr(struct rds_ib_mr *ibmr, bool drop)
+{
+	struct rds_ib_mr_pool *pool = ibmr->pool;
+
+	if (drop)
+		llist_add(&ibmr->llnode, &pool->drop_list);
+	else
+		llist_add(&ibmr->llnode, &pool->free_list);
+	atomic_add(ibmr->sg_len, &pool->free_pinned);
+	atomic_inc(&pool->dirty_count);
+
+	/* If we've pinned too many pages, request a flush */
+	if (atomic_read(&pool->free_pinned) >= pool->max_free_pinned ||
+	    atomic_read(&pool->dirty_count) >= pool->max_items / 5)
+		queue_delayed_work(rds_ib_mr_wq, &pool->flush_worker, 10);
+}
+
+static int rds_ib_post_reg_frmr(struct rds_ib_mr *ibmr)
+{
+	struct rds_ib_frmr *frmr = &ibmr->u.frmr;
+	struct ib_send_wr *failed_wr;
+	struct ib_reg_wr reg_wr;
+	int ret;
+
+	while (atomic_dec_return(&ibmr->ic->i_fastreg_wrs) <= 0) {
+		atomic_inc(&ibmr->ic->i_fastreg_wrs);
+		cpu_relax();
+	}
+
+	ret = ib_map_mr_sg_zbva(frmr->mr, ibmr->sg, ibmr->sg_len, PAGE_SIZE);
+	if (unlikely(ret != ibmr->sg_len))
+		return ret < 0 ? ret : -EINVAL;
+
+	/* Perform a WR for the fast_reg_mr. Each individual page
+	 * in the sg list is added to the fast reg page list and placed
+	 * inside the fast_reg_mr WR.  The key used is a rolling 8bit
+	 * counter, which should guarantee uniqueness.
+	 */
+	ib_update_fast_reg_key(frmr->mr, ibmr->remap_count++);
+	frmr->fr_state = FRMR_IS_INUSE;
+
+	memset(&reg_wr, 0, sizeof(reg_wr));
+	reg_wr.wr.wr_id = (unsigned long)(void *)ibmr;
+	reg_wr.wr.opcode = IB_WR_REG_MR;
+	reg_wr.wr.num_sge = 0;
+	reg_wr.mr = frmr->mr;
+	reg_wr.key = frmr->mr->rkey;
+	reg_wr.access = IB_ACCESS_LOCAL_WRITE |
+			IB_ACCESS_REMOTE_READ |
+			IB_ACCESS_REMOTE_WRITE;
+	reg_wr.wr.send_flags = IB_SEND_SIGNALED;
+
+	failed_wr = &reg_wr.wr;
+	ret = ib_post_send(ibmr->ic->i_cm_id->qp, &reg_wr.wr, &failed_wr);
+	WARN_ON(failed_wr != &reg_wr.wr);
+	if (unlikely(ret)) {
+		/* Failure here can be because of -ENOMEM as well */
+		frmr->fr_state = FRMR_IS_STALE;
+		atomic_inc(&ibmr->ic->i_fastreg_wrs);
+		if (printk_ratelimit())
+			pr_warn("RDS/IB: %s returned error(%d)\n",
+				__func__, ret);
+	}
+	return ret;
+}
+
+static int rds_ib_map_frmr(struct rds_ib_device *rds_ibdev,
+			   struct rds_ib_mr_pool *pool,
+			   struct rds_ib_mr *ibmr,
+			   struct scatterlist *sg, unsigned int sg_len)
+{
+	struct ib_device *dev = rds_ibdev->dev;
+	struct rds_ib_frmr *frmr = &ibmr->u.frmr;
+	int i;
+	u32 len;
+	int ret = 0;
+
+	/* We want to teardown old ibmr values here and fill it up with
+	 * new sg values
+	 */
+	rds_ib_teardown_mr(ibmr);
+
+	ibmr->sg = sg;
+	ibmr->sg_len = sg_len;
+	ibmr->sg_dma_len = 0;
+	frmr->sg_byte_len = 0;
+	WARN_ON(ibmr->sg_dma_len);
+	ibmr->sg_dma_len = ib_dma_map_sg(dev, ibmr->sg, ibmr->sg_len,
+					 DMA_BIDIRECTIONAL);
+	if (unlikely(!ibmr->sg_dma_len)) {
+		pr_warn("RDS/IB: %s failed!\n", __func__);
+		return -EBUSY;
+	}
+
+	frmr->sg_byte_len = 0;
+	frmr->dma_npages = 0;
+	len = 0;
+
+	ret = -EINVAL;
+	for (i = 0; i < ibmr->sg_dma_len; ++i) {
+		unsigned int dma_len = ib_sg_dma_len(dev, &ibmr->sg[i]);
+		u64 dma_addr = ib_sg_dma_address(dev, &ibmr->sg[i]);
+
+		frmr->sg_byte_len += dma_len;
+		if (dma_addr & ~PAGE_MASK) {
+			if (i > 0)
+				goto out_unmap;
+			else
+				++frmr->dma_npages;
+		}
+
+		if ((dma_addr + dma_len) & ~PAGE_MASK) {
+			if (i < ibmr->sg_dma_len - 1)
+				goto out_unmap;
+			else
+				++frmr->dma_npages;
+		}
+
+		len += dma_len;
+	}
+	frmr->dma_npages += len >> PAGE_SHIFT;
+
+	if (frmr->dma_npages > ibmr->pool->fmr_attr.max_pages) {
+		ret = -EMSGSIZE;
+		goto out_unmap;
+	}
+
+	ret = rds_ib_post_reg_frmr(ibmr);
+	if (ret)
+		goto out_unmap;
+
+	if (ibmr->pool->pool_type == RDS_IB_MR_8K_POOL)
+		rds_ib_stats_inc(s_ib_rdma_mr_8k_used);
+	else
+		rds_ib_stats_inc(s_ib_rdma_mr_1m_used);
+
+	return ret;
+
+out_unmap:
+	ib_dma_unmap_sg(rds_ibdev->dev, ibmr->sg, ibmr->sg_len,
+			DMA_BIDIRECTIONAL);
+	ibmr->sg_dma_len = 0;
+	return ret;
+}
+
+static int rds_ib_post_inv(struct rds_ib_mr *ibmr)
+{
+	struct ib_send_wr *s_wr, *failed_wr;
+	struct rds_ib_frmr *frmr = &ibmr->u.frmr;
+	struct rdma_cm_id *i_cm_id = ibmr->ic->i_cm_id;
+	int ret = -EINVAL;
+
+	if (!i_cm_id || !i_cm_id->qp || !frmr->mr)
+		goto out;
+
+	if (frmr->fr_state != FRMR_IS_INUSE)
+		goto out;
+
+	while (atomic_dec_return(&ibmr->ic->i_fastreg_wrs) <= 0) {
+		atomic_inc(&ibmr->ic->i_fastreg_wrs);
+		cpu_relax();
+	}
+
+	frmr->fr_inv = true;
+	s_wr = &frmr->fr_wr;
+
+	memset(s_wr, 0, sizeof(*s_wr));
+	s_wr->wr_id = (unsigned long)(void *)ibmr;
+	s_wr->opcode = IB_WR_LOCAL_INV;
+	s_wr->ex.invalidate_rkey = frmr->mr->rkey;
+	s_wr->send_flags = IB_SEND_SIGNALED;
+
+	failed_wr = s_wr;
+	ret = ib_post_send(i_cm_id->qp, s_wr, &failed_wr);
+	WARN_ON(failed_wr != s_wr);
+	if (unlikely(ret)) {
+		frmr->fr_state = FRMR_IS_STALE;
+		frmr->fr_inv = false;
+		atomic_inc(&ibmr->ic->i_fastreg_wrs);
+		pr_err("RDS/IB: %s returned error(%d)\n", __func__, ret);
+		goto out;
+	}
+out:
+	return ret;
+}
+
+void rds_ib_mr_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc)
+{
+	struct rds_ib_mr *ibmr = (void *)(unsigned long)wc->wr_id;
+	struct rds_ib_frmr *frmr = &ibmr->u.frmr;
+
+	if (wc->status != IB_WC_SUCCESS) {
+		frmr->fr_state = FRMR_IS_STALE;
+		if (rds_conn_up(ic->conn))
+			rds_ib_conn_error(ic->conn,
+					  "frmr completion <%pI4,%pI4> status %u(%s), vendor_err 0x%x, disconnecting and reconnecting\n",
+					  &ic->conn->c_laddr,
+					  &ic->conn->c_faddr,
+					  wc->status,
+					  ib_wc_status_msg(wc->status),
+					  wc->vendor_err);
+	}
+
+	if (frmr->fr_inv) {
+		frmr->fr_state = FRMR_IS_FREE;
+		frmr->fr_inv = false;
+	}
+
+	atomic_inc(&ic->i_fastreg_wrs);
+}
+
+void rds_ib_unreg_frmr(struct list_head *list, unsigned int *nfreed,
+		       unsigned long *unpinned, unsigned int goal)
+{
+	struct rds_ib_mr *ibmr, *next;
+	struct rds_ib_frmr *frmr;
+	int ret = 0;
+	unsigned int freed = *nfreed;
+
+	/* String all ib_mr's onto one list and hand them to ib_unmap_fmr */
+	list_for_each_entry(ibmr, list, unmap_list) {
+		if (ibmr->sg_dma_len)
+			ret |= rds_ib_post_inv(ibmr);
+	}
+	if (ret)
+		pr_warn("RDS/IB: %s failed (err=%d)\n", __func__, ret);
+
+	/* Now we can destroy the DMA mapping and unpin any pages */
+	list_for_each_entry_safe(ibmr, next, list, unmap_list) {
+		*unpinned += ibmr->sg_len;
+		frmr = &ibmr->u.frmr;
+		__rds_ib_teardown_mr(ibmr);
+		if (freed < goal || frmr->fr_state == FRMR_IS_STALE) {
+			/* Don't de-allocate if the MR is not free yet */
+			if (frmr->fr_state == FRMR_IS_INUSE)
+				continue;
+
+			if (ibmr->pool->pool_type == RDS_IB_MR_8K_POOL)
+				rds_ib_stats_inc(s_ib_rdma_mr_8k_free);
+			else
+				rds_ib_stats_inc(s_ib_rdma_mr_1m_free);
+			list_del(&ibmr->unmap_list);
+			if (frmr->mr)
+				ib_dereg_mr(frmr->mr);
+			kfree(ibmr);
+			freed++;
+		}
+	}
+	*nfreed = freed;
+}
+
+struct rds_ib_mr *rds_ib_reg_frmr(struct rds_ib_device *rds_ibdev,
+				  struct rds_ib_connection *ic,
+				  struct scatterlist *sg,
+				  unsigned long nents, u32 *key)
+{
+	struct rds_ib_mr *ibmr = NULL;
+	struct rds_ib_frmr *frmr;
+	int ret;
+
+	do {
+		if (ibmr)
+			rds_ib_free_frmr(ibmr, true);
+		ibmr = rds_ib_alloc_frmr(rds_ibdev, nents);
+		if (IS_ERR(ibmr))
+			return ibmr;
+		frmr = &ibmr->u.frmr;
+	} while (frmr->fr_state != FRMR_IS_FREE);
+
+	ibmr->ic = ic;
+	ibmr->device = rds_ibdev;
+	ret = rds_ib_map_frmr(rds_ibdev, ibmr->pool, ibmr, sg, nents);
+	if (ret == 0) {
+		*key = frmr->mr->rkey;
+	} else {
+		rds_ib_free_frmr(ibmr, false);
+		ibmr = ERR_PTR(ret);
+	}
+
+	return ibmr;
+}
+
+void rds_ib_free_frmr_list(struct rds_ib_mr *ibmr)
+{
+	struct rds_ib_mr_pool *pool = ibmr->pool;
+	struct rds_ib_frmr *frmr = &ibmr->u.frmr;
+
+	if (frmr->fr_state == FRMR_IS_STALE)
+		llist_add(&ibmr->llnode, &pool->drop_list);
+	else
+		llist_add(&ibmr->llnode, &pool->free_list);
+}
diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h
index 2f9b9c3..1c754f4 100644
--- a/net/rds/ib_mr.h
+++ b/net/rds/ib_mr.h
@@ -48,6 +48,21 @@ struct rds_ib_fmr {
 	u64			*dma;
 };
 
+enum rds_ib_fr_state {
+	FRMR_IS_FREE,	/* mr invalidated & ready for use */
+	FRMR_IS_INUSE,	/* mr is in use or used & can be invalidated */
+	FRMR_IS_STALE,	/* Stale MR and needs to be dropped  */
+};
+
+struct rds_ib_frmr {
+	struct ib_mr		*mr;
+	enum rds_ib_fr_state	fr_state;
+	bool			fr_inv;
+	struct ib_send_wr	fr_wr;
+	unsigned int		dma_npages;
+	unsigned int		sg_byte_len;
+};
+
 /* This is stored as mr->r_trans_private. */
 struct rds_ib_mr {
 	struct rds_ib_device		*device;
@@ -66,6 +81,7 @@ struct rds_ib_mr {
 
 	union {
 		struct rds_ib_fmr	fmr;
+		struct rds_ib_frmr	frmr;
 	} u;
 };
 
@@ -88,6 +104,7 @@ struct rds_ib_mr_pool {
 	unsigned long		max_items_soft;
 	unsigned long		max_free_pinned;
 	struct ib_fmr_attr	fmr_attr;
+	bool			use_fastreg;
 };
 
 extern struct workqueue_struct *rds_ib_mr_wq;
@@ -121,4 +138,11 @@ struct rds_ib_mr *rds_ib_try_reuse_ibmr(struct rds_ib_mr_pool *);
 void rds_ib_unreg_fmr(struct list_head *, unsigned int *,
 		      unsigned long *, unsigned int);
 void rds_ib_free_fmr_list(struct rds_ib_mr *);
+struct rds_ib_mr *rds_ib_reg_frmr(struct rds_ib_device *rds_ibdev,
+				  struct rds_ib_connection *ic,
+				  struct scatterlist *sg,
+				  unsigned long nents, u32 *key);
+void rds_ib_unreg_frmr(struct list_head *list, unsigned int *nfreed,
+		       unsigned long *unpinned, unsigned int goal);
+void rds_ib_free_frmr_list(struct rds_ib_mr *);
 #endif
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index ec7ea32..f7164ac 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -397,7 +397,10 @@ int rds_ib_flush_mr_pool(struct rds_ib_mr_pool *pool,
 	if (list_empty(&unmap_list))
 		goto out;
 
-	rds_ib_unreg_fmr(&unmap_list, &nfreed, &unpinned, free_goal);
+	if (pool->use_fastreg)
+		rds_ib_unreg_frmr(&unmap_list, &nfreed, &unpinned, free_goal);
+	else
+		rds_ib_unreg_fmr(&unmap_list, &nfreed, &unpinned, free_goal);
 
 	if (!list_empty(&unmap_list)) {
 		/* we have to make sure that none of the things we're about
@@ -489,7 +492,10 @@ void rds_ib_free_mr(void *trans_private, int invalidate)
 	rdsdebug("RDS/IB: free_mr nents %u\n", ibmr->sg_len);
 
 	/* Return it to the pool's free list */
-	rds_ib_free_fmr_list(ibmr);
+	if (rds_ibdev->use_fastreg)
+		rds_ib_free_frmr_list(ibmr);
+	else
+		rds_ib_free_fmr_list(ibmr);
 
 	atomic_add(ibmr->sg_len, &pool->free_pinned);
 	atomic_inc(&pool->dirty_count);
@@ -534,6 +540,7 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 {
 	struct rds_ib_device *rds_ibdev;
 	struct rds_ib_mr *ibmr = NULL;
+	struct rds_ib_connection *ic = rs->rs_conn->c_transport_data;
 	int ret;
 
 	rds_ibdev = rds_ib_get_device(rs->rs_bound_addr);
@@ -547,7 +554,10 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		goto out;
 	}
 
-	ibmr = rds_ib_reg_fmr(rds_ibdev, sg, nents, key_ret);
+	if (rds_ibdev->use_fastreg)
+		ibmr = rds_ib_reg_frmr(rds_ibdev, ic, sg, nents, key_ret);
+	else
+		ibmr = rds_ib_reg_fmr(rds_ibdev, sg, nents, key_ret);
 	if (ibmr)
 		rds_ibdev = NULL;
 
@@ -601,6 +611,7 @@ struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_ibdev,
 	pool->fmr_attr.max_maps = rds_ibdev->fmr_max_remaps;
 	pool->fmr_attr.page_shift = PAGE_SHIFT;
 	pool->max_items_soft = rds_ibdev->max_mrs * 3 / 4;
+	pool->use_fastreg = rds_ibdev->use_fastreg;
 
 	return pool;
 }
-- 
cgit v0.10.2


From 495db27302a88fcabb105c3d30a617beda913945 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:27 +0100
Subject: stmmac: share reset function between dwmac100 and dwmac1000

This patch is to share the same reset procedure between dwmac100 and
dwmac1000 chips.
This will also help on enhancing the driver and support new chips.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 1e19c8f..bac0e44 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -376,7 +376,8 @@ extern const struct stmmac_desc_ops ndesc_ops;
 /* Specific DMA helpers */
 struct stmmac_dma_ops {
 	/* DMA core initialization */
-	int (*init) (void __iomem *ioaddr, int pbl, int fb, int mb,
+	int (*reset)(void __iomem *ioaddr);
+	void (*init)(void __iomem *ioaddr, int pbl, int fb, int mb,
 		     int burst_len, u32 dma_tx, u32 dma_rx, int atds);
 	/* Dump DMA registers */
 	void (*dump_regs) (void __iomem *ioaddr);
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100.h b/drivers/net/ethernet/stmicro/stmmac/dwmac100.h
index 2ec6aea..1657acf 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100.h
@@ -95,7 +95,6 @@
 #define DMA_BUS_MODE_DSL_MASK	0x0000007c	/* Descriptor Skip Length */
 #define DMA_BUS_MODE_DSL_SHIFT	2	/*   (in DWORDS)      */
 #define DMA_BUS_MODE_BAR_BUS	0x00000002	/* Bar-Bus Arbitration */
-#define DMA_BUS_MODE_SFT_RESET	0x00000001	/* Software Reset */
 #define DMA_BUS_MODE_DEFAULT	0x00000000
 
 /* DMA Control register defines */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
index 8831a05..9d36ae7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
@@ -221,7 +221,6 @@ enum inter_frame_gap {
 
 /*--- DMA BLOCK defines ---*/
 /* DMA Bus Mode register defines */
-#define DMA_BUS_MODE_SFT_RESET	0x00000001	/* Software Reset */
 #define DMA_BUS_MODE_DA		0x00000002	/* Arbitration scheme */
 #define DMA_BUS_MODE_DSL_MASK	0x0000007c	/* Descriptor Skip Length */
 #define DMA_BUS_MODE_DSL_SHIFT	2		/*   (in DWORDS)      */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index 0e8937c..5f0aea5 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -30,23 +30,10 @@
 #include "dwmac1000.h"
 #include "dwmac_dma.h"
 
-static int dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
-			      int burst_len, u32 dma_tx, u32 dma_rx, int atds)
+static void dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
+			       int burst_len, u32 dma_tx, u32 dma_rx, int atds)
 {
-	u32 value = readl(ioaddr + DMA_BUS_MODE);
-	int limit;
-
-	/* DMA SW reset */
-	value |= DMA_BUS_MODE_SFT_RESET;
-	writel(value, ioaddr + DMA_BUS_MODE);
-	limit = 10;
-	while (limit--) {
-		if (!(readl(ioaddr + DMA_BUS_MODE) & DMA_BUS_MODE_SFT_RESET))
-			break;
-		mdelay(10);
-	}
-	if (limit < 0)
-		return -EBUSY;
+	u32 value;
 
 	/*
 	 * Set the DMA PBL (Programmable Burst Length) mode
@@ -102,8 +89,6 @@ static int dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
 	 */
 	writel(dma_tx, ioaddr + DMA_TX_BASE_ADDR);
 	writel(dma_rx, ioaddr + DMA_RCV_BASE_ADDR);
-
-	return 0;
 }
 
 static u32 dwmac1000_configure_fc(u32 csr6, int rxfifosz)
@@ -205,6 +190,7 @@ static void dwmac1000_rx_watchdog(void __iomem *ioaddr, u32 riwt)
 }
 
 const struct stmmac_dma_ops dwmac1000_dma_ops = {
+	.reset = dwmac_dma_reset,
 	.init = dwmac1000_dma_init,
 	.dump_regs = dwmac1000_dump_dma_regs,
 	.dma_mode = dwmac1000_dma_operation_mode,
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index 9d0971c..c40582a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -32,24 +32,9 @@
 #include "dwmac100.h"
 #include "dwmac_dma.h"
 
-static int dwmac100_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
-			     int burst_len, u32 dma_tx, u32 dma_rx, int atds)
+static void dwmac100_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
+			      int burst_len, u32 dma_tx, u32 dma_rx, int atds)
 {
-	u32 value = readl(ioaddr + DMA_BUS_MODE);
-	int limit;
-
-	/* DMA SW reset */
-	value |= DMA_BUS_MODE_SFT_RESET;
-	writel(value, ioaddr + DMA_BUS_MODE);
-	limit = 10;
-	while (limit--) {
-		if (!(readl(ioaddr + DMA_BUS_MODE) & DMA_BUS_MODE_SFT_RESET))
-			break;
-		mdelay(10);
-	}
-	if (limit < 0)
-		return -EBUSY;
-
 	/* Enable Application Access by writing to DMA CSR0 */
 	writel(DMA_BUS_MODE_DEFAULT | (pbl << DMA_BUS_MODE_PBL_SHIFT),
 	       ioaddr + DMA_BUS_MODE);
@@ -62,8 +47,6 @@ static int dwmac100_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
 	 */
 	writel(dma_tx, ioaddr + DMA_TX_BASE_ADDR);
 	writel(dma_rx, ioaddr + DMA_RCV_BASE_ADDR);
-
-	return 0;
 }
 
 /* Store and Forward capability is not used at all.
@@ -131,6 +114,7 @@ static void dwmac100_dma_diagnostic_fr(void *data, struct stmmac_extra_stats *x,
 }
 
 const struct stmmac_dma_ops dwmac100_dma_ops = {
+	.reset = dwmac_dma_reset,
 	.init = dwmac100_dma_init,
 	.dump_regs = dwmac100_dump_dma_regs,
 	.dma_mode = dwmac100_dma_operation_mode,
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
index def266d..13ca90e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
@@ -35,6 +35,10 @@
 #define DMA_CONTROL		0x00001018	/* Ctrl (Operational Mode) */
 #define DMA_INTR_ENA		0x0000101c	/* Interrupt Enable */
 #define DMA_MISSED_FRAME_CTR	0x00001020	/* Missed Frame Counter */
+
+/* SW Reset */
+#define DMA_BUS_MODE_SFT_RESET	0x00000001	/* Software Reset */
+
 /* Rx watchdog register */
 #define DMA_RX_WATCHDOG		0x00001024
 /* AXI Bus Mode */
@@ -112,5 +116,6 @@ void dwmac_dma_stop_tx(void __iomem *ioaddr);
 void dwmac_dma_start_rx(void __iomem *ioaddr);
 void dwmac_dma_stop_rx(void __iomem *ioaddr);
 int dwmac_dma_interrupt(void __iomem *ioaddr, struct stmmac_extra_stats *x);
+int dwmac_dma_reset(void __iomem *ioaddr);
 
 #endif /* __DWMAC_DMA_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
index 484e3cf..84e3e84 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
@@ -26,6 +26,27 @@
 
 #define GMAC_HI_REG_AE		0x80000000
 
+int dwmac_dma_reset(void __iomem *ioaddr)
+{
+	u32 value = readl(ioaddr + DMA_BUS_MODE);
+	int limit;
+
+	/* DMA SW reset */
+	value |= DMA_BUS_MODE_SFT_RESET;
+	writel(value, ioaddr + DMA_BUS_MODE);
+	limit = 10;
+	while (limit--) {
+		if (!(readl(ioaddr + DMA_BUS_MODE) & DMA_BUS_MODE_SFT_RESET))
+			break;
+		mdelay(10);
+	}
+
+	if (limit < 0)
+		return -EBUSY;
+
+	return 0;
+}
+
 /* CSR1 enables the transmit DMA to check for new descriptor */
 void dwmac_enable_dma_transmission(void __iomem *ioaddr)
 {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index c21015b..13752e9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1638,6 +1638,7 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
 	int pbl = DEFAULT_DMA_PBL, fixed_burst = 0, burst_len = 0;
 	int mixed_burst = 0;
 	int atds = 0;
+	int ret = 0;
 
 	if (priv->plat->dma_cfg) {
 		pbl = priv->plat->dma_cfg->pbl;
@@ -1649,9 +1650,16 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
 	if (priv->extend_desc && (priv->mode == STMMAC_RING_MODE))
 		atds = 1;
 
-	return priv->hw->dma->init(priv->ioaddr, pbl, fixed_burst, mixed_burst,
-				   burst_len, priv->dma_tx_phy,
-				   priv->dma_rx_phy, atds);
+	ret = priv->hw->dma->reset(priv->ioaddr);
+	if (ret) {
+		dev_err(priv->device, "Failed to reset the dma\n");
+		return ret;
+	}
+
+	priv->hw->dma->init(priv->ioaddr, pbl, fixed_burst, mixed_burst,
+			    burst_len, priv->dma_tx_phy,
+			    priv->dma_rx_phy, atds);
+	return ret;
 }
 
 /**
-- 
cgit v0.10.2


From afea03656add70a0e00f5b0039f87288c7af8b9f Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:28 +0100
Subject: stmmac: rework DMA bus setting and introduce new platform AXI
 structure

This patch restructures the DMA bus settings and this is done
by introducing a new platform structure used for programming
the AXI Bus Mode Register inside the DMA module.
This structure can be populated from device-tree as documented in the
binding txt file.

After initializing the DMA, the AXI register can be optionally tuned
for platform drivers based.
This patch also reworks some parameters to make coherent the DMA
configuration now that AXI register is introduced.
For example, the burst_len is managed by using the mentioned axi
support above; so the snps,burst-len parameter has been removed.
It makes sense to provide the AAL parameter from DT to Address-Aligned
Beats inside the Register0 and review the PBL settings when initialize
the engine.

For PCI glue, rebuilding the story of this setting, it
was added to align a configuration so not for fixing some
known problem. No issue raised after this patch.
It is safe to use the default burst length instead of
tuning it to the maximum value

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/stmmac.txt b/Documentation/devicetree/bindings/net/stmmac.txt
index e862a92..6605d19 100644
--- a/Documentation/devicetree/bindings/net/stmmac.txt
+++ b/Documentation/devicetree/bindings/net/stmmac.txt
@@ -17,7 +17,25 @@ Required properties:
 	The 1st cell is reset pre-delay in micro seconds.
 	The 2nd cell is reset pulse in micro seconds.
 	The 3rd cell is reset post-delay in micro seconds.
+
+Optional properties:
+- resets: Should contain a phandle to the STMMAC reset signal, if any
+- reset-names: Should contain the reset signal name "stmmaceth", if a
+	reset phandle is given
+- max-frame-size: See ethernet.txt file in the same directory
+- clocks: If present, the first clock should be the GMAC main clock and
+  the second clock should be peripheral's register interface clock. Further
+  clocks may be specified in derived bindings.
+- clock-names: One name for each entry in the clocks property, the
+  first one should be "stmmaceth" and the second one should be "pclk".
+- clk_ptp_ref: this is the PTP reference clock; in case of the PTP is
+  available this clock is used for programming the Timestamp Addend Register.
+  If not passed then the system clock will be used and this is fine on some
+  platforms.
+- tx-fifo-depth: See ethernet.txt file in the same directory
+- rx-fifo-depth: See ethernet.txt file in the same directory
 - snps,pbl		Programmable Burst Length
+- snps,aal		Address-Aligned Beats
 - snps,fixed-burst	Program the DMA to use the fixed burst mode
 - snps,mixed-burst	Program the DMA to use the mixed burst mode
 - snps,force_thresh_dma_mode	Force DMA to use the threshold mode for
@@ -29,27 +47,28 @@ Required properties:
 				supported by this device instance
 - snps,perfect-filter-entries:	Number of perfect filter entries supported
 				by this device instance
-
-Optional properties:
-- resets: Should contain a phandle to the STMMAC reset signal, if any
-- reset-names: Should contain the reset signal name "stmmaceth", if a
-	reset phandle is given
-- max-frame-size: See ethernet.txt file in the same directory
-- clocks: If present, the first clock should be the GMAC main clock
-  The optional second clock should be peripheral's register interface clock.
-  The third optional clock should be the ptp reference clock.
-  Further clocks may be specified in derived bindings.
-- clock-names: One name for each entry in the clocks property.
-  The first one should be "stmmaceth".
-  The optional second one should be "pclk".
-  The optional third one should be "clk_ptp_ref".
-- snps,burst_len: The AXI burst lenth value of the AXI BUS MODE register.
-- tx-fifo-depth: See ethernet.txt file in the same directory
-- rx-fifo-depth: See ethernet.txt file in the same directory
+- AXI BUS Mode parameters: below the list of all the parameters to program the
+			   AXI register inside the DMA module:
+	- snps,lpi_en: enable Low Power Interface
+	- snps,xit_frm: unlock on WoL
+	- snps,wr_osr_lmt: max write oustanding req. limit
+	- snps,rd_osr_lmt: max read oustanding req. limit
+	- snps,kbbe: do not cross 1KiB boundary.
+	- snps,axi_all: align address
+	- snps,blen: this is a vector of supported burst length.
+	- snps,fb: fixed-burst
+	- snps,mb: mixed-burst
+	- snps,rb: rebuild INCRx Burst
 - mdio: with compatible = "snps,dwmac-mdio", create and register mdio bus.
 
 Examples:
 
+	stmmac_axi_setup: stmmac-axi-config {
+		snps,wr_osr_lmt = <0xf>;
+		snps,rd_osr_lmt = <0xf>;
+		snps,blen = <256 128 64 32 0 0 0>;
+	};
+
 	gmac0: ethernet@e0800000 {
 		compatible = "st,spear600-gmac";
 		reg = <0xe0800000 0x8000>;
@@ -65,6 +84,7 @@ Examples:
 		tx-fifo-depth = <16384>;
 		clocks = <&clock>;
 		clock-names = "stmmaceth";
+		snps,axi-config = <&stmmac_axi_setup>;
 		mdio0 {
 			#address-cells = <1>;
 			#size-cells = <0>;
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index bac0e44..586a336 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -27,6 +27,7 @@
 
 #include <linux/etherdevice.h>
 #include <linux/netdevice.h>
+#include <linux/stmmac.h>
 #include <linux/phy.h>
 #include <linux/module.h>
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
@@ -378,7 +379,9 @@ struct stmmac_dma_ops {
 	/* DMA core initialization */
 	int (*reset)(void __iomem *ioaddr);
 	void (*init)(void __iomem *ioaddr, int pbl, int fb, int mb,
-		     int burst_len, u32 dma_tx, u32 dma_rx, int atds);
+		     int aal, u32 dma_tx, u32 dma_rx, int atds);
+	/* Configure the AXI Bus Mode Register */
+	void (*axi)(void __iomem *ioaddr, struct stmmac_axi *axi);
 	/* Dump DMA registers */
 	void (*dump_regs) (void __iomem *ioaddr);
 	/* Set tx/rx threshold in the csr6 register
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
index 9d36ae7..b0593a4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
@@ -240,7 +240,7 @@ enum rx_tx_priority_ratio {
 #define DMA_BUS_MODE_RPBL_MASK	0x003e0000	/* Rx-Programmable Burst Len */
 #define DMA_BUS_MODE_RPBL_SHIFT	17
 #define DMA_BUS_MODE_USP	0x00800000
-#define DMA_BUS_MODE_PBL	0x01000000
+#define DMA_BUS_MODE_MAXPBL	0x01000000
 #define DMA_BUS_MODE_AAL	0x02000000
 
 /* DMA CRS Control and Status Register Mapping */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index 5f0aea5..da32d60 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -30,24 +30,76 @@
 #include "dwmac1000.h"
 #include "dwmac_dma.h"
 
+static void dwmac1000_dma_axi(void __iomem *ioaddr, struct stmmac_axi *axi)
+{
+	u32 value = readl(ioaddr + DMA_AXI_BUS_MODE);
+	int i;
+
+	pr_info("dwmac1000: Master AXI performs %s burst length\n",
+		!(value & DMA_AXI_UNDEF) ? "fixed" : "any");
+
+	if (axi->axi_lpi_en)
+		value |= DMA_AXI_EN_LPI;
+	if (axi->axi_xit_frm)
+		value |= DMA_AXI_LPI_XIT_FRM;
+
+	value |= (axi->axi_wr_osr_lmt & DMA_AXI_WR_OSR_LMT_MASK) <<
+		 DMA_AXI_WR_OSR_LMT_SHIFT;
+
+	value |= (axi->axi_rd_osr_lmt & DMA_AXI_RD_OSR_LMT_MASK) <<
+		 DMA_AXI_RD_OSR_LMT_SHIFT;
+
+	/* Depending on the UNDEF bit the Master AXI will perform any burst
+	 * length according to the BLEN programmed (by default all BLEN are
+	 * set).
+	 */
+	for (i = 0; i < AXI_BLEN; i++) {
+		switch (axi->axi_blen[i]) {
+		case 256:
+			value |= DMA_AXI_BLEN256;
+			break;
+		case 128:
+			value |= DMA_AXI_BLEN128;
+			break;
+		case 64:
+			value |= DMA_AXI_BLEN64;
+			break;
+		case 32:
+			value |= DMA_AXI_BLEN32;
+			break;
+		case 16:
+			value |= DMA_AXI_BLEN16;
+			break;
+		case 8:
+			value |= DMA_AXI_BLEN8;
+			break;
+		case 4:
+			value |= DMA_AXI_BLEN4;
+			break;
+		}
+	}
+
+	writel(value, ioaddr + DMA_AXI_BUS_MODE);
+}
+
 static void dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
-			       int burst_len, u32 dma_tx, u32 dma_rx, int atds)
+			       int aal, u32 dma_tx, u32 dma_rx, int atds)
 {
-	u32 value;
+	u32 value = readl(ioaddr + DMA_BUS_MODE);
 
 	/*
-	 * Set the DMA PBL (Programmable Burst Length) mode
-	 * Before stmmac core 3.50 this mode bit was 4xPBL, and
+	 * Set the DMA PBL (Programmable Burst Length) mode.
+	 *
+	 * Note: before stmmac core 3.50 this mode bit was 4xPBL, and
 	 * post 3.5 mode bit acts as 8*PBL.
-	 * For core rev < 3.5, when the core is set for 4xPBL mode, the
-	 * DMA transfers the data in 4, 8, 16, 32, 64 & 128 beats
-	 * depending on pbl value.
-	 * For core rev > 3.5, when the core is set for 8xPBL mode, the
-	 * DMA transfers the data in 8, 16, 32, 64, 128 & 256 beats
-	 * depending on pbl value.
+	 *
+	 * This configuration doesn't take care about the Separate PBL
+	 * so only the bits: 13-8 are programmed with the PBL passed from the
+	 * platform.
 	 */
-	value = DMA_BUS_MODE_PBL | ((pbl << DMA_BUS_MODE_PBL_SHIFT) |
-				    (pbl << DMA_BUS_MODE_RPBL_SHIFT));
+	value |= DMA_BUS_MODE_MAXPBL;
+	value &= ~DMA_BUS_MODE_PBL_MASK;
+	value |= (pbl << DMA_BUS_MODE_PBL_SHIFT);
 
 	/* Set the Fixed burst mode */
 	if (fb)
@@ -60,26 +112,10 @@ static void dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
 	if (atds)
 		value |= DMA_BUS_MODE_ATDS;
 
-	writel(value, ioaddr + DMA_BUS_MODE);
+	if (aal)
+		value |= DMA_BUS_MODE_AAL;
 
-	/* In case of GMAC AXI configuration, program the DMA_AXI_BUS_MODE
-	 * for supported bursts.
-	 *
-	 * Note: This is applicable only for revision GMACv3.61a. For
-	 * older version this register is reserved and shall have no
-	 * effect.
-	 *
-	 * Note:
-	 *  For Fixed Burst Mode: if we directly write 0xFF to this
-	 *  register using the configurations pass from platform code,
-	 *  this would ensure that all bursts supported by core are set
-	 *  and those which are not supported would remain ineffective.
-	 *
-	 *  For Non Fixed Burst Mode: provide the maximum value of the
-	 *  burst length. Any burst equal or below the provided burst
-	 *  length would be allowed to perform.
-	 */
-	writel(burst_len, ioaddr + DMA_AXI_BUS_MODE);
+	writel(value, ioaddr + DMA_BUS_MODE);
 
 	/* Mask interrupts by writing to CSR7 */
 	writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
@@ -192,6 +228,7 @@ static void dwmac1000_rx_watchdog(void __iomem *ioaddr, u32 riwt)
 const struct stmmac_dma_ops dwmac1000_dma_ops = {
 	.reset = dwmac_dma_reset,
 	.init = dwmac1000_dma_init,
+	.axi = dwmac1000_dma_axi,
 	.dump_regs = dwmac1000_dump_dma_regs,
 	.dma_mode = dwmac1000_dma_operation_mode,
 	.enable_dma_transmission = dwmac_enable_dma_transmission,
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index c40582a..61f54c9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -33,7 +33,7 @@
 #include "dwmac_dma.h"
 
 static void dwmac100_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
-			      int burst_len, u32 dma_tx, u32 dma_rx, int atds)
+			      int aal, u32 dma_tx, u32 dma_rx, int atds)
 {
 	/* Enable Application Access by writing to DMA CSR0 */
 	writel(DMA_BUS_MODE_DEFAULT | (pbl << DMA_BUS_MODE_PBL_SHIFT),
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
index 13ca90e..726d9d9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
@@ -41,8 +41,40 @@
 
 /* Rx watchdog register */
 #define DMA_RX_WATCHDOG		0x00001024
-/* AXI Bus Mode */
+
+/* AXI Master Bus Mode */
 #define DMA_AXI_BUS_MODE	0x00001028
+
+#define DMA_AXI_EN_LPI		BIT(31)
+#define DMA_AXI_LPI_XIT_FRM	BIT(30)
+#define DMA_AXI_WR_OSR_LMT	GENMASK(23, 20)
+#define DMA_AXI_WR_OSR_LMT_SHIFT	20
+#define DMA_AXI_WR_OSR_LMT_MASK	0xf
+#define DMA_AXI_RD_OSR_LMT	GENMASK(19, 16)
+#define DMA_AXI_RD_OSR_LMT_SHIFT	16
+#define DMA_AXI_RD_OSR_LMT_MASK	0xf
+
+#define DMA_AXI_OSR_MAX		0xf
+#define DMA_AXI_MAX_OSR_LIMIT ((DMA_AXI_OSR_MAX << DMA_AXI_WR_OSR_LMT_SHIFT) | \
+			       (DMA_AXI_OSR_MAX << DMA_AXI_RD_OSR_LMT_SHIFT))
+#define	DMA_AXI_1KBBE		BIT(13)
+#define DMA_AXI_AAL		BIT(12)
+#define DMA_AXI_BLEN256		BIT(7)
+#define DMA_AXI_BLEN128		BIT(6)
+#define DMA_AXI_BLEN64		BIT(5)
+#define DMA_AXI_BLEN32		BIT(4)
+#define DMA_AXI_BLEN16		BIT(3)
+#define DMA_AXI_BLEN8		BIT(2)
+#define DMA_AXI_BLEN4		BIT(1)
+#define DMA_BURST_LEN_DEFAULT	(DMA_AXI_BLEN256 | DMA_AXI_BLEN128 | \
+				 DMA_AXI_BLEN64 | DMA_AXI_BLEN32 | \
+				 DMA_AXI_BLEN16 | DMA_AXI_BLEN8 | \
+				 DMA_AXI_BLEN4)
+
+#define DMA_AXI_UNDEF		BIT(0)
+
+#define DMA_AXI_BURST_LEN_MASK	0x000000FE
+
 #define DMA_CUR_TX_BUF_ADDR	0x00001050	/* Current Host Tx Buffer */
 #define DMA_CUR_RX_BUF_ADDR	0x00001054	/* Current Host Rx Buffer */
 #define DMA_HW_FEATURE		0x00001058	/* HW Feature Register */
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 13752e9..89c2626 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1635,7 +1635,7 @@ static void stmmac_check_ether_addr(struct stmmac_priv *priv)
  */
 static int stmmac_init_dma_engine(struct stmmac_priv *priv)
 {
-	int pbl = DEFAULT_DMA_PBL, fixed_burst = 0, burst_len = 0;
+	int pbl = DEFAULT_DMA_PBL, fixed_burst = 0, aal = 0;
 	int mixed_burst = 0;
 	int atds = 0;
 	int ret = 0;
@@ -1644,7 +1644,7 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
 		pbl = priv->plat->dma_cfg->pbl;
 		fixed_burst = priv->plat->dma_cfg->fixed_burst;
 		mixed_burst = priv->plat->dma_cfg->mixed_burst;
-		burst_len = priv->plat->dma_cfg->burst_len;
+		aal = priv->plat->dma_cfg->aal;
 	}
 
 	if (priv->extend_desc && (priv->mode == STMMAC_RING_MODE))
@@ -1657,8 +1657,12 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
 	}
 
 	priv->hw->dma->init(priv->ioaddr, pbl, fixed_burst, mixed_burst,
-			    burst_len, priv->dma_tx_phy,
-			    priv->dma_rx_phy, atds);
+			    aal, priv->dma_tx_phy, priv->dma_rx_phy, atds);
+
+	if ((priv->synopsys_id >= DWMAC_CORE_3_50) &&
+	    (priv->plat->axi && priv->hw->dma->axi))
+		priv->hw->dma->axi(priv->ioaddr, priv->plat->axi);
+
 	return ret;
 }
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
index d71a721..ae43887 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
@@ -81,7 +81,7 @@ static void stmmac_default_data(struct plat_stmmacenet_data *plat)
 	plat->mdio_bus_data->phy_mask = 0;
 
 	plat->dma_cfg->pbl = 32;
-	plat->dma_cfg->burst_len = DMA_AXI_BLEN_256;
+	/* TODO: AXI */
 
 	/* Set default value for multicast hash bins */
 	plat->multicast_filter_bins = HASH_TABLE_SIZE;
@@ -115,8 +115,8 @@ static int quark_default_data(struct plat_stmmacenet_data *plat,
 	plat->mdio_bus_data->phy_mask = 0;
 
 	plat->dma_cfg->pbl = 16;
-	plat->dma_cfg->burst_len = DMA_AXI_BLEN_256;
 	plat->dma_cfg->fixed_burst = 1;
+	/* AXI (TODO) */
 
 	/* Set default value for multicast hash bins */
 	plat->multicast_filter_bins = HASH_TABLE_SIZE;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
index 6a52fa1..69ccf48 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
@@ -96,6 +96,42 @@ static int dwmac1000_validate_ucast_entries(int ucast_entries)
 }
 
 /**
+ * stmmac_axi_setup - parse DT parameters for programming the AXI register
+ * @pdev: platform device
+ * @priv: driver private struct.
+ * Description:
+ * if required, from device-tree the AXI internal register can be tuned
+ * by using platform parameters.
+ */
+static struct stmmac_axi *stmmac_axi_setup(struct platform_device *pdev)
+{
+	struct device_node *np;
+	struct stmmac_axi *axi;
+
+	np = of_parse_phandle(pdev->dev.of_node, "snps,axi-config", 0);
+	if (!np)
+		return NULL;
+
+	axi = kzalloc(sizeof(axi), GFP_KERNEL);
+	if (!axi)
+		return ERR_PTR(-ENOMEM);
+
+	axi->axi_lpi_en = of_property_read_bool(np, "snps,lpi_en");
+	axi->axi_xit_frm = of_property_read_bool(np, "snps,xit_frm");
+	axi->axi_kbbe = of_property_read_bool(np, "snps,axi_kbbe");
+	axi->axi_axi_all = of_property_read_bool(np, "snps,axi_all");
+	axi->axi_fb = of_property_read_bool(np, "snps,axi_fb");
+	axi->axi_mb = of_property_read_bool(np, "snps,axi_mb");
+	axi->axi_rb =  of_property_read_bool(np, "snps,axi_rb");
+
+	of_property_read_u32(np, "snps,wr_osr_lmt", &axi->axi_wr_osr_lmt);
+	of_property_read_u32(np, "snps,rd_osr_lmt", &axi->axi_rd_osr_lmt);
+	of_property_read_u32_array(np, "snps,blen", axi->axi_blen, AXI_BLEN);
+
+	return axi;
+}
+
+/**
  * stmmac_probe_config_dt - parse device-tree driver parameters
  * @pdev: platform_device structure
  * @plat: driver data platform structure
@@ -216,13 +252,11 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac)
 		}
 		plat->dma_cfg = dma_cfg;
 		of_property_read_u32(np, "snps,pbl", &dma_cfg->pbl);
+		dma_cfg->aal = of_property_read_bool(np, "snps,aal");
 		dma_cfg->fixed_burst =
 			of_property_read_bool(np, "snps,fixed-burst");
 		dma_cfg->mixed_burst =
 			of_property_read_bool(np, "snps,mixed-burst");
-		of_property_read_u32(np, "snps,burst_len", &dma_cfg->burst_len);
-		if (dma_cfg->burst_len < 0 || dma_cfg->burst_len > 256)
-			dma_cfg->burst_len = 0;
 	}
 	plat->force_thresh_dma_mode = of_property_read_bool(np, "snps,force_thresh_dma_mode");
 	if (plat->force_thresh_dma_mode) {
@@ -230,6 +264,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac)
 		pr_warn("force_sf_dma_mode is ignored if force_thresh_dma_mode is set.");
 	}
 
+	plat->axi = stmmac_axi_setup(pdev);
+
 	return plat;
 }
 #else
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index eead8ab..6e53fa8 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -90,7 +90,21 @@ struct stmmac_dma_cfg {
 	int pbl;
 	int fixed_burst;
 	int mixed_burst;
-	int burst_len;
+	bool aal;
+};
+
+#define AXI_BLEN	7
+struct stmmac_axi {
+	bool axi_lpi_en;
+	bool axi_xit_frm;
+	u32 axi_wr_osr_lmt;
+	u32 axi_rd_osr_lmt;
+	bool axi_kbbe;
+	bool axi_axi_all;
+	u32 axi_blen[AXI_BLEN];
+	bool axi_fb;
+	bool axi_mb;
+	bool axi_rb;
 };
 
 struct plat_stmmacenet_data {
@@ -122,5 +136,6 @@ struct plat_stmmacenet_data {
 	int (*init)(struct platform_device *pdev, void *priv);
 	void (*exit)(struct platform_device *pdev, void *priv);
 	void *bsp_priv;
+	struct stmmac_axi *axi;
 };
 #endif
-- 
cgit v0.10.2


From 293e4365a1adb1d3747dcb7d12a52950e0e232a2 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:29 +0100
Subject: stmmac: change descriptor layout

This patch completely changes the descriptor layout to improve
the whole performances due to the single read usage of the
descriptors in critical paths.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/descs.h b/drivers/net/ethernet/stmicro/stmmac/descs.h
index 799c292..2e4c171 100644
--- a/drivers/net/ethernet/stmicro/stmmac/descs.h
+++ b/drivers/net/ethernet/stmicro/stmmac/descs.h
@@ -1,6 +1,6 @@
 /*******************************************************************************
-  Header File to describe the DMA descriptors.
-  Enhanced descriptors have been in case of DWMAC1000 Cores.
+  Header File to describe the DMA descriptors and related definitions.
+  This is for DWMAC100 and 1000 cores.
 
   This program is free software; you can redistribute it and/or modify it
   under the terms and conditions of the GNU General Public License,
@@ -24,198 +24,164 @@
 #ifndef __DESCS_H__
 #define __DESCS_H__
 
+#include <linux/bitops.h>
+
+/* Normal receive descriptor defines */
+
+/* RDES0 */
+#define	RDES0_PAYLOAD_CSUM_ERR	BIT(0)
+#define	RDES0_CRC_ERROR		BIT(1)
+#define	RDES0_DRIBBLING		BIT(2)
+#define	RDES0_MII_ERROR		BIT(3)
+#define	RDES0_RECEIVE_WATCHDOG	BIT(4)
+#define	RDES0_FRAME_TYPE	BIT(5)
+#define	RDES0_COLLISION		BIT(6)
+#define	RDES0_IPC_CSUM_ERROR	BIT(7)
+#define	RDES0_LAST_DESCRIPTOR	BIT(8)
+#define	RDES0_FIRST_DESCRIPTOR	BIT(9)
+#define	RDES0_VLAN_TAG		BIT(10)
+#define	RDES0_OVERFLOW_ERROR	BIT(11)
+#define	RDES0_LENGTH_ERROR	BIT(12)
+#define	RDES0_SA_FILTER_FAIL	BIT(13)
+#define	RDES0_DESCRIPTOR_ERROR	BIT(14)
+#define	RDES0_ERROR_SUMMARY	BIT(15)
+#define	RDES0_FRAME_LEN_MASK	GENMASK(29, 16)
+#define RDES0_FRAME_LEN_SHIFT	16
+#define	RDES0_DA_FILTER_FAIL	BIT(30)
+#define	RDES0_OWN		BIT(31)
+			/* RDES1 */
+#define	RDES1_BUFFER1_SIZE_MASK		GENMASK(10, 0)
+#define	RDES1_BUFFER2_SIZE_MASK		GENMASK(21, 11)
+#define	RDES1_BUFFER2_SIZE_SHIFT	11
+#define	RDES1_SECOND_ADDRESS_CHAINED	BIT(24)
+#define	RDES1_END_RING			BIT(25)
+#define	RDES1_DISABLE_IC		BIT(31)
+
+/* Enhanced receive descriptor defines */
+
+/* RDES0 (similar to normal RDES) */
+#define	 ERDES0_RX_MAC_ADDR	BIT(0)
+
+/* RDES1: completely differ from normal desc definitions */
+#define	ERDES1_BUFFER1_SIZE_MASK	GENMASK(12, 0)
+#define	ERDES1_SECOND_ADDRESS_CHAINED	BIT(14)
+#define	ERDES1_END_RING			BIT(15)
+#define	ERDES1_BUFFER2_SIZE_MASK	GENMASK(28, 16)
+#define ERDES1_BUFFER2_SIZE_SHIFT	16
+#define	ERDES1_DISABLE_IC		BIT(31)
+
+/* Normal transmit descriptor defines */
+/* TDES0 */
+#define	TDES0_DEFERRED			BIT(0)
+#define	TDES0_UNDERFLOW_ERROR		BIT(1)
+#define	TDES0_EXCESSIVE_DEFERRAL	BIT(2)
+#define	TDES0_COLLISION_COUNT_MASK	GENMASK(6, 3)
+#define	TDES0_VLAN_FRAME		BIT(7)
+#define	TDES0_EXCESSIVE_COLLISIONS	BIT(8)
+#define	TDES0_LATE_COLLISION		BIT(9)
+#define	TDES0_NO_CARRIER		BIT(10)
+#define	TDES0_LOSS_CARRIER		BIT(11)
+#define	TDES0_PAYLOAD_ERROR		BIT(12)
+#define	TDES0_FRAME_FLUSHED		BIT(13)
+#define	TDES0_JABBER_TIMEOUT		BIT(14)
+#define	TDES0_ERROR_SUMMARY		BIT(15)
+#define	TDES0_IP_HEADER_ERROR		BIT(16)
+#define	TDES0_TIME_STAMP_STATUS		BIT(17)
+#define	TDES0_OWN			BIT(31)
+/* TDES1 */
+#define	TDES1_BUFFER1_SIZE_MASK		GENMASK(10, 0)
+#define	TDES1_BUFFER2_SIZE_MASK		GENMASK(21, 11)
+#define	TDES1_BUFFER2_SIZE_SHIFT	11
+#define	TDES1_TIME_STAMP_ENABLE		BIT(22)
+#define	TDES1_DISABLE_PADDING		BIT(23)
+#define	TDES1_SECOND_ADDRESS_CHAINED	BIT(24)
+#define	TDES1_END_RING			BIT(25)
+#define	TDES1_CRC_DISABLE		BIT(26)
+#define	TDES1_CHECKSUM_INSERTION_MASK	GENMASK(28, 27)
+#define	TDES1_CHECKSUM_INSERTION_SHIFT	27
+#define	TDES1_FIRST_SEGMENT		BIT(29)
+#define	TDES1_LAST_SEGMENT		BIT(30)
+#define	TDES1_INTERRUPT			BIT(31)
+
+/* Enhanced transmit descriptor defines */
+/* TDES0 */
+#define	ETDES0_DEFERRED			BIT(0)
+#define	ETDES0_UNDERFLOW_ERROR		BIT(1)
+#define	ETDES0_EXCESSIVE_DEFERRAL	BIT(2)
+#define	ETDES0_COLLISION_COUNT_MASK	GENMASK(6, 3)
+#define	ETDES0_VLAN_FRAME		BIT(7)
+#define	ETDES0_EXCESSIVE_COLLISIONS	BIT(8)
+#define	ETDES0_LATE_COLLISION		BIT(9)
+#define	ETDES0_NO_CARRIER		BIT(10)
+#define	ETDES0_LOSS_CARRIER		BIT(11)
+#define	ETDES0_PAYLOAD_ERROR		BIT(12)
+#define	ETDES0_FRAME_FLUSHED		BIT(13)
+#define	ETDES0_JABBER_TIMEOUT		BIT(14)
+#define	ETDES0_ERROR_SUMMARY		BIT(15)
+#define	ETDES0_IP_HEADER_ERROR		BIT(16)
+#define	ETDES0_TIME_STAMP_STATUS	BIT(17)
+#define	ETDES0_SECOND_ADDRESS_CHAINED	BIT(20)
+#define	ETDES0_END_RING			BIT(21)
+#define	ETDES0_CHECKSUM_INSERTION_MASK	GENMASK(23, 22)
+#define	ETDES0_CHECKSUM_INSERTION_SHIFT	22
+#define	ETDES0_TIME_STAMP_ENABLE	BIT(25)
+#define	ETDES0_DISABLE_PADDING		BIT(26)
+#define	ETDES0_CRC_DISABLE		BIT(27)
+#define	ETDES0_FIRST_SEGMENT		BIT(28)
+#define	ETDES0_LAST_SEGMENT		BIT(29)
+#define	ETDES0_INTERRUPT		BIT(30)
+#define	ETDES0_OWN			BIT(31)
+/* TDES1 */
+#define	ETDES1_BUFFER1_SIZE_MASK	GENMASK(12, 0)
+#define	ETDES1_BUFFER2_SIZE_MASK	GENMASK(28, 16)
+#define	ETDES1_BUFFER2_SIZE_SHIFT	16
+
+/* Extended Receive descriptor definitions */
+#define	ERDES4_IP_PAYLOAD_TYPE_MASK	GENMASK(2, 6)
+#define	ERDES4_IP_HDR_ERR		BIT(3)
+#define	ERDES4_IP_PAYLOAD_ERR		BIT(4)
+#define	ERDES4_IP_CSUM_BYPASSED		BIT(5)
+#define	ERDES4_IPV4_PKT_RCVD		BIT(6)
+#define	ERDES4_IPV6_PKT_RCVD		BIT(7)
+#define	ERDES4_MSG_TYPE_MASK		GENMASK(11, 8)
+#define	ERDES4_PTP_FRAME_TYPE		BIT(12)
+#define	ERDES4_PTP_VER			BIT(13)
+#define	ERDES4_TIMESTAMP_DROPPED	BIT(14)
+#define	ERDES4_AV_PKT_RCVD		BIT(16)
+#define	ERDES4_AV_TAGGED_PKT_RCVD	BIT(17)
+#define	ERDES4_VLAN_TAG_PRI_VAL_MASK	GENMASK(20, 18)
+#define	ERDES4_L3_FILTER_MATCH		BIT(24)
+#define	ERDES4_L4_FILTER_MATCH		BIT(25)
+#define	ERDES4_L3_L4_FILT_NO_MATCH_MASK	GENMASK(27, 26)
+
+/* Extended RDES4 message type definitions */
+#define RDES_EXT_NO_PTP			0
+#define RDES_EXT_SYNC			1
+#define RDES_EXT_FOLLOW_UP		2
+#define RDES_EXT_DELAY_REQ		3
+#define RDES_EXT_DELAY_RESP		4
+#define RDES_EXT_PDELAY_REQ		5
+#define RDES_EXT_PDELAY_RESP		6
+#define RDES_EXT_PDELAY_FOLLOW_UP	7
+
 /* Basic descriptor structure for normal and alternate descriptors */
 struct dma_desc {
-	/* Receive descriptor */
-	union {
-		struct {
-			/* RDES0 */
-			u32 payload_csum_error:1;
-			u32 crc_error:1;
-			u32 dribbling:1;
-			u32 mii_error:1;
-			u32 receive_watchdog:1;
-			u32 frame_type:1;
-			u32 collision:1;
-			u32 ipc_csum_error:1;
-			u32 last_descriptor:1;
-			u32 first_descriptor:1;
-			u32 vlan_tag:1;
-			u32 overflow_error:1;
-			u32 length_error:1;
-			u32 sa_filter_fail:1;
-			u32 descriptor_error:1;
-			u32 error_summary:1;
-			u32 frame_length:14;
-			u32 da_filter_fail:1;
-			u32 own:1;
-			/* RDES1 */
-			u32 buffer1_size:11;
-			u32 buffer2_size:11;
-			u32 reserved1:2;
-			u32 second_address_chained:1;
-			u32 end_ring:1;
-			u32 reserved2:5;
-			u32 disable_ic:1;
-
-		} rx;
-		struct {
-			/* RDES0 */
-			u32 rx_mac_addr:1;
-			u32 crc_error:1;
-			u32 dribbling:1;
-			u32 error_gmii:1;
-			u32 receive_watchdog:1;
-			u32 frame_type:1;
-			u32 late_collision:1;
-			u32 ipc_csum_error:1;
-			u32 last_descriptor:1;
-			u32 first_descriptor:1;
-			u32 vlan_tag:1;
-			u32 overflow_error:1;
-			u32 length_error:1;
-			u32 sa_filter_fail:1;
-			u32 descriptor_error:1;
-			u32 error_summary:1;
-			u32 frame_length:14;
-			u32 da_filter_fail:1;
-			u32 own:1;
-			/* RDES1 */
-			u32 buffer1_size:13;
-			u32 reserved1:1;
-			u32 second_address_chained:1;
-			u32 end_ring:1;
-			u32 buffer2_size:13;
-			u32 reserved2:2;
-			u32 disable_ic:1;
-		} erx;		/* -- enhanced -- */
-
-		/* Transmit descriptor */
-		struct {
-			/* TDES0 */
-			u32 deferred:1;
-			u32 underflow_error:1;
-			u32 excessive_deferral:1;
-			u32 collision_count:4;
-			u32 vlan_frame:1;
-			u32 excessive_collisions:1;
-			u32 late_collision:1;
-			u32 no_carrier:1;
-			u32 loss_carrier:1;
-			u32 payload_error:1;
-			u32 frame_flushed:1;
-			u32 jabber_timeout:1;
-			u32 error_summary:1;
-			u32 ip_header_error:1;
-			u32 time_stamp_status:1;
-			u32 reserved1:13;
-			u32 own:1;
-			/* TDES1 */
-			u32 buffer1_size:11;
-			u32 buffer2_size:11;
-			u32 time_stamp_enable:1;
-			u32 disable_padding:1;
-			u32 second_address_chained:1;
-			u32 end_ring:1;
-			u32 crc_disable:1;
-			u32 checksum_insertion:2;
-			u32 first_segment:1;
-			u32 last_segment:1;
-			u32 interrupt:1;
-		} tx;
-		struct {
-			/* TDES0 */
-			u32 deferred:1;
-			u32 underflow_error:1;
-			u32 excessive_deferral:1;
-			u32 collision_count:4;
-			u32 vlan_frame:1;
-			u32 excessive_collisions:1;
-			u32 late_collision:1;
-			u32 no_carrier:1;
-			u32 loss_carrier:1;
-			u32 payload_error:1;
-			u32 frame_flushed:1;
-			u32 jabber_timeout:1;
-			u32 error_summary:1;
-			u32 ip_header_error:1;
-			u32 time_stamp_status:1;
-			u32 reserved1:2;
-			u32 second_address_chained:1;
-			u32 end_ring:1;
-			u32 checksum_insertion:2;
-			u32 reserved2:1;
-			u32 time_stamp_enable:1;
-			u32 disable_padding:1;
-			u32 crc_disable:1;
-			u32 first_segment:1;
-			u32 last_segment:1;
-			u32 interrupt:1;
-			u32 own:1;
-			/* TDES1 */
-			u32 buffer1_size:13;
-			u32 reserved3:3;
-			u32 buffer2_size:13;
-			u32 reserved4:3;
-		} etx;		/* -- enhanced -- */
-
-		u64 all_flags;
-	} des01;
+	unsigned int des0;
+	unsigned int des1;
 	unsigned int des2;
 	unsigned int des3;
 };
 
-/* Extended descriptor structure (supported by new SYNP GMAC generations) */
+/* Extended descriptor structure (e.g. >= databook 3.50a) */
 struct dma_extended_desc {
-	struct dma_desc basic;
-	union {
-		struct {
-			u32 ip_payload_type:3;
-			u32 ip_hdr_err:1;
-			u32 ip_payload_err:1;
-			u32 ip_csum_bypassed:1;
-			u32 ipv4_pkt_rcvd:1;
-			u32 ipv6_pkt_rcvd:1;
-			u32 msg_type:4;
-			u32 ptp_frame_type:1;
-			u32 ptp_ver:1;
-			u32 timestamp_dropped:1;
-			u32 reserved:1;
-			u32 av_pkt_rcvd:1;
-			u32 av_tagged_pkt_rcvd:1;
-			u32 vlan_tag_priority_val:3;
-			u32 reserved3:3;
-			u32 l3_filter_match:1;
-			u32 l4_filter_match:1;
-			u32 l3_l4_filter_no_match:2;
-			u32 reserved4:4;
-		} erx;
-		struct {
-			u32 reserved;
-		} etx;
-	} des4;
+	struct dma_desc basic;	/* Basic descriptors */
+	unsigned int des4;	/* Extended Status */
 	unsigned int des5;	/* Reserved */
 	unsigned int des6;	/* Tx/Rx Timestamp Low */
 	unsigned int des7;	/* Tx/Rx Timestamp High */
 };
 
 /* Transmit checksum insertion control */
-enum tdes_csum_insertion {
-	cic_disabled = 0,	/* Checksum Insertion Control */
-	cic_only_ip = 1,	/* Only IP header */
-	/* IP header but pseudoheader is not calculated */
-	cic_no_pseudoheader = 2,
-	cic_full = 3,		/* IP header and pseudoheader */
-};
-
-/* Extended RDES4 definitions */
-#define RDES_EXT_NO_PTP			0
-#define RDES_EXT_SYNC			0x1
-#define RDES_EXT_FOLLOW_UP		0x2
-#define RDES_EXT_DELAY_REQ		0x3
-#define RDES_EXT_DELAY_RESP		0x4
-#define RDES_EXT_PDELAY_REQ		0x5
-#define RDES_EXT_PDELAY_RESP		0x6
-#define RDES_EXT_PDELAY_FOLLOW_UP	0x7
+#define	TX_CIC_FULL	3	/* Include IP header and pseudoheader */
 
 #endif /* __DESCS_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/descs_com.h b/drivers/net/ethernet/stmicro/stmmac/descs_com.h
index 6f2cc78..7635a46 100644
--- a/drivers/net/ethernet/stmicro/stmmac/descs_com.h
+++ b/drivers/net/ethernet/stmicro/stmmac/descs_com.h
@@ -35,100 +35,91 @@
 /* Enhanced descriptors */
 static inline void ehn_desc_rx_set_on_ring(struct dma_desc *p, int end)
 {
-	p->des01.erx.buffer2_size = BUF_SIZE_8KiB - 1;
-	if (end)
-		p->des01.erx.end_ring = 1;
-}
+	p->des1 |= ((BUF_SIZE_8KiB - 1) << ERDES1_BUFFER2_SIZE_SHIFT)
+		   & ERDES1_BUFFER2_SIZE_MASK;
 
-static inline void ehn_desc_tx_set_on_ring(struct dma_desc *p, int end)
-{
 	if (end)
-		p->des01.etx.end_ring = 1;
+		p->des1 |= ERDES1_END_RING;
 }
 
-static inline void enh_desc_end_tx_desc_on_ring(struct dma_desc *p, int ter)
+static inline void enh_desc_end_tx_desc_on_ring(struct dma_desc *p, int end)
 {
-	p->des01.etx.end_ring = ter;
+	if (end)
+		p->des0 |= ETDES0_END_RING;
+	else
+		p->des0 &= ~ETDES0_END_RING;
 }
 
 static inline void enh_set_tx_desc_len_on_ring(struct dma_desc *p, int len)
 {
 	if (unlikely(len > BUF_SIZE_4KiB)) {
-		p->des01.etx.buffer1_size = BUF_SIZE_4KiB;
-		p->des01.etx.buffer2_size = len - BUF_SIZE_4KiB;
+		p->des1 |= (((len - BUF_SIZE_4KiB) << ETDES1_BUFFER2_SIZE_SHIFT)
+			    & ETDES1_BUFFER2_SIZE_MASK) | (BUF_SIZE_4KiB
+			    & ETDES1_BUFFER1_SIZE_MASK);
 	} else
-		p->des01.etx.buffer1_size = len;
+		p->des1 |= (len & ETDES1_BUFFER1_SIZE_MASK);
 }
 
 /* Normal descriptors */
 static inline void ndesc_rx_set_on_ring(struct dma_desc *p, int end)
 {
-	p->des01.rx.buffer2_size = BUF_SIZE_2KiB - 1;
-	if (end)
-		p->des01.rx.end_ring = 1;
-}
+	p->des1 |= ((BUF_SIZE_2KiB - 1) << RDES1_BUFFER2_SIZE_SHIFT)
+		    & RDES1_BUFFER2_SIZE_MASK;
 
-static inline void ndesc_tx_set_on_ring(struct dma_desc *p, int end)
-{
 	if (end)
-		p->des01.tx.end_ring = 1;
+		p->des1 |= RDES1_END_RING;
 }
 
-static inline void ndesc_end_tx_desc_on_ring(struct dma_desc *p, int ter)
+static inline void ndesc_end_tx_desc_on_ring(struct dma_desc *p, int end)
 {
-	p->des01.tx.end_ring = ter;
+	if (end)
+		p->des1 |= TDES1_END_RING;
+	else
+		p->des1 &= ~TDES1_END_RING;
 }
 
 static inline void norm_set_tx_desc_len_on_ring(struct dma_desc *p, int len)
 {
 	if (unlikely(len > BUF_SIZE_2KiB)) {
-		p->des01.etx.buffer1_size = BUF_SIZE_2KiB - 1;
-		p->des01.etx.buffer2_size = len - p->des01.etx.buffer1_size;
+		unsigned int buffer1 = (BUF_SIZE_2KiB - 1)
+					& TDES1_BUFFER1_SIZE_MASK;
+		p->des1 |= ((((len - buffer1) << TDES1_BUFFER2_SIZE_SHIFT)
+			    & TDES1_BUFFER2_SIZE_MASK) | buffer1);
 	} else
-		p->des01.tx.buffer1_size = len;
+		p->des1 |= (len & TDES1_BUFFER1_SIZE_MASK);
 }
 
 /* Specific functions used for Chain mode */
 
 /* Enhanced descriptors */
-static inline void ehn_desc_rx_set_on_chain(struct dma_desc *p, int end)
-{
-	p->des01.erx.second_address_chained = 1;
-}
-
-static inline void ehn_desc_tx_set_on_chain(struct dma_desc *p, int end)
+static inline void ehn_desc_rx_set_on_chain(struct dma_desc *p)
 {
-	p->des01.etx.second_address_chained = 1;
+	p->des1 |= ERDES1_SECOND_ADDRESS_CHAINED;
 }
 
-static inline void enh_desc_end_tx_desc_on_chain(struct dma_desc *p, int ter)
+static inline void enh_desc_end_tx_desc_on_chain(struct dma_desc *p)
 {
-	p->des01.etx.second_address_chained = 1;
+	p->des0 |= ETDES0_SECOND_ADDRESS_CHAINED;
 }
 
 static inline void enh_set_tx_desc_len_on_chain(struct dma_desc *p, int len)
 {
-	p->des01.etx.buffer1_size = len;
+	p->des1 |= (len & ETDES1_BUFFER1_SIZE_MASK);
 }
 
 /* Normal descriptors */
 static inline void ndesc_rx_set_on_chain(struct dma_desc *p, int end)
 {
-	p->des01.rx.second_address_chained = 1;
-}
-
-static inline void ndesc_tx_set_on_chain(struct dma_desc *p, int ring_size)
-{
-	p->des01.tx.second_address_chained = 1;
+	p->des1 |= RDES1_SECOND_ADDRESS_CHAINED;
 }
 
-static inline void ndesc_end_tx_desc_on_chain(struct dma_desc *p, int ter)
+static inline void ndesc_tx_set_on_chain(struct dma_desc *p)
 {
-	p->des01.tx.second_address_chained = 1;
+	p->des1 |= TDES1_SECOND_ADDRESS_CHAINED;
 }
 
 static inline void norm_set_tx_desc_len_on_chain(struct dma_desc *p, int len)
 {
-	p->des01.tx.buffer1_size = len;
+	p->des1 |= len & TDES1_BUFFER1_SIZE_MASK;
 }
 #endif /* __DESC_COM_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 7d94444..716b807 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -1,7 +1,7 @@
 /*******************************************************************************
   This contains the functions to handle the enhanced descriptors.
 
-  Copyright (C) 2007-2009  STMicroelectronics Ltd
+  Copyright (C) 2007-2014  STMicroelectronics Ltd
 
   This program is free software; you can redistribute it and/or modify it
   under the terms and conditions of the GNU General Public License,
@@ -29,44 +29,44 @@
 static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 				  struct dma_desc *p, void __iomem *ioaddr)
 {
-	int ret = 0;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
+	unsigned int tdes0 = p->des0;
+	int ret = 0;
 
-	if (unlikely(p->des01.etx.error_summary)) {
-		if (unlikely(p->des01.etx.jabber_timeout))
+	if (unlikely(tdes0 & ETDES0_ERROR_SUMMARY)) {
+		if (unlikely(tdes0 & ETDES0_JABBER_TIMEOUT))
 			x->tx_jabber++;
 
-		if (unlikely(p->des01.etx.frame_flushed)) {
+		if (unlikely(tdes0 & ETDES0_FRAME_FLUSHED)) {
 			x->tx_frame_flushed++;
 			dwmac_dma_flush_tx_fifo(ioaddr);
 		}
 
-		if (unlikely(p->des01.etx.loss_carrier)) {
+		if (unlikely(tdes0 & ETDES0_LOSS_CARRIER)) {
 			x->tx_losscarrier++;
 			stats->tx_carrier_errors++;
 		}
-		if (unlikely(p->des01.etx.no_carrier)) {
+		if (unlikely(tdes0 & ETDES0_NO_CARRIER)) {
 			x->tx_carrier++;
 			stats->tx_carrier_errors++;
 		}
-		if (unlikely(p->des01.etx.late_collision))
-			stats->collisions += p->des01.etx.collision_count;
-
-		if (unlikely(p->des01.etx.excessive_collisions))
-			stats->collisions += p->des01.etx.collision_count;
+		if (unlikely((tdes0 & ETDES0_LATE_COLLISION) ||
+			     (tdes0 & ETDES0_EXCESSIVE_COLLISIONS)))
+			stats->collisions +=
+				(tdes0 & ETDES0_COLLISION_COUNT_MASK) >> 3;
 
-		if (unlikely(p->des01.etx.excessive_deferral))
+		if (unlikely(tdes0 & ETDES0_EXCESSIVE_DEFERRAL))
 			x->tx_deferred++;
 
-		if (unlikely(p->des01.etx.underflow_error)) {
+		if (unlikely(tdes0 & ETDES0_UNDERFLOW_ERROR)) {
 			dwmac_dma_flush_tx_fifo(ioaddr);
 			x->tx_underflow++;
 		}
 
-		if (unlikely(p->des01.etx.ip_header_error))
+		if (unlikely(tdes0 & ETDES0_IP_HEADER_ERROR))
 			x->tx_ip_header_error++;
 
-		if (unlikely(p->des01.etx.payload_error)) {
+		if (unlikely(tdes0 & ETDES0_PAYLOAD_ERROR)) {
 			x->tx_payload_error++;
 			dwmac_dma_flush_tx_fifo(ioaddr);
 		}
@@ -74,11 +74,11 @@ static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 		ret = -1;
 	}
 
-	if (unlikely(p->des01.etx.deferred))
+	if (unlikely(tdes0 & ETDES0_DEFERRED))
 		x->tx_deferred++;
 
 #ifdef STMMAC_VLAN_TAG_USED
-	if (p->des01.etx.vlan_frame)
+	if (tdes0 & ETDES0_VLAN_FRAME)
 		x->tx_vlan++;
 #endif
 
@@ -87,7 +87,7 @@ static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 
 static int enh_desc_get_tx_len(struct dma_desc *p)
 {
-	return p->des01.etx.buffer1_size;
+	return (p->des1 & ETDES1_BUFFER1_SIZE_MASK);
 }
 
 static int enh_desc_coe_rdes0(int ipc_err, int type, int payload_err)
@@ -126,50 +126,55 @@ static int enh_desc_coe_rdes0(int ipc_err, int type, int payload_err)
 static void enh_desc_get_ext_status(void *data, struct stmmac_extra_stats *x,
 				    struct dma_extended_desc *p)
 {
-	if (unlikely(p->basic.des01.erx.rx_mac_addr)) {
-		if (p->des4.erx.ip_hdr_err)
+	unsigned int rdes0 = p->basic.des0;
+	unsigned int rdes4 = p->des4;
+
+	if (unlikely(rdes0 & ERDES0_RX_MAC_ADDR)) {
+		int message_type = (rdes4 & ERDES4_MSG_TYPE_MASK) >> 8;
+
+		if (rdes4 & ERDES4_IP_HDR_ERR)
 			x->ip_hdr_err++;
-		if (p->des4.erx.ip_payload_err)
+		if (rdes4 & ERDES4_IP_PAYLOAD_ERR)
 			x->ip_payload_err++;
-		if (p->des4.erx.ip_csum_bypassed)
+		if (rdes4 & ERDES4_IP_CSUM_BYPASSED)
 			x->ip_csum_bypassed++;
-		if (p->des4.erx.ipv4_pkt_rcvd)
+		if (rdes4 & ERDES4_IPV4_PKT_RCVD)
 			x->ipv4_pkt_rcvd++;
-		if (p->des4.erx.ipv6_pkt_rcvd)
+		if (rdes4 & ERDES4_IPV6_PKT_RCVD)
 			x->ipv6_pkt_rcvd++;
-		if (p->des4.erx.msg_type == RDES_EXT_SYNC)
+		if (message_type == RDES_EXT_SYNC)
 			x->rx_msg_type_sync++;
-		else if (p->des4.erx.msg_type == RDES_EXT_FOLLOW_UP)
+		else if (message_type == RDES_EXT_FOLLOW_UP)
 			x->rx_msg_type_follow_up++;
-		else if (p->des4.erx.msg_type == RDES_EXT_DELAY_REQ)
+		else if (message_type == RDES_EXT_DELAY_REQ)
 			x->rx_msg_type_delay_req++;
-		else if (p->des4.erx.msg_type == RDES_EXT_DELAY_RESP)
+		else if (message_type == RDES_EXT_DELAY_RESP)
 			x->rx_msg_type_delay_resp++;
-		else if (p->des4.erx.msg_type == RDES_EXT_PDELAY_REQ)
+		else if (message_type == RDES_EXT_PDELAY_REQ)
 			x->rx_msg_type_pdelay_req++;
-		else if (p->des4.erx.msg_type == RDES_EXT_PDELAY_RESP)
+		else if (message_type == RDES_EXT_PDELAY_RESP)
 			x->rx_msg_type_pdelay_resp++;
-		else if (p->des4.erx.msg_type == RDES_EXT_PDELAY_FOLLOW_UP)
+		else if (message_type == RDES_EXT_PDELAY_FOLLOW_UP)
 			x->rx_msg_type_pdelay_follow_up++;
 		else
 			x->rx_msg_type_ext_no_ptp++;
-		if (p->des4.erx.ptp_frame_type)
+		if (rdes4 & ERDES4_PTP_FRAME_TYPE)
 			x->ptp_frame_type++;
-		if (p->des4.erx.ptp_ver)
+		if (rdes4 & ERDES4_PTP_VER)
 			x->ptp_ver++;
-		if (p->des4.erx.timestamp_dropped)
+		if (rdes4 & ERDES4_TIMESTAMP_DROPPED)
 			x->timestamp_dropped++;
-		if (p->des4.erx.av_pkt_rcvd)
+		if (rdes4 & ERDES4_AV_PKT_RCVD)
 			x->av_pkt_rcvd++;
-		if (p->des4.erx.av_tagged_pkt_rcvd)
+		if (rdes4 & ERDES4_AV_TAGGED_PKT_RCVD)
 			x->av_tagged_pkt_rcvd++;
-		if (p->des4.erx.vlan_tag_priority_val)
+		if ((rdes4 & ERDES4_VLAN_TAG_PRI_VAL_MASK) >> 18)
 			x->vlan_tag_priority_val++;
-		if (p->des4.erx.l3_filter_match)
+		if (rdes4 & ERDES4_L3_FILTER_MATCH)
 			x->l3_filter_match++;
-		if (p->des4.erx.l4_filter_match)
+		if (rdes4 & ERDES4_L4_FILTER_MATCH)
 			x->l4_filter_match++;
-		if (p->des4.erx.l3_l4_filter_no_match)
+		if ((rdes4 & ERDES4_L3_L4_FILT_NO_MATCH_MASK) >> 26)
 			x->l3_l4_filter_no_match++;
 	}
 }
@@ -177,30 +182,30 @@ static void enh_desc_get_ext_status(void *data, struct stmmac_extra_stats *x,
 static int enh_desc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 				  struct dma_desc *p)
 {
-	int ret = good_frame;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
+	unsigned int rdes0 = p->des0;
+	int ret = good_frame;
 
-	if (unlikely(p->des01.erx.error_summary)) {
-		if (unlikely(p->des01.erx.descriptor_error)) {
+	if (unlikely(rdes0 & RDES0_ERROR_SUMMARY)) {
+		if (unlikely(rdes0 & RDES0_DESCRIPTOR_ERROR)) {
 			x->rx_desc++;
 			stats->rx_length_errors++;
 		}
-		if (unlikely(p->des01.erx.overflow_error))
+		if (unlikely(rdes0 & RDES0_OVERFLOW_ERROR))
 			x->rx_gmac_overflow++;
 
-		if (unlikely(p->des01.erx.ipc_csum_error))
+		if (unlikely(rdes0 & RDES0_IPC_CSUM_ERROR))
 			pr_err("\tIPC Csum Error/Giant frame\n");
 
-		if (unlikely(p->des01.erx.late_collision)) {
+		if (unlikely(rdes0 & RDES0_COLLISION))
 			stats->collisions++;
-		}
-		if (unlikely(p->des01.erx.receive_watchdog))
+		if (unlikely(rdes0 & RDES0_RECEIVE_WATCHDOG))
 			x->rx_watchdog++;
 
-		if (unlikely(p->des01.erx.error_gmii))
+		if (unlikely(rdes0 & RDES0_MII_ERROR))	/* GMII */
 			x->rx_mii++;
 
-		if (unlikely(p->des01.erx.crc_error)) {
+		if (unlikely(rdes0 & RDES0_CRC_ERROR)) {
 			x->rx_crc++;
 			stats->rx_crc_errors++;
 		}
@@ -211,26 +216,27 @@ static int enh_desc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 	 * It doesn't match with the information reported into the databook.
 	 * At any rate, we need to understand if the CSUM hw computation is ok
 	 * and report this info to the upper layers. */
-	ret = enh_desc_coe_rdes0(p->des01.erx.ipc_csum_error,
-		p->des01.erx.frame_type, p->des01.erx.rx_mac_addr);
+	ret = enh_desc_coe_rdes0(!!(rdes0 & RDES0_IPC_CSUM_ERROR),
+				 !!(rdes0 & RDES0_FRAME_TYPE),
+				 !!(rdes0 & ERDES0_RX_MAC_ADDR));
 
-	if (unlikely(p->des01.erx.dribbling))
+	if (unlikely(rdes0 & RDES0_DRIBBLING))
 		x->dribbling_bit++;
 
-	if (unlikely(p->des01.erx.sa_filter_fail)) {
+	if (unlikely(rdes0 & RDES0_SA_FILTER_FAIL)) {
 		x->sa_rx_filter_fail++;
 		ret = discard_frame;
 	}
-	if (unlikely(p->des01.erx.da_filter_fail)) {
+	if (unlikely(rdes0 & RDES0_DA_FILTER_FAIL)) {
 		x->da_rx_filter_fail++;
 		ret = discard_frame;
 	}
-	if (unlikely(p->des01.erx.length_error)) {
+	if (unlikely(rdes0 & RDES0_LENGTH_ERROR)) {
 		x->rx_length++;
 		ret = discard_frame;
 	}
 #ifdef STMMAC_VLAN_TAG_USED
-	if (p->des01.erx.vlan_tag)
+	if (rdes0 & RDES0_VLAN_TAG)
 		x->rx_vlan++;
 #endif
 
@@ -240,60 +246,59 @@ static int enh_desc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 static void enh_desc_init_rx_desc(struct dma_desc *p, int disable_rx_ic,
 				  int mode, int end)
 {
-	p->des01.all_flags = 0;
-	p->des01.erx.own = 1;
-	p->des01.erx.buffer1_size = BUF_SIZE_8KiB - 1;
+	p->des0 |= RDES0_OWN;
+	p->des1 |= ((BUF_SIZE_8KiB - 1) & ERDES1_BUFFER1_SIZE_MASK);
 
 	if (mode == STMMAC_CHAIN_MODE)
-		ehn_desc_rx_set_on_chain(p, end);
+		ehn_desc_rx_set_on_chain(p);
 	else
 		ehn_desc_rx_set_on_ring(p, end);
 
 	if (disable_rx_ic)
-		p->des01.erx.disable_ic = 1;
+		p->des1 |= ERDES1_DISABLE_IC;
 }
 
 static void enh_desc_init_tx_desc(struct dma_desc *p, int mode, int end)
 {
-	p->des01.all_flags = 0;
+	p->des0 &= ~ETDES0_OWN;
 	if (mode == STMMAC_CHAIN_MODE)
-		ehn_desc_tx_set_on_chain(p, end);
+		enh_desc_end_tx_desc_on_chain(p);
 	else
-		ehn_desc_tx_set_on_ring(p, end);
+		enh_desc_end_tx_desc_on_ring(p, end);
 }
 
 static int enh_desc_get_tx_owner(struct dma_desc *p)
 {
-	return p->des01.etx.own;
+	return (p->des0 & ETDES0_OWN) >> 31;
 }
 
 static int enh_desc_get_rx_owner(struct dma_desc *p)
 {
-	return p->des01.erx.own;
+	return (p->des0 & RDES0_OWN) >> 31;
 }
 
 static void enh_desc_set_tx_owner(struct dma_desc *p)
 {
-	p->des01.etx.own = 1;
+	p->des0 |= ETDES0_OWN;
 }
 
 static void enh_desc_set_rx_owner(struct dma_desc *p)
 {
-	p->des01.erx.own = 1;
+	p->des0 |= RDES0_OWN;
 }
 
 static int enh_desc_get_tx_ls(struct dma_desc *p)
 {
-	return p->des01.etx.last_segment;
+	return (p->des0 & ETDES0_LAST_SEGMENT) >> 29;
 }
 
 static void enh_desc_release_tx_desc(struct dma_desc *p, int mode)
 {
-	int ter = p->des01.etx.end_ring;
+	int ter = (p->des0 & ETDES0_END_RING) >> 21;
 
 	memset(p, 0, offsetof(struct dma_desc, des2));
 	if (mode == STMMAC_CHAIN_MODE)
-		enh_desc_end_tx_desc_on_chain(p, ter);
+		enh_desc_end_tx_desc_on_chain(p);
 	else
 		enh_desc_end_tx_desc_on_ring(p, ter);
 }
@@ -301,49 +306,60 @@ static void enh_desc_release_tx_desc(struct dma_desc *p, int mode)
 static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 				     int csum_flag, int mode)
 {
-	p->des01.etx.first_segment = is_fs;
+	unsigned int tdes0 = p->des0;
+
+	if (is_fs)
+		tdes0 |= ETDES0_FIRST_SEGMENT;
+	else
+		tdes0 &= ~ETDES0_FIRST_SEGMENT;
+
+	if (likely(csum_flag))
+		tdes0 |= (TX_CIC_FULL << ETDES0_CHECKSUM_INSERTION_SHIFT);
+	else
+		tdes0 &= ~(TX_CIC_FULL << ETDES0_CHECKSUM_INSERTION_SHIFT);
+
+	p->des0 = tdes0;
 
 	if (mode == STMMAC_CHAIN_MODE)
 		enh_set_tx_desc_len_on_chain(p, len);
 	else
 		enh_set_tx_desc_len_on_ring(p, len);
-
-	if (likely(csum_flag))
-		p->des01.etx.checksum_insertion = cic_full;
 }
 
 static void enh_desc_clear_tx_ic(struct dma_desc *p)
 {
-	p->des01.etx.interrupt = 0;
+	p->des0 &= ~ETDES0_INTERRUPT;
 }
 
 static void enh_desc_close_tx_desc(struct dma_desc *p)
 {
-	p->des01.etx.last_segment = 1;
-	p->des01.etx.interrupt = 1;
+	p->des0 |= ETDES0_LAST_SEGMENT | ETDES0_INTERRUPT;
 }
 
 static int enh_desc_get_rx_frame_len(struct dma_desc *p, int rx_coe_type)
 {
+	unsigned int csum = 0;
 	/* The type-1 checksum offload engines append the checksum at
 	 * the end of frame and the two bytes of checksum are added in
 	 * the length.
 	 * Adjust for that in the framelen for type-1 checksum offload
-	 * engines. */
+	 * engines.
+	 */
 	if (rx_coe_type == STMMAC_RX_COE_TYPE1)
-		return p->des01.erx.frame_length - 2;
-	else
-		return p->des01.erx.frame_length;
+		csum = 2;
+
+	return (((p->des0 & RDES0_FRAME_LEN_MASK) >> RDES0_FRAME_LEN_SHIFT) -
+		csum);
 }
 
 static void enh_desc_enable_tx_timestamp(struct dma_desc *p)
 {
-	p->des01.etx.time_stamp_enable = 1;
+	p->des0 |= ETDES0_TIME_STAMP_ENABLE;
 }
 
 static int enh_desc_get_tx_timestamp_status(struct dma_desc *p)
 {
-	return p->des01.etx.time_stamp_status;
+	return (p->des0 & ETDES0_TIME_STAMP_STATUS) >> 17;
 }
 
 static u64 enh_desc_get_timestamp(void *desc, u32 ats)
@@ -368,7 +384,7 @@ static int enh_desc_get_rx_timestamp_status(void *desc, u32 ats)
 {
 	if (ats) {
 		struct dma_extended_desc *p = (struct dma_extended_desc *)desc;
-		return p->basic.des01.erx.ipc_csum_error;
+		return (p->basic.des0 & RDES0_IPC_CSUM_ERROR) >> 7;
 	} else {
 		struct dma_desc *p = (struct dma_desc *)desc;
 		if ((p->des2 == 0xffffffff) && (p->des3 == 0xffffffff))
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index 48c3456..460c573 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -29,33 +29,38 @@
 static int ndesc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 			       struct dma_desc *p, void __iomem *ioaddr)
 {
-	int ret = 0;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
+	unsigned int tdes0 = p->des0;
+	int ret = 0;
 
-	if (unlikely(p->des01.tx.error_summary)) {
-		if (unlikely(p->des01.tx.underflow_error)) {
+	if (unlikely(tdes0 & TDES0_ERROR_SUMMARY)) {
+		if (unlikely(tdes0 & TDES0_UNDERFLOW_ERROR)) {
 			x->tx_underflow++;
 			stats->tx_fifo_errors++;
 		}
-		if (unlikely(p->des01.tx.no_carrier)) {
+		if (unlikely(tdes0 & TDES0_NO_CARRIER)) {
 			x->tx_carrier++;
 			stats->tx_carrier_errors++;
 		}
-		if (unlikely(p->des01.tx.loss_carrier)) {
+		if (unlikely(tdes0 & TDES0_LOSS_CARRIER)) {
 			x->tx_losscarrier++;
 			stats->tx_carrier_errors++;
 		}
-		if (unlikely((p->des01.tx.excessive_deferral) ||
-			     (p->des01.tx.excessive_collisions) ||
-			     (p->des01.tx.late_collision)))
-			stats->collisions += p->des01.tx.collision_count;
+		if (unlikely((tdes0 & TDES0_EXCESSIVE_DEFERRAL) ||
+			     (tdes0 & TDES0_EXCESSIVE_COLLISIONS) ||
+			     (tdes0 & TDES0_LATE_COLLISION))) {
+			unsigned int collisions;
+
+			collisions = (tdes0 & TDES0_COLLISION_COUNT_MASK) >> 3;
+			stats->collisions += collisions;
+		}
 		ret = -1;
 	}
 
-	if (p->des01.etx.vlan_frame)
+	if (tdes0 & TDES0_VLAN_FRAME)
 		x->tx_vlan++;
 
-	if (unlikely(p->des01.tx.deferred))
+	if (unlikely(tdes0 & TDES0_DEFERRED))
 		x->tx_deferred++;
 
 	return ret;
@@ -63,7 +68,7 @@ static int ndesc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 
 static int ndesc_get_tx_len(struct dma_desc *p)
 {
-	return p->des01.tx.buffer1_size;
+	return (p->des1 & RDES1_BUFFER1_SIZE_MASK);
 }
 
 /* This function verifies if each incoming frame has some errors
@@ -74,47 +79,48 @@ static int ndesc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 			       struct dma_desc *p)
 {
 	int ret = good_frame;
+	unsigned int rdes0 = p->des0;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
 
-	if (unlikely(p->des01.rx.last_descriptor == 0)) {
+	if (unlikely(!(rdes0 & RDES0_LAST_DESCRIPTOR))) {
 		pr_warn("%s: Oversized frame spanned multiple buffers\n",
 			__func__);
 		stats->rx_length_errors++;
 		return discard_frame;
 	}
 
-	if (unlikely(p->des01.rx.error_summary)) {
-		if (unlikely(p->des01.rx.descriptor_error))
+	if (unlikely(rdes0 & RDES0_ERROR_SUMMARY)) {
+		if (unlikely(rdes0 & RDES0_DESCRIPTOR_ERROR))
 			x->rx_desc++;
-		if (unlikely(p->des01.rx.sa_filter_fail))
+		if (unlikely(rdes0 & RDES0_SA_FILTER_FAIL))
 			x->sa_filter_fail++;
-		if (unlikely(p->des01.rx.overflow_error))
+		if (unlikely(rdes0 & RDES0_OVERFLOW_ERROR))
 			x->overflow_error++;
-		if (unlikely(p->des01.rx.ipc_csum_error))
+		if (unlikely(rdes0 & RDES0_IPC_CSUM_ERROR))
 			x->ipc_csum_error++;
-		if (unlikely(p->des01.rx.collision)) {
+		if (unlikely(rdes0 & RDES0_COLLISION)) {
 			x->rx_collision++;
 			stats->collisions++;
 		}
-		if (unlikely(p->des01.rx.crc_error)) {
+		if (unlikely(rdes0 & RDES0_CRC_ERROR)) {
 			x->rx_crc++;
 			stats->rx_crc_errors++;
 		}
 		ret = discard_frame;
 	}
-	if (unlikely(p->des01.rx.dribbling))
+	if (unlikely(rdes0 & RDES0_DRIBBLING))
 		x->dribbling_bit++;
 
-	if (unlikely(p->des01.rx.length_error)) {
+	if (unlikely(rdes0 & RDES0_LENGTH_ERROR)) {
 		x->rx_length++;
 		ret = discard_frame;
 	}
-	if (unlikely(p->des01.rx.mii_error)) {
+	if (unlikely(rdes0 & RDES0_MII_ERROR)) {
 		x->rx_mii++;
 		ret = discard_frame;
 	}
 #ifdef STMMAC_VLAN_TAG_USED
-	if (p->des01.rx.vlan_tag)
+	if (rdes0 & RDES0_VLAN_TAG)
 		x->vlan_tag++;
 #endif
 	return ret;
@@ -123,9 +129,8 @@ static int ndesc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 static void ndesc_init_rx_desc(struct dma_desc *p, int disable_rx_ic, int mode,
 			       int end)
 {
-	p->des01.all_flags = 0;
-	p->des01.rx.own = 1;
-	p->des01.rx.buffer1_size = BUF_SIZE_2KiB - 1;
+	p->des0 |= RDES0_OWN;
+	p->des1 |= (BUF_SIZE_2KiB - 1) & RDES1_BUFFER1_SIZE_MASK;
 
 	if (mode == STMMAC_CHAIN_MODE)
 		ndesc_rx_set_on_chain(p, end);
@@ -133,50 +138,50 @@ static void ndesc_init_rx_desc(struct dma_desc *p, int disable_rx_ic, int mode,
 		ndesc_rx_set_on_ring(p, end);
 
 	if (disable_rx_ic)
-		p->des01.rx.disable_ic = 1;
+		p->des1 |= RDES1_DISABLE_IC;
 }
 
 static void ndesc_init_tx_desc(struct dma_desc *p, int mode, int end)
 {
-	p->des01.all_flags = 0;
+	p->des0 &= ~TDES0_OWN;
 	if (mode == STMMAC_CHAIN_MODE)
-		ndesc_tx_set_on_chain(p, end);
+		ndesc_tx_set_on_chain(p);
 	else
-		ndesc_tx_set_on_ring(p, end);
+		ndesc_end_tx_desc_on_ring(p, end);
 }
 
 static int ndesc_get_tx_owner(struct dma_desc *p)
 {
-	return p->des01.tx.own;
+	return (p->des0 & TDES0_OWN) >> 31;
 }
 
 static int ndesc_get_rx_owner(struct dma_desc *p)
 {
-	return p->des01.rx.own;
+	return (p->des0 & RDES0_OWN) >> 31;
 }
 
 static void ndesc_set_tx_owner(struct dma_desc *p)
 {
-	p->des01.tx.own = 1;
+	p->des0 |= TDES0_OWN;
 }
 
 static void ndesc_set_rx_owner(struct dma_desc *p)
 {
-	p->des01.rx.own = 1;
+	p->des0 |= RDES0_OWN;
 }
 
 static int ndesc_get_tx_ls(struct dma_desc *p)
 {
-	return p->des01.tx.last_segment;
+	return (p->des1 & TDES1_LAST_SEGMENT) >> 30;
 }
 
 static void ndesc_release_tx_desc(struct dma_desc *p, int mode)
 {
-	int ter = p->des01.tx.end_ring;
+	int ter = (p->des1 & TDES1_END_RING) >> 25;
 
 	memset(p, 0, offsetof(struct dma_desc, des2));
 	if (mode == STMMAC_CHAIN_MODE)
-		ndesc_end_tx_desc_on_chain(p, ter);
+		ndesc_tx_set_on_chain(p);
 	else
 		ndesc_end_tx_desc_on_ring(p, ter);
 }
@@ -184,48 +189,62 @@ static void ndesc_release_tx_desc(struct dma_desc *p, int mode)
 static void ndesc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 				  int csum_flag, int mode)
 {
-	p->des01.tx.first_segment = is_fs;
+	unsigned int tdes1 = p->des1;
+
+	if (is_fs)
+		tdes1 |= TDES1_FIRST_SEGMENT;
+	else
+		tdes1 &= ~TDES1_FIRST_SEGMENT;
+
+	if (likely(csum_flag))
+		tdes1 |= (TX_CIC_FULL) << TDES1_CHECKSUM_INSERTION_SHIFT;
+	else
+		tdes1 &= ~(TX_CIC_FULL << TDES1_CHECKSUM_INSERTION_SHIFT);
+
+	p->des1 = tdes1;
+
 	if (mode == STMMAC_CHAIN_MODE)
 		norm_set_tx_desc_len_on_chain(p, len);
 	else
 		norm_set_tx_desc_len_on_ring(p, len);
-
-	if (likely(csum_flag))
-		p->des01.tx.checksum_insertion = cic_full;
 }
 
 static void ndesc_clear_tx_ic(struct dma_desc *p)
 {
-	p->des01.tx.interrupt = 0;
+	p->des1 &= ~TDES1_INTERRUPT;
 }
 
 static void ndesc_close_tx_desc(struct dma_desc *p)
 {
-	p->des01.tx.last_segment = 1;
-	p->des01.tx.interrupt = 1;
+	p->des1 |= TDES1_LAST_SEGMENT | TDES1_INTERRUPT;
 }
 
 static int ndesc_get_rx_frame_len(struct dma_desc *p, int rx_coe_type)
 {
+	unsigned int csum = 0;
+
 	/* The type-1 checksum offload engines append the checksum at
 	 * the end of frame and the two bytes of checksum are added in
 	 * the length.
 	 * Adjust for that in the framelen for type-1 checksum offload
-	 * engines. */
+	 * engines
+	 */
 	if (rx_coe_type == STMMAC_RX_COE_TYPE1)
-		return p->des01.rx.frame_length - 2;
-	else
-		return p->des01.rx.frame_length;
+		csum = 2;
+
+	return (((p->des0 & RDES0_FRAME_LEN_MASK) >> RDES0_FRAME_LEN_SHIFT) -
+		csum);
+
 }
 
 static void ndesc_enable_tx_timestamp(struct dma_desc *p)
 {
-	p->des01.tx.time_stamp_enable = 1;
+	p->des1 |= TDES1_TIME_STAMP_ENABLE;
 }
 
 static int ndesc_get_tx_timestamp_status(struct dma_desc *p)
 {
-	return p->des01.tx.time_stamp_status;
+	return (p->des0 & TDES0_TIME_STAMP_STATUS) >> 17;
 }
 
 static u64 ndesc_get_timestamp(void *desc, u32 ats)
-- 
cgit v0.10.2


From e3ad57c96715df2989ce6c18e58faf2913b305cb Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:30 +0100
Subject: stmmac: review RX/TX ring management

This patch is to rework the ring management now optimized.
The indexes into the ring buffer are always incremented, and
the entry is accessed via doing a modulo to find the "real"
position in the ring.
It is inefficient, modulo is an expensive operation.

The formula [(entry + 1) & (size - 1)] is now adopted on
a ring that is power-of-2 in size.
Then, the number of elements cannot be set by command line but
it is fixed.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index cf28dab..2763772 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -31,8 +31,7 @@
 static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 {
 	struct stmmac_priv *priv = (struct stmmac_priv *)p;
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int entry = priv->cur_tx % txsize;
+	unsigned int entry = priv->cur_tx;
 	struct dma_desc *desc = priv->dma_tx + entry;
 	unsigned int nopaged_len = skb_headlen(skb);
 	unsigned int bmax;
@@ -54,7 +53,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 
 	while (len != 0) {
 		priv->tx_skbuff[entry] = NULL;
-		entry = (++priv->cur_tx) % txsize;
+		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
 		desc = priv->dma_tx + entry;
 
 		if (len > bmax) {
@@ -82,6 +81,9 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			len = 0;
 		}
 	}
+
+	priv->cur_tx = entry;
+
 	return entry;
 }
 
@@ -138,7 +140,7 @@ static void stmmac_refill_desc3(void *priv_ptr, struct dma_desc *p)
 		 */
 		p->des3 = (unsigned int)(priv->dma_rx_phy +
 					 (((priv->dirty_rx) + 1) %
-					  priv->dma_rx_size) *
+					  DMA_RX_SIZE) *
 					 sizeof(struct dma_desc));
 }
 
@@ -151,10 +153,9 @@ static void stmmac_clean_desc3(void *priv_ptr, struct dma_desc *p)
 		 * 1588-2002 time stamping is enabled, hence reinitialize it
 		 * to keep explicit chaining in the descriptor.
 		 */
-		p->des3 = (unsigned int)(priv->dma_tx_phy +
-					 (((priv->dirty_tx + 1) %
-					   priv->dma_tx_size) *
-					  sizeof(struct dma_desc)));
+		p->des3 = (unsigned int)((priv->dma_tx_phy +
+					  ((priv->dirty_tx + 1) % DMA_TX_SIZE))
+					  * sizeof(struct dma_desc));
 }
 
 const struct stmmac_mode_ops chain_mode_ops = {
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 586a336..09291af 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -42,6 +42,10 @@
 #define	DWMAC_CORE_3_40	0x34
 #define	DWMAC_CORE_3_50	0x35
 
+#define DMA_TX_SIZE 512
+#define DMA_RX_SIZE 512
+#define STMMAC_GET_ENTRY(x, size)	((x + 1) & (size - 1))
+
 #undef FRAME_FILTER_DEBUG
 /* #define FRAME_FILTER_DEBUG */
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index 5dd50c6..4358a87 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -31,8 +31,7 @@
 static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 {
 	struct stmmac_priv *priv = (struct stmmac_priv *)p;
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int entry = priv->cur_tx % txsize;
+	unsigned int entry = priv->cur_tx;
 	struct dma_desc *desc;
 	unsigned int nopaged_len = skb_headlen(skb);
 	unsigned int bmax, len;
@@ -62,7 +61,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 						STMMAC_RING_MODE);
 		wmb();
 		priv->tx_skbuff[entry] = NULL;
-		entry = (++priv->cur_tx) % txsize;
+		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
 
 		if (priv->extend_desc)
 			desc = (struct dma_desc *)(priv->dma_etx + entry);
@@ -90,6 +89,8 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 						STMMAC_RING_MODE);
 	}
 
+	priv->cur_tx = entry;
+
 	return entry;
 }
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 1f3b33a..7ae7c64 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -54,7 +54,6 @@ struct stmmac_priv {
 	struct sk_buff **tx_skbuff;
 	unsigned int cur_tx;
 	unsigned int dirty_tx;
-	unsigned int dma_tx_size;
 	u32 tx_count_frames;
 	u32 tx_coal_frames;
 	u32 tx_coal_timer;
@@ -71,7 +70,6 @@ struct stmmac_priv {
 	struct sk_buff **rx_skbuff;
 	unsigned int cur_rx;
 	unsigned int dirty_rx;
-	unsigned int dma_rx_size;
 	unsigned int dma_buf_sz;
 	u32 rx_riwt;
 	int hwts_rx_en;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 89c2626..eb555f0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -71,15 +71,7 @@ static int phyaddr = -1;
 module_param(phyaddr, int, S_IRUGO);
 MODULE_PARM_DESC(phyaddr, "Physical device address");
 
-#define DMA_TX_SIZE 256
-static int dma_txsize = DMA_TX_SIZE;
-module_param(dma_txsize, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(dma_txsize, "Number of descriptors in the TX list");
-
-#define DMA_RX_SIZE 256
-static int dma_rxsize = DMA_RX_SIZE;
-module_param(dma_rxsize, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(dma_rxsize, "Number of descriptors in the RX list");
+#define STMMAC_TX_THRESH	(DMA_TX_SIZE / 4)
 
 static int flow_ctrl = FLOW_OFF;
 module_param(flow_ctrl, int, S_IRUGO | S_IWUSR);
@@ -134,10 +126,6 @@ static void stmmac_verify_args(void)
 {
 	if (unlikely(watchdog < 0))
 		watchdog = TX_TIMEO;
-	if (unlikely(dma_rxsize < 0))
-		dma_rxsize = DMA_RX_SIZE;
-	if (unlikely(dma_txsize < 0))
-		dma_txsize = DMA_TX_SIZE;
 	if (unlikely((buf_sz < DEFAULT_BUFSIZE) || (buf_sz > BUF_SIZE_16KiB)))
 		buf_sz = DEFAULT_BUFSIZE;
 	if (unlikely(flow_ctrl > 1))
@@ -197,12 +185,28 @@ static void print_pkt(unsigned char *buf, int len)
 	print_hex_dump_bytes("", DUMP_PREFIX_OFFSET, buf, len);
 }
 
-/* minimum number of free TX descriptors required to wake up TX process */
-#define STMMAC_TX_THRESH(x)	(x->dma_tx_size/4)
-
 static inline u32 stmmac_tx_avail(struct stmmac_priv *priv)
 {
-	return priv->dirty_tx + priv->dma_tx_size - priv->cur_tx - 1;
+	unsigned avail;
+
+	if (priv->dirty_tx > priv->cur_tx)
+		avail = priv->dirty_tx - priv->cur_tx - 1;
+	else
+		avail = DMA_TX_SIZE - priv->cur_tx + priv->dirty_tx - 1;
+
+	return avail;
+}
+
+static inline u32 stmmac_rx_dirty(struct stmmac_priv *priv)
+{
+	unsigned dirty;
+
+	if (priv->dirty_rx <= priv->cur_rx)
+		dirty = priv->cur_rx - priv->dirty_rx;
+	else
+		dirty = DMA_RX_SIZE - priv->dirty_rx + priv->cur_rx;
+
+	return dirty;
 }
 
 /**
@@ -906,19 +910,16 @@ static void stmmac_display_ring(void *head, int size, int extend_desc)
 
 static void stmmac_display_rings(struct stmmac_priv *priv)
 {
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int rxsize = priv->dma_rx_size;
-
 	if (priv->extend_desc) {
 		pr_info("Extended RX descriptor ring:\n");
-		stmmac_display_ring((void *)priv->dma_erx, rxsize, 1);
+		stmmac_display_ring((void *)priv->dma_erx, DMA_RX_SIZE, 1);
 		pr_info("Extended TX descriptor ring:\n");
-		stmmac_display_ring((void *)priv->dma_etx, txsize, 1);
+		stmmac_display_ring((void *)priv->dma_etx, DMA_TX_SIZE, 1);
 	} else {
 		pr_info("RX descriptor ring:\n");
-		stmmac_display_ring((void *)priv->dma_rx, rxsize, 0);
+		stmmac_display_ring((void *)priv->dma_rx, DMA_RX_SIZE, 0);
 		pr_info("TX descriptor ring:\n");
-		stmmac_display_ring((void *)priv->dma_tx, txsize, 0);
+		stmmac_display_ring((void *)priv->dma_tx, DMA_TX_SIZE, 0);
 	}
 }
 
@@ -947,28 +948,26 @@ static int stmmac_set_bfsize(int mtu, int bufsize)
 static void stmmac_clear_descriptors(struct stmmac_priv *priv)
 {
 	int i;
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int rxsize = priv->dma_rx_size;
 
 	/* Clear the Rx/Tx descriptors */
-	for (i = 0; i < rxsize; i++)
+	for (i = 0; i < DMA_RX_SIZE; i++)
 		if (priv->extend_desc)
 			priv->hw->desc->init_rx_desc(&priv->dma_erx[i].basic,
 						     priv->use_riwt, priv->mode,
-						     (i == rxsize - 1));
+						     (i == DMA_RX_SIZE - 1));
 		else
 			priv->hw->desc->init_rx_desc(&priv->dma_rx[i],
 						     priv->use_riwt, priv->mode,
-						     (i == rxsize - 1));
-	for (i = 0; i < txsize; i++)
+						     (i == DMA_RX_SIZE - 1));
+	for (i = 0; i < DMA_TX_SIZE; i++)
 		if (priv->extend_desc)
 			priv->hw->desc->init_tx_desc(&priv->dma_etx[i].basic,
 						     priv->mode,
-						     (i == txsize - 1));
+						     (i == DMA_TX_SIZE - 1));
 		else
 			priv->hw->desc->init_tx_desc(&priv->dma_tx[i],
 						     priv->mode,
-						     (i == txsize - 1));
+						     (i == DMA_TX_SIZE - 1));
 }
 
 /**
@@ -1031,8 +1030,6 @@ static int init_dma_desc_rings(struct net_device *dev, gfp_t flags)
 {
 	int i;
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int rxsize = priv->dma_rx_size;
 	unsigned int bfsize = 0;
 	int ret = -ENOMEM;
 
@@ -1044,10 +1041,6 @@ static int init_dma_desc_rings(struct net_device *dev, gfp_t flags)
 
 	priv->dma_buf_sz = bfsize;
 
-	if (netif_msg_probe(priv))
-		pr_debug("%s: txsize %d, rxsize %d, bfsize %d\n", __func__,
-			 txsize, rxsize, bfsize);
-
 	if (netif_msg_probe(priv)) {
 		pr_debug("(%s) dma_rx_phy=0x%08x dma_tx_phy=0x%08x\n", __func__,
 			 (u32) priv->dma_rx_phy, (u32) priv->dma_tx_phy);
@@ -1055,7 +1048,7 @@ static int init_dma_desc_rings(struct net_device *dev, gfp_t flags)
 		/* RX INITIALIZATION */
 		pr_debug("\tSKB addresses:\nskb\t\tskb data\tdma data\n");
 	}
-	for (i = 0; i < rxsize; i++) {
+	for (i = 0; i < DMA_RX_SIZE; i++) {
 		struct dma_desc *p;
 		if (priv->extend_desc)
 			p = &((priv->dma_erx + i)->basic);
@@ -1072,26 +1065,26 @@ static int init_dma_desc_rings(struct net_device *dev, gfp_t flags)
 				 (unsigned int)priv->rx_skbuff_dma[i]);
 	}
 	priv->cur_rx = 0;
-	priv->dirty_rx = (unsigned int)(i - rxsize);
+	priv->dirty_rx = (unsigned int)(i - DMA_RX_SIZE);
 	buf_sz = bfsize;
 
 	/* Setup the chained descriptor addresses */
 	if (priv->mode == STMMAC_CHAIN_MODE) {
 		if (priv->extend_desc) {
 			priv->hw->mode->init(priv->dma_erx, priv->dma_rx_phy,
-					     rxsize, 1);
+					     DMA_RX_SIZE, 1);
 			priv->hw->mode->init(priv->dma_etx, priv->dma_tx_phy,
-					     txsize, 1);
+					     DMA_TX_SIZE, 1);
 		} else {
 			priv->hw->mode->init(priv->dma_rx, priv->dma_rx_phy,
-					     rxsize, 0);
+					     DMA_RX_SIZE, 0);
 			priv->hw->mode->init(priv->dma_tx, priv->dma_tx_phy,
-					     txsize, 0);
+					     DMA_TX_SIZE, 0);
 		}
 	}
 
 	/* TX INITIALIZATION */
-	for (i = 0; i < txsize; i++) {
+	for (i = 0; i < DMA_TX_SIZE; i++) {
 		struct dma_desc *p;
 		if (priv->extend_desc)
 			p = &((priv->dma_etx + i)->basic);
@@ -1123,7 +1116,7 @@ static void dma_free_rx_skbufs(struct stmmac_priv *priv)
 {
 	int i;
 
-	for (i = 0; i < priv->dma_rx_size; i++)
+	for (i = 0; i < DMA_RX_SIZE; i++)
 		stmmac_free_rx_buffers(priv, i);
 }
 
@@ -1131,7 +1124,7 @@ static void dma_free_tx_skbufs(struct stmmac_priv *priv)
 {
 	int i;
 
-	for (i = 0; i < priv->dma_tx_size; i++) {
+	for (i = 0; i < DMA_TX_SIZE; i++) {
 		struct dma_desc *p;
 
 		if (priv->extend_desc)
@@ -1171,33 +1164,31 @@ static void dma_free_tx_skbufs(struct stmmac_priv *priv)
  */
 static int alloc_dma_desc_resources(struct stmmac_priv *priv)
 {
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int rxsize = priv->dma_rx_size;
 	int ret = -ENOMEM;
 
-	priv->rx_skbuff_dma = kmalloc_array(rxsize, sizeof(dma_addr_t),
+	priv->rx_skbuff_dma = kmalloc_array(DMA_RX_SIZE, sizeof(dma_addr_t),
 					    GFP_KERNEL);
 	if (!priv->rx_skbuff_dma)
 		return -ENOMEM;
 
-	priv->rx_skbuff = kmalloc_array(rxsize, sizeof(struct sk_buff *),
+	priv->rx_skbuff = kmalloc_array(DMA_RX_SIZE, sizeof(struct sk_buff *),
 					GFP_KERNEL);
 	if (!priv->rx_skbuff)
 		goto err_rx_skbuff;
 
-	priv->tx_skbuff_dma = kmalloc_array(txsize,
+	priv->tx_skbuff_dma = kmalloc_array(DMA_TX_SIZE,
 					    sizeof(*priv->tx_skbuff_dma),
 					    GFP_KERNEL);
 	if (!priv->tx_skbuff_dma)
 		goto err_tx_skbuff_dma;
 
-	priv->tx_skbuff = kmalloc_array(txsize, sizeof(struct sk_buff *),
+	priv->tx_skbuff = kmalloc_array(DMA_TX_SIZE, sizeof(struct sk_buff *),
 					GFP_KERNEL);
 	if (!priv->tx_skbuff)
 		goto err_tx_skbuff;
 
 	if (priv->extend_desc) {
-		priv->dma_erx = dma_zalloc_coherent(priv->device, rxsize *
+		priv->dma_erx = dma_zalloc_coherent(priv->device, DMA_RX_SIZE *
 						    sizeof(struct
 							   dma_extended_desc),
 						    &priv->dma_rx_phy,
@@ -1205,31 +1196,31 @@ static int alloc_dma_desc_resources(struct stmmac_priv *priv)
 		if (!priv->dma_erx)
 			goto err_dma;
 
-		priv->dma_etx = dma_zalloc_coherent(priv->device, txsize *
+		priv->dma_etx = dma_zalloc_coherent(priv->device, DMA_TX_SIZE *
 						    sizeof(struct
 							   dma_extended_desc),
 						    &priv->dma_tx_phy,
 						    GFP_KERNEL);
 		if (!priv->dma_etx) {
-			dma_free_coherent(priv->device, priv->dma_rx_size *
+			dma_free_coherent(priv->device, DMA_RX_SIZE *
 					  sizeof(struct dma_extended_desc),
 					  priv->dma_erx, priv->dma_rx_phy);
 			goto err_dma;
 		}
 	} else {
-		priv->dma_rx = dma_zalloc_coherent(priv->device, rxsize *
+		priv->dma_rx = dma_zalloc_coherent(priv->device, DMA_RX_SIZE *
 						   sizeof(struct dma_desc),
 						   &priv->dma_rx_phy,
 						   GFP_KERNEL);
 		if (!priv->dma_rx)
 			goto err_dma;
 
-		priv->dma_tx = dma_zalloc_coherent(priv->device, txsize *
+		priv->dma_tx = dma_zalloc_coherent(priv->device, DMA_TX_SIZE *
 						   sizeof(struct dma_desc),
 						   &priv->dma_tx_phy,
 						   GFP_KERNEL);
 		if (!priv->dma_tx) {
-			dma_free_coherent(priv->device, priv->dma_rx_size *
+			dma_free_coherent(priv->device, DMA_RX_SIZE *
 					  sizeof(struct dma_desc),
 					  priv->dma_rx, priv->dma_rx_phy);
 			goto err_dma;
@@ -1258,16 +1249,16 @@ static void free_dma_desc_resources(struct stmmac_priv *priv)
 	/* Free DMA regions of consistent memory previously allocated */
 	if (!priv->extend_desc) {
 		dma_free_coherent(priv->device,
-				  priv->dma_tx_size * sizeof(struct dma_desc),
+				  DMA_TX_SIZE * sizeof(struct dma_desc),
 				  priv->dma_tx, priv->dma_tx_phy);
 		dma_free_coherent(priv->device,
-				  priv->dma_rx_size * sizeof(struct dma_desc),
+				  DMA_RX_SIZE * sizeof(struct dma_desc),
 				  priv->dma_rx, priv->dma_rx_phy);
 	} else {
-		dma_free_coherent(priv->device, priv->dma_tx_size *
+		dma_free_coherent(priv->device, DMA_TX_SIZE *
 				  sizeof(struct dma_extended_desc),
 				  priv->dma_etx, priv->dma_tx_phy);
-		dma_free_coherent(priv->device, priv->dma_rx_size *
+		dma_free_coherent(priv->device, DMA_RX_SIZE *
 				  sizeof(struct dma_extended_desc),
 				  priv->dma_erx, priv->dma_rx_phy);
 	}
@@ -1312,16 +1303,15 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
  */
 static void stmmac_tx_clean(struct stmmac_priv *priv)
 {
-	unsigned int txsize = priv->dma_tx_size;
 	unsigned int bytes_compl = 0, pkts_compl = 0;
+	unsigned int entry = priv->dirty_tx;
 
 	spin_lock(&priv->tx_lock);
 
 	priv->xstats.tx_clean++;
 
-	while (priv->dirty_tx != priv->cur_tx) {
+	while (entry != priv->cur_tx) {
 		int last;
-		unsigned int entry = priv->dirty_tx % txsize;
 		struct sk_buff *skb = priv->tx_skbuff[entry];
 		struct dma_desc *p;
 
@@ -1378,16 +1368,17 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 
 		priv->hw->desc->release_tx_desc(p, priv->mode);
 
-		priv->dirty_tx++;
+		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
+		priv->dirty_tx = entry;
 	}
 
 	netdev_completed_queue(priv->dev, pkts_compl, bytes_compl);
 
 	if (unlikely(netif_queue_stopped(priv->dev) &&
-		     stmmac_tx_avail(priv) > STMMAC_TX_THRESH(priv))) {
+		     stmmac_tx_avail(priv) > STMMAC_TX_THRESH)) {
 		netif_tx_lock(priv->dev);
 		if (netif_queue_stopped(priv->dev) &&
-		    stmmac_tx_avail(priv) > STMMAC_TX_THRESH(priv)) {
+		    stmmac_tx_avail(priv) > STMMAC_TX_THRESH) {
 			if (netif_msg_tx_done(priv))
 				pr_debug("%s: restart transmit\n", __func__);
 			netif_wake_queue(priv->dev);
@@ -1421,20 +1412,19 @@ static inline void stmmac_disable_dma_irq(struct stmmac_priv *priv)
 static void stmmac_tx_err(struct stmmac_priv *priv)
 {
 	int i;
-	int txsize = priv->dma_tx_size;
 	netif_stop_queue(priv->dev);
 
 	priv->hw->dma->stop_tx(priv->ioaddr);
 	dma_free_tx_skbufs(priv);
-	for (i = 0; i < txsize; i++)
+	for (i = 0; i < DMA_TX_SIZE; i++)
 		if (priv->extend_desc)
 			priv->hw->desc->init_tx_desc(&priv->dma_etx[i].basic,
 						     priv->mode,
-						     (i == txsize - 1));
+						     (i == DMA_TX_SIZE - 1));
 		else
 			priv->hw->desc->init_tx_desc(&priv->dma_tx[i],
 						     priv->mode,
-						     (i == txsize - 1));
+						     (i == DMA_TX_SIZE - 1));
 	priv->dirty_tx = 0;
 	priv->cur_tx = 0;
 	netdev_reset_queue(priv->dev);
@@ -1811,9 +1801,6 @@ static int stmmac_open(struct net_device *dev)
 	memset(&priv->xstats, 0, sizeof(struct stmmac_extra_stats));
 	priv->xstats.threshold = tc;
 
-	/* Create and initialize the TX/RX descriptors chains. */
-	priv->dma_tx_size = STMMAC_ALIGN(dma_txsize);
-	priv->dma_rx_size = STMMAC_ALIGN(dma_rxsize);
 	priv->dma_buf_sz = STMMAC_ALIGN(buf_sz);
 
 	ret = alloc_dma_desc_resources(priv);
@@ -1955,7 +1942,6 @@ static int stmmac_release(struct net_device *dev)
 static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned int txsize = priv->dma_tx_size;
 	int entry;
 	int i, csum_insertion = 0, is_jumbo = 0;
 	int nfrags = skb_shinfo(skb)->nr_frags;
@@ -1978,7 +1964,8 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (priv->tx_path_in_lpi_mode)
 		stmmac_disable_eee_mode(priv);
 
-	entry = priv->cur_tx % txsize;
+	entry = priv->cur_tx;
+
 
 	csum_insertion = (skb->ip_summed == CHECKSUM_PARTIAL);
 
@@ -2013,7 +2000,8 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		int len = skb_frag_size(frag);
 
 		priv->tx_skbuff[entry] = NULL;
-		entry = (++priv->cur_tx) % txsize;
+		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
+
 		if (priv->extend_desc)
 			desc = (struct dma_desc *)(priv->dma_etx + entry);
 		else
@@ -2056,17 +2044,21 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	priv->hw->desc->set_tx_owner(first);
 	wmb();
 
-	priv->cur_tx++;
+	entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
+
+	priv->cur_tx = entry;
 
 	if (netif_msg_pktdata(priv)) {
 		pr_debug("%s: curr %d dirty=%d entry=%d, first=%p, nfrags=%d",
-			__func__, (priv->cur_tx % txsize),
-			(priv->dirty_tx % txsize), entry, first, nfrags);
+			__func__, (priv->cur_tx % DMA_TX_SIZE),
+			(priv->dirty_tx % DMA_TX_SIZE), entry, first, nfrags);
 
 		if (priv->extend_desc)
-			stmmac_display_ring((void *)priv->dma_etx, txsize, 1);
+			stmmac_display_ring((void *)priv->dma_etx,
+					    DMA_TX_SIZE, 1);
 		else
-			stmmac_display_ring((void *)priv->dma_tx, txsize, 0);
+			stmmac_display_ring((void *)priv->dma_tx,
+					    DMA_TX_SIZE, 0);
 
 		pr_debug(">>> frame to be transmitted: ");
 		print_pkt(skb->data, skb->len);
@@ -2128,11 +2120,11 @@ static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb)
  */
 static inline void stmmac_rx_refill(struct stmmac_priv *priv)
 {
-	unsigned int rxsize = priv->dma_rx_size;
 	int bfsize = priv->dma_buf_sz;
+	unsigned int entry = priv->dirty_rx;
+	int dirty = stmmac_rx_dirty(priv);
 
-	for (; priv->cur_rx - priv->dirty_rx > 0; priv->dirty_rx++) {
-		unsigned int entry = priv->dirty_rx % rxsize;
+	while (dirty-- > 0) {
 		struct dma_desc *p;
 
 		if (priv->extend_desc)
@@ -2168,7 +2160,10 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
 		wmb();
 		priv->hw->desc->set_rx_owner(p);
 		wmb();
+
+		entry = STMMAC_GET_ENTRY(entry, DMA_RX_SIZE);
 	}
+	priv->dirty_rx = entry;
 }
 
 /**
@@ -2180,8 +2175,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
  */
 static int stmmac_rx(struct stmmac_priv *priv, int limit)
 {
-	unsigned int rxsize = priv->dma_rx_size;
-	unsigned int entry = priv->cur_rx % rxsize;
+	unsigned int entry = priv->cur_rx;
 	unsigned int next_entry;
 	unsigned int count = 0;
 	int coe = priv->hw->rx_csum;
@@ -2189,9 +2183,11 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 	if (netif_msg_rx_status(priv)) {
 		pr_debug("%s: descriptor ring:\n", __func__);
 		if (priv->extend_desc)
-			stmmac_display_ring((void *)priv->dma_erx, rxsize, 1);
+			stmmac_display_ring((void *)priv->dma_erx,
+					    DMA_RX_SIZE, 1);
 		else
-			stmmac_display_ring((void *)priv->dma_rx, rxsize, 0);
+			stmmac_display_ring((void *)priv->dma_rx,
+					    DMA_RX_SIZE, 0);
 	}
 	while (count < limit) {
 		int status;
@@ -2207,7 +2203,9 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 
 		count++;
 
-		next_entry = (++priv->cur_rx) % rxsize;
+		priv->cur_rx = STMMAC_GET_ENTRY(priv->cur_rx, DMA_RX_SIZE);
+		next_entry = priv->cur_rx;
+
 		if (priv->extend_desc)
 			prefetch(priv->dma_erx + next_entry);
 		else
@@ -2567,19 +2565,17 @@ static int stmmac_sysfs_ring_read(struct seq_file *seq, void *v)
 {
 	struct net_device *dev = seq->private;
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned int txsize = priv->dma_tx_size;
-	unsigned int rxsize = priv->dma_rx_size;
 
 	if (priv->extend_desc) {
 		seq_printf(seq, "Extended RX descriptor ring:\n");
-		sysfs_display_ring((void *)priv->dma_erx, rxsize, 1, seq);
+		sysfs_display_ring((void *)priv->dma_erx, DMA_RX_SIZE, 1, seq);
 		seq_printf(seq, "Extended TX descriptor ring:\n");
-		sysfs_display_ring((void *)priv->dma_etx, txsize, 1, seq);
+		sysfs_display_ring((void *)priv->dma_etx, DMA_TX_SIZE, 1, seq);
 	} else {
 		seq_printf(seq, "RX descriptor ring:\n");
-		sysfs_display_ring((void *)priv->dma_rx, rxsize, 0, seq);
+		sysfs_display_ring((void *)priv->dma_rx, DMA_RX_SIZE, 0, seq);
 		seq_printf(seq, "TX descriptor ring:\n");
-		sysfs_display_ring((void *)priv->dma_tx, txsize, 0, seq);
+		sysfs_display_ring((void *)priv->dma_tx, DMA_TX_SIZE, 0, seq);
 	}
 
 	return 0;
@@ -3149,12 +3145,6 @@ static int __init stmmac_cmdline_opt(char *str)
 		} else if (!strncmp(opt, "phyaddr:", 8)) {
 			if (kstrtoint(opt + 8, 0, &phyaddr))
 				goto err;
-		} else if (!strncmp(opt, "dma_txsize:", 11)) {
-			if (kstrtoint(opt + 11, 0, &dma_txsize))
-				goto err;
-		} else if (!strncmp(opt, "dma_rxsize:", 11)) {
-			if (kstrtoint(opt + 11, 0, &dma_rxsize))
-				goto err;
 		} else if (!strncmp(opt, "buf_sz:", 7)) {
 			if (kstrtoint(opt + 7, 0, &buf_sz))
 				goto err;
-- 
cgit v0.10.2


From 553e2ab3130e9c3d234bee0c80a2a1e5057c8f9a Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:31 +0100
Subject: stmmac: add length field to dma data

Currently, the code pulls out the length field when
unmapping a buffer directly from the descriptor. This will result
in an uncached read to a dma_alloc_coherent() region. There is no
need to do this, so this patch simply puts the value directly into
a data structure which will hit the cache.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index 2763772..7fa7ab0d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -49,6 +49,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 	if (dma_mapping_error(priv->device, desc->des2))
 		return -1;
 	priv->tx_skbuff_dma[entry].buf = desc->des2;
+	priv->tx_skbuff_dma[entry].len = bmax;
 	priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum, STMMAC_CHAIN_MODE);
 
 	while (len != 0) {
@@ -63,6 +64,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			if (dma_mapping_error(priv->device, desc->des2))
 				return -1;
 			priv->tx_skbuff_dma[entry].buf = desc->des2;
+			priv->tx_skbuff_dma[entry].len = bmax;
 			priv->hw->desc->prepare_tx_desc(desc, 0, bmax, csum,
 							STMMAC_CHAIN_MODE);
 			priv->hw->desc->set_tx_owner(desc);
@@ -75,6 +77,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			if (dma_mapping_error(priv->device, desc->des2))
 				return -1;
 			priv->tx_skbuff_dma[entry].buf = desc->des2;
+			priv->tx_skbuff_dma[entry].len = len;
 			priv->hw->desc->prepare_tx_desc(desc, 0, len, csum,
 							STMMAC_CHAIN_MODE);
 			priv->hw->desc->set_tx_owner(desc);
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index 4358a87..cfc2f24 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -56,6 +56,8 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			return -1;
 
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
+		priv->tx_skbuff_dma[entry].len = bmax;
+
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum,
 						STMMAC_RING_MODE);
@@ -73,6 +75,8 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 		if (dma_mapping_error(priv->device, desc->des2))
 			return -1;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
+		priv->tx_skbuff_dma[entry].len = len;
+
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum,
 						STMMAC_RING_MODE);
@@ -84,6 +88,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 		if (dma_mapping_error(priv->device, desc->des2))
 			return -1;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
+		priv->tx_skbuff_dma[entry].len = nopaged_len;
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len, csum,
 						STMMAC_RING_MODE);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 7ae7c64..c497460 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -45,6 +45,7 @@ struct stmmac_resources {
 struct stmmac_tx_info {
 	dma_addr_t buf;
 	bool map_as_page;
+	unsigned len;
 };
 
 struct stmmac_priv {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index eb555f0..90a946f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1093,6 +1093,7 @@ static int init_dma_desc_rings(struct net_device *dev, gfp_t flags)
 		p->des2 = 0;
 		priv->tx_skbuff_dma[i].buf = 0;
 		priv->tx_skbuff_dma[i].map_as_page = false;
+		priv->tx_skbuff_dma[i].len = 0;
 		priv->tx_skbuff[i] = NULL;
 	}
 
@@ -1136,12 +1137,12 @@ static void dma_free_tx_skbufs(struct stmmac_priv *priv)
 			if (priv->tx_skbuff_dma[i].map_as_page)
 				dma_unmap_page(priv->device,
 					       priv->tx_skbuff_dma[i].buf,
-					       priv->hw->desc->get_tx_len(p),
+					       priv->tx_skbuff_dma[i].len,
 					       DMA_TO_DEVICE);
 			else
 				dma_unmap_single(priv->device,
 						 priv->tx_skbuff_dma[i].buf,
-						 priv->hw->desc->get_tx_len(p),
+						 priv->tx_skbuff_dma[i].len,
 						 DMA_TO_DEVICE);
 		}
 
@@ -1347,12 +1348,12 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 			if (priv->tx_skbuff_dma[entry].map_as_page)
 				dma_unmap_page(priv->device,
 					       priv->tx_skbuff_dma[entry].buf,
-					       priv->hw->desc->get_tx_len(p),
+					       priv->tx_skbuff_dma[entry].len,
 					       DMA_TO_DEVICE);
 			else
 				dma_unmap_single(priv->device,
 						 priv->tx_skbuff_dma[entry].buf,
-						 priv->hw->desc->get_tx_len(p),
+						 priv->tx_skbuff_dma[entry].len,
 						 DMA_TO_DEVICE);
 			priv->tx_skbuff_dma[entry].buf = 0;
 			priv->tx_skbuff_dma[entry].map_as_page = false;
@@ -1986,6 +1987,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		if (dma_mapping_error(priv->device, desc->des2))
 			goto dma_map_err;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
+		priv->tx_skbuff_dma[entry].len = nopaged_len;
 		priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len,
 						csum_insertion, priv->mode);
 	} else {
@@ -2014,6 +2016,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
 		priv->tx_skbuff_dma[entry].map_as_page = true;
+		priv->tx_skbuff_dma[entry].len = len;
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
 						priv->mode);
 		wmb();
-- 
cgit v0.10.2


From 2a6d8e172639b49a81afc76140dd7ee7c9a24da1 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:32 +0100
Subject: stmmac: add last_segment field to dma data

last_segment field is read twice from dma descriptors in stmmac_clean().
Add last_segment to dma data so that this flag is from priv
structure in cache instead of memory.
It avoids reading twice from memory for each loop in stmmac_clean().

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index 7fa7ab0d..355eafb 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -150,8 +150,9 @@ static void stmmac_refill_desc3(void *priv_ptr, struct dma_desc *p)
 static void stmmac_clean_desc3(void *priv_ptr, struct dma_desc *p)
 {
 	struct stmmac_priv *priv = (struct stmmac_priv *)priv_ptr;
+	unsigned int entry = priv->dirty_tx;
 
-	if (priv->hw->desc->get_tx_ls(p) && !priv->extend_desc)
+	if (priv->tx_skbuff_dma[entry].last_segment && !priv->extend_desc)
 		/* NOTE: Device will overwrite des3 with timestamp value if
 		 * 1588-2002 time stamping is enabled, hence reinitialize it
 		 * to keep explicit chaining in the descriptor.
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index c497460..0436918 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -46,6 +46,7 @@ struct stmmac_tx_info {
 	dma_addr_t buf;
 	bool map_as_page;
 	unsigned len;
+	bool last_segment;
 };
 
 struct stmmac_priv {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 90a946f..feae0de 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1094,6 +1094,7 @@ static int init_dma_desc_rings(struct net_device *dev, gfp_t flags)
 		priv->tx_skbuff_dma[i].buf = 0;
 		priv->tx_skbuff_dma[i].map_as_page = false;
 		priv->tx_skbuff_dma[i].len = 0;
+		priv->tx_skbuff_dma[i].last_segment = false;
 		priv->tx_skbuff[i] = NULL;
 	}
 
@@ -1326,7 +1327,7 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 			break;
 
 		/* Verify tx error by looking at the last segment. */
-		last = priv->hw->desc->get_tx_ls(p);
+		last = priv->tx_skbuff_dma[entry].last_segment;
 		if (likely(last)) {
 			int tx_error =
 			    priv->hw->desc->tx_status(&priv->dev->stats,
@@ -1359,6 +1360,7 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 			priv->tx_skbuff_dma[entry].map_as_page = false;
 		}
 		priv->hw->mode->clean_desc3(priv, p);
+		priv->tx_skbuff_dma[entry].last_segment = false;
 
 		if (likely(skb != NULL)) {
 			pkts_compl++;
@@ -2028,6 +2030,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/* Finalize the latest segment. */
 	priv->hw->desc->close_tx_desc(desc);
+	priv->tx_skbuff_dma[entry].last_segment = true;
 
 	wmb();
 	/* According to the coalesce parameter the IC bit for the latest
-- 
cgit v0.10.2


From 96951366ce8546662de56f58a3885b94326f9670 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:33 +0100
Subject: stmmac: add is_jumbo field to dma data

Optimize tx_clean by avoiding a des3 read in stmmac_clean_desc3().

In ring mode, TX, des3 seems only used when xmit a jumbo frame.
In case of normal descriptors, it may also be used for time
stamping.
Clean it in the above two case, without reading it.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index 355eafb..dacb654 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -152,7 +152,8 @@ static void stmmac_clean_desc3(void *priv_ptr, struct dma_desc *p)
 	struct stmmac_priv *priv = (struct stmmac_priv *)priv_ptr;
 	unsigned int entry = priv->dirty_tx;
 
-	if (priv->tx_skbuff_dma[entry].last_segment && !priv->extend_desc)
+	if (priv->tx_skbuff_dma[entry].last_segment && !priv->extend_desc &&
+	    priv->hwts_tx_en)
 		/* NOTE: Device will overwrite des3 with timestamp value if
 		 * 1588-2002 time stamping is enabled, hence reinitialize it
 		 * to keep explicit chaining in the descriptor.
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index cfc2f24..c648774 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -57,6 +57,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
 		priv->tx_skbuff_dma[entry].len = bmax;
+		priv->tx_skbuff_dma[entry].is_jumbo = true;
 
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum,
@@ -76,6 +77,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			return -1;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
 		priv->tx_skbuff_dma[entry].len = len;
+		priv->tx_skbuff_dma[entry].is_jumbo = true;
 
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum,
@@ -89,6 +91,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			return -1;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
 		priv->tx_skbuff_dma[entry].len = nopaged_len;
+		priv->tx_skbuff_dma[entry].is_jumbo = true;
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len, csum,
 						STMMAC_RING_MODE);
@@ -126,7 +129,13 @@ static void stmmac_init_desc3(struct dma_desc *p)
 
 static void stmmac_clean_desc3(void *priv_ptr, struct dma_desc *p)
 {
-	if (unlikely(p->des3))
+	struct stmmac_priv *priv = (struct stmmac_priv *)priv_ptr;
+	unsigned int entry = priv->dirty_tx;
+
+	/* des3 is only used for jumbo frames tx or time stamping */
+	if (unlikely(priv->tx_skbuff_dma[entry].is_jumbo ||
+		     (priv->tx_skbuff_dma[entry].last_segment &&
+		      !priv->extend_desc && priv->hwts_tx_en)))
 		p->des3 = 0;
 }
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 0436918..0d01f3e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -47,6 +47,7 @@ struct stmmac_tx_info {
 	bool map_as_page;
 	unsigned len;
 	bool last_segment;
+	bool is_jumbo;
 };
 
 struct stmmac_priv {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index feae0de..0194a8f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1361,6 +1361,7 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 		}
 		priv->hw->mode->clean_desc3(priv, p);
 		priv->tx_skbuff_dma[entry].last_segment = false;
+		priv->tx_skbuff_dma[entry].is_jumbo = false;
 
 		if (likely(skb != NULL)) {
 			pkts_compl++;
-- 
cgit v0.10.2


From c1fa3212be5503d802a5c4c451dd4e673fdc603a Mon Sep 17 00:00:00 2001
From: Fabrice Gasnier <fabrice.gasnier@st.com>
Date: Mon, 29 Feb 2016 14:27:34 +0100
Subject: stmmac: merge get_rx_owner into rx_status routine.

The RDES0 register can be read several times while doing RX of a
packet.
This patch slightly improves RX path performance by reading rdes0
once for two operation: check rx owner, get rx status bits.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Acked-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 09291af..3ba268e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -238,10 +238,11 @@ struct stmmac_extra_stats {
 
 /* Rx IPC status */
 enum rx_frame_status {
-	good_frame = 0,
-	discard_frame = 1,
-	csum_none = 2,
-	llc_snap = 4,
+	good_frame = 0x0,
+	discard_frame = 0x1,
+	csum_none = 0x2,
+	llc_snap = 0x4,
+	dma_own = 0x8,
 };
 
 enum dma_irq_status {
@@ -356,7 +357,6 @@ struct stmmac_desc_ops {
 	/* Get the buffer size from the descriptor */
 	int (*get_tx_len) (struct dma_desc *p);
 	/* Handle extra events on specific interrupts hw dependent */
-	int (*get_rx_owner) (struct dma_desc *p);
 	void (*set_rx_owner) (struct dma_desc *p);
 	/* Get the receive frame size */
 	int (*get_rx_frame_len) (struct dma_desc *p, int rx_coe_type);
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 716b807..1a2fce9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -186,6 +186,9 @@ static int enh_desc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 	unsigned int rdes0 = p->des0;
 	int ret = good_frame;
 
+	if (unlikely(rdes0 & RDES0_OWN))
+		return dma_own;
+
 	if (unlikely(rdes0 & RDES0_ERROR_SUMMARY)) {
 		if (unlikely(rdes0 & RDES0_DESCRIPTOR_ERROR)) {
 			x->rx_desc++;
@@ -272,11 +275,6 @@ static int enh_desc_get_tx_owner(struct dma_desc *p)
 	return (p->des0 & ETDES0_OWN) >> 31;
 }
 
-static int enh_desc_get_rx_owner(struct dma_desc *p)
-{
-	return (p->des0 & RDES0_OWN) >> 31;
-}
-
 static void enh_desc_set_tx_owner(struct dma_desc *p)
 {
 	p->des0 |= ETDES0_OWN;
@@ -402,7 +400,6 @@ const struct stmmac_desc_ops enh_desc_ops = {
 	.init_rx_desc = enh_desc_init_rx_desc,
 	.init_tx_desc = enh_desc_init_tx_desc,
 	.get_tx_owner = enh_desc_get_tx_owner,
-	.get_rx_owner = enh_desc_get_rx_owner,
 	.release_tx_desc = enh_desc_release_tx_desc,
 	.prepare_tx_desc = enh_desc_prepare_tx_desc,
 	.clear_tx_ic = enh_desc_clear_tx_ic,
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index 460c573..5a91932 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -82,6 +82,9 @@ static int ndesc_get_rx_status(void *data, struct stmmac_extra_stats *x,
 	unsigned int rdes0 = p->des0;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
 
+	if (unlikely(rdes0 & RDES0_OWN))
+		return dma_own;
+
 	if (unlikely(!(rdes0 & RDES0_LAST_DESCRIPTOR))) {
 		pr_warn("%s: Oversized frame spanned multiple buffers\n",
 			__func__);
@@ -155,11 +158,6 @@ static int ndesc_get_tx_owner(struct dma_desc *p)
 	return (p->des0 & TDES0_OWN) >> 31;
 }
 
-static int ndesc_get_rx_owner(struct dma_desc *p)
-{
-	return (p->des0 & RDES0_OWN) >> 31;
-}
-
 static void ndesc_set_tx_owner(struct dma_desc *p)
 {
 	p->des0 |= TDES0_OWN;
@@ -277,7 +275,6 @@ const struct stmmac_desc_ops ndesc_ops = {
 	.init_rx_desc = ndesc_init_rx_desc,
 	.init_tx_desc = ndesc_init_tx_desc,
 	.get_tx_owner = ndesc_get_tx_owner,
-	.get_rx_owner = ndesc_get_rx_owner,
 	.release_tx_desc = ndesc_release_tx_desc,
 	.prepare_tx_desc = ndesc_prepare_tx_desc,
 	.clear_tx_ic = ndesc_clear_tx_ic,
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 0194a8f..796d7c6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -2205,7 +2205,11 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 		else
 			p = priv->dma_rx + entry;
 
-		if (priv->hw->desc->get_rx_owner(p))
+		/* read the status of the incoming frame */
+		status = priv->hw->desc->rx_status(&priv->dev->stats,
+						   &priv->xstats, p);
+		/* check if managed by the DMA otherwise go ahead */
+		if (unlikely(status & dma_own))
 			break;
 
 		count++;
@@ -2218,9 +2222,6 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 		else
 			prefetch(priv->dma_rx + next_entry);
 
-		/* read the status of the incoming frame */
-		status = priv->hw->desc->rx_status(&priv->dev->stats,
-						   &priv->xstats, p);
 		if ((priv->extend_desc) && (priv->hw->desc->rx_extended_status))
 			priv->hw->desc->rx_extended_status(&priv->dev->stats,
 							   &priv->xstats,
-- 
cgit v0.10.2


From be434d5075d6be0cda996200b2a20035e1565215 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:35 +0100
Subject: stmmac: optimize tx desc management

This patch is to optimize the way to manage the TDES inside the
xmit function. When prepare the frame, some settings (e.g. OWN
bit) can be merged. This has been reworked to improve the tx
performances.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index dacb654..b3e669a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -50,7 +50,9 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 		return -1;
 	priv->tx_skbuff_dma[entry].buf = desc->des2;
 	priv->tx_skbuff_dma[entry].len = bmax;
-	priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum, STMMAC_CHAIN_MODE);
+	/* do not close the descriptor and do not set own bit */
+	priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum, STMMAC_CHAIN_MODE,
+					0, false);
 
 	while (len != 0) {
 		priv->tx_skbuff[entry] = NULL;
@@ -66,8 +68,8 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 			priv->tx_skbuff_dma[entry].buf = desc->des2;
 			priv->tx_skbuff_dma[entry].len = bmax;
 			priv->hw->desc->prepare_tx_desc(desc, 0, bmax, csum,
-							STMMAC_CHAIN_MODE);
-			priv->hw->desc->set_tx_owner(desc);
+							STMMAC_CHAIN_MODE, 1,
+							false);
 			len -= bmax;
 			i++;
 		} else {
@@ -78,9 +80,10 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 				return -1;
 			priv->tx_skbuff_dma[entry].buf = desc->des2;
 			priv->tx_skbuff_dma[entry].len = len;
+			/* last descriptor can be set now */
 			priv->hw->desc->prepare_tx_desc(desc, 0, len, csum,
-							STMMAC_CHAIN_MODE);
-			priv->hw->desc->set_tx_owner(desc);
+							STMMAC_CHAIN_MODE, 1,
+							true);
 			len = 0;
 		}
 	}
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 3ba268e..885c0f9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -338,12 +338,11 @@ struct stmmac_desc_ops {
 
 	/* Invoked by the xmit function to prepare the tx descriptor */
 	void (*prepare_tx_desc) (struct dma_desc *p, int is_fs, int len,
-				 int csum_flag, int mode);
+				 bool csum_flag, int mode, bool tx_own,
+				 bool ls_ic);
 	/* Set/get the owner of the descriptor */
 	void (*set_tx_owner) (struct dma_desc *p);
 	int (*get_tx_owner) (struct dma_desc *p);
-	/* Invoked by the xmit function to close the tx descriptor */
-	void (*close_tx_desc) (struct dma_desc *p);
 	/* Clean the tx descriptor as soon as the tx irq is received */
 	void (*release_tx_desc) (struct dma_desc *p, int mode);
 	/* Clear interrupt on tx frame completion. When this bit is
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 1a2fce9..1abd80e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -302,7 +302,8 @@ static void enh_desc_release_tx_desc(struct dma_desc *p, int mode)
 }
 
 static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
-				     int csum_flag, int mode)
+				     bool csum_flag, int mode, bool tx_own,
+				     bool ls_ic)
 {
 	unsigned int tdes0 = p->des0;
 
@@ -316,6 +317,19 @@ static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 	else
 		tdes0 &= ~(TX_CIC_FULL << ETDES0_CHECKSUM_INSERTION_SHIFT);
 
+	if (tx_own)
+		tdes0 |= ETDES0_OWN;
+
+	if (is_fs & tx_own)
+		/* When the own bit, for the first frame, has to be set, all
+		 * descriptors for the same frame has to be set before, to
+		 * avoid race condition.
+		 */
+		wmb();
+
+	if (ls_ic)
+		tdes0 |= ETDES0_LAST_SEGMENT | ETDES0_INTERRUPT;
+
 	p->des0 = tdes0;
 
 	if (mode == STMMAC_CHAIN_MODE)
@@ -329,11 +343,6 @@ static void enh_desc_clear_tx_ic(struct dma_desc *p)
 	p->des0 &= ~ETDES0_INTERRUPT;
 }
 
-static void enh_desc_close_tx_desc(struct dma_desc *p)
-{
-	p->des0 |= ETDES0_LAST_SEGMENT | ETDES0_INTERRUPT;
-}
-
 static int enh_desc_get_rx_frame_len(struct dma_desc *p, int rx_coe_type)
 {
 	unsigned int csum = 0;
@@ -403,7 +412,6 @@ const struct stmmac_desc_ops enh_desc_ops = {
 	.release_tx_desc = enh_desc_release_tx_desc,
 	.prepare_tx_desc = enh_desc_prepare_tx_desc,
 	.clear_tx_ic = enh_desc_clear_tx_ic,
-	.close_tx_desc = enh_desc_close_tx_desc,
 	.get_tx_ls = enh_desc_get_tx_ls,
 	.set_tx_owner = enh_desc_set_tx_owner,
 	.set_rx_owner = enh_desc_set_rx_owner,
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index 5a91932..19cc12d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -185,7 +185,8 @@ static void ndesc_release_tx_desc(struct dma_desc *p, int mode)
 }
 
 static void ndesc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
-				  int csum_flag, int mode)
+				  bool csum_flag, int mode, bool tx_own,
+				  bool ls_ic)
 {
 	unsigned int tdes1 = p->des1;
 
@@ -199,6 +200,12 @@ static void ndesc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 	else
 		tdes1 &= ~(TX_CIC_FULL << TDES1_CHECKSUM_INSERTION_SHIFT);
 
+	if (tx_own)
+		tdes1 |= TDES0_OWN;
+
+	if (ls_ic)
+		tdes1 |= TDES1_LAST_SEGMENT | TDES1_INTERRUPT;
+
 	p->des1 = tdes1;
 
 	if (mode == STMMAC_CHAIN_MODE)
@@ -212,11 +219,6 @@ static void ndesc_clear_tx_ic(struct dma_desc *p)
 	p->des1 &= ~TDES1_INTERRUPT;
 }
 
-static void ndesc_close_tx_desc(struct dma_desc *p)
-{
-	p->des1 |= TDES1_LAST_SEGMENT | TDES1_INTERRUPT;
-}
-
 static int ndesc_get_rx_frame_len(struct dma_desc *p, int rx_coe_type)
 {
 	unsigned int csum = 0;
@@ -278,7 +280,6 @@ const struct stmmac_desc_ops ndesc_ops = {
 	.release_tx_desc = ndesc_release_tx_desc,
 	.prepare_tx_desc = ndesc_prepare_tx_desc,
 	.clear_tx_ic = ndesc_clear_tx_ic,
-	.close_tx_desc = ndesc_close_tx_desc,
 	.get_tx_ls = ndesc_get_tx_ls,
 	.set_tx_owner = ndesc_set_tx_owner,
 	.set_rx_owner = ndesc_set_rx_owner,
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index c648774..11c7164 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -61,7 +61,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum,
-						STMMAC_RING_MODE);
+						STMMAC_RING_MODE, 0, false);
 		wmb();
 		priv->tx_skbuff[entry] = NULL;
 		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
@@ -81,9 +81,8 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum,
-						STMMAC_RING_MODE);
+						STMMAC_RING_MODE, 1, true);
 		wmb();
-		priv->hw->desc->set_tx_owner(desc);
 	} else {
 		desc->des2 = dma_map_single(priv->device, skb->data,
 					    nopaged_len, DMA_TO_DEVICE);
@@ -94,7 +93,7 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 		priv->tx_skbuff_dma[entry].is_jumbo = true;
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len, csum,
-						STMMAC_RING_MODE);
+						STMMAC_RING_MODE, 0, true);
 	}
 
 	priv->cur_tx = entry;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 796d7c6..24c3608 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1991,8 +1991,10 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 			goto dma_map_err;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
 		priv->tx_skbuff_dma[entry].len = nopaged_len;
+		/* do not set the own at this stage */
 		priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len,
-						csum_insertion, priv->mode);
+						csum_insertion, priv->mode, 0,
+						nfrags == 0);
 	} else {
 		desc = first;
 		entry = priv->hw->mode->jumbo_frm(priv, skb, csum_insertion);
@@ -2003,6 +2005,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 	for (i = 0; i < nfrags; i++) {
 		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		int len = skb_frag_size(frag);
+		bool last_segment = (i == (nfrags - 1));
 
 		priv->tx_skbuff[entry] = NULL;
 		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
@@ -2021,19 +2024,12 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		priv->tx_skbuff_dma[entry].map_as_page = true;
 		priv->tx_skbuff_dma[entry].len = len;
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
-						priv->mode);
-		wmb();
-		priv->hw->desc->set_tx_owner(desc);
-		wmb();
+						priv->mode, 1, last_segment);
+		priv->tx_skbuff_dma[entry].last_segment = last_segment;
 	}
 
 	priv->tx_skbuff[entry] = skb;
 
-	/* Finalize the latest segment. */
-	priv->hw->desc->close_tx_desc(desc);
-	priv->tx_skbuff_dma[entry].last_segment = true;
-
-	wmb();
 	/* According to the coalesce parameter the IC bit for the latest
 	 * segment could be reset and the timer re-started to invoke the
 	 * stmmac_tx function. This approach takes care about the fragments.
-- 
cgit v0.10.2


From c363b6586cd424431e84d921267e101ec67207f5 Mon Sep 17 00:00:00 2001
From: Fabrice Gasnier <fabrice.gasnier@st.com>
Date: Mon, 29 Feb 2016 14:27:36 +0100
Subject: stmmac: optimize tx clean function

This patch "inline" get_tx_owner and get_ls routines. It Results in a
unique read to tdes0, instead of three, to check TX_OWN and LS bits,
and other status bits.

It helps improve driver TX path by removing two uncached read/writes
inside TX clean loop for enhanced descriptors but not for normal ones
because the des1 must be read in any case.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Acked-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 885c0f9..7ccb147 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -245,6 +245,14 @@ enum rx_frame_status {
 	dma_own = 0x8,
 };
 
+/* Tx status */
+enum tx_frame_status {
+	tx_done = 0x0,
+	tx_not_ls = 0x1,
+	tx_err = 0x2,
+	tx_dma_own = 0x4,
+};
+
 enum dma_irq_status {
 	tx_hard_error = 0x1,
 	tx_hard_error_bump_tc = 0x2,
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 1abd80e..957610b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -31,7 +31,15 @@ static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 {
 	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int tdes0 = p->des0;
-	int ret = 0;
+	int ret = tx_done;
+
+	/* Get tx owner first */
+	if (unlikely(tdes0 & ETDES0_OWN))
+		return tx_dma_own;
+
+	/* Verify tx error by looking at the last segment. */
+	if (likely(!(tdes0 & ETDES0_LAST_SEGMENT)))
+		return tx_not_ls;
 
 	if (unlikely(tdes0 & ETDES0_ERROR_SUMMARY)) {
 		if (unlikely(tdes0 & ETDES0_JABBER_TIMEOUT))
@@ -71,7 +79,7 @@ static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 			dwmac_dma_flush_tx_fifo(ioaddr);
 		}
 
-		ret = -1;
+		ret = tx_err;
 	}
 
 	if (unlikely(tdes0 & ETDES0_DEFERRED))
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index 19cc12d..122fb5a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -31,7 +31,16 @@ static int ndesc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 {
 	struct net_device_stats *stats = (struct net_device_stats *)data;
 	unsigned int tdes0 = p->des0;
-	int ret = 0;
+	unsigned int tdes1 = p->des1;
+	int ret = tx_done;
+
+	/* Get tx owner first */
+	if (unlikely(tdes0 & TDES0_OWN))
+		return tx_dma_own;
+
+	/* Verify tx error by looking at the last segment. */
+	if (likely(!(tdes1 & TDES1_LAST_SEGMENT)))
+		return tx_not_ls;
 
 	if (unlikely(tdes0 & TDES0_ERROR_SUMMARY)) {
 		if (unlikely(tdes0 & TDES0_UNDERFLOW_ERROR)) {
@@ -54,7 +63,7 @@ static int ndesc_get_tx_status(void *data, struct stmmac_extra_stats *x,
 			collisions = (tdes0 & TDES0_COLLISION_COUNT_MASK) >> 3;
 			stats->collisions += collisions;
 		}
-		ret = -1;
+		ret = tx_err;
 	}
 
 	if (tdes0 & TDES0_VLAN_FRAME)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 24c3608..d31179f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1313,32 +1313,31 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 	priv->xstats.tx_clean++;
 
 	while (entry != priv->cur_tx) {
-		int last;
 		struct sk_buff *skb = priv->tx_skbuff[entry];
 		struct dma_desc *p;
+		int status;
 
 		if (priv->extend_desc)
 			p = (struct dma_desc *)(priv->dma_etx + entry);
 		else
 			p = priv->dma_tx + entry;
 
-		/* Check if the descriptor is owned by the DMA. */
-		if (priv->hw->desc->get_tx_owner(p))
-			break;
-
-		/* Verify tx error by looking at the last segment. */
-		last = priv->tx_skbuff_dma[entry].last_segment;
-		if (likely(last)) {
-			int tx_error =
-			    priv->hw->desc->tx_status(&priv->dev->stats,
+		status = priv->hw->desc->tx_status(&priv->dev->stats,
 						      &priv->xstats, p,
 						      priv->ioaddr);
-			if (likely(tx_error == 0)) {
+		/* Check if the descriptor is owned by the DMA */
+		if (unlikely(status & tx_dma_own))
+			break;
+
+		/* Just consider the last segment and ...*/
+		if (likely(!(status & tx_not_ls))) {
+			/* ... verify the status error condition */
+			if (unlikely(status & tx_err)) {
+				priv->dev->stats.tx_errors++;
+			} else {
 				priv->dev->stats.tx_packets++;
 				priv->xstats.tx_pkt_n++;
-			} else
-				priv->dev->stats.tx_errors++;
-
+			}
 			stmmac_get_tx_hwtstamp(priv, entry, skb);
 		}
 		if (netif_msg_tx_done(priv))
-- 
cgit v0.10.2


From fbc80823a93c57f2310dacfabfea6c76424552b6 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:37 +0100
Subject: stmmac: set dirty index out of the loop

The dirty index can be updated out of the loop where all the
tx resources are claimed. This will help on performances too.
Also a useless debug printk has been removed from the main loop.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index d31179f..2e4c10a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1340,9 +1340,6 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 			}
 			stmmac_get_tx_hwtstamp(priv, entry, skb);
 		}
-		if (netif_msg_tx_done(priv))
-			pr_debug("%s: curr %d, dirty %d\n", __func__,
-				 priv->cur_tx, priv->dirty_tx);
 
 		if (likely(priv->tx_skbuff_dma[entry].buf)) {
 			if (priv->tx_skbuff_dma[entry].map_as_page)
@@ -1372,8 +1369,8 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
 		priv->hw->desc->release_tx_desc(p, priv->mode);
 
 		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
-		priv->dirty_tx = entry;
 	}
+	priv->dirty_tx = entry;
 
 	netdev_completed_queue(priv->dev, pkts_compl, bytes_compl);
 
-- 
cgit v0.10.2


From 0e80bdc9a72df3b31a9fc2012102a6cc8d664e93 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:38 +0100
Subject: stmmac: first frame prep at the end of xmit routine

This patch is to fill the first descriptor just before granting
the DMA engine so at the end of the xmit.
The patch takes care about the algorithm adopted to mitigate the
interrupts, then it fixes the last segment in case of no fragments.
Moreover, this new implementation does not pass any "ter" field when
prepare the descriptors because this is not necessary.
The patch also details the memory barrier in the xmit.

As final results, this patch guarantees the same performances
but fixing a case if small datagram are sent. In fact, this
kind of test is impacted if no coalesce is done.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 7ccb147..f96d257 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -100,7 +100,7 @@ struct stmmac_extra_stats {
 	unsigned long napi_poll;
 	unsigned long tx_normal_irq_n;
 	unsigned long tx_clean;
-	unsigned long tx_reset_ic_bit;
+	unsigned long tx_set_ic_bit;
 	unsigned long irq_receive_pmt_irq_n;
 	/* MMC info */
 	unsigned long mmc_tx_irq_n;
@@ -347,7 +347,7 @@ struct stmmac_desc_ops {
 	/* Invoked by the xmit function to prepare the tx descriptor */
 	void (*prepare_tx_desc) (struct dma_desc *p, int is_fs, int len,
 				 bool csum_flag, int mode, bool tx_own,
-				 bool ls_ic);
+				 bool ls);
 	/* Set/get the owner of the descriptor */
 	void (*set_tx_owner) (struct dma_desc *p);
 	int (*get_tx_owner) (struct dma_desc *p);
@@ -355,7 +355,7 @@ struct stmmac_desc_ops {
 	void (*release_tx_desc) (struct dma_desc *p, int mode);
 	/* Clear interrupt on tx frame completion. When this bit is
 	 * set an interrupt happens as soon as the frame is transmitted */
-	void (*clear_tx_ic) (struct dma_desc *p);
+	void (*set_tx_ic)(struct dma_desc *p);
 	/* Last tx segment reports the transmit status */
 	int (*get_tx_ls) (struct dma_desc *p);
 	/* Return the transmit status looking at the TDES1 */
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 957610b..cfb018c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -311,10 +311,15 @@ static void enh_desc_release_tx_desc(struct dma_desc *p, int mode)
 
 static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 				     bool csum_flag, int mode, bool tx_own,
-				     bool ls_ic)
+				     bool ls)
 {
 	unsigned int tdes0 = p->des0;
 
+	if (mode == STMMAC_CHAIN_MODE)
+		enh_set_tx_desc_len_on_chain(p, len);
+	else
+		enh_set_tx_desc_len_on_ring(p, len);
+
 	if (is_fs)
 		tdes0 |= ETDES0_FIRST_SEGMENT;
 	else
@@ -325,6 +330,10 @@ static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 	else
 		tdes0 &= ~(TX_CIC_FULL << ETDES0_CHECKSUM_INSERTION_SHIFT);
 
+	if (ls)
+		tdes0 |= ETDES0_LAST_SEGMENT;
+
+	/* Finally set the OWN bit. Later the DMA will start! */
 	if (tx_own)
 		tdes0 |= ETDES0_OWN;
 
@@ -335,20 +344,12 @@ static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 		 */
 		wmb();
 
-	if (ls_ic)
-		tdes0 |= ETDES0_LAST_SEGMENT | ETDES0_INTERRUPT;
-
 	p->des0 = tdes0;
-
-	if (mode == STMMAC_CHAIN_MODE)
-		enh_set_tx_desc_len_on_chain(p, len);
-	else
-		enh_set_tx_desc_len_on_ring(p, len);
 }
 
-static void enh_desc_clear_tx_ic(struct dma_desc *p)
+static void enh_desc_set_tx_ic(struct dma_desc *p)
 {
-	p->des0 &= ~ETDES0_INTERRUPT;
+	p->des0 |= ETDES0_INTERRUPT;
 }
 
 static int enh_desc_get_rx_frame_len(struct dma_desc *p, int rx_coe_type)
@@ -419,7 +420,7 @@ const struct stmmac_desc_ops enh_desc_ops = {
 	.get_tx_owner = enh_desc_get_tx_owner,
 	.release_tx_desc = enh_desc_release_tx_desc,
 	.prepare_tx_desc = enh_desc_prepare_tx_desc,
-	.clear_tx_ic = enh_desc_clear_tx_ic,
+	.set_tx_ic = enh_desc_set_tx_ic,
 	.get_tx_ls = enh_desc_get_tx_ls,
 	.set_tx_owner = enh_desc_set_tx_owner,
 	.set_rx_owner = enh_desc_set_rx_owner,
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index 122fb5a..e13228f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -195,10 +195,15 @@ static void ndesc_release_tx_desc(struct dma_desc *p, int mode)
 
 static void ndesc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 				  bool csum_flag, int mode, bool tx_own,
-				  bool ls_ic)
+				  bool ls)
 {
 	unsigned int tdes1 = p->des1;
 
+	if (mode == STMMAC_CHAIN_MODE)
+		norm_set_tx_desc_len_on_chain(p, len);
+	else
+		norm_set_tx_desc_len_on_ring(p, len);
+
 	if (is_fs)
 		tdes1 |= TDES1_FIRST_SEGMENT;
 	else
@@ -209,23 +214,18 @@ static void ndesc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
 	else
 		tdes1 &= ~(TX_CIC_FULL << TDES1_CHECKSUM_INSERTION_SHIFT);
 
+	if (ls)
+		tdes1 |= TDES1_LAST_SEGMENT;
+
 	if (tx_own)
 		tdes1 |= TDES0_OWN;
 
-	if (ls_ic)
-		tdes1 |= TDES1_LAST_SEGMENT | TDES1_INTERRUPT;
-
 	p->des1 = tdes1;
-
-	if (mode == STMMAC_CHAIN_MODE)
-		norm_set_tx_desc_len_on_chain(p, len);
-	else
-		norm_set_tx_desc_len_on_ring(p, len);
 }
 
-static void ndesc_clear_tx_ic(struct dma_desc *p)
+static void ndesc_set_tx_ic(struct dma_desc *p)
 {
-	p->des1 &= ~TDES1_INTERRUPT;
+	p->des1 |= TDES1_INTERRUPT;
 }
 
 static int ndesc_get_rx_frame_len(struct dma_desc *p, int rx_coe_type)
@@ -288,7 +288,7 @@ const struct stmmac_desc_ops ndesc_ops = {
 	.get_tx_owner = ndesc_get_tx_owner,
 	.release_tx_desc = ndesc_release_tx_desc,
 	.prepare_tx_desc = ndesc_prepare_tx_desc,
-	.clear_tx_ic = ndesc_clear_tx_ic,
+	.set_tx_ic = ndesc_set_tx_ic,
 	.get_tx_ls = ndesc_get_tx_ls,
 	.set_tx_owner = ndesc_set_tx_owner,
 	.set_rx_owner = ndesc_set_rx_owner,
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index 11c7164..7723b5d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -62,7 +62,6 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 1, bmax, csum,
 						STMMAC_RING_MODE, 0, false);
-		wmb();
 		priv->tx_skbuff[entry] = NULL;
 		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
 
@@ -82,7 +81,6 @@ static int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
 		desc->des3 = desc->des2 + BUF_SIZE_4KiB;
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum,
 						STMMAC_RING_MODE, 1, true);
-		wmb();
 	} else {
 		desc->des2 = dma_map_single(priv->device, skb->data,
 					    nopaged_len, DMA_TO_DEVICE);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 4c6486c..c803d4c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -97,7 +97,7 @@ static const struct stmmac_stats stmmac_gstrings_stats[] = {
 	STMMAC_STAT(napi_poll),
 	STMMAC_STAT(tx_normal_irq_n),
 	STMMAC_STAT(tx_clean),
-	STMMAC_STAT(tx_reset_ic_bit),
+	STMMAC_STAT(tx_set_ic_bit),
 	STMMAC_STAT(irq_receive_pmt_irq_n),
 	/* MMC info */
 	STMMAC_STAT(mmc_tx_irq_n),
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 2e4c10a..90b2612 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1942,12 +1942,12 @@ static int stmmac_release(struct net_device *dev)
 static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
-	int entry;
+	unsigned int nopaged_len = skb_headlen(skb);
 	int i, csum_insertion = 0, is_jumbo = 0;
 	int nfrags = skb_shinfo(skb)->nr_frags;
+	unsigned int entry, first_entry;
 	struct dma_desc *desc, *first;
-	unsigned int nopaged_len = skb_headlen(skb);
-	unsigned int enh_desc = priv->plat->enh_desc;
+	unsigned int enh_desc;
 
 	spin_lock(&priv->tx_lock);
 
@@ -1965,34 +1965,25 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		stmmac_disable_eee_mode(priv);
 
 	entry = priv->cur_tx;
-
+	first_entry = entry;
 
 	csum_insertion = (skb->ip_summed == CHECKSUM_PARTIAL);
 
-	if (priv->extend_desc)
+	if (likely(priv->extend_desc))
 		desc = (struct dma_desc *)(priv->dma_etx + entry);
 	else
 		desc = priv->dma_tx + entry;
 
 	first = desc;
 
+	priv->tx_skbuff[first_entry] = skb;
+
+	enh_desc = priv->plat->enh_desc;
 	/* To program the descriptors according to the size of the frame */
 	if (enh_desc)
 		is_jumbo = priv->hw->mode->is_jumbo_frm(skb->len, enh_desc);
 
-	if (likely(!is_jumbo)) {
-		desc->des2 = dma_map_single(priv->device, skb->data,
-					    nopaged_len, DMA_TO_DEVICE);
-		if (dma_mapping_error(priv->device, desc->des2))
-			goto dma_map_err;
-		priv->tx_skbuff_dma[entry].buf = desc->des2;
-		priv->tx_skbuff_dma[entry].len = nopaged_len;
-		/* do not set the own at this stage */
-		priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len,
-						csum_insertion, priv->mode, 0,
-						nfrags == 0);
-	} else {
-		desc = first;
+	if (unlikely(is_jumbo)) {
 		entry = priv->hw->mode->jumbo_frm(priv, skb, csum_insertion);
 		if (unlikely(entry < 0))
 			goto dma_map_err;
@@ -2003,10 +1994,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		int len = skb_frag_size(frag);
 		bool last_segment = (i == (nfrags - 1));
 
-		priv->tx_skbuff[entry] = NULL;
 		entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
 
-		if (priv->extend_desc)
+		if (likely(priv->extend_desc))
 			desc = (struct dma_desc *)(priv->dma_etx + entry);
 		else
 			desc = priv->dma_tx + entry;
@@ -2016,41 +2006,25 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		if (dma_mapping_error(priv->device, desc->des2))
 			goto dma_map_err; /* should reuse desc w/o issues */
 
+		priv->tx_skbuff[entry] = NULL;
 		priv->tx_skbuff_dma[entry].buf = desc->des2;
 		priv->tx_skbuff_dma[entry].map_as_page = true;
 		priv->tx_skbuff_dma[entry].len = len;
+		priv->tx_skbuff_dma[entry].last_segment = last_segment;
+
+		/* Prepare the descriptor and set the own bit too */
 		priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
 						priv->mode, 1, last_segment);
-		priv->tx_skbuff_dma[entry].last_segment = last_segment;
 	}
 
-	priv->tx_skbuff[entry] = skb;
-
-	/* According to the coalesce parameter the IC bit for the latest
-	 * segment could be reset and the timer re-started to invoke the
-	 * stmmac_tx function. This approach takes care about the fragments.
-	 */
-	priv->tx_count_frames += nfrags + 1;
-	if (priv->tx_coal_frames > priv->tx_count_frames) {
-		priv->hw->desc->clear_tx_ic(desc);
-		priv->xstats.tx_reset_ic_bit++;
-		mod_timer(&priv->txtimer,
-			  STMMAC_COAL_TIMER(priv->tx_coal_timer));
-	} else
-		priv->tx_count_frames = 0;
-
-	/* To avoid raise condition */
-	priv->hw->desc->set_tx_owner(first);
-	wmb();
-
 	entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE);
 
 	priv->cur_tx = entry;
 
 	if (netif_msg_pktdata(priv)) {
-		pr_debug("%s: curr %d dirty=%d entry=%d, first=%p, nfrags=%d",
-			__func__, (priv->cur_tx % DMA_TX_SIZE),
-			(priv->dirty_tx % DMA_TX_SIZE), entry, first, nfrags);
+		pr_debug("%s: curr=%d dirty=%d f=%d, e=%d, first=%p, nfrags=%d",
+			 __func__, priv->cur_tx, priv->dirty_tx, first_entry,
+			 entry, first, nfrags);
 
 		if (priv->extend_desc)
 			stmmac_display_ring((void *)priv->dma_etx,
@@ -2062,6 +2036,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		pr_debug(">>> frame to be transmitted: ");
 		print_pkt(skb->data, skb->len);
 	}
+
 	if (unlikely(stmmac_tx_avail(priv) <= (MAX_SKB_FRAGS + 1))) {
 		if (netif_msg_hw(priv))
 			pr_debug("%s: stop transmitted packets\n", __func__);
@@ -2070,16 +2045,59 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	dev->stats.tx_bytes += skb->len;
 
-	if (unlikely((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
-		     priv->hwts_tx_en)) {
-		/* declare that device is doing timestamping */
-		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
-		priv->hw->desc->enable_tx_timestamp(first);
+	/* According to the coalesce parameter the IC bit for the latest
+	 * segment is reset and the timer re-started to clean the tx status.
+	 * This approach takes care about the fragments: desc is the first
+	 * element in case of no SG.
+	 */
+	priv->tx_count_frames += nfrags + 1;
+	if (likely(priv->tx_coal_frames > priv->tx_count_frames)) {
+		mod_timer(&priv->txtimer,
+			  STMMAC_COAL_TIMER(priv->tx_coal_timer));
+	} else {
+		priv->tx_count_frames = 0;
+		priv->hw->desc->set_tx_ic(desc);
+		priv->xstats.tx_set_ic_bit++;
 	}
 
 	if (!priv->hwts_tx_en)
 		skb_tx_timestamp(skb);
 
+	/* Ready to fill the first descriptor and set the OWN bit w/o any
+	 * problems because all the descriptors are actually ready to be
+	 * passed to the DMA engine.
+	 */
+	if (likely(!is_jumbo)) {
+		bool last_segment = (nfrags == 0);
+
+		first->des2 = dma_map_single(priv->device, skb->data,
+					     nopaged_len, DMA_TO_DEVICE);
+		if (dma_mapping_error(priv->device, first->des2))
+			goto dma_map_err;
+
+		priv->tx_skbuff_dma[first_entry].buf = first->des2;
+		priv->tx_skbuff_dma[first_entry].len = nopaged_len;
+		priv->tx_skbuff_dma[first_entry].last_segment = last_segment;
+
+		if (unlikely((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+			     priv->hwts_tx_en)) {
+			/* declare that device is doing timestamping */
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+			priv->hw->desc->enable_tx_timestamp(first);
+		}
+
+		/* Prepare the first descriptor setting the OWN bit too */
+		priv->hw->desc->prepare_tx_desc(first, 1, nopaged_len,
+						csum_insertion, priv->mode, 1,
+						last_segment);
+
+		/* The own bit must be the latest setting done when prepare the
+		 * descriptor and then barrier is needed to make sure that
+		 * all is coherent before granting the DMA engine.
+		 */
+		smp_wmb();
+	}
+
 	netdev_sent_queue(dev, skb->len);
 	priv->hw->dma->enable_dma_transmission(priv->ioaddr);
 
-- 
cgit v0.10.2


From 8e99fc5f887a072ee023eb9a3240eba85e768734 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:39 +0100
Subject: stmmac: do not poll phy handler when attach a switch

This patch avoids to call the stmmac_adjust_link when
the driver is connected to a switch by using the FIXED_PHY
support. Prior this patch the phydev->irq was set as PHY_POLL
so periodically the phy handler was invoked spending useless
time because the link cannot actually change.
Note that the stmmac_adjust_link will be called just one
time and this guarantees that the ST glue logic will be
setup according to the mode and speed fixed.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 90b2612..eab7ac0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -866,6 +866,11 @@ static int stmmac_init_phy(struct net_device *dev)
 		phy_disconnect(phydev);
 		return -ENODEV;
 	}
+
+	/* If attached to a switch, there is no reason to poll phy handler */
+	if (!strcmp(priv->plat->phy_bus_name, "fixed"))
+		phydev->irq = PHY_IGNORE_INTERRUPT;
+
 	pr_debug("stmmac_init_phy:  %s: attached to PHY (UID 0x%x)"
 		 " Link = %d\n", dev->name, phydev->phy_id, phydev->link);
 
-- 
cgit v0.10.2


From 8ecd80a5f6a7e31824c0d49e10c3cf6825c6c291 Mon Sep 17 00:00:00 2001
From: Fabrice Gasnier <fabrice.gasnier@st.com>
Date: Mon, 29 Feb 2016 14:27:40 +0100
Subject: stmmac: fix phy init when attached to a phy

phy_bus_name can be NULL when "fixed-link" property isn't used.
Then, since "stmmac: do not poll phy handler when attach a switch",
phy_bus_name ptr needs to be checked before strcmp is called.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index eab7ac0..3cc1355 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -868,8 +868,9 @@ static int stmmac_init_phy(struct net_device *dev)
 	}
 
 	/* If attached to a switch, there is no reason to poll phy handler */
-	if (!strcmp(priv->plat->phy_bus_name, "fixed"))
-		phydev->irq = PHY_IGNORE_INTERRUPT;
+	if (priv->plat->phy_bus_name)
+		if (!strcmp(priv->plat->phy_bus_name, "fixed"))
+			phydev->irq = PHY_IGNORE_INTERRUPT;
 
 	pr_debug("stmmac_init_phy:  %s: attached to PHY (UID 0x%x)"
 		 " Link = %d\n", dev->name, phydev->phy_id, phydev->link);
-- 
cgit v0.10.2


From 22ad38381547950c64ab8590ebf04a5d3693cbb7 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:41 +0100
Subject: stmmac: do not perform zero-copy for rx frames

This patch is to allow this driver to copy tiny frames during the reception
process. This is giving more stability while stressing the driver on STi
embedded systems.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 0d01f3e..221f5cd 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -74,6 +74,7 @@ struct stmmac_priv {
 	unsigned int cur_rx;
 	unsigned int dirty_rx;
 	unsigned int dma_buf_sz;
+	unsigned int rx_copybreak;
 	u32 rx_riwt;
 	int hwts_rx_en;
 	dma_addr_t *rx_skbuff_dma;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index c803d4c..3c7928e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -781,6 +781,43 @@ static int stmmac_get_ts_info(struct net_device *dev,
 		return ethtool_op_get_ts_info(dev, info);
 }
 
+static int stmmac_get_tunable(struct net_device *dev,
+			      const struct ethtool_tunable *tuna, void *data)
+{
+	struct stmmac_priv *priv = netdev_priv(dev);
+	int ret = 0;
+
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		*(u32 *)data = priv->rx_copybreak;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
+static int stmmac_set_tunable(struct net_device *dev,
+			      const struct ethtool_tunable *tuna,
+			      const void *data)
+{
+	struct stmmac_priv *priv = netdev_priv(dev);
+	int ret = 0;
+
+	switch (tuna->id) {
+	case ETHTOOL_RX_COPYBREAK:
+		priv->rx_copybreak = *(u32 *)data;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
 static const struct ethtool_ops stmmac_ethtool_ops = {
 	.begin = stmmac_check_if_running,
 	.get_drvinfo = stmmac_ethtool_getdrvinfo,
@@ -803,6 +840,8 @@ static const struct ethtool_ops stmmac_ethtool_ops = {
 	.get_ts_info = stmmac_get_ts_info,
 	.get_coalesce = stmmac_get_coalesce,
 	.set_coalesce = stmmac_set_coalesce,
+	.get_tunable = stmmac_get_tunable,
+	.set_tunable = stmmac_set_tunable,
 };
 
 void stmmac_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 3cc1355..2ffe8dd 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -91,6 +91,8 @@ static int buf_sz = DEFAULT_BUFSIZE;
 module_param(buf_sz, int, S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(buf_sz, "DMA buffer size");
 
+#define	STMMAC_RX_COPYBREAK	256
+
 static const u32 default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
 				      NETIF_MSG_LINK | NETIF_MSG_IFUP |
 				      NETIF_MSG_IFDOWN | NETIF_MSG_TIMER);
@@ -1808,6 +1810,7 @@ static int stmmac_open(struct net_device *dev)
 	priv->xstats.threshold = tc;
 
 	priv->dma_buf_sz = STMMAC_ALIGN(buf_sz);
+	priv->rx_copybreak = STMMAC_RX_COPYBREAK;
 
 	ret = alloc_dma_desc_resources(priv);
 	if (ret < 0) {
@@ -2159,8 +2162,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
 			struct sk_buff *skb;
 
 			skb = netdev_alloc_skb_ip_align(priv->dev, bfsize);
-
-			if (unlikely(skb == NULL))
+			if (unlikely(!skb))
 				break;
 
 			priv->rx_skbuff[entry] = skb;
@@ -2282,23 +2284,52 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 					pr_debug("\tframe size %d, COE: %d\n",
 						 frame_len, status);
 			}
-			skb = priv->rx_skbuff[entry];
-			if (unlikely(!skb)) {
-				pr_err("%s: Inconsistent Rx descriptor chain\n",
-				       priv->dev->name);
-				priv->dev->stats.rx_dropped++;
-				break;
+
+			if (unlikely(frame_len < priv->rx_copybreak)) {
+				skb = netdev_alloc_skb_ip_align(priv->dev,
+								frame_len);
+				if (unlikely(!skb)) {
+					if (net_ratelimit())
+						dev_warn(priv->device,
+							 "packet dropped\n");
+					priv->dev->stats.rx_dropped++;
+					break;
+				}
+
+				dma_sync_single_for_cpu(priv->device,
+							priv->rx_skbuff_dma
+							[entry], frame_len,
+							DMA_FROM_DEVICE);
+				skb_copy_to_linear_data(skb,
+							priv->
+							rx_skbuff[entry]->data,
+							frame_len);
+
+				skb_put(skb, frame_len);
+				dma_sync_single_for_device(priv->device,
+							   priv->rx_skbuff_dma
+							   [entry], frame_len,
+							   DMA_FROM_DEVICE);
+			} else {
+				skb = priv->rx_skbuff[entry];
+				if (unlikely(!skb)) {
+					pr_err("%s: Inconsistent Rx chain\n",
+					       priv->dev->name);
+					priv->dev->stats.rx_dropped++;
+					break;
+				}
+				prefetch(skb->data - NET_IP_ALIGN);
+				priv->rx_skbuff[entry] = NULL;
+
+				skb_put(skb, frame_len);
+				dma_unmap_single(priv->device,
+						 priv->rx_skbuff_dma[entry],
+						 priv->dma_buf_sz,
+						 DMA_FROM_DEVICE);
 			}
-			prefetch(skb->data - NET_IP_ALIGN);
-			priv->rx_skbuff[entry] = NULL;
 
 			stmmac_get_rx_hwtstamp(priv, entry, skb);
 
-			skb_put(skb, frame_len);
-			dma_unmap_single(priv->device,
-					 priv->rx_skbuff_dma[entry],
-					 priv->dma_buf_sz, DMA_FROM_DEVICE);
-
 			if (netif_msg_pktdata(priv)) {
 				pr_debug("frame received (%dbytes)", frame_len);
 				print_pkt(skb->data, frame_len);
-- 
cgit v0.10.2


From 120e87f91e2980e17356f96055023681f1a4d45b Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:42 +0100
Subject: stmmac: tune rx copy via threshold.

There is a threshold now used to also limit the skb allocation
when use zero-copy. This is to avoid that there are incoherence
in the ring due to a failure on skb allocation under very
aggressive testing and under low memory conditions.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 221f5cd..d6c244f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -75,6 +75,7 @@ struct stmmac_priv {
 	unsigned int dirty_rx;
 	unsigned int dma_buf_sz;
 	unsigned int rx_copybreak;
+	unsigned int rx_zeroc_thresh;
 	u32 rx_riwt;
 	int hwts_rx_en;
 	dma_addr_t *rx_skbuff_dma;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 2ffe8dd..4c5ce98 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -72,6 +72,7 @@ module_param(phyaddr, int, S_IRUGO);
 MODULE_PARM_DESC(phyaddr, "Physical device address");
 
 #define STMMAC_TX_THRESH	(DMA_TX_SIZE / 4)
+#define STMMAC_RX_THRESH	(DMA_RX_SIZE / 4)
 
 static int flow_ctrl = FLOW_OFF;
 module_param(flow_ctrl, int, S_IRUGO | S_IWUSR);
@@ -2138,6 +2139,14 @@ static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb)
 }
 
 
+static inline int stmmac_rx_threshold_count(struct stmmac_priv *priv)
+{
+	if (priv->rx_zeroc_thresh < STMMAC_RX_THRESH)
+		return 0;
+
+	return 1;
+}
+
 /**
  * stmmac_rx_refill - refill used skb preallocated buffers
  * @priv: driver private structure
@@ -2162,8 +2171,15 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
 			struct sk_buff *skb;
 
 			skb = netdev_alloc_skb_ip_align(priv->dev, bfsize);
-			if (unlikely(!skb))
+			if (unlikely(!skb)) {
+				/* so for a while no zero-copy! */
+				priv->rx_zeroc_thresh = STMMAC_RX_THRESH;
+				if (unlikely(net_ratelimit()))
+					dev_err(priv->device,
+						"fail to alloc skb entry %d\n",
+						entry);
 				break;
+			}
 
 			priv->rx_skbuff[entry] = skb;
 			priv->rx_skbuff_dma[entry] =
@@ -2179,9 +2195,13 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
 
 			priv->hw->mode->refill_desc3(priv, p);
 
+			if (priv->rx_zeroc_thresh > 0)
+				priv->rx_zeroc_thresh--;
+
 			if (netif_msg_rx_status(priv))
 				pr_debug("\trefill entry #%d\n", entry);
 		}
+
 		wmb();
 		priv->hw->desc->set_rx_owner(p);
 		wmb();
@@ -2285,7 +2305,8 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 						 frame_len, status);
 			}
 
-			if (unlikely(frame_len < priv->rx_copybreak)) {
+			if (unlikely((frame_len < priv->rx_copybreak) ||
+				     stmmac_rx_threshold_count(priv))) {
 				skb = netdev_alloc_skb_ip_align(priv->dev,
 								frame_len);
 				if (unlikely(!skb)) {
@@ -2320,6 +2341,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 				}
 				prefetch(skb->data - NET_IP_ALIGN);
 				priv->rx_skbuff[entry] = NULL;
+				priv->rx_zeroc_thresh++;
 
 				skb_put(skb, frame_len);
 				dma_unmap_single(priv->device,
-- 
cgit v0.10.2


From 3796e44ddc1f08230ee920c70ffb5c2e4d1670f6 Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Mon, 29 Feb 2016 14:27:43 +0100
Subject: stmmac: update version to Oct_2015

This patch just updates the driver to the version fully
tested on STi platforms. This version is Oct_2015.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index d6c244f..8bbab97 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -24,7 +24,7 @@
 #define __STMMAC_H__
 
 #define STMMAC_RESOURCE_NAME   "stmmaceth"
-#define DRV_MODULE_VERSION	"March_2013"
+#define DRV_MODULE_VERSION	"Oct_2015"
 
 #include <linux/clk.h>
 #include <linux/stmmac.h>
-- 
cgit v0.10.2


From da08e4259fbfd769d1e825a685d44132c8576450 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 1 Mar 2016 17:19:32 +0530
Subject: cxgb4/cxgb4vf: Use fl capacity to check if fl needs to be replenished

Use freelist capacity instead of freelist size while checking, if
freelist needs to be refilled

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index b4eb468..22d9720 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -2226,7 +2226,7 @@ static int process_responses(struct sge_rspq *q, int budget)
 		budget_left--;
 	}
 
-	if (q->offset >= 0 && rxq->fl.size - rxq->fl.avail >= 16)
+	if (q->offset >= 0 && fl_cap(&rxq->fl) - rxq->fl.avail >= 16)
 		__refill_fl(q->adap, &rxq->fl);
 	return budget - budget_left;
 }
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index 6528231..9772aad 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -1864,7 +1864,7 @@ static int process_responses(struct sge_rspq *rspq, int budget)
 	 * for new buffer pointers, refill the Free List.
 	 */
 	if (rspq->offset >= 0 &&
-	    rxq->fl.size - rxq->fl.avail >= 2*FL_PER_EQ_UNIT)
+	    fl_cap(&rxq->fl) - rxq->fl.avail >= 2*FL_PER_EQ_UNIT)
 		__refill_fl(rspq->adapter, &rxq->fl);
 	return budget - budget_left;
 }
-- 
cgit v0.10.2


From edadad80d65bf9c7aa9f2605dbd2eef94ccd47c0 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 1 Mar 2016 17:19:33 +0530
Subject: cxgb4/cxgb4vf: For T6 adapter, set FBMIN to 64 bytes

T4 and T5 hardware will not coalesce Free List PCI-E Fetch Requests if
the Host Driver provides more Free List Pointers than the Fetch Burst
Minimum value.  So if we set FBMIN to 64 bytes and the Host Driver
supplies 128 bytes of Free List Pointer data, the hardware will issue two
64-byte PCI-E Fetch Requests rather than a single coallesced 128-byte
Fetch Request. T6 fixes this. So, for T4/T5 we set the FBMIN value to 128

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index 22d9720..deca4a2 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -2611,8 +2611,18 @@ int t4_sge_alloc_rxq(struct adapter *adap, struct sge_rspq *iq, bool fwevtq,
 				htonl(FW_IQ_CMD_FL0CNGCHMAP_V(cong) |
 				      FW_IQ_CMD_FL0CONGCIF_F |
 				      FW_IQ_CMD_FL0CONGEN_F);
+		/* In T6, for egress queue type FL there is internal overhead
+		 * of 16B for header going into FLM module.  Hence the maximum
+		 * allowed burst size is 448 bytes.  For T4/T5, the hardware
+		 * doesn't coalesce fetch requests if more than 64 bytes of
+		 * Free List pointers are provided, so we use a 128-byte Fetch
+		 * Burst Minimum there (T6 implements coalescing so we can use
+		 * the smaller 64-byte value there).
+		 */
 		c.fl0dcaen_to_fl0cidxfthresh =
-			htons(FW_IQ_CMD_FL0FBMIN_V(FETCHBURSTMIN_64B_X) |
+			htons(FW_IQ_CMD_FL0FBMIN_V(chip <= CHELSIO_T5 ?
+						   FETCHBURSTMIN_128B_X :
+						   FETCHBURSTMIN_64B_X) |
 			      FW_IQ_CMD_FL0FBMAX_V((chip <= CHELSIO_T5) ?
 						   FETCHBURSTMAX_512B_X :
 						   FETCHBURSTMAX_256B_X));
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h
index a5231fa..36cf307 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_values.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_values.h
@@ -65,6 +65,7 @@
 #define TIMERREG_COUNTER0_X		0
 
 #define FETCHBURSTMIN_64B_X		2
+#define FETCHBURSTMIN_128B_X		3
 
 #define FETCHBURSTMAX_256B_X		2
 #define FETCHBURSTMAX_512B_X		3
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index 9772aad..ba6a4e3 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -2300,9 +2300,20 @@ int t4vf_sge_alloc_rxq(struct adapter *adapter, struct sge_rspq *rspq,
 				FW_IQ_CMD_FL0HOSTFCMODE_V(SGE_HOSTFCMODE_NONE) |
 				FW_IQ_CMD_FL0PACKEN_F |
 				FW_IQ_CMD_FL0PADEN_F);
+
+		/* In T6, for egress queue type FL there is internal overhead
+		 * of 16B for header going into FLM module.  Hence the maximum
+		 * allowed burst size is 448 bytes.  For T4/T5, the hardware
+		 * doesn't coalesce fetch requests if more than 64 bytes of
+		 * Free List pointers are provided, so we use a 128-byte Fetch
+		 * Burst Minimum there (T6 implements coalescing so we can use
+		 * the smaller 64-byte value there).
+		 */
 		cmd.fl0dcaen_to_fl0cidxfthresh =
 			cpu_to_be16(
-				FW_IQ_CMD_FL0FBMIN_V(SGE_FETCHBURSTMIN_64B) |
+				FW_IQ_CMD_FL0FBMIN_V(chip <= CHELSIO_T5 ?
+						     FETCHBURSTMIN_128B_X :
+						     FETCHBURSTMIN_64B_X) |
 				FW_IQ_CMD_FL0FBMAX_V((chip <= CHELSIO_T5) ?
 						     FETCHBURSTMAX_512B_X :
 						     FETCHBURSTMAX_256B_X));
-- 
cgit v0.10.2


From cb440364c72cbbf3dde084cf65b997b40ecb8efd Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 1 Mar 2016 17:19:34 +0530
Subject: cxgb4vf: Make sge init code more readable

Adds a new function t4vf_fl_pkt_align() and use the same in SGE
initialization code to find out freelist packet alignment

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index ba6a4e3..1ccd282 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -2618,7 +2618,6 @@ int t4vf_sge_init(struct adapter *adapter)
 	u32 fl0 = sge_params->sge_fl_buffer_size[0];
 	u32 fl1 = sge_params->sge_fl_buffer_size[1];
 	struct sge *s = &adapter->sge;
-	unsigned int ingpadboundary, ingpackboundary, ingpad_shift;
 
 	/*
 	 * Start by vetting the basic SGE parameters which have been set up by
@@ -2630,7 +2629,8 @@ int t4vf_sge_init(struct adapter *adapter)
 			fl0, fl1);
 		return -EINVAL;
 	}
-	if ((sge_params->sge_control & RXPKTCPLMODE_F) == 0) {
+	if ((sge_params->sge_control & RXPKTCPLMODE_F) !=
+	    RXPKTCPLMODE_V(RXPKTCPLMODE_SPLIT_X)) {
 		dev_err(adapter->pdev_dev, "bad SGE CPL MODE\n");
 		return -EINVAL;
 	}
@@ -2643,41 +2643,7 @@ int t4vf_sge_init(struct adapter *adapter)
 	s->stat_len = ((sge_params->sge_control & EGRSTATUSPAGESIZE_F)
 			? 128 : 64);
 	s->pktshift = PKTSHIFT_G(sge_params->sge_control);
-
-	/* T4 uses a single control field to specify both the PCIe Padding and
-	 * Packing Boundary.  T5 introduced the ability to specify these
-	 * separately.  The actual Ingress Packet Data alignment boundary
-	 * within Packed Buffer Mode is the maximum of these two
-	 * specifications.  (Note that it makes no real practical sense to
-	 * have the Pading Boudary be larger than the Packing Boundary but you
-	 * could set the chip up that way and, in fact, legacy T4 code would
-	 * end doing this because it would initialize the Padding Boundary and
-	 * leave the Packing Boundary initialized to 0 (16 bytes).)
-	 * Padding Boundary values in T6 starts from 8B,
-	 * where as it is 32B for T4 and T5.
-	 */
-	if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5)
-		ingpad_shift = INGPADBOUNDARY_SHIFT_X;
-	else
-		ingpad_shift = T6_INGPADBOUNDARY_SHIFT_X;
-
-	ingpadboundary = 1 << (INGPADBOUNDARY_G(sge_params->sge_control) +
-			       ingpad_shift);
-	if (is_t4(adapter->params.chip)) {
-		s->fl_align = ingpadboundary;
-	} else {
-		/* T5 has a different interpretation of one of the PCIe Packing
-		 * Boundary values.
-		 */
-		ingpackboundary = INGPACKBOUNDARY_G(sge_params->sge_control2);
-		if (ingpackboundary == INGPACKBOUNDARY_16B_X)
-			ingpackboundary = 16;
-		else
-			ingpackboundary = 1 << (ingpackboundary +
-						INGPACKBOUNDARY_SHIFT_X);
-
-		s->fl_align = max(ingpadboundary, ingpackboundary);
-	}
+	s->fl_align = t4vf_fl_pkt_align(adapter);
 
 	/* A FL with <= fl_starve_thres buffers is starving and a periodic
 	 * timer will attempt to refill it.  This needs to be larger than the
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
index 6ce302fe..9b40a85 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h
@@ -309,6 +309,7 @@ int t4vf_port_init(struct adapter *, int);
 int t4vf_fw_reset(struct adapter *);
 int t4vf_set_params(struct adapter *, unsigned int, const u32 *, const u32 *);
 
+int t4vf_fl_pkt_align(struct adapter *adapter);
 enum t4_bar2_qtype { T4_BAR2_QTYPE_EGRESS, T4_BAR2_QTYPE_INGRESS };
 int t4vf_bar2_sge_qregs(struct adapter *adapter,
 			unsigned int qid,
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
index 5422011..fed83d8 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
@@ -418,6 +418,61 @@ int t4vf_set_params(struct adapter *adapter, unsigned int nparams,
 }
 
 /**
+ *	t4vf_fl_pkt_align - return the fl packet alignment
+ *	@adapter: the adapter
+ *
+ *	T4 has a single field to specify the packing and padding boundary.
+ *	T5 onwards has separate fields for this and hence the alignment for
+ *	next packet offset is maximum of these two.  And T6 changes the
+ *	Ingress Padding Boundary Shift, so it's all a mess and it's best
+ *	if we put this in low-level Common Code ...
+ *
+ */
+int t4vf_fl_pkt_align(struct adapter *adapter)
+{
+	u32 sge_control, sge_control2;
+	unsigned int ingpadboundary, ingpackboundary, fl_align, ingpad_shift;
+
+	sge_control = adapter->params.sge.sge_control;
+
+	/* T4 uses a single control field to specify both the PCIe Padding and
+	 * Packing Boundary.  T5 introduced the ability to specify these
+	 * separately.  The actual Ingress Packet Data alignment boundary
+	 * within Packed Buffer Mode is the maximum of these two
+	 * specifications.  (Note that it makes no real practical sense to
+	 * have the Pading Boudary be larger than the Packing Boundary but you
+	 * could set the chip up that way and, in fact, legacy T4 code would
+	 * end doing this because it would initialize the Padding Boundary and
+	 * leave the Packing Boundary initialized to 0 (16 bytes).)
+	 * Padding Boundary values in T6 starts from 8B,
+	 * where as it is 32B for T4 and T5.
+	 */
+	if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5)
+		ingpad_shift = INGPADBOUNDARY_SHIFT_X;
+	else
+		ingpad_shift = T6_INGPADBOUNDARY_SHIFT_X;
+
+	ingpadboundary = 1 << (INGPADBOUNDARY_G(sge_control) + ingpad_shift);
+
+	fl_align = ingpadboundary;
+	if (!is_t4(adapter->params.chip)) {
+		/* T5 has a different interpretation of one of the PCIe Packing
+		 * Boundary values.
+		 */
+		sge_control2 = adapter->params.sge.sge_control2;
+		ingpackboundary = INGPACKBOUNDARY_G(sge_control2);
+		if (ingpackboundary == INGPACKBOUNDARY_16B_X)
+			ingpackboundary = 16;
+		else
+			ingpackboundary = 1 << (ingpackboundary +
+						INGPACKBOUNDARY_SHIFT_X);
+
+		fl_align = max(ingpadboundary, ingpackboundary);
+	}
+	return fl_align;
+}
+
+/**
  *	t4vf_bar2_sge_qregs - return BAR2 SGE Queue register information
  *	@adapter: the adapter
  *	@qid: the Queue ID
-- 
cgit v0.10.2


From 5d7b80522b3f1c60147b8c0405ca57e015be26e5 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 1 Mar 2016 17:19:35 +0530
Subject: cxgb4vf: Remove redundant adapter ready check during probe

Function t4vf_wait_dev_ready() is already called in t4vf_prep_adapter(),
no need to call it again in adap_init0().

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 8337514..5d989e4 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -2237,16 +2237,6 @@ static int adap_init0(struct adapter *adapter)
 	u32 param, val = 0;
 
 	/*
-	 * Wait for the device to become ready before proceeding ...
-	 */
-	err = t4vf_wait_dev_ready(adapter);
-	if (err) {
-		dev_err(adapter->pdev_dev, "device didn't become ready:"
-			" err=%d\n", err);
-		return err;
-	}
-
-	/*
 	 * Some environments do not properly handle PCIE FLRs -- e.g. in Linux
 	 * 2.6.31 and later we can't call pci_reset_function() in order to
 	 * issue an FLR because of a self- deadlock on the device semaphore.
-- 
cgit v0.10.2


From f1ea2fe05dcc0104b445c127bf4bcf2b1b498b76 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 1 Mar 2016 17:19:36 +0530
Subject: cxgb4vf: Remove dead functions collect_netdev_[um]c_list_addrs

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 5d989e4..91857b8 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -862,52 +862,6 @@ static struct net_device_stats *cxgb4vf_get_stats(struct net_device *dev)
 	return ns;
 }
 
-/*
- * Collect up to maxaddrs worth of a netdevice's unicast addresses, starting
- * at a specified offset within the list, into an array of addrss pointers and
- * return the number collected.
- */
-static inline unsigned int collect_netdev_uc_list_addrs(const struct net_device *dev,
-							const u8 **addr,
-							unsigned int offset,
-							unsigned int maxaddrs)
-{
-	unsigned int index = 0;
-	unsigned int naddr = 0;
-	const struct netdev_hw_addr *ha;
-
-	for_each_dev_addr(dev, ha)
-		if (index++ >= offset) {
-			addr[naddr++] = ha->addr;
-			if (naddr >= maxaddrs)
-				break;
-		}
-	return naddr;
-}
-
-/*
- * Collect up to maxaddrs worth of a netdevice's multicast addresses, starting
- * at a specified offset within the list, into an array of addrss pointers and
- * return the number collected.
- */
-static inline unsigned int collect_netdev_mc_list_addrs(const struct net_device *dev,
-							const u8 **addr,
-							unsigned int offset,
-							unsigned int maxaddrs)
-{
-	unsigned int index = 0;
-	unsigned int naddr = 0;
-	const struct netdev_hw_addr *ha;
-
-	netdev_for_each_mc_addr(ha, dev)
-		if (index++ >= offset) {
-			addr[naddr++] = ha->addr;
-			if (naddr >= maxaddrs)
-				break;
-		}
-	return naddr;
-}
-
 static inline int cxgb4vf_set_addr_hash(struct port_info *pi)
 {
 	struct adapter *adapter = pi->adapter;
-- 
cgit v0.10.2


From 6353e1875dd775fc915cff3a2d53cb163bdeb63e Mon Sep 17 00:00:00 2001
From: Eric Engestrom <eric.engestrom@imgtec.com>
Date: Mon, 29 Feb 2016 16:36:38 +0000
Subject: net/rtnetlink: remove dead code

3b766cd832328fcb87db3507e7b98cf42f21689d ("net/core: Add reading VF
statistics through the PF netdevice") added that variable but it's never
been used.

Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 62737f4..6128aac 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1391,15 +1391,6 @@ static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = {
 	[IFLA_VF_TRUST]		= { .len = sizeof(struct ifla_vf_trust) },
 };
 
-static const struct nla_policy ifla_vf_stats_policy[IFLA_VF_STATS_MAX + 1] = {
-	[IFLA_VF_STATS_RX_PACKETS]	= { .type = NLA_U64 },
-	[IFLA_VF_STATS_TX_PACKETS]	= { .type = NLA_U64 },
-	[IFLA_VF_STATS_RX_BYTES]	= { .type = NLA_U64 },
-	[IFLA_VF_STATS_TX_BYTES]	= { .type = NLA_U64 },
-	[IFLA_VF_STATS_BROADCAST]	= { .type = NLA_U64 },
-	[IFLA_VF_STATS_MULTICAST]	= { .type = NLA_U64 },
-};
-
 static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = {
 	[IFLA_PORT_VF]		= { .type = NLA_U32 },
 	[IFLA_PORT_PROFILE]	= { .type = NLA_STRING,
-- 
cgit v0.10.2


From a9d562358b5c12a3d1c343f45a3c56df47dad753 Mon Sep 17 00:00:00 2001
From: Eric Engestrom <eric.engestrom@imgtec.com>
Date: Mon, 29 Feb 2016 16:38:06 +0000
Subject: net/ipv4: remove left over dead code

8cc785f6f429c2a3fb81745dc142cbd72a462c4a ("net: ipv4: make the ping
/proc code AF-independent") removed the code using it, but renamed this
variable instead of removing it.

Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 76dce90..cf9700b 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -1142,13 +1142,6 @@ static int ping_v4_seq_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static const struct seq_operations ping_v4_seq_ops = {
-	.show		= ping_v4_seq_show,
-	.start		= ping_v4_seq_start,
-	.next		= ping_seq_next,
-	.stop		= ping_seq_stop,
-};
-
 static int ping_seq_open(struct inode *inode, struct file *file)
 {
 	struct ping_seq_afinfo *afinfo = PDE_DATA(inode);
-- 
cgit v0.10.2


From 998fc1d0803bff10d6d8a6f6ed67689f327d9315 Mon Sep 17 00:00:00 2001
From: Eric Engestrom <eric.engestrom@imgtec.com>
Date: Mon, 29 Feb 2016 16:40:23 +0000
Subject: ethernet/atl1c: remove left over dead code

Left over from c24588afc536a35c924d014f13b669b20ccf8553
("atl1c: using fixed TXQ configuration for l2cb and l1c")

Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 8b5988e..d0084d4 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -65,10 +65,6 @@ static void atl1c_reset_dma_ring(struct atl1c_adapter *adapter);
 static int atl1c_configure(struct atl1c_adapter *adapter);
 static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter);
 
-static const u16 atl1c_pay_load_size[] = {
-	128, 256, 512, 1024, 2048, 4096,
-};
-
 
 static const u32 atl1c_default_msg = NETIF_MSG_DRV | NETIF_MSG_PROBE |
 	NETIF_MSG_LINK | NETIF_MSG_TIMER | NETIF_MSG_IFDOWN | NETIF_MSG_IFUP;
-- 
cgit v0.10.2


From dd215430dc9132fc0505cb4d5aa6dc6243ce8f9c Mon Sep 17 00:00:00 2001
From: Mika Westerberg <mika.westerberg@linux.intel.com>
Date: Thu, 3 Mar 2016 11:26:18 +0200
Subject: NFC: pn544: Drop two useless checks in ACPI probe path

When pn544_hci_i2c_acpi_request_resources() gets called we
already know that the entries in ->acpi_match_table have
matched ACPI ID of the device.
In addition I2C client pointer cannot be NULL in any case
(otherwise I2C core would not call ->probe() for the driver
in the first place).

Drop the two useless checks from the driver.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

diff --git a/drivers/nfc/pn544/i2c.c b/drivers/nfc/pn544/i2c.c
index 76c3184..45d0e66 100644
--- a/drivers/nfc/pn544/i2c.c
+++ b/drivers/nfc/pn544/i2c.c
@@ -877,20 +877,8 @@ exit_state_wait_secure_write_answer:
 static int pn544_hci_i2c_acpi_request_resources(struct i2c_client *client)
 {
 	struct pn544_i2c_phy *phy = i2c_get_clientdata(client);
-	const struct acpi_device_id *id;
 	struct gpio_desc *gpiod_en, *gpiod_fw;
-	struct device *dev;
-
-	if (!client)
-		return -EINVAL;
-
-	dev = &client->dev;
-
-	/* Match the struct device against a given list of ACPI IDs */
-	id = acpi_match_device(dev->driver->acpi_match_table, dev);
-
-	if (!id)
-		return -ENODEV;
+	struct device *dev = &client->dev;
 
 	/* Get EN GPIO from ACPI */
 	gpiod_en = devm_gpiod_get_index(dev, PN544_GPIO_NAME_EN, 1,
-- 
cgit v0.10.2


From e0b6ce00b1ec0434e96629ab64f3e16dcb4f33b7 Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Wed, 3 Feb 2016 21:07:43 +0530
Subject: ath10k: fix pointless update of peer stats list

We periodically receive f/w stats event for updating
the rx duration and there is no reason to keep on appending
the f/w stats peer list, as this gets completely cleaned up when
the user polls for f/w stats {pdev, vdev, peer stats}. Only don't
print the warning message in the case PEER_STATS service is enabled

Fixes: 856e7c3 ("ath10k: add debugfs support for Per STA total rx duration")
Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 848a0dd..a633152 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -351,8 +351,10 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb)
 	if (peer_stats_svc)
 		ath10k_sta_update_rx_duration(ar, &stats.peers);
 
-	if (ar->debug.fw_stats_done && !peer_stats_svc) {
-		ath10k_warn(ar, "received unsolicited stats update event\n");
+	if (ar->debug.fw_stats_done) {
+		if (!peer_stats_svc)
+			ath10k_warn(ar, "received unsolicited stats update event\n");
+
 		goto free;
 	}
 
-- 
cgit v0.10.2


From 53a5c9bc53ce51f65699a43c67ab167436d28083 Mon Sep 17 00:00:00 2001
From: Ashok Raj Nagarajan <arnagara@qti.qualcomm.com>
Date: Fri, 5 Feb 2016 21:12:48 +0530
Subject: ath10k: fix pktlog in QCA99X0

Currently, we are providing wrong payload data of pktlog to trace points.
Data we receive from FW through copy engine 8 contains pktlog data alone.
We don't need to parse anything in driver before handing it to trace
points.

Fixes: afb0bf7f530b ("ath10k: add support for pktlog in QCA99X0")
Signed-off-by: Ashok Raj Nagarajan <arnagara@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index cc957a6..bf778f2 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2145,11 +2145,7 @@ EXPORT_SYMBOL(ath10k_htt_t2h_msg_handler);
 void ath10k_htt_rx_pktlog_completion_handler(struct ath10k *ar,
 					     struct sk_buff *skb)
 {
-	struct ath10k_pktlog_10_4_hdr *hdr =
-		(struct ath10k_pktlog_10_4_hdr *)skb->data;
-
-	trace_ath10k_htt_pktlog(ar, hdr->payload,
-				sizeof(*hdr) + __le16_to_cpu(hdr->size));
+	trace_ath10k_htt_pktlog(ar, skb->data, skb->len);
 	dev_kfree_skb_any(skb);
 }
 EXPORT_SYMBOL(ath10k_htt_rx_pktlog_completion_handler);
-- 
cgit v0.10.2


From 18353749170033405790ed62a3ae2aa2d3d8a82a Mon Sep 17 00:00:00 2001
From: Ashok Raj Nagarajan <arnagara@qti.qualcomm.com>
Date: Fri, 5 Feb 2016 21:12:49 +0530
Subject: ath10k: add hw_rev to trace events to support pktlog

pktlog data is different between firmware variants (eg. 10.2 vs 10.4). To
have a unified user space script to decode pktlog trace events generated,
it is desirable to know which firmware variant has provided the events and
thereby decode the pktlogs appropriately. Hardware revision (hw_rev) helps
to determine the firmware variant sending these trace events. So add hw_rev
to trace events.

Signed-off-by: Ashok Raj Nagarajan <arnagara@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/trace.h b/drivers/net/wireless/ath/ath10k/trace.h
index 71bdb36..e0d00ce 100644
--- a/drivers/net/wireless/ath/ath10k/trace.h
+++ b/drivers/net/wireless/ath/ath10k/trace.h
@@ -250,6 +250,7 @@ TRACE_EVENT(ath10k_wmi_dbglog,
 	TP_STRUCT__entry(
 		__string(device, dev_name(ar->dev))
 		__string(driver, dev_driver_string(ar->dev))
+		__field(u8, hw_type);
 		__field(size_t, buf_len)
 		__dynamic_array(u8, buf, buf_len)
 	),
@@ -257,14 +258,16 @@ TRACE_EVENT(ath10k_wmi_dbglog,
 	TP_fast_assign(
 		__assign_str(device, dev_name(ar->dev));
 		__assign_str(driver, dev_driver_string(ar->dev));
+		__entry->hw_type = ar->hw_rev;
 		__entry->buf_len = buf_len;
 		memcpy(__get_dynamic_array(buf), buf, buf_len);
 	),
 
 	TP_printk(
-		"%s %s len %zu",
+		"%s %s %d len %zu",
 		__get_str(driver),
 		__get_str(device),
+		__entry->hw_type,
 		__entry->buf_len
 	)
 );
@@ -277,6 +280,7 @@ TRACE_EVENT(ath10k_htt_pktlog,
 	TP_STRUCT__entry(
 		__string(device, dev_name(ar->dev))
 		__string(driver, dev_driver_string(ar->dev))
+		__field(u8, hw_type);
 		__field(u16, buf_len)
 		__dynamic_array(u8, pktlog, buf_len)
 	),
@@ -284,14 +288,16 @@ TRACE_EVENT(ath10k_htt_pktlog,
 	TP_fast_assign(
 		__assign_str(device, dev_name(ar->dev));
 		__assign_str(driver, dev_driver_string(ar->dev));
+		__entry->hw_type = ar->hw_rev;
 		__entry->buf_len = buf_len;
 		memcpy(__get_dynamic_array(pktlog), buf, buf_len);
 	),
 
 	TP_printk(
-		"%s %s size %hu",
+		"%s %s %d size %hu",
 		__get_str(driver),
 		__get_str(device),
+		__entry->hw_type,
 		__entry->buf_len
 	 )
 );
@@ -440,6 +446,7 @@ TRACE_EVENT(ath10k_htt_rx_desc,
 	TP_STRUCT__entry(
 		__string(device, dev_name(ar->dev))
 		__string(driver, dev_driver_string(ar->dev))
+		__field(u8, hw_type);
 		__field(u16, len)
 		__dynamic_array(u8, rxdesc, len)
 	),
@@ -447,14 +454,16 @@ TRACE_EVENT(ath10k_htt_rx_desc,
 	TP_fast_assign(
 		__assign_str(device, dev_name(ar->dev));
 		__assign_str(driver, dev_driver_string(ar->dev));
+		__entry->hw_type = ar->hw_rev;
 		__entry->len = len;
 		memcpy(__get_dynamic_array(rxdesc), data, len);
 	),
 
 	TP_printk(
-		"%s %s rxdesc len %d",
+		"%s %s %d rxdesc len %d",
 		__get_str(driver),
 		__get_str(device),
+		__entry->hw_type,
 		__entry->len
 	 )
 );
-- 
cgit v0.10.2


From 22baa98097df3eb92a51e8661fda5dd7c0f1eb93 Mon Sep 17 00:00:00 2001
From: Anton Protopopov <a.s.protopopov@gmail.com>
Date: Wed, 10 Feb 2016 11:58:55 -0500
Subject: ath10k: fix erroneous return value

The ath10k_pci_hif_exchange_bmi_msg() function may return the positive
value EIO instead of -EIO in case of error.

Signed-off-by: Anton Protopopov <a.s.protopopov@gmail.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c
index 0e338b6..b3cff1d 100644
--- a/drivers/net/wireless/ath/ath10k/pci.c
+++ b/drivers/net/wireless/ath/ath10k/pci.c
@@ -1772,7 +1772,7 @@ int ath10k_pci_hif_exchange_bmi_msg(struct ath10k *ar,
 					    DMA_FROM_DEVICE);
 		ret = dma_mapping_error(ar->dev, resp_paddr);
 		if (ret) {
-			ret = EIO;
+			ret = -EIO;
 			goto err_req;
 		}
 
-- 
cgit v0.10.2


From e7827e512af01c6220e2f1cc53853f11cd742eef Mon Sep 17 00:00:00 2001
From: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Date: Fri, 12 Feb 2016 11:40:58 +0530
Subject: ath10k: reduce rx_lock contention for htt rx indication

Received frame indications are queued into a skb list and latest
processed by txrx tasklet. This skb queue is protected by htt rx lock.
Since the entire rx processing till delivering frame to mac80211 and
replenish tasks are processed under rx_lock protection, there might be
some delay in queuing newly received rx frame into that list on
multicore systems. Optimize this by using skb list lock while accessing
rx completion queue instead of htt rx lock.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index bf778f2..a31de7b 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2011,9 +2011,7 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		break;
 	}
 	case HTT_T2H_MSG_TYPE_RX_IND:
-		spin_lock_bh(&htt->rx_ring.lock);
-		__skb_queue_tail(&htt->rx_compl_q, skb);
-		spin_unlock_bh(&htt->rx_ring.lock);
+		skb_queue_tail(&htt->rx_compl_q, skb);
 		tasklet_schedule(&htt->txrx_compl_task);
 		return;
 	case HTT_T2H_MSG_TYPE_PEER_MAP: {
@@ -2111,9 +2109,7 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		break;
 	}
 	case HTT_T2H_MSG_TYPE_RX_IN_ORD_PADDR_IND: {
-		spin_lock_bh(&htt->rx_ring.lock);
-		__skb_queue_tail(&htt->rx_in_ord_compl_q, skb);
-		spin_unlock_bh(&htt->rx_ring.lock);
+		skb_queue_tail(&htt->rx_in_ord_compl_q, skb);
 		tasklet_schedule(&htt->txrx_compl_task);
 		return;
 	}
@@ -2170,16 +2166,18 @@ static void ath10k_htt_txrx_compl_task(unsigned long ptr)
 		dev_kfree_skb_any(skb);
 	}
 
-	spin_lock_bh(&htt->rx_ring.lock);
-	while ((skb = __skb_dequeue(&htt->rx_compl_q))) {
+	while ((skb = skb_dequeue(&htt->rx_compl_q))) {
 		resp = (struct htt_resp *)skb->data;
+		spin_lock_bh(&htt->rx_ring.lock);
 		ath10k_htt_rx_handler(htt, &resp->rx_ind);
+		spin_unlock_bh(&htt->rx_ring.lock);
 		dev_kfree_skb_any(skb);
 	}
 
-	while ((skb = __skb_dequeue(&htt->rx_in_ord_compl_q))) {
+	while ((skb = skb_dequeue(&htt->rx_in_ord_compl_q))) {
+		spin_lock_bh(&htt->rx_ring.lock);
 		ath10k_htt_rx_in_ord_ind(ar, skb);
+		spin_unlock_bh(&htt->rx_ring.lock);
 		dev_kfree_skb_any(skb);
 	}
-	spin_unlock_bh(&htt->rx_ring.lock);
 }
-- 
cgit v0.10.2


From da6416cac6b9439167964cae1302d9cfe764bf1d Mon Sep 17 00:00:00 2001
From: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Date: Fri, 12 Feb 2016 11:40:59 +0530
Subject: ath10k: process htt rx indication as batch mode

On multicore systems, it is possible that txrx tasket can run
in parallel with pci tasklet (i.e smp affinity of ath10k irq is
assigned to multiple CPUs). Feeding and consuming from the same
rx completion list leads to txrx tasklet runs for longer period.
Prevent this by processing a snapshot of rx queue by moving list
into temporary list. Consecutive received frames will be processed
in next batch.

Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c
index a31de7b..ae9b686 100644
--- a/drivers/net/wireless/ath/ath10k/htt_rx.c
+++ b/drivers/net/wireless/ath/ath10k/htt_rx.c
@@ -2151,22 +2151,34 @@ static void ath10k_htt_txrx_compl_task(unsigned long ptr)
 	struct ath10k_htt *htt = (struct ath10k_htt *)ptr;
 	struct ath10k *ar = htt->ar;
 	struct sk_buff_head tx_q;
+	struct sk_buff_head rx_q;
+	struct sk_buff_head rx_ind_q;
 	struct htt_resp *resp;
 	struct sk_buff *skb;
 	unsigned long flags;
 
 	__skb_queue_head_init(&tx_q);
+	__skb_queue_head_init(&rx_q);
+	__skb_queue_head_init(&rx_ind_q);
 
 	spin_lock_irqsave(&htt->tx_compl_q.lock, flags);
 	skb_queue_splice_init(&htt->tx_compl_q, &tx_q);
 	spin_unlock_irqrestore(&htt->tx_compl_q.lock, flags);
 
+	spin_lock_irqsave(&htt->rx_compl_q.lock, flags);
+	skb_queue_splice_init(&htt->rx_compl_q, &rx_q);
+	spin_unlock_irqrestore(&htt->rx_compl_q.lock, flags);
+
+	spin_lock_irqsave(&htt->rx_in_ord_compl_q.lock, flags);
+	skb_queue_splice_init(&htt->rx_in_ord_compl_q, &rx_ind_q);
+	spin_unlock_irqrestore(&htt->rx_in_ord_compl_q.lock, flags);
+
 	while ((skb = __skb_dequeue(&tx_q))) {
 		ath10k_htt_rx_frm_tx_compl(htt->ar, skb);
 		dev_kfree_skb_any(skb);
 	}
 
-	while ((skb = skb_dequeue(&htt->rx_compl_q))) {
+	while ((skb = __skb_dequeue(&rx_q))) {
 		resp = (struct htt_resp *)skb->data;
 		spin_lock_bh(&htt->rx_ring.lock);
 		ath10k_htt_rx_handler(htt, &resp->rx_ind);
@@ -2174,7 +2186,7 @@ static void ath10k_htt_txrx_compl_task(unsigned long ptr)
 		dev_kfree_skb_any(skb);
 	}
 
-	while ((skb = skb_dequeue(&htt->rx_in_ord_compl_q))) {
+	while ((skb = __skb_dequeue(&rx_ind_q))) {
 		spin_lock_bh(&htt->rx_ring.lock);
 		ath10k_htt_rx_in_ord_ind(ar, skb);
 		spin_unlock_bh(&htt->rx_ring.lock);
-- 
cgit v0.10.2


From af9a6a3ad04cb042fe1b723335c411c1ab9b7959 Mon Sep 17 00:00:00 2001
From: Anilkumar Kolli <akolli@qti.qualcomm.com>
Date: Tue, 23 Feb 2016 12:19:57 +0530
Subject: ath10k: reduce number of peers to support peer stats feature

To enable per peer stats feature we are reducing the number of peers.
Firmware has introduced tx stats feature. We have memory limitation in
firmware to add these additional bytes.

These are the new variables introduced in the firmware.
========		=======================
Variable	      	Bytes required/per rate
========		=======================
TX success packets 	1
TX failed packets	1
Retry packets		1
Success bytes		2
TX failed bytes		2
Retry bytes		2
Tx duration		4
Rate			1
Bw and AMPDU flags	1
Total			16 (because of allocation in word pattern)

Firmware sends these tx_stats in pktlog.
If we consider 4 feedbacks at a time, Frimware need about ~1K memory for coding
and 8192 bytes required / per rate [ 4*16*128(peers)].
To accommodate this firmware needs to reduce 10 peers.

This fixes a firmware crash with firmware-5.bin_10.2.4.70.22-2.

Signed-off-by: Anilkumar Kolli <akolli@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index 4d31764..c84c2d3 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -1509,8 +1509,13 @@ static int ath10k_core_init_firmware_features(struct ath10k *ar)
 	case ATH10K_FW_WMI_OP_VERSION_10_1:
 	case ATH10K_FW_WMI_OP_VERSION_10_2:
 	case ATH10K_FW_WMI_OP_VERSION_10_2_4:
-		ar->max_num_peers = TARGET_10X_NUM_PEERS;
-		ar->max_num_stations = TARGET_10X_NUM_STATIONS;
+		if (test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map)) {
+			ar->max_num_peers = TARGET_10X_TX_STATS_NUM_PEERS;
+			ar->max_num_stations = TARGET_10X_TX_STATS_NUM_STATIONS;
+		} else {
+			ar->max_num_peers = TARGET_10X_NUM_PEERS;
+			ar->max_num_stations = TARGET_10X_NUM_STATIONS;
+		}
 		ar->max_num_vdevs = TARGET_10X_NUM_VDEVS;
 		ar->htt.max_num_pending_tx = TARGET_10X_NUM_MSDU_DESC;
 		ar->fw_stats_req_mask = WMI_STAT_PEER;
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 2dece8d..f0cfbc7 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -375,14 +375,19 @@ enum ath10k_hw_4addr_pad {
 #define TARGET_10X_MAC_AGGR_DELIM		0
 #define TARGET_10X_AST_SKID_LIMIT		128
 #define TARGET_10X_NUM_STATIONS			128
+#define TARGET_10X_TX_STATS_NUM_STATIONS	118
 #define TARGET_10X_NUM_PEERS			((TARGET_10X_NUM_STATIONS) + \
 						 (TARGET_10X_NUM_VDEVS))
+#define TARGET_10X_TX_STATS_NUM_PEERS		((TARGET_10X_TX_STATS_NUM_STATIONS) + \
+						 (TARGET_10X_NUM_VDEVS))
 #define TARGET_10X_NUM_OFFLOAD_PEERS		0
 #define TARGET_10X_NUM_OFFLOAD_REORDER_BUFS	0
 #define TARGET_10X_NUM_PEER_KEYS		2
 #define TARGET_10X_NUM_TIDS_MAX			256
 #define TARGET_10X_NUM_TIDS			min((TARGET_10X_NUM_TIDS_MAX), \
 						    (TARGET_10X_NUM_PEERS) * 2)
+#define TARGET_10X_TX_STATS_NUM_TIDS		min((TARGET_10X_NUM_TIDS_MAX), \
+						    (TARGET_10X_TX_STATS_NUM_PEERS) * 2)
 #define TARGET_10X_TX_CHAIN_MASK		(BIT(0) | BIT(1) | BIT(2))
 #define TARGET_10X_RX_CHAIN_MASK		(BIT(0) | BIT(1) | BIT(2))
 #define TARGET_10X_RX_TIMEOUT_LO_PRI		100
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 1ce6742..7026138 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -5516,9 +5516,15 @@ static struct sk_buff *ath10k_wmi_10_2_op_gen_init(struct ath10k *ar)
 	u32 len, val, features;
 
 	config.num_vdevs = __cpu_to_le32(TARGET_10X_NUM_VDEVS);
-	config.num_peers = __cpu_to_le32(TARGET_10X_NUM_PEERS);
 	config.num_peer_keys = __cpu_to_le32(TARGET_10X_NUM_PEER_KEYS);
-	config.num_tids = __cpu_to_le32(TARGET_10X_NUM_TIDS);
+	if (test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map)) {
+		config.num_peers = __cpu_to_le32(TARGET_10X_TX_STATS_NUM_PEERS);
+		config.num_tids = __cpu_to_le32(TARGET_10X_TX_STATS_NUM_TIDS);
+	} else {
+		config.num_peers = __cpu_to_le32(TARGET_10X_NUM_PEERS);
+		config.num_tids = __cpu_to_le32(TARGET_10X_NUM_TIDS);
+	}
+
 	config.ast_skid_limit = __cpu_to_le32(TARGET_10X_AST_SKID_LIMIT);
 	config.tx_chain_mask = __cpu_to_le32(TARGET_10X_TX_CHAIN_MASK);
 	config.rx_chain_mask = __cpu_to_le32(TARGET_10X_RX_CHAIN_MASK);
-- 
cgit v0.10.2


From c28e6f06ff401ce8f37d58ff6d8123664cd65a51 Mon Sep 17 00:00:00 2001
From: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Date: Tue, 23 Feb 2016 12:58:36 +0530
Subject: ath10k: fix sanity check on enabling btcoex via debugfs

First check for the device state before enabling / disabling
btcoex, also return a proper error value. Enabling / disabling
btcoex ideally does a f/w + ath10k_core_restart so the checks
that are applicable for 'simulate_fw_crash' shall be applicable
for this as well

Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index a633152..076d29b 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -2114,6 +2114,7 @@ static ssize_t ath10k_write_btcoex(struct file *file,
 	struct ath10k *ar = file->private_data;
 	char buf[32];
 	size_t buf_size;
+	int ret = 0;
 	bool val;
 
 	buf_size = min(count, (sizeof(buf) - 1));
@@ -2127,6 +2128,12 @@ static ssize_t ath10k_write_btcoex(struct file *file,
 
 	mutex_lock(&ar->conf_mutex);
 
+	if (ar->state != ATH10K_STATE_ON &&
+	    ar->state != ATH10K_STATE_RESTARTED) {
+		ret = -ENETDOWN;
+		goto exit;
+	}
+
 	if (!(test_bit(ATH10K_FLAG_BTCOEX, &ar->dev_flags) ^ val))
 		goto exit;
 
@@ -2135,17 +2142,15 @@ static ssize_t ath10k_write_btcoex(struct file *file,
 	else
 		clear_bit(ATH10K_FLAG_BTCOEX, &ar->dev_flags);
 
-	if (ar->state != ATH10K_STATE_ON)
-		goto exit;
-
 	ath10k_info(ar, "restarting firmware due to btcoex change");
 
 	queue_work(ar->workqueue, &ar->restart_work);
+	ret = count;
 
 exit:
 	mutex_unlock(&ar->conf_mutex);
 
-	return count;
+	return ret;
 }
 
 static ssize_t ath10k_read_btcoex(struct file *file, char __user *ubuf,
-- 
cgit v0.10.2


From 82def495d118c6d36300c1cf2448c276273be603 Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Fri, 5 Feb 2016 09:45:50 +0800
Subject: ath9k: make NF load complete quickly and reliably

Make NF load complete quickly and reliably. NF load execution
is delayed by HW to end of frame if frame Rx or Tx is ongoing.
Increasing timeout to max frame duration. If NF cal is ongoing
before NF load, stop it before load, and restart it afterwards.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/calib.c b/drivers/net/wireless/ath/ath9k/calib.c
index 3e2e24e..37f6d66 100644
--- a/drivers/net/wireless/ath/ath9k/calib.c
+++ b/drivers/net/wireless/ath/ath9k/calib.c
@@ -241,6 +241,7 @@ int ath9k_hw_loadnf(struct ath_hw *ah, struct ath9k_channel *chan)
 	u8 chainmask = (ah->rxchainmask << 3) | ah->rxchainmask;
 	struct ath_common *common = ath9k_hw_common(ah);
 	s16 default_nf = ath9k_hw_get_default_nf(ah, chan);
+	u32 bb_agc_ctl = REG_READ(ah, AR_PHY_AGC_CONTROL);
 
 	if (ah->caldata)
 		h = ah->caldata->nfCalHist;
@@ -264,6 +265,16 @@ int ath9k_hw_loadnf(struct ath_hw *ah, struct ath9k_channel *chan)
 	}
 
 	/*
+	 * stop NF cal if ongoing to ensure NF load completes immediately
+	 * (or after end rx/tx frame if ongoing)
+	 */
+	if (bb_agc_ctl & AR_PHY_AGC_CONTROL_NF) {
+		REG_CLR_BIT(ah, AR_PHY_AGC_CONTROL, AR_PHY_AGC_CONTROL_NF);
+		REG_RMW_BUFFER_FLUSH(ah);
+		ENABLE_REG_RMW_BUFFER(ah);
+	}
+
+	/*
 	 * Load software filtered NF value into baseband internal minCCApwr
 	 * variable.
 	 */
@@ -276,18 +287,33 @@ int ath9k_hw_loadnf(struct ath_hw *ah, struct ath9k_channel *chan)
 
 	/*
 	 * Wait for load to complete, should be fast, a few 10s of us.
-	 * The max delay was changed from an original 250us to 10000us
-	 * since 250us often results in NF load timeout and causes deaf
-	 * condition during stress testing 12/12/2009
+	 * The max delay was changed from an original 250us to 22.2 msec.
+	 * This would increase timeout to the longest possible frame
+	 * (11n max length 22.1 msec)
 	 */
-	for (j = 0; j < 10000; j++) {
+	for (j = 0; j < 22200; j++) {
 		if ((REG_READ(ah, AR_PHY_AGC_CONTROL) &
-		     AR_PHY_AGC_CONTROL_NF) == 0)
+			      AR_PHY_AGC_CONTROL_NF) == 0)
 			break;
 		udelay(10);
 	}
 
 	/*
+	 * Restart NF so it can continue.
+	 */
+	if (bb_agc_ctl & AR_PHY_AGC_CONTROL_NF) {
+		ENABLE_REG_RMW_BUFFER(ah);
+		if (bb_agc_ctl & AR_PHY_AGC_CONTROL_ENABLE_NF)
+			REG_SET_BIT(ah, AR_PHY_AGC_CONTROL,
+				    AR_PHY_AGC_CONTROL_ENABLE_NF);
+		if (bb_agc_ctl & AR_PHY_AGC_CONTROL_NO_UPDATE_NF)
+			REG_SET_BIT(ah, AR_PHY_AGC_CONTROL,
+				    AR_PHY_AGC_CONTROL_NO_UPDATE_NF);
+		REG_SET_BIT(ah, AR_PHY_AGC_CONTROL, AR_PHY_AGC_CONTROL_NF);
+		REG_RMW_BUFFER_FLUSH(ah);
+	}
+
+	/*
 	 * We timed out waiting for the noisefloor to load, probably due to an
 	 * in-progress rx. Simply return here and allow the load plenty of time
 	 * to complete before the next calibration interval.  We need to avoid
@@ -296,7 +322,7 @@ int ath9k_hw_loadnf(struct ath_hw *ah, struct ath9k_channel *chan)
 	 * here, the baseband nf cal will just be capped by our present
 	 * noisefloor until the next calibration timer.
 	 */
-	if (j == 10000) {
+	if (j == 22200) {
 		ath_dbg(common, ANY,
 			"Timeout while waiting for nf to load: AR_PHY_AGC_CONTROL=0x%x\n",
 			REG_READ(ah, AR_PHY_AGC_CONTROL));
-- 
cgit v0.10.2


From e9a26010f607621597f25e55d94faf85a2d0d5a4 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 12 Feb 2016 12:55:21 +0100
Subject: ath9k: reduce stack usage in ar9003_aic_cal_post_process

In some configurations, this function uses more than the warning limit
of 1024 bytes:

drivers/net/wireless/ath/ath9k/ar9003_aic.c: In function 'ar9003_aic_cal_post_process':
drivers/net/wireless/ath/ath9k/ar9003_aic.c:434:1: error: the frame size of 1040 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]

It turns out that there are two large arrays on the stack here, but
almost all the data in them is never used outside of the loop in
which it gets written, so we can replace the array with a single
instance.

The .valid flag is used later, so I'm replacing the array of structures
with an array of bools. An obvious follow-up optimization would be
to replace it with a bitmask and set_bit()/find_first_bit()/
find_last_bit()/... operations. However, I have not tested this patch,
so I sticked to the simpler transformation that does the job of
reducing the stack usage to a harmless level.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_aic.c b/drivers/net/wireless/ath/ath9k/ar9003_aic.c
index 1db119d..547cd46 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_aic.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_aic.c
@@ -53,19 +53,19 @@ static bool ar9003_hw_is_aic_enabled(struct ath_hw *ah)
 	return true;
 }
 
-static int16_t ar9003_aic_find_valid(struct ath_aic_sram_info *cal_sram,
+static int16_t ar9003_aic_find_valid(bool *cal_sram_valid,
 				     bool dir, u8 index)
 {
 	int16_t i;
 
 	if (dir) {
 		for (i = index + 1; i < ATH_AIC_MAX_BT_CHANNEL; i++) {
-			if (cal_sram[i].valid)
+			if (cal_sram_valid[i])
 				break;
 		}
 	} else {
 		for (i = index - 1; i >= 0; i--) {
-			if (cal_sram[i].valid)
+			if (cal_sram_valid[i])
 				break;
 		}
 	}
@@ -264,7 +264,7 @@ static u8 ar9003_aic_cal_start(struct ath_hw *ah, u8 min_valid_count)
 static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 {
 	struct ath9k_hw_aic *aic = &ah->btcoex_hw.aic;
-	struct ath_aic_sram_info cal_sram[ATH_AIC_MAX_BT_CHANNEL];
+	bool cal_sram_valid[ATH_AIC_MAX_BT_CHANNEL];
 	struct ath_aic_out_info aic_sram[ATH_AIC_MAX_BT_CHANNEL];
 	u32 dir_path_gain_idx, quad_path_gain_idx, value;
 	u32 fixed_com_att_db;
@@ -272,33 +272,34 @@ static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 	int16_t i;
 	bool ret = true;
 
-	memset(&cal_sram, 0, sizeof(cal_sram));
+	memset(&cal_sram_valid, 0, sizeof(cal_sram_valid));
 	memset(&aic_sram, 0, sizeof(aic_sram));
 
 	for (i = 0; i < ATH_AIC_MAX_BT_CHANNEL; i++) {
+		struct ath_aic_sram_info sram;
 		value = aic->aic_sram[i];
 
-		cal_sram[i].valid =
+		cal_sram_valid[i] = sram.valid =
 			MS(value, AR_PHY_AIC_SRAM_VALID);
-		cal_sram[i].rot_quad_att_db =
+		sram.rot_quad_att_db =
 			MS(value, AR_PHY_AIC_SRAM_ROT_QUAD_ATT_DB);
-		cal_sram[i].vga_quad_sign =
+		sram.vga_quad_sign =
 			MS(value, AR_PHY_AIC_SRAM_VGA_QUAD_SIGN);
-		cal_sram[i].rot_dir_att_db =
+		sram.rot_dir_att_db =
 			MS(value, AR_PHY_AIC_SRAM_ROT_DIR_ATT_DB);
-		cal_sram[i].vga_dir_sign =
+		sram.vga_dir_sign =
 			MS(value, AR_PHY_AIC_SRAM_VGA_DIR_SIGN);
-		cal_sram[i].com_att_6db =
+		sram.com_att_6db =
 			MS(value, AR_PHY_AIC_SRAM_COM_ATT_6DB);
 
-		if (cal_sram[i].valid) {
-			dir_path_gain_idx = cal_sram[i].rot_dir_att_db +
-				com_att_db_table[cal_sram[i].com_att_6db];
-			quad_path_gain_idx = cal_sram[i].rot_quad_att_db +
-				com_att_db_table[cal_sram[i].com_att_6db];
+		if (sram.valid) {
+			dir_path_gain_idx = sram.rot_dir_att_db +
+				com_att_db_table[sram.com_att_6db];
+			quad_path_gain_idx = sram.rot_quad_att_db +
+				com_att_db_table[sram.com_att_6db];
 
-			dir_path_sign = (cal_sram[i].vga_dir_sign) ? 1 : -1;
-			quad_path_sign = (cal_sram[i].vga_quad_sign) ? 1 : -1;
+			dir_path_sign = (sram.vga_dir_sign) ? 1 : -1;
+			quad_path_sign = (sram.vga_quad_sign) ? 1 : -1;
 
 			aic_sram[i].dir_path_gain_lin = dir_path_sign *
 				aic_lin_table[dir_path_gain_idx];
@@ -310,16 +311,16 @@ static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 	for (i = 0; i < ATH_AIC_MAX_BT_CHANNEL; i++) {
 		int16_t start_idx, end_idx;
 
-		if (cal_sram[i].valid)
+		if (cal_sram_valid[i])
 			continue;
 
-		start_idx = ar9003_aic_find_valid(cal_sram, 0, i);
-		end_idx = ar9003_aic_find_valid(cal_sram, 1, i);
+		start_idx = ar9003_aic_find_valid(cal_sram_valid, 0, i);
+		end_idx = ar9003_aic_find_valid(cal_sram_valid, 1, i);
 
 		if (start_idx < 0) {
 			/* extrapolation */
 			start_idx = end_idx;
-			end_idx = ar9003_aic_find_valid(cal_sram, 1, start_idx);
+			end_idx = ar9003_aic_find_valid(cal_sram_valid, 1, start_idx);
 
 			if (end_idx < 0) {
 				ret = false;
@@ -342,7 +343,7 @@ static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 
 		if (end_idx < 0) {
 			/* extrapolation */
-			end_idx = ar9003_aic_find_valid(cal_sram, 0, start_idx);
+			end_idx = ar9003_aic_find_valid(cal_sram_valid, 0, start_idx);
 
 			if (end_idx < 0) {
 				ret = false;
@@ -378,19 +379,21 @@ static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 	}
 
 	/* From dir/quad_path_gain_lin to sram. */
-	i = ar9003_aic_find_valid(cal_sram, 1, 0);
+	i = ar9003_aic_find_valid(cal_sram_valid, 1, 0);
 	if (i < 0) {
 		i = 0;
 		ret = false;
 	}
-	fixed_com_att_db = com_att_db_table[cal_sram[i].com_att_6db];
+	fixed_com_att_db = com_att_db_table[MS(aic->aic_sram[i],
+					    AR_PHY_AIC_SRAM_COM_ATT_6DB)];
 
 	for (i = 0; i < ATH_AIC_MAX_BT_CHANNEL; i++) {
 		int16_t rot_dir_path_att_db, rot_quad_path_att_db;
+		struct ath_aic_sram_info sram;
 
-		aic_sram[i].sram.vga_dir_sign =
+		sram.vga_dir_sign =
 			(aic_sram[i].dir_path_gain_lin >= 0) ? 1 : 0;
-		aic_sram[i].sram.vga_quad_sign=
+		sram.vga_quad_sign =
 			(aic_sram[i].quad_path_gain_lin >= 0) ? 1 : 0;
 
 		rot_dir_path_att_db =
@@ -400,33 +403,31 @@ static bool ar9003_aic_cal_post_process(struct ath_hw *ah)
 			ar9003_aic_find_index(0, abs(aic_sram[i].quad_path_gain_lin)) -
 			fixed_com_att_db;
 
-		aic_sram[i].sram.com_att_6db =
+		sram.com_att_6db =
 			ar9003_aic_find_index(1, fixed_com_att_db);
 
-		aic_sram[i].sram.valid = 1;
+		sram.valid = 1;
 
-		aic_sram[i].sram.rot_dir_att_db =
+		sram.rot_dir_att_db =
 			min(max(rot_dir_path_att_db,
 				(int16_t)ATH_AIC_MIN_ROT_DIR_ATT_DB),
 			    ATH_AIC_MAX_ROT_DIR_ATT_DB);
-		aic_sram[i].sram.rot_quad_att_db =
+		sram.rot_quad_att_db =
 			min(max(rot_quad_path_att_db,
 				(int16_t)ATH_AIC_MIN_ROT_QUAD_ATT_DB),
 			    ATH_AIC_MAX_ROT_QUAD_ATT_DB);
-	}
 
-	for (i = 0; i < ATH_AIC_MAX_BT_CHANNEL; i++) {
-		aic->aic_sram[i] = (SM(aic_sram[i].sram.vga_dir_sign,
+		aic->aic_sram[i] = (SM(sram.vga_dir_sign,
 				       AR_PHY_AIC_SRAM_VGA_DIR_SIGN) |
-				    SM(aic_sram[i].sram.vga_quad_sign,
+				    SM(sram.vga_quad_sign,
 				       AR_PHY_AIC_SRAM_VGA_QUAD_SIGN) |
-				    SM(aic_sram[i].sram.com_att_6db,
+				    SM(sram.com_att_6db,
 				       AR_PHY_AIC_SRAM_COM_ATT_6DB) |
-				    SM(aic_sram[i].sram.valid,
+				    SM(sram.valid,
 				       AR_PHY_AIC_SRAM_VALID) |
-				    SM(aic_sram[i].sram.rot_dir_att_db,
+				    SM(sram.rot_dir_att_db,
 				       AR_PHY_AIC_SRAM_ROT_DIR_ATT_DB) |
-				    SM(aic_sram[i].sram.rot_quad_att_db,
+				    SM(sram.rot_quad_att_db,
 				       AR_PHY_AIC_SRAM_ROT_QUAD_ATT_DB));
 	}
 
diff --git a/drivers/net/wireless/ath/ath9k/ar9003_aic.h b/drivers/net/wireless/ath/ath9k/ar9003_aic.h
index 86f4064..9512c63 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_aic.h
+++ b/drivers/net/wireless/ath/ath9k/ar9003_aic.h
@@ -50,7 +50,6 @@ struct ath_aic_sram_info {
 struct ath_aic_out_info {
 	int16_t dir_path_gain_lin;
 	int16_t quad_path_gain_lin;
-	struct ath_aic_sram_info sram;
 };
 
 u8 ar9003_aic_calibration(struct ath_hw *ah);
-- 
cgit v0.10.2


From 25c0f301425cecdc51d7cc6eb4fb163d464236bc Mon Sep 17 00:00:00 2001
From: Miaoqing Pan <miaoqing@codeaurora.org>
Date: Thu, 18 Feb 2016 17:20:02 +0800
Subject: ath9k: clear bb filter calibration power threshold

JP WiFi certification for bandwidth of channel 14 failed, the OBW
is lower than the requirement. Clear the bb filter calibration power
threshold to increase OBW(+2). The fix only for qca9531 chip now.

Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_phy.c b/drivers/net/wireless/ath/ath9k/ar9003_phy.c
index abd9646..06c1ca6 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_phy.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_phy.c
@@ -976,9 +976,14 @@ static int ar9003_hw_process_ini(struct ath_hw *ah,
 	/*
 	 * JAPAN regulatory.
 	 */
-	if (chan->channel == 2484)
+	if (chan->channel == 2484) {
 		ar9003_hw_prog_ini(ah, &ah->iniCckfirJapan2484, 1);
 
+		if (AR_SREV_9531(ah))
+			REG_RMW_FIELD(ah, AR_PHY_FCAL_2_0,
+				      AR_PHY_FLC_PWR_THRESH, 0);
+	}
+
 	ah->modes_index = modesIndex;
 	ar9003_hw_override_ini(ah);
 	ar9003_hw_set_channel_regs(ah, chan);
diff --git a/drivers/net/wireless/ath/ath9k/ar9003_phy.h b/drivers/net/wireless/ath/ath9k/ar9003_phy.h
index c5f8bc4..566da78 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_phy.h
+++ b/drivers/net/wireless/ath/ath9k/ar9003_phy.h
@@ -487,6 +487,9 @@
 #define AR_PHY_ADDAC_PARA_CTL    (AR_SM_BASE + 0x150)
 #define AR_PHY_XPA_CFG           (AR_SM_BASE + 0x158)
 
+#define AR_PHY_FLC_PWR_THRESH		7
+#define AR_PHY_FLC_PWR_THRESH_S		0
+
 #define AR_PHY_FRAME_CTL_CF_OVERLAP_WINDOW  3
 #define AR_PHY_FRAME_CTL_CF_OVERLAP_WINDOW_S    0
 
-- 
cgit v0.10.2


From 5eb4dce3b3471ec9d1ea2945fa3d2bab4ac7e100 Mon Sep 17 00:00:00 2001
From: John Fastabend <john.fastabend@gmail.com>
Date: Mon, 29 Feb 2016 11:26:13 -0800
Subject: net: relax setup_tc ndo op handle restriction

I added this check in setup_tc to multiple drivers,

 if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)

Unfortunately restricting to TC_H_ROOT like this breaks the old
instantiation of mqprio to setup a hardware qdisc. This patch
relaxes the test to only check the type to make it equivalent
to the check before I broke it. With this the old instantiation
continues to work.

A good smoke test is to setup mqprio with,

# tc qdisc add dev eth4 root mqprio num_tc 8 \
  map 0 1 2 3 4 5 6 7 \
  queues 0@0 1@1 2@2 3@3 4@4 5@5 6@6 7@7

Fixes: e4c6734eaab9 ("net: rework ndo tc op to consume additional qdisc handle paramete")
Reported-by: Singh Krishneil <krishneil.k.singh@intel.com>
Reported-by: Jake Keller <jacob.e.keller@intel.com>
CC: Murali Karicheri <m-karicheri2@ti.com>
CC: Shradha Shah <sshah@solarflare.com>
CC: Or Gerlitz <ogerlitz@mellanox.com>
CC: Ariel Elior <ariel.elior@qlogic.com>
CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
CC: Bruce Allan <bruce.w.allan@intel.com>
CC: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 3360684..ebf9224 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -1632,7 +1632,7 @@ static int xgbe_setup_tc(struct net_device *netdev, u32 handle, __be16 proto,
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	u8 tc;
 
-	if (handle != TC_H_ROOT || tc_to_netdev->type != TC_SETUP_MQPRIO)
+	if (tc_to_netdev->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	tc = tc_to_netdev->tc;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 45843d1..a949783 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -4275,7 +4275,7 @@ int bnx2x_setup_tc(struct net_device *dev, u8 num_tc)
 int __bnx2x_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 		     struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 	return bnx2x_setup_tc(dev, tc->tc);
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index aa6a318..ce6b075 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -5447,7 +5447,7 @@ static int bnxt_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 	struct bnxt *bp = netdev_priv(dev);
 	u8 tc;
 
-	if (handle != TC_H_ROOT || ntc->type != TC_SETUP_MQPRIO)
+	if (ntc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	tc = ntc->tc;
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index dc1a821..d09a8dd 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -1207,7 +1207,7 @@ err_queueing_scheme:
 static int __fm10k_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 			    struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	return fm10k_setup_tc(dev, tc->tc);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index b893ff8..4d6223d 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -8419,7 +8419,7 @@ int __ixgbe_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 		}
 	}
 
-	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	return ixgbe_setup_tc(dev, tc->tc);
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index e26b110..16b26d1 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -73,7 +73,7 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up)
 static int __mlx4_en_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 			      struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	return mlx4_en_setup_tc(dev, tc->tc);
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index 2cdb571..2337789 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -571,7 +571,7 @@ int efx_setup_tc(struct net_device *net_dev, u32 handle, __be16 proto,
 	unsigned tc, num_tc;
 	int rc;
 
-	if (handle != TC_H_ROOT || ntc->type != TC_SETUP_MQPRIO)
+	if (ntc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	num_tc = ntc->tc;
diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index ed0c30f..1d0942c 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -1860,7 +1860,7 @@ static int netcp_setup_tc(struct net_device *dev, u32 handle, __be16 proto,
 	/* setup tc must be called under rtnl lock */
 	ASSERT_RTNL();
 
-	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	/* Sanity-check the number of traffic classes requested */
-- 
cgit v0.10.2


From 3d1cbe839ac3aa7b23a274b691092c90f9bf1b8e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 2 Mar 2016 10:40:54 +0100
Subject: net: mellanox: add DEVLINK dependencies

The new NET_DEVLINK infrastructure can be a loadable module, but the drivers
using it might be built-in, which causes link errors like:

drivers/net/built-in.o: In function `mlx4_load_one':
:(.text+0x2fbfda): undefined reference to `devlink_port_register'
:(.text+0x2fc084): undefined reference to `devlink_port_unregister'
drivers/net/built-in.o: In function `mlxsw_sx_port_remove':
:(.text+0x33a03a): undefined reference to `devlink_port_type_clear'
:(.text+0x33a04e): undefined reference to `devlink_port_unregister'

There are multiple ways to avoid this:

a) add 'depends on NET_DEVLINK || !NET_DEVLINK' dependencies
   for each user
b) use 'select NET_DEVLINK' from each driver that uses it
   and hide the symbol in Kconfig.
c) make NET_DEVLINK a 'bool' option so we don't have to
   list it as a dependency, and rely on the APIs to be
   stubbed out when it is disabled
d) use IS_REACHABLE() rather than IS_ENABLED() to check for
   NET_DEVLINK in include/net/devlink.h

This implements a variation of approach a) by adding an
intermediate symbol that drivers can depend on, and changes
the three drivers using it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 09d4d087cd48 ("mlx4: Implement devlink interface")
Fixes: c4745500e988 ("mlxsw: Implement devlink interface")
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/infiniband/hw/mlx4/Kconfig b/drivers/infiniband/hw/mlx4/Kconfig
index fc01dea..db4aa13 100644
--- a/drivers/infiniband/hw/mlx4/Kconfig
+++ b/drivers/infiniband/hw/mlx4/Kconfig
@@ -1,6 +1,7 @@
 config MLX4_INFINIBAND
 	tristate "Mellanox ConnectX HCA support"
 	depends on NETDEVICES && ETHERNET && PCI && INET
+	depends on MAY_USE_DEVLINK
 	select NET_VENDOR_MELLANOX
 	select MLX4_CORE
 	---help---
diff --git a/drivers/net/ethernet/mellanox/mlx4/Kconfig b/drivers/net/ethernet/mellanox/mlx4/Kconfig
index 1486ce9..9ca3734 100644
--- a/drivers/net/ethernet/mellanox/mlx4/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx4/Kconfig
@@ -4,6 +4,7 @@
 
 config MLX4_EN
 	tristate "Mellanox Technologies 1/10/40Gbit Ethernet support"
+	depends on MAY_USE_DEVLINK
 	depends on PCI
 	select MLX4_CORE
 	select PTP_1588_CLOCK
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
index ce26adc..2ad7f67 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
@@ -4,6 +4,7 @@
 
 config MLXSW_CORE
 	tristate "Mellanox Technologies Switch ASICs support"
+	depends on MAY_USE_DEVLINK
 	---help---
 	  This driver supports Mellanox Technologies Switch ASICs family.
 
diff --git a/net/Kconfig b/net/Kconfig
index 6c9cfb0..2760825 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -403,6 +403,15 @@ config NET_DEVLINK
 	  infrastructure to support access to physical chip-wide config and
 	  monitoring.
 
+config MAY_USE_DEVLINK
+	tristate
+	default m if NET_DEVLINK=m
+	default y if NET_DEVLINK=y || NET_DEVLINK=n
+	help
+	  Drivers using the devlink infrastructure should have a dependency
+	  on MAY_USE_DEVLINK to ensure they do not cause link errors when
+	  devlink is a loadable module and the driver using it is built-in.
+
 endif   # if NET
 
 # Used by archs to tell that they support BPF_JIT
-- 
cgit v0.10.2


From 274ba628a32ee15bb8e08b8a9240c354066ed24a Mon Sep 17 00:00:00 2001
From: Simon Horman <horms+renesas@verge.net.au>
Date: Wed, 2 Mar 2016 10:28:13 +0900
Subject: sh_eth, ravb: Use ARCH_RENESAS

Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.

This is part of an ongoing process to migrate from ARCH_SHMOBILE to
ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.

Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/Kconfig b/drivers/net/ethernet/renesas/Kconfig
index 270c4c9..4f132cf 100644
--- a/drivers/net/ethernet/renesas/Kconfig
+++ b/drivers/net/ethernet/renesas/Kconfig
@@ -18,7 +18,7 @@ if NET_VENDOR_RENESAS
 config SH_ETH
 	tristate "Renesas SuperH Ethernet support"
 	depends on HAS_DMA
-	depends on ARCH_SHMOBILE || SUPERH || COMPILE_TEST
+	depends on ARCH_RENESAS || SUPERH || COMPILE_TEST
 	select CRC32
 	select MII
 	select MDIO_BITBANG
@@ -32,7 +32,7 @@ config SH_ETH
 config RAVB
 	tristate "Renesas Ethernet AVB support"
 	depends on HAS_DMA
-	depends on ARCH_SHMOBILE || COMPILE_TEST
+	depends on ARCH_RENESAS || COMPILE_TEST
 	select CRC32
 	select MII
 	select MDIO_BITBANG
-- 
cgit v0.10.2


From b54b8c2d6e3cd1db17405a402ba42da5a9c8bf1d Mon Sep 17 00:00:00 2001
From: Lada Trimasova <Lada.Trimasova@synopsys.com>
Date: Thu, 3 Mar 2016 17:07:46 +0300
Subject: net: ezchip: adapt driver to little endian architecture

Since ezchip network driver is written with big endian EZChip platform it
is necessary to add support for little endian architecture.

The first issue is that the order of the bits in a bit field is
implementation specific. So all the bit fields are removed.
Named constants are used to access necessary fields.

And the second one is that network byte order is big endian.
For example, data on ethernet is transmitted with most-significant
octet (byte) first. So in case of little endian architecture
it is important to swap data byte order when we read it from
register. In case of unaligned access we can use "get_unaligned_be32"
and in other case we can use function "ioread32_rep" which reads all
data from register and works either with little endian or big endian
architecture.

And then when we are going to write data to register we need to restore
byte order using the function "put_unaligned_be32" in case of
unaligned access and in other case "iowrite32_rep".

The last little fix is a space between type and pointer to observe
coding style.

Signed-off-by: Lada Trimasova <ltrimas@synopsys.com>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Noam Camus <noamc@ezchip.com>
Cc: Tal Zilcer <talz@ezchip.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/ezchip/nps_enet.c b/drivers/net/ethernet/ezchip/nps_enet.c
index b102668..1f23845a 100644
--- a/drivers/net/ethernet/ezchip/nps_enet.c
+++ b/drivers/net/ethernet/ezchip/nps_enet.c
@@ -43,20 +43,21 @@ static void nps_enet_read_rx_fifo(struct net_device *ndev,
 	bool dst_is_aligned = IS_ALIGNED((unsigned long)dst, sizeof(u32));
 
 	/* In case dst is not aligned we need an intermediate buffer */
-	if (dst_is_aligned)
-		for (i = 0; i < len; i++, reg++)
-			*reg = nps_enet_reg_get(priv, NPS_ENET_REG_RX_BUF);
+	if (dst_is_aligned) {
+		ioread32_rep(priv->regs_base + NPS_ENET_REG_RX_BUF, reg, len);
+		reg += len;
+	}
 	else { /* !dst_is_aligned */
 		for (i = 0; i < len; i++, reg++) {
 			u32 buf = nps_enet_reg_get(priv, NPS_ENET_REG_RX_BUF);
-			put_unaligned(buf, reg);
+			put_unaligned_be32(buf, reg);
 		}
 	}
-
 	/* copy last bytes (if any) */
 	if (last) {
-		u32 buf = nps_enet_reg_get(priv, NPS_ENET_REG_RX_BUF);
-		memcpy((u8*)reg, &buf, last);
+		u32 buf;
+		ioread32_rep(priv->regs_base + NPS_ENET_REG_RX_BUF, &buf, 1);
+		memcpy((u8 *)reg, &buf, last);
 	}
 }
 
@@ -66,26 +67,28 @@ static u32 nps_enet_rx_handler(struct net_device *ndev)
 	u32 work_done = 0;
 	struct nps_enet_priv *priv = netdev_priv(ndev);
 	struct sk_buff *skb;
-	struct nps_enet_rx_ctl rx_ctrl;
+	u32 rx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_RX_CTL);
+	u32 rx_ctrl_cr = (rx_ctrl_value & RX_CTL_CR_MASK) >> RX_CTL_CR_SHIFT;
+	u32 rx_ctrl_er = (rx_ctrl_value & RX_CTL_ER_MASK) >> RX_CTL_ER_SHIFT;
+	u32 rx_ctrl_crc = (rx_ctrl_value & RX_CTL_CRC_MASK) >> RX_CTL_CRC_SHIFT;
 
-	rx_ctrl.value = nps_enet_reg_get(priv, NPS_ENET_REG_RX_CTL);
-	frame_len = rx_ctrl.nr;
+	frame_len = (rx_ctrl_value & RX_CTL_NR_MASK) >> RX_CTL_NR_SHIFT;
 
 	/* Check if we got RX */
-	if (!rx_ctrl.cr)
+	if (!rx_ctrl_cr)
 		return work_done;
 
 	/* If we got here there is a work for us */
 	work_done++;
 
 	/* Check Rx error */
-	if (rx_ctrl.er) {
+	if (rx_ctrl_er) {
 		ndev->stats.rx_errors++;
 		err = 1;
 	}
 
 	/* Check Rx CRC error */
-	if (rx_ctrl.crc) {
+	if (rx_ctrl_crc) {
 		ndev->stats.rx_crc_errors++;
 		ndev->stats.rx_dropped++;
 		err = 1;
@@ -136,23 +139,24 @@ rx_irq_frame_done:
 static void nps_enet_tx_handler(struct net_device *ndev)
 {
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_tx_ctl tx_ctrl;
-
-	tx_ctrl.value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL);
+	u32 tx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL);
+	u32 tx_ctrl_ct = (tx_ctrl_value & TX_CTL_CT_MASK) >> TX_CTL_CT_SHIFT;
+	u32 tx_ctrl_et = (tx_ctrl_value & TX_CTL_ET_MASK) >> TX_CTL_ET_SHIFT;
+	u32 tx_ctrl_nt = (tx_ctrl_value & TX_CTL_NT_MASK) >> TX_CTL_NT_SHIFT;
 
 	/* Check if we got TX */
-	if (!priv->tx_packet_sent || tx_ctrl.ct)
+	if (!priv->tx_packet_sent || tx_ctrl_ct)
 		return;
 
 	/* Ack Tx ctrl register */
 	nps_enet_reg_set(priv, NPS_ENET_REG_TX_CTL, 0);
 
 	/* Check Tx transmit error */
-	if (unlikely(tx_ctrl.et)) {
+	if (unlikely(tx_ctrl_et)) {
 		ndev->stats.tx_errors++;
 	} else {
 		ndev->stats.tx_packets++;
-		ndev->stats.tx_bytes += tx_ctrl.nt;
+		ndev->stats.tx_bytes += tx_ctrl_nt;
 	}
 
 	dev_kfree_skb(priv->tx_skb);
@@ -178,13 +182,16 @@ static int nps_enet_poll(struct napi_struct *napi, int budget)
 	nps_enet_tx_handler(ndev);
 	work_done = nps_enet_rx_handler(ndev);
 	if (work_done < budget) {
-		struct nps_enet_buf_int_enable buf_int_enable;
+		u32 buf_int_enable_value = 0;
 
 		napi_complete(napi);
-		buf_int_enable.rx_rdy = NPS_ENET_ENABLE;
-		buf_int_enable.tx_done = NPS_ENET_ENABLE;
+
+		/* set tx_done and rx_rdy bits */
+		buf_int_enable_value |= NPS_ENET_ENABLE << RX_RDY_SHIFT;
+		buf_int_enable_value |= NPS_ENET_ENABLE << TX_DONE_SHIFT;
+
 		nps_enet_reg_set(priv, NPS_ENET_REG_BUF_INT_ENABLE,
-				 buf_int_enable.value);
+				 buf_int_enable_value);
 	}
 
 	return work_done;
@@ -205,13 +212,12 @@ static irqreturn_t nps_enet_irq_handler(s32 irq, void *dev_instance)
 {
 	struct net_device *ndev = dev_instance;
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_rx_ctl rx_ctrl;
-	struct nps_enet_tx_ctl tx_ctrl;
-
-	rx_ctrl.value = nps_enet_reg_get(priv, NPS_ENET_REG_RX_CTL);
-	tx_ctrl.value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL);
+	u32 rx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_RX_CTL);
+	u32 tx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL);
+	u32 tx_ctrl_ct = (tx_ctrl_value & TX_CTL_CT_MASK) >> TX_CTL_CT_SHIFT;
+	u32 rx_ctrl_cr = (rx_ctrl_value & RX_CTL_CR_MASK) >> RX_CTL_CR_SHIFT;
 
-	if ((!tx_ctrl.ct && priv->tx_packet_sent) || rx_ctrl.cr)
+	if ((!tx_ctrl_ct && priv->tx_packet_sent) || rx_ctrl_cr)
 		if (likely(napi_schedule_prep(&priv->napi))) {
 			nps_enet_reg_set(priv, NPS_ENET_REG_BUF_INT_ENABLE, 0);
 			__napi_schedule(&priv->napi);
@@ -223,22 +229,24 @@ static irqreturn_t nps_enet_irq_handler(s32 irq, void *dev_instance)
 static void nps_enet_set_hw_mac_address(struct net_device *ndev)
 {
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_ge_mac_cfg_1 ge_mac_cfg_1;
-	struct nps_enet_ge_mac_cfg_2 *ge_mac_cfg_2 = &priv->ge_mac_cfg_2;
+	u32 ge_mac_cfg_1_value = 0;
+	u32 *ge_mac_cfg_2_value = &priv->ge_mac_cfg_2_value;
 
 	/* set MAC address in HW */
-	ge_mac_cfg_1.octet_0 = ndev->dev_addr[0];
-	ge_mac_cfg_1.octet_1 = ndev->dev_addr[1];
-	ge_mac_cfg_1.octet_2 = ndev->dev_addr[2];
-	ge_mac_cfg_1.octet_3 = ndev->dev_addr[3];
-	ge_mac_cfg_2->octet_4 = ndev->dev_addr[4];
-	ge_mac_cfg_2->octet_5 = ndev->dev_addr[5];
+	ge_mac_cfg_1_value |= ndev->dev_addr[0] << CFG_1_OCTET_0_SHIFT;
+	ge_mac_cfg_1_value |= ndev->dev_addr[1] << CFG_1_OCTET_1_SHIFT;
+	ge_mac_cfg_1_value |= ndev->dev_addr[2] << CFG_1_OCTET_2_SHIFT;
+	ge_mac_cfg_1_value |= ndev->dev_addr[3] << CFG_1_OCTET_3_SHIFT;
+	*ge_mac_cfg_2_value = (*ge_mac_cfg_2_value & ~CFG_2_OCTET_4_MASK)
+		 | ndev->dev_addr[4] << CFG_2_OCTET_4_SHIFT;
+	*ge_mac_cfg_2_value = (*ge_mac_cfg_2_value & ~CFG_2_OCTET_5_MASK)
+		 | ndev->dev_addr[5] << CFG_2_OCTET_5_SHIFT;
 
 	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_1,
-			 ge_mac_cfg_1.value);
+			 ge_mac_cfg_1_value);
 
 	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_2,
-			 ge_mac_cfg_2->value);
+			 *ge_mac_cfg_2_value);
 }
 
 /**
@@ -254,93 +262,97 @@ static void nps_enet_set_hw_mac_address(struct net_device *ndev)
 static void nps_enet_hw_reset(struct net_device *ndev)
 {
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_ge_rst ge_rst;
-	struct nps_enet_phase_fifo_ctl phase_fifo_ctl;
+	u32 ge_rst_value = 0, phase_fifo_ctl_value = 0;
 
-	ge_rst.value = 0;
-	phase_fifo_ctl.value = 0;
 	/* Pcs reset sequence*/
-	ge_rst.gmac_0 = NPS_ENET_ENABLE;
-	nps_enet_reg_set(priv, NPS_ENET_REG_GE_RST, ge_rst.value);
+	ge_rst_value |= NPS_ENET_ENABLE << RST_GMAC_0_SHIFT;
+	nps_enet_reg_set(priv, NPS_ENET_REG_GE_RST, ge_rst_value);
 	usleep_range(10, 20);
-	ge_rst.value = 0;
-	nps_enet_reg_set(priv, NPS_ENET_REG_GE_RST, ge_rst.value);
+	nps_enet_reg_set(priv, NPS_ENET_REG_GE_RST, ge_rst_value);
 
 	/* Tx fifo reset sequence */
-	phase_fifo_ctl.rst = NPS_ENET_ENABLE;
-	phase_fifo_ctl.init = NPS_ENET_ENABLE;
+	phase_fifo_ctl_value |= NPS_ENET_ENABLE << PHASE_FIFO_CTL_RST_SHIFT;
+	phase_fifo_ctl_value |= NPS_ENET_ENABLE << PHASE_FIFO_CTL_INIT_SHIFT;
 	nps_enet_reg_set(priv, NPS_ENET_REG_PHASE_FIFO_CTL,
-			 phase_fifo_ctl.value);
+			 phase_fifo_ctl_value);
 	usleep_range(10, 20);
-	phase_fifo_ctl.value = 0;
+	phase_fifo_ctl_value = 0;
 	nps_enet_reg_set(priv, NPS_ENET_REG_PHASE_FIFO_CTL,
-			 phase_fifo_ctl.value);
+			 phase_fifo_ctl_value);
 }
 
 static void nps_enet_hw_enable_control(struct net_device *ndev)
 {
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_ge_mac_cfg_0 ge_mac_cfg_0;
-	struct nps_enet_buf_int_enable buf_int_enable;
-	struct nps_enet_ge_mac_cfg_2 *ge_mac_cfg_2 = &priv->ge_mac_cfg_2;
-	struct nps_enet_ge_mac_cfg_3 *ge_mac_cfg_3 = &priv->ge_mac_cfg_3;
+	u32 ge_mac_cfg_0_value = 0, buf_int_enable_value = 0;
+	u32 *ge_mac_cfg_2_value = &priv->ge_mac_cfg_2_value;
+	u32 *ge_mac_cfg_3_value = &priv->ge_mac_cfg_3_value;
 	s32 max_frame_length;
 
-	ge_mac_cfg_0.value = 0;
-	buf_int_enable.value = 0;
 	/* Enable Rx and Tx statistics */
-	ge_mac_cfg_2->stat_en = NPS_ENET_GE_MAC_CFG_2_STAT_EN;
+	*ge_mac_cfg_2_value = (*ge_mac_cfg_2_value & ~CFG_2_STAT_EN_MASK)
+		 | NPS_ENET_GE_MAC_CFG_2_STAT_EN << CFG_2_STAT_EN_SHIFT;
 
 	/* Discard packets with different MAC address */
-	ge_mac_cfg_2->disc_da = NPS_ENET_ENABLE;
+	*ge_mac_cfg_2_value = (*ge_mac_cfg_2_value & ~CFG_2_DISK_DA_MASK)
+		 | NPS_ENET_ENABLE << CFG_2_DISK_DA_SHIFT;
 
 	/* Discard multicast packets */
-	ge_mac_cfg_2->disc_mc = NPS_ENET_ENABLE;
+	*ge_mac_cfg_2_value = (*ge_mac_cfg_2_value & ~CFG_2_DISK_MC_MASK)
+		 | NPS_ENET_ENABLE << CFG_2_DISK_MC_SHIFT;
 
 	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_2,
-			 ge_mac_cfg_2->value);
+			 *ge_mac_cfg_2_value);
 
 	/* Discard Packets bigger than max frame length */
 	max_frame_length = ETH_HLEN + ndev->mtu + ETH_FCS_LEN;
-	if (max_frame_length <= NPS_ENET_MAX_FRAME_LENGTH)
-		ge_mac_cfg_3->max_len = max_frame_length;
+	if (max_frame_length <= NPS_ENET_MAX_FRAME_LENGTH) {
+		*ge_mac_cfg_3_value =
+			 (*ge_mac_cfg_3_value & ~CFG_3_MAX_LEN_MASK)
+			 | max_frame_length << CFG_3_MAX_LEN_SHIFT;
+	}
 
 	/* Enable interrupts */
-	buf_int_enable.rx_rdy = NPS_ENET_ENABLE;
-	buf_int_enable.tx_done = NPS_ENET_ENABLE;
+	buf_int_enable_value |= NPS_ENET_ENABLE << RX_RDY_SHIFT;
+	buf_int_enable_value |= NPS_ENET_ENABLE << TX_DONE_SHIFT;
 	nps_enet_reg_set(priv, NPS_ENET_REG_BUF_INT_ENABLE,
-			 buf_int_enable.value);
+			 buf_int_enable_value);
 
 	/* Write device MAC address to HW */
 	nps_enet_set_hw_mac_address(ndev);
 
 	/* Rx and Tx HW features */
-	ge_mac_cfg_0.tx_pad_en = NPS_ENET_ENABLE;
-	ge_mac_cfg_0.tx_crc_en = NPS_ENET_ENABLE;
-	ge_mac_cfg_0.rx_crc_strip = NPS_ENET_ENABLE;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_TX_PAD_EN_SHIFT;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_TX_CRC_EN_SHIFT;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_RX_CRC_STRIP_SHIFT;
 
 	/* IFG configuration */
-	ge_mac_cfg_0.rx_ifg = NPS_ENET_GE_MAC_CFG_0_RX_IFG;
-	ge_mac_cfg_0.tx_ifg = NPS_ENET_GE_MAC_CFG_0_TX_IFG;
+	ge_mac_cfg_0_value |=
+		 NPS_ENET_GE_MAC_CFG_0_RX_IFG << CFG_0_RX_IFG_SHIFT;
+	ge_mac_cfg_0_value |=
+		 NPS_ENET_GE_MAC_CFG_0_TX_IFG << CFG_0_TX_IFG_SHIFT;
 
 	/* preamble configuration */
-	ge_mac_cfg_0.rx_pr_check_en = NPS_ENET_ENABLE;
-	ge_mac_cfg_0.tx_pr_len = NPS_ENET_GE_MAC_CFG_0_TX_PR_LEN;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_RX_PR_CHECK_EN_SHIFT;
+	ge_mac_cfg_0_value |=
+		 NPS_ENET_GE_MAC_CFG_0_TX_PR_LEN << CFG_0_TX_PR_LEN_SHIFT;
 
 	/* enable flow control frames */
-	ge_mac_cfg_0.tx_fc_en = NPS_ENET_ENABLE;
-	ge_mac_cfg_0.rx_fc_en = NPS_ENET_ENABLE;
-	ge_mac_cfg_0.tx_fc_retr = NPS_ENET_GE_MAC_CFG_0_TX_FC_RETR;
-	ge_mac_cfg_3->cf_drop = NPS_ENET_ENABLE;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_TX_FC_EN_SHIFT;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_RX_FC_EN_SHIFT;
+	ge_mac_cfg_0_value |=
+		 NPS_ENET_GE_MAC_CFG_0_TX_FC_RETR << CFG_0_TX_FC_RETR_SHIFT;
+	*ge_mac_cfg_3_value = (*ge_mac_cfg_3_value & ~CFG_3_CF_DROP_MASK)
+		 | NPS_ENET_ENABLE << CFG_3_CF_DROP_SHIFT;
 
 	/* Enable Rx and Tx */
-	ge_mac_cfg_0.rx_en = NPS_ENET_ENABLE;
-	ge_mac_cfg_0.tx_en = NPS_ENET_ENABLE;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_RX_EN_SHIFT;
+	ge_mac_cfg_0_value |= NPS_ENET_ENABLE << CFG_0_TX_EN_SHIFT;
 
 	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_3,
-			 ge_mac_cfg_3->value);
+			 *ge_mac_cfg_3_value);
 	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_0,
-			 ge_mac_cfg_0.value);
+			 ge_mac_cfg_0_value);
 }
 
 static void nps_enet_hw_disable_control(struct net_device *ndev)
@@ -358,31 +370,28 @@ static void nps_enet_send_frame(struct net_device *ndev,
 				struct sk_buff *skb)
 {
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_tx_ctl tx_ctrl;
+	u32 tx_ctrl_value = 0;
 	short length = skb->len;
 	u32 i, len = DIV_ROUND_UP(length, sizeof(u32));
 	u32 *src = (void *)skb->data;
 	bool src_is_aligned = IS_ALIGNED((unsigned long)src, sizeof(u32));
 
-	tx_ctrl.value = 0;
 	/* In case src is not aligned we need an intermediate buffer */
 	if (src_is_aligned)
-		for (i = 0; i < len; i++, src++)
-			nps_enet_reg_set(priv, NPS_ENET_REG_TX_BUF, *src);
+		iowrite32_rep(priv->regs_base + NPS_ENET_REG_TX_BUF, src, len);
 	else /* !src_is_aligned */
 		for (i = 0; i < len; i++, src++)
 			nps_enet_reg_set(priv, NPS_ENET_REG_TX_BUF,
-					 get_unaligned(src));
+					 get_unaligned_be32(src));
 
 	/* Write the length of the Frame */
-	tx_ctrl.nt = length;
+	tx_ctrl_value |= length << TX_CTL_NT_SHIFT;
 
 	/* Indicate SW is done */
 	priv->tx_packet_sent = true;
-	tx_ctrl.ct = NPS_ENET_ENABLE;
-
+	tx_ctrl_value |= NPS_ENET_ENABLE << TX_CTL_CT_SHIFT;
 	/* Send Frame */
-	nps_enet_reg_set(priv, NPS_ENET_REG_TX_CTL, tx_ctrl.value);
+	nps_enet_reg_set(priv, NPS_ENET_REG_TX_CTL, tx_ctrl_value);
 }
 
 /**
@@ -422,19 +431,23 @@ static s32 nps_enet_set_mac_address(struct net_device *ndev, void *p)
 static void nps_enet_set_rx_mode(struct net_device *ndev)
 {
 	struct nps_enet_priv *priv = netdev_priv(ndev);
-	struct nps_enet_ge_mac_cfg_2 ge_mac_cfg_2;
-
-	ge_mac_cfg_2.value = priv->ge_mac_cfg_2.value;
+	u32 ge_mac_cfg_2_value = priv->ge_mac_cfg_2_value;
 
 	if (ndev->flags & IFF_PROMISC) {
-		ge_mac_cfg_2.disc_da = NPS_ENET_DISABLE;
-		ge_mac_cfg_2.disc_mc = NPS_ENET_DISABLE;
+		ge_mac_cfg_2_value = (ge_mac_cfg_2_value & ~CFG_2_DISK_DA_MASK)
+			 | NPS_ENET_DISABLE << CFG_2_DISK_DA_SHIFT;
+		ge_mac_cfg_2_value = (ge_mac_cfg_2_value & ~CFG_2_DISK_MC_MASK)
+			 | NPS_ENET_DISABLE << CFG_2_DISK_MC_SHIFT;
+
 	} else {
-		ge_mac_cfg_2.disc_da = NPS_ENET_ENABLE;
-		ge_mac_cfg_2.disc_mc = NPS_ENET_ENABLE;
+		ge_mac_cfg_2_value = (ge_mac_cfg_2_value & ~CFG_2_DISK_DA_MASK)
+			 | NPS_ENET_ENABLE << CFG_2_DISK_DA_SHIFT;
+		ge_mac_cfg_2_value = (ge_mac_cfg_2_value & ~CFG_2_DISK_MC_MASK)
+			 | NPS_ENET_ENABLE << CFG_2_DISK_MC_SHIFT;
+
 	}
 
-	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_2, ge_mac_cfg_2.value);
+	nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_2, ge_mac_cfg_2_value);
 }
 
 /**
@@ -453,12 +466,15 @@ static s32 nps_enet_open(struct net_device *ndev)
 
 	/* Reset private variables */
 	priv->tx_packet_sent = false;
-	priv->ge_mac_cfg_2.value = 0;
-	priv->ge_mac_cfg_3.value = 0;
+	priv->ge_mac_cfg_2_value = 0;
+	priv->ge_mac_cfg_3_value = 0;
 
 	/* ge_mac_cfg_3 default values */
-	priv->ge_mac_cfg_3.rx_ifg_th = NPS_ENET_GE_MAC_CFG_3_RX_IFG_TH;
-	priv->ge_mac_cfg_3.max_len = NPS_ENET_GE_MAC_CFG_3_MAX_LEN;
+	priv->ge_mac_cfg_3_value |=
+		 NPS_ENET_GE_MAC_CFG_3_RX_IFG_TH << CFG_3_RX_IFG_TH_SHIFT;
+
+	priv->ge_mac_cfg_3_value |=
+		 NPS_ENET_GE_MAC_CFG_3_MAX_LEN << CFG_3_MAX_LEN_SHIFT;
 
 	/* Disable HW device */
 	nps_enet_hw_disable_control(ndev);
diff --git a/drivers/net/ethernet/ezchip/nps_enet.h b/drivers/net/ethernet/ezchip/nps_enet.h
index 6703674..d0cab60 100644
--- a/drivers/net/ethernet/ezchip/nps_enet.h
+++ b/drivers/net/ethernet/ezchip/nps_enet.h
@@ -43,233 +43,123 @@
 #define NPS_ENET_REG_GE_RST		0x1400
 #define NPS_ENET_REG_PHASE_FIFO_CTL	0x1404
 
-/* Tx control register */
-struct nps_enet_tx_ctl {
-	union {
-		/* ct: SW sets to indicate frame ready in Tx buffer for
-		 *     transmission. HW resets to when transmission done
-		 * et: Transmit error
-		 * nt: Length in bytes of Tx frame loaded to Tx buffer
-		 */
-		struct {
-			u32
-			__reserved_1:16,
-			ct:1,
-			et:1,
-			__reserved_2:3,
-			nt:11;
-		};
-
-		u32 value;
-	};
-};
-
-/* Rx control register */
-struct nps_enet_rx_ctl {
-	union {
-		/* cr:  HW sets to indicate frame ready in Rx buffer.
-		 *      SW resets to indicate host read received frame
-		 *      and new frames can be written to Rx buffer
-		 * er:  Rx error indication
-		 * crc: Rx CRC error indication
-		 * nr:  Length in bytes of Rx frame loaded by MAC to Rx buffer
-		 */
-		struct {
-			u32
-			__reserved_1:16,
-			cr:1,
-			er:1,
-			crc:1,
-			__reserved_2:2,
-			nr:11;
-		};
-
-		u32 value;
-	};
-};
-
-/* Interrupt enable for data buffer events register */
-struct nps_enet_buf_int_enable {
-	union {
-		/* tx_done: Interrupt generation in the case when new frame
-		 *          is ready in Rx buffer
-		 * rx_rdy:  Interrupt generation in the case when current frame
-		 *          was read from TX buffer
-		 */
-		struct {
-			u32
-			__reserved:30,
-			tx_done:1,
-			rx_rdy:1;
-		};
-
-		u32 value;
-	};
-};
-
-/* Gbps Eth MAC Configuration 0 register */
-struct nps_enet_ge_mac_cfg_0 {
-	union {
-		/* tx_pr_len:          Transmit preamble length in bytes
-		 * tx_ifg_nib:         Tx idle pattern
-		 * nib_mode:           Nibble (4-bit) Mode
-		 * rx_pr_check_en:     Receive preamble Check Enable
-		 * tx_ifg:             Transmit inter-Frame Gap
-		 * rx_ifg:             Receive inter-Frame Gap
-		 * tx_fc_retr:         Transmit Flow Control Retransmit Mode
-		 * rx_length_check_en: Receive Length Check Enable
-		 * rx_crc_ignore:      Results of the CRC check are ignored
-		 * rx_crc_strip:       MAC strips the CRC from received frames
-		 * rx_fc_en:           Receive Flow Control Enable
-		 * tx_crc_en:          Transmit CRC Enabled
-		 * tx_pad_en:          Transmit Padding Enable
-		 * tx_cf_en:           Transmit Flow Control Enable
-		 * tx_en:              Transmit Enable
-		 * rx_en:              Receive Enable
-		 */
-		struct {
-			u32
-			tx_pr_len:4,
-			tx_ifg_nib:4,
-			nib_mode:1,
-			rx_pr_check_en:1,
-			tx_ifg:6,
-			rx_ifg:4,
-			tx_fc_retr:3,
-			rx_length_check_en:1,
-			rx_crc_ignore:1,
-			rx_crc_strip:1,
-			rx_fc_en:1,
-			tx_crc_en:1,
-			tx_pad_en:1,
-			tx_fc_en:1,
-			tx_en:1,
-			rx_en:1;
-		};
-
-		u32 value;
-	};
-};
-
-/* Gbps Eth MAC Configuration 1 register */
-struct nps_enet_ge_mac_cfg_1 {
-	union {
-		/* octet_3: MAC address octet 3
-		 * octet_2: MAC address octet 2
-		 * octet_1: MAC address octet 1
-		 * octet_0: MAC address octet 0
-		 */
-		struct {
-			u32
-			octet_3:8,
-			octet_2:8,
-			octet_1:8,
-			octet_0:8;
-		};
-
-		u32 value;
-	};
-};
-
-/* Gbps Eth MAC Configuration 2 register */
-struct nps_enet_ge_mac_cfg_2 {
-	union {
-		/* transmit_flush_en: MAC flush enable
-		 * stat_en:           RMON statistics interface enable
-		 * disc_da:           Discard frames with DA different
-		 *                    from MAC address
-		 * disc_bc:           Discard broadcast frames
-		 * disc_mc:           Discard multicast frames
-		 * octet_5:           MAC address octet 5
-		 * octet_4:           MAC address octet 4
-		 */
-		struct {
-			u32
-			transmit_flush_en:1,
-			__reserved_1:5,
-			stat_en:2,
-			__reserved_2:1,
-			disc_da:1,
-			disc_bc:1,
-			disc_mc:1,
-			__reserved_3:4,
-			octet_5:8,
-			octet_4:8;
-		};
-
-		u32 value;
-	};
-};
-
-/* Gbps Eth MAC Configuration 3 register */
-struct nps_enet_ge_mac_cfg_3 {
-	union {
-		/* ext_oob_cbfc_sel:  Selects one of the 4 profiles for
-		 *                    extended OOB in-flow-control indication
-		 * max_len:           Maximum receive frame length in bytes
-		 * tx_cbfc_en:        Enable transmission of class-based
-		 *                    flow control packets
-		 * rx_ifg_th:         Threshold for IFG status reporting via OOB
-		 * cf_timeout:        Configurable time to decrement FC counters
-		 * cf_drop:           Drop control frames
-		 * redirect_cbfc_sel: Selects one of CBFC redirect profiles
-		 * rx_cbfc_redir_en:  Enable Rx class-based flow
-		 *                    control redirect
-		 * rx_cbfc_en:        Enable Rx class-based flow control
-		 * tm_hd_mode:        TM header mode
-		 */
-		struct {
-			u32
-			ext_oob_cbfc_sel:2,
-			max_len:14,
-			tx_cbfc_en:1,
-			rx_ifg_th:5,
-			cf_timeout:4,
-			cf_drop:1,
-			redirect_cbfc_sel:2,
-			rx_cbfc_redir_en:1,
-			rx_cbfc_en:1,
-			tm_hd_mode:1;
-		};
-
-		u32 value;
-	};
-};
-
-/* GE MAC, PCS reset control register */
-struct nps_enet_ge_rst {
-	union {
-		/* gmac_0: GE MAC reset
-		 * spcs_0: SGMII PCS reset
-		 */
-		struct {
-			u32
-			__reserved_1:23,
-			gmac_0:1,
-			__reserved_2:7,
-			spcs_0:1;
-		};
-
-		u32 value;
-	};
-};
-
-/* Tx phase sync FIFO control register */
-struct nps_enet_phase_fifo_ctl {
-	union {
-		/* init: initialize serdes TX phase sync FIFO pointers
-		 * rst:  reset serdes TX phase sync FIFO
-		 */
-		struct {
-			u32
-			__reserved:30,
-			init:1,
-			rst:1;
-		};
-
-		u32 value;
-	};
-};
+/* Tx control register masks and shifts */
+#define TX_CTL_NT_MASK 0x7FF
+#define TX_CTL_NT_SHIFT 0
+#define TX_CTL_ET_MASK 0x4000
+#define TX_CTL_ET_SHIFT 14
+#define TX_CTL_CT_MASK 0x8000
+#define TX_CTL_CT_SHIFT 15
+
+/* Rx control register masks and shifts */
+#define RX_CTL_NR_MASK 0x7FF
+#define RX_CTL_NR_SHIFT 0
+#define RX_CTL_CRC_MASK 0x2000
+#define RX_CTL_CRC_SHIFT 13
+#define RX_CTL_ER_MASK 0x4000
+#define RX_CTL_ER_SHIFT 14
+#define RX_CTL_CR_MASK 0x8000
+#define RX_CTL_CR_SHIFT 15
+
+/* Interrupt enable for data buffer events register masks and shifts */
+#define RX_RDY_MASK 0x1
+#define RX_RDY_SHIFT 0
+#define TX_DONE_MASK 0x2
+#define TX_DONE_SHIFT 1
+
+/* Gbps Eth MAC Configuration 0 register masks and shifts */
+#define CFG_0_RX_EN_MASK 0x1
+#define CFG_0_RX_EN_SHIFT 0
+#define CFG_0_TX_EN_MASK 0x2
+#define CFG_0_TX_EN_SHIFT 1
+#define CFG_0_TX_FC_EN_MASK 0x4
+#define CFG_0_TX_FC_EN_SHIFT 2
+#define CFG_0_TX_PAD_EN_MASK 0x8
+#define CFG_0_TX_PAD_EN_SHIFT 3
+#define CFG_0_TX_CRC_EN_MASK 0x10
+#define CFG_0_TX_CRC_EN_SHIFT 4
+#define CFG_0_RX_FC_EN_MASK 0x20
+#define CFG_0_RX_FC_EN_SHIFT 5
+#define CFG_0_RX_CRC_STRIP_MASK 0x40
+#define CFG_0_RX_CRC_STRIP_SHIFT 6
+#define CFG_0_RX_CRC_IGNORE_MASK 0x80
+#define CFG_0_RX_CRC_IGNORE_SHIFT 7
+#define CFG_0_RX_LENGTH_CHECK_EN_MASK 0x100
+#define CFG_0_RX_LENGTH_CHECK_EN_SHIFT 8
+#define CFG_0_TX_FC_RETR_MASK 0xE00
+#define CFG_0_TX_FC_RETR_SHIFT 9
+#define CFG_0_RX_IFG_MASK 0xF000
+#define CFG_0_RX_IFG_SHIFT 12
+#define CFG_0_TX_IFG_MASK 0x3F0000
+#define CFG_0_TX_IFG_SHIFT 16
+#define CFG_0_RX_PR_CHECK_EN_MASK 0x400000
+#define CFG_0_RX_PR_CHECK_EN_SHIFT 22
+#define CFG_0_NIB_MODE_MASK 0x800000
+#define CFG_0_NIB_MODE_SHIFT 23
+#define CFG_0_TX_IFG_NIB_MASK 0xF000000
+#define CFG_0_TX_IFG_NIB_SHIFT 24
+#define CFG_0_TX_PR_LEN_MASK 0xF0000000
+#define CFG_0_TX_PR_LEN_SHIFT 28
+
+/* Gbps Eth MAC Configuration 1 register masks and shifts */
+#define CFG_1_OCTET_0_MASK 0x000000FF
+#define CFG_1_OCTET_0_SHIFT 0
+#define CFG_1_OCTET_1_MASK 0x0000FF00
+#define CFG_1_OCTET_1_SHIFT 8
+#define CFG_1_OCTET_2_MASK 0x00FF0000
+#define CFG_1_OCTET_2_SHIFT 16
+#define CFG_1_OCTET_3_MASK 0xFF000000
+#define CFG_1_OCTET_3_SHIFT 24
+
+/* Gbps Eth MAC Configuration 2 register masks and shifts */
+#define CFG_2_OCTET_4_MASK 0x000000FF
+#define CFG_2_OCTET_4_SHIFT 0
+#define CFG_2_OCTET_5_MASK 0x0000FF00
+#define CFG_2_OCTET_5_SHIFT 8
+#define CFG_2_DISK_MC_MASK 0x00100000
+#define CFG_2_DISK_MC_SHIFT 20
+#define CFG_2_DISK_BC_MASK 0x00200000
+#define CFG_2_DISK_BC_SHIFT 21
+#define CFG_2_DISK_DA_MASK 0x00400000
+#define CFG_2_DISK_DA_SHIFT 22
+#define CFG_2_STAT_EN_MASK 0x3000000
+#define CFG_2_STAT_EN_SHIFT 24
+#define CFG_2_TRANSMIT_FLUSH_EN_MASK 0x80000000
+#define CFG_2_TRANSMIT_FLUSH_EN_SHIFT 31
+
+/* Gbps Eth MAC Configuration 3 register masks and shifts */
+#define CFG_3_TM_HD_MODE_MASK 0x1
+#define CFG_3_TM_HD_MODE_SHIFT 0
+#define CFG_3_RX_CBFC_EN_MASK 0x2
+#define CFG_3_RX_CBFC_EN_SHIFT 1
+#define CFG_3_RX_CBFC_REDIR_EN_MASK 0x4
+#define CFG_3_RX_CBFC_REDIR_EN_SHIFT 2
+#define CFG_3_REDIRECT_CBFC_SEL_MASK 0x18
+#define CFG_3_REDIRECT_CBFC_SEL_SHIFT 3
+#define CFG_3_CF_DROP_MASK 0x20
+#define CFG_3_CF_DROP_SHIFT 5
+#define CFG_3_CF_TIMEOUT_MASK 0x3C0
+#define CFG_3_CF_TIMEOUT_SHIFT 6
+#define CFG_3_RX_IFG_TH_MASK 0x7C00
+#define CFG_3_RX_IFG_TH_SHIFT 10
+#define CFG_3_TX_CBFC_EN_MASK 0x8000
+#define CFG_3_TX_CBFC_EN_SHIFT 15
+#define CFG_3_MAX_LEN_MASK 0x3FFF0000
+#define CFG_3_MAX_LEN_SHIFT 16
+#define CFG_3_EXT_OOB_CBFC_SEL_MASK 0xC0000000
+#define CFG_3_EXT_OOB_CBFC_SEL_SHIFT 30
+
+/* GE MAC, PCS reset control register masks and shifts */
+#define RST_SPCS_MASK 0x1
+#define RST_SPCS_SHIFT 0
+#define RST_GMAC_0_MASK 0x100
+#define RST_GMAC_0_SHIFT 8
+
+/* Tx phase sync FIFO control register masks and shifts */
+#define PHASE_FIFO_CTL_RST_MASK 0x1
+#define PHASE_FIFO_CTL_RST_SHIFT 0
+#define PHASE_FIFO_CTL_INIT_MASK 0x2
+#define PHASE_FIFO_CTL_INIT_SHIFT 1
 
 /**
  * struct nps_enet_priv - Storage of ENET's private information.
@@ -285,8 +175,8 @@ struct nps_enet_priv {
 	bool tx_packet_sent;
 	struct sk_buff *tx_skb;
 	struct napi_struct napi;
-	struct nps_enet_ge_mac_cfg_2 ge_mac_cfg_2;
-	struct nps_enet_ge_mac_cfg_3 ge_mac_cfg_3;
+	u32 ge_mac_cfg_2_value;
+	u32 ge_mac_cfg_3_value;
 };
 
 /**
-- 
cgit v0.10.2


From 799977d9aafbf0ca0b9c39b04cbfb16db71302c9 Mon Sep 17 00:00:00 2001
From: David Ahern <dsa@cumulusnetworks.com>
Date: Wed, 2 Mar 2016 11:30:07 -0800
Subject: net: ipv6: Fix refcnt on host routes

Andrew and Ying Huang's test robot both reported usage count problems that
trace back to the 'keep address on ifdown' patch.

>From Andrew:
We execute CRIU test on linux-next. On the current linux-next kernel
they hangs on creating a network namespace.

The kernel log contains many massages like this:
[ 1036.122108] unregister_netdevice: waiting for lo to become free.
Usage count = 2
[ 1046.165156] unregister_netdevice: waiting for lo to become free.
Usage count = 2
[ 1056.210287] unregister_netdevice: waiting for lo to become free.
Usage count = 2

I tried to revert this patch and the bug disappeared.

Here is a set of commands to reproduce this bug:

[root@linux-next-test linux-next]# uname -a
Linux linux-next-test 4.5.0-rc6-next-20160301+ #3 SMP Wed Mar 2
17:32:18 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux

[root@linux-next-test ~]# unshare -n
[root@linux-next-test ~]# ip link set up dev lo
[root@linux-next-test ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
group default qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
[root@linux-next-test ~]# logout
[root@linux-next-test ~]# unshare -n

 -----

The problem is a change made to RTM_DELADDR case in __ipv6_ifa_notify that
was added in an early version of the offending patch and is no longer
needed.

Fixes: f1705ec197e7 ("net: ipv6: Make address flushing on ifdown optional")
Cc: Andrey Wagin <avagin@gmail.com>
Cc: Ying Huang <ying.huang@linux.intel.com>
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Tested-by: Jeremiah Mahler <jmmahler@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index b1169d5..8c0dab2 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -5319,12 +5319,10 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 			if (rt)
 				ip6_del_rt(rt);
 		}
-		if (ifp->rt) {
-			dst_hold(&ifp->rt->dst);
+		dst_hold(&ifp->rt->dst);
+
+		ip6_del_rt(ifp->rt);
 
-			ip6_del_rt(ifp->rt);
-			ifp->rt = NULL;
-		}
 		rt_genid_bump_ipv6(net);
 		break;
 	}
-- 
cgit v0.10.2


From 1f27cde313d72d6b44a73ba89c8b2c6a99c628cf Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 2 Mar 2016 08:21:43 -0800
Subject: net: sched: use pfifo_fast for non real queues

Some devices declare a high number of TX queues, then set a much
lower real_num_tx_queues

This cause setups using fq_codel, sfq or fq as the default qdisc to consume
more memory than really needed.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index e5bba89..46e55f0 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -345,6 +345,12 @@ extern struct Qdisc_ops pfifo_fast_ops;
 extern struct Qdisc_ops mq_qdisc_ops;
 extern struct Qdisc_ops noqueue_qdisc_ops;
 extern const struct Qdisc_ops *default_qdisc_ops;
+static inline const struct Qdisc_ops *
+get_default_qdisc_ops(const struct net_device *dev, int ntx)
+{
+	return ntx < dev->real_num_tx_queues ?
+			default_qdisc_ops : &pfifo_fast_ops;
+}
 
 struct Qdisc_class_common {
 	u32			classid;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 16bc83b..f18c350 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -567,6 +567,7 @@ struct Qdisc_ops pfifo_fast_ops __read_mostly = {
 	.dump		=	pfifo_fast_dump,
 	.owner		=	THIS_MODULE,
 };
+EXPORT_SYMBOL(pfifo_fast_ops);
 
 static struct lock_class_key qdisc_tx_busylock;
 
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index 3e82f04..56a77b8 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -57,7 +57,7 @@ static int mq_init(struct Qdisc *sch, struct nlattr *opt)
 
 	for (ntx = 0; ntx < dev->num_tx_queues; ntx++) {
 		dev_queue = netdev_get_tx_queue(dev, ntx);
-		qdisc = qdisc_create_dflt(dev_queue, default_qdisc_ops,
+		qdisc = qdisc_create_dflt(dev_queue, get_default_qdisc_ops(dev, ntx),
 					  TC_H_MAKE(TC_H_MAJ(sch->handle),
 						    TC_H_MIN(ntx + 1)));
 		if (qdisc == NULL)
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 02ffb3f..b8002ce 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -125,7 +125,8 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt)
 
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		dev_queue = netdev_get_tx_queue(dev, i);
-		qdisc = qdisc_create_dflt(dev_queue, default_qdisc_ops,
+		qdisc = qdisc_create_dflt(dev_queue,
+					  get_default_qdisc_ops(dev, i),
 					  TC_H_MAKE(TC_H_MAJ(sch->handle),
 						    TC_H_MIN(i + 1)));
 		if (qdisc == NULL) {
-- 
cgit v0.10.2


From e721498a63147618283d2a8093664809f300778c Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:53:46 +0000
Subject: rxrpc: Fix a case where a call event bit is being used as a flag bit

Fix a case where RXRPC_CALL_RELEASE (an event) is being used to specify a
flag bit.  RXRPC_CALL_RELEASED should be used instead.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/ar-accept.c b/net/rxrpc/ar-accept.c
index 6d79310..65ea813 100644
--- a/net/rxrpc/ar-accept.c
+++ b/net/rxrpc/ar-accept.c
@@ -185,7 +185,7 @@ invalid_service:
 	read_unlock_bh(&local->services_lock);
 
 	read_lock_bh(&call->state_lock);
-	if (!test_bit(RXRPC_CALL_RELEASE, &call->flags) &&
+	if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
 	    !test_and_set_bit(RXRPC_CALL_RELEASE, &call->events)) {
 		rxrpc_get_call(call);
 		rxrpc_queue_call(call);
-- 
cgit v0.10.2


From 5b8848d14912d5e9974d5d83d71ed6cba2b49345 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:53:46 +0000
Subject: rxrpc: Convert call flag and event numbers into enums

Convert call flag and event numbers into enums and move their definitions
outside of the struct.

Also move the call state enum outside of the struct and add an extra
element to count the number of states.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/ar-ack.c b/net/rxrpc/ar-ack.c
index adc555e..7cb7e45 100644
--- a/net/rxrpc/ar-ack.c
+++ b/net/rxrpc/ar-ack.c
@@ -886,10 +886,11 @@ void rxrpc_process_call(struct work_struct *work)
 	struct rxrpc_header hdr;
 	struct msghdr msg;
 	struct kvec iov[5];
+	enum rxrpc_call_event genbit;
 	unsigned long bits;
 	__be32 data, pad;
 	size_t len;
-	int genbit, loop, nbit, ioc, ret, mtu;
+	int loop, nbit, ioc, ret, mtu;
 	u32 abort_code = RX_PROTOCOL_ERROR;
 	u8 *acks = NULL;
 
diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c
index a9e05db..d9a8eeb 100644
--- a/net/rxrpc/ar-call.c
+++ b/net/rxrpc/ar-call.c
@@ -28,7 +28,7 @@ unsigned rxrpc_max_call_lifetime = 60 * HZ;
  */
 unsigned rxrpc_dead_call_expiry = 2 * HZ;
 
-const char *const rxrpc_call_states[] = {
+const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = {
 	[RXRPC_CALL_CLIENT_SEND_REQUEST]	= "ClSndReq",
 	[RXRPC_CALL_CLIENT_AWAIT_REPLY]		= "ClAwtRpl",
 	[RXRPC_CALL_CLIENT_RECV_REPLY]		= "ClRcvRpl",
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 2934a73..3aea424 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -293,6 +293,67 @@ struct rxrpc_connection {
 };
 
 /*
+ * Flags in call->flags.
+ */
+enum rxrpc_call_flag {
+	RXRPC_CALL_RELEASED,		/* call has been released - no more message to userspace */
+	RXRPC_CALL_TERMINAL_MSG,	/* call has given the socket its final message */
+	RXRPC_CALL_RCVD_LAST,		/* all packets received */
+	RXRPC_CALL_RUN_RTIMER,		/* Tx resend timer started */
+	RXRPC_CALL_TX_SOFT_ACK,		/* sent some soft ACKs */
+	RXRPC_CALL_PROC_BUSY,		/* the processor is busy */
+	RXRPC_CALL_INIT_ACCEPT,		/* acceptance was initiated */
+	RXRPC_CALL_HAS_USERID,		/* has a user ID attached */
+	RXRPC_CALL_EXPECT_OOS,		/* expect out of sequence packets */
+};
+
+/*
+ * Events that can be raised on a call.
+ */
+enum rxrpc_call_event {
+	RXRPC_CALL_RCVD_ACKALL,		/* ACKALL or reply received */
+	RXRPC_CALL_RCVD_BUSY,		/* busy packet received */
+	RXRPC_CALL_RCVD_ABORT,		/* abort packet received */
+	RXRPC_CALL_RCVD_ERROR,		/* network error received */
+	RXRPC_CALL_ACK_FINAL,		/* need to generate final ACK (and release call) */
+	RXRPC_CALL_ACK,			/* need to generate ACK */
+	RXRPC_CALL_REJECT_BUSY,		/* need to generate busy message */
+	RXRPC_CALL_ABORT,		/* need to generate abort */
+	RXRPC_CALL_CONN_ABORT,		/* local connection abort generated */
+	RXRPC_CALL_RESEND_TIMER,	/* Tx resend timer expired */
+	RXRPC_CALL_RESEND,		/* Tx resend required */
+	RXRPC_CALL_DRAIN_RX_OOS,	/* drain the Rx out of sequence queue */
+	RXRPC_CALL_LIFE_TIMER,		/* call's lifetimer ran out */
+	RXRPC_CALL_ACCEPTED,		/* incoming call accepted by userspace app */
+	RXRPC_CALL_SECURED,		/* incoming call's connection is now secure */
+	RXRPC_CALL_POST_ACCEPT,		/* need to post an "accept?" message to the app */
+	RXRPC_CALL_RELEASE,		/* need to release the call's resources */
+};
+
+/*
+ * The states that a call can be in.
+ */
+enum rxrpc_call_state {
+	RXRPC_CALL_CLIENT_SEND_REQUEST,	/* - client sending request phase */
+	RXRPC_CALL_CLIENT_AWAIT_REPLY,	/* - client awaiting reply */
+	RXRPC_CALL_CLIENT_RECV_REPLY,	/* - client receiving reply phase */
+	RXRPC_CALL_CLIENT_FINAL_ACK,	/* - client sending final ACK phase */
+	RXRPC_CALL_SERVER_SECURING,	/* - server securing request connection */
+	RXRPC_CALL_SERVER_ACCEPTING,	/* - server accepting request */
+	RXRPC_CALL_SERVER_RECV_REQUEST,	/* - server receiving request */
+	RXRPC_CALL_SERVER_ACK_REQUEST,	/* - server pending ACK of request */
+	RXRPC_CALL_SERVER_SEND_REPLY,	/* - server sending reply */
+	RXRPC_CALL_SERVER_AWAIT_ACK,	/* - server awaiting final ACK */
+	RXRPC_CALL_COMPLETE,		/* - call completed */
+	RXRPC_CALL_SERVER_BUSY,		/* - call rejected by busy server */
+	RXRPC_CALL_REMOTELY_ABORTED,	/* - call aborted by peer */
+	RXRPC_CALL_LOCALLY_ABORTED,	/* - call aborted locally on error or close */
+	RXRPC_CALL_NETWORK_ERROR,	/* - call terminated by network error */
+	RXRPC_CALL_DEAD,		/* - call is dead */
+	NR__RXRPC_CALL_STATES
+};
+
+/*
  * RxRPC call definition
  * - matched by { connection, call_id }
  */
@@ -317,57 +378,13 @@ struct rxrpc_call {
 	unsigned long		user_call_ID;	/* user-defined call ID */
 	unsigned long		creation_jif;	/* time of call creation */
 	unsigned long		flags;
-#define RXRPC_CALL_RELEASED	0	/* call has been released - no more message to userspace */
-#define RXRPC_CALL_TERMINAL_MSG	1	/* call has given the socket its final message */
-#define RXRPC_CALL_RCVD_LAST	2	/* all packets received */
-#define RXRPC_CALL_RUN_RTIMER	3	/* Tx resend timer started */
-#define RXRPC_CALL_TX_SOFT_ACK	4	/* sent some soft ACKs */
-#define RXRPC_CALL_PROC_BUSY	5	/* the processor is busy */
-#define RXRPC_CALL_INIT_ACCEPT	6	/* acceptance was initiated */
-#define RXRPC_CALL_HAS_USERID	7	/* has a user ID attached */
-#define RXRPC_CALL_EXPECT_OOS	8	/* expect out of sequence packets */
 	unsigned long		events;
-#define RXRPC_CALL_RCVD_ACKALL	0	/* ACKALL or reply received */
-#define RXRPC_CALL_RCVD_BUSY	1	/* busy packet received */
-#define RXRPC_CALL_RCVD_ABORT	2	/* abort packet received */
-#define RXRPC_CALL_RCVD_ERROR	3	/* network error received */
-#define RXRPC_CALL_ACK_FINAL	4	/* need to generate final ACK (and release call) */
-#define RXRPC_CALL_ACK		5	/* need to generate ACK */
-#define RXRPC_CALL_REJECT_BUSY	6	/* need to generate busy message */
-#define RXRPC_CALL_ABORT	7	/* need to generate abort */
-#define RXRPC_CALL_CONN_ABORT	8	/* local connection abort generated */
-#define RXRPC_CALL_RESEND_TIMER	9	/* Tx resend timer expired */
-#define RXRPC_CALL_RESEND	10	/* Tx resend required */
-#define RXRPC_CALL_DRAIN_RX_OOS	11	/* drain the Rx out of sequence queue */
-#define RXRPC_CALL_LIFE_TIMER	12	/* call's lifetimer ran out */
-#define RXRPC_CALL_ACCEPTED	13	/* incoming call accepted by userspace app */
-#define RXRPC_CALL_SECURED	14	/* incoming call's connection is now secure */
-#define RXRPC_CALL_POST_ACCEPT	15	/* need to post an "accept?" message to the app */
-#define RXRPC_CALL_RELEASE	16	/* need to release the call's resources */
-
 	spinlock_t		lock;
 	rwlock_t		state_lock;	/* lock for state transition */
 	atomic_t		usage;
 	atomic_t		sequence;	/* Tx data packet sequence counter */
 	u32			abort_code;	/* local/remote abort code */
-	enum {					/* current state of call */
-		RXRPC_CALL_CLIENT_SEND_REQUEST,	/* - client sending request phase */
-		RXRPC_CALL_CLIENT_AWAIT_REPLY,	/* - client awaiting reply */
-		RXRPC_CALL_CLIENT_RECV_REPLY,	/* - client receiving reply phase */
-		RXRPC_CALL_CLIENT_FINAL_ACK,	/* - client sending final ACK phase */
-		RXRPC_CALL_SERVER_SECURING,	/* - server securing request connection */
-		RXRPC_CALL_SERVER_ACCEPTING,	/* - server accepting request */
-		RXRPC_CALL_SERVER_RECV_REQUEST,	/* - server receiving request */
-		RXRPC_CALL_SERVER_ACK_REQUEST,	/* - server pending ACK of request */
-		RXRPC_CALL_SERVER_SEND_REPLY,	/* - server sending reply */
-		RXRPC_CALL_SERVER_AWAIT_ACK,	/* - server awaiting final ACK */
-		RXRPC_CALL_COMPLETE,		/* - call completed */
-		RXRPC_CALL_SERVER_BUSY,		/* - call rejected by busy server */
-		RXRPC_CALL_REMOTELY_ABORTED,	/* - call aborted by peer */
-		RXRPC_CALL_LOCALLY_ABORTED,	/* - call aborted locally on error or close */
-		RXRPC_CALL_NETWORK_ERROR,	/* - call terminated by network error */
-		RXRPC_CALL_DEAD,		/* - call is dead */
-	} state;
+	enum rxrpc_call_state	state : 8;	/* current state of call */
 	int			debug_id;	/* debug ID for printks */
 	u8			channel;	/* connection channel occupied by this call */
 
-- 
cgit v0.10.2


From 4c198ad17a7253cc8ef3ff39bfe73d6b5e65ceef Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:53:46 +0000
Subject: rxrpc: Rename call events to begin RXRPC_CALL_EV_

Rename call event names to begin RXRPC_CALL_EV_ to distinguish them from the
flags.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/ar-accept.c b/net/rxrpc/ar-accept.c
index 65ea813..9a49f32 100644
--- a/net/rxrpc/ar-accept.c
+++ b/net/rxrpc/ar-accept.c
@@ -186,7 +186,7 @@ invalid_service:
 
 	read_lock_bh(&call->state_lock);
 	if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
-	    !test_and_set_bit(RXRPC_CALL_RELEASE, &call->events)) {
+	    !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) {
 		rxrpc_get_call(call);
 		rxrpc_queue_call(call);
 	}
@@ -379,7 +379,7 @@ struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx,
 	rb_insert_color(&call->sock_node, &rx->calls);
 	if (test_and_set_bit(RXRPC_CALL_HAS_USERID, &call->flags))
 		BUG();
-	if (test_and_set_bit(RXRPC_CALL_ACCEPTED, &call->events))
+	if (test_and_set_bit(RXRPC_CALL_EV_ACCEPTED, &call->events))
 		BUG();
 	rxrpc_queue_call(call);
 
@@ -395,7 +395,7 @@ struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx,
 out_release:
 	_debug("release %p", call);
 	if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
-	    !test_and_set_bit(RXRPC_CALL_RELEASE, &call->events))
+	    !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events))
 		rxrpc_queue_call(call);
 out_discard:
 	write_unlock_bh(&call->state_lock);
@@ -434,7 +434,7 @@ int rxrpc_reject_call(struct rxrpc_sock *rx)
 	switch (call->state) {
 	case RXRPC_CALL_SERVER_ACCEPTING:
 		call->state = RXRPC_CALL_SERVER_BUSY;
-		if (test_and_set_bit(RXRPC_CALL_REJECT_BUSY, &call->events))
+		if (test_and_set_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events))
 			rxrpc_queue_call(call);
 		ret = 0;
 		goto out_release;
@@ -458,7 +458,7 @@ int rxrpc_reject_call(struct rxrpc_sock *rx)
 out_release:
 	_debug("release %p", call);
 	if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
-	    !test_and_set_bit(RXRPC_CALL_RELEASE, &call->events))
+	    !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events))
 		rxrpc_queue_call(call);
 out_discard:
 	write_unlock_bh(&call->state_lock);
diff --git a/net/rxrpc/ar-ack.c b/net/rxrpc/ar-ack.c
index 7cb7e45..9183da7 100644
--- a/net/rxrpc/ar-ack.c
+++ b/net/rxrpc/ar-ack.c
@@ -161,7 +161,7 @@ cancel_timer:
 	try_to_del_timer_sync(&call->ack_timer);
 	read_lock_bh(&call->state_lock);
 	if (call->state <= RXRPC_CALL_COMPLETE &&
-	    !test_and_set_bit(RXRPC_CALL_ACK, &call->events))
+	    !test_and_set_bit(RXRPC_CALL_EV_ACK, &call->events))
 		rxrpc_queue_call(call);
 	read_unlock_bh(&call->state_lock);
 }
@@ -193,7 +193,7 @@ static void rxrpc_set_resend(struct rxrpc_call *call, u8 resend,
 
 	if (resend & 1) {
 		_debug("SET RESEND");
-		set_bit(RXRPC_CALL_RESEND, &call->events);
+		set_bit(RXRPC_CALL_EV_RESEND, &call->events);
 	}
 
 	if (resend & 2) {
@@ -203,7 +203,7 @@ static void rxrpc_set_resend(struct rxrpc_call *call, u8 resend,
 	} else {
 		_debug("KILL RESEND TIMER");
 		del_timer_sync(&call->resend_timer);
-		clear_bit(RXRPC_CALL_RESEND_TIMER, &call->events);
+		clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events);
 		clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
 	}
 	read_unlock_bh(&call->state_lock);
@@ -555,7 +555,7 @@ inserted:
 	if (call->state < RXRPC_CALL_COMPLETE &&
 	    call->rx_data_post == call->rx_first_oos) {
 		_debug("drain rx oos now");
-		set_bit(RXRPC_CALL_DRAIN_RX_OOS, &call->events);
+		set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events);
 	}
 	read_unlock(&call->state_lock);
 
@@ -793,7 +793,7 @@ all_acked:
 
 	del_timer_sync(&call->resend_timer);
 	clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
-	clear_bit(RXRPC_CALL_RESEND_TIMER, &call->events);
+	clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events);
 
 	if (call->acks_window)
 		rxrpc_zap_tx_window(call);
@@ -928,17 +928,17 @@ void rxrpc_process_call(struct work_struct *work)
 	iov[0].iov_len	= sizeof(hdr);
 
 	/* deal with events of a final nature */
-	if (test_bit(RXRPC_CALL_RELEASE, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) {
 		rxrpc_release_call(call);
-		clear_bit(RXRPC_CALL_RELEASE, &call->events);
+		clear_bit(RXRPC_CALL_EV_RELEASE, &call->events);
 	}
 
-	if (test_bit(RXRPC_CALL_RCVD_ERROR, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events)) {
 		int error;
 
-		clear_bit(RXRPC_CALL_CONN_ABORT, &call->events);
-		clear_bit(RXRPC_CALL_REJECT_BUSY, &call->events);
-		clear_bit(RXRPC_CALL_ABORT, &call->events);
+		clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events);
+		clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events);
+		clear_bit(RXRPC_CALL_EV_ABORT, &call->events);
 
 		error = call->conn->trans->peer->net_error;
 		_debug("post net error %d", error);
@@ -946,32 +946,32 @@ void rxrpc_process_call(struct work_struct *work)
 		if (rxrpc_post_message(call, RXRPC_SKB_MARK_NET_ERROR,
 				       error, true) < 0)
 			goto no_mem;
-		clear_bit(RXRPC_CALL_RCVD_ERROR, &call->events);
+		clear_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events);
 		goto kill_ACKs;
 	}
 
-	if (test_bit(RXRPC_CALL_CONN_ABORT, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events)) {
 		ASSERTCMP(call->state, >, RXRPC_CALL_COMPLETE);
 
-		clear_bit(RXRPC_CALL_REJECT_BUSY, &call->events);
-		clear_bit(RXRPC_CALL_ABORT, &call->events);
+		clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events);
+		clear_bit(RXRPC_CALL_EV_ABORT, &call->events);
 
 		_debug("post conn abort");
 
 		if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR,
 				       call->conn->error, true) < 0)
 			goto no_mem;
-		clear_bit(RXRPC_CALL_CONN_ABORT, &call->events);
+		clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events);
 		goto kill_ACKs;
 	}
 
-	if (test_bit(RXRPC_CALL_REJECT_BUSY, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events)) {
 		hdr.type = RXRPC_PACKET_TYPE_BUSY;
-		genbit = RXRPC_CALL_REJECT_BUSY;
+		genbit = RXRPC_CALL_EV_REJECT_BUSY;
 		goto send_message;
 	}
 
-	if (test_bit(RXRPC_CALL_ABORT, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_ABORT, &call->events)) {
 		ASSERTCMP(call->state, >, RXRPC_CALL_COMPLETE);
 
 		if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR,
@@ -981,12 +981,12 @@ void rxrpc_process_call(struct work_struct *work)
 		data = htonl(call->abort_code);
 		iov[1].iov_base = &data;
 		iov[1].iov_len = sizeof(data);
-		genbit = RXRPC_CALL_ABORT;
+		genbit = RXRPC_CALL_EV_ABORT;
 		goto send_message;
 	}
 
-	if (test_bit(RXRPC_CALL_ACK_FINAL, &call->events)) {
-		genbit = RXRPC_CALL_ACK_FINAL;
+	if (test_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events)) {
+		genbit = RXRPC_CALL_EV_ACK_FINAL;
 
 		ack.bufferSpace	= htons(8);
 		ack.maxSkew	= 0;
@@ -1012,12 +1012,12 @@ void rxrpc_process_call(struct work_struct *work)
 		goto send_ACK;
 	}
 
-	if (call->events & ((1 << RXRPC_CALL_RCVD_BUSY) |
-			    (1 << RXRPC_CALL_RCVD_ABORT))
+	if (call->events & ((1 << RXRPC_CALL_EV_RCVD_BUSY) |
+			    (1 << RXRPC_CALL_EV_RCVD_ABORT))
 	    ) {
 		u32 mark;
 
-		if (test_bit(RXRPC_CALL_RCVD_ABORT, &call->events))
+		if (test_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events))
 			mark = RXRPC_SKB_MARK_REMOTE_ABORT;
 		else
 			mark = RXRPC_SKB_MARK_BUSY;
@@ -1027,22 +1027,22 @@ void rxrpc_process_call(struct work_struct *work)
 		if (rxrpc_post_message(call, mark, ECONNABORTED, true) < 0)
 			goto no_mem;
 
-		clear_bit(RXRPC_CALL_RCVD_BUSY, &call->events);
-		clear_bit(RXRPC_CALL_RCVD_ABORT, &call->events);
+		clear_bit(RXRPC_CALL_EV_RCVD_BUSY, &call->events);
+		clear_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events);
 		goto kill_ACKs;
 	}
 
-	if (test_and_clear_bit(RXRPC_CALL_RCVD_ACKALL, &call->events)) {
+	if (test_and_clear_bit(RXRPC_CALL_EV_RCVD_ACKALL, &call->events)) {
 		_debug("do implicit ackall");
 		rxrpc_clear_tx_window(call);
 	}
 
-	if (test_bit(RXRPC_CALL_LIFE_TIMER, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events)) {
 		write_lock_bh(&call->state_lock);
 		if (call->state <= RXRPC_CALL_COMPLETE) {
 			call->state = RXRPC_CALL_LOCALLY_ABORTED;
 			call->abort_code = RX_CALL_TIMEOUT;
-			set_bit(RXRPC_CALL_ABORT, &call->events);
+			set_bit(RXRPC_CALL_EV_ABORT, &call->events);
 		}
 		write_unlock_bh(&call->state_lock);
 
@@ -1051,7 +1051,7 @@ void rxrpc_process_call(struct work_struct *work)
 				       ETIME, true) < 0)
 			goto no_mem;
 
-		clear_bit(RXRPC_CALL_LIFE_TIMER, &call->events);
+		clear_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events);
 		goto kill_ACKs;
 	}
 
@@ -1072,13 +1072,13 @@ void rxrpc_process_call(struct work_struct *work)
 	}
 
 	/* handle resending */
-	if (test_and_clear_bit(RXRPC_CALL_RESEND_TIMER, &call->events))
+	if (test_and_clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events))
 		rxrpc_resend_timer(call);
-	if (test_and_clear_bit(RXRPC_CALL_RESEND, &call->events))
+	if (test_and_clear_bit(RXRPC_CALL_EV_RESEND, &call->events))
 		rxrpc_resend(call);
 
 	/* consider sending an ordinary ACK */
-	if (test_bit(RXRPC_CALL_ACK, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_ACK, &call->events)) {
 		_debug("send ACK: window: %d - %d { %lx }",
 		       call->rx_data_eaten, call->ackr_win_top,
 		       call->ackr_window[0]);
@@ -1086,11 +1086,11 @@ void rxrpc_process_call(struct work_struct *work)
 		if (call->state > RXRPC_CALL_SERVER_ACK_REQUEST &&
 		    call->ackr_reason != RXRPC_ACK_PING_RESPONSE) {
 			/* ACK by sending reply DATA packet in this state */
-			clear_bit(RXRPC_CALL_ACK, &call->events);
+			clear_bit(RXRPC_CALL_EV_ACK, &call->events);
 			goto maybe_reschedule;
 		}
 
-		genbit = RXRPC_CALL_ACK;
+		genbit = RXRPC_CALL_EV_ACK;
 
 		acks = kzalloc(call->ackr_win_top - call->rx_data_eaten,
 			       GFP_NOFS);
@@ -1153,7 +1153,7 @@ void rxrpc_process_call(struct work_struct *work)
 
 	/* handle completion of security negotiations on an incoming
 	 * connection */
-	if (test_and_clear_bit(RXRPC_CALL_SECURED, &call->events)) {
+	if (test_and_clear_bit(RXRPC_CALL_EV_SECURED, &call->events)) {
 		_debug("secured");
 		spin_lock_bh(&call->lock);
 
@@ -1161,7 +1161,7 @@ void rxrpc_process_call(struct work_struct *work)
 			_debug("securing");
 			write_lock(&call->conn->lock);
 			if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
-			    !test_bit(RXRPC_CALL_RELEASE, &call->events)) {
+			    !test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) {
 				_debug("not released");
 				call->state = RXRPC_CALL_SERVER_ACCEPTING;
 				list_move_tail(&call->accept_link,
@@ -1170,39 +1170,39 @@ void rxrpc_process_call(struct work_struct *work)
 			write_unlock(&call->conn->lock);
 			read_lock(&call->state_lock);
 			if (call->state < RXRPC_CALL_COMPLETE)
-				set_bit(RXRPC_CALL_POST_ACCEPT, &call->events);
+				set_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events);
 			read_unlock(&call->state_lock);
 		}
 
 		spin_unlock_bh(&call->lock);
-		if (!test_bit(RXRPC_CALL_POST_ACCEPT, &call->events))
+		if (!test_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events))
 			goto maybe_reschedule;
 	}
 
 	/* post a notification of an acceptable connection to the app */
-	if (test_bit(RXRPC_CALL_POST_ACCEPT, &call->events)) {
+	if (test_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events)) {
 		_debug("post accept");
 		if (rxrpc_post_message(call, RXRPC_SKB_MARK_NEW_CALL,
 				       0, false) < 0)
 			goto no_mem;
-		clear_bit(RXRPC_CALL_POST_ACCEPT, &call->events);
+		clear_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events);
 		goto maybe_reschedule;
 	}
 
 	/* handle incoming call acceptance */
-	if (test_and_clear_bit(RXRPC_CALL_ACCEPTED, &call->events)) {
+	if (test_and_clear_bit(RXRPC_CALL_EV_ACCEPTED, &call->events)) {
 		_debug("accepted");
 		ASSERTCMP(call->rx_data_post, ==, 0);
 		call->rx_data_post = 1;
 		read_lock_bh(&call->state_lock);
 		if (call->state < RXRPC_CALL_COMPLETE)
-			set_bit(RXRPC_CALL_DRAIN_RX_OOS, &call->events);
+			set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events);
 		read_unlock_bh(&call->state_lock);
 	}
 
 	/* drain the out of sequence received packet queue into the packet Rx
 	 * queue */
-	if (test_and_clear_bit(RXRPC_CALL_DRAIN_RX_OOS, &call->events)) {
+	if (test_and_clear_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events)) {
 		while (call->rx_data_post == call->rx_first_oos)
 			if (rxrpc_drain_rx_oos_queue(call) < 0)
 				break;
@@ -1281,12 +1281,12 @@ send_message_2:
 	}
 
 	switch (genbit) {
-	case RXRPC_CALL_ABORT:
+	case RXRPC_CALL_EV_ABORT:
 		clear_bit(genbit, &call->events);
-		clear_bit(RXRPC_CALL_RCVD_ABORT, &call->events);
+		clear_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events);
 		goto kill_ACKs;
 
-	case RXRPC_CALL_ACK_FINAL:
+	case RXRPC_CALL_EV_ACK_FINAL:
 		write_lock_bh(&call->state_lock);
 		if (call->state == RXRPC_CALL_CLIENT_FINAL_ACK)
 			call->state = RXRPC_CALL_COMPLETE;
@@ -1311,9 +1311,9 @@ send_message_2:
 
 kill_ACKs:
 	del_timer_sync(&call->ack_timer);
-	if (test_and_clear_bit(RXRPC_CALL_ACK_FINAL, &call->events))
+	if (test_and_clear_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events))
 		rxrpc_put_call(call);
-	clear_bit(RXRPC_CALL_ACK, &call->events);
+	clear_bit(RXRPC_CALL_EV_ACK, &call->events);
 
 maybe_reschedule:
 	if (call->events || !skb_queue_empty(&call->rx_queue)) {
@@ -1332,7 +1332,7 @@ maybe_reschedule:
 
 		read_lock_bh(&call->state_lock);
 		if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
-		    !test_and_set_bit(RXRPC_CALL_RELEASE, &call->events))
+		    !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events))
 			rxrpc_queue_call(call);
 		read_unlock_bh(&call->state_lock);
 	}
diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c
index d9a8eeb..3468a07 100644
--- a/net/rxrpc/ar-call.c
+++ b/net/rxrpc/ar-call.c
@@ -452,7 +452,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
 		read_lock(&call->state_lock);
 		switch (call->state) {
 		case RXRPC_CALL_LOCALLY_ABORTED:
-			if (!test_and_set_bit(RXRPC_CALL_ABORT, &call->events))
+			if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events))
 				rxrpc_queue_call(call);
 		case RXRPC_CALL_REMOTELY_ABORTED:
 			read_unlock(&call->state_lock);
@@ -686,7 +686,7 @@ void rxrpc_release_call(struct rxrpc_call *call)
 		_debug("+++ ABORTING STATE %d +++\n", call->state);
 		call->state = RXRPC_CALL_LOCALLY_ABORTED;
 		call->abort_code = RX_CALL_DEAD;
-		set_bit(RXRPC_CALL_ABORT, &call->events);
+		set_bit(RXRPC_CALL_EV_ABORT, &call->events);
 		rxrpc_queue_call(call);
 	}
 	write_unlock(&call->state_lock);
@@ -763,10 +763,10 @@ static void rxrpc_mark_call_released(struct rxrpc_call *call)
 			_debug("abort call %p", call);
 			call->state = RXRPC_CALL_LOCALLY_ABORTED;
 			call->abort_code = RX_CALL_DEAD;
-			if (!test_and_set_bit(RXRPC_CALL_ABORT, &call->events))
+			if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events))
 				sched = true;
 		}
-		if (!test_and_set_bit(RXRPC_CALL_RELEASE, &call->events))
+		if (!test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events))
 			sched = true;
 		if (sched)
 			rxrpc_queue_call(call);
@@ -975,7 +975,7 @@ static void rxrpc_call_life_expired(unsigned long _call)
 	_enter("{%d}", call->debug_id);
 	read_lock_bh(&call->state_lock);
 	if (call->state < RXRPC_CALL_COMPLETE) {
-		set_bit(RXRPC_CALL_LIFE_TIMER, &call->events);
+		set_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events);
 		rxrpc_queue_call(call);
 	}
 	read_unlock_bh(&call->state_lock);
@@ -995,7 +995,7 @@ static void rxrpc_resend_time_expired(unsigned long _call)
 		return;
 
 	clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
-	if (!test_and_set_bit(RXRPC_CALL_RESEND_TIMER, &call->events))
+	if (!test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events))
 		rxrpc_queue_call(call);
 }
 
@@ -1013,7 +1013,7 @@ static void rxrpc_ack_time_expired(unsigned long _call)
 
 	read_lock_bh(&call->state_lock);
 	if (call->state < RXRPC_CALL_COMPLETE &&
-	    !test_and_set_bit(RXRPC_CALL_ACK, &call->events))
+	    !test_and_set_bit(RXRPC_CALL_EV_ACK, &call->events))
 		rxrpc_queue_call(call);
 	read_unlock_bh(&call->state_lock);
 }
diff --git a/net/rxrpc/ar-connevent.c b/net/rxrpc/ar-connevent.c
index e7ed43a..45e81b7 100644
--- a/net/rxrpc/ar-connevent.c
+++ b/net/rxrpc/ar-connevent.c
@@ -42,9 +42,9 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn, int state,
 			call->state = state;
 			call->abort_code = abort_code;
 			if (state == RXRPC_CALL_LOCALLY_ABORTED)
-				set_bit(RXRPC_CALL_CONN_ABORT, &call->events);
+				set_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events);
 			else
-				set_bit(RXRPC_CALL_RCVD_ABORT, &call->events);
+				set_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events);
 			rxrpc_queue_call(call);
 		}
 		write_unlock(&call->state_lock);
@@ -132,7 +132,7 @@ static void rxrpc_call_is_secure(struct rxrpc_call *call)
 	if (call) {
 		read_lock(&call->state_lock);
 		if (call->state < RXRPC_CALL_COMPLETE &&
-		    !test_and_set_bit(RXRPC_CALL_SECURED, &call->events))
+		    !test_and_set_bit(RXRPC_CALL_EV_SECURED, &call->events))
 			rxrpc_queue_call(call);
 		read_unlock(&call->state_lock);
 	}
diff --git a/net/rxrpc/ar-error.c b/net/rxrpc/ar-error.c
index 0610efa..3c47441 100644
--- a/net/rxrpc/ar-error.c
+++ b/net/rxrpc/ar-error.c
@@ -222,7 +222,7 @@ void rxrpc_UDP_error_handler(struct work_struct *work)
 			if (call->state != RXRPC_CALL_COMPLETE &&
 			    call->state < RXRPC_CALL_NETWORK_ERROR) {
 				call->state = RXRPC_CALL_NETWORK_ERROR;
-				set_bit(RXRPC_CALL_RCVD_ERROR, &call->events);
+				set_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events);
 				rxrpc_queue_call(call);
 			}
 			write_unlock(&call->state_lock);
diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c
index 4505a69..9185535 100644
--- a/net/rxrpc/ar-input.c
+++ b/net/rxrpc/ar-input.c
@@ -231,7 +231,7 @@ static int rxrpc_fast_process_data(struct rxrpc_call *call,
 		_debug("drain rx oos now");
 		read_lock(&call->state_lock);
 		if (call->state < RXRPC_CALL_COMPLETE &&
-		    !test_and_set_bit(RXRPC_CALL_DRAIN_RX_OOS, &call->events))
+		    !test_and_set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events))
 			rxrpc_queue_call(call);
 		read_unlock(&call->state_lock);
 	}
@@ -287,12 +287,12 @@ static void rxrpc_assume_implicit_ackall(struct rxrpc_call *call, u32 serial)
 		call->acks_latest = serial;
 
 		_debug("implicit ACKALL %%%u", call->acks_latest);
-		set_bit(RXRPC_CALL_RCVD_ACKALL, &call->events);
+		set_bit(RXRPC_CALL_EV_RCVD_ACKALL, &call->events);
 		write_unlock_bh(&call->state_lock);
 
 		if (try_to_del_timer_sync(&call->resend_timer) >= 0) {
-			clear_bit(RXRPC_CALL_RESEND_TIMER, &call->events);
-			clear_bit(RXRPC_CALL_RESEND, &call->events);
+			clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events);
+			clear_bit(RXRPC_CALL_EV_RESEND, &call->events);
 			clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
 		}
 		break;
@@ -358,7 +358,7 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 		if (call->state < RXRPC_CALL_COMPLETE) {
 			call->state = RXRPC_CALL_REMOTELY_ABORTED;
 			call->abort_code = abort_code;
-			set_bit(RXRPC_CALL_RCVD_ABORT, &call->events);
+			set_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events);
 			rxrpc_queue_call(call);
 		}
 		goto free_packet_unlock;
@@ -373,7 +373,7 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 		switch (call->state) {
 		case RXRPC_CALL_CLIENT_SEND_REQUEST:
 			call->state = RXRPC_CALL_SERVER_BUSY;
-			set_bit(RXRPC_CALL_RCVD_BUSY, &call->events);
+			set_bit(RXRPC_CALL_EV_RCVD_BUSY, &call->events);
 			rxrpc_queue_call(call);
 		case RXRPC_CALL_SERVER_BUSY:
 			goto free_packet_unlock;
@@ -433,7 +433,7 @@ protocol_error_locked:
 	if (call->state <= RXRPC_CALL_COMPLETE) {
 		call->state = RXRPC_CALL_LOCALLY_ABORTED;
 		call->abort_code = RX_PROTOCOL_ERROR;
-		set_bit(RXRPC_CALL_ABORT, &call->events);
+		set_bit(RXRPC_CALL_EV_ABORT, &call->events);
 		rxrpc_queue_call(call);
 	}
 free_packet_unlock:
@@ -505,7 +505,7 @@ protocol_error:
 	if (call->state <= RXRPC_CALL_COMPLETE) {
 		call->state = RXRPC_CALL_LOCALLY_ABORTED;
 		call->abort_code = RX_PROTOCOL_ERROR;
-		set_bit(RXRPC_CALL_ABORT, &call->events);
+		set_bit(RXRPC_CALL_EV_ABORT, &call->events);
 		rxrpc_queue_call(call);
 	}
 	write_unlock_bh(&call->state_lock);
@@ -530,7 +530,7 @@ static void rxrpc_post_packet_to_call(struct rxrpc_call *call,
 	read_lock(&call->state_lock);
 	switch (call->state) {
 	case RXRPC_CALL_LOCALLY_ABORTED:
-		if (!test_and_set_bit(RXRPC_CALL_ABORT, &call->events)) {
+		if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) {
 			rxrpc_queue_call(call);
 			goto free_unlock;
 		}
@@ -546,7 +546,7 @@ static void rxrpc_post_packet_to_call(struct rxrpc_call *call,
 		/* resend last packet of a completed call */
 		_debug("final ack again");
 		rxrpc_get_call(call);
-		set_bit(RXRPC_CALL_ACK_FINAL, &call->events);
+		set_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events);
 		rxrpc_queue_call(call);
 		goto free_unlock;
 	default:
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 3aea424..3f29406 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -311,23 +311,23 @@ enum rxrpc_call_flag {
  * Events that can be raised on a call.
  */
 enum rxrpc_call_event {
-	RXRPC_CALL_RCVD_ACKALL,		/* ACKALL or reply received */
-	RXRPC_CALL_RCVD_BUSY,		/* busy packet received */
-	RXRPC_CALL_RCVD_ABORT,		/* abort packet received */
-	RXRPC_CALL_RCVD_ERROR,		/* network error received */
-	RXRPC_CALL_ACK_FINAL,		/* need to generate final ACK (and release call) */
-	RXRPC_CALL_ACK,			/* need to generate ACK */
-	RXRPC_CALL_REJECT_BUSY,		/* need to generate busy message */
-	RXRPC_CALL_ABORT,		/* need to generate abort */
-	RXRPC_CALL_CONN_ABORT,		/* local connection abort generated */
-	RXRPC_CALL_RESEND_TIMER,	/* Tx resend timer expired */
-	RXRPC_CALL_RESEND,		/* Tx resend required */
-	RXRPC_CALL_DRAIN_RX_OOS,	/* drain the Rx out of sequence queue */
-	RXRPC_CALL_LIFE_TIMER,		/* call's lifetimer ran out */
-	RXRPC_CALL_ACCEPTED,		/* incoming call accepted by userspace app */
-	RXRPC_CALL_SECURED,		/* incoming call's connection is now secure */
-	RXRPC_CALL_POST_ACCEPT,		/* need to post an "accept?" message to the app */
-	RXRPC_CALL_RELEASE,		/* need to release the call's resources */
+	RXRPC_CALL_EV_RCVD_ACKALL,	/* ACKALL or reply received */
+	RXRPC_CALL_EV_RCVD_BUSY,	/* busy packet received */
+	RXRPC_CALL_EV_RCVD_ABORT,	/* abort packet received */
+	RXRPC_CALL_EV_RCVD_ERROR,	/* network error received */
+	RXRPC_CALL_EV_ACK_FINAL,	/* need to generate final ACK (and release call) */
+	RXRPC_CALL_EV_ACK,		/* need to generate ACK */
+	RXRPC_CALL_EV_REJECT_BUSY,	/* need to generate busy message */
+	RXRPC_CALL_EV_ABORT,		/* need to generate abort */
+	RXRPC_CALL_EV_CONN_ABORT,	/* local connection abort generated */
+	RXRPC_CALL_EV_RESEND_TIMER,	/* Tx resend timer expired */
+	RXRPC_CALL_EV_RESEND,		/* Tx resend required */
+	RXRPC_CALL_EV_DRAIN_RX_OOS,	/* drain the Rx out of sequence queue */
+	RXRPC_CALL_EV_LIFE_TIMER,	/* call's lifetimer ran out */
+	RXRPC_CALL_EV_ACCEPTED,		/* incoming call accepted by userspace app */
+	RXRPC_CALL_EV_SECURED,		/* incoming call's connection is now secure */
+	RXRPC_CALL_EV_POST_ACCEPT,	/* need to post an "accept?" message to the app */
+	RXRPC_CALL_EV_RELEASE,		/* need to release the call's resources */
 };
 
 /*
@@ -440,7 +440,7 @@ static inline void rxrpc_abort_call(struct rxrpc_call *call, u32 abort_code)
 	if (call->state < RXRPC_CALL_COMPLETE) {
 		call->abort_code = abort_code;
 		call->state = RXRPC_CALL_LOCALLY_ABORTED;
-		set_bit(RXRPC_CALL_ABORT, &call->events);
+		set_bit(RXRPC_CALL_EV_ABORT, &call->events);
 	}
 	write_unlock_bh(&call->state_lock);
 }
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 14c4e12..9e1527a 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -111,11 +111,11 @@ static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code)
 	if (call->state <= RXRPC_CALL_COMPLETE) {
 		call->state = RXRPC_CALL_LOCALLY_ABORTED;
 		call->abort_code = abort_code;
-		set_bit(RXRPC_CALL_ABORT, &call->events);
+		set_bit(RXRPC_CALL_EV_ABORT, &call->events);
 		del_timer_sync(&call->resend_timer);
 		del_timer_sync(&call->ack_timer);
-		clear_bit(RXRPC_CALL_RESEND_TIMER, &call->events);
-		clear_bit(RXRPC_CALL_ACK, &call->events);
+		clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events);
+		clear_bit(RXRPC_CALL_EV_ACK, &call->events);
 		clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
 		rxrpc_queue_call(call);
 	}
@@ -437,7 +437,7 @@ static inline void rxrpc_instant_resend(struct rxrpc_call *call)
 	if (try_to_del_timer_sync(&call->resend_timer) >= 0) {
 		clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags);
 		if (call->state < RXRPC_CALL_COMPLETE &&
-		    !test_and_set_bit(RXRPC_CALL_RESEND_TIMER, &call->events))
+		    !test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events))
 			rxrpc_queue_call(call);
 	}
 	read_unlock_bh(&call->state_lock);
diff --git a/net/rxrpc/ar-recvmsg.c b/net/rxrpc/ar-recvmsg.c
index b92bede..70f4703 100644
--- a/net/rxrpc/ar-recvmsg.c
+++ b/net/rxrpc/ar-recvmsg.c
@@ -33,7 +33,7 @@ void rxrpc_remove_user_ID(struct rxrpc_sock *rx, struct rxrpc_call *call)
 
 	read_lock_bh(&call->state_lock);
 	if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
-	    !test_and_set_bit(RXRPC_CALL_RELEASE, &call->events))
+	    !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events))
 		rxrpc_queue_call(call);
 	read_unlock_bh(&call->state_lock);
 }
diff --git a/net/rxrpc/ar-skbuff.c b/net/rxrpc/ar-skbuff.c
index 4cfab49..81f3c02 100644
--- a/net/rxrpc/ar-skbuff.c
+++ b/net/rxrpc/ar-skbuff.c
@@ -34,7 +34,7 @@ static void rxrpc_request_final_ACK(struct rxrpc_call *call)
 		/* get an extra ref on the call for the final-ACK generator to
 		 * release */
 		rxrpc_get_call(call);
-		set_bit(RXRPC_CALL_ACK_FINAL, &call->events);
+		set_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events);
 		if (try_to_del_timer_sync(&call->ack_timer) >= 0)
 			rxrpc_queue_call(call);
 		break;
-- 
cgit v0.10.2


From 0d12f8a4027d021c9cc942f09f38d28288020c5d Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:53:46 +0000
Subject: rxrpc: Keep the skb private record of the Rx header in host byte
 order

Currently, a copy of the Rx packet header is copied into the the sk_buff
private data so that we can advance the pointer into the buffer,
potentially discarding the original.  At the moment, this copy is held in
network byte order, but this means we're doing a lot of unnecessary
translations.

The reasons it was done this way are that we need the values in network
byte order occasionally and we can use the copy, slightly modified, as part
of an iov array when sending an ack or an abort packet.

However, it seems more reasonable on review that it would be better kept in
host byte order and that we make up a new header when we want to send
another packet.

To this end, rename the original header struct to rxrpc_wire_header (with
BE fields) and institute a variant called rxrpc_host_header that has host
order fields.  Change the struct in the sk_buff private data into an
rxrpc_host_header and translate the values when filling it in.

This further allows us to keep values kept in various structures in host
byte order rather than network byte order and allows removal of some fields
that are byteswapped duplicates.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/include/rxrpc/packet.h b/include/rxrpc/packet.h
index 4dce116..de1e679 100644
--- a/include/rxrpc/packet.h
+++ b/include/rxrpc/packet.h
@@ -22,7 +22,7 @@ typedef __be32	rxrpc_serial_net_t; /* on-the-wire Rx message serial number */
  * on-the-wire Rx packet header
  * - all multibyte fields should be in network byte order
  */
-struct rxrpc_header {
+struct rxrpc_wire_header {
 	__be32		epoch;		/* client boot timestamp */
 
 	__be32		cid;		/* connection and channel ID */
@@ -68,8 +68,6 @@ struct rxrpc_header {
 
 } __packed;
 
-#define __rxrpc_header_off(X) offsetof(struct rxrpc_header,X)
-
 extern const char *rxrpc_pkts[];
 
 /*****************************************************************************/
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 7e2d105..7bb5cca 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -37,7 +37,7 @@ static struct proto rxrpc_proto;
 static const struct proto_ops rxrpc_rpc_ops;
 
 /* local epoch for detecting local-end reset */
-__be32 rxrpc_epoch;
+u32 rxrpc_epoch;
 
 /* current debugging ID */
 atomic_t rxrpc_debug_id;
@@ -125,7 +125,6 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len)
 	struct sock *sk = sock->sk;
 	struct rxrpc_local *local;
 	struct rxrpc_sock *rx = rxrpc_sk(sk), *prx;
-	__be16 service_id;
 	int ret;
 
 	_enter("%p,%p,%d", rx, saddr, len);
@@ -152,14 +151,12 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len)
 
 	rx->local = local;
 	if (srx->srx_service) {
-		service_id = htons(srx->srx_service);
 		write_lock_bh(&local->services_lock);
 		list_for_each_entry(prx, &local->services, listen_link) {
-			if (prx->service_id == service_id)
+			if (prx->srx.srx_service == srx->srx_service)
 				goto service_in_use;
 		}
 
-		rx->service_id = service_id;
 		list_add_tail(&rx->listen_link, &local->services);
 		write_unlock_bh(&local->services_lock);
 
@@ -276,7 +273,6 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
 	struct rxrpc_transport *trans;
 	struct rxrpc_call *call;
 	struct rxrpc_sock *rx = rxrpc_sk(sock->sk);
-	__be16 service_id;
 
 	_enter(",,%x,%lx", key_serial(key), user_call_ID);
 
@@ -299,16 +295,15 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
 		atomic_inc(&trans->usage);
 	}
 
-	service_id = rx->service_id;
-	if (srx)
-		service_id = htons(srx->srx_service);
+	if (!srx)
+		srx = &rx->srx;
 
 	if (!key)
 		key = rx->key;
 	if (key && !key->payload.data[0])
 		key = NULL; /* a no-security key */
 
-	bundle = rxrpc_get_bundle(rx, trans, key, service_id, gfp);
+	bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, gfp);
 	if (IS_ERR(bundle)) {
 		call = ERR_CAST(bundle);
 		goto out;
@@ -425,7 +420,6 @@ static int rxrpc_connect(struct socket *sock, struct sockaddr *addr,
 	}
 
 	rx->trans = trans;
-	rx->service_id = htons(srx->srx_service);
 	rx->sk.sk_state = RXRPC_CLIENT_CONNECTED;
 
 	release_sock(&rx->sk);
@@ -778,7 +772,7 @@ static struct proto rxrpc_proto = {
 	.name		= "RXRPC",
 	.owner		= THIS_MODULE,
 	.obj_size	= sizeof(struct rxrpc_sock),
-	.max_header	= sizeof(struct rxrpc_header),
+	.max_header	= sizeof(struct rxrpc_wire_header),
 };
 
 static const struct net_proto_family rxrpc_family_ops = {
@@ -796,7 +790,7 @@ static int __init af_rxrpc_init(void)
 
 	BUILD_BUG_ON(sizeof(struct rxrpc_skb_priv) > FIELD_SIZEOF(struct sk_buff, cb));
 
-	rxrpc_epoch = htonl(get_seconds());
+	rxrpc_epoch = get_seconds();
 
 	ret = -ENOMEM;
 	rxrpc_call_jar = kmem_cache_create(
diff --git a/net/rxrpc/ar-accept.c b/net/rxrpc/ar-accept.c
index 9a49f32..73c9054 100644
--- a/net/rxrpc/ar-accept.c
+++ b/net/rxrpc/ar-accept.c
@@ -27,7 +27,7 @@
  * generate a connection-level abort
  */
 static int rxrpc_busy(struct rxrpc_local *local, struct sockaddr_rxrpc *srx,
-		      struct rxrpc_header *hdr)
+		      struct rxrpc_wire_header *whdr)
 {
 	struct msghdr msg;
 	struct kvec iov[1];
@@ -36,25 +36,21 @@ static int rxrpc_busy(struct rxrpc_local *local, struct sockaddr_rxrpc *srx,
 
 	_enter("%d,,", local->debug_id);
 
+	whdr->type	= RXRPC_PACKET_TYPE_BUSY;
+	whdr->serial	= htonl(1);
+
 	msg.msg_name	= &srx->transport.sin;
 	msg.msg_namelen	= sizeof(srx->transport.sin);
 	msg.msg_control	= NULL;
 	msg.msg_controllen = 0;
 	msg.msg_flags	= 0;
 
-	hdr->seq	= 0;
-	hdr->type	= RXRPC_PACKET_TYPE_BUSY;
-	hdr->flags	= 0;
-	hdr->userStatus	= 0;
-	hdr->_rsvd	= 0;
-
-	iov[0].iov_base	= hdr;
-	iov[0].iov_len	= sizeof(*hdr);
+	iov[0].iov_base	= whdr;
+	iov[0].iov_len	= sizeof(*whdr);
 
 	len = iov[0].iov_len;
 
-	hdr->serial = htonl(1);
-	_proto("Tx BUSY %%%u", ntohl(hdr->serial));
+	_proto("Tx BUSY %%1");
 
 	ret = kernel_sendmsg(local->socket, &msg, iov, 1, len);
 	if (ret < 0) {
@@ -211,8 +207,8 @@ void rxrpc_accept_incoming_calls(struct work_struct *work)
 	struct rxrpc_skb_priv *sp;
 	struct sockaddr_rxrpc srx;
 	struct rxrpc_sock *rx;
+	struct rxrpc_wire_header whdr;
 	struct sk_buff *skb;
-	__be16 service_id;
 	int ret;
 
 	_enter("%d", local->debug_id);
@@ -240,6 +236,19 @@ process_next_packet:
 
 	sp = rxrpc_skb(skb);
 
+	/* Set up a response packet header in case we need it */
+	whdr.epoch	= htonl(sp->hdr.epoch);
+	whdr.cid	= htonl(sp->hdr.cid);
+	whdr.callNumber	= htonl(sp->hdr.callNumber);
+	whdr.seq	= htonl(sp->hdr.seq);
+	whdr.serial	= 0;
+	whdr.flags	= 0;
+	whdr.type	= 0;
+	whdr.userStatus	= 0;
+	whdr.securityIndex = sp->hdr.securityIndex;
+	whdr._rsvd	= 0;
+	whdr.serviceId	= htons(sp->hdr.serviceId);
+
 	/* determine the remote address */
 	memset(&srx, 0, sizeof(srx));
 	srx.srx_family = AF_RXRPC;
@@ -256,10 +265,9 @@ process_next_packet:
 	}
 
 	/* get the socket providing the service */
-	service_id = sp->hdr.serviceId;
 	read_lock_bh(&local->services_lock);
 	list_for_each_entry(rx, &local->services, listen_link) {
-		if (rx->service_id == service_id &&
+		if (rx->srx.srx_service == sp->hdr.serviceId &&
 		    rx->sk.sk_state != RXRPC_CLOSE)
 			goto found_service;
 	}
@@ -267,7 +275,7 @@ process_next_packet:
 	goto invalid_service;
 
 found_service:
-	_debug("found service %hd", ntohs(rx->service_id));
+	_debug("found service %hd", rx->srx.srx_service);
 	if (sk_acceptq_is_full(&rx->sk))
 		goto backlog_full;
 	sk_acceptq_added(&rx->sk);
@@ -296,7 +304,7 @@ found_service:
 backlog_full:
 	read_unlock_bh(&local->services_lock);
 busy:
-	rxrpc_busy(local, &srx, &sp->hdr);
+	rxrpc_busy(local, &srx, &whdr);
 	rxrpc_free_skb(skb);
 	goto process_next_packet;
 
diff --git a/net/rxrpc/ar-ack.c b/net/rxrpc/ar-ack.c
index 9183da7..20f3f00 100644
--- a/net/rxrpc/ar-ack.c
+++ b/net/rxrpc/ar-ack.c
@@ -91,7 +91,7 @@ static const s8 rxrpc_ack_priority[] = {
  * propose an ACK be sent
  */
 void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason,
-			 __be32 serial, bool immediate)
+			 u32 serial, bool immediate)
 {
 	unsigned long expiry;
 	s8 prior = rxrpc_ack_priority[ack_reason];
@@ -99,8 +99,7 @@ void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason,
 	ASSERTCMP(prior, >, 0);
 
 	_enter("{%d},%s,%%%x,%u",
-	       call->debug_id, rxrpc_acks(ack_reason), ntohl(serial),
-	       immediate);
+	       call->debug_id, rxrpc_acks(ack_reason), serial, immediate);
 
 	if (prior < rxrpc_ack_priority[call->ackr_reason]) {
 		if (immediate)
@@ -139,7 +138,7 @@ void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason,
 		expiry = rxrpc_requested_ack_delay;
 		if (!expiry)
 			goto cancel_timer;
-		if (!immediate || serial == cpu_to_be32(1)) {
+		if (!immediate || serial == 1) {
 			_debug("run defer timer");
 			goto run_timer;
 		}
@@ -157,7 +156,7 @@ run_timer:
 	return;
 
 cancel_timer:
-	_debug("cancel timer %%%u", ntohl(serial));
+	_debug("cancel timer %%%u", serial);
 	try_to_del_timer_sync(&call->ack_timer);
 	read_lock_bh(&call->state_lock);
 	if (call->state <= RXRPC_CALL_COMPLETE &&
@@ -170,7 +169,7 @@ cancel_timer:
  * propose an ACK be sent, locking the call structure
  */
 void rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason,
-		       __be32 serial, bool immediate)
+		       u32 serial, bool immediate)
 {
 	s8 prior = rxrpc_ack_priority[ack_reason];
 
@@ -214,8 +213,8 @@ static void rxrpc_set_resend(struct rxrpc_call *call, u8 resend,
  */
 static void rxrpc_resend(struct rxrpc_call *call)
 {
+	struct rxrpc_wire_header *whdr;
 	struct rxrpc_skb_priv *sp;
-	struct rxrpc_header *hdr;
 	struct sk_buff *txb;
 	unsigned long *p_txb, resend_at;
 	bool stop;
@@ -247,14 +246,13 @@ static void rxrpc_resend(struct rxrpc_call *call)
 			sp->need_resend = false;
 
 			/* each Tx packet has a new serial number */
-			sp->hdr.serial =
-				htonl(atomic_inc_return(&call->conn->serial));
+			sp->hdr.serial = atomic_inc_return(&call->conn->serial);
 
-			hdr = (struct rxrpc_header *) txb->head;
-			hdr->serial = sp->hdr.serial;
+			whdr = (struct rxrpc_wire_header *)txb->head;
+			whdr->serial = htonl(sp->hdr.serial);
 
 			_proto("Tx DATA %%%u { #%d }",
-			       ntohl(sp->hdr.serial), ntohl(sp->hdr.seq));
+			       sp->hdr.serial, sp->hdr.seq);
 			if (rxrpc_send_packet(call->conn->trans, txb) < 0) {
 				stop = true;
 				sp->resend_at = jiffies + 3;
@@ -428,7 +426,7 @@ static void rxrpc_rotate_tx_window(struct rxrpc_call *call, u32 hard)
 	int tail = call->acks_tail, old_tail;
 	int win = CIRC_CNT(call->acks_head, tail, call->acks_winsz);
 
-	_enter("{%u,%u},%u", call->acks_hard, win, hard);
+	kenter("{%u,%u},%u", call->acks_hard, win, hard);
 
 	ASSERTCMP(hard - call->acks_hard, <=, win);
 
@@ -478,11 +476,11 @@ static int rxrpc_drain_rx_oos_queue(struct rxrpc_call *call)
 		sp = rxrpc_skb(skb);
 
 		_debug("drain OOS packet %d [%d]",
-		       ntohl(sp->hdr.seq), call->rx_first_oos);
+		       sp->hdr.seq, call->rx_first_oos);
 
-		if (ntohl(sp->hdr.seq) != call->rx_first_oos) {
+		if (sp->hdr.seq != call->rx_first_oos) {
 			skb_queue_head(&call->rx_oos_queue, skb);
-			call->rx_first_oos = ntohl(rxrpc_skb(skb)->hdr.seq);
+			call->rx_first_oos = rxrpc_skb(skb)->hdr.seq;
 			_debug("requeue %p {%u}", skb, call->rx_first_oos);
 		} else {
 			skb->mark = RXRPC_SKB_MARK_DATA;
@@ -496,8 +494,7 @@ static int rxrpc_drain_rx_oos_queue(struct rxrpc_call *call)
 			/* find out what the next packet is */
 			skb = skb_peek(&call->rx_oos_queue);
 			if (skb)
-				call->rx_first_oos =
-					ntohl(rxrpc_skb(skb)->hdr.seq);
+				call->rx_first_oos = rxrpc_skb(skb)->hdr.seq;
 			else
 				call->rx_first_oos = 0;
 			_debug("peek %p {%u}", skb, call->rx_first_oos);
@@ -522,7 +519,7 @@ static void rxrpc_insert_oos_packet(struct rxrpc_call *call,
 	u32 seq;
 
 	sp = rxrpc_skb(skb);
-	seq = ntohl(sp->hdr.seq);
+	seq = sp->hdr.seq;
 	_enter(",,{%u}", seq);
 
 	skb->destructor = rxrpc_packet_destructor;
@@ -535,9 +532,8 @@ static void rxrpc_insert_oos_packet(struct rxrpc_call *call,
 
 	skb_queue_walk(&call->rx_oos_queue, p) {
 		psp = rxrpc_skb(p);
-		if (ntohl(psp->hdr.seq) > seq) {
-			_debug("insert oos #%u before #%u",
-			       seq, ntohl(psp->hdr.seq));
+		if (psp->hdr.seq > seq) {
+			_debug("insert oos #%u before #%u", seq, psp->hdr.seq);
 			skb_insert(p, skb, &call->rx_oos_queue);
 			goto inserted;
 		}
@@ -586,7 +582,7 @@ static void rxrpc_zap_tx_window(struct rxrpc_call *call)
 
 		skb = (struct sk_buff *) _skb;
 		sp = rxrpc_skb(skb);
-		_debug("+++ clear Tx %u", ntohl(sp->hdr.seq));
+		_debug("+++ clear Tx %u", sp->hdr.seq);
 		rxrpc_free_skb(skb);
 	}
 
@@ -657,8 +653,7 @@ process_further:
 		/* data packets that wind up here have been received out of
 		 * order, need security processing or are jumbo packets */
 	case RXRPC_PACKET_TYPE_DATA:
-		_proto("OOSQ DATA %%%u { #%u }",
-		       ntohl(sp->hdr.serial), ntohl(sp->hdr.seq));
+		_proto("OOSQ DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq);
 
 		/* secured packets must be verified and possibly decrypted */
 		if (rxrpc_verify_packet(call, skb, _abort_code) < 0)
@@ -676,7 +671,7 @@ process_further:
 		if (!skb_pull(skb, sizeof(ack)))
 			BUG();
 
-		latest = ntohl(sp->hdr.serial);
+		latest = sp->hdr.serial;
 		hard = ntohl(ack.firstPacket);
 		tx = atomic_read(&call->sequence);
 
@@ -881,9 +876,9 @@ void rxrpc_process_call(struct work_struct *work)
 {
 	struct rxrpc_call *call =
 		container_of(work, struct rxrpc_call, processor);
+	struct rxrpc_wire_header whdr;
 	struct rxrpc_ackpacket ack;
 	struct rxrpc_ackinfo ackinfo;
-	struct rxrpc_header hdr;
 	struct msghdr msg;
 	struct kvec iov[5];
 	enum rxrpc_call_event genbit;
@@ -891,7 +886,7 @@ void rxrpc_process_call(struct work_struct *work)
 	__be32 data, pad;
 	size_t len;
 	int loop, nbit, ioc, ret, mtu;
-	u32 abort_code = RX_PROTOCOL_ERROR;
+	u32 serial, abort_code = RX_PROTOCOL_ERROR;
 	u8 *acks = NULL;
 
 	//printk("\n--------------------\n");
@@ -912,20 +907,20 @@ void rxrpc_process_call(struct work_struct *work)
 	msg.msg_controllen = 0;
 	msg.msg_flags	= 0;
 
-	hdr.epoch	= call->conn->epoch;
-	hdr.cid		= call->cid;
-	hdr.callNumber	= call->call_id;
-	hdr.seq		= 0;
-	hdr.type	= RXRPC_PACKET_TYPE_ACK;
-	hdr.flags	= call->conn->out_clientflag;
-	hdr.userStatus	= 0;
-	hdr.securityIndex = call->conn->security_ix;
-	hdr._rsvd	= 0;
-	hdr.serviceId	= call->conn->service_id;
+	whdr.epoch	= htonl(call->conn->epoch);
+	whdr.cid	= htonl(call->cid);
+	whdr.callNumber	= htonl(call->call_id);
+	whdr.seq	= 0;
+	whdr.type	= RXRPC_PACKET_TYPE_ACK;
+	whdr.flags	= call->conn->out_clientflag;
+	whdr.userStatus	= 0;
+	whdr.securityIndex = call->conn->security_ix;
+	whdr._rsvd	= 0;
+	whdr.serviceId	= htons(call->service_id);
 
 	memset(iov, 0, sizeof(iov));
-	iov[0].iov_base	= &hdr;
-	iov[0].iov_len	= sizeof(hdr);
+	iov[0].iov_base	= &whdr;
+	iov[0].iov_len	= sizeof(whdr);
 
 	/* deal with events of a final nature */
 	if (test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) {
@@ -966,7 +961,7 @@ void rxrpc_process_call(struct work_struct *work)
 	}
 
 	if (test_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events)) {
-		hdr.type = RXRPC_PACKET_TYPE_BUSY;
+		whdr.type = RXRPC_PACKET_TYPE_BUSY;
 		genbit = RXRPC_CALL_EV_REJECT_BUSY;
 		goto send_message;
 	}
@@ -977,7 +972,7 @@ void rxrpc_process_call(struct work_struct *work)
 		if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR,
 				       ECONNABORTED, true) < 0)
 			goto no_mem;
-		hdr.type = RXRPC_PACKET_TYPE_ABORT;
+		whdr.type = RXRPC_PACKET_TYPE_ABORT;
 		data = htonl(call->abort_code);
 		iov[1].iov_base = &data;
 		iov[1].iov_len = sizeof(data);
@@ -996,9 +991,9 @@ void rxrpc_process_call(struct work_struct *work)
 		call->ackr_reason = 0;
 
 		spin_lock_bh(&call->lock);
-		ack.serial = call->ackr_serial;
-		ack.previousPacket = call->ackr_prev_seq;
-		ack.firstPacket = htonl(call->rx_data_eaten + 1);
+		ack.serial	= htonl(call->ackr_serial);
+		ack.previousPacket = htonl(call->ackr_prev_seq);
+		ack.firstPacket	= htonl(call->rx_data_eaten + 1);
 		spin_unlock_bh(&call->lock);
 
 		pad = 0;
@@ -1100,13 +1095,11 @@ void rxrpc_process_call(struct work_struct *work)
 		//hdr.flags	= RXRPC_SLOW_START_OK;
 		ack.bufferSpace	= htons(8);
 		ack.maxSkew	= 0;
-		ack.serial	= 0;
-		ack.reason	= 0;
 
 		spin_lock_bh(&call->lock);
-		ack.reason = call->ackr_reason;
-		ack.serial = call->ackr_serial;
-		ack.previousPacket = call->ackr_prev_seq;
+		ack.reason	= call->ackr_reason;
+		ack.serial	= htonl(call->ackr_serial);
+		ack.previousPacket = htonl(call->ackr_prev_seq);
 		ack.firstPacket = htonl(call->rx_data_eaten + 1);
 
 		ack.nAcks = 0;
@@ -1225,9 +1218,10 @@ send_ACK:
 	ackinfo.rxMTU	= htonl(rxrpc_rx_mtu);
 	ackinfo.jumbo_max = htonl(rxrpc_rx_jumbo_max);
 
-	hdr.serial = htonl(atomic_inc_return(&call->conn->serial));
+	serial = atomic_inc_return(&call->conn->serial);
+	whdr.serial = htonl(serial);
 	_proto("Tx ACK %%%u { m=%hu f=#%u p=#%u s=%%%u r=%s n=%u }",
-	       ntohl(hdr.serial),
+	       serial,
 	       ntohs(ack.maxSkew),
 	       ntohl(ack.firstPacket),
 	       ntohl(ack.previousPacket),
@@ -1243,8 +1237,9 @@ send_ACK:
 send_message:
 	_debug("send message");
 
-	hdr.serial = htonl(atomic_inc_return(&call->conn->serial));
-	_proto("Tx %s %%%u", rxrpc_pkts[hdr.type], ntohl(hdr.serial));
+	serial = atomic_inc_return(&call->conn->serial);
+	whdr.serial = htonl(serial);
+	_proto("Tx %s %%%u", rxrpc_pkts[whdr.type], serial);
 send_message_2:
 
 	len = iov[0].iov_len;
@@ -1327,8 +1322,7 @@ maybe_reschedule:
 	if (call->state >= RXRPC_CALL_COMPLETE &&
 	    !list_empty(&call->accept_link)) {
 		_debug("X unlinking once-pending call %p { e=%lx f=%lx c=%x }",
-		       call, call->events, call->flags,
-		       ntohl(call->conn->cid));
+		       call, call->events, call->flags, call->conn->cid);
 
 		read_lock_bh(&call->state_lock);
 		if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
@@ -1346,7 +1340,7 @@ error:
 	 * this means there's a race between clearing the flag and setting the
 	 * work pending bit and the work item being processed again */
 	if (call->events && !work_pending(&call->processor)) {
-		_debug("jumpstart %x", ntohl(call->conn->cid));
+		_debug("jumpstart %x", call->conn->cid);
 		rxrpc_queue_call(call);
 	}
 
diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c
index 3468a07..4a499e0 100644
--- a/net/rxrpc/ar-call.c
+++ b/net/rxrpc/ar-call.c
@@ -64,11 +64,11 @@ static DEFINE_HASHTABLE(rxrpc_call_hash, 10);
  * Hash function for rxrpc_call_hash
  */
 static unsigned long rxrpc_call_hashfunc(
-	u8		clientflag,
-	__be32		cid,
-	__be32		call_id,
-	__be32		epoch,
-	__be16		service_id,
+	u8		in_clientflag,
+	u32		cid,
+	u32		call_id,
+	u32		epoch,
+	u16		service_id,
 	sa_family_t	proto,
 	void		*localptr,
 	unsigned int	addr_size,
@@ -77,7 +77,6 @@ static unsigned long rxrpc_call_hashfunc(
 	const u16 *p;
 	unsigned int i;
 	unsigned long key;
-	u32 hcid = ntohl(cid);
 
 	_enter("");
 
@@ -85,12 +84,12 @@ static unsigned long rxrpc_call_hashfunc(
 	/* We just want to add up the __be32 values, so forcing the
 	 * cast should be okay.
 	 */
-	key += (__force u32)epoch;
-	key += (__force u16)service_id;
-	key += (__force u32)call_id;
-	key += (hcid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT;
-	key += hcid & RXRPC_CHANNELMASK;
-	key += clientflag;
+	key += epoch;
+	key += service_id;
+	key += call_id;
+	key += (cid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT;
+	key += cid & RXRPC_CHANNELMASK;
+	key += in_clientflag;
 	key += proto;
 	/* Step through the peer address in 16-bit portions for speed */
 	for (i = 0, p = (const u16 *)peer_addr; i < addr_size >> 1; i++, p++)
@@ -148,19 +147,16 @@ static void rxrpc_call_hash_del(struct rxrpc_call *call)
  * isn't there.
  */
 struct rxrpc_call *rxrpc_find_call_hash(
-	u8		clientflag,
-	__be32		cid,
-	__be32		call_id,
-	__be32		epoch,
-	__be16		service_id,
+	struct rxrpc_host_header *hdr,
 	void		*localptr,
 	sa_family_t	proto,
-	const u8	*peer_addr)
+	const void	*peer_addr)
 {
 	unsigned long key;
 	unsigned int addr_size = 0;
 	struct rxrpc_call *call = NULL;
 	struct rxrpc_call *ret = NULL;
+	u8 in_clientflag = hdr->flags & RXRPC_CLIENT_INITIATED;
 
 	_enter("");
 	switch (proto) {
@@ -174,20 +170,21 @@ struct rxrpc_call *rxrpc_find_call_hash(
 		break;
 	}
 
-	key = rxrpc_call_hashfunc(clientflag, cid, call_id, epoch,
-				  service_id, proto, localptr, addr_size,
+	key = rxrpc_call_hashfunc(in_clientflag, hdr->cid, hdr->callNumber,
+				  hdr->epoch, hdr->serviceId,
+				  proto, localptr, addr_size,
 				  peer_addr);
 	hash_for_each_possible_rcu(rxrpc_call_hash, call, hash_node, key) {
 		if (call->hash_key == key &&
-		    call->call_id == call_id &&
-		    call->cid == cid &&
-		    call->in_clientflag == clientflag &&
-		    call->service_id == service_id &&
+		    call->call_id == hdr->callNumber &&
+		    call->cid == hdr->cid &&
+		    call->in_clientflag == in_clientflag &&
+		    call->service_id == hdr->serviceId &&
 		    call->proto == proto &&
 		    call->local == localptr &&
 		    memcmp(call->peer_ip.ipv6_addr, peer_addr,
-			      addr_size) == 0 &&
-		    call->epoch == epoch) {
+			   addr_size) == 0 &&
+		    call->epoch == hdr->epoch) {
 			ret = call;
 			break;
 		}
@@ -414,12 +411,12 @@ found_extant_second:
  */
 struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
 				       struct rxrpc_connection *conn,
-				       struct rxrpc_header *hdr,
+				       struct rxrpc_host_header *hdr,
 				       gfp_t gfp)
 {
 	struct rxrpc_call *call, *candidate;
 	struct rb_node **p, *parent;
-	__be32 call_id;
+	u32 call_id;
 
 	_enter(",%d,,%x", conn->debug_id, gfp);
 
@@ -433,7 +430,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
 	candidate->conn = conn;
 	candidate->cid = hdr->cid;
 	candidate->call_id = hdr->callNumber;
-	candidate->channel = ntohl(hdr->cid) & RXRPC_CHANNELMASK;
+	candidate->channel = hdr->cid & RXRPC_CHANNELMASK;
 	candidate->rx_data_post = 0;
 	candidate->state = RXRPC_CALL_SERVER_ACCEPTING;
 	if (conn->security_ix > 0)
@@ -492,9 +489,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
 		/* The tree is sorted in order of the __be32 value without
 		 * turning it into host order.
 		 */
-		if ((__force u32)call_id < (__force u32)call->call_id)
+		if (call_id < call->call_id)
 			p = &(*p)->rb_left;
-		else if ((__force u32)call_id > (__force u32)call->call_id)
+		else if (call_id > call->call_id)
 			p = &(*p)->rb_right;
 		else
 			goto old_call;
@@ -714,8 +711,7 @@ void rxrpc_release_call(struct rxrpc_call *call)
 
 			_debug("- zap %s %%%u #%u",
 			       rxrpc_pkts[sp->hdr.type],
-			       ntohl(sp->hdr.serial),
-			       ntohl(sp->hdr.seq));
+			       sp->hdr.serial, sp->hdr.seq);
 			rxrpc_free_skb(skb);
 			spin_lock_bh(&call->lock);
 		}
@@ -873,9 +869,9 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call)
 			unsigned long _skb;
 
 			_skb = call->acks_window[call->acks_tail] & ~1;
-			sp = rxrpc_skb((struct sk_buff *) _skb);
-			_debug("+++ clear Tx %u", ntohl(sp->hdr.seq));
-			rxrpc_free_skb((struct sk_buff *) _skb);
+			sp = rxrpc_skb((struct sk_buff *)_skb);
+			_debug("+++ clear Tx %u", sp->hdr.seq);
+			rxrpc_free_skb((struct sk_buff *)_skb);
 			call->acks_tail =
 				(call->acks_tail + 1) & (call->acks_winsz - 1);
 		}
diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c
index 6c71ed1..53df14c 100644
--- a/net/rxrpc/ar-connection.c
+++ b/net/rxrpc/ar-connection.c
@@ -57,10 +57,10 @@ static struct rxrpc_conn_bundle *rxrpc_alloc_bundle(gfp_t gfp)
  */
 static inline
 int rxrpc_cmp_bundle(const struct rxrpc_conn_bundle *bundle,
-		     struct key *key, __be16 service_id)
+		     struct key *key, u16 service_id)
 {
 	return (bundle->service_id - service_id) ?:
-		((unsigned long) bundle->key - (unsigned long) key);
+		((unsigned long)bundle->key - (unsigned long)key);
 }
 
 /*
@@ -69,14 +69,14 @@ int rxrpc_cmp_bundle(const struct rxrpc_conn_bundle *bundle,
 struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx,
 					   struct rxrpc_transport *trans,
 					   struct key *key,
-					   __be16 service_id,
+					   u16 service_id,
 					   gfp_t gfp)
 {
 	struct rxrpc_conn_bundle *bundle, *candidate;
 	struct rb_node *p, *parent, **pp;
 
 	_enter("%p{%x},%x,%hx,",
-	       rx, key_serial(key), trans->debug_id, ntohs(service_id));
+	       rx, key_serial(key), trans->debug_id, service_id);
 
 	if (rx->trans == trans && rx->bundle) {
 		atomic_inc(&rx->bundle->usage);
@@ -213,7 +213,7 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp)
 		conn->debug_id = atomic_inc_return(&rxrpc_debug_id);
 		conn->avail_calls = RXRPC_MAXCALLS;
 		conn->size_align = 4;
-		conn->header_size = sizeof(struct rxrpc_header);
+		conn->header_size = sizeof(struct rxrpc_wire_header);
 	}
 
 	_leave(" = %p{%d}", conn, conn ? conn->debug_id : 0);
@@ -230,7 +230,7 @@ static void rxrpc_assign_connection_id(struct rxrpc_connection *conn)
 	struct rxrpc_connection *xconn;
 	struct rb_node *parent, **p;
 	__be32 epoch;
-	u32 real_conn_id;
+	u32 cid;
 
 	_enter("");
 
@@ -241,7 +241,7 @@ static void rxrpc_assign_connection_id(struct rxrpc_connection *conn)
 	conn->trans->conn_idcounter += RXRPC_CID_INC;
 	if (conn->trans->conn_idcounter < RXRPC_CID_INC)
 		conn->trans->conn_idcounter = RXRPC_CID_INC;
-	real_conn_id = conn->trans->conn_idcounter;
+	cid = conn->trans->conn_idcounter;
 
 attempt_insertion:
 	parent = NULL;
@@ -255,9 +255,9 @@ attempt_insertion:
 			p = &(*p)->rb_left;
 		else if (epoch > xconn->epoch)
 			p = &(*p)->rb_right;
-		else if (real_conn_id < xconn->real_conn_id)
+		else if (cid < xconn->cid)
 			p = &(*p)->rb_left;
-		else if (real_conn_id > xconn->real_conn_id)
+		else if (cid > xconn->cid)
 			p = &(*p)->rb_right;
 		else
 			goto id_exists;
@@ -268,20 +268,19 @@ attempt_insertion:
 	rb_link_node(&conn->node, parent, p);
 	rb_insert_color(&conn->node, &conn->trans->client_conns);
 
-	conn->real_conn_id = real_conn_id;
-	conn->cid = htonl(real_conn_id);
+	conn->cid = cid;
 	write_unlock_bh(&conn->trans->conn_lock);
-	_leave(" [CONNID %x CID %x]", real_conn_id, ntohl(conn->cid));
+	_leave(" [CID %x]", cid);
 	return;
 
 	/* we found a connection with the proposed ID - walk the tree from that
 	 * point looking for the next unused ID */
 id_exists:
 	for (;;) {
-		real_conn_id += RXRPC_CID_INC;
-		if (real_conn_id < RXRPC_CID_INC) {
-			real_conn_id = RXRPC_CID_INC;
-			conn->trans->conn_idcounter = real_conn_id;
+		cid += RXRPC_CID_INC;
+		if (cid < RXRPC_CID_INC) {
+			cid = RXRPC_CID_INC;
+			conn->trans->conn_idcounter = cid;
 			goto attempt_insertion;
 		}
 
@@ -291,7 +290,7 @@ id_exists:
 
 		xconn = rb_entry(parent, struct rxrpc_connection, node);
 		if (epoch < xconn->epoch ||
-		    real_conn_id < xconn->real_conn_id)
+		    cid < xconn->cid)
 			goto attempt_insertion;
 	}
 }
@@ -334,7 +333,7 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn,
  */
 static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
 				   struct rxrpc_transport *trans,
-				   __be16 service_id,
+				   u16 service_id,
 				   struct rxrpc_call *call,
 				   gfp_t gfp)
 {
@@ -404,11 +403,11 @@ found_channel:
 	conn->channels[chan] = call;
 	call->conn = conn;
 	call->channel = chan;
-	call->cid = conn->cid | htonl(chan);
-	call->call_id = htonl(++conn->call_counter);
+	call->cid = conn->cid | chan;
+	call->call_id = ++conn->call_counter;
 
 	_net("CONNECT client on conn %d chan %d as call %x",
-	     conn->debug_id, chan, ntohl(call->call_id));
+	     conn->debug_id, chan, call->call_id);
 
 	spin_unlock(&trans->client_lock);
 
@@ -593,11 +592,11 @@ found_channel:
 	conn->channels[chan] = call;
 	call->conn = conn;
 	call->channel = chan;
-	call->cid = conn->cid | htonl(chan);
-	call->call_id = htonl(++conn->call_counter);
+	call->cid = conn->cid | chan;
+	call->call_id = ++conn->call_counter;
 
 	_net("CONNECT client on conn %d chan %d as call %x",
-	     conn->debug_id, chan, ntohl(call->call_id));
+	     conn->debug_id, chan, call->call_id);
 
 	ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS);
 	spin_unlock(&trans->client_lock);
@@ -620,21 +619,21 @@ interrupted:
  */
 struct rxrpc_connection *
 rxrpc_incoming_connection(struct rxrpc_transport *trans,
-			  struct rxrpc_header *hdr,
+			  struct rxrpc_host_header *hdr,
 			  gfp_t gfp)
 {
 	struct rxrpc_connection *conn, *candidate = NULL;
 	struct rb_node *p, **pp;
 	const char *new = "old";
 	__be32 epoch;
-	u32 conn_id;
+	u32 cid;
 
 	_enter("");
 
 	ASSERT(hdr->flags & RXRPC_CLIENT_INITIATED);
 
 	epoch = hdr->epoch;
-	conn_id = ntohl(hdr->cid) & RXRPC_CIDMASK;
+	cid = hdr->cid & RXRPC_CIDMASK;
 
 	/* search the connection list first */
 	read_lock_bh(&trans->conn_lock);
@@ -643,15 +642,15 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
 	while (p) {
 		conn = rb_entry(p, struct rxrpc_connection, node);
 
-		_debug("maybe %x", conn->real_conn_id);
+		_debug("maybe %x", conn->cid);
 
 		if (epoch < conn->epoch)
 			p = p->rb_left;
 		else if (epoch > conn->epoch)
 			p = p->rb_right;
-		else if (conn_id < conn->real_conn_id)
+		else if (cid < conn->cid)
 			p = p->rb_left;
-		else if (conn_id > conn->real_conn_id)
+		else if (cid > conn->cid)
 			p = p->rb_right;
 		else
 			goto found_extant_connection;
@@ -668,12 +667,11 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
 
 	candidate->trans = trans;
 	candidate->epoch = hdr->epoch;
-	candidate->cid = hdr->cid & cpu_to_be32(RXRPC_CIDMASK);
+	candidate->cid = hdr->cid & RXRPC_CIDMASK;
 	candidate->service_id = hdr->serviceId;
 	candidate->security_ix = hdr->securityIndex;
 	candidate->in_clientflag = RXRPC_CLIENT_INITIATED;
 	candidate->out_clientflag = 0;
-	candidate->real_conn_id = conn_id;
 	candidate->state = RXRPC_CONN_SERVER;
 	if (candidate->service_id)
 		candidate->state = RXRPC_CONN_SERVER_UNSECURED;
@@ -690,9 +688,9 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
 			pp = &(*pp)->rb_left;
 		else if (epoch > conn->epoch)
 			pp = &(*pp)->rb_right;
-		else if (conn_id < conn->real_conn_id)
+		else if (cid < conn->cid)
 			pp = &(*pp)->rb_left;
-		else if (conn_id > conn->real_conn_id)
+		else if (cid > conn->cid)
 			pp = &(*pp)->rb_right;
 		else
 			goto found_extant_second;
@@ -714,7 +712,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
 	new = "new";
 
 success:
-	_net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->real_conn_id);
+	_net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->cid);
 
 	_leave(" = %p {u=%d}", conn, atomic_read(&conn->usage));
 	return conn;
@@ -751,18 +749,17 @@ security_mismatch:
  * packet
  */
 struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
-					       struct rxrpc_header *hdr)
+					       struct rxrpc_host_header *hdr)
 {
 	struct rxrpc_connection *conn;
 	struct rb_node *p;
-	__be32 epoch;
-	u32 conn_id;
+	u32 epoch, cid;
 
-	_enter(",{%x,%x}", ntohl(hdr->cid), hdr->flags);
+	_enter(",{%x,%x}", hdr->cid, hdr->flags);
 
 	read_lock_bh(&trans->conn_lock);
 
-	conn_id = ntohl(hdr->cid) & RXRPC_CIDMASK;
+	cid = hdr->cid & RXRPC_CIDMASK;
 	epoch = hdr->epoch;
 
 	if (hdr->flags & RXRPC_CLIENT_INITIATED)
@@ -773,15 +770,15 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
 	while (p) {
 		conn = rb_entry(p, struct rxrpc_connection, node);
 
-		_debug("maybe %x", conn->real_conn_id);
+		_debug("maybe %x", conn->cid);
 
 		if (epoch < conn->epoch)
 			p = p->rb_left;
 		else if (epoch > conn->epoch)
 			p = p->rb_right;
-		else if (conn_id < conn->real_conn_id)
+		else if (cid < conn->cid)
 			p = p->rb_left;
-		else if (conn_id > conn->real_conn_id)
+		else if (cid > conn->cid)
 			p = p->rb_right;
 		else
 			goto found;
diff --git a/net/rxrpc/ar-connevent.c b/net/rxrpc/ar-connevent.c
index 45e81b7..1bdaaed 100644
--- a/net/rxrpc/ar-connevent.c
+++ b/net/rxrpc/ar-connevent.c
@@ -60,11 +60,12 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn, int state,
 static int rxrpc_abort_connection(struct rxrpc_connection *conn,
 				  u32 error, u32 abort_code)
 {
-	struct rxrpc_header hdr;
+	struct rxrpc_wire_header whdr;
 	struct msghdr msg;
 	struct kvec iov[2];
 	__be32 word;
 	size_t len;
+	u32 serial;
 	int ret;
 
 	_enter("%d,,%u,%u", conn->debug_id, error, abort_code);
@@ -89,28 +90,29 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn,
 	msg.msg_controllen = 0;
 	msg.msg_flags	= 0;
 
-	hdr.epoch	= conn->epoch;
-	hdr.cid		= conn->cid;
-	hdr.callNumber	= 0;
-	hdr.seq		= 0;
-	hdr.type	= RXRPC_PACKET_TYPE_ABORT;
-	hdr.flags	= conn->out_clientflag;
-	hdr.userStatus	= 0;
-	hdr.securityIndex = conn->security_ix;
-	hdr._rsvd	= 0;
-	hdr.serviceId	= conn->service_id;
+	whdr.epoch	= htonl(conn->epoch);
+	whdr.cid	= htonl(conn->cid);
+	whdr.callNumber	= 0;
+	whdr.seq	= 0;
+	whdr.type	= RXRPC_PACKET_TYPE_ABORT;
+	whdr.flags	= conn->out_clientflag;
+	whdr.userStatus	= 0;
+	whdr.securityIndex = conn->security_ix;
+	whdr._rsvd	= 0;
+	whdr.serviceId	= htons(conn->service_id);
 
 	word = htonl(abort_code);
 
-	iov[0].iov_base	= &hdr;
-	iov[0].iov_len	= sizeof(hdr);
+	iov[0].iov_base	= &whdr;
+	iov[0].iov_len	= sizeof(whdr);
 	iov[1].iov_base	= &word;
 	iov[1].iov_len	= sizeof(word);
 
 	len = iov[0].iov_len + iov[1].iov_len;
 
-	hdr.serial = htonl(atomic_inc_return(&conn->serial));
-	_proto("Tx CONN ABORT %%%u { %d }", ntohl(hdr.serial), abort_code);
+	serial = atomic_inc_return(&conn->serial);
+	whdr.serial = htonl(serial);
+	_proto("Tx CONN ABORT %%%u { %d }", serial, abort_code);
 
 	ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len);
 	if (ret < 0) {
@@ -146,8 +148,8 @@ static int rxrpc_process_event(struct rxrpc_connection *conn,
 			       u32 *_abort_code)
 {
 	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
-	__be32 tmp;
-	u32 serial;
+	__be32 wtmp;
+	u32 abort_code;
 	int loop, ret;
 
 	if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) {
@@ -155,19 +157,18 @@ static int rxrpc_process_event(struct rxrpc_connection *conn,
 		return -ECONNABORTED;
 	}
 
-	serial = ntohl(sp->hdr.serial);
-
-	_enter("{%d},{%u,%%%u},", conn->debug_id, sp->hdr.type, serial);
+	_enter("{%d},{%u,%%%u},", conn->debug_id, sp->hdr.type, sp->hdr.serial);
 
 	switch (sp->hdr.type) {
 	case RXRPC_PACKET_TYPE_ABORT:
-		if (skb_copy_bits(skb, 0, &tmp, sizeof(tmp)) < 0)
+		if (skb_copy_bits(skb, 0, &wtmp, sizeof(wtmp)) < 0)
 			return -EPROTO;
-		_proto("Rx ABORT %%%u { ac=%d }", serial, ntohl(tmp));
+		abort_code = ntohl(wtmp);
+		_proto("Rx ABORT %%%u { ac=%d }", sp->hdr.serial, abort_code);
 
 		conn->state = RXRPC_CONN_REMOTELY_ABORTED;
 		rxrpc_abort_calls(conn, RXRPC_CALL_REMOTELY_ABORTED,
-				  ntohl(tmp));
+				  abort_code);
 		return -ECONNABORTED;
 
 	case RXRPC_PACKET_TYPE_CHALLENGE:
@@ -335,7 +336,7 @@ void rxrpc_reject_packets(struct work_struct *work)
 		struct sockaddr_in sin;
 	} sa;
 	struct rxrpc_skb_priv *sp;
-	struct rxrpc_header hdr;
+	struct rxrpc_wire_header whdr;
 	struct rxrpc_local *local;
 	struct sk_buff *skb;
 	struct msghdr msg;
@@ -348,11 +349,11 @@ void rxrpc_reject_packets(struct work_struct *work)
 
 	_enter("%d", local->debug_id);
 
-	iov[0].iov_base = &hdr;
-	iov[0].iov_len = sizeof(hdr);
+	iov[0].iov_base = &whdr;
+	iov[0].iov_len = sizeof(whdr);
 	iov[1].iov_base = &code;
 	iov[1].iov_len = sizeof(code);
-	size = sizeof(hdr) + sizeof(code);
+	size = sizeof(whdr) + sizeof(code);
 
 	msg.msg_name = &sa;
 	msg.msg_control = NULL;
@@ -370,8 +371,8 @@ void rxrpc_reject_packets(struct work_struct *work)
 		break;
 	}
 
-	memset(&hdr, 0, sizeof(hdr));
-	hdr.type = RXRPC_PACKET_TYPE_ABORT;
+	memset(&whdr, 0, sizeof(whdr));
+	whdr.type = RXRPC_PACKET_TYPE_ABORT;
 
 	while ((skb = skb_dequeue(&local->reject_queue))) {
 		sp = rxrpc_skb(skb);
@@ -381,13 +382,13 @@ void rxrpc_reject_packets(struct work_struct *work)
 			sa.sin.sin_addr.s_addr = ip_hdr(skb)->saddr;
 			code = htonl(skb->priority);
 
-			hdr.epoch = sp->hdr.epoch;
-			hdr.cid = sp->hdr.cid;
-			hdr.callNumber = sp->hdr.callNumber;
-			hdr.serviceId = sp->hdr.serviceId;
-			hdr.flags = sp->hdr.flags;
-			hdr.flags ^= RXRPC_CLIENT_INITIATED;
-			hdr.flags &= RXRPC_CLIENT_INITIATED;
+			whdr.epoch	= htonl(sp->hdr.epoch);
+			whdr.cid	= htonl(sp->hdr.cid);
+			whdr.callNumber	= htonl(sp->hdr.callNumber);
+			whdr.serviceId	= htons(sp->hdr.serviceId);
+			whdr.flags	= sp->hdr.flags;
+			whdr.flags	^= RXRPC_CLIENT_INITIATED;
+			whdr.flags	&= RXRPC_CLIENT_INITIATED;
 
 			kernel_sendmsg(local->socket, &msg, iov, 2, size);
 			break;
diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c
index 9185535..e6396a8 100644
--- a/net/rxrpc/ar-input.c
+++ b/net/rxrpc/ar-input.c
@@ -310,8 +310,8 @@ static void rxrpc_assume_implicit_ackall(struct rxrpc_call *call, u32 serial)
 void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 {
 	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
-	__be32 _abort_code;
-	u32 serial, hi_serial, seq, abort_code;
+	__be32 wtmp;
+	u32 hi_serial, abort_code;
 
 	_enter("%p,%p", call, skb);
 
@@ -330,16 +330,15 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 
 	/* track the latest serial number on this connection for ACK packet
 	 * information */
-	serial = ntohl(sp->hdr.serial);
 	hi_serial = atomic_read(&call->conn->hi_serial);
-	while (serial > hi_serial)
+	while (sp->hdr.serial > hi_serial)
 		hi_serial = atomic_cmpxchg(&call->conn->hi_serial, hi_serial,
-					   serial);
+					   sp->hdr.serial);
 
 	/* request ACK generation for any ACK or DATA packet that requests
 	 * it */
 	if (sp->hdr.flags & RXRPC_REQUEST_ACK) {
-		_proto("ACK Requested on %%%u", serial);
+		_proto("ACK Requested on %%%u", sp->hdr.serial);
 		rxrpc_propose_ACK(call, RXRPC_ACK_REQUESTED, sp->hdr.serial, false);
 	}
 
@@ -347,12 +346,11 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 	case RXRPC_PACKET_TYPE_ABORT:
 		_debug("abort");
 
-		if (skb_copy_bits(skb, 0, &_abort_code,
-				  sizeof(_abort_code)) < 0)
+		if (skb_copy_bits(skb, 0, &wtmp, sizeof(wtmp)) < 0)
 			goto protocol_error;
 
-		abort_code = ntohl(_abort_code);
-		_proto("Rx ABORT %%%u { %x }", serial, abort_code);
+		abort_code = ntohl(wtmp);
+		_proto("Rx ABORT %%%u { %x }", sp->hdr.serial, abort_code);
 
 		write_lock_bh(&call->state_lock);
 		if (call->state < RXRPC_CALL_COMPLETE) {
@@ -364,7 +362,7 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 		goto free_packet_unlock;
 
 	case RXRPC_PACKET_TYPE_BUSY:
-		_proto("Rx BUSY %%%u", serial);
+		_proto("Rx BUSY %%%u", sp->hdr.serial);
 
 		if (call->conn->out_clientflag)
 			goto protocol_error;
@@ -382,15 +380,13 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 		}
 
 	default:
-		_proto("Rx %s %%%u", rxrpc_pkts[sp->hdr.type], serial);
+		_proto("Rx %s %%%u", rxrpc_pkts[sp->hdr.type], sp->hdr.serial);
 		goto protocol_error;
 
 	case RXRPC_PACKET_TYPE_DATA:
-		seq = ntohl(sp->hdr.seq);
+		_proto("Rx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq);
 
-		_proto("Rx DATA %%%u { #%u }", serial, seq);
-
-		if (seq == 0)
+		if (sp->hdr.seq == 0)
 			goto protocol_error;
 
 		call->ackr_prev_seq = sp->hdr.seq;
@@ -398,9 +394,9 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
 		/* received data implicitly ACKs all of the request packets we
 		 * sent when we're acting as a client */
 		if (call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY)
-			rxrpc_assume_implicit_ackall(call, serial);
+			rxrpc_assume_implicit_ackall(call, sp->hdr.serial);
 
-		switch (rxrpc_fast_process_data(call, skb, seq)) {
+		switch (rxrpc_fast_process_data(call, skb, sp->hdr.seq)) {
 		case 0:
 			skb = NULL;
 			goto done;
@@ -481,12 +477,12 @@ static void rxrpc_process_jumbo_packet(struct rxrpc_call *call,
 		if (!pskb_pull(jumbo, sizeof(jhdr)))
 			BUG();
 
-		sp->hdr.seq	= htonl(ntohl(sp->hdr.seq) + 1);
-		sp->hdr.serial	= htonl(ntohl(sp->hdr.serial) + 1);
+		sp->hdr.seq	+= 1;
+		sp->hdr.serial	+= 1;
 		sp->hdr.flags	= jhdr.flags;
 		sp->hdr._rsvd	= jhdr._rsvd;
 
-		_proto("Rx DATA Jumbo %%%u", ntohl(sp->hdr.serial) - 1);
+		_proto("Rx DATA Jumbo %%%u", sp->hdr.serial - 1);
 
 		rxrpc_fast_process_packet(call, part);
 		part = NULL;
@@ -607,6 +603,35 @@ static void rxrpc_post_packet_to_local(struct rxrpc_local *local,
 	rxrpc_queue_work(&local->event_processor);
 }
 
+/*
+ * Extract the wire header from a packet and translate the byte order.
+ */
+static noinline
+int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb)
+{
+	struct rxrpc_wire_header whdr;
+
+	/* dig out the RxRPC connection details */
+	if (skb_copy_bits(skb, sizeof(struct udphdr), &whdr, sizeof(whdr)) < 0)
+		return -EBADMSG;
+	if (!pskb_pull(skb, sizeof(struct udphdr) + sizeof(whdr)))
+		BUG();
+
+	memset(sp, 0, sizeof(*sp));
+	sp->hdr.epoch		= ntohl(whdr.epoch);
+	sp->hdr.cid		= ntohl(whdr.cid);
+	sp->hdr.callNumber	= ntohl(whdr.callNumber);
+	sp->hdr.seq		= ntohl(whdr.seq);
+	sp->hdr.serial		= ntohl(whdr.serial);
+	sp->hdr.flags		= whdr.flags;
+	sp->hdr.type		= whdr.type;
+	sp->hdr.userStatus	= whdr.userStatus;
+	sp->hdr.securityIndex	= whdr.securityIndex;
+	sp->hdr._rsvd		= ntohs(whdr._rsvd);
+	sp->hdr.serviceId	= ntohs(whdr.serviceId);
+	return 0;
+}
+
 static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
 					       struct sk_buff *skb,
 					       struct rxrpc_skb_priv *sp)
@@ -686,27 +711,22 @@ void rxrpc_data_ready(struct sock *sk)
 
 	UDP_INC_STATS_BH(&init_net, UDP_MIB_INDATAGRAMS, 0);
 
-	/* the socket buffer we have is owned by UDP, with UDP's data all over
-	 * it, but we really want our own */
+	/* The socket buffer we have is owned by UDP, with UDP's data all over
+	 * it, but we really want our own data there.
+	 */
 	skb_orphan(skb);
 	sp = rxrpc_skb(skb);
-	memset(sp, 0, sizeof(*sp));
 
 	_net("Rx UDP packet from %08x:%04hu",
 	     ntohl(ip_hdr(skb)->saddr), ntohs(udp_hdr(skb)->source));
 
 	/* dig out the RxRPC connection details */
-	if (skb_copy_bits(skb, sizeof(struct udphdr), &sp->hdr,
-			  sizeof(sp->hdr)) < 0)
+	if (rxrpc_extract_header(sp, skb) < 0)
 		goto bad_message;
-	if (!pskb_pull(skb, sizeof(struct udphdr) + sizeof(sp->hdr)))
-		BUG();
 
 	_net("Rx RxRPC %s ep=%x call=%x:%x",
 	     sp->hdr.flags & RXRPC_CLIENT_INITIATED ? "ToServer" : "ToClient",
-	     ntohl(sp->hdr.epoch),
-	     ntohl(sp->hdr.cid),
-	     ntohl(sp->hdr.callNumber));
+	     sp->hdr.epoch, sp->hdr.cid, sp->hdr.callNumber);
 
 	if (sp->hdr.type == 0 || sp->hdr.type >= RXRPC_N_PACKET_TYPES) {
 		_proto("Rx Bad Packet Type %u", sp->hdr.type);
@@ -737,14 +757,9 @@ void rxrpc_data_ready(struct sock *sk)
 		rxrpc_put_connection(conn);
 	} else {
 		struct rxrpc_call *call;
-		u8 in_clientflag = 0;
-
-		if (sp->hdr.flags & RXRPC_CLIENT_INITIATED)
-			in_clientflag = RXRPC_CLIENT_INITIATED;
-		call = rxrpc_find_call_hash(in_clientflag, sp->hdr.cid,
-					    sp->hdr.callNumber, sp->hdr.epoch,
-					    sp->hdr.serviceId, local, AF_INET,
-					    (u8 *)&ip_hdr(skb)->saddr);
+
+		call = rxrpc_find_call_hash(&sp->hdr, local,
+					    AF_INET, &ip_hdr(skb)->saddr);
 		if (call)
 			rxrpc_post_packet_to_call(call, skb);
 		else
@@ -759,7 +774,7 @@ cant_route_call:
 	_debug("can't route call");
 	if (sp->hdr.flags & RXRPC_CLIENT_INITIATED &&
 	    sp->hdr.type == RXRPC_PACKET_TYPE_DATA) {
-		if (sp->hdr.seq == cpu_to_be32(1)) {
+		if (sp->hdr.seq == 1) {
 			_debug("first packet");
 			skb_queue_tail(&local->accept_queue, skb);
 			rxrpc_queue_work(&local->acceptor);
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 3f29406..06bf5ab 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -70,12 +70,31 @@ struct rxrpc_sock {
 #define RXRPC_SECURITY_MAX	RXRPC_SECURITY_ENCRYPT
 	struct sockaddr_rxrpc	srx;		/* local address */
 	sa_family_t		proto;		/* protocol created with */
-	__be16			service_id;	/* service ID of local/remote service */
 };
 
 #define rxrpc_sk(__sk) container_of((__sk), struct rxrpc_sock, sk)
 
 /*
+ * CPU-byteorder normalised Rx packet header.
+ */
+struct rxrpc_host_header {
+	u32		epoch;		/* client boot timestamp */
+	u32		cid;		/* connection and channel ID */
+	u32		callNumber;	/* call ID (0 for connection-level packets) */
+	u32		seq;		/* sequence number of pkt in call stream */
+	u32		serial;		/* serial number of pkt sent to network */
+	u8		type;		/* packet type */
+	u8		flags;		/* packet flags */
+	u8		userStatus;	/* app-layer defined status */
+	u8		securityIndex;	/* security protocol ID */
+	union {
+		u16	_rsvd;		/* reserved */
+		u16	cksum;		/* kerberos security checksum */
+	};
+	u16		serviceId;	/* service ID */
+} __packed;
+
+/*
  * RxRPC socket buffer private variables
  * - max 48 bytes (struct sk_buff::cb)
  */
@@ -89,7 +108,7 @@ struct rxrpc_skb_priv {
 		bool		need_resend;	/* T if needs resending */
 	};
 
-	struct rxrpc_header	hdr;		/* RxRPC packet header from this packet */
+	struct rxrpc_host_header hdr;		/* RxRPC packet header from this packet */
 };
 
 #define rxrpc_skb(__skb) ((struct rxrpc_skb_priv *) &(__skb)->cb)
@@ -230,7 +249,7 @@ struct rxrpc_conn_bundle {
 	atomic_t		usage;
 	int			debug_id;	/* debug ID for printks */
 	unsigned short		num_conns;	/* number of connections in this bundle */
-	__be16			service_id;	/* service ID */
+	u16			service_id;	/* Service ID for this bundle */
 	u8			security_ix;	/* security type */
 };
 
@@ -260,7 +279,6 @@ struct rxrpc_connection {
 	rwlock_t		lock;		/* access lock */
 	spinlock_t		state_lock;	/* state-change lock */
 	atomic_t		usage;
-	u32			real_conn_id;	/* connection ID (host-endian) */
 	enum {					/* current state of connection */
 		RXRPC_CONN_UNUSED,		/* - connection not yet attempted */
 		RXRPC_CONN_CLIENT,		/* - client connection */
@@ -282,11 +300,9 @@ struct rxrpc_connection {
 	u8			security_size;	/* security header size */
 	u32			security_level;	/* security level negotiated */
 	u32			security_nonce;	/* response re-use preventer */
-
-	/* the following are all in net order */
-	__be32			epoch;		/* epoch of this connection */
-	__be32			cid;		/* connection ID */
-	__be16			service_id;	/* service ID */
+	u32			epoch;		/* epoch of this connection */
+	u32			cid;		/* connection ID */
+	u16			service_id;	/* service ID for this connection */
 	u8			security_ix;	/* security type */
 	u8			in_clientflag;	/* RXRPC_CLIENT_INITIATED if we are server */
 	u8			out_clientflag;	/* RXRPC_CLIENT_INITIATED if we are client */
@@ -406,9 +422,9 @@ struct rxrpc_call {
 	rxrpc_seq_t		rx_data_eaten;	/* last data seq ID consumed by recvmsg */
 	rxrpc_seq_t		rx_first_oos;	/* first packet in rx_oos_queue (or 0) */
 	rxrpc_seq_t		ackr_win_top;	/* top of ACK window (rx_data_eaten is bottom) */
-	rxrpc_seq_net_t		ackr_prev_seq;	/* previous sequence number received */
+	rxrpc_seq_t		ackr_prev_seq;	/* previous sequence number received */
 	u8			ackr_reason;	/* reason to ACK */
-	__be32			ackr_serial;	/* serial of packet being ACK'd */
+	rxrpc_serial_t		ackr_serial;	/* serial of packet being ACK'd */
 	atomic_t		ackr_not_idle;	/* number of packets in Rx queue */
 
 	/* received packet records, 1 bit per record */
@@ -420,11 +436,10 @@ struct rxrpc_call {
 	u8			in_clientflag;	/* Copy of conn->in_clientflag for hashing */
 	struct rxrpc_local	*local;		/* Local endpoint. Used for hashing. */
 	sa_family_t		proto;		/* Frame protocol */
-	/* the following should all be in net order */
-	__be32			cid;		/* connection ID + channel index  */
-	__be32			call_id;	/* call ID on connection  */
-	__be32			epoch;		/* epoch of this connection */
-	__be16			service_id;	/* service ID */
+	u32			call_id;	/* call ID on connection  */
+	u32			cid;		/* connection ID plus channel index */
+	u32			epoch;		/* epoch of this connection */
+	u16			service_id;	/* service ID */
 	union {					/* Peer IP address for hashing */
 		__be32	ipv4_addr;
 		__u8	ipv6_addr[16];		/* Anticipates eventual IPv6 support */
@@ -449,7 +464,7 @@ static inline void rxrpc_abort_call(struct rxrpc_call *call, u32 abort_code)
  * af_rxrpc.c
  */
 extern atomic_t rxrpc_n_skbs;
-extern __be32 rxrpc_epoch;
+extern u32 rxrpc_epoch;
 extern atomic_t rxrpc_debug_id;
 extern struct workqueue_struct *rxrpc_workqueue;
 
@@ -470,8 +485,8 @@ extern unsigned rxrpc_rx_window_size;
 extern unsigned rxrpc_rx_mtu;
 extern unsigned rxrpc_rx_jumbo_max;
 
-void __rxrpc_propose_ACK(struct rxrpc_call *, u8, __be32, bool);
-void rxrpc_propose_ACK(struct rxrpc_call *, u8, __be32, bool);
+void __rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool);
+void rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool);
 void rxrpc_process_call(struct work_struct *);
 
 /*
@@ -483,15 +498,15 @@ extern struct kmem_cache *rxrpc_call_jar;
 extern struct list_head rxrpc_calls;
 extern rwlock_t rxrpc_call_lock;
 
-struct rxrpc_call *rxrpc_find_call_hash(u8,  __be32, __be32, __be32,
-					__be16, void *, sa_family_t, const u8 *);
+struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *,
+					void *, sa_family_t, const void *);
 struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *,
 					 struct rxrpc_transport *,
 					 struct rxrpc_conn_bundle *,
 					 unsigned long, int, gfp_t);
 struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *,
 				       struct rxrpc_connection *,
-				       struct rxrpc_header *, gfp_t);
+				       struct rxrpc_host_header *, gfp_t);
 struct rxrpc_call *rxrpc_find_server_call(struct rxrpc_sock *, unsigned long);
 void rxrpc_release_call(struct rxrpc_call *);
 void rxrpc_release_calls_on_socket(struct rxrpc_sock *);
@@ -507,16 +522,16 @@ extern rwlock_t rxrpc_connection_lock;
 
 struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *,
 					   struct rxrpc_transport *,
-					   struct key *, __be16, gfp_t);
+					   struct key *, u16, gfp_t);
 void rxrpc_put_bundle(struct rxrpc_transport *, struct rxrpc_conn_bundle *);
 int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_transport *,
 		       struct rxrpc_conn_bundle *, struct rxrpc_call *, gfp_t);
 void rxrpc_put_connection(struct rxrpc_connection *);
 void __exit rxrpc_destroy_all_connections(void);
 struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *,
-					       struct rxrpc_header *);
+					       struct rxrpc_host_header *);
 extern struct rxrpc_connection *
-rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_header *,
+rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_host_header *,
 			  gfp_t);
 
 /*
diff --git a/net/rxrpc/ar-local.c b/net/rxrpc/ar-local.c
index 78483b4..4e1e6db 100644
--- a/net/rxrpc/ar-local.c
+++ b/net/rxrpc/ar-local.c
@@ -323,9 +323,11 @@ void __exit rxrpc_destroy_all_locals(void)
  * Reply to a version request
  */
 static void rxrpc_send_version_request(struct rxrpc_local *local,
-				       struct rxrpc_header *hdr,
+				       struct rxrpc_host_header *hdr,
 				       struct sk_buff *skb)
 {
+	struct rxrpc_wire_header whdr;
+	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
 	struct sockaddr_in sin;
 	struct msghdr msg;
 	struct kvec iov[2];
@@ -344,15 +346,20 @@ static void rxrpc_send_version_request(struct rxrpc_local *local,
 	msg.msg_controllen = 0;
 	msg.msg_flags	= 0;
 
-	hdr->seq	= 0;
-	hdr->serial	= 0;
-	hdr->type	= RXRPC_PACKET_TYPE_VERSION;
-	hdr->flags	= RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED);
-	hdr->userStatus	= 0;
-	hdr->_rsvd	= 0;
-
-	iov[0].iov_base	= hdr;
-	iov[0].iov_len	= sizeof(*hdr);
+	whdr.epoch	= htonl(sp->hdr.epoch);
+	whdr.cid	= htonl(sp->hdr.cid);
+	whdr.callNumber	= htonl(sp->hdr.callNumber);
+	whdr.seq	= 0;
+	whdr.serial	= 0;
+	whdr.type	= RXRPC_PACKET_TYPE_VERSION;
+	whdr.flags	= RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED);
+	whdr.userStatus	= 0;
+	whdr.securityIndex = 0;
+	whdr._rsvd	= 0;
+	whdr.serviceId	= htons(sp->hdr.serviceId);
+
+	iov[0].iov_base	= &whdr;
+	iov[0].iov_len	= sizeof(whdr);
 	iov[1].iov_base	= (char *)rxrpc_version_string;
 	iov[1].iov_len	= sizeof(rxrpc_version_string);
 
@@ -383,7 +390,7 @@ static void rxrpc_process_local_events(struct work_struct *work)
 	while ((skb = skb_dequeue(&local->event_queue))) {
 		struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
 
-		kdebug("{%d},{%u}", local->debug_id, sp->hdr.type);
+		_debug("{%d},{%u}", local->debug_id, sp->hdr.type);
 
 		switch (sp->hdr.type) {
 		case RXRPC_PACKET_TYPE_VERSION:
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 9e1527a..353f5c9 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -136,7 +136,7 @@ int rxrpc_client_sendmsg(struct rxrpc_sock *rx, struct rxrpc_transport *trans,
 	struct rxrpc_call *call;
 	unsigned long user_call_ID = 0;
 	struct key *key;
-	__be16 service_id;
+	u16 service_id;
 	u32 abort_code = 0;
 	int ret;
 
@@ -151,11 +151,11 @@ int rxrpc_client_sendmsg(struct rxrpc_sock *rx, struct rxrpc_transport *trans,
 
 	bundle = NULL;
 	if (trans) {
-		service_id = rx->service_id;
+		service_id = rx->srx.srx_service;
 		if (msg->msg_name) {
 			DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx,
 					 msg->msg_name);
-			service_id = htons(srx->srx_service);
+			service_id = srx->srx_service;
 		}
 		key = rx->key;
 		if (key && !rx->key->payload.data[0])
@@ -348,7 +348,7 @@ int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb)
 
 	/* send the packet with the don't fragment bit set if we currently
 	 * think it's small enough */
-	if (skb->len - sizeof(struct rxrpc_header) < trans->peer->maxdata) {
+	if (skb->len - sizeof(struct rxrpc_wire_header) < trans->peer->maxdata) {
 		down_read(&trans->local->defrag_sem);
 		/* send the packet by UDP
 		 * - returns -EMSGSIZE if UDP would have to fragment the packet
@@ -480,8 +480,7 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb,
 		write_unlock_bh(&call->state_lock);
 	}
 
-	_proto("Tx DATA %%%u { #%u }",
-	       ntohl(sp->hdr.serial), ntohl(sp->hdr.seq));
+	_proto("Tx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq);
 
 	sp->need_resend = false;
 	sp->resend_at = jiffies + rxrpc_resend_timeout;
@@ -513,6 +512,29 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb,
 }
 
 /*
+ * Convert a host-endian header into a network-endian header.
+ */
+static void rxrpc_insert_header(struct sk_buff *skb)
+{
+	struct rxrpc_wire_header whdr;
+	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
+
+	whdr.epoch	= htonl(sp->hdr.epoch);
+	whdr.cid	= htonl(sp->hdr.cid);
+	whdr.callNumber	= htonl(sp->hdr.callNumber);
+	whdr.seq	= htonl(sp->hdr.seq);
+	whdr.serial	= htonl(sp->hdr.serial);
+	whdr.type	= sp->hdr.type;
+	whdr.flags	= sp->hdr.flags;
+	whdr.userStatus	= sp->hdr.userStatus;
+	whdr.securityIndex = sp->hdr.securityIndex;
+	whdr._rsvd	= htons(sp->hdr._rsvd);
+	whdr.serviceId	= htons(sp->hdr.serviceId);
+
+	memcpy(skb->head, &whdr, sizeof(whdr));
+}
+
+/*
  * send data through a socket
  * - must be called in process context
  * - caller holds the socket locked
@@ -650,17 +672,16 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 
 			seq = atomic_inc_return(&call->sequence);
 
-			sp->hdr.epoch = conn->epoch;
-			sp->hdr.cid = call->cid;
+			sp->hdr.epoch	= conn->epoch;
+			sp->hdr.cid	= call->cid;
 			sp->hdr.callNumber = call->call_id;
-			sp->hdr.seq = htonl(seq);
-			sp->hdr.serial =
-				htonl(atomic_inc_return(&conn->serial));
-			sp->hdr.type = RXRPC_PACKET_TYPE_DATA;
+			sp->hdr.seq	= seq;
+			sp->hdr.serial	= atomic_inc_return(&conn->serial);
+			sp->hdr.type	= RXRPC_PACKET_TYPE_DATA;
 			sp->hdr.userStatus = 0;
 			sp->hdr.securityIndex = conn->security_ix;
-			sp->hdr._rsvd = 0;
-			sp->hdr.serviceId = conn->service_id;
+			sp->hdr._rsvd	= 0;
+			sp->hdr.serviceId = call->service_id;
 
 			sp->hdr.flags = conn->out_clientflag;
 			if (msg_data_left(msg) == 0 && !more)
@@ -673,12 +694,11 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 
 			ret = rxrpc_secure_packet(
 				call, skb, skb->mark,
-				skb->head + sizeof(struct rxrpc_header));
+				skb->head + sizeof(struct rxrpc_wire_header));
 			if (ret < 0)
 				goto out;
 
-			memcpy(skb->head, &sp->hdr,
-			       sizeof(struct rxrpc_header));
+			rxrpc_insert_header(skb);
 			rxrpc_queue_packet(call, skb, !msg_data_left(msg) && !more);
 			skb = NULL;
 		}
diff --git a/net/rxrpc/ar-peer.c b/net/rxrpc/ar-peer.c
index bebaa43..dc089b1 100644
--- a/net/rxrpc/ar-peer.c
+++ b/net/rxrpc/ar-peer.c
@@ -92,7 +92,7 @@ static struct rxrpc_peer *rxrpc_alloc_peer(struct sockaddr_rxrpc *srx,
 			BUG();
 		}
 
-		peer->hdrsize += sizeof(struct rxrpc_header);
+		peer->hdrsize += sizeof(struct rxrpc_wire_header);
 		peer->maxdata = peer->mtu - peer->hdrsize;
 	}
 
diff --git a/net/rxrpc/ar-proc.c b/net/rxrpc/ar-proc.c
index 38047f7..525b2ba 100644
--- a/net/rxrpc/ar-proc.c
+++ b/net/rxrpc/ar-proc.c
@@ -74,9 +74,9 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
 		   " %-8.8s %08x %lx\n",
 		   lbuff,
 		   rbuff,
-		   ntohs(call->conn->service_id),
-		   ntohl(call->conn->cid),
-		   ntohl(call->call_id),
+		   call->conn->service_id,
+		   call->cid,
+		   call->call_id,
 		   call->conn->in_clientflag ? "Svc" : "Clt",
 		   atomic_read(&call->usage),
 		   rxrpc_call_states[call->state],
@@ -157,8 +157,8 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v)
 		   " %s %08x %08x %08x\n",
 		   lbuff,
 		   rbuff,
-		   ntohs(conn->service_id),
-		   ntohl(conn->cid),
+		   conn->service_id,
+		   conn->cid,
 		   conn->call_counter,
 		   conn->in_clientflag ? "Svc" : "Clt",
 		   atomic_read(&conn->usage),
diff --git a/net/rxrpc/ar-recvmsg.c b/net/rxrpc/ar-recvmsg.c
index 70f4703..64facba 100644
--- a/net/rxrpc/ar-recvmsg.c
+++ b/net/rxrpc/ar-recvmsg.c
@@ -158,7 +158,7 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 			goto receive_non_data_message;
 
 		_debug("recvmsg DATA #%u { %d, %d }",
-		       ntohl(sp->hdr.seq), skb->len, sp->offset);
+		       sp->hdr.seq, skb->len, sp->offset);
 
 		if (!continue_call) {
 			/* only set the control data once per recvmsg() */
@@ -169,11 +169,11 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 			ASSERT(test_bit(RXRPC_CALL_HAS_USERID, &call->flags));
 		}
 
-		ASSERTCMP(ntohl(sp->hdr.seq), >=, call->rx_data_recv);
-		ASSERTCMP(ntohl(sp->hdr.seq), <=, call->rx_data_recv + 1);
-		call->rx_data_recv = ntohl(sp->hdr.seq);
+		ASSERTCMP(sp->hdr.seq, >=, call->rx_data_recv);
+		ASSERTCMP(sp->hdr.seq, <=, call->rx_data_recv + 1);
+		call->rx_data_recv = sp->hdr.seq;
 
-		ASSERTCMP(ntohl(sp->hdr.seq), >, call->rx_data_eaten);
+		ASSERTCMP(sp->hdr.seq, >, call->rx_data_eaten);
 
 		offset = sp->offset;
 		copy = skb->len - offset;
@@ -364,11 +364,11 @@ void rxrpc_kernel_data_delivered(struct sk_buff *skb)
 	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
 	struct rxrpc_call *call = sp->call;
 
-	ASSERTCMP(ntohl(sp->hdr.seq), >=, call->rx_data_recv);
-	ASSERTCMP(ntohl(sp->hdr.seq), <=, call->rx_data_recv + 1);
-	call->rx_data_recv = ntohl(sp->hdr.seq);
+	ASSERTCMP(sp->hdr.seq, >=, call->rx_data_recv);
+	ASSERTCMP(sp->hdr.seq, <=, call->rx_data_recv + 1);
+	call->rx_data_recv = sp->hdr.seq;
 
-	ASSERTCMP(ntohl(sp->hdr.seq), >, call->rx_data_eaten);
+	ASSERTCMP(sp->hdr.seq, >, call->rx_data_eaten);
 	rxrpc_free_skb(skb);
 }
 
diff --git a/net/rxrpc/ar-security.c b/net/rxrpc/ar-security.c
index 8334474..e2f4c49 100644
--- a/net/rxrpc/ar-security.c
+++ b/net/rxrpc/ar-security.c
@@ -171,7 +171,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn)
 
 	_enter("");
 
-	sprintf(kdesc, "%u:%u", ntohs(conn->service_id), conn->security_ix);
+	sprintf(kdesc, "%u:%u", conn->service_id, conn->security_ix);
 
 	sec = rxrpc_security_lookup(conn->security_ix);
 	if (!sec) {
@@ -182,7 +182,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn)
 	/* find the service */
 	read_lock_bh(&local->services_lock);
 	list_for_each_entry(rx, &local->services, listen_link) {
-		if (rx->service_id == conn->service_id)
+		if (rx->srx.srx_service == conn->service_id)
 			goto found_service;
 	}
 
diff --git a/net/rxrpc/ar-skbuff.c b/net/rxrpc/ar-skbuff.c
index 81f3c02..ae9f93f 100644
--- a/net/rxrpc/ar-skbuff.c
+++ b/net/rxrpc/ar-skbuff.c
@@ -59,7 +59,7 @@ static void rxrpc_hard_ACK_data(struct rxrpc_call *call,
 
 	spin_lock_bh(&call->lock);
 
-	_debug("hard ACK #%u", ntohl(sp->hdr.seq));
+	_debug("hard ACK #%u", sp->hdr.seq);
 
 	for (loop = 0; loop < RXRPC_ACKR_WINDOW_ASZ; loop++) {
 		call->ackr_window[loop] >>= 1;
@@ -67,7 +67,7 @@ static void rxrpc_hard_ACK_data(struct rxrpc_call *call,
 			call->ackr_window[loop + 1] << (BITS_PER_LONG - 1);
 	}
 
-	seq = ntohl(sp->hdr.seq);
+	seq = sp->hdr.seq;
 	ASSERTCMP(seq, ==, call->rx_data_eaten + 1);
 	call->rx_data_eaten = seq;
 
diff --git a/net/rxrpc/ar-transport.c b/net/rxrpc/ar-transport.c
index 9946467..5f9b9d4 100644
--- a/net/rxrpc/ar-transport.c
+++ b/net/rxrpc/ar-transport.c
@@ -51,6 +51,7 @@ static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local,
 		spin_lock_init(&trans->client_lock);
 		rwlock_init(&trans->conn_lock);
 		atomic_set(&trans->usage, 1);
+		trans->conn_idcounter = peer->srx.srx_service << 16;
 		trans->debug_id = atomic_inc_return(&rxrpc_debug_id);
 
 		if (peer->srx.transport.family == AF_INET) {
diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index d7a9ab5..1604802 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -132,8 +132,8 @@ static void rxkad_prime_packet_security(struct rxrpc_connection *conn)
 	desc.info = iv.x;
 	desc.flags = 0;
 
-	tmpbuf.x[0] = conn->epoch;
-	tmpbuf.x[1] = conn->cid;
+	tmpbuf.x[0] = htonl(conn->epoch);
+	tmpbuf.x[1] = htonl(conn->cid);
 	tmpbuf.x[2] = 0;
 	tmpbuf.x[3] = htonl(conn->security_ix);
 
@@ -169,8 +169,8 @@ static int rxkad_secure_packet_auth(const struct rxrpc_call *call,
 
 	_enter("");
 
-	check = ntohl(sp->hdr.seq ^ sp->hdr.callNumber);
-	data_size |= (u32) check << 16;
+	check = sp->hdr.seq ^ sp->hdr.callNumber;
+	data_size |= (u32)check << 16;
 
 	tmpbuf.hdr.data_size = htonl(data_size);
 	memcpy(&tmpbuf.first, sechdr + 4, sizeof(tmpbuf.first));
@@ -215,9 +215,9 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call,
 
 	_enter("");
 
-	check = ntohl(sp->hdr.seq ^ sp->hdr.callNumber);
+	check = sp->hdr.seq ^ sp->hdr.callNumber;
 
-	rxkhdr.data_size = htonl(data_size | (u32) check << 16);
+	rxkhdr.data_size = htonl(data_size | (u32)check << 16);
 	rxkhdr.checksum = 0;
 
 	/* encrypt from the session key */
@@ -262,14 +262,13 @@ static int rxkad_secure_packet(const struct rxrpc_call *call,
 	struct {
 		__be32 x[2];
 	} tmpbuf __attribute__((aligned(8))); /* must all be in same page */
-	__be32 x;
-	u32 y;
+	u32 x, y;
 	int ret;
 
 	sp = rxrpc_skb(skb);
 
 	_enter("{%d{%x}},{#%u},%zu,",
-	       call->debug_id, key_serial(call->conn->key), ntohl(sp->hdr.seq),
+	       call->debug_id, key_serial(call->conn->key), sp->hdr.seq,
 	       data_size);
 
 	if (!call->conn->cipher)
@@ -286,10 +285,10 @@ static int rxkad_secure_packet(const struct rxrpc_call *call,
 	desc.flags = 0;
 
 	/* calculate the security checksum */
-	x = htonl(call->channel << (32 - RXRPC_CIDSHIFT));
-	x |= sp->hdr.seq & cpu_to_be32(0x3fffffff);
-	tmpbuf.x[0] = sp->hdr.callNumber;
-	tmpbuf.x[1] = x;
+	x = call->channel << (32 - RXRPC_CIDSHIFT);
+	x |= sp->hdr.seq & 0x3fffffff;
+	tmpbuf.x[0] = htonl(sp->hdr.callNumber);
+	tmpbuf.x[1] = htonl(x);
 
 	sg_init_one(&sg[0], &tmpbuf, sizeof(tmpbuf));
 	sg_init_one(&sg[1], &tmpbuf, sizeof(tmpbuf));
@@ -299,7 +298,7 @@ static int rxkad_secure_packet(const struct rxrpc_call *call,
 	y = (y >> 16) & 0xffff;
 	if (y == 0)
 		y = 1; /* zero checksums are not permitted */
-	sp->hdr.cksum = htons(y);
+	sp->hdr.cksum = y;
 
 	switch (call->conn->security_level) {
 	case RXRPC_SECURITY_PLAIN:
@@ -368,7 +367,7 @@ static int rxkad_verify_packet_auth(const struct rxrpc_call *call,
 	data_size = buf & 0xffff;
 
 	check = buf >> 16;
-	check ^= ntohl(sp->hdr.seq ^ sp->hdr.callNumber);
+	check ^= sp->hdr.seq ^ sp->hdr.callNumber;
 	check &= 0xffff;
 	if (check != 0) {
 		*_abort_code = RXKADSEALEDINCON;
@@ -453,7 +452,7 @@ static int rxkad_verify_packet_encrypt(const struct rxrpc_call *call,
 	data_size = buf & 0xffff;
 
 	check = buf >> 16;
-	check ^= ntohl(sp->hdr.seq ^ sp->hdr.callNumber);
+	check ^= sp->hdr.seq ^ sp->hdr.callNumber;
 	check &= 0xffff;
 	if (check != 0) {
 		*_abort_code = RXKADSEALEDINCON;
@@ -494,16 +493,14 @@ static int rxkad_verify_packet(const struct rxrpc_call *call,
 	struct {
 		__be32 x[2];
 	} tmpbuf __attribute__((aligned(8))); /* must all be in same page */
-	__be32 x;
-	__be16 cksum;
-	u32 y;
+	u16 cksum;
+	u32 x, y;
 	int ret;
 
 	sp = rxrpc_skb(skb);
 
 	_enter("{%d{%x}},{#%u}",
-	       call->debug_id, key_serial(call->conn->key),
-	       ntohl(sp->hdr.seq));
+	       call->debug_id, key_serial(call->conn->key), sp->hdr.seq);
 
 	if (!call->conn->cipher)
 		return 0;
@@ -521,21 +518,20 @@ static int rxkad_verify_packet(const struct rxrpc_call *call,
 	desc.flags = 0;
 
 	/* validate the security checksum */
-	x = htonl(call->channel << (32 - RXRPC_CIDSHIFT));
-	x |= sp->hdr.seq & cpu_to_be32(0x3fffffff);
-	tmpbuf.x[0] = call->call_id;
-	tmpbuf.x[1] = x;
+	x = call->channel << (32 - RXRPC_CIDSHIFT);
+	x |= sp->hdr.seq & 0x3fffffff;
+	tmpbuf.x[0] = htonl(call->call_id);
+	tmpbuf.x[1] = htonl(x);
 
 	sg_init_one(&sg[0], &tmpbuf, sizeof(tmpbuf));
 	sg_init_one(&sg[1], &tmpbuf, sizeof(tmpbuf));
 	crypto_blkcipher_encrypt_iv(&desc, &sg[0], &sg[1], sizeof(tmpbuf));
 
 	y = ntohl(tmpbuf.x[1]);
-	y = (y >> 16) & 0xffff;
-	if (y == 0)
-		y = 1; /* zero checksums are not permitted */
+	cksum = (y >> 16) & 0xffff;
+	if (cksum == 0)
+		cksum = 1; /* zero checksums are not permitted */
 
-	cksum = htons(y);
 	if (sp->hdr.cksum != cksum) {
 		*_abort_code = RXKADSEALEDINCON;
 		_leave(" = -EPROTO [csum failed]");
@@ -567,10 +563,11 @@ static int rxkad_verify_packet(const struct rxrpc_call *call,
 static int rxkad_issue_challenge(struct rxrpc_connection *conn)
 {
 	struct rxkad_challenge challenge;
-	struct rxrpc_header hdr;
+	struct rxrpc_wire_header whdr;
 	struct msghdr msg;
 	struct kvec iov[2];
 	size_t len;
+	u32 serial;
 	int ret;
 
 	_enter("{%d,%x}", conn->debug_id, key_serial(conn->key));
@@ -592,26 +589,27 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
 	msg.msg_controllen = 0;
 	msg.msg_flags	= 0;
 
-	hdr.epoch	= conn->epoch;
-	hdr.cid		= conn->cid;
-	hdr.callNumber	= 0;
-	hdr.seq		= 0;
-	hdr.type	= RXRPC_PACKET_TYPE_CHALLENGE;
-	hdr.flags	= conn->out_clientflag;
-	hdr.userStatus	= 0;
-	hdr.securityIndex = conn->security_ix;
-	hdr._rsvd	= 0;
-	hdr.serviceId	= conn->service_id;
-
-	iov[0].iov_base	= &hdr;
-	iov[0].iov_len	= sizeof(hdr);
+	whdr.epoch	= htonl(conn->epoch);
+	whdr.cid	= htonl(conn->cid);
+	whdr.callNumber	= 0;
+	whdr.seq	= 0;
+	whdr.type	= RXRPC_PACKET_TYPE_CHALLENGE;
+	whdr.flags	= conn->out_clientflag;
+	whdr.userStatus	= 0;
+	whdr.securityIndex = conn->security_ix;
+	whdr._rsvd	= 0;
+	whdr.serviceId	= htons(conn->service_id);
+
+	iov[0].iov_base	= &whdr;
+	iov[0].iov_len	= sizeof(whdr);
 	iov[1].iov_base	= &challenge;
 	iov[1].iov_len	= sizeof(challenge);
 
 	len = iov[0].iov_len + iov[1].iov_len;
 
-	hdr.serial = htonl(atomic_inc_return(&conn->serial));
-	_proto("Tx CHALLENGE %%%u", ntohl(hdr.serial));
+	serial = atomic_inc_return(&conn->serial);
+	whdr.serial = htonl(serial);
+	_proto("Tx CHALLENGE %%%u", serial);
 
 	ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len);
 	if (ret < 0) {
@@ -627,13 +625,15 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
  * send a Kerberos security response
  */
 static int rxkad_send_response(struct rxrpc_connection *conn,
-			       struct rxrpc_header *hdr,
+			       struct rxrpc_host_header *hdr,
 			       struct rxkad_response *resp,
 			       const struct rxkad_key *s2)
 {
+	struct rxrpc_wire_header whdr;
 	struct msghdr msg;
 	struct kvec iov[3];
 	size_t len;
+	u32 serial;
 	int ret;
 
 	_enter("");
@@ -644,24 +644,26 @@ static int rxkad_send_response(struct rxrpc_connection *conn,
 	msg.msg_controllen = 0;
 	msg.msg_flags	= 0;
 
-	hdr->epoch	= conn->epoch;
-	hdr->seq	= 0;
-	hdr->type	= RXRPC_PACKET_TYPE_RESPONSE;
-	hdr->flags	= conn->out_clientflag;
-	hdr->userStatus	= 0;
-	hdr->_rsvd	= 0;
+	memset(&whdr, 0, sizeof(whdr));
+	whdr.epoch	= htonl(hdr->epoch);
+	whdr.cid	= htonl(hdr->cid);
+	whdr.type	= RXRPC_PACKET_TYPE_RESPONSE;
+	whdr.flags	= conn->out_clientflag;
+	whdr.securityIndex = hdr->securityIndex;
+	whdr.serviceId	= htons(hdr->serviceId);
 
-	iov[0].iov_base	= hdr;
-	iov[0].iov_len	= sizeof(*hdr);
+	iov[0].iov_base	= &whdr;
+	iov[0].iov_len	= sizeof(whdr);
 	iov[1].iov_base	= resp;
 	iov[1].iov_len	= sizeof(*resp);
-	iov[2].iov_base	= (void *) s2->ticket;
+	iov[2].iov_base	= (void *)s2->ticket;
 	iov[2].iov_len	= s2->ticket_len;
 
 	len = iov[0].iov_len + iov[1].iov_len + iov[2].iov_len;
 
-	hdr->serial = htonl(atomic_inc_return(&conn->serial));
-	_proto("Tx RESPONSE %%%u", ntohl(hdr->serial));
+	serial = atomic_inc_return(&conn->serial);
+	whdr.serial = htonl(serial);
+	_proto("Tx RESPONSE %%%u", serial);
 
 	ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 3, len);
 	if (ret < 0) {
@@ -770,7 +772,7 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn,
 	min_level = ntohl(challenge.min_level);
 
 	_proto("Rx CHALLENGE %%%u { v=%u n=%u ml=%u }",
-	       ntohl(sp->hdr.serial), version, nonce, min_level);
+	       sp->hdr.serial, version, nonce, min_level);
 
 	abort_code = RXKADINCONSISTENCY;
 	if (version != RXKAD_VERSION)
@@ -786,17 +788,17 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn,
 	memset(&resp, 0, sizeof(resp));
 
 	resp.version = RXKAD_VERSION;
-	resp.encrypted.epoch = conn->epoch;
-	resp.encrypted.cid = conn->cid;
+	resp.encrypted.epoch = htonl(conn->epoch);
+	resp.encrypted.cid = htonl(conn->cid);
 	resp.encrypted.securityIndex = htonl(conn->security_ix);
 	resp.encrypted.call_id[0] =
-		(conn->channels[0] ? conn->channels[0]->call_id : 0);
+		htonl(conn->channels[0] ? conn->channels[0]->call_id : 0);
 	resp.encrypted.call_id[1] =
-		(conn->channels[1] ? conn->channels[1]->call_id : 0);
+		htonl(conn->channels[1] ? conn->channels[1]->call_id : 0);
 	resp.encrypted.call_id[2] =
-		(conn->channels[2] ? conn->channels[2]->call_id : 0);
+		htonl(conn->channels[2] ? conn->channels[2]->call_id : 0);
 	resp.encrypted.call_id[3] =
-		(conn->channels[3] ? conn->channels[3]->call_id : 0);
+		htonl(conn->channels[3] ? conn->channels[3]->call_id : 0);
 	resp.encrypted.inc_nonce = htonl(nonce + 1);
 	resp.encrypted.level = htonl(conn->security_level);
 	resp.kvno = htonl(token->kad->kvno);
@@ -1022,7 +1024,7 @@ static int rxkad_verify_response(struct rxrpc_connection *conn,
 	kvno = ntohl(response.kvno);
 	sp = rxrpc_skb(skb);
 	_proto("Rx RESPONSE %%%u { v=%u kv=%u tl=%u }",
-	       ntohl(sp->hdr.serial), version, kvno, ticket_len);
+	       sp->hdr.serial, version, kvno, ticket_len);
 
 	abort_code = RXKADINCONSISTENCY;
 	if (version != RXKAD_VERSION)
@@ -1058,9 +1060,9 @@ static int rxkad_verify_response(struct rxrpc_connection *conn,
 	rxkad_decrypt_response(conn, &response, &session_key);
 
 	abort_code = RXKADSEALEDINCON;
-	if (response.encrypted.epoch != conn->epoch)
+	if (ntohl(response.encrypted.epoch) != conn->epoch)
 		goto protocol_error_free;
-	if (response.encrypted.cid != conn->cid)
+	if (ntohl(response.encrypted.cid) != conn->cid)
 		goto protocol_error_free;
 	if (ntohl(response.encrypted.securityIndex) != conn->security_ix)
 		goto protocol_error_free;
@@ -1077,7 +1079,7 @@ static int rxkad_verify_response(struct rxrpc_connection *conn,
 		goto protocol_error_free;
 
 	abort_code = RXKADOUTOFSEQUENCE;
-	if (response.encrypted.inc_nonce != htonl(conn->security_nonce + 1))
+	if (ntohl(response.encrypted.inc_nonce) != conn->security_nonce + 1)
 		goto protocol_error_free;
 
 	abort_code = RXKADLEVELFAIL;
-- 
cgit v0.10.2


From e33b3d97bc443aa330f086ca90a209a0d35dfa86 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:54:27 +0000
Subject: rxrpc: The protocol family should be set to PF_RXRPC not PF_UNIX

Fix the protocol family set in the proto_ops for rxrpc to be PF_RXRPC not
PF_UNIX.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 7bb5cca..5e810b8 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -748,7 +748,7 @@ static int rxrpc_release(struct socket *sock)
  * RxRPC network protocol
  */
 static const struct proto_ops rxrpc_rpc_ops = {
-	.family		= PF_UNIX,
+	.family		= PF_RXRPC,
 	.owner		= THIS_MODULE,
 	.release	= rxrpc_release,
 	.bind		= rxrpc_bind,
-- 
cgit v0.10.2


From ee6fe085a992d40d2b5d3e7e2b9e0eec1cdacaac Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:55:32 +0000
Subject: rxrpc: Fix defined range for /proc/sys/net/rxrpc/rx_mtu

The upper bound of the defined range for rx_mtu is being set in the same
member as the lower bound (extra1) rather than the correct place (extra2).
I'm not entirely sure why this compiles.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c
index 50a98a9..093547a 100644
--- a/net/rxrpc/sysctl.c
+++ b/net/rxrpc/sysctl.c
@@ -115,7 +115,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= (void *)&one,
-		.extra1		= (void *)&n_65535,
+		.extra2		= (void *)&n_65535,
 	},
 	{
 		.procname	= "rx_jumbo_max",
-- 
cgit v0.10.2


From 351c1e648623b742fe1687636117306adc8b561c Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:56:06 +0000
Subject: rxrpc: Be more selective about the types of received packets we
 accept

Currently, received RxRPC packets outside the range 1-13 are rejected.
There are, however, holes in the range that should also be rejected - plus
at least one type we don't yet support - so reject these also.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/include/rxrpc/packet.h b/include/rxrpc/packet.h
index de1e679..9ebab3a 100644
--- a/include/rxrpc/packet.h
+++ b/include/rxrpc/packet.h
@@ -70,6 +70,17 @@ struct rxrpc_wire_header {
 
 extern const char *rxrpc_pkts[];
 
+#define RXRPC_SUPPORTED_PACKET_TYPES (			\
+		(1 << RXRPC_PACKET_TYPE_DATA) |		\
+		(1 << RXRPC_PACKET_TYPE_ACK) |		\
+		(1 << RXRPC_PACKET_TYPE_BUSY) |		\
+		(1 << RXRPC_PACKET_TYPE_ABORT) |	\
+		(1 << RXRPC_PACKET_TYPE_ACKALL) |	\
+		(1 << RXRPC_PACKET_TYPE_CHALLENGE) |	\
+		(1 << RXRPC_PACKET_TYPE_RESPONSE) |	\
+		/*(1 << RXRPC_PACKET_TYPE_DEBUG) | */	\
+		(1 << RXRPC_PACKET_TYPE_VERSION))
+
 /*****************************************************************************/
 /*
  * jumbo packet secondary header
diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c
index e6396a8..63ed75c 100644
--- a/net/rxrpc/ar-input.c
+++ b/net/rxrpc/ar-input.c
@@ -728,7 +728,8 @@ void rxrpc_data_ready(struct sock *sk)
 	     sp->hdr.flags & RXRPC_CLIENT_INITIATED ? "ToServer" : "ToClient",
 	     sp->hdr.epoch, sp->hdr.cid, sp->hdr.callNumber);
 
-	if (sp->hdr.type == 0 || sp->hdr.type >= RXRPC_N_PACKET_TYPES) {
+	if (sp->hdr.type >= RXRPC_N_PACKET_TYPES ||
+	    !((RXRPC_SUPPORTED_PACKET_TYPES >> sp->hdr.type) & 1)) {
 		_proto("Rx Bad Packet Type %u", sp->hdr.type);
 		goto bad_message;
 	}
-- 
cgit v0.10.2


From b4f1342f915201ee15ef6890857b5469879ee402 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:56:19 +0000
Subject: rxrpc: Adjust some whitespace and comments

Remove some excess whitespace, insert some missing spaces and adjust a
couple of comments.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 5e810b8..b46f774 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -121,7 +121,7 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx,
  */
 static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len)
 {
-	struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) saddr;
+	struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *)saddr;
 	struct sock *sk = sock->sk;
 	struct rxrpc_local *local;
 	struct rxrpc_sock *rx = rxrpc_sk(sk), *prx;
@@ -142,7 +142,7 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len)
 
 	memcpy(&rx->srx, srx, sizeof(rx->srx));
 
-	/* find a local transport endpoint if we don't have one already */
+	/* Find or create a local transport endpoint to use */
 	local = rxrpc_lookup_local(&rx->srx);
 	if (IS_ERR(local)) {
 		ret = PTR_ERR(local);
@@ -297,7 +297,6 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
 
 	if (!srx)
 		srx = &rx->srx;
-
 	if (!key)
 		key = rx->key;
 	if (key && !key->payload.data[0])
@@ -319,7 +318,6 @@ out_notrans:
 	_leave(" = %p", call);
 	return call;
 }
-
 EXPORT_SYMBOL(rxrpc_kernel_begin_call);
 
 /**
@@ -335,7 +333,6 @@ void rxrpc_kernel_end_call(struct rxrpc_call *call)
 	rxrpc_remove_user_ID(call->socket, call);
 	rxrpc_put_call(call);
 }
-
 EXPORT_SYMBOL(rxrpc_kernel_end_call);
 
 /**
@@ -616,7 +613,7 @@ static int rxrpc_create(struct net *net, struct socket *sock, int protocol,
 	if (!net_eq(net, &init_net))
 		return -EAFNOSUPPORT;
 
-	/* we support transport protocol UDP only */
+	/* we support transport protocol UDP/UDP6 only */
 	if (protocol != PF_INET)
 		return -EPROTONOSUPPORT;
 
diff --git a/net/rxrpc/ar-accept.c b/net/rxrpc/ar-accept.c
index 73c9054..277731a 100644
--- a/net/rxrpc/ar-accept.c
+++ b/net/rxrpc/ar-accept.c
@@ -415,7 +415,7 @@ out:
 }
 
 /*
- * handle rejectance of a call by userspace
+ * Handle rejection of a call by userspace
  * - reject the call at the front of the queue
  */
 int rxrpc_reject_call(struct rxrpc_sock *rx)
@@ -495,7 +495,6 @@ struct rxrpc_call *rxrpc_kernel_accept_call(struct socket *sock,
 	_leave(" = %p", call);
 	return call;
 }
-
 EXPORT_SYMBOL(rxrpc_kernel_accept_call);
 
 /**
@@ -514,5 +513,4 @@ int rxrpc_kernel_reject_call(struct socket *sock)
 	_leave(" = %d", ret);
 	return ret;
 }
-
 EXPORT_SYMBOL(rxrpc_kernel_reject_call);
diff --git a/net/rxrpc/ar-error.c b/net/rxrpc/ar-error.c
index 3c47441..5948191 100644
--- a/net/rxrpc/ar-error.c
+++ b/net/rxrpc/ar-error.c
@@ -115,7 +115,6 @@ void rxrpc_UDP_error_report(struct sock *sk)
 	/* pass the transport ref to error_handler to release */
 	skb_queue_tail(&trans->error_queue, skb);
 	rxrpc_queue_work(&trans->error_handler);
-
 	_leave("");
 }
 
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 06bf5ab..8b495ae 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -16,7 +16,7 @@
 	BUG_ON(atomic_read((X)) >> (sizeof(atomic_t) - 2) == \
 	       (POISON_FREE << 8 | POISON_FREE))
 #else
-#define CHECK_SLAB_OKAY(X) do {} while(0)
+#define CHECK_SLAB_OKAY(X) do {} while (0)
 #endif
 
 #define FCRYPT_BSIZE 8
@@ -726,7 +726,7 @@ do {								\
 		printk(KERN_ERR "RxRPC: Assertion failed\n");	\
 		BUG();						\
 	}							\
-} while(0)
+} while (0)
 
 #define ASSERTCMP(X, OP, Y)						\
 do {									\
@@ -739,7 +739,7 @@ do {									\
 		       (unsigned long)(X), (unsigned long)(Y));		\
 		BUG();							\
 	}								\
-} while(0)
+} while (0)
 
 #define ASSERTIF(C, X)						\
 do {								\
@@ -748,7 +748,7 @@ do {								\
 		printk(KERN_ERR "RxRPC: Assertion failed\n");	\
 		BUG();						\
 	}							\
-} while(0)
+} while (0)
 
 #define ASSERTIFCMP(C, X, OP, Y)					\
 do {									\
@@ -761,25 +761,25 @@ do {									\
 		       (unsigned long)(X), (unsigned long)(Y));		\
 		BUG();							\
 	}								\
-} while(0)
+} while (0)
 
 #else
 
 #define ASSERT(X)				\
 do {						\
-} while(0)
+} while (0)
 
 #define ASSERTCMP(X, OP, Y)			\
 do {						\
-} while(0)
+} while (0)
 
 #define ASSERTIF(C, X)				\
 do {						\
-} while(0)
+} while (0)
 
 #define ASSERTIFCMP(C, X, OP, Y)		\
 do {						\
-} while(0)
+} while (0)
 
 #endif /* __KDEBUGALL */
 
@@ -836,9 +836,9 @@ do {							\
 	CHECK_SLAB_OKAY(&(CALL)->usage);		\
 	if (atomic_inc_return(&(CALL)->usage) == 1)	\
 		BUG();					\
-} while(0)
+} while (0)
 
 #define rxrpc_put_call(CALL)				\
 do {							\
 	__rxrpc_put_call(CALL);				\
-} while(0)
+} while (0)
diff --git a/net/rxrpc/ar-security.c b/net/rxrpc/ar-security.c
index e2f4c49..ceff639 100644
--- a/net/rxrpc/ar-security.c
+++ b/net/rxrpc/ar-security.c
@@ -167,7 +167,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn)
 	struct rxrpc_sock *rx;
 	struct key *key;
 	key_ref_t kref;
-	char kdesc[5+1+3+1];
+	char kdesc[5 + 1 + 3 + 1];
 
 	_enter("");
 
diff --git a/net/rxrpc/ar-skbuff.c b/net/rxrpc/ar-skbuff.c
index ae9f93f..62a2674 100644
--- a/net/rxrpc/ar-skbuff.c
+++ b/net/rxrpc/ar-skbuff.c
@@ -133,5 +133,4 @@ void rxrpc_kernel_free_skb(struct sk_buff *skb)
 {
 	rxrpc_free_skb(skb);
 }
-
 EXPORT_SYMBOL(rxrpc_kernel_free_skb);
diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index 1604802..0f27524 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -195,9 +195,9 @@ static int rxkad_secure_packet_auth(const struct rxrpc_call *call,
  * wholly encrypt a packet (level 2 security)
  */
 static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call,
-					struct sk_buff *skb,
-					u32 data_size,
-					void *sechdr)
+				       struct sk_buff *skb,
+				       u32 data_size,
+				       void *sechdr)
 {
 	const struct rxrpc_key_token *token;
 	struct rxkad_level2_hdr rxkhdr
@@ -251,9 +251,9 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call,
  * checksum an RxRPC packet header
  */
 static int rxkad_secure_packet(const struct rxrpc_call *call,
-				struct sk_buff *skb,
-				size_t data_size,
-				void *sechdr)
+			       struct sk_buff *skb,
+			       size_t data_size,
+			       void *sechdr)
 {
 	struct rxrpc_skb_priv *sp;
 	struct blkcipher_desc desc;
-- 
cgit v0.10.2


From ee72b9fddb41a2087b4a7aa74791ef1700f3f6cc Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:58:06 +0000
Subject: rxrpc: Use ACCESS_ONCE() when accessing circular buffer pointers

Use ACCESS_ONCE() when accessing the other-end pointer into a circular
buffer as it's possible the other-end pointer might change whilst we're
doing this, and if we access it twice, we might get some weird things
happening.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 353f5c9..14c8df6 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -401,7 +401,8 @@ static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx,
 	int ret;
 
 	_enter(",{%d},%ld",
-	       CIRC_SPACE(call->acks_head, call->acks_tail, call->acks_winsz),
+	       CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail),
+			  call->acks_winsz),
 	       *timeo);
 
 	add_wait_queue(&call->tx_waitq, &myself);
@@ -409,7 +410,7 @@ static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx,
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
 		ret = 0;
-		if (CIRC_SPACE(call->acks_head, call->acks_tail,
+		if (CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail),
 			       call->acks_winsz) > 0)
 			break;
 		if (signal_pending(current)) {
@@ -570,7 +571,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 
 			_debug("alloc");
 
-			if (CIRC_SPACE(call->acks_head, call->acks_tail,
+			if (CIRC_SPACE(call->acks_head,
+				       ACCESS_ONCE(call->acks_tail),
 				       call->acks_winsz) <= 0) {
 				ret = -EAGAIN;
 				if (msg->msg_flags & MSG_DONTWAIT)
@@ -686,7 +688,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
 			sp->hdr.flags = conn->out_clientflag;
 			if (msg_data_left(msg) == 0 && !more)
 				sp->hdr.flags |= RXRPC_LAST_PACKET;
-			else if (CIRC_SPACE(call->acks_head, call->acks_tail,
+			else if (CIRC_SPACE(call->acks_head,
+					    ACCESS_ONCE(call->acks_tail),
 					    call->acks_winsz) > 1)
 				sp->hdr.flags |= RXRPC_MORE_PACKETS;
 			if (more && seq & 1)
-- 
cgit v0.10.2


From 098a20991d15b76a7e6835b95bd58269e10e3a60 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:59:00 +0000
Subject: rxrpc: rxkad: The version number in the response should be net byte
 order

The version number rxkad places in the response should be network byte
order.

Whilst we're at it, rearrange the code to be more readable.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index 0f27524..ca746f3 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -787,10 +787,15 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn,
 	/* build the response packet */
 	memset(&resp, 0, sizeof(resp));
 
-	resp.version = RXKAD_VERSION;
-	resp.encrypted.epoch = htonl(conn->epoch);
-	resp.encrypted.cid = htonl(conn->cid);
-	resp.encrypted.securityIndex = htonl(conn->security_ix);
+	resp.version			= htonl(RXKAD_VERSION);
+	resp.encrypted.epoch		= htonl(conn->epoch);
+	resp.encrypted.cid		= htonl(conn->cid);
+	resp.encrypted.securityIndex	= htonl(conn->security_ix);
+	resp.encrypted.inc_nonce	= htonl(nonce + 1);
+	resp.encrypted.level		= htonl(conn->security_level);
+	resp.kvno			= htonl(token->kad->kvno);
+	resp.ticket_len			= htonl(token->kad->ticket_len);
+
 	resp.encrypted.call_id[0] =
 		htonl(conn->channels[0] ? conn->channels[0]->call_id : 0);
 	resp.encrypted.call_id[1] =
@@ -799,10 +804,6 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn,
 		htonl(conn->channels[2] ? conn->channels[2]->call_id : 0);
 	resp.encrypted.call_id[3] =
 		htonl(conn->channels[3] ? conn->channels[3]->call_id : 0);
-	resp.encrypted.inc_nonce = htonl(nonce + 1);
-	resp.encrypted.level = htonl(conn->security_level);
-	resp.kvno = htonl(token->kad->kvno);
-	resp.ticket_len = htonl(token->kad->ticket_len);
 
 	/* calculate the response checksum and then do the encryption */
 	rxkad_calc_response_checksum(&resp);
-- 
cgit v0.10.2


From 2b15ef15bca689d60be27f6ffdaa2c92a59ae6e7 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:59:13 +0000
Subject: rxrpc: rxkad: Casts are needed when comparing be32 values

Forced casts are needed to avoid sparse warning when directly comparing
be32 values.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index ca746f3..3106a0c 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -142,7 +142,7 @@ static void rxkad_prime_packet_security(struct rxrpc_connection *conn)
 	crypto_blkcipher_encrypt_iv(&desc, &sg[0], &sg[1], sizeof(tmpbuf));
 
 	memcpy(&conn->csum_iv, &tmpbuf.x[2], sizeof(conn->csum_iv));
-	ASSERTCMP(conn->csum_iv.n[0], ==, tmpbuf.x[2]);
+	ASSERTCMP((u32 __force)conn->csum_iv.n[0], ==, (u32 __force)tmpbuf.x[2]);
 
 	_leave("");
 }
-- 
cgit v0.10.2


From ab802ee0ab5f4e1e7ee17632627726fbd7881e6c Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 15:59:49 +0000
Subject: rxrpc: Clear the unused part of a sockaddr_rxrpc for memcmp() use

Clear the unused part of a sockaddr_rxrpc structs so that memcmp() can be
used to compare them.

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index b46f774..a765017 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -81,6 +81,8 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx,
 				  struct sockaddr_rxrpc *srx,
 				  int len)
 {
+	unsigned tail;
+
 	if (len < sizeof(struct sockaddr_rxrpc))
 		return -EINVAL;
 
@@ -103,9 +105,7 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx,
 		_debug("INET: %x @ %pI4",
 		       ntohs(srx->transport.sin.sin_port),
 		       &srx->transport.sin.sin_addr);
-		if (srx->transport_len > 8)
-			memset((void *)&srx->transport + 8, 0,
-			       srx->transport_len - 8);
+		tail = offsetof(struct sockaddr_rxrpc, transport.sin.__pad);
 		break;
 
 	case AF_INET6:
@@ -113,6 +113,8 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx,
 		return -EAFNOSUPPORT;
 	}
 
+	if (tail < len)
+		memset((void *)srx + tail, 0, len - tail);
 	return 0;
 }
 
-- 
cgit v0.10.2


From a4373a489e87e2bf8794194bc275b6c05f1af2f7 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 4 Mar 2016 16:02:03 +0000
Subject: rxrpc: Don't try to map ICMP to error as the lower layer already did
 that

In the ICMP message processing code, don't try to map ICMP codes to UNIX
error codes as the caller (IPv4/IPv6) already did that for us (ee_errno).

Signed-off-by: David Howells <dhowells@redhat.com>

diff --git a/net/rxrpc/ar-error.c b/net/rxrpc/ar-error.c
index 5948191..3e82d6f 100644
--- a/net/rxrpc/ar-error.c
+++ b/net/rxrpc/ar-error.c
@@ -151,28 +151,18 @@ void rxrpc_UDP_error_handler(struct work_struct *work)
 			switch (ee->ee_code) {
 			case ICMP_NET_UNREACH:
 				_net("Rx Received ICMP Network Unreachable");
-				err = ENETUNREACH;
 				break;
 			case ICMP_HOST_UNREACH:
 				_net("Rx Received ICMP Host Unreachable");
-				err = EHOSTUNREACH;
 				break;
 			case ICMP_PORT_UNREACH:
 				_net("Rx Received ICMP Port Unreachable");
-				err = ECONNREFUSED;
-				break;
-			case ICMP_FRAG_NEEDED:
-				_net("Rx Received ICMP Fragmentation Needed (%d)",
-				     ee->ee_info);
-				err = 0; /* dealt with elsewhere */
 				break;
 			case ICMP_NET_UNKNOWN:
 				_net("Rx Received ICMP Unknown Network");
-				err = ENETUNREACH;
 				break;
 			case ICMP_HOST_UNKNOWN:
 				_net("Rx Received ICMP Unknown Host");
-				err = EHOSTUNREACH;
 				break;
 			default:
 				_net("Rx Received ICMP DestUnreach code=%u",
-- 
cgit v0.10.2


From 9b2f8f616a8d992243be31ff47d3f64031bf168a Mon Sep 17 00:00:00 2001
From: Simon Horman <simon.horman@netronome.com>
Date: Fri, 4 Mar 2016 15:58:07 +0900
Subject: nfp: remove Rolf Neugebauer as co-maintainer

Rolf is no longer in his previous role at Netronome and as far as I know no
longer working on the NFP driver. Thus it does not seem appropriate for him
to be a co-maintainer anymore.

Reviewed-by: Dinan Gunawardena <dinan.gunawardena@netronome.com>
Signed-off-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/MAINTAINERS b/MAINTAINERS
index f5d7726..be0b56b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7501,7 +7501,6 @@ F:	net/netrom/
 
 NETRONOME ETHERNET DRIVERS
 M:	Jakub Kicinski <jakub.kicinski@netronome.com>
-M:	Rolf Neugebauer <rolf.neugebauer@netronome.com>
 L:	oss-drivers@netronome.com
 S:	Maintained
 F:	drivers/net/ethernet/netronome/
-- 
cgit v0.10.2


From b5d3755a22e0cc4c369c0985aef0c52c2477c1e7 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Fri, 4 Mar 2016 11:52:16 +0100
Subject: uapi: define DIV_ROUND_UP for userland

DIV_ROUND_UP is defined in linux/kernel.h only for the kernel.
When ethtool.h is included by a userland app, we got the following error:

include/linux/ethtool.h:1218:8: error: variably modified 'queue_mask' at file scope
  __u32 queue_mask[DIV_ROUND_UP(MAX_NUM_QUEUE, 32)];
        ^

Let's add a common definition in uapi and use it everywhere.

Fixes: ac2c7ad0e5d6 ("net/ethtool: introduce a new ioctl for per queue setting")
CC: Kan Liang <kan.liang@intel.com>
Suggested-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index f31638c..ac19239 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -64,7 +64,7 @@
 #define round_down(x, y) ((x) & ~__round_mask(x, y))
 
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
-#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
+#define DIV_ROUND_UP __KERNEL_DIV_ROUND_UP
 #define DIV_ROUND_UP_ULL(ll,d) \
 	({ unsigned long long _tmp = (ll)+(d)-1; do_div(_tmp, d); _tmp; })
 
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 37fd6dc..9c22249 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -13,6 +13,7 @@
 #ifndef _UAPI_LINUX_ETHTOOL_H
 #define _UAPI_LINUX_ETHTOOL_H
 
+#include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/if_ether.h>
 
@@ -1215,7 +1216,7 @@ enum ethtool_sfeatures_retval_bits {
 struct ethtool_per_queue_op {
 	__u32	cmd;
 	__u32	sub_command;
-	__u32	queue_mask[DIV_ROUND_UP(MAX_NUM_QUEUE, 32)];
+	__u32	queue_mask[__KERNEL_DIV_ROUND_UP(MAX_NUM_QUEUE, 32)];
 	char	data[];
 };
 
diff --git a/include/uapi/linux/kernel.h b/include/uapi/linux/kernel.h
index 321e399..466073f 100644
--- a/include/uapi/linux/kernel.h
+++ b/include/uapi/linux/kernel.h
@@ -9,5 +9,6 @@
 #define __ALIGN_KERNEL(x, a)		__ALIGN_KERNEL_MASK(x, (typeof(x))(a) - 1)
 #define __ALIGN_KERNEL_MASK(x, mask)	(((x) + (mask)) & ~(mask))
 
+#define __KERNEL_DIV_ROUND_UP(n, d) (((n) + (d) - 1) / (d))
 
 #endif /* _UAPI_LINUX_KERNEL_H */
diff --git a/include/uapi/linux/mroute6.h b/include/uapi/linux/mroute6.h
index ce91215..5062fb5 100644
--- a/include/uapi/linux/mroute6.h
+++ b/include/uapi/linux/mroute6.h
@@ -1,6 +1,7 @@
 #ifndef _UAPI__LINUX_MROUTE6_H
 #define _UAPI__LINUX_MROUTE6_H
 
+#include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/sockios.h>
 
@@ -46,14 +47,8 @@ typedef unsigned short mifi_t;
 typedef	__u32		if_mask;
 #define NIFBITS (sizeof(if_mask) * 8)        /* bits per mask */
 
-#if !defined(__KERNEL__)
-#if !defined(DIV_ROUND_UP)
-#define	DIV_ROUND_UP(x,y)	(((x) + ((y) - 1)) / (y))
-#endif
-#endif
-
 typedef struct if_set {
-	if_mask ifs_bits[DIV_ROUND_UP(IF_SETSIZE, NIFBITS)];
+	if_mask ifs_bits[__KERNEL_DIV_ROUND_UP(IF_SETSIZE, NIFBITS)];
 } if_set;
 
 #define IF_SET(n, p)    ((p)->ifs_bits[(n)/NIFBITS] |= (1 << ((n) % NIFBITS)))
-- 
cgit v0.10.2


From 28466e0f2463a1084781c9312cd9148b2530eea7 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Fri, 4 Mar 2016 11:52:17 +0100
Subject: cxgb4i: don't redefine DIV_ROUND_UP

let's use the common definition to avoid the following warning during the
compilation:

drivers/scsi/cxgbi/cxgb4i/cxgb4i.c:161:0: warning: "DIV_ROUND_UP" redefined
 #define DIV_ROUND_UP(n, d) (((n) + (d) - 1) / (d))
 ^
In file included from include/linux/list.h:8:0,
                 from include/linux/module.h:9,
                 from drivers/scsi/cxgbi/cxgb4i/cxgb4i.c:16:
include/linux/kernel.h:67:0: note: this is the location of the previous definition
 #define DIV_ROUND_UP __KERNEL_DIV_ROUND_UP
 ^

Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
index 804806e..339f6b7 100644
--- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
+++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
@@ -13,6 +13,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ":%s: " fmt, __func__
 
+#include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <scsi/scsi_host.h>
@@ -158,7 +159,6 @@ static struct scsi_transport_template *cxgb4i_stt;
  * open/close/abort and data send/receive.
  */
 
-#define DIV_ROUND_UP(n, d)	(((n) + (d) - 1) / (d))
 #define RCV_BUFSIZ_MASK		0x3FFU
 #define MAX_IMM_TX_PKT_LEN	256
 
-- 
cgit v0.10.2


From 166cc7136752192bb6e1f2be2e56614762292a00 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Fri, 4 Mar 2016 11:52:18 +0100
Subject: drm/vmwgfx: remove userland definition of DIV_ROUND_UP

Let's use __KERNEL_DIV_ROUND_UP, which is defined in uapi/linux/kernel.h.

Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h b/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h
index 58704f0..531d220 100644
--- a/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h
+++ b/drivers/gpu/drm/vmwgfx/device_include/svga3d_surfacedefs.h
@@ -25,6 +25,8 @@
  *
  **************************************************************************/
 
+#include <linux/kernel.h>
+
 #ifdef __KERNEL__
 
 #include <drm/vmwgfx_drm.h>
@@ -36,7 +38,6 @@
 #define ARRAY_SIZE(_A) (sizeof(_A) / sizeof((_A)[0]))
 #endif /* ARRAY_SIZE */
 
-#define DIV_ROUND_UP(x, y)  (((x) + (y) - 1) / (y))
 #define max_t(type, x, y)  ((x) > (y) ? (x) : (y))
 #define surf_size_struct SVGA3dSize
 #define u32 uint32
@@ -987,12 +988,12 @@ svga3dsurface_get_size_in_blocks(const struct svga3d_surface_desc *desc,
 				 const surf_size_struct *pixel_size,
 				 surf_size_struct *block_size)
 {
-	block_size->width = DIV_ROUND_UP(pixel_size->width,
-					 desc->block_size.width);
-	block_size->height = DIV_ROUND_UP(pixel_size->height,
-					  desc->block_size.height);
-	block_size->depth = DIV_ROUND_UP(pixel_size->depth,
-					 desc->block_size.depth);
+	block_size->width = __KERNEL_DIV_ROUND_UP(pixel_size->width,
+						  desc->block_size.width);
+	block_size->height = __KERNEL_DIV_ROUND_UP(pixel_size->height,
+						   desc->block_size.height);
+	block_size->depth = __KERNEL_DIV_ROUND_UP(pixel_size->depth,
+						  desc->block_size.depth);
 }
 
 static inline bool
@@ -1100,8 +1101,9 @@ svga3dsurface_get_pixel_offset(SVGA3dSurfaceFormat format,
 	const struct svga3d_surface_desc *desc = svga3dsurface_get_desc(format);
 	const u32 bw = desc->block_size.width, bh = desc->block_size.height;
 	const u32 bd = desc->block_size.depth;
-	const u32 rowstride = DIV_ROUND_UP(width, bw) * desc->bytes_per_block;
-	const u32 imgstride = DIV_ROUND_UP(height, bh) * rowstride;
+	const u32 rowstride = __KERNEL_DIV_ROUND_UP(width, bw) *
+			      desc->bytes_per_block;
+	const u32 imgstride = __KERNEL_DIV_ROUND_UP(height, bh) * rowstride;
 	const u32 offset = (z / bd * imgstride +
 			    y / bh * rowstride +
 			    x / bw * desc->bytes_per_block);
-- 
cgit v0.10.2


From 14e2037902d65213842b4e40305ff54a64abbcb6 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Fri, 4 Mar 2016 11:52:19 +0100
Subject: ethtool.h: define INT_MAX for userland

INT_MAX needs limits.h in userland.
When ethtool.h is included by a userland app, we got the following error:

.../usr/include/linux/ethtool.h: In function 'ethtool_validate_speed':
.../usr/include/linux/ethtool.h:1471:18: error: 'INT_MAX' undeclared (first use in this function)
  return speed <= INT_MAX || speed == SPEED_UNKNOWN
                  ^

Fixes: e02564ee334a ("ethtool: make validate_speed accept all speeds between 0 and INT_MAX")
CC: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 9c22249..2835b07 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -17,6 +17,10 @@
 #include <linux/types.h>
 #include <linux/if_ether.h>
 
+#ifndef __KERNEL__
+#include <limits.h> /* for INT_MAX */
+#endif
+
 /* All structures exposed to userland should be defined such that they
  * have the same layout for 32-bit and 64-bit userland.
  */
-- 
cgit v0.10.2


From 6297b91c7fb530d3f2543a8e25d68ccf75bfb139 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 3 Mar 2016 01:16:54 +0000
Subject: vxlan: use reset to set header pointers

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add operation
in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 775ddb4..8ca243d 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1460,7 +1460,7 @@ static struct sk_buff *vxlan_na_create(struct sk_buff *request,
 	reply->dev = dev;
 	skb_reserve(reply, LL_RESERVED_SPACE(request->dev));
 	skb_push(reply, sizeof(struct ethhdr));
-	skb_set_mac_header(reply, 0);
+	skb_reset_mac_header(reply);
 
 	ns = (struct nd_msg *)skb_transport_header(request);
 
@@ -1480,7 +1480,7 @@ static struct sk_buff *vxlan_na_create(struct sk_buff *request,
 	reply->protocol = htons(ETH_P_IPV6);
 
 	skb_pull(reply, sizeof(struct ethhdr));
-	skb_set_network_header(reply, 0);
+	skb_reset_network_header(reply);
 	skb_put(reply, sizeof(struct ipv6hdr));
 
 	/* IPv6 header */
@@ -1495,7 +1495,7 @@ static struct sk_buff *vxlan_na_create(struct sk_buff *request,
 	pip6->saddr = *(struct in6_addr *)n->primary_key;
 
 	skb_pull(reply, sizeof(struct ipv6hdr));
-	skb_set_transport_header(reply, 0);
+	skb_reset_transport_header(reply);
 
 	na = (struct nd_msg *)skb_put(reply, sizeof(*na) + na_olen);
 
-- 
cgit v0.10.2


From 6b163a85478b27748bad219fe3ead7ac87217265 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 3 Mar 2016 01:16:55 +0000
Subject: mac80211_hwsim: use reset to set mac header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c
index a723a85..e85e073 100644
--- a/drivers/net/wireless/mac80211_hwsim.c
+++ b/drivers/net/wireless/mac80211_hwsim.c
@@ -844,7 +844,7 @@ static void mac80211_hwsim_monitor_rx(struct ieee80211_hw *hw,
 	hdr->rt_chbitmask = cpu_to_le16(flags);
 
 	skb->dev = hwsim_mon;
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->pkt_type = PACKET_OTHERHOST;
 	skb->protocol = htons(ETH_P_802_2);
@@ -887,7 +887,7 @@ static void mac80211_hwsim_monitor_ack(struct ieee80211_channel *chan,
 	memcpy(hdr11->addr1, addr, ETH_ALEN);
 
 	skb->dev = hwsim_mon;
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->pkt_type = PACKET_OTHERHOST;
 	skb->protocol = htons(ETH_P_802_2);
-- 
cgit v0.10.2


From d57a544d71092a81c2064ff242009a38e7b1a631 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 3 Mar 2016 01:16:56 +0000
Subject: mac80211: use reset to set header pointer

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c
index c6be0b4..5b6aec1 100644
--- a/net/mac80211/mesh_hwmp.c
+++ b/net/mac80211/mesh_hwmp.c
@@ -205,9 +205,9 @@ static void prepare_frame_for_deferred_tx(struct ieee80211_sub_if_data *sdata,
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 
-	skb_set_mac_header(skb, 0);
-	skb_set_network_header(skb, 0);
-	skb_set_transport_header(skb, 0);
+	skb_reset_mac_header(skb);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
 
 	/* Send all internal mgmt frames on VO. Accordingly set TID to 7. */
 	skb_set_queue_mapping(skb, IEEE80211_AC_VO);
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 9127957..5690e4c 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -3101,7 +3101,7 @@ static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx,
 	ieee80211_add_rx_radiotap_header(local, skb, rate, needed_headroom,
 					 false);
 
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->pkt_type = PACKET_OTHERHOST;
 	skb->protocol = htons(ETH_P_802_2);
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 6101deb..8b1b2ea 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -697,7 +697,7 @@ void ieee80211_tx_monitor(struct ieee80211_local *local, struct sk_buff *skb,
 					 rtap_len, shift);
 
 	/* XXX: is this sufficient for BPF? */
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	skb->pkt_type = PACKET_OTHERHOST;
 	skb->protocol = htons(ETH_P_802_2);
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 3a7475f..62ad532 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -2458,7 +2458,7 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata,
 	/* Update skb pointers to various headers since this modified frame
 	 * is going to go through Linux networking code that may potentially
 	 * need things like pointer to IP header. */
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, nh_pos);
 	skb_set_transport_header(skb, h_pos);
 
@@ -3973,9 +3973,9 @@ void __ieee80211_tx_skb_tid_band(struct ieee80211_sub_if_data *sdata,
 {
 	int ac = ieee802_1d_to_ac[tid & 7];
 
-	skb_set_mac_header(skb, 0);
-	skb_set_network_header(skb, 0);
-	skb_set_transport_header(skb, 0);
+	skb_reset_mac_header(skb);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
 
 	skb_set_queue_mapping(skb, ac);
 	skb->priority = tid;
-- 
cgit v0.10.2


From 3ef523aeeee88130e5de10adac719dbd6fa266e5 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 3 Mar 2016 01:16:57 +0000
Subject: wireless: use reset to set mac header

Since offset is zero, it's not necessary to use set function. Reset
function is straightforward, and will remove the unnecessary add
operation in set function.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/wireless/util.c b/net/wireless/util.c
index c7f6820..9f440a9 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -636,7 +636,7 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr,
 	/* Update skb pointers to various headers since this modified frame
 	 * is going to go through Linux networking code that may potentially
 	 * need things like pointer to IP header. */
-	skb_set_mac_header(skb, 0);
+	skb_reset_mac_header(skb);
 	skb_set_network_header(skb, nh_pos);
 	skb_set_transport_header(skb, h_pos);
 
-- 
cgit v0.10.2


From 4479a899f75cdbd0406d29701d459be5cadaf00b Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Mon, 8 Feb 2016 11:18:46 +0200
Subject: iwlwifi: pcie: forbid RTPM on device removal

The pci driver keeps any unbound device in active state and forbids
runtime PM.  When our driver gets probed, we take control of the
state.  When the device is released (i.e. during unbind or module
removal), we should return the state to what it was before.  To do so,
we need to forbid RTPM in the driver remove op.

Additionally, remove an unnecessary pm_runtime_disable() call, move
the initial ref_count setting to a better place and add some comments
explaining what is going on.

Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index d33b6ba..05b9685 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -631,13 +631,31 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* if RTPM is in use, enable it in our device */
 	if (iwl_trans->runtime_pm_mode != IWL_PLAT_PM_MODE_DISABLED) {
+		/* We explicitly set the device to active here to
+		 * clear contingent errors.
+		 */
 		pm_runtime_set_active(&pdev->dev);
+
 		pm_runtime_set_autosuspend_delay(&pdev->dev,
 					 iwlwifi_mod_params.d0i3_entry_delay);
 		pm_runtime_use_autosuspend(&pdev->dev);
+
+		/* We are not supposed to call pm_runtime_allow() by
+		 * ourselves, but let userspace enable runtime PM via
+		 * sysfs.  However, since we don't enable this from
+		 * userspace yet, we need to allow/forbid() ourselves.
+		*/
 		pm_runtime_allow(&pdev->dev);
 	}
 
+	/* The PCI device starts with a reference taken and we are
+	 * supposed to release it here.  But to simplify the
+	 * interaction with the opmode, we don't do it now, but let
+	 * the opmode release it when it's ready.  To account for this
+	 * reference, we start with ref_count set to 1.
+	 */
+	trans_pcie->ref_count = 1;
+
 	return 0;
 
 out_free_drv:
@@ -652,7 +670,17 @@ static void iwl_pci_remove(struct pci_dev *pdev)
 	struct iwl_trans *trans = pci_get_drvdata(pdev);
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
+	/* if RTPM was in use, restore it to the state before probe */
+	if (trans->runtime_pm_mode != IWL_PLAT_PM_MODE_DISABLED) {
+		/* We should not call forbid here, but we do for now.
+		 * Check the comment to pm_runtime_allow() in
+		 * iwl_pci_probe().
+		 */
+		pm_runtime_forbid(trans->dev);
+	}
+
 	iwl_drv_stop(trans_pcie->drv);
+
 	iwl_trans_pcie_free(trans);
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index e67957d..eb39c7e 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1646,9 +1646,6 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 	trans->command_groups = trans_cfg->command_groups;
 	trans->command_groups_size = trans_cfg->command_groups_size;
 
-	/* init ref_count to 1 (should be cleared when ucode is loaded) */
-	trans_pcie->ref_count = 1;
-
 	/* Initialize NAPI here - it should be before registering to mac80211
 	 * in the opmode but after the HW struct is allocated.
 	 * As this function may be called again in some corner cases don't
@@ -1663,9 +1660,6 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
-	/* TODO: check if this is really needed */
-	pm_runtime_disable(trans->dev);
-
 	iwl_pcie_synchronize_irqs(trans);
 
 	iwl_pcie_tx_free(trans);
-- 
cgit v0.10.2


From d56daea43c5a5d72e67ac60d97734d5823b457d6 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 15 Feb 2016 19:30:49 +0200
Subject: iwlwifi: pcie: refactor RXBs reclaiming code

Change the code to move rxbs directly from the allocator's
list to the queue's free list. This makes the code more
readable, saves the interim array and the double loop over
the free RBs.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 489b07a..a310fd2 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -539,40 +539,46 @@ static void iwl_pcie_rx_allocator(struct iwl_trans *trans)
 }
 
 /*
- * iwl_pcie_rx_allocator_get - Returns the pre-allocated pages
+ * iwl_pcie_rx_allocator_get - returns the pre-allocated pages
 .*
 .* Called by queue when the queue posted allocation request and
  * has freed 8 RBDs in order to restock itself.
+ * This function directly moves the allocated RBs to the queue's ownership
+ * and updates the relevant counters.
  */
-static int iwl_pcie_rx_allocator_get(struct iwl_trans *trans,
-				     struct iwl_rx_mem_buffer
-				     *out[RX_CLAIM_REQ_ALLOC])
+static void iwl_pcie_rx_allocator_get(struct iwl_trans *trans,
+				      struct iwl_rxq *rxq)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
 	int i;
 
+	lockdep_assert_held(&rxq->lock);
+
 	/*
 	 * atomic_dec_if_positive returns req_ready - 1 for any scenario.
 	 * If req_ready is 0 atomic_dec_if_positive will return -1 and this
-	 * function will return -ENOMEM, as there are no ready requests.
+	 * function will return early, as there are no ready requests.
 	 * atomic_dec_if_positive will perofrm the *actual* decrement only if
 	 * req_ready > 0, i.e. - there are ready requests and the function
 	 * hands one request to the caller.
 	 */
 	if (atomic_dec_if_positive(&rba->req_ready) < 0)
-		return -ENOMEM;
+		return;
 
 	spin_lock(&rba->lock);
 	for (i = 0; i < RX_CLAIM_REQ_ALLOC; i++) {
 		/* Get next free Rx buffer, remove it from free list */
-		out[i] = list_first_entry(&rba->rbd_allocated,
-			       struct iwl_rx_mem_buffer, list);
-		list_del(&out[i]->list);
+		struct iwl_rx_mem_buffer *rxb =
+			list_first_entry(&rba->rbd_allocated,
+					 struct iwl_rx_mem_buffer, list);
+
+		list_move(&rxb->list, &rxq->rx_free);
 	}
 	spin_unlock(&rba->lock);
 
-	return 0;
+	rxq->used_count -= RX_CLAIM_REQ_ALLOC;
+	rxq->free_count += RX_CLAIM_REQ_ALLOC;
 }
 
 static void iwl_pcie_rx_allocator_work(struct work_struct *data)
@@ -1149,7 +1155,7 @@ static void iwl_pcie_rx_handle(struct iwl_trans *trans, int queue)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rxq *rxq = &trans_pcie->rxq[queue];
-	u32 r, i, j, count = 0;
+	u32 r, i, count = 0;
 	bool emergency = false;
 
 restart:
@@ -1193,39 +1199,24 @@ restart:
 
 		i = (i + 1) & (rxq->queue_size - 1);
 
-		/* If we have RX_CLAIM_REQ_ALLOC released rx buffers -
-		 * try to claim the pre-allocated buffers from the allocator */
-		if (rxq->used_count >= RX_CLAIM_REQ_ALLOC) {
+		/*
+		 * If we have RX_CLAIM_REQ_ALLOC released rx buffers -
+		 * try to claim the pre-allocated buffers from the allocator.
+		 * If not ready - will try to reclaim next time.
+		 * There is no need to reschedule work - allocator exits only
+		 * on success
+		 */
+		if (rxq->used_count >= RX_CLAIM_REQ_ALLOC)
+			iwl_pcie_rx_allocator_get(trans, rxq);
+
+		if (rxq->used_count % RX_CLAIM_REQ_ALLOC == 0 && !emergency) {
 			struct iwl_rb_allocator *rba = &trans_pcie->rba;
-			struct iwl_rx_mem_buffer *out[RX_CLAIM_REQ_ALLOC];
-
-			if (rxq->used_count % RX_CLAIM_REQ_ALLOC == 0 &&
-			    !emergency) {
-				/* Add the remaining 6 empty RBDs
-				* for allocator use
-				 */
-				spin_lock(&rba->lock);
-				list_splice_tail_init(&rxq->rx_used,
-						      &rba->rbd_empty);
-				spin_unlock(&rba->lock);
-			}
 
-			/* If not ready - continue, will try to reclaim later.
-			* No need to reschedule work - allocator exits only on
-			* success */
-			if (!iwl_pcie_rx_allocator_get(trans, out)) {
-				/* If success - then RX_CLAIM_REQ_ALLOC
-				 * buffers were retrieved and should be added
-				 * to free list */
-				rxq->used_count -= RX_CLAIM_REQ_ALLOC;
-				for (j = 0; j < RX_CLAIM_REQ_ALLOC; j++) {
-					list_add_tail(&out[j]->list,
-						      &rxq->rx_free);
-					rxq->free_count++;
-				}
-			}
-		}
-		if (emergency) {
+			/* Add the remaining empty RBDs for allocator use */
+			spin_lock(&rba->lock);
+			list_splice_tail_init(&rxq->rx_used, &rba->rbd_empty);
+			spin_unlock(&rba->lock);
+		} else if (emergency) {
 			count++;
 			if (count == 8) {
 				count = 0;
-- 
cgit v0.10.2


From e5f91d91ac2e09f93f58ab8f6813d12f2b3afa03 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Wed, 2 Mar 2016 15:17:28 +0200
Subject: iwlwifi: pcie: set RB chunk size back to 64

128 byte chunk size is supported only on PCIe and not
on IOSF. For now, change it back to 64 byte.

Reported-by: Oren Givon <oren.givon@intel.com>
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index a310fd2..509f790 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -801,11 +801,10 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans)
 
 	/*
 	 * Activate DMA snooping.
-	 * Set RX DMA chunk size to 128 bit
+	 * Set RX DMA chunk size to 64B
 	 * Default queue is 0
 	 */
 	iwl_write_prph(trans, RFH_GEN_CFG, RFH_GEN_CFG_RFH_DMA_SNOOP |
-		       RFH_GEN_CFG_RB_CHUNK_SIZE |
 		       (DEFAULT_RXQ_NUM << RFH_GEN_CFG_DEFAULT_RXQ_NUM_POS) |
 		       RFH_GEN_CFG_SERVICE_DMA_SNOOP);
 	/* Enable the relevant rx queues */
-- 
cgit v0.10.2


From 91f66a3c672065a3aca387ca7e0028504bb8f457 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 28 Feb 2016 10:15:08 +0200
Subject: iwlwifi: mvm: avoid panics with thermal device usage

Thermal zone device registration can fail, and in this case
we don't want to remove WiFi functionality. This is why the
thermal zone registration function is void, and the flows
continue even if the thermal zone device registration failed.
Same applies for the cooling device.

This means that we at least need to remember that the thermal
zone device didn't register properly and take the minimal
precautions to avoid panic'ing when we access it.

This was missing.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ab410b4..b869db9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1575,7 +1575,6 @@ void iwl_mvm_set_hw_ctkill_state(struct iwl_mvm *mvm, bool state);
 int iwl_mvm_get_temp(struct iwl_mvm *mvm, s32 *temp);
 void iwl_mvm_ct_kill_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 int iwl_mvm_send_temp_report_ths_cmd(struct iwl_mvm *mvm);
-int iwl_mvm_cooling_device_register(struct iwl_mvm *mvm);
 int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget);
 
 /* Location Aware Regulatory */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 999bcb8..0a02e98 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -211,10 +211,14 @@ void iwl_mvm_temp_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb)
 	 * the firmware and hence to take the mutex.
 	 * Avoid the deadlock by unlocking the mutex here.
 	 */
-	mutex_unlock(&mvm->mutex);
-	thermal_notify_framework(mvm->tz_device.tzone,
-				 mvm->tz_device.fw_trips_index[ths_crossed]);
-	mutex_lock(&mvm->mutex);
+	if (mvm->tz_device.tzone) {
+		struct iwl_mvm_thermal_device *tz_dev = &mvm->tz_device;
+
+		mutex_unlock(&mvm->mutex);
+		thermal_notify_framework(tz_dev->tzone,
+					 tz_dev->fw_trips_index[ths_crossed]);
+		mutex_lock(&mvm->mutex);
+	}
 #endif /* CONFIG_THERMAL */
 }
 
@@ -520,16 +524,20 @@ int iwl_mvm_send_temp_report_ths_cmd(struct iwl_mvm *mvm)
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (!mvm->tz_device.tzone)
+		return -EINVAL;
+
 	/* The driver holds array of temperature trips that are unsorted
 	 * and uncompressed, the FW should get it compressed and sorted
 	 */
 
 	/* compress temp_trips to cmd array, remove uninitialized values*/
-	for (i = 0; i < IWL_MAX_DTS_TRIPS; i++)
+	for (i = 0; i < IWL_MAX_DTS_TRIPS; i++) {
 		if (mvm->tz_device.temp_trips[i] != S16_MIN) {
 			cmd.thresholds[idx++] =
 				cpu_to_le16(mvm->tz_device.temp_trips[i]);
 		}
+	}
 	cmd.num_temps = cpu_to_le32(idx);
 
 	if (!idx)
@@ -696,6 +704,7 @@ static void iwl_mvm_thermal_zone_register(struct iwl_mvm *mvm)
 		IWL_DEBUG_TEMP(mvm,
 			       "Failed to register to thermal zone (err = %ld)\n",
 			       PTR_ERR(mvm->tz_device.tzone));
+		mvm->tz_device.tzone = NULL;
 		return;
 	}
 
@@ -750,6 +759,10 @@ int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget)
 		return ret;
 	}
 
+	/* can happen if the registration failed */
+	if (!mvm->cooling_dev.cdev)
+		return -EINVAL;
+
 	if (op == CTDP_CMD_OPERATION_START)
 		mvm->cooling_dev.cur_state = budget;
 
@@ -812,15 +825,12 @@ static struct thermal_cooling_device_ops tcooling_ops = {
 	.set_cur_state = iwl_mvm_tcool_set_cur_state,
 };
 
-int iwl_mvm_cooling_device_register(struct iwl_mvm *mvm)
+static void iwl_mvm_cooling_device_register(struct iwl_mvm *mvm)
 {
 	char name[] = "iwlwifi";
 
-	if (!iwl_mvm_is_ctdp_supported(mvm)) {
-		mvm->cooling_dev.cdev = NULL;
-
-		return 0;
-	}
+	if (!iwl_mvm_is_ctdp_supported(mvm))
+		return;
 
 	BUILD_BUG_ON(ARRAY_SIZE(name) >= THERMAL_NAME_LENGTH);
 
@@ -833,34 +843,29 @@ int iwl_mvm_cooling_device_register(struct iwl_mvm *mvm)
 		IWL_DEBUG_TEMP(mvm,
 			       "Failed to register to cooling device (err = %ld)\n",
 			       PTR_ERR(mvm->cooling_dev.cdev));
-		return PTR_ERR(mvm->cooling_dev.cdev);
+		mvm->cooling_dev.cdev = NULL;
+		return;
 	}
-
-	return 0;
 }
 
 static void iwl_mvm_thermal_zone_unregister(struct iwl_mvm *mvm)
 {
-	if (!iwl_mvm_is_tt_in_fw(mvm))
+	if (!iwl_mvm_is_tt_in_fw(mvm) || !mvm->tz_device.tzone)
 		return;
 
-	if (mvm->tz_device.tzone) {
-		IWL_DEBUG_TEMP(mvm, "Thermal zone device unregister\n");
-		thermal_zone_device_unregister(mvm->tz_device.tzone);
-		mvm->tz_device.tzone = NULL;
-	}
+	IWL_DEBUG_TEMP(mvm, "Thermal zone device unregister\n");
+	thermal_zone_device_unregister(mvm->tz_device.tzone);
+	mvm->tz_device.tzone = NULL;
 }
 
 static void iwl_mvm_cooling_device_unregister(struct iwl_mvm *mvm)
 {
-	if (!iwl_mvm_is_ctdp_supported(mvm))
+	if (!iwl_mvm_is_ctdp_supported(mvm) || !mvm->cooling_dev.cdev)
 		return;
 
-	if (mvm->cooling_dev.cdev) {
-		IWL_DEBUG_TEMP(mvm, "Cooling device unregister\n");
-		thermal_cooling_device_unregister(mvm->cooling_dev.cdev);
-		mvm->cooling_dev.cdev = NULL;
-	}
+	IWL_DEBUG_TEMP(mvm, "Cooling device unregister\n");
+	thermal_cooling_device_unregister(mvm->cooling_dev.cdev);
+	mvm->cooling_dev.cdev = NULL;
 }
 #endif /* CONFIG_THERMAL */
 
-- 
cgit v0.10.2


From 3f20efba41916ee17ce82f0fdd02581ada2872b2 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 27 Jan 2016 14:52:02 +0100
Subject: ipvs: handle ip_vs_fill_iph_skb_off failure

ip_vs_fill_iph_skb_off() may not find an IP header, and gcc has
determined that ip_vs_sip_fill_param() then incorrectly accesses
the protocol fields:

net/netfilter/ipvs/ip_vs_pe_sip.c: In function 'ip_vs_sip_fill_param':
net/netfilter/ipvs/ip_vs_pe_sip.c:76:5: error: 'iph.protocol' may be used uninitialized in this function [-Werror=maybe-uninitialized]
  if (iph.protocol != IPPROTO_UDP)
     ^
net/netfilter/ipvs/ip_vs_pe_sip.c:81:10: error: 'iph.len' may be used uninitialized in this function [-Werror=maybe-uninitialized]
  dataoff = iph.len + sizeof(struct udphdr);
          ^

This adds a check for the ip_vs_fill_iph_skb_off() return code
before looking at the ip header data returned from it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: b0e010c527de ("ipvs: replace ip_vs_fill_ip4hdr with ip_vs_fill_iph_skb_off")
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>

diff --git a/net/netfilter/ipvs/ip_vs_pe_sip.c b/net/netfilter/ipvs/ip_vs_pe_sip.c
index 1b8d594..c4e9ca0 100644
--- a/net/netfilter/ipvs/ip_vs_pe_sip.c
+++ b/net/netfilter/ipvs/ip_vs_pe_sip.c
@@ -70,10 +70,10 @@ ip_vs_sip_fill_param(struct ip_vs_conn_param *p, struct sk_buff *skb)
 	const char *dptr;
 	int retc;
 
-	ip_vs_fill_iph_skb(p->af, skb, false, &iph);
+	retc = ip_vs_fill_iph_skb(p->af, skb, false, &iph);
 
 	/* Only useful with UDP */
-	if (iph.protocol != IPPROTO_UDP)
+	if (!retc || iph.protocol != IPPROTO_UDP)
 		return -EINVAL;
 	/* todo: IPv6 fragments:
 	 *       I think this only should be done for the first fragment. /HS
-- 
cgit v0.10.2


From f719e3754ee2f7275437e61a6afd520181fdd43b Mon Sep 17 00:00:00 2001
From: Julian Anastasov <ja@ssi.bg>
Date: Sat, 5 Mar 2016 15:03:22 +0200
Subject: ipvs: drop first packet to redirect conntrack

Jiri Bohac is reporting for a problem where the attempt
to reschedule existing connection to another real server
needs proper redirect for the conntrack used by the IPVS
connection. For example, when IPVS connection is created
to NAT-ed real server we alter the reply direction of
conntrack. If we later decide to select different real
server we can not alter again the conntrack. And if we
expire the old connection, the new connection is left
without conntrack.

So, the only way to redirect both the IPVS connection and
the Netfilter's conntrack is to drop the SYN packet that
hits existing connection, to wait for the next jiffie
to expire the old connection and its conntrack and to rely
on client's retransmission to create new connection as
usually.

Jiri Bohac provided a fix that drops all SYNs on rescheduling,
I extended his patch to do such drops only for connections
that use conntrack. Here is the original report from Jiri Bohac:

Since commit dc7b3eb900aa ("ipvs: Fix reuse connection if real server
is dead"), new connections to dead servers are redistributed
immediately to new servers.  The old connection is expired using
ip_vs_conn_expire_now() which sets the connection timer to expire
immediately.

However, before the timer callback, ip_vs_conn_expire(), is run
to clean the connection's conntrack entry, the new redistributed
connection may already be established and its conntrack removed
instead.

Fix this by dropping the first packet of the new connection
instead, like we do when the destination server is not available.
The timer will have deleted the old conntrack entry long before
the first packet of the new connection is retransmitted.

Fixes: dc7b3eb900aa ("ipvs: Fix reuse connection if real server is dead")
Signed-off-by: Jiri Bohac <jbohac@suse.cz>
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 0816c87..a6cc576 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -1588,6 +1588,23 @@ static inline void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp)
 }
 #endif /* CONFIG_IP_VS_NFCT */
 
+/* Really using conntrack? */
+static inline bool ip_vs_conn_uses_conntrack(struct ip_vs_conn *cp,
+					     struct sk_buff *skb)
+{
+#ifdef CONFIG_IP_VS_NFCT
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct;
+
+	if (!(cp->flags & IP_VS_CONN_F_NFCT))
+		return false;
+	ct = nf_ct_get(skb, &ctinfo);
+	if (ct && !nf_ct_is_untracked(ct))
+		return true;
+#endif
+	return false;
+}
+
 static inline int
 ip_vs_dest_conn_overhead(struct ip_vs_dest *dest)
 {
diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index f57b4dc..4da5600 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -1757,15 +1757,34 @@ ip_vs_in(struct netns_ipvs *ipvs, unsigned int hooknum, struct sk_buff *skb, int
 	cp = pp->conn_in_get(ipvs, af, skb, &iph);
 
 	conn_reuse_mode = sysctl_conn_reuse_mode(ipvs);
-	if (conn_reuse_mode && !iph.fragoffs &&
-	    is_new_conn(skb, &iph) && cp &&
-	    ((unlikely(sysctl_expire_nodest_conn(ipvs)) && cp->dest &&
-	      unlikely(!atomic_read(&cp->dest->weight))) ||
-	     unlikely(is_new_conn_expected(cp, conn_reuse_mode)))) {
-		if (!atomic_read(&cp->n_control))
-			ip_vs_conn_expire_now(cp);
-		__ip_vs_conn_put(cp);
-		cp = NULL;
+	if (conn_reuse_mode && !iph.fragoffs && is_new_conn(skb, &iph) && cp) {
+		bool uses_ct = false, resched = false;
+
+		if (unlikely(sysctl_expire_nodest_conn(ipvs)) && cp->dest &&
+		    unlikely(!atomic_read(&cp->dest->weight))) {
+			resched = true;
+			uses_ct = ip_vs_conn_uses_conntrack(cp, skb);
+		} else if (is_new_conn_expected(cp, conn_reuse_mode)) {
+			uses_ct = ip_vs_conn_uses_conntrack(cp, skb);
+			if (!atomic_read(&cp->n_control)) {
+				resched = true;
+			} else {
+				/* Do not reschedule controlling connection
+				 * that uses conntrack while it is still
+				 * referenced by controlled connection(s).
+				 */
+				resched = !uses_ct;
+			}
+		}
+
+		if (resched) {
+			if (!atomic_read(&cp->n_control))
+				ip_vs_conn_expire_now(cp);
+			__ip_vs_conn_put(cp);
+			if (uses_ct)
+				return NF_DROP;
+			cp = NULL;
+		}
 	}
 
 	if (unlikely(!cp)) {
-- 
cgit v0.10.2


From f911b675a06d05211da52cf8267db80c81b6aee4 Mon Sep 17 00:00:00 2001
From: Julian Anastasov <ja@ssi.bg>
Date: Sat, 5 Mar 2016 15:03:23 +0200
Subject: ipvs: allow rescheduling after RST

"RFC 5961, 4.2. Mitigation" describes a mechanism to request
client to confirm with RST the restart of TCP connection
before resending its SYN. As result, IPVS can see SYNs for
existing connection in CLOSE state. Add check to allow
rescheduling in this state.

Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>

diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 4da5600..b9a4082 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -1089,6 +1089,7 @@ static inline bool is_new_conn_expected(const struct ip_vs_conn *cp,
 	switch (cp->protocol) {
 	case IPPROTO_TCP:
 		return (cp->state == IP_VS_TCP_S_TIME_WAIT) ||
+		       (cp->state == IP_VS_TCP_S_CLOSE) ||
 			((conn_reuse_mode & 2) &&
 			 (cp->state == IP_VS_TCP_S_FIN_WAIT) &&
 			 (cp->flags & IP_VS_CONN_F_NOOUTPUT));
-- 
cgit v0.10.2


From 7617a24f83b5d67f4dab1844956be1cebc44aec8 Mon Sep 17 00:00:00 2001
From: Marco Angaroni <marcoangaroni@gmail.com>
Date: Sat, 5 Mar 2016 12:10:02 +0100
Subject: ipvs: correct initial offset of Call-ID header search in SIP
 persistence engine

The IPVS SIP persistence engine is not able to parse the SIP header
"Call-ID" when such header is inserted in the first positions of
the SIP message.

When IPVS is configured with "--pe sip" option, like for example:
ipvsadm -A -u 1.2.3.4:5060 -s rr --pe sip -p 120 -o
some particular messages (see below for details) do not create entries
in the connection template table, which can be listed with:
ipvsadm -Lcn --persistent-conn

Problematic SIP messages are SIP responses having "Call-ID" header
positioned just after message first line:
SIP/2.0 200 OK
[Call-ID header here]
[rest of the headers]

When "Call-ID" header is positioned down (after a few other headers)
it is correctly recognized.

This is due to the data offset used in get_callid function call inside
ip_vs_pe_sip.c file: since dptr already points to the start of the
SIP message, the value of dataoff should be initially 0.
Otherwise the header is searched starting from some bytes after the
first character of the SIP message.

Fixes: 758ff0338722 ("IPVS: sip persistence engine")
Signed-off-by: Marco Angaroni <marcoangaroni@gmail.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>

diff --git a/net/netfilter/ipvs/ip_vs_pe_sip.c b/net/netfilter/ipvs/ip_vs_pe_sip.c
index c4e9ca0..0a6eb5c 100644
--- a/net/netfilter/ipvs/ip_vs_pe_sip.c
+++ b/net/netfilter/ipvs/ip_vs_pe_sip.c
@@ -88,7 +88,7 @@ ip_vs_sip_fill_param(struct ip_vs_conn_param *p, struct sk_buff *skb)
 	dptr = skb->data + dataoff;
 	datalen = skb->len - dataoff;
 
-	if (get_callid(dptr, dataoff, datalen, &matchoff, &matchlen))
+	if (get_callid(dptr, 0, datalen, &matchoff, &matchlen))
 		return -EINVAL;
 
 	/* N.B: pe_data is only set on success,
-- 
cgit v0.10.2


From d1491fa54ceecfc32147626375e27937cb54b091 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Wed, 2 Mar 2016 11:20:36 -0800
Subject: act_ife: fix a typo in kmemdup() parameters

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c
index 6e7ec25..c589a9b 100644
--- a/net/sched/act_ife.c
+++ b/net/sched/act_ife.c
@@ -108,7 +108,7 @@ EXPORT_SYMBOL_GPL(ife_get_meta_u16);
 
 int ife_alloc_meta_u32(struct tcf_meta_info *mi, void *metaval)
 {
-	mi->metaval = kmemdup(&metaval, sizeof(u32), GFP_KERNEL);
+	mi->metaval = kmemdup(metaval, sizeof(u32), GFP_KERNEL);
 	if (!mi->metaval)
 		return -ENOMEM;
 
@@ -118,7 +118,7 @@ EXPORT_SYMBOL_GPL(ife_alloc_meta_u32);
 
 int ife_alloc_meta_u16(struct tcf_meta_info *mi, void *metaval)
 {
-	mi->metaval = kmemdup(&metaval, sizeof(u16), GFP_KERNEL);
+	mi->metaval = kmemdup(metaval, sizeof(u16), GFP_KERNEL);
 	if (!mi->metaval)
 		return -ENOMEM;
 
-- 
cgit v0.10.2


From 4c3e962df044c34d25403f8f58ff7b34c4966b4a Mon Sep 17 00:00:00 2001
From: Wu Fengguang <fengguang.wu@intel.com>
Date: Thu, 3 Mar 2016 09:55:19 +0800
Subject: stmmac: fix noderef.cocci warnings

drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c:115:15-21: ERROR: application of sizeof to pointer

 sizeof when applied to a pointer typed expression gives the size of
 the pointer

Generated by: scripts/coccinelle/misc/noderef.cocci

CC: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
index 69ccf48..9cf181f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
@@ -112,7 +112,7 @@ static struct stmmac_axi *stmmac_axi_setup(struct platform_device *pdev)
 	if (!np)
 		return NULL;
 
-	axi = kzalloc(sizeof(axi), GFP_KERNEL);
+	axi = kzalloc(sizeof(*axi), GFP_KERNEL);
 	if (!axi)
 		return ERR_PTR(-ENOMEM);
 
-- 
cgit v0.10.2


From 962d8cdc3133435aed2928637f73e272128a326c Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bernhard@bwalle.de>
Date: Thu, 3 Mar 2016 10:15:55 +0100
Subject: net: fec: Rename "phy-reset-active-low" property

is actually "active high". Thanks for Troy Kisky for pointing
that out.

Since the patch is in linux-next, this patch is incremental and doesn't
replace the original patch.

Signed-off-by: Bernhard Walle <bernhard@bwalle.de>
Acked-by: Fugang Duan <fugang.duan@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/fsl-fec.txt b/Documentation/devicetree/bindings/net/fsl-fec.txt
index a4799ff..b037a9d 100644
--- a/Documentation/devicetree/bindings/net/fsl-fec.txt
+++ b/Documentation/devicetree/bindings/net/fsl-fec.txt
@@ -12,7 +12,7 @@ Optional properties:
   only if property "phy-reset-gpios" is available.  Missing the property
   will have the duration be 1 millisecond.  Numbers greater than 1000 are
   invalid and 1 millisecond will be used instead.
-- phy-reset-active-low : If present then the reset sequence using the GPIO
+- phy-reset-active-high : If present then the reset sequence using the GPIO
   specified in the "phy-reset-gpios" property is reversed (H=reset state,
   L=operation state).
 - phy-supply : regulator that powers the Ethernet PHY.
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index bad0ba2..37c0815 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -3191,7 +3191,7 @@ static int fec_enet_init(struct net_device *ndev)
 static void fec_reset_phy(struct platform_device *pdev)
 {
 	int err, phy_reset;
-	bool active_low = false;
+	bool active_high = false;
 	int msec = 1;
 	struct device_node *np = pdev->dev.of_node;
 
@@ -3207,17 +3207,17 @@ static void fec_reset_phy(struct platform_device *pdev)
 	if (!gpio_is_valid(phy_reset))
 		return;
 
-	active_low = of_property_read_bool(np, "phy-reset-active-low");
+	active_high = of_property_read_bool(np, "phy-reset-active-high");
 
 	err = devm_gpio_request_one(&pdev->dev, phy_reset,
-			active_low ? GPIOF_OUT_INIT_HIGH : GPIOF_OUT_INIT_LOW,
+			active_high ? GPIOF_OUT_INIT_HIGH : GPIOF_OUT_INIT_LOW,
 			"phy-reset");
 	if (err) {
 		dev_err(&pdev->dev, "failed to get phy-reset-gpios: %d\n", err);
 		return;
 	}
 	msleep(msec);
-	gpio_set_value_cansleep(phy_reset, !active_low);
+	gpio_set_value_cansleep(phy_reset, !active_high);
 }
 #else /* CONFIG_OF */
 static void fec_reset_phy(struct platform_device *pdev)
-- 
cgit v0.10.2


From 9b3009604b8e305c32125530a3cbcc57d88f3bc1 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Thu, 3 Mar 2016 14:20:40 +0100
Subject: tipc: add net device to skb before UDP xmit

Prior to this patch enabling a IPv4 UDP bearer caused a null pointer
dereference in iptunnel_xmit_stats(), when it tried to dereference the
net device from the skb. To resolve this we now point the skb device
to the net device resolved from the routing table.

Fixes: 039f50629b7f (ip_tunnel: Move stats update to iptunnel_xmit())
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Erik Hugne <erik.hugne@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index d63a911..f22a5bb1 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -181,6 +181,8 @@ static int tipc_udp_send_msg(struct net *net, struct sk_buff *skb,
 			err = PTR_ERR(rt);
 			goto tx_error;
 		}
+
+		skb->dev = rt->dst.dev;
 		ttl = ip4_dst_hoplimit(&rt->dst);
 		udp_tunnel_xmit_skb(rt, ub->ubsock->sk, skb, src->ipv4.s_addr,
 				    dst->ipv4.s_addr, 0, ttl, 0, src->udp_port,
-- 
cgit v0.10.2


From 2837f39c7cdbd209ab04d1c1f4eca015a40d5cd6 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Thu, 3 Mar 2016 14:20:41 +0100
Subject: tipc: don't check link reset on non existing link

Make sure we have a link before checking if it has been reset or not.

Prior to this patch tipc_link_is_reset() could be called with a non
existing link, resulting in a null pointer dereference.

Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Erik Hugne <erik.hugne@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/node.c b/net/tipc/node.c
index cdb7950..590d597 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -843,7 +843,7 @@ void tipc_node_check_dest(struct net *net, u32 onode,
 	memcpy(&le->maddr, maddr, sizeof(*maddr));
 exit:
 	tipc_node_write_unlock(n);
-	if (reset && !tipc_link_is_reset(l))
+	if (reset && l && !tipc_link_is_reset(l))
 		tipc_node_link_down(n, b->identity, false);
 	tipc_node_put(n);
 }
-- 
cgit v0.10.2


From ddb3712552c8807c75576fb4fbdbb16f0d48b161 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Thu, 3 Mar 2016 14:20:42 +0100
Subject: tipc: safely copy UDP netlink data from user

The netlink policy for TIPC_NLA_UDP_LOCAL and TIPC_NLA_UDP_REMOTE
is of type binary with a defined length. This causes the policy
framework to threat the defined length as maximum length.

There is however no protection against a user sending a smaller
amount of data. Prior to this patch this wasn't handled which could
result in a partially incomplete sockaddr_storage struct containing
uninitialized data.

In this patch we use nla_memcpy() when copying the user data. This
ensures a potential gap at the end is cleared out properly.

This was found by Julia with Coccinelle tool.

Reported-by: Daniel Borkmann <daniel@iogearbox.net>
Reported-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Erik Hugne <erik.hugne@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index f22a5bb1..6fe8740 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -276,7 +276,7 @@ static int parse_options(struct nlattr *attrs[], struct udp_bearer *ub,
 			 struct udp_media_addr *remote)
 {
 	struct nlattr *opts[TIPC_NLA_UDP_MAX + 1];
-	struct sockaddr_storage *sa_local, *sa_remote;
+	struct sockaddr_storage sa_local, sa_remote;
 
 	if (!attrs[TIPC_NLA_BEARER_UDP_OPTS])
 		goto err;
@@ -285,41 +285,43 @@ static int parse_options(struct nlattr *attrs[], struct udp_bearer *ub,
 			     tipc_nl_udp_policy))
 		goto err;
 	if (opts[TIPC_NLA_UDP_LOCAL] && opts[TIPC_NLA_UDP_REMOTE]) {
-		sa_local = nla_data(opts[TIPC_NLA_UDP_LOCAL]);
-		sa_remote = nla_data(opts[TIPC_NLA_UDP_REMOTE]);
+		nla_memcpy(&sa_local, opts[TIPC_NLA_UDP_LOCAL],
+			   sizeof(sa_local));
+		nla_memcpy(&sa_remote, opts[TIPC_NLA_UDP_REMOTE],
+			   sizeof(sa_remote));
 	} else {
 err:
 		pr_err("Invalid UDP bearer configuration");
 		return -EINVAL;
 	}
-	if ((sa_local->ss_family & sa_remote->ss_family) == AF_INET) {
+	if ((sa_local.ss_family & sa_remote.ss_family) == AF_INET) {
 		struct sockaddr_in *ip4;
 
-		ip4 = (struct sockaddr_in *)sa_local;
+		ip4 = (struct sockaddr_in *)&sa_local;
 		local->proto = htons(ETH_P_IP);
 		local->udp_port = ip4->sin_port;
 		local->ipv4.s_addr = ip4->sin_addr.s_addr;
 
-		ip4 = (struct sockaddr_in *)sa_remote;
+		ip4 = (struct sockaddr_in *)&sa_remote;
 		remote->proto = htons(ETH_P_IP);
 		remote->udp_port = ip4->sin_port;
 		remote->ipv4.s_addr = ip4->sin_addr.s_addr;
 		return 0;
 
 #if IS_ENABLED(CONFIG_IPV6)
-	} else if ((sa_local->ss_family & sa_remote->ss_family) == AF_INET6) {
+	} else if ((sa_local.ss_family & sa_remote.ss_family) == AF_INET6) {
 		struct sockaddr_in6 *ip6;
 
-		ip6 = (struct sockaddr_in6 *)sa_local;
+		ip6 = (struct sockaddr_in6 *)&sa_local;
 		local->proto = htons(ETH_P_IPV6);
 		local->udp_port = ip6->sin6_port;
-		local->ipv6 = ip6->sin6_addr;
+		memcpy(&local->ipv6, &ip6->sin6_addr, sizeof(struct in6_addr));
 		ub->ifindex = ip6->sin6_scope_id;
 
-		ip6 = (struct sockaddr_in6 *)sa_remote;
+		ip6 = (struct sockaddr_in6 *)&sa_remote;
 		remote->proto = htons(ETH_P_IPV6);
 		remote->udp_port = ip6->sin6_port;
-		remote->ipv6 = ip6->sin6_addr;
+		memcpy(&remote->ipv6, &ip6->sin6_addr, sizeof(struct in6_addr));
 		return 0;
 #endif
 	}
-- 
cgit v0.10.2


From 34f65dbb6c7572b826d836d62b99de32ecf2af48 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Thu, 3 Mar 2016 14:20:43 +0100
Subject: tipc: make sure required IPv6 addresses are scoped

Make sure the user has provided a scope for multicast and link local
addresses used locally by a UDP bearer.

Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Erik Hugne <erik.hugne@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index 6fe8740..fb2f7ec6 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -310,9 +310,14 @@ err:
 
 #if IS_ENABLED(CONFIG_IPV6)
 	} else if ((sa_local.ss_family & sa_remote.ss_family) == AF_INET6) {
+		int atype;
 		struct sockaddr_in6 *ip6;
 
 		ip6 = (struct sockaddr_in6 *)&sa_local;
+		atype = ipv6_addr_type(&ip6->sin6_addr);
+		if (__ipv6_addr_needs_scope_id(atype) && !ip6->sin6_scope_id)
+			return -EINVAL;
+
 		local->proto = htons(ETH_P_IPV6);
 		local->udp_port = ip6->sin6_port;
 		memcpy(&local->ipv6, &ip6->sin6_addr, sizeof(struct in6_addr));
-- 
cgit v0.10.2


From e74a386d70c793ef44678b44b021c99094b28c47 Mon Sep 17 00:00:00 2001
From: Jon Paul Maloy <jon.maloy@ericsson.com>
Date: Thu, 3 Mar 2016 14:23:21 -0500
Subject: tipc: remove pre-allocated message header in link struct

Until now, we have kept a pre-allocated protocol message header
aggregated into struct tipc_link. Apart from adding unnecessary
footprint to the link instances, this requires extra code both to
initialize and re-initialize it.

We now remove this sub-optimization. This change also makes it
possible to clean up the function tipc_build_proto_msg() and remove
a couple of small functions that were accessing the mentioned header.
In particular, we can replace all occurrences of the local function
call link_own_addr(link) with the generic tipc_own_addr(net).

Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index e401108..ae469b3 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -412,11 +412,6 @@ enomem:
 	return -ENOMEM;
 }
 
-void tipc_bcast_reinit(struct net *net)
-{
-	tipc_link_reinit(tipc_bc_sndlink(net), tipc_own_addr(net));
-}
-
 void tipc_bcast_stop(struct net *net)
 {
 	struct tipc_net *tn = net_generic(net, tipc_net_id);
diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h
index 1944c6c..d5e79b3 100644
--- a/net/tipc/bcast.h
+++ b/net/tipc/bcast.h
@@ -46,7 +46,6 @@ struct tipc_node_map;
 extern const char tipc_bclink_name[];
 
 int tipc_bcast_init(struct net *net);
-void tipc_bcast_reinit(struct net *net);
 void tipc_bcast_stop(struct net *net);
 void tipc_bcast_add_peer(struct net *net, struct tipc_link *l,
 			 struct sk_buff_head *xmitq);
diff --git a/net/tipc/link.c b/net/tipc/link.c
index e31d92f..fc2fa83 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1,7 +1,7 @@
 /*
  * net/tipc/link.c: TIPC link code
  *
- * Copyright (c) 1996-2007, 2012-2015, Ericsson AB
+ * Copyright (c) 1996-2007, 2012-2016, Ericsson AB
  * Copyright (c) 2004-2007, 2010-2013, Wind River Systems
  * All rights reserved.
  *
@@ -127,6 +127,7 @@ struct tipc_link {
 
 	/* Management and link supervision data */
 	u32 peer_session;
+	u32 session;
 	u32 peer_bearer_id;
 	u32 bearer_id;
 	u32 tolerance;
@@ -136,11 +137,7 @@ struct tipc_link {
 	u16 peer_caps;
 	bool active;
 	u32 silent_intv_cnt;
-	struct {
-		unchar hdr[INT_H_SIZE];
-		unchar body[TIPC_MAX_IF_NAME];
-	} proto_msg;
-	struct tipc_msg *pmsg;
+	char if_name[TIPC_MAX_IF_NAME];
 	u32 priority;
 	char net_plane;
 
@@ -215,10 +212,11 @@ enum {
  * Interval between NACKs when packets arrive out of order
  */
 #define TIPC_NACK_INTV (TIPC_MIN_LINK_WIN * 2)
-/*
- * Out-of-range value for link session numbers
+
+/* Wildcard value for link session numbers. When it is known that
+ * peer endpoint is down, any session number must be accepted.
  */
-#define WILDCARD_SESSION 0x10000
+#define ANY_SESSION 0x10000
 
 /* Link FSM states:
  */
@@ -398,16 +396,6 @@ char *tipc_link_name(struct tipc_link *l)
 	return l->name;
 }
 
-static u32 link_own_addr(struct tipc_link *l)
-{
-	return msg_prevnode(l->pmsg);
-}
-
-void tipc_link_reinit(struct tipc_link *l, u32 addr)
-{
-	msg_set_prevnode(l->pmsg, addr);
-}
-
 /**
  * tipc_link_create - create a new link
  * @n: pointer to associated node
@@ -441,29 +429,22 @@ bool tipc_link_create(struct net *net, char *if_name, int bearer_id,
 		      struct tipc_link **link)
 {
 	struct tipc_link *l;
-	struct tipc_msg *hdr;
 
 	l = kzalloc(sizeof(*l), GFP_ATOMIC);
 	if (!l)
 		return false;
 	*link = l;
-	l->pmsg = (struct tipc_msg *)&l->proto_msg;
-	hdr = l->pmsg;
-	tipc_msg_init(ownnode, hdr, LINK_PROTOCOL, RESET_MSG, INT_H_SIZE, peer);
-	msg_set_size(hdr, sizeof(l->proto_msg));
-	msg_set_session(hdr, session);
-	msg_set_bearer_id(hdr, l->bearer_id);
+	l->session = session;
 
 	/* Note: peer i/f name is completed by reset/activate message */
 	sprintf(l->name, "%u.%u.%u:%s-%u.%u.%u:unknown",
 		tipc_zone(ownnode), tipc_cluster(ownnode), tipc_node(ownnode),
 		if_name, tipc_zone(peer), tipc_cluster(peer), tipc_node(peer));
-	strcpy((char *)msg_data(hdr), if_name);
-
+	strcpy(l->if_name, if_name);
 	l->addr = peer;
 	l->peer_caps = peer_caps;
 	l->net = net;
-	l->peer_session = WILDCARD_SESSION;
+	l->peer_session = ANY_SESSION;
 	l->bearer_id = bearer_id;
 	l->tolerance = tolerance;
 	l->net_plane = net_plane;
@@ -790,7 +771,7 @@ static int link_schedule_user(struct tipc_link *link, struct sk_buff_head *list)
 	struct tipc_msg *msg = buf_msg(skb_peek(list));
 	int imp = msg_importance(msg);
 	u32 oport = msg_origport(msg);
-	u32 addr = link_own_addr(link);
+	u32 addr = tipc_own_addr(link->net);
 	struct sk_buff *skb;
 
 	/* This really cannot happen...  */
@@ -839,16 +820,9 @@ void link_prepare_wakeup(struct tipc_link *l)
 
 void tipc_link_reset(struct tipc_link *l)
 {
-	/* Link is down, accept any session */
-	l->peer_session = WILDCARD_SESSION;
-
-	/* If peer is up, it only accepts an incremented session number */
-	msg_set_session(l->pmsg, msg_session(l->pmsg) + 1);
-
-	/* Prepare for renewed mtu size negotiation */
+	l->peer_session = ANY_SESSION;
+	l->session++;
 	l->mtu = l->advertised_mtu;
-
-	/* Clean up all queues and counters: */
 	__skb_queue_purge(&l->transmq);
 	__skb_queue_purge(&l->deferdq);
 	skb_queue_splice_init(&l->wakeupq, l->inputq);
@@ -1156,7 +1130,7 @@ int tipc_link_build_ack_msg(struct tipc_link *l, struct sk_buff_head *xmitq)
 
 	/* Broadcast ACK must be sent via a unicast link => defer to caller */
 	if (link_is_bc_rcvlink(l)) {
-		if (((l->rcv_nxt ^ link_own_addr(l)) & 0xf) != 0xf)
+		if (((l->rcv_nxt ^ tipc_own_addr(l->net)) & 0xf) != 0xf)
 			return 0;
 		l->rcv_unacked = 0;
 		return TIPC_LINK_SND_BC_ACK;
@@ -1268,15 +1242,30 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
 				      u16 rcvgap, int tolerance, int priority,
 				      struct sk_buff_head *xmitq)
 {
-	struct sk_buff *skb = NULL;
-	struct tipc_msg *hdr = l->pmsg;
+	struct sk_buff *skb;
+	struct tipc_msg *hdr;
+	struct sk_buff_head *dfq = &l->deferdq;
 	bool node_up = link_is_up(l->bc_rcvlink);
 
 	/* Don't send protocol message during reset or link failover */
 	if (tipc_link_is_blocked(l))
 		return;
 
-	msg_set_type(hdr, mtyp);
+	if (!tipc_link_is_up(l) && (mtyp == STATE_MSG))
+		return;
+
+	if (!skb_queue_empty(dfq))
+		rcvgap = buf_seqno(skb_peek(dfq)) - l->rcv_nxt;
+
+	skb = tipc_msg_create(LINK_PROTOCOL, mtyp, INT_H_SIZE,
+			      TIPC_MAX_IF_NAME, l->addr,
+			      tipc_own_addr(l->net), 0, 0, 0);
+	if (!skb)
+		return;
+
+	hdr = buf_msg(skb);
+	msg_set_session(hdr, l->session);
+	msg_set_bearer_id(hdr, l->bearer_id);
 	msg_set_net_plane(hdr, l->net_plane);
 	msg_set_next_sent(hdr, l->snd_nxt);
 	msg_set_ack(hdr, l->rcv_nxt - 1);
@@ -1286,36 +1275,23 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
 	msg_set_linkprio(hdr, priority);
 	msg_set_redundant_link(hdr, node_up);
 	msg_set_seq_gap(hdr, 0);
-
-	/* Compatibility: created msg must not be in sequence with pkt flow */
 	msg_set_seqno(hdr, l->snd_nxt + U16_MAX / 2);
 
 	if (mtyp == STATE_MSG) {
-		if (!tipc_link_is_up(l))
-			return;
-
-		/* Override rcvgap if there are packets in deferred queue */
-		if (!skb_queue_empty(&l->deferdq))
-			rcvgap = buf_seqno(skb_peek(&l->deferdq)) - l->rcv_nxt;
-		if (rcvgap) {
-			msg_set_seq_gap(hdr, rcvgap);
-			l->stats.sent_nacks++;
-		}
+		msg_set_seq_gap(hdr, rcvgap);
+		msg_set_size(hdr, INT_H_SIZE);
 		msg_set_probe(hdr, probe);
-		if (probe)
-			l->stats.sent_probes++;
 		l->stats.sent_states++;
 		l->rcv_unacked = 0;
 	} else {
 		/* RESET_MSG or ACTIVATE_MSG */
 		msg_set_max_pkt(hdr, l->advertised_mtu);
-		msg_set_ack(hdr, l->rcv_nxt - 1);
-		msg_set_next_sent(hdr, 1);
+		strcpy(msg_data(hdr), l->if_name);
 	}
-	skb = tipc_buf_acquire(msg_size(hdr));
-	if (!skb)
-		return;
-	skb_copy_to_linear_data(skb, hdr, msg_size(hdr));
+	if (probe)
+		l->stats.sent_probes++;
+	if (rcvgap)
+		l->stats.sent_nacks++;
 	skb->priority = TC_PRIO_CONTROL;
 	__skb_queue_tail(xmitq, skb);
 }
@@ -1340,7 +1316,7 @@ void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl,
 
 	/* At least one packet required for safe algorithm => add dummy */
 	skb = tipc_msg_create(TIPC_LOW_IMPORTANCE, TIPC_DIRECT_MSG,
-			      BASIC_H_SIZE, 0, l->addr, link_own_addr(l),
+			      BASIC_H_SIZE, 0, l->addr, tipc_own_addr(l->net),
 			      0, 0, TIPC_ERR_NO_PORT);
 	if (!skb) {
 		pr_warn("%sunable to create tunnel packet\n", link_co_err);
@@ -1351,7 +1327,7 @@ void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl,
 	__skb_queue_purge(&tmpxq);
 
 	/* Initialize reusable tunnel packet header */
-	tipc_msg_init(link_own_addr(l), &tnlhdr, TUNNEL_PROTOCOL,
+	tipc_msg_init(tipc_own_addr(l->net), &tnlhdr, TUNNEL_PROTOCOL,
 		      mtyp, INT_H_SIZE, l->addr);
 	pktcnt = skb_queue_len(&l->transmq) + skb_queue_len(&l->backlogq);
 	msg_set_msgcnt(&tnlhdr, pktcnt);
@@ -1410,7 +1386,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 	if (tipc_link_is_blocked(l) || !xmitq)
 		goto exit;
 
-	if (link_own_addr(l) > msg_prevnode(hdr))
+	if (tipc_own_addr(l->net) > msg_prevnode(hdr))
 		l->net_plane = msg_net_plane(hdr);
 
 	switch (mtyp) {
@@ -1418,7 +1394,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 
 		/* Ignore duplicate RESET with old session number */
 		if ((less_eq(msg_session(hdr), l->peer_session)) &&
-		    (l->peer_session != WILDCARD_SESSION))
+		    (l->peer_session != ANY_SESSION))
 			break;
 		/* fall thru' */
 
@@ -1515,7 +1491,7 @@ static bool tipc_link_build_bc_proto_msg(struct tipc_link *l, bool bcast,
 	u16 gap_to = peers_snd_nxt - 1;
 
 	skb = tipc_msg_create(BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE,
-			      0, l->addr, link_own_addr(l), 0, 0, 0);
+			      0, l->addr, tipc_own_addr(l->net), 0, 0, 0);
 	if (!skb)
 		return false;
 	hdr = buf_msg(skb);
@@ -1670,7 +1646,7 @@ int tipc_link_bc_nack_rcv(struct tipc_link *l, struct sk_buff *skb,
 	if (mtyp != STATE_MSG)
 		return 0;
 
-	if (dnode == link_own_addr(l)) {
+	if (dnode == tipc_own_addr(l->net)) {
 		tipc_link_bc_ack_rcv(l, acked, xmitq);
 		rc = tipc_link_retrans(l->bc_sndlink, from, to, xmitq);
 		l->stats.recv_nacks++;
diff --git a/net/tipc/link.h b/net/tipc/link.h
index b4ee9d6..6a94175 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -86,7 +86,6 @@ bool tipc_link_bc_create(struct net *net, u32 ownnode, u32 peer,
 			 struct sk_buff_head *namedq,
 			 struct tipc_link *bc_sndlink,
 			 struct tipc_link **link);
-void tipc_link_reinit(struct tipc_link *l, u32 addr);
 void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl,
 			   int mtyp, struct sk_buff_head *xmitq);
 void tipc_link_build_reset_msg(struct tipc_link *l, struct sk_buff_head *xmitq);
diff --git a/net/tipc/net.c b/net/tipc/net.c
index 77bf911..86d68b3 100644
--- a/net/tipc/net.c
+++ b/net/tipc/net.c
@@ -116,7 +116,6 @@ int tipc_net_start(struct net *net, u32 addr)
 	tn->own_addr = addr;
 	tipc_named_reinit(net);
 	tipc_sk_reinit(net);
-	tipc_bcast_reinit(net);
 
 	tipc_nametbl_publish(net, TIPC_CFG_SRV, tn->own_addr, tn->own_addr,
 			     TIPC_ZONE_SCOPE, 0, tn->own_addr);
-- 
cgit v0.10.2


From afd5b1704cb15eabc4358b7b0dd7f33099408eaa Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 1 Mar 2016 12:18:22 +0200
Subject: iwlwifi: refactor the code that reads the MAC address from the NVM

It makes it slightly easier to follow. Pass the pointer to
the transport which allows to read WFMP_MAC_ADDR_X register
only when needed and to use IWL_ERR instead of the less
commonly used IWL_ERR_DEV logger macro.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 3481357..5e6b90d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2008 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -69,6 +70,7 @@
 #include "iwl-drv.h"
 #include "iwl-modparams.h"
 #include "iwl-nvm-parse.h"
+#include "iwl-prph.h"
 
 /* NVM offsets (in words) definitions */
 enum wkp_nvm_offsets {
@@ -522,27 +524,11 @@ static void iwl_set_radio_cfg(const struct iwl_cfg *cfg,
 	data->valid_rx_ant = NVM_RF_CFG_RX_ANT_MSK_FAMILY_8000(radio_cfg);
 }
 
-static void iwl_set_hw_address(const struct iwl_cfg *cfg,
-			       struct iwl_nvm_data *data,
-			       const __le16 *nvm_sec)
-{
-	const u8 *hw_addr = (const u8 *)(nvm_sec + HW_ADDR);
-
-	/* The byte order is little endian 16 bit, meaning 214365 */
-	data->hw_addr[0] = hw_addr[1];
-	data->hw_addr[1] = hw_addr[0];
-	data->hw_addr[2] = hw_addr[3];
-	data->hw_addr[3] = hw_addr[2];
-	data->hw_addr[4] = hw_addr[5];
-	data->hw_addr[5] = hw_addr[4];
-}
-
-static void iwl_set_hw_address_family_8000(struct device *dev,
+static void iwl_set_hw_address_family_8000(struct iwl_trans *trans,
 					   const struct iwl_cfg *cfg,
 					   struct iwl_nvm_data *data,
 					   const __le16 *mac_override,
-					   const __le16 *nvm_hw,
-					   __le32 mac_addr0, __le32 mac_addr1)
+					   const __le16 *nvm_hw)
 {
 	const u8 *hw_addr;
 
@@ -568,11 +554,16 @@ static void iwl_set_hw_address_family_8000(struct device *dev,
 		    memcmp(reserved_mac, hw_addr, ETH_ALEN) != 0)
 			return;
 
-		IWL_ERR_DEV(dev,
-			    "mac address from nvm override section is not valid\n");
+		IWL_ERR(trans,
+			"mac address from nvm override section is not valid\n");
 	}
 
 	if (nvm_hw) {
+		/* read the mac address from WFMP registers */
+		__le32 mac_addr0 = cpu_to_le32(iwl_trans_read_prph(trans,
+						WFMP_MAC_ADDR_0));
+		__le32 mac_addr1 = cpu_to_le32(iwl_trans_read_prph(trans,
+						WFMP_MAC_ADDR_1));
 		/* read the MAC address from HW resisters */
 		hw_addr = (const u8 *)&mac_addr0;
 		data->hw_addr[0] = hw_addr[3];
@@ -585,28 +576,50 @@ static void iwl_set_hw_address_family_8000(struct device *dev,
 		data->hw_addr[5] = hw_addr[0];
 
 		if (!is_valid_ether_addr(data->hw_addr))
-			IWL_ERR_DEV(dev,
-				    "mac address (%pM) from hw section is not valid\n",
-				    data->hw_addr);
+			IWL_ERR(trans,
+				"mac address (%pM) from hw section is not valid\n",
+				data->hw_addr);
 
 		return;
 	}
 
-	IWL_ERR_DEV(dev, "mac address is not found\n");
+	IWL_ERR(trans, "mac address is not found\n");
+}
+
+static void iwl_set_hw_address(struct iwl_trans *trans,
+			       const struct iwl_cfg *cfg,
+			       struct iwl_nvm_data *data, const __le16 *nvm_hw,
+			       const __le16 *mac_override)
+{
+	if (cfg->device_family != IWL_DEVICE_FAMILY_8000) {
+		const u8 *hw_addr = (const u8 *)(nvm_hw + HW_ADDR);
+
+		/* The byte order is little endian 16 bit, meaning 214365 */
+		data->hw_addr[0] = hw_addr[1];
+		data->hw_addr[1] = hw_addr[0];
+		data->hw_addr[2] = hw_addr[3];
+		data->hw_addr[3] = hw_addr[2];
+		data->hw_addr[4] = hw_addr[5];
+		data->hw_addr[5] = hw_addr[4];
+	} else {
+		iwl_set_hw_address_family_8000(trans, cfg, data,
+					       mac_override, nvm_hw);
+	}
 }
 
 struct iwl_nvm_data *
-iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
+iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		   const __le16 *nvm_hw, const __le16 *nvm_sw,
 		   const __le16 *nvm_calib, const __le16 *regulatory,
 		   const __le16 *mac_override, const __le16 *phy_sku,
-		   u8 tx_chains, u8 rx_chains, bool lar_fw_supported,
-		   __le32 mac_addr0, __le32 mac_addr1)
+		   u8 tx_chains, u8 rx_chains, bool lar_fw_supported)
 {
+	struct device *dev = trans->dev;
 	struct iwl_nvm_data *data;
-	u32 sku;
-	u32 radio_cfg;
+	bool lar_enabled;
+	u32 sku, radio_cfg;
 	u16 lar_config;
+	const __le16 *ch_section;
 
 	if (cfg->device_family != IWL_DEVICE_FAMILY_8000)
 		data = kzalloc(sizeof(*data) +
@@ -645,21 +658,16 @@ iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
 	if (cfg->device_family != IWL_DEVICE_FAMILY_8000) {
 		/* Checking for required sections */
 		if (!nvm_calib) {
-			IWL_ERR_DEV(dev,
-				    "Can't parse empty Calib NVM sections\n");
+			IWL_ERR(trans,
+				"Can't parse empty Calib NVM sections\n");
 			kfree(data);
 			return NULL;
 		}
 		/* in family 8000 Xtal calibration values moved to OTP */
 		data->xtal_calib[0] = *(nvm_calib + XTAL_CALIB);
 		data->xtal_calib[1] = *(nvm_calib + XTAL_CALIB + 1);
-	}
-
-	if (cfg->device_family != IWL_DEVICE_FAMILY_8000) {
-		iwl_set_hw_address(cfg, data, nvm_hw);
-
-		iwl_init_sbands(dev, cfg, data, nvm_sw,
-				tx_chains, rx_chains, lar_fw_supported);
+		lar_enabled = true;
+		ch_section = nvm_sw;
 	} else {
 		u16 lar_offset = data->nvm_version < 0xE39 ?
 				 NVM_LAR_OFFSET_FAMILY_8000_OLD :
@@ -668,16 +676,13 @@ iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
 		lar_config = le16_to_cpup(regulatory + lar_offset);
 		data->lar_enabled = !!(lar_config &
 				       NVM_LAR_ENABLED_FAMILY_8000);
-
-		/* MAC address in family 8000 */
-		iwl_set_hw_address_family_8000(dev, cfg, data, mac_override,
-					       nvm_hw, mac_addr0, mac_addr1);
-
-		iwl_init_sbands(dev, cfg, data, regulatory,
-				tx_chains, rx_chains,
-				lar_fw_supported && data->lar_enabled);
+		lar_enabled = data->lar_enabled;
+		ch_section = regulatory;
 	}
 
+	iwl_set_hw_address(trans, cfg, data, nvm_hw, mac_override);
+	iwl_init_sbands(dev, cfg, data, ch_section, tx_chains, rx_chains,
+			lar_fw_supported && lar_enabled);
 	data->calib_version = 255;
 
 	return data;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
index 4e8e0dc..d704d52 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
@@ -74,12 +74,11 @@
  * later with iwl_free_nvm_data().
  */
 struct iwl_nvm_data *
-iwl_parse_nvm_data(struct device *dev, const struct iwl_cfg *cfg,
+iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		   const __le16 *nvm_hw, const __le16 *nvm_sw,
 		   const __le16 *nvm_calib, const __le16 *regulatory,
 		   const __le16 *mac_override, const __le16 *phy_sku,
-		   u8 tx_chains, u8 rx_chains, bool lar_fw_supported,
-		   __le32 mac_addr0, __le32 mac_addr1);
+		   u8 tx_chains, u8 rx_chains, bool lar_fw_supported);
 
 /**
  * iwl_parse_mcc_info - parse MCC (mobile country code) info coming from FW
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
index c446e0d..25a9840 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
@@ -300,7 +300,6 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm)
 	struct iwl_nvm_section *sections = mvm->nvm_sections;
 	const __le16 *hw, *sw, *calib, *regulatory, *mac_override, *phy_sku;
 	bool lar_enabled;
-	__le32 mac_addr0, mac_addr1;
 
 	/* Checking for required sections */
 	if (mvm->trans->cfg->device_family != IWL_DEVICE_FAMILY_8000) {
@@ -336,12 +335,6 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm)
 	if (WARN_ON(!mvm->cfg))
 		return NULL;
 
-	/* read the mac address from WFMP registers */
-	mac_addr0 = cpu_to_le32(iwl_trans_read_prph(mvm->trans,
-						    WFMP_MAC_ADDR_0));
-	mac_addr1 = cpu_to_le32(iwl_trans_read_prph(mvm->trans,
-						    WFMP_MAC_ADDR_1));
-
 	hw = (const __le16 *)sections[mvm->cfg->nvm_hw_section_num].data;
 	sw = (const __le16 *)sections[NVM_SECTION_TYPE_SW].data;
 	calib = (const __le16 *)sections[NVM_SECTION_TYPE_CALIBRATION].data;
@@ -354,10 +347,10 @@ iwl_parse_nvm_sections(struct iwl_mvm *mvm)
 		      fw_has_capa(&mvm->fw->ucode_capa,
 				  IWL_UCODE_TLV_CAPA_LAR_SUPPORT);
 
-	return iwl_parse_nvm_data(mvm->trans->dev, mvm->cfg, hw, sw, calib,
+	return iwl_parse_nvm_data(mvm->trans, mvm->cfg, hw, sw, calib,
 				  regulatory, mac_override, phy_sku,
 				  mvm->fw->valid_tx_ant, mvm->fw->valid_rx_ant,
-				  lar_enabled, mac_addr0, mac_addr1);
+				  lar_enabled);
 }
 
 #define MAX_NVM_FILE_LEN	16384
-- 
cgit v0.10.2


From 04ddc2aacc1c52793c1e30db5073bef6fbc08942 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Thu, 3 Mar 2016 13:31:39 +0200
Subject: iwlwifi: mvm: fix unregistration of thermal in some error flows

The call to iwl_mvm_thermal_initialize() was too early in the
function.
Unregister will be performed when goto out_unregister is called,
but as the code was - out_free may be called and leave without
unregistering from thermal.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 699a808..4322e8e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -611,9 +611,6 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	IWL_INFO(mvm, "Detected %s, REV=0x%X\n",
 		 mvm->cfg->name, mvm->trans->hw_rev);
 
-	min_backoff = calc_min_backoff(trans, cfg);
-	iwl_mvm_thermal_initialize(mvm, min_backoff);
-
 	if (iwlwifi_mod_params.nvm_file)
 		mvm->nvm_file_name = iwlwifi_mod_params.nvm_file;
 	else
@@ -666,6 +663,9 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	if (err)
 		goto out_free;
 
+	min_backoff = calc_min_backoff(trans, cfg);
+	iwl_mvm_thermal_initialize(mvm, min_backoff);
+
 	err = iwl_mvm_dbgfs_register(mvm, dbgfs_dir);
 	if (err)
 		goto out_unregister;
-- 
cgit v0.10.2


From 30d915c2ecbb60d222f49e8f249d04e208d11ebc Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Thu, 3 Mar 2016 16:49:05 +0200
Subject: iwlwifi: mvm: set the correct amsdu enum values

The amsdu enum values are off by 1 bit. Fix it.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
index eb9b870..7a16e55 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h
@@ -264,9 +264,8 @@ enum iwl_rx_mpdu_mac_flags2 {
 };
 
 enum iwl_rx_mpdu_amsdu_info {
-	IWL_RX_MPDU_AMSDU_SUBFRAME_IDX_MASK	= 0x3f,
-	IWL_RX_MPDU_AMSDU_LAST_SUBFRAME		= 0x40,
-	/* 0x80 bit reserved for now */
+	IWL_RX_MPDU_AMSDU_SUBFRAME_IDX_MASK	= 0x7f,
+	IWL_RX_MPDU_AMSDU_LAST_SUBFRAME		= 0x80,
 };
 
 enum iwl_rx_l3l4_flags {
-- 
cgit v0.10.2


From 7c70fee5ae5c4fb542e432599cb85c8031b952a0 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Tue, 2 Feb 2016 11:55:53 +0200
Subject: iwlwifi: mvm: extend time event duration

Before authentication, we start a time event during
which we wait for a beacon in order to sync our timers.
If we didn't hear the beacon during this time - we abandon
the connection. However, in congested environment, it was
observed we might not hear beacons in that time slot.
Extend the time event to give the connection a better chance.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index ec6b072..3590835 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2556,10 +2556,8 @@ static void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
 				      struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	u32 duration = min(IWL_MVM_TE_SESSION_PROTECTION_MAX_TIME_MS,
-			   200 + vif->bss_conf.beacon_int);
-	u32 min_duration = min(IWL_MVM_TE_SESSION_PROTECTION_MIN_TIME_MS,
-			       100 + vif->bss_conf.beacon_int);
+	u32 duration = IWL_MVM_TE_SESSION_PROTECTION_MAX_TIME_MS;
+	u32 min_duration = IWL_MVM_TE_SESSION_PROTECTION_MIN_TIME_MS;
 
 	if (WARN_ON_ONCE(vif->bss_conf.assoc))
 		return;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.h b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.h
index 99d9a35..3d2e8b6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.h
@@ -115,7 +115,7 @@
  * needed by the driver.
  */
 
-#define IWL_MVM_TE_SESSION_PROTECTION_MAX_TIME_MS 500
+#define IWL_MVM_TE_SESSION_PROTECTION_MAX_TIME_MS 600
 #define IWL_MVM_TE_SESSION_PROTECTION_MIN_TIME_MS 400
 
 /**
-- 
cgit v0.10.2


From 7b5acd11a3ffec027abf4df2b30690a974873174 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 11 Feb 2016 10:56:22 -0600
Subject: rtlwifi: rtl8821ae: Silence useless log output

When driver rtl8821ae is loaded but not connected to any AP, it logs
a "firmware not ready to run" message roughly once a minute. To
eliminate logging this massage under normal debug conditions, the
degug level needed to print this message is increased.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.c
index 525eb23..a4fc70e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/fw.c
@@ -271,7 +271,7 @@ int rtl8821ae_download_fw(struct ieee80211_hw *hw, bool buse_wake_on_wlan_fw)
 
 	err = _rtl8821ae_fw_free_to_go(hw);
 	if (err) {
-		RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
+		RT_TRACE(rtlpriv, COMP_ERR, DBG_DMESG,
 			 "Firmware is not ready to run!\n");
 	} else {
 		RT_TRACE(rtlpriv, COMP_FW, DBG_LOUD,
-- 
cgit v0.10.2


From 9cc3fdc86c6006bae878da6d2d14c1633064a12f Mon Sep 17 00:00:00 2001
From: One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk>
Date: Mon, 15 Feb 2016 19:04:56 +0000
Subject: rt2x00: unterminated strlen of user data

The buffer needs to be zero terminated in case the user data is not.
Otherwise we run off the end of the buffer.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c b/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c
index 25ee3cb..72ae530 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c
@@ -478,7 +478,7 @@ static ssize_t rt2x00debug_write_##__name(struct file *file,	\
 {								\
 	struct rt2x00debug_intf *intf = file->private_data;	\
 	const struct rt2x00debug *debug = intf->debug;		\
-	char line[16];						\
+	char line[17];						\
 	size_t size;						\
 	unsigned int index = intf->offset_##__name;		\
 	__type value;						\
@@ -494,7 +494,8 @@ static ssize_t rt2x00debug_write_##__name(struct file *file,	\
 								\
 	if (copy_from_user(line, buf, length))			\
 		return -EFAULT;					\
-								\
+	line[16] = 0;						\
+						\
 	size = strlen(line);					\
 	value = simple_strtoul(line, NULL, 0);			\
 								\
-- 
cgit v0.10.2


From 2f10e50e3d284ba2a4e787bfe19e89770f87a8b3 Mon Sep 17 00:00:00 2001
From: Christian Lamparter <chunkeey@googlemail.com>
Date: Mon, 15 Feb 2016 21:16:06 +0100
Subject: carl9170: import 1.9.9 firmware headers

Import new headers from my firmware branch:
<https://github.com/chunkeey/carl9170fw>

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ath/carl9170/fwcmd.h b/drivers/net/wireless/ath/carl9170/fwcmd.h
index 9111d4f..ea1d80f 100644
--- a/drivers/net/wireless/ath/carl9170/fwcmd.h
+++ b/drivers/net/wireless/ath/carl9170/fwcmd.h
@@ -56,6 +56,7 @@ enum carl9170_cmd_oids {
 	CARL9170_CMD_RX_FILTER		= 0x07,
 	CARL9170_CMD_WOL		= 0x08,
 	CARL9170_CMD_TALLY		= 0x09,
+	CARL9170_CMD_WREGB		= 0x0a,
 
 	/* CAM */
 	CARL9170_CMD_EKEY		= 0x10,
@@ -123,6 +124,12 @@ struct carl9170_write_reg {
 	} regs[0] __packed;
 } __packed;
 
+struct carl9170_write_reg_byte {
+	__le32	addr;
+	__le32  count;
+	u8	val[0];
+} __packed;
+
 #define	CARL9170FW_PHY_HT_ENABLE		0x4
 #define	CARL9170FW_PHY_HT_DYN2040		0x8
 #define	CARL9170FW_PHY_HT_EXT_CHAN_OFF		0x3
@@ -226,6 +233,7 @@ struct carl9170_cmd {
 		struct carl9170_u32_list	echo;
 		struct carl9170_reg_list	rreg;
 		struct carl9170_write_reg	wreg;
+		struct carl9170_write_reg_byte	wregb;
 		struct carl9170_rf_init		rf_init;
 		struct carl9170_psm		psm;
 		struct carl9170_wol_cmd		wol;
diff --git a/drivers/net/wireless/ath/carl9170/fwdesc.h b/drivers/net/wireless/ath/carl9170/fwdesc.h
index 66848d4..0533f79c 100644
--- a/drivers/net/wireless/ath/carl9170/fwdesc.h
+++ b/drivers/net/wireless/ath/carl9170/fwdesc.h
@@ -81,6 +81,12 @@ enum carl9170fw_feature_list {
 	/* Firmware will pass BA when BARs are queued */
 	CARL9170FW_RX_BA_FILTER,
 
+	/* Firmware has support to write a byte at a time */
+	CARL9170FW_HAS_WREGB_CMD,
+
+	/* Pattern generator */
+	CARL9170FW_PATTERN_GENERATOR,
+
 	/* KEEP LAST */
 	__CARL9170FW_FEATURE_NUM
 };
diff --git a/drivers/net/wireless/ath/carl9170/hw.h b/drivers/net/wireless/ath/carl9170/hw.h
index 0db874a..08e0ae9 100644
--- a/drivers/net/wireless/ath/carl9170/hw.h
+++ b/drivers/net/wireless/ath/carl9170/hw.h
@@ -453,9 +453,74 @@
 #define	AR9170_MC_REG_BASE			0x1d1000
 
 #define	AR9170_MC_REG_FLASH_WAIT_STATE		(AR9170_MC_REG_BASE + 0x000)
-#define	AR9170_MC_REG_SEEPROM_WP0		(AR9170_MC_REG_BASE + 0x400)
-#define	AR9170_MC_REG_SEEPROM_WP1		(AR9170_MC_REG_BASE + 0x404)
-#define	AR9170_MC_REG_SEEPROM_WP2		(AR9170_MC_REG_BASE + 0x408)
+
+#define AR9170_SPI_REG_BASE			(AR9170_MC_REG_BASE + 0x200)
+#define AR9170_SPI_REG_CONTROL0			(AR9170_SPI_REG_BASE + 0x000)
+#define		AR9170_SPI_CONTROL0_BUSY		BIT(0)
+#define		AR9170_SPI_CONTROL0_CMD_GO		BIT(1)
+#define		AR9170_SPI_CONTROL0_PAGE_WR		BIT(2)
+#define		AR9170_SPI_CONTROL0_SEQ_RD		BIT(3)
+#define		AR9170_SPI_CONTROL0_CMD_ABORT		BIT(4)
+#define		AR9170_SPI_CONTROL0_CMD_LEN_S		8
+#define		AR9170_SPI_CONTROL0_CMD_LEN		0x00000f00
+#define		AR9170_SPI_CONTROL0_RD_LEN_S		12
+#define		AR9170_SPI_CONTROL0_RD_LEN		0x00007000
+
+#define	AR9170_SPI_REG_CONTROL1			(AR9170_SPI_REG_BASE + 0x004)
+#define		AR9170_SPI_CONTROL1_SCK_RATE		BIT(0)
+#define		AR9170_SPI_CONTROL1_DRIVE_SDO		BIT(1)
+#define		AR9170_SPI_CONTROL1_MODE_SEL_S		2
+#define		AR9170_SPI_CONTROL1_MODE_SEL		0x000000c0
+#define		AR9170_SPI_CONTROL1_WRITE_PROTECT	BIT(4)
+
+#define AR9170_SPI_REG_COMMAND_PORT0		(AR9170_SPI_REG_BASE + 0x008)
+#define		AR9170_SPI_COMMAND_PORT0_CMD0_S		0
+#define		AR9170_SPI_COMMAND_PORT0_CMD0		0x000000ff
+#define		AR9170_SPI_COMMAND_PORT0_CMD1_S		8
+#define		AR9170_SPI_COMMAND_PORT0_CMD1		0x0000ff00
+#define		AR9170_SPI_COMMAND_PORT0_CMD2_S		16
+#define		AR9170_SPI_COMMAND_PORT0_CMD2		0x00ff0000
+#define		AR9170_SPI_COMMAND_PORT0_CMD3_S		24
+#define		AR9170_SPI_COMMAND_PORT0_CMD3		0xff000000
+
+#define AR9170_SPI_REG_COMMAND_PORT1		(AR9170_SPI_REG_BASE + 0x00C)
+#define		AR9170_SPI_COMMAND_PORT1_CMD4_S		0
+#define		AR9170_SPI_COMMAND_PORT1_CMD4		0x000000ff
+#define		AR9170_SPI_COMMAND_PORT1_CMD5_S		8
+#define		AR9170_SPI_COMMAND_PORT1_CMD5		0x0000ff00
+#define		AR9170_SPI_COMMAND_PORT1_CMD6_S		16
+#define		AR9170_SPI_COMMAND_PORT1_CMD6		0x00ff0000
+#define		AR9170_SPI_COMMAND_PORT1_CMD7_S		24
+#define		AR9170_SPI_COMMAND_PORT1_CMD7		0xff000000
+
+#define AR9170_SPI_REG_DATA_PORT		(AR9170_SPI_REG_BASE + 0x010)
+#define AR9170_SPI_REG_PAGE_WRITE_LEN		(AR9170_SPI_REG_BASE + 0x014)
+
+#define AR9170_EEPROM_REG_BASE			(AR9170_MC_REG_BASE + 0x400)
+#define	AR9170_EEPROM_REG_WP_MAGIC1		(AR9170_EEPROM_REG_BASE + 0x000)
+#define		AR9170_EEPROM_WP_MAGIC1			0x12345678
+
+#define	AR9170_EEPROM_REG_WP_MAGIC2		(AR9170_EEPROM_REG_BASE + 0x004)
+#define		AR9170_EEPROM_WP_MAGIC2			0x55aa00ff
+
+#define	AR9170_EEPROM_REG_WP_MAGIC3		(AR9170_EEPROM_REG_BASE + 0x008)
+#define		AR9170_EEPROM_WP_MAGIC3			0x13579ace
+
+#define	AR9170_EEPROM_REG_CLOCK_DIV		(AR9170_EEPROM_REG_BASE + 0x00C)
+#define		AR9170_EEPROM_CLOCK_DIV_FAC_S		0
+#define		AR9170_EEPROM_CLOCK_DIV_FAC		0x000001ff
+#define		AR9170_EEPROM_CLOCK_DIV_FAC_39KHZ	0xff
+#define		AR9170_EEPROM_CLOCK_DIV_FAC_78KHZ	0x7f
+#define		AR9170_EEPROM_CLOCK_DIV_FAC_312KHZ	0x1f
+#define		AR9170_EEPROM_CLOCK_DIV_FAC_10MHZ	0x0
+#define	AR9170_EEPROM_CLOCK_DIV_SOFT_RST		BIT(9)
+
+#define AR9170_EEPROM_REG_MODE			(AR9170_EEPROM_REG_BASE + 0x010)
+#define	AR9170_EEPROM_MODE_EEPROM_SIZE_16K_PLUS		BIT(31)
+
+#define AR9170_EEPROM_REG_WRITE_PROTECT		(AR9170_EEPROM_REG_BASE + 0x014)
+#define		AR9170_EEPROM_WRITE_PROTECT_WP_STATUS	BIT(0)
+#define		AR9170_EEPROM_WRITE_PROTECT_WP_SET	BIT(8)
 
 /* Interrupt Controller */
 #define	AR9170_MAX_INT_SRC			9
@@ -589,11 +654,13 @@
 #define	AR9170_USB_REG_EP10_MAP			(AR9170_USB_REG_BASE + 0x039)
 
 #define	AR9170_USB_REG_EP_IN_MAX_SIZE_HIGH	(AR9170_USB_REG_BASE + 0x03f)
+#define		AR9170_USB_EP_IN_STALL			0x8
 #define		AR9170_USB_EP_IN_TOGGLE			0x10
 
 #define	AR9170_USB_REG_EP_IN_MAX_SIZE_LOW	(AR9170_USB_REG_BASE + 0x03e)
 
 #define	AR9170_USB_REG_EP_OUT_MAX_SIZE_HIGH	(AR9170_USB_REG_BASE + 0x05f)
+#define		AR9170_USB_EP_OUT_STALL			0x8
 #define		AR9170_USB_EP_OUT_TOGGLE		0x10
 
 #define	AR9170_USB_REG_EP_OUT_MAX_SIZE_LOW	(AR9170_USB_REG_BASE + 0x05e)
diff --git a/drivers/net/wireless/ath/carl9170/version.h b/drivers/net/wireless/ath/carl9170/version.h
index 2282847..a0410fe 100644
--- a/drivers/net/wireless/ath/carl9170/version.h
+++ b/drivers/net/wireless/ath/carl9170/version.h
@@ -1,7 +1,7 @@
 #ifndef __CARL9170_SHARED_VERSION_H
 #define __CARL9170_SHARED_VERSION_H
-#define CARL9170FW_VERSION_YEAR 12
-#define CARL9170FW_VERSION_MONTH 12
+#define CARL9170FW_VERSION_YEAR 16
+#define CARL9170FW_VERSION_MONTH 2
 #define CARL9170FW_VERSION_DAY 15
-#define CARL9170FW_VERSION_GIT "1.9.7"
+#define CARL9170FW_VERSION_GIT "1.9.9"
 #endif /* __CARL9170_SHARED_VERSION_H */
-- 
cgit v0.10.2


From 49f86ec21c01b654f6ec47f2f4567f4f9ebaa26b Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 15 Feb 2016 16:12:07 -0600
Subject: rtlwifi: Change long delays to sleeps

Routine rtl_addr_delay() uses delay statements in code that can
sleep. To improve system responsiveness, the various delay statements
are changed.

In addition, routines rtl_rfreg_delay() and rtl_bb_delay() are
rewritten to use the code in rtl_addr_delay() for most of their
input values.

Suggested-by: Byeoungwook Kim <quddnr145@gmail.com>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c
index 16bb57c..0f48048 100644
--- a/drivers/net/wireless/realtek/rtlwifi/core.c
+++ b/drivers/net/wireless/realtek/rtlwifi/core.c
@@ -54,59 +54,39 @@ EXPORT_SYMBOL(channel5g_80m);
 void rtl_addr_delay(u32 addr)
 {
 	if (addr == 0xfe)
-		mdelay(50);
+		msleep(50);
 	else if (addr == 0xfd)
-		mdelay(5);
+		msleep(5);
 	else if (addr == 0xfc)
-		mdelay(1);
+		msleep(1);
 	else if (addr == 0xfb)
-		udelay(50);
+		usleep_range(50, 100);
 	else if (addr == 0xfa)
-		udelay(5);
+		usleep_range(5, 10);
 	else if (addr == 0xf9)
-		udelay(1);
+		usleep_range(1, 2);
 }
 EXPORT_SYMBOL(rtl_addr_delay);
 
 void rtl_rfreg_delay(struct ieee80211_hw *hw, enum radio_path rfpath, u32 addr,
 		     u32 mask, u32 data)
 {
-	if (addr == 0xfe) {
-		mdelay(50);
-	} else if (addr == 0xfd) {
-		mdelay(5);
-	} else if (addr == 0xfc) {
-		mdelay(1);
-	} else if (addr == 0xfb) {
-		udelay(50);
-	} else if (addr == 0xfa) {
-		udelay(5);
-	} else if (addr == 0xf9) {
-		udelay(1);
+	if (addr >= 0xf9 && addr <= 0xfe) {
+		rtl_addr_delay(addr);
 	} else {
 		rtl_set_rfreg(hw, rfpath, addr, mask, data);
-		udelay(1);
+		usleep_range(1, 2);
 	}
 }
 EXPORT_SYMBOL(rtl_rfreg_delay);
 
 void rtl_bb_delay(struct ieee80211_hw *hw, u32 addr, u32 data)
 {
-	if (addr == 0xfe) {
-		mdelay(50);
-	} else if (addr == 0xfd) {
-		mdelay(5);
-	} else if (addr == 0xfc) {
-		mdelay(1);
-	} else if (addr == 0xfb) {
-		udelay(50);
-	} else if (addr == 0xfa) {
-		udelay(5);
-	} else if (addr == 0xf9) {
-		udelay(1);
+	if (addr >= 0xf9 && addr <= 0xfe) {
+		rtl_addr_delay(addr);
 	} else {
 		rtl_set_bbreg(hw, addr, MASKDWORD, data);
-		udelay(1);
+		usleep_range(1, 2);
 	}
 }
 EXPORT_SYMBOL(rtl_bb_delay);
-- 
cgit v0.10.2


From a9eb0c4b73e7287074dfabd84337ea1a49976a3d Mon Sep 17 00:00:00 2001
From: Arend van Spriel <arend@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:50 +0100
Subject: brcmfmac: change function name for
 brcmf_cfg80211_wait_vif_event_timeout()

Dropping the '_timeout' from the function name as the fact that a timeout
value is passed makes it obvious a timeout is used. Also helps to keep code
lines a bit shorter and easier to stick to 80 char boundary.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index d00c5c1..5a4a0c4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -564,8 +564,8 @@ struct wireless_dev *brcmf_ap_add_vif(struct wiphy *wiphy, const char *name,
 	}
 
 	/* wait for firmware event */
-	err = brcmf_cfg80211_wait_vif_event_timeout(cfg, BRCMF_E_IF_ADD,
-						    BRCMF_VIF_EVENT_TIMEOUT);
+	err = brcmf_cfg80211_wait_vif_event(cfg, BRCMF_E_IF_ADD,
+					    BRCMF_VIF_EVENT_TIMEOUT);
 	brcmf_cfg80211_arm_vif_event(cfg, NULL);
 	if (!err) {
 		brcmf_err("timeout occurred\n");
@@ -6395,8 +6395,9 @@ bool brcmf_cfg80211_vif_event_armed(struct brcmf_cfg80211_info *cfg)
 
 	return armed;
 }
-int brcmf_cfg80211_wait_vif_event_timeout(struct brcmf_cfg80211_info *cfg,
-					  u8 action, ulong timeout)
+
+int brcmf_cfg80211_wait_vif_event(struct brcmf_cfg80211_info *cfg,
+				  u8 action, ulong timeout)
 {
 	struct brcmf_cfg80211_vif_event *event = &cfg->vif_event;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
index 40efb53..97c327d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
@@ -402,8 +402,8 @@ bool brcmf_get_vif_state_any(struct brcmf_cfg80211_info *cfg,
 void brcmf_cfg80211_arm_vif_event(struct brcmf_cfg80211_info *cfg,
 				  struct brcmf_cfg80211_vif *vif);
 bool brcmf_cfg80211_vif_event_armed(struct brcmf_cfg80211_info *cfg);
-int brcmf_cfg80211_wait_vif_event_timeout(struct brcmf_cfg80211_info *cfg,
-					  u8 action, ulong timeout);
+int brcmf_cfg80211_wait_vif_event(struct brcmf_cfg80211_info *cfg,
+				  u8 action, ulong timeout);
 s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg,
 				struct brcmf_if *ifp, bool aborted,
 				bool fw_abort);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
index 821b649..de5892a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
@@ -1988,8 +1988,8 @@ int brcmf_p2p_ifchange(struct brcmf_cfg80211_info *cfg,
 		brcmf_cfg80211_arm_vif_event(cfg, NULL);
 		return err;
 	}
-	err = brcmf_cfg80211_wait_vif_event_timeout(cfg, BRCMF_E_IF_CHANGE,
-						    BRCMF_VIF_EVENT_TIMEOUT);
+	err = brcmf_cfg80211_wait_vif_event(cfg, BRCMF_E_IF_CHANGE,
+					    BRCMF_VIF_EVENT_TIMEOUT);
 	brcmf_cfg80211_arm_vif_event(cfg, NULL);
 	if (!err)  {
 		brcmf_err("No BRCMF_E_IF_CHANGE event received\n");
@@ -2090,8 +2090,8 @@ static struct wireless_dev *brcmf_p2p_create_p2pdev(struct brcmf_p2p_info *p2p,
 	}
 
 	/* wait for firmware event */
-	err = brcmf_cfg80211_wait_vif_event_timeout(p2p->cfg, BRCMF_E_IF_ADD,
-						    BRCMF_VIF_EVENT_TIMEOUT);
+	err = brcmf_cfg80211_wait_vif_event(p2p->cfg, BRCMF_E_IF_ADD,
+					    BRCMF_VIF_EVENT_TIMEOUT);
 	brcmf_cfg80211_arm_vif_event(p2p->cfg, NULL);
 	brcmf_fweh_p2pdev_setup(pri_ifp, false);
 	if (!err) {
@@ -2180,8 +2180,8 @@ struct wireless_dev *brcmf_p2p_add_vif(struct wiphy *wiphy, const char *name,
 	}
 
 	/* wait for firmware event */
-	err = brcmf_cfg80211_wait_vif_event_timeout(cfg, BRCMF_E_IF_ADD,
-						    BRCMF_VIF_EVENT_TIMEOUT);
+	err = brcmf_cfg80211_wait_vif_event(cfg, BRCMF_E_IF_ADD,
+					    BRCMF_VIF_EVENT_TIMEOUT);
 	brcmf_cfg80211_arm_vif_event(cfg, NULL);
 	if (!err) {
 		brcmf_err("timeout occurred\n");
@@ -2272,8 +2272,8 @@ int brcmf_p2p_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev)
 	}
 	if (!err) {
 		/* wait for firmware event */
-		err = brcmf_cfg80211_wait_vif_event_timeout(cfg, BRCMF_E_IF_DEL,
-							BRCMF_VIF_EVENT_TIMEOUT);
+		err = brcmf_cfg80211_wait_vif_event(cfg, BRCMF_E_IF_DEL,
+						    BRCMF_VIF_EVENT_TIMEOUT);
 		if (!err)
 			err = -EIO;
 		else
-- 
cgit v0.10.2


From d536733442d14664437dc9c3ec993d97b5a4f1b1 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:51 +0100
Subject: brcmfmac: Limit memory allocs to <64K

Some systems have problems with allocating memory allocation larger
then 64K. Often on unload/load or suspend/resume a failure is
reported: Could not allocate wiphy device. This patch makes the
escan intermediate storage buf dynamically allocated, and smaller
than 64K.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5a4a0c4..5478128 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -1125,7 +1125,7 @@ brcmf_cfg80211_escan(struct wiphy *wiphy, struct brcmf_cfg80211_vif *vif,
 
 	/* Arm scan timeout timer */
 	mod_timer(&cfg->escan_timeout, jiffies +
-			WL_ESCAN_TIMER_INTERVAL_MS * HZ / 1000);
+			BRCMF_ESCAN_TIMER_INTERVAL_MS * HZ / 1000);
 
 	return 0;
 
@@ -3020,7 +3020,7 @@ brcmf_cfg80211_escan_handler(struct brcmf_if *ifp,
 
 		list = (struct brcmf_scan_results *)
 				cfg->escan_info.escan_buf;
-		if (bi_length > WL_ESCAN_BUF_SIZE - list->buflen) {
+		if (bi_length > BRCMF_ESCAN_BUF_SIZE - list->buflen) {
 			brcmf_err("Buffer is too small: ignoring\n");
 			goto exit;
 		}
@@ -3033,8 +3033,8 @@ brcmf_cfg80211_escan_handler(struct brcmf_if *ifp,
 							  bss_info_le))
 				goto exit;
 		}
-		memcpy(&(cfg->escan_info.escan_buf[list->buflen]),
-			bss_info_le, bi_length);
+		memcpy(&cfg->escan_info.escan_buf[list->buflen], bss_info_le,
+		       bi_length);
 		list->version = le32_to_cpu(bss_info_le->version);
 		list->buflen += bi_length;
 		list->count++;
@@ -5402,14 +5402,14 @@ static void brcmf_deinit_priv_mem(struct brcmf_cfg80211_info *cfg)
 {
 	kfree(cfg->conf);
 	cfg->conf = NULL;
-	kfree(cfg->escan_ioctl_buf);
-	cfg->escan_ioctl_buf = NULL;
 	kfree(cfg->extra_buf);
 	cfg->extra_buf = NULL;
 	kfree(cfg->wowl.nd);
 	cfg->wowl.nd = NULL;
 	kfree(cfg->wowl.nd_info);
 	cfg->wowl.nd_info = NULL;
+	kfree(cfg->escan_info.escan_buf);
+	cfg->escan_info.escan_buf = NULL;
 }
 
 static s32 brcmf_init_priv_mem(struct brcmf_cfg80211_info *cfg)
@@ -5417,9 +5417,6 @@ static s32 brcmf_init_priv_mem(struct brcmf_cfg80211_info *cfg)
 	cfg->conf = kzalloc(sizeof(*cfg->conf), GFP_KERNEL);
 	if (!cfg->conf)
 		goto init_priv_mem_out;
-	cfg->escan_ioctl_buf = kzalloc(BRCMF_DCMD_MEDLEN, GFP_KERNEL);
-	if (!cfg->escan_ioctl_buf)
-		goto init_priv_mem_out;
 	cfg->extra_buf = kzalloc(WL_EXTRA_BUF_MAX, GFP_KERNEL);
 	if (!cfg->extra_buf)
 		goto init_priv_mem_out;
@@ -5431,6 +5428,9 @@ static s32 brcmf_init_priv_mem(struct brcmf_cfg80211_info *cfg)
 				    GFP_KERNEL);
 	if (!cfg->wowl.nd_info)
 		goto init_priv_mem_out;
+	cfg->escan_info.escan_buf = kzalloc(BRCMF_ESCAN_BUF_SIZE, GFP_KERNEL);
+	if (!cfg->escan_info.escan_buf)
+		goto init_priv_mem_out;
 
 	return 0;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
index 97c327d..01f096f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
@@ -28,8 +28,11 @@
 #define WL_ROAM_TRIGGER_LEVEL		-75
 #define WL_ROAM_DELTA			20
 
-#define WL_ESCAN_BUF_SIZE		(1024 * 64)
-#define WL_ESCAN_TIMER_INTERVAL_MS	10000 /* E-Scan timeout */
+/* Keep BRCMF_ESCAN_BUF_SIZE below 64K (65536). Allocing over 64K can be
+ * problematic on some systems and should be avoided.
+ */
+#define BRCMF_ESCAN_BUF_SIZE		65000
+#define BRCMF_ESCAN_TIMER_INTERVAL_MS	10000	/* E-Scan timeout */
 
 #define WL_ESCAN_ACTION_START		1
 #define WL_ESCAN_ACTION_CONTINUE	2
@@ -205,7 +208,7 @@ enum wl_escan_state {
 
 struct escan_info {
 	u32 escan_state;
-	u8 escan_buf[WL_ESCAN_BUF_SIZE];
+	u8 *escan_buf;
 	struct wiphy *wiphy;
 	struct brcmf_if *ifp;
 	s32 (*run)(struct brcmf_cfg80211_info *cfg, struct brcmf_if *ifp,
@@ -278,7 +281,6 @@ struct brcmf_cfg80211_wowl {
  * @escan_info: escan information.
  * @escan_timeout: Timer for catch scan timeout.
  * @escan_timeout_work: scan timeout worker.
- * @escan_ioctl_buf: dongle command buffer for escan commands.
  * @vif_list: linked list of vif instances.
  * @vif_cnt: number of vif instances.
  * @vif_event: vif event signalling.
@@ -309,7 +311,6 @@ struct brcmf_cfg80211_info {
 	struct escan_info escan_info;
 	struct timer_list escan_timeout;
 	struct work_struct escan_timeout_work;
-	u8 *escan_ioctl_buf;
 	struct list_head vif_list;
 	struct brcmf_cfg80211_vif_event vif_event;
 	struct completion vif_disabled;
-- 
cgit v0.10.2


From 6ea09153b6ccbee79869537ce61f5c16f736d72b Mon Sep 17 00:00:00 2001
From: Franky Lin <frankyl@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:52 +0100
Subject: brcmfmac: check for wowl support before enumerating feature flag

In some cases wiphy->wowlan could be NULL if firmware doesn't have the
support. Driver should check for support before walking down the feature
flags.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5478128..d66f08f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6594,7 +6594,8 @@ struct brcmf_cfg80211_info *brcmf_cfg80211_attach(struct brcmf_pub *drvr,
 	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_SCAN_RANDOM_MAC)) {
 		wiphy->features |= NL80211_FEATURE_SCHED_SCAN_RANDOM_MAC_ADDR;
 #ifdef CONFIG_PM
-		if (wiphy->wowlan->flags & WIPHY_WOWLAN_NET_DETECT)
+		if (wiphy->wowlan &&
+		    wiphy->wowlan->flags & WIPHY_WOWLAN_NET_DETECT)
 			wiphy->features |= NL80211_FEATURE_ND_RANDOM_MAC_ADDR;
 #endif
 	}
-- 
cgit v0.10.2


From 73345fd212980d2e28a5c6d83801c903bd773680 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:53 +0100
Subject: brcmfmac: Configure country code using device specific settings

Country code configuration in a device is a device specific
operation. For this the country code as specified by reg notifier
(iso3166 alpha2) needs to be translated to a device specific
country locale and revision number. This patch adds this
translation and puts a placeholder in the device specific settings
where the translation table can be stored. Additional patches will
be needed to read these tables from for example device platform
data.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index d66f08f..2fd18d0 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6405,28 +6405,85 @@ int brcmf_cfg80211_wait_vif_event(struct brcmf_cfg80211_info *cfg,
 				  vif_event_equals(event, action), timeout);
 }
 
+static s32 brcmf_translate_country_code(struct brcmf_pub *drvr, char alpha2[2],
+					struct brcmf_fil_country_le *ccreq)
+{
+	struct cc_translate *country_codes;
+	struct cc_entry *cc;
+	s32 found_index;
+	int i;
+
+	country_codes = drvr->settings->country_codes;
+	if (!country_codes) {
+		brcmf_dbg(TRACE, "No country codes configured for device\n");
+		return -EINVAL;
+	}
+
+	if ((alpha2[0] == ccreq->country_abbrev[0]) &&
+	    (alpha2[1] == ccreq->country_abbrev[1])) {
+		brcmf_dbg(TRACE, "Country code already set\n");
+		return -EAGAIN;
+	}
+
+	found_index = -1;
+	for (i = 0; i < country_codes->table_size; i++) {
+		cc = &country_codes->table[i];
+		if ((cc->iso3166[0] == '\0') && (found_index == -1))
+			found_index = i;
+		if ((cc->iso3166[0] == alpha2[0]) &&
+		    (cc->iso3166[1] == alpha2[1])) {
+			found_index = i;
+			break;
+		}
+	}
+	if (found_index == -1) {
+		brcmf_dbg(TRACE, "No country code match found\n");
+		return -EINVAL;
+	}
+	memset(ccreq, 0, sizeof(*ccreq));
+	ccreq->rev = cpu_to_le32(country_codes->table[found_index].rev);
+	memcpy(ccreq->ccode, country_codes->table[found_index].cc,
+	       BRCMF_COUNTRY_BUF_SZ);
+	ccreq->country_abbrev[0] = alpha2[0];
+	ccreq->country_abbrev[1] = alpha2[1];
+	ccreq->country_abbrev[2] = 0;
+
+	return 0;
+}
+
 static void brcmf_cfg80211_reg_notifier(struct wiphy *wiphy,
 					struct regulatory_request *req)
 {
 	struct brcmf_cfg80211_info *cfg = wiphy_priv(wiphy);
 	struct brcmf_if *ifp = netdev_priv(cfg_to_ndev(cfg));
 	struct brcmf_fil_country_le ccreq;
+	s32 err;
 	int i;
 
-	brcmf_dbg(TRACE, "enter: initiator=%d, alpha=%c%c\n", req->initiator,
-		  req->alpha2[0], req->alpha2[1]);
-
 	/* ignore non-ISO3166 country codes */
 	for (i = 0; i < sizeof(req->alpha2); i++)
 		if (req->alpha2[i] < 'A' || req->alpha2[i] > 'Z') {
-			brcmf_err("not a ISO3166 code\n");
+			brcmf_err("not a ISO3166 code (0x%02x 0x%02x)\n",
+				  req->alpha2[0], req->alpha2[1]);
 			return;
 		}
-	memset(&ccreq, 0, sizeof(ccreq));
-	ccreq.rev = cpu_to_le32(-1);
-	memcpy(ccreq.ccode, req->alpha2, sizeof(req->alpha2));
-	if (brcmf_fil_iovar_data_set(ifp, "country", &ccreq, sizeof(ccreq))) {
-		brcmf_err("firmware rejected country setting\n");
+
+	brcmf_dbg(TRACE, "Enter: initiator=%d, alpha=%c%c\n", req->initiator,
+		  req->alpha2[0], req->alpha2[1]);
+
+	err = brcmf_fil_iovar_data_get(ifp, "country", &ccreq, sizeof(ccreq));
+	if (err) {
+		brcmf_err("Country code iovar returned err = %d\n", err);
+		return;
+	}
+
+	err = brcmf_translate_country_code(ifp->drvr, req->alpha2, &ccreq);
+	if (err)
+		return;
+
+	err = brcmf_fil_iovar_data_set(ifp, "country", &ccreq, sizeof(ccreq));
+	if (err) {
+		brcmf_err("Firmware rejected country setting\n");
 		return;
 	}
 	brcmf_setup_wiphybands(wiphy);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index cfee477..8199862 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -230,10 +230,8 @@ void brcmf_mp_attach(void)
 int brcmf_mp_device_attach(struct brcmf_pub *drvr)
 {
 	drvr->settings = kzalloc(sizeof(*drvr->settings), GFP_ATOMIC);
-	if (!drvr->settings) {
-		brcmf_err("Failed to alloca storage space for settings\n");
+	if (!drvr->settings)
 		return -ENOMEM;
-	}
 
 	drvr->settings->sdiod_txglomsz = brcmf_sdiod_txglomsz;
 	drvr->settings->p2p_enable = !!brcmf_p2p_enable;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index 3b0a63b..b678b2d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -15,6 +15,8 @@
 #ifndef BRCMFMAC_COMMON_H
 #define BRCMFMAC_COMMON_H
 
+#include "fwil_types.h"
+
 extern const u8 ALLFFMAC[ETH_ALEN];
 
 #define BRCMF_FW_ALTPATH_LEN			256
@@ -39,6 +41,33 @@ struct brcmf_mp_global_t {
 extern struct brcmf_mp_global_t brcmf_mp_global;
 
 /**
+ * struct cc_entry - Struct for translating user space country code (iso3166) to
+ *		     firmware country code and revision.
+ *
+ * @iso3166: iso3166 alpha 2 country code string.
+ * @cc: firmware country code string.
+ * @rev: firmware country code revision.
+ */
+struct cc_entry {
+	char	iso3166[BRCMF_COUNTRY_BUF_SZ];
+	char	cc[BRCMF_COUNTRY_BUF_SZ];
+	s32	rev;
+};
+
+/**
+ * struct cc_translate - Struct for translating country codes as set by user
+ *			 space to a country code and rev which can be used by
+ *			 firmware.
+ *
+ * @table_size: number of entries in table (> 0)
+ * @table: dynamic array of 1 or more elements with translation information.
+ */
+struct cc_translate {
+	int	table_size;
+	struct cc_entry table[0];
+};
+
+/**
  * struct brcmf_mp_device - Device module paramaters.
  *
  * @sdiod_txglomsz: SDIO txglom size.
@@ -47,6 +76,7 @@ extern struct brcmf_mp_global_t brcmf_mp_global;
  * @feature_disable: Feature_disable bitmask.
  * @fcmode: FWS flow control.
  * @roamoff: Firmware roaming off?
+ * @country_codes: If available, pointer to struct for translating country codes
  */
 struct brcmf_mp_device {
 	int	sdiod_txglomsz;
@@ -56,6 +86,7 @@ struct brcmf_mp_device {
 	int	fcmode;
 	bool	roamoff;
 	bool	ignore_probe_fail;
+	struct cc_translate *country_codes;
 };
 
 void brcmf_mp_attach(void);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
index 1afc2ad..e9e177d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
@@ -134,6 +134,8 @@
 #define BRCMF_PFN_MAC_OUI_ONLY		BIT(0)
 #define BRCMF_PFN_SET_MAC_UNASSOC	BIT(1)
 
+#define BRCMF_MCSSET_LEN		16
+
 /* join preference types for join_pref iovar */
 enum brcmf_join_pref_types {
 	BRCMF_JOIN_PREF_RSSI = 1,
@@ -279,7 +281,7 @@ struct brcmf_bss_info_le {
 	__le32 reserved32[1];	/* Reserved for expansion of BSS properties */
 	u8 flags;		/* flags */
 	u8 reserved[3];	/* Reserved for expansion of BSS properties */
-	u8 basic_mcs[MCSSET_LEN];	/* 802.11N BSS required MCS set */
+	u8 basic_mcs[BRCMF_MCSSET_LEN];	/* 802.11N BSS required MCS set */
 
 	__le16 ie_offset;	/* offset at which IEs start, from beginning */
 	__le32 ie_length;	/* byte length of Information Elements */
-- 
cgit v0.10.2


From 0aedbcaf6f182690790d98d90d5fe1e64c846c34 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:54 +0100
Subject: brcmfmac: Add length checks on firmware events

Add additional length checks on firmware events to create more
robust code.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Lei Zhang <leizh@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 2fd18d0..6cc8fdc 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -3092,6 +3092,11 @@ brcmf_notify_sched_scan_results(struct brcmf_if *ifp,
 
 	brcmf_dbg(SCAN, "Enter\n");
 
+	if (e->datalen < (sizeof(*pfn_result) + sizeof(*netinfo))) {
+		brcmf_dbg(SCAN, "Event data to small. Ignore\n");
+		return 0;
+	}
+
 	if (e->event_code == BRCMF_E_PFN_NET_LOST) {
 		brcmf_dbg(SCAN, "PFN NET LOST event. Do Nothing\n");
 		return 0;
@@ -3415,6 +3420,11 @@ brcmf_wowl_nd_results(struct brcmf_if *ifp, const struct brcmf_event_msg *e,
 
 	brcmf_dbg(SCAN, "Enter\n");
 
+	if (e->datalen < (sizeof(*pfn_result) + sizeof(*netinfo))) {
+		brcmf_dbg(SCAN, "Event data to small. Ignore\n");
+		return 0;
+	}
+
 	pfn_result = (struct brcmf_pno_scanresults_le *)data;
 
 	if (e->event_code == BRCMF_E_PFN_NET_LOST) {
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c
index 7b26fb1..d414fbb 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c
@@ -26,50 +26,6 @@
 #include "fwil.h"
 
 /**
- * struct brcm_ethhdr - broadcom specific ether header.
- *
- * @subtype: subtype for this packet.
- * @length: TODO: length of appended data.
- * @version: version indication.
- * @oui: OUI of this packet.
- * @usr_subtype: subtype for this OUI.
- */
-struct brcm_ethhdr {
-	__be16 subtype;
-	__be16 length;
-	u8 version;
-	u8 oui[3];
-	__be16 usr_subtype;
-} __packed;
-
-struct brcmf_event_msg_be {
-	__be16 version;
-	__be16 flags;
-	__be32 event_type;
-	__be32 status;
-	__be32 reason;
-	__be32 auth_type;
-	__be32 datalen;
-	u8 addr[ETH_ALEN];
-	char ifname[IFNAMSIZ];
-	u8 ifidx;
-	u8 bsscfgidx;
-} __packed;
-
-/**
- * struct brcmf_event - contents of broadcom event packet.
- *
- * @eth: standard ether header.
- * @hdr: broadcom specific ether header.
- * @msg: common part of the actual event message.
- */
-struct brcmf_event {
-	struct ethhdr eth;
-	struct brcm_ethhdr hdr;
-	struct brcmf_event_msg_be msg;
-} __packed;
-
-/**
  * struct brcmf_fweh_queue_item - event item on event queue.
  *
  * @q: list element for queuing.
@@ -85,6 +41,7 @@ struct brcmf_fweh_queue_item {
 	u8 ifidx;
 	u8 ifaddr[ETH_ALEN];
 	struct brcmf_event_msg_be emsg;
+	u32 datalen;
 	u8 data[0];
 };
 
@@ -294,6 +251,11 @@ static void brcmf_fweh_event_worker(struct work_struct *work)
 		brcmf_dbg_hex_dump(BRCMF_EVENT_ON(), event->data,
 				   min_t(u32, emsg.datalen, 64),
 				   "event payload, len=%d\n", emsg.datalen);
+		if (emsg.datalen > event->datalen) {
+			brcmf_err("event invalid length header=%d, msg=%d\n",
+				  event->datalen, emsg.datalen);
+			goto event_free;
+		}
 
 		/* special handling of interface event */
 		if (event->code == BRCMF_E_IF) {
@@ -439,7 +401,8 @@ int brcmf_fweh_activate_events(struct brcmf_if *ifp)
  * dispatch the event to a registered handler (using worker).
  */
 void brcmf_fweh_process_event(struct brcmf_pub *drvr,
-			      struct brcmf_event *event_packet)
+			      struct brcmf_event *event_packet,
+			      u32 packet_len)
 {
 	enum brcmf_fweh_event_code code;
 	struct brcmf_fweh_info *fweh = &drvr->fweh;
@@ -459,6 +422,9 @@ void brcmf_fweh_process_event(struct brcmf_pub *drvr,
 	if (code != BRCMF_E_IF && !fweh->evt_handler[code])
 		return;
 
+	if (datalen > BRCMF_DCMD_MAXLEN)
+		return;
+
 	if (in_interrupt())
 		alloc_flag = GFP_ATOMIC;
 
@@ -472,6 +438,7 @@ void brcmf_fweh_process_event(struct brcmf_pub *drvr,
 	/* use memcpy to get aligned event message */
 	memcpy(&event->emsg, &event_packet->msg, sizeof(event->emsg));
 	memcpy(event->data, data, datalen);
+	event->datalen = datalen;
 	memcpy(event->ifaddr, event_packet->eth.h_dest, ETH_ALEN);
 
 	brcmf_fweh_queue_event(fweh, event);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.h
index 5e39e2a..26ff5a9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.h
@@ -27,7 +27,6 @@
 struct brcmf_pub;
 struct brcmf_if;
 struct brcmf_cfg80211_info;
-struct brcmf_event;
 
 /* list of firmware events */
 #define BRCMF_FWEH_EVENT_ENUM_DEFLIST \
@@ -180,13 +179,55 @@ enum brcmf_fweh_event_code {
 /**
  * definitions for event packet validation.
  */
-#define BRCMF_EVENT_OUI_OFFSET		19
-#define BRCM_OUI			"\x00\x10\x18"
-#define DOT11_OUI_LEN			3
-#define BCMILCP_BCM_SUBTYPE_EVENT	1
+#define BRCM_OUI				"\x00\x10\x18"
+#define BCMILCP_BCM_SUBTYPE_EVENT		1
 
 
 /**
+ * struct brcm_ethhdr - broadcom specific ether header.
+ *
+ * @subtype: subtype for this packet.
+ * @length: TODO: length of appended data.
+ * @version: version indication.
+ * @oui: OUI of this packet.
+ * @usr_subtype: subtype for this OUI.
+ */
+struct brcm_ethhdr {
+	__be16 subtype;
+	__be16 length;
+	u8 version;
+	u8 oui[3];
+	__be16 usr_subtype;
+} __packed;
+
+struct brcmf_event_msg_be {
+	__be16 version;
+	__be16 flags;
+	__be32 event_type;
+	__be32 status;
+	__be32 reason;
+	__be32 auth_type;
+	__be32 datalen;
+	u8 addr[ETH_ALEN];
+	char ifname[IFNAMSIZ];
+	u8 ifidx;
+	u8 bsscfgidx;
+} __packed;
+
+/**
+ * struct brcmf_event - contents of broadcom event packet.
+ *
+ * @eth: standard ether header.
+ * @hdr: broadcom specific ether header.
+ * @msg: common part of the actual event message.
+ */
+struct brcmf_event {
+	struct ethhdr eth;
+	struct brcm_ethhdr hdr;
+	struct brcmf_event_msg_be msg;
+} __packed;
+
+/**
  * struct brcmf_event_msg - firmware event message.
  *
  * @version: version information.
@@ -256,34 +297,35 @@ void brcmf_fweh_unregister(struct brcmf_pub *drvr,
 			   enum brcmf_fweh_event_code code);
 int brcmf_fweh_activate_events(struct brcmf_if *ifp);
 void brcmf_fweh_process_event(struct brcmf_pub *drvr,
-			      struct brcmf_event *event_packet);
+			      struct brcmf_event *event_packet,
+			      u32 packet_len);
 void brcmf_fweh_p2pdev_setup(struct brcmf_if *ifp, bool ongoing);
 
 static inline void brcmf_fweh_process_skb(struct brcmf_pub *drvr,
 					  struct sk_buff *skb)
 {
 	struct brcmf_event *event_packet;
-	u8 *data;
 	u16 usr_stype;
 
 	/* only process events when protocol matches */
 	if (skb->protocol != cpu_to_be16(ETH_P_LINK_CTL))
 		return;
 
+	if ((skb->len + ETH_HLEN) < sizeof(*event_packet))
+		return;
+
 	/* check for BRCM oui match */
 	event_packet = (struct brcmf_event *)skb_mac_header(skb);
-	data = (u8 *)event_packet;
-	data += BRCMF_EVENT_OUI_OFFSET;
-	if (memcmp(BRCM_OUI, data, DOT11_OUI_LEN))
+	if (memcmp(BRCM_OUI, &event_packet->hdr.oui[0],
+		   sizeof(event_packet->hdr.oui)))
 		return;
 
 	/* final match on usr_subtype */
-	data += DOT11_OUI_LEN;
-	usr_stype = get_unaligned_be16(data);
+	usr_stype = get_unaligned_be16(&event_packet->hdr.usr_subtype);
 	if (usr_stype != BCMILCP_BCM_SUBTYPE_EVENT)
 		return;
 
-	brcmf_fweh_process_event(drvr, event_packet);
+	brcmf_fweh_process_event(drvr, event_packet, skb->len + ETH_HLEN);
 }
 
 #endif /* FWEH_H_ */
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
index de5892a..b5a49e5 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c
@@ -1361,6 +1361,11 @@ int brcmf_p2p_notify_action_frame_rx(struct brcmf_if *ifp,
 	u16 mgmt_type;
 	u8 action;
 
+	if (e->datalen < sizeof(*rxframe)) {
+		brcmf_dbg(SCAN, "Event data to small. Ignore\n");
+		return 0;
+	}
+
 	ch.chspec = be16_to_cpu(rxframe->chanspec);
 	cfg->d11inf.decchspec(&ch);
 	/* Check if wpa_supplicant has registered for this frame */
@@ -1858,6 +1863,11 @@ s32 brcmf_p2p_notify_rx_mgmt_p2p_probereq(struct brcmf_if *ifp,
 	brcmf_dbg(INFO, "Enter: event %d reason %d\n", e->event_code,
 		  e->reason);
 
+	if (e->datalen < sizeof(*rxframe)) {
+		brcmf_dbg(SCAN, "Event data to small. Ignore\n");
+		return 0;
+	}
+
 	ch.chspec = be16_to_cpu(rxframe->chanspec);
 	cfg->d11inf.decchspec(&ch);
 
-- 
cgit v0.10.2


From 52f22fb21764f083d697b3e35c18f37e471eba76 Mon Sep 17 00:00:00 2001
From: Franky Lin <frankyl@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:55 +0100
Subject: brcmfmac: add neighbor discovery offload ip address table
 configuration

Configure ipv6 address for neighbor discovery offload ip table in
firmware obtained through ipv6 address notification callback.

Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 6cc8fdc..71f1fdf 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -456,7 +456,7 @@ send_key_to_dongle(struct brcmf_if *ifp, struct brcmf_wsec_key *key)
 }
 
 static s32
-brcmf_configure_arp_offload(struct brcmf_if *ifp, bool enable)
+brcmf_configure_arp_nd_offload(struct brcmf_if *ifp, bool enable)
 {
 	s32 err;
 	u32 mode;
@@ -484,6 +484,15 @@ brcmf_configure_arp_offload(struct brcmf_if *ifp, bool enable)
 				  enable, mode);
 	}
 
+	err = brcmf_fil_iovar_int_set(ifp, "ndoe", enable);
+	if (err) {
+		brcmf_dbg(TRACE, "failed to configure (%d) ND offload err = %d\n",
+			  enable, err);
+		err = 0;
+	} else
+		brcmf_dbg(TRACE, "successfully configured (%d) ND offload to 0x%x\n",
+			  enable, mode);
+
 	return err;
 }
 
@@ -3543,7 +3552,7 @@ static s32 brcmf_cfg80211_resume(struct wiphy *wiphy)
 		brcmf_report_wowl_wakeind(wiphy, ifp);
 		brcmf_fil_iovar_int_set(ifp, "wowl_clear", 0);
 		brcmf_config_wowl_pattern(ifp, "clr", NULL, 0, NULL, 0);
-		brcmf_configure_arp_offload(ifp, true);
+		brcmf_configure_arp_nd_offload(ifp, true);
 		brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_PM,
 				      cfg->wowl.pre_pmmode);
 		cfg->wowl.active = false;
@@ -3567,7 +3576,7 @@ static void brcmf_configure_wowl(struct brcmf_cfg80211_info *cfg,
 
 	brcmf_dbg(TRACE, "Suspend, wowl config.\n");
 
-	brcmf_configure_arp_offload(ifp, false);
+	brcmf_configure_arp_nd_offload(ifp, false);
 	brcmf_fil_cmd_int_get(ifp, BRCMF_C_GET_PM, &cfg->wowl.pre_pmmode);
 	brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_PM, PM_MAX);
 
@@ -4336,7 +4345,7 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev,
 
 	if (!mbss) {
 		brcmf_set_mpc(ifp, 0);
-		brcmf_configure_arp_offload(ifp, false);
+		brcmf_configure_arp_nd_offload(ifp, false);
 	}
 
 	/* find the RSN_IE */
@@ -4482,7 +4491,7 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev,
 exit:
 	if ((err) && (!mbss)) {
 		brcmf_set_mpc(ifp, 1);
-		brcmf_configure_arp_offload(ifp, true);
+		brcmf_configure_arp_nd_offload(ifp, true);
 	}
 	return err;
 }
@@ -4540,7 +4549,7 @@ static int brcmf_cfg80211_stop_ap(struct wiphy *wiphy, struct net_device *ndev)
 			brcmf_err("bss_enable config failed %d\n", err);
 	}
 	brcmf_set_mpc(ifp, 1);
-	brcmf_configure_arp_offload(ifp, true);
+	brcmf_configure_arp_nd_offload(ifp, true);
 	clear_bit(BRCMF_VIF_STATUS_AP_CREATED, &ifp->vif->sme_state);
 	brcmf_net_setcarrier(ifp, false);
 
@@ -6287,7 +6296,7 @@ static s32 brcmf_config_dongle(struct brcmf_cfg80211_info *cfg)
 	if (err)
 		goto default_conf_out;
 
-	brcmf_configure_arp_offload(ifp, true);
+	brcmf_configure_arp_nd_offload(ifp, true);
 
 	cfg->dongle_up = true;
 default_conf_out:
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index ed9998b..9507cc9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -20,6 +20,8 @@
 #include <linux/inetdevice.h>
 #include <net/cfg80211.h>
 #include <net/rtnetlink.h>
+#include <net/addrconf.h>
+#include <net/ipv6.h>
 #include <brcmu_utils.h>
 #include <brcmu_wifi.h>
 
@@ -172,6 +174,35 @@ _brcmf_set_mac_address(struct work_struct *work)
 	}
 }
 
+#if IS_ENABLED(CONFIG_IPV6)
+static void _brcmf_update_ndtable(struct work_struct *work)
+{
+	struct brcmf_if *ifp;
+	int i, ret;
+
+	ifp = container_of(work, struct brcmf_if, ndoffload_work);
+
+	/* clear the table in firmware */
+	ret = brcmf_fil_iovar_data_set(ifp, "nd_hostip_clear", NULL, 0);
+	if (ret) {
+		brcmf_dbg(TRACE, "fail to clear nd ip table err:%d\n", ret);
+		return;
+	}
+
+	for (i = 0; i < ifp->ipv6addr_idx; i++) {
+		ret = brcmf_fil_iovar_data_set(ifp, "nd_hostip",
+					       &ifp->ipv6_addr_tbl[i],
+					       sizeof(struct in6_addr));
+		if (ret)
+			brcmf_err("add nd ip err %d\n", ret);
+	}
+}
+#else
+static void _brcmf_update_ndtable(struct work_struct *work)
+{
+}
+#endif
+
 static int brcmf_netdev_set_mac_address(struct net_device *ndev, void *addr)
 {
 	struct brcmf_if *ifp = netdev_priv(ndev);
@@ -685,6 +716,7 @@ int brcmf_net_attach(struct brcmf_if *ifp, bool rtnl_locked)
 
 	INIT_WORK(&ifp->setmacaddr_work, _brcmf_set_mac_address);
 	INIT_WORK(&ifp->multicast_work, _brcmf_set_multicast_list);
+	INIT_WORK(&ifp->ndoffload_work, _brcmf_update_ndtable);
 
 	if (rtnl_locked)
 		err = register_netdevice(ndev);
@@ -884,6 +916,7 @@ static void brcmf_del_if(struct brcmf_pub *drvr, s32 bsscfgidx)
 		if (ifp->ndev->netdev_ops == &brcmf_netdev_ops_pri) {
 			cancel_work_sync(&ifp->setmacaddr_work);
 			cancel_work_sync(&ifp->multicast_work);
+			cancel_work_sync(&ifp->ndoffload_work);
 		}
 		brcmf_net_detach(ifp->ndev);
 	} else {
@@ -1025,6 +1058,56 @@ static int brcmf_inetaddr_changed(struct notifier_block *nb,
 }
 #endif
 
+#if IS_ENABLED(CONFIG_IPV6)
+static int brcmf_inet6addr_changed(struct notifier_block *nb,
+				   unsigned long action, void *data)
+{
+	struct brcmf_pub *drvr = container_of(nb, struct brcmf_pub,
+					      inet6addr_notifier);
+	struct inet6_ifaddr *ifa = data;
+	struct brcmf_if *ifp;
+	int i;
+	struct in6_addr *table;
+
+	/* Only handle primary interface */
+	ifp = drvr->iflist[0];
+	if (!ifp)
+		return NOTIFY_DONE;
+	if (ifp->ndev != ifa->idev->dev)
+		return NOTIFY_DONE;
+
+	table = ifp->ipv6_addr_tbl;
+	for (i = 0; i < NDOL_MAX_ENTRIES; i++)
+		if (ipv6_addr_equal(&ifa->addr, &table[i]))
+			break;
+
+	switch (action) {
+	case NETDEV_UP:
+		if (i == NDOL_MAX_ENTRIES) {
+			if (ifp->ipv6addr_idx < NDOL_MAX_ENTRIES) {
+				table[ifp->ipv6addr_idx++] = ifa->addr;
+			} else {
+				for (i = 0; i < NDOL_MAX_ENTRIES - 1; i++)
+					table[i] = table[i + 1];
+				table[NDOL_MAX_ENTRIES - 1] = ifa->addr;
+			}
+		}
+		break;
+	case NETDEV_DOWN:
+		if (i < NDOL_MAX_ENTRIES)
+			for (; i < ifp->ipv6addr_idx; i++)
+				table[i] = table[i + 1];
+		break;
+	default:
+		break;
+	}
+
+	schedule_work(&ifp->ndoffload_work);
+
+	return NOTIFY_OK;
+}
+#endif
+
 int brcmf_attach(struct device *dev)
 {
 	struct brcmf_pub *drvr = NULL;
@@ -1164,30 +1247,41 @@ int brcmf_bus_start(struct device *dev)
 #ifdef CONFIG_INET
 	drvr->inetaddr_notifier.notifier_call = brcmf_inetaddr_changed;
 	ret = register_inetaddr_notifier(&drvr->inetaddr_notifier);
+	if (ret)
+		goto fail;
+
+#if IS_ENABLED(CONFIG_IPV6)
+	drvr->inet6addr_notifier.notifier_call = brcmf_inet6addr_changed;
+	ret = register_inet6addr_notifier(&drvr->inet6addr_notifier);
+	if (ret) {
+		unregister_inetaddr_notifier(&drvr->inetaddr_notifier);
+		goto fail;
+	}
 #endif
+#endif /* CONFIG_INET */
+
+	return 0;
 
 fail:
-	if (ret < 0) {
-		brcmf_err("failed: %d\n", ret);
-		if (drvr->config) {
-			brcmf_cfg80211_detach(drvr->config);
-			drvr->config = NULL;
-		}
-		if (drvr->fws) {
-			brcmf_fws_del_interface(ifp);
-			brcmf_fws_deinit(drvr);
-		}
-		if (ifp)
-			brcmf_net_detach(ifp->ndev);
-		if (p2p_ifp)
-			brcmf_net_detach(p2p_ifp->ndev);
-		drvr->iflist[0] = NULL;
-		drvr->iflist[1] = NULL;
-		if (brcmf_ignoring_probe_fail(drvr))
-			ret = 0;
-		return ret;
+	brcmf_err("failed: %d\n", ret);
+	if (drvr->config) {
+		brcmf_cfg80211_detach(drvr->config);
+		drvr->config = NULL;
 	}
-	return 0;
+	if (drvr->fws) {
+		brcmf_fws_del_interface(ifp);
+		brcmf_fws_deinit(drvr);
+	}
+	if (ifp)
+		brcmf_net_detach(ifp->ndev);
+	if (p2p_ifp)
+		brcmf_net_detach(p2p_ifp->ndev);
+	drvr->iflist[0] = NULL;
+	drvr->iflist[1] = NULL;
+	if (brcmf_ignoring_probe_fail(drvr))
+		ret = 0;
+
+	return ret;
 }
 
 void brcmf_bus_add_txhdrlen(struct device *dev, uint len)
@@ -1237,6 +1331,10 @@ void brcmf_detach(struct device *dev)
 	unregister_inetaddr_notifier(&drvr->inetaddr_notifier);
 #endif
 
+#if IS_ENABLED(CONFIG_IPV6)
+	unregister_inet6addr_notifier(&drvr->inet6addr_notifier);
+#endif
+
 	/* stop firmware event handling */
 	brcmf_fweh_detach(drvr);
 	if (drvr->config)
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
index 8f39435..500dac6 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
@@ -48,6 +48,8 @@
  */
 #define BRCMF_DRIVER_FIRMWARE_VERSION_LEN	32
 
+#define NDOL_MAX_ENTRIES	8
+
 /**
  * struct brcmf_ampdu_rx_reorder - AMPDU receive reorder info
  *
@@ -143,6 +145,7 @@ struct brcmf_pub {
 #endif
 
 	struct notifier_block inetaddr_notifier;
+	struct notifier_block inet6addr_notifier;
 	struct brcmf_mp_device *settings;
 };
 
@@ -175,6 +178,7 @@ enum brcmf_netif_stop_reason {
  * @stats: interface specific network statistics.
  * @setmacaddr_work: worker object for setting mac address.
  * @multicast_work: worker object for multicast provisioning.
+ * @ndoffload_work: worker object for neighbor discovery offload configuration.
  * @fws_desc: interface specific firmware-signalling descriptor.
  * @ifidx: interface index in device firmware.
  * @bsscfgidx: index of bss associated with this interface.
@@ -191,6 +195,7 @@ struct brcmf_if {
 	struct net_device_stats stats;
 	struct work_struct setmacaddr_work;
 	struct work_struct multicast_work;
+	struct work_struct ndoffload_work;
 	struct brcmf_fws_mac_descriptor *fws_desc;
 	int ifidx;
 	s32 bsscfgidx;
@@ -199,6 +204,8 @@ struct brcmf_if {
 	spinlock_t netif_stop_lock;
 	atomic_t pend_8021x_cnt;
 	wait_queue_head_t pend_8021x_wait;
+	struct in6_addr ipv6_addr_tbl[NDOL_MAX_ENTRIES];
+	u8 ipv6addr_idx;
 };
 
 struct brcmf_skb_reorder_data {
-- 
cgit v0.10.2


From cd2bc19c61b2da81ce310edb20f98e51fa7a5889 Mon Sep 17 00:00:00 2001
From: Franky Lin <frankyl@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:56 +0100
Subject: brcmfmac: check return for ARP ip setting iovar

The return value of iovar set function should be saved and checked.

Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Reviewed-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Franky Lin <frankyl@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index 9507cc9..dcf9393 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -1039,14 +1039,14 @@ static int brcmf_inetaddr_changed(struct notifier_block *nb,
 				return NOTIFY_OK;
 			}
 			for (i = 0; i < ARPOL_MAX_ENTRIES; i++) {
-				if (addr_table[i] != 0) {
-					brcmf_fil_iovar_data_set(ifp,
-						"arp_hostip", &addr_table[i],
-						sizeof(addr_table[i]));
-					if (ret)
-						brcmf_err("add arp ip err %d\n",
-							  ret);
-				}
+				if (addr_table[i] == 0)
+					continue;
+				ret = brcmf_fil_iovar_data_set(ifp, "arp_hostip",
+							       &addr_table[i],
+							       sizeof(addr_table[i]));
+				if (ret)
+					brcmf_err("add arp ip err %d\n",
+						  ret);
 			}
 		}
 		break;
-- 
cgit v0.10.2


From 6ac27689b01e23e21f08d6f55d23a94eb10a8efc Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:57 +0100
Subject: brcmfmac: use device memsize config from fw if defined

Newer type pcie devices have memory which get shared between fw and
hw. The division of this memory is done firmware compile time. As a
result the ramsize as used by driver needs to be adjusted for this.
This is done by reading the memory size from the firmware.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index d5f9ef4..d89212b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -207,6 +207,10 @@ static struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 #define BRCMF_PCIE_CFGREG_REG_BAR3_CONFIG	0x4F4
 #define BRCMF_PCIE_LINK_STATUS_CTRL_ASPM_ENAB	3
 
+/* Magic number at a magic location to find RAM size */
+#define BRCMF_RAMSIZE_MAGIC			0x534d4152	/* SMAR */
+#define BRCMF_RAMSIZE_OFFSET			0x6c
+
 
 struct brcmf_pcie_console {
 	u32 base_addr;
@@ -1412,6 +1416,28 @@ static const struct brcmf_bus_ops brcmf_pcie_bus_ops = {
 };
 
 
+static void
+brcmf_pcie_adjust_ramsize(struct brcmf_pciedev_info *devinfo, u8 *data,
+			  u32 data_len)
+{
+	__le32 *field;
+	u32 newsize;
+
+	if (data_len < BRCMF_RAMSIZE_OFFSET + 8)
+		return;
+
+	field = (__le32 *)&data[BRCMF_RAMSIZE_OFFSET];
+	if (le32_to_cpup(field) != BRCMF_RAMSIZE_MAGIC)
+		return;
+	field++;
+	newsize = le32_to_cpup(field);
+
+	brcmf_dbg(PCIE, "Found ramsize info in FW, adjusting to 0x%x\n",
+		  newsize);
+	devinfo->ci->ramsize = newsize;
+}
+
+
 static int
 brcmf_pcie_init_share_ram_info(struct brcmf_pciedev_info *devinfo,
 			       u32 sharedram_addr)
@@ -1694,6 +1720,13 @@ static void brcmf_pcie_setup(struct device *dev, const struct firmware *fw,
 
 	brcmf_pcie_attach(devinfo);
 
+	/* Some of the firmwares have the size of the memory of the device
+	 * defined inside the firmware. This is because part of the memory in
+	 * the device is shared and the devision is determined by FW. Parse
+	 * the firmware and adjust the chip memory size now.
+	 */
+	brcmf_pcie_adjust_ramsize(devinfo, (u8 *)fw->data, fw->size);
+
 	ret = brcmf_pcie_download_fw_nvram(devinfo, fw, nvram, nvram_len);
 	if (ret)
 		goto fail;
-- 
cgit v0.10.2


From 9300bf8610fd90a63a04dcc6be8223552321edad Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:58 +0100
Subject: brcmfmac: use bar1 window size as provided by pci subsystem

The PCIE bar1 window size is specified by chip. Currently the
ioremap of bar1 was using a define which always matched the size
of bar1, but newer chips can have a different bar1 sizes. With
this patch the ioremap will be called with the by chip provided
window size.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index d89212b..3fd4e3b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -72,7 +72,6 @@ static struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 
 #define BRCMF_PCIE_FW_UP_TIMEOUT		2000 /* msec */
 
-#define BRCMF_PCIE_TCM_MAP_SIZE			(4096 * 1024)
 #define BRCMF_PCIE_REG_MAP_SIZE			(32 * 1024)
 
 /* backplane addres space accessed by BAR0 */
@@ -252,7 +251,6 @@ struct brcmf_pciedev_info {
 	char nvram_name[BRCMF_FW_NAME_LEN];
 	void __iomem *regs;
 	void __iomem *tcm;
-	u32 tcm_size;
 	u32 ram_base;
 	u32 ram_size;
 	struct brcmf_chip *ci;
@@ -1592,8 +1590,7 @@ static int brcmf_pcie_get_resource(struct brcmf_pciedev_info *devinfo)
 	}
 
 	devinfo->regs = ioremap_nocache(bar0_addr, BRCMF_PCIE_REG_MAP_SIZE);
-	devinfo->tcm = ioremap_nocache(bar1_addr, BRCMF_PCIE_TCM_MAP_SIZE);
-	devinfo->tcm_size = BRCMF_PCIE_TCM_MAP_SIZE;
+	devinfo->tcm = ioremap_nocache(bar1_addr, bar1_size);
 
 	if (!devinfo->regs || !devinfo->tcm) {
 		brcmf_err("ioremap() failed (%p,%p)\n", devinfo->regs,
@@ -1602,8 +1599,9 @@ static int brcmf_pcie_get_resource(struct brcmf_pciedev_info *devinfo)
 	}
 	brcmf_dbg(PCIE, "Phys addr : reg space = %p base addr %#016llx\n",
 		  devinfo->regs, (unsigned long long)bar0_addr);
-	brcmf_dbg(PCIE, "Phys addr : mem space = %p base addr %#016llx\n",
-		  devinfo->tcm, (unsigned long long)bar1_addr);
+	brcmf_dbg(PCIE, "Phys addr : mem space = %p base addr %#016llx size 0x%x\n",
+		  devinfo->tcm, (unsigned long long)bar1_addr,
+		  (unsigned int)bar1_size);
 
 	return 0;
 }
-- 
cgit v0.10.2


From bc86fdb9ac02c77b9f55325f64fb70decc425962 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:26:59 +0100
Subject: brcmfmac: add support for the PCIE 4366c0 chip

A newer version of the 4366 PCIE chip has been released. Add
support for this version of the chip.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 3fd4e3b..3cb32b7 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -53,6 +53,7 @@ BRCMF_FW_NVRAM_DEF(4358, "brcmfmac4358-pcie.bin", "brcmfmac4358-pcie.txt");
 BRCMF_FW_NVRAM_DEF(4359, "brcmfmac4359-pcie.bin", "brcmfmac4359-pcie.txt");
 BRCMF_FW_NVRAM_DEF(4365B, "brcmfmac4365b-pcie.bin", "brcmfmac4365b-pcie.txt");
 BRCMF_FW_NVRAM_DEF(4366B, "brcmfmac4366b-pcie.bin", "brcmfmac4366b-pcie.txt");
+BRCMF_FW_NVRAM_DEF(4366C, "brcmfmac4366c-pcie.bin", "brcmfmac4366c-pcie.txt");
 BRCMF_FW_NVRAM_DEF(4371, "brcmfmac4371-pcie.bin", "brcmfmac4371-pcie.txt");
 
 static struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
@@ -66,7 +67,8 @@ static struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4358_CHIP_ID, 0xFFFFFFFF, 4358),
 	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4359_CHIP_ID, 0xFFFFFFFF, 4359),
 	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4365_CHIP_ID, 0xFFFFFFFF, 4365B),
-	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4366_CHIP_ID, 0xFFFFFFFF, 4366B),
+	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4366_CHIP_ID, 0x0000000F, 4366B),
+	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4366_CHIP_ID, 0xFFFFFFF0, 4366C),
 	BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4371_CHIP_ID, 0xFFFFFFFF, 4371),
 };
 
-- 
cgit v0.10.2


From d457a44fd85cd766c8bb48d6b542467bf89601b4 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:00 +0100
Subject: brcmfmac: remove pcie gen1 support

The PCIE bus driver supports older gen1 (v1) chips, but there is no
actual device which is using this older pcie core which is supported
by brcmfmac. Remove all gen1 related code.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 3cb32b7..1866b66 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -100,9 +100,6 @@ static struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = {
 #define BRCMF_PCIE_PCIE2REG_CONFIGDATA		0x124
 #define BRCMF_PCIE_PCIE2REG_H2D_MAILBOX		0x140
 
-#define BRCMF_PCIE_GENREV1			1
-#define BRCMF_PCIE_GENREV2			2
-
 #define BRCMF_PCIE2_INTA			0x01
 #define BRCMF_PCIE2_INTB			0x02
 
@@ -257,9 +254,7 @@ struct brcmf_pciedev_info {
 	u32 ram_size;
 	struct brcmf_chip *ci;
 	u32 coreid;
-	u32 generic_corerev;
 	struct brcmf_pcie_shared_info shared;
-	void (*ringbell)(struct brcmf_pciedev_info *devinfo);
 	wait_queue_head_t mbdata_resp_wait;
 	bool mbdata_completed;
 	bool irq_allocated;
@@ -746,68 +741,22 @@ static void brcmf_pcie_bus_console_read(struct brcmf_pciedev_info *devinfo)
 }
 
 
-static __used void brcmf_pcie_ringbell_v1(struct brcmf_pciedev_info *devinfo)
-{
-	u32 reg_value;
-
-	brcmf_dbg(PCIE, "RING !\n");
-	reg_value = brcmf_pcie_read_reg32(devinfo,
-					  BRCMF_PCIE_PCIE2REG_MAILBOXINT);
-	reg_value |= BRCMF_PCIE2_INTB;
-	brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXINT,
-			       reg_value);
-}
-
-
-static void brcmf_pcie_ringbell_v2(struct brcmf_pciedev_info *devinfo)
-{
-	brcmf_dbg(PCIE, "RING !\n");
-	/* Any arbitrary value will do, lets use 1 */
-	brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_H2D_MAILBOX, 1);
-}
-
-
 static void brcmf_pcie_intr_disable(struct brcmf_pciedev_info *devinfo)
 {
-	if (devinfo->generic_corerev == BRCMF_PCIE_GENREV1)
-		pci_write_config_dword(devinfo->pdev, BRCMF_PCIE_REG_INTMASK,
-				       0);
-	else
-		brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXMASK,
-				       0);
+	brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXMASK, 0);
 }
 
 
 static void brcmf_pcie_intr_enable(struct brcmf_pciedev_info *devinfo)
 {
-	if (devinfo->generic_corerev == BRCMF_PCIE_GENREV1)
-		pci_write_config_dword(devinfo->pdev, BRCMF_PCIE_REG_INTMASK,
-				       BRCMF_PCIE_INT_DEF);
-	else
-		brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXMASK,
-				       BRCMF_PCIE_MB_INT_D2H_DB |
-				       BRCMF_PCIE_MB_INT_FN0_0 |
-				       BRCMF_PCIE_MB_INT_FN0_1);
+	brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXMASK,
+			       BRCMF_PCIE_MB_INT_D2H_DB |
+			       BRCMF_PCIE_MB_INT_FN0_0 |
+			       BRCMF_PCIE_MB_INT_FN0_1);
 }
 
 
-static irqreturn_t brcmf_pcie_quick_check_isr_v1(int irq, void *arg)
-{
-	struct brcmf_pciedev_info *devinfo = (struct brcmf_pciedev_info *)arg;
-	u32 status;
-
-	status = 0;
-	pci_read_config_dword(devinfo->pdev, BRCMF_PCIE_REG_INTSTATUS, &status);
-	if (status) {
-		brcmf_pcie_intr_disable(devinfo);
-		brcmf_dbg(PCIE, "Enter\n");
-		return IRQ_WAKE_THREAD;
-	}
-	return IRQ_NONE;
-}
-
-
-static irqreturn_t brcmf_pcie_quick_check_isr_v2(int irq, void *arg)
+static irqreturn_t brcmf_pcie_quick_check_isr(int irq, void *arg)
 {
 	struct brcmf_pciedev_info *devinfo = (struct brcmf_pciedev_info *)arg;
 
@@ -820,29 +769,7 @@ static irqreturn_t brcmf_pcie_quick_check_isr_v2(int irq, void *arg)
 }
 
 
-static irqreturn_t brcmf_pcie_isr_thread_v1(int irq, void *arg)
-{
-	struct brcmf_pciedev_info *devinfo = (struct brcmf_pciedev_info *)arg;
-	const struct pci_dev *pdev = devinfo->pdev;
-	u32 status;
-
-	devinfo->in_irq = true;
-	status = 0;
-	pci_read_config_dword(pdev, BRCMF_PCIE_REG_INTSTATUS, &status);
-	brcmf_dbg(PCIE, "Enter %x\n", status);
-	if (status) {
-		pci_write_config_dword(pdev, BRCMF_PCIE_REG_INTSTATUS, status);
-		if (devinfo->state == BRCMFMAC_PCIE_STATE_UP)
-			brcmf_proto_msgbuf_rx_trigger(&devinfo->pdev->dev);
-	}
-	if (devinfo->state == BRCMFMAC_PCIE_STATE_UP)
-		brcmf_pcie_intr_enable(devinfo);
-	devinfo->in_irq = false;
-	return IRQ_HANDLED;
-}
-
-
-static irqreturn_t brcmf_pcie_isr_thread_v2(int irq, void *arg)
+static irqreturn_t brcmf_pcie_isr_thread(int irq, void *arg)
 {
 	struct brcmf_pciedev_info *devinfo = (struct brcmf_pciedev_info *)arg;
 	u32 status;
@@ -879,28 +806,14 @@ static int brcmf_pcie_request_irq(struct brcmf_pciedev_info *devinfo)
 	brcmf_pcie_intr_disable(devinfo);
 
 	brcmf_dbg(PCIE, "Enter\n");
-	/* is it a v1 or v2 implementation */
+
 	pci_enable_msi(pdev);
-	if (devinfo->generic_corerev == BRCMF_PCIE_GENREV1) {
-		if (request_threaded_irq(pdev->irq,
-					 brcmf_pcie_quick_check_isr_v1,
-					 brcmf_pcie_isr_thread_v1,
-					 IRQF_SHARED, "brcmf_pcie_intr",
-					 devinfo)) {
-			pci_disable_msi(pdev);
-			brcmf_err("Failed to request IRQ %d\n", pdev->irq);
-			return -EIO;
-		}
-	} else {
-		if (request_threaded_irq(pdev->irq,
-					 brcmf_pcie_quick_check_isr_v2,
-					 brcmf_pcie_isr_thread_v2,
-					 IRQF_SHARED, "brcmf_pcie_intr",
-					 devinfo)) {
-			pci_disable_msi(pdev);
-			brcmf_err("Failed to request IRQ %d\n", pdev->irq);
-			return -EIO;
-		}
+	if (request_threaded_irq(pdev->irq, brcmf_pcie_quick_check_isr,
+				 brcmf_pcie_isr_thread, IRQF_SHARED,
+				 "brcmf_pcie_intr", devinfo)) {
+		pci_disable_msi(pdev);
+		brcmf_err("Failed to request IRQ %d\n", pdev->irq);
+		return -EIO;
 	}
 	devinfo->irq_allocated = true;
 	return 0;
@@ -931,16 +844,9 @@ static void brcmf_pcie_release_irq(struct brcmf_pciedev_info *devinfo)
 	if (devinfo->in_irq)
 		brcmf_err("Still in IRQ (processing) !!!\n");
 
-	if (devinfo->generic_corerev == BRCMF_PCIE_GENREV1) {
-		status = 0;
-		pci_read_config_dword(pdev, BRCMF_PCIE_REG_INTSTATUS, &status);
-		pci_write_config_dword(pdev, BRCMF_PCIE_REG_INTSTATUS, status);
-	} else {
-		status = brcmf_pcie_read_reg32(devinfo,
-					       BRCMF_PCIE_PCIE2REG_MAILBOXINT);
-		brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXINT,
-				       status);
-	}
+	status = brcmf_pcie_read_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXINT);
+	brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_MAILBOXINT, status);
+
 	devinfo->irq_allocated = false;
 }
 
@@ -989,7 +895,9 @@ static int brcmf_pcie_ring_mb_ring_bell(void *ctx)
 	if (devinfo->state != BRCMFMAC_PCIE_STATE_UP)
 		return -EIO;
 
-	devinfo->ringbell(devinfo);
+	brcmf_dbg(PCIE, "RING !\n");
+	/* Any arbitrary value will do, lets use 1 */
+	brcmf_pcie_write_reg32(devinfo, BRCMF_PCIE_PCIE2REG_H2D_MAILBOX, 1);
 
 	return 0;
 }
@@ -1503,9 +1411,6 @@ static int brcmf_pcie_download_fw_nvram(struct brcmf_pciedev_info *devinfo,
 	u32 address;
 	u32 resetintr;
 
-	devinfo->ringbell = brcmf_pcie_ringbell_v2;
-	devinfo->generic_corerev = BRCMF_PCIE_GENREV2;
-
 	brcmf_dbg(PCIE, "Halt ARM.\n");
 	err = brcmf_pcie_enter_download_state(devinfo);
 	if (err)
-- 
cgit v0.10.2


From e9217b4b62a7296b5c008fd607579dbaf40bd754 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:01 +0100
Subject: brcmfmac: increase timeout for tx eapol

When keys get set and updated this has to happen after eapol got
transmitted (without key or old key) before the key can be updated.
To make sure the order of sending eapol and configuring key is done
correctly a timeout for tx of eapol is applied. This timeout is set
to 50 msec, which is not always enough. Especially in AP mode and
key updates the timeout may need to be much longer because client(s)
can be in powersave. Increase the timeout from 50 to 950 msec.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index dcf9393..3087d10 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -42,7 +42,7 @@ MODULE_AUTHOR("Broadcom Corporation");
 MODULE_DESCRIPTION("Broadcom 802.11 wireless LAN fullmac driver.");
 MODULE_LICENSE("Dual BSD/GPL");
 
-#define MAX_WAIT_FOR_8021X_TX			msecs_to_jiffies(50)
+#define MAX_WAIT_FOR_8021X_TX			msecs_to_jiffies(950)
 
 /* AMPDU rx reordering definitions */
 #define BRCMF_RXREORDER_FLOWID_OFFSET		0
-- 
cgit v0.10.2


From d84d99e007770b4dd9b9e8be99da043c6555e274 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:02 +0100
Subject: brcmfmac: move module init and exit to common

In preparation of module parameters for all devices the module init
and exit routines are moved to the common file.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index 8199862..b8dc68d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -28,6 +28,10 @@
 #include "tracepoint.h"
 #include "common.h"
 
+MODULE_AUTHOR("Broadcom Corporation");
+MODULE_DESCRIPTION("Broadcom 802.11 wireless LAN fullmac driver.");
+MODULE_LICENSE("Dual BSD/GPL");
+
 const u8 ALLFFMAC[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
 
 #define BRCMF_DEFAULT_SCAN_CHANNEL_TIME	40
@@ -221,7 +225,7 @@ void __brcmf_dbg(u32 level, const char *func, const char *fmt, ...)
 }
 #endif
 
-void brcmf_mp_attach(void)
+static void brcmf_mp_attach(void)
 {
 	strlcpy(brcmf_mp_global.firmware_path, brcmf_firmware_path,
 		BRCMF_FW_ALTPATH_LEN);
@@ -249,3 +253,33 @@ void brcmf_mp_device_detach(struct brcmf_pub *drvr)
 	kfree(drvr->settings);
 }
 
+static int __init brcmfmac_module_init(void)
+{
+	int err;
+
+	/* Initialize debug system first */
+	brcmf_debugfs_init();
+
+#ifdef CONFIG_BRCMFMAC_SDIO
+	brcmf_sdio_init();
+#endif
+	/* Initialize global module paramaters */
+	brcmf_mp_attach();
+
+	/* Continue the initialization by registering the different busses */
+	err = brcmf_core_init();
+	if (err)
+		brcmf_debugfs_exit();
+
+	return err;
+}
+
+static void __exit brcmfmac_module_exit(void)
+{
+	brcmf_core_exit();
+	brcmf_debugfs_exit();
+}
+
+module_init(brcmfmac_module_init);
+module_exit(brcmfmac_module_exit);
+
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index b678b2d..256be1b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -89,7 +89,6 @@ struct brcmf_mp_device {
 	struct cc_translate *country_codes;
 };
 
-void brcmf_mp_attach(void);
 int brcmf_mp_device_attach(struct brcmf_pub *drvr);
 void brcmf_mp_device_detach(struct brcmf_pub *drvr);
 #ifdef DEBUG
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index 3087d10..7b24e2b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -38,10 +38,6 @@
 #include "pcie.h"
 #include "common.h"
 
-MODULE_AUTHOR("Broadcom Corporation");
-MODULE_DESCRIPTION("Broadcom 802.11 wireless LAN fullmac driver.");
-MODULE_LICENSE("Dual BSD/GPL");
-
 #define MAX_WAIT_FOR_8021X_TX			msecs_to_jiffies(950)
 
 /* AMPDU rx reordering definitions */
@@ -1422,19 +1418,15 @@ static void brcmf_driver_register(struct work_struct *work)
 }
 static DECLARE_WORK(brcmf_driver_work, brcmf_driver_register);
 
-static int __init brcmfmac_module_init(void)
+int __init brcmf_core_init(void)
 {
-	brcmf_debugfs_init();
-#ifdef CONFIG_BRCMFMAC_SDIO
-	brcmf_sdio_init();
-#endif
 	if (!schedule_work(&brcmf_driver_work))
 		return -EBUSY;
 
 	return 0;
 }
 
-static void __exit brcmfmac_module_exit(void)
+void __exit brcmf_core_exit(void)
 {
 	cancel_work_sync(&brcmf_driver_work);
 
@@ -1447,8 +1439,5 @@ static void __exit brcmfmac_module_exit(void)
 #ifdef CONFIG_BRCMFMAC_PCIE
 	brcmf_pcie_exit();
 #endif
-	brcmf_debugfs_exit();
 }
 
-module_init(brcmfmac_module_init);
-module_exit(brcmfmac_module_exit);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
index 500dac6..7bdb6fe 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h
@@ -227,5 +227,7 @@ void brcmf_txflowblock_if(struct brcmf_if *ifp,
 void brcmf_txfinalize(struct brcmf_if *ifp, struct sk_buff *txp, bool success);
 void brcmf_netif_rx(struct brcmf_if *ifp, struct sk_buff *skb);
 void brcmf_net_setcarrier(struct brcmf_if *ifp, bool on);
+int __init brcmf_core_init(void);
+void __exit brcmf_core_exit(void);
 
 #endif /* BRCMFMAC_CORE_H */
-- 
cgit v0.10.2


From 5c22fb85102a751e5a305d8fd13a1856a725bf01 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:03 +0100
Subject: brcmfmac: add wowl gtk rekeying offload support

This patch adds support for gtk rekeying offload and for gtk
rekeying failure during wowl mode.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 71f1fdf..6d849bb 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -3526,6 +3526,10 @@ static void brcmf_report_wowl_wakeind(struct wiphy *wiphy, struct brcmf_if *ifp)
 			else
 				wakeup_data.net_detect = cfg->wowl.nd_info;
 		}
+		if (wakeind & BRCMF_WOWL_GTK_FAILURE) {
+			brcmf_dbg(INFO, "WOWL Wake indicator: BRCMF_WOWL_GTK_FAILURE\n");
+			wakeup_data.gtk_rekey_failure = true;
+		}
 	} else {
 		wakeup = NULL;
 	}
@@ -3607,6 +3611,8 @@ static void brcmf_configure_wowl(struct brcmf_cfg80211_info *cfg,
 		brcmf_fweh_register(cfg->pub, BRCMF_E_PFN_NET_FOUND,
 				    brcmf_wowl_nd_results);
 	}
+	if (wowl->gtk_rekey_failure)
+		wowl_config |= BRCMF_WOWL_GTK_FAILURE;
 	if (!test_bit(BRCMF_VIF_STATUS_CONNECTED, &ifp->vif->sme_state))
 		wowl_config |= BRCMF_WOWL_UNASSOC;
 
@@ -4874,7 +4880,32 @@ static int brcmf_cfg80211_tdls_oper(struct wiphy *wiphy,
 	return ret;
 }
 
-static struct cfg80211_ops wl_cfg80211_ops = {
+#ifdef CONFIG_PM
+static int
+brcmf_cfg80211_set_rekey_data(struct wiphy *wiphy, struct net_device *ndev,
+			      struct cfg80211_gtk_rekey_data *gtk)
+{
+	struct brcmf_if *ifp = netdev_priv(ndev);
+	struct brcmf_gtk_keyinfo_le gtk_le;
+	int ret;
+
+	brcmf_dbg(TRACE, "Enter, bssidx=%d\n", ifp->bsscfgidx);
+
+	memcpy(gtk_le.kck, gtk->kck, sizeof(gtk_le.kck));
+	memcpy(gtk_le.kek, gtk->kek, sizeof(gtk_le.kek));
+	memcpy(gtk_le.replay_counter, gtk->replay_ctr,
+	       sizeof(gtk_le.replay_counter));
+
+	ret = brcmf_fil_iovar_data_set(ifp, "gtk_key_info", &gtk_le,
+				       sizeof(gtk_le));
+	if (ret < 0)
+		brcmf_err("gtk_key_info iovar failed: ret=%d\n", ret);
+
+	return ret;
+}
+#endif
+
+static struct cfg80211_ops brcmf_cfg80211_ops = {
 	.add_virtual_intf = brcmf_cfg80211_add_iface,
 	.del_virtual_intf = brcmf_cfg80211_del_iface,
 	.change_virtual_intf = brcmf_cfg80211_change_iface,
@@ -6139,19 +6170,18 @@ static void brcmf_wiphy_wowl_params(struct wiphy *wiphy, struct brcmf_if *ifp)
 {
 #ifdef CONFIG_PM
 	struct brcmf_cfg80211_info *cfg = wiphy_to_cfg(wiphy);
-	s32 err;
-	u32 wowl_cap;
 
 	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_PNO)) {
-		err = brcmf_fil_iovar_int_get(ifp, "wowl_cap", &wowl_cap);
-		if (!err) {
-			if (wowl_cap & BRCMF_WOWL_PFN_FOUND) {
-				brcmf_wowlan_support.flags |=
-							WIPHY_WOWLAN_NET_DETECT;
-				init_waitqueue_head(&cfg->wowl.nd_data_wait);
-			}
+		if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL_ND)) {
+			brcmf_wowlan_support.flags |= WIPHY_WOWLAN_NET_DETECT;
+			init_waitqueue_head(&cfg->wowl.nd_data_wait);
 		}
 	}
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL_GTK)) {
+		brcmf_wowlan_support.flags |= WIPHY_WOWLAN_SUPPORTS_GTK_REKEY;
+		brcmf_wowlan_support.flags |= WIPHY_WOWLAN_GTK_REKEY_FAILURE;
+	}
+
 	wiphy->wowlan = &brcmf_wowlan_support;
 #endif
 }
@@ -6538,6 +6568,7 @@ struct brcmf_cfg80211_info *brcmf_cfg80211_attach(struct brcmf_pub *drvr,
 	struct net_device *ndev = brcmf_get_ifp(drvr, 0)->ndev;
 	struct brcmf_cfg80211_info *cfg;
 	struct wiphy *wiphy;
+	struct cfg80211_ops *ops;
 	struct brcmf_cfg80211_vif *vif;
 	struct brcmf_if *ifp;
 	s32 err = 0;
@@ -6549,8 +6580,17 @@ struct brcmf_cfg80211_info *brcmf_cfg80211_attach(struct brcmf_pub *drvr,
 		return NULL;
 	}
 
+	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
+	if (!ops)
+		return NULL;
+
+	memcpy(ops, &brcmf_cfg80211_ops, sizeof(*ops));
 	ifp = netdev_priv(ndev);
-	wiphy = wiphy_new(&wl_cfg80211_ops, sizeof(struct brcmf_cfg80211_info));
+#ifdef CONFIG_PM
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL_GTK))
+		ops->set_rekey_data = brcmf_cfg80211_set_rekey_data;
+#endif
+	wiphy = wiphy_new(ops, sizeof(struct brcmf_cfg80211_info));
 	if (!wiphy) {
 		brcmf_err("Could not allocate wiphy device\n");
 		return NULL;
@@ -6560,6 +6600,7 @@ struct brcmf_cfg80211_info *brcmf_cfg80211_attach(struct brcmf_pub *drvr,
 
 	cfg = wiphy_priv(wiphy);
 	cfg->wiphy = wiphy;
+	cfg->ops = ops;
 	cfg->pub = drvr;
 	init_vif_event(&cfg->vif_event);
 	INIT_LIST_HEAD(&cfg->vif_list);
@@ -6686,6 +6727,7 @@ priv_out:
 	ifp->vif = NULL;
 wiphy_out:
 	brcmf_free_wiphy(wiphy);
+	kfree(ops);
 	return NULL;
 }
 
@@ -6696,6 +6738,7 @@ void brcmf_cfg80211_detach(struct brcmf_cfg80211_info *cfg)
 
 	brcmf_btcoex_detach(cfg);
 	wiphy_unregister(cfg->wiphy);
+	kfree(cfg->ops);
 	wl_deinit_priv(cfg);
 	brcmf_free_wiphy(cfg->wiphy);
 }
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
index 01f096f..e4c1d43 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
@@ -256,6 +256,7 @@ struct brcmf_cfg80211_wowl {
  * struct brcmf_cfg80211_info - dongle private data of cfg80211 interface
  *
  * @wiphy: wiphy object for cfg80211 interface.
+ * @ops: pointer to copy of ops as registered with wiphy object.
  * @conf: dongle configuration.
  * @p2p: peer-to-peer specific information.
  * @btcoex: Bluetooth coexistence information.
@@ -288,6 +289,7 @@ struct brcmf_cfg80211_wowl {
  */
 struct brcmf_cfg80211_info {
 	struct wiphy *wiphy;
+	struct cfg80211_ops *ops;
 	struct brcmf_cfg80211_conf *conf;
 	struct brcmf_p2p_info p2p;
 	struct brcmf_btcoex_info *btcoex;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
index 1ffa95f..098732a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
@@ -136,6 +136,7 @@ void brcmf_feat_attach(struct brcmf_pub *drvr)
 {
 	struct brcmf_if *ifp = brcmf_get_ifp(drvr, 0);
 	struct brcmf_pno_macaddr_le pfn_mac;
+	u32 wowl_cap;
 	s32 err;
 
 	brcmf_feat_firmware_capabilities(ifp);
@@ -143,6 +144,17 @@ void brcmf_feat_attach(struct brcmf_pub *drvr)
 	brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_PNO, "pfn");
 	if (drvr->bus_if->wowl_supported)
 		brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_WOWL, "wowl");
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL)) {
+		err = brcmf_fil_iovar_int_get(ifp, "wowl_cap", &wowl_cap);
+		if (!err) {
+			if (wowl_cap & BRCMF_WOWL_PFN_FOUND)
+				ifp->drvr->feat_flags |=
+					BIT(BRCMF_FEAT_WOWL_ND);
+			if (wowl_cap & BRCMF_WOWL_GTK_FAILURE)
+				ifp->drvr->feat_flags |=
+					BIT(BRCMF_FEAT_WOWL_GTK);
+		}
+	}
 	/* MBSS does not work for 43362 */
 	if (drvr->bus_if->chip == BRCM_CC_43362_CHIP_ID)
 		ifp->drvr->feat_flags &= ~BIT(BRCMF_FEAT_MBSS);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
index 2e2479d..f940c29 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
@@ -27,6 +27,8 @@
  * RSDB: Real Simultaneous Dual Band
  * TDLS: Tunneled Direct Link Setup
  * SCAN_RANDOM_MAC: Random MAC during (net detect) scheduled scan.
+ * WOWL_ND: WOWL net detect (PNO)
+ * WOWL_GTK: (WOWL) GTK rekeying offload
  */
 #define BRCMF_FEAT_LIST \
 	BRCMF_FEAT_DEF(MBSS) \
@@ -36,7 +38,9 @@
 	BRCMF_FEAT_DEF(P2P) \
 	BRCMF_FEAT_DEF(RSDB) \
 	BRCMF_FEAT_DEF(TDLS) \
-	BRCMF_FEAT_DEF(SCAN_RANDOM_MAC)
+	BRCMF_FEAT_DEF(SCAN_RANDOM_MAC) \
+	BRCMF_FEAT_DEF(WOWL_ND) \
+	BRCMF_FEAT_DEF(WOWL_GTK)
 
 /*
  * Quirks:
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
index e9e177d..6d41ae3 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
@@ -111,7 +111,9 @@
 /* Wakeup if received matched secured pattern: */
 #define BRCMF_WOWL_SECURE		(1 << 25)
 /* Wakeup on finding preferred network */
-#define BRCMF_WOWL_PFN_FOUND		(1 << 26)
+#define BRCMF_WOWL_PFN_FOUND		(1 << 27)
+/* Wakeup on receiving pairwise key EAP packets: */
+#define WIPHY_WOWL_EAP_PK		(1 << 28)
 /* Link Down indication in WoWL mode: */
 #define BRCMF_WOWL_LINKDOWN		(1 << 31)
 
@@ -136,6 +138,10 @@
 
 #define BRCMF_MCSSET_LEN		16
 
+#define BRCMF_RSN_KCK_LENGTH		16
+#define BRCMF_RSN_KEK_LENGTH		16
+#define BRCMF_RSN_REPLAY_LEN		8
+
 /* join preference types for join_pref iovar */
 enum brcmf_join_pref_types {
 	BRCMF_JOIN_PREF_RSSI = 1,
@@ -789,4 +795,17 @@ struct brcmf_pktcnt_le {
 	__le32 rx_ocast_good_pkt;
 };
 
+/**
+ * struct brcmf_gtk_keyinfo_le - GTP rekey data
+ *
+ * @kck: key confirmation key.
+ * @kek: key encryption key.
+ * @replay_counter: replay counter.
+ */
+struct brcmf_gtk_keyinfo_le {
+	u8 kck[BRCMF_RSN_KCK_LENGTH];
+	u8 kek[BRCMF_RSN_KEK_LENGTH];
+	u8 replay_counter[BRCMF_RSN_REPLAY_LEN];
+};
+
 #endif /* FWIL_TYPES_H_ */
-- 
cgit v0.10.2


From 8ea56be0869f8230ed7b2779397225fe45080dd1 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:04 +0100
Subject: brcmfmac: move platform data retrieval code to common

In preparation of module parameters for all devices the module
platform data retrieval is moved from sdio to common. It is still
only used for sdio devices.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index b98db8a..25cd712 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -27,8 +27,6 @@
 #include <linux/mmc/sdio_func.h>
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
-#include <linux/platform_device.h>
-#include <linux/platform_data/brcmfmac-sdio.h>
 #include <linux/pm_runtime.h>
 #include <linux/suspend.h>
 #include <linux/errno.h>
@@ -46,7 +44,6 @@
 #include "bus.h"
 #include "debug.h"
 #include "sdio.h"
-#include "of.h"
 #include "core.h"
 #include "common.h"
 
@@ -106,18 +103,18 @@ static void brcmf_sdiod_dummy_irqhandler(struct sdio_func *func)
 
 int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 {
+	struct brcmfmac_sdio_platform_data *pdata;
 	int ret = 0;
 	u8 data;
 	u32 addr, gpiocontrol;
 	unsigned long flags;
 
-	if ((sdiodev->pdata) && (sdiodev->pdata->oob_irq_supported)) {
+	pdata = sdiodev->pdata;
+	if ((pdata) && (pdata->oob_irq_supported)) {
 		brcmf_dbg(SDIO, "Enter, register OOB IRQ %d\n",
-			  sdiodev->pdata->oob_irq_nr);
-		ret = request_irq(sdiodev->pdata->oob_irq_nr,
-				  brcmf_sdiod_oob_irqhandler,
-				  sdiodev->pdata->oob_irq_flags,
-				  "brcmf_oob_intr",
+			  pdata->oob_irq_nr);
+		ret = request_irq(pdata->oob_irq_nr, brcmf_sdiod_oob_irqhandler,
+				  pdata->oob_irq_flags, "brcmf_oob_intr",
 				  &sdiodev->func[1]->dev);
 		if (ret != 0) {
 			brcmf_err("request_irq failed %d\n", ret);
@@ -129,7 +126,7 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 		sdiodev->irq_en = true;
 		spin_unlock_irqrestore(&sdiodev->irq_en_lock, flags);
 
-		ret = enable_irq_wake(sdiodev->pdata->oob_irq_nr);
+		ret = enable_irq_wake(pdata->oob_irq_nr);
 		if (ret != 0) {
 			brcmf_err("enable_irq_wake failed %d\n", ret);
 			return ret;
@@ -158,7 +155,7 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 
 		/* redirect, configure and enable io for interrupt signal */
 		data = SDIO_SEPINT_MASK | SDIO_SEPINT_OE;
-		if (sdiodev->pdata->oob_irq_flags & IRQF_TRIGGER_HIGH)
+		if (pdata->oob_irq_flags & IRQF_TRIGGER_HIGH)
 			data |= SDIO_SEPINT_ACT_HI;
 		brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_BRCM_SEPINT, data, &ret);
 
@@ -176,9 +173,12 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 
 int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev)
 {
+	struct brcmfmac_sdio_platform_data *pdata;
+
 	brcmf_dbg(SDIO, "Entering\n");
 
-	if ((sdiodev->pdata) && (sdiodev->pdata->oob_irq_supported)) {
+	pdata = sdiodev->pdata;
+	if ((pdata) && (pdata->oob_irq_supported)) {
 		sdio_claim_host(sdiodev->func[1]);
 		brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_BRCM_SEPINT, 0, NULL);
 		brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_IENx, 0, NULL);
@@ -187,11 +187,10 @@ int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev)
 		if (sdiodev->oob_irq_requested) {
 			sdiodev->oob_irq_requested = false;
 			if (sdiodev->irq_wake) {
-				disable_irq_wake(sdiodev->pdata->oob_irq_nr);
+				disable_irq_wake(pdata->oob_irq_nr);
 				sdiodev->irq_wake = false;
 			}
-			free_irq(sdiodev->pdata->oob_irq_nr,
-				 &sdiodev->func[1]->dev);
+			free_irq(pdata->oob_irq_nr, &sdiodev->func[1]->dev);
 			sdiodev->irq_en = false;
 		}
 	} else {
@@ -1103,8 +1102,6 @@ static const struct sdio_device_id brcmf_sdmmc_ids[] = {
 };
 MODULE_DEVICE_TABLE(sdio, brcmf_sdmmc_ids);
 
-static struct brcmfmac_sdio_platform_data *brcmfmac_sdio_pdata;
-
 
 static void brcmf_sdiod_acpi_set_power_manageable(struct device *dev,
 						  int val)
@@ -1167,10 +1164,7 @@ static int brcmf_ops_sdio_probe(struct sdio_func *func,
 	dev_set_drvdata(&func->dev, bus_if);
 	dev_set_drvdata(&sdiodev->func[1]->dev, bus_if);
 	sdiodev->dev = &sdiodev->func[1]->dev;
-	sdiodev->pdata = brcmfmac_sdio_pdata;
-
-	if (!sdiodev->pdata)
-		brcmf_of_probe(sdiodev);
+	sdiodev->pdata = brcmf_get_module_param(sdiodev->dev);
 
 #ifdef CONFIG_PM_SLEEP
 	/* wowl can be supported when KEEP_POWER is true and (WAKE_SDIO_IRQ
@@ -1296,7 +1290,7 @@ static const struct dev_pm_ops brcmf_sdio_pm_ops = {
 static struct sdio_driver brcmf_sdmmc_driver = {
 	.probe = brcmf_ops_sdio_probe,
 	.remove = brcmf_ops_sdio_remove,
-	.name = BRCMFMAC_SDIO_PDATA_NAME,
+	.name = KBUILD_MODNAME,
 	.id_table = brcmf_sdmmc_ids,
 	.drv = {
 		.owner = THIS_MODULE,
@@ -1306,37 +1300,6 @@ static struct sdio_driver brcmf_sdmmc_driver = {
 	},
 };
 
-static int __init brcmf_sdio_pd_probe(struct platform_device *pdev)
-{
-	brcmf_dbg(SDIO, "Enter\n");
-
-	brcmfmac_sdio_pdata = dev_get_platdata(&pdev->dev);
-
-	if (brcmfmac_sdio_pdata->power_on)
-		brcmfmac_sdio_pdata->power_on();
-
-	return 0;
-}
-
-static int brcmf_sdio_pd_remove(struct platform_device *pdev)
-{
-	brcmf_dbg(SDIO, "Enter\n");
-
-	if (brcmfmac_sdio_pdata->power_off)
-		brcmfmac_sdio_pdata->power_off();
-
-	sdio_unregister_driver(&brcmf_sdmmc_driver);
-
-	return 0;
-}
-
-static struct platform_driver brcmf_sdio_pd = {
-	.remove		= brcmf_sdio_pd_remove,
-	.driver		= {
-		.name	= BRCMFMAC_SDIO_PDATA_NAME,
-	}
-};
-
 void brcmf_sdio_register(void)
 {
 	int ret;
@@ -1350,19 +1313,6 @@ void brcmf_sdio_exit(void)
 {
 	brcmf_dbg(SDIO, "Enter\n");
 
-	if (brcmfmac_sdio_pdata)
-		platform_driver_unregister(&brcmf_sdio_pd);
-	else
-		sdio_unregister_driver(&brcmf_sdmmc_driver);
+	sdio_unregister_driver(&brcmf_sdmmc_driver);
 }
 
-void __init brcmf_sdio_init(void)
-{
-	int ret;
-
-	brcmf_dbg(SDIO, "Enter\n");
-
-	ret = platform_driver_probe(&brcmf_sdio_pd, brcmf_sdio_pd_probe);
-	if (ret == -ENODEV)
-		brcmf_dbg(SDIO, "No platform data available.\n");
-}
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index b8dc68d..020901c 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -27,6 +27,7 @@
 #include "fwil_types.h"
 #include "tracepoint.h"
 #include "common.h"
+#include "of.h"
 
 MODULE_AUTHOR("Broadcom Corporation");
 MODULE_DESCRIPTION("Broadcom 802.11 wireless LAN fullmac driver.");
@@ -79,6 +80,7 @@ module_param_named(ignore_probe_fail, brcmf_ignore_probe_fail, int, 0);
 MODULE_PARM_DESC(ignore_probe_fail, "always succeed probe for debugging");
 #endif
 
+static struct brcmfmac_sdio_platform_data *brcmfmac_pdata;
 struct brcmf_mp_global_t brcmf_mp_global;
 
 int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
@@ -231,6 +233,13 @@ static void brcmf_mp_attach(void)
 		BRCMF_FW_ALTPATH_LEN);
 }
 
+struct brcmfmac_sdio_platform_data *brcmf_get_module_param(struct device *dev)
+{
+	if (!brcmfmac_pdata)
+		brcmf_of_probe(dev, &brcmfmac_pdata);
+	return brcmfmac_pdata;
+}
+
 int brcmf_mp_device_attach(struct brcmf_pub *drvr)
 {
 	drvr->settings = kzalloc(sizeof(*drvr->settings), GFP_ATOMIC);
@@ -253,6 +262,35 @@ void brcmf_mp_device_detach(struct brcmf_pub *drvr)
 	kfree(drvr->settings);
 }
 
+static int __init brcmf_common_pd_probe(struct platform_device *pdev)
+{
+	brcmf_dbg(INFO, "Enter\n");
+
+	brcmfmac_pdata = dev_get_platdata(&pdev->dev);
+
+	if (brcmfmac_pdata->power_on)
+		brcmfmac_pdata->power_on();
+
+	return 0;
+}
+
+static int brcmf_common_pd_remove(struct platform_device *pdev)
+{
+	brcmf_dbg(INFO, "Enter\n");
+
+	if (brcmfmac_pdata->power_off)
+		brcmfmac_pdata->power_off();
+
+	return 0;
+}
+
+static struct platform_driver brcmf_pd = {
+	.remove		= brcmf_common_pd_remove,
+	.driver		= {
+		.name	= BRCMFMAC_SDIO_PDATA_NAME,
+	}
+};
+
 static int __init brcmfmac_module_init(void)
 {
 	int err;
@@ -260,16 +298,21 @@ static int __init brcmfmac_module_init(void)
 	/* Initialize debug system first */
 	brcmf_debugfs_init();
 
-#ifdef CONFIG_BRCMFMAC_SDIO
-	brcmf_sdio_init();
-#endif
+	/* Get the platform data (if available) for our devices */
+	err = platform_driver_probe(&brcmf_pd, brcmf_common_pd_probe);
+	if (err == -ENODEV)
+		brcmf_dbg(INFO, "No platform data available.\n");
+
 	/* Initialize global module paramaters */
 	brcmf_mp_attach();
 
 	/* Continue the initialization by registering the different busses */
 	err = brcmf_core_init();
-	if (err)
+	if (err) {
 		brcmf_debugfs_exit();
+		if (brcmfmac_pdata)
+			platform_driver_unregister(&brcmf_pd);
+	}
 
 	return err;
 }
@@ -277,6 +320,8 @@ static int __init brcmfmac_module_init(void)
 static void __exit brcmfmac_module_exit(void)
 {
 	brcmf_core_exit();
+	if (brcmfmac_pdata)
+		platform_driver_unregister(&brcmf_pd);
 	brcmf_debugfs_exit();
 }
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index 256be1b..54a26ed 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -15,6 +15,8 @@
 #ifndef BRCMFMAC_COMMON_H
 #define BRCMFMAC_COMMON_H
 
+#include <linux/platform_device.h>
+#include <linux/platform_data/brcmfmac-sdio.h>
 #include "fwil_types.h"
 
 extern const u8 ALLFFMAC[ETH_ALEN];
@@ -89,6 +91,7 @@ struct brcmf_mp_device {
 	struct cc_translate *country_codes;
 };
 
+struct brcmfmac_sdio_platform_data *brcmf_get_module_param(struct device *dev);
 int brcmf_mp_device_attach(struct brcmf_pub *drvr);
 void brcmf_mp_device_detach(struct brcmf_pub *drvr);
 #ifdef DEBUG
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index 03f35e0..8201d93 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -16,17 +16,16 @@
 #include <linux/init.h>
 #include <linux/of.h>
 #include <linux/of_irq.h>
-#include <linux/mmc/card.h>
-#include <linux/platform_data/brcmfmac-sdio.h>
-#include <linux/mmc/sdio_func.h>
 
 #include <defs.h>
 #include "debug.h"
-#include "sdio.h"
+#include "core.h"
+#include "common.h"
+#include "of.h"
 
-void brcmf_of_probe(struct brcmf_sdio_dev *sdiodev)
+void
+brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_platform_data **sdio)
 {
-	struct device *dev = sdiodev->dev;
 	struct device_node *np = dev->of_node;
 	int irq;
 	u32 irqf;
@@ -35,12 +34,12 @@ void brcmf_of_probe(struct brcmf_sdio_dev *sdiodev)
 	if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
 		return;
 
-	sdiodev->pdata = devm_kzalloc(dev, sizeof(*sdiodev->pdata), GFP_KERNEL);
-	if (!sdiodev->pdata)
+	*sdio = devm_kzalloc(dev, sizeof(*sdio), GFP_KERNEL);
+	if (!(*sdio))
 		return;
 
 	if (of_property_read_u32(np, "brcm,drive-strength", &val) == 0)
-		sdiodev->pdata->drive_strength = val;
+		(*sdio)->drive_strength = val;
 
 	/* make sure there are interrupts defined in the node */
 	if (!of_find_property(np, "interrupts", NULL))
@@ -53,7 +52,7 @@ void brcmf_of_probe(struct brcmf_sdio_dev *sdiodev)
 	}
 	irqf = irqd_get_trigger_type(irq_get_irq_data(irq));
 
-	sdiodev->pdata->oob_irq_supported = true;
-	sdiodev->pdata->oob_irq_nr = irq;
-	sdiodev->pdata->oob_irq_flags = irqf;
+	(*sdio)->oob_irq_supported = true;
+	(*sdio)->oob_irq_nr = irq;
+	(*sdio)->oob_irq_flags = irqf;
 }
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
index 5f7c355..84b4744 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
@@ -14,9 +14,11 @@
  * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
 #ifdef CONFIG_OF
-void brcmf_of_probe(struct brcmf_sdio_dev *sdiodev);
+void
+brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_platform_data **sdio);
 #else
-static void brcmf_of_probe(struct brcmf_sdio_dev *sdiodev)
+static void brcmf_of_probe(struct device *dev,
+			   struct brcmfmac_sdio_platform_data **sdio)
 {
 }
 #endif /* CONFIG_OF */
-- 
cgit v0.10.2


From 73ef9e640e94ab6205c0bb92dc8bac53a40d952e Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:05 +0100
Subject: brcmfmac: keep ARP and ND offload enabled during WOWL

Currently ARP and ND (IPv6 Neigbor Discovery) offload get disabled
on entering suspend. However when firmwares support the wowl_cap
iovar then these offload routines can be kept enabled as they
will work during WOWL as well.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 6d849bb..5609a79 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -3556,7 +3556,8 @@ static s32 brcmf_cfg80211_resume(struct wiphy *wiphy)
 		brcmf_report_wowl_wakeind(wiphy, ifp);
 		brcmf_fil_iovar_int_set(ifp, "wowl_clear", 0);
 		brcmf_config_wowl_pattern(ifp, "clr", NULL, 0, NULL, 0);
-		brcmf_configure_arp_nd_offload(ifp, true);
+		if (!brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL_ARP_ND))
+			brcmf_configure_arp_nd_offload(ifp, true);
 		brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_PM,
 				      cfg->wowl.pre_pmmode);
 		cfg->wowl.active = false;
@@ -3580,7 +3581,8 @@ static void brcmf_configure_wowl(struct brcmf_cfg80211_info *cfg,
 
 	brcmf_dbg(TRACE, "Suspend, wowl config.\n");
 
-	brcmf_configure_arp_nd_offload(ifp, false);
+	if (!brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL_ARP_ND))
+		brcmf_configure_arp_nd_offload(ifp, false);
 	brcmf_fil_cmd_int_get(ifp, BRCMF_C_GET_PM, &cfg->wowl.pre_pmmode);
 	brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_PM, PM_MAX);
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
index 098732a..bfa19bb 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
@@ -147,6 +147,7 @@ void brcmf_feat_attach(struct brcmf_pub *drvr)
 	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL)) {
 		err = brcmf_fil_iovar_int_get(ifp, "wowl_cap", &wowl_cap);
 		if (!err) {
+			ifp->drvr->feat_flags |= BIT(BRCMF_FEAT_WOWL_ARP_ND);
 			if (wowl_cap & BRCMF_WOWL_PFN_FOUND)
 				ifp->drvr->feat_flags |=
 					BIT(BRCMF_FEAT_WOWL_ND);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
index f940c29..a3efa35 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
@@ -29,6 +29,7 @@
  * SCAN_RANDOM_MAC: Random MAC during (net detect) scheduled scan.
  * WOWL_ND: WOWL net detect (PNO)
  * WOWL_GTK: (WOWL) GTK rekeying offload
+ * WOWL_ARP_ND: ARP and Neighbor Discovery offload support during WOWL.
  */
 #define BRCMF_FEAT_LIST \
 	BRCMF_FEAT_DEF(MBSS) \
@@ -40,7 +41,8 @@
 	BRCMF_FEAT_DEF(TDLS) \
 	BRCMF_FEAT_DEF(SCAN_RANDOM_MAC) \
 	BRCMF_FEAT_DEF(WOWL_ND) \
-	BRCMF_FEAT_DEF(WOWL_GTK)
+	BRCMF_FEAT_DEF(WOWL_GTK) \
+	BRCMF_FEAT_DEF(WOWL_ARP_ND)
 
 /*
  * Quirks:
-- 
cgit v0.10.2


From 4d7928959832ea41f7f91456b76da19cad01bd09 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:07 +0100
Subject: brcmfmac: switch to new platform data

Platform data is only available for sdio. With this patch a new
platform data structure is being used which allows for platform
data for any device and configurable per device. This patch only
switches to the new structure and adds support for SDIO devices.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index 25cd712..bb4aece 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -103,7 +103,7 @@ static void brcmf_sdiod_dummy_irqhandler(struct sdio_func *func)
 
 int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 {
-	struct brcmfmac_sdio_platform_data *pdata;
+	struct brcmfmac_sdio_pd *pdata;
 	int ret = 0;
 	u8 data;
 	u32 addr, gpiocontrol;
@@ -173,7 +173,7 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 
 int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev)
 {
-	struct brcmfmac_sdio_platform_data *pdata;
+	struct brcmfmac_sdio_pd *pdata;
 
 	brcmf_dbg(SDIO, "Entering\n");
 
@@ -1164,17 +1164,6 @@ static int brcmf_ops_sdio_probe(struct sdio_func *func,
 	dev_set_drvdata(&func->dev, bus_if);
 	dev_set_drvdata(&sdiodev->func[1]->dev, bus_if);
 	sdiodev->dev = &sdiodev->func[1]->dev;
-	sdiodev->pdata = brcmf_get_module_param(sdiodev->dev);
-
-#ifdef CONFIG_PM_SLEEP
-	/* wowl can be supported when KEEP_POWER is true and (WAKE_SDIO_IRQ
-	 * is true or when platform data OOB irq is true).
-	 */
-	if ((sdio_get_host_pm_caps(sdiodev->func[1]) & MMC_PM_KEEP_POWER) &&
-	    ((sdio_get_host_pm_caps(sdiodev->func[1]) & MMC_PM_WAKE_SDIO_IRQ) ||
-	     (sdiodev->pdata && sdiodev->pdata->oob_irq_supported)))
-		bus_if->wowl_supported = true;
-#endif
 
 	brcmf_sdiod_change_state(sdiodev, BRCMF_SDIOD_DOWN);
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5609a79..5e3acac 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -6459,8 +6459,8 @@ int brcmf_cfg80211_wait_vif_event(struct brcmf_cfg80211_info *cfg,
 static s32 brcmf_translate_country_code(struct brcmf_pub *drvr, char alpha2[2],
 					struct brcmf_fil_country_le *ccreq)
 {
-	struct cc_translate *country_codes;
-	struct cc_entry *cc;
+	struct brcmfmac_pd_cc *country_codes;
+	struct brcmfmac_pd_cc_entry *cc;
 	s32 found_index;
 	int i;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index 020901c..4bd3225 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -80,7 +80,7 @@ module_param_named(ignore_probe_fail, brcmf_ignore_probe_fail, int, 0);
 MODULE_PARM_DESC(ignore_probe_fail, "always succeed probe for debugging");
 #endif
 
-static struct brcmfmac_sdio_platform_data *brcmfmac_pdata;
+static struct brcmfmac_platform_data *brcmfmac_pdata;
 struct brcmf_mp_global_t brcmf_mp_global;
 
 int brcmf_c_preinit_dcmds(struct brcmf_if *ifp)
@@ -229,15 +229,46 @@ void __brcmf_dbg(u32 level, const char *func, const char *fmt, ...)
 
 static void brcmf_mp_attach(void)
 {
+	/* If module param firmware path is set then this will always be used,
+	 * if not set then if available use the platform data version. To make
+	 * sure it gets initialized at all, always copy the module param version
+	 */
 	strlcpy(brcmf_mp_global.firmware_path, brcmf_firmware_path,
 		BRCMF_FW_ALTPATH_LEN);
+	if ((brcmfmac_pdata) && (brcmfmac_pdata->fw_alternative_path) &&
+	    (brcmf_mp_global.firmware_path[0] == '\0')) {
+		strlcpy(brcmf_mp_global.firmware_path,
+			brcmfmac_pdata->fw_alternative_path,
+			BRCMF_FW_ALTPATH_LEN);
+	}
 }
 
-struct brcmfmac_sdio_platform_data *brcmf_get_module_param(struct device *dev)
+struct brcmfmac_sdio_pd *brcmf_get_module_param(struct device *dev,
+						enum brcmf_bus_type bus_type,
+						u32 chip, u32 chiprev)
 {
-	if (!brcmfmac_pdata)
-		brcmf_of_probe(dev, &brcmfmac_pdata);
-	return brcmfmac_pdata;
+	struct brcmfmac_sdio_pd *pdata;
+	struct brcmfmac_pd_device *device_pd;
+	int i;
+
+	if (brcmfmac_pdata) {
+		for (i = 0; i < brcmfmac_pdata->device_count; i++) {
+			device_pd = &brcmfmac_pdata->devices[i];
+			if ((device_pd->bus_type == bus_type) &&
+			    (device_pd->id == chip) &&
+			    ((device_pd->rev == chiprev) ||
+			     (device_pd->rev == -1))) {
+				brcmf_dbg(INFO, "Platform data for device found\n");
+				if (device_pd->bus_type == BRCMF_BUSTYPE_SDIO)
+					return &device_pd->bus.sdio;
+				break;
+			}
+		}
+	}
+	pdata = NULL;
+	brcmf_of_probe(dev, &pdata);
+
+	return pdata;
 }
 
 int brcmf_mp_device_attach(struct brcmf_pub *drvr)
@@ -287,7 +318,7 @@ static int brcmf_common_pd_remove(struct platform_device *pdev)
 static struct platform_driver brcmf_pd = {
 	.remove		= brcmf_common_pd_remove,
 	.driver		= {
-		.name	= BRCMFMAC_SDIO_PDATA_NAME,
+		.name	= BRCMFMAC_PDATA_NAME,
 	}
 };
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index 54a26ed..a64e40e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -16,7 +16,7 @@
 #define BRCMFMAC_COMMON_H
 
 #include <linux/platform_device.h>
-#include <linux/platform_data/brcmfmac-sdio.h>
+#include <linux/platform_data/brcmfmac.h>
 #include "fwil_types.h"
 
 extern const u8 ALLFFMAC[ETH_ALEN];
@@ -43,33 +43,6 @@ struct brcmf_mp_global_t {
 extern struct brcmf_mp_global_t brcmf_mp_global;
 
 /**
- * struct cc_entry - Struct for translating user space country code (iso3166) to
- *		     firmware country code and revision.
- *
- * @iso3166: iso3166 alpha 2 country code string.
- * @cc: firmware country code string.
- * @rev: firmware country code revision.
- */
-struct cc_entry {
-	char	iso3166[BRCMF_COUNTRY_BUF_SZ];
-	char	cc[BRCMF_COUNTRY_BUF_SZ];
-	s32	rev;
-};
-
-/**
- * struct cc_translate - Struct for translating country codes as set by user
- *			 space to a country code and rev which can be used by
- *			 firmware.
- *
- * @table_size: number of entries in table (> 0)
- * @table: dynamic array of 1 or more elements with translation information.
- */
-struct cc_translate {
-	int	table_size;
-	struct cc_entry table[0];
-};
-
-/**
  * struct brcmf_mp_device - Device module paramaters.
  *
  * @sdiod_txglomsz: SDIO txglom size.
@@ -88,10 +61,12 @@ struct brcmf_mp_device {
 	int	fcmode;
 	bool	roamoff;
 	bool	ignore_probe_fail;
-	struct cc_translate *country_codes;
+	struct brcmfmac_pd_cc *country_codes;
 };
 
-struct brcmfmac_sdio_platform_data *brcmf_get_module_param(struct device *dev);
+struct brcmfmac_sdio_pd *brcmf_get_module_param(struct device *dev,
+						enum brcmf_bus_type bus_type,
+						u32 chip, u32 chiprev);
 int brcmf_mp_device_attach(struct brcmf_pub *drvr);
 void brcmf_mp_device_detach(struct brcmf_pub *drvr);
 #ifdef DEBUG
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index 8201d93..ece0b65 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -23,8 +23,7 @@
 #include "common.h"
 #include "of.h"
 
-void
-brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_platform_data **sdio)
+void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio)
 {
 	struct device_node *np = dev->of_node;
 	int irq;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
index 84b4744..1ba951f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
@@ -15,10 +15,9 @@
  */
 #ifdef CONFIG_OF
 void
-brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_platform_data **sdio);
+brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio);
 #else
-static void brcmf_of_probe(struct device *dev,
-			   struct brcmfmac_sdio_platform_data **sdio)
+static void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio)
 {
 }
 #endif /* CONFIG_OF */
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index c790fa8..6e36704 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -33,8 +33,6 @@
 #include <linux/bcma/bcma.h>
 #include <linux/debugfs.h>
 #include <linux/vmalloc.h>
-#include <linux/platform_data/brcmfmac-sdio.h>
-#include <linux/moduleparam.h>
 #include <asm/unaligned.h>
 #include <defs.h>
 #include <brcmu_wifi.h>
@@ -44,6 +42,8 @@
 #include "sdio.h"
 #include "chip.h"
 #include "firmware.h"
+#include "core.h"
+#include "common.h"
 
 #define DCMD_RESP_TIMEOUT	msecs_to_jiffies(2500)
 #define CTL_DONE_TIMEOUT	msecs_to_jiffies(2500)
@@ -3775,26 +3775,28 @@ static const struct brcmf_buscore_ops brcmf_sdio_buscore_ops = {
 static bool
 brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 {
+	struct brcmf_sdio_dev *sdiodev;
 	u8 clkctl = 0;
 	int err = 0;
 	int reg_addr;
 	u32 reg_val;
 	u32 drivestrength;
 
-	sdio_claim_host(bus->sdiodev->func[1]);
+	sdiodev = bus->sdiodev;
+	sdio_claim_host(sdiodev->func[1]);
 
 	pr_debug("F1 signature read @0x18000000=0x%4x\n",
-		 brcmf_sdiod_regrl(bus->sdiodev, SI_ENUM_BASE, NULL));
+		 brcmf_sdiod_regrl(sdiodev, SI_ENUM_BASE, NULL));
 
 	/*
 	 * Force PLL off until brcmf_chip_attach()
 	 * programs PLL control regs
 	 */
 
-	brcmf_sdiod_regwb(bus->sdiodev, SBSDIO_FUNC1_CHIPCLKCSR,
+	brcmf_sdiod_regwb(sdiodev, SBSDIO_FUNC1_CHIPCLKCSR,
 			  BRCMF_INIT_CLKCTL1, &err);
 	if (!err)
-		clkctl = brcmf_sdiod_regrb(bus->sdiodev,
+		clkctl = brcmf_sdiod_regrb(sdiodev,
 					   SBSDIO_FUNC1_CHIPCLKCSR, &err);
 
 	if (err || ((clkctl & ~SBSDIO_AVBITS) != BRCMF_INIT_CLKCTL1)) {
@@ -3803,50 +3805,77 @@ brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 		goto fail;
 	}
 
-	bus->ci = brcmf_chip_attach(bus->sdiodev, &brcmf_sdio_buscore_ops);
+	bus->ci = brcmf_chip_attach(sdiodev, &brcmf_sdio_buscore_ops);
 	if (IS_ERR(bus->ci)) {
 		brcmf_err("brcmf_chip_attach failed!\n");
 		bus->ci = NULL;
 		goto fail;
 	}
+	sdiodev->pdata = brcmf_get_module_param(sdiodev->dev,
+						   BRCMF_BUSTYPE_SDIO,
+						   bus->ci->chip,
+						   bus->ci->chiprev);
+	/* platform specific configuration:
+	 *   alignments must be at least 4 bytes for ADMA
+	 */
+	bus->head_align = ALIGNMENT;
+	bus->sgentry_align = ALIGNMENT;
+	if (sdiodev->pdata) {
+		if (sdiodev->pdata->sd_head_align > ALIGNMENT)
+			bus->head_align = sdiodev->pdata->sd_head_align;
+		if (sdiodev->pdata->sd_sgentry_align > ALIGNMENT)
+			bus->sgentry_align = sdiodev->pdata->sd_sgentry_align;
+	}
+	/* allocate scatter-gather table. sg support
+	 * will be disabled upon allocation failure.
+	 */
+	brcmf_sdiod_sgtable_alloc(sdiodev);
+
+#ifdef CONFIG_PM_SLEEP
+	/* wowl can be supported when KEEP_POWER is true and (WAKE_SDIO_IRQ
+	 * is true or when platform data OOB irq is true).
+	 */
+	if ((sdio_get_host_pm_caps(sdiodev->func[1]) & MMC_PM_KEEP_POWER) &&
+	    ((sdio_get_host_pm_caps(sdiodev->func[1]) & MMC_PM_WAKE_SDIO_IRQ) ||
+	     (sdiodev->pdata && sdiodev->pdata->oob_irq_supported)))
+		sdiodev->bus_if->wowl_supported = true;
+#endif
 
 	if (brcmf_sdio_kso_init(bus)) {
 		brcmf_err("error enabling KSO\n");
 		goto fail;
 	}
 
-	if ((bus->sdiodev->pdata) && (bus->sdiodev->pdata->drive_strength))
-		drivestrength = bus->sdiodev->pdata->drive_strength;
+	if ((sdiodev->pdata) && (sdiodev->pdata->drive_strength))
+		drivestrength = sdiodev->pdata->drive_strength;
 	else
 		drivestrength = DEFAULT_SDIO_DRIVE_STRENGTH;
-	brcmf_sdio_drivestrengthinit(bus->sdiodev, bus->ci, drivestrength);
+	brcmf_sdio_drivestrengthinit(sdiodev, bus->ci, drivestrength);
 
 	/* Set card control so an SDIO card reset does a WLAN backplane reset */
-	reg_val = brcmf_sdiod_regrb(bus->sdiodev,
-				    SDIO_CCCR_BRCM_CARDCTRL, &err);
+	reg_val = brcmf_sdiod_regrb(sdiodev, SDIO_CCCR_BRCM_CARDCTRL, &err);
 	if (err)
 		goto fail;
 
 	reg_val |= SDIO_CCCR_BRCM_CARDCTRL_WLANRESET;
 
-	brcmf_sdiod_regwb(bus->sdiodev,
-			  SDIO_CCCR_BRCM_CARDCTRL, reg_val, &err);
+	brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_BRCM_CARDCTRL, reg_val, &err);
 	if (err)
 		goto fail;
 
 	/* set PMUControl so a backplane reset does PMU state reload */
 	reg_addr = CORE_CC_REG(brcmf_chip_get_pmu(bus->ci)->base, pmucontrol);
-	reg_val = brcmf_sdiod_regrl(bus->sdiodev, reg_addr, &err);
+	reg_val = brcmf_sdiod_regrl(sdiodev, reg_addr, &err);
 	if (err)
 		goto fail;
 
 	reg_val |= (BCMA_CC_PMU_CTL_RES_RELOAD << BCMA_CC_PMU_CTL_RES_SHIFT);
 
-	brcmf_sdiod_regwl(bus->sdiodev, reg_addr, reg_val, &err);
+	brcmf_sdiod_regwl(sdiodev, reg_addr, reg_val, &err);
 	if (err)
 		goto fail;
 
-	sdio_release_host(bus->sdiodev->func[1]);
+	sdio_release_host(sdiodev->func[1]);
 
 	brcmu_pktq_init(&bus->txq, (PRIOMASK + 1), TXQLEN);
 
@@ -3867,7 +3896,7 @@ brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 	return true;
 
 fail:
-	sdio_release_host(bus->sdiodev->func[1]);
+	sdio_release_host(sdiodev->func[1]);
 	return false;
 }
 
@@ -4045,18 +4074,6 @@ struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 	bus->txminmax = BRCMF_TXMINMAX;
 	bus->tx_seq = SDPCM_SEQ_WRAP - 1;
 
-	/* platform specific configuration:
-	 *   alignments must be at least 4 bytes for ADMA
-	 */
-	bus->head_align = ALIGNMENT;
-	bus->sgentry_align = ALIGNMENT;
-	if (sdiodev->pdata) {
-		if (sdiodev->pdata->sd_head_align > ALIGNMENT)
-			bus->head_align = sdiodev->pdata->sd_head_align;
-		if (sdiodev->pdata->sd_sgentry_align > ALIGNMENT)
-			bus->sgentry_align = sdiodev->pdata->sd_sgentry_align;
-	}
-
 	/* single-threaded workqueue */
 	wq = alloc_ordered_workqueue("brcmf_wq/%s", WQ_MEM_RECLAIM,
 				     dev_name(&sdiodev->func[1]->dev));
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
index 23f2231..50df9cb 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
@@ -184,7 +184,7 @@ struct brcmf_sdio_dev {
 	struct brcmf_sdio *bus;
 	struct device *dev;
 	struct brcmf_bus *bus_if;
-	struct brcmfmac_sdio_platform_data *pdata;
+	struct brcmfmac_sdio_pd *pdata;
 	bool oob_irq_requested;
 	bool irq_en;			/* irq enable flags */
 	spinlock_t irq_en_lock;
diff --git a/include/linux/platform_data/brcmfmac-sdio.h b/include/linux/platform_data/brcmfmac-sdio.h
deleted file mode 100644
index e75dcbf..0000000
--- a/include/linux/platform_data/brcmfmac-sdio.h
+++ /dev/null
@@ -1,135 +0,0 @@
-/*
- * Copyright (c) 2013 Broadcom Corporation
- *
- * Permission to use, copy, modify, and/or distribute this software for any
- * purpose with or without fee is hereby granted, provided that the above
- * copyright notice and this permission notice appear in all copies.
- *
- * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
- * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
- * SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
- * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
- * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
- */
-
-#ifndef _LINUX_BRCMFMAC_PLATFORM_H
-#define _LINUX_BRCMFMAC_PLATFORM_H
-
-/*
- * Platform specific driver functions and data. Through the platform specific
- * device data functions can be provided to help the brcmfmac driver to
- * operate with the device in combination with the used platform.
- *
- * Use the platform data in the following (similar) way:
- *
- *
-#include <brcmfmac_platform.h>
-
-
-static void brcmfmac_power_on(void)
-{
-}
-
-static void brcmfmac_power_off(void)
-{
-}
-
-static void brcmfmac_reset(void)
-{
-}
-
-static struct brcmfmac_sdio_platform_data brcmfmac_sdio_pdata = {
-	.power_on		= brcmfmac_power_on,
-	.power_off		= brcmfmac_power_off,
-	.reset			= brcmfmac_reset
-};
-
-static struct platform_device brcmfmac_device = {
-	.name			= BRCMFMAC_SDIO_PDATA_NAME,
-	.id			= PLATFORM_DEVID_NONE,
-	.dev.platform_data	= &brcmfmac_sdio_pdata
-};
-
-void __init brcmfmac_init_pdata(void)
-{
-	brcmfmac_sdio_pdata.oob_irq_supported = true;
-	brcmfmac_sdio_pdata.oob_irq_nr = gpio_to_irq(GPIO_BRCMF_SDIO_OOB);
-	brcmfmac_sdio_pdata.oob_irq_flags = IORESOURCE_IRQ |
-					    IORESOURCE_IRQ_HIGHLEVEL;
-	platform_device_register(&brcmfmac_device);
-}
- *
- *
- * Note: the brcmfmac can be loaded as module or be statically built-in into
- * the kernel. If built-in then do note that it uses module_init (and
- * module_exit) routines which equal device_initcall. So if you intend to
- * create a module with the platform specific data for the brcmfmac and have
- * it built-in to the kernel then use a higher initcall then device_initcall
- * (see init.h). If this is not done then brcmfmac will load without problems
- * but will not pickup the platform data.
- *
- * When the driver does not "detect" platform driver data then it will continue
- * without reporting anything and just assume there is no data needed. Which is
- * probably true for most platforms.
- *
- * Explanation of the platform_data fields:
- *
- * drive_strength: is the preferred drive_strength to be used for the SDIO
- * pins. If 0 then a default value will be used. This is the target drive
- * strength, the exact drive strength which will be used depends on the
- * capabilities of the device.
- *
- * oob_irq_supported: does the board have support for OOB interrupts. SDIO
- * in-band interrupts are relatively slow and for having less overhead on
- * interrupt processing an out of band interrupt can be used. If the HW
- * supports this then enable this by setting this field to true and configure
- * the oob related fields.
- *
- * oob_irq_nr, oob_irq_flags: the OOB interrupt information. The values are
- * used for registering the irq using request_irq function.
- *
- * broken_sg_support: flag for broken sg list support of SDIO host controller.
- * Set this to true if the SDIO host controller has higher align requirement
- * than 32 bytes for each scatterlist item.
- *
- * sd_head_align: alignment requirement for start of data buffer
- *
- * sd_sgentry_align: length alignment requirement for each sg entry
- *
- * power_on: This function is called by the brcmfmac when the module gets
- * loaded. This can be particularly useful for low power devices. The platform
- * spcific routine may for example decide to power up the complete device.
- * If there is no use-case for this function then provide NULL.
- *
- * power_off: This function is called by the brcmfmac when the module gets
- * unloaded. At this point the device can be powered down or otherwise be reset.
- * So if an actual power_off is not supported but reset is then reset the device
- * when this function gets called. This can be particularly useful for low power
- * devices. If there is no use-case for this function (either power-down or
- * reset) then provide NULL.
- *
- * reset: This function can get called if the device communication broke down.
- * This functionality is particularly useful in case of SDIO type devices. It is
- * possible to reset a dongle via sdio data interface, but it requires that
- * this is fully functional. This function is chip/module specific and this
- * function should return only after the complete reset has completed.
- */
-
-#define BRCMFMAC_SDIO_PDATA_NAME	"brcmfmac_sdio"
-
-struct brcmfmac_sdio_platform_data {
-	unsigned int drive_strength;
-	bool oob_irq_supported;
-	unsigned int oob_irq_nr;
-	unsigned long oob_irq_flags;
-	bool broken_sg_support;
-	unsigned short sd_head_align;
-	unsigned short sd_sgentry_align;
-	void (*power_on)(void);
-	void (*power_off)(void);
-	void (*reset)(void);
-};
-
-#endif /* _LINUX_BRCMFMAC_PLATFORM_H */
diff --git a/include/linux/platform_data/brcmfmac.h b/include/linux/platform_data/brcmfmac.h
new file mode 100644
index 0000000..1d30bf2
--- /dev/null
+++ b/include/linux/platform_data/brcmfmac.h
@@ -0,0 +1,185 @@
+/*
+ * Copyright (c) 201 Broadcom Corporation
+ *
+ * Permission to use, copy, modify, and/or distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
+ * SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
+ * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
+ * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+
+#ifndef _LINUX_BRCMFMAC_PLATFORM_H
+#define _LINUX_BRCMFMAC_PLATFORM_H
+
+
+#define BRCMFMAC_PDATA_NAME		"brcmfmac"
+
+#define BRCMFMAC_COUNTRY_BUF_SZ		4
+
+
+/*
+ * Platform specific driver functions and data. Through the platform specific
+ * device data functions and data can be provided to help the brcmfmac driver to
+ * operate with the device in combination with the used platform.
+ */
+
+
+/**
+ * Note: the brcmfmac can be loaded as module or be statically built-in into
+ * the kernel. If built-in then do note that it uses module_init (and
+ * module_exit) routines which equal device_initcall. So if you intend to
+ * create a module with the platform specific data for the brcmfmac and have
+ * it built-in to the kernel then use a higher initcall then device_initcall
+ * (see init.h). If this is not done then brcmfmac will load without problems
+ * but will not pickup the platform data.
+ *
+ * When the driver does not "detect" platform driver data then it will continue
+ * without reporting anything and just assume there is no data needed. Which is
+ * probably true for most platforms.
+ */
+
+/**
+ * enum brcmf_bus_type - Bus type identifier. Currently SDIO, USB and PCIE are
+ *			 supported.
+ */
+enum brcmf_bus_type {
+	BRCMF_BUSTYPE_SDIO,
+	BRCMF_BUSTYPE_USB,
+	BRCMF_BUSTYPE_PCIE
+};
+
+
+/**
+ * struct brcmfmac_sdio_pd - SDIO Device specific platform data.
+ *
+ * @txglomsz:		SDIO txglom size. Use 0 if default of driver is to be
+ *			used.
+ * @drive_strength:	is the preferred drive_strength to be used for the SDIO
+ *			pins. If 0 then a default value will be used. This is
+ *			the target drive strength, the exact drive strength
+ *			which will be used depends on the capabilities of the
+ *			device.
+ * @oob_irq_supported:	does the board have support for OOB interrupts. SDIO
+ *			in-band interrupts are relatively slow and for having
+ *			less overhead on interrupt processing an out of band
+ *			interrupt can be used. If the HW supports this then
+ *			enable this by setting this field to true and configure
+ *			the oob related fields.
+ * @oob_irq_nr,
+ * @oob_irq_flags:	the OOB interrupt information. The values are used for
+ *			registering the irq using request_irq function.
+ * @broken_sg_support:	flag for broken sg list support of SDIO host controller.
+ *			Set this to true if the SDIO host controller has higher
+ *			align requirement than 32 bytes for each scatterlist
+ *			item.
+ * @sd_head_align:	alignment requirement for start of data buffer.
+ * @sd_sgentry_align:	length alignment requirement for each sg entry.
+ * @reset:		This function can get called if the device communication
+ *			broke down. This functionality is particularly useful in
+ *			case of SDIO type devices. It is possible to reset a
+ *			dongle via sdio data interface, but it requires that
+ *			this is fully functional. This function is chip/module
+ *			specific and this function should return only after the
+ *			complete reset has completed.
+ */
+struct brcmfmac_sdio_pd {
+	int		txglomsz;
+	unsigned int	drive_strength;
+	bool		oob_irq_supported;
+	unsigned int	oob_irq_nr;
+	unsigned long	oob_irq_flags;
+	bool		broken_sg_support;
+	unsigned short	sd_head_align;
+	unsigned short	sd_sgentry_align;
+	void		(*reset)(void);
+};
+
+/**
+ * struct brcmfmac_pd_cc_entry - Struct for translating user space country code
+ *				 (iso3166) to firmware country code and
+ *				 revision.
+ *
+ * @iso3166:	iso3166 alpha 2 country code string.
+ * @cc:		firmware country code string.
+ * @rev:	firmware country code revision.
+ */
+struct brcmfmac_pd_cc_entry {
+	char	iso3166[BRCMFMAC_COUNTRY_BUF_SZ];
+	char	cc[BRCMFMAC_COUNTRY_BUF_SZ];
+	s32	rev;
+};
+
+/**
+ * struct brcmfmac_pd_cc - Struct for translating country codes as set by user
+ *			   space to a country code and rev which can be used by
+ *			   firmware.
+ *
+ * @table_size:	number of entries in table (> 0)
+ * @table:	array of 1 or more elements with translation information.
+ */
+struct brcmfmac_pd_cc {
+	int				table_size;
+	struct brcmfmac_pd_cc_entry	table[0];
+};
+
+/**
+ * struct brcmfmac_pd_device - Device specific platform data. (id/rev/bus_type)
+ *			       is the unique identifier of the device.
+ *
+ * @id:			ID of the device for which this data is. In case of SDIO
+ *			or PCIE this is the chipid as identified by chip.c In
+ *			case of USB this is the chipid as identified by the
+ *			device query.
+ * @rev:		chip revision, see id.
+ * @bus_type:		The type of bus. Some chipid/rev exist for different bus
+ *			types. Each bus type has its own set of settings.
+ * @feature_disable:	Bitmask of features to disable (override), See feature.c
+ *			in brcmfmac for details.
+ * @country_codes:	If available, pointer to struct for translating country
+ *			codes.
+ * @bus:		Bus specific (union) device settings. Currently only
+ *			SDIO.
+ */
+struct brcmfmac_pd_device {
+	unsigned int		id;
+	unsigned int		rev;
+	enum brcmf_bus_type	bus_type;
+	unsigned int		feature_disable;
+	struct brcmfmac_pd_cc	*country_codes;
+	union {
+		struct brcmfmac_sdio_pd sdio;
+	} bus;
+};
+
+/**
+ * struct brcmfmac_platform_data - BRCMFMAC specific platform data.
+ *
+ * @power_on:	This function is called by the brcmfmac driver when the module
+ *		gets loaded. This can be particularly useful for low power
+ *		devices. The platform spcific routine may for example decide to
+ *		power up the complete device. If there is no use-case for this
+ *		function then provide NULL.
+ * @power_off:	This function is called by the brcmfmac when the module gets
+ *		unloaded. At this point the devices can be powered down or
+ *		otherwise be reset. So if an actual power_off is not supported
+ *		but reset is supported by the devices then reset the devices
+ *		when this function gets called. This can be particularly useful
+ *		for low power devices. If there is no use-case for this
+ *		function then provide NULL.
+ */
+struct brcmfmac_platform_data {
+	void	(*power_on)(void);
+	void	(*power_off)(void);
+	char	*fw_alternative_path;
+	int	device_count;
+	struct brcmfmac_pd_device devices[0];
+};
+
+
+#endif /* _LINUX_BRCMFMAC_PLATFORM_H */
-- 
cgit v0.10.2


From af5b5e62f72e08e6e7e3d3cd78ee7e4cec96f7c8 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:08 +0100
Subject: brcmfmac: merge platform data and module paramaters

Merge module parameters and platform data in one struct. This is the
last step to move to the new platform data per device. Now parameters
of platform data will be merged with module parameters per device.

Reviewed-by: Arend Van Spriel <arend@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Signed-off-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index bb4aece..da0cdd3 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -109,8 +109,8 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev)
 	u32 addr, gpiocontrol;
 	unsigned long flags;
 
-	pdata = sdiodev->pdata;
-	if ((pdata) && (pdata->oob_irq_supported)) {
+	pdata = &sdiodev->settings->bus.sdio;
+	if (pdata->oob_irq_supported) {
 		brcmf_dbg(SDIO, "Enter, register OOB IRQ %d\n",
 			  pdata->oob_irq_nr);
 		ret = request_irq(pdata->oob_irq_nr, brcmf_sdiod_oob_irqhandler,
@@ -177,8 +177,8 @@ int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev)
 
 	brcmf_dbg(SDIO, "Entering\n");
 
-	pdata = sdiodev->pdata;
-	if ((pdata) && (pdata->oob_irq_supported)) {
+	pdata = &sdiodev->settings->bus.sdio;
+	if (pdata->oob_irq_supported) {
 		sdio_claim_host(sdiodev->func[1]);
 		brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_BRCM_SEPINT, 0, NULL);
 		brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_IENx, 0, NULL);
@@ -522,7 +522,7 @@ static int brcmf_sdiod_sglist_rw(struct brcmf_sdio_dev *sdiodev, uint fn,
 	target_list = pktlist;
 	/* for host with broken sg support, prepare a page aligned list */
 	__skb_queue_head_init(&local_list);
-	if (sdiodev->pdata && sdiodev->pdata->broken_sg_support && !write) {
+	if (!write && sdiodev->settings->bus.sdio.broken_sg_support) {
 		req_sz = 0;
 		skb_queue_walk(pktlist, pkt_next)
 			req_sz += pkt_next->len;
@@ -629,7 +629,7 @@ static int brcmf_sdiod_sglist_rw(struct brcmf_sdio_dev *sdiodev, uint fn,
 		}
 	}
 
-	if (sdiodev->pdata && sdiodev->pdata->broken_sg_support && !write) {
+	if (!write && sdiodev->settings->bus.sdio.broken_sg_support) {
 		local_pkt_next = local_list.next;
 		orig_offset = 0;
 		skb_queue_walk(pktlist, pkt_next) {
@@ -900,7 +900,7 @@ void brcmf_sdiod_sgtable_alloc(struct brcmf_sdio_dev *sdiodev)
 		return;
 
 	nents = max_t(uint, BRCMF_DEFAULT_RXGLOM_SIZE,
-		      sdiodev->bus_if->drvr->settings->sdiod_txglomsz);
+		      sdiodev->settings->bus.sdio.txglomsz);
 	nents += (nents >> 4) + 1;
 
 	WARN_ON(nents > sdiodev->max_segment_count);
@@ -912,7 +912,7 @@ void brcmf_sdiod_sgtable_alloc(struct brcmf_sdio_dev *sdiodev)
 		sdiodev->sg_support = false;
 	}
 
-	sdiodev->txglomsz = sdiodev->bus_if->drvr->settings->sdiod_txglomsz;
+	sdiodev->txglomsz = sdiodev->settings->bus.sdio.txglomsz;
 }
 
 #ifdef CONFIG_PM_SLEEP
@@ -1246,8 +1246,8 @@ static int brcmf_ops_sdio_suspend(struct device *dev)
 
 	sdio_flags = MMC_PM_KEEP_POWER;
 	if (sdiodev->wowl_enabled) {
-		if (sdiodev->pdata->oob_irq_supported)
-			enable_irq_wake(sdiodev->pdata->oob_irq_nr);
+		if (sdiodev->settings->bus.sdio.oob_irq_supported)
+			enable_irq_wake(sdiodev->settings->bus.sdio.oob_irq_nr);
 		else
 			sdio_flags |= MMC_PM_WAKE_SDIO_IRQ;
 	}
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h
index 36093f9..8e02a47 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bus.h
@@ -43,6 +43,8 @@ enum brcmf_bus_protocol_type {
 	BRCMF_PROTO_MSGBUF
 };
 
+struct brcmf_mp_device;
+
 struct brcmf_bus_dcmd {
 	char *name;
 	char *param;
@@ -217,7 +219,7 @@ bool brcmf_c_prec_enq(struct device *dev, struct pktq *q, struct sk_buff *pkt,
 void brcmf_rx_frame(struct device *dev, struct sk_buff *rxp);
 
 /* Indication from bus module regarding presence/insertion of dongle. */
-int brcmf_attach(struct device *dev);
+int brcmf_attach(struct device *dev, struct brcmf_mp_device *settings);
 /* Indication from bus module regarding removal/absence of dongle */
 void brcmf_detach(struct device *dev);
 /* Indication from bus module that dongle should be reset */
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index 4bd3225..9e909e3 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -243,14 +243,35 @@ static void brcmf_mp_attach(void)
 	}
 }
 
-struct brcmfmac_sdio_pd *brcmf_get_module_param(struct device *dev,
-						enum brcmf_bus_type bus_type,
-						u32 chip, u32 chiprev)
+struct brcmf_mp_device *brcmf_get_module_param(struct device *dev,
+					       enum brcmf_bus_type bus_type,
+					       u32 chip, u32 chiprev)
 {
-	struct brcmfmac_sdio_pd *pdata;
+	struct brcmf_mp_device *settings;
 	struct brcmfmac_pd_device *device_pd;
+	bool found;
 	int i;
 
+	brcmf_dbg(INFO, "Enter, bus=%d, chip=%d, rev=%d\n", bus_type, chip,
+		  chiprev);
+	settings = kzalloc(sizeof(*settings), GFP_ATOMIC);
+	if (!settings)
+		return NULL;
+
+	/* start by using the module paramaters */
+	settings->p2p_enable = !!brcmf_p2p_enable;
+	settings->feature_disable = brcmf_feature_disable;
+	settings->fcmode = brcmf_fcmode;
+	settings->roamoff = !!brcmf_roamoff;
+#ifdef DEBUG
+	settings->ignore_probe_fail = !!brcmf_ignore_probe_fail;
+#endif
+
+	if (bus_type == BRCMF_BUSTYPE_SDIO)
+		settings->bus.sdio.txglomsz = brcmf_sdiod_txglomsz;
+
+	/* See if there is any device specific platform data configured */
+	found = false;
 	if (brcmfmac_pdata) {
 		for (i = 0; i < brcmfmac_pdata->device_count; i++) {
 			device_pd = &brcmfmac_pdata->devices[i];
@@ -259,38 +280,29 @@ struct brcmfmac_sdio_pd *brcmf_get_module_param(struct device *dev,
 			    ((device_pd->rev == chiprev) ||
 			     (device_pd->rev == -1))) {
 				brcmf_dbg(INFO, "Platform data for device found\n");
+				settings->country_codes =
+						device_pd->country_codes;
 				if (device_pd->bus_type == BRCMF_BUSTYPE_SDIO)
-					return &device_pd->bus.sdio;
+					memcpy(&settings->bus.sdio,
+					       &device_pd->bus.sdio,
+					       sizeof(settings->bus.sdio));
+				found = true;
 				break;
 			}
 		}
 	}
-	pdata = NULL;
-	brcmf_of_probe(dev, &pdata);
-
-	return pdata;
-}
-
-int brcmf_mp_device_attach(struct brcmf_pub *drvr)
-{
-	drvr->settings = kzalloc(sizeof(*drvr->settings), GFP_ATOMIC);
-	if (!drvr->settings)
-		return -ENOMEM;
-
-	drvr->settings->sdiod_txglomsz = brcmf_sdiod_txglomsz;
-	drvr->settings->p2p_enable = !!brcmf_p2p_enable;
-	drvr->settings->feature_disable = brcmf_feature_disable;
-	drvr->settings->fcmode = brcmf_fcmode;
-	drvr->settings->roamoff = !!brcmf_roamoff;
-#ifdef DEBUG
-	drvr->settings->ignore_probe_fail = !!brcmf_ignore_probe_fail;
-#endif
-	return 0;
+	if ((bus_type == BRCMF_BUSTYPE_SDIO) && (!found)) {
+		/* No platform data for this device. In case of SDIO try OF
+		 * (Open Firwmare) Device Tree.
+		 */
+		brcmf_of_probe(dev, &settings->bus.sdio);
+	}
+	return settings;
 }
 
-void brcmf_mp_device_detach(struct brcmf_pub *drvr)
+void brcmf_release_module_param(struct brcmf_mp_device *module_param)
 {
-	kfree(drvr->settings);
+	kfree(module_param);
 }
 
 static int __init brcmf_common_pd_probe(struct platform_device *pdev)
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
index a64e40e..bd095ab 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.h
@@ -45,41 +45,30 @@ extern struct brcmf_mp_global_t brcmf_mp_global;
 /**
  * struct brcmf_mp_device - Device module paramaters.
  *
- * @sdiod_txglomsz: SDIO txglom size.
- * @joinboost_5g_rssi: 5g rssi booost for preferred join selection.
  * @p2p_enable: Legacy P2P0 enable (old wpa_supplicant).
  * @feature_disable: Feature_disable bitmask.
  * @fcmode: FWS flow control.
  * @roamoff: Firmware roaming off?
+ * @ignore_probe_fail: Ignore probe failure.
  * @country_codes: If available, pointer to struct for translating country codes
+ * @bus: Bus specific platform data. Only SDIO at the mmoment.
  */
 struct brcmf_mp_device {
-	int	sdiod_txglomsz;
-	int	joinboost_5g_rssi;
-	bool	p2p_enable;
-	int	feature_disable;
-	int	fcmode;
-	bool	roamoff;
-	bool	ignore_probe_fail;
+	bool		p2p_enable;
+	unsigned int	feature_disable;
+	int		fcmode;
+	bool		roamoff;
+	bool		ignore_probe_fail;
 	struct brcmfmac_pd_cc *country_codes;
+	union {
+		struct brcmfmac_sdio_pd sdio;
+	} bus;
 };
 
-struct brcmfmac_sdio_pd *brcmf_get_module_param(struct device *dev,
-						enum brcmf_bus_type bus_type,
-						u32 chip, u32 chiprev);
-int brcmf_mp_device_attach(struct brcmf_pub *drvr);
-void brcmf_mp_device_detach(struct brcmf_pub *drvr);
-#ifdef DEBUG
-static inline bool brcmf_ignoring_probe_fail(struct brcmf_pub *drvr)
-{
-	return drvr->settings->ignore_probe_fail;
-}
-#else
-static inline bool brcmf_ignoring_probe_fail(struct brcmf_pub *drvr)
-{
-	return false;
-}
-#endif
+struct brcmf_mp_device *brcmf_get_module_param(struct device *dev,
+					       enum brcmf_bus_type bus_type,
+					       u32 chip, u32 chiprev);
+void brcmf_release_module_param(struct brcmf_mp_device *module_param);
 
 /* Sets dongle media info (drv_version, mac address). */
 int brcmf_c_preinit_dcmds(struct brcmf_if *ifp);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index 7b24e2b..ff825cd 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -1104,7 +1104,7 @@ static int brcmf_inet6addr_changed(struct notifier_block *nb,
 }
 #endif
 
-int brcmf_attach(struct device *dev)
+int brcmf_attach(struct device *dev, struct brcmf_mp_device *settings)
 {
 	struct brcmf_pub *drvr = NULL;
 	int ret = 0;
@@ -1126,10 +1126,7 @@ int brcmf_attach(struct device *dev)
 	drvr->hdrlen = 0;
 	drvr->bus_if = dev_get_drvdata(dev);
 	drvr->bus_if->drvr = drvr;
-
-	/* Initialize device specific settings */
-	if (brcmf_mp_device_attach(drvr))
-		goto fail;
+	drvr->settings = settings;
 
 	/* attach debug facilities */
 	brcmf_debug_attach(drvr);
@@ -1274,7 +1271,7 @@ fail:
 		brcmf_net_detach(p2p_ifp->ndev);
 	drvr->iflist[0] = NULL;
 	drvr->iflist[1] = NULL;
-	if (brcmf_ignoring_probe_fail(drvr))
+	if (drvr->settings->ignore_probe_fail)
 		ret = 0;
 
 	return ret;
@@ -1350,8 +1347,6 @@ void brcmf_detach(struct device *dev)
 
 	brcmf_proto_detach(drvr);
 
-	brcmf_mp_device_detach(drvr);
-
 	brcmf_debug_detach(drvr);
 	bus_if->drvr = NULL;
 	kfree(drvr);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index ece0b65..425c41d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -23,7 +23,7 @@
 #include "common.h"
 #include "of.h"
 
-void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio)
+void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd *sdio)
 {
 	struct device_node *np = dev->of_node;
 	int irq;
@@ -33,12 +33,8 @@ void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio)
 	if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
 		return;
 
-	*sdio = devm_kzalloc(dev, sizeof(*sdio), GFP_KERNEL);
-	if (!(*sdio))
-		return;
-
 	if (of_property_read_u32(np, "brcm,drive-strength", &val) == 0)
-		(*sdio)->drive_strength = val;
+		sdio->drive_strength = val;
 
 	/* make sure there are interrupts defined in the node */
 	if (!of_find_property(np, "interrupts", NULL))
@@ -51,7 +47,7 @@ void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio)
 	}
 	irqf = irqd_get_trigger_type(irq_get_irq_data(irq));
 
-	(*sdio)->oob_irq_supported = true;
-	(*sdio)->oob_irq_nr = irq;
-	(*sdio)->oob_irq_flags = irqf;
+	sdio->oob_irq_supported = true;
+	sdio->oob_irq_nr = irq;
+	sdio->oob_irq_flags = irqf;
 }
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
index 1ba951f..a9d94c1 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.h
@@ -14,10 +14,9 @@
  * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
 #ifdef CONFIG_OF
-void
-brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio);
+void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd *sdio);
 #else
-static void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd **sdio)
+static void brcmf_of_probe(struct device *dev, struct brcmfmac_sdio_pd *sdio)
 {
 }
 #endif /* CONFIG_OF */
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 1866b66..52fef5e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -37,6 +37,8 @@
 #include "pcie.h"
 #include "firmware.h"
 #include "chip.h"
+#include "core.h"
+#include "common.h"
 
 
 enum brcmf_pcie_state {
@@ -266,6 +268,7 @@ struct brcmf_pciedev_info {
 	u16 (*read_ptr)(struct brcmf_pciedev_info *devinfo, u32 mem_offset);
 	void (*write_ptr)(struct brcmf_pciedev_info *devinfo, u32 mem_offset,
 			  u16 value);
+	struct brcmf_mp_device *settings;
 };
 
 struct brcmf_pcie_ringbuf {
@@ -1525,16 +1528,16 @@ static void brcmf_pcie_release_resource(struct brcmf_pciedev_info *devinfo)
 }
 
 
-static int brcmf_pcie_attach_bus(struct device *dev)
+static int brcmf_pcie_attach_bus(struct brcmf_pciedev_info *devinfo)
 {
 	int ret;
 
 	/* Attach to the common driver interface */
-	ret = brcmf_attach(dev);
+	ret = brcmf_attach(&devinfo->pdev->dev, devinfo->settings);
 	if (ret) {
 		brcmf_err("brcmf_attach failed\n");
 	} else {
-		ret = brcmf_bus_start(dev);
+		ret = brcmf_bus_start(&devinfo->pdev->dev);
 		if (ret)
 			brcmf_err("dongle is not responding\n");
 	}
@@ -1672,7 +1675,7 @@ static void brcmf_pcie_setup(struct device *dev, const struct firmware *fw,
 	init_waitqueue_head(&devinfo->mbdata_resp_wait);
 
 	brcmf_pcie_intr_enable(devinfo);
-	if (brcmf_pcie_attach_bus(bus->dev) == 0)
+	if (brcmf_pcie_attach_bus(devinfo) == 0)
 		return;
 
 	brcmf_pcie_bus_console_read(devinfo);
@@ -1716,6 +1719,15 @@ brcmf_pcie_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto fail;
 	}
 
+	devinfo->settings = brcmf_get_module_param(&devinfo->pdev->dev,
+						   BRCMF_BUSTYPE_PCIE,
+						   devinfo->ci->chip,
+						   devinfo->ci->chiprev);
+	if (!devinfo->settings) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
 	bus = kzalloc(sizeof(*bus), GFP_KERNEL);
 	if (!bus) {
 		ret = -ENOMEM;
@@ -1760,6 +1772,8 @@ fail:
 	brcmf_pcie_release_resource(devinfo);
 	if (devinfo->ci)
 		brcmf_chip_detach(devinfo->ci);
+	if (devinfo->settings)
+		brcmf_release_module_param(devinfo->settings);
 	kfree(pcie_bus_dev);
 	kfree(devinfo);
 	return ret;
@@ -1799,6 +1813,8 @@ brcmf_pcie_remove(struct pci_dev *pdev)
 
 	if (devinfo->ci)
 		brcmf_chip_detach(devinfo->ci);
+	if (devinfo->settings)
+		brcmf_release_module_param(devinfo->settings);
 
 	kfree(devinfo);
 	dev_set_drvdata(&pdev->dev, NULL);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index 6e36704..43fd3f4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -2442,15 +2442,17 @@ static void brcmf_sdio_bus_stop(struct device *dev)
 
 static inline void brcmf_sdio_clrintr(struct brcmf_sdio *bus)
 {
+	struct brcmf_sdio_dev *sdiodev;
 	unsigned long flags;
 
-	if (bus->sdiodev->oob_irq_requested) {
-		spin_lock_irqsave(&bus->sdiodev->irq_en_lock, flags);
-		if (!bus->sdiodev->irq_en && !atomic_read(&bus->ipend)) {
-			enable_irq(bus->sdiodev->pdata->oob_irq_nr);
-			bus->sdiodev->irq_en = true;
+	sdiodev = bus->sdiodev;
+	if (sdiodev->oob_irq_requested) {
+		spin_lock_irqsave(&sdiodev->irq_en_lock, flags);
+		if (!sdiodev->irq_en && !atomic_read(&bus->ipend)) {
+			enable_irq(sdiodev->settings->bus.sdio.oob_irq_nr);
+			sdiodev->irq_en = true;
 		}
-		spin_unlock_irqrestore(&bus->sdiodev->irq_en_lock, flags);
+		spin_unlock_irqrestore(&sdiodev->irq_en_lock, flags);
 	}
 }
 
@@ -3394,9 +3396,7 @@ static int brcmf_sdio_bus_preinit(struct device *dev)
 					   sizeof(u32));
 	} else {
 		/* otherwise, set txglomalign */
-		value = 4;
-		if (sdiodev->pdata)
-			value = sdiodev->pdata->sd_sgentry_align;
+		value = sdiodev->settings->bus.sdio.sd_sgentry_align;
 		/* SDIO ADMA requires at least 32 bit alignment */
 		value = max_t(u32, value, 4);
 		err = brcmf_iovar_data_set(dev, "bus:txglomalign", &value,
@@ -3811,21 +3811,25 @@ brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 		bus->ci = NULL;
 		goto fail;
 	}
-	sdiodev->pdata = brcmf_get_module_param(sdiodev->dev,
+	sdiodev->settings = brcmf_get_module_param(sdiodev->dev,
 						   BRCMF_BUSTYPE_SDIO,
 						   bus->ci->chip,
 						   bus->ci->chiprev);
+	if (!sdiodev->settings) {
+		brcmf_err("Failed to get device parameters\n");
+		goto fail;
+	}
 	/* platform specific configuration:
 	 *   alignments must be at least 4 bytes for ADMA
 	 */
 	bus->head_align = ALIGNMENT;
 	bus->sgentry_align = ALIGNMENT;
-	if (sdiodev->pdata) {
-		if (sdiodev->pdata->sd_head_align > ALIGNMENT)
-			bus->head_align = sdiodev->pdata->sd_head_align;
-		if (sdiodev->pdata->sd_sgentry_align > ALIGNMENT)
-			bus->sgentry_align = sdiodev->pdata->sd_sgentry_align;
-	}
+	if (sdiodev->settings->bus.sdio.sd_head_align > ALIGNMENT)
+		bus->head_align = sdiodev->settings->bus.sdio.sd_head_align;
+	if (sdiodev->settings->bus.sdio.sd_sgentry_align > ALIGNMENT)
+		bus->sgentry_align =
+				sdiodev->settings->bus.sdio.sd_sgentry_align;
+
 	/* allocate scatter-gather table. sg support
 	 * will be disabled upon allocation failure.
 	 */
@@ -3837,7 +3841,7 @@ brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 	 */
 	if ((sdio_get_host_pm_caps(sdiodev->func[1]) & MMC_PM_KEEP_POWER) &&
 	    ((sdio_get_host_pm_caps(sdiodev->func[1]) & MMC_PM_WAKE_SDIO_IRQ) ||
-	     (sdiodev->pdata && sdiodev->pdata->oob_irq_supported)))
+	     (sdiodev->settings->bus.sdio.oob_irq_supported)))
 		sdiodev->bus_if->wowl_supported = true;
 #endif
 
@@ -3846,8 +3850,8 @@ brcmf_sdio_probe_attach(struct brcmf_sdio *bus)
 		goto fail;
 	}
 
-	if ((sdiodev->pdata) && (sdiodev->pdata->drive_strength))
-		drivestrength = sdiodev->pdata->drive_strength;
+	if (sdiodev->settings->bus.sdio.drive_strength)
+		drivestrength = sdiodev->settings->bus.sdio.drive_strength;
 	else
 		drivestrength = DEFAULT_SDIO_DRIVE_STRENGTH;
 	brcmf_sdio_drivestrengthinit(sdiodev, bus->ci, drivestrength);
@@ -4124,7 +4128,7 @@ struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 	bus->tx_hdrlen = SDPCM_HWHDR_LEN + SDPCM_SWHDR_LEN;
 
 	/* Attach to the common layer, reserve hdr space */
-	ret = brcmf_attach(bus->sdiodev->dev);
+	ret = brcmf_attach(bus->sdiodev->dev, bus->sdiodev->settings);
 	if (ret != 0) {
 		brcmf_err("brcmf_attach failed\n");
 		goto fail;
@@ -4228,6 +4232,8 @@ void brcmf_sdio_remove(struct brcmf_sdio *bus)
 			}
 			brcmf_chip_detach(bus->ci);
 		}
+		if (bus->sdiodev->settings)
+			brcmf_release_module_param(bus->sdiodev->settings);
 
 		kfree(bus->rxbuf);
 		kfree(bus->hdrbuf);
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
index 50df9cb..dcf0ce8 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h
@@ -184,7 +184,7 @@ struct brcmf_sdio_dev {
 	struct brcmf_sdio *bus;
 	struct device *dev;
 	struct brcmf_bus *bus_if;
-	struct brcmfmac_sdio_pd *pdata;
+	struct brcmf_mp_device *settings;
 	bool oob_irq_requested;
 	bool irq_en;			/* irq enable flags */
 	spinlock_t irq_en_lock;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
index c72b7b3..869eb82 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
@@ -27,6 +27,8 @@
 #include "debug.h"
 #include "firmware.h"
 #include "usb.h"
+#include "core.h"
+#include "common.h"
 
 
 #define IOCTL_RESP_TIMEOUT		msecs_to_jiffies(2000)
@@ -171,6 +173,7 @@ struct brcmf_usbdev_info {
 	struct urb *bulk_urb; /* used for FW download */
 
 	bool wowl_enabled;
+	struct brcmf_mp_device *settings;
 };
 
 static void brcmf_usb_rx_refill(struct brcmf_usbdev_info *devinfo,
@@ -1027,6 +1030,9 @@ static void brcmf_usb_detach(struct brcmf_usbdev_info *devinfo)
 
 	kfree(devinfo->tx_reqs);
 	kfree(devinfo->rx_reqs);
+
+	if (devinfo->settings)
+		brcmf_release_module_param(devinfo->settings);
 }
 
 
@@ -1136,7 +1142,7 @@ static int brcmf_usb_bus_setup(struct brcmf_usbdev_info *devinfo)
 	int ret;
 
 	/* Attach to the common driver interface */
-	ret = brcmf_attach(devinfo->dev);
+	ret = brcmf_attach(devinfo->dev, devinfo->settings);
 	if (ret) {
 		brcmf_err("brcmf_attach failed\n");
 		return ret;
@@ -1223,6 +1229,14 @@ static int brcmf_usb_probe_cb(struct brcmf_usbdev_info *devinfo)
 	bus->wowl_supported = true;
 #endif
 
+	devinfo->settings = brcmf_get_module_param(bus->dev, BRCMF_BUSTYPE_USB,
+						   bus_pub->devid,
+						   bus_pub->chiprev);
+	if (!devinfo->settings) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
 	if (!brcmf_usb_dlneeded(devinfo)) {
 		ret = brcmf_usb_bus_setup(devinfo);
 		if (ret)
-- 
cgit v0.10.2


From 219e0f747ad6d2b504199dbbaee6b9ad269dde07 Mon Sep 17 00:00:00 2001
From: Hante Meuleman <hante.meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:09 +0100
Subject: brcmfmac: integrate add_keyext in add_key

brcmf_add_keyext is called when a key is configured for a specific
mac address. This function is very similar to the calling function
brcmf_add_key. Integrate this function and also use existing del_key
function in case key is to be cleared.

Reviewed-by: Arend Van Spriel <arend.van@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <franky.lin@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5e3acac..5203d8f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -2073,84 +2073,34 @@ done:
 }
 
 static s32
-brcmf_add_keyext(struct wiphy *wiphy, struct net_device *ndev,
-	      u8 key_idx, const u8 *mac_addr, struct key_params *params)
+brcmf_cfg80211_del_key(struct wiphy *wiphy, struct net_device *ndev,
+		       u8 key_idx, bool pairwise, const u8 *mac_addr)
 {
 	struct brcmf_if *ifp = netdev_priv(ndev);
 	struct brcmf_wsec_key key;
 	s32 err = 0;
-	u8 keybuf[8];
+
+	brcmf_dbg(TRACE, "Enter\n");
+	if (!check_vif_up(ifp->vif))
+		return -EIO;
+
+	if (key_idx >= BRCMF_MAX_DEFAULT_KEYS) {
+		/* we ignore this key index in this case */
+		return -EINVAL;
+	}
 
 	memset(&key, 0, sizeof(key));
-	key.index = (u32) key_idx;
-	/* Instead of bcast for ea address for default wep keys,
-		 driver needs it to be Null */
-	if (!is_multicast_ether_addr(mac_addr))
-		memcpy((char *)&key.ea, (void *)mac_addr, ETH_ALEN);
-	key.len = (u32) params->key_len;
-	/* check for key index change */
-	if (key.len == 0) {
-		/* key delete */
-		err = send_key_to_dongle(ifp, &key);
-		if (err)
-			brcmf_err("key delete error (%d)\n", err);
-	} else {
-		if (key.len > sizeof(key.data)) {
-			brcmf_err("Invalid key length (%d)\n", key.len);
-			return -EINVAL;
-		}
 
-		brcmf_dbg(CONN, "Setting the key index %d\n", key.index);
-		memcpy(key.data, params->key, key.len);
+	key.index = (u32)key_idx;
+	key.flags = BRCMF_PRIMARY_KEY;
+	key.algo = CRYPTO_ALGO_OFF;
 
-		if (!brcmf_is_apmode(ifp->vif) &&
-		    (params->cipher == WLAN_CIPHER_SUITE_TKIP)) {
-			brcmf_dbg(CONN, "Swapping RX/TX MIC key\n");
-			memcpy(keybuf, &key.data[24], sizeof(keybuf));
-			memcpy(&key.data[24], &key.data[16], sizeof(keybuf));
-			memcpy(&key.data[16], keybuf, sizeof(keybuf));
-		}
+	brcmf_dbg(CONN, "key index (%d)\n", key_idx);
 
-		/* if IW_ENCODE_EXT_RX_SEQ_VALID set */
-		if (params->seq && params->seq_len == 6) {
-			/* rx iv */
-			u8 *ivptr;
-			ivptr = (u8 *) params->seq;
-			key.rxiv.hi = (ivptr[5] << 24) | (ivptr[4] << 16) |
-			    (ivptr[3] << 8) | ivptr[2];
-			key.rxiv.lo = (ivptr[1] << 8) | ivptr[0];
-			key.iv_initialized = true;
-		}
+	/* Set the new key/index */
+	err = send_key_to_dongle(ifp, &key);
 
-		switch (params->cipher) {
-		case WLAN_CIPHER_SUITE_WEP40:
-			key.algo = CRYPTO_ALGO_WEP1;
-			brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_WEP40\n");
-			break;
-		case WLAN_CIPHER_SUITE_WEP104:
-			key.algo = CRYPTO_ALGO_WEP128;
-			brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_WEP104\n");
-			break;
-		case WLAN_CIPHER_SUITE_TKIP:
-			key.algo = CRYPTO_ALGO_TKIP;
-			brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_TKIP\n");
-			break;
-		case WLAN_CIPHER_SUITE_AES_CMAC:
-			key.algo = CRYPTO_ALGO_AES_CCM;
-			brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_AES_CMAC\n");
-			break;
-		case WLAN_CIPHER_SUITE_CCMP:
-			key.algo = CRYPTO_ALGO_AES_CCM;
-			brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_CCMP\n");
-			break;
-		default:
-			brcmf_err("Invalid cipher (0x%x)\n", params->cipher);
-			return -EINVAL;
-		}
-		err = send_key_to_dongle(ifp, &key);
-		if (err)
-			brcmf_err("wsec_key error (%d)\n", err);
-	}
+	brcmf_dbg(TRACE, "Exit\n");
 	return err;
 }
 
@@ -2163,8 +2113,9 @@ brcmf_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
 	struct brcmf_wsec_key *key;
 	s32 val;
 	s32 wsec;
-	s32 err = 0;
+	s32 err;
 	u8 keybuf[8];
+	bool ext_key;
 
 	brcmf_dbg(TRACE, "Enter\n");
 	brcmf_dbg(CONN, "key index (%d)\n", key_idx);
@@ -2177,27 +2128,32 @@ brcmf_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
 		return -EINVAL;
 	}
 
-	if (mac_addr &&
-		(params->cipher != WLAN_CIPHER_SUITE_WEP40) &&
-		(params->cipher != WLAN_CIPHER_SUITE_WEP104)) {
-		brcmf_dbg(TRACE, "Exit");
-		return brcmf_add_keyext(wiphy, ndev, key_idx, mac_addr, params);
-	}
-
-	key = &ifp->vif->profile.key[key_idx];
-	memset(key, 0, sizeof(*key));
+	if (params->key_len == 0)
+		return brcmf_cfg80211_del_key(wiphy, ndev, key_idx, pairwise,
+					      mac_addr);
 
 	if (params->key_len > sizeof(key->data)) {
 		brcmf_err("Too long key length (%u)\n", params->key_len);
-		err = -EINVAL;
-		goto done;
+		return -EINVAL;
 	}
+
+	ext_key = false;
+	if (mac_addr && (params->cipher != WLAN_CIPHER_SUITE_WEP40) &&
+	    (params->cipher != WLAN_CIPHER_SUITE_WEP104)) {
+		brcmf_dbg(TRACE, "Ext key, mac %pM", mac_addr);
+		ext_key = true;
+	}
+
+	key = &ifp->vif->profile.key[key_idx];
+	memset(key, 0, sizeof(*key));
+	if ((ext_key) && (!is_multicast_ether_addr(mac_addr)))
+		memcpy((char *)&key->ea, (void *)mac_addr, ETH_ALEN);
 	key->len = params->key_len;
 	key->index = key_idx;
-
 	memcpy(key->data, params->key, key->len);
+	if (!ext_key)
+		key->flags = BRCMF_PRIMARY_KEY;
 
-	key->flags = BRCMF_PRIMARY_KEY;
 	switch (params->cipher) {
 	case WLAN_CIPHER_SUITE_WEP40:
 		key->algo = CRYPTO_ALGO_WEP1;
@@ -2237,7 +2193,7 @@ brcmf_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
 	}
 
 	err = send_key_to_dongle(ifp, key);
-	if (err)
+	if (ext_key || err)
 		goto done;
 
 	err = brcmf_fil_bsscfg_int_get(ifp, "wsec", &wsec);
@@ -2258,38 +2214,6 @@ done:
 }
 
 static s32
-brcmf_cfg80211_del_key(struct wiphy *wiphy, struct net_device *ndev,
-		    u8 key_idx, bool pairwise, const u8 *mac_addr)
-{
-	struct brcmf_if *ifp = netdev_priv(ndev);
-	struct brcmf_wsec_key key;
-	s32 err = 0;
-
-	brcmf_dbg(TRACE, "Enter\n");
-	if (!check_vif_up(ifp->vif))
-		return -EIO;
-
-	if (key_idx >= BRCMF_MAX_DEFAULT_KEYS) {
-		/* we ignore this key index in this case */
-		return -EINVAL;
-	}
-
-	memset(&key, 0, sizeof(key));
-
-	key.index = (u32) key_idx;
-	key.flags = BRCMF_PRIMARY_KEY;
-	key.algo = CRYPTO_ALGO_OFF;
-
-	brcmf_dbg(CONN, "key index (%d)\n", key_idx);
-
-	/* Set the new key/index */
-	err = send_key_to_dongle(ifp, &key);
-
-	brcmf_dbg(TRACE, "Exit\n");
-	return err;
-}
-
-static s32
 brcmf_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev,
 		    u8 key_idx, bool pairwise, const u8 *mac_addr, void *cookie,
 		    void (*callback) (void *cookie, struct key_params * params))
-- 
cgit v0.10.2


From 240d61a9ddeb8a77f11369f67f6ef060354b909d Mon Sep 17 00:00:00 2001
From: Hante Meuleman <hante.meuleman@broadcom.com>
Date: Wed, 17 Feb 2016 11:27:10 +0100
Subject: brcmfmac: add 802.11w management frame protection support

Add full support for both AP and STA for management frame protection.

Reviewed-by: Arend Van Spriel <arend.van@broadcom.com>
Reviewed-by: Franky (Zhenhui) Lin <franky.lin@broadcom.com>
Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com>
Signed-off-by: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5203d8f..d5c2a27 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -72,8 +72,13 @@
 #define RSN_AKM_NONE			0	/* None (IBSS) */
 #define RSN_AKM_UNSPECIFIED		1	/* Over 802.1x */
 #define RSN_AKM_PSK			2	/* Pre-shared Key */
+#define RSN_AKM_SHA256_1X		5	/* SHA256, 802.1X */
+#define RSN_AKM_SHA256_PSK		6	/* SHA256, Pre-shared Key */
 #define RSN_CAP_LEN			2	/* Length of RSN capabilities */
-#define RSN_CAP_PTK_REPLAY_CNTR_MASK	0x000C
+#define RSN_CAP_PTK_REPLAY_CNTR_MASK	(BIT(2) | BIT(3))
+#define RSN_CAP_MFPR_MASK		BIT(6)
+#define RSN_CAP_MFPC_MASK		BIT(7)
+#define RSN_PMKID_COUNT_LEN		2
 
 #define VNDR_IE_CMD_LEN			4	/* length of the set command
 						 * string :"add", "del" (+ NUL)
@@ -211,12 +216,19 @@ static const struct ieee80211_regdomain brcmf_regdom = {
 		REG_RULE(5470-10, 5850+10, 80, 6, 20, 0), }
 };
 
-static const u32 __wl_cipher_suites[] = {
+/* Note: brcmf_cipher_suites is an array of int defining which cipher suites
+ * are supported. A pointer to this array and the number of entries is passed
+ * on to upper layers. AES_CMAC defines whether or not the driver supports MFP.
+ * So the cipher suite AES_CMAC has to be the last one in the array, and when
+ * device does not support MFP then the number of suites will be decreased by 1
+ */
+static const u32 brcmf_cipher_suites[] = {
 	WLAN_CIPHER_SUITE_WEP40,
 	WLAN_CIPHER_SUITE_WEP104,
 	WLAN_CIPHER_SUITE_TKIP,
 	WLAN_CIPHER_SUITE_CCMP,
-	WLAN_CIPHER_SUITE_AES_CMAC,
+	/* Keep as last entry: */
+	WLAN_CIPHER_SUITE_AES_CMAC
 };
 
 /* Vendor specific ie. id = 221, oui and type defines exact ie */
@@ -1533,7 +1545,7 @@ static s32 brcmf_set_auth_type(struct net_device *ndev,
 
 static s32
 brcmf_set_wsec_mode(struct net_device *ndev,
-		     struct cfg80211_connect_params *sme, bool mfp)
+		    struct cfg80211_connect_params *sme)
 {
 	struct brcmf_cfg80211_profile *profile = ndev_to_prof(ndev);
 	struct brcmf_cfg80211_security *sec;
@@ -1592,10 +1604,7 @@ brcmf_set_wsec_mode(struct net_device *ndev,
 	    sme->privacy)
 		pval = AES_ENABLED;
 
-	if (mfp)
-		wsec = pval | gval | MFP_CAPABLE;
-	else
-		wsec = pval | gval;
+	wsec = pval | gval;
 	err = brcmf_fil_bsscfg_int_set(netdev_priv(ndev), "wsec", wsec);
 	if (err) {
 		brcmf_err("error (%d)\n", err);
@@ -1612,56 +1621,100 @@ brcmf_set_wsec_mode(struct net_device *ndev,
 static s32
 brcmf_set_key_mgmt(struct net_device *ndev, struct cfg80211_connect_params *sme)
 {
-	struct brcmf_cfg80211_profile *profile = ndev_to_prof(ndev);
-	struct brcmf_cfg80211_security *sec;
-	s32 val = 0;
-	s32 err = 0;
+	struct brcmf_if *ifp = netdev_priv(ndev);
+	s32 val;
+	s32 err;
+	const struct brcmf_tlv *rsn_ie;
+	const u8 *ie;
+	u32 ie_len;
+	u32 offset;
+	u16 rsn_cap;
+	u32 mfp;
+	u16 count;
 
-	if (sme->crypto.n_akm_suites) {
-		err = brcmf_fil_bsscfg_int_get(netdev_priv(ndev),
-					       "wpa_auth", &val);
-		if (err) {
-			brcmf_err("could not get wpa_auth (%d)\n", err);
-			return err;
+	if (!sme->crypto.n_akm_suites)
+		return 0;
+
+	err = brcmf_fil_bsscfg_int_get(netdev_priv(ndev), "wpa_auth", &val);
+	if (err) {
+		brcmf_err("could not get wpa_auth (%d)\n", err);
+		return err;
+	}
+	if (val & (WPA_AUTH_PSK | WPA_AUTH_UNSPECIFIED)) {
+		switch (sme->crypto.akm_suites[0]) {
+		case WLAN_AKM_SUITE_8021X:
+			val = WPA_AUTH_UNSPECIFIED;
+			break;
+		case WLAN_AKM_SUITE_PSK:
+			val = WPA_AUTH_PSK;
+			break;
+		default:
+			brcmf_err("invalid cipher group (%d)\n",
+				  sme->crypto.cipher_group);
+			return -EINVAL;
 		}
-		if (val & (WPA_AUTH_PSK | WPA_AUTH_UNSPECIFIED)) {
-			switch (sme->crypto.akm_suites[0]) {
-			case WLAN_AKM_SUITE_8021X:
-				val = WPA_AUTH_UNSPECIFIED;
-				break;
-			case WLAN_AKM_SUITE_PSK:
-				val = WPA_AUTH_PSK;
-				break;
-			default:
-				brcmf_err("invalid cipher group (%d)\n",
-					  sme->crypto.cipher_group);
-				return -EINVAL;
-			}
-		} else if (val & (WPA2_AUTH_PSK | WPA2_AUTH_UNSPECIFIED)) {
-			switch (sme->crypto.akm_suites[0]) {
-			case WLAN_AKM_SUITE_8021X:
-				val = WPA2_AUTH_UNSPECIFIED;
-				break;
-			case WLAN_AKM_SUITE_PSK:
-				val = WPA2_AUTH_PSK;
-				break;
-			default:
-				brcmf_err("invalid cipher group (%d)\n",
-					  sme->crypto.cipher_group);
-				return -EINVAL;
-			}
+	} else if (val & (WPA2_AUTH_PSK | WPA2_AUTH_UNSPECIFIED)) {
+		switch (sme->crypto.akm_suites[0]) {
+		case WLAN_AKM_SUITE_8021X:
+			val = WPA2_AUTH_UNSPECIFIED;
+			break;
+		case WLAN_AKM_SUITE_8021X_SHA256:
+			val = WPA2_AUTH_1X_SHA256;
+			break;
+		case WLAN_AKM_SUITE_PSK_SHA256:
+			val = WPA2_AUTH_PSK_SHA256;
+			break;
+		case WLAN_AKM_SUITE_PSK:
+			val = WPA2_AUTH_PSK;
+			break;
+		default:
+			brcmf_err("invalid cipher group (%d)\n",
+				  sme->crypto.cipher_group);
+			return -EINVAL;
 		}
+	}
 
-		brcmf_dbg(CONN, "setting wpa_auth to %d\n", val);
-		err = brcmf_fil_bsscfg_int_set(netdev_priv(ndev),
-					       "wpa_auth", val);
-		if (err) {
-			brcmf_err("could not set wpa_auth (%d)\n", err);
-			return err;
-		}
+	if (!brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MFP))
+		goto skip_mfp_config;
+	/* The MFP mode (1 or 2) needs to be determined, parse IEs. The
+	 * IE will not be verified, just a quick search for MFP config
+	 */
+	rsn_ie = brcmf_parse_tlvs((const u8 *)sme->ie, sme->ie_len,
+				  WLAN_EID_RSN);
+	if (!rsn_ie)
+		goto skip_mfp_config;
+	ie = (const u8 *)rsn_ie;
+	ie_len = rsn_ie->len + TLV_HDR_LEN;
+	/* Skip unicast suite */
+	offset = TLV_HDR_LEN + WPA_IE_VERSION_LEN + WPA_IE_MIN_OUI_LEN;
+	if (offset + WPA_IE_SUITE_COUNT_LEN >= ie_len)
+		goto skip_mfp_config;
+	/* Skip multicast suite */
+	count = ie[offset] + (ie[offset + 1] << 8);
+	offset += WPA_IE_SUITE_COUNT_LEN + (count * WPA_IE_MIN_OUI_LEN);
+	if (offset + WPA_IE_SUITE_COUNT_LEN >= ie_len)
+		goto skip_mfp_config;
+	/* Skip auth key management suite(s) */
+	count = ie[offset] + (ie[offset + 1] << 8);
+	offset += WPA_IE_SUITE_COUNT_LEN + (count * WPA_IE_MIN_OUI_LEN);
+	if (offset + WPA_IE_SUITE_COUNT_LEN > ie_len)
+		goto skip_mfp_config;
+	/* Ready to read capabilities */
+	mfp = BRCMF_MFP_NONE;
+	rsn_cap = ie[offset] + (ie[offset + 1] << 8);
+	if (rsn_cap & RSN_CAP_MFPR_MASK)
+		mfp = BRCMF_MFP_REQUIRED;
+	else if (rsn_cap & RSN_CAP_MFPC_MASK)
+		mfp = BRCMF_MFP_CAPABLE;
+	brcmf_fil_bsscfg_int_set(netdev_priv(ndev), "mfp", mfp);
+
+skip_mfp_config:
+	brcmf_dbg(CONN, "setting wpa_auth to %d\n", val);
+	err = brcmf_fil_bsscfg_int_set(netdev_priv(ndev), "wpa_auth", val);
+	if (err) {
+		brcmf_err("could not set wpa_auth (%d)\n", err);
+		return err;
 	}
-	sec = &profile->sec;
-	sec->wpa_auth = sme->crypto.akm_suites[0];
 
 	return err;
 }
@@ -1827,7 +1880,7 @@ brcmf_cfg80211_connect(struct wiphy *wiphy, struct net_device *ndev,
 		goto done;
 	}
 
-	err = brcmf_set_wsec_mode(ndev, sme, sme->mfp == NL80211_MFP_REQUIRED);
+	err = brcmf_set_wsec_mode(ndev, sme);
 	if (err) {
 		brcmf_err("wl_set_set_cipher failed (%d)\n", err);
 		goto done;
@@ -2077,10 +2130,12 @@ brcmf_cfg80211_del_key(struct wiphy *wiphy, struct net_device *ndev,
 		       u8 key_idx, bool pairwise, const u8 *mac_addr)
 {
 	struct brcmf_if *ifp = netdev_priv(ndev);
-	struct brcmf_wsec_key key;
-	s32 err = 0;
+	struct brcmf_wsec_key *key;
+	s32 err;
 
 	brcmf_dbg(TRACE, "Enter\n");
+	brcmf_dbg(CONN, "key index (%d)\n", key_idx);
+
 	if (!check_vif_up(ifp->vif))
 		return -EIO;
 
@@ -2089,16 +2144,19 @@ brcmf_cfg80211_del_key(struct wiphy *wiphy, struct net_device *ndev,
 		return -EINVAL;
 	}
 
-	memset(&key, 0, sizeof(key));
+	key = &ifp->vif->profile.key[key_idx];
 
-	key.index = (u32)key_idx;
-	key.flags = BRCMF_PRIMARY_KEY;
-	key.algo = CRYPTO_ALGO_OFF;
+	if (key->algo == CRYPTO_ALGO_OFF) {
+		brcmf_dbg(CONN, "Ignore clearing of (never configured) key\n");
+		return -EINVAL;
+	}
 
-	brcmf_dbg(CONN, "key index (%d)\n", key_idx);
+	memset(key, 0, sizeof(*key));
+	key->index = (u32)key_idx;
+	key->flags = BRCMF_PRIMARY_KEY;
 
-	/* Set the new key/index */
-	err = send_key_to_dongle(ifp, &key);
+	/* Clear the key/index */
+	err = send_key_to_dongle(ifp, key);
 
 	brcmf_dbg(TRACE, "Exit\n");
 	return err;
@@ -2106,8 +2164,8 @@ brcmf_cfg80211_del_key(struct wiphy *wiphy, struct net_device *ndev,
 
 static s32
 brcmf_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
-		    u8 key_idx, bool pairwise, const u8 *mac_addr,
-		    struct key_params *params)
+		       u8 key_idx, bool pairwise, const u8 *mac_addr,
+		       struct key_params *params)
 {
 	struct brcmf_if *ifp = netdev_priv(ndev);
 	struct brcmf_wsec_key *key;
@@ -2214,9 +2272,10 @@ done:
 }
 
 static s32
-brcmf_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev,
-		    u8 key_idx, bool pairwise, const u8 *mac_addr, void *cookie,
-		    void (*callback) (void *cookie, struct key_params * params))
+brcmf_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev, u8 key_idx,
+		       bool pairwise, const u8 *mac_addr, void *cookie,
+		       void (*callback)(void *cookie,
+					struct key_params *params))
 {
 	struct key_params params;
 	struct brcmf_if *ifp = netdev_priv(ndev);
@@ -2268,8 +2327,15 @@ done:
 
 static s32
 brcmf_cfg80211_config_default_mgmt_key(struct wiphy *wiphy,
-				    struct net_device *ndev, u8 key_idx)
+				       struct net_device *ndev, u8 key_idx)
 {
+	struct brcmf_if *ifp = netdev_priv(ndev);
+
+	brcmf_dbg(TRACE, "Enter key_idx %d\n", key_idx);
+
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MFP))
+		return 0;
+
 	brcmf_dbg(INFO, "Not supported\n");
 
 	return -EOPNOTSUPP;
@@ -3769,7 +3835,7 @@ brcmf_configure_wpaie(struct brcmf_if *ifp,
 	u32 auth = 0; /* d11 open authentication */
 	u16 count;
 	s32 err = 0;
-	s32 len = 0;
+	s32 len;
 	u32 i;
 	u32 wsec;
 	u32 pval = 0;
@@ -3779,6 +3845,7 @@ brcmf_configure_wpaie(struct brcmf_if *ifp,
 	u8 *data;
 	u16 rsn_cap;
 	u32 wme_bss_disable;
+	u32 mfp;
 
 	brcmf_dbg(TRACE, "Enter\n");
 	if (wpa_ie == NULL)
@@ -3893,19 +3960,53 @@ brcmf_configure_wpaie(struct brcmf_if *ifp,
 			is_rsn_ie ? (wpa_auth |= WPA2_AUTH_PSK) :
 				    (wpa_auth |= WPA_AUTH_PSK);
 			break;
+		case RSN_AKM_SHA256_PSK:
+			brcmf_dbg(TRACE, "RSN_AKM_MFP_PSK\n");
+			wpa_auth |= WPA2_AUTH_PSK_SHA256;
+			break;
+		case RSN_AKM_SHA256_1X:
+			brcmf_dbg(TRACE, "RSN_AKM_MFP_1X\n");
+			wpa_auth |= WPA2_AUTH_1X_SHA256;
+			break;
 		default:
 			brcmf_err("Ivalid key mgmt info\n");
 		}
 		offset++;
 	}
 
+	mfp = BRCMF_MFP_NONE;
 	if (is_rsn_ie) {
 		wme_bss_disable = 1;
 		if ((offset + RSN_CAP_LEN) <= len) {
 			rsn_cap = data[offset] + (data[offset + 1] << 8);
 			if (rsn_cap & RSN_CAP_PTK_REPLAY_CNTR_MASK)
 				wme_bss_disable = 0;
+			if (rsn_cap & RSN_CAP_MFPR_MASK) {
+				brcmf_dbg(TRACE, "MFP Required\n");
+				mfp = BRCMF_MFP_REQUIRED;
+				/* Firmware only supports mfp required in
+				 * combination with WPA2_AUTH_PSK_SHA256 or
+				 * WPA2_AUTH_1X_SHA256.
+				 */
+				if (!(wpa_auth & (WPA2_AUTH_PSK_SHA256 |
+						  WPA2_AUTH_1X_SHA256))) {
+					err = -EINVAL;
+					goto exit;
+				}
+				/* Firmware has requirement that WPA2_AUTH_PSK/
+				 * WPA2_AUTH_UNSPECIFIED be set, if SHA256 OUI
+				 * is to be included in the rsn ie.
+				 */
+				if (wpa_auth & WPA2_AUTH_PSK_SHA256)
+					wpa_auth |= WPA2_AUTH_PSK;
+				else if (wpa_auth & WPA2_AUTH_1X_SHA256)
+					wpa_auth |= WPA2_AUTH_UNSPECIFIED;
+			} else if (rsn_cap & RSN_CAP_MFPC_MASK) {
+				brcmf_dbg(TRACE, "MFP Capable\n");
+				mfp = BRCMF_MFP_CAPABLE;
+			}
 		}
+		offset += RSN_CAP_LEN;
 		/* set wme_bss_disable to sync RSN Capabilities */
 		err = brcmf_fil_bsscfg_int_set(ifp, "wme_bss_disable",
 					       wme_bss_disable);
@@ -3913,6 +4014,21 @@ brcmf_configure_wpaie(struct brcmf_if *ifp,
 			brcmf_err("wme_bss_disable error %d\n", err);
 			goto exit;
 		}
+
+		/* Skip PMKID cnt as it is know to be 0 for AP. */
+		offset += RSN_PMKID_COUNT_LEN;
+
+		/* See if there is BIP wpa suite left for MFP */
+		if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MFP) &&
+		    ((offset + WPA_IE_MIN_OUI_LEN) <= len)) {
+			err = brcmf_fil_bsscfg_data_set(ifp, "bip",
+							&data[offset],
+							WPA_IE_MIN_OUI_LEN);
+			if (err < 0) {
+				brcmf_err("bip error %d\n", err);
+				goto exit;
+			}
+		}
 	}
 	/* FOR WPS , set SES_OW_ENABLED */
 	wsec = (pval | gval | SES_OW_ENABLED);
@@ -3929,6 +4045,16 @@ brcmf_configure_wpaie(struct brcmf_if *ifp,
 		brcmf_err("wsec error %d\n", err);
 		goto exit;
 	}
+	/* Configure MFP, this needs to go after wsec otherwise the wsec command
+	 * will overwrite the values set by MFP
+	 */
+	if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MFP)) {
+		err = brcmf_fil_bsscfg_int_set(ifp, "mfp", mfp);
+		if (err < 0) {
+			brcmf_err("mfp error %d\n", err);
+			goto exit;
+		}
+	}
 	/* set upper-layer auth */
 	err = brcmf_fil_bsscfg_int_set(ifp, "wpa_auth", wpa_auth);
 	if (err < 0) {
@@ -6149,8 +6275,10 @@ static int brcmf_setup_wiphy(struct wiphy *wiphy, struct brcmf_if *ifp)
 	wiphy->n_addresses = i;
 
 	wiphy->signal_type = CFG80211_SIGNAL_TYPE_MBM;
-	wiphy->cipher_suites = __wl_cipher_suites;
-	wiphy->n_cipher_suites = ARRAY_SIZE(__wl_cipher_suites);
+	wiphy->cipher_suites = brcmf_cipher_suites;
+	wiphy->n_cipher_suites = ARRAY_SIZE(brcmf_cipher_suites);
+	if (!brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MFP))
+		wiphy->n_cipher_suites--;
 	wiphy->flags |= WIPHY_FLAG_PS_ON_BY_DEFAULT |
 			WIPHY_FLAG_OFFCHAN_TX |
 			WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
index e4c1d43..95e35bc 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h
@@ -72,7 +72,7 @@
 
 #define BRCMF_VNDR_IE_P2PAF_SHIFT	12
 
-#define BRCMF_MAX_DEFAULT_KEYS		4
+#define BRCMF_MAX_DEFAULT_KEYS		6
 
 /* beacon loss timeout defaults */
 #define BRCMF_DEFAULT_BCN_TIMEOUT_ROAM_ON	2
@@ -107,7 +107,6 @@ struct brcmf_cfg80211_security {
 	u32 auth_type;
 	u32 cipher_pairwise;
 	u32 cipher_group;
-	u32 wpa_auth;
 };
 
 /**
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
index bfa19bb..62985f2 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.c
@@ -161,6 +161,7 @@ void brcmf_feat_attach(struct brcmf_pub *drvr)
 		ifp->drvr->feat_flags &= ~BIT(BRCMF_FEAT_MBSS);
 	brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_RSDB, "rsdb_mode");
 	brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_TDLS, "tdls_enable");
+	brcmf_feat_iovar_int_get(ifp, BRCMF_FEAT_MFP, "mfp");
 
 	pfn_mac.version = BRCMF_PFN_MACADDR_CFG_VER;
 	err = brcmf_fil_iovar_data_get(ifp, "pfn_macaddr", &pfn_mac,
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
index a3efa35..db4733a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/feature.h
@@ -30,6 +30,7 @@
  * WOWL_ND: WOWL net detect (PNO)
  * WOWL_GTK: (WOWL) GTK rekeying offload
  * WOWL_ARP_ND: ARP and Neighbor Discovery offload support during WOWL.
+ * MFP: 802.11w Management Frame Protection.
  */
 #define BRCMF_FEAT_LIST \
 	BRCMF_FEAT_DEF(MBSS) \
@@ -42,7 +43,8 @@
 	BRCMF_FEAT_DEF(SCAN_RANDOM_MAC) \
 	BRCMF_FEAT_DEF(WOWL_ND) \
 	BRCMF_FEAT_DEF(WOWL_GTK) \
-	BRCMF_FEAT_DEF(WOWL_ARP_ND)
+	BRCMF_FEAT_DEF(WOWL_ARP_ND) \
+	BRCMF_FEAT_DEF(MFP)
 
 /*
  * Quirks:
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
index 6d41ae3..a4118c0 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwil_types.h
@@ -142,6 +142,10 @@
 #define BRCMF_RSN_KEK_LENGTH		16
 #define BRCMF_RSN_REPLAY_LEN		8
 
+#define BRCMF_MFP_NONE			0
+#define BRCMF_MFP_CAPABLE		1
+#define BRCMF_MFP_REQUIRED		2
+
 /* join preference types for join_pref iovar */
 enum brcmf_join_pref_types {
 	BRCMF_JOIN_PREF_RSSI = 1,
diff --git a/drivers/net/wireless/broadcom/brcm80211/include/brcmu_wifi.h b/drivers/net/wireless/broadcom/brcm80211/include/brcmu_wifi.h
index 3f68dd5..7b9a779 100644
--- a/drivers/net/wireless/broadcom/brcm80211/include/brcmu_wifi.h
+++ b/drivers/net/wireless/broadcom/brcm80211/include/brcmu_wifi.h
@@ -236,6 +236,8 @@ static inline bool ac_bitmap_tst(u8 bitmap, int prec)
 #define WPA2_AUTH_RESERVED3	0x0200
 #define WPA2_AUTH_RESERVED4	0x0400
 #define WPA2_AUTH_RESERVED5	0x0800
+#define WPA2_AUTH_1X_SHA256	0x1000  /* 1X with SHA256 key derivation */
+#define WPA2_AUTH_PSK_SHA256	0x8000	/* PSK with SHA256 key derivation */
 
 #define DOT11_DEFAULT_RTS_LEN		2347
 #define DOT11_DEFAULT_FRAG_LEN		2346
-- 
cgit v0.10.2


From 8dd37c7cd442411f52b90e8b556d38324450de46 Mon Sep 17 00:00:00 2001
From: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Date: Wed, 17 Feb 2016 18:03:37 +0530
Subject: mwifiex: Use to_delayed_work()

Introduce the use of to_delayed_work() helper function instead of open
coding it with container_of()

A simplified version of the Coccinelle semantic patch used to make
this change is:

//<smpl>
@@
expression a;
symbol work;
@@
- container_of(a, struct delayed_work, work)
+ to_delayed_work(a)
//</smpl>

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/11h.c b/drivers/net/wireless/marvell/mwifiex/11h.c
index 71a1b58..81c60d0 100644
--- a/drivers/net/wireless/marvell/mwifiex/11h.c
+++ b/drivers/net/wireless/marvell/mwifiex/11h.c
@@ -123,8 +123,7 @@ void mwifiex_11h_process_join(struct mwifiex_private *priv, u8 **buffer,
 void mwifiex_dfs_cac_work_queue(struct work_struct *work)
 {
 	struct cfg80211_chan_def chandef;
-	struct delayed_work *delayed_work =
-			container_of(work, struct delayed_work, work);
+	struct delayed_work *delayed_work = to_delayed_work(work);
 	struct mwifiex_private *priv =
 			container_of(delayed_work, struct mwifiex_private,
 				     dfs_cac_work);
@@ -289,8 +288,7 @@ int mwifiex_11h_handle_radar_detected(struct mwifiex_private *priv,
 void mwifiex_dfs_chan_sw_work_queue(struct work_struct *work)
 {
 	struct mwifiex_uap_bss_param *bss_cfg;
-	struct delayed_work *delayed_work =
-			container_of(work, struct delayed_work, work);
+	struct delayed_work *delayed_work = to_delayed_work(work);
 	struct mwifiex_private *priv =
 			container_of(delayed_work, struct mwifiex_private,
 				     dfs_chan_sw_work);
-- 
cgit v0.10.2


From 064b32f822afd9a578a4ec6bf754cd61cbe0c324 Mon Sep 17 00:00:00 2001
From: Alexey Khoroshilov <khoroshilov@ispras.ru>
Date: Sun, 21 Feb 2016 01:30:29 +0300
Subject: at76c50x-usb: avoid double usb_put_dev() after downloading internal
 firmware in at76_probe()

There is no need in usb_put_dev() if at76_load_internal_fw() succeed.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c
index dab2513..1efb1d6 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.c
+++ b/drivers/net/wireless/atmel/at76c50x-usb.c
@@ -2481,9 +2481,7 @@ static int at76_probe(struct usb_interface *interface,
 			dev_err(&interface->dev,
 				"error %d downloading internal firmware\n",
 				ret);
-			goto exit;
 		}
-		usb_put_dev(udev);
 		goto exit;
 	}
 
-- 
cgit v0.10.2


From ce2542435aadc9953656e8afddaab0caf0079a3a Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Mon, 22 Feb 2016 11:35:46 +0000
Subject: rtlwifi: pass struct rtl_stats by reference as it is more efficient

passing rtl_stats by value is inefficient; the structure is over 300
bytes in size and generally just one field (packet_report_type)
is being accessed, so the pass by value is a relatively large overhead.
This change just affects just the rx_command_packet calls.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/pci.c b/drivers/net/wireless/realtek/rtlwifi/pci.c
index 140d254..283d608 100644
--- a/drivers/net/wireless/realtek/rtlwifi/pci.c
+++ b/drivers/net/wireless/realtek/rtlwifi/pci.c
@@ -855,7 +855,7 @@ static void _rtl_pci_rx_interrupt(struct ieee80211_hw *hw)
 		}
 		/* handle command packet here */
 		if (rtlpriv->cfg->ops->rx_command_packet &&
-		    rtlpriv->cfg->ops->rx_command_packet(hw, stats, skb)) {
+		    rtlpriv->cfg->ops->rx_command_packet(hw, &stats, skb)) {
 				dev_kfree_skb_any(skb);
 				goto new_trx_end;
 		}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c
index 791efbe..1170106 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c
@@ -851,7 +851,7 @@ void rtl88ee_tx_polling(struct ieee80211_hw *hw, u8 hw_queue)
 }
 
 u32 rtl88ee_rx_command_packet(struct ieee80211_hw *hw,
-			      struct rtl_stats status,
+			      const struct rtl_stats *status,
 			      struct sk_buff *skb)
 {
 	return 0;
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h
index eab5ae0..5a24d19 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h
@@ -790,7 +790,7 @@ void rtl88ee_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc,
 			     bool firstseg, bool lastseg,
 			     struct sk_buff *skb);
 u32 rtl88ee_rx_command_packet(struct ieee80211_hw *hw,
-			      struct rtl_stats status,
+			      const struct rtl_stats *status,
 			      struct sk_buff *skb);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c
index d39ee67..24eff8e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c
@@ -1105,13 +1105,13 @@ void rtl92ee_tx_polling(struct ieee80211_hw *hw, u8 hw_queue)
 }
 
 u32 rtl92ee_rx_command_packet(struct ieee80211_hw *hw,
-			      struct rtl_stats status,
+			      const struct rtl_stats *status,
 			      struct sk_buff *skb)
 {
 	u32 result = 0;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
-	switch (status.packet_report_type) {
+	switch (status->packet_report_type) {
 	case NORMAL_RX:
 		result = 0;
 		break;
@@ -1121,7 +1121,7 @@ u32 rtl92ee_rx_command_packet(struct ieee80211_hw *hw,
 		break;
 	default:
 		RT_TRACE(rtlpriv, COMP_RECV, DBG_TRACE,
-			 "Unknown packet type %d\n", status.packet_report_type);
+			 "Unknown packet type %d\n", status->packet_report_type);
 		break;
 	}
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h
index 8f78ac9..a4c3834 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h
@@ -857,6 +857,6 @@ void rtl92ee_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc,
 			     bool firstseg, bool lastseg,
 			     struct sk_buff *skb);
 u32 rtl92ee_rx_command_packet(struct ieee80211_hw *hw,
-			      struct rtl_stats status,
+			      const struct rtl_stats *status,
 			      struct sk_buff *skb);
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c
index 2f7c144..7b4a9b6 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c
@@ -710,7 +710,7 @@ void rtl8723e_tx_polling(struct ieee80211_hw *hw, u8 hw_queue)
 }
 
 u32 rtl8723e_rx_command_packet(struct ieee80211_hw *hw,
-			       struct rtl_stats status,
+			       const struct rtl_stats *status,
 			       struct sk_buff *skb)
 {
 	return 0;
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h
index 017da7e..32970bf 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h
@@ -716,6 +716,6 @@ void rtl8723e_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc,
 			      bool firstseg, bool lastseg,
 			      struct sk_buff *skb);
 u32 rtl8723e_rx_command_packet(struct ieee80211_hw *hw,
-			       struct rtl_stats status,
+			       const struct rtl_stats *status,
 			       struct sk_buff *skb);
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c
index 338ec9a..6034597 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c
@@ -758,13 +758,13 @@ void rtl8723be_tx_polling(struct ieee80211_hw *hw, u8 hw_queue)
 }
 
 u32 rtl8723be_rx_command_packet(struct ieee80211_hw *hw,
-				struct rtl_stats status,
+				const struct rtl_stats *status,
 				struct sk_buff *skb)
 {
 	u32 result = 0;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
-	switch (status.packet_report_type) {
+	switch (status->packet_report_type) {
 	case NORMAL_RX:
 			result = 0;
 			break;
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h
index 45949ac..40c36607 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h
@@ -620,6 +620,6 @@ void rtl8723be_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc,
 			       bool firstseg, bool lastseg,
 			       struct sk_buff *skb);
 u32 rtl8723be_rx_command_packet(struct ieee80211_hw *hw,
-				struct rtl_stats status,
+				const struct rtl_stats *status,
 				struct sk_buff *skb);
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c
index 174743a..41efaa1 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c
@@ -998,13 +998,13 @@ void rtl8821ae_tx_polling(struct ieee80211_hw *hw, u8 hw_queue)
 }
 
 u32 rtl8821ae_rx_command_packet(struct ieee80211_hw *hw,
-				struct rtl_stats status,
+				const struct rtl_stats *status,
 				struct sk_buff *skb)
 {
 	u32 result = 0;
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
-	switch (status.packet_report_type) {
+	switch (status->packet_report_type) {
 	case NORMAL_RX:
 		result = 0;
 		break;
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h
index 3140904..ad565be 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h
@@ -615,6 +615,6 @@ void rtl8821ae_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc,
 			       bool firstseg, bool lastseg,
 			       struct sk_buff *skb);
 u32 rtl8821ae_rx_command_packet(struct ieee80211_hw *hw,
-				struct rtl_stats status,
+				const struct rtl_stats *status,
 				struct sk_buff *skb);
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index 9e3cdd7..b07b364 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -2189,7 +2189,7 @@ struct rtl_hal_ops {
 	bool (*get_btc_status) (void);
 	bool (*is_fw_header)(struct rtlwifi_firmware_header *hdr);
 	u32 (*rx_command_packet)(struct ieee80211_hw *hw,
-				 struct rtl_stats status, struct sk_buff *skb);
+				 const struct rtl_stats *status, struct sk_buff *skb);
 	void (*add_wowlan_pattern)(struct ieee80211_hw *hw,
 				   struct rtl_wow_pattern *rtl_pattern,
 				   u8 index);
-- 
cgit v0.10.2


From de651ce3d750c19a7b8b7cef22407ecc496619f4 Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 23 Feb 2016 05:16:16 -0800
Subject: mwifiex: add delay when tdls confirm frame is queued

It is observed that driver may send the data packet to tdls peer
before tdls peer receives tdls setup confirm frame.
Similar race condition exists during tdls teardown procedure also.
This patch adds 10 milliseconds delay to resolve the race.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/tdls.c b/drivers/net/wireless/marvell/mwifiex/tdls.c
index 9275f9c..1506496 100644
--- a/drivers/net/wireless/marvell/mwifiex/tdls.c
+++ b/drivers/net/wireless/marvell/mwifiex/tdls.c
@@ -680,6 +680,13 @@ int mwifiex_send_tdls_data_frame(struct mwifiex_private *priv, const u8 *peer,
 	__net_timestamp(skb);
 	mwifiex_queue_tx_pkt(priv, skb);
 
+	/* Delay 10ms to make sure tdls setup confirm/teardown frame
+	 * is received by peer
+	*/
+	if (action_code == WLAN_TDLS_SETUP_CONFIRM ||
+	    action_code == WLAN_TDLS_TEARDOWN)
+		msleep_interruptible(10);
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From a6139b6271f9f95377fe3486aed6120c9142779b Mon Sep 17 00:00:00 2001
From: Amitkumar Karwar <akarwar@marvell.com>
Date: Tue, 23 Feb 2016 05:16:17 -0800
Subject: mwifiex: fix corner case association failure

This patch corrects the error case in association path by returning
-1. Earlier "media_connected" used to remain on in this error case
causing failure for further association attempts.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Fixes: b887664d882ee4 ('mwifiex: channel switch handling for station')
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
index 5cbee58..d5c56eb 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
@@ -314,6 +314,7 @@ int mwifiex_bss_start(struct mwifiex_private *priv, struct cfg80211_bss *bss,
 			mwifiex_dbg(adapter, ERROR,
 				    "Attempt to reconnect on csa closed chan(%d)\n",
 				    bss_desc->channel);
+			ret = -1;
 			goto done;
 		}
 
-- 
cgit v0.10.2


From 8b7ef8b66eb9127b6d2ee46348b3f743bb40facb Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 23 Feb 2016 05:16:18 -0800
Subject: mwifiex: add sdio multiport aggregation debug information

This patch sdio multi port aggregation statistics which can be
used for debugging. This debug data is collected in
/sys/kernel/debug/mwifiex/mlan0/debug.

Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/ioctl.h b/drivers/net/wireless/marvell/mwifiex/ioctl.h
index 14cfa37..a5a48c1 100644
--- a/drivers/net/wireless/marvell/mwifiex/ioctl.h
+++ b/drivers/net/wireless/marvell/mwifiex/ioctl.h
@@ -184,6 +184,7 @@ struct mwifiex_ds_tx_ba_stream_tbl {
 };
 
 #define DBG_CMD_NUM    5
+#define MWIFIEX_DBG_SDIO_MP_NUM    10
 
 struct tdls_peer_info {
 	u8 peer_addr[ETH_ALEN];
@@ -235,6 +236,11 @@ struct mwifiex_debug_info {
 	u8 cmd_sent;
 	u8 cmd_resp_received;
 	u8 event_received;
+	u32 last_mp_wr_bitmap[MWIFIEX_DBG_SDIO_MP_NUM];
+	u32 last_mp_wr_ports[MWIFIEX_DBG_SDIO_MP_NUM];
+	u32 last_mp_wr_len[MWIFIEX_DBG_SDIO_MP_NUM];
+	u32 last_mp_curr_wr_port[MWIFIEX_DBG_SDIO_MP_NUM];
+	u8 last_sdio_mp_index;
 };
 
 #define MWIFIEX_KEY_INDEX_UNICAST	0x40000000
diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index aea7aee..aafc4ab 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -223,6 +223,11 @@ struct mwifiex_dbg {
 	u16 last_cmd_resp_index;
 	u16 last_event[DBG_CMD_NUM];
 	u16 last_event_index;
+	u32 last_mp_wr_bitmap[MWIFIEX_DBG_SDIO_MP_NUM];
+	u32 last_mp_wr_ports[MWIFIEX_DBG_SDIO_MP_NUM];
+	u32 last_mp_wr_len[MWIFIEX_DBG_SDIO_MP_NUM];
+	u32 last_mp_curr_wr_port[MWIFIEX_DBG_SDIO_MP_NUM];
+	u8 last_sdio_mp_index;
 };
 
 enum MWIFIEX_HARDWARE_STATUS {
diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c
index abf15db..838b260 100644
--- a/drivers/net/wireless/marvell/mwifiex/sdio.c
+++ b/drivers/net/wireless/marvell/mwifiex/sdio.c
@@ -1684,6 +1684,7 @@ static int mwifiex_host_to_card_mp_aggr(struct mwifiex_adapter *adapter,
 	s32 f_precopy_cur_buf = 0;
 	s32 f_postcopy_cur_buf = 0;
 	u32 mport;
+	int index;
 
 	if (!card->mpa_tx.enabled ||
 	    (card->has_control_mask && (port == CTRL_PORT)) ||
@@ -1788,6 +1789,15 @@ static int mwifiex_host_to_card_mp_aggr(struct mwifiex_adapter *adapter,
 		ret = mwifiex_write_data_to_card(adapter, card->mpa_tx.buf,
 						 card->mpa_tx.buf_len, mport);
 
+		/* Save the last multi port tx aggreagation info to debug log */
+		index = adapter->dbg.last_sdio_mp_index;
+		index = (index + 1) % MWIFIEX_DBG_SDIO_MP_NUM;
+		adapter->dbg.last_sdio_mp_index = index;
+		adapter->dbg.last_mp_wr_ports[index] = mport;
+		adapter->dbg.last_mp_wr_bitmap[index] = card->mp_wr_bitmap;
+		adapter->dbg.last_mp_wr_len[index] = card->mpa_tx.buf_len;
+		adapter->dbg.last_mp_curr_wr_port[index] = card->curr_wr_port;
+
 		MP_TX_AGGR_BUF_RESET(card);
 	}
 
diff --git a/drivers/net/wireless/marvell/mwifiex/util.c b/drivers/net/wireless/marvell/mwifiex/util.c
index 0cec8a6..6681be0 100644
--- a/drivers/net/wireless/marvell/mwifiex/util.c
+++ b/drivers/net/wireless/marvell/mwifiex/util.c
@@ -78,6 +78,16 @@ static struct mwifiex_debug_data items[] = {
 	 item_addr(last_event), DBG_CMD_NUM},
 	{"last_event_index", item_size(last_event_index),
 	 item_addr(last_event_index), 1},
+	{"last_mp_wr_bitmap", item_size(last_mp_wr_bitmap),
+	 item_addr(last_mp_wr_bitmap), MWIFIEX_DBG_SDIO_MP_NUM},
+	{"last_mp_wr_ports", item_size(last_mp_wr_ports),
+	 item_addr(last_mp_wr_ports), MWIFIEX_DBG_SDIO_MP_NUM},
+	{"last_mp_wr_len", item_size(last_mp_wr_len),
+	 item_addr(last_mp_wr_len), MWIFIEX_DBG_SDIO_MP_NUM},
+	{"last_mp_curr_wr_port", item_size(last_mp_curr_wr_port),
+	 item_addr(last_mp_curr_wr_port), MWIFIEX_DBG_SDIO_MP_NUM},
+	{"last_sdio_mp_index", item_size(last_sdio_mp_index),
+	 item_addr(last_sdio_mp_index), 1},
 	{"num_cmd_h2c_fail", item_size(num_cmd_host_to_card_failure),
 	 item_addr(num_cmd_host_to_card_failure), 1},
 	{"num_cmd_sleep_cfm_fail",
@@ -233,6 +243,16 @@ int mwifiex_get_debug_info(struct mwifiex_private *priv,
 		memcpy(info->last_event, adapter->dbg.last_event,
 		       sizeof(adapter->dbg.last_event));
 		info->last_event_index = adapter->dbg.last_event_index;
+		memcpy(info->last_mp_wr_bitmap, adapter->dbg.last_mp_wr_bitmap,
+		       sizeof(adapter->dbg.last_mp_wr_bitmap));
+		memcpy(info->last_mp_wr_ports, adapter->dbg.last_mp_wr_ports,
+		       sizeof(adapter->dbg.last_mp_wr_ports));
+		memcpy(info->last_mp_curr_wr_port,
+		       adapter->dbg.last_mp_curr_wr_port,
+		       sizeof(adapter->dbg.last_mp_curr_wr_port));
+		memcpy(info->last_mp_wr_len, adapter->dbg.last_mp_wr_len,
+		       sizeof(adapter->dbg.last_mp_wr_len));
+		info->last_sdio_mp_index = adapter->dbg.last_sdio_mp_index;
 		info->data_sent = adapter->data_sent;
 		info->cmd_sent = adapter->cmd_sent;
 		info->cmd_resp_received = adapter->cmd_resp_received;
-- 
cgit v0.10.2


From 0cb52aac4d19510d13410ee4232c0248fabe750e Mon Sep 17 00:00:00 2001
From: Xinming Hu <huxm@marvell.com>
Date: Tue, 23 Feb 2016 05:16:19 -0800
Subject: mwifiex: do not set multiport flag for tx/rx single packet

multiport address flag(0x1000) should not be set during sdio cmd53,
if we have only one packet to read/write.

Signed-off-by: Cathy Luo <cluo@marvell.com>
Signed-off-by: Xinming Hu <huxm@marvell.com>
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c
index 838b260..b2c839a 100644
--- a/drivers/net/wireless/marvell/mwifiex/sdio.c
+++ b/drivers/net/wireless/marvell/mwifiex/sdio.c
@@ -1355,6 +1355,9 @@ static int mwifiex_sdio_card_to_host_mp_aggr(struct mwifiex_adapter *adapter,
 				 card->mpa_rx.start_port;
 		}
 
+		if (card->mpa_rx.pkt_cnt == 1)
+			mport = adapter->ioport + port;
+
 		if (mwifiex_read_data_sync(adapter, card->mpa_rx.buf,
 					   card->mpa_rx.buf_len, mport, 1))
 			goto error;
@@ -1786,6 +1789,9 @@ static int mwifiex_host_to_card_mp_aggr(struct mwifiex_adapter *adapter,
 				 card->mpa_tx.start_port;
 		}
 
+		if (card->mpa_tx.pkt_cnt == 1)
+			mport = adapter->ioport + port;
+
 		ret = mwifiex_write_data_to_card(adapter, card->mpa_tx.buf,
 						 card->mpa_tx.buf_len, mport);
 
-- 
cgit v0.10.2


From f36f299068794ffc5026f25b6a1b3ed615ea832d Mon Sep 17 00:00:00 2001
From: Anthony Wong <anthony.wong@ubuntu.com>
Date: Tue, 23 Feb 2016 23:09:22 +0800
Subject: rt2x00: add new rt2800usb device Buffalo WLI-UC-G450

Add USB ID 0411:01fd for Buffalo WLI-UC-G450 wireless adapter,
RT chipset 3593

Signed-off-by: Anthony Wong <anthony.wong@ubuntu.com>
Cc: stable@vger.kernel.org
Acked-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800usb.c b/drivers/net/wireless/ralink/rt2x00/rt2800usb.c
index bf9afbf..4b0bb6b 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2800usb.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2800usb.c
@@ -1026,6 +1026,7 @@ static struct usb_device_id rt2800usb_device_table[] = {
 	{ USB_DEVICE(0x0411, 0x01a2) },
 	{ USB_DEVICE(0x0411, 0x01ee) },
 	{ USB_DEVICE(0x0411, 0x01a8) },
+	{ USB_DEVICE(0x0411, 0x01fd) },
 	/* Corega */
 	{ USB_DEVICE(0x07aa, 0x002f) },
 	{ USB_DEVICE(0x07aa, 0x003c) },
-- 
cgit v0.10.2


From 32962d5b43b748c43b9b0b32d02674c1ce6997a9 Mon Sep 17 00:00:00 2001
From: Ujjal Roy <royujjal@gmail.com>
Date: Thu, 25 Feb 2016 00:49:23 +0530
Subject: mwifiex: Fixed incorrect indentation issue

This patch fixes the incorrect indentation of the case label.

Signed-off-by: Ujjal Roy <royujjal@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index f2dce81..29b7f6e 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -676,7 +676,7 @@ mwifiex_cfg80211_set_wiphy_params(struct wiphy *wiphy, u32 changed)
 		}
 		break;
 
-		case MWIFIEX_BSS_ROLE_STA:
+	case MWIFIEX_BSS_ROLE_STA:
 		if (priv->media_connected) {
 			mwifiex_dbg(adapter, ERROR,
 				    "cannot change wiphy params when connected");
-- 
cgit v0.10.2


From 3a968d766a63267b3d6b7a66b9d7023b7d93eaae Mon Sep 17 00:00:00 2001
From: Ujjal Roy <royujjal@gmail.com>
Date: Thu, 25 Feb 2016 00:49:43 +0530
Subject: mwifiex: Removed extra spaces before commas

This patch fixes spaces before commas issue in coding style.

Signed-off-by: Ujjal Roy <royujjal@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c
index efb19e2..de36438 100644
--- a/drivers/net/wireless/marvell/mwifiex/pcie.c
+++ b/drivers/net/wireless/marvell/mwifiex/pcie.c
@@ -1401,7 +1401,7 @@ mwifiex_pcie_send_boot_cmd(struct mwifiex_adapter *adapter, struct sk_buff *skb)
 		return -1;
 	}
 
-	if (mwifiex_map_pci_memory(adapter, skb, skb->len , PCI_DMA_TODEVICE))
+	if (mwifiex_map_pci_memory(adapter, skb, skb->len, PCI_DMA_TODEVICE))
 		return -1;
 
 	buf_pa = MWIFIEX_SKB_DMA_ADDR(skb);
diff --git a/drivers/net/wireless/marvell/mwifiex/uap_cmd.c b/drivers/net/wireless/marvell/mwifiex/uap_cmd.c
index e791166..16d95b2 100644
--- a/drivers/net/wireless/marvell/mwifiex/uap_cmd.c
+++ b/drivers/net/wireless/marvell/mwifiex/uap_cmd.c
@@ -192,7 +192,7 @@ mwifiex_set_ht_params(struct mwifiex_private *priv,
 		}
 		priv->ap_11n_enabled = 1;
 	} else {
-		memset(&bss_cfg->ht_cap , 0, sizeof(struct ieee80211_ht_cap));
+		memset(&bss_cfg->ht_cap, 0, sizeof(struct ieee80211_ht_cap));
 		bss_cfg->ht_cap.cap_info = cpu_to_le16(MWIFIEX_DEF_HT_CAP);
 		bss_cfg->ht_cap.ampdu_params_info = MWIFIEX_DEF_AMPDU;
 	}
-- 
cgit v0.10.2


From 354a1947a02528b19ee326e230cb44399cd31865 Mon Sep 17 00:00:00 2001
From: Ujjal Roy <royujjal@gmail.com>
Date: Thu, 25 Feb 2016 00:49:56 +0530
Subject: mwifiex: Added missing spaces around brackets

This patch fixes the missing spaces issue in coding style.

Signed-off-by: Ujjal Roy <royujjal@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/usb.c b/drivers/net/wireless/marvell/mwifiex/usb.c
index e43aff9..0510861 100644
--- a/drivers/net/wireless/marvell/mwifiex/usb.c
+++ b/drivers/net/wireless/marvell/mwifiex/usb.c
@@ -244,9 +244,9 @@ setup_for_next:
 	if (card->rx_cmd_ep == context->ep) {
 		mwifiex_usb_submit_rx_urb(context, size);
 	} else {
-		if (atomic_read(&adapter->rx_pending) <= HIGH_RX_PENDING){
+		if (atomic_read(&adapter->rx_pending) <= HIGH_RX_PENDING) {
 			mwifiex_usb_submit_rx_urb(context, size);
-		}else{
+		} else {
 			context->skb = NULL;
 		}
 	}
-- 
cgit v0.10.2


From 2b02a36d12192f2a86388913143cd1e399eb971c Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Thu, 25 Feb 2016 23:24:27 +0000
Subject: mt7601u: do not free dma_buf when ivp allocation fails

If the allocation of ivp fails the error handling attempts to
free an uninitialized dma_buf; this data structure just contains
garbage on the stack, so the freeing will cause issues when the
urb, buf and dma fields are free'd. Fix this by not free'ing the
dma_buf if the ivp allocation fails.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/mediatek/mt7601u/mcu.c b/drivers/net/wireless/mediatek/mt7601u/mcu.c
index fbb1986..91c4b34 100644
--- a/drivers/net/wireless/mediatek/mt7601u/mcu.c
+++ b/drivers/net/wireless/mediatek/mt7601u/mcu.c
@@ -362,7 +362,9 @@ mt7601u_upload_firmware(struct mt7601u_dev *dev, const struct mt76_fw *fw)
 	int i, ret;
 
 	ivb = kmemdup(fw->ivb, sizeof(fw->ivb), GFP_KERNEL);
-	if (!ivb || mt7601u_usb_alloc_buf(dev, MCU_FW_URB_SIZE, &dma_buf)) {
+	if (!ivb)
+		return -ENOMEM;
+	if (mt7601u_usb_alloc_buf(dev, MCU_FW_URB_SIZE, &dma_buf)) {
 		ret = -ENOMEM;
 		goto error;
 	}
-- 
cgit v0.10.2


From 2e62f9b2a41e4ade1a0bb3c1bbda4defe4c67243 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Fri, 12 Feb 2016 10:15:43 +0100
Subject: bcma: drop unneeded fields from bcma_pflash struct
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Most of info stored in this struct wasn't really used anywhere as we put
all that data in platform data & resource as well.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c
index a40a203..20c134c 100644
--- a/drivers/bcma/driver_mips.c
+++ b/drivers/bcma/driver_mips.c
@@ -288,18 +288,15 @@ static void bcma_core_mips_flash_detect(struct bcma_drv_mips *mcore)
 	case BCMA_CC_FLASHT_PARA:
 		bcma_debug(bus, "Found parallel flash\n");
 		pflash->present = true;
-		pflash->window = BCMA_SOC_FLASH2;
-		pflash->window_size = BCMA_SOC_FLASH2_SZ;
 
 		if ((bcma_read32(cc->core, BCMA_CC_FLASH_CFG) &
 		     BCMA_CC_FLASH_CFG_DS) == 0)
-			pflash->buswidth = 1;
+			bcma_pflash_data.width = 1;
 		else
-			pflash->buswidth = 2;
+			bcma_pflash_data.width = 2;
 
-		bcma_pflash_data.width = pflash->buswidth;
-		bcma_pflash_resource.start = pflash->window;
-		bcma_pflash_resource.end = pflash->window + pflash->window_size;
+		bcma_pflash_resource.start = BCMA_SOC_FLASH2;
+		bcma_pflash_resource.end = BCMA_SOC_FLASH2 + BCMA_SOC_FLASH2_SZ;
 
 		break;
 	default:
diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h
index 700d0c6..16eaaad 100644
--- a/include/linux/bcma/bcma_driver_chipcommon.h
+++ b/include/linux/bcma/bcma_driver_chipcommon.h
@@ -579,9 +579,6 @@ struct bcma_chipcommon_pmu {
 #ifdef CONFIG_BCMA_DRIVER_MIPS
 struct bcma_pflash {
 	bool present;
-	u8 buswidth;
-	u32 window;
-	u32 window_size;
 };
 
 #ifdef CONFIG_BCMA_SFLASH
-- 
cgit v0.10.2


From d6a3b51ada68c2bd3e184f4729ce626a1721cf74 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Fri, 12 Feb 2016 10:15:44 +0100
Subject: bcma: move parallel flash support to separated file
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This follows the way of handling other flashes and cleans code a bit. As
next task we will want to move flash code to ChipCommon driver as:
1) Flash controllers are accesible using ChipCommon registers
2) This code isn't MIPS specific
This change prepares bcma for that.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/Kconfig b/drivers/bcma/Kconfig
index 023d448..efdc2ae 100644
--- a/drivers/bcma/Kconfig
+++ b/drivers/bcma/Kconfig
@@ -70,6 +70,11 @@ config BCMA_DRIVER_MIPS
 
 	  If unsure, say N
 
+config BCMA_PFLASH
+	bool
+	depends on BCMA_DRIVER_MIPS
+	default y
+
 config BCMA_SFLASH
 	bool
 	depends on BCMA_DRIVER_MIPS
diff --git a/drivers/bcma/Makefile b/drivers/bcma/Makefile
index f32af9b..087948a 100644
--- a/drivers/bcma/Makefile
+++ b/drivers/bcma/Makefile
@@ -1,6 +1,7 @@
 bcma-y					+= main.o scan.o core.o sprom.o
 bcma-y					+= driver_chipcommon.o driver_chipcommon_pmu.o
 bcma-y					+= driver_chipcommon_b.o
+bcma-$(CONFIG_BCMA_PFLASH)		+= driver_chipcommon_pflash.o
 bcma-$(CONFIG_BCMA_SFLASH)		+= driver_chipcommon_sflash.o
 bcma-$(CONFIG_BCMA_NFLASH)		+= driver_chipcommon_nflash.o
 bcma-$(CONFIG_BCMA_DRIVER_PCI)		+= driver_pci.o
diff --git a/drivers/bcma/bcma_private.h b/drivers/bcma/bcma_private.h
index 7e4ddfb..eda0909 100644
--- a/drivers/bcma/bcma_private.h
+++ b/drivers/bcma/bcma_private.h
@@ -47,9 +47,6 @@ int bcma_sprom_get(struct bcma_bus *bus);
 void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc);
 void bcma_core_chipcommon_init(struct bcma_drv_cc *cc);
 void bcma_chipco_bcm4331_ext_pa_lines_ctl(struct bcma_drv_cc *cc, bool enable);
-#ifdef CONFIG_BCMA_DRIVER_MIPS
-extern struct platform_device bcma_pflash_dev;
-#endif /* CONFIG_BCMA_DRIVER_MIPS */
 
 /* driver_chipcommon_b.c */
 int bcma_core_chipcommon_b_init(struct bcma_drv_cc_b *ccb);
@@ -61,6 +58,21 @@ void bcma_pmu_init(struct bcma_drv_cc *cc);
 u32 bcma_pmu_get_alp_clock(struct bcma_drv_cc *cc);
 u32 bcma_pmu_get_cpu_clock(struct bcma_drv_cc *cc);
 
+/**************************************************
+ * driver_chipcommon_sflash.c
+ **************************************************/
+
+#ifdef CONFIG_BCMA_PFLASH
+extern struct platform_device bcma_pflash_dev;
+int bcma_pflash_init(struct bcma_drv_cc *cc);
+#else
+static inline int bcma_pflash_init(struct bcma_drv_cc *cc)
+{
+	bcma_err(cc->core->bus, "Parallel flash not supported\n");
+	return 0;
+}
+#endif /* CONFIG_BCMA_PFLASH */
+
 #ifdef CONFIG_BCMA_SFLASH
 /* driver_chipcommon_sflash.c */
 int bcma_sflash_init(struct bcma_drv_cc *cc);
diff --git a/drivers/bcma/driver_chipcommon_pflash.c b/drivers/bcma/driver_chipcommon_pflash.c
new file mode 100644
index 0000000..3b497c9
--- /dev/null
+++ b/drivers/bcma/driver_chipcommon_pflash.c
@@ -0,0 +1,49 @@
+/*
+ * Broadcom specific AMBA
+ * ChipCommon parallel flash
+ *
+ * Licensed under the GNU/GPL. See COPYING for details.
+ */
+
+#include "bcma_private.h"
+
+#include <linux/bcma/bcma.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+
+static const char * const part_probes[] = { "bcm47xxpart", NULL };
+
+static struct physmap_flash_data bcma_pflash_data = {
+	.part_probe_types	= part_probes,
+};
+
+static struct resource bcma_pflash_resource = {
+	.name	= "bcma_pflash",
+	.flags  = IORESOURCE_MEM,
+};
+
+struct platform_device bcma_pflash_dev = {
+	.name		= "physmap-flash",
+	.dev		= {
+		.platform_data  = &bcma_pflash_data,
+	},
+	.resource	= &bcma_pflash_resource,
+	.num_resources	= 1,
+};
+
+int bcma_pflash_init(struct bcma_drv_cc *cc)
+{
+	struct bcma_pflash *pflash = &cc->pflash;
+
+	pflash->present = true;
+
+	if (!(bcma_read32(cc->core, BCMA_CC_FLASH_CFG) & BCMA_CC_FLASH_CFG_DS))
+		bcma_pflash_data.width = 1;
+	else
+		bcma_pflash_data.width = 2;
+
+	bcma_pflash_resource.start = BCMA_SOC_FLASH2;
+	bcma_pflash_resource.end = BCMA_SOC_FLASH2 + BCMA_SOC_FLASH2_SZ;
+
+	return 0;
+}
diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c
index 20c134c..967b0e8 100644
--- a/drivers/bcma/driver_mips.c
+++ b/drivers/bcma/driver_mips.c
@@ -14,8 +14,6 @@
 
 #include <linux/bcma/bcma.h>
 
-#include <linux/mtd/physmap.h>
-#include <linux/platform_device.h>
 #include <linux/serial.h>
 #include <linux/serial_core.h>
 #include <linux/serial_reg.h>
@@ -32,26 +30,6 @@ enum bcma_boot_dev {
 	BCMA_BOOT_DEV_NAND,
 };
 
-static const char * const part_probes[] = { "bcm47xxpart", NULL };
-
-static struct physmap_flash_data bcma_pflash_data = {
-	.part_probe_types	= part_probes,
-};
-
-static struct resource bcma_pflash_resource = {
-	.name	= "bcma_pflash",
-	.flags  = IORESOURCE_MEM,
-};
-
-struct platform_device bcma_pflash_dev = {
-	.name		= "physmap-flash",
-	.dev		= {
-		.platform_data  = &bcma_pflash_data,
-	},
-	.resource	= &bcma_pflash_resource,
-	.num_resources	= 1,
-};
-
 /* The 47162a0 hangs when reading MIPS DMP registers registers */
 static inline bool bcma_core_mips_bcm47162a0_quirk(struct bcma_device *dev)
 {
@@ -276,7 +254,6 @@ static void bcma_core_mips_flash_detect(struct bcma_drv_mips *mcore)
 {
 	struct bcma_bus *bus = mcore->core->bus;
 	struct bcma_drv_cc *cc = &bus->drv_cc;
-	struct bcma_pflash *pflash = &cc->pflash;
 	enum bcma_boot_dev boot_dev;
 
 	switch (cc->capabilities & BCMA_CC_CAP_FLASHT) {
@@ -287,17 +264,7 @@ static void bcma_core_mips_flash_detect(struct bcma_drv_mips *mcore)
 		break;
 	case BCMA_CC_FLASHT_PARA:
 		bcma_debug(bus, "Found parallel flash\n");
-		pflash->present = true;
-
-		if ((bcma_read32(cc->core, BCMA_CC_FLASH_CFG) &
-		     BCMA_CC_FLASH_CFG_DS) == 0)
-			bcma_pflash_data.width = 1;
-		else
-			bcma_pflash_data.width = 2;
-
-		bcma_pflash_resource.start = BCMA_SOC_FLASH2;
-		bcma_pflash_resource.end = BCMA_SOC_FLASH2 + BCMA_SOC_FLASH2_SZ;
-
+		bcma_pflash_init(cc);
 		break;
 	default:
 		bcma_err(bus, "Flash type not supported\n");
diff --git a/drivers/bcma/main.c b/drivers/bcma/main.c
index c466f75..786be8f 100644
--- a/drivers/bcma/main.c
+++ b/drivers/bcma/main.c
@@ -350,7 +350,7 @@ static int bcma_register_devices(struct bcma_bus *bus)
 		bcma_register_core(bus, core);
 	}
 
-#ifdef CONFIG_BCMA_DRIVER_MIPS
+#ifdef CONFIG_BCMA_PFLASH
 	if (bus->drv_cc.pflash.present) {
 		err = platform_device_register(&bcma_pflash_dev);
 		if (err)
diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h
index 16eaaad..846513c 100644
--- a/include/linux/bcma/bcma_driver_chipcommon.h
+++ b/include/linux/bcma/bcma_driver_chipcommon.h
@@ -576,10 +576,11 @@ struct bcma_chipcommon_pmu {
 	u32 crystalfreq;	/* The active crystal frequency (in kHz) */
 };
 
-#ifdef CONFIG_BCMA_DRIVER_MIPS
+#ifdef CONFIG_BCMA_PFLASH
 struct bcma_pflash {
 	bool present;
 };
+#endif
 
 #ifdef CONFIG_BCMA_SFLASH
 struct mtd_info;
@@ -603,6 +604,7 @@ struct bcma_nflash {
 };
 #endif
 
+#ifdef CONFIG_BCMA_DRIVER_MIPS
 struct bcma_serial_port {
 	void *regs;
 	unsigned long clockspeed;
@@ -622,8 +624,9 @@ struct bcma_drv_cc {
 	/* Fast Powerup Delay constant */
 	u16 fast_pwrup_delay;
 	struct bcma_chipcommon_pmu pmu;
-#ifdef CONFIG_BCMA_DRIVER_MIPS
+#ifdef CONFIG_BCMA_PFLASH
 	struct bcma_pflash pflash;
+#endif
 #ifdef CONFIG_BCMA_SFLASH
 	struct bcma_sflash sflash;
 #endif
@@ -631,6 +634,7 @@ struct bcma_drv_cc {
 	struct bcma_nflash nflash;
 #endif
 
+#ifdef CONFIG_BCMA_DRIVER_MIPS
 	int nr_serial_ports;
 	struct bcma_serial_port serial_ports[4];
 #endif /* CONFIG_BCMA_DRIVER_MIPS */
-- 
cgit v0.10.2


From 0ea6f0c582c2675285ad094df0137f4f0de47869 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <zajec5@gmail.com>
Date: Fri, 12 Feb 2016 10:15:45 +0100
Subject: bcma: move flash detection code to ChipCommon core driver
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Flashes are attached and accessible through ChipCommon core, they aren't
connected to MIPS one. Moreover some new ARM devices (e.g. BCM47189) may
have serial flash accessibility using ChipCommon registers as well. To
support them we can't keep this code in MIPS core driver.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/bcma/driver_chipcommon.c b/drivers/bcma/driver_chipcommon.c
index b0f44a2..921ce18 100644
--- a/drivers/bcma/driver_chipcommon.c
+++ b/drivers/bcma/driver_chipcommon.c
@@ -115,6 +115,33 @@ int bcma_chipco_watchdog_register(struct bcma_drv_cc *cc)
 	return 0;
 }
 
+static void bcma_core_chipcommon_flash_detect(struct bcma_drv_cc *cc)
+{
+	struct bcma_bus *bus = cc->core->bus;
+
+	switch (cc->capabilities & BCMA_CC_CAP_FLASHT) {
+	case BCMA_CC_FLASHT_STSER:
+	case BCMA_CC_FLASHT_ATSER:
+		bcma_debug(bus, "Found serial flash\n");
+		bcma_sflash_init(cc);
+		break;
+	case BCMA_CC_FLASHT_PARA:
+		bcma_debug(bus, "Found parallel flash\n");
+		bcma_pflash_init(cc);
+		break;
+	default:
+		bcma_err(bus, "Flash type not supported\n");
+	}
+
+	if (cc->core->id.rev == 38 ||
+	    bus->chipinfo.id == BCMA_CHIP_ID_BCM4706) {
+		if (cc->capabilities & BCMA_CC_CAP_NFLASH) {
+			bcma_debug(bus, "Found NAND flash\n");
+			bcma_nflash_init(cc);
+		}
+	}
+}
+
 void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc)
 {
 	struct bcma_bus *bus = cc->core->bus;
@@ -136,6 +163,9 @@ void bcma_core_chipcommon_early_init(struct bcma_drv_cc *cc)
 	if (IS_BUILTIN(CONFIG_BCM47XX) && bus->hosttype == BCMA_HOSTTYPE_SOC)
 		bcma_chipco_serial_init(cc);
 
+	if (bus->hosttype == BCMA_HOSTTYPE_SOC)
+		bcma_core_chipcommon_flash_detect(cc);
+
 	cc->early_setup_done = true;
 }
 
diff --git a/drivers/bcma/driver_mips.c b/drivers/bcma/driver_mips.c
index 967b0e8..96f1713 100644
--- a/drivers/bcma/driver_mips.c
+++ b/drivers/bcma/driver_mips.c
@@ -250,34 +250,11 @@ static enum bcma_boot_dev bcma_boot_dev(struct bcma_bus *bus)
 	return BCMA_BOOT_DEV_SERIAL;
 }
 
-static void bcma_core_mips_flash_detect(struct bcma_drv_mips *mcore)
+static void bcma_core_mips_nvram_init(struct bcma_drv_mips *mcore)
 {
 	struct bcma_bus *bus = mcore->core->bus;
-	struct bcma_drv_cc *cc = &bus->drv_cc;
 	enum bcma_boot_dev boot_dev;
 
-	switch (cc->capabilities & BCMA_CC_CAP_FLASHT) {
-	case BCMA_CC_FLASHT_STSER:
-	case BCMA_CC_FLASHT_ATSER:
-		bcma_debug(bus, "Found serial flash\n");
-		bcma_sflash_init(cc);
-		break;
-	case BCMA_CC_FLASHT_PARA:
-		bcma_debug(bus, "Found parallel flash\n");
-		bcma_pflash_init(cc);
-		break;
-	default:
-		bcma_err(bus, "Flash type not supported\n");
-	}
-
-	if (cc->core->id.rev == 38 ||
-	    bus->chipinfo.id == BCMA_CHIP_ID_BCM4706) {
-		if (cc->capabilities & BCMA_CC_CAP_NFLASH) {
-			bcma_debug(bus, "Found NAND flash\n");
-			bcma_nflash_init(cc);
-		}
-	}
-
 	/* Determine flash type this SoC boots from */
 	boot_dev = bcma_boot_dev(bus);
 	switch (boot_dev) {
@@ -304,7 +281,7 @@ void bcma_core_mips_early_init(struct bcma_drv_mips *mcore)
 	if (mcore->early_setup_done)
 		return;
 
-	bcma_core_mips_flash_detect(mcore);
+	bcma_core_mips_nvram_init(mcore);
 
 	mcore->early_setup_done = true;
 }
-- 
cgit v0.10.2


From 795f3512ca952965dd2546ceed3d9636abcb7920 Mon Sep 17 00:00:00 2001
From: Wei Tang <tangwei@cmss.chinamobile.com>
Date: Mon, 7 Mar 2016 14:25:10 +0800
Subject: netlink: do not initialise statics to 0 or NULL

This patch fixes the checkpatch.pl error to netlabel_unlabeled.c:

ERROR: do not initialise statics to 0 or NULL

Signed-off-by: Wei Tang <tangwei@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index b038092..9eaa9a1 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -116,11 +116,11 @@ struct netlbl_unlhsh_walk_arg {
 static DEFINE_SPINLOCK(netlbl_unlhsh_lock);
 #define netlbl_unlhsh_rcu_deref(p) \
 	rcu_dereference_check(p, lockdep_is_held(&netlbl_unlhsh_lock))
-static struct netlbl_unlhsh_tbl *netlbl_unlhsh = NULL;
-static struct netlbl_unlhsh_iface *netlbl_unlhsh_def = NULL;
+static struct netlbl_unlhsh_tbl *netlbl_unlhsh;
+static struct netlbl_unlhsh_iface *netlbl_unlhsh_def;
 
 /* Accept unlabeled packets flag */
-static u8 netlabel_unlabel_acceptflg = 0;
+static u8 netlabel_unlabel_acceptflg;
 
 /* NetLabel Generic NETLINK unlabeled family */
 static struct genl_family netlbl_unlabel_gnl_family = {
-- 
cgit v0.10.2


From 8303394d811ad4e1245b2d72e15c39880ec3d0ff Mon Sep 17 00:00:00 2001
From: Wei Tang <tangwei@cmss.chinamobile.com>
Date: Mon, 7 Mar 2016 14:29:53 +0800
Subject: netlabel: do not initialise statics to NULL

This patch fixes the checkpatch.pl error to netlabel_domainhash.c:

ERROR: do not initialise statics to NULL

Signed-off-by: Wei Tang <tangwei@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index f0cb92f..ada6742 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -55,8 +55,8 @@ struct netlbl_domhsh_tbl {
 static DEFINE_SPINLOCK(netlbl_domhsh_lock);
 #define netlbl_domhsh_rcu_deref(p) \
 	rcu_dereference_check(p, lockdep_is_held(&netlbl_domhsh_lock))
-static struct netlbl_domhsh_tbl *netlbl_domhsh = NULL;
-static struct netlbl_dom_map *netlbl_domhsh_def = NULL;
+static struct netlbl_domhsh_tbl *netlbl_domhsh;
+static struct netlbl_dom_map *netlbl_domhsh_def;
 
 /*
  * Domain Hash Table Helper Functions
-- 
cgit v0.10.2


From 8dfd329fbc240729938d24bf87aca49ea89289c5 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Fri, 4 Mar 2016 14:07:54 +0000
Subject: arp: correct return value of arp_rcv

Currently, arp_rcv() always return zero on a packet delivery upcall.

To make its behavior more compliant with the way this API should be
used, this patch changes this to let it return NET_RX_SUCCESS when the
packet is proper handled, and NET_RX_DROP otherwise.

v1->v2:
If sanity check is failed, call kfree_skb() instead of consume_skb(), then
return the correct return value.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index c102eb5..c34c754 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -665,7 +665,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 	 */
 
 	if (!in_dev)
-		goto out;
+		goto out_free_skb;
 
 	arp = arp_hdr(skb);
 
@@ -673,7 +673,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 	default:
 		if (arp->ar_pro != htons(ETH_P_IP) ||
 		    htons(dev_type) != arp->ar_hrd)
-			goto out;
+			goto out_free_skb;
 		break;
 	case ARPHRD_ETHER:
 	case ARPHRD_FDDI:
@@ -690,17 +690,17 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 		if ((arp->ar_hrd != htons(ARPHRD_ETHER) &&
 		     arp->ar_hrd != htons(ARPHRD_IEEE802)) ||
 		    arp->ar_pro != htons(ETH_P_IP))
-			goto out;
+			goto out_free_skb;
 		break;
 	case ARPHRD_AX25:
 		if (arp->ar_pro != htons(AX25_P_IP) ||
 		    arp->ar_hrd != htons(ARPHRD_AX25))
-			goto out;
+			goto out_free_skb;
 		break;
 	case ARPHRD_NETROM:
 		if (arp->ar_pro != htons(AX25_P_IP) ||
 		    arp->ar_hrd != htons(ARPHRD_NETROM))
-			goto out;
+			goto out_free_skb;
 		break;
 	}
 
@@ -708,7 +708,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 
 	if (arp->ar_op != htons(ARPOP_REPLY) &&
 	    arp->ar_op != htons(ARPOP_REQUEST))
-		goto out;
+		goto out_free_skb;
 
 /*
  *	Extract fields
@@ -733,7 +733,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
  */
 	if (ipv4_is_multicast(tip) ||
 	    (!IN_DEV_ROUTE_LOCALNET(in_dev) && ipv4_is_loopback(tip)))
-		goto out;
+		goto out_free_skb;
 
  /*
   *	For some 802.11 wireless deployments (and possibly other networks),
@@ -741,7 +741,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
   *	and thus should not be accepted.
   */
 	if (sip == tip && IN_DEV_ORCONF(in_dev, DROP_GRATUITOUS_ARP))
-		goto out;
+		goto out_free_skb;
 
 /*
  *     Special case: We must set Frame Relay source Q.922 address
@@ -778,7 +778,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 		    !arp_ignore(in_dev, sip, tip))
 			arp_send_dst(ARPOP_REPLY, ETH_P_ARP, sip, dev, tip,
 				     sha, dev->dev_addr, sha, reply_dst);
-		goto out;
+		goto out_consume_skb;
 	}
 
 	if (arp->ar_op == htons(ARPOP_REQUEST) &&
@@ -803,7 +803,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 					neigh_release(n);
 				}
 			}
-			goto out;
+			goto out_consume_skb;
 		} else if (IN_DEV_FORWARD(in_dev)) {
 			if (addr_type == RTN_UNICAST  &&
 			    (arp_fwd_proxy(in_dev, dev, rt) ||
@@ -826,7 +826,7 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 						       in_dev->arp_parms, skb);
 					goto out_free_dst;
 				}
-				goto out;
+				goto out_consume_skb;
 			}
 		}
 	}
@@ -876,11 +876,16 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb)
 		neigh_release(n);
 	}
 
-out:
+out_consume_skb:
 	consume_skb(skb);
+
 out_free_dst:
 	dst_release(reply_dst);
-	return 0;
+	return NET_RX_SUCCESS;
+
+out_free_skb:
+	kfree_skb(skb);
+	return NET_RX_DROP;
 }
 
 static void parp_redo(struct sk_buff *skb)
@@ -924,11 +929,11 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev,
 
 consumeskb:
 	consume_skb(skb);
-	return 0;
+	return NET_RX_SUCCESS;
 freeskb:
 	kfree_skb(skb);
 out_of_mem:
-	return 0;
+	return NET_RX_DROP;
 }
 
 /*
-- 
cgit v0.10.2


From 49cc66eaee19e772997b63b057ea4b4bf7d48db0 Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Fri, 4 Mar 2016 17:04:42 +0100
Subject: tipc: move netlink policies to netlink.c

Make the c files less cluttered and enable netlink attributes to be
shared between files.

Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index 802ffad..27a5406 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -40,6 +40,7 @@
 #include "link.h"
 #include "discover.h"
 #include "bcast.h"
+#include "netlink.h"
 
 #define MAX_ADDR_STR 60
 
@@ -54,23 +55,6 @@ static struct tipc_media * const media_info_array[] = {
 	NULL
 };
 
-static const struct nla_policy
-tipc_nl_bearer_policy[TIPC_NLA_BEARER_MAX + 1]	= {
-	[TIPC_NLA_BEARER_UNSPEC]		= { .type = NLA_UNSPEC },
-	[TIPC_NLA_BEARER_NAME] = {
-		.type = NLA_STRING,
-		.len = TIPC_MAX_BEARER_NAME
-	},
-	[TIPC_NLA_BEARER_PROP]			= { .type = NLA_NESTED },
-	[TIPC_NLA_BEARER_DOMAIN]		= { .type = NLA_U32 }
-};
-
-static const struct nla_policy tipc_nl_media_policy[TIPC_NLA_MEDIA_MAX + 1] = {
-	[TIPC_NLA_MEDIA_UNSPEC]		= { .type = NLA_UNSPEC },
-	[TIPC_NLA_MEDIA_NAME]		= { .type = NLA_STRING },
-	[TIPC_NLA_MEDIA_PROP]		= { .type = NLA_NESTED }
-};
-
 static void bearer_disable(struct net *net, struct tipc_bearer *b);
 
 /**
diff --git a/net/tipc/link.c b/net/tipc/link.c
index fc2fa83..7d2bb3e 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -192,14 +192,6 @@ struct tipc_link {
 static const char *link_co_err = "Link tunneling error, ";
 static const char *link_rst_msg = "Resetting link ";
 
-/* Properties valid for media, bearar and link */
-static const struct nla_policy tipc_nl_prop_policy[TIPC_NLA_PROP_MAX + 1] = {
-	[TIPC_NLA_PROP_UNSPEC]		= { .type = NLA_UNSPEC },
-	[TIPC_NLA_PROP_PRIO]		= { .type = NLA_U32 },
-	[TIPC_NLA_PROP_TOL]		= { .type = NLA_U32 },
-	[TIPC_NLA_PROP_WIN]		= { .type = NLA_U32 }
-};
-
 /* Send states for broadcast NACKs
  */
 enum {
diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c
index 777b979..e190460 100644
--- a/net/tipc/name_table.c
+++ b/net/tipc/name_table.c
@@ -47,12 +47,6 @@
 
 #define TIPC_NAMETBL_SIZE 1024		/* must be a power of 2 */
 
-static const struct nla_policy
-tipc_nl_name_table_policy[TIPC_NLA_NAME_TABLE_MAX + 1] = {
-	[TIPC_NLA_NAME_TABLE_UNSPEC]	= { .type = NLA_UNSPEC },
-	[TIPC_NLA_NAME_TABLE_PUBL]	= { .type = NLA_NESTED }
-};
-
 /**
  * struct name_info - name sequence publication info
  * @node_list: circular list of publications made by own node
diff --git a/net/tipc/net.c b/net/tipc/net.c
index 86d68b3..28bf4fe 100644
--- a/net/tipc/net.c
+++ b/net/tipc/net.c
@@ -41,11 +41,7 @@
 #include "socket.h"
 #include "node.h"
 #include "bcast.h"
-
-static const struct nla_policy tipc_nl_net_policy[TIPC_NLA_NET_MAX + 1] = {
-	[TIPC_NLA_NET_UNSPEC]	= { .type = NLA_UNSPEC },
-	[TIPC_NLA_NET_ID]	= { .type = NLA_U32 }
-};
+#include "netlink.h"
 
 /*
  * The TIPC locking policy is designed to ensure a very fine locking
diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c
index 8975b01..56935df2 100644
--- a/net/tipc/netlink.c
+++ b/net/tipc/netlink.c
@@ -55,6 +55,75 @@ static const struct nla_policy tipc_nl_policy[TIPC_NLA_MAX + 1] = {
 	[TIPC_NLA_NAME_TABLE]	= { .type = NLA_NESTED, }
 };
 
+const struct nla_policy
+tipc_nl_name_table_policy[TIPC_NLA_NAME_TABLE_MAX + 1] = {
+	[TIPC_NLA_NAME_TABLE_UNSPEC]	= { .type = NLA_UNSPEC },
+	[TIPC_NLA_NAME_TABLE_PUBL]	= { .type = NLA_NESTED }
+};
+
+const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = {
+	[TIPC_NLA_SOCK_UNSPEC]		= { .type = NLA_UNSPEC },
+	[TIPC_NLA_SOCK_ADDR]		= { .type = NLA_U32 },
+	[TIPC_NLA_SOCK_REF]		= { .type = NLA_U32 },
+	[TIPC_NLA_SOCK_CON]		= { .type = NLA_NESTED },
+	[TIPC_NLA_SOCK_HAS_PUBL]	= { .type = NLA_FLAG }
+};
+
+const struct nla_policy tipc_nl_net_policy[TIPC_NLA_NET_MAX + 1] = {
+	[TIPC_NLA_NET_UNSPEC]		= { .type = NLA_UNSPEC },
+	[TIPC_NLA_NET_ID]		= { .type = NLA_U32 }
+};
+
+const struct nla_policy tipc_nl_link_policy[TIPC_NLA_LINK_MAX + 1] = {
+	[TIPC_NLA_LINK_UNSPEC]		= { .type = NLA_UNSPEC },
+	[TIPC_NLA_LINK_NAME]		= { .type = NLA_STRING,
+					    .len = TIPC_MAX_LINK_NAME },
+	[TIPC_NLA_LINK_MTU]		= { .type = NLA_U32 },
+	[TIPC_NLA_LINK_BROADCAST]	= { .type = NLA_FLAG },
+	[TIPC_NLA_LINK_UP]		= { .type = NLA_FLAG },
+	[TIPC_NLA_LINK_ACTIVE]		= { .type = NLA_FLAG },
+	[TIPC_NLA_LINK_PROP]		= { .type = NLA_NESTED },
+	[TIPC_NLA_LINK_STATS]		= { .type = NLA_NESTED },
+	[TIPC_NLA_LINK_RX]		= { .type = NLA_U32 },
+	[TIPC_NLA_LINK_TX]		= { .type = NLA_U32 }
+};
+
+const struct nla_policy tipc_nl_node_policy[TIPC_NLA_NODE_MAX + 1] = {
+	[TIPC_NLA_NODE_UNSPEC]		= { .type = NLA_UNSPEC },
+	[TIPC_NLA_NODE_ADDR]		= { .type = NLA_U32 },
+	[TIPC_NLA_NODE_UP]		= { .type = NLA_FLAG }
+};
+
+/* Properties valid for media, bearer and link */
+const struct nla_policy tipc_nl_prop_policy[TIPC_NLA_PROP_MAX + 1] = {
+	[TIPC_NLA_PROP_UNSPEC]		= { .type = NLA_UNSPEC },
+	[TIPC_NLA_PROP_PRIO]		= { .type = NLA_U32 },
+	[TIPC_NLA_PROP_TOL]		= { .type = NLA_U32 },
+	[TIPC_NLA_PROP_WIN]		= { .type = NLA_U32 }
+};
+
+const struct nla_policy tipc_nl_bearer_policy[TIPC_NLA_BEARER_MAX + 1]	= {
+	[TIPC_NLA_BEARER_UNSPEC]	= { .type = NLA_UNSPEC },
+	[TIPC_NLA_BEARER_NAME]		= { .type = NLA_STRING,
+					    .len = TIPC_MAX_BEARER_NAME },
+	[TIPC_NLA_BEARER_PROP]		= { .type = NLA_NESTED },
+	[TIPC_NLA_BEARER_DOMAIN]	= { .type = NLA_U32 }
+};
+
+const struct nla_policy tipc_nl_media_policy[TIPC_NLA_MEDIA_MAX + 1] = {
+	[TIPC_NLA_MEDIA_UNSPEC]		= { .type = NLA_UNSPEC },
+	[TIPC_NLA_MEDIA_NAME]		= { .type = NLA_STRING },
+	[TIPC_NLA_MEDIA_PROP]		= { .type = NLA_NESTED }
+};
+
+const struct nla_policy tipc_nl_udp_policy[TIPC_NLA_UDP_MAX + 1] = {
+	[TIPC_NLA_UDP_UNSPEC]	= {.type = NLA_UNSPEC},
+	[TIPC_NLA_UDP_LOCAL]	= {.type = NLA_BINARY,
+				   .len = sizeof(struct sockaddr_storage)},
+	[TIPC_NLA_UDP_REMOTE]	= {.type = NLA_BINARY,
+				   .len = sizeof(struct sockaddr_storage)},
+};
+
 /* Users of the legacy API (tipc-config) can't handle that we add operations,
  * so we have a separate genl handling for the new API.
  */
diff --git a/net/tipc/netlink.h b/net/tipc/netlink.h
index 08a1db6..ed1dbcb 100644
--- a/net/tipc/netlink.h
+++ b/net/tipc/netlink.h
@@ -35,6 +35,7 @@
 
 #ifndef _TIPC_NETLINK_H
 #define _TIPC_NETLINK_H
+#include <net/netlink.h>
 
 extern struct genl_family tipc_genl_family;
 int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***buf);
@@ -45,6 +46,16 @@ struct tipc_nl_msg {
 	u32 seq;
 };
 
+extern const struct nla_policy tipc_nl_name_table_policy[];
+extern const struct nla_policy tipc_nl_sock_policy[];
+extern const struct nla_policy tipc_nl_net_policy[];
+extern const struct nla_policy tipc_nl_link_policy[];
+extern const struct nla_policy tipc_nl_node_policy[];
+extern const struct nla_policy tipc_nl_prop_policy[];
+extern const struct nla_policy tipc_nl_bearer_policy[];
+extern const struct nla_policy tipc_nl_media_policy[];
+extern const struct nla_policy tipc_nl_udp_policy[];
+
 int tipc_netlink_start(void);
 int tipc_netlink_compat_start(void);
 void tipc_netlink_stop(void);
diff --git a/net/tipc/node.c b/net/tipc/node.c
index 590d597..ace178f 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -41,6 +41,7 @@
 #include "socket.h"
 #include "bcast.h"
 #include "discover.h"
+#include "netlink.h"
 
 #define INVALID_NODE_SIG	0x10000
 
@@ -164,28 +165,6 @@ struct tipc_sock_conn {
 	struct list_head list;
 };
 
-static const struct nla_policy tipc_nl_link_policy[TIPC_NLA_LINK_MAX + 1] = {
-	[TIPC_NLA_LINK_UNSPEC]		= { .type = NLA_UNSPEC },
-	[TIPC_NLA_LINK_NAME] = {
-		.type = NLA_STRING,
-		.len = TIPC_MAX_LINK_NAME
-	},
-	[TIPC_NLA_LINK_MTU]		= { .type = NLA_U32 },
-	[TIPC_NLA_LINK_BROADCAST]	= { .type = NLA_FLAG },
-	[TIPC_NLA_LINK_UP]		= { .type = NLA_FLAG },
-	[TIPC_NLA_LINK_ACTIVE]		= { .type = NLA_FLAG },
-	[TIPC_NLA_LINK_PROP]		= { .type = NLA_NESTED },
-	[TIPC_NLA_LINK_STATS]		= { .type = NLA_NESTED },
-	[TIPC_NLA_LINK_RX]		= { .type = NLA_U32 },
-	[TIPC_NLA_LINK_TX]		= { .type = NLA_U32 }
-};
-
-static const struct nla_policy tipc_nl_node_policy[TIPC_NLA_NODE_MAX + 1] = {
-	[TIPC_NLA_NODE_UNSPEC]		= { .type = NLA_UNSPEC },
-	[TIPC_NLA_NODE_ADDR]		= { .type = NLA_U32 },
-	[TIPC_NLA_NODE_UP]		= { .type = NLA_FLAG }
-};
-
 static struct tipc_link *node_active_link(struct tipc_node *n, int sel)
 {
 	int bearer_id = n->active_links[sel & 1];
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 69c2905..56b8a96 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -42,6 +42,7 @@
 #include "name_distr.h"
 #include "socket.h"
 #include "bcast.h"
+#include "netlink.h"
 
 #define SS_LISTENING		-1	/* socket is listening */
 #define SS_READY		-2	/* socket is connectionless */
@@ -126,14 +127,6 @@ static const struct proto_ops stream_ops;
 static const struct proto_ops msg_ops;
 static struct proto tipc_proto;
 
-static const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = {
-	[TIPC_NLA_SOCK_UNSPEC]		= { .type = NLA_UNSPEC },
-	[TIPC_NLA_SOCK_ADDR]		= { .type = NLA_U32 },
-	[TIPC_NLA_SOCK_REF]		= { .type = NLA_U32 },
-	[TIPC_NLA_SOCK_CON]		= { .type = NLA_NESTED },
-	[TIPC_NLA_SOCK_HAS_PUBL]	= { .type = NLA_FLAG }
-};
-
 static const struct rhashtable_params tsk_rht_params;
 
 /*
diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index fb2f7ec6..49b3c2e 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -48,20 +48,13 @@
 #include <linux/tipc_netlink.h>
 #include "core.h"
 #include "bearer.h"
+#include "netlink.h"
 
 /* IANA assigned UDP port */
 #define UDP_PORT_DEFAULT	6118
 
 #define UDP_MIN_HEADROOM        28
 
-static const struct nla_policy tipc_nl_udp_policy[TIPC_NLA_UDP_MAX + 1] = {
-	[TIPC_NLA_UDP_UNSPEC]	= {.type = NLA_UNSPEC},
-	[TIPC_NLA_UDP_LOCAL]	= {.type = NLA_BINARY,
-				   .len = sizeof(struct sockaddr_storage)},
-	[TIPC_NLA_UDP_REMOTE]	= {.type = NLA_BINARY,
-				   .len = sizeof(struct sockaddr_storage)},
-};
-
 /**
  * struct udp_media_addr - IP/UDP addressing information
  *
-- 
cgit v0.10.2


From 088c86183012495b53ecc1c734909e5712a40b66 Mon Sep 17 00:00:00 2001
From: Manish Chopra <manish.chopra@qlogic.com>
Date: Fri, 4 Mar 2016 12:35:05 -0500
Subject: qed/qede: Add infrastructure support for hardware GRO

This patch adds mainly structures and APIs prototype changes
in order to give support for qede slowpath/fastpath support
for the same.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 592e0e6..236db8a 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -2919,7 +2919,19 @@ struct eth_vport_rx_mode {
 };
 
 struct eth_vport_tpa_param {
-	u64     reserved[2];
+	u8	tpa_ipv4_en_flg;
+	u8	tpa_ipv6_en_flg;
+	u8	tpa_ipv4_tunn_en_flg;
+	u8	tpa_ipv6_tunn_en_flg;
+	u8	tpa_pkt_split_flg;
+	u8	tpa_hdr_data_split_flg;
+	u8	tpa_gro_consistent_flg;
+	u8	tpa_max_aggs_num;
+	u16	tpa_max_size;
+	u16	tpa_min_size_to_start;
+	u16	tpa_min_size_to_cont;
+	u8	max_buff_num;
+	u8	reserved;
 };
 
 struct eth_vport_tx_mode {
diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index 102ddc7..3f35c6c 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -132,16 +132,29 @@ struct qed_sp_vport_update_params {
 	struct qed_filter_accept_flags	accept_flags;
 };
 
+enum qed_tpa_mode {
+	QED_TPA_MODE_NONE,
+	QED_TPA_MODE_UNUSED,
+	QED_TPA_MODE_GRO,
+	QED_TPA_MODE_MAX
+};
+
+struct qed_sp_vport_start_params {
+	enum qed_tpa_mode	tpa_mode;
+	bool			remove_inner_vlan;
+	bool			drop_ttl0;
+	u8			max_buffers_per_cqe;
+	u32			concrete_fid;
+	u16			opaque_fid;
+	u8			vport_id;
+	u16			mtu;
+};
+
 #define QED_MAX_SGES_NUM 16
 #define CRC32_POLY 0x1edc6f41
 
 static int qed_sp_vport_start(struct qed_hwfn *p_hwfn,
-			      u32 concrete_fid,
-			      u16 opaque_fid,
-			      u8 vport_id,
-			      u16 mtu,
-			      u8 drop_ttl0_flg,
-			      u8 inner_vlan_removal_en_flg)
+			      struct qed_sp_vport_start_params *p_params)
 {
 	struct vport_start_ramrod_data *p_ramrod = NULL;
 	struct qed_spq_entry *p_ent =  NULL;
@@ -150,13 +163,13 @@ static int qed_sp_vport_start(struct qed_hwfn *p_hwfn,
 	u16 rx_mode = 0;
 	u8 abs_vport_id = 0;
 
-	rc = qed_fw_vport(p_hwfn, vport_id, &abs_vport_id);
+	rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_vport_id);
 	if (rc != 0)
 		return rc;
 
 	memset(&init_data, 0, sizeof(init_data));
 	init_data.cid = qed_spq_get_cid(p_hwfn);
-	init_data.opaque_fid = opaque_fid;
+	init_data.opaque_fid = p_params->opaque_fid;
 	init_data.comp_mode = QED_SPQ_MODE_EBLOCK;
 
 	rc = qed_sp_init_request(p_hwfn, &p_ent,
@@ -168,9 +181,9 @@ static int qed_sp_vport_start(struct qed_hwfn *p_hwfn,
 	p_ramrod		= &p_ent->ramrod.vport_start;
 	p_ramrod->vport_id	= abs_vport_id;
 
-	p_ramrod->mtu			= cpu_to_le16(mtu);
-	p_ramrod->inner_vlan_removal_en = inner_vlan_removal_en_flg;
-	p_ramrod->drop_ttl0_en		= drop_ttl0_flg;
+	p_ramrod->mtu			= cpu_to_le16(p_params->mtu);
+	p_ramrod->inner_vlan_removal_en	= p_params->remove_inner_vlan;
+	p_ramrod->drop_ttl0_en		= p_params->drop_ttl0;
 
 	SET_FIELD(rx_mode, ETH_VPORT_RX_MODE_UCAST_DROP_ALL, 1);
 	SET_FIELD(rx_mode, ETH_VPORT_RX_MODE_MCAST_DROP_ALL, 1);
@@ -181,9 +194,26 @@ static int qed_sp_vport_start(struct qed_hwfn *p_hwfn,
 	memset(&p_ramrod->tpa_param, 0,
 	       sizeof(struct eth_vport_tpa_param));
 
+	p_ramrod->tpa_param.max_buff_num = p_params->max_buffers_per_cqe;
+
+	switch (p_params->tpa_mode) {
+	case QED_TPA_MODE_GRO:
+		p_ramrod->tpa_param.tpa_max_aggs_num = ETH_TPA_MAX_AGGS_NUM;
+		p_ramrod->tpa_param.tpa_max_size = (u16)-1;
+		p_ramrod->tpa_param.tpa_min_size_to_cont = p_params->mtu / 2;
+		p_ramrod->tpa_param.tpa_min_size_to_start = p_params->mtu / 2;
+		p_ramrod->tpa_param.tpa_ipv4_en_flg = 1;
+		p_ramrod->tpa_param.tpa_ipv6_en_flg = 1;
+		p_ramrod->tpa_param.tpa_pkt_split_flg = 1;
+		p_ramrod->tpa_param.tpa_gro_consistent_flg = 1;
+		break;
+	default:
+		break;
+	}
+
 	/* Software Function ID in hwfn (PFs are 0 - 15, VFs are 16 - 135) */
 	p_ramrod->sw_fid = qed_concrete_to_sw_fid(p_hwfn->cdev,
-						  concrete_fid);
+						  p_params->concrete_fid);
 
 	return qed_spq_post(p_hwfn, p_ent, NULL);
 }
@@ -1592,24 +1622,25 @@ static void qed_register_eth_ops(struct qed_dev *cdev,
 }
 
 static int qed_start_vport(struct qed_dev *cdev,
-			   u8 vport_id,
-			   u16 mtu,
-			   u8 drop_ttl0_flg,
-			   u8 inner_vlan_removal_en_flg)
+			   struct qed_start_vport_params *params)
 {
 	int rc, i;
 
 	for_each_hwfn(cdev, i) {
+		struct qed_sp_vport_start_params start = { 0 };
 		struct qed_hwfn *p_hwfn = &cdev->hwfns[i];
 
-		rc = qed_sp_vport_start(p_hwfn,
-					p_hwfn->hw_info.concrete_fid,
-					p_hwfn->hw_info.opaque_fid,
-					vport_id,
-					mtu,
-					drop_ttl0_flg,
-					inner_vlan_removal_en_flg);
-
+		start.tpa_mode = params->gro_enable ? QED_TPA_MODE_GRO :
+							QED_TPA_MODE_NONE;
+		start.remove_inner_vlan = params->remove_inner_vlan;
+		start.drop_ttl0 = params->drop_ttl0;
+		start.opaque_fid = p_hwfn->hw_info.opaque_fid;
+		start.concrete_fid = p_hwfn->hw_info.concrete_fid;
+		start.vport_id = params->vport_id;
+		start.max_buffers_per_cqe = 16;
+		start.mtu = params->mtu;
+
+		rc = qed_sp_vport_start(p_hwfn, &start);
 		if (rc) {
 			DP_ERR(cdev, "Failed to start VPORT\n");
 			return rc;
@@ -1619,7 +1650,7 @@ static int qed_start_vport(struct qed_dev *cdev,
 
 		DP_VERBOSE(cdev, (QED_MSG_SPQ | NETIF_MSG_IFUP),
 			   "Started V-PORT %d with MTU %d\n",
-			   vport_id, mtu);
+			   start.vport_id, start.mtu);
 	}
 
 	qed_reset_vport_stats(cdev);
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index ddd9e4a..f75f334 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -2466,11 +2466,12 @@ static int qede_stop_queues(struct qede_dev *edev)
 static int qede_start_queues(struct qede_dev *edev)
 {
 	int rc, tc, i;
-	int vport_id = 0, drop_ttl0_flg = 1, vlan_removal_en = 1;
+	int vlan_removal_en = 1;
 	struct qed_dev *cdev = edev->cdev;
 	struct qed_update_vport_rss_params *rss_params = &edev->rss_params;
 	struct qed_update_vport_params vport_update_params;
 	struct qed_queue_start_common_params q_params;
+	struct qed_start_vport_params start = {0};
 
 	if (!edev->num_rss) {
 		DP_ERR(edev,
@@ -2478,10 +2479,12 @@ static int qede_start_queues(struct qede_dev *edev)
 		return -EINVAL;
 	}
 
-	rc = edev->ops->vport_start(cdev, vport_id,
-				    edev->ndev->mtu,
-				    drop_ttl0_flg,
-				    vlan_removal_en);
+	start.mtu = edev->ndev->mtu;
+	start.vport_id = 0;
+	start.drop_ttl0 = true;
+	start.remove_inner_vlan = vlan_removal_en;
+
+	rc = edev->ops->vport_start(cdev, &start);
 
 	if (rc) {
 		DP_ERR(edev, "Start V-PORT failed %d\n", rc);
@@ -2490,7 +2493,7 @@ static int qede_start_queues(struct qede_dev *edev)
 
 	DP_VERBOSE(edev, NETIF_MSG_IFUP,
 		   "Start vport ramrod passed, vport_id = %d, MTU = %d, vlan_removal_en = %d\n",
-		   vport_id, edev->ndev->mtu + 0xe, vlan_removal_en);
+		   start.vport_id, edev->ndev->mtu + 0xe, vlan_removal_en);
 
 	for_each_rss(i) {
 		struct qede_fastpath *fp = &edev->fp_array[i];
@@ -2555,7 +2558,7 @@ static int qede_start_queues(struct qede_dev *edev)
 
 	/* Prepare and send the vport enable */
 	memset(&vport_update_params, 0, sizeof(vport_update_params));
-	vport_update_params.vport_id = vport_id;
+	vport_update_params.vport_id = start.vport_id;
 	vport_update_params.update_vport_active_flg = 1;
 	vport_update_params.vport_active_flg = 1;
 
diff --git a/include/linux/qed/qed_eth_if.h b/include/linux/qed/qed_eth_if.h
index e53b0ca..e1d6983 100644
--- a/include/linux/qed/qed_eth_if.h
+++ b/include/linux/qed/qed_eth_if.h
@@ -39,6 +39,14 @@ struct qed_update_vport_params {
 	struct qed_update_vport_rss_params rss_params;
 };
 
+struct qed_start_vport_params {
+	bool remove_inner_vlan;
+	bool gro_enable;
+	bool drop_ttl0;
+	u8 vport_id;
+	u16 mtu;
+};
+
 struct qed_stop_rxq_params {
 	u8 rss_id;
 	u8 rx_queue_id;
@@ -118,9 +126,7 @@ struct qed_eth_ops {
 			     void *cookie);
 
 	int (*vport_start)(struct qed_dev *cdev,
-			   u8 vport_id, u16 mtu,
-			   u8 drop_ttl0_flg,
-			   u8 inner_vlan_removal_en_flg);
+			   struct qed_start_vport_params *params);
 
 	int (*vport_stop)(struct qed_dev *cdev,
 			  u8 vport_id);
-- 
cgit v0.10.2


From 55482edc25f0606851de42e73618f813f310d009 Mon Sep 17 00:00:00 2001
From: Manish Chopra <manish.chopra@qlogic.com>
Date: Fri, 4 Mar 2016 12:35:06 -0500
Subject: qede: Add slowpath/fastpath support and enable hardware GRO

This patch configures hardware to use GRO and adds support
for fastpath APIs to handle HW aggregated packets.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index 02e17d3..d023251 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -160,6 +160,7 @@ struct qede_dev {
 	u16			q_num_rx_buffers; /* Must be a power of two */
 	u16			q_num_tx_buffers; /* Must be a power of two */
 
+	bool gro_disable;
 	struct list_head vlan_list;
 	u16 configured_vlans;
 	u16 non_configured_vlans;
@@ -188,6 +189,24 @@ struct sw_rx_data {
 	unsigned int page_offset;
 };
 
+enum qede_agg_state {
+	QEDE_AGG_STATE_NONE  = 0,
+	QEDE_AGG_STATE_START = 1,
+	QEDE_AGG_STATE_ERROR = 2
+};
+
+struct qede_agg_info {
+	struct sw_rx_data replace_buf;
+	dma_addr_t replace_buf_mapping;
+	struct sw_rx_data start_buf;
+	dma_addr_t start_buf_mapping;
+	struct eth_fast_path_rx_tpa_start_cqe start_cqe;
+	enum qede_agg_state agg_state;
+	struct sk_buff *skb;
+	int frag_id;
+	u16 vlan_tag;
+};
+
 struct qede_rx_queue {
 	__le16			*hw_cons_ptr;
 	struct sw_rx_data	*sw_rx_ring;
@@ -197,6 +216,9 @@ struct qede_rx_queue {
 	struct qed_chain	rx_comp_ring;
 	void __iomem		*hw_rxq_prod_addr;
 
+	/* GRO */
+	struct qede_agg_info	tpa_info[ETH_TPA_MAX_AGGS_NUM];
+
 	int			rx_buf_size;
 	unsigned int		rx_buf_seg_size;
 
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index f75f334..5728625 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -866,6 +866,278 @@ static inline void qede_skb_receive(struct qede_dev *edev,
 	napi_gro_receive(&fp->napi, skb);
 }
 
+static void qede_set_gro_params(struct qede_dev *edev,
+				struct sk_buff *skb,
+				struct eth_fast_path_rx_tpa_start_cqe *cqe)
+{
+	u16 parsing_flags = le16_to_cpu(cqe->pars_flags.flags);
+
+	if (((parsing_flags >> PARSING_AND_ERR_FLAGS_L3TYPE_SHIFT) &
+	    PARSING_AND_ERR_FLAGS_L3TYPE_MASK) == 2)
+		skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
+	else
+		skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+
+	skb_shinfo(skb)->gso_size = __le16_to_cpu(cqe->len_on_first_bd) -
+					cqe->header_len;
+}
+
+static int qede_fill_frag_skb(struct qede_dev *edev,
+			      struct qede_rx_queue *rxq,
+			      u8 tpa_agg_index,
+			      u16 len_on_bd)
+{
+	struct sw_rx_data *current_bd = &rxq->sw_rx_ring[rxq->sw_rx_cons &
+							 NUM_RX_BDS_MAX];
+	struct qede_agg_info *tpa_info = &rxq->tpa_info[tpa_agg_index];
+	struct sk_buff *skb = tpa_info->skb;
+
+	if (unlikely(tpa_info->agg_state != QEDE_AGG_STATE_START))
+		goto out;
+
+	/* Add one frag and update the appropriate fields in the skb */
+	skb_fill_page_desc(skb, tpa_info->frag_id++,
+			   current_bd->data, current_bd->page_offset,
+			   len_on_bd);
+
+	if (unlikely(qede_realloc_rx_buffer(edev, rxq, current_bd))) {
+		tpa_info->agg_state = QEDE_AGG_STATE_ERROR;
+		goto out;
+	}
+
+	qed_chain_consume(&rxq->rx_bd_ring);
+	rxq->sw_rx_cons++;
+
+	skb->data_len += len_on_bd;
+	skb->truesize += rxq->rx_buf_seg_size;
+	skb->len += len_on_bd;
+
+	return 0;
+
+out:
+	return -ENOMEM;
+}
+
+static void qede_tpa_start(struct qede_dev *edev,
+			   struct qede_rx_queue *rxq,
+			   struct eth_fast_path_rx_tpa_start_cqe *cqe)
+{
+	struct qede_agg_info *tpa_info = &rxq->tpa_info[cqe->tpa_agg_index];
+	struct eth_rx_bd *rx_bd_cons = qed_chain_consume(&rxq->rx_bd_ring);
+	struct eth_rx_bd *rx_bd_prod = qed_chain_produce(&rxq->rx_bd_ring);
+	struct sw_rx_data *replace_buf = &tpa_info->replace_buf;
+	dma_addr_t mapping = tpa_info->replace_buf_mapping;
+	struct sw_rx_data *sw_rx_data_cons;
+	struct sw_rx_data *sw_rx_data_prod;
+	enum pkt_hash_types rxhash_type;
+	u32 rxhash;
+
+	sw_rx_data_cons = &rxq->sw_rx_ring[rxq->sw_rx_cons & NUM_RX_BDS_MAX];
+	sw_rx_data_prod = &rxq->sw_rx_ring[rxq->sw_rx_prod & NUM_RX_BDS_MAX];
+
+	/* Use pre-allocated replacement buffer - we can't release the agg.
+	 * start until its over and we don't want to risk allocation failing
+	 * here, so re-allocate when aggregation will be over.
+	 */
+	dma_unmap_addr_set(sw_rx_data_prod, mapping,
+			   dma_unmap_addr(replace_buf, mapping));
+
+	sw_rx_data_prod->data = replace_buf->data;
+	rx_bd_prod->addr.hi = cpu_to_le32(upper_32_bits(mapping));
+	rx_bd_prod->addr.lo = cpu_to_le32(lower_32_bits(mapping));
+	sw_rx_data_prod->page_offset = replace_buf->page_offset;
+
+	rxq->sw_rx_prod++;
+
+	/* move partial skb from cons to pool (don't unmap yet)
+	 * save mapping, incase we drop the packet later on.
+	 */
+	tpa_info->start_buf = *sw_rx_data_cons;
+	mapping = HILO_U64(le32_to_cpu(rx_bd_cons->addr.hi),
+			   le32_to_cpu(rx_bd_cons->addr.lo));
+
+	tpa_info->start_buf_mapping = mapping;
+	rxq->sw_rx_cons++;
+
+	/* set tpa state to start only if we are able to allocate skb
+	 * for this aggregation, otherwise mark as error and aggregation will
+	 * be dropped
+	 */
+	tpa_info->skb = netdev_alloc_skb(edev->ndev,
+					 le16_to_cpu(cqe->len_on_first_bd));
+	if (unlikely(!tpa_info->skb)) {
+		tpa_info->agg_state = QEDE_AGG_STATE_ERROR;
+		return;
+	}
+
+	skb_put(tpa_info->skb, le16_to_cpu(cqe->len_on_first_bd));
+	memcpy(&tpa_info->start_cqe, cqe, sizeof(tpa_info->start_cqe));
+
+	/* Start filling in the aggregation info */
+	tpa_info->frag_id = 0;
+	tpa_info->agg_state = QEDE_AGG_STATE_START;
+
+	rxhash = qede_get_rxhash(edev, cqe->bitfields,
+				 cqe->rss_hash, &rxhash_type);
+	skb_set_hash(tpa_info->skb, rxhash, rxhash_type);
+	if ((le16_to_cpu(cqe->pars_flags.flags) >>
+	     PARSING_AND_ERR_FLAGS_TAG8021QEXIST_SHIFT) &
+		    PARSING_AND_ERR_FLAGS_TAG8021QEXIST_MASK)
+		tpa_info->vlan_tag = le16_to_cpu(cqe->vlan_tag);
+	else
+		tpa_info->vlan_tag = 0;
+
+	/* This is needed in order to enable forwarding support */
+	qede_set_gro_params(edev, tpa_info->skb, cqe);
+
+	if (likely(cqe->ext_bd_len_list[0]))
+		qede_fill_frag_skb(edev, rxq, cqe->tpa_agg_index,
+				   le16_to_cpu(cqe->ext_bd_len_list[0]));
+
+	if (unlikely(cqe->ext_bd_len_list[1])) {
+		DP_ERR(edev,
+		       "Unlikely - got a TPA aggregation with more than one ext_bd_len_list entry in the TPA start\n");
+		tpa_info->agg_state = QEDE_AGG_STATE_ERROR;
+	}
+}
+
+static void qede_gro_ip_csum(struct sk_buff *skb)
+{
+	const struct iphdr *iph = ip_hdr(skb);
+	struct tcphdr *th;
+
+	skb_set_network_header(skb, 0);
+	skb_set_transport_header(skb, sizeof(struct iphdr));
+	th = tcp_hdr(skb);
+
+	th->check = ~tcp_v4_check(skb->len - skb_transport_offset(skb),
+				  iph->saddr, iph->daddr, 0);
+
+	tcp_gro_complete(skb);
+}
+
+static void qede_gro_ipv6_csum(struct sk_buff *skb)
+{
+	struct ipv6hdr *iph = ipv6_hdr(skb);
+	struct tcphdr *th;
+
+	skb_set_network_header(skb, 0);
+	skb_set_transport_header(skb, sizeof(struct ipv6hdr));
+	th = tcp_hdr(skb);
+
+	th->check = ~tcp_v6_check(skb->len - skb_transport_offset(skb),
+				  &iph->saddr, &iph->daddr, 0);
+	tcp_gro_complete(skb);
+}
+
+static void qede_gro_receive(struct qede_dev *edev,
+			     struct qede_fastpath *fp,
+			     struct sk_buff *skb,
+			     u16 vlan_tag)
+{
+	if (skb_shinfo(skb)->gso_size) {
+		switch (skb->protocol) {
+		case htons(ETH_P_IP):
+			qede_gro_ip_csum(skb);
+			break;
+		case htons(ETH_P_IPV6):
+			qede_gro_ipv6_csum(skb);
+			break;
+		default:
+			DP_ERR(edev,
+			       "Error: FW GRO supports only IPv4/IPv6, not 0x%04x\n",
+			       ntohs(skb->protocol));
+		}
+	}
+
+	skb_record_rx_queue(skb, fp->rss_id);
+	qede_skb_receive(edev, fp, skb, vlan_tag);
+}
+
+static inline void qede_tpa_cont(struct qede_dev *edev,
+				 struct qede_rx_queue *rxq,
+				 struct eth_fast_path_rx_tpa_cont_cqe *cqe)
+{
+	int i;
+
+	for (i = 0; cqe->len_list[i]; i++)
+		qede_fill_frag_skb(edev, rxq, cqe->tpa_agg_index,
+				   le16_to_cpu(cqe->len_list[i]));
+
+	if (unlikely(i > 1))
+		DP_ERR(edev,
+		       "Strange - TPA cont with more than a single len_list entry\n");
+}
+
+static void qede_tpa_end(struct qede_dev *edev,
+			 struct qede_fastpath *fp,
+			 struct eth_fast_path_rx_tpa_end_cqe *cqe)
+{
+	struct qede_rx_queue *rxq = fp->rxq;
+	struct qede_agg_info *tpa_info;
+	struct sk_buff *skb;
+	int i;
+
+	tpa_info = &rxq->tpa_info[cqe->tpa_agg_index];
+	skb = tpa_info->skb;
+
+	for (i = 0; cqe->len_list[i]; i++)
+		qede_fill_frag_skb(edev, rxq, cqe->tpa_agg_index,
+				   le16_to_cpu(cqe->len_list[i]));
+	if (unlikely(i > 1))
+		DP_ERR(edev,
+		       "Strange - TPA emd with more than a single len_list entry\n");
+
+	if (unlikely(tpa_info->agg_state != QEDE_AGG_STATE_START))
+		goto err;
+
+	/* Sanity */
+	if (unlikely(cqe->num_of_bds != tpa_info->frag_id + 1))
+		DP_ERR(edev,
+		       "Strange - TPA had %02x BDs, but SKB has only %d frags\n",
+		       cqe->num_of_bds, tpa_info->frag_id);
+	if (unlikely(skb->len != le16_to_cpu(cqe->total_packet_len)))
+		DP_ERR(edev,
+		       "Strange - total packet len [cqe] is %4x but SKB has len %04x\n",
+		       le16_to_cpu(cqe->total_packet_len), skb->len);
+
+	memcpy(skb->data,
+	       page_address(tpa_info->start_buf.data) +
+		tpa_info->start_cqe.placement_offset +
+		tpa_info->start_buf.page_offset,
+	       le16_to_cpu(tpa_info->start_cqe.len_on_first_bd));
+
+	/* Recycle [mapped] start buffer for the next replacement */
+	tpa_info->replace_buf = tpa_info->start_buf;
+	tpa_info->replace_buf_mapping = tpa_info->start_buf_mapping;
+
+	/* Finalize the SKB */
+	skb->protocol = eth_type_trans(skb, edev->ndev);
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	/* tcp_gro_complete() will copy NAPI_GRO_CB(skb)->count
+	 * to skb_shinfo(skb)->gso_segs
+	 */
+	NAPI_GRO_CB(skb)->count = le16_to_cpu(cqe->num_of_coalesced_segs);
+
+	qede_gro_receive(edev, fp, skb, tpa_info->vlan_tag);
+
+	tpa_info->agg_state = QEDE_AGG_STATE_NONE;
+
+	return;
+err:
+	/* The BD starting the aggregation is still mapped; Re-use it for
+	 * future aggregations [as replacement buffer]
+	 */
+	memcpy(&tpa_info->replace_buf, &tpa_info->start_buf,
+	       sizeof(struct sw_rx_data));
+	tpa_info->replace_buf_mapping = tpa_info->start_buf_mapping;
+	tpa_info->start_buf.data = NULL;
+	tpa_info->agg_state = QEDE_AGG_STATE_NONE;
+	dev_kfree_skb_any(tpa_info->skb);
+	tpa_info->skb = NULL;
+}
+
 static u8 qede_check_csum(u16 flag)
 {
 	u16 csum_flag = 0;
@@ -931,6 +1203,25 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget)
 			goto next_cqe;
 		}
 
+		if (cqe_type != ETH_RX_CQE_TYPE_REGULAR) {
+			switch (cqe_type) {
+			case ETH_RX_CQE_TYPE_TPA_START:
+				qede_tpa_start(edev, rxq,
+					       &cqe->fast_path_tpa_start);
+				goto next_cqe;
+			case ETH_RX_CQE_TYPE_TPA_CONT:
+				qede_tpa_cont(edev, rxq,
+					      &cqe->fast_path_tpa_cont);
+				goto next_cqe;
+			case ETH_RX_CQE_TYPE_TPA_END:
+				qede_tpa_end(edev, fp,
+					     &cqe->fast_path_tpa_end);
+				goto next_rx_only;
+			default:
+				break;
+			}
+		}
+
 		/* Get the data from the SW ring */
 		sw_rx_index = rxq->sw_rx_cons & NUM_RX_BDS_MAX;
 		sw_rx_data = &rxq->sw_rx_ring[sw_rx_index];
@@ -1057,9 +1348,9 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget)
 		qede_skb_receive(edev, fp, skb, le16_to_cpu(fp_cqe->vlan_tag));
 
 		qed_chain_consume(&rxq->rx_bd_ring);
-
 next_rx:
 		rxq->sw_rx_cons++;
+next_rx_only:
 		rx_pkt++;
 
 next_cqe: /* don't consume bd rx buffer */
@@ -1952,9 +2243,31 @@ static void qede_free_rx_buffers(struct qede_dev *edev,
 	}
 }
 
+static void qede_free_sge_mem(struct qede_dev *edev,
+			      struct qede_rx_queue *rxq) {
+	int i;
+
+	if (edev->gro_disable)
+		return;
+
+	for (i = 0; i < ETH_TPA_MAX_AGGS_NUM; i++) {
+		struct qede_agg_info *tpa_info = &rxq->tpa_info[i];
+		struct sw_rx_data *replace_buf = &tpa_info->replace_buf;
+
+		if (replace_buf) {
+			dma_unmap_page(&edev->pdev->dev,
+				       dma_unmap_addr(replace_buf, mapping),
+				       PAGE_SIZE, DMA_FROM_DEVICE);
+			__free_page(replace_buf->data);
+		}
+	}
+}
+
 static void qede_free_mem_rxq(struct qede_dev *edev,
 			      struct qede_rx_queue *rxq)
 {
+	qede_free_sge_mem(edev, rxq);
+
 	/* Free rx buffers */
 	qede_free_rx_buffers(edev, rxq);
 
@@ -2010,6 +2323,53 @@ static int qede_alloc_rx_buffer(struct qede_dev *edev,
 	return 0;
 }
 
+static int qede_alloc_sge_mem(struct qede_dev *edev,
+			      struct qede_rx_queue *rxq)
+{
+	dma_addr_t mapping;
+	int i;
+
+	if (edev->gro_disable)
+		return 0;
+
+	if (edev->ndev->mtu > PAGE_SIZE) {
+		edev->gro_disable = 1;
+		return 0;
+	}
+
+	for (i = 0; i < ETH_TPA_MAX_AGGS_NUM; i++) {
+		struct qede_agg_info *tpa_info = &rxq->tpa_info[i];
+		struct sw_rx_data *replace_buf = &tpa_info->replace_buf;
+
+		replace_buf->data = alloc_pages(GFP_ATOMIC, 0);
+		if (unlikely(!replace_buf->data)) {
+			DP_NOTICE(edev,
+				  "Failed to allocate TPA skb pool [replacement buffer]\n");
+			goto err;
+		}
+
+		mapping = dma_map_page(&edev->pdev->dev, replace_buf->data, 0,
+				       rxq->rx_buf_size, DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) {
+			DP_NOTICE(edev,
+				  "Failed to map TPA replacement buffer\n");
+			goto err;
+		}
+
+		dma_unmap_addr_set(replace_buf, mapping, mapping);
+		tpa_info->replace_buf.page_offset = 0;
+
+		tpa_info->replace_buf_mapping = mapping;
+		tpa_info->agg_state = QEDE_AGG_STATE_NONE;
+	}
+
+	return 0;
+err:
+	qede_free_sge_mem(edev, rxq);
+	edev->gro_disable = 1;
+	return -ENOMEM;
+}
+
 /* This function allocates all memory needed per Rx queue */
 static int qede_alloc_mem_rxq(struct qede_dev *edev,
 			      struct qede_rx_queue *rxq)
@@ -2071,6 +2431,8 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev,
 			  num_allocated);
 	}
 
+	qede_alloc_sge_mem(edev, rxq);
+
 	return 0;
 
 err:
@@ -2233,6 +2595,8 @@ static void qede_init_fp(struct qede_dev *edev)
 		snprintf(fp->name, sizeof(fp->name), "%s-fp-%d",
 			 edev->ndev->name, rss_id);
 	}
+
+	edev->gro_disable = !(edev->ndev->features & NETIF_F_GRO);
 }
 
 static int qede_set_real_num_queues(struct qede_dev *edev)
@@ -2479,6 +2843,7 @@ static int qede_start_queues(struct qede_dev *edev)
 		return -EINVAL;
 	}
 
+	start.gro_enable = !edev->gro_disable;
 	start.mtu = edev->ndev->mtu;
 	start.vport_id = 0;
 	start.drop_ttl0 = true;
-- 
cgit v0.10.2


From d66ab51442211158b677c2f12310c314d9587f74 Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Fri, 4 Mar 2016 15:07:43 -0800
Subject: hv_netvsc: Move subchannel waiting to rndis_filter_device_remove()

During hot add, vmbus_device_register() is called from vmbus_onoffer(), on
the same workqueue as the subchannel offer message work-queue, so
subchannel offer won't be processed until the vmbus_device_register()/...
/netvsc_probe() is done.
Also, vmbus_device_register() is called with channel_mutex locked, which
prevents subchannel processing too. So the "waiting for sub-channel
processing" will not success in hot add case. But, in usual module loading,
the netvsc_probe() is called from different code path, and doesn't fail.

This patch resolves the deadlock during NIC hot-add, and speeds up NIC
loading time.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index a37bbda..47d07c5 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -1175,22 +1175,18 @@ int rndis_filter_device_add(struct hv_device *dev,
 	ret = rndis_filter_set_rss_param(rndis_device, net_device->num_chn);
 
 	/*
-	 * Wait for the host to send us the sub-channel offers.
+	 * Set the number of sub-channels to be received.
 	 */
 	spin_lock_irqsave(&net_device->sc_lock, flags);
 	sc_delta = num_rss_qs - (net_device->num_chn - 1);
 	net_device->num_sc_offered -= sc_delta;
 	spin_unlock_irqrestore(&net_device->sc_lock, flags);
 
-	while (net_device->num_sc_offered != 0) {
-		t = wait_for_completion_timeout(&net_device->channel_init_wait, 10*HZ);
-		if (t == 0)
-			WARN(1, "Netvsc: Waiting for sub-channel processing");
-	}
 out:
 	if (ret) {
 		net_device->max_chn = 1;
 		net_device->num_chn = 1;
+		net_device->num_sc_offered = 0;
 	}
 
 	return 0; /* return 0 because primary channel can be used alone */
@@ -1204,6 +1200,17 @@ void rndis_filter_device_remove(struct hv_device *dev)
 {
 	struct netvsc_device *net_dev = hv_get_drvdata(dev);
 	struct rndis_device *rndis_dev = net_dev->extension;
+	unsigned long t;
+
+	/* If not all subchannel offers are complete, wait for them until
+	 * completion to avoid race.
+	 */
+	while (net_dev->num_sc_offered > 0) {
+		t = wait_for_completion_timeout(&net_dev->channel_init_wait,
+						10 * HZ);
+		if (t == 0)
+			WARN(1, "Netvsc: Waiting for sub-channel processing");
+	}
 
 	/* Halt and release the rndis device */
 	rndis_filter_halt_device(rndis_dev);
-- 
cgit v0.10.2


From 73fb270592164b1917442f8bff4c791d095ee2ef Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 25 Feb 2016 11:03:01 -0600
Subject: rtlwifi: Fix size of wireless mode variable

Smatch reports the following warning:

  CHECK   drivers/net/wireless/realtek/rtlwifi/rc.c
drivers/net/wireless/realtek/rtlwifi/rc.c:144 _rtl_rc_rate_set_series() warn: impossible condition '(wireless_mode == 256) => (0-255 == 256)'

This warning arises because commit acc6907b87a9 ("rtlwifi: Fix warning
from ieee80211_get_tx_rates() when using 5G") now checks the wireless
mode for WIRELESS_MODE_AC_ONLY (BIT(8)) in _rtl_rc_rate_set_series().
As a result, all quantities used to store the wireless mode must be u16.

This patch also reorders struct rtl_sta_info to save a little space.

Fixes: d76d65fd2695 ("rtlwifi: fix broken VHT support")
Reported-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtlwifi/rc.c b/drivers/net/wireless/realtek/rtlwifi/rc.c
index 28f7010..1aca777 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rc.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rc.c
@@ -41,7 +41,7 @@ static u8 _rtl_rc_get_highest_rix(struct rtl_priv *rtlpriv,
 	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
 	struct rtl_phy *rtlphy = &(rtlpriv->phy);
 	struct rtl_sta_info *sta_entry = NULL;
-	u8 wireless_mode = 0;
+	u16 wireless_mode = 0;
 
 	/*
 	 *this rate is no use for true rate, firmware
@@ -99,7 +99,7 @@ static void _rtl_rc_rate_set_series(struct rtl_priv *rtlpriv,
 {
 	struct rtl_mac *mac = rtl_mac(rtlpriv);
 	struct rtl_sta_info *sta_entry = NULL;
-	u8 wireless_mode = 0;
+	u16 wireless_mode = 0;
 	u8 sgi_20 = 0, sgi_40 = 0, sgi_80 = 0;
 
 	if (sta) {
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c
index 5da9bd0..fe900ba 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c
@@ -3837,7 +3837,7 @@ void rtl8821ae_update_channel_access_setting(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
-	u8 wireless_mode = mac->mode;
+	u16 wireless_mode = mac->mode;
 	u8 sifs_timer, r2t_sifs;
 
 	rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_SLOT_TIME,
diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index b07b364..554d814 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -1318,14 +1318,13 @@ struct rtl_tid_data {
 
 struct rtl_sta_info {
 	struct list_head list;
-	u8 ratr_index;
-	u8 wireless_mode;
-	u8 mimo_ps;
-	u8 mac_addr[ETH_ALEN];
 	struct rtl_tid_data tids[MAX_TID_COUNT];
-
 	/* just used for ap adhoc or mesh*/
 	struct rssi_sta rssi_stat;
+	u16 wireless_mode;
+	u8 ratr_index;
+	u8 mimo_ps;
+	u8 mac_addr[ETH_ALEN];
 } __packed;
 
 struct rtl_priv;
-- 
cgit v0.10.2


From 836856e3bd61d0644e5178a2c1b51d90459e2788 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 2 Mar 2016 16:59:02 +0100
Subject: wireless: cw1200: use __maybe_unused to hide pm functions_

The cw1200 uses #ifdef to check for CONFIG_PM, but then
uses SIMPLE_DEV_PM_OPS, which leaves the references out when
CONFIG_PM_SLEEP is not defined, so we get a warning with
PM=y && PM_SLEEP=n:

drivers/net/wireless/st/cw1200/cw1200_spi.c:450:12: error: 'cw1200_spi_suspend' defined but not used [-Werror=unused-function]

This removes the incorrect #ifdef and instead uses a __maybe_unused
annotation to let the compiler know it can silently drop
the function definition.

For the DEV_PM_OPS definition, we can use an IS_ENABLED() check
to avoid defining the structure when CONFIG_PM is not set without
the #ifdef.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/st/cw1200/cw1200_spi.c b/drivers/net/wireless/st/cw1200/cw1200_spi.c
index a740083..63f95e9 100644
--- a/drivers/net/wireless/st/cw1200/cw1200_spi.c
+++ b/drivers/net/wireless/st/cw1200/cw1200_spi.c
@@ -446,8 +446,7 @@ static int cw1200_spi_disconnect(struct spi_device *func)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int cw1200_spi_suspend(struct device *dev)
+static int __maybe_unused cw1200_spi_suspend(struct device *dev)
 {
 	struct hwbus_priv *self = spi_get_drvdata(to_spi_device(dev));
 
@@ -460,16 +459,12 @@ static int cw1200_spi_suspend(struct device *dev)
 
 static SIMPLE_DEV_PM_OPS(cw1200_pm_ops, cw1200_spi_suspend, NULL);
 
-#endif
-
 static struct spi_driver spi_driver = {
 	.probe		= cw1200_spi_probe,
 	.remove		= cw1200_spi_disconnect,
 	.driver = {
 		.name		= "cw1200_wlan_spi",
-#ifdef CONFIG_PM
-		.pm		= &cw1200_pm_ops,
-#endif
+		.pm		= IS_ENABLED(CONFIG_PM) ? &cw1200_pm_ops : NULL,
 	},
 };
 
diff --git a/drivers/net/wireless/st/cw1200/pm.h b/drivers/net/wireless/st/cw1200/pm.h
index 3ed90ff..5345484 100644
--- a/drivers/net/wireless/st/cw1200/pm.h
+++ b/drivers/net/wireless/st/cw1200/pm.h
@@ -31,13 +31,18 @@ int cw1200_pm_init(struct cw1200_pm_state *pm,
 void cw1200_pm_deinit(struct cw1200_pm_state *pm);
 int cw1200_wow_suspend(struct ieee80211_hw *hw,
 		       struct cfg80211_wowlan *wowlan);
-int cw1200_wow_resume(struct ieee80211_hw *hw);
 int cw1200_can_suspend(struct cw1200_common *priv);
+int cw1200_wow_resume(struct ieee80211_hw *hw);
 void cw1200_pm_stay_awake(struct cw1200_pm_state *pm,
 			  unsigned long tmo);
 #else
 static inline void cw1200_pm_stay_awake(struct cw1200_pm_state *pm,
-					unsigned long tmo) {
+					unsigned long tmo)
+{
+}
+static inline int cw1200_can_suspend(struct cw1200_common *priv)
+{
+	return 0;
 }
 #endif
 #endif
-- 
cgit v0.10.2


From 8050c0f0274a15841756968857cfb07b3ab809ae Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:02 +0100
Subject: bpf: allow bpf_csum_diff to feed bpf_l3_csum_replace as well

Commit 7d672345ed29 ("bpf: add generic bpf_csum_diff helper") added a
generic checksum diff helper that can feed bpf_l4_csum_replace() with
a target __wsum diff that is to be applied to the L4 checksum. This
facility is very flexible, can be cascaded, allows for adding, removing,
or diffing data, or for calculating the pseudo header checksum from
scratch, but it can also be reused for working with the IPv4 header
checksum.

Thus, analogous to bpf_l4_csum_replace(), add a case for header field
value of 0 to change the checksum at a given offset through a new helper
csum_replace_by_diff(). Also, in addition to that, this provides an
easy to use interface for feeding precalculated diffs f.e. coming from
a map. It nicely complements bpf_l3_csum_replace() that currently allows
only for csum updates of 2 and 4 byte diffs.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/checksum.h b/include/net/checksum.h
index 10a16b5..abffc64 100644
--- a/include/net/checksum.h
+++ b/include/net/checksum.h
@@ -120,6 +120,11 @@ static inline __wsum csum_partial_ext(const void *buff, int len, __wsum sum)
 
 #define CSUM_MANGLED_0 ((__force __sum16)0xffff)
 
+static inline void csum_replace_by_diff(__sum16 *sum, __wsum diff)
+{
+	*sum = csum_fold(csum_add(diff, ~csum_unfold(*sum)));
+}
+
 static inline void csum_replace4(__sum16 *sum, __be32 from, __be32 to)
 {
 	__wsum tmp = csum_sub(~csum_unfold(*sum), (__force __wsum)from);
diff --git a/net/core/filter.c b/net/core/filter.c
index 69f4ffc..356a251 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1447,6 +1447,12 @@ static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 		return -EFAULT;
 
 	switch (flags & BPF_F_HDR_FIELD_MASK) {
+	case 0:
+		if (unlikely(from != 0))
+			return -EINVAL;
+
+		csum_replace_by_diff(ptr, to);
+		break;
 	case 2:
 		csum_replace2(ptr, from, to);
 		break;
-- 
cgit v0.10.2


From 8afd54c87ad7089734ef0527937a256586ba828a Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:03 +0100
Subject: bpf: add flags to bpf_skb_store_bytes for clearing hash

When overwriting parts of the packet with bpf_skb_store_bytes() that
were fed previously into skb->hash calculation, we should clear the
current hash with skb_clear_hash(), so that a next skb_get_hash() call
can determine the correct hash related to this skb.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index ee21932..2e3e903 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -305,6 +305,7 @@ enum bpf_func_id {
 
 /* BPF_FUNC_skb_store_bytes flags. */
 #define BPF_F_RECOMPUTE_CSUM		(1ULL << 0)
+#define BPF_F_INVALIDATE_HASH		(1ULL << 1)
 
 /* BPF_FUNC_l3_csum_replace and BPF_FUNC_l4_csum_replace flags.
  * First 4 bits are for passing the header field size.
diff --git a/net/core/filter.c b/net/core/filter.c
index 356a251..a1fe246 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1353,7 +1353,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 	unsigned int len = (unsigned int) r4;
 	void *ptr;
 
-	if (unlikely(flags & ~(BPF_F_RECOMPUTE_CSUM)))
+	if (unlikely(flags & ~(BPF_F_RECOMPUTE_CSUM | BPF_F_INVALIDATE_HASH)))
 		return -EINVAL;
 
 	/* bpf verifier guarantees that:
@@ -1384,6 +1384,8 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 
 	if (flags & BPF_F_RECOMPUTE_CSUM)
 		skb_postpush_rcsum(skb, ptr, len);
+	if (flags & BPF_F_INVALIDATE_HASH)
+		skb_clear_hash(skb);
 
 	return 0;
 }
-- 
cgit v0.10.2


From 577c50aade0f34926e4a47f61629739e6da91af6 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:04 +0100
Subject: bpf: make helper function protos static

They are only used here, so there's no reason they should not be static.
Only the vlan push/pop protos are used in the test_bpf suite.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/filter.c b/net/core/filter.c
index a1fe246..ce4e18d 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1390,7 +1390,7 @@ static u64 bpf_skb_store_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 flags)
 	return 0;
 }
 
-const struct bpf_func_proto bpf_skb_store_bytes_proto = {
+static const struct bpf_func_proto bpf_skb_store_bytes_proto = {
 	.func		= bpf_skb_store_bytes,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -1421,7 +1421,7 @@ static u64 bpf_skb_load_bytes(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
 	return 0;
 }
 
-const struct bpf_func_proto bpf_skb_load_bytes_proto = {
+static const struct bpf_func_proto bpf_skb_load_bytes_proto = {
 	.func		= bpf_skb_load_bytes,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -1472,7 +1472,7 @@ static u64 bpf_l3_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 	return 0;
 }
 
-const struct bpf_func_proto bpf_l3_csum_replace_proto = {
+static const struct bpf_func_proto bpf_l3_csum_replace_proto = {
 	.func		= bpf_l3_csum_replace,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -1531,7 +1531,7 @@ static u64 bpf_l4_csum_replace(u64 r1, u64 r2, u64 from, u64 to, u64 flags)
 	return 0;
 }
 
-const struct bpf_func_proto bpf_l4_csum_replace_proto = {
+static const struct bpf_func_proto bpf_l4_csum_replace_proto = {
 	.func		= bpf_l4_csum_replace,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -1570,7 +1570,7 @@ static u64 bpf_csum_diff(u64 r1, u64 from_size, u64 r3, u64 to_size, u64 seed)
 	return csum_partial(sp->diff, diff_size, seed);
 }
 
-const struct bpf_func_proto bpf_csum_diff_proto = {
+static const struct bpf_func_proto bpf_csum_diff_proto = {
 	.func		= bpf_csum_diff,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -1608,7 +1608,7 @@ static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5)
 	return dev_queue_xmit(skb2);
 }
 
-const struct bpf_func_proto bpf_clone_redirect_proto = {
+static const struct bpf_func_proto bpf_clone_redirect_proto = {
 	.func           = bpf_clone_redirect,
 	.gpl_only       = false,
 	.ret_type       = RET_INTEGER,
@@ -1660,7 +1660,7 @@ int skb_do_redirect(struct sk_buff *skb)
 	return dev_queue_xmit(skb);
 }
 
-const struct bpf_func_proto bpf_redirect_proto = {
+static const struct bpf_func_proto bpf_redirect_proto = {
 	.func           = bpf_redirect,
 	.gpl_only       = false,
 	.ret_type       = RET_INTEGER,
@@ -1799,7 +1799,7 @@ static u64 bpf_skb_get_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	return 0;
 }
 
-const struct bpf_func_proto bpf_skb_get_tunnel_key_proto = {
+static const struct bpf_func_proto bpf_skb_get_tunnel_key_proto = {
 	.func		= bpf_skb_get_tunnel_key,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -1861,7 +1861,7 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	return 0;
 }
 
-const struct bpf_func_proto bpf_skb_set_tunnel_key_proto = {
+static const struct bpf_func_proto bpf_skb_set_tunnel_key_proto = {
 	.func		= bpf_skb_set_tunnel_key,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
-- 
cgit v0.10.2


From 2208087061c4ad88de188911367effc550144836 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:05 +0100
Subject: bpf: allow to propagate df in bpf_skb_set_tunnel_key

Added by 9a628224a61b ("ip_tunnel: Add dont fragment flag."), allow to
feed df flag into tunneling facilities (currently supported on TX by
vxlan, geneve and gre) as a hint from eBPF's bpf_skb_set_tunnel_key()
helper.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 2e3e903..21ee6d5 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -330,6 +330,7 @@ enum bpf_func_id {
 
 /* BPF_FUNC_skb_set_tunnel_key flags. */
 #define BPF_F_ZERO_CSUM_TX		(1ULL << 1)
+#define BPF_F_DONT_FRAGMENT		(1ULL << 2)
 
 /* user accessible mirror of in-kernel sk_buff.
  * new fields can only be added to the end of this structure
diff --git a/net/core/filter.c b/net/core/filter.c
index ce4e18d..6c9d155 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1819,7 +1819,8 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	u8 compat[sizeof(struct bpf_tunnel_key)];
 	struct ip_tunnel_info *info;
 
-	if (unlikely(flags & ~(BPF_F_TUNINFO_IPV6 | BPF_F_ZERO_CSUM_TX)))
+	if (unlikely(flags & ~(BPF_F_TUNINFO_IPV6 | BPF_F_ZERO_CSUM_TX |
+			       BPF_F_DONT_FRAGMENT)))
 		return -EINVAL;
 	if (unlikely(size != sizeof(struct bpf_tunnel_key))) {
 		switch (size) {
@@ -1844,6 +1845,9 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	info->mode = IP_TUNNEL_INFO_TX;
 
 	info->key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM;
+	if (flags & BPF_F_DONT_FRAGMENT)
+		info->key.tun_flags |= TUNNEL_DONT_FRAGMENT;
+
 	info->key.tun_id = cpu_to_be64(from->tunnel_id);
 	info->key.tos = from->tunnel_tos;
 	info->key.ttl = from->tunnel_ttl;
-- 
cgit v0.10.2


From 14ca0751c96f8d3d0f52e8ed3b3236f8b34d3460 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:06 +0100
Subject: bpf: support for access to tunnel options

After eBPF being able to programmatically access/manage tunnel key meta
data via commit d3aa45ce6b94 ("bpf: add helpers to access tunnel metadata")
and more recently also for IPv6 through c6c33454072f ("bpf: support ipv6
for bpf_skb_{set,get}_tunnel_key"), this work adds two complementary
helpers to generically access their auxiliary tunnel options.

Geneve and vxlan support this facility. For geneve, TLVs can be pushed,
and for the vxlan case its GBP extension. I.e. setting tunnel key for geneve
case only makes sense, if we can also read/write TLVs into it. In the GBP
case, it provides the flexibility to easily map the group policy ID in
combination with other helpers or maps.

I chose to model this as two separate helpers, bpf_skb_{set,get}_tunnel_opt(),
for a couple of reasons. bpf_skb_{set,get}_tunnel_key() is already rather
complex by itself, and there may be cases for tunnel key backends where
tunnel options are not always needed. If we would have integrated this
into bpf_skb_{set,get}_tunnel_key() nevertheless, we are very limited with
remaining helper arguments, so keeping compatibility on structs in case of
passing in a flat buffer gets more cumbersome. Separating both also allows
for more flexibility and future extensibility, f.e. options could be fed
directly from a map, etc.

Moreover, change geneve's xmit path to test only for info->options_len
instead of TUNNEL_GENEVE_OPT flag. This makes it more consistent with vxlan's
xmit path and allows for avoiding to specify a protocol flag in the API on
xmit, so it can be protocol agnostic. Having info->options_len is enough
information that is needed. Tested with vxlan and geneve.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index bc5da35..36db4cf 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -940,7 +940,7 @@ static netdev_tx_t geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 		u8 vni[3];
 
 		tunnel_id_to_vni(key->tun_id, vni);
-		if (key->tun_flags & TUNNEL_GENEVE_OPT)
+		if (info->options_len)
 			opts = ip_tunnel_info_opts(info);
 
 		if (key->tun_flags & TUNNEL_CSUM)
@@ -1027,7 +1027,7 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 		u8 vni[3];
 
 		tunnel_id_to_vni(key->tun_id, vni);
-		if (key->tun_flags & TUNNEL_GENEVE_OPT)
+		if (info->options_len)
 			opts = ip_tunnel_info_opts(info);
 
 		if (key->tun_flags & TUNNEL_CSUM)
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 21ee6d5..9221f65 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -298,6 +298,17 @@ enum bpf_func_id {
 	 * Return: csum result
 	 */
 	BPF_FUNC_csum_diff,
+
+	/**
+	 * bpf_skb_[gs]et_tunnel_opt(skb, opt, size)
+	 * retrieve or populate tunnel options metadata
+	 * @skb: pointer to skb
+	 * @opt: pointer to raw tunnel option data
+	 * @size: size of @opt
+	 * Return: 0 on success for set, option size for get
+	 */
+	BPF_FUNC_skb_get_tunnel_opt,
+	BPF_FUNC_skb_set_tunnel_opt,
 	__BPF_FUNC_MAX_ID,
 };
 
diff --git a/net/core/filter.c b/net/core/filter.c
index 6c9d155..012a10c 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1809,6 +1809,32 @@ static const struct bpf_func_proto bpf_skb_get_tunnel_key_proto = {
 	.arg4_type	= ARG_ANYTHING,
 };
 
+static u64 bpf_skb_get_tunnel_opt(u64 r1, u64 r2, u64 size, u64 r4, u64 r5)
+{
+	struct sk_buff *skb = (struct sk_buff *) (long) r1;
+	u8 *to = (u8 *) (long) r2;
+	const struct ip_tunnel_info *info = skb_tunnel_info(skb);
+
+	if (unlikely(!info ||
+		     !(info->key.tun_flags & TUNNEL_OPTIONS_PRESENT)))
+		return -ENOENT;
+	if (unlikely(size < info->options_len))
+		return -ENOMEM;
+
+	ip_tunnel_info_opts_get(to, info);
+
+	return info->options_len;
+}
+
+static const struct bpf_func_proto bpf_skb_get_tunnel_opt_proto = {
+	.func		= bpf_skb_get_tunnel_opt,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_PTR_TO_STACK,
+	.arg3_type	= ARG_CONST_STACK_SIZE,
+};
+
 static struct metadata_dst __percpu *md_dst;
 
 static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
@@ -1875,17 +1901,58 @@ static const struct bpf_func_proto bpf_skb_set_tunnel_key_proto = {
 	.arg4_type	= ARG_ANYTHING,
 };
 
-static const struct bpf_func_proto *bpf_get_skb_set_tunnel_key_proto(void)
+#define BPF_TUNLEN_MAX	255
+
+static u64 bpf_skb_set_tunnel_opt(u64 r1, u64 r2, u64 size, u64 r4, u64 r5)
+{
+	struct sk_buff *skb = (struct sk_buff *) (long) r1;
+	u8 *from = (u8 *) (long) r2;
+	struct ip_tunnel_info *info = skb_tunnel_info(skb);
+	const struct metadata_dst *md = this_cpu_ptr(md_dst);
+
+	if (unlikely(info != &md->u.tun_info || (size & (sizeof(u32) - 1))))
+		return -EINVAL;
+	if (unlikely(size > BPF_TUNLEN_MAX))
+		return -ENOMEM;
+
+	ip_tunnel_info_opts_set(info, from, size);
+
+	return 0;
+}
+
+static const struct bpf_func_proto bpf_skb_set_tunnel_opt_proto = {
+	.func		= bpf_skb_set_tunnel_opt,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_PTR_TO_STACK,
+	.arg3_type	= ARG_CONST_STACK_SIZE,
+};
+
+static const struct bpf_func_proto *
+bpf_get_skb_set_tunnel_proto(enum bpf_func_id which)
 {
 	if (!md_dst) {
-		/* race is not possible, since it's called from
-		 * verifier that is holding verifier mutex
+		BUILD_BUG_ON(FIELD_SIZEOF(struct ip_tunnel_info,
+					  options_len) != 1);
+
+		/* Race is not possible, since it's called from verifier
+		 * that is holding verifier mutex.
 		 */
-		md_dst = metadata_dst_alloc_percpu(0, GFP_KERNEL);
+		md_dst = metadata_dst_alloc_percpu(BPF_TUNLEN_MAX,
+						   GFP_KERNEL);
 		if (!md_dst)
 			return NULL;
 	}
-	return &bpf_skb_set_tunnel_key_proto;
+
+	switch (which) {
+	case BPF_FUNC_skb_set_tunnel_key:
+		return &bpf_skb_set_tunnel_key_proto;
+	case BPF_FUNC_skb_set_tunnel_opt:
+		return &bpf_skb_set_tunnel_opt_proto;
+	default:
+		return NULL;
+	}
 }
 
 static const struct bpf_func_proto *
@@ -1939,7 +2006,11 @@ tc_cls_act_func_proto(enum bpf_func_id func_id)
 	case BPF_FUNC_skb_get_tunnel_key:
 		return &bpf_skb_get_tunnel_key_proto;
 	case BPF_FUNC_skb_set_tunnel_key:
-		return bpf_get_skb_set_tunnel_key_proto();
+		return bpf_get_skb_set_tunnel_proto(func_id);
+	case BPF_FUNC_skb_get_tunnel_opt:
+		return &bpf_skb_get_tunnel_opt_proto;
+	case BPF_FUNC_skb_set_tunnel_opt:
+		return bpf_get_skb_set_tunnel_proto(func_id);
 	case BPF_FUNC_redirect:
 		return &bpf_redirect_proto;
 	case BPF_FUNC_get_route_realm:
-- 
cgit v0.10.2


From db3c6139e6ead91b42e7c2ad044ed8beaee884e6 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:07 +0100
Subject: bpf, vxlan, geneve, gre: fix usage of dst_cache on xmit

The assumptions from commit 0c1d70af924b ("net: use dst_cache for vxlan
device"), 468dfffcd762 ("geneve: add dst caching support") and 3c1cb4d2604c
("net/ipv4: add dst cache support for gre lwtunnels") on dst_cache usage
when ip_tunnel_info is used is unfortunately not always valid as assumed.

While it seems correct for ip_tunnel_info front-ends such as OVS, eBPF
however can fill in ip_tunnel_info for consumers like vxlan, geneve or gre
with different remote dsts, tos, etc, therefore they cannot be assumed as
packet independent.

Right now vxlan, geneve, gre would cache the dst for eBPF and every packet
would reuse the same entry that was first created on the initial route
lookup. eBPF doesn't store/cache the ip_tunnel_info, so each skb may have
a different one.

Fix it by adding a flag that checks the ip_tunnel_info. Also the !tos test
in vxlan needs to be handeled differently in this context as it is currently
inferred from ip_tunnel_info as well if present. ip_tunnel_dst_cache_usable()
helper is added for the three tunnel cases, which checks if we can use dst
cache.

Fixes: 0c1d70af924b ("net: use dst_cache for vxlan device")
Fixes: 468dfffcd762 ("geneve: add dst caching support")
Fixes: 3c1cb4d2604c ("net/ipv4: add dst cache support for gre lwtunnels")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 36db4cf..6a0cbbe 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -775,10 +775,10 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb,
 				       struct flowi4 *fl4,
 				       struct ip_tunnel_info *info)
 {
+	bool use_cache = ip_tunnel_dst_cache_usable(skb, info);
 	struct geneve_dev *geneve = netdev_priv(dev);
 	struct dst_cache *dst_cache;
 	struct rtable *rt = NULL;
-	bool use_cache = true;
 	__u8 tos;
 
 	memset(fl4, 0, sizeof(*fl4));
@@ -804,7 +804,6 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb,
 		dst_cache = &geneve->dst_cache;
 	}
 
-	use_cache = use_cache && !skb->mark;
 	if (use_cache) {
 		rt = dst_cache_get_ip4(dst_cache, &fl4->saddr);
 		if (rt)
@@ -832,11 +831,11 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 					   struct flowi6 *fl6,
 					   struct ip_tunnel_info *info)
 {
+	bool use_cache = ip_tunnel_dst_cache_usable(skb, info);
 	struct geneve_dev *geneve = netdev_priv(dev);
 	struct geneve_sock *gs6 = geneve->sock6;
 	struct dst_entry *dst = NULL;
 	struct dst_cache *dst_cache;
-	bool use_cache = true;
 	__u8 prio;
 
 	memset(fl6, 0, sizeof(*fl6));
@@ -862,7 +861,6 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 		dst_cache = &geneve->dst_cache;
 	}
 
-	use_cache = use_cache && !skb->mark;
 	if (use_cache) {
 		dst = dst_cache_get_ip6(dst_cache, &fl6->saddr);
 		if (dst)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index fc998a3..7294a45 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1756,17 +1756,15 @@ static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 				      struct sk_buff *skb, int oif, u8 tos,
 				      __be32 daddr, __be32 *saddr,
 				      struct dst_cache *dst_cache,
-				      struct ip_tunnel_info *info)
+				      const struct ip_tunnel_info *info)
 {
+	bool use_cache = ip_tunnel_dst_cache_usable(skb, info);
 	struct rtable *rt = NULL;
-	bool use_cache = false;
 	struct flowi4 fl4;
 
-	/* when the ip_tunnel_info is availble, the tos used for lookup is
-	 * packet independent, so we can use the cache
-	 */
-	if (!skb->mark && (!tos || info)) {
-		use_cache = true;
+	if (tos && !info)
+		use_cache = false;
+	if (use_cache) {
 		rt = dst_cache_get_ip4(dst_cache, saddr);
 		if (rt)
 			return rt;
@@ -1794,13 +1792,15 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 					  struct sk_buff *skb, int oif,
 					  const struct in6_addr *daddr,
 					  struct in6_addr *saddr,
-					  struct dst_cache *dst_cache)
+					  struct dst_cache *dst_cache,
+					  const struct ip_tunnel_info *info)
 {
+	bool use_cache = ip_tunnel_dst_cache_usable(skb, info);
 	struct dst_entry *ndst;
 	struct flowi6 fl6;
 	int err;
 
-	if (!skb->mark) {
+	if (use_cache) {
 		ndst = dst_cache_get_ip6(dst_cache, saddr);
 		if (ndst)
 			return ndst;
@@ -1820,7 +1820,7 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 		return ERR_PTR(err);
 
 	*saddr = fl6.saddr;
-	if (!skb->mark)
+	if (use_cache)
 		dst_cache_set_ip6(dst_cache, ndst, saddr);
 	return ndst;
 }
@@ -2018,7 +2018,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		ndst = vxlan6_get_route(vxlan, skb,
 					rdst ? rdst->remote_ifindex : 0,
 					&dst->sin6.sin6_addr, &saddr,
-					dst_cache);
+					dst_cache, info);
 		if (IS_ERR(ndst)) {
 			netdev_dbg(dev, "no route to %pI6\n",
 				   &dst->sin6.sin6_addr);
@@ -2387,7 +2387,7 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 			return -EINVAL;
 		ndst = vxlan6_get_route(vxlan, skb, 0,
 					&info->key.u.ipv6.dst,
-					&info->key.u.ipv6.src, NULL);
+					&info->key.u.ipv6.src, NULL, info);
 		if (IS_ERR(ndst))
 			return PTR_ERR(ndst);
 		dst_release(ndst);
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 5f28b60..e1395d7 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -140,6 +140,7 @@ struct ip_tunnel {
 #define TUNNEL_CRIT_OPT		__cpu_to_be16(0x0400)
 #define TUNNEL_GENEVE_OPT	__cpu_to_be16(0x0800)
 #define TUNNEL_VXLAN_OPT	__cpu_to_be16(0x1000)
+#define TUNNEL_NOCACHE		__cpu_to_be16(0x2000)
 
 #define TUNNEL_OPTIONS_PRESENT	(TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT)
 
@@ -206,6 +207,20 @@ static inline void ip_tunnel_key_init(struct ip_tunnel_key *key,
 		       0, sizeof(*key) - IP_TUNNEL_KEY_SIZE);
 }
 
+static inline bool
+ip_tunnel_dst_cache_usable(const struct sk_buff *skb,
+			   const struct ip_tunnel_info *info)
+{
+	if (skb->mark)
+		return false;
+	if (!info)
+		return true;
+	if (info->key.tun_flags & TUNNEL_NOCACHE)
+		return false;
+
+	return true;
+}
+
 static inline unsigned short ip_tunnel_info_af(const struct ip_tunnel_info
 					       *tun_info)
 {
diff --git a/net/core/filter.c b/net/core/filter.c
index 012a10c..a66dc03 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1870,7 +1870,7 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	info = &md->u.tun_info;
 	info->mode = IP_TUNNEL_INFO_TX;
 
-	info->key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM;
+	info->key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM | TUNNEL_NOCACHE;
 	if (flags & BPF_F_DONT_FRAGMENT)
 		info->key.tun_flags |= TUNNEL_DONT_FRAGMENT;
 
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 202437d..31936d3 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -527,11 +527,12 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ip_tunnel_info *tun_info;
 	const struct ip_tunnel_key *key;
+	struct rtable *rt = NULL;
 	struct flowi4 fl;
-	struct rtable *rt;
 	int min_headroom;
 	int tunnel_hlen;
 	__be16 df, flags;
+	bool use_cache;
 	int err;
 
 	tun_info = skb_tunnel_info(skb);
@@ -540,13 +541,14 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto err_free_skb;
 
 	key = &tun_info->key;
-	rt = !skb->mark ? dst_cache_get_ip4(&tun_info->dst_cache, &fl.saddr) :
-			 NULL;
+	use_cache = ip_tunnel_dst_cache_usable(skb, tun_info);
+	if (use_cache)
+		rt = dst_cache_get_ip4(&tun_info->dst_cache, &fl.saddr);
 	if (!rt) {
 		rt = gre_get_rt(skb, dev, &fl, key);
 		if (IS_ERR(rt))
 				goto err_free_skb;
-		if (!skb->mark)
+		if (use_cache)
 			dst_cache_set_ip4(&tun_info->dst_cache, &rt->dst,
 					  fl.saddr);
 	}
-- 
cgit v0.10.2


From 1400615d64cf5afee533aff8234c837da465841b Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 4 Mar 2016 15:15:08 +0100
Subject: vxlan: allow setting ipv6 traffic class

We can already do that for IPv4, but IPv6 support was missing. Add
it for vxlan, so it can be used with collect metadata frontends.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 7294a45..2399099 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1789,7 +1789,7 @@ static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 
 #if IS_ENABLED(CONFIG_IPV6)
 static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
-					  struct sk_buff *skb, int oif,
+					  struct sk_buff *skb, int oif, u8 tos,
 					  const struct in6_addr *daddr,
 					  struct in6_addr *saddr,
 					  struct dst_cache *dst_cache,
@@ -1800,6 +1800,8 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 	struct flowi6 fl6;
 	int err;
 
+	if (tos && !info)
+		use_cache = false;
 	if (use_cache) {
 		ndst = dst_cache_get_ip6(dst_cache, saddr);
 		if (ndst)
@@ -1808,6 +1810,7 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 
 	memset(&fl6, 0, sizeof(fl6));
 	fl6.flowi6_oif = oif;
+	fl6.flowi6_tos = RT_TOS(tos);
 	fl6.daddr = *daddr;
 	fl6.saddr = vxlan->cfg.saddr.sin6.sin6_addr;
 	fl6.flowi6_mark = skb->mark;
@@ -2016,7 +2019,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		sk = vxlan->vn6_sock->sock->sk;
 
 		ndst = vxlan6_get_route(vxlan, skb,
-					rdst ? rdst->remote_ifindex : 0,
+					rdst ? rdst->remote_ifindex : 0, tos,
 					&dst->sin6.sin6_addr, &saddr,
 					dst_cache, info);
 		if (IS_ERR(ndst)) {
@@ -2053,6 +2056,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 		if (!info)
 			udp_sum = !(flags & VXLAN_F_UDP_ZERO_CSUM6_TX);
 
+		tos = ip_tunnel_ecn_encap(tos, old_iph, skb);
 		ttl = ttl ? : ip6_dst_hoplimit(ndst);
 		skb_scrub_packet(skb, xnet);
 		err = vxlan_build_skb(skb, ndst, sizeof(struct ipv6hdr),
@@ -2062,8 +2066,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			return;
 		}
 		udp_tunnel6_xmit_skb(ndst, sk, skb, dev,
-				     &saddr, &dst->sin6.sin6_addr,
-				     0, ttl, src_port, dst_port, !udp_sum);
+				     &saddr, &dst->sin6.sin6_addr, tos, ttl,
+				     src_port, dst_port, !udp_sum);
 #endif
 	}
 
@@ -2385,7 +2389,7 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 
 		if (!vxlan->vn6_sock)
 			return -EINVAL;
-		ndst = vxlan6_get_route(vxlan, skb, 0,
+		ndst = vxlan6_get_route(vxlan, skb, 0, info->key.tos,
 					&info->key.u.ipv6.dst,
 					&info->key.u.ipv6.src, NULL, info);
 		if (IS_ERR(ndst))
-- 
cgit v0.10.2


From f8b33d8e870758ccff13e5f81fd5050b52a42d35 Mon Sep 17 00:00:00 2001
From: Kyeong Yoo <Kyeong.Yoo@alliedtelesis.co.nz>
Date: Mon, 7 Mar 2016 17:07:57 +1300
Subject: net_sched: dsmark: use qdisc_dequeue_peeked()

This fix is for dsmark similar to commit 3557619f0f6f7496ed453d4825e249
("net_sched: prio: use qdisc_dequeue_peeked")
and makes use of qdisc_dequeue_peeked() instead of direct dequeue() call.

First time, wrr peeks dsmark, which will then peek into sfq.
sfq dequeues an skb and it's stored in sch->gso_skb.
Next time, wrr tries to dequeue from dsmark, which will call sfq dequeue
directly. This results skipping the previously peeked skb.

So changed dsmark dequeue to call qdisc_dequeue_peeked() instead to use
peeked skb if exists.

Signed-off-by: Kyeong Yoo <kyeong.yoo@alliedtelesis.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index d0dff0c..34b4dda 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -276,7 +276,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
 
 	pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p);
 
-	skb = p->q->ops->dequeue(p->q);
+	skb = qdisc_dequeue_peeked(p->q);
 	if (skb == NULL)
 		return NULL;
 
-- 
cgit v0.10.2


From d8aacd87180141ff6b812b53de77a4336e87c91a Mon Sep 17 00:00:00 2001
From: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Date: Tue, 8 Mar 2016 20:29:10 +0100
Subject: netfilter: ipset: Check IPSET_ATTR_ETHER netlink attribute length

Julia Lawall pointed out that IPSET_ATTR_ETHER netlink attribute length
was not checked explicitly, just for the maximum possible size. Malicious
netlink clients could send shorter attribute and thus resulting a kernel
read after the buffer.

The patch adds the explicit length checkings.

Reported-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>

diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
index 29dde20..9a065f6 100644
--- a/net/netfilter/ipset/ip_set_bitmap_ipmac.c
+++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c
@@ -267,6 +267,8 @@ bitmap_ipmac_uadt(struct ip_set *set, struct nlattr *tb[],
 
 	e.id = ip_to_id(map, ip);
 	if (tb[IPSET_ATTR_ETHER]) {
+		if (nla_len(tb[IPSET_ATTR_ETHER]) != ETH_ALEN)
+			return -IPSET_ERR_PROTOCOL;
 		memcpy(e.ether, nla_data(tb[IPSET_ATTR_ETHER]), ETH_ALEN);
 		e.add_mac = 1;
 	}
diff --git a/net/netfilter/ipset/ip_set_hash_mac.c b/net/netfilter/ipset/ip_set_hash_mac.c
index f1e7d2c..8f004ed 100644
--- a/net/netfilter/ipset/ip_set_hash_mac.c
+++ b/net/netfilter/ipset/ip_set_hash_mac.c
@@ -110,7 +110,8 @@ hash_mac4_uadt(struct ip_set *set, struct nlattr *tb[],
 	if (tb[IPSET_ATTR_LINENO])
 		*lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]);
 
-	if (unlikely(!tb[IPSET_ATTR_ETHER]))
+	if (unlikely(!tb[IPSET_ATTR_ETHER] ||
+		     nla_len(tb[IPSET_ATTR_ETHER]) != ETH_ALEN))
 		return -IPSET_ERR_PROTOCOL;
 
 	ret = ip_set_get_extensions(set, tb, &ext);
-- 
cgit v0.10.2


From 27c4d578600c401c119c012a90920805fab05cc9 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:41 -0500
Subject: bnxt_en: Refactor bnxt_fw_to_ethtool_advertised_spds().

Include the conversion of pause bits and add one extra call layer so
that the same refactored function can be reused to get the link partner
advertisement bits.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 84ea26d..a6ee26a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -515,9 +515,8 @@ static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info)
 	return speed_mask;
 }
 
-static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info)
+static u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause)
 {
-	u16 fw_speeds = link_info->auto_link_speeds;
 	u32 speed_mask = 0;
 
 	/* TODO: support 25GB, 40GB, 50GB with different cable type */
@@ -532,9 +531,28 @@ static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info)
 		speed_mask |= ADVERTISED_10000baseT_Full;
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_40GB)
 		speed_mask |= ADVERTISED_40000baseCR4_Full;
+
+	if ((fw_pause & BNXT_LINK_PAUSE_BOTH) == BNXT_LINK_PAUSE_BOTH)
+		speed_mask |= ADVERTISED_Pause;
+	else if (fw_pause & BNXT_LINK_PAUSE_TX)
+		speed_mask |= ADVERTISED_Asym_Pause;
+	else if (fw_pause & BNXT_LINK_PAUSE_RX)
+		speed_mask |= ADVERTISED_Pause | ADVERTISED_Asym_Pause;
+
 	return speed_mask;
 }
 
+static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info)
+{
+	u16 fw_speeds = link_info->auto_link_speeds;
+	u8 fw_pause = 0;
+
+	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
+		fw_pause = link_info->auto_pause_setting;
+
+	return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause);
+}
+
 u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed)
 {
 	switch (fw_link_speed) {
@@ -580,17 +598,6 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		cmd->autoneg = AUTONEG_DISABLE;
 		cmd->advertising = 0;
 	}
-	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) {
-		if ((link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH) ==
-		    BNXT_LINK_PAUSE_BOTH) {
-			cmd->advertising |= ADVERTISED_Pause;
-		} else {
-			cmd->advertising |= ADVERTISED_Asym_Pause;
-			if (link_info->auto_pause_setting &
-			    BNXT_LINK_PAUSE_RX)
-				cmd->advertising |= ADVERTISED_Pause;
-		}
-	}
 
 	cmd->port = PORT_NONE;
 	if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP) {
-- 
cgit v0.10.2


From 3277360eb29c6e482391975717d983060ecbd28d Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:42 -0500
Subject: bnxt_en: Add reporting of link partner advertisement.

And report actual pause settings to ETHTOOL_GPAUSEPARAM to let ethtool
resolve the actual pause settings.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 4dfc25f..b740489 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4468,6 +4468,7 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 	link_info->pause = resp->pause;
 	link_info->auto_mode = resp->auto_mode;
 	link_info->auto_pause_setting = resp->auto_pause;
+	link_info->lp_pause = resp->link_partner_adv_pause;
 	link_info->force_pause_setting = resp->force_pause;
 	link_info->duplex_setting = resp->duplex;
 	if (link_info->phy_link_status == BNXT_LINK_LINK)
@@ -4478,6 +4479,8 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
 	link_info->auto_link_speed = le16_to_cpu(resp->auto_link_speed);
 	link_info->support_speeds = le16_to_cpu(resp->support_speeds);
 	link_info->auto_link_speeds = le16_to_cpu(resp->auto_link_speed_mask);
+	link_info->lp_auto_link_speeds =
+		le16_to_cpu(resp->link_partner_adv_speeds);
 	link_info->preemphasis = le32_to_cpu(resp->preemphasis);
 	link_info->phy_ver[0] = resp->phy_maj;
 	link_info->phy_ver[1] = resp->phy_min;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 9aa38f5..2f24e4e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -780,6 +780,7 @@ struct bnxt_link_info {
 #define BNXT_LINK_PAUSE_RX	PORT_PHY_QCFG_RESP_PAUSE_RX
 #define BNXT_LINK_PAUSE_BOTH	(PORT_PHY_QCFG_RESP_PAUSE_RX | \
 				 PORT_PHY_QCFG_RESP_PAUSE_TX)
+	u8			lp_pause;
 	u8			auto_pause_setting;
 	u8			force_pause_setting;
 	u8			duplex_setting;
@@ -814,6 +815,7 @@ struct bnxt_link_info {
 #define BNXT_LINK_SPEED_MSK_25GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_25GB
 #define BNXT_LINK_SPEED_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB
 #define BNXT_LINK_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB
+	u16			lp_auto_link_speeds;
 	u16			auto_link_speed;
 	u16			force_link_speed;
 	u32			preemphasis;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index a6ee26a..591c290 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -553,6 +553,17 @@ static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info)
 	return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause);
 }
 
+static u32 bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info)
+{
+	u16 fw_speeds = link_info->lp_auto_link_speeds;
+	u8 fw_pause = 0;
+
+	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
+		fw_pause = link_info->lp_pause;
+
+	return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause);
+}
+
 u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed)
 {
 	switch (fw_link_speed) {
@@ -594,6 +605,9 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 			bnxt_fw_to_ethtool_advertised_spds(link_info);
 		cmd->advertising |= ADVERTISED_Autoneg;
 		cmd->autoneg = AUTONEG_ENABLE;
+		if (link_info->phy_link_status == BNXT_LINK_LINK)
+			cmd->lp_advertising =
+				bnxt_fw_to_ethtool_lp_adv(link_info);
 	} else {
 		cmd->autoneg = AUTONEG_DISABLE;
 		cmd->advertising = 0;
@@ -757,8 +771,10 @@ static void bnxt_get_pauseparam(struct net_device *dev,
 	if (BNXT_VF(bp))
 		return;
 	epause->autoneg = !!(link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL);
-	epause->rx_pause = ((link_info->pause & BNXT_LINK_PAUSE_RX) != 0);
-	epause->tx_pause = ((link_info->pause & BNXT_LINK_PAUSE_TX) != 0);
+	epause->rx_pause =
+		((link_info->auto_pause_setting & BNXT_LINK_PAUSE_RX) != 0);
+	epause->tx_pause =
+		((link_info->auto_pause_setting & BNXT_LINK_PAUSE_TX) != 0);
 }
 
 static int bnxt_set_pauseparam(struct net_device *dev,
-- 
cgit v0.10.2


From 4b32cacca28fe8b29bf266feff19b6fc2180402e Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:43 -0500
Subject: bnxt_en: Use common function to get ethtool supported flags.

The supported bits and advertising bits in ethtool have the same
definitions.  The same is true for the firmware bits.  So use the
common function to handle the conversion for both supported and
advertising bits.

v2: Don't use parentheses on function return.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 591c290..ac9d6e5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -496,25 +496,6 @@ static void bnxt_get_drvinfo(struct net_device *dev,
 	kfree(pkglog);
 }
 
-static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info)
-{
-	u16 fw_speeds = link_info->support_speeds;
-	u32 speed_mask = 0;
-
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_100MB)
-		speed_mask |= SUPPORTED_100baseT_Full;
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_1GB)
-		speed_mask |= SUPPORTED_1000baseT_Full;
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_2_5GB)
-		speed_mask |= SUPPORTED_2500baseX_Full;
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_10GB)
-		speed_mask |= SUPPORTED_10000baseT_Full;
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_40GB)
-		speed_mask |= SUPPORTED_40000baseCR4_Full;
-
-	return speed_mask;
-}
-
 static u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause)
 {
 	u32 speed_mask = 0;
@@ -564,6 +545,15 @@ static u32 bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info)
 	return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause);
 }
 
+static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info)
+{
+	u16 fw_speeds = link_info->support_speeds;
+	u32 supported;
+
+	supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+	return supported | SUPPORTED_Pause | SUPPORTED_Asym_Pause;
+}
+
 u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed)
 {
 	switch (fw_link_speed) {
@@ -595,7 +585,6 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	u16 ethtool_speed;
 
 	cmd->supported = bnxt_fw_to_ethtool_support_spds(link_info);
-	cmd->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
 
 	if (link_info->auto_link_speeds)
 		cmd->supported |= SUPPORTED_Autoneg;
-- 
cgit v0.10.2


From f1a082a6f79fd5f06b27ef05a5ba7ec8d6e83b4c Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:44 -0500
Subject: bnxt_en: Extend autoneg to all speeds.

Allow all autoneg speeds aupported by firmware to be advertised.  If
the advertising parameter is 0, then all supported speeds will be
advertised.

Remove BNXT_ALL_COPPER_ETHTOOL_SPEED which is no longer used as all
supported speeds can be advertised.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 2f24e4e..c4424b6 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -757,10 +757,6 @@ struct bnxt_ntuple_filter {
 #define BNXT_FLTR_UPDATE	1
 };
 
-#define BNXT_ALL_COPPER_ETHTOOL_SPEED				\
-	(ADVERTISED_100baseT_Full | ADVERTISED_1000baseT_Full |	\
-	 ADVERTISED_10000baseT_Full)
-
 struct bnxt_link_info {
 	u8			media_type;
 	u8			transceiver;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index ac9d6e5..c099802 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -696,16 +696,10 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		return rc;
 
 	if (cmd->autoneg == AUTONEG_ENABLE) {
-		if (link_info->media_type != PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP) {
-			netdev_err(dev, "Media type doesn't support autoneg\n");
-			rc = -EINVAL;
-			goto set_setting_exit;
-		}
-		if (cmd->advertising & ~(BNXT_ALL_COPPER_ETHTOOL_SPEED |
-					 ADVERTISED_Autoneg |
-					 ADVERTISED_TP |
-					 ADVERTISED_Pause |
-					 ADVERTISED_Asym_Pause)) {
+		u32 supported_spds = bnxt_fw_to_ethtool_support_spds(link_info);
+
+		if (cmd->advertising & ~(supported_spds | ADVERTISED_Autoneg |
+					 ADVERTISED_TP | ADVERTISED_FIBRE)) {
 			netdev_err(dev, "Unsupported advertising mask (adv: 0x%x)\n",
 				   cmd->advertising);
 			rc = -EINVAL;
-- 
cgit v0.10.2


From 3bdf56c47dfcd819ab1e73644c2eb9c72c08f29e Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:45 -0500
Subject: bnxt_en: Add port statistics support.

Gather periodic port statistics if the device is PF and link is up.  This
is triggered in bnxt_timer() every one second to request firmware to DMA
the counters.

Signed-off-by: Michael Chan <michael.chan@broadocm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index b740489..ae5f08e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -2362,6 +2362,14 @@ static void bnxt_free_stats(struct bnxt *bp)
 	u32 size, i;
 	struct pci_dev *pdev = bp->pdev;
 
+	if (bp->hw_rx_port_stats) {
+		dma_free_coherent(&pdev->dev, bp->hw_port_stats_size,
+				  bp->hw_rx_port_stats,
+				  bp->hw_rx_port_stats_map);
+		bp->hw_rx_port_stats = NULL;
+		bp->flags &= ~BNXT_FLAG_PORT_STATS;
+	}
+
 	if (!bp->bnapi)
 		return;
 
@@ -2398,6 +2406,24 @@ static int bnxt_alloc_stats(struct bnxt *bp)
 
 		cpr->hw_stats_ctx_id = INVALID_STATS_CTX_ID;
 	}
+
+	if (BNXT_PF(bp)) {
+		bp->hw_port_stats_size = sizeof(struct rx_port_stats) +
+					 sizeof(struct tx_port_stats) + 1024;
+
+		bp->hw_rx_port_stats =
+			dma_alloc_coherent(&pdev->dev, bp->hw_port_stats_size,
+					   &bp->hw_rx_port_stats_map,
+					   GFP_KERNEL);
+		if (!bp->hw_rx_port_stats)
+			return -ENOMEM;
+
+		bp->hw_tx_port_stats = (void *)(bp->hw_rx_port_stats + 1) +
+				       512;
+		bp->hw_tx_port_stats_map = bp->hw_rx_port_stats_map +
+					   sizeof(struct rx_port_stats) + 512;
+		bp->flags |= BNXT_FLAG_PORT_STATS;
+	}
 	return 0;
 }
 
@@ -3834,6 +3860,23 @@ hwrm_ver_get_exit:
 	return rc;
 }
 
+static int bnxt_hwrm_port_qstats(struct bnxt *bp)
+{
+	int rc;
+	struct bnxt_pf_info *pf = &bp->pf;
+	struct hwrm_port_qstats_input req = {0};
+
+	if (!(bp->flags & BNXT_FLAG_PORT_STATS))
+		return 0;
+
+	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_QSTATS, -1, -1);
+	req.port_id = cpu_to_le16(pf->port_id);
+	req.tx_stat_host_addr = cpu_to_le64(bp->hw_tx_port_stats_map);
+	req.rx_stat_host_addr = cpu_to_le64(bp->hw_rx_port_stats_map);
+	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	return rc;
+}
+
 static void bnxt_hwrm_free_tunnel_ports(struct bnxt *bp)
 {
 	if (bp->vxlan_port_cnt) {
@@ -5232,6 +5275,10 @@ static void bnxt_timer(unsigned long data)
 	if (atomic_read(&bp->intr_sem) != 0)
 		goto bnxt_restart_timer;
 
+	if (bp->link_info.link_up && (bp->flags & BNXT_FLAG_PORT_STATS)) {
+		set_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event);
+		schedule_work(&bp->sp_task);
+	}
 bnxt_restart_timer:
 	mod_timer(&bp->timer, jiffies + bp->current_interval);
 }
@@ -5283,6 +5330,9 @@ static void bnxt_sp_task(struct work_struct *work)
 		rtnl_unlock();
 	}
 
+	if (test_and_clear_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event))
+		bnxt_hwrm_port_qstats(bp);
+
 	smp_mb__before_atomic();
 	clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state);
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index c4424b6..ec04c47 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -873,6 +873,7 @@ struct bnxt {
 	#define BNXT_FLAG_MSIX_CAP	0x80
 	#define BNXT_FLAG_RFS		0x100
 	#define BNXT_FLAG_SHARED_RINGS	0x200
+	#define BNXT_FLAG_PORT_STATS	0x400
 
 	#define BNXT_FLAG_ALL_CONFIG_FEATS (BNXT_FLAG_TPA |		\
 					    BNXT_FLAG_RFS |		\
@@ -925,7 +926,7 @@ struct bnxt {
 	struct bnxt_queue_info	q_info[BNXT_MAX_QUEUE];
 
 	unsigned int		current_interval;
-#define BNXT_TIMER_INTERVAL	(HZ / 2)
+#define BNXT_TIMER_INTERVAL	HZ
 
 	struct timer_list	timer;
 
@@ -945,6 +946,13 @@ struct bnxt {
 	void			*hwrm_dbg_resp_addr;
 	dma_addr_t		hwrm_dbg_resp_dma_addr;
 #define HWRM_DBG_REG_BUF_SIZE	128
+
+	struct rx_port_stats	*hw_rx_port_stats;
+	struct tx_port_stats	*hw_tx_port_stats;
+	dma_addr_t		hw_rx_port_stats_map;
+	dma_addr_t		hw_tx_port_stats_map;
+	int			hw_port_stats_size;
+
 	int			hwrm_cmd_timeout;
 	struct mutex		hwrm_cmd_lock;	/* serialize hwrm messages */
 	struct hwrm_ver_get_output	ver_resp;
@@ -980,6 +988,7 @@ struct bnxt {
 #define BNXT_RESET_TASK_SP_EVENT	6
 #define BNXT_RST_RING_SP_EVENT		7
 #define BNXT_HWRM_PF_UNLOAD_SP_EVENT	8
+#define BNXT_PERIODIC_STATS_SP_EVENT	9
 
 	struct bnxt_pf_info	pf;
 #ifdef CONFIG_BNXT_SRIOV
-- 
cgit v0.10.2


From 9947f83fb79ca501f5ab24c370211bfb78b6b364 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:46 -0500
Subject: bnxt_en: Include some hardware port statistics in ndo_get_stats64().

Include some of the port error counters (e.g. crc) in ->ndo_get_stats64()
for the PF device.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index ae5f08e..1cd00a0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4935,6 +4935,22 @@ bnxt_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 		stats->tx_dropped += le64_to_cpu(hw_stats->tx_drop_pkts);
 	}
 
+	if (bp->flags & BNXT_FLAG_PORT_STATS) {
+		struct rx_port_stats *rx = bp->hw_rx_port_stats;
+		struct tx_port_stats *tx = bp->hw_tx_port_stats;
+
+		stats->rx_crc_errors = le64_to_cpu(rx->rx_fcs_err_frames);
+		stats->rx_frame_errors = le64_to_cpu(rx->rx_align_err_frames);
+		stats->rx_length_errors = le64_to_cpu(rx->rx_undrsz_frames) +
+					  le64_to_cpu(rx->rx_ovrsz_frames) +
+					  le64_to_cpu(rx->rx_runt_frames);
+		stats->rx_errors = le64_to_cpu(rx->rx_false_carrier_frames) +
+				   le64_to_cpu(rx->rx_jbr_frames);
+		stats->collisions = le64_to_cpu(tx->tx_total_collisions);
+		stats->tx_fifo_errors = le64_to_cpu(tx->tx_fifo_underruns);
+		stats->tx_errors = le64_to_cpu(tx->tx_err);
+	}
+
 	return stats;
 }
 
-- 
cgit v0.10.2


From 8ddc9aaa725a9337fc7bbe95fe1d1499769fb9b2 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:47 -0500
Subject: bnxt_en: Include hardware port statistics in ethtool -S.

Include the more useful port statistics in ethtool -S for the PF device.

Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index c099802..9ada166 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/ctype.h>
+#include <linux/stringify.h>
 #include <linux/ethtool.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
@@ -83,13 +84,99 @@ static int bnxt_set_coalesce(struct net_device *dev,
 
 #define BNXT_NUM_STATS	21
 
+#define BNXT_RX_STATS_OFFSET(counter)	\
+	(offsetof(struct rx_port_stats, counter) / 8)
+
+#define BNXT_RX_STATS_ENTRY(counter)	\
+	{ BNXT_RX_STATS_OFFSET(counter), __stringify(counter) }
+
+#define BNXT_TX_STATS_OFFSET(counter)			\
+	((offsetof(struct tx_port_stats, counter) +	\
+	  sizeof(struct rx_port_stats) + 512) / 8)
+
+#define BNXT_TX_STATS_ENTRY(counter)	\
+	{ BNXT_TX_STATS_OFFSET(counter), __stringify(counter) }
+
+static const struct {
+	long offset;
+	char string[ETH_GSTRING_LEN];
+} bnxt_port_stats_arr[] = {
+	BNXT_RX_STATS_ENTRY(rx_64b_frames),
+	BNXT_RX_STATS_ENTRY(rx_65b_127b_frames),
+	BNXT_RX_STATS_ENTRY(rx_128b_255b_frames),
+	BNXT_RX_STATS_ENTRY(rx_256b_511b_frames),
+	BNXT_RX_STATS_ENTRY(rx_512b_1023b_frames),
+	BNXT_RX_STATS_ENTRY(rx_1024b_1518_frames),
+	BNXT_RX_STATS_ENTRY(rx_good_vlan_frames),
+	BNXT_RX_STATS_ENTRY(rx_1519b_2047b_frames),
+	BNXT_RX_STATS_ENTRY(rx_2048b_4095b_frames),
+	BNXT_RX_STATS_ENTRY(rx_4096b_9216b_frames),
+	BNXT_RX_STATS_ENTRY(rx_9217b_16383b_frames),
+	BNXT_RX_STATS_ENTRY(rx_total_frames),
+	BNXT_RX_STATS_ENTRY(rx_ucast_frames),
+	BNXT_RX_STATS_ENTRY(rx_mcast_frames),
+	BNXT_RX_STATS_ENTRY(rx_bcast_frames),
+	BNXT_RX_STATS_ENTRY(rx_fcs_err_frames),
+	BNXT_RX_STATS_ENTRY(rx_ctrl_frames),
+	BNXT_RX_STATS_ENTRY(rx_pause_frames),
+	BNXT_RX_STATS_ENTRY(rx_pfc_frames),
+	BNXT_RX_STATS_ENTRY(rx_align_err_frames),
+	BNXT_RX_STATS_ENTRY(rx_ovrsz_frames),
+	BNXT_RX_STATS_ENTRY(rx_jbr_frames),
+	BNXT_RX_STATS_ENTRY(rx_mtu_err_frames),
+	BNXT_RX_STATS_ENTRY(rx_tagged_frames),
+	BNXT_RX_STATS_ENTRY(rx_double_tagged_frames),
+	BNXT_RX_STATS_ENTRY(rx_good_frames),
+	BNXT_RX_STATS_ENTRY(rx_undrsz_frames),
+	BNXT_RX_STATS_ENTRY(rx_eee_lpi_events),
+	BNXT_RX_STATS_ENTRY(rx_eee_lpi_duration),
+	BNXT_RX_STATS_ENTRY(rx_bytes),
+	BNXT_RX_STATS_ENTRY(rx_runt_bytes),
+	BNXT_RX_STATS_ENTRY(rx_runt_frames),
+
+	BNXT_TX_STATS_ENTRY(tx_64b_frames),
+	BNXT_TX_STATS_ENTRY(tx_65b_127b_frames),
+	BNXT_TX_STATS_ENTRY(tx_128b_255b_frames),
+	BNXT_TX_STATS_ENTRY(tx_256b_511b_frames),
+	BNXT_TX_STATS_ENTRY(tx_512b_1023b_frames),
+	BNXT_TX_STATS_ENTRY(tx_1024b_1518_frames),
+	BNXT_TX_STATS_ENTRY(tx_good_vlan_frames),
+	BNXT_TX_STATS_ENTRY(tx_1519b_2047_frames),
+	BNXT_TX_STATS_ENTRY(tx_2048b_4095b_frames),
+	BNXT_TX_STATS_ENTRY(tx_4096b_9216b_frames),
+	BNXT_TX_STATS_ENTRY(tx_9217b_16383b_frames),
+	BNXT_TX_STATS_ENTRY(tx_good_frames),
+	BNXT_TX_STATS_ENTRY(tx_total_frames),
+	BNXT_TX_STATS_ENTRY(tx_ucast_frames),
+	BNXT_TX_STATS_ENTRY(tx_mcast_frames),
+	BNXT_TX_STATS_ENTRY(tx_bcast_frames),
+	BNXT_TX_STATS_ENTRY(tx_pause_frames),
+	BNXT_TX_STATS_ENTRY(tx_pfc_frames),
+	BNXT_TX_STATS_ENTRY(tx_jabber_frames),
+	BNXT_TX_STATS_ENTRY(tx_fcs_err_frames),
+	BNXT_TX_STATS_ENTRY(tx_err),
+	BNXT_TX_STATS_ENTRY(tx_fifo_underruns),
+	BNXT_TX_STATS_ENTRY(tx_eee_lpi_events),
+	BNXT_TX_STATS_ENTRY(tx_eee_lpi_duration),
+	BNXT_TX_STATS_ENTRY(tx_total_collisions),
+	BNXT_TX_STATS_ENTRY(tx_bytes),
+};
+
+#define BNXT_NUM_PORT_STATS ARRAY_SIZE(bnxt_port_stats_arr)
+
 static int bnxt_get_sset_count(struct net_device *dev, int sset)
 {
 	struct bnxt *bp = netdev_priv(dev);
 
 	switch (sset) {
-	case ETH_SS_STATS:
-		return BNXT_NUM_STATS * bp->cp_nr_rings;
+	case ETH_SS_STATS: {
+		int num_stats = BNXT_NUM_STATS * bp->cp_nr_rings;
+
+		if (bp->flags & BNXT_FLAG_PORT_STATS)
+			num_stats += BNXT_NUM_PORT_STATS;
+
+		return num_stats;
+	}
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -118,6 +205,14 @@ static void bnxt_get_ethtool_stats(struct net_device *dev,
 			buf[j] = le64_to_cpu(hw_stats[k]);
 		buf[j++] = cpr->rx_l4_csum_errors;
 	}
+	if (bp->flags & BNXT_FLAG_PORT_STATS) {
+		__le64 *port_stats = (__le64 *)bp->hw_rx_port_stats;
+
+		for (i = 0; i < BNXT_NUM_PORT_STATS; i++, j++) {
+			buf[j] = le64_to_cpu(*(port_stats +
+					       bnxt_port_stats_arr[i].offset));
+		}
+	}
 }
 
 static void bnxt_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
@@ -172,6 +267,12 @@ static void bnxt_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
 			sprintf(buf, "[%d]: rx_l4_csum_errors", i);
 			buf += ETH_GSTRING_LEN;
 		}
+		if (bp->flags & BNXT_FLAG_PORT_STATS) {
+			for (i = 0; i < BNXT_NUM_PORT_STATS; i++) {
+				strcpy(buf, bnxt_port_stats_arr[i].string);
+				buf += ETH_GSTRING_LEN;
+			}
+		}
 		break;
 	default:
 		netdev_err(bp->dev, "bnxt_get_strings invalid request %x\n",
-- 
cgit v0.10.2


From 6316ea6db93d875df333e7ab205bf1aa3b3616d7 Mon Sep 17 00:00:00 2001
From: Satish Baddipadige <sbaddipa@broadcom.com>
Date: Mon, 7 Mar 2016 15:38:48 -0500
Subject: bnxt_en: Enable AER support.

Add pci_error_handler callbacks to support for pcie advanced error
recovery.

Signed-off-by: Satish Baddipadige <sbaddipa@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 1cd00a0..aabbd51 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -5412,6 +5412,8 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev)
 		goto init_err_release;
 	}
 
+	pci_enable_pcie_error_reporting(pdev);
+
 	INIT_WORK(&bp->sp_task, bnxt_sp_task);
 
 	spin_lock_init(&bp->ntp_fltr_lock);
@@ -5791,6 +5793,7 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 	if (BNXT_PF(bp))
 		bnxt_sriov_disable(bp);
 
+	pci_disable_pcie_error_reporting(pdev);
 	unregister_netdev(dev);
 	cancel_work_sync(&bp->sp_task);
 	bp->sp_event = 0;
@@ -6030,11 +6033,117 @@ init_err_free:
 	return rc;
 }
 
+/**
+ * bnxt_io_error_detected - called when PCI error is detected
+ * @pdev: Pointer to PCI device
+ * @state: The current pci connection state
+ *
+ * This function is called after a PCI bus error affecting
+ * this device has been detected.
+ */
+static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
+					       pci_channel_state_t state)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+
+	netdev_info(netdev, "PCI I/O error detected\n");
+
+	rtnl_lock();
+	netif_device_detach(netdev);
+
+	if (state == pci_channel_io_perm_failure) {
+		rtnl_unlock();
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	if (netif_running(netdev))
+		bnxt_close(netdev);
+
+	pci_disable_device(pdev);
+	rtnl_unlock();
+
+	/* Request a slot slot reset. */
+	return PCI_ERS_RESULT_NEED_RESET;
+}
+
+/**
+ * bnxt_io_slot_reset - called after the pci bus has been reset.
+ * @pdev: Pointer to PCI device
+ *
+ * Restart the card from scratch, as if from a cold-boot.
+ * At this point, the card has exprienced a hard reset,
+ * followed by fixups by BIOS, and has its config space
+ * set up identically to what it was at cold boot.
+ */
+static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct bnxt *bp = netdev_priv(netdev);
+	int err = 0;
+	pci_ers_result_t result = PCI_ERS_RESULT_DISCONNECT;
+
+	netdev_info(bp->dev, "PCI Slot Reset\n");
+
+	rtnl_lock();
+
+	if (pci_enable_device(pdev)) {
+		dev_err(&pdev->dev,
+			"Cannot re-enable PCI device after reset.\n");
+	} else {
+		pci_set_master(pdev);
+
+		if (netif_running(netdev))
+			err = bnxt_open(netdev);
+
+		if (!err)
+			result = PCI_ERS_RESULT_RECOVERED;
+	}
+
+	if (result != PCI_ERS_RESULT_RECOVERED && netif_running(netdev))
+		dev_close(netdev);
+
+	rtnl_unlock();
+
+	err = pci_cleanup_aer_uncorrect_error_status(pdev);
+	if (err) {
+		dev_err(&pdev->dev,
+			"pci_cleanup_aer_uncorrect_error_status failed 0x%0x\n",
+			 err); /* non-fatal, continue */
+	}
+
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+/**
+ * bnxt_io_resume - called when traffic can start flowing again.
+ * @pdev: Pointer to PCI device
+ *
+ * This callback is called when the error recovery driver tells
+ * us that its OK to resume normal operation.
+ */
+static void bnxt_io_resume(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+
+	rtnl_lock();
+
+	netif_device_attach(netdev);
+
+	rtnl_unlock();
+}
+
+static const struct pci_error_handlers bnxt_err_handler = {
+	.error_detected	= bnxt_io_error_detected,
+	.slot_reset	= bnxt_io_slot_reset,
+	.resume		= bnxt_io_resume
+};
+
 static struct pci_driver bnxt_pci_driver = {
 	.name		= DRV_MODULE_NAME,
 	.id_table	= bnxt_pci_tbl,
 	.probe		= bnxt_init_one,
 	.remove		= bnxt_remove_one,
+	.err_handler	= &bnxt_err_handler,
 #if defined(CONFIG_BNXT_SRIOV)
 	.sriov_configure = bnxt_sriov_configure,
 #endif
-- 
cgit v0.10.2


From 3570df914f9cae15df7f6ba472a51eaf798fdb46 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michal=20Kube=C4=8Dek?= <mkubecek@suse.cz>
Date: Tue, 8 Mar 2016 14:44:25 +0100
Subject: ipv6: replace global gc_args with local variable

Global variable gc_args is only used in fib6_run_gc() and functions
called from it. As fib6_run_gc() makes sure there is at most one
instance of fib6_clean_all() running at any moment, we can replace
gc_args with a local variable which will be needed once multiple
instances (per netns) of garbage collector are allowed.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 0c7e276..d7c715a 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1725,14 +1725,15 @@ static void fib6_flush_trees(struct net *net)
  *	Garbage collection
  */
 
-static struct fib6_gc_args
+struct fib6_gc_args
 {
 	int			timeout;
 	int			more;
-} gc_args;
+};
 
 static int fib6_age(struct rt6_info *rt, void *arg)
 {
+	struct fib6_gc_args *gc_args = arg;
 	unsigned long now = jiffies;
 
 	/*
@@ -1748,10 +1749,10 @@ static int fib6_age(struct rt6_info *rt, void *arg)
 			RT6_TRACE("expiring %p\n", rt);
 			return -1;
 		}
-		gc_args.more++;
+		gc_args->more++;
 	} else if (rt->rt6i_flags & RTF_CACHE) {
 		if (atomic_read(&rt->dst.__refcnt) == 0 &&
-		    time_after_eq(now, rt->dst.lastuse + gc_args.timeout)) {
+		    time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) {
 			RT6_TRACE("aging clone %p\n", rt);
 			return -1;
 		} else if (rt->rt6i_flags & RTF_GATEWAY) {
@@ -1769,7 +1770,7 @@ static int fib6_age(struct rt6_info *rt, void *arg)
 				return -1;
 			}
 		}
-		gc_args.more++;
+		gc_args->more++;
 	}
 
 	return 0;
@@ -1779,6 +1780,7 @@ static DEFINE_SPINLOCK(fib6_gc_lock);
 
 void fib6_run_gc(unsigned long expires, struct net *net, bool force)
 {
+	struct fib6_gc_args gc_args;
 	unsigned long now;
 
 	if (force) {
@@ -1792,7 +1794,7 @@ void fib6_run_gc(unsigned long expires, struct net *net, bool force)
 
 	gc_args.more = icmp6_dst_gc();
 
-	fib6_clean_all(net, fib6_age, NULL);
+	fib6_clean_all(net, fib6_age, &gc_args);
 	now = jiffies;
 	net->ipv6.ip6_rt_last_gc = now;
 
-- 
cgit v0.10.2


From 9a03cd8f38efb83c13fbe62aff50eea4efff93da Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michal=20Kube=C4=8Dek?= <mkubecek@suse.cz>
Date: Tue, 8 Mar 2016 14:44:35 +0100
Subject: ipv6: per netns fib6 walkers

The IPv6 FIB data structures are separated per network namespace but
there is still only one global walkers list and one global walker list
lock. This means changes in one namespace unnecessarily interfere with
walkers in other namespaces.

Replace the global list with per-netns lists (and give each its own
lock).

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index c0368db..f0109b9 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -58,7 +58,9 @@ struct netns_ipv6 {
 	struct timer_list       ip6_fib_timer;
 	struct hlist_head       *fib_table_hash;
 	struct fib6_table       *fib6_main_tbl;
+	struct list_head	fib6_walkers;
 	struct dst_ops		ip6_dst_ops;
+	rwlock_t		fib6_walker_lock;
 	unsigned int		 ip6_rt_gc_expire;
 	unsigned long		 ip6_rt_last_gc;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index d7c715a..883f283 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -55,8 +55,6 @@ struct fib6_cleaner {
 	void *arg;
 };
 
-static DEFINE_RWLOCK(fib6_walker_lock);
-
 #ifdef CONFIG_IPV6_SUBTREES
 #define FWS_INIT FWS_S
 #else
@@ -66,7 +64,7 @@ static DEFINE_RWLOCK(fib6_walker_lock);
 static void fib6_prune_clones(struct net *net, struct fib6_node *fn);
 static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn);
 static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_node *fn);
-static int fib6_walk(struct fib6_walker *w);
+static int fib6_walk(struct net *net, struct fib6_walker *w);
 static int fib6_walk_continue(struct fib6_walker *w);
 
 /*
@@ -78,21 +76,21 @@ static int fib6_walk_continue(struct fib6_walker *w);
 
 static void fib6_gc_timer_cb(unsigned long arg);
 
-static LIST_HEAD(fib6_walkers);
-#define FOR_WALKERS(w) list_for_each_entry(w, &fib6_walkers, lh)
+#define FOR_WALKERS(net, w) \
+	list_for_each_entry(w, &(net)->ipv6.fib6_walkers, lh)
 
-static void fib6_walker_link(struct fib6_walker *w)
+static void fib6_walker_link(struct net *net, struct fib6_walker *w)
 {
-	write_lock_bh(&fib6_walker_lock);
-	list_add(&w->lh, &fib6_walkers);
-	write_unlock_bh(&fib6_walker_lock);
+	write_lock_bh(&net->ipv6.fib6_walker_lock);
+	list_add(&w->lh, &net->ipv6.fib6_walkers);
+	write_unlock_bh(&net->ipv6.fib6_walker_lock);
 }
 
-static void fib6_walker_unlink(struct fib6_walker *w)
+static void fib6_walker_unlink(struct net *net, struct fib6_walker *w)
 {
-	write_lock_bh(&fib6_walker_lock);
+	write_lock_bh(&net->ipv6.fib6_walker_lock);
 	list_del(&w->lh);
-	write_unlock_bh(&fib6_walker_lock);
+	write_unlock_bh(&net->ipv6.fib6_walker_lock);
 }
 
 static int fib6_new_sernum(struct net *net)
@@ -325,12 +323,13 @@ static int fib6_dump_node(struct fib6_walker *w)
 
 static void fib6_dump_end(struct netlink_callback *cb)
 {
+	struct net *net = sock_net(cb->skb->sk);
 	struct fib6_walker *w = (void *)cb->args[2];
 
 	if (w) {
 		if (cb->args[4]) {
 			cb->args[4] = 0;
-			fib6_walker_unlink(w);
+			fib6_walker_unlink(net, w);
 		}
 		cb->args[2] = 0;
 		kfree(w);
@@ -348,6 +347,7 @@ static int fib6_dump_done(struct netlink_callback *cb)
 static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb,
 			   struct netlink_callback *cb)
 {
+	struct net *net = sock_net(skb->sk);
 	struct fib6_walker *w;
 	int res;
 
@@ -359,7 +359,7 @@ static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb,
 		w->skip = 0;
 
 		read_lock_bh(&table->tb6_lock);
-		res = fib6_walk(w);
+		res = fib6_walk(net, w);
 		read_unlock_bh(&table->tb6_lock);
 		if (res > 0) {
 			cb->args[4] = 1;
@@ -379,7 +379,7 @@ static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb,
 		res = fib6_walk_continue(w);
 		read_unlock_bh(&table->tb6_lock);
 		if (res <= 0) {
-			fib6_walker_unlink(w);
+			fib6_walker_unlink(net, w);
 			cb->args[4] = 0;
 		}
 	}
@@ -1340,8 +1340,8 @@ static struct fib6_node *fib6_repair_tree(struct net *net,
 		}
 #endif
 
-		read_lock(&fib6_walker_lock);
-		FOR_WALKERS(w) {
+		read_lock(&net->ipv6.fib6_walker_lock);
+		FOR_WALKERS(net, w) {
 			if (!child) {
 				if (w->root == fn) {
 					w->root = w->node = NULL;
@@ -1368,7 +1368,7 @@ static struct fib6_node *fib6_repair_tree(struct net *net,
 				}
 			}
 		}
-		read_unlock(&fib6_walker_lock);
+		read_unlock(&net->ipv6.fib6_walker_lock);
 
 		node_free(fn);
 		if (pn->fn_flags & RTN_RTINFO || FIB6_SUBTREE(pn))
@@ -1411,8 +1411,8 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	}
 
 	/* Adjust walkers */
-	read_lock(&fib6_walker_lock);
-	FOR_WALKERS(w) {
+	read_lock(&net->ipv6.fib6_walker_lock);
+	FOR_WALKERS(net, w) {
 		if (w->state == FWS_C && w->leaf == rt) {
 			RT6_TRACE("walker %p adjusted by delroute\n", w);
 			w->leaf = rt->dst.rt6_next;
@@ -1420,7 +1420,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 				w->state = FWS_U;
 		}
 	}
-	read_unlock(&fib6_walker_lock);
+	read_unlock(&net->ipv6.fib6_walker_lock);
 
 	rt->dst.rt6_next = NULL;
 
@@ -1588,17 +1588,17 @@ skip:
 	}
 }
 
-static int fib6_walk(struct fib6_walker *w)
+static int fib6_walk(struct net *net, struct fib6_walker *w)
 {
 	int res;
 
 	w->state = FWS_INIT;
 	w->node = w->root;
 
-	fib6_walker_link(w);
+	fib6_walker_link(net, w);
 	res = fib6_walk_continue(w);
 	if (res <= 0)
-		fib6_walker_unlink(w);
+		fib6_walker_unlink(net, w);
 	return res;
 }
 
@@ -1668,7 +1668,7 @@ static void fib6_clean_tree(struct net *net, struct fib6_node *root,
 	c.arg = arg;
 	c.net = net;
 
-	fib6_walk(&c.w);
+	fib6_walk(net, &c.w);
 }
 
 static void __fib6_clean_all(struct net *net,
@@ -1816,6 +1816,8 @@ static int __net_init fib6_net_init(struct net *net)
 {
 	size_t size = sizeof(struct hlist_head) * FIB6_TABLE_HASHSZ;
 
+	rwlock_init(&net->ipv6.fib6_walker_lock);
+	INIT_LIST_HEAD(&net->ipv6.fib6_walkers);
 	setup_timer(&net->ipv6.ip6_fib_timer, fib6_gc_timer_cb, (unsigned long)net);
 
 	net->ipv6.rt6_stats = kzalloc(sizeof(*net->ipv6.rt6_stats), GFP_KERNEL);
@@ -1976,7 +1978,8 @@ static int ipv6_route_yield(struct fib6_walker *w)
 	return 0;
 }
 
-static void ipv6_route_seq_setup_walk(struct ipv6_route_iter *iter)
+static void ipv6_route_seq_setup_walk(struct ipv6_route_iter *iter,
+				      struct net *net)
 {
 	memset(&iter->w, 0, sizeof(iter->w));
 	iter->w.func = ipv6_route_yield;
@@ -1986,7 +1989,7 @@ static void ipv6_route_seq_setup_walk(struct ipv6_route_iter *iter)
 	iter->w.args = iter;
 	iter->sernum = iter->w.root->fn_sernum;
 	INIT_LIST_HEAD(&iter->w.lh);
-	fib6_walker_link(&iter->w);
+	fib6_walker_link(net, &iter->w);
 }
 
 static struct fib6_table *ipv6_route_seq_next_table(struct fib6_table *tbl,
@@ -2047,16 +2050,16 @@ iter_table:
 			++*pos;
 		return iter->w.leaf;
 	} else if (r < 0) {
-		fib6_walker_unlink(&iter->w);
+		fib6_walker_unlink(net, &iter->w);
 		return NULL;
 	}
-	fib6_walker_unlink(&iter->w);
+	fib6_walker_unlink(net, &iter->w);
 
 	iter->tbl = ipv6_route_seq_next_table(iter->tbl, net);
 	if (!iter->tbl)
 		return NULL;
 
-	ipv6_route_seq_setup_walk(iter);
+	ipv6_route_seq_setup_walk(iter, net);
 	goto iter_table;
 }
 
@@ -2071,7 +2074,7 @@ static void *ipv6_route_seq_start(struct seq_file *seq, loff_t *pos)
 	iter->skip = *pos;
 
 	if (iter->tbl) {
-		ipv6_route_seq_setup_walk(iter);
+		ipv6_route_seq_setup_walk(iter, net);
 		return ipv6_route_seq_next(seq, NULL, pos);
 	} else {
 		return NULL;
@@ -2087,10 +2090,11 @@ static bool ipv6_route_iter_active(struct ipv6_route_iter *iter)
 static void ipv6_route_seq_stop(struct seq_file *seq, void *v)
 	__releases(RCU_BH)
 {
+	struct net *net = seq_file_net(seq);
 	struct ipv6_route_iter *iter = seq->private;
 
 	if (ipv6_route_iter_active(iter))
-		fib6_walker_unlink(&iter->w);
+		fib6_walker_unlink(net, &iter->w);
 
 	rcu_read_unlock_bh();
 }
-- 
cgit v0.10.2


From 3dc94f93be161ec4203673de9a34b7362d8985b5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michal=20Kube=C4=8Dek?= <mkubecek@suse.cz>
Date: Tue, 8 Mar 2016 14:44:45 +0100
Subject: ipv6: per netns FIB garbage collection

One of our customers observed issues with FIB6 garbage collectors
running in different network namespaces blocking each other, resulting
in soft lockups (fib6_run_gc() initiated from timer runs always in
forced mode).

Now that FIB6 walkers are separated per namespace, there is no more need
for instances of fib6_run_gc() in different namespaces blocking each
other. There is still a call to icmp6_dst_gc() which operates on shared
data but this function is protected by its own shared lock.

Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index f0109b9..10d0848 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -61,6 +61,7 @@ struct netns_ipv6 {
 	struct list_head	fib6_walkers;
 	struct dst_ops		ip6_dst_ops;
 	rwlock_t		fib6_walker_lock;
+	spinlock_t		fib6_gc_lock;
 	unsigned int		 ip6_rt_gc_expire;
 	unsigned long		 ip6_rt_last_gc;
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 883f283..ea071fa 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -1776,16 +1776,14 @@ static int fib6_age(struct rt6_info *rt, void *arg)
 	return 0;
 }
 
-static DEFINE_SPINLOCK(fib6_gc_lock);
-
 void fib6_run_gc(unsigned long expires, struct net *net, bool force)
 {
 	struct fib6_gc_args gc_args;
 	unsigned long now;
 
 	if (force) {
-		spin_lock_bh(&fib6_gc_lock);
-	} else if (!spin_trylock_bh(&fib6_gc_lock)) {
+		spin_lock_bh(&net->ipv6.fib6_gc_lock);
+	} else if (!spin_trylock_bh(&net->ipv6.fib6_gc_lock)) {
 		mod_timer(&net->ipv6.ip6_fib_timer, jiffies + HZ);
 		return;
 	}
@@ -1804,7 +1802,7 @@ void fib6_run_gc(unsigned long expires, struct net *net, bool force)
 					+ net->ipv6.sysctl.ip6_rt_gc_interval));
 	else
 		del_timer(&net->ipv6.ip6_fib_timer);
-	spin_unlock_bh(&fib6_gc_lock);
+	spin_unlock_bh(&net->ipv6.fib6_gc_lock);
 }
 
 static void fib6_gc_timer_cb(unsigned long arg)
@@ -1816,6 +1814,7 @@ static int __net_init fib6_net_init(struct net *net)
 {
 	size_t size = sizeof(struct hlist_head) * FIB6_TABLE_HASHSZ;
 
+	spin_lock_init(&net->ipv6.fib6_gc_lock);
 	rwlock_init(&net->ipv6.fib6_walker_lock);
 	INIT_LIST_HEAD(&net->ipv6.fib6_walkers);
 	setup_timer(&net->ipv6.ip6_fib_timer, fib6_gc_timer_cb, (unsigned long)net);
-- 
cgit v0.10.2


From b121d1e74d1f24654bdc3165d3db1ca149501356 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:13 -0800
Subject: bpf: prevent kprobe+bpf deadlocks

if kprobe is placed within update or delete hash map helpers
that hold bucket spin lock and triggered bpf program is trying to
grab the spinlock for the same bucket on the same cpu, it will
deadlock.
Fix it by extending existing recursion prevention mechanism.

Note, map_lookup and other tracing helpers don't have this problem,
since they don't hold any locks and don't modify global data.
bpf_trace_printk has its own recursive check and ok as well.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 51e498e..4b07082 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -10,6 +10,7 @@
 #include <uapi/linux/bpf.h>
 #include <linux/workqueue.h>
 #include <linux/file.h>
+#include <linux/percpu.h>
 
 struct bpf_map;
 
@@ -163,6 +164,8 @@ bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *f
 const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
 
 #ifdef CONFIG_BPF_SYSCALL
+DECLARE_PER_CPU(int, bpf_prog_active);
+
 void bpf_register_prog_type(struct bpf_prog_type_list *tl);
 void bpf_register_map_type(struct bpf_map_type_list *tl);
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index c95a753..dc99f6a 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -18,6 +18,8 @@
 #include <linux/filter.h>
 #include <linux/version.h>
 
+DEFINE_PER_CPU(int, bpf_prog_active);
+
 int sysctl_unprivileged_bpf_disabled __read_mostly;
 
 static LIST_HEAD(bpf_map_types);
@@ -347,6 +349,11 @@ static int map_update_elem(union bpf_attr *attr)
 	if (copy_from_user(value, uvalue, value_size) != 0)
 		goto free_value;
 
+	/* must increment bpf_prog_active to avoid kprobe+bpf triggering from
+	 * inside bpf map update or delete otherwise deadlocks are possible
+	 */
+	preempt_disable();
+	__this_cpu_inc(bpf_prog_active);
 	if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH) {
 		err = bpf_percpu_hash_update(map, key, value, attr->flags);
 	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
@@ -356,6 +363,8 @@ static int map_update_elem(union bpf_attr *attr)
 		err = map->ops->map_update_elem(map, key, value, attr->flags);
 		rcu_read_unlock();
 	}
+	__this_cpu_dec(bpf_prog_active);
+	preempt_enable();
 
 free_value:
 	kfree(value);
@@ -394,9 +403,13 @@ static int map_delete_elem(union bpf_attr *attr)
 	if (copy_from_user(key, ukey, map->key_size) != 0)
 		goto free_key;
 
+	preempt_disable();
+	__this_cpu_inc(bpf_prog_active);
 	rcu_read_lock();
 	err = map->ops->map_delete_elem(map, key);
 	rcu_read_unlock();
+	__this_cpu_dec(bpf_prog_active);
+	preempt_enable();
 
 free_key:
 	kfree(key);
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 4b8caa3..3e4ffb3 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -13,8 +13,6 @@
 #include <linux/ctype.h>
 #include "trace.h"
 
-static DEFINE_PER_CPU(int, bpf_prog_active);
-
 /**
  * trace_call_bpf - invoke BPF program
  * @prog: BPF program
-- 
cgit v0.10.2


From e19494edab82f55a633911f25094581891bdc351 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:14 -0800
Subject: bpf: introduce percpu_freelist

Introduce simple percpu_freelist to keep single list of elements
spread across per-cpu singly linked lists.

/* push element into the list */
void pcpu_freelist_push(struct pcpu_freelist *, struct pcpu_freelist_node *);

/* pop element from the list */
struct pcpu_freelist_node *pcpu_freelist_pop(struct pcpu_freelist *);

The object is pushed to the current cpu list.
Pop first trying to get the object from the current cpu list,
if it's empty goes to the neigbour cpu list.

For bpf program usage pattern the collision rate is very low,
since programs push and pop the objects typically on the same cpu.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index 8a932d0..eed911d 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -1,7 +1,7 @@
 obj-y := core.o
 
 obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o
-obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o
+obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o percpu_freelist.o
 ifeq ($(CONFIG_PERF_EVENTS),y)
 obj-$(CONFIG_BPF_SYSCALL) += stackmap.o
 endif
diff --git a/kernel/bpf/percpu_freelist.c b/kernel/bpf/percpu_freelist.c
new file mode 100644
index 0000000..5c51d19
--- /dev/null
+++ b/kernel/bpf/percpu_freelist.c
@@ -0,0 +1,100 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include "percpu_freelist.h"
+
+int pcpu_freelist_init(struct pcpu_freelist *s)
+{
+	int cpu;
+
+	s->freelist = alloc_percpu(struct pcpu_freelist_head);
+	if (!s->freelist)
+		return -ENOMEM;
+
+	for_each_possible_cpu(cpu) {
+		struct pcpu_freelist_head *head = per_cpu_ptr(s->freelist, cpu);
+
+		raw_spin_lock_init(&head->lock);
+		head->first = NULL;
+	}
+	return 0;
+}
+
+void pcpu_freelist_destroy(struct pcpu_freelist *s)
+{
+	free_percpu(s->freelist);
+}
+
+static inline void __pcpu_freelist_push(struct pcpu_freelist_head *head,
+					struct pcpu_freelist_node *node)
+{
+	raw_spin_lock(&head->lock);
+	node->next = head->first;
+	head->first = node;
+	raw_spin_unlock(&head->lock);
+}
+
+void pcpu_freelist_push(struct pcpu_freelist *s,
+			struct pcpu_freelist_node *node)
+{
+	struct pcpu_freelist_head *head = this_cpu_ptr(s->freelist);
+
+	__pcpu_freelist_push(head, node);
+}
+
+void pcpu_freelist_populate(struct pcpu_freelist *s, void *buf, u32 elem_size,
+			    u32 nr_elems)
+{
+	struct pcpu_freelist_head *head;
+	unsigned long flags;
+	int i, cpu, pcpu_entries;
+
+	pcpu_entries = nr_elems / num_possible_cpus() + 1;
+	i = 0;
+
+	/* disable irq to workaround lockdep false positive
+	 * in bpf usage pcpu_freelist_populate() will never race
+	 * with pcpu_freelist_push()
+	 */
+	local_irq_save(flags);
+	for_each_possible_cpu(cpu) {
+again:
+		head = per_cpu_ptr(s->freelist, cpu);
+		__pcpu_freelist_push(head, buf);
+		i++;
+		buf += elem_size;
+		if (i == nr_elems)
+			break;
+		if (i % pcpu_entries)
+			goto again;
+	}
+	local_irq_restore(flags);
+}
+
+struct pcpu_freelist_node *pcpu_freelist_pop(struct pcpu_freelist *s)
+{
+	struct pcpu_freelist_head *head;
+	struct pcpu_freelist_node *node;
+	int orig_cpu, cpu;
+
+	orig_cpu = cpu = raw_smp_processor_id();
+	while (1) {
+		head = per_cpu_ptr(s->freelist, cpu);
+		raw_spin_lock(&head->lock);
+		node = head->first;
+		if (node) {
+			head->first = node->next;
+			raw_spin_unlock(&head->lock);
+			return node;
+		}
+		raw_spin_unlock(&head->lock);
+		cpu = cpumask_next(cpu, cpu_possible_mask);
+		if (cpu >= nr_cpu_ids)
+			cpu = 0;
+		if (cpu == orig_cpu)
+			return NULL;
+	}
+}
diff --git a/kernel/bpf/percpu_freelist.h b/kernel/bpf/percpu_freelist.h
new file mode 100644
index 0000000..3049aae
--- /dev/null
+++ b/kernel/bpf/percpu_freelist.h
@@ -0,0 +1,31 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#ifndef __PERCPU_FREELIST_H__
+#define __PERCPU_FREELIST_H__
+#include <linux/spinlock.h>
+#include <linux/percpu.h>
+
+struct pcpu_freelist_head {
+	struct pcpu_freelist_node *first;
+	raw_spinlock_t lock;
+};
+
+struct pcpu_freelist {
+	struct pcpu_freelist_head __percpu *freelist;
+};
+
+struct pcpu_freelist_node {
+	struct pcpu_freelist_node *next;
+};
+
+void pcpu_freelist_push(struct pcpu_freelist *, struct pcpu_freelist_node *);
+struct pcpu_freelist_node *pcpu_freelist_pop(struct pcpu_freelist *);
+void pcpu_freelist_populate(struct pcpu_freelist *s, void *buf, u32 elem_size,
+			    u32 nr_elems);
+int pcpu_freelist_init(struct pcpu_freelist *);
+void pcpu_freelist_destroy(struct pcpu_freelist *s);
+#endif
-- 
cgit v0.10.2


From 6c90598174322b8888029e40dd84a4eb01f56afe Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:15 -0800
Subject: bpf: pre-allocate hash map elements

If kprobe is placed on spin_unlock then calling kmalloc/kfree from
bpf programs is not safe, since the following dead lock is possible:
kfree->spin_lock(kmem_cache_node->lock)...spin_unlock->kprobe->
bpf_prog->map_update->kmalloc->spin_lock(of the same kmem_cache_node->lock)
and deadlocks.

The following solutions were considered and some implemented, but
eventually discarded
- kmem_cache_create for every map
- add recursion check to slow-path of slub
- use reserved memory in bpf_map_update for in_irq or in preempt_disabled
- kmalloc via irq_work

At the end pre-allocation of all map elements turned out to be the simplest
solution and since the user is charged upfront for all the memory, such
pre-allocation doesn't affect the user space visible behavior.

Since it's impossible to tell whether kprobe is triggered in a safe
location from kmalloc point of view, use pre-allocation by default
and introduce new BPF_F_NO_PREALLOC flag.

While testing of per-cpu hash maps it was discovered
that alloc_percpu(GFP_ATOMIC) has odd corner cases and often
fails to allocate memory even when 90% of it is free.
The pre-allocation of per-cpu hash elements solves this problem as well.

Turned out that bpf_map_update() quickly followed by
bpf_map_lookup()+bpf_map_delete() is very common pattern used
in many of iovisor/bcc/tools, so there is additional benefit of
pre-allocation, since such use cases are must faster.

Since all hash map elements are now pre-allocated we can remove
atomic increment of htab->count and save few more cycles.

Also add bpf_map_precharge_memlock() to check rlimit_memlock early to avoid
large malloc/free done by users who don't have sufficient limits.

Pre-allocation is done with vmalloc and alloc/free is done
via percpu_freelist. Here are performance numbers for different
pre-allocation algorithms that were implemented, but discarded
in favor of percpu_freelist:

1 cpu:
pcpu_ida	2.1M
pcpu_ida nolock	2.3M
bt		2.4M
kmalloc		1.8M
hlist+spinlock	2.3M
pcpu_freelist	2.6M

4 cpu:
pcpu_ida	1.5M
pcpu_ida nolock	1.8M
bt w/smp_align	1.7M
bt no/smp_align	1.1M
kmalloc		0.7M
hlist+spinlock	0.2M
pcpu_freelist	2.0M

8 cpu:
pcpu_ida	0.7M
bt w/smp_align	0.8M
kmalloc		0.4M
pcpu_freelist	1.5M

32 cpu:
kmalloc		0.13M
pcpu_freelist	0.49M

pcpu_ida nolock is a modified percpu_ida algorithm without
percpu_ida_cpu locks and without cross-cpu tag stealing.
It's faster than existing percpu_ida, but not as fast as pcpu_freelist.

bt is a variant of block/blk-mq-tag.c simlified and customized
for bpf use case. bt w/smp_align is using cache line for every 'long'
(similar to blk-mq-tag). bt no/smp_align allocates 'long'
bitmasks continuously to save memory. It's comparable to percpu_ida
and in some cases faster, but slower than percpu_freelist

hlist+spinlock is the simplest free list with single spinlock.
As expeceted it has very bad scaling in SMP.

kmalloc is existing implementation which is still available via
BPF_F_NO_PREALLOC flag. It's significantly slower in single cpu and
in 8 cpu setup it's 3 times slower than pre-allocation with pcpu_freelist,
but saves memory, so in cases where map->max_entries can be large
and number of map update/delete per second is low, it may make
sense to use it.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 4b07082..efd1d4c 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -37,6 +37,7 @@ struct bpf_map {
 	u32 key_size;
 	u32 value_size;
 	u32 max_entries;
+	u32 map_flags;
 	u32 pages;
 	struct user_struct *user;
 	const struct bpf_map_ops *ops;
@@ -178,6 +179,7 @@ struct bpf_map *__bpf_map_get(struct fd f);
 void bpf_map_inc(struct bpf_map *map, bool uref);
 void bpf_map_put_with_uref(struct bpf_map *map);
 void bpf_map_put(struct bpf_map *map);
+int bpf_map_precharge_memlock(u32 pages);
 
 extern int sysctl_unprivileged_bpf_disabled;
 
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 9221f65..0e30b19 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -101,12 +101,15 @@ enum bpf_prog_type {
 #define BPF_NOEXIST	1 /* create new element if it didn't exist */
 #define BPF_EXIST	2 /* update existing element */
 
+#define BPF_F_NO_PREALLOC	(1U << 0)
+
 union bpf_attr {
 	struct { /* anonymous struct used by BPF_MAP_CREATE command */
 		__u32	map_type;	/* one of enum bpf_map_type */
 		__u32	key_size;	/* size of key in bytes */
 		__u32	value_size;	/* size of value in bytes */
 		__u32	max_entries;	/* max number of entries in a map */
+		__u32	map_flags;	/* prealloc or not */
 	};
 
 	struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index a68e951..fff3650 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -1,4 +1,5 @@
 /* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
+ * Copyright (c) 2016 Facebook
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of version 2 of the GNU General Public
@@ -13,6 +14,7 @@
 #include <linux/jhash.h>
 #include <linux/filter.h>
 #include <linux/vmalloc.h>
+#include "percpu_freelist.h"
 
 struct bucket {
 	struct hlist_head head;
@@ -22,6 +24,8 @@ struct bucket {
 struct bpf_htab {
 	struct bpf_map map;
 	struct bucket *buckets;
+	void *elems;
+	struct pcpu_freelist freelist;
 	atomic_t count;	/* number of elements in this hashtable */
 	u32 n_buckets;	/* number of hash buckets */
 	u32 elem_size;	/* size of each element in bytes */
@@ -29,15 +33,86 @@ struct bpf_htab {
 
 /* each htab element is struct htab_elem + key + value */
 struct htab_elem {
-	struct hlist_node hash_node;
-	struct rcu_head rcu;
 	union {
-		u32 hash;
-		u32 key_size;
+		struct hlist_node hash_node;
+		struct bpf_htab *htab;
+		struct pcpu_freelist_node fnode;
 	};
+	struct rcu_head rcu;
+	u32 hash;
 	char key[0] __aligned(8);
 };
 
+static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size,
+				     void __percpu *pptr)
+{
+	*(void __percpu **)(l->key + key_size) = pptr;
+}
+
+static inline void __percpu *htab_elem_get_ptr(struct htab_elem *l, u32 key_size)
+{
+	return *(void __percpu **)(l->key + key_size);
+}
+
+static struct htab_elem *get_htab_elem(struct bpf_htab *htab, int i)
+{
+	return (struct htab_elem *) (htab->elems + i * htab->elem_size);
+}
+
+static void htab_free_elems(struct bpf_htab *htab)
+{
+	int i;
+
+	if (htab->map.map_type != BPF_MAP_TYPE_PERCPU_HASH)
+		goto free_elems;
+
+	for (i = 0; i < htab->map.max_entries; i++) {
+		void __percpu *pptr;
+
+		pptr = htab_elem_get_ptr(get_htab_elem(htab, i),
+					 htab->map.key_size);
+		free_percpu(pptr);
+	}
+free_elems:
+	vfree(htab->elems);
+}
+
+static int prealloc_elems_and_freelist(struct bpf_htab *htab)
+{
+	int err = -ENOMEM, i;
+
+	htab->elems = vzalloc(htab->elem_size * htab->map.max_entries);
+	if (!htab->elems)
+		return -ENOMEM;
+
+	if (htab->map.map_type != BPF_MAP_TYPE_PERCPU_HASH)
+		goto skip_percpu_elems;
+
+	for (i = 0; i < htab->map.max_entries; i++) {
+		u32 size = round_up(htab->map.value_size, 8);
+		void __percpu *pptr;
+
+		pptr = __alloc_percpu_gfp(size, 8, GFP_USER | __GFP_NOWARN);
+		if (!pptr)
+			goto free_elems;
+		htab_elem_set_ptr(get_htab_elem(htab, i), htab->map.key_size,
+				  pptr);
+	}
+
+skip_percpu_elems:
+	err = pcpu_freelist_init(&htab->freelist);
+	if (err)
+		goto free_elems;
+
+	pcpu_freelist_populate(&htab->freelist, htab->elems, htab->elem_size,
+			       htab->map.max_entries);
+	return 0;
+
+free_elems:
+	htab_free_elems(htab);
+	return err;
+}
+
 /* Called from syscall */
 static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 {
@@ -46,6 +121,10 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 	int err, i;
 	u64 cost;
 
+	if (attr->map_flags & ~BPF_F_NO_PREALLOC)
+		/* reserved bits should not be used */
+		return ERR_PTR(-EINVAL);
+
 	htab = kzalloc(sizeof(*htab), GFP_USER);
 	if (!htab)
 		return ERR_PTR(-ENOMEM);
@@ -55,6 +134,7 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 	htab->map.key_size = attr->key_size;
 	htab->map.value_size = attr->value_size;
 	htab->map.max_entries = attr->max_entries;
+	htab->map.map_flags = attr->map_flags;
 
 	/* check sanity of attributes.
 	 * value_size == 0 may be allowed in the future to use map as a set
@@ -92,7 +172,7 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 	if (percpu)
 		htab->elem_size += sizeof(void *);
 	else
-		htab->elem_size += htab->map.value_size;
+		htab->elem_size += round_up(htab->map.value_size, 8);
 
 	/* prevent zero size kmalloc and check for u32 overflow */
 	if (htab->n_buckets == 0 ||
@@ -112,6 +192,11 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 
 	htab->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT;
 
+	/* if map size is larger than memlock limit, reject it early */
+	err = bpf_map_precharge_memlock(htab->map.pages);
+	if (err)
+		goto free_htab;
+
 	err = -ENOMEM;
 	htab->buckets = kmalloc_array(htab->n_buckets, sizeof(struct bucket),
 				      GFP_USER | __GFP_NOWARN);
@@ -127,10 +212,16 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr)
 		raw_spin_lock_init(&htab->buckets[i].lock);
 	}
 
-	atomic_set(&htab->count, 0);
+	if (!(attr->map_flags & BPF_F_NO_PREALLOC)) {
+		err = prealloc_elems_and_freelist(htab);
+		if (err)
+			goto free_buckets;
+	}
 
 	return &htab->map;
 
+free_buckets:
+	kvfree(htab->buckets);
 free_htab:
 	kfree(htab);
 	return ERR_PTR(err);
@@ -249,42 +340,42 @@ find_first_elem:
 		}
 	}
 
-	/* itereated over all buckets and all elements */
+	/* iterated over all buckets and all elements */
 	return -ENOENT;
 }
 
-
-static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size,
-				     void __percpu *pptr)
-{
-	*(void __percpu **)(l->key + key_size) = pptr;
-}
-
-static inline void __percpu *htab_elem_get_ptr(struct htab_elem *l, u32 key_size)
-{
-	return *(void __percpu **)(l->key + key_size);
-}
-
-static void htab_percpu_elem_free(struct htab_elem *l)
+static void htab_elem_free(struct bpf_htab *htab, struct htab_elem *l)
 {
-	free_percpu(htab_elem_get_ptr(l, l->key_size));
+	if (htab->map.map_type == BPF_MAP_TYPE_PERCPU_HASH)
+		free_percpu(htab_elem_get_ptr(l, htab->map.key_size));
 	kfree(l);
+
 }
 
-static void htab_percpu_elem_free_rcu(struct rcu_head *head)
+static void htab_elem_free_rcu(struct rcu_head *head)
 {
 	struct htab_elem *l = container_of(head, struct htab_elem, rcu);
+	struct bpf_htab *htab = l->htab;
 
-	htab_percpu_elem_free(l);
+	/* must increment bpf_prog_active to avoid kprobe+bpf triggering while
+	 * we're calling kfree, otherwise deadlock is possible if kprobes
+	 * are placed somewhere inside of slub
+	 */
+	preempt_disable();
+	__this_cpu_inc(bpf_prog_active);
+	htab_elem_free(htab, l);
+	__this_cpu_dec(bpf_prog_active);
+	preempt_enable();
 }
 
-static void free_htab_elem(struct htab_elem *l, bool percpu, u32 key_size)
+static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l)
 {
-	if (percpu) {
-		l->key_size = key_size;
-		call_rcu(&l->rcu, htab_percpu_elem_free_rcu);
+	if (!(htab->map.map_flags & BPF_F_NO_PREALLOC)) {
+		pcpu_freelist_push(&htab->freelist, &l->fnode);
 	} else {
-		kfree_rcu(l, rcu);
+		atomic_dec(&htab->count);
+		l->htab = htab;
+		call_rcu(&l->rcu, htab_elem_free_rcu);
 	}
 }
 
@@ -293,23 +384,39 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 					 bool percpu, bool onallcpus)
 {
 	u32 size = htab->map.value_size;
+	bool prealloc = !(htab->map.map_flags & BPF_F_NO_PREALLOC);
 	struct htab_elem *l_new;
 	void __percpu *pptr;
 
-	l_new = kmalloc(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN);
-	if (!l_new)
-		return NULL;
+	if (prealloc) {
+		l_new = (struct htab_elem *)pcpu_freelist_pop(&htab->freelist);
+		if (!l_new)
+			return ERR_PTR(-E2BIG);
+	} else {
+		if (atomic_inc_return(&htab->count) > htab->map.max_entries) {
+			atomic_dec(&htab->count);
+			return ERR_PTR(-E2BIG);
+		}
+		l_new = kmalloc(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN);
+		if (!l_new)
+			return ERR_PTR(-ENOMEM);
+	}
 
 	memcpy(l_new->key, key, key_size);
 	if (percpu) {
 		/* round up value_size to 8 bytes */
 		size = round_up(size, 8);
 
-		/* alloc_percpu zero-fills */
-		pptr = __alloc_percpu_gfp(size, 8, GFP_ATOMIC | __GFP_NOWARN);
-		if (!pptr) {
-			kfree(l_new);
-			return NULL;
+		if (prealloc) {
+			pptr = htab_elem_get_ptr(l_new, key_size);
+		} else {
+			/* alloc_percpu zero-fills */
+			pptr = __alloc_percpu_gfp(size, 8,
+						  GFP_ATOMIC | __GFP_NOWARN);
+			if (!pptr) {
+				kfree(l_new);
+				return ERR_PTR(-ENOMEM);
+			}
 		}
 
 		if (!onallcpus) {
@@ -324,7 +431,8 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 				off += size;
 			}
 		}
-		htab_elem_set_ptr(l_new, key_size, pptr);
+		if (!prealloc)
+			htab_elem_set_ptr(l_new, key_size, pptr);
 	} else {
 		memcpy(l_new->key + round_up(key_size, 8), value, size);
 	}
@@ -336,12 +444,6 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 static int check_flags(struct bpf_htab *htab, struct htab_elem *l_old,
 		       u64 map_flags)
 {
-	if (!l_old && unlikely(atomic_read(&htab->count) >= htab->map.max_entries))
-		/* if elem with this 'key' doesn't exist and we've reached
-		 * max_entries limit, fail insertion of new elem
-		 */
-		return -E2BIG;
-
 	if (l_old && map_flags == BPF_NOEXIST)
 		/* elem already exists */
 		return -EEXIST;
@@ -375,13 +477,6 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 
 	hash = htab_map_hash(key, key_size);
 
-	/* allocate new element outside of the lock, since
-	 * we're most likley going to insert it
-	 */
-	l_new = alloc_htab_elem(htab, key, value, key_size, hash, false, false);
-	if (!l_new)
-		return -ENOMEM;
-
 	b = __select_bucket(htab, hash);
 	head = &b->head;
 
@@ -394,21 +489,24 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value,
 	if (ret)
 		goto err;
 
+	l_new = alloc_htab_elem(htab, key, value, key_size, hash, false, false);
+	if (IS_ERR(l_new)) {
+		/* all pre-allocated elements are in use or memory exhausted */
+		ret = PTR_ERR(l_new);
+		goto err;
+	}
+
 	/* add new element to the head of the list, so that
 	 * concurrent search will find it before old elem
 	 */
 	hlist_add_head_rcu(&l_new->hash_node, head);
 	if (l_old) {
 		hlist_del_rcu(&l_old->hash_node);
-		kfree_rcu(l_old, rcu);
-	} else {
-		atomic_inc(&htab->count);
+		free_htab_elem(htab, l_old);
 	}
-	raw_spin_unlock_irqrestore(&b->lock, flags);
-	return 0;
+	ret = 0;
 err:
 	raw_spin_unlock_irqrestore(&b->lock, flags);
-	kfree(l_new);
 	return ret;
 }
 
@@ -466,12 +564,11 @@ static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key,
 	} else {
 		l_new = alloc_htab_elem(htab, key, value, key_size,
 					hash, true, onallcpus);
-		if (!l_new) {
-			ret = -ENOMEM;
+		if (IS_ERR(l_new)) {
+			ret = PTR_ERR(l_new);
 			goto err;
 		}
 		hlist_add_head_rcu(&l_new->hash_node, head);
-		atomic_inc(&htab->count);
 	}
 	ret = 0;
 err:
@@ -489,7 +586,6 @@ static int htab_percpu_map_update_elem(struct bpf_map *map, void *key,
 static int htab_map_delete_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
-	bool percpu = map->map_type == BPF_MAP_TYPE_PERCPU_HASH;
 	struct hlist_head *head;
 	struct bucket *b;
 	struct htab_elem *l;
@@ -511,8 +607,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key)
 
 	if (l) {
 		hlist_del_rcu(&l->hash_node);
-		atomic_dec(&htab->count);
-		free_htab_elem(l, percpu, key_size);
+		free_htab_elem(htab, l);
 		ret = 0;
 	}
 
@@ -531,17 +626,10 @@ static void delete_all_elements(struct bpf_htab *htab)
 
 		hlist_for_each_entry_safe(l, n, head, hash_node) {
 			hlist_del_rcu(&l->hash_node);
-			atomic_dec(&htab->count);
-			if (htab->map.map_type == BPF_MAP_TYPE_PERCPU_HASH) {
-				l->key_size = htab->map.key_size;
-				htab_percpu_elem_free(l);
-			} else {
-				kfree(l);
-			}
+			htab_elem_free(htab, l);
 		}
 	}
 }
-
 /* Called when map->refcnt goes to zero, either from workqueue or from syscall */
 static void htab_map_free(struct bpf_map *map)
 {
@@ -554,10 +642,16 @@ static void htab_map_free(struct bpf_map *map)
 	 */
 	synchronize_rcu();
 
-	/* some of kfree_rcu() callbacks for elements of this map may not have
-	 * executed. It's ok. Proceed to free residual elements and map itself
+	/* some of free_htab_elem() callbacks for elements of this map may
+	 * not have executed. Wait for them.
 	 */
-	delete_all_elements(htab);
+	rcu_barrier();
+	if (htab->map.map_flags & BPF_F_NO_PREALLOC) {
+		delete_all_elements(htab);
+	} else {
+		htab_free_elems(htab);
+		pcpu_freelist_destroy(&htab->freelist);
+	}
 	kvfree(htab->buckets);
 	kfree(htab);
 }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index dc99f6a..cbd94b2 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -48,6 +48,19 @@ void bpf_register_map_type(struct bpf_map_type_list *tl)
 	list_add(&tl->list_node, &bpf_map_types);
 }
 
+int bpf_map_precharge_memlock(u32 pages)
+{
+	struct user_struct *user = get_current_user();
+	unsigned long memlock_limit, cur;
+
+	memlock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
+	cur = atomic_long_read(&user->locked_vm);
+	free_uid(user);
+	if (cur + pages > memlock_limit)
+		return -EPERM;
+	return 0;
+}
+
 static int bpf_map_charge_memlock(struct bpf_map *map)
 {
 	struct user_struct *user = get_current_user();
@@ -153,7 +166,7 @@ int bpf_map_new_fd(struct bpf_map *map)
 		   offsetof(union bpf_attr, CMD##_LAST_FIELD) - \
 		   sizeof(attr->CMD##_LAST_FIELD)) != NULL
 
-#define BPF_MAP_CREATE_LAST_FIELD max_entries
+#define BPF_MAP_CREATE_LAST_FIELD map_flags
 /* called via syscall */
 static int map_create(union bpf_attr *attr)
 {
-- 
cgit v0.10.2


From 823707b68d6e6c4b1be619b039c7045fef1740e6 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:16 -0800
Subject: bpf: check for reserved flag bits in array and stack maps

Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index bd3bdf2..76d5a79 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -53,7 +53,7 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 
 	/* check sanity of attributes */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
-	    attr->value_size == 0)
+	    attr->value_size == 0 || attr->map_flags)
 		return ERR_PTR(-EINVAL);
 
 	if (attr->value_size >= 1 << (KMALLOC_SHIFT_MAX - 1))
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 8a60ee1..f0a02c3 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -35,6 +35,9 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 	if (!capable(CAP_SYS_ADMIN))
 		return ERR_PTR(-EPERM);
 
+	if (attr->map_flags)
+		return ERR_PTR(-EINVAL);
+
 	/* check sanity of attributes */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
 	    value_size < 8 || value_size % 8 ||
-- 
cgit v0.10.2


From 557c0c6e7df8e14a46bd7560d193fa5bbc00a858 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:17 -0800
Subject: bpf: convert stackmap to pre-allocation

It was observed that calling bpf_get_stackid() from a kprobe inside
slub or from spin_unlock causes similar deadlock as with hashmap,
therefore convert stackmap to use pre-allocated memory.

The call_rcu is no longer feasible mechanism, since delayed freeing
causes bpf_get_stackid() to fail unpredictably when number of actual
stacks is significantly less than user requested max_entries.
Since elements are no longer freed into slub, we can push elements into
freelist immediately and let them be recycled.
However the very unlikley race between user space map_lookup() and
program-side recycling is possible:
     cpu0                          cpu1
     ----                          ----
user does lookup(stackidX)
starts copying ips into buffer
                                   delete(stackidX)
                                   calls bpf_get_stackid()
				   which recyles the element and
                                   overwrites with new stack trace

To avoid user space seeing a partial stack trace consisting of two
merged stack traces, do bucket = xchg(, NULL); copy; xchg(,bucket);
to preserve consistent stack trace delivery to user space.
Now we can move memset(,0) of left-over element value from critical
path of bpf_get_stackid() into slow-path of user space lookup.
Also disallow lookup() from bpf program, since it's useless and
program shouldn't be messing with collected stack trace.

Note that similar race between user space lookup and kernel side updates
is also present in hashmap, but it's not a new race. bpf programs were
always allowed to modify hash and array map elements while user space
is copying them.

Fixes: d5a3b1f69186 ("bpf: introduce BPF_MAP_TYPE_STACK_TRACE")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index efd1d4c..21ee41b 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -195,6 +195,7 @@ int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value,
 			   u64 flags);
 int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value,
 			    u64 flags);
+int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
 
 /* memcpy that is used with 8-byte aligned pointers, power-of-8 size and
  * forced to use 'long' read/writes to try to atomically copy long counters.
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index f0a02c3..499d9e9 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -10,9 +10,10 @@
 #include <linux/vmalloc.h>
 #include <linux/stacktrace.h>
 #include <linux/perf_event.h>
+#include "percpu_freelist.h"
 
 struct stack_map_bucket {
-	struct rcu_head rcu;
+	struct pcpu_freelist_node fnode;
 	u32 hash;
 	u32 nr;
 	u64 ip[];
@@ -20,10 +21,34 @@ struct stack_map_bucket {
 
 struct bpf_stack_map {
 	struct bpf_map map;
+	void *elems;
+	struct pcpu_freelist freelist;
 	u32 n_buckets;
-	struct stack_map_bucket __rcu *buckets[];
+	struct stack_map_bucket *buckets[];
 };
 
+static int prealloc_elems_and_freelist(struct bpf_stack_map *smap)
+{
+	u32 elem_size = sizeof(struct stack_map_bucket) + smap->map.value_size;
+	int err;
+
+	smap->elems = vzalloc(elem_size * smap->map.max_entries);
+	if (!smap->elems)
+		return -ENOMEM;
+
+	err = pcpu_freelist_init(&smap->freelist);
+	if (err)
+		goto free_elems;
+
+	pcpu_freelist_populate(&smap->freelist, smap->elems, elem_size,
+			       smap->map.max_entries);
+	return 0;
+
+free_elems:
+	vfree(smap->elems);
+	return err;
+}
+
 /* Called from syscall */
 static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 {
@@ -70,12 +95,22 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 	smap->n_buckets = n_buckets;
 	smap->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT;
 
+	err = bpf_map_precharge_memlock(smap->map.pages);
+	if (err)
+		goto free_smap;
+
 	err = get_callchain_buffers();
 	if (err)
 		goto free_smap;
 
+	err = prealloc_elems_and_freelist(smap);
+	if (err)
+		goto put_buffers;
+
 	return &smap->map;
 
+put_buffers:
+	put_callchain_buffers();
 free_smap:
 	kvfree(smap);
 	return ERR_PTR(err);
@@ -121,7 +156,7 @@ static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5)
 	ips = trace->ip + skip + init_nr;
 	hash = jhash2((u32 *)ips, trace_len / sizeof(u32), 0);
 	id = hash & (smap->n_buckets - 1);
-	bucket = rcu_dereference(smap->buckets[id]);
+	bucket = READ_ONCE(smap->buckets[id]);
 
 	if (bucket && bucket->hash == hash) {
 		if (flags & BPF_F_FAST_STACK_CMP)
@@ -135,19 +170,18 @@ static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5)
 	if (bucket && !(flags & BPF_F_REUSE_STACKID))
 		return -EEXIST;
 
-	new_bucket = kmalloc(sizeof(struct stack_map_bucket) + map->value_size,
-			     GFP_ATOMIC | __GFP_NOWARN);
+	new_bucket = (struct stack_map_bucket *)
+		pcpu_freelist_pop(&smap->freelist);
 	if (unlikely(!new_bucket))
 		return -ENOMEM;
 
 	memcpy(new_bucket->ip, ips, trace_len);
-	memset(new_bucket->ip + trace_len / 8, 0, map->value_size - trace_len);
 	new_bucket->hash = hash;
 	new_bucket->nr = trace_nr;
 
 	old_bucket = xchg(&smap->buckets[id], new_bucket);
 	if (old_bucket)
-		kfree_rcu(old_bucket, rcu);
+		pcpu_freelist_push(&smap->freelist, &old_bucket->fnode);
 	return id;
 }
 
@@ -160,17 +194,34 @@ const struct bpf_func_proto bpf_get_stackid_proto = {
 	.arg3_type	= ARG_ANYTHING,
 };
 
-/* Called from syscall or from eBPF program */
+/* Called from eBPF program */
 static void *stack_map_lookup_elem(struct bpf_map *map, void *key)
 {
+	return NULL;
+}
+
+/* Called from syscall */
+int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value)
+{
 	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
-	struct stack_map_bucket *bucket;
-	u32 id = *(u32 *)key;
+	struct stack_map_bucket *bucket, *old_bucket;
+	u32 id = *(u32 *)key, trace_len;
 
 	if (unlikely(id >= smap->n_buckets))
-		return NULL;
-	bucket = rcu_dereference(smap->buckets[id]);
-	return bucket ? bucket->ip : NULL;
+		return -ENOENT;
+
+	bucket = xchg(&smap->buckets[id], NULL);
+	if (!bucket)
+		return -ENOENT;
+
+	trace_len = bucket->nr * sizeof(u64);
+	memcpy(value, bucket->ip, trace_len);
+	memset(value + trace_len, 0, map->value_size - trace_len);
+
+	old_bucket = xchg(&smap->buckets[id], bucket);
+	if (old_bucket)
+		pcpu_freelist_push(&smap->freelist, &old_bucket->fnode);
+	return 0;
 }
 
 static int stack_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
@@ -196,7 +247,7 @@ static int stack_map_delete_elem(struct bpf_map *map, void *key)
 
 	old_bucket = xchg(&smap->buckets[id], NULL);
 	if (old_bucket) {
-		kfree_rcu(old_bucket, rcu);
+		pcpu_freelist_push(&smap->freelist, &old_bucket->fnode);
 		return 0;
 	} else {
 		return -ENOENT;
@@ -207,13 +258,12 @@ static int stack_map_delete_elem(struct bpf_map *map, void *key)
 static void stack_map_free(struct bpf_map *map)
 {
 	struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map);
-	int i;
 
+	/* wait for bpf programs to complete before freeing stack map */
 	synchronize_rcu();
 
-	for (i = 0; i < smap->n_buckets; i++)
-		if (smap->buckets[i])
-			kfree_rcu(smap->buckets[i], rcu);
+	vfree(smap->elems);
+	pcpu_freelist_destroy(&smap->freelist);
 	kvfree(smap);
 	put_callchain_buffers();
 }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index cbd94b2..2978d0d 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -290,6 +290,8 @@ static int map_lookup_elem(union bpf_attr *attr)
 		err = bpf_percpu_hash_copy(map, key, value);
 	} else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) {
 		err = bpf_percpu_array_copy(map, key, value);
+	} else if (map->map_type == BPF_MAP_TYPE_STACK_TRACE) {
+		err = bpf_stackmap_copy(map, key, value);
 	} else {
 		rcu_read_lock();
 		ptr = map->ops->map_lookup_elem(map, key);
-- 
cgit v0.10.2


From 618ec9a7b1fda46f8dd4a630ded983aeb51218f3 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:18 -0800
Subject: samples/bpf: make map creation more verbose

map creation is typically the first one to fail when rlimits are
too low, not enough memory, etc
Make this failure scenario more verbose

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/bpf_load.c b/samples/bpf/bpf_load.c
index da86a8e..816bca5 100644
--- a/samples/bpf/bpf_load.c
+++ b/samples/bpf/bpf_load.c
@@ -158,8 +158,11 @@ static int load_maps(struct bpf_map_def *maps, int len)
 					   maps[i].key_size,
 					   maps[i].value_size,
 					   maps[i].max_entries);
-		if (map_fd[i] < 0)
+		if (map_fd[i] < 0) {
+			printf("failed to create a map: %d %s\n",
+			       errno, strerror(errno));
 			return 1;
+		}
 
 		if (maps[i].type == BPF_MAP_TYPE_PROG_ARRAY)
 			prog_array_fd = map_fd[i];
-- 
cgit v0.10.2


From 3622e7e4935105991dc648bca650c858576aecda Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:19 -0800
Subject: samples/bpf: move ksym_search() into library

move ksym search from offwaketime into library to be reused
in other tests

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/bpf_load.c b/samples/bpf/bpf_load.c
index 816bca5..d168642 100644
--- a/samples/bpf/bpf_load.c
+++ b/samples/bpf/bpf_load.c
@@ -346,3 +346,65 @@ void read_trace_pipe(void)
 		}
 	}
 }
+
+#define MAX_SYMS 300000
+static struct ksym syms[MAX_SYMS];
+static int sym_cnt;
+
+static int ksym_cmp(const void *p1, const void *p2)
+{
+	return ((struct ksym *)p1)->addr - ((struct ksym *)p2)->addr;
+}
+
+int load_kallsyms(void)
+{
+	FILE *f = fopen("/proc/kallsyms", "r");
+	char func[256], buf[256];
+	char symbol;
+	void *addr;
+	int i = 0;
+
+	if (!f)
+		return -ENOENT;
+
+	while (!feof(f)) {
+		if (!fgets(buf, sizeof(buf), f))
+			break;
+		if (sscanf(buf, "%p %c %s", &addr, &symbol, func) != 3)
+			break;
+		if (!addr)
+			continue;
+		syms[i].addr = (long) addr;
+		syms[i].name = strdup(func);
+		i++;
+	}
+	sym_cnt = i;
+	qsort(syms, sym_cnt, sizeof(struct ksym), ksym_cmp);
+	return 0;
+}
+
+struct ksym *ksym_search(long key)
+{
+	int start = 0, end = sym_cnt;
+	int result;
+
+	while (start < end) {
+		size_t mid = start + (end - start) / 2;
+
+		result = key - syms[mid].addr;
+		if (result < 0)
+			end = mid;
+		else if (result > 0)
+			start = mid + 1;
+		else
+			return &syms[mid];
+	}
+
+	if (start >= 1 && syms[start - 1].addr < key &&
+	    key < syms[start].addr)
+		/* valid ksym */
+		return &syms[start - 1];
+
+	/* out of range. return _stext */
+	return &syms[0];
+}
diff --git a/samples/bpf/bpf_load.h b/samples/bpf/bpf_load.h
index cbd7c2b..dfa57fe 100644
--- a/samples/bpf/bpf_load.h
+++ b/samples/bpf/bpf_load.h
@@ -23,5 +23,11 @@ extern int event_fd[MAX_PROGS];
 int load_bpf_file(char *path);
 
 void read_trace_pipe(void);
+struct ksym {
+	long addr;
+	char *name;
+};
 
+int load_kallsyms(void);
+struct ksym *ksym_search(long key);
 #endif
diff --git a/samples/bpf/offwaketime_user.c b/samples/bpf/offwaketime_user.c
index 17cf302..6f002a9 100644
--- a/samples/bpf/offwaketime_user.c
+++ b/samples/bpf/offwaketime_user.c
@@ -18,80 +18,15 @@
 #include "libbpf.h"
 #include "bpf_load.h"
 
-#define MAX_SYMS 300000
 #define PRINT_RAW_ADDR 0
 
-static struct ksym {
-	long addr;
-	char *name;
-} syms[MAX_SYMS];
-static int sym_cnt;
-
-static int ksym_cmp(const void *p1, const void *p2)
-{
-	return ((struct ksym *)p1)->addr - ((struct ksym *)p2)->addr;
-}
-
-static int load_kallsyms(void)
-{
-	FILE *f = fopen("/proc/kallsyms", "r");
-	char func[256], buf[256];
-	char symbol;
-	void *addr;
-	int i = 0;
-
-	if (!f)
-		return -ENOENT;
-
-	while (!feof(f)) {
-		if (!fgets(buf, sizeof(buf), f))
-			break;
-		if (sscanf(buf, "%p %c %s", &addr, &symbol, func) != 3)
-			break;
-		if (!addr)
-			continue;
-		syms[i].addr = (long) addr;
-		syms[i].name = strdup(func);
-		i++;
-	}
-	sym_cnt = i;
-	qsort(syms, sym_cnt, sizeof(struct ksym), ksym_cmp);
-	return 0;
-}
-
-static void *search(long key)
-{
-	int start = 0, end = sym_cnt;
-	int result;
-
-	while (start < end) {
-		size_t mid = start + (end - start) / 2;
-
-		result = key - syms[mid].addr;
-		if (result < 0)
-			end = mid;
-		else if (result > 0)
-			start = mid + 1;
-		else
-			return &syms[mid];
-	}
-
-	if (start >= 1 && syms[start - 1].addr < key &&
-	    key < syms[start].addr)
-		/* valid ksym */
-		return &syms[start - 1];
-
-	/* out of range. return _stext */
-	return &syms[0];
-}
-
 static void print_ksym(__u64 addr)
 {
 	struct ksym *sym;
 
 	if (!addr)
 		return;
-	sym = search(addr);
+	sym = ksym_search(addr);
 	if (PRINT_RAW_ADDR)
 		printf("%s/%llx;", sym->name, addr);
 	else
-- 
cgit v0.10.2


From 89b976070190eb9dd14943c0d6ca4b7209f61405 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:20 -0800
Subject: samples/bpf: add map_flags to bpf loader

note old loader is compatible with new kernel.
map_flags are optional

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h
index 811bcca..9363500 100644
--- a/samples/bpf/bpf_helpers.h
+++ b/samples/bpf/bpf_helpers.h
@@ -61,6 +61,7 @@ struct bpf_map_def {
 	unsigned int key_size;
 	unsigned int value_size;
 	unsigned int max_entries;
+	unsigned int map_flags;
 };
 
 static int (*bpf_skb_store_bytes)(void *ctx, int off, void *from, int len, int flags) =
diff --git a/samples/bpf/bpf_load.c b/samples/bpf/bpf_load.c
index d168642..58f86bd 100644
--- a/samples/bpf/bpf_load.c
+++ b/samples/bpf/bpf_load.c
@@ -157,7 +157,8 @@ static int load_maps(struct bpf_map_def *maps, int len)
 		map_fd[i] = bpf_create_map(maps[i].type,
 					   maps[i].key_size,
 					   maps[i].value_size,
-					   maps[i].max_entries);
+					   maps[i].max_entries,
+					   maps[i].map_flags);
 		if (map_fd[i] < 0) {
 			printf("failed to create a map: %d %s\n",
 			       errno, strerror(errno));
diff --git a/samples/bpf/fds_example.c b/samples/bpf/fds_example.c
index e2fd16c..625e797 100644
--- a/samples/bpf/fds_example.c
+++ b/samples/bpf/fds_example.c
@@ -44,7 +44,7 @@ static void usage(void)
 static int bpf_map_create(void)
 {
 	return bpf_create_map(BPF_MAP_TYPE_ARRAY, sizeof(uint32_t),
-			      sizeof(uint32_t), 1024);
+			      sizeof(uint32_t), 1024, 0);
 }
 
 static int bpf_prog_create(const char *object)
diff --git a/samples/bpf/libbpf.c b/samples/bpf/libbpf.c
index 65a8d48..9969e35 100644
--- a/samples/bpf/libbpf.c
+++ b/samples/bpf/libbpf.c
@@ -19,13 +19,14 @@ static __u64 ptr_to_u64(void *ptr)
 }
 
 int bpf_create_map(enum bpf_map_type map_type, int key_size, int value_size,
-		   int max_entries)
+		   int max_entries, int map_flags)
 {
 	union bpf_attr attr = {
 		.map_type = map_type,
 		.key_size = key_size,
 		.value_size = value_size,
-		.max_entries = max_entries
+		.max_entries = max_entries,
+		.map_flags = map_flags,
 	};
 
 	return syscall(__NR_bpf, BPF_MAP_CREATE, &attr, sizeof(attr));
diff --git a/samples/bpf/libbpf.h b/samples/bpf/libbpf.h
index 014aacf..364582b 100644
--- a/samples/bpf/libbpf.h
+++ b/samples/bpf/libbpf.h
@@ -5,7 +5,7 @@
 struct bpf_insn;
 
 int bpf_create_map(enum bpf_map_type map_type, int key_size, int value_size,
-		   int max_entries);
+		   int max_entries, int map_flags);
 int bpf_update_elem(int fd, void *key, void *value, unsigned long long flags);
 int bpf_lookup_elem(int fd, void *key, void *value);
 int bpf_delete_elem(int fd, void *key);
diff --git a/samples/bpf/sock_example.c b/samples/bpf/sock_example.c
index a0ce251..28b60ba 100644
--- a/samples/bpf/sock_example.c
+++ b/samples/bpf/sock_example.c
@@ -34,7 +34,7 @@ static int test_sock(void)
 	long long value = 0, tcp_cnt, udp_cnt, icmp_cnt;
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, sizeof(key), sizeof(value),
-				256);
+				256, 0);
 	if (map_fd < 0) {
 		printf("failed to create map '%s'\n", strerror(errno));
 		goto cleanup;
diff --git a/samples/bpf/test_maps.c b/samples/bpf/test_maps.c
index ad466ed..7bd9edd 100644
--- a/samples/bpf/test_maps.c
+++ b/samples/bpf/test_maps.c
@@ -2,6 +2,7 @@
  * Testsuite for eBPF maps
  *
  * Copyright (c) 2014 PLUMgrid, http://plumgrid.com
+ * Copyright (c) 2016 Facebook
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of version 2 of the GNU General Public
@@ -17,13 +18,16 @@
 #include <stdlib.h>
 #include "libbpf.h"
 
+static int map_flags;
+
 /* sanity tests for map API */
 static void test_hashmap_sanity(int i, void *data)
 {
 	long long key, next_key, value;
 	int map_fd;
 
-	map_fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(key), sizeof(value), 2);
+	map_fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(key), sizeof(value),
+				2, map_flags);
 	if (map_fd < 0) {
 		printf("failed to create hashmap '%s'\n", strerror(errno));
 		exit(1);
@@ -99,7 +103,7 @@ static void test_percpu_hashmap_sanity(int task, void *data)
 	int map_fd, i;
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_PERCPU_HASH, sizeof(key),
-				sizeof(value[0]), 2);
+				sizeof(value[0]), 2, map_flags);
 	if (map_fd < 0) {
 		printf("failed to create hashmap '%s'\n", strerror(errno));
 		exit(1);
@@ -188,7 +192,8 @@ static void test_arraymap_sanity(int i, void *data)
 	int key, next_key, map_fd;
 	long long value;
 
-	map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, sizeof(key), sizeof(value), 2);
+	map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, sizeof(key), sizeof(value),
+				2, 0);
 	if (map_fd < 0) {
 		printf("failed to create arraymap '%s'\n", strerror(errno));
 		exit(1);
@@ -244,7 +249,7 @@ static void test_percpu_arraymap_many_keys(void)
 	int key, map_fd, i;
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_PERCPU_ARRAY, sizeof(key),
-				sizeof(values[0]), nr_keys);
+				sizeof(values[0]), nr_keys, 0);
 	if (map_fd < 0) {
 		printf("failed to create per-cpu arraymap '%s'\n",
 		       strerror(errno));
@@ -275,7 +280,7 @@ static void test_percpu_arraymap_sanity(int i, void *data)
 	int key, next_key, map_fd;
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_PERCPU_ARRAY, sizeof(key),
-				sizeof(values[0]), 2);
+				sizeof(values[0]), 2, 0);
 	if (map_fd < 0) {
 		printf("failed to create arraymap '%s'\n", strerror(errno));
 		exit(1);
@@ -336,7 +341,7 @@ static void test_map_large(void)
 
 	/* allocate 4Mbyte of memory */
 	map_fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(key), sizeof(value),
-				MAP_SIZE);
+				MAP_SIZE, map_flags);
 	if (map_fd < 0) {
 		printf("failed to create large map '%s'\n", strerror(errno));
 		exit(1);
@@ -421,7 +426,7 @@ static void test_map_parallel(void)
 	int data[2];
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(key), sizeof(value),
-				MAP_SIZE);
+				MAP_SIZE, map_flags);
 	if (map_fd < 0) {
 		printf("failed to create map for parallel test '%s'\n",
 		       strerror(errno));
diff --git a/samples/bpf/test_verifier.c b/samples/bpf/test_verifier.c
index 563c507..4b51a90 100644
--- a/samples/bpf/test_verifier.c
+++ b/samples/bpf/test_verifier.c
@@ -1198,7 +1198,7 @@ static int create_map(void)
 	int map_fd;
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_HASH,
-				sizeof(long long), sizeof(long long), 1024);
+				sizeof(long long), sizeof(long long), 1024, 0);
 	if (map_fd < 0)
 		printf("failed to create map '%s'\n", strerror(errno));
 
@@ -1210,7 +1210,7 @@ static int create_prog_array(void)
 	int map_fd;
 
 	map_fd = bpf_create_map(BPF_MAP_TYPE_PROG_ARRAY,
-				sizeof(int), sizeof(int), 4);
+				sizeof(int), sizeof(int), 4, 0);
 	if (map_fd < 0)
 		printf("failed to create prog_array '%s'\n", strerror(errno));
 
-- 
cgit v0.10.2


From c3f85cffc50d2f259903555979581a632b945ec2 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Mon, 7 Mar 2016 21:57:21 -0800
Subject: samples/bpf: test both pre-alloc and normal maps

extend test coveraged to include pre-allocated and run-time alloc maps

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/test_maps.c b/samples/bpf/test_maps.c
index 7bd9edd..47bf085 100644
--- a/samples/bpf/test_maps.c
+++ b/samples/bpf/test_maps.c
@@ -468,7 +468,7 @@ static void test_map_parallel(void)
 	assert(bpf_get_next_key(map_fd, &key, &key) == -1 && errno == ENOENT);
 }
 
-int main(void)
+static void run_all_tests(void)
 {
 	test_hashmap_sanity(0, NULL);
 	test_percpu_hashmap_sanity(0, NULL);
@@ -479,6 +479,14 @@ int main(void)
 	test_map_large();
 	test_map_parallel();
 	test_map_stress();
+}
+
+int main(void)
+{
+	map_flags = 0;
+	run_all_tests();
+	map_flags = BPF_F_NO_PREALLOC;
+	run_all_tests();
 	printf("test_maps: OK\n");
 	return 0;
 }
-- 
cgit v0.10.2


From e28e87ed474c5a0b378c66fb85efc8e487f4f63f Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Tue, 8 Mar 2016 23:36:03 +0100
Subject: ip_tunnel, bpf: ip_tunnel_info_opts_{get, set} depends on CONFIG_INET

Helpers like ip_tunnel_info_opts_{get,set}() are only available if
CONFIG_INET is set, thus add an empty definition into the header for
the !CONFIG_INET case, where already other empty inline helpers are
defined.

This avoids ifdef kludge inside filter.c, but also vxlan and geneve
themself where this facility can only be used with, depend on INET
being set. For the !INET case TUNNEL_OPTIONS_PRESENT would never be
set in flags.

Fixes: 14ca0751c96f ("bpf: support for access to tunnel options")
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index e1395d7..0acd80f 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -369,6 +369,17 @@ static inline void ip_tunnel_unneed_metadata(void)
 {
 }
 
+static inline void ip_tunnel_info_opts_get(void *to,
+					   const struct ip_tunnel_info *info)
+{
+}
+
+static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info,
+					   const void *from, int len)
+{
+	info->options_len = 0;
+}
+
 #endif /* CONFIG_INET */
 
 #endif /* __NET_IP_TUNNELS_H */
-- 
cgit v0.10.2


From 9d8b612d88e81c2cf7dbc0d4a827da6ca0d848e0 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Tue, 8 Mar 2016 15:07:52 -0800
Subject: samples/bpf: add bpf map stress test

this test calls bpf programs from different contexts:
from inside of slub, from rcu, from pretty much everywhere,
since it kprobes all spin_lock functions.
It stresses the bpf hash and percpu map pre-allocation,
deallocation logic and call_rcu mechanisms.
User space part adding more stress by walking and deleting map elements.

Note that due to nature bpf_load.c the earlier kprobe+bpf programs are
already active while loader loads new programs, creates new kprobes and
attaches them.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index c4f8ae0..75a13e7 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -17,6 +17,7 @@ hostprogs-y += tracex6
 hostprogs-y += trace_output
 hostprogs-y += lathist
 hostprogs-y += offwaketime
+hostprogs-y += spintest
 
 test_verifier-objs := test_verifier.o libbpf.o
 test_maps-objs := test_maps.o libbpf.o
@@ -34,6 +35,7 @@ tracex6-objs := bpf_load.o libbpf.o tracex6_user.o
 trace_output-objs := bpf_load.o libbpf.o trace_output_user.o
 lathist-objs := bpf_load.o libbpf.o lathist_user.o
 offwaketime-objs := bpf_load.o libbpf.o offwaketime_user.o
+spintest-objs := bpf_load.o libbpf.o spintest_user.o
 
 # Tell kbuild to always build the programs
 always := $(hostprogs-y)
@@ -50,6 +52,7 @@ always += trace_output_kern.o
 always += tcbpf1_kern.o
 always += lathist_kern.o
 always += offwaketime_kern.o
+always += spintest_kern.o
 
 HOSTCFLAGS += -I$(objtree)/usr/include
 
@@ -67,6 +70,7 @@ HOSTLOADLIBES_tracex6 += -lelf
 HOSTLOADLIBES_trace_output += -lelf -lrt
 HOSTLOADLIBES_lathist += -lelf
 HOSTLOADLIBES_offwaketime += -lelf
+HOSTLOADLIBES_spintest += -lelf
 
 # point this to your LLVM backend with bpf support
 LLC=$(srctree)/tools/bpf/llvm/bld/Debug+Asserts/bin/llc
diff --git a/samples/bpf/spintest_kern.c b/samples/bpf/spintest_kern.c
new file mode 100644
index 0000000..ef8ac33
--- /dev/null
+++ b/samples/bpf/spintest_kern.c
@@ -0,0 +1,59 @@
+/* Copyright (c) 2016, Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/version.h>
+#include <uapi/linux/bpf.h>
+#include "bpf_helpers.h"
+
+struct bpf_map_def SEC("maps") my_map = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(long),
+	.value_size = sizeof(long),
+	.max_entries = 1024,
+};
+struct bpf_map_def SEC("maps") my_map2 = {
+	.type = BPF_MAP_TYPE_PERCPU_HASH,
+	.key_size = sizeof(long),
+	.value_size = sizeof(long),
+	.max_entries = 1024,
+};
+
+#define PROG(foo) \
+int foo(struct pt_regs *ctx) \
+{ \
+	long v = ctx->ip, *val; \
+\
+	val = bpf_map_lookup_elem(&my_map, &v); \
+	bpf_map_update_elem(&my_map, &v, &v, BPF_ANY); \
+	bpf_map_update_elem(&my_map2, &v, &v, BPF_ANY); \
+	bpf_map_delete_elem(&my_map2, &v); \
+	return 0; \
+}
+
+/* add kprobes to all possible *spin* functions */
+SEC("kprobe/spin_unlock")PROG(p1)
+SEC("kprobe/spin_lock")PROG(p2)
+SEC("kprobe/mutex_spin_on_owner")PROG(p3)
+SEC("kprobe/rwsem_spin_on_owner")PROG(p4)
+SEC("kprobe/spin_unlock_irqrestore")PROG(p5)
+SEC("kprobe/_raw_spin_unlock_irqrestore")PROG(p6)
+SEC("kprobe/_raw_spin_unlock_bh")PROG(p7)
+SEC("kprobe/_raw_spin_unlock")PROG(p8)
+SEC("kprobe/_raw_spin_lock_irqsave")PROG(p9)
+SEC("kprobe/_raw_spin_trylock_bh")PROG(p10)
+SEC("kprobe/_raw_spin_lock_irq")PROG(p11)
+SEC("kprobe/_raw_spin_trylock")PROG(p12)
+SEC("kprobe/_raw_spin_lock")PROG(p13)
+SEC("kprobe/_raw_spin_lock_bh")PROG(p14)
+/* and to inner bpf helpers */
+SEC("kprobe/htab_map_update_elem")PROG(p15)
+SEC("kprobe/__htab_percpu_map_update_elem")PROG(p16)
+SEC("kprobe/htab_map_alloc")PROG(p17)
+
+char _license[] SEC("license") = "GPL";
+u32 _version SEC("version") = LINUX_VERSION_CODE;
diff --git a/samples/bpf/spintest_user.c b/samples/bpf/spintest_user.c
new file mode 100644
index 0000000..311ede5
--- /dev/null
+++ b/samples/bpf/spintest_user.c
@@ -0,0 +1,50 @@
+#include <stdio.h>
+#include <unistd.h>
+#include <linux/bpf.h>
+#include <string.h>
+#include <assert.h>
+#include <sys/resource.h>
+#include "libbpf.h"
+#include "bpf_load.h"
+
+int main(int ac, char **argv)
+{
+	struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+	long key, next_key, value;
+	char filename[256];
+	struct ksym *sym;
+	int i;
+
+	snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+	setrlimit(RLIMIT_MEMLOCK, &r);
+
+	if (load_kallsyms()) {
+		printf("failed to process /proc/kallsyms\n");
+		return 2;
+	}
+
+	if (load_bpf_file(filename)) {
+		printf("%s", bpf_log_buf);
+		return 1;
+	}
+
+	for (i = 0; i < 5; i++) {
+		key = 0;
+		printf("kprobing funcs:");
+		while (bpf_get_next_key(map_fd[0], &key, &next_key) == 0) {
+			bpf_lookup_elem(map_fd[0], &next_key, &value);
+			assert(next_key == value);
+			sym = ksym_search(value);
+			printf(" %s", sym->name);
+			key = next_key;
+		}
+		if (key)
+			printf("\n");
+		key = 0;
+		while (bpf_get_next_key(map_fd[0], &key, &next_key) == 0)
+			bpf_delete_elem(map_fd[0], &next_key);
+		sleep(1);
+	}
+
+	return 0;
+}
-- 
cgit v0.10.2


From 7dcc42b685b4acf7b2f3f08d157f1574e1bec7f7 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Tue, 8 Mar 2016 15:07:53 -0800
Subject: samples/bpf: stress test bpf_get_stackid

increase stress by also calling bpf_get_stackid() from
various *spin* functions

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/spintest_kern.c b/samples/bpf/spintest_kern.c
index ef8ac33..4b27619 100644
--- a/samples/bpf/spintest_kern.c
+++ b/samples/bpf/spintest_kern.c
@@ -8,6 +8,7 @@
 #include <linux/netdevice.h>
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
+#include <uapi/linux/perf_event.h>
 #include "bpf_helpers.h"
 
 struct bpf_map_def SEC("maps") my_map = {
@@ -23,6 +24,13 @@ struct bpf_map_def SEC("maps") my_map2 = {
 	.max_entries = 1024,
 };
 
+struct bpf_map_def SEC("maps") stackmap = {
+	.type = BPF_MAP_TYPE_STACK_TRACE,
+	.key_size = sizeof(u32),
+	.value_size = PERF_MAX_STACK_DEPTH * sizeof(u64),
+	.max_entries = 10000,
+};
+
 #define PROG(foo) \
 int foo(struct pt_regs *ctx) \
 { \
@@ -32,6 +40,7 @@ int foo(struct pt_regs *ctx) \
 	bpf_map_update_elem(&my_map, &v, &v, BPF_ANY); \
 	bpf_map_update_elem(&my_map2, &v, &v, BPF_ANY); \
 	bpf_map_delete_elem(&my_map2, &v); \
+	bpf_get_stackid(ctx, &stackmap, BPF_F_REUSE_STACKID); \
 	return 0; \
 }
 
-- 
cgit v0.10.2


From 26e9093110fb9ceb10093e4914b129b58d49a425 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Tue, 8 Mar 2016 15:07:54 -0800
Subject: samples/bpf: add map performance test

performance tests for hash map and per-cpu hash map
with and without pre-allocation

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 75a13e7..502c9fc 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -18,6 +18,7 @@ hostprogs-y += trace_output
 hostprogs-y += lathist
 hostprogs-y += offwaketime
 hostprogs-y += spintest
+hostprogs-y += map_perf_test
 
 test_verifier-objs := test_verifier.o libbpf.o
 test_maps-objs := test_maps.o libbpf.o
@@ -36,6 +37,7 @@ trace_output-objs := bpf_load.o libbpf.o trace_output_user.o
 lathist-objs := bpf_load.o libbpf.o lathist_user.o
 offwaketime-objs := bpf_load.o libbpf.o offwaketime_user.o
 spintest-objs := bpf_load.o libbpf.o spintest_user.o
+map_perf_test-objs := bpf_load.o libbpf.o map_perf_test_user.o
 
 # Tell kbuild to always build the programs
 always := $(hostprogs-y)
@@ -53,6 +55,7 @@ always += tcbpf1_kern.o
 always += lathist_kern.o
 always += offwaketime_kern.o
 always += spintest_kern.o
+always += map_perf_test_kern.o
 
 HOSTCFLAGS += -I$(objtree)/usr/include
 
@@ -71,6 +74,7 @@ HOSTLOADLIBES_trace_output += -lelf -lrt
 HOSTLOADLIBES_lathist += -lelf
 HOSTLOADLIBES_offwaketime += -lelf
 HOSTLOADLIBES_spintest += -lelf
+HOSTLOADLIBES_map_perf_test += -lelf -lrt
 
 # point this to your LLVM backend with bpf support
 LLC=$(srctree)/tools/bpf/llvm/bld/Debug+Asserts/bin/llc
diff --git a/samples/bpf/map_perf_test_kern.c b/samples/bpf/map_perf_test_kern.c
new file mode 100644
index 0000000..311538e
--- /dev/null
+++ b/samples/bpf/map_perf_test_kern.c
@@ -0,0 +1,100 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/version.h>
+#include <uapi/linux/bpf.h>
+#include "bpf_helpers.h"
+
+#define MAX_ENTRIES 1000
+
+struct bpf_map_def SEC("maps") hash_map = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(u32),
+	.value_size = sizeof(long),
+	.max_entries = MAX_ENTRIES,
+};
+
+struct bpf_map_def SEC("maps") percpu_hash_map = {
+	.type = BPF_MAP_TYPE_PERCPU_HASH,
+	.key_size = sizeof(u32),
+	.value_size = sizeof(long),
+	.max_entries = MAX_ENTRIES,
+};
+
+struct bpf_map_def SEC("maps") hash_map_alloc = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(u32),
+	.value_size = sizeof(long),
+	.max_entries = MAX_ENTRIES,
+	.map_flags = BPF_F_NO_PREALLOC,
+};
+
+struct bpf_map_def SEC("maps") percpu_hash_map_alloc = {
+	.type = BPF_MAP_TYPE_PERCPU_HASH,
+	.key_size = sizeof(u32),
+	.value_size = sizeof(long),
+	.max_entries = MAX_ENTRIES,
+	.map_flags = BPF_F_NO_PREALLOC,
+};
+
+SEC("kprobe/sys_getuid")
+int stress_hmap(struct pt_regs *ctx)
+{
+	u32 key = bpf_get_current_pid_tgid();
+	long init_val = 1;
+	long *value;
+
+	bpf_map_update_elem(&hash_map, &key, &init_val, BPF_ANY);
+	value = bpf_map_lookup_elem(&hash_map, &key);
+	if (value)
+		bpf_map_delete_elem(&hash_map, &key);
+	return 0;
+}
+
+SEC("kprobe/sys_geteuid")
+int stress_percpu_hmap(struct pt_regs *ctx)
+{
+	u32 key = bpf_get_current_pid_tgid();
+	long init_val = 1;
+	long *value;
+
+	bpf_map_update_elem(&percpu_hash_map, &key, &init_val, BPF_ANY);
+	value = bpf_map_lookup_elem(&percpu_hash_map, &key);
+	if (value)
+		bpf_map_delete_elem(&percpu_hash_map, &key);
+	return 0;
+}
+SEC("kprobe/sys_getgid")
+int stress_hmap_alloc(struct pt_regs *ctx)
+{
+	u32 key = bpf_get_current_pid_tgid();
+	long init_val = 1;
+	long *value;
+
+	bpf_map_update_elem(&hash_map_alloc, &key, &init_val, BPF_ANY);
+	value = bpf_map_lookup_elem(&hash_map_alloc, &key);
+	if (value)
+		bpf_map_delete_elem(&hash_map_alloc, &key);
+	return 0;
+}
+
+SEC("kprobe/sys_getegid")
+int stress_percpu_hmap_alloc(struct pt_regs *ctx)
+{
+	u32 key = bpf_get_current_pid_tgid();
+	long init_val = 1;
+	long *value;
+
+	bpf_map_update_elem(&percpu_hash_map_alloc, &key, &init_val, BPF_ANY);
+	value = bpf_map_lookup_elem(&percpu_hash_map_alloc, &key);
+	if (value)
+		bpf_map_delete_elem(&percpu_hash_map_alloc, &key);
+	return 0;
+}
+char _license[] SEC("license") = "GPL";
+u32 _version SEC("version") = LINUX_VERSION_CODE;
diff --git a/samples/bpf/map_perf_test_user.c b/samples/bpf/map_perf_test_user.c
new file mode 100644
index 0000000..95af56e
--- /dev/null
+++ b/samples/bpf/map_perf_test_user.c
@@ -0,0 +1,155 @@
+/* Copyright (c) 2016 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#define _GNU_SOURCE
+#include <sched.h>
+#include <stdio.h>
+#include <sys/types.h>
+#include <asm/unistd.h>
+#include <unistd.h>
+#include <assert.h>
+#include <sys/wait.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <linux/bpf.h>
+#include <string.h>
+#include <time.h>
+#include "libbpf.h"
+#include "bpf_load.h"
+
+#define MAX_CNT 1000000
+
+static __u64 time_get_ns(void)
+{
+	struct timespec ts;
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+	return ts.tv_sec * 1000000000ull + ts.tv_nsec;
+}
+
+#define HASH_PREALLOC		(1 << 0)
+#define PERCPU_HASH_PREALLOC	(1 << 1)
+#define HASH_KMALLOC		(1 << 2)
+#define PERCPU_HASH_KMALLOC	(1 << 3)
+
+static int test_flags = ~0;
+
+static void test_hash_prealloc(int cpu)
+{
+	__u64 start_time;
+	int i;
+
+	start_time = time_get_ns();
+	for (i = 0; i < MAX_CNT; i++)
+		syscall(__NR_getuid);
+	printf("%d:hash_map_perf pre-alloc %lld events per sec\n",
+	       cpu, MAX_CNT * 1000000000ll / (time_get_ns() - start_time));
+}
+
+static void test_percpu_hash_prealloc(int cpu)
+{
+	__u64 start_time;
+	int i;
+
+	start_time = time_get_ns();
+	for (i = 0; i < MAX_CNT; i++)
+		syscall(__NR_geteuid);
+	printf("%d:percpu_hash_map_perf pre-alloc %lld events per sec\n",
+	       cpu, MAX_CNT * 1000000000ll / (time_get_ns() - start_time));
+}
+
+static void test_hash_kmalloc(int cpu)
+{
+	__u64 start_time;
+	int i;
+
+	start_time = time_get_ns();
+	for (i = 0; i < MAX_CNT; i++)
+		syscall(__NR_getgid);
+	printf("%d:hash_map_perf kmalloc %lld events per sec\n",
+	       cpu, MAX_CNT * 1000000000ll / (time_get_ns() - start_time));
+}
+
+static void test_percpu_hash_kmalloc(int cpu)
+{
+	__u64 start_time;
+	int i;
+
+	start_time = time_get_ns();
+	for (i = 0; i < MAX_CNT; i++)
+		syscall(__NR_getegid);
+	printf("%d:percpu_hash_map_perf kmalloc %lld events per sec\n",
+	       cpu, MAX_CNT * 1000000000ll / (time_get_ns() - start_time));
+}
+
+static void loop(int cpu)
+{
+	cpu_set_t cpuset;
+
+	CPU_ZERO(&cpuset);
+	CPU_SET(cpu, &cpuset);
+	sched_setaffinity(0, sizeof(cpuset), &cpuset);
+
+	if (test_flags & HASH_PREALLOC)
+		test_hash_prealloc(cpu);
+
+	if (test_flags & PERCPU_HASH_PREALLOC)
+		test_percpu_hash_prealloc(cpu);
+
+	if (test_flags & HASH_KMALLOC)
+		test_hash_kmalloc(cpu);
+
+	if (test_flags & PERCPU_HASH_KMALLOC)
+		test_percpu_hash_kmalloc(cpu);
+}
+
+static void run_perf_test(int tasks)
+{
+	pid_t pid[tasks];
+	int i;
+
+	for (i = 0; i < tasks; i++) {
+		pid[i] = fork();
+		if (pid[i] == 0) {
+			loop(i);
+			exit(0);
+		} else if (pid[i] == -1) {
+			printf("couldn't spawn #%d process\n", i);
+			exit(1);
+		}
+	}
+	for (i = 0; i < tasks; i++) {
+		int status;
+
+		assert(waitpid(pid[i], &status, 0) == pid[i]);
+		assert(status == 0);
+	}
+}
+
+int main(int argc, char **argv)
+{
+	struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+	char filename[256];
+	int num_cpu = 8;
+
+	snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+	setrlimit(RLIMIT_MEMLOCK, &r);
+
+	if (argc > 1)
+		test_flags = atoi(argv[1]) ? : test_flags;
+
+	if (argc > 2)
+		num_cpu = atoi(argv[2]) ? : num_cpu;
+
+	if (load_bpf_file(filename)) {
+		printf("%s", bpf_log_buf);
+		return 1;
+	}
+
+	run_perf_test(num_cpu);
+
+	return 0;
+}
-- 
cgit v0.10.2


From 00f481bd895a826058d301b3093e86e819497b51 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Wed, 24 Feb 2016 12:19:22 +0200
Subject: iwlwifi: mvm: add ctdp operations to debugfs

Add debugfs entries to get the ctdp budget average
and to stop ctdp.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 56e6b0b..a43b392 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -73,6 +73,44 @@
 #include "debugfs.h"
 #include "iwl-fw-error-dump.h"
 
+static ssize_t iwl_dbgfs_ctdp_budget_read(struct file *file,
+					  char __user *user_buf,
+					  size_t count, loff_t *ppos)
+{
+	struct iwl_mvm *mvm = file->private_data;
+	char buf[16];
+	int pos, budget;
+
+	if (!mvm->ucode_loaded || mvm->cur_ucode != IWL_UCODE_REGULAR)
+		return -EIO;
+
+	mutex_lock(&mvm->mutex);
+	budget = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_REPORT, 0);
+	mutex_unlock(&mvm->mutex);
+
+	if (budget < 0)
+		return budget;
+
+	pos = scnprintf(buf, sizeof(buf), "%d\n", budget);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
+}
+
+static ssize_t iwl_dbgfs_stop_ctdp_write(struct iwl_mvm *mvm, char *buf,
+					 size_t count, loff_t *ppos)
+{
+	int ret;
+
+	if (!mvm->ucode_loaded || mvm->cur_ucode != IWL_UCODE_REGULAR)
+		return -EIO;
+
+	mutex_lock(&mvm->mutex);
+	ret = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_STOP, 0);
+	mutex_unlock(&mvm->mutex);
+
+	return ret ?: count;
+}
+
 static ssize_t iwl_dbgfs_tx_flush_write(struct iwl_mvm *mvm, char *buf,
 					size_t count, loff_t *ppos)
 {
@@ -1493,6 +1531,8 @@ iwl_dbgfs_send_echo_cmd_write(struct iwl_mvm *mvm, char *buf,
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(prph_reg, 64);
 
 /* Device wide debugfs entries */
+MVM_DEBUGFS_READ_FILE_OPS(ctdp_budget);
+MVM_DEBUGFS_WRITE_FILE_OPS(stop_ctdp, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(tx_flush, 16);
 MVM_DEBUGFS_WRITE_FILE_OPS(sta_drain, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(send_echo_cmd, 8);
@@ -1542,6 +1582,8 @@ int iwl_mvm_dbgfs_register(struct iwl_mvm *mvm, struct dentry *dbgfs_dir)
 	MVM_DEBUGFS_ADD_FILE(set_nic_temperature, mvm->debugfs_dir,
 			     S_IWUSR | S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(nic_temp, dbgfs_dir, S_IRUSR);
+	MVM_DEBUGFS_ADD_FILE(ctdp_budget, dbgfs_dir, S_IRUSR);
+	MVM_DEBUGFS_ADD_FILE(stop_ctdp, dbgfs_dir, S_IWUSR);
 	MVM_DEBUGFS_ADD_FILE(stations, dbgfs_dir, S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(bt_notif, dbgfs_dir, S_IRUSR);
 	MVM_DEBUGFS_ADD_FILE(bt_cmd, dbgfs_dir, S_IRUSR);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 0a02e98..ce0c6fd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -510,6 +510,50 @@ static const struct iwl_tt_params iwl_mvm_default_tt_params = {
 	.support_tx_backoff = true,
 };
 
+int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget)
+{
+	struct iwl_mvm_ctdp_cmd cmd = {
+		.operation = cpu_to_le32(op),
+		.budget = cpu_to_le32(budget),
+		.window_size = 0,
+	};
+	int ret;
+	u32 status;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	ret = iwl_mvm_send_cmd_pdu_status(mvm, WIDE_ID(PHY_OPS_GROUP,
+						       CTDP_CONFIG_CMD),
+					  sizeof(cmd), &cmd, &status);
+
+	if (ret) {
+		IWL_ERR(mvm, "cTDP command failed (err=%d)\n", ret);
+		return ret;
+	}
+
+	switch (op) {
+	case CTDP_CMD_OPERATION_START:
+#ifdef CONFIG_THERMAL
+		mvm->cooling_dev.cur_state = budget;
+#endif /* CONFIG_THERMAL */
+		break;
+	case CTDP_CMD_OPERATION_REPORT:
+		IWL_DEBUG_TEMP(mvm, "cTDP avg energy in mWatt = %d\n", status);
+		/* when the function is called with CTDP_CMD_OPERATION_REPORT
+		 * option the function should return the average budget value
+		 * that is received from the FW.
+		 * The budget can't be less or equal to 0, so it's possible
+		 * to distinguish between error values and budgets.
+		 */
+		return status;
+	case CTDP_CMD_OPERATION_STOP:
+		IWL_DEBUG_TEMP(mvm, "cTDP stopped successfully\n");
+		break;
+	}
+
+	return 0;
+}
+
 #ifdef CONFIG_THERMAL
 static int compare_temps(const void *a, const void *b)
 {
@@ -738,40 +782,6 @@ static const u32 iwl_mvm_cdev_budgets[] = {
 	150,	/* cooling state 19 */
 };
 
-int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget)
-{
-	struct iwl_mvm_ctdp_cmd cmd = {
-		.operation = cpu_to_le32(op),
-		.budget = cpu_to_le32(budget),
-		.window_size = 0,
-	};
-	int ret;
-	u32 status;
-
-	lockdep_assert_held(&mvm->mutex);
-
-	ret = iwl_mvm_send_cmd_pdu_status(mvm, WIDE_ID(PHY_OPS_GROUP,
-						       CTDP_CONFIG_CMD),
-					  sizeof(cmd), &cmd, &status);
-
-	if (ret) {
-		IWL_ERR(mvm, "cTDP command failed (err=%d)\n", ret);
-		return ret;
-	}
-
-	/* can happen if the registration failed */
-	if (!mvm->cooling_dev.cdev)
-		return -EINVAL;
-
-	if (op == CTDP_CMD_OPERATION_START)
-		mvm->cooling_dev.cur_state = budget;
-
-	else if (op == CTDP_CMD_OPERATION_REPORT)
-		IWL_DEBUG_TEMP(mvm, "cTDP avg energy in mWatt = %d\n", status);
-
-	return 0;
-}
-
 static int iwl_mvm_tcool_get_max_state(struct thermal_cooling_device *cdev,
 				       unsigned long *state)
 {
-- 
cgit v0.10.2


From 62d23403d4a201db117975d46c1889c79987069e Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Sun, 6 Mar 2016 09:51:29 +0200
Subject: iwlwifi: mvm: turn off AMSDU bit in QoS control for de-aggregated
 AMSDUs

Our hardware de-aggregates AMSDUs but copies the mac header
as it to the de-aggregated MPDUs. We need to turn off the AMSDU
bit in the QoS control ourselves.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index cd6ca37..9a54f2d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -519,6 +519,18 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			rcu_read_unlock();
 			return;
 		}
+
+		/*
+		 * Our hardware de-aggregates AMSDUs but copies the mac header
+		 * as it to the de-aggregated MPDUs. We need to turn off the
+		 * AMSDU bit in the QoS control ourselves.
+		 */
+		if ((desc->mac_flags2 & IWL_RX_MPDU_MFLG2_AMSDU) &&
+		    !WARN_ON(!ieee80211_is_data_qos(hdr->frame_control))) {
+			u8 *qc = ieee80211_get_qos_ctl(hdr);
+
+			*qc &= ~IEEE80211_QOS_CTL_A_MSDU_PRESENT;
+		}
 	}
 
 	/*
-- 
cgit v0.10.2


From 5151ad953c68f2af0b50cf8d074a902ccf001ae0 Mon Sep 17 00:00:00 2001
From: Matti Gottlieb <matti.gottlieb@intel.com>
Date: Wed, 9 Mar 2016 14:46:28 +0200
Subject: iwlwifi: mvm: ROC: cleanup time event info on FW failure

Currently when the FW sends start/stop aux roc time event
notification with an error status, the driver returns an
error value, but does not remove the time event, and does
not notify the stack above that the time event is over.

This causes problems that the stack above assumes we are still
in the middle of a time event, and therefore can block different
events, such as scanning.

On FW failure notification, cleanup the time event parameters and
notify the stack above that the time event is over.

Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index 924dd6a..2c12789 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -371,20 +371,13 @@ static int iwl_mvm_aux_roc_te_handle_notif(struct iwl_mvm *mvm,
 
 	iwl_mvm_te_check_trigger(mvm, notif, te_data);
 
-	if (!le32_to_cpu(notif->status)) {
-		IWL_DEBUG_TE(mvm,
-			     "ERROR: Aux ROC Time Event %s notification failure\n",
-			     (le32_to_cpu(notif->action) &
-			      TE_V2_NOTIF_HOST_EVENT_START) ? "start" : "end");
-		return -EINVAL;
-	}
-
 	IWL_DEBUG_TE(mvm,
-		     "Aux ROC time event notification  - UID = 0x%x action %d\n",
+		     "Aux ROC time event notification  - UID = 0x%x action %d (error = %d)\n",
 		     le32_to_cpu(notif->unique_id),
-		     le32_to_cpu(notif->action));
+		     le32_to_cpu(notif->action), le32_to_cpu(notif->status));
 
-	if (le32_to_cpu(notif->action) == TE_V2_NOTIF_HOST_EVENT_END) {
+	if (!le32_to_cpu(notif->status) ||
+	    le32_to_cpu(notif->action) == TE_V2_NOTIF_HOST_EVENT_END) {
 		/* End TE, notify mac80211 */
 		ieee80211_remain_on_channel_expired(mvm->hw);
 		iwl_mvm_roc_finished(mvm); /* flush aux queue */
-- 
cgit v0.10.2


From c9cb14a64c32f890d5a0649cb7d81dbfece33056 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Thu, 3 Mar 2016 15:35:34 +0200
Subject: iwlwifi: mvm: add support for async rx handler without hold the mutex

When running async rx handler the framework holds the mvm->mutex
before starting the async handler, that might cause a deadlock in case
the handler calls to ops that lock the mutex as well.
Add support for running async rx handler without hold the mutex before
activating the handler.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 4322e8e..5e8ab79 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -205,79 +205,107 @@ static void iwl_mvm_nic_config(struct iwl_op_mode *op_mode)
 				       ~APMG_PS_CTRL_EARLY_PWR_OFF_RESET_DIS);
 }
 
+/**
+ * enum iwl_rx_handler_context context for Rx handler
+ * @RX_HANDLER_SYNC : this means that it will be called in the Rx path
+ *	which can't acquire mvm->mutex.
+ * @RX_HANDLER_ASYNC_LOCKED : If the handler needs to hold mvm->mutex
+ *	(and only in this case!), it should be set as ASYNC. In that case,
+ *	it will be called from a worker with mvm->mutex held.
+ * @RX_HANDLER_ASYNC_UNLOCKED : in case the handler needs to lock the
+ *	mutex itself, it will be called from a worker without mvm->mutex held.
+ */
+enum iwl_rx_handler_context {
+	RX_HANDLER_SYNC,
+	RX_HANDLER_ASYNC_LOCKED,
+	RX_HANDLER_ASYNC_UNLOCKED,
+};
+
+/**
+ * struct iwl_rx_handlers handler for FW notification
+ * @cmd_id: command id
+ * @context: see &iwl_rx_handler_context
+ * @fn: the function is called when notification is received
+ */
 struct iwl_rx_handlers {
 	u16 cmd_id;
-	bool async;
+	enum iwl_rx_handler_context context;
 	void (*fn)(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 };
 
-#define RX_HANDLER(_cmd_id, _fn, _async)	\
-	{ .cmd_id = _cmd_id , .fn = _fn , .async = _async }
-#define RX_HANDLER_GRP(_grp, _cmd, _fn, _async)	\
-	{ .cmd_id = WIDE_ID(_grp, _cmd), .fn = _fn, .async = _async }
+#define RX_HANDLER(_cmd_id, _fn, _context)	\
+	{ .cmd_id = _cmd_id, .fn = _fn, .context = _context }
+#define RX_HANDLER_GRP(_grp, _cmd, _fn, _context)	\
+	{ .cmd_id = WIDE_ID(_grp, _cmd), .fn = _fn, .context = _context }
 
 /*
  * Handlers for fw notifications
  * Convention: RX_HANDLER(CMD_NAME, iwl_mvm_rx_CMD_NAME
  * This list should be in order of frequency for performance purposes.
  *
- * The handler can be SYNC - this means that it will be called in the Rx path
- * which can't acquire mvm->mutex. If the handler needs to hold mvm->mutex (and
- * only in this case!), it should be set as ASYNC. In that case, it will be
- * called from a worker with mvm->mutex held.
+ * The handler can be one from three contexts, see &iwl_rx_handler_context
  */
 static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
-	RX_HANDLER(TX_CMD, iwl_mvm_rx_tx_cmd, false),
-	RX_HANDLER(BA_NOTIF, iwl_mvm_rx_ba_notif, false),
-
-	RX_HANDLER(BT_PROFILE_NOTIFICATION, iwl_mvm_rx_bt_coex_notif, true),
-	RX_HANDLER(BEACON_NOTIFICATION, iwl_mvm_rx_beacon_notif, true),
-	RX_HANDLER(STATISTICS_NOTIFICATION, iwl_mvm_rx_statistics, true),
+	RX_HANDLER(TX_CMD, iwl_mvm_rx_tx_cmd, RX_HANDLER_SYNC),
+	RX_HANDLER(BA_NOTIF, iwl_mvm_rx_ba_notif, RX_HANDLER_SYNC),
+
+	RX_HANDLER(BT_PROFILE_NOTIFICATION, iwl_mvm_rx_bt_coex_notif,
+		   RX_HANDLER_ASYNC_LOCKED),
+	RX_HANDLER(BEACON_NOTIFICATION, iwl_mvm_rx_beacon_notif,
+		   RX_HANDLER_ASYNC_LOCKED),
+	RX_HANDLER(STATISTICS_NOTIFICATION, iwl_mvm_rx_statistics,
+		   RX_HANDLER_ASYNC_LOCKED),
 	RX_HANDLER(ANTENNA_COUPLING_NOTIFICATION,
-		   iwl_mvm_rx_ant_coupling_notif, true),
+		   iwl_mvm_rx_ant_coupling_notif, RX_HANDLER_ASYNC_LOCKED),
 
 	RX_HANDLER(BA_WINDOW_STATUS_NOTIFICATION_ID,
-		   iwl_mvm_window_status_notif, false),
+		   iwl_mvm_window_status_notif, RX_HANDLER_SYNC),
 
-	RX_HANDLER(TIME_EVENT_NOTIFICATION, iwl_mvm_rx_time_event_notif, false),
-	RX_HANDLER(MCC_CHUB_UPDATE_CMD, iwl_mvm_rx_chub_update_mcc, true),
+	RX_HANDLER(TIME_EVENT_NOTIFICATION, iwl_mvm_rx_time_event_notif,
+		   RX_HANDLER_SYNC),
+	RX_HANDLER(MCC_CHUB_UPDATE_CMD, iwl_mvm_rx_chub_update_mcc,
+		   RX_HANDLER_ASYNC_LOCKED),
 
-	RX_HANDLER(EOSP_NOTIFICATION, iwl_mvm_rx_eosp_notif, false),
+	RX_HANDLER(EOSP_NOTIFICATION, iwl_mvm_rx_eosp_notif, RX_HANDLER_SYNC),
 
 	RX_HANDLER(SCAN_ITERATION_COMPLETE,
-		   iwl_mvm_rx_lmac_scan_iter_complete_notif, false),
+		   iwl_mvm_rx_lmac_scan_iter_complete_notif, RX_HANDLER_SYNC),
 	RX_HANDLER(SCAN_OFFLOAD_COMPLETE,
-		   iwl_mvm_rx_lmac_scan_complete_notif, true),
+		   iwl_mvm_rx_lmac_scan_complete_notif,
+		   RX_HANDLER_ASYNC_LOCKED),
 	RX_HANDLER(MATCH_FOUND_NOTIFICATION, iwl_mvm_rx_scan_match_found,
-		   false),
+		   RX_HANDLER_SYNC),
 	RX_HANDLER(SCAN_COMPLETE_UMAC, iwl_mvm_rx_umac_scan_complete_notif,
-		   true),
+		   RX_HANDLER_ASYNC_LOCKED),
 	RX_HANDLER(SCAN_ITERATION_COMPLETE_UMAC,
-		   iwl_mvm_rx_umac_scan_iter_complete_notif, false),
+		   iwl_mvm_rx_umac_scan_iter_complete_notif, RX_HANDLER_SYNC),
 
-	RX_HANDLER(CARD_STATE_NOTIFICATION, iwl_mvm_rx_card_state_notif, false),
+	RX_HANDLER(CARD_STATE_NOTIFICATION, iwl_mvm_rx_card_state_notif,
+		   RX_HANDLER_SYNC),
 
 	RX_HANDLER(MISSED_BEACONS_NOTIFICATION, iwl_mvm_rx_missed_beacons_notif,
-		   false),
+		   RX_HANDLER_SYNC),
 
-	RX_HANDLER(REPLY_ERROR, iwl_mvm_rx_fw_error, false),
+	RX_HANDLER(REPLY_ERROR, iwl_mvm_rx_fw_error, RX_HANDLER_SYNC),
 	RX_HANDLER(PSM_UAPSD_AP_MISBEHAVING_NOTIFICATION,
-		   iwl_mvm_power_uapsd_misbehaving_ap_notif, false),
-	RX_HANDLER(DTS_MEASUREMENT_NOTIFICATION, iwl_mvm_temp_notif, true),
+		   iwl_mvm_power_uapsd_misbehaving_ap_notif, RX_HANDLER_SYNC),
+	RX_HANDLER(DTS_MEASUREMENT_NOTIFICATION, iwl_mvm_temp_notif,
+		   RX_HANDLER_ASYNC_LOCKED),
 	RX_HANDLER_GRP(PHY_OPS_GROUP, DTS_MEASUREMENT_NOTIF_WIDE,
-		       iwl_mvm_temp_notif, true),
+		       iwl_mvm_temp_notif, RX_HANDLER_ASYNC_LOCKED),
 	RX_HANDLER_GRP(PHY_OPS_GROUP, CT_KILL_NOTIFICATION,
-		       iwl_mvm_ct_kill_notif, false),
+		       iwl_mvm_ct_kill_notif, RX_HANDLER_SYNC),
 
 	RX_HANDLER(TDLS_CHANNEL_SWITCH_NOTIFICATION, iwl_mvm_rx_tdls_notif,
-		   true),
-	RX_HANDLER(MFUART_LOAD_NOTIFICATION, iwl_mvm_rx_mfuart_notif, false),
-	RX_HANDLER(TOF_NOTIFICATION, iwl_mvm_tof_resp_handler, true),
+		   RX_HANDLER_ASYNC_LOCKED),
+	RX_HANDLER(MFUART_LOAD_NOTIFICATION, iwl_mvm_rx_mfuart_notif,
+		   RX_HANDLER_SYNC),
+	RX_HANDLER(TOF_NOTIFICATION, iwl_mvm_tof_resp_handler,
+		   RX_HANDLER_ASYNC_LOCKED),
 	RX_HANDLER_GRP(PROT_OFFLOAD_GROUP, STORED_BEACON_NTF,
-		       iwl_mvm_rx_stored_beacon_notif, false),
+		       iwl_mvm_rx_stored_beacon_notif, RX_HANDLER_SYNC),
 	RX_HANDLER_GRP(DATA_PATH_GROUP, MU_GROUP_MGMT_NOTIF,
-		       iwl_mvm_mu_mimo_grp_notif, false),
-
+		       iwl_mvm_mu_mimo_grp_notif, RX_HANDLER_SYNC),
 };
 #undef RX_HANDLER
 #undef RX_HANDLER_GRP
@@ -743,6 +771,7 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode)
 struct iwl_async_handler_entry {
 	struct list_head list;
 	struct iwl_rx_cmd_buffer rxb;
+	enum iwl_rx_handler_context context;
 	void (*fn)(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb);
 };
 
@@ -769,7 +798,6 @@ static void iwl_mvm_async_handlers_wk(struct work_struct *wk)
 	INIT_LIST_HEAD(&local_list);
 
 	/* Ensure that we are not in stop flow (check iwl_mvm_mac_stop) */
-	mutex_lock(&mvm->mutex);
 
 	/*
 	 * Sync with Rx path with a lock. Remove all the entries from this list,
@@ -780,12 +808,15 @@ static void iwl_mvm_async_handlers_wk(struct work_struct *wk)
 	spin_unlock_bh(&mvm->async_handlers_lock);
 
 	list_for_each_entry_safe(entry, tmp, &local_list, list) {
+		if (entry->context == RX_HANDLER_ASYNC_LOCKED)
+			mutex_lock(&mvm->mutex);
 		entry->fn(mvm, &entry->rxb);
 		iwl_free_rxb(&entry->rxb);
 		list_del(&entry->list);
+		if (entry->context == RX_HANDLER_ASYNC_LOCKED)
+			mutex_unlock(&mvm->mutex);
 		kfree(entry);
 	}
-	mutex_unlock(&mvm->mutex);
 }
 
 static inline void iwl_mvm_rx_check_trigger(struct iwl_mvm *mvm,
@@ -842,7 +873,7 @@ static void iwl_mvm_rx_common(struct iwl_mvm *mvm,
 		if (rx_h->cmd_id != WIDE_ID(pkt->hdr.group_id, pkt->hdr.cmd))
 			continue;
 
-		if (!rx_h->async) {
+		if (rx_h->context == RX_HANDLER_SYNC) {
 			rx_h->fn(mvm, rxb);
 			return;
 		}
@@ -856,6 +887,7 @@ static void iwl_mvm_rx_common(struct iwl_mvm *mvm,
 		entry->rxb._offset = rxb->_offset;
 		entry->rxb._rx_page_order = rxb->_rx_page_order;
 		entry->fn = rx_h->fn;
+		entry->context = rx_h->context;
 		spin_lock(&mvm->async_handlers_lock);
 		list_add_tail(&entry->list, &mvm->async_handlers_list);
 		spin_unlock(&mvm->async_handlers_lock);
-- 
cgit v0.10.2


From 7b5424361ec9270f40c3e23cb747cc8b9ee66235 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 1 Feb 2016 13:46:06 +0200
Subject: iwlwifi: pcie: fine tune number of rxbs

We kick the allocator when we have 2 RBDs that don't have
attached RBs, and the allocator allocates 8 RBs meaning
that it needs another 6 RBDs to attach the RBs to.
The design is that allocator should always have enough RBDs
to fulfill requests, so we give in advance 6 RBDs to the
allocator so that when it is kicked, it gets additional 2 RBDs
and has enough RBDs.
These RBDs were taken from the Rx queue itself, meaning
that each Rx queue didn't have the maximal number of
RBDs, but MAX - 6.
Change initial number of RBDs in the system to include both
queue size and allocator reserves.
Note the multi-queue is always 511 instead of 512 to avoid a
full queue since we cannot detect this state easily enough in
the 9000 arch.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
index 8af818b..582008a 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
@@ -511,9 +511,12 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl)
  */
 #define FH_TX_CHICKEN_BITS_SCD_AUTO_RETRY_EN	(0x00000002)
 
-#define MQ_RX_TABLE_SIZE		512
-#define MQ_RX_TABLE_MASK		(MQ_RX_TABLE_SIZE - 1)
-#define MQ_RX_POOL_SIZE			MQ_RX_TABLE_MASK
+#define MQ_RX_TABLE_SIZE	512
+#define MQ_RX_TABLE_MASK	(MQ_RX_TABLE_SIZE - 1)
+#define MQ_RX_NUM_RBDS		(MQ_RX_TABLE_SIZE - 1)
+#define RX_POOL_SIZE		(MQ_RX_NUM_RBDS +	\
+				 IWL_MAX_RX_HW_QUEUES *	\
+				 (RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC))
 
 #define RX_QUEUE_SIZE                         256
 #define RX_QUEUE_MASK                         255
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 6677f31..dadafbd 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -347,7 +347,7 @@ struct iwl_tso_hdr_page {
  */
 struct iwl_trans_pcie {
 	struct iwl_rxq *rxq;
-	struct iwl_rx_mem_buffer rx_pool[MQ_RX_POOL_SIZE];
+	struct iwl_rx_mem_buffer rx_pool[RX_POOL_SIZE];
 	struct iwl_rx_mem_buffer *global_table[MQ_RX_TABLE_SIZE];
 	struct iwl_rb_allocator rba;
 	struct iwl_trans *trans;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 509f790..98524a0 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -434,7 +434,7 @@ static void iwl_pcie_free_rbs_pool(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
-	for (i = 0; i < MQ_RX_POOL_SIZE; i++) {
+	for (i = 0; i < RX_POOL_SIZE; i++) {
 		if (!trans_pcie->rx_pool[i].page)
 			continue;
 		dma_unmap_page(trans->dev, trans_pcie->rx_pool[i].page_dma,
@@ -835,7 +835,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rxq *def_rxq;
 	struct iwl_rb_allocator *rba = &trans_pcie->rba;
-	int i, err, num_rbds, allocator_pool_size;
+	int i, err, queue_size, allocator_pool_size, num_alloc;
 
 	if (!trans_pcie->rxq) {
 		err = iwl_pcie_rx_alloc(trans);
@@ -887,11 +887,12 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	}
 
 	/* move the pool to the default queue and allocator ownerships */
-	num_rbds = trans->cfg->mq_rx_supported ?
-		     MQ_RX_POOL_SIZE : RX_QUEUE_SIZE;
+	queue_size = trans->cfg->mq_rx_supported ?
+		     MQ_RX_NUM_RBDS : RX_QUEUE_SIZE;
 	allocator_pool_size = trans->num_rx_queues *
 		(RX_CLAIM_REQ_ALLOC - RX_POST_REQ_ALLOC);
-	for (i = 0; i < num_rbds; i++) {
+	num_alloc = queue_size + allocator_pool_size;
+	for (i = 0; i < num_alloc; i++) {
 		struct iwl_rx_mem_buffer *rxb = &trans_pcie->rx_pool[i];
 
 		if (i < allocator_pool_size)
-- 
cgit v0.10.2


From 17c867bfe89bd2e089752ac938468900387acbe2 Mon Sep 17 00:00:00 2001
From: Sara Sharon <sara.sharon@intel.com>
Date: Mon, 7 Mar 2016 14:18:29 +0200
Subject: iwlwifi: add support for getting HW address from CSR

From 9000 family on, we need to get HW address from host
CSR registers.
OEM can override it by fusing the override registers - read
those first, and if those are 0 - read the OTP registers instead.

In addition - bail out if no valid mac address is present. Make
it shared for all NICs.

Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
index 8e32a57..318b1dc 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c
@@ -140,7 +140,8 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 	.thermal_params = &iwl9000_tt_params,				\
 	.apmg_not_supported = true,					\
 	.mq_rx_supported = true,					\
-	.vht_mu_mimo_supported = true
+	.vht_mu_mimo_supported = true,					\
+	.mac_addr_from_csr = true
 
 const struct iwl_cfg iwl9260_2ac_cfg = {
 		.name = "Intel(R) Dual Band Wireless AC 9260",
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 4f2b57e..3e4d346 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -297,6 +297,7 @@ struct iwl_pwr_tx_backoff {
  * @host_interrupt_operation_mode: device needs host interrupt operation
  *	mode set
  * @nvm_hw_section_num: the ID of the HW NVM section
+ * @mac_addr_from_csr: read HW address from CSR registers
  * @features: hw features, any combination of feature_whitelist
  * @pwr_tx_backoffs: translation table between power limits and backoffs
  * @max_rx_agg_size: max RX aggregation size of the ADDBA request/response
@@ -345,6 +346,7 @@ struct iwl_cfg {
 	const bool host_interrupt_operation_mode;
 	bool high_temp;
 	u8   nvm_hw_section_num;
+	bool mac_addr_from_csr;
 	bool lp_xtal_workaround;
 	const struct iwl_pwr_tx_backoff *pwr_tx_backoffs;
 	bool no_power_up_nic_in_init;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index a79c4f6..b978f6c 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -598,4 +598,14 @@ enum msix_hw_int_causes {
 #define MSIX_AUTO_CLEAR_CAUSE			0
 #define MSIX_NON_AUTO_CLEAR_CAUSE		BIT(7)
 
+/*****************************************************************************
+ *                     HW address related registers                          *
+ *****************************************************************************/
+
+#define CSR_ADDR_BASE			(0x380)
+#define CSR_MAC_ADDR0_OTP		(CSR_ADDR_BASE)
+#define CSR_MAC_ADDR1_OTP		(CSR_ADDR_BASE + 4)
+#define CSR_MAC_ADDR0_STRAP		(CSR_ADDR_BASE + 8)
+#define CSR_MAC_ADDR1_STRAP		(CSR_ADDR_BASE + 0xC)
+
 #endif /* !__iwl_csr_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 5e6b90d..93a6895 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -71,6 +71,8 @@
 #include "iwl-modparams.h"
 #include "iwl-nvm-parse.h"
 #include "iwl-prph.h"
+#include "iwl-io.h"
+#include "iwl-csr.h"
 
 /* NVM offsets (in words) definitions */
 enum wkp_nvm_offsets {
@@ -524,6 +526,36 @@ static void iwl_set_radio_cfg(const struct iwl_cfg *cfg,
 	data->valid_rx_ant = NVM_RF_CFG_RX_ANT_MSK_FAMILY_8000(radio_cfg);
 }
 
+static void iwl_flip_hw_address(__le32 mac_addr0, __le32 mac_addr1, u8 *dest)
+{
+	const u8 *hw_addr;
+
+	hw_addr = (const u8 *)&mac_addr0;
+	dest[0] = hw_addr[3];
+	dest[1] = hw_addr[2];
+	dest[2] = hw_addr[1];
+	dest[3] = hw_addr[0];
+
+	hw_addr = (const u8 *)&mac_addr1;
+	dest[4] = hw_addr[1];
+	dest[5] = hw_addr[0];
+}
+
+static void iwl_set_hw_address_from_csr(struct iwl_trans *trans,
+					struct iwl_nvm_data *data)
+{
+	__le32 mac_addr0 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR0_STRAP));
+	__le32 mac_addr1 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR1_STRAP));
+
+	/* If OEM did not fuse address - get it from OTP */
+	if (!mac_addr0 && !mac_addr1) {
+		mac_addr0 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR0_OTP));
+		mac_addr1 = cpu_to_le32(iwl_read32(trans, CSR_MAC_ADDR1_OTP));
+	}
+
+	iwl_flip_hw_address(mac_addr0, mac_addr1, data->hw_addr);
+}
+
 static void iwl_set_hw_address_family_8000(struct iwl_trans *trans,
 					   const struct iwl_cfg *cfg,
 					   struct iwl_nvm_data *data,
@@ -564,21 +596,8 @@ static void iwl_set_hw_address_family_8000(struct iwl_trans *trans,
 						WFMP_MAC_ADDR_0));
 		__le32 mac_addr1 = cpu_to_le32(iwl_trans_read_prph(trans,
 						WFMP_MAC_ADDR_1));
-		/* read the MAC address from HW resisters */
-		hw_addr = (const u8 *)&mac_addr0;
-		data->hw_addr[0] = hw_addr[3];
-		data->hw_addr[1] = hw_addr[2];
-		data->hw_addr[2] = hw_addr[1];
-		data->hw_addr[3] = hw_addr[0];
-
-		hw_addr = (const u8 *)&mac_addr1;
-		data->hw_addr[4] = hw_addr[1];
-		data->hw_addr[5] = hw_addr[0];
-
-		if (!is_valid_ether_addr(data->hw_addr))
-			IWL_ERR(trans,
-				"mac address (%pM) from hw section is not valid\n",
-				data->hw_addr);
+
+		iwl_flip_hw_address(mac_addr0, mac_addr1, data->hw_addr);
 
 		return;
 	}
@@ -586,12 +605,14 @@ static void iwl_set_hw_address_family_8000(struct iwl_trans *trans,
 	IWL_ERR(trans, "mac address is not found\n");
 }
 
-static void iwl_set_hw_address(struct iwl_trans *trans,
-			       const struct iwl_cfg *cfg,
-			       struct iwl_nvm_data *data, const __le16 *nvm_hw,
-			       const __le16 *mac_override)
+static int iwl_set_hw_address(struct iwl_trans *trans,
+			      const struct iwl_cfg *cfg,
+			      struct iwl_nvm_data *data, const __le16 *nvm_hw,
+			      const __le16 *mac_override)
 {
-	if (cfg->device_family != IWL_DEVICE_FAMILY_8000) {
+	if (cfg->mac_addr_from_csr) {
+		iwl_set_hw_address_from_csr(trans, data);
+	} else if (cfg->device_family != IWL_DEVICE_FAMILY_8000) {
 		const u8 *hw_addr = (const u8 *)(nvm_hw + HW_ADDR);
 
 		/* The byte order is little endian 16 bit, meaning 214365 */
@@ -605,6 +626,13 @@ static void iwl_set_hw_address(struct iwl_trans *trans,
 		iwl_set_hw_address_family_8000(trans, cfg, data,
 					       mac_override, nvm_hw);
 	}
+
+	if (!is_valid_ether_addr(data->hw_addr)) {
+		IWL_ERR(trans, "no valid mac address was found\n");
+		return -EINVAL;
+	}
+
+	return 0;
 }
 
 struct iwl_nvm_data *
@@ -680,7 +708,12 @@ iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		ch_section = regulatory;
 	}
 
-	iwl_set_hw_address(trans, cfg, data, nvm_hw, mac_override);
+	/* If no valid mac address was found - bail out */
+	if (iwl_set_hw_address(trans, cfg, data, nvm_hw, mac_override)) {
+		kfree(data);
+		return NULL;
+	}
+
 	iwl_init_sbands(dev, cfg, data, ch_section, tx_chains, rx_chains,
 			lar_fw_supported && lar_enabled);
 	data->calib_version = 255;
-- 
cgit v0.10.2


From 532beba378d26d5bd9bbb1b485e969c13bf72009 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Mon, 7 Mar 2016 22:23:52 +0200
Subject: iwlwifi: mvm: don't let NDPs mess the packet tracking

We need to track the next packet that we will reclaim in
order to know when the Tx queues are empty. This is useful
when we open or tear down an A-MPDU session which requires
to switch queue.
The next packet being reclaimed is identified by its WiFi
sequence number and this is relevant only when we use QoS.
QoS NDPs do have a TID but have a meaningless sequence
number. The spec mandates the receiver to ignore the
sequence number in this case, allowing the transmitter to
put any sequence number. Our implementation leaves it 0.
When we reclaim a QoS NDP, we can't update the next_relcaim
counter since the sequence number of the QoS NDP itself is
invalid.
We used to update the next_reclaim based on the sequence
number of the QoS NDP which reset it to 1 (0 + 1) and
because of this, we never knew when the queue got empty.
This had to sad consequence to stuck the A-MPDU state
machine in a transient state.
To fix this, don't update next_reclaim when we reclaim
a QoS NDP.

Alesya saw this bug when testing u-APSD. Because the
A-MPDU state machine was stuck in EMPTYING_DELBA, we
updated mac80211 that we still have frames for that
station when it got back to sleep. mac80211 then wrongly
set the TIM bit in the beacon and requested to release
non-existent frames from the A-MPDU queue. This led to
a situation where the client was trying to poll frames
but we had no frames to send.

Reported-by: Alesya Shapira <alesya.shapira@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 271e8da..75870e6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -963,6 +964,7 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 	struct sk_buff_head skbs;
 	u8 skb_freed = 0;
 	u16 next_reclaimed, seq_ctl;
+	bool is_ndp = false;
 
 	__skb_queue_head_init(&skbs);
 
@@ -1016,6 +1018,20 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 			seq_ctl = le16_to_cpu(hdr->seq_ctrl);
 		}
 
+		if (unlikely(!seq_ctl)) {
+			struct ieee80211_hdr *hdr = (void *)skb->data;
+
+			/*
+			 * If it is an NDP, we can't update next_reclaim since
+			 * its sequence control is 0. Note that for that same
+			 * reason, NDPs are never sent to A-MPDU'able queues
+			 * so that we can never have more than one freed frame
+			 * for a single Tx resonse (see WARN_ON below).
+			 */
+			if (ieee80211_is_qos_nullfunc(hdr->frame_control))
+				is_ndp = true;
+		}
+
 		/*
 		 * TODO: this is not accurate if we are freeing more than one
 		 * packet.
@@ -1079,9 +1095,16 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm,
 			bool send_eosp_ndp = false;
 
 			spin_lock_bh(&mvmsta->lock);
-			tid_data->next_reclaimed = next_reclaimed;
-			IWL_DEBUG_TX_REPLY(mvm, "Next reclaimed packet:%d\n",
-					   next_reclaimed);
+			if (!is_ndp) {
+				tid_data->next_reclaimed = next_reclaimed;
+				IWL_DEBUG_TX_REPLY(mvm,
+						   "Next reclaimed packet:%d\n",
+						   next_reclaimed);
+			} else {
+				IWL_DEBUG_TX_REPLY(mvm,
+						   "NDP - don't update next_reclaimed\n");
+			}
+
 			iwl_mvm_check_ratid_empty(mvm, sta, tid);
 
 			if (mvmsta->sleep_tx_count) {
-- 
cgit v0.10.2


From 416cb2467bba25a6729b95e1ada0254a9477a360 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 8 Mar 2016 13:53:05 +0200
Subject: iwlwifi: mvm: remove RRM advertisement

mac80211 advertises this feature for all its drivers.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 3590835..39b9c38 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -610,8 +610,6 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 			IWL_UCODE_TLV_CAPA_WFA_TPC_REP_IE_SUPPORT))
 		hw->wiphy->features |= NL80211_FEATURE_WFA_TPC_IE_IN_PROBES;
 
-	wiphy_ext_feature_set(hw->wiphy, NL80211_EXT_FEATURE_RRM);
-
 	mvm->rts_threshold = IEEE80211_MAX_RTS_THRESHOLD;
 
 #ifdef CONFIG_PM_SLEEP
-- 
cgit v0.10.2


From b358993b3f582f43a6098028b5b08d084dff4d24 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Sun, 14 Feb 2016 14:03:10 +0200
Subject: iwlwifi: mvm: return the cooling state index instead of the budget

iwl_mvm_tcool_get_cur_state is the function that returns the
cooling state index to the sysfs handler. This function returns
mvm->cooling_dev.cur_state but that variable was set to the
budget and not the cooling state index. Fix that.
Add a missing blank line while at it.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index b869db9..9abbc93 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -543,8 +543,8 @@ struct iwl_mvm_thermal_device {
 };
 
 /*
- * iwl_mvm_cooling_device
- * @cur_state: current state in milliwatts
+ * struct iwl_mvm_cooling_device
+ * @cur_state: current state
  * @cdev: struct thermal cooling device
  */
 struct iwl_mvm_cooling_device {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index ce0c6fd..f1f2825 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -510,11 +510,35 @@ static const struct iwl_tt_params iwl_mvm_default_tt_params = {
 	.support_tx_backoff = true,
 };
 
-int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget)
+/* budget in mWatt */
+static const u32 iwl_mvm_cdev_budgets[] = {
+	2000,	/* cooling state 0 */
+	1800,	/* cooling state 1 */
+	1600,	/* cooling state 2 */
+	1400,	/* cooling state 3 */
+	1200,	/* cooling state 4 */
+	1000,	/* cooling state 5 */
+	900,	/* cooling state 6 */
+	800,	/* cooling state 7 */
+	700,	/* cooling state 8 */
+	650,	/* cooling state 9 */
+	600,	/* cooling state 10 */
+	550,	/* cooling state 11 */
+	500,	/* cooling state 12 */
+	450,	/* cooling state 13 */
+	400,	/* cooling state 14 */
+	350,	/* cooling state 15 */
+	300,	/* cooling state 16 */
+	250,	/* cooling state 17 */
+	200,	/* cooling state 18 */
+	150,	/* cooling state 19 */
+};
+
+int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 state)
 {
 	struct iwl_mvm_ctdp_cmd cmd = {
 		.operation = cpu_to_le32(op),
-		.budget = cpu_to_le32(budget),
+		.budget = cpu_to_le32(iwl_mvm_cdev_budgets[state]),
 		.window_size = 0,
 	};
 	int ret;
@@ -534,7 +558,7 @@ int iwl_mvm_ctdp_command(struct iwl_mvm *mvm, u32 op, u32 budget)
 	switch (op) {
 	case CTDP_CMD_OPERATION_START:
 #ifdef CONFIG_THERMAL
-		mvm->cooling_dev.cur_state = budget;
+		mvm->cooling_dev.cur_state = state;
 #endif /* CONFIG_THERMAL */
 		break;
 	case CTDP_CMD_OPERATION_REPORT:
@@ -759,29 +783,6 @@ static void iwl_mvm_thermal_zone_register(struct iwl_mvm *mvm)
 		mvm->tz_device.temp_trips[i] = S16_MIN;
 }
 
-static const u32 iwl_mvm_cdev_budgets[] = {
-	2000,	/* cooling state 0 */
-	1800,	/* cooling state 1 */
-	1600,	/* cooling state 2 */
-	1400,	/* cooling state 3 */
-	1200,	/* cooling state 4 */
-	1000,	/* cooling state 5 */
-	900,	/* cooling state 6 */
-	800,	/* cooling state 7 */
-	700,	/* cooling state 8 */
-	650,	/* cooling state 9 */
-	600,	/* cooling state 10 */
-	550,	/* cooling state 11 */
-	500,	/* cooling state 12 */
-	450,	/* cooling state 13 */
-	400,	/* cooling state 14 */
-	350,	/* cooling state 15 */
-	300,	/* cooling state 16 */
-	250,	/* cooling state 17 */
-	200,	/* cooling state 18 */
-	150,	/* cooling state 19 */
-};
-
 static int iwl_mvm_tcool_get_max_state(struct thermal_cooling_device *cdev,
 				       unsigned long *state)
 {
@@ -799,6 +800,7 @@ static int iwl_mvm_tcool_get_cur_state(struct thermal_cooling_device *cdev,
 		return -EBUSY;
 
 	*state = mvm->cooling_dev.cur_state;
+
 	return 0;
 }
 
@@ -822,7 +824,7 @@ static int iwl_mvm_tcool_set_cur_state(struct thermal_cooling_device *cdev,
 	}
 
 	ret = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_START,
-				   iwl_mvm_cdev_budgets[new_state]);
+				   new_state);
 
 unlock:
 	mutex_unlock(&mvm->mutex);
-- 
cgit v0.10.2


From e0e168dc8c4e41bab6ab94a838cfee50dd2b5f25 Mon Sep 17 00:00:00 2001
From: Gregory Greenman <gregory.greenman@intel.com>
Date: Mon, 29 Feb 2016 15:34:25 +0200
Subject: iwlwifi: pcie: avoid restocks inside rx loop if not emergency

When trying to reach high Rx throughput of more than 500Mbps on
a device with a relatively weak CPU (Atom x5-Z8500), CPU utilization
may become a bottleneck. Analysis showed that we are looping in
iwl_pcie_rx_handle for very long periods which led to starvation
of other threads (iwl_pcie_rx_handle runs with _bh disabled).
We were handling Rx and allocating new buffers and the new buffers
were ready quickly enough to be available before we had finished
handling all the buffers available in the hardware. As a
consequence, we called iwl_pcie_rxq_restock to refill the hardware
with the new buffers, and start again handling new buffers without
exiting the function. Since we read the hardware pointer again when
we goto restart, new buffers were handled immediately instead of
exiting the function.

This patch avoids refilling RBs inside rx handling loop, unless an
emergency situation is reached. It also doesn't read the hardware
pointer again unless we are in an emergency (unlikely) case.
This significantly reduce the maximal time we spend in
iwl_pcie_rx_handle with _bh disabled.

Signed-off-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 98524a0..4be3c35 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -231,6 +231,9 @@ static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans)
 	}
 }
 
+/*
+ * iwl_pcie_rxq_mq_restock - restock implementation for multi-queue rx
+ */
 static void iwl_pcie_rxq_mq_restock(struct iwl_trans *trans,
 				    struct iwl_rxq *rxq)
 {
@@ -277,17 +280,10 @@ static void iwl_pcie_rxq_mq_restock(struct iwl_trans *trans,
 }
 
 /*
- * iwl_pcie_rxq_restock - refill RX queue from pre-allocated pool
- *
- * If there are slots in the RX queue that need to be restocked,
- * and we have free pre-allocated buffers, fill the ranks as much
- * as we can, pulling from rx_free.
- *
- * This moves the 'write' index forward to catch up with 'processed', and
- * also updates the memory address in the firmware to reference the new
- * target buffer.
+ * iwl_pcie_rxq_sq_restock - restock implementation for single queue rx
  */
-static void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
+static void iwl_pcie_rxq_sq_restock(struct iwl_trans *trans,
+				    struct iwl_rxq *rxq)
 {
 	struct iwl_rx_mem_buffer *rxb;
 
@@ -332,6 +328,26 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
 }
 
 /*
+ * iwl_pcie_rxq_restock - refill RX queue from pre-allocated pool
+ *
+ * If there are slots in the RX queue that need to be restocked,
+ * and we have free pre-allocated buffers, fill the ranks as much
+ * as we can, pulling from rx_free.
+ *
+ * This moves the 'write' index forward to catch up with 'processed', and
+ * also updates the memory address in the firmware to reference the new
+ * target buffer.
+ */
+static
+void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq)
+{
+	if (trans->cfg->mq_rx_supported)
+		iwl_pcie_rxq_mq_restock(trans, rxq);
+	else
+		iwl_pcie_rxq_sq_restock(trans, rxq);
+}
+
+/*
  * iwl_pcie_rx_alloc_page - allocates and returns a page.
  *
  */
@@ -907,7 +923,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans)
 	if (trans->cfg->mq_rx_supported) {
 		iwl_pcie_rx_mq_hw_init(trans);
 	} else {
-		iwl_pcie_rxq_restock(trans, def_rxq);
+		iwl_pcie_rxq_sq_restock(trans, def_rxq);
 		iwl_pcie_rx_hw_init(trans, def_rxq);
 	}
 
@@ -1222,24 +1238,13 @@ restart:
 				count = 0;
 				if (rxq->used_count < rxq->queue_size / 3)
 					emergency = false;
+
+				rxq->read = i;
 				spin_unlock(&rxq->lock);
 				iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
-				spin_lock(&rxq->lock);
-			}
-		}
-		/* handle restock for three cases, can be all of them at once:
-		* - we just pulled buffers from the allocator
-		* - we have 8+ unstolen pages accumulated
-		* - we are in emergency and allocated buffers
-		 */
-		if (rxq->free_count >=  RX_CLAIM_REQ_ALLOC) {
-			rxq->read = i;
-			spin_unlock(&rxq->lock);
-			if (trans->cfg->mq_rx_supported)
-				iwl_pcie_rxq_mq_restock(trans, rxq);
-			else
 				iwl_pcie_rxq_restock(trans, rxq);
-			goto restart;
+				goto restart;
+			}
 		}
 	}
 out:
@@ -1264,6 +1269,8 @@ out:
 
 	if (rxq->napi.poll)
 		napi_gro_flush(&rxq->napi, false);
+
+	iwl_pcie_rxq_restock(trans, rxq);
 }
 
 static struct iwl_trans_pcie *iwl_pcie_get_trans_pcie(struct msix_entry *entry)
-- 
cgit v0.10.2


From 7d3ca7f4b1d5a43a1e0dbd112bb4c3bcf9a81fab Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Mon, 7 Mar 2016 10:16:38 +0200
Subject: iwlwifi: mvm: adapt the firmware assert log to new firmware

Newer firmware versions put different data in the memory
which is read by the driver upon firmware crash. Just
change the variable names in the code and the name of the
data in the log that we print withouth any functional
change.
On older firmware, there will be a mismatch between the
names that are printed and the content itself, but that's
harmless.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/main.c b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
index f62c2d7..85628127 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/main.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
@@ -1652,10 +1652,10 @@ static void iwl_dump_nic_error_log(struct iwl_priv *priv)
 
 	trace_iwlwifi_dev_ucode_error(trans->dev, table.error_id, table.tsf_low,
 				      table.data1, table.data2, table.line,
-				      table.blink1, table.blink2, table.ilink1,
-				      table.ilink2, table.bcon_time, table.gp1,
-				      table.gp2, table.gp3, table.ucode_ver,
-				      table.hw_ver, 0, table.brd_ver);
+				      table.blink2, table.ilink1, table.ilink2,
+				      table.bcon_time, table.gp1, table.gp2,
+				      table.gp3, table.ucode_ver, table.hw_ver,
+				      0, table.brd_ver);
 	IWL_ERR(priv, "0x%08X | %-28s\n", table.error_id,
 		desc_lookup(table.error_id));
 	IWL_ERR(priv, "0x%08X | uPc\n", table.pc);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
index 2a0703f..f02e2c8 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
@@ -2,6 +2,7 @@
  *
  * Copyright(c) 2009 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License as
@@ -121,13 +122,12 @@ TRACE_EVENT(iwlwifi_dev_tx,
 
 TRACE_EVENT(iwlwifi_dev_ucode_error,
 	TP_PROTO(const struct device *dev, u32 desc, u32 tsf_low,
-		 u32 data1, u32 data2, u32 line, u32 blink1,
-		 u32 blink2, u32 ilink1, u32 ilink2, u32 bcon_time,
-		 u32 gp1, u32 gp2, u32 gp3, u32 major, u32 minor, u32 hw_ver,
-		 u32 brd_ver),
+		 u32 data1, u32 data2, u32 line, u32 blink2, u32 ilink1,
+		 u32 ilink2, u32 bcon_time, u32 gp1, u32 gp2, u32 rev_type,
+		 u32 major, u32 minor, u32 hw_ver, u32 brd_ver),
 	TP_ARGS(dev, desc, tsf_low, data1, data2, line,
-		blink1, blink2, ilink1, ilink2, bcon_time, gp1, gp2,
-		gp3, major, minor, hw_ver, brd_ver),
+		 blink2, ilink1, ilink2, bcon_time, gp1, gp2,
+		 rev_type, major, minor, hw_ver, brd_ver),
 	TP_STRUCT__entry(
 		DEV_ENTRY
 		__field(u32, desc)
@@ -135,14 +135,13 @@ TRACE_EVENT(iwlwifi_dev_ucode_error,
 		__field(u32, data1)
 		__field(u32, data2)
 		__field(u32, line)
-		__field(u32, blink1)
 		__field(u32, blink2)
 		__field(u32, ilink1)
 		__field(u32, ilink2)
 		__field(u32, bcon_time)
 		__field(u32, gp1)
 		__field(u32, gp2)
-		__field(u32, gp3)
+		__field(u32, rev_type)
 		__field(u32, major)
 		__field(u32, minor)
 		__field(u32, hw_ver)
@@ -155,29 +154,27 @@ TRACE_EVENT(iwlwifi_dev_ucode_error,
 		__entry->data1 = data1;
 		__entry->data2 = data2;
 		__entry->line = line;
-		__entry->blink1 = blink1;
 		__entry->blink2 = blink2;
 		__entry->ilink1 = ilink1;
 		__entry->ilink2 = ilink2;
 		__entry->bcon_time = bcon_time;
 		__entry->gp1 = gp1;
 		__entry->gp2 = gp2;
-		__entry->gp3 = gp3;
+		__entry->rev_type = rev_type;
 		__entry->major = major;
 		__entry->minor = minor;
 		__entry->hw_ver = hw_ver;
 		__entry->brd_ver = brd_ver;
 	),
 	TP_printk("[%s] #%02d %010u data 0x%08X 0x%08X line %u, "
-		  "blink 0x%05X 0x%05X ilink 0x%05X 0x%05X "
-		  "bcon_tm %010u gp 0x%08X 0x%08X 0x%08X major 0x%08X "
+		  "blink2 0x%05X ilink 0x%05X 0x%05X "
+		  "bcon_tm %010u gp 0x%08X 0x%08X rev_type 0x%08X major 0x%08X "
 		  "minor 0x%08X hw 0x%08X brd 0x%08X",
 		  __get_str(dev), __entry->desc, __entry->tsf_low,
-		  __entry->data1,
-		  __entry->data2, __entry->line, __entry->blink1,
+		  __entry->data1, __entry->data2, __entry->line,
 		  __entry->blink2, __entry->ilink1, __entry->ilink2,
 		  __entry->bcon_time, __entry->gp1, __entry->gp2,
-		  __entry->gp3, __entry->major, __entry->minor,
+		  __entry->rev_type, __entry->major, __entry->minor,
 		  __entry->hw_ver, __entry->brd_ver)
 );
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 59453c1..53cdc57 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -376,8 +376,8 @@ struct iwl_error_event_table_v1 {
 struct iwl_error_event_table {
 	u32 valid;		/* (nonzero) valid, (0) log is empty */
 	u32 error_id;		/* type of error */
-	u32 pc;			/* program counter */
-	u32 blink1;		/* branch link */
+	u32 trm_hw_status0;	/* TRM HW status */
+	u32 trm_hw_status1;	/* TRM HW status */
 	u32 blink2;		/* branch link */
 	u32 ilink1;		/* interrupt link */
 	u32 ilink2;		/* interrupt link */
@@ -389,7 +389,7 @@ struct iwl_error_event_table {
 	u32 tsf_hi;		/* network timestamp function timer */
 	u32 gp1;		/* GP1 timer register */
 	u32 gp2;		/* GP2 timer register */
-	u32 gp3;		/* GP3 timer register */
+	u32 fw_rev_type;	/* firmware revision type */
 	u32 major;		/* uCode version major */
 	u32 minor;		/* uCode version minor */
 	u32 hw_ver;		/* HW Silicon version */
@@ -408,7 +408,7 @@ struct iwl_error_event_table {
 				 * time_flag */
 	u32 isr4;		/* isr status register LMPM_NIC_ISR4:
 				 * wico interrupt */
-	u32 isr_pref;		/* isr status register LMPM_NIC_PREF_STAT */
+	u32 last_cmd_id;	/* last HCMD id handled by the firmware */
 	u32 wait_event;		/* wait event() caller address */
 	u32 l2p_control;	/* L2pControlField */
 	u32 l2p_duration;	/* L2pDurationField */
@@ -419,7 +419,7 @@ struct iwl_error_event_table {
 	u32 u_timestamp;	/* indicate when the date and time of the
 				 * compilation */
 	u32 flow_handler;	/* FH read/write pointers, RX credit */
-} __packed /* LOG_ERROR_TABLE_API_S_VER_2 */;
+} __packed /* LOG_ERROR_TABLE_API_S_VER_3 */;
 
 /*
  * UMAC error struct - relevant starting from family 8000 chip.
@@ -529,9 +529,9 @@ static void iwl_mvm_dump_nic_error_log_old(struct iwl_mvm *mvm)
 
 	trace_iwlwifi_dev_ucode_error(trans->dev, table.error_id, table.tsf_low,
 				      table.data1, table.data2, table.data3,
-				      table.blink1, table.blink2, table.ilink1,
-				      table.ilink2, table.bcon_time, table.gp1,
-				      table.gp2, table.gp3, table.ucode_ver, 0,
+				      table.blink2, table.ilink1, table.ilink2,
+				      table.bcon_time, table.gp1, table.gp2,
+				      table.gp3, table.ucode_ver, 0,
 				      table.hw_ver, table.brd_ver);
 	IWL_ERR(mvm, "0x%08X | %-28s\n", table.error_id,
 		desc_lookup(table.error_id));
@@ -615,14 +615,14 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
 
 	trace_iwlwifi_dev_ucode_error(trans->dev, table.error_id, table.tsf_low,
 				      table.data1, table.data2, table.data3,
-				      table.blink1, table.blink2, table.ilink1,
+				      table.blink2, table.ilink1,
 				      table.ilink2, table.bcon_time, table.gp1,
-				      table.gp2, table.gp3, table.major,
+				      table.gp2, table.fw_rev_type, table.major,
 				      table.minor, table.hw_ver, table.brd_ver);
 	IWL_ERR(mvm, "0x%08X | %-28s\n", table.error_id,
 		desc_lookup(table.error_id));
-	IWL_ERR(mvm, "0x%08X | uPc\n", table.pc);
-	IWL_ERR(mvm, "0x%08X | branchlink1\n", table.blink1);
+	IWL_ERR(mvm, "0x%08X | trm_hw_status0\n", table.trm_hw_status0);
+	IWL_ERR(mvm, "0x%08X | trm_hw_status1\n", table.trm_hw_status1);
 	IWL_ERR(mvm, "0x%08X | branchlink2\n", table.blink2);
 	IWL_ERR(mvm, "0x%08X | interruptlink1\n", table.ilink1);
 	IWL_ERR(mvm, "0x%08X | interruptlink2\n", table.ilink2);
@@ -634,7 +634,7 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
 	IWL_ERR(mvm, "0x%08X | tsf hi\n", table.tsf_hi);
 	IWL_ERR(mvm, "0x%08X | time gp1\n", table.gp1);
 	IWL_ERR(mvm, "0x%08X | time gp2\n", table.gp2);
-	IWL_ERR(mvm, "0x%08X | time gp3\n", table.gp3);
+	IWL_ERR(mvm, "0x%08X | uCode revision type\n", table.fw_rev_type);
 	IWL_ERR(mvm, "0x%08X | uCode version major\n", table.major);
 	IWL_ERR(mvm, "0x%08X | uCode version minor\n", table.minor);
 	IWL_ERR(mvm, "0x%08X | hw version\n", table.hw_ver);
@@ -645,7 +645,7 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm)
 	IWL_ERR(mvm, "0x%08X | isr2\n", table.isr2);
 	IWL_ERR(mvm, "0x%08X | isr3\n", table.isr3);
 	IWL_ERR(mvm, "0x%08X | isr4\n", table.isr4);
-	IWL_ERR(mvm, "0x%08X | isr_pref\n", table.isr_pref);
+	IWL_ERR(mvm, "0x%08X | last cmd Id\n", table.last_cmd_id);
 	IWL_ERR(mvm, "0x%08X | wait_event\n", table.wait_event);
 	IWL_ERR(mvm, "0x%08X | l2p_control\n", table.l2p_control);
 	IWL_ERR(mvm, "0x%08X | l2p_duration\n", table.l2p_duration);
-- 
cgit v0.10.2


From 81279c49ce472bf5900ef8b19bf91e2ad6bd1c34 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 9 Mar 2016 14:58:47 +0100
Subject: iwlwifi: mvm: don't try to offload AES-CMAC in AP/IBSS modes

The firmware/hardware only supports checking AES-CMAC on RX, not
using it on TX. For station mode this is fine, since it's the only
thing it will ever do. For AP mode, it never receives such frames,
but must be able to transmit them. This is currently broken since
we try to enable them for hardware crypto (for RX only) and then
treat them as TX_CMD_SEC_EXT, leading to FIFO underruns during TX
so the frames never go out to the air.

To fix this, simply use software on TX in AP (and IBSS) mode.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 39b9c38..76e649c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2686,8 +2686,12 @@ static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
 			 * GTK on AP interface is a TX-only key, return 0;
 			 * on IBSS they're per-station and because we're lazy
 			 * we don't support them for RX, so do the same.
+			 * CMAC in AP/IBSS modes must be done in software.
 			 */
-			ret = 0;
+			if (key->cipher == WLAN_CIPHER_SUITE_AES_CMAC)
+				ret = -EOPNOTSUPP;
+			else
+				ret = 0;
 			key->hw_key_idx = STA_KEY_IDX_INVALID;
 			break;
 		}
-- 
cgit v0.10.2


From 5ed47226e06195935c5a2b88604b340a4364867e Mon Sep 17 00:00:00 2001
From: Ayala Beker <ayala.beker@intel.com>
Date: Wed, 3 Feb 2016 15:36:52 +0200
Subject: iwlwifi: mvm: update GSCAN capabilities

Gscan capabilities were updated with new capabilities supported
by the device. While at it, simplify the firmware support
conditional and move both conditions into the WARN() to make it
easier to undertand and use the unlikely() for both.

Signed-off-by: Ayala Beker <ayala.beker@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 184c0fe..f899666 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2007 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -374,15 +376,12 @@ static int iwl_store_cscheme(struct iwl_fw *fw, const u8 *data, const u32 len)
 	return 0;
 }
 
-static int iwl_store_gscan_capa(struct iwl_fw *fw, const u8 *data,
-				const u32 len)
+static void iwl_store_gscan_capa(struct iwl_fw *fw, const u8 *data,
+				 const u32 len)
 {
 	struct iwl_fw_gscan_capabilities *fw_capa = (void *)data;
 	struct iwl_gscan_capabilities *capa = &fw->gscan_capa;
 
-	if (len < sizeof(*fw_capa))
-		return -EINVAL;
-
 	capa->max_scan_cache_size = le32_to_cpu(fw_capa->max_scan_cache_size);
 	capa->max_scan_buckets = le32_to_cpu(fw_capa->max_scan_buckets);
 	capa->max_ap_cache_per_scan =
@@ -395,7 +394,15 @@ static int iwl_store_gscan_capa(struct iwl_fw *fw, const u8 *data,
 		le32_to_cpu(fw_capa->max_significant_change_aps);
 	capa->max_bssid_history_entries =
 		le32_to_cpu(fw_capa->max_bssid_history_entries);
-	return 0;
+	capa->max_hotlist_ssids = le32_to_cpu(fw_capa->max_hotlist_ssids);
+	capa->max_number_epno_networks =
+		le32_to_cpu(fw_capa->max_number_epno_networks);
+	capa->max_number_epno_networks_by_ssid =
+		le32_to_cpu(fw_capa->max_number_epno_networks_by_ssid);
+	capa->max_number_of_white_listed_ssid =
+		le32_to_cpu(fw_capa->max_number_of_white_listed_ssid);
+	capa->max_number_of_black_listed_ssid =
+		le32_to_cpu(fw_capa->max_number_of_black_listed_ssid);
 }
 
 /*
@@ -1023,8 +1030,15 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 				le32_to_cpup((__le32 *)tlv_data);
 			break;
 		case IWL_UCODE_TLV_FW_GSCAN_CAPA:
-			if (iwl_store_gscan_capa(&drv->fw, tlv_data, tlv_len))
-				goto invalid_tlv_len;
+			/*
+			 * Don't return an error in case of a shorter tlv_len
+			 * to enable loading of FW that has an old format
+			 * of GSCAN capabilities TLV.
+			 */
+			if (tlv_len < sizeof(struct iwl_fw_gscan_capabilities))
+				break;
+
+			iwl_store_gscan_capa(&drv->fw, tlv_data, tlv_len);
 			gscan_capa = true;
 			break;
 		default:
@@ -1046,12 +1060,8 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 		return -EINVAL;
 	}
 
-	/*
-	 * If ucode advertises that it supports GSCAN but GSCAN
-	 * capabilities TLV is not present, warn and continue without GSCAN.
-	 */
-	if (fw_has_capa(capa, IWL_UCODE_TLV_CAPA_GSCAN_SUPPORT) &&
-	    WARN(!gscan_capa,
+	if (WARN(fw_has_capa(capa, IWL_UCODE_TLV_CAPA_GSCAN_SUPPORT) &&
+		 !gscan_capa,
 		 "GSCAN is supported but capabilities TLV is unavailable\n"))
 		__clear_bit((__force long)IWL_UCODE_TLV_CAPA_GSCAN_SUPPORT,
 			    capa->_capa);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
index 5f69bf5..15ec4e2 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h
@@ -809,6 +809,12 @@ struct iwl_fw_dbg_conf_tlv {
  *	change APs.
  * @max_bssid_history_entries: number of BSSID/RSSI entries that the device can
  *	hold.
+ * @max_hotlist_ssids: maximum number of entries for hotlist SSIDs.
+ * @max_number_epno_networks: max number of epno entries.
+ * @max_number_epno_networks_by_ssid: max number of epno entries if ssid is
+ *	specified.
+ * @max_number_of_white_listed_ssid: max number of white listed SSIDs.
+ * @max_number_of_black_listed_ssid: max number of black listed SSIDs.
  */
 struct iwl_fw_gscan_capabilities {
 	__le32 max_scan_cache_size;
@@ -819,6 +825,11 @@ struct iwl_fw_gscan_capabilities {
 	__le32 max_hotlist_aps;
 	__le32 max_significant_change_aps;
 	__le32 max_bssid_history_entries;
+	__le32 max_hotlist_ssids;
+	__le32 max_number_epno_networks;
+	__le32 max_number_epno_networks_by_ssid;
+	__le32 max_number_of_white_listed_ssid;
+	__le32 max_number_of_black_listed_ssid;
 } __packed;
 
 #endif  /* __iwl_fw_file_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw.h
index 85d6d6d..2942571 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fw.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw.h
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2008 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016        Intel Deutschland GmbH
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -205,6 +207,12 @@ struct iwl_fw_cscheme_list {
  *	change APs.
  * @max_bssid_history_entries: number of BSSID/RSSI entries that the device can
  *	hold.
+ * @max_hotlist_ssids: maximum number of entries for hotlist SSIDs.
+ * @max_number_epno_networks: max number of epno entries.
+ * @max_number_epno_networks_by_ssid: max number of epno entries if ssid is
+ *	specified.
+ * @max_number_of_white_listed_ssid: max number of white listed SSIDs.
+ * @max_number_of_black_listed_ssid: max number of black listed SSIDs.
  */
 struct iwl_gscan_capabilities {
 	u32 max_scan_cache_size;
@@ -215,6 +223,11 @@ struct iwl_gscan_capabilities {
 	u32 max_hotlist_aps;
 	u32 max_significant_change_aps;
 	u32 max_bssid_history_entries;
+	u32 max_hotlist_ssids;
+	u32 max_number_epno_networks;
+	u32 max_number_epno_networks_by_ssid;
+	u32 max_number_of_white_listed_ssid;
+	u32 max_number_of_black_listed_ssid;
 };
 
 /**
-- 
cgit v0.10.2


From ff3c44e675054533403909ecb76e78c1d4efbd26 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:00 -0800
Subject: rcu: Add list_next_or_null_rcu

This is a convenience function that returns the next entry in an RCU
list or NULL if at the end of the list.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index 14ec165..17d4f84 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -319,6 +319,27 @@ static inline void list_splice_tail_init_rcu(struct list_head *list,
 })
 
 /**
+ * list_next_or_null_rcu - get the first element from a list
+ * @head:	the head for the list.
+ * @ptr:        the list head to take the next element from.
+ * @type:       the type of the struct this is embedded in.
+ * @member:     the name of the list_head within the struct.
+ *
+ * Note that if the ptr is at the end of the list, NULL is returned.
+ *
+ * This primitive may safely run concurrently with the _rcu list-mutation
+ * primitives such as list_add_rcu() as long as it's guarded by rcu_read_lock().
+ */
+#define list_next_or_null_rcu(head, ptr, type, member) \
+({ \
+	struct list_head *__head = (head); \
+	struct list_head *__ptr = (ptr); \
+	struct list_head *__next = READ_ONCE(__ptr->next); \
+	likely(__next != __head) ? list_entry_rcu(__next, type, \
+						  member) : NULL; \
+})
+
+/**
  * list_for_each_entry_rcu	-	iterate over rcu list of given type
  * @pos:	the type * to use as a loop cursor.
  * @head:	the head for your list.
-- 
cgit v0.10.2


From f4a00aacdb5f6784d46e8c999b6bb52ece4b306b Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:01 -0800
Subject: net: Make sock_alloc exportable

Export it for cases where we want to create sockets by hand.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/net.h b/include/linux/net.h
index 0b4ac7d..49175e4 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -215,6 +215,7 @@ int __sock_create(struct net *net, int family, int type, int proto,
 int sock_create(int family, int type, int proto, struct socket **res);
 int sock_create_kern(struct net *net, int family, int type, int proto, struct socket **res);
 int sock_create_lite(int family, int type, int proto, struct socket **res);
+struct socket *sock_alloc(void);
 void sock_release(struct socket *sock);
 int sock_sendmsg(struct socket *sock, struct msghdr *msg);
 int sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
diff --git a/net/socket.c b/net/socket.c
index c044d1e..38a78d4 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -533,7 +533,7 @@ static const struct inode_operations sockfs_inode_ops = {
  *	NULL is returned.
  */
 
-static struct socket *sock_alloc(void)
+struct socket *sock_alloc(void)
 {
 	struct inode *inode;
 	struct socket *sock;
@@ -554,6 +554,7 @@ static struct socket *sock_alloc(void)
 	this_cpu_add(sockets_in_use, 1);
 	return sock;
 }
+EXPORT_SYMBOL(sock_alloc);
 
 /**
  *	sock_release	-	close a socket
-- 
cgit v0.10.2


From 28a94d8fb35b3a75b802f368ae6f4a9f6b0d435a Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:02 -0800
Subject: net: Allow MSG_EOR in each msghdr of sendmmsg

This patch allows setting MSG_EOR in each individual msghdr passed
in sendmmsg. This allows a sendmmsg to send multiple messages when
using SOCK_SEQPACKET.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/socket.c b/net/socket.c
index 38a78d4..0dd4dd8 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1875,7 +1875,8 @@ static int copy_msghdr_from_user(struct msghdr *kmsg,
 
 static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
 			 struct msghdr *msg_sys, unsigned int flags,
-			 struct used_address *used_address)
+			 struct used_address *used_address,
+			 unsigned int allowed_msghdr_flags)
 {
 	struct compat_msghdr __user *msg_compat =
 	    (struct compat_msghdr __user *)msg;
@@ -1901,6 +1902,7 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
 
 	if (msg_sys->msg_controllen > INT_MAX)
 		goto out_freeiov;
+	flags |= (msg_sys->msg_flags & allowed_msghdr_flags);
 	ctl_len = msg_sys->msg_controllen;
 	if ((MSG_CMSG_COMPAT & flags) && ctl_len) {
 		err =
@@ -1979,7 +1981,7 @@ long __sys_sendmsg(int fd, struct user_msghdr __user *msg, unsigned flags)
 	if (!sock)
 		goto out;
 
-	err = ___sys_sendmsg(sock, msg, &msg_sys, flags, NULL);
+	err = ___sys_sendmsg(sock, msg, &msg_sys, flags, NULL, 0);
 
 	fput_light(sock->file, fput_needed);
 out:
@@ -2024,7 +2026,7 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen,
 	while (datagrams < vlen) {
 		if (MSG_CMSG_COMPAT & flags) {
 			err = ___sys_sendmsg(sock, (struct user_msghdr __user *)compat_entry,
-					     &msg_sys, flags, &used_address);
+					     &msg_sys, flags, &used_address, MSG_EOR);
 			if (err < 0)
 				break;
 			err = __put_user(err, &compat_entry->msg_len);
@@ -2032,7 +2034,7 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen,
 		} else {
 			err = ___sys_sendmsg(sock,
 					     (struct user_msghdr __user *)entry,
-					     &msg_sys, flags, &used_address);
+					     &msg_sys, flags, &used_address, MSG_EOR);
 			if (err < 0)
 				break;
 			err = put_user(err, &entry->msg_len);
-- 
cgit v0.10.2


From f092276d85b82504e8a07498f4e9e0c51f06745c Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:03 -0800
Subject: net: Add MSG_BATCH flag

Add a new msg flag called MSG_BATCH. This flag is used in sendmsg to
indicate that more messages will follow (i.e. a batch of messages is
being sent). This is similar to MSG_MORE except that the following
messages are not merged into one packet, they are sent individually.
sendmmsg is updated so that each contained message except for the
last one is marked as MSG_BATCH.

MSG_BATCH is a performance optimization in cases where a socket
implementation can benefit by transmitting packets in a batch.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/socket.h b/include/linux/socket.h
index 5bf59c8..d834af2 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -274,6 +274,7 @@ struct ucred {
 #define MSG_MORE	0x8000	/* Sender will send more */
 #define MSG_WAITFORONE	0x10000	/* recvmmsg(): block until 1+ packets avail */
 #define MSG_SENDPAGE_NOTLAST 0x20000 /* sendpage() internal : not the last page */
+#define MSG_BATCH	0x40000 /* sendmmsg(): more messages coming */
 #define MSG_EOF         MSG_FIN
 
 #define MSG_FASTOPEN	0x20000000	/* Send data in TCP SYN */
diff --git a/net/socket.c b/net/socket.c
index 0dd4dd8..886649c 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2008,6 +2008,7 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen,
 	struct compat_mmsghdr __user *compat_entry;
 	struct msghdr msg_sys;
 	struct used_address used_address;
+	unsigned int oflags = flags;
 
 	if (vlen > UIO_MAXIOV)
 		vlen = UIO_MAXIOV;
@@ -2022,8 +2023,12 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen,
 	entry = mmsg;
 	compat_entry = (struct compat_mmsghdr __user *)mmsg;
 	err = 0;
+	flags |= MSG_BATCH;
 
 	while (datagrams < vlen) {
+		if (datagrams == vlen - 1)
+			flags = oflags;
+
 		if (MSG_CMSG_COMPAT & flags) {
 			err = ___sys_sendmsg(sock, (struct user_msghdr __user *)compat_entry,
 					     &msg_sys, flags, &used_address, MSG_EOR);
-- 
cgit v0.10.2


From fa9835e52e3ea946916c2ce6c625c86421131740 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:04 -0800
Subject: net: Walk fragments in __skb_splice_bits

Add walking of fragments in __skb_splice_bits.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 9d7be61..51d768e 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1918,6 +1918,7 @@ static bool __skb_splice_bits(struct sk_buff *skb, struct pipe_inode_info *pipe,
 			      struct splice_pipe_desc *spd, struct sock *sk)
 {
 	int seg;
+	struct sk_buff *iter;
 
 	/* map the linear part :
 	 * If skb->head_frag is set, this 'linear' part is backed by a
@@ -1944,6 +1945,19 @@ static bool __skb_splice_bits(struct sk_buff *skb, struct pipe_inode_info *pipe,
 			return true;
 	}
 
+	skb_walk_frags(skb, iter) {
+		if (*offset >= iter->len) {
+			*offset -= iter->len;
+			continue;
+		}
+		/* __skb_splice_bits() only fails if the output has no room
+		 * left, so no point in going over the frag_list for the error
+		 * case.
+		 */
+		if (__skb_splice_bits(iter, pipe, offset, len, spd, sk))
+			return true;
+	}
+
 	return false;
 }
 
@@ -1970,9 +1984,7 @@ ssize_t skb_socket_splice(struct sock *sk,
 
 /*
  * Map data from the skb to a pipe. Should handle both the linear part,
- * the fragments, and the frag list. It does NOT handle frag lists within
- * the frag list, if such a thing exists. We'd probably need to recurse to
- * handle that cleanly.
+ * the fragments, and the frag list.
  */
 int skb_splice_bits(struct sk_buff *skb, struct sock *sk, unsigned int offset,
 		    struct pipe_inode_info *pipe, unsigned int tlen,
@@ -1991,29 +2003,10 @@ int skb_splice_bits(struct sk_buff *skb, struct sock *sk, unsigned int offset,
 		.ops = &nosteal_pipe_buf_ops,
 		.spd_release = sock_spd_release,
 	};
-	struct sk_buff *frag_iter;
 	int ret = 0;
 
-	/*
-	 * __skb_splice_bits() only fails if the output has no room left,
-	 * so no point in going over the frag_list for the error case.
-	 */
-	if (__skb_splice_bits(skb, pipe, &offset, &tlen, &spd, sk))
-		goto done;
-	else if (!tlen)
-		goto done;
+	__skb_splice_bits(skb, pipe, &offset, &tlen, &spd, sk);
 
-	/*
-	 * now see if we have a frag_list to map
-	 */
-	skb_walk_frags(skb, frag_iter) {
-		if (!tlen)
-			break;
-		if (__skb_splice_bits(frag_iter, pipe, &offset, &tlen, &spd, sk))
-			break;
-	}
-
-done:
 	if (spd.nr_pages)
 		ret = splice_cb(sk, pipe, &spd);
 
-- 
cgit v0.10.2


From 473bd239b808a8af5241ce9996a16d283d88ddff Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:05 -0800
Subject: tcp: Add tcp_inq to get available receive bytes on socket

Create a common kernel function to get the number of bytes available
on a TCP socket. This is based on code in INQ getsockopt and we now call
the function for that getsockopt.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/tcp.h b/include/net/tcp.h
index e90db85..0302636 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1816,4 +1816,28 @@ static inline void skb_set_tcp_pure_ack(struct sk_buff *skb)
 	skb->truesize = 2;
 }
 
+static inline int tcp_inq(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	int answ;
+
+	if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) {
+		answ = 0;
+	} else if (sock_flag(sk, SOCK_URGINLINE) ||
+		   !tp->urg_data ||
+		   before(tp->urg_seq, tp->copied_seq) ||
+		   !before(tp->urg_seq, tp->rcv_nxt)) {
+
+		answ = tp->rcv_nxt - tp->copied_seq;
+
+		/* Subtract 1, if FIN was received */
+		if (answ && sock_flag(sk, SOCK_DONE))
+			answ--;
+	} else {
+		answ = tp->urg_seq - tp->copied_seq;
+	}
+
+	return answ;
+}
+
 #endif	/* _TCP_H */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f9faadb..a265f00 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -556,20 +556,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg)
 			return -EINVAL;
 
 		slow = lock_sock_fast(sk);
-		if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))
-			answ = 0;
-		else if (sock_flag(sk, SOCK_URGINLINE) ||
-			 !tp->urg_data ||
-			 before(tp->urg_seq, tp->copied_seq) ||
-			 !before(tp->urg_seq, tp->rcv_nxt)) {
-
-			answ = tp->rcv_nxt - tp->copied_seq;
-
-			/* Subtract 1, if FIN was received */
-			if (answ && sock_flag(sk, SOCK_DONE))
-				answ--;
-		} else
-			answ = tp->urg_seq - tp->copied_seq;
+		answ = tcp_inq(sk);
 		unlock_sock_fast(sk, slow);
 		break;
 	case SIOCATMARK:
-- 
cgit v0.10.2


From ab7ac4eb9832e32a09f4e8042705484d2fb0aad3 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:06 -0800
Subject: kcm: Kernel Connection Multiplexor module

This module implements the Kernel Connection Multiplexor.

Kernel Connection Multiplexor (KCM) is a facility that provides a
message based interface over TCP for generic application protocols.
With KCM an application can efficiently send and receive application
protocol messages over TCP using datagram sockets.

For more information see the included Documentation/networking/kcm.txt

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/socket.h b/include/linux/socket.h
index d834af2..73bf6c6 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -200,7 +200,9 @@ struct ucred {
 #define AF_ALG		38	/* Algorithm sockets		*/
 #define AF_NFC		39	/* NFC sockets			*/
 #define AF_VSOCK	40	/* vSockets			*/
-#define AF_MAX		41	/* For now.. */
+#define AF_KCM		41	/* Kernel Connection Multiplexor*/
+
+#define AF_MAX		42	/* For now.. */
 
 /* Protocol families, same as address families. */
 #define PF_UNSPEC	AF_UNSPEC
@@ -246,6 +248,7 @@ struct ucred {
 #define PF_ALG		AF_ALG
 #define PF_NFC		AF_NFC
 #define PF_VSOCK	AF_VSOCK
+#define PF_KCM		AF_KCM
 #define PF_MAX		AF_MAX
 
 /* Maximum queue length specifiable by listen.  */
@@ -323,6 +326,7 @@ struct ucred {
 #define SOL_CAIF	278
 #define SOL_ALG		279
 #define SOL_NFC		280
+#define SOL_KCM		281
 
 /* IPX options */
 #define IPX_TYPE	1
diff --git a/include/net/kcm.h b/include/net/kcm.h
new file mode 100644
index 0000000..1bcae39
--- /dev/null
+++ b/include/net/kcm.h
@@ -0,0 +1,125 @@
+/*
+ * Kernel Connection Multiplexor
+ *
+ * Copyright (c) 2016 Tom Herbert <tom@herbertland.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ */
+
+#ifndef __NET_KCM_H_
+#define __NET_KCM_H_
+
+#include <linux/skbuff.h>
+#include <net/sock.h>
+#include <uapi/linux/kcm.h>
+
+extern unsigned int kcm_net_id;
+
+struct kcm_tx_msg {
+	unsigned int sent;
+	unsigned int fragidx;
+	unsigned int frag_offset;
+	unsigned int msg_flags;
+	struct sk_buff *frag_skb;
+	struct sk_buff *last_skb;
+};
+
+struct kcm_rx_msg {
+	int full_len;
+	int accum_len;
+	int offset;
+};
+
+/* Socket structure for KCM client sockets */
+struct kcm_sock {
+	struct sock sk;
+	struct kcm_mux *mux;
+	struct list_head kcm_sock_list;
+	int index;
+	u32 done : 1;
+	struct work_struct done_work;
+
+	/* Transmit */
+	struct kcm_psock *tx_psock;
+	struct work_struct tx_work;
+	struct list_head wait_psock_list;
+	struct sk_buff *seq_skb;
+
+	/* Don't use bit fields here, these are set under different locks */
+	bool tx_wait;
+	bool tx_wait_more;
+
+	/* Receive */
+	struct kcm_psock *rx_psock;
+	struct list_head wait_rx_list; /* KCMs waiting for receiving */
+	bool rx_wait;
+	u32 rx_disabled : 1;
+};
+
+struct bpf_prog;
+
+/* Structure for an attached lower socket */
+struct kcm_psock {
+	struct sock *sk;
+	struct kcm_mux *mux;
+	int index;
+
+	u32 tx_stopped : 1;
+	u32 rx_stopped : 1;
+	u32 done : 1;
+	u32 unattaching : 1;
+
+	void (*save_state_change)(struct sock *sk);
+	void (*save_data_ready)(struct sock *sk);
+	void (*save_write_space)(struct sock *sk);
+
+	struct list_head psock_list;
+
+	/* Receive */
+	struct sk_buff *rx_skb_head;
+	struct sk_buff **rx_skb_nextp;
+	struct sk_buff *ready_rx_msg;
+	struct list_head psock_ready_list;
+	struct work_struct rx_work;
+	struct delayed_work rx_delayed_work;
+	struct bpf_prog *bpf_prog;
+	struct kcm_sock *rx_kcm;
+
+	/* Transmit */
+	struct kcm_sock *tx_kcm;
+	struct list_head psock_avail_list;
+};
+
+/* Per net MUX list */
+struct kcm_net {
+	struct mutex mutex;
+	struct list_head mux_list;
+	int count;
+};
+
+/* Structure for a MUX */
+struct kcm_mux {
+	struct list_head kcm_mux_list;
+	struct rcu_head rcu;
+	struct kcm_net *knet;
+
+	struct list_head kcm_socks;	/* All KCM sockets on MUX */
+	int kcm_socks_cnt;		/* Total KCM socket count for MUX */
+	struct list_head psocks;	/* List of all psocks on MUX */
+	int psocks_cnt;		/* Total attached sockets */
+
+	/* Receive */
+	spinlock_t rx_lock ____cacheline_aligned_in_smp;
+	struct list_head kcm_rx_waiters; /* KCMs waiting for receiving */
+	struct list_head psocks_ready;	/* List of psocks with a msg ready */
+	struct sk_buff_head rx_hold_queue;
+
+	/* Transmit */
+	spinlock_t  lock ____cacheline_aligned_in_smp;	/* TX and mux locking */
+	struct list_head psocks_avail;	/* List of available psocks */
+	struct list_head kcm_tx_waiters; /* KCMs waiting for a TX psock */
+};
+
+#endif /* __NET_KCM_H_ */
diff --git a/include/uapi/linux/kcm.h b/include/uapi/linux/kcm.h
new file mode 100644
index 0000000..a5a53094
--- /dev/null
+++ b/include/uapi/linux/kcm.h
@@ -0,0 +1,40 @@
+/*
+ * Kernel Connection Multiplexor
+ *
+ * Copyright (c) 2016 Tom Herbert <tom@herbertland.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ *
+ * User API to clone KCM sockets and attach transport socket to a KCM
+ * multiplexor.
+ */
+
+#ifndef KCM_KERNEL_H
+#define KCM_KERNEL_H
+
+struct kcm_attach {
+	int fd;
+	int bpf_fd;
+};
+
+struct kcm_unattach {
+	int fd;
+};
+
+struct kcm_clone {
+	int fd;
+};
+
+#define SIOCKCMATTACH	(SIOCPROTOPRIVATE + 0)
+#define SIOCKCMUNATTACH	(SIOCPROTOPRIVATE + 1)
+#define SIOCKCMCLONE	(SIOCPROTOPRIVATE + 2)
+
+#define KCMPROTO_CONNECTED	0
+
+/* Socket options */
+#define KCM_RECV_DISABLE	1
+
+#endif
+
diff --git a/net/Kconfig b/net/Kconfig
index 2760825..10640d5 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -360,6 +360,7 @@ source "net/can/Kconfig"
 source "net/irda/Kconfig"
 source "net/bluetooth/Kconfig"
 source "net/rxrpc/Kconfig"
+source "net/kcm/Kconfig"
 
 config FIB_RULES
 	bool
diff --git a/net/Makefile b/net/Makefile
index a5d0409..81d1411 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_IRDA)		+= irda/
 obj-$(CONFIG_BT)		+= bluetooth/
 obj-$(CONFIG_SUNRPC)		+= sunrpc/
 obj-$(CONFIG_AF_RXRPC)		+= rxrpc/
+obj-$(CONFIG_AF_KCM)		+= kcm/
 obj-$(CONFIG_ATM)		+= atm/
 obj-$(CONFIG_L2TP)		+= l2tp/
 obj-$(CONFIG_DECNET)		+= decnet/
diff --git a/net/kcm/Kconfig b/net/kcm/Kconfig
new file mode 100644
index 0000000..5db94d9
--- /dev/null
+++ b/net/kcm/Kconfig
@@ -0,0 +1,10 @@
+
+config AF_KCM
+	tristate "KCM sockets"
+	depends on INET
+	select BPF_SYSCALL
+	---help---
+	  KCM (Kernel Connection Multiplexor) sockets provide a method
+	  for multiplexing messages of a message based application
+	  protocol over kernel connectons (e.g. TCP connections).
+
diff --git a/net/kcm/Makefile b/net/kcm/Makefile
new file mode 100644
index 0000000..cb525f7
--- /dev/null
+++ b/net/kcm/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_AF_KCM) += kcm.o
+
+kcm-y := kcmsock.o
diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c
new file mode 100644
index 0000000..30ef69a
--- /dev/null
+++ b/net/kcm/kcmsock.c
@@ -0,0 +1,2016 @@
+#include <linux/bpf.h>
+#include <linux/errno.h>
+#include <linux/errqueue.h>
+#include <linux/file.h>
+#include <linux/in.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/net.h>
+#include <linux/netdevice.h>
+#include <linux/poll.h>
+#include <linux/rculist.h>
+#include <linux/skbuff.h>
+#include <linux/socket.h>
+#include <linux/uaccess.h>
+#include <linux/workqueue.h>
+#include <net/kcm.h>
+#include <net/netns/generic.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include <uapi/linux/kcm.h>
+
+unsigned int kcm_net_id;
+
+static struct kmem_cache *kcm_psockp __read_mostly;
+static struct kmem_cache *kcm_muxp __read_mostly;
+static struct workqueue_struct *kcm_wq;
+
+static inline struct kcm_sock *kcm_sk(const struct sock *sk)
+{
+	return (struct kcm_sock *)sk;
+}
+
+static inline struct kcm_tx_msg *kcm_tx_msg(struct sk_buff *skb)
+{
+	return (struct kcm_tx_msg *)skb->cb;
+}
+
+static inline struct kcm_rx_msg *kcm_rx_msg(struct sk_buff *skb)
+{
+	return (struct kcm_rx_msg *)((void *)skb->cb +
+				     offsetof(struct qdisc_skb_cb, data));
+}
+
+static void report_csk_error(struct sock *csk, int err)
+{
+	csk->sk_err = EPIPE;
+	csk->sk_error_report(csk);
+}
+
+/* Callback lock held */
+static void kcm_abort_rx_psock(struct kcm_psock *psock, int err,
+			       struct sk_buff *skb)
+{
+	struct sock *csk = psock->sk;
+
+	/* Unrecoverable error in receive */
+
+	if (psock->rx_stopped)
+		return;
+
+	psock->rx_stopped = 1;
+
+	/* Report an error on the lower socket */
+	report_csk_error(csk, err);
+}
+
+static void kcm_abort_tx_psock(struct kcm_psock *psock, int err,
+			       bool wakeup_kcm)
+{
+	struct sock *csk = psock->sk;
+	struct kcm_mux *mux = psock->mux;
+
+	/* Unrecoverable error in transmit */
+
+	spin_lock_bh(&mux->lock);
+
+	if (psock->tx_stopped) {
+		spin_unlock_bh(&mux->lock);
+		return;
+	}
+
+	psock->tx_stopped = 1;
+
+	if (!psock->tx_kcm) {
+		/* Take off psocks_avail list */
+		list_del(&psock->psock_avail_list);
+	} else if (wakeup_kcm) {
+		/* In this case psock is being aborted while outside of
+		 * write_msgs and psock is reserved. Schedule tx_work
+		 * to handle the failure there. Need to commit tx_stopped
+		 * before queuing work.
+		 */
+		smp_mb();
+
+		queue_work(kcm_wq, &psock->tx_kcm->tx_work);
+	}
+
+	spin_unlock_bh(&mux->lock);
+
+	/* Report error on lower socket */
+	report_csk_error(csk, err);
+}
+
+static int kcm_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
+
+/* KCM is ready to receive messages on its queue-- either the KCM is new or
+ * has become unblocked after being blocked on full socket buffer. Queue any
+ * pending ready messages on a psock. RX mux lock held.
+ */
+static void kcm_rcv_ready(struct kcm_sock *kcm)
+{
+	struct kcm_mux *mux = kcm->mux;
+	struct kcm_psock *psock;
+	struct sk_buff *skb;
+
+	if (unlikely(kcm->rx_wait || kcm->rx_psock || kcm->rx_disabled))
+		return;
+
+	while (unlikely((skb = __skb_dequeue(&mux->rx_hold_queue)))) {
+		if (kcm_queue_rcv_skb(&kcm->sk, skb)) {
+			/* Assuming buffer limit has been reached */
+			skb_queue_head(&mux->rx_hold_queue, skb);
+			WARN_ON(!sk_rmem_alloc_get(&kcm->sk));
+			return;
+		}
+	}
+
+	while (!list_empty(&mux->psocks_ready)) {
+		psock = list_first_entry(&mux->psocks_ready, struct kcm_psock,
+					 psock_ready_list);
+
+		if (kcm_queue_rcv_skb(&kcm->sk, psock->ready_rx_msg)) {
+			/* Assuming buffer limit has been reached */
+			WARN_ON(!sk_rmem_alloc_get(&kcm->sk));
+			return;
+		}
+
+		/* Consumed the ready message on the psock. Schedule rx_work to
+		 * get more messages.
+		 */
+		list_del(&psock->psock_ready_list);
+		psock->ready_rx_msg = NULL;
+
+		/* Commit clearing of ready_rx_msg for queuing work */
+		smp_mb();
+
+		queue_work(kcm_wq, &psock->rx_work);
+	}
+
+	/* Buffer limit is okay now, add to ready list */
+	list_add_tail(&kcm->wait_rx_list,
+		      &kcm->mux->kcm_rx_waiters);
+	kcm->rx_wait = true;
+}
+
+static void kcm_rfree(struct sk_buff *skb)
+{
+	struct sock *sk = skb->sk;
+	struct kcm_sock *kcm = kcm_sk(sk);
+	struct kcm_mux *mux = kcm->mux;
+	unsigned int len = skb->truesize;
+
+	sk_mem_uncharge(sk, len);
+	atomic_sub(len, &sk->sk_rmem_alloc);
+
+	/* For reading rx_wait and rx_psock without holding lock */
+	smp_mb__after_atomic();
+
+	if (!kcm->rx_wait && !kcm->rx_psock &&
+	    sk_rmem_alloc_get(sk) < sk->sk_rcvlowat) {
+		spin_lock_bh(&mux->rx_lock);
+		kcm_rcv_ready(kcm);
+		spin_unlock_bh(&mux->rx_lock);
+	}
+}
+
+static int kcm_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
+{
+	struct sk_buff_head *list = &sk->sk_receive_queue;
+
+	if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
+		return -ENOMEM;
+
+	if (!sk_rmem_schedule(sk, skb, skb->truesize))
+		return -ENOBUFS;
+
+	skb->dev = NULL;
+
+	skb_orphan(skb);
+	skb->sk = sk;
+	skb->destructor = kcm_rfree;
+	atomic_add(skb->truesize, &sk->sk_rmem_alloc);
+	sk_mem_charge(sk, skb->truesize);
+
+	skb_queue_tail(list, skb);
+
+	if (!sock_flag(sk, SOCK_DEAD))
+		sk->sk_data_ready(sk);
+
+	return 0;
+}
+
+/* Requeue received messages for a kcm socket to other kcm sockets. This is
+ * called with a kcm socket is receive disabled.
+ * RX mux lock held.
+ */
+static void requeue_rx_msgs(struct kcm_mux *mux, struct sk_buff_head *head)
+{
+	struct sk_buff *skb;
+	struct kcm_sock *kcm;
+
+	while ((skb = __skb_dequeue(head))) {
+		/* Reset destructor to avoid calling kcm_rcv_ready */
+		skb->destructor = sock_rfree;
+		skb_orphan(skb);
+try_again:
+		if (list_empty(&mux->kcm_rx_waiters)) {
+			skb_queue_tail(&mux->rx_hold_queue, skb);
+			continue;
+		}
+
+		kcm = list_first_entry(&mux->kcm_rx_waiters,
+				       struct kcm_sock, wait_rx_list);
+
+		if (kcm_queue_rcv_skb(&kcm->sk, skb)) {
+			/* Should mean socket buffer full */
+			list_del(&kcm->wait_rx_list);
+			kcm->rx_wait = false;
+
+			/* Commit rx_wait to read in kcm_free */
+			smp_wmb();
+
+			goto try_again;
+		}
+	}
+}
+
+/* Lower sock lock held */
+static struct kcm_sock *reserve_rx_kcm(struct kcm_psock *psock,
+				       struct sk_buff *head)
+{
+	struct kcm_mux *mux = psock->mux;
+	struct kcm_sock *kcm;
+
+	WARN_ON(psock->ready_rx_msg);
+
+	if (psock->rx_kcm)
+		return psock->rx_kcm;
+
+	spin_lock_bh(&mux->rx_lock);
+
+	if (psock->rx_kcm) {
+		spin_unlock_bh(&mux->rx_lock);
+		return psock->rx_kcm;
+	}
+
+	if (list_empty(&mux->kcm_rx_waiters)) {
+		psock->ready_rx_msg = head;
+		list_add_tail(&psock->psock_ready_list,
+			      &mux->psocks_ready);
+		spin_unlock_bh(&mux->rx_lock);
+		return NULL;
+	}
+
+	kcm = list_first_entry(&mux->kcm_rx_waiters,
+			       struct kcm_sock, wait_rx_list);
+	list_del(&kcm->wait_rx_list);
+	kcm->rx_wait = false;
+
+	psock->rx_kcm = kcm;
+	kcm->rx_psock = psock;
+
+	spin_unlock_bh(&mux->rx_lock);
+
+	return kcm;
+}
+
+static void kcm_done(struct kcm_sock *kcm);
+
+static void kcm_done_work(struct work_struct *w)
+{
+	kcm_done(container_of(w, struct kcm_sock, done_work));
+}
+
+/* Lower sock held */
+static void unreserve_rx_kcm(struct kcm_psock *psock,
+			     bool rcv_ready)
+{
+	struct kcm_sock *kcm = psock->rx_kcm;
+	struct kcm_mux *mux = psock->mux;
+
+	if (!kcm)
+		return;
+
+	spin_lock_bh(&mux->rx_lock);
+
+	psock->rx_kcm = NULL;
+	kcm->rx_psock = NULL;
+
+	/* Commit kcm->rx_psock before sk_rmem_alloc_get to sync with
+	 * kcm_rfree
+	 */
+	smp_mb();
+
+	if (unlikely(kcm->done)) {
+		spin_unlock_bh(&mux->rx_lock);
+
+		/* Need to run kcm_done in a task since we need to qcquire
+		 * callback locks which may already be held here.
+		 */
+		INIT_WORK(&kcm->done_work, kcm_done_work);
+		schedule_work(&kcm->done_work);
+		return;
+	}
+
+	if (unlikely(kcm->rx_disabled)) {
+		requeue_rx_msgs(mux, &kcm->sk.sk_receive_queue);
+	} else if (rcv_ready || unlikely(!sk_rmem_alloc_get(&kcm->sk))) {
+		/* Check for degenerative race with rx_wait that all
+		 * data was dequeued (accounted for in kcm_rfree).
+		 */
+		kcm_rcv_ready(kcm);
+	}
+	spin_unlock_bh(&mux->rx_lock);
+}
+
+/* Macro to invoke filter function. */
+#define KCM_RUN_FILTER(prog, ctx) \
+	(*prog->bpf_func)(ctx, prog->insnsi)
+
+/* Lower socket lock held */
+static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
+			unsigned int orig_offset, size_t orig_len)
+{
+	struct kcm_psock *psock = (struct kcm_psock *)desc->arg.data;
+	struct kcm_rx_msg *rxm;
+	struct kcm_sock *kcm;
+	struct sk_buff *head, *skb;
+	size_t eaten = 0, cand_len;
+	ssize_t extra;
+	int err;
+	bool cloned_orig = false;
+
+	if (psock->ready_rx_msg)
+		return 0;
+
+	head = psock->rx_skb_head;
+	if (head) {
+		/* Message already in progress */
+
+		if (unlikely(orig_offset)) {
+			/* Getting data with a non-zero offset when a message is
+			 * in progress is not expected. If it does happen, we
+			 * need to clone and pull since we can't deal with
+			 * offsets in the skbs for a message expect in the head.
+			 */
+			orig_skb = skb_clone(orig_skb, GFP_ATOMIC);
+			if (!orig_skb) {
+				desc->error = -ENOMEM;
+				return 0;
+			}
+			if (!pskb_pull(orig_skb, orig_offset)) {
+				kfree_skb(orig_skb);
+				desc->error = -ENOMEM;
+				return 0;
+			}
+			cloned_orig = true;
+			orig_offset = 0;
+		}
+
+		if (!psock->rx_skb_nextp) {
+			/* We are going to append to the frags_list of head.
+			 * Need to unshare the frag_list.
+			 */
+			err = skb_unclone(head, GFP_ATOMIC);
+			if (err) {
+				desc->error = err;
+				return 0;
+			}
+
+			if (unlikely(skb_shinfo(head)->frag_list)) {
+				/* We can't append to an sk_buff that already
+				 * has a frag_list. We create a new head, point
+				 * the frag_list of that to the old head, and
+				 * then are able to use the old head->next for
+				 * appending to the message.
+				 */
+				if (WARN_ON(head->next)) {
+					desc->error = -EINVAL;
+					return 0;
+				}
+
+				skb = alloc_skb(0, GFP_ATOMIC);
+				if (!skb) {
+					desc->error = -ENOMEM;
+					return 0;
+				}
+				skb->len = head->len;
+				skb->data_len = head->len;
+				skb->truesize = head->truesize;
+				*kcm_rx_msg(skb) = *kcm_rx_msg(head);
+				psock->rx_skb_nextp = &head->next;
+				skb_shinfo(skb)->frag_list = head;
+				psock->rx_skb_head = skb;
+				head = skb;
+			} else {
+				psock->rx_skb_nextp =
+				    &skb_shinfo(head)->frag_list;
+			}
+		}
+	}
+
+	while (eaten < orig_len) {
+		/* Always clone since we will consume something */
+		skb = skb_clone(orig_skb, GFP_ATOMIC);
+		if (!skb) {
+			desc->error = -ENOMEM;
+			break;
+		}
+
+		cand_len = orig_len - eaten;
+
+		head = psock->rx_skb_head;
+		if (!head) {
+			head = skb;
+			psock->rx_skb_head = head;
+			/* Will set rx_skb_nextp on next packet if needed */
+			psock->rx_skb_nextp = NULL;
+			rxm = kcm_rx_msg(head);
+			memset(rxm, 0, sizeof(*rxm));
+			rxm->offset = orig_offset + eaten;
+		} else {
+			/* Unclone since we may be appending to an skb that we
+			 * already share a frag_list with.
+			 */
+			err = skb_unclone(skb, GFP_ATOMIC);
+			if (err) {
+				desc->error = err;
+				break;
+			}
+
+			rxm = kcm_rx_msg(head);
+			*psock->rx_skb_nextp = skb;
+			psock->rx_skb_nextp = &skb->next;
+			head->data_len += skb->len;
+			head->len += skb->len;
+			head->truesize += skb->truesize;
+		}
+
+		if (!rxm->full_len) {
+			ssize_t len;
+
+			len = KCM_RUN_FILTER(psock->bpf_prog, head);
+
+			if (!len) {
+				/* Need more header to determine length */
+				rxm->accum_len += cand_len;
+				eaten += cand_len;
+				WARN_ON(eaten != orig_len);
+				break;
+			} else if (len <= (ssize_t)head->len -
+					  skb->len - rxm->offset) {
+				/* Length must be into new skb (and also
+				 * greater than zero)
+				 */
+				desc->error = -EPROTO;
+				psock->rx_skb_head = NULL;
+				kcm_abort_rx_psock(psock, EPROTO, head);
+				break;
+			}
+
+			rxm->full_len = len;
+		}
+
+		extra = (ssize_t)(rxm->accum_len + cand_len) - rxm->full_len;
+
+		if (extra < 0) {
+			/* Message not complete yet. */
+			rxm->accum_len += cand_len;
+			eaten += cand_len;
+			WARN_ON(eaten != orig_len);
+			break;
+		}
+
+		/* Positive extra indicates ore bytes than needed for the
+		 * message
+		 */
+
+		WARN_ON(extra > cand_len);
+
+		eaten += (cand_len - extra);
+
+		/* Hurray, we have a new message! */
+		psock->rx_skb_head = NULL;
+
+try_queue:
+		kcm = reserve_rx_kcm(psock, head);
+		if (!kcm) {
+			/* Unable to reserve a KCM, message is held in psock. */
+			break;
+		}
+
+		if (kcm_queue_rcv_skb(&kcm->sk, head)) {
+			/* Should mean socket buffer full */
+			unreserve_rx_kcm(psock, false);
+			goto try_queue;
+		}
+	}
+
+	if (cloned_orig)
+		kfree_skb(orig_skb);
+
+	return eaten;
+}
+
+/* Called with lock held on lower socket */
+static int psock_tcp_read_sock(struct kcm_psock *psock)
+{
+	read_descriptor_t desc;
+
+	desc.arg.data = psock;
+	desc.error = 0;
+	desc.count = 1; /* give more than one skb per call */
+
+	/* sk should be locked here, so okay to do tcp_read_sock */
+	tcp_read_sock(psock->sk, &desc, kcm_tcp_recv);
+
+	unreserve_rx_kcm(psock, true);
+
+	return desc.error;
+}
+
+/* Lower sock lock held */
+static void psock_tcp_data_ready(struct sock *sk)
+{
+	struct kcm_psock *psock;
+
+	read_lock_bh(&sk->sk_callback_lock);
+
+	psock = (struct kcm_psock *)sk->sk_user_data;
+	if (unlikely(!psock || psock->rx_stopped))
+		goto out;
+
+	if (psock->ready_rx_msg)
+		goto out;
+
+	if (psock_tcp_read_sock(psock) == -ENOMEM)
+		queue_delayed_work(kcm_wq, &psock->rx_delayed_work, 0);
+
+out:
+	read_unlock_bh(&sk->sk_callback_lock);
+}
+
+static void do_psock_rx_work(struct kcm_psock *psock)
+{
+	read_descriptor_t rd_desc;
+	struct sock *csk = psock->sk;
+
+	/* We need the read lock to synchronize with psock_tcp_data_ready. We
+	 * need the socket lock for calling tcp_read_sock.
+	 */
+	lock_sock(csk);
+	read_lock_bh(&csk->sk_callback_lock);
+
+	if (unlikely(csk->sk_user_data != psock))
+		goto out;
+
+	if (unlikely(psock->rx_stopped))
+		goto out;
+
+	if (psock->ready_rx_msg)
+		goto out;
+
+	rd_desc.arg.data = psock;
+
+	if (psock_tcp_read_sock(psock) == -ENOMEM)
+		queue_delayed_work(kcm_wq, &psock->rx_delayed_work, 0);
+
+out:
+	read_unlock_bh(&csk->sk_callback_lock);
+	release_sock(csk);
+}
+
+static void psock_rx_work(struct work_struct *w)
+{
+	do_psock_rx_work(container_of(w, struct kcm_psock, rx_work));
+}
+
+static void psock_rx_delayed_work(struct work_struct *w)
+{
+	do_psock_rx_work(container_of(w, struct kcm_psock,
+				      rx_delayed_work.work));
+}
+
+static void psock_tcp_state_change(struct sock *sk)
+{
+	/* TCP only does a POLLIN for a half close. Do a POLLHUP here
+	 * since application will normally not poll with POLLIN
+	 * on the TCP sockets.
+	 */
+
+	report_csk_error(sk, EPIPE);
+}
+
+static void psock_tcp_write_space(struct sock *sk)
+{
+	struct kcm_psock *psock;
+	struct kcm_mux *mux;
+	struct kcm_sock *kcm;
+
+	read_lock_bh(&sk->sk_callback_lock);
+
+	psock = (struct kcm_psock *)sk->sk_user_data;
+	if (unlikely(!psock))
+		goto out;
+
+	mux = psock->mux;
+
+	spin_lock_bh(&mux->lock);
+
+	/* Check if the socket is reserved so someone is waiting for sending. */
+	kcm = psock->tx_kcm;
+	if (kcm)
+		queue_work(kcm_wq, &kcm->tx_work);
+
+	spin_unlock_bh(&mux->lock);
+out:
+	read_unlock_bh(&sk->sk_callback_lock);
+}
+
+static void unreserve_psock(struct kcm_sock *kcm);
+
+/* kcm sock is locked. */
+static struct kcm_psock *reserve_psock(struct kcm_sock *kcm)
+{
+	struct kcm_mux *mux = kcm->mux;
+	struct kcm_psock *psock;
+
+	psock = kcm->tx_psock;
+
+	smp_rmb(); /* Must read tx_psock before tx_wait */
+
+	if (psock) {
+		WARN_ON(kcm->tx_wait);
+		if (unlikely(psock->tx_stopped))
+			unreserve_psock(kcm);
+		else
+			return kcm->tx_psock;
+	}
+
+	spin_lock_bh(&mux->lock);
+
+	/* Check again under lock to see if psock was reserved for this
+	 * psock via psock_unreserve.
+	 */
+	psock = kcm->tx_psock;
+	if (unlikely(psock)) {
+		WARN_ON(kcm->tx_wait);
+		spin_unlock_bh(&mux->lock);
+		return kcm->tx_psock;
+	}
+
+	if (!list_empty(&mux->psocks_avail)) {
+		psock = list_first_entry(&mux->psocks_avail,
+					 struct kcm_psock,
+					 psock_avail_list);
+		list_del(&psock->psock_avail_list);
+		if (kcm->tx_wait) {
+			list_del(&kcm->wait_psock_list);
+			kcm->tx_wait = false;
+		}
+		kcm->tx_psock = psock;
+		psock->tx_kcm = kcm;
+	} else if (!kcm->tx_wait) {
+		list_add_tail(&kcm->wait_psock_list,
+			      &mux->kcm_tx_waiters);
+		kcm->tx_wait = true;
+	}
+
+	spin_unlock_bh(&mux->lock);
+
+	return psock;
+}
+
+/* mux lock held */
+static void psock_now_avail(struct kcm_psock *psock)
+{
+	struct kcm_mux *mux = psock->mux;
+	struct kcm_sock *kcm;
+
+	if (list_empty(&mux->kcm_tx_waiters)) {
+		list_add_tail(&psock->psock_avail_list,
+			      &mux->psocks_avail);
+	} else {
+		kcm = list_first_entry(&mux->kcm_tx_waiters,
+				       struct kcm_sock,
+				       wait_psock_list);
+		list_del(&kcm->wait_psock_list);
+		kcm->tx_wait = false;
+		psock->tx_kcm = kcm;
+
+		/* Commit before changing tx_psock since that is read in
+		 * reserve_psock before queuing work.
+		 */
+		smp_mb();
+
+		kcm->tx_psock = psock;
+		queue_work(kcm_wq, &kcm->tx_work);
+	}
+}
+
+/* kcm sock is locked. */
+static void unreserve_psock(struct kcm_sock *kcm)
+{
+	struct kcm_psock *psock;
+	struct kcm_mux *mux = kcm->mux;
+
+	spin_lock_bh(&mux->lock);
+
+	psock = kcm->tx_psock;
+
+	if (WARN_ON(!psock)) {
+		spin_unlock_bh(&mux->lock);
+		return;
+	}
+
+	smp_rmb(); /* Read tx_psock before tx_wait */
+
+	WARN_ON(kcm->tx_wait);
+
+	kcm->tx_psock = NULL;
+	psock->tx_kcm = NULL;
+
+	if (unlikely(psock->tx_stopped)) {
+		if (psock->done) {
+			/* Deferred free */
+			list_del(&psock->psock_list);
+			mux->psocks_cnt--;
+			sock_put(psock->sk);
+			fput(psock->sk->sk_socket->file);
+			kmem_cache_free(kcm_psockp, psock);
+		}
+
+		/* Don't put back on available list */
+
+		spin_unlock_bh(&mux->lock);
+
+		return;
+	}
+
+	psock_now_avail(psock);
+
+	spin_unlock_bh(&mux->lock);
+}
+
+/* Write any messages ready on the kcm socket.  Called with kcm sock lock
+ * held.  Return bytes actually sent or error.
+ */
+static int kcm_write_msgs(struct kcm_sock *kcm)
+{
+	struct sock *sk = &kcm->sk;
+	struct kcm_psock *psock;
+	struct sk_buff *skb, *head;
+	struct kcm_tx_msg *txm;
+	unsigned short fragidx, frag_offset;
+	unsigned int sent, total_sent = 0;
+	int ret = 0;
+
+	kcm->tx_wait_more = false;
+	psock = kcm->tx_psock;
+	if (unlikely(psock && psock->tx_stopped)) {
+		/* A reserved psock was aborted asynchronously. Unreserve
+		 * it and we'll retry the message.
+		 */
+		unreserve_psock(kcm);
+		if (skb_queue_empty(&sk->sk_write_queue))
+			return 0;
+
+		kcm_tx_msg(skb_peek(&sk->sk_write_queue))->sent = 0;
+
+	} else if (skb_queue_empty(&sk->sk_write_queue)) {
+		return 0;
+	}
+
+	head = skb_peek(&sk->sk_write_queue);
+	txm = kcm_tx_msg(head);
+
+	if (txm->sent) {
+		/* Send of first skbuff in queue already in progress */
+		if (WARN_ON(!psock)) {
+			ret = -EINVAL;
+			goto out;
+		}
+		sent = txm->sent;
+		frag_offset = txm->frag_offset;
+		fragidx = txm->fragidx;
+		skb = txm->frag_skb;
+
+		goto do_frag;
+	}
+
+try_again:
+	psock = reserve_psock(kcm);
+	if (!psock)
+		goto out;
+
+	do {
+		skb = head;
+		txm = kcm_tx_msg(head);
+		sent = 0;
+
+do_frag_list:
+		if (WARN_ON(!skb_shinfo(skb)->nr_frags)) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		for (fragidx = 0; fragidx < skb_shinfo(skb)->nr_frags;
+		     fragidx++) {
+			skb_frag_t *frag;
+
+			frag_offset = 0;
+do_frag:
+			frag = &skb_shinfo(skb)->frags[fragidx];
+			if (WARN_ON(!frag->size)) {
+				ret = -EINVAL;
+				goto out;
+			}
+
+			ret = kernel_sendpage(psock->sk->sk_socket,
+					      frag->page.p,
+					      frag->page_offset + frag_offset,
+					      frag->size - frag_offset,
+					      MSG_DONTWAIT);
+			if (ret <= 0) {
+				if (ret == -EAGAIN) {
+					/* Save state to try again when there's
+					 * write space on the socket
+					 */
+					txm->sent = sent;
+					txm->frag_offset = frag_offset;
+					txm->fragidx = fragidx;
+					txm->frag_skb = skb;
+
+					ret = 0;
+					goto out;
+				}
+
+				/* Hard failure in sending message, abort this
+				 * psock since it has lost framing
+				 * synchonization and retry sending the
+				 * message from the beginning.
+				 */
+				kcm_abort_tx_psock(psock, ret ? -ret : EPIPE,
+						   true);
+				unreserve_psock(kcm);
+
+				txm->sent = 0;
+				ret = 0;
+
+				goto try_again;
+			}
+
+			sent += ret;
+			frag_offset += ret;
+			if (frag_offset < frag->size) {
+				/* Not finished with this frag */
+				goto do_frag;
+			}
+		}
+
+		if (skb == head) {
+			if (skb_has_frag_list(skb)) {
+				skb = skb_shinfo(skb)->frag_list;
+				goto do_frag_list;
+			}
+		} else if (skb->next) {
+			skb = skb->next;
+			goto do_frag_list;
+		}
+
+		/* Successfully sent the whole packet, account for it. */
+		skb_dequeue(&sk->sk_write_queue);
+		kfree_skb(head);
+		sk->sk_wmem_queued -= sent;
+		total_sent += sent;
+	} while ((head = skb_peek(&sk->sk_write_queue)));
+out:
+	if (!head) {
+		/* Done with all queued messages. */
+		WARN_ON(!skb_queue_empty(&sk->sk_write_queue));
+		unreserve_psock(kcm);
+	}
+
+	/* Check if write space is available */
+	sk->sk_write_space(sk);
+
+	return total_sent ? : ret;
+}
+
+static void kcm_tx_work(struct work_struct *w)
+{
+	struct kcm_sock *kcm = container_of(w, struct kcm_sock, tx_work);
+	struct sock *sk = &kcm->sk;
+	int err;
+
+	lock_sock(sk);
+
+	/* Primarily for SOCK_DGRAM sockets, also handle asynchronous tx
+	 * aborts
+	 */
+	err = kcm_write_msgs(kcm);
+	if (err < 0) {
+		/* Hard failure in write, report error on KCM socket */
+		pr_warn("KCM: Hard failure on kcm_write_msgs %d\n", err);
+		report_csk_error(&kcm->sk, -err);
+		goto out;
+	}
+
+	/* Primarily for SOCK_SEQPACKET sockets */
+	if (likely(sk->sk_socket) &&
+	    test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
+		clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+		sk->sk_write_space(sk);
+	}
+
+out:
+	release_sock(sk);
+}
+
+static void kcm_push(struct kcm_sock *kcm)
+{
+	if (kcm->tx_wait_more)
+		kcm_write_msgs(kcm);
+}
+
+static int kcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
+{
+	struct sock *sk = sock->sk;
+	struct kcm_sock *kcm = kcm_sk(sk);
+	struct sk_buff *skb = NULL, *head = NULL;
+	size_t copy, copied = 0;
+	long timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
+	int eor = (sock->type == SOCK_DGRAM) ?
+		  !(msg->msg_flags & MSG_MORE) : !!(msg->msg_flags & MSG_EOR);
+	int err = -EPIPE;
+
+	lock_sock(sk);
+
+	/* Per tcp_sendmsg this should be in poll */
+	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+
+	if (sk->sk_err)
+		goto out_error;
+
+	if (kcm->seq_skb) {
+		/* Previously opened message */
+		head = kcm->seq_skb;
+		skb = kcm_tx_msg(head)->last_skb;
+		goto start;
+	}
+
+	/* Call the sk_stream functions to manage the sndbuf mem. */
+	if (!sk_stream_memory_free(sk)) {
+		kcm_push(kcm);
+		set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+		err = sk_stream_wait_memory(sk, &timeo);
+		if (err)
+			goto out_error;
+	}
+
+	/* New message, alloc head skb */
+	head = alloc_skb(0, sk->sk_allocation);
+	while (!head) {
+		kcm_push(kcm);
+		err = sk_stream_wait_memory(sk, &timeo);
+		if (err)
+			goto out_error;
+
+		head = alloc_skb(0, sk->sk_allocation);
+	}
+
+	skb = head;
+
+	/* Set ip_summed to CHECKSUM_UNNECESSARY to avoid calling
+	 * csum_and_copy_from_iter from skb_do_copy_data_nocache.
+	 */
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+start:
+	while (msg_data_left(msg)) {
+		bool merge = true;
+		int i = skb_shinfo(skb)->nr_frags;
+		struct page_frag *pfrag = sk_page_frag(sk);
+
+		if (!sk_page_frag_refill(sk, pfrag))
+			goto wait_for_memory;
+
+		if (!skb_can_coalesce(skb, i, pfrag->page,
+				      pfrag->offset)) {
+			if (i == MAX_SKB_FRAGS) {
+				struct sk_buff *tskb;
+
+				tskb = alloc_skb(0, sk->sk_allocation);
+				if (!tskb)
+					goto wait_for_memory;
+
+				if (head == skb)
+					skb_shinfo(head)->frag_list = tskb;
+				else
+					skb->next = tskb;
+
+				skb = tskb;
+				skb->ip_summed = CHECKSUM_UNNECESSARY;
+				continue;
+			}
+			merge = false;
+		}
+
+		copy = min_t(int, msg_data_left(msg),
+			     pfrag->size - pfrag->offset);
+
+		if (!sk_wmem_schedule(sk, copy))
+			goto wait_for_memory;
+
+		err = skb_copy_to_page_nocache(sk, &msg->msg_iter, skb,
+					       pfrag->page,
+					       pfrag->offset,
+					       copy);
+		if (err)
+			goto out_error;
+
+		/* Update the skb. */
+		if (merge) {
+			skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy);
+		} else {
+			skb_fill_page_desc(skb, i, pfrag->page,
+					   pfrag->offset, copy);
+			get_page(pfrag->page);
+		}
+
+		pfrag->offset += copy;
+		copied += copy;
+		if (head != skb) {
+			head->len += copy;
+			head->data_len += copy;
+		}
+
+		continue;
+
+wait_for_memory:
+		kcm_push(kcm);
+		err = sk_stream_wait_memory(sk, &timeo);
+		if (err)
+			goto out_error;
+	}
+
+	if (eor) {
+		bool not_busy = skb_queue_empty(&sk->sk_write_queue);
+
+		/* Message complete, queue it on send buffer */
+		__skb_queue_tail(&sk->sk_write_queue, head);
+		kcm->seq_skb = NULL;
+
+		if (msg->msg_flags & MSG_BATCH) {
+			kcm->tx_wait_more = true;
+		} else if (kcm->tx_wait_more || not_busy) {
+			err = kcm_write_msgs(kcm);
+			if (err < 0) {
+				/* We got a hard error in write_msgs but have
+				 * already queued this message. Report an error
+				 * in the socket, but don't affect return value
+				 * from sendmsg
+				 */
+				pr_warn("KCM: Hard failure on kcm_write_msgs\n");
+				report_csk_error(&kcm->sk, -err);
+			}
+		}
+	} else {
+		/* Message not complete, save state */
+partial_message:
+		kcm->seq_skb = head;
+		kcm_tx_msg(head)->last_skb = skb;
+	}
+
+	release_sock(sk);
+	return copied;
+
+out_error:
+	kcm_push(kcm);
+
+	if (copied && sock->type == SOCK_SEQPACKET) {
+		/* Wrote some bytes before encountering an
+		 * error, return partial success.
+		 */
+		goto partial_message;
+	}
+
+	if (head != kcm->seq_skb)
+		kfree_skb(head);
+
+	err = sk_stream_error(sk, msg->msg_flags, err);
+
+	/* make sure we wake any epoll edge trigger waiter */
+	if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && err == -EAGAIN))
+		sk->sk_write_space(sk);
+
+	release_sock(sk);
+	return err;
+}
+
+static struct sk_buff *kcm_wait_data(struct sock *sk, int flags,
+				     long timeo, int *err)
+{
+	struct sk_buff *skb;
+
+	while (!(skb = skb_peek(&sk->sk_receive_queue))) {
+		if (sk->sk_err) {
+			*err = sock_error(sk);
+			return NULL;
+		}
+
+		if (sock_flag(sk, SOCK_DONE))
+			return NULL;
+
+		if ((flags & MSG_DONTWAIT) || !timeo) {
+			*err = -EAGAIN;
+			return NULL;
+		}
+
+		sk_wait_data(sk, &timeo, NULL);
+
+		/* Handle signals */
+		if (signal_pending(current)) {
+			*err = sock_intr_errno(timeo);
+			return NULL;
+		}
+	}
+
+	return skb;
+}
+
+static int kcm_recvmsg(struct socket *sock, struct msghdr *msg,
+		       size_t len, int flags)
+{
+	struct sock *sk = sock->sk;
+	int err = 0;
+	long timeo;
+	struct kcm_rx_msg *rxm;
+	int copied = 0;
+	struct sk_buff *skb;
+
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+
+	lock_sock(sk);
+
+	skb = kcm_wait_data(sk, flags, timeo, &err);
+	if (!skb)
+		goto out;
+
+	/* Okay, have a message on the receive queue */
+
+	rxm = kcm_rx_msg(skb);
+
+	if (len > rxm->full_len)
+		len = rxm->full_len;
+
+	err = skb_copy_datagram_msg(skb, rxm->offset, msg, len);
+	if (err < 0)
+		goto out;
+
+	copied = len;
+	if (likely(!(flags & MSG_PEEK))) {
+		if (copied < rxm->full_len) {
+			if (sock->type == SOCK_DGRAM) {
+				/* Truncated message */
+				msg->msg_flags |= MSG_TRUNC;
+				goto msg_finished;
+			}
+			rxm->offset += copied;
+			rxm->full_len -= copied;
+		} else {
+msg_finished:
+			/* Finished with message */
+			msg->msg_flags |= MSG_EOR;
+			skb_unlink(skb, &sk->sk_receive_queue);
+			kfree_skb(skb);
+		}
+	}
+
+out:
+	release_sock(sk);
+
+	return copied ? : err;
+}
+
+/* kcm sock lock held */
+static void kcm_recv_disable(struct kcm_sock *kcm)
+{
+	struct kcm_mux *mux = kcm->mux;
+
+	if (kcm->rx_disabled)
+		return;
+
+	spin_lock_bh(&mux->rx_lock);
+
+	kcm->rx_disabled = 1;
+
+	/* If a psock is reserved we'll do cleanup in unreserve */
+	if (!kcm->rx_psock) {
+		if (kcm->rx_wait) {
+			list_del(&kcm->wait_rx_list);
+			kcm->rx_wait = false;
+		}
+
+		requeue_rx_msgs(mux, &kcm->sk.sk_receive_queue);
+	}
+
+	spin_unlock_bh(&mux->rx_lock);
+}
+
+/* kcm sock lock held */
+static void kcm_recv_enable(struct kcm_sock *kcm)
+{
+	struct kcm_mux *mux = kcm->mux;
+
+	if (!kcm->rx_disabled)
+		return;
+
+	spin_lock_bh(&mux->rx_lock);
+
+	kcm->rx_disabled = 0;
+	kcm_rcv_ready(kcm);
+
+	spin_unlock_bh(&mux->rx_lock);
+}
+
+static int kcm_setsockopt(struct socket *sock, int level, int optname,
+			  char __user *optval, unsigned int optlen)
+{
+	struct kcm_sock *kcm = kcm_sk(sock->sk);
+	int val, valbool;
+	int err = 0;
+
+	if (level != SOL_KCM)
+		return -ENOPROTOOPT;
+
+	if (optlen < sizeof(int))
+		return -EINVAL;
+
+	if (get_user(val, (int __user *)optval))
+		return -EINVAL;
+
+	valbool = val ? 1 : 0;
+
+	switch (optname) {
+	case KCM_RECV_DISABLE:
+		lock_sock(&kcm->sk);
+		if (valbool)
+			kcm_recv_disable(kcm);
+		else
+			kcm_recv_enable(kcm);
+		release_sock(&kcm->sk);
+		break;
+	default:
+		err = -ENOPROTOOPT;
+	}
+
+	return err;
+}
+
+static int kcm_getsockopt(struct socket *sock, int level, int optname,
+			  char __user *optval, int __user *optlen)
+{
+	struct kcm_sock *kcm = kcm_sk(sock->sk);
+	int val, len;
+
+	if (level != SOL_KCM)
+		return -ENOPROTOOPT;
+
+	if (get_user(len, optlen))
+		return -EFAULT;
+
+	len = min_t(unsigned int, len, sizeof(int));
+	if (len < 0)
+		return -EINVAL;
+
+	switch (optname) {
+	case KCM_RECV_DISABLE:
+		val = kcm->rx_disabled;
+		break;
+	default:
+		return -ENOPROTOOPT;
+	}
+
+	if (put_user(len, optlen))
+		return -EFAULT;
+	if (copy_to_user(optval, &val, len))
+		return -EFAULT;
+	return 0;
+}
+
+static void init_kcm_sock(struct kcm_sock *kcm, struct kcm_mux *mux)
+{
+	struct kcm_sock *tkcm;
+	struct list_head *head;
+	int index = 0;
+
+	/* For SOCK_SEQPACKET sock type, datagram_poll checks the sk_state, so
+	 * we set sk_state, otherwise epoll_wait always returns right away with
+	 * POLLHUP
+	 */
+	kcm->sk.sk_state = TCP_ESTABLISHED;
+
+	/* Add to mux's kcm sockets list */
+	kcm->mux = mux;
+	spin_lock_bh(&mux->lock);
+
+	head = &mux->kcm_socks;
+	list_for_each_entry(tkcm, &mux->kcm_socks, kcm_sock_list) {
+		if (tkcm->index != index)
+			break;
+		head = &tkcm->kcm_sock_list;
+		index++;
+	}
+
+	list_add(&kcm->kcm_sock_list, head);
+	kcm->index = index;
+
+	mux->kcm_socks_cnt++;
+	spin_unlock_bh(&mux->lock);
+
+	INIT_WORK(&kcm->tx_work, kcm_tx_work);
+
+	spin_lock_bh(&mux->rx_lock);
+	kcm_rcv_ready(kcm);
+	spin_unlock_bh(&mux->rx_lock);
+}
+
+static int kcm_attach(struct socket *sock, struct socket *csock,
+		      struct bpf_prog *prog)
+{
+	struct kcm_sock *kcm = kcm_sk(sock->sk);
+	struct kcm_mux *mux = kcm->mux;
+	struct sock *csk;
+	struct kcm_psock *psock = NULL, *tpsock;
+	struct list_head *head;
+	int index = 0;
+
+	if (csock->ops->family != PF_INET &&
+	    csock->ops->family != PF_INET6)
+		return -EINVAL;
+
+	csk = csock->sk;
+	if (!csk)
+		return -EINVAL;
+
+	/* Only support TCP for now */
+	if (csk->sk_protocol != IPPROTO_TCP)
+		return -EINVAL;
+
+	psock = kmem_cache_zalloc(kcm_psockp, GFP_KERNEL);
+	if (!psock)
+		return -ENOMEM;
+
+	psock->mux = mux;
+	psock->sk = csk;
+	psock->bpf_prog = prog;
+	INIT_WORK(&psock->rx_work, psock_rx_work);
+	INIT_DELAYED_WORK(&psock->rx_delayed_work, psock_rx_delayed_work);
+
+	sock_hold(csk);
+
+	write_lock_bh(&csk->sk_callback_lock);
+	psock->save_data_ready = csk->sk_data_ready;
+	psock->save_write_space = csk->sk_write_space;
+	psock->save_state_change = csk->sk_state_change;
+	csk->sk_user_data = psock;
+	csk->sk_data_ready = psock_tcp_data_ready;
+	csk->sk_write_space = psock_tcp_write_space;
+	csk->sk_state_change = psock_tcp_state_change;
+	write_unlock_bh(&csk->sk_callback_lock);
+
+	/* Finished initialization, now add the psock to the MUX. */
+	spin_lock_bh(&mux->lock);
+	head = &mux->psocks;
+	list_for_each_entry(tpsock, &mux->psocks, psock_list) {
+		if (tpsock->index != index)
+			break;
+		head = &tpsock->psock_list;
+		index++;
+	}
+
+	list_add(&psock->psock_list, head);
+	psock->index = index;
+
+	mux->psocks_cnt++;
+	psock_now_avail(psock);
+	spin_unlock_bh(&mux->lock);
+
+	/* Schedule RX work in case there are already bytes queued */
+	queue_work(kcm_wq, &psock->rx_work);
+
+	return 0;
+}
+
+static int kcm_attach_ioctl(struct socket *sock, struct kcm_attach *info)
+{
+	struct socket *csock;
+	struct bpf_prog *prog;
+	int err;
+
+	csock = sockfd_lookup(info->fd, &err);
+	if (!csock)
+		return -ENOENT;
+
+	prog = bpf_prog_get(info->bpf_fd);
+	if (IS_ERR(prog)) {
+		err = PTR_ERR(prog);
+		goto out;
+	}
+
+	if (prog->type != BPF_PROG_TYPE_SOCKET_FILTER) {
+		bpf_prog_put(prog);
+		err = -EINVAL;
+		goto out;
+	}
+
+	err = kcm_attach(sock, csock, prog);
+	if (err) {
+		bpf_prog_put(prog);
+		goto out;
+	}
+
+	/* Keep reference on file also */
+
+	return 0;
+out:
+	fput(csock->file);
+	return err;
+}
+
+static void kcm_unattach(struct kcm_psock *psock)
+{
+	struct sock *csk = psock->sk;
+	struct kcm_mux *mux = psock->mux;
+
+	/* Stop getting callbacks from TCP socket. After this there should
+	 * be no way to reserve a kcm for this psock.
+	 */
+	write_lock_bh(&csk->sk_callback_lock);
+	csk->sk_user_data = NULL;
+	csk->sk_data_ready = psock->save_data_ready;
+	csk->sk_write_space = psock->save_write_space;
+	csk->sk_state_change = psock->save_state_change;
+	psock->rx_stopped = 1;
+
+	if (WARN_ON(psock->rx_kcm)) {
+		write_unlock_bh(&csk->sk_callback_lock);
+		return;
+	}
+
+	spin_lock_bh(&mux->rx_lock);
+
+	/* Stop receiver activities. After this point psock should not be
+	 * able to get onto ready list either through callbacks or work.
+	 */
+	if (psock->ready_rx_msg) {
+		list_del(&psock->psock_ready_list);
+		kfree_skb(psock->ready_rx_msg);
+		psock->ready_rx_msg = NULL;
+	}
+
+	spin_unlock_bh(&mux->rx_lock);
+
+	write_unlock_bh(&csk->sk_callback_lock);
+
+	cancel_work_sync(&psock->rx_work);
+	cancel_delayed_work_sync(&psock->rx_delayed_work);
+
+	bpf_prog_put(psock->bpf_prog);
+
+	kfree_skb(psock->rx_skb_head);
+	psock->rx_skb_head = NULL;
+
+	spin_lock_bh(&mux->lock);
+
+	if (psock->tx_kcm) {
+		/* psock was reserved.  Just mark it finished and we will clean
+		 * up in the kcm paths, we need kcm lock which can not be
+		 * acquired here.
+		 */
+		spin_unlock_bh(&mux->lock);
+
+		/* We are unattaching a socket that is reserved. Abort the
+		 * socket since we may be out of sync in sending on it. We need
+		 * to do this without the mux lock.
+		 */
+		kcm_abort_tx_psock(psock, EPIPE, false);
+
+		spin_lock_bh(&mux->lock);
+		if (!psock->tx_kcm) {
+			/* psock now unreserved in window mux was unlocked */
+			goto no_reserved;
+		}
+		psock->done = 1;
+
+		/* Commit done before queuing work to process it */
+		smp_mb();
+
+		/* Queue tx work to make sure psock->done is handled */
+		queue_work(kcm_wq, &psock->tx_kcm->tx_work);
+		spin_unlock_bh(&mux->lock);
+	} else {
+no_reserved:
+		if (!psock->tx_stopped)
+			list_del(&psock->psock_avail_list);
+		list_del(&psock->psock_list);
+		mux->psocks_cnt--;
+		spin_unlock_bh(&mux->lock);
+
+		sock_put(csk);
+		fput(csk->sk_socket->file);
+		kmem_cache_free(kcm_psockp, psock);
+	}
+}
+
+static int kcm_unattach_ioctl(struct socket *sock, struct kcm_unattach *info)
+{
+	struct kcm_sock *kcm = kcm_sk(sock->sk);
+	struct kcm_mux *mux = kcm->mux;
+	struct kcm_psock *psock;
+	struct socket *csock;
+	struct sock *csk;
+	int err;
+
+	csock = sockfd_lookup(info->fd, &err);
+	if (!csock)
+		return -ENOENT;
+
+	csk = csock->sk;
+	if (!csk) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	err = -ENOENT;
+
+	spin_lock_bh(&mux->lock);
+
+	list_for_each_entry(psock, &mux->psocks, psock_list) {
+		if (psock->sk != csk)
+			continue;
+
+		/* Found the matching psock */
+
+		if (psock->unattaching || WARN_ON(psock->done)) {
+			err = -EALREADY;
+			break;
+		}
+
+		psock->unattaching = 1;
+
+		spin_unlock_bh(&mux->lock);
+
+		kcm_unattach(psock);
+
+		err = 0;
+		goto out;
+	}
+
+	spin_unlock_bh(&mux->lock);
+
+out:
+	fput(csock->file);
+	return err;
+}
+
+static struct proto kcm_proto = {
+	.name	= "KCM",
+	.owner	= THIS_MODULE,
+	.obj_size = sizeof(struct kcm_sock),
+};
+
+/* Clone a kcm socket. */
+static int kcm_clone(struct socket *osock, struct kcm_clone *info,
+		     struct socket **newsockp)
+{
+	struct socket *newsock;
+	struct sock *newsk;
+	struct file *newfile;
+	int err, newfd;
+
+	err = -ENFILE;
+	newsock = sock_alloc();
+	if (!newsock)
+		goto out;
+
+	newsock->type = osock->type;
+	newsock->ops = osock->ops;
+
+	__module_get(newsock->ops->owner);
+
+	newfd = get_unused_fd_flags(0);
+	if (unlikely(newfd < 0)) {
+		err = newfd;
+		goto out_fd_fail;
+	}
+
+	newfile = sock_alloc_file(newsock, 0, osock->sk->sk_prot_creator->name);
+	if (unlikely(IS_ERR(newfile))) {
+		err = PTR_ERR(newfile);
+		goto out_sock_alloc_fail;
+	}
+
+	newsk = sk_alloc(sock_net(osock->sk), PF_KCM, GFP_KERNEL,
+			 &kcm_proto, true);
+	if (!newsk) {
+		err = -ENOMEM;
+		goto out_sk_alloc_fail;
+	}
+
+	sock_init_data(newsock, newsk);
+	init_kcm_sock(kcm_sk(newsk), kcm_sk(osock->sk)->mux);
+
+	fd_install(newfd, newfile);
+	*newsockp = newsock;
+	info->fd = newfd;
+
+	return 0;
+
+out_sk_alloc_fail:
+	fput(newfile);
+out_sock_alloc_fail:
+	put_unused_fd(newfd);
+out_fd_fail:
+	sock_release(newsock);
+out:
+	return err;
+}
+
+static int kcm_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	int err;
+
+	switch (cmd) {
+	case SIOCKCMATTACH: {
+		struct kcm_attach info;
+
+		if (copy_from_user(&info, (void __user *)arg, sizeof(info)))
+			err = -EFAULT;
+
+		err = kcm_attach_ioctl(sock, &info);
+
+		break;
+	}
+	case SIOCKCMUNATTACH: {
+		struct kcm_unattach info;
+
+		if (copy_from_user(&info, (void __user *)arg, sizeof(info)))
+			err = -EFAULT;
+
+		err = kcm_unattach_ioctl(sock, &info);
+
+		break;
+	}
+	case SIOCKCMCLONE: {
+		struct kcm_clone info;
+		struct socket *newsock = NULL;
+
+		if (copy_from_user(&info, (void __user *)arg, sizeof(info)))
+			err = -EFAULT;
+
+		err = kcm_clone(sock, &info, &newsock);
+
+		if (!err) {
+			if (copy_to_user((void __user *)arg, &info,
+					 sizeof(info))) {
+				err = -EFAULT;
+				sock_release(newsock);
+			}
+		}
+
+		break;
+	}
+	default:
+		err = -ENOIOCTLCMD;
+		break;
+	}
+
+	return err;
+}
+
+static void free_mux(struct rcu_head *rcu)
+{
+	struct kcm_mux *mux = container_of(rcu,
+	    struct kcm_mux, rcu);
+
+	kmem_cache_free(kcm_muxp, mux);
+}
+
+static void release_mux(struct kcm_mux *mux)
+{
+	struct kcm_net *knet = mux->knet;
+	struct kcm_psock *psock, *tmp_psock;
+
+	/* Release psocks */
+	list_for_each_entry_safe(psock, tmp_psock,
+				 &mux->psocks, psock_list) {
+		if (!WARN_ON(psock->unattaching))
+			kcm_unattach(psock);
+	}
+
+	if (WARN_ON(mux->psocks_cnt))
+		return;
+
+	__skb_queue_purge(&mux->rx_hold_queue);
+
+	mutex_lock(&knet->mutex);
+	list_del_rcu(&mux->kcm_mux_list);
+	knet->count--;
+	mutex_unlock(&knet->mutex);
+
+	call_rcu(&mux->rcu, free_mux);
+}
+
+static void kcm_done(struct kcm_sock *kcm)
+{
+	struct kcm_mux *mux = kcm->mux;
+	struct sock *sk = &kcm->sk;
+	int socks_cnt;
+
+	spin_lock_bh(&mux->rx_lock);
+	if (kcm->rx_psock) {
+		/* Cleanup in unreserve_rx_kcm */
+		WARN_ON(kcm->done);
+		kcm->rx_disabled = 1;
+		kcm->done = 1;
+		spin_unlock_bh(&mux->rx_lock);
+		return;
+	}
+
+	if (kcm->rx_wait) {
+		list_del(&kcm->wait_rx_list);
+		kcm->rx_wait = false;
+	}
+	/* Move any pending receive messages to other kcm sockets */
+	requeue_rx_msgs(mux, &sk->sk_receive_queue);
+
+	spin_unlock_bh(&mux->rx_lock);
+
+	if (WARN_ON(sk_rmem_alloc_get(sk)))
+		return;
+
+	/* Detach from MUX */
+	spin_lock_bh(&mux->lock);
+
+	list_del(&kcm->kcm_sock_list);
+	mux->kcm_socks_cnt--;
+	socks_cnt = mux->kcm_socks_cnt;
+
+	spin_unlock_bh(&mux->lock);
+
+	if (!socks_cnt) {
+		/* We are done with the mux now. */
+		release_mux(mux);
+	}
+
+	WARN_ON(kcm->rx_wait);
+
+	sock_put(&kcm->sk);
+}
+
+/* Called by kcm_release to close a KCM socket.
+ * If this is the last KCM socket on the MUX, destroy the MUX.
+ */
+static int kcm_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	struct kcm_sock *kcm;
+	struct kcm_mux *mux;
+	struct kcm_psock *psock;
+
+	if (!sk)
+		return 0;
+
+	kcm = kcm_sk(sk);
+	mux = kcm->mux;
+
+	sock_orphan(sk);
+	kfree_skb(kcm->seq_skb);
+
+	lock_sock(sk);
+	/* Purge queue under lock to avoid race condition with tx_work trying
+	 * to act when queue is nonempty. If tx_work runs after this point
+	 * it will just return.
+	 */
+	__skb_queue_purge(&sk->sk_write_queue);
+	release_sock(sk);
+
+	spin_lock_bh(&mux->lock);
+	if (kcm->tx_wait) {
+		/* Take of tx_wait list, after this point there should be no way
+		 * that a psock will be assigned to this kcm.
+		 */
+		list_del(&kcm->wait_psock_list);
+		kcm->tx_wait = false;
+	}
+	spin_unlock_bh(&mux->lock);
+
+	/* Cancel work. After this point there should be no outside references
+	 * to the kcm socket.
+	 */
+	cancel_work_sync(&kcm->tx_work);
+
+	lock_sock(sk);
+	psock = kcm->tx_psock;
+	if (psock) {
+		/* A psock was reserved, so we need to kill it since it
+		 * may already have some bytes queued from a message. We
+		 * need to do this after removing kcm from tx_wait list.
+		 */
+		kcm_abort_tx_psock(psock, EPIPE, false);
+		unreserve_psock(kcm);
+	}
+	release_sock(sk);
+
+	WARN_ON(kcm->tx_wait);
+	WARN_ON(kcm->tx_psock);
+
+	sock->sk = NULL;
+
+	kcm_done(kcm);
+
+	return 0;
+}
+
+static const struct proto_ops kcm_ops = {
+	.family =	PF_KCM,
+	.owner =	THIS_MODULE,
+	.release =	kcm_release,
+	.bind =		sock_no_bind,
+	.connect =	sock_no_connect,
+	.socketpair =	sock_no_socketpair,
+	.accept =	sock_no_accept,
+	.getname =	sock_no_getname,
+	.poll =		datagram_poll,
+	.ioctl =	kcm_ioctl,
+	.listen =	sock_no_listen,
+	.shutdown =	sock_no_shutdown,
+	.setsockopt =	kcm_setsockopt,
+	.getsockopt =	kcm_getsockopt,
+	.sendmsg =	kcm_sendmsg,
+	.recvmsg =	kcm_recvmsg,
+	.mmap =		sock_no_mmap,
+	.sendpage =	sock_no_sendpage,
+};
+
+/* Create proto operation for kcm sockets */
+static int kcm_create(struct net *net, struct socket *sock,
+		      int protocol, int kern)
+{
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+	struct sock *sk;
+	struct kcm_mux *mux;
+
+	switch (sock->type) {
+	case SOCK_DGRAM:
+	case SOCK_SEQPACKET:
+		sock->ops = &kcm_ops;
+		break;
+	default:
+		return -ESOCKTNOSUPPORT;
+	}
+
+	if (protocol != KCMPROTO_CONNECTED)
+		return -EPROTONOSUPPORT;
+
+	sk = sk_alloc(net, PF_KCM, GFP_KERNEL, &kcm_proto, kern);
+	if (!sk)
+		return -ENOMEM;
+
+	/* Allocate a kcm mux, shared between KCM sockets */
+	mux = kmem_cache_zalloc(kcm_muxp, GFP_KERNEL);
+	if (!mux) {
+		sk_free(sk);
+		return -ENOMEM;
+	}
+
+	spin_lock_init(&mux->lock);
+	spin_lock_init(&mux->rx_lock);
+	INIT_LIST_HEAD(&mux->kcm_socks);
+	INIT_LIST_HEAD(&mux->kcm_rx_waiters);
+	INIT_LIST_HEAD(&mux->kcm_tx_waiters);
+
+	INIT_LIST_HEAD(&mux->psocks);
+	INIT_LIST_HEAD(&mux->psocks_ready);
+	INIT_LIST_HEAD(&mux->psocks_avail);
+
+	mux->knet = knet;
+
+	/* Add new MUX to list */
+	mutex_lock(&knet->mutex);
+	list_add_rcu(&mux->kcm_mux_list, &knet->mux_list);
+	knet->count++;
+	mutex_unlock(&knet->mutex);
+
+	skb_queue_head_init(&mux->rx_hold_queue);
+
+	/* Init KCM socket */
+	sock_init_data(sock, sk);
+	init_kcm_sock(kcm_sk(sk), mux);
+
+	return 0;
+}
+
+static struct net_proto_family kcm_family_ops = {
+	.family = PF_KCM,
+	.create = kcm_create,
+	.owner  = THIS_MODULE,
+};
+
+static __net_init int kcm_init_net(struct net *net)
+{
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+
+	INIT_LIST_HEAD_RCU(&knet->mux_list);
+	mutex_init(&knet->mutex);
+
+	return 0;
+}
+
+static __net_exit void kcm_exit_net(struct net *net)
+{
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+
+	/* All KCM sockets should be closed at this point, which should mean
+	 * that all multiplexors and psocks have been destroyed.
+	 */
+	WARN_ON(!list_empty(&knet->mux_list));
+}
+
+static struct pernet_operations kcm_net_ops = {
+	.init = kcm_init_net,
+	.exit = kcm_exit_net,
+	.id   = &kcm_net_id,
+	.size = sizeof(struct kcm_net),
+};
+
+static int __init kcm_init(void)
+{
+	int err = -ENOMEM;
+
+	kcm_muxp = kmem_cache_create("kcm_mux_cache",
+				     sizeof(struct kcm_mux), 0,
+				     SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
+	if (!kcm_muxp)
+		goto fail;
+
+	kcm_psockp = kmem_cache_create("kcm_psock_cache",
+				       sizeof(struct kcm_psock), 0,
+					SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
+	if (!kcm_psockp)
+		goto fail;
+
+	kcm_wq = create_singlethread_workqueue("kkcmd");
+	if (!kcm_wq)
+		goto fail;
+
+	err = proto_register(&kcm_proto, 1);
+	if (err)
+		goto fail;
+
+	err = sock_register(&kcm_family_ops);
+	if (err)
+		goto sock_register_fail;
+
+	err = register_pernet_device(&kcm_net_ops);
+	if (err)
+		goto net_ops_fail;
+
+	return 0;
+
+net_ops_fail:
+	sock_unregister(PF_KCM);
+
+sock_register_fail:
+	proto_unregister(&kcm_proto);
+
+fail:
+	kmem_cache_destroy(kcm_muxp);
+	kmem_cache_destroy(kcm_psockp);
+
+	if (kcm_wq)
+		destroy_workqueue(kcm_wq);
+
+	return err;
+}
+
+static void __exit kcm_exit(void)
+{
+	unregister_pernet_device(&kcm_net_ops);
+	sock_unregister(PF_KCM);
+	proto_unregister(&kcm_proto);
+	destroy_workqueue(kcm_wq);
+
+	kmem_cache_destroy(kcm_muxp);
+	kmem_cache_destroy(kcm_psockp);
+}
+
+module_init(kcm_init);
+module_exit(kcm_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_NETPROTO(PF_KCM);
+
-- 
cgit v0.10.2


From cd6e111bf5be5c70aef96a86d791ee7be0c0e137 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:07 -0800
Subject: kcm: Add statistics and proc interfaces

This patch adds various counters for KCM. These include counters for
messages and bytes received or sent, as well as counters for number of
attached/unattached TCP sockets and other error or edge events.

The statistics are exposed via a proc interface. /proc/net/kcm provides
statistics per KCM socket and per psock (attached TCP sockets).
/proc/net/kcm_stats provides aggregate statistics.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/kcm.h b/include/net/kcm.h
index 1bcae39..39c7abe 100644
--- a/include/net/kcm.h
+++ b/include/net/kcm.h
@@ -17,6 +17,42 @@
 
 extern unsigned int kcm_net_id;
 
+#define KCM_STATS_ADD(stat, count) ((stat) += (count))
+#define KCM_STATS_INCR(stat) ((stat)++)
+
+struct kcm_psock_stats {
+	unsigned long long rx_msgs;
+	unsigned long long rx_bytes;
+	unsigned long long tx_msgs;
+	unsigned long long tx_bytes;
+	unsigned int rx_aborts;
+	unsigned int rx_mem_fail;
+	unsigned int rx_need_more_hdr;
+	unsigned int rx_bad_hdr_len;
+	unsigned long long reserved;
+	unsigned long long unreserved;
+	unsigned int tx_aborts;
+};
+
+struct kcm_mux_stats {
+	unsigned long long rx_msgs;
+	unsigned long long rx_bytes;
+	unsigned long long tx_msgs;
+	unsigned long long tx_bytes;
+	unsigned int rx_ready_drops;
+	unsigned int tx_retries;
+	unsigned int psock_attach;
+	unsigned int psock_unattach_rsvd;
+	unsigned int psock_unattach;
+};
+
+struct kcm_stats {
+	unsigned long long rx_msgs;
+	unsigned long long rx_bytes;
+	unsigned long long tx_msgs;
+	unsigned long long tx_bytes;
+};
+
 struct kcm_tx_msg {
 	unsigned int sent;
 	unsigned int fragidx;
@@ -41,6 +77,8 @@ struct kcm_sock {
 	u32 done : 1;
 	struct work_struct done_work;
 
+	struct kcm_stats stats;
+
 	/* Transmit */
 	struct kcm_psock *tx_psock;
 	struct work_struct tx_work;
@@ -77,6 +115,8 @@ struct kcm_psock {
 
 	struct list_head psock_list;
 
+	struct kcm_psock_stats stats;
+
 	/* Receive */
 	struct sk_buff *rx_skb_head;
 	struct sk_buff **rx_skb_nextp;
@@ -86,15 +126,21 @@ struct kcm_psock {
 	struct delayed_work rx_delayed_work;
 	struct bpf_prog *bpf_prog;
 	struct kcm_sock *rx_kcm;
+	unsigned long long saved_rx_bytes;
+	unsigned long long saved_rx_msgs;
 
 	/* Transmit */
 	struct kcm_sock *tx_kcm;
 	struct list_head psock_avail_list;
+	unsigned long long saved_tx_bytes;
+	unsigned long long saved_tx_msgs;
 };
 
 /* Per net MUX list */
 struct kcm_net {
 	struct mutex mutex;
+	struct kcm_psock_stats aggregate_psock_stats;
+	struct kcm_mux_stats aggregate_mux_stats;
 	struct list_head mux_list;
 	int count;
 };
@@ -110,6 +156,9 @@ struct kcm_mux {
 	struct list_head psocks;	/* List of all psocks on MUX */
 	int psocks_cnt;		/* Total attached sockets */
 
+	struct kcm_mux_stats stats;
+	struct kcm_psock_stats aggregate_psock_stats;
+
 	/* Receive */
 	spinlock_t rx_lock ____cacheline_aligned_in_smp;
 	struct list_head kcm_rx_waiters; /* KCMs waiting for receiving */
@@ -122,4 +171,49 @@ struct kcm_mux {
 	struct list_head kcm_tx_waiters; /* KCMs waiting for a TX psock */
 };
 
+#ifdef CONFIG_PROC_FS
+int kcm_proc_init(void);
+void kcm_proc_exit(void);
+#else
+static int kcm_proc_init(void) { return 0; }
+static void kcm_proc_exit(void) { }
+#endif
+
+static inline void aggregate_psock_stats(struct kcm_psock_stats *stats,
+					 struct kcm_psock_stats *agg_stats)
+{
+	/* Save psock statistics in the mux when psock is being unattached. */
+
+#define SAVE_PSOCK_STATS(_stat) (agg_stats->_stat += stats->_stat)
+	SAVE_PSOCK_STATS(rx_msgs);
+	SAVE_PSOCK_STATS(rx_bytes);
+	SAVE_PSOCK_STATS(rx_aborts);
+	SAVE_PSOCK_STATS(rx_mem_fail);
+	SAVE_PSOCK_STATS(rx_need_more_hdr);
+	SAVE_PSOCK_STATS(rx_bad_hdr_len);
+	SAVE_PSOCK_STATS(tx_msgs);
+	SAVE_PSOCK_STATS(tx_bytes);
+	SAVE_PSOCK_STATS(reserved);
+	SAVE_PSOCK_STATS(unreserved);
+	SAVE_PSOCK_STATS(tx_aborts);
+#undef SAVE_PSOCK_STATS
+}
+
+static inline void aggregate_mux_stats(struct kcm_mux_stats *stats,
+				       struct kcm_mux_stats *agg_stats)
+{
+	/* Save psock statistics in the mux when psock is being unattached. */
+
+#define SAVE_MUX_STATS(_stat) (agg_stats->_stat += stats->_stat)
+	SAVE_MUX_STATS(rx_msgs);
+	SAVE_MUX_STATS(rx_bytes);
+	SAVE_MUX_STATS(tx_msgs);
+	SAVE_MUX_STATS(tx_bytes);
+	SAVE_MUX_STATS(rx_ready_drops);
+	SAVE_MUX_STATS(psock_attach);
+	SAVE_MUX_STATS(psock_unattach_rsvd);
+	SAVE_MUX_STATS(psock_unattach);
+#undef SAVE_MUX_STATS
+}
+
 #endif /* __NET_KCM_H_ */
diff --git a/net/kcm/Makefile b/net/kcm/Makefile
index cb525f7..7125613 100644
--- a/net/kcm/Makefile
+++ b/net/kcm/Makefile
@@ -1,3 +1,3 @@
 obj-$(CONFIG_AF_KCM) += kcm.o
 
-kcm-y := kcmsock.o
+kcm-y := kcmsock.o kcmproc.o
diff --git a/net/kcm/kcmproc.c b/net/kcm/kcmproc.c
new file mode 100644
index 0000000..5eb9809
--- /dev/null
+++ b/net/kcm/kcmproc.c
@@ -0,0 +1,422 @@
+#include <linux/in.h>
+#include <linux/inet.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/net.h>
+#include <linux/proc_fs.h>
+#include <linux/rculist.h>
+#include <linux/seq_file.h>
+#include <linux/socket.h>
+#include <net/inet_sock.h>
+#include <net/kcm.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <net/tcp.h>
+
+#ifdef CONFIG_PROC_FS
+struct kcm_seq_muxinfo {
+	char				*name;
+	const struct file_operations	*seq_fops;
+	const struct seq_operations	seq_ops;
+};
+
+static struct kcm_mux *kcm_get_first(struct seq_file *seq)
+{
+	struct net *net = seq_file_net(seq);
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+
+	return list_first_or_null_rcu(&knet->mux_list,
+				      struct kcm_mux, kcm_mux_list);
+}
+
+static struct kcm_mux *kcm_get_next(struct kcm_mux *mux)
+{
+	struct kcm_net *knet = mux->knet;
+
+	return list_next_or_null_rcu(&knet->mux_list, &mux->kcm_mux_list,
+				     struct kcm_mux, kcm_mux_list);
+}
+
+static struct kcm_mux *kcm_get_idx(struct seq_file *seq, loff_t pos)
+{
+	struct net *net = seq_file_net(seq);
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+	struct kcm_mux *m;
+
+	list_for_each_entry_rcu(m, &knet->mux_list, kcm_mux_list) {
+		if (!pos)
+			return m;
+		--pos;
+	}
+	return NULL;
+}
+
+static void *kcm_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	void *p;
+
+	if (v == SEQ_START_TOKEN)
+		p = kcm_get_first(seq);
+	else
+		p = kcm_get_next(v);
+	++*pos;
+	return p;
+}
+
+static void *kcm_seq_start(struct seq_file *seq, loff_t *pos)
+	__acquires(rcu)
+{
+	rcu_read_lock();
+
+	if (!*pos)
+		return SEQ_START_TOKEN;
+	else
+		return kcm_get_idx(seq, *pos - 1);
+}
+
+static void kcm_seq_stop(struct seq_file *seq, void *v)
+	__releases(rcu)
+{
+	rcu_read_unlock();
+}
+
+struct kcm_proc_mux_state {
+	struct seq_net_private p;
+	int idx;
+};
+
+static int kcm_seq_open(struct inode *inode, struct file *file)
+{
+	struct kcm_seq_muxinfo *muxinfo = PDE_DATA(inode);
+	int err;
+
+	err = seq_open_net(inode, file, &muxinfo->seq_ops,
+			   sizeof(struct kcm_proc_mux_state));
+	if (err < 0)
+		return err;
+	return err;
+}
+
+static void kcm_format_mux_header(struct seq_file *seq)
+{
+	struct net *net = seq_file_net(seq);
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+
+	seq_printf(seq,
+		   "*** KCM statistics (%d MUX) ****\n",
+		   knet->count);
+
+	seq_printf(seq,
+		   "%-14s %-10s %-16s %-10s %-16s %-8s %-8s %-8s %-8s %s",
+		   "Object",
+		   "RX-Msgs",
+		   "RX-Bytes",
+		   "TX-Msgs",
+		   "TX-Bytes",
+		   "Recv-Q",
+		   "Rmem",
+		   "Send-Q",
+		   "Smem",
+		   "Status");
+
+	/* XXX: pdsts header stuff here */
+	seq_puts(seq, "\n");
+}
+
+static void kcm_format_sock(struct kcm_sock *kcm, struct seq_file *seq,
+			    int i, int *len)
+{
+	seq_printf(seq,
+		   "   kcm-%-7u %-10llu %-16llu %-10llu %-16llu %-8d %-8d %-8d %-8s ",
+		   kcm->index,
+		   kcm->stats.rx_msgs,
+		   kcm->stats.rx_bytes,
+		   kcm->stats.tx_msgs,
+		   kcm->stats.tx_bytes,
+		   kcm->sk.sk_receive_queue.qlen,
+		   sk_rmem_alloc_get(&kcm->sk),
+		   kcm->sk.sk_write_queue.qlen,
+		   "-");
+
+	if (kcm->tx_psock)
+		seq_printf(seq, "Psck-%u ", kcm->tx_psock->index);
+
+	if (kcm->tx_wait)
+		seq_puts(seq, "TxWait ");
+
+	if (kcm->tx_wait_more)
+		seq_puts(seq, "WMore ");
+
+	if (kcm->rx_wait)
+		seq_puts(seq, "RxWait ");
+
+	seq_puts(seq, "\n");
+}
+
+static void kcm_format_psock(struct kcm_psock *psock, struct seq_file *seq,
+			     int i, int *len)
+{
+	seq_printf(seq,
+		   "   psock-%-5u %-10llu %-16llu %-10llu %-16llu %-8d %-8d %-8d %-8d ",
+		   psock->index,
+		   psock->stats.rx_msgs,
+		   psock->stats.rx_bytes,
+		   psock->stats.tx_msgs,
+		   psock->stats.tx_bytes,
+		   psock->sk->sk_receive_queue.qlen,
+		   atomic_read(&psock->sk->sk_rmem_alloc),
+		   psock->sk->sk_write_queue.qlen,
+		   atomic_read(&psock->sk->sk_wmem_alloc));
+
+	if (psock->done)
+		seq_puts(seq, "Done ");
+
+	if (psock->tx_stopped)
+		seq_puts(seq, "TxStop ");
+
+	if (psock->rx_stopped)
+		seq_puts(seq, "RxStop ");
+
+	if (psock->tx_kcm)
+		seq_printf(seq, "Rsvd-%d ", psock->tx_kcm->index);
+
+	if (psock->ready_rx_msg)
+		seq_puts(seq, "RdyRx ");
+
+	seq_puts(seq, "\n");
+}
+
+static void
+kcm_format_mux(struct kcm_mux *mux, loff_t idx, struct seq_file *seq)
+{
+	int i, len;
+	struct kcm_sock *kcm;
+	struct kcm_psock *psock;
+
+	/* mux information */
+	seq_printf(seq,
+		   "%-6s%-8s %-10llu %-16llu %-10llu %-16llu %-8s %-8s %-8s %-8s ",
+		   "mux", "",
+		   mux->stats.rx_msgs,
+		   mux->stats.rx_bytes,
+		   mux->stats.tx_msgs,
+		   mux->stats.tx_bytes,
+		   "-", "-", "-", "-");
+
+	seq_printf(seq, "KCMs: %d, Psocks %d\n",
+		   mux->kcm_socks_cnt, mux->psocks_cnt);
+
+	/* kcm sock information */
+	i = 0;
+	spin_lock_bh(&mux->lock);
+	list_for_each_entry(kcm, &mux->kcm_socks, kcm_sock_list) {
+		kcm_format_sock(kcm, seq, i, &len);
+		i++;
+	}
+	i = 0;
+	list_for_each_entry(psock, &mux->psocks, psock_list) {
+		kcm_format_psock(psock, seq, i, &len);
+		i++;
+	}
+	spin_unlock_bh(&mux->lock);
+}
+
+static int kcm_seq_show(struct seq_file *seq, void *v)
+{
+	struct kcm_proc_mux_state *mux_state;
+
+	mux_state = seq->private;
+	if (v == SEQ_START_TOKEN) {
+		mux_state->idx = 0;
+		kcm_format_mux_header(seq);
+	} else {
+		kcm_format_mux(v, mux_state->idx, seq);
+		mux_state->idx++;
+	}
+	return 0;
+}
+
+static const struct file_operations kcm_seq_fops = {
+	.owner		= THIS_MODULE,
+	.open		= kcm_seq_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+};
+
+static struct kcm_seq_muxinfo kcm_seq_muxinfo = {
+	.name		= "kcm",
+	.seq_fops	= &kcm_seq_fops,
+	.seq_ops	= {
+		.show	= kcm_seq_show,
+		.start	= kcm_seq_start,
+		.next	= kcm_seq_next,
+		.stop	= kcm_seq_stop,
+	}
+};
+
+static int kcm_proc_register(struct net *net, struct kcm_seq_muxinfo *muxinfo)
+{
+	struct proc_dir_entry *p;
+	int rc = 0;
+
+	p = proc_create_data(muxinfo->name, S_IRUGO, net->proc_net,
+			     muxinfo->seq_fops, muxinfo);
+	if (!p)
+		rc = -ENOMEM;
+	return rc;
+}
+EXPORT_SYMBOL(kcm_proc_register);
+
+static void kcm_proc_unregister(struct net *net,
+				struct kcm_seq_muxinfo *muxinfo)
+{
+	remove_proc_entry(muxinfo->name, net->proc_net);
+}
+EXPORT_SYMBOL(kcm_proc_unregister);
+
+static int kcm_stats_seq_show(struct seq_file *seq, void *v)
+{
+	struct kcm_psock_stats psock_stats;
+	struct kcm_mux_stats mux_stats;
+	struct kcm_mux *mux;
+	struct kcm_psock *psock;
+	struct net *net = seq->private;
+	struct kcm_net *knet = net_generic(net, kcm_net_id);
+
+	memset(&mux_stats, 0, sizeof(mux_stats));
+	memset(&psock_stats, 0, sizeof(psock_stats));
+
+	mutex_lock(&knet->mutex);
+
+	aggregate_mux_stats(&knet->aggregate_mux_stats, &mux_stats);
+	aggregate_psock_stats(&knet->aggregate_psock_stats,
+			      &psock_stats);
+
+	list_for_each_entry_rcu(mux, &knet->mux_list, kcm_mux_list) {
+		spin_lock_bh(&mux->lock);
+		aggregate_mux_stats(&mux->stats, &mux_stats);
+		aggregate_psock_stats(&mux->aggregate_psock_stats,
+				      &psock_stats);
+		list_for_each_entry(psock, &mux->psocks, psock_list)
+			aggregate_psock_stats(&psock->stats, &psock_stats);
+		spin_unlock_bh(&mux->lock);
+	}
+
+	mutex_unlock(&knet->mutex);
+
+	seq_printf(seq,
+		   "%-8s %-10s %-16s %-10s %-16s %-10s %-10s %-10s %-10s %-10s\n",
+		   "MUX",
+		   "RX-Msgs",
+		   "RX-Bytes",
+		   "TX-Msgs",
+		   "TX-Bytes",
+		   "TX-Retries",
+		   "Attach",
+		   "Unattach",
+		   "UnattchRsvd",
+		   "RX-RdyDrops");
+
+	seq_printf(seq,
+		   "%-8s %-10llu %-16llu %-10llu %-16llu %-10u %-10u %-10u %-10u %-10u\n",
+		   "",
+		   mux_stats.rx_msgs,
+		   mux_stats.rx_bytes,
+		   mux_stats.tx_msgs,
+		   mux_stats.tx_bytes,
+		   mux_stats.tx_retries,
+		   mux_stats.psock_attach,
+		   mux_stats.psock_unattach_rsvd,
+		   mux_stats.psock_unattach,
+		   mux_stats.rx_ready_drops);
+
+	seq_printf(seq,
+		   "%-8s %-10s %-16s %-10s %-16s %-10s %-10s %-10s %-10s %-10s %-10s %-10s\n",
+		   "Psock",
+		   "RX-Msgs",
+		   "RX-Bytes",
+		   "TX-Msgs",
+		   "TX-Bytes",
+		   "Reserved",
+		   "Unreserved",
+		   "RX-Aborts",
+		   "RX-MemFail",
+		   "RX-NeedMor",
+		   "RX-BadLen",
+		   "TX-Aborts");
+
+	seq_printf(seq,
+		   "%-8s %-10llu %-16llu %-10llu %-16llu %-10llu %-10llu %-10u %-10u %-10u %-10u %-10u\n",
+		   "",
+		   psock_stats.rx_msgs,
+		   psock_stats.rx_bytes,
+		   psock_stats.tx_msgs,
+		   psock_stats.tx_bytes,
+		   psock_stats.reserved,
+		   psock_stats.unreserved,
+		   psock_stats.rx_aborts,
+		   psock_stats.rx_mem_fail,
+		   psock_stats.rx_need_more_hdr,
+		   psock_stats.rx_bad_hdr_len,
+		   psock_stats.tx_aborts);
+
+	return 0;
+}
+
+static int kcm_stats_seq_open(struct inode *inode, struct file *file)
+{
+	return single_open_net(inode, file, kcm_stats_seq_show);
+}
+
+static const struct file_operations kcm_stats_seq_fops = {
+	.owner   = THIS_MODULE,
+	.open    = kcm_stats_seq_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release_net,
+};
+
+static int kcm_proc_init_net(struct net *net)
+{
+	int err;
+
+	if (!proc_create("kcm_stats", S_IRUGO, net->proc_net,
+			 &kcm_stats_seq_fops)) {
+		err = -ENOMEM;
+		goto out_kcm_stats;
+	}
+
+	err = kcm_proc_register(net, &kcm_seq_muxinfo);
+	if (err)
+		goto out_kcm;
+
+	return 0;
+
+out_kcm:
+	remove_proc_entry("kcm_stats", net->proc_net);
+out_kcm_stats:
+	return err;
+}
+
+static void kcm_proc_exit_net(struct net *net)
+{
+	kcm_proc_unregister(net, &kcm_seq_muxinfo);
+	remove_proc_entry("kcm_stats", net->proc_net);
+}
+
+static struct pernet_operations kcm_net_ops = {
+	.init = kcm_proc_init_net,
+	.exit = kcm_proc_exit_net,
+};
+
+int __init kcm_proc_init(void)
+{
+	return register_pernet_subsys(&kcm_net_ops);
+}
+
+void __exit kcm_proc_exit(void)
+{
+	unregister_pernet_subsys(&kcm_net_ops);
+}
+
+#endif /* CONFIG_PROC_FS */
diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c
index 30ef69a..f938d7d 100644
--- a/net/kcm/kcmsock.c
+++ b/net/kcm/kcmsock.c
@@ -59,6 +59,7 @@ static void kcm_abort_rx_psock(struct kcm_psock *psock, int err,
 		return;
 
 	psock->rx_stopped = 1;
+	KCM_STATS_INCR(psock->stats.rx_aborts);
 
 	/* Report an error on the lower socket */
 	report_csk_error(csk, err);
@@ -80,6 +81,7 @@ static void kcm_abort_tx_psock(struct kcm_psock *psock, int err,
 	}
 
 	psock->tx_stopped = 1;
+	KCM_STATS_INCR(psock->stats.tx_aborts);
 
 	if (!psock->tx_kcm) {
 		/* Take off psocks_avail list */
@@ -101,6 +103,29 @@ static void kcm_abort_tx_psock(struct kcm_psock *psock, int err,
 	report_csk_error(csk, err);
 }
 
+/* RX mux lock held. */
+static void kcm_update_rx_mux_stats(struct kcm_mux *mux,
+				    struct kcm_psock *psock)
+{
+	KCM_STATS_ADD(mux->stats.rx_bytes,
+		      psock->stats.rx_bytes - psock->saved_rx_bytes);
+	mux->stats.rx_msgs +=
+		psock->stats.rx_msgs - psock->saved_rx_msgs;
+	psock->saved_rx_msgs = psock->stats.rx_msgs;
+	psock->saved_rx_bytes = psock->stats.rx_bytes;
+}
+
+static void kcm_update_tx_mux_stats(struct kcm_mux *mux,
+				    struct kcm_psock *psock)
+{
+	KCM_STATS_ADD(mux->stats.tx_bytes,
+		      psock->stats.tx_bytes - psock->saved_tx_bytes);
+	mux->stats.tx_msgs +=
+		psock->stats.tx_msgs - psock->saved_tx_msgs;
+	psock->saved_tx_msgs = psock->stats.tx_msgs;
+	psock->saved_tx_bytes = psock->stats.tx_bytes;
+}
+
 static int kcm_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
 
 /* KCM is ready to receive messages on its queue-- either the KCM is new or
@@ -254,6 +279,8 @@ static struct kcm_sock *reserve_rx_kcm(struct kcm_psock *psock,
 		return psock->rx_kcm;
 	}
 
+	kcm_update_rx_mux_stats(mux, psock);
+
 	if (list_empty(&mux->kcm_rx_waiters)) {
 		psock->ready_rx_msg = head;
 		list_add_tail(&psock->psock_ready_list,
@@ -356,10 +383,12 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 			 */
 			orig_skb = skb_clone(orig_skb, GFP_ATOMIC);
 			if (!orig_skb) {
+				KCM_STATS_INCR(psock->stats.rx_mem_fail);
 				desc->error = -ENOMEM;
 				return 0;
 			}
 			if (!pskb_pull(orig_skb, orig_offset)) {
+				KCM_STATS_INCR(psock->stats.rx_mem_fail);
 				kfree_skb(orig_skb);
 				desc->error = -ENOMEM;
 				return 0;
@@ -374,6 +403,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 			 */
 			err = skb_unclone(head, GFP_ATOMIC);
 			if (err) {
+				KCM_STATS_INCR(psock->stats.rx_mem_fail);
 				desc->error = err;
 				return 0;
 			}
@@ -392,6 +422,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 
 				skb = alloc_skb(0, GFP_ATOMIC);
 				if (!skb) {
+					KCM_STATS_INCR(psock->stats.rx_mem_fail);
 					desc->error = -ENOMEM;
 					return 0;
 				}
@@ -414,6 +445,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 		/* Always clone since we will consume something */
 		skb = skb_clone(orig_skb, GFP_ATOMIC);
 		if (!skb) {
+			KCM_STATS_INCR(psock->stats.rx_mem_fail);
 			desc->error = -ENOMEM;
 			break;
 		}
@@ -435,6 +467,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 			 */
 			err = skb_unclone(skb, GFP_ATOMIC);
 			if (err) {
+				KCM_STATS_INCR(psock->stats.rx_mem_fail);
 				desc->error = err;
 				break;
 			}
@@ -456,6 +489,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 				/* Need more header to determine length */
 				rxm->accum_len += cand_len;
 				eaten += cand_len;
+				KCM_STATS_INCR(psock->stats.rx_need_more_hdr);
 				WARN_ON(eaten != orig_len);
 				break;
 			} else if (len <= (ssize_t)head->len -
@@ -463,6 +497,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 				/* Length must be into new skb (and also
 				 * greater than zero)
 				 */
+				KCM_STATS_INCR(psock->stats.rx_bad_hdr_len);
 				desc->error = -EPROTO;
 				psock->rx_skb_head = NULL;
 				kcm_abort_rx_psock(psock, EPROTO, head);
@@ -492,6 +527,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 
 		/* Hurray, we have a new message! */
 		psock->rx_skb_head = NULL;
+		KCM_STATS_INCR(psock->stats.rx_msgs);
 
 try_queue:
 		kcm = reserve_rx_kcm(psock, head);
@@ -510,6 +546,8 @@ try_queue:
 	if (cloned_orig)
 		kfree_skb(orig_skb);
 
+	KCM_STATS_ADD(psock->stats.rx_bytes, eaten);
+
 	return eaten;
 }
 
@@ -671,6 +709,7 @@ static struct kcm_psock *reserve_psock(struct kcm_sock *kcm)
 		}
 		kcm->tx_psock = psock;
 		psock->tx_kcm = kcm;
+		KCM_STATS_INCR(psock->stats.reserved);
 	} else if (!kcm->tx_wait) {
 		list_add_tail(&kcm->wait_psock_list,
 			      &mux->kcm_tx_waiters);
@@ -705,6 +744,7 @@ static void psock_now_avail(struct kcm_psock *psock)
 		smp_mb();
 
 		kcm->tx_psock = psock;
+		KCM_STATS_INCR(psock->stats.reserved);
 		queue_work(kcm_wq, &kcm->tx_work);
 	}
 }
@@ -726,10 +766,13 @@ static void unreserve_psock(struct kcm_sock *kcm)
 
 	smp_rmb(); /* Read tx_psock before tx_wait */
 
+	kcm_update_tx_mux_stats(mux, psock);
+
 	WARN_ON(kcm->tx_wait);
 
 	kcm->tx_psock = NULL;
 	psock->tx_kcm = NULL;
+	KCM_STATS_INCR(psock->stats.unreserved);
 
 	if (unlikely(psock->tx_stopped)) {
 		if (psock->done) {
@@ -753,6 +796,15 @@ static void unreserve_psock(struct kcm_sock *kcm)
 	spin_unlock_bh(&mux->lock);
 }
 
+static void kcm_report_tx_retry(struct kcm_sock *kcm)
+{
+	struct kcm_mux *mux = kcm->mux;
+
+	spin_lock_bh(&mux->lock);
+	KCM_STATS_INCR(mux->stats.tx_retries);
+	spin_unlock_bh(&mux->lock);
+}
+
 /* Write any messages ready on the kcm socket.  Called with kcm sock lock
  * held.  Return bytes actually sent or error.
  */
@@ -773,6 +825,7 @@ static int kcm_write_msgs(struct kcm_sock *kcm)
 		 * it and we'll retry the message.
 		 */
 		unreserve_psock(kcm);
+		kcm_report_tx_retry(kcm);
 		if (skb_queue_empty(&sk->sk_write_queue))
 			return 0;
 
@@ -856,6 +909,7 @@ do_frag:
 				unreserve_psock(kcm);
 
 				txm->sent = 0;
+				kcm_report_tx_retry(kcm);
 				ret = 0;
 
 				goto try_again;
@@ -863,6 +917,7 @@ do_frag:
 
 			sent += ret;
 			frag_offset += ret;
+			KCM_STATS_ADD(psock->stats.tx_bytes, ret);
 			if (frag_offset < frag->size) {
 				/* Not finished with this frag */
 				goto do_frag;
@@ -884,6 +939,7 @@ do_frag:
 		kfree_skb(head);
 		sk->sk_wmem_queued -= sent;
 		total_sent += sent;
+		KCM_STATS_INCR(psock->stats.tx_msgs);
 	} while ((head = skb_peek(&sk->sk_write_queue)));
 out:
 	if (!head) {
@@ -1061,6 +1117,7 @@ wait_for_memory:
 		/* Message complete, queue it on send buffer */
 		__skb_queue_tail(&sk->sk_write_queue, head);
 		kcm->seq_skb = NULL;
+		KCM_STATS_INCR(kcm->stats.tx_msgs);
 
 		if (msg->msg_flags & MSG_BATCH) {
 			kcm->tx_wait_more = true;
@@ -1083,6 +1140,8 @@ partial_message:
 		kcm_tx_msg(head)->last_skb = skb;
 	}
 
+	KCM_STATS_ADD(kcm->stats.tx_bytes, copied);
+
 	release_sock(sk);
 	return copied;
 
@@ -1144,6 +1203,7 @@ static int kcm_recvmsg(struct socket *sock, struct msghdr *msg,
 		       size_t len, int flags)
 {
 	struct sock *sk = sock->sk;
+	struct kcm_sock *kcm = kcm_sk(sk);
 	int err = 0;
 	long timeo;
 	struct kcm_rx_msg *rxm;
@@ -1171,6 +1231,7 @@ static int kcm_recvmsg(struct socket *sock, struct msghdr *msg,
 
 	copied = len;
 	if (likely(!(flags & MSG_PEEK))) {
+		KCM_STATS_ADD(kcm->stats.rx_bytes, copied);
 		if (copied < rxm->full_len) {
 			if (sock->type == SOCK_DGRAM) {
 				/* Truncated message */
@@ -1183,6 +1244,7 @@ static int kcm_recvmsg(struct socket *sock, struct msghdr *msg,
 msg_finished:
 			/* Finished with message */
 			msg->msg_flags |= MSG_EOR;
+			KCM_STATS_INCR(kcm->stats.rx_msgs);
 			skb_unlink(skb, &sk->sk_receive_queue);
 			kfree_skb(skb);
 		}
@@ -1394,6 +1456,7 @@ static int kcm_attach(struct socket *sock, struct socket *csock,
 	list_add(&psock->psock_list, head);
 	psock->index = index;
 
+	KCM_STATS_INCR(mux->stats.psock_attach);
 	mux->psocks_cnt++;
 	psock_now_avail(psock);
 	spin_unlock_bh(&mux->lock);
@@ -1469,6 +1532,7 @@ static void kcm_unattach(struct kcm_psock *psock)
 		list_del(&psock->psock_ready_list);
 		kfree_skb(psock->ready_rx_msg);
 		psock->ready_rx_msg = NULL;
+		KCM_STATS_INCR(mux->stats.rx_ready_drops);
 	}
 
 	spin_unlock_bh(&mux->rx_lock);
@@ -1485,11 +1549,16 @@ static void kcm_unattach(struct kcm_psock *psock)
 
 	spin_lock_bh(&mux->lock);
 
+	aggregate_psock_stats(&psock->stats, &mux->aggregate_psock_stats);
+
+	KCM_STATS_INCR(mux->stats.psock_unattach);
+
 	if (psock->tx_kcm) {
 		/* psock was reserved.  Just mark it finished and we will clean
 		 * up in the kcm paths, we need kcm lock which can not be
 		 * acquired here.
 		 */
+		KCM_STATS_INCR(mux->stats.psock_unattach_rsvd);
 		spin_unlock_bh(&mux->lock);
 
 		/* We are unattaching a socket that is reserved. Abort the
@@ -1717,6 +1786,9 @@ static void release_mux(struct kcm_mux *mux)
 	__skb_queue_purge(&mux->rx_hold_queue);
 
 	mutex_lock(&knet->mutex);
+	aggregate_mux_stats(&mux->stats, &knet->aggregate_mux_stats);
+	aggregate_psock_stats(&mux->aggregate_psock_stats,
+			      &knet->aggregate_psock_stats);
 	list_del_rcu(&mux->kcm_mux_list);
 	knet->count--;
 	mutex_unlock(&knet->mutex);
@@ -1979,8 +2051,15 @@ static int __init kcm_init(void)
 	if (err)
 		goto net_ops_fail;
 
+	err = kcm_proc_init();
+	if (err)
+		goto proc_init_fail;
+
 	return 0;
 
+proc_init_fail:
+	unregister_pernet_device(&kcm_net_ops);
+
 net_ops_fail:
 	sock_unregister(PF_KCM);
 
@@ -1999,6 +2078,7 @@ fail:
 
 static void __exit kcm_exit(void)
 {
+	kcm_proc_exit();
 	unregister_pernet_device(&kcm_net_ops);
 	sock_unregister(PF_KCM);
 	proto_unregister(&kcm_proto);
-- 
cgit v0.10.2


From 91687355b92735e5f247ed163b3b0b4d14c3cab6 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:08 -0800
Subject: kcm: Splice support

Implement kcm_splice_read. This is supported only for seqpacket.
Add kcm_seqpacket_ops and set splice read to kcm_splice_read.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c
index f938d7d..982ea5f 100644
--- a/net/kcm/kcmsock.c
+++ b/net/kcm/kcmsock.c
@@ -1256,6 +1256,76 @@ out:
 	return copied ? : err;
 }
 
+static ssize_t kcm_sock_splice(struct sock *sk,
+			       struct pipe_inode_info *pipe,
+			       struct splice_pipe_desc *spd)
+{
+	int ret;
+
+	release_sock(sk);
+	ret = splice_to_pipe(pipe, spd);
+	lock_sock(sk);
+
+	return ret;
+}
+
+static ssize_t kcm_splice_read(struct socket *sock, loff_t *ppos,
+			       struct pipe_inode_info *pipe, size_t len,
+			       unsigned int flags)
+{
+	struct sock *sk = sock->sk;
+	struct kcm_sock *kcm = kcm_sk(sk);
+	long timeo;
+	struct kcm_rx_msg *rxm;
+	int err = 0;
+	size_t copied;
+	struct sk_buff *skb;
+
+	/* Only support splice for SOCKSEQPACKET */
+
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+
+	lock_sock(sk);
+
+	skb = kcm_wait_data(sk, flags, timeo, &err);
+	if (!skb)
+		goto err_out;
+
+	/* Okay, have a message on the receive queue */
+
+	rxm = kcm_rx_msg(skb);
+
+	if (len > rxm->full_len)
+		len = rxm->full_len;
+
+	copied = skb_splice_bits(skb, sk, rxm->offset, pipe, len, flags,
+				 kcm_sock_splice);
+	if (copied < 0) {
+		err = copied;
+		goto err_out;
+	}
+
+	KCM_STATS_ADD(kcm->stats.rx_bytes, copied);
+
+	rxm->offset += copied;
+	rxm->full_len -= copied;
+
+	/* We have no way to return MSG_EOR. If all the bytes have been
+	 * read we still leave the message in the receive socket buffer.
+	 * A subsequent recvmsg needs to be done to return MSG_EOR and
+	 * finish reading the message.
+	 */
+
+	release_sock(sk);
+
+	return copied;
+
+err_out:
+	release_sock(sk);
+
+	return err;
+}
+
 /* kcm sock lock held */
 static void kcm_recv_disable(struct kcm_sock *kcm)
 {
@@ -1907,7 +1977,7 @@ static int kcm_release(struct socket *sock)
 	return 0;
 }
 
-static const struct proto_ops kcm_ops = {
+static const struct proto_ops kcm_dgram_ops = {
 	.family =	PF_KCM,
 	.owner =	THIS_MODULE,
 	.release =	kcm_release,
@@ -1928,6 +1998,28 @@ static const struct proto_ops kcm_ops = {
 	.sendpage =	sock_no_sendpage,
 };
 
+static const struct proto_ops kcm_seqpacket_ops = {
+	.family =	PF_KCM,
+	.owner =	THIS_MODULE,
+	.release =	kcm_release,
+	.bind =		sock_no_bind,
+	.connect =	sock_no_connect,
+	.socketpair =	sock_no_socketpair,
+	.accept =	sock_no_accept,
+	.getname =	sock_no_getname,
+	.poll =		datagram_poll,
+	.ioctl =	kcm_ioctl,
+	.listen =	sock_no_listen,
+	.shutdown =	sock_no_shutdown,
+	.setsockopt =	kcm_setsockopt,
+	.getsockopt =	kcm_getsockopt,
+	.sendmsg =	kcm_sendmsg,
+	.recvmsg =	kcm_recvmsg,
+	.mmap =		sock_no_mmap,
+	.sendpage =	sock_no_sendpage,
+	.splice_read =	kcm_splice_read,
+};
+
 /* Create proto operation for kcm sockets */
 static int kcm_create(struct net *net, struct socket *sock,
 		      int protocol, int kern)
@@ -1938,8 +2030,10 @@ static int kcm_create(struct net *net, struct socket *sock,
 
 	switch (sock->type) {
 	case SOCK_DGRAM:
+		sock->ops = &kcm_dgram_ops;
+		break;
 	case SOCK_SEQPACKET:
-		sock->ops = &kcm_ops;
+		sock->ops = &kcm_seqpacket_ops;
 		break;
 	default:
 		return -ESOCKTNOSUPPORT;
-- 
cgit v0.10.2


From f29698fc6b3a45a5c6147eca8379f38be8232117 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:09 -0800
Subject: kcm: Sendpage support

Implement kcm_sendpage. Set in sendpage to kcm_sendpage in both
dgram and seqpacket ops.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c
index 982ea5f..9ac2499 100644
--- a/net/kcm/kcmsock.c
+++ b/net/kcm/kcmsock.c
@@ -990,6 +990,149 @@ static void kcm_push(struct kcm_sock *kcm)
 		kcm_write_msgs(kcm);
 }
 
+static ssize_t kcm_sendpage(struct socket *sock, struct page *page,
+			    int offset, size_t size, int flags)
+
+{
+	struct sock *sk = sock->sk;
+	struct kcm_sock *kcm = kcm_sk(sk);
+	struct sk_buff *skb = NULL, *head = NULL;
+	long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
+	bool eor;
+	int err = 0;
+	int i;
+
+	if (flags & MSG_SENDPAGE_NOTLAST)
+		flags |= MSG_MORE;
+
+	/* No MSG_EOR from splice, only look at MSG_MORE */
+	eor = !(flags & MSG_MORE);
+
+	lock_sock(sk);
+
+	sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
+
+	err = -EPIPE;
+	if (sk->sk_err)
+		goto out_error;
+
+	if (kcm->seq_skb) {
+		/* Previously opened message */
+		head = kcm->seq_skb;
+		skb = kcm_tx_msg(head)->last_skb;
+		i = skb_shinfo(skb)->nr_frags;
+
+		if (skb_can_coalesce(skb, i, page, offset)) {
+			skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], size);
+			skb_shinfo(skb)->tx_flags |= SKBTX_SHARED_FRAG;
+			goto coalesced;
+		}
+
+		if (i >= MAX_SKB_FRAGS) {
+			struct sk_buff *tskb;
+
+			tskb = alloc_skb(0, sk->sk_allocation);
+			while (!tskb) {
+				kcm_push(kcm);
+				err = sk_stream_wait_memory(sk, &timeo);
+				if (err)
+					goto out_error;
+			}
+
+			if (head == skb)
+				skb_shinfo(head)->frag_list = tskb;
+			else
+				skb->next = tskb;
+
+			skb = tskb;
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			i = 0;
+		}
+	} else {
+		/* Call the sk_stream functions to manage the sndbuf mem. */
+		if (!sk_stream_memory_free(sk)) {
+			kcm_push(kcm);
+			set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+			err = sk_stream_wait_memory(sk, &timeo);
+			if (err)
+				goto out_error;
+		}
+
+		head = alloc_skb(0, sk->sk_allocation);
+		while (!head) {
+			kcm_push(kcm);
+			err = sk_stream_wait_memory(sk, &timeo);
+			if (err)
+				goto out_error;
+		}
+
+		skb = head;
+		i = 0;
+	}
+
+	get_page(page);
+	skb_fill_page_desc(skb, i, page, offset, size);
+	skb_shinfo(skb)->tx_flags |= SKBTX_SHARED_FRAG;
+
+coalesced:
+	skb->len += size;
+	skb->data_len += size;
+	skb->truesize += size;
+	sk->sk_wmem_queued += size;
+	sk_mem_charge(sk, size);
+
+	if (head != skb) {
+		head->len += size;
+		head->data_len += size;
+		head->truesize += size;
+	}
+
+	if (eor) {
+		bool not_busy = skb_queue_empty(&sk->sk_write_queue);
+
+		/* Message complete, queue it on send buffer */
+		__skb_queue_tail(&sk->sk_write_queue, head);
+		kcm->seq_skb = NULL;
+		KCM_STATS_INCR(kcm->stats.tx_msgs);
+
+		if (flags & MSG_BATCH) {
+			kcm->tx_wait_more = true;
+		} else if (kcm->tx_wait_more || not_busy) {
+			err = kcm_write_msgs(kcm);
+			if (err < 0) {
+				/* We got a hard error in write_msgs but have
+				 * already queued this message. Report an error
+				 * in the socket, but don't affect return value
+				 * from sendmsg
+				 */
+				pr_warn("KCM: Hard failure on kcm_write_msgs\n");
+				report_csk_error(&kcm->sk, -err);
+			}
+		}
+	} else {
+		/* Message not complete, save state */
+		kcm->seq_skb = head;
+		kcm_tx_msg(head)->last_skb = skb;
+	}
+
+	KCM_STATS_ADD(kcm->stats.tx_bytes, size);
+
+	release_sock(sk);
+	return size;
+
+out_error:
+	kcm_push(kcm);
+
+	err = sk_stream_error(sk, flags, err);
+
+	/* make sure we wake any epoll edge trigger waiter */
+	if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && err == -EAGAIN))
+		sk->sk_write_space(sk);
+
+	release_sock(sk);
+	return err;
+}
+
 static int kcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 {
 	struct sock *sk = sock->sk;
@@ -1995,7 +2138,7 @@ static const struct proto_ops kcm_dgram_ops = {
 	.sendmsg =	kcm_sendmsg,
 	.recvmsg =	kcm_recvmsg,
 	.mmap =		sock_no_mmap,
-	.sendpage =	sock_no_sendpage,
+	.sendpage =	kcm_sendpage,
 };
 
 static const struct proto_ops kcm_seqpacket_ops = {
@@ -2016,7 +2159,7 @@ static const struct proto_ops kcm_seqpacket_ops = {
 	.sendmsg =	kcm_sendmsg,
 	.recvmsg =	kcm_recvmsg,
 	.mmap =		sock_no_mmap,
-	.sendpage =	sock_no_sendpage,
+	.sendpage =	kcm_sendpage,
 	.splice_read =	kcm_splice_read,
 };
 
-- 
cgit v0.10.2


From 7ced95ef525c329f947c424859cf2b0a3b731f8c Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:10 -0800
Subject: kcm: Add memory limit for receive message construction

Message assembly is performed on the TCP socket. This is logically
equivalent of an application that performs a peek on the socket to find
out how much memory is needed for a receive buffer. The receive socket
buffer also provides the maximum message size which is checked.

The receive algorithm is something like:

   1) Receive the first skbuf for a message (or skbufs if multiple are
      needed to determine message length).
   2) Check the message length against the number of bytes in the TCP
      receive queue (tcp_inq()).
	- If all the bytes of the message are in the queue (incluing the
	  skbuf received), then proceed with message assembly (it should
	  complete with the tcp_read_sock)
        - Else, mark the psock with the number of bytes needed to
	  complete the message.
   3) In TCP data ready function, if the psock indicates that we are
      waiting for the rest of the bytes of a messages, check the number
      of queued bytes against that.
        - If there are still not enough bytes for the message, just
	  return
        - Else, clear the waiting bytes and proceed to receive the
	  skbufs.  The message should now be received in one
	  tcp_read_sock

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/kcm.h b/include/net/kcm.h
index 39c7abe..d892956 100644
--- a/include/net/kcm.h
+++ b/include/net/kcm.h
@@ -28,6 +28,7 @@ struct kcm_psock_stats {
 	unsigned int rx_aborts;
 	unsigned int rx_mem_fail;
 	unsigned int rx_need_more_hdr;
+	unsigned int rx_msg_too_big;
 	unsigned int rx_bad_hdr_len;
 	unsigned long long reserved;
 	unsigned long long unreserved;
@@ -66,6 +67,7 @@ struct kcm_rx_msg {
 	int full_len;
 	int accum_len;
 	int offset;
+	int early_eaten;
 };
 
 /* Socket structure for KCM client sockets */
@@ -128,6 +130,7 @@ struct kcm_psock {
 	struct kcm_sock *rx_kcm;
 	unsigned long long saved_rx_bytes;
 	unsigned long long saved_rx_msgs;
+	unsigned int rx_need_bytes;
 
 	/* Transmit */
 	struct kcm_sock *tx_kcm;
@@ -190,6 +193,7 @@ static inline void aggregate_psock_stats(struct kcm_psock_stats *stats,
 	SAVE_PSOCK_STATS(rx_aborts);
 	SAVE_PSOCK_STATS(rx_mem_fail);
 	SAVE_PSOCK_STATS(rx_need_more_hdr);
+	SAVE_PSOCK_STATS(rx_msg_too_big);
 	SAVE_PSOCK_STATS(rx_bad_hdr_len);
 	SAVE_PSOCK_STATS(tx_msgs);
 	SAVE_PSOCK_STATS(tx_bytes);
diff --git a/net/kcm/kcmproc.c b/net/kcm/kcmproc.c
index 5eb9809..7638b35 100644
--- a/net/kcm/kcmproc.c
+++ b/net/kcm/kcmproc.c
@@ -331,7 +331,7 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v)
 		   mux_stats.rx_ready_drops);
 
 	seq_printf(seq,
-		   "%-8s %-10s %-16s %-10s %-16s %-10s %-10s %-10s %-10s %-10s %-10s %-10s\n",
+		   "%-8s %-10s %-16s %-10s %-16s %-10s %-10s %-10s %-10s %-10s %-10s %-10s %-10s\n",
 		   "Psock",
 		   "RX-Msgs",
 		   "RX-Bytes",
@@ -343,10 +343,11 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v)
 		   "RX-MemFail",
 		   "RX-NeedMor",
 		   "RX-BadLen",
+		   "RX-TooBig",
 		   "TX-Aborts");
 
 	seq_printf(seq,
-		   "%-8s %-10llu %-16llu %-10llu %-16llu %-10llu %-10llu %-10u %-10u %-10u %-10u %-10u\n",
+		   "%-8s %-10llu %-16llu %-10llu %-16llu %-10llu %-10llu %-10u %-10u %-10u %-10u %-10u %-10u\n",
 		   "",
 		   psock_stats.rx_msgs,
 		   psock_stats.rx_bytes,
@@ -358,6 +359,7 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v)
 		   psock_stats.rx_mem_fail,
 		   psock_stats.rx_need_more_hdr,
 		   psock_stats.rx_bad_hdr_len,
+		   psock_stats.rx_msg_too_big,
 		   psock_stats.tx_aborts);
 
 	return 0;
diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c
index 9ac2499..8bc38d3 100644
--- a/net/kcm/kcmsock.c
+++ b/net/kcm/kcmsock.c
@@ -375,6 +375,19 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 	if (head) {
 		/* Message already in progress */
 
+		rxm = kcm_rx_msg(head);
+		if (unlikely(rxm->early_eaten)) {
+			/* Already some number of bytes on the receive sock
+			 * data saved in rx_skb_head, just indicate they
+			 * are consumed.
+			 */
+			eaten = orig_len <= rxm->early_eaten ?
+				orig_len : rxm->early_eaten;
+			rxm->early_eaten -= eaten;
+
+			return eaten;
+		}
+
 		if (unlikely(orig_offset)) {
 			/* Getting data with a non-zero offset when a message is
 			 * in progress is not expected. If it does happen, we
@@ -492,6 +505,13 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 				KCM_STATS_INCR(psock->stats.rx_need_more_hdr);
 				WARN_ON(eaten != orig_len);
 				break;
+			} else if (len > psock->sk->sk_rcvbuf) {
+				/* Message length exceeds maximum allowed */
+				KCM_STATS_INCR(psock->stats.rx_msg_too_big);
+				desc->error = -EMSGSIZE;
+				psock->rx_skb_head = NULL;
+				kcm_abort_rx_psock(psock, EMSGSIZE, head);
+				break;
 			} else if (len <= (ssize_t)head->len -
 					  skb->len - rxm->offset) {
 				/* Length must be into new skb (and also
@@ -511,6 +531,23 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 
 		if (extra < 0) {
 			/* Message not complete yet. */
+			if (rxm->full_len - rxm->accum_len >
+			    tcp_inq(psock->sk)) {
+				/* Don't have the whole messages in the socket
+				 * buffer. Set psock->rx_need_bytes to wait for
+				 * the rest of the message. Also, set "early
+				 * eaten" since we've already buffered the skb
+				 * but don't consume yet per tcp_read_sock.
+				 */
+
+				psock->rx_need_bytes = rxm->full_len -
+						       rxm->accum_len;
+				rxm->accum_len += cand_len;
+				rxm->early_eaten = cand_len;
+				KCM_STATS_ADD(psock->stats.rx_bytes, cand_len);
+				desc->count = 0; /* Stop reading socket */
+				break;
+			}
 			rxm->accum_len += cand_len;
 			eaten += cand_len;
 			WARN_ON(eaten != orig_len);
@@ -582,6 +619,13 @@ static void psock_tcp_data_ready(struct sock *sk)
 	if (psock->ready_rx_msg)
 		goto out;
 
+	if (psock->rx_need_bytes) {
+		if (tcp_inq(sk) >= psock->rx_need_bytes)
+			psock->rx_need_bytes = 0;
+		else
+			goto out;
+	}
+
 	if (psock_tcp_read_sock(psock) == -ENOMEM)
 		queue_delayed_work(kcm_wq, &psock->rx_delayed_work, 0);
 
-- 
cgit v0.10.2


From 29152a34f72cb4d7ab32885ad2f20a482c92a8f3 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:11 -0800
Subject: kcm: Add receive message timeout

This patch adds receive timeout for message assembly on the attached TCP
sockets. The timeout is set when a new messages is started and the whole
message has not been received by TCP (not in the receive queue). If the
completely message is subsequently received the timer is cancelled, if the
timer expires the RX side is aborted.

The timeout value is taken from the socket timeout (SO_RCVTIMEO) that is
set on a TCP socket (i.e. set by get sockopt before attaching a TCP socket
to KCM.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/kcm.h b/include/net/kcm.h
index d892956..95c425c 100644
--- a/include/net/kcm.h
+++ b/include/net/kcm.h
@@ -29,6 +29,7 @@ struct kcm_psock_stats {
 	unsigned int rx_mem_fail;
 	unsigned int rx_need_more_hdr;
 	unsigned int rx_msg_too_big;
+	unsigned int rx_msg_timeouts;
 	unsigned int rx_bad_hdr_len;
 	unsigned long long reserved;
 	unsigned long long unreserved;
@@ -130,6 +131,7 @@ struct kcm_psock {
 	struct kcm_sock *rx_kcm;
 	unsigned long long saved_rx_bytes;
 	unsigned long long saved_rx_msgs;
+	struct timer_list rx_msg_timer;
 	unsigned int rx_need_bytes;
 
 	/* Transmit */
@@ -194,6 +196,7 @@ static inline void aggregate_psock_stats(struct kcm_psock_stats *stats,
 	SAVE_PSOCK_STATS(rx_mem_fail);
 	SAVE_PSOCK_STATS(rx_need_more_hdr);
 	SAVE_PSOCK_STATS(rx_msg_too_big);
+	SAVE_PSOCK_STATS(rx_msg_timeouts);
 	SAVE_PSOCK_STATS(rx_bad_hdr_len);
 	SAVE_PSOCK_STATS(tx_msgs);
 	SAVE_PSOCK_STATS(tx_bytes);
diff --git a/net/kcm/kcmproc.c b/net/kcm/kcmproc.c
index 7638b35..7380087 100644
--- a/net/kcm/kcmproc.c
+++ b/net/kcm/kcmproc.c
@@ -331,7 +331,7 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v)
 		   mux_stats.rx_ready_drops);
 
 	seq_printf(seq,
-		   "%-8s %-10s %-16s %-10s %-16s %-10s %-10s %-10s %-10s %-10s %-10s %-10s %-10s\n",
+		   "%-8s %-10s %-16s %-10s %-16s %-10s %-10s %-10s %-10s %-10s %-10s %-10s %-10s %-10s\n",
 		   "Psock",
 		   "RX-Msgs",
 		   "RX-Bytes",
@@ -344,10 +344,11 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v)
 		   "RX-NeedMor",
 		   "RX-BadLen",
 		   "RX-TooBig",
+		   "RX-Timeout",
 		   "TX-Aborts");
 
 	seq_printf(seq,
-		   "%-8s %-10llu %-16llu %-10llu %-16llu %-10llu %-10llu %-10u %-10u %-10u %-10u %-10u %-10u\n",
+		   "%-8s %-10llu %-16llu %-10llu %-16llu %-10llu %-10llu %-10u %-10u %-10u %-10u %-10u %-10u %-10u\n",
 		   "",
 		   psock_stats.rx_msgs,
 		   psock_stats.rx_bytes,
@@ -360,6 +361,7 @@ static int kcm_stats_seq_show(struct seq_file *seq, void *v)
 		   psock_stats.rx_need_more_hdr,
 		   psock_stats.rx_bad_hdr_len,
 		   psock_stats.rx_msg_too_big,
+		   psock_stats.rx_msg_timeouts,
 		   psock_stats.tx_aborts);
 
 	return 0;
diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c
index 8bc38d3..40662d73 100644
--- a/net/kcm/kcmsock.c
+++ b/net/kcm/kcmsock.c
@@ -55,6 +55,8 @@ static void kcm_abort_rx_psock(struct kcm_psock *psock, int err,
 
 	/* Unrecoverable error in receive */
 
+	del_timer(&psock->rx_msg_timer);
+
 	if (psock->rx_stopped)
 		return;
 
@@ -351,6 +353,12 @@ static void unreserve_rx_kcm(struct kcm_psock *psock,
 	spin_unlock_bh(&mux->rx_lock);
 }
 
+static void kcm_start_rx_timer(struct kcm_psock *psock)
+{
+	if (psock->sk->sk_rcvtimeo)
+		mod_timer(&psock->rx_msg_timer, psock->sk->sk_rcvtimeo);
+}
+
 /* Macro to invoke filter function. */
 #define KCM_RUN_FILTER(prog, ctx) \
 	(*prog->bpf_func)(ctx, prog->insnsi)
@@ -500,6 +508,10 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 
 			if (!len) {
 				/* Need more header to determine length */
+				if (!rxm->accum_len) {
+					/* Start RX timer for new message */
+					kcm_start_rx_timer(psock);
+				}
 				rxm->accum_len += cand_len;
 				eaten += cand_len;
 				KCM_STATS_INCR(psock->stats.rx_need_more_hdr);
@@ -540,6 +552,11 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 				 * but don't consume yet per tcp_read_sock.
 				 */
 
+				if (!rxm->accum_len) {
+					/* Start RX timer for new message */
+					kcm_start_rx_timer(psock);
+				}
+
 				psock->rx_need_bytes = rxm->full_len -
 						       rxm->accum_len;
 				rxm->accum_len += cand_len;
@@ -563,6 +580,7 @@ static int kcm_tcp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
 		eaten += (cand_len - extra);
 
 		/* Hurray, we have a new message! */
+		del_timer(&psock->rx_msg_timer);
 		psock->rx_skb_head = NULL;
 		KCM_STATS_INCR(psock->stats.rx_msgs);
 
@@ -1656,6 +1674,15 @@ static void init_kcm_sock(struct kcm_sock *kcm, struct kcm_mux *mux)
 	spin_unlock_bh(&mux->rx_lock);
 }
 
+static void kcm_rx_msg_timeout(unsigned long arg)
+{
+	struct kcm_psock *psock = (struct kcm_psock *)arg;
+
+	/* Message assembly timed out */
+	KCM_STATS_INCR(psock->stats.rx_msg_timeouts);
+	kcm_abort_rx_psock(psock, ETIMEDOUT, NULL);
+}
+
 static int kcm_attach(struct socket *sock, struct socket *csock,
 		      struct bpf_prog *prog)
 {
@@ -1685,6 +1712,10 @@ static int kcm_attach(struct socket *sock, struct socket *csock,
 	psock->mux = mux;
 	psock->sk = csk;
 	psock->bpf_prog = prog;
+
+	setup_timer(&psock->rx_msg_timer, kcm_rx_msg_timeout,
+		    (unsigned long)psock);
+
 	INIT_WORK(&psock->rx_work, psock_rx_work);
 	INIT_DELAYED_WORK(&psock->rx_delayed_work, psock_rx_delayed_work);
 
@@ -1796,6 +1827,7 @@ static void kcm_unattach(struct kcm_psock *psock)
 
 	write_unlock_bh(&csk->sk_callback_lock);
 
+	del_timer_sync(&psock->rx_msg_timer);
 	cancel_work_sync(&psock->rx_work);
 	cancel_delayed_work_sync(&psock->rx_delayed_work);
 
-- 
cgit v0.10.2


From 10016594f4c6b3ef34c5de97d8ab62205d9d26a5 Mon Sep 17 00:00:00 2001
From: Tom Herbert <tom@herbertland.com>
Date: Mon, 7 Mar 2016 14:11:12 -0800
Subject: kcm: Add description in Documentation

Add kcm.txt to desribe KCM and interfaces.

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/kcm.txt b/Documentation/networking/kcm.txt
new file mode 100644
index 0000000..3476ede
--- /dev/null
+++ b/Documentation/networking/kcm.txt
@@ -0,0 +1,285 @@
+Kernel Connection Mulitplexor
+-----------------------------
+
+Kernel Connection Multiplexor (KCM) is a mechanism that provides a message based
+interface over TCP for generic application protocols. With KCM an application
+can efficiently send and receive application protocol messages over TCP using
+datagram sockets.
+
+KCM implements an NxM multiplexor in the kernel as diagrammed below:
+
++------------+   +------------+   +------------+   +------------+
+| KCM socket |   | KCM socket |   | KCM socket |   | KCM socket |
++------------+   +------------+   +------------+   +------------+
+      |                 |               |                |
+      +-----------+     |               |     +----------+
+                  |     |               |     |
+               +----------------------------------+
+               |           Multiplexor            |
+               +----------------------------------+
+                 |   |           |           |  |
+       +---------+   |           |           |  ------------+
+       |             |           |           |              |
++----------+  +----------+  +----------+  +----------+ +----------+
+|  Psock   |  |  Psock   |  |  Psock   |  |  Psock   | |  Psock   |
++----------+  +----------+  +----------+  +----------+ +----------+
+      |              |           |            |             |
++----------+  +----------+  +----------+  +----------+ +----------+
+| TCP sock |  | TCP sock |  | TCP sock |  | TCP sock | | TCP sock |
++----------+  +----------+  +----------+  +----------+ +----------+
+
+KCM sockets
+-----------
+
+The KCM sockets provide the user interface to the muliplexor. All the KCM sockets
+bound to a multiplexor are considered to have equivalent function, and I/O
+operations in different sockets may be done in parallel without the need for
+synchronization between threads in userspace.
+
+Multiplexor
+-----------
+
+The multiplexor provides the message steering. In the transmit path, messages
+written on a KCM socket are sent atomically on an appropriate TCP socket.
+Similarly, in the receive path, messages are constructed on each TCP socket
+(Psock) and complete messages are steered to a KCM socket.
+
+TCP sockets & Psocks
+--------------------
+
+TCP sockets may be bound to a KCM multiplexor. A Psock structure is allocated
+for each bound TCP socket, this structure holds the state for constructing
+messages on receive as well as other connection specific information for KCM.
+
+Connected mode semantics
+------------------------
+
+Each multiplexor assumes that all attached TCP connections are to the same
+destination and can use the different connections for load balancing when
+transmitting. The normal send and recv calls (include sendmmsg and recvmmsg)
+can be used to send and receive messages from the KCM socket.
+
+Socket types
+------------
+
+KCM supports SOCK_DGRAM and SOCK_SEQPACKET socket types.
+
+Message delineation
+-------------------
+
+Messages are sent over a TCP stream with some application protocol message
+format that typically includes a header which frames the messages. The length
+of a received message can be deduced from the application protocol header
+(often just a simple length field).
+
+A TCP stream must be parsed to determine message boundaries. Berkeley Packet
+Filter (BPF) is used for this. When attaching a TCP socket to a multiplexor a
+BPF program must be specified. The program is called at the start of receiving
+a new message and is given an skbuff that contains the bytes received so far.
+It parses the message header and returns the length of the message. Given this
+information, KCM will construct the message of the stated length and deliver it
+to a KCM socket.
+
+TCP socket management
+---------------------
+
+When a TCP socket is attached to a KCM multiplexor data ready (POLLIN) and
+write space available (POLLOUT) events are handled by the multiplexor. If there
+is a state change (disconnection) or other error on a TCP socket, an error is
+posted on the TCP socket so that a POLLERR event happens and KCM discontinues
+using the socket. When the application gets the error notification for a
+TCP socket, it should unattach the socket from KCM and then handle the error
+condition (the typical response is to close the socket and create a new
+connection if necessary).
+
+KCM limits the maximum receive message size to be the size of the receive
+socket buffer on the attached TCP socket (the socket buffer size can be set by
+SO_RCVBUF). If the length of a new message reported by the BPF program is
+greater than this limit a corresponding error (EMSGSIZE) is posted on the TCP
+socket. The BPF program may also enforce a maximum messages size and report an
+error when it is exceeded.
+
+A timeout may be set for assembling messages on a receive socket. The timeout
+value is taken from the receive timeout of the attached TCP socket (this is set
+by SO_RCVTIMEO). If the timer expires before assembly is complete an error
+(ETIMEDOUT) is posted on the socket.
+
+User interface
+==============
+
+Creating a multiplexor
+----------------------
+
+A new multiplexor and initial KCM socket is created by a socket call:
+
+  socket(AF_KCM, type, protocol)
+
+  - type is either SOCK_DGRAM or SOCK_SEQPACKET
+  - protocol is KCMPROTO_CONNECTED
+
+Cloning KCM sockets
+-------------------
+
+After the first KCM socket is created using the socket call as described
+above, additional sockets for the multiplexor can be created by cloning
+a KCM socket. This is accomplished by an ioctl on a KCM socket:
+
+  /* From linux/kcm.h */
+  struct kcm_clone {
+        int fd;
+  };
+
+  struct kcm_clone info;
+
+  memset(&info, 0, sizeof(info));
+
+  err = ioctl(kcmfd, SIOCKCMCLONE, &info);
+
+  if (!err)
+    newkcmfd = info.fd;
+
+Attach transport sockets
+------------------------
+
+Attaching of transport sockets to a multiplexor is performed by calling an
+ioctl on a KCM socket for the multiplexor. e.g.:
+
+  /* From linux/kcm.h */
+  struct kcm_attach {
+        int fd;
+	int bpf_fd;
+  };
+
+  struct kcm_attach info;
+
+  memset(&info, 0, sizeof(info));
+
+  info.fd = tcpfd;
+  info.bpf_fd = bpf_prog_fd;
+
+  ioctl(kcmfd, SIOCKCMATTACH, &info);
+
+The kcm_attach structure contains:
+  fd: file descriptor for TCP socket being attached
+  bpf_prog_fd: file descriptor for compiled BPF program downloaded
+
+Unattach transport sockets
+--------------------------
+
+Unattaching a transport socket from a multiplexor is straightforward. An
+"unattach" ioctl is done with the kcm_unattach structure as the argument:
+
+  /* From linux/kcm.h */
+  struct kcm_unattach {
+        int fd;
+  };
+
+  struct kcm_unattach info;
+
+  memset(&info, 0, sizeof(info));
+
+  info.fd = cfd;
+
+  ioctl(fd, SIOCKCMUNATTACH, &info);
+
+Disabling receive on KCM socket
+-------------------------------
+
+A setsockopt is used to disable or enable receiving on a KCM socket.
+When receive is disabled, any pending messages in the socket's
+receive buffer are moved to other sockets. This feature is useful
+if an application thread knows that it will be doing a lot of
+work on a request and won't be able to service new messages for a
+while. Example use:
+
+  int val = 1;
+
+  setsockopt(kcmfd, SOL_KCM, KCM_RECV_DISABLE, &val, sizeof(val))
+
+BFP programs for message delineation
+------------------------------------
+
+BPF programs can be compiled using the BPF LLVM backend. For exmple,
+the BPF program for parsing Thrift is:
+
+  #include "bpf.h" /* for __sk_buff */
+  #include "bpf_helpers.h" /* for load_word intrinsic */
+
+  SEC("socket_kcm")
+  int bpf_prog1(struct __sk_buff *skb)
+  {
+       return load_word(skb, 0) + 4;
+  }
+
+  char _license[] SEC("license") = "GPL";
+
+Use in applications
+===================
+
+KCM accelerates application layer protocols. Specifically, it allows
+applications to use a message based interface for sending and receiving
+messages. The kernel provides necessary assurances that messages are sent
+and received atomically. This relieves much of the burden applications have
+in mapping a message based protocol onto the TCP stream. KCM also make
+application layer messages a unit of work in the kernel for the purposes of
+steerng and scheduling, which in turn allows a simpler networking model in
+multithreaded applications.
+
+Configurations
+--------------
+
+In an Nx1 configuration, KCM logically provides multiple socket handles
+to the same TCP connection. This allows parallelism between in I/O
+operations on the TCP socket (for instance copyin and copyout of data is
+parallelized). In an application, a KCM socket can be opened for each
+processing thread and inserted into the epoll (similar to how SO_REUSEPORT
+is used to allow multiple listener sockets on the same port).
+
+In a MxN configuration, multiple connections are established to the
+same destination. These are used for simple load balancing.
+
+Message batching
+----------------
+
+The primary purpose of KCM is load balancing between KCM sockets and hence
+threads in a nominal use case. Perfect load balancing, that is steering
+each received message to a different KCM socket or steering each sent
+message to a different TCP socket, can negatively impact performance
+since this doesn't allow for affinities to be established. Balancing
+based on groups, or batches of messages, can be beneficial for performance.
+
+On transmit, there are three ways an application can batch (pipeline)
+messages on a KCM socket.
+  1) Send multiple messages in a single sendmmsg.
+  2) Send a group of messages each with a sendmsg call, where all messages
+     except the last have MSG_BATCH in the flags of sendmsg call.
+  3) Create "super message" composed of multiple messages and send this
+     with a single sendmsg.
+
+On receive, the KCM module attempts to queue messages received on the
+same KCM socket during each TCP ready callback. The targeted KCM socket
+changes at each receive ready callback on the KCM socket. The application
+does not need to configure this.
+
+Error handling
+--------------
+
+An application should include a thread to monitor errors raised on
+the TCP connection. Normally, this will be done by placing each
+TCP socket attached to a KCM multiplexor in epoll set for POLLERR
+event. If an error occurs on an attached TCP socket, KCM sets an EPIPE
+on the socket thus waking up the application thread. When the application
+sees the error (which may just be a disconnect) it should unattach the
+socket from KCM and then close it. It is assumed that once an error is
+posted on the TCP socket the data stream is unrecoverable (i.e. an error
+may have occurred in in the middle of receiving a messssge).
+
+TCP connection monitoring
+-------------------------
+
+In KCM there is no means to correlate a message to the TCP socket that
+was used to send or receive the message (except in the case there is
+only one attached TCP socket). However, the application does retain
+an open file descriptor to the socket so it will be able to get statistics
+from the socket which can be used in detecting issues (such as high
+retransmissions on the socket).
-- 
cgit v0.10.2


From 87aca73737e379f079993802d2c43606f7c5d26c Mon Sep 17 00:00:00 2001
From: Jean Delvare <jdelvare@suse.de>
Date: Thu, 21 Jan 2016 09:20:12 +0100
Subject: NFC: microread: Drop platform data header file

Originally I only wanted to drop the unneeded inclusion of
<linux/i2c.h>, but then noticed that struct
microread_nfc_platform_data isn't actually used, and
MICROREAD_DRIVER_NAME is redefined in the only file where it is used,
so we can get rid of the header file and dead code altogether.

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org>
Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

diff --git a/MAINTAINERS b/MAINTAINERS
index 355e1c8..5e4e50f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7644,7 +7644,6 @@ F:	net/nfc/
 F:	include/net/nfc/
 F:	include/uapi/linux/nfc.h
 F:	drivers/nfc/
-F:	include/linux/platform_data/microread.h
 F:	include/linux/platform_data/nfcmrvl.h
 F:	include/linux/platform_data/nxp-nci.h
 F:	include/linux/platform_data/pn544.h
diff --git a/drivers/nfc/microread/i2c.c b/drivers/nfc/microread/i2c.c
index 918e8f2..e0e8afd 100644
--- a/drivers/nfc/microread/i2c.c
+++ b/drivers/nfc/microread/i2c.c
@@ -246,18 +246,10 @@ static int microread_i2c_probe(struct i2c_client *client,
 			       const struct i2c_device_id *id)
 {
 	struct microread_i2c_phy *phy;
-	struct microread_nfc_platform_data *pdata =
-		dev_get_platdata(&client->dev);
 	int r;
 
 	dev_dbg(&client->dev, "client %p\n", client);
 
-	if (!pdata) {
-		nfc_err(&client->dev, "client %p: missing platform data\n",
-			client);
-		return -EINVAL;
-	}
-
 	phy = devm_kzalloc(&client->dev, sizeof(struct microread_i2c_phy),
 			   GFP_KERNEL);
 	if (!phy)
diff --git a/include/linux/platform_data/microread.h b/include/linux/platform_data/microread.h
deleted file mode 100644
index ca13992..0000000
--- a/include/linux/platform_data/microread.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * Driver include for the Inside Secure microread NFC Chip.
- *
- * Copyright (C) 2011 Tieto Poland
- * Copyright (C) 2012 Intel Corporation. All rights reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * version 2 as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-#ifndef _MICROREAD_H
-#define _MICROREAD_H
-
-#include <linux/i2c.h>
-
-#define MICROREAD_DRIVER_NAME	"microread"
-
-/* board config platform data for microread */
-struct microread_nfc_platform_data {
-	unsigned int rst_gpio;
-	unsigned int irq_gpio;
-	unsigned int ioh_gpio;
-};
-
-#endif /* _MICROREAD_H */
-- 
cgit v0.10.2


From 2793a23aacbd754dbbb5cb75093deb7e4103bace Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 9 Mar 2016 21:58:32 -0500
Subject: net: validate variable length ll headers

Netdevice parameter hard_header_len is variously interpreted both as
an upper and lower bound on link layer header length. The field is
used as upper bound when reserving room at allocation, as lower bound
when validating user input in PF_PACKET.

Clarify the definition to be maximum header length. For validation
of untrusted headers, add an optional validate member to header_ops.

Allow bypassing of validation by passing CAP_SYS_RAWIO, for instance
for deliberate testing of corrupt input. In this case, pad trailing
bytes, as some device drivers expect completely initialized headers.

See also http://comments.gmane.org/gmane.linux.network/401064

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index efe7cec..fd30cb5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -268,6 +268,7 @@ struct header_ops {
 	void	(*cache_update)(struct hh_cache *hh,
 				const struct net_device *dev,
 				const unsigned char *haddr);
+	bool	(*validate)(const char *ll_header, unsigned int len);
 };
 
 /* These flag bits are private to the generic network queueing
@@ -1459,8 +1460,7 @@ enum netdev_priv_flags {
  *	@dma:		DMA channel
  *	@mtu:		Interface MTU value
  *	@type:		Interface hardware type
- *	@hard_header_len: Hardware header length, which means that this is the
- *			  minimum size of a packet.
+ *	@hard_header_len: Maximum hardware header length.
  *
  *	@needed_headroom: Extra headroom the hardware may need, but not in all
  *			  cases can this be guaranteed
@@ -2687,6 +2687,24 @@ static inline int dev_parse_header(const struct sk_buff *skb,
 	return dev->header_ops->parse(skb, haddr);
 }
 
+/* ll_header must have at least hard_header_len allocated */
+static inline bool dev_validate_header(const struct net_device *dev,
+				       char *ll_header, int len)
+{
+	if (likely(len >= dev->hard_header_len))
+		return true;
+
+	if (capable(CAP_SYS_RAWIO)) {
+		memset(ll_header + len, 0, dev->hard_header_len - len);
+		return true;
+	}
+
+	if (dev->header_ops && dev->header_ops->validate)
+		return dev->header_ops->validate(ll_header, len);
+
+	return false;
+}
+
 typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
 int register_gifconf(unsigned int family, gifconf_func_t *gifconf);
 static inline int unregister_gifconf(unsigned int family)
-- 
cgit v0.10.2


From ea47781c26510e5d97f80f9aceafe9065bd5e3aa Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 9 Mar 2016 21:58:33 -0500
Subject: ax25: add link layer header validation function

As variable length protocol, AX25 fails link layer header validation
tests based on a minimum length. header_ops.validate allows protocols
to validate headers that are shorter than hard_header_len. Implement
this callback for AX25.

See also http://comments.gmane.org/gmane.linux.network/401064

Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ax25/ax25_ip.c b/net/ax25/ax25_ip.c
index b563a3f..2fa3be9 100644
--- a/net/ax25/ax25_ip.c
+++ b/net/ax25/ax25_ip.c
@@ -228,8 +228,23 @@ netdev_tx_t ax25_ip_xmit(struct sk_buff *skb)
 }
 #endif
 
+static bool ax25_validate_header(const char *header, unsigned int len)
+{
+	ax25_digi digi;
+
+	if (!len)
+		return false;
+
+	if (header[0])
+		return true;
+
+	return ax25_addr_parse(header + 1, len - 1, NULL, NULL, &digi, NULL,
+			       NULL);
+}
+
 const struct header_ops ax25_header_ops = {
 	.create = ax25_hard_header,
+	.validate = ax25_validate_header,
 };
 
 EXPORT_SYMBOL(ax25_header_ops);
-- 
cgit v0.10.2


From 9ed988cd591500c040b2a6257bc68543e08ceeef Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Wed, 9 Mar 2016 21:58:34 -0500
Subject: packet: validate variable length ll headers

Replace link layer header validation check ll_header_truncate with
more generic dev_validate_header.

Validation based on hard_header_len incorrectly drops valid packets
in variable length protocols, such as AX25. dev_validate_header
calls header_ops.validate for such protocols to ensure correctness
below hard_header_len.

See also http://comments.gmane.org/gmane.linux.network/401064

Fixes 9c7077622dd9 ("packet: make packet_snd fail on len smaller than l2 header")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index d41b107..1ecfa71 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -1915,6 +1915,10 @@ retry:
 		goto retry;
 	}
 
+	if (!dev_validate_header(dev, skb->data, len)) {
+		err = -EINVAL;
+		goto out_unlock;
+	}
 	if (len > (dev->mtu + dev->hard_header_len + extra_len) &&
 	    !packet_extra_vlan_len_allowed(dev, skb)) {
 		err = -EMSGSIZE;
@@ -2393,18 +2397,6 @@ static void tpacket_destruct_skb(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 
-static bool ll_header_truncated(const struct net_device *dev, int len)
-{
-	/* net device doesn't like empty head */
-	if (unlikely(len < dev->hard_header_len)) {
-		net_warn_ratelimited("%s: packet size is too short (%d < %d)\n",
-				     current->comm, len, dev->hard_header_len);
-		return true;
-	}
-
-	return false;
-}
-
 static void tpacket_set_protocol(const struct net_device *dev,
 				 struct sk_buff *skb)
 {
@@ -2522,16 +2514,20 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 		if (unlikely(err < 0))
 			return -EINVAL;
 	} else if (copylen) {
+		int hdrlen = min_t(int, copylen, tp_len);
+
 		skb_push(skb, dev->hard_header_len);
 		skb_put(skb, copylen - dev->hard_header_len);
-		err = skb_store_bits(skb, 0, data, copylen);
+		err = skb_store_bits(skb, 0, data, hdrlen);
 		if (unlikely(err))
 			return err;
+		if (!dev_validate_header(dev, skb->data, hdrlen))
+			return -EINVAL;
 		if (!skb->protocol)
 			tpacket_set_protocol(dev, skb);
 
-		data += copylen;
-		to_write -= copylen;
+		data += hdrlen;
+		to_write -= hdrlen;
 	}
 
 	offset = offset_in_page(data);
@@ -2703,13 +2699,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 			copylen = __virtio16_to_cpu(vio_le(),
 						    vnet_hdr->hdr_len);
 		}
-		if (dev->hard_header_len) {
-			if (ll_header_truncated(dev, tp_len)) {
-				tp_len = -EINVAL;
-				goto tpacket_error;
-			}
-			copylen = max_t(int, copylen, dev->hard_header_len);
-		}
+		copylen = max_t(int, copylen, dev->hard_header_len);
 		skb = sock_alloc_send_skb(&po->sk,
 				hlen + tlen + sizeof(struct sockaddr_ll) +
 				(copylen - dev->hard_header_len),
@@ -2905,9 +2895,6 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 		offset = dev_hard_header(skb, dev, ntohs(proto), addr, NULL, len);
 		if (unlikely(offset < 0))
 			goto out_free;
-	} else {
-		if (ll_header_truncated(dev, len))
-			goto out_free;
 	}
 
 	/* Returns -EFAULT on error */
@@ -2915,6 +2902,12 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	if (err)
 		goto out_free;
 
+	if (sock->type == SOCK_RAW &&
+	    !dev_validate_header(dev, skb->data, len)) {
+		err = -EINVAL;
+		goto out_free;
+	}
+
 	sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags);
 
 	if (!vnet_hdr.gso_type && (len > dev->mtu + reserve + extra_len) &&
-- 
cgit v0.10.2


From b8cdc05173f05d212627b7aba7ec47fa334a79f2 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Wed, 9 Mar 2016 18:56:49 -0800
Subject: bpf: bpf_stackmap_copy depends on CONFIG_PERF_EVENTS

0-day bot reported build error:
kernel/built-in.o: In function `map_lookup_elem':
>> kernel/bpf/.tmp_syscall.o:(.text+0x329b3c): undefined reference to `bpf_stackmap_copy'
when CONFIG_BPF_SYSCALL is set and CONFIG_PERF_EVENTS is not.
Add weak definition to resolve it.
This code path in map_lookup_elem() is never taken
when CONFIG_PERF_EVENTS is not set.

Fixes: 557c0c6e7df8 ("bpf: convert stackmap to pre-allocation")
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 2978d0d..2a2efe1 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -244,6 +244,11 @@ static void __user *u64_to_ptr(__u64 val)
 	return (void __user *) (unsigned long) val;
 }
 
+int __weak bpf_stackmap_copy(struct bpf_map *map, void *key, void *value)
+{
+	return -ENOTSUPP;
+}
+
 /* last field in 'union bpf_attr' used by this command */
 #define BPF_MAP_LOOKUP_ELEM_LAST_FIELD value
 
-- 
cgit v0.10.2


From cdc4e47da8f4c32eeb6b2061a8a834f4362a12b7 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@fb.com>
Date: Wed, 9 Mar 2016 20:02:33 -0800
Subject: bpf: avoid copying junk bytes in bpf_get_current_comm()

Lots of places in the kernel use memcpy(buf, comm, TASK_COMM_LEN); but
the result is typically passed to print("%s", buf) and extra bytes
after zero don't cause any harm.
In bpf the result of bpf_get_current_comm() is used as the part of
map key and was causing spurious hash map mismatches.
Use strlcpy() to guarantee zero-terminated string.
bpf verifier checks that output buffer is zero-initialized,
so even for short task names the output buffer don't have junk bytes.
Note it's not a security concern, since kprobe+bpf is root only.

Fixes: ffeedafbf023 ("bpf: introduce current->pid, tgid, uid, gid, comm accessors")
Reported-by: Tobias Waldekranz <tobias@waldekranz.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 4504ca6..50da680 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -166,7 +166,7 @@ static u64 bpf_get_current_comm(u64 r1, u64 size, u64 r3, u64 r4, u64 r5)
 	if (!task)
 		return -EINVAL;
 
-	memcpy(buf, task->comm, min_t(size_t, size, sizeof(task->comm)));
+	strlcpy(buf, task->comm, min_t(size_t, size, sizeof(task->comm)));
 	return 0;
 }
 
-- 
cgit v0.10.2


From 99312c377f82f91e7754aefe42f64b619b295fd6 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Thu, 3 Mar 2016 20:45:55 +0100
Subject: can: ifi: Fix clock generator configuration

The clock generation does not match reality when using the CAN IP
core outside of the FPGA design. This patch fixes the computation
of values which are programmed into the clock generator registers.

First, there are some off-by-one errors which manifest themselves
only when communicating with different controller, so those are
fixed.

Second, the bits in the clock generator registers have different
meaning depending on whether the core is in ISO CANFD mode or any
of the other modes (BOSCH CANFD or CAN2.0). Detect the ISO CANFD
mode and fix handling of this special case of clock configuration.

Finally, the CAN clock speed is in CANCLOCK register, not SYSCLOCK
register, so fix this as well.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Reviewed-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c
index 0d1c164..57581cb 100644
--- a/drivers/net/can/ifi_canfd/ifi_canfd.c
+++ b/drivers/net/can/ifi_canfd/ifi_canfd.c
@@ -514,25 +514,25 @@ static irqreturn_t ifi_canfd_isr(int irq, void *dev_id)
 
 static const struct can_bittiming_const ifi_canfd_bittiming_const = {
 	.name		= KBUILD_MODNAME,
-	.tseg1_min	= 2,	/* Time segment 1 = prop_seg + phase_seg1 */
+	.tseg1_min	= 1,	/* Time segment 1 = prop_seg + phase_seg1 */
 	.tseg1_max	= 64,
-	.tseg2_min	= 1,	/* Time segment 2 = phase_seg2 */
-	.tseg2_max	= 16,
+	.tseg2_min	= 2,	/* Time segment 2 = phase_seg2 */
+	.tseg2_max	= 64,
 	.sjw_max	= 16,
-	.brp_min	= 1,
-	.brp_max	= 1024,
+	.brp_min	= 2,
+	.brp_max	= 256,
 	.brp_inc	= 1,
 };
 
 static const struct can_bittiming_const ifi_canfd_data_bittiming_const = {
 	.name		= KBUILD_MODNAME,
-	.tseg1_min	= 2,	/* Time segment 1 = prop_seg + phase_seg1 */
-	.tseg1_max	= 16,
-	.tseg2_min	= 1,	/* Time segment 2 = phase_seg2 */
-	.tseg2_max	= 8,
-	.sjw_max	= 4,
-	.brp_min	= 1,
-	.brp_max	= 32,
+	.tseg1_min	= 1,	/* Time segment 1 = prop_seg + phase_seg1 */
+	.tseg1_max	= 64,
+	.tseg2_min	= 2,	/* Time segment 2 = phase_seg2 */
+	.tseg2_max	= 64,
+	.sjw_max	= 16,
+	.brp_min	= 2,
+	.brp_max	= 256,
 	.brp_inc	= 1,
 };
 
@@ -545,32 +545,34 @@ static void ifi_canfd_set_bittiming(struct net_device *ndev)
 	u32 noniso_arg = 0;
 	u32 time_off;
 
-	if (priv->can.ctrlmode & CAN_CTRLMODE_FD_NON_ISO) {
+	if ((priv->can.ctrlmode & CAN_CTRLMODE_FD) &&
+	    !(priv->can.ctrlmode & CAN_CTRLMODE_FD_NON_ISO)) {
+		time_off = IFI_CANFD_TIME_SJW_OFF_ISO;
+	} else {
 		noniso_arg = IFI_CANFD_TIME_SET_TIMEB_BOSCH |
 			     IFI_CANFD_TIME_SET_TIMEA_BOSCH |
 			     IFI_CANFD_TIME_SET_PRESC_BOSCH |
 			     IFI_CANFD_TIME_SET_SJW_BOSCH;
 		time_off = IFI_CANFD_TIME_SJW_OFF_BOSCH;
-	} else {
-		time_off = IFI_CANFD_TIME_SJW_OFF_ISO;
 	}
 
 	/* Configure bit timing */
-	brp = bt->brp - 1;
+	brp = bt->brp - 2;
 	sjw = bt->sjw - 1;
 	tseg1 = bt->prop_seg + bt->phase_seg1 - 1;
-	tseg2 = bt->phase_seg2 - 1;
+	tseg2 = bt->phase_seg2 - 2;
 	writel((tseg2 << IFI_CANFD_TIME_TIMEB_OFF) |
 	       (tseg1 << IFI_CANFD_TIME_TIMEA_OFF) |
 	       (brp << IFI_CANFD_TIME_PRESCALE_OFF) |
-	       (sjw << time_off),
+	       (sjw << time_off) |
+	       noniso_arg,
 	       priv->base + IFI_CANFD_TIME);
 
 	/* Configure data bit timing */
-	brp = dbt->brp - 1;
+	brp = dbt->brp - 2;
 	sjw = dbt->sjw - 1;
 	tseg1 = dbt->prop_seg + dbt->phase_seg1 - 1;
-	tseg2 = dbt->phase_seg2 - 1;
+	tseg2 = dbt->phase_seg2 - 2;
 	writel((tseg2 << IFI_CANFD_TIME_TIMEB_OFF) |
 	       (tseg1 << IFI_CANFD_TIME_TIMEA_OFF) |
 	       (brp << IFI_CANFD_TIME_PRESCALE_OFF) |
@@ -847,7 +849,7 @@ static int ifi_canfd_plat_probe(struct platform_device *pdev)
 
 	priv->can.state = CAN_STATE_STOPPED;
 
-	priv->can.clock.freq = readl(addr + IFI_CANFD_SYSCLOCK);
+	priv->can.clock.freq = readl(addr + IFI_CANFD_CANCLOCK);
 
 	priv->can.bittiming_const	= &ifi_canfd_bittiming_const;
 	priv->can.data_bittiming_const	= &ifi_canfd_data_bittiming_const;
-- 
cgit v0.10.2


From f1deaee0c3ab1950987d7207d613df0417bd251c Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Thu, 3 Mar 2016 20:45:56 +0100
Subject: can: ifi: Fix TX DLC configuration

The TX DLC, the transmission length information, was not written
into the transmit configuration register. When using the CAN core
with different CAN controller, the receiving CAN controller will
receive only the ID part of the CAN frame, but no data at all.

This patch adds the TX DLC into the register to fix this issue.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Reviewed-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c
index 57581cb..fb93122 100644
--- a/drivers/net/can/ifi_canfd/ifi_canfd.c
+++ b/drivers/net/can/ifi_canfd/ifi_canfd.c
@@ -749,8 +749,7 @@ static netdev_tx_t ifi_canfd_start_xmit(struct sk_buff *skb,
 {
 	struct ifi_canfd_priv *priv = netdev_priv(ndev);
 	struct canfd_frame *cf = (struct canfd_frame *)skb->data;
-	u32 txst, txid;
-	u32 txdlc = 0;
+	u32 txst, txid, txdlc;
 	int i;
 
 	if (can_dropped_invalid_skb(ndev, skb))
@@ -773,12 +772,11 @@ static netdev_tx_t ifi_canfd_start_xmit(struct sk_buff *skb,
 		txid = cf->can_id & CAN_SFF_MASK;
 	}
 
-	if (priv->can.ctrlmode & (CAN_CTRLMODE_FD | CAN_CTRLMODE_FD_NON_ISO)) {
-		if (can_is_canfd_skb(skb)) {
-			txdlc |= IFI_CANFD_TXFIFO_DLC_EDL;
-			if (cf->flags & CANFD_BRS)
-				txdlc |= IFI_CANFD_TXFIFO_DLC_BRS;
-		}
+	txdlc = can_len2dlc(cf->len);
+	if ((priv->can.ctrlmode & CAN_CTRLMODE_FD) && can_is_canfd_skb(skb)) {
+		txdlc |= IFI_CANFD_TXFIFO_DLC_EDL;
+		if (cf->flags & CANFD_BRS)
+			txdlc |= IFI_CANFD_TXFIFO_DLC_BRS;
 	}
 
 	if (cf->can_id & CAN_RTR_FLAG)
-- 
cgit v0.10.2


From 223654355c4eb35f890233e6a52b0f8d79bc3b44 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Thu, 3 Mar 2016 20:45:57 +0100
Subject: can: ifi: Fix RX and TX ID mask

The RX and TX ID mask for CAN2.0 is 11 bits wide. This patch fixes
the incorrect mask, which caused the CAN IDs to miss the MSBit both
on receive and transmit.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Reviewed-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c
index fb93122..aee43c0 100644
--- a/drivers/net/can/ifi_canfd/ifi_canfd.c
+++ b/drivers/net/can/ifi_canfd/ifi_canfd.c
@@ -135,8 +135,8 @@
 
 #define IFI_CANFD_RXFIFO_ID			0x6c
 #define IFI_CANFD_RXFIFO_ID_ID_OFFSET		0
-#define IFI_CANFD_RXFIFO_ID_ID_STD_MASK		0x3ff
-#define IFI_CANFD_RXFIFO_ID_ID_XTD_MASK		0x1fffffff
+#define IFI_CANFD_RXFIFO_ID_ID_STD_MASK		CAN_SFF_MASK
+#define IFI_CANFD_RXFIFO_ID_ID_XTD_MASK		CAN_EFF_MASK
 #define IFI_CANFD_RXFIFO_ID_IDE			BIT(29)
 
 #define IFI_CANFD_RXFIFO_DATA			0x70	/* 0x70..0xac */
@@ -156,8 +156,8 @@
 
 #define IFI_CANFD_TXFIFO_ID			0xbc
 #define IFI_CANFD_TXFIFO_ID_ID_OFFSET		0
-#define IFI_CANFD_TXFIFO_ID_ID_STD_MASK		0x3ff
-#define IFI_CANFD_TXFIFO_ID_ID_XTD_MASK		0x1fffffff
+#define IFI_CANFD_TXFIFO_ID_ID_STD_MASK		CAN_SFF_MASK
+#define IFI_CANFD_TXFIFO_ID_ID_XTD_MASK		CAN_EFF_MASK
 #define IFI_CANFD_TXFIFO_ID_IDE			BIT(29)
 
 #define IFI_CANFD_TXFIFO_DATA			0xc0	/* 0xb0..0xfc */
-- 
cgit v0.10.2


From 6cc6426605b7edabde41c72add723979d1d509e7 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Thu, 3 Mar 2016 20:45:58 +0100
Subject: can: ifi: Add obscure bit swap for EFF frame IDs

In case of CAN2.0 EFF frame, the controller handles frame IDs in a
rather bizzare way. The ID is split into an extended part, IDX[28:11]
and standard part, ID[10:0]. In the TX path, the core first sends the
top 11 bits of the IDX, followed by ID and finally the rest of IDX.
In the RX path, the core stores the ID the LSbit part of IDX field,
followed by the LSbit parts of real IDX. The MSbit parts of IDX are
stored in ID field of the register.

This patch implements the necessary bit shuffling to mitigate this
obscure behavior. In case two of these controllers are connected
together, the RX and TX bit swapping nullifies itself and the issue
does not manifest. The issue only manifests when talking to another
different CAN controller.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Wolfgang Grandegger <wg@grandegger.com>
Reviewed-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c
index aee43c0..a1bd54f 100644
--- a/drivers/net/can/ifi_canfd/ifi_canfd.c
+++ b/drivers/net/can/ifi_canfd/ifi_canfd.c
@@ -136,7 +136,11 @@
 #define IFI_CANFD_RXFIFO_ID			0x6c
 #define IFI_CANFD_RXFIFO_ID_ID_OFFSET		0
 #define IFI_CANFD_RXFIFO_ID_ID_STD_MASK		CAN_SFF_MASK
+#define IFI_CANFD_RXFIFO_ID_ID_STD_OFFSET	0
+#define IFI_CANFD_RXFIFO_ID_ID_STD_WIDTH	10
 #define IFI_CANFD_RXFIFO_ID_ID_XTD_MASK		CAN_EFF_MASK
+#define IFI_CANFD_RXFIFO_ID_ID_XTD_OFFSET	11
+#define IFI_CANFD_RXFIFO_ID_ID_XTD_WIDTH	18
 #define IFI_CANFD_RXFIFO_ID_IDE			BIT(29)
 
 #define IFI_CANFD_RXFIFO_DATA			0x70	/* 0x70..0xac */
@@ -157,7 +161,11 @@
 #define IFI_CANFD_TXFIFO_ID			0xbc
 #define IFI_CANFD_TXFIFO_ID_ID_OFFSET		0
 #define IFI_CANFD_TXFIFO_ID_ID_STD_MASK		CAN_SFF_MASK
+#define IFI_CANFD_TXFIFO_ID_ID_STD_OFFSET	0
+#define IFI_CANFD_TXFIFO_ID_ID_STD_WIDTH	10
 #define IFI_CANFD_TXFIFO_ID_ID_XTD_MASK		CAN_EFF_MASK
+#define IFI_CANFD_TXFIFO_ID_ID_XTD_OFFSET	11
+#define IFI_CANFD_TXFIFO_ID_ID_XTD_WIDTH	18
 #define IFI_CANFD_TXFIFO_ID_IDE			BIT(29)
 
 #define IFI_CANFD_TXFIFO_DATA			0xc0	/* 0xb0..0xfc */
@@ -229,10 +237,20 @@ static void ifi_canfd_read_fifo(struct net_device *ndev)
 
 	rxid = readl(priv->base + IFI_CANFD_RXFIFO_ID);
 	id = (rxid >> IFI_CANFD_RXFIFO_ID_ID_OFFSET);
-	if (id & IFI_CANFD_RXFIFO_ID_IDE)
+	if (id & IFI_CANFD_RXFIFO_ID_IDE) {
 		id &= IFI_CANFD_RXFIFO_ID_ID_XTD_MASK;
-	else
+		/*
+		 * In case the Extended ID frame is received, the standard
+		 * and extended part of the ID are swapped in the register,
+		 * so swap them back to obtain the correct ID.
+		 */
+		id = (id >> IFI_CANFD_RXFIFO_ID_ID_XTD_OFFSET) |
+		     ((id & IFI_CANFD_RXFIFO_ID_ID_STD_MASK) <<
+		       IFI_CANFD_RXFIFO_ID_ID_XTD_WIDTH);
+		id |= CAN_EFF_FLAG;
+	} else {
 		id &= IFI_CANFD_RXFIFO_ID_ID_STD_MASK;
+	}
 	cf->can_id = id;
 
 	if (rxdlc & IFI_CANFD_RXFIFO_DLC_ESI) {
@@ -767,6 +785,15 @@ static netdev_tx_t ifi_canfd_start_xmit(struct sk_buff *skb,
 
 	if (cf->can_id & CAN_EFF_FLAG) {
 		txid = cf->can_id & CAN_EFF_MASK;
+		/*
+		 * In case the Extended ID frame is transmitted, the
+		 * standard and extended part of the ID are swapped
+		 * in the register, so swap them back to send the
+		 * correct ID.
+		 */
+		txid = (txid >> IFI_CANFD_TXFIFO_ID_ID_XTD_WIDTH) |
+		       ((txid & IFI_CANFD_TXFIFO_ID_ID_XTD_MASK) <<
+		         IFI_CANFD_TXFIFO_ID_ID_XTD_OFFSET);
 		txid |= IFI_CANFD_TXFIFO_ID_IDE;
 	} else {
 		txid = cf->can_id & CAN_SFF_MASK;
-- 
cgit v0.10.2


From e481ab23c57b37c989fa27e0a6b3e941a908775a Mon Sep 17 00:00:00 2001
From: Ramesh Shanmugasundaram <ramesh.shanmugasundaram@bp.renesas.com>
Date: Mon, 29 Feb 2016 12:37:25 +0000
Subject: can: rcar_can: Add r8a7795 support

Added r8a7795 SoC support.

Signed-off-by: Ramesh Shanmugasundaram <ramesh.shanmugasundaram@bp.renesas.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>

diff --git a/Documentation/devicetree/bindings/net/can/rcar_can.txt b/Documentation/devicetree/bindings/net/can/rcar_can.txt
index 65edc05..8d40ab2 100644
--- a/Documentation/devicetree/bindings/net/can/rcar_can.txt
+++ b/Documentation/devicetree/bindings/net/can/rcar_can.txt
@@ -9,8 +9,10 @@ Required properties:
 	      "renesas,can-r8a7792" if CAN controller is a part of R8A7792 SoC.
 	      "renesas,can-r8a7793" if CAN controller is a part of R8A7793 SoC.
 	      "renesas,can-r8a7794" if CAN controller is a part of R8A7794 SoC.
+	      "renesas,can-r8a7795" if CAN controller is a part of R8A7795 SoC.
 	      "renesas,rcar-gen1-can" for a generic R-Car Gen1 compatible device.
 	      "renesas,rcar-gen2-can" for a generic R-Car Gen2 compatible device.
+	      "renesas,rcar-gen3-can" for a generic R-Car Gen3 compatible device.
 	      When compatible with the generic version, nodes must list the
 	      SoC-specific version corresponding to the platform first
 	      followed by the generic version.
@@ -22,6 +24,15 @@ Required properties:
 - pinctrl-0: pin control group to be used for this controller.
 - pinctrl-names: must be "default".
 
+Required properties for "renesas,can-r8a7795" compatible:
+In R8A7795 SoC, "clkp2" can be CANFD clock. This is a div6 clock and can be
+used by both CAN and CAN FD controller at the same time. It needs to be scaled
+to maximum frequency if any of these controllers use it. This is done using
+the below properties.
+
+- assigned-clocks: phandle of clkp2(CANFD) clock.
+- assigned-clock-rates: maximum frequency of this clock.
+
 Optional properties:
 - renesas,can-clock-select: R-Car CAN Clock Source Select. Valid values are:
 			    <0x0> (default) : Peripheral clock (clkp1)
diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index 164ccde..0d40aef 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -106,7 +106,7 @@ config CAN_JANZ_ICAN3
 
 config CAN_RCAR
 	tristate "Renesas R-Car CAN controller"
-	depends on ARM
+	depends on ARCH_RENESAS || ARM
 	---help---
 	  Say Y here if you want to use CAN controller found on Renesas R-Car
 	  SoCs.
diff --git a/drivers/net/can/rcar_can.c b/drivers/net/can/rcar_can.c
index ad3d2e0..788459f 100644
--- a/drivers/net/can/rcar_can.c
+++ b/drivers/net/can/rcar_can.c
@@ -906,6 +906,7 @@ static const struct of_device_id rcar_can_of_table[] __maybe_unused = {
 	{ .compatible = "renesas,can-r8a7791" },
 	{ .compatible = "renesas,rcar-gen1-can" },
 	{ .compatible = "renesas,rcar-gen2-can" },
+	{ .compatible = "renesas,rcar-gen3-can" },
 	{ }
 };
 MODULE_DEVICE_TABLE(of, rcar_can_of_table);
-- 
cgit v0.10.2


From 3691ac4a9c957c2896f8ebfb8a92081301f1682e Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 8 Mar 2016 15:04:08 +0300
Subject: libertas: fix an error code in probe

We accidentally return success instead of a negative error code.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/libertas/main.c b/drivers/net/wireless/marvell/libertas/main.c
index b35b8bc..8541cbe 100644
--- a/drivers/net/wireless/marvell/libertas/main.c
+++ b/drivers/net/wireless/marvell/libertas/main.c
@@ -1118,7 +1118,8 @@ int lbs_start_card(struct lbs_private *priv)
 	else
 		pr_info("%s: mesh disabled\n", dev->name);
 
-	if (lbs_cfg_register(priv)) {
+	ret = lbs_cfg_register(priv);
+	if (ret) {
 		pr_err("cannot register device\n");
 		goto done;
 	}
-- 
cgit v0.10.2


From 107b87133361aa5502d7d3c82cdfff77bf8ebe18 Mon Sep 17 00:00:00 2001
From: Hui Wang <hui.wang@canonical.com>
Date: Wed, 9 Mar 2016 15:25:26 +0800
Subject: brcmfmac: Remove waitqueue_active check

We met a problem of pm_suspend  when repeated closing/opening the lid
on a Lenovo laptop (1/20 reproduce rate), below is the log:

[ 199.735876] PM: Entering mem sleep
[ 199.750516] e1000e: EEE TX LPI TIMER: 00000011
[ 199.856638] Trying to free nonexistent resource <000000000000d000-000000000000d0ff>
[ 201.753566] brcmfmac: brcmf_pcie_suspend: Timeout on response for entering D3 substate
[ 201.753581] pci_legacy_suspend(): brcmf_pcie_suspend+0x0/0x1f0 [brcmfmac] returns -5
[ 201.753585] dpm_run_callback(): pci_pm_suspend+0x0/0x160 returns -5
[ 201.753589] PM: Device 0000:04:00.0 failed to suspend async: error -5

Through debugging, we found when problem happens, it is not the device
fails to enter D3, but the signal D3_ACK comes too early to pass the
waitqueue_active() check.

Just like this:
brcmf_pcie_send_mb_data(devinfo, BRCMF_H2D_HOST_D3_INFORM);
// signal is triggered here
wait_event_timeout(devinfo->mbdata_resp_wait, devinfo->mbdata_completed,
		   BRCMF_PCIE_MBDATA_TIMEOUT);

So far I think it is safe to remove waitqueue_active check since there
is only one place to trigger this signal (sending
BRCMF_H2D_HOST_D3_INFORM). And it is not a problem calling wake_up
event earlier than calling wait_event.

Cc: Brett Rudley <brudley@broadcom.com>
Cc: Hante Meuleman <meuleman@broadcom.com>
Cc: Franky (Zhenhui) Lin <frankyl@broadcom.com>
Cc: Pieter-Paul Giesberts <pieterpg@broadcom.com>
Cc: Arend van Spriel <arend@broadcom.com>
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 52fef5e..0af8db8 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -677,10 +677,8 @@ static void brcmf_pcie_handle_mb_data(struct brcmf_pciedev_info *devinfo)
 		brcmf_dbg(PCIE, "D2H_MB_DATA: DEEP SLEEP EXIT\n");
 	if (dtoh_mb_data & BRCMF_D2H_DEV_D3_ACK) {
 		brcmf_dbg(PCIE, "D2H_MB_DATA: D3 ACK\n");
-		if (waitqueue_active(&devinfo->mbdata_resp_wait)) {
-			devinfo->mbdata_completed = true;
-			wake_up(&devinfo->mbdata_resp_wait);
-		}
+		devinfo->mbdata_completed = true;
+		wake_up(&devinfo->mbdata_resp_wait);
 	}
 }
 
-- 
cgit v0.10.2


From 54f008497b9f09f0880dd84ce5608be4ad7cff13 Mon Sep 17 00:00:00 2001
From: Amitkumar Karwar <akarwar@marvell.com>
Date: Wed, 9 Mar 2016 04:21:11 -0800
Subject: mwifiex: Empty Tx queue during suspend

In cfg80211 suspend handler, stop the netif queue and
wait until all the Tx queues become empty. Start the
queues in resume handler.

Signed-off-by: Amitkumar Karwar <akarwar@marvell.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index 29b7f6e..bb7235e 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -20,6 +20,7 @@
 #include "cfg80211.h"
 #include "main.h"
 #include "11n.h"
+#include "wmm.h"
 
 static char *reg_alpha2;
 module_param(reg_alpha2, charp, 0);
@@ -3259,7 +3260,7 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 {
 	struct mwifiex_adapter *adapter = mwifiex_cfg80211_get_adapter(wiphy);
 	struct mwifiex_ds_hs_cfg hs_cfg;
-	int i, ret = 0;
+	int i, ret = 0, retry_num = 10;
 	struct mwifiex_private *priv;
 
 	for (i = 0; i < adapter->priv_num; i++) {
@@ -3269,6 +3270,21 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 
 	mwifiex_cancel_all_pending_cmd(adapter);
 
+	for (i = 0; i < adapter->priv_num; i++) {
+		priv = adapter->priv[i];
+		if (priv && priv->netdev)
+			mwifiex_stop_net_dev_queue(priv->netdev, adapter);
+	}
+
+	for (i = 0; i < retry_num; i++) {
+		if (!mwifiex_wmm_lists_empty(adapter) ||
+		    !mwifiex_bypass_txlist_empty(adapter) ||
+		    !skb_queue_empty(&adapter->tx_data_q))
+			usleep_range(10000, 15000);
+		else
+			break;
+	}
+
 	if (!wowlan) {
 		mwifiex_dbg(adapter, ERROR,
 			    "None of the WOWLAN triggers enabled\n");
@@ -3321,12 +3337,18 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy,
 static int mwifiex_cfg80211_resume(struct wiphy *wiphy)
 {
 	struct mwifiex_adapter *adapter = mwifiex_cfg80211_get_adapter(wiphy);
-	struct mwifiex_private *priv =
-		mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA);
+	struct mwifiex_private *priv;
 	struct mwifiex_ds_wakeup_reason wakeup_reason;
 	struct cfg80211_wowlan_wakeup wakeup_report;
 	int i;
 
+	for (i = 0; i < adapter->priv_num; i++) {
+		priv = adapter->priv[i];
+		if (priv && priv->netdev)
+			mwifiex_wake_up_net_dev_queue(priv->netdev, adapter);
+	}
+
+	priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA);
 	mwifiex_get_wakeup_reason(priv, HostCmd_ACT_GEN_GET, MWIFIEX_SYNC_CMD,
 				  &wakeup_reason);
 	memset(&wakeup_report, 0, sizeof(struct cfg80211_wowlan_wakeup));
-- 
cgit v0.10.2


From 87cba1696045cbf963d09c82ceca98db92b7ace9 Mon Sep 17 00:00:00 2001
From: Eliad Peller <eliad@wizery.com>
Date: Mon, 7 Mar 2016 00:28:08 +0200
Subject: wlcore: don't WARN_ON in case of existing ROC

When working with AP + P2P, it's possible to get into
a state when the AP is in ROC (due to assiciating station)
while trying to ROC on the P2P interface.

Replace the WARN_ON with wl1271_error to avoid warnings
in this case.

Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index 45662cf..dde3620 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -5495,7 +5495,7 @@ static int wlcore_op_remain_on_channel(struct ieee80211_hw *hw,
 {
 	struct wl12xx_vif *wlvif = wl12xx_vif_to_data(vif);
 	struct wl1271 *wl = hw->priv;
-	int channel, ret = 0;
+	int channel, active_roc, ret = 0;
 
 	channel = ieee80211_frequency_to_channel(chan->center_freq);
 
@@ -5508,9 +5508,9 @@ static int wlcore_op_remain_on_channel(struct ieee80211_hw *hw,
 		goto out;
 
 	/* return EBUSY if we can't ROC right now */
-	if (WARN_ON(wl->roc_vif ||
-		    find_first_bit(wl->roc_map,
-				   WL12XX_MAX_ROLES) < WL12XX_MAX_ROLES)) {
+	active_roc = find_first_bit(wl->roc_map, WL12XX_MAX_ROLES);
+	if (wl->roc_vif || active_roc < WL12XX_MAX_ROLES) {
+		wl1271_warning("active roc on role %d", active_roc);
 		ret = -EBUSY;
 		goto out;
 	}
-- 
cgit v0.10.2


From 8cf77e176f921d20a1e61d288eaee74a2d1d2ae1 Mon Sep 17 00:00:00 2001
From: Eliad Peller <eliad@wizery.com>
Date: Mon, 7 Mar 2016 00:28:09 +0200
Subject: wlcore/wl18xx: add radar_debug_mode handling

Add debugfs key (under CFG80211_CERTIFICATION_ONUS
configuration) to set/clear radar_debug_mode.
In this mode, the driver simply ignores radar
events (but prints them).

The fw is notified about this mode through
a special generic_cfg_feature command.

This mode is relevant only for ap mode. look for
it when initializing ap vif.

Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/ti/wl18xx/debugfs.c b/drivers/net/wireless/ti/wl18xx/debugfs.c
index 4edfe28..86ccf84 100644
--- a/drivers/net/wireless/ti/wl18xx/debugfs.c
+++ b/drivers/net/wireless/ti/wl18xx/debugfs.c
@@ -345,6 +345,69 @@ static const struct file_operations dynamic_fw_traces_ops = {
 	.llseek = default_llseek,
 };
 
+#ifdef CONFIG_CFG80211_CERTIFICATION_ONUS
+static ssize_t radar_debug_mode_write(struct file *file,
+				      const char __user *user_buf,
+				      size_t count, loff_t *ppos)
+{
+	struct wl1271 *wl = file->private_data;
+	struct wl12xx_vif *wlvif;
+	unsigned long value;
+	int ret;
+
+	ret = kstrtoul_from_user(user_buf, count, 10, &value);
+	if (ret < 0) {
+		wl1271_warning("illegal radar_debug_mode value!");
+		return -EINVAL;
+	}
+
+	/* valid values: 0/1 */
+	if (!(value == 0 || value == 1)) {
+		wl1271_warning("value is not in valid!");
+		return -EINVAL;
+	}
+
+	mutex_lock(&wl->mutex);
+
+	wl->radar_debug_mode = value;
+
+	if (unlikely(wl->state != WLCORE_STATE_ON))
+		goto out;
+
+	ret = wl1271_ps_elp_wakeup(wl);
+	if (ret < 0)
+		goto out;
+
+	wl12xx_for_each_wlvif_ap(wl, wlvif) {
+		wlcore_cmd_generic_cfg(wl, wlvif,
+				       WLCORE_CFG_FEATURE_RADAR_DEBUG,
+				       wl->radar_debug_mode, 0);
+	}
+
+	wl1271_ps_elp_sleep(wl);
+out:
+	mutex_unlock(&wl->mutex);
+	return count;
+}
+
+static ssize_t radar_debug_mode_read(struct file *file,
+				     char __user *userbuf,
+				     size_t count, loff_t *ppos)
+{
+	struct wl1271 *wl = file->private_data;
+
+	return wl1271_format_buffer(userbuf, count, ppos,
+				    "%d\n", wl->radar_debug_mode);
+}
+
+static const struct file_operations radar_debug_mode_ops = {
+	.write = radar_debug_mode_write,
+	.read = radar_debug_mode_read,
+	.open = simple_open,
+	.llseek = default_llseek,
+};
+#endif /* CFG80211_CERTIFICATION_ONUS */
+
 int wl18xx_debugfs_add_files(struct wl1271 *wl,
 			     struct dentry *rootdir)
 {
@@ -510,6 +573,9 @@ int wl18xx_debugfs_add_files(struct wl1271 *wl,
 
 	DEBUGFS_ADD(conf, moddir);
 	DEBUGFS_ADD(radar_detection, moddir);
+#ifdef CONFIG_CFG80211_CERTIFICATION_ONUS
+	DEBUGFS_ADD(radar_debug_mode, moddir);
+#endif
 	DEBUGFS_ADD(dynamic_fw_traces, moddir);
 
 	return 0;
diff --git a/drivers/net/wireless/ti/wl18xx/event.c b/drivers/net/wireless/ti/wl18xx/event.c
index 719907a..ff6e46d 100644
--- a/drivers/net/wireless/ti/wl18xx/event.c
+++ b/drivers/net/wireless/ti/wl18xx/event.c
@@ -146,7 +146,8 @@ int wl18xx_process_mailbox_events(struct wl1271 *wl)
 			    mbox->radar_channel,
 			    wl18xx_radar_type_decode(mbox->radar_type));
 
-		ieee80211_radar_detected(wl->hw);
+		if (!wl->radar_debug_mode)
+			ieee80211_radar_detected(wl->hw);
 	}
 
 	if (vector & PERIODIC_SCAN_REPORT_EVENT_ID) {
diff --git a/drivers/net/wireless/ti/wlcore/init.c b/drivers/net/wireless/ti/wlcore/init.c
index e92f263..d0b7734 100644
--- a/drivers/net/wireless/ti/wlcore/init.c
+++ b/drivers/net/wireless/ti/wlcore/init.c
@@ -558,6 +558,11 @@ static int wl12xx_init_ap_role(struct wl1271 *wl, struct wl12xx_vif *wlvif)
 	if (ret < 0)
 		return ret;
 
+	if (wl->radar_debug_mode)
+		wlcore_cmd_generic_cfg(wl, wlvif,
+				       WLCORE_CFG_FEATURE_RADAR_DEBUG,
+				       wl->radar_debug_mode, 0);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ti/wlcore/wlcore.h b/drivers/net/wireless/ti/wlcore/wlcore.h
index dda01b1..72c31a8 100644
--- a/drivers/net/wireless/ti/wlcore/wlcore.h
+++ b/drivers/net/wireless/ti/wlcore/wlcore.h
@@ -463,6 +463,7 @@ struct wl1271 {
 
 	/* the current dfs region */
 	enum nl80211_dfs_regions dfs_region;
+	bool radar_debug_mode;
 
 	/* size of the private FW status data */
 	size_t fw_status_len;
-- 
cgit v0.10.2


From 3307d84024d2ed9c8aee0a4bfa694319613734bf Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:03:59 -0500
Subject: rtl8xxxu: Add initial code to parse rtl8192eu efuse

This is the start of 8192eu support. For now just detect the device
and parse the efuse.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e654bd3..bb95d00 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -42,7 +42,7 @@
 
 #define DRIVER_NAME "rtl8xxxu"
 
-static int rtl8xxxu_debug;
+static int rtl8xxxu_debug = RTL8XXXU_DEBUG_EFUSE;
 static bool rtl8xxxu_ht40_2g;
 
 MODULE_AUTHOR("Jes Sorensen <Jes.Sorensen@redhat.com>");
@@ -1890,6 +1890,53 @@ static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 #endif
 
+static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
+{
+	int i;
+
+	if (priv->efuse_wifi.efuse8192eu.rtl_id != cpu_to_le16(0x8129))
+		return -EINVAL;
+
+	ether_addr_copy(priv->mac_addr, priv->efuse_wifi.efuse8192eu.mac_addr);
+
+	memcpy(priv->cck_tx_power_index_A,
+	       priv->efuse_wifi.efuse8192eu.cck_tx_power_index_A,
+	       sizeof(priv->cck_tx_power_index_A));
+	memcpy(priv->cck_tx_power_index_B,
+	       priv->efuse_wifi.efuse8192eu.cck_tx_power_index_B,
+	       sizeof(priv->cck_tx_power_index_B));
+
+	memcpy(priv->ht40_1s_tx_power_index_A,
+	       priv->efuse_wifi.efuse8192eu.ht40_1s_tx_power_index_A,
+	       sizeof(priv->ht40_1s_tx_power_index_A));
+	memcpy(priv->ht40_1s_tx_power_index_B,
+	       priv->efuse_wifi.efuse8192eu.ht40_1s_tx_power_index_B,
+	       sizeof(priv->ht40_1s_tx_power_index_B));
+
+	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
+		 priv->efuse_wifi.efuse8192eu.vendor_name);
+	dev_info(&priv->udev->dev, "Product: %.11s\n",
+		 priv->efuse_wifi.efuse8192eu.device_name);
+	dev_info(&priv->udev->dev, "Serial: %.11s\n",
+		 priv->efuse_wifi.efuse8192eu.serial);
+
+	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_EFUSE) {
+		unsigned char *raw = priv->efuse_wifi.raw;
+
+		dev_info(&priv->udev->dev,
+			 "%s: dumping efuse (0x%02zx bytes):\n",
+			 __func__, sizeof(struct rtl8192eu_efuse));
+		for (i = 0; i < sizeof(struct rtl8192eu_efuse); i += 8) {
+			dev_info(&priv->udev->dev, "%02x: "
+				 "%02x %02x %02x %02x %02x %02x %02x %02x\n", i,
+				 raw[i], raw[i + 1], raw[i + 2],
+				 raw[i + 3], raw[i + 4], raw[i + 5],
+				 raw[i + 6], raw[i + 7]);
+		}
+	}
+	return -EINVAL;
+}
+
 static int
 rtl8xxxu_read_efuse8(struct rtl8xxxu_priv *priv, u16 offset, u8 *data)
 {
@@ -1973,7 +2020,7 @@ static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 	}
 
 	/* Default value is 0xff */
-	memset(priv->efuse_wifi.raw, 0xff, EFUSE_MAP_LEN_8723A);
+	memset(priv->efuse_wifi.raw, 0xff, EFUSE_MAP_LEN);
 
 	efuse_addr = 0;
 	while (efuse_addr < EFUSE_REAL_CONTENT_LEN_8723A) {
@@ -2005,7 +2052,7 @@ static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 
 			/* We have 8 bits to indicate validity */
 			map_addr = offset * 8;
-			if (map_addr >= EFUSE_MAP_LEN_8723A) {
+			if (map_addr >= EFUSE_MAP_LEN) {
 				dev_warn(dev, "%s: Illegal map_addr (%04x), "
 					 "efuse corrupt!\n",
 					 __func__, map_addr);
@@ -2286,6 +2333,24 @@ static int rtl8192cu_load_firmware(struct rtl8xxxu_priv *priv)
 
 #endif
 
+static int rtl8192eu_load_firmware(struct rtl8xxxu_priv *priv)
+{
+	char *fw_name;
+	int ret;
+
+	return -EBUSY;
+	if (!priv->vendor_umc)
+		fw_name = "rtlwifi/rtl8192cufw_TMSC.bin";
+	else if (priv->chip_cut || priv->rtlchip == 0x8192c)
+		fw_name = "rtlwifi/rtl8192cufw_B.bin";
+	else
+		fw_name = "rtlwifi/rtl8192cufw_A.bin";
+
+	ret = rtl8xxxu_load_firmware(priv, fw_name);
+
+	return ret;
+}
+
 static void rtl8xxxu_firmware_self_reset(struct rtl8xxxu_priv *priv)
 {
 	u16 val16;
@@ -5863,6 +5928,13 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 
 #endif
 
+static struct rtl8xxxu_fileops rtl8192eu_fops = {
+	.parse_efuse = rtl8192eu_parse_efuse,
+	.load_firmware = rtl8192eu_load_firmware,
+	.power_on = rtl8192cu_power_on,
+	.writeN_block_size = 128,
+};
+
 static struct usb_device_id dev_table[] = {
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x8724, 0xff, 0xff, 0xff),
 	.driver_info = (unsigned long)&rtl8723au_fops},
@@ -5870,6 +5942,8 @@ static struct usb_device_id dev_table[] = {
 	.driver_info = (unsigned long)&rtl8723au_fops},
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x0724, 0xff, 0xff, 0xff),
 	.driver_info = (unsigned long)&rtl8723au_fops},
+{USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x818b, 0xff, 0xff, 0xff),
+	.driver_info = (unsigned long)&rtl8192eu_fops},
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 /* Still supported by rtlwifi */
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x8176, 0xff, 0xff, 0xff),
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index bbd0f6b..8168a35 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -55,8 +55,8 @@
 #define RTL8723A_MAX_RF_PATHS		2
 #define RF6052_MAX_TX_PWR		0x3f
 
-#define EFUSE_MAP_LEN_8723A		256
-#define EFUSE_MAX_SECTION_8723A		32
+#define EFUSE_MAP_LEN			512
+#define EFUSE_MAX_SECTION_8723A		64
 #define EFUSE_REAL_CONTENT_LEN_8723A	512
 #define EFUSE_BT_MAP_LEN_8723A		1024
 #define EFUSE_MAX_WORD_UNIT		4
@@ -500,6 +500,49 @@ struct rtl8192cu_efuse {
 	u8 customer_id;
 };
 
+struct rtl8192eu_efuse {
+	__le16 rtl_id;
+	u8 res0[0x0e];
+	u8 cck_tx_power_index_A[3];	/* 0x10 */
+	u8 cck_tx_power_index_B[3];
+	u8 ht40_1s_tx_power_index_A[3];	/* 0x16 */
+	u8 ht40_1s_tx_power_index_B[3];
+	u8 res1[0x9c];
+	u8 channel_plan;		/* 0xb8 */
+	u8 xtal_k;
+	u8 thermal_meter;
+	u8 iqk_lck;
+	u8 pa_type;			/* 0xbc */
+	u8 lna_type_2g;			/* 0xbd */
+	u8 res2[1];
+	u8 lna_type_5g;			/* 0xbf */
+	u8 res13[1];
+	u8 rf_board_option;
+	u8 rf_feature_option;
+	u8 rf_bt_setting;
+	u8 eeprom_version;
+	u8 eeprom_customer_id;
+	u8 res3[3];
+	u8 rf_antenna_option;		/* 0xc9 */
+	u8 res4[6];
+	u8 vid;				/* 0xd0 */
+	u8 res5[1];
+	u8 pid;				/* 0xd2 */
+	u8 res6[1];
+	u8 usb_optional_function;
+	u8 res7[2];
+	u8 mac_addr[ETH_ALEN];		/* 0xd7 */
+	u8 res8[2];
+	u8 vendor_name[7];
+	u8 res9[2];
+	u8 device_name[0x0b];		/* 0xe8 */
+	u8 res10[2];
+	u8 serial[0x0b];		/* 0xf5 */
+	u8 res11[0x30];
+	u8 unknown[0x0d];		/* 0x130 */
+	u8 res12[0xc3];
+};
+
 struct rtl8xxxu_reg8val {
 	u16 reg;
 	u8 val;
@@ -643,9 +686,10 @@ struct rtl8xxxu_priv {
 		u8 val8;
 	} usb_buf;
 	union {
-		u8 raw[EFUSE_MAP_LEN_8723A];
+		u8 raw[EFUSE_MAP_LEN];
 		struct rtl8723au_efuse efuse8723;
 		struct rtl8192cu_efuse efuse8192;
+		struct rtl8192eu_efuse efuse8192eu;
 	} efuse_wifi;
 	u32 adda_backup[RTL8XXXU_ADDA_REGS];
 	u32 mac_backup[RTL8XXXU_MAC_REGS];
-- 
cgit v0.10.2


From 0e5d435a619fadb475f3a260eda2252acd1662f0 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:00 -0500
Subject: rtl8xxxu: Identify chip vendors correctly

This identifies the chip vendors correctly and also picks the correct
firmware for rtl8192eu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index bb95d00..7459422 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1662,16 +1662,24 @@ static void rtl8xxxu_print_chipinfo(struct rtl8xxxu_priv *priv)
 	case 1:
 		cut = "B";
 		break;
+	case 2:
+		cut = "C";
+		break;
+	case 3:
+		cut = "D";
+		break;
+	case 4:
+		cut = "E";
+		break;
 	default:
 		cut = "unknown";
 	}
 
 	dev_info(dev,
 		 "RTL%s rev %s (%s) %iT%iR, TX queues %i, WiFi=%i, BT=%i, GPS=%i, HI PA=%i\n",
-		 priv->chip_name, cut, priv->vendor_umc ? "UMC" : "TSMC",
-		 priv->tx_paths, priv->rx_paths, priv->ep_tx_count,
-		 priv->has_wifi, priv->has_bluetooth, priv->has_gps,
-		 priv->hi_pa);
+		 priv->chip_name, cut, priv->chip_vendor, priv->tx_paths,
+		 priv->rx_paths, priv->ep_tx_count, priv->has_wifi,
+		 priv->has_bluetooth, priv->has_gps, priv->hi_pa);
 
 	dev_info(dev, "RTL%s MAC: %pM\n", priv->chip_name, priv->mac_addr);
 }
@@ -1708,7 +1716,21 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 	} else if (val32 & SYS_CFG_TYPE_ID) {
 		bonding = rtl8xxxu_read32(priv, REG_HPON_FSM);
 		bonding &= HPON_FSM_BONDING_MASK;
-		if (bonding == HPON_FSM_BONDING_1T2R) {
+		if (priv->chip_cut >= 3) {
+			if (bonding == HPON_FSM_BONDING_1T2R) {
+				sprintf(priv->chip_name, "8191EU");
+				priv->rf_paths = 2;
+				priv->rx_paths = 2;
+				priv->tx_paths = 1;
+				priv->rtlchip = 0x8191e;
+			} else {
+				sprintf(priv->chip_name, "8192EU");
+				priv->rf_paths = 2;
+				priv->rx_paths = 2;
+				priv->tx_paths = 2;
+				priv->rtlchip = 0x8192e;
+			}
+		} else if (bonding == HPON_FSM_BONDING_1T2R) {
 			sprintf(priv->chip_name, "8191CU");
 			priv->rf_paths = 2;
 			priv->rx_paths = 2;
@@ -1731,8 +1753,34 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 		priv->has_wifi = 1;
 	}
 
-	if (val32 & SYS_CFG_VENDOR_ID)
-		priv->vendor_umc = 1;
+	switch (priv->rtlchip) {
+	case 0x8188e:
+	case 0x8192e:
+	case 0x8723b:
+		switch (val32 & SYS_CFG_VENDOR_EXT_MASK) {
+		case SYS_CFG_VENDOR_ID_TSMC:
+			sprintf(priv->chip_vendor, "TSMC");
+			break;
+		case SYS_CFG_VENDOR_ID_SMIC:
+			sprintf(priv->chip_vendor, "SMIC");
+			priv->vendor_smic = 1;
+			break;
+		case SYS_CFG_VENDOR_ID_UMC:
+			sprintf(priv->chip_vendor, "UMC");
+			priv->vendor_umc = 1;
+			break;
+		default:
+			sprintf(priv->chip_vendor, "unknown");
+		}
+		break;
+	default:
+		if (val32 & SYS_CFG_VENDOR_ID) {
+			sprintf(priv->chip_vendor, "UMC");
+			priv->vendor_umc = 1;
+		} else {
+			sprintf(priv->chip_vendor, "TSMC");
+		}
+	}
 
 	val32 = rtl8xxxu_read32(priv, REG_GPIO_OUTSTS);
 	priv->rom_rev = (val32 & GPIO_RF_RL_ID) >> 28;
@@ -1934,7 +1982,7 @@ static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 				 raw[i + 6], raw[i + 7]);
 		}
 	}
-	return -EINVAL;
+	return 0;
 }
 
 static int
@@ -2269,6 +2317,7 @@ static int rtl8xxxu_load_firmware(struct rtl8xxxu_priv *priv, char *fw_name)
 
 	signature = le16_to_cpu(priv->fw_data->signature);
 	switch (signature & 0xfff0) {
+	case 0x92e0:
 	case 0x92c0:
 	case 0x88c0:
 	case 0x2300:
@@ -2338,13 +2387,7 @@ static int rtl8192eu_load_firmware(struct rtl8xxxu_priv *priv)
 	char *fw_name;
 	int ret;
 
-	return -EBUSY;
-	if (!priv->vendor_umc)
-		fw_name = "rtlwifi/rtl8192cufw_TMSC.bin";
-	else if (priv->chip_cut || priv->rtlchip == 0x8192c)
-		fw_name = "rtlwifi/rtl8192cufw_B.bin";
-	else
-		fw_name = "rtlwifi/rtl8192cufw_A.bin";
+	fw_name = "rtlwifi/rtl8192eu_nic.bin";
 
 	ret = rtl8xxxu_load_firmware(priv, fw_name);
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 8168a35..0d4ff80 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -625,6 +625,7 @@ struct rtl8xxxu_priv {
 
 	u8 mac_addr[ETH_ALEN];
 	char chip_name[8];
+	char chip_vendor[8];
 	u8 cck_tx_power_index_A[3];	/* 0x10 */
 	u8 cck_tx_power_index_B[3];
 	u8 ht40_1s_tx_power_index_A[3];	/* 0x16 */
@@ -647,6 +648,7 @@ struct rtl8xxxu_priv {
 	u32 has_gps:1;
 	u32 hi_pa:1;
 	u32 vendor_umc:1;
+	u32 vendor_smic:1;
 	u32 has_polarity_ctrl:1;
 	u32 has_eeprom:1;
 	u32 boot_eeprom:1;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 8f6c9c6..a934c0d 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -222,6 +222,10 @@
 #define  SYS_CFG_CHIP_VER		(BIT(12) | BIT(13) | BIT(14) | BIT(15))
 #define  SYS_CFG_BT_FUNC		BIT(16)
 #define  SYS_CFG_VENDOR_ID		BIT(19)
+#define  SYS_CFG_VENDOR_EXT_MASK	(BIT(18) | BIT(19))
+#define   SYS_CFG_VENDOR_ID_TSMC	0
+#define   SYS_CFG_VENDOR_ID_SMIC	BIT(18)
+#define   SYS_CFG_VENDOR_ID_UMC		BIT(19)
 #define  SYS_CFG_PAD_HWPD_IDN		BIT(22)
 #define  SYS_CFG_TRP_VAUX_EN		BIT(23)
 #define  SYS_CFG_TRP_BT_EN		BIT(24)
-- 
cgit v0.10.2


From 4a82ffe3f8040c5b3dd50888e00077d1cc8cbf29 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:01 -0500
Subject: rtl8xxxu: Use 1024 byte block loads for 8192eu firmware

The rtl8192eu can handle 1024 byte block writes, unlike it's
predecessors (8192cu/8188cu).

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 7459422..35c0625 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5975,7 +5975,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.parse_efuse = rtl8192eu_parse_efuse,
 	.load_firmware = rtl8192eu_load_firmware,
 	.power_on = rtl8192cu_power_on,
-	.writeN_block_size = 128,
+	.writeN_block_size = 1024,
 };
 
 static struct usb_device_id dev_table[] = {
-- 
cgit v0.10.2


From b001e086c4c8a5e60a2a28d40ded47ab124458f6 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:02 -0500
Subject: rtl8xxxu: Add rtl8192eu_nic.bin to the MODULE_FIRMWARE list

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 35c0625..e48815f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -54,6 +54,7 @@ MODULE_FIRMWARE("rtlwifi/rtl8723aufw_B_NoBT.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8192cufw_A.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8192cufw_B.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8192cufw_TMSC.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8192eu_nic.bin");
 
 module_param_named(debug, rtl8xxxu_debug, int, 0600);
 MODULE_PARM_DESC(debug, "Set debug mask");
-- 
cgit v0.10.2


From c05a9dbfb2b1f30b13126b0901fe4f1bdaf37406 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:03 -0500
Subject: rtl8xxxu: Implment rtl8192eu_power_on()

This implements the rtl8192eu power on sequence, and splits it off
from the rtl8192cu/rtl8723au power on sequence.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e48815f..55d12fb 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -3791,7 +3791,7 @@ exit:
 	return ret;
 }
 
-static void rtl8xxxu_disabled_to_emu(struct rtl8xxxu_priv *priv)
+static void rtl8723a_disabled_to_emu(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
 
@@ -3811,7 +3811,82 @@ static void rtl8xxxu_disabled_to_emu(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8);
 }
 
-static int rtl8xxxu_emu_to_active(struct rtl8xxxu_priv *priv)
+static void rtl8192e_disabled_to_emu(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+
+	/* Clear suspend enable and power down enable*/
+	val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1);
+	val8 &= ~(BIT(3) | BIT(4));
+	rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8);
+}
+
+static int rtl8192e_emu_to_active(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u32 val32;
+	int count, ret = 0;
+
+	/* disable HWPDN 0x04[15]=0*/
+	val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1);
+	val8 &= ~BIT(7);
+	rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8);
+
+	/* disable SW LPS 0x04[10]= 0 */
+	val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1);
+	val8 &= ~BIT(2);
+	rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8);
+
+	/* disable WL suspend*/
+	val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1);
+	val8 &= ~(BIT(3) | BIT(4));
+	rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8);
+
+	/* wait till 0x04[17] = 1 power ready*/
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+		if (val32 & BIT(17))
+			break;
+
+		udelay(10);
+	}
+
+	if (!count) {
+		ret = -EBUSY;
+		goto exit;
+	}
+
+	/* We should be able to optimize the following three entries into one */
+
+	/* release WLON reset 0x04[16]= 1*/
+	val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 2);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, REG_APS_FSMCO + 2, val8);
+
+	/* set, then poll until 0 */
+	val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+	val32 |= APS_FSMCO_MAC_ENABLE;
+	rtl8xxxu_write32(priv, REG_APS_FSMCO, val32);
+
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+		if ((val32 & APS_FSMCO_MAC_ENABLE) == 0) {
+			ret = 0;
+			break;
+		}
+		udelay(10);
+	}
+
+	if (!count) {
+		ret = -EBUSY;
+		goto exit;
+	}
+
+exit:
+	return ret;
+}
+
+static int rtl8723a_emu_to_active(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
 	u32 val32;
@@ -3940,9 +4015,9 @@ static int rtl8723au_power_on(struct rtl8xxxu_priv *priv)
 	 */
 	rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0);
 
-	rtl8xxxu_disabled_to_emu(priv);
+	rtl8723a_disabled_to_emu(priv);
 
-	ret = rtl8xxxu_emu_to_active(priv);
+	ret = rtl8723a_emu_to_active(priv);
 	if (ret)
 		goto exit;
 
@@ -4081,6 +4156,52 @@ static int rtl8192cu_power_on(struct rtl8xxxu_priv *priv)
 
 #endif
 
+static int rtl8192eu_power_on(struct rtl8xxxu_priv *priv)
+{
+	u16 val16;
+	u32 val32;
+	int ret;
+
+	ret = 0;
+
+	val32 = rtl8xxxu_read32(priv, REG_SYS_CFG);
+	if (val32 & SYS_CFG_SPS_LDO_SEL) {
+		rtl8xxxu_write8(priv, REG_LDO_SW_CTRL, 0xc3);
+	} else {
+		/*
+		 * Raise 1.2V voltage
+		 */
+		val32 = rtl8xxxu_read32(priv, REG_8192E_LDOV12_CTRL);
+		val32 &= 0xff0fffff;
+		val32 |= 0x00500000;
+		rtl8xxxu_write32(priv, REG_8192E_LDOV12_CTRL, val32);
+		rtl8xxxu_write8(priv, REG_LDO_SW_CTRL, 0x83);
+	}
+
+	rtl8192e_disabled_to_emu(priv);
+
+	ret = rtl8192e_emu_to_active(priv);
+	if (ret)
+		goto exit;
+
+	rtl8xxxu_write16(priv, REG_CR, 0x0000);
+
+	/*
+	 * Enable MAC DMA/WMAC/SCHEDULE/SEC block
+	 * Set CR bit10 to enable 32k calibration.
+	 */
+	val16 = rtl8xxxu_read16(priv, REG_CR);
+	val16 |= (CR_HCI_TXDMA_ENABLE | CR_HCI_RXDMA_ENABLE |
+		  CR_TXDMA_ENABLE | CR_RXDMA_ENABLE |
+		  CR_PROTOCOL_ENABLE | CR_SCHEDULE_ENABLE |
+		  CR_MAC_TX_ENABLE | CR_MAC_RX_ENABLE |
+		  CR_SECURITY_ENABLE | CR_CALTIMER_ENABLE);
+	rtl8xxxu_write16(priv, REG_CR, val16);
+
+exit:
+	return ret;
+}
+
 static void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
@@ -5975,8 +6096,8 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.parse_efuse = rtl8192eu_parse_efuse,
 	.load_firmware = rtl8192eu_load_firmware,
-	.power_on = rtl8192cu_power_on,
-	.writeN_block_size = 1024,
+	.power_on = rtl8192eu_power_on,
+	.writeN_block_size = 128,
 };
 
 static struct usb_device_id dev_table[] = {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index a934c0d..02c7d6c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -72,6 +72,7 @@
 #define REG_AFE_MISC			0x0010
 #define REG_SPS0_CTRL			0x0011
 #define REG_SPS_OCP_CFG			0x0018
+#define REG_8192E_LDOV12_CTRL		0x0014
 #define REG_RSV_CTRL			0x001c
 
 #define REG_RF_CTRL			0x001f
@@ -178,6 +179,8 @@
 						   control */
 #define  MULTI_GPS_FUNC_EN		BIT(22)	/* GPS function enable */
 
+#define REG_LDO_SW_CTRL			0x007c	/* 8192eu */
+
 #define REG_MCU_FW_DL			0x0080
 #define  MCU_FW_DL_ENABLE		BIT(0)
 #define  MCU_FW_DL_READY		BIT(1)
@@ -229,6 +232,7 @@
 #define  SYS_CFG_PAD_HWPD_IDN		BIT(22)
 #define  SYS_CFG_TRP_VAUX_EN		BIT(23)
 #define  SYS_CFG_TRP_BT_EN		BIT(24)
+#define  SYS_CFG_SPS_LDO_SEL		BIT(24)	/* 8192eu */
 #define  SYS_CFG_BD_PKG_SEL		BIT(25)
 #define  SYS_CFG_BD_HCI_SEL		BIT(26)
 #define  SYS_CFG_TYPE_ID		BIT(27)
@@ -261,6 +265,8 @@
 #define  GPIO_USB_SUSEN			BIT(23)
 #define  GPIO_RF_RL_ID			(BIT(31) | BIT(30) | BIT(29) | BIT(28))
 
+#define REG_SYS_CFG2			0x00fc	/* 8192eu */
+
 /* 0x0100 ~ 0x01FF	MACTOP General Configuration */
 #define REG_CR				0x0100
 #define  CR_HCI_TXDMA_ENABLE		BIT(0)
-- 
cgit v0.10.2


From 74b99bed874f5a17336515b63f776a8cbbbceb93 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:04 -0500
Subject: rtl8xxxu: Add rtl8xxxu_auto_llt_table()

Newer chips can auto load the LLT table, it is no longer necessary to
build it manually in the driver.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 55d12fb..b2e4eac4 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2774,6 +2774,34 @@ exit:
 	return ret;
 }
 
+static int rtl8xxxu_auto_llt_table(struct rtl8xxxu_priv *priv, u8 last_tx_page)
+{
+	u32 val32;
+	int ret = 0;
+	int i;
+
+	val32 = rtl8xxxu_read32(priv, REG_AUTO_LLT);
+	pr_info("AUTO_LLT = %08x\n", val32);
+	val32 |= AUTO_LLT_INIT_LLT;
+	rtl8xxxu_write32(priv, REG_AUTO_LLT, val32);
+
+	for (i = 500; i; i--) {
+		val32 = rtl8xxxu_read32(priv, REG_AUTO_LLT);
+		if (!(val32 & AUTO_LLT_INIT_LLT))
+			break;
+		usleep_range(2, 4);
+	}
+
+	if (i) {
+		ret = -EBUSY;
+		dev_warn(&priv->udev->dev, "LLT table init failed\n");
+	}
+	else
+		dev_warn(&priv->udev->dev, "LLT table init success\n");
+
+	return ret;
+}
+
 static int rtl8xxxu_init_queue_priority(struct rtl8xxxu_priv *priv)
 {
 	u16 val16, hi, lo;
@@ -4287,7 +4315,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	dev_dbg(dev, "%s: macpower %i\n", __func__, macpower);
 	if (!macpower) {
-		ret = rtl8xxxu_init_llt_table(priv, TX_TOTAL_PAGE_NUM);
+		ret = priv->fops->llt_init(priv, TX_TOTAL_PAGE_NUM);
 		if (ret) {
 			dev_warn(dev, "%s: LLT table init failed\n", __func__);
 			goto exit;
@@ -6079,6 +6107,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.parse_efuse = rtl8723au_parse_efuse,
 	.load_firmware = rtl8723au_load_firmware,
 	.power_on = rtl8723au_power_on,
+	.llt_init = rtl8xxxu_init_llt_table,
 	.writeN_block_size = 1024,
 };
 
@@ -6088,6 +6117,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.parse_efuse = rtl8192cu_parse_efuse,
 	.load_firmware = rtl8192cu_load_firmware,
 	.power_on = rtl8192cu_power_on,
+	.llt_init = rtl8xxxu_init_llt_table,
 	.writeN_block_size = 128,
 };
 
@@ -6097,6 +6127,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.parse_efuse = rtl8192eu_parse_efuse,
 	.load_firmware = rtl8192eu_load_firmware,
 	.power_on = rtl8192eu_power_on,
+	.llt_init = rtl8xxxu_auto_llt_table,
 	.writeN_block_size = 128,
 };
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 0d4ff80..56c4e0a 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -718,5 +718,6 @@ struct rtl8xxxu_fileops {
 	int (*parse_efuse) (struct rtl8xxxu_priv *priv);
 	int (*load_firmware) (struct rtl8xxxu_priv *priv);
 	int (*power_on) (struct rtl8xxxu_priv *priv);
+	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
 	int writeN_block_size;
 };
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 02c7d6c..b83bd34 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -372,6 +372,9 @@
 #define  RQPN_NPQ_SHIFT			0
 #define  RQPN_EPQ_SHIFT			16
 
+#define REG_AUTO_LLT			0x0224
+#define  AUTO_LLT_INIT_LLT		BIT(16)
+
 /* 0x0280 ~ 0x02FF	RXDMA Configuration */
 #define REG_RXDMA_AGG_PG_TH		0x0280
 #define REG_RXPKT_NUM			0x0284
-- 
cgit v0.10.2


From 07bb46be5306d27c8d8ce7b287688554b9ca1569 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:05 -0500
Subject: rtl8xxxu: Init page boundaries before starting the firmware

This reorganizes the device initialization to init page boundaries
before starting the firmware. This matches the flow in the 8192eu
vendor driver.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index b2e4eac4..65994e9 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4315,6 +4315,40 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	dev_dbg(dev, "%s: macpower %i\n", __func__, macpower);
 	if (!macpower) {
+		if (priv->ep_tx_normal_queue)
+			val8 = TX_PAGE_NUM_NORM_PQ;
+		else
+			val8 = 0;
+
+		rtl8xxxu_write8(priv, REG_RQPN_NPQ, val8);
+
+		val32 = (TX_PAGE_NUM_PUBQ << RQPN_NORM_PQ_SHIFT) | RQPN_LOAD;
+
+		if (priv->ep_tx_high_queue)
+			val32 |= (TX_PAGE_NUM_HI_PQ << RQPN_HI_PQ_SHIFT);
+		if (priv->ep_tx_low_queue)
+			val32 |= (TX_PAGE_NUM_LO_PQ << RQPN_LO_PQ_SHIFT);
+
+		rtl8xxxu_write32(priv, REG_RQPN, val32);
+
+		/*
+		 * Set TX buffer boundary
+		 */
+		val8 = TX_TOTAL_PAGE_NUM + 1;
+		rtl8xxxu_write8(priv, REG_TXPKTBUF_BCNQ_BDNY, val8);
+		rtl8xxxu_write8(priv, REG_TXPKTBUF_MGQ_BDNY, val8);
+		rtl8xxxu_write8(priv, REG_TXPKTBUF_WMAC_LBK_BF_HD, val8);
+		rtl8xxxu_write8(priv, REG_TRXFF_BNDY, val8);
+		rtl8xxxu_write8(priv, REG_TDECTRL + 1, val8);
+	}
+
+	ret = rtl8xxxu_init_queue_priority(priv);
+	dev_dbg(dev, "%s: init_queue_priority %i\n", __func__, ret);
+	if (ret)
+		goto exit;
+
+	dev_dbg(dev, "%s: macpower %i\n", __func__, macpower);
+	if (!macpower) {
 		ret = priv->fops->llt_init(priv, TX_TOTAL_PAGE_NUM);
 		if (ret) {
 			dev_warn(dev, "%s: LLT table init failed\n", __func__);
@@ -4392,40 +4426,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	priv->rf_mode_ag[0] = rtl8xxxu_read_rfreg(priv, RF_A,
 						  RF6052_REG_MODE_AG);
 
-	dev_dbg(dev, "%s: macpower %i\n", __func__, macpower);
-	if (!macpower) {
-		if (priv->ep_tx_normal_queue)
-			val8 = TX_PAGE_NUM_NORM_PQ;
-		else
-			val8 = 0;
-
-		rtl8xxxu_write8(priv, REG_RQPN_NPQ, val8);
-
-		val32 = (TX_PAGE_NUM_PUBQ << RQPN_NORM_PQ_SHIFT) | RQPN_LOAD;
-
-		if (priv->ep_tx_high_queue)
-			val32 |= (TX_PAGE_NUM_HI_PQ << RQPN_HI_PQ_SHIFT);
-		if (priv->ep_tx_low_queue)
-			val32 |= (TX_PAGE_NUM_LO_PQ << RQPN_LO_PQ_SHIFT);
-
-		rtl8xxxu_write32(priv, REG_RQPN, val32);
-
-		/*
-		 * Set TX buffer boundary
-		 */
-		val8 = TX_TOTAL_PAGE_NUM + 1;
-		rtl8xxxu_write8(priv, REG_TXPKTBUF_BCNQ_BDNY, val8);
-		rtl8xxxu_write8(priv, REG_TXPKTBUF_MGQ_BDNY, val8);
-		rtl8xxxu_write8(priv, REG_TXPKTBUF_WMAC_LBK_BF_HD, val8);
-		rtl8xxxu_write8(priv, REG_TRXFF_BNDY, val8);
-		rtl8xxxu_write8(priv, REG_TDECTRL + 1, val8);
-	}
-
-	ret = rtl8xxxu_init_queue_priority(priv);
-	dev_dbg(dev, "%s: init_queue_priority %i\n", __func__, ret);
-	if (ret)
-		goto exit;
-
 	/*
 	 * Set RX page boundary
 	 */
-- 
cgit v0.10.2


From a47b9d477c792f8c0c9b6419a25528566260916a Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:06 -0500
Subject: rtl8xxxu: Init the LLT after we start the firmware

To match the flow of the vendor driver, move the LLT init to after the
firmware is started.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 65994e9..ad5b15c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4342,6 +4342,15 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write8(priv, REG_TDECTRL + 1, val8);
 	}
 
+	ret = rtl8xxxu_download_firmware(priv);
+	dev_dbg(dev, "%s: download_fiwmare %i\n", __func__, ret);
+	if (ret)
+		goto exit;
+	ret = rtl8xxxu_start_firmware(priv);
+	dev_dbg(dev, "%s: start_fiwmare %i\n", __func__, ret);
+	if (ret)
+		goto exit;
+
 	ret = rtl8xxxu_init_queue_priority(priv);
 	dev_dbg(dev, "%s: init_queue_priority %i\n", __func__, ret);
 	if (ret)
@@ -4356,15 +4365,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		}
 	}
 
-	ret = rtl8xxxu_download_firmware(priv);
-	dev_dbg(dev, "%s: download_fiwmare %i\n", __func__, ret);
-	if (ret)
-		goto exit;
-	ret = rtl8xxxu_start_firmware(priv);
-	dev_dbg(dev, "%s: start_fiwmare %i\n", __func__, ret);
-	if (ret)
-		goto exit;
-
 	ret = rtl8xxxu_init_mac(priv, rtl8723a_mac_init_table);
 	dev_dbg(dev, "%s: init_mac %i\n", __func__, ret);
 	if (ret)
-- 
cgit v0.10.2


From 4de24819192eb507525fa24182c86c2073accd12 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:07 -0500
Subject: rtl8xxxu: Fix incorrect test for auto LLT failure

The logic for testing auto load failure in rtl8xxxu_auto_llt_table()
was inverted.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ad5b15c..68d6bb8 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2781,7 +2781,6 @@ static int rtl8xxxu_auto_llt_table(struct rtl8xxxu_priv *priv, u8 last_tx_page)
 	int i;
 
 	val32 = rtl8xxxu_read32(priv, REG_AUTO_LLT);
-	pr_info("AUTO_LLT = %08x\n", val32);
 	val32 |= AUTO_LLT_INIT_LLT;
 	rtl8xxxu_write32(priv, REG_AUTO_LLT, val32);
 
@@ -2792,12 +2791,10 @@ static int rtl8xxxu_auto_llt_table(struct rtl8xxxu_priv *priv, u8 last_tx_page)
 		usleep_range(2, 4);
 	}
 
-	if (i) {
+	if (!i) {
 		ret = -EBUSY;
 		dev_warn(&priv->udev->dev, "LLT table init failed\n");
 	}
-	else
-		dev_warn(&priv->udev->dev, "LLT table init success\n");
 
 	return ret;
 }
-- 
cgit v0.10.2


From b63d0aaca62be6eed15d3829174307ac2d608623 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:08 -0500
Subject: rtl8xxxu: Kludge to drop incorrect USB OUT EP for 8192EU

The 8192eu (and some other parts) will report an incorrect USB OUT
EP. This tells the chip to drop it - as per the vendor driver.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 68d6bb8..b6d8014 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4362,6 +4362,12 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		}
 	}
 
+	if (priv->rtlchip == 0x8192e) {
+		val32 = rtl8xxxu_read32(priv, REG_TXDMA_OFFSET_CHK);
+		val32 |= TXDMA_OFFSET_DROP_DATA_EN;
+		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32);
+	}
+
 	ret = rtl8xxxu_init_mac(priv, rtl8723a_mac_init_table);
 	dev_dbg(dev, "%s: init_mac %i\n", __func__, ret);
 	if (ret)
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index b83bd34..226a1fa 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -367,6 +367,7 @@
 #define REG_FIFOPAGE			0x0204
 #define REG_TDECTRL			0x0208
 #define REG_TXDMA_OFFSET_CHK		0x020c
+#define  TXDMA_OFFSET_DROP_DATA_EN	BIT(9)
 #define REG_TXDMA_STATUS		0x0210
 #define REG_RQPN_NPQ			0x0214
 #define  RQPN_NPQ_SHIFT			0
-- 
cgit v0.10.2


From 99ad16cbea15a6e61acca5662eff8d195cc6fbf6 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:09 -0500
Subject: rtl8xxxu: Init REG_HIMR[01] for 8192eu parts

The newer generation chips have different interrupt registers.
Initialize this correct registers on 8192eu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index b6d8014..e4d8127 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4366,6 +4366,9 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		val32 = rtl8xxxu_read32(priv, REG_TXDMA_OFFSET_CHK);
 		val32 |= TXDMA_OFFSET_DROP_DATA_EN;
 		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32);
+
+		rtl8xxxu_write32(priv, REG_HIMR0, 0x00);
+		rtl8xxxu_write32(priv, REG_HIMR1, 0x00);
 	}
 
 	ret = rtl8xxxu_init_mac(priv, rtl8723a_mac_init_table);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 226a1fa..3cbb549 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -196,6 +196,10 @@
 #define REG_HMBOX_EXT_1			0x008a
 #define REG_HMBOX_EXT_2			0x008c
 #define REG_HMBOX_EXT_3			0x008e
+/* Interrupt registers for 8192e/8812 */
+#define REG_HIMR0			0x00b0
+#define REG_HIMR1			0x00b8
+
 /*  Host suspend counter on FPGA platform */
 #define REG_HOST_SUSP_CNT		0x00bc
 /*  Efuse access protection for RTL8723 */
-- 
cgit v0.10.2


From 35a741febfae3cfc2a27d3b4935e255585ecfd81 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:10 -0500
Subject: rtl8xxxu: Initial rtl8723bu chip identification

This provides initial detection of 8723bu devices, and selects the
correct firmware image to load.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e4d8127..729f6d2 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -55,6 +55,8 @@ MODULE_FIRMWARE("rtlwifi/rtl8192cufw_A.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8192cufw_B.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8192cufw_TMSC.bin");
 MODULE_FIRMWARE("rtlwifi/rtl8192eu_nic.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8723bu_nic.bin");
+MODULE_FIRMWARE("rtlwifi/rtl8723bu_bt.bin");
 
 module_param_named(debug, rtl8xxxu_debug, int, 0600);
 MODULE_PARM_DESC(debug, "Set debug mask");
@@ -1700,11 +1702,17 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 	}
 
 	if (val32 & SYS_CFG_BT_FUNC) {
-		sprintf(priv->chip_name, "8723AU");
+		if (priv->chip_cut >= 3) {
+			sprintf(priv->chip_name, "8723BU");
+			priv->rtlchip = 0x8723b;
+		} else {
+			sprintf(priv->chip_name, "8723AU");
+			priv->rtlchip = 0x8723a;
+		}
+
 		priv->rf_paths = 1;
 		priv->rx_paths = 1;
 		priv->tx_paths = 1;
-		priv->rtlchip = 0x8723a;
 
 		val32 = rtl8xxxu_read32(priv, REG_MULTI_FUNC_CTRL);
 		if (val32 & MULTI_WIFI_FUNC_EN)
@@ -1807,6 +1815,7 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 	 */
 	if (!priv->ep_tx_count) {
 		switch (priv->nr_out_eps) {
+		case 4:
 		case 3:
 			priv->ep_tx_low_queue = 1;
 			priv->ep_tx_count++;
@@ -2321,6 +2330,7 @@ static int rtl8xxxu_load_firmware(struct rtl8xxxu_priv *priv, char *fw_name)
 	case 0x92e0:
 	case 0x92c0:
 	case 0x88c0:
+	case 0x5300:
 	case 0x2300:
 		break;
 	default:
@@ -2362,6 +2372,20 @@ static int rtl8723au_load_firmware(struct rtl8xxxu_priv *priv)
 	return ret;
 }
 
+static int rtl8723bu_load_firmware(struct rtl8xxxu_priv *priv)
+{
+	char *fw_name;
+	int ret;
+
+	if (priv->enable_bluetooth)
+		fw_name = "rtlwifi/rtl8723bu_bt.bin";
+	else
+		fw_name = "rtlwifi/rtl8723bu_nic.bin";
+
+	ret = rtl8xxxu_load_firmware(priv, fw_name);
+	return ret;
+}
+
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 
 static int rtl8192cu_load_firmware(struct rtl8xxxu_priv *priv)
@@ -2587,7 +2611,7 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 	else
 		rtl8xxxu_init_phy_regs(priv, rtl8xxx_agc_standard_table);
 
-	if (priv->rtlchip == 0x8723a &&
+	if ((priv->rtlchip == 0x8723a || priv->rtlchip == 0x8723b) &&
 	    priv->efuse_wifi.efuse8723.version >= 0x01) {
 		val32 = rtl8xxxu_read32(priv, REG_MAC_PHY_CTRL);
 
@@ -6117,6 +6141,14 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.writeN_block_size = 1024,
 };
 
+static struct rtl8xxxu_fileops rtl8723bu_fops = {
+	.parse_efuse = rtl8723au_parse_efuse,
+	.load_firmware = rtl8723bu_load_firmware,
+	.power_on = rtl8723au_power_on,
+	.llt_init = rtl8xxxu_auto_llt_table,
+	.writeN_block_size = 128,
+};
+
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 
 static struct rtl8xxxu_fileops rtl8192cu_fops = {
@@ -6146,6 +6178,8 @@ static struct usb_device_id dev_table[] = {
 	.driver_info = (unsigned long)&rtl8723au_fops},
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x818b, 0xff, 0xff, 0xff),
 	.driver_info = (unsigned long)&rtl8192eu_fops},
+{USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0xb720, 0xff, 0xff, 0xff),
+	.driver_info = (unsigned long)&rtl8723bu_fops},
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 /* Still supported by rtlwifi */
 {USB_DEVICE_AND_INTERFACE_INFO(USB_VENDOR_ID_REALTEK, 0x8176, 0xff, 0xff, 0xff),
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 56c4e0a..dd4c563 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -34,7 +34,7 @@
 #define RTL8XXXU_MAX_REG_POLL		500
 #define	USB_INTR_CONTENT_LENGTH		56
 
-#define RTL8XXXU_OUT_ENDPOINTS		3
+#define RTL8XXXU_OUT_ENDPOINTS		4
 
 #define REALTEK_USB_READ		0xc0
 #define REALTEK_USB_WRITE		0x40
-- 
cgit v0.10.2


From 3c836d6014e5057fc5845829ae762190008bded4 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:11 -0500
Subject: rtl8xxxu: Add rtl8723bu_parse_efuse() and 8723bu efuse definition

Implement first stab at parsing the 8723bu's efuse.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 729f6d2..255300c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1877,6 +1877,51 @@ static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 	return 0;
 }
 
+static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
+{
+	if (priv->efuse_wifi.efuse8723bu.rtl_id != cpu_to_le16(0x8129))
+		return -EINVAL;
+
+	ether_addr_copy(priv->mac_addr, priv->efuse_wifi.efuse8723bu.mac_addr);
+
+	memcpy(priv->cck_tx_power_index_A,
+	       priv->efuse_wifi.efuse8723bu.cck_tx_power_index_A,
+	       sizeof(priv->cck_tx_power_index_A));
+	memcpy(priv->cck_tx_power_index_B,
+	       priv->efuse_wifi.efuse8723bu.cck_tx_power_index_B,
+	       sizeof(priv->cck_tx_power_index_B));
+
+	memcpy(priv->ht40_1s_tx_power_index_A,
+	       priv->efuse_wifi.efuse8723bu.ht40_1s_tx_power_index_A,
+	       sizeof(priv->ht40_1s_tx_power_index_A));
+	memcpy(priv->ht40_1s_tx_power_index_B,
+	       priv->efuse_wifi.efuse8723bu.ht40_1s_tx_power_index_B,
+	       sizeof(priv->ht40_1s_tx_power_index_B));
+
+	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
+		 priv->efuse_wifi.efuse8723bu.vendor_name);
+	dev_info(&priv->udev->dev, "Product: %.41s\n",
+		 priv->efuse_wifi.efuse8723bu.device_name);
+
+	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_EFUSE) {
+		int i;
+		unsigned char *raw = priv->efuse_wifi.raw;
+
+		dev_info(&priv->udev->dev,
+			 "%s: dumping efuse (0x%02zx bytes):\n",
+			 __func__, sizeof(struct rtl8723bu_efuse));
+		for (i = 0; i < sizeof(struct rtl8723bu_efuse); i += 8) {
+			dev_info(&priv->udev->dev, "%02x: "
+				 "%02x %02x %02x %02x %02x %02x %02x %02x\n", i,
+				 raw[i], raw[i + 1], raw[i + 2],
+				 raw[i + 3], raw[i + 4], raw[i + 5],
+				 raw[i + 6], raw[i + 7]);
+		}
+	}
+
+	return 0;
+}
+
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 
 static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv)
@@ -6142,7 +6187,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 };
 
 static struct rtl8xxxu_fileops rtl8723bu_fops = {
-	.parse_efuse = rtl8723au_parse_efuse,
+	.parse_efuse = rtl8723bu_parse_efuse,
 	.load_firmware = rtl8723bu_load_firmware,
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index dd4c563..bf424db 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -500,6 +500,47 @@ struct rtl8192cu_efuse {
 	u8 customer_id;
 };
 
+struct rtl8723bu_efuse {
+	__le16 rtl_id;
+	u8 res0[0x0e];
+	u8 cck_tx_power_index_A[3];	/* 0x10 */
+	u8 cck_tx_power_index_B[3];
+	u8 ht40_1s_tx_power_index_A[3];	/* 0x16 */
+	u8 ht40_1s_tx_power_index_B[3];
+	u8 res1[0x9c];
+	u8 channel_plan;		/* 0xb8 */
+	u8 xtal_k;
+	u8 thermal_meter;
+	u8 iqk_lck;
+	u8 pa_type;			/* 0xbc */
+	u8 lna_type_2g;			/* 0xbd */
+	u8 res2[3];
+	u8 rf_board_option;
+	u8 rf_feature_option;
+	u8 rf_bt_setting;
+	u8 eeprom_version;
+	u8 eeprom_customer_id;
+	u8 res3[2];
+	u8 tx_pwr_calibrate_rate;
+	u8 rf_antenna_option;		/* 0xc9 */
+	u8 rfe_option;
+	u8 res4[9];
+	u8 usb_optional_function;
+	u8 res5[0x1e];
+	u8 res6[2];
+	u8 serial[0x0b];		/* 0xf5 */
+	u8 vid;				/* 0x100 */
+	u8 res7;
+	u8 pid;
+	u8 res8[4];
+	u8 mac_addr[ETH_ALEN];		/* 0x107 */
+	u8 res9[2];
+	u8 vendor_name[0x07];
+	u8 res10[2];
+	u8 device_name[0x14];		/* 0xe8 */
+	u8 res11[0xd4];
+};
+
 struct rtl8192eu_efuse {
 	__le16 rtl_id;
 	u8 res0[0x0e];
@@ -690,6 +731,7 @@ struct rtl8xxxu_priv {
 	union {
 		u8 raw[EFUSE_MAP_LEN];
 		struct rtl8723au_efuse efuse8723;
+		struct rtl8723bu_efuse efuse8723bu;
 		struct rtl8192cu_efuse efuse8192;
 		struct rtl8192eu_efuse efuse8192eu;
 	} efuse_wifi;
-- 
cgit v0.10.2


From adfc01243f0fc891c630f27ef45320fea71bb98e Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:12 -0500
Subject: rtl8xxxu: Use 1024 byte writes for writing 8723bu firmware

The 8723bu, like the 8192eu, can also handle 1024 byte block writes.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 255300c..93f6b71 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6191,7 +6191,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.load_firmware = rtl8723bu_load_firmware,
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
-	.writeN_block_size = 128,
+	.writeN_block_size = 1024,
 };
 
 #ifdef CONFIG_RTL8XXXU_UNTESTED
-- 
cgit v0.10.2


From 0e28b9753a9cf462a535c56e2d31ee17521a4b73 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:13 -0500
Subject: rtl8xxxu: Only setup USB interrupts for parts which support it

Only 1st generation chips do provide USB interrupts, so do not try to
setup interrupts for newer chips (8192eu and 8723bu).

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 93f6b71..467c0bca 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1707,6 +1707,7 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 			priv->rtlchip = 0x8723b;
 		} else {
 			sprintf(priv->chip_name, "8723AU");
+			priv->usb_interrupts = 1;
 			priv->rtlchip = 0x8723a;
 		}
 
@@ -1744,12 +1745,14 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 			priv->rf_paths = 2;
 			priv->rx_paths = 2;
 			priv->tx_paths = 1;
+			priv->usb_interrupts = 1;
 			priv->rtlchip = 0x8191c;
 		} else {
 			sprintf(priv->chip_name, "8192CU");
 			priv->rf_paths = 2;
 			priv->rx_paths = 2;
 			priv->tx_paths = 2;
+			priv->usb_interrupts = 1;
 			priv->rtlchip = 0x8192c;
 		}
 		priv->has_wifi = 1;
@@ -1759,6 +1762,7 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 		priv->rx_paths = 1;
 		priv->tx_paths = 1;
 		priv->rtlchip = 0x8188c;
+		priv->usb_interrupts = 1;
 		priv->has_wifi = 1;
 	}
 
@@ -5825,9 +5829,11 @@ static int rtl8xxxu_start(struct ieee80211_hw *hw)
 	init_usb_anchor(&priv->int_anchor);
 
 	rtl8723a_enable_rf(priv);
-	ret = rtl8xxxu_submit_int_urb(hw);
-	if (ret)
-		goto exit;
+	if (priv->usb_interrupts) {
+		ret = rtl8xxxu_submit_int_urb(hw);
+		if (ret)
+			goto exit;
+	}
 
 	for (i = 0; i < RTL8XXXU_TX_URBS; i++) {
 		tx_urb = kmalloc(sizeof(struct rtl8xxxu_tx_urb), GFP_KERNEL);
@@ -5902,14 +5908,16 @@ static void rtl8xxxu_stop(struct ieee80211_hw *hw)
 
 	usb_kill_anchored_urbs(&priv->rx_anchor);
 	usb_kill_anchored_urbs(&priv->tx_anchor);
-	usb_kill_anchored_urbs(&priv->int_anchor);
+	if (priv->usb_interrupts)
+		usb_kill_anchored_urbs(&priv->int_anchor);
 
 	rtl8723a_disable_rf(priv);
 
 	/*
 	 * Disable interrupts
 	 */
-	rtl8xxxu_write32(priv, REG_USB_HIMR, 0);
+	if (priv->usb_interrupts)
+		rtl8xxxu_write32(priv, REG_USB_HIMR, 0);
 
 	rtl8xxxu_free_rx_resources(priv);
 	rtl8xxxu_free_tx_resources(priv);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index bf424db..2055048 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -693,6 +693,7 @@ struct rtl8xxxu_priv {
 	u32 has_polarity_ctrl:1;
 	u32 has_eeprom:1;
 	u32 boot_eeprom:1;
+	u32 usb_interrupts:1;
 	u32 ep_tx_high_queue:1;
 	u32 ep_tx_normal_queue:1;
 	u32 ep_tx_low_queue:1;
-- 
cgit v0.10.2


From 36c32588c83df156e80e485e48a827cd57589620 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:14 -0500
Subject: rtl8xxxu: Add rtl8723b_phy_1t_init_table

This adds the 8723bu PHY 1T init table.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 467c0bca..cc91247 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -251,6 +251,107 @@ static struct rtl8xxxu_reg32val rtl8723a_phy_1t_init_table[] = {
 	{0xffff, 0xffffffff},
 };
 
+static struct rtl8xxxu_reg32val rtl8723b_phy_1t_init_table[] = {
+	{0x800, 0x80040000}, {0x804, 0x00000003},
+	{0x808, 0x0000fc00}, {0x80c, 0x0000000a},
+	{0x810, 0x10001331}, {0x814, 0x020c3d10},
+	{0x818, 0x02200385}, {0x81c, 0x00000000},
+	{0x820, 0x01000100}, {0x824, 0x00190204},
+	{0x828, 0x00000000}, {0x82c, 0x00000000},
+	{0x830, 0x00000000}, {0x834, 0x00000000},
+	{0x838, 0x00000000}, {0x83c, 0x00000000},
+	{0x840, 0x00010000}, {0x844, 0x00000000},
+	{0x848, 0x00000000}, {0x84c, 0x00000000},
+	{0x850, 0x00000000}, {0x854, 0x00000000},
+	{0x858, 0x569a11a9}, {0x85c, 0x01000014},
+	{0x860, 0x66f60110}, {0x864, 0x061f0649},
+	{0x868, 0x00000000}, {0x86c, 0x27272700},
+	{0x870, 0x07000760}, {0x874, 0x25004000},
+	{0x878, 0x00000808}, {0x87c, 0x00000000},
+	{0x880, 0xb0000c1c}, {0x884, 0x00000001},
+	{0x888, 0x00000000}, {0x88c, 0xccc000c0},
+	{0x890, 0x00000800}, {0x894, 0xfffffffe},
+	{0x898, 0x40302010}, {0x89c, 0x00706050},
+	{0x900, 0x00000000}, {0x904, 0x00000023},
+	{0x908, 0x00000000}, {0x90c, 0x81121111},
+	{0x910, 0x00000002}, {0x914, 0x00000201},
+	{0xa00, 0x00d047c8}, {0xa04, 0x80ff800c},
+	{0xa08, 0x8c838300}, {0xa0c, 0x2e7f120f},
+	{0xa10, 0x9500bb78}, {0xa14, 0x1114d028},
+	{0xa18, 0x00881117}, {0xa1c, 0x89140f00},
+	{0xa20, 0x1a1b0000}, {0xa24, 0x090e1317},
+	{0xa28, 0x00000204}, {0xa2c, 0x00d30000},
+	{0xa70, 0x101fbf00}, {0xa74, 0x00000007},
+	{0xa78, 0x00000900}, {0xa7c, 0x225b0606},
+	{0xa80, 0x21806490}, {0xb2c, 0x00000000},
+	{0xc00, 0x48071d40}, {0xc04, 0x03a05611},
+	{0xc08, 0x000000e4}, {0xc0c, 0x6c6c6c6c},
+	{0xc10, 0x08800000}, {0xc14, 0x40000100},
+	{0xc18, 0x08800000}, {0xc1c, 0x40000100},
+	{0xc20, 0x00000000}, {0xc24, 0x00000000},
+	{0xc28, 0x00000000}, {0xc2c, 0x00000000},
+	{0xc30, 0x69e9ac44}, {0xc34, 0x469652af},
+	{0xc38, 0x49795994}, {0xc3c, 0x0a97971c},
+	{0xc40, 0x1f7c403f}, {0xc44, 0x000100b7},
+	{0xc48, 0xec020107}, {0xc4c, 0x007f037f},
+	{0xc50, 0x69553420}, {0xc54, 0x43bc0094},
+	{0xc58, 0x00013149}, {0xc5c, 0x00250492},
+	{0xc60, 0x00000000}, {0xc64, 0x7112848b},
+	{0xc68, 0x47c00bff}, {0xc6c, 0x00000036},
+	{0xc70, 0x2c7f000d}, {0xc74, 0x020610db},
+	{0xc78, 0x0000001f}, {0xc7c, 0x00b91612},
+	{0xc80, 0x390000e4}, {0xc84, 0x20f60000},
+	{0xc88, 0x40000100}, {0xc8c, 0x20200000},
+	{0xc90, 0x00020e1a}, {0xc94, 0x00000000},
+	{0xc98, 0x00020e1a}, {0xc9c, 0x00007f7f},
+	{0xca0, 0x00000000}, {0xca4, 0x000300a0},
+	{0xca8, 0x00000000}, {0xcac, 0x00000000},
+	{0xcb0, 0x00000000}, {0xcb4, 0x00000000},
+	{0xcb8, 0x00000000}, {0xcbc, 0x28000000},
+	{0xcc0, 0x00000000}, {0xcc4, 0x00000000},
+	{0xcc8, 0x00000000}, {0xccc, 0x00000000},
+	{0xcd0, 0x00000000}, {0xcd4, 0x00000000},
+	{0xcd8, 0x64b22427}, {0xcdc, 0x00766932},
+	{0xce0, 0x00222222}, {0xce4, 0x00000000},
+	{0xce8, 0x37644302}, {0xcec, 0x2f97d40c},
+	{0xd00, 0x00000740}, {0xd04, 0x40020401},
+	{0xd08, 0x0000907f}, {0xd0c, 0x20010201},
+	{0xd10, 0xa0633333}, {0xd14, 0x3333bc53},
+	{0xd18, 0x7a8f5b6f}, {0xd2c, 0xcc979975},
+	{0xd30, 0x00000000}, {0xd34, 0x80608000},
+	{0xd38, 0x00000000}, {0xd3c, 0x00127353},
+	{0xd40, 0x00000000}, {0xd44, 0x00000000},
+	{0xd48, 0x00000000}, {0xd4c, 0x00000000},
+	{0xd50, 0x6437140a}, {0xd54, 0x00000000},
+	{0xd58, 0x00000282}, {0xd5c, 0x30032064},
+	{0xd60, 0x4653de68}, {0xd64, 0x04518a3c},
+	{0xd68, 0x00002101}, {0xd6c, 0x2a201c16},
+	{0xd70, 0x1812362e}, {0xd74, 0x322c2220},
+	{0xd78, 0x000e3c24}, {0xe00, 0x2d2d2d2d},
+	{0xe04, 0x2d2d2d2d}, {0xe08, 0x0390272d},
+	{0xe10, 0x2d2d2d2d}, {0xe14, 0x2d2d2d2d},
+	{0xe18, 0x2d2d2d2d}, {0xe1c, 0x2d2d2d2d},
+	{0xe28, 0x00000000}, {0xe30, 0x1000dc1f},
+	{0xe34, 0x10008c1f}, {0xe38, 0x02140102},
+	{0xe3c, 0x681604c2}, {0xe40, 0x01007c00},
+	{0xe44, 0x01004800}, {0xe48, 0xfb000000},
+	{0xe4c, 0x000028d1}, {0xe50, 0x1000dc1f},
+	{0xe54, 0x10008c1f}, {0xe58, 0x02140102},
+	{0xe5c, 0x28160d05}, {0xe60, 0x00000008},
+	{0xe68, 0x001b2556}, {0xe6c, 0x00c00096},
+	{0xe70, 0x00c00096}, {0xe74, 0x01000056},
+	{0xe78, 0x01000014}, {0xe7c, 0x01000056},
+	{0xe80, 0x01000014}, {0xe84, 0x00c00096},
+	{0xe88, 0x01000056}, {0xe8c, 0x00c00096},
+	{0xed0, 0x00c00096}, {0xed4, 0x00c00096},
+	{0xed8, 0x00c00096}, {0xedc, 0x000000d6},
+	{0xee0, 0x000000d6}, {0xeec, 0x01c00016},
+	{0xf14, 0x00000003}, {0xf4c, 0x00000000},
+	{0xf00, 0x00000300},
+	{0x820, 0x01000100}, {0x800, 0x83040000},
+	{0xffff, 0xffffffff},
+};
+
 static struct rtl8xxxu_reg32val rtl8192cu_phy_2t_init_table[] = {
 	{0x024, 0x0011800f}, {0x028, 0x00ffdb83},
 	{0x800, 0x80040002}, {0x804, 0x00000003},
@@ -2589,6 +2690,8 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 		rtl8xxxu_init_phy_regs(priv, rtl8188ru_phy_1t_highpa_table);
 	else if (priv->tx_paths == 2)
 		rtl8xxxu_init_phy_regs(priv, rtl8192cu_phy_2t_init_table);
+	else if (priv->rtlchip == 0x8723b)
+		rtl8xxxu_init_phy_regs(priv, rtl8723b_phy_1t_init_table);
 	else
 		rtl8xxxu_init_phy_regs(priv, rtl8723a_phy_1t_init_table);
 
-- 
cgit v0.10.2


From 22a31d455ce20737e647f8c9cc37ac9080deee96 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:15 -0500
Subject: rtl8xxxu: Add rtl8723bu_radioa_1t_init_table

Add 8723bu 1T radio init table. The vendor driver indicates that some
registers need special treatment for TFBGA90, TFBGA80, and TFBGA79
packaging. However the vendor driver never actually checks the package
type, so just stick to default values here.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index cc91247..7d62440 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -792,6 +792,75 @@ static struct rtl8xxxu_rfregval rtl8723au_radioa_1t_init_table[] = {
 	{0xff, 0xffffffff}
 };
 
+static struct rtl8xxxu_rfregval rtl8723bu_radioa_1t_init_table[] = {
+	{0x00, 0x00010000}, {0xb0, 0x000dffe0},
+	{0xfe, 0x00000000}, {0xfe, 0x00000000},
+	{0xfe, 0x00000000}, {0xb1, 0x00000018},
+	{0xfe, 0x00000000}, {0xfe, 0x00000000},
+	{0xfe, 0x00000000}, {0xb2, 0x00084c00},
+	{0xb5, 0x0000d2cc}, {0xb6, 0x000925aa},
+	{0xb7, 0x00000010}, {0xb8, 0x0000907f},
+	{0x5c, 0x00000002}, {0x7c, 0x00000002},
+	{0x7e, 0x00000005}, {0x8b, 0x0006fc00},
+	{0xb0, 0x000ff9f0}, {0x1c, 0x000739d2},
+	{0x1e, 0x00000000}, {0xdf, 0x00000780},
+	{0x50, 0x00067435},
+	/*
+	 * The 8723bu vendor driver indicates that bit 8 should be set in
+	 * 0x51 for package types TFBGA90, TFBGA80, and TFBGA79. However
+	 * they never actually check the package type - and just default
+	 * to not setting it.
+	 */
+	{0x51, 0x0006b04e},
+	{0x52, 0x000007d2}, {0x53, 0x00000000},
+	{0x54, 0x00050400}, {0x55, 0x0004026e},
+	{0xdd, 0x0000004c}, {0x70, 0x00067435},
+	/*
+	 * 0x71 has same package type condition as for register 0x51
+	 */
+	{0x71, 0x0006b04e},
+	{0x72, 0x000007d2}, {0x73, 0x00000000},
+	{0x74, 0x00050400}, {0x75, 0x0004026e},
+	{0xef, 0x00000100}, {0x34, 0x0000add7},
+	{0x35, 0x00005c00}, {0x34, 0x00009dd4},
+	{0x35, 0x00005000}, {0x34, 0x00008dd1},
+	{0x35, 0x00004400}, {0x34, 0x00007dce},
+	{0x35, 0x00003800}, {0x34, 0x00006cd1},
+	{0x35, 0x00004400}, {0x34, 0x00005cce},
+	{0x35, 0x00003800}, {0x34, 0x000048ce},
+	{0x35, 0x00004400}, {0x34, 0x000034ce},
+	{0x35, 0x00003800}, {0x34, 0x00002451},
+	{0x35, 0x00004400}, {0x34, 0x0000144e},
+	{0x35, 0x00003800}, {0x34, 0x00000051},
+	{0x35, 0x00004400}, {0xef, 0x00000000},
+	{0xef, 0x00000100}, {0xed, 0x00000010},
+	{0x44, 0x0000add7}, {0x44, 0x00009dd4},
+	{0x44, 0x00008dd1}, {0x44, 0x00007dce},
+	{0x44, 0x00006cc1}, {0x44, 0x00005cce},
+	{0x44, 0x000044d1}, {0x44, 0x000034ce},
+	{0x44, 0x00002451}, {0x44, 0x0000144e},
+	{0x44, 0x00000051}, {0xef, 0x00000000},
+	{0xed, 0x00000000}, {0x7f, 0x00020080},
+	{0xef, 0x00002000}, {0x3b, 0x000380ef},
+	{0x3b, 0x000302fe}, {0x3b, 0x00028ce6},
+	{0x3b, 0x000200bc}, {0x3b, 0x000188a5},
+	{0x3b, 0x00010fbc}, {0x3b, 0x00008f71},
+	{0x3b, 0x00000900}, {0xef, 0x00000000},
+	{0xed, 0x00000001}, {0x40, 0x000380ef},
+	{0x40, 0x000302fe}, {0x40, 0x00028ce6},
+	{0x40, 0x000200bc}, {0x40, 0x000188a5},
+	{0x40, 0x00010fbc}, {0x40, 0x00008f71},
+	{0x40, 0x00000900}, {0xed, 0x00000000},
+	{0x82, 0x00080000}, {0x83, 0x00008000},
+	{0x84, 0x00048d80}, {0x85, 0x00068000},
+	{0xa2, 0x00080000}, {0xa3, 0x00008000},
+	{0xa4, 0x00048d80}, {0xa5, 0x00068000},
+	{0xed, 0x00000002}, {0xef, 0x00000002},
+	{0x56, 0x00000032}, {0x76, 0x00000032},
+	{0x01, 0x00000780},
+	{0xff, 0xffffffff}
+};
+
 static struct rtl8xxxu_rfregval rtl8192cu_radioa_2t_init_table[] = {
 	{0x00, 0x00030159}, {0x01, 0x00031284},
 	{0x02, 0x00098000}, {0x03, 0x00018c63},
@@ -1270,6 +1339,11 @@ static u32 rtl8xxxu_read_rfreg(struct rtl8xxxu_priv *priv,
 	return retval;
 }
 
+/*
+ * The RTL8723BU driver indicates that registers 0xb2 and 0xb6 can
+ * have write issues in high temperature conditions. We may have to
+ * retry writing them.
+ */
 static int rtl8xxxu_write_rfreg(struct rtl8xxxu_priv *priv,
 				enum rtl8xxxu_rfpath path, u8 reg, u32 data)
 {
@@ -4562,6 +4636,10 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rftable = rtl8723au_radioa_1t_init_table;
 		ret = rtl8xxxu_init_phy_rf(priv, rftable, RF_A);
 		break;
+	case 0x8723b:
+		rftable = rtl8723bu_radioa_1t_init_table;
+		ret = rtl8xxxu_init_phy_rf(priv, rftable, RF_A);
+		break;
 	case 0x8188c:
 		if (priv->hi_pa)
 			rftable = rtl8188ru_radioa_1t_highpa_table;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 2055048..326fda5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -537,8 +537,10 @@ struct rtl8723bu_efuse {
 	u8 res9[2];
 	u8 vendor_name[0x07];
 	u8 res10[2];
-	u8 device_name[0x14];		/* 0xe8 */
-	u8 res11[0xd4];
+	u8 device_name[0x14];
+	u8 res11[0xcf];
+	u8 package_type;		/* 0x1fb */
+	u8 res12[0x4];
 };
 
 struct rtl8192eu_efuse {
-- 
cgit v0.10.2


From f0d9f5e90717a5e1ab014d63590d10c806ab15b0 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:16 -0500
Subject: rtl8xxxu: Add rtl8723bu_phy_init_antenna_selection()

So far this is just for 8723BU. It includes writing to a number of
registers I have seen no description for so far.

0x0064
0x0930
0x0944

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 7d62440..c55193b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2670,6 +2670,38 @@ static void rtl8xxxu_firmware_self_reset(struct rtl8xxxu_priv *priv)
 	}
 }
 
+static void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+
+	val32 = rtl8xxxu_read32(priv, 0x64);
+	val32 &= ~(BIT(20) | BIT(24));
+	rtl8xxxu_write32(priv, 0x64, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_GPIO_MUXCFG);
+	val32 &= ~BIT(4);
+	val32 |= BIT(3);
+	rtl8xxxu_write32(priv, REG_GPIO_MUXCFG, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_LEDCFG0);
+	val32 &= ~BIT(23);
+	val32 |= BIT(24);
+	rtl8xxxu_write32(priv, REG_LEDCFG0, val32);
+
+	val32 = rtl8xxxu_read32(priv, 0x0944);
+	val32 |= (BIT(0) | BIT(1));
+	rtl8xxxu_write32(priv, 0x0944, val32);
+
+	val32 = rtl8xxxu_read32(priv, 0x0930);
+	val32 &= 0xffffff00;
+	val32 |= 0x77;
+	rtl8xxxu_write32(priv, 0x0930, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_PWR_DATA);
+	val32 |= BIT(11);
+	rtl8xxxu_write32(priv, REG_PWR_DATA, val32);
+}
+
 static int
 rtl8xxxu_init_mac(struct rtl8xxxu_priv *priv, struct rtl8xxxu_reg8val *array)
 {
@@ -4621,6 +4653,9 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write32(priv, REG_HIMR1, 0x00);
 	}
 
+	if (priv->fops->phy_init_antenna_selection)
+		priv->fops->phy_init_antenna_selection(priv);
+
 	ret = rtl8xxxu_init_mac(priv, rtl8723a_mac_init_table);
 	dev_dbg(dev, "%s: init_mac %i\n", __func__, ret);
 	if (ret)
@@ -6380,6 +6415,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.load_firmware = rtl8723bu_load_firmware,
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
+	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.writeN_block_size = 1024,
 };
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 326fda5..43300bd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -764,5 +764,6 @@ struct rtl8xxxu_fileops {
 	int (*load_firmware) (struct rtl8xxxu_priv *priv);
 	int (*power_on) (struct rtl8xxxu_priv *priv);
 	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
+	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
 };
-- 
cgit v0.10.2


From b7dd8ff916574f77d40914321c08b7d78f101530 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:17 -0500
Subject: rtl8xxxu: Add rtl8723b_mac_init_table

Newer chips seem to have some different mac registers, requiring
a different init table.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index c55193b..813aadb 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -153,6 +153,37 @@ static struct rtl8xxxu_reg8val rtl8723a_mac_init_table[] = {
 	{0x70a, 0x65}, {0x70b, 0x87}, {0xffff, 0xff},
 };
 
+static struct rtl8xxxu_reg8val rtl8723b_mac_init_table[] = {
+	{0x02f, 0x30}, {0x035, 0x00}, {0x039, 0x08}, {0x04e, 0xe0},
+	{0x064, 0x00}, {0x067, 0x20}, {0x428, 0x0a}, {0x429, 0x10},
+	{0x430, 0x00}, {0x431, 0x00},
+	{0x432, 0x00}, {0x433, 0x01}, {0x434, 0x04}, {0x435, 0x05},
+	{0x436, 0x07}, {0x437, 0x08}, {0x43c, 0x04}, {0x43d, 0x05},
+	{0x43e, 0x07}, {0x43f, 0x08}, {0x440, 0x5d}, {0x441, 0x01},
+	{0x442, 0x00}, {0x444, 0x10}, {0x445, 0x00}, {0x446, 0x00},
+	{0x447, 0x00}, {0x448, 0x00}, {0x449, 0xf0}, {0x44a, 0x0f},
+	{0x44b, 0x3e}, {0x44c, 0x10}, {0x44d, 0x00}, {0x44e, 0x00},
+	{0x44f, 0x00}, {0x450, 0x00}, {0x451, 0xf0}, {0x452, 0x0f},
+	{0x453, 0x00}, {0x456, 0x5e}, {0x460, 0x66}, {0x461, 0x66},
+	{0x4c8, 0xff}, {0x4c9, 0x08}, {0x4cc, 0xff},
+	{0x4cd, 0xff}, {0x4ce, 0x01}, {0x500, 0x26}, {0x501, 0xa2},
+	{0x502, 0x2f}, {0x503, 0x00}, {0x504, 0x28}, {0x505, 0xa3},
+	{0x506, 0x5e}, {0x507, 0x00}, {0x508, 0x2b}, {0x509, 0xa4},
+	{0x50a, 0x5e}, {0x50b, 0x00}, {0x50c, 0x4f}, {0x50d, 0xa4},
+	{0x50e, 0x00}, {0x50f, 0x00}, {0x512, 0x1c}, {0x514, 0x0a},
+	{0x516, 0x0a}, {0x525, 0x4f},
+	{0x550, 0x10}, {0x551, 0x10}, {0x559, 0x02}, {0x55c, 0x50},
+	{0x55d, 0xff}, {0x605, 0x30}, {0x608, 0x0e}, {0x609, 0x2a},
+	{0x620, 0xff}, {0x621, 0xff}, {0x622, 0xff}, {0x623, 0xff},
+	{0x624, 0xff}, {0x625, 0xff}, {0x626, 0xff}, {0x627, 0xff},
+	{0x638, 0x50}, {0x63c, 0x0a}, {0x63d, 0x0a}, {0x63e, 0x0e},
+	{0x63f, 0x0e}, {0x640, 0x40}, {0x642, 0x40}, {0x643, 0x00},
+	{0x652, 0xc8}, {0x66e, 0x05}, {0x700, 0x21}, {0x701, 0x43},
+	{0x702, 0x65}, {0x703, 0x87}, {0x708, 0x21}, {0x709, 0x43},
+	{0x70a, 0x65}, {0x70b, 0x87}, {0x765, 0x18}, {0x76e, 0x04},
+	{0xffff, 0xff},
+};
+
 static struct rtl8xxxu_reg32val rtl8723a_phy_1t_init_table[] = {
 	{0x800, 0x80040000}, {0x804, 0x00000003},
 	{0x808, 0x0000fc00}, {0x80c, 0x0000000a},
@@ -4656,7 +4687,11 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (priv->fops->phy_init_antenna_selection)
 		priv->fops->phy_init_antenna_selection(priv);
 
-	ret = rtl8xxxu_init_mac(priv, rtl8723a_mac_init_table);
+	if (priv->rtlchip == 0x8723b)
+		ret = rtl8xxxu_init_mac(priv, rtl8723b_mac_init_table);
+	else
+		ret = rtl8xxxu_init_mac(priv, rtl8723a_mac_init_table);
+
 	dev_dbg(dev, "%s: init_mac %i\n", __func__, ret);
 	if (ret)
 		goto exit;
-- 
cgit v0.10.2


From b9f498e11fe9ae508c716d5cb4e468cf176018f2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:18 -0500
Subject: rtl8xxxu: Add 8723by AGC table

The different RF module seems to require a different AGC table as well

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 813aadb..89c66a2 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -748,6 +748,77 @@ static struct rtl8xxxu_reg32val rtl8xxx_agc_highpa_table[] = {
 	{0xffff, 0xffffffff}
 };
 
+static struct rtl8xxxu_reg32val rtl8xxx_agc_8723bu_table[] = {
+	{0xc78, 0xfd000001}, {0xc78, 0xfc010001},
+	{0xc78, 0xfb020001}, {0xc78, 0xfa030001},
+	{0xc78, 0xf9040001}, {0xc78, 0xf8050001},
+	{0xc78, 0xf7060001}, {0xc78, 0xf6070001},
+	{0xc78, 0xf5080001}, {0xc78, 0xf4090001},
+	{0xc78, 0xf30a0001}, {0xc78, 0xf20b0001},
+	{0xc78, 0xf10c0001}, {0xc78, 0xf00d0001},
+	{0xc78, 0xef0e0001}, {0xc78, 0xee0f0001},
+	{0xc78, 0xed100001}, {0xc78, 0xec110001},
+	{0xc78, 0xeb120001}, {0xc78, 0xea130001},
+	{0xc78, 0xe9140001}, {0xc78, 0xe8150001},
+	{0xc78, 0xe7160001}, {0xc78, 0xe6170001},
+	{0xc78, 0xe5180001}, {0xc78, 0xe4190001},
+	{0xc78, 0xe31a0001}, {0xc78, 0xa51b0001},
+	{0xc78, 0xa41c0001}, {0xc78, 0xa31d0001},
+	{0xc78, 0x671e0001}, {0xc78, 0x661f0001},
+	{0xc78, 0x65200001}, {0xc78, 0x64210001},
+	{0xc78, 0x63220001}, {0xc78, 0x4a230001},
+	{0xc78, 0x49240001}, {0xc78, 0x48250001},
+	{0xc78, 0x47260001}, {0xc78, 0x46270001},
+	{0xc78, 0x45280001}, {0xc78, 0x44290001},
+	{0xc78, 0x432a0001}, {0xc78, 0x422b0001},
+	{0xc78, 0x292c0001}, {0xc78, 0x282d0001},
+	{0xc78, 0x272e0001}, {0xc78, 0x262f0001},
+	{0xc78, 0x0a300001}, {0xc78, 0x09310001},
+	{0xc78, 0x08320001}, {0xc78, 0x07330001},
+	{0xc78, 0x06340001}, {0xc78, 0x05350001},
+	{0xc78, 0x04360001}, {0xc78, 0x03370001},
+	{0xc78, 0x02380001}, {0xc78, 0x01390001},
+	{0xc78, 0x013a0001}, {0xc78, 0x013b0001},
+	{0xc78, 0x013c0001}, {0xc78, 0x013d0001},
+	{0xc78, 0x013e0001}, {0xc78, 0x013f0001},
+	{0xc78, 0xfc400001}, {0xc78, 0xfb410001},
+	{0xc78, 0xfa420001}, {0xc78, 0xf9430001},
+	{0xc78, 0xf8440001}, {0xc78, 0xf7450001},
+	{0xc78, 0xf6460001}, {0xc78, 0xf5470001},
+	{0xc78, 0xf4480001}, {0xc78, 0xf3490001},
+	{0xc78, 0xf24a0001}, {0xc78, 0xf14b0001},
+	{0xc78, 0xf04c0001}, {0xc78, 0xef4d0001},
+	{0xc78, 0xee4e0001}, {0xc78, 0xed4f0001},
+	{0xc78, 0xec500001}, {0xc78, 0xeb510001},
+	{0xc78, 0xea520001}, {0xc78, 0xe9530001},
+	{0xc78, 0xe8540001}, {0xc78, 0xe7550001},
+	{0xc78, 0xe6560001}, {0xc78, 0xe5570001},
+	{0xc78, 0xe4580001}, {0xc78, 0xe3590001},
+	{0xc78, 0xa65a0001}, {0xc78, 0xa55b0001},
+	{0xc78, 0xa45c0001}, {0xc78, 0xa35d0001},
+	{0xc78, 0x675e0001}, {0xc78, 0x665f0001},
+	{0xc78, 0x65600001}, {0xc78, 0x64610001},
+	{0xc78, 0x63620001}, {0xc78, 0x62630001},
+	{0xc78, 0x61640001}, {0xc78, 0x48650001},
+	{0xc78, 0x47660001}, {0xc78, 0x46670001},
+	{0xc78, 0x45680001}, {0xc78, 0x44690001},
+	{0xc78, 0x436a0001}, {0xc78, 0x426b0001},
+	{0xc78, 0x286c0001}, {0xc78, 0x276d0001},
+	{0xc78, 0x266e0001}, {0xc78, 0x256f0001},
+	{0xc78, 0x24700001}, {0xc78, 0x09710001},
+	{0xc78, 0x08720001}, {0xc78, 0x07730001},
+	{0xc78, 0x06740001}, {0xc78, 0x05750001},
+	{0xc78, 0x04760001}, {0xc78, 0x03770001},
+	{0xc78, 0x02780001}, {0xc78, 0x01790001},
+	{0xc78, 0x017a0001}, {0xc78, 0x017b0001},
+	{0xc78, 0x017c0001}, {0xc78, 0x017d0001},
+	{0xc78, 0x017e0001}, {0xc78, 0x017f0001},
+	{0xc50, 0x69553422},
+	{0xc50, 0x69553420},
+	{0x824, 0x00390204},
+	{0xffff, 0xffffffff}
+};
+
 static struct rtl8xxxu_rfregval rtl8723au_radioa_1t_init_table[] = {
 	{0x00, 0x00030159}, {0x01, 0x00031284},
 	{0x02, 0x00098000}, {0x03, 0x00039c63},
@@ -2895,7 +2966,9 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 		rtl8xxxu_write32(priv, REG_TX_TO_TX, val32);
 	}
 
-	if (priv->hi_pa)
+	if (priv->rtlchip == 0x8723b)
+		rtl8xxxu_init_phy_regs(priv, rtl8xxx_agc_8723bu_table);
+	else if (priv->hi_pa)
 		rtl8xxxu_init_phy_regs(priv, rtl8xxx_agc_highpa_table);
 	else
 		rtl8xxxu_init_phy_regs(priv, rtl8xxx_agc_standard_table);
-- 
cgit v0.10.2


From ed35d09469e55b3324ef85b33e0355bd86cdd066 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:19 -0500
Subject: rtl8xxxu: Handle 32 bit mailbox extension regs found on
 8723bu/8192eu/8812

Gen1 chips use a 16 bit mailbox extension register, for upto 48 bit
mailbox commands. The newer generation chips use a 32 bit mailbox
extension register instead, for upto 64 bit mailbox commands.

Handle writing the larger mailboxes.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 89c66a2..eb70fa2 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1482,7 +1482,8 @@ static int rtl8723a_h2c_cmd(struct rtl8xxxu_priv *priv, struct h2c_cmd *h2c)
 
 	mbox_nr = priv->next_mbox;
 	mbox_reg = REG_HMBOX_0 + (mbox_nr * 4);
-	mbox_ext_reg = REG_HMBOX_EXT_0 + (mbox_nr * 2);
+	mbox_ext_reg = priv->fops->mbox_ext_reg +
+		(mbox_nr * priv->fops->mbox_ext_width);
 
 	/*
 	 * MBOX ready?
@@ -1504,11 +1505,19 @@ static int rtl8723a_h2c_cmd(struct rtl8xxxu_priv *priv, struct h2c_cmd *h2c)
 	 * Need to swap as it's being swapped again by rtl8xxxu_write16/32()
 	 */
 	if (h2c->cmd.cmd & H2C_EXT) {
-		rtl8xxxu_write16(priv, mbox_ext_reg,
-				 le16_to_cpu(h2c->raw.ext));
-		if (rtl8xxxu_debug & RTL8XXXU_DEBUG_H2C)
-			dev_info(dev, "H2C_EXT %04x\n",
-				 le16_to_cpu(h2c->raw.ext));
+		if (priv->fops->mbox_ext_width == 4) {
+			rtl8xxxu_write32(priv, mbox_ext_reg,
+					 le32_to_cpu(h2c->raw_wide.ext));
+			if (rtl8xxxu_debug & RTL8XXXU_DEBUG_H2C)
+				dev_info(dev, "H2C_EXT %08x\n",
+					 le32_to_cpu(h2c->raw_wide.ext));
+		} else {
+			rtl8xxxu_write16(priv, mbox_ext_reg,
+					 le16_to_cpu(h2c->raw.ext));
+			if (rtl8xxxu_debug & RTL8XXXU_DEBUG_H2C)
+				dev_info(dev, "H2C_EXT %04x\n",
+					 le16_to_cpu(h2c->raw.ext));
+		}
 	}
 	rtl8xxxu_write32(priv, mbox_reg, le32_to_cpu(h2c->raw.data));
 	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_H2C)
@@ -6516,6 +6525,8 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.writeN_block_size = 1024,
+	.mbox_ext_reg = REG_HMBOX_EXT_0,
+	.mbox_ext_width = 2,
 };
 
 static struct rtl8xxxu_fileops rtl8723bu_fops = {
@@ -6525,6 +6536,8 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.writeN_block_size = 1024,
+	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
+	.mbox_ext_width = 4,
 };
 
 #ifdef CONFIG_RTL8XXXU_UNTESTED
@@ -6535,6 +6548,8 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.power_on = rtl8192cu_power_on,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.writeN_block_size = 128,
+	.mbox_ext_reg = REG_HMBOX_EXT_0,
+	.mbox_ext_width = 2,
 };
 
 #endif
@@ -6545,6 +6560,8 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.power_on = rtl8192eu_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.writeN_block_size = 128,
+	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
+	.mbox_ext_width = 4,
 };
 
 static struct usb_device_id dev_table[] = {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 43300bd..7aed304 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -628,13 +628,17 @@ struct h2c_cmd {
 	union {
 		struct {
 			u8 cmd;
-			u8 data[5];
+			u8 data[7];
 		} __packed cmd;
 		struct {
 			__le32 data;
 			__le16 ext;
 		} __packed raw;
 		struct {
+			__le32 data;
+			__le32 ext;
+		} __packed raw_wide;
+		struct {
 			u8 cmd;
 			u8 data;
 			u8 pad[4];
@@ -766,4 +770,6 @@ struct rtl8xxxu_fileops {
 	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
+	u16 mbox_ext_reg;
+	char mbox_ext_width;
 };
-- 
cgit v0.10.2


From 14d8856082c5fdb320b108af690c79d4d560d9b2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:20 -0500
Subject: rtl8xxxu: Add some missing register definitions for 8723bu

This introduces additional register definitions for newer generation
chips.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 3cbb549..f1375b8 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -155,6 +155,8 @@
 /*  RTL8723 WIFI/BT/GPS Multi-Function GPIO Select. */
 #define REG_GPIO_IO_SEL_2		0x0062
 
+/*  RTL8723B */
+#define REG_PAD_CTRL1			0x0064
 /*  RTL8723 only WIFI/BT/GPS Multi-Function control source. */
 #define REG_MULTI_FUNC_CTRL		0x0068
 
@@ -196,9 +198,11 @@
 #define REG_HMBOX_EXT_1			0x008a
 #define REG_HMBOX_EXT_2			0x008c
 #define REG_HMBOX_EXT_3			0x008e
-/* Interrupt registers for 8192e/8812 */
+/* Interrupt registers for 8192e/8723bu/8812 */
 #define REG_HIMR0			0x00b0
+#define REG_HISR0			0x00b4
 #define REG_HIMR1			0x00b8
+#define REG_HISR1			0x00bc
 
 /*  Host suspend counter on FPGA platform */
 #define REG_HOST_SUSP_CNT		0x00bc
@@ -337,6 +341,8 @@
 #define REG_MBIST_DONE			0x0178
 #define REG_MBIST_FAIL			0x017c
 #define REG_C2HEVT_MSG_NORMAL		0x01a0
+/* 8192EU/8723BU/8812 */
+#define REG_C2HEVT_CMD_ID_8723B		0x01ae
 #define REG_C2HEVT_CLEAR		0x01af
 #define REG_C2HEVT_MSG_TEST		0x01b8
 #define REG_MCUTST_1			0x01c0
-- 
cgit v0.10.2


From 6431ea00dfc5a7a3cdca1320034f20b69846a432 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:21 -0500
Subject: rtl8xxxu: Group USB fixups together for all chips

In addition do not apply fixups for 8188/8191/8192 A-cut UMC parts.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index eb70fa2..be95ef3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4757,11 +4757,42 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		}
 	}
 
-	if (priv->rtlchip == 0x8192e) {
+	/* Fix USB interface interference issue */
+	if (priv->rtlchip == 0x8723a) {
+		rtl8xxxu_write8(priv, 0xfe40, 0xe0);
+		rtl8xxxu_write8(priv, 0xfe41, 0x8d);
+		rtl8xxxu_write8(priv, 0xfe42, 0x80);
+		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, 0xfd0320);
+	} else {
 		val32 = rtl8xxxu_read32(priv, REG_TXDMA_OFFSET_CHK);
 		val32 |= TXDMA_OFFSET_DROP_DATA_EN;
 		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32);
+	}
 
+	/* Solve too many protocol error on USB bus */
+	/* Can't do this for 8188/8192 UMC A cut parts */
+	if (priv->rtlchip == 0x8723a ||
+	    ((priv->rtlchip == 0x8192c || priv->rtlchip == 0x8191c ||
+	      priv->rtlchip == 0x8188c) &&
+	     (priv->chip_cut || !priv->vendor_umc))) {
+		rtl8xxxu_write8(priv, 0xfe40, 0xe6);
+		rtl8xxxu_write8(priv, 0xfe41, 0x94);
+		rtl8xxxu_write8(priv, 0xfe42, 0x80);
+
+		rtl8xxxu_write8(priv, 0xfe40, 0xe0);
+		rtl8xxxu_write8(priv, 0xfe41, 0x19);
+		rtl8xxxu_write8(priv, 0xfe42, 0x80);
+
+		rtl8xxxu_write8(priv, 0xfe40, 0xe5);
+		rtl8xxxu_write8(priv, 0xfe41, 0x91);
+		rtl8xxxu_write8(priv, 0xfe42, 0x80);
+
+		rtl8xxxu_write8(priv, 0xfe40, 0xe2);
+		rtl8xxxu_write8(priv, 0xfe41, 0x81);
+		rtl8xxxu_write8(priv, 0xfe42, 0x80);
+	}
+
+	if (priv->rtlchip == 0x8192e || priv->rtlchip == 0x8723b) {
 		rtl8xxxu_write32(priv, REG_HIMR0, 0x00);
 		rtl8xxxu_write32(priv, REG_HIMR1, 0x00);
 	}
@@ -4968,30 +4999,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	rtl8723a_phy_lc_calibrate(priv);
 
-	/* fix USB interface interference issue */
-	rtl8xxxu_write8(priv, 0xfe40, 0xe0);
-	rtl8xxxu_write8(priv, 0xfe41, 0x8d);
-	rtl8xxxu_write8(priv, 0xfe42, 0x80);
-	rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, 0xfd0320);
-
-	/* Solve too many protocol error on USB bus */
-	/* Can't do this for 8188/8192 UMC A cut parts */
-	rtl8xxxu_write8(priv, 0xfe40, 0xe6);
-	rtl8xxxu_write8(priv, 0xfe41, 0x94);
-	rtl8xxxu_write8(priv, 0xfe42, 0x80);
-
-	rtl8xxxu_write8(priv, 0xfe40, 0xe0);
-	rtl8xxxu_write8(priv, 0xfe41, 0x19);
-	rtl8xxxu_write8(priv, 0xfe42, 0x80);
-
-	rtl8xxxu_write8(priv, 0xfe40, 0xe5);
-	rtl8xxxu_write8(priv, 0xfe41, 0x91);
-	rtl8xxxu_write8(priv, 0xfe42, 0x80);
-
-	rtl8xxxu_write8(priv, 0xfe40, 0xe2);
-	rtl8xxxu_write8(priv, 0xfe41, 0x81);
-	rtl8xxxu_write8(priv, 0xfe42, 0x80);
-
 	/* Init BT hw config. */
 	rtl8xxxu_init_bt(priv);
 
-- 
cgit v0.10.2


From d940c247adbd44c885777099e3e6c64b7ca2d078 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:22 -0500
Subject: rtl8xxxu: Add definitions for new generation h2c commands

The larger mailboxes also use a different set of mailbox commands.
This provides a list of the 64 bit commands.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 7aed304..198b1d8 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -617,12 +617,91 @@ struct rtl8xxxu_rfregs {
 
 #define H2C_MAX_MBOX			4
 #define H2C_EXT				BIT(7)
-#define H2C_SET_POWER_MODE		1
-#define H2C_JOIN_BSS_REPORT		2
 #define  H2C_JOIN_BSS_DISCONNECT	0
 #define  H2C_JOIN_BSS_CONNECT		1
-#define H2C_SET_RSSI			5
-#define H2C_SET_RATE_MASK		(6 | H2C_EXT)
+
+/*
+ * H2C (firmware) commands differ between the older generation chips
+ * 8188[cr]u, 819[12]cu, and 8723au, and the more recent chips 8723bu,
+ * 8192[de]u, 8192eu, and 8812.
+ */
+enum h2c_cmd_8723a {
+	H2C_SET_POWER_MODE = 1,
+	H2C_JOIN_BSS_REPORT = 2,
+	H2C_SET_RSSI = 5,
+	H2C_SET_RATE_MASK = (6 | H2C_EXT),
+};
+
+enum h2c_cmd_8723b {
+	/*
+	 * Common Class: 000
+	 */
+	H2C_8723B_RSVD_PAGE = 0x00,
+	H2C_8723B_MEDIA_STATUS_RPT = 0x01,
+	H2C_8723B_SCAN_ENABLE = 0x02,
+	H2C_8723B_KEEP_ALIVE = 0x03,
+	H2C_8723B_DISCON_DECISION = 0x04,
+	H2C_8723B_PSD_OFFLOAD = 0x05,
+	H2C_8723B_AP_OFFLOAD = 0x08,
+	H2C_8723B_BCN_RSVDPAGE = 0x09,
+	H2C_8723B_PROBERSP_RSVDPAGE = 0x0A,
+	H2C_8723B_FCS_RSVDPAGE = 0x10,
+	H2C_8723B_FCS_INFO = 0x11,
+	H2C_8723B_AP_WOW_GPIO_CTRL = 0x13,
+
+	/*
+	 * PoweSave Class: 001
+	 */
+	H2C_8723B_SET_PWR_MODE = 0x20,
+	H2C_8723B_PS_TUNING_PARA = 0x21,
+	H2C_8723B_PS_TUNING_PARA2 = 0x22,
+	H2C_8723B_P2P_LPS_PARAM = 0x23,
+	H2C_8723B_P2P_PS_OFFLOAD = 0x24,
+	H2C_8723B_PS_SCAN_ENABLE = 0x25,
+	H2C_8723B_SAP_PS_ = 0x26,
+	H2C_8723B_INACTIVE_PS_ = 0x27,
+	H2C_8723B_FWLPS_IN_IPS_ = 0x28,
+
+	/*
+	 * Dynamic Mechanism Class: 010
+	 */
+	H2C_8723B_MACID_CFG = 0x40,
+	H2C_8723B_TXBF = 0x41,
+	H2C_8723B_RSSI_SETTING = 0x42,
+	H2C_8723B_AP_REQ_TXRPT = 0x43,
+	H2C_8723B_INIT_RATE_COLLECT = 0x44,
+
+	/*
+	 * BT Class: 011
+	 */
+	H2C_8723B_B_TYPE_TDMA = 0x60,
+	H2C_8723B_BT_INFO = 0x61,
+	H2C_8723B_FORCE_BT_TXPWR = 0x62,
+	H2C_8723B_BT_IGNORE_WLANACT = 0x63,
+	H2C_8723B_DAC_SWING_VALUE = 0x64,
+	H2C_8723B_ANT_SEL_RSV = 0x65,
+	H2C_8723B_WL_OPMODE = 0x66,
+	H2C_8723B_BT_MP_OPER = 0x67,
+	H2C_8723B_BT_CONTROL = 0x68,
+	H2C_8723B_BT_WIFI_CTRL = 0x69,
+	H2C_8723B_BT_FW_PATCH = 0x6A,
+	H2C_8723B_BT_WLAN_CALIBRATION = 0x6D,
+
+	/*
+	 * WOWLAN Class: 100
+	 */
+	H2C_8723B_WOWLAN = 0x80,
+	H2C_8723B_REMOTE_WAKE_CTRL = 0x81,
+	H2C_8723B_AOAC_GLOBAL_INFO = 0x82,
+	H2C_8723B_AOAC_RSVD_PAGE = 0x83,
+	H2C_8723B_AOAC_RSVD_PAGE2 = 0x84,
+	H2C_8723B_D0_SCAN_OFFLOAD_CTRL = 0x85,
+	H2C_8723B_D0_SCAN_OFFLOAD_INFO = 0x86,
+	H2C_8723B_CHNL_SWITCH_OFFLOAD = 0x87,
+
+	H2C_8723B_RESET_TSF = 0xC0,
+};
+
 
 struct h2c_cmd {
 	union {
-- 
cgit v0.10.2


From f6c47702ede3edc1dbf73355c134d7e33d9c86ca Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jsitnicki@gmail.com>
Date: Mon, 29 Feb 2016 17:04:23 -0500
Subject: rtl8xxxu: Don't check for illegal offset when reading from efuse

It is enough to check for either illegal offset or illegal map address
because map address is a value derived from an offset:

  map_addr = offset * 8
  EFUSE_MAP_LEN = EFUSE_MAX_SECTION_8723A * 8

Leave just the check for an illegal map address because its upper
bound (EFUSE_MAP_LEN) is used also in a couple other places.

Signed-off-by: Jakub Sitnicki <jsitnicki@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index be95ef3..c279d6d 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2417,6 +2417,8 @@ static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 
 	efuse_addr = 0;
 	while (efuse_addr < EFUSE_REAL_CONTENT_LEN_8723A) {
+		u16 map_addr;
+
 		ret = rtl8xxxu_read_efuse8(priv, efuse_addr++, &header);
 		if (ret || header == 0xff)
 			goto exit;
@@ -2439,45 +2441,36 @@ static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 			word_mask = header & 0x0f;
 		}
 
-		if (offset < EFUSE_MAX_SECTION_8723A) {
-			u16 map_addr;
-			/* Get word enable value from PG header */
+		/* Get word enable value from PG header */
 
-			/* We have 8 bits to indicate validity */
-			map_addr = offset * 8;
-			if (map_addr >= EFUSE_MAP_LEN) {
-				dev_warn(dev, "%s: Illegal map_addr (%04x), "
-					 "efuse corrupt!\n",
-					 __func__, map_addr);
-				ret = -EINVAL;
-				goto exit;
-			}
-			for (i = 0; i < EFUSE_MAX_WORD_UNIT; i++) {
-				/* Check word enable condition in the section */
-				if (!(word_mask & BIT(i))) {
-					ret = rtl8xxxu_read_efuse8(priv,
-								   efuse_addr++,
-								   &val8);
-					if (ret)
-						goto exit;
-					priv->efuse_wifi.raw[map_addr++] = val8;
-
-					ret = rtl8xxxu_read_efuse8(priv,
-								   efuse_addr++,
-								   &val8);
-					if (ret)
-						goto exit;
-					priv->efuse_wifi.raw[map_addr++] = val8;
-				} else
-					map_addr += 2;
-			}
-		} else {
-			dev_warn(dev,
-				 "%s: Illegal offset (%04x), efuse corrupt!\n",
-				 __func__, offset);
+		/* We have 8 bits to indicate validity */
+		map_addr = offset * 8;
+		if (map_addr >= EFUSE_MAP_LEN) {
+			dev_warn(dev, "%s: Illegal map_addr (%04x), "
+				 "efuse corrupt!\n",
+				 __func__, map_addr);
 			ret = -EINVAL;
 			goto exit;
 		}
+		for (i = 0; i < EFUSE_MAX_WORD_UNIT; i++) {
+			/* Check word enable condition in the section */
+			if (!(word_mask & BIT(i))) {
+				ret = rtl8xxxu_read_efuse8(priv,
+							   efuse_addr++,
+							   &val8);
+				if (ret)
+					goto exit;
+				priv->efuse_wifi.raw[map_addr++] = val8;
+
+				ret = rtl8xxxu_read_efuse8(priv,
+							   efuse_addr++,
+							   &val8);
+				if (ret)
+					goto exit;
+				priv->efuse_wifi.raw[map_addr++] = val8;
+			} else
+				map_addr += 2;
+		}
 	}
 
 exit:
-- 
cgit v0.10.2


From 32a39dd4b574abd32304857e8a5b24de113b81ee Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jsitnicki@gmail.com>
Date: Mon, 29 Feb 2016 17:04:24 -0500
Subject: rtl8xxxu: Skip disabled efuse words early

Avoid a negative conditional and an extra level of indentation in the
bigger part of the loop body.

Signed-off-by: Jakub Sitnicki <jsitnicki@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index c279d6d..7a1c094 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2454,22 +2454,20 @@ static int rtl8xxxu_read_efuse(struct rtl8xxxu_priv *priv)
 		}
 		for (i = 0; i < EFUSE_MAX_WORD_UNIT; i++) {
 			/* Check word enable condition in the section */
-			if (!(word_mask & BIT(i))) {
-				ret = rtl8xxxu_read_efuse8(priv,
-							   efuse_addr++,
-							   &val8);
-				if (ret)
-					goto exit;
-				priv->efuse_wifi.raw[map_addr++] = val8;
-
-				ret = rtl8xxxu_read_efuse8(priv,
-							   efuse_addr++,
-							   &val8);
-				if (ret)
-					goto exit;
-				priv->efuse_wifi.raw[map_addr++] = val8;
-			} else
+			if (word_mask & BIT(i)) {
 				map_addr += 2;
+				continue;
+			}
+
+			ret = rtl8xxxu_read_efuse8(priv, efuse_addr++, &val8);
+			if (ret)
+				goto exit;
+			priv->efuse_wifi.raw[map_addr++] = val8;
+
+			ret = rtl8xxxu_read_efuse8(priv, efuse_addr++, &val8);
+			if (ret)
+				goto exit;
+			priv->efuse_wifi.raw[map_addr++] = val8;
 		}
 	}
 
-- 
cgit v0.10.2


From d38f1c3715d6de00d3b7a09254f9a0930d611bed Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jsitnicki@gmail.com>
Date: Mon, 29 Feb 2016 17:04:25 -0500
Subject: rtl8xxxu: rtl8723au: Introduce a pointer to efuse

Signed-off-by: Jakub Sitnicki <jsitnicki@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 7a1c094..172d76c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2127,43 +2127,45 @@ static int rtl8xxxu_identify_chip(struct rtl8xxxu_priv *priv)
 
 static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 {
-	if (priv->efuse_wifi.efuse8723.rtl_id != cpu_to_le16(0x8129))
+	struct rtl8723au_efuse *efuse = &priv->efuse_wifi.efuse8723;
+
+	if (efuse->rtl_id != cpu_to_le16(0x8129))
 		return -EINVAL;
 
-	ether_addr_copy(priv->mac_addr, priv->efuse_wifi.efuse8723.mac_addr);
+	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
 	memcpy(priv->cck_tx_power_index_A,
-	       priv->efuse_wifi.efuse8723.cck_tx_power_index_A,
+	       efuse->cck_tx_power_index_A,
 	       sizeof(priv->cck_tx_power_index_A));
 	memcpy(priv->cck_tx_power_index_B,
-	       priv->efuse_wifi.efuse8723.cck_tx_power_index_B,
+	       efuse->cck_tx_power_index_B,
 	       sizeof(priv->cck_tx_power_index_B));
 
 	memcpy(priv->ht40_1s_tx_power_index_A,
-	       priv->efuse_wifi.efuse8723.ht40_1s_tx_power_index_A,
+	       efuse->ht40_1s_tx_power_index_A,
 	       sizeof(priv->ht40_1s_tx_power_index_A));
 	memcpy(priv->ht40_1s_tx_power_index_B,
-	       priv->efuse_wifi.efuse8723.ht40_1s_tx_power_index_B,
+	       efuse->ht40_1s_tx_power_index_B,
 	       sizeof(priv->ht40_1s_tx_power_index_B));
 
 	memcpy(priv->ht20_tx_power_index_diff,
-	       priv->efuse_wifi.efuse8723.ht20_tx_power_index_diff,
+	       efuse->ht20_tx_power_index_diff,
 	       sizeof(priv->ht20_tx_power_index_diff));
 	memcpy(priv->ofdm_tx_power_index_diff,
-	       priv->efuse_wifi.efuse8723.ofdm_tx_power_index_diff,
+	       efuse->ofdm_tx_power_index_diff,
 	       sizeof(priv->ofdm_tx_power_index_diff));
 
 	memcpy(priv->ht40_max_power_offset,
-	       priv->efuse_wifi.efuse8723.ht40_max_power_offset,
+	       efuse->ht40_max_power_offset,
 	       sizeof(priv->ht40_max_power_offset));
 	memcpy(priv->ht20_max_power_offset,
-	       priv->efuse_wifi.efuse8723.ht20_max_power_offset,
+	       efuse->ht20_max_power_offset,
 	       sizeof(priv->ht20_max_power_offset));
 
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
-		 priv->efuse_wifi.efuse8723.vendor_name);
+		 efuse->vendor_name);
 	dev_info(&priv->udev->dev, "Product: %.41s\n",
-		 priv->efuse_wifi.efuse8723.device_name);
+		 efuse->device_name);
 	return 0;
 }
 
-- 
cgit v0.10.2


From 4959444165523e68118035bebb727b9431746765 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jsitnicki@gmail.com>
Date: Mon, 29 Feb 2016 17:04:26 -0500
Subject: rtl8xxxu: rtl8192cu: Introduce a pointer to efuse

Signed-off-by: Jakub Sitnicki <jsitnicki@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 172d76c..fcae9f0 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2218,50 +2218,51 @@ static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv)
 {
+	struct rtl8192cu_efuse *efuse = &priv->efuse_wifi.efuse8192;
 	int i;
 
-	if (priv->efuse_wifi.efuse8192.rtl_id != cpu_to_le16(0x8129))
+	if (efuse->rtl_id != cpu_to_le16(0x8129))
 		return -EINVAL;
 
-	ether_addr_copy(priv->mac_addr, priv->efuse_wifi.efuse8192.mac_addr);
+	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
 	memcpy(priv->cck_tx_power_index_A,
-	       priv->efuse_wifi.efuse8192.cck_tx_power_index_A,
+	       efuse->cck_tx_power_index_A,
 	       sizeof(priv->cck_tx_power_index_A));
 	memcpy(priv->cck_tx_power_index_B,
-	       priv->efuse_wifi.efuse8192.cck_tx_power_index_B,
+	       efuse->cck_tx_power_index_B,
 	       sizeof(priv->cck_tx_power_index_B));
 
 	memcpy(priv->ht40_1s_tx_power_index_A,
-	       priv->efuse_wifi.efuse8192.ht40_1s_tx_power_index_A,
+	       efuse->ht40_1s_tx_power_index_A,
 	       sizeof(priv->ht40_1s_tx_power_index_A));
 	memcpy(priv->ht40_1s_tx_power_index_B,
-	       priv->efuse_wifi.efuse8192.ht40_1s_tx_power_index_B,
+	       efuse->ht40_1s_tx_power_index_B,
 	       sizeof(priv->ht40_1s_tx_power_index_B));
 	memcpy(priv->ht40_2s_tx_power_index_diff,
-	       priv->efuse_wifi.efuse8192.ht40_2s_tx_power_index_diff,
+	       efuse->ht40_2s_tx_power_index_diff,
 	       sizeof(priv->ht40_2s_tx_power_index_diff));
 
 	memcpy(priv->ht20_tx_power_index_diff,
-	       priv->efuse_wifi.efuse8192.ht20_tx_power_index_diff,
+	       efuse->ht20_tx_power_index_diff,
 	       sizeof(priv->ht20_tx_power_index_diff));
 	memcpy(priv->ofdm_tx_power_index_diff,
-	       priv->efuse_wifi.efuse8192.ofdm_tx_power_index_diff,
+	       efuse->ofdm_tx_power_index_diff,
 	       sizeof(priv->ofdm_tx_power_index_diff));
 
 	memcpy(priv->ht40_max_power_offset,
-	       priv->efuse_wifi.efuse8192.ht40_max_power_offset,
+	       efuse->ht40_max_power_offset,
 	       sizeof(priv->ht40_max_power_offset));
 	memcpy(priv->ht20_max_power_offset,
-	       priv->efuse_wifi.efuse8192.ht20_max_power_offset,
+	       efuse->ht20_max_power_offset,
 	       sizeof(priv->ht20_max_power_offset));
 
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
-		 priv->efuse_wifi.efuse8192.vendor_name);
+		 efuse->vendor_name);
 	dev_info(&priv->udev->dev, "Product: %.20s\n",
-		 priv->efuse_wifi.efuse8192.device_name);
+		 efuse->device_name);
 
-	if (priv->efuse_wifi.efuse8192.rf_regulatory & 0x20) {
+	if (efuse->rf_regulatory & 0x20) {
 		sprintf(priv->chip_name, "8188RU");
 		priv->hi_pa = 1;
 	}
-- 
cgit v0.10.2


From b7dda34d1ecda7e3243137a6ffcb7e9173e5d651 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:27 -0500
Subject: rtl8xxxu: rtl8192eu_parse_efuse(): Use a pointer to the struct
 rtl8192eu_efuse

Make the code easier to read and less error prone by using a pointer
to the efuse.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index fcae9f0..a8f1eae 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2288,33 +2288,27 @@ static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 {
+	struct rtl8192eu_efuse *efuse = &priv->efuse_wifi.efuse8192eu;
 	int i;
 
-	if (priv->efuse_wifi.efuse8192eu.rtl_id != cpu_to_le16(0x8129))
+	if (efuse->rtl_id != cpu_to_le16(0x8129))
 		return -EINVAL;
 
-	ether_addr_copy(priv->mac_addr, priv->efuse_wifi.efuse8192eu.mac_addr);
+	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
-	memcpy(priv->cck_tx_power_index_A,
-	       priv->efuse_wifi.efuse8192eu.cck_tx_power_index_A,
+	memcpy(priv->cck_tx_power_index_A, efuse->cck_tx_power_index_A,
 	       sizeof(priv->cck_tx_power_index_A));
-	memcpy(priv->cck_tx_power_index_B,
-	       priv->efuse_wifi.efuse8192eu.cck_tx_power_index_B,
+	memcpy(priv->cck_tx_power_index_B, efuse->cck_tx_power_index_B,
 	       sizeof(priv->cck_tx_power_index_B));
 
-	memcpy(priv->ht40_1s_tx_power_index_A,
-	       priv->efuse_wifi.efuse8192eu.ht40_1s_tx_power_index_A,
+	memcpy(priv->ht40_1s_tx_power_index_A, efuse->ht40_1s_tx_power_index_A,
 	       sizeof(priv->ht40_1s_tx_power_index_A));
-	memcpy(priv->ht40_1s_tx_power_index_B,
-	       priv->efuse_wifi.efuse8192eu.ht40_1s_tx_power_index_B,
+	memcpy(priv->ht40_1s_tx_power_index_B, efuse->ht40_1s_tx_power_index_B,
 	       sizeof(priv->ht40_1s_tx_power_index_B));
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
-		 priv->efuse_wifi.efuse8192eu.vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.11s\n",
-		 priv->efuse_wifi.efuse8192eu.device_name);
-	dev_info(&priv->udev->dev, "Serial: %.11s\n",
-		 priv->efuse_wifi.efuse8192eu.serial);
+	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
+	dev_info(&priv->udev->dev, "Product: %.11s\n", efuse->device_name);
+	dev_info(&priv->udev->dev, "Serial: %.11s\n", efuse->serial);
 
 	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_EFUSE) {
 		unsigned char *raw = priv->efuse_wifi.raw;
-- 
cgit v0.10.2


From b8ba8602b81a019c67bc1f34b6b68e8965086532 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:28 -0500
Subject: rtl8xxxu: rtl8723bu_parse_efuse(): Use a pointer to the struct
 rtl8723bu_efuse

Likewise for 8723bu, use a pointer to the efuse.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index a8f1eae..43bbbf0 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2171,29 +2171,25 @@ static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 
 static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
 {
-	if (priv->efuse_wifi.efuse8723bu.rtl_id != cpu_to_le16(0x8129))
+	struct rtl8723bu_efuse *efuse = &priv->efuse_wifi.efuse8723bu;
+
+	if (efuse->rtl_id != cpu_to_le16(0x8129))
 		return -EINVAL;
 
-	ether_addr_copy(priv->mac_addr, priv->efuse_wifi.efuse8723bu.mac_addr);
+	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
-	memcpy(priv->cck_tx_power_index_A,
-	       priv->efuse_wifi.efuse8723bu.cck_tx_power_index_A,
+	memcpy(priv->cck_tx_power_index_A, efuse->cck_tx_power_index_A,
 	       sizeof(priv->cck_tx_power_index_A));
-	memcpy(priv->cck_tx_power_index_B,
-	       priv->efuse_wifi.efuse8723bu.cck_tx_power_index_B,
+	memcpy(priv->cck_tx_power_index_B, efuse->cck_tx_power_index_B,
 	       sizeof(priv->cck_tx_power_index_B));
 
-	memcpy(priv->ht40_1s_tx_power_index_A,
-	       priv->efuse_wifi.efuse8723bu.ht40_1s_tx_power_index_A,
+	memcpy(priv->ht40_1s_tx_power_index_A, efuse->ht40_1s_tx_power_index_A,
 	       sizeof(priv->ht40_1s_tx_power_index_A));
-	memcpy(priv->ht40_1s_tx_power_index_B,
-	       priv->efuse_wifi.efuse8723bu.ht40_1s_tx_power_index_B,
+	memcpy(priv->ht40_1s_tx_power_index_B, efuse->ht40_1s_tx_power_index_B,
 	       sizeof(priv->ht40_1s_tx_power_index_B));
 
-	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
-		 priv->efuse_wifi.efuse8723bu.vendor_name);
-	dev_info(&priv->udev->dev, "Product: %.41s\n",
-		 priv->efuse_wifi.efuse8723bu.device_name);
+	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
+	dev_info(&priv->udev->dev, "Product: %.41s\n", efuse->device_name);
 
 	if (rtl8xxxu_debug & RTL8XXXU_DEBUG_EFUSE) {
 		int i;
-- 
cgit v0.10.2


From 8da91571bb401a3b9ffa36d7c6ca9c2476777e4d Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:29 -0500
Subject: rtl8xxxu: rtl8xxxu_h2c_cmd(): Add size argument

The firmware command API differs slightly between new and old
devices. The new generation requires the size since there is no
extension bit encoded into the command number.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 43bbbf0..b42d71b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1471,7 +1471,8 @@ static int rtl8xxxu_write_rfreg(struct rtl8xxxu_priv *priv,
 	return retval;
 }
 
-static int rtl8723a_h2c_cmd(struct rtl8xxxu_priv *priv, struct h2c_cmd *h2c)
+static int rtl8723a_h2c_cmd(struct rtl8xxxu_priv *priv,
+			    struct h2c_cmd *h2c, int len)
 {
 	struct device *dev = &priv->udev->dev;
 	int mbox_nr, retry, retval = 0;
@@ -1504,7 +1505,7 @@ static int rtl8723a_h2c_cmd(struct rtl8xxxu_priv *priv, struct h2c_cmd *h2c)
 	/*
 	 * Need to swap as it's being swapped again by rtl8xxxu_write16/32()
 	 */
-	if (h2c->cmd.cmd & H2C_EXT) {
+	if (len > sizeof(u32)) {
 		if (priv->fops->mbox_ext_width == 4) {
 			rtl8xxxu_write32(priv, mbox_ext_reg,
 					 le32_to_cpu(h2c->raw_wide.ext));
@@ -5110,9 +5111,9 @@ static void rtl8xxxu_update_rate_mask(struct rtl8xxxu_priv *priv,
 	if (sgi)
 		h2c.ramask.arg |= 0x20;
 
-	dev_dbg(&priv->udev->dev, "%s: rate mask %08x, arg %02x\n", __func__,
-		ramask, h2c.ramask.arg);
-	rtl8723a_h2c_cmd(priv, &h2c);
+	dev_dbg(&priv->udev->dev, "%s: rate mask %08x, arg %02x, size %li\n",
+		__func__, ramask, h2c.ramask.arg, sizeof(h2c.ramask));
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ramask));
 }
 
 static void rtl8xxxu_set_basic_rates(struct rtl8xxxu_priv *priv, u32 rate_cfg)
@@ -5200,7 +5201,7 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			h2c.joinbss.data = H2C_JOIN_BSS_DISCONNECT;
 		}
 		h2c.joinbss.cmd = H2C_JOIN_BSS_REPORT;
-		rtl8723a_h2c_cmd(priv, &h2c);
+		rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.joinbss));
 	}
 
 	if (changed & BSS_CHANGED_ERP_PREAMBLE) {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 198b1d8..47898ff 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -720,7 +720,6 @@ struct h2c_cmd {
 		struct {
 			u8 cmd;
 			u8 data;
-			u8 pad[4];
 		} __packed joinbss;
 		struct {
 			u8 cmd;
-- 
cgit v0.10.2


From c7a5a190df232b833b7867ef1679dff561b73c84 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:30 -0500
Subject: rtl8xxxu: Do BT_WLAN_CALIBRATION before doing IQK calibration

Newer generation chips require the firmware be notified before we
start the IQK calibration.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index b42d71b..e0653b2 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1497,7 +1497,7 @@ static int rtl8723a_h2c_cmd(struct rtl8xxxu_priv *priv,
 	} while (retry--);
 
 	if (!retry) {
-		dev_dbg(dev, "%s: Mailbox busy\n", __func__);
+		dev_info(dev, "%s: Mailbox busy\n", __func__);
 		retval = -EBUSY;
 		goto error;
 	}
@@ -3877,6 +3877,20 @@ static void rtl8xxxu_phy_iqcalibrate(struct rtl8xxxu_priv *priv,
 	}
 }
 
+static void rtl8xxxu_prepare_calibrate(struct rtl8xxxu_priv *priv, u8 start)
+{
+	struct h2c_cmd h2c;
+
+	if (priv->fops->mbox_ext_width < 4)
+		return;
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.bt_wlan_calibration.cmd = H2C_8723B_BT_WLAN_CALIBRATION;
+	h2c.bt_wlan_calibration.data = start;
+
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_wlan_calibration));
+}
+
 static void rtl8723a_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
 {
 	struct device *dev = &priv->udev->dev;
@@ -3888,6 +3902,8 @@ static void rtl8723a_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
 	s32 reg_tmp = 0;
 	bool simu;
 
+	rtl8xxxu_prepare_calibrate(priv, 1);
+
 	memset(result, 0, sizeof(result));
 	candidate = -1;
 
@@ -3975,6 +3991,8 @@ static void rtl8723a_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
 
 	rtl8xxxu_save_regs(priv, rtl8723au_iqk_phy_iq_bb_reg,
 			   priv->bb_recovery_backup, RTL8XXXU_BB_REGS);
+
+	rtl8xxxu_prepare_calibrate(priv, 0);
 }
 
 static void rtl8723a_phy_lc_calibrate(struct rtl8xxxu_priv *priv)
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 47898ff..07533ac 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -727,6 +727,10 @@ struct h2c_cmd {
 			u8 arg;
 			__le16 mask_lo;
 		} __packed ramask;
+		struct {
+			u8 cmd;
+			u8 data;
+		} __packed bt_wlan_calibration;
 	};
 };
 
-- 
cgit v0.10.2


From eaa4d14c97c1990327fd4e9fc49beca0ce7baa14 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:31 -0500
Subject: rtl8xxxu: Do not overwrite rtl8xxxu_debug for untested chips

Fix a silly bug where the debug level was overwritten rather than
amended for untested chips.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e0653b2..cfea68b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6391,7 +6391,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
 	}
 
 	if (untested) {
-		rtl8xxxu_debug = RTL8XXXU_DEBUG_EFUSE;
+		rtl8xxxu_debug |= RTL8XXXU_DEBUG_EFUSE;
 		dev_info(&udev->dev,
 			 "This Realtek USB WiFi dongle (0x%04x:0x%04x) is untested!\n",
 			 id->idVendor, id->idProduct);
-- 
cgit v0.10.2


From 7ff8c1ae610159fcb6e7188224520cf0f8b7a336 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:32 -0500
Subject: rtl8xxxu: Use correct formatting type to print sizeof()

Usual gcc i386 issue reported by kbuildbot

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index cfea68b..171e038 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5129,7 +5129,7 @@ static void rtl8xxxu_update_rate_mask(struct rtl8xxxu_priv *priv,
 	if (sgi)
 		h2c.ramask.arg |= 0x20;
 
-	dev_dbg(&priv->udev->dev, "%s: rate mask %08x, arg %02x, size %li\n",
+	dev_dbg(&priv->udev->dev, "%s: rate mask %08x, arg %02x, size %zi\n",
 		__func__, ramask, h2c.ramask.arg, sizeof(h2c.ramask));
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ramask));
 }
-- 
cgit v0.10.2


From 8634af5e6a8dea0f7110f32c00ceffee71be8a80 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:33 -0500
Subject: rtl8xxxu: Make rtl8xxxu_add_path_on() use device specific init values

rtl8192cu/rtl8188cu/rtl8723au use the same values, but 8723bu and
8192eu have their own.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 171e038..a16dabf 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -3557,11 +3557,13 @@ static void rtl8xxxu_path_adda_on(struct rtl8xxxu_priv *priv, const u32 *regs,
 	u32 path_on;
 	int i;
 
-	path_on = path_a_on ? 0x04db25a4 : 0x0b1b25a4;
 	if (priv->tx_paths == 1) {
-		path_on = 0x0bdb25a0;
-		rtl8xxxu_write32(priv, regs[0], 0x0b1b25a0);
+		path_on = priv->fops->adda_1t_path_on;
+		rtl8xxxu_write32(priv, regs[0], priv->fops->adda_1t_init);
 	} else {
+		path_on = path_a_on ? priv->fops->adda_2t_path_on_a :
+			priv->fops->adda_2t_path_on_b;
+
 		rtl8xxxu_write32(priv, regs[0], path_on);
 	}
 
@@ -6537,6 +6539,10 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
+	.adda_1t_init = 0x0b1b25a0,
+	.adda_1t_path_on = 0x0bdb25a0,
+	.adda_2t_path_on_a = 0x04db25a4,
+	.adda_2t_path_on_b = 0x0b1b25a4,
 };
 
 static struct rtl8xxxu_fileops rtl8723bu_fops = {
@@ -6548,6 +6554,10 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
+	.adda_1t_init = 0x01c00014,
+	.adda_1t_path_on = 0x01c00014,
+	.adda_2t_path_on_a = 0x01c00014,
+	.adda_2t_path_on_b = 0x01c00014,
 };
 
 #ifdef CONFIG_RTL8XXXU_UNTESTED
@@ -6560,6 +6570,10 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
+	.adda_1t_init = 0x0b1b25a0,
+	.adda_1t_path_on = 0x0bdb25a0,
+	.adda_2t_path_on_a = 0x04db25a4,
+	.adda_2t_path_on_b = 0x0b1b25a4,
 };
 
 #endif
@@ -6572,6 +6586,10 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
+	.adda_1t_init = 0x0fc01616,
+	.adda_1t_path_on = 0x0fc01616,
+	.adda_2t_path_on_a = 0x0fc01616,
+	.adda_2t_path_on_b = 0x0fc01616,
 };
 
 static struct usb_device_id dev_table[] = {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 07533ac..4b8431d 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -854,4 +854,8 @@ struct rtl8xxxu_fileops {
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
+	u32 adda_1t_init;
+	u32 adda_1t_path_on;
+	u32 adda_2t_path_on_a;
+	u32 adda_2t_path_on_b;
 };
-- 
cgit v0.10.2


From c6594ffd46a4b26d2a2d59c54da399ec563b6b2f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:34 -0500
Subject: rtl8xxxu: Add a couple of new register definitions

This adds some additional register definitions for 8723bu, as well as
a bit define for USB RXDMA aggregation in REG_RXDMA_AGG_PG_TH.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index f1375b8..162a9e8 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -388,9 +388,14 @@
 
 /* 0x0280 ~ 0x02FF	RXDMA Configuration */
 #define REG_RXDMA_AGG_PG_TH		0x0280
+#define  RXDMA_USB_AGG_ENABLE		BIT(31)
 #define REG_RXPKT_NUM			0x0284
 #define REG_RXDMA_STATUS		0x0288
 
+/* Presumably only found on newer chips such as 8723bu */
+#define REG_RX_DMA_CTRL_8723B		0x0286
+#define REG_RX_DMA_MODE_CTRL_8723B	0x0290
+
 #define REG_RF_BB_CMD_ADDR		0x02c0
 #define REG_RF_BB_CMD_DATA		0x02c4
 
@@ -473,6 +478,7 @@
 #define REG_POWER_STATUS		0x04a4
 #define REG_POWER_STAGE1		0x04b4
 #define REG_POWER_STAGE2		0x04b8
+#define REG_AMPDU_BURST_MODE_8723B	0x04bc
 #define REG_PKT_VO_VI_LIFE_TIME		0x04c0
 #define REG_PKT_BE_BK_LIFE_TIME		0x04c2
 #define REG_STBC_SETTING		0x04c4
-- 
cgit v0.10.2


From e1547c535ede26c84b72d51e031dd57551037bec Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:35 -0500
Subject: rtl8xxxu: First stab at adding IQK calibration for 8723bu parts

The 8723bu also has it's own IQK calibration process. This is similar
in flow, but still different enough to warrent it's own
implementation, at least for now.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index a16dabf..a40ef21 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -3510,6 +3510,91 @@ static bool rtl8xxxu_simularity_compare(struct rtl8xxxu_priv *priv,
 	return false;
 }
 
+static bool rtl8723bu_simularity_compare(struct rtl8xxxu_priv *priv,
+					 int result[][8], int c1, int c2)
+{
+	u32 i, j, diff, simubitmap, bound = 0;
+	int candidate[2] = {-1, -1};	/* for path A and path B */
+	int tmp1, tmp2;
+	bool retval = true;
+
+	if (priv->tx_paths > 1)
+		bound = 8;
+	else
+		bound = 4;
+
+	simubitmap = 0;
+
+	for (i = 0; i < bound; i++) {
+		if (i & 1) {
+			if ((result[c1][i] & 0x00000200))
+				tmp1 = result[c1][i] | 0xfffffc00;
+			else
+				tmp1 = result[c1][i];
+
+			if ((result[c2][i]& 0x00000200))
+				tmp2 = result[c2][i] | 0xfffffc00;
+			else
+				tmp2 = result[c2][i];
+		} else {
+			tmp1 = result[c1][i];
+			tmp2 = result[c2][i];
+		}
+
+		diff = (tmp1 > tmp2) ? (tmp1 - tmp2) : (tmp2 - tmp1);
+
+		if (diff > MAX_TOLERANCE) {
+			if ((i == 2 || i == 6) && !simubitmap) {
+				if (result[c1][i] + result[c1][i + 1] == 0)
+					candidate[(i / 4)] = c2;
+				else if (result[c2][i] + result[c2][i + 1] == 0)
+					candidate[(i / 4)] = c1;
+				else
+					simubitmap = simubitmap | (1 << i);
+			} else {
+				simubitmap = simubitmap | (1 << i);
+			}
+		}
+	}
+
+	if (simubitmap == 0) {
+		for (i = 0; i < (bound / 4); i++) {
+			if (candidate[i] >= 0) {
+				for (j = i * 4; j < (i + 1) * 4 - 2; j++)
+					result[3][j] = result[candidate[i]][j];
+				retval = false;
+			}
+		}
+		return retval;
+	} else {
+		if (!(simubitmap & 0x03)) {
+			/* path A TX OK */
+			for (i = 0; i < 2; i++)
+				result[3][i] = result[c1][i];
+		}
+
+		if (!(simubitmap & 0x0c)) {
+			/* path A RX OK */
+			for (i = 2; i < 4; i++)
+				result[3][i] = result[c1][i];
+		}
+
+		if (!(simubitmap & 0x30) && priv->tx_paths > 1) {
+			/* path B RX OK */
+			for (i = 4; i < 6; i++)
+				result[3][i] = result[c1][i];
+		}
+
+		if (!(simubitmap & 0x30) && priv->tx_paths > 1) {
+			/* path B RX OK */
+			for (i = 6; i < 8; i++)
+				result[3][i] = result[c1][i];
+		}
+	}
+
+	return false;
+}
+
 static void
 rtl8xxxu_save_mac_regs(struct rtl8xxxu_priv *priv, const u32 *reg, u32 *backup)
 {
@@ -3677,6 +3762,369 @@ out:
 	return result;
 }
 
+static int rtl8723bu_iqk_path_a(struct rtl8xxxu_priv *priv)
+{
+	u32 reg_eac, reg_e94, reg_e9c, path_sel, val32;
+	int result = 0;
+
+	path_sel = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/*
+	 * Enable path A PA in TX IQK mode
+	 */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x20000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0003f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xc7f87);
+
+	/*
+	 * Tx IQK setting
+	 */
+	rtl8xxxu_write32(priv, REG_TX_IQK, 0x01007c00);
+	rtl8xxxu_write32(priv, REG_RX_IQK, 0x01004800);
+
+	/* path-A IQK setting */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x18008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_B, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_B, 0x38008c1c);
+
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_A, 0x821403ea);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_A, 0x28110000);
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_B, 0x82110000);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_B, 0x28110000);
+
+	/* LO calibration setting */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_RSP, 0x00462911);
+
+	/*
+	 * Enter IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	val32 |= 0x80800000;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/*
+	 * The vendor driver indicates the USB module is always using
+	 * S0S1 path 1 for the 8723bu. This may be different for 8192eu
+	 */
+	if (priv->rf_paths > 1)
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00000000);
+	else
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00000280);
+
+	/*
+	 * Bit 12 seems to be BT_GRANT, and is only found in the 8723bu.
+	 * No trace of this in the 8192eu or 8188eu vendor drivers.
+	 */
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, 0x00000800);
+
+	/* One shot, path A LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf9000000);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000);
+
+	mdelay(1);
+
+	/* Restore Ant Path */
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel);
+#ifdef RTL8723BU_BT
+	/* GNT_BT = 1 */
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, 0x00001800);
+#endif
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_e94 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_A);
+	reg_e9c = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_A);
+
+	val32 = (reg_e9c >> 16) & 0x3ff;
+	if (val32 & 0x200)
+		val32 = 0x400 - val32;
+
+	if (!(reg_eac & BIT(28)) &&
+	    ((reg_e94 & 0x03ff0000) != 0x01420000) &&
+	    ((reg_e9c & 0x03ff0000) != 0x00420000) &&
+	    ((reg_e94 & 0x03ff0000)  < 0x01100000) &&
+	    ((reg_e94 & 0x03ff0000)  > 0x00f00000) &&
+	    val32 < 0xf)
+		result |= 0x01;
+	else	/* If TX not OK, ignore RX */
+		goto out;
+
+out:
+	return result;
+}
+
+static int rtl8723bu_rx_iqk_path_a(struct rtl8xxxu_priv *priv)
+{
+	u32 reg_ea4, reg_eac, reg_e94, reg_e9c, path_sel, val32;
+	int result = 0;
+
+	path_sel = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/*
+	 * Enable path A PA in TX IQK mode
+	 */
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x30000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0001f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xf7fb7);
+
+	/*
+	 * Tx IQK setting
+	 */
+	rtl8xxxu_write32(priv, REG_TX_IQK, 0x01007c00);
+	rtl8xxxu_write32(priv, REG_RX_IQK, 0x01004800);
+
+	/* path-A IQK setting */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x18008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_B, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_B, 0x38008c1c);
+
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_A, 0x82160ff0);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_A, 0x28110000);
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_B, 0x82110000);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_B, 0x28110000);
+
+	/* LO calibration setting */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_RSP, 0x0046a911);
+
+	/*
+	 * Enter IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	val32 |= 0x80800000;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/*
+	 * The vendor driver indicates the USB module is always using
+	 * S0S1 path 1 for the 8723bu. This may be different for 8192eu
+	 */
+	if (priv->rf_paths > 1)
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00000000);
+	else
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00000280);
+
+	/*
+	 * Bit 12 seems to be BT_GRANT, and is only found in the 8723bu.
+	 * No trace of this in the 8192eu or 8188eu vendor drivers.
+	 */
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, 0x00000800);
+
+	/* One shot, path A LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf9000000);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000);
+
+	mdelay(1);
+
+	/* Restore Ant Path */
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel);
+#ifdef RTL8723BU_BT
+	/* GNT_BT = 1 */
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, 0x00001800);
+#endif
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_e94 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_A);
+	reg_e9c = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_A);
+
+	val32 = (reg_e9c >> 16) & 0x3ff;
+	if (val32 & 0x200)
+		val32 = 0x400 - val32;
+
+	if (!(reg_eac & BIT(28)) &&
+	    ((reg_e94 & 0x03ff0000) != 0x01420000) &&
+	    ((reg_e9c & 0x03ff0000) != 0x00420000) &&
+	    ((reg_e94 & 0x03ff0000)  < 0x01100000) &&
+	    ((reg_e94 & 0x03ff0000)  > 0x00f00000) &&
+	    val32 < 0xf)
+		result |= 0x01;
+	else	/* If TX not OK, ignore RX */
+		goto out;
+
+	val32 = 0x80007c00 | (reg_e94 &0x3ff0000) |
+		((reg_e9c & 0x3ff0000) >> 16);
+	rtl8xxxu_write32(priv, REG_TX_IQK, val32);
+
+	/*
+	 * Modify RX IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x30000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0001f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xf7d77);
+
+	/*
+	 * PA, PAD setting
+	 */
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, 0xf80);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_55, 0x4021f);
+
+	/*
+	 * RX IQK setting
+	 */
+	rtl8xxxu_write32(priv, REG_RX_IQK, 0x01004800);
+
+	/* path-A IQK setting */
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x18008c1c);
+	rtl8xxxu_write32(priv, REG_TX_IQK_TONE_B, 0x38008c1c);
+	rtl8xxxu_write32(priv, REG_RX_IQK_TONE_B, 0x38008c1c);
+
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_A, 0x82110000);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_A, 0x2816001f);
+	rtl8xxxu_write32(priv, REG_TX_IQK_PI_B, 0x82110000);
+	rtl8xxxu_write32(priv, REG_RX_IQK_PI_B, 0x28110000);
+
+	/* LO calibration setting */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_RSP, 0x0046a8d1);
+
+	/*
+	 * Enter IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	val32 |= 0x80800000;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	if (priv->rf_paths > 1)
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00000000);
+	else
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00000280);
+
+	/*
+	 * Disable BT
+	 */
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, 0x00000800);
+
+	/* One shot, path A LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf9000000);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_PTS, 0xf8000000);
+
+	mdelay(1);
+
+	/* Restore Ant Path */
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, path_sel);
+#ifdef RTL8723BU_BT
+	/* GNT_BT = 1 */
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, 0x00001800);
+#endif
+
+	/*
+	 * Leave IQK mode
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_ea4 = rtl8xxxu_read32(priv, REG_RX_POWER_BEFORE_IQK_A_2);
+
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_DF, 0x780);
+
+	val32 = (reg_eac >> 16) & 0x3ff;
+	if (val32 & 0x200)
+		val32 = 0x400 - val32;
+
+	if (!(reg_eac & BIT(27)) &&
+	    ((reg_ea4 & 0x03ff0000) != 0x01320000) &&
+	    ((reg_eac & 0x03ff0000) != 0x00360000) &&
+	    ((reg_ea4 & 0x03ff0000)  < 0x01100000) &&
+	    ((reg_ea4 & 0x03ff0000)  > 0x00f00000) &&
+	    val32 < 0xf)
+		result |= 0x02;
+	else	/* If TX not OK, ignore RX */
+		goto out;
+out:
+	return result;
+}
+
+#ifdef RTL8723BU_PATH_B
+static int rtl8723bu_iqk_path_b(struct rtl8xxxu_priv *priv)
+{
+	u32 reg_eac, reg_eb4, reg_ebc, reg_ec4, reg_ecc, path_sel;
+	int result = 0;
+
+	path_sel = rtl8xxxu_read32(priv, REG_S0S1_PATH_SWITCH);
+
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	/* One shot, path B LOK & IQK */
+	rtl8xxxu_write32(priv, REG_IQK_AGC_CONT, 0x00000002);
+	rtl8xxxu_write32(priv, REG_IQK_AGC_CONT, 0x00000000);
+
+	mdelay(1);
+
+	/* Check failed */
+	reg_eac = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_A_2);
+	reg_eb4 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_B);
+	reg_ebc = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_B);
+	reg_ec4 = rtl8xxxu_read32(priv, REG_RX_POWER_BEFORE_IQK_B_2);
+	reg_ecc = rtl8xxxu_read32(priv, REG_RX_POWER_AFTER_IQK_B_2);
+
+	if (!(reg_eac & BIT(31)) &&
+	    ((reg_eb4 & 0x03ff0000) != 0x01420000) &&
+	    ((reg_ebc & 0x03ff0000) != 0x00420000))
+		result |= 0x01;
+	else
+		goto out;
+
+	if (!(reg_eac & BIT(30)) &&
+	    (((reg_ec4 & 0x03ff0000) >> 16) != 0x132) &&
+	    (((reg_ecc & 0x03ff0000) >> 16) != 0x36))
+		result |= 0x02;
+	else
+		dev_warn(&priv->udev->dev, "%s: Path B RX IQK failed!\n",
+			 __func__);
+out:
+	return result;
+}
+#endif
+
 static void rtl8xxxu_phy_iqcalibrate(struct rtl8xxxu_priv *priv,
 				     int result[][8], int t)
 {
@@ -3879,6 +4327,234 @@ static void rtl8xxxu_phy_iqcalibrate(struct rtl8xxxu_priv *priv,
 	}
 }
 
+static void rtl8723bu_phy_iqcalibrate(struct rtl8xxxu_priv *priv,
+				      int result[][8], int t)
+{
+	struct device *dev = &priv->udev->dev;
+	u32 i, val32;
+	int path_a_ok /*, path_b_ok */;
+	int retry = 2;
+	const u32 adda_regs[RTL8XXXU_ADDA_REGS] = {
+		REG_FPGA0_XCD_SWITCH_CTRL, REG_BLUETOOTH,
+		REG_RX_WAIT_CCA, REG_TX_CCK_RFON,
+		REG_TX_CCK_BBON, REG_TX_OFDM_RFON,
+		REG_TX_OFDM_BBON, REG_TX_TO_RX,
+		REG_TX_TO_TX, REG_RX_CCK,
+		REG_RX_OFDM, REG_RX_WAIT_RIFS,
+		REG_RX_TO_RX, REG_STANDBY,
+		REG_SLEEP, REG_PMPD_ANAEN
+	};
+	const u32 iqk_mac_regs[RTL8XXXU_MAC_REGS] = {
+		REG_TXPAUSE, REG_BEACON_CTRL,
+		REG_BEACON_CTRL_1, REG_GPIO_MUXCFG
+	};
+	const u32 iqk_bb_regs[RTL8XXXU_BB_REGS] = {
+		REG_OFDM0_TRX_PATH_ENABLE, REG_OFDM0_TR_MUX_PAR,
+		REG_FPGA0_XCD_RF_SW_CTRL, REG_CONFIG_ANT_A, REG_CONFIG_ANT_B,
+		REG_FPGA0_XAB_RF_SW_CTRL, REG_FPGA0_XA_RF_INT_OE,
+		REG_FPGA0_XB_RF_INT_OE, REG_FPGA0_RF_MODE
+	};
+	u8 xa_agc = rtl8xxxu_read32(priv, REG_OFDM0_XA_AGC_CORE1) & 0xff;
+	u8 xb_agc = rtl8xxxu_read32(priv, REG_OFDM0_XB_AGC_CORE1) & 0xff;
+
+	/*
+	 * Note: IQ calibration must be performed after loading
+	 *       PHY_REG.txt , and radio_a, radio_b.txt
+	 */
+
+	if (t == 0) {
+		/* Save ADDA parameters, turn Path A ADDA on */
+		rtl8xxxu_save_regs(priv, adda_regs, priv->adda_backup,
+				   RTL8XXXU_ADDA_REGS);
+		rtl8xxxu_save_mac_regs(priv, iqk_mac_regs, priv->mac_backup);
+		rtl8xxxu_save_regs(priv, iqk_bb_regs,
+				   priv->bb_backup, RTL8XXXU_BB_REGS);
+	}
+
+	rtl8xxxu_path_adda_on(priv, adda_regs, true);
+
+	/* MAC settings */
+	rtl8xxxu_mac_calibration(priv, iqk_mac_regs, priv->mac_backup);
+
+	val32 = rtl8xxxu_read32(priv, REG_CCK0_AFE_SETTING);
+	val32 |= 0x0f000000;
+	rtl8xxxu_write32(priv, REG_CCK0_AFE_SETTING, val32);
+
+	rtl8xxxu_write32(priv, REG_OFDM0_TRX_PATH_ENABLE, 0x03a05600);
+	rtl8xxxu_write32(priv, REG_OFDM0_TR_MUX_PAR, 0x000800e4);
+	rtl8xxxu_write32(priv, REG_FPGA0_XCD_RF_SW_CTRL, 0x22204000);
+
+#ifdef RTL8723BU_PATH_B
+	/* Set RF mode to standby Path B */
+	if (priv->tx_paths > 1)
+		rtl8xxxu_write_rfreg(priv, RF_B, RF6052_REG_AC, 0x10000);
+#endif
+
+#if 0
+	/* Page B init */
+	rtl8xxxu_write32(priv, REG_CONFIG_ANT_A, 0x0f600000);
+
+	if (priv->tx_paths > 1)
+		rtl8xxxu_write32(priv, REG_CONFIG_ANT_B, 0x0f600000);
+#endif
+
+	/*
+	 * RX IQ calibration setting for 8723B D cut large current issue
+	 * when leaving IPS
+	 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x30000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0001f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xf7fb7);
+
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_ED);
+	val32 |= 0x20;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_ED, val32);
+
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_43, 0x60fbd);
+
+	for (i = 0; i < retry; i++) {
+		path_a_ok = rtl8723bu_iqk_path_a(priv);
+		if (path_a_ok == 0x01) {
+			val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+			val32 &= 0x000000ff;
+			rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+#if 0 /* Only needed in restore case, we may need this when going to suspend */
+			priv->RFCalibrateInfo.TxLOK[RF_A] =
+				rtl8xxxu_read_rfreg(priv, RF_A,
+						    RF6052_REG_TXM_IDAC);
+#endif
+
+			val32 = rtl8xxxu_read32(priv,
+						REG_TX_POWER_BEFORE_IQK_A);
+			result[t][0] = (val32 >> 16) & 0x3ff;
+			val32 = rtl8xxxu_read32(priv,
+						REG_TX_POWER_AFTER_IQK_A);
+			result[t][1] = (val32 >> 16) & 0x3ff;
+
+			break;
+		}
+	}
+
+	if (!path_a_ok)
+		dev_dbg(dev, "%s: Path A TX IQK failed!\n", __func__);
+
+	for (i = 0; i < retry; i++) {
+		path_a_ok = rtl8723bu_rx_iqk_path_a(priv);
+		if (path_a_ok == 0x03) {
+			val32 = rtl8xxxu_read32(priv,
+						REG_RX_POWER_BEFORE_IQK_A_2);
+			result[t][2] = (val32 >> 16) & 0x3ff;
+			val32 = rtl8xxxu_read32(priv,
+						REG_RX_POWER_AFTER_IQK_A_2);
+			result[t][3] = (val32 >> 16) & 0x3ff;
+
+			break;
+		}
+	}
+
+	if (!path_a_ok)
+		dev_dbg(dev, "%s: Path A RX IQK failed!\n", __func__);
+
+	if (priv->tx_paths > 1) {
+#if 1
+		dev_warn(dev, "%s: Path B not supported\n", __func__);
+#else
+
+		/*
+		 * Path A into standby
+		 */
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+		val32 &= 0x000000ff;
+		rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_AC, 0x10000);
+
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+		val32 &= 0x000000ff;
+		val32 |= 0x80800000;
+		rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+		/* Turn Path B ADDA on */
+		rtl8xxxu_path_adda_on(priv, adda_regs, false);
+
+		for (i = 0; i < retry; i++) {
+			path_b_ok = rtl8xxxu_iqk_path_b(priv);
+			if (path_b_ok == 0x03) {
+				val32 = rtl8xxxu_read32(priv, REG_TX_POWER_BEFORE_IQK_B);
+				result[t][4] = (val32 >> 16) & 0x3ff;
+				val32 = rtl8xxxu_read32(priv, REG_TX_POWER_AFTER_IQK_B);
+				result[t][5] = (val32 >> 16) & 0x3ff;
+				break;
+			}
+		}
+
+		if (!path_b_ok)
+			dev_dbg(dev, "%s: Path B IQK failed!\n", __func__);
+
+		for (i = 0; i < retry; i++) {
+			path_b_ok = rtl8723bu_rx_iqk_path_b(priv);
+			if (path_a_ok == 0x03) {
+				val32 = rtl8xxxu_read32(priv,
+							REG_RX_POWER_BEFORE_IQK_B_2);
+				result[t][6] = (val32 >> 16) & 0x3ff;
+				val32 = rtl8xxxu_read32(priv,
+							REG_RX_POWER_AFTER_IQK_B_2);
+				result[t][7] = (val32 >> 16) & 0x3ff;
+				break;
+			}
+		}
+
+		if (!path_b_ok)
+			dev_dbg(dev, "%s: Path B RX IQK failed!\n", __func__);
+#endif
+	}
+
+	/* Back to BB mode, load original value */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 &= 0x000000ff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+
+	if (t) {
+		/* Reload ADDA power saving parameters */
+		rtl8xxxu_restore_regs(priv, adda_regs, priv->adda_backup,
+				      RTL8XXXU_ADDA_REGS);
+
+		/* Reload MAC parameters */
+		rtl8xxxu_restore_mac_regs(priv, iqk_mac_regs, priv->mac_backup);
+
+		/* Reload BB parameters */
+		rtl8xxxu_restore_regs(priv, iqk_bb_regs,
+				      priv->bb_backup, RTL8XXXU_BB_REGS);
+
+		/* Restore RX initial gain */
+		val32 = rtl8xxxu_read32(priv, REG_OFDM0_XA_AGC_CORE1);
+		val32 &= 0xffffff00;
+		rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, val32 | 0x50);
+		rtl8xxxu_write32(priv, REG_OFDM0_XA_AGC_CORE1, val32 | xa_agc);
+
+		if (priv->tx_paths > 1) {
+			val32 = rtl8xxxu_read32(priv, REG_OFDM0_XB_AGC_CORE1);
+			val32 &= 0xffffff00;
+			rtl8xxxu_write32(priv, REG_OFDM0_XB_AGC_CORE1,
+					 val32 | 0x50);
+			rtl8xxxu_write32(priv, REG_OFDM0_XB_AGC_CORE1,
+					 val32 | xb_agc);
+		}
+
+		/* Load 0xe30 IQC default value */
+		rtl8xxxu_write32(priv, REG_TX_IQK_TONE_A, 0x01008c00);
+		rtl8xxxu_write32(priv, REG_RX_IQK_TONE_A, 0x01008c00);
+	}
+}
+
 static void rtl8xxxu_prepare_calibrate(struct rtl8xxxu_priv *priv, u8 start)
 {
 	struct h2c_cmd h2c;
@@ -3893,7 +4569,7 @@ static void rtl8xxxu_prepare_calibrate(struct rtl8xxxu_priv *priv, u8 start)
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_wlan_calibration));
 }
 
-static void rtl8723a_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
+static void rtl8723au_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
 {
 	struct device *dev = &priv->udev->dev;
 	int result[4][8];	/* last is final result */
@@ -3997,6 +4673,133 @@ static void rtl8723a_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_prepare_calibrate(priv, 0);
 }
 
+static void rtl8723bu_phy_iq_calibrate(struct rtl8xxxu_priv *priv)
+{
+	struct device *dev = &priv->udev->dev;
+	int result[4][8];	/* last is final result */
+	int i, candidate;
+	bool path_a_ok, path_b_ok;
+	u32 reg_e94, reg_e9c, reg_ea4, reg_eac;
+	u32 reg_eb4, reg_ebc, reg_ec4, reg_ecc;
+	u32 val32, bt_control;
+	s32 reg_tmp = 0;
+	bool simu;
+
+	rtl8xxxu_prepare_calibrate(priv, 1);
+
+	memset(result, 0, sizeof(result));
+	candidate = -1;
+
+	path_a_ok = false;
+	path_b_ok = false;
+
+	bt_control = rtl8xxxu_read32(priv, REG_BT_CONTROL_8723BU);
+
+	for (i = 0; i < 3; i++) {
+		rtl8723bu_phy_iqcalibrate(priv, result, i);
+
+		if (i == 1) {
+			simu = rtl8723bu_simularity_compare(priv, result, 0, 1);
+			if (simu) {
+				candidate = 0;
+				break;
+			}
+		}
+
+		if (i == 2) {
+			simu = rtl8723bu_simularity_compare(priv, result, 0, 2);
+			if (simu) {
+				candidate = 0;
+				break;
+			}
+
+			simu = rtl8723bu_simularity_compare(priv, result, 1, 2);
+			if (simu) {
+				candidate = 1;
+			} else {
+				for (i = 0; i < 8; i++)
+					reg_tmp += result[3][i];
+
+				if (reg_tmp)
+					candidate = 3;
+				else
+					candidate = -1;
+			}
+		}
+	}
+
+	for (i = 0; i < 4; i++) {
+		reg_e94 = result[i][0];
+		reg_e9c = result[i][1];
+		reg_ea4 = result[i][2];
+		reg_eac = result[i][3];
+		reg_eb4 = result[i][4];
+		reg_ebc = result[i][5];
+		reg_ec4 = result[i][6];
+		reg_ecc = result[i][7];
+	}
+
+	if (candidate >= 0) {
+		reg_e94 = result[candidate][0];
+		priv->rege94 =  reg_e94;
+		reg_e9c = result[candidate][1];
+		priv->rege9c = reg_e9c;
+		reg_ea4 = result[candidate][2];
+		reg_eac = result[candidate][3];
+		reg_eb4 = result[candidate][4];
+		priv->regeb4 = reg_eb4;
+		reg_ebc = result[candidate][5];
+		priv->regebc = reg_ebc;
+		reg_ec4 = result[candidate][6];
+		reg_ecc = result[candidate][7];
+		dev_dbg(dev, "%s: candidate is %x\n", __func__, candidate);
+		dev_dbg(dev,
+			"%s: e94 =%x e9c=%x ea4=%x eac=%x eb4=%x ebc=%x ec4=%x "
+			"ecc=%x\n ", __func__, reg_e94, reg_e9c,
+			reg_ea4, reg_eac, reg_eb4, reg_ebc, reg_ec4, reg_ecc);
+		path_a_ok = true;
+		path_b_ok = true;
+	} else {
+		reg_e94 = reg_eb4 = priv->rege94 = priv->regeb4 = 0x100;
+		reg_e9c = reg_ebc = priv->rege9c = priv->regebc = 0x0;
+	}
+
+	if (reg_e94 && candidate >= 0)
+		rtl8xxxu_fill_iqk_matrix_a(priv, path_a_ok, result,
+					   candidate, (reg_ea4 == 0));
+
+	if (priv->tx_paths > 1 && reg_eb4)
+		rtl8xxxu_fill_iqk_matrix_b(priv, path_b_ok, result,
+					   candidate, (reg_ec4 == 0));
+
+	rtl8xxxu_save_regs(priv, rtl8723au_iqk_phy_iq_bb_reg,
+			   priv->bb_recovery_backup, RTL8XXXU_BB_REGS);
+
+	rtl8xxxu_write32(priv, REG_BT_CONTROL_8723BU, bt_control);
+
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_WE_LUT);
+	val32 |= 0x80000;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_WE_LUT, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_RCK_OS, 0x18000);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G1, 0x0001f);
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_TXPA_G2, 0xe6177);
+	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_ED);
+	val32 |= 0x20;
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_UNKNOWN_ED, val32);
+	rtl8xxxu_write_rfreg(priv, RF_A, 0x43, 0x300bd);
+
+	if (priv->rf_paths > 1) {
+		dev_dbg(dev, "%s: beware 2T not yet supported\n", __func__);
+#ifdef RTL8723BU_PATH_B
+		if (RF_Path == 0x0)	//S1
+			ODM_SetIQCbyRFpath(pDM_Odm, 0);
+		else	//S0
+			ODM_SetIQCbyRFpath(pDM_Odm, 1);
+#endif
+	}
+	rtl8xxxu_prepare_calibrate(priv, 0);
+}
+
 static void rtl8723a_phy_lc_calibrate(struct rtl8xxxu_priv *priv)
 {
 	u32 val32;
@@ -4995,7 +5798,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	rtl8xxxu_write16(priv, REG_FAST_EDCA_CTRL, 0);
 
-	rtl8723a_phy_iq_calibrate(priv);
+	priv->fops->phy_iq_calibrate(priv);
 
 	/*
 	 * This should enable thermal meter
@@ -6536,6 +7339,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.load_firmware = rtl8723au_load_firmware,
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_init_llt_table,
+	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -6551,6 +7355,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
+	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -6567,6 +7372,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.load_firmware = rtl8192cu_load_firmware,
 	.power_on = rtl8192cu_power_on,
 	.llt_init = rtl8xxxu_init_llt_table,
+	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -6583,6 +7389,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.load_firmware = rtl8192eu_load_firmware,
 	.power_on = rtl8192eu_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
+	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 4b8431d..76398de 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -851,6 +851,7 @@ struct rtl8xxxu_fileops {
 	int (*power_on) (struct rtl8xxxu_priv *priv);
 	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
+	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 162a9e8..499210f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -708,6 +708,12 @@
 #define REG_MACID1			0x0700
 #define REG_BSSID1			0x0708
 
+/*
+ * This seems to be 8723bu specific
+ */
+#define REG_BT_CONTROL_8723BU		0x0764
+#define  BT_CONTROL_BT_GRANT		BIT(12)
+
 #define REG_FPGA0_RF_MODE		0x0800
 #define  FPGA_RF_MODE			BIT(0)
 #define  FPGA_RF_MODE_JAPAN		BIT(1)
@@ -810,6 +816,7 @@
 #define REG_RFE_CTRL_ANTA_SRC		0x0930	/* 8723BU */
 #define REG_RFE_PATH_SELECT		0x0940	/* 8723BU */
 #define REG_RFE_BUFFER			0x0944	/* 8723BU */
+#define REG_S0S1_PATH_SWITCH		0x0948	/* 8723BU */
 
 #define REG_CCK0_SYSTEM			0x0a00
 #define  CCK0_SIDEBAND			BIT(4)
@@ -1050,3 +1057,13 @@
 #define RF6052_REG_TXPA_G1		0x31	/* RF TX PA control */
 #define RF6052_REG_TXPA_G2		0x32	/* RF TX PA control */
 #define RF6052_REG_TXPA_G3		0x33	/* RF TX PA control */
+
+/*
+ * NextGen regs: 8723BU
+ */
+#define	RF6052_REG_UNKNOWN_43		0x43
+#define	RF6052_REG_UNKNOWN_55		0x55
+#define	RF6052_REG_S0S1			0xb0
+#define	RF6052_REG_UNKNOWN_DF		0xdf
+#define	RF6052_REG_UNKNOWN_ED		0xed
+#define	RF6052_REG_WE_LUT		0xef
-- 
cgit v0.10.2


From 0d698dec06f828480a95a0c7a91c13bc4ad374e4 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:36 -0500
Subject: rtl8xxxu: Handle S0S1 register in lc_calibrate()

Newer chips (8723bu/8192eu) has S0S1 settings which needs to be dealt
with during LC calibration.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index a40ef21..cfd016c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -4835,12 +4835,17 @@ static void rtl8723a_phy_lc_calibrate(struct rtl8xxxu_priv *priv)
 	}
 
 	/* Start LC calibration */
+	if (priv->fops->has_s0s1)
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_S0S1, 0xdfbe0);
 	val32 = rtl8xxxu_read_rfreg(priv, RF_A, RF6052_REG_MODE_AG);
 	val32 |= 0x08000;
 	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_MODE_AG, val32);
 
 	msleep(100);
 
+	if (priv->fops->has_s0s1)
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_S0S1, 0xdffe0);
+
 	/* Restore original parameters */
 	if (lstf & OFDM_LSTF_MASK) {
 		/* Path-A */
@@ -7359,6 +7364,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
+	.has_s0s1 = 1,
 	.adda_1t_init = 0x01c00014,
 	.adda_1t_path_on = 0x01c00014,
 	.adda_2t_path_on_a = 0x01c00014,
@@ -7393,6 +7399,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
+	.has_s0s1 = 1,
 	.adda_1t_init = 0x0fc01616,
 	.adda_1t_path_on = 0x0fc01616,
 	.adda_2t_path_on_a = 0x0fc01616,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 76398de..4a974a6 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -855,6 +855,7 @@ struct rtl8xxxu_fileops {
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
+	char has_s0s1;
 	u32 adda_1t_init;
 	u32 adda_1t_path_on;
 	u32 adda_2t_path_on_a;
-- 
cgit v0.10.2


From fa0f2d481d178fa2bf44f9488a695ab56d4e78f6 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:37 -0500
Subject: rtl8xxxu: Do LC calibration before IQK calibration

This matches the flow of the vendor driver for newer hardware, and
doesn't seem to cause issues for the older parts.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index cfd016c..62537bd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5803,6 +5803,8 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	rtl8xxxu_write16(priv, REG_FAST_EDCA_CTRL, 0);
 
+	rtl8723a_phy_lc_calibrate(priv);
+
 	priv->fops->phy_iq_calibrate(priv);
 
 	/*
@@ -5810,8 +5812,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	 */
 	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_T_METER, 0x60);
 
-	rtl8723a_phy_lc_calibrate(priv);
-
 	/* Init BT hw config. */
 	rtl8xxxu_init_bt(priv);
 
-- 
cgit v0.10.2


From 60c76cbadb61084139d710ca710959274efdde5b Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:38 -0500
Subject: rtl8xxxu: Remove backing up certain registers, which was never used

This was inspired by the vendor driver, but in the end never used for
anything.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 62537bd..4579a3e 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5815,20 +5815,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/* Init BT hw config. */
 	rtl8xxxu_init_bt(priv);
 
-	/*
-	 * Not sure if we really need to save these parameters, but the
-	 * vendor driver does
-	 */
-	val32 = rtl8xxxu_read32(priv, REG_FPGA0_XA_HSSI_PARM2);
-	if (val32 & FPGA0_HSSI_PARM2_CCK_HIGH_PWR)
-		priv->path_a_hi_power = 1;
-
-	val32 = rtl8xxxu_read32(priv, REG_OFDM0_TRX_PATH_ENABLE);
-	priv->path_a_rf_paths = val32 & OFDM_RF_PATH_RX_MASK;
-
-	val32 = rtl8xxxu_read32(priv, REG_OFDM0_XA_AGC_CORE1);
-	priv->path_a_ig_value = val32 & OFDM0_X_AGC_CORE1_IGI_MASK;
-
 	/* Set NAV_UPPER to 30000us */
 	val8 = ((30000 + NAV_UPPER_UNIT - 1) / NAV_UPPER_UNIT);
 	rtl8xxxu_write8(priv, REG_NAV_UPPER, val8);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 4a974a6..7b76e11 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -785,13 +785,10 @@ struct rtl8xxxu_priv {
 	u32 ep_tx_high_queue:1;
 	u32 ep_tx_normal_queue:1;
 	u32 ep_tx_low_queue:1;
-	u32 path_a_hi_power:1;
-	u32 path_a_rf_paths:4;
 	unsigned int pipe_interrupt;
 	unsigned int pipe_in;
 	unsigned int pipe_out[TXDESC_QUEUE_MAX];
 	u8 out_ep[RTL8XXXU_OUT_ENDPOINTS];
-	u8 path_a_ig_value;
 	u8 ep_tx_count;
 	u8 rf_paths;
 	u8 rx_paths;
-- 
cgit v0.10.2


From e6f9a9c3b5235f31434a9de9034ea5eda1532113 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jsitnicki@gmail.com>
Date: Mon, 29 Feb 2016 17:04:39 -0500
Subject: rtl8xxxu: rtl8192eu: Map out EFUSE TX power area

TX power values are laid out differently in EFUSE found in RTL8192EU &
RTL8188EU devices.  TX power indices and differences for each RF path
are not interleaved (A, B, A, B), as in other chips, but follow one
another (A, B, C, D).

Signed-off-by: Jakub Sitnicki <jsitnicki@gmail.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4579a3e..4edd9c7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2293,16 +2293,6 @@ static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
-	memcpy(priv->cck_tx_power_index_A, efuse->cck_tx_power_index_A,
-	       sizeof(priv->cck_tx_power_index_A));
-	memcpy(priv->cck_tx_power_index_B, efuse->cck_tx_power_index_B,
-	       sizeof(priv->cck_tx_power_index_B));
-
-	memcpy(priv->ht40_1s_tx_power_index_A, efuse->ht40_1s_tx_power_index_A,
-	       sizeof(priv->ht40_1s_tx_power_index_A));
-	memcpy(priv->ht40_1s_tx_power_index_B, efuse->ht40_1s_tx_power_index_B,
-	       sizeof(priv->ht40_1s_tx_power_index_B));
-
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
 	dev_info(&priv->udev->dev, "Product: %.11s\n", efuse->device_name);
 	dev_info(&priv->udev->dev, "Serial: %.11s\n", efuse->serial);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 7b76e11..6c6bac2 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -543,14 +543,26 @@ struct rtl8723bu_efuse {
 	u8 res12[0x4];
 };
 
+struct rtl8192eu_efuse_tx_power {
+	u8 cck_base[6];
+	u8 ht40_base[5];
+	struct rtl8723au_idx ht20_ofdm_1s_diff;
+	struct rtl8723au_idx ht40_ht20_2s_diff;
+	struct rtl8723au_idx ofdm_cck_2s_diff; /* not used */
+	struct rtl8723au_idx ht40_ht20_3s_diff;
+	struct rtl8723au_idx ofdm_cck_3s_diff; /* not used */
+	struct rtl8723au_idx ht40_ht20_4s_diff;
+	struct rtl8723au_idx ofdm_cck_4s_diff; /* not used */
+};
+
 struct rtl8192eu_efuse {
 	__le16 rtl_id;
 	u8 res0[0x0e];
-	u8 cck_tx_power_index_A[3];	/* 0x10 */
-	u8 cck_tx_power_index_B[3];
-	u8 ht40_1s_tx_power_index_A[3];	/* 0x16 */
-	u8 ht40_1s_tx_power_index_B[3];
-	u8 res1[0x9c];
+	struct rtl8192eu_efuse_tx_power tx_power_index_A;	/* 0x10 */
+	struct rtl8192eu_efuse_tx_power tx_power_index_B;	/* 0x22 */
+	struct rtl8192eu_efuse_tx_power tx_power_index_C;	/* 0x34 */
+	struct rtl8192eu_efuse_tx_power tx_power_index_D;	/* 0x46 */
+	u8 res1[0x60];
 	u8 channel_plan;		/* 0xb8 */
 	u8 xtal_k;
 	u8 thermal_meter;
-- 
cgit v0.10.2


From c3f9506f2374772e579e1c19b86d79f1da71dea2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:40 -0500
Subject: rtl8xxxu: Initial implementation of rtl8723bu_config_channel()

This is a first stab of implementing rtl8723bu_config_channel(). For
now this will only do 20MHz channels.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4edd9c7..ada0f04 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1777,6 +1777,136 @@ static void rtl8723au_config_channel(struct ieee80211_hw *hw)
 	}
 }
 
+static void rtl8723bu_config_channel(struct ieee80211_hw *hw)
+{
+	struct rtl8xxxu_priv *priv = hw->priv;
+	u32 val32, rsr;
+	u8 val8, opmode, subchannel;
+	u16 rf_mode_bw;
+	bool ht = true;
+	int sec_ch_above, channel;
+	int i;
+
+	rf_mode_bw = rtl8xxxu_read16(priv, REG_WMAC_TRXPTCL_CTL);
+	rf_mode_bw &= ~WMAC_TRXPTCL_CTL_BW_MASK;
+	rsr = rtl8xxxu_read32(priv, REG_RESPONSE_RATE_SET);
+	channel = hw->conf.chandef.chan->hw_value;
+
+/* Hack */
+	subchannel = 0;
+
+	switch (hw->conf.chandef.width) {
+	case NL80211_CHAN_WIDTH_20_NOHT:
+		ht = false;
+	case NL80211_CHAN_WIDTH_20:
+		rf_mode_bw |= WMAC_TRXPTCL_CTL_BW_20;
+		subchannel = 0;
+
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
+		val32 &= ~FPGA_RF_MODE;
+		rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_FPGA1_RF_MODE);
+		val32 &= ~FPGA_RF_MODE;
+		rtl8xxxu_write32(priv, REG_FPGA1_RF_MODE, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_OFDM0_TX_PSDO_NOISE_WEIGHT);
+		val32 &= ~(BIT(30) | BIT(31));
+		rtl8xxxu_write32(priv, REG_OFDM0_TX_PSDO_NOISE_WEIGHT, val32);
+
+		break;
+	case NL80211_CHAN_WIDTH_40:
+		rf_mode_bw |= WMAC_TRXPTCL_CTL_BW_40;
+
+		if (hw->conf.chandef.center_freq1 >
+		    hw->conf.chandef.chan->center_freq) {
+			sec_ch_above = 1;
+			channel += 2;
+		} else {
+			sec_ch_above = 0;
+			channel -= 2;
+		}
+
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
+		val32 |= FPGA_RF_MODE;
+		rtl8xxxu_write32(priv, REG_FPGA0_RF_MODE, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_FPGA1_RF_MODE);
+		val32 |= FPGA_RF_MODE;
+		rtl8xxxu_write32(priv, REG_FPGA1_RF_MODE, val32);
+
+		/*
+		 * Set Control channel to upper or lower. These settings
+		 * are required only for 40MHz
+		 */
+		val32 = rtl8xxxu_read32(priv, REG_CCK0_SYSTEM);
+		val32 &= ~CCK0_SIDEBAND;
+		if (!sec_ch_above)
+			val32 |= CCK0_SIDEBAND;
+		rtl8xxxu_write32(priv, REG_CCK0_SYSTEM, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_OFDM1_LSTF);
+		val32 &= ~OFDM_LSTF_PRIME_CH_MASK; /* 0xc00 */
+		if (sec_ch_above)
+			val32 |= OFDM_LSTF_PRIME_CH_LOW;
+		else
+			val32 |= OFDM_LSTF_PRIME_CH_HIGH;
+		rtl8xxxu_write32(priv, REG_OFDM1_LSTF, val32);
+
+		val32 = rtl8xxxu_read32(priv, REG_FPGA0_POWER_SAVE);
+		val32 &= ~(FPGA0_PS_LOWER_CHANNEL | FPGA0_PS_UPPER_CHANNEL);
+		if (sec_ch_above)
+			val32 |= FPGA0_PS_UPPER_CHANNEL;
+		else
+			val32 |= FPGA0_PS_LOWER_CHANNEL;
+		rtl8xxxu_write32(priv, REG_FPGA0_POWER_SAVE, val32);
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		rf_mode_bw |= WMAC_TRXPTCL_CTL_BW_80;
+		break;
+	default:
+		break;
+	}
+
+	for (i = RF_A; i < priv->rf_paths; i++) {
+		val32 = rtl8xxxu_read_rfreg(priv, i, RF6052_REG_MODE_AG);
+		val32 &= ~MODE_AG_CHANNEL_MASK;
+		val32 |= channel;
+		rtl8xxxu_write_rfreg(priv, i, RF6052_REG_MODE_AG, val32);
+	}
+
+	rtl8xxxu_write16(priv, REG_WMAC_TRXPTCL_CTL, rf_mode_bw);
+	rtl8xxxu_write8(priv, REG_DATA_SUBCHANNEL, subchannel);
+
+	if (ht)
+		val8 = 0x0e;
+	else
+		val8 = 0x0a;
+
+	rtl8xxxu_write8(priv, REG_SIFS_CCK + 1, val8);
+	rtl8xxxu_write8(priv, REG_SIFS_OFDM + 1, val8);
+
+	rtl8xxxu_write16(priv, REG_R2T_SIFS, 0x0808);
+	rtl8xxxu_write16(priv, REG_T2T_SIFS, 0x0a0a);
+
+	for (i = RF_A; i < priv->rf_paths; i++) {
+		val32 = rtl8xxxu_read_rfreg(priv, i, RF6052_REG_MODE_AG);
+		val32 &= ~MODE_AG_BW_MASK;
+		switch(hw->conf.chandef.width) {
+		case NL80211_CHAN_WIDTH_80:
+			val32 |= MODE_AG_BW_80MHZ_8723B;
+			break;
+		case NL80211_CHAN_WIDTH_40:
+			val32 |= MODE_AG_BW_40MHZ_8723B;
+			break;
+		default:
+			val32 |= MODE_AG_BW_20MHZ_8723B;
+			break;
+		}
+		rtl8xxxu_write_rfreg(priv, i, RF6052_REG_MODE_AG, val32);
+	}
+}
+
 static void
 rtl8723a_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 {
@@ -7321,6 +7451,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.power_on = rtl8723au_power_on,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
+	.config_channel = rtl8723au_config_channel,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -7337,6 +7468,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
+	.config_channel = rtl8723bu_config_channel,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -7355,6 +7487,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.power_on = rtl8192cu_power_on,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
+	.config_channel = rtl8723au_config_channel,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -7372,6 +7505,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.power_on = rtl8192eu_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
+	.config_channel = rtl8723bu_config_channel,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 6c6bac2..bce52cd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -861,6 +861,7 @@ struct rtl8xxxu_fileops {
 	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
+	void (*config_channel) (struct ieee80211_hw *hw);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 499210f..147deba 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -473,6 +473,9 @@
 #define REG_FAST_EDCA_CTRL		0x0460
 #define REG_RD_RESP_PKT_TH		0x0463
 #define REG_INIRTS_RATE_SEL		0x0480
+/* 8723bu */
+#define REG_DATA_SUBCHANNEL		0x0483
+/* 8723au */
 #define REG_INIDATA_RATE_SEL		0x0484
 
 #define REG_POWER_STATUS		0x04a4
@@ -658,6 +661,10 @@
 #define REG_FWDLY			0x0661
 #define REG_RXERR_RPT			0x0664
 #define REG_WMAC_TRXPTCL_CTL		0x0668
+#define  WMAC_TRXPTCL_CTL_BW_MASK	(BIT(7) | BIT(8))
+#define  WMAC_TRXPTCL_CTL_BW_20		0
+#define  WMAC_TRXPTCL_CTL_BW_40		BIT(7)
+#define  WMAC_TRXPTCL_CTL_BW_80		BIT(8)
 
 /*  Security */
 #define REG_CAM_CMD			0x0670
@@ -872,6 +879,9 @@
 
 #define REG_OFDM0_RX_IQ_EXT_ANTA	0x0ca0
 
+/* 8723bu */
+#define REG_OFDM0_TX_PSDO_NOISE_WEIGHT	0x0ce4
+
 #define REG_OFDM1_LSTF			0x0d00
 #define  OFDM_LSTF_PRIME_CH_LOW		BIT(10)
 #define  OFDM_LSTF_PRIME_CH_HIGH	BIT(11)
@@ -1030,6 +1040,10 @@
 #define RF6052_REG_MODE_AG		0x18	/* RF channel and BW switch */
 #define  MODE_AG_CHANNEL_MASK		0x3ff
 #define  MODE_AG_CHANNEL_20MHZ		BIT(10)
+#define  MODE_AG_BW_MASK		(BIT(10) | BIT(11))
+#define  MODE_AG_BW_20MHZ_8723B		(BIT(10) | BIT(11))
+#define  MODE_AG_BW_40MHZ_8723B		BIT(10)
+#define  MODE_AG_BW_80MHZ_8723B		0
 
 #define RF6052_REG_TOP			0x19
 #define RF6052_REG_RX_G1		0x1a
-- 
cgit v0.10.2


From f37e9228aecaa9c92ee0433cd45ca0c545b90ce4 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:41 -0500
Subject: rtl8xxxu: Initial rtl8723bu_init_bt() code

This should initialize the antennas on the 8723bu, but so far I am
still not receiving anything :( More work is needed.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ada0f04..fa8a92a 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5603,10 +5603,87 @@ static void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e);
 }
 
-static void rtl8xxxu_init_bt(struct rtl8xxxu_priv *priv)
+static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 {
-	if (!priv->has_bluetooth)
-		return;
+	struct h2c_cmd h2c;
+	u32 val32;
+	u8 val8;
+
+	/*
+	 * No indication anywhere as to what 0x0790 does. The 2 antenna
+	 * vendor code preserves bits 6-7 here.
+	 */
+	rtl8xxxu_write8(priv, 0x0790, 0x05);
+	/*
+	 * 0x0778 seems to be related to enabling the number of antennas
+	 * In the vendor driver halbtc8723b2ant_InitHwConfig() sets it
+	 * to 0x03, while halbtc8723b1ant_InitHwConfig() sets it to 0x01
+	 */
+	rtl8xxxu_write8(priv, 0x0778, 0x01);
+
+	val8 = rtl8xxxu_read8(priv, REG_GPIO_MUXCFG);
+	val8 |= BIT(5);
+	rtl8xxxu_write8(priv, REG_GPIO_MUXCFG, val8);
+
+	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_IQADJ_G1, 0x780);
+
+	/*
+	 * Set BT grant to low
+	 */
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.bt_grant.cmd = H2C_8723B_BT_GRANT;
+	h2c.bt_grant.data = 0;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_grant));
+
+	/*
+	 * WLAN action by PTA
+	 */
+	rtl8xxxu_write8(priv, REG_WLAN_ACT_CONTROL_8723B, 0x0c);
+
+	/*
+	 * BT select S0/S1 controlled by WiFi
+	 */
+	val8 = rtl8xxxu_read8(priv, 0x0067);
+	val8 |= BIT(5);
+	rtl8xxxu_write8(priv, 0x0067, val8);
+
+	val32 = rtl8xxxu_read32(priv, REG_PWR_DATA);
+	val32 |= BIT(11);
+	rtl8xxxu_write32(priv, REG_PWR_DATA, val32);
+
+	/*
+	 * Bits 6/7 are marked in/out ... but for what?
+	 */
+	rtl8xxxu_write8(priv, 0x0974, 0xff);
+
+	val32 = rtl8xxxu_read32(priv, 0x0944);
+	val32 |= (BIT(0) | BIT(1));
+	rtl8xxxu_write32(priv, 0x0944, val32);
+
+	rtl8xxxu_write8(priv, REG_RFE_CTRL_ANTA_SRC, 0x77);
+
+	val32 = rtl8xxxu_read32(priv, REG_LEDCFG0);
+	val32 &= ~BIT(24);
+	val32 |= BIT(23);
+	rtl8xxxu_write32(priv, REG_LEDCFG0, val32);
+
+	/*
+	 * Fix external switch Main->S1, Aux->S0
+	 */
+	val8 = rtl8xxxu_read8(priv, REG_PAD_CTRL1);
+	val8 &= ~BIT(0);
+	rtl8xxxu_write8(priv, REG_PAD_CTRL1, val8);
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.ant_sel_rsv.cmd = H2C_8723B_ANT_SEL_RSV;
+	h2c.ant_sel_rsv.ant_inverse = 1;
+	h2c.ant_sel_rsv.int_switch_type = 0;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ant_sel_rsv));
+
+	/*
+	 * 0x280, 0x00, 0x200, 0x80 - not clear
+	 */
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x280);
 }
 
 static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
@@ -5933,7 +6010,8 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_T_METER, 0x60);
 
 	/* Init BT hw config. */
-	rtl8xxxu_init_bt(priv);
+	if (priv->fops->init_bt)
+		priv->fops->init_bt(priv);
 
 	/* Set NAV_UPPER to 30000us */
 	val8 = ((30000 + NAV_UPPER_UNIT - 1) / NAV_UPPER_UNIT);
@@ -7469,6 +7547,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
+	.init_bt = rtl8723bu_init_bt,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index bce52cd..bcbdc78 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -696,8 +696,9 @@ enum h2c_cmd_8723b {
 	H2C_8723B_BT_MP_OPER = 0x67,
 	H2C_8723B_BT_CONTROL = 0x68,
 	H2C_8723B_BT_WIFI_CTRL = 0x69,
-	H2C_8723B_BT_FW_PATCH = 0x6A,
-	H2C_8723B_BT_WLAN_CALIBRATION = 0x6D,
+	H2C_8723B_BT_FW_PATCH = 0x6a,
+	H2C_8723B_BT_WLAN_CALIBRATION = 0x6d,
+	H2C_8723B_BT_GRANT = 0x6e,
 
 	/*
 	 * WOWLAN Class: 100
@@ -743,6 +744,15 @@ struct h2c_cmd {
 			u8 cmd;
 			u8 data;
 		} __packed bt_wlan_calibration;
+		struct {
+			u8 cmd;
+			u8 ant_inverse;
+			u8 int_switch_type;
+		} __packed ant_sel_rsv;
+		struct {
+			u8 cmd;
+			u8 data;
+		} __packed bt_grant;
 	};
 };
 
@@ -862,6 +872,7 @@ struct rtl8xxxu_fileops {
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
 	void (*config_channel) (struct ieee80211_hw *hw);
+	void (*init_bt) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 147deba..248f64c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -721,6 +721,8 @@
 #define REG_BT_CONTROL_8723BU		0x0764
 #define  BT_CONTROL_BT_GRANT		BIT(12)
 
+#define REG_WLAN_ACT_CONTROL_8723B	0x076e
+
 #define REG_FPGA0_RF_MODE		0x0800
 #define  FPGA_RF_MODE			BIT(0)
 #define  FPGA_RF_MODE_JAPAN		BIT(1)
-- 
cgit v0.10.2


From 368633ce6853bb2ba678c7cb5bfeb103e7bc56ba Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:42 -0500
Subject: rtl8xxxu: Remove unused variable

Remove an unused variable to make the compiler happy.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index fa8a92a..7054394 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1781,7 +1781,7 @@ static void rtl8723bu_config_channel(struct ieee80211_hw *hw)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
 	u32 val32, rsr;
-	u8 val8, opmode, subchannel;
+	u8 val8, subchannel;
 	u16 rf_mode_bw;
 	bool ht = true;
 	int sec_ch_above, channel;
-- 
cgit v0.10.2


From 3ca7b32c9d6da17a2660bc40b24f5dc0cb433cc9 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:43 -0500
Subject: rtl8xxxu: Improve 8723bu init code

Implement additional init sequence code for the 8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 7054394..91f189a 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2984,14 +2984,18 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 	 *       addresses, which is initialized here. Do we need this?
 	 */
 
-	val8 = rtl8xxxu_read8(priv, REG_AFE_PLL_CTRL);
-	udelay(2);
-	val8 |= AFE_PLL_320_ENABLE;
-	rtl8xxxu_write8(priv, REG_AFE_PLL_CTRL, val8);
-	udelay(2);
+	if (priv->rtlchip == 0x8723b) {
+		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00);
+	} else {
+		val8 = rtl8xxxu_read8(priv, REG_AFE_PLL_CTRL);
+		udelay(2);
+		val8 |= AFE_PLL_320_ENABLE;
+		rtl8xxxu_write8(priv, REG_AFE_PLL_CTRL, val8);
+		udelay(2);
 
-	rtl8xxxu_write8(priv, REG_AFE_PLL_CTRL + 1, 0xff);
-	udelay(2);
+		rtl8xxxu_write8(priv, REG_AFE_PLL_CTRL + 1, 0xff);
+		udelay(2);
+	}
 
 	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC);
 	val8 |= SYS_FUNC_BB_GLB_RSTN | SYS_FUNC_BBRSTB;
@@ -5603,6 +5607,21 @@ static void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e);
 }
 
+static void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv,
+				  u8 arg1, u8 arg2, u8 arg3, u8 arg4, u8 arg5)
+{
+	struct h2c_cmd h2c;
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.b_type_dma.cmd = H2C_8723B_B_TYPE_TDMA;
+	h2c.b_type_dma.data1 = arg1;
+	h2c.b_type_dma.data2 = arg2;
+	h2c.b_type_dma.data3 = arg3;
+	h2c.b_type_dma.data4 = arg4;
+	h2c.b_type_dma.data5 = arg5;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.b_type_dma));
+}
+
 static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 {
 	struct h2c_cmd h2c;
@@ -5683,7 +5702,17 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	/*
 	 * 0x280, 0x00, 0x200, 0x80 - not clear
 	 */
-	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x280);
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00);
+
+	/*
+	 * Software control, antenna at WiFi side
+	 */
+	rtl8723bu_set_ps_tdma(priv, 0x00, 0x00, 0x00, 0x00, 0x00);
+
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE1, 0x55555555);
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE2, 0x5a5a5a5a);
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE3, 0x00ffffff);
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE4, 0x00000003);
 }
 
 static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index bcbdc78..c4c0b28 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -742,6 +742,14 @@ struct h2c_cmd {
 		} __packed ramask;
 		struct {
 			u8 cmd;
+			u8 data1;
+			u8 data2;
+			u8 data3;
+			u8 data4;
+			u8 data5;
+		} __packed b_type_dma;
+		struct {
+			u8 cmd;
 			u8 data;
 		} __packed bt_wlan_calibration;
 		struct {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 248f64c..a82c0ba 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -709,7 +709,10 @@
 #define REG_BCN_PSR_RPT			0x06a8
 #define REG_CALB32K_CTRL		0x06ac
 #define REG_PKT_MON_CTRL		0x06b4
-#define REG_BT_COEX_TABLE		0x06c0
+#define REG_BT_COEX_TABLE1		0x06c0
+#define REG_BT_COEX_TABLE2		0x06c4
+#define REG_BT_COEX_TABLE3		0x06c8
+#define REG_BT_COEX_TABLE4		0x06cc
 #define REG_WMAC_RESP_TXINFO		0x06d8
 
 #define REG_MACID1			0x0700
-- 
cgit v0.10.2


From 7297f49c3761826242cd8a742f0f78a139d6653a Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:44 -0500
Subject: rtl8xxxu: Do not ignore wlan activity on 8723bu

The 8723bu is a WiFi/BT combo part. When initializing it for WiFi,
make sure to tell it not to ignore WiFi activity.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 91f189a..c9cbea7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5704,6 +5704,11 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	 */
 	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00);
 
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.ignore_wlan.cmd = H2C_8723B_BT_IGNORE_WLANACT;
+	h2c.ignore_wlan.data = 0;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ignore_wlan));
+
 	/*
 	 * Software control, antenna at WiFi side
 	 */
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index c4c0b28..5a6b223 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -754,6 +754,10 @@ struct h2c_cmd {
 		} __packed bt_wlan_calibration;
 		struct {
 			u8 cmd;
+			u8 data;
+		} __packed ignore_wlan;
+		struct {
+			u8 cmd;
 			u8 ant_inverse;
 			u8 int_switch_type;
 		} __packed ant_sel_rsv;
-- 
cgit v0.10.2


From a228a5db4d19b05da7315320e0b4a87ed7c472aa Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:45 -0500
Subject: rtl8xxxu: Set the right type for ps tdma on 8723bu

Use the correct type when setting PS TDMA for 8723bu. This matches the
vendor driver.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index c9cbea7..0638d46 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5712,7 +5712,7 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	/*
 	 * Software control, antenna at WiFi side
 	 */
-	rtl8723bu_set_ps_tdma(priv, 0x00, 0x00, 0x00, 0x00, 0x00);
+	rtl8723bu_set_ps_tdma(priv, 0x08, 0x00, 0x00, 0x00, 0x00);
 
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE1, 0x55555555);
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE2, 0x5a5a5a5a);
-- 
cgit v0.10.2


From a6c80d211cdf0748a8ae4fd1ab41cf43d3ecbd71 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:46 -0500
Subject: rtl8xxxu: Add rtl8723bu (nextgen) rx descriptor definition

The nextgen chips use a slightly different RX descriptor format. This
adds support for the new format.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 5a6b223..a8a4f2b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -197,6 +197,131 @@ struct rtl8xxxu_rx_desc {
 #endif
 };
 
+struct rtl8723bu_rx_desc {
+#ifdef __LITTLE_ENDIAN
+	u32 pktlen:14;
+	u32 crc32:1;
+	u32 icverr:1;
+	u32 drvinfo_sz:4;
+	u32 security:3;
+	u32 qos:1;
+	u32 shift:2;
+	u32 phy_stats:1;
+	u32 swdec:1;
+	u32 ls:1;
+	u32 fs:1;
+	u32 eor:1;
+	u32 own:1;
+
+	u32 macid:7;
+	u32 dummy1_0:1;
+	u32 tid:4;
+	u32 dummy1_1:1;
+	u32 amsdu:1;
+	u32 rxid_match:1;
+	u32 paggr:1;
+	u32 a1fit:4;	/* 16 */
+	u32 chkerr:1;
+	u32 ipver:1;
+	u32 tcpudp:1;
+	u32 chkvld:1;
+	u32 pam:1;
+	u32 pwr:1;
+	u32 more_data:1;
+	u32 more_frag:1;
+	u32 type:2;
+	u32 mc:1;
+	u32 bc:1;
+
+	u32 seq:12;
+	u32 frag:4;
+	u32 rx_is_qos:1;	/* 16 */
+	u32 dummy2_0:1;
+	u32 wlanhd_iv_len:6;
+	u32 dummy2_1:4;
+	u32 rpt_sel:1;
+	u32 dummy2_2:3;
+
+	u32 rxmcs:7;
+	u32 dummy3_0:3;
+	u32 htc:1;
+	u32 eosp:1;
+	u32 bssidfit:2;
+	u32 dummy3_1:2;
+	u32 usb_agg_pktnum:8;	/* 16 */
+	u32 dummy3_2:5;
+	u32 pattern_match:1;
+	u32 unicast_match:1;
+	u32 magic_match:1;
+
+	u32 splcp:1;
+	u32 ldcp:1;
+	u32 stbc:1;
+	u32 dummy4_0:1;
+	u32 bw:2;
+	u32 dummy4_1:26;
+#else
+	u32 own:1;
+	u32 eor:1;
+	u32 fs:1;
+	u32 ls:1;
+	u32 swdec:1;
+	u32 phy_stats:1;
+	u32 shift:2;
+	u32 qos:1;
+	u32 security:3;
+	u32 drvinfo_sz:4;
+	u32 icverr:1;
+	u32 crc32:1;
+	u32 pktlen:14;
+
+	u32 bc:1;
+	u32 mc:1;
+	u32 type:2;
+	u32 mf:1;
+	u32 md:1;
+	u32 pwr:1;
+	u32 pam:1;
+	u32 a2fit:4;
+	u32 a1fit:4;
+	u32 faggr:1;
+	u32 paggr:1;
+	u32 amsdu:1;
+	u32 hwrsvd:4;
+	u32 tid:4;
+	u32 macid:5;
+
+	u32 dummy2_2:3;
+	u32 rpt_sel:1;
+	u32 dummy2_1:4;
+	u32 wlanhd_iv_len:6;
+	u32 dummy2_0:1;
+	u32 rx_is_qos:1;
+	u32 frag:4;		/* 16 */
+	u32 seq:12;
+
+	u32 magic_match:1;
+	u32 unicast_match:1;
+	u32 pattern_match:1;
+	u32 dummy3_2:5;
+	u32 usb_agg_pktnum:8;
+	u32 dummy3_1:2;		/* 16 */
+	u32 bssidfit:2;
+	u32 eosp:1;
+	u32 htc:1;
+	u32 dummy3_0:3;
+	u32 rxmcs:7;
+
+	u32 dumm4_1:26;
+	u32 bw:2;
+	u32 dummy4_0:1;
+	u32 stbc:1;
+	u32 ldcp:1;
+	u32 splcp:1;
+#endif
+	__le32 tsfl;
+};
+
 struct rtl8xxxu_tx_desc {
 	__le16 pkt_size;
 	u8 pkt_offset;
-- 
cgit v0.10.2


From b18cdfdb67b258c01947e611b6bbb745188eb131 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:47 -0500
Subject: rtl8xxxu: Handle 8723bu style rx descriptors

This adds code to parse the new RX descriptor format used by the
8723bu/8192eu parts.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 0638d46..8e6d54f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6730,6 +6730,91 @@ static void rtl8xxxu_rx_urb_work(struct work_struct *work)
 	}
 }
 
+static int rtl8723au_parse_rx_desc(struct rtl8xxxu_priv *priv,
+				   struct sk_buff *skb,
+				   struct ieee80211_rx_status *rx_status)
+{
+	struct rtl8xxxu_rx_desc *rx_desc = (struct rtl8xxxu_rx_desc *)skb->data;
+	struct rtl8723au_phy_stats *phy_stats;
+	int drvinfo_sz, desc_shift;
+
+	skb_pull(skb, sizeof(struct rtl8xxxu_rx_desc));
+
+	phy_stats = (struct rtl8723au_phy_stats *)skb->data;
+
+	drvinfo_sz = rx_desc->drvinfo_sz * 8;
+	desc_shift = rx_desc->shift;
+	skb_pull(skb, drvinfo_sz + desc_shift);
+
+	if (rx_desc->phy_stats)
+		rtl8xxxu_rx_parse_phystats(priv, rx_status, rx_desc, phy_stats);
+
+	rx_status->mactime = le32_to_cpu(rx_desc->tsfl);
+	rx_status->flag |= RX_FLAG_MACTIME_START;
+
+	if (!rx_desc->swdec)
+		rx_status->flag |= RX_FLAG_DECRYPTED;
+	if (rx_desc->crc32)
+		rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
+	if (rx_desc->bw)
+		rx_status->flag |= RX_FLAG_40MHZ;
+
+	if (rx_desc->rxht) {
+		rx_status->flag |= RX_FLAG_HT;
+		rx_status->rate_idx = rx_desc->rxmcs - DESC_RATE_MCS0;
+	} else {
+		rx_status->rate_idx = rx_desc->rxmcs;
+	}
+
+	return RX_TYPE_DATA_PKT;
+}
+
+static int rtl8723bu_parse_rx_desc(struct rtl8xxxu_priv *priv,
+				   struct sk_buff *skb,
+				   struct ieee80211_rx_status *rx_status)
+{
+	struct rtl8723bu_rx_desc *rx_desc =
+		(struct rtl8723bu_rx_desc *)skb->data;
+	struct rtl8723au_phy_stats *phy_stats;
+	int drvinfo_sz, desc_shift;
+	int rx_type;
+
+	skb_pull(skb, sizeof(struct rtl8723bu_rx_desc));
+
+	phy_stats = (struct rtl8723au_phy_stats *)skb->data;
+
+	drvinfo_sz = rx_desc->drvinfo_sz * 8;
+	desc_shift = rx_desc->shift;
+	skb_pull(skb, drvinfo_sz + desc_shift);
+
+	rx_status->mactime = le32_to_cpu(rx_desc->tsfl);
+	rx_status->flag |= RX_FLAG_MACTIME_START;
+
+	if (!rx_desc->swdec)
+		rx_status->flag |= RX_FLAG_DECRYPTED;
+	if (rx_desc->crc32)
+		rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
+	if (rx_desc->bw)
+		rx_status->flag |= RX_FLAG_40MHZ;
+
+	if (rx_desc->rxmcs >= DESC_RATE_MCS0) {
+		rx_status->flag |= RX_FLAG_HT;
+		rx_status->rate_idx = rx_desc->rxmcs - DESC_RATE_MCS0;
+	} else {
+		rx_status->rate_idx = rx_desc->rxmcs;
+	}
+
+	if (rx_desc->rpt_sel) {
+		struct device *dev = &priv->udev->dev;
+		dev_dbg(dev, "%s: C2H packet\n", __func__);
+		rx_type = RX_TYPE_C2H;
+	} else {
+		rx_type = RX_TYPE_DATA_PKT;
+	}
+
+	return rx_type;
+}
+
 static void rtl8xxxu_rx_complete(struct urb *urb)
 {
 	struct rtl8xxxu_rx_urb *rx_urb =
@@ -6737,54 +6822,30 @@ static void rtl8xxxu_rx_complete(struct urb *urb)
 	struct ieee80211_hw *hw = rx_urb->hw;
 	struct rtl8xxxu_priv *priv = hw->priv;
 	struct sk_buff *skb = (struct sk_buff *)urb->context;
-	struct rtl8xxxu_rx_desc *rx_desc = (struct rtl8xxxu_rx_desc *)skb->data;
-	struct rtl8723au_phy_stats *phy_stats;
 	struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb);
 	struct device *dev = &priv->udev->dev;
 	__le32 *_rx_desc_le = (__le32 *)skb->data;
 	u32 *_rx_desc = (u32 *)skb->data;
-	int drvinfo_sz, desc_shift, i;
+	int rx_type, i;
 
 	for (i = 0; i < (sizeof(struct rtl8xxxu_rx_desc) / sizeof(u32)); i++)
 		_rx_desc[i] = le32_to_cpu(_rx_desc_le[i]);
 
-	drvinfo_sz = rx_desc->drvinfo_sz * 8;
-	desc_shift = rx_desc->shift;
 	skb_put(skb, urb->actual_length);
 
 	if (urb->status == 0) {
-		skb_pull(skb, sizeof(struct rtl8xxxu_rx_desc));
-		phy_stats = (struct rtl8723au_phy_stats *)skb->data;
-
-		skb_pull(skb, drvinfo_sz + desc_shift);
-
 		memset(rx_status, 0, sizeof(struct ieee80211_rx_status));
 
-		if (rx_desc->phy_stats)
-			rtl8xxxu_rx_parse_phystats(priv, rx_status,
-						   rx_desc, phy_stats);
+		rx_type = priv->fops->parse_rx_desc(priv, skb, rx_status);
 
 		rx_status->freq = hw->conf.chandef.chan->center_freq;
 		rx_status->band = hw->conf.chandef.chan->band;
 
-		rx_status->mactime = le32_to_cpu(rx_desc->tsfl);
-		rx_status->flag |= RX_FLAG_MACTIME_START;
-
-		if (!rx_desc->swdec)
-			rx_status->flag |= RX_FLAG_DECRYPTED;
-		if (rx_desc->crc32)
-			rx_status->flag |= RX_FLAG_FAILED_FCS_CRC;
-		if (rx_desc->bw)
-			rx_status->flag |= RX_FLAG_40MHZ;
-
-		if (rx_desc->rxht) {
-			rx_status->flag |= RX_FLAG_HT;
-			rx_status->rate_idx = rx_desc->rxmcs - DESC_RATE_MCS0;
-		} else {
-			rx_status->rate_idx = rx_desc->rxmcs;
-		}
+		if (rx_type == RX_TYPE_DATA_PKT)
+			ieee80211_rx_irqsafe(hw, skb);
+		else
+			dev_kfree_skb(skb);
 
-		ieee80211_rx_irqsafe(hw, skb);
 		skb = NULL;
 		rx_urb->urb.context = NULL;
 		rtl8xxxu_queue_rx_urb(priv, rx_urb);
@@ -7564,6 +7625,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
+	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -7582,6 +7644,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
 	.init_bt = rtl8723bu_init_bt,
+	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -7601,6 +7664,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
+	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -7619,6 +7683,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
+	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index a8a4f2b..4a3afe3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -61,6 +61,12 @@
 #define EFUSE_BT_MAP_LEN_8723A		1024
 #define EFUSE_MAX_WORD_UNIT		4
 
+enum rtl8xxxu_rx_type {
+	RX_TYPE_DATA_PKT = 0,
+	RX_TYPE_C2H = 1,
+	RX_TYPE_ERROR = -1
+};
+
 struct rtl8xxxu_rx_desc {
 #ifdef __LITTLE_ENDIAN
 	u32 pktlen:14;
@@ -1010,6 +1016,8 @@ struct rtl8xxxu_fileops {
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
 	void (*config_channel) (struct ieee80211_hw *hw);
 	void (*init_bt) (struct rtl8xxxu_priv *priv);
+	int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb,
+			      struct ieee80211_rx_status *rx_status);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
-- 
cgit v0.10.2


From b2b43b7837ba2e097bb019ee488ca5f4e0056fa4 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:48 -0500
Subject: rtl8xxxu: Initial functionality to handle C2H events for 8723bu

The 64 bit mailbox commands also provide a different method for
mailbox command responses (C2H events).

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 8e6d54f..1f28093 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6815,6 +6815,36 @@ static int rtl8723bu_parse_rx_desc(struct rtl8xxxu_priv *priv,
 	return rx_type;
 }
 
+static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
+				 struct sk_buff *skb)
+{
+	struct rtl8723bu_c2h *c2h = (struct rtl8723bu_c2h *)skb->data;
+	struct device *dev = &priv->udev->dev;
+	int len;
+
+	len = skb->len - 2;
+
+	pr_info("%s: C2H ID %02x seq %02x, len %02x %02x\n", __func__,
+		c2h->id, c2h->seq, len, c2h->bt_info.response_source);
+
+	switch(c2h->id) {
+	case C2H_8723B_BT_INFO:
+		if (c2h->bt_info.response_source >
+		    BT_INFO_SRC_8723B_BT_ACTIVE_SEND)
+			dev_info(dev, "C2H_BT_INFO WiFi only firmware\n");
+		else
+			dev_info(dev, "C2H_BT_INFO BT/WiFi coexist firmware\n");
+
+		if (c2h->bt_info.bt_has_reset)
+			dev_info(dev, "BT has been reset\n");
+
+		break;
+	default:
+		pr_info("%s: Unhandled C2H event %02x\n", __func__, c2h->id);
+		break;
+	}
+}
+
 static void rtl8xxxu_rx_complete(struct urb *urb)
 {
 	struct rtl8xxxu_rx_urb *rx_urb =
@@ -6843,8 +6873,10 @@ static void rtl8xxxu_rx_complete(struct urb *urb)
 
 		if (rx_type == RX_TYPE_DATA_PKT)
 			ieee80211_rx_irqsafe(hw, skb);
-		else
+		else {
+			rtl8723bu_handle_c2h(priv, skb);
 			dev_kfree_skb(skb);
+		}
 
 		skb = NULL;
 		rx_urb->urb.context = NULL;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 4a3afe3..e4d5c4b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -899,6 +899,60 @@ struct h2c_cmd {
 	};
 };
 
+enum c2h_evt_8723b {
+	C2H_8723B_DEBUG = 0,
+	C2H_8723B_TSF = 1,
+	C2H_8723B_AP_RPT_RSP = 2,
+	C2H_8723B_CCX_TX_RPT = 3,
+	C2H_8723B_BT_RSSI = 4,
+	C2H_8723B_BT_OP_MODE = 5,
+	C2H_8723B_EXT_RA_RPT = 6,
+	C2H_8723B_BT_INFO = 9,
+	C2H_8723B_HW_INFO_EXCH = 10,
+	C2H_8723B_BT_MP_INFO = 11,
+	C2H_8723B_FW_DEBUG = 0xff,
+};
+
+enum bt_info_src_8723b {
+	BT_INFO_SRC_8723B_WIFI_FW = 0x0,
+        BT_INFO_SRC_8723B_BT_RSP = 0x1,
+        BT_INFO_SRC_8723B_BT_ACTIVE_SEND = 0x2,
+};
+
+struct rtl8723bu_c2h {
+	u8 id;
+	u8 seq;
+	union {
+		struct {
+			u8 payload[0];
+		} __packed raw;
+		struct {
+			u8 response_source:4;
+			u8 dummy0_0:4;
+
+			u8 bt_info;
+
+			u8 retry_count:4;
+			u8 dummy2_0:1;
+			u8 bt_page:1;
+			u8 tx_rx_mask:1;
+			u8 dummy2_2:1;
+
+			u8 rssi;
+
+			u8 basic_rate:1;
+			u8 bt_has_reset:1;
+			u8 dummy4_1:1;;
+			u8 ignore_wlan:1;
+			u8 auto_report:1;
+			u8 dummy4_2:3;
+
+			u8 a4;
+			u8 a5;
+		} __packed bt_info;
+	};
+};
+
 struct rtl8xxxu_fileops;
 
 struct rtl8xxxu_priv {
-- 
cgit v0.10.2


From 394f1bd314cd08915fee9d5b36a2105dc1a543b7 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:49 -0500
Subject: rtl8xxxu: Handle BT register writes and MP_OPER events

8723bu BT registers are written via the mailbox interface. Add support
for writing these and corresponding C2H event responses.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 1f28093..ec03e91 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1531,6 +1531,27 @@ error:
 	return retval;
 }
 
+static void rtl8723bu_write_btreg(struct rtl8xxxu_priv *priv, u8 reg, u8 data)
+{
+	struct h2c_cmd h2c;
+	int reqnum = 0;
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.bt_mp_oper.cmd = H2C_8723B_BT_MP_OPER;
+	h2c.bt_mp_oper.operreq = 0 | (reqnum << 4);
+	h2c.bt_mp_oper.opcode = BT_MP_OP_WRITE_REG_VALUE;
+	h2c.bt_mp_oper.data = data;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_mp_oper));
+
+	reqnum++;
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.bt_mp_oper.cmd = H2C_8723B_BT_MP_OPER;
+	h2c.bt_mp_oper.operreq = 0 | (reqnum << 4);
+	h2c.bt_mp_oper.opcode = BT_MP_OP_WRITE_REG_VALUE;
+	h2c.bt_mp_oper.addr = reg;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_mp_oper));
+}
+
 static void rtl8723a_enable_rf(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
@@ -5646,6 +5667,8 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 
 	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_IQADJ_G1, 0x780);
 
+	rtl8723bu_write_btreg(priv, 0x3c, 0x15); /* BT TRx Mask on */
+
 	/*
 	 * Set BT grant to low
 	 */
@@ -6824,8 +6847,9 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
 
 	len = skb->len - 2;
 
-	pr_info("%s: C2H ID %02x seq %02x, len %02x %02x\n", __func__,
-		c2h->id, c2h->seq, len, c2h->bt_info.response_source);
+	dev_info(dev, "%s: C2H ID %02x seq %02x, len %02x source %02x\n",
+		 __func__,
+		 c2h->id, c2h->seq, len, c2h->bt_info.response_source);
 
 	switch(c2h->id) {
 	case C2H_8723B_BT_INFO:
@@ -6837,8 +6861,14 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
 
 		if (c2h->bt_info.bt_has_reset)
 			dev_info(dev, "BT has been reset\n");
+		if (c2h->bt_info.tx_rx_mask)
+			dev_info(dev, "BT TRx mask\n");
 
 		break;
+	case C2H_8723B_BT_MP_INFO:
+		dev_info(dev, "C2H_MP_INFO ext ID %02x, status %02x\n",
+			 c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status);
+		break;
 	default:
 		pr_info("%s: Unhandled C2H event %02x\n", __func__, c2h->id);
 		break;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index e4d5c4b..3e4e355 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -881,6 +881,13 @@ struct h2c_cmd {
 		} __packed b_type_dma;
 		struct {
 			u8 cmd;
+			u8 operreq;
+			u8 opcode;
+			u8 data;
+			u8 addr;
+		} __packed bt_mp_oper;
+		struct {
+			u8 cmd;
 			u8 data;
 		} __packed bt_wlan_calibration;
 		struct {
@@ -908,8 +915,8 @@ enum c2h_evt_8723b {
 	C2H_8723B_BT_OP_MODE = 5,
 	C2H_8723B_EXT_RA_RPT = 6,
 	C2H_8723B_BT_INFO = 9,
-	C2H_8723B_HW_INFO_EXCH = 10,
-	C2H_8723B_BT_MP_INFO = 11,
+	C2H_8723B_HW_INFO_EXCH = 0x0a,
+	C2H_8723B_BT_MP_INFO = 0x0b,
 	C2H_8723B_FW_DEBUG = 0xff,
 };
 
@@ -919,6 +926,46 @@ enum bt_info_src_8723b {
         BT_INFO_SRC_8723B_BT_ACTIVE_SEND = 0x2,
 };
 
+enum bt_mp_oper_opcode_8723b {
+	BT_MP_OP_GET_BT_VERSION	= 0x00,
+	BT_MP_OP_RESET = 0x01,
+	BT_MP_OP_TEST_CTRL = 0x02,
+	BT_MP_OP_SET_BT_MODE = 0x03,
+	BT_MP_OP_SET_CHNL_TX_GAIN = 0x04,
+	BT_MP_OP_SET_PKT_TYPE_LEN = 0x05,
+	BT_MP_OP_SET_PKT_CNT_L_PL_TYPE = 0x06,
+	BT_MP_OP_SET_PKT_CNT_H_PKT_INTV = 0x07,
+	BT_MP_OP_SET_PKT_HEADER = 0x08,
+	BT_MP_OP_SET_WHITENCOEFF = 0x09,
+	BT_MP_OP_SET_BD_ADDR_L = 0x0a,
+	BT_MP_OP_SET_BD_ADDR_H = 0x0b,
+	BT_MP_OP_WRITE_REG_ADDR = 0x0c,
+	BT_MP_OP_WRITE_REG_VALUE = 0x0d,
+	BT_MP_OP_GET_BT_STATUS = 0x0e,
+	BT_MP_OP_GET_BD_ADDR_L = 0x0f,
+	BT_MP_OP_GET_BD_ADDR_H = 0x10,
+	BT_MP_OP_READ_REG = 0x11,
+	BT_MP_OP_SET_TARGET_BD_ADDR_L = 0x12,
+	BT_MP_OP_SET_TARGET_BD_ADDR_H = 0x13,
+	BT_MP_OP_SET_TX_POWER_CALIBRATION = 0x14,
+	BT_MP_OP_GET_RX_PKT_CNT_L = 0x15,
+	BT_MP_OP_GET_RX_PKT_CNT_H = 0x16,
+	BT_MP_OP_GET_RX_ERROR_BITS_L = 0x17,
+	BT_MP_OP_GET_RX_ERROR_BITS_H = 0x18,
+	BT_MP_OP_GET_RSSI = 0x19,
+	BT_MP_OP_GET_CFO_HDR_QUALITY_L = 0x1a,
+	BT_MP_OP_GET_CFO_HDR_QUALITY_H = 0x1b,
+	BT_MP_OP_GET_TARGET_BD_ADDR_L = 0x1c,
+	BT_MP_OP_GET_TARGET_BD_ADDR_H = 0x1d,
+	BT_MP_OP_GET_AFH_MAP_L = 0x1e,
+	BT_MP_OP_GET_AFH_MAP_M = 0x1f,
+	BT_MP_OP_GET_AFH_MAP_H = 0x20,
+	BT_MP_OP_GET_AFH_STATUS = 0x21,
+	BT_MP_OP_SET_TRACKING_INTERVAL = 0x22,
+	BT_MP_OP_SET_THERMAL_METER = 0x23,
+	BT_MP_OP_ENABLE_CFO_TRACKING = 0x24,
+};
+
 struct rtl8723bu_c2h {
 	u8 id;
 	u8 seq;
@@ -927,6 +974,14 @@ struct rtl8723bu_c2h {
 			u8 payload[0];
 		} __packed raw;
 		struct {
+			u8 ext_id;
+			u8 status:4;
+			u8 retlen:4;
+			u8 opcode_ver:4;
+			u8 req_num:4;
+			u8 payload[2];
+		} __packed bt_mp_info;
+		struct {
 			u8 response_source:4;
 			u8 dummy0_0:4;
 
-- 
cgit v0.10.2


From 6b9eae0129f43b8e691cec91a99601b0829fb841 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:50 -0500
Subject: rtl8xxxu: Issue BT_INFO command

Issue a BT_INFO command to verify the status of BT/WiFi settings.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ec03e91..9b5b768 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5727,20 +5727,25 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	 */
 	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00);
 
-	memset(&h2c, 0, sizeof(struct h2c_cmd));
-	h2c.ignore_wlan.cmd = H2C_8723B_BT_IGNORE_WLANACT;
-	h2c.ignore_wlan.data = 0;
-	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ignore_wlan));
-
 	/*
 	 * Software control, antenna at WiFi side
 	 */
 	rtl8723bu_set_ps_tdma(priv, 0x08, 0x00, 0x00, 0x00, 0x00);
 
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.bt_info.cmd = H2C_8723B_BT_INFO;
+	h2c.bt_info.data = BIT(0);
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_info));
+
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE1, 0x55555555);
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE2, 0x5a5a5a5a);
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE3, 0x00ffffff);
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE4, 0x00000003);
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+	h2c.ignore_wlan.cmd = H2C_8723B_BT_IGNORE_WLANACT;
+	h2c.ignore_wlan.data = 0;
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ignore_wlan));
 }
 
 static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
@@ -6847,8 +6852,7 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
 
 	len = skb->len - 2;
 
-	dev_info(dev, "%s: C2H ID %02x seq %02x, len %02x source %02x\n",
-		 __func__,
+	dev_info(dev, "C2H ID %02x seq %02x, len %02x source %02x\n",
 		 c2h->id, c2h->seq, len, c2h->bt_info.response_source);
 
 	switch(c2h->id) {
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 3e4e355..ea0b28a 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -881,6 +881,10 @@ struct h2c_cmd {
 		} __packed b_type_dma;
 		struct {
 			u8 cmd;
+			u8 data;
+		} __packed bt_info;
+		struct {
+			u8 cmd;
 			u8 operreq;
 			u8 opcode;
 			u8 data;
-- 
cgit v0.10.2


From 04313eb4f2225cb5dba2fffdb0f5f1abd71e611d Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:51 -0500
Subject: rtl8xxxu: Do not set REG_AFE_XTAL_CTRL on 8723bu

The 8723bu does not like REG_AFE_XTAL_CTRL being set, so skip this for
now, to match the vendor driver.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 9b5b768..a41f7f0 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2998,6 +2998,7 @@ static int rtl8xxxu_init_phy_regs(struct rtl8xxxu_priv *priv,
 static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 {
 	u8 val8, ldoa15, ldov12d, lpldo, ldohci12;
+	u16 val16;
 	u32 val32;
 
 	/*
@@ -3018,16 +3019,18 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 		udelay(2);
 	}
 
-	val8 = rtl8xxxu_read8(priv, REG_SYS_FUNC);
-	val8 |= SYS_FUNC_BB_GLB_RSTN | SYS_FUNC_BBRSTB;
-	rtl8xxxu_write8(priv, REG_SYS_FUNC, val8);
+	val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+	val16 |= SYS_FUNC_BB_GLB_RSTN | SYS_FUNC_BBRSTB;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
 
-	/* AFE_XTAL_RF_GATE (bit 14) if addressing as 32 bit register */
-	val32 = rtl8xxxu_read32(priv, REG_AFE_XTAL_CTRL);
-	val32 &= ~AFE_XTAL_RF_GATE;
-	if (priv->has_bluetooth)
-		val32 &= ~AFE_XTAL_BT_GATE;
-	rtl8xxxu_write32(priv, REG_AFE_XTAL_CTRL, val32);
+	if (priv->rtlchip != 0x8723b) {
+		/* AFE_XTAL_RF_GATE (bit 14) if addressing as 32 bit register */
+		val32 = rtl8xxxu_read32(priv, REG_AFE_XTAL_CTRL);
+		val32 &= ~AFE_XTAL_RF_GATE;
+		if (priv->has_bluetooth)
+			val32 &= ~AFE_XTAL_BT_GATE;
+		rtl8xxxu_write32(priv, REG_AFE_XTAL_CTRL, val32);
+	}
 
 	/* 6. 0x1f[7:0] = 0x07 */
 	val8 = RF_ENABLE | RF_RSTB | RF_SDMRSTB;
-- 
cgit v0.10.2


From 42836db1f609b66fdb1b307243fa01451748ab68 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:52 -0500
Subject: rtl8xxxu: Implement 8723bu power on sequence

This implements the 8723bu specific power on sequence as it is
different from that of the 8723au chips.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index a41f7f0..1f30b3b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2933,10 +2933,6 @@ static void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv)
 	val32 &= 0xffffff00;
 	val32 |= 0x77;
 	rtl8xxxu_write32(priv, 0x0930, val32);
-
-	val32 = rtl8xxxu_read32(priv, REG_PWR_DATA);
-	val32 |= BIT(11);
-	rtl8xxxu_write32(priv, REG_PWR_DATA, val32);
 }
 
 static int
@@ -5359,6 +5355,127 @@ exit:
 	return ret;
 }
 
+static int rtl8723b_emu_to_active(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u32 val32;
+	int count, ret = 0;
+
+	/* 0x20[0] = 1 enable LDOA12 MACRO block for all interface */
+	val8 = rtl8xxxu_read8(priv, REG_LDOA15_CTRL);
+	val8 |= LDOA15_ENABLE;
+	rtl8xxxu_write8(priv, REG_LDOA15_CTRL, val8);
+
+	/* 0x67[0] = 0 to disable BT_GPS_SEL pins*/
+	val8 = rtl8xxxu_read8(priv, 0x0067);
+	val8 &= ~BIT(4);
+	rtl8xxxu_write8(priv, 0x0067, val8);
+
+	mdelay(1);
+
+	/* 0x00[5] = 0 release analog Ips to digital, 1:isolation */
+	val8 = rtl8xxxu_read8(priv, REG_SYS_ISO_CTRL);
+	val8 &= ~SYS_ISO_ANALOG_IPS;
+	rtl8xxxu_write8(priv, REG_SYS_ISO_CTRL, val8);
+
+	/* Disable SW LPS 0x04[10]= 0 */
+	val32 = rtl8xxxu_read8(priv, REG_APS_FSMCO);
+	val32 &= ~APS_FSMCO_SW_LPS;
+	rtl8xxxu_write32(priv, REG_APS_FSMCO, val32);
+
+	/* Wait until 0x04[17] = 1 power ready */
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+		if (val32 & BIT(17))
+			break;
+
+		udelay(10);
+	}
+
+	if (!count) {
+		ret = -EBUSY;
+		goto exit;
+	}
+
+	/* We should be able to optimize the following three entries into one */
+
+	/* Release WLON reset 0x04[16]= 1*/
+	val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+	val32 |= APS_FSMCO_WLON_RESET;
+	rtl8xxxu_write32(priv, REG_APS_FSMCO, val32);
+
+	/* Disable HWPDN 0x04[15]= 0*/
+	val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+	val32 &= ~APS_FSMCO_HW_POWERDOWN;
+	rtl8xxxu_write32(priv, REG_APS_FSMCO, val32);
+
+	/* Disable WL suspend*/
+	val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+	val32 &= ~(APS_FSMCO_HW_SUSPEND | APS_FSMCO_PCIE);
+	rtl8xxxu_write32(priv, REG_APS_FSMCO, val32);
+
+	/* Set, then poll until 0 */
+	val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+	val32 |= APS_FSMCO_MAC_ENABLE;
+	rtl8xxxu_write32(priv, REG_APS_FSMCO, val32);
+
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val32 = rtl8xxxu_read32(priv, REG_APS_FSMCO);
+		if ((val32 & APS_FSMCO_MAC_ENABLE) == 0) {
+			ret = 0;
+			break;
+		}
+		udelay(10);
+	}
+
+	if (!count) {
+		ret = -EBUSY;
+		goto exit;
+	}
+
+	/* Enable WL control XTAL setting */
+	val8 = rtl8xxxu_read8(priv, REG_AFE_MISC);
+	val8 |= AFE_MISC_WL_XTAL_CTRL;
+	rtl8xxxu_write8(priv, REG_AFE_MISC, val8);
+
+	/* Enable falling edge triggering interrupt */
+	val8 = rtl8xxxu_read8(priv, REG_GPIO_INTM + 1);
+	val8 |= BIT(1);
+	rtl8xxxu_write8(priv, REG_GPIO_INTM + 1, val8);
+
+	/* Enable GPIO9 interrupt mode */
+	val8 = rtl8xxxu_read8(priv, REG_GPIO_IO_SEL_2 + 1);
+	val8 |= BIT(1);
+	rtl8xxxu_write8(priv, REG_GPIO_IO_SEL_2 + 1, val8);
+
+	/* Enable GPIO9 input mode */
+	val8 = rtl8xxxu_read8(priv, REG_GPIO_IO_SEL_2);
+	val8 &= ~BIT(1);
+	rtl8xxxu_write8(priv, REG_GPIO_IO_SEL_2, val8);
+
+	/* Enable HSISR GPIO[C:0] interrupt */
+	val8 = rtl8xxxu_read8(priv, REG_HSIMR);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, REG_HSIMR, val8);
+
+	/* Enable HSISR GPIO9 interrupt */
+	val8 = rtl8xxxu_read8(priv, REG_HSIMR + 2);
+	val8 |= BIT(1);
+	rtl8xxxu_write8(priv, REG_HSIMR + 2, val8);
+
+	val8 = rtl8xxxu_read8(priv, REG_MULTI_FUNC_CTRL);
+	val8 |= MULTI_WIFI_HW_ROF_EN;
+	rtl8xxxu_write8(priv, REG_MULTI_FUNC_CTRL, val8);
+
+	/* For GPIO9 internal pull high setting BIT(14) */
+	val8 = rtl8xxxu_read8(priv, REG_MULTI_FUNC_CTRL + 1);
+	val8 |= BIT(6);
+	rtl8xxxu_write8(priv, REG_MULTI_FUNC_CTRL + 1, val8);
+
+exit:
+	return ret;
+}
+
 static int rtl8xxxu_emu_to_disabled(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
@@ -5430,6 +5547,62 @@ exit:
 	return ret;
 }
 
+static int rtl8723bu_power_on(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u16 val16;
+	u32 val32;
+	int ret;
+
+	rtl8723a_disabled_to_emu(priv);
+
+	ret = rtl8723b_emu_to_active(priv);
+	if (ret)
+		goto exit;
+
+	/*
+	 * Enable MAC DMA/WMAC/SCHEDULE/SEC block
+	 * Set CR bit10 to enable 32k calibration.
+	 */
+	val16 = rtl8xxxu_read16(priv, REG_CR);
+	val16 |= (CR_HCI_TXDMA_ENABLE | CR_HCI_RXDMA_ENABLE |
+		  CR_TXDMA_ENABLE | CR_RXDMA_ENABLE |
+		  CR_PROTOCOL_ENABLE | CR_SCHEDULE_ENABLE |
+		  CR_MAC_TX_ENABLE | CR_MAC_RX_ENABLE |
+		  CR_SECURITY_ENABLE | CR_CALTIMER_ENABLE);
+	rtl8xxxu_write16(priv, REG_CR, val16);
+
+	/*
+	 * BT coexist power on settings. This is identical for 1 and 2
+	 * antenna parts.
+	 */
+	rtl8xxxu_write8(priv, REG_PAD_CTRL1 + 3, 0x20);
+
+	val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+	val16 |= SYS_FUNC_BBRSTB | SYS_FUNC_BB_GLB_RSTN;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+
+	rtl8xxxu_write8(priv, REG_BT_CONTROL_8723BU + 1, 0x18);
+	rtl8xxxu_write8(priv, REG_WLAN_ACT_CONTROL_8723B, 0x04);
+	rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00);
+	/* Antenna inverse */
+	rtl8xxxu_write8(priv, 0xfe08, 0x01);
+
+	val16 = rtl8xxxu_read16(priv, REG_PWR_DATA);
+	val16 |= PWR_DATA_EEPRPAD_RFE_CTRL_EN;
+	rtl8xxxu_write16(priv, REG_PWR_DATA, val16);
+
+	val32 = rtl8xxxu_read32(priv, REG_LEDCFG0);
+	val32 |= LEDCFG0_DPDT_SELECT;
+	rtl8xxxu_write32(priv, REG_LEDCFG0, val32);
+
+	val8 = rtl8xxxu_read8(priv, REG_PAD_CTRL1);
+	val8 &= ~PAD_CTRL1_SW_DPDT_SEL_DATA;
+	rtl8xxxu_write8(priv, REG_PAD_CTRL1, val8);
+exit:
+	return ret;
+}
+
 #ifdef CONFIG_RTL8XXXU_UNTESTED
 
 static int rtl8192cu_power_on(struct rtl8xxxu_priv *priv)
@@ -7707,7 +7880,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.parse_efuse = rtl8723bu_parse_efuse,
 	.load_firmware = rtl8723bu_load_firmware,
-	.power_on = rtl8723au_power_on,
+	.power_on = rtl8723bu_power_on,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index a82c0ba..5250388 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -70,6 +70,8 @@
 
 #define REG_EE_VPD			0x000c
 #define REG_AFE_MISC			0x0010
+#define  AFE_MISC_WL_XTAL_CTRL		BIT(6)
+
 #define REG_SPS0_CTRL			0x0011
 #define REG_SPS_OCP_CFG			0x0018
 #define REG_8192E_LDOV12_CTRL		0x0014
@@ -133,6 +135,8 @@
 #define  EFUSE_ACCESS_DISABLE		0x00	/* RTL8723 only */
 
 #define REG_PWR_DATA			0x0038
+#define PWR_DATA_EEPRPAD_RFE_CTRL_EN	BIT(11)
+
 #define REG_CAL_TIMER			0x003c
 #define REG_ACLK_MON			0x003e
 #define REG_GPIO_MUXCFG			0x0040
@@ -140,7 +144,10 @@
 #define REG_MAC_PINMUX_CFG		0x0043
 #define REG_GPIO_PIN_CTRL		0x0044
 #define REG_GPIO_INTM			0x0048
+#define  GPIO_INTM_EDGE_TRIG_IRQ	BIT(9)
+
 #define REG_LEDCFG0			0x004c
+#define  LEDCFG0_DPDT_SELECT		BIT(23)
 #define REG_LEDCFG1			0x004d
 #define REG_LEDCFG2			0x004e
 #define  LEDCFG2_DPDT_SELECT		BIT(7)
@@ -154,9 +161,13 @@
 #define REG_GPIO_PIN_CTRL_2		0x0060
 /*  RTL8723 WIFI/BT/GPS Multi-Function GPIO Select. */
 #define REG_GPIO_IO_SEL_2		0x0062
+#define  GPIO_IO_SEL_2_GPIO09_INPUT	BIT(1)
+#define  GPIO_IO_SEL_2_GPIO09_IRQ	BIT(9)
 
 /*  RTL8723B */
 #define REG_PAD_CTRL1			0x0064
+#define  PAD_CTRL1_SW_DPDT_SEL_DATA	BIT(0)
+
 /*  RTL8723 only WIFI/BT/GPS Multi-Function control source. */
 #define REG_MULTI_FUNC_CTRL		0x0068
 
-- 
cgit v0.10.2


From 79fb5fe9edb78bd28969dd0f6bacc8c12550c4cc Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:53 -0500
Subject: rtl8xxxu: Setup LLT before downloading firmware

This matches the order of the 8723bu vendor driver

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 1f30b3b..e836382 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5955,6 +5955,26 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	dev_dbg(dev, "%s: macpower %i\n", __func__, macpower);
 	if (!macpower) {
+		ret = priv->fops->llt_init(priv, TX_TOTAL_PAGE_NUM);
+		if (ret) {
+			dev_warn(dev, "%s: LLT table init failed\n", __func__);
+			goto exit;
+		}
+
+		/*
+		 * Presumably this is for 8188EU as well
+		 * Enable TX report and TX report timer
+		 */
+		if (priv->rtlchip == 0x8723bu) {
+			val8 = rtl8xxxu_read8(priv, REG_TX_REPORT_CTRL);
+			val8 |= BIT(1);
+			rtl8xxxu_write8(priv, REG_TX_REPORT_CTRL, val8);
+			/* Set MAX RPT MACID */
+			rtl8xxxu_write8(priv, REG_TX_REPORT_CTRL + 1, 0x02);
+			/* TX report Timer. Unit: 32us */
+			rtl8xxxu_write16(priv, REG_TX_REPORT_TIME, 0xcdf0);
+		}
+
 		if (priv->ep_tx_normal_queue)
 			val8 = TX_PAGE_NUM_NORM_PQ;
 		else
@@ -5996,15 +6016,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (ret)
 		goto exit;
 
-	dev_dbg(dev, "%s: macpower %i\n", __func__, macpower);
-	if (!macpower) {
-		ret = priv->fops->llt_init(priv, TX_TOTAL_PAGE_NUM);
-		if (ret) {
-			dev_warn(dev, "%s: LLT table init failed\n", __func__);
-			goto exit;
-		}
-	}
-
 	/* Fix USB interface interference issue */
 	if (priv->rtlchip == 0x8723a) {
 		rtl8xxxu_write8(priv, 0xfe40, 0xe0);
-- 
cgit v0.10.2


From 8baf670b8928893bacf95a67c86ec2d6c42263f2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:54 -0500
Subject: rtl8xxxu: Additional fixes for 8723bu

Additional tweaks to further map the init sequence for the 8723bu to
that of the vendor driver.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e836382..e879fe2 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2957,7 +2957,8 @@ rtl8xxxu_init_mac(struct rtl8xxxu_priv *priv, struct rtl8xxxu_reg8val *array)
 		}
 	}
 
-	rtl8xxxu_write8(priv, REG_MAX_AGGR_NUM, 0x0a);
+	if (priv->rtlchip != 0x8723b)
+		rtl8xxxu_write8(priv, REG_MAX_AGGR_NUM, 0x0a);
 
 	return 0;
 }
@@ -3003,6 +3004,11 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 	 */
 
 	if (priv->rtlchip == 0x8723b) {
+		val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+		val16 |= SYS_FUNC_BB_GLB_RSTN | SYS_FUNC_BBRSTB |
+			SYS_FUNC_DIO_RF;
+		rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+
 		rtl8xxxu_write32(priv, REG_S0S1_PATH_SWITCH, 0x00);
 	} else {
 		val8 = rtl8xxxu_read8(priv, REG_AFE_PLL_CTRL);
@@ -3013,11 +3019,11 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 
 		rtl8xxxu_write8(priv, REG_AFE_PLL_CTRL + 1, 0xff);
 		udelay(2);
-	}
 
-	val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
-	val16 |= SYS_FUNC_BB_GLB_RSTN | SYS_FUNC_BBRSTB;
-	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+		val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+		val16 |= SYS_FUNC_BB_GLB_RSTN | SYS_FUNC_BBRSTB;
+		rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+	}
 
 	if (priv->rtlchip != 0x8723b) {
 		/* AFE_XTAL_RF_GATE (bit 14) if addressing as 32 bit register */
@@ -3036,9 +3042,14 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 		rtl8xxxu_init_phy_regs(priv, rtl8188ru_phy_1t_highpa_table);
 	else if (priv->tx_paths == 2)
 		rtl8xxxu_init_phy_regs(priv, rtl8192cu_phy_2t_init_table);
-	else if (priv->rtlchip == 0x8723b)
+	else if (priv->rtlchip == 0x8723b) {
+		/*
+		 * Why?
+		 */
+		rtl8xxxu_write8(priv, REG_SYS_FUNC, 0xe3);
+		rtl8xxxu_write8(priv, REG_AFE_XTAL_CTRL + 1, 0x80);
 		rtl8xxxu_init_phy_regs(priv, rtl8723b_phy_1t_init_table);
-	else
+	} else
 		rtl8xxxu_init_phy_regs(priv, rtl8723a_phy_1t_init_table);
 
 
-- 
cgit v0.10.2


From 4ef22eb933c21f307413f9fc402f83e7384cf726 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:55 -0500
Subject: rtl8xxxu: Handle XTAL_K value in efuse specific location

Retrieve the XTAL_K value in the parse_efuse() functions as it's
location various on a per device basis. For parts that do not provide
an XTAL_K value, skip setting it.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e879fe2..7f69a1b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2314,6 +2314,10 @@ static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 	       efuse->ht20_max_power_offset,
 	       sizeof(priv->ht20_max_power_offset));
 
+	if (priv->efuse_wifi.efuse8723.version >= 0x01) {
+		priv->has_xtalk = 1;
+		priv->xtalk = priv->efuse_wifi.efuse8723.xtal_k & 0x3f;
+	}
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
 		 efuse->vendor_name);
 	dev_info(&priv->udev->dev, "Product: %.41s\n",
@@ -2340,6 +2344,9 @@ static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
 	memcpy(priv->ht40_1s_tx_power_index_B, efuse->ht40_1s_tx_power_index_B,
 	       sizeof(priv->ht40_1s_tx_power_index_B));
 
+	priv->has_xtalk = 1;
+	priv->xtalk = priv->efuse_wifi.efuse8723bu.xtal_k & 0x3f;
+
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
 	dev_info(&priv->udev->dev, "Product: %.41s\n", efuse->device_name);
 
@@ -2444,6 +2451,9 @@ static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
+	priv->has_xtalk = 1;
+	priv->xtalk = priv->efuse_wifi.efuse8192eu.xtal_k & 0x3f;
+
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n", efuse->vendor_name);
 	dev_info(&priv->udev->dev, "Product: %.11s\n", efuse->device_name);
 	dev_info(&priv->udev->dev, "Serial: %.11s\n", efuse->serial);
@@ -3122,11 +3132,10 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 	else
 		rtl8xxxu_init_phy_regs(priv, rtl8xxx_agc_standard_table);
 
-	if ((priv->rtlchip == 0x8723a || priv->rtlchip == 0x8723b) &&
-	    priv->efuse_wifi.efuse8723.version >= 0x01) {
+	if (priv->has_xtalk) {
 		val32 = rtl8xxxu_read32(priv, REG_MAC_PHY_CTRL);
 
-		val8 = priv->efuse_wifi.efuse8723.xtal_k & 0x3f;
+		val8 = priv->xtalk;
 		val32 &= 0xff000fff;
 		val32 |= ((val8 | (val8 << 6)) << 12);
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index ea0b28a..2250e1b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1063,6 +1063,8 @@ struct rtl8xxxu_priv {
 	u32 ep_tx_high_queue:1;
 	u32 ep_tx_normal_queue:1;
 	u32 ep_tx_low_queue:1;
+	u32 has_xtalk:1;
+	u8 xtalk;
 	unsigned int pipe_interrupt;
 	unsigned int pipe_in;
 	unsigned int pipe_out[TXDESC_QUEUE_MAX];
-- 
cgit v0.10.2


From a0e262bcbe71892c88abb12b6d409553863efacc Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:56 -0500
Subject: rtl8xxxu: Another 8723bu patch for rtl8xxxu_init_phy_bb()

This function is going to need to be split up into chip specific
variants.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 7f69a1b..3f742b7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -3142,13 +3142,16 @@ static int rtl8xxxu_init_phy_bb(struct rtl8xxxu_priv *priv)
 		rtl8xxxu_write32(priv, REG_MAC_PHY_CTRL, val32);
 	}
 
-	ldoa15 = LDOA15_ENABLE | LDOA15_OBUF;
-	ldov12d = LDOV12D_ENABLE | BIT(2) | (2 << LDOV12D_VADJ_SHIFT);
-	ldohci12 = 0x57;
-	lpldo = 1;
-	val32 = (lpldo << 24) | (ldohci12 << 16) | (ldov12d << 8) | ldoa15;
+	if (priv->rtlchip != 0x8723bu) {
+		ldoa15 = LDOA15_ENABLE | LDOA15_OBUF;
+		ldov12d = LDOV12D_ENABLE | BIT(2) | (2 << LDOV12D_VADJ_SHIFT);
+		ldohci12 = 0x57;
+		lpldo = 1;
+		val32 = (lpldo << 24) | (ldohci12 << 16) |
+			(ldov12d << 8) | ldoa15;
 
-	rtl8xxxu_write32(priv, REG_LDOA15_CTRL, val32);
+		rtl8xxxu_write32(priv, REG_LDOA15_CTRL, val32);
+	}
 
 	return 0;
 }
-- 
cgit v0.10.2


From 360157eb25d4088c28ff114a613d2c95b61e93f4 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:57 -0500
Subject: rtl8xxxu: Another 8723bu magic register set during init

No indication of what register 0xa3 does anywhere in the vendor source.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 3f742b7..f5f38a5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5996,6 +5996,11 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 			rtl8xxxu_write8(priv, REG_TX_REPORT_CTRL + 1, 0x02);
 			/* TX report Timer. Unit: 32us */
 			rtl8xxxu_write16(priv, REG_TX_REPORT_TIME, 0xcdf0);
+
+			/* tmp ps ? */
+			val8 = rtl8xxxu_read8(priv, 0xa3);
+			val8 &= 0xf8;
+			rtl8xxxu_write8(priv, 0xa3, val8);
 		}
 
 		if (priv->ep_tx_normal_queue)
-- 
cgit v0.10.2


From 3a4be6a092c89ad7a94b147c3529d6626e5a59ff Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:58 -0500
Subject: rtl8xxxu: Init H2C command register for 8723bu

In addition make register read/write flow match closer to vendor
driver flow. This is mainly to be able to compare the register write
log with the vendor driver, and can be optimized later once 8723bu
support is working.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index f5f38a5..1d0869b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2685,6 +2685,11 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 		goto exit;
 	}
 
+	/*
+	 * Init H2C command
+	 */
+	if (priv->rtlchip == 0x8723b)
+		rtl8xxxu_write8(priv, REG_HMTFR, 0x0f);
 exit:
 	return ret;
 }
@@ -2927,14 +2932,20 @@ static void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv)
 
 	val32 = rtl8xxxu_read32(priv, REG_GPIO_MUXCFG);
 	val32 &= ~BIT(4);
+	rtl8xxxu_write32(priv, REG_GPIO_MUXCFG, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_GPIO_MUXCFG);
 	val32 |= BIT(3);
 	rtl8xxxu_write32(priv, REG_GPIO_MUXCFG, val32);
 
 	val32 = rtl8xxxu_read32(priv, REG_LEDCFG0);
-	val32 &= ~BIT(23);
 	val32 |= BIT(24);
 	rtl8xxxu_write32(priv, REG_LEDCFG0, val32);
 
+	val32 = rtl8xxxu_read32(priv, REG_LEDCFG0);
+	val32 &= ~BIT(23);
+	rtl8xxxu_write32(priv, REG_LEDCFG0, val32);
+
 	val32 = rtl8xxxu_read32(priv, 0x0944);
 	val32 |= (BIT(0) | BIT(1));
 	rtl8xxxu_write32(priv, 0x0944, val32);
@@ -2943,6 +2954,10 @@ static void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv)
 	val32 &= 0xffffff00;
 	val32 |= 0x77;
 	rtl8xxxu_write32(priv, 0x0930, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_PWR_DATA);
+	val32 |= PWR_DATA_EEPRPAD_RFE_CTRL_EN;
+	rtl8xxxu_write32(priv, REG_PWR_DATA, val32);
 }
 
 static int
@@ -6079,7 +6094,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write8(priv, 0xfe42, 0x80);
 	}
 
-	if (priv->rtlchip == 0x8192e || priv->rtlchip == 0x8723b) {
+	if (priv->rtlchip == 0x8192e) {
 		rtl8xxxu_write32(priv, REG_HIMR0, 0x00);
 		rtl8xxxu_write32(priv, REG_HIMR1, 0x00);
 	}
-- 
cgit v0.10.2


From f30ed675545dd4b9f32d3d674054b6024b413f1d Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:04:59 -0500
Subject: rtl8xxxu: 80M spur hack is for 8723au only

Only apply the 80M spur hack for 8723au parts.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 1d0869b..9ba4cad 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6151,11 +6151,12 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (ret)
 		goto exit;
 
-	/* Reduce 80M spur */
-	rtl8xxxu_write32(priv, REG_AFE_XTAL_CTRL, 0x0381808d);
-	rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
-	rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff82);
-	rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
+	if (priv->rtlchip == 0x8723a) { /* Reduce 80M spur */
+		rtl8xxxu_write32(priv, REG_AFE_XTAL_CTRL, 0x0381808d);
+		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
+		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff82);
+		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
+	}
 
 	/* RFSW Control - clear bit 14 ?? */
 	rtl8xxxu_write32(priv, REG_FPGA0_TX_INFO, 0x00000003);
-- 
cgit v0.10.2


From 1f1b20f11ab4ce38a338b278bb6b538bbfdc831a Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:00 -0500
Subject: rtl8xxxu: Do queue init in same order as 8723bu vendor driver

Reorganize the init sequence in order to be able to compare to the
8723bu vendor driver's init sequence.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 9ba4cad..27fb89f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6017,32 +6017,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 			val8 &= 0xf8;
 			rtl8xxxu_write8(priv, 0xa3, val8);
 		}
-
-		if (priv->ep_tx_normal_queue)
-			val8 = TX_PAGE_NUM_NORM_PQ;
-		else
-			val8 = 0;
-
-		rtl8xxxu_write8(priv, REG_RQPN_NPQ, val8);
-
-		val32 = (TX_PAGE_NUM_PUBQ << RQPN_NORM_PQ_SHIFT) | RQPN_LOAD;
-
-		if (priv->ep_tx_high_queue)
-			val32 |= (TX_PAGE_NUM_HI_PQ << RQPN_HI_PQ_SHIFT);
-		if (priv->ep_tx_low_queue)
-			val32 |= (TX_PAGE_NUM_LO_PQ << RQPN_LO_PQ_SHIFT);
-
-		rtl8xxxu_write32(priv, REG_RQPN, val32);
-
-		/*
-		 * Set TX buffer boundary
-		 */
-		val8 = TX_TOTAL_PAGE_NUM + 1;
-		rtl8xxxu_write8(priv, REG_TXPKTBUF_BCNQ_BDNY, val8);
-		rtl8xxxu_write8(priv, REG_TXPKTBUF_MGQ_BDNY, val8);
-		rtl8xxxu_write8(priv, REG_TXPKTBUF_WMAC_LBK_BF_HD, val8);
-		rtl8xxxu_write8(priv, REG_TRXFF_BNDY, val8);
-		rtl8xxxu_write8(priv, REG_TDECTRL + 1, val8);
 	}
 
 	ret = rtl8xxxu_download_firmware(priv);
@@ -6054,11 +6028,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (ret)
 		goto exit;
 
-	ret = rtl8xxxu_init_queue_priority(priv);
-	dev_dbg(dev, "%s: init_queue_priority %i\n", __func__, ret);
-	if (ret)
-		goto exit;
-
 	/* Fix USB interface interference issue */
 	if (priv->rtlchip == 0x8723a) {
 		rtl8xxxu_write8(priv, 0xfe40, 0xe0);
@@ -6158,6 +6127,43 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
 	}
 
+	if (!macpower){
+		if (priv->ep_tx_normal_queue)
+			val8 = TX_PAGE_NUM_NORM_PQ;
+		else
+			val8 = 0;
+
+		rtl8xxxu_write8(priv, REG_RQPN_NPQ, val8);
+
+		val32 = (TX_PAGE_NUM_PUBQ << RQPN_NORM_PQ_SHIFT) | RQPN_LOAD;
+
+		if (priv->ep_tx_high_queue)
+			val32 |= (TX_PAGE_NUM_HI_PQ << RQPN_HI_PQ_SHIFT);
+		if (priv->ep_tx_low_queue)
+			val32 |= (TX_PAGE_NUM_LO_PQ << RQPN_LO_PQ_SHIFT);
+
+		rtl8xxxu_write32(priv, REG_RQPN, val32);
+
+		/*
+		 * Set TX buffer boundary
+		 */
+		val8 = TX_TOTAL_PAGE_NUM + 1;
+
+		if (priv->rtlchip == 0x8723b)
+			val8 -= 1;
+
+		rtl8xxxu_write8(priv, REG_TXPKTBUF_BCNQ_BDNY, val8);
+		rtl8xxxu_write8(priv, REG_TXPKTBUF_MGQ_BDNY, val8);
+		rtl8xxxu_write8(priv, REG_TXPKTBUF_WMAC_LBK_BF_HD, val8);
+		rtl8xxxu_write8(priv, REG_TRXFF_BNDY, val8);
+		rtl8xxxu_write8(priv, REG_TDECTRL + 1, val8);
+	}
+
+	ret = rtl8xxxu_init_queue_priority(priv);
+	dev_dbg(dev, "%s: init_queue_priority %i\n", __func__, ret);
+	if (ret)
+		goto exit;
+
 	/* RFSW Control - clear bit 14 ?? */
 	rtl8xxxu_write32(priv, REG_FPGA0_TX_INFO, 0x00000003);
 	/* 0x07000760 */
-- 
cgit v0.10.2


From b87212ceceef655dae563efd2df13cb59dbd1711 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:01 -0500
Subject: rtl8xxxu: Do not set FPGA0_TX_INFO for 8723bu and use a larger PBP
 page size

The vendor driver does not set FPGA0_TX_INFO here. In additiona the
8723bu can handler a larger PBP page size.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 27fb89f..bb43937 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6165,7 +6165,8 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		goto exit;
 
 	/* RFSW Control - clear bit 14 ?? */
-	rtl8xxxu_write32(priv, REG_FPGA0_TX_INFO, 0x00000003);
+	if (priv->rtlchip != 0x8723b)
+		rtl8xxxu_write32(priv, REG_FPGA0_TX_INFO, 0x00000003);
 	/* 0x07000760 */
 	val32 = FPGA0_RF_TRSW | FPGA0_RF_TRSWB | FPGA0_RF_ANTSW |
 		FPGA0_RF_ANTSWB | FPGA0_RF_PAPE |
@@ -6185,8 +6186,12 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/*
 	 * Transfer page size is always 128
 	 */
-	val8 = (PBP_PAGE_SIZE_128 << PBP_PAGE_SIZE_RX_SHIFT) |
-		(PBP_PAGE_SIZE_128 << PBP_PAGE_SIZE_TX_SHIFT);
+	if (priv->rtlchip == 0x8723b)
+		val8 = (PBP_PAGE_SIZE_256 << PBP_PAGE_SIZE_RX_SHIFT) |
+			(PBP_PAGE_SIZE_256 << PBP_PAGE_SIZE_TX_SHIFT);
+	else
+		val8 = (PBP_PAGE_SIZE_128 << PBP_PAGE_SIZE_RX_SHIFT) |
+			(PBP_PAGE_SIZE_128 << PBP_PAGE_SIZE_TX_SHIFT);
 	rtl8xxxu_write8(priv, REG_PBP, val8);
 
 	/*
-- 
cgit v0.10.2


From fadfa041546924c28ca2a3bf1ed8fd25686e8cf5 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:02 -0500
Subject: rtl8xxxu: Set RX boundary for 8723bu

Set the correct TRXFF boundary for 8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index bb43937..6611128 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6182,7 +6182,10 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/*
 	 * Set RX page boundary
 	 */
-	rtl8xxxu_write16(priv, REG_TRXFF_BNDY + 2, 0x27ff);
+	if (priv->rtlchip == 0x8723b)
+		rtl8xxxu_write16(priv, REG_TRXFF_BNDY + 2, 0x3f7f);
+	else
+		rtl8xxxu_write16(priv, REG_TRXFF_BNDY + 2, 0x27ff);
 	/*
 	 * Transfer page size is always 128
 	 */
-- 
cgit v0.10.2


From c36906044104813304a9c40252f9b2a573ccd187 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:03 -0500
Subject: rtl8xxxu: Initialize burst parameters for 8723bu

Implement burst parameter sequence for 8723bu parts. Eventually this
should be moved into device specific sections.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 6611128..6314825 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6279,6 +6279,39 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	rtl8xxxu_write16(priv, REG_BEACON_TCFG, 0x660F);
 
 	/*
+	 * Initialize burst parameters
+	 */
+	if (priv->rtlchip == 0x8723b) {
+		/*
+		 * For USB high speed set 512B packets
+		 */
+		val8 = rtl8xxxu_read8(priv, REG_RXDMA_PRO_8723B);
+		val8 &= ~(BIT(4) | BIT(5));
+		val8 |= BIT(4);
+		val8 |= BIT(1) | BIT(2) | BIT(3);
+		rtl8xxxu_write8(priv, REG_RXDMA_PRO_8723B, val8);
+
+		/*
+		 * For USB high speed set 512B packets
+		 */
+		val8 = rtl8xxxu_read8(priv, REG_HT_SINGLE_AMPDU_8723B);
+		val8 |= BIT(7);
+		rtl8xxxu_write8(priv, REG_HT_SINGLE_AMPDU_8723B, val8);
+
+		rtl8xxxu_write16(priv, REG_MAX_AGGR_NUM, 0x0c14);
+		rtl8xxxu_write8(priv, REG_AMPDU_MAX_TIME_8723B, 0x5e);
+		rtl8xxxu_write32(priv, REG_AGGLEN_LMT, 0xffffffff);
+		rtl8xxxu_write8(priv, REG_RX_PKT_LIMIT, 0x18);
+		rtl8xxxu_write8(priv, REG_PIFS, 0x00);
+		rtl8xxxu_write8(priv, REG_USTIME_TSF_8723B, 0x50);
+		rtl8xxxu_write8(priv, REG_USTIME_EDCA, 0x50);
+
+		val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL);
+		val8 |= BIT(5) | BIT(6);
+		rtl8xxxu_write8(priv, REG_RSV_CTRL, val8);
+	}
+
+	/*
 	 * Enable CCK and OFDM block
 	 */
 	val32 = rtl8xxxu_read32(priv, REG_FPGA0_RF_MODE);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 5250388..8209830 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -405,7 +405,7 @@
 
 /* Presumably only found on newer chips such as 8723bu */
 #define REG_RX_DMA_CTRL_8723B		0x0286
-#define REG_RX_DMA_MODE_CTRL_8723B	0x0290
+#define REG_RXDMA_PRO_8723B		0x0290
 
 #define REG_RF_BB_CMD_ADDR		0x02c0
 #define REG_RF_BB_CMD_DATA		0x02c4
@@ -478,6 +478,7 @@
 #define REG_ARFR1			0x0448
 #define REG_ARFR2			0x044c
 #define REG_ARFR3			0x0450
+#define REG_AMPDU_MAX_TIME_8723B	0x0456
 #define REG_AGGLEN_LMT			0x0458
 #define REG_AMPDU_MIN_SPACE		0x045c
 #define REG_TXPKTBUF_WMAC_LBK_BF_HD	0x045d
@@ -496,6 +497,7 @@
 #define REG_PKT_VO_VI_LIFE_TIME		0x04c0
 #define REG_PKT_BE_BK_LIFE_TIME		0x04c2
 #define REG_STBC_SETTING		0x04c4
+#define REG_HT_SINGLE_AMPDU_8723B	0x04c7
 #define REG_PROT_MODE_CTRL		0x04c8
 #define REG_MAX_AGGR_NUM		0x04ca
 #define REG_RTS_MAX_AGGR_NUM		0x04cb
@@ -560,6 +562,7 @@
 #define  BEACON_DMA_ATIME_INT_TIME	2
 
 #define REG_ATIMWND			0x055a
+#define REG_USTIME_TSF_8723B		0x055c
 #define REG_BCN_MAX_ERR			0x055d
 #define REG_RXTSF_OFFSET_CCK		0x055e
 #define REG_RXTSF_OFFSET_OFDM		0x055f
-- 
cgit v0.10.2


From 1ea8e846c9eab861db1108302a391182c3fce528 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:04 -0500
Subject: rtl8xxxu: Call device specific _config_channel()

Having a version for the newer chips without calling it doesn't do
much good.....

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 6314825..960e1ac 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7348,7 +7348,7 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed)
 
 		rtl8723a_set_tx_power(priv, channel, ht40);
 
-		rtl8723au_config_channel(hw);
+		priv->fops->config_channel(hw);
 	}
 
 exit:
-- 
cgit v0.10.2


From 5ac61789e864ca7fbfe4425afe53837f535363cc Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:05 -0500
Subject: rtl8xxxu: 8723bu lock phy after RF init

Set PHY lock after running the RF init sequence on 8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 960e1ac..f20568b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6093,6 +6093,13 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	case 0x8723b:
 		rftable = rtl8723bu_radioa_1t_init_table;
 		ret = rtl8xxxu_init_phy_rf(priv, rftable, RF_A);
+		/*
+		 * PHY LCK
+		 */
+		rtl8xxxu_write_rfreg(priv, RF_A, 0xb0, 0xdfbe0);
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_MODE_AG, 0x8c01);
+		msleep(200);
+		rtl8xxxu_write_rfreg(priv, RF_A, 0xb0, 0xdffe0);
 		break;
 	case 0x8188c:
 		if (priv->hi_pa)
-- 
cgit v0.10.2


From 7e9567ff98a6b7312a6eb1fe0f9c0f81b42147eb Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:06 -0500
Subject: rtl8xxxu: Add REG_DWBCN1_CTRL_8723B define

List yet another new register found on the 8723b.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index 8209830..ec6f40f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -397,6 +397,8 @@
 #define REG_AUTO_LLT			0x0224
 #define  AUTO_LLT_INIT_LLT		BIT(16)
 
+#define REG_DWBCN1_CTRL_8723B		0x0228
+
 /* 0x0280 ~ 0x02FF	RXDMA Configuration */
 #define REG_RXDMA_AGG_PG_TH		0x0280
 #define  RXDMA_USB_AGG_ENABLE		BIT(31)
-- 
cgit v0.10.2


From 2f109c8e51d6302a18687c9381a62401d6b69f24 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:07 -0500
Subject: rtl8xxxu: Group chip quirks together

Group chip quirks together instead of having them scattered all over
in the init code.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index f20568b..ad63719 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6028,18 +6028,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (ret)
 		goto exit;
 
-	/* Fix USB interface interference issue */
-	if (priv->rtlchip == 0x8723a) {
-		rtl8xxxu_write8(priv, 0xfe40, 0xe0);
-		rtl8xxxu_write8(priv, 0xfe41, 0x8d);
-		rtl8xxxu_write8(priv, 0xfe42, 0x80);
-		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, 0xfd0320);
-	} else {
-		val32 = rtl8xxxu_read32(priv, REG_TXDMA_OFFSET_CHK);
-		val32 |= TXDMA_OFFSET_DROP_DATA_EN;
-		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32);
-	}
-
 	/* Solve too many protocol error on USB bus */
 	/* Can't do this for 8188/8192 UMC A cut parts */
 	if (priv->rtlchip == 0x8723a ||
@@ -6127,11 +6115,25 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	if (ret)
 		goto exit;
 
-	if (priv->rtlchip == 0x8723a) { /* Reduce 80M spur */
+	/*
+	 * Chip specific quirks
+	 */
+	if (priv->rtlchip == 0x8723a) {
+		/* Fix USB interface interference issue */
+		rtl8xxxu_write8(priv, 0xfe40, 0xe0);
+		rtl8xxxu_write8(priv, 0xfe41, 0x8d);
+		rtl8xxxu_write8(priv, 0xfe42, 0x80);
+		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, 0xfd0320);
+
+		/* Reduce 80M spur */
 		rtl8xxxu_write32(priv, REG_AFE_XTAL_CTRL, 0x0381808d);
 		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
 		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff82);
 		rtl8xxxu_write32(priv, REG_AFE_PLL_CTRL, 0xf0ffff83);
+	} else {
+		val32 = rtl8xxxu_read32(priv, REG_TXDMA_OFFSET_CHK);
+		val32 |= TXDMA_OFFSET_DROP_DATA_EN;
+		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32);
 	}
 
 	if (!macpower){
-- 
cgit v0.10.2


From 3e88ca447a8a17e122ba58d79c725c63b3efcd70 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:08 -0500
Subject: rtl8xxxu: Setup RX aggregation

This initializes RX DMA aggregation on 8723bu. We should do this for
all parts eventually, and also init TX aggregation.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ad63719..04dad194 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5962,6 +5962,25 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ignore_wlan));
 }
 
+static void rtl8723bu_init_aggregation(struct rtl8xxxu_priv *priv)
+{
+	u32 agg_rx;
+	u8 agg_ctrl;
+
+	/*
+	 * For now simply disable RX aggregation
+	 */
+	agg_ctrl = rtl8xxxu_read8(priv, REG_TRXDMA_CTRL);
+	agg_ctrl &= ~TRXDMA_CTRL_RXDMA_AGG_EN;
+
+	agg_rx = rtl8xxxu_read32(priv, REG_RXDMA_AGG_PG_TH);
+	agg_rx &= ~RXDMA_USB_AGG_ENABLE;
+	agg_rx &= ~0xff0f;
+
+	rtl8xxxu_write8(priv, REG_TRXDMA_CTRL, agg_ctrl);
+	rtl8xxxu_write32(priv, REG_RXDMA_AGG_PG_TH, agg_rx);
+}
+
 static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
@@ -6320,6 +6339,9 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write8(priv, REG_RSV_CTRL, val8);
 	}
 
+	if (priv->fops->init_aggregation)
+		priv->fops->init_aggregation(priv);
+
 	/*
 	 * Enable CCK and OFDM block
 	 */
@@ -7998,6 +8020,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.config_channel = rtl8723bu_config_channel,
 	.init_bt = rtl8723bu_init_bt,
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
+	.init_aggregation = rtl8723bu_init_aggregation,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 2250e1b..14b7a35 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1133,6 +1133,7 @@ struct rtl8xxxu_fileops {
 	void (*init_bt) (struct rtl8xxxu_priv *priv);
 	int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb,
 			      struct ieee80211_rx_status *rx_status);
+	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index ec6f40f..d6c7ad3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -318,6 +318,7 @@
 #define  PBP_PAGE_SIZE_1024		0x4
 
 #define REG_TRXDMA_CTRL			0x010c
+#define  TRXDMA_CTRL_RXDMA_AGG_EN	BIT(2)
 #define  TRXDMA_CTRL_VOQ_SHIFT		4
 #define  TRXDMA_CTRL_VIQ_SHIFT		6
 #define  TRXDMA_CTRL_BEQ_SHIFT		8
-- 
cgit v0.10.2


From f2a4163a22c66d178084ff722c490f5ecf91a088 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:09 -0500
Subject: rtl8xxxu: Add missing blank space in front of bracket

Keep the automated tools happy

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 04dad194..1fa0dbf 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6155,7 +6155,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		rtl8xxxu_write32(priv, REG_TXDMA_OFFSET_CHK, val32);
 	}
 
-	if (!macpower){
+	if (!macpower) {
 		if (priv->ep_tx_normal_queue)
 			val8 = TX_PAGE_NUM_NORM_PQ;
 		else
-- 
cgit v0.10.2


From 9c79bf95d08a920b1c1219a2a9d5fce2f6172bbe Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:10 -0500
Subject: rtl8xxxu: Implement init_statistics for 8723bu

Vendor driver implements this for 8723b and 8821 series

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 1fa0dbf..6b6fa1f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5981,6 +5981,29 @@ static void rtl8723bu_init_aggregation(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write32(priv, REG_RXDMA_AGG_PG_TH, agg_rx);
 }
 
+static void rtl8723bu_init_statistics(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+
+	/* Time duration for NHM unit: 4us, 0x2710=40ms */
+	rtl8xxxu_write16(priv, REG_NHM_TIMER_8723B + 2, 0x2710);
+	rtl8xxxu_write16(priv, REG_NHM_TH9_TH10_8723B + 2, 0xffff);
+	rtl8xxxu_write32(priv, REG_NHM_TH3_TO_TH0_8723B, 0xffffff52);
+	rtl8xxxu_write32(priv, REG_NHM_TH7_TO_TH4_8723B, 0xffffffff);
+	/* TH8 */
+	val32 = rtl8xxxu_read32(priv, REG_FPGA0_IQK);
+	val32 |= 0xff;
+	rtl8xxxu_write32(priv, REG_FPGA0_IQK, val32);
+	/* Enable CCK */
+	val32 = rtl8xxxu_read32(priv, REG_NHM_TH9_TH10_8723B);
+	val32 |= BIT(8) | BIT(9) | BIT(10);
+	rtl8xxxu_write32(priv, REG_NHM_TH9_TH10_8723B, val32);
+	/* Max power amongst all RX antennas */
+	val32 = rtl8xxxu_read32(priv, REG_OFDM0_FA_RSTC);
+	val32 |= BIT(7);
+	rtl8xxxu_write32(priv, REG_OFDM0_FA_RSTC, val32);
+}
+
 static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
@@ -6371,6 +6394,9 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 
 	rtl8xxxu_write16(priv, REG_FAST_EDCA_CTRL, 0);
 
+	if (priv->fops->init_statistics)
+		priv->fops->init_statistics(priv);
+
 	rtl8723a_phy_lc_calibrate(priv);
 
 	priv->fops->phy_iq_calibrate(priv);
@@ -8021,6 +8047,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.init_bt = rtl8723bu_init_bt,
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.init_aggregation = rtl8723bu_init_aggregation,
+	.init_statistics = rtl8723bu_init_statistics,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 14b7a35..b75678b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1134,6 +1134,7 @@ struct rtl8xxxu_fileops {
 	int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb,
 			      struct ieee80211_rx_status *rx_status);
 	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
+	void (*init_statistics) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index d6c7ad3..d3f6fa3 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -833,6 +833,11 @@
 #define REG_FPGA0_ANALOG3		0x0888
 #define REG_FPGA0_ANALOG4		0x088c
 
+#define REG_NHM_TH9_TH10_8723B		0x0890
+#define REG_NHM_TIMER_8723B		0x0894
+#define REG_NHM_TH3_TO_TH0_8723B	0x0898
+#define REG_NHM_TH7_TO_TH4_8723B	0x089c
+
 #define REG_FPGA0_XA_LSSI_READBACK	0x08a0	/* Tranceiver LSSI Readback */
 #define REG_FPGA0_XB_LSSI_READBACK	0x08a4
 #define REG_HSPI_XA_READBACK		0x08b8	/* Transceiver A HSPI read */
@@ -869,6 +874,8 @@
 
 #define REG_OFDM0_TR_MUX_PAR		0x0c08
 
+#define REG_OFDM0_FA_RSTC		0x0c0c
+
 #define REG_OFDM0_XA_RX_IQ_IMBALANCE	0x0c14
 #define REG_OFDM0_XB_RX_IQ_IMBALANCE	0x0c1c
 
-- 
cgit v0.10.2


From 541bca7f5d8ffbd4dba8b432f3a5162cb7ba41a6 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:11 -0500
Subject: rtl8xxxu: RF_T_METER is different on the newer chips

Provide RF_T_METER register location for nextgen chips.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index d3f6fa3..a055362 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -1104,9 +1104,10 @@
 /*
  * NextGen regs: 8723BU
  */
-#define	RF6052_REG_UNKNOWN_43		0x43
-#define	RF6052_REG_UNKNOWN_55		0x55
-#define	RF6052_REG_S0S1			0xb0
-#define	RF6052_REG_UNKNOWN_DF		0xdf
-#define	RF6052_REG_UNKNOWN_ED		0xed
-#define	RF6052_REG_WE_LUT		0xef
+#define RF6052_REG_T_METER_8723B	0x42
+#define RF6052_REG_UNKNOWN_43		0x43
+#define RF6052_REG_UNKNOWN_55		0x55
+#define RF6052_REG_S0S1			0xb0
+#define RF6052_REG_UNKNOWN_DF		0xdf
+#define RF6052_REG_UNKNOWN_ED		0xed
+#define RF6052_REG_WE_LUT		0xef
-- 
cgit v0.10.2


From fc1c89b3407fe0b3ebc8fa651332a62ed773872d Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:12 -0500
Subject: rtl8xxxu: Set WLAN_ACT_CONTROL per vendor driver setting

The initial code set the wrong setting in WLAN_ACT_CONTROL for the
8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 6b6fa1f..52c7567 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5894,7 +5894,7 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	/*
 	 * WLAN action by PTA
 	 */
-	rtl8xxxu_write8(priv, REG_WLAN_ACT_CONTROL_8723B, 0x0c);
+	rtl8xxxu_write8(priv, REG_WLAN_ACT_CONTROL_8723B, 0x04);
 
 	/*
 	 * BT select S0/S1 controlled by WiFi
-- 
cgit v0.10.2


From 499cfc02a08a27ba29d22938ddbe5140806316c5 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:13 -0500
Subject: rtl8xxxu: 8723bu: REG_BT_COEX_TABLE4 is only 8 bits

The BT_COEX_TABLE register list contains 3 32 bit registers and one 8
bit register. Hence, use rtl8xxxu_write8() when writing the 8 bit
register.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 52c7567..47d85d5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5954,7 +5954,7 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE1, 0x55555555);
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE2, 0x5a5a5a5a);
 	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE3, 0x00ffffff);
-	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE4, 0x00000003);
+	rtl8xxxu_write8(priv, REG_BT_COEX_TABLE4, 0x03);
 
 	memset(&h2c, 0, sizeof(struct h2c_cmd));
 	h2c.ignore_wlan.cmd = H2C_8723B_BT_IGNORE_WLANACT;
-- 
cgit v0.10.2


From 120e627f658b3fe2be2f3699527dd687168b3693 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:14 -0500
Subject: rtl8xxxu: Use name for REG_RFE_BUFFER rather than hard coded value

Register 0x0944 is REG_RFE_BUFFER. Use the name rather than hard coded
value when accessing it.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 47d85d5..3764f3e 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2946,9 +2946,9 @@ static void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv)
 	val32 &= ~BIT(23);
 	rtl8xxxu_write32(priv, REG_LEDCFG0, val32);
 
-	val32 = rtl8xxxu_read32(priv, 0x0944);
+	val32 = rtl8xxxu_read32(priv, REG_RFE_BUFFER);
 	val32 |= (BIT(0) | BIT(1));
-	rtl8xxxu_write32(priv, 0x0944, val32);
+	rtl8xxxu_write32(priv, REG_RFE_BUFFER, val32);
 
 	val32 = rtl8xxxu_read32(priv, 0x0930);
 	val32 &= 0xffffff00;
@@ -5912,9 +5912,9 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	 */
 	rtl8xxxu_write8(priv, 0x0974, 0xff);
 
-	val32 = rtl8xxxu_read32(priv, 0x0944);
+	val32 = rtl8xxxu_read32(priv, REG_RFE_BUFFER);
 	val32 |= (BIT(0) | BIT(1));
-	rtl8xxxu_write32(priv, 0x0944, val32);
+	rtl8xxxu_write32(priv, REG_RFE_BUFFER, val32);
 
 	rtl8xxxu_write8(priv, REG_RFE_CTRL_ANTA_SRC, 0x77);
 
-- 
cgit v0.10.2


From 59b743979c178b780bc57f179493fdbf3ad4f0b7 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:15 -0500
Subject: rtl8xxxu: Use REG_RFE_CTRL_ANTA_SRC rather than hard coded value

Another case where we should use the register name rather than the
hard coded value when accessing it.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 3764f3e..5b74ba1 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2950,10 +2950,10 @@ static void rtl8723bu_phy_init_antenna_selection(struct rtl8xxxu_priv *priv)
 	val32 |= (BIT(0) | BIT(1));
 	rtl8xxxu_write32(priv, REG_RFE_BUFFER, val32);
 
-	val32 = rtl8xxxu_read32(priv, 0x0930);
+	val32 = rtl8xxxu_read32(priv, REG_RFE_CTRL_ANTA_SRC);
 	val32 &= 0xffffff00;
 	val32 |= 0x77;
-	rtl8xxxu_write32(priv, 0x0930, val32);
+	rtl8xxxu_write32(priv, REG_RFE_CTRL_ANTA_SRC, val32);
 
 	val32 = rtl8xxxu_read32(priv, REG_PWR_DATA);
 	val32 |= PWR_DATA_EEPRPAD_RFE_CTRL_EN;
-- 
cgit v0.10.2


From a3a5dac6b1bf69b1d90f3293abfb4392ec2fc68f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:16 -0500
Subject: rtl8xxxu: Setup coex table correctly (hopefully)

Use the same values as the vendor driver when setting up the BTCOEX
table for 8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 5b74ba1..3e99a17 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5842,6 +5842,7 @@ static void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e);
 }
 
+#ifdef NEED_PS_TDMA
 static void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv,
 				  u8 arg1, u8 arg2, u8 arg3, u8 arg4, u8 arg5)
 {
@@ -5856,6 +5857,7 @@ static void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv,
 	h2c.b_type_dma.data5 = arg5;
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.b_type_dma));
 }
+#endif
 
 static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 {
@@ -5944,18 +5946,20 @@ static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
 	/*
 	 * Software control, antenna at WiFi side
 	 */
+#ifdef NEED_PS_TDMA
 	rtl8723bu_set_ps_tdma(priv, 0x08, 0x00, 0x00, 0x00, 0x00);
+#endif
+
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE1, 0x55555555);
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE2, 0x55555555);
+	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE3, 0x00ffffff);
+	rtl8xxxu_write8(priv, REG_BT_COEX_TABLE4, 0x03);
 
 	memset(&h2c, 0, sizeof(struct h2c_cmd));
 	h2c.bt_info.cmd = H2C_8723B_BT_INFO;
 	h2c.bt_info.data = BIT(0);
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.bt_info));
 
-	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE1, 0x55555555);
-	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE2, 0x5a5a5a5a);
-	rtl8xxxu_write32(priv, REG_BT_COEX_TABLE3, 0x00ffffff);
-	rtl8xxxu_write8(priv, REG_BT_COEX_TABLE4, 0x03);
-
 	memset(&h2c, 0, sizeof(struct h2c_cmd));
 	h2c.ignore_wlan.cmd = H2C_8723B_BT_IGNORE_WLANACT;
 	h2c.ignore_wlan.data = 0;
-- 
cgit v0.10.2


From db08de9443be5d4eef724621d3d51cab4daf25d3 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:17 -0500
Subject: rtl8xxxu: Do not use hard-wired RF enable settings for 8723bu

These settings simply block the 8723bu, for now leave an empty
function.

With this change we can finally communicate with aliens using the
8723bu!

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 3e99a17..5b6b9c5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1596,6 +1596,10 @@ static void rtl8723a_enable_rf(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_TXPAUSE, 0x00);
 }
 
+static void rtl8723b_enable_rf(struct rtl8xxxu_priv *priv)
+{
+}
+
 static void rtl8723a_disable_rf(struct rtl8xxxu_priv *priv)
 {
 	u8 sps0;
@@ -7665,7 +7669,7 @@ static int rtl8xxxu_start(struct ieee80211_hw *hw)
 	init_usb_anchor(&priv->tx_anchor);
 	init_usb_anchor(&priv->int_anchor);
 
-	rtl8723a_enable_rf(priv);
+	priv->fops->enable_rf(priv);
 	if (priv->usb_interrupts) {
 		ret = rtl8xxxu_submit_int_urb(hw);
 		if (ret)
@@ -8031,6 +8035,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
+	.enable_rf = rtl8723a_enable_rf,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8052,6 +8057,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.init_aggregation = rtl8723bu_init_aggregation,
 	.init_statistics = rtl8723bu_init_statistics,
+	.enable_rf = rtl8723b_enable_rf,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -8072,6 +8078,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
+	.enable_rf = rtl8723a_enable_rf,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8091,6 +8098,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
+	.enable_rf = rtl8723b_enable_rf,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index b75678b..ac8f47f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1135,6 +1135,7 @@ struct rtl8xxxu_fileops {
 			      struct ieee80211_rx_status *rx_status);
 	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
 	void (*init_statistics) (struct rtl8xxxu_priv *priv);
+	void (*enable_rf) (struct rtl8xxxu_priv *priv);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
-- 
cgit v0.10.2


From 4a0d7db53124dc81b03926bcec864a0575d94f8e Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:18 -0500
Subject: rtl8xxxu: Correct struct rtl8723bu_efuse to list power bases
 correctly

Correct TX power definitions in rtl8723bu_efuse

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 5b6b9c5..40447e4 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2338,16 +2338,6 @@ static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
-	memcpy(priv->cck_tx_power_index_A, efuse->cck_tx_power_index_A,
-	       sizeof(priv->cck_tx_power_index_A));
-	memcpy(priv->cck_tx_power_index_B, efuse->cck_tx_power_index_B,
-	       sizeof(priv->cck_tx_power_index_B));
-
-	memcpy(priv->ht40_1s_tx_power_index_A, efuse->ht40_1s_tx_power_index_A,
-	       sizeof(priv->ht40_1s_tx_power_index_A));
-	memcpy(priv->ht40_1s_tx_power_index_B, efuse->ht40_1s_tx_power_index_B,
-	       sizeof(priv->ht40_1s_tx_power_index_B));
-
 	priv->has_xtalk = 1;
 	priv->xtalk = priv->efuse_wifi.efuse8723bu.xtal_k & 0x3f;
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index ac8f47f..6d56581 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -53,6 +53,7 @@
 
 #define RTL8723A_CHANNEL_GROUPS		3
 #define RTL8723A_MAX_RF_PATHS		2
+#define RTL8723B_MAX_RF_PATHS		4
 #define RF6052_MAX_TX_PWR		0x3f
 
 #define EFUSE_MAP_LEN			512
@@ -631,14 +632,26 @@ struct rtl8192cu_efuse {
 	u8 customer_id;
 };
 
+struct rtl8723bu_efuse_tx_power {
+	u8 cck_base[6];
+	u8 ht40_base[5];
+	struct rtl8723au_idx ht20_ofdm_1s_diff;
+	struct rtl8723au_idx ht40_ht20_2s_diff;
+	struct rtl8723au_idx ofdm_cck_2s_diff; /* not used */
+	struct rtl8723au_idx ht40_ht20_3s_diff;
+	struct rtl8723au_idx ofdm_cck_3s_diff; /* not used */
+	struct rtl8723au_idx ht40_ht20_4s_diff;
+	struct rtl8723au_idx ofdm_cck_4s_diff; /* not used */
+	u8 dummy5g[24]; /* max channel group (14) + power diff offset (10) */
+};
+
 struct rtl8723bu_efuse {
 	__le16 rtl_id;
 	u8 res0[0x0e];
-	u8 cck_tx_power_index_A[3];	/* 0x10 */
-	u8 cck_tx_power_index_B[3];
-	u8 ht40_1s_tx_power_index_A[3];	/* 0x16 */
-	u8 ht40_1s_tx_power_index_B[3];
-	u8 res1[0x9c];
+	struct rtl8723bu_efuse_tx_power tx_power_index_A;	/* 0x10 */
+	struct rtl8723bu_efuse_tx_power tx_power_index_B;	/* 0x3a */
+	struct rtl8723bu_efuse_tx_power tx_power_index_C;	/* 0x64 */
+	struct rtl8723bu_efuse_tx_power tx_power_index_D;	/* 0x8e */
 	u8 channel_plan;		/* 0xb8 */
 	u8 xtal_k;
 	u8 thermal_meter;
-- 
cgit v0.10.2


From e796dab4b987503913418260bce782d230df60db Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:19 -0500
Subject: rtl8xxxu: Introduce set_tx_power() fileop and a new 8723b dummy
 derivative

The 8723b series is significantly different from the older generation
in this sense. So far the 8723b version doesn't do anything useful.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 40447e4..de1e654 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1680,6 +1680,24 @@ static int rtl8723a_channel_to_group(int channel)
 	return group;
 }
 
+static int rtl8723b_channel_to_group(int channel)
+{
+	int group;
+
+	if (channel < 3)
+		group = 0;
+	else if (channel < 6)
+		group = 1;
+	else if (channel < 9)
+		group = 2;
+	else if (channel < 12)
+		group = 3;
+	else
+		group = 4;
+
+	return group;
+}
+
 static void rtl8723au_config_channel(struct ieee80211_hw *hw)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
@@ -2041,6 +2059,14 @@ rtl8723a_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 	}
 }
 
+static void
+rtl8723b_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
+{
+	int group;
+
+	group = rtl8723b_channel_to_group(channel);
+}
+
 static void rtl8xxxu_set_linktype(struct rtl8xxxu_priv *priv,
 				  enum nl80211_iftype linktype)
 {
@@ -6378,7 +6404,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/*
 	 * Start out with default power levels for channel 6, 20MHz
 	 */
-	rtl8723a_set_tx_power(priv, 1, false);
+	priv->fops->set_tx_power(priv, 1, false);
 
 	/* Let the 8051 take control of antenna setting */
 	val8 = rtl8xxxu_read8(priv, REG_LEDCFG2);
@@ -7401,7 +7427,7 @@ static int rtl8xxxu_config(struct ieee80211_hw *hw, u32 changed)
 
 		channel = hw->conf.chandef.chan->hw_value;
 
-		rtl8723a_set_tx_power(priv, channel, ht40);
+		priv->fops->set_tx_power(priv, channel, ht40);
 
 		priv->fops->config_channel(hw);
 	}
@@ -8026,6 +8052,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.config_channel = rtl8723au_config_channel,
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.enable_rf = rtl8723a_enable_rf,
+	.set_tx_power = rtl8723a_set_tx_power,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8048,6 +8075,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.init_aggregation = rtl8723bu_init_aggregation,
 	.init_statistics = rtl8723bu_init_statistics,
 	.enable_rf = rtl8723b_enable_rf,
+	.set_tx_power = rtl8723b_set_tx_power,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -8069,6 +8097,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.config_channel = rtl8723au_config_channel,
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.enable_rf = rtl8723a_enable_rf,
+	.set_tx_power = rtl8723a_set_tx_power,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8089,6 +8118,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.config_channel = rtl8723bu_config_channel,
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.enable_rf = rtl8723b_enable_rf,
+	.set_tx_power = rtl8723b_set_tx_power,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 6d56581..c7b4c92 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1149,6 +1149,8 @@ struct rtl8xxxu_fileops {
 	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
 	void (*init_statistics) (struct rtl8xxxu_priv *priv);
 	void (*enable_rf) (struct rtl8xxxu_priv *priv);
+	void (*set_tx_power) (struct rtl8xxxu_priv *priv, int channel,
+			      bool ht40);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
-- 
cgit v0.10.2


From 3e84f9386162878deb146a7e4f05710693f1a9c7 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:20 -0500
Subject: rtl8xxxu: Use size of source pointer when copying efuse data

Some newer chips have more channel groups in their efuse parameter
tables, so use the size of the source, rather than the destination
when copying them out. This avoids copying garbage when increasing the
common array sizes.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index de1e654..4c37bd4 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2318,31 +2318,31 @@ static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	memcpy(priv->cck_tx_power_index_A,
 	       efuse->cck_tx_power_index_A,
-	       sizeof(priv->cck_tx_power_index_A));
+	       sizeof(efuse->cck_tx_power_index_A));
 	memcpy(priv->cck_tx_power_index_B,
 	       efuse->cck_tx_power_index_B,
-	       sizeof(priv->cck_tx_power_index_B));
+	       sizeof(efuse->cck_tx_power_index_B));
 
 	memcpy(priv->ht40_1s_tx_power_index_A,
 	       efuse->ht40_1s_tx_power_index_A,
-	       sizeof(priv->ht40_1s_tx_power_index_A));
+	       sizeof(efuse->ht40_1s_tx_power_index_A));
 	memcpy(priv->ht40_1s_tx_power_index_B,
 	       efuse->ht40_1s_tx_power_index_B,
-	       sizeof(priv->ht40_1s_tx_power_index_B));
+	       sizeof(efuse->ht40_1s_tx_power_index_B));
 
 	memcpy(priv->ht20_tx_power_index_diff,
 	       efuse->ht20_tx_power_index_diff,
-	       sizeof(priv->ht20_tx_power_index_diff));
+	       sizeof(efuse->ht20_tx_power_index_diff));
 	memcpy(priv->ofdm_tx_power_index_diff,
 	       efuse->ofdm_tx_power_index_diff,
-	       sizeof(priv->ofdm_tx_power_index_diff));
+	       sizeof(efuse->ofdm_tx_power_index_diff));
 
 	memcpy(priv->ht40_max_power_offset,
 	       efuse->ht40_max_power_offset,
-	       sizeof(priv->ht40_max_power_offset));
+	       sizeof(efuse->ht40_max_power_offset));
 	memcpy(priv->ht20_max_power_offset,
 	       efuse->ht20_max_power_offset,
-	       sizeof(priv->ht20_max_power_offset));
+	       sizeof(efuse->ht20_max_power_offset));
 
 	if (priv->efuse_wifi.efuse8723.version >= 0x01) {
 		priv->has_xtalk = 1;
@@ -2403,34 +2403,34 @@ static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv)
 
 	memcpy(priv->cck_tx_power_index_A,
 	       efuse->cck_tx_power_index_A,
-	       sizeof(priv->cck_tx_power_index_A));
+	       sizeof(efuse->cck_tx_power_index_A));
 	memcpy(priv->cck_tx_power_index_B,
 	       efuse->cck_tx_power_index_B,
-	       sizeof(priv->cck_tx_power_index_B));
+	       sizeof(efuse->cck_tx_power_index_B));
 
 	memcpy(priv->ht40_1s_tx_power_index_A,
 	       efuse->ht40_1s_tx_power_index_A,
-	       sizeof(priv->ht40_1s_tx_power_index_A));
+	       sizeof(efuse->ht40_1s_tx_power_index_A));
 	memcpy(priv->ht40_1s_tx_power_index_B,
 	       efuse->ht40_1s_tx_power_index_B,
-	       sizeof(priv->ht40_1s_tx_power_index_B));
+	       sizeof(efuse->ht40_1s_tx_power_index_B));
 	memcpy(priv->ht40_2s_tx_power_index_diff,
 	       efuse->ht40_2s_tx_power_index_diff,
-	       sizeof(priv->ht40_2s_tx_power_index_diff));
+	       sizeof(efuse->ht40_2s_tx_power_index_diff));
 
 	memcpy(priv->ht20_tx_power_index_diff,
 	       efuse->ht20_tx_power_index_diff,
-	       sizeof(priv->ht20_tx_power_index_diff));
+	       sizeof(efuse->ht20_tx_power_index_diff));
 	memcpy(priv->ofdm_tx_power_index_diff,
 	       efuse->ofdm_tx_power_index_diff,
-	       sizeof(priv->ofdm_tx_power_index_diff));
+	       sizeof(efuse->ofdm_tx_power_index_diff));
 
 	memcpy(priv->ht40_max_power_offset,
 	       efuse->ht40_max_power_offset,
-	       sizeof(priv->ht40_max_power_offset));
+	       sizeof(efuse->ht40_max_power_offset));
 	memcpy(priv->ht20_max_power_offset,
 	       efuse->ht20_max_power_offset,
-	       sizeof(priv->ht20_max_power_offset));
+	       sizeof(efuse->ht20_max_power_offset));
 
 	dev_info(&priv->udev->dev, "Vendor: %.7s\n",
 		 efuse->vendor_name);
-- 
cgit v0.10.2


From 21db9973303caebcbe2fde6dcc668c1aac2ed55f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:21 -0500
Subject: rtl8xxxu: Bump TX power arrays to handle larger channel groups

Newer generation chips have more channels groups. In order to carry
the larger arrays in common structures, bump the array sizes to
match.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index c7b4c92..92768f5 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -53,7 +53,9 @@
 
 #define RTL8723A_CHANNEL_GROUPS		3
 #define RTL8723A_MAX_RF_PATHS		2
+#define RTL8723B_CHANNEL_GROUPS		6
 #define RTL8723B_MAX_RF_PATHS		4
+#define RTL8XXXU_MAX_CHANNEL_GROUPS	6
 #define RF6052_MAX_TX_PWR		0x3f
 
 #define EFUSE_MAP_LEN			512
@@ -1046,19 +1048,24 @@ struct rtl8xxxu_priv {
 	u8 mac_addr[ETH_ALEN];
 	char chip_name[8];
 	char chip_vendor[8];
-	u8 cck_tx_power_index_A[3];	/* 0x10 */
-	u8 cck_tx_power_index_B[3];
-	u8 ht40_1s_tx_power_index_A[3];	/* 0x16 */
-	u8 ht40_1s_tx_power_index_B[3];
+	u8 cck_tx_power_index_A[RTL8XXXU_MAX_CHANNEL_GROUPS];
+	u8 cck_tx_power_index_B[RTL8XXXU_MAX_CHANNEL_GROUPS];
+	u8 ht40_1s_tx_power_index_A[RTL8XXXU_MAX_CHANNEL_GROUPS];
+	u8 ht40_1s_tx_power_index_B[RTL8XXXU_MAX_CHANNEL_GROUPS];
 	/*
 	 * The following entries are half-bytes split as:
 	 * bits 0-3: path A, bits 4-7: path B, all values 4 bits signed
 	 */
-	struct rtl8723au_idx ht40_2s_tx_power_index_diff[3];
-	struct rtl8723au_idx ht20_tx_power_index_diff[3];
-	struct rtl8723au_idx ofdm_tx_power_index_diff[3];
-	struct rtl8723au_idx ht40_max_power_offset[3];
-	struct rtl8723au_idx ht20_max_power_offset[3];
+	struct rtl8723au_idx ht40_2s_tx_power_index_diff[
+		RTL8XXXU_MAX_CHANNEL_GROUPS];
+	struct rtl8723au_idx ht20_tx_power_index_diff[
+		RTL8XXXU_MAX_CHANNEL_GROUPS];
+	struct rtl8723au_idx ofdm_tx_power_index_diff[
+		RTL8XXXU_MAX_CHANNEL_GROUPS];
+	struct rtl8723au_idx ht40_max_power_offset[
+		RTL8XXXU_MAX_CHANNEL_GROUPS];
+	struct rtl8723au_idx ht20_max_power_offset[
+		RTL8XXXU_MAX_CHANNEL_GROUPS];
 	u32 chip_cut:4;
 	u32 rom_rev:4;
 	u32 is_multi_func:1;
-- 
cgit v0.10.2


From 3be269990814e7f052e134cfc1d538b7010c0600 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:22 -0500
Subject: rtl8xxxu: Parse efuse power indices for 8723bu

This should (hopefully) parse the power indices correctly for the
8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4c37bd4..0723008 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2358,12 +2358,55 @@ static int rtl8723au_parse_efuse(struct rtl8xxxu_priv *priv)
 static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv)
 {
 	struct rtl8723bu_efuse *efuse = &priv->efuse_wifi.efuse8723bu;
+	int i;
 
 	if (efuse->rtl_id != cpu_to_le16(0x8129))
 		return -EINVAL;
 
 	ether_addr_copy(priv->mac_addr, efuse->mac_addr);
 
+	memcpy(priv->cck_tx_power_index_A, efuse->tx_power_index_A.cck_base,
+	       sizeof(efuse->tx_power_index_A.cck_base));
+	memcpy(priv->cck_tx_power_index_B, efuse->tx_power_index_B.cck_base,
+	       sizeof(efuse->tx_power_index_B.cck_base));
+
+	memcpy(priv->ht40_1s_tx_power_index_A,
+	       efuse->tx_power_index_A.ht40_base,
+	       sizeof(efuse->tx_power_index_A.ht40_base));
+	memcpy(priv->ht40_1s_tx_power_index_B,
+	       efuse->tx_power_index_B.ht40_base,
+	       sizeof(efuse->tx_power_index_B.ht40_base));
+
+	priv->ofdm_tx_power_diff[0].a =
+		efuse->tx_power_index_A.ht20_ofdm_1s_diff.a;
+	priv->ofdm_tx_power_diff[0].b =
+		efuse->tx_power_index_B.ht20_ofdm_1s_diff.a;
+
+	priv->ht20_tx_power_diff[0].a =
+		efuse->tx_power_index_A.ht20_ofdm_1s_diff.b;
+	priv->ht20_tx_power_diff[0].b =
+		efuse->tx_power_index_B.ht20_ofdm_1s_diff.b;
+
+	priv->ht40_tx_power_diff[0].a = 0;
+	priv->ht40_tx_power_diff[0].b = 0;
+
+	for (i = 1; i < RTL8723B_TX_COUNT; i++) {
+		priv->ofdm_tx_power_diff[i].a =
+			efuse->tx_power_index_A.pwr_diff[i - 1].ofdm;
+		priv->ofdm_tx_power_diff[i].b =
+			efuse->tx_power_index_B.pwr_diff[i - 1].ofdm;
+
+		priv->ht20_tx_power_diff[i].a =
+			efuse->tx_power_index_A.pwr_diff[i - 1].ht20;
+		priv->ht20_tx_power_diff[i].b =
+			efuse->tx_power_index_B.pwr_diff[i - 1].ht20;
+
+		priv->ht40_tx_power_diff[i].a =
+			efuse->tx_power_index_A.pwr_diff[i - 1].ht40;
+		priv->ht40_tx_power_diff[i].b =
+			efuse->tx_power_index_B.pwr_diff[i - 1].ht40;
+	}
+
 	priv->has_xtalk = 1;
 	priv->xtalk = priv->efuse_wifi.efuse8723bu.xtal_k & 0x3f;
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 92768f5..fe23561 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -54,6 +54,7 @@
 #define RTL8723A_CHANNEL_GROUPS		3
 #define RTL8723A_MAX_RF_PATHS		2
 #define RTL8723B_CHANNEL_GROUPS		6
+#define RTL8723B_TX_COUNT		4
 #define RTL8723B_MAX_RF_PATHS		4
 #define RTL8XXXU_MAX_CHANNEL_GROUPS	6
 #define RF6052_MAX_TX_PWR		0x3f
@@ -634,16 +635,25 @@ struct rtl8192cu_efuse {
 	u8 customer_id;
 };
 
+struct rtl8723bu_pwr_idx {
+#ifdef __LITTLE_ENDIAN
+	int	ht20:4;
+	int	ht40:4;
+	int	ofdm:4;
+	int	cck:4;
+#else
+	int	cck:4;
+	int	ofdm:4;
+	int	ht40:4;
+	int	ht20:4;
+#endif
+} __attribute__((packed));
+
 struct rtl8723bu_efuse_tx_power {
 	u8 cck_base[6];
 	u8 ht40_base[5];
 	struct rtl8723au_idx ht20_ofdm_1s_diff;
-	struct rtl8723au_idx ht40_ht20_2s_diff;
-	struct rtl8723au_idx ofdm_cck_2s_diff; /* not used */
-	struct rtl8723au_idx ht40_ht20_3s_diff;
-	struct rtl8723au_idx ofdm_cck_3s_diff; /* not used */
-	struct rtl8723au_idx ht40_ht20_4s_diff;
-	struct rtl8723au_idx ofdm_cck_4s_diff; /* not used */
+	struct rtl8723bu_pwr_idx pwr_diff[3];
 	u8 dummy5g[24]; /* max channel group (14) + power diff offset (10) */
 };
 
@@ -1057,15 +1067,18 @@ struct rtl8xxxu_priv {
 	 * bits 0-3: path A, bits 4-7: path B, all values 4 bits signed
 	 */
 	struct rtl8723au_idx ht40_2s_tx_power_index_diff[
-		RTL8XXXU_MAX_CHANNEL_GROUPS];
-	struct rtl8723au_idx ht20_tx_power_index_diff[
-		RTL8XXXU_MAX_CHANNEL_GROUPS];
-	struct rtl8723au_idx ofdm_tx_power_index_diff[
-		RTL8XXXU_MAX_CHANNEL_GROUPS];
-	struct rtl8723au_idx ht40_max_power_offset[
-		RTL8XXXU_MAX_CHANNEL_GROUPS];
-	struct rtl8723au_idx ht20_max_power_offset[
-		RTL8XXXU_MAX_CHANNEL_GROUPS];
+		RTL8723A_CHANNEL_GROUPS];
+	struct rtl8723au_idx ht20_tx_power_index_diff[RTL8723A_CHANNEL_GROUPS];
+	struct rtl8723au_idx ofdm_tx_power_index_diff[RTL8723A_CHANNEL_GROUPS];
+	struct rtl8723au_idx ht40_max_power_offset[RTL8723A_CHANNEL_GROUPS];
+	struct rtl8723au_idx ht20_max_power_offset[RTL8723A_CHANNEL_GROUPS];
+	/*
+	 * Newer generation chips only keep power diffs per TX count,
+	 * not per channel group.
+	 */
+	struct rtl8723au_idx ofdm_tx_power_diff[RTL8723B_TX_COUNT];
+	struct rtl8723au_idx ht20_tx_power_diff[RTL8723B_TX_COUNT];
+	struct rtl8723au_idx ht40_tx_power_diff[RTL8723B_TX_COUNT];
 	u32 chip_cut:4;
 	u32 rom_rev:4;
 	u32 is_multi_func:1;
-- 
cgit v0.10.2


From 54bed43f3a67351b1ad75ddcad82587d9e5471e0 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:23 -0500
Subject: rtl8xxxu: Set 8723bu TX power for CCK and OFDM rates

This implements support for setting TX power for CCK and OFDM rates on
8723bu. MCS rates is still pending.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 0723008..e800605 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2062,9 +2062,30 @@ rtl8723a_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 static void
 rtl8723b_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 {
-	int group;
+	u32 val32, ofdm;
+	u8 cck, ofdmbase;
+	int group, tx_idx;
 
+	tx_idx = 0;
 	group = rtl8723b_channel_to_group(channel);
+
+	cck = priv->cck_tx_power_index_B[group];
+	val32 = rtl8xxxu_read32(priv, REG_TX_AGC_A_CCK1_MCS32);
+	val32 &= 0xffff00ff;
+	val32 |= (cck << 8);
+	rtl8xxxu_write32(priv, REG_TX_AGC_A_CCK1_MCS32, val32);
+
+	val32 = rtl8xxxu_read32(priv, REG_TX_AGC_B_CCK11_A_CCK2_11);
+	val32 &= 0xff;
+	val32 |= ((cck << 8) | (cck << 16) | (cck << 24));
+	rtl8xxxu_write32(priv, REG_TX_AGC_B_CCK11_A_CCK2_11, val32);
+
+	ofdmbase = priv->ht40_1s_tx_power_index_B[group];
+	ofdmbase += priv->ofdm_tx_power_diff[tx_idx].b;
+	ofdm = ofdmbase | ofdmbase << 8 | ofdmbase << 16 | ofdmbase << 24;
+
+	rtl8xxxu_write32(priv, REG_TX_AGC_A_RATE18_06, ofdm);
+	rtl8xxxu_write32(priv, REG_TX_AGC_A_RATE54_24, ofdm);
 }
 
 static void rtl8xxxu_set_linktype(struct rtl8xxxu_priv *priv,
-- 
cgit v0.10.2


From 1d3cc44dddf8501934172ea141397647f01bb0c6 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:24 -0500
Subject: rtl8xxxu: Set 8723bu MCS TX power

This adds the missing support for setting MCS TX power rates on 8723bu.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e800605..5681339 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2062,8 +2062,8 @@ rtl8723a_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 static void
 rtl8723b_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 {
-	u32 val32, ofdm;
-	u8 cck, ofdmbase;
+	u32 val32, ofdm, mcs;
+	u8 cck, ofdmbase, mcsbase;
 	int group, tx_idx;
 
 	tx_idx = 0;
@@ -2086,6 +2086,16 @@ rtl8723b_set_tx_power(struct rtl8xxxu_priv *priv, int channel, bool ht40)
 
 	rtl8xxxu_write32(priv, REG_TX_AGC_A_RATE18_06, ofdm);
 	rtl8xxxu_write32(priv, REG_TX_AGC_A_RATE54_24, ofdm);
+
+	mcsbase = priv->ht40_1s_tx_power_index_B[group];
+	if (ht40)
+		mcsbase += priv->ht40_tx_power_diff[tx_idx++].b;
+	else
+		mcsbase += priv->ht20_tx_power_diff[tx_idx++].b;
+	mcs = mcsbase | mcsbase << 8 | mcsbase << 16 | mcsbase << 24;
+
+	rtl8xxxu_write32(priv, REG_TX_AGC_A_MCS03_MCS00, mcs);
+	rtl8xxxu_write32(priv, REG_TX_AGC_A_MCS07_MCS04, mcs);
 }
 
 static void rtl8xxxu_set_linktype(struct rtl8xxxu_priv *priv,
-- 
cgit v0.10.2


From 72143b9e947c593555fa2a869a4bc08b9745b105 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:25 -0500
Subject: rtl8xxxu: Set the correct thermal meter register for 8723bu

Older chips use RF register 0x24 to set the thermal meter. Newer chips
use register 0x42.

This change makes sure to set the correct thermal meter register
depending on the chip.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 5681339..241d68d 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6502,7 +6502,11 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	/*
 	 * This should enable thermal meter
 	 */
-	rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_T_METER, 0x60);
+	if (priv->fops->has_s0s1)
+		rtl8xxxu_write_rfreg(priv,
+				     RF_A, RF6052_REG_T_METER_8723B, 0x37cf8);
+	else
+		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_T_METER, 0x60);
 
 	/* Init BT hw config. */
 	if (priv->fops->init_bt)
-- 
cgit v0.10.2


From 80491a1f3c878be44efa9d25fda1bb7b3388ca79 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:26 -0500
Subject: rtl8xxxu: Add definition for 8723bu tx descriptor

Newer generation chips use a 40 byte TX descriptor, compared to the
32 byte descriptor used on older chips.

This adds the definition for the 40 byte descriptor.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index fe23561..f06b88b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -346,6 +346,22 @@ struct rtl8xxxu_tx_desc {
 	__le16 txdw7;
 };
 
+struct rtl8723bu_tx_desc {
+	__le16 pkt_size;
+	u8 pkt_offset;
+	u8 txdw0;
+	__le32 txdw1;
+	__le32 txdw2;
+	__le32 txdw3;
+	__le32 txdw4;
+	__le32 txdw5;
+	__le32 txdw6;
+	__le16 csum;
+	__le16 txdw7;
+	__le32 txdw8;
+	__le32 txdw9;
+};
+
 /*  CCK Rates, TxHT = 0 */
 #define DESC_RATE_1M			0x00
 #define DESC_RATE_2M			0x01
-- 
cgit v0.10.2


From 179e1742569620aed495f5d106d019b2dd6057d1 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:27 -0500
Subject: rtl8xxxu: Handle 40 byte TX descriptors for rtl8723bu

Note the descriptor checksum is still only calculated over the initial
32 bytes of the descriptor, ignoring the last 8 bytes of the
descriptor.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 241d68d..19bc5d7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6787,7 +6787,12 @@ static u32 rtl8xxxu_queue_select(struct ieee80211_hw *hw, struct sk_buff *skb)
 	return queue;
 }
 
-static void rtl8xxxu_calc_tx_desc_csum(struct rtl8xxxu_tx_desc *tx_desc)
+/*
+ * Despite newer chips 8723b/8812/8821 having a larger TX descriptor
+ * format. The descriptor checksum is still only calculated over the
+ * initial 32 bytes of the descriptor!
+ */
+static void rtl8xxxu_calc_tx_desc_csum(struct rtl8723au_tx_desc *tx_desc)
 {
 	__le16 *ptr = (__le16 *)tx_desc;
 	u16 csum = 0;
@@ -6799,7 +6804,7 @@ static void rtl8xxxu_calc_tx_desc_csum(struct rtl8xxxu_tx_desc *tx_desc)
 	 */
 	tx_desc->csum = cpu_to_le16(0);
 
-	for (i = 0; i < (sizeof(struct rtl8xxxu_tx_desc) / sizeof(u16)); i++)
+	for (i = 0; i < (sizeof(struct rtl8723au_tx_desc) / sizeof(u16)); i++)
 		csum = csum ^ le16_to_cpu(ptr[i]);
 
 	tx_desc->csum |= cpu_to_le16(csum);
@@ -6868,13 +6873,15 @@ static void rtl8xxxu_tx_complete(struct urb *urb)
 	struct sk_buff *skb = (struct sk_buff *)urb->context;
 	struct ieee80211_tx_info *tx_info;
 	struct ieee80211_hw *hw;
+	struct rtl8xxxu_priv *priv;
 	struct rtl8xxxu_tx_urb *tx_urb =
 		container_of(urb, struct rtl8xxxu_tx_urb, urb);
 
 	tx_info = IEEE80211_SKB_CB(skb);
 	hw = tx_info->rate_driver_data[0];
+	priv = hw->priv;
 
-	skb_pull(skb, sizeof(struct rtl8xxxu_tx_desc));
+	skb_pull(skb, priv->fops->tx_desc_size);
 
 	ieee80211_tx_info_clear_status(tx_info);
 	tx_info->status.rates[0].idx = -1;
@@ -6885,7 +6892,7 @@ static void rtl8xxxu_tx_complete(struct urb *urb)
 
 	ieee80211_tx_status_irqsafe(hw, skb);
 
-	rtl8xxxu_free_tx_urb(hw->priv, tx_urb);
+	rtl8xxxu_free_tx_urb(priv, tx_urb);
 }
 
 static void rtl8xxxu_dump_action(struct device *dev,
@@ -6935,7 +6942,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	struct ieee80211_tx_info *tx_info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_rate *tx_rate = ieee80211_get_tx_rate(hw, tx_info);
 	struct rtl8xxxu_priv *priv = hw->priv;
-	struct rtl8xxxu_tx_desc *tx_desc;
+	struct rtl8723au_tx_desc *tx_desc;
 	struct rtl8xxxu_tx_urb *tx_urb;
 	struct ieee80211_sta *sta = NULL;
 	struct ieee80211_vif *vif = tx_info->control.vif;
@@ -6944,16 +6951,17 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	u16 pktlen = skb->len;
 	u16 seq_number;
 	u16 rate_flag = tx_info->control.rates[0].flags;
+	int tx_desc_size = priv->fops->tx_desc_size;
 	int ret;
 
-	if (skb_headroom(skb) < sizeof(struct rtl8xxxu_tx_desc)) {
+	if (skb_headroom(skb) < tx_desc_size) {
 		dev_warn(dev,
 			 "%s: Not enough headroom (%i) for tx descriptor\n",
 			 __func__, skb_headroom(skb));
 		goto error;
 	}
 
-	if (unlikely(skb->len > (65535 - sizeof(struct rtl8xxxu_tx_desc)))) {
+	if (unlikely(skb->len > (65535 - tx_desc_size))) {
 		dev_warn(dev, "%s: Trying to send over-sized skb (%i)\n",
 			 __func__, skb->len);
 		goto error;
@@ -6977,12 +6985,11 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	if (control && control->sta)
 		sta = control->sta;
 
-	tx_desc = (struct rtl8xxxu_tx_desc *)
-		skb_push(skb, sizeof(struct rtl8xxxu_tx_desc));
+	tx_desc = (struct rtl8723au_tx_desc *)skb_push(skb, tx_desc_size);
 
-	memset(tx_desc, 0, sizeof(struct rtl8xxxu_tx_desc));
+	memset(tx_desc, 0, tx_desc_size);
 	tx_desc->pkt_size = cpu_to_le16(pktlen);
-	tx_desc->pkt_offset = sizeof(struct rtl8xxxu_tx_desc);
+	tx_desc->pkt_offset = tx_desc_size;
 
 	tx_desc->txdw0 =
 		TXDESC_OWN | TXDESC_FIRST_SEGMENT | TXDESC_LAST_SEGMENT;
@@ -8077,7 +8084,7 @@ static int rtl8xxxu_probe(struct usb_interface *interface,
 	SET_IEEE80211_DEV(priv->hw, &interface->dev);
 	SET_IEEE80211_PERM_ADDR(hw, priv->mac_addr);
 
-	hw->extra_tx_headroom = sizeof(struct rtl8xxxu_tx_desc);
+	hw->extra_tx_headroom = priv->fops->tx_desc_size;
 	ieee80211_hw_set(hw, SIGNAL_DBM);
 	/*
 	 * The firmware handles rate control
@@ -8134,6 +8141,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
+	.tx_desc_size = sizeof(struct rtl8723au_tx_desc),
 	.adda_1t_init = 0x0b1b25a0,
 	.adda_1t_path_on = 0x0bdb25a0,
 	.adda_2t_path_on_a = 0x04db25a4,
@@ -8157,6 +8165,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
+	.tx_desc_size = sizeof(struct rtl8723bu_tx_desc),
 	.has_s0s1 = 1,
 	.adda_1t_init = 0x01c00014,
 	.adda_1t_path_on = 0x01c00014,
@@ -8179,6 +8188,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
+	.tx_desc_size = sizeof(struct rtl8723au_tx_desc),
 	.adda_1t_init = 0x0b1b25a0,
 	.adda_1t_path_on = 0x0bdb25a0,
 	.adda_2t_path_on_a = 0x04db25a4,
@@ -8200,6 +8210,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
+	.tx_desc_size = sizeof(struct rtl8723au_tx_desc),
 	.has_s0s1 = 1,
 	.adda_1t_init = 0x0fc01616,
 	.adda_1t_path_on = 0x0fc01616,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index f06b88b..d53912a 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -332,7 +332,7 @@ struct rtl8723bu_rx_desc {
 	__le32 tsfl;
 };
 
-struct rtl8xxxu_tx_desc {
+struct rtl8723au_tx_desc {
 	__le16 pkt_size;
 	u8 pkt_offset;
 	u8 txdw0;
@@ -1190,6 +1190,7 @@ struct rtl8xxxu_fileops {
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
+	char tx_desc_size;
 	char has_s0s1;
 	u32 adda_1t_init;
 	u32 adda_1t_path_on;
-- 
cgit v0.10.2


From 5e00d5034abf816aa1ca5033ed7c0d6ff04cba87 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:28 -0500
Subject: rtl8xxxu: Do not unconditionally print debug info in
 rtl8723bu_handle_c2h()

Reduce the log level in rtl8723bu_handle_c2h()

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 19bc5d7..49fb5bd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7296,29 +7296,29 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
 
 	len = skb->len - 2;
 
-	dev_info(dev, "C2H ID %02x seq %02x, len %02x source %02x\n",
-		 c2h->id, c2h->seq, len, c2h->bt_info.response_source);
+	dev_dbg(dev, "C2H ID %02x seq %02x, len %02x source %02x\n",
+		c2h->id, c2h->seq, len, c2h->bt_info.response_source);
 
 	switch(c2h->id) {
 	case C2H_8723B_BT_INFO:
 		if (c2h->bt_info.response_source >
 		    BT_INFO_SRC_8723B_BT_ACTIVE_SEND)
-			dev_info(dev, "C2H_BT_INFO WiFi only firmware\n");
+			dev_dbg(dev, "C2H_BT_INFO WiFi only firmware\n");
 		else
-			dev_info(dev, "C2H_BT_INFO BT/WiFi coexist firmware\n");
+			dev_dbg(dev, "C2H_BT_INFO BT/WiFi coexist firmware\n");
 
 		if (c2h->bt_info.bt_has_reset)
-			dev_info(dev, "BT has been reset\n");
+			dev_dbg(dev, "BT has been reset\n");
 		if (c2h->bt_info.tx_rx_mask)
-			dev_info(dev, "BT TRx mask\n");
+			dev_dbg(dev, "BT TRx mask\n");
 
 		break;
 	case C2H_8723B_BT_MP_INFO:
-		dev_info(dev, "C2H_MP_INFO ext ID %02x, status %02x\n",
-			 c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status);
+		dev_dbg(dev, "C2H_MP_INFO ext ID %02x, status %02x\n",
+			c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status);
 		break;
 	default:
-		pr_info("%s: Unhandled C2H event %02x\n", __func__, c2h->id);
+		dev_info(dev, "Unhandled C2H event %02x\n", c2h->id);
 		break;
 	}
 }
-- 
cgit v0.10.2


From 0249258db457b1279f40c7adf46bfb2eea817884 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:29 -0500
Subject: rtl8xxxu: Add additional tx descriptor bits for data word 0

This adds documentation for some additional bits in TX descriptor word
0.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index d53912a..9364a31 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -407,8 +407,12 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_OWN			BIT(31)
 #else
 #define TXDESC_BROADMULTICAST		BIT(0)
+#define TXDESC_HTC			BIT(1)
 #define TXDESC_LAST_SEGMENT		BIT(2)
 #define TXDESC_FIRST_SEGMENT		BIT(3)
+#define TXDESC_LINIP			BIT(4)
+#define TXDESC_NO_ACM			BIT(5)
+#define TXDESC_GF			BIT(6)
 #define TXDESC_OWN			BIT(7)
 #endif
 
-- 
cgit v0.10.2


From ce2d1dbbb4b7b72cfe64d07fbd282f472ea51bbd Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:30 -0500
Subject: rtl8xxxu: Add more 40 byte TX desc bit definitions

Add additional bit definitions for 40 byte TX descriptors, and rename
bits for 32 byte descriptors that are located differently in the 40
byte descriptor format.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 49fb5bd..4cff2c9 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7035,11 +7035,12 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 			ampdu = (u32)sta->ht_cap.ampdu_density;
 			val32 = ampdu << TXDESC_AMPDU_DENSITY_SHIFT;
 			tx_desc->txdw2 |= cpu_to_le32(val32);
-			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_ENABLE);
+
+			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_ENABLE_8723A);
 		} else
-			tx_desc->txdw1 |= cpu_to_le32(TXDESC_BK);
+			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_BREAK_8723A);
 	} else
-		tx_desc->txdw1 |= cpu_to_le32(TXDESC_BK);
+		tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_BREAK_8723A);
 
 	if (ieee80211_is_data_qos(hdr->frame_control))
 		tx_desc->txdw4 |= cpu_to_le32(TXDESC_QOS);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 9364a31..6ba12ab 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -417,9 +417,15 @@ struct rtl8723bu_tx_desc {
 #endif
 
 /* Word 1 */
+/*
+ * Bits 0-7 differ dependent on chip generation. For 8723au bits 5/6 are
+ * aggregation enable and break respectively. For 8723bu, bits 0-7 are macid.
+ */
 #define TXDESC_PKT_OFFSET_SZ		0
-#define TXDESC_AGG_ENABLE		BIT(5)
-#define TXDESC_BK			BIT(6)
+#define TXDESC_AGG_ENABLE_8723A		BIT(5)
+#define TXDESC_AGG_BREAK_8723A		BIT(6)
+#define TXDESC_MACID_SHIFT_8723B	0
+#define TXDESC_MACID_MASK_8723B		0x00f0
 #define TXDESC_QUEUE_SHIFT		8
 #define TXDESC_QUEUE_MASK		0x1f00
 #define TXDESC_QUEUE_BK			0x2
@@ -431,6 +437,9 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_QUEUE_MGNT		0x12
 #define TXDESC_QUEUE_CMD		0x13
 #define TXDESC_QUEUE_MAX		(TXDESC_QUEUE_CMD + 1)
+#define TXDESC_RDG_NAV_EXT_8723B	BIT(13)
+#define TXDESC_LSIG_TXOP_ENABLE_8723B	BIT(14)
+#define TXDESC_PIFS_8723B		BIT(15)
 
 #define DESC_RATE_ID_SHIFT		16
 #define DESC_RATE_ID_MASK		0xf
@@ -442,8 +451,20 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_HWPC			BIT(31)
 
 /* Word 2 */
-#define TXDESC_ACK_REPORT		BIT(19)
+#define TXDESC_PAID_SHIFT_8723B		0
+#define TXDESC_PAID_MASK_8723B		0x1ff
+#define TXDESC_CCA_RTS_SHIFT_8723B	10
+#define TXDESC_CCA_RTS_MASK_8723B	0xc00
+#define TXDESC_AGG_ENABLE_8723B		BIT(12)
+#define TXDESC_RDG_ENABLE_8723B		BIT(13)
+#define TXDESC_AGG_BREAK_8723B		BIT(16)
+#define TXDESC_MORE_FRAG_8723B		BIT(17)
+#define TXDESC_RAW_8723B		BIT(18)
+#define TXDESC_ACK_REPORT_8723A		BIT(19)
+#define TXDESC_SPE_RPT_8723B		BIT(19)
 #define TXDESC_AMPDU_DENSITY_SHIFT	20
+#define TXDESC_BT_INT_8723B		BIT(23)
+#define TXDESC_GID_8723B		BIT(24)
 
 /* Word 3 */
 #define TXDESC_SEQ_SHIFT		16
-- 
cgit v0.10.2


From a40ace4f01c77362553455a819eba407f7fccc76 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:31 -0500
Subject: rtl8xxxu: Set the correct TX descriptor bits for agg and break on
 8723b

Fixup victim of the relocated bits for AGG_ENABLE/AGG_BREAK in the 40
byte TX descriptor

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4cff2c9..dd9addf 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6953,6 +6953,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	u16 rate_flag = tx_info->control.rates[0].flags;
 	int tx_desc_size = priv->fops->tx_desc_size;
 	int ret;
+	bool txdesc40, ampdu_enable;
 
 	if (skb_headroom(skb) < tx_desc_size) {
 		dev_warn(dev,
@@ -6980,6 +6981,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	if (ieee80211_is_action(hdr->frame_control))
 		rtl8xxxu_dump_action(dev, hdr);
 
+	txdesc40 = (tx_desc_size == 40);
 	tx_info->rate_driver_data[0] = hw;
 
 	if (control && control->sta)
@@ -7028,6 +7030,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 		tx_desc->txdw5 |= cpu_to_le32(0x0001ff00);
 
 	/* (tx_info->flags & IEEE80211_TX_CTL_AMPDU) && */
+	ampdu_enable = false;
 	if (ieee80211_is_data_qos(hdr->frame_control) && sta) {
 		if (sta->ht_cap.ht_supported) {
 			u32 ampdu, val32;
@@ -7036,11 +7039,21 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 			val32 = ampdu << TXDESC_AMPDU_DENSITY_SHIFT;
 			tx_desc->txdw2 |= cpu_to_le32(val32);
 
+			ampdu_enable = true;
+		}
+	}
+
+	if (!txdesc40) {
+		if (ampdu_enable)
 			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_ENABLE_8723A);
-		} else
+		else
 			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_BREAK_8723A);
-	} else
-		tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_BREAK_8723A);
+	} else {
+		if (ampdu_enable)
+			tx_desc->txdw2 |= cpu_to_le32(TXDESC_AGG_ENABLE_8723B);
+		else
+			tx_desc->txdw2 |= cpu_to_le32(TXDESC_AGG_BREAK_8723B);
+	};
 
 	if (ieee80211_is_data_qos(hdr->frame_control))
 		tx_desc->txdw4 |= cpu_to_le32(TXDESC_QOS);
-- 
cgit v0.10.2


From cc2646d4bed4f85bbb72871717f86b625677a874 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:32 -0500
Subject: rtl8xxxu: Set sequence number correctly for 40 byte TX descriptors

SEQ changed location in the 40 byte TX descriptor. Set it correctly.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index dd9addf..968eac7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6943,6 +6943,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	struct ieee80211_rate *tx_rate = ieee80211_get_tx_rate(hw, tx_info);
 	struct rtl8xxxu_priv *priv = hw->priv;
 	struct rtl8723au_tx_desc *tx_desc;
+	struct rtl8723bu_tx_desc *tx_desc40;
 	struct rtl8xxxu_tx_urb *tx_urb;
 	struct ieee80211_sta *sta = NULL;
 	struct ieee80211_vif *vif = tx_info->control.vif;
@@ -6953,7 +6954,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	u16 rate_flag = tx_info->control.rates[0].flags;
 	int tx_desc_size = priv->fops->tx_desc_size;
 	int ret;
-	bool txdesc40, ampdu_enable;
+	bool usedesc40, ampdu_enable;
 
 	if (skb_headroom(skb) < tx_desc_size) {
 		dev_warn(dev,
@@ -6981,7 +6982,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	if (ieee80211_is_action(hdr->frame_control))
 		rtl8xxxu_dump_action(dev, hdr);
 
-	txdesc40 = (tx_desc_size == 40);
+	usedesc40 = (tx_desc_size == 40);
 	tx_info->rate_driver_data[0] = hw;
 
 	if (control && control->sta)
@@ -7017,9 +7018,6 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 		}
 	}
 
-	seq_number = IEEE80211_SEQ_TO_SN(le16_to_cpu(hdr->seq_ctrl));
-	tx_desc->txdw3 = cpu_to_le32((u32)seq_number << TXDESC_SEQ_SHIFT);
-
 	if (rate_flag & IEEE80211_TX_RC_MCS)
 		rate = tx_info->control.rates[0].idx + DESC_RATE_MCS0;
 	else
@@ -7043,16 +7041,26 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 		}
 	}
 
-	if (!txdesc40) {
+	seq_number = IEEE80211_SEQ_TO_SN(le16_to_cpu(hdr->seq_ctrl));
+	if (!usedesc40) {
+		tx_desc->txdw3 =
+			cpu_to_le32((u32)seq_number << TXDESC_SEQ_SHIFT_8723A);
+
 		if (ampdu_enable)
 			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_ENABLE_8723A);
 		else
 			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_BREAK_8723A);
 	} else {
+		tx_desc40 = (struct rtl8723bu_tx_desc *)tx_desc;
+
+		tx_desc40->txdw9 =
+			cpu_to_le32((u32)seq_number << TXDESC_SEQ_SHIFT_8723B);
+
 		if (ampdu_enable)
-			tx_desc->txdw2 |= cpu_to_le32(TXDESC_AGG_ENABLE_8723B);
+			tx_desc40->txdw2 |=
+				cpu_to_le32(TXDESC_AGG_ENABLE_8723B);
 		else
-			tx_desc->txdw2 |= cpu_to_le32(TXDESC_AGG_BREAK_8723B);
+			tx_desc40->txdw2 |= cpu_to_le32(TXDESC_AGG_BREAK_8723B);
 	};
 
 	if (ieee80211_is_data_qos(hdr->frame_control))
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 6ba12ab..8f95896 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -467,8 +467,8 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_GID_8723B		BIT(24)
 
 /* Word 3 */
-#define TXDESC_SEQ_SHIFT		16
-#define TXDESC_SEQ_MASK			0x0fff0000
+#define TXDESC_SEQ_SHIFT_8723A		16
+#define TXDESC_SEQ_MASK_8723A		0x0fff0000
 
 /* Word 4 */
 #define TXDESC_QOS			BIT(6)
@@ -498,6 +498,10 @@ struct rtl8723bu_tx_desc {
 /* Word 6 */
 #define TXDESC_MAX_AGG_SHIFT		11
 
+/* Word 9 */
+#define TXDESC_SEQ_SHIFT_8723B		12
+#define TXDESC_SEQ_MASK_8723B		0x00fff000
+
 struct phy_rx_agc_info {
 #ifdef __LITTLE_ENDIAN
 	u8	gain:7, trsw:1;
-- 
cgit v0.10.2


From 2098bfb5f3277b12dc2f9dc61efd57161c000f77 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:33 -0500
Subject: rtl8723au: Update TX descriptor words 4 and 5 definitions

TX data words 4 and 5 differ significantly between 32 byte and 40 byte
descriptors.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 968eac7..2fa587e 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7076,16 +7076,17 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 	}
 	if (ieee80211_is_mgmt(hdr->frame_control)) {
 		tx_desc->txdw5 = cpu_to_le32(tx_rate->hw_value);
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_USE_DRIVER_RATE);
-		tx_desc->txdw5 |= cpu_to_le32(6 << TXDESC_RETRY_LIMIT_SHIFT);
-		tx_desc->txdw5 |= cpu_to_le32(TXDESC_RETRY_LIMIT_ENABLE);
+		tx_desc->txdw4 |= cpu_to_le32(TXDESC_USE_DRIVER_RATE_8723A);
+		tx_desc->txdw5 |=
+			cpu_to_le32(6 << TXDESC_RETRY_LIMIT_SHIFT_8723A);
+		tx_desc->txdw5 |= cpu_to_le32(TXDESC_RETRY_LIMIT_ENABLE_8723A);
 	}
 
 	if (rate_flag & IEEE80211_TX_RC_USE_RTS_CTS) {
 		/* Use RTS rate 24M - does the mac80211 tell us which to use? */
 		tx_desc->txdw4 |= cpu_to_le32(DESC_RATE_24M);
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_RTS_CTS_ENABLE);
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_HW_RTS_ENABLE);
+		tx_desc->txdw4 |= cpu_to_le32(TXDESC_RTS_CTS_ENABLE_8723A);
+		tx_desc->txdw4 |= cpu_to_le32(TXDESC_HW_RTS_ENABLE_8723A);
 	}
 
 	rtl8xxxu_calc_tx_desc_csum(tx_desc);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 8f95896..5f60f19 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -467,17 +467,21 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_GID_8723B		BIT(24)
 
 /* Word 3 */
+#define TXDESC_USE_DRIVER_RATE_8723B	BIT(8)
+#define TXDESC_CTS_SELF_ENABLE_8723B	BIT(11)
+#define TXDESC_RTS_CTS_ENABLE_8723B	BIT(12)
+#define TXDESC_HW_RTS_ENABLE_8723B	BIT(13)
 #define TXDESC_SEQ_SHIFT_8723A		16
 #define TXDESC_SEQ_MASK_8723A		0x0fff0000
 
 /* Word 4 */
 #define TXDESC_QOS			BIT(6)
-#define TXDESC_HW_SEQ_ENABLE		BIT(7)
-#define TXDESC_USE_DRIVER_RATE		BIT(8)
+#define TXDESC_HW_SEQ_ENABLE_8723A	BIT(7)
+#define TXDESC_USE_DRIVER_RATE_8723A	BIT(8)
 #define TXDESC_DISABLE_DATA_FB		BIT(10)
-#define TXDESC_CTS_SELF_ENABLE		BIT(11)
-#define TXDESC_RTS_CTS_ENABLE		BIT(12)
-#define TXDESC_HW_RTS_ENABLE		BIT(13)
+#define TXDESC_CTS_SELF_ENABLE_8723A	BIT(11)
+#define TXDESC_RTS_CTS_ENABLE_8723A	BIT(12)
+#define TXDESC_HW_RTS_ENABLE_8723A	BIT(13)
 #define TXDESC_PRIME_CH_OFF_LOWER	BIT(20)
 #define TXDESC_PRIME_CH_OFF_UPPER	BIT(21)
 #define TXDESC_SHORT_PREAMBLE		BIT(24)
@@ -485,19 +489,27 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_RTS_DATA_BW		BIT(27)
 #define TXDESC_RTS_PRIME_CH_OFF_LOWER	BIT(28)
 #define TXDESC_RTS_PRIME_CH_OFF_UPPER	BIT(29)
+#define TXDESC_RETRY_LIMIT_ENABLE_8723B	BIT(17)
+#define TXDESC_RETRY_LIMIT_SHIFT_8723B	18
+#define TXDESC_RETRY_LIMIT_MASK_8723B	0x00fc0000
+#define TXDESC_RTS_RATE_SHIFT_8723B	24
+#define TXDESC_RTS_RATE_MASK_8723B	0x3f000000
 
 /* Word 5 */
-#define TXDESC_RTS_RATE_SHIFT		0
-#define TXDESC_RTS_RATE_MASK		0x3f
+#define TXDESC_RTS_RATE_SHIFT_8723A	0
+#define TXDESC_RTS_RATE_MASK_8723A	0x3f
 #define TXDESC_SHORT_GI			BIT(6)
 #define TXDESC_CCX_TAG			BIT(7)
-#define TXDESC_RETRY_LIMIT_ENABLE	BIT(17)
-#define TXDESC_RETRY_LIMIT_SHIFT	18
-#define TXDESC_RETRY_LIMIT_MASK		0x00fc0000
+#define TXDESC_RETRY_LIMIT_ENABLE_8723A	BIT(17)
+#define TXDESC_RETRY_LIMIT_SHIFT_8723A	18
+#define TXDESC_RETRY_LIMIT_MASK_8723A	0x00fc0000
 
 /* Word 6 */
 #define TXDESC_MAX_AGG_SHIFT		11
 
+/* Word 8 */
+#define TXDESC_HW_SEQ_ENABLE_8723B	BIT(15)
+
 /* Word 9 */
 #define TXDESC_SEQ_SHIFT_8723B		12
 #define TXDESC_SEQ_MASK_8723B		0x00fff000
-- 
cgit v0.10.2


From 2c6670b2a8e6dcd17bedd4dfbac663b2aa3d6949 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:34 -0500
Subject: rtl8xxxu: TX RTS rate is word 4 for 8723a

Correct the setting of TX RTS for 8723a generation chips. In addition
update documentation to match that this is part of data word 4, note
data word 5.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 2fa587e..4fbf445 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7084,7 +7084,8 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 
 	if (rate_flag & IEEE80211_TX_RC_USE_RTS_CTS) {
 		/* Use RTS rate 24M - does the mac80211 tell us which to use? */
-		tx_desc->txdw4 |= cpu_to_le32(DESC_RATE_24M);
+		tx_desc->txdw4 |= cpu_to_le32(DESC_RATE_24M <<
+					      TXDESC_RTS_RATE_SHIFT_8723A);
 		tx_desc->txdw4 |= cpu_to_le32(TXDESC_RTS_CTS_ENABLE_8723A);
 		tx_desc->txdw4 |= cpu_to_le32(TXDESC_HW_RTS_ENABLE_8723A);
 	}
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 5f60f19..b015c75 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -475,6 +475,8 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_SEQ_MASK_8723A		0x0fff0000
 
 /* Word 4 */
+#define TXDESC_RTS_RATE_SHIFT_8723A	0
+#define TXDESC_RTS_RATE_MASK_8723A	0x3f
 #define TXDESC_QOS			BIT(6)
 #define TXDESC_HW_SEQ_ENABLE_8723A	BIT(7)
 #define TXDESC_USE_DRIVER_RATE_8723A	BIT(8)
@@ -496,8 +498,6 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_RTS_RATE_MASK_8723B	0x3f000000
 
 /* Word 5 */
-#define TXDESC_RTS_RATE_SHIFT_8723A	0
-#define TXDESC_RTS_RATE_MASK_8723A	0x3f
 #define TXDESC_SHORT_GI			BIT(6)
 #define TXDESC_CCX_TAG			BIT(7)
 #define TXDESC_RETRY_LIMIT_ENABLE_8723A	BIT(17)
-- 
cgit v0.10.2


From 4c683607145d22c872d6b473c5f5465f8459c5de Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:35 -0500
Subject: rtl8xxxu: Improve handling of txdesc32 vs txdesc40 handling

Further correct the handling of 40 byte TX descriptors.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4fbf445..927fd3b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7018,15 +7018,6 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 		}
 	}
 
-	if (rate_flag & IEEE80211_TX_RC_MCS)
-		rate = tx_info->control.rates[0].idx + DESC_RATE_MCS0;
-	else
-		rate = tx_rate->hw_value;
-	tx_desc->txdw5 = cpu_to_le32(rate);
-
-	if (ieee80211_is_data(hdr->frame_control))
-		tx_desc->txdw5 |= cpu_to_le32(0x0001ff00);
-
 	/* (tx_info->flags & IEEE80211_TX_CTL_AMPDU) && */
 	ampdu_enable = false;
 	if (ieee80211_is_data_qos(hdr->frame_control) && sta) {
@@ -7041,8 +7032,18 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 		}
 	}
 
+	if (rate_flag & IEEE80211_TX_RC_MCS)
+		rate = tx_info->control.rates[0].idx + DESC_RATE_MCS0;
+	else
+		rate = tx_rate->hw_value;
+
 	seq_number = IEEE80211_SEQ_TO_SN(le16_to_cpu(hdr->seq_ctrl));
 	if (!usedesc40) {
+		tx_desc->txdw5 = cpu_to_le32(rate);
+
+		if (ieee80211_is_data(hdr->frame_control))
+			tx_desc->txdw5 |= cpu_to_le32(0x0001ff00);
+
 		tx_desc->txdw3 =
 			cpu_to_le32((u32)seq_number << TXDESC_SEQ_SHIFT_8723A);
 
@@ -7050,9 +7051,56 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_ENABLE_8723A);
 		else
 			tx_desc->txdw1 |= cpu_to_le32(TXDESC_AGG_BREAK_8723A);
+
+		if (ieee80211_is_mgmt(hdr->frame_control)) {
+			tx_desc->txdw5 = cpu_to_le32(tx_rate->hw_value);
+			tx_desc->txdw4 |=
+				cpu_to_le32(TXDESC_USE_DRIVER_RATE_8723A);
+			tx_desc->txdw5 |=
+				cpu_to_le32(6 <<
+					    TXDESC_RETRY_LIMIT_SHIFT_8723A);
+			tx_desc->txdw5 |=
+				cpu_to_le32(TXDESC_RETRY_LIMIT_ENABLE_8723A);
+		}
+
+		if (ieee80211_is_data_qos(hdr->frame_control))
+			tx_desc->txdw4 |= cpu_to_le32(TXDESC_QOS_8723A);
+
+		if (rate_flag & IEEE80211_TX_RC_USE_SHORT_PREAMBLE ||
+		    (sta && vif && vif->bss_conf.use_short_preamble))
+			tx_desc->txdw4 |=
+				cpu_to_le32(TXDESC_SHORT_PREAMBLE_8723A);
+
+		if (rate_flag & IEEE80211_TX_RC_SHORT_GI ||
+		    (ieee80211_is_data_qos(hdr->frame_control) &&
+		     sta && sta->ht_cap.cap &
+		     (IEEE80211_HT_CAP_SGI_40 | IEEE80211_HT_CAP_SGI_20))) {
+			tx_desc->txdw5 |= cpu_to_le32(TXDESC_SHORT_GI);
+		}
+
+		if (rate_flag & IEEE80211_TX_RC_USE_RTS_CTS) {
+			/*
+			 * Use RTS rate 24M - does the mac80211 tell
+			 * us which to use?
+			 */
+			tx_desc->txdw4 |=
+				cpu_to_le32(DESC_RATE_24M <<
+					    TXDESC_RTS_RATE_SHIFT_8723A);
+			tx_desc->txdw4 |=
+				cpu_to_le32(TXDESC_RTS_CTS_ENABLE_8723A);
+			tx_desc->txdw4 |=
+				cpu_to_le32(TXDESC_HW_RTS_ENABLE_8723A);
+		}
 	} else {
 		tx_desc40 = (struct rtl8723bu_tx_desc *)tx_desc;
 
+		tx_desc40->txdw4 = cpu_to_le32(rate);
+		if (ieee80211_is_data(hdr->frame_control)) {
+			tx_desc->txdw4 |=
+				cpu_to_le32(0x1f <<
+					    TXDESC_DATA_RATE_FB_SHIFT_8723B);
+		}
+
 		tx_desc40->txdw9 =
 			cpu_to_le32((u32)seq_number << TXDESC_SEQ_SHIFT_8723B);
 
@@ -7061,34 +7109,37 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 				cpu_to_le32(TXDESC_AGG_ENABLE_8723B);
 		else
 			tx_desc40->txdw2 |= cpu_to_le32(TXDESC_AGG_BREAK_8723B);
-	};
 
-	if (ieee80211_is_data_qos(hdr->frame_control))
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_QOS);
-	if (rate_flag & IEEE80211_TX_RC_USE_SHORT_PREAMBLE ||
-	    (sta && vif && vif->bss_conf.use_short_preamble))
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_SHORT_PREAMBLE);
-	if (rate_flag & IEEE80211_TX_RC_SHORT_GI ||
-	    (ieee80211_is_data_qos(hdr->frame_control) &&
-	     sta && sta->ht_cap.cap &
-	     (IEEE80211_HT_CAP_SGI_40 | IEEE80211_HT_CAP_SGI_20))) {
-		tx_desc->txdw5 |= cpu_to_le32(TXDESC_SHORT_GI);
-	}
-	if (ieee80211_is_mgmt(hdr->frame_control)) {
-		tx_desc->txdw5 = cpu_to_le32(tx_rate->hw_value);
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_USE_DRIVER_RATE_8723A);
-		tx_desc->txdw5 |=
-			cpu_to_le32(6 << TXDESC_RETRY_LIMIT_SHIFT_8723A);
-		tx_desc->txdw5 |= cpu_to_le32(TXDESC_RETRY_LIMIT_ENABLE_8723A);
-	}
-
-	if (rate_flag & IEEE80211_TX_RC_USE_RTS_CTS) {
-		/* Use RTS rate 24M - does the mac80211 tell us which to use? */
-		tx_desc->txdw4 |= cpu_to_le32(DESC_RATE_24M <<
-					      TXDESC_RTS_RATE_SHIFT_8723A);
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_RTS_CTS_ENABLE_8723A);
-		tx_desc->txdw4 |= cpu_to_le32(TXDESC_HW_RTS_ENABLE_8723A);
-	}
+		if (ieee80211_is_mgmt(hdr->frame_control)) {
+			tx_desc40->txdw4 = cpu_to_le32(tx_rate->hw_value);
+			tx_desc40->txdw3 |=
+				cpu_to_le32(TXDESC_USE_DRIVER_RATE_8723B);
+			tx_desc40->txdw4 |=
+				cpu_to_le32(6 <<
+					    TXDESC_RETRY_LIMIT_SHIFT_8723B);
+			tx_desc40->txdw4 |=
+				cpu_to_le32(TXDESC_RETRY_LIMIT_ENABLE_8723B);
+		}
+
+		if (rate_flag & IEEE80211_TX_RC_USE_SHORT_PREAMBLE ||
+		    (sta && vif && vif->bss_conf.use_short_preamble))
+			tx_desc40->txdw5 |=
+				cpu_to_le32(TXDESC_SHORT_PREAMBLE_8723B);
+
+		if (rate_flag & IEEE80211_TX_RC_USE_RTS_CTS) {
+			/*
+			 * Use RTS rate 24M - does the mac80211 tell
+			 * us which to use?
+			 */
+			tx_desc->txdw4 |=
+				cpu_to_le32(DESC_RATE_24M <<
+					    TXDESC_RTS_RATE_SHIFT_8723B);
+			tx_desc->txdw3 |=
+				cpu_to_le32(TXDESC_RTS_CTS_ENABLE_8723B);
+			tx_desc->txdw3 |=
+				cpu_to_le32(TXDESC_HW_RTS_ENABLE_8723B);
+		}
+	};
 
 	rtl8xxxu_calc_tx_desc_csum(tx_desc);
 
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index b015c75..687b9a9 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -477,7 +477,7 @@ struct rtl8723bu_tx_desc {
 /* Word 4 */
 #define TXDESC_RTS_RATE_SHIFT_8723A	0
 #define TXDESC_RTS_RATE_MASK_8723A	0x3f
-#define TXDESC_QOS			BIT(6)
+#define TXDESC_QOS_8723A		BIT(6)
 #define TXDESC_HW_SEQ_ENABLE_8723A	BIT(7)
 #define TXDESC_USE_DRIVER_RATE_8723A	BIT(8)
 #define TXDESC_DISABLE_DATA_FB		BIT(10)
@@ -486,11 +486,13 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_HW_RTS_ENABLE_8723A	BIT(13)
 #define TXDESC_PRIME_CH_OFF_LOWER	BIT(20)
 #define TXDESC_PRIME_CH_OFF_UPPER	BIT(21)
-#define TXDESC_SHORT_PREAMBLE		BIT(24)
+#define TXDESC_SHORT_PREAMBLE_8723A	BIT(24)
 #define TXDESC_DATA_BW			BIT(25)
 #define TXDESC_RTS_DATA_BW		BIT(27)
 #define TXDESC_RTS_PRIME_CH_OFF_LOWER	BIT(28)
 #define TXDESC_RTS_PRIME_CH_OFF_UPPER	BIT(29)
+#define TXDESC_DATA_RATE_FB_SHIFT_8723B	8
+#define TXDESC_DATA_RATE_FB_MASK_8723B	0x00001f00
 #define TXDESC_RETRY_LIMIT_ENABLE_8723B	BIT(17)
 #define TXDESC_RETRY_LIMIT_SHIFT_8723B	18
 #define TXDESC_RETRY_LIMIT_MASK_8723B	0x00fc0000
@@ -498,6 +500,7 @@ struct rtl8723bu_tx_desc {
 #define TXDESC_RTS_RATE_MASK_8723B	0x3f000000
 
 /* Word 5 */
+#define TXDESC_SHORT_PREAMBLE_8723B	BIT(4)
 #define TXDESC_SHORT_GI			BIT(6)
 #define TXDESC_CCX_TAG			BIT(7)
 #define TXDESC_RETRY_LIMIT_ENABLE_8723A	BIT(17)
-- 
cgit v0.10.2


From e975b87cafd24d943f395823b1a000d1389f48ef Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:36 -0500
Subject: rtl8xxxu: Do not parse RX descriptor info for C2H packets

C2H events are delivered as RX packets on 8723bu/8192eu. When
receiving a C2H event, do not parse the rest of the RX descriptor as
the info isn't valid.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 927fd3b..76793e9 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7324,7 +7324,6 @@ static int rtl8723bu_parse_rx_desc(struct rtl8xxxu_priv *priv,
 		(struct rtl8723bu_rx_desc *)skb->data;
 	struct rtl8723au_phy_stats *phy_stats;
 	int drvinfo_sz, desc_shift;
-	int rx_type;
 
 	skb_pull(skb, sizeof(struct rtl8723bu_rx_desc));
 
@@ -7334,6 +7333,12 @@ static int rtl8723bu_parse_rx_desc(struct rtl8xxxu_priv *priv,
 	desc_shift = rx_desc->shift;
 	skb_pull(skb, drvinfo_sz + desc_shift);
 
+	if (rx_desc->rpt_sel) {
+		struct device *dev = &priv->udev->dev;
+		dev_dbg(dev, "%s: C2H packet\n", __func__);
+		return RX_TYPE_C2H;
+	}
+
 	rx_status->mactime = le32_to_cpu(rx_desc->tsfl);
 	rx_status->flag |= RX_FLAG_MACTIME_START;
 
@@ -7351,15 +7356,7 @@ static int rtl8723bu_parse_rx_desc(struct rtl8xxxu_priv *priv,
 		rx_status->rate_idx = rx_desc->rxmcs;
 	}
 
-	if (rx_desc->rpt_sel) {
-		struct device *dev = &priv->udev->dev;
-		dev_dbg(dev, "%s: C2H packet\n", __func__);
-		rx_type = RX_TYPE_C2H;
-	} else {
-		rx_type = RX_TYPE_DATA_PKT;
-	}
-
-	return rx_type;
+	return RX_TYPE_DATA_PKT;
 }
 
 static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
-- 
cgit v0.10.2


From 80b30b2af59af9400568af05b8b409e5a20da930 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:37 -0500
Subject: rtl8xxxu: Define 8723b H2C ramask command structure

Define H2C command structure for setting the rate mask.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 687b9a9..37657fe 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -893,7 +893,7 @@ enum h2c_cmd_8723b {
 	/*
 	 * Dynamic Mechanism Class: 010
 	 */
-	H2C_8723B_MACID_CFG = 0x40,
+	H2C_8723B_MACID_CFG_RAID = 0x40,
 	H2C_8723B_TXBF = 0x41,
 	H2C_8723B_RSSI_SETTING = 0x42,
 	H2C_8723B_AP_REQ_TXRPT = 0x43,
@@ -960,6 +960,15 @@ struct h2c_cmd {
 			u8 cmd;
 			u8 data1;
 			u8 data2;
+			u8 ramask1;
+			u8 ramask2;
+			u8 ramask3;
+			u8 ramask4;
+		} __packed b_macid_cfg;
+		struct {
+			u8 cmd;
+			u8 data1;
+			u8 data2;
 			u8 data3;
 			u8 data4;
 			u8 data5;
-- 
cgit v0.10.2


From f653e69009c633d099b34f7473b037c570c256f8 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:38 -0500
Subject: rtl8xxxu: Implement basic 8723b specific update_rate_mask() function

Support for setting bandwidth and VHT parameters is still missing

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 76793e9..5486ad7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6609,11 +6609,13 @@ static void rtl8xxxu_sw_scan_complete(struct ieee80211_hw *hw,
 	rtl8xxxu_write8(priv, REG_BEACON_CTRL, val8);
 }
 
-static void rtl8xxxu_update_rate_mask(struct rtl8xxxu_priv *priv,
-				      u32 ramask, int sgi)
+static void rtl8723au_update_rate_mask(struct rtl8xxxu_priv *priv,
+				       u32 ramask, int sgi)
 {
 	struct h2c_cmd h2c;
 
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+
 	h2c.ramask.cmd = H2C_SET_RATE_MASK;
 	h2c.ramask.mask_lo = cpu_to_le16(ramask & 0xffff);
 	h2c.ramask.mask_hi = cpu_to_le16(ramask >> 16);
@@ -6627,6 +6629,32 @@ static void rtl8xxxu_update_rate_mask(struct rtl8xxxu_priv *priv,
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ramask));
 }
 
+static void rtl8723bu_update_rate_mask(struct rtl8xxxu_priv *priv,
+				       u32 ramask, int sgi)
+{
+	struct h2c_cmd h2c;
+	u8 bw = 0;
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+
+	h2c.b_macid_cfg.cmd = H2C_8723B_MACID_CFG_RAID;
+	h2c.b_macid_cfg.ramask0 = ramask & 0xff;
+	h2c.b_macid_cfg.ramask1 = (ramask >> 8) & 0xff;
+	h2c.b_macid_cfg.ramask2 = (ramask >> 16) & 0xff;
+	h2c.b_macid_cfg.ramask3 = (ramask >> 24) & 0xff;
+
+	h2c.ramask.arg = 0x80;
+	h2c.b_macid_cfg.data1 = 0;
+	if (sgi)
+		h2c.b_macid_cfg.data1 |= BIT(7);
+
+	h2c.b_macid_cfg.data2 = bw;
+
+	dev_dbg(&priv->udev->dev, "%s: rate mask %08x, arg %02x, size %zi\n",
+		__func__, ramask, h2c.ramask.arg, sizeof(h2c.b_macid_cfg));
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.b_macid_cfg));
+}
+
 static void rtl8xxxu_set_basic_rates(struct rtl8xxxu_priv *priv, u32 rate_cfg)
 {
 	u32 val32;
@@ -6693,7 +6721,7 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 				sgi = 1;
 			rcu_read_unlock();
 
-			rtl8xxxu_update_rate_mask(priv, ramask, sgi);
+			priv->fops->update_rate_mask(priv, ramask, sgi);
 
 			rtl8xxxu_write8(priv, REG_BCN_MAX_ERR, 0xff);
 
@@ -8210,6 +8238,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.enable_rf = rtl8723a_enable_rf,
 	.set_tx_power = rtl8723a_set_tx_power,
+	.update_rate_mask = rtl8723au_update_rate_mask,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8234,6 +8263,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.init_statistics = rtl8723bu_init_statistics,
 	.enable_rf = rtl8723b_enable_rf,
 	.set_tx_power = rtl8723b_set_tx_power,
+	.update_rate_mask = rtl8723bu_update_rate_mask,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -8257,6 +8287,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.enable_rf = rtl8723a_enable_rf,
 	.set_tx_power = rtl8723a_set_tx_power,
+	.update_rate_mask = rtl8723au_update_rate_mask,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8279,6 +8310,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.enable_rf = rtl8723b_enable_rf,
 	.set_tx_power = rtl8723b_set_tx_power,
+	.update_rate_mask = rtl8723au_update_rate_mask,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 37657fe..eea62f1 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -958,12 +958,24 @@ struct h2c_cmd {
 		} __packed ramask;
 		struct {
 			u8 cmd;
+			u8 macid;
+			/*
+			 * [0:4] - RAID
+			 * [7]   - SGI
+			 */
 			u8 data1;
+			/*
+			 * [0:1] - Bandwidth
+			 * [3]   - No Update
+			 * [4:5] - VHT enable
+			 * [6]   - DISPT
+			 * [7]   - DISRA
+			 */
 			u8 data2;
+			u8 ramask0;
 			u8 ramask1;
 			u8 ramask2;
 			u8 ramask3;
-			u8 ramask4;
 		} __packed b_macid_cfg;
 		struct {
 			u8 cmd;
@@ -1240,6 +1252,8 @@ struct rtl8xxxu_fileops {
 	void (*enable_rf) (struct rtl8xxxu_priv *priv);
 	void (*set_tx_power) (struct rtl8xxxu_priv *priv, int channel,
 			      bool ht40);
+	void (*update_rate_mask) (struct rtl8xxxu_priv *priv,
+				  u32 ramask, int sgi);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
-- 
cgit v0.10.2


From 7d794eaa857e9475e8b02f5a53100cad569194e3 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:39 -0500
Subject: rtl8xxxu: Report media status using the correct H2C command for
 8723bu

Implement support for nextgen devices reporting connectition to the
firmware.

The H2C API for reporting connection to the firmware is different
between the two device generations. Use the fileops structure to
determine which one to call.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 5486ad7..3441361 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6655,6 +6655,39 @@ static void rtl8723bu_update_rate_mask(struct rtl8xxxu_priv *priv,
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.b_macid_cfg));
 }
 
+static void rtl8723au_report_connect(struct rtl8xxxu_priv *priv,
+				     u8 macid, bool connect)
+{
+	struct h2c_cmd h2c;
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+
+	h2c.joinbss.cmd = H2C_JOIN_BSS_REPORT;
+
+	if (connect)
+		h2c.joinbss.data = H2C_JOIN_BSS_CONNECT;
+	else
+		h2c.joinbss.data = H2C_JOIN_BSS_DISCONNECT;
+
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.joinbss));
+}
+
+static void rtl8723bu_report_connect(struct rtl8xxxu_priv *priv,
+				     u8 macid, bool connect)
+{
+	struct h2c_cmd h2c;
+
+	memset(&h2c, 0, sizeof(struct h2c_cmd));
+
+	h2c.media_status_rpt.cmd = H2C_8723B_MEDIA_STATUS_RPT;
+	if (connect)
+		h2c.media_status_rpt.parm |= BIT(0);
+	else
+		h2c.media_status_rpt.parm &= ~BIT(0);
+
+	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.media_status_rpt));
+}
+
 static void rtl8xxxu_set_basic_rates(struct rtl8xxxu_priv *priv, u32 rate_cfg)
 {
 	u32 val32;
@@ -6687,11 +6720,8 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	u8 val8;
 
 	if (changed & BSS_CHANGED_ASSOC) {
-		struct h2c_cmd h2c;
-
 		dev_dbg(dev, "Changed ASSOC: %i!\n", bss_conf->assoc);
 
-		memset(&h2c, 0, sizeof(struct h2c_cmd));
 		rtl8xxxu_set_linktype(priv, vif->type);
 
 		if (bss_conf->assoc) {
@@ -6731,16 +6761,14 @@ rtl8xxxu_bss_info_changed(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			rtl8xxxu_write16(priv, REG_BCN_PSR_RPT,
 					 0xc000 | bss_conf->aid);
 
-			h2c.joinbss.data = H2C_JOIN_BSS_CONNECT;
+			priv->fops->report_connect(priv, 0, true);
 		} else {
 			val8 = rtl8xxxu_read8(priv, REG_BEACON_CTRL);
 			val8 |= BEACON_DISABLE_TSF_UPDATE;
 			rtl8xxxu_write8(priv, REG_BEACON_CTRL, val8);
 
-			h2c.joinbss.data = H2C_JOIN_BSS_DISCONNECT;
+			priv->fops->report_connect(priv, 0, false);
 		}
-		h2c.joinbss.cmd = H2C_JOIN_BSS_REPORT;
-		rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.joinbss));
 	}
 
 	if (changed & BSS_CHANGED_ERP_PREAMBLE) {
@@ -8239,6 +8267,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.enable_rf = rtl8723a_enable_rf,
 	.set_tx_power = rtl8723a_set_tx_power,
 	.update_rate_mask = rtl8723au_update_rate_mask,
+	.report_connect = rtl8723au_report_connect,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8264,6 +8293,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.enable_rf = rtl8723b_enable_rf,
 	.set_tx_power = rtl8723b_set_tx_power,
 	.update_rate_mask = rtl8723bu_update_rate_mask,
+	.report_connect = rtl8723bu_report_connect,
 	.writeN_block_size = 1024,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
@@ -8288,6 +8318,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.enable_rf = rtl8723a_enable_rf,
 	.set_tx_power = rtl8723a_set_tx_power,
 	.update_rate_mask = rtl8723au_update_rate_mask,
+	.report_connect = rtl8723au_report_connect,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT_0,
 	.mbox_ext_width = 2,
@@ -8311,6 +8342,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.enable_rf = rtl8723b_enable_rf,
 	.set_tx_power = rtl8723b_set_tx_power,
 	.update_rate_mask = rtl8723au_update_rate_mask,
+	.report_connect = rtl8723au_report_connect,
 	.writeN_block_size = 128,
 	.mbox_ext_reg = REG_HMBOX_EXT0_8723B,
 	.mbox_ext_width = 4,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index eea62f1..2191b6b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -958,6 +958,12 @@ struct h2c_cmd {
 		} __packed ramask;
 		struct {
 			u8 cmd;
+			u8 parm;
+			u8 macid;
+			u8 macid_end;
+		} __packed media_status_rpt;
+		struct {
+			u8 cmd;
 			u8 macid;
 			/*
 			 * [0:4] - RAID
@@ -1254,6 +1260,8 @@ struct rtl8xxxu_fileops {
 			      bool ht40);
 	void (*update_rate_mask) (struct rtl8xxxu_priv *priv,
 				  u32 ramask, int sgi);
+	void (*report_connect) (struct rtl8xxxu_priv *priv,
+				u8 macid, bool connect);
 	int writeN_block_size;
 	u16 mbox_ext_reg;
 	char mbox_ext_width;
-- 
cgit v0.10.2


From 739dc9f2f5c01dafc7336ffebd9d36bba4f9174e Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:40 -0500
Subject: rtl8xxxu: Dump contents of unhandled C2H events

Dump the contents of unhandled C2H events. We should be handling all
expected events, so this is debugging help in case an unexpected event
happens.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 3441361..07f4ee0 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7446,7 +7446,10 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
 			c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status);
 		break;
 	default:
-		dev_info(dev, "Unhandled C2H event %02x\n", c2h->id);
+		dev_info(dev, "Unhandled C2H event %02x seq %02x\n",
+			 c2h->id, c2h->seq);
+		print_hex_dump(KERN_INFO, "C2H content: ", DUMP_PREFIX_NONE,
+			       16, 1, c2h->raw.payload, len, false);
 		break;
 	}
 }
-- 
cgit v0.10.2


From 55a18dd18052eaef908efbb27d5c6d7e0569048f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:41 -0500
Subject: rtl8xxxu: Process C2H RA_REPORT events for 8723bu

Handle RA_REPORTS events for 8723bu to not have them show up as
unhandled.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 07f4ee0..4164b1a 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7445,6 +7445,12 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv,
 		dev_dbg(dev, "C2H_MP_INFO ext ID %02x, status %02x\n",
 			c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status);
 		break;
+	case C2H_8723B_RA_REPORT:
+		dev_dbg(dev,
+			"C2H RA RPT: rate %02x, unk %i, macid %02x, noise %i\n",
+			c2h->ra_report.rate, c2h->ra_report.dummy0_0,
+			c2h->ra_report.macid, c2h->ra_report.noisy_state);
+		break;
 	default:
 		dev_info(dev, "Unhandled C2H event %02x seq %02x\n",
 			 c2h->id, c2h->seq);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 2191b6b..b44cb59 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1033,6 +1033,7 @@ enum c2h_evt_8723b {
 	C2H_8723B_BT_INFO = 9,
 	C2H_8723B_HW_INFO_EXCH = 0x0a,
 	C2H_8723B_BT_MP_INFO = 0x0b,
+	C2H_8723B_RA_REPORT = 0x0c,
 	C2H_8723B_FW_DEBUG = 0xff,
 };
 
@@ -1121,6 +1122,16 @@ struct rtl8723bu_c2h {
 			u8 a4;
 			u8 a5;
 		} __packed bt_info;
+		struct {
+			u8 rate:7;
+			u8 dummy0_0:1;
+			u8 macid;
+			u8 ldpc:1;
+			u8 txbf:1;
+			u8 noisy_state:1;
+			u8 dummy2_0:5;
+			u8 dummy3_0;
+		} __packed ra_report;
 	};
 };
 
-- 
cgit v0.10.2


From 87957081b749f3445fc6f10fae57280d45d6c6d5 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:42 -0500
Subject: rtl8xxxu: Pass RX rate to rx_parse_phystats and enable phystats for
 rtl8723bu

rtl8xxxu_rx_parse_phystats() only needs the RX rate to determine
whether to handle the stats as CCK or not. Parsing in the rate rather
than the rx descriptor elimantes the need to handle multiple rx
descriptor formats in the function.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4164b1a..4e08f5e 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7216,13 +7216,13 @@ error:
 
 static void rtl8xxxu_rx_parse_phystats(struct rtl8xxxu_priv *priv,
 				       struct ieee80211_rx_status *rx_status,
-				       struct rtl8xxxu_rx_desc *rx_desc,
-				       struct rtl8723au_phy_stats *phy_stats)
+				       struct rtl8723au_phy_stats *phy_stats,
+				       u32 rxmcs)
 {
 	if (phy_stats->sgi_en)
 		rx_status->flag |= RX_FLAG_SHORT_GI;
 
-	if (rx_desc->rxmcs < DESC_RATE_6M) {
+	if (rxmcs < DESC_RATE_6M) {
 		/*
 		 * Handle PHY stats for CCK rates
 		 */
@@ -7350,7 +7350,8 @@ static int rtl8723au_parse_rx_desc(struct rtl8xxxu_priv *priv,
 	skb_pull(skb, drvinfo_sz + desc_shift);
 
 	if (rx_desc->phy_stats)
-		rtl8xxxu_rx_parse_phystats(priv, rx_status, rx_desc, phy_stats);
+		rtl8xxxu_rx_parse_phystats(priv, rx_status, phy_stats,
+					   rx_desc->rxmcs);
 
 	rx_status->mactime = le32_to_cpu(rx_desc->tsfl);
 	rx_status->flag |= RX_FLAG_MACTIME_START;
@@ -7395,6 +7396,10 @@ static int rtl8723bu_parse_rx_desc(struct rtl8xxxu_priv *priv,
 		return RX_TYPE_C2H;
 	}
 
+	if (rx_desc->phy_stats)
+		rtl8xxxu_rx_parse_phystats(priv, rx_status, phy_stats,
+					   rx_desc->rxmcs);
+
 	rx_status->mactime = le32_to_cpu(rx_desc->tsfl);
 	rx_status->flag |= RX_FLAG_MACTIME_START;
 
-- 
cgit v0.10.2


From 6979494adf57afff0b6a5ad7ba77bdaef63274ad Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:43 -0500
Subject: rtl8xxxu: Remove unncessary semicolon

This removes an superfluous semicolon.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 4e08f5e..ce45c190 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -7195,7 +7195,7 @@ static void rtl8xxxu_tx(struct ieee80211_hw *hw,
 			tx_desc->txdw3 |=
 				cpu_to_le32(TXDESC_HW_RTS_ENABLE_8723B);
 		}
-	};
+	}
 
 	rtl8xxxu_calc_tx_desc_csum(tx_desc);
 
-- 
cgit v0.10.2


From 0290e7d0fd1e91d86289abbdd2d785c945e6bb2e Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:44 -0500
Subject: rtl8xxxu: convert rtl8723bu_init_bt() into rtl8723b_enable_rf()

rtl8723bu_init_bt() is effectively the function enabling RF, so name
it appropriately.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ce45c190..aae4509 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -1596,10 +1596,6 @@ static void rtl8723a_enable_rf(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_TXPAUSE, 0x00);
 }
 
-static void rtl8723b_enable_rf(struct rtl8xxxu_priv *priv)
-{
-}
-
 static void rtl8723a_disable_rf(struct rtl8xxxu_priv *priv)
 {
 	u8 sps0;
@@ -5953,7 +5949,7 @@ static void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv,
 }
 #endif
 
-static void rtl8723bu_init_bt(struct rtl8xxxu_priv *priv)
+static void rtl8723b_enable_rf(struct rtl8xxxu_priv *priv)
 {
 	struct h2c_cmd h2c;
 	u32 val32;
@@ -6508,10 +6504,6 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 	else
 		rtl8xxxu_write_rfreg(priv, RF_A, RF6052_REG_T_METER, 0x60);
 
-	/* Init BT hw config. */
-	if (priv->fops->init_bt)
-		priv->fops->init_bt(priv);
-
 	/* Set NAV_UPPER to 30000us */
 	val8 = ((30000 + NAV_UPPER_UNIT - 1) / NAV_UPPER_UNIT);
 	rtl8xxxu_write8(priv, REG_NAV_UPPER, val8);
@@ -8300,7 +8292,6 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
-	.init_bt = rtl8723bu_init_bt,
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.init_aggregation = rtl8723bu_init_aggregation,
 	.init_statistics = rtl8723bu_init_statistics,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index b44cb59..776fd17 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1261,7 +1261,6 @@ struct rtl8xxxu_fileops {
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
 	void (*config_channel) (struct ieee80211_hw *hw);
-	void (*init_bt) (struct rtl8xxxu_priv *priv);
 	int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb,
 			      struct ieee80211_rx_status *rx_status);
 	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
-- 
cgit v0.10.2


From 37f44dc79a385fed307cdaad6003fd3c0ca99c07 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:45 -0500
Subject: rtl8xxxu: Use define for REG_PWR_DATA bits

Use the bit define rather than hard code the value for REG_PWR_DATA bits.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index aae4509..f936e03 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5996,7 +5996,7 @@ static void rtl8723b_enable_rf(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, 0x0067, val8);
 
 	val32 = rtl8xxxu_read32(priv, REG_PWR_DATA);
-	val32 |= BIT(11);
+	val32 |= PWR_DATA_EEPRPAD_RFE_CTRL_EN;
 	rtl8xxxu_write32(priv, REG_PWR_DATA, val32);
 
 	/*
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index a055362..e0b9ea38 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -135,7 +135,7 @@
 #define  EFUSE_ACCESS_DISABLE		0x00	/* RTL8723 only */
 
 #define REG_PWR_DATA			0x0038
-#define PWR_DATA_EEPRPAD_RFE_CTRL_EN	BIT(11)
+#define  PWR_DATA_EEPRPAD_RFE_CTRL_EN	BIT(11)
 
 #define REG_CAL_TIMER			0x003c
 #define REG_ACLK_MON			0x003e
-- 
cgit v0.10.2


From fc89a41fa686b41b5339b3eaaa9851fa4c5551a9 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:46 -0500
Subject: rtl8xxxu: Implement 8723bu specific disable_rf() function

Powering up the 8723bu RF should probably be matched by the ability to
power it down again.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index f936e03..50d2c5c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -6056,6 +6056,17 @@ static void rtl8723b_enable_rf(struct rtl8xxxu_priv *priv)
 	rtl8723a_h2c_cmd(priv, &h2c, sizeof(h2c.ignore_wlan));
 }
 
+static void rtl8723b_disable_rf(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+
+	rtl8xxxu_write8(priv, REG_TXPAUSE, 0xff);
+
+	val32 = rtl8xxxu_read32(priv, REG_RX_WAIT_CCA);
+	val32 &= ~(BIT(22) | BIT(23));
+	rtl8xxxu_write32(priv, REG_RX_WAIT_CCA, val32);
+}
+
 static void rtl8723bu_init_aggregation(struct rtl8xxxu_priv *priv)
 {
 	u32 agg_rx;
@@ -7987,7 +7998,7 @@ static void rtl8xxxu_stop(struct ieee80211_hw *hw)
 	if (priv->usb_interrupts)
 		usb_kill_anchored_urbs(&priv->int_anchor);
 
-	rtl8723a_disable_rf(priv);
+	priv->fops->disable_rf(priv);
 
 	/*
 	 * Disable interrupts
@@ -8271,6 +8282,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.config_channel = rtl8723au_config_channel,
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.enable_rf = rtl8723a_enable_rf,
+	.disable_rf = rtl8723a_disable_rf,
 	.set_tx_power = rtl8723a_set_tx_power,
 	.update_rate_mask = rtl8723au_update_rate_mask,
 	.report_connect = rtl8723au_report_connect,
@@ -8296,6 +8308,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.init_aggregation = rtl8723bu_init_aggregation,
 	.init_statistics = rtl8723bu_init_statistics,
 	.enable_rf = rtl8723b_enable_rf,
+	.disable_rf = rtl8723b_disable_rf,
 	.set_tx_power = rtl8723b_set_tx_power,
 	.update_rate_mask = rtl8723bu_update_rate_mask,
 	.report_connect = rtl8723bu_report_connect,
@@ -8321,6 +8334,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.config_channel = rtl8723au_config_channel,
 	.parse_rx_desc = rtl8723au_parse_rx_desc,
 	.enable_rf = rtl8723a_enable_rf,
+	.disable_rf = rtl8723a_disable_rf,
 	.set_tx_power = rtl8723a_set_tx_power,
 	.update_rate_mask = rtl8723au_update_rate_mask,
 	.report_connect = rtl8723au_report_connect,
@@ -8345,6 +8359,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.config_channel = rtl8723bu_config_channel,
 	.parse_rx_desc = rtl8723bu_parse_rx_desc,
 	.enable_rf = rtl8723b_enable_rf,
+	.disable_rf = rtl8723b_disable_rf,
 	.set_tx_power = rtl8723b_set_tx_power,
 	.update_rate_mask = rtl8723au_update_rate_mask,
 	.report_connect = rtl8723au_report_connect,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 776fd17..9c535e8 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1266,6 +1266,7 @@ struct rtl8xxxu_fileops {
 	void (*init_aggregation) (struct rtl8xxxu_priv *priv);
 	void (*init_statistics) (struct rtl8xxxu_priv *priv);
 	void (*enable_rf) (struct rtl8xxxu_priv *priv);
+	void (*disable_rf) (struct rtl8xxxu_priv *priv);
 	void (*set_tx_power) (struct rtl8xxxu_priv *priv, int channel,
 			      bool ht40);
 	void (*update_rate_mask) (struct rtl8xxxu_priv *priv,
-- 
cgit v0.10.2


From fe37d5f644e5fb7179ff6464482eca9a4ebf555c Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:47 -0500
Subject: rtl8xxxu: Implement device specific power_off function

Implment 8723bu specific device power down, and make power_off() a
fileops function.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 50d2c5c..ef60e83 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5240,6 +5240,64 @@ exit:
 	return ret;
 }
 
+static int rtl8723bu_active_to_emu(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u16 val16;
+	u32 val32;
+	int count, ret;
+
+	/* Turn off RF */
+	rtl8xxxu_write8(priv, REG_RF_CTRL, 0);
+
+	/* Enable rising edge triggering interrupt */
+	val16 = rtl8xxxu_read16(priv, REG_GPIO_INTM);
+	val16 &= ~GPIO_INTM_EDGE_TRIG_IRQ;
+	rtl8xxxu_write16(priv, REG_GPIO_INTM, val16);
+
+	/* Release WLON reset 0x04[16]= 1*/
+	val32 = rtl8xxxu_read32(priv, REG_GPIO_INTM);
+	val32 |= APS_FSMCO_WLON_RESET;
+	rtl8xxxu_write32(priv, REG_GPIO_INTM, val32);
+
+	/* 0x0005[1] = 1 turn off MAC by HW state machine*/
+	val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1);
+	val8 |= BIT(1);
+	rtl8xxxu_write8(priv, REG_APS_FSMCO + 1, val8);
+
+	for (count = RTL8XXXU_MAX_REG_POLL; count; count--) {
+		val8 = rtl8xxxu_read8(priv, REG_APS_FSMCO + 1);
+		if ((val8 & BIT(1)) == 0)
+			break;
+		udelay(10);
+	}
+
+	if (!count) {
+		dev_warn(&priv->udev->dev, "%s: Disabling MAC timed out\n",
+			 __func__);
+		ret = -EBUSY;
+		goto exit;
+	}
+
+	/* Enable BT control XTAL setting */
+	val8 = rtl8xxxu_read8(priv, REG_AFE_MISC);
+	val8 &= ~AFE_MISC_WL_XTAL_CTRL;
+	rtl8xxxu_write8(priv, REG_AFE_MISC, val8);
+
+	/* 0x0000[5] = 1 analog Ips to digital, 1:isolation */
+	val8 = rtl8xxxu_read8(priv, REG_SYS_ISO_CTRL);
+	val8 |= SYS_ISO_ANALOG_IPS;
+	rtl8xxxu_write8(priv, REG_SYS_ISO_CTRL, val8);
+
+	/* 0x0020[0] = 0 disable LDOA12 MACRO block*/
+	val8 = rtl8xxxu_read8(priv, REG_LDOA15_CTRL);
+	val8 &= ~LDOA15_ENABLE;
+	rtl8xxxu_write8(priv, REG_LDOA15_CTRL, val8);
+
+exit:
+	return ret;
+}
+
 static int rtl8xxxu_active_to_lps(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
@@ -5932,6 +5990,38 @@ static void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write8(priv, REG_RSV_CTRL, 0x0e);
 }
 
+static void rtl8723bu_power_off(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u16 val16;
+
+	/*
+	 * Disable TX report timer
+	 */
+	val8 = rtl8xxxu_read8(priv, REG_TX_REPORT_CTRL);
+	val8 &= ~TX_REPORT_CTRL_TIMER_ENABLE;
+	rtl8xxxu_write8(priv, REG_TX_REPORT_CTRL, val8);
+
+	rtl8xxxu_write16(priv, REG_CR, 0x0000);
+
+	rtl8xxxu_active_to_lps(priv);
+
+	/* Reset Firmware if running in RAM */
+	if (rtl8xxxu_read8(priv, REG_MCU_FW_DL) & MCU_FW_RAM_SEL)
+		rtl8xxxu_firmware_self_reset(priv);
+
+	/* Reset MCU */
+	val16 = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+	val16 &= ~SYS_FUNC_CPU_ENABLE;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, val16);
+
+	/* Reset MCU ready status */
+	rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00);
+
+	rtl8723bu_active_to_emu(priv);
+	rtl8xxxu_emu_to_disabled(priv);
+}
+
 #ifdef NEED_PS_TDMA
 static void rtl8723bu_set_ps_tdma(struct rtl8xxxu_priv *priv,
 				  u8 arg1, u8 arg2, u8 arg3, u8 arg4, u8 arg5)
@@ -6152,7 +6242,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw)
 		 */
 		if (priv->rtlchip == 0x8723bu) {
 			val8 = rtl8xxxu_read8(priv, REG_TX_REPORT_CTRL);
-			val8 |= BIT(1);
+			val8 |= TX_REPORT_CTRL_TIMER_ENABLE;
 			rtl8xxxu_write8(priv, REG_TX_REPORT_CTRL, val8);
 			/* Set MAX RPT MACID */
 			rtl8xxxu_write8(priv, REG_TX_REPORT_CTRL + 1, 0x02);
@@ -6545,7 +6635,7 @@ static void rtl8xxxu_disable_device(struct ieee80211_hw *hw)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
 
-	rtl8xxxu_power_off(priv);
+	priv->fops->power_off(priv);
 }
 
 static void rtl8xxxu_cam_write(struct rtl8xxxu_priv *priv,
@@ -8277,6 +8367,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.parse_efuse = rtl8723au_parse_efuse,
 	.load_firmware = rtl8723au_load_firmware,
 	.power_on = rtl8723au_power_on,
+	.power_off = rtl8xxxu_power_off,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
@@ -8300,6 +8391,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.parse_efuse = rtl8723bu_parse_efuse,
 	.load_firmware = rtl8723bu_load_firmware,
 	.power_on = rtl8723bu_power_on,
+	.power_off = rtl8723bu_power_off,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
@@ -8329,6 +8421,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.parse_efuse = rtl8192cu_parse_efuse,
 	.load_firmware = rtl8192cu_load_firmware,
 	.power_on = rtl8192cu_power_on,
+	.power_off = rtl8xxxu_power_off,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
@@ -8354,6 +8447,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.parse_efuse = rtl8192eu_parse_efuse,
 	.load_firmware = rtl8192eu_load_firmware,
 	.power_on = rtl8192eu_power_on,
+	.power_off = rtl8xxxu_power_off,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index 9c535e8..bfa9d6c 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1257,6 +1257,7 @@ struct rtl8xxxu_fileops {
 	int (*parse_efuse) (struct rtl8xxxu_priv *priv);
 	int (*load_firmware) (struct rtl8xxxu_priv *priv);
 	int (*power_on) (struct rtl8xxxu_priv *priv);
+	void (*power_off) (struct rtl8xxxu_priv *priv);
 	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index e0b9ea38..a8d5c9f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -512,6 +512,8 @@
 #define REG_PKT_LOSE_RPT		0x04e1
 #define REG_PTCL_ERR_STATUS		0x04e2
 #define REG_TX_REPORT_CTRL		0x04ec
+#define  TX_REPORT_CTRL_TIMER_ENABLE	BIT(1)
+
 #define REG_TX_REPORT_TIME		0x04f0
 #define REG_DUMMY			0x04fc
 
-- 
cgit v0.10.2


From 430b454c5a3951ffaf117d4e613e9cea03f86a90 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:48 -0500
Subject: rtl8xxxu: Flush FIFO before powering down devices

This should help when reloading the driver for 8723bu devices

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index ef60e83..e444e2b 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5687,6 +5687,36 @@ static int rtl8xxxu_emu_to_disabled(struct rtl8xxxu_priv *priv)
 	return 0;
 }
 
+static int rtl8xxxu_flush_fifo(struct rtl8xxxu_priv *priv)
+{
+	u32 val32;
+	int retry, retval;
+
+	rtl8xxxu_write8(priv, REG_TXPAUSE, 0xff);
+
+	val32 = rtl8xxxu_read32(priv, REG_RXPKT_NUM);
+	val32 |= RXPKT_NUM_RW_RELEASE_EN;
+	rtl8xxxu_write32(priv, REG_RXPKT_NUM, val32);
+
+	retry = 100;
+	retval = -EBUSY;
+
+	do {
+		val32 = rtl8xxxu_read32(priv, REG_RXPKT_NUM);
+		if (val32 & RXPKT_NUM_RXDMA_IDLE) {
+			retval = 0;
+			break;
+		}
+	} while (retry--);
+
+	rtl8xxxu_write16(priv, REG_RQPN_NPQ, 0);
+	rtl8xxxu_write32(priv, REG_RQPN, 0x80000000);
+	mdelay(2);
+	pr_info("%s: retry %i\n", __func__, retry);
+
+	return retval;
+}
+
 static int rtl8723au_power_on(struct rtl8xxxu_priv *priv)
 {
 	u8 val8;
@@ -5957,6 +5987,8 @@ static void rtl8xxxu_power_off(struct rtl8xxxu_priv *priv)
 		rtl8xxxu_write32(priv, REG_FPGA0_XCD_RF_PARM, val32);
 	}
 
+	rtl8xxxu_flush_fifo(priv);
+
 	rtl8xxxu_active_to_lps(priv);
 
 	/* Turn off RF */
@@ -5995,6 +6027,8 @@ static void rtl8723bu_power_off(struct rtl8xxxu_priv *priv)
 	u8 val8;
 	u16 val16;
 
+	rtl8xxxu_flush_fifo(priv);
+
 	/*
 	 * Disable TX report timer
 	 */
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
index a8d5c9f..e545e84 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h
@@ -404,6 +404,8 @@
 #define REG_RXDMA_AGG_PG_TH		0x0280
 #define  RXDMA_USB_AGG_ENABLE		BIT(31)
 #define REG_RXPKT_NUM			0x0284
+#define  RXPKT_NUM_RXDMA_IDLE		BIT(17)
+#define  RXPKT_NUM_RW_RELEASE_EN	BIT(18)
 #define REG_RXDMA_STATUS		0x0288
 
 /* Presumably only found on newer chips such as 8723bu */
-- 
cgit v0.10.2


From 145428ec7c6794c3ac78ddce2b073076c9c870d2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:49 -0500
Subject: rtl8xxxu: Print a warning if flushing the FIFO fails

Only print a warning if the FIFO flush fails, as opposed to printing
the status unconditionally.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index e444e2b..5c08db7 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -5689,6 +5689,7 @@ static int rtl8xxxu_emu_to_disabled(struct rtl8xxxu_priv *priv)
 
 static int rtl8xxxu_flush_fifo(struct rtl8xxxu_priv *priv)
 {
+	struct device *dev = &priv->udev->dev;
 	u32 val32;
 	int retry, retval;
 
@@ -5712,7 +5713,9 @@ static int rtl8xxxu_flush_fifo(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write16(priv, REG_RQPN_NPQ, 0);
 	rtl8xxxu_write32(priv, REG_RQPN, 0x80000000);
 	mdelay(2);
-	pr_info("%s: retry %i\n", __func__, retry);
+
+	if (!retry)
+		dev_warn(dev, "Failed to flush FIFO\n");
 
 	return retval;
 }
-- 
cgit v0.10.2


From 7d4ccb8bae85550035328bddd20197614dfe415a Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:50 -0500
Subject: rtl8xxxu: Use correct 8051 reset function for 8723b parts

8723b needs more action, so implement support for device specific
reset functions.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 5c08db7..81777c4 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2720,12 +2720,44 @@ static void rtl8xxxu_reset_8051(struct rtl8xxxu_priv *priv)
 	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
 	val8 &= ~BIT(0);
 	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+
 	sys_func = rtl8xxxu_read16(priv, REG_SYS_FUNC);
 	sys_func &= ~SYS_FUNC_CPU_ENABLE;
 	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
+
 	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
 	val8 |= BIT(0);
 	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+
+	sys_func |= SYS_FUNC_CPU_ENABLE;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
+}
+
+static void rtl8723bu_reset_8051(struct rtl8xxxu_priv *priv)
+{
+	u8 val8;
+	u16 sys_func;
+
+	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL);
+	val8 &= ~BIT(1);
+	rtl8xxxu_write8(priv, REG_RSV_CTRL, val8);
+
+	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
+	val8 &= ~BIT(0);
+	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+
+	sys_func = rtl8xxxu_read16(priv, REG_SYS_FUNC);
+	sys_func &= ~SYS_FUNC_CPU_ENABLE;
+	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
+
+	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL);
+	val8 &= ~BIT(1);
+	rtl8xxxu_write8(priv, REG_RSV_CTRL, val8);
+
+	val8 = rtl8xxxu_read8(priv, REG_RSV_CTRL + 1);
+	val8 |= BIT(0);
+	rtl8xxxu_write8(priv, REG_RSV_CTRL + 1, val8);
+
 	sys_func |= SYS_FUNC_CPU_ENABLE;
 	rtl8xxxu_write16(priv, REG_SYS_FUNC, sys_func);
 }
@@ -2758,7 +2790,7 @@ static int rtl8xxxu_start_firmware(struct rtl8xxxu_priv *priv)
 	 * Reset the 8051 in order for the firmware to start running,
 	 * otherwise it won't come up on the 8192eu
 	 */
-	rtl8xxxu_reset_8051(priv);
+	priv->fops->reset_8051(priv);
 
 	/* Wait for firmware to become ready */
 	for (i = 0; i < RTL8XXXU_FIRMWARE_POLL_MAX; i++) {
@@ -2805,7 +2837,7 @@ static int rtl8xxxu_download_firmware(struct rtl8xxxu_priv *priv)
 	if (val8 & MCU_FW_RAM_SEL) {
 		pr_info("do the RAM reset\n");
 		rtl8xxxu_write8(priv, REG_MCU_FW_DL, 0x00);
-		rtl8xxxu_reset_8051(priv);
+		priv->fops->reset_8051(priv);
 	}
 
 	/* MCU firmware download enable */
@@ -8405,6 +8437,7 @@ static struct rtl8xxxu_fileops rtl8723au_fops = {
 	.load_firmware = rtl8723au_load_firmware,
 	.power_on = rtl8723au_power_on,
 	.power_off = rtl8xxxu_power_off,
+	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
@@ -8429,6 +8462,7 @@ static struct rtl8xxxu_fileops rtl8723bu_fops = {
 	.load_firmware = rtl8723bu_load_firmware,
 	.power_on = rtl8723bu_power_on,
 	.power_off = rtl8723bu_power_off,
+	.reset_8051 = rtl8723bu_reset_8051,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_init_antenna_selection = rtl8723bu_phy_init_antenna_selection,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
@@ -8459,6 +8493,7 @@ static struct rtl8xxxu_fileops rtl8192cu_fops = {
 	.load_firmware = rtl8192cu_load_firmware,
 	.power_on = rtl8192cu_power_on,
 	.power_off = rtl8xxxu_power_off,
+	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_init_llt_table,
 	.phy_iq_calibrate = rtl8723au_phy_iq_calibrate,
 	.config_channel = rtl8723au_config_channel,
@@ -8485,6 +8520,7 @@ static struct rtl8xxxu_fileops rtl8192eu_fops = {
 	.load_firmware = rtl8192eu_load_firmware,
 	.power_on = rtl8192eu_power_on,
 	.power_off = rtl8xxxu_power_off,
+	.reset_8051 = rtl8xxxu_reset_8051,
 	.llt_init = rtl8xxxu_auto_llt_table,
 	.phy_iq_calibrate = rtl8723bu_phy_iq_calibrate,
 	.config_channel = rtl8723bu_config_channel,
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
index bfa9d6c..7b73654 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h
@@ -1258,6 +1258,7 @@ struct rtl8xxxu_fileops {
 	int (*load_firmware) (struct rtl8xxxu_priv *priv);
 	int (*power_on) (struct rtl8xxxu_priv *priv);
 	void (*power_off) (struct rtl8xxxu_priv *priv);
+	void (*reset_8051) (struct rtl8xxxu_priv *priv);
 	int (*llt_init) (struct rtl8xxxu_priv *priv, u8 last_tx_page);
 	void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv);
 	void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv);
-- 
cgit v0.10.2


From ccfe1e85322090649d2fae599e55300c1512bf15 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <Jes.Sorensen@redhat.com>
Date: Mon, 29 Feb 2016 17:05:51 -0500
Subject: rtl8xxxu: Temporarily disable 8192eu device init

To reduce the patch volume, temporariliy disable 8192eu device init.

Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
index 81777c4..abdff45 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.c
@@ -2562,6 +2562,10 @@ static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv)
 				 raw[i + 6], raw[i + 7]);
 		}
 	}
+	/*
+	 * Temporarily disable 8192eu support
+	 */
+	return -EINVAL;
 	return 0;
 }
 
-- 
cgit v0.10.2


From 079c2652e5af648db6bf4f54bcafdafcc57a0d2c Mon Sep 17 00:00:00 2001
From: Robert Baldyga <r.baldyga@samsung.com>
Date: Thu, 10 Mar 2016 15:22:43 +0100
Subject: MAINTAINERS: nfc: s3fwrn5: Add second maintainer

Add Krzysztof Opasiak as maintainer of S3FWRN5 driver.

Signed-off-by: Robert Baldyga <r.baldyga@samsung.com>
Acked-by: Krzysztof Opasiak <k.opasiak@samsung.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>

diff --git a/MAINTAINERS b/MAINTAINERS
index 5e4e50f..4dd2d37 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9482,6 +9482,7 @@ F:	drivers/media/i2c/s5k5baf.c
 
 SAMSUNG S3FWRN5 NFC DRIVER
 M:	Robert Baldyga <r.baldyga@samsung.com>
+M:	Krzysztof Opasiak <k.opasiak@samsung.com>
 L:	linux-nfc@lists.01.org (moderated for non-subscribers)
 S:	Supported
 F:	drivers/nfc/s3fwrn5
-- 
cgit v0.10.2


From c5fd9cbb9dc40e0d273d6731030e65096adceeda Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A9r=C3=B4me=20de=20Bretagne?=
 <jerome.debretagne@gmail.com>
Date: Thu, 3 Mar 2016 01:46:28 +0100
Subject: Bluetooth: hci_bcm: Add BCM2E55 ACPI ID used in Lenovo ThinkPad
 Tablet 8
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Lenovo ThinkPad Tablet 8 with BCM43241 rev B5 chipset uses the BCM2E55
ACPI ID for Bluetooth. Add it to the list of supported devices.

Signed-off-by: Jérôme de Bretagne <jerome.debretagne@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/hci_bcm.c b/drivers/bluetooth/hci_bcm.c
index bb4c5a0..d8881dc 100644
--- a/drivers/bluetooth/hci_bcm.c
+++ b/drivers/bluetooth/hci_bcm.c
@@ -821,6 +821,7 @@ static const struct acpi_device_id bcm_acpi_match[] = {
 	{ "BCM2E3F", 0 },
 	{ "BCM2E40", 0 },
 	{ "BCM2E54", 0 },
+	{ "BCM2E55", 0 },
 	{ "BCM2E64", 0 },
 	{ "BCM2E65", 0 },
 	{ "BCM2E67", 0 },
-- 
cgit v0.10.2


From 75c6aca4765dbe3d0c1507ab5052f2e373dc2331 Mon Sep 17 00:00:00 2001
From: Dmitry Tunin <hanipouspilot@gmail.com>
Date: Fri, 4 Mar 2016 01:32:19 +0300
Subject: Bluetooth: btusb: Add a new AR3012 ID 13d3:3472

T: Bus=01 Lev=01 Prnt=01 Port=04 Cnt=01 Dev#= 4 Spd=12 MxCh= 0
D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=13d3 ProdID=3472 Rev=00.01
C: #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb

BugLink: https://bugs.launchpad.net/bugs/1552925

Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Cc: stable@vger.kernel.org

diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c
index 9374738..47ca4b3 100644
--- a/drivers/bluetooth/ath3k.c
+++ b/drivers/bluetooth/ath3k.c
@@ -120,6 +120,7 @@ static const struct usb_device_id ath3k_table[] = {
 	{ USB_DEVICE(0x13d3, 0x3408) },
 	{ USB_DEVICE(0x13d3, 0x3423) },
 	{ USB_DEVICE(0x13d3, 0x3432) },
+	{ USB_DEVICE(0x13d3, 0x3472) },
 	{ USB_DEVICE(0x13d3, 0x3474) },
 
 	/* Atheros AR5BBU12 with sflash firmware */
@@ -185,6 +186,7 @@ static const struct usb_device_id ath3k_blist_tbl[] = {
 	{ USB_DEVICE(0x13d3, 0x3408), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3423), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3432), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x13d3, 0x3472), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3474), .driver_info = BTUSB_ATH3012 },
 
 	/* Atheros AR5BBU22 with sflash firmware */
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 97f3bba..0d4e372 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -234,6 +234,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x13d3, 0x3408), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3423), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3432), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x13d3, 0x3472), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x13d3, 0x3474), .driver_info = BTUSB_ATH3012 },
 
 	/* Atheros AR5BBU12 with sflash firmware */
-- 
cgit v0.10.2


From f16089209e1029d45ae78dd238b6ab9b2c9a886c Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 4 Mar 2016 10:10:20 +0100
Subject: mac802154: use put and get unaligned functions

This patch removes the swap pointer and memmove functionality. Instead
we use the well known put/get unaligned access with specific byte order
handling.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Suggested-by: Marc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/include/net/mac802154.h b/include/net/mac802154.h
index 2e3cdd20..6cd7a70 100644
--- a/include/net/mac802154.h
+++ b/include/net/mac802154.h
@@ -16,10 +16,10 @@
 #ifndef NET_MAC802154_H
 #define NET_MAC802154_H
 
+#include <asm/unaligned.h>
 #include <net/af_ieee802154.h>
 #include <linux/ieee802154.h>
 #include <linux/skbuff.h>
-#include <linux/unaligned/memmove.h>
 
 #include <net/cfg802154.h>
 
@@ -254,7 +254,7 @@ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb)
 		return cpu_to_le16(0);
 	}
 
-	return (__force __le16)__get_unaligned_memmove16(skb_mac_header(skb));
+	return get_unaligned_le16(skb_mac_header(skb));
 }
 
 /**
@@ -264,7 +264,7 @@ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb)
  */
 static inline void ieee802154_be64_to_le64(void *le64_dst, const void *be64_src)
 {
-	__put_unaligned_memmove64(swab64p(be64_src), le64_dst);
+	put_unaligned_le64(get_unaligned_be64(be64_src), le64_dst);
 }
 
 /**
@@ -274,7 +274,7 @@ static inline void ieee802154_be64_to_le64(void *le64_dst, const void *be64_src)
  */
 static inline void ieee802154_le64_to_be64(void *be64_dst, const void *le64_src)
 {
-	__put_unaligned_memmove64(swab64p(le64_src), be64_dst);
+	put_unaligned_be64(get_unaligned_le64(le64_src), be64_dst);
 }
 
 /**
@@ -284,7 +284,7 @@ static inline void ieee802154_le64_to_be64(void *be64_dst, const void *le64_src)
  */
 static inline void ieee802154_le16_to_be16(void *be16_dst, const void *le16_src)
 {
-	__put_unaligned_memmove16(swab16p(le16_src), be16_dst);
+	put_unaligned_be16(get_unaligned_le16(le16_src), be16_dst);
 }
 
 /**
-- 
cgit v0.10.2


From 6aaf37b41a37b184e509aa3ec29d96b7712dfd43 Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Fri, 4 Mar 2016 10:10:21 +0100
Subject: 6lowpan: debugfs: add missing static

This patch solves the sparse warning:

net/6lowpan/debugfs.c:164:30: warning: symbol 'lowpan_ctx_pfx_fops' was
not declared. Should it be static?
net/6lowpan/debugfs.c:241:30: warning: symbol 'lowpan_context_fops' was
not declared. Should it be static?

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/6lowpan/debugfs.c b/net/6lowpan/debugfs.c
index aa49ff4..0793a81 100644
--- a/net/6lowpan/debugfs.c
+++ b/net/6lowpan/debugfs.c
@@ -161,7 +161,7 @@ out:
 	return status;
 }
 
-const struct file_operations lowpan_ctx_pfx_fops = {
+static const struct file_operations lowpan_ctx_pfx_fops = {
 	.open		= lowpan_ctx_pfx_open,
 	.read		= seq_read,
 	.write		= lowpan_ctx_pfx_write,
@@ -238,7 +238,7 @@ static int lowpan_context_open(struct inode *inode, struct file *file)
 	return single_open(file, lowpan_context_show, inode->i_private);
 }
 
-const struct file_operations lowpan_context_fops = {
+static const struct file_operations lowpan_context_fops = {
 	.open		= lowpan_context_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-- 
cgit v0.10.2


From 24c4a81d765738655f94d242eb821cb26e6c143a Mon Sep 17 00:00:00 2001
From: Alexander Aring <aar@pengutronix.de>
Date: Mon, 7 Mar 2016 20:07:31 +0100
Subject: 6lowpan: iphc: fix SAM/DAM bit comment

This patch fixes the comments for SAM/DAM value.

Signed-off-by: Alexander Aring <aar@pengutronix.de>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c
index 7217251..99bb22a 100644
--- a/net/6lowpan/iphc.c
+++ b/net/6lowpan/iphc.c
@@ -788,7 +788,7 @@ static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct in6_addr *ipaddr,
 	}
 
 	memset(&tmp, 0, sizeof(tmp));
-	/* check for SAM/DAM = 01 */
+	/* check for SAM/DAM = 10 */
 	tmp.s6_addr[11] = 0xFF;
 	tmp.s6_addr[12] = 0xFE;
 	memcpy(&tmp.s6_addr[14], &ipaddr->s6_addr[14], 2);
@@ -801,7 +801,7 @@ static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct in6_addr *ipaddr,
 	}
 
 	memset(&tmp, 0, sizeof(tmp));
-	/* check for SAM/DAM = 10, should always match */
+	/* check for SAM/DAM = 01, should always match */
 	memcpy(&tmp.s6_addr[8], &ipaddr->s6_addr[8], 8);
 	/* context information are always used */
 	ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen);
-- 
cgit v0.10.2


From 9a01242dc7fc4d5fe3f722afbf35b33aa414cd2f Mon Sep 17 00:00:00 2001
From: Wei-Ning Huang <wnhuang@chromium.org>
Date: Tue, 8 Mar 2016 11:40:06 +0800
Subject: Bluetooth: btmrvl_sdio: fix firmware activation failure

In some case, the btmrvl_sdio firmware would fail to active within the
polling time. Increase the polling interval to 100 msec to fix the
issue.

Signed-off-by: Wei-Ning Huang <wnhuang@chromium.org>
Signed-off-by: Wei-Ning Huang <wnhuang@google.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c
index 6ed8acf..c6ef248 100644
--- a/drivers/bluetooth/btmrvl_sdio.c
+++ b/drivers/bluetooth/btmrvl_sdio.c
@@ -371,7 +371,7 @@ static int btmrvl_sdio_verify_fw_download(struct btmrvl_sdio_card *card,
 		if (firmwarestat == FIRMWARE_READY)
 			return 0;
 
-		msleep(10);
+		msleep(100);
 	}
 
 	return -ETIMEDOUT;
-- 
cgit v0.10.2


From eec7a01dc8366f7e43fc0417f4aee70eaeaca9a9 Mon Sep 17 00:00:00 2001
From: Johan Hedberg <johan.hedberg@intel.com>
Date: Wed, 9 Mar 2016 17:30:32 +0200
Subject: Bluetooth: Move memset closer to where it's needed

Minor fix to not do the memset until the variable it clears is
actually used.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 32575b4..7264025 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -726,14 +726,14 @@ static void hci_req_add_le_create_conn(struct hci_request *req,
 	struct hci_dev *hdev = conn->hdev;
 	u8 own_addr_type;
 
-	memset(&cp, 0, sizeof(cp));
-
 	/* Update random address, but set require_privacy to false so
 	 * that we never connect with an non-resolvable address.
 	 */
 	if (hci_update_random_address(req, false, &own_addr_type))
 		return;
 
+	memset(&cp, 0, sizeof(cp));
+
 	/* Set window to be the same value as the interval to enable
 	 * continuous scanning.
 	 */
-- 
cgit v0.10.2


From d43efbd0d545f476be0bbef30cbe18234f8a9631 Mon Sep 17 00:00:00 2001
From: Johan Hedberg <johan.hedberg@intel.com>
Date: Wed, 9 Mar 2016 17:30:33 +0200
Subject: Bluetooth: Fix adding discoverable to adv instance flags

When lookup up the advertising instance flags for the default
advertising instance (0) the discoverable flag should be filled in
based on the HCI_DISCOVERABLE flag.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
index c78ee2d..77be344 100644
--- a/net/bluetooth/hci_request.c
+++ b/net/bluetooth/hci_request.c
@@ -866,6 +866,9 @@ static u32 get_adv_instance_flags(struct hci_dev *hdev, u8 instance)
 		if (hci_dev_test_flag(hdev, HCI_ADVERTISING_CONNECTABLE))
 			flags |= MGMT_ADV_FLAG_CONNECTABLE;
 
+		if (hci_dev_test_flag(hdev, HCI_DISCOVERABLE))
+			flags |= MGMT_ADV_FLAG_DISCOV;
+
 		return flags;
 	}
 
-- 
cgit v0.10.2


From 82a37adeedd38880940e2772ec1ae27a09353e5a Mon Sep 17 00:00:00 2001
From: Johan Hedberg <johan.hedberg@intel.com>
Date: Wed, 9 Mar 2016 17:30:34 +0200
Subject: Bluetooth: Add support for limited privacy mode

Introduce a limited privacy mode indicated by value 0x02 to the mgmt
Set Privacy command.

With value 0x02 the kernel will use privacy mode with a resolvable
private address. In case the controller is bondable and discoverable
the identity address will be used.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index 339ea57..5d38d98 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -233,6 +233,7 @@ enum {
 	HCI_SC_ENABLED,
 	HCI_SC_ONLY,
 	HCI_PRIVACY,
+	HCI_LIMITED_PRIVACY,
 	HCI_RPA_EXPIRED,
 	HCI_RPA_RESOLVING,
 	HCI_HS_ENABLED,
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 7264025..bf9f8a8 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -719,6 +719,13 @@ done:
 	hci_dev_unlock(hdev);
 }
 
+static bool conn_use_rpa(struct hci_conn *conn)
+{
+	struct hci_dev *hdev = conn->hdev;
+
+	return hci_dev_test_flag(hdev, HCI_PRIVACY);
+}
+
 static void hci_req_add_le_create_conn(struct hci_request *req,
 				       struct hci_conn *conn)
 {
@@ -729,7 +736,8 @@ static void hci_req_add_le_create_conn(struct hci_request *req,
 	/* Update random address, but set require_privacy to false so
 	 * that we never connect with an non-resolvable address.
 	 */
-	if (hci_update_random_address(req, false, &own_addr_type))
+	if (hci_update_random_address(req, false, conn_use_rpa(conn),
+				      &own_addr_type))
 		return;
 
 	memset(&cp, 0, sizeof(cp));
@@ -774,7 +782,8 @@ static void hci_req_directed_advertising(struct hci_request *req,
 	/* Set require_privacy to false so that the remote device has a
 	 * chance of identifying us.
 	 */
-	if (hci_update_random_address(req, false, &own_addr_type) < 0)
+	if (hci_update_random_address(req, false, conn_use_rpa(conn),
+				      &own_addr_type) < 0)
 		return;
 
 	memset(&cp, 0, sizeof(cp));
diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
index 77be344..95a545c 100644
--- a/net/bluetooth/hci_request.c
+++ b/net/bluetooth/hci_request.c
@@ -771,6 +771,11 @@ static u8 update_white_list(struct hci_request *req)
 	return 0x01;
 }
 
+static bool scan_use_rpa(struct hci_dev *hdev)
+{
+	return hci_dev_test_flag(hdev, HCI_PRIVACY);
+}
+
 void hci_req_add_le_passive_scan(struct hci_request *req)
 {
 	struct hci_cp_le_set_scan_param param_cp;
@@ -785,7 +790,8 @@ void hci_req_add_le_passive_scan(struct hci_request *req)
 	 * advertising with our address will be correctly reported
 	 * by the controller.
 	 */
-	if (hci_update_random_address(req, false, &own_addr_type))
+	if (hci_update_random_address(req, false, scan_use_rpa(hdev),
+				      &own_addr_type))
 		return;
 
 	/* Adding or removing entries from the white list must
@@ -881,6 +887,29 @@ static u32 get_adv_instance_flags(struct hci_dev *hdev, u8 instance)
 	return adv_instance->flags;
 }
 
+static bool adv_use_rpa(struct hci_dev *hdev, uint32_t flags)
+{
+	/* If privacy is not enabled don't use RPA */
+	if (!hci_dev_test_flag(hdev, HCI_PRIVACY))
+		return false;
+
+	/* If basic privacy mode is enabled use RPA */
+	if (!hci_dev_test_flag(hdev, HCI_LIMITED_PRIVACY))
+		return true;
+
+	/* If limited privacy mode is enabled don't use RPA if we're
+	 * both discoverable and bondable.
+	 */
+	if ((flags & MGMT_ADV_FLAG_DISCOV) &&
+	    hci_dev_test_flag(hdev, HCI_BONDABLE))
+		return false;
+
+	/* We're neither bondable nor discoverable in the limited
+	 * privacy mode, therefore use RPA.
+	 */
+	return true;
+}
+
 void __hci_req_enable_advertising(struct hci_request *req)
 {
 	struct hci_dev *hdev = req->hdev;
@@ -914,7 +943,9 @@ void __hci_req_enable_advertising(struct hci_request *req)
 	 * advertising is used. In that case it is fine to use a
 	 * non-resolvable private address.
 	 */
-	if (hci_update_random_address(req, !connectable, &own_addr_type) < 0)
+	if (hci_update_random_address(req, !connectable,
+				      adv_use_rpa(hdev, flags),
+				      &own_addr_type) < 0)
 		return;
 
 	memset(&cp, 0, sizeof(cp));
@@ -1328,7 +1359,7 @@ static void set_random_addr(struct hci_request *req, bdaddr_t *rpa)
 }
 
 int hci_update_random_address(struct hci_request *req, bool require_privacy,
-			      u8 *own_addr_type)
+			      bool use_rpa, u8 *own_addr_type)
 {
 	struct hci_dev *hdev = req->hdev;
 	int err;
@@ -1337,7 +1368,7 @@ int hci_update_random_address(struct hci_request *req, bool require_privacy,
 	 * current RPA has expired or there is something else than
 	 * the current RPA in use, then generate a new one.
 	 */
-	if (hci_dev_test_flag(hdev, HCI_PRIVACY)) {
+	if (use_rpa) {
 		int to;
 
 		*own_addr_type = ADDR_LE_DEV_RANDOM;
@@ -1599,9 +1630,16 @@ static int discoverable_update(struct hci_request *req, unsigned long opt)
 	/* Advertising instances don't use the global discoverable setting, so
 	 * only update AD if advertising was enabled using Set Advertising.
 	 */
-	if (hci_dev_test_flag(hdev, HCI_ADVERTISING))
+	if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) {
 		__hci_req_update_adv_data(req, 0x00);
 
+		/* Discoverable mode affects the local advertising
+		 * address in limited privacy mode.
+		 */
+		if (hci_dev_test_flag(hdev, HCI_LIMITED_PRIVACY))
+			__hci_req_enable_advertising(req);
+	}
+
 	hci_dev_unlock(hdev);
 
 	return 0;
@@ -1944,7 +1982,8 @@ static int active_scan(struct hci_request *req, unsigned long opt)
 	 * address (when privacy feature has been enabled) or non-resolvable
 	 * private address.
 	 */
-	err = hci_update_random_address(req, true, &own_addr_type);
+	err = hci_update_random_address(req, true, scan_use_rpa(hdev),
+					&own_addr_type);
 	if (err < 0)
 		own_addr_type = ADDR_LE_DEV_PUBLIC;
 
diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h
index 64ff8c0..b2d044b 100644
--- a/net/bluetooth/hci_request.h
+++ b/net/bluetooth/hci_request.h
@@ -89,7 +89,7 @@ static inline void hci_req_update_scan(struct hci_dev *hdev)
 void __hci_req_update_scan(struct hci_request *req);
 
 int hci_update_random_address(struct hci_request *req, bool require_privacy,
-			      u8 *own_addr_type);
+			      bool use_rpa, u8 *own_addr_type);
 
 int hci_abort_conn(struct hci_conn *conn, u8 reason);
 void __hci_abort_conn(struct hci_request *req, struct hci_conn *conn,
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 5a5089c..2ca3555 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -1382,8 +1382,19 @@ static int set_bondable(struct sock *sk, struct hci_dev *hdev, void *data,
 	if (err < 0)
 		goto unlock;
 
-	if (changed)
+	if (changed) {
+		/* In limited privacy mode the change of bondable mode
+		 * may affect the local advertising address.
+		 */
+		if (hdev_is_powered(hdev) &&
+		    hci_dev_test_flag(hdev, HCI_ADVERTISING) &&
+		    hci_dev_test_flag(hdev, HCI_DISCOVERABLE) &&
+		    hci_dev_test_flag(hdev, HCI_LIMITED_PRIVACY))
+			queue_work(hdev->req_workqueue,
+				   &hdev->discoverable_update);
+
 		err = new_settings(hdev, sk);
+	}
 
 unlock:
 	hci_dev_unlock(hdev);
@@ -4423,7 +4434,7 @@ static int set_privacy(struct sock *sk, struct hci_dev *hdev, void *cp_data,
 		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_PRIVACY,
 				       MGMT_STATUS_NOT_SUPPORTED);
 
-	if (cp->privacy != 0x00 && cp->privacy != 0x01)
+	if (cp->privacy != 0x00 && cp->privacy != 0x01 && cp->privacy != 0x02)
 		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_SET_PRIVACY,
 				       MGMT_STATUS_INVALID_PARAMS);
 
@@ -4442,10 +4453,15 @@ static int set_privacy(struct sock *sk, struct hci_dev *hdev, void *cp_data,
 		changed = !hci_dev_test_and_set_flag(hdev, HCI_PRIVACY);
 		memcpy(hdev->irk, cp->irk, sizeof(hdev->irk));
 		hci_dev_set_flag(hdev, HCI_RPA_EXPIRED);
+		if (cp->privacy == 0x02)
+			hci_dev_set_flag(hdev, HCI_LIMITED_PRIVACY);
+		else
+			hci_dev_clear_flag(hdev, HCI_LIMITED_PRIVACY);
 	} else {
 		changed = hci_dev_test_and_clear_flag(hdev, HCI_PRIVACY);
 		memset(hdev->irk, 0, sizeof(hdev->irk));
 		hci_dev_clear_flag(hdev, HCI_RPA_EXPIRED);
+		hci_dev_clear_flag(hdev, HCI_LIMITED_PRIVACY);
 	}
 
 	err = send_settings_rsp(sk, MGMT_OP_SET_PRIVACY, hdev);
-- 
cgit v0.10.2


From 9c176cec42e4b0deeed901fc363bf4235c5ca96a Mon Sep 17 00:00:00 2001
From: Johan Hedberg <johan.hedberg@intel.com>
Date: Wed, 9 Mar 2016 17:52:24 +0200
Subject: Bluetooth: Increment management interface revision

Increment the mgmt revision due to the recently added limited
privacy mode.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 2ca3555..c856fb6 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -38,7 +38,7 @@
 #include "mgmt_util.h"
 
 #define MGMT_VERSION	1
-#define MGMT_REVISION	11
+#define MGMT_REVISION	12
 
 static const u16 mgmt_commands[] = {
 	MGMT_OP_READ_INDEX_LIST,
-- 
cgit v0.10.2


From f720d0caa0af2c33ad15310974c7320345ab4468 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 10 Mar 2016 19:31:12 +0100
Subject: kcm: mark helper functions inline

The stub helper functions for the newly added kcm_proc_init/exit interfaces
are defined as 'static' in a header file, which leads to build warnings for
each file that includes them without calling them:

include/net/kcm.h:183:12: error: 'kcm_proc_init' defined but not used [-Werror=unused-function]
include/net/kcm.h:184:13: error: 'kcm_proc_exit' defined but not used [-Werror=unused-function]

This marks the two functions as 'static inline' instead, which avoids the
warnings and is obviously what was meant here.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: cd6e111bf5be ("kcm: Add statistics and proc interfaces")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/kcm.h b/include/net/kcm.h
index 95c425c..2840b58 100644
--- a/include/net/kcm.h
+++ b/include/net/kcm.h
@@ -180,8 +180,8 @@ struct kcm_mux {
 int kcm_proc_init(void);
 void kcm_proc_exit(void);
 #else
-static int kcm_proc_init(void) { return 0; }
-static void kcm_proc_exit(void) { }
+static inline int kcm_proc_init(void) { return 0; }
+static inline void kcm_proc_exit(void) { }
 #endif
 
 static inline void aggregate_psock_stats(struct kcm_psock_stats *stats,
-- 
cgit v0.10.2


From c1b7fca65070bfadca94dd53a4e6b71cd4f69715 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Tue, 8 Mar 2016 01:36:28 +0300
Subject: sh_eth: fix NULL pointer dereference in sh_eth_ring_format()

In a low memory situation, if netdev_alloc_skb() fails on a first RX ring
loop iteration  in sh_eth_ring_format(), 'rxdesc' is still NULL.  Avoid
kernel oops by adding the 'rxdesc' check after the loop.

Reported-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 9c64489..6fb75a1 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -1154,7 +1154,8 @@ static void sh_eth_ring_format(struct net_device *ndev)
 	mdp->dirty_rx = (u32) (i - mdp->num_rx_ring);
 
 	/* Mark the last entry as wrapping the ring. */
-	rxdesc->status |= cpu_to_le32(RD_RDLE);
+	if (rxdesc)
+		rxdesc->status |= cpu_to_le32(RD_RDLE);
 
 	memset(mdp->tx_ring, 0, tx_ringsize);
 
-- 
cgit v0.10.2


From d0ba913488dc8c55d1880f5ed34f096dc45fb05d Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Tue, 8 Mar 2016 01:37:09 +0300
Subject: sh_eth: advance 'rxdesc' later in sh_eth_ring_format()

Iff dma_map_single() fails, 'rxdesc'  should point  to the last filled RX
descriptor, so  that it can be marked as the last one, however the driver
would have  already  advanced it by that time. In order to fix that, only
fill  an RX descriptor  once all the data for it is ready.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 6fb75a1..2e9a781 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -1127,11 +1127,8 @@ static void sh_eth_ring_format(struct net_device *ndev)
 			break;
 		sh_eth_set_receive_align(skb);
 
-		/* RX descriptor */
-		rxdesc = &mdp->rx_ring[i];
 		/* The size of the buffer is a multiple of 32 bytes. */
 		buf_len = ALIGN(mdp->rx_buf_sz, 32);
-		rxdesc->len = cpu_to_le32(buf_len << 16);
 		dma_addr = dma_map_single(&ndev->dev, skb->data, buf_len,
 					  DMA_FROM_DEVICE);
 		if (dma_mapping_error(&ndev->dev, dma_addr)) {
@@ -1139,6 +1136,10 @@ static void sh_eth_ring_format(struct net_device *ndev)
 			break;
 		}
 		mdp->rx_skbuff[i] = skb;
+
+		/* RX descriptor */
+		rxdesc = &mdp->rx_ring[i];
+		rxdesc->len = cpu_to_le32(buf_len << 16);
 		rxdesc->addr = cpu_to_le32(dma_addr);
 		rxdesc->status = cpu_to_le32(RD_RACT | RD_RFP);
 
-- 
cgit v0.10.2


From 2d9deae4aedee2be2205e22440ac357c37013658 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Mon, 7 Mar 2016 18:24:17 -0500
Subject: net: dsa: mv88e6xxx: rework port state setter

Apply a few non-functional changes on the port state setter:

  * add a dynamic debug message with state names to track changes
  * explicit states checking instead of assuming their numeric values
  * lock mutex only once when changing several port states
  * use bitmap macros to declare and access port_state_update_mask

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index d11c9d5..3a58a8a 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1051,39 +1051,49 @@ static int _mv88e6xxx_atu_remove(struct dsa_switch *ds, u16 fid, int port,
 	return _mv88e6xxx_atu_move(ds, fid, port, 0x0f, static_too);
 }
 
-static int mv88e6xxx_set_port_state(struct dsa_switch *ds, int port, u8 state)
+static const char * const mv88e6xxx_port_state_names[] = {
+	[PORT_CONTROL_STATE_DISABLED] = "Disabled",
+	[PORT_CONTROL_STATE_BLOCKING] = "Blocking/Listening",
+	[PORT_CONTROL_STATE_LEARNING] = "Learning",
+	[PORT_CONTROL_STATE_FORWARDING] = "Forwarding",
+};
+
+static int _mv88e6xxx_port_state(struct dsa_switch *ds, int port, u8 state)
 {
-	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	int reg, ret = 0;
 	u8 oldstate;
 
-	mutex_lock(&ps->smi_mutex);
-
 	reg = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_CONTROL);
-	if (reg < 0) {
-		ret = reg;
-		goto abort;
-	}
+	if (reg < 0)
+		return reg;
 
 	oldstate = reg & PORT_CONTROL_STATE_MASK;
+
 	if (oldstate != state) {
 		/* Flush forwarding database if we're moving a port
 		 * from Learning or Forwarding state to Disabled or
 		 * Blocking or Listening state.
 		 */
-		if (oldstate >= PORT_CONTROL_STATE_LEARNING &&
-		    state <= PORT_CONTROL_STATE_BLOCKING) {
+		if ((oldstate == PORT_CONTROL_STATE_LEARNING ||
+		     oldstate == PORT_CONTROL_STATE_FORWARDING)
+		    && (state == PORT_CONTROL_STATE_DISABLED ||
+			state == PORT_CONTROL_STATE_BLOCKING)) {
 			ret = _mv88e6xxx_atu_remove(ds, 0, port, false);
 			if (ret)
-				goto abort;
+				return ret;
 		}
+
 		reg = (reg & ~PORT_CONTROL_STATE_MASK) | state;
 		ret = _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_CONTROL,
 					   reg);
+		if (ret)
+			return ret;
+
+		netdev_dbg(ds->ports[port], "PortState %s (was %s)\n",
+			   mv88e6xxx_port_state_names[state],
+			   mv88e6xxx_port_state_names[oldstate]);
 	}
 
-abort:
-	mutex_unlock(&ps->smi_mutex);
 	return ret;
 }
 
@@ -1146,13 +1156,11 @@ int mv88e6xxx_port_stp_update(struct dsa_switch *ds, int port, u8 state)
 		break;
 	}
 
-	netdev_dbg(ds->ports[port], "port state %d [%d]\n", state, stp_state);
-
 	/* mv88e6xxx_port_stp_update may be called with softirqs disabled,
 	 * so we can not update the port state directly but need to schedule it.
 	 */
 	ps->ports[port].state = stp_state;
-	set_bit(port, &ps->port_state_update_mask);
+	set_bit(port, ps->port_state_update_mask);
 	schedule_work(&ps->bridge_work);
 
 	return 0;
@@ -2228,11 +2236,15 @@ static void mv88e6xxx_bridge_work(struct work_struct *work)
 	ps = container_of(work, struct mv88e6xxx_priv_state, bridge_work);
 	ds = ((struct dsa_switch *)ps) - 1;
 
-	while (ps->port_state_update_mask) {
-		port = __ffs(ps->port_state_update_mask);
-		clear_bit(port, &ps->port_state_update_mask);
-		mv88e6xxx_set_port_state(ds, port, ps->ports[port].state);
-	}
+	mutex_lock(&ps->smi_mutex);
+
+	for (port = 0; port < ps->num_ports; ++port)
+		if (test_and_clear_bit(port, ps->port_state_update_mask) &&
+		    _mv88e6xxx_port_state(ds, port, ps->ports[port].state))
+			netdev_warn(ds->ports[port], "failed to update state to %s\n",
+				    mv88e6xxx_port_state_names[ps->ports[port].state]);
+
+	mutex_unlock(&ps->smi_mutex);
 }
 
 static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port)
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index d7b088d..3425616 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -426,7 +426,7 @@ struct mv88e6xxx_priv_state {
 
 	struct mv88e6xxx_priv_port	ports[DSA_MAX_PORTS];
 
-	unsigned long port_state_update_mask;
+	DECLARE_BITMAP(port_state_update_mask, DSA_MAX_PORTS);
 
 	struct work_struct bridge_work;
 };
-- 
cgit v0.10.2


From 5da96031834a65e064b97c8d9f7df958c818a4cc Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Mon, 7 Mar 2016 18:24:39 -0500
Subject: net: dsa: mv88e6xxx: read then write PVID

The port register 0x07 contains more options than just the default VID,
even though they are not used yet. So prefer a read then write operation
over a direct write.

This also allows to keep track of the change through dynamic debug.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 3a58a8a..1aee42d 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1166,23 +1166,45 @@ int mv88e6xxx_port_stp_update(struct dsa_switch *ds, int port, u8 state)
 	return 0;
 }
 
-static int _mv88e6xxx_port_pvid_get(struct dsa_switch *ds, int port, u16 *pvid)
+static int _mv88e6xxx_port_pvid(struct dsa_switch *ds, int port, u16 *new,
+				u16 *old)
 {
+	u16 pvid;
 	int ret;
 
 	ret = _mv88e6xxx_reg_read(ds, REG_PORT(port), PORT_DEFAULT_VLAN);
 	if (ret < 0)
 		return ret;
 
-	*pvid = ret & PORT_DEFAULT_VLAN_MASK;
+	pvid = ret & PORT_DEFAULT_VLAN_MASK;
+
+	if (new) {
+		ret &= ~PORT_DEFAULT_VLAN_MASK;
+		ret |= *new & PORT_DEFAULT_VLAN_MASK;
+
+		ret = _mv88e6xxx_reg_write(ds, REG_PORT(port),
+					   PORT_DEFAULT_VLAN, ret);
+		if (ret < 0)
+			return ret;
+
+		netdev_dbg(ds->ports[port], "DefaultVID %d (was %d)\n", *new,
+			   pvid);
+	}
+
+	if (old)
+		*old = pvid;
 
 	return 0;
 }
 
+static int _mv88e6xxx_port_pvid_get(struct dsa_switch *ds, int port, u16 *pvid)
+{
+	return _mv88e6xxx_port_pvid(ds, port, NULL, pvid);
+}
+
 static int _mv88e6xxx_port_pvid_set(struct dsa_switch *ds, int port, u16 pvid)
 {
-	return _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_DEFAULT_VLAN,
-				   pvid & PORT_DEFAULT_VLAN_MASK);
+	return _mv88e6xxx_port_pvid(ds, port, &pvid, NULL);
 }
 
 static int _mv88e6xxx_vtu_wait(struct dsa_switch *ds)
-- 
cgit v0.10.2


From 5220ef1e39577fbe20989e97eac708d8896966f3 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Mon, 7 Mar 2016 18:24:52 -0500
Subject: net: dsa: mv88e6xxx: avoid writing the same mode

There is no need to change the 802.1Q port mode for the same value.
Thus avoid such message:

    [  401.954836] dsa dsa@0 lan0: 802.1Q Mode: Disabled (was Disabled)

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 1aee42d..5f07524 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1765,16 +1765,21 @@ int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port,
 
 	old = ret & PORT_CONTROL_2_8021Q_MASK;
 
-	ret &= ~PORT_CONTROL_2_8021Q_MASK;
-	ret |= new & PORT_CONTROL_2_8021Q_MASK;
+	if (new != old) {
+		ret &= ~PORT_CONTROL_2_8021Q_MASK;
+		ret |= new & PORT_CONTROL_2_8021Q_MASK;
 
-	ret = _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_CONTROL_2, ret);
-	if (ret < 0)
-		goto unlock;
+		ret = _mv88e6xxx_reg_write(ds, REG_PORT(port), PORT_CONTROL_2,
+					   ret);
+		if (ret < 0)
+			goto unlock;
+
+		netdev_dbg(ds->ports[port], "802.1Q Mode %s (was %s)\n",
+			   mv88e6xxx_port_8021q_mode_names[new],
+			   mv88e6xxx_port_8021q_mode_names[old]);
+	}
 
-	netdev_dbg(ds->ports[port], "802.1Q Mode: %s (was %s)\n",
-		   mv88e6xxx_port_8021q_mode_names[new],
-		   mv88e6xxx_port_8021q_mode_names[old]);
+	ret = 0;
 unlock:
 	mutex_unlock(&ps->smi_mutex);
 
-- 
cgit v0.10.2


From 84f670189b8c0dfd3dbaf12da8c946225f4011d3 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 8 Mar 2016 10:50:17 +0530
Subject: cxgb4vf: Enable interrupts before we register our network devices

This avoids a race condition where a system that has network devices set up
to be automatically configured and we get the first Port Link Status
message from the firmware on the Asynchronous Firmware Event Queue before
we've enabled interrupts. If that happens, we end up losing the interrupt
and never realizing that the links has actually come up.

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 91857b8..fcafe34 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -2771,6 +2771,24 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 		}
 	}
 
+	/* See what interrupts we'll be using.  If we've been configured to
+	 * use MSI-X interrupts, try to enable them but fall back to using
+	 * MSI interrupts if we can't enable MSI-X interrupts.  If we can't
+	 * get MSI interrupts we bail with the error.
+	 */
+	if (msi == MSI_MSIX && enable_msix(adapter) == 0)
+		adapter->flags |= USING_MSIX;
+	else {
+		err = pci_enable_msi(pdev);
+		if (err) {
+			dev_err(&pdev->dev, "Unable to allocate %s interrupts;"
+				" err=%d\n",
+				msi == MSI_MSIX ? "MSI-X or MSI" : "MSI", err);
+			goto err_free_dev;
+		}
+		adapter->flags |= USING_MSI;
+	}
+
 	/*
 	 * The "card" is now ready to go.  If any errors occur during device
 	 * registration we do not fail the whole "card" but rather proceed
@@ -2793,7 +2811,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	}
 	if (adapter->registered_device_map == 0) {
 		dev_err(&pdev->dev, "could not register any net devices\n");
-		goto err_free_dev;
+		goto err_disable_interrupts;
 	}
 
 	/*
@@ -2811,25 +2829,6 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	}
 
 	/*
-	 * See what interrupts we'll be using.  If we've been configured to
-	 * use MSI-X interrupts, try to enable them but fall back to using
-	 * MSI interrupts if we can't enable MSI-X interrupts.  If we can't
-	 * get MSI interrupts we bail with the error.
-	 */
-	if (msi == MSI_MSIX && enable_msix(adapter) == 0)
-		adapter->flags |= USING_MSIX;
-	else {
-		err = pci_enable_msi(pdev);
-		if (err) {
-			dev_err(&pdev->dev, "Unable to allocate %s interrupts;"
-				" err=%d\n",
-				msi == MSI_MSIX ? "MSI-X or MSI" : "MSI", err);
-			goto err_free_debugfs;
-		}
-		adapter->flags |= USING_MSI;
-	}
-
-	/*
 	 * Now that we know how many "ports" we have and what their types are,
 	 * and how many Queue Sets we can support, we can configure our queue
 	 * resources.
@@ -2856,11 +2855,13 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	 * Error recovery and exit code.  Unwind state that's been created
 	 * so far and return the error.
 	 */
-
-err_free_debugfs:
-	if (!IS_ERR_OR_NULL(adapter->debugfs_root)) {
-		cleanup_debugfs(adapter);
-		debugfs_remove_recursive(adapter->debugfs_root);
+err_disable_interrupts:
+	if (adapter->flags & USING_MSIX) {
+		pci_disable_msix(adapter->pdev);
+		adapter->flags &= ~USING_MSIX;
+	} else if (adapter->flags & USING_MSI) {
+		pci_disable_msi(adapter->pdev);
+		adapter->flags &= ~USING_MSI;
 	}
 
 err_free_dev:
-- 
cgit v0.10.2


From 495c22bbb2b03548f6aa870faa7f6b601cb41c85 Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 8 Mar 2016 10:50:18 +0530
Subject: cxgb4vf: Configure queue based on resource and interrupt type

The Queue Set Configuration code was always reserving room for a
Forwarded interrupt Queue even in the cases where we weren't using it.
Figure out how many Ports and Queue Sets we can support. This depends on
knowing our Virtual Function Resources and may be called a second time
if we fall back from MSI-X to MSI Interrupt Mode. This change fixes that
problem.

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index fcafe34..17a3153 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -2176,6 +2176,73 @@ static void cleanup_debugfs(struct adapter *adapter)
 	/* nothing to do */
 }
 
+/* Figure out how many Ports and Queue Sets we can support.  This depends on
+ * knowing our Virtual Function Resources and may be called a second time if
+ * we fall back from MSI-X to MSI Interrupt Mode.
+ */
+static void size_nports_qsets(struct adapter *adapter)
+{
+	struct vf_resources *vfres = &adapter->params.vfres;
+	unsigned int ethqsets, pmask_nports;
+
+	/* The number of "ports" which we support is equal to the number of
+	 * Virtual Interfaces with which we've been provisioned.
+	 */
+	adapter->params.nports = vfres->nvi;
+	if (adapter->params.nports > MAX_NPORTS) {
+		dev_warn(adapter->pdev_dev, "only using %d of %d maximum"
+			 " allowed virtual interfaces\n", MAX_NPORTS,
+			 adapter->params.nports);
+		adapter->params.nports = MAX_NPORTS;
+	}
+
+	/* We may have been provisioned with more VIs than the number of
+	 * ports we're allowed to access (our Port Access Rights Mask).
+	 * This is obviously a configuration conflict but we don't want to
+	 * crash the kernel or anything silly just because of that.
+	 */
+	pmask_nports = hweight32(adapter->params.vfres.pmask);
+	if (pmask_nports < adapter->params.nports) {
+		dev_warn(adapter->pdev_dev, "only using %d of %d provissioned"
+			 " virtual interfaces; limited by Port Access Rights"
+			 " mask %#x\n", pmask_nports, adapter->params.nports,
+			 adapter->params.vfres.pmask);
+		adapter->params.nports = pmask_nports;
+	}
+
+	/* We need to reserve an Ingress Queue for the Asynchronous Firmware
+	 * Event Queue.  And if we're using MSI Interrupts, we'll also need to
+	 * reserve an Ingress Queue for a Forwarded Interrupts.
+	 *
+	 * The rest of the FL/Intr-capable ingress queues will be matched up
+	 * one-for-one with Ethernet/Control egress queues in order to form
+	 * "Queue Sets" which will be aportioned between the "ports".  For
+	 * each Queue Set, we'll need the ability to allocate two Egress
+	 * Contexts -- one for the Ingress Queue Free List and one for the TX
+	 * Ethernet Queue.
+	 *
+	 * Note that even if we're currently configured to use MSI-X
+	 * Interrupts (module variable msi == MSI_MSIX) we may get downgraded
+	 * to MSI Interrupts if we can't get enough MSI-X Interrupts.  If that
+	 * happens we'll need to adjust things later.
+	 */
+	ethqsets = vfres->niqflint - 1 - (msi == MSI_MSI);
+	if (vfres->nethctrl != ethqsets)
+		ethqsets = min(vfres->nethctrl, ethqsets);
+	if (vfres->neq < ethqsets*2)
+		ethqsets = vfres->neq/2;
+	if (ethqsets > MAX_ETH_QSETS)
+		ethqsets = MAX_ETH_QSETS;
+	adapter->sge.max_ethqsets = ethqsets;
+
+	if (adapter->sge.max_ethqsets < adapter->params.nports) {
+		dev_warn(adapter->pdev_dev, "only using %d of %d available"
+			 " virtual interfaces (too few Queue Sets)\n",
+			 adapter->sge.max_ethqsets, adapter->params.nports);
+		adapter->params.nports = adapter->sge.max_ethqsets;
+	}
+}
+
 /*
  * Perform early "adapter" initialization.  This is where we discover what
  * adapter parameters we're going to be using and initialize basic adapter
@@ -2183,10 +2250,8 @@ static void cleanup_debugfs(struct adapter *adapter)
  */
 static int adap_init0(struct adapter *adapter)
 {
-	struct vf_resources *vfres = &adapter->params.vfres;
 	struct sge_params *sge_params = &adapter->params.sge;
 	struct sge *s = &adapter->sge;
-	unsigned int ethqsets;
 	int err;
 	u32 param, val = 0;
 
@@ -2295,69 +2360,18 @@ static int adap_init0(struct adapter *adapter)
 		return err;
 	}
 
-	/*
-	 * The number of "ports" which we support is equal to the number of
-	 * Virtual Interfaces with which we've been provisioned.
-	 */
-	adapter->params.nports = vfres->nvi;
-	if (adapter->params.nports > MAX_NPORTS) {
-		dev_warn(adapter->pdev_dev, "only using %d of %d allowed"
-			 " virtual interfaces\n", MAX_NPORTS,
-			 adapter->params.nports);
-		adapter->params.nports = MAX_NPORTS;
-	}
-
-	/*
-	 * We need to reserve a number of the ingress queues with Free List
-	 * and Interrupt capabilities for special interrupt purposes (like
-	 * asynchronous firmware messages, or forwarded interrupts if we're
-	 * using MSI).  The rest of the FL/Intr-capable ingress queues will be
-	 * matched up one-for-one with Ethernet/Control egress queues in order
-	 * to form "Queue Sets" which will be aportioned between the "ports".
-	 * For each Queue Set, we'll need the ability to allocate two Egress
-	 * Contexts -- one for the Ingress Queue Free List and one for the TX
-	 * Ethernet Queue.
-	 */
-	ethqsets = vfres->niqflint - INGQ_EXTRAS;
-	if (vfres->nethctrl != ethqsets) {
-		dev_warn(adapter->pdev_dev, "unequal number of [available]"
-			 " ingress/egress queues (%d/%d); using minimum for"
-			 " number of Queue Sets\n", ethqsets, vfres->nethctrl);
-		ethqsets = min(vfres->nethctrl, ethqsets);
-	}
-	if (vfres->neq < ethqsets*2) {
-		dev_warn(adapter->pdev_dev, "Not enough Egress Contexts (%d)"
-			 " to support Queue Sets (%d); reducing allowed Queue"
-			 " Sets\n", vfres->neq, ethqsets);
-		ethqsets = vfres->neq/2;
-	}
-	if (ethqsets > MAX_ETH_QSETS) {
-		dev_warn(adapter->pdev_dev, "only using %d of %d allowed Queue"
-			 " Sets\n", MAX_ETH_QSETS, adapter->sge.max_ethqsets);
-		ethqsets = MAX_ETH_QSETS;
-	}
-	if (vfres->niq != 0 || vfres->neq > ethqsets*2) {
-		dev_warn(adapter->pdev_dev, "unused resources niq/neq (%d/%d)"
-			 " ignored\n", vfres->niq, vfres->neq - ethqsets*2);
-	}
-	adapter->sge.max_ethqsets = ethqsets;
-
-	/*
-	 * Check for various parameter sanity issues.  Most checks simply
-	 * result in us using fewer resources than our provissioning but we
-	 * do need at least  one "port" with which to work ...
-	 */
-	if (adapter->sge.max_ethqsets < adapter->params.nports) {
-		dev_warn(adapter->pdev_dev, "only using %d of %d available"
-			 " virtual interfaces (too few Queue Sets)\n",
-			 adapter->sge.max_ethqsets, adapter->params.nports);
-		adapter->params.nports = adapter->sge.max_ethqsets;
-	}
-	if (adapter->params.nports == 0) {
+	/* Check for various parameter sanity issues */
+	if (adapter->params.vfres.nvi == 0) {
 		dev_err(adapter->pdev_dev, "no virtual interfaces configured/"
 			"usable!\n");
 		return -EINVAL;
 	}
+
+	/* Initialize nports and max_ethqsets now that we have our Virtual
+	 * Function Resources.
+	 */
+	size_nports_qsets(adapter);
+
 	return 0;
 }
 
@@ -2779,16 +2793,32 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	if (msi == MSI_MSIX && enable_msix(adapter) == 0)
 		adapter->flags |= USING_MSIX;
 	else {
+		if (msi == MSI_MSIX) {
+			dev_info(adapter->pdev_dev,
+				 "Unable to use MSI-X Interrupts; falling "
+				 "back to MSI Interrupts\n");
+
+			/* We're going to need a Forwarded Interrupt Queue so
+			 * that may cut into how many Queue Sets we can
+			 * support.
+			 */
+			msi = MSI_MSI;
+			size_nports_qsets(adapter);
+		}
 		err = pci_enable_msi(pdev);
 		if (err) {
-			dev_err(&pdev->dev, "Unable to allocate %s interrupts;"
-				" err=%d\n",
-				msi == MSI_MSIX ? "MSI-X or MSI" : "MSI", err);
+			dev_err(&pdev->dev, "Unable to allocate MSI Interrupts;"
+				" err=%d\n", err);
 			goto err_free_dev;
 		}
 		adapter->flags |= USING_MSI;
 	}
 
+	/* Now that we know how many "ports" we have and what interrupt
+	 * mechanism we're going to use, we can configure our queue resources.
+	 */
+	cfg_queues(adapter);
+
 	/*
 	 * The "card" is now ready to go.  If any errors occur during device
 	 * registration we do not fail the whole "card" but rather proceed
@@ -2829,13 +2859,6 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	}
 
 	/*
-	 * Now that we know how many "ports" we have and what their types are,
-	 * and how many Queue Sets we can support, we can configure our queue
-	 * resources.
-	 */
-	cfg_queues(adapter);
-
-	/*
 	 * Print a short notice on the existence and configuration of the new
 	 * VF network device ...
 	 */
-- 
cgit v0.10.2


From 28f71c6df4c5399e4f30ce37c1fff0f6af3d364d Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 8 Mar 2016 10:50:19 +0530
Subject: cxgb4vf: Add a couple more checks for invalid provisioning
 configurations

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 17a3153..5a3b883 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -2361,6 +2361,11 @@ static int adap_init0(struct adapter *adapter)
 	}
 
 	/* Check for various parameter sanity issues */
+	if (adapter->params.vfres.pmask == 0) {
+		dev_err(adapter->pdev_dev, "no port access configured\n"
+			"usable!\n");
+		return -EINVAL;
+	}
 	if (adapter->params.vfres.nvi == 0) {
 		dev_err(adapter->pdev_dev, "no virtual interfaces configured/"
 			"usable!\n");
-- 
cgit v0.10.2


From a8d16d08065f8d3f2fca4a4c377fc4bfc5bdfccd Mon Sep 17 00:00:00 2001
From: Hariprasad Shenai <hariprasad@chelsio.com>
Date: Tue, 8 Mar 2016 10:50:20 +0530
Subject: cxgb4vf: Set number of queues in pci probe only

Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 5a3b883..1cc8a7a 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -790,10 +790,6 @@ static int cxgb4vf_open(struct net_device *dev)
 	/*
 	 * Note that this interface is up and start everything up ...
 	 */
-	netif_set_real_num_tx_queues(dev, pi->nqsets);
-	err = netif_set_real_num_rx_queues(dev, pi->nqsets);
-	if (err)
-		goto err_unwind;
 	err = link_start(dev);
 	if (err)
 		goto err_unwind;
@@ -2831,10 +2827,14 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev,
 	 * must register at least one net device.
 	 */
 	for_each_port(adapter, pidx) {
+		struct port_info *pi = netdev_priv(adapter->port[pidx]);
 		netdev = adapter->port[pidx];
 		if (netdev == NULL)
 			continue;
 
+		netif_set_real_num_tx_queues(netdev, pi->nqsets);
+		netif_set_real_num_rx_queues(netdev, pi->nqsets);
+
 		err = register_netdev(netdev);
 		if (err) {
 			dev_warn(&pdev->dev, "cannot register net device %s,"
-- 
cgit v0.10.2


From 5bf93251cee1fb66141d1d2eaff86e04a9397bdf Mon Sep 17 00:00:00 2001
From: Rajesh Borundia <rajesh.borundia@qlogic.com>
Date: Tue, 8 Mar 2016 02:39:57 -0500
Subject: qlcnic: Remove unnecessary usage of atomic_t

o atomic_t usage is incorrect as we are not implementing
any atomicity.

Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
index 46bbea8..d18667b 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
@@ -1099,7 +1099,7 @@ struct qlcnic_mailbox {
 	unsigned long		status;
 	spinlock_t		queue_lock;	/* Mailbox queue lock */
 	spinlock_t		aen_lock;	/* Mailbox response/AEN lock */
-	atomic_t		rsp_status;
+	u32			rsp_status;
 	u32			num_cmds;
 };
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
index 37a731b..e3d1bb7 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
@@ -491,7 +491,7 @@ irqreturn_t qlcnic_83xx_clear_legacy_intr(struct qlcnic_adapter *adapter)
 
 static inline void qlcnic_83xx_notify_mbx_response(struct qlcnic_mailbox *mbx)
 {
-	atomic_set(&mbx->rsp_status, QLC_83XX_MBX_RESPONSE_ARRIVED);
+	mbx->rsp_status = QLC_83XX_MBX_RESPONSE_ARRIVED;
 	complete(&mbx->completion);
 }
 
@@ -510,7 +510,7 @@ static void qlcnic_83xx_poll_process_aen(struct qlcnic_adapter *adapter)
 	if (event &  QLCNIC_MBX_ASYNC_EVENT) {
 		__qlcnic_83xx_process_aen(adapter);
 	} else {
-		if (atomic_read(&mbx->rsp_status) != rsp_status)
+		if (mbx->rsp_status != rsp_status)
 			qlcnic_83xx_notify_mbx_response(mbx);
 	}
 out:
@@ -1023,7 +1023,7 @@ static void qlcnic_83xx_process_aen(struct qlcnic_adapter *adapter)
 		if (event &  QLCNIC_MBX_ASYNC_EVENT) {
 			__qlcnic_83xx_process_aen(adapter);
 		} else {
-			if (atomic_read(&mbx->rsp_status) != rsp_status)
+			if (mbx->rsp_status != rsp_status)
 				qlcnic_83xx_notify_mbx_response(mbx);
 		}
 	}
@@ -4050,7 +4050,6 @@ static void qlcnic_83xx_mailbox_worker(struct work_struct *work)
 	struct qlcnic_adapter *adapter = mbx->adapter;
 	const struct qlcnic_mbx_ops *mbx_ops = mbx->ops;
 	struct device *dev = &adapter->pdev->dev;
-	atomic_t *rsp_status = &mbx->rsp_status;
 	struct list_head *head = &mbx->cmd_q;
 	struct qlcnic_hardware_context *ahw;
 	struct qlcnic_cmd_args *cmd = NULL;
@@ -4063,7 +4062,7 @@ static void qlcnic_83xx_mailbox_worker(struct work_struct *work)
 			return;
 		}
 
-		atomic_set(rsp_status, QLC_83XX_MBX_RESPONSE_WAIT);
+		mbx->rsp_status = QLC_83XX_MBX_RESPONSE_WAIT;
 
 		spin_lock(&mbx->queue_lock);
 
-- 
cgit v0.10.2


From 819bfe764dceec2f6b4551768453f374b4c60443 Mon Sep 17 00:00:00 2001
From: Rajesh Borundia <rajesh.borundia@qlogic.com>
Date: Tue, 8 Mar 2016 02:39:58 -0500
Subject: qlcnic: Fix mailbox completion handling during spurious interrupt

o While the driver is in the middle of a MB completion processing
and it receives a spurious MB interrupt, it is mistaken as a good MB
completion interrupt leading to premature completion of the next MB
request. Fix the driver to guard against this by checking the current
state of MB processing and ignore the spurious interrupt.
Also added a stats counter to record this condition.

Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
index d18667b..55007f1 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
@@ -566,6 +566,7 @@ struct qlcnic_adapter_stats {
 	u64  tx_dma_map_error;
 	u64  spurious_intr;
 	u64  mac_filter_limit_overrun;
+	u64  mbx_spurious_intr;
 };
 
 /*
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
index e3d1bb7..f9640d5ce 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c
@@ -2338,9 +2338,9 @@ static void qlcnic_83xx_handle_link_aen(struct qlcnic_adapter *adapter,
 
 static irqreturn_t qlcnic_83xx_handle_aen(int irq, void *data)
 {
+	u32 mask, resp, event, rsp_status = QLC_83XX_MBX_RESPONSE_ARRIVED;
 	struct qlcnic_adapter *adapter = data;
 	struct qlcnic_mailbox *mbx;
-	u32 mask, resp, event;
 	unsigned long flags;
 
 	mbx = adapter->ahw->mailbox;
@@ -2350,10 +2350,14 @@ static irqreturn_t qlcnic_83xx_handle_aen(int irq, void *data)
 		goto out;
 
 	event = readl(QLCNIC_MBX_FW(adapter->ahw, 0));
-	if (event &  QLCNIC_MBX_ASYNC_EVENT)
+	if (event &  QLCNIC_MBX_ASYNC_EVENT) {
 		__qlcnic_83xx_process_aen(adapter);
-	else
-		qlcnic_83xx_notify_mbx_response(mbx);
+	} else {
+		if (mbx->rsp_status != rsp_status)
+			qlcnic_83xx_notify_mbx_response(mbx);
+		else
+			adapter->stats.mbx_spurious_intr++;
+	}
 
 out:
 	mask = QLCRDX(adapter->ahw, QLCNIC_DEF_INT_MASK);
@@ -4053,6 +4057,7 @@ static void qlcnic_83xx_mailbox_worker(struct work_struct *work)
 	struct list_head *head = &mbx->cmd_q;
 	struct qlcnic_hardware_context *ahw;
 	struct qlcnic_cmd_args *cmd = NULL;
+	unsigned long flags;
 
 	ahw = adapter->ahw;
 
@@ -4062,7 +4067,9 @@ static void qlcnic_83xx_mailbox_worker(struct work_struct *work)
 			return;
 		}
 
+		spin_lock_irqsave(&mbx->aen_lock, flags);
 		mbx->rsp_status = QLC_83XX_MBX_RESPONSE_WAIT;
+		spin_unlock_irqrestore(&mbx->aen_lock, flags);
 
 		spin_lock(&mbx->queue_lock);
 
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
index 494e810..0a2318c 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c
@@ -59,7 +59,8 @@ static const struct qlcnic_stats qlcnic_gstrings_stats[] = {
 	 QLC_OFF(stats.mac_filter_limit_overrun)},
 	{"spurious intr", QLC_SIZEOF(stats.spurious_intr),
 	 QLC_OFF(stats.spurious_intr)},
-
+	{"mbx spurious intr", QLC_SIZEOF(stats.mbx_spurious_intr),
+	 QLC_OFF(stats.mbx_spurious_intr)},
 };
 
 static const char qlcnic_device_gstrings_stats[][ETH_GSTRING_LEN] = {
-- 
cgit v0.10.2


From 88f09bd5b9875a0fbf2075221590d9c4418cdbbc Mon Sep 17 00:00:00 2001
From: Manish Chopra <manish.chopra@qlogic.com>
Date: Tue, 8 Mar 2016 04:09:44 -0500
Subject: qede: Fix net-next "make ARCH=x86_64"

'commit 55482edc25f0606851de42e73618f813f310d009
("qede: Add slowpath/fastpath support and enable hardware GRO")'
introduces below error when compiling net-next with "make ARCH=x86_64"

drivers/built-in.o: In function `qede_rx_int':
qede_main.c:(.text+0x6101a0): undefined reference to `tcp_gro_complete'

Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index 5728625..518af32 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c
@@ -1001,6 +1001,7 @@ static void qede_tpa_start(struct qede_dev *edev,
 	}
 }
 
+#ifdef CONFIG_INET
 static void qede_gro_ip_csum(struct sk_buff *skb)
 {
 	const struct iphdr *iph = ip_hdr(skb);
@@ -1029,12 +1030,14 @@ static void qede_gro_ipv6_csum(struct sk_buff *skb)
 				  &iph->saddr, &iph->daddr, 0);
 	tcp_gro_complete(skb);
 }
+#endif
 
 static void qede_gro_receive(struct qede_dev *edev,
 			     struct qede_fastpath *fp,
 			     struct sk_buff *skb,
 			     u16 vlan_tag)
 {
+#ifdef CONFIG_INET
 	if (skb_shinfo(skb)->gso_size) {
 		switch (skb->protocol) {
 		case htons(ETH_P_IP):
@@ -1049,7 +1052,7 @@ static void qede_gro_receive(struct qede_dev *edev,
 			       ntohs(skb->protocol));
 		}
 	}
-
+#endif
 	skb_record_rx_queue(skb, fp->rss_id);
 	qede_skb_receive(edev, fp, skb, vlan_tag);
 }
-- 
cgit v0.10.2


From 04761890a7cec6a1ff9aafd909004da4fe8059db Mon Sep 17 00:00:00 2001
From: Neil Armstrong <narmstrong@baylibre.com>
Date: Tue, 8 Mar 2016 10:36:20 +0100
Subject: net: dsa: Fix cleanup resources upon module removal

The initial commit badly merged into the dsa_resume method instead
of the dsa_remove_dst method.
As consequence, the dst->master_netdev->dsa_ptr is not set to NULL on
removal and re-bind of the dsa device fails with error -17.

Fixes: b0dc635d923c ("net: dsa: cleanup resources upon module removal ")
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Acked-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index fa4daba..d8fb47f 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -935,6 +935,14 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst)
 {
 	int i;
 
+	dst->master_netdev->dsa_ptr = NULL;
+
+	/* If we used a tagging format that doesn't have an ethertype
+	 * field, make sure that all packets from this point get sent
+	 * without the tag and go through the regular receive path.
+	 */
+	wmb();
+
 	for (i = 0; i < dst->pd->nr_chips; i++) {
 		struct dsa_switch *ds = dst->ds[i];
 
@@ -988,14 +996,6 @@ static int dsa_suspend(struct device *d)
 	struct dsa_switch_tree *dst = platform_get_drvdata(pdev);
 	int i, ret = 0;
 
-	dst->master_netdev->dsa_ptr = NULL;
-
-	/* If we used a tagging format that doesn't have an ethertype
-	 * field, make sure that all packets from this point get sent
-	 * without the tag and go through the regular receive path.
-	 */
-	wmb();
-
 	for (i = 0; i < dst->pd->nr_chips; i++) {
 		struct dsa_switch *ds = dst->ds[i];
 
-- 
cgit v0.10.2


From 58ff9865b75dab73ccfae89bc8313ca2497b4c8f Mon Sep 17 00:00:00 2001
From: John Crispin <blogic@openwrt.org>
Date: Tue, 8 Mar 2016 11:29:54 +0100
Subject: net-next: mediatek: document MediaTek SoC ethernet binding

This adds the binding documentation for the MediaTek Ethernet
controller.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Rob Herring <robh@kernel.org>
Cc: devicetree@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/mediatek-net.txt b/Documentation/devicetree/bindings/net/mediatek-net.txt
new file mode 100644
index 0000000..5ca7929
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/mediatek-net.txt
@@ -0,0 +1,77 @@
+MediaTek Frame Engine Ethernet controller
+=========================================
+
+The frame engine ethernet controller can be found on MediaTek SoCs. These SoCs
+have dual GMAC each represented by a child node..
+
+* Ethernet controller node
+
+Required properties:
+- compatible: Should be "mediatek,mt7623-eth"
+- reg: Address and length of the register set for the device
+- interrupts: Should contain the frame engines interrupt
+- clocks: the clock used by the core
+- clock-names: the names of the clock listed in the clocks property. These are
+	"ethif", "esw", "gp2", "gp1"
+- power-domains: phandle to the power domain that the ethernet is part of
+- resets: Should contain a phandle to the ethsys reset signal
+- reset-names: Should contain the reset signal name "eth"
+- mediatek,ethsys: phandle to the syscon node that handles the port setup
+- mediatek,pctl: phandle to the syscon node that handles the ports slew rate
+	and driver current
+
+Optional properties:
+- interrupt-parent: Should be the phandle for the interrupt controller
+  that services interrupts for this device
+
+
+* Ethernet MAC node
+
+Required properties:
+- compatible: Should be "mediatek,eth-mac"
+- reg: The number of the MAC
+- phy-handle: see ethernet.txt file in the same directory.
+
+Example:
+
+eth: ethernet@1b100000 {
+	compatible = "mediatek,mt7623-eth";
+	reg = <0 0x1b100000 0 0x20000>;
+	clocks = <&topckgen CLK_TOP_ETHIF_SEL>,
+		 <&ethsys CLK_ETHSYS_ESW>,
+		 <&ethsys CLK_ETHSYS_GP2>,
+		 <&ethsys CLK_ETHSYS_GP1>;
+	clock-names = "ethif", "esw", "gp2", "gp1";
+	interrupts = <GIC_SPI 200 IRQ_TYPE_LEVEL_LOW>;
+	power-domains = <&scpsys MT2701_POWER_DOMAIN_ETH>;
+	resets = <&ethsys MT2701_ETHSYS_ETH_RST>;
+	reset-names = "eth";
+	mediatek,ethsys = <&ethsys>;
+	mediatek,pctl = <&syscfg_pctl_a>;
+	#address-cells = <1>;
+	#size-cells = <0>;
+
+	gmac1: mac@0 {
+		compatible = "mediatek,eth-mac";
+		reg = <0>;
+		phy-handle = <&phy0>;
+	};
+
+	gmac2: mac@1 {
+		compatible = "mediatek,eth-mac";
+		reg = <1>;
+		phy-handle = <&phy1>;
+	};
+
+	mdio-bus {
+		phy0: ethernet-phy@0 {
+			reg = <0>;
+			phy-mode = "rgmii";
+		};
+
+		phy1: ethernet-phy@1 {
+			reg = <1>;
+			phy-mode = "rgmii";
+		};
+	};
+};
-- 
cgit v0.10.2


From 656e705243fd0c2864b89634ea16ed444ef64dc6 Mon Sep 17 00:00:00 2001
From: John Crispin <blogic@openwrt.org>
Date: Tue, 8 Mar 2016 11:29:55 +0100
Subject: net-next: mediatek: add support for MT7623 ethernet

Add ethernet support for MediaTek SoCs from the MT7623 family. These have
dual GMAC. Depending on the exact version, there might be a built-in
Gigabit switch (MT7530). The core does not have the typical DMA ring setup.
Instead there is a linked list that we add descriptors to. There is only
one linked list that both MACs use together. There is a special field
inside the TX descriptors called the VQID. This allows us to assign packets
to different internal queues. By using a separate id for each MAC we are
able to get deterministic results for BQL. Additionally we need to
provide the core with a block of scratch memory that is the same size as
the RX ring and data buffer. This is really needed to make the HW datapath
work. Although the driver does not support this yet, we still need to
assign the memory and tell the core about it for RX to work.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Michael Lee <igvtee@gmail.com>
Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
new file mode 100644
index 0000000..ba3afa5
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -0,0 +1,1807 @@
+/*   This program is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; version 2 of the License
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ *
+ *   Copyright (C) 2009-2016 John Crispin <blogic@openwrt.org>
+ *   Copyright (C) 2009-2016 Felix Fietkau <nbd@openwrt.org>
+ *   Copyright (C) 2013-2016 Michael Lee <igvtee@gmail.com>
+ */
+
+#include <linux/of_device.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
+#include <linux/clk.h>
+#include <linux/if_vlan.h>
+#include <linux/reset.h>
+#include <linux/tcp.h>
+
+#include "mtk_eth_soc.h"
+
+static int mtk_msg_level = -1;
+module_param_named(msg_level, mtk_msg_level, int, 0);
+MODULE_PARM_DESC(msg_level, "Message level (-1=defaults,0=none,...,16=all)");
+
+#define MTK_ETHTOOL_STAT(x) { #x, \
+			      offsetof(struct mtk_hw_stats, x) / sizeof(u64) }
+
+/* strings used by ethtool */
+static const struct mtk_ethtool_stats {
+	char str[ETH_GSTRING_LEN];
+	u32 offset;
+} mtk_ethtool_stats[] = {
+	MTK_ETHTOOL_STAT(tx_bytes),
+	MTK_ETHTOOL_STAT(tx_packets),
+	MTK_ETHTOOL_STAT(tx_skip),
+	MTK_ETHTOOL_STAT(tx_collisions),
+	MTK_ETHTOOL_STAT(rx_bytes),
+	MTK_ETHTOOL_STAT(rx_packets),
+	MTK_ETHTOOL_STAT(rx_overflow),
+	MTK_ETHTOOL_STAT(rx_fcs_errors),
+	MTK_ETHTOOL_STAT(rx_short_errors),
+	MTK_ETHTOOL_STAT(rx_long_errors),
+	MTK_ETHTOOL_STAT(rx_checksum_errors),
+	MTK_ETHTOOL_STAT(rx_flow_control_packets),
+};
+
+void mtk_w32(struct mtk_eth *eth, u32 val, unsigned reg)
+{
+	__raw_writel(val, eth->base + reg);
+}
+
+u32 mtk_r32(struct mtk_eth *eth, unsigned reg)
+{
+	return __raw_readl(eth->base + reg);
+}
+
+static int mtk_mdio_busy_wait(struct mtk_eth *eth)
+{
+	unsigned long t_start = jiffies;
+
+	while (1) {
+		if (!(mtk_r32(eth, MTK_PHY_IAC) & PHY_IAC_ACCESS))
+			return 0;
+		if (time_after(jiffies, t_start + PHY_IAC_TIMEOUT))
+			break;
+		usleep_range(10, 20);
+	}
+
+	dev_err(eth->dev, "mdio: MDIO timeout\n");
+	return -1;
+}
+
+u32 _mtk_mdio_write(struct mtk_eth *eth, u32 phy_addr,
+		    u32 phy_register, u32 write_data)
+{
+	if (mtk_mdio_busy_wait(eth))
+		return -1;
+
+	write_data &= 0xffff;
+
+	mtk_w32(eth, PHY_IAC_ACCESS | PHY_IAC_START | PHY_IAC_WRITE |
+		(phy_register << PHY_IAC_REG_SHIFT) |
+		(phy_addr << PHY_IAC_ADDR_SHIFT) | write_data,
+		MTK_PHY_IAC);
+
+	if (mtk_mdio_busy_wait(eth))
+		return -1;
+
+	return 0;
+}
+
+u32 _mtk_mdio_read(struct mtk_eth *eth, int phy_addr, int phy_reg)
+{
+	u32 d;
+
+	if (mtk_mdio_busy_wait(eth))
+		return 0xffff;
+
+	mtk_w32(eth, PHY_IAC_ACCESS | PHY_IAC_START | PHY_IAC_READ |
+		(phy_reg << PHY_IAC_REG_SHIFT) |
+		(phy_addr << PHY_IAC_ADDR_SHIFT),
+		MTK_PHY_IAC);
+
+	if (mtk_mdio_busy_wait(eth))
+		return 0xffff;
+
+	d = mtk_r32(eth, MTK_PHY_IAC) & 0xffff;
+
+	return d;
+}
+
+static int mtk_mdio_write(struct mii_bus *bus, int phy_addr,
+			  int phy_reg, u16 val)
+{
+	struct mtk_eth *eth = bus->priv;
+
+	return _mtk_mdio_write(eth, phy_addr, phy_reg, val);
+}
+
+static int mtk_mdio_read(struct mii_bus *bus, int phy_addr, int phy_reg)
+{
+	struct mtk_eth *eth = bus->priv;
+
+	return _mtk_mdio_read(eth, phy_addr, phy_reg);
+}
+
+static void mtk_phy_link_adjust(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	u32 mcr = MAC_MCR_MAX_RX_1536 | MAC_MCR_IPG_CFG |
+		  MAC_MCR_FORCE_MODE | MAC_MCR_TX_EN |
+		  MAC_MCR_RX_EN | MAC_MCR_BACKOFF_EN |
+		  MAC_MCR_BACKPR_EN;
+
+	switch (mac->phy_dev->speed) {
+	case SPEED_1000:
+		mcr |= MAC_MCR_SPEED_1000;
+		break;
+	case SPEED_100:
+		mcr |= MAC_MCR_SPEED_100;
+		break;
+	};
+
+	if (mac->phy_dev->link)
+		mcr |= MAC_MCR_FORCE_LINK;
+
+	if (mac->phy_dev->duplex)
+		mcr |= MAC_MCR_FORCE_DPX;
+
+	if (mac->phy_dev->pause)
+		mcr |= MAC_MCR_FORCE_RX_FC | MAC_MCR_FORCE_TX_FC;
+
+	mtk_w32(mac->hw, mcr, MTK_MAC_MCR(mac->id));
+
+	if (mac->phy_dev->link)
+		netif_carrier_on(dev);
+	else
+		netif_carrier_off(dev);
+}
+
+static int mtk_phy_connect_node(struct mtk_eth *eth, struct mtk_mac *mac,
+				struct device_node *phy_node)
+{
+	const __be32 *_addr = NULL;
+	struct phy_device *phydev;
+	int phy_mode, addr;
+
+	_addr = of_get_property(phy_node, "reg", NULL);
+
+	if (!_addr || (be32_to_cpu(*_addr) >= 0x20)) {
+		pr_err("%s: invalid phy address\n", phy_node->name);
+		return -EINVAL;
+	}
+	addr = be32_to_cpu(*_addr);
+	phy_mode = of_get_phy_mode(phy_node);
+	if (phy_mode < 0) {
+		dev_err(eth->dev, "incorrect phy-mode %d\n", phy_mode);
+		return -EINVAL;
+	}
+
+	phydev = of_phy_connect(eth->netdev[mac->id], phy_node,
+				mtk_phy_link_adjust, 0, phy_mode);
+	if (IS_ERR(phydev)) {
+		dev_err(eth->dev, "could not connect to PHY\n");
+		return PTR_ERR(phydev);
+	}
+
+	dev_info(eth->dev,
+		 "connected mac %d to PHY at %s [uid=%08x, driver=%s]\n",
+		 mac->id, phydev_name(phydev), phydev->phy_id,
+		 phydev->drv->name);
+
+	mac->phy_dev = phydev;
+
+	return 0;
+}
+
+static int mtk_phy_connect(struct mtk_mac *mac)
+{
+	struct mtk_eth *eth = mac->hw;
+	struct device_node *np;
+	u32 val, ge_mode;
+
+	np = of_parse_phandle(mac->of_node, "phy-handle", 0);
+	if (!np)
+		return -ENODEV;
+
+	switch (of_get_phy_mode(np)) {
+	case PHY_INTERFACE_MODE_RGMII:
+		ge_mode = 0;
+		break;
+	case PHY_INTERFACE_MODE_MII:
+		ge_mode = 1;
+		break;
+	case PHY_INTERFACE_MODE_RMII:
+		ge_mode = 2;
+		break;
+	default:
+		dev_err(eth->dev, "invalid phy_mode\n");
+		return -1;
+	}
+
+	/* put the gmac into the right mode */
+	regmap_read(eth->ethsys, ETHSYS_SYSCFG0, &val);
+	val &= ~SYSCFG0_GE_MODE(SYSCFG0_GE_MASK, mac->id);
+	val |= SYSCFG0_GE_MODE(ge_mode, mac->id);
+	regmap_write(eth->ethsys, ETHSYS_SYSCFG0, val);
+
+	mtk_phy_connect_node(eth, mac, np);
+	mac->phy_dev->autoneg = AUTONEG_ENABLE;
+	mac->phy_dev->speed = 0;
+	mac->phy_dev->duplex = 0;
+	mac->phy_dev->supported &= PHY_BASIC_FEATURES;
+	mac->phy_dev->advertising = mac->phy_dev->supported |
+				    ADVERTISED_Autoneg;
+	phy_start_aneg(mac->phy_dev);
+
+	return 0;
+}
+
+static int mtk_mdio_init(struct mtk_eth *eth)
+{
+	struct device_node *mii_np;
+	int err;
+
+	mii_np = of_get_child_by_name(eth->dev->of_node, "mdio-bus");
+	if (!mii_np) {
+		dev_err(eth->dev, "no %s child node found", "mdio-bus");
+		return -ENODEV;
+	}
+
+	if (!of_device_is_available(mii_np)) {
+		err = 0;
+		goto err_put_node;
+	}
+
+	eth->mii_bus = mdiobus_alloc();
+	if (!eth->mii_bus) {
+		err = -ENOMEM;
+		goto err_put_node;
+	}
+
+	eth->mii_bus->name = "mdio";
+	eth->mii_bus->read = mtk_mdio_read;
+	eth->mii_bus->write = mtk_mdio_write;
+	eth->mii_bus->priv = eth;
+	eth->mii_bus->parent = eth->dev;
+
+	snprintf(eth->mii_bus->id, MII_BUS_ID_SIZE, "%s", mii_np->name);
+	err = of_mdiobus_register(eth->mii_bus, mii_np);
+	if (err)
+		goto err_free_bus;
+
+	return 0;
+
+err_free_bus:
+	kfree(eth->mii_bus);
+
+err_put_node:
+	of_node_put(mii_np);
+	eth->mii_bus = NULL;
+	return err;
+}
+
+static void mtk_mdio_cleanup(struct mtk_eth *eth)
+{
+	if (!eth->mii_bus)
+		return;
+
+	mdiobus_unregister(eth->mii_bus);
+	of_node_put(eth->mii_bus->dev.of_node);
+	kfree(eth->mii_bus);
+}
+
+static inline void mtk_irq_disable(struct mtk_eth *eth, u32 mask)
+{
+	u32 val;
+
+	val = mtk_r32(eth, MTK_QDMA_INT_MASK);
+	mtk_w32(eth, val & ~mask, MTK_QDMA_INT_MASK);
+	/* flush write */
+	mtk_r32(eth, MTK_QDMA_INT_MASK);
+}
+
+static inline void mtk_irq_enable(struct mtk_eth *eth, u32 mask)
+{
+	u32 val;
+
+	val = mtk_r32(eth, MTK_QDMA_INT_MASK);
+	mtk_w32(eth, val | mask, MTK_QDMA_INT_MASK);
+	/* flush write */
+	mtk_r32(eth, MTK_QDMA_INT_MASK);
+}
+
+static int mtk_set_mac_address(struct net_device *dev, void *p)
+{
+	int ret = eth_mac_addr(dev, p);
+	struct mtk_mac *mac = netdev_priv(dev);
+	const char *macaddr = dev->dev_addr;
+	unsigned long flags;
+
+	if (ret)
+		return ret;
+
+	spin_lock_irqsave(&mac->hw->page_lock, flags);
+	mtk_w32(mac->hw, (macaddr[0] << 8) | macaddr[1],
+		MTK_GDMA_MAC_ADRH(mac->id));
+	mtk_w32(mac->hw, (macaddr[2] << 24) | (macaddr[3] << 16) |
+		(macaddr[4] << 8) | macaddr[5],
+		MTK_GDMA_MAC_ADRL(mac->id));
+	spin_unlock_irqrestore(&mac->hw->page_lock, flags);
+
+	return 0;
+}
+
+void mtk_stats_update_mac(struct mtk_mac *mac)
+{
+	struct mtk_hw_stats *hw_stats = mac->hw_stats;
+	unsigned int base = MTK_GDM1_TX_GBCNT;
+	u64 stats;
+
+	base += hw_stats->reg_offset;
+
+	u64_stats_update_begin(&hw_stats->syncp);
+
+	hw_stats->rx_bytes += mtk_r32(mac->hw, base);
+	stats =  mtk_r32(mac->hw, base + 0x04);
+	if (stats)
+		hw_stats->rx_bytes += (stats << 32);
+	hw_stats->rx_packets += mtk_r32(mac->hw, base + 0x08);
+	hw_stats->rx_overflow += mtk_r32(mac->hw, base + 0x10);
+	hw_stats->rx_fcs_errors += mtk_r32(mac->hw, base + 0x14);
+	hw_stats->rx_short_errors += mtk_r32(mac->hw, base + 0x18);
+	hw_stats->rx_long_errors += mtk_r32(mac->hw, base + 0x1c);
+	hw_stats->rx_checksum_errors += mtk_r32(mac->hw, base + 0x20);
+	hw_stats->rx_flow_control_packets +=
+					mtk_r32(mac->hw, base + 0x24);
+	hw_stats->tx_skip += mtk_r32(mac->hw, base + 0x28);
+	hw_stats->tx_collisions += mtk_r32(mac->hw, base + 0x2c);
+	hw_stats->tx_bytes += mtk_r32(mac->hw, base + 0x30);
+	stats =  mtk_r32(mac->hw, base + 0x34);
+	if (stats)
+		hw_stats->tx_bytes += (stats << 32);
+	hw_stats->tx_packets += mtk_r32(mac->hw, base + 0x38);
+	u64_stats_update_end(&hw_stats->syncp);
+}
+
+static void mtk_stats_update(struct mtk_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < MTK_MAC_COUNT; i++) {
+		if (!eth->mac[i] || !eth->mac[i]->hw_stats)
+			continue;
+		if (spin_trylock(&eth->mac[i]->hw_stats->stats_lock)) {
+			mtk_stats_update_mac(eth->mac[i]);
+			spin_unlock(&eth->mac[i]->hw_stats->stats_lock);
+		}
+	}
+}
+
+static struct rtnl_link_stats64 *mtk_get_stats64(struct net_device *dev,
+					struct rtnl_link_stats64 *storage)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_hw_stats *hw_stats = mac->hw_stats;
+	unsigned int start;
+
+	if (netif_running(dev) && netif_device_present(dev)) {
+		if (spin_trylock(&hw_stats->stats_lock)) {
+			mtk_stats_update_mac(mac);
+			spin_unlock(&hw_stats->stats_lock);
+		}
+	}
+
+	do {
+		start = u64_stats_fetch_begin_irq(&hw_stats->syncp);
+		storage->rx_packets = hw_stats->rx_packets;
+		storage->tx_packets = hw_stats->tx_packets;
+		storage->rx_bytes = hw_stats->rx_bytes;
+		storage->tx_bytes = hw_stats->tx_bytes;
+		storage->collisions = hw_stats->tx_collisions;
+		storage->rx_length_errors = hw_stats->rx_short_errors +
+			hw_stats->rx_long_errors;
+		storage->rx_over_errors = hw_stats->rx_overflow;
+		storage->rx_crc_errors = hw_stats->rx_fcs_errors;
+		storage->rx_errors = hw_stats->rx_checksum_errors;
+		storage->tx_aborted_errors = hw_stats->tx_skip;
+	} while (u64_stats_fetch_retry_irq(&hw_stats->syncp, start));
+
+	storage->tx_errors = dev->stats.tx_errors;
+	storage->rx_dropped = dev->stats.rx_dropped;
+	storage->tx_dropped = dev->stats.tx_dropped;
+
+	return storage;
+}
+
+static inline int mtk_max_frag_size(int mtu)
+{
+	/* make sure buf_size will be at least MTK_MAX_RX_LENGTH */
+	if (mtu + MTK_RX_ETH_HLEN < MTK_MAX_RX_LENGTH)
+		mtu = MTK_MAX_RX_LENGTH - MTK_RX_ETH_HLEN;
+
+	return SKB_DATA_ALIGN(MTK_RX_HLEN + mtu) +
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+}
+
+static inline int mtk_max_buf_size(int frag_size)
+{
+	int buf_size = frag_size - NET_SKB_PAD - NET_IP_ALIGN -
+		       SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+
+	WARN_ON(buf_size < MTK_MAX_RX_LENGTH);
+
+	return buf_size;
+}
+
+static inline void mtk_rx_get_desc(struct mtk_rx_dma *rxd,
+				   struct mtk_rx_dma *dma_rxd)
+{
+	rxd->rxd1 = READ_ONCE(dma_rxd->rxd1);
+	rxd->rxd2 = READ_ONCE(dma_rxd->rxd2);
+	rxd->rxd3 = READ_ONCE(dma_rxd->rxd3);
+	rxd->rxd4 = READ_ONCE(dma_rxd->rxd4);
+}
+
+/* the qdma core needs scratch memory to be setup */
+static int mtk_init_fq_dma(struct mtk_eth *eth)
+{
+	unsigned int phy_ring_head, phy_ring_tail;
+	int cnt = MTK_DMA_SIZE;
+	dma_addr_t dma_addr;
+	int i;
+
+	eth->scratch_ring = dma_alloc_coherent(eth->dev,
+					       cnt * sizeof(struct mtk_tx_dma),
+					       &phy_ring_head,
+					       GFP_ATOMIC | __GFP_ZERO);
+	if (unlikely(!eth->scratch_ring))
+		return -ENOMEM;
+
+	eth->scratch_head = kcalloc(cnt, MTK_QDMA_PAGE_SIZE,
+				    GFP_KERNEL);
+	dma_addr = dma_map_single(eth->dev,
+				  eth->scratch_head, cnt * MTK_QDMA_PAGE_SIZE,
+				  DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(eth->dev, dma_addr)))
+		return -ENOMEM;
+
+	memset(eth->scratch_ring, 0x0, sizeof(struct mtk_tx_dma) * cnt);
+	phy_ring_tail = phy_ring_head +
+			(sizeof(struct mtk_tx_dma) * (cnt - 1));
+
+	for (i = 0; i < cnt; i++) {
+		eth->scratch_ring[i].txd1 =
+					(dma_addr + (i * MTK_QDMA_PAGE_SIZE));
+		if (i < cnt - 1)
+			eth->scratch_ring[i].txd2 = (phy_ring_head +
+				((i + 1) * sizeof(struct mtk_tx_dma)));
+		eth->scratch_ring[i].txd3 = TX_DMA_SDL(MTK_QDMA_PAGE_SIZE);
+	}
+
+	mtk_w32(eth, phy_ring_head, MTK_QDMA_FQ_HEAD);
+	mtk_w32(eth, phy_ring_tail, MTK_QDMA_FQ_TAIL);
+	mtk_w32(eth, (cnt << 16) | cnt, MTK_QDMA_FQ_CNT);
+	mtk_w32(eth, MTK_QDMA_PAGE_SIZE << 16, MTK_QDMA_FQ_BLEN);
+
+	return 0;
+}
+
+static inline void *mtk_qdma_phys_to_virt(struct mtk_tx_ring *ring, u32 desc)
+{
+	void *ret = ring->dma;
+
+	return ret + (desc - ring->phys);
+}
+
+static inline struct mtk_tx_buf *mtk_desc_to_tx_buf(struct mtk_tx_ring *ring,
+						    struct mtk_tx_dma *txd)
+{
+	int idx = txd - ring->dma;
+
+	return &ring->buf[idx];
+}
+
+static void mtk_tx_unmap(struct device *dev, struct mtk_tx_buf *tx_buf)
+{
+	if (tx_buf->flags & MTK_TX_FLAGS_SINGLE0) {
+		dma_unmap_single(dev,
+				 dma_unmap_addr(tx_buf, dma_addr0),
+				 dma_unmap_len(tx_buf, dma_len0),
+				 DMA_TO_DEVICE);
+	} else if (tx_buf->flags & MTK_TX_FLAGS_PAGE0) {
+		dma_unmap_page(dev,
+			       dma_unmap_addr(tx_buf, dma_addr0),
+			       dma_unmap_len(tx_buf, dma_len0),
+			       DMA_TO_DEVICE);
+	}
+	tx_buf->flags = 0;
+	if (tx_buf->skb &&
+	    (tx_buf->skb != (struct sk_buff *)MTK_DMA_DUMMY_DESC))
+		dev_kfree_skb_any(tx_buf->skb);
+	tx_buf->skb = NULL;
+}
+
+static int mtk_tx_map(struct sk_buff *skb, struct net_device *dev,
+		      int tx_num, struct mtk_tx_ring *ring, bool gso)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+	struct mtk_tx_dma *itxd, *txd;
+	struct mtk_tx_buf *tx_buf;
+	unsigned long flags;
+	dma_addr_t mapped_addr;
+	unsigned int nr_frags;
+	int i, n_desc = 1;
+	u32 txd4 = 0;
+
+	itxd = ring->next_free;
+	if (itxd == ring->last_free)
+		return -ENOMEM;
+
+	/* set the forward port */
+	txd4 |= (mac->id + 1) << TX_DMA_FPORT_SHIFT;
+
+	tx_buf = mtk_desc_to_tx_buf(ring, itxd);
+	memset(tx_buf, 0, sizeof(*tx_buf));
+
+	if (gso)
+		txd4 |= TX_DMA_TSO;
+
+	/* TX Checksum offload */
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		txd4 |= TX_DMA_CHKSUM;
+
+	/* VLAN header offload */
+	if (skb_vlan_tag_present(skb))
+		txd4 |= TX_DMA_INS_VLAN | skb_vlan_tag_get(skb);
+
+	mapped_addr = dma_map_single(&dev->dev, skb->data,
+				     skb_headlen(skb), DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(&dev->dev, mapped_addr)))
+		return -ENOMEM;
+
+	/* normally we can rely on the stack not calling this more than once,
+	 * however we have 2 queues running ont he same ring so we need to lock
+	 * the ring access
+	 */
+	spin_lock_irqsave(&eth->page_lock, flags);
+	WRITE_ONCE(itxd->txd1, mapped_addr);
+	tx_buf->flags |= MTK_TX_FLAGS_SINGLE0;
+	dma_unmap_addr_set(tx_buf, dma_addr0, mapped_addr);
+	dma_unmap_len_set(tx_buf, dma_len0, skb_headlen(skb));
+
+	/* TX SG offload */
+	txd = itxd;
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	for (i = 0; i < nr_frags; i++) {
+		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
+		unsigned int offset = 0;
+		int frag_size = skb_frag_size(frag);
+
+		while (frag_size) {
+			bool last_frag = false;
+			unsigned int frag_map_size;
+
+			txd = mtk_qdma_phys_to_virt(ring, txd->txd2);
+			if (txd == ring->last_free)
+				goto err_dma;
+
+			n_desc++;
+			frag_map_size = min(frag_size, MTK_TX_DMA_BUF_LEN);
+			mapped_addr = skb_frag_dma_map(&dev->dev, frag, offset,
+						       frag_map_size,
+						       DMA_TO_DEVICE);
+			if (unlikely(dma_mapping_error(&dev->dev, mapped_addr)))
+				goto err_dma;
+
+			if (i == nr_frags - 1 &&
+			    (frag_size - frag_map_size) == 0)
+				last_frag = true;
+
+			WRITE_ONCE(txd->txd1, mapped_addr);
+			WRITE_ONCE(txd->txd3, (TX_DMA_SWC |
+					       TX_DMA_PLEN0(frag_map_size) |
+					       last_frag * TX_DMA_LS0) |
+					       mac->id);
+			WRITE_ONCE(txd->txd4, 0);
+
+			tx_buf->skb = (struct sk_buff *)MTK_DMA_DUMMY_DESC;
+			tx_buf = mtk_desc_to_tx_buf(ring, txd);
+			memset(tx_buf, 0, sizeof(*tx_buf));
+
+			tx_buf->flags |= MTK_TX_FLAGS_PAGE0;
+			dma_unmap_addr_set(tx_buf, dma_addr0, mapped_addr);
+			dma_unmap_len_set(tx_buf, dma_len0, frag_map_size);
+			frag_size -= frag_map_size;
+			offset += frag_map_size;
+		}
+	}
+
+	/* store skb to cleanup */
+	tx_buf->skb = skb;
+
+	WRITE_ONCE(itxd->txd4, txd4);
+	WRITE_ONCE(itxd->txd3, (TX_DMA_SWC | TX_DMA_PLEN0(skb_headlen(skb)) |
+				(!nr_frags * TX_DMA_LS0)));
+
+	spin_unlock_irqrestore(&eth->page_lock, flags);
+
+	netdev_sent_queue(dev, skb->len);
+	skb_tx_timestamp(skb);
+
+	ring->next_free = mtk_qdma_phys_to_virt(ring, txd->txd2);
+	atomic_sub(n_desc, &ring->free_count);
+
+	/* make sure that all changes to the dma ring are flushed before we
+	 * continue
+	 */
+	wmb();
+
+	if (netif_xmit_stopped(netdev_get_tx_queue(dev, 0)) || !skb->xmit_more)
+		mtk_w32(eth, txd->txd2, MTK_QTX_CTX_PTR);
+
+	return 0;
+
+err_dma:
+	do {
+		tx_buf = mtk_desc_to_tx_buf(ring, txd);
+
+		/* unmap dma */
+		mtk_tx_unmap(&dev->dev, tx_buf);
+
+		itxd->txd3 = TX_DMA_LS0 | TX_DMA_OWNER_CPU;
+		itxd = mtk_qdma_phys_to_virt(ring, itxd->txd2);
+	} while (itxd != txd);
+
+	return -ENOMEM;
+}
+
+static inline int mtk_cal_txd_req(struct sk_buff *skb)
+{
+	int i, nfrags;
+	struct skb_frag_struct *frag;
+
+	nfrags = 1;
+	if (skb_is_gso(skb)) {
+		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+			frag = &skb_shinfo(skb)->frags[i];
+			nfrags += DIV_ROUND_UP(frag->size, MTK_TX_DMA_BUF_LEN);
+		}
+	} else {
+		nfrags += skb_shinfo(skb)->nr_frags;
+	}
+
+	return DIV_ROUND_UP(nfrags, 2);
+}
+
+static int mtk_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+	struct mtk_tx_ring *ring = &eth->tx_ring;
+	struct net_device_stats *stats = &dev->stats;
+	bool gso = false;
+	int tx_num;
+
+	tx_num = mtk_cal_txd_req(skb);
+	if (unlikely(atomic_read(&ring->free_count) <= tx_num)) {
+		netif_stop_queue(dev);
+		netif_err(eth, tx_queued, dev,
+			  "Tx Ring full when queue awake!\n");
+		return NETDEV_TX_BUSY;
+	}
+
+	/* TSO: fill MSS info in tcp checksum field */
+	if (skb_is_gso(skb)) {
+		if (skb_cow_head(skb, 0)) {
+			netif_warn(eth, tx_err, dev,
+				   "GSO expand head fail.\n");
+			goto drop;
+		}
+
+		if (skb_shinfo(skb)->gso_type &
+				(SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) {
+			gso = true;
+			tcp_hdr(skb)->check = htons(skb_shinfo(skb)->gso_size);
+		}
+	}
+
+	if (mtk_tx_map(skb, dev, tx_num, ring, gso) < 0)
+		goto drop;
+
+	if (unlikely(atomic_read(&ring->free_count) <= ring->thresh)) {
+		netif_stop_queue(dev);
+		if (unlikely(atomic_read(&ring->free_count) >
+			     ring->thresh))
+			netif_wake_queue(dev);
+	}
+
+	return NETDEV_TX_OK;
+
+drop:
+	stats->tx_dropped++;
+	dev_kfree_skb(skb);
+	return NETDEV_TX_OK;
+}
+
+static int mtk_poll_rx(struct napi_struct *napi, int budget,
+		       struct mtk_eth *eth, u32 rx_intr)
+{
+	struct mtk_rx_ring *ring = &eth->rx_ring;
+	int idx = ring->calc_idx;
+	struct sk_buff *skb;
+	u8 *data, *new_data;
+	struct mtk_rx_dma *rxd, trxd;
+	int done = 0;
+
+	while (done < budget) {
+		struct net_device *netdev;
+		unsigned int pktlen;
+		dma_addr_t dma_addr;
+		int mac = 0;
+
+		idx = NEXT_RX_DESP_IDX(idx);
+		rxd = &ring->dma[idx];
+		data = ring->data[idx];
+
+		mtk_rx_get_desc(&trxd, rxd);
+		if (!(trxd.rxd2 & RX_DMA_DONE))
+			break;
+
+		/* find out which mac the packet come from. values start at 1 */
+		mac = (trxd.rxd4 >> RX_DMA_FPORT_SHIFT) &
+		      RX_DMA_FPORT_MASK;
+		mac--;
+
+		netdev = eth->netdev[mac];
+
+		/* alloc new buffer */
+		new_data = napi_alloc_frag(ring->frag_size);
+		if (unlikely(!new_data)) {
+			netdev->stats.rx_dropped++;
+			goto release_desc;
+		}
+		dma_addr = dma_map_single(&eth->netdev[mac]->dev,
+					  new_data + NET_SKB_PAD,
+					  ring->buf_size,
+					  DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(&netdev->dev, dma_addr))) {
+			skb_free_frag(new_data);
+			goto release_desc;
+		}
+
+		/* receive data */
+		skb = build_skb(data, ring->frag_size);
+		if (unlikely(!skb)) {
+			put_page(virt_to_head_page(new_data));
+			goto release_desc;
+		}
+		skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN);
+
+		dma_unmap_single(&netdev->dev, trxd.rxd1,
+				 ring->buf_size, DMA_FROM_DEVICE);
+		pktlen = RX_DMA_GET_PLEN0(trxd.rxd2);
+		skb->dev = netdev;
+		skb_put(skb, pktlen);
+		if (trxd.rxd4 & RX_DMA_L4_VALID)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+		else
+			skb_checksum_none_assert(skb);
+		skb->protocol = eth_type_trans(skb, netdev);
+
+		if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX &&
+		    RX_DMA_VID(trxd.rxd3))
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+					       RX_DMA_VID(trxd.rxd3));
+		napi_gro_receive(napi, skb);
+
+		ring->data[idx] = new_data;
+		rxd->rxd1 = (unsigned int)dma_addr;
+
+release_desc:
+		rxd->rxd2 = RX_DMA_PLEN0(ring->buf_size);
+
+		ring->calc_idx = idx;
+		/* make sure that all changes to the dma ring are flushed before
+		 * we continue
+		 */
+		wmb();
+		mtk_w32(eth, ring->calc_idx, MTK_QRX_CRX_IDX0);
+		done++;
+	}
+
+	if (done < budget)
+		mtk_w32(eth, rx_intr, MTK_QMTK_INT_STATUS);
+
+	return done;
+}
+
+static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again)
+{
+	struct mtk_tx_ring *ring = &eth->tx_ring;
+	struct mtk_tx_dma *desc;
+	struct sk_buff *skb;
+	struct mtk_tx_buf *tx_buf;
+	int total = 0, done[MTK_MAX_DEVS];
+	unsigned int bytes[MTK_MAX_DEVS];
+	u32 cpu, dma;
+	static int condition;
+	int i;
+
+	memset(done, 0, sizeof(done));
+	memset(bytes, 0, sizeof(bytes));
+
+	cpu = mtk_r32(eth, MTK_QTX_CRX_PTR);
+	dma = mtk_r32(eth, MTK_QTX_DRX_PTR);
+
+	desc = mtk_qdma_phys_to_virt(ring, cpu);
+
+	while ((cpu != dma) && budget) {
+		u32 next_cpu = desc->txd2;
+		int mac;
+
+		desc = mtk_qdma_phys_to_virt(ring, desc->txd2);
+		if ((desc->txd3 & TX_DMA_OWNER_CPU) == 0)
+			break;
+
+		mac = (desc->txd4 >> TX_DMA_FPORT_SHIFT) &
+		       TX_DMA_FPORT_MASK;
+		mac--;
+
+		tx_buf = mtk_desc_to_tx_buf(ring, desc);
+		skb = tx_buf->skb;
+		if (!skb) {
+			condition = 1;
+			break;
+		}
+
+		if (skb != (struct sk_buff *)MTK_DMA_DUMMY_DESC) {
+			bytes[mac] += skb->len;
+			done[mac]++;
+			budget--;
+		}
+		mtk_tx_unmap(eth->dev, tx_buf);
+
+		ring->last_free->txd2 = next_cpu;
+		ring->last_free = desc;
+		atomic_inc(&ring->free_count);
+
+		cpu = next_cpu;
+	}
+
+	mtk_w32(eth, cpu, MTK_QTX_CRX_PTR);
+
+	for (i = 0; i < MTK_MAC_COUNT; i++) {
+		if (!eth->netdev[i] || !done[i])
+			continue;
+		netdev_completed_queue(eth->netdev[i], done[i], bytes[i]);
+		total += done[i];
+	}
+
+	/* read hw index again make sure no new tx packet */
+	if (cpu != dma || cpu != mtk_r32(eth, MTK_QTX_DRX_PTR))
+		*tx_again = true;
+	else
+		mtk_w32(eth, MTK_TX_DONE_INT, MTK_QMTK_INT_STATUS);
+
+	if (!total)
+		return 0;
+
+	for (i = 0; i < MTK_MAC_COUNT; i++) {
+		if (!eth->netdev[i] ||
+		    unlikely(!netif_queue_stopped(eth->netdev[i])))
+			continue;
+		if (atomic_read(&ring->free_count) > ring->thresh)
+			netif_wake_queue(eth->netdev[i]);
+	}
+
+	return total;
+}
+
+static int mtk_poll(struct napi_struct *napi, int budget)
+{
+	struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi);
+	u32 status, status2, mask, tx_intr, rx_intr, status_intr;
+	int tx_done, rx_done;
+	bool tx_again = false;
+
+	status = mtk_r32(eth, MTK_QMTK_INT_STATUS);
+	status2 = mtk_r32(eth, MTK_INT_STATUS2);
+	tx_intr = MTK_TX_DONE_INT;
+	rx_intr = MTK_RX_DONE_INT;
+	status_intr = (MTK_GDM1_AF | MTK_GDM2_AF);
+	tx_done = 0;
+	rx_done = 0;
+	tx_again = 0;
+
+	if (status & tx_intr)
+		tx_done = mtk_poll_tx(eth, budget, &tx_again);
+
+	if (status & rx_intr)
+		rx_done = mtk_poll_rx(napi, budget, eth, rx_intr);
+
+	if (unlikely(status2 & status_intr)) {
+		mtk_stats_update(eth);
+		mtk_w32(eth, status_intr, MTK_INT_STATUS2);
+	}
+
+	if (unlikely(netif_msg_intr(eth))) {
+		mask = mtk_r32(eth, MTK_QDMA_INT_MASK);
+		netdev_info(eth->netdev[0],
+			    "done tx %d, rx %d, intr 0x%08x/0x%x\n",
+			    tx_done, rx_done, status, mask);
+	}
+
+	if (tx_again || rx_done == budget)
+		return budget;
+
+	status = mtk_r32(eth, MTK_QMTK_INT_STATUS);
+	if (status & (tx_intr | rx_intr))
+		return budget;
+
+	napi_complete(napi);
+	mtk_irq_enable(eth, tx_intr | rx_intr);
+
+	return rx_done;
+}
+
+static int mtk_tx_alloc(struct mtk_eth *eth)
+{
+	struct mtk_tx_ring *ring = &eth->tx_ring;
+	int i, sz = sizeof(*ring->dma);
+
+	ring->buf = kcalloc(MTK_DMA_SIZE, sizeof(*ring->buf),
+			       GFP_KERNEL);
+	if (!ring->buf)
+		goto no_tx_mem;
+
+	ring->dma = dma_alloc_coherent(eth->dev,
+					  MTK_DMA_SIZE * sz,
+					  &ring->phys,
+					  GFP_ATOMIC | __GFP_ZERO);
+	if (!ring->dma)
+		goto no_tx_mem;
+
+	memset(ring->dma, 0, MTK_DMA_SIZE * sz);
+	for (i = 0; i < MTK_DMA_SIZE; i++) {
+		int next = (i + 1) % MTK_DMA_SIZE;
+		u32 next_ptr = ring->phys + next * sz;
+
+		ring->dma[i].txd2 = next_ptr;
+		ring->dma[i].txd3 = TX_DMA_LS0 | TX_DMA_OWNER_CPU;
+	}
+
+	atomic_set(&ring->free_count, MTK_DMA_SIZE - 2);
+	ring->next_free = &ring->dma[0];
+	ring->last_free = &ring->dma[MTK_DMA_SIZE - 2];
+	ring->thresh = max((unsigned long)MTK_DMA_SIZE >> 2,
+			      MAX_SKB_FRAGS);
+
+	/* make sure that all changes to the dma ring are flushed before we
+	 * continue
+	 */
+	wmb();
+
+	mtk_w32(eth, ring->phys, MTK_QTX_CTX_PTR);
+	mtk_w32(eth, ring->phys, MTK_QTX_DTX_PTR);
+	mtk_w32(eth,
+		ring->phys + ((MTK_DMA_SIZE - 1) * sz),
+		MTK_QTX_CRX_PTR);
+	mtk_w32(eth,
+		ring->phys + ((MTK_DMA_SIZE - 1) * sz),
+		MTK_QTX_DRX_PTR);
+
+	return 0;
+
+no_tx_mem:
+	return -ENOMEM;
+}
+
+static void mtk_tx_clean(struct mtk_eth *eth)
+{
+	struct mtk_tx_ring *ring = &eth->tx_ring;
+	int i;
+
+	if (ring->buf) {
+		for (i = 0; i < MTK_DMA_SIZE; i++)
+			mtk_tx_unmap(eth->dev, &ring->buf[i]);
+		kfree(ring->buf);
+		ring->buf = NULL;
+	}
+
+	if (ring->dma) {
+		dma_free_coherent(eth->dev,
+				  MTK_DMA_SIZE * sizeof(*ring->dma),
+				  ring->dma,
+				  ring->phys);
+		ring->dma = NULL;
+	}
+}
+
+static int mtk_rx_alloc(struct mtk_eth *eth)
+{
+	struct mtk_rx_ring *ring = &eth->rx_ring;
+	int i;
+
+	ring->frag_size = mtk_max_frag_size(ETH_DATA_LEN);
+	ring->buf_size = mtk_max_buf_size(ring->frag_size);
+	ring->data = kcalloc(MTK_DMA_SIZE, sizeof(*ring->data),
+			     GFP_KERNEL);
+	if (!ring->data)
+		return -ENOMEM;
+
+	for (i = 0; i < MTK_DMA_SIZE; i++) {
+		ring->data[i] = netdev_alloc_frag(ring->frag_size);
+		if (!ring->data[i])
+			return -ENOMEM;
+	}
+
+	ring->dma = dma_alloc_coherent(eth->dev,
+				       MTK_DMA_SIZE * sizeof(*ring->dma),
+				       &ring->phys,
+				       GFP_ATOMIC | __GFP_ZERO);
+	if (!ring->dma)
+		return -ENOMEM;
+
+	for (i = 0; i < MTK_DMA_SIZE; i++) {
+		dma_addr_t dma_addr = dma_map_single(eth->dev,
+				ring->data[i] + NET_SKB_PAD,
+				ring->buf_size,
+				DMA_FROM_DEVICE);
+		if (unlikely(dma_mapping_error(eth->dev, dma_addr)))
+			return -ENOMEM;
+		ring->dma[i].rxd1 = (unsigned int)dma_addr;
+
+		ring->dma[i].rxd2 = RX_DMA_PLEN0(ring->buf_size);
+	}
+	ring->calc_idx = MTK_DMA_SIZE - 1;
+	/* make sure that all changes to the dma ring are flushed before we
+	 * continue
+	 */
+	wmb();
+
+	mtk_w32(eth, eth->rx_ring.phys, MTK_QRX_BASE_PTR0);
+	mtk_w32(eth, MTK_DMA_SIZE, MTK_QRX_MAX_CNT0);
+	mtk_w32(eth, eth->rx_ring.calc_idx, MTK_QRX_CRX_IDX0);
+	mtk_w32(eth, MTK_PST_DRX_IDX0, MTK_QDMA_RST_IDX);
+	mtk_w32(eth, (QDMA_RES_THRES << 8) | QDMA_RES_THRES, MTK_QTX_CFG(0));
+
+	return 0;
+}
+
+static void mtk_rx_clean(struct mtk_eth *eth)
+{
+	struct mtk_rx_ring *ring = &eth->rx_ring;
+	int i;
+
+	if (ring->data && ring->dma) {
+		for (i = 0; i < MTK_DMA_SIZE; i++) {
+			if (!ring->data[i])
+				continue;
+			if (!ring->dma[i].rxd1)
+				continue;
+			dma_unmap_single(eth->dev,
+					 ring->dma[i].rxd1,
+					 ring->buf_size,
+					 DMA_FROM_DEVICE);
+			skb_free_frag(ring->data[i]);
+		}
+		kfree(ring->data);
+		ring->data = NULL;
+	}
+
+	if (ring->dma) {
+		dma_free_coherent(eth->dev,
+				  MTK_DMA_SIZE * sizeof(*ring->dma),
+				  ring->dma,
+				  ring->phys);
+		ring->dma = NULL;
+	}
+}
+
+/* wait for DMA to finish whatever it is doing before we start using it again */
+static int mtk_dma_busy_wait(struct mtk_eth *eth)
+{
+	unsigned long t_start = jiffies;
+
+	while (1) {
+		if (!(mtk_r32(eth, MTK_QDMA_GLO_CFG) &
+		      (MTK_RX_DMA_BUSY | MTK_TX_DMA_BUSY)))
+			return 0;
+		if (time_after(jiffies, t_start + MTK_DMA_BUSY_TIMEOUT))
+			break;
+	}
+
+	dev_err(eth->dev, "DMA init timeout\n");
+	return -1;
+}
+
+static int mtk_dma_init(struct mtk_eth *eth)
+{
+	int err;
+
+	if (mtk_dma_busy_wait(eth))
+		return -EBUSY;
+
+	/* QDMA needs scratch memory for internal reordering of the
+	 * descriptors
+	 */
+	err = mtk_init_fq_dma(eth);
+	if (err)
+		return err;
+
+	err = mtk_tx_alloc(eth);
+	if (err)
+		return err;
+
+	err = mtk_rx_alloc(eth);
+	if (err)
+		return err;
+
+	/* Enable random early drop and set drop threshold automatically */
+	mtk_w32(eth, FC_THRES_DROP_MODE | FC_THRES_DROP_EN | FC_THRES_MIN,
+		MTK_QDMA_FC_THRES);
+	mtk_w32(eth, 0x0, MTK_QDMA_HRED2);
+
+	return 0;
+}
+
+static void mtk_dma_free(struct mtk_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < MTK_MAC_COUNT; i++)
+		if (eth->netdev[i])
+			netdev_reset_queue(eth->netdev[i]);
+	mtk_tx_clean(eth);
+	mtk_rx_clean(eth);
+	kfree(eth->scratch_head);
+}
+
+static void mtk_tx_timeout(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+
+	eth->netdev[mac->id]->stats.tx_errors++;
+	netif_err(eth, tx_err, dev,
+		  "transmit timed out\n");
+	schedule_work(&mac->pending_work);
+}
+
+static irqreturn_t mtk_handle_irq(int irq, void *_eth)
+{
+	struct mtk_eth *eth = _eth;
+	u32 status;
+
+	status = mtk_r32(eth, MTK_QMTK_INT_STATUS);
+	if (unlikely(!status))
+		return IRQ_NONE;
+
+	if (likely(status & (MTK_RX_DONE_INT | MTK_TX_DONE_INT))) {
+		if (likely(napi_schedule_prep(&eth->rx_napi)))
+			__napi_schedule(&eth->rx_napi);
+	} else {
+		mtk_w32(eth, status, MTK_QMTK_INT_STATUS);
+	}
+	mtk_irq_disable(eth, (MTK_RX_DONE_INT | MTK_TX_DONE_INT));
+
+	return IRQ_HANDLED;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void mtk_poll_controller(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+	u32 int_mask = MTK_TX_DONE_INT | MTK_RX_DONE_INT;
+
+	mtk_irq_disable(eth, int_mask);
+	mtk_handle_irq(dev->irq, dev);
+	mtk_irq_enable(eth, int_mask);
+}
+#endif
+
+static int mtk_start_dma(struct mtk_eth *eth)
+{
+	int err;
+
+	err = mtk_dma_init(eth);
+	if (err) {
+		mtk_dma_free(eth);
+		return err;
+	}
+
+	mtk_w32(eth,
+		MTK_TX_WB_DDONE | MTK_RX_DMA_EN | MTK_TX_DMA_EN |
+		MTK_RX_2B_OFFSET | MTK_DMA_SIZE_16DWORDS |
+		MTK_RX_BT_32DWORDS,
+		MTK_QDMA_GLO_CFG);
+
+	return 0;
+}
+
+static int mtk_open(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+
+	/* we run 2 netdevs on the same dma ring so we only bring it up once */
+	if (!atomic_read(&eth->dma_refcnt)) {
+		int err = mtk_start_dma(eth);
+
+		if (err)
+			return err;
+
+		napi_enable(&eth->rx_napi);
+		mtk_irq_enable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT);
+	}
+	atomic_inc(&eth->dma_refcnt);
+
+	phy_start(mac->phy_dev);
+	netif_start_queue(dev);
+
+	return 0;
+}
+
+static void mtk_stop_dma(struct mtk_eth *eth, u32 glo_cfg)
+{
+	unsigned long flags;
+	u32 val;
+	int i;
+
+	/* stop the dma engine */
+	spin_lock_irqsave(&eth->page_lock, flags);
+	val = mtk_r32(eth, glo_cfg);
+	mtk_w32(eth, val & ~(MTK_TX_WB_DDONE | MTK_RX_DMA_EN | MTK_TX_DMA_EN),
+		glo_cfg);
+	spin_unlock_irqrestore(&eth->page_lock, flags);
+
+	/* wait for dma stop */
+	for (i = 0; i < 10; i++) {
+		val = mtk_r32(eth, glo_cfg);
+		if (val & (MTK_TX_DMA_BUSY | MTK_RX_DMA_BUSY)) {
+			msleep(20);
+			continue;
+		}
+		break;
+	}
+}
+
+static int mtk_stop(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+
+	netif_tx_disable(dev);
+	phy_stop(mac->phy_dev);
+
+	/* only shutdown DMA if this is the last user */
+	if (!atomic_dec_and_test(&eth->dma_refcnt))
+		return 0;
+
+	mtk_irq_disable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT);
+	napi_disable(&eth->rx_napi);
+
+	mtk_stop_dma(eth, MTK_QDMA_GLO_CFG);
+
+	mtk_dma_free(eth);
+
+	return 0;
+}
+
+static int __init mtk_hw_init(struct mtk_eth *eth)
+{
+	int err, i;
+
+	/* reset the frame engine */
+	reset_control_assert(eth->rstc);
+	usleep_range(10, 20);
+	reset_control_deassert(eth->rstc);
+	usleep_range(10, 20);
+
+	/* Set GE2 driving and slew rate */
+	regmap_write(eth->pctl, GPIO_DRV_SEL10, 0xa00);
+
+	/* set GE2 TDSEL */
+	regmap_write(eth->pctl, GPIO_OD33_CTRL8, 0x5);
+
+	/* set GE2 TUNE */
+	regmap_write(eth->pctl, GPIO_BIAS_CTRL, 0x0);
+
+	/* GE1, Force 1000M/FD, FC ON */
+	mtk_w32(eth, MAC_MCR_FIXED_LINK, MTK_MAC_MCR(0));
+
+	/* GE2, Force 1000M/FD, FC ON */
+	mtk_w32(eth, MAC_MCR_FIXED_LINK, MTK_MAC_MCR(1));
+
+	/* Enable RX VLan Offloading */
+	mtk_w32(eth, 1, MTK_CDMP_EG_CTRL);
+
+	err = devm_request_irq(eth->dev, eth->irq, mtk_handle_irq, 0,
+			       dev_name(eth->dev), eth);
+	if (err)
+		return err;
+
+	err = mtk_mdio_init(eth);
+	if (err)
+		return err;
+
+	/* disable delay and normal interrupt */
+	mtk_w32(eth, 0, MTK_QDMA_DELAY_INT);
+	mtk_irq_disable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT);
+	mtk_w32(eth, RST_GL_PSE, MTK_RST_GL);
+	mtk_w32(eth, 0, MTK_RST_GL);
+
+	/* FE int grouping */
+	mtk_w32(eth, 0, MTK_FE_INT_GRP);
+
+	for (i = 0; i < 2; i++) {
+		u32 val = mtk_r32(eth, MTK_GDMA_FWD_CFG(i));
+
+		/* setup the forward port to send frame to QDMA */
+		val &= ~0xffff;
+		val |= 0x5555;
+
+		/* Enable RX checksum */
+		val |= MTK_GDMA_ICS_EN | MTK_GDMA_TCS_EN | MTK_GDMA_UCS_EN;
+
+		/* setup the mac dma */
+		mtk_w32(eth, val, MTK_GDMA_FWD_CFG(i));
+	}
+
+	return 0;
+}
+
+static int __init mtk_init(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+	const char *mac_addr;
+
+	mac_addr = of_get_mac_address(mac->of_node);
+	if (mac_addr)
+		ether_addr_copy(dev->dev_addr, mac_addr);
+
+	/* If the mac address is invalid, use random mac address  */
+	if (!is_valid_ether_addr(dev->dev_addr)) {
+		random_ether_addr(dev->dev_addr);
+		dev_err(eth->dev, "generated random MAC address %pM\n",
+			dev->dev_addr);
+		dev->addr_assign_type = NET_ADDR_RANDOM;
+	}
+
+	return mtk_phy_connect(mac);
+}
+
+static void mtk_uninit(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_eth *eth = mac->hw;
+
+	phy_disconnect(mac->phy_dev);
+	mtk_mdio_cleanup(eth);
+	mtk_irq_disable(eth, ~0);
+	free_irq(dev->irq, dev);
+}
+
+static int mtk_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	switch (cmd) {
+	case SIOCGMIIPHY:
+	case SIOCGMIIREG:
+	case SIOCSMIIREG:
+		return phy_mii_ioctl(mac->phy_dev, ifr, cmd);
+	default:
+		break;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static void mtk_pending_work(struct work_struct *work)
+{
+	struct mtk_mac *mac = container_of(work, struct mtk_mac, pending_work);
+	struct mtk_eth *eth = mac->hw;
+	struct net_device *dev = eth->netdev[mac->id];
+	int err;
+
+	rtnl_lock();
+	mtk_stop(dev);
+
+	err = mtk_open(dev);
+	if (err) {
+		netif_alert(eth, ifup, dev,
+			    "Driver up/down cycle failed, closing device.\n");
+		dev_close(dev);
+	}
+	rtnl_unlock();
+}
+
+static int mtk_cleanup(struct mtk_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < MTK_MAC_COUNT; i++) {
+		struct mtk_mac *mac = netdev_priv(eth->netdev[i]);
+
+		if (!eth->netdev[i])
+			continue;
+
+		unregister_netdev(eth->netdev[i]);
+		free_netdev(eth->netdev[i]);
+		cancel_work_sync(&mac->pending_work);
+	}
+
+	return 0;
+}
+
+static int mtk_get_settings(struct net_device *dev,
+			    struct ethtool_cmd *cmd)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	int err;
+
+	err = phy_read_status(mac->phy_dev);
+	if (err)
+		return -ENODEV;
+
+	return phy_ethtool_gset(mac->phy_dev, cmd);
+}
+
+static int mtk_set_settings(struct net_device *dev,
+			    struct ethtool_cmd *cmd)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	if (cmd->phy_address != mac->phy_dev->mdio.addr) {
+		mac->phy_dev = mdiobus_get_phy(mac->hw->mii_bus,
+					       cmd->phy_address);
+		if (!mac->phy_dev)
+			return -ENODEV;
+	}
+
+	return phy_ethtool_sset(mac->phy_dev, cmd);
+}
+
+static void mtk_get_drvinfo(struct net_device *dev,
+			    struct ethtool_drvinfo *info)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	strlcpy(info->driver, mac->hw->dev->driver->name, sizeof(info->driver));
+	strlcpy(info->bus_info, dev_name(mac->hw->dev), sizeof(info->bus_info));
+	info->n_stats = ARRAY_SIZE(mtk_ethtool_stats);
+}
+
+static u32 mtk_get_msglevel(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	return mac->hw->msg_enable;
+}
+
+static void mtk_set_msglevel(struct net_device *dev, u32 value)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	mac->hw->msg_enable = value;
+}
+
+static int mtk_nway_reset(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	return genphy_restart_aneg(mac->phy_dev);
+}
+
+static u32 mtk_get_link(struct net_device *dev)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	int err;
+
+	err = genphy_update_link(mac->phy_dev);
+	if (err)
+		return ethtool_op_get_link(dev);
+
+	return mac->phy_dev->link;
+}
+
+static void mtk_get_strings(struct net_device *dev, u32 stringset, u8 *data)
+{
+	int i;
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < ARRAY_SIZE(mtk_ethtool_stats); i++) {
+			memcpy(data, mtk_ethtool_stats[i].str, ETH_GSTRING_LEN);
+			data += ETH_GSTRING_LEN;
+		}
+		break;
+	}
+}
+
+static int mtk_get_sset_count(struct net_device *dev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return ARRAY_SIZE(mtk_ethtool_stats);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void mtk_get_ethtool_stats(struct net_device *dev,
+				  struct ethtool_stats *stats, u64 *data)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+	struct mtk_hw_stats *hwstats = mac->hw_stats;
+	u64 *data_src, *data_dst;
+	unsigned int start;
+	int i;
+
+	if (netif_running(dev) && netif_device_present(dev)) {
+		if (spin_trylock(&hwstats->stats_lock)) {
+			mtk_stats_update_mac(mac);
+			spin_unlock(&hwstats->stats_lock);
+		}
+	}
+
+	do {
+		data_src = (u64*)hwstats;
+		data_dst = data;
+		start = u64_stats_fetch_begin_irq(&hwstats->syncp);
+
+		for (i = 0; i < ARRAY_SIZE(mtk_ethtool_stats); i++)
+			*data_dst++ = *(data_src + mtk_ethtool_stats[i].offset);
+	} while (u64_stats_fetch_retry_irq(&hwstats->syncp, start));
+}
+
+static struct ethtool_ops mtk_ethtool_ops = {
+	.get_settings		= mtk_get_settings,
+	.set_settings		= mtk_set_settings,
+	.get_drvinfo		= mtk_get_drvinfo,
+	.get_msglevel		= mtk_get_msglevel,
+	.set_msglevel		= mtk_set_msglevel,
+	.nway_reset		= mtk_nway_reset,
+	.get_link		= mtk_get_link,
+	.get_strings		= mtk_get_strings,
+	.get_sset_count		= mtk_get_sset_count,
+	.get_ethtool_stats	= mtk_get_ethtool_stats,
+};
+
+static const struct net_device_ops mtk_netdev_ops = {
+	.ndo_init		= mtk_init,
+	.ndo_uninit		= mtk_uninit,
+	.ndo_open		= mtk_open,
+	.ndo_stop		= mtk_stop,
+	.ndo_start_xmit		= mtk_start_xmit,
+	.ndo_set_mac_address	= mtk_set_mac_address,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_do_ioctl		= mtk_do_ioctl,
+	.ndo_change_mtu		= eth_change_mtu,
+	.ndo_tx_timeout		= mtk_tx_timeout,
+	.ndo_get_stats64        = mtk_get_stats64,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= mtk_poll_controller,
+#endif
+};
+
+static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np)
+{
+	struct mtk_mac *mac;
+	const __be32 *_id = of_get_property(np, "reg", NULL);
+	int id, err;
+
+	if (!_id) {
+		dev_err(eth->dev, "missing mac id\n");
+		return -EINVAL;
+	}
+
+	id = be32_to_cpup(_id);
+	if (id >= MTK_MAC_COUNT) {
+		dev_err(eth->dev, "%d is not a valid mac id\n", id);
+		return -EINVAL;
+	}
+
+	if (eth->netdev[id]) {
+		dev_err(eth->dev, "duplicate mac id found: %d\n", id);
+		return -EINVAL;
+	}
+
+	eth->netdev[id] = alloc_etherdev(sizeof(*mac));
+	if (!eth->netdev[id]) {
+		dev_err(eth->dev, "alloc_etherdev failed\n");
+		return -ENOMEM;
+	}
+	mac = netdev_priv(eth->netdev[id]);
+	eth->mac[id] = mac;
+	mac->id = id;
+	mac->hw = eth;
+	mac->of_node = np;
+	INIT_WORK(&mac->pending_work, mtk_pending_work);
+
+	mac->hw_stats = devm_kzalloc(eth->dev,
+				     sizeof(*mac->hw_stats),
+				     GFP_KERNEL);
+	if (!mac->hw_stats) {
+		dev_err(eth->dev, "failed to allocate counter memory\n");
+		err = -ENOMEM;
+		goto free_netdev;
+	}
+	spin_lock_init(&mac->hw_stats->stats_lock);
+	mac->hw_stats->reg_offset = id * MTK_STAT_OFFSET;
+
+	SET_NETDEV_DEV(eth->netdev[id], eth->dev);
+	eth->netdev[id]->netdev_ops = &mtk_netdev_ops;
+	eth->netdev[id]->base_addr = (unsigned long)eth->base;
+	eth->netdev[id]->vlan_features = MTK_HW_FEATURES &
+		~(NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX);
+	eth->netdev[id]->features |= MTK_HW_FEATURES;
+	eth->netdev[id]->ethtool_ops = &mtk_ethtool_ops;
+
+	err = register_netdev(eth->netdev[id]);
+	if (err) {
+		dev_err(eth->dev, "error bringing up device\n");
+		goto free_netdev;
+	}
+	eth->netdev[id]->irq = eth->irq;
+	netif_info(eth, probe, eth->netdev[id],
+		   "mediatek frame engine at 0x%08lx, irq %d\n",
+		   eth->netdev[id]->base_addr, eth->netdev[id]->irq);
+
+	return 0;
+
+free_netdev:
+	free_netdev(eth->netdev[id]);
+	return err;
+}
+
+static int mtk_probe(struct platform_device *pdev)
+{
+	struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	struct device_node *mac_np;
+	const struct of_device_id *match;
+	struct mtk_soc_data *soc;
+	struct mtk_eth *eth;
+	int err;
+
+	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
+	pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask;
+
+	device_reset(&pdev->dev);
+
+	match = of_match_device(of_mtk_match, &pdev->dev);
+	soc = (struct mtk_soc_data *)match->data;
+
+	eth = devm_kzalloc(&pdev->dev, sizeof(*eth), GFP_KERNEL);
+	if (!eth)
+		return -ENOMEM;
+
+	eth->base = devm_ioremap_resource(&pdev->dev, res);
+	if (!eth->base)
+		return -EADDRNOTAVAIL;
+
+	spin_lock_init(&eth->page_lock);
+
+	eth->ethsys = syscon_regmap_lookup_by_phandle(pdev->dev.of_node,
+						      "mediatek,ethsys");
+	if (IS_ERR(eth->ethsys)) {
+		dev_err(&pdev->dev, "no ethsys regmap found\n");
+		return PTR_ERR(eth->ethsys);
+	}
+
+	eth->pctl = syscon_regmap_lookup_by_phandle(pdev->dev.of_node,
+						    "mediatek,pctl");
+	if (IS_ERR(eth->pctl)) {
+		dev_err(&pdev->dev, "no pctl regmap found\n");
+		return PTR_ERR(eth->pctl);
+	}
+
+	eth->rstc = devm_reset_control_get(&pdev->dev, "eth");
+	if (IS_ERR(eth->rstc)) {
+		dev_err(&pdev->dev, "no eth reset found\n");
+		return PTR_ERR(eth->rstc);
+	}
+
+	eth->irq = platform_get_irq(pdev, 0);
+	if (eth->irq < 0) {
+		dev_err(&pdev->dev, "no IRQ resource found\n");
+		return -ENXIO;
+	}
+
+	eth->clk_ethif = devm_clk_get(&pdev->dev, "ethif");
+	eth->clk_esw = devm_clk_get(&pdev->dev, "esw");
+	eth->clk_gp1 = devm_clk_get(&pdev->dev, "gp1");
+	eth->clk_gp2 = devm_clk_get(&pdev->dev, "gp2");
+	if (IS_ERR(eth->clk_esw) || IS_ERR(eth->clk_gp1) ||
+	    IS_ERR(eth->clk_gp2) || IS_ERR(eth->clk_ethif))
+		return -ENODEV;
+
+	clk_prepare_enable(eth->clk_ethif);
+	clk_prepare_enable(eth->clk_esw);
+	clk_prepare_enable(eth->clk_gp1);
+	clk_prepare_enable(eth->clk_gp2);
+
+	eth->dev = &pdev->dev;
+	eth->msg_enable = netif_msg_init(mtk_msg_level, MTK_DEFAULT_MSG_ENABLE);
+
+	err = mtk_hw_init(eth);
+	if (err)
+		return err;
+
+	for_each_child_of_node(pdev->dev.of_node, mac_np) {
+		if (!of_device_is_compatible(mac_np,
+					     "mediatek,eth-mac"))
+			continue;
+
+		if (!of_device_is_available(mac_np))
+			continue;
+
+		err = mtk_add_mac(eth, mac_np);
+		if (err)
+			goto err_free_dev;
+	}
+
+	/* we run 2 devices on the same DMA ring so we need a dummy device
+	 * for NAPI to work
+	 */
+	init_dummy_netdev(&eth->dummy_dev);
+	netif_napi_add(&eth->dummy_dev, &eth->rx_napi, mtk_poll,
+		       MTK_NAPI_WEIGHT);
+
+	platform_set_drvdata(pdev, eth);
+
+	return 0;
+
+err_free_dev:
+	mtk_cleanup(eth);
+	return err;
+}
+
+static int mtk_remove(struct platform_device *pdev)
+{
+	struct mtk_eth *eth = platform_get_drvdata(pdev);
+
+	clk_disable_unprepare(eth->clk_ethif);
+	clk_disable_unprepare(eth->clk_esw);
+	clk_disable_unprepare(eth->clk_gp1);
+	clk_disable_unprepare(eth->clk_gp2);
+
+	netif_napi_del(&eth->rx_napi);
+	mtk_cleanup(eth);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+const struct of_device_id of_mtk_match[] = {
+	{ .compatible = "mediatek,mt7623-eth" },
+	{},
+};
+
+static struct platform_driver mtk_driver = {
+	.probe = mtk_probe,
+	.remove = mtk_remove,
+	.driver = {
+		.name = "mtk_soc_eth",
+		.owner = THIS_MODULE,
+		.of_match_table = of_mtk_match,
+	},
+};
+
+module_platform_driver(mtk_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("John Crispin <blogic@openwrt.org>");
+MODULE_DESCRIPTION("Ethernet driver for MediaTek SoC");
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
new file mode 100644
index 0000000..48a5292
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -0,0 +1,421 @@
+/*   This program is free software; you can redistribute it and/or modify
+ *   it under the terms of the GNU General Public License as published by
+ *   the Free Software Foundation; version 2 of the License
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ *
+ *   Copyright (C) 2009-2016 John Crispin <blogic@openwrt.org>
+ *   Copyright (C) 2009-2016 Felix Fietkau <nbd@openwrt.org>
+ *   Copyright (C) 2013-2016 Michael Lee <igvtee@gmail.com>
+ */
+
+#ifndef MTK_ETH_H
+#define MTK_ETH_H
+
+#define MTK_QDMA_PAGE_SIZE	2048
+#define	MTK_MAX_RX_LENGTH	1536
+#define MTK_TX_DMA_BUF_LEN	0x3fff
+#define MTK_DMA_SIZE		256
+#define MTK_NAPI_WEIGHT		64
+#define MTK_MAC_COUNT		2
+#define MTK_RX_ETH_HLEN		(VLAN_ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN)
+#define MTK_RX_HLEN		(NET_SKB_PAD + MTK_RX_ETH_HLEN + NET_IP_ALIGN)
+#define MTK_DMA_DUMMY_DESC	0xffffffff
+#define MTK_DEFAULT_MSG_ENABLE	(NETIF_MSG_DRV | \
+				 NETIF_MSG_PROBE | \
+				 NETIF_MSG_LINK | \
+				 NETIF_MSG_TIMER | \
+				 NETIF_MSG_IFDOWN | \
+				 NETIF_MSG_IFUP | \
+				 NETIF_MSG_RX_ERR | \
+				 NETIF_MSG_TX_ERR)
+#define MTK_HW_FEATURES		(NETIF_F_IP_CSUM | \
+				 NETIF_F_RXCSUM | \
+				 NETIF_F_HW_VLAN_CTAG_TX | \
+				 NETIF_F_HW_VLAN_CTAG_RX | \
+				 NETIF_F_SG | NETIF_F_TSO | \
+				 NETIF_F_TSO6 | \
+				 NETIF_F_IPV6_CSUM)
+#define NEXT_RX_DESP_IDX(X)	(((X) + 1) & (MTK_DMA_SIZE - 1))
+
+/* Frame Engine Global Reset Register */
+#define MTK_RST_GL		0x04
+#define RST_GL_PSE		BIT(0)
+
+/* Frame Engine Interrupt Status Register */
+#define MTK_INT_STATUS2		0x08
+#define MTK_GDM1_AF		BIT(28)
+#define MTK_GDM2_AF		BIT(29)
+
+/* Frame Engine Interrupt Grouping Register */
+#define MTK_FE_INT_GRP		0x20
+
+/* CDMP Exgress Control Register */
+#define MTK_CDMP_EG_CTRL	0x404
+
+/* GDM Exgress Control Register */
+#define MTK_GDMA_FWD_CFG(x)	(0x500 + (x * 0x1000))
+#define MTK_GDMA_ICS_EN		BIT(22)
+#define MTK_GDMA_TCS_EN		BIT(21)
+#define MTK_GDMA_UCS_EN		BIT(20)
+
+/* Unicast Filter MAC Address Register - Low */
+#define MTK_GDMA_MAC_ADRL(x)	(0x508 + (x * 0x1000))
+
+/* Unicast Filter MAC Address Register - High */
+#define MTK_GDMA_MAC_ADRH(x)	(0x50C + (x * 0x1000))
+
+/* QDMA TX Queue Configuration Registers */
+#define MTK_QTX_CFG(x)		(0x1800 + (x * 0x10))
+#define QDMA_RES_THRES		4
+
+/* QDMA TX Queue Scheduler Registers */
+#define MTK_QTX_SCH(x)		(0x1804 + (x * 0x10))
+
+/* QDMA RX Base Pointer Register */
+#define MTK_QRX_BASE_PTR0	0x1900
+
+/* QDMA RX Maximum Count Register */
+#define MTK_QRX_MAX_CNT0	0x1904
+
+/* QDMA RX CPU Pointer Register */
+#define MTK_QRX_CRX_IDX0	0x1908
+
+/* QDMA RX DMA Pointer Register */
+#define MTK_QRX_DRX_IDX0	0x190C
+
+/* QDMA Global Configuration Register */
+#define MTK_QDMA_GLO_CFG	0x1A04
+#define MTK_RX_2B_OFFSET	BIT(31)
+#define MTK_RX_BT_32DWORDS	(3 << 11)
+#define MTK_TX_WB_DDONE		BIT(6)
+#define MTK_DMA_SIZE_16DWORDS	(2 << 4)
+#define MTK_RX_DMA_BUSY		BIT(3)
+#define MTK_TX_DMA_BUSY		BIT(1)
+#define MTK_RX_DMA_EN		BIT(2)
+#define MTK_TX_DMA_EN		BIT(0)
+#define MTK_DMA_BUSY_TIMEOUT	HZ
+
+/* QDMA Reset Index Register */
+#define MTK_QDMA_RST_IDX	0x1A08
+#define MTK_PST_DRX_IDX0	BIT(16)
+
+/* QDMA Delay Interrupt Register */
+#define MTK_QDMA_DELAY_INT	0x1A0C
+
+/* QDMA Flow Control Register */
+#define MTK_QDMA_FC_THRES	0x1A10
+#define FC_THRES_DROP_MODE	BIT(20)
+#define FC_THRES_DROP_EN	(7 << 16)
+#define FC_THRES_MIN		0x4444
+
+/* QDMA Interrupt Status Register */
+#define MTK_QMTK_INT_STATUS	0x1A18
+#define MTK_RX_DONE_INT1	BIT(17)
+#define MTK_RX_DONE_INT0	BIT(16)
+#define MTK_TX_DONE_INT3	BIT(3)
+#define MTK_TX_DONE_INT2	BIT(2)
+#define MTK_TX_DONE_INT1	BIT(1)
+#define MTK_TX_DONE_INT0	BIT(0)
+#define MTK_RX_DONE_INT		(MTK_RX_DONE_INT0 | MTK_RX_DONE_INT1)
+#define MTK_TX_DONE_INT		(MTK_TX_DONE_INT0 | MTK_TX_DONE_INT1 | \
+				 MTK_TX_DONE_INT2 | MTK_TX_DONE_INT3)
+
+/* QDMA Interrupt Status Register */
+#define MTK_QDMA_INT_MASK	0x1A1C
+
+/* QDMA Interrupt Mask Register */
+#define MTK_QDMA_HRED2		0x1A44
+
+/* QDMA TX Forward CPU Pointer Register */
+#define MTK_QTX_CTX_PTR		0x1B00
+
+/* QDMA TX Forward DMA Pointer Register */
+#define MTK_QTX_DTX_PTR		0x1B04
+
+/* QDMA TX Release CPU Pointer Register */
+#define MTK_QTX_CRX_PTR		0x1B10
+
+/* QDMA TX Release DMA Pointer Register */
+#define MTK_QTX_DRX_PTR		0x1B14
+
+/* QDMA FQ Head Pointer Register */
+#define MTK_QDMA_FQ_HEAD	0x1B20
+
+/* QDMA FQ Head Pointer Register */
+#define MTK_QDMA_FQ_TAIL	0x1B24
+
+/* QDMA FQ Free Page Counter Register */
+#define MTK_QDMA_FQ_CNT		0x1B28
+
+/* QDMA FQ Free Page Buffer Length Register */
+#define MTK_QDMA_FQ_BLEN	0x1B2C
+
+/* GMA1 Received Good Byte Count Register */
+#define MTK_GDM1_TX_GBCNT	0x2400
+#define MTK_STAT_OFFSET		0x40
+
+/* QDMA descriptor txd4 */
+#define TX_DMA_CHKSUM		(0x7 << 29)
+#define TX_DMA_TSO		BIT(28)
+#define TX_DMA_FPORT_SHIFT	25
+#define TX_DMA_FPORT_MASK	0x7
+#define TX_DMA_INS_VLAN		BIT(16)
+
+/* QDMA descriptor txd3 */
+#define TX_DMA_OWNER_CPU	BIT(31)
+#define TX_DMA_LS0		BIT(30)
+#define TX_DMA_PLEN0(_x)	(((_x) & MTK_TX_DMA_BUF_LEN) << 16)
+#define TX_DMA_SWC		BIT(14)
+#define TX_DMA_SDL(_x)		(((_x) & 0x3fff) << 16)
+
+/* QDMA descriptor rxd2 */
+#define RX_DMA_DONE		BIT(31)
+#define RX_DMA_PLEN0(_x)	(((_x) & 0x3fff) << 16)
+#define RX_DMA_GET_PLEN0(_x)	(((_x) >> 16) & 0x3fff)
+
+/* QDMA descriptor rxd3 */
+#define RX_DMA_VID(_x)		((_x) & 0xfff)
+
+/* QDMA descriptor rxd4 */
+#define RX_DMA_L4_VALID		BIT(24)
+#define RX_DMA_FPORT_SHIFT	19
+#define RX_DMA_FPORT_MASK	0x7
+
+/* PHY Indirect Access Control registers */
+#define MTK_PHY_IAC		0x10004
+#define PHY_IAC_ACCESS		BIT(31)
+#define PHY_IAC_READ		BIT(19)
+#define PHY_IAC_WRITE		BIT(18)
+#define PHY_IAC_START		BIT(16)
+#define PHY_IAC_ADDR_SHIFT	20
+#define PHY_IAC_REG_SHIFT	25
+#define PHY_IAC_TIMEOUT		HZ
+
+/* Mac control registers */
+#define MTK_MAC_MCR(x)		(0x10100 + (x * 0x100))
+#define MAC_MCR_MAX_RX_1536	BIT(24)
+#define MAC_MCR_IPG_CFG		(BIT(18) | BIT(16))
+#define MAC_MCR_FORCE_MODE	BIT(15)
+#define MAC_MCR_TX_EN		BIT(14)
+#define MAC_MCR_RX_EN		BIT(13)
+#define MAC_MCR_BACKOFF_EN	BIT(9)
+#define MAC_MCR_BACKPR_EN	BIT(8)
+#define MAC_MCR_FORCE_RX_FC	BIT(5)
+#define MAC_MCR_FORCE_TX_FC	BIT(4)
+#define MAC_MCR_SPEED_1000	BIT(3)
+#define MAC_MCR_SPEED_100	BIT(2)
+#define MAC_MCR_FORCE_DPX	BIT(1)
+#define MAC_MCR_FORCE_LINK	BIT(0)
+#define MAC_MCR_FIXED_LINK	(MAC_MCR_MAX_RX_1536 | MAC_MCR_IPG_CFG | \
+				 MAC_MCR_FORCE_MODE | MAC_MCR_TX_EN | \
+				 MAC_MCR_RX_EN | MAC_MCR_BACKOFF_EN | \
+				 MAC_MCR_BACKPR_EN | MAC_MCR_FORCE_RX_FC | \
+				 MAC_MCR_FORCE_TX_FC | MAC_MCR_SPEED_1000 | \
+				 MAC_MCR_FORCE_DPX | MAC_MCR_FORCE_LINK)
+
+/* GPIO port control registers for GMAC 2*/
+#define GPIO_OD33_CTRL8		0x4c0
+#define GPIO_BIAS_CTRL		0xed0
+#define GPIO_DRV_SEL10		0xf00
+
+/* ethernet subsystem config register */
+#define ETHSYS_SYSCFG0		0x14
+#define SYSCFG0_GE_MASK		0x3
+#define SYSCFG0_GE_MODE(x, y)	(x << (12 + (y * 2)))
+
+struct mtk_rx_dma {
+	unsigned int rxd1;
+	unsigned int rxd2;
+	unsigned int rxd3;
+	unsigned int rxd4;
+} __packed __aligned(4);
+
+struct mtk_tx_dma {
+	unsigned int txd1;
+	unsigned int txd2;
+	unsigned int txd3;
+	unsigned int txd4;
+} __packed __aligned(4);
+
+struct mtk_eth;
+struct mtk_mac;
+
+/* struct mtk_hw_stats - the structure that holds the traffic statistics.
+ * @stats_lock:		make sure that stats operations are atomic
+ * @reg_offset:		the status register offset of the SoC
+ * @syncp:		the refcount
+ *
+ * All of the supported SoCs have hardware counters for traffic statistics.
+ * Whenever the status IRQ triggers we can read the latest stats from these
+ * counters and store them in this struct.
+ */
+struct mtk_hw_stats {
+	u64 tx_bytes;
+	u64 tx_packets;
+	u64 tx_skip;
+	u64 tx_collisions;
+	u64 rx_bytes;
+	u64 rx_packets;
+	u64 rx_overflow;
+	u64 rx_fcs_errors;
+	u64 rx_short_errors;
+	u64 rx_long_errors;
+	u64 rx_checksum_errors;
+	u64 rx_flow_control_packets;
+
+	spinlock_t		stats_lock;
+	u32			reg_offset;
+	struct u64_stats_sync	syncp;
+};
+
+/* PDMA descriptor can point at 1-2 segments. This enum allows us to track how
+ * memory was allocated so that it can be freed properly
+ */
+enum mtk_tx_flags {
+	MTK_TX_FLAGS_SINGLE0	= 0x01,
+	MTK_TX_FLAGS_PAGE0	= 0x02,
+};
+
+/* struct mtk_tx_buf -	This struct holds the pointers to the memory pointed at
+ *			by the TX descriptor	s
+ * @skb:		The SKB pointer of the packet being sent
+ * @dma_addr0:		The base addr of the first segment
+ * @dma_len0:		The length of the first segment
+ * @dma_addr1:		The base addr of the second segment
+ * @dma_len1:		The length of the second segment
+ */
+struct mtk_tx_buf {
+	struct sk_buff *skb;
+	u32 flags;
+	DEFINE_DMA_UNMAP_ADDR(dma_addr0);
+	DEFINE_DMA_UNMAP_LEN(dma_len0);
+	DEFINE_DMA_UNMAP_ADDR(dma_addr1);
+	DEFINE_DMA_UNMAP_LEN(dma_len1);
+};
+
+/* struct mtk_tx_ring -	This struct holds info describing a TX ring
+ * @dma:		The descriptor ring
+ * @buf:		The memory pointed at by the ring
+ * @phys:		The physical addr of tx_buf
+ * @next_free:		Pointer to the next free descriptor
+ * @last_free:		Pointer to the last free descriptor
+ * @thresh:		The threshold of minimum amount of free descriptors
+ * @free_count:		QDMA uses a linked list. Track how many free descriptors
+ *			are present
+ */
+struct mtk_tx_ring {
+	struct mtk_tx_dma *dma;
+	struct mtk_tx_buf *buf;
+	dma_addr_t phys;
+	struct mtk_tx_dma *next_free;
+	struct mtk_tx_dma *last_free;
+	u16 thresh;
+	atomic_t free_count;
+};
+
+/* struct mtk_rx_ring -	This struct holds info describing a RX ring
+ * @dma:		The descriptor ring
+ * @data:		The memory pointed at by the ring
+ * @phys:		The physical addr of rx_buf
+ * @frag_size:		How big can each fragment be
+ * @buf_size:		The size of each packet buffer
+ * @calc_idx:		The current head of ring
+ */
+struct mtk_rx_ring {
+	struct mtk_rx_dma *dma;
+	u8 **data;
+	dma_addr_t phys;
+	u16 frag_size;
+	u16 buf_size;
+	u16 calc_idx;
+};
+
+/* currently no SoC has more than 2 macs */
+#define MTK_MAX_DEVS			2
+
+/* struct mtk_eth -	This is the main datasructure for holding the state
+ *			of the driver
+ * @dev:		The device pointer
+ * @base:		The mapped register i/o base
+ * @page_lock:		Make sure that register operations are atomic
+ * @dummy_dev:		we run 2 netdevs on 1 physical DMA ring and need a
+ *			dummy for NAPI to work
+ * @netdev:		The netdev instances
+ * @mac:		Each netdev is linked to a physical MAC
+ * @irq:		The IRQ that we are using
+ * @msg_enable:		Ethtool msg level
+ * @ethsys:		The register map pointing at the range used to setup
+ *			MII modes
+ * @pctl:		The register map pointing at the range used to setup
+ *			GMAC port drive/slew values
+ * @dma_refcnt:		track how many netdevs are using the DMA engine
+ * @tx_ring:		Pointer to the memore holding info about the TX ring
+ * @rx_ring:		Pointer to the memore holding info about the RX ring
+ * @rx_napi:		The NAPI struct
+ * @scratch_ring:	Newer SoCs need memory for a second HW managed TX ring
+ * @scratch_head:	The scratch memory that scratch_ring points to.
+ * @clk_ethif:		The ethif clock
+ * @clk_esw:		The switch clock
+ * @clk_gp1:		The gmac1 clock
+ * @clk_gp2:		The gmac2 clock
+ * @mii_bus:		If there is a bus we need to create an instance for it
+ */
+
+struct mtk_eth {
+	struct device			*dev;
+	void __iomem			*base;
+	struct reset_control		*rstc;
+	spinlock_t			page_lock;
+	struct net_device		dummy_dev;
+	struct net_device		*netdev[MTK_MAX_DEVS];
+	struct mtk_mac			*mac[MTK_MAX_DEVS];
+	int				irq;
+	u32				msg_enable;
+	unsigned long			sysclk;
+	struct regmap			*ethsys;
+	struct regmap			*pctl;
+	atomic_t			dma_refcnt;
+	struct mtk_tx_ring		tx_ring;
+	struct mtk_rx_ring		rx_ring;
+	struct napi_struct		rx_napi;
+	struct mtk_tx_dma		*scratch_ring;
+	void				*scratch_head;
+	struct clk			*clk_ethif;
+	struct clk			*clk_esw;
+	struct clk			*clk_gp1;
+	struct clk			*clk_gp2;
+	struct mii_bus			*mii_bus;
+};
+
+/* struct mtk_mac -	the structure that holds the info about the MACs of the
+ *			SoC
+ * @id:			The number of the MAC
+ * @of_node:		Our devicetree node
+ * @hw:			Backpointer to our main datastruture
+ * @hw_stats:		Packet statistics counter
+ * @phy_dev:		The attached PHY if available
+ * @pending_work:	The workqueue used to reset the dma ring
+ */
+struct mtk_mac {
+	int				id;
+	struct device_node		*of_node;
+	struct mtk_eth			*hw;
+	struct mtk_hw_stats		*hw_stats;
+	struct phy_device		*phy_dev;
+	struct work_struct		pending_work;
+};
+
+/* the struct describing the SoC. these are declared in the soc_xyz.c files */
+extern const struct of_device_id of_mtk_match[];
+
+/* read the hardware status register */
+void mtk_stats_update_mac(struct mtk_mac *mac);
+
+void mtk_w32(struct mtk_eth *eth, u32 val, unsigned reg);
+u32 mtk_r32(struct mtk_eth *eth, unsigned reg);
+
+#endif /* MTK_ETH_H */
-- 
cgit v0.10.2


From 009fb0978ca4b5362cf875869c42893cc43b71a1 Mon Sep 17 00:00:00 2001
From: John Crispin <blogic@openwrt.org>
Date: Tue, 8 Mar 2016 11:29:56 +0100
Subject: net-next: mediatek: add Kconfig and Makefile

This patch adds the Makefile and Kconfig required to make the driver build.

Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 0b13af8..be67a19 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -106,6 +106,7 @@ config LANTIQ_ETOP
 	  Support for the MII0 inside the Lantiq SoC
 
 source "drivers/net/ethernet/marvell/Kconfig"
+source "drivers/net/ethernet/mediatek/Kconfig"
 source "drivers/net/ethernet/mellanox/Kconfig"
 source "drivers/net/ethernet/micrel/Kconfig"
 source "drivers/net/ethernet/microchip/Kconfig"
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 38dc1a7..6ffcc80 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -46,6 +46,7 @@ obj-$(CONFIG_JME) += jme.o
 obj-$(CONFIG_KORINA) += korina.o
 obj-$(CONFIG_LANTIQ_ETOP) += lantiq_etop.o
 obj-$(CONFIG_NET_VENDOR_MARVELL) += marvell/
+obj-$(CONFIG_NET_VENDOR_MEDIATEK) += mediatek/
 obj-$(CONFIG_NET_VENDOR_MELLANOX) += mellanox/
 obj-$(CONFIG_NET_VENDOR_MICREL) += micrel/
 obj-$(CONFIG_NET_VENDOR_MICROCHIP) += microchip/
diff --git a/drivers/net/ethernet/mediatek/Kconfig b/drivers/net/ethernet/mediatek/Kconfig
new file mode 100644
index 0000000..698bb89
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/Kconfig
@@ -0,0 +1,17 @@
+config NET_VENDOR_MEDIATEK
+	bool "MediaTek ethernet driver"
+	depends on ARCH_MEDIATEK
+	---help---
+	  If you have a Mediatek SoC with ethernet, say Y.
+
+if NET_VENDOR_MEDIATEK
+
+config NET_MEDIATEK_SOC
+	tristate "MediaTek MT7623 Gigabit ethernet support"
+	depends on NET_VENDOR_MEDIATEK && (MACH_MT7623 || MACH_MT2701)
+	select PHYLIB
+	---help---
+	  This driver supports the gigabit ethernet MACs in the
+	  MediaTek MT2701/MT7623 chipset family.
+
+endif #NET_VENDOR_MEDIATEK
diff --git a/drivers/net/ethernet/mediatek/Makefile b/drivers/net/ethernet/mediatek/Makefile
new file mode 100644
index 0000000..aa3f1c8
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the Mediatek SoCs built-in ethernet macs
+#
+
+obj-$(CONFIG_NET_MEDIATEK_SOC)			+= mtk_eth_soc.o
-- 
cgit v0.10.2


From 0c272fc9b0d6e559835ebfe6bdf3b8c12a570d17 Mon Sep 17 00:00:00 2001
From: John Crispin <blogic@openwrt.org>
Date: Tue, 8 Mar 2016 11:29:57 +0100
Subject: net-next: mediatek: add an entry to MAINTAINERS

Add myself and Felix as the Maintainers for the MediaTek ethernet driver.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/MAINTAINERS b/MAINTAINERS
index 2132c99..1663ad4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7035,6 +7035,13 @@ F:	include/uapi/linux/meye.h
 F:	include/uapi/linux/ivtv*
 F:	include/uapi/linux/uvcvideo.h
 
+MEDIATEK ETHERNET DRIVER
+M:	Felix Fietkau <nbd@openwrt.org>
+M:	John Crispin <blogic@openwrt.org>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/net/ethernet/mediatek/
+
 MEDIATEK MT7601U WIRELESS LAN DRIVER
 M:	Jakub Kicinski <kubakici@wp.pl>
 L:	linux-wireless@vger.kernel.org
-- 
cgit v0.10.2


From 5b33f48842fa1e13e9c0ea8cc59c1d0df19042db Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:29 +0200
Subject: net/flower: Introduce hardware offload support

This patch is based on a patch made by John Fastabend.
It adds support for offloading cls_flower.
when NETIF_F_HW_TC is on:
  flags = 0       => Rule will be processed twice - by hardware, and if
                     still relevant, by software.
  flags = SKIP_HW => Rull will be processed by software only

If hardware fail/not capabale to apply the rule, operation will NOT
fail. Filter will be processed by SW only.

Acked-by: Jiri Pirko <jiri@mellanox.com>
Suggested-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fd30cb5..41df0b4 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -786,6 +786,7 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
 enum {
 	TC_SETUP_MQPRIO,
 	TC_SETUP_CLSU32,
+	TC_SETUP_CLSFLOWER,
 };
 
 struct tc_cls_u32_offload;
@@ -795,6 +796,7 @@ struct tc_to_netdev {
 	union {
 		u8 tc;
 		struct tc_cls_u32_offload *cls_u32;
+		struct tc_cls_flower_offload *cls_flower;
 	};
 };
 
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index bea14ee..5b4e8f0 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -409,4 +409,18 @@ static inline bool tc_should_offload(struct net_device *dev, u32 flags)
 	return true;
 }
 
+enum tc_fl_command {
+	TC_CLSFLOWER_REPLACE,
+	TC_CLSFLOWER_DESTROY,
+};
+
+struct tc_cls_flower_offload {
+	enum tc_fl_command command;
+	u64 cookie;
+	struct flow_dissector *dissector;
+	struct fl_flow_key *mask;
+	struct fl_flow_key *key;
+	struct tcf_exts *exts;
+};
+
 #endif
diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index 9874f568..c43c5f7 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -417,6 +417,8 @@ enum {
 	TCA_FLOWER_KEY_TCP_DST,		/* be16 */
 	TCA_FLOWER_KEY_UDP_SRC,		/* be16 */
 	TCA_FLOWER_KEY_UDP_DST,		/* be16 */
+
+	TCA_FLOWER_FLAGS,
 	__TCA_FLOWER_MAX,
 };
 
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 95b0212..25d8766 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -165,6 +165,51 @@ static void fl_destroy_filter(struct rcu_head *head)
 	kfree(f);
 }
 
+static void fl_hw_destroy_filter(struct tcf_proto *tp, u64 cookie)
+{
+	struct net_device *dev = tp->q->dev_queue->dev;
+	struct tc_cls_flower_offload offload = {0};
+	struct tc_to_netdev tc;
+
+	if (!tc_should_offload(dev, 0))
+		return;
+
+	offload.command = TC_CLSFLOWER_DESTROY;
+	offload.cookie = cookie;
+
+	tc.type = TC_SETUP_CLSFLOWER;
+	tc.cls_flower = &offload;
+
+	dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, &tc);
+}
+
+static void fl_hw_replace_filter(struct tcf_proto *tp,
+				 struct flow_dissector *dissector,
+				 struct fl_flow_key *mask,
+				 struct fl_flow_key *key,
+				 struct tcf_exts *actions,
+				 u64 cookie, u32 flags)
+{
+	struct net_device *dev = tp->q->dev_queue->dev;
+	struct tc_cls_flower_offload offload = {0};
+	struct tc_to_netdev tc;
+
+	if (!tc_should_offload(dev, flags))
+		return;
+
+	offload.command = TC_CLSFLOWER_REPLACE;
+	offload.cookie = cookie;
+	offload.dissector = dissector;
+	offload.mask = mask;
+	offload.key = key;
+	offload.exts = actions;
+
+	tc.type = TC_SETUP_CLSFLOWER;
+	tc.cls_flower = &offload;
+
+	dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, &tc);
+}
+
 static bool fl_destroy(struct tcf_proto *tp, bool force)
 {
 	struct cls_fl_head *head = rtnl_dereference(tp->root);
@@ -174,6 +219,7 @@ static bool fl_destroy(struct tcf_proto *tp, bool force)
 		return false;
 
 	list_for_each_entry_safe(f, next, &head->filters, list) {
+		fl_hw_destroy_filter(tp, (u64)f);
 		list_del_rcu(&f->list);
 		call_rcu(&f->rcu, fl_destroy_filter);
 	}
@@ -459,6 +505,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 	struct cls_fl_filter *fnew;
 	struct nlattr *tb[TCA_FLOWER_MAX + 1];
 	struct fl_flow_mask mask = {};
+	u32 flags = 0;
 	int err;
 
 	if (!tca[TCA_OPTIONS])
@@ -486,6 +533,9 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 	}
 	fnew->handle = handle;
 
+	if (tb[TCA_FLOWER_FLAGS])
+		flags = nla_get_u32(tb[TCA_FLOWER_FLAGS]);
+
 	err = fl_set_parms(net, tp, fnew, &mask, base, tb, tca[TCA_RATE], ovr);
 	if (err)
 		goto errout;
@@ -498,9 +548,20 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 				     head->ht_params);
 	if (err)
 		goto errout;
-	if (fold)
+
+	fl_hw_replace_filter(tp,
+			     &head->dissector,
+			     &mask.key,
+			     &fnew->key,
+			     &fnew->exts,
+			     (u64)fnew,
+			     flags);
+
+	if (fold) {
 		rhashtable_remove_fast(&head->ht, &fold->ht_node,
 				       head->ht_params);
+		fl_hw_destroy_filter(tp, (u64)fold);
+	}
 
 	*arg = (unsigned long) fnew;
 
@@ -527,6 +588,7 @@ static int fl_delete(struct tcf_proto *tp, unsigned long arg)
 	rhashtable_remove_fast(&head->ht, &f->ht_node,
 			       head->ht_params);
 	list_del_rcu(&f->list);
+	fl_hw_destroy_filter(tp, (u64)f);
 	tcf_unbind_filter(tp, &f->res);
 	call_rcu(&f->rcu, fl_destroy_filter);
 	return 0;
-- 
cgit v0.10.2


From 8de2d793daf784f8f109565bcc023a6d198bad85 Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:30 +0200
Subject: net/flow_dissector: Make dissector_uses_key() and
 skb_flow_dissector_target() public

Will be used in a following patch to query if a key is being used, and
what it's value in the target object.

Acked-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index 8c8548c..d3d60dc 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -184,4 +184,17 @@ static inline bool flow_keys_have_l4(struct flow_keys *keys)
 
 u32 flow_hash_from_keys(struct flow_keys *keys);
 
+static inline bool dissector_uses_key(const struct flow_dissector *flow_dissector,
+				      enum flow_dissector_key_id key_id)
+{
+	return flow_dissector->used_keys & (1 << key_id);
+}
+
+static inline void *skb_flow_dissector_target(struct flow_dissector *flow_dissector,
+					      enum flow_dissector_key_id key_id,
+					      void *target_container)
+{
+	return ((char *)target_container) + flow_dissector->offset[key_id];
+}
+
 #endif
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 7c7b873..a669dea 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -19,25 +19,12 @@
 #include <net/flow_dissector.h>
 #include <scsi/fc/fc_fcoe.h>
 
-static bool dissector_uses_key(const struct flow_dissector *flow_dissector,
-			       enum flow_dissector_key_id key_id)
-{
-	return flow_dissector->used_keys & (1 << key_id);
-}
-
 static void dissector_set_key(struct flow_dissector *flow_dissector,
 			      enum flow_dissector_key_id key_id)
 {
 	flow_dissector->used_keys |= (1 << key_id);
 }
 
-static void *skb_flow_dissector_target(struct flow_dissector *flow_dissector,
-				       enum flow_dissector_key_id key_id,
-				       void *target_container)
-{
-	return ((char *) target_container) + flow_dissector->offset[key_id];
-}
-
 void skb_flow_dissector_init(struct flow_dissector *flow_dissector,
 			     const struct flow_dissector_key *key,
 			     unsigned int key_count)
-- 
cgit v0.10.2


From 00175aec941e9c306d8a5ce930b2d91f7c04468c Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:31 +0200
Subject: net/sched: Macro instead of CONFIG_NET_CLS_ACT ifdef

Introduce the macros tc_no_actions and tc_for_each_action to make code
clearer.
Extracted struct tc_action out of the ifdef to make calls to
is_tcf_gact_shot() and similar functions valid, even when it is a nop.

Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Suggested-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/act_api.h b/include/net/act_api.h
index 342be6c..2a19fe1 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -78,11 +78,6 @@ static inline void tcf_lastuse_update(struct tcf_t *tm)
 		tm->lastuse = now;
 }
 
-#ifdef CONFIG_NET_CLS_ACT
-
-#define ACT_P_CREATED 1
-#define ACT_P_DELETED 1
-
 struct tc_action {
 	void			*priv;
 	const struct tc_action_ops	*ops;
@@ -92,6 +87,11 @@ struct tc_action {
 	struct tcf_hashinfo	*hinfo;
 };
 
+#ifdef CONFIG_NET_CLS_ACT
+
+#define ACT_P_CREATED 1
+#define ACT_P_DELETED 1
+
 struct tc_action_ops {
 	struct list_head head;
 	char    kind[IFNAMSIZ];
@@ -171,5 +171,16 @@ int tcf_action_dump(struct sk_buff *skb, struct list_head *, int, int);
 int tcf_action_dump_old(struct sk_buff *skb, struct tc_action *a, int, int);
 int tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int, int);
 int tcf_action_copy_stats(struct sk_buff *, struct tc_action *, int);
+
+#define tc_no_actions(_exts) \
+	(list_empty(&(_exts)->actions))
+
+#define tc_for_each_action(_a, _exts) \
+	list_for_each_entry(a, &(_exts)->actions, list)
+#else /* CONFIG_NET_CLS_ACT */
+
+#define tc_no_actions(_exts) true
+#define tc_for_each_action(_a, _exts) while (0)
+
 #endif /* CONFIG_NET_CLS_ACT */
 #endif
diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h
index 04a3183..93c520b 100644
--- a/include/net/tc_act/tc_gact.h
+++ b/include/net/tc_act/tc_gact.h
@@ -16,9 +16,9 @@ struct tcf_gact {
 #define to_gact(a) \
 	container_of(a->priv, struct tcf_gact, common)
 
-#ifdef CONFIG_NET_CLS_ACT
 static inline bool is_tcf_gact_shot(const struct tc_action *a)
 {
+#ifdef CONFIG_NET_CLS_ACT
 	struct tcf_gact *gact;
 
 	if (a->ops && a->ops->type != TCA_ACT_GACT)
@@ -28,7 +28,7 @@ static inline bool is_tcf_gact_shot(const struct tc_action *a)
 	if (gact->tcf_action == TC_ACT_SHOT)
 		return true;
 
+#endif
 	return false;
 }
-#endif
 #endif /* __NET_TC_GACT_H */
-- 
cgit v0.10.2


From 519afb1813eab066a0c9995a08861fd0af75d5ae Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:32 +0200
Subject: net/act_skbedit: Utility functions for mark action

Enable device drivers to query the action, if and only if is a mark
action and what value to use for marking.

Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/tc_act/tc_skbedit.h b/include/net/tc_act/tc_skbedit.h
index 0df9a0d..b496d5a 100644
--- a/include/net/tc_act/tc_skbedit.h
+++ b/include/net/tc_act/tc_skbedit.h
@@ -20,6 +20,7 @@
 #define __NET_TC_SKBEDIT_H
 
 #include <net/act_api.h>
+#include <linux/tc_act/tc_skbedit.h>
 
 struct tcf_skbedit {
 	struct tcf_common	common;
@@ -32,4 +33,19 @@ struct tcf_skbedit {
 #define to_skbedit(a) \
 	container_of(a->priv, struct tcf_skbedit, common)
 
+/* Return true iff action is mark */
+static inline bool is_tcf_skbedit_mark(const struct tc_action *a)
+{
+#ifdef CONFIG_NET_CLS_ACT
+	if (a->ops && a->ops->type == TCA_ACT_SKBEDIT)
+		return to_skbedit(a)->flags == SKBEDIT_F_MARK;
+#endif
+	return false;
+}
+
+static inline u32 tcf_skbedit_mark(const struct tc_action *a)
+{
+	return to_skbedit(a)->mark;
+}
+
 #endif /* __NET_TC_SKBEDIT_H */
-- 
cgit v0.10.2


From 60ab4584f5bf4e496baec7b090a4b382d03b2100 Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:33 +0200
Subject: net/mlx5_core: Set flow steering dest only for forward rules

We need to handle flow table entry destinations only if the action
associated with the rule is forwarding (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST).

Fixes: 26a8145390b3 ('net/mlx5_core: Introduce flow steering firmware commands')
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index a9894d2..f46f1db 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -218,19 +218,22 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
 				      match_value);
 	memcpy(in_match_value, &fte->val, MLX5_ST_SZ_BYTES(fte_match_param));
 
-	in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination);
-	list_for_each_entry(dst, &fte->node.children, node.list) {
-		unsigned int id;
-
-		MLX5_SET(dest_format_struct, in_dests, destination_type,
-			 dst->dest_attr.type);
-		if (dst->dest_attr.type ==
-		    MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE)
-			id = dst->dest_attr.ft->id;
-		else
-			id = dst->dest_attr.tir_num;
-		MLX5_SET(dest_format_struct, in_dests, destination_id, id);
-		in_dests += MLX5_ST_SZ_BYTES(dest_format_struct);
+	if (fte->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
+		in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination);
+		list_for_each_entry(dst, &fte->node.children, node.list) {
+			unsigned int id;
+
+			MLX5_SET(dest_format_struct, in_dests, destination_type,
+				 dst->dest_attr.type);
+			if (dst->dest_attr.type ==
+			    MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE) {
+				id = dst->dest_attr.ft->id;
+			} else {
+				id = dst->dest_attr.tir_num;
+			}
+			MLX5_SET(dest_format_struct, in_dests, destination_id, id);
+			in_dests += MLX5_ST_SZ_BYTES(dest_format_struct);
+		}
 	}
 	memset(out, 0, sizeof(out));
 	err = mlx5_cmd_exec_check_status(dev, in, inlen, out,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 6f68dba..f0e67d2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -360,8 +360,8 @@ static void del_rule(struct fs_node *node)
 	memcpy(match_value, fte->val, sizeof(fte->val));
 	fs_get_obj(ft, fg->node.parent);
 	list_del(&rule->node.list);
-	fte->dests_size--;
-	if (fte->dests_size) {
+	if ((fte->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) &&
+	    --fte->dests_size) {
 		err = mlx5_cmd_update_fte(dev, ft,
 					  fg->id, fte);
 		if (err)
@@ -763,7 +763,8 @@ static struct mlx5_flow_rule *alloc_rule(struct mlx5_flow_destination *dest)
 		return NULL;
 
 	rule->node.type = FS_TYPE_FLOW_DEST;
-	memcpy(&rule->dest_attr, dest, sizeof(*dest));
+	if (dest)
+		memcpy(&rule->dest_attr, dest, sizeof(*dest));
 
 	return rule;
 }
@@ -785,8 +786,9 @@ static struct mlx5_flow_rule *add_rule_fte(struct fs_fte *fte,
 	/* Add dest to dests list- added as first element after the head */
 	tree_init_node(&rule->node, 1, del_rule);
 	list_add_tail(&rule->node.list, &fte->node.children);
-	fte->dests_size++;
-	if (fte->dests_size == 1)
+	if (dest)
+		fte->dests_size++;
+	if (fte->dests_size == 1 || !dest)
 		err = mlx5_cmd_create_fte(get_dev(&ft->node),
 					  ft, fg->id, fte);
 	else
@@ -802,7 +804,8 @@ static struct mlx5_flow_rule *add_rule_fte(struct fs_fte *fte,
 free_rule:
 	list_del(&rule->node.list);
 	kfree(rule);
-	fte->dests_size--;
+	if (dest)
+		fte->dests_size--;
 	return ERR_PTR(err);
 }
 
@@ -996,6 +999,9 @@ mlx5_add_flow_rule(struct mlx5_flow_table *ft,
 	struct mlx5_flow_group *g;
 	struct mlx5_flow_rule *rule;
 
+	if ((action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) && !dest)
+		return ERR_PTR(-EINVAL);
+
 	nested_lock_ref_node(&ft->node, FS_MUTEX_GRANDPARENT);
 	fs_for_each_fg(g, ft)
 		if (compare_match_criteria(g->mask.match_criteria_enable,
-- 
cgit v0.10.2


From 67ba422e95edfbd42f1d9782868880b364eea173 Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:34 +0200
Subject: net/mlx5e: Relax ndo_setup_tc handle restriction

Restricting handle to TC_H_ROOT breaks the old instantiation of mqprio
to setup a hardware qdisc. This patch relaxes the test, to only check the
type.

Fixes: 08fb1da ("net/mlx5e: Support DCBNL IEEE ETS")
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 19e5dae..f1ebe06 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1883,7 +1883,7 @@ static int mlx5e_setup_tc(struct net_device *netdev, u8 tc)
 static int mlx5e_ndo_setup_tc(struct net_device *dev, u32 handle,
 			      __be16 proto, struct tc_to_netdev *tc)
 {
-	if (handle != TC_H_ROOT || tc->type != TC_SETUP_MQPRIO)
+	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
 	return mlx5e_setup_tc(dev, tc->tc);
-- 
cgit v0.10.2


From b6172aac71e9fba0981e013839ced5fcd97746fb Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:35 +0200
Subject: net/mlx5e: Add a new priority for kernel flow tables

Move the vlan and main flow tables to use priority 1. This will allow
the upcoming TC offload logic to use a higher priority (0) for the
offload steering table.

Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index 80d81ab..d00a242 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -1041,7 +1041,7 @@ static int mlx5e_create_main_flow_table(struct mlx5e_priv *priv)
 	int err;
 
 	ft->num_groups = 0;
-	ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_MAIN_TABLE_SIZE);
+	ft->t = mlx5_create_flow_table(priv->fts.ns, 1, MLX5E_MAIN_TABLE_SIZE);
 
 	if (IS_ERR(ft->t)) {
 		err = PTR_ERR(ft->t);
@@ -1150,7 +1150,7 @@ static int mlx5e_create_vlan_flow_table(struct mlx5e_priv *priv)
 	int err;
 
 	ft->num_groups = 0;
-	ft->t = mlx5_create_flow_table(priv->fts.ns, 0, MLX5E_VLAN_TABLE_SIZE);
+	ft->t = mlx5_create_flow_table(priv->fts.ns, 1, MLX5E_VLAN_TABLE_SIZE);
 
 	if (IS_ERR(ft->t)) {
 		err = PTR_ERR(ft->t);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index f0e67d2..e848d70 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -73,8 +73,8 @@
 #define BY_PASS_MIN_LEVEL (KENREL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\
 			   LEFTOVERS_MAX_FT)
 
-#define KERNEL_MAX_FT 2
-#define KERNEL_NUM_PRIOS 1
+#define KERNEL_MAX_FT 3
+#define KERNEL_NUM_PRIOS 2
 #define KENREL_MIN_LEVEL 2
 
 struct node_caps {
-- 
cgit v0.10.2


From e8f887ac6a45ecb7f881f278a3fed1fbf002ef0b Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:36 +0200
Subject: net/mlx5e: Introduce tc offload support

Extend ndo_setup_tc() to support ingress tc offloading. Will be used by
later patches to offload tc flower filter.

Feature is off by default and could be enabled by issuing:
 # ethtool  -K eth0 hw-tc-offload on

Offloads flow table is dynamically created when first filter is
added.
Rules are saved in a hash table that is maintained by the consumer (for
example - the flower offload in the next patch).
When last filter is removed and no filters exist in the hash table, the
offload flow table is destroyed.

Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 11b592d..4fc45ee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -6,6 +6,6 @@ mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 
 mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \
 		en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \
-		en_txrx.o en_clock.o vxlan.o
+		en_txrx.o en_clock.o vxlan.o en_tc.o
 
 mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) +=  en_dcbnl.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index dbc2fb8..0f76d32 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -43,6 +43,7 @@
 #include <linux/mlx5/port.h>
 #include <linux/mlx5/vport.h>
 #include <linux/mlx5/transobj.h>
+#include <linux/rhashtable.h>
 #include "wq.h"
 #include "mlx5_core.h"
 
@@ -527,8 +528,16 @@ struct mlx5e_flow_table {
 	struct mlx5_flow_group		**g;
 };
 
+struct mlx5e_tc_flow_table {
+	struct mlx5_flow_table		*t;
+
+	struct rhashtable_params        ht_params;
+	struct rhashtable               ht;
+};
+
 struct mlx5e_flow_tables {
 	struct mlx5_flow_namespace	*ns;
+	struct mlx5e_tc_flow_table	tc;
 	struct mlx5e_flow_table		vlan;
 	struct mlx5e_flow_table		main;
 };
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index f1ebe06..e621048 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -30,9 +30,12 @@
  * SOFTWARE.
  */
 
+#include <net/tc_act/tc_gact.h>
+#include <net/pkt_cls.h>
 #include <linux/mlx5/fs.h>
 #include <net/vxlan.h>
 #include "en.h"
+#include "en_tc.h"
 #include "eswitch.h"
 #include "vxlan.h"
 
@@ -1883,6 +1886,17 @@ static int mlx5e_setup_tc(struct net_device *netdev, u8 tc)
 static int mlx5e_ndo_setup_tc(struct net_device *dev, u32 handle,
 			      __be16 proto, struct tc_to_netdev *tc)
 {
+	struct mlx5e_priv *priv = netdev_priv(dev);
+
+	if (TC_H_MAJ(handle) != TC_H_MAJ(TC_H_INGRESS))
+		goto mqprio;
+
+	switch (tc->type) {
+	default:
+		return -EOPNOTSUPP;
+	}
+
+mqprio:
 	if (tc->type != TC_SETUP_MQPRIO)
 		return -EINVAL;
 
@@ -1968,6 +1982,13 @@ static int mlx5e_set_features(struct net_device *netdev,
 			mlx5e_disable_vlan_filter(priv);
 	}
 
+	if ((changes & NETIF_F_HW_TC) && !(features & NETIF_F_HW_TC) &&
+	    mlx5e_tc_num_filters(priv)) {
+		netdev_err(netdev,
+			   "Active offloaded tc filters, can't turn hw_tc_offload off\n");
+		return -EINVAL;
+	}
+
 	return err;
 }
 
@@ -2375,6 +2396,13 @@ static void mlx5e_build_netdev(struct net_device *netdev)
 	if (!priv->params.lro_en)
 		netdev->features  &= ~NETIF_F_LRO;
 
+#define FT_CAP(f) MLX5_CAP_FLOWTABLE(mdev, flow_table_properties_nic_receive.f)
+	if (FT_CAP(flow_modify_en) &&
+	    FT_CAP(modify_root) &&
+	    FT_CAP(identified_miss_table_mode) &&
+	    FT_CAP(flow_table_modify))
+		priv->netdev->hw_features      |= NETIF_F_HW_TC;
+
 	netdev->features         |= NETIF_F_HIGHDMA;
 
 	netdev->priv_flags       |= IFF_UNICAST_FLT;
@@ -2496,6 +2524,10 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 
 	mlx5e_vxlan_init(priv);
 
+	err = mlx5e_tc_init(priv);
+	if (err)
+		goto err_destroy_flow_tables;
+
 #ifdef CONFIG_MLX5_CORE_EN_DCB
 	mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets);
 #endif
@@ -2503,7 +2535,7 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 	err = register_netdev(netdev);
 	if (err) {
 		mlx5_core_err(mdev, "register_netdev failed, %d\n", err);
-		goto err_destroy_flow_tables;
+		goto err_tc_cleanup;
 	}
 
 	if (mlx5e_vxlan_allowed(mdev))
@@ -2514,6 +2546,9 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev)
 
 	return priv;
 
+err_tc_cleanup:
+	mlx5e_tc_cleanup(priv);
+
 err_destroy_flow_tables:
 	mlx5e_destroy_flow_tables(priv);
 
@@ -2561,6 +2596,7 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv)
 	mlx5e_disable_async_events(priv);
 	flush_scheduled_work();
 	unregister_netdev(netdev);
+	mlx5e_tc_cleanup(priv);
 	mlx5e_vxlan_cleanup(priv);
 	mlx5e_destroy_flow_tables(priv);
 	mlx5e_destroy_tirs(priv);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
new file mode 100644
index 0000000..1dc715d
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -0,0 +1,131 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/mlx5/fs.h>
+#include <linux/mlx5/device.h>
+#include <linux/rhashtable.h>
+#include "en.h"
+#include "en_tc.h"
+
+struct mlx5e_tc_flow {
+	struct rhash_head	node;
+	u64			cookie;
+	struct mlx5_flow_rule	*rule;
+};
+
+#define MLX5E_TC_FLOW_TABLE_NUM_ENTRIES 1024
+#define MLX5E_TC_FLOW_TABLE_NUM_GROUPS 4
+
+static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv,
+						u32 *match_c, u32 *match_v,
+						u32 action, u32 flow_tag)
+{
+	struct mlx5_flow_destination dest = {
+		.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE,
+		{.ft = priv->fts.vlan.t},
+	};
+	struct mlx5_flow_rule *rule;
+	bool table_created = false;
+
+	if (IS_ERR_OR_NULL(priv->fts.tc.t)) {
+		priv->fts.tc.t =
+			mlx5_create_auto_grouped_flow_table(priv->fts.ns, 0,
+							    MLX5E_TC_FLOW_TABLE_NUM_ENTRIES,
+							    MLX5E_TC_FLOW_TABLE_NUM_GROUPS);
+		if (IS_ERR(priv->fts.tc.t)) {
+			netdev_err(priv->netdev,
+				   "Failed to create tc offload table\n");
+			return ERR_CAST(priv->fts.tc.t);
+		}
+
+		table_created = true;
+	}
+
+	rule = mlx5_add_flow_rule(priv->fts.tc.t, MLX5_MATCH_OUTER_HEADERS,
+				  match_c, match_v,
+				  action, flow_tag,
+				  action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST ? &dest : NULL);
+
+	if (IS_ERR(rule) && table_created) {
+		mlx5_destroy_flow_table(priv->fts.tc.t);
+		priv->fts.tc.t = NULL;
+	}
+
+	return rule;
+}
+
+static void mlx5e_tc_del_flow(struct mlx5e_priv *priv,
+			      struct mlx5_flow_rule *rule)
+{
+	mlx5_del_flow_rule(rule);
+
+	if (!mlx5e_tc_num_filters(priv)) {
+		mlx5_destroy_flow_table(priv->fts.tc.t);
+		priv->fts.tc.t = NULL;
+	}
+}
+
+static const struct rhashtable_params mlx5e_tc_flow_ht_params = {
+	.head_offset = offsetof(struct mlx5e_tc_flow, node),
+	.key_offset = offsetof(struct mlx5e_tc_flow, cookie),
+	.key_len = sizeof(((struct mlx5e_tc_flow *)0)->cookie),
+	.automatic_shrinking = true,
+};
+
+int mlx5e_tc_init(struct mlx5e_priv *priv)
+{
+	struct mlx5e_tc_flow_table *tc = &priv->fts.tc;
+
+	tc->ht_params = mlx5e_tc_flow_ht_params;
+	return rhashtable_init(&tc->ht, &tc->ht_params);
+}
+
+static void _mlx5e_tc_del_flow(void *ptr, void *arg)
+{
+	struct mlx5e_tc_flow *flow = ptr;
+	struct mlx5e_priv *priv = arg;
+
+	mlx5e_tc_del_flow(priv, flow->rule);
+	kfree(flow);
+}
+
+void mlx5e_tc_cleanup(struct mlx5e_priv *priv)
+{
+	struct mlx5e_tc_flow_table *tc = &priv->fts.tc;
+
+	rhashtable_free_and_destroy(&tc->ht, _mlx5e_tc_del_flow, priv);
+
+	if (!IS_ERR_OR_NULL(priv->fts.tc.t)) {
+		mlx5_destroy_flow_table(priv->fts.tc.t);
+		priv->fts.tc.t = NULL;
+	}
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
new file mode 100644
index 0000000..46eacc5
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
@@ -0,0 +1,44 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __MLX5_EN_TC_H__
+#define __MLX5_EN_TC_H__
+
+int mlx5e_tc_init(struct mlx5e_priv *priv);
+void mlx5e_tc_cleanup(struct mlx5e_priv *priv);
+
+static inline int mlx5e_tc_num_filters(struct mlx5e_priv *priv)
+{
+	return atomic_read(&priv->fts.tc.ht.nelems);
+}
+
+#endif /* __MLX5_EN_TC_H__ */
-- 
cgit v0.10.2


From e3a2b7ed018e885721a1b664ed7bf0f3fe29c81c Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:37 +0200
Subject: net/mlx5e: Support offload cls_flower with drop action

Parse tc_cls_flower_offload into device specific commands and program
the hardware to classify and act accordingly.

For example, to drop ICMP (ip_proto 1) packets from specific smac, dmac,
src_ip, src_ip, arriving to interface ens9:

 # tc qdisc add dev ens9 ingress

 # tc filter add dev ens9 protocol ip parent ffff: \
     flower ip_proto 1 \
     dst_mac 7c:fe:90:69:81:62 src_mac 7c:fe:90:69:81:56 \
     dst_ip 11.11.11.11 src_ip 11.11.11.12 indev ens9 \
     action drop

Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index e621048..ac58078 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1892,6 +1892,13 @@ static int mlx5e_ndo_setup_tc(struct net_device *dev, u32 handle,
 		goto mqprio;
 
 	switch (tc->type) {
+	case TC_SETUP_CLSFLOWER:
+		switch (tc->cls_flower->command) {
+		case TC_CLSFLOWER_REPLACE:
+			return mlx5e_configure_flower(priv, proto, tc->cls_flower);
+		case TC_CLSFLOWER_DESTROY:
+			return mlx5e_delete_flower(priv, tc->cls_flower);
+		}
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 1dc715d..3ed4d96 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -30,6 +30,9 @@
  * SOFTWARE.
  */
 
+#include <net/flow_dissector.h>
+#include <net/pkt_cls.h>
+#include <net/tc_act/tc_gact.h>
 #include <linux/mlx5/fs.h>
 #include <linux/mlx5/device.h>
 #include <linux/rhashtable.h>
@@ -94,6 +97,300 @@ static void mlx5e_tc_del_flow(struct mlx5e_priv *priv,
 	}
 }
 
+static int parse_cls_flower(struct mlx5e_priv *priv,
+			    u32 *match_c, u32 *match_v,
+			    struct tc_cls_flower_offload *f)
+{
+	void *headers_c = MLX5_ADDR_OF(fte_match_param, match_c, outer_headers);
+	void *headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers);
+	u16 addr_type = 0;
+	u8 ip_proto = 0;
+
+	if (f->dissector->used_keys &
+	    ~(BIT(FLOW_DISSECTOR_KEY_CONTROL) |
+	      BIT(FLOW_DISSECTOR_KEY_BASIC) |
+	      BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
+	      BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
+	      BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
+	      BIT(FLOW_DISSECTOR_KEY_PORTS))) {
+		netdev_warn(priv->netdev, "Unsupported key used: 0x%x\n",
+			    f->dissector->used_keys);
+		return -EOPNOTSUPP;
+	}
+
+	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_CONTROL)) {
+		struct flow_dissector_key_control *key =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_BASIC,
+						  f->key);
+		addr_type = key->addr_type;
+	}
+
+	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) {
+		struct flow_dissector_key_basic *key =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_BASIC,
+						  f->key);
+		struct flow_dissector_key_basic *mask =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_BASIC,
+						  f->mask);
+		ip_proto = key->ip_proto;
+
+		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype,
+			 ntohs(mask->n_proto));
+		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype,
+			 ntohs(key->n_proto));
+
+		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol,
+			 mask->ip_proto);
+		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol,
+			 key->ip_proto);
+	}
+
+	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+		struct flow_dissector_key_eth_addrs *key =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_ETH_ADDRS,
+						  f->key);
+		struct flow_dissector_key_eth_addrs *mask =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_ETH_ADDRS,
+						  f->mask);
+
+		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+					     dmac_47_16),
+				mask->dst);
+		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+					     dmac_47_16),
+				key->dst);
+
+		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+					     smac_47_16),
+				mask->src);
+		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+					     smac_47_16),
+				key->src);
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+		struct flow_dissector_key_ipv4_addrs *key =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_IPV4_ADDRS,
+						  f->key);
+		struct flow_dissector_key_ipv4_addrs *mask =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_IPV4_ADDRS,
+						  f->mask);
+
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+				    src_ipv4_src_ipv6.ipv4_layout.ipv4),
+		       &mask->src, sizeof(mask->src));
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+				    src_ipv4_src_ipv6.ipv4_layout.ipv4),
+		       &key->src, sizeof(key->src));
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+				    dst_ipv4_dst_ipv6.ipv4_layout.ipv4),
+		       &mask->dst, sizeof(mask->dst));
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+				    dst_ipv4_dst_ipv6.ipv4_layout.ipv4),
+		       &key->dst, sizeof(key->dst));
+	}
+
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+		struct flow_dissector_key_ipv6_addrs *key =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_IPV6_ADDRS,
+						  f->key);
+		struct flow_dissector_key_ipv6_addrs *mask =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_IPV6_ADDRS,
+						  f->mask);
+
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+				    src_ipv4_src_ipv6.ipv6_layout.ipv6),
+		       &mask->src, sizeof(mask->src));
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+				    src_ipv4_src_ipv6.ipv6_layout.ipv6),
+		       &key->src, sizeof(key->src));
+
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+				    dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
+		       &mask->dst, sizeof(mask->dst));
+		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+				    dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
+		       &key->dst, sizeof(key->dst));
+	}
+
+	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_PORTS)) {
+		struct flow_dissector_key_ports *key =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_PORTS,
+						  f->key);
+		struct flow_dissector_key_ports *mask =
+			skb_flow_dissector_target(f->dissector,
+						  FLOW_DISSECTOR_KEY_PORTS,
+						  f->mask);
+		switch (ip_proto) {
+		case IPPROTO_TCP:
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+				 tcp_sport, ntohs(mask->src));
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+				 tcp_sport, ntohs(key->src));
+
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+				 tcp_dport, ntohs(mask->dst));
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+				 tcp_dport, ntohs(key->dst));
+			break;
+
+		case IPPROTO_UDP:
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+				 udp_sport, ntohs(mask->src));
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+				 udp_sport, ntohs(key->src));
+
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+				 udp_dport, ntohs(mask->dst));
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+				 udp_dport, ntohs(key->dst));
+			break;
+		default:
+			netdev_err(priv->netdev,
+				   "Only UDP and TCP transport are supported\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int parse_tc_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
+			    u32 *action, u32 *flow_tag)
+{
+	const struct tc_action *a;
+
+	if (tc_no_actions(exts))
+		return -EINVAL;
+
+	*flow_tag = MLX5_FS_DEFAULT_FLOW_TAG;
+	*action = 0;
+
+	tc_for_each_action(a, exts) {
+		/* Only support a single action per rule */
+		if (*action)
+			return -EINVAL;
+
+		if (is_tcf_gact_shot(a)) {
+			*action |= MLX5_FLOW_CONTEXT_ACTION_DROP;
+			continue;
+		}
+
+		if (is_tcf_skbedit_mark(a)) {
+			u32 mark = tcf_skbedit_mark(a);
+
+			if (mark & ~MLX5E_TC_FLOW_ID_MASK) {
+				netdev_warn(priv->netdev, "Bad flow mark - only 16 bit is supported: 0x%x\n",
+					    mark);
+				return -EINVAL;
+			}
+
+			*flow_tag = mark;
+			*action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+			continue;
+		}
+
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol,
+			   struct tc_cls_flower_offload *f)
+{
+	struct mlx5e_tc_flow_table *tc = &priv->fts.tc;
+	u32 *match_c;
+	u32 *match_v;
+	int err = 0;
+	u32 flow_tag;
+	u32 action;
+	struct mlx5e_tc_flow *flow;
+	struct mlx5_flow_rule *old = NULL;
+
+	flow = rhashtable_lookup_fast(&tc->ht, &f->cookie,
+				      tc->ht_params);
+	if (flow)
+		old = flow->rule;
+	else
+		flow = kzalloc(sizeof(*flow), GFP_KERNEL);
+
+	match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL);
+	match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL);
+	if (!match_c || !match_v || !flow) {
+		err = -ENOMEM;
+		goto err_free;
+	}
+
+	flow->cookie = f->cookie;
+
+	err = parse_cls_flower(priv, match_c, match_v, f);
+	if (err < 0)
+		goto err_free;
+
+	err = parse_tc_actions(priv, f->exts, &action, &flow_tag);
+	if (err < 0)
+		goto err_free;
+
+	err = rhashtable_insert_fast(&tc->ht, &flow->node,
+				     tc->ht_params);
+	if (err)
+		goto err_free;
+
+	flow->rule = mlx5e_tc_add_flow(priv, match_c, match_v, action,
+				       flow_tag);
+	if (IS_ERR(flow->rule)) {
+		err = PTR_ERR(flow->rule);
+		goto err_hash_del;
+	}
+
+	if (old)
+		mlx5e_tc_del_flow(priv, old);
+
+	goto out;
+
+err_hash_del:
+	rhashtable_remove_fast(&tc->ht, &flow->node, tc->ht_params);
+
+err_free:
+	if (!old)
+		kfree(flow);
+out:
+	kfree(match_c);
+	kfree(match_v);
+	return err;
+}
+
+int mlx5e_delete_flower(struct mlx5e_priv *priv,
+			struct tc_cls_flower_offload *f)
+{
+	struct mlx5e_tc_flow *flow;
+	struct mlx5e_tc_flow_table *tc = &priv->fts.tc;
+
+	flow = rhashtable_lookup_fast(&tc->ht, &f->cookie,
+				      tc->ht_params);
+	if (!flow)
+		return -EINVAL;
+
+	rhashtable_remove_fast(&tc->ht, &flow->node, tc->ht_params);
+
+	mlx5e_tc_del_flow(priv, flow->rule);
+
+	kfree(flow);
+
+	return 0;
+}
+
 static const struct rhashtable_params mlx5e_tc_flow_ht_params = {
 	.head_offset = offsetof(struct mlx5e_tc_flow, node),
 	.key_offset = offsetof(struct mlx5e_tc_flow, cookie),
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
index 46eacc5..70642f4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
@@ -36,6 +36,11 @@
 int mlx5e_tc_init(struct mlx5e_priv *priv);
 void mlx5e_tc_cleanup(struct mlx5e_priv *priv);
 
+int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol,
+			   struct tc_cls_flower_offload *f);
+int mlx5e_delete_flower(struct mlx5e_priv *priv,
+			struct tc_cls_flower_offload *f);
+
 static inline int mlx5e_tc_num_filters(struct mlx5e_priv *priv)
 {
 	return atomic_read(&priv->fts.tc.ht.nelems);
-- 
cgit v0.10.2


From 12185a9fafa9cf39b73588c92aa49300ff3bf191 Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Tue, 8 Mar 2016 12:42:38 +0200
Subject: net/mlx5e: Support offload cls_flower with skbedit mark action

Introduce offloading of skbedit mark action.

For example, to mark with 0x1234, all TCP (ip_proto 6) packets arriving
to interface ens9:

 # tc qdisc add dev ens9 ingress
 # tc filter add dev ens9 protocol ip parent ffff: \
     flower ip_proto 6 \
     indev ens9 \
     action skbedit mark 0x1234

Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 884ed19..58d4e2f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -35,6 +35,7 @@
 #include <linux/tcp.h>
 #include <net/busy_poll.h>
 #include "en.h"
+#include "en_tc.h"
 
 static inline bool mlx5e_rx_hw_stamp(struct mlx5e_tstamp *tstamp)
 {
@@ -224,6 +225,8 @@ static inline void mlx5e_build_rx_skb(struct mlx5_cqe64 *cqe,
 	if (cqe_has_vlan(cqe))
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
 				       be16_to_cpu(cqe->vlan_info));
+
+	skb->mark = be32_to_cpu(cqe->sop_drop_qpn) & MLX5E_TC_FLOW_ID_MASK;
 }
 
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 3ed4d96..b3de09f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -33,6 +33,7 @@
 #include <net/flow_dissector.h>
 #include <net/pkt_cls.h>
 #include <net/tc_act/tc_gact.h>
+#include <net/tc_act/tc_skbedit.h>
 #include <linux/mlx5/fs.h>
 #include <linux/mlx5/device.h>
 #include <linux/rhashtable.h>
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
index 70642f4..d677428 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h
@@ -33,6 +33,8 @@
 #ifndef __MLX5_EN_TC_H__
 #define __MLX5_EN_TC_H__
 
+#define MLX5E_TC_FLOW_ID_MASK 0x0000ffff
+
 int mlx5e_tc_init(struct mlx5e_priv *priv);
 void mlx5e_tc_cleanup(struct mlx5e_priv *priv);
 
-- 
cgit v0.10.2


From f0716cd6eb89f769bff3b37120219720c994bb7e Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 9 Mar 2016 00:04:21 +0100
Subject: netfilter: nft_compat: check match/targetinfo attr size

We copy according to ->target|matchsize, so check that the netlink attribute
(which can include padding and might be larger) contains enough data.

Reported-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c
index 454841b..6228c42 100644
--- a/net/netfilter/nft_compat.c
+++ b/net/netfilter/nft_compat.c
@@ -660,6 +660,9 @@ nft_match_select_ops(const struct nft_ctx *ctx,
 	if (IS_ERR(match))
 		return ERR_PTR(-ENOENT);
 
+	if (match->matchsize > nla_len(tb[NFTA_MATCH_INFO]))
+		return ERR_PTR(-EINVAL);
+
 	/* This is the first time we use this match, allocate operations */
 	nft_match = kzalloc(sizeof(struct nft_xt), GFP_KERNEL);
 	if (nft_match == NULL)
@@ -740,6 +743,9 @@ nft_target_select_ops(const struct nft_ctx *ctx,
 	if (IS_ERR(target))
 		return ERR_PTR(-ENOENT);
 
+	if (target->targetsize > nla_len(tb[NFTA_TARGET_INFO]))
+		return ERR_PTR(-EINVAL);
+
 	/* This is the first time we use this target, allocate operations */
 	nft_target = kzalloc(sizeof(struct nft_xt), GFP_KERNEL);
 	if (nft_target == NULL)
-- 
cgit v0.10.2


From 6a19cc8c892b113fb84947207346f783de76dbd1 Mon Sep 17 00:00:00 2001
From: Johan Hedberg <johan.hedberg@intel.com>
Date: Fri, 11 Mar 2016 09:56:32 +0200
Subject: Bluetooth: Fix setting correct flags in AD

A recent change added MGMT_ADV_FLAG_DISCOV to the flags returned by
get_adv_instance_flags(), however failed to take into account limited
discoverable mode. This patch fixes the issue by setting the correct
discoverability flag in the AD data.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
index 95a545c..6e125d7 100644
--- a/net/bluetooth/hci_request.c
+++ b/net/bluetooth/hci_request.c
@@ -872,7 +872,9 @@ static u32 get_adv_instance_flags(struct hci_dev *hdev, u8 instance)
 		if (hci_dev_test_flag(hdev, HCI_ADVERTISING_CONNECTABLE))
 			flags |= MGMT_ADV_FLAG_CONNECTABLE;
 
-		if (hci_dev_test_flag(hdev, HCI_DISCOVERABLE))
+		if (hci_dev_test_flag(hdev, HCI_LIMITED_DISCOVERABLE))
+			flags |= MGMT_ADV_FLAG_LIMITED_DISCOV;
+		else if (hci_dev_test_flag(hdev, HCI_DISCOVERABLE))
 			flags |= MGMT_ADV_FLAG_DISCOV;
 
 		return flags;
-- 
cgit v0.10.2


From 6a0e78072c2ae7b20b14e0249d8108441ea928d2 Mon Sep 17 00:00:00 2001
From: Johan Hedberg <johan.hedberg@intel.com>
Date: Fri, 11 Mar 2016 09:56:33 +0200
Subject: Bluetooth: Fix potential buffer overflow with Add Advertising

The Add Advertising command handler does the appropriate checks for
the AD and Scan Response data, however fails to take into account the
general length of the mgmt command itself, which could lead to
potential buffer overflows. This patch adds the necessary check that
the mgmt command length is consistent with the given ad and scan_rsp
lengths.

Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Cc: stable@vger.kernel.org

diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index c856fb6..9e4b931 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -5995,6 +5995,10 @@ static int add_advertising(struct sock *sk, struct hci_dev *hdev,
 		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_ADVERTISING,
 				       MGMT_STATUS_INVALID_PARAMS);
 
+	if (data_len != sizeof(*cp) + cp->adv_data_len + cp->scan_rsp_len)
+		return mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_ADVERTISING,
+				       MGMT_STATUS_INVALID_PARAMS);
+
 	flags = __le32_to_cpu(cp->flags);
 	timeout = __le16_to_cpu(cp->timeout);
 	duration = __le16_to_cpu(cp->duration);
-- 
cgit v0.10.2


From 8208d21bf309551686b7a76d19059ae182a956d0 Mon Sep 17 00:00:00 2001
From: Amir Vadai <amir@vadai.me>
Date: Fri, 11 Mar 2016 11:08:45 +0200
Subject: net/flower: Fix pointer cast

Cast pointer to unsigned long instead of u64, to fix compilation warning
on 32 bit arch, spotted by 0day build.

Fixes: 5b33f48 ("net/flower: Introduce hardware offload support")
Signed-off-by: Amir Vadai <amir@vadai.me>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 5b4e8f0..caa5e18 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -416,7 +416,7 @@ enum tc_fl_command {
 
 struct tc_cls_flower_offload {
 	enum tc_fl_command command;
-	u64 cookie;
+	unsigned long cookie;
 	struct flow_dissector *dissector;
 	struct fl_flow_key *mask;
 	struct fl_flow_key *key;
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 25d8766..2181ffc 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -165,7 +165,7 @@ static void fl_destroy_filter(struct rcu_head *head)
 	kfree(f);
 }
 
-static void fl_hw_destroy_filter(struct tcf_proto *tp, u64 cookie)
+static void fl_hw_destroy_filter(struct tcf_proto *tp, unsigned long cookie)
 {
 	struct net_device *dev = tp->q->dev_queue->dev;
 	struct tc_cls_flower_offload offload = {0};
@@ -188,7 +188,7 @@ static void fl_hw_replace_filter(struct tcf_proto *tp,
 				 struct fl_flow_key *mask,
 				 struct fl_flow_key *key,
 				 struct tcf_exts *actions,
-				 u64 cookie, u32 flags)
+				 unsigned long cookie, u32 flags)
 {
 	struct net_device *dev = tp->q->dev_queue->dev;
 	struct tc_cls_flower_offload offload = {0};
@@ -219,7 +219,7 @@ static bool fl_destroy(struct tcf_proto *tp, bool force)
 		return false;
 
 	list_for_each_entry_safe(f, next, &head->filters, list) {
-		fl_hw_destroy_filter(tp, (u64)f);
+		fl_hw_destroy_filter(tp, (unsigned long)f);
 		list_del_rcu(&f->list);
 		call_rcu(&f->rcu, fl_destroy_filter);
 	}
@@ -554,13 +554,13 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 			     &mask.key,
 			     &fnew->key,
 			     &fnew->exts,
-			     (u64)fnew,
+			     (unsigned long)fnew,
 			     flags);
 
 	if (fold) {
 		rhashtable_remove_fast(&head->ht, &fold->ht_node,
 				       head->ht_params);
-		fl_hw_destroy_filter(tp, (u64)fold);
+		fl_hw_destroy_filter(tp, (unsigned long)fold);
 	}
 
 	*arg = (unsigned long) fnew;
@@ -588,7 +588,7 @@ static int fl_delete(struct tcf_proto *tp, unsigned long arg)
 	rhashtable_remove_fast(&head->ht, &f->ht_node,
 			       head->ht_params);
 	list_del_rcu(&f->list);
-	fl_hw_destroy_filter(tp, (u64)f);
+	fl_hw_destroy_filter(tp, (unsigned long)f);
 	tcf_unbind_filter(tp, &f->res);
 	call_rcu(&f->rcu, fl_destroy_filter);
 	return 0;
-- 
cgit v0.10.2


From 8e2ad4113ce4671686740f808ff2795395c39eef Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Tue, 8 Mar 2016 15:18:54 -0500
Subject: macvtap: always pass ethernet header in linear

The stack expects link layer headers in the skb linear section.
Macvtap can create skbs with llheader in frags in edge cases:
when (IFF_VNET_HDR is off or vnet_hdr.hdr_len < ETH_HLEN) and
prepad + len > PAGE_SIZE and vnet_hdr.flags has no or bad csum.

Add checks to ensure linear is always at least ETH_HLEN.
At this point, len is already ensured to be >= ETH_HLEN.

For backwards compatiblity, rounds up short vnet_hdr.hdr_len.
This differs from tap and packet, which return an error.

Fixes b9fb9ee07e67 ("macvtap: add GSO/csum offload support")
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index d636d05..95394ed 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -760,6 +760,8 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m,
 			macvtap16_to_cpu(q, vnet_hdr.hdr_len) : GOODCOPY_LEN;
 		if (copylen > good_linear)
 			copylen = good_linear;
+		else if (copylen < ETH_HLEN)
+			copylen = ETH_HLEN;
 		linear = copylen;
 		i = *from;
 		iov_iter_advance(&i, copylen);
@@ -769,10 +771,11 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m,
 
 	if (!zerocopy) {
 		copylen = len;
-		if (macvtap16_to_cpu(q, vnet_hdr.hdr_len) > good_linear)
+		linear = macvtap16_to_cpu(q, vnet_hdr.hdr_len);
+		if (linear > good_linear)
 			linear = good_linear;
-		else
-			linear = macvtap16_to_cpu(q, vnet_hdr.hdr_len);
+		else if (linear < ETH_HLEN)
+			linear = ETH_HLEN;
 	}
 
 	skb = macvtap_alloc_skb(&q->sk, MACVTAP_RESERVE, copylen,
-- 
cgit v0.10.2


From 869f63a4d28144c03c8f4a4c0d1e8f31f8c11a10 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 8 Mar 2016 12:59:33 -0800
Subject: mlxsw: spectrum: Check requested ageing time is valid

Commit c62987bbd8a1 ("bridge: push bridge setting ageing_time down to
switchdev") added a check for minimum and maximum ageing time, but this
breaks existing behaviour where one can set ageing time to 0 for a
non-learning bridge.

Push this check down to the driver and allow the check in the bridge
layer to be removed. Currently ageing time 0 is refused by the driver,
but we can later add support for this functionality.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 1b691d7..4b8abaf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -123,6 +123,8 @@ struct mlxsw_sp {
 #define MLXSW_SP_DEFAULT_LEARNING_INTERVAL 100
 		unsigned int interval; /* ms */
 	} fdb_notify;
+#define MLXSW_SP_MIN_AGEING_TIME 10
+#define MLXSW_SP_MAX_AGEING_TIME 1000000
 #define MLXSW_SP_DEFAULT_AGEING_TIME 300
 	u32 ageing_time;
 	struct mlxsw_sp_upper master_bridge;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index 7b56098..e1c74ef 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -311,8 +311,13 @@ static int mlxsw_sp_port_attr_br_ageing_set(struct mlxsw_sp_port *mlxsw_sp_port,
 	unsigned long ageing_jiffies = clock_t_to_jiffies(ageing_clock_t);
 	u32 ageing_time = jiffies_to_msecs(ageing_jiffies) / 1000;
 
-	if (switchdev_trans_ph_prepare(trans))
-		return 0;
+	if (switchdev_trans_ph_prepare(trans)) {
+		if (ageing_time < MLXSW_SP_MIN_AGEING_TIME ||
+		    ageing_time > MLXSW_SP_MAX_AGEING_TIME)
+			return -ERANGE;
+		else
+			return 0;
+	}
 
 	return mlxsw_sp_ageing_set(mlxsw_sp, ageing_time);
 }
-- 
cgit v0.10.2


From 88de1cd457e5cb664d6d437e2ea4750d089165f5 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@mellanox.com>
Date: Tue, 8 Mar 2016 12:59:34 -0800
Subject: rocker: set FDB cleanup timer according to lowest ageing time

In rocker, ageing time is a per-port attribute, so the next time the FDB
cleanup timer fires should be set according to the lowest ageing time.

This will later allow us to delete the BR_MIN_AGEING_TIME macro, which was
added to guarantee minimum ageing time in the bridge layer, thereby breaking
existing behavior.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
index 1ab995f..011f1b6 100644
--- a/drivers/net/ethernet/rocker/rocker.h
+++ b/drivers/net/ethernet/rocker/rocker.h
@@ -63,6 +63,7 @@ struct rocker {
 	struct {
 		u64 id;
 	} hw;
+	unsigned long ageing_time;
 	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
 	struct rocker_dma_ring_info cmd_ring;
 	struct rocker_dma_ring_info event_ring;
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 28b775e..c8b1bf4 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -2733,6 +2733,8 @@ static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	rocker->hw.id = rocker_read64(rocker, SWITCH_ID);
 
+	rocker->ageing_time = BR_DEFAULT_AGEING_TIME;
+
 	err = rocker_probe_ports(rocker);
 	if (err) {
 		dev_err(&pdev->dev, "failed to probe ports\n");
diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
index 07218c3..cfde525 100644
--- a/drivers/net/ethernet/rocker/rocker_ofdpa.c
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -2105,7 +2105,7 @@ static void ofdpa_fdb_cleanup(unsigned long data)
 	struct ofdpa_port *ofdpa_port;
 	struct ofdpa_fdb_tbl_entry *entry;
 	struct hlist_node *tmp;
-	unsigned long next_timer = jiffies + BR_MIN_AGEING_TIME;
+	unsigned long next_timer = jiffies + ofdpa->rocker->ageing_time;
 	unsigned long expires;
 	unsigned long lock_flags;
 	int flags = OFDPA_OP_FLAG_NOWAIT | OFDPA_OP_FLAG_REMOVE |
@@ -2648,9 +2648,12 @@ ofdpa_port_attr_bridge_ageing_time_set(struct rocker_port *rocker_port,
 				       struct switchdev_trans *trans)
 {
 	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
+	struct rocker *rocker = rocker_port->rocker;
 
 	if (!switchdev_trans_ph_prepare(trans)) {
 		ofdpa_port->ageing_time = clock_t_to_jiffies(ageing_time);
+		if (ofdpa_port->ageing_time < rocker->ageing_time)
+			rocker->ageing_time = ofdpa_port->ageing_time;
 		mod_timer(&ofdpa_port->ofdpa->fdb_cleanup_timer, jiffies);
 	}
 
-- 
cgit v0.10.2


From 4c656c13b254d598e83e586b7b4d36a2043dad85 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemming@brocade.com>
Date: Tue, 8 Mar 2016 12:59:35 -0800
Subject: bridge: allow zero ageing time

This fixes a regression in the bridge ageing time caused by:
commit c62987bbd8a1 ("bridge: push bridge setting ageing_time down to switchdev")

There are users of Linux bridge which use the feature that if ageing time
is set to 0 it causes entries to never expire. See:
  https://www.linuxfoundation.org/collaborate/workgroups/networking/bridge

For a pure software bridge, it is unnecessary for the code to have
arbitrary restrictions on what values are allowable.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h
index a338a68..dcb89e3 100644
--- a/include/linux/if_bridge.h
+++ b/include/linux/if_bridge.h
@@ -46,10 +46,6 @@ struct br_ip_list {
 #define BR_LEARNING_SYNC	BIT(9)
 #define BR_PROXYARP_WIFI	BIT(10)
 
-/* values as per ieee8021QBridgeFdbAgingTime */
-#define BR_MIN_AGEING_TIME	(10 * HZ)
-#define BR_MAX_AGEING_TIME	(1000000 * HZ)
-
 #define BR_DEFAULT_AGEING_TIME	(300 * HZ)
 
 extern void brioctl_set(int (*ioctl_hook)(struct net *, unsigned int, void __user *));
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index c22816a..e234490 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -562,6 +562,14 @@ int br_set_max_age(struct net_bridge *br, unsigned long val)
 
 }
 
+/* Set time interval that dynamic forwarding entries live
+ * For pure software bridge, allow values outside the 802.1
+ * standard specification for special cases:
+ *  0 - entry never ages (all permanant)
+ *  1 - entry disappears (no persistance)
+ *
+ * Offloaded switch entries maybe more restrictive
+ */
 int br_set_ageing_time(struct net_bridge *br, u32 ageing_time)
 {
 	struct switchdev_attr attr = {
@@ -573,9 +581,6 @@ int br_set_ageing_time(struct net_bridge *br, u32 ageing_time)
 	unsigned long t = clock_t_to_jiffies(ageing_time);
 	int err;
 
-	if (t < BR_MIN_AGEING_TIME || t > BR_MAX_AGEING_TIME)
-		return -ERANGE;
-
 	err = switchdev_port_attr_set(br->dev, &attr);
 	if (err)
 		return err;
-- 
cgit v0.10.2


From e327f4e193b6a176f4dc3b19faf6371a8917588d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 8 Mar 2016 13:54:56 -0800
Subject: cisco: enic: Update logging macros and uses

Don't hide varibles used by the logging macros.

Miscellanea:

o Use the more common ##__VA_ARGS__ extension
o Add missing newlines to formats
o Realign arguments

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cisco/enic/enic.h b/drivers/net/ethernet/cisco/enic/enic.h
index 7ba6d53..130f910 100644
--- a/drivers/net/ethernet/cisco/enic/enic.h
+++ b/drivers/net/ethernet/cisco/enic/enic.h
@@ -201,16 +201,20 @@ static inline struct net_device *vnic_get_netdev(struct vnic_dev *vdev)
 }
 
 /* wrappers function for kernel log
- * Make sure variable vdev of struct vnic_dev is available in the block where
- * these macros are used
  */
-#define vdev_info(args...)	dev_info(&vdev->pdev->dev, args)
-#define vdev_warn(args...)	dev_warn(&vdev->pdev->dev, args)
-#define vdev_err(args...)	dev_err(&vdev->pdev->dev, args)
-
-#define vdev_netinfo(args...)	netdev_info(vnic_get_netdev(vdev), args)
-#define vdev_netwarn(args...)	netdev_warn(vnic_get_netdev(vdev), args)
-#define vdev_neterr(args...)	netdev_err(vnic_get_netdev(vdev), args)
+#define vdev_err(vdev, fmt, ...)					\
+	dev_err(&(vdev)->pdev->dev, fmt, ##__VA_ARGS__)
+#define vdev_warn(vdev, fmt, ...)					\
+	dev_warn(&(vdev)->pdev->dev, fmt, ##__VA_ARGS__)
+#define vdev_info(vdev, fmt, ...)					\
+	dev_info(&(vdev)->pdev->dev, fmt, ##__VA_ARGS__)
+
+#define vdev_neterr(vdev, fmt, ...)					\
+	netdev_err(vnic_get_netdev(vdev), fmt, ##__VA_ARGS__)
+#define vdev_netwarn(vdev, fmt, ...)					\
+	netdev_warn(vnic_get_netdev(vdev), fmt, ##__VA_ARGS__)
+#define vdev_netinfo(vdev, fmt, ...)					\
+	netdev_info(vnic_get_netdev(vdev), fmt, ##__VA_ARGS__)
 
 static inline struct device *enic_get_dev(struct enic *enic)
 {
diff --git a/drivers/net/ethernet/cisco/enic/vnic_cq.c b/drivers/net/ethernet/cisco/enic/vnic_cq.c
index abeda2a..9c682af 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_cq.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_cq.c
@@ -43,7 +43,7 @@ int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index,
 
 	cq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_CQ, index);
 	if (!cq->ctrl) {
-		vdev_err("Failed to hook CQ[%d] resource\n", index);
+		vdev_err(vdev, "Failed to hook CQ[%d] resource\n", index);
 		return -EINVAL;
 	}
 
diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.c b/drivers/net/ethernet/cisco/enic/vnic_dev.c
index 1fdf5fe..8f27df3 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_dev.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_dev.c
@@ -53,14 +53,14 @@ static int vnic_dev_discover_res(struct vnic_dev *vdev,
 		return -EINVAL;
 
 	if (bar->len < VNIC_MAX_RES_HDR_SIZE) {
-		vdev_err("vNIC BAR0 res hdr length error\n");
+		vdev_err(vdev, "vNIC BAR0 res hdr length error\n");
 		return -EINVAL;
 	}
 
 	rh  = bar->vaddr;
 	mrh = bar->vaddr;
 	if (!rh) {
-		vdev_err("vNIC BAR0 res hdr not mem-mapped\n");
+		vdev_err(vdev, "vNIC BAR0 res hdr not mem-mapped\n");
 		return -EINVAL;
 	}
 
@@ -69,7 +69,7 @@ static int vnic_dev_discover_res(struct vnic_dev *vdev,
 		(ioread32(&rh->version) != VNIC_RES_VERSION)) {
 		if ((ioread32(&mrh->magic) != MGMTVNIC_MAGIC) ||
 			(ioread32(&mrh->version) != MGMTVNIC_VERSION)) {
-			vdev_err("vNIC BAR0 res magic/version error exp (%lx/%lx) or (%lx/%lx), curr (%x/%x)\n",
+			vdev_err(vdev, "vNIC BAR0 res magic/version error exp (%lx/%lx) or (%lx/%lx), curr (%x/%x)\n",
 				 VNIC_RES_MAGIC, VNIC_RES_VERSION,
 				 MGMTVNIC_MAGIC, MGMTVNIC_VERSION,
 				 ioread32(&rh->magic), ioread32(&rh->version));
@@ -106,7 +106,7 @@ static int vnic_dev_discover_res(struct vnic_dev *vdev,
 			/* each count is stride bytes long */
 			len = count * VNIC_RES_STRIDE;
 			if (len + bar_offset > bar[bar_num].len) {
-				vdev_err("vNIC BAR0 resource %d out-of-bounds, offset 0x%x + size 0x%x > bar len 0x%lx\n",
+				vdev_err(vdev, "vNIC BAR0 resource %d out-of-bounds, offset 0x%x + size 0x%x > bar len 0x%lx\n",
 					 type, bar_offset, len,
 					 bar[bar_num].len);
 				return -EINVAL;
@@ -198,7 +198,7 @@ int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring,
 		&ring->base_addr_unaligned);
 
 	if (!ring->descs_unaligned) {
-		vdev_err("Failed to allocate ring (size=%d), aborting\n",
+		vdev_err(vdev, "Failed to allocate ring (size=%d), aborting\n",
 			 (int)ring->size);
 		return -ENOMEM;
 	}
@@ -241,7 +241,7 @@ static int _vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 		return -ENODEV;
 	}
 	if (status & STAT_BUSY) {
-		vdev_neterr("Busy devcmd %d\n", _CMD_N(cmd));
+		vdev_neterr(vdev, "Busy devcmd %d\n", _CMD_N(cmd));
 		return -EBUSY;
 	}
 
@@ -275,7 +275,7 @@ static int _vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 					return -err;
 				if (err != ERR_ECMDUNKNOWN ||
 				    cmd != CMD_CAPABILITY)
-					vdev_neterr("Error %d devcmd %d\n",
+					vdev_neterr(vdev, "Error %d devcmd %d\n",
 						    err, _CMD_N(cmd));
 				return -err;
 			}
@@ -290,7 +290,7 @@ static int _vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 		}
 	}
 
-	vdev_neterr("Timedout devcmd %d\n", _CMD_N(cmd));
+	vdev_neterr(vdev, "Timedout devcmd %d\n", _CMD_N(cmd));
 	return -ETIMEDOUT;
 }
 
@@ -313,7 +313,7 @@ static int _vnic_dev_cmd2(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 	new_posted = (posted + 1) % DEVCMD2_RING_SIZE;
 
 	if (new_posted == fetch_index) {
-		vdev_neterr("devcmd2 %d: wq is full. fetch index: %u, posted index: %u\n",
+		vdev_neterr(vdev, "devcmd2 %d: wq is full. fetch index: %u, posted index: %u\n",
 			    _CMD_N(cmd), fetch_index, posted);
 		return -EBUSY;
 	}
@@ -352,7 +352,7 @@ static int _vnic_dev_cmd2(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 				err = result->error;
 				if (err != ERR_ECMDUNKNOWN ||
 				    cmd != CMD_CAPABILITY)
-					vdev_neterr("Error %d devcmd %d\n",
+					vdev_neterr(vdev, "Error %d devcmd %d\n",
 						    err, _CMD_N(cmd));
 				return -err;
 			}
@@ -365,7 +365,7 @@ static int _vnic_dev_cmd2(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 		udelay(100);
 	}
 
-	vdev_neterr("devcmd %d timed out\n", _CMD_N(cmd));
+	vdev_neterr(vdev, "devcmd %d timed out\n", _CMD_N(cmd));
 
 	return -ETIMEDOUT;
 }
@@ -401,7 +401,7 @@ static int vnic_dev_init_devcmd2(struct vnic_dev *vdev)
 
 	fetch_index = ioread32(&vdev->devcmd2->wq.ctrl->fetch_index);
 	if (fetch_index == 0xFFFFFFFF) { /* check for hardware gone  */
-		vdev_err("Fatal error in devcmd2 init - hardware surprise removal");
+		vdev_err(vdev, "Fatal error in devcmd2 init - hardware surprise removal\n");
 
 		return -ENODEV;
 	}
@@ -474,8 +474,8 @@ static int vnic_dev_cmd_proxy(struct vnic_dev *vdev,
 		err = (int)vdev->args[1];
 		if (err != ERR_ECMDUNKNOWN ||
 		    cmd != CMD_CAPABILITY)
-			vdev_neterr("Error %d proxy devcmd %d\n", err,
-				    _CMD_N(cmd));
+			vdev_neterr(vdev, "Error %d proxy devcmd %d\n",
+				    err, _CMD_N(cmd));
 		return err;
 	}
 
@@ -768,7 +768,7 @@ int vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast,
 
 	err = vnic_dev_cmd(vdev, CMD_PACKET_FILTER, &a0, &a1, wait);
 	if (err)
-		vdev_neterr("Can't set packet filter\n");
+		vdev_neterr(vdev, "Can't set packet filter\n");
 
 	return err;
 }
@@ -785,7 +785,7 @@ int vnic_dev_add_addr(struct vnic_dev *vdev, const u8 *addr)
 
 	err = vnic_dev_cmd(vdev, CMD_ADDR_ADD, &a0, &a1, wait);
 	if (err)
-		vdev_neterr("Can't add addr [%pM], %d\n", addr, err);
+		vdev_neterr(vdev, "Can't add addr [%pM], %d\n", addr, err);
 
 	return err;
 }
@@ -802,7 +802,7 @@ int vnic_dev_del_addr(struct vnic_dev *vdev, const u8 *addr)
 
 	err = vnic_dev_cmd(vdev, CMD_ADDR_DEL, &a0, &a1, wait);
 	if (err)
-		vdev_neterr("Can't del addr [%pM], %d\n", addr, err);
+		vdev_neterr(vdev, "Can't del addr [%pM], %d\n", addr, err);
 
 	return err;
 }
@@ -846,7 +846,8 @@ int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr)
 	dma_addr_t notify_pa;
 
 	if (vdev->notify || vdev->notify_pa) {
-		vdev_neterr("notify block %p still allocated", vdev->notify);
+		vdev_neterr(vdev, "notify block %p still allocated\n",
+			    vdev->notify);
 		return -EINVAL;
 	}
 
@@ -965,7 +966,7 @@ int vnic_dev_intr_coal_timer_info(struct vnic_dev *vdev)
 	 */
 	if ((err == ERR_ECMDUNKNOWN) ||
 		(!err && !(vdev->args[0] && vdev->args[1] && vdev->args[2]))) {
-		vdev_netwarn("Using default conversion factor for interrupt coalesce timer\n");
+		vdev_netwarn(vdev, "Using default conversion factor for interrupt coalesce timer\n");
 		vnic_dev_intr_coal_timer_info_default(vdev);
 		return 0;
 	}
@@ -1103,16 +1104,16 @@ int vnic_devcmd_init(struct vnic_dev *vdev)
 	if (res) {
 		err = vnic_dev_init_devcmd2(vdev);
 		if (err)
-			vdev_warn("DEVCMD2 init failed: %d, Using DEVCMD1",
+			vdev_warn(vdev, "DEVCMD2 init failed: %d, Using DEVCMD1\n",
 				  err);
 		else
 			return 0;
 	} else {
-		vdev_warn("DEVCMD2 resource not found (old firmware?) Using DEVCMD1\n");
+		vdev_warn(vdev, "DEVCMD2 resource not found (old firmware?) Using DEVCMD1\n");
 	}
 	err = vnic_dev_init_devcmd1(vdev);
 	if (err)
-		vdev_err("DEVCMD1 initialization failed: %d", err);
+		vdev_err(vdev, "DEVCMD1 initialization failed: %d\n", err);
 
 	return err;
 }
diff --git a/drivers/net/ethernet/cisco/enic/vnic_intr.c b/drivers/net/ethernet/cisco/enic/vnic_intr.c
index 942759d..23604e3 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_intr.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_intr.c
@@ -40,7 +40,8 @@ int vnic_intr_alloc(struct vnic_dev *vdev, struct vnic_intr *intr,
 
 	intr->ctrl = vnic_dev_get_res(vdev, RES_TYPE_INTR_CTRL, index);
 	if (!intr->ctrl) {
-		vdev_err("Failed to hook INTR[%d].ctrl resource\n", index);
+		vdev_err(vdev, "Failed to hook INTR[%d].ctrl resource\n",
+			 index);
 		return -EINVAL;
 	}
 
diff --git a/drivers/net/ethernet/cisco/enic/vnic_rq.c b/drivers/net/ethernet/cisco/enic/vnic_rq.c
index cce2777..e572a52 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_rq.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_rq.c
@@ -92,7 +92,7 @@ int vnic_rq_alloc(struct vnic_dev *vdev, struct vnic_rq *rq, unsigned int index,
 
 	rq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_RQ, index);
 	if (!rq->ctrl) {
-		vdev_err("Failed to hook RQ[%d] resource\n", index);
+		vdev_err(vdev, "Failed to hook RQ[%d] resource\n", index);
 		return -EINVAL;
 	}
 
@@ -179,7 +179,7 @@ int vnic_rq_disable(struct vnic_rq *rq)
 		udelay(10);
 	}
 
-	vdev_neterr("Failed to disable RQ[%d]\n", rq->index);
+	vdev_neterr(vdev, "Failed to disable RQ[%d]\n", rq->index);
 
 	return -ETIMEDOUT;
 }
diff --git a/drivers/net/ethernet/cisco/enic/vnic_wq.c b/drivers/net/ethernet/cisco/enic/vnic_wq.c
index 05ad16a..090cc65 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_wq.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_wq.c
@@ -95,7 +95,7 @@ int vnic_wq_alloc(struct vnic_dev *vdev, struct vnic_wq *wq, unsigned int index,
 
 	wq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_WQ, index);
 	if (!wq->ctrl) {
-		vdev_err("Failed to hook WQ[%d] resource\n", index);
+		vdev_err(vdev, "Failed to hook WQ[%d] resource\n", index);
 		return -EINVAL;
 	}
 
@@ -187,7 +187,7 @@ int vnic_wq_disable(struct vnic_wq *wq)
 		udelay(10);
 	}
 
-	vdev_neterr("Failed to disable WQ[%d]\n", wq->index);
+	vdev_neterr(vdev, "Failed to disable WQ[%d]\n", wq->index);
 
 	return -ETIMEDOUT;
 }
-- 
cgit v0.10.2


From 134611446dc657e1bbc73ca0e4e6b599df687db0 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 9 Mar 2016 03:00:02 +0100
Subject: ip_tunnel: add support for setting flow label via collect metadata

This patch extends udp_tunnel6_xmit_skb() to pass in the IPv6 flow label
from call sites. Currently, there's no such option and it's always set to
zero when writing ip6_flow_hdr(). Add a label member to ip_tunnel_key, so
that flow-based tunnels via collect metadata frontends can make use of it.
vxlan and geneve will be converted to add flow label support separately.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 6a0cbbe..89ccff7 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -1054,7 +1054,7 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 		ttl = ttl ? : ip6_dst_hoplimit(dst);
 	}
 	udp_tunnel6_xmit_skb(dst, gs6->sock->sk, skb, dev,
-			     &fl6.saddr, &fl6.daddr, prio, ttl,
+			     &fl6.saddr, &fl6.daddr, prio, ttl, 0,
 			     sport, geneve->dst_port,
 			     !!(flags & GENEVE_F_UDP_ZERO_CSUM6_TX));
 	return NETDEV_TX_OK;
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 2399099..8bdcd5e 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -2066,7 +2066,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			return;
 		}
 		udp_tunnel6_xmit_skb(ndst, sk, skb, dev,
-				     &saddr, &dst->sin6.sin6_addr, tos, ttl,
+				     &saddr, &dst->sin6.sin6_addr, tos, ttl, 0,
 				     src_port, dst_port, !udp_sum);
 #endif
 	}
diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h
index 84b833a..5db9f59 100644
--- a/include/net/dst_metadata.h
+++ b/include/net/dst_metadata.h
@@ -126,7 +126,7 @@ static inline struct metadata_dst *ip_tun_rx_dst(struct sk_buff *skb,
 
 	ip_tunnel_key_init(&tun_dst->u.tun_info.key,
 			   iph->saddr, iph->daddr, iph->tos, iph->ttl,
-			   0, 0, tunnel_id, flags);
+			   0, 0, 0, tunnel_id, flags);
 	return tun_dst;
 }
 
@@ -152,8 +152,11 @@ static inline struct metadata_dst *ipv6_tun_rx_dst(struct sk_buff *skb,
 
 	info->key.u.ipv6.src = ip6h->saddr;
 	info->key.u.ipv6.dst = ip6h->daddr;
+
 	info->key.tos = ipv6_get_dsfield(ip6h);
 	info->key.ttl = ip6h->hop_limit;
+	info->key.label = ip6_flowlabel(ip6h);
+
 	return tun_dst;
 }
 
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 0acd80f..5dc2e45 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -48,6 +48,7 @@ struct ip_tunnel_key {
 	__be16			tun_flags;
 	u8			tos;		/* TOS for IPv4, TC for IPv6 */
 	u8			ttl;		/* TTL for IPv4, HL for IPv6 */
+	__be32			label;		/* Flow Label for IPv6 */
 	__be16			tp_src;
 	__be16			tp_dst;
 };
@@ -181,7 +182,7 @@ int ip_tunnel_encap_del_ops(const struct ip_tunnel_encap_ops *op,
 
 static inline void ip_tunnel_key_init(struct ip_tunnel_key *key,
 				      __be32 saddr, __be32 daddr,
-				      u8 tos, u8 ttl,
+				      u8 tos, u8 ttl, __be32 label,
 				      __be16 tp_src, __be16 tp_dst,
 				      __be64 tun_id, __be16 tun_flags)
 {
@@ -192,6 +193,7 @@ static inline void ip_tunnel_key_init(struct ip_tunnel_key *key,
 	       0, IP_TUNNEL_KEY_IPV4_PAD_LEN);
 	key->tos = tos;
 	key->ttl = ttl;
+	key->label = label;
 	key->tun_flags = tun_flags;
 
 	/* For the tunnel types on the top of IPsec, the tp_src and tp_dst of
diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h
index 97f5adb..b831140 100644
--- a/include/net/udp_tunnel.h
+++ b/include/net/udp_tunnel.h
@@ -88,8 +88,8 @@ int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk,
 			 struct sk_buff *skb,
 			 struct net_device *dev, struct in6_addr *saddr,
 			 struct in6_addr *daddr,
-			 __u8 prio, __u8 ttl, __be16 src_port,
-			 __be16 dst_port, bool nocheck);
+			 __u8 prio, __u8 ttl, __be32 label,
+			 __be16 src_port, __be16 dst_port, bool nocheck);
 #endif
 
 void udp_tunnel_sock_release(struct socket *sock);
diff --git a/net/ipv6/ip6_udp_tunnel.c b/net/ipv6/ip6_udp_tunnel.c
index 14dacf1..a752052 100644
--- a/net/ipv6/ip6_udp_tunnel.c
+++ b/net/ipv6/ip6_udp_tunnel.c
@@ -73,8 +73,8 @@ int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk,
 			 struct sk_buff *skb,
 			 struct net_device *dev, struct in6_addr *saddr,
 			 struct in6_addr *daddr,
-			 __u8 prio, __u8 ttl, __be16 src_port,
-			 __be16 dst_port, bool nocheck)
+			 __u8 prio, __u8 ttl, __be32 label,
+			 __be16 src_port, __be16 dst_port, bool nocheck)
 {
 	struct udphdr *uh;
 	struct ipv6hdr *ip6h;
@@ -98,7 +98,7 @@ int udp_tunnel6_xmit_skb(struct dst_entry *dst, struct sock *sk,
 	__skb_push(skb, sizeof(*ip6h));
 	skb_reset_network_header(skb);
 	ip6h		  = ipv6_hdr(skb);
-	ip6_flow_hdr(ip6h, prio, htonl(0));
+	ip6_flow_hdr(ip6h, prio, label);
 	ip6h->payload_len = htons(skb->len);
 	ip6h->nexthdr     = IPPROTO_UDP;
 	ip6h->hop_limit   = ttl;
diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index 49b3c2e..c94f9a1 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -196,7 +196,7 @@ static int tipc_udp_send_msg(struct net *net, struct sk_buff *skb,
 		ttl = ip6_dst_hoplimit(ndst);
 		err = udp_tunnel6_xmit_skb(ndst, ub->ubsock->sk, skb,
 					   ndst->dev, &src->ipv6,
-					   &dst->ipv6, 0, ttl, src->udp_port,
+					   &dst->ipv6, 0, ttl, 0, src->udp_port,
 					   dst->udp_port, false);
 #endif
 	}
-- 
cgit v0.10.2


From e7f70af111f086a20800ad2e17f544b2e3e0f375 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 9 Mar 2016 03:00:03 +0100
Subject: vxlan: support setting IPv6 flow label

This work adds support for setting the IPv6 flow label for vxlan per
device and through collect metadata (ip_tunnel_key) frontends. The
vxlan dst cache does not need any special considerations here, for
the cases where caches can be used, the label is static per cache.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 8bdcd5e..8eda76f 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1790,6 +1790,7 @@ static struct rtable *vxlan_get_route(struct vxlan_dev *vxlan,
 #if IS_ENABLED(CONFIG_IPV6)
 static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 					  struct sk_buff *skb, int oif, u8 tos,
+					  __be32 label,
 					  const struct in6_addr *daddr,
 					  struct in6_addr *saddr,
 					  struct dst_cache *dst_cache,
@@ -1813,6 +1814,7 @@ static struct dst_entry *vxlan6_get_route(struct vxlan_dev *vxlan,
 	fl6.flowi6_tos = RT_TOS(tos);
 	fl6.daddr = *daddr;
 	fl6.saddr = vxlan->cfg.saddr.sin6.sin6_addr;
+	fl6.flowlabel = label;
 	fl6.flowi6_mark = skb->mark;
 	fl6.flowi6_proto = IPPROTO_UDP;
 
@@ -1888,7 +1890,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	struct vxlan_metadata _md;
 	struct vxlan_metadata *md = &_md;
 	__be16 src_port = 0, dst_port;
-	__be32 vni;
+	__be32 vni, label;
 	__be16 df = 0;
 	__u8 tos, ttl;
 	int err;
@@ -1939,12 +1941,14 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 	if (tos == 1)
 		tos = ip_tunnel_get_dsfield(old_iph, skb);
 
+	label = vxlan->cfg.label;
 	src_port = udp_flow_src_port(dev_net(dev), skb, vxlan->cfg.port_min,
 				     vxlan->cfg.port_max, true);
 
 	if (info) {
 		ttl = info->key.ttl;
 		tos = info->key.tos;
+		label = info->key.label;
 		udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM);
 
 		if (info->options_len)
@@ -2020,7 +2024,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 
 		ndst = vxlan6_get_route(vxlan, skb,
 					rdst ? rdst->remote_ifindex : 0, tos,
-					&dst->sin6.sin6_addr, &saddr,
+					label, &dst->sin6.sin6_addr, &saddr,
 					dst_cache, info);
 		if (IS_ERR(ndst)) {
 			netdev_dbg(dev, "no route to %pI6\n",
@@ -2066,8 +2070,8 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev,
 			return;
 		}
 		udp_tunnel6_xmit_skb(ndst, sk, skb, dev,
-				     &saddr, &dst->sin6.sin6_addr, tos, ttl, 0,
-				     src_port, dst_port, !udp_sum);
+				     &saddr, &dst->sin6.sin6_addr, tos, ttl,
+				     label, src_port, dst_port, !udp_sum);
 #endif
 	}
 
@@ -2390,7 +2394,7 @@ static int vxlan_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 		if (!vxlan->vn6_sock)
 			return -EINVAL;
 		ndst = vxlan6_get_route(vxlan, skb, 0, info->key.tos,
-					&info->key.u.ipv6.dst,
+					info->key.label, &info->key.u.ipv6.dst,
 					&info->key.u.ipv6.src, NULL, info);
 		if (IS_ERR(ndst))
 			return PTR_ERR(ndst);
@@ -2505,6 +2509,7 @@ static const struct nla_policy vxlan_policy[IFLA_VXLAN_MAX + 1] = {
 	[IFLA_VXLAN_LOCAL6]	= { .len = sizeof(struct in6_addr) },
 	[IFLA_VXLAN_TOS]	= { .type = NLA_U8 },
 	[IFLA_VXLAN_TTL]	= { .type = NLA_U8 },
+	[IFLA_VXLAN_LABEL]	= { .type = NLA_U32 },
 	[IFLA_VXLAN_LEARNING]	= { .type = NLA_U8 },
 	[IFLA_VXLAN_AGEING]	= { .type = NLA_U32 },
 	[IFLA_VXLAN_LIMIT]	= { .type = NLA_U32 },
@@ -2739,6 +2744,11 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev,
 		vxlan->flags |= VXLAN_F_IPV6;
 	}
 
+	if (conf->label && !use_ipv6) {
+		pr_info("label only supported in use with IPv6\n");
+		return -EINVAL;
+	}
+
 	if (conf->remote_ifindex) {
 		lowerdev = __dev_get_by_index(src_net, conf->remote_ifindex);
 		dst->remote_ifindex = conf->remote_ifindex;
@@ -2887,6 +2897,10 @@ static int vxlan_newlink(struct net *src_net, struct net_device *dev,
 	if (data[IFLA_VXLAN_TTL])
 		conf.ttl = nla_get_u8(data[IFLA_VXLAN_TTL]);
 
+	if (data[IFLA_VXLAN_LABEL])
+		conf.label = nla_get_be32(data[IFLA_VXLAN_LABEL]) &
+			     IPV6_FLOWLABEL_MASK;
+
 	if (!data[IFLA_VXLAN_LEARNING] || nla_get_u8(data[IFLA_VXLAN_LEARNING]))
 		conf.flags |= VXLAN_F_LEARN;
 
@@ -2990,6 +3004,7 @@ static size_t vxlan_get_size(const struct net_device *dev)
 		nla_total_size(sizeof(struct in6_addr)) + /* IFLA_VXLAN_LOCAL{6} */
 		nla_total_size(sizeof(__u8)) +	/* IFLA_VXLAN_TTL */
 		nla_total_size(sizeof(__u8)) +	/* IFLA_VXLAN_TOS */
+		nla_total_size(sizeof(__be32)) + /* IFLA_VXLAN_LABEL */
 		nla_total_size(sizeof(__u8)) +	/* IFLA_VXLAN_LEARNING */
 		nla_total_size(sizeof(__u8)) +	/* IFLA_VXLAN_PROXY */
 		nla_total_size(sizeof(__u8)) +	/* IFLA_VXLAN_RSC */
@@ -3053,6 +3068,7 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
 
 	if (nla_put_u8(skb, IFLA_VXLAN_TTL, vxlan->cfg.ttl) ||
 	    nla_put_u8(skb, IFLA_VXLAN_TOS, vxlan->cfg.tos) ||
+	    nla_put_be32(skb, IFLA_VXLAN_LABEL, vxlan->cfg.label) ||
 	    nla_put_u8(skb, IFLA_VXLAN_LEARNING,
 			!!(vxlan->flags & VXLAN_F_LEARN)) ||
 	    nla_put_u8(skb, IFLA_VXLAN_PROXY,
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 6eda4ed..a763c96 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -162,6 +162,7 @@ struct vxlan_config {
 	u16			port_max;
 	u8			tos;
 	u8			ttl;
+	__be32			label;
 	u32			flags;
 	unsigned long		age_interval;
 	unsigned int		addrmax;
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index d452cea..6bebc97 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -456,6 +456,7 @@ enum {
 	IFLA_VXLAN_GBP,
 	IFLA_VXLAN_REMCSUM_NOPARTIAL,
 	IFLA_VXLAN_COLLECT_METADATA,
+	IFLA_VXLAN_LABEL,
 	__IFLA_VXLAN_MAX
 };
 #define IFLA_VXLAN_MAX	(__IFLA_VXLAN_MAX - 1)
-- 
cgit v0.10.2


From 8eb3b99554b82da968d1fbc00df9f3156c5e2d63 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 9 Mar 2016 03:00:04 +0100
Subject: geneve: support setting IPv6 flow label

This work adds support for setting the IPv6 flow label for geneve per
device and through collect metadata (ip_tunnel_key) frontends. Also here,
the geneve dst cache does not need any special considerations, for the
cases where caches can be used, the label is static per cache.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 89ccff7..33185b9 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -68,6 +68,7 @@ struct geneve_dev {
 	u8                 tos;		/* TOS override */
 	union geneve_addr  remote;	/* IP address for link partner */
 	struct list_head   next;	/* geneve's per namespace list */
+	__be32		   label;	/* IPv6 flowlabel override */
 	__be16		   dst_port;
 	bool		   collect_md;
 	struct gro_cells   gro_cells;
@@ -846,6 +847,7 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 		fl6->daddr = info->key.u.ipv6.dst;
 		fl6->saddr = info->key.u.ipv6.src;
 		fl6->flowi6_tos = RT_TOS(info->key.tos);
+		fl6->flowlabel = info->key.label;
 		dst_cache = &info->dst_cache;
 	} else {
 		prio = geneve->tos;
@@ -857,6 +859,7 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb,
 		}
 
 		fl6->flowi6_tos = RT_TOS(prio);
+		fl6->flowlabel = geneve->label;
 		fl6->daddr = geneve->remote.sin6.sin6_addr;
 		dst_cache = &geneve->dst_cache;
 	}
@@ -998,6 +1001,7 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 	struct flowi6 fl6;
 	__u8 prio, ttl;
 	__be16 sport;
+	__be32 label;
 	bool xnet = !net_eq(geneve->net, dev_net(geneve->dev));
 	u32 flags = geneve->flags;
 
@@ -1041,6 +1045,7 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 
 		prio = ip_tunnel_ecn_encap(key->tos, iip, skb);
 		ttl = key->ttl;
+		label = info->key.label;
 	} else {
 		err = geneve6_build_skb(dst, skb, 0, geneve->vni,
 					0, NULL, flags, xnet);
@@ -1052,9 +1057,11 @@ static netdev_tx_t geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev,
 		if (!ttl && ipv6_addr_is_multicast(&fl6.daddr))
 			ttl = 1;
 		ttl = ttl ? : ip6_dst_hoplimit(dst);
+		label = geneve->label;
 	}
+
 	udp_tunnel6_xmit_skb(dst, gs6->sock->sk, skb, dev,
-			     &fl6.saddr, &fl6.daddr, prio, ttl, 0,
+			     &fl6.saddr, &fl6.daddr, prio, ttl, label,
 			     sport, geneve->dst_port,
 			     !!(flags & GENEVE_F_UDP_ZERO_CSUM6_TX));
 	return NETDEV_TX_OK;
@@ -1238,6 +1245,7 @@ static const struct nla_policy geneve_policy[IFLA_GENEVE_MAX + 1] = {
 	[IFLA_GENEVE_REMOTE6]		= { .len = sizeof(struct in6_addr) },
 	[IFLA_GENEVE_TTL]		= { .type = NLA_U8 },
 	[IFLA_GENEVE_TOS]		= { .type = NLA_U8 },
+	[IFLA_GENEVE_LABEL]		= { .type = NLA_U32 },
 	[IFLA_GENEVE_PORT]		= { .type = NLA_U16 },
 	[IFLA_GENEVE_COLLECT_METADATA]	= { .type = NLA_FLAG },
 	[IFLA_GENEVE_UDP_CSUM]		= { .type = NLA_U8 },
@@ -1295,8 +1303,8 @@ static struct geneve_dev *geneve_find_dev(struct geneve_net *gn,
 
 static int geneve_configure(struct net *net, struct net_device *dev,
 			    union geneve_addr *remote,
-			    __u32 vni, __u8 ttl, __u8 tos, __be16 dst_port,
-			    bool metadata, u32 flags)
+			    __u32 vni, __u8 ttl, __u8 tos, __be32 label,
+			    __be16 dst_port, bool metadata, u32 flags)
 {
 	struct geneve_net *gn = net_generic(net, geneve_net_id);
 	struct geneve_dev *t, *geneve = netdev_priv(dev);
@@ -1306,7 +1314,7 @@ static int geneve_configure(struct net *net, struct net_device *dev,
 	if (!remote)
 		return -EINVAL;
 	if (metadata &&
-	    (remote->sa.sa_family != AF_UNSPEC || vni || tos || ttl))
+	    (remote->sa.sa_family != AF_UNSPEC || vni || tos || ttl || label))
 		return -EINVAL;
 
 	geneve->net = net;
@@ -1321,10 +1329,14 @@ static int geneve_configure(struct net *net, struct net_device *dev,
 	    (remote->sa.sa_family == AF_INET6 &&
 	     ipv6_addr_is_multicast(&remote->sin6.sin6_addr)))
 		return -EINVAL;
+	if (label && remote->sa.sa_family != AF_INET6)
+		return -EINVAL;
+
 	geneve->remote = *remote;
 
 	geneve->ttl = ttl;
 	geneve->tos = tos;
+	geneve->label = label;
 	geneve->dst_port = dst_port;
 	geneve->collect_md = metadata;
 	geneve->flags = flags;
@@ -1367,6 +1379,7 @@ static int geneve_newlink(struct net *net, struct net_device *dev,
 	__u8 ttl = 0, tos = 0;
 	bool metadata = false;
 	union geneve_addr remote = geneve_remote_unspec;
+	__be32 label = 0;
 	__u32 vni = 0;
 	u32 flags = 0;
 
@@ -1403,6 +1416,10 @@ static int geneve_newlink(struct net *net, struct net_device *dev,
 	if (data[IFLA_GENEVE_TOS])
 		tos = nla_get_u8(data[IFLA_GENEVE_TOS]);
 
+	if (data[IFLA_GENEVE_LABEL])
+		label = nla_get_be32(data[IFLA_GENEVE_LABEL]) &
+			IPV6_FLOWLABEL_MASK;
+
 	if (data[IFLA_GENEVE_PORT])
 		dst_port = nla_get_be16(data[IFLA_GENEVE_PORT]);
 
@@ -1421,8 +1438,8 @@ static int geneve_newlink(struct net *net, struct net_device *dev,
 	    nla_get_u8(data[IFLA_GENEVE_UDP_ZERO_CSUM6_RX]))
 		flags |= GENEVE_F_UDP_ZERO_CSUM6_RX;
 
-	return geneve_configure(net, dev, &remote, vni, ttl, tos, dst_port,
-				metadata, flags);
+	return geneve_configure(net, dev, &remote, vni, ttl, tos, label,
+				dst_port, metadata, flags);
 }
 
 static void geneve_dellink(struct net_device *dev, struct list_head *head)
@@ -1439,6 +1456,7 @@ static size_t geneve_get_size(const struct net_device *dev)
 		nla_total_size(sizeof(struct in6_addr)) + /* IFLA_GENEVE_REMOTE{6} */
 		nla_total_size(sizeof(__u8)) +  /* IFLA_GENEVE_TTL */
 		nla_total_size(sizeof(__u8)) +  /* IFLA_GENEVE_TOS */
+		nla_total_size(sizeof(__be32)) +  /* IFLA_GENEVE_LABEL */
 		nla_total_size(sizeof(__be16)) +  /* IFLA_GENEVE_PORT */
 		nla_total_size(0) +	 /* IFLA_GENEVE_COLLECT_METADATA */
 		nla_total_size(sizeof(__u8)) + /* IFLA_GENEVE_UDP_CSUM */
@@ -1469,7 +1487,8 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev)
 	}
 
 	if (nla_put_u8(skb, IFLA_GENEVE_TTL, geneve->ttl) ||
-	    nla_put_u8(skb, IFLA_GENEVE_TOS, geneve->tos))
+	    nla_put_u8(skb, IFLA_GENEVE_TOS, geneve->tos) ||
+	    nla_put_be32(skb, IFLA_GENEVE_LABEL, geneve->label))
 		goto nla_put_failure;
 
 	if (nla_put_be16(skb, IFLA_GENEVE_PORT, geneve->dst_port))
@@ -1521,7 +1540,7 @@ struct net_device *geneve_dev_create_fb(struct net *net, const char *name,
 		return dev;
 
 	err = geneve_configure(net, dev, &geneve_remote_unspec,
-			       0, 0, 0, htons(dst_port), true,
+			       0, 0, 0, 0, htons(dst_port), true,
 			       GENEVE_F_UDP_ZERO_CSUM6_RX);
 	if (err)
 		goto err;
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 6bebc97..249eef9 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -479,6 +479,7 @@ enum {
 	IFLA_GENEVE_UDP_CSUM,
 	IFLA_GENEVE_UDP_ZERO_CSUM6_TX,
 	IFLA_GENEVE_UDP_ZERO_CSUM6_RX,
+	IFLA_GENEVE_LABEL,
 	__IFLA_GENEVE_MAX
 };
 #define IFLA_GENEVE_MAX	(__IFLA_GENEVE_MAX - 1)
-- 
cgit v0.10.2


From 4018ab1875e0d00b84ac61bc15427136ad55849e Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 9 Mar 2016 03:00:05 +0100
Subject: bpf: support flow label for bpf_skb_{set, get}_tunnel_key

This patch extends bpf_tunnel_key with a tunnel_label member, that maps
to ip_tunnel_key's label so underlying backends like vxlan and geneve
can propagate the label to udp_tunnel6_xmit_skb(), where it's being set
in the IPv6 header. It allows for having 20 more bits to encode/decode
flow related meta information programmatically. Tested with vxlan and
geneve.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 0e30b19..924f537 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -375,6 +375,7 @@ struct bpf_tunnel_key {
 	};
 	__u8 tunnel_tos;
 	__u8 tunnel_ttl;
+	__u32 tunnel_label;
 };
 
 #endif /* _UAPI__LINUX_BPF_H__ */
diff --git a/net/core/filter.c b/net/core/filter.c
index a66dc03..6fc3893 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1770,12 +1770,15 @@ static u64 bpf_skb_get_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 		return -EPROTO;
 	if (unlikely(size != sizeof(struct bpf_tunnel_key))) {
 		switch (size) {
+		case offsetof(struct bpf_tunnel_key, tunnel_label):
+			goto set_compat;
 		case offsetof(struct bpf_tunnel_key, remote_ipv6[1]):
 			/* Fixup deprecated structure layouts here, so we have
 			 * a common path later on.
 			 */
 			if (ip_tunnel_info_af(info) != AF_INET)
 				return -EINVAL;
+set_compat:
 			to = (struct bpf_tunnel_key *)compat;
 			break;
 		default:
@@ -1787,11 +1790,13 @@ static u64 bpf_skb_get_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	to->tunnel_tos = info->key.tos;
 	to->tunnel_ttl = info->key.ttl;
 
-	if (flags & BPF_F_TUNINFO_IPV6)
+	if (flags & BPF_F_TUNINFO_IPV6) {
 		memcpy(to->remote_ipv6, &info->key.u.ipv6.src,
 		       sizeof(to->remote_ipv6));
-	else
+		to->tunnel_label = be32_to_cpu(info->key.label);
+	} else {
 		to->remote_ipv4 = be32_to_cpu(info->key.u.ipv4.src);
+	}
 
 	if (unlikely(size != sizeof(struct bpf_tunnel_key)))
 		memcpy((void *)(long) r2, to, size);
@@ -1850,6 +1855,7 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 		return -EINVAL;
 	if (unlikely(size != sizeof(struct bpf_tunnel_key))) {
 		switch (size) {
+		case offsetof(struct bpf_tunnel_key, tunnel_label):
 		case offsetof(struct bpf_tunnel_key, remote_ipv6[1]):
 			/* Fixup deprecated structure layouts here, so we have
 			 * a common path later on.
@@ -1862,6 +1868,8 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 			return -EINVAL;
 		}
 	}
+	if (unlikely(!(flags & BPF_F_TUNINFO_IPV6) && from->tunnel_label))
+		return -EINVAL;
 
 	skb_dst_drop(skb);
 	dst_hold((struct dst_entry *) md);
@@ -1882,6 +1890,8 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 		info->mode |= IP_TUNNEL_INFO_IPV6;
 		memcpy(&info->key.u.ipv6.dst, from->remote_ipv6,
 		       sizeof(from->remote_ipv6));
+		info->key.label = cpu_to_be32(from->tunnel_label) &
+				  IPV6_FLOWLABEL_MASK;
 	} else {
 		info->key.u.ipv4.dst = cpu_to_be32(from->remote_ipv4);
 		if (flags & BPF_F_ZERO_CSUM_TX)
-- 
cgit v0.10.2


From fc916ff2027760900f3b373c388a2098253f3c2a Mon Sep 17 00:00:00 2001
From: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Date: Wed, 9 Mar 2016 09:16:23 +0200
Subject: qed: Prevent MF link notifications

When device is configured for Multi-function mode, some older management
firmware might incorrectly notify interfaces of link changes while they
haven't requested the physical link configuration to be set.
This can create bizzare race conditions where unloading interfaces are
getting notified that the link is up.

Let the driver compensate - store the logical requested state of the link
and don't propagate notifications after protocol driver explicitly
requires the link to be unset.

Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h
index e5604ee..fcb8e9b 100644
--- a/drivers/net/ethernet/qlogic/qed/qed.h
+++ b/drivers/net/ethernet/qlogic/qed/qed.h
@@ -303,6 +303,9 @@ struct qed_hwfn {
 	bool				b_int_enabled;
 	bool				b_int_requested;
 
+	/* True if the driver requests for the link */
+	bool				b_drv_link_init;
+
 	struct qed_mcp_info		*mcp_info;
 
 	struct qed_hw_cid_data		*p_tx_cids;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index f23ce73..2bf9824 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -390,7 +390,10 @@ static void qed_mcp_handle_link_change(struct qed_hwfn *p_hwfn,
 		return;
 	}
 
-	p_link->link_up = !!(status & LINK_STATUS_LINK_UP);
+	if (p_hwfn->b_drv_link_init)
+		p_link->link_up = !!(status & LINK_STATUS_LINK_UP);
+	else
+		p_link->link_up = false;
 
 	p_link->full_duplex = true;
 	switch ((status & LINK_STATUS_SPEED_AND_DUPLEX_MASK)) {
@@ -520,6 +523,8 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn,
 		       offsetof(struct public_drv_mb, union_data) + i,
 		       ((u32 *)&phy_cfg)[i >> 2]);
 
+	p_hwfn->b_drv_link_init = b_up;
+
 	if (b_up) {
 		DP_VERBOSE(p_hwfn, NETIF_MSG_LINK,
 			   "Configuring Link: Speed 0x%08x, Pause 0x%08x, adv_speed 0x%08x, loopback 0x%08x, features 0x%08x\n",
-- 
cgit v0.10.2


From 5529bad98f10f742e3ab490733bed8e108508759 Mon Sep 17 00:00:00 2001
From: Tomer Tayar <Tomer.Tayar@qlogic.com>
Date: Wed, 9 Mar 2016 09:16:24 +0200
Subject: qed: Major changes to MB locking

Driver interaction with the managemnt firmware is done via mailbox
commands which the management firmware periodically sample, as well
as placing of additional data in set places in the shared memory.
Each PF has a single designated mailbox address, and all flows that
require messaging to the management should use it.

This patch does 2 things:
 1. It re-defines the critical section surrounding the mailbox sending -
that section should include the setting of the shared memory as well as
the sending of the command [otherwise a race might send a command with
the data of a different command].
 2. It moves the locking scheme from using mutices into using spinlocks.
This lays the groundwork for sending MFW commands from non-sleepable
contexts.

Signed-off-by: Tomer Tayar <Tomer.Tayar@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 2bf9824..6937c88 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -11,8 +11,8 @@
 #include <linux/delay.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
-#include <linux/mutex.h>
 #include <linux/slab.h>
+#include <linux/spinlock.h>
 #include <linux/string.h>
 #include "qed.h"
 #include "qed_hsi.h"
@@ -168,8 +168,8 @@ int qed_mcp_cmd_init(struct qed_hwfn *p_hwfn,
 	if (!p_info->mfw_mb_shadow || !p_info->mfw_mb_addr)
 		goto err;
 
-	/* Initialize the MFW mutex */
-	mutex_init(&p_info->mutex);
+	/* Initialize the MFW spinlock */
+	spin_lock_init(&p_info->lock);
 
 	return 0;
 
@@ -179,6 +179,52 @@ err:
 	return -ENOMEM;
 }
 
+/* Locks the MFW mailbox of a PF to ensure a single access.
+ * The lock is achieved in most cases by holding a spinlock, causing other
+ * threads to wait till a previous access is done.
+ * In some cases (currently when a [UN]LOAD_REQ commands are sent), the single
+ * access is achieved by setting a blocking flag, which will fail other
+ * competing contexts to send their mailboxes.
+ */
+static int qed_mcp_mb_lock(struct qed_hwfn *p_hwfn,
+			   u32 cmd)
+{
+	spin_lock_bh(&p_hwfn->mcp_info->lock);
+
+	/* The spinlock shouldn't be acquired when the mailbox command is
+	 * [UN]LOAD_REQ, since the engine is locked by the MFW, and a parallel
+	 * pending [UN]LOAD_REQ command of another PF together with a spinlock
+	 * (i.e. interrupts are disabled) - can lead to a deadlock.
+	 * It is assumed that for a single PF, no other mailbox commands can be
+	 * sent from another context while sending LOAD_REQ, and that any
+	 * parallel commands to UNLOAD_REQ can be cancelled.
+	 */
+	if (cmd == DRV_MSG_CODE_LOAD_DONE || cmd == DRV_MSG_CODE_UNLOAD_DONE)
+		p_hwfn->mcp_info->block_mb_sending = false;
+
+	if (p_hwfn->mcp_info->block_mb_sending) {
+		DP_NOTICE(p_hwfn,
+			  "Trying to send a MFW mailbox command [0x%x] in parallel to [UN]LOAD_REQ. Aborting.\n",
+			  cmd);
+		spin_unlock_bh(&p_hwfn->mcp_info->lock);
+		return -EBUSY;
+	}
+
+	if (cmd == DRV_MSG_CODE_LOAD_REQ || cmd == DRV_MSG_CODE_UNLOAD_REQ) {
+		p_hwfn->mcp_info->block_mb_sending = true;
+		spin_unlock_bh(&p_hwfn->mcp_info->lock);
+	}
+
+	return 0;
+}
+
+static void qed_mcp_mb_unlock(struct qed_hwfn	*p_hwfn,
+			      u32		cmd)
+{
+	if (cmd != DRV_MSG_CODE_LOAD_REQ && cmd != DRV_MSG_CODE_UNLOAD_REQ)
+		spin_unlock_bh(&p_hwfn->mcp_info->lock);
+}
+
 int qed_mcp_reset(struct qed_hwfn *p_hwfn,
 		  struct qed_ptt *p_ptt)
 {
@@ -187,6 +233,13 @@ int qed_mcp_reset(struct qed_hwfn *p_hwfn,
 	u32 org_mcp_reset_seq, cnt = 0;
 	int rc = 0;
 
+	/* Ensure that only a single thread is accessing the mailbox at a
+	 * certain time.
+	 */
+	rc = qed_mcp_mb_lock(p_hwfn, DRV_MSG_CODE_MCP_RESET);
+	if (rc != 0)
+		return rc;
+
 	/* Set drv command along with the updated sequence */
 	org_mcp_reset_seq = qed_rd(p_hwfn, p_ptt, MISCS_REG_GENERIC_POR_0);
 	DRV_MB_WR(p_hwfn, p_ptt, drv_mb_header,
@@ -209,6 +262,8 @@ int qed_mcp_reset(struct qed_hwfn *p_hwfn,
 		rc = -EAGAIN;
 	}
 
+	qed_mcp_mb_unlock(p_hwfn, DRV_MSG_CODE_MCP_RESET);
+
 	return rc;
 }
 
@@ -275,14 +330,12 @@ static int qed_do_mcp_cmd(struct qed_hwfn *p_hwfn,
 	return rc;
 }
 
-int qed_mcp_cmd(struct qed_hwfn *p_hwfn,
-		struct qed_ptt *p_ptt,
-		u32 cmd,
-		u32 param,
-		u32 *o_mcp_resp,
-		u32 *o_mcp_param)
+static int qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
+				 struct qed_ptt *p_ptt,
+				 struct qed_mcp_mb_params *p_mb_params)
 {
-	int rc = 0;
+	u32 union_data_addr;
+	int rc;
 
 	/* MCP not initialized */
 	if (!qed_mcp_is_init(p_hwfn)) {
@@ -290,28 +343,56 @@ int qed_mcp_cmd(struct qed_hwfn *p_hwfn,
 		return -EBUSY;
 	}
 
-	/* Lock Mutex to ensure only single thread is
-	 * accessing the MCP at one time
+	union_data_addr = p_hwfn->mcp_info->drv_mb_addr +
+			  offsetof(struct public_drv_mb, union_data);
+
+	/* Ensure that only a single thread is accessing the mailbox at a
+	 * certain time.
 	 */
-	mutex_lock(&p_hwfn->mcp_info->mutex);
-	rc = qed_do_mcp_cmd(p_hwfn, p_ptt, cmd, param,
-			    o_mcp_resp, o_mcp_param);
-	/* Release Mutex */
-	mutex_unlock(&p_hwfn->mcp_info->mutex);
+	rc = qed_mcp_mb_lock(p_hwfn, p_mb_params->cmd);
+	if (rc)
+		return rc;
+
+	if (p_mb_params->p_data_src != NULL)
+		qed_memcpy_to(p_hwfn, p_ptt, union_data_addr,
+			      p_mb_params->p_data_src,
+			      sizeof(*p_mb_params->p_data_src));
+
+	rc = qed_do_mcp_cmd(p_hwfn, p_ptt, p_mb_params->cmd,
+			    p_mb_params->param, &p_mb_params->mcp_resp,
+			    &p_mb_params->mcp_param);
+
+	if (p_mb_params->p_data_dst != NULL)
+		qed_memcpy_from(p_hwfn, p_ptt, p_mb_params->p_data_dst,
+				union_data_addr,
+				sizeof(*p_mb_params->p_data_dst));
+
+	qed_mcp_mb_unlock(p_hwfn, p_mb_params->cmd);
 
 	return rc;
 }
 
-static void qed_mcp_set_drv_ver(struct qed_dev *cdev,
-				struct qed_hwfn *p_hwfn,
-				struct qed_ptt *p_ptt)
+int qed_mcp_cmd(struct qed_hwfn *p_hwfn,
+		struct qed_ptt *p_ptt,
+		u32 cmd,
+		u32 param,
+		u32 *o_mcp_resp,
+		u32 *o_mcp_param)
 {
-	u32 i;
+	struct qed_mcp_mb_params mb_params;
+	int rc;
+
+	memset(&mb_params, 0, sizeof(mb_params));
+	mb_params.cmd = cmd;
+	mb_params.param = param;
+	rc = qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params);
+	if (rc)
+		return rc;
+
+	*o_mcp_resp = mb_params.mcp_resp;
+	*o_mcp_param = mb_params.mcp_param;
 
-	/* Copy version string to MCP */
-	for (i = 0; i < MCP_DRV_VER_STR_SIZE_DWORD; i++)
-		DRV_MB_WR(p_hwfn, p_ptt, union_data.ver_str[i],
-			  *(u32 *)&cdev->ver_str[i * sizeof(u32)]);
+	return 0;
 }
 
 int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
@@ -319,26 +400,18 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
 		     u32 *p_load_code)
 {
 	struct qed_dev *cdev = p_hwfn->cdev;
-	u32 param;
+	struct qed_mcp_mb_params mb_params;
+	union drv_union_data union_data;
 	int rc;
 
-	if (!qed_mcp_is_init(p_hwfn)) {
-		DP_NOTICE(p_hwfn, "MFW is not initialized !\n");
-		return -EBUSY;
-	}
-
-	/* Save driver's version to shmem */
-	qed_mcp_set_drv_ver(cdev, p_hwfn, p_ptt);
-
-	DP_VERBOSE(p_hwfn, QED_MSG_SP, "fw_seq 0x%08x, drv_pulse 0x%x\n",
-		   p_hwfn->mcp_info->drv_mb_seq,
-		   p_hwfn->mcp_info->drv_pulse_seq);
-
+	memset(&mb_params, 0, sizeof(mb_params));
 	/* Load Request */
-	rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_LOAD_REQ,
-			 (PDA_COMP | DRV_ID_MCP_HSI_VER_CURRENT |
-			  cdev->drv_type),
-			 p_load_code, &param);
+	mb_params.cmd = DRV_MSG_CODE_LOAD_REQ;
+	mb_params.param = PDA_COMP | DRV_ID_MCP_HSI_VER_CURRENT |
+			  cdev->drv_type;
+	memcpy(&union_data.ver_str, cdev->ver_str, MCP_DRV_VER_STR_SIZE);
+	mb_params.p_data_src = &union_data;
+	rc = qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params);
 
 	/* if mcp fails to respond we must abort */
 	if (rc) {
@@ -346,6 +419,8 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
 		return rc;
 	}
 
+	*p_load_code = mb_params.mcp_resp;
+
 	/* If MFW refused (e.g. other port is in diagnostic mode) we
 	 * must abort. This can happen in the following cases:
 	 * - Other port is in diagnostic mode
@@ -495,55 +570,43 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn,
 		     bool b_up)
 {
 	struct qed_mcp_link_params *params = &p_hwfn->mcp_info->link_input;
-	u32 param = 0, reply = 0, cmd;
-	struct pmm_phy_cfg phy_cfg;
+	struct qed_mcp_mb_params mb_params;
+	union drv_union_data union_data;
+	struct pmm_phy_cfg *phy_cfg;
 	int rc = 0;
-	u32 i;
-
-	if (!qed_mcp_is_init(p_hwfn)) {
-		DP_NOTICE(p_hwfn, "MFW is not initialized !\n");
-		return -EBUSY;
-	}
+	u32 cmd;
 
 	/* Set the shmem configuration according to params */
-	memset(&phy_cfg, 0, sizeof(phy_cfg));
+	phy_cfg = &union_data.drv_phy_cfg;
+	memset(phy_cfg, 0, sizeof(*phy_cfg));
 	cmd = b_up ? DRV_MSG_CODE_INIT_PHY : DRV_MSG_CODE_LINK_RESET;
 	if (!params->speed.autoneg)
-		phy_cfg.speed = params->speed.forced_speed;
-	phy_cfg.pause |= (params->pause.autoneg) ? PMM_PAUSE_AUTONEG : 0;
-	phy_cfg.pause |= (params->pause.forced_rx) ? PMM_PAUSE_RX : 0;
-	phy_cfg.pause |= (params->pause.forced_tx) ? PMM_PAUSE_TX : 0;
-	phy_cfg.adv_speed = params->speed.advertised_speeds;
-	phy_cfg.loopback_mode = params->loopback_mode;
-
-	/* Write the requested configuration to shmem */
-	for (i = 0; i < sizeof(phy_cfg); i += 4)
-		qed_wr(p_hwfn, p_ptt,
-		       p_hwfn->mcp_info->drv_mb_addr +
-		       offsetof(struct public_drv_mb, union_data) + i,
-		       ((u32 *)&phy_cfg)[i >> 2]);
+		phy_cfg->speed = params->speed.forced_speed;
+	phy_cfg->pause |= (params->pause.autoneg) ? PMM_PAUSE_AUTONEG : 0;
+	phy_cfg->pause |= (params->pause.forced_rx) ? PMM_PAUSE_RX : 0;
+	phy_cfg->pause |= (params->pause.forced_tx) ? PMM_PAUSE_TX : 0;
+	phy_cfg->adv_speed = params->speed.advertised_speeds;
+	phy_cfg->loopback_mode = params->loopback_mode;
 
 	p_hwfn->b_drv_link_init = b_up;
 
 	if (b_up) {
 		DP_VERBOSE(p_hwfn, NETIF_MSG_LINK,
 			   "Configuring Link: Speed 0x%08x, Pause 0x%08x, adv_speed 0x%08x, loopback 0x%08x, features 0x%08x\n",
-			   phy_cfg.speed,
-			   phy_cfg.pause,
-			   phy_cfg.adv_speed,
-			   phy_cfg.loopback_mode,
-			   phy_cfg.feature_config_flags);
+			   phy_cfg->speed,
+			   phy_cfg->pause,
+			   phy_cfg->adv_speed,
+			   phy_cfg->loopback_mode,
+			   phy_cfg->feature_config_flags);
 	} else {
 		DP_VERBOSE(p_hwfn, NETIF_MSG_LINK,
 			   "Resetting link\n");
 	}
 
-	DP_VERBOSE(p_hwfn, QED_MSG_SP, "fw_seq 0x%08x, drv_pulse 0x%x\n",
-		   p_hwfn->mcp_info->drv_mb_seq,
-		   p_hwfn->mcp_info->drv_pulse_seq);
-
-	/* Load Request */
-	rc = qed_mcp_cmd(p_hwfn, p_ptt, cmd, 0, &reply, &param);
+	memset(&mb_params, 0, sizeof(mb_params));
+	mb_params.cmd = cmd;
+	mb_params.p_data_src = &union_data;
+	rc = qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params);
 
 	/* if mcp fails to respond we must abort */
 	if (rc) {
@@ -836,31 +899,28 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn,
 			 struct qed_ptt *p_ptt,
 			 struct qed_mcp_drv_version *p_ver)
 {
-	int rc = 0;
-	u32 param = 0, reply = 0, i;
-
-	if (!qed_mcp_is_init(p_hwfn)) {
-		DP_NOTICE(p_hwfn, "MFW is not initialized !\n");
-		return -EBUSY;
-	}
+	struct drv_version_stc *p_drv_version;
+	struct qed_mcp_mb_params mb_params;
+	union drv_union_data union_data;
+	__be32 val;
+	u32 i;
+	int rc;
 
-	DRV_MB_WR(p_hwfn, p_ptt, union_data.drv_version.version,
-		  p_ver->version);
-	/* Copy version string to shmem */
-	for (i = 0; i < (MCP_DRV_VER_STR_SIZE - 4) / 4; i++) {
-		DRV_MB_WR(p_hwfn, p_ptt,
-			  union_data.drv_version.name[i * sizeof(u32)],
-			  *(u32 *)&p_ver->name[i * sizeof(u32)]);
+	p_drv_version = &union_data.drv_version;
+	p_drv_version->version = p_ver->version;
+	for (i = 0; i < MCP_DRV_VER_STR_SIZE - 1; i += 4) {
+		val = cpu_to_be32(p_ver->name[i]);
+		*(u32 *)&p_drv_version->name[i * sizeof(u32)] = val;
 	}
 
-	rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_SET_VERSION, 0, &reply,
-			 &param);
-	if (rc) {
+	memset(&mb_params, 0, sizeof(mb_params));
+	mb_params.cmd = DRV_MSG_CODE_SET_VERSION;
+	mb_params.p_data_src = &union_data;
+	rc = qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params);
+	if (rc)
 		DP_ERR(p_hwfn, "MCP response failure, aborting\n");
-		return rc;
-	}
 
-	return 0;
+	return rc;
 }
 
 int qed_mcp_set_led(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
index 506197d..50917a2 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
@@ -11,8 +11,8 @@
 
 #include <linux/types.h>
 #include <linux/delay.h>
-#include <linux/mutex.h>
 #include <linux/slab.h>
+#include <linux/spinlock.h>
 #include "qed_hsi.h"
 
 struct qed_mcp_link_speed_params {
@@ -255,7 +255,8 @@ int qed_mcp_set_led(struct qed_hwfn *p_hwfn,
 #define MFW_PORT(_p_hwfn)       ((_p_hwfn)->abs_pf_id %	\
 				 ((_p_hwfn)->cdev->num_ports_in_engines * 2))
 struct qed_mcp_info {
-	struct mutex				mutex; /* MCP access lock */
+	spinlock_t				lock;
+	bool					block_mb_sending;
 	u32					public_base;
 	u32					drv_mb_addr;
 	u32					mfw_mb_addr;
@@ -272,6 +273,15 @@ struct qed_mcp_info {
 	u16					mcp_hist;
 };
 
+struct qed_mcp_mb_params {
+	u32			cmd;
+	u32			param;
+	union drv_union_data	*p_data_src;
+	union drv_union_data	*p_data_dst;
+	u32			mcp_resp;
+	u32			mcp_param;
+};
+
 /**
  * @brief Initialize the interface with the MCP
  *
-- 
cgit v0.10.2


From 334c03b59ba28aa76384474e705a8a0d47f0bb39 Mon Sep 17 00:00:00 2001
From: Zvi Nachmani <Zvi.Nachmani@qlogic.com>
Date: Wed, 9 Mar 2016 09:16:25 +0200
Subject: qed: Notify of transciever changes

Handle a new message from the MFW, one that indicate that the transciever
state has changed, and log that into the system logs.

Signed-off-by: Zvi Nachmani <Zvi.Nachmani@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
index 236db8a..a368f5e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h
@@ -3621,6 +3621,9 @@ struct public_port {
 	u32					fc_npiv_nvram_tbl_addr;
 	u32					fc_npiv_nvram_tbl_size;
 	u32					transceiver_data;
+#define PMM_TRANSCEIVER_STATE_MASK		0x000000FF
+#define PMM_TRANSCEIVER_STATE_SHIFT		0x00000000
+#define PMM_TRANSCEIVER_STATE_PRESENT		0x00000001
 };
 
 /**************************************/
@@ -3955,6 +3958,14 @@ enum MFW_DRV_MSG_TYPE {
 	MFW_DRV_MSG_DCBX_REMOTE_MIB_UPDATED,
 	MFW_DRV_MSG_DCBX_OPERATIONAL_MIB_UPDATED,
 	MFW_DRV_MSG_ERROR_RECOVERY,
+	MFW_DRV_MSG_BW_UPDATE,
+	MFW_DRV_MSG_S_TAG_UPDATE,
+	MFW_DRV_MSG_GET_LAN_STATS,
+	MFW_DRV_MSG_GET_FCOE_STATS,
+	MFW_DRV_MSG_GET_ISCSI_STATS,
+	MFW_DRV_MSG_GET_RDMA_STATS,
+	MFW_DRV_MSG_FAILURE_DETECTED,
+	MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE,
 	MFW_DRV_MSG_MAX
 };
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 6937c88..39c831d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -440,6 +440,33 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
 	return 0;
 }
 
+static void qed_mcp_handle_transceiver_change(struct qed_hwfn *p_hwfn,
+					      struct qed_ptt *p_ptt)
+{
+	u32 transceiver_state;
+
+	transceiver_state = qed_rd(p_hwfn, p_ptt,
+				   p_hwfn->mcp_info->port_addr +
+				   offsetof(struct public_port,
+					    transceiver_data));
+
+	DP_VERBOSE(p_hwfn,
+		   (NETIF_MSG_HW | QED_MSG_SP),
+		   "Received transceiver state update [0x%08x] from mfw [Addr 0x%x]\n",
+		   transceiver_state,
+		   (u32)(p_hwfn->mcp_info->port_addr +
+			 offsetof(struct public_port,
+				  transceiver_data)));
+
+	transceiver_state = GET_FIELD(transceiver_state,
+				      PMM_TRANSCEIVER_STATE);
+
+	if (transceiver_state == PMM_TRANSCEIVER_STATE_PRESENT)
+		DP_NOTICE(p_hwfn, "Transceiver is present.\n");
+	else
+		DP_NOTICE(p_hwfn, "Transceiver is unplugged.\n");
+}
+
 static void qed_mcp_handle_link_change(struct qed_hwfn *p_hwfn,
 				       struct qed_ptt *p_ptt,
 				       bool b_reset)
@@ -649,6 +676,9 @@ int qed_mcp_handle_events(struct qed_hwfn *p_hwfn,
 		case MFW_DRV_MSG_LINK_CHANGE:
 			qed_mcp_handle_link_change(p_hwfn, p_ptt, false);
 			break;
+		case MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE:
+			qed_mcp_handle_transceiver_change(p_hwfn, p_ptt);
+			break;
 		default:
 			DP_NOTICE(p_hwfn, "Unimplemented MFW message %d\n", i);
 			rc = -EINVAL;
-- 
cgit v0.10.2


From 8f60bafec368191594a8a6ed5f0ce14968131754 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 9 Mar 2016 09:16:26 +0200
Subject: qed: Enlrage the drain timeout

In the scenario where slowpath configuration isn't passing due to
various pause configurations affecting the chip, the theoretical time
required in worst-case-scenario to empty hw fifos sufficiently to
guarantee that slowpath configuration would flow is currently
insufficient.

This increases such a drain request to the theoretical maximum.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 39c831d..b89c9a8 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -899,11 +899,11 @@ int qed_mcp_drain(struct qed_hwfn *p_hwfn,
 	int rc;
 
 	rc = qed_mcp_cmd(p_hwfn, p_ptt,
-			 DRV_MSG_CODE_NIG_DRAIN, 100,
+			 DRV_MSG_CODE_NIG_DRAIN, 1000,
 			 &resp, &param);
 
 	/* Wait for the drain to complete before returning */
-	msleep(120);
+	msleep(1020);
 
 	return rc;
 }
-- 
cgit v0.10.2


From d157bd761585605b7882935ffb86286919f62ea1 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 10 Mar 2016 01:56:23 +0100
Subject: netfilter: x_tables: check for size overflow

Ben Hawkes says:
 integer overflow in xt_alloc_table_info, which on 32-bit systems can
 lead to small structure allocation and a copy_from_user based heap
 corruption.

Reported-by: Ben Hawkes <hawkes@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index d0cd2b9..582c9cf 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -659,6 +659,9 @@ struct xt_table_info *xt_alloc_table_info(unsigned int size)
 	struct xt_table_info *info = NULL;
 	size_t sz = sizeof(*info) + size;
 
+	if (sz < sizeof(*info))
+		return NULL;
+
 	/* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */
 	if ((SMP_ALIGN(size) >> PAGE_SHIFT) + 2 > totalram_pages)
 		return NULL;
-- 
cgit v0.10.2


From 3a8befcd7872f572882ad7e14994a17f9b55dd4e Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Sat, 12 Mar 2016 12:03:27 +0100
Subject: rocker: move ageing_time from struct rocker to struct ofdpa

This is OF-DPA specific, used only there, similar to
ofdpa_port->ageing_time. So move it to OF-DPA code.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/rocker/rocker.h b/drivers/net/ethernet/rocker/rocker.h
index 011f1b6..1ab995f 100644
--- a/drivers/net/ethernet/rocker/rocker.h
+++ b/drivers/net/ethernet/rocker/rocker.h
@@ -63,7 +63,6 @@ struct rocker {
 	struct {
 		u64 id;
 	} hw;
-	unsigned long ageing_time;
 	spinlock_t cmd_ring_lock;		/* for cmd ring accesses */
 	struct rocker_dma_ring_info cmd_ring;
 	struct rocker_dma_ring_info event_ring;
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index c8b1bf4..28b775e 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -2733,8 +2733,6 @@ static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	rocker->hw.id = rocker_read64(rocker, SWITCH_ID);
 
-	rocker->ageing_time = BR_DEFAULT_AGEING_TIME;
-
 	err = rocker_probe_ports(rocker);
 	if (err) {
 		dev_err(&pdev->dev, "failed to probe ports\n");
diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
index cfde525..0e758bc 100644
--- a/drivers/net/ethernet/rocker/rocker_ofdpa.c
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -188,6 +188,7 @@ struct ofdpa {
 	DECLARE_HASHTABLE(neigh_tbl, 16);
 	spinlock_t neigh_tbl_lock;		/* for neigh tbl accesses */
 	u32 neigh_tbl_next_index;
+	unsigned long ageing_time;
 };
 
 struct ofdpa_port {
@@ -2105,7 +2106,7 @@ static void ofdpa_fdb_cleanup(unsigned long data)
 	struct ofdpa_port *ofdpa_port;
 	struct ofdpa_fdb_tbl_entry *entry;
 	struct hlist_node *tmp;
-	unsigned long next_timer = jiffies + ofdpa->rocker->ageing_time;
+	unsigned long next_timer = jiffies + ofdpa->ageing_time;
 	unsigned long expires;
 	unsigned long lock_flags;
 	int flags = OFDPA_OP_FLAG_NOWAIT | OFDPA_OP_FLAG_REMOVE |
@@ -2492,6 +2493,8 @@ static int ofdpa_init(struct rocker *rocker)
 		    (unsigned long) ofdpa);
 	mod_timer(&ofdpa->fdb_cleanup_timer, jiffies);
 
+	ofdpa->ageing_time = BR_DEFAULT_AGEING_TIME;
+
 	return 0;
 }
 
@@ -2648,12 +2651,12 @@ ofdpa_port_attr_bridge_ageing_time_set(struct rocker_port *rocker_port,
 				       struct switchdev_trans *trans)
 {
 	struct ofdpa_port *ofdpa_port = rocker_port->wpriv;
-	struct rocker *rocker = rocker_port->rocker;
+	struct ofdpa *ofdpa = ofdpa_port->ofdpa;
 
 	if (!switchdev_trans_ph_prepare(trans)) {
 		ofdpa_port->ageing_time = clock_t_to_jiffies(ageing_time);
-		if (ofdpa_port->ageing_time < rocker->ageing_time)
-			rocker->ageing_time = ofdpa_port->ageing_time;
+		if (ofdpa_port->ageing_time < ofdpa->ageing_time)
+			ofdpa->ageing_time = ofdpa_port->ageing_time;
 		mod_timer(&ofdpa_port->ofdpa->fdb_cleanup_timer, jiffies);
 	}
 
-- 
cgit v0.10.2


From c194cf93c164ed1c71142485ee0f70f9f2d1fe35 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 9 Mar 2016 09:24:23 -0800
Subject: gro: Defer clearing of flush bit in tunnel paths

This patch updates the GRO handlers for GRE, VXLAN, GENEVE, and FOU so that
we do not clear the flush bit until after we have called the next level GRO
handler.  Previously this was being cleared before parsing through the list
of frames, however this resulted in several paths where either the bit
needed to be reset but wasn't as in the case of FOU, or cases where it was
being set as in GENEVE.  By just deferring the clearing of the bit until
after the next level protocol has been parsed we can avoid any unnecessary
bit twiddling and avoid bugs.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 33185b9..192631a 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -463,8 +463,6 @@ static struct sk_buff **geneve_gro_receive(struct sk_buff **head,
 			goto out;
 	}
 
-	flush = 0;
-
 	for (p = *head; p; p = p->next) {
 		if (!NAPI_GRO_CB(p)->same_flow)
 			continue;
@@ -481,14 +479,13 @@ static struct sk_buff **geneve_gro_receive(struct sk_buff **head,
 
 	rcu_read_lock();
 	ptype = gro_find_receive_by_type(type);
-	if (!ptype) {
-		flush = 1;
+	if (!ptype)
 		goto out_unlock;
-	}
 
 	skb_gro_pull(skb, gh_len);
 	skb_gro_postpull_rcsum(skb, gh, gh_len);
 	pp = ptype->callbacks.gro_receive(head, skb);
+	flush = 0;
 
 out_unlock:
 	rcu_read_unlock();
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 8eda76f..800106a7 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -591,8 +591,6 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head,
 
 	skb_gro_pull(skb, sizeof(struct vxlanhdr)); /* pull vxlan header */
 
-	flush = 0;
-
 	for (p = *head; p; p = p->next) {
 		if (!NAPI_GRO_CB(p)->same_flow)
 			continue;
@@ -606,6 +604,7 @@ static struct sk_buff **vxlan_gro_receive(struct sk_buff **head,
 	}
 
 	pp = eth_gro_receive(head, skb);
+	flush = 0;
 
 out:
 	skb_gro_remcsum_cleanup(skb, &grc);
diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c
index 88dab0c..7804842 100644
--- a/net/ipv4/fou.c
+++ b/net/ipv4/fou.c
@@ -319,8 +319,6 @@ static struct sk_buff **gue_gro_receive(struct sk_buff **head,
 
 	skb_gro_pull(skb, hdrlen);
 
-	flush = 0;
-
 	for (p = *head; p; p = p->next) {
 		const struct guehdr *guehdr2;
 
@@ -352,6 +350,7 @@ static struct sk_buff **gue_gro_receive(struct sk_buff **head,
 		goto out_unlock;
 
 	pp = ops->callbacks.gro_receive(head, skb);
+	flush = 0;
 
 out_unlock:
 	rcu_read_unlock();
diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 47f4c54..540866d 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -175,8 +175,6 @@ static struct sk_buff **gre_gro_receive(struct sk_buff **head,
 					     null_compute_pseudo);
 	}
 
-	flush = 0;
-
 	for (p = *head; p; p = p->next) {
 		const struct gre_base_hdr *greh2;
 
@@ -213,6 +211,7 @@ static struct sk_buff **gre_gro_receive(struct sk_buff **head,
 	skb_gro_postpull_rcsum(skb, greh, grehlen);
 
 	pp = ptype->callbacks.gro_receive(head, skb);
+	flush = 0;
 
 out_unlock:
 	rcu_read_unlock();
-- 
cgit v0.10.2


From 338039635d01524090e7bd706a3e555e20d5b337 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Wed, 9 Mar 2016 09:25:26 -0800
Subject: csum: Update csum_block_add to use rotate instead of byteswap

The code for csum_block_add was doing a funky byteswap to swap the even and
odd bytes of the checksum if the offset was odd.  Instead of doing this we
can save ourselves some trouble and just shift by 8 as this should have the
same effect in terms of the final checksum value and only requires one
instruction.

In addition we can update csum_block_sub to just use csum_block_add with a
inverse value for csum2.  This way we follow the same code path as
csum_block_add without having to duplicate it.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/checksum.h b/include/net/checksum.h
index abffc64..5c30891 100644
--- a/include/net/checksum.h
+++ b/include/net/checksum.h
@@ -88,8 +88,11 @@ static inline __wsum
 csum_block_add(__wsum csum, __wsum csum2, int offset)
 {
 	u32 sum = (__force u32)csum2;
-	if (offset&1)
-		sum = ((sum&0xFF00FF)<<8)+((sum>>8)&0xFF00FF);
+
+	/* rotate sum to align it with a 16b boundary */
+	if (offset & 1)
+		sum = ror32(sum, 8);
+
 	return csum_add(csum, (__force __wsum)sum);
 }
 
@@ -102,10 +105,7 @@ csum_block_add_ext(__wsum csum, __wsum csum2, int offset, int len)
 static inline __wsum
 csum_block_sub(__wsum csum, __wsum csum2, int offset)
 {
-	u32 sum = (__force u32)csum2;
-	if (offset&1)
-		sum = ((sum&0xFF00FF)<<8)+((sum>>8)&0xFF00FF);
-	return csum_sub(csum, (__force __wsum)sum);
+	return csum_block_add(csum, ~csum2, offset);
 }
 
 static inline __wsum csum_unfold(__sum16 n)
-- 
cgit v0.10.2


From dad8aff754247781514c9d38f6db89557685136d Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 9 Mar 2016 23:22:56 +0000
Subject: rxrpc: Replace all unsigned with unsigned int

Replace all "unsigned" types with "unsigned int" types.

Reported-by: David Miller <davem@davemloft.net>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index a765017..9d935fa 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -81,7 +81,7 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx,
 				  struct sockaddr_rxrpc *srx,
 				  int len)
 {
-	unsigned tail;
+	unsigned int tail;
 
 	if (len < sizeof(struct sockaddr_rxrpc))
 		return -EINVAL;
diff --git a/net/rxrpc/ar-ack.c b/net/rxrpc/ar-ack.c
index 20f3f00..16d9670 100644
--- a/net/rxrpc/ar-ack.c
+++ b/net/rxrpc/ar-ack.c
@@ -23,7 +23,7 @@
  * How long to wait before scheduling ACK generation after seeing a
  * packet with RXRPC_REQUEST_ACK set (in jiffies).
  */
-unsigned rxrpc_requested_ack_delay = 1;
+unsigned int rxrpc_requested_ack_delay = 1;
 
 /*
  * How long to wait before scheduling an ACK with subtype DELAY (in jiffies).
@@ -32,7 +32,7 @@ unsigned rxrpc_requested_ack_delay = 1;
  * all consumed within this time we will send a DELAY ACK if an ACK was not
  * requested to let the sender know it doesn't need to resend.
  */
-unsigned rxrpc_soft_ack_delay = 1 * HZ;
+unsigned int rxrpc_soft_ack_delay = 1 * HZ;
 
 /*
  * How long to wait before scheduling an ACK with subtype IDLE (in jiffies).
@@ -41,7 +41,7 @@ unsigned rxrpc_soft_ack_delay = 1 * HZ;
  * further packets aren't immediately received to decide when to send an IDLE
  * ACK let the other end know that it can free up its Tx buffer space.
  */
-unsigned rxrpc_idle_ack_delay = 0.5 * HZ;
+unsigned int rxrpc_idle_ack_delay = 0.5 * HZ;
 
 /*
  * Receive window size in packets.  This indicates the maximum number of
@@ -49,19 +49,19 @@ unsigned rxrpc_idle_ack_delay = 0.5 * HZ;
  * limit is hit, we should generate an EXCEEDS_WINDOW ACK and discard further
  * packets.
  */
-unsigned rxrpc_rx_window_size = 32;
+unsigned int rxrpc_rx_window_size = 32;
 
 /*
  * Maximum Rx MTU size.  This indicates to the sender the size of jumbo packet
  * made by gluing normal packets together that we're willing to handle.
  */
-unsigned rxrpc_rx_mtu = 5692;
+unsigned int rxrpc_rx_mtu = 5692;
 
 /*
  * The maximum number of fragments in a received jumbo packet that we tell the
  * sender that we're willing to handle.
  */
-unsigned rxrpc_rx_jumbo_max = 4;
+unsigned int rxrpc_rx_jumbo_max = 4;
 
 static const char *rxrpc_acks(u8 reason)
 {
diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c
index 4a499e0..7c8d300 100644
--- a/net/rxrpc/ar-call.c
+++ b/net/rxrpc/ar-call.c
@@ -21,12 +21,12 @@
 /*
  * Maximum lifetime of a call (in jiffies).
  */
-unsigned rxrpc_max_call_lifetime = 60 * HZ;
+unsigned int rxrpc_max_call_lifetime = 60 * HZ;
 
 /*
  * Time till dead call expires after last use (in jiffies).
  */
-unsigned rxrpc_dead_call_expiry = 2 * HZ;
+unsigned int rxrpc_dead_call_expiry = 2 * HZ;
 
 const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = {
 	[RXRPC_CALL_CLIENT_SEND_REQUEST]	= "ClSndReq",
diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c
index 53df14c..9942da1 100644
--- a/net/rxrpc/ar-connection.c
+++ b/net/rxrpc/ar-connection.c
@@ -21,7 +21,7 @@
 /*
  * Time till a connection expires after last use (in seconds).
  */
-unsigned rxrpc_connection_expiry = 10 * 60;
+unsigned int rxrpc_connection_expiry = 10 * 60;
 
 static void rxrpc_connection_reaper(struct work_struct *work);
 
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 8b495ae..a3002f4 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -478,12 +478,12 @@ int rxrpc_reject_call(struct rxrpc_sock *);
 /*
  * ar-ack.c
  */
-extern unsigned rxrpc_requested_ack_delay;
-extern unsigned rxrpc_soft_ack_delay;
-extern unsigned rxrpc_idle_ack_delay;
-extern unsigned rxrpc_rx_window_size;
-extern unsigned rxrpc_rx_mtu;
-extern unsigned rxrpc_rx_jumbo_max;
+extern unsigned int rxrpc_requested_ack_delay;
+extern unsigned int rxrpc_soft_ack_delay;
+extern unsigned int rxrpc_idle_ack_delay;
+extern unsigned int rxrpc_rx_window_size;
+extern unsigned int rxrpc_rx_mtu;
+extern unsigned int rxrpc_rx_jumbo_max;
 
 void __rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool);
 void rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool);
@@ -492,8 +492,8 @@ void rxrpc_process_call(struct work_struct *);
 /*
  * ar-call.c
  */
-extern unsigned rxrpc_max_call_lifetime;
-extern unsigned rxrpc_dead_call_expiry;
+extern unsigned int rxrpc_max_call_lifetime;
+extern unsigned int rxrpc_dead_call_expiry;
 extern struct kmem_cache *rxrpc_call_jar;
 extern struct list_head rxrpc_calls;
 extern rwlock_t rxrpc_call_lock;
@@ -516,7 +516,7 @@ void __exit rxrpc_destroy_all_calls(void);
 /*
  * ar-connection.c
  */
-extern unsigned rxrpc_connection_expiry;
+extern unsigned int rxrpc_connection_expiry;
 extern struct list_head rxrpc_connections;
 extern rwlock_t rxrpc_connection_lock;
 
@@ -579,7 +579,7 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t,
 /*
  * ar-output.c
  */
-extern unsigned rxrpc_resend_timeout;
+extern unsigned int rxrpc_resend_timeout;
 
 int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *);
 int rxrpc_client_sendmsg(struct rxrpc_sock *, struct rxrpc_transport *,
@@ -627,7 +627,7 @@ void rxrpc_packet_destructor(struct sk_buff *);
 /*
  * ar-transport.c
  */
-extern unsigned rxrpc_transport_expiry;
+extern unsigned int rxrpc_transport_expiry;
 
 struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *,
 					    struct rxrpc_peer *, gfp_t);
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 14c8df6..d36fb6e 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -21,7 +21,7 @@
 /*
  * Time till packet resend (in jiffies).
  */
-unsigned rxrpc_resend_timeout = 4 * HZ;
+unsigned int rxrpc_resend_timeout = 4 * HZ;
 
 static int rxrpc_send_data(struct rxrpc_sock *rx,
 			   struct rxrpc_call *call,
diff --git a/net/rxrpc/ar-transport.c b/net/rxrpc/ar-transport.c
index 5f9b9d4..66a1a56 100644
--- a/net/rxrpc/ar-transport.c
+++ b/net/rxrpc/ar-transport.c
@@ -20,7 +20,7 @@
 /*
  * Time after last use at which transport record is cleaned up.
  */
-unsigned rxrpc_transport_expiry = 3600 * 24;
+unsigned int rxrpc_transport_expiry = 3600 * 24;
 
 static void rxrpc_transport_reaper(struct work_struct *work);
 
diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c
index 093547a..d20ed57 100644
--- a/net/rxrpc/sysctl.c
+++ b/net/rxrpc/sysctl.c
@@ -15,11 +15,11 @@
 #include "ar-internal.h"
 
 static struct ctl_table_header *rxrpc_sysctl_reg_table;
-static const unsigned zero = 0;
-static const unsigned one = 1;
-static const unsigned four = 4;
-static const unsigned n_65535 = 65535;
-static const unsigned n_max_acks = RXRPC_MAXACKS;
+static const unsigned int zero = 0;
+static const unsigned int one = 1;
+static const unsigned int four = 4;
+static const unsigned int n_65535 = 65535;
+static const unsigned int n_max_acks = RXRPC_MAXACKS;
 
 /*
  * RxRPC operating parameters.
@@ -32,7 +32,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "req_ack_delay",
 		.data		= &rxrpc_requested_ack_delay,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_ms_jiffies,
 		.extra1		= (void *)&zero,
@@ -40,7 +40,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "soft_ack_delay",
 		.data		= &rxrpc_soft_ack_delay,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_ms_jiffies,
 		.extra1		= (void *)&one,
@@ -48,7 +48,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "idle_ack_delay",
 		.data		= &rxrpc_idle_ack_delay,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_ms_jiffies,
 		.extra1		= (void *)&one,
@@ -56,7 +56,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "resend_timeout",
 		.data		= &rxrpc_resend_timeout,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_ms_jiffies,
 		.extra1		= (void *)&one,
@@ -66,7 +66,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "max_call_lifetime",
 		.data		= &rxrpc_max_call_lifetime,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 		.extra1		= (void *)&one,
@@ -74,7 +74,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "dead_call_expiry",
 		.data		= &rxrpc_dead_call_expiry,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 		.extra1		= (void *)&one,
@@ -84,7 +84,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "connection_expiry",
 		.data		= &rxrpc_connection_expiry,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= (void *)&one,
@@ -92,7 +92,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "transport_expiry",
 		.data		= &rxrpc_transport_expiry,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= (void *)&one,
@@ -102,7 +102,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "rx_window_size",
 		.data		= &rxrpc_rx_window_size,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= (void *)&one,
@@ -111,7 +111,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "rx_mtu",
 		.data		= &rxrpc_rx_mtu,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= (void *)&one,
@@ -120,7 +120,7 @@ static struct ctl_table rxrpc_sysctl_table[] = {
 	{
 		.procname	= "rx_jumbo_max",
 		.data		= &rxrpc_rx_jumbo_max,
-		.maxlen		= sizeof(unsigned),
+		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= (void *)&one,
-- 
cgit v0.10.2


From 39d2adebf137de5f900843f69f5e500932e31047 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Thu, 10 Mar 2016 15:31:57 +0800
Subject: sctp: fix the transports round robin issue when init is retransmitted

prior to this patch, at the beginning if we have two paths in one assoc,
they may have the same params other than the last_time_heard, it will try
the paths like this:

1st cycle
  try trans1 fail.
  then trans2 is selected.(cause it's last_time_heard is after trans1).

2nd cycle:
  try  trans2 fail
  then trans2 is selected.(cause it's last_time_heard is after trans1).

3rd cycle:
  try  trans2 fail
  then trans2 is selected.(cause it's last_time_heard is after trans1).

....

trans1 will never have change to be selected, which is not what we expect.
we should keeping round robin all the paths if they are just added at the
beginning.

So at first every tranport's last_time_heard should be initialized 0, so
that we ensure they have the same value at the beginning, only by this,
all the transports could get equal chance to be selected.

Then for sctp_trans_elect_best, it should return the trans_next one when
*trans == *trans_next, so that we can try next if it fails,  but now it
always return trans. so we can fix it by exchanging these two params when
we calls sctp_trans_elect_tie().

Fixes: 4c47af4d5eb2 ('net: sctp: rework multihoming retransmission path selection to rfc4960')
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 2bf8ec9..cd87344 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -1263,7 +1263,7 @@ static struct sctp_transport *sctp_trans_elect_best(struct sctp_transport *curr,
 	if (score_curr > score_best)
 		return curr;
 	else if (score_curr == score_best)
-		return sctp_trans_elect_tie(curr, best);
+		return sctp_trans_elect_tie(best, curr);
 	else
 		return best;
 }
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index a431c14..d517153 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -72,7 +72,7 @@ static struct sctp_transport *sctp_transport_init(struct net *net,
 	 */
 	peer->rto = msecs_to_jiffies(net->sctp.rto_initial);
 
-	peer->last_time_heard = ktime_get();
+	peer->last_time_heard = ktime_set(0, 0);
 	peer->last_time_ecne_reduced = jiffies;
 
 	peer->param_flags = SPP_HB_DISABLE |
-- 
cgit v0.10.2


From 136ba622de49a6bf1f6e5eab3391ed5d5dbe30e3 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Thu, 10 Mar 2016 08:55:50 +0000
Subject: netconf: add macro to represent all attributes

This patch adds macro NETCONFA_ALL to represent all type of netconf
attributes for IPv4 and IPv6.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/netconf.h b/include/uapi/linux/netconf.h
index 23cbd34..45dfad5 100644
--- a/include/uapi/linux/netconf.h
+++ b/include/uapi/linux/netconf.h
@@ -19,6 +19,7 @@ enum {
 	__NETCONFA_MAX
 };
 #define NETCONFA_MAX	(__NETCONFA_MAX - 1)
+#define NETCONFA_ALL	-1
 
 #define NETCONFA_IFINDEX_ALL		-1
 #define NETCONFA_IFINDEX_DEFAULT	-2
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 8c3df2c..65e76a4 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1753,17 +1753,20 @@ static int inet_netconf_msgsize_devconf(int type)
 {
 	int size = NLMSG_ALIGN(sizeof(struct netconfmsg))
 		   + nla_total_size(4);	/* NETCONFA_IFINDEX */
+	bool all = false;
 
-	/* type -1 is used for ALL */
-	if (type == -1 || type == NETCONFA_FORWARDING)
+	if (type == NETCONFA_ALL)
+		all = true;
+
+	if (all || type == NETCONFA_FORWARDING)
 		size += nla_total_size(4);
-	if (type == -1 || type == NETCONFA_RP_FILTER)
+	if (all || type == NETCONFA_RP_FILTER)
 		size += nla_total_size(4);
-	if (type == -1 || type == NETCONFA_MC_FORWARDING)
+	if (all || type == NETCONFA_MC_FORWARDING)
 		size += nla_total_size(4);
-	if (type == -1 || type == NETCONFA_PROXY_NEIGH)
+	if (all || type == NETCONFA_PROXY_NEIGH)
 		size += nla_total_size(4);
-	if (type == -1 || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN)
+	if (all || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN)
 		size += nla_total_size(4);
 
 	return size;
@@ -1776,36 +1779,39 @@ static int inet_netconf_fill_devconf(struct sk_buff *skb, int ifindex,
 {
 	struct nlmsghdr  *nlh;
 	struct netconfmsg *ncm;
+	bool all = false;
 
 	nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct netconfmsg),
 			flags);
 	if (!nlh)
 		return -EMSGSIZE;
 
+	if (type == NETCONFA_ALL)
+		all = true;
+
 	ncm = nlmsg_data(nlh);
 	ncm->ncm_family = AF_INET;
 
 	if (nla_put_s32(skb, NETCONFA_IFINDEX, ifindex) < 0)
 		goto nla_put_failure;
 
-	/* type -1 is used for ALL */
-	if ((type == -1 || type == NETCONFA_FORWARDING) &&
+	if ((all || type == NETCONFA_FORWARDING) &&
 	    nla_put_s32(skb, NETCONFA_FORWARDING,
 			IPV4_DEVCONF(*devconf, FORWARDING)) < 0)
 		goto nla_put_failure;
-	if ((type == -1 || type == NETCONFA_RP_FILTER) &&
+	if ((all || type == NETCONFA_RP_FILTER) &&
 	    nla_put_s32(skb, NETCONFA_RP_FILTER,
 			IPV4_DEVCONF(*devconf, RP_FILTER)) < 0)
 		goto nla_put_failure;
-	if ((type == -1 || type == NETCONFA_MC_FORWARDING) &&
+	if ((all || type == NETCONFA_MC_FORWARDING) &&
 	    nla_put_s32(skb, NETCONFA_MC_FORWARDING,
 			IPV4_DEVCONF(*devconf, MC_FORWARDING)) < 0)
 		goto nla_put_failure;
-	if ((type == -1 || type == NETCONFA_PROXY_NEIGH) &&
+	if ((all || type == NETCONFA_PROXY_NEIGH) &&
 	    nla_put_s32(skb, NETCONFA_PROXY_NEIGH,
 			IPV4_DEVCONF(*devconf, PROXY_ARP)) < 0)
 		goto nla_put_failure;
-	if ((type == -1 || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN) &&
+	if ((all || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN) &&
 	    nla_put_s32(skb, NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN,
 			IPV4_DEVCONF(*devconf, IGNORE_ROUTES_WITH_LINKDOWN)) < 0)
 		goto nla_put_failure;
@@ -1893,14 +1899,14 @@ static int inet_netconf_get_devconf(struct sk_buff *in_skb,
 	}
 
 	err = -ENOBUFS;
-	skb = nlmsg_new(inet_netconf_msgsize_devconf(-1), GFP_ATOMIC);
+	skb = nlmsg_new(inet_netconf_msgsize_devconf(NETCONFA_ALL), GFP_ATOMIC);
 	if (!skb)
 		goto errout;
 
 	err = inet_netconf_fill_devconf(skb, ifindex, devconf,
 					NETLINK_CB(in_skb).portid,
 					nlh->nlmsg_seq, RTM_NEWNETCONF, 0,
-					-1);
+					NETCONFA_ALL);
 	if (err < 0) {
 		/* -EMSGSIZE implies BUG in inet_netconf_msgsize_devconf() */
 		WARN_ON(err == -EMSGSIZE);
@@ -1944,7 +1950,7 @@ static int inet_netconf_dump_devconf(struct sk_buff *skb,
 						      cb->nlh->nlmsg_seq,
 						      RTM_NEWNETCONF,
 						      NLM_F_MULTI,
-						      -1) < 0) {
+						      NETCONFA_ALL) < 0) {
 				rcu_read_unlock();
 				goto done;
 			}
@@ -1960,7 +1966,7 @@ cont:
 					      NETLINK_CB(cb->skb).portid,
 					      cb->nlh->nlmsg_seq,
 					      RTM_NEWNETCONF, NLM_F_MULTI,
-					      -1) < 0)
+					      NETCONFA_ALL) < 0)
 			goto done;
 		else
 			h++;
@@ -1971,7 +1977,7 @@ cont:
 					      NETLINK_CB(cb->skb).portid,
 					      cb->nlh->nlmsg_seq,
 					      RTM_NEWNETCONF, NLM_F_MULTI,
-					      -1) < 0)
+					      NETCONFA_ALL) < 0)
 			goto done;
 		else
 			h++;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8c0dab2..27aed1a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -473,18 +473,21 @@ static int inet6_netconf_msgsize_devconf(int type)
 {
 	int size =  NLMSG_ALIGN(sizeof(struct netconfmsg))
 		    + nla_total_size(4);	/* NETCONFA_IFINDEX */
+	bool all = false;
 
-	/* type -1 is used for ALL */
-	if (type == -1 || type == NETCONFA_FORWARDING)
+	if (type == NETCONFA_ALL)
+		all = true;
+
+	if (all || type == NETCONFA_FORWARDING)
 		size += nla_total_size(4);
 #ifdef CONFIG_IPV6_MROUTE
-	if (type == -1 || type == NETCONFA_MC_FORWARDING)
+	if (all || type == NETCONFA_MC_FORWARDING)
 		size += nla_total_size(4);
 #endif
-	if (type == -1 || type == NETCONFA_PROXY_NEIGH)
+	if (all || type == NETCONFA_PROXY_NEIGH)
 		size += nla_total_size(4);
 
-	if (type == -1 || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN)
+	if (all || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN)
 		size += nla_total_size(4);
 
 	return size;
@@ -497,33 +500,36 @@ static int inet6_netconf_fill_devconf(struct sk_buff *skb, int ifindex,
 {
 	struct nlmsghdr  *nlh;
 	struct netconfmsg *ncm;
+	bool all = false;
 
 	nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct netconfmsg),
 			flags);
 	if (!nlh)
 		return -EMSGSIZE;
 
+	if (type == NETCONFA_ALL)
+		all = true;
+
 	ncm = nlmsg_data(nlh);
 	ncm->ncm_family = AF_INET6;
 
 	if (nla_put_s32(skb, NETCONFA_IFINDEX, ifindex) < 0)
 		goto nla_put_failure;
 
-	/* type -1 is used for ALL */
-	if ((type == -1 || type == NETCONFA_FORWARDING) &&
+	if ((all || type == NETCONFA_FORWARDING) &&
 	    nla_put_s32(skb, NETCONFA_FORWARDING, devconf->forwarding) < 0)
 		goto nla_put_failure;
 #ifdef CONFIG_IPV6_MROUTE
-	if ((type == -1 || type == NETCONFA_MC_FORWARDING) &&
+	if ((all || type == NETCONFA_MC_FORWARDING) &&
 	    nla_put_s32(skb, NETCONFA_MC_FORWARDING,
 			devconf->mc_forwarding) < 0)
 		goto nla_put_failure;
 #endif
-	if ((type == -1 || type == NETCONFA_PROXY_NEIGH) &&
+	if ((all || type == NETCONFA_PROXY_NEIGH) &&
 	    nla_put_s32(skb, NETCONFA_PROXY_NEIGH, devconf->proxy_ndp) < 0)
 		goto nla_put_failure;
 
-	if ((type == -1 || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN) &&
+	if ((all || type == NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN) &&
 	    nla_put_s32(skb, NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN,
 			devconf->ignore_routes_with_linkdown) < 0)
 		goto nla_put_failure;
@@ -609,14 +615,14 @@ static int inet6_netconf_get_devconf(struct sk_buff *in_skb,
 	}
 
 	err = -ENOBUFS;
-	skb = nlmsg_new(inet6_netconf_msgsize_devconf(-1), GFP_ATOMIC);
+	skb = nlmsg_new(inet6_netconf_msgsize_devconf(NETCONFA_ALL), GFP_ATOMIC);
 	if (!skb)
 		goto errout;
 
 	err = inet6_netconf_fill_devconf(skb, ifindex, devconf,
 					 NETLINK_CB(in_skb).portid,
 					 nlh->nlmsg_seq, RTM_NEWNETCONF, 0,
-					 -1);
+					 NETCONFA_ALL);
 	if (err < 0) {
 		/* -EMSGSIZE implies BUG in inet6_netconf_msgsize_devconf() */
 		WARN_ON(err == -EMSGSIZE);
@@ -660,7 +666,7 @@ static int inet6_netconf_dump_devconf(struct sk_buff *skb,
 						       cb->nlh->nlmsg_seq,
 						       RTM_NEWNETCONF,
 						       NLM_F_MULTI,
-						       -1) < 0) {
+						       NETCONFA_ALL) < 0) {
 				rcu_read_unlock();
 				goto done;
 			}
@@ -676,7 +682,7 @@ cont:
 					       NETLINK_CB(cb->skb).portid,
 					       cb->nlh->nlmsg_seq,
 					       RTM_NEWNETCONF, NLM_F_MULTI,
-					       -1) < 0)
+					       NETCONFA_ALL) < 0)
 			goto done;
 		else
 			h++;
@@ -687,7 +693,7 @@ cont:
 					       NETLINK_CB(cb->skb).portid,
 					       cb->nlh->nlmsg_seq,
 					       RTM_NEWNETCONF, NLM_F_MULTI,
-					       -1) < 0)
+					       NETCONFA_ALL) < 0)
 			goto done;
 		else
 			h++;
-- 
cgit v0.10.2


From 6b8abef5f833b03be1b5af491193477ad609ad35 Mon Sep 17 00:00:00 2001
From: Paul Durrant <Paul.Durrant@citrix.com>
Date: Thu, 10 Mar 2016 12:30:26 +0000
Subject: xen-netback: re-import canonical netif header

The canonical netif header (in the Xen source repo) and the Linux variant
have diverged significantly. Recently much documentation has been added to
the canonical header which is highly useful for developers making
modifications to either xen-netfront or xen-netback. This patch therefore
re-imports the canonical header in its entirity.

To maintain compatibility and some style consistency with the old Linux
variant, the header was stripped of its emacs boilerplate, and
post-processed and copied into place with the following commands:

ed -s netif.h << EOF
H
,s/NETTXF_/XEN_NETTXF_/g
,s/NETRXF_/XEN_NETRXF_/g
,s/NETIF_/XEN_NETIF_/g
,s/XEN_XEN_/XEN_/g
,s/netif/xen_netif/g
,s/xen_xen_/xen_/g
,s/^typedef.*$//g
,s/^    /${TAB}/g
w
$
w
EOF

indent --line-length 80 --linux-style netif.h \
-o include/xen/interface/io/netif.h

Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: David Vrabel <david.vrabel@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/xen/interface/io/netif.h b/include/xen/interface/io/netif.h
index 252ffd4..4f20dbc 100644
--- a/include/xen/interface/io/netif.h
+++ b/include/xen/interface/io/netif.h
@@ -1,16 +1,34 @@
 /******************************************************************************
- * netif.h
+ * xen_netif.h
  *
  * Unified network-device I/O interface for Xen guest OSes.
  *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
  * Copyright (c) 2003-2004, Keir Fraser
  */
 
-#ifndef __XEN_PUBLIC_IO_NETIF_H__
-#define __XEN_PUBLIC_IO_NETIF_H__
+#ifndef __XEN_PUBLIC_IO_XEN_NETIF_H__
+#define __XEN_PUBLIC_IO_XEN_NETIF_H__
 
-#include <xen/interface/io/ring.h>
-#include <xen/interface/grant_table.h>
+#include "ring.h"
+#include "../grant_table.h"
 
 /*
  * Older implementation of Xen network frontend / backend has an
@@ -38,10 +56,10 @@
  * that it cannot safely queue packets (as it may not be kicked to send them).
  */
 
- /*
+/*
  * "feature-split-event-channels" is introduced to separate guest TX
- * and RX notificaion. Backend either doesn't support this feature or
- * advertise it via xenstore as 0 (disabled) or 1 (enabled).
+ * and RX notification. Backend either doesn't support this feature or
+ * advertises it via xenstore as 0 (disabled) or 1 (enabled).
  *
  * To make use of this feature, frontend should allocate two event
  * channels for TX and RX, advertise them to backend as
@@ -118,151 +136,804 @@
  */
 
 /*
- * This is the 'wire' format for packets:
- *  Request 1: xen_netif_tx_request  -- XEN_NETTXF_* (any flags)
- * [Request 2: xen_netif_extra_info]    (only if request 1 has XEN_NETTXF_extra_info)
- * [Request 3: xen_netif_extra_info]    (only if request 2 has XEN_NETIF_EXTRA_MORE)
- *  Request 4: xen_netif_tx_request  -- XEN_NETTXF_more_data
- *  Request 5: xen_netif_tx_request  -- XEN_NETTXF_more_data
+ * "feature-multicast-control" and "feature-dynamic-multicast-control"
+ * advertise the capability to filter ethernet multicast packets in the
+ * backend. If the frontend wishes to take advantage of this feature then
+ * it may set "request-multicast-control". If the backend only advertises
+ * "feature-multicast-control" then "request-multicast-control" must be set
+ * before the frontend moves into the connected state. The backend will
+ * sample the value on this state transition and any subsequent change in
+ * value will have no effect. However, if the backend also advertises
+ * "feature-dynamic-multicast-control" then "request-multicast-control"
+ * may be set by the frontend at any time. In this case, the backend will
+ * watch the value and re-sample on watch events.
+ *
+ * If the sampled value of "request-multicast-control" is set then the
+ * backend transmit side should no longer flood multicast packets to the
+ * frontend, it should instead drop any multicast packet that does not
+ * match in a filter list.
+ * The list is amended by the frontend by sending dummy transmit requests
+ * containing XEN_NETIF_EXTRA_TYPE_MCAST_{ADD,DEL} extra-info fragments as
+ * specified below.
+ * Note that the filter list may be amended even if the sampled value of
+ * "request-multicast-control" is not set, however the filter should only
+ * be applied if it is set.
+ */
+
+/*
+ * Control ring
+ * ============
+ *
+ * Some features, such as hashing (detailed below), require a
+ * significant amount of out-of-band data to be passed from frontend to
+ * backend. Use of xenstore is not suitable for large quantities of data
+ * because of quota limitations and so a dedicated 'control ring' is used.
+ * The ability of the backend to use a control ring is advertised by
+ * setting:
+ *
+ * /local/domain/X/backend/<domid>/<vif>/feature-ctrl-ring = "1"
+ *
+ * The frontend provides a control ring to the backend by setting:
+ *
+ * /local/domain/<domid>/device/vif/<vif>/ctrl-ring-ref = <gref>
+ * /local/domain/<domid>/device/vif/<vif>/event-channel-ctrl = <port>
+ *
+ * where <gref> is the grant reference of the shared page used to
+ * implement the control ring and <port> is an event channel to be used
+ * as a mailbox interrupt. These keys must be set before the frontend
+ * moves into the connected state.
+ *
+ * The control ring uses a fixed request/response message size and is
+ * balanced (i.e. one request to one response), so operationally it is much
+ * the same as a transmit or receive ring.
+ * Note that there is no requirement that responses are issued in the same
+ * order as requests.
+ */
+
+/*
+ * Hash types
+ * ==========
+ *
+ * For the purposes of the definitions below, 'Packet[]' is an array of
+ * octets containing an IP packet without options, 'Array[X..Y]' means a
+ * sub-array of 'Array' containing bytes X thru Y inclusive, and '+' is
+ * used to indicate concatenation of arrays.
+ */
+
+/*
+ * A hash calculated over an IP version 4 header as follows:
+ *
+ * Buffer[0..8] = Packet[12..15] (source address) +
+ *                Packet[16..19] (destination address)
+ *
+ * Result = Hash(Buffer, 8)
+ */
+#define _XEN_NETIF_CTRL_HASH_TYPE_IPV4 0
+#define XEN_NETIF_CTRL_HASH_TYPE_IPV4 \
+	(1 << _XEN_NETIF_CTRL_HASH_TYPE_IPV4)
+
+/*
+ * A hash calculated over an IP version 4 header and TCP header as
+ * follows:
+ *
+ * Buffer[0..12] = Packet[12..15] (source address) +
+ *                 Packet[16..19] (destination address) +
+ *                 Packet[20..21] (source port) +
+ *                 Packet[22..23] (destination port)
+ *
+ * Result = Hash(Buffer, 12)
+ */
+#define _XEN_NETIF_CTRL_HASH_TYPE_IPV4_TCP 1
+#define XEN_NETIF_CTRL_HASH_TYPE_IPV4_TCP \
+	(1 << _XEN_NETIF_CTRL_HASH_TYPE_IPV4_TCP)
+
+/*
+ * A hash calculated over an IP version 6 header as follows:
+ *
+ * Buffer[0..32] = Packet[8..23]  (source address ) +
+ *                 Packet[24..39] (destination address)
+ *
+ * Result = Hash(Buffer, 32)
+ */
+#define _XEN_NETIF_CTRL_HASH_TYPE_IPV6 2
+#define XEN_NETIF_CTRL_HASH_TYPE_IPV6 \
+	(1 << _XEN_NETIF_CTRL_HASH_TYPE_IPV6)
+
+/*
+ * A hash calculated over an IP version 6 header and TCP header as
+ * follows:
+ *
+ * Buffer[0..36] = Packet[8..23]  (source address) +
+ *                 Packet[24..39] (destination address) +
+ *                 Packet[40..41] (source port) +
+ *                 Packet[42..43] (destination port)
+ *
+ * Result = Hash(Buffer, 36)
+ */
+#define _XEN_NETIF_CTRL_HASH_TYPE_IPV6_TCP 3
+#define XEN_NETIF_CTRL_HASH_TYPE_IPV6_TCP \
+	(1 << _XEN_NETIF_CTRL_HASH_TYPE_IPV6_TCP)
+
+/*
+ * Hash algorithms
+ * ===============
+ */
+
+#define XEN_NETIF_CTRL_HASH_ALGORITHM_NONE 0
+
+/*
+ * Toeplitz hash:
+ */
+
+#define XEN_NETIF_CTRL_HASH_ALGORITHM_TOEPLITZ 1
+
+/*
+ * This algorithm uses a 'key' as well as the data buffer itself.
+ * (Buffer[] and Key[] are treated as shift-registers where the MSB of
+ * Buffer/Key[0] is considered 'left-most' and the LSB of Buffer/Key[N-1]
+ * is the 'right-most').
+ *
+ * Value = 0
+ * For number of bits in Buffer[]
+ *    If (left-most bit of Buffer[] is 1)
+ *        Value ^= left-most 32 bits of Key[]
+ *    Key[] << 1
+ *    Buffer[] << 1
+ *
+ * The code below is provided for convenience where an operating system
+ * does not already provide an implementation.
+ */
+#ifdef XEN_NETIF_DEFINE_TOEPLITZ
+static uint32_t xen_netif_toeplitz_hash(const uint8_t *key,
+					unsigned int keylen,
+					const uint8_t *buf, unsigned int buflen)
+{
+	unsigned int keyi, bufi;
+	uint64_t prefix = 0;
+	uint64_t hash = 0;
+
+	/* Pre-load prefix with the first 8 bytes of the key */
+	for (keyi = 0; keyi < 8; keyi++) {
+		prefix <<= 8;
+		prefix |= (keyi < keylen) ? key[keyi] : 0;
+	}
+
+	for (bufi = 0; bufi < buflen; bufi++) {
+		uint8_t byte = buf[bufi];
+		unsigned int bit;
+
+		for (bit = 0; bit < 8; bit++) {
+			if (byte & 0x80)
+				hash ^= prefix;
+			prefix <<= 1;
+			byte <<= 1;
+		}
+
+		/*
+		 * 'prefix' has now been left-shifted by 8, so
+		 * OR in the next byte.
+		 */
+		prefix |= (keyi < keylen) ? key[keyi] : 0;
+		keyi++;
+	}
+
+	/* The valid part of the hash is in the upper 32 bits. */
+	return hash >> 32;
+}
+#endif				/* XEN_NETIF_DEFINE_TOEPLITZ */
+
+/*
+ * Control requests (struct xen_netif_ctrl_request)
+ * ================================================
+ *
+ * All requests have the following format:
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |    id     |   type    |         data[0]       |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |         data[1]       |         data[2]       |
+ * +-----+-----+-----+-----+-----------------------+
+ *
+ * id: the request identifier, echoed in response.
+ * type: the type of request (see below)
+ * data[]: any data associated with the request (determined by type)
+ */
+
+struct xen_netif_ctrl_request {
+	uint16_t id;
+	uint16_t type;
+
+#define XEN_NETIF_CTRL_TYPE_INVALID               0
+#define XEN_NETIF_CTRL_TYPE_GET_HASH_FLAGS        1
+#define XEN_NETIF_CTRL_TYPE_SET_HASH_FLAGS        2
+#define XEN_NETIF_CTRL_TYPE_SET_HASH_KEY          3
+#define XEN_NETIF_CTRL_TYPE_GET_HASH_MAPPING_SIZE 4
+#define XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING_SIZE 5
+#define XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING      6
+#define XEN_NETIF_CTRL_TYPE_SET_HASH_ALGORITHM    7
+
+	uint32_t data[3];
+};
+
+/*
+ * Control responses (struct xen_netif_ctrl_response)
+ * ==================================================
+ *
+ * All responses have the following format:
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |    id     |   type    |         status        |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |         data          |
+ * +-----+-----+-----+-----+
+ *
+ * id: the corresponding request identifier
+ * type: the type of the corresponding request
+ * status: the status of request processing
+ * data: any data associated with the response (determined by type and
+ *       status)
+ */
+
+struct xen_netif_ctrl_response {
+	uint16_t id;
+	uint16_t type;
+	uint32_t status;
+
+#define XEN_NETIF_CTRL_STATUS_SUCCESS           0
+#define XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED     1
+#define XEN_NETIF_CTRL_STATUS_INVALID_PARAMETER 2
+#define XEN_NETIF_CTRL_STATUS_BUFFER_OVERFLOW   3
+
+	uint32_t data;
+};
+
+/*
+ * Control messages
+ * ================
+ *
+ * XEN_NETIF_CTRL_TYPE_SET_HASH_ALGORITHM
+ * --------------------------------------
+ *
+ * This is sent by the frontend to set the desired hash algorithm.
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_SET_HASH_ALGORITHM
+ *  data[0] = a XEN_NETIF_CTRL_HASH_ALGORITHM_* value
+ *  data[1] = 0
+ *  data[2] = 0
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED     - Operation not
+ *                                                     supported
+ *           XEN_NETIF_CTRL_STATUS_INVALID_PARAMETER - The algorithm is not
+ *                                                     supported
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS           - Operation successful
+ *
+ * NOTE: Setting data[0] to XEN_NETIF_CTRL_HASH_ALGORITHM_NONE disables
+ *       hashing and the backend is free to choose how it steers packets
+ *       to queues (which is the default behaviour).
+ *
+ * XEN_NETIF_CTRL_TYPE_GET_HASH_FLAGS
+ * ----------------------------------
+ *
+ * This is sent by the frontend to query the types of hash supported by
+ * the backend.
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_GET_HASH_FLAGS
+ *  data[0] = 0
+ *  data[1] = 0
+ *  data[2] = 0
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED - Operation not supported
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS       - Operation successful
+ *  data   = supported hash types (if operation was successful)
+ *
+ * NOTE: A valid hash algorithm must be selected before this operation can
+ *       succeed.
+ *
+ * XEN_NETIF_CTRL_TYPE_SET_HASH_FLAGS
+ * ----------------------------------
+ *
+ * This is sent by the frontend to set the types of hash that the backend
+ * should calculate. (See above for hash type definitions).
+ * Note that the 'maximal' type of hash should always be chosen. For
+ * example, if the frontend sets both IPV4 and IPV4_TCP hash types then
+ * the latter hash type should be calculated for any TCP packet and the
+ * former only calculated for non-TCP packets.
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_SET_HASH_FLAGS
+ *  data[0] = bitwise OR of XEN_NETIF_CTRL_HASH_TYPE_* values
+ *  data[1] = 0
+ *  data[2] = 0
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED     - Operation not
+ *                                                     supported
+ *           XEN_NETIF_CTRL_STATUS_INVALID_PARAMETER - One or more flag
+ *                                                     value is invalid or
+ *                                                     unsupported
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS           - Operation successful
+ *  data   = 0
+ *
+ * NOTE: A valid hash algorithm must be selected before this operation can
+ *       succeed.
+ *       Also, setting data[0] to zero disables hashing and the backend
+ *       is free to choose how it steers packets to queues.
+ *
+ * XEN_NETIF_CTRL_TYPE_SET_HASH_KEY
+ * --------------------------------
+ *
+ * This is sent by the frontend to set the key of the hash if the algorithm
+ * requires it. (See hash algorithms above).
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_SET_HASH_KEY
+ *  data[0] = grant reference of page containing the key (assumed to
+ *            start at beginning of grant)
+ *  data[1] = size of key in octets
+ *  data[2] = 0
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED     - Operation not
+ *                                                     supported
+ *           XEN_NETIF_CTRL_STATUS_INVALID_PARAMETER - Key size is invalid
+ *           XEN_NETIF_CTRL_STATUS_BUFFER_OVERFLOW   - Key size is larger
+ *                                                     than the backend
+ *                                                     supports
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS           - Operation successful
+ *  data   = 0
+ *
+ * NOTE: Any key octets not specified are assumed to be zero (the key
+ *       is assumed to be empty by default) and specifying a new key
+ *       invalidates any previous key, hence specifying a key size of
+ *       zero will clear the key (which ensures that the calculated hash
+ *       will always be zero).
+ *       The maximum size of key is algorithm and backend specific, but
+ *       is also limited by the single grant reference.
+ *       The grant reference may be read-only and must remain valid until
+ *       the response has been processed.
+ *
+ * XEN_NETIF_CTRL_TYPE_GET_HASH_MAPPING_SIZE
+ * -----------------------------------------
+ *
+ * This is sent by the frontend to query the maximum size of mapping
+ * table supported by the backend. The size is specified in terms of
+ * table entries.
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_GET_HASH_MAPPING_SIZE
+ *  data[0] = 0
+ *  data[1] = 0
+ *  data[2] = 0
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED - Operation not supported
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS       - Operation successful
+ *  data   = maximum number of entries allowed in the mapping table
+ *           (if operation was successful) or zero if a mapping table is
+ *           not supported (i.e. hash mapping is done only by modular
+ *           arithmetic).
+ *
+ * XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING_SIZE
+ * -------------------------------------
+ *
+ * This is sent by the frontend to set the actual size of the mapping
+ * table to be used by the backend. The size is specified in terms of
+ * table entries.
+ * Any previous table is invalidated by this message and any new table
+ * is assumed to be zero filled.
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING_SIZE
+ *  data[0] = number of entries in mapping table
+ *  data[1] = 0
+ *  data[2] = 0
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED     - Operation not
+ *                                                     supported
+ *           XEN_NETIF_CTRL_STATUS_INVALID_PARAMETER - Table size is invalid
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS           - Operation successful
+ *  data   = 0
+ *
+ * NOTE: Setting data[0] to 0 means that hash mapping should be done
+ *       using modular arithmetic.
+ *
+ * XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING
+ * ------------------------------------
+ *
+ * This is sent by the frontend to set the content of the table mapping
+ * hash value to queue number. The backend should calculate the hash from
+ * the packet header, use it as an index into the table (modulo the size
+ * of the table) and then steer the packet to the queue number found at
+ * that index.
+ *
+ * Request:
+ *
+ *  type    = XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING
+ *  data[0] = grant reference of page containing the mapping (sub-)table
+ *            (assumed to start at beginning of grant)
+ *  data[1] = size of (sub-)table in entries
+ *  data[2] = offset, in entries, of sub-table within overall table
+ *
+ * Response:
+ *
+ *  status = XEN_NETIF_CTRL_STATUS_NOT_SUPPORTED     - Operation not
+ *                                                     supported
+ *           XEN_NETIF_CTRL_STATUS_INVALID_PARAMETER - Table size or content
+ *                                                     is invalid
+ *           XEN_NETIF_CTRL_STATUS_BUFFER_OVERFLOW   - Table size is larger
+ *                                                     than the backend
+ *                                                     supports
+ *           XEN_NETIF_CTRL_STATUS_SUCCESS           - Operation successful
+ *  data   = 0
+ *
+ * NOTE: The overall table has the following format:
+ *
+ *          0     1     2     3     4     5     6     7  octet
+ *       +-----+-----+-----+-----+-----+-----+-----+-----+
+ *       |       mapping[0]      |       mapping[1]      |
+ *       +-----+-----+-----+-----+-----+-----+-----+-----+
+ *       |                       .                       |
+ *       |                       .                       |
+ *       |                       .                       |
+ *       +-----+-----+-----+-----+-----+-----+-----+-----+
+ *       |      mapping[N-2]     |      mapping[N-1]     |
+ *       +-----+-----+-----+-----+-----+-----+-----+-----+
+ *
+ *       where N is specified by a XEN_NETIF_CTRL_TYPE_SET_HASH_MAPPING_SIZE
+ *       message and each  mapping must specifies a queue between 0 and
+ *       "multi-queue-num-queues" (see above).
+ *       The backend may support a mapping table larger than can be
+ *       mapped by a single grant reference. Thus sub-tables within a
+ *       larger table can be individually set by sending multiple messages
+ *       with differing offset values. Specifying a new sub-table does not
+ *       invalidate any table data outside that range.
+ *       The grant reference may be read-only and must remain valid until
+ *       the response has been processed.
+ */
+
+DEFINE_RING_TYPES(xen_netif_ctrl,
+		  struct xen_netif_ctrl_request,
+		  struct xen_netif_ctrl_response);
+
+/*
+ * Guest transmit
+ * ==============
+ *
+ * This is the 'wire' format for transmit (frontend -> backend) packets:
+ *
+ *  Fragment 1: xen_netif_tx_request_t  - flags = XEN_NETTXF_*
+ *                                    size = total packet size
+ * [Extra 1: xen_netif_extra_info_t]    - (only if fragment 1 flags include
+ *                                     XEN_NETTXF_extra_info)
+ *  ...
+ * [Extra N: xen_netif_extra_info_t]    - (only if extra N-1 flags include
+ *                                     XEN_NETIF_EXTRA_MORE)
  *  ...
- *  Request N: xen_netif_tx_request  -- 0
+ *  Fragment N: xen_netif_tx_request_t  - (only if fragment N-1 flags include
+ *                                     XEN_NETTXF_more_data - flags on preceding
+ *                                     extras are not relevant here)
+ *                                    flags = 0
+ *                                    size = fragment size
+ *
+ * NOTE:
+ *
+ * This format slightly is different from that used for receive
+ * (backend -> frontend) packets. Specifically, in a multi-fragment
+ * packet the actual size of fragment 1 can only be determined by
+ * subtracting the sizes of fragments 2..N from the total packet size.
+ *
+ * Ring slot size is 12 octets, however not all request/response
+ * structs use the full size.
+ *
+ * tx request data (xen_netif_tx_request_t)
+ * ------------------------------------
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | grant ref             | offset    | flags     |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | id        | size      |
+ * +-----+-----+-----+-----+
+ *
+ * grant ref: Reference to buffer page.
+ * offset: Offset within buffer page.
+ * flags: XEN_NETTXF_*.
+ * id: request identifier, echoed in response.
+ * size: packet size in bytes.
+ *
+ * tx response (xen_netif_tx_response_t)
+ * ---------------------------------
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | id        | status    | unused                |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | unused                |
+ * +-----+-----+-----+-----+
+ *
+ * id: reflects id in transmit request
+ * status: XEN_NETIF_RSP_*
+ *
+ * Guest receive
+ * =============
+ *
+ * This is the 'wire' format for receive (backend -> frontend) packets:
+ *
+ *  Fragment 1: xen_netif_rx_request_t  - flags = XEN_NETRXF_*
+ *                                    size = fragment size
+ * [Extra 1: xen_netif_extra_info_t]    - (only if fragment 1 flags include
+ *                                     XEN_NETRXF_extra_info)
+ *  ...
+ * [Extra N: xen_netif_extra_info_t]    - (only if extra N-1 flags include
+ *                                     XEN_NETIF_EXTRA_MORE)
+ *  ...
+ *  Fragment N: xen_netif_rx_request_t  - (only if fragment N-1 flags include
+ *                                     XEN_NETRXF_more_data - flags on preceding
+ *                                     extras are not relevant here)
+ *                                    flags = 0
+ *                                    size = fragment size
+ *
+ * NOTE:
+ *
+ * This format slightly is different from that used for transmit
+ * (frontend -> backend) packets. Specifically, in a multi-fragment
+ * packet the size of the packet can only be determined by summing the
+ * sizes of fragments 1..N.
+ *
+ * Ring slot size is 8 octets.
+ *
+ * rx request (xen_netif_rx_request_t)
+ * -------------------------------
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | id        | pad       | gref                  |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ *
+ * id: request identifier, echoed in response.
+ * gref: reference to incoming granted frame.
+ *
+ * rx response (xen_netif_rx_response_t)
+ * ---------------------------------
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | id        | offset    | flags     | status    |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ *
+ * id: reflects id in receive request
+ * offset: offset in page of start of received packet
+ * flags: XEN_NETRXF_*
+ * status: -ve: XEN_NETIF_RSP_*; +ve: Rx'ed pkt size.
+ *
+ * NOTE: Historically, to support GSO on the frontend receive side, Linux
+ *       netfront does not make use of the rx response id (because, as
+ *       described below, extra info structures overlay the id field).
+ *       Instead it assumes that responses always appear in the same ring
+ *       slot as their corresponding request. Thus, to maintain
+ *       compatibility, backends must make sure this is the case.
+ *
+ * Extra Info
+ * ==========
+ *
+ * Can be present if initial request or response has NET{T,R}XF_extra_info,
+ * or previous extra request has XEN_NETIF_EXTRA_MORE.
+ *
+ * The struct therefore needs to fit into either a tx or rx slot and
+ * is therefore limited to 8 octets.
+ *
+ * NOTE: Because extra info data overlays the usual request/response
+ *       structures, there is no id information in the opposite direction.
+ *       So, if an extra info overlays an rx response the frontend can
+ *       assume that it is in the same ring slot as the request that was
+ *       consumed to make the slot available, and the backend must ensure
+ *       this assumption is true.
+ *
+ * extra info (xen_netif_extra_info_t)
+ * -------------------------------
+ *
+ * General format:
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |type |flags| type specific data                |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * | padding for tx        |
+ * +-----+-----+-----+-----+
+ *
+ * type: XEN_NETIF_EXTRA_TYPE_*
+ * flags: XEN_NETIF_EXTRA_FLAG_*
+ * padding for tx: present only in the tx case due to 8 octet limit
+ *                 from rx case. Not shown in type specific entries
+ *                 below.
+ *
+ * XEN_NETIF_EXTRA_TYPE_GSO:
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |type |flags| size      |type | pad | features  |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ *
+ * type: Must be XEN_NETIF_EXTRA_TYPE_GSO
+ * flags: XEN_NETIF_EXTRA_FLAG_*
+ * size: Maximum payload size of each segment. For example,
+ *       for TCP this is just the path MSS.
+ * type: XEN_NETIF_GSO_TYPE_*: This determines the protocol of
+ *       the packet and any extra features required to segment the
+ *       packet properly.
+ * features: EN_XEN_NETIF_GSO_FEAT_*: This specifies any extra GSO
+ *           features required to process this packet, such as ECN
+ *           support for TCPv4.
+ *
+ * XEN_NETIF_EXTRA_TYPE_MCAST_{ADD,DEL}:
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |type |flags| addr                              |
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ *
+ * type: Must be XEN_NETIF_EXTRA_TYPE_MCAST_{ADD,DEL}
+ * flags: XEN_NETIF_EXTRA_FLAG_*
+ * addr: address to add/remove
+ *
+ * XEN_NETIF_EXTRA_TYPE_HASH:
+ *
+ * A backend that supports teoplitz hashing is assumed to accept
+ * this type of extra info in transmit packets.
+ * A frontend that enables hashing is assumed to accept
+ * this type of extra info in receive packets.
+ *
+ *    0     1     2     3     4     5     6     7  octet
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ * |type |flags|htype| alg |LSB ---- value ---- MSB|
+ * +-----+-----+-----+-----+-----+-----+-----+-----+
+ *
+ * type: Must be XEN_NETIF_EXTRA_TYPE_HASH
+ * flags: XEN_NETIF_EXTRA_FLAG_*
+ * htype: Hash type (one of _XEN_NETIF_CTRL_HASH_TYPE_* - see above)
+ * alg: The algorithm used to calculate the hash (one of
+ *      XEN_NETIF_CTRL_HASH_TYPE_ALGORITHM_* - see above)
+ * value: Hash value
  */
 
 /* Protocol checksum field is blank in the packet (hardware offload)? */
-#define _XEN_NETTXF_csum_blank		(0)
-#define  XEN_NETTXF_csum_blank		(1U<<_XEN_NETTXF_csum_blank)
+#define _XEN_NETTXF_csum_blank     (0)
+#define  XEN_NETTXF_csum_blank     (1U<<_XEN_NETTXF_csum_blank)
 
 /* Packet data has been validated against protocol checksum. */
-#define _XEN_NETTXF_data_validated	(1)
-#define  XEN_NETTXF_data_validated	(1U<<_XEN_NETTXF_data_validated)
+#define _XEN_NETTXF_data_validated (1)
+#define  XEN_NETTXF_data_validated (1U<<_XEN_NETTXF_data_validated)
 
 /* Packet continues in the next request descriptor. */
-#define _XEN_NETTXF_more_data		(2)
-#define  XEN_NETTXF_more_data		(1U<<_XEN_NETTXF_more_data)
+#define _XEN_NETTXF_more_data      (2)
+#define  XEN_NETTXF_more_data      (1U<<_XEN_NETTXF_more_data)
 
 /* Packet to be followed by extra descriptor(s). */
-#define _XEN_NETTXF_extra_info		(3)
-#define  XEN_NETTXF_extra_info		(1U<<_XEN_NETTXF_extra_info)
+#define _XEN_NETTXF_extra_info     (3)
+#define  XEN_NETTXF_extra_info     (1U<<_XEN_NETTXF_extra_info)
 
 #define XEN_NETIF_MAX_TX_SIZE 0xFFFF
 struct xen_netif_tx_request {
-    grant_ref_t gref;      /* Reference to buffer page */
-    uint16_t offset;       /* Offset within buffer page */
-    uint16_t flags;        /* XEN_NETTXF_* */
-    uint16_t id;           /* Echoed in response message. */
-    uint16_t size;         /* Packet size in bytes.       */
+	grant_ref_t gref;
+	uint16_t offset;
+	uint16_t flags;
+	uint16_t id;
+	uint16_t size;
 };
 
 /* Types of xen_netif_extra_info descriptors. */
-#define XEN_NETIF_EXTRA_TYPE_NONE	(0)  /* Never used - invalid */
-#define XEN_NETIF_EXTRA_TYPE_GSO	(1)  /* u.gso */
-#define XEN_NETIF_EXTRA_TYPE_MCAST_ADD	(2)  /* u.mcast */
-#define XEN_NETIF_EXTRA_TYPE_MCAST_DEL	(3)  /* u.mcast */
-#define XEN_NETIF_EXTRA_TYPE_MAX	(4)
+#define XEN_NETIF_EXTRA_TYPE_NONE      (0)	/* Never used - invalid */
+#define XEN_NETIF_EXTRA_TYPE_GSO       (1)	/* u.gso */
+#define XEN_NETIF_EXTRA_TYPE_MCAST_ADD (2)	/* u.mcast */
+#define XEN_NETIF_EXTRA_TYPE_MCAST_DEL (3)	/* u.mcast */
+#define XEN_NETIF_EXTRA_TYPE_HASH      (4)	/* u.hash */
+#define XEN_NETIF_EXTRA_TYPE_MAX       (5)
 
-/* xen_netif_extra_info flags. */
-#define _XEN_NETIF_EXTRA_FLAG_MORE	(0)
-#define  XEN_NETIF_EXTRA_FLAG_MORE	(1U<<_XEN_NETIF_EXTRA_FLAG_MORE)
+/* xen_netif_extra_info_t flags. */
+#define _XEN_NETIF_EXTRA_FLAG_MORE (0)
+#define XEN_NETIF_EXTRA_FLAG_MORE  (1U<<_XEN_NETIF_EXTRA_FLAG_MORE)
 
 /* GSO types */
-#define XEN_NETIF_GSO_TYPE_NONE		(0)
-#define XEN_NETIF_GSO_TYPE_TCPV4	(1)
-#define XEN_NETIF_GSO_TYPE_TCPV6	(2)
+#define XEN_NETIF_GSO_TYPE_NONE         (0)
+#define XEN_NETIF_GSO_TYPE_TCPV4        (1)
+#define XEN_NETIF_GSO_TYPE_TCPV6        (2)
 
 /*
- * This structure needs to fit within both netif_tx_request and
- * netif_rx_response for compatibility.
+ * This structure needs to fit within both xen_netif_tx_request_t and
+ * xen_netif_rx_response_t for compatibility.
  */
 struct xen_netif_extra_info {
-	uint8_t type;  /* XEN_NETIF_EXTRA_TYPE_* */
-	uint8_t flags; /* XEN_NETIF_EXTRA_FLAG_* */
-
+	uint8_t type;
+	uint8_t flags;
 	union {
 		struct {
-			/*
-			 * Maximum payload size of each segment. For
-			 * example, for TCP this is just the path MSS.
-			 */
 			uint16_t size;
-
-			/*
-			 * GSO type. This determines the protocol of
-			 * the packet and any extra features required
-			 * to segment the packet properly.
-			 */
-			uint8_t type; /* XEN_NETIF_GSO_TYPE_* */
-
-			/* Future expansion. */
+			uint8_t type;
 			uint8_t pad;
-
-			/*
-			 * GSO features. This specifies any extra GSO
-			 * features required to process this packet,
-			 * such as ECN support for TCPv4.
-			 */
-			uint16_t features; /* XEN_NETIF_GSO_FEAT_* */
+			uint16_t features;
 		} gso;
-
 		struct {
-			uint8_t addr[6]; /* Address to add/remove. */
+			uint8_t addr[6];
 		} mcast;
-
+		struct {
+			uint8_t type;
+			uint8_t algorithm;
+			uint8_t value[4];
+		} hash;
 		uint16_t pad[3];
 	} u;
 };
 
 struct xen_netif_tx_response {
 	uint16_t id;
-	int16_t  status;       /* XEN_NETIF_RSP_* */
+	int16_t status;
 };
 
 struct xen_netif_rx_request {
-	uint16_t    id;        /* Echoed in response message.        */
-	grant_ref_t gref;      /* Reference to incoming granted frame */
+	uint16_t id;		/* Echoed in response message.        */
+	uint16_t pad;
+	grant_ref_t gref;
 };
 
 /* Packet data has been validated against protocol checksum. */
-#define _XEN_NETRXF_data_validated	(0)
-#define  XEN_NETRXF_data_validated	(1U<<_XEN_NETRXF_data_validated)
+#define _XEN_NETRXF_data_validated (0)
+#define  XEN_NETRXF_data_validated (1U<<_XEN_NETRXF_data_validated)
 
 /* Protocol checksum field is blank in the packet (hardware offload)? */
-#define _XEN_NETRXF_csum_blank		(1)
-#define  XEN_NETRXF_csum_blank		(1U<<_XEN_NETRXF_csum_blank)
+#define _XEN_NETRXF_csum_blank     (1)
+#define  XEN_NETRXF_csum_blank     (1U<<_XEN_NETRXF_csum_blank)
 
 /* Packet continues in the next request descriptor. */
-#define _XEN_NETRXF_more_data		(2)
-#define  XEN_NETRXF_more_data		(1U<<_XEN_NETRXF_more_data)
+#define _XEN_NETRXF_more_data      (2)
+#define  XEN_NETRXF_more_data      (1U<<_XEN_NETRXF_more_data)
 
 /* Packet to be followed by extra descriptor(s). */
-#define _XEN_NETRXF_extra_info		(3)
-#define  XEN_NETRXF_extra_info		(1U<<_XEN_NETRXF_extra_info)
+#define _XEN_NETRXF_extra_info     (3)
+#define  XEN_NETRXF_extra_info     (1U<<_XEN_NETRXF_extra_info)
 
-/* GSO Prefix descriptor. */
-#define _XEN_NETRXF_gso_prefix		(4)
-#define  XEN_NETRXF_gso_prefix		(1U<<_XEN_NETRXF_gso_prefix)
+/* Packet has GSO prefix. Deprecated but included for compatibility */
+#define _XEN_NETRXF_gso_prefix     (4)
+#define  XEN_NETRXF_gso_prefix     (1U<<_XEN_NETRXF_gso_prefix)
 
 struct xen_netif_rx_response {
-    uint16_t id;
-    uint16_t offset;       /* Offset in page of start of received packet  */
-    uint16_t flags;        /* XEN_NETRXF_* */
-    int16_t  status;       /* -ve: BLKIF_RSP_* ; +ve: Rx'ed pkt size. */
+	uint16_t id;
+	uint16_t offset;
+	uint16_t flags;
+	int16_t status;
 };
 
 /*
- * Generate netif ring structures and types.
+ * Generate xen_netif ring structures and types.
  */
 
-DEFINE_RING_TYPES(xen_netif_tx,
-		  struct xen_netif_tx_request,
+DEFINE_RING_TYPES(xen_netif_tx, struct xen_netif_tx_request,
 		  struct xen_netif_tx_response);
-DEFINE_RING_TYPES(xen_netif_rx,
-		  struct xen_netif_rx_request,
+DEFINE_RING_TYPES(xen_netif_rx, struct xen_netif_rx_request,
 		  struct xen_netif_rx_response);
 
-#define XEN_NETIF_RSP_DROPPED	-2
-#define XEN_NETIF_RSP_ERROR	-1
-#define XEN_NETIF_RSP_OKAY	 0
-/* No response: used for auxiliary requests (e.g., xen_netif_extra_info). */
-#define XEN_NETIF_RSP_NULL	 1
+#define XEN_NETIF_RSP_DROPPED         -2
+#define XEN_NETIF_RSP_ERROR           -1
+#define XEN_NETIF_RSP_OKAY             0
+/* No response: used for auxiliary requests (e.g., xen_netif_extra_info_t). */
+#define XEN_NETIF_RSP_NULL             1
 
 #endif
-- 
cgit v0.10.2


From 562abd39a1902745bdcab266c7824cd6c5bc34d3 Mon Sep 17 00:00:00 2001
From: Paul Durrant <Paul.Durrant@citrix.com>
Date: Thu, 10 Mar 2016 12:30:27 +0000
Subject: xen-netback: support multiple extra info fragments passed from
 frontend

The code does not currently support a frontend passing multiple extra info
fragments to the backend in a tx request. The xenvif_get_extras() function
handles multiple extra_info fragments but make_tx_response() assumes there
is only ever a single extra info fragment.

This patch modifies xenvif_get_extras() to pass back a count of extra
info fragments, which is then passed to make_tx_response() (after
possibly being stashed in pending_tx_info for deferred responses).

Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 1128252..f44b388 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -52,6 +52,7 @@ typedef unsigned int pending_ring_idx_t;
 
 struct pending_tx_info {
 	struct xen_netif_tx_request req; /* tx request */
+	unsigned int extra_count;
 	/* Callback data for released SKBs. The callback is always
 	 * xenvif_zerocopy_callback, desc contains the pending_idx, which is
 	 * also an index in pending_tx_info array. It is initialized in
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 61b97c3..b42f260 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -95,6 +95,7 @@ static void xenvif_idx_release(struct xenvif_queue *queue, u16 pending_idx,
 
 static void make_tx_response(struct xenvif_queue *queue,
 			     struct xen_netif_tx_request *txp,
+			     unsigned int extra_count,
 			     s8       st);
 static void push_tx_responses(struct xenvif_queue *queue);
 
@@ -696,14 +697,15 @@ void xenvif_tx_credit_callback(unsigned long data)
 }
 
 static void xenvif_tx_err(struct xenvif_queue *queue,
-			  struct xen_netif_tx_request *txp, RING_IDX end)
+			  struct xen_netif_tx_request *txp,
+			  unsigned int extra_count, RING_IDX end)
 {
 	RING_IDX cons = queue->tx.req_cons;
 	unsigned long flags;
 
 	do {
 		spin_lock_irqsave(&queue->response_lock, flags);
-		make_tx_response(queue, txp, XEN_NETIF_RSP_ERROR);
+		make_tx_response(queue, txp, extra_count, XEN_NETIF_RSP_ERROR);
 		push_tx_responses(queue);
 		spin_unlock_irqrestore(&queue->response_lock, flags);
 		if (cons == end)
@@ -724,6 +726,7 @@ static void xenvif_fatal_tx_err(struct xenvif *vif)
 
 static int xenvif_count_requests(struct xenvif_queue *queue,
 				 struct xen_netif_tx_request *first,
+				 unsigned int extra_count,
 				 struct xen_netif_tx_request *txp,
 				 int work_to_do)
 {
@@ -812,7 +815,7 @@ static int xenvif_count_requests(struct xenvif_queue *queue,
 	} while (more_data);
 
 	if (drop_err) {
-		xenvif_tx_err(queue, first, cons + slots);
+		xenvif_tx_err(queue, first, extra_count, cons + slots);
 		return drop_err;
 	}
 
@@ -827,9 +830,10 @@ struct xenvif_tx_cb {
 #define XENVIF_TX_CB(skb) ((struct xenvif_tx_cb *)(skb)->cb)
 
 static inline void xenvif_tx_create_map_op(struct xenvif_queue *queue,
-					  u16 pending_idx,
-					  struct xen_netif_tx_request *txp,
-					  struct gnttab_map_grant_ref *mop)
+					   u16 pending_idx,
+					   struct xen_netif_tx_request *txp,
+					   unsigned int extra_count,
+					   struct gnttab_map_grant_ref *mop)
 {
 	queue->pages_to_map[mop-queue->tx_map_ops] = queue->mmap_pages[pending_idx];
 	gnttab_set_map_op(mop, idx_to_kaddr(queue, pending_idx),
@@ -838,6 +842,7 @@ static inline void xenvif_tx_create_map_op(struct xenvif_queue *queue,
 
 	memcpy(&queue->pending_tx_info[pending_idx].req, txp,
 	       sizeof(*txp));
+	queue->pending_tx_info[pending_idx].extra_count = extra_count;
 }
 
 static inline struct sk_buff *xenvif_alloc_skb(unsigned int size)
@@ -880,7 +885,7 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif_queue *que
 	     shinfo->nr_frags++, txp++, gop++) {
 		index = pending_index(queue->pending_cons++);
 		pending_idx = queue->pending_ring[index];
-		xenvif_tx_create_map_op(queue, pending_idx, txp, gop);
+		xenvif_tx_create_map_op(queue, pending_idx, txp, 0, gop);
 		frag_set_pending_idx(&frags[shinfo->nr_frags], pending_idx);
 	}
 
@@ -893,7 +898,8 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif_queue *que
 		     shinfo->nr_frags++, txp++, gop++) {
 			index = pending_index(queue->pending_cons++);
 			pending_idx = queue->pending_ring[index];
-			xenvif_tx_create_map_op(queue, pending_idx, txp, gop);
+			xenvif_tx_create_map_op(queue, pending_idx, txp, 0,
+						gop);
 			frag_set_pending_idx(&frags[shinfo->nr_frags],
 					     pending_idx);
 		}
@@ -1095,8 +1101,9 @@ static void xenvif_fill_frags(struct xenvif_queue *queue, struct sk_buff *skb)
 }
 
 static int xenvif_get_extras(struct xenvif_queue *queue,
-				struct xen_netif_extra_info *extras,
-				int work_to_do)
+			     struct xen_netif_extra_info *extras,
+			     unsigned int *extra_count,
+			     int work_to_do)
 {
 	struct xen_netif_extra_info extra;
 	RING_IDX cons = queue->tx.req_cons;
@@ -1109,9 +1116,12 @@ static int xenvif_get_extras(struct xenvif_queue *queue,
 		}
 
 		RING_COPY_REQUEST(&queue->tx, cons, &extra);
+
+		queue->tx.req_cons = ++cons;
+		(*extra_count)++;
+
 		if (unlikely(!extra.type ||
 			     extra.type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
-			queue->tx.req_cons = ++cons;
 			netdev_err(queue->vif->dev,
 				   "Invalid extra type: %d\n", extra.type);
 			xenvif_fatal_tx_err(queue->vif);
@@ -1119,7 +1129,6 @@ static int xenvif_get_extras(struct xenvif_queue *queue,
 		}
 
 		memcpy(&extras[extra.type - 1], &extra, sizeof(extra));
-		queue->tx.req_cons = ++cons;
 	} while (extra.flags & XEN_NETIF_EXTRA_FLAG_MORE);
 
 	return work_to_do;
@@ -1294,6 +1303,7 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 		struct xen_netif_tx_request txreq;
 		struct xen_netif_tx_request txfrags[XEN_NETBK_LEGACY_SLOTS_MAX];
 		struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX-1];
+		unsigned int extra_count;
 		u16 pending_idx;
 		RING_IDX idx;
 		int work_to_do;
@@ -1330,8 +1340,10 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 		queue->tx.req_cons = ++idx;
 
 		memset(extras, 0, sizeof(extras));
+		extra_count = 0;
 		if (txreq.flags & XEN_NETTXF_extra_info) {
 			work_to_do = xenvif_get_extras(queue, extras,
+						       &extra_count,
 						       work_to_do);
 			idx = queue->tx.req_cons;
 			if (unlikely(work_to_do < 0))
@@ -1344,7 +1356,7 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 			extra = &extras[XEN_NETIF_EXTRA_TYPE_MCAST_ADD - 1];
 			ret = xenvif_mcast_add(queue->vif, extra->u.mcast.addr);
 
-			make_tx_response(queue, &txreq,
+			make_tx_response(queue, &txreq, extra_count,
 					 (ret == 0) ?
 					 XEN_NETIF_RSP_OKAY :
 					 XEN_NETIF_RSP_ERROR);
@@ -1358,12 +1370,14 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 			extra = &extras[XEN_NETIF_EXTRA_TYPE_MCAST_DEL - 1];
 			xenvif_mcast_del(queue->vif, extra->u.mcast.addr);
 
-			make_tx_response(queue, &txreq, XEN_NETIF_RSP_OKAY);
+			make_tx_response(queue, &txreq, extra_count,
+					 XEN_NETIF_RSP_OKAY);
 			push_tx_responses(queue);
 			continue;
 		}
 
-		ret = xenvif_count_requests(queue, &txreq, txfrags, work_to_do);
+		ret = xenvif_count_requests(queue, &txreq, extra_count,
+					    txfrags, work_to_do);
 		if (unlikely(ret < 0))
 			break;
 
@@ -1372,7 +1386,7 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 		if (unlikely(txreq.size < ETH_HLEN)) {
 			netdev_dbg(queue->vif->dev,
 				   "Bad packet size: %d\n", txreq.size);
-			xenvif_tx_err(queue, &txreq, idx);
+			xenvif_tx_err(queue, &txreq, extra_count, idx);
 			break;
 		}
 
@@ -1397,7 +1411,7 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 		if (unlikely(skb == NULL)) {
 			netdev_dbg(queue->vif->dev,
 				   "Can't allocate a skb in start_xmit.\n");
-			xenvif_tx_err(queue, &txreq, idx);
+			xenvif_tx_err(queue, &txreq, extra_count, idx);
 			break;
 		}
 
@@ -1416,7 +1430,7 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 			nskb = xenvif_alloc_skb(0);
 			if (unlikely(nskb == NULL)) {
 				kfree_skb(skb);
-				xenvif_tx_err(queue, &txreq, idx);
+				xenvif_tx_err(queue, &txreq, extra_count, idx);
 				if (net_ratelimit())
 					netdev_err(queue->vif->dev,
 						   "Can't allocate the frag_list skb.\n");
@@ -1457,13 +1471,16 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue,
 		if (data_len < txreq.size) {
 			frag_set_pending_idx(&skb_shinfo(skb)->frags[0],
 					     pending_idx);
-			xenvif_tx_create_map_op(queue, pending_idx, &txreq, gop);
+			xenvif_tx_create_map_op(queue, pending_idx, &txreq,
+						extra_count, gop);
 			gop++;
 		} else {
 			frag_set_pending_idx(&skb_shinfo(skb)->frags[0],
 					     INVALID_PENDING_IDX);
-			memcpy(&queue->pending_tx_info[pending_idx].req, &txreq,
-			       sizeof(txreq));
+			memcpy(&queue->pending_tx_info[pending_idx].req,
+			       &txreq, sizeof(txreq));
+			queue->pending_tx_info[pending_idx].extra_count =
+				extra_count;
 		}
 
 		queue->pending_cons++;
@@ -1804,7 +1821,8 @@ static void xenvif_idx_release(struct xenvif_queue *queue, u16 pending_idx,
 
 	spin_lock_irqsave(&queue->response_lock, flags);
 
-	make_tx_response(queue, &pending_tx_info->req, status);
+	make_tx_response(queue, &pending_tx_info->req,
+			 pending_tx_info->extra_count, status);
 
 	/* Release the pending index before pusing the Tx response so
 	 * its available before a new Tx request is pushed by the
@@ -1821,6 +1839,7 @@ static void xenvif_idx_release(struct xenvif_queue *queue, u16 pending_idx,
 
 static void make_tx_response(struct xenvif_queue *queue,
 			     struct xen_netif_tx_request *txp,
+			     unsigned int extra_count,
 			     s8       st)
 {
 	RING_IDX i = queue->tx.rsp_prod_pvt;
@@ -1830,7 +1849,7 @@ static void make_tx_response(struct xenvif_queue *queue,
 	resp->id     = txp->id;
 	resp->status = st;
 
-	if (txp->flags & XEN_NETTXF_extra_info)
+	while (extra_count-- != 0)
 		RING_GET_RESPONSE(&queue->tx, ++i)->status = XEN_NETIF_RSP_NULL;
 
 	queue->tx.rsp_prod_pvt = ++i;
-- 
cgit v0.10.2


From 8e4ee59c1e75b74966476dcc3552c3b30d2768e7 Mon Sep 17 00:00:00 2001
From: Paul Durrant <Paul.Durrant@citrix.com>
Date: Thu, 10 Mar 2016 12:30:28 +0000
Subject: xen-netback: reduce log spam

Remove the "prepare for reconnect" pr_info in xenbus.c. It's largely
uninteresting and the states of the frontend and backend can easily be
observed by watching the (o)xenstored log.

Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
index 39a303d..bd182cd 100644
--- a/drivers/net/xen-netback/xenbus.c
+++ b/drivers/net/xen-netback/xenbus.c
@@ -511,8 +511,6 @@ static void set_backend_state(struct backend_info *be,
 			switch (state) {
 			case XenbusStateInitWait:
 			case XenbusStateConnected:
-				pr_info("%s: prepare for reconnect\n",
-					be->dev->nodename);
 				backend_switch_state(be, XenbusStateInitWait);
 				break;
 			case XenbusStateClosing:
-- 
cgit v0.10.2


From 470c3822d2ab7fadcbb1ac317ef27b31caac370e Mon Sep 17 00:00:00 2001
From: LABBE Corentin <clabbe.montjoie@gmail.com>
Date: Thu, 10 Mar 2016 13:58:58 +0100
Subject: phy: remove documentation of removed members of phy_device structure

Commit e5a03bfd873c ("phy: Add an mdio_device structure") removed addr,
bus and dev member of the phy_device structure.
This patch remove the documentation about those members.

Signed-off-by: LABBE Corentin <clabbe.montjoie@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/phy.h b/include/linux/phy.h
index d6f3641..2abd791 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -327,8 +327,6 @@ struct phy_c45_device_ids {
 /* phy_device: An instance of a PHY
  *
  * drv: Pointer to the driver for this PHY instance
- * bus: Pointer to the bus this PHY is on
- * dev: driver model device structure for this PHY
  * phy_id: UID for this device found during discovery
  * c45_ids: 802.3-c45 Device Identifers if is_c45.
  * is_c45:  Set to true if this phy uses clause 45 addressing.
@@ -338,7 +336,6 @@ struct phy_c45_device_ids {
  * suspended: Set to true if this phy has been suspended successfully.
  * state: state of the PHY for management purposes
  * dev_flags: Device-specific flags used by the PHY driver.
- * addr: Bus address of PHY
  * link_timeout: The number of timer firings to wait before the
  * giving up on the current attempt at acquiring a link
  * irq: IRQ number of the PHY's interrupt (-1 if none)
-- 
cgit v0.10.2


From 6bdaa5e9ed39b3b3328f35d218e8ad5a99cfc4d2 Mon Sep 17 00:00:00 2001
From: Nicolas Ferre <nicolas.ferre@atmel.com>
Date: Thu, 10 Mar 2016 16:44:32 +0100
Subject: net: macb: fix default configuration for GMAC on AT91

On AT91 SoCs, the User Register (USRIO) exposes a switch to configure the
"Reduced" or "Traditional" version of the Media Independent Interface
(RMII vs. MII or RGMII vs. GMII).
As on the older EMAC version, on GMAC, this switch is set by default to the
non-reduced type of interface, so use the existing capability and extend it to
GMII as well. We then keep the current logic in the macb_init() function.

The capabilities of sama5d2, sama5d4 and sama5d3 GEM interface are updated in
the macb_config structure to be able to properly enable them with a traditional
interface (GMII or MII).

Reported-by: Romain HENRIET <romain.henriet@l-acoustics.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c
index 7ccf229..3ce6095 100644
--- a/drivers/net/ethernet/cadence/macb.c
+++ b/drivers/net/ethernet/cadence/macb.c
@@ -2440,9 +2440,9 @@ static int macb_init(struct platform_device *pdev)
 		if (bp->phy_interface == PHY_INTERFACE_MODE_RGMII)
 			val = GEM_BIT(RGMII);
 		else if (bp->phy_interface == PHY_INTERFACE_MODE_RMII &&
-			 (bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII))
+			 (bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII))
 			val = MACB_BIT(RMII);
-		else if (!(bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII))
+		else if (!(bp->caps & MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII))
 			val = MACB_BIT(MII);
 
 		if (bp->caps & MACB_CAPS_USRIO_HAS_CLKEN)
@@ -2774,7 +2774,7 @@ static int at91ether_init(struct platform_device *pdev)
 }
 
 static const struct macb_config at91sam9260_config = {
-	.caps = MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_USRIO_DEFAULT_IS_MII,
+	.caps = MACB_CAPS_USRIO_HAS_CLKEN | MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
 	.clk_init = macb_clk_init,
 	.init = macb_init,
 };
@@ -2787,21 +2787,22 @@ static const struct macb_config pc302gem_config = {
 };
 
 static const struct macb_config sama5d2_config = {
-	.caps = 0,
+	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
 	.dma_burst_length = 16,
 	.clk_init = macb_clk_init,
 	.init = macb_init,
 };
 
 static const struct macb_config sama5d3_config = {
-	.caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE,
+	.caps = MACB_CAPS_SG_DISABLED | MACB_CAPS_GIGABIT_MODE_AVAILABLE
+	      | MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
 	.dma_burst_length = 16,
 	.clk_init = macb_clk_init,
 	.init = macb_init,
 };
 
 static const struct macb_config sama5d4_config = {
-	.caps = 0,
+	.caps = MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII,
 	.dma_burst_length = 4,
 	.clk_init = macb_clk_init,
 	.init = macb_init,
diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 9ba416d..8a13824 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -400,7 +400,7 @@
 /* Capability mask bits */
 #define MACB_CAPS_ISR_CLEAR_ON_WRITE		0x00000001
 #define MACB_CAPS_USRIO_HAS_CLKEN		0x00000002
-#define MACB_CAPS_USRIO_DEFAULT_IS_MII		0x00000004
+#define MACB_CAPS_USRIO_DEFAULT_IS_MII_GMII	0x00000004
 #define MACB_CAPS_NO_GIGABIT_HALF		0x00000008
 #define MACB_CAPS_USRIO_DISABLED		0x00000010
 #define MACB_CAPS_FIFO_MODE			0x10000000
-- 
cgit v0.10.2


From 6f15cdbf8a8ac2e22767cc8b1eae225702733c95 Mon Sep 17 00:00:00 2001
From: Samuel Gauthier <samuel.gauthier@6wind.com>
Date: Thu, 10 Mar 2016 17:14:59 +0100
Subject: ovs: allow nl 'flow set' to use ufid without flow key

When we want to change a flow using netlink, we have to identify it to
be able to perform a lookup. Both the flow key and unique flow ID
(ufid) are valid identifiers, but we always have to specify the flow
key in the netlink message. When both attributes are there, the ufid
is used. The flow key is used to validate the actions provided by
the userland.

This commit allows to use the ufid without having to provide the flow
key, as it is already done in the netlink 'flow get' and 'flow del'
path. The flow key remains mandatory when an action is provided.

Signed-off-by: Samuel Gauthier <samuel.gauthier@6wind.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Acked-by: Pravin B Shelar <pshelar@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index e6a7d49..0cc66a4 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1096,26 +1096,32 @@ static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info)
 	struct sw_flow_match match;
 	struct sw_flow_id sfid;
 	u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]);
-	int error;
+	int error = 0;
 	bool log = !a[OVS_FLOW_ATTR_PROBE];
 	bool ufid_present;
 
-	/* Extract key. */
-	error = -EINVAL;
-	if (!a[OVS_FLOW_ATTR_KEY]) {
-		OVS_NLERR(log, "Flow key attribute not present in set flow.");
-		goto error;
-	}
-
 	ufid_present = ovs_nla_get_ufid(&sfid, a[OVS_FLOW_ATTR_UFID], log);
-	ovs_match_init(&match, &key, &mask);
-	error = ovs_nla_get_match(net, &match, a[OVS_FLOW_ATTR_KEY],
-				  a[OVS_FLOW_ATTR_MASK], log);
+	if (a[OVS_FLOW_ATTR_KEY]) {
+		ovs_match_init(&match, &key, &mask);
+		error = ovs_nla_get_match(net, &match, a[OVS_FLOW_ATTR_KEY],
+					  a[OVS_FLOW_ATTR_MASK], log);
+	} else if (!ufid_present) {
+		OVS_NLERR(log,
+			  "Flow set message rejected, Key attribute missing.");
+		error = -EINVAL;
+	}
 	if (error)
 		goto error;
 
 	/* Validate actions. */
 	if (a[OVS_FLOW_ATTR_ACTIONS]) {
+		if (!a[OVS_FLOW_ATTR_KEY]) {
+			OVS_NLERR(log,
+				  "Flow key attribute not present in set flow.");
+			error = -EINVAL;
+			goto error;
+		}
+
 		acts = get_flow_actions(net, a[OVS_FLOW_ATTR_ACTIONS], &key,
 					&mask, log);
 		if (IS_ERR(acts)) {
-- 
cgit v0.10.2


From cea8768f333e3f0bc231d8b815aa4a9e63fa990c Mon Sep 17 00:00:00 2001
From: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Date: Thu, 10 Mar 2016 18:33:07 -0300
Subject: sctp: allow sctp_transmit_packet and others to use gfp

Currently sctp_sendmsg() triggers some calls that will allocate memory
with GFP_ATOMIC even when not necessary. In the case of
sctp_packet_transmit it will allocate a linear skb that will be used to
construct the packet and this may cause sends to fail due to ENOMEM more
often than anticipated specially with big MTUs.

This patch thus allows it to inherit gfp flags from upper calls so that
it can use GFP_KERNEL if it was triggered by a sctp_sendmsg call or
similar. All others, like retransmits or flushes started from BH, are
still allocated using GFP_ATOMIC.

In netperf tests this didn't result in any performance drawbacks when
memory is not too fragmented and made it trigger ENOMEM way less often.

Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h
index 487ef34..efc0174 100644
--- a/include/net/sctp/sm.h
+++ b/include/net/sctp/sm.h
@@ -201,7 +201,7 @@ struct sctp_chunk *sctp_make_cwr(const struct sctp_association *,
 struct sctp_chunk * sctp_make_datafrag_empty(struct sctp_association *,
 					const struct sctp_sndrcvinfo *sinfo,
 					int len, const __u8 flags,
-					__u16 ssn);
+					__u16 ssn, gfp_t gfp);
 struct sctp_chunk *sctp_make_ecne(const struct sctp_association *,
 				  const __u32);
 struct sctp_chunk *sctp_make_sack(const struct sctp_association *);
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index d05b566..9d23766 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -655,7 +655,7 @@ void sctp_chunk_free(struct sctp_chunk *);
 void  *sctp_addto_chunk(struct sctp_chunk *, int len, const void *data);
 struct sctp_chunk *sctp_chunkify(struct sk_buff *,
 				 const struct sctp_association *,
-				 struct sock *);
+				 struct sock *, gfp_t gfp);
 void sctp_init_addrs(struct sctp_chunk *, union sctp_addr *,
 		     union sctp_addr *);
 const union sctp_addr *sctp_source(const struct sctp_chunk *chunk);
@@ -717,10 +717,10 @@ struct sctp_packet *sctp_packet_init(struct sctp_packet *,
 				     __u16 sport, __u16 dport);
 struct sctp_packet *sctp_packet_config(struct sctp_packet *, __u32 vtag, int);
 sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *,
-                                       struct sctp_chunk *, int);
+				       struct sctp_chunk *, int, gfp_t);
 sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *,
                                      struct sctp_chunk *);
-int sctp_packet_transmit(struct sctp_packet *);
+int sctp_packet_transmit(struct sctp_packet *, gfp_t);
 void sctp_packet_free(struct sctp_packet *);
 
 static inline int sctp_packet_empty(struct sctp_packet *packet)
@@ -1053,7 +1053,7 @@ struct sctp_outq {
 void sctp_outq_init(struct sctp_association *, struct sctp_outq *);
 void sctp_outq_teardown(struct sctp_outq *);
 void sctp_outq_free(struct sctp_outq*);
-int sctp_outq_tail(struct sctp_outq *, struct sctp_chunk *chunk);
+int sctp_outq_tail(struct sctp_outq *, struct sctp_chunk *chunk, gfp_t);
 int sctp_outq_sack(struct sctp_outq *, struct sctp_chunk *);
 int sctp_outq_is_empty(const struct sctp_outq *);
 void sctp_outq_restart(struct sctp_outq *);
@@ -1061,7 +1061,7 @@ void sctp_outq_restart(struct sctp_outq *);
 void sctp_retransmit(struct sctp_outq *, struct sctp_transport *,
 		     sctp_retransmit_reason_t);
 void sctp_retransmit_mark(struct sctp_outq *, struct sctp_transport *, __u8);
-int sctp_outq_uncork(struct sctp_outq *);
+int sctp_outq_uncork(struct sctp_outq *, gfp_t gfp);
 /* Uncork and flush an outqueue.  */
 static inline void sctp_outq_cork(struct sctp_outq *q)
 {
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index cd87344..a19b3e6 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -1493,7 +1493,7 @@ void sctp_assoc_rwnd_increase(struct sctp_association *asoc, unsigned int len)
 
 		asoc->peer.sack_needed = 0;
 
-		sctp_outq_tail(&asoc->outqueue, sack);
+		sctp_outq_tail(&asoc->outqueue, sack, GFP_ATOMIC);
 
 		/* Stop the SACK timer.  */
 		timer = &asoc->timers[SCTP_EVENT_TIMEOUT_SACK];
diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c
index 3aa4307..958ef5f 100644
--- a/net/sctp/chunk.c
+++ b/net/sctp/chunk.c
@@ -260,7 +260,8 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *asoc,
 				frag |= SCTP_DATA_SACK_IMM;
 		}
 
-		chunk = sctp_make_datafrag_empty(asoc, sinfo, len, frag, 0);
+		chunk = sctp_make_datafrag_empty(asoc, sinfo, len, frag,
+						 0, GFP_KERNEL);
 
 		if (!chunk) {
 			err = -ENOMEM;
@@ -296,7 +297,8 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *asoc,
 		    (sinfo->sinfo_flags & SCTP_SACK_IMMEDIATELY))
 			frag |= SCTP_DATA_SACK_IMM;
 
-		chunk = sctp_make_datafrag_empty(asoc, sinfo, over, frag, 0);
+		chunk = sctp_make_datafrag_empty(asoc, sinfo, over, frag,
+						 0, GFP_KERNEL);
 
 		if (!chunk) {
 			err = -ENOMEM;
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 21a2d6b..db76f1a 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -221,7 +221,7 @@ int sctp_rcv(struct sk_buff *skb)
 		goto discard_release;
 
 	/* Create an SCTP packet structure. */
-	chunk = sctp_chunkify(skb, asoc, sk);
+	chunk = sctp_chunkify(skb, asoc, sk, GFP_ATOMIC);
 	if (!chunk)
 		goto discard_release;
 	SCTP_INPUT_CB(skb)->chunk = chunk;
diff --git a/net/sctp/output.c b/net/sctp/output.c
index 9d610ed..736c004 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -153,7 +153,7 @@ void sctp_packet_free(struct sctp_packet *packet)
  */
 sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet,
 				       struct sctp_chunk *chunk,
-				       int one_packet)
+				       int one_packet, gfp_t gfp)
 {
 	sctp_xmit_t retval;
 	int error = 0;
@@ -163,7 +163,7 @@ sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet,
 	switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) {
 	case SCTP_XMIT_PMTU_FULL:
 		if (!packet->has_cookie_echo) {
-			error = sctp_packet_transmit(packet);
+			error = sctp_packet_transmit(packet, gfp);
 			if (error < 0)
 				chunk->skb->sk->sk_err = -error;
 
@@ -376,7 +376,7 @@ static void sctp_packet_set_owner_w(struct sk_buff *skb, struct sock *sk)
  *
  * The return value is a normal kernel error return value.
  */
-int sctp_packet_transmit(struct sctp_packet *packet)
+int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp)
 {
 	struct sctp_transport *tp = packet->transport;
 	struct sctp_association *asoc = tp->asoc;
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index c0380cf..f03541d 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -68,7 +68,7 @@ static void sctp_mark_missing(struct sctp_outq *q,
 
 static void sctp_generate_fwdtsn(struct sctp_outq *q, __u32 sack_ctsn);
 
-static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout);
+static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp);
 
 /* Add data to the front of the queue. */
 static inline void sctp_outq_head_data(struct sctp_outq *q,
@@ -285,7 +285,7 @@ void sctp_outq_free(struct sctp_outq *q)
 }
 
 /* Put a new chunk in an sctp_outq.  */
-int sctp_outq_tail(struct sctp_outq *q, struct sctp_chunk *chunk)
+int sctp_outq_tail(struct sctp_outq *q, struct sctp_chunk *chunk, gfp_t gfp)
 {
 	struct net *net = sock_net(q->asoc->base.sk);
 	int error = 0;
@@ -341,7 +341,7 @@ int sctp_outq_tail(struct sctp_outq *q, struct sctp_chunk *chunk)
 		return error;
 
 	if (!q->cork)
-		error = sctp_outq_flush(q, 0);
+		error = sctp_outq_flush(q, 0, gfp);
 
 	return error;
 }
@@ -510,7 +510,7 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport,
 	 * will be flushed at the end.
 	 */
 	if (reason != SCTP_RTXR_FAST_RTX)
-		error = sctp_outq_flush(q, /* rtx_timeout */ 1);
+		error = sctp_outq_flush(q, /* rtx_timeout */ 1, GFP_ATOMIC);
 
 	if (error)
 		q->asoc->base.sk->sk_err = -error;
@@ -601,12 +601,12 @@ redo:
 				 * control chunks are already freed so there
 				 * is nothing we can do.
 				 */
-				sctp_packet_transmit(pkt);
+				sctp_packet_transmit(pkt, GFP_ATOMIC);
 				goto redo;
 			}
 
 			/* Send this packet.  */
-			error = sctp_packet_transmit(pkt);
+			error = sctp_packet_transmit(pkt, GFP_ATOMIC);
 
 			/* If we are retransmitting, we should only
 			 * send a single packet.
@@ -622,7 +622,7 @@ redo:
 
 		case SCTP_XMIT_RWND_FULL:
 			/* Send this packet. */
-			error = sctp_packet_transmit(pkt);
+			error = sctp_packet_transmit(pkt, GFP_ATOMIC);
 
 			/* Stop sending DATA as there is no more room
 			 * at the receiver.
@@ -632,7 +632,7 @@ redo:
 
 		case SCTP_XMIT_DELAY:
 			/* Send this packet. */
-			error = sctp_packet_transmit(pkt);
+			error = sctp_packet_transmit(pkt, GFP_ATOMIC);
 
 			/* Stop sending DATA because of nagle delay. */
 			done = 1;
@@ -685,12 +685,12 @@ redo:
 }
 
 /* Cork the outqueue so queued chunks are really queued. */
-int sctp_outq_uncork(struct sctp_outq *q)
+int sctp_outq_uncork(struct sctp_outq *q, gfp_t gfp)
 {
 	if (q->cork)
 		q->cork = 0;
 
-	return sctp_outq_flush(q, 0);
+	return sctp_outq_flush(q, 0, gfp);
 }
 
 
@@ -703,7 +703,7 @@ int sctp_outq_uncork(struct sctp_outq *q)
  * locking concerns must be made.  Today we use the sock lock to protect
  * this function.
  */
-static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
+static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp)
 {
 	struct sctp_packet *packet;
 	struct sctp_packet singleton;
@@ -825,7 +825,7 @@ static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 			sctp_packet_init(&singleton, transport, sport, dport);
 			sctp_packet_config(&singleton, vtag, 0);
 			sctp_packet_append_chunk(&singleton, chunk);
-			error = sctp_packet_transmit(&singleton);
+			error = sctp_packet_transmit(&singleton, gfp);
 			if (error < 0)
 				return error;
 			break;
@@ -856,7 +856,7 @@ static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 		case SCTP_CID_ASCONF:
 		case SCTP_CID_FWD_TSN:
 			status = sctp_packet_transmit_chunk(packet, chunk,
-							    one_packet);
+							    one_packet, gfp);
 			if (status  != SCTP_XMIT_OK) {
 				/* put the chunk back */
 				list_add(&chunk->list, &q->control_chunk_list);
@@ -1011,7 +1011,7 @@ static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
 				 atomic_read(&chunk->skb->users) : -1);
 
 			/* Add the chunk to the packet.  */
-			status = sctp_packet_transmit_chunk(packet, chunk, 0);
+			status = sctp_packet_transmit_chunk(packet, chunk, 0, gfp);
 
 			switch (status) {
 			case SCTP_XMIT_PMTU_FULL:
@@ -1088,7 +1088,7 @@ sctp_flush_out:
 						      send_ready);
 		packet = &t->packet;
 		if (!sctp_packet_empty(packet))
-			error = sctp_packet_transmit(packet);
+			error = sctp_packet_transmit(packet, gfp);
 
 		/* Clear the burst limited state, if any */
 		sctp_transport_burst_reset(t);
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 5d6a03f..8449ca2 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -62,11 +62,13 @@
 #include <net/sctp/sm.h>
 
 static struct sctp_chunk *sctp_make_control(const struct sctp_association *asoc,
-					    __u8 type, __u8 flags, int paylen);
+					    __u8 type, __u8 flags, int paylen,
+					    gfp_t gfp);
 static struct sctp_chunk *sctp_make_data(const struct sctp_association *asoc,
-					 __u8 flags, int paylen);
+					 __u8 flags, int paylen, gfp_t gfp);
 static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc,
-					   __u8 type, __u8 flags, int paylen);
+					   __u8 type, __u8 flags, int paylen,
+					   gfp_t gfp);
 static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep,
 					const struct sctp_association *asoc,
 					const struct sctp_chunk *init_chunk,
@@ -318,7 +320,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc,
 	 * PLEASE DO NOT FIXME [This version does not support Host Name.]
 	 */
 
-	retval = sctp_make_control(asoc, SCTP_CID_INIT, 0, chunksize);
+	retval = sctp_make_control(asoc, SCTP_CID_INIT, 0, chunksize, gfp);
 	if (!retval)
 		goto nodata;
 
@@ -465,7 +467,7 @@ struct sctp_chunk *sctp_make_init_ack(const struct sctp_association *asoc,
 					num_ext);
 
 	/* Now allocate and fill out the chunk.  */
-	retval = sctp_make_control(asoc, SCTP_CID_INIT_ACK, 0, chunksize);
+	retval = sctp_make_control(asoc, SCTP_CID_INIT_ACK, 0, chunksize, gfp);
 	if (!retval)
 		goto nomem_chunk;
 
@@ -570,7 +572,8 @@ struct sctp_chunk *sctp_make_cookie_echo(const struct sctp_association *asoc,
 	cookie_len = asoc->peer.cookie_len;
 
 	/* Build a cookie echo chunk.  */
-	retval = sctp_make_control(asoc, SCTP_CID_COOKIE_ECHO, 0, cookie_len);
+	retval = sctp_make_control(asoc, SCTP_CID_COOKIE_ECHO, 0,
+				   cookie_len, GFP_ATOMIC);
 	if (!retval)
 		goto nodata;
 	retval->subh.cookie_hdr =
@@ -615,7 +618,7 @@ struct sctp_chunk *sctp_make_cookie_ack(const struct sctp_association *asoc,
 {
 	struct sctp_chunk *retval;
 
-	retval = sctp_make_control(asoc, SCTP_CID_COOKIE_ACK, 0, 0);
+	retval = sctp_make_control(asoc, SCTP_CID_COOKIE_ACK, 0, 0, GFP_ATOMIC);
 
 	/* RFC 2960 6.4 Multi-homed SCTP Endpoints
 	 *
@@ -664,7 +667,7 @@ struct sctp_chunk *sctp_make_cwr(const struct sctp_association *asoc,
 
 	cwr.lowest_tsn = htonl(lowest_tsn);
 	retval = sctp_make_control(asoc, SCTP_CID_ECN_CWR, 0,
-				   sizeof(sctp_cwrhdr_t));
+				   sizeof(sctp_cwrhdr_t), GFP_ATOMIC);
 
 	if (!retval)
 		goto nodata;
@@ -698,7 +701,7 @@ struct sctp_chunk *sctp_make_ecne(const struct sctp_association *asoc,
 
 	ecne.lowest_tsn = htonl(lowest_tsn);
 	retval = sctp_make_control(asoc, SCTP_CID_ECN_ECNE, 0,
-				   sizeof(sctp_ecnehdr_t));
+				   sizeof(sctp_ecnehdr_t), GFP_ATOMIC);
 	if (!retval)
 		goto nodata;
 	retval->subh.ecne_hdr =
@@ -713,7 +716,8 @@ nodata:
  */
 struct sctp_chunk *sctp_make_datafrag_empty(struct sctp_association *asoc,
 				       const struct sctp_sndrcvinfo *sinfo,
-				       int data_len, __u8 flags, __u16 ssn)
+				       int data_len, __u8 flags, __u16 ssn,
+				       gfp_t gfp)
 {
 	struct sctp_chunk *retval;
 	struct sctp_datahdr dp;
@@ -734,7 +738,7 @@ struct sctp_chunk *sctp_make_datafrag_empty(struct sctp_association *asoc,
 		dp.ssn = htons(ssn);
 
 	chunk_len = sizeof(dp) + data_len;
-	retval = sctp_make_data(asoc, flags, chunk_len);
+	retval = sctp_make_data(asoc, flags, chunk_len, gfp);
 	if (!retval)
 		goto nodata;
 
@@ -781,7 +785,7 @@ struct sctp_chunk *sctp_make_sack(const struct sctp_association *asoc)
 		+ sizeof(__u32) * num_dup_tsns;
 
 	/* Create the chunk.  */
-	retval = sctp_make_control(asoc, SCTP_CID_SACK, 0, len);
+	retval = sctp_make_control(asoc, SCTP_CID_SACK, 0, len, GFP_ATOMIC);
 	if (!retval)
 		goto nodata;
 
@@ -861,7 +865,7 @@ struct sctp_chunk *sctp_make_shutdown(const struct sctp_association *asoc,
 	shut.cum_tsn_ack = htonl(ctsn);
 
 	retval = sctp_make_control(asoc, SCTP_CID_SHUTDOWN, 0,
-				   sizeof(sctp_shutdownhdr_t));
+				   sizeof(sctp_shutdownhdr_t), GFP_ATOMIC);
 	if (!retval)
 		goto nodata;
 
@@ -879,7 +883,8 @@ struct sctp_chunk *sctp_make_shutdown_ack(const struct sctp_association *asoc,
 {
 	struct sctp_chunk *retval;
 
-	retval = sctp_make_control(asoc, SCTP_CID_SHUTDOWN_ACK, 0, 0);
+	retval = sctp_make_control(asoc, SCTP_CID_SHUTDOWN_ACK, 0, 0,
+				   GFP_ATOMIC);
 
 	/* RFC 2960 6.4 Multi-homed SCTP Endpoints
 	 *
@@ -908,7 +913,8 @@ struct sctp_chunk *sctp_make_shutdown_complete(
 	 */
 	flags |= asoc ? 0 : SCTP_CHUNK_FLAG_T;
 
-	retval = sctp_make_control(asoc, SCTP_CID_SHUTDOWN_COMPLETE, flags, 0);
+	retval = sctp_make_control(asoc, SCTP_CID_SHUTDOWN_COMPLETE, flags,
+				   0, GFP_ATOMIC);
 
 	/* RFC 2960 6.4 Multi-homed SCTP Endpoints
 	 *
@@ -947,7 +953,8 @@ struct sctp_chunk *sctp_make_abort(const struct sctp_association *asoc,
 			flags = SCTP_CHUNK_FLAG_T;
 	}
 
-	retval = sctp_make_control(asoc, SCTP_CID_ABORT, flags, hint);
+	retval = sctp_make_control(asoc, SCTP_CID_ABORT, flags, hint,
+				   GFP_ATOMIC);
 
 	/* RFC 2960 6.4 Multi-homed SCTP Endpoints
 	 *
@@ -1139,7 +1146,8 @@ struct sctp_chunk *sctp_make_heartbeat(const struct sctp_association *asoc,
 	struct sctp_chunk *retval;
 	sctp_sender_hb_info_t hbinfo;
 
-	retval = sctp_make_control(asoc, SCTP_CID_HEARTBEAT, 0, sizeof(hbinfo));
+	retval = sctp_make_control(asoc, SCTP_CID_HEARTBEAT, 0,
+				   sizeof(hbinfo), GFP_ATOMIC);
 
 	if (!retval)
 		goto nodata;
@@ -1167,7 +1175,8 @@ struct sctp_chunk *sctp_make_heartbeat_ack(const struct sctp_association *asoc,
 {
 	struct sctp_chunk *retval;
 
-	retval  = sctp_make_control(asoc, SCTP_CID_HEARTBEAT_ACK, 0, paylen);
+	retval  = sctp_make_control(asoc, SCTP_CID_HEARTBEAT_ACK, 0, paylen,
+				    GFP_ATOMIC);
 	if (!retval)
 		goto nodata;
 
@@ -1200,7 +1209,7 @@ static struct sctp_chunk *sctp_make_op_error_space(
 	struct sctp_chunk *retval;
 
 	retval = sctp_make_control(asoc, SCTP_CID_ERROR, 0,
-				   sizeof(sctp_errhdr_t) + size);
+				   sizeof(sctp_errhdr_t) + size, GFP_ATOMIC);
 	if (!retval)
 		goto nodata;
 
@@ -1271,7 +1280,8 @@ struct sctp_chunk *sctp_make_auth(const struct sctp_association *asoc)
 		return NULL;
 
 	retval = sctp_make_control(asoc, SCTP_CID_AUTH, 0,
-			hmac_desc->hmac_len + sizeof(sctp_authhdr_t));
+			hmac_desc->hmac_len + sizeof(sctp_authhdr_t),
+			GFP_ATOMIC);
 	if (!retval)
 		return NULL;
 
@@ -1309,11 +1319,11 @@ struct sctp_chunk *sctp_make_auth(const struct sctp_association *asoc)
  */
 struct sctp_chunk *sctp_chunkify(struct sk_buff *skb,
 			    const struct sctp_association *asoc,
-			    struct sock *sk)
+			    struct sock *sk, gfp_t gfp)
 {
 	struct sctp_chunk *retval;
 
-	retval = kmem_cache_zalloc(sctp_chunk_cachep, GFP_ATOMIC);
+	retval = kmem_cache_zalloc(sctp_chunk_cachep, gfp);
 
 	if (!retval)
 		goto nodata;
@@ -1361,7 +1371,8 @@ const union sctp_addr *sctp_source(const struct sctp_chunk *chunk)
  * arguments, reserving enough space for a 'paylen' byte payload.
  */
 static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc,
-					    __u8 type, __u8 flags, int paylen)
+					    __u8 type, __u8 flags, int paylen,
+					    gfp_t gfp)
 {
 	struct sctp_chunk *retval;
 	sctp_chunkhdr_t *chunk_hdr;
@@ -1369,8 +1380,7 @@ static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc,
 	struct sock *sk;
 
 	/* No need to allocate LL here, as this is only a chunk. */
-	skb = alloc_skb(WORD_ROUND(sizeof(sctp_chunkhdr_t) + paylen),
-			GFP_ATOMIC);
+	skb = alloc_skb(WORD_ROUND(sizeof(sctp_chunkhdr_t) + paylen), gfp);
 	if (!skb)
 		goto nodata;
 
@@ -1381,7 +1391,7 @@ static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc,
 	chunk_hdr->length = htons(sizeof(sctp_chunkhdr_t));
 
 	sk = asoc ? asoc->base.sk : NULL;
-	retval = sctp_chunkify(skb, asoc, sk);
+	retval = sctp_chunkify(skb, asoc, sk, gfp);
 	if (!retval) {
 		kfree_skb(skb);
 		goto nodata;
@@ -1400,16 +1410,18 @@ nodata:
 }
 
 static struct sctp_chunk *sctp_make_data(const struct sctp_association *asoc,
-					 __u8 flags, int paylen)
+					 __u8 flags, int paylen, gfp_t gfp)
 {
-	return _sctp_make_chunk(asoc, SCTP_CID_DATA, flags, paylen);
+	return _sctp_make_chunk(asoc, SCTP_CID_DATA, flags, paylen, gfp);
 }
 
 static struct sctp_chunk *sctp_make_control(const struct sctp_association *asoc,
-					    __u8 type, __u8 flags, int paylen)
+					    __u8 type, __u8 flags, int paylen,
+					    gfp_t gfp)
 {
-	struct sctp_chunk *chunk = _sctp_make_chunk(asoc, type, flags, paylen);
+	struct sctp_chunk *chunk;
 
+	chunk = _sctp_make_chunk(asoc, type, flags, paylen, gfp);
 	if (chunk)
 		sctp_control_set_owner_w(chunk);
 
@@ -2756,7 +2768,8 @@ static struct sctp_chunk *sctp_make_asconf(struct sctp_association *asoc,
 	length += addrlen;
 
 	/* Create the chunk.  */
-	retval = sctp_make_control(asoc, SCTP_CID_ASCONF, 0, length);
+	retval = sctp_make_control(asoc, SCTP_CID_ASCONF, 0, length,
+				   GFP_ATOMIC);
 	if (!retval)
 		return NULL;
 
@@ -2940,7 +2953,8 @@ static struct sctp_chunk *sctp_make_asconf_ack(const struct sctp_association *as
 	int			length = sizeof(asconf) + vparam_len;
 
 	/* Create the chunk.  */
-	retval = sctp_make_control(asoc, SCTP_CID_ASCONF_ACK, 0, length);
+	retval = sctp_make_control(asoc, SCTP_CID_ASCONF_ACK, 0, length,
+				   GFP_ATOMIC);
 	if (!retval)
 		return NULL;
 
@@ -3500,7 +3514,7 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc,
 
 	hint = (nstreams + 1) * sizeof(__u32);
 
-	retval = sctp_make_control(asoc, SCTP_CID_FWD_TSN, 0, hint);
+	retval = sctp_make_control(asoc, SCTP_CID_FWD_TSN, 0, hint, GFP_ATOMIC);
 
 	if (!retval)
 		return NULL;
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index b5327bb..3c22c41 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -1019,13 +1019,13 @@ static void sctp_cmd_t1_timer_update(struct sctp_association *asoc,
  * encouraged for small fragments.
  */
 static int sctp_cmd_send_msg(struct sctp_association *asoc,
-				struct sctp_datamsg *msg)
+				struct sctp_datamsg *msg, gfp_t gfp)
 {
 	struct sctp_chunk *chunk;
 	int error = 0;
 
 	list_for_each_entry(chunk, &msg->chunks, frag_list) {
-		error = sctp_outq_tail(&asoc->outqueue, chunk);
+		error = sctp_outq_tail(&asoc->outqueue, chunk, gfp);
 		if (error)
 			break;
 	}
@@ -1249,7 +1249,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 		case SCTP_CMD_NEW_ASOC:
 			/* Register a new association.  */
 			if (local_cork) {
-				sctp_outq_uncork(&asoc->outqueue);
+				sctp_outq_uncork(&asoc->outqueue, gfp);
 				local_cork = 0;
 			}
 
@@ -1269,7 +1269,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 
 		case SCTP_CMD_DELETE_TCB:
 			if (local_cork) {
-				sctp_outq_uncork(&asoc->outqueue);
+				sctp_outq_uncork(&asoc->outqueue, gfp);
 				local_cork = 0;
 			}
 			/* Delete the current association.  */
@@ -1423,13 +1423,14 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 				local_cork = 1;
 			}
 			/* Send a chunk to our peer.  */
-			error = sctp_outq_tail(&asoc->outqueue, cmd->obj.chunk);
+			error = sctp_outq_tail(&asoc->outqueue, cmd->obj.chunk,
+					       gfp);
 			break;
 
 		case SCTP_CMD_SEND_PKT:
 			/* Send a full packet to our peer.  */
 			packet = cmd->obj.packet;
-			sctp_packet_transmit(packet);
+			sctp_packet_transmit(packet, gfp);
 			sctp_ootb_pkt_free(packet);
 			break;
 
@@ -1639,7 +1640,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 			 */
 			chunk->pdiscard = 1;
 			if (asoc) {
-				sctp_outq_uncork(&asoc->outqueue);
+				sctp_outq_uncork(&asoc->outqueue, gfp);
 				local_cork = 0;
 			}
 			break;
@@ -1677,7 +1678,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 		case SCTP_CMD_FORCE_PRIM_RETRAN:
 			t = asoc->peer.retran_path;
 			asoc->peer.retran_path = asoc->peer.primary_path;
-			error = sctp_outq_uncork(&asoc->outqueue);
+			error = sctp_outq_uncork(&asoc->outqueue, gfp);
 			local_cork = 0;
 			asoc->peer.retran_path = t;
 			break;
@@ -1704,7 +1705,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
 				sctp_outq_cork(&asoc->outqueue);
 				local_cork = 1;
 			}
-			error = sctp_cmd_send_msg(asoc, cmd->obj.msg);
+			error = sctp_cmd_send_msg(asoc, cmd->obj.msg, gfp);
 			break;
 		case SCTP_CMD_SEND_NEXT_ASCONF:
 			sctp_cmd_send_asconf(asoc);
@@ -1734,9 +1735,9 @@ out:
 	 */
 	if (asoc && SCTP_EVENT_T_CHUNK == event_type && chunk) {
 		if (chunk->end_of_packet || chunk->singleton)
-			error = sctp_outq_uncork(&asoc->outqueue);
+			error = sctp_outq_uncork(&asoc->outqueue, gfp);
 	} else if (local_cork)
-		error = sctp_outq_uncork(&asoc->outqueue);
+		error = sctp_outq_uncork(&asoc->outqueue, gfp);
 	return error;
 nomem:
 	error = -ENOMEM;
-- 
cgit v0.10.2


From 233fa44bd67ae0c6cbc01f96ab87d66a6c57812e Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@mellanox.com>
Date: Thu, 10 Mar 2016 23:10:21 +0100
Subject: mlxsw: pci: Implement reset done check

Firmware now tells us that the reset is done by passing a magic value
via register. Use it to shorten the wait in case this is supported.
With old firmware, we still wait until the timeout is reached.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 7992c55..7f4173c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -1681,11 +1681,18 @@ static const struct mlxsw_bus mlxsw_pci_bus = {
 
 static int mlxsw_pci_sw_reset(struct mlxsw_pci *mlxsw_pci)
 {
+	unsigned long end;
+
 	mlxsw_pci_write32(mlxsw_pci, SW_RESET, MLXSW_PCI_SW_RESET_RST_BIT);
-	/* Current firware does not let us know when the reset is done.
-	 * So we just wait here for constant time and hope for the best.
-	 */
-	msleep(MLXSW_PCI_SW_RESET_TIMEOUT_MSECS);
+	wmb(); /* reset needs to be written before we read control register */
+	end = jiffies + msecs_to_jiffies(MLXSW_PCI_SW_RESET_TIMEOUT_MSECS);
+	do {
+		u32 val = mlxsw_pci_read32(mlxsw_pci, FW_READY);
+
+		if ((val & MLXSW_PCI_FW_READY_MASK) == MLXSW_PCI_FW_READY_MAGIC)
+			break;
+		cond_resched();
+	} while (time_before(jiffies, end));
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.h b/drivers/net/ethernet/mellanox/mlxsw/pci.h
index 9121060..d942a3e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.h
@@ -61,6 +61,9 @@
 #define MLXSW_PCI_SW_RESET			0xF0010
 #define MLXSW_PCI_SW_RESET_RST_BIT		BIT(0)
 #define MLXSW_PCI_SW_RESET_TIMEOUT_MSECS	5000
+#define MLXSW_PCI_FW_READY			0xA1844
+#define MLXSW_PCI_FW_READY_MASK			0xFF
+#define MLXSW_PCI_FW_READY_MAGIC		0x5E
 
 #define MLXSW_PCI_DOORBELL_SDQ_OFFSET		0x000
 #define MLXSW_PCI_DOORBELL_RDQ_OFFSET		0x200
-- 
cgit v0.10.2


From c45569755e828a8458d125bb93feb1d90ac9be6f Mon Sep 17 00:00:00 2001
From: Chun-Hao Lin <hau@realtek.com>
Date: Fri, 11 Mar 2016 14:21:14 +0800
Subject: r8169:Remove unnecessary phy reset for pcie nic when setting link
 spped.

For pcie nic, after setting link speed and there is no link driver does not need
to do phy reset until link up.

For some pcie nics, to do this will also reset phy speed down counter and prevent
phy from auto speed down.

This patch fix the issue reported in following link.
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1547151

Signed-off-by: Chunhao Lin <hau@realtek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index dd2cf37..94f08f1 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -1999,7 +1999,8 @@ static int rtl8169_set_speed(struct net_device *dev,
 		goto out;
 
 	if (netif_running(dev) && (autoneg == AUTONEG_ENABLE) &&
-	    (advertising & ADVERTISED_1000baseT_Full)) {
+	    (advertising & ADVERTISED_1000baseT_Full) &&
+	    !pci_is_pcie(tp->pci_dev)) {
 		mod_timer(&tp->timer, jiffies + RTL8169_PHY_TIMEOUT);
 	}
 out:
-- 
cgit v0.10.2


From 885eb0a516e4d686849b91c5a1ba25c70b7a6540 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Fri, 11 Mar 2016 09:43:58 +0100
Subject: net: adjust napi_consume_skb to handle non-NAPI callers

Some drivers reuse/share code paths that free SKBs between NAPI
and non-NAPI calls. Adjust napi_consume_skb to handle this
use-case.

Before, calls from netpoll (w/ IRQs disabled) was handled and
indicated with a budget zero indication.  Use the same zero
indication to handle calls not originating from NAPI/softirq.
Simply handled by using dev_consume_skb_any().

This adds an extra branch+call for the netpoll case (checking
in_irq() + irqs_disabled()), but that is okay as this is a slowpath.

Suggested-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 51d768e..f044f97 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -801,9 +801,9 @@ void napi_consume_skb(struct sk_buff *skb, int budget)
 	if (unlikely(!skb))
 		return;
 
-	/* if budget is 0 assume netpoll w/ IRQs disabled */
+	/* Zero budget indicate non-NAPI context called us, like netpoll */
 	if (unlikely(!budget)) {
-		dev_consume_skb_irq(skb);
+		dev_consume_skb_any(skb);
 		return;
 	}
 
-- 
cgit v0.10.2


From b4a53379a0c19df784e30b1f35af5841f2b74f30 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Fri, 11 Mar 2016 09:44:08 +0100
Subject: mlx4: use napi_consume_skb API to get bulk free operations

Bulk free of SKBs happen transparently by the API call napi_consume_skb().
The napi budget parameter is usually needed by napi_consume_skb()
to detect if called from netpoll.  In this patch it has an extra meaning.

For mlx4 driver, the mlx4_en_stop_port() call is done outside
NAPI/softirq context, and cleanup the entire TX ring via
mlx4_en_free_tx_buf().  The code mlx4_en_free_tx_desc() for
freeing SKBs are shared with NAPI calls.

To handle this shared use the zero budget indication is reused,
and handled appropriately in napi_consume_skb(). To reflect this,
variable is called napi_mode for the function call that needed
this distinction.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index e0946ab..c0d7b72 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -276,7 +276,8 @@ static void mlx4_en_stamp_wqe(struct mlx4_en_priv *priv,
 
 static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
 				struct mlx4_en_tx_ring *ring,
-				int index, u8 owner, u64 timestamp)
+				int index, u8 owner, u64 timestamp,
+				int napi_mode)
 {
 	struct mlx4_en_tx_info *tx_info = &ring->tx_info[index];
 	struct mlx4_en_tx_desc *tx_desc = ring->buf + index * TXBB_SIZE;
@@ -347,7 +348,8 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
 			}
 		}
 	}
-	dev_consume_skb_any(skb);
+	napi_consume_skb(skb, napi_mode);
+
 	return tx_info->nr_txbb;
 }
 
@@ -371,7 +373,8 @@ int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring)
 	while (ring->cons != ring->prod) {
 		ring->last_nr_txbb = mlx4_en_free_tx_desc(priv, ring,
 						ring->cons & ring->size_mask,
-						!!(ring->cons & ring->size), 0);
+						!!(ring->cons & ring->size), 0,
+						0 /* Non-NAPI caller */);
 		ring->cons += ring->last_nr_txbb;
 		cnt++;
 	}
@@ -385,7 +388,7 @@ int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring)
 }
 
 static bool mlx4_en_process_tx_cq(struct net_device *dev,
-				 struct mlx4_en_cq *cq)
+				  struct mlx4_en_cq *cq, int napi_budget)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_cq *mcq = &cq->mcq;
@@ -451,7 +454,7 @@ static bool mlx4_en_process_tx_cq(struct net_device *dev,
 			last_nr_txbb = mlx4_en_free_tx_desc(
 					priv, ring, ring_index,
 					!!((ring_cons + txbbs_skipped) &
-					ring->size), timestamp);
+					ring->size), timestamp, napi_budget);
 
 			mlx4_en_stamp_wqe(priv, ring, stamp_index,
 					  !!((ring_cons + txbbs_stamp) &
@@ -511,7 +514,7 @@ int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	int clean_complete;
 
-	clean_complete = mlx4_en_process_tx_cq(dev, cq);
+	clean_complete = mlx4_en_process_tx_cq(dev, cq, budget);
 	if (!clean_complete)
 		return budget;
 
-- 
cgit v0.10.2


From 8ec736e556e3340b4b4295c7567b0766d6629702 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <brouer@redhat.com>
Date: Fri, 11 Mar 2016 09:44:17 +0100
Subject: mlx5: use napi_consume_skb API to get bulk free operations

Bulk free of SKBs happen transparently by the API call napi_consume_skb().
The napi budget parameter is needed by napi_consume_skb() to detect
if called from netpoll.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 0f76d32..0cb4a09 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -629,7 +629,7 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
 void mlx5e_completion_event(struct mlx5_core_cq *mcq);
 void mlx5e_cq_error_event(struct mlx5_core_cq *mcq, enum mlx5_event event);
 int mlx5e_napi_poll(struct napi_struct *napi, int budget);
-bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq);
+bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget);
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
 bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq);
 struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 94a14f8..1ffc7cb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -339,7 +339,7 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev)
 	return mlx5e_sq_xmit(sq, skb);
 }
 
-bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
+bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 {
 	struct mlx5e_sq *sq;
 	u32 dma_fifo_cc;
@@ -411,7 +411,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 			npkts++;
 			nbytes += wi->num_bytes;
 			sqcc += wi->num_wqebbs;
-			dev_kfree_skb(skb);
+			napi_consume_skb(skb, napi_budget);
 		} while (!last_wqe);
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
index 66d51a7..9bb4395 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
@@ -60,7 +60,7 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget)
 	clear_bit(MLX5E_CHANNEL_NAPI_SCHED, &c->flags);
 
 	for (i = 0; i < c->num_tc; i++)
-		busy |= mlx5e_poll_tx_cq(&c->sq[i].cq);
+		busy |= mlx5e_poll_tx_cq(&c->sq[i].cq, budget);
 
 	work_done = mlx5e_poll_rx_cq(&c->rq.cq, budget);
 	busy |= work_done == budget;
-- 
cgit v0.10.2


From 98267311fe3b334ae7c107fa0e2413adcf3ba735 Mon Sep 17 00:00:00 2001
From: Zefir Kurtisi <zefir.kurtisi@neratec.com>
Date: Fri, 11 Mar 2016 15:31:53 +0100
Subject: at803x: fix suspend/resume for SGMII link

When operating the at803x in SGMII mode, resuming the chip
from power down brings up the copper-side link but leaves
the SGMII link in unconnected state (tested with at8031
attached to gianfar). In effect, this caused a permanent
link loss once the related interface was put down.

This patch ensures that power down handling in supspend()
and resume() is also applied to the SGMII link.

Signed-off-by: Zefir Kurtisi <zefir.kurtisi@neratec.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/at803x.c b/drivers/net/phy/at803x.c
index 2174ec9..1e901c7 100644
--- a/drivers/net/phy/at803x.c
+++ b/drivers/net/phy/at803x.c
@@ -52,6 +52,9 @@
 #define AT803X_DEBUG_REG_5			0x05
 #define AT803X_DEBUG_TX_CLK_DLY_EN		BIT(8)
 
+#define AT803X_REG_CHIP_CONFIG			0x1f
+#define AT803X_BT_BX_REG_SEL			0x8000
+
 #define ATH8030_PHY_ID 0x004dd076
 #define ATH8031_PHY_ID 0x004dd074
 #define ATH8035_PHY_ID 0x004dd072
@@ -206,6 +209,7 @@ static int at803x_suspend(struct phy_device *phydev)
 {
 	int value;
 	int wol_enabled;
+	int ccr;
 
 	mutex_lock(&phydev->lock);
 
@@ -221,6 +225,16 @@ static int at803x_suspend(struct phy_device *phydev)
 
 	phy_write(phydev, MII_BMCR, value);
 
+	if (phydev->interface != PHY_INTERFACE_MODE_SGMII)
+		goto done;
+
+	/* also power-down SGMII interface */
+	ccr = phy_read(phydev, AT803X_REG_CHIP_CONFIG);
+	phy_write(phydev, AT803X_REG_CHIP_CONFIG, ccr & ~AT803X_BT_BX_REG_SEL);
+	phy_write(phydev, MII_BMCR, phy_read(phydev, MII_BMCR) | BMCR_PDOWN);
+	phy_write(phydev, AT803X_REG_CHIP_CONFIG, ccr | AT803X_BT_BX_REG_SEL);
+
+done:
 	mutex_unlock(&phydev->lock);
 
 	return 0;
@@ -229,6 +243,7 @@ static int at803x_suspend(struct phy_device *phydev)
 static int at803x_resume(struct phy_device *phydev)
 {
 	int value;
+	int ccr;
 
 	mutex_lock(&phydev->lock);
 
@@ -236,6 +251,17 @@ static int at803x_resume(struct phy_device *phydev)
 	value &= ~(BMCR_PDOWN | BMCR_ISOLATE);
 	phy_write(phydev, MII_BMCR, value);
 
+	if (phydev->interface != PHY_INTERFACE_MODE_SGMII)
+		goto done;
+
+	/* also power-up SGMII interface */
+	ccr = phy_read(phydev, AT803X_REG_CHIP_CONFIG);
+	phy_write(phydev, AT803X_REG_CHIP_CONFIG, ccr & ~AT803X_BT_BX_REG_SEL);
+	value = phy_read(phydev, MII_BMCR) & ~(BMCR_PDOWN | BMCR_ISOLATE);
+	phy_write(phydev, MII_BMCR, value);
+	phy_write(phydev, AT803X_REG_CHIP_CONFIG, ccr | AT803X_BT_BX_REG_SEL);
+
+done:
 	mutex_unlock(&phydev->lock);
 
 	return 0;
-- 
cgit v0.10.2


From f3c986908cc3b369b57f75de306c635a0074b76b Mon Sep 17 00:00:00 2001
From: "liping.zhang" <liping.zhang@spreadtrum.com>
Date: Fri, 11 Mar 2016 23:08:36 +0800
Subject: net: socket: use pr_info_once to tip the obsolete usage of PF_PACKET

There is no need to use the static variable here, pr_info_once is more
concise.

Signed-off-by: Liping Zhang <liping.zhang@spreadtrum.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/socket.c b/net/socket.c
index 886649c..c5ddc52 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1107,12 +1107,8 @@ int __sock_create(struct net *net, int family, int type, int protocol,
 	   deadlock in module load.
 	 */
 	if (family == PF_INET && type == SOCK_PACKET) {
-		static int warned;
-		if (!warned) {
-			warned = 1;
-			pr_info("%s uses obsolete (PF_INET,SOCK_PACKET)\n",
-				current->comm);
-		}
+		pr_info_once("%s uses obsolete (PF_INET,SOCK_PACKET)\n",
+			     current->comm);
 		family = PF_PACKET;
 	}
 
-- 
cgit v0.10.2


From dece8d2b78d19df7fe5e4e965f1f0d1a3e188d1b Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Fri, 11 Mar 2016 18:07:31 +0100
Subject: uapi: add MACsec bits

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild
index ebd10e6..e25ebcf 100644
--- a/include/uapi/linux/Kbuild
+++ b/include/uapi/linux/Kbuild
@@ -173,6 +173,7 @@ header-y += if_hippi.h
 header-y += if_infiniband.h
 header-y += if_link.h
 header-y += if_ltalk.h
+header-y += if_macsec.h
 header-y += if_packet.h
 header-y += if_phonet.h
 header-y += if_plip.h
diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h
index ea9221b..4a93051 100644
--- a/include/uapi/linux/if_ether.h
+++ b/include/uapi/linux/if_ether.h
@@ -83,6 +83,7 @@
 #define ETH_P_8021AD	0x88A8          /* 802.1ad Service VLAN		*/
 #define ETH_P_802_EX1	0x88B5		/* 802.1 Local Experimental 1.  */
 #define ETH_P_TIPC	0x88CA		/* TIPC 			*/
+#define ETH_P_MACSEC	0x88E5		/* 802.1ae MACsec */
 #define ETH_P_8021AH	0x88E7          /* 802.1ah Backbone Service Tag */
 #define ETH_P_MVRP	0x88F5          /* 802.1Q MVRP                  */
 #define ETH_P_1588	0x88F7		/* IEEE 1588 Timesync */
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 249eef9..8e3f88f 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -413,6 +413,35 @@ enum {
 
 #define IFLA_VRF_PORT_MAX (__IFLA_VRF_PORT_MAX - 1)
 
+/* MACSEC section */
+enum {
+	IFLA_MACSEC_UNSPEC,
+	IFLA_MACSEC_SCI,
+	IFLA_MACSEC_PORT,
+	IFLA_MACSEC_ICV_LEN,
+	IFLA_MACSEC_CIPHER_SUITE,
+	IFLA_MACSEC_WINDOW,
+	IFLA_MACSEC_ENCODING_SA,
+	IFLA_MACSEC_ENCRYPT,
+	IFLA_MACSEC_PROTECT,
+	IFLA_MACSEC_INC_SCI,
+	IFLA_MACSEC_ES,
+	IFLA_MACSEC_SCB,
+	IFLA_MACSEC_REPLAY_PROTECT,
+	IFLA_MACSEC_VALIDATION,
+	__IFLA_MACSEC_MAX,
+};
+
+#define IFLA_MACSEC_MAX (__IFLA_MACSEC_MAX - 1)
+
+enum macsec_validation_type {
+	MACSEC_VALIDATE_DISABLED = 0,
+	MACSEC_VALIDATE_CHECK = 1,
+	MACSEC_VALIDATE_STRICT = 2,
+	__MACSEC_VALIDATE_END,
+	MACSEC_VALIDATE_MAX = __MACSEC_VALIDATE_END - 1,
+};
+
 /* IPVLAN section */
 enum {
 	IFLA_IPVLAN_UNSPEC,
diff --git a/include/uapi/linux/if_macsec.h b/include/uapi/linux/if_macsec.h
new file mode 100644
index 0000000..26b0d1e
--- /dev/null
+++ b/include/uapi/linux/if_macsec.h
@@ -0,0 +1,161 @@
+/*
+ * include/uapi/linux/if_macsec.h - MACsec device
+ *
+ * Copyright (c) 2015 Sabrina Dubroca <sd@queasysnail.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _UAPI_MACSEC_H
+#define _UAPI_MACSEC_H
+
+#include <linux/types.h>
+
+#define MACSEC_GENL_NAME "macsec"
+#define MACSEC_GENL_VERSION 1
+
+#define MACSEC_MAX_KEY_LEN 128
+
+#define DEFAULT_CIPHER_ID   0x0080020001000001ULL
+#define DEFAULT_CIPHER_ALT  0x0080C20001000001ULL
+
+#define MACSEC_MIN_ICV_LEN 8
+#define MACSEC_MAX_ICV_LEN 32
+
+enum macsec_attrs {
+	MACSEC_ATTR_UNSPEC,
+	MACSEC_ATTR_IFINDEX,     /* u32, ifindex of the MACsec netdevice */
+	MACSEC_ATTR_RXSC_CONFIG, /* config, nested macsec_rxsc_attrs */
+	MACSEC_ATTR_SA_CONFIG,   /* config, nested macsec_sa_attrs */
+	MACSEC_ATTR_SECY,        /* dump, nested macsec_secy_attrs */
+	MACSEC_ATTR_TXSA_LIST,   /* dump, nested, macsec_sa_attrs for each TXSA */
+	MACSEC_ATTR_RXSC_LIST,   /* dump, nested, macsec_rxsc_attrs for each RXSC */
+	MACSEC_ATTR_TXSC_STATS,  /* dump, nested, macsec_txsc_stats_attr */
+	MACSEC_ATTR_SECY_STATS,  /* dump, nested, macsec_secy_stats_attr */
+	__MACSEC_ATTR_END,
+	NUM_MACSEC_ATTR = __MACSEC_ATTR_END,
+	MACSEC_ATTR_MAX = __MACSEC_ATTR_END - 1,
+};
+
+enum macsec_secy_attrs {
+	MACSEC_SECY_ATTR_UNSPEC,
+	MACSEC_SECY_ATTR_SCI,
+	MACSEC_SECY_ATTR_ENCODING_SA,
+	MACSEC_SECY_ATTR_WINDOW,
+	MACSEC_SECY_ATTR_CIPHER_SUITE,
+	MACSEC_SECY_ATTR_ICV_LEN,
+	MACSEC_SECY_ATTR_PROTECT,
+	MACSEC_SECY_ATTR_REPLAY,
+	MACSEC_SECY_ATTR_OPER,
+	MACSEC_SECY_ATTR_VALIDATE,
+	MACSEC_SECY_ATTR_ENCRYPT,
+	MACSEC_SECY_ATTR_INC_SCI,
+	MACSEC_SECY_ATTR_ES,
+	MACSEC_SECY_ATTR_SCB,
+	__MACSEC_SECY_ATTR_END,
+	NUM_MACSEC_SECY_ATTR = __MACSEC_SECY_ATTR_END,
+	MACSEC_SECY_ATTR_MAX = __MACSEC_SECY_ATTR_END - 1,
+};
+
+enum macsec_rxsc_attrs {
+	MACSEC_RXSC_ATTR_UNSPEC,
+	MACSEC_RXSC_ATTR_SCI,     /* config/dump, u64 */
+	MACSEC_RXSC_ATTR_ACTIVE,  /* config/dump, u8 0..1 */
+	MACSEC_RXSC_ATTR_SA_LIST, /* dump, nested */
+	MACSEC_RXSC_ATTR_STATS,   /* dump, nested, macsec_rxsc_stats_attr */
+	__MACSEC_RXSC_ATTR_END,
+	NUM_MACSEC_RXSC_ATTR = __MACSEC_RXSC_ATTR_END,
+	MACSEC_RXSC_ATTR_MAX = __MACSEC_RXSC_ATTR_END - 1,
+};
+
+enum macsec_sa_attrs {
+	MACSEC_SA_ATTR_UNSPEC,
+	MACSEC_SA_ATTR_AN,     /* config/dump, u8 0..3 */
+	MACSEC_SA_ATTR_ACTIVE, /* config/dump, u8 0..1 */
+	MACSEC_SA_ATTR_PN,     /* config/dump, u32 */
+	MACSEC_SA_ATTR_KEY,    /* config, data */
+	MACSEC_SA_ATTR_KEYID,  /* config/dump, u64 */
+	MACSEC_SA_ATTR_STATS,  /* dump, nested, macsec_sa_stats_attr */
+	__MACSEC_SA_ATTR_END,
+	NUM_MACSEC_SA_ATTR = __MACSEC_SA_ATTR_END,
+	MACSEC_SA_ATTR_MAX = __MACSEC_SA_ATTR_END - 1,
+};
+
+enum macsec_nl_commands {
+	MACSEC_CMD_GET_TXSC,
+	MACSEC_CMD_ADD_RXSC,
+	MACSEC_CMD_DEL_RXSC,
+	MACSEC_CMD_UPD_RXSC,
+	MACSEC_CMD_ADD_TXSA,
+	MACSEC_CMD_DEL_TXSA,
+	MACSEC_CMD_UPD_TXSA,
+	MACSEC_CMD_ADD_RXSA,
+	MACSEC_CMD_DEL_RXSA,
+	MACSEC_CMD_UPD_RXSA,
+};
+
+/* u64 per-RXSC stats */
+enum macsec_rxsc_stats_attr {
+	MACSEC_RXSC_STATS_ATTR_UNSPEC,
+	MACSEC_RXSC_STATS_ATTR_IN_OCTETS_VALIDATED,
+	MACSEC_RXSC_STATS_ATTR_IN_OCTETS_DECRYPTED,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_UNCHECKED,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_DELAYED,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_OK,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_INVALID,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_LATE,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_NOT_VALID,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_NOT_USING_SA,
+	MACSEC_RXSC_STATS_ATTR_IN_PKTS_UNUSED_SA,
+	__MACSEC_RXSC_STATS_ATTR_END,
+	NUM_MACSEC_RXSC_STATS_ATTR = __MACSEC_RXSC_STATS_ATTR_END,
+	MACSEC_RXSC_STATS_ATTR_MAX = __MACSEC_RXSC_STATS_ATTR_END - 1,
+};
+
+/* u32 per-{RX,TX}SA stats */
+enum macsec_sa_stats_attr {
+	MACSEC_SA_STATS_ATTR_UNSPEC,
+	MACSEC_SA_STATS_ATTR_IN_PKTS_OK,
+	MACSEC_SA_STATS_ATTR_IN_PKTS_INVALID,
+	MACSEC_SA_STATS_ATTR_IN_PKTS_NOT_VALID,
+	MACSEC_SA_STATS_ATTR_IN_PKTS_NOT_USING_SA,
+	MACSEC_SA_STATS_ATTR_IN_PKTS_UNUSED_SA,
+	MACSEC_SA_STATS_ATTR_OUT_PKTS_PROTECTED,
+	MACSEC_SA_STATS_ATTR_OUT_PKTS_ENCRYPTED,
+	__MACSEC_SA_STATS_ATTR_END,
+	NUM_MACSEC_SA_STATS_ATTR = __MACSEC_SA_STATS_ATTR_END,
+	MACSEC_SA_STATS_ATTR_MAX = __MACSEC_SA_STATS_ATTR_END - 1,
+};
+
+/* u64 per-TXSC stats */
+enum macsec_txsc_stats_attr {
+	MACSEC_TXSC_STATS_ATTR_UNSPEC,
+	MACSEC_TXSC_STATS_ATTR_OUT_PKTS_PROTECTED,
+	MACSEC_TXSC_STATS_ATTR_OUT_PKTS_ENCRYPTED,
+	MACSEC_TXSC_STATS_ATTR_OUT_OCTETS_PROTECTED,
+	MACSEC_TXSC_STATS_ATTR_OUT_OCTETS_ENCRYPTED,
+	__MACSEC_TXSC_STATS_ATTR_END,
+	NUM_MACSEC_TXSC_STATS_ATTR = __MACSEC_TXSC_STATS_ATTR_END,
+	MACSEC_TXSC_STATS_ATTR_MAX = __MACSEC_TXSC_STATS_ATTR_END - 1,
+};
+
+/* u64 per-SecY stats */
+enum macsec_secy_stats_attr {
+	MACSEC_SECY_STATS_ATTR_UNSPEC,
+	MACSEC_SECY_STATS_ATTR_OUT_PKTS_UNTAGGED,
+	MACSEC_SECY_STATS_ATTR_IN_PKTS_UNTAGGED,
+	MACSEC_SECY_STATS_ATTR_OUT_PKTS_TOO_LONG,
+	MACSEC_SECY_STATS_ATTR_IN_PKTS_NO_TAG,
+	MACSEC_SECY_STATS_ATTR_IN_PKTS_BAD_TAG,
+	MACSEC_SECY_STATS_ATTR_IN_PKTS_UNKNOWN_SCI,
+	MACSEC_SECY_STATS_ATTR_IN_PKTS_NO_SCI,
+	MACSEC_SECY_STATS_ATTR_IN_PKTS_OVERRUN,
+	__MACSEC_SECY_STATS_ATTR_END,
+	NUM_MACSEC_SECY_STATS_ATTR = __MACSEC_SECY_STATS_ATTR_END,
+	MACSEC_SECY_STATS_ATTR_MAX = __MACSEC_SECY_STATS_ATTR_END - 1,
+};
+
+#endif /* _UAPI_MACSEC_H */
-- 
cgit v0.10.2


From 3c17578473b9be5a6e7680a45ea97e1d56e13249 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Fri, 11 Mar 2016 18:07:32 +0100
Subject: net: add MACsec netdevice priv_flags and helper

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 41df0b4..be693b3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1328,6 +1328,7 @@ struct net_device_ops {
  * @IFF_RXFH_CONFIGURED: device has had Rx Flow indirection table configured
  * @IFF_PHONY_HEADROOM: the headroom value is controlled by an external
  *	entity (i.e. the master device for bridged veth)
+ * @IFF_MACSEC: device is a MACsec device
  */
 enum netdev_priv_flags {
 	IFF_802_1Q_VLAN			= 1<<0,
@@ -1357,6 +1358,7 @@ enum netdev_priv_flags {
 	IFF_TEAM			= 1<<24,
 	IFF_RXFH_CONFIGURED		= 1<<25,
 	IFF_PHONY_HEADROOM		= 1<<26,
+	IFF_MACSEC			= 1<<27,
 };
 
 #define IFF_802_1Q_VLAN			IFF_802_1Q_VLAN
@@ -1385,6 +1387,7 @@ enum netdev_priv_flags {
 #define IFF_L3MDEV_SLAVE		IFF_L3MDEV_SLAVE
 #define IFF_TEAM			IFF_TEAM
 #define IFF_RXFH_CONFIGURED		IFF_RXFH_CONFIGURED
+#define IFF_MACSEC			IFF_MACSEC
 
 /**
  *	struct net_device - The DEVICE structure.
@@ -4045,6 +4048,11 @@ static inline void skb_gso_error_unwind(struct sk_buff *skb, __be16 protocol,
 	skb->mac_len = mac_len;
 }
 
+static inline bool netif_is_macsec(const struct net_device *dev)
+{
+	return dev->priv_flags & IFF_MACSEC;
+}
+
 static inline bool netif_is_macvlan(const struct net_device *dev)
 {
 	return dev->priv_flags & IFF_MACVLAN;
-- 
cgit v0.10.2


From c09440f7dcb304002dfced8c0fea289eb25f2da0 Mon Sep 17 00:00:00 2001
From: Sabrina Dubroca <sd@queasysnail.net>
Date: Fri, 11 Mar 2016 18:07:33 +0100
Subject: macsec: introduce IEEE 802.1AE driver

This is an implementation of MACsec/IEEE 802.1AE.  This driver
provides authentication and encryption of traffic in a LAN, typically
with GCM-AES-128, and optional replay protection.

http://standards.ieee.org/getieee802/download/802.1AE-2006.pdf

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index f184fb5..2a1ba62b 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -193,6 +193,13 @@ config GENEVE
 	  To compile this driver as a module, choose M here: the module
 	  will be called geneve.
 
+config MACSEC
+	tristate "IEEE 802.1AE MAC-level encryption (MACsec)"
+	select CRYPTO_AES
+	select CRYPTO_GCM
+	---help---
+	   MACsec is an encryption standard for Ethernet.
+
 config NETCONSOLE
 	tristate "Network console logging support"
 	---help---
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 900b0c5..1aa7cb8 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -10,6 +10,7 @@ obj-$(CONFIG_IPVLAN) += ipvlan/
 obj-$(CONFIG_DUMMY) += dummy.o
 obj-$(CONFIG_EQUALIZER) += eql.o
 obj-$(CONFIG_IFB) += ifb.o
+obj-$(CONFIG_MACSEC) += macsec.o
 obj-$(CONFIG_MACVLAN) += macvlan.o
 obj-$(CONFIG_MACVTAP) += macvtap.o
 obj-$(CONFIG_MII) += mii.o
diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
new file mode 100644
index 0000000..84d3e5c
--- /dev/null
+++ b/drivers/net/macsec.c
@@ -0,0 +1,3297 @@
+/*
+ * drivers/net/macsec.c - MACsec device
+ *
+ * Copyright (c) 2015 Sabrina Dubroca <sd@queasysnail.net>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/types.h>
+#include <linux/skbuff.h>
+#include <linux/socket.h>
+#include <linux/module.h>
+#include <crypto/aead.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <net/genetlink.h>
+#include <net/sock.h>
+
+#include <uapi/linux/if_macsec.h>
+
+typedef u64 __bitwise sci_t;
+
+#define MACSEC_SCI_LEN 8
+
+/* SecTAG length = macsec_eth_header without the optional SCI */
+#define MACSEC_TAG_LEN 6
+
+struct macsec_eth_header {
+	struct ethhdr eth;
+	/* SecTAG */
+	u8  tci_an;
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	u8  short_length:6,
+		  unused:2;
+#elif defined(__BIG_ENDIAN_BITFIELD)
+	u8        unused:2,
+	    short_length:6;
+#else
+#error	"Please fix <asm/byteorder.h>"
+#endif
+	__be32 packet_number;
+	u8 secure_channel_id[8]; /* optional */
+} __packed;
+
+#define MACSEC_TCI_VERSION 0x80
+#define MACSEC_TCI_ES      0x40 /* end station */
+#define MACSEC_TCI_SC      0x20 /* SCI present */
+#define MACSEC_TCI_SCB     0x10 /* epon */
+#define MACSEC_TCI_E       0x08 /* encryption */
+#define MACSEC_TCI_C       0x04 /* changed text */
+#define MACSEC_AN_MASK     0x03 /* association number */
+#define MACSEC_TCI_CONFID  (MACSEC_TCI_E | MACSEC_TCI_C)
+
+/* minimum secure data length deemed "not short", see IEEE 802.1AE-2006 9.7 */
+#define MIN_NON_SHORT_LEN 48
+
+#define GCM_AES_IV_LEN 12
+#define DEFAULT_ICV_LEN 16
+
+#define MACSEC_NUM_AN 4 /* 2 bits for the association number */
+
+#define for_each_rxsc(secy, sc)			\
+	for (sc = rcu_dereference_bh(secy->rx_sc);	\
+	     sc;				\
+	     sc = rcu_dereference_bh(sc->next))
+#define for_each_rxsc_rtnl(secy, sc)			\
+	for (sc = rtnl_dereference(secy->rx_sc);	\
+	     sc;					\
+	     sc = rtnl_dereference(sc->next))
+
+struct gcm_iv {
+	union {
+		u8 secure_channel_id[8];
+		sci_t sci;
+	};
+	__be32 pn;
+};
+
+/**
+ * struct macsec_key - SA key
+ * @id: user-provided key identifier
+ * @tfm: crypto struct, key storage
+ */
+struct macsec_key {
+	u64 id;
+	struct crypto_aead *tfm;
+};
+
+struct macsec_rx_sc_stats {
+	__u64 InOctetsValidated;
+	__u64 InOctetsDecrypted;
+	__u64 InPktsUnchecked;
+	__u64 InPktsDelayed;
+	__u64 InPktsOK;
+	__u64 InPktsInvalid;
+	__u64 InPktsLate;
+	__u64 InPktsNotValid;
+	__u64 InPktsNotUsingSA;
+	__u64 InPktsUnusedSA;
+};
+
+struct macsec_rx_sa_stats {
+	__u32 InPktsOK;
+	__u32 InPktsInvalid;
+	__u32 InPktsNotValid;
+	__u32 InPktsNotUsingSA;
+	__u32 InPktsUnusedSA;
+};
+
+struct macsec_tx_sa_stats {
+	__u32 OutPktsProtected;
+	__u32 OutPktsEncrypted;
+};
+
+struct macsec_tx_sc_stats {
+	__u64 OutPktsProtected;
+	__u64 OutPktsEncrypted;
+	__u64 OutOctetsProtected;
+	__u64 OutOctetsEncrypted;
+};
+
+struct macsec_dev_stats {
+	__u64 OutPktsUntagged;
+	__u64 InPktsUntagged;
+	__u64 OutPktsTooLong;
+	__u64 InPktsNoTag;
+	__u64 InPktsBadTag;
+	__u64 InPktsUnknownSCI;
+	__u64 InPktsNoSCI;
+	__u64 InPktsOverrun;
+};
+
+/**
+ * struct macsec_rx_sa - receive secure association
+ * @active:
+ * @next_pn: packet number expected for the next packet
+ * @lock: protects next_pn manipulations
+ * @key: key structure
+ * @stats: per-SA stats
+ */
+struct macsec_rx_sa {
+	struct macsec_key key;
+	spinlock_t lock;
+	u32 next_pn;
+	atomic_t refcnt;
+	bool active;
+	struct macsec_rx_sa_stats __percpu *stats;
+	struct macsec_rx_sc *sc;
+	struct rcu_head rcu;
+};
+
+struct pcpu_rx_sc_stats {
+	struct macsec_rx_sc_stats stats;
+	struct u64_stats_sync syncp;
+};
+
+/**
+ * struct macsec_rx_sc - receive secure channel
+ * @sci: secure channel identifier for this SC
+ * @active: channel is active
+ * @sa: array of secure associations
+ * @stats: per-SC stats
+ */
+struct macsec_rx_sc {
+	struct macsec_rx_sc __rcu *next;
+	sci_t sci;
+	bool active;
+	struct macsec_rx_sa __rcu *sa[MACSEC_NUM_AN];
+	struct pcpu_rx_sc_stats __percpu *stats;
+	atomic_t refcnt;
+	struct rcu_head rcu_head;
+};
+
+/**
+ * struct macsec_tx_sa - transmit secure association
+ * @active:
+ * @next_pn: packet number to use for the next packet
+ * @lock: protects next_pn manipulations
+ * @key: key structure
+ * @stats: per-SA stats
+ */
+struct macsec_tx_sa {
+	struct macsec_key key;
+	spinlock_t lock;
+	u32 next_pn;
+	atomic_t refcnt;
+	bool active;
+	struct macsec_tx_sa_stats __percpu *stats;
+	struct rcu_head rcu;
+};
+
+struct pcpu_tx_sc_stats {
+	struct macsec_tx_sc_stats stats;
+	struct u64_stats_sync syncp;
+};
+
+/**
+ * struct macsec_tx_sc - transmit secure channel
+ * @active:
+ * @encoding_sa: association number of the SA currently in use
+ * @encrypt: encrypt packets on transmit, or authenticate only
+ * @send_sci: always include the SCI in the SecTAG
+ * @end_station:
+ * @scb: single copy broadcast flag
+ * @sa: array of secure associations
+ * @stats: stats for this TXSC
+ */
+struct macsec_tx_sc {
+	bool active;
+	u8 encoding_sa;
+	bool encrypt;
+	bool send_sci;
+	bool end_station;
+	bool scb;
+	struct macsec_tx_sa __rcu *sa[MACSEC_NUM_AN];
+	struct pcpu_tx_sc_stats __percpu *stats;
+};
+
+#define MACSEC_VALIDATE_DEFAULT MACSEC_VALIDATE_STRICT
+
+/**
+ * struct macsec_secy - MACsec Security Entity
+ * @netdev: netdevice for this SecY
+ * @n_rx_sc: number of receive secure channels configured on this SecY
+ * @sci: secure channel identifier used for tx
+ * @key_len: length of keys used by the cipher suite
+ * @icv_len: length of ICV used by the cipher suite
+ * @validate_frames: validation mode
+ * @operational: MAC_Operational flag
+ * @protect_frames: enable protection for this SecY
+ * @replay_protect: enable packet number checks on receive
+ * @replay_window: size of the replay window
+ * @tx_sc: transmit secure channel
+ * @rx_sc: linked list of receive secure channels
+ */
+struct macsec_secy {
+	struct net_device *netdev;
+	unsigned int n_rx_sc;
+	sci_t sci;
+	u16 key_len;
+	u16 icv_len;
+	enum macsec_validation_type validate_frames;
+	bool operational;
+	bool protect_frames;
+	bool replay_protect;
+	u32 replay_window;
+	struct macsec_tx_sc tx_sc;
+	struct macsec_rx_sc __rcu *rx_sc;
+};
+
+struct pcpu_secy_stats {
+	struct macsec_dev_stats stats;
+	struct u64_stats_sync syncp;
+};
+
+/**
+ * struct macsec_dev - private data
+ * @secy: SecY config
+ * @real_dev: pointer to underlying netdevice
+ * @stats: MACsec device stats
+ * @secys: linked list of SecY's on the underlying device
+ */
+struct macsec_dev {
+	struct macsec_secy secy;
+	struct net_device *real_dev;
+	struct pcpu_secy_stats __percpu *stats;
+	struct list_head secys;
+};
+
+/**
+ * struct macsec_rxh_data - rx_handler private argument
+ * @secys: linked list of SecY's on this underlying device
+ */
+struct macsec_rxh_data {
+	struct list_head secys;
+};
+
+static struct macsec_dev *macsec_priv(const struct net_device *dev)
+{
+	return (struct macsec_dev *)netdev_priv(dev);
+}
+
+static struct macsec_rxh_data *macsec_data_rcu(const struct net_device *dev)
+{
+	return rcu_dereference_bh(dev->rx_handler_data);
+}
+
+static struct macsec_rxh_data *macsec_data_rtnl(const struct net_device *dev)
+{
+	return rtnl_dereference(dev->rx_handler_data);
+}
+
+struct macsec_cb {
+	struct aead_request *req;
+	union {
+		struct macsec_tx_sa *tx_sa;
+		struct macsec_rx_sa *rx_sa;
+	};
+	u8 assoc_num;
+	bool valid;
+	bool has_sci;
+};
+
+static struct macsec_rx_sa *macsec_rxsa_get(struct macsec_rx_sa __rcu *ptr)
+{
+	struct macsec_rx_sa *sa = rcu_dereference_bh(ptr);
+
+	if (!sa || !sa->active)
+		return NULL;
+
+	if (!atomic_inc_not_zero(&sa->refcnt))
+		return NULL;
+
+	return sa;
+}
+
+static void free_rx_sc_rcu(struct rcu_head *head)
+{
+	struct macsec_rx_sc *rx_sc = container_of(head, struct macsec_rx_sc, rcu_head);
+
+	free_percpu(rx_sc->stats);
+	kfree(rx_sc);
+}
+
+static struct macsec_rx_sc *macsec_rxsc_get(struct macsec_rx_sc *sc)
+{
+	return atomic_inc_not_zero(&sc->refcnt) ? sc : NULL;
+}
+
+static void macsec_rxsc_put(struct macsec_rx_sc *sc)
+{
+	if (atomic_dec_and_test(&sc->refcnt))
+		call_rcu(&sc->rcu_head, free_rx_sc_rcu);
+}
+
+static void free_rxsa(struct rcu_head *head)
+{
+	struct macsec_rx_sa *sa = container_of(head, struct macsec_rx_sa, rcu);
+
+	crypto_free_aead(sa->key.tfm);
+	free_percpu(sa->stats);
+	macsec_rxsc_put(sa->sc);
+	kfree(sa);
+}
+
+static void macsec_rxsa_put(struct macsec_rx_sa *sa)
+{
+	if (atomic_dec_and_test(&sa->refcnt))
+		call_rcu(&sa->rcu, free_rxsa);
+}
+
+static struct macsec_tx_sa *macsec_txsa_get(struct macsec_tx_sa __rcu *ptr)
+{
+	struct macsec_tx_sa *sa = rcu_dereference_bh(ptr);
+
+	if (!sa || !sa->active)
+		return NULL;
+
+	if (!atomic_inc_not_zero(&sa->refcnt))
+		return NULL;
+
+	return sa;
+}
+
+static void free_txsa(struct rcu_head *head)
+{
+	struct macsec_tx_sa *sa = container_of(head, struct macsec_tx_sa, rcu);
+
+	crypto_free_aead(sa->key.tfm);
+	free_percpu(sa->stats);
+	kfree(sa);
+}
+
+static void macsec_txsa_put(struct macsec_tx_sa *sa)
+{
+	if (atomic_dec_and_test(&sa->refcnt))
+		call_rcu(&sa->rcu, free_txsa);
+}
+
+static struct macsec_cb *macsec_skb_cb(struct sk_buff *skb)
+{
+	BUILD_BUG_ON(sizeof(struct macsec_cb) > sizeof(skb->cb));
+	return (struct macsec_cb *)skb->cb;
+}
+
+#define MACSEC_PORT_ES (htons(0x0001))
+#define MACSEC_PORT_SCB (0x0000)
+#define MACSEC_UNDEF_SCI ((__force sci_t)0xffffffffffffffffULL)
+
+#define DEFAULT_SAK_LEN 16
+#define DEFAULT_SEND_SCI true
+#define DEFAULT_ENCRYPT false
+#define DEFAULT_ENCODING_SA 0
+
+static sci_t make_sci(u8 *addr, __be16 port)
+{
+	sci_t sci;
+
+	memcpy(&sci, addr, ETH_ALEN);
+	memcpy(((char *)&sci) + ETH_ALEN, &port, sizeof(port));
+
+	return sci;
+}
+
+static sci_t macsec_frame_sci(struct macsec_eth_header *hdr, bool sci_present)
+{
+	sci_t sci;
+
+	if (sci_present)
+		memcpy(&sci, hdr->secure_channel_id,
+		       sizeof(hdr->secure_channel_id));
+	else
+		sci = make_sci(hdr->eth.h_source, MACSEC_PORT_ES);
+
+	return sci;
+}
+
+static unsigned int macsec_sectag_len(bool sci_present)
+{
+	return MACSEC_TAG_LEN + (sci_present ? MACSEC_SCI_LEN : 0);
+}
+
+static unsigned int macsec_hdr_len(bool sci_present)
+{
+	return macsec_sectag_len(sci_present) + ETH_HLEN;
+}
+
+static unsigned int macsec_extra_len(bool sci_present)
+{
+	return macsec_sectag_len(sci_present) + sizeof(__be16);
+}
+
+/* Fill SecTAG according to IEEE 802.1AE-2006 10.5.3 */
+static void macsec_fill_sectag(struct macsec_eth_header *h,
+			       const struct macsec_secy *secy, u32 pn)
+{
+	const struct macsec_tx_sc *tx_sc = &secy->tx_sc;
+
+	memset(&h->tci_an, 0, macsec_sectag_len(tx_sc->send_sci));
+	h->eth.h_proto = htons(ETH_P_MACSEC);
+
+	if (tx_sc->send_sci ||
+	    (secy->n_rx_sc > 1 && !tx_sc->end_station && !tx_sc->scb)) {
+		h->tci_an |= MACSEC_TCI_SC;
+		memcpy(&h->secure_channel_id, &secy->sci,
+		       sizeof(h->secure_channel_id));
+	} else {
+		if (tx_sc->end_station)
+			h->tci_an |= MACSEC_TCI_ES;
+		if (tx_sc->scb)
+			h->tci_an |= MACSEC_TCI_SCB;
+	}
+
+	h->packet_number = htonl(pn);
+
+	/* with GCM, C/E clear for !encrypt, both set for encrypt */
+	if (tx_sc->encrypt)
+		h->tci_an |= MACSEC_TCI_CONFID;
+	else if (secy->icv_len != DEFAULT_ICV_LEN)
+		h->tci_an |= MACSEC_TCI_C;
+
+	h->tci_an |= tx_sc->encoding_sa;
+}
+
+static void macsec_set_shortlen(struct macsec_eth_header *h, size_t data_len)
+{
+	if (data_len < MIN_NON_SHORT_LEN)
+		h->short_length = data_len;
+}
+
+/* validate MACsec packet according to IEEE 802.1AE-2006 9.12 */
+static bool macsec_validate_skb(struct sk_buff *skb, u16 icv_len)
+{
+	struct macsec_eth_header *h = (struct macsec_eth_header *)skb->data;
+	int len = skb->len - 2 * ETH_ALEN;
+	int extra_len = macsec_extra_len(!!(h->tci_an & MACSEC_TCI_SC)) + icv_len;
+
+	/* a) It comprises at least 17 octets */
+	if (skb->len <= 16)
+		return false;
+
+	/* b) MACsec EtherType: already checked */
+
+	/* c) V bit is clear */
+	if (h->tci_an & MACSEC_TCI_VERSION)
+		return false;
+
+	/* d) ES or SCB => !SC */
+	if ((h->tci_an & MACSEC_TCI_ES || h->tci_an & MACSEC_TCI_SCB) &&
+	    (h->tci_an & MACSEC_TCI_SC))
+		return false;
+
+	/* e) Bits 7 and 8 of octet 4 of the SecTAG are clear */
+	if (h->unused)
+		return false;
+
+	/* rx.pn != 0 (figure 10-5) */
+	if (!h->packet_number)
+		return false;
+
+	/* length check, f) g) h) i) */
+	if (h->short_length)
+		return len == extra_len + h->short_length;
+	return len >= extra_len + MIN_NON_SHORT_LEN;
+}
+
+#define MACSEC_NEEDED_HEADROOM (macsec_extra_len(true))
+#define MACSEC_NEEDED_TAILROOM MACSEC_MAX_ICV_LEN
+
+static void macsec_fill_iv(unsigned char *iv, sci_t sci, u32 pn)
+{
+	struct gcm_iv *gcm_iv = (struct gcm_iv *)iv;
+
+	gcm_iv->sci = sci;
+	gcm_iv->pn = htonl(pn);
+}
+
+static struct macsec_eth_header *macsec_ethhdr(struct sk_buff *skb)
+{
+	return (struct macsec_eth_header *)skb_mac_header(skb);
+}
+
+static u32 tx_sa_update_pn(struct macsec_tx_sa *tx_sa, struct macsec_secy *secy)
+{
+	u32 pn;
+
+	spin_lock_bh(&tx_sa->lock);
+	pn = tx_sa->next_pn;
+
+	tx_sa->next_pn++;
+	if (tx_sa->next_pn == 0) {
+		pr_debug("PN wrapped, transitioning to !oper\n");
+		tx_sa->active = false;
+		if (secy->protect_frames)
+			secy->operational = false;
+	}
+	spin_unlock_bh(&tx_sa->lock);
+
+	return pn;
+}
+
+static void macsec_encrypt_finish(struct sk_buff *skb, struct net_device *dev)
+{
+	struct macsec_dev *macsec = netdev_priv(dev);
+
+	skb->dev = macsec->real_dev;
+	skb_reset_mac_header(skb);
+	skb->protocol = eth_hdr(skb)->h_proto;
+}
+
+static void macsec_count_tx(struct sk_buff *skb, struct macsec_tx_sc *tx_sc,
+			    struct macsec_tx_sa *tx_sa)
+{
+	struct pcpu_tx_sc_stats *txsc_stats = this_cpu_ptr(tx_sc->stats);
+
+	u64_stats_update_begin(&txsc_stats->syncp);
+	if (tx_sc->encrypt) {
+		txsc_stats->stats.OutOctetsEncrypted += skb->len;
+		txsc_stats->stats.OutPktsEncrypted++;
+		this_cpu_inc(tx_sa->stats->OutPktsEncrypted);
+	} else {
+		txsc_stats->stats.OutOctetsProtected += skb->len;
+		txsc_stats->stats.OutPktsProtected++;
+		this_cpu_inc(tx_sa->stats->OutPktsProtected);
+	}
+	u64_stats_update_end(&txsc_stats->syncp);
+}
+
+static void count_tx(struct net_device *dev, int ret, int len)
+{
+	if (likely(ret == NET_XMIT_SUCCESS || ret == NET_XMIT_CN)) {
+		struct pcpu_sw_netstats *stats = this_cpu_ptr(dev->tstats);
+
+		u64_stats_update_begin(&stats->syncp);
+		stats->tx_packets++;
+		stats->tx_bytes += len;
+		u64_stats_update_end(&stats->syncp);
+	} else {
+		dev->stats.tx_dropped++;
+	}
+}
+
+static void macsec_encrypt_done(struct crypto_async_request *base, int err)
+{
+	struct sk_buff *skb = base->data;
+	struct net_device *dev = skb->dev;
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct macsec_tx_sa *sa = macsec_skb_cb(skb)->tx_sa;
+	int len, ret;
+
+	aead_request_free(macsec_skb_cb(skb)->req);
+
+	rcu_read_lock_bh();
+	macsec_encrypt_finish(skb, dev);
+	macsec_count_tx(skb, &macsec->secy.tx_sc, macsec_skb_cb(skb)->tx_sa);
+	len = skb->len;
+	ret = dev_queue_xmit(skb);
+	count_tx(dev, ret, len);
+	rcu_read_unlock_bh();
+
+	macsec_txsa_put(sa);
+	dev_put(dev);
+}
+
+static struct sk_buff *macsec_encrypt(struct sk_buff *skb,
+				      struct net_device *dev)
+{
+	int ret;
+	struct scatterlist sg[MAX_SKB_FRAGS + 1];
+	unsigned char iv[GCM_AES_IV_LEN];
+	struct ethhdr *eth;
+	struct macsec_eth_header *hh;
+	size_t unprotected_len;
+	struct aead_request *req;
+	struct macsec_secy *secy;
+	struct macsec_tx_sc *tx_sc;
+	struct macsec_tx_sa *tx_sa;
+	struct macsec_dev *macsec = macsec_priv(dev);
+	u32 pn;
+
+	secy = &macsec->secy;
+	tx_sc = &secy->tx_sc;
+
+	/* 10.5.1 TX SA assignment */
+	tx_sa = macsec_txsa_get(tx_sc->sa[tx_sc->encoding_sa]);
+	if (!tx_sa) {
+		secy->operational = false;
+		kfree_skb(skb);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (unlikely(skb_headroom(skb) < MACSEC_NEEDED_HEADROOM ||
+		     skb_tailroom(skb) < MACSEC_NEEDED_TAILROOM)) {
+		struct sk_buff *nskb = skb_copy_expand(skb,
+						       MACSEC_NEEDED_HEADROOM,
+						       MACSEC_NEEDED_TAILROOM,
+						       GFP_ATOMIC);
+		if (likely(nskb)) {
+			consume_skb(skb);
+			skb = nskb;
+		} else {
+			macsec_txsa_put(tx_sa);
+			kfree_skb(skb);
+			return ERR_PTR(-ENOMEM);
+		}
+	} else {
+		skb = skb_unshare(skb, GFP_ATOMIC);
+		if (!skb) {
+			macsec_txsa_put(tx_sa);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+
+	unprotected_len = skb->len;
+	eth = eth_hdr(skb);
+	hh = (struct macsec_eth_header *)skb_push(skb, macsec_extra_len(tx_sc->send_sci));
+	memmove(hh, eth, 2 * ETH_ALEN);
+
+	pn = tx_sa_update_pn(tx_sa, secy);
+	if (pn == 0) {
+		macsec_txsa_put(tx_sa);
+		kfree_skb(skb);
+		return ERR_PTR(-ENOLINK);
+	}
+	macsec_fill_sectag(hh, secy, pn);
+	macsec_set_shortlen(hh, unprotected_len - 2 * ETH_ALEN);
+
+	macsec_fill_iv(iv, secy->sci, pn);
+
+	skb_put(skb, secy->icv_len);
+
+	if (skb->len - ETH_HLEN > macsec_priv(dev)->real_dev->mtu) {
+		struct pcpu_secy_stats *secy_stats = this_cpu_ptr(macsec->stats);
+
+		u64_stats_update_begin(&secy_stats->syncp);
+		secy_stats->stats.OutPktsTooLong++;
+		u64_stats_update_end(&secy_stats->syncp);
+
+		macsec_txsa_put(tx_sa);
+		kfree_skb(skb);
+		return ERR_PTR(-EINVAL);
+	}
+
+	req = aead_request_alloc(tx_sa->key.tfm, GFP_ATOMIC);
+	if (!req) {
+		macsec_txsa_put(tx_sa);
+		kfree_skb(skb);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	sg_init_table(sg, MAX_SKB_FRAGS + 1);
+	skb_to_sgvec(skb, sg, 0, skb->len);
+
+	if (tx_sc->encrypt) {
+		int len = skb->len - macsec_hdr_len(tx_sc->send_sci) -
+			  secy->icv_len;
+		aead_request_set_crypt(req, sg, sg, len, iv);
+		aead_request_set_ad(req, macsec_hdr_len(tx_sc->send_sci));
+	} else {
+		aead_request_set_crypt(req, sg, sg, 0, iv);
+		aead_request_set_ad(req, skb->len - secy->icv_len);
+	}
+
+	macsec_skb_cb(skb)->req = req;
+	macsec_skb_cb(skb)->tx_sa = tx_sa;
+	aead_request_set_callback(req, 0, macsec_encrypt_done, skb);
+
+	dev_hold(skb->dev);
+	ret = crypto_aead_encrypt(req);
+	if (ret == -EINPROGRESS) {
+		return ERR_PTR(ret);
+	} else if (ret != 0) {
+		dev_put(skb->dev);
+		kfree_skb(skb);
+		aead_request_free(req);
+		macsec_txsa_put(tx_sa);
+		return ERR_PTR(-EINVAL);
+	}
+
+	dev_put(skb->dev);
+	aead_request_free(req);
+	macsec_txsa_put(tx_sa);
+
+	return skb;
+}
+
+static bool macsec_post_decrypt(struct sk_buff *skb, struct macsec_secy *secy, u32 pn)
+{
+	struct macsec_rx_sa *rx_sa = macsec_skb_cb(skb)->rx_sa;
+	struct pcpu_rx_sc_stats *rxsc_stats = this_cpu_ptr(rx_sa->sc->stats);
+	struct macsec_eth_header *hdr = macsec_ethhdr(skb);
+	u32 lowest_pn = 0;
+
+	spin_lock(&rx_sa->lock);
+	if (rx_sa->next_pn >= secy->replay_window)
+		lowest_pn = rx_sa->next_pn - secy->replay_window;
+
+	/* Now perform replay protection check again
+	 * (see IEEE 802.1AE-2006 figure 10-5)
+	 */
+	if (secy->replay_protect && pn < lowest_pn) {
+		spin_unlock(&rx_sa->lock);
+		u64_stats_update_begin(&rxsc_stats->syncp);
+		rxsc_stats->stats.InPktsLate++;
+		u64_stats_update_end(&rxsc_stats->syncp);
+		return false;
+	}
+
+	if (secy->validate_frames != MACSEC_VALIDATE_DISABLED) {
+		u64_stats_update_begin(&rxsc_stats->syncp);
+		if (hdr->tci_an & MACSEC_TCI_E)
+			rxsc_stats->stats.InOctetsDecrypted += skb->len;
+		else
+			rxsc_stats->stats.InOctetsValidated += skb->len;
+		u64_stats_update_end(&rxsc_stats->syncp);
+	}
+
+	if (!macsec_skb_cb(skb)->valid) {
+		spin_unlock(&rx_sa->lock);
+
+		/* 10.6.5 */
+		if (hdr->tci_an & MACSEC_TCI_C ||
+		    secy->validate_frames == MACSEC_VALIDATE_STRICT) {
+			u64_stats_update_begin(&rxsc_stats->syncp);
+			rxsc_stats->stats.InPktsNotValid++;
+			u64_stats_update_end(&rxsc_stats->syncp);
+			return false;
+		}
+
+		u64_stats_update_begin(&rxsc_stats->syncp);
+		if (secy->validate_frames == MACSEC_VALIDATE_CHECK) {
+			rxsc_stats->stats.InPktsInvalid++;
+			this_cpu_inc(rx_sa->stats->InPktsInvalid);
+		} else if (pn < lowest_pn) {
+			rxsc_stats->stats.InPktsDelayed++;
+		} else {
+			rxsc_stats->stats.InPktsUnchecked++;
+		}
+		u64_stats_update_end(&rxsc_stats->syncp);
+	} else {
+		u64_stats_update_begin(&rxsc_stats->syncp);
+		if (pn < lowest_pn) {
+			rxsc_stats->stats.InPktsDelayed++;
+		} else {
+			rxsc_stats->stats.InPktsOK++;
+			this_cpu_inc(rx_sa->stats->InPktsOK);
+		}
+		u64_stats_update_end(&rxsc_stats->syncp);
+
+		if (pn >= rx_sa->next_pn)
+			rx_sa->next_pn = pn + 1;
+		spin_unlock(&rx_sa->lock);
+	}
+
+	return true;
+}
+
+static void macsec_reset_skb(struct sk_buff *skb, struct net_device *dev)
+{
+	skb->pkt_type = PACKET_HOST;
+	skb->protocol = eth_type_trans(skb, dev);
+
+	skb_reset_network_header(skb);
+	if (!skb_transport_header_was_set(skb))
+		skb_reset_transport_header(skb);
+	skb_reset_mac_len(skb);
+}
+
+static void macsec_finalize_skb(struct sk_buff *skb, u8 icv_len, u8 hdr_len)
+{
+	memmove(skb->data + hdr_len, skb->data, 2 * ETH_ALEN);
+	skb_pull(skb, hdr_len);
+	pskb_trim_unique(skb, skb->len - icv_len);
+}
+
+static void count_rx(struct net_device *dev, int len)
+{
+	struct pcpu_sw_netstats *stats = this_cpu_ptr(dev->tstats);
+
+	u64_stats_update_begin(&stats->syncp);
+	stats->rx_packets++;
+	stats->rx_bytes += len;
+	u64_stats_update_end(&stats->syncp);
+}
+
+static void macsec_decrypt_done(struct crypto_async_request *base, int err)
+{
+	struct sk_buff *skb = base->data;
+	struct net_device *dev = skb->dev;
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct macsec_rx_sa *rx_sa = macsec_skb_cb(skb)->rx_sa;
+	int len, ret;
+	u32 pn;
+
+	aead_request_free(macsec_skb_cb(skb)->req);
+
+	rcu_read_lock_bh();
+	pn = ntohl(macsec_ethhdr(skb)->packet_number);
+	if (!macsec_post_decrypt(skb, &macsec->secy, pn)) {
+		rcu_read_unlock_bh();
+		kfree_skb(skb);
+		goto out;
+	}
+
+	macsec_finalize_skb(skb, macsec->secy.icv_len,
+			    macsec_extra_len(macsec_skb_cb(skb)->has_sci));
+	macsec_reset_skb(skb, macsec->secy.netdev);
+
+	len = skb->len;
+	ret = netif_rx(skb);
+	if (ret == NET_RX_SUCCESS)
+		count_rx(dev, len);
+	else
+		macsec->secy.netdev->stats.rx_dropped++;
+
+	rcu_read_unlock_bh();
+
+out:
+	macsec_rxsa_put(rx_sa);
+	dev_put(dev);
+	return;
+}
+
+static struct sk_buff *macsec_decrypt(struct sk_buff *skb,
+				      struct net_device *dev,
+				      struct macsec_rx_sa *rx_sa,
+				      sci_t sci,
+				      struct macsec_secy *secy)
+{
+	int ret;
+	struct scatterlist sg[MAX_SKB_FRAGS + 1];
+	unsigned char iv[GCM_AES_IV_LEN];
+	struct aead_request *req;
+	struct macsec_eth_header *hdr;
+	u16 icv_len = secy->icv_len;
+
+	macsec_skb_cb(skb)->valid = false;
+	skb = skb_share_check(skb, GFP_ATOMIC);
+	if (!skb)
+		return NULL;
+
+	req = aead_request_alloc(rx_sa->key.tfm, GFP_ATOMIC);
+	if (!req) {
+		kfree_skb(skb);
+		return NULL;
+	}
+
+	hdr = (struct macsec_eth_header *)skb->data;
+	macsec_fill_iv(iv, sci, ntohl(hdr->packet_number));
+
+	sg_init_table(sg, MAX_SKB_FRAGS + 1);
+	skb_to_sgvec(skb, sg, 0, skb->len);
+
+	if (hdr->tci_an & MACSEC_TCI_E) {
+		/* confidentiality: ethernet + macsec header
+		 * authenticated, encrypted payload
+		 */
+		int len = skb->len - macsec_hdr_len(macsec_skb_cb(skb)->has_sci);
+
+		aead_request_set_crypt(req, sg, sg, len, iv);
+		aead_request_set_ad(req, macsec_hdr_len(macsec_skb_cb(skb)->has_sci));
+		skb = skb_unshare(skb, GFP_ATOMIC);
+		if (!skb) {
+			aead_request_free(req);
+			return NULL;
+		}
+	} else {
+		/* integrity only: all headers + data authenticated */
+		aead_request_set_crypt(req, sg, sg, icv_len, iv);
+		aead_request_set_ad(req, skb->len - icv_len);
+	}
+
+	macsec_skb_cb(skb)->req = req;
+	macsec_skb_cb(skb)->rx_sa = rx_sa;
+	skb->dev = dev;
+	aead_request_set_callback(req, 0, macsec_decrypt_done, skb);
+
+	dev_hold(dev);
+	ret = crypto_aead_decrypt(req);
+	if (ret == -EINPROGRESS) {
+		return NULL;
+	} else if (ret != 0) {
+		/* decryption/authentication failed
+		 * 10.6 if validateFrames is disabled, deliver anyway
+		 */
+		if (ret != -EBADMSG) {
+			kfree_skb(skb);
+			skb = NULL;
+		}
+	} else {
+		macsec_skb_cb(skb)->valid = true;
+	}
+	dev_put(dev);
+
+	aead_request_free(req);
+
+	return skb;
+}
+
+static struct macsec_rx_sc *find_rx_sc(struct macsec_secy *secy, sci_t sci)
+{
+	struct macsec_rx_sc *rx_sc;
+
+	for_each_rxsc(secy, rx_sc) {
+		if (rx_sc->sci == sci)
+			return rx_sc;
+	}
+
+	return NULL;
+}
+
+static struct macsec_rx_sc *find_rx_sc_rtnl(struct macsec_secy *secy, sci_t sci)
+{
+	struct macsec_rx_sc *rx_sc;
+
+	for_each_rxsc_rtnl(secy, rx_sc) {
+		if (rx_sc->sci == sci)
+			return rx_sc;
+	}
+
+	return NULL;
+}
+
+static void handle_not_macsec(struct sk_buff *skb)
+{
+	struct macsec_rxh_data *rxd;
+	struct macsec_dev *macsec;
+
+	rcu_read_lock();
+	rxd = macsec_data_rcu(skb->dev);
+
+	/* 10.6 If the management control validateFrames is not
+	 * Strict, frames without a SecTAG are received, counted, and
+	 * delivered to the Controlled Port
+	 */
+	list_for_each_entry_rcu(macsec, &rxd->secys, secys) {
+		struct sk_buff *nskb;
+		int ret;
+		struct pcpu_secy_stats *secy_stats = this_cpu_ptr(macsec->stats);
+
+		if (macsec->secy.validate_frames == MACSEC_VALIDATE_STRICT) {
+			u64_stats_update_begin(&secy_stats->syncp);
+			secy_stats->stats.InPktsNoTag++;
+			u64_stats_update_end(&secy_stats->syncp);
+			continue;
+		}
+
+		/* deliver on this port */
+		nskb = skb_clone(skb, GFP_ATOMIC);
+		if (!nskb)
+			break;
+
+		nskb->dev = macsec->secy.netdev;
+
+		ret = netif_rx(nskb);
+		if (ret == NET_RX_SUCCESS) {
+			u64_stats_update_begin(&secy_stats->syncp);
+			secy_stats->stats.InPktsUntagged++;
+			u64_stats_update_end(&secy_stats->syncp);
+		} else {
+			macsec->secy.netdev->stats.rx_dropped++;
+		}
+	}
+
+	rcu_read_unlock();
+}
+
+static rx_handler_result_t macsec_handle_frame(struct sk_buff **pskb)
+{
+	struct sk_buff *skb = *pskb;
+	struct net_device *dev = skb->dev;
+	struct macsec_eth_header *hdr;
+	struct macsec_secy *secy = NULL;
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_rx_sa *rx_sa;
+	struct macsec_rxh_data *rxd;
+	struct macsec_dev *macsec;
+	sci_t sci;
+	u32 pn;
+	bool cbit;
+	struct pcpu_rx_sc_stats *rxsc_stats;
+	struct pcpu_secy_stats *secy_stats;
+	bool pulled_sci;
+
+	if (skb_headroom(skb) < ETH_HLEN)
+		goto drop_direct;
+
+	hdr = macsec_ethhdr(skb);
+	if (hdr->eth.h_proto != htons(ETH_P_MACSEC)) {
+		handle_not_macsec(skb);
+
+		/* and deliver to the uncontrolled port */
+		return RX_HANDLER_PASS;
+	}
+
+	skb = skb_unshare(skb, GFP_ATOMIC);
+	if (!skb) {
+		*pskb = NULL;
+		return RX_HANDLER_CONSUMED;
+	}
+
+	pulled_sci = pskb_may_pull(skb, macsec_extra_len(true));
+	if (!pulled_sci) {
+		if (!pskb_may_pull(skb, macsec_extra_len(false)))
+			goto drop_direct;
+	}
+
+	hdr = macsec_ethhdr(skb);
+
+	/* Frames with a SecTAG that has the TCI E bit set but the C
+	 * bit clear are discarded, as this reserved encoding is used
+	 * to identify frames with a SecTAG that are not to be
+	 * delivered to the Controlled Port.
+	 */
+	if ((hdr->tci_an & (MACSEC_TCI_C | MACSEC_TCI_E)) == MACSEC_TCI_E)
+		return RX_HANDLER_PASS;
+
+	/* now, pull the extra length */
+	if (hdr->tci_an & MACSEC_TCI_SC) {
+		if (!pulled_sci)
+			goto drop_direct;
+	}
+
+	/* ethernet header is part of crypto processing */
+	skb_push(skb, ETH_HLEN);
+
+	macsec_skb_cb(skb)->has_sci = !!(hdr->tci_an & MACSEC_TCI_SC);
+	macsec_skb_cb(skb)->assoc_num = hdr->tci_an & MACSEC_AN_MASK;
+	sci = macsec_frame_sci(hdr, macsec_skb_cb(skb)->has_sci);
+
+	rcu_read_lock();
+	rxd = macsec_data_rcu(skb->dev);
+
+	list_for_each_entry_rcu(macsec, &rxd->secys, secys) {
+		struct macsec_rx_sc *sc = find_rx_sc(&macsec->secy, sci);
+
+		if (sc) {
+			secy = &macsec->secy;
+			rx_sc = sc;
+			break;
+		}
+	}
+
+	if (!secy)
+		goto nosci;
+
+	dev = secy->netdev;
+	macsec = macsec_priv(dev);
+	secy_stats = this_cpu_ptr(macsec->stats);
+	rxsc_stats = this_cpu_ptr(rx_sc->stats);
+
+	if (!macsec_validate_skb(skb, secy->icv_len)) {
+		u64_stats_update_begin(&secy_stats->syncp);
+		secy_stats->stats.InPktsBadTag++;
+		u64_stats_update_end(&secy_stats->syncp);
+		goto drop_nosa;
+	}
+
+	rx_sa = macsec_rxsa_get(rx_sc->sa[macsec_skb_cb(skb)->assoc_num]);
+	if (!rx_sa) {
+		/* 10.6.1 if the SA is not in use */
+
+		/* If validateFrames is Strict or the C bit in the
+		 * SecTAG is set, discard
+		 */
+		if (hdr->tci_an & MACSEC_TCI_C ||
+		    secy->validate_frames == MACSEC_VALIDATE_STRICT) {
+			u64_stats_update_begin(&rxsc_stats->syncp);
+			rxsc_stats->stats.InPktsNotUsingSA++;
+			u64_stats_update_end(&rxsc_stats->syncp);
+			goto drop_nosa;
+		}
+
+		/* not Strict, the frame (with the SecTAG and ICV
+		 * removed) is delivered to the Controlled Port.
+		 */
+		u64_stats_update_begin(&rxsc_stats->syncp);
+		rxsc_stats->stats.InPktsUnusedSA++;
+		u64_stats_update_end(&rxsc_stats->syncp);
+		goto deliver;
+	}
+
+	/* First, PN check to avoid decrypting obviously wrong packets */
+	pn = ntohl(hdr->packet_number);
+	if (secy->replay_protect) {
+		bool late;
+
+		spin_lock(&rx_sa->lock);
+		late = rx_sa->next_pn >= secy->replay_window &&
+		       pn < (rx_sa->next_pn - secy->replay_window);
+		spin_unlock(&rx_sa->lock);
+
+		if (late) {
+			u64_stats_update_begin(&rxsc_stats->syncp);
+			rxsc_stats->stats.InPktsLate++;
+			u64_stats_update_end(&rxsc_stats->syncp);
+			goto drop;
+		}
+	}
+
+	/* Disabled && !changed text => skip validation */
+	if (hdr->tci_an & MACSEC_TCI_C ||
+	    secy->validate_frames != MACSEC_VALIDATE_DISABLED)
+		skb = macsec_decrypt(skb, dev, rx_sa, sci, secy);
+
+	if (!skb) {
+		macsec_rxsa_put(rx_sa);
+		rcu_read_unlock();
+		*pskb = NULL;
+		return RX_HANDLER_CONSUMED;
+	}
+
+	if (!macsec_post_decrypt(skb, secy, pn))
+		goto drop;
+
+deliver:
+	macsec_finalize_skb(skb, secy->icv_len,
+			    macsec_extra_len(macsec_skb_cb(skb)->has_sci));
+	macsec_reset_skb(skb, secy->netdev);
+
+	macsec_rxsa_put(rx_sa);
+	count_rx(dev, skb->len);
+
+	rcu_read_unlock();
+
+	*pskb = skb;
+	return RX_HANDLER_ANOTHER;
+
+drop:
+	macsec_rxsa_put(rx_sa);
+drop_nosa:
+	rcu_read_unlock();
+drop_direct:
+	kfree_skb(skb);
+	*pskb = NULL;
+	return RX_HANDLER_CONSUMED;
+
+nosci:
+	/* 10.6.1 if the SC is not found */
+	cbit = !!(hdr->tci_an & MACSEC_TCI_C);
+	if (!cbit)
+		macsec_finalize_skb(skb, DEFAULT_ICV_LEN,
+				    macsec_extra_len(macsec_skb_cb(skb)->has_sci));
+
+	list_for_each_entry_rcu(macsec, &rxd->secys, secys) {
+		struct sk_buff *nskb;
+		int ret;
+
+		secy_stats = this_cpu_ptr(macsec->stats);
+
+		/* If validateFrames is Strict or the C bit in the
+		 * SecTAG is set, discard
+		 */
+		if (cbit ||
+		    macsec->secy.validate_frames == MACSEC_VALIDATE_STRICT) {
+			u64_stats_update_begin(&secy_stats->syncp);
+			secy_stats->stats.InPktsNoSCI++;
+			u64_stats_update_end(&secy_stats->syncp);
+			continue;
+		}
+
+		/* not strict, the frame (with the SecTAG and ICV
+		 * removed) is delivered to the Controlled Port.
+		 */
+		nskb = skb_clone(skb, GFP_ATOMIC);
+		if (!nskb)
+			break;
+
+		macsec_reset_skb(nskb, macsec->secy.netdev);
+
+		ret = netif_rx(nskb);
+		if (ret == NET_RX_SUCCESS) {
+			u64_stats_update_begin(&secy_stats->syncp);
+			secy_stats->stats.InPktsUnknownSCI++;
+			u64_stats_update_end(&secy_stats->syncp);
+		} else {
+			macsec->secy.netdev->stats.rx_dropped++;
+		}
+	}
+
+	rcu_read_unlock();
+	*pskb = skb;
+	return RX_HANDLER_PASS;
+}
+
+static struct crypto_aead *macsec_alloc_tfm(char *key, int key_len, int icv_len)
+{
+	struct crypto_aead *tfm;
+	int ret;
+
+	tfm = crypto_alloc_aead("gcm(aes)", 0, CRYPTO_ALG_ASYNC);
+	if (!tfm || IS_ERR(tfm))
+		return NULL;
+
+	ret = crypto_aead_setkey(tfm, key, key_len);
+	if (ret < 0) {
+		crypto_free_aead(tfm);
+		return NULL;
+	}
+
+	ret = crypto_aead_setauthsize(tfm, icv_len);
+	if (ret < 0) {
+		crypto_free_aead(tfm);
+		return NULL;
+	}
+
+	return tfm;
+}
+
+static int init_rx_sa(struct macsec_rx_sa *rx_sa, char *sak, int key_len,
+		      int icv_len)
+{
+	rx_sa->stats = alloc_percpu(struct macsec_rx_sa_stats);
+	if (!rx_sa->stats)
+		return -1;
+
+	rx_sa->key.tfm = macsec_alloc_tfm(sak, key_len, icv_len);
+	if (!rx_sa->key.tfm) {
+		free_percpu(rx_sa->stats);
+		return -1;
+	}
+
+	rx_sa->active = false;
+	rx_sa->next_pn = 1;
+	atomic_set(&rx_sa->refcnt, 1);
+	spin_lock_init(&rx_sa->lock);
+
+	return 0;
+}
+
+static void clear_rx_sa(struct macsec_rx_sa *rx_sa)
+{
+	rx_sa->active = false;
+
+	macsec_rxsa_put(rx_sa);
+}
+
+static void free_rx_sc(struct macsec_rx_sc *rx_sc)
+{
+	int i;
+
+	for (i = 0; i < MACSEC_NUM_AN; i++) {
+		struct macsec_rx_sa *sa = rtnl_dereference(rx_sc->sa[i]);
+
+		RCU_INIT_POINTER(rx_sc->sa[i], NULL);
+		if (sa)
+			clear_rx_sa(sa);
+	}
+
+	macsec_rxsc_put(rx_sc);
+}
+
+static struct macsec_rx_sc *del_rx_sc(struct macsec_secy *secy, sci_t sci)
+{
+	struct macsec_rx_sc *rx_sc, __rcu **rx_scp;
+
+	for (rx_scp = &secy->rx_sc, rx_sc = rtnl_dereference(*rx_scp);
+	     rx_sc;
+	     rx_scp = &rx_sc->next, rx_sc = rtnl_dereference(*rx_scp)) {
+		if (rx_sc->sci == sci) {
+			if (rx_sc->active)
+				secy->n_rx_sc--;
+			rcu_assign_pointer(*rx_scp, rx_sc->next);
+			return rx_sc;
+		}
+	}
+
+	return NULL;
+}
+
+static struct macsec_rx_sc *create_rx_sc(struct net_device *dev, sci_t sci)
+{
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_dev *macsec;
+	struct net_device *real_dev = macsec_priv(dev)->real_dev;
+	struct macsec_rxh_data *rxd = macsec_data_rtnl(real_dev);
+	struct macsec_secy *secy;
+
+	list_for_each_entry(macsec, &rxd->secys, secys) {
+		if (find_rx_sc_rtnl(&macsec->secy, sci))
+			return ERR_PTR(-EEXIST);
+	}
+
+	rx_sc = kzalloc(sizeof(*rx_sc), GFP_KERNEL);
+	if (!rx_sc)
+		return ERR_PTR(-ENOMEM);
+
+	rx_sc->stats = netdev_alloc_pcpu_stats(struct pcpu_rx_sc_stats);
+	if (!rx_sc->stats) {
+		kfree(rx_sc);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	rx_sc->sci = sci;
+	rx_sc->active = true;
+	atomic_set(&rx_sc->refcnt, 1);
+
+	secy = &macsec_priv(dev)->secy;
+	rcu_assign_pointer(rx_sc->next, secy->rx_sc);
+	rcu_assign_pointer(secy->rx_sc, rx_sc);
+
+	if (rx_sc->active)
+		secy->n_rx_sc++;
+
+	return rx_sc;
+}
+
+static int init_tx_sa(struct macsec_tx_sa *tx_sa, char *sak, int key_len,
+		      int icv_len)
+{
+	tx_sa->stats = alloc_percpu(struct macsec_tx_sa_stats);
+	if (!tx_sa->stats)
+		return -1;
+
+	tx_sa->key.tfm = macsec_alloc_tfm(sak, key_len, icv_len);
+	if (!tx_sa->key.tfm) {
+		free_percpu(tx_sa->stats);
+		return -1;
+	}
+
+	tx_sa->active = false;
+	atomic_set(&tx_sa->refcnt, 1);
+	spin_lock_init(&tx_sa->lock);
+
+	return 0;
+}
+
+static void clear_tx_sa(struct macsec_tx_sa *tx_sa)
+{
+	tx_sa->active = false;
+
+	macsec_txsa_put(tx_sa);
+}
+
+static struct genl_family macsec_fam = {
+	.id		= GENL_ID_GENERATE,
+	.name		= MACSEC_GENL_NAME,
+	.hdrsize	= 0,
+	.version	= MACSEC_GENL_VERSION,
+	.maxattr	= MACSEC_ATTR_MAX,
+	.netnsok	= true,
+};
+
+static struct net_device *get_dev_from_nl(struct net *net,
+					  struct nlattr **attrs)
+{
+	int ifindex = nla_get_u32(attrs[MACSEC_ATTR_IFINDEX]);
+	struct net_device *dev;
+
+	dev = __dev_get_by_index(net, ifindex);
+	if (!dev)
+		return ERR_PTR(-ENODEV);
+
+	if (!netif_is_macsec(dev))
+		return ERR_PTR(-ENODEV);
+
+	return dev;
+}
+
+static sci_t nla_get_sci(const struct nlattr *nla)
+{
+	return (__force sci_t)nla_get_u64(nla);
+}
+
+static int nla_put_sci(struct sk_buff *skb, int attrtype, sci_t value)
+{
+	return nla_put_u64(skb, attrtype, (__force u64)value);
+}
+
+static struct macsec_tx_sa *get_txsa_from_nl(struct net *net,
+					     struct nlattr **attrs,
+					     struct nlattr **tb_sa,
+					     struct net_device **devp,
+					     struct macsec_secy **secyp,
+					     struct macsec_tx_sc **scp,
+					     u8 *assoc_num)
+{
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_tx_sc *tx_sc;
+	struct macsec_tx_sa *tx_sa;
+
+	if (!tb_sa[MACSEC_SA_ATTR_AN])
+		return ERR_PTR(-EINVAL);
+
+	*assoc_num = nla_get_u8(tb_sa[MACSEC_SA_ATTR_AN]);
+
+	dev = get_dev_from_nl(net, attrs);
+	if (IS_ERR(dev))
+		return ERR_CAST(dev);
+
+	if (*assoc_num >= MACSEC_NUM_AN)
+		return ERR_PTR(-EINVAL);
+
+	secy = &macsec_priv(dev)->secy;
+	tx_sc = &secy->tx_sc;
+
+	tx_sa = rtnl_dereference(tx_sc->sa[*assoc_num]);
+	if (!tx_sa)
+		return ERR_PTR(-ENODEV);
+
+	*devp = dev;
+	*scp = tx_sc;
+	*secyp = secy;
+	return tx_sa;
+}
+
+static struct macsec_rx_sc *get_rxsc_from_nl(struct net *net,
+					     struct nlattr **attrs,
+					     struct nlattr **tb_rxsc,
+					     struct net_device **devp,
+					     struct macsec_secy **secyp)
+{
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	sci_t sci;
+
+	dev = get_dev_from_nl(net, attrs);
+	if (IS_ERR(dev))
+		return ERR_CAST(dev);
+
+	secy = &macsec_priv(dev)->secy;
+
+	if (!tb_rxsc[MACSEC_RXSC_ATTR_SCI])
+		return ERR_PTR(-EINVAL);
+
+	sci = nla_get_sci(tb_rxsc[MACSEC_RXSC_ATTR_SCI]);
+	rx_sc = find_rx_sc_rtnl(secy, sci);
+	if (!rx_sc)
+		return ERR_PTR(-ENODEV);
+
+	*secyp = secy;
+	*devp = dev;
+
+	return rx_sc;
+}
+
+static struct macsec_rx_sa *get_rxsa_from_nl(struct net *net,
+					     struct nlattr **attrs,
+					     struct nlattr **tb_rxsc,
+					     struct nlattr **tb_sa,
+					     struct net_device **devp,
+					     struct macsec_secy **secyp,
+					     struct macsec_rx_sc **scp,
+					     u8 *assoc_num)
+{
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_rx_sa *rx_sa;
+
+	if (!tb_sa[MACSEC_SA_ATTR_AN])
+		return ERR_PTR(-EINVAL);
+
+	*assoc_num = nla_get_u8(tb_sa[MACSEC_SA_ATTR_AN]);
+	if (*assoc_num >= MACSEC_NUM_AN)
+		return ERR_PTR(-EINVAL);
+
+	rx_sc = get_rxsc_from_nl(net, attrs, tb_rxsc, devp, secyp);
+	if (IS_ERR(rx_sc))
+		return ERR_CAST(rx_sc);
+
+	rx_sa = rtnl_dereference(rx_sc->sa[*assoc_num]);
+	if (!rx_sa)
+		return ERR_PTR(-ENODEV);
+
+	*scp = rx_sc;
+	return rx_sa;
+}
+
+
+static const struct nla_policy macsec_genl_policy[NUM_MACSEC_ATTR] = {
+	[MACSEC_ATTR_IFINDEX] = { .type = NLA_U32 },
+	[MACSEC_ATTR_RXSC_CONFIG] = { .type = NLA_NESTED },
+	[MACSEC_ATTR_SA_CONFIG] = { .type = NLA_NESTED },
+};
+
+static const struct nla_policy macsec_genl_rxsc_policy[NUM_MACSEC_RXSC_ATTR] = {
+	[MACSEC_RXSC_ATTR_SCI] = { .type = NLA_U64 },
+	[MACSEC_RXSC_ATTR_ACTIVE] = { .type = NLA_U8 },
+};
+
+static const struct nla_policy macsec_genl_sa_policy[NUM_MACSEC_SA_ATTR] = {
+	[MACSEC_SA_ATTR_AN] = { .type = NLA_U8 },
+	[MACSEC_SA_ATTR_ACTIVE] = { .type = NLA_U8 },
+	[MACSEC_SA_ATTR_PN] = { .type = NLA_U32 },
+	[MACSEC_SA_ATTR_KEYID] = { .type = NLA_U64 },
+	[MACSEC_SA_ATTR_KEY] = { .type = NLA_BINARY,
+				 .len = MACSEC_MAX_KEY_LEN, },
+};
+
+static int parse_sa_config(struct nlattr **attrs, struct nlattr **tb_sa)
+{
+	if (!attrs[MACSEC_ATTR_SA_CONFIG])
+		return -EINVAL;
+
+	if (nla_parse_nested(tb_sa, MACSEC_SA_ATTR_MAX, attrs[MACSEC_ATTR_SA_CONFIG],
+			     macsec_genl_sa_policy))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int parse_rxsc_config(struct nlattr **attrs, struct nlattr **tb_rxsc)
+{
+	if (!attrs[MACSEC_ATTR_RXSC_CONFIG])
+		return -EINVAL;
+
+	if (nla_parse_nested(tb_rxsc, MACSEC_RXSC_ATTR_MAX, attrs[MACSEC_ATTR_RXSC_CONFIG],
+			     macsec_genl_rxsc_policy))
+		return -EINVAL;
+
+	return 0;
+}
+
+static bool validate_add_rxsa(struct nlattr **attrs)
+{
+	if (!attrs[MACSEC_SA_ATTR_AN] ||
+	    !attrs[MACSEC_SA_ATTR_KEY] ||
+	    !attrs[MACSEC_SA_ATTR_KEYID])
+		return false;
+
+	if (nla_get_u8(attrs[MACSEC_SA_ATTR_AN]) >= MACSEC_NUM_AN)
+		return false;
+
+	if (attrs[MACSEC_SA_ATTR_PN] && nla_get_u32(attrs[MACSEC_SA_ATTR_PN]) == 0)
+		return false;
+
+	if (attrs[MACSEC_SA_ATTR_ACTIVE]) {
+		if (nla_get_u8(attrs[MACSEC_SA_ATTR_ACTIVE]) > 1)
+			return false;
+	}
+
+	return true;
+}
+
+static int macsec_add_rxsa(struct sk_buff *skb, struct genl_info *info)
+{
+	struct net_device *dev;
+	struct nlattr **attrs = info->attrs;
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_rx_sa *rx_sa;
+	unsigned char assoc_num;
+	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_sa_config(attrs, tb_sa))
+		return -EINVAL;
+
+	if (parse_rxsc_config(attrs, tb_rxsc))
+		return -EINVAL;
+
+	if (!validate_add_rxsa(tb_sa))
+		return -EINVAL;
+
+	rtnl_lock();
+	rx_sc = get_rxsc_from_nl(genl_info_net(info), attrs, tb_rxsc, &dev, &secy);
+	if (IS_ERR(rx_sc) || !macsec_rxsc_get(rx_sc)) {
+		rtnl_unlock();
+		return PTR_ERR(rx_sc);
+	}
+
+	assoc_num = nla_get_u8(tb_sa[MACSEC_SA_ATTR_AN]);
+
+	if (nla_len(tb_sa[MACSEC_SA_ATTR_KEY]) != secy->key_len) {
+		pr_notice("macsec: nl: add_rxsa: bad key length: %d != %d\n",
+			  nla_len(tb_sa[MACSEC_SA_ATTR_KEY]), secy->key_len);
+		rtnl_unlock();
+		return -EINVAL;
+	}
+
+	rx_sa = rtnl_dereference(rx_sc->sa[assoc_num]);
+	if (rx_sa) {
+		rtnl_unlock();
+		return -EBUSY;
+	}
+
+	rx_sa = kmalloc(sizeof(*rx_sa), GFP_KERNEL);
+	if (init_rx_sa(rx_sa, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]), secy->key_len,
+		       secy->icv_len)) {
+		rtnl_unlock();
+		return -ENOMEM;
+	}
+
+	if (tb_sa[MACSEC_SA_ATTR_PN]) {
+		spin_lock_bh(&rx_sa->lock);
+		rx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
+		spin_unlock_bh(&rx_sa->lock);
+	}
+
+	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
+		rx_sa->active = !!nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
+
+	rx_sa->key.id = nla_get_u64(tb_sa[MACSEC_SA_ATTR_KEYID]);
+	rx_sa->sc = rx_sc;
+	rcu_assign_pointer(rx_sc->sa[assoc_num], rx_sa);
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static bool validate_add_rxsc(struct nlattr **attrs)
+{
+	if (!attrs[MACSEC_RXSC_ATTR_SCI])
+		return false;
+
+	if (attrs[MACSEC_RXSC_ATTR_ACTIVE]) {
+		if (nla_get_u8(attrs[MACSEC_RXSC_ATTR_ACTIVE]) > 1)
+			return false;
+	}
+
+	return true;
+}
+
+static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info)
+{
+	struct net_device *dev;
+	sci_t sci = MACSEC_UNDEF_SCI;
+	struct nlattr **attrs = info->attrs;
+	struct macsec_rx_sc *rx_sc;
+	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_rxsc_config(attrs, tb_rxsc))
+		return -EINVAL;
+
+	if (!validate_add_rxsc(tb_rxsc))
+		return -EINVAL;
+
+	rtnl_lock();
+	dev = get_dev_from_nl(genl_info_net(info), attrs);
+	if (IS_ERR(dev)) {
+		rtnl_unlock();
+		return PTR_ERR(dev);
+	}
+
+	sci = nla_get_sci(tb_rxsc[MACSEC_RXSC_ATTR_SCI]);
+
+	rx_sc = create_rx_sc(dev, sci);
+	if (IS_ERR(rx_sc)) {
+		rtnl_unlock();
+		return PTR_ERR(rx_sc);
+	}
+
+	if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE])
+		rx_sc->active = !!nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]);
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static bool validate_add_txsa(struct nlattr **attrs)
+{
+	if (!attrs[MACSEC_SA_ATTR_AN] ||
+	    !attrs[MACSEC_SA_ATTR_PN] ||
+	    !attrs[MACSEC_SA_ATTR_KEY] ||
+	    !attrs[MACSEC_SA_ATTR_KEYID])
+		return false;
+
+	if (nla_get_u8(attrs[MACSEC_SA_ATTR_AN]) >= MACSEC_NUM_AN)
+		return false;
+
+	if (nla_get_u32(attrs[MACSEC_SA_ATTR_PN]) == 0)
+		return false;
+
+	if (attrs[MACSEC_SA_ATTR_ACTIVE]) {
+		if (nla_get_u8(attrs[MACSEC_SA_ATTR_ACTIVE]) > 1)
+			return false;
+	}
+
+	return true;
+}
+
+static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info)
+{
+	struct net_device *dev;
+	struct nlattr **attrs = info->attrs;
+	struct macsec_secy *secy;
+	struct macsec_tx_sc *tx_sc;
+	struct macsec_tx_sa *tx_sa;
+	unsigned char assoc_num;
+	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_sa_config(attrs, tb_sa))
+		return -EINVAL;
+
+	if (!validate_add_txsa(tb_sa))
+		return -EINVAL;
+
+	rtnl_lock();
+	dev = get_dev_from_nl(genl_info_net(info), attrs);
+	if (IS_ERR(dev)) {
+		rtnl_unlock();
+		return PTR_ERR(dev);
+	}
+
+	secy = &macsec_priv(dev)->secy;
+	tx_sc = &secy->tx_sc;
+
+	assoc_num = nla_get_u8(tb_sa[MACSEC_SA_ATTR_AN]);
+
+	if (nla_len(tb_sa[MACSEC_SA_ATTR_KEY]) != secy->key_len) {
+		pr_notice("macsec: nl: add_txsa: bad key length: %d != %d\n",
+			  nla_len(tb_sa[MACSEC_SA_ATTR_KEY]), secy->key_len);
+		rtnl_unlock();
+		return -EINVAL;
+	}
+
+	tx_sa = rtnl_dereference(tx_sc->sa[assoc_num]);
+	if (tx_sa) {
+		rtnl_unlock();
+		return -EBUSY;
+	}
+
+	tx_sa = kmalloc(sizeof(*tx_sa), GFP_KERNEL);
+	if (!tx_sa || init_tx_sa(tx_sa, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]),
+				 secy->key_len, secy->icv_len)) {
+		rtnl_unlock();
+		return -ENOMEM;
+	}
+
+	tx_sa->key.id = nla_get_u64(tb_sa[MACSEC_SA_ATTR_KEYID]);
+
+	spin_lock_bh(&tx_sa->lock);
+	tx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
+	spin_unlock_bh(&tx_sa->lock);
+
+	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
+		tx_sa->active = !!nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
+
+	if (assoc_num == tx_sc->encoding_sa && tx_sa->active)
+		secy->operational = true;
+
+	rcu_assign_pointer(tx_sc->sa[assoc_num], tx_sa);
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static int macsec_del_rxsa(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_rx_sa *rx_sa;
+	u8 assoc_num;
+	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_sa_config(attrs, tb_sa))
+		return -EINVAL;
+
+	if (parse_rxsc_config(attrs, tb_rxsc))
+		return -EINVAL;
+
+	rtnl_lock();
+	rx_sa = get_rxsa_from_nl(genl_info_net(info), attrs, tb_rxsc, tb_sa,
+				 &dev, &secy, &rx_sc, &assoc_num);
+	if (IS_ERR(rx_sa)) {
+		rtnl_unlock();
+		return PTR_ERR(rx_sa);
+	}
+
+	if (rx_sa->active) {
+		rtnl_unlock();
+		return -EBUSY;
+	}
+
+	RCU_INIT_POINTER(rx_sc->sa[assoc_num], NULL);
+	clear_rx_sa(rx_sa);
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static int macsec_del_rxsc(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	sci_t sci;
+	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_rxsc_config(attrs, tb_rxsc))
+		return -EINVAL;
+
+	if (!tb_rxsc[MACSEC_RXSC_ATTR_SCI])
+		return -EINVAL;
+
+	rtnl_lock();
+	dev = get_dev_from_nl(genl_info_net(info), info->attrs);
+	if (IS_ERR(dev)) {
+		rtnl_unlock();
+		return PTR_ERR(dev);
+	}
+
+	secy = &macsec_priv(dev)->secy;
+	sci = nla_get_sci(tb_rxsc[MACSEC_RXSC_ATTR_SCI]);
+
+	rx_sc = del_rx_sc(secy, sci);
+	if (!rx_sc) {
+		rtnl_unlock();
+		return -ENODEV;
+	}
+
+	free_rx_sc(rx_sc);
+	rtnl_unlock();
+
+	return 0;
+}
+
+static int macsec_del_txsa(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_tx_sc *tx_sc;
+	struct macsec_tx_sa *tx_sa;
+	u8 assoc_num;
+	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_sa_config(attrs, tb_sa))
+		return -EINVAL;
+
+	rtnl_lock();
+	tx_sa = get_txsa_from_nl(genl_info_net(info), attrs, tb_sa,
+				 &dev, &secy, &tx_sc, &assoc_num);
+	if (IS_ERR(tx_sa)) {
+		rtnl_unlock();
+		return PTR_ERR(tx_sa);
+	}
+
+	if (tx_sa->active) {
+		rtnl_unlock();
+		return -EBUSY;
+	}
+
+	RCU_INIT_POINTER(tx_sc->sa[assoc_num], NULL);
+	clear_tx_sa(tx_sa);
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static bool validate_upd_sa(struct nlattr **attrs)
+{
+	if (!attrs[MACSEC_SA_ATTR_AN] ||
+	    attrs[MACSEC_SA_ATTR_KEY] ||
+	    attrs[MACSEC_SA_ATTR_KEYID])
+		return false;
+
+	if (nla_get_u8(attrs[MACSEC_SA_ATTR_AN]) >= MACSEC_NUM_AN)
+		return false;
+
+	if (attrs[MACSEC_SA_ATTR_PN] && nla_get_u32(attrs[MACSEC_SA_ATTR_PN]) == 0)
+		return false;
+
+	if (attrs[MACSEC_SA_ATTR_ACTIVE]) {
+		if (nla_get_u8(attrs[MACSEC_SA_ATTR_ACTIVE]) > 1)
+			return false;
+	}
+
+	return true;
+}
+
+static int macsec_upd_txsa(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_tx_sc *tx_sc;
+	struct macsec_tx_sa *tx_sa;
+	u8 assoc_num;
+	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_sa_config(attrs, tb_sa))
+		return -EINVAL;
+
+	if (!validate_upd_sa(tb_sa))
+		return -EINVAL;
+
+	rtnl_lock();
+	tx_sa = get_txsa_from_nl(genl_info_net(info), attrs, tb_sa,
+				 &dev, &secy, &tx_sc, &assoc_num);
+	if (IS_ERR(tx_sa)) {
+		rtnl_unlock();
+		return PTR_ERR(tx_sa);
+	}
+
+	if (tb_sa[MACSEC_SA_ATTR_PN]) {
+		spin_lock_bh(&tx_sa->lock);
+		tx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
+		spin_unlock_bh(&tx_sa->lock);
+	}
+
+	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
+		tx_sa->active = nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
+
+	if (assoc_num == tx_sc->encoding_sa)
+		secy->operational = tx_sa->active;
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static int macsec_upd_rxsa(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_rx_sa *rx_sa;
+	u8 assoc_num;
+	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+	struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_rxsc_config(attrs, tb_rxsc))
+		return -EINVAL;
+
+	if (parse_sa_config(attrs, tb_sa))
+		return -EINVAL;
+
+	if (!validate_upd_sa(tb_sa))
+		return -EINVAL;
+
+	rtnl_lock();
+	rx_sa = get_rxsa_from_nl(genl_info_net(info), attrs, tb_rxsc, tb_sa,
+				 &dev, &secy, &rx_sc, &assoc_num);
+	if (IS_ERR(rx_sa)) {
+		rtnl_unlock();
+		return PTR_ERR(rx_sa);
+	}
+
+	if (tb_sa[MACSEC_SA_ATTR_PN]) {
+		spin_lock_bh(&rx_sa->lock);
+		rx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]);
+		spin_unlock_bh(&rx_sa->lock);
+	}
+
+	if (tb_sa[MACSEC_SA_ATTR_ACTIVE])
+		rx_sa->active = nla_get_u8(tb_sa[MACSEC_SA_ATTR_ACTIVE]);
+
+	rtnl_unlock();
+	return 0;
+}
+
+static int macsec_upd_rxsc(struct sk_buff *skb, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+	struct net_device *dev;
+	struct macsec_secy *secy;
+	struct macsec_rx_sc *rx_sc;
+	struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1];
+
+	if (!attrs[MACSEC_ATTR_IFINDEX])
+		return -EINVAL;
+
+	if (parse_rxsc_config(attrs, tb_rxsc))
+		return -EINVAL;
+
+	if (!validate_add_rxsc(tb_rxsc))
+		return -EINVAL;
+
+	rtnl_lock();
+	rx_sc = get_rxsc_from_nl(genl_info_net(info), attrs, tb_rxsc, &dev, &secy);
+	if (IS_ERR(rx_sc)) {
+		rtnl_unlock();
+		return PTR_ERR(rx_sc);
+	}
+
+	if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]) {
+		bool new = !!nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]);
+
+		if (rx_sc->active != new)
+			secy->n_rx_sc += new ? 1 : -1;
+
+		rx_sc->active = new;
+	}
+
+	rtnl_unlock();
+
+	return 0;
+}
+
+static int copy_tx_sa_stats(struct sk_buff *skb,
+			     struct macsec_tx_sa_stats __percpu *pstats)
+{
+	struct macsec_tx_sa_stats sum = {0, };
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		const struct macsec_tx_sa_stats *stats = per_cpu_ptr(pstats, cpu);
+
+		sum.OutPktsProtected += stats->OutPktsProtected;
+		sum.OutPktsEncrypted += stats->OutPktsEncrypted;
+	}
+
+	if (nla_put_u32(skb, MACSEC_SA_STATS_ATTR_OUT_PKTS_PROTECTED, sum.OutPktsProtected) ||
+	    nla_put_u32(skb, MACSEC_SA_STATS_ATTR_OUT_PKTS_ENCRYPTED, sum.OutPktsEncrypted))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int copy_rx_sa_stats(struct sk_buff *skb,
+			     struct macsec_rx_sa_stats __percpu *pstats)
+{
+	struct macsec_rx_sa_stats sum = {0, };
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		const struct macsec_rx_sa_stats *stats = per_cpu_ptr(pstats, cpu);
+
+		sum.InPktsOK         += stats->InPktsOK;
+		sum.InPktsInvalid    += stats->InPktsInvalid;
+		sum.InPktsNotValid   += stats->InPktsNotValid;
+		sum.InPktsNotUsingSA += stats->InPktsNotUsingSA;
+		sum.InPktsUnusedSA   += stats->InPktsUnusedSA;
+	}
+
+	if (nla_put_u32(skb, MACSEC_SA_STATS_ATTR_IN_PKTS_OK, sum.InPktsOK) ||
+	    nla_put_u32(skb, MACSEC_SA_STATS_ATTR_IN_PKTS_INVALID, sum.InPktsInvalid) ||
+	    nla_put_u32(skb, MACSEC_SA_STATS_ATTR_IN_PKTS_NOT_VALID, sum.InPktsNotValid) ||
+	    nla_put_u32(skb, MACSEC_SA_STATS_ATTR_IN_PKTS_NOT_USING_SA, sum.InPktsNotUsingSA) ||
+	    nla_put_u32(skb, MACSEC_SA_STATS_ATTR_IN_PKTS_UNUSED_SA, sum.InPktsUnusedSA))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int copy_rx_sc_stats(struct sk_buff *skb,
+			     struct pcpu_rx_sc_stats __percpu *pstats)
+{
+	struct macsec_rx_sc_stats sum = {0, };
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		const struct pcpu_rx_sc_stats *stats;
+		struct macsec_rx_sc_stats tmp;
+		unsigned int start;
+
+		stats = per_cpu_ptr(pstats, cpu);
+		do {
+			start = u64_stats_fetch_begin_irq(&stats->syncp);
+			memcpy(&tmp, &stats->stats, sizeof(tmp));
+		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+
+		sum.InOctetsValidated += tmp.InOctetsValidated;
+		sum.InOctetsDecrypted += tmp.InOctetsDecrypted;
+		sum.InPktsUnchecked   += tmp.InPktsUnchecked;
+		sum.InPktsDelayed     += tmp.InPktsDelayed;
+		sum.InPktsOK          += tmp.InPktsOK;
+		sum.InPktsInvalid     += tmp.InPktsInvalid;
+		sum.InPktsLate        += tmp.InPktsLate;
+		sum.InPktsNotValid    += tmp.InPktsNotValid;
+		sum.InPktsNotUsingSA  += tmp.InPktsNotUsingSA;
+		sum.InPktsUnusedSA    += tmp.InPktsUnusedSA;
+	}
+
+	if (nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_OCTETS_VALIDATED, sum.InOctetsValidated) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_OCTETS_DECRYPTED, sum.InOctetsDecrypted) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_UNCHECKED, sum.InPktsUnchecked) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_DELAYED, sum.InPktsDelayed) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_OK, sum.InPktsOK) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_INVALID, sum.InPktsInvalid) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_LATE, sum.InPktsLate) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_NOT_VALID, sum.InPktsNotValid) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_NOT_USING_SA, sum.InPktsNotUsingSA) ||
+	    nla_put_u64(skb, MACSEC_RXSC_STATS_ATTR_IN_PKTS_UNUSED_SA, sum.InPktsUnusedSA))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int copy_tx_sc_stats(struct sk_buff *skb,
+			     struct pcpu_tx_sc_stats __percpu *pstats)
+{
+	struct macsec_tx_sc_stats sum = {0, };
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		const struct pcpu_tx_sc_stats *stats;
+		struct macsec_tx_sc_stats tmp;
+		unsigned int start;
+
+		stats = per_cpu_ptr(pstats, cpu);
+		do {
+			start = u64_stats_fetch_begin_irq(&stats->syncp);
+			memcpy(&tmp, &stats->stats, sizeof(tmp));
+		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+
+		sum.OutPktsProtected   += tmp.OutPktsProtected;
+		sum.OutPktsEncrypted   += tmp.OutPktsEncrypted;
+		sum.OutOctetsProtected += tmp.OutOctetsProtected;
+		sum.OutOctetsEncrypted += tmp.OutOctetsEncrypted;
+	}
+
+	if (nla_put_u64(skb, MACSEC_TXSC_STATS_ATTR_OUT_PKTS_PROTECTED, sum.OutPktsProtected) ||
+	    nla_put_u64(skb, MACSEC_TXSC_STATS_ATTR_OUT_PKTS_ENCRYPTED, sum.OutPktsEncrypted) ||
+	    nla_put_u64(skb, MACSEC_TXSC_STATS_ATTR_OUT_OCTETS_PROTECTED, sum.OutOctetsProtected) ||
+	    nla_put_u64(skb, MACSEC_TXSC_STATS_ATTR_OUT_OCTETS_ENCRYPTED, sum.OutOctetsEncrypted))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int copy_secy_stats(struct sk_buff *skb,
+			    struct pcpu_secy_stats __percpu *pstats)
+{
+	struct macsec_dev_stats sum = {0, };
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		const struct pcpu_secy_stats *stats;
+		struct macsec_dev_stats tmp;
+		unsigned int start;
+
+		stats = per_cpu_ptr(pstats, cpu);
+		do {
+			start = u64_stats_fetch_begin_irq(&stats->syncp);
+			memcpy(&tmp, &stats->stats, sizeof(tmp));
+		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+
+		sum.OutPktsUntagged  += tmp.OutPktsUntagged;
+		sum.InPktsUntagged   += tmp.InPktsUntagged;
+		sum.OutPktsTooLong   += tmp.OutPktsTooLong;
+		sum.InPktsNoTag      += tmp.InPktsNoTag;
+		sum.InPktsBadTag     += tmp.InPktsBadTag;
+		sum.InPktsUnknownSCI += tmp.InPktsUnknownSCI;
+		sum.InPktsNoSCI      += tmp.InPktsNoSCI;
+		sum.InPktsOverrun    += tmp.InPktsOverrun;
+	}
+
+	if (nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_OUT_PKTS_UNTAGGED, sum.OutPktsUntagged) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_IN_PKTS_UNTAGGED, sum.InPktsUntagged) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_OUT_PKTS_TOO_LONG, sum.OutPktsTooLong) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_IN_PKTS_NO_TAG, sum.InPktsNoTag) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_IN_PKTS_BAD_TAG, sum.InPktsBadTag) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_IN_PKTS_UNKNOWN_SCI, sum.InPktsUnknownSCI) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_IN_PKTS_NO_SCI, sum.InPktsNoSCI) ||
+	    nla_put_u64(skb, MACSEC_SECY_STATS_ATTR_IN_PKTS_OVERRUN, sum.InPktsOverrun))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+static int nla_put_secy(struct macsec_secy *secy, struct sk_buff *skb)
+{
+	struct macsec_tx_sc *tx_sc = &secy->tx_sc;
+	struct nlattr *secy_nest = nla_nest_start(skb, MACSEC_ATTR_SECY);
+
+	if (!secy_nest)
+		return 1;
+
+	if (nla_put_sci(skb, MACSEC_SECY_ATTR_SCI, secy->sci) ||
+	    nla_put_u64(skb, MACSEC_SECY_ATTR_CIPHER_SUITE, DEFAULT_CIPHER_ID) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_ICV_LEN, secy->icv_len) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_OPER, secy->operational) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_PROTECT, secy->protect_frames) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_REPLAY, secy->replay_protect) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_VALIDATE, secy->validate_frames) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_ENCRYPT, tx_sc->encrypt) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_INC_SCI, tx_sc->send_sci) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_ES, tx_sc->end_station) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_SCB, tx_sc->scb) ||
+	    nla_put_u8(skb, MACSEC_SECY_ATTR_ENCODING_SA, tx_sc->encoding_sa))
+		goto cancel;
+
+	if (secy->replay_protect) {
+		if (nla_put_u32(skb, MACSEC_SECY_ATTR_WINDOW, secy->replay_window))
+			goto cancel;
+	}
+
+	nla_nest_end(skb, secy_nest);
+	return 0;
+
+cancel:
+	nla_nest_cancel(skb, secy_nest);
+	return 1;
+}
+
+static int dump_secy(struct macsec_secy *secy, struct net_device *dev,
+		     struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct macsec_rx_sc *rx_sc;
+	struct macsec_tx_sc *tx_sc = &secy->tx_sc;
+	struct nlattr *txsa_list, *rxsc_list;
+	int i, j;
+	void *hdr;
+	struct nlattr *attr;
+
+	hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
+			  &macsec_fam, NLM_F_MULTI, MACSEC_CMD_GET_TXSC);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	rtnl_lock();
+
+	if (nla_put_u32(skb, MACSEC_ATTR_IFINDEX, dev->ifindex))
+		goto nla_put_failure;
+
+	if (nla_put_secy(secy, skb))
+		goto nla_put_failure;
+
+	attr = nla_nest_start(skb, MACSEC_ATTR_TXSC_STATS);
+	if (!attr)
+		goto nla_put_failure;
+	if (copy_tx_sc_stats(skb, tx_sc->stats)) {
+		nla_nest_cancel(skb, attr);
+		goto nla_put_failure;
+	}
+	nla_nest_end(skb, attr);
+
+	attr = nla_nest_start(skb, MACSEC_ATTR_SECY_STATS);
+	if (!attr)
+		goto nla_put_failure;
+	if (copy_secy_stats(skb, macsec_priv(dev)->stats)) {
+		nla_nest_cancel(skb, attr);
+		goto nla_put_failure;
+	}
+	nla_nest_end(skb, attr);
+
+	txsa_list = nla_nest_start(skb, MACSEC_ATTR_TXSA_LIST);
+	if (!txsa_list)
+		goto nla_put_failure;
+	for (i = 0, j = 1; i < MACSEC_NUM_AN; i++) {
+		struct macsec_tx_sa *tx_sa = rtnl_dereference(tx_sc->sa[i]);
+		struct nlattr *txsa_nest;
+
+		if (!tx_sa)
+			continue;
+
+		txsa_nest = nla_nest_start(skb, j++);
+		if (!txsa_nest) {
+			nla_nest_cancel(skb, txsa_list);
+			goto nla_put_failure;
+		}
+
+		if (nla_put_u8(skb, MACSEC_SA_ATTR_AN, i) ||
+		    nla_put_u32(skb, MACSEC_SA_ATTR_PN, tx_sa->next_pn) ||
+		    nla_put_u64(skb, MACSEC_SA_ATTR_KEYID, tx_sa->key.id) ||
+		    nla_put_u8(skb, MACSEC_SA_ATTR_ACTIVE, tx_sa->active)) {
+			nla_nest_cancel(skb, txsa_nest);
+			nla_nest_cancel(skb, txsa_list);
+			goto nla_put_failure;
+		}
+
+		attr = nla_nest_start(skb, MACSEC_SA_ATTR_STATS);
+		if (!attr) {
+			nla_nest_cancel(skb, txsa_nest);
+			nla_nest_cancel(skb, txsa_list);
+			goto nla_put_failure;
+		}
+		if (copy_tx_sa_stats(skb, tx_sa->stats)) {
+			nla_nest_cancel(skb, attr);
+			nla_nest_cancel(skb, txsa_nest);
+			nla_nest_cancel(skb, txsa_list);
+			goto nla_put_failure;
+		}
+		nla_nest_end(skb, attr);
+
+		nla_nest_end(skb, txsa_nest);
+	}
+	nla_nest_end(skb, txsa_list);
+
+	rxsc_list = nla_nest_start(skb, MACSEC_ATTR_RXSC_LIST);
+	if (!rxsc_list)
+		goto nla_put_failure;
+
+	j = 1;
+	for_each_rxsc_rtnl(secy, rx_sc) {
+		int k;
+		struct nlattr *rxsa_list;
+		struct nlattr *rxsc_nest = nla_nest_start(skb, j++);
+
+		if (!rxsc_nest) {
+			nla_nest_cancel(skb, rxsc_list);
+			goto nla_put_failure;
+		}
+
+		if (nla_put_u8(skb, MACSEC_RXSC_ATTR_ACTIVE, rx_sc->active) ||
+		    nla_put_sci(skb, MACSEC_RXSC_ATTR_SCI, rx_sc->sci)) {
+			nla_nest_cancel(skb, rxsc_nest);
+			nla_nest_cancel(skb, rxsc_list);
+			goto nla_put_failure;
+		}
+
+		attr = nla_nest_start(skb, MACSEC_RXSC_ATTR_STATS);
+		if (!attr) {
+			nla_nest_cancel(skb, rxsc_nest);
+			nla_nest_cancel(skb, rxsc_list);
+			goto nla_put_failure;
+		}
+		if (copy_rx_sc_stats(skb, rx_sc->stats)) {
+			nla_nest_cancel(skb, attr);
+			nla_nest_cancel(skb, rxsc_nest);
+			nla_nest_cancel(skb, rxsc_list);
+			goto nla_put_failure;
+		}
+		nla_nest_end(skb, attr);
+
+		rxsa_list = nla_nest_start(skb, MACSEC_RXSC_ATTR_SA_LIST);
+		if (!rxsa_list) {
+			nla_nest_cancel(skb, rxsc_nest);
+			nla_nest_cancel(skb, rxsc_list);
+			goto nla_put_failure;
+		}
+
+		for (i = 0, k = 1; i < MACSEC_NUM_AN; i++) {
+			struct macsec_rx_sa *rx_sa = rtnl_dereference(rx_sc->sa[i]);
+			struct nlattr *rxsa_nest;
+
+			if (!rx_sa)
+				continue;
+
+			rxsa_nest = nla_nest_start(skb, k++);
+			if (!rxsa_nest) {
+				nla_nest_cancel(skb, rxsa_list);
+				nla_nest_cancel(skb, rxsc_nest);
+				nla_nest_cancel(skb, rxsc_list);
+				goto nla_put_failure;
+			}
+
+			attr = nla_nest_start(skb, MACSEC_SA_ATTR_STATS);
+			if (!attr) {
+				nla_nest_cancel(skb, rxsa_list);
+				nla_nest_cancel(skb, rxsc_nest);
+				nla_nest_cancel(skb, rxsc_list);
+				goto nla_put_failure;
+			}
+			if (copy_rx_sa_stats(skb, rx_sa->stats)) {
+				nla_nest_cancel(skb, attr);
+				nla_nest_cancel(skb, rxsa_list);
+				nla_nest_cancel(skb, rxsc_nest);
+				nla_nest_cancel(skb, rxsc_list);
+				goto nla_put_failure;
+			}
+			nla_nest_end(skb, attr);
+
+			if (nla_put_u8(skb, MACSEC_SA_ATTR_AN, i) ||
+			    nla_put_u32(skb, MACSEC_SA_ATTR_PN, rx_sa->next_pn) ||
+			    nla_put_u64(skb, MACSEC_SA_ATTR_KEYID, rx_sa->key.id) ||
+			    nla_put_u8(skb, MACSEC_SA_ATTR_ACTIVE, rx_sa->active)) {
+				nla_nest_cancel(skb, rxsa_nest);
+				nla_nest_cancel(skb, rxsc_nest);
+				nla_nest_cancel(skb, rxsc_list);
+				goto nla_put_failure;
+			}
+			nla_nest_end(skb, rxsa_nest);
+		}
+
+		nla_nest_end(skb, rxsa_list);
+		nla_nest_end(skb, rxsc_nest);
+	}
+
+	nla_nest_end(skb, rxsc_list);
+
+	rtnl_unlock();
+
+	genlmsg_end(skb, hdr);
+
+	return 0;
+
+nla_put_failure:
+	rtnl_unlock();
+	genlmsg_cancel(skb, hdr);
+	return -EMSGSIZE;
+}
+
+static int macsec_dump_txsc(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct net *net = sock_net(skb->sk);
+	struct net_device *dev;
+	int dev_idx, d;
+
+	dev_idx = cb->args[0];
+
+	d = 0;
+	for_each_netdev(net, dev) {
+		struct macsec_secy *secy;
+
+		if (d < dev_idx)
+			goto next;
+
+		if (!netif_is_macsec(dev))
+			goto next;
+
+		secy = &macsec_priv(dev)->secy;
+		if (dump_secy(secy, dev, skb, cb) < 0)
+			goto done;
+next:
+		d++;
+	}
+
+done:
+	cb->args[0] = d;
+	return skb->len;
+}
+
+static const struct genl_ops macsec_genl_ops[] = {
+	{
+		.cmd = MACSEC_CMD_GET_TXSC,
+		.dumpit = macsec_dump_txsc,
+		.policy = macsec_genl_policy,
+	},
+	{
+		.cmd = MACSEC_CMD_ADD_RXSC,
+		.doit = macsec_add_rxsc,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_DEL_RXSC,
+		.doit = macsec_del_rxsc,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_UPD_RXSC,
+		.doit = macsec_upd_rxsc,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_ADD_TXSA,
+		.doit = macsec_add_txsa,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_DEL_TXSA,
+		.doit = macsec_del_txsa,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_UPD_TXSA,
+		.doit = macsec_upd_txsa,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_ADD_RXSA,
+		.doit = macsec_add_rxsa,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_DEL_RXSA,
+		.doit = macsec_del_rxsa,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = MACSEC_CMD_UPD_RXSA,
+		.doit = macsec_upd_rxsa,
+		.policy = macsec_genl_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
+};
+
+static netdev_tx_t macsec_start_xmit(struct sk_buff *skb,
+				     struct net_device *dev)
+{
+	struct macsec_dev *macsec = netdev_priv(dev);
+	struct macsec_secy *secy = &macsec->secy;
+	struct pcpu_secy_stats *secy_stats;
+	int ret, len;
+
+	/* 10.5 */
+	if (!secy->protect_frames) {
+		secy_stats = this_cpu_ptr(macsec->stats);
+		u64_stats_update_begin(&secy_stats->syncp);
+		secy_stats->stats.OutPktsUntagged++;
+		u64_stats_update_end(&secy_stats->syncp);
+		len = skb->len;
+		ret = dev_queue_xmit(skb);
+		count_tx(dev, ret, len);
+		return ret;
+	}
+
+	if (!secy->operational) {
+		kfree_skb(skb);
+		dev->stats.tx_dropped++;
+		return NETDEV_TX_OK;
+	}
+
+	skb = macsec_encrypt(skb, dev);
+	if (IS_ERR(skb)) {
+		if (PTR_ERR(skb) != -EINPROGRESS)
+			dev->stats.tx_dropped++;
+		return NETDEV_TX_OK;
+	}
+
+	macsec_count_tx(skb, &macsec->secy.tx_sc, macsec_skb_cb(skb)->tx_sa);
+
+	macsec_encrypt_finish(skb, dev);
+	len = skb->len;
+	ret = dev_queue_xmit(skb);
+	count_tx(dev, ret, len);
+	return ret;
+}
+
+#define MACSEC_FEATURES \
+	(NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST)
+static int macsec_dev_init(struct net_device *dev)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+
+	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
+	if (!dev->tstats)
+		return -ENOMEM;
+
+	dev->features = real_dev->features & MACSEC_FEATURES;
+	dev->features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE;
+
+	dev->needed_headroom = real_dev->needed_headroom +
+			       MACSEC_NEEDED_HEADROOM;
+	dev->needed_tailroom = real_dev->needed_tailroom +
+			       MACSEC_NEEDED_TAILROOM;
+
+	if (is_zero_ether_addr(dev->dev_addr))
+		eth_hw_addr_inherit(dev, real_dev);
+	if (is_zero_ether_addr(dev->broadcast))
+		memcpy(dev->broadcast, real_dev->broadcast, dev->addr_len);
+
+	return 0;
+}
+
+static void macsec_dev_uninit(struct net_device *dev)
+{
+	free_percpu(dev->tstats);
+}
+
+static netdev_features_t macsec_fix_features(struct net_device *dev,
+					     netdev_features_t features)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+
+	features &= real_dev->features & MACSEC_FEATURES;
+	features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE;
+
+	return features;
+}
+
+static int macsec_dev_open(struct net_device *dev)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+	int err;
+
+	if (!(real_dev->flags & IFF_UP))
+		return -ENETDOWN;
+
+	err = dev_uc_add(real_dev, dev->dev_addr);
+	if (err < 0)
+		return err;
+
+	if (dev->flags & IFF_ALLMULTI) {
+		err = dev_set_allmulti(real_dev, 1);
+		if (err < 0)
+			goto del_unicast;
+	}
+
+	if (dev->flags & IFF_PROMISC) {
+		err = dev_set_promiscuity(real_dev, 1);
+		if (err < 0)
+			goto clear_allmulti;
+	}
+
+	if (netif_carrier_ok(real_dev))
+		netif_carrier_on(dev);
+
+	return 0;
+clear_allmulti:
+	if (dev->flags & IFF_ALLMULTI)
+		dev_set_allmulti(real_dev, -1);
+del_unicast:
+	dev_uc_del(real_dev, dev->dev_addr);
+	netif_carrier_off(dev);
+	return err;
+}
+
+static int macsec_dev_stop(struct net_device *dev)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+
+	netif_carrier_off(dev);
+
+	dev_mc_unsync(real_dev, dev);
+	dev_uc_unsync(real_dev, dev);
+
+	if (dev->flags & IFF_ALLMULTI)
+		dev_set_allmulti(real_dev, -1);
+
+	if (dev->flags & IFF_PROMISC)
+		dev_set_promiscuity(real_dev, -1);
+
+	dev_uc_del(real_dev, dev->dev_addr);
+
+	return 0;
+}
+
+static void macsec_dev_change_rx_flags(struct net_device *dev, int change)
+{
+	struct net_device *real_dev = macsec_priv(dev)->real_dev;
+
+	if (!(dev->flags & IFF_UP))
+		return;
+
+	if (change & IFF_ALLMULTI)
+		dev_set_allmulti(real_dev, dev->flags & IFF_ALLMULTI ? 1 : -1);
+
+	if (change & IFF_PROMISC)
+		dev_set_promiscuity(real_dev,
+				    dev->flags & IFF_PROMISC ? 1 : -1);
+}
+
+static void macsec_dev_set_rx_mode(struct net_device *dev)
+{
+	struct net_device *real_dev = macsec_priv(dev)->real_dev;
+
+	dev_mc_sync(real_dev, dev);
+	dev_uc_sync(real_dev, dev);
+}
+
+static int macsec_set_mac_address(struct net_device *dev, void *p)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+	struct sockaddr *addr = p;
+	int err;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	if (!(dev->flags & IFF_UP))
+		goto out;
+
+	err = dev_uc_add(real_dev, addr->sa_data);
+	if (err < 0)
+		return err;
+
+	dev_uc_del(real_dev, dev->dev_addr);
+
+out:
+	ether_addr_copy(dev->dev_addr, addr->sa_data);
+	return 0;
+}
+
+static int macsec_change_mtu(struct net_device *dev, int new_mtu)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	unsigned int extra = macsec->secy.icv_len + macsec_extra_len(true);
+
+	if (macsec->real_dev->mtu - extra < new_mtu)
+		return -ERANGE;
+
+	dev->mtu = new_mtu;
+
+	return 0;
+}
+
+static struct rtnl_link_stats64 *macsec_get_stats64(struct net_device *dev,
+						    struct rtnl_link_stats64 *s)
+{
+	int cpu;
+
+	if (!dev->tstats)
+		return s;
+
+	for_each_possible_cpu(cpu) {
+		struct pcpu_sw_netstats *stats;
+		struct pcpu_sw_netstats tmp;
+		int start;
+
+		stats = per_cpu_ptr(dev->tstats, cpu);
+		do {
+			start = u64_stats_fetch_begin_irq(&stats->syncp);
+			tmp.rx_packets = stats->rx_packets;
+			tmp.rx_bytes   = stats->rx_bytes;
+			tmp.tx_packets = stats->tx_packets;
+			tmp.tx_bytes   = stats->tx_bytes;
+		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+
+		s->rx_packets += tmp.rx_packets;
+		s->rx_bytes   += tmp.rx_bytes;
+		s->tx_packets += tmp.tx_packets;
+		s->tx_bytes   += tmp.tx_bytes;
+	}
+
+	s->rx_dropped = dev->stats.rx_dropped;
+	s->tx_dropped = dev->stats.tx_dropped;
+
+	return s;
+}
+
+static int macsec_get_iflink(const struct net_device *dev)
+{
+	return macsec_priv(dev)->real_dev->ifindex;
+}
+
+static const struct net_device_ops macsec_netdev_ops = {
+	.ndo_init		= macsec_dev_init,
+	.ndo_uninit		= macsec_dev_uninit,
+	.ndo_open		= macsec_dev_open,
+	.ndo_stop		= macsec_dev_stop,
+	.ndo_fix_features	= macsec_fix_features,
+	.ndo_change_mtu		= macsec_change_mtu,
+	.ndo_set_rx_mode	= macsec_dev_set_rx_mode,
+	.ndo_change_rx_flags	= macsec_dev_change_rx_flags,
+	.ndo_set_mac_address	= macsec_set_mac_address,
+	.ndo_start_xmit		= macsec_start_xmit,
+	.ndo_get_stats64	= macsec_get_stats64,
+	.ndo_get_iflink		= macsec_get_iflink,
+};
+
+static const struct device_type macsec_type = {
+	.name = "macsec",
+};
+
+static const struct nla_policy macsec_rtnl_policy[IFLA_MACSEC_MAX + 1] = {
+	[IFLA_MACSEC_SCI] = { .type = NLA_U64 },
+	[IFLA_MACSEC_ICV_LEN] = { .type = NLA_U8 },
+	[IFLA_MACSEC_CIPHER_SUITE] = { .type = NLA_U64 },
+	[IFLA_MACSEC_WINDOW] = { .type = NLA_U32 },
+	[IFLA_MACSEC_ENCODING_SA] = { .type = NLA_U8 },
+	[IFLA_MACSEC_ENCRYPT] = { .type = NLA_U8 },
+	[IFLA_MACSEC_PROTECT] = { .type = NLA_U8 },
+	[IFLA_MACSEC_INC_SCI] = { .type = NLA_U8 },
+	[IFLA_MACSEC_ES] = { .type = NLA_U8 },
+	[IFLA_MACSEC_SCB] = { .type = NLA_U8 },
+	[IFLA_MACSEC_REPLAY_PROTECT] = { .type = NLA_U8 },
+	[IFLA_MACSEC_VALIDATION] = { .type = NLA_U8 },
+};
+
+static void macsec_free_netdev(struct net_device *dev)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+
+	free_percpu(macsec->stats);
+	free_percpu(macsec->secy.tx_sc.stats);
+
+	dev_put(real_dev);
+	free_netdev(dev);
+}
+
+static void macsec_setup(struct net_device *dev)
+{
+	ether_setup(dev);
+	dev->tx_queue_len = 0;
+	dev->netdev_ops = &macsec_netdev_ops;
+	dev->destructor = macsec_free_netdev;
+
+	eth_zero_addr(dev->broadcast);
+}
+
+static void macsec_changelink_common(struct net_device *dev,
+				     struct nlattr *data[])
+{
+	struct macsec_secy *secy;
+	struct macsec_tx_sc *tx_sc;
+
+	secy = &macsec_priv(dev)->secy;
+	tx_sc = &secy->tx_sc;
+
+	if (data[IFLA_MACSEC_ENCODING_SA]) {
+		struct macsec_tx_sa *tx_sa;
+
+		tx_sc->encoding_sa = nla_get_u8(data[IFLA_MACSEC_ENCODING_SA]);
+		tx_sa = rtnl_dereference(tx_sc->sa[tx_sc->encoding_sa]);
+
+		secy->operational = tx_sa && tx_sa->active;
+	}
+
+	if (data[IFLA_MACSEC_WINDOW])
+		secy->replay_window = nla_get_u32(data[IFLA_MACSEC_WINDOW]);
+
+	if (data[IFLA_MACSEC_ENCRYPT])
+		tx_sc->encrypt = !!nla_get_u8(data[IFLA_MACSEC_ENCRYPT]);
+
+	if (data[IFLA_MACSEC_PROTECT])
+		secy->protect_frames = !!nla_get_u8(data[IFLA_MACSEC_PROTECT]);
+
+	if (data[IFLA_MACSEC_INC_SCI])
+		tx_sc->send_sci = !!nla_get_u8(data[IFLA_MACSEC_INC_SCI]);
+
+	if (data[IFLA_MACSEC_ES])
+		tx_sc->end_station = !!nla_get_u8(data[IFLA_MACSEC_ES]);
+
+	if (data[IFLA_MACSEC_SCB])
+		tx_sc->scb = !!nla_get_u8(data[IFLA_MACSEC_SCB]);
+
+	if (data[IFLA_MACSEC_REPLAY_PROTECT])
+		secy->replay_protect = !!nla_get_u8(data[IFLA_MACSEC_REPLAY_PROTECT]);
+
+	if (data[IFLA_MACSEC_VALIDATION])
+		secy->validate_frames = nla_get_u8(data[IFLA_MACSEC_VALIDATION]);
+}
+
+static int macsec_changelink(struct net_device *dev, struct nlattr *tb[],
+			     struct nlattr *data[])
+{
+	if (!data)
+		return 0;
+
+	if (data[IFLA_MACSEC_CIPHER_SUITE] ||
+	    data[IFLA_MACSEC_ICV_LEN] ||
+	    data[IFLA_MACSEC_SCI] ||
+	    data[IFLA_MACSEC_PORT])
+		return -EINVAL;
+
+	macsec_changelink_common(dev, data);
+
+	return 0;
+}
+
+static void macsec_del_dev(struct macsec_dev *macsec)
+{
+	int i;
+
+	while (macsec->secy.rx_sc) {
+		struct macsec_rx_sc *rx_sc = rtnl_dereference(macsec->secy.rx_sc);
+
+		rcu_assign_pointer(macsec->secy.rx_sc, rx_sc->next);
+		free_rx_sc(rx_sc);
+	}
+
+	for (i = 0; i < MACSEC_NUM_AN; i++) {
+		struct macsec_tx_sa *sa = rtnl_dereference(macsec->secy.tx_sc.sa[i]);
+
+		if (sa) {
+			RCU_INIT_POINTER(macsec->secy.tx_sc.sa[i], NULL);
+			clear_tx_sa(sa);
+		}
+	}
+}
+
+static void macsec_dellink(struct net_device *dev, struct list_head *head)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev = macsec->real_dev;
+	struct macsec_rxh_data *rxd = macsec_data_rtnl(real_dev);
+
+	unregister_netdevice_queue(dev, head);
+	list_del_rcu(&macsec->secys);
+	if (list_empty(&rxd->secys))
+		netdev_rx_handler_unregister(real_dev);
+
+	macsec_del_dev(macsec);
+}
+
+static int register_macsec_dev(struct net_device *real_dev,
+			       struct net_device *dev)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct macsec_rxh_data *rxd = macsec_data_rtnl(real_dev);
+
+	if (!rxd) {
+		int err;
+
+		rxd = kmalloc(sizeof(*rxd), GFP_KERNEL);
+		if (!rxd)
+			return -ENOMEM;
+
+		INIT_LIST_HEAD(&rxd->secys);
+
+		err = netdev_rx_handler_register(real_dev, macsec_handle_frame,
+						 rxd);
+		if (err < 0)
+			return err;
+	}
+
+	list_add_tail_rcu(&macsec->secys, &rxd->secys);
+	return 0;
+}
+
+static bool sci_exists(struct net_device *dev, sci_t sci)
+{
+	struct macsec_rxh_data *rxd = macsec_data_rtnl(dev);
+	struct macsec_dev *macsec;
+
+	list_for_each_entry(macsec, &rxd->secys, secys) {
+		if (macsec->secy.sci == sci)
+			return true;
+	}
+
+	return false;
+}
+
+static sci_t dev_to_sci(struct net_device *dev, __be16 port)
+{
+	return make_sci(dev->dev_addr, port);
+}
+
+static int macsec_add_dev(struct net_device *dev, sci_t sci, u8 icv_len)
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct macsec_secy *secy = &macsec->secy;
+
+	macsec->stats = netdev_alloc_pcpu_stats(struct pcpu_secy_stats);
+	if (!macsec->stats)
+		return -ENOMEM;
+
+	secy->tx_sc.stats = netdev_alloc_pcpu_stats(struct pcpu_tx_sc_stats);
+	if (!secy->tx_sc.stats) {
+		free_percpu(macsec->stats);
+		return -ENOMEM;
+	}
+
+	if (sci == MACSEC_UNDEF_SCI)
+		sci = dev_to_sci(dev, MACSEC_PORT_ES);
+
+	secy->netdev = dev;
+	secy->operational = true;
+	secy->key_len = DEFAULT_SAK_LEN;
+	secy->icv_len = icv_len;
+	secy->validate_frames = MACSEC_VALIDATE_DEFAULT;
+	secy->protect_frames = true;
+	secy->replay_protect = false;
+
+	secy->sci = sci;
+	secy->tx_sc.active = true;
+	secy->tx_sc.encoding_sa = DEFAULT_ENCODING_SA;
+	secy->tx_sc.encrypt = DEFAULT_ENCRYPT;
+	secy->tx_sc.send_sci = DEFAULT_SEND_SCI;
+	secy->tx_sc.end_station = false;
+	secy->tx_sc.scb = false;
+
+	return 0;
+}
+
+static int macsec_newlink(struct net *net, struct net_device *dev,
+			  struct nlattr *tb[], struct nlattr *data[])
+{
+	struct macsec_dev *macsec = macsec_priv(dev);
+	struct net_device *real_dev;
+	int err;
+	sci_t sci;
+	u8 icv_len = DEFAULT_ICV_LEN;
+	rx_handler_func_t *rx_handler;
+
+	if (!tb[IFLA_LINK])
+		return -EINVAL;
+	real_dev = __dev_get_by_index(net, nla_get_u32(tb[IFLA_LINK]));
+	if (!real_dev)
+		return -ENODEV;
+
+	dev->priv_flags |= IFF_MACSEC;
+
+	macsec->real_dev = real_dev;
+
+	if (data && data[IFLA_MACSEC_ICV_LEN])
+		icv_len = nla_get_u8(data[IFLA_MACSEC_ICV_LEN]);
+	dev->mtu = real_dev->mtu - icv_len - macsec_extra_len(true);
+
+	rx_handler = rtnl_dereference(real_dev->rx_handler);
+	if (rx_handler && rx_handler != macsec_handle_frame)
+		return -EBUSY;
+
+	err = register_netdevice(dev);
+	if (err < 0)
+		return err;
+
+	/* need to be already registered so that ->init has run and
+	 * the MAC addr is set
+	 */
+	if (data && data[IFLA_MACSEC_SCI])
+		sci = nla_get_sci(data[IFLA_MACSEC_SCI]);
+	else if (data && data[IFLA_MACSEC_PORT])
+		sci = dev_to_sci(dev, nla_get_be16(data[IFLA_MACSEC_PORT]));
+	else
+		sci = dev_to_sci(dev, MACSEC_PORT_ES);
+
+	if (rx_handler && sci_exists(real_dev, sci)) {
+		err = -EBUSY;
+		goto unregister;
+	}
+
+	err = macsec_add_dev(dev, sci, icv_len);
+	if (err)
+		goto unregister;
+
+	if (data)
+		macsec_changelink_common(dev, data);
+
+	err = register_macsec_dev(real_dev, dev);
+	if (err < 0)
+		goto del_dev;
+
+	dev_hold(real_dev);
+
+	return 0;
+
+del_dev:
+	macsec_del_dev(macsec);
+unregister:
+	unregister_netdevice(dev);
+	return err;
+}
+
+static int macsec_validate_attr(struct nlattr *tb[], struct nlattr *data[])
+{
+	u64 csid = DEFAULT_CIPHER_ID;
+	u8 icv_len = DEFAULT_ICV_LEN;
+	int flag;
+	bool es, scb, sci;
+
+	if (!data)
+		return 0;
+
+	if (data[IFLA_MACSEC_CIPHER_SUITE])
+		csid = nla_get_u64(data[IFLA_MACSEC_CIPHER_SUITE]);
+
+	if (data[IFLA_MACSEC_ICV_LEN])
+		icv_len = nla_get_u8(data[IFLA_MACSEC_ICV_LEN]);
+
+	switch (csid) {
+	case DEFAULT_CIPHER_ID:
+	case DEFAULT_CIPHER_ALT:
+		if (icv_len < MACSEC_MIN_ICV_LEN ||
+		    icv_len > MACSEC_MAX_ICV_LEN)
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (data[IFLA_MACSEC_ENCODING_SA]) {
+		if (nla_get_u8(data[IFLA_MACSEC_ENCODING_SA]) >= MACSEC_NUM_AN)
+			return -EINVAL;
+	}
+
+	for (flag = IFLA_MACSEC_ENCODING_SA + 1;
+	     flag < IFLA_MACSEC_VALIDATION;
+	     flag++) {
+		if (data[flag]) {
+			if (nla_get_u8(data[flag]) > 1)
+				return -EINVAL;
+		}
+	}
+
+	es  = data[IFLA_MACSEC_ES] ? nla_get_u8(data[IFLA_MACSEC_ES]) : false;
+	sci = data[IFLA_MACSEC_INC_SCI] ? nla_get_u8(data[IFLA_MACSEC_INC_SCI]) : false;
+	scb = data[IFLA_MACSEC_SCB] ? nla_get_u8(data[IFLA_MACSEC_SCB]) : false;
+
+	if ((sci && (scb || es)) || (scb && es))
+		return -EINVAL;
+
+	if (data[IFLA_MACSEC_VALIDATION] &&
+	    nla_get_u8(data[IFLA_MACSEC_VALIDATION]) > MACSEC_VALIDATE_MAX)
+		return -EINVAL;
+
+	if ((data[IFLA_MACSEC_PROTECT] &&
+	     nla_get_u8(data[IFLA_MACSEC_PROTECT])) &&
+	    !data[IFLA_MACSEC_WINDOW])
+		return -EINVAL;
+
+	return 0;
+}
+
+static struct net *macsec_get_link_net(const struct net_device *dev)
+{
+	return dev_net(macsec_priv(dev)->real_dev);
+}
+
+static size_t macsec_get_size(const struct net_device *dev)
+{
+	return 0 +
+		nla_total_size(8) + /* SCI */
+		nla_total_size(1) + /* ICV_LEN */
+		nla_total_size(8) + /* CIPHER_SUITE */
+		nla_total_size(4) + /* WINDOW */
+		nla_total_size(1) + /* ENCODING_SA */
+		nla_total_size(1) + /* ENCRYPT */
+		nla_total_size(1) + /* PROTECT */
+		nla_total_size(1) + /* INC_SCI */
+		nla_total_size(1) + /* ES */
+		nla_total_size(1) + /* SCB */
+		nla_total_size(1) + /* REPLAY_PROTECT */
+		nla_total_size(1) + /* VALIDATION */
+		0;
+}
+
+static int macsec_fill_info(struct sk_buff *skb,
+			    const struct net_device *dev)
+{
+	struct macsec_secy *secy = &macsec_priv(dev)->secy;
+	struct macsec_tx_sc *tx_sc = &secy->tx_sc;
+
+	if (nla_put_sci(skb, IFLA_MACSEC_SCI, secy->sci) ||
+	    nla_put_u8(skb, IFLA_MACSEC_ICV_LEN, secy->icv_len) ||
+	    nla_put_u64(skb, IFLA_MACSEC_CIPHER_SUITE, DEFAULT_CIPHER_ID) ||
+	    nla_put_u8(skb, IFLA_MACSEC_ENCODING_SA, tx_sc->encoding_sa) ||
+	    nla_put_u8(skb, IFLA_MACSEC_ENCRYPT, tx_sc->encrypt) ||
+	    nla_put_u8(skb, IFLA_MACSEC_PROTECT, secy->protect_frames) ||
+	    nla_put_u8(skb, IFLA_MACSEC_INC_SCI, tx_sc->send_sci) ||
+	    nla_put_u8(skb, IFLA_MACSEC_ES, tx_sc->end_station) ||
+	    nla_put_u8(skb, IFLA_MACSEC_SCB, tx_sc->scb) ||
+	    nla_put_u8(skb, IFLA_MACSEC_REPLAY_PROTECT, secy->replay_protect) ||
+	    nla_put_u8(skb, IFLA_MACSEC_VALIDATION, secy->validate_frames) ||
+	    0)
+		goto nla_put_failure;
+
+	if (secy->replay_protect) {
+		if (nla_put_u32(skb, IFLA_MACSEC_WINDOW, secy->replay_window))
+			goto nla_put_failure;
+	}
+
+	return 0;
+
+nla_put_failure:
+	return -EMSGSIZE;
+}
+
+static struct rtnl_link_ops macsec_link_ops __read_mostly = {
+	.kind		= "macsec",
+	.priv_size	= sizeof(struct macsec_dev),
+	.maxtype	= IFLA_MACSEC_MAX,
+	.policy		= macsec_rtnl_policy,
+	.setup		= macsec_setup,
+	.validate	= macsec_validate_attr,
+	.newlink	= macsec_newlink,
+	.changelink	= macsec_changelink,
+	.dellink	= macsec_dellink,
+	.get_size	= macsec_get_size,
+	.fill_info	= macsec_fill_info,
+	.get_link_net	= macsec_get_link_net,
+};
+
+static bool is_macsec_master(struct net_device *dev)
+{
+	return rcu_access_pointer(dev->rx_handler) == macsec_handle_frame;
+}
+
+static int macsec_notify(struct notifier_block *this, unsigned long event,
+			 void *ptr)
+{
+	struct net_device *real_dev = netdev_notifier_info_to_dev(ptr);
+	LIST_HEAD(head);
+
+	if (!is_macsec_master(real_dev))
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_UNREGISTER: {
+		struct macsec_dev *m, *n;
+		struct macsec_rxh_data *rxd;
+
+		rxd = macsec_data_rtnl(real_dev);
+		list_for_each_entry_safe(m, n, &rxd->secys, secys) {
+			macsec_dellink(m->secy.netdev, &head);
+		}
+		unregister_netdevice_many(&head);
+		break;
+	}
+	case NETDEV_CHANGEMTU: {
+		struct macsec_dev *m;
+		struct macsec_rxh_data *rxd;
+
+		rxd = macsec_data_rtnl(real_dev);
+		list_for_each_entry(m, &rxd->secys, secys) {
+			struct net_device *dev = m->secy.netdev;
+			unsigned int mtu = real_dev->mtu - (m->secy.icv_len +
+							    macsec_extra_len(true));
+
+			if (dev->mtu > mtu)
+				dev_set_mtu(dev, mtu);
+		}
+	}
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block macsec_notifier = {
+	.notifier_call = macsec_notify,
+};
+
+static int __init macsec_init(void)
+{
+	int err;
+
+	pr_info("MACsec IEEE 802.1AE\n");
+	err = register_netdevice_notifier(&macsec_notifier);
+	if (err)
+		return err;
+
+	err = rtnl_link_register(&macsec_link_ops);
+	if (err)
+		goto notifier;
+
+	err = genl_register_family_with_ops(&macsec_fam, macsec_genl_ops);
+	if (err)
+		goto rtnl;
+
+	return 0;
+
+rtnl:
+	rtnl_link_unregister(&macsec_link_ops);
+notifier:
+	unregister_netdevice_notifier(&macsec_notifier);
+	return err;
+}
+
+static void __exit macsec_exit(void)
+{
+	genl_unregister_family(&macsec_fam);
+	rtnl_link_unregister(&macsec_link_ops);
+	unregister_netdevice_notifier(&macsec_notifier);
+}
+
+module_init(macsec_init);
+module_exit(macsec_exit);
+
+MODULE_ALIAS_RTNL_LINK("macsec");
+
+MODULE_DESCRIPTION("MACsec IEEE 802.1AE");
+MODULE_LICENSE("GPL v2");
-- 
cgit v0.10.2


From fbd40ea0180a2d328c5adc61414dc8bab9335ce2 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sun, 13 Mar 2016 23:28:00 -0400
Subject: ipv4: Don't do expensive useless work during inetdev destroy.

When an inetdev is destroyed, every address assigned to the interface
is removed.  And in this scenerio we do two pointless things which can
be very expensive if the number of assigned interfaces is large:

1) Address promotion.  We are deleting all addresses, so there is no
   point in doing this.

2) A full nf conntrack table purge for every address.  We only need to
   do this once, as is already caught by the existing
   masq_dev_notifier so masq_inet_event() can skip this.

Reported-by: Solar Designer <solar@openwall.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tested-by: Cyrill Gorcunov <gorcunov@openvz.org>

diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 65e76a4..e333bc8 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -334,6 +334,9 @@ static void __inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap,
 
 	ASSERT_RTNL();
 
+	if (in_dev->dead)
+		goto no_promotions;
+
 	/* 1. Deleting primary ifaddr forces deletion all secondaries
 	 * unless alias promotion is set
 	 **/
@@ -380,6 +383,7 @@ static void __inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap,
 			fib_del_ifaddr(ifa, ifa1);
 	}
 
+no_promotions:
 	/* 2. Unlink it */
 
 	*ifap = ifa1->ifa_next;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 4734475..21add55 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -922,6 +922,9 @@ void fib_del_ifaddr(struct in_ifaddr *ifa, struct in_ifaddr *iprim)
 		subnet = 1;
 	}
 
+	if (in_dev->dead)
+		goto no_promotions;
+
 	/* Deletion is more complicated than add.
 	 * We should take care of not to delete too much :-)
 	 *
@@ -997,6 +1000,7 @@ void fib_del_ifaddr(struct in_ifaddr *ifa, struct in_ifaddr *iprim)
 		}
 	}
 
+no_promotions:
 	if (!(ok & BRD_OK))
 		fib_magic(RTM_DELROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim);
 	if (subnet && ifa->ifa_prefixlen < 31) {
diff --git a/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c b/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c
index c6eb421..ea91058 100644
--- a/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c
+++ b/net/ipv4/netfilter/nf_nat_masquerade_ipv4.c
@@ -108,10 +108,18 @@ static int masq_inet_event(struct notifier_block *this,
 			   unsigned long event,
 			   void *ptr)
 {
-	struct net_device *dev = ((struct in_ifaddr *)ptr)->ifa_dev->dev;
+	struct in_device *idev = ((struct in_ifaddr *)ptr)->ifa_dev;
 	struct netdev_notifier_info info;
 
-	netdev_notifier_info_init(&info, dev);
+	/* The masq_dev_notifier will catch the case of the device going
+	 * down.  So if the inetdev is dead and being destroyed we have
+	 * no work to do.  Otherwise this is an individual address removal
+	 * and we have to perform the flush.
+	 */
+	if (idev->dead)
+		return NOTIFY_DONE;
+
+	netdev_notifier_info_init(&info, idev->dev);
 	return masq_device_event(this, event, &info);
 }
 
-- 
cgit v0.10.2


From 01cfbad79a5e2b835abf6a8154a341d75a6fc8cd Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 11 Mar 2016 14:05:34 -0800
Subject: ipv4: Update parameters for csum_tcpudp_magic to their original types

This patch updates all instances of csum_tcpudp_magic and
csum_tcpudp_nofold to reflect the types that are usually used as the source
inputs.  For example the protocol field is populated based on nexthdr which
is actually an unsigned 8 bit value.  The length is usually populated based
on skb->len which is an unsigned integer.

This addresses an issue in which the IPv6 function csum_ipv6_magic was
generating a checksum using the full 32b of skb->len while
csum_tcpudp_magic was only using the lower 16 bits.  As a result we could
run into issues when attempting to adjust the checksum as there was no
protocol agnostic way to update it.

With this change the value is still truncated as many architectures use
"(len + proto) << 8", however this truncation only occurs for values
greater than 16776960 in length and as such is unlikely to occur as we stop
the inner headers at ~64K in size.

I did have to make a few minor changes in the arm, mn10300, nios2, and
score versions of the function in order to support these changes as they
were either using things such as an OR to combine the protocol and length,
or were using ntohs to convert the length which would have truncated the
value.

I also updated a few spots in terms of whitespace and type differences for
the addresses.  Most of this was just to make sure all of the definitions
were in sync going forward.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/alpha/include/asm/checksum.h b/arch/alpha/include/asm/checksum.h
index d3854bb..cba34b1 100644
--- a/arch/alpha/include/asm/checksum.h
+++ b/arch/alpha/include/asm/checksum.h
@@ -13,14 +13,11 @@ extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl);
  * computes the checksum of the TCP/UDP pseudo-header
  * returns a 16-bit checksum, already complemented
  */
-extern __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					   unsigned short len,
-					   unsigned short proto,
-					   __wsum sum);
+__sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum);
 
 __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-				unsigned short len, unsigned short proto,
-				__wsum sum);
+			  __u32 len, __u8 proto, __wsum sum);
 
 /*
  * computes the checksum of a memory block at buff, length len,
diff --git a/arch/alpha/lib/checksum.c b/arch/alpha/lib/checksum.c
index 199f6ef..377f9e3 100644
--- a/arch/alpha/lib/checksum.c
+++ b/arch/alpha/lib/checksum.c
@@ -42,9 +42,7 @@ static inline unsigned short from64to16(unsigned long x)
  * returns a 16-bit checksum, already complemented.
  */
 __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-				   unsigned short len,
-				   unsigned short proto,
-				   __wsum sum)
+			  __u32 len, __u8 proto, __wsum sum)
 {
 	return (__force __sum16)~from64to16(
 		(__force u64)saddr + (__force u64)daddr +
@@ -52,9 +50,7 @@ __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
 }
 
 __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-				   unsigned short len,
-				   unsigned short proto,
-				   __wsum sum)
+			  __u32 len, __u8 proto, __wsum sum)
 {
 	unsigned long result;
 
diff --git a/arch/arc/include/asm/checksum.h b/arch/arc/include/asm/checksum.h
index 1095729..913eb4a 100644
--- a/arch/arc/include/asm/checksum.h
+++ b/arch/arc/include/asm/checksum.h
@@ -70,8 +70,8 @@ ip_fast_csum(const void *iph, unsigned int ihl)
  * SA [4], DA [4], zeroes [1], Proto[1], TCP Seg(hdr+data) Len [2]
  */
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	__asm__ __volatile__(
 	"	add.f %0, %0, %1	\n"
diff --git a/arch/arm/include/asm/checksum.h b/arch/arm/include/asm/checksum.h
index 5233151..42d020b 100644
--- a/arch/arm/include/asm/checksum.h
+++ b/arch/arm/include/asm/checksum.h
@@ -84,10 +84,10 @@ ip_fast_csum(const void *iph, unsigned int ihl)
 }
 
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
-	u32 lenprot = len | proto << 16;
+	u32 lenprot = len + proto;
 	if (__builtin_constant_p(sum) && sum == 0) {
 		__asm__(
 		"adds	%0, %1, %2	@ csum_tcpudp_nofold0	\n\t"
@@ -121,8 +121,8 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len,
-		  unsigned short proto, __wsum sum)
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
diff --git a/arch/avr32/include/asm/checksum.h b/arch/avr32/include/asm/checksum.h
index 4ddbfd2..4ab7d5b 100644
--- a/arch/avr32/include/asm/checksum.h
+++ b/arch/avr32/include/asm/checksum.h
@@ -111,9 +111,8 @@ static inline __sum16 csum_fold(__wsum sum)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					       unsigned short len,
-					       unsigned short proto,
-					       __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	asm("	add	%0, %1\n"
 	    "	adc	%0, %0, %2\n"
@@ -132,9 +131,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/blackfin/include/asm/checksum.h b/arch/blackfin/include/asm/checksum.h
index 623cc7f..e7134bf 100644
--- a/arch/blackfin/include/asm/checksum.h
+++ b/arch/blackfin/include/asm/checksum.h
@@ -14,8 +14,8 @@
  */
 
 static inline __wsum
-__csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+__csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		     __u8 proto, __wsum sum)
 {
 	unsigned int carry;
 
diff --git a/arch/c6x/include/asm/checksum.h b/arch/c6x/include/asm/checksum.h
index 7246816..249b0e4 100644
--- a/arch/c6x/include/asm/checksum.h
+++ b/arch/c6x/include/asm/checksum.h
@@ -10,8 +10,8 @@
 #define _ASM_C6X_CHECKSUM_H
 
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	unsigned long long tmp;
 
diff --git a/arch/cris/include/arch-v10/arch/checksum.h b/arch/cris/include/arch-v10/arch/checksum.h
index b8000c5..d1d1bd9 100644
--- a/arch/cris/include/arch-v10/arch/checksum.h
+++ b/arch/cris/include/arch-v10/arch/checksum.h
@@ -9,8 +9,8 @@
  */
 
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	__wsum res;
 	__asm__ ("add.d %2, %0\n\t"
diff --git a/arch/cris/include/arch-v32/arch/checksum.h b/arch/cris/include/arch-v32/arch/checksum.h
index e5dcfce..65cf205 100644
--- a/arch/cris/include/arch-v32/arch/checksum.h
+++ b/arch/cris/include/arch-v32/arch/checksum.h
@@ -11,7 +11,7 @@
  */
 static inline __wsum
 csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-		   unsigned short len, unsigned short proto, __wsum sum)
+		   __u32 len, __u8 proto, __wsum sum)
 {
 	__wsum res;
 
diff --git a/arch/cris/include/asm/checksum.h b/arch/cris/include/asm/checksum.h
index 75dcb77d..ea949c6 100644
--- a/arch/cris/include/asm/checksum.h
+++ b/arch/cris/include/asm/checksum.h
@@ -63,9 +63,8 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
  */
 
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/frv/include/asm/checksum.h b/arch/frv/include/asm/checksum.h
index 269da09..cd59cd4 100644
--- a/arch/frv/include/asm/checksum.h
+++ b/arch/frv/include/asm/checksum.h
@@ -105,8 +105,8 @@ static inline __sum16 csum_fold(__wsum sum)
  * returns a 16-bit checksum, already complemented
  */
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		  unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	asm("	addcc		%1,%0,%0,icc0	\n"
 	    "	addxcc		%2,%0,%0,icc0	\n"
@@ -120,8 +120,8 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
 }
 
 static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len,
-		  unsigned short proto, __wsum sum)
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/hexagon/include/asm/checksum.h b/arch/hexagon/include/asm/checksum.h
index 46ec8a7..d9f58d6 100644
--- a/arch/hexagon/include/asm/checksum.h
+++ b/arch/hexagon/include/asm/checksum.h
@@ -38,12 +38,12 @@ __wsum csum_partial_copy_nocheck(const void *src, void *dst,
  * returns a 16-bit checksum, already complemented
  */
 #define csum_tcpudp_nofold csum_tcpudp_nofold
-__wsum csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
-	unsigned short len, unsigned short proto, __wsum sum);
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum);
 
 #define csum_tcpudp_magic csum_tcpudp_magic
-__sum16 csum_tcpudp_magic(unsigned long saddr, unsigned long daddr,
-	unsigned short len, unsigned short proto, __wsum sum);
+__sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum);
 
 #include <asm-generic/checksum.h>
 
diff --git a/arch/hexagon/lib/checksum.c b/arch/hexagon/lib/checksum.c
index 8169f78..617506d 100644
--- a/arch/hexagon/lib/checksum.c
+++ b/arch/hexagon/lib/checksum.c
@@ -60,18 +60,16 @@ static inline unsigned short from64to16(u64 x)
  * computes the checksum of the TCP/UDP pseudo-header
  * returns a 16-bit checksum, already complemented.
  */
-__sum16 csum_tcpudp_magic(unsigned long saddr, unsigned long daddr,
-			  unsigned short len, unsigned short proto,
-			  __wsum sum)
+__sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum)
 {
 	return (__force __sum16)~from64to16(
 		(__force u64)saddr + (__force u64)daddr +
 		(__force u64)sum + ((len + proto) << 8));
 }
 
-__wsum csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
-			  unsigned short len, unsigned short proto,
-			  __wsum sum)
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum)
 {
 	u64 result;
 
diff --git a/arch/ia64/include/asm/checksum.h b/arch/ia64/include/asm/checksum.h
index 97af155..ac9c687 100644
--- a/arch/ia64/include/asm/checksum.h
+++ b/arch/ia64/include/asm/checksum.h
@@ -16,15 +16,11 @@ extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl);
  * Computes the checksum of the TCP/UDP pseudo-header returns a 16-bit
  * checksum, already complemented
  */
-extern __sum16 csum_tcpudp_magic (__be32 saddr, __be32 daddr,
-					     unsigned short len,
-					     unsigned short proto,
-					     __wsum sum);
+extern __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+				 __u32 len, __u8 proto, __wsum sum);
 
-extern __wsum csum_tcpudp_nofold (__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
-					__wsum sum);
+extern __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+				 __u32 len, __u8 proto, __wsum sum);
 
 /*
  * Computes the checksum of a memory block at buff, length len,
diff --git a/arch/ia64/lib/checksum.c b/arch/ia64/lib/checksum.c
index 9fc9550..2cb23cb 100644
--- a/arch/ia64/lib/checksum.c
+++ b/arch/ia64/lib/checksum.c
@@ -34,8 +34,8 @@ from64to16 (unsigned long x)
  * returns a 16-bit checksum, already complemented.
  */
 __sum16
-csum_tcpudp_magic (__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	return (__force __sum16)~from64to16(
 		(__force u64)saddr + (__force u64)daddr +
@@ -45,8 +45,8 @@ csum_tcpudp_magic (__be32 saddr, __be32 daddr, unsigned short len,
 EXPORT_SYMBOL(csum_tcpudp_magic);
 
 __wsum
-csum_tcpudp_nofold (__be32 saddr, __be32 daddr, unsigned short len,
-		    unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	unsigned long result;
 
diff --git a/arch/m32r/include/asm/checksum.h b/arch/m32r/include/asm/checksum.h
index a7a7c4f..d68e93c 100644
--- a/arch/m32r/include/asm/checksum.h
+++ b/arch/m32r/include/asm/checksum.h
@@ -114,9 +114,8 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					       unsigned short len,
-					       unsigned short proto,
-					       __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 #if defined(__LITTLE_ENDIAN)
 	unsigned long len_proto = (proto + len) << 8;
@@ -145,9 +144,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/metag/include/asm/checksum.h b/arch/metag/include/asm/checksum.h
index 08dd1cc..f65fe83 100644
--- a/arch/metag/include/asm/checksum.h
+++ b/arch/metag/include/asm/checksum.h
@@ -59,8 +59,7 @@ extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl);
  * returns a 16-bit checksum, already complemented
  */
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	unsigned long len_proto = (proto + len) << 8;
@@ -78,8 +77,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
 }
 
 static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len,
-		  unsigned short proto, __wsum sum)
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
diff --git a/arch/microblaze/include/asm/checksum.h b/arch/microblaze/include/asm/checksum.h
index 0185cbe..adeeceb 100644
--- a/arch/microblaze/include/asm/checksum.h
+++ b/arch/microblaze/include/asm/checksum.h
@@ -16,8 +16,8 @@
  */
 #define csum_tcpudp_nofold	csum_tcpudp_nofold
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	__asm__("add %0, %0, %1\n\t"
 		"addc %0, %0, %2\n\t"
diff --git a/arch/mips/include/asm/checksum.h b/arch/mips/include/asm/checksum.h
index 3ceacde..c635541 100644
--- a/arch/mips/include/asm/checksum.h
+++ b/arch/mips/include/asm/checksum.h
@@ -160,9 +160,9 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 }
 #define ip_fast_csum ip_fast_csum
 
-static inline __wsum csum_tcpudp_nofold(__be32 saddr,
-	__be32 daddr, unsigned short len, unsigned short proto,
-	__wsum sum)
+static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	__asm__(
 	"	.set	push		# csum_tcpudp_nofold\n"
diff --git a/arch/mn10300/include/asm/checksum.h b/arch/mn10300/include/asm/checksum.h
index 9fb2a8d..c80df5b 100644
--- a/arch/mn10300/include/asm/checksum.h
+++ b/arch/mn10300/include/asm/checksum.h
@@ -37,16 +37,11 @@ static inline __sum16 csum_fold(__wsum sum)
 	return (~sum) >> 16;
 }
 
-static inline __wsum csum_tcpudp_nofold(unsigned long saddr,
-					unsigned long daddr,
-					unsigned short len,
-					unsigned short proto,
+static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
-	__wsum tmp;
-
-	tmp = (__wsum) ntohs(len) << 16;
-	tmp += (__wsum) proto << 8;
+	__wsum tmp = (__wsum)((len + proto) << 8);
 
 	asm(
 		"	add	%1,%0		\n"
@@ -64,10 +59,8 @@ static inline __wsum csum_tcpudp_nofold(unsigned long saddr,
  * computes the checksum of the TCP/UDP pseudo-header
  * returns a 16-bit checksum, already complemented
  */
-static inline __sum16 csum_tcpudp_magic(unsigned long saddr,
-					unsigned long daddr,
-					unsigned short len,
-					unsigned short proto,
+static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
diff --git a/arch/nios2/include/asm/checksum.h b/arch/nios2/include/asm/checksum.h
index 6bc1f0d..703c5ee 100644
--- a/arch/nios2/include/asm/checksum.h
+++ b/arch/nios2/include/asm/checksum.h
@@ -45,8 +45,7 @@ static inline __sum16 csum_fold(__wsum sum)
  */
 #define csum_tcpudp_nofold csum_tcpudp_nofold
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	__asm__ __volatile__(
@@ -60,7 +59,7 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
 		"cmpltu	r8, %0, %3\n"
 		"add	%0, %0, r8\n"	/* add carry */
 		: "=r" (sum), "=r" (saddr)
-		: "r" (daddr), "r" ((ntohs(len) << 16) + (proto * 256)),
+		: "r" (daddr), "r" ((len + proto) << 8),
 		  "0" (sum),
 		  "1" (saddr)
 		: "r8");
@@ -69,8 +68,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
 }
 
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto, __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
diff --git a/arch/parisc/include/asm/checksum.h b/arch/parisc/include/asm/checksum.h
index c84b2fc..9815ab1 100644
--- a/arch/parisc/include/asm/checksum.h
+++ b/arch/parisc/include/asm/checksum.h
@@ -85,9 +85,8 @@ static inline __sum16 csum_fold(__wsum csum)
 }
  
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					       unsigned short len,
-					       unsigned short proto,
-					       __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	__asm__(
 	"	add  %1, %0, %0\n"
@@ -104,9 +103,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/s390/include/asm/checksum.h b/arch/s390/include/asm/checksum.h
index 7403648..d7f100c 100644
--- a/arch/s390/include/asm/checksum.h
+++ b/arch/s390/include/asm/checksum.h
@@ -91,8 +91,7 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
  * returns a 32-bit checksum
  */
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-                   unsigned short len, unsigned short proto,
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len, __u8 proto,
                    __wsum sum)
 {
 	__u32 csum = (__force __u32)sum;
@@ -118,8 +117,7 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  */
 
 static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-                  unsigned short len, unsigned short proto,
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len, __u8 proto,
                   __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
diff --git a/arch/score/include/asm/checksum.h b/arch/score/include/asm/checksum.h
index 961bd64..a375bc2 100644
--- a/arch/score/include/asm/checksum.h
+++ b/arch/score/include/asm/checksum.h
@@ -127,10 +127,10 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 }
 
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
-	unsigned long tmp = (ntohs(len) << 16) + proto * 256;
+	unsigned long tmp = (len + proto) << 8;
 	__asm__ __volatile__(
 		".set volatile\n\t"
 		"add\t%0, %0, %2\n\t"
@@ -161,8 +161,8 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len,
-		unsigned short proto, __wsum sum)
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
diff --git a/arch/sh/include/asm/checksum_32.h b/arch/sh/include/asm/checksum_32.h
index 14b7ac2..fd730f1 100644
--- a/arch/sh/include/asm/checksum_32.h
+++ b/arch/sh/include/asm/checksum_32.h
@@ -115,8 +115,7 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 #ifdef __LITTLE_ENDIAN__
@@ -142,8 +141,7 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
diff --git a/arch/sparc/include/asm/checksum_32.h b/arch/sparc/include/asm/checksum_32.h
index 426b238..86ae655 100644
--- a/arch/sparc/include/asm/checksum_32.h
+++ b/arch/sparc/include/asm/checksum_32.h
@@ -170,9 +170,8 @@ static inline __sum16 csum_fold(__wsum sum)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					       unsigned short len,
-					       unsigned short proto,
-					       __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	__asm__ __volatile__("addcc\t%1, %0, %0\n\t"
 			     "addxcc\t%2, %0, %0\n\t"
@@ -190,9 +189,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/sparc/include/asm/checksum_64.h b/arch/sparc/include/asm/checksum_64.h
index b8779a6..ef0c6f4 100644
--- a/arch/sparc/include/asm/checksum_64.h
+++ b/arch/sparc/include/asm/checksum_64.h
@@ -96,8 +96,7 @@ static inline __sum16 csum_fold(__wsum sum)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					unsigned int len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	__asm__ __volatile__(
@@ -116,8 +115,7 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
diff --git a/arch/unicore32/include/asm/checksum.h b/arch/unicore32/include/asm/checksum.h
index f55c3f9..23ceb9e 100644
--- a/arch/unicore32/include/asm/checksum.h
+++ b/arch/unicore32/include/asm/checksum.h
@@ -20,8 +20,8 @@
  */
 
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	__asm__(
 	"add.a	%0, %1, %2\n"
diff --git a/arch/x86/include/asm/checksum_32.h b/arch/x86/include/asm/checksum_32.h
index f50de69..6f38060 100644
--- a/arch/x86/include/asm/checksum_32.h
+++ b/arch/x86/include/asm/checksum_32.h
@@ -112,8 +112,7 @@ static inline __sum16 csum_fold(__wsum sum)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	asm("addl %1, %0	;\n"
@@ -131,8 +130,7 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto,
+					__u32 len, __u8 proto,
 					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
diff --git a/arch/x86/include/asm/checksum_64.h b/arch/x86/include/asm/checksum_64.h
index cd00e17..97b98e2 100644
--- a/arch/x86/include/asm/checksum_64.h
+++ b/arch/x86/include/asm/checksum_64.h
@@ -84,8 +84,8 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
  * 32bit unfolded.
  */
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum)
 {
 	asm("  addl %1, %0\n"
 	    "  adcl %2, %0\n"
@@ -110,8 +110,8 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
  * complemented and ready to be filled in.
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					unsigned short len,
-					unsigned short proto, __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
diff --git a/arch/x86/um/asm/checksum.h b/arch/x86/um/asm/checksum.h
index ee94018..54d96f1 100644
--- a/arch/x86/um/asm/checksum.h
+++ b/arch/x86/um/asm/checksum.h
@@ -87,8 +87,8 @@ static inline __sum16 csum_fold(__wsum sum)
  * 32bit unfolded.
  */
 static inline __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		   unsigned short proto, __wsum sum)
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	asm("  addl %1, %0\n"
 	    "  adcl %2, %0\n"
@@ -104,9 +104,8 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-					   unsigned short len,
-					   unsigned short proto,
-					   __wsum sum)
+					__u32 len, __u8 proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/arch/xtensa/include/asm/checksum.h b/arch/xtensa/include/asm/checksum.h
index 0593de6..62254e6 100644
--- a/arch/xtensa/include/asm/checksum.h
+++ b/arch/xtensa/include/asm/checksum.h
@@ -123,9 +123,8 @@ static __inline__ __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
 }
 
 static __inline__ __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					    __u32 len, __u8 proto,
+					    __wsum sum)
 {
 
 #ifdef __XTENSA_EL__
@@ -157,9 +156,8 @@ static __inline__ __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static __inline__ __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						       unsigned short len,
-						       unsigned short proto,
-						       __wsum sum)
+					    __u32 len, __u8 proto,
+					    __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
diff --git a/include/asm-generic/checksum.h b/include/asm-generic/checksum.h
index 59811df..3150cbd 100644
--- a/include/asm-generic/checksum.h
+++ b/include/asm-generic/checksum.h
@@ -65,14 +65,14 @@ static inline __sum16 csum_fold(__wsum csum)
  * returns a 16-bit checksum, already complemented
  */
 extern __wsum
-csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
-		unsigned short proto, __wsum sum);
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len,
+		   __u8 proto, __wsum sum);
 #endif
 
 #ifndef csum_tcpudp_magic
 static inline __sum16
-csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len,
-		  unsigned short proto, __wsum sum)
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
 }
diff --git a/lib/checksum.c b/lib/checksum.c
index 8b39e86..d3ec93f 100644
--- a/lib/checksum.c
+++ b/lib/checksum.c
@@ -191,9 +191,7 @@ static inline u32 from64to32(u64 x)
 }
 
 __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-			unsigned short len,
-			unsigned short proto,
-			__wsum sum)
+			  __u32 len, __u8 proto, __wsum sum)
 {
 	unsigned long long s = (__force u32)sum;
 
-- 
cgit v0.10.2


From 1e94082963747b551b129528714827f76a090e93 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 11 Mar 2016 14:05:41 -0800
Subject: ipv6: Pass proto to csum_ipv6_magic as __u8 instead of unsigned short

This patch updates csum_ipv6_magic so that it correctly recognizes that
protocol is a unsigned 8 bit value.

This will allow us to better understand what limitations may or may not be
present in how we handle the data.  For example there are a number of
places that call htonl on the protocol value.  This is likely not necessary
and can be replaced with a multiplication by ntohl(1) which will be
converted to a shift by the compiler.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/alpha/include/asm/checksum.h b/arch/alpha/include/asm/checksum.h
index cba34b1..f2bbdd2 100644
--- a/arch/alpha/include/asm/checksum.h
+++ b/arch/alpha/include/asm/checksum.h
@@ -67,6 +67,5 @@ static inline __sum16 csum_fold(__wsum csum)
 #define _HAVE_ARCH_IPV6_CSUM
 extern __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 			       const struct in6_addr *daddr,
-			       __u32 len, unsigned short proto,
-			       __wsum sum);
+			       __u32 len, __u8 proto, __wsum sum);
 #endif
diff --git a/arch/arm/include/asm/checksum.h b/arch/arm/include/asm/checksum.h
index 42d020b..524692f 100644
--- a/arch/arm/include/asm/checksum.h
+++ b/arch/arm/include/asm/checksum.h
@@ -144,8 +144,8 @@ __csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, __
 		__be32 proto, __wsum sum);
 
 static inline __sum16
-csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr, __u32 len,
-		unsigned short proto, __wsum sum)
+csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr,
+		__u32 len, __u8 proto, __wsum sum)
 {
 	return csum_fold(__csum_ipv6_magic(saddr, daddr, htonl(len),
 					   htonl(proto), sum));
diff --git a/arch/frv/include/asm/checksum.h b/arch/frv/include/asm/checksum.h
index cd59cd4..b77388c 100644
--- a/arch/frv/include/asm/checksum.h
+++ b/arch/frv/include/asm/checksum.h
@@ -135,7 +135,7 @@ extern __sum16 ip_compute_csum(const void *buff, int len);
 #define _HAVE_ARCH_IPV6_CSUM
 static inline __sum16
 csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr,
-		__u32 len, unsigned short proto, __wsum sum)
+		__u32 len, __u8 proto, __wsum sum)
 {
 	unsigned long tmp, tmp2;
 
diff --git a/arch/ia64/include/asm/checksum.h b/arch/ia64/include/asm/checksum.h
index ac9c687..7accf54 100644
--- a/arch/ia64/include/asm/checksum.h
+++ b/arch/ia64/include/asm/checksum.h
@@ -69,7 +69,7 @@ static inline __sum16 csum_fold(__wsum csum)
 #define _HAVE_ARCH_IPV6_CSUM	1
 struct in6_addr;
 extern __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
-	const struct in6_addr *daddr, __u32 len, unsigned short proto,
-	__wsum csum);
+			       const struct in6_addr *daddr,
+			       __u32 len, __u8 proto, __wsum csum);
 
 #endif /* _ASM_IA64_CHECKSUM_H */
diff --git a/arch/m68k/include/asm/checksum.h b/arch/m68k/include/asm/checksum.h
index 2f88d86..75e91f0 100644
--- a/arch/m68k/include/asm/checksum.h
+++ b/arch/m68k/include/asm/checksum.h
@@ -117,7 +117,7 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static __inline__ __sum16
 csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr,
-		__u32 len, unsigned short proto, __wsum sum)
+		__u32 len, __u8 proto, __wsum sum)
 {
 	register unsigned long tmp;
 	__asm__("addl %2@,%0\n\t"
diff --git a/arch/mips/include/asm/checksum.h b/arch/mips/include/asm/checksum.h
index c635541..bce1ce5 100644
--- a/arch/mips/include/asm/checksum.h
+++ b/arch/mips/include/asm/checksum.h
@@ -215,7 +215,7 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 					  const struct in6_addr *daddr,
-					  __u32 len, unsigned short proto,
+					  __u32 len, __u8 proto,
 					  __wsum sum)
 {
 	__wsum tmp;
diff --git a/arch/parisc/include/asm/checksum.h b/arch/parisc/include/asm/checksum.h
index 9815ab1..60c2c42 100644
--- a/arch/parisc/include/asm/checksum.h
+++ b/arch/parisc/include/asm/checksum.h
@@ -122,7 +122,7 @@ static inline __sum16 ip_compute_csum(const void *buf, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 					  const struct in6_addr *daddr,
-					  __u32 len, unsigned short proto,
+					  __u32 len, __u8 proto,
 					  __wsum sum)
 {
 	__asm__ __volatile__ (
diff --git a/arch/score/include/asm/checksum.h b/arch/score/include/asm/checksum.h
index a375bc2..539d9fd 100644
--- a/arch/score/include/asm/checksum.h
+++ b/arch/score/include/asm/checksum.h
@@ -179,9 +179,8 @@ static inline unsigned short ip_compute_csum(const void *buff, int len)
 
 #define _HAVE_ARCH_IPV6_CSUM
 static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
-				const struct in6_addr *daddr,
-				__u32 len, unsigned short proto,
-				__wsum sum)
+				      const struct in6_addr *daddr,
+				      __u32 len, __u8 proto, __wsum sum)
 {
 	__asm__ __volatile__(
 		".set\tvolatile\t\t\t# csum_ipv6_magic\n\t"
diff --git a/arch/sh/include/asm/checksum_32.h b/arch/sh/include/asm/checksum_32.h
index fd730f1..9c84386 100644
--- a/arch/sh/include/asm/checksum_32.h
+++ b/arch/sh/include/asm/checksum_32.h
@@ -159,8 +159,7 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 				      const struct in6_addr *daddr,
-				      __u32 len, unsigned short proto,
-				      __wsum sum)
+				      __u32 len, __u8 proto, __wsum sum)
 {
 	unsigned int __dummy;
 	__asm__("clrt\n\t"
diff --git a/arch/sparc/include/asm/checksum_32.h b/arch/sparc/include/asm/checksum_32.h
index 86ae655..eff748c 100644
--- a/arch/sparc/include/asm/checksum_32.h
+++ b/arch/sparc/include/asm/checksum_32.h
@@ -199,8 +199,7 @@ static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
 
 static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 				      const struct in6_addr *daddr,
-				      __u32 len, unsigned short proto,
-				      __wsum sum)
+				      __u32 len, __u8 proto, __wsum sum)
 {
 	__asm__ __volatile__ (
 		"addcc	%3, %4, %%g4\n\t"
diff --git a/arch/sparc/include/asm/checksum_64.h b/arch/sparc/include/asm/checksum_64.h
index ef0c6f4..0395d75 100644
--- a/arch/sparc/include/asm/checksum_64.h
+++ b/arch/sparc/include/asm/checksum_64.h
@@ -125,8 +125,7 @@ static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
 
 static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 				      const struct in6_addr *daddr,
-				      __u32 len, unsigned short proto,
-				      __wsum sum)
+				      __u32 len, __u8 proto, __wsum sum)
 {
 	__asm__ __volatile__ (
 "	addcc		%3, %4, %%g7\n"
diff --git a/arch/x86/include/asm/checksum_32.h b/arch/x86/include/asm/checksum_32.h
index 6f38060..532f85e 100644
--- a/arch/x86/include/asm/checksum_32.h
+++ b/arch/x86/include/asm/checksum_32.h
@@ -149,8 +149,7 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 				      const struct in6_addr *daddr,
-				      __u32 len, unsigned short proto,
-				      __wsum sum)
+				      __u32 len, __u8 proto, __wsum sum)
 {
 	asm("addl 0(%1), %0	;\n"
 	    "adcl 4(%1), %0	;\n"
diff --git a/arch/x86/include/asm/checksum_64.h b/arch/x86/include/asm/checksum_64.h
index 97b98e2..c020ee7 100644
--- a/arch/x86/include/asm/checksum_64.h
+++ b/arch/x86/include/asm/checksum_64.h
@@ -177,7 +177,7 @@ struct in6_addr;
 #define _HAVE_ARCH_IPV6_CSUM 1
 extern __sum16
 csum_ipv6_magic(const struct in6_addr *saddr, const struct in6_addr *daddr,
-		__u32 len, unsigned short proto, __wsum sum);
+		__u32 len, __u8 proto, __wsum sum);
 
 static inline unsigned add32_with_carry(unsigned a, unsigned b)
 {
diff --git a/arch/x86/lib/csum-wrappers_64.c b/arch/x86/lib/csum-wrappers_64.c
index 1318f75..28a6654 100644
--- a/arch/x86/lib/csum-wrappers_64.c
+++ b/arch/x86/lib/csum-wrappers_64.c
@@ -135,7 +135,7 @@ EXPORT_SYMBOL(csum_partial_copy_nocheck);
 
 __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 			const struct in6_addr *daddr,
-			__u32 len, unsigned short proto, __wsum sum)
+			__u32 len, __u8 proto, __wsum sum)
 {
 	__u64 rest, sum64;
 
diff --git a/arch/x86/um/asm/checksum_32.h b/arch/x86/um/asm/checksum_32.h
index ab77b6f..83a75f8 100644
--- a/arch/x86/um/asm/checksum_32.h
+++ b/arch/x86/um/asm/checksum_32.h
@@ -13,7 +13,7 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 					  const struct in6_addr *daddr,
-					  __u32 len, unsigned short proto,
+					  __u32 len, __u8 proto,
 					  __wsum sum)
 {
 	__asm__(
diff --git a/arch/xtensa/include/asm/checksum.h b/arch/xtensa/include/asm/checksum.h
index 62254e6..ec35074 100644
--- a/arch/xtensa/include/asm/checksum.h
+++ b/arch/xtensa/include/asm/checksum.h
@@ -175,7 +175,7 @@ static __inline__ __sum16 ip_compute_csum(const void *buff, int len)
 #define _HAVE_ARCH_IPV6_CSUM
 static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 					  const struct in6_addr *daddr,
-					  __u32 len, unsigned short proto,
+					  __u32 len, __u8 proto,
 					  __wsum sum)
 {
 	unsigned int __dummy;
diff --git a/include/net/ip6_checksum.h b/include/net/ip6_checksum.h
index 1a49b73..cca8405 100644
--- a/include/net/ip6_checksum.h
+++ b/include/net/ip6_checksum.h
@@ -37,8 +37,7 @@
 #ifndef _HAVE_ARCH_IPV6_CSUM
 __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 			const struct in6_addr *daddr,
-			__u32 len, unsigned short proto,
-			__wsum csum);
+			__u32 len, __u8 proto, __wsum csum);
 #endif
 
 static inline __wsum ip6_compute_pseudo(struct sk_buff *skb, int proto)
diff --git a/net/ipv6/ip6_checksum.c b/net/ipv6/ip6_checksum.c
index 8f92058..b2025bf 100644
--- a/net/ipv6/ip6_checksum.c
+++ b/net/ipv6/ip6_checksum.c
@@ -6,8 +6,7 @@
 #ifndef _HAVE_ARCH_IPV6_CSUM
 __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
 			const struct in6_addr *daddr,
-			__u32 len, unsigned short proto,
-			__wsum csum)
+			__u32 len, __u8 proto, __wsum csum)
 {
 
 	int carry;
-- 
cgit v0.10.2


From 08334824951dd6d1295860da07b1236d18b0b8df Mon Sep 17 00:00:00 2001
From: Alexander Duyck <aduyck@mirantis.com>
Date: Fri, 11 Mar 2016 14:05:47 -0800
Subject: GSO/UDP: Use skb->len instead of udph->len to determine length of
 original skb

It is possible for tunnels to end up generating IP or IPv6 datagrams that
are larger than 64K and expecting to be segmented.  As such we need to deal
with length values greater than 64K.  In order to accommodate this we need
to update the code to work with a 32b length value instead of a 16b one.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index f5abb1a..8a3405a 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -40,13 +40,19 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	__be16 protocol = skb->protocol;
 	u16 mac_len = skb->mac_len;
 	int udp_offset, outer_hlen;
-	u32 partial;
+	__wsum partial;
 
 	if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
 		goto out;
 
-	/* adjust partial header checksum to negate old length */
-	partial = (__force u32)uh->check + (__force u16)~uh->len;
+	/* Adjust partial header checksum to negate old length.
+	 * We cannot rely on the value contained in uh->len as it is
+	 * possible that the actual value exceeds the boundaries of the
+	 * 16 bit length field due to the header being added outside of an
+	 * IP or IPv6 frame that was already limited to 64K - 1.
+	 */
+	partial = csum_sub(csum_unfold(uh->check),
+			   (__force __wsum)htonl(skb->len));
 
 	/* setup inner skb. */
 	skb->encapsulation = 0;
@@ -119,8 +125,7 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 		if (!need_csum)
 			continue;
 
-		uh->check = ~csum_fold((__force __wsum)
-				       ((__force u32)len + partial));
+		uh->check = ~csum_fold(csum_add(partial, (__force __wsum)len));
 
 		if (skb->encapsulation || !offload_csum) {
 			uh->check = gso_make_checksum(skb, ~uh->check);
-- 
cgit v0.10.2


From eb43e023130b5021c1ffad4d5c84cb310ffcb0f6 Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:38:56 +0100
Subject: misc: sram: add optional ioremap without write combining

Some SRAM users may require non-bufferable access to the memory, which is
impossible, because devm_ioremap_wc() is used for setting sram->virt_base.

This commit adds optional flag 'no-memory-wc', which allow to choose remap
method, using DT property. Documentation is updated accordingly.

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/sram/sram.txt b/Documentation/devicetree/bindings/sram/sram.txt
index 42ee943..227e3a3 100644
--- a/Documentation/devicetree/bindings/sram/sram.txt
+++ b/Documentation/devicetree/bindings/sram/sram.txt
@@ -25,6 +25,11 @@ Required properties in the sram node:
 - ranges : standard definition, should translate from local addresses
            within the sram to bus addresses
 
+Optional properties in the sram node:
+
+- no-memory-wc : the flag indicating, that SRAM memory region has not to
+                 be remapped as write combining. WC is used by default.
+
 Required properties in the area nodes:
 
 - reg : iomem address range, relative to the SRAM range
diff --git a/drivers/misc/sram.c b/drivers/misc/sram.c
index 736dae7..69cdabe 100644
--- a/drivers/misc/sram.c
+++ b/drivers/misc/sram.c
@@ -360,7 +360,10 @@ static int sram_probe(struct platform_device *pdev)
 		return -EBUSY;
 	}
 
-	sram->virt_base = devm_ioremap_wc(sram->dev, res->start, size);
+	if (of_property_read_bool(pdev->dev.of_node, "no-memory-wc"))
+		sram->virt_base = devm_ioremap(sram->dev, res->start, size);
+	else
+		sram->virt_base = devm_ioremap_wc(sram->dev, res->start, size);
 	if (IS_ERR(sram->virt_base))
 		return PTR_ERR(sram->virt_base);
 
-- 
cgit v0.10.2


From 4a547a5a46d05442a767b8ea8b8feb2a5bd76585 Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:38:57 +0100
Subject: ARM: dts: armada-38x: add buffer manager nodes

Armada 38x network controller supports hardware buffer management (BM).
Since it is now enabled in mvneta driver, appropriate nodes can be added
to armada-38x.dtsi - for the actual common BM unit (bm@c8000) and its
internal SRAM (bm-bppi), which is used for indirect access to buffer
pointer ring residing in DRAM.

Pools - ports mapping, bm-bppi entry in 'soc' node's ranges and optional
parameters are supposed to be set in board files.

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/armada-38x.dtsi b/arch/arm/boot/dts/armada-38x.dtsi
index e8b7f67..066a8f0 100644
--- a/arch/arm/boot/dts/armada-38x.dtsi
+++ b/arch/arm/boot/dts/armada-38x.dtsi
@@ -540,6 +540,14 @@
 				status = "disabled";
 			};
 
+			bm: bm@c8000 {
+				compatible = "marvell,armada-380-neta-bm";
+				reg = <0xc8000 0xac>;
+				clocks = <&gateclk 13>;
+				internal-mem = <&bm_bppi>;
+				status = "disabled";
+			};
+
 			sata@e0000 {
 				compatible = "marvell,armada-380-ahci";
 				reg = <0xe0000 0x2000>;
@@ -618,6 +626,17 @@
 			#size-cells = <1>;
 			ranges = <0 MBUS_ID(0x09, 0x15) 0 0x800>;
 		};
+
+		bm_bppi: bm-bppi {
+			compatible = "mmio-sram";
+			reg = <MBUS_ID(0x0c, 0x04) 0 0x100000>;
+			ranges = <0 MBUS_ID(0x0c, 0x04) 0 0x100000>;
+			#address-cells = <1>;
+			#size-cells = <1>;
+			clocks = <&gateclk 13>;
+			no-memory-wc;
+			status = "disabled";
+		};
 	};
 
 	clocks {
-- 
cgit v0.10.2


From c49e99c2b25a412623412a461bb751239208b9b3 Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:38:58 +0100
Subject: ARM: dts: armada-38x: enable buffer manager support on Armada 38x
 boards

Since mvneta driver supports using hardware buffer management (BM), in
order to use it, board files have to be adjusted accordingly. This commit
enables BM on:
* A385-DB-AP - each port has its own pool for long and common pool for
short packets,
* A388-ClearFog - same as above,
* A388-DB - to each port unique 'short' and 'long' pools are mapped,
* A388-GP - same as above.

Moreover appropriate entry is added to 'soc' node ranges, as well as "okay"
status for 'bm' and 'bm-bppi' (internal SRAM) nodes.

[gregory.clement@free-electrons.com: add suppport for the ClearFog board]

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/armada-385-db-ap.dts b/arch/arm/boot/dts/armada-385-db-ap.dts
index acd5b15..5f9451b 100644
--- a/arch/arm/boot/dts/armada-385-db-ap.dts
+++ b/arch/arm/boot/dts/armada-385-db-ap.dts
@@ -61,7 +61,8 @@
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0xfff00000 0x100000
 			  MBUS_ID(0x09, 0x19) 0 0xf1100000 0x10000
-			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000>;
+			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0xf1200000 0x100000>;
 
 		internal-regs {
 			spi1: spi@10680 {
@@ -138,12 +139,18 @@
 				status = "okay";
 				phy = <&phy2>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <1>;
+				bm,pool-short = <3>;
 			};
 
 			ethernet@34000 {
 				status = "okay";
 				phy = <&phy1>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
+				bm,pool-short = <3>;
 			};
 
 			ethernet@70000 {
@@ -157,6 +164,13 @@
 				status = "okay";
 				phy = <&phy0>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
+				bm,pool-short = <3>;
+			};
+
+			bm@c8000 {
+				status = "okay";
 			};
 
 			nfc: flash@d0000 {
@@ -178,6 +192,10 @@
 			};
 		};
 
+		bm-bppi {
+			status = "okay";
+		};
+
 		pcie-controller {
 			status = "okay";
 
diff --git a/arch/arm/boot/dts/armada-388-clearfog.dts b/arch/arm/boot/dts/armada-388-clearfog.dts
index c6e180e..c60206e 100644
--- a/arch/arm/boot/dts/armada-388-clearfog.dts
+++ b/arch/arm/boot/dts/armada-388-clearfog.dts
@@ -78,6 +78,9 @@
 		internal-regs {
 			ethernet@30000 {
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
+				bm,pool-short = <1>;
 				status = "okay";
 
 				fixed-link {
@@ -88,6 +91,9 @@
 
 			ethernet@34000 {
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <3>;
+				bm,pool-short = <1>;
 				status = "okay";
 
 				fixed-link {
diff --git a/arch/arm/boot/dts/armada-388-db.dts b/arch/arm/boot/dts/armada-388-db.dts
index ff47af5..ea93ed7 100644
--- a/arch/arm/boot/dts/armada-388-db.dts
+++ b/arch/arm/boot/dts/armada-388-db.dts
@@ -66,7 +66,8 @@
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0xfff00000 0x100000
 			  MBUS_ID(0x09, 0x19) 0 0xf1100000 0x10000
-			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000>;
+			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0xf1200000 0x100000>;
 
 		internal-regs {
 			spi@10600 {
@@ -99,6 +100,9 @@
 				status = "okay";
 				phy = <&phy1>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
+				bm,pool-short = <3>;
 			};
 
 			usb@58000 {
@@ -109,6 +113,9 @@
 				status = "okay";
 				phy = <&phy0>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
+				bm,pool-short = <1>;
 			};
 
 			mdio@72004 {
@@ -129,6 +136,10 @@
 				status = "okay";
 			};
 
+			bm@c8000 {
+				status = "okay";
+			};
+
 			flash@d0000 {
 				status = "okay";
 				num-cs = <1>;
@@ -169,6 +180,10 @@
 			};
 		};
 
+		bm-bppi {
+			status = "okay";
+		};
+
 		pcie-controller {
 			status = "okay";
 			/*
diff --git a/arch/arm/boot/dts/armada-388-gp.dts b/arch/arm/boot/dts/armada-388-gp.dts
index cd31602..466b01e 100644
--- a/arch/arm/boot/dts/armada-388-gp.dts
+++ b/arch/arm/boot/dts/armada-388-gp.dts
@@ -60,7 +60,8 @@
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0xfff00000 0x100000
 			  MBUS_ID(0x09, 0x19) 0 0xf1100000 0x10000
-			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000>;
+			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0xf1200000 0x100000>;
 
 		internal-regs {
 			spi@10600 {
@@ -133,6 +134,9 @@
 				status = "okay";
 				phy = <&phy1>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
+				bm,pool-short = <3>;
 			};
 
 			/* CON4 */
@@ -152,6 +156,9 @@
 				status = "okay";
 				phy = <&phy0>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
+				bm,pool-short = <1>;
 			};
 
 
@@ -186,6 +193,10 @@
 				};
 			};
 
+			bm@c8000 {
+				status = "okay";
+			};
+
 			sata@e0000 {
 				pinctrl-names = "default";
 				pinctrl-0 = <&sata2_pins>, <&sata3_pins>;
@@ -240,6 +251,10 @@
 			};
 		};
 
+		bm-bppi {
+			status = "okay";
+		};
+
 		pcie-controller {
 			status = "okay";
 			/*
diff --git a/arch/arm/boot/dts/armada-38x-solidrun-microsom.dtsi b/arch/arm/boot/dts/armada-38x-solidrun-microsom.dtsi
index 3f792a5..8c98422 100644
--- a/arch/arm/boot/dts/armada-38x-solidrun-microsom.dtsi
+++ b/arch/arm/boot/dts/armada-38x-solidrun-microsom.dtsi
@@ -58,7 +58,8 @@
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0xfff00000 0x100000
 			  MBUS_ID(0x09, 0x19) 0 0xf1100000 0x10000
-			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000>;
+			  MBUS_ID(0x09, 0x15) 0 0xf1110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0xf1200000 0x100000>;
 
 		internal-regs {
 			ethernet@70000 {
@@ -66,6 +67,9 @@
 				pinctrl-names = "default";
 				phy = <&phy_dedicated>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
+				bm,pool-short = <1>;
 				status = "okay";
 			};
 
@@ -110,6 +114,15 @@
 				pinctrl-names = "default";
 				status = "okay";
 			};
+
+			bm@c8000 {
+				status = "okay";
+			};
 		};
+
+		bm-bppi {
+			status = "okay";
+		};
+
 	};
 };
-- 
cgit v0.10.2


From ebae1376fd9ff2cb31bebb04deaa4a4bc04372e5 Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:38:59 +0100
Subject: ARM: dts: armada-xp: add buffer manager nodes

Armada XP network controller supports hardware buffer management (BM).
Since it is now enabled in mvneta driver, appropriate nodes can be added
to armada-xp.dtsi - for the actual common BM unit (bm@c0000) and its
internal SRAM (bm-bppi), which is used for indirect access to buffer
pointer ring residing in DRAM.

Pools - ports mapping, bm-bppi entry in 'soc' node's ranges and optional
parameters are supposed to be set in board files.

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/armada-xp.dtsi b/arch/arm/boot/dts/armada-xp.dtsi
index be23196..553349c 100644
--- a/arch/arm/boot/dts/armada-xp.dtsi
+++ b/arch/arm/boot/dts/armada-xp.dtsi
@@ -253,6 +253,14 @@
 				marvell,crypto-sram-size = <0x800>;
 			};
 
+			bm: bm@c0000 {
+				compatible = "marvell,armada-380-neta-bm";
+				reg = <0xc0000 0xac>;
+				clocks = <&gateclk 13>;
+				internal-mem = <&bm_bppi>;
+				status = "disabled";
+			};
+
 			xor@f0900 {
 				compatible = "marvell,orion-xor";
 				reg = <0xF0900 0x100
@@ -291,6 +299,17 @@
 			#size-cells = <1>;
 			ranges = <0 MBUS_ID(0x09, 0x05) 0 0x800>;
 		};
+
+		bm_bppi: bm-bppi {
+			compatible = "mmio-sram";
+			reg = <MBUS_ID(0x0c, 0x04) 0 0x100000>;
+			ranges = <0 MBUS_ID(0x0c, 0x04) 0 0x100000>;
+			#address-cells = <1>;
+			#size-cells = <1>;
+			clocks = <&gateclk 13>;
+			no-memory-wc;
+			status = "disabled";
+		};
 	};
 
 	clocks {
-- 
cgit v0.10.2


From 9dd7a57e2cbf33d1f13be8e2d07875c63beb71ac Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:39:00 +0100
Subject: ARM: dts: armada-xp: enable buffer manager support on Armada XP
 boards

Since mvneta driver supports using hardware buffer management (BM), in
order to use it, board files have to be adjusted accordingly. This commit
enables BM on AXP-DB and AXP-GP in same manner - because number of ports
on those boards is the same as number of possible pools, each port is
supposed to use single pool for all kind of packets.

Moreover appropriate entry is added to 'soc' node ranges, as well as "okay"
status for 'bm' and 'bm-bppi' (internal SRAM) nodes.

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/armada-xp-db.dts b/arch/arm/boot/dts/armada-xp-db.dts
index f774101..3065730 100644
--- a/arch/arm/boot/dts/armada-xp-db.dts
+++ b/arch/arm/boot/dts/armada-xp-db.dts
@@ -77,7 +77,8 @@
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
 			  MBUS_ID(0x01, 0x2f) 0 0 0xf0000000 0x1000000
 			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0 0xf1200000 0x100000>;
 
 		devbus-bootcs {
 			status = "okay";
@@ -181,21 +182,33 @@
 				status = "okay";
 				phy = <&phy0>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
 			};
 			ethernet@74000 {
 				status = "okay";
 				phy = <&phy1>;
 				phy-mode = "rgmii-id";
+				buffer-manager = <&bm>;
+				bm,pool-long = <1>;
 			};
 			ethernet@30000 {
 				status = "okay";
 				phy = <&phy2>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
 			};
 			ethernet@34000 {
 				status = "okay";
 				phy = <&phy3>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <3>;
+			};
+
+			bm@c0000 {
+				status = "okay";
 			};
 
 			mvsdio@d4000 {
@@ -230,5 +243,9 @@
 				};
 			};
 		};
+
+		bm-bppi {
+			status = "okay";
+		};
 	};
 };
diff --git a/arch/arm/boot/dts/armada-xp-gp.dts b/arch/arm/boot/dts/armada-xp-gp.dts
index 4878d73..a1ded01 100644
--- a/arch/arm/boot/dts/armada-xp-gp.dts
+++ b/arch/arm/boot/dts/armada-xp-gp.dts
@@ -96,7 +96,8 @@
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
 			  MBUS_ID(0x01, 0x2f) 0 0 0xf0000000 0x1000000
 			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0 0xf1200000 0x100000>;
 
 		devbus-bootcs {
 			status = "okay";
@@ -196,21 +197,29 @@
 				status = "okay";
 				phy = <&phy0>;
 				phy-mode = "qsgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
 			};
 			ethernet@74000 {
 				status = "okay";
 				phy = <&phy1>;
 				phy-mode = "qsgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <1>;
 			};
 			ethernet@30000 {
 				status = "okay";
 				phy = <&phy2>;
 				phy-mode = "qsgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
 			};
 			ethernet@34000 {
 				status = "okay";
 				phy = <&phy3>;
 				phy-mode = "qsgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <3>;
 			};
 
 			/* Front-side USB slot */
@@ -235,6 +244,10 @@
 				};
 			};
 
+			bm@c0000 {
+				status = "okay";
+			};
+
 			nand@d0000 {
 				status = "okay";
 				num-cs = <1>;
@@ -243,5 +256,9 @@
 				nand-on-flash-bbt;
 			};
 		};
+
+		bm-bppi {
+			status = "okay";
+		};
 	};
 };
-- 
cgit v0.10.2


From 293fdc24fcc9532e2a10120daa1c38b057f37209 Mon Sep 17 00:00:00 2001
From: Gregory CLEMENT <gregory.clement@free-electrons.com>
Date: Mon, 14 Mar 2016 09:39:01 +0100
Subject: ARM: dts: armada-xp-openblocks-ax3-4: Add BM support

Allow Openblock AX3 using hardware buffer management with mvneta.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
index a5db177..3aa29a9 100644
--- a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
+++ b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
@@ -67,7 +67,8 @@
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
 			  MBUS_ID(0x01, 0x2f) 0 0 0xf0000000 0x8000000
 			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000
+			  MBUS_ID(0x0c, 0x04) 0 0 0xd1200000 0x100000>;
 
 		devbus-bootcs {
 			status = "okay";
@@ -176,21 +177,29 @@
 				status = "okay";
 				phy = <&phy0>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <0>;
 			};
 			ethernet@74000 {
 				status = "okay";
 				phy = <&phy1>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <1>;
 			};
 			ethernet@30000 {
 				status = "okay";
 				phy = <&phy2>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <2>;
 			};
 			ethernet@34000 {
 				status = "okay";
 				phy = <&phy3>;
 				phy-mode = "sgmii";
+				buffer-manager = <&bm>;
+				bm,pool-long = <3>;
 			};
 			i2c@11000 {
 				status = "okay";
@@ -219,6 +228,14 @@
 			usb@51000 {
 				status = "okay";
 			};
+
+			bm@c0000 {
+				status = "okay";
+			};
+		};
+
+		bm-bppi {
+			status = "okay";
 		};
 	};
 };
-- 
cgit v0.10.2


From f2900acea8018c4525ddaa86c7f7cd8afd3f0cc4 Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:39:02 +0100
Subject: bus: mvebu-mbus: provide api for obtaining IO and DRAM window
 information

This commit enables finding appropriate mbus window and obtaining its
target id and attribute for given physical address in two separate
routines, both for IO and DRAM windows. This functionality
is needed for Armada XP/38x Network Controller's Buffer Manager and
PnC configuration.

[gregory.clement@free-electrons.com: Fix size test for
mvebu_mbus_get_dram_win_info]

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
[DRAM window information reference in LKv3.10]
Signed-off-by: Evan Wang <xswang@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/bus/mvebu-mbus.c b/drivers/bus/mvebu-mbus.c
index c43c3d2..c2e5286 100644
--- a/drivers/bus/mvebu-mbus.c
+++ b/drivers/bus/mvebu-mbus.c
@@ -948,6 +948,58 @@ void mvebu_mbus_get_pcie_io_aperture(struct resource *res)
 	*res = mbus_state.pcie_io_aperture;
 }
 
+int mvebu_mbus_get_dram_win_info(phys_addr_t phyaddr, u8 *target, u8 *attr)
+{
+	const struct mbus_dram_target_info *dram;
+	int i;
+
+	/* Get dram info */
+	dram = mv_mbus_dram_info();
+	if (!dram) {
+		pr_err("missing DRAM information\n");
+		return -ENODEV;
+	}
+
+	/* Try to find matching DRAM window for phyaddr */
+	for (i = 0; i < dram->num_cs; i++) {
+		const struct mbus_dram_window *cs = dram->cs + i;
+
+		if (cs->base <= phyaddr &&
+			phyaddr <= (cs->base + cs->size - 1)) {
+			*target = dram->mbus_dram_target_id;
+			*attr = cs->mbus_attr;
+			return 0;
+		}
+	}
+
+	pr_err("invalid dram address 0x%x\n", phyaddr);
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(mvebu_mbus_get_dram_win_info);
+
+int mvebu_mbus_get_io_win_info(phys_addr_t phyaddr, u32 *size, u8 *target,
+			       u8 *attr)
+{
+	int win;
+
+	for (win = 0; win < mbus_state.soc->num_wins; win++) {
+		u64 wbase;
+		int enabled;
+
+		mvebu_mbus_read_window(&mbus_state, win, &enabled, &wbase,
+				       size, target, attr, NULL);
+
+		if (!enabled)
+			continue;
+
+		if (wbase <= phyaddr && phyaddr <= wbase + *size)
+			return win;
+	}
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(mvebu_mbus_get_io_win_info);
+
 static __init int mvebu_mbus_debugfs_init(void)
 {
 	struct mvebu_mbus_state *s = &mbus_state;
diff --git a/include/linux/mbus.h b/include/linux/mbus.h
index 1f7bc63..ea34a86 100644
--- a/include/linux/mbus.h
+++ b/include/linux/mbus.h
@@ -69,6 +69,9 @@ static inline const struct mbus_dram_target_info *mv_mbus_dram_info_nooverlap(vo
 int mvebu_mbus_save_cpu_target(u32 *store_addr);
 void mvebu_mbus_get_pcie_mem_aperture(struct resource *res);
 void mvebu_mbus_get_pcie_io_aperture(struct resource *res);
+int mvebu_mbus_get_dram_win_info(phys_addr_t phyaddr, u8 *target, u8 *attr);
+int mvebu_mbus_get_io_win_info(phys_addr_t phyaddr, u32 *size, u8 *target,
+			       u8 *attr);
 int mvebu_mbus_add_window_remap_by_id(unsigned int target,
 				      unsigned int attribute,
 				      phys_addr_t base, size_t size,
-- 
cgit v0.10.2


From dc35a10f68d3781c2345d60b22103785985ca849 Mon Sep 17 00:00:00 2001
From: Marcin Wojtas <mw@semihalf.com>
Date: Mon, 14 Mar 2016 09:39:03 +0100
Subject: net: mvneta: bm: add support for hardware buffer management

Buffer manager (BM) is a dedicated hardware unit that can be used by all
ethernet ports of Armada XP and 38x SoC's. It allows to offload CPU on RX
path by sparing DRAM access on refilling buffer pool, hardware-based
filling of descriptor ring data and better memory utilization due to HW
arbitration for using 'short' pools for small packets.

Tests performed with A388 SoC working as a network bridge between two
packet generators showed increase of maximum processed 64B packets by
~20k (~555k packets with BM enabled vs ~535 packets without BM). Also
when pushing 1500B-packets with a line rate achieved, CPU load decreased
from around 25% without BM to 20% with BM.

BM comprise up to 4 buffer pointers' (BP) rings kept in DRAM, which
are called external BP pools - BPPE. Allocating and releasing buffer
pointers (BP) to/from BPPE is performed indirectly by write/read access
to a dedicated internal SRAM, where internal BP pools (BPPI) are placed.
BM hardware controls status of BPPE automatically, as well as assigning
proper buffers to RX descriptors. For more details please refer to
Functional Specification of Armada XP or 38x SoC.

In order to enable support for a separate hardware block, common for all
ports, a new driver has to be implemented ('mvneta_bm'). It provides
initialization sequence of address space, clocks, registers, SRAM,
empty pools' structures and also obtaining optional configuration
from DT (please refer to device tree binding documentation). mvneta_bm
exposes also a necessary API to mvneta driver, as well as a dedicated
structure with BM information (bm_priv), whose presence is used as a
flag notifying of BM usage by port. It has to be ensured that mvneta_bm
probe is executed prior to the ones in ports' driver. In case BM is not
used or its probe fails, mvneta falls back to use software buffer
management.

A sequence executed in mvneta_probe function is modified in order to have
an access to needed resources before possible port's BM initialization is
done. According to port-pools mapping provided by DT appropriate registers
are configured and the buffer pools are filled. RX path is modified
accordingly. Becaues the hardware allows a wide variety of configuration
options, following assumptions are made:
* using BM mechanisms can be selectively disabled/enabled basing
  on DT configuration among the ports
* 'long' pool's single buffer size is tied to port's MTU
* using 'long' pool by port is obligatory and it cannot be shared
* using 'short' pool for smaller packets is optional
* one 'short' pool can be shared among all ports

This commit enables hardware buffer management operation cooperating with
existing mvneta driver. New device tree binding documentation is added and
the one of mvneta is updated accordingly.

[gregory.clement@free-electrons.com: removed the suspend/resume part]

Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt b/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt
index d0cb869..73be897 100644
--- a/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt
+++ b/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt
@@ -18,15 +18,30 @@ Optional properties:
   "core" for core clock and "bus" for the optional bus clock.
 
 
+Optional properties (valid only for Armada XP/38x):
+
+- buffer-manager: a phandle to a buffer manager node. Please refer to
+  Documentation/devicetree/bindings/net/marvell-neta-bm.txt
+- bm,pool-long: ID of a pool, that will accept all packets of a size
+  higher than 'short' pool's threshold (if set) and up to MTU value.
+  Obligatory, when the port is supposed to use hardware
+  buffer management.
+- bm,pool-short: ID of a pool, that will be used for accepting
+  packets of a size lower than given threshold. If not set, the port
+  will use a single 'long' pool for all packets, as defined above.
+
 Example:
 
-ethernet@d0070000 {
+ethernet@70000 {
 	compatible = "marvell,armada-370-neta";
-	reg = <0xd0070000 0x2500>;
+	reg = <0x70000 0x2500>;
 	interrupts = <8>;
 	clocks = <&gate_clk 4>;
 	tx-csum-limit = <9800>
 	status = "okay";
 	phy = <&phy0>;
 	phy-mode = "rgmii-id";
+	buffer-manager = <&bm>;
+	bm,pool-long = <0>;
+	bm,pool-short = <1>;
 };
diff --git a/Documentation/devicetree/bindings/net/marvell-neta-bm.txt b/Documentation/devicetree/bindings/net/marvell-neta-bm.txt
new file mode 100644
index 0000000..c1b1d7c3
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/marvell-neta-bm.txt
@@ -0,0 +1,49 @@
+* Marvell Armada 380/XP Buffer Manager driver (BM)
+
+Required properties:
+
+- compatible: should be "marvell,armada-380-neta-bm".
+- reg: address and length of the register set for the device.
+- clocks: a pointer to the reference clock for this device.
+- internal-mem: a phandle to BM internal SRAM definition.
+
+Optional properties (port):
+
+- pool<0 : 3>,capacity: size of external buffer pointers' ring maintained
+  in DRAM. Can be set for each pool (id 0 : 3) separately. The value has
+  to be chosen between 128 and 16352 and it also has to be aligned to 32.
+  Otherwise the driver would adjust a given number or choose default if
+  not set.
+- pool<0 : 3>,pkt-size: maximum size of a packet accepted by a given buffer
+  pointers' pool (id 0 : 3). It will be taken into consideration only when pool
+  type is 'short'. For 'long' ones it would be overridden by port's MTU.
+  If not set a driver will choose a default value.
+
+In order to see how to hook the BM to a given ethernet port, please
+refer to Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt.
+
+Example:
+
+- main node:
+
+bm: bm@c8000 {
+	compatible = "marvell,armada-380-neta-bm";
+	reg = <0xc8000 0xac>;
+	clocks = <&gateclk 13>;
+	internal-mem = <&bm_bppi>;
+	status = "okay";
+	pool2,capacity = <4096>;
+	pool1,pkt-size = <512>;
+};
+
+- internal SRAM node:
+
+bm_bppi: bm-bppi {
+	compatible = "mmio-sram";
+	reg = <MBUS_ID(0x0c, 0x04) 0 0x100000>;
+	ranges = <0 MBUS_ID(0x0c, 0x04) 0 0x100000>;
+	#address-cells = <1>;
+	#size-cells = <1>;
+	clocks = <&gateclk 13>;
+	status = "okay";
+};
diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index a1c862b..ac6605c 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -40,6 +40,19 @@ config MVMDIO
 
 	  This driver is used by the MV643XX_ETH and MVNETA drivers.
 
+config MVNETA_BM
+	tristate "Marvell Armada 38x/XP network interface BM support"
+	depends on MVNETA
+	---help---
+	  This driver supports auxiliary block of the network
+	  interface units in the Marvell ARMADA XP and ARMADA 38x SoC
+	  family, which is called buffer manager.
+
+	  This driver, when enabled, strictly cooperates with mvneta
+	  driver and is common for all network ports of the devices,
+	  even for Armada 370 SoC, which doesn't support hardware
+	  buffer management.
+
 config MVNETA
 	tristate "Marvell Armada 370/38x/XP network interface support"
 	depends on PLAT_ORION
diff --git a/drivers/net/ethernet/marvell/Makefile b/drivers/net/ethernet/marvell/Makefile
index f6425bd..ff1bffa 100644
--- a/drivers/net/ethernet/marvell/Makefile
+++ b/drivers/net/ethernet/marvell/Makefile
@@ -4,6 +4,7 @@
 
 obj-$(CONFIG_MVMDIO) += mvmdio.o
 obj-$(CONFIG_MV643XX_ETH) += mv643xx_eth.o
+obj-$(CONFIG_MVNETA_BM) += mvneta_bm.o
 obj-$(CONFIG_MVNETA) += mvneta.o
 obj-$(CONFIG_MVPP2) += mvpp2.o
 obj-$(CONFIG_PXA168_ETH) += pxa168_eth.o
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index b0ae69f..2847c0c 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -30,6 +30,7 @@
 #include <linux/phy.h>
 #include <linux/platform_device.h>
 #include <linux/skbuff.h>
+#include "mvneta_bm.h"
 #include <net/ip.h>
 #include <net/ipv6.h>
 #include <net/tso.h>
@@ -37,6 +38,10 @@
 /* Registers */
 #define MVNETA_RXQ_CONFIG_REG(q)                (0x1400 + ((q) << 2))
 #define      MVNETA_RXQ_HW_BUF_ALLOC            BIT(0)
+#define      MVNETA_RXQ_SHORT_POOL_ID_SHIFT	4
+#define      MVNETA_RXQ_SHORT_POOL_ID_MASK	0x30
+#define      MVNETA_RXQ_LONG_POOL_ID_SHIFT	6
+#define      MVNETA_RXQ_LONG_POOL_ID_MASK	0xc0
 #define      MVNETA_RXQ_PKT_OFFSET_ALL_MASK     (0xf    << 8)
 #define      MVNETA_RXQ_PKT_OFFSET_MASK(offs)   ((offs) << 8)
 #define MVNETA_RXQ_THRESHOLD_REG(q)             (0x14c0 + ((q) << 2))
@@ -50,6 +55,9 @@
 #define MVNETA_RXQ_STATUS_UPDATE_REG(q)         (0x1500 + ((q) << 2))
 #define      MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT  16
 #define      MVNETA_RXQ_ADD_NON_OCCUPIED_MAX    255
+#define MVNETA_PORT_POOL_BUFFER_SZ_REG(pool)	(0x1700 + ((pool) << 2))
+#define      MVNETA_PORT_POOL_BUFFER_SZ_SHIFT	3
+#define      MVNETA_PORT_POOL_BUFFER_SZ_MASK	0xfff8
 #define MVNETA_PORT_RX_RESET                    0x1cc0
 #define      MVNETA_PORT_RX_DMA_RESET           BIT(0)
 #define MVNETA_PHY_ADDR                         0x2000
@@ -107,6 +115,7 @@
 #define MVNETA_GMAC_CLOCK_DIVIDER                0x24f4
 #define      MVNETA_GMAC_1MS_CLOCK_ENABLE        BIT(31)
 #define MVNETA_ACC_MODE                          0x2500
+#define MVNETA_BM_ADDRESS                        0x2504
 #define MVNETA_CPU_MAP(cpu)                      (0x2540 + ((cpu) << 2))
 #define      MVNETA_CPU_RXQ_ACCESS_ALL_MASK      0x000000ff
 #define      MVNETA_CPU_TXQ_ACCESS_ALL_MASK      0x0000ff00
@@ -253,7 +262,10 @@
 #define MVNETA_CPU_D_CACHE_LINE_SIZE    32
 #define MVNETA_TX_CSUM_DEF_SIZE		1600
 #define MVNETA_TX_CSUM_MAX_SIZE		9800
-#define MVNETA_ACC_MODE_EXT		1
+#define MVNETA_ACC_MODE_EXT1		1
+#define MVNETA_ACC_MODE_EXT2		2
+
+#define MVNETA_MAX_DECODE_WIN		6
 
 /* Timeout constants */
 #define MVNETA_TX_DISABLE_TIMEOUT_MSEC	1000
@@ -293,7 +305,8 @@
 	((addr >= txq->tso_hdrs_phys) && \
 	 (addr < txq->tso_hdrs_phys + txq->size * TSO_HEADER_SIZE))
 
-#define MVNETA_RX_BUF_SIZE(pkt_size)   ((pkt_size) + NET_SKB_PAD)
+#define MVNETA_RX_GET_BM_POOL_ID(rxd) \
+	(((rxd)->status & MVNETA_RXD_BM_POOL_MASK) >> MVNETA_RXD_BM_POOL_SHIFT)
 
 struct mvneta_statistic {
 	unsigned short offset;
@@ -359,6 +372,7 @@ struct mvneta_pcpu_port {
 };
 
 struct mvneta_port {
+	u8 id;
 	struct mvneta_pcpu_port __percpu	*ports;
 	struct mvneta_pcpu_stats __percpu	*stats;
 
@@ -394,6 +408,11 @@ struct mvneta_port {
 	unsigned int tx_csum_limit;
 	unsigned int use_inband_status:1;
 
+	struct mvneta_bm *bm_priv;
+	struct mvneta_bm_pool *pool_long;
+	struct mvneta_bm_pool *pool_short;
+	int bm_win_id;
+
 	u64 ethtool_stats[ARRAY_SIZE(mvneta_statistics)];
 
 	u32 indir[MVNETA_RSS_LU_TABLE_SIZE];
@@ -419,6 +438,8 @@ struct mvneta_port {
 #define MVNETA_TX_L4_CSUM_NOT	BIT(31)
 
 #define MVNETA_RXD_ERR_CRC		0x0
+#define MVNETA_RXD_BM_POOL_SHIFT	13
+#define MVNETA_RXD_BM_POOL_MASK		(BIT(13) | BIT(14))
 #define MVNETA_RXD_ERR_SUMMARY		BIT(16)
 #define MVNETA_RXD_ERR_OVERRUN		BIT(17)
 #define MVNETA_RXD_ERR_LEN		BIT(18)
@@ -563,6 +584,9 @@ static int rxq_def;
 
 static int rx_copybreak __read_mostly = 256;
 
+/* HW BM need that each port be identify by a unique ID */
+static int global_port_id;
+
 #define MVNETA_DRIVER_NAME "mvneta"
 #define MVNETA_DRIVER_VERSION "1.0"
 
@@ -829,6 +853,214 @@ static void mvneta_rxq_bm_disable(struct mvneta_port *pp,
 	mvreg_write(pp, MVNETA_RXQ_CONFIG_REG(rxq->id), val);
 }
 
+/* Enable buffer management (BM) */
+static void mvneta_rxq_bm_enable(struct mvneta_port *pp,
+				 struct mvneta_rx_queue *rxq)
+{
+	u32 val;
+
+	val = mvreg_read(pp, MVNETA_RXQ_CONFIG_REG(rxq->id));
+	val |= MVNETA_RXQ_HW_BUF_ALLOC;
+	mvreg_write(pp, MVNETA_RXQ_CONFIG_REG(rxq->id), val);
+}
+
+/* Notify HW about port's assignment of pool for bigger packets */
+static void mvneta_rxq_long_pool_set(struct mvneta_port *pp,
+				     struct mvneta_rx_queue *rxq)
+{
+	u32 val;
+
+	val = mvreg_read(pp, MVNETA_RXQ_CONFIG_REG(rxq->id));
+	val &= ~MVNETA_RXQ_LONG_POOL_ID_MASK;
+	val |= (pp->pool_long->id << MVNETA_RXQ_LONG_POOL_ID_SHIFT);
+
+	mvreg_write(pp, MVNETA_RXQ_CONFIG_REG(rxq->id), val);
+}
+
+/* Notify HW about port's assignment of pool for smaller packets */
+static void mvneta_rxq_short_pool_set(struct mvneta_port *pp,
+				      struct mvneta_rx_queue *rxq)
+{
+	u32 val;
+
+	val = mvreg_read(pp, MVNETA_RXQ_CONFIG_REG(rxq->id));
+	val &= ~MVNETA_RXQ_SHORT_POOL_ID_MASK;
+	val |= (pp->pool_short->id << MVNETA_RXQ_SHORT_POOL_ID_SHIFT);
+
+	mvreg_write(pp, MVNETA_RXQ_CONFIG_REG(rxq->id), val);
+}
+
+/* Set port's receive buffer size for assigned BM pool */
+static inline void mvneta_bm_pool_bufsize_set(struct mvneta_port *pp,
+					      int buf_size,
+					      u8 pool_id)
+{
+	u32 val;
+
+	if (!IS_ALIGNED(buf_size, 8)) {
+		dev_warn(pp->dev->dev.parent,
+			 "illegal buf_size value %d, round to %d\n",
+			 buf_size, ALIGN(buf_size, 8));
+		buf_size = ALIGN(buf_size, 8);
+	}
+
+	val = mvreg_read(pp, MVNETA_PORT_POOL_BUFFER_SZ_REG(pool_id));
+	val |= buf_size & MVNETA_PORT_POOL_BUFFER_SZ_MASK;
+	mvreg_write(pp, MVNETA_PORT_POOL_BUFFER_SZ_REG(pool_id), val);
+}
+
+/* Configure MBUS window in order to enable access BM internal SRAM */
+static int mvneta_mbus_io_win_set(struct mvneta_port *pp, u32 base, u32 wsize,
+				  u8 target, u8 attr)
+{
+	u32 win_enable, win_protect;
+	int i;
+
+	win_enable = mvreg_read(pp, MVNETA_BASE_ADDR_ENABLE);
+
+	if (pp->bm_win_id < 0) {
+		/* Find first not occupied window */
+		for (i = 0; i < MVNETA_MAX_DECODE_WIN; i++) {
+			if (win_enable & (1 << i)) {
+				pp->bm_win_id = i;
+				break;
+			}
+		}
+		if (i == MVNETA_MAX_DECODE_WIN)
+			return -ENOMEM;
+	} else {
+		i = pp->bm_win_id;
+	}
+
+	mvreg_write(pp, MVNETA_WIN_BASE(i), 0);
+	mvreg_write(pp, MVNETA_WIN_SIZE(i), 0);
+
+	if (i < 4)
+		mvreg_write(pp, MVNETA_WIN_REMAP(i), 0);
+
+	mvreg_write(pp, MVNETA_WIN_BASE(i), (base & 0xffff0000) |
+		    (attr << 8) | target);
+
+	mvreg_write(pp, MVNETA_WIN_SIZE(i), (wsize - 1) & 0xffff0000);
+
+	win_protect = mvreg_read(pp, MVNETA_ACCESS_PROTECT_ENABLE);
+	win_protect |= 3 << (2 * i);
+	mvreg_write(pp, MVNETA_ACCESS_PROTECT_ENABLE, win_protect);
+
+	win_enable &= ~(1 << i);
+	mvreg_write(pp, MVNETA_BASE_ADDR_ENABLE, win_enable);
+
+	return 0;
+}
+
+/* Assign and initialize pools for port. In case of fail
+ * buffer manager will remain disabled for current port.
+ */
+static int mvneta_bm_port_init(struct platform_device *pdev,
+			       struct mvneta_port *pp)
+{
+	struct device_node *dn = pdev->dev.of_node;
+	u32 long_pool_id, short_pool_id, wsize;
+	u8 target, attr;
+	int err;
+
+	/* Get BM window information */
+	err = mvebu_mbus_get_io_win_info(pp->bm_priv->bppi_phys_addr, &wsize,
+					 &target, &attr);
+	if (err < 0)
+		return err;
+
+	pp->bm_win_id = -1;
+
+	/* Open NETA -> BM window */
+	err = mvneta_mbus_io_win_set(pp, pp->bm_priv->bppi_phys_addr, wsize,
+				     target, attr);
+	if (err < 0) {
+		netdev_info(pp->dev, "fail to configure mbus window to BM\n");
+		return err;
+	}
+
+	if (of_property_read_u32(dn, "bm,pool-long", &long_pool_id)) {
+		netdev_info(pp->dev, "missing long pool id\n");
+		return -EINVAL;
+	}
+
+	/* Create port's long pool depending on mtu */
+	pp->pool_long = mvneta_bm_pool_use(pp->bm_priv, long_pool_id,
+					   MVNETA_BM_LONG, pp->id,
+					   MVNETA_RX_PKT_SIZE(pp->dev->mtu));
+	if (!pp->pool_long) {
+		netdev_info(pp->dev, "fail to obtain long pool for port\n");
+		return -ENOMEM;
+	}
+
+	pp->pool_long->port_map |= 1 << pp->id;
+
+	mvneta_bm_pool_bufsize_set(pp, pp->pool_long->buf_size,
+				   pp->pool_long->id);
+
+	/* If short pool id is not defined, assume using single pool */
+	if (of_property_read_u32(dn, "bm,pool-short", &short_pool_id))
+		short_pool_id = long_pool_id;
+
+	/* Create port's short pool */
+	pp->pool_short = mvneta_bm_pool_use(pp->bm_priv, short_pool_id,
+					    MVNETA_BM_SHORT, pp->id,
+					    MVNETA_BM_SHORT_PKT_SIZE);
+	if (!pp->pool_short) {
+		netdev_info(pp->dev, "fail to obtain short pool for port\n");
+		mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_long, 1 << pp->id);
+		return -ENOMEM;
+	}
+
+	if (short_pool_id != long_pool_id) {
+		pp->pool_short->port_map |= 1 << pp->id;
+		mvneta_bm_pool_bufsize_set(pp, pp->pool_short->buf_size,
+					   pp->pool_short->id);
+	}
+
+	return 0;
+}
+
+/* Update settings of a pool for bigger packets */
+static void mvneta_bm_update_mtu(struct mvneta_port *pp, int mtu)
+{
+	struct mvneta_bm_pool *bm_pool = pp->pool_long;
+	int num;
+
+	/* Release all buffers from long pool */
+	mvneta_bm_bufs_free(pp->bm_priv, bm_pool, 1 << pp->id);
+	if (bm_pool->buf_num) {
+		WARN(1, "cannot free all buffers in pool %d\n",
+		     bm_pool->id);
+		goto bm_mtu_err;
+	}
+
+	bm_pool->pkt_size = MVNETA_RX_PKT_SIZE(mtu);
+	bm_pool->buf_size = MVNETA_RX_BUF_SIZE(bm_pool->pkt_size);
+	bm_pool->frag_size = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
+			  SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(bm_pool->pkt_size));
+
+	/* Fill entire long pool */
+	num = mvneta_bm_bufs_add(pp->bm_priv, bm_pool, bm_pool->size);
+	if (num != bm_pool->size) {
+		WARN(1, "pool %d: %d of %d allocated\n",
+		     bm_pool->id, num, bm_pool->size);
+		goto bm_mtu_err;
+	}
+	mvneta_bm_pool_bufsize_set(pp, bm_pool->buf_size, bm_pool->id);
+
+	return;
+
+bm_mtu_err:
+	mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_long, 1 << pp->id);
+	mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_short, 1 << pp->id);
+
+	pp->bm_priv = NULL;
+	mvreg_write(pp, MVNETA_ACC_MODE, MVNETA_ACC_MODE_EXT1);
+	netdev_info(pp->dev, "fail to update MTU, fall back to software BM\n");
+}
+
 /* Start the Ethernet port RX and TX activity */
 static void mvneta_port_up(struct mvneta_port *pp)
 {
@@ -1149,9 +1381,17 @@ static void mvneta_defaults_set(struct mvneta_port *pp)
 	mvreg_write(pp, MVNETA_PORT_RX_RESET, 0);
 
 	/* Set Port Acceleration Mode */
-	val = MVNETA_ACC_MODE_EXT;
+	if (pp->bm_priv)
+		/* HW buffer management + legacy parser */
+		val = MVNETA_ACC_MODE_EXT2;
+	else
+		/* SW buffer management + legacy parser */
+		val = MVNETA_ACC_MODE_EXT1;
 	mvreg_write(pp, MVNETA_ACC_MODE, val);
 
+	if (pp->bm_priv)
+		mvreg_write(pp, MVNETA_BM_ADDRESS, pp->bm_priv->bppi_phys_addr);
+
 	/* Update val of portCfg register accordingly with all RxQueue types */
 	val = MVNETA_PORT_CONFIG_DEFL_VALUE(pp->rxq_def);
 	mvreg_write(pp, MVNETA_PORT_CONFIG, val);
@@ -1518,23 +1758,25 @@ static void mvneta_txq_done(struct mvneta_port *pp,
 	}
 }
 
-static void *mvneta_frag_alloc(const struct mvneta_port *pp)
+void *mvneta_frag_alloc(unsigned int frag_size)
 {
-	if (likely(pp->frag_size <= PAGE_SIZE))
-		return netdev_alloc_frag(pp->frag_size);
+	if (likely(frag_size <= PAGE_SIZE))
+		return netdev_alloc_frag(frag_size);
 	else
-		return kmalloc(pp->frag_size, GFP_ATOMIC);
+		return kmalloc(frag_size, GFP_ATOMIC);
 }
+EXPORT_SYMBOL_GPL(mvneta_frag_alloc);
 
-static void mvneta_frag_free(const struct mvneta_port *pp, void *data)
+void mvneta_frag_free(unsigned int frag_size, void *data)
 {
-	if (likely(pp->frag_size <= PAGE_SIZE))
+	if (likely(frag_size <= PAGE_SIZE))
 		skb_free_frag(data);
 	else
 		kfree(data);
 }
+EXPORT_SYMBOL_GPL(mvneta_frag_free);
 
-/* Refill processing */
+/* Refill processing for SW buffer management */
 static int mvneta_rx_refill(struct mvneta_port *pp,
 			    struct mvneta_rx_desc *rx_desc)
 
@@ -1542,7 +1784,7 @@ static int mvneta_rx_refill(struct mvneta_port *pp,
 	dma_addr_t phys_addr;
 	void *data;
 
-	data = mvneta_frag_alloc(pp);
+	data = mvneta_frag_alloc(pp->frag_size);
 	if (!data)
 		return -ENOMEM;
 
@@ -1550,7 +1792,7 @@ static int mvneta_rx_refill(struct mvneta_port *pp,
 				   MVNETA_RX_BUF_SIZE(pp->pkt_size),
 				   DMA_FROM_DEVICE);
 	if (unlikely(dma_mapping_error(pp->dev->dev.parent, phys_addr))) {
-		mvneta_frag_free(pp, data);
+		mvneta_frag_free(pp->frag_size, data);
 		return -ENOMEM;
 	}
 
@@ -1596,22 +1838,156 @@ static void mvneta_rxq_drop_pkts(struct mvneta_port *pp,
 	int rx_done, i;
 
 	rx_done = mvneta_rxq_busy_desc_num_get(pp, rxq);
+	if (rx_done)
+		mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_done);
+
+	if (pp->bm_priv) {
+		for (i = 0; i < rx_done; i++) {
+			struct mvneta_rx_desc *rx_desc =
+						  mvneta_rxq_next_desc_get(rxq);
+			u8 pool_id = MVNETA_RX_GET_BM_POOL_ID(rx_desc);
+			struct mvneta_bm_pool *bm_pool;
+
+			bm_pool = &pp->bm_priv->bm_pools[pool_id];
+			/* Return dropped buffer to the pool */
+			mvneta_bm_pool_put_bp(pp->bm_priv, bm_pool,
+					      rx_desc->buf_phys_addr);
+		}
+		return;
+	}
+
 	for (i = 0; i < rxq->size; i++) {
 		struct mvneta_rx_desc *rx_desc = rxq->descs + i;
 		void *data = (void *)rx_desc->buf_cookie;
 
 		dma_unmap_single(pp->dev->dev.parent, rx_desc->buf_phys_addr,
 				 MVNETA_RX_BUF_SIZE(pp->pkt_size), DMA_FROM_DEVICE);
-		mvneta_frag_free(pp, data);
+		mvneta_frag_free(pp->frag_size, data);
 	}
+}
 
-	if (rx_done)
-		mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_done);
+/* Main rx processing when using software buffer management */
+static int mvneta_rx_swbm(struct mvneta_port *pp, int rx_todo,
+			  struct mvneta_rx_queue *rxq)
+{
+	struct mvneta_pcpu_port *port = this_cpu_ptr(pp->ports);
+	struct net_device *dev = pp->dev;
+	int rx_done;
+	u32 rcvd_pkts = 0;
+	u32 rcvd_bytes = 0;
+
+	/* Get number of received packets */
+	rx_done = mvneta_rxq_busy_desc_num_get(pp, rxq);
+
+	if (rx_todo > rx_done)
+		rx_todo = rx_done;
+
+	rx_done = 0;
+
+	/* Fairness NAPI loop */
+	while (rx_done < rx_todo) {
+		struct mvneta_rx_desc *rx_desc = mvneta_rxq_next_desc_get(rxq);
+		struct sk_buff *skb;
+		unsigned char *data;
+		dma_addr_t phys_addr;
+		u32 rx_status, frag_size;
+		int rx_bytes, err;
+
+		rx_done++;
+		rx_status = rx_desc->status;
+		rx_bytes = rx_desc->data_size - (ETH_FCS_LEN + MVNETA_MH_SIZE);
+		data = (unsigned char *)rx_desc->buf_cookie;
+		phys_addr = rx_desc->buf_phys_addr;
+
+		if (!mvneta_rxq_desc_is_first_last(rx_status) ||
+		    (rx_status & MVNETA_RXD_ERR_SUMMARY)) {
+err_drop_frame:
+			dev->stats.rx_errors++;
+			mvneta_rx_error(pp, rx_desc);
+			/* leave the descriptor untouched */
+			continue;
+		}
+
+		if (rx_bytes <= rx_copybreak) {
+		/* better copy a small frame and not unmap the DMA region */
+			skb = netdev_alloc_skb_ip_align(dev, rx_bytes);
+			if (unlikely(!skb))
+				goto err_drop_frame;
+
+			dma_sync_single_range_for_cpu(dev->dev.parent,
+						      rx_desc->buf_phys_addr,
+						      MVNETA_MH_SIZE + NET_SKB_PAD,
+						      rx_bytes,
+						      DMA_FROM_DEVICE);
+			memcpy(skb_put(skb, rx_bytes),
+			       data + MVNETA_MH_SIZE + NET_SKB_PAD,
+			       rx_bytes);
+
+			skb->protocol = eth_type_trans(skb, dev);
+			mvneta_rx_csum(pp, rx_status, skb);
+			napi_gro_receive(&port->napi, skb);
+
+			rcvd_pkts++;
+			rcvd_bytes += rx_bytes;
+
+			/* leave the descriptor and buffer untouched */
+			continue;
+		}
+
+		/* Refill processing */
+		err = mvneta_rx_refill(pp, rx_desc);
+		if (err) {
+			netdev_err(dev, "Linux processing - Can't refill\n");
+			rxq->missed++;
+			goto err_drop_frame;
+		}
+
+		frag_size = pp->frag_size;
+
+		skb = build_skb(data, frag_size > PAGE_SIZE ? 0 : frag_size);
+
+		/* After refill old buffer has to be unmapped regardless
+		 * the skb is successfully built or not.
+		 */
+		dma_unmap_single(dev->dev.parent, phys_addr,
+				 MVNETA_RX_BUF_SIZE(pp->pkt_size),
+				 DMA_FROM_DEVICE);
+
+		if (!skb)
+			goto err_drop_frame;
+
+		rcvd_pkts++;
+		rcvd_bytes += rx_bytes;
+
+		/* Linux processing */
+		skb_reserve(skb, MVNETA_MH_SIZE + NET_SKB_PAD);
+		skb_put(skb, rx_bytes);
+
+		skb->protocol = eth_type_trans(skb, dev);
+
+		mvneta_rx_csum(pp, rx_status, skb);
+
+		napi_gro_receive(&port->napi, skb);
+	}
+
+	if (rcvd_pkts) {
+		struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats);
+
+		u64_stats_update_begin(&stats->syncp);
+		stats->rx_packets += rcvd_pkts;
+		stats->rx_bytes   += rcvd_bytes;
+		u64_stats_update_end(&stats->syncp);
+	}
+
+	/* Update rxq management counters */
+	mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_done);
+
+	return rx_done;
 }
 
-/* Main rx processing */
-static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
-		     struct mvneta_rx_queue *rxq)
+/* Main rx processing when using hardware buffer management */
+static int mvneta_rx_hwbm(struct mvneta_port *pp, int rx_todo,
+			  struct mvneta_rx_queue *rxq)
 {
 	struct mvneta_pcpu_port *port = this_cpu_ptr(pp->ports);
 	struct net_device *dev = pp->dev;
@@ -1630,21 +2006,29 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
 	/* Fairness NAPI loop */
 	while (rx_done < rx_todo) {
 		struct mvneta_rx_desc *rx_desc = mvneta_rxq_next_desc_get(rxq);
+		struct mvneta_bm_pool *bm_pool = NULL;
 		struct sk_buff *skb;
 		unsigned char *data;
 		dma_addr_t phys_addr;
-		u32 rx_status;
+		u32 rx_status, frag_size;
 		int rx_bytes, err;
+		u8 pool_id;
 
 		rx_done++;
 		rx_status = rx_desc->status;
 		rx_bytes = rx_desc->data_size - (ETH_FCS_LEN + MVNETA_MH_SIZE);
 		data = (unsigned char *)rx_desc->buf_cookie;
 		phys_addr = rx_desc->buf_phys_addr;
+		pool_id = MVNETA_RX_GET_BM_POOL_ID(rx_desc);
+		bm_pool = &pp->bm_priv->bm_pools[pool_id];
 
 		if (!mvneta_rxq_desc_is_first_last(rx_status) ||
 		    (rx_status & MVNETA_RXD_ERR_SUMMARY)) {
-		err_drop_frame:
+err_drop_frame_ret_pool:
+			/* Return the buffer to the pool */
+			mvneta_bm_pool_put_bp(pp->bm_priv, bm_pool,
+					      rx_desc->buf_phys_addr);
+err_drop_frame:
 			dev->stats.rx_errors++;
 			mvneta_rx_error(pp, rx_desc);
 			/* leave the descriptor untouched */
@@ -1655,7 +2039,7 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
 			/* better copy a small frame and not unmap the DMA region */
 			skb = netdev_alloc_skb_ip_align(dev, rx_bytes);
 			if (unlikely(!skb))
-				goto err_drop_frame;
+				goto err_drop_frame_ret_pool;
 
 			dma_sync_single_range_for_cpu(dev->dev.parent,
 			                              rx_desc->buf_phys_addr,
@@ -1673,26 +2057,31 @@ static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
 			rcvd_pkts++;
 			rcvd_bytes += rx_bytes;
 
+			/* Return the buffer to the pool */
+			mvneta_bm_pool_put_bp(pp->bm_priv, bm_pool,
+					      rx_desc->buf_phys_addr);
+
 			/* leave the descriptor and buffer untouched */
 			continue;
 		}
 
 		/* Refill processing */
-		err = mvneta_rx_refill(pp, rx_desc);
+		err = mvneta_bm_pool_refill(pp->bm_priv, bm_pool);
 		if (err) {
 			netdev_err(dev, "Linux processing - Can't refill\n");
 			rxq->missed++;
-			goto err_drop_frame;
+			goto err_drop_frame_ret_pool;
 		}
 
-		skb = build_skb(data, pp->frag_size > PAGE_SIZE ? 0 : pp->frag_size);
+		frag_size = bm_pool->frag_size;
+
+		skb = build_skb(data, frag_size > PAGE_SIZE ? 0 : frag_size);
 
 		/* After refill old buffer has to be unmapped regardless
 		 * the skb is successfully built or not.
 		 */
-		dma_unmap_single(dev->dev.parent, phys_addr,
-				 MVNETA_RX_BUF_SIZE(pp->pkt_size), DMA_FROM_DEVICE);
-
+		dma_unmap_single(&pp->bm_priv->pdev->dev, phys_addr,
+				 bm_pool->buf_size, DMA_FROM_DEVICE);
 		if (!skb)
 			goto err_drop_frame;
 
@@ -2297,7 +2686,10 @@ static int mvneta_poll(struct napi_struct *napi, int budget)
 
 	if (rx_queue) {
 		rx_queue = rx_queue - 1;
-		rx_done = mvneta_rx(pp, budget, &pp->rxqs[rx_queue]);
+		if (pp->bm_priv)
+			rx_done = mvneta_rx_hwbm(pp, budget, &pp->rxqs[rx_queue]);
+		else
+			rx_done = mvneta_rx_swbm(pp, budget, &pp->rxqs[rx_queue]);
 	}
 
 	budget -= rx_done;
@@ -2386,9 +2778,17 @@ static int mvneta_rxq_init(struct mvneta_port *pp,
 	mvneta_rx_pkts_coal_set(pp, rxq, rxq->pkts_coal);
 	mvneta_rx_time_coal_set(pp, rxq, rxq->time_coal);
 
-	/* Fill RXQ with buffers from RX pool */
-	mvneta_rxq_buf_size_set(pp, rxq, MVNETA_RX_BUF_SIZE(pp->pkt_size));
-	mvneta_rxq_bm_disable(pp, rxq);
+	if (!pp->bm_priv) {
+		/* Fill RXQ with buffers from RX pool */
+		mvneta_rxq_buf_size_set(pp, rxq,
+					MVNETA_RX_BUF_SIZE(pp->pkt_size));
+		mvneta_rxq_bm_disable(pp, rxq);
+	} else {
+		mvneta_rxq_bm_enable(pp, rxq);
+		mvneta_rxq_long_pool_set(pp, rxq);
+		mvneta_rxq_short_pool_set(pp, rxq);
+	}
+
 	mvneta_rxq_fill(pp, rxq, rxq->size);
 
 	return 0;
@@ -2661,6 +3061,9 @@ static int mvneta_change_mtu(struct net_device *dev, int mtu)
 	dev->mtu = mtu;
 
 	if (!netif_running(dev)) {
+		if (pp->bm_priv)
+			mvneta_bm_update_mtu(pp, mtu);
+
 		netdev_update_features(dev);
 		return 0;
 	}
@@ -2673,6 +3076,9 @@ static int mvneta_change_mtu(struct net_device *dev, int mtu)
 	mvneta_cleanup_txqs(pp);
 	mvneta_cleanup_rxqs(pp);
 
+	if (pp->bm_priv)
+		mvneta_bm_update_mtu(pp, mtu);
+
 	pp->pkt_size = MVNETA_RX_PKT_SIZE(dev->mtu);
 	pp->frag_size = SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(pp->pkt_size)) +
 	                SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
@@ -3557,6 +3963,7 @@ static int mvneta_probe(struct platform_device *pdev)
 	struct resource *res;
 	struct device_node *dn = pdev->dev.of_node;
 	struct device_node *phy_node;
+	struct device_node *bm_node;
 	struct mvneta_port *pp;
 	struct net_device *dev;
 	const char *dt_mac_addr;
@@ -3690,26 +4097,39 @@ static int mvneta_probe(struct platform_device *pdev)
 
 	pp->tx_csum_limit = tx_csum_limit;
 
+	dram_target_info = mv_mbus_dram_info();
+	if (dram_target_info)
+		mvneta_conf_mbus_windows(pp, dram_target_info);
+
 	pp->tx_ring_size = MVNETA_MAX_TXD;
 	pp->rx_ring_size = MVNETA_MAX_RXD;
 
 	pp->dev = dev;
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
+	pp->id = global_port_id++;
+
+	/* Obtain access to BM resources if enabled and already initialized */
+	bm_node = of_parse_phandle(dn, "buffer-manager", 0);
+	if (bm_node && bm_node->data) {
+		pp->bm_priv = bm_node->data;
+		err = mvneta_bm_port_init(pdev, pp);
+		if (err < 0) {
+			dev_info(&pdev->dev, "use SW buffer management\n");
+			pp->bm_priv = NULL;
+		}
+	}
+
 	err = mvneta_init(&pdev->dev, pp);
 	if (err < 0)
-		goto err_free_stats;
+		goto err_netdev;
 
 	err = mvneta_port_power_up(pp, phy_mode);
 	if (err < 0) {
 		dev_err(&pdev->dev, "can't power up port\n");
-		goto err_free_stats;
+		goto err_netdev;
 	}
 
-	dram_target_info = mv_mbus_dram_info();
-	if (dram_target_info)
-		mvneta_conf_mbus_windows(pp, dram_target_info);
-
 	for_each_present_cpu(cpu) {
 		struct mvneta_pcpu_port *port = per_cpu_ptr(pp->ports, cpu);
 
@@ -3744,6 +4164,13 @@ static int mvneta_probe(struct platform_device *pdev)
 
 	return 0;
 
+err_netdev:
+	unregister_netdev(dev);
+	if (pp->bm_priv) {
+		mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_long, 1 << pp->id);
+		mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_short,
+				       1 << pp->id);
+	}
 err_free_stats:
 	free_percpu(pp->stats);
 err_free_ports:
@@ -3775,6 +4202,12 @@ static int mvneta_remove(struct platform_device *pdev)
 	of_node_put(pp->phy_node);
 	free_netdev(dev);
 
+	if (pp->bm_priv) {
+		mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_long, 1 << pp->id);
+		mvneta_bm_pool_destroy(pp->bm_priv, pp->pool_short,
+				       1 << pp->id);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/marvell/mvneta_bm.c b/drivers/net/ethernet/marvell/mvneta_bm.c
new file mode 100644
index 0000000..8c968e7
--- /dev/null
+++ b/drivers/net/ethernet/marvell/mvneta_bm.c
@@ -0,0 +1,546 @@
+/*
+ * Driver for Marvell NETA network controller Buffer Manager.
+ *
+ * Copyright (C) 2015 Marvell
+ *
+ * Marcin Wojtas <mw@semihalf.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/genalloc.h>
+#include <linux/platform_device.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+#include <linux/mbus.h>
+#include <linux/module.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/clk.h>
+#include "mvneta_bm.h"
+
+#define MVNETA_BM_DRIVER_NAME "mvneta_bm"
+#define MVNETA_BM_DRIVER_VERSION "1.0"
+
+static void mvneta_bm_write(struct mvneta_bm *priv, u32 offset, u32 data)
+{
+	writel(data, priv->reg_base + offset);
+}
+
+static u32 mvneta_bm_read(struct mvneta_bm *priv, u32 offset)
+{
+	return readl(priv->reg_base + offset);
+}
+
+static void mvneta_bm_pool_enable(struct mvneta_bm *priv, int pool_id)
+{
+	u32 val;
+
+	val = mvneta_bm_read(priv, MVNETA_BM_POOL_BASE_REG(pool_id));
+	val |= MVNETA_BM_POOL_ENABLE_MASK;
+	mvneta_bm_write(priv, MVNETA_BM_POOL_BASE_REG(pool_id), val);
+
+	/* Clear BM cause register */
+	mvneta_bm_write(priv, MVNETA_BM_INTR_CAUSE_REG, 0);
+}
+
+static void mvneta_bm_pool_disable(struct mvneta_bm *priv, int pool_id)
+{
+	u32 val;
+
+	val = mvneta_bm_read(priv, MVNETA_BM_POOL_BASE_REG(pool_id));
+	val &= ~MVNETA_BM_POOL_ENABLE_MASK;
+	mvneta_bm_write(priv, MVNETA_BM_POOL_BASE_REG(pool_id), val);
+}
+
+static inline void mvneta_bm_config_set(struct mvneta_bm *priv, u32 mask)
+{
+	u32 val;
+
+	val = mvneta_bm_read(priv, MVNETA_BM_CONFIG_REG);
+	val |= mask;
+	mvneta_bm_write(priv, MVNETA_BM_CONFIG_REG, val);
+}
+
+static inline void mvneta_bm_config_clear(struct mvneta_bm *priv, u32 mask)
+{
+	u32 val;
+
+	val = mvneta_bm_read(priv, MVNETA_BM_CONFIG_REG);
+	val &= ~mask;
+	mvneta_bm_write(priv, MVNETA_BM_CONFIG_REG, val);
+}
+
+static void mvneta_bm_pool_target_set(struct mvneta_bm *priv, int pool_id,
+				      u8 target_id, u8 attr)
+{
+	u32 val;
+
+	val = mvneta_bm_read(priv, MVNETA_BM_XBAR_POOL_REG(pool_id));
+	val &= ~MVNETA_BM_TARGET_ID_MASK(pool_id);
+	val &= ~MVNETA_BM_XBAR_ATTR_MASK(pool_id);
+	val |= MVNETA_BM_TARGET_ID_VAL(pool_id, target_id);
+	val |= MVNETA_BM_XBAR_ATTR_VAL(pool_id, attr);
+
+	mvneta_bm_write(priv, MVNETA_BM_XBAR_POOL_REG(pool_id), val);
+}
+
+/* Allocate skb for BM pool */
+void *mvneta_buf_alloc(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+		       dma_addr_t *buf_phys_addr)
+{
+	void *buf;
+	dma_addr_t phys_addr;
+
+	buf = mvneta_frag_alloc(bm_pool->frag_size);
+	if (!buf)
+		return NULL;
+
+	/* In order to update buf_cookie field of RX descriptor properly,
+	 * BM hardware expects buf virtual address to be placed in the
+	 * first four bytes of mapped buffer.
+	 */
+	*(u32 *)buf = (u32)buf;
+	phys_addr = dma_map_single(&priv->pdev->dev, buf, bm_pool->buf_size,
+				   DMA_FROM_DEVICE);
+	if (unlikely(dma_mapping_error(&priv->pdev->dev, phys_addr))) {
+		mvneta_frag_free(bm_pool->frag_size, buf);
+		return NULL;
+	}
+	*buf_phys_addr = phys_addr;
+
+	return buf;
+}
+
+/* Refill processing for HW buffer management */
+int mvneta_bm_pool_refill(struct mvneta_bm *priv,
+			  struct mvneta_bm_pool *bm_pool)
+{
+	dma_addr_t buf_phys_addr;
+	void *buf;
+
+	buf = mvneta_buf_alloc(priv, bm_pool, &buf_phys_addr);
+	if (!buf)
+		return -ENOMEM;
+
+	mvneta_bm_pool_put_bp(priv, bm_pool, buf_phys_addr);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mvneta_bm_pool_refill);
+
+/* Allocate buffers for the pool */
+int mvneta_bm_bufs_add(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+		       int buf_num)
+{
+	int err, i;
+
+	if (bm_pool->buf_num == bm_pool->size) {
+		dev_dbg(&priv->pdev->dev, "pool %d already filled\n",
+			bm_pool->id);
+		return bm_pool->buf_num;
+	}
+
+	if (buf_num < 0 ||
+	    (buf_num + bm_pool->buf_num > bm_pool->size)) {
+		dev_err(&priv->pdev->dev,
+			"cannot allocate %d buffers for pool %d\n",
+			buf_num, bm_pool->id);
+		return 0;
+	}
+
+	for (i = 0; i < buf_num; i++) {
+		err = mvneta_bm_pool_refill(priv, bm_pool);
+		if (err < 0)
+			break;
+	}
+
+	/* Update BM driver with number of buffers added to pool */
+	bm_pool->buf_num += i;
+
+	dev_dbg(&priv->pdev->dev,
+		"%s pool %d: pkt_size=%4d, buf_size=%4d, frag_size=%4d\n",
+		bm_pool->type == MVNETA_BM_SHORT ? "short" : "long",
+		bm_pool->id, bm_pool->pkt_size, bm_pool->buf_size,
+		bm_pool->frag_size);
+
+	dev_dbg(&priv->pdev->dev,
+		"%s pool %d: %d of %d buffers added\n",
+		bm_pool->type == MVNETA_BM_SHORT ? "short" : "long",
+		bm_pool->id, i, buf_num);
+
+	return i;
+}
+EXPORT_SYMBOL_GPL(mvneta_bm_bufs_add);
+
+/* Create pool */
+static int mvneta_bm_pool_create(struct mvneta_bm *priv,
+				 struct mvneta_bm_pool *bm_pool)
+{
+	struct platform_device *pdev = priv->pdev;
+	u8 target_id, attr;
+	int size_bytes, err;
+
+	size_bytes = sizeof(u32) * bm_pool->size;
+	bm_pool->virt_addr = dma_alloc_coherent(&pdev->dev, size_bytes,
+						&bm_pool->phys_addr,
+						GFP_KERNEL);
+	if (!bm_pool->virt_addr)
+		return -ENOMEM;
+
+	if (!IS_ALIGNED((u32)bm_pool->virt_addr, MVNETA_BM_POOL_PTR_ALIGN)) {
+		dma_free_coherent(&pdev->dev, size_bytes, bm_pool->virt_addr,
+				  bm_pool->phys_addr);
+		dev_err(&pdev->dev, "BM pool %d is not %d bytes aligned\n",
+			bm_pool->id, MVNETA_BM_POOL_PTR_ALIGN);
+		return -ENOMEM;
+	}
+
+	err = mvebu_mbus_get_dram_win_info(bm_pool->phys_addr, &target_id,
+					   &attr);
+	if (err < 0) {
+		dma_free_coherent(&pdev->dev, size_bytes, bm_pool->virt_addr,
+				  bm_pool->phys_addr);
+		return err;
+	}
+
+	/* Set pool address */
+	mvneta_bm_write(priv, MVNETA_BM_POOL_BASE_REG(bm_pool->id),
+			bm_pool->phys_addr);
+
+	mvneta_bm_pool_target_set(priv, bm_pool->id, target_id,  attr);
+	mvneta_bm_pool_enable(priv, bm_pool->id);
+
+	return 0;
+}
+
+/* Notify the driver that BM pool is being used as specific type and return the
+ * pool pointer on success
+ */
+struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
+					  enum mvneta_bm_type type, u8 port_id,
+					  int pkt_size)
+{
+	struct mvneta_bm_pool *new_pool = &priv->bm_pools[pool_id];
+	int num, err;
+
+	if (new_pool->type == MVNETA_BM_LONG &&
+	    new_pool->port_map != 1 << port_id) {
+		dev_err(&priv->pdev->dev,
+			"long pool cannot be shared by the ports\n");
+		return NULL;
+	}
+
+	if (new_pool->type == MVNETA_BM_SHORT && new_pool->type != type) {
+		dev_err(&priv->pdev->dev,
+			"mixing pools' types between the ports is forbidden\n");
+		return NULL;
+	}
+
+	if (new_pool->pkt_size == 0 || type != MVNETA_BM_SHORT)
+		new_pool->pkt_size = pkt_size;
+
+	/* Allocate buffers in case BM pool hasn't been used yet */
+	if (new_pool->type == MVNETA_BM_FREE) {
+		new_pool->type = type;
+		new_pool->buf_size = MVNETA_RX_BUF_SIZE(new_pool->pkt_size);
+		new_pool->frag_size =
+			SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(new_pool->pkt_size)) +
+			SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+
+		/* Create new pool */
+		err = mvneta_bm_pool_create(priv, new_pool);
+		if (err) {
+			dev_err(&priv->pdev->dev, "fail to create pool %d\n",
+				new_pool->id);
+			return NULL;
+		}
+
+		/* Allocate buffers for this pool */
+		num = mvneta_bm_bufs_add(priv, new_pool, new_pool->size);
+		if (num != new_pool->size) {
+			WARN(1, "pool %d: %d of %d allocated\n",
+			     new_pool->id, num, new_pool->size);
+			return NULL;
+		}
+	}
+
+	return new_pool;
+}
+EXPORT_SYMBOL_GPL(mvneta_bm_pool_use);
+
+/* Free all buffers from the pool */
+void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+			 u8 port_map)
+{
+	int i;
+
+	bm_pool->port_map &= ~port_map;
+	if (bm_pool->port_map)
+		return;
+
+	mvneta_bm_config_set(priv, MVNETA_BM_EMPTY_LIMIT_MASK);
+
+	for (i = 0; i < bm_pool->buf_num; i++) {
+		dma_addr_t buf_phys_addr;
+		u32 *vaddr;
+
+		/* Get buffer physical address (indirect access) */
+		buf_phys_addr = mvneta_bm_pool_get_bp(priv, bm_pool);
+
+		/* Work-around to the problems when destroying the pool,
+		 * when it occurs that a read access to BPPI returns 0.
+		 */
+		if (buf_phys_addr == 0)
+			continue;
+
+		vaddr = phys_to_virt(buf_phys_addr);
+		if (!vaddr)
+			break;
+
+		dma_unmap_single(&priv->pdev->dev, buf_phys_addr,
+				 bm_pool->buf_size, DMA_FROM_DEVICE);
+		mvneta_frag_free(bm_pool->frag_size, vaddr);
+	}
+
+	mvneta_bm_config_clear(priv, MVNETA_BM_EMPTY_LIMIT_MASK);
+
+	/* Update BM driver with number of buffers removed from pool */
+	bm_pool->buf_num -= i;
+}
+EXPORT_SYMBOL_GPL(mvneta_bm_bufs_free);
+
+/* Cleanup pool */
+void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
+			    struct mvneta_bm_pool *bm_pool, u8 port_map)
+{
+	bm_pool->port_map &= ~port_map;
+	if (bm_pool->port_map)
+		return;
+
+	bm_pool->type = MVNETA_BM_FREE;
+
+	mvneta_bm_bufs_free(priv, bm_pool, port_map);
+	if (bm_pool->buf_num)
+		WARN(1, "cannot free all buffers in pool %d\n", bm_pool->id);
+
+	if (bm_pool->virt_addr) {
+		dma_free_coherent(&priv->pdev->dev, sizeof(u32) * bm_pool->size,
+				  bm_pool->virt_addr, bm_pool->phys_addr);
+		bm_pool->virt_addr = NULL;
+	}
+
+	mvneta_bm_pool_disable(priv, bm_pool->id);
+}
+EXPORT_SYMBOL_GPL(mvneta_bm_pool_destroy);
+
+static void mvneta_bm_pools_init(struct mvneta_bm *priv)
+{
+	struct device_node *dn = priv->pdev->dev.of_node;
+	struct mvneta_bm_pool *bm_pool;
+	char prop[15];
+	u32 size;
+	int i;
+
+	/* Activate BM unit */
+	mvneta_bm_write(priv, MVNETA_BM_COMMAND_REG, MVNETA_BM_START_MASK);
+
+	/* Create all pools with maximum size */
+	for (i = 0; i < MVNETA_BM_POOLS_NUM; i++) {
+		bm_pool = &priv->bm_pools[i];
+		bm_pool->id = i;
+		bm_pool->type = MVNETA_BM_FREE;
+
+		/* Reset read pointer */
+		mvneta_bm_write(priv, MVNETA_BM_POOL_READ_PTR_REG(i), 0);
+
+		/* Reset write pointer */
+		mvneta_bm_write(priv, MVNETA_BM_POOL_WRITE_PTR_REG(i), 0);
+
+		/* Configure pool size according to DT or use default value */
+		sprintf(prop, "pool%d,capacity", i);
+		if (of_property_read_u32(dn, prop, &size)) {
+			size = MVNETA_BM_POOL_CAP_DEF;
+		} else if (size > MVNETA_BM_POOL_CAP_MAX) {
+			dev_warn(&priv->pdev->dev,
+				 "Illegal pool %d capacity %d, set to %d\n",
+				 i, size, MVNETA_BM_POOL_CAP_MAX);
+			size = MVNETA_BM_POOL_CAP_MAX;
+		} else if (size < MVNETA_BM_POOL_CAP_MIN) {
+			dev_warn(&priv->pdev->dev,
+				 "Illegal pool %d capacity %d, set to %d\n",
+				 i, size, MVNETA_BM_POOL_CAP_MIN);
+			size = MVNETA_BM_POOL_CAP_MIN;
+		} else if (!IS_ALIGNED(size, MVNETA_BM_POOL_CAP_ALIGN)) {
+			dev_warn(&priv->pdev->dev,
+				 "Illegal pool %d capacity %d, round to %d\n",
+				 i, size, ALIGN(size,
+				 MVNETA_BM_POOL_CAP_ALIGN));
+			size = ALIGN(size, MVNETA_BM_POOL_CAP_ALIGN);
+		}
+		bm_pool->size = size;
+
+		mvneta_bm_write(priv, MVNETA_BM_POOL_SIZE_REG(i),
+				bm_pool->size);
+
+		/* Obtain custom pkt_size from DT */
+		sprintf(prop, "pool%d,pkt-size", i);
+		if (of_property_read_u32(dn, prop, &bm_pool->pkt_size))
+			bm_pool->pkt_size = 0;
+	}
+}
+
+static void mvneta_bm_default_set(struct mvneta_bm *priv)
+{
+	u32 val;
+
+	/* Mask BM all interrupts */
+	mvneta_bm_write(priv, MVNETA_BM_INTR_MASK_REG, 0);
+
+	/* Clear BM cause register */
+	mvneta_bm_write(priv, MVNETA_BM_INTR_CAUSE_REG, 0);
+
+	/* Set BM configuration register */
+	val = mvneta_bm_read(priv, MVNETA_BM_CONFIG_REG);
+
+	/* Reduce MaxInBurstSize from 32 BPs to 16 BPs */
+	val &= ~MVNETA_BM_MAX_IN_BURST_SIZE_MASK;
+	val |= MVNETA_BM_MAX_IN_BURST_SIZE_16BP;
+	mvneta_bm_write(priv, MVNETA_BM_CONFIG_REG, val);
+}
+
+static int mvneta_bm_init(struct mvneta_bm *priv)
+{
+	mvneta_bm_default_set(priv);
+
+	/* Allocate and initialize BM pools structures */
+	priv->bm_pools = devm_kcalloc(&priv->pdev->dev, MVNETA_BM_POOLS_NUM,
+				      sizeof(struct mvneta_bm_pool),
+				      GFP_KERNEL);
+	if (!priv->bm_pools)
+		return -ENOMEM;
+
+	mvneta_bm_pools_init(priv);
+
+	return 0;
+}
+
+static int mvneta_bm_get_sram(struct device_node *dn,
+			      struct mvneta_bm *priv)
+{
+	priv->bppi_pool = of_gen_pool_get(dn, "internal-mem", 0);
+	if (!priv->bppi_pool)
+		return -ENOMEM;
+
+	priv->bppi_virt_addr = gen_pool_dma_alloc(priv->bppi_pool,
+						  MVNETA_BM_BPPI_SIZE,
+						  &priv->bppi_phys_addr);
+	if (!priv->bppi_virt_addr)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void mvneta_bm_put_sram(struct mvneta_bm *priv)
+{
+	gen_pool_free(priv->bppi_pool, priv->bppi_phys_addr,
+		      MVNETA_BM_BPPI_SIZE);
+}
+
+static int mvneta_bm_probe(struct platform_device *pdev)
+{
+	struct device_node *dn = pdev->dev.of_node;
+	struct mvneta_bm *priv;
+	struct resource *res;
+	int err;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(struct mvneta_bm), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	priv->reg_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(priv->reg_base))
+		return PTR_ERR(priv->reg_base);
+
+	priv->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(priv->clk))
+		return PTR_ERR(priv->clk);
+	err = clk_prepare_enable(priv->clk);
+	if (err < 0)
+		return err;
+
+	err = mvneta_bm_get_sram(dn, priv);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to allocate internal memory\n");
+		goto err_clk;
+	}
+
+	priv->pdev = pdev;
+
+	/* Initialize buffer manager internals */
+	err = mvneta_bm_init(priv);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to initialize controller\n");
+		goto err_sram;
+	}
+
+	dn->data = priv;
+	platform_set_drvdata(pdev, priv);
+
+	dev_info(&pdev->dev, "Buffer Manager for network controller enabled\n");
+
+	return 0;
+
+err_sram:
+	mvneta_bm_put_sram(priv);
+err_clk:
+	clk_disable_unprepare(priv->clk);
+	return err;
+}
+
+static int mvneta_bm_remove(struct platform_device *pdev)
+{
+	struct mvneta_bm *priv = platform_get_drvdata(pdev);
+	u8 all_ports_map = 0xff;
+	int i = 0;
+
+	for (i = 0; i < MVNETA_BM_POOLS_NUM; i++) {
+		struct mvneta_bm_pool *bm_pool = &priv->bm_pools[i];
+
+		mvneta_bm_pool_destroy(priv, bm_pool, all_ports_map);
+	}
+
+	mvneta_bm_put_sram(priv);
+
+	/* Dectivate BM unit */
+	mvneta_bm_write(priv, MVNETA_BM_COMMAND_REG, MVNETA_BM_STOP_MASK);
+
+	clk_disable_unprepare(priv->clk);
+
+	return 0;
+}
+
+static const struct of_device_id mvneta_bm_match[] = {
+	{ .compatible = "marvell,armada-380-neta-bm" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, mvneta_bm_match);
+
+static struct platform_driver mvneta_bm_driver = {
+	.probe = mvneta_bm_probe,
+	.remove = mvneta_bm_remove,
+	.driver = {
+		.name = MVNETA_BM_DRIVER_NAME,
+		.of_match_table = mvneta_bm_match,
+	},
+};
+
+module_platform_driver(mvneta_bm_driver);
+
+MODULE_DESCRIPTION("Marvell NETA Buffer Manager Driver - www.marvell.com");
+MODULE_AUTHOR("Marcin Wojtas <mw@semihalf.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/ethernet/marvell/mvneta_bm.h b/drivers/net/ethernet/marvell/mvneta_bm.h
new file mode 100644
index 0000000..db239e0
--- /dev/null
+++ b/drivers/net/ethernet/marvell/mvneta_bm.h
@@ -0,0 +1,189 @@
+/*
+ * Driver for Marvell NETA network controller Buffer Manager.
+ *
+ * Copyright (C) 2015 Marvell
+ *
+ * Marcin Wojtas <mw@semihalf.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef _MVNETA_BM_H_
+#define _MVNETA_BM_H_
+
+/* BM Configuration Register */
+#define MVNETA_BM_CONFIG_REG			0x0
+#define    MVNETA_BM_STATUS_MASK		0x30
+#define    MVNETA_BM_ACTIVE_MASK		BIT(4)
+#define    MVNETA_BM_MAX_IN_BURST_SIZE_MASK	0x60000
+#define    MVNETA_BM_MAX_IN_BURST_SIZE_16BP	BIT(18)
+#define    MVNETA_BM_EMPTY_LIMIT_MASK		BIT(19)
+
+/* BM Activation Register */
+#define MVNETA_BM_COMMAND_REG			0x4
+#define    MVNETA_BM_START_MASK			BIT(0)
+#define    MVNETA_BM_STOP_MASK			BIT(1)
+#define    MVNETA_BM_PAUSE_MASK			BIT(2)
+
+/* BM Xbar interface Register */
+#define MVNETA_BM_XBAR_01_REG			0x8
+#define MVNETA_BM_XBAR_23_REG			0xc
+#define MVNETA_BM_XBAR_POOL_REG(pool)		\
+		(((pool) < 2) ? MVNETA_BM_XBAR_01_REG : MVNETA_BM_XBAR_23_REG)
+#define     MVNETA_BM_TARGET_ID_OFFS(pool)	(((pool) & 1) ? 16 : 0)
+#define     MVNETA_BM_TARGET_ID_MASK(pool)	\
+		(0xf << MVNETA_BM_TARGET_ID_OFFS(pool))
+#define     MVNETA_BM_TARGET_ID_VAL(pool, id)	\
+		((id) << MVNETA_BM_TARGET_ID_OFFS(pool))
+#define     MVNETA_BM_XBAR_ATTR_OFFS(pool)	(((pool) & 1) ? 20 : 4)
+#define     MVNETA_BM_XBAR_ATTR_MASK(pool)	\
+		(0xff << MVNETA_BM_XBAR_ATTR_OFFS(pool))
+#define     MVNETA_BM_XBAR_ATTR_VAL(pool, attr)	\
+		((attr) << MVNETA_BM_XBAR_ATTR_OFFS(pool))
+
+/* Address of External Buffer Pointers Pool Register */
+#define MVNETA_BM_POOL_BASE_REG(pool)		(0x10 + ((pool) << 4))
+#define     MVNETA_BM_POOL_ENABLE_MASK		BIT(0)
+
+/* External Buffer Pointers Pool RD pointer Register */
+#define MVNETA_BM_POOL_READ_PTR_REG(pool)	(0x14 + ((pool) << 4))
+#define     MVNETA_BM_POOL_SET_READ_PTR_MASK	0xfffc
+#define     MVNETA_BM_POOL_GET_READ_PTR_OFFS	16
+#define     MVNETA_BM_POOL_GET_READ_PTR_MASK	0xfffc0000
+
+/* External Buffer Pointers Pool WR pointer */
+#define MVNETA_BM_POOL_WRITE_PTR_REG(pool)	(0x18 + ((pool) << 4))
+#define     MVNETA_BM_POOL_SET_WRITE_PTR_OFFS	0
+#define     MVNETA_BM_POOL_SET_WRITE_PTR_MASK	0xfffc
+#define     MVNETA_BM_POOL_GET_WRITE_PTR_OFFS	16
+#define     MVNETA_BM_POOL_GET_WRITE_PTR_MASK	0xfffc0000
+
+/* External Buffer Pointers Pool Size Register */
+#define MVNETA_BM_POOL_SIZE_REG(pool)		(0x1c + ((pool) << 4))
+#define     MVNETA_BM_POOL_SIZE_MASK		0x3fff
+
+/* BM Interrupt Cause Register */
+#define MVNETA_BM_INTR_CAUSE_REG		(0x50)
+
+/* BM interrupt Mask Register */
+#define MVNETA_BM_INTR_MASK_REG			(0x54)
+
+/* Other definitions */
+#define MVNETA_BM_SHORT_PKT_SIZE		256
+#define MVNETA_BM_POOLS_NUM			4
+#define MVNETA_BM_POOL_CAP_MIN			128
+#define MVNETA_BM_POOL_CAP_DEF			2048
+#define MVNETA_BM_POOL_CAP_MAX			\
+		(16 * 1024 - MVNETA_BM_POOL_CAP_ALIGN)
+#define MVNETA_BM_POOL_CAP_ALIGN		32
+#define MVNETA_BM_POOL_PTR_ALIGN		32
+
+#define MVNETA_BM_POOL_ACCESS_OFFS		8
+
+#define MVNETA_BM_BPPI_SIZE			0x100000
+
+#define MVNETA_RX_BUF_SIZE(pkt_size)   ((pkt_size) + NET_SKB_PAD)
+
+enum mvneta_bm_type {
+	MVNETA_BM_FREE,
+	MVNETA_BM_LONG,
+	MVNETA_BM_SHORT
+};
+
+struct mvneta_bm {
+	void __iomem *reg_base;
+	struct clk *clk;
+	struct platform_device *pdev;
+
+	struct gen_pool *bppi_pool;
+	/* BPPI virtual base address */
+	void __iomem *bppi_virt_addr;
+	/* BPPI physical base address */
+	dma_addr_t bppi_phys_addr;
+
+	/* BM pools */
+	struct mvneta_bm_pool *bm_pools;
+};
+
+struct mvneta_bm_pool {
+	/* Pool number in the range 0-3 */
+	u8 id;
+	enum mvneta_bm_type type;
+
+	/* Buffer Pointers Pool External (BPPE) size in number of bytes */
+	int size;
+	/* Number of buffers used by this pool */
+	int buf_num;
+	/* Pool buffer size */
+	int buf_size;
+	/* Packet size */
+	int pkt_size;
+	/* Single frag size */
+	u32 frag_size;
+
+	/* BPPE virtual base address */
+	u32 *virt_addr;
+	/* BPPE physical base address */
+	dma_addr_t phys_addr;
+
+	/* Ports using BM pool */
+	u8 port_map;
+
+	struct mvneta_bm *priv;
+};
+
+/* Declarations and definitions */
+void *mvneta_frag_alloc(unsigned int frag_size);
+void mvneta_frag_free(unsigned int frag_size, void *data);
+
+#if defined(CONFIG_MVNETA_BM) || defined(CONFIG_MVNETA_BM_MODULE)
+void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
+			    struct mvneta_bm_pool *bm_pool, u8 port_map);
+void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+			 u8 port_map);
+int mvneta_bm_bufs_add(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+		       int buf_num);
+int mvneta_bm_pool_refill(struct mvneta_bm *priv,
+			  struct mvneta_bm_pool *bm_pool);
+struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
+					  enum mvneta_bm_type type, u8 port_id,
+					  int pkt_size);
+
+static inline void mvneta_bm_pool_put_bp(struct mvneta_bm *priv,
+					 struct mvneta_bm_pool *bm_pool,
+					 dma_addr_t buf_phys_addr)
+{
+	writel_relaxed(buf_phys_addr, priv->bppi_virt_addr +
+		       (bm_pool->id << MVNETA_BM_POOL_ACCESS_OFFS));
+}
+
+static inline u32 mvneta_bm_pool_get_bp(struct mvneta_bm *priv,
+					struct mvneta_bm_pool *bm_pool)
+{
+	return readl_relaxed(priv->bppi_virt_addr +
+			     (bm_pool->id << MVNETA_BM_POOL_ACCESS_OFFS));
+}
+#else
+void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
+			    struct mvneta_bm_pool *bm_pool, u8 port_map) {}
+void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+			 u8 port_map) {}
+int mvneta_bm_bufs_add(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
+		       int buf_num) { return 0; }
+int mvneta_bm_pool_refill(struct mvneta_bm *priv,
+			  struct mvneta_bm_pool *bm_pool) {return 0; }
+struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
+					  enum mvneta_bm_type type, u8 port_id,
+					  int pkt_size) { return NULL; }
+
+static inline void mvneta_bm_pool_put_bp(struct mvneta_bm *priv,
+					 struct mvneta_bm_pool *bm_pool,
+					 dma_addr_t buf_phys_addr) {}
+
+static inline u32 mvneta_bm_pool_get_bp(struct mvneta_bm *priv,
+					struct mvneta_bm_pool *bm_pool)
+{ return 0; }
+#endif /* CONFIG_MVNETA_BM */
+#endif
-- 
cgit v0.10.2


From 8cb2d8bf57e6e004c37db2fb4ce74f4d032b7cd0 Mon Sep 17 00:00:00 2001
From: Gregory CLEMENT <gregory.clement@free-electrons.com>
Date: Mon, 14 Mar 2016 09:39:04 +0100
Subject: net: add a hardware buffer management helper API

This basic implementation allows to share code between driver using
hardware buffer management. As the code is hardware agnostic, there is
few helpers, most of the optimization brought by the an HW BM has to be
done at driver level.

Tested-by: Sebastian Careba <nitroshift@yahoo.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/hwbm.h b/include/net/hwbm.h
new file mode 100644
index 0000000..47d0866
--- /dev/null
+++ b/include/net/hwbm.h
@@ -0,0 +1,28 @@
+#ifndef _HWBM_H
+#define _HWBM_H
+
+struct hwbm_pool {
+	/* Capacity of the pool */
+	int size;
+	/* Size of the buffers managed */
+	int frag_size;
+	/* Number of buffers currently used by this pool */
+	int buf_num;
+	/* constructor called during alocation */
+	int (*construct)(struct hwbm_pool *bm_pool, void *buf);
+	/* protect acces to the buffer counter*/
+	spinlock_t lock;
+	/* private data */
+	void *priv;
+};
+#ifdef CONFIG_HWBM
+void hwbm_buf_free(struct hwbm_pool *bm_pool, void *buf);
+int hwbm_pool_refill(struct hwbm_pool *bm_pool, gfp_t gfp);
+int hwbm_pool_add(struct hwbm_pool *bm_pool, unsigned int buf_num, gfp_t gfp);
+#else
+void hwbm_buf_free(struct hwbm_pool *bm_pool, void *buf) {}
+int hwbm_pool_refill(struct hwbm_pool *bm_pool, gfp_t gfp) { return 0; }
+int hwbm_pool_add(struct hwbm_pool *bm_pool, unsigned int buf_num, gfp_t gfp)
+{ return 0; }
+#endif /* CONFIG_HWBM */
+#endif /* _HWBM_H */
diff --git a/net/Kconfig b/net/Kconfig
index 10640d5..e134498 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -253,6 +253,9 @@ config XPS
 	depends on SMP
 	default y
 
+config HWBM
+       bool
+
 config SOCK_CGROUP_DATA
 	bool
 	default n
diff --git a/net/core/Makefile b/net/core/Makefile
index 014422e..d6508c2 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -25,4 +25,5 @@ obj-$(CONFIG_CGROUP_NET_PRIO) += netprio_cgroup.o
 obj-$(CONFIG_CGROUP_NET_CLASSID) += netclassid_cgroup.o
 obj-$(CONFIG_LWTUNNEL) += lwtunnel.o
 obj-$(CONFIG_DST_CACHE) += dst_cache.o
+obj-$(CONFIG_HWBM) += hwbm.o
 obj-$(CONFIG_NET_DEVLINK) += devlink.o
diff --git a/net/core/hwbm.c b/net/core/hwbm.c
new file mode 100644
index 0000000..941c284
--- /dev/null
+++ b/net/core/hwbm.c
@@ -0,0 +1,87 @@
+/* Support for hardware buffer manager.
+ *
+ * Copyright (C) 2016 Marvell
+ *
+ * Gregory CLEMENT <gregory.clement@free-electrons.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/printk.h>
+#include <linux/skbuff.h>
+#include <net/hwbm.h>
+
+void hwbm_buf_free(struct hwbm_pool *bm_pool, void *buf)
+{
+	if (likely(bm_pool->frag_size <= PAGE_SIZE))
+		skb_free_frag(buf);
+	else
+		kfree(buf);
+}
+EXPORT_SYMBOL_GPL(hwbm_buf_free);
+
+/* Refill processing for HW buffer management */
+int hwbm_pool_refill(struct hwbm_pool *bm_pool, gfp_t gfp)
+{
+	int frag_size = bm_pool->frag_size;
+	void *buf;
+
+	if (likely(frag_size <= PAGE_SIZE))
+		buf = netdev_alloc_frag(frag_size);
+	else
+		buf = kmalloc(frag_size, gfp);
+
+	if (!buf)
+		return -ENOMEM;
+
+	if (bm_pool->construct)
+		if (bm_pool->construct(bm_pool, buf)) {
+			hwbm_buf_free(bm_pool, buf);
+			return -ENOMEM;
+		}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hwbm_pool_refill);
+
+int hwbm_pool_add(struct hwbm_pool *bm_pool, unsigned int buf_num, gfp_t gfp)
+{
+	int err, i;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bm_pool->lock, flags);
+	if (bm_pool->buf_num == bm_pool->size) {
+		pr_warn("pool already filled\n");
+		return bm_pool->buf_num;
+	}
+
+	if (buf_num + bm_pool->buf_num > bm_pool->size) {
+		pr_warn("cannot allocate %d buffers for pool\n",
+			buf_num);
+		return 0;
+	}
+
+	if ((buf_num + bm_pool->buf_num) < bm_pool->buf_num) {
+		pr_warn("Adding %d buffers to the %d current buffers will overflow\n",
+			buf_num,  bm_pool->buf_num);
+		return 0;
+	}
+
+	for (i = 0; i < buf_num; i++) {
+		err = hwbm_pool_refill(bm_pool, gfp);
+		if (err < 0)
+			break;
+	}
+
+	/* Update BM driver with number of buffers added to pool */
+	bm_pool->buf_num += i;
+
+	pr_debug("hwpm pool: %d of %d buffers added\n", i, buf_num);
+	spin_unlock_irqrestore(&bm_pool->lock, flags);
+
+	return i;
+}
+EXPORT_SYMBOL_GPL(hwbm_pool_add);
-- 
cgit v0.10.2


From baa11ebc0c7680861f74f8a11dca903e4b421262 Mon Sep 17 00:00:00 2001
From: Gregory CLEMENT <gregory.clement@free-electrons.com>
Date: Mon, 14 Mar 2016 09:39:05 +0100
Subject: net: mvneta: Use the new hwbm framework

Now that the hardware buffer management framework had been introduced,
let's use it.

Tested-by: Sebastian Careba <nitroshift@yahoo.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index ac6605c..62d80fd 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -43,6 +43,7 @@ config MVMDIO
 config MVNETA_BM
 	tristate "Marvell Armada 38x/XP network interface BM support"
 	depends on MVNETA
+	select HWBM
 	---help---
 	  This driver supports auxiliary block of the network
 	  interface units in the Marvell ARMADA XP and ARMADA 38x SoC
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 2847c0c..3d8e7d3 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -30,6 +30,7 @@
 #include <linux/phy.h>
 #include <linux/platform_device.h>
 #include <linux/skbuff.h>
+#include <net/hwbm.h>
 #include "mvneta_bm.h"
 #include <net/ip.h>
 #include <net/ipv6.h>
@@ -1026,11 +1027,12 @@ static int mvneta_bm_port_init(struct platform_device *pdev,
 static void mvneta_bm_update_mtu(struct mvneta_port *pp, int mtu)
 {
 	struct mvneta_bm_pool *bm_pool = pp->pool_long;
+	struct hwbm_pool *hwbm_pool = &bm_pool->hwbm_pool;
 	int num;
 
 	/* Release all buffers from long pool */
 	mvneta_bm_bufs_free(pp->bm_priv, bm_pool, 1 << pp->id);
-	if (bm_pool->buf_num) {
+	if (hwbm_pool->buf_num) {
 		WARN(1, "cannot free all buffers in pool %d\n",
 		     bm_pool->id);
 		goto bm_mtu_err;
@@ -1038,14 +1040,14 @@ static void mvneta_bm_update_mtu(struct mvneta_port *pp, int mtu)
 
 	bm_pool->pkt_size = MVNETA_RX_PKT_SIZE(mtu);
 	bm_pool->buf_size = MVNETA_RX_BUF_SIZE(bm_pool->pkt_size);
-	bm_pool->frag_size = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
-			  SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(bm_pool->pkt_size));
+	hwbm_pool->frag_size = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
+			SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(bm_pool->pkt_size));
 
 	/* Fill entire long pool */
-	num = mvneta_bm_bufs_add(pp->bm_priv, bm_pool, bm_pool->size);
-	if (num != bm_pool->size) {
+	num = hwbm_pool_add(hwbm_pool, hwbm_pool->size, GFP_ATOMIC);
+	if (num != hwbm_pool->size) {
 		WARN(1, "pool %d: %d of %d allocated\n",
-		     bm_pool->id, num, bm_pool->size);
+		     bm_pool->id, num, hwbm_pool->size);
 		goto bm_mtu_err;
 	}
 	mvneta_bm_pool_bufsize_set(pp, bm_pool->buf_size, bm_pool->id);
@@ -2066,14 +2068,14 @@ err_drop_frame:
 		}
 
 		/* Refill processing */
-		err = mvneta_bm_pool_refill(pp->bm_priv, bm_pool);
+		err = hwbm_pool_refill(&bm_pool->hwbm_pool, GFP_ATOMIC);
 		if (err) {
 			netdev_err(dev, "Linux processing - Can't refill\n");
 			rxq->missed++;
 			goto err_drop_frame_ret_pool;
 		}
 
-		frag_size = bm_pool->frag_size;
+		frag_size = bm_pool->hwbm_pool.frag_size;
 
 		skb = build_skb(data, frag_size > PAGE_SIZE ? 0 : frag_size);
 
diff --git a/drivers/net/ethernet/marvell/mvneta_bm.c b/drivers/net/ethernet/marvell/mvneta_bm.c
index 8c968e7..01fccec 100644
--- a/drivers/net/ethernet/marvell/mvneta_bm.c
+++ b/drivers/net/ethernet/marvell/mvneta_bm.c
@@ -10,16 +10,17 @@
  * warranty of any kind, whether express or implied.
  */
 
-#include <linux/kernel.h>
+#include <linux/clk.h>
 #include <linux/genalloc.h>
-#include <linux/platform_device.h>
-#include <linux/netdevice.h>
-#include <linux/skbuff.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
 #include <linux/mbus.h>
 #include <linux/module.h>
-#include <linux/io.h>
+#include <linux/netdevice.h>
 #include <linux/of.h>
-#include <linux/clk.h>
+#include <linux/platform_device.h>
+#include <linux/skbuff.h>
+#include <net/hwbm.h>
 #include "mvneta_bm.h"
 
 #define MVNETA_BM_DRIVER_NAME "mvneta_bm"
@@ -88,17 +89,13 @@ static void mvneta_bm_pool_target_set(struct mvneta_bm *priv, int pool_id,
 	mvneta_bm_write(priv, MVNETA_BM_XBAR_POOL_REG(pool_id), val);
 }
 
-/* Allocate skb for BM pool */
-void *mvneta_buf_alloc(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
-		       dma_addr_t *buf_phys_addr)
+int mvneta_bm_construct(struct hwbm_pool *hwbm_pool, void *buf)
 {
-	void *buf;
+	struct mvneta_bm_pool *bm_pool =
+		(struct mvneta_bm_pool *)hwbm_pool->priv;
+	struct mvneta_bm *priv = bm_pool->priv;
 	dma_addr_t phys_addr;
 
-	buf = mvneta_frag_alloc(bm_pool->frag_size);
-	if (!buf)
-		return NULL;
-
 	/* In order to update buf_cookie field of RX descriptor properly,
 	 * BM hardware expects buf virtual address to be placed in the
 	 * first four bytes of mapped buffer.
@@ -106,75 +103,13 @@ void *mvneta_buf_alloc(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
 	*(u32 *)buf = (u32)buf;
 	phys_addr = dma_map_single(&priv->pdev->dev, buf, bm_pool->buf_size,
 				   DMA_FROM_DEVICE);
-	if (unlikely(dma_mapping_error(&priv->pdev->dev, phys_addr))) {
-		mvneta_frag_free(bm_pool->frag_size, buf);
-		return NULL;
-	}
-	*buf_phys_addr = phys_addr;
-
-	return buf;
-}
-
-/* Refill processing for HW buffer management */
-int mvneta_bm_pool_refill(struct mvneta_bm *priv,
-			  struct mvneta_bm_pool *bm_pool)
-{
-	dma_addr_t buf_phys_addr;
-	void *buf;
-
-	buf = mvneta_buf_alloc(priv, bm_pool, &buf_phys_addr);
-	if (!buf)
+	if (unlikely(dma_mapping_error(&priv->pdev->dev, phys_addr)))
 		return -ENOMEM;
 
-	mvneta_bm_pool_put_bp(priv, bm_pool, buf_phys_addr);
-
+	mvneta_bm_pool_put_bp(priv, bm_pool, phys_addr);
 	return 0;
 }
-EXPORT_SYMBOL_GPL(mvneta_bm_pool_refill);
-
-/* Allocate buffers for the pool */
-int mvneta_bm_bufs_add(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
-		       int buf_num)
-{
-	int err, i;
-
-	if (bm_pool->buf_num == bm_pool->size) {
-		dev_dbg(&priv->pdev->dev, "pool %d already filled\n",
-			bm_pool->id);
-		return bm_pool->buf_num;
-	}
-
-	if (buf_num < 0 ||
-	    (buf_num + bm_pool->buf_num > bm_pool->size)) {
-		dev_err(&priv->pdev->dev,
-			"cannot allocate %d buffers for pool %d\n",
-			buf_num, bm_pool->id);
-		return 0;
-	}
-
-	for (i = 0; i < buf_num; i++) {
-		err = mvneta_bm_pool_refill(priv, bm_pool);
-		if (err < 0)
-			break;
-	}
-
-	/* Update BM driver with number of buffers added to pool */
-	bm_pool->buf_num += i;
-
-	dev_dbg(&priv->pdev->dev,
-		"%s pool %d: pkt_size=%4d, buf_size=%4d, frag_size=%4d\n",
-		bm_pool->type == MVNETA_BM_SHORT ? "short" : "long",
-		bm_pool->id, bm_pool->pkt_size, bm_pool->buf_size,
-		bm_pool->frag_size);
-
-	dev_dbg(&priv->pdev->dev,
-		"%s pool %d: %d of %d buffers added\n",
-		bm_pool->type == MVNETA_BM_SHORT ? "short" : "long",
-		bm_pool->id, i, buf_num);
-
-	return i;
-}
-EXPORT_SYMBOL_GPL(mvneta_bm_bufs_add);
+EXPORT_SYMBOL_GPL(mvneta_bm_construct);
 
 /* Create pool */
 static int mvneta_bm_pool_create(struct mvneta_bm *priv,
@@ -183,8 +118,7 @@ static int mvneta_bm_pool_create(struct mvneta_bm *priv,
 	struct platform_device *pdev = priv->pdev;
 	u8 target_id, attr;
 	int size_bytes, err;
-
-	size_bytes = sizeof(u32) * bm_pool->size;
+	size_bytes = sizeof(u32) * bm_pool->hwbm_pool.size;
 	bm_pool->virt_addr = dma_alloc_coherent(&pdev->dev, size_bytes,
 						&bm_pool->phys_addr,
 						GFP_KERNEL);
@@ -245,11 +179,16 @@ struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
 
 	/* Allocate buffers in case BM pool hasn't been used yet */
 	if (new_pool->type == MVNETA_BM_FREE) {
+		struct hwbm_pool *hwbm_pool = &new_pool->hwbm_pool;
+
+		new_pool->priv = priv;
 		new_pool->type = type;
 		new_pool->buf_size = MVNETA_RX_BUF_SIZE(new_pool->pkt_size);
-		new_pool->frag_size =
+		hwbm_pool->frag_size =
 			SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(new_pool->pkt_size)) +
 			SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+		hwbm_pool->construct = mvneta_bm_construct;
+		hwbm_pool->priv = new_pool;
 
 		/* Create new pool */
 		err = mvneta_bm_pool_create(priv, new_pool);
@@ -260,10 +199,10 @@ struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
 		}
 
 		/* Allocate buffers for this pool */
-		num = mvneta_bm_bufs_add(priv, new_pool, new_pool->size);
-		if (num != new_pool->size) {
+		num = hwbm_pool_add(hwbm_pool, hwbm_pool->size, GFP_ATOMIC);
+		if (num != hwbm_pool->size) {
 			WARN(1, "pool %d: %d of %d allocated\n",
-			     new_pool->id, num, new_pool->size);
+			     new_pool->id, num, hwbm_pool->size);
 			return NULL;
 		}
 	}
@@ -284,7 +223,7 @@ void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
 
 	mvneta_bm_config_set(priv, MVNETA_BM_EMPTY_LIMIT_MASK);
 
-	for (i = 0; i < bm_pool->buf_num; i++) {
+	for (i = 0; i < bm_pool->hwbm_pool.buf_num; i++) {
 		dma_addr_t buf_phys_addr;
 		u32 *vaddr;
 
@@ -303,13 +242,13 @@ void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
 
 		dma_unmap_single(&priv->pdev->dev, buf_phys_addr,
 				 bm_pool->buf_size, DMA_FROM_DEVICE);
-		mvneta_frag_free(bm_pool->frag_size, vaddr);
+		hwbm_buf_free(&bm_pool->hwbm_pool, vaddr);
 	}
 
 	mvneta_bm_config_clear(priv, MVNETA_BM_EMPTY_LIMIT_MASK);
 
 	/* Update BM driver with number of buffers removed from pool */
-	bm_pool->buf_num -= i;
+	bm_pool->hwbm_pool.buf_num -= i;
 }
 EXPORT_SYMBOL_GPL(mvneta_bm_bufs_free);
 
@@ -317,6 +256,7 @@ EXPORT_SYMBOL_GPL(mvneta_bm_bufs_free);
 void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
 			    struct mvneta_bm_pool *bm_pool, u8 port_map)
 {
+	struct hwbm_pool *hwbm_pool = &bm_pool->hwbm_pool;
 	bm_pool->port_map &= ~port_map;
 	if (bm_pool->port_map)
 		return;
@@ -324,11 +264,12 @@ void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
 	bm_pool->type = MVNETA_BM_FREE;
 
 	mvneta_bm_bufs_free(priv, bm_pool, port_map);
-	if (bm_pool->buf_num)
+	if (hwbm_pool->buf_num)
 		WARN(1, "cannot free all buffers in pool %d\n", bm_pool->id);
 
 	if (bm_pool->virt_addr) {
-		dma_free_coherent(&priv->pdev->dev, sizeof(u32) * bm_pool->size,
+		dma_free_coherent(&priv->pdev->dev,
+				  sizeof(u32) * hwbm_pool->size,
 				  bm_pool->virt_addr, bm_pool->phys_addr);
 		bm_pool->virt_addr = NULL;
 	}
@@ -381,10 +322,10 @@ static void mvneta_bm_pools_init(struct mvneta_bm *priv)
 				 MVNETA_BM_POOL_CAP_ALIGN));
 			size = ALIGN(size, MVNETA_BM_POOL_CAP_ALIGN);
 		}
-		bm_pool->size = size;
+		bm_pool->hwbm_pool.size = size;
 
 		mvneta_bm_write(priv, MVNETA_BM_POOL_SIZE_REG(i),
-				bm_pool->size);
+				bm_pool->hwbm_pool.size);
 
 		/* Obtain custom pkt_size from DT */
 		sprintf(prop, "pool%d,pkt-size", i);
diff --git a/drivers/net/ethernet/marvell/mvneta_bm.h b/drivers/net/ethernet/marvell/mvneta_bm.h
index db239e0..e74fd44 100644
--- a/drivers/net/ethernet/marvell/mvneta_bm.h
+++ b/drivers/net/ethernet/marvell/mvneta_bm.h
@@ -108,20 +108,15 @@ struct mvneta_bm {
 };
 
 struct mvneta_bm_pool {
+	struct hwbm_pool hwbm_pool;
 	/* Pool number in the range 0-3 */
 	u8 id;
 	enum mvneta_bm_type type;
 
-	/* Buffer Pointers Pool External (BPPE) size in number of bytes */
-	int size;
-	/* Number of buffers used by this pool */
-	int buf_num;
-	/* Pool buffer size */
-	int buf_size;
 	/* Packet size */
 	int pkt_size;
-	/* Single frag size */
-	u32 frag_size;
+	/* Size of the buffer acces through DMA*/
+	u32 buf_size;
 
 	/* BPPE virtual base address */
 	u32 *virt_addr;
@@ -143,8 +138,7 @@ void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
 			    struct mvneta_bm_pool *bm_pool, u8 port_map);
 void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
 			 u8 port_map);
-int mvneta_bm_bufs_add(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
-		       int buf_num);
+int mvneta_bm_construct(struct hwbm_pool *hwbm_pool, void *buf);
 int mvneta_bm_pool_refill(struct mvneta_bm *priv,
 			  struct mvneta_bm_pool *bm_pool);
 struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
@@ -170,8 +164,7 @@ void mvneta_bm_pool_destroy(struct mvneta_bm *priv,
 			    struct mvneta_bm_pool *bm_pool, u8 port_map) {}
 void mvneta_bm_bufs_free(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
 			 u8 port_map) {}
-int mvneta_bm_bufs_add(struct mvneta_bm *priv, struct mvneta_bm_pool *bm_pool,
-		       int buf_num) { return 0; }
+int mvneta_bm_construct(struct hwbm_pool *hwbm_pool, void *buf) { return 0; }
 int mvneta_bm_pool_refill(struct mvneta_bm *priv,
 			  struct mvneta_bm_pool *bm_pool) {return 0; }
 struct mvneta_bm_pool *mvneta_bm_pool_use(struct mvneta_bm *priv, u8 pool_id,
-- 
cgit v0.10.2


From 9bd160bfa27fa41927dbbce7ee0ea779700e09ef Mon Sep 17 00:00:00 2001
From: Richard Alpe <richard.alpe@ericsson.com>
Date: Mon, 14 Mar 2016 09:43:52 +0100
Subject: tipc: make sure IPv6 header fits in skb headroom

Expand headroom further in order to be able to fit the larger IPv6
header. Prior to this patch this caused a skb under panic for certain
tipc packets when using IPv6 UDP bearer(s).

Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c
index c94f9a1..c9cf2be 100644
--- a/net/tipc/udp_media.c
+++ b/net/tipc/udp_media.c
@@ -53,7 +53,7 @@
 /* IANA assigned UDP port */
 #define UDP_PORT_DEFAULT	6118
 
-#define UDP_MIN_HEADROOM        28
+#define UDP_MIN_HEADROOM        48
 
 /**
  * struct udp_media_addr - IP/UDP addressing information
-- 
cgit v0.10.2


From 5c2e26f6f674ee93719769d024cc72fad3ba9134 Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@cavium.com>
Date: Mon, 14 Mar 2016 16:36:14 +0530
Subject: net: thunderx: Set recevie buffer page usage count in bulk

Instead of calling get_page() for every receive buffer carved out
of page, set page's usage count at the end, to reduce no of atomic
calls.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h
index 092f097..872b22d 100644
--- a/drivers/net/ethernet/cavium/thunder/nic.h
+++ b/drivers/net/ethernet/cavium/thunder/nic.h
@@ -294,6 +294,7 @@ struct nicvf {
 	u32			speed;
 	struct page		*rb_page;
 	u32			rb_page_offset;
+	u16			rb_pageref;
 	bool			rb_alloc_fail;
 	bool			rb_work_scheduled;
 	struct delayed_work	rbdr_work;
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
index 0dd1abf..fa05e34 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
@@ -18,6 +18,15 @@
 #include "q_struct.h"
 #include "nicvf_queues.h"
 
+static void nicvf_get_page(struct nicvf *nic)
+{
+	if (!nic->rb_pageref || !nic->rb_page)
+		return;
+
+	atomic_add(nic->rb_pageref, &nic->rb_page->_count);
+	nic->rb_pageref = 0;
+}
+
 /* Poll a register for a specific value */
 static int nicvf_poll_reg(struct nicvf *nic, int qidx,
 			  u64 reg, int bit_pos, int bits, int val)
@@ -81,16 +90,15 @@ static inline int nicvf_alloc_rcv_buffer(struct nicvf *nic, gfp_t gfp,
 	int order = (PAGE_SIZE <= 4096) ?  PAGE_ALLOC_COSTLY_ORDER : 0;
 
 	/* Check if request can be accomodated in previous allocated page */
-	if (nic->rb_page) {
-		if ((nic->rb_page_offset + buf_len + buf_len) >
-		    (PAGE_SIZE << order)) {
-			nic->rb_page = NULL;
-		} else {
-			nic->rb_page_offset += buf_len;
-			get_page(nic->rb_page);
-		}
+	if (nic->rb_page &&
+	    ((nic->rb_page_offset + buf_len) < (PAGE_SIZE << order))) {
+		nic->rb_pageref++;
+		goto ret;
 	}
 
+	nicvf_get_page(nic);
+	nic->rb_page = NULL;
+
 	/* Allocate a new page */
 	if (!nic->rb_page) {
 		nic->rb_page = alloc_pages(gfp | __GFP_COMP | __GFP_NOWARN,
@@ -102,7 +110,9 @@ static inline int nicvf_alloc_rcv_buffer(struct nicvf *nic, gfp_t gfp,
 		nic->rb_page_offset = 0;
 	}
 
+ret:
 	*rbuf = (u64 *)((u64)page_address(nic->rb_page) + nic->rb_page_offset);
+	nic->rb_page_offset += buf_len;
 
 	return 0;
 }
@@ -158,6 +168,9 @@ static int  nicvf_init_rbdr(struct nicvf *nic, struct rbdr *rbdr,
 		desc = GET_RBDR_DESC(rbdr, idx);
 		desc->buf_addr = virt_to_phys(rbuf) >> NICVF_RCV_BUF_ALIGN;
 	}
+
+	nicvf_get_page(nic);
+
 	return 0;
 }
 
@@ -241,6 +254,8 @@ refill:
 		new_rb++;
 	}
 
+	nicvf_get_page(nic);
+
 	/* make sure all memory stores are done before ringing doorbell */
 	smp_wmb();
 
-- 
cgit v0.10.2


From 1d368790bc2d4111feae6fcc0b3b68d6ae92ff0f Mon Sep 17 00:00:00 2001
From: Sunil Goutham <sgoutham@cavium.com>
Date: Mon, 14 Mar 2016 16:36:15 +0530
Subject: net: thunderx: Adjust nicvf structure to reduce cache misses

Adjusted nicvf structure such that all elements used in hot
path like napi, xmit e.t.c fall into same cache line. This reduced
no of cache misses and resulted in ~2% increase in no of packets
handled on a core.

Also modified elements with :1 notation to boolean, to be
consistent with other element definitions.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h
index 872b22d..83025bb 100644
--- a/drivers/net/ethernet/cavium/thunder/nic.h
+++ b/drivers/net/ethernet/cavium/thunder/nic.h
@@ -272,46 +272,54 @@ struct nicvf {
 	struct nicvf		*pnicvf;
 	struct net_device	*netdev;
 	struct pci_dev		*pdev;
+	void __iomem		*reg_base;
+	struct queue_set	*qs;
+	struct nicvf_cq_poll	*napi[8];
 	u8			vf_id;
-	u8			node;
-	u8			tns_mode:1;
-	u8			sqs_mode:1;
-	u8			loopback_supported:1;
+	u8			sqs_id;
+	bool                    sqs_mode;
 	bool			hw_tso;
-	u16			mtu;
-	struct queue_set	*qs;
+
+	/* Receive buffer alloc */
+	u32			rb_page_offset;
+	u16			rb_pageref;
+	bool			rb_alloc_fail;
+	bool			rb_work_scheduled;
+	struct page		*rb_page;
+	struct delayed_work	rbdr_work;
+	struct tasklet_struct	rbdr_task;
+
+	/* Secondary Qset */
+	u8			sqs_count;
 #define	MAX_SQS_PER_VF_SINGLE_NODE		5
 #define	MAX_SQS_PER_VF				11
-	u8			sqs_id;
-	u8			sqs_count; /* Secondary Qset count */
 	struct nicvf		*snicvf[MAX_SQS_PER_VF];
+
+	/* Queue count */
 	u8			rx_queues;
 	u8			tx_queues;
 	u8			max_queues;
-	void __iomem		*reg_base;
+
+	u8			node;
+	u8			cpi_alg;
+	u16			mtu;
 	bool			link_up;
 	u8			duplex;
 	u32			speed;
-	struct page		*rb_page;
-	u32			rb_page_offset;
-	u16			rb_pageref;
-	bool			rb_alloc_fail;
-	bool			rb_work_scheduled;
-	struct delayed_work	rbdr_work;
-	struct tasklet_struct	rbdr_task;
-	struct tasklet_struct	qs_err_task;
-	struct tasklet_struct	cq_task;
-	struct nicvf_cq_poll	*napi[8];
+	bool			tns_mode;
+	bool			loopback_supported;
 	struct nicvf_rss_info	rss_info;
-	u8			cpi_alg;
+	struct tasklet_struct	qs_err_task;
+	struct work_struct	reset_task;
+
 	/* Interrupt coalescing settings */
 	u32			cq_coalesce_usecs;
-
 	u32			msg_enable;
+
+	/* Stats */
 	struct nicvf_hw_stats   hw_stats;
 	struct nicvf_drv_stats  drv_stats;
 	struct bgx_stats	bgx_stats;
-	struct work_struct	reset_task;
 
 	/* MSI-X  */
 	bool			msix_enabled;
-- 
cgit v0.10.2


From 34b88a68f26a75e4fded796f1a49c40f82234b7d Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Mon, 14 Mar 2016 09:56:35 -0300
Subject: net: Fix use after free in the recvmmsg exit path

The syzkaller fuzzer hit the following use-after-free:

  Call Trace:
   [<ffffffff8175ea0e>] __asan_report_load8_noabort+0x3e/0x40 mm/kasan/report.c:295
   [<ffffffff851cc31a>] __sys_recvmmsg+0x6fa/0x7f0 net/socket.c:2261
   [<     inline     >] SYSC_recvmmsg net/socket.c:2281
   [<ffffffff851cc57f>] SyS_recvmmsg+0x16f/0x180 net/socket.c:2270
   [<ffffffff86332bb6>] entry_SYSCALL_64_fastpath+0x16/0x7a
  arch/x86/entry/entry_64.S:185

And, as Dmitry rightly assessed, that is because we can drop the
reference and then touch it when the underlying recvmsg calls return
some packets and then hit an error, which will make recvmmsg to set
sock->sk->sk_err, oops, fix it.

Reported-and-Tested-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Kostya Serebryany <kcc@google.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Fixes: a2e2725541fa ("net: Introduce recvmmsg socket syscall")
http://lkml.kernel.org/r/20160122211644.GC2470@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/socket.c b/net/socket.c
index c5ddc52..5f77a8e 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2244,31 +2244,31 @@ int __sys_recvmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen,
 		cond_resched();
 	}
 
-out_put:
-	fput_light(sock->file, fput_needed);
-
 	if (err == 0)
-		return datagrams;
+		goto out_put;
 
-	if (datagrams != 0) {
+	if (datagrams == 0) {
+		datagrams = err;
+		goto out_put;
+	}
+
+	/*
+	 * We may return less entries than requested (vlen) if the
+	 * sock is non block and there aren't enough datagrams...
+	 */
+	if (err != -EAGAIN) {
 		/*
-		 * We may return less entries than requested (vlen) if the
-		 * sock is non block and there aren't enough datagrams...
+		 * ... or  if recvmsg returns an error after we
+		 * received some datagrams, where we record the
+		 * error to return on the next call or if the
+		 * app asks about it using getsockopt(SO_ERROR).
 		 */
-		if (err != -EAGAIN) {
-			/*
-			 * ... or  if recvmsg returns an error after we
-			 * received some datagrams, where we record the
-			 * error to return on the next call or if the
-			 * app asks about it using getsockopt(SO_ERROR).
-			 */
-			sock->sk->sk_err = -err;
-		}
-
-		return datagrams;
+		sock->sk->sk_err = -err;
 	}
+out_put:
+	fput_light(sock->file, fput_needed);
 
-	return err;
+	return datagrams;
 }
 
 SYSCALL_DEFINE5(recvmmsg, int, fd, struct mmsghdr __user *, mmsg,
-- 
cgit v0.10.2


From 6aab1a628b99bc1350673e5191a21571626cade3 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:07:10 +0100
Subject: net: mediatek: use dma_addr_t correctly

dma_alloc_coherent() expects a dma_addr_t pointer as its argument,
not an 'unsigned int', and gcc correctly warns about broken
code in the mtk_init_fq_dma function:

drivers/net/ethernet/mediatek/mtk_eth_soc.c: In function 'mtk_init_fq_dma':
drivers/net/ethernet/mediatek/mtk_eth_soc.c:463:13: error: passing argument 3 of 'dma_alloc_coherent' from incompatible pointer type [-Werror=incompatible-pointer-types]

This changes the type of the local variable to dma_addr_t.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index ba3afa5..3e42204 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -453,7 +453,7 @@ static inline void mtk_rx_get_desc(struct mtk_rx_dma *rxd,
 /* the qdma core needs scratch memory to be setup */
 static int mtk_init_fq_dma(struct mtk_eth *eth)
 {
-	unsigned int phy_ring_head, phy_ring_tail;
+	dma_addr_t phy_ring_head, phy_ring_tail;
 	int cnt = MTK_DMA_SIZE;
 	dma_addr_t dma_addr;
 	int i;
-- 
cgit v0.10.2


From a25cdc0dd87d308b508ebea436cb43ee49d54a0e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:07:11 +0100
Subject: net: mediatek: remove incorrect dma_mask assignment

Device drivers should not mess with the DMA mask directly,
but instead call dma_set_mask() etc if needed.

In case of the mtk_eth_soc driver, the mask already gets set
correctly when the device is created, and setting it again
is against the documented API.

This removes the incorrect setting.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 3e42204..87f4177 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -1676,9 +1676,6 @@ static int mtk_probe(struct platform_device *pdev)
 	struct mtk_eth *eth;
 	int err;
 
-	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
-	pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask;
-
 	device_reset(&pdev->dev);
 
 	match = of_match_device(of_mtk_match, &pdev->dev);
-- 
cgit v0.10.2


From 916848ca6f10e1bbf0c2af72ef51e494aaa5c466 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:07:12 +0100
Subject: net: mediatek: check device_reset return code

The device_reset() function may fail, so we have to check
its return value, e.g. to make deferred probing work correctly.
gcc warns about it because of the warn_unused_result attribute:

drivers/net/ethernet/mediatek/mtk_eth_soc.c: In function 'mtk_probe':
drivers/net/ethernet/mediatek/mtk_eth_soc.c:1679:2: error: ignoring return value of 'device_reset', declared with attribute warn_unused_result [-Werror=unused-result]

This adds the trivial error check to propagate the return value
to the generic platform device probe code.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 87f4177..1e6c549 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -1676,7 +1676,9 @@ static int mtk_probe(struct platform_device *pdev)
 	struct mtk_eth *eth;
 	int err;
 
-	device_reset(&pdev->dev);
+	err = device_reset(&pdev->dev);
+	if (err)
+		return err;
 
 	match = of_match_device(of_mtk_match, &pdev->dev);
 	soc = (struct mtk_soc_data *)match->data;
-- 
cgit v0.10.2


From baefd7015cdb304ce6c94f9679d0486c71954766 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:18:34 +0100
Subject: mlx4: add missing braces in verify_qp_parameters

The implementation of QP paravirtualization back in linux-3.7 included
some code that looks very dubious, and gcc-6 has grown smart enough
to warn about it:

drivers/net/ethernet/mellanox/mlx4/resource_tracker.c: In function 'verify_qp_parameters':
drivers/net/ethernet/mellanox/mlx4/resource_tracker.c:3154:5: error: statement is indented as if it were guarded by... [-Werror=misleading-indentation]
     if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH) {
     ^~
drivers/net/ethernet/mellanox/mlx4/resource_tracker.c:3144:4: note: ...this 'if' clause, but it is not
    if (slave != mlx4_master_func_num(dev))

>From looking at the context, I'm reasonably sure that the indentation
is correct but that it should have contained curly braces from the
start, as the update_gid() function in the same patch correctly does.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 54679e148287 ("mlx4: Implement QP paravirtualization and maintain phys_pkey_cache for smp_snoop")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index 25ce1b0..cd9b2b2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -3141,7 +3141,7 @@ static int verify_qp_parameters(struct mlx4_dev *dev,
 		case QP_TRANS_RTS2RTS:
 		case QP_TRANS_SQD2SQD:
 		case QP_TRANS_SQD2RTS:
-			if (slave != mlx4_master_func_num(dev))
+			if (slave != mlx4_master_func_num(dev)) {
 				if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH) {
 					port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
 					if (dev->caps.port_mask[port] != MLX4_PORT_TYPE_IB)
@@ -3160,6 +3160,7 @@ static int verify_qp_parameters(struct mlx4_dev *dev,
 					if (qp_ctx->alt_path.mgid_index >= num_gids)
 						return -EINVAL;
 				}
+			}
 			break;
 		default:
 			break;
-- 
cgit v0.10.2


From e725a66c0202b5f36c2f9d59d26a65c53bbf21f7 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:18:35 +0100
Subject: farsync: fix off-by-one bug in fst_add_one

gcc-6 finds an out of bounds access in the fst_add_one function
when calculating the end of the mmio area:

drivers/net/wan/farsync.c: In function 'fst_add_one':
drivers/net/wan/farsync.c:418:53: error: index 2 denotes an offset greater than size of 'u8[2][8192] {aka unsigned char[2][8192]}' [-Werror=array-bounds]
 #define BUF_OFFSET(X)   (BFM_BASE + offsetof(struct buf_window, X))
                                                     ^
include/linux/compiler-gcc.h:158:21: note: in definition of macro '__compiler_offsetof'
  __builtin_offsetof(a, b)
                     ^
drivers/net/wan/farsync.c:418:37: note: in expansion of macro 'offsetof'
 #define BUF_OFFSET(X)   (BFM_BASE + offsetof(struct buf_window, X))
                                     ^~~~~~~~
drivers/net/wan/farsync.c:2519:36: note: in expansion of macro 'BUF_OFFSET'
                                  + BUF_OFFSET ( txBuffer[i][NUM_TX_BUFFER][0]);
                                    ^~~~~~~~~~

The warning is correct, but not critical because this appears
to be a write-only variable that is set by each WAN driver but
never accessed afterwards.

I'm taking the minimal fix here, using the correct pointer by
pointing 'mem_end' to the last byte inside of the register area
as all other WAN drivers do, rather than the first byte outside of
it. An alternative would be to just remove the mem_end member
entirely.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/wan/farsync.c b/drivers/net/wan/farsync.c
index 44541db..69b994f 100644
--- a/drivers/net/wan/farsync.c
+++ b/drivers/net/wan/farsync.c
@@ -2516,7 +2516,7 @@ fst_add_one(struct pci_dev *pdev, const struct pci_device_id *ent)
                 dev->mem_start   = card->phys_mem
                                  + BUF_OFFSET ( txBuffer[i][0][0]);
                 dev->mem_end     = card->phys_mem
-                                 + BUF_OFFSET ( txBuffer[i][NUM_TX_BUFFER][0]);
+                                 + BUF_OFFSET ( txBuffer[i][NUM_TX_BUFFER - 1][LEN_RX_BUFFER - 1]);
                 dev->base_addr   = card->pci_conf;
                 dev->irq         = card->irq;
 
-- 
cgit v0.10.2


From 83d6f1f15f8cce844b0a131cbc63e444620e48b5 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:18:36 +0100
Subject: ath9k: fix buffer overrun for ar9287

Code that was added back in 2.6.38 has an obvious overflow
when accessing a static array, and at the time it was added
only a code comment was put in front of it as a reminder
to have it reviewed properly.

This has not happened, but gcc-6 now points to the specific
overflow:

drivers/net/wireless/ath/ath9k/eeprom.c: In function 'ath9k_hw_get_gain_boundaries_pdadcs':
drivers/net/wireless/ath/ath9k/eeprom.c:483:44: error: array subscript is above array bounds [-Werror=array-bounds]
     maxPwrT4[i] = data_9287[idxL].pwrPdg[i][4];
                   ~~~~~~~~~~~~~~~~~~~~~~~~~^~~

It turns out that the correct array length exists in the local
'intercepts' variable of this function, so we can just use that
instead of hardcoding '4', so this patch changes all three
instances to use that variable. The other two instances were
already correct, but it's more consistent this way.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 940cd2c12ebf ("ath9k_hw: merge the ar9287 version of ath9k_hw_get_gain_boundaries_pdadcs")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/wireless/ath/ath9k/eeprom.c b/drivers/net/wireless/ath/ath9k/eeprom.c
index 73fb423..a794157 100644
--- a/drivers/net/wireless/ath/ath9k/eeprom.c
+++ b/drivers/net/wireless/ath/ath9k/eeprom.c
@@ -477,10 +477,9 @@ void ath9k_hw_get_gain_boundaries_pdadcs(struct ath_hw *ah,
 
 	if (match) {
 		if (AR_SREV_9287(ah)) {
-			/* FIXME: array overrun? */
 			for (i = 0; i < numXpdGains; i++) {
 				minPwrT4[i] = data_9287[idxL].pwrPdg[i][0];
-				maxPwrT4[i] = data_9287[idxL].pwrPdg[i][4];
+				maxPwrT4[i] = data_9287[idxL].pwrPdg[i][intercepts - 1];
 				ath9k_hw_fill_vpd_table(minPwrT4[i], maxPwrT4[i],
 						data_9287[idxL].pwrPdg[i],
 						data_9287[idxL].vpdPdg[i],
@@ -490,7 +489,7 @@ void ath9k_hw_get_gain_boundaries_pdadcs(struct ath_hw *ah,
 		} else if (eeprom_4k) {
 			for (i = 0; i < numXpdGains; i++) {
 				minPwrT4[i] = data_4k[idxL].pwrPdg[i][0];
-				maxPwrT4[i] = data_4k[idxL].pwrPdg[i][4];
+				maxPwrT4[i] = data_4k[idxL].pwrPdg[i][intercepts - 1];
 				ath9k_hw_fill_vpd_table(minPwrT4[i], maxPwrT4[i],
 						data_4k[idxL].pwrPdg[i],
 						data_4k[idxL].vpdPdg[i],
@@ -500,7 +499,7 @@ void ath9k_hw_get_gain_boundaries_pdadcs(struct ath_hw *ah,
 		} else {
 			for (i = 0; i < numXpdGains; i++) {
 				minPwrT4[i] = data_def[idxL].pwrPdg[i][0];
-				maxPwrT4[i] = data_def[idxL].pwrPdg[i][4];
+				maxPwrT4[i] = data_def[idxL].pwrPdg[i][intercepts - 1];
 				ath9k_hw_fill_vpd_table(minPwrT4[i], maxPwrT4[i],
 						data_def[idxL].pwrPdg[i],
 						data_def[idxL].vpdPdg[i],
-- 
cgit v0.10.2


From 362210e0dff4eb7bb36a9b34dbef3b39d779d95e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:18:37 +0100
Subject: ath9k: fix misleading indentation

A cleanup patch in linux-3.18 moved around some code in the ath9k
driver and left some code to be indented in a misleading way,
made worse by the addition of some new code for p2p mode, as
discovered by a new gcc-6 warning:

drivers/net/wireless/ath/ath9k/init.c: In function 'ath9k_set_hw_capab':
drivers/net/wireless/ath/ath9k/init.c:851:4: warning: statement is indented as if it were guarded by... [-Wmisleading-indentation]
    hw->wiphy->iface_combinations = if_comb;
    ^~
drivers/net/wireless/ath/ath9k/init.c:847:3: note: ...this 'if' clause, but it is not
   if (ath9k_is_chanctx_enabled())
   ^~

The code is in fact correct, but the indentation is not, so I'm
reformatting it as it should have been after the original cleanup.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 499afaccf6f3 ("ath9k: Isolate ath9k_use_chanctx module parameter")
Fixes: eb61f9f623f7 ("ath9k: advertise p2p dev support when chanctx")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c
index d4e0ac9..1c226d6 100644
--- a/drivers/net/wireless/ath/ath9k/init.c
+++ b/drivers/net/wireless/ath/ath9k/init.c
@@ -848,8 +848,8 @@ static void ath9k_set_hw_capab(struct ath_softc *sc, struct ieee80211_hw *hw)
 			hw->wiphy->interface_modes |=
 					BIT(NL80211_IFTYPE_P2P_DEVICE);
 
-			hw->wiphy->iface_combinations = if_comb;
-			hw->wiphy->n_iface_combinations = ARRAY_SIZE(if_comb);
+		hw->wiphy->iface_combinations = if_comb;
+		hw->wiphy->n_iface_combinations = ARRAY_SIZE(if_comb);
 	}
 
 	hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
-- 
cgit v0.10.2


From 8e0cc8c326d99e41468c96fea9785ab78883a281 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:18:38 +0100
Subject: net: caif: fix misleading indentation

gcc points out code that is not indented the way it is
interpreted:

net/caif/cfpkt_skbuff.c: In function 'cfpkt_setlen':
net/caif/cfpkt_skbuff.c:289:4: error: statement is indented as if it were guarded by... [-Werror=misleading-indentation]
    return cfpkt_getlen(pkt);
    ^~~~~~
net/caif/cfpkt_skbuff.c:286:3: note: ...this 'else' clause, but it is not
   else
   ^~~~

It is clear from the context that not returning here would be
a bug, as we'd end up passing a negative length into a function
that takes a u16 length, so it is not missing curly braces
here, and I'm assuming that the indentation is the only part
that's wrong about it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/caif/cfpkt_skbuff.c b/net/caif/cfpkt_skbuff.c
index f6c3b21..59ce1fc 100644
--- a/net/caif/cfpkt_skbuff.c
+++ b/net/caif/cfpkt_skbuff.c
@@ -286,7 +286,7 @@ int cfpkt_setlen(struct cfpkt *pkt, u16 len)
 		else
 			skb_trim(skb, len);
 
-			return cfpkt_getlen(pkt);
+		return cfpkt_getlen(pkt);
 	}
 
 	/* Need to expand SKB */
-- 
cgit v0.10.2


From efc21d95067f658a20e51e24c4c86d68f23b7f7f Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 15:53:57 +0100
Subject: vmxnet3: fix lock imbalance in vmxnet3_tq_xmit()

A recent bug fix rearranged the code in vmxnet3_tq_xmit() in a
way that left the error handling for oversized headers unlock
a lock that had not been taken yet. Gcc warns about the incorrect
use of the 'flags' variable because of that:

drivers/net/vmxnet3/vmxnet3_drv.c: In function 'vmxnet3_tq_xmit.constprop':
include/linux/spinlock.h:246:3: error: 'flags' may be used uninitialized in this function [-Werror=maybe-uninitialized]

This changes the error handling path to 'goto' the end of the function
beyond the lock/unlock pair.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: cec05562fb1d ("vmxnet3: avoid calling pskb_may_pull with interrupts disabled")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index fc895d0..b2348f6 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -1022,14 +1022,16 @@ vmxnet3_tq_xmit(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 		if (ctx.mss) {
 			if (unlikely(ctx.eth_ip_hdr_size + ctx.l4_hdr_size >
 				     VMXNET3_MAX_TX_BUF_SIZE)) {
-				goto hdr_too_big;
+				tq->stats.drop_oversized_hdr++;
+				goto drop_pkt;
 			}
 		} else {
 			if (skb->ip_summed == CHECKSUM_PARTIAL) {
 				if (unlikely(ctx.eth_ip_hdr_size +
 					     skb->csum_offset >
 					     VMXNET3_MAX_CSUM_OFFSET)) {
-					goto hdr_too_big;
+					tq->stats.drop_oversized_hdr++;
+					goto drop_pkt;
 				}
 			}
 		}
@@ -1123,8 +1125,6 @@ vmxnet3_tq_xmit(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 
 	return NETDEV_TX_OK;
 
-hdr_too_big:
-	tq->stats.drop_oversized_hdr++;
 unlock_drop_pkt:
 	spin_unlock_irqrestore(&tq->tx_lock, flags);
 drop_pkt:
-- 
cgit v0.10.2


From a44d6eacdaf56f74fad699af7f4925a5f5ac0e7f Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Mon, 14 Mar 2016 10:52:15 -0700
Subject: tcp: Add RFC4898 tcpEStatsPerfDataSegsOut/In

Per RFC4898, they count segments sent/received
containing a positive length data segment (that includes
retransmission segments carrying data).  Unlike
tcpi_segs_out/in, tcpi_data_segs_out/in excludes segments
carrying no data (e.g. pure ack).

The patch also updates the segs_in in tcp_fastopen_add_skb()
so that segs_in >= data_segs_in property is kept.

Together with retransmission data, tcpi_data_segs_out
gives a better signal on the rxmit rate.

v6: Rebase on the latest net-next

v5: Eric pointed out that checking skb->len is still needed in
tcp_fastopen_add_skb() because skb can carry a FIN without data.
Hence, instead of open coding segs_in and data_segs_in, tcp_segs_in()
helper is used.  Comment is added to the fastopen case to explain why
segs_in has to be reset and tcp_segs_in() has to be called before
__skb_pull().

v4: Add comment to the changes in tcp_fastopen_add_skb()
and also add remark on this case in the commit message.

v3: Add const modifier to the skb parameter in tcp_segs_in()

v2: Rework based on recent fix by Eric:
commit a9d99ce28ed3 ("tcp: fix tcpi_segs_in after connection establishment")

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Cc: Chris Rapier <rapier@psc.edu>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Marcelo Ricardo Leitner <mleitner@redhat.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index bcbf51d..7be9b12 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -158,6 +158,9 @@ struct tcp_sock {
 	u32	segs_in;	/* RFC4898 tcpEStatsPerfSegsIn
 				 * total number of segments in.
 				 */
+	u32	data_segs_in;	/* RFC4898 tcpEStatsPerfDataSegsIn
+				 * total number of data segments in.
+				 */
  	u32	rcv_nxt;	/* What we want to receive next 	*/
 	u32	copied_seq;	/* Head of yet unread data		*/
 	u32	rcv_wup;	/* rcv_nxt on last window update sent	*/
@@ -165,6 +168,9 @@ struct tcp_sock {
 	u32	segs_out;	/* RFC4898 tcpEStatsPerfSegsOut
 				 * The total number of segments sent.
 				 */
+	u32	data_segs_out;	/* RFC4898 tcpEStatsPerfDataSegsOut
+				 * total number of data segments sent.
+				 */
 	u64	bytes_acked;	/* RFC4898 tcpEStatsAppHCThruOctetsAcked
 				 * sum(delta(snd_una)), or how many bytes
 				 * were acked.
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 0302636..c8dbd29 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1840,4 +1840,14 @@ static inline int tcp_inq(struct sock *sk)
 	return answ;
 }
 
+static inline void tcp_segs_in(struct tcp_sock *tp, const struct sk_buff *skb)
+{
+	u16 segs_in;
+
+	segs_in = max_t(u16, 1, skb_shinfo(skb)->gso_segs);
+	tp->segs_in += segs_in;
+	if (skb->len > tcp_hdrlen(skb))
+		tp->data_segs_in += segs_in;
+}
+
 #endif	/* _TCP_H */
diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index fe95446..53e8e3f 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -199,6 +199,8 @@ struct tcp_info {
 
 	__u32	tcpi_notsent_bytes;
 	__u32	tcpi_min_rtt;
+	__u32	tcpi_data_segs_in;	/* RFC4898 tcpEStatsDataSegsIn */
+	__u32	tcpi_data_segs_out;	/* RFC4898 tcpEStatsDataSegsOut */
 };
 
 /* for TCP_MD5SIG socket option */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index a265f00..992b310 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2715,6 +2715,8 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info)
 	info->tcpi_notsent_bytes = max(0, notsent_bytes);
 
 	info->tcpi_min_rtt = tcp_min_rtt(tp);
+	info->tcpi_data_segs_in = tp->data_segs_in;
+	info->tcpi_data_segs_out = tp->data_segs_out;
 }
 EXPORT_SYMBOL_GPL(tcp_get_info);
 
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index fdb286d..4fc0061 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -140,6 +140,14 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
 		return;
 
 	skb_dst_drop(skb);
+	/* segs_in has been initialized to 1 in tcp_create_openreq_child().
+	 * Hence, reset segs_in to 0 before calling tcp_segs_in()
+	 * to avoid double counting.  Also, tcp_segs_in() expects
+	 * skb->len to include the tcp_hdrlen.  Hence, it should
+	 * be called before __skb_pull().
+	 */
+	tp->segs_in = 0;
+	tcp_segs_in(tp, skb);
 	__skb_pull(skb, tcp_hdrlen(skb));
 	skb_set_owner_r(skb, sk);
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 4c8d58d..0b02ef7 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1650,7 +1650,7 @@ process:
 	sk_incoming_cpu_update(sk);
 
 	bh_lock_sock_nested(sk);
-	tcp_sk(sk)->segs_in += max_t(u16, 1, skb_shinfo(skb)->gso_segs);
+	tcp_segs_in(tcp_sk(sk), skb);
 	ret = 0;
 	if (!sock_owned_by_user(sk)) {
 		if (!tcp_prequeue(sk, skb))
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index ae90e4b..acb366d 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -812,7 +812,7 @@ int tcp_child_process(struct sock *parent, struct sock *child,
 	int ret = 0;
 	int state = child->sk_state;
 
-	tcp_sk(child)->segs_in += max_t(u16, 1, skb_shinfo(skb)->gso_segs);
+	tcp_segs_in(tcp_sk(child), skb);
 	if (!sock_owned_by_user(child)) {
 		ret = tcp_rcv_state_process(child, skb);
 		/* Wakeup parent, send SIGIO */
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 7d2c7a4..7d2dc01 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1003,8 +1003,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
 	if (likely(tcb->tcp_flags & TCPHDR_ACK))
 		tcp_event_ack_sent(sk, tcp_skb_pcount(skb));
 
-	if (skb->len != tcp_header_size)
+	if (skb->len != tcp_header_size) {
 		tcp_event_data_sent(tp, sk);
+		tp->data_segs_out += tcp_skb_pcount(skb);
+	}
 
 	if (after(tcb->end_seq, tp->snd_nxt) || tcb->seq == tcb->end_seq)
 		TCP_ADD_STATS(sock_net(sk), TCP_MIB_OUTSEGS,
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 33f2820..9c16565 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1443,7 +1443,7 @@ process:
 	sk_incoming_cpu_update(sk);
 
 	bh_lock_sock_nested(sk);
-	tcp_sk(sk)->segs_in += max_t(u16, 1, skb_shinfo(skb)->gso_segs);
+	tcp_segs_in(tcp_sk(sk), skb);
 	ret = 0;
 	if (!sock_owned_by_user(sk)) {
 		if (!tcp_prequeue(sk, skb))
-- 
cgit v0.10.2


From 387178ec26332e2a35529a716ff3e83a3a457fca Mon Sep 17 00:00:00 2001
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Mon, 14 Mar 2016 10:55:50 -0700
Subject: Documentation: networking: phy.txt: Add missing functions

Some new development in PHYLIB added new function pointers to the struct
phy_driver, document these.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/phy.txt b/Documentation/networking/phy.txt
index e839e7e..7ab9404 100644
--- a/Documentation/networking/phy.txt
+++ b/Documentation/networking/phy.txt
@@ -267,13 +267,23 @@ Writing a PHY driver
    config_intr: Enable or disable interrupts
    remove: Does any driver take-down
    ts_info: Queries about the HW timestamping status
+   match_phy_device: used for Clause 45 capable PHYs to match devices
+   in package and ensure they are compatible
    hwtstamp: Set the PHY HW timestamping configuration
    rxtstamp: Requests a receive timestamp at the PHY level for a 'skb'
    txtsamp: Requests a transmit timestamp at the PHY level for a 'skb'
    set_wol: Enable Wake-on-LAN at the PHY level
    get_wol: Get the Wake-on-LAN status at the PHY level
+   link_change_notify: called to inform the core is about to change the
+   link state, can be used to work around bogus PHY between state changes
    read_mmd_indirect: Read PHY MMD indirect register
    write_mmd_indirect: Write PHY MMD indirect register
+   module_info: Get the size and type of an EEPROM contained in an plug-in
+   module
+   module_eeprom: Get EEPROM information of a plug-in module
+   get_sset_count: Get number of strings sets that get_strings will count
+   get_strings: Get strings from requested objects (statistics)
+   get_stats: Get the extended statistics from the PHY device
 
  Of these, only config_aneg and read_status are required to be
  assigned by the driver code.  The rest are optional.  Also, it is
-- 
cgit v0.10.2


From 7c82a7b998c6c6a2cb8a42aeb1d5d3450e51afb3 Mon Sep 17 00:00:00 2001
From: Igal Liberman <igal.liberman@freescale.com>
Date: Sun, 13 Mar 2016 21:14:43 +0200
Subject: fsl/fman: fix dtsec_set_tx_pause_frames

Fix a bug introduced in e06a03b (fsl/fman: fix the pause_time test)
When pause_time is set to '0' - pause frames are disabled and
there's no need to apply dTSEC-A003 Errata workaround.

Signed-off-by: Igal Liberman <igal.liberman@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
index 7c92eb8..c88918c 100644
--- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c
+++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
@@ -932,15 +932,14 @@ int dtsec_set_tx_pause_frames(struct fman_mac *dtsec,
 	if (!is_init_done(dtsec->dtsec_drv_param))
 		return -EINVAL;
 
-	/* FM_BAD_TX_TS_IN_B_2_B_ERRATA_DTSEC_A003 Errata workaround */
-	if (dtsec->fm_rev_info.major == 2)
-		if (pause_time <= 320) {
+	if (pause_time) {
+		/* FM_BAD_TX_TS_IN_B_2_B_ERRATA_DTSEC_A003 Errata workaround */
+		if (dtsec->fm_rev_info.major == 2 && pause_time <= 320) {
 			pr_warn("pause-time: %d illegal.Should be > 320\n",
 				pause_time);
 			return -EINVAL;
 		}
 
-	if (pause_time) {
 		ptv = ioread32be(&regs->ptv);
 		ptv &= PTV_PTE_MASK;
 		ptv |= pause_time & PTV_PT_MASK;
-- 
cgit v0.10.2


From 0df83e7a6b58674683677f572e4e2361b2a69eea Mon Sep 17 00:00:00 2001
From: Anna-Maria Gleixner <anna-maria@linutronix.de>
Date: Fri, 11 Mar 2016 10:10:23 +0100
Subject: net: mvneta: Add missing hotplug notifier transition

The mvneta_percpu_notifier() hotplug callback lacks handling of the
CPU_DOWN_FAILED case. That means, if CPU_DOWN_PREPARE failes, the
driver is not well configured on the CPU.

Add handling for CPU_DOWN_FAILED[_FROZEN] hotplug notifier transition
to setup the driver.

Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: netdev@vger.kernel.org
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 3d8e7d3..be92668 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3328,6 +3328,8 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 	switch (action) {
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
+	case CPU_DOWN_FAILED:
+	case CPU_DOWN_FAILED_FROZEN:
 		spin_lock(&pp->lock);
 		/* Configuring the driver for a new CPU while the
 		 * driver is stopping is racy, so just avoid it.
-- 
cgit v0.10.2


From 5fc7cf179449502ad4ad67845ded2df94b680de2 Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Fri, 11 Mar 2016 09:53:09 -0800
Subject: net: thunderx: Cleanup PHY probing code.

Remove the call to force the octeon-mdio driver to be loaded.  Allow
the standard driver loading mechanisms to load the PHY drivers, and
use -EPROBE_DEFER to cause the BGX driver to be probed only after the
PHY drivers are available.

Reorder the setting of MAC addresses and PHY probing to allow BGX
LMACs with no attached PHY to still be assigned a MAC address.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
index f8abdff..feed231 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
@@ -978,27 +978,37 @@ static int bgx_init_of_phy(struct bgx *bgx)
 	const char *mac;
 
 	device_for_each_child_node(&bgx->pdev->dev, fwn) {
+		struct phy_device *pd;
 		struct device_node *phy_np;
 		struct device_node *node = to_of_node(fwn);
 
-		/* If it is not an OF node we cannot handle it yet, so
-		 * exit the loop.
+		/* Should always be an OF node.  But if it is not, we
+		 * cannot handle it, so exit the loop.
 		 */
 		if (!node)
 			break;
 
-		phy_np = of_parse_phandle(node, "phy-handle", 0);
-		if (!phy_np)
-			continue;
-
-		bgx->lmac[lmac].phydev = of_phy_find_device(phy_np);
-
 		mac = of_get_mac_address(node);
 		if (mac)
 			ether_addr_copy(bgx->lmac[lmac].mac, mac);
 
 		SET_NETDEV_DEV(&bgx->lmac[lmac].netdev, &bgx->pdev->dev);
 		bgx->lmac[lmac].lmacid = lmac;
+
+		phy_np = of_parse_phandle(node, "phy-handle", 0);
+		/* If there is no phy or defective firmware presents
+		 * this cortina phy, for which there is no driver
+		 * support, ignore it.
+		 */
+		if (phy_np &&
+		    !of_device_is_compatible(phy_np, "cortina,cs4223-slice")) {
+			/* Wait until the phy drivers are available */
+			pd = of_phy_find_device(phy_np);
+			if (!pd)
+				return -EPROBE_DEFER;
+			bgx->lmac[lmac].phydev = pd;
+		}
+
 		lmac++;
 		if (lmac == MAX_LMAC_PER_BGX) {
 			of_node_put(node);
@@ -1032,9 +1042,6 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct bgx *bgx = NULL;
 	u8 lmac;
 
-	/* Load octeon mdio driver */
-	octeon_mdiobus_force_mod_depencency();
-
 	bgx = devm_kzalloc(dev, sizeof(*bgx), GFP_KERNEL);
 	if (!bgx)
 		return -ENOMEM;
-- 
cgit v0.10.2


From 1eefee901fca0208b8a56f20cdc134e2b8638ae7 Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Fri, 11 Mar 2016 09:53:10 -0800
Subject: phy: mdio-octeon: Refactor into two files/modules

A follow-on patch uses PCI probing to find the Thunder MDIO hardware.
In preparation for this, split out the common code into a new file
mdio-cavium.c, which will be used by both the existing OCTEON driver,
and the new Thunder PCI based driver.

As part of the refactoring simplify the struct cavium_mdiobus by
removing fields that are only ever used in the probe function and can
just as well be local variables.

Use readq/writeq in preference to readq_relaxed/writeq_relaxed as the
relaxed form was an optimization for an early chip revision, and the
MDIO drivers are not performance bottlenecks that need optimization in
the first place.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index f0a7702..40faec9 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -183,15 +183,18 @@ config MDIO_GPIO
 	  To compile this driver as a module, choose M here: the module
 	  will be called mdio-gpio.
 
+config MDIO_CAVIUM
+	tristate
+
 config MDIO_OCTEON
-	tristate "Support for MDIO buses on Octeon and ThunderX SOCs"
+	tristate "Support for MDIO buses on Octeon and some ThunderX SOCs"
 	depends on 64BIT
 	depends on HAS_IOMEM
+	select MDIO_CAVIUM
 	help
-
 	  This module provides a driver for the Octeon and ThunderX MDIO
-	  busses. It is required by the Octeon and ThunderX ethernet device
-	  drivers.
+	  buses. It is required by the Octeon and ThunderX ethernet device
+	  drivers on some systems.
 
 config MDIO_SUN4I
 	tristate "Allwinner sun4i MDIO interface support"
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index 680e88f9..041b3d9 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_DP83867_PHY)	+= dp83867.o
 obj-$(CONFIG_STE10XP)		+= ste10Xp.o
 obj-$(CONFIG_MICREL_PHY)	+= micrel.o
 obj-$(CONFIG_MDIO_OCTEON)	+= mdio-octeon.o
+obj-$(CONFIG_MDIO_CAVIUM)	+= mdio-cavium.o
 obj-$(CONFIG_MICREL_KS8995MA)	+= spi_ks8995.o
 obj-$(CONFIG_AT803X_PHY)	+= at803x.o
 obj-$(CONFIG_AMD_PHY)		+= amd.o
diff --git a/drivers/net/phy/mdio-cavium.c b/drivers/net/phy/mdio-cavium.c
new file mode 100644
index 0000000..e796ee1
--- /dev/null
+++ b/drivers/net/phy/mdio-cavium.c
@@ -0,0 +1,149 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2009-2016 Cavium, Inc.
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/phy.h>
+#include <linux/io.h>
+
+#include "mdio-cavium.h"
+
+static void cavium_mdiobus_set_mode(struct cavium_mdiobus *p,
+				    enum cavium_mdiobus_mode m)
+{
+	union cvmx_smix_clk smi_clk;
+
+	if (m == p->mode)
+		return;
+
+	smi_clk.u64 = oct_mdio_readq(p->register_base + SMI_CLK);
+	smi_clk.s.mode = (m == C45) ? 1 : 0;
+	smi_clk.s.preamble = 1;
+	oct_mdio_writeq(smi_clk.u64, p->register_base + SMI_CLK);
+	p->mode = m;
+}
+
+static int cavium_mdiobus_c45_addr(struct cavium_mdiobus *p,
+				   int phy_id, int regnum)
+{
+	union cvmx_smix_cmd smi_cmd;
+	union cvmx_smix_wr_dat smi_wr;
+	int timeout = 1000;
+
+	cavium_mdiobus_set_mode(p, C45);
+
+	smi_wr.u64 = 0;
+	smi_wr.s.dat = regnum & 0xffff;
+	oct_mdio_writeq(smi_wr.u64, p->register_base + SMI_WR_DAT);
+
+	regnum = (regnum >> 16) & 0x1f;
+
+	smi_cmd.u64 = 0;
+	smi_cmd.s.phy_op = 0; /* MDIO_CLAUSE_45_ADDRESS */
+	smi_cmd.s.phy_adr = phy_id;
+	smi_cmd.s.reg_adr = regnum;
+	oct_mdio_writeq(smi_cmd.u64, p->register_base + SMI_CMD);
+
+	do {
+		/* Wait 1000 clocks so we don't saturate the RSL bus
+		 * doing reads.
+		 */
+		__delay(1000);
+		smi_wr.u64 = oct_mdio_readq(p->register_base + SMI_WR_DAT);
+	} while (smi_wr.s.pending && --timeout);
+
+	if (timeout <= 0)
+		return -EIO;
+	return 0;
+}
+
+int cavium_mdiobus_read(struct mii_bus *bus, int phy_id, int regnum)
+{
+	struct cavium_mdiobus *p = bus->priv;
+	union cvmx_smix_cmd smi_cmd;
+	union cvmx_smix_rd_dat smi_rd;
+	unsigned int op = 1; /* MDIO_CLAUSE_22_READ */
+	int timeout = 1000;
+
+	if (regnum & MII_ADDR_C45) {
+		int r = cavium_mdiobus_c45_addr(p, phy_id, regnum);
+
+		if (r < 0)
+			return r;
+
+		regnum = (regnum >> 16) & 0x1f;
+		op = 3; /* MDIO_CLAUSE_45_READ */
+	} else {
+		cavium_mdiobus_set_mode(p, C22);
+	}
+
+	smi_cmd.u64 = 0;
+	smi_cmd.s.phy_op = op;
+	smi_cmd.s.phy_adr = phy_id;
+	smi_cmd.s.reg_adr = regnum;
+	oct_mdio_writeq(smi_cmd.u64, p->register_base + SMI_CMD);
+
+	do {
+		/* Wait 1000 clocks so we don't saturate the RSL bus
+		 * doing reads.
+		 */
+		__delay(1000);
+		smi_rd.u64 = oct_mdio_readq(p->register_base + SMI_RD_DAT);
+	} while (smi_rd.s.pending && --timeout);
+
+	if (smi_rd.s.val)
+		return smi_rd.s.dat;
+	else
+		return -EIO;
+}
+EXPORT_SYMBOL(cavium_mdiobus_read);
+
+int cavium_mdiobus_write(struct mii_bus *bus, int phy_id, int regnum, u16 val)
+{
+	struct cavium_mdiobus *p = bus->priv;
+	union cvmx_smix_cmd smi_cmd;
+	union cvmx_smix_wr_dat smi_wr;
+	unsigned int op = 0; /* MDIO_CLAUSE_22_WRITE */
+	int timeout = 1000;
+
+	if (regnum & MII_ADDR_C45) {
+		int r = cavium_mdiobus_c45_addr(p, phy_id, regnum);
+
+		if (r < 0)
+			return r;
+
+		regnum = (regnum >> 16) & 0x1f;
+		op = 1; /* MDIO_CLAUSE_45_WRITE */
+	} else {
+		cavium_mdiobus_set_mode(p, C22);
+	}
+
+	smi_wr.u64 = 0;
+	smi_wr.s.dat = val;
+	oct_mdio_writeq(smi_wr.u64, p->register_base + SMI_WR_DAT);
+
+	smi_cmd.u64 = 0;
+	smi_cmd.s.phy_op = op;
+	smi_cmd.s.phy_adr = phy_id;
+	smi_cmd.s.reg_adr = regnum;
+	oct_mdio_writeq(smi_cmd.u64, p->register_base + SMI_CMD);
+
+	do {
+		/* Wait 1000 clocks so we don't saturate the RSL bus
+		 * doing reads.
+		 */
+		__delay(1000);
+		smi_wr.u64 = oct_mdio_readq(p->register_base + SMI_WR_DAT);
+	} while (smi_wr.s.pending && --timeout);
+
+	if (timeout <= 0)
+		return -EIO;
+
+	return 0;
+}
+EXPORT_SYMBOL(cavium_mdiobus_write);
diff --git a/drivers/net/phy/mdio-cavium.h b/drivers/net/phy/mdio-cavium.h
new file mode 100644
index 0000000..4bccd45
--- /dev/null
+++ b/drivers/net/phy/mdio-cavium.h
@@ -0,0 +1,119 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2009-2016 Cavium, Inc.
+ */
+
+enum cavium_mdiobus_mode {
+	UNINIT = 0,
+	C22,
+	C45
+};
+
+#define SMI_CMD		0x0
+#define SMI_WR_DAT	0x8
+#define SMI_RD_DAT	0x10
+#define SMI_CLK		0x18
+#define SMI_EN		0x20
+
+#ifdef __BIG_ENDIAN_BITFIELD
+#define OCT_MDIO_BITFIELD_FIELD(field, more)	\
+	field;					\
+	more
+
+#else
+#define OCT_MDIO_BITFIELD_FIELD(field, more)	\
+	more					\
+	field;
+
+#endif
+
+union cvmx_smix_clk {
+	u64 u64;
+	struct cvmx_smix_clk_s {
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_25_63:39,
+	  OCT_MDIO_BITFIELD_FIELD(u64 mode:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_21_23:3,
+	  OCT_MDIO_BITFIELD_FIELD(u64 sample_hi:5,
+	  OCT_MDIO_BITFIELD_FIELD(u64 sample_mode:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_14_14:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 clk_idle:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 preamble:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 sample:4,
+	  OCT_MDIO_BITFIELD_FIELD(u64 phase:8,
+	  ;))))))))))
+	} s;
+};
+
+union cvmx_smix_cmd {
+	u64 u64;
+	struct cvmx_smix_cmd_s {
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_18_63:46,
+	  OCT_MDIO_BITFIELD_FIELD(u64 phy_op:2,
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_13_15:3,
+	  OCT_MDIO_BITFIELD_FIELD(u64 phy_adr:5,
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_5_7:3,
+	  OCT_MDIO_BITFIELD_FIELD(u64 reg_adr:5,
+	  ;))))))
+	} s;
+};
+
+union cvmx_smix_en {
+	u64 u64;
+	struct cvmx_smix_en_s {
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_1_63:63,
+	  OCT_MDIO_BITFIELD_FIELD(u64 en:1,
+	  ;))
+	} s;
+};
+
+union cvmx_smix_rd_dat {
+	u64 u64;
+	struct cvmx_smix_rd_dat_s {
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_18_63:46,
+	  OCT_MDIO_BITFIELD_FIELD(u64 pending:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 val:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 dat:16,
+	  ;))))
+	} s;
+};
+
+union cvmx_smix_wr_dat {
+	u64 u64;
+	struct cvmx_smix_wr_dat_s {
+	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_18_63:46,
+	  OCT_MDIO_BITFIELD_FIELD(u64 pending:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 val:1,
+	  OCT_MDIO_BITFIELD_FIELD(u64 dat:16,
+	  ;))))
+	} s;
+};
+
+struct cavium_mdiobus {
+	struct mii_bus *mii_bus;
+	u64 register_base;
+	enum cavium_mdiobus_mode mode;
+};
+
+#ifdef CONFIG_CAVIUM_OCTEON_SOC
+
+#include <asm/octeon/octeon.h>
+
+static inline void oct_mdio_writeq(u64 val, u64 addr)
+{
+	cvmx_write_csr(addr, val);
+}
+
+static inline u64 oct_mdio_readq(u64 addr)
+{
+	return cvmx_read_csr(addr);
+}
+#else
+#define oct_mdio_writeq(val, addr)	writeq(val, (void *)addr)
+#define oct_mdio_readq(addr)		readq((void *)addr)
+#endif
+
+int cavium_mdiobus_read(struct mii_bus *bus, int phy_id, int regnum);
+int cavium_mdiobus_write(struct mii_bus *bus, int phy_id, int regnum, u16 val);
diff --git a/drivers/net/phy/mdio-octeon.c b/drivers/net/phy/mdio-octeon.c
index 47d4f2f..ab6914f 100644
--- a/drivers/net/phy/mdio-octeon.c
+++ b/drivers/net/phy/mdio-octeon.c
@@ -3,272 +3,26 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2009-2012 Cavium, Inc.
+ * Copyright (C) 2009-2015 Cavium, Inc.
  */
 
 #include <linux/platform_device.h>
 #include <linux/of_address.h>
 #include <linux/of_mdio.h>
-#include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/gfp.h>
 #include <linux/phy.h>
 #include <linux/io.h>
 
-#ifdef CONFIG_CAVIUM_OCTEON_SOC
-#include <asm/octeon/octeon.h>
-#endif
-
-#define DRV_VERSION "1.1"
-#define DRV_DESCRIPTION "Cavium Networks Octeon/ThunderX SMI/MDIO driver"
-
-#define SMI_CMD		0x0
-#define SMI_WR_DAT	0x8
-#define SMI_RD_DAT	0x10
-#define SMI_CLK		0x18
-#define SMI_EN		0x20
-
-#ifdef __BIG_ENDIAN_BITFIELD
-#define OCT_MDIO_BITFIELD_FIELD(field, more)	\
-	field;					\
-	more
-
-#else
-#define OCT_MDIO_BITFIELD_FIELD(field, more)	\
-	more					\
-	field;
-
-#endif
-
-union cvmx_smix_clk {
-	u64 u64;
-	struct cvmx_smix_clk_s {
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_25_63:39,
-	  OCT_MDIO_BITFIELD_FIELD(u64 mode:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_21_23:3,
-	  OCT_MDIO_BITFIELD_FIELD(u64 sample_hi:5,
-	  OCT_MDIO_BITFIELD_FIELD(u64 sample_mode:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_14_14:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 clk_idle:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 preamble:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 sample:4,
-	  OCT_MDIO_BITFIELD_FIELD(u64 phase:8,
-	  ;))))))))))
-	} s;
-};
-
-union cvmx_smix_cmd {
-	u64 u64;
-	struct cvmx_smix_cmd_s {
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_18_63:46,
-	  OCT_MDIO_BITFIELD_FIELD(u64 phy_op:2,
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_13_15:3,
-	  OCT_MDIO_BITFIELD_FIELD(u64 phy_adr:5,
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_5_7:3,
-	  OCT_MDIO_BITFIELD_FIELD(u64 reg_adr:5,
-	  ;))))))
-	} s;
-};
-
-union cvmx_smix_en {
-	u64 u64;
-	struct cvmx_smix_en_s {
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_1_63:63,
-	  OCT_MDIO_BITFIELD_FIELD(u64 en:1,
-	  ;))
-	} s;
-};
-
-union cvmx_smix_rd_dat {
-	u64 u64;
-	struct cvmx_smix_rd_dat_s {
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_18_63:46,
-	  OCT_MDIO_BITFIELD_FIELD(u64 pending:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 val:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 dat:16,
-	  ;))))
-	} s;
-};
-
-union cvmx_smix_wr_dat {
-	u64 u64;
-	struct cvmx_smix_wr_dat_s {
-	  OCT_MDIO_BITFIELD_FIELD(u64 reserved_18_63:46,
-	  OCT_MDIO_BITFIELD_FIELD(u64 pending:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 val:1,
-	  OCT_MDIO_BITFIELD_FIELD(u64 dat:16,
-	  ;))))
-	} s;
-};
-
-enum octeon_mdiobus_mode {
-	UNINIT = 0,
-	C22,
-	C45
-};
-
-struct octeon_mdiobus {
-	struct mii_bus *mii_bus;
-	u64 register_base;
-	resource_size_t mdio_phys;
-	resource_size_t regsize;
-	enum octeon_mdiobus_mode mode;
-};
-
-#ifdef CONFIG_CAVIUM_OCTEON_SOC
-static void oct_mdio_writeq(u64 val, u64 addr)
-{
-	cvmx_write_csr(addr, val);
-}
-
-static u64 oct_mdio_readq(u64 addr)
-{
-	return cvmx_read_csr(addr);
-}
-#else
-#define oct_mdio_writeq(val, addr)	writeq_relaxed(val, (void *)addr)
-#define oct_mdio_readq(addr)		readq_relaxed((void *)addr)
-#endif
-
-static void octeon_mdiobus_set_mode(struct octeon_mdiobus *p,
-				    enum octeon_mdiobus_mode m)
-{
-	union cvmx_smix_clk smi_clk;
-
-	if (m == p->mode)
-		return;
-
-	smi_clk.u64 = oct_mdio_readq(p->register_base + SMI_CLK);
-	smi_clk.s.mode = (m == C45) ? 1 : 0;
-	smi_clk.s.preamble = 1;
-	oct_mdio_writeq(smi_clk.u64, p->register_base + SMI_CLK);
-	p->mode = m;
-}
-
-static int octeon_mdiobus_c45_addr(struct octeon_mdiobus *p,
-				   int phy_id, int regnum)
-{
-	union cvmx_smix_cmd smi_cmd;
-	union cvmx_smix_wr_dat smi_wr;
-	int timeout = 1000;
-
-	octeon_mdiobus_set_mode(p, C45);
-
-	smi_wr.u64 = 0;
-	smi_wr.s.dat = regnum & 0xffff;
-	oct_mdio_writeq(smi_wr.u64, p->register_base + SMI_WR_DAT);
-
-	regnum = (regnum >> 16) & 0x1f;
-
-	smi_cmd.u64 = 0;
-	smi_cmd.s.phy_op = 0; /* MDIO_CLAUSE_45_ADDRESS */
-	smi_cmd.s.phy_adr = phy_id;
-	smi_cmd.s.reg_adr = regnum;
-	oct_mdio_writeq(smi_cmd.u64, p->register_base + SMI_CMD);
-
-	do {
-		/* Wait 1000 clocks so we don't saturate the RSL bus
-		 * doing reads.
-		 */
-		__delay(1000);
-		smi_wr.u64 = oct_mdio_readq(p->register_base + SMI_WR_DAT);
-	} while (smi_wr.s.pending && --timeout);
-
-	if (timeout <= 0)
-		return -EIO;
-	return 0;
-}
-
-static int octeon_mdiobus_read(struct mii_bus *bus, int phy_id, int regnum)
-{
-	struct octeon_mdiobus *p = bus->priv;
-	union cvmx_smix_cmd smi_cmd;
-	union cvmx_smix_rd_dat smi_rd;
-	unsigned int op = 1; /* MDIO_CLAUSE_22_READ */
-	int timeout = 1000;
-
-	if (regnum & MII_ADDR_C45) {
-		int r = octeon_mdiobus_c45_addr(p, phy_id, regnum);
-		if (r < 0)
-			return r;
-
-		regnum = (regnum >> 16) & 0x1f;
-		op = 3; /* MDIO_CLAUSE_45_READ */
-	} else {
-		octeon_mdiobus_set_mode(p, C22);
-	}
-
-
-	smi_cmd.u64 = 0;
-	smi_cmd.s.phy_op = op;
-	smi_cmd.s.phy_adr = phy_id;
-	smi_cmd.s.reg_adr = regnum;
-	oct_mdio_writeq(smi_cmd.u64, p->register_base + SMI_CMD);
-
-	do {
-		/* Wait 1000 clocks so we don't saturate the RSL bus
-		 * doing reads.
-		 */
-		__delay(1000);
-		smi_rd.u64 = oct_mdio_readq(p->register_base + SMI_RD_DAT);
-	} while (smi_rd.s.pending && --timeout);
-
-	if (smi_rd.s.val)
-		return smi_rd.s.dat;
-	else
-		return -EIO;
-}
-
-static int octeon_mdiobus_write(struct mii_bus *bus, int phy_id,
-				int regnum, u16 val)
-{
-	struct octeon_mdiobus *p = bus->priv;
-	union cvmx_smix_cmd smi_cmd;
-	union cvmx_smix_wr_dat smi_wr;
-	unsigned int op = 0; /* MDIO_CLAUSE_22_WRITE */
-	int timeout = 1000;
-
-
-	if (regnum & MII_ADDR_C45) {
-		int r = octeon_mdiobus_c45_addr(p, phy_id, regnum);
-		if (r < 0)
-			return r;
-
-		regnum = (regnum >> 16) & 0x1f;
-		op = 1; /* MDIO_CLAUSE_45_WRITE */
-	} else {
-		octeon_mdiobus_set_mode(p, C22);
-	}
-
-	smi_wr.u64 = 0;
-	smi_wr.s.dat = val;
-	oct_mdio_writeq(smi_wr.u64, p->register_base + SMI_WR_DAT);
-
-	smi_cmd.u64 = 0;
-	smi_cmd.s.phy_op = op;
-	smi_cmd.s.phy_adr = phy_id;
-	smi_cmd.s.reg_adr = regnum;
-	oct_mdio_writeq(smi_cmd.u64, p->register_base + SMI_CMD);
-
-	do {
-		/* Wait 1000 clocks so we don't saturate the RSL bus
-		 * doing reads.
-		 */
-		__delay(1000);
-		smi_wr.u64 = oct_mdio_readq(p->register_base + SMI_WR_DAT);
-	} while (smi_wr.s.pending && --timeout);
-
-	if (timeout <= 0)
-		return -EIO;
-
-	return 0;
-}
+#include "mdio-cavium.h"
 
 static int octeon_mdiobus_probe(struct platform_device *pdev)
 {
-	struct octeon_mdiobus *bus;
+	struct cavium_mdiobus *bus;
 	struct mii_bus *mii_bus;
 	struct resource *res_mem;
+	resource_size_t mdio_phys;
+	resource_size_t regsize;
 	union cvmx_smix_en smi_en;
 	int err = -ENOENT;
 
@@ -284,17 +38,17 @@ static int octeon_mdiobus_probe(struct platform_device *pdev)
 
 	bus = mii_bus->priv;
 	bus->mii_bus = mii_bus;
-	bus->mdio_phys = res_mem->start;
-	bus->regsize = resource_size(res_mem);
+	mdio_phys = res_mem->start;
+	regsize = resource_size(res_mem);
 
-	if (!devm_request_mem_region(&pdev->dev, bus->mdio_phys, bus->regsize,
+	if (!devm_request_mem_region(&pdev->dev, mdio_phys, regsize,
 				     res_mem->name)) {
 		dev_err(&pdev->dev, "request_mem_region failed\n");
 		return -ENXIO;
 	}
 
 	bus->register_base =
-		(u64)devm_ioremap(&pdev->dev, bus->mdio_phys, bus->regsize);
+		(u64)devm_ioremap(&pdev->dev, mdio_phys, regsize);
 	if (!bus->register_base) {
 		dev_err(&pdev->dev, "dev_ioremap failed\n");
 		return -ENOMEM;
@@ -304,13 +58,12 @@ static int octeon_mdiobus_probe(struct platform_device *pdev)
 	smi_en.s.en = 1;
 	oct_mdio_writeq(smi_en.u64, bus->register_base + SMI_EN);
 
-	bus->mii_bus->priv = bus;
-	bus->mii_bus->name = "mdio-octeon";
+	bus->mii_bus->name = KBUILD_MODNAME;
 	snprintf(bus->mii_bus->id, MII_BUS_ID_SIZE, "%llx", bus->register_base);
 	bus->mii_bus->parent = &pdev->dev;
 
-	bus->mii_bus->read = octeon_mdiobus_read;
-	bus->mii_bus->write = octeon_mdiobus_write;
+	bus->mii_bus->read = cavium_mdiobus_read;
+	bus->mii_bus->write = cavium_mdiobus_write;
 
 	platform_set_drvdata(pdev, bus);
 
@@ -318,7 +71,7 @@ static int octeon_mdiobus_probe(struct platform_device *pdev)
 	if (err)
 		goto fail_register;
 
-	dev_info(&pdev->dev, "Version " DRV_VERSION "\n");
+	dev_info(&pdev->dev, "Probed\n");
 
 	return 0;
 fail_register:
@@ -330,7 +83,7 @@ fail_register:
 
 static int octeon_mdiobus_remove(struct platform_device *pdev)
 {
-	struct octeon_mdiobus *bus;
+	struct cavium_mdiobus *bus;
 	union cvmx_smix_en smi_en;
 
 	bus = platform_get_drvdata(pdev);
@@ -352,7 +105,7 @@ MODULE_DEVICE_TABLE(of, octeon_mdiobus_match);
 
 static struct platform_driver octeon_mdiobus_driver = {
 	.driver = {
-		.name		= "mdio-octeon",
+		.name		= KBUILD_MODNAME,
 		.of_match_table = octeon_mdiobus_match,
 	},
 	.probe		= octeon_mdiobus_probe,
@@ -367,7 +120,6 @@ EXPORT_SYMBOL(octeon_mdiobus_force_mod_depencency);
 
 module_platform_driver(octeon_mdiobus_driver);
 
-MODULE_DESCRIPTION(DRV_DESCRIPTION);
-MODULE_VERSION(DRV_VERSION);
+MODULE_DESCRIPTION("Cavium OCTEON MDIO bus driver");
 MODULE_AUTHOR("David Daney");
 MODULE_LICENSE("GPL");
-- 
cgit v0.10.2


From 379d7ac7ca31722a1fb488ae3e98b274c9db568c Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Fri, 11 Mar 2016 09:53:11 -0800
Subject: phy: mdio-thunder: Add driver for Cavium Thunder SoC MDIO buses.

The Cavium Thunder SoCs have multiple MIDO buses that are part of a
single PCI device.  To model this in the device tree we call the PCI
parent device a "cavium,thunder-8890-mdio-nexus", it has several
children, one for each MDIO bus.

The MDIO bus hardware is identical to that found in the OCTEON SoCs,
so we use that code for things that are not part of the PCI driver
probe/remove

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/cavium-mdio.txt b/Documentation/devicetree/bindings/net/cavium-mdio.txt
index 04cb749..020df08 100644
--- a/Documentation/devicetree/bindings/net/cavium-mdio.txt
+++ b/Documentation/devicetree/bindings/net/cavium-mdio.txt
@@ -1,9 +1,12 @@
 * System Management Interface (SMI) / MDIO
 
 Properties:
-- compatible: "cavium,octeon-3860-mdio"
+- compatible: One of:
 
-  Compatibility with all cn3XXX, cn5XXX and cn6XXX SOCs.
+   "cavium,octeon-3860-mdio": Compatibility with all cn3XXX, cn5XXX
+                       and cn6XXX SOCs.
+
+   "cavium,thunder-8890-mdio": Compatibility with all cn8XXX SOCs.
 
 - reg: The base address of the MDIO bus controller register bank.
 
@@ -25,3 +28,57 @@ Example:
 			reg = <0>;
 		};
 	};
+
+
+* System Management Interface (SMI) / MDIO Nexus
+
+  Several mdio buses may be gathered as children of a single PCI
+  device, this PCI device is the nexus of the buses.
+
+Properties:
+
+- compatible: "cavium,thunder-8890-mdio-nexus";
+
+- reg: The PCI device and function numbers of the nexus device.
+
+- #address-cells: Must be <2>.
+
+- #size-cells: Must be <2>.
+
+- ranges: As needed for mapping of the MDIO bus device registers.
+
+- assigned-addresses: As needed for mapping of the MDIO bus device registers.
+
+Example:
+
+        mdio-nexus@1,3 {
+                compatible = "cavium,thunder-8890-mdio-nexus";
+                #address-cells = <2>;
+                #size-cells = <2>;
+                reg = <0x0b00 0 0 0 0>; /* DEVFN = 0x0b (1:3) */
+                assigned-addresses = <0x03000000 0x87e0 0x05000000 0x0 0x800000>;
+                ranges = <0x87e0 0x05000000 0x03000000 0x87e0 0x05000000 0x0 0x800000>;
+
+                mdio0@87e0,05003800 {
+                        compatible = "cavium,thunder-8890-mdio";
+                        #address-cells = <1>;
+                        #size-cells = <0>;
+                        reg = <0x87e0 0x05003800 0x0 0x30>;
+
+                        ethernet-phy@0 {
+                                ...
+                                reg = <0>;
+                        };
+                };
+                mdio0@87e0,05003880 {
+                        compatible = "cavium,thunder-8890-mdio";
+                        #address-cells = <1>;
+                        #size-cells = <0>;
+                        reg = <0x87e0 0x05003880 0x0 0x30>;
+
+                        ethernet-phy@0 {
+                                ...
+                                reg = <0>;
+                        };
+                };
+        };
diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 40faec9..075a4cc 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -196,6 +196,17 @@ config MDIO_OCTEON
 	  buses. It is required by the Octeon and ThunderX ethernet device
 	  drivers on some systems.
 
+config MDIO_THUNDER
+	tristate "Support for MDIO buses on on ThunderX SOCs"
+	depends on 64BIT
+	depends on PCI
+	select MDIO_CAVIUM
+	help
+	  This driver supports the MDIO interfaces found on Cavium
+	  ThunderX SoCs when the MDIO bus device appears on as a PCI
+	  device.
+
+
 config MDIO_SUN4I
 	tristate "Allwinner sun4i MDIO interface support"
 	depends on ARCH_SUNXI
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index 041b3d9..fcdbb92 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_DP83867_PHY)	+= dp83867.o
 obj-$(CONFIG_STE10XP)		+= ste10Xp.o
 obj-$(CONFIG_MICREL_PHY)	+= micrel.o
 obj-$(CONFIG_MDIO_OCTEON)	+= mdio-octeon.o
+obj-$(CONFIG_MDIO_THUNDER)	+= mdio-thunder.o
 obj-$(CONFIG_MDIO_CAVIUM)	+= mdio-cavium.o
 obj-$(CONFIG_MICREL_KS8995MA)	+= spi_ks8995.o
 obj-$(CONFIG_AT803X_PHY)	+= at803x.o
diff --git a/drivers/net/phy/mdio-thunder.c b/drivers/net/phy/mdio-thunder.c
new file mode 100644
index 0000000..5646169
--- /dev/null
+++ b/drivers/net/phy/mdio-thunder.c
@@ -0,0 +1,154 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2009-2016 Cavium, Inc.
+ */
+
+#include <linux/of_address.h>
+#include <linux/of_mdio.h>
+#include <linux/module.h>
+#include <linux/gfp.h>
+#include <linux/phy.h>
+#include <linux/io.h>
+#include <linux/acpi.h>
+#include <linux/pci.h>
+
+#include "mdio-cavium.h"
+
+struct thunder_mdiobus_nexus {
+	void __iomem *bar0;
+	struct cavium_mdiobus *buses[4];
+};
+
+static int thunder_mdiobus_pci_probe(struct pci_dev *pdev,
+				     const struct pci_device_id *ent)
+{
+	struct device_node *node;
+	struct fwnode_handle *fwn;
+	struct thunder_mdiobus_nexus *nexus;
+	int err;
+	int i;
+
+	nexus = devm_kzalloc(&pdev->dev, sizeof(*nexus), GFP_KERNEL);
+	if (!nexus)
+		return -ENOMEM;
+
+	pci_set_drvdata(pdev, nexus);
+
+	err = pcim_enable_device(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to enable PCI device\n");
+		pci_set_drvdata(pdev, NULL);
+		return err;
+	}
+
+	err = pci_request_regions(pdev, KBUILD_MODNAME);
+	if (err) {
+		dev_err(&pdev->dev, "pci_request_regions failed\n");
+		goto err_disable_device;
+	}
+
+	nexus->bar0 = pcim_iomap(pdev, 0, pci_resource_len(pdev, 0));
+	if (!nexus->bar0) {
+		err = -ENOMEM;
+		goto err_release_regions;
+	}
+
+	i = 0;
+	device_for_each_child_node(&pdev->dev, fwn) {
+		struct resource r;
+		struct mii_bus *mii_bus;
+		struct cavium_mdiobus *bus;
+		union cvmx_smix_en smi_en;
+
+		/* If it is not an OF node we cannot handle it yet, so
+		 * exit the loop.
+		 */
+		node = to_of_node(fwn);
+		if (!node)
+			break;
+
+		err = of_address_to_resource(node, 0, &r);
+		if (err) {
+			dev_err(&pdev->dev,
+				"Couldn't translate address for \"%s\"\n",
+				node->name);
+			break;
+		}
+
+		mii_bus = devm_mdiobus_alloc_size(&pdev->dev, sizeof(*bus));
+		if (!mii_bus)
+			break;
+		bus = mii_bus->priv;
+		bus->mii_bus = mii_bus;
+
+		nexus->buses[i] = bus;
+		i++;
+
+		bus->register_base = (u64)nexus->bar0 +
+			r.start - pci_resource_start(pdev, 0);
+
+		smi_en.u64 = 0;
+		smi_en.s.en = 1;
+		oct_mdio_writeq(smi_en.u64, bus->register_base + SMI_EN);
+		bus->mii_bus->name = KBUILD_MODNAME;
+		snprintf(bus->mii_bus->id, MII_BUS_ID_SIZE, "%llx", r.start);
+		bus->mii_bus->parent = &pdev->dev;
+		bus->mii_bus->read = cavium_mdiobus_read;
+		bus->mii_bus->write = cavium_mdiobus_write;
+
+		err = of_mdiobus_register(bus->mii_bus, node);
+		if (err)
+			dev_err(&pdev->dev, "of_mdiobus_register failed\n");
+
+		dev_info(&pdev->dev, "Added bus at %llx\n", r.start);
+		if (i >= ARRAY_SIZE(nexus->buses))
+			break;
+	}
+	return 0;
+
+err_release_regions:
+	pci_release_regions(pdev);
+
+err_disable_device:
+	pci_set_drvdata(pdev, NULL);
+	return err;
+}
+
+static void thunder_mdiobus_pci_remove(struct pci_dev *pdev)
+{
+	int i;
+	struct thunder_mdiobus_nexus *nexus = pci_get_drvdata(pdev);
+
+	for (i = 0; i < ARRAY_SIZE(nexus->buses); i++) {
+		struct cavium_mdiobus *bus = nexus->buses[i];
+
+		if (!bus)
+			continue;
+
+		mdiobus_unregister(bus->mii_bus);
+		mdiobus_free(bus->mii_bus);
+		oct_mdio_writeq(0, bus->register_base + SMI_EN);
+	}
+	pci_set_drvdata(pdev, NULL);
+}
+
+static const struct pci_device_id thunder_mdiobus_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, 0xa02b) },
+	{ 0, } /* End of table. */
+};
+MODULE_DEVICE_TABLE(pci, thunder_mdiobus_id_table);
+
+static struct pci_driver thunder_mdiobus_driver = {
+	.name = KBUILD_MODNAME,
+	.id_table = thunder_mdiobus_id_table,
+	.probe = thunder_mdiobus_pci_probe,
+	.remove = thunder_mdiobus_pci_remove,
+};
+
+module_pci_driver(thunder_mdiobus_driver);
+
+MODULE_DESCRIPTION("Cavium ThunderX MDIO bus driver");
+MODULE_LICENSE("GPL");
-- 
cgit v0.10.2


From a3478bae7a2751500386816ca567af4fc0f5eb10 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 13 Mar 2016 00:33:13 +0300
Subject: of_mdio: mdio_device_create() never returns NULL

mdio_device_create() never returns NULL, thus checking for it in
of_mdiobus_register_device() is pointless...

Suggested-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 5e78382..3bc68ac 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -98,7 +98,7 @@ static int of_mdiobus_register_device(struct mii_bus *mdio,
 	int rc;
 
 	mdiodev = mdio_device_create(mdio, addr);
-	if (!mdiodev || IS_ERR(mdiodev))
+	if (IS_ERR(mdiodev))
 		return 1;
 
 	/* Associate the OF node with the device structure so it
-- 
cgit v0.10.2


From ac044b902e50ba50713e257e331bc0e0c1c7d4d9 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 13 Mar 2016 00:34:02 +0300
Subject: of_mdio: use IS_ERR_OR_NULL()

IS_ERR_OR_NULL() is open coded in of_mdiobus_register_phy(), so just call
it directly...

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 3bc68ac..52d9b14 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -56,7 +56,7 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio, struct device_node *chi
 		phy = phy_device_create(mdio, addr, phy_id, 0, NULL);
 	else
 		phy = get_phy_device(mdio, addr, is_c45);
-	if (!phy || IS_ERR(phy))
+	if (IS_ERR_OR_NULL(phy))
 		return 1;
 
 	rc = irq_of_parse_and_map(child, 0);
-- 
cgit v0.10.2


From 5189b1d82f1ce6c2749fa3499d28ffd3f5075543 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 13 Mar 2016 00:34:58 +0300
Subject: of_mdio: use PTR_ERR_OR_ZERO()

PTR_ERR_OR_ZERO() is open coded in of_phy_register_fixed_link(), so just
call it directly...

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 52d9b14..8453f08 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -412,7 +412,7 @@ int of_phy_register_fixed_link(struct device_node *np)
 		if (strcmp(managed, "in-band-status") == 0) {
 			/* status is zeroed, namely its .link member */
 			phy = fixed_phy_register(PHY_POLL, &status, -1, np);
-			return IS_ERR(phy) ? PTR_ERR(phy) : 0;
+			return PTR_ERR_OR_ZERO(phy);
 		}
 	}
 
@@ -434,7 +434,7 @@ int of_phy_register_fixed_link(struct device_node *np)
 			return -EPROBE_DEFER;
 
 		phy = fixed_phy_register(PHY_POLL, &status, link_gpio, np);
-		return IS_ERR(phy) ? PTR_ERR(phy) : 0;
+		return PTR_ERR_OR_ZERO(phy);
 	}
 
 	/* Old binding */
@@ -446,7 +446,7 @@ int of_phy_register_fixed_link(struct device_node *np)
 		status.pause = be32_to_cpu(fixed_link_prop[3]);
 		status.asym_pause = be32_to_cpu(fixed_link_prop[4]);
 		phy = fixed_phy_register(PHY_POLL, &status, -1, np);
-		return IS_ERR(phy) ? PTR_ERR(phy) : 0;
+		return PTR_ERR_OR_ZERO(phy);
 	}
 
 	return -ENODEV;
-- 
cgit v0.10.2


From ca3dfa51e67cd41f6514b84a88bc101e8b1a139a Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 12 Mar 2016 00:01:36 +0100
Subject: dsa: Rename mv88e6123_61_65 to mv88e6123 to be consistent

All the drivers support multiple chips, but mv88e6123_61_65 is the
only one that reflects this in its naming. Change it to be consistent
with the other drivers.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/configs/multi_v5_defconfig b/arch/arm/configs/multi_v5_defconfig
index 1f9ca47..afb1f62 100644
--- a/arch/arm/configs/multi_v5_defconfig
+++ b/arch/arm/configs/multi_v5_defconfig
@@ -91,7 +91,7 @@ CONFIG_SATA_MV=y
 CONFIG_NETDEVICES=y
 CONFIG_NET_DSA_MV88E6060=y
 CONFIG_NET_DSA_MV88E6131=y
-CONFIG_NET_DSA_MV88E6123_61_65=y
+CONFIG_NET_DSA_MV88E6123=y
 CONFIG_NET_DSA_MV88E6171=y
 CONFIG_NET_DSA_MV88E6352=y
 CONFIG_MV643XX_ETH=y
diff --git a/arch/arm/configs/mvebu_v5_defconfig b/arch/arm/configs/mvebu_v5_defconfig
index af29780..6c4c540 100644
--- a/arch/arm/configs/mvebu_v5_defconfig
+++ b/arch/arm/configs/mvebu_v5_defconfig
@@ -92,7 +92,7 @@ CONFIG_SATA_MV=y
 CONFIG_NETDEVICES=y
 CONFIG_NET_DSA_MV88E6060=y
 CONFIG_NET_DSA_MV88E6131=y
-CONFIG_NET_DSA_MV88E6123_61_65=y
+CONFIG_NET_DSA_MV88E6123=y
 CONFIG_NET_DSA_MV88E6171=y
 CONFIG_NET_DSA_MV88E6352=y
 CONFIG_MV643XX_ETH=y
diff --git a/arch/arm/configs/orion5x_defconfig b/arch/arm/configs/orion5x_defconfig
index 5876ce7..6a5bc27 100644
--- a/arch/arm/configs/orion5x_defconfig
+++ b/arch/arm/configs/orion5x_defconfig
@@ -86,7 +86,7 @@ CONFIG_SATA_MV=y
 CONFIG_NETDEVICES=y
 CONFIG_MII=y
 CONFIG_NET_DSA_MV88E6131=y
-CONFIG_NET_DSA_MV88E6123_61_65=y
+CONFIG_NET_DSA_MV88E6123=y
 CONFIG_MV643XX_ETH=y
 CONFIG_MARVELL_PHY=y
 # CONFIG_INPUT_MOUSEDEV is not set
diff --git a/arch/tile/configs/tilegx_defconfig b/arch/tile/configs/tilegx_defconfig
index 37dc936..984fa00 100644
--- a/arch/tile/configs/tilegx_defconfig
+++ b/arch/tile/configs/tilegx_defconfig
@@ -222,7 +222,7 @@ CONFIG_TUN=y
 CONFIG_VETH=m
 CONFIG_NET_DSA_MV88E6060=y
 CONFIG_NET_DSA_MV88E6131=y
-CONFIG_NET_DSA_MV88E6123_61_65=y
+CONFIG_NET_DSA_MV88E6123=y
 CONFIG_SKY2=y
 CONFIG_PTP_1588_CLOCK_TILEGX=y
 # CONFIG_WLAN is not set
diff --git a/arch/tile/configs/tilepro_defconfig b/arch/tile/configs/tilepro_defconfig
index 76a2781..71ad9f7 100644
--- a/arch/tile/configs/tilepro_defconfig
+++ b/arch/tile/configs/tilepro_defconfig
@@ -341,7 +341,7 @@ CONFIG_TUN=y
 CONFIG_VETH=m
 CONFIG_NET_DSA_MV88E6060=y
 CONFIG_NET_DSA_MV88E6131=y
-CONFIG_NET_DSA_MV88E6123_61_65=y
+CONFIG_NET_DSA_MV88E6123=y
 # CONFIG_NET_VENDOR_3COM is not set
 CONFIG_E1000E=y
 # CONFIG_WLAN is not set
diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index 4c483d9..90ba003 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -27,7 +27,7 @@ config NET_DSA_MV88E6131
 	  This enables support for the Marvell 88E6085/6095/6095F/6131
 	  ethernet switch chips.
 
-config NET_DSA_MV88E6123_61_65
+config NET_DSA_MV88E6123
 	tristate "Marvell 88E6123/6161/6165 ethernet switch chip support"
 	depends on NET_DSA
 	select NET_DSA_MV88E6XXX
diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile
index e2d51c4..a6e0993 100644
--- a/drivers/net/dsa/Makefile
+++ b/drivers/net/dsa/Makefile
@@ -1,8 +1,8 @@
 obj-$(CONFIG_NET_DSA_MV88E6060) += mv88e6060.o
 obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx_drv.o
 mv88e6xxx_drv-y += mv88e6xxx.o
-ifdef CONFIG_NET_DSA_MV88E6123_61_65
-mv88e6xxx_drv-y += mv88e6123_61_65.o
+ifdef CONFIG_NET_DSA_MV88E6123
+mv88e6xxx_drv-y += mv88e6123.o
 endif
 ifdef CONFIG_NET_DSA_MV88E6131
 mv88e6xxx_drv-y += mv88e6131.o
diff --git a/drivers/net/dsa/mv88e6123.c b/drivers/net/dsa/mv88e6123.c
new file mode 100644
index 0000000..69a6f79
--- /dev/null
+++ b/drivers/net/dsa/mv88e6123.c
@@ -0,0 +1,124 @@
+/*
+ * net/dsa/mv88e6123_61_65.c - Marvell 88e6123/6161/6165 switch chip support
+ * Copyright (c) 2008-2009 Marvell Semiconductor
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/delay.h>
+#include <linux/jiffies.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include <net/dsa.h>
+#include "mv88e6xxx.h"
+
+static const struct mv88e6xxx_switch_id mv88e6123_table[] = {
+	{ PORT_SWITCH_ID_6123, "Marvell 88E6123" },
+	{ PORT_SWITCH_ID_6123_A1, "Marvell 88E6123 (A1)" },
+	{ PORT_SWITCH_ID_6123_A2, "Marvell 88E6123 (A2)" },
+	{ PORT_SWITCH_ID_6161, "Marvell 88E6161" },
+	{ PORT_SWITCH_ID_6161_A1, "Marvell 88E6161 (A1)" },
+	{ PORT_SWITCH_ID_6161_A2, "Marvell 88E6161 (A2)" },
+	{ PORT_SWITCH_ID_6165, "Marvell 88E6165" },
+	{ PORT_SWITCH_ID_6165_A1, "Marvell 88E6165 (A1)" },
+	{ PORT_SWITCH_ID_6165_A2, "Marvell 88e6165 (A2)" },
+};
+
+static char *mv88e6123_probe(struct device *host_dev, int sw_addr)
+{
+	return mv88e6xxx_lookup_name(host_dev, sw_addr, mv88e6123_table,
+				     ARRAY_SIZE(mv88e6123_table));
+}
+
+static int mv88e6123_setup_global(struct dsa_switch *ds)
+{
+	u32 upstream_port = dsa_upstream_port(ds);
+	int ret;
+	u32 reg;
+
+	ret = mv88e6xxx_setup_global(ds);
+	if (ret)
+		return ret;
+
+	/* Disable the PHY polling unit (since there won't be any
+	 * external PHYs to poll), don't discard packets with
+	 * excessive collisions, and mask all interrupt sources.
+	 */
+	REG_WRITE(REG_GLOBAL, GLOBAL_CONTROL, 0x0000);
+
+	/* Configure the upstream port, and configure the upstream
+	 * port as the port to which ingress and egress monitor frames
+	 * are to be sent.
+	 */
+	reg = upstream_port << GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT |
+		upstream_port << GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT |
+		upstream_port << GLOBAL_MONITOR_CONTROL_ARP_SHIFT;
+	REG_WRITE(REG_GLOBAL, GLOBAL_MONITOR_CONTROL, reg);
+
+	/* Disable remote management for now, and set the switch's
+	 * DSA device number.
+	 */
+	REG_WRITE(REG_GLOBAL, GLOBAL_CONTROL_2, ds->index & 0x1f);
+
+	return 0;
+}
+
+static int mv88e6123_setup(struct dsa_switch *ds)
+{
+	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	int ret;
+
+	ret = mv88e6xxx_setup_common(ds);
+	if (ret < 0)
+		return ret;
+
+	switch (ps->id) {
+	case PORT_SWITCH_ID_6123:
+		ps->num_ports = 3;
+		break;
+	case PORT_SWITCH_ID_6161:
+	case PORT_SWITCH_ID_6165:
+		ps->num_ports = 6;
+		break;
+	default:
+		return -ENODEV;
+	}
+
+	ret = mv88e6xxx_switch_reset(ds, false);
+	if (ret < 0)
+		return ret;
+
+	ret = mv88e6123_setup_global(ds);
+	if (ret < 0)
+		return ret;
+
+	return mv88e6xxx_setup_ports(ds);
+}
+
+struct dsa_switch_driver mv88e6123_switch_driver = {
+	.tag_protocol		= DSA_TAG_PROTO_EDSA,
+	.priv_size		= sizeof(struct mv88e6xxx_priv_state),
+	.probe			= mv88e6123_probe,
+	.setup			= mv88e6123_setup,
+	.set_addr		= mv88e6xxx_set_addr_indirect,
+	.phy_read		= mv88e6xxx_phy_read,
+	.phy_write		= mv88e6xxx_phy_write,
+	.get_strings		= mv88e6xxx_get_strings,
+	.get_ethtool_stats	= mv88e6xxx_get_ethtool_stats,
+	.get_sset_count		= mv88e6xxx_get_sset_count,
+	.adjust_link		= mv88e6xxx_adjust_link,
+#ifdef CONFIG_NET_DSA_HWMON
+	.get_temp		= mv88e6xxx_get_temp,
+#endif
+	.get_regs_len		= mv88e6xxx_get_regs_len,
+	.get_regs		= mv88e6xxx_get_regs,
+};
+
+MODULE_ALIAS("platform:mv88e6123");
+MODULE_ALIAS("platform:mv88e6161");
+MODULE_ALIAS("platform:mv88e6165");
diff --git a/drivers/net/dsa/mv88e6123_61_65.c b/drivers/net/dsa/mv88e6123_61_65.c
deleted file mode 100644
index d4fcf45..0000000
--- a/drivers/net/dsa/mv88e6123_61_65.c
+++ /dev/null
@@ -1,124 +0,0 @@
-/*
- * net/dsa/mv88e6123_61_65.c - Marvell 88e6123/6161/6165 switch chip support
- * Copyright (c) 2008-2009 Marvell Semiconductor
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#include <linux/delay.h>
-#include <linux/jiffies.h>
-#include <linux/list.h>
-#include <linux/module.h>
-#include <linux/netdevice.h>
-#include <linux/phy.h>
-#include <net/dsa.h>
-#include "mv88e6xxx.h"
-
-static const struct mv88e6xxx_switch_id mv88e6123_61_65_table[] = {
-	{ PORT_SWITCH_ID_6123, "Marvell 88E6123" },
-	{ PORT_SWITCH_ID_6123_A1, "Marvell 88E6123 (A1)" },
-	{ PORT_SWITCH_ID_6123_A2, "Marvell 88E6123 (A2)" },
-	{ PORT_SWITCH_ID_6161, "Marvell 88E6161" },
-	{ PORT_SWITCH_ID_6161_A1, "Marvell 88E6161 (A1)" },
-	{ PORT_SWITCH_ID_6161_A2, "Marvell 88E6161 (A2)" },
-	{ PORT_SWITCH_ID_6165, "Marvell 88E6165" },
-	{ PORT_SWITCH_ID_6165_A1, "Marvell 88E6165 (A1)" },
-	{ PORT_SWITCH_ID_6165_A2, "Marvell 88e6165 (A2)" },
-};
-
-static char *mv88e6123_61_65_probe(struct device *host_dev, int sw_addr)
-{
-	return mv88e6xxx_lookup_name(host_dev, sw_addr, mv88e6123_61_65_table,
-				     ARRAY_SIZE(mv88e6123_61_65_table));
-}
-
-static int mv88e6123_61_65_setup_global(struct dsa_switch *ds)
-{
-	u32 upstream_port = dsa_upstream_port(ds);
-	int ret;
-	u32 reg;
-
-	ret = mv88e6xxx_setup_global(ds);
-	if (ret)
-		return ret;
-
-	/* Disable the PHY polling unit (since there won't be any
-	 * external PHYs to poll), don't discard packets with
-	 * excessive collisions, and mask all interrupt sources.
-	 */
-	REG_WRITE(REG_GLOBAL, GLOBAL_CONTROL, 0x0000);
-
-	/* Configure the upstream port, and configure the upstream
-	 * port as the port to which ingress and egress monitor frames
-	 * are to be sent.
-	 */
-	reg = upstream_port << GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT |
-		upstream_port << GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT |
-		upstream_port << GLOBAL_MONITOR_CONTROL_ARP_SHIFT;
-	REG_WRITE(REG_GLOBAL, GLOBAL_MONITOR_CONTROL, reg);
-
-	/* Disable remote management for now, and set the switch's
-	 * DSA device number.
-	 */
-	REG_WRITE(REG_GLOBAL, GLOBAL_CONTROL_2, ds->index & 0x1f);
-
-	return 0;
-}
-
-static int mv88e6123_61_65_setup(struct dsa_switch *ds)
-{
-	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
-	int ret;
-
-	ret = mv88e6xxx_setup_common(ds);
-	if (ret < 0)
-		return ret;
-
-	switch (ps->id) {
-	case PORT_SWITCH_ID_6123:
-		ps->num_ports = 3;
-		break;
-	case PORT_SWITCH_ID_6161:
-	case PORT_SWITCH_ID_6165:
-		ps->num_ports = 6;
-		break;
-	default:
-		return -ENODEV;
-	}
-
-	ret = mv88e6xxx_switch_reset(ds, false);
-	if (ret < 0)
-		return ret;
-
-	ret = mv88e6123_61_65_setup_global(ds);
-	if (ret < 0)
-		return ret;
-
-	return mv88e6xxx_setup_ports(ds);
-}
-
-struct dsa_switch_driver mv88e6123_61_65_switch_driver = {
-	.tag_protocol		= DSA_TAG_PROTO_EDSA,
-	.priv_size		= sizeof(struct mv88e6xxx_priv_state),
-	.probe			= mv88e6123_61_65_probe,
-	.setup			= mv88e6123_61_65_setup,
-	.set_addr		= mv88e6xxx_set_addr_indirect,
-	.phy_read		= mv88e6xxx_phy_read,
-	.phy_write		= mv88e6xxx_phy_write,
-	.get_strings		= mv88e6xxx_get_strings,
-	.get_ethtool_stats	= mv88e6xxx_get_ethtool_stats,
-	.get_sset_count		= mv88e6xxx_get_sset_count,
-	.adjust_link		= mv88e6xxx_adjust_link,
-#ifdef CONFIG_NET_DSA_HWMON
-	.get_temp		= mv88e6xxx_get_temp,
-#endif
-	.get_regs_len		= mv88e6xxx_get_regs_len,
-	.get_regs		= mv88e6xxx_get_regs,
-};
-
-MODULE_ALIAS("platform:mv88e6123");
-MODULE_ALIAS("platform:mv88e6161");
-MODULE_ALIAS("platform:mv88e6165");
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 5f07524..5309c738 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -2989,8 +2989,8 @@ static int __init mv88e6xxx_init(void)
 #if IS_ENABLED(CONFIG_NET_DSA_MV88E6131)
 	register_switch_driver(&mv88e6131_switch_driver);
 #endif
-#if IS_ENABLED(CONFIG_NET_DSA_MV88E6123_61_65)
-	register_switch_driver(&mv88e6123_61_65_switch_driver);
+#if IS_ENABLED(CONFIG_NET_DSA_MV88E6123)
+	register_switch_driver(&mv88e6123_switch_driver);
 #endif
 #if IS_ENABLED(CONFIG_NET_DSA_MV88E6352)
 	register_switch_driver(&mv88e6352_switch_driver);
@@ -3010,8 +3010,8 @@ static void __exit mv88e6xxx_cleanup(void)
 #if IS_ENABLED(CONFIG_NET_DSA_MV88E6352)
 	unregister_switch_driver(&mv88e6352_switch_driver);
 #endif
-#if IS_ENABLED(CONFIG_NET_DSA_MV88E6123_61_65)
-	unregister_switch_driver(&mv88e6123_61_65_switch_driver);
+#if IS_ENABLED(CONFIG_NET_DSA_MV88E6123)
+	unregister_switch_driver(&mv88e6123_switch_driver);
 #endif
 #if IS_ENABLED(CONFIG_NET_DSA_MV88E6131)
 	unregister_switch_driver(&mv88e6131_switch_driver);
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 3425616..281cefe 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -519,7 +519,7 @@ int mv88e6xxx_phy_page_write(struct dsa_switch *ds, int port, int page,
 			     int reg, int val);
 
 extern struct dsa_switch_driver mv88e6131_switch_driver;
-extern struct dsa_switch_driver mv88e6123_61_65_switch_driver;
+extern struct dsa_switch_driver mv88e6123_switch_driver;
 extern struct dsa_switch_driver mv88e6352_switch_driver;
 extern struct dsa_switch_driver mv88e6171_switch_driver;
 
-- 
cgit v0.10.2


From b71be352f76ac3c0559feec2fd960ab9c31404c9 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 12 Mar 2016 00:01:37 +0100
Subject: dsa: slave: Don't reference NULL pointer during phy_disconnect

When the phy is disconnected, the parent pointer to the netdev it was
attached to is set to NULL. The code then tries to suspend the phy,
but dsa_slave_fixed_link_update needs the parent pointer to determine
which switch the phy is connected to. So it dereferenced a NULL
pointer. Check for this condition.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 27bf03d..49056d9 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -896,11 +896,15 @@ static void dsa_slave_adjust_link(struct net_device *dev)
 static int dsa_slave_fixed_link_update(struct net_device *dev,
 				       struct fixed_phy_status *status)
 {
-	struct dsa_slave_priv *p = netdev_priv(dev);
-	struct dsa_switch *ds = p->parent;
+	struct dsa_slave_priv *p;
+	struct dsa_switch *ds;
 
-	if (ds->drv->fixed_link_update)
-		ds->drv->fixed_link_update(ds, p->port, status);
+	if (dev) {
+		p = netdev_priv(dev);
+		ds = p->parent;
+		if (ds->drv->fixed_link_update)
+			ds->drv->fixed_link_update(ds, p->port, status);
+	}
 
 	return 0;
 }
-- 
cgit v0.10.2


From 3a44514ff979d7b0192a9f67445509f4ea7dfe2f Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 12 Mar 2016 00:01:38 +0100
Subject: dsa: Destroy fixed link phys after the phy has been disconnected

The phy is disconnected from the slave in dsa_slave_destroy(). Don't
destroy fixed link phys until after this, since there can be fixed
linked phys connected to ports.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index d8fb47f..1018e7d 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -430,7 +430,18 @@ static void dsa_switch_destroy(struct dsa_switch *ds)
 		hwmon_device_unregister(ds->hwmon_dev);
 #endif
 
-	/* Disable configuration of the CPU and DSA ports */
+	/* Destroy network devices for physical switch ports. */
+	for (port = 0; port < DSA_MAX_PORTS; port++) {
+		if (!(ds->phys_port_mask & (1 << port)))
+			continue;
+
+		if (!ds->ports[port])
+			continue;
+
+		dsa_slave_destroy(ds->ports[port]);
+	}
+
+	/* Remove any fixed link PHYs */
 	for (port = 0; port < DSA_MAX_PORTS; port++) {
 		if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)))
 			continue;
@@ -448,17 +459,6 @@ static void dsa_switch_destroy(struct dsa_switch *ds)
 		}
 	}
 
-	/* Destroy network devices for physical switch ports. */
-	for (port = 0; port < DSA_MAX_PORTS; port++) {
-		if (!(ds->phys_port_mask & (1 << port)))
-			continue;
-
-		if (!ds->ports[port])
-			continue;
-
-		dsa_slave_destroy(ds->ports[port]);
-	}
-
 	mdiobus_unregister(ds->slave_mii_bus);
 }
 
-- 
cgit v0.10.2


From ec777e6b4254648be51146d4e7d735ee3d06bd13 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 12 Mar 2016 00:01:39 +0100
Subject: dsa: dsa: Fix freeing of fixed-phys from user ports.

All ports types can have a fixed PHY associated with it. Remove the
check which limits removal to only CPU and DSA ports.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 1018e7d..f100f34 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -443,9 +443,6 @@ static void dsa_switch_destroy(struct dsa_switch *ds)
 
 	/* Remove any fixed link PHYs */
 	for (port = 0; port < DSA_MAX_PORTS; port++) {
-		if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)))
-			continue;
-
 		port_dn = cd->port_dn[port];
 		if (of_phy_is_fixed_link(port_dn)) {
 			phydev = of_phy_find_device(port_dn);
-- 
cgit v0.10.2


From 5bcbe0f35fb13e31fdd0b2dc9695f19ab0208207 Mon Sep 17 00:00:00 2001
From: Andrew Lunn <andrew@lunn.ch>
Date: Sat, 12 Mar 2016 00:01:40 +0100
Subject: phy: fixed: Fix removal of phys.

The fixed phys delete function simply removed the fixed phy from the
internal linked list and freed the memory. It however did not
unregister the associated phy device. This meant it was still possible
to find the phy device on the mdio bus.

Make fixed_phy_del() an internal function and add a
fixed_phy_unregister() to unregisters the phy device and then uses
fixed_phy_del() to free resources.

Modify DSA to use this new API function, so we don't leak phys.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c
index ab9c473..fc07a88 100644
--- a/drivers/net/phy/fixed_phy.c
+++ b/drivers/net/phy/fixed_phy.c
@@ -285,7 +285,7 @@ err_regs:
 }
 EXPORT_SYMBOL_GPL(fixed_phy_add);
 
-void fixed_phy_del(int phy_addr)
+static void fixed_phy_del(int phy_addr)
 {
 	struct fixed_mdio_bus *fmb = &platform_fmb;
 	struct fixed_phy *fp, *tmp;
@@ -300,7 +300,6 @@ void fixed_phy_del(int phy_addr)
 		}
 	}
 }
-EXPORT_SYMBOL_GPL(fixed_phy_del);
 
 static int phy_fixed_addr;
 static DEFINE_SPINLOCK(phy_fixed_addr_lock);
@@ -371,6 +370,14 @@ struct phy_device *fixed_phy_register(unsigned int irq,
 }
 EXPORT_SYMBOL_GPL(fixed_phy_register);
 
+void fixed_phy_unregister(struct phy_device *phy)
+{
+	phy_device_remove(phy);
+
+	fixed_phy_del(phy->mdio.addr);
+}
+EXPORT_SYMBOL_GPL(fixed_phy_unregister);
+
 static int __init fixed_mdio_bus_init(void)
 {
 	struct fixed_mdio_bus *fmb = &platform_fmb;
diff --git a/include/linux/phy_fixed.h b/include/linux/phy_fixed.h
index 2400d2e..1d41ec4 100644
--- a/include/linux/phy_fixed.h
+++ b/include/linux/phy_fixed.h
@@ -19,7 +19,7 @@ extern struct phy_device *fixed_phy_register(unsigned int irq,
 					     struct fixed_phy_status *status,
 					     int link_gpio,
 					     struct device_node *np);
-extern void fixed_phy_del(int phy_addr);
+extern void fixed_phy_unregister(struct phy_device *phydev);
 extern int fixed_phy_set_link_update(struct phy_device *phydev,
 			int (*link_update)(struct net_device *,
 					   struct fixed_phy_status *));
@@ -40,9 +40,8 @@ static inline struct phy_device *fixed_phy_register(unsigned int irq,
 {
 	return ERR_PTR(-ENODEV);
 }
-static inline int fixed_phy_del(int phy_addr)
+static inline void fixed_phy_unregister(struct phy_device *phydev)
 {
-	return -ENODEV;
 }
 static inline int fixed_phy_set_link_update(struct phy_device *phydev,
 			int (*link_update)(struct net_device *,
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index f100f34..c28c474 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -447,11 +447,9 @@ static void dsa_switch_destroy(struct dsa_switch *ds)
 		if (of_phy_is_fixed_link(port_dn)) {
 			phydev = of_phy_find_device(port_dn);
 			if (phydev) {
-				int addr = phydev->mdio.addr;
-
 				phy_device_free(phydev);
 				of_node_put(port_dn);
-				fixed_phy_del(addr);
+				fixed_phy_unregister(phydev);
 			}
 		}
 	}
-- 
cgit v0.10.2


From 8626c56c8279b88b91b8df55345cdf89542d5e90 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sat, 12 Mar 2016 11:14:42 +0100
Subject: bridge: fix potential use-after-free when hook returns QUEUE or
 STOLEN verdict

Zefir Kurtisi reported kernel panic with an openwrt specific patch.
However, it turns out that mainline has a similar bug waiting to happen.

Once NF_HOOK() returns the skb is in undefined state and must not be
used.   Moreover, the okfn must consume the skb to support async
processing (NF_QUEUE).

Current okfn in this spot doesn't consume it and caller assumes that
NF_HOOK return value tells us if skb was freed or not, but thats wrong.

It "works" because no in-tree user registers a NFPROTO_BRIDGE hook at
LOCAL_IN that returns STOLEN or NF_QUEUE verdicts.

Once we add NF_QUEUE support for nftables bridge this will break --
NF_QUEUE holds the skb for async processing, caller will erronoulsy
return RX_HANDLER_PASS and on reinject netfilter will access free'd skb.

Fix this by pushing skb up the stack in the okfn instead.

NB: It also seems dubious to use LOCAL_IN while bypassing PRE_ROUTING
completely in this case but this is how its been forever so it seems
preferable to not change this.

Cc: Felix Fietkau <nbd@openwrt.org>
Cc: Zefir Kurtisi <zefir.kurtisi@neratec.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Tested-by: Zefir Kurtisi <zefir.kurtisi@neratec.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index f7fba74..1607977 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -222,7 +222,10 @@ static int br_handle_local_finish(struct net *net, struct sock *sk, struct sk_bu
 	/* check if vlan is allowed, to avoid spoofing */
 	if (p->flags & BR_LEARNING && br_should_learn(p, skb, &vid))
 		br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid, false);
-	return 0;	 /* process further */
+
+	BR_INPUT_SKB_CB(skb)->brdev = p->br->dev;
+	br_pass_frame_up(skb);
+	return 0;
 }
 
 /*
@@ -284,14 +287,9 @@ rx_handler_result_t br_handle_frame(struct sk_buff **pskb)
 		}
 
 		/* Deliver packet to local host only */
-		if (NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN,
-			    dev_net(skb->dev), NULL, skb, skb->dev, NULL,
-			    br_handle_local_finish)) {
-			return RX_HANDLER_CONSUMED; /* consumed by filter */
-		} else {
-			*pskb = skb;
-			return RX_HANDLER_PASS;	/* continue processing */
-		}
+		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, dev_net(skb->dev),
+			NULL, skb, skb->dev, NULL, br_handle_local_finish);
+		return RX_HANDLER_CONSUMED;
 	}
 
 forward:
-- 
cgit v0.10.2


From 1c2722a975fdb8c90bc6ba8570b7fb62db4e2e9c Mon Sep 17 00:00:00 2001
From: Gregory CLEMENT <gregory.clement@free-electrons.com>
Date: Sat, 12 Mar 2016 18:44:17 +0100
Subject: net: mvneta: Fix spinlock usage

In the previous patch, the spinlock was not initialized. While it didn't
cause any trouble yet it could be a problem to use it uninitialized.

The most annoying part was the critical section protected by the spinlock
in mvneta_stop(). Some of the functions could sleep as pointed when
activated CONFIG_DEBUG_ATOMIC_SLEEP. Actually, in mvneta_stop() we only
need to protect the is_stopped flagged, indeed the code of the notifier
for CPU online is protected by the same spinlock, so when we get the
lock, the notifer work is done.

Reported-by: Patrick Uiterwijk <patrick@puiterwijk.org>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index be92668..665eac2 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3480,17 +3480,17 @@ static int mvneta_stop(struct net_device *dev)
 	struct mvneta_port *pp = netdev_priv(dev);
 
 	/* Inform that we are stopping so we don't want to setup the
-	 * driver for new CPUs in the notifiers
+	 * driver for new CPUs in the notifiers. The code of the
+	 * notifier for CPU online is protected by the same spinlock,
+	 * so when we get the lock, the notifer work is done.
 	 */
 	spin_lock(&pp->lock);
 	pp->is_stopped = true;
+	spin_unlock(&pp->lock);
+
 	mvneta_stop_dev(pp);
 	mvneta_mdio_remove(pp);
 	unregister_cpu_notifier(&pp->cpu_notifier);
-	/* Now that the notifier are unregistered, we can release le
-	 * lock
-	 */
-	spin_unlock(&pp->lock);
 	on_each_cpu(mvneta_percpu_disable, pp, true);
 	free_percpu_irq(dev->irq, pp->ports);
 	mvneta_cleanup_rxqs(pp);
@@ -4023,6 +4023,7 @@ static int mvneta_probe(struct platform_device *pdev)
 	dev->ethtool_ops = &mvneta_eth_tool_ops;
 
 	pp = netdev_priv(dev);
+	spin_lock_init(&pp->lock);
 	pp->phy_node = phy_node;
 	pp->phy_interface = phy_mode;
 
-- 
cgit v0.10.2


From 928b6519afeb2a5e2dc61154380b545ed66c476a Mon Sep 17 00:00:00 2001
From: Dmitri Epshtein <dima@marvell.com>
Date: Sat, 12 Mar 2016 18:44:18 +0100
Subject: net: mvneta: enable change MAC address when interface is up

Function eth_prepare_mac_addr_change() is called as part of MAC
address change. This function check if interface is running.
To enable change MAC address when interface is running:
IFF_LIVE_ADDR_CHANGE flag must be set to dev->priv_flags field

Fixes: c5aff18204da ("net: mvneta: driver for Marvell Armada 370/XP
network unit")
Cc: stable@vger.kernel.org
Signed-off-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 665eac2..09a3066 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -4145,7 +4145,7 @@ static int mvneta_probe(struct platform_device *pdev)
 	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO;
 	dev->hw_features |= dev->features;
 	dev->vlan_features |= dev->features;
-	dev->priv_flags |= IFF_UNICAST_FLT;
+	dev->priv_flags |= IFF_UNICAST_FLT | IFF_LIVE_ADDR_CHANGE;
 	dev->gso_max_segs = MVNETA_MAX_TSO_SEGS;
 
 	err = register_netdev(dev);
-- 
cgit v0.10.2


From 0838abb3c08cff83719b461adffefc83721af34b Mon Sep 17 00:00:00 2001
From: Dmitri Epshtein <dima@marvell.com>
Date: Sat, 12 Mar 2016 18:44:19 +0100
Subject: net: mvneta: fix error messages in mvneta_port_down function

This commit corrects error printing when shutting down the port.

[gregory.clement@free-electrons.com: split initial commit in two
individual changes]
Signed-off-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 09a3066..1af60bd 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -1107,7 +1107,7 @@ static void mvneta_port_down(struct mvneta_port *pp)
 	do {
 		if (count++ >= MVNETA_RX_DISABLE_TIMEOUT_MSEC) {
 			netdev_warn(pp->dev,
-				    "TIMEOUT for RX stopped ! rx_queue_cmd: 0x08%x\n",
+				    "TIMEOUT for RX stopped ! rx_queue_cmd: 0x%08x\n",
 				    val);
 			break;
 		}
@@ -1146,7 +1146,7 @@ static void mvneta_port_down(struct mvneta_port *pp)
 	do {
 		if (count++ >= MVNETA_TX_FIFO_EMPTY_TIMEOUT) {
 			netdev_warn(pp->dev,
-				    "TX FIFO empty timeout status=0x08%x\n",
+				    "TX FIFO empty timeout status=0x%08x\n",
 				    val);
 			break;
 		}
-- 
cgit v0.10.2


From a3703fb31a4779e5c515f51f602dda1ffcfbfec9 Mon Sep 17 00:00:00 2001
From: Dmitri Epshtein <dima@marvell.com>
Date: Sat, 12 Mar 2016 18:44:20 +0100
Subject: net: mvneta: replace magic numbers by existing macros

Some literal values are actually already defined by macros, so let's use
them.

[gregory.clement@free-electrons.com: split intial commit in two
individual changes]
Signed-off-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 1af60bd..577f7ca 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -1114,7 +1114,7 @@ static void mvneta_port_down(struct mvneta_port *pp)
 		mdelay(1);
 
 		val = mvreg_read(pp, MVNETA_RXQ_CMD);
-	} while (val & 0xff);
+	} while (val & MVNETA_RXQ_ENABLE_MASK);
 
 	/* Stop Tx port activity. Check port Tx activity. Issue stop
 	 * command for active channels only
@@ -1139,7 +1139,7 @@ static void mvneta_port_down(struct mvneta_port *pp)
 		/* Check TX Command reg that all Txqs are stopped */
 		val = mvreg_read(pp, MVNETA_TXQ_CMD);
 
-	} while (val & 0xff);
+	} while (val & MVNETA_TXQ_ENABLE_MASK);
 
 	/* Double check to verify that TX FIFO is empty */
 	count = 0;
-- 
cgit v0.10.2


From 4fa8c3cc70b2a0aa090bce174c6f2ac148453690 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Sun, 13 Mar 2016 01:29:45 +0300
Subject: sh_eth: kill useless initializers

Some of the local variable intializers in the driver turned out to be
pointless,  kill 'em.

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 2e9a781..8385d92 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -447,8 +447,8 @@ static bool sh_eth_is_rz_fast_ether(struct sh_eth_private *mdp)
 
 static void sh_eth_select_mii(struct net_device *ndev)
 {
-	u32 value = 0x0;
 	struct sh_eth_private *mdp = netdev_priv(ndev);
+	u32 value;
 
 	switch (mdp->phy_interface) {
 	case PHY_INTERFACE_MODE_GMII:
@@ -1231,8 +1231,8 @@ ring_free:
 
 static int sh_eth_dev_init(struct net_device *ndev, bool start)
 {
-	int ret = 0;
 	struct sh_eth_private *mdp = netdev_priv(ndev);
+	int ret;
 
 	/* Soft Reset */
 	ret = sh_eth_reset(ndev);
@@ -1355,7 +1355,7 @@ static int sh_eth_txfree(struct net_device *ndev)
 	struct sh_eth_private *mdp = netdev_priv(ndev);
 	struct sh_eth_txdesc *txdesc;
 	int free_num = 0;
-	int entry = 0;
+	int entry;
 
 	for (; mdp->cur_tx - mdp->dirty_tx > 0; mdp->dirty_tx++) {
 		entry = mdp->dirty_tx % mdp->num_tx_ring;
@@ -1396,10 +1396,10 @@ static int sh_eth_rx(struct net_device *ndev, u32 intr_status, int *quota)
 	int boguscnt = (mdp->dirty_rx + mdp->num_rx_ring) - mdp->cur_rx;
 	int limit;
 	struct sk_buff *skb;
-	u16 pkt_len = 0;
 	u32 desc_status;
 	int skbuff_size = mdp->rx_buf_sz + SH_ETH_RX_ALIGN + 32 - 1;
 	dma_addr_t dma_addr;
+	u16 pkt_len;
 	u32 buf_len;
 
 	boguscnt = min(boguscnt, *quota);
@@ -1778,7 +1778,7 @@ static int sh_eth_phy_init(struct net_device *ndev)
 {
 	struct device_node *np = ndev->dev.parent->of_node;
 	struct sh_eth_private *mdp = netdev_priv(ndev);
-	struct phy_device *phydev = NULL;
+	struct phy_device *phydev;
 
 	mdp->link = 0;
 	mdp->speed = 0;
@@ -2232,8 +2232,8 @@ static const struct ethtool_ops sh_eth_ethtool_ops = {
 /* network device open function */
 static int sh_eth_open(struct net_device *ndev)
 {
-	int ret = 0;
 	struct sh_eth_private *mdp = netdev_priv(ndev);
+	int ret;
 
 	pm_runtime_get_sync(&mdp->pdev->dev);
 
@@ -2988,12 +2988,12 @@ static inline struct sh_eth_plat_data *sh_eth_parse_dt(struct device *dev)
 
 static int sh_eth_drv_probe(struct platform_device *pdev)
 {
-	int ret, devno = 0;
 	struct resource *res;
-	struct net_device *ndev = NULL;
-	struct sh_eth_private *mdp = NULL;
 	struct sh_eth_plat_data *pd = dev_get_platdata(&pdev->dev);
 	const struct platform_device_id *id = platform_get_device_id(pdev);
+	struct sh_eth_private *mdp;
+	struct net_device *ndev;
+	int ret, devno;
 
 	/* get base addr */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-- 
cgit v0.10.2


From 464be1e0be687ffbad64014099f7d0c1a5f3723e Mon Sep 17 00:00:00 2001
From: "Maciej S. Szmigiero" <mail@maciej.szmigiero.name>
Date: Sun, 13 Mar 2016 00:18:06 +0100
Subject: mISDN: Order IPAC register defines

It looks like IPAC/ISAC chips register defines weren't in any particular
order.

Order them by their number to make it easier to spot holes.

Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name>
Acked-by: Karsten Keil <keil@b1-systems.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/isdn/hardware/mISDN/ipac.h b/drivers/isdn/hardware/mISDN/ipac.h
index 8121e04..76aa6f8 100644
--- a/drivers/isdn/hardware/mISDN/ipac.h
+++ b/drivers/isdn/hardware/mISDN/ipac.h
@@ -99,32 +99,32 @@ struct ipac_hw {
 
 /* All registers original Siemens Spec  */
 /* IPAC/ISAC registers */
-#define ISAC_MASK		0x20
 #define ISAC_ISTA		0x20
-#define ISAC_STAR		0x21
+#define ISAC_MASK		0x20
 #define ISAC_CMDR		0x21
+#define ISAC_STAR		0x21
+#define ISAC_MODE		0x22
+#define ISAC_TIMR		0x23
 #define ISAC_EXIR		0x24
-#define ISAC_ADF2		0x39
+#define ISAC_RBCL		0x25
+#define ISAC_RSTA		0x27
+#define ISAC_RBCH		0x2A
 #define ISAC_SPCR		0x30
-#define ISAC_ADF1		0x38
 #define ISAC_CIR0		0x31
 #define ISAC_CIX0		0x31
-#define ISAC_CIR1		0x33
-#define ISAC_CIX1		0x33
-#define ISAC_STCR		0x37
-#define ISAC_MODE		0x22
-#define ISAC_RSTA		0x27
-#define ISAC_RBCL		0x25
-#define ISAC_RBCH		0x2A
-#define ISAC_TIMR		0x23
-#define ISAC_SQXR		0x3b
-#define ISAC_SQRR		0x3b
-#define ISAC_MOSR		0x3a
-#define ISAC_MOCR		0x3a
 #define ISAC_MOR0		0x32
 #define ISAC_MOX0		0x32
+#define ISAC_CIR1		0x33
+#define ISAC_CIX1		0x33
 #define ISAC_MOR1		0x34
 #define ISAC_MOX1		0x34
+#define ISAC_STCR		0x37
+#define ISAC_ADF1		0x38
+#define ISAC_ADF2		0x39
+#define ISAC_MOCR		0x3a
+#define ISAC_MOSR		0x3a
+#define ISAC_SQRR		0x3b
+#define ISAC_SQXR		0x3b
 
 #define ISAC_RBCH_XAC		0x80
 
@@ -212,13 +212,13 @@ struct ipac_hw {
 #define ISAC_CMD_DUI		0xF
 
 /* ISAC/ISACX/IPAC/IPACX L1 indications */
-#define ISAC_IND_RS		0x1
-#define ISAC_IND_PU		0x7
 #define ISAC_IND_DR		0x0
+#define ISAC_IND_RS		0x1
 #define ISAC_IND_SD		0x2
 #define ISAC_IND_DIS		0x3
-#define ISAC_IND_EI		0x6
 #define ISAC_IND_RSY		0x4
+#define ISAC_IND_EI		0x6
+#define ISAC_IND_PU		0x7
 #define ISAC_IND_ARD		0x8
 #define ISAC_IND_TI		0xA
 #define ISAC_IND_ATI		0xB
@@ -339,9 +339,9 @@ struct ipac_hw {
 #define ISACX__AUX		0x08
 #define ISACX__CIC		0x10
 #define ISACX__ST		0x20
+#define IPACX__ON		0x2C
 #define IPACX__ICB		0x40
 #define IPACX__ICA		0x80
-#define IPACX__ON		0x2C
 
 /* ISACX/IPACX _CMDRD (W) */
 #define ISACX_CMDRD_XRES	0x01
-- 
cgit v0.10.2


From 1e1589ad8b5cb5b8a6781ba5850cf710ada0e919 Mon Sep 17 00:00:00 2001
From: "Maciej S. Szmigiero" <mail@maciej.szmigiero.name>
Date: Sun, 13 Mar 2016 00:19:07 +0100
Subject: mISDN: Support DR6 indication in mISDNipac driver

According to figure 39 in PEB3086 data sheet, version 1.4 this indication
replaces DR when layer 1 transition source state is F6.

This fixes mISDN layer 1 getting stuck in F6 state in TE mode on
Dialogic Diva 2.02 card (and possibly others) when NT deactivates it.

Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name>
Acked-by: Karsten Keil <keil@b1-systems.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/isdn/hardware/mISDN/ipac.h b/drivers/isdn/hardware/mISDN/ipac.h
index 76aa6f8..720ee72 100644
--- a/drivers/isdn/hardware/mISDN/ipac.h
+++ b/drivers/isdn/hardware/mISDN/ipac.h
@@ -217,6 +217,7 @@ struct ipac_hw {
 #define ISAC_IND_SD		0x2
 #define ISAC_IND_DIS		0x3
 #define ISAC_IND_RSY		0x4
+#define ISAC_IND_DR6		0x5
 #define ISAC_IND_EI		0x6
 #define ISAC_IND_PU		0x7
 #define ISAC_IND_ARD		0x8
diff --git a/drivers/isdn/hardware/mISDN/mISDNipac.c b/drivers/isdn/hardware/mISDN/mISDNipac.c
index cb428b9..aa9b6c3 100644
--- a/drivers/isdn/hardware/mISDN/mISDNipac.c
+++ b/drivers/isdn/hardware/mISDN/mISDNipac.c
@@ -80,6 +80,7 @@ isac_ph_state_bh(struct dchannel *dch)
 		l1_event(dch->l1, HW_DEACT_CNF);
 		break;
 	case ISAC_IND_DR:
+	case ISAC_IND_DR6:
 		dch->state = 3;
 		l1_event(dch->l1, HW_DEACT_IND);
 		break;
@@ -660,6 +661,7 @@ isac_l1cmd(struct dchannel *dch, u32 cmd)
 		spin_lock_irqsave(isac->hwlock, flags);
 		if ((isac->state == ISAC_IND_EI) ||
 		    (isac->state == ISAC_IND_DR) ||
+		    (isac->state == ISAC_IND_DR6) ||
 		    (isac->state == ISAC_IND_RS))
 			ph_command(isac, ISAC_CMD_TIM);
 		else
-- 
cgit v0.10.2


From 71327a4e7d997276d49db92fd3d30008389ee6d5 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Sun, 13 Mar 2016 16:21:32 -0400
Subject: net: dsa: rename port_*_bridge routines

Rename DSA port_join_bridge and port_leave_bridge routines to
respectively port_bridge_join and port_bridge_leave in order to respect
an implicit Port::Bridge namespace.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/networking/dsa/dsa.txt b/Documentation/networking/dsa/dsa.txt
index 974e9c3..3b196c3 100644
--- a/Documentation/networking/dsa/dsa.txt
+++ b/Documentation/networking/dsa/dsa.txt
@@ -521,12 +521,12 @@ See Documentation/hwmon/sysfs-interface for details.
 Bridge layer
 ------------
 
-- port_join_bridge: bridge layer function invoked when a given switch port is
+- port_bridge_join: bridge layer function invoked when a given switch port is
   added to a bridge, this function should be doing the necessary at the switch
   level to permit the joining port from being added to the relevant logical
   domain for it to ingress/egress traffic with other members of the bridge.
 
-- port_leave_bridge: bridge layer function invoked when a given switch port is
+- port_bridge_leave: bridge layer function invoked when a given switch port is
   removed from a bridge, this function should be doing the necessary at the
   switch level to deny the leaving port from ingress/egress traffic from the
   remaining bridge members. When the port leaves the bridge, it should be aged
diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 3f62759..4bcc9eb 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -1387,8 +1387,8 @@ static struct dsa_switch_driver bcm_sf2_switch_driver = {
 	.port_disable		= bcm_sf2_port_disable,
 	.get_eee		= bcm_sf2_sw_get_eee,
 	.set_eee		= bcm_sf2_sw_set_eee,
-	.port_join_bridge	= bcm_sf2_sw_br_join,
-	.port_leave_bridge	= bcm_sf2_sw_br_leave,
+	.port_bridge_join	= bcm_sf2_sw_br_join,
+	.port_bridge_leave	= bcm_sf2_sw_br_leave,
 	.port_stp_update	= bcm_sf2_sw_br_set_stp_state,
 	.port_fdb_prepare	= bcm_sf2_sw_fdb_prepare,
 	.port_fdb_add		= bcm_sf2_sw_fdb_add,
diff --git a/drivers/net/dsa/mv88e6171.c b/drivers/net/dsa/mv88e6171.c
index d72ccbd..c0164b9 100644
--- a/drivers/net/dsa/mv88e6171.c
+++ b/drivers/net/dsa/mv88e6171.c
@@ -103,8 +103,8 @@ struct dsa_switch_driver mv88e6171_switch_driver = {
 #endif
 	.get_regs_len		= mv88e6xxx_get_regs_len,
 	.get_regs		= mv88e6xxx_get_regs,
-	.port_join_bridge	= mv88e6xxx_port_bridge_join,
-	.port_leave_bridge	= mv88e6xxx_port_bridge_leave,
+	.port_bridge_join	= mv88e6xxx_port_bridge_join,
+	.port_bridge_leave	= mv88e6xxx_port_bridge_leave,
 	.port_stp_update        = mv88e6xxx_port_stp_update,
 	.port_vlan_filtering	= mv88e6xxx_port_vlan_filtering,
 	.port_vlan_prepare	= mv88e6xxx_port_vlan_prepare,
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
index a41fa50..5f528ab 100644
--- a/drivers/net/dsa/mv88e6352.c
+++ b/drivers/net/dsa/mv88e6352.c
@@ -324,8 +324,8 @@ struct dsa_switch_driver mv88e6352_switch_driver = {
 	.set_eeprom		= mv88e6352_set_eeprom,
 	.get_regs_len		= mv88e6xxx_get_regs_len,
 	.get_regs		= mv88e6xxx_get_regs,
-	.port_join_bridge	= mv88e6xxx_port_bridge_join,
-	.port_leave_bridge	= mv88e6xxx_port_bridge_leave,
+	.port_bridge_join	= mv88e6xxx_port_bridge_join,
+	.port_bridge_leave	= mv88e6xxx_port_bridge_leave,
 	.port_stp_update	= mv88e6xxx_port_stp_update,
 	.port_vlan_filtering	= mv88e6xxx_port_vlan_filtering,
 	.port_vlan_prepare	= mv88e6xxx_port_vlan_prepare,
diff --git a/include/net/dsa.h b/include/net/dsa.h
index 26c0a3f..004e034 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -296,9 +296,9 @@ struct dsa_switch_driver {
 	/*
 	 * Bridge integration
 	 */
-	int	(*port_join_bridge)(struct dsa_switch *ds, int port,
+	int	(*port_bridge_join)(struct dsa_switch *ds, int port,
 				    struct net_device *bridge);
-	int	(*port_leave_bridge)(struct dsa_switch *ds, int port);
+	int	(*port_bridge_leave)(struct dsa_switch *ds, int port);
 	int	(*port_stp_update)(struct dsa_switch *ds, int port,
 				   u8 state);
 
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 49056d9..52653d7 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -448,8 +448,8 @@ static int dsa_slave_bridge_port_join(struct net_device *dev,
 
 	p->bridge_dev = br;
 
-	if (ds->drv->port_join_bridge)
-		ret = ds->drv->port_join_bridge(ds, p->port, br);
+	if (ds->drv->port_bridge_join)
+		ret = ds->drv->port_bridge_join(ds, p->port, br);
 
 	return ret;
 }
@@ -461,8 +461,8 @@ static int dsa_slave_bridge_port_leave(struct net_device *dev)
 	int ret = -EOPNOTSUPP;
 
 
-	if (ds->drv->port_leave_bridge)
-		ret = ds->drv->port_leave_bridge(ds, p->port);
+	if (ds->drv->port_bridge_leave)
+		ret = ds->drv->port_bridge_leave(ds, p->port);
 
 	p->bridge_dev = NULL;
 
-- 
cgit v0.10.2


From 16bfa7024eba5e36aff38ba62086b9027373007d Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Sun, 13 Mar 2016 16:21:33 -0400
Subject: net: dsa: make port_bridge_leave return void

netdev_upper_dev_unlink() which notifies NETDEV_CHANGEUPPER, returns
void, as well as del_nbp(). So there's no advantage to catch an eventual
error from the port_bridge_leave routine at the DSA level.

Make this routine void for the DSA layer and its existing drivers.

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 4bcc9eb..95944d5 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -516,7 +516,7 @@ static int bcm_sf2_sw_br_join(struct dsa_switch *ds, int port,
 	return 0;
 }
 
-static int bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port)
+static void bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port)
 {
 	struct bcm_sf2_priv *priv = ds_to_priv(ds);
 	struct net_device *bridge = priv->port_sts[port].bridge_dev;
@@ -543,8 +543,6 @@ static int bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port)
 	core_writel(priv, p_ctl, CORE_PORT_VLAN_CTL_PORT(port));
 	priv->port_sts[port].vlan_ctl_mask = p_ctl;
 	priv->port_sts[port].bridge_dev = NULL;
-
-	return 0;
 }
 
 static int bcm_sf2_sw_br_set_stp_state(struct dsa_switch *ds, int port,
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index 5309c738..fa086e0 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -2219,39 +2219,29 @@ unlock:
 	return err;
 }
 
-int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port)
+void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	struct net_device *bridge = ps->ports[port].bridge_dev;
 	u16 fid;
-	int i, err;
+	int i;
 
 	mutex_lock(&ps->smi_mutex);
 
 	/* Give the port a fresh Filtering Information Database */
-	err = _mv88e6xxx_fid_new(ds, &fid);
-	if (err)
-		goto unlock;
-
-	err = _mv88e6xxx_port_fid_set(ds, port, fid);
-	if (err)
-		goto unlock;
+	if (_mv88e6xxx_fid_new(ds, &fid) ||
+	    _mv88e6xxx_port_fid_set(ds, port, fid))
+		netdev_warn(ds->ports[port], "failed to assign a new FID\n");
 
 	/* Unassign the bridge and remap each port's VLANTable */
 	ps->ports[port].bridge_dev = NULL;
 
-	for (i = 0; i < ps->num_ports; ++i) {
-		if (i == port || ps->ports[i].bridge_dev == bridge) {
-			err = _mv88e6xxx_port_based_vlan_map(ds, i);
-			if (err)
-				break;
-		}
-	}
+	for (i = 0; i < ps->num_ports; ++i)
+		if (i == port || ps->ports[i].bridge_dev == bridge)
+			if (_mv88e6xxx_port_based_vlan_map(ds, i))
+				netdev_warn(ds->ports[i], "failed to remap\n");
 
-unlock:
 	mutex_unlock(&ps->smi_mutex);
-
-	return err;
 }
 
 static void mv88e6xxx_bridge_work(struct work_struct *work)
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 281cefe..9a038ab 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -488,7 +488,7 @@ int mv88e6xxx_set_eee(struct dsa_switch *ds, int port,
 		      struct phy_device *phydev, struct ethtool_eee *e);
 int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port,
 			       struct net_device *bridge);
-int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port);
+void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port);
 int mv88e6xxx_port_stp_update(struct dsa_switch *ds, int port, u8 state);
 int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port,
 				  bool vlan_filtering);
diff --git a/include/net/dsa.h b/include/net/dsa.h
index 004e034..6463bb2 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -298,7 +298,7 @@ struct dsa_switch_driver {
 	 */
 	int	(*port_bridge_join)(struct dsa_switch *ds, int port,
 				    struct net_device *bridge);
-	int	(*port_bridge_leave)(struct dsa_switch *ds, int port);
+	void	(*port_bridge_leave)(struct dsa_switch *ds, int port);
 	int	(*port_stp_update)(struct dsa_switch *ds, int port,
 				   u8 state);
 
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 52653d7..8e00f1d 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -454,15 +454,14 @@ static int dsa_slave_bridge_port_join(struct net_device *dev,
 	return ret;
 }
 
-static int dsa_slave_bridge_port_leave(struct net_device *dev)
+static void dsa_slave_bridge_port_leave(struct net_device *dev)
 {
 	struct dsa_slave_priv *p = netdev_priv(dev);
 	struct dsa_switch *ds = p->parent;
-	int ret = -EOPNOTSUPP;
 
 
 	if (ds->drv->port_bridge_leave)
-		ret = ds->drv->port_bridge_leave(ds, p->port);
+		ds->drv->port_bridge_leave(ds, p->port);
 
 	p->bridge_dev = NULL;
 
@@ -470,8 +469,6 @@ static int dsa_slave_bridge_port_leave(struct net_device *dev)
 	 * so allow it to be in BR_STATE_FORWARDING to be kept functional
 	 */
 	dsa_slave_stp_update(dev, BR_STATE_FORWARDING);
-
-	return ret;
 }
 
 static int dsa_slave_port_attr_get(struct net_device *dev,
@@ -1156,7 +1153,7 @@ static int dsa_slave_master_changed(struct net_device *dev)
 	    !strcmp(master->rtnl_link_ops->kind, "bridge"))
 		err = dsa_slave_bridge_port_join(dev, master);
 	else if (dsa_port_is_bridged(p))
-		err = dsa_slave_bridge_port_leave(dev);
+		dsa_slave_bridge_port_leave(dev);
 
 	return err;
 }
-- 
cgit v0.10.2


From 6debb68a2d1acd1f5eca1d9d1c79a54dc998b6d6 Mon Sep 17 00:00:00 2001
From: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Date: Sun, 13 Mar 2016 16:21:34 -0400
Subject: net: dsa: refine netdev event notifier

Rework the netdev event handler, similar to what the Mellanox Spectrum
driver does, to easily welcome more events later (for example
NETDEV_PRECHANGEUPPER) and use netdev helpers (such as
netif_is_bridge_master).

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 8e00f1d..a575f03 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -451,7 +451,7 @@ static int dsa_slave_bridge_port_join(struct net_device *dev,
 	if (ds->drv->port_bridge_join)
 		ret = ds->drv->port_bridge_join(ds, p->port, br);
 
-	return ret;
+	return ret == -EOPNOTSUPP ? 0 : ret;
 }
 
 static void dsa_slave_bridge_port_leave(struct net_device *dev)
@@ -1143,40 +1143,46 @@ static bool dsa_slave_dev_check(struct net_device *dev)
 	return dev->netdev_ops == &dsa_slave_netdev_ops;
 }
 
-static int dsa_slave_master_changed(struct net_device *dev)
+static int dsa_slave_port_upper_event(struct net_device *dev,
+				      unsigned long event, void *ptr)
 {
-	struct net_device *master = netdev_master_upper_dev_get(dev);
-	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct netdev_notifier_changeupper_info *info = ptr;
+	struct net_device *upper = info->upper_dev;
 	int err = 0;
 
-	if (master && master->rtnl_link_ops &&
-	    !strcmp(master->rtnl_link_ops->kind, "bridge"))
-		err = dsa_slave_bridge_port_join(dev, master);
-	else if (dsa_port_is_bridged(p))
-		dsa_slave_bridge_port_leave(dev);
+	switch (event) {
+	case NETDEV_CHANGEUPPER:
+		if (netif_is_bridge_master(upper)) {
+			if (info->linking)
+				err = dsa_slave_bridge_port_join(dev, upper);
+			else
+				dsa_slave_bridge_port_leave(dev);
+		}
 
-	return err;
+		break;
+	}
+
+	return notifier_from_errno(err);
 }
 
-int dsa_slave_netdevice_event(struct notifier_block *unused,
-			      unsigned long event, void *ptr)
+static int dsa_slave_port_event(struct net_device *dev, unsigned long event,
+				void *ptr)
 {
-	struct net_device *dev;
-	int err = 0;
-
 	switch (event) {
 	case NETDEV_CHANGEUPPER:
-		dev = netdev_notifier_info_to_dev(ptr);
-		if (!dsa_slave_dev_check(dev))
-			goto out;
+		return dsa_slave_port_upper_event(dev, event, ptr);
+	}
 
-		err = dsa_slave_master_changed(dev);
-		if (err && err != -EOPNOTSUPP)
-			netdev_warn(dev, "failed to reflect master change\n");
+	return NOTIFY_DONE;
+}
 
-		break;
-	}
+int dsa_slave_netdevice_event(struct notifier_block *unused,
+			      unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+
+	if (dsa_slave_dev_check(dev))
+		return dsa_slave_port_event(dev, event, ptr);
 
-out:
 	return NOTIFY_DONE;
 }
-- 
cgit v0.10.2


From bfa3f9d7f3b349acea8982d2248e33a0ed84c687 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:16 -0800
Subject: netfilter: Remove IP_CT_NEW_REPLY definition.

Remove the definition of IP_CT_NEW_REPLY from the kernel as it does
not make sense.  This allows the definition of IP_CT_NUMBER to be
simplified as well.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/uapi/linux/netfilter/nf_conntrack_common.h b/include/uapi/linux/netfilter/nf_conntrack_common.h
index 319f471..6d074d1 100644
--- a/include/uapi/linux/netfilter/nf_conntrack_common.h
+++ b/include/uapi/linux/netfilter/nf_conntrack_common.h
@@ -20,9 +20,15 @@ enum ip_conntrack_info {
 
 	IP_CT_ESTABLISHED_REPLY = IP_CT_ESTABLISHED + IP_CT_IS_REPLY,
 	IP_CT_RELATED_REPLY = IP_CT_RELATED + IP_CT_IS_REPLY,
-	IP_CT_NEW_REPLY = IP_CT_NEW + IP_CT_IS_REPLY,	
-	/* Number of distinct IP_CT types (no NEW in reply dirn). */
-	IP_CT_NUMBER = IP_CT_IS_REPLY * 2 - 1
+	/* No NEW in reply direction. */
+
+	/* Number of distinct IP_CT types. */
+	IP_CT_NUMBER,
+
+	/* only for userspace compatibility */
+#ifndef __KERNEL__
+	IP_CT_NEW_REPLY = IP_CT_NUMBER,
+#endif
 };
 
 #define NF_CT_STATE_INVALID_BIT			(1 << 0)
diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index ee6ff8f..3045290 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -75,7 +75,6 @@ static u8 ovs_ct_get_state(enum ip_conntrack_info ctinfo)
 	switch (ctinfo) {
 	case IP_CT_ESTABLISHED_REPLY:
 	case IP_CT_RELATED_REPLY:
-	case IP_CT_NEW_REPLY:
 		ct_state |= OVS_CS_F_REPLY_DIR;
 		break;
 	default:
@@ -92,7 +91,6 @@ static u8 ovs_ct_get_state(enum ip_conntrack_info ctinfo)
 		ct_state |= OVS_CS_F_RELATED;
 		break;
 	case IP_CT_NEW:
-	case IP_CT_NEW_REPLY:
 		ct_state |= OVS_CS_F_NEW;
 		break;
 	default:
-- 
cgit v0.10.2


From 264619055bd52bc2278af848472176642d759874 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:17 -0800
Subject: netfilter: Allow calling into nat helper without skb_dst.

NAT checksum recalculation code assumes existence of skb_dst, which
becomes a problem for a later patch in the series ("openvswitch:
Interface with NAT.").  Simplify this by removing the check on
skb_dst, as the checksum will be dealt with later in the stack.

Suggested-by: Pravin Shelar <pshelar@nicira.com>
Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c
index 61c7cc2..f8aad03 100644
--- a/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_nat_l3proto_ipv4.c
@@ -127,29 +127,15 @@ static void nf_nat_ipv4_csum_recalc(struct sk_buff *skb,
 				    u8 proto, void *data, __sum16 *check,
 				    int datalen, int oldlen)
 {
-	const struct iphdr *iph = ip_hdr(skb);
-	struct rtable *rt = skb_rtable(skb);
-
 	if (skb->ip_summed != CHECKSUM_PARTIAL) {
-		if (!(rt->rt_flags & RTCF_LOCAL) &&
-		    (!skb->dev || skb->dev->features &
-		     (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM))) {
-			skb->ip_summed = CHECKSUM_PARTIAL;
-			skb->csum_start = skb_headroom(skb) +
-					  skb_network_offset(skb) +
-					  ip_hdrlen(skb);
-			skb->csum_offset = (void *)check - data;
-			*check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
-						    datalen, proto, 0);
-		} else {
-			*check = 0;
-			*check = csum_tcpudp_magic(iph->saddr, iph->daddr,
-						   datalen, proto,
-						   csum_partial(data, datalen,
-								0));
-			if (proto == IPPROTO_UDP && !*check)
-				*check = CSUM_MANGLED_0;
-		}
+		const struct iphdr *iph = ip_hdr(skb);
+
+		skb->ip_summed = CHECKSUM_PARTIAL;
+		skb->csum_start = skb_headroom(skb) + skb_network_offset(skb) +
+			ip_hdrlen(skb);
+		skb->csum_offset = (void *)check - data;
+		*check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, datalen,
+					    proto, 0);
 	} else
 		inet_proto_csum_replace2(check, skb,
 					 htons(oldlen), htons(datalen), true);
diff --git a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c
index 6ce3099..e0be97e 100644
--- a/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c
+++ b/net/ipv6/netfilter/nf_nat_l3proto_ipv6.c
@@ -131,29 +131,15 @@ static void nf_nat_ipv6_csum_recalc(struct sk_buff *skb,
 				    u8 proto, void *data, __sum16 *check,
 				    int datalen, int oldlen)
 {
-	const struct ipv6hdr *ipv6h = ipv6_hdr(skb);
-	struct rt6_info *rt = (struct rt6_info *)skb_dst(skb);
-
 	if (skb->ip_summed != CHECKSUM_PARTIAL) {
-		if (!(rt->rt6i_flags & RTF_LOCAL) &&
-		    (!skb->dev || skb->dev->features &
-		     (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM))) {
-			skb->ip_summed = CHECKSUM_PARTIAL;
-			skb->csum_start = skb_headroom(skb) +
-					  skb_network_offset(skb) +
-					  (data - (void *)skb->data);
-			skb->csum_offset = (void *)check - data;
-			*check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
-						  datalen, proto, 0);
-		} else {
-			*check = 0;
-			*check = csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
-						 datalen, proto,
-						 csum_partial(data, datalen,
-							      0));
-			if (proto == IPPROTO_UDP && !*check)
-				*check = CSUM_MANGLED_0;
-		}
+		const struct ipv6hdr *ipv6h = ipv6_hdr(skb);
+
+		skb->ip_summed = CHECKSUM_PARTIAL;
+		skb->csum_start = skb_headroom(skb) + skb_network_offset(skb) +
+			(data - (void *)skb->data);
+		skb->csum_offset = (void *)check - data;
+		*check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
+					  datalen, proto, 0);
 	} else
 		inet_proto_csum_replace2(check, skb,
 					 htons(oldlen), htons(datalen), true);
-- 
cgit v0.10.2


From 9f13ded8d3c715147c4759f937cfb712c185ca13 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:18 -0800
Subject: openvswitch: Add commentary to conntrack.c

This makes the code easier to understand and the following patches
more focused.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Acked-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index 3045290..2c2bf07 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -152,8 +152,12 @@ static void ovs_ct_update_key(const struct sk_buff *skb,
 	ct = nf_ct_get(skb, &ctinfo);
 	if (ct) {
 		state = ovs_ct_get_state(ctinfo);
+		/* All unconfirmed entries are NEW connections. */
 		if (!nf_ct_is_confirmed(ct))
 			state |= OVS_CS_F_NEW;
+		/* OVS persists the related flag for the duration of the
+		 * connection.
+		 */
 		if (ct->master)
 			state |= OVS_CS_F_RELATED;
 		zone = nf_ct_zone(ct);
@@ -165,6 +169,9 @@ static void ovs_ct_update_key(const struct sk_buff *skb,
 	__ovs_ct_update_key(key, state, zone, ct);
 }
 
+/* This is called to initialize CT key fields possibly coming in from the local
+ * stack.
+ */
 void ovs_ct_fill_key(const struct sk_buff *skb, struct sw_flow_key *key)
 {
 	ovs_ct_update_key(skb, NULL, key, false);
@@ -199,7 +206,6 @@ static int ovs_ct_set_mark(struct sk_buff *skb, struct sw_flow_key *key,
 	struct nf_conn *ct;
 	u32 new_mark;
 
-
 	/* The connection could be invalid, in which case set_mark is no-op. */
 	ct = nf_ct_get(skb, &ctinfo);
 	if (!ct)
@@ -375,6 +381,11 @@ static bool skb_nfct_cached(const struct net *net, const struct sk_buff *skb,
 	return true;
 }
 
+/* Pass 'skb' through conntrack in 'net', using zone configured in 'info', if
+ * not done already.  Update key with new CT state.
+ * Note that if the packet is deemed invalid by conntrack, skb->nfct will be
+ * set to NULL and 0 will be returned.
+ */
 static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 			   const struct ovs_conntrack_info *info,
 			   struct sk_buff *skb)
@@ -418,6 +429,13 @@ static int ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 {
 	struct nf_conntrack_expect *exp;
 
+	/* If we pass an expected packet through nf_conntrack_in() the
+	 * expectation is typically removed, but the packet could still be
+	 * lost in upcall processing.  To prevent this from happening we
+	 * perform an explicit expectation lookup.  Expected connections are
+	 * always new, and will be passed through conntrack only when they are
+	 * committed, as it is OK to remove the expectation at that time.
+	 */
 	exp = ovs_ct_expect_find(net, &info->zone, info->family, skb);
 	if (exp) {
 		u8 state;
@@ -455,6 +473,7 @@ static int ovs_ct_commit(struct net *net, struct sw_flow_key *key,
 	err = __ovs_ct_lookup(net, key, info, skb);
 	if (err)
 		return err;
+	/* This is a no-op if the connection has already been confirmed. */
 	if (nf_conntrack_confirm(skb) != NF_ACCEPT)
 		return -EINVAL;
 
-- 
cgit v0.10.2


From 394e910e909b174270b8231fd51942eb2f541fb9 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:19 -0800
Subject: openvswitch: Update the CT state key only after nf_conntrack_in().

Only a successful nf_conntrack_in() call can effect a connection state
change, so it suffices to update the key only after the
nf_conntrack_in() returns.

This change is needed for the later NAT patches.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Acked-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index 2c2bf07..a487bb3 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -382,7 +382,8 @@ static bool skb_nfct_cached(const struct net *net, const struct sk_buff *skb,
 }
 
 /* Pass 'skb' through conntrack in 'net', using zone configured in 'info', if
- * not done already.  Update key with new CT state.
+ * not done already.  Update key with new CT state after passing the packet
+ * through conntrack.
  * Note that if the packet is deemed invalid by conntrack, skb->nfct will be
  * set to NULL and 0 will be returned.
  */
@@ -411,14 +412,14 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 				    skb) != NF_ACCEPT)
 			return -ENOENT;
 
+		ovs_ct_update_key(skb, info, key, true);
+
 		if (ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
 			WARN_ONCE(1, "helper rejected packet");
 			return -EINVAL;
 		}
 	}
 
-	ovs_ct_update_key(skb, info, key, true);
-
 	return 0;
 }
 
-- 
cgit v0.10.2


From 289f225349cb2a97448fd14599ab34b741f706f3 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:20 -0800
Subject: openvswitch: Find existing conntrack entry after upcall.

Add a new function ovs_ct_find_existing() to find an existing
conntrack entry for which this packet was already applied to.  This is
only to be called when there is evidence that the packet was already
tracked and committed, but we lost the ct reference due to an
userspace upcall.

ovs_ct_find_existing() is called from skb_nfct_cached(), which can now
hide the fact that the ct reference may have been lost due to an
upcall.  This allows ovs_ct_commit() to be simplified.

This patch is needed by later "openvswitch: Interface with NAT" patch,
as we need to be able to pass the packet through NAT using the
original ct reference also after the reference is lost after an
upcall.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Acked-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index a487bb3..ae36fe2 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -356,14 +356,101 @@ ovs_ct_expect_find(struct net *net, const struct nf_conntrack_zone *zone,
 	return __nf_ct_expect_find(net, zone, &tuple);
 }
 
+/* This replicates logic from nf_conntrack_core.c that is not exported. */
+static enum ip_conntrack_info
+ovs_ct_get_info(const struct nf_conntrack_tuple_hash *h)
+{
+	const struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
+
+	if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY)
+		return IP_CT_ESTABLISHED_REPLY;
+	/* Once we've had two way comms, always ESTABLISHED. */
+	if (test_bit(IPS_SEEN_REPLY_BIT, &ct->status))
+		return IP_CT_ESTABLISHED;
+	if (test_bit(IPS_EXPECTED_BIT, &ct->status))
+		return IP_CT_RELATED;
+	return IP_CT_NEW;
+}
+
+/* Find an existing connection which this packet belongs to without
+ * re-attributing statistics or modifying the connection state.  This allows an
+ * skb->nfct lost due to an upcall to be recovered during actions execution.
+ *
+ * Must be called with rcu_read_lock.
+ *
+ * On success, populates skb->nfct and skb->nfctinfo, and returns the
+ * connection.  Returns NULL if there is no existing entry.
+ */
+static struct nf_conn *
+ovs_ct_find_existing(struct net *net, const struct nf_conntrack_zone *zone,
+		     u8 l3num, struct sk_buff *skb)
+{
+	struct nf_conntrack_l3proto *l3proto;
+	struct nf_conntrack_l4proto *l4proto;
+	struct nf_conntrack_tuple tuple;
+	struct nf_conntrack_tuple_hash *h;
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct;
+	unsigned int dataoff;
+	u8 protonum;
+
+	l3proto = __nf_ct_l3proto_find(l3num);
+	if (!l3proto) {
+		pr_debug("ovs_ct_find_existing: Can't get l3proto\n");
+		return NULL;
+	}
+	if (l3proto->get_l4proto(skb, skb_network_offset(skb), &dataoff,
+				 &protonum) <= 0) {
+		pr_debug("ovs_ct_find_existing: Can't get protonum\n");
+		return NULL;
+	}
+	l4proto = __nf_ct_l4proto_find(l3num, protonum);
+	if (!l4proto) {
+		pr_debug("ovs_ct_find_existing: Can't get l4proto\n");
+		return NULL;
+	}
+	if (!nf_ct_get_tuple(skb, skb_network_offset(skb), dataoff, l3num,
+			     protonum, net, &tuple, l3proto, l4proto)) {
+		pr_debug("ovs_ct_find_existing: Can't get tuple\n");
+		return NULL;
+	}
+
+	/* look for tuple match */
+	h = nf_conntrack_find_get(net, zone, &tuple);
+	if (!h)
+		return NULL;   /* Not found. */
+
+	ct = nf_ct_tuplehash_to_ctrack(h);
+
+	ctinfo = ovs_ct_get_info(h);
+	if (ctinfo == IP_CT_NEW) {
+		/* This should not happen. */
+		WARN_ONCE(1, "ovs_ct_find_existing: new packet for %p\n", ct);
+	}
+	skb->nfct = &ct->ct_general;
+	skb->nfctinfo = ctinfo;
+	return ct;
+}
+
 /* Determine whether skb->nfct is equal to the result of conntrack lookup. */
-static bool skb_nfct_cached(const struct net *net, const struct sk_buff *skb,
-			    const struct ovs_conntrack_info *info)
+static bool skb_nfct_cached(struct net *net,
+			    const struct sw_flow_key *key,
+			    const struct ovs_conntrack_info *info,
+			    struct sk_buff *skb)
 {
 	enum ip_conntrack_info ctinfo;
 	struct nf_conn *ct;
 
 	ct = nf_ct_get(skb, &ctinfo);
+	/* If no ct, check if we have evidence that an existing conntrack entry
+	 * might be found for this skb.  This happens when we lose a skb->nfct
+	 * due to an upcall.  If the connection was not confirmed, it is not
+	 * cached and needs to be run through conntrack again.
+	 */
+	if (!ct && key->ct.state & OVS_CS_F_TRACKED &&
+	    !(key->ct.state & OVS_CS_F_INVALID) &&
+	    key->ct.zone == info->zone.id)
+		ct = ovs_ct_find_existing(net, &info->zone, info->family, skb);
 	if (!ct)
 		return false;
 	if (!net_eq(net, read_pnet(&ct->ct_net)))
@@ -396,7 +483,7 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 	 * actually run the packet through conntrack twice unless it's for a
 	 * different zone.
 	 */
-	if (!skb_nfct_cached(net, skb, info)) {
+	if (!skb_nfct_cached(net, key, info, skb)) {
 		struct nf_conn *tmpl = info->ct;
 
 		/* Associate skb with specified zone. */
@@ -459,18 +546,8 @@ static int ovs_ct_commit(struct net *net, struct sw_flow_key *key,
 			 const struct ovs_conntrack_info *info,
 			 struct sk_buff *skb)
 {
-	u8 state;
 	int err;
 
-	state = key->ct.state;
-	if (key->ct.zone == info->zone.id &&
-	    ((state & OVS_CS_F_TRACKED) && !(state & OVS_CS_F_NEW))) {
-		/* Previous lookup has shown that this connection is already
-		 * tracked and committed. Skip committing.
-		 */
-		return 0;
-	}
-
 	err = __ovs_ct_lookup(net, key, info, skb);
 	if (err)
 		return err;
-- 
cgit v0.10.2


From 5b6b929376a621e2bd3367f5de563d7123506597 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:21 -0800
Subject: openvswitch: Handle NF_REPEAT in conntrack action.

Repeat the nf_conntrack_in() call when it returns NF_REPEAT.  This
avoids dropping a SYN packet re-opening an existing TCP connection.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Acked-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index ae36fe2..85256b3 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -485,6 +485,7 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 	 */
 	if (!skb_nfct_cached(net, key, info, skb)) {
 		struct nf_conn *tmpl = info->ct;
+		int err;
 
 		/* Associate skb with specified zone. */
 		if (tmpl) {
@@ -495,8 +496,13 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 			skb->nfctinfo = IP_CT_NEW;
 		}
 
-		if (nf_conntrack_in(net, info->family, NF_INET_PRE_ROUTING,
-				    skb) != NF_ACCEPT)
+		/* Repeat if requested, see nf_iterate(). */
+		do {
+			err = nf_conntrack_in(net, info->family,
+					      NF_INET_PRE_ROUTING, skb);
+		} while (err == NF_REPEAT);
+
+		if (err != NF_ACCEPT)
 			return -ENOENT;
 
 		ovs_ct_update_key(skb, info, key, true);
-- 
cgit v0.10.2


From 28b6e0c1ace45779c60e7cefe6d469b7ecb520b8 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:22 -0800
Subject: openvswitch: Delay conntrack helper call for new connections.

There is no need to help connections that are not confirmed, so we can
delay helping new connections to the time when they are confirmed.
This change is needed for NAT support, and having this as a separate
patch will make the following NAT patch a bit easier to review.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Acked-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index 85256b3..f718b72 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -483,7 +483,11 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 	 * actually run the packet through conntrack twice unless it's for a
 	 * different zone.
 	 */
-	if (!skb_nfct_cached(net, key, info, skb)) {
+	bool cached = skb_nfct_cached(net, key, info, skb);
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct;
+
+	if (!cached) {
 		struct nf_conn *tmpl = info->ct;
 		int err;
 
@@ -506,11 +510,18 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 			return -ENOENT;
 
 		ovs_ct_update_key(skb, info, key, true);
+	}
 
-		if (ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
-			WARN_ONCE(1, "helper rejected packet");
-			return -EINVAL;
-		}
+	/* Call the helper only if:
+	 * - nf_conntrack_in() was executed above ("!cached") for a confirmed
+	 *   connection, or
+	 * - When committing an unconfirmed connection.
+	 */
+	ct = nf_ct_get(skb, &ctinfo);
+	if (ct && (nf_ct_is_confirmed(ct) ? !cached : info->commit) &&
+	    ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
+		WARN_ONCE(1, "helper rejected packet");
+		return -EINVAL;
 	}
 
 	return 0;
-- 
cgit v0.10.2


From 05752523e56502cd9975aec0a2ded465d51a71f3 Mon Sep 17 00:00:00 2001
From: Jarno Rajahalme <jarno@ovn.org>
Date: Thu, 10 Mar 2016 10:54:23 -0800
Subject: openvswitch: Interface with NAT.

Extend OVS conntrack interface to cover NAT.  New nested
OVS_CT_ATTR_NAT attribute may be used to include NAT with a CT action.
A bare OVS_CT_ATTR_NAT only mangles existing and expected connections.
If OVS_NAT_ATTR_SRC or OVS_NAT_ATTR_DST is included within the nested
attributes, new (non-committed/non-confirmed) connections are mangled
according to the rest of the nested attributes.

The corresponding OVS userspace patch series includes test cases (in
tests/system-traffic.at) that also serve as example uses.

This work extends on a branch by Thomas Graf at
https://github.com/tgraf/ovs/tree/nat.

Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Acked-by: Thomas Graf <tgraf@suug.ch>
Acked-by: Joe Stringer <joe@ovn.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index a27222d..616d047 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -454,6 +454,14 @@ struct ovs_key_ct_labels {
 #define OVS_CS_F_REPLY_DIR         0x08 /* Flow is in the reply direction. */
 #define OVS_CS_F_INVALID           0x10 /* Could not track connection. */
 #define OVS_CS_F_TRACKED           0x20 /* Conntrack has occurred. */
+#define OVS_CS_F_SRC_NAT           0x40 /* Packet's source address/port was
+					 * mangled by NAT.
+					 */
+#define OVS_CS_F_DST_NAT           0x80 /* Packet's destination address/port
+					 * was mangled by NAT.
+					 */
+
+#define OVS_CS_F_NAT_MASK (OVS_CS_F_SRC_NAT | OVS_CS_F_DST_NAT)
 
 /**
  * enum ovs_flow_attr - attributes for %OVS_FLOW_* commands.
@@ -632,6 +640,8 @@ struct ovs_action_hash {
  * mask. For each bit set in the mask, the corresponding bit in the value is
  * copied to the connection tracking label field in the connection.
  * @OVS_CT_ATTR_HELPER: variable length string defining conntrack ALG.
+ * @OVS_CT_ATTR_NAT: Nested OVS_NAT_ATTR_* for performing L3 network address
+ * translation (NAT) on the packet.
  */
 enum ovs_ct_attr {
 	OVS_CT_ATTR_UNSPEC,
@@ -641,12 +651,51 @@ enum ovs_ct_attr {
 	OVS_CT_ATTR_LABELS,     /* labels to associate with this connection. */
 	OVS_CT_ATTR_HELPER,     /* netlink helper to assist detection of
 				   related connections. */
+	OVS_CT_ATTR_NAT,        /* Nested OVS_NAT_ATTR_* */
 	__OVS_CT_ATTR_MAX
 };
 
 #define OVS_CT_ATTR_MAX (__OVS_CT_ATTR_MAX - 1)
 
 /**
+ * enum ovs_nat_attr - Attributes for %OVS_CT_ATTR_NAT.
+ *
+ * @OVS_NAT_ATTR_SRC: Flag for Source NAT (mangle source address/port).
+ * @OVS_NAT_ATTR_DST: Flag for Destination NAT (mangle destination
+ * address/port).  Only one of (@OVS_NAT_ATTR_SRC, @OVS_NAT_ATTR_DST) may be
+ * specified.  Effective only for packets for ct_state NEW connections.
+ * Packets of committed connections are mangled by the NAT action according to
+ * the committed NAT type regardless of the flags specified.  As a corollary, a
+ * NAT action without a NAT type flag will only mangle packets of committed
+ * connections.  The following NAT attributes only apply for NEW
+ * (non-committed) connections, and they may be included only when the CT
+ * action has the @OVS_CT_ATTR_COMMIT flag and either @OVS_NAT_ATTR_SRC or
+ * @OVS_NAT_ATTR_DST is also included.
+ * @OVS_NAT_ATTR_IP_MIN: struct in_addr or struct in6_addr
+ * @OVS_NAT_ATTR_IP_MAX: struct in_addr or struct in6_addr
+ * @OVS_NAT_ATTR_PROTO_MIN: u16 L4 protocol specific lower boundary (port)
+ * @OVS_NAT_ATTR_PROTO_MAX: u16 L4 protocol specific upper boundary (port)
+ * @OVS_NAT_ATTR_PERSISTENT: Flag for persistent IP mapping across reboots
+ * @OVS_NAT_ATTR_PROTO_HASH: Flag for pseudo random L4 port mapping (MD5)
+ * @OVS_NAT_ATTR_PROTO_RANDOM: Flag for fully randomized L4 port mapping
+ */
+enum ovs_nat_attr {
+	OVS_NAT_ATTR_UNSPEC,
+	OVS_NAT_ATTR_SRC,
+	OVS_NAT_ATTR_DST,
+	OVS_NAT_ATTR_IP_MIN,
+	OVS_NAT_ATTR_IP_MAX,
+	OVS_NAT_ATTR_PROTO_MIN,
+	OVS_NAT_ATTR_PROTO_MAX,
+	OVS_NAT_ATTR_PERSISTENT,
+	OVS_NAT_ATTR_PROTO_HASH,
+	OVS_NAT_ATTR_PROTO_RANDOM,
+	__OVS_NAT_ATTR_MAX,
+};
+
+#define OVS_NAT_ATTR_MAX (__OVS_NAT_ATTR_MAX - 1)
+
+/**
  * enum ovs_action_attr - Action types.
  *
  * @OVS_ACTION_ATTR_OUTPUT: Output packet to port.
diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig
index cd5fd9d..234a733 100644
--- a/net/openvswitch/Kconfig
+++ b/net/openvswitch/Kconfig
@@ -6,7 +6,8 @@ config OPENVSWITCH
 	tristate "Open vSwitch"
 	depends on INET
 	depends on !NF_CONNTRACK || \
-		   (NF_CONNTRACK && (!NF_DEFRAG_IPV6 || NF_DEFRAG_IPV6))
+		   (NF_CONNTRACK && ((!NF_DEFRAG_IPV6 || NF_DEFRAG_IPV6) && \
+				     (!NF_NAT || NF_NAT)))
 	select LIBCRC32C
 	select MPLS
 	select NET_MPLS_GSO
diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index f718b72..dc5eb29 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -13,21 +13,31 @@
 
 #include <linux/module.h>
 #include <linux/openvswitch.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+#include <linux/sctp.h>
 #include <net/ip.h>
 #include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/nf_conntrack_helper.h>
 #include <net/netfilter/nf_conntrack_labels.h>
+#include <net/netfilter/nf_conntrack_seqadj.h>
 #include <net/netfilter/nf_conntrack_zones.h>
 #include <net/netfilter/ipv6/nf_defrag_ipv6.h>
 
+#ifdef CONFIG_NF_NAT_NEEDED
+#include <linux/netfilter/nf_nat.h>
+#include <net/netfilter/nf_nat_core.h>
+#include <net/netfilter/nf_nat_l3proto.h>
+#endif
+
 #include "datapath.h"
 #include "conntrack.h"
 #include "flow.h"
 #include "flow_netlink.h"
 
 struct ovs_ct_len_tbl {
-	size_t maxlen;
-	size_t minlen;
+	int maxlen;
+	int minlen;
 };
 
 /* Metadata mark for masked write to conntrack mark */
@@ -42,15 +52,25 @@ struct md_labels {
 	struct ovs_key_ct_labels mask;
 };
 
+enum ovs_ct_nat {
+	OVS_CT_NAT = 1 << 0,     /* NAT for committed connections only. */
+	OVS_CT_SRC_NAT = 1 << 1, /* Source NAT for NEW connections. */
+	OVS_CT_DST_NAT = 1 << 2, /* Destination NAT for NEW connections. */
+};
+
 /* Conntrack action context for execution. */
 struct ovs_conntrack_info {
 	struct nf_conntrack_helper *helper;
 	struct nf_conntrack_zone zone;
 	struct nf_conn *ct;
 	u8 commit : 1;
+	u8 nat : 3;                 /* enum ovs_ct_nat */
 	u16 family;
 	struct md_mark mark;
 	struct md_labels labels;
+#ifdef CONFIG_NF_NAT_NEEDED
+	struct nf_nat_range range;  /* Only present for SRC NAT and DST NAT. */
+#endif
 };
 
 static void __ovs_ct_free_action(struct ovs_conntrack_info *ct_info);
@@ -137,12 +157,15 @@ static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state,
 	ovs_ct_get_labels(ct, &key->ct.labels);
 }
 
-/* Update 'key' based on skb->nfct. If 'post_ct' is true, then OVS has
- * previously sent the packet to conntrack via the ct action.
+/* Update 'key' based on skb->nfct.  If 'post_ct' is true, then OVS has
+ * previously sent the packet to conntrack via the ct action.  If
+ * 'keep_nat_flags' is true, the existing NAT flags retained, else they are
+ * initialized from the connection status.
  */
 static void ovs_ct_update_key(const struct sk_buff *skb,
 			      const struct ovs_conntrack_info *info,
-			      struct sw_flow_key *key, bool post_ct)
+			      struct sw_flow_key *key, bool post_ct,
+			      bool keep_nat_flags)
 {
 	const struct nf_conntrack_zone *zone = &nf_ct_zone_dflt;
 	enum ip_conntrack_info ctinfo;
@@ -160,6 +183,14 @@ static void ovs_ct_update_key(const struct sk_buff *skb,
 		 */
 		if (ct->master)
 			state |= OVS_CS_F_RELATED;
+		if (keep_nat_flags) {
+			state |= key->ct.state & OVS_CS_F_NAT_MASK;
+		} else {
+			if (ct->status & IPS_SRC_NAT)
+				state |= OVS_CS_F_SRC_NAT;
+			if (ct->status & IPS_DST_NAT)
+				state |= OVS_CS_F_DST_NAT;
+		}
 		zone = nf_ct_zone(ct);
 	} else if (post_ct) {
 		state = OVS_CS_F_TRACKED | OVS_CS_F_INVALID;
@@ -174,7 +205,7 @@ static void ovs_ct_update_key(const struct sk_buff *skb,
  */
 void ovs_ct_fill_key(const struct sk_buff *skb, struct sw_flow_key *key)
 {
-	ovs_ct_update_key(skb, NULL, key, false);
+	ovs_ct_update_key(skb, NULL, key, false, false);
 }
 
 int ovs_ct_put_key(const struct sw_flow_key *key, struct sk_buff *skb)
@@ -263,6 +294,7 @@ static int ovs_ct_helper(struct sk_buff *skb, u16 proto)
 	enum ip_conntrack_info ctinfo;
 	unsigned int protoff;
 	struct nf_conn *ct;
+	int err;
 
 	ct = nf_ct_get(skb, &ctinfo);
 	if (!ct || ctinfo == IP_CT_RELATED_REPLY)
@@ -299,7 +331,18 @@ static int ovs_ct_helper(struct sk_buff *skb, u16 proto)
 		return NF_DROP;
 	}
 
-	return helper->help(skb, protoff, ct, ctinfo);
+	err = helper->help(skb, protoff, ct, ctinfo);
+	if (err != NF_ACCEPT)
+		return err;
+
+	/* Adjust seqs after helper.  This is needed due to some helpers (e.g.,
+	 * FTP with NAT) adusting the TCP payload size when mangling IP
+	 * addresses and/or port numbers in the text-based control connection.
+	 */
+	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
+	    !nf_ct_seq_adjust(skb, ct, ctinfo, protoff))
+		return NF_DROP;
+	return NF_ACCEPT;
 }
 
 /* Returns 0 on success, -EINPROGRESS if 'skb' is stolen, or other nonzero
@@ -468,6 +511,200 @@ static bool skb_nfct_cached(struct net *net,
 	return true;
 }
 
+#ifdef CONFIG_NF_NAT_NEEDED
+/* Modelled after nf_nat_ipv[46]_fn().
+ * range is only used for new, uninitialized NAT state.
+ * Returns either NF_ACCEPT or NF_DROP.
+ */
+static int ovs_ct_nat_execute(struct sk_buff *skb, struct nf_conn *ct,
+			      enum ip_conntrack_info ctinfo,
+			      const struct nf_nat_range *range,
+			      enum nf_nat_manip_type maniptype)
+{
+	int hooknum, nh_off, err = NF_ACCEPT;
+
+	nh_off = skb_network_offset(skb);
+	skb_pull(skb, nh_off);
+
+	/* See HOOK2MANIP(). */
+	if (maniptype == NF_NAT_MANIP_SRC)
+		hooknum = NF_INET_LOCAL_IN; /* Source NAT */
+	else
+		hooknum = NF_INET_LOCAL_OUT; /* Destination NAT */
+
+	switch (ctinfo) {
+	case IP_CT_RELATED:
+	case IP_CT_RELATED_REPLY:
+		if (skb->protocol == htons(ETH_P_IP) &&
+		    ip_hdr(skb)->protocol == IPPROTO_ICMP) {
+			if (!nf_nat_icmp_reply_translation(skb, ct, ctinfo,
+							   hooknum))
+				err = NF_DROP;
+			goto push;
+#if IS_ENABLED(CONFIG_NF_NAT_IPV6)
+		} else if (skb->protocol == htons(ETH_P_IPV6)) {
+			__be16 frag_off;
+			u8 nexthdr = ipv6_hdr(skb)->nexthdr;
+			int hdrlen = ipv6_skip_exthdr(skb,
+						      sizeof(struct ipv6hdr),
+						      &nexthdr, &frag_off);
+
+			if (hdrlen >= 0 && nexthdr == IPPROTO_ICMPV6) {
+				if (!nf_nat_icmpv6_reply_translation(skb, ct,
+								     ctinfo,
+								     hooknum,
+								     hdrlen))
+					err = NF_DROP;
+				goto push;
+			}
+#endif
+		}
+		/* Non-ICMP, fall thru to initialize if needed. */
+	case IP_CT_NEW:
+		/* Seen it before?  This can happen for loopback, retrans,
+		 * or local packets.
+		 */
+		if (!nf_nat_initialized(ct, maniptype)) {
+			/* Initialize according to the NAT action. */
+			err = (range && range->flags & NF_NAT_RANGE_MAP_IPS)
+				/* Action is set up to establish a new
+				 * mapping.
+				 */
+				? nf_nat_setup_info(ct, range, maniptype)
+				: nf_nat_alloc_null_binding(ct, hooknum);
+			if (err != NF_ACCEPT)
+				goto push;
+		}
+		break;
+
+	case IP_CT_ESTABLISHED:
+	case IP_CT_ESTABLISHED_REPLY:
+		break;
+
+	default:
+		err = NF_DROP;
+		goto push;
+	}
+
+	err = nf_nat_packet(ct, ctinfo, hooknum, skb);
+push:
+	skb_push(skb, nh_off);
+
+	return err;
+}
+
+static void ovs_nat_update_key(struct sw_flow_key *key,
+			       const struct sk_buff *skb,
+			       enum nf_nat_manip_type maniptype)
+{
+	if (maniptype == NF_NAT_MANIP_SRC) {
+		__be16 src;
+
+		key->ct.state |= OVS_CS_F_SRC_NAT;
+		if (key->eth.type == htons(ETH_P_IP))
+			key->ipv4.addr.src = ip_hdr(skb)->saddr;
+		else if (key->eth.type == htons(ETH_P_IPV6))
+			memcpy(&key->ipv6.addr.src, &ipv6_hdr(skb)->saddr,
+			       sizeof(key->ipv6.addr.src));
+		else
+			return;
+
+		if (key->ip.proto == IPPROTO_UDP)
+			src = udp_hdr(skb)->source;
+		else if (key->ip.proto == IPPROTO_TCP)
+			src = tcp_hdr(skb)->source;
+		else if (key->ip.proto == IPPROTO_SCTP)
+			src = sctp_hdr(skb)->source;
+		else
+			return;
+
+		key->tp.src = src;
+	} else {
+		__be16 dst;
+
+		key->ct.state |= OVS_CS_F_DST_NAT;
+		if (key->eth.type == htons(ETH_P_IP))
+			key->ipv4.addr.dst = ip_hdr(skb)->daddr;
+		else if (key->eth.type == htons(ETH_P_IPV6))
+			memcpy(&key->ipv6.addr.dst, &ipv6_hdr(skb)->daddr,
+			       sizeof(key->ipv6.addr.dst));
+		else
+			return;
+
+		if (key->ip.proto == IPPROTO_UDP)
+			dst = udp_hdr(skb)->dest;
+		else if (key->ip.proto == IPPROTO_TCP)
+			dst = tcp_hdr(skb)->dest;
+		else if (key->ip.proto == IPPROTO_SCTP)
+			dst = sctp_hdr(skb)->dest;
+		else
+			return;
+
+		key->tp.dst = dst;
+	}
+}
+
+/* Returns NF_DROP if the packet should be dropped, NF_ACCEPT otherwise. */
+static int ovs_ct_nat(struct net *net, struct sw_flow_key *key,
+		      const struct ovs_conntrack_info *info,
+		      struct sk_buff *skb, struct nf_conn *ct,
+		      enum ip_conntrack_info ctinfo)
+{
+	enum nf_nat_manip_type maniptype;
+	int err;
+
+	if (nf_ct_is_untracked(ct)) {
+		/* A NAT action may only be performed on tracked packets. */
+		return NF_ACCEPT;
+	}
+
+	/* Add NAT extension if not confirmed yet. */
+	if (!nf_ct_is_confirmed(ct) && !nf_ct_nat_ext_add(ct))
+		return NF_ACCEPT;   /* Can't NAT. */
+
+	/* Determine NAT type.
+	 * Check if the NAT type can be deduced from the tracked connection.
+	 * Make sure expected traffic is NATted only when committing.
+	 */
+	if (info->nat & OVS_CT_NAT && ctinfo != IP_CT_NEW &&
+	    ct->status & IPS_NAT_MASK &&
+	    (!(ct->status & IPS_EXPECTED_BIT) || info->commit)) {
+		/* NAT an established or related connection like before. */
+		if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY)
+			/* This is the REPLY direction for a connection
+			 * for which NAT was applied in the forward
+			 * direction.  Do the reverse NAT.
+			 */
+			maniptype = ct->status & IPS_SRC_NAT
+				? NF_NAT_MANIP_DST : NF_NAT_MANIP_SRC;
+		else
+			maniptype = ct->status & IPS_SRC_NAT
+				? NF_NAT_MANIP_SRC : NF_NAT_MANIP_DST;
+	} else if (info->nat & OVS_CT_SRC_NAT) {
+		maniptype = NF_NAT_MANIP_SRC;
+	} else if (info->nat & OVS_CT_DST_NAT) {
+		maniptype = NF_NAT_MANIP_DST;
+	} else {
+		return NF_ACCEPT; /* Connection is not NATed. */
+	}
+	err = ovs_ct_nat_execute(skb, ct, ctinfo, &info->range, maniptype);
+
+	/* Mark NAT done if successful and update the flow key. */
+	if (err == NF_ACCEPT)
+		ovs_nat_update_key(key, skb, maniptype);
+
+	return err;
+}
+#else /* !CONFIG_NF_NAT_NEEDED */
+static int ovs_ct_nat(struct net *net, struct sw_flow_key *key,
+		      const struct ovs_conntrack_info *info,
+		      struct sk_buff *skb, struct nf_conn *ct,
+		      enum ip_conntrack_info ctinfo)
+{
+	return NF_ACCEPT;
+}
+#endif
+
 /* Pass 'skb' through conntrack in 'net', using zone configured in 'info', if
  * not done already.  Update key with new CT state after passing the packet
  * through conntrack.
@@ -509,19 +746,43 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 		if (err != NF_ACCEPT)
 			return -ENOENT;
 
-		ovs_ct_update_key(skb, info, key, true);
+		/* Clear CT state NAT flags to mark that we have not yet done
+		 * NAT after the nf_conntrack_in() call.  We can actually clear
+		 * the whole state, as it will be re-initialized below.
+		 */
+		key->ct.state = 0;
+
+		/* Update the key, but keep the NAT flags. */
+		ovs_ct_update_key(skb, info, key, true, true);
 	}
 
-	/* Call the helper only if:
-	 * - nf_conntrack_in() was executed above ("!cached") for a confirmed
-	 *   connection, or
-	 * - When committing an unconfirmed connection.
-	 */
 	ct = nf_ct_get(skb, &ctinfo);
-	if (ct && (nf_ct_is_confirmed(ct) ? !cached : info->commit) &&
-	    ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
-		WARN_ONCE(1, "helper rejected packet");
-		return -EINVAL;
+	if (ct) {
+		/* Packets starting a new connection must be NATted before the
+		 * helper, so that the helper knows about the NAT.  We enforce
+		 * this by delaying both NAT and helper calls for unconfirmed
+		 * connections until the committing CT action.  For later
+		 * packets NAT and Helper may be called in either order.
+		 *
+		 * NAT will be done only if the CT action has NAT, and only
+		 * once per packet (per zone), as guarded by the NAT bits in
+		 * the key->ct.state.
+		 */
+		if (info->nat && !(key->ct.state & OVS_CS_F_NAT_MASK) &&
+		    (nf_ct_is_confirmed(ct) || info->commit) &&
+		    ovs_ct_nat(net, key, info, skb, ct, ctinfo) != NF_ACCEPT) {
+			return -EINVAL;
+		}
+
+		/* Call the helper only if:
+		 * - nf_conntrack_in() was executed above ("!cached") for a
+		 *   confirmed connection, or
+		 * - When committing an unconfirmed connection.
+		 */
+		if ((nf_ct_is_confirmed(ct) ? !cached : info->commit) &&
+		    ovs_ct_helper(skb, info->family) != NF_ACCEPT) {
+			return -EINVAL;
+		}
 	}
 
 	return 0;
@@ -545,15 +806,13 @@ static int ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 	if (exp) {
 		u8 state;
 
+		/* NOTE: New connections are NATted and Helped only when
+		 * committed, so we are not calling into NAT here.
+		 */
 		state = OVS_CS_F_TRACKED | OVS_CS_F_NEW | OVS_CS_F_RELATED;
 		__ovs_ct_update_key(key, state, &info->zone, exp->master);
-	} else {
-		int err;
-
-		err = __ovs_ct_lookup(net, key, info, skb);
-		if (err)
-			return err;
-	}
+	} else
+		return __ovs_ct_lookup(net, key, info, skb);
 
 	return 0;
 }
@@ -653,6 +912,135 @@ static int ovs_ct_add_helper(struct ovs_conntrack_info *info, const char *name,
 	return 0;
 }
 
+#ifdef CONFIG_NF_NAT_NEEDED
+static int parse_nat(const struct nlattr *attr,
+		     struct ovs_conntrack_info *info, bool log)
+{
+	struct nlattr *a;
+	int rem;
+	bool have_ip_max = false;
+	bool have_proto_max = false;
+	bool ip_vers = (info->family == NFPROTO_IPV6);
+
+	nla_for_each_nested(a, attr, rem) {
+		static const int ovs_nat_attr_lens[OVS_NAT_ATTR_MAX + 1][2] = {
+			[OVS_NAT_ATTR_SRC] = {0, 0},
+			[OVS_NAT_ATTR_DST] = {0, 0},
+			[OVS_NAT_ATTR_IP_MIN] = {sizeof(struct in_addr),
+						 sizeof(struct in6_addr)},
+			[OVS_NAT_ATTR_IP_MAX] = {sizeof(struct in_addr),
+						 sizeof(struct in6_addr)},
+			[OVS_NAT_ATTR_PROTO_MIN] = {sizeof(u16), sizeof(u16)},
+			[OVS_NAT_ATTR_PROTO_MAX] = {sizeof(u16), sizeof(u16)},
+			[OVS_NAT_ATTR_PERSISTENT] = {0, 0},
+			[OVS_NAT_ATTR_PROTO_HASH] = {0, 0},
+			[OVS_NAT_ATTR_PROTO_RANDOM] = {0, 0},
+		};
+		int type = nla_type(a);
+
+		if (type > OVS_NAT_ATTR_MAX) {
+			OVS_NLERR(log,
+				  "Unknown NAT attribute (type=%d, max=%d).\n",
+				  type, OVS_NAT_ATTR_MAX);
+			return -EINVAL;
+		}
+
+		if (nla_len(a) != ovs_nat_attr_lens[type][ip_vers]) {
+			OVS_NLERR(log,
+				  "NAT attribute type %d has unexpected length (%d != %d).\n",
+				  type, nla_len(a),
+				  ovs_nat_attr_lens[type][ip_vers]);
+			return -EINVAL;
+		}
+
+		switch (type) {
+		case OVS_NAT_ATTR_SRC:
+		case OVS_NAT_ATTR_DST:
+			if (info->nat) {
+				OVS_NLERR(log,
+					  "Only one type of NAT may be specified.\n"
+					  );
+				return -ERANGE;
+			}
+			info->nat |= OVS_CT_NAT;
+			info->nat |= ((type == OVS_NAT_ATTR_SRC)
+					? OVS_CT_SRC_NAT : OVS_CT_DST_NAT);
+			break;
+
+		case OVS_NAT_ATTR_IP_MIN:
+			nla_memcpy(&info->range.min_addr, a, nla_len(a));
+			info->range.flags |= NF_NAT_RANGE_MAP_IPS;
+			break;
+
+		case OVS_NAT_ATTR_IP_MAX:
+			have_ip_max = true;
+			nla_memcpy(&info->range.max_addr, a,
+				   sizeof(info->range.max_addr));
+			info->range.flags |= NF_NAT_RANGE_MAP_IPS;
+			break;
+
+		case OVS_NAT_ATTR_PROTO_MIN:
+			info->range.min_proto.all = htons(nla_get_u16(a));
+			info->range.flags |= NF_NAT_RANGE_PROTO_SPECIFIED;
+			break;
+
+		case OVS_NAT_ATTR_PROTO_MAX:
+			have_proto_max = true;
+			info->range.max_proto.all = htons(nla_get_u16(a));
+			info->range.flags |= NF_NAT_RANGE_PROTO_SPECIFIED;
+			break;
+
+		case OVS_NAT_ATTR_PERSISTENT:
+			info->range.flags |= NF_NAT_RANGE_PERSISTENT;
+			break;
+
+		case OVS_NAT_ATTR_PROTO_HASH:
+			info->range.flags |= NF_NAT_RANGE_PROTO_RANDOM;
+			break;
+
+		case OVS_NAT_ATTR_PROTO_RANDOM:
+			info->range.flags |= NF_NAT_RANGE_PROTO_RANDOM_FULLY;
+			break;
+
+		default:
+			OVS_NLERR(log, "Unknown nat attribute (%d).\n", type);
+			return -EINVAL;
+		}
+	}
+
+	if (rem > 0) {
+		OVS_NLERR(log, "NAT attribute has %d unknown bytes.\n", rem);
+		return -EINVAL;
+	}
+	if (!info->nat) {
+		/* Do not allow flags if no type is given. */
+		if (info->range.flags) {
+			OVS_NLERR(log,
+				  "NAT flags may be given only when NAT range (SRC or DST) is also specified.\n"
+				  );
+			return -EINVAL;
+		}
+		info->nat = OVS_CT_NAT;   /* NAT existing connections. */
+	} else if (!info->commit) {
+		OVS_NLERR(log,
+			  "NAT attributes may be specified only when CT COMMIT flag is also specified.\n"
+			  );
+		return -EINVAL;
+	}
+	/* Allow missing IP_MAX. */
+	if (info->range.flags & NF_NAT_RANGE_MAP_IPS && !have_ip_max) {
+		memcpy(&info->range.max_addr, &info->range.min_addr,
+		       sizeof(info->range.max_addr));
+	}
+	/* Allow missing PROTO_MAX. */
+	if (info->range.flags & NF_NAT_RANGE_PROTO_SPECIFIED &&
+	    !have_proto_max) {
+		info->range.max_proto.all = info->range.min_proto.all;
+	}
+	return 0;
+}
+#endif
+
 static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = {
 	[OVS_CT_ATTR_COMMIT]	= { .minlen = 0, .maxlen = 0 },
 	[OVS_CT_ATTR_ZONE]	= { .minlen = sizeof(u16),
@@ -662,7 +1050,11 @@ static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = {
 	[OVS_CT_ATTR_LABELS]	= { .minlen = sizeof(struct md_labels),
 				    .maxlen = sizeof(struct md_labels) },
 	[OVS_CT_ATTR_HELPER]	= { .minlen = 1,
-				    .maxlen = NF_CT_HELPER_NAME_LEN }
+				    .maxlen = NF_CT_HELPER_NAME_LEN },
+#ifdef CONFIG_NF_NAT_NEEDED
+	/* NAT length is checked when parsing the nested attributes. */
+	[OVS_CT_ATTR_NAT]	= { .minlen = 0, .maxlen = INT_MAX },
+#endif
 };
 
 static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info,
@@ -729,6 +1121,15 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info,
 				return -EINVAL;
 			}
 			break;
+#ifdef CONFIG_NF_NAT_NEEDED
+		case OVS_CT_ATTR_NAT: {
+			int err = parse_nat(a, info, log);
+
+			if (err)
+				return err;
+			break;
+		}
+#endif
 		default:
 			OVS_NLERR(log, "Unknown conntrack attr (%d)",
 				  type);
@@ -816,6 +1217,74 @@ err_free_ct:
 	return err;
 }
 
+#ifdef CONFIG_NF_NAT_NEEDED
+static bool ovs_ct_nat_to_attr(const struct ovs_conntrack_info *info,
+			       struct sk_buff *skb)
+{
+	struct nlattr *start;
+
+	start = nla_nest_start(skb, OVS_CT_ATTR_NAT);
+	if (!start)
+		return false;
+
+	if (info->nat & OVS_CT_SRC_NAT) {
+		if (nla_put_flag(skb, OVS_NAT_ATTR_SRC))
+			return false;
+	} else if (info->nat & OVS_CT_DST_NAT) {
+		if (nla_put_flag(skb, OVS_NAT_ATTR_DST))
+			return false;
+	} else {
+		goto out;
+	}
+
+	if (info->range.flags & NF_NAT_RANGE_MAP_IPS) {
+		if (info->family == NFPROTO_IPV4) {
+			if (nla_put_in_addr(skb, OVS_NAT_ATTR_IP_MIN,
+					    info->range.min_addr.ip) ||
+			    (info->range.max_addr.ip
+			     != info->range.min_addr.ip &&
+			     (nla_put_in_addr(skb, OVS_NAT_ATTR_IP_MAX,
+					      info->range.max_addr.ip))))
+				return false;
+#if IS_ENABLED(CONFIG_NF_NAT_IPV6)
+		} else if (info->family == NFPROTO_IPV6) {
+			if (nla_put_in6_addr(skb, OVS_NAT_ATTR_IP_MIN,
+					     &info->range.min_addr.in6) ||
+			    (memcmp(&info->range.max_addr.in6,
+				    &info->range.min_addr.in6,
+				    sizeof(info->range.max_addr.in6)) &&
+			     (nla_put_in6_addr(skb, OVS_NAT_ATTR_IP_MAX,
+					       &info->range.max_addr.in6))))
+				return false;
+#endif
+		} else {
+			return false;
+		}
+	}
+	if (info->range.flags & NF_NAT_RANGE_PROTO_SPECIFIED &&
+	    (nla_put_u16(skb, OVS_NAT_ATTR_PROTO_MIN,
+			 ntohs(info->range.min_proto.all)) ||
+	     (info->range.max_proto.all != info->range.min_proto.all &&
+	      nla_put_u16(skb, OVS_NAT_ATTR_PROTO_MAX,
+			  ntohs(info->range.max_proto.all)))))
+		return false;
+
+	if (info->range.flags & NF_NAT_RANGE_PERSISTENT &&
+	    nla_put_flag(skb, OVS_NAT_ATTR_PERSISTENT))
+		return false;
+	if (info->range.flags & NF_NAT_RANGE_PROTO_RANDOM &&
+	    nla_put_flag(skb, OVS_NAT_ATTR_PROTO_HASH))
+		return false;
+	if (info->range.flags & NF_NAT_RANGE_PROTO_RANDOM_FULLY &&
+	    nla_put_flag(skb, OVS_NAT_ATTR_PROTO_RANDOM))
+		return false;
+out:
+	nla_nest_end(skb, start);
+
+	return true;
+}
+#endif
+
 int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info,
 			  struct sk_buff *skb)
 {
@@ -844,7 +1313,10 @@ int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info,
 				   ct_info->helper->name))
 			return -EMSGSIZE;
 	}
-
+#ifdef CONFIG_NF_NAT_NEEDED
+	if (ct_info->nat && !ovs_ct_nat_to_attr(ct_info, skb))
+		return -EMSGSIZE;
+#endif
 	nla_nest_end(skb, start);
 
 	return 0;
diff --git a/net/openvswitch/conntrack.h b/net/openvswitch/conntrack.h
index a7544f4..8f6230b 100644
--- a/net/openvswitch/conntrack.h
+++ b/net/openvswitch/conntrack.h
@@ -37,7 +37,8 @@ void ovs_ct_free_action(const struct nlattr *a);
 
 #define CT_SUPPORTED_MASK (OVS_CS_F_NEW | OVS_CS_F_ESTABLISHED | \
 			   OVS_CS_F_RELATED | OVS_CS_F_REPLY_DIR | \
-			   OVS_CS_F_INVALID | OVS_CS_F_TRACKED)
+			   OVS_CS_F_INVALID | OVS_CS_F_TRACKED | \
+			   OVS_CS_F_SRC_NAT | OVS_CS_F_DST_NAT)
 #else
 #include <linux/errno.h>
 
-- 
cgit v0.10.2


From e3ebd894f084255fde19116955ba7054858ff5d6 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Mar 2016 23:45:12 +0100
Subject: smc91x: avoid self-comparison warning

The smc91x driver defines a macro that compares its argument to
itself, apparently to get a true result while using its argument
to avoid a warning about unused local variables.

Unfortunately, this triggers a warning with gcc-6, as the comparison
is obviously useless:

drivers/net/ethernet/smsc/smc91x.c: In function 'smc_hardware_send_pkt':
drivers/net/ethernet/smsc/smc91x.c:563:14: error: self-comparison always evaluates to true [-Werror=tautological-compare]
  if (!smc_special_trylock(&lp->lock, flags)) {

This replaces the macro with another one that behaves similarly,
with a cast to (void) to ensure the argument is used, and using
a literal 'true' as its value.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c
index db7db8a..c5ed27c 100644
--- a/drivers/net/ethernet/smsc/smc91x.c
+++ b/drivers/net/ethernet/smsc/smc91x.c
@@ -540,7 +540,7 @@ static inline void  smc_rcv(struct net_device *dev)
 #define smc_special_lock(lock, flags)		spin_lock_irqsave(lock, flags)
 #define smc_special_unlock(lock, flags) 	spin_unlock_irqrestore(lock, flags)
 #else
-#define smc_special_trylock(lock, flags)	(flags == flags)
+#define smc_special_trylock(lock, flags)	((void)flags, true)
 #define smc_special_lock(lock, flags)   	do { flags = 0; } while (0)
 #define smc_special_unlock(lock, flags)	do { flags = 0; } while (0)
 #endif
-- 
cgit v0.10.2


From acffb584cda7069b0c2c83045503ccd07516a891 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 14 Mar 2016 15:40:00 -0700
Subject: net: diag: add a scheduling point in inet_diag_dump_icsk()

On loaded TCP servers, looking at millions of sockets can hold
cpu for many seconds, if the lookup condition is very narrow.

(eg : ss dst 1.2.3.4 )

Better add a cond_resched() to allow other processes to access
the cpu.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 50c0d96..5fdb02f 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -879,6 +879,7 @@ next_normal:
 		}
 
 		spin_unlock_bh(lock);
+		cond_resched();
 	}
 
 done:
-- 
cgit v0.10.2


From e39365be031e37b229f745ea49db0b25e82436fa Mon Sep 17 00:00:00 2001
From: Nicholas Mc Guire <hofrat@osadl.org>
Date: Mon, 14 Mar 2016 12:39:02 +0100
Subject: netfilter: nf_conntrack: consolidate lock/unlock into unlock_wait

The spin_lock()/spin_unlock() is synchronizing on the
nf_conntrack_locks_all_lock which is equivalent to
spin_unlock_wait() but the later should be more efficient.

Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index f60b4fd..afde5f5 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -74,8 +74,7 @@ void nf_conntrack_lock(spinlock_t *lock) __acquires(lock)
 	spin_lock(lock);
 	while (unlikely(nf_conntrack_locks_all)) {
 		spin_unlock(lock);
-		spin_lock(&nf_conntrack_locks_all_lock);
-		spin_unlock(&nf_conntrack_locks_all_lock);
+		spin_unlock_wait(&nf_conntrack_locks_all_lock);
 		spin_lock(lock);
 	}
 }
@@ -121,8 +120,7 @@ static void nf_conntrack_all_lock(void)
 	nf_conntrack_locks_all = true;
 
 	for (i = 0; i < CONNTRACK_LOCKS; i++) {
-		spin_lock(&nf_conntrack_locks[i]);
-		spin_unlock(&nf_conntrack_locks[i]);
+		spin_unlock_wait(&nf_conntrack_locks[i]);
 	}
 }
 
-- 
cgit v0.10.2


From 489ce5f441e4094a3419d7e8e29fd756b7f391d4 Mon Sep 17 00:00:00 2001
From: Nicholas Mc Guire <hofrat@osadl.org>
Date: Sun, 13 Mar 2016 11:48:24 +0100
Subject: sctp: consolidate local_bh_disable/enable + spin_lock/unlock to _bh
 variant

local_bh_disable() + spin_lock() is equivalent to spin_lock_bh(), same for
the unlock/enable case, so replace the calls by the appropriate wrappers.

Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index b89501e..f28ecf8 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -7254,14 +7254,12 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk,
 	/* Hook this new socket in to the bind_hash list. */
 	head = &sctp_port_hashtable[sctp_phashfn(sock_net(oldsk),
 						 inet_sk(oldsk)->inet_num)];
-	local_bh_disable();
-	spin_lock(&head->lock);
+	spin_lock_bh(&head->lock);
 	pp = sctp_sk(oldsk)->bind_hash;
 	sk_add_bind_node(newsk, &pp->owner);
 	sctp_sk(newsk)->bind_hash = pp;
 	inet_sk(newsk)->inet_num = inet_sk(oldsk)->inet_num;
-	spin_unlock(&head->lock);
-	local_bh_enable();
+	spin_unlock_bh(&head->lock);
 
 	/* Copy the bind_addr list from the original endpoint to the new
 	 * endpoint so that we can handle restarts properly
-- 
cgit v0.10.2


From d78a1f08458f8118cf42154e55a3006cc17502bc Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Sun, 13 Mar 2016 21:21:48 +0200
Subject: bnx2x: don't wait for Tx completion on recovery

When driver has hit a parity event, HW can no longer write to host memory.
As a result, Tx completions cannot be written to the host SB memory, and
waiting for Tx completions eventually timeout.
As driver is willing to delay as much as 1-2 seconds per Tx queue for its
draining and this delay is sequential, the time to recover might greatly
lengthen needlessly in case the recovery is done under multi-connection
traffic.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index a949783..0a9108c 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -3042,8 +3042,12 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
 		bnx2x_save_statistics(bp);
 	}
 
-	/* wait till consumers catch up with producers in all queues */
-	bnx2x_drain_tx_queues(bp);
+	/* wait till consumers catch up with producers in all queues.
+	 * If we're recovering, FW can't write to host so no reason
+	 * to wait for the queues to complete all Tx.
+	 */
+	if (unload_mode != UNLOAD_RECOVERY)
+		bnx2x_drain_tx_queues(bp);
 
 	/* if VF indicate to PF this function is going down (PF will delete sp
 	 * elements and clear initializations
-- 
cgit v0.10.2


From ad846aa52ca7b7ec6b99a8b13c4ce0bbe7b15b7f Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Mon, 14 Mar 2016 01:09:53 +0300
Subject: sh_eth: do not call netif_start_queue() from sh_eth_dev_init()

Iff  sh_eth_phy_start() call fails in sh_eth_open(), the netif_start_queue()
call done by sh_eth_dev_init()  is not undone.  In order to deal with that,
stop calling netif_start_queue()  from there, so that it can be called only
when the device is fully opened and sh_eth_dev_init() only deals with the
hardware initialization, symmetrically to sh_eth_dev_exit()...

Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index 8385d92..004e2d7 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -1312,8 +1312,6 @@ static int sh_eth_dev_init(struct net_device *ndev, bool start)
 	if (start) {
 		/* Setting the Rx mode will start the Rx process. */
 		sh_eth_write(ndev, EDRRR_R, EDRRR);
-
-		netif_start_queue(ndev);
 	}
 
 	return ret;
@@ -2261,6 +2259,8 @@ static int sh_eth_open(struct net_device *ndev)
 	if (ret)
 		goto out_free_irq;
 
+	netif_start_queue(ndev);
+
 	mdp->is_opened = 1;
 
 	return ret;
@@ -2304,6 +2304,8 @@ static void sh_eth_tx_timeout(struct net_device *ndev)
 
 	/* device init */
 	sh_eth_dev_init(ndev, true);
+
+	netif_start_queue(ndev);
 }
 
 /* Packet transmit function */
-- 
cgit v0.10.2


From 793cf87de9d1a62dc9079c3ec5fcc01cfc62fafb Mon Sep 17 00:00:00 2001
From: Ben Hutchings <ben@decadent.org.uk>
Date: Mon, 14 Mar 2016 01:05:38 +0000
Subject: ethtool: Set cmd field in ETHTOOL_GLINKSETTINGS response to wrong
 nwords

When the ETHTOOL_GLINKSETTINGS implementation finds that userland is
using the wrong number of words of link mode bitmaps (or is trying to
find out the right numbers) it sets the cmd field to 0 in the response
structure.

This is inconsistent with the implementation of every other ethtool
command, so let's remove that inconsistency before it gets into a
stable release.

Fixes: 3f1ac7a700d03 ("net: ethtool: add new ETHTOOL_xLINKSETTINGS API")
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 2966cd0..f426c5a 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -655,7 +655,7 @@ static int ethtool_get_link_ksettings(struct net_device *dev,
 	    != link_ksettings.base.link_mode_masks_nwords) {
 		/* wrong link mode nbits requested */
 		memset(&link_ksettings, 0, sizeof(link_ksettings));
-		/* keep cmd field reset to 0 */
+		link_ksettings.base.cmd = ETHTOOL_GLINKSETTINGS;
 		/* send back number of words required as negative val */
 		compiletime_assert(__ETHTOOL_LINK_MODE_MASK_NU32 <= S8_MAX,
 				   "need too many bits for link modes!");
-- 
cgit v0.10.2


From 434242cd6080dde0103d3b4f28daec5c742c465e Mon Sep 17 00:00:00 2001
From: Caesar Wang <wxt@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:52 +0800
Subject: net: arc_emac: make the rockchip emac document more compatible

Add the rk3036 SoCs to match driver for document since the emac driver
has supported the rk3036 SoCs.

This patch adds the rk3036/rk3066/rk3188 SoCS to compatible for rockchip
emac ducument. Also, that will suit for other SoCs in the future.

Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: devicetree@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/emac_rockchip.txt b/Documentation/devicetree/bindings/net/emac_rockchip.txt
index 8dc1c79..05bd7da 100644
--- a/Documentation/devicetree/bindings/net/emac_rockchip.txt
+++ b/Documentation/devicetree/bindings/net/emac_rockchip.txt
@@ -1,8 +1,10 @@
-* ARC EMAC 10/100 Ethernet platform driver for Rockchip Rk3066/RK3188 SoCs
+* ARC EMAC 10/100 Ethernet platform driver for Rockchip RK3036/RK3066/RK3188 SoCs
 
 Required properties:
-- compatible: Should be "rockchip,rk3066-emac" or "rockchip,rk3188-emac"
-  according to the target SoC.
+- compatible: should be "rockchip,<name>-emac"
+   "rockchip,rk3036-emac": found on RK3036 SoCs
+   "rockchip,rk3066-emac": found on RK3066 SoCs
+   "rockchip,rk3188-emac": found on RK3188 SoCs
 - reg: Address and length of the register set for the device
 - interrupts: Should contain the EMAC interrupts
 - rockchip,grf: phandle to the syscon grf used to control speed and mode
-- 
cgit v0.10.2


From 8700eee6271c932b2747a6b157655f546c27e7ad Mon Sep 17 00:00:00 2001
From: Caesar Wang <wxt@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:53 +0800
Subject: net: arc_emac: add phy reset is optional for device tree

This patch adds the following property for arc_emac.

1) phy-reset-gpios:
The phy-reset-gpio is an optional property for arc emac device tree boot.
Change the binding document to match the driver code.

2) phy-reset-duration:
Different boards may require different phy reset duration. Add property
phy-reset-duration for device tree probe, so that the boards that need
a longer reset duration can specify it in their device tree.

Anyway, we can add the above property for arc emac.

Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: devicetree@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Cc; Alexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/Documentation/devicetree/bindings/net/arc_emac.txt b/Documentation/devicetree/bindings/net/arc_emac.txt
index a1d71eb..c73a0e9 100644
--- a/Documentation/devicetree/bindings/net/arc_emac.txt
+++ b/Documentation/devicetree/bindings/net/arc_emac.txt
@@ -7,6 +7,13 @@ Required properties:
 - max-speed: see ethernet.txt file in the same directory.
 - phy: see ethernet.txt file in the same directory.
 
+Optional properties:
+- phy-reset-gpios : Should specify the gpio for phy reset
+- phy-reset-duration : Reset duration in milliseconds.  Should present
+  only if property "phy-reset-gpios" is available.  Missing the property
+  will have the duration be 1 millisecond.  Numbers greater than 1000 are
+  invalid and 1 millisecond will be used instead.
+
 Clock handling:
 The clock frequency is needed to calculate and set polling period of EMAC.
 It must be provided by one of:
-- 
cgit v0.10.2


From 1bddd96cba03da0a14b3e5144e98c9a6ff17e983 Mon Sep 17 00:00:00 2001
From: Caesar Wang <wxt@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:54 +0800
Subject: net: arc_emac: support the phy reset for emac driver

This patch adds to support the emac phy reset.

Different boards may require different phy reset duration. Add property
phy-reset-duration for emac driver, so that the boards that need
a longer reset duration can specify it in their device tree.

Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: Alexander Kochetkov <al.kochet@gmail.com>
Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/arc/emac.h b/drivers/net/ethernet/arc/emac.h
index dae1ac3..1a40403 100644
--- a/drivers/net/ethernet/arc/emac.h
+++ b/drivers/net/ethernet/arc/emac.h
@@ -102,6 +102,11 @@ struct buffer_state {
 	DEFINE_DMA_UNMAP_LEN(len);
 };
 
+struct arc_emac_mdio_bus_data {
+	struct gpio_desc *reset_gpio;
+	int msec;
+};
+
 /**
  * struct arc_emac_priv - Storage of EMAC's private information.
  * @dev:	Pointer to the current device.
@@ -131,6 +136,7 @@ struct arc_emac_priv {
 	struct device *dev;
 	struct phy_device *phy_dev;
 	struct mii_bus *bus;
+	struct arc_emac_mdio_bus_data bus_data;
 
 	void __iomem *regs;
 	struct clk *clk;
diff --git a/drivers/net/ethernet/arc/emac_mdio.c b/drivers/net/ethernet/arc/emac_mdio.c
index d5ee986..caf7042 100644
--- a/drivers/net/ethernet/arc/emac_mdio.c
+++ b/drivers/net/ethernet/arc/emac_mdio.c
@@ -7,6 +7,7 @@
 #include <linux/delay.h>
 #include <linux/of_mdio.h>
 #include <linux/platform_device.h>
+#include <linux/gpio/consumer.h>
 
 #include "emac.h"
 
@@ -99,6 +100,25 @@ static int arc_mdio_write(struct mii_bus *bus, int phy_addr,
 }
 
 /**
+ * arc_mdio_reset
+ * @bus: points to the mii_bus structure
+ * Description: reset the MII bus
+ */
+int arc_mdio_reset(struct mii_bus *bus)
+{
+	struct arc_emac_priv *priv = bus->priv;
+	struct arc_emac_mdio_bus_data *data = &priv->bus_data;
+
+	if (data->reset_gpio) {
+		gpiod_set_value_cansleep(data->reset_gpio, 1);
+		msleep(data->msec);
+		gpiod_set_value_cansleep(data->reset_gpio, 0);
+	}
+
+	return 0;
+}
+
+/**
  * arc_mdio_probe - MDIO probe function.
  * @priv:	Pointer to ARC EMAC private data structure.
  *
@@ -109,6 +129,8 @@ static int arc_mdio_write(struct mii_bus *bus, int phy_addr,
  */
 int arc_mdio_probe(struct arc_emac_priv *priv)
 {
+	struct arc_emac_mdio_bus_data *data = &priv->bus_data;
+	struct device_node *np = priv->dev->of_node;
 	struct mii_bus *bus;
 	int error;
 
@@ -122,6 +144,21 @@ int arc_mdio_probe(struct arc_emac_priv *priv)
 	bus->name = "Synopsys MII Bus",
 	bus->read = &arc_mdio_read;
 	bus->write = &arc_mdio_write;
+	bus->reset = &arc_mdio_reset;
+
+	/* optional reset-related properties */
+	data->reset_gpio = devm_gpiod_get_optional(priv->dev, "phy-reset",
+						   GPIOD_OUT_LOW);
+	if (IS_ERR(data->reset_gpio)) {
+		error = PTR_ERR(data->reset_gpio);
+		dev_err(priv->dev, "Failed to request gpio: %d\n", error);
+		return error;
+	}
+
+	of_property_read_u32(np, "phy-reset-duration", &data->msec);
+	/* A sane reset duration should not be longer than 1s */
+	if (data->msec > 1000)
+		data->msec = 1;
 
 	snprintf(bus->id, MII_BUS_ID_SIZE, "%s", bus->name);
 
-- 
cgit v0.10.2


From 663713eb477b9b916426124d6f5d5c5e9db919cf Mon Sep 17 00:00:00 2001
From: Caesar Wang <wxt@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:55 +0800
Subject: net: arc: trivial: cleanup the emac driver

This patch will make the driver more readability

The emac has the error and warnings if you run
'scripts/checkpatch.pl -f --subjective xxx' to check.

Let's clean up such trivial details.

Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: Jiri Kosina <trivial@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexander Kochetkov <al.kochet@gmail.com>
Cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/arc/emac.h b/drivers/net/ethernet/arc/emac.h
index 1a40403..ca562bc 100644
--- a/drivers/net/ethernet/arc/emac.h
+++ b/drivers/net/ethernet/arc/emac.h
@@ -14,36 +14,36 @@
 #include <linux/clk.h>
 
 /* STATUS and ENABLE Register bit masks */
-#define TXINT_MASK	(1<<0)	/* Transmit interrupt */
-#define RXINT_MASK	(1<<1)	/* Receive interrupt */
-#define ERR_MASK	(1<<2)	/* Error interrupt */
-#define TXCH_MASK	(1<<3)	/* Transmit chaining error interrupt */
-#define MSER_MASK	(1<<4)	/* Missed packet counter error */
-#define RXCR_MASK	(1<<8)	/* RXCRCERR counter rolled over  */
-#define RXFR_MASK	(1<<9)	/* RXFRAMEERR counter rolled over */
-#define RXFL_MASK	(1<<10)	/* RXOFLOWERR counter rolled over */
-#define MDIO_MASK	(1<<12)	/* MDIO complete interrupt */
-#define TXPL_MASK	(1<<31)	/* Force polling of BD by EMAC */
+#define TXINT_MASK	(1 << 0)	/* Transmit interrupt */
+#define RXINT_MASK	(1 << 1)	/* Receive interrupt */
+#define ERR_MASK	(1 << 2)	/* Error interrupt */
+#define TXCH_MASK	(1 << 3)	/* Transmit chaining error interrupt */
+#define MSER_MASK	(1 << 4)	/* Missed packet counter error */
+#define RXCR_MASK	(1 << 8)	/* RXCRCERR counter rolled over  */
+#define RXFR_MASK	(1 << 9)	/* RXFRAMEERR counter rolled over */
+#define RXFL_MASK	(1 << 10)	/* RXOFLOWERR counter rolled over */
+#define MDIO_MASK	(1 << 12)	/* MDIO complete interrupt */
+#define TXPL_MASK	(1 << 31)	/* Force polling of BD by EMAC */
 
 /* CONTROL Register bit masks */
-#define EN_MASK		(1<<0)	/* VMAC enable */
-#define TXRN_MASK	(1<<3)	/* TX enable */
-#define RXRN_MASK	(1<<4)	/* RX enable */
-#define DSBC_MASK	(1<<8)	/* Disable receive broadcast */
-#define ENFL_MASK	(1<<10)	/* Enable Full-duplex */
-#define PROM_MASK	(1<<11)	/* Promiscuous mode */
+#define EN_MASK		(1 << 0)	/* VMAC enable */
+#define TXRN_MASK	(1 << 3)	/* TX enable */
+#define RXRN_MASK	(1 << 4)	/* RX enable */
+#define DSBC_MASK	(1 << 8)	/* Disable receive broadcast */
+#define ENFL_MASK	(1 << 10)	/* Enable Full-duplex */
+#define PROM_MASK	(1 << 11)	/* Promiscuous mode */
 
 /* Buffer descriptor INFO bit masks */
-#define OWN_MASK	(1<<31)	/* 0-CPU owns buffer, 1-EMAC owns buffer */
-#define FIRST_MASK	(1<<16)	/* First buffer in chain */
-#define LAST_MASK	(1<<17)	/* Last buffer in chain */
+#define OWN_MASK	(1 << 31)	/* 0-CPU or 1-EMAC owns buffer */
+#define FIRST_MASK	(1 << 16)	/* First buffer in chain */
+#define LAST_MASK	(1 << 17)	/* Last buffer in chain */
 #define LEN_MASK	0x000007FF	/* last 11 bits */
-#define CRLS		(1<<21)
-#define DEFR		(1<<22)
-#define DROP		(1<<23)
-#define RTRY		(1<<24)
-#define LTCL		(1<<28)
-#define UFLO		(1<<29)
+#define CRLS		(1 << 21)
+#define DEFR		(1 << 22)
+#define DROP		(1 << 23)
+#define RTRY		(1 << 24)
+#define LTCL		(1 << 28)
+#define UFLO		(1 << 29)
 
 #define FOR_EMAC	OWN_MASK
 #define FOR_CPU		0
@@ -66,7 +66,7 @@ enum {
 	R_MDIO,
 };
 
-#define TX_TIMEOUT		(400*HZ/1000)	/* Transmission timeout */
+#define TX_TIMEOUT		(400 * HZ / 1000) /* Transmission timeout */
 
 #define ARC_EMAC_NAPI_WEIGHT	40		/* Workload for NAPI */
 
@@ -196,6 +196,7 @@ static inline unsigned int arc_reg_get(struct arc_emac_priv *priv, int reg)
 static inline void arc_reg_or(struct arc_emac_priv *priv, int reg, int mask)
 {
 	unsigned int value = arc_reg_get(priv, reg);
+
 	arc_reg_set(priv, reg, value | mask);
 }
 
@@ -211,6 +212,7 @@ static inline void arc_reg_or(struct arc_emac_priv *priv, int reg, int mask)
 static inline void arc_reg_clr(struct arc_emac_priv *priv, int reg, int mask)
 {
 	unsigned int value = arc_reg_get(priv, reg);
+
 	arc_reg_set(priv, reg, value & ~mask);
 }
 
diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c
index 6446af1..a3a9392 100644
--- a/drivers/net/ethernet/arc/emac_main.c
+++ b/drivers/net/ethernet/arc/emac_main.c
@@ -26,7 +26,6 @@
 
 #include "emac.h"
 
-
 /**
  * arc_emac_tx_avail - Return the number of available slots in the tx ring.
  * @priv: Pointer to ARC EMAC private data structure.
@@ -66,7 +65,7 @@ static void arc_emac_adjust_link(struct net_device *ndev)
 	if (priv->duplex != phy_dev->duplex) {
 		reg = arc_reg_get(priv, R_CTRL);
 
-		if (DUPLEX_FULL == phy_dev->duplex)
+		if (phy_dev->duplex == DUPLEX_FULL)
 			reg |= ENFL_MASK;
 		else
 			reg &= ~ENFL_MASK;
@@ -466,9 +465,9 @@ static int arc_emac_open(struct net_device *ndev)
 
 	/* Set CONTROL */
 	arc_reg_set(priv, R_CTRL,
-		     (RX_BD_NUM << 24) |	/* RX BD table length */
-		     (TX_BD_NUM << 16) |	/* TX BD table length */
-		     TXRN_MASK | RXRN_MASK);
+		    (RX_BD_NUM << 24) |	/* RX BD table length */
+		    (TX_BD_NUM << 16) |	/* TX BD table length */
+		    TXRN_MASK | RXRN_MASK);
 
 	napi_enable(&priv->napi);
 
@@ -533,8 +532,10 @@ static void arc_free_tx_queue(struct net_device *ndev)
 		struct buffer_state *tx_buff = &priv->tx_buff[i];
 
 		if (tx_buff->skb) {
-			dma_unmap_single(&ndev->dev, dma_unmap_addr(tx_buff, addr),
-					 dma_unmap_len(tx_buff, len), DMA_TO_DEVICE);
+			dma_unmap_single(&ndev->dev,
+					 dma_unmap_addr(tx_buff, addr),
+					 dma_unmap_len(tx_buff, len),
+					 DMA_TO_DEVICE);
 
 			/* return the sk_buff to system */
 			dev_kfree_skb_irq(tx_buff->skb);
@@ -562,8 +563,10 @@ static void arc_free_rx_queue(struct net_device *ndev)
 		struct buffer_state *rx_buff = &priv->rx_buff[i];
 
 		if (rx_buff->skb) {
-			dma_unmap_single(&ndev->dev, dma_unmap_addr(rx_buff, addr),
-					dma_unmap_len(rx_buff, len), DMA_FROM_DEVICE);
+			dma_unmap_single(&ndev->dev,
+					 dma_unmap_addr(rx_buff, addr),
+					 dma_unmap_len(rx_buff, len),
+					 DMA_FROM_DEVICE);
 
 			/* return the sk_buff to system */
 			dev_kfree_skb_irq(rx_buff->skb);
@@ -717,8 +720,8 @@ static void arc_emac_set_address_internal(struct net_device *ndev)
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 	unsigned int addr_low, addr_hi;
 
-	addr_low = le32_to_cpu(*(__le32 *) &ndev->dev_addr[0]);
-	addr_hi = le16_to_cpu(*(__le16 *) &ndev->dev_addr[4]);
+	addr_low = le32_to_cpu(*(__le32 *)&ndev->dev_addr[0]);
+	addr_hi = le16_to_cpu(*(__le16 *)&ndev->dev_addr[4]);
 
 	arc_reg_set(priv, R_ADDRL, addr_low);
 	arc_reg_set(priv, R_ADDRH, addr_hi);
@@ -774,7 +777,6 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 	unsigned int id, clock_frequency, irq;
 	int err;
 
-
 	/* Get PHY from device tree */
 	phy_node = of_parse_phandle(dev->of_node, "phy", 0);
 	if (!phy_node) {
@@ -796,7 +798,6 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 		return -ENODEV;
 	}
 
-
 	ndev->netdev_ops = &arc_emac_netdev_ops;
 	ndev->ethtool_ops = &arc_emac_ethtool_ops;
 	ndev->watchdog_timeo = TX_TIMEOUT;
@@ -807,9 +808,9 @@ int arc_emac_probe(struct net_device *ndev, int interface)
 	priv->dev = dev;
 
 	priv->regs = devm_ioremap_resource(dev, &res_regs);
-	if (IS_ERR(priv->regs)) {
+	if (IS_ERR(priv->regs))
 		return PTR_ERR(priv->regs);
-	}
+
 	dev_dbg(dev, "Registers base address is 0x%p\n", priv->regs);
 
 	if (priv->clk) {
@@ -930,10 +931,8 @@ int arc_emac_remove(struct net_device *ndev)
 	unregister_netdev(ndev);
 	netif_napi_del(&priv->napi);
 
-	if (!IS_ERR(priv->clk)) {
+	if (!IS_ERR(priv->clk))
 		clk_disable_unprepare(priv->clk);
-	}
-
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/arc/emac_mdio.c b/drivers/net/ethernet/arc/emac_mdio.c
index caf7042..16419f5 100644
--- a/drivers/net/ethernet/arc/emac_mdio.c
+++ b/drivers/net/ethernet/arc/emac_mdio.c
@@ -94,7 +94,7 @@ static int arc_mdio_write(struct mii_bus *bus, int phy_addr,
 		phy_addr, reg_num, value);
 
 	arc_reg_set(priv, R_MDIO,
-		     0x50020000 | (phy_addr << 23) | (reg_num << 18) | value);
+		    0x50020000 | (phy_addr << 23) | (reg_num << 18) | value);
 
 	return arc_mdio_complete_wait(priv);
 }
diff --git a/drivers/net/ethernet/arc/emac_rockchip.c b/drivers/net/ethernet/arc/emac_rockchip.c
index 85e821c..e278e3d 100644
--- a/drivers/net/ethernet/arc/emac_rockchip.c
+++ b/drivers/net/ethernet/arc/emac_rockchip.c
@@ -50,7 +50,7 @@ static void emac_rockchip_set_mac_speed(void *priv, unsigned int speed)
 	u32 data;
 	int err = 0;
 
-	switch(speed) {
+	switch (speed) {
 	case 10:
 		data = (1 << (speed_offset + 16)) | (0 << speed_offset);
 		break;
@@ -83,9 +83,18 @@ static const struct emac_rockchip_soc_data emac_rk3188_emac_data = {
 };
 
 static const struct of_device_id emac_rockchip_dt_ids[] = {
-	{ .compatible = "rockchip,rk3036-emac", .data = &emac_rk3036_emac_data },
-	{ .compatible = "rockchip,rk3066-emac", .data = &emac_rk3066_emac_data },
-	{ .compatible = "rockchip,rk3188-emac", .data = &emac_rk3188_emac_data },
+	{
+		.compatible = "rockchip,rk3036-emac",
+		.data = &emac_rk3036_emac_data,
+	},
+	{
+		.compatible = "rockchip,rk3066-emac",
+		.data = &emac_rk3066_emac_data,
+	},
+	{
+		.compatible = "rockchip,rk3188-emac",
+		.data = &emac_rk3188_emac_data,
+	},
 	{ /* Sentinel */ }
 };
 
@@ -123,9 +132,11 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 		goto out_netdev;
 	}
 
-	priv->grf = syscon_regmap_lookup_by_phandle(dev->of_node, "rockchip,grf");
+	priv->grf = syscon_regmap_lookup_by_phandle(dev->of_node,
+						    "rockchip,grf");
 	if (IS_ERR(priv->grf)) {
-		dev_err(dev, "failed to retrieve global register file (%ld)\n", PTR_ERR(priv->grf));
+		dev_err(dev, "failed to retrieve global register file (%ld)\n",
+			PTR_ERR(priv->grf));
 		err = PTR_ERR(priv->grf);
 		goto out_netdev;
 	}
@@ -135,14 +146,16 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 
 	priv->emac.clk = devm_clk_get(dev, "hclk");
 	if (IS_ERR(priv->emac.clk)) {
-		dev_err(dev, "failed to retrieve host clock (%ld)\n", PTR_ERR(priv->emac.clk));
+		dev_err(dev, "failed to retrieve host clock (%ld)\n",
+			PTR_ERR(priv->emac.clk));
 		err = PTR_ERR(priv->emac.clk);
 		goto out_netdev;
 	}
 
 	priv->refclk = devm_clk_get(dev, "macref");
 	if (IS_ERR(priv->refclk)) {
-		dev_err(dev, "failed to retrieve reference clock (%ld)\n", PTR_ERR(priv->refclk));
+		dev_err(dev, "failed to retrieve reference clock (%ld)\n",
+			PTR_ERR(priv->refclk));
 		err = PTR_ERR(priv->refclk);
 		goto out_netdev;
 	}
@@ -179,19 +192,22 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 
 	err = regmap_write(priv->grf, priv->soc_data->grf_offset, data);
 	if (err) {
-		dev_err(dev, "unable to apply initial settings to grf (%d)\n", err);
+		dev_err(dev, "unable to apply initial settings to grf (%d)\n",
+			err);
 		goto out_regulator_disable;
 	}
 
 	/* RMII interface needs always a rate of 50MHz */
 	err = clk_set_rate(priv->refclk, 50000000);
 	if (err)
-		dev_err(dev, "failed to change reference clock rate (%d)\n", err);
+		dev_err(dev,
+			"failed to change reference clock rate (%d)\n", err);
 
 	if (priv->soc_data->need_div_macclk) {
 		priv->macclk = devm_clk_get(dev, "macclk");
 		if (IS_ERR(priv->macclk)) {
-			dev_err(dev, "failed to retrieve mac clock (%ld)\n", PTR_ERR(priv->macclk));
+			dev_err(dev, "failed to retrieve mac clock (%ld)\n",
+				PTR_ERR(priv->macclk));
 			err = PTR_ERR(priv->macclk);
 			goto out_regulator_disable;
 		}
@@ -205,7 +221,8 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 		/* RMII TX/RX needs always a rate of 25MHz */
 		err = clk_set_rate(priv->macclk, 25000000);
 		if (err)
-			dev_err(dev, "failed to change mac clock rate (%d)\n", err);
+			dev_err(dev,
+				"failed to change mac clock rate (%d)\n", err);
 	}
 
 	err = arc_emac_probe(ndev, interface);
-- 
cgit v0.10.2


From fb781c8e2a370d67acf7b8a8826e6f5e3ae1d7c6 Mon Sep 17 00:00:00 2001
From: Xing Zheng <zhengxing@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:56 +0800
Subject: clk: rockchip: add node-id for rk3036 emac hclk

Add the node-id for the emac hclk to the binding header.

Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: Xing Zheng <zhengxing@rock-chips.com>
Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org
Cc: linux-rockchip@lists.infradead.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/dt-bindings/clock/rk3036-cru.h b/include/dt-bindings/clock/rk3036-cru.h
index ebc7a7b..3396591 100644
--- a/include/dt-bindings/clock/rk3036-cru.h
+++ b/include/dt-bindings/clock/rk3036-cru.h
@@ -92,6 +92,7 @@
 #define HCLK_SDMMC		456
 #define HCLK_SDIO		457
 #define HCLK_EMMC		459
+#define HCLK_MAC		460
 #define HCLK_I2S		462
 #define HCLK_LCDC		465
 #define HCLK_ROM		467
-- 
cgit v0.10.2


From e764b93924b47cd53b818c1cf8708a35bdfbb83d Mon Sep 17 00:00:00 2001
From: Xing Zheng <zhengxing@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:57 +0800
Subject: clk: rockchip: associate the rk3036 HCLK_EMAC clock-id

Associate the new clock id the clock.

Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: Xing Zheng <zhengxing@rock-chips.com>
Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org
Cc: linux-rockchip@lists.infradead.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/clk/rockchip/clk-rk3036.c b/drivers/clk/rockchip/clk-rk3036.c
index bc7fbac..37f4292 100644
--- a/drivers/clk/rockchip/clk-rk3036.c
+++ b/drivers/clk/rockchip/clk-rk3036.c
@@ -404,7 +404,7 @@ static struct rockchip_clk_branch rk3036_clk_branches[] __initdata = {
 	GATE(HCLK_OTG1, "hclk_otg1", "hclk_peri", CLK_IGNORE_UNUSED, RK2928_CLKGATE_CON(7), 3, GFLAGS),
 	GATE(HCLK_I2S, "hclk_i2s", "hclk_peri", 0, RK2928_CLKGATE_CON(7), 2, GFLAGS),
 	GATE(0, "hclk_sfc", "hclk_peri", CLK_IGNORE_UNUSED, RK2928_CLKGATE_CON(3), 14, GFLAGS),
-	GATE(0, "hclk_mac", "hclk_peri", CLK_IGNORE_UNUSED, RK2928_CLKGATE_CON(3), 15, GFLAGS),
+	GATE(HCLK_MAC, "hclk_mac", "hclk_peri", 0, RK2928_CLKGATE_CON(3), 5, GFLAGS),
 
 	/* pclk_peri gates */
 	GATE(0, "pclk_peri_matrix", "pclk_peri", CLK_IGNORE_UNUSED, RK2928_CLKGATE_CON(4), 1, GFLAGS),
-- 
cgit v0.10.2


From f7e180222b973a0b363564b281a314276cb2b594 Mon Sep 17 00:00:00 2001
From: Xing Zheng <zhengxing@rock-chips.com>
Date: Mon, 14 Mar 2016 16:01:58 +0800
Subject: clk: rockchip: add clock-id for rk3036 emac pll source clock

Suitable PLLs for the emac on the rk3036 are difficult to find
and one of them is the (continuously changing) APLL. So in most
cases it will be necessary to select a PLL manually.
So add a clock-id for it.

Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: Xing Zheng <zhengxing@rock-chips.com>
Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org
Cc: linux-rockchip@lists.infradead.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/dt-bindings/clock/rk3036-cru.h b/include/dt-bindings/clock/rk3036-cru.h
index 3396591..de44109 100644
--- a/include/dt-bindings/clock/rk3036-cru.h
+++ b/include/dt-bindings/clock/rk3036-cru.h
@@ -54,6 +54,7 @@
 #define SCLK_PVTM_VIDEO		125
 #define SCLK_MAC		151
 #define SCLK_MACREF		152
+#define SCLK_MACPLL		153
 #define SCLK_SFC		160
 
 /* aclk gates */
-- 
cgit v0.10.2


From 2c6fae2501d87ca94b5249df38797f02d4e39add Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 14 Mar 2016 16:01:59 +0800
Subject: clk: rockchip: associate SCLK_MAC_PLL and disable reparenting on
 rk3036

The emac needs constant and very specific rate but the possible PLL-sources
are very limited, so we expect the PLL source to be set manually on per
board and don't want it to get changed in an automatic way later.
So add the necessary clock-id and disable reparenting on set_rate calls.

Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Cc: Michael Turquette <mturquette@baylibre.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org

Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/clk/rockchip/clk-rk3036.c b/drivers/clk/rockchip/clk-rk3036.c
index 37f4292..53e9c39 100644
--- a/drivers/clk/rockchip/clk-rk3036.c
+++ b/drivers/clk/rockchip/clk-rk3036.c
@@ -343,7 +343,7 @@ static struct rockchip_clk_branch rk3036_clk_branches[] __initdata = {
 			RK2928_CLKSEL_CON(16), 0, 2, MFLAGS, 2, 5, DFLAGS,
 			RK2928_CLKGATE_CON(10), 5, GFLAGS),
 
-	COMPOSITE_NOGATE(0, "mac_pll_src", mux_pll_src_3plls_p, 0,
+	COMPOSITE_NOGATE(SCLK_MACPLL, "mac_pll_src", mux_pll_src_3plls_p, CLK_SET_RATE_NO_REPARENT,
 			RK2928_CLKSEL_CON(21), 0, 2, MFLAGS, 9, 5, DFLAGS),
 	MUX(SCLK_MACREF, "mac_clk_ref", mux_mac_p, CLK_SET_RATE_PARENT,
 			RK2928_CLKSEL_CON(21), 3, 1, MFLAGS),
-- 
cgit v0.10.2


From af671e7bd96bc9bde623b0e6f75bfa4269c2c57f Mon Sep 17 00:00:00 2001
From: Xing Zheng <zhengxing@rock-chips.com>
Date: Mon, 14 Mar 2016 16:02:00 +0800
Subject: ARM: dts: rockchip: add to support emac for rk3036 SoCs

This patch adds the emac device node for rk3036 SoCs.
We need to let mac clock under the DPLL which is able to provide
the accurate 50MHz what mac_ref need, since that will cause some
unstable things if the cpufreq is working.

Signed-off-by: Xing Zheng <zhengxing@rock-chips.com>
Signed-off-by: Caesar Wang <wxt@rock-chips.com>
Cc: linux-rockchip@lists.infradead.org
Cc: Xing Zheng <zhengxing@rock-chips.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/arm/boot/dts/rk3036-evb.dts b/arch/arm/boot/dts/rk3036-evb.dts
index 28a0336..b3d6ec8 100644
--- a/arch/arm/boot/dts/rk3036-evb.dts
+++ b/arch/arm/boot/dts/rk3036-evb.dts
@@ -47,6 +47,20 @@
 	compatible = "rockchip,rk3036-evb", "rockchip,rk3036";
 };
 
+&emac {
+	pinctrl-names = "default";
+	pinctrl-0 = <&emac_xfer>, <&emac_mdio>;
+	phy = <&phy0>;
+	phy-reset-gpios = <&gpio2 22 GPIO_ACTIVE_LOW>; /* PHY_RST */
+	phy-reset-duration = <10>; /* millisecond */
+
+	status = "okay";
+
+	phy0: ethernet-phy@0 {
+		reg = <0>;
+	};
+};
+
 &i2c1 {
 	status = "okay";
 
diff --git a/arch/arm/boot/dts/rk3036-kylin.dts b/arch/arm/boot/dts/rk3036-kylin.dts
index 992f9ca..6b4ffc3 100644
--- a/arch/arm/boot/dts/rk3036-kylin.dts
+++ b/arch/arm/boot/dts/rk3036-kylin.dts
@@ -60,6 +60,20 @@
 	status = "okay";
 };
 
+&emac {
+	pinctrl-names = "default";
+	pinctrl-0 = <&emac_xfer>, <&emac_mdio>;
+	phy = <&phy0>;
+	phy-reset-gpios = <&gpio2 22 GPIO_ACTIVE_LOW>; /* PHY_RST */
+	phy-reset-duration = <10>; /* millisecond */
+
+	status = "okay";
+
+	phy0: ethernet-phy@0 {
+		reg = <0>;
+	};
+};
+
 &emmc {
 	status = "okay";
 };
diff --git a/arch/arm/boot/dts/rk3036.dtsi b/arch/arm/boot/dts/rk3036.dtsi
index b9567c1..609319c 100644
--- a/arch/arm/boot/dts/rk3036.dtsi
+++ b/arch/arm/boot/dts/rk3036.dtsi
@@ -186,6 +186,27 @@
 		status = "disabled";
 	};
 
+	emac: ethernet@10200000 {
+		compatible = "rockchip,rk3036-emac", "snps,arc-emac";
+		reg = <0x10200000 0x4000>;
+		interrupts = <GIC_SPI 8 IRQ_TYPE_LEVEL_HIGH>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+		rockchip,grf = <&grf>;
+		clocks = <&cru HCLK_MAC>, <&cru SCLK_MACREF>, <&cru SCLK_MAC>;
+		clock-names = "hclk", "macref", "macclk";
+		/*
+		 * Fix the emac parent clock is DPLL instead of APLL.
+		 * since that will cause some unstable things if the cpufreq
+		 * is working. (e.g: the accurate 50MHz what mac_ref need)
+		 */
+		assigned-clocks = <&cru SCLK_MACPLL>;
+		assigned-clock-parents = <&cru PLL_DPLL>;
+		max-speed = <100>;
+		phy-mode = "rmii";
+		status = "disabled";
+	};
+
 	sdmmc: dwmmc@10214000 {
 		compatible = "rockchip,rk3036-dw-mshc", "rockchip,rk3288-dw-mshc";
 		reg = <0x10214000 0x4000>;
@@ -556,6 +577,24 @@
 			};
 		};
 
+		emac {
+			emac_xfer: emac-xfer {
+				rockchip,pins = <2 10 RK_FUNC_1 &pcfg_pull_default>, /* crs_dvalid */
+						<2 13 RK_FUNC_1 &pcfg_pull_default>, /* tx_en */
+						<2 14 RK_FUNC_1 &pcfg_pull_default>, /* mac_clk */
+						<2 15 RK_FUNC_1 &pcfg_pull_default>, /* rx_err */
+						<2 16 RK_FUNC_1 &pcfg_pull_default>, /* rxd1 */
+						<2 17 RK_FUNC_1 &pcfg_pull_default>, /* rxd0 */
+						<2 18 RK_FUNC_1 &pcfg_pull_default>, /* txd1 */
+						<2 19 RK_FUNC_1 &pcfg_pull_default>; /* txd0 */
+			};
+
+			emac_mdio: emac-mdio {
+				rockchip,pins = <2 12 RK_FUNC_1 &pcfg_pull_default>, /* mac_md */
+						<2 25 RK_FUNC_1 &pcfg_pull_default>; /* mac_mdclk */
+			};
+		};
+
 		i2c0 {
 			i2c0_xfer: i2c0-xfer {
 				rockchip,pins = <0 0 RK_FUNC_1 &pcfg_pull_none>,
-- 
cgit v0.10.2


From e8e56ffd9d2973398b60ece1f1bebb8d67b4d032 Mon Sep 17 00:00:00 2001
From: Guillaume Nault <g.nault@alphalink.fr>
Date: Mon, 14 Mar 2016 21:17:16 +0100
Subject: ppp: ensure file->private_data can't be overridden

Locking ppp_mutex must be done before dereferencing file->private_data,
otherwise it could be modified before ppp_unattached_ioctl() takes the
lock. This could lead ppp_unattached_ioctl() to override ->private_data,
thus leaking reference to the ppp_file previously pointed to.

v2: lock all ppp_ioctl() instead of just checking private_data in
    ppp_unattached_ioctl(), to avoid ambiguous behaviour.

Fixes: f3ff8a4d80e8 ("ppp: push BKL down into the driver")
Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index 931836e..4fd8610 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -575,7 +575,7 @@ static int get_filter(void __user *arg, struct sock_filter **p)
 
 static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
-	struct ppp_file *pf = file->private_data;
+	struct ppp_file *pf;
 	struct ppp *ppp;
 	int err = -EFAULT, val, val2, i;
 	struct ppp_idle idle;
@@ -585,9 +585,14 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	void __user *argp = (void __user *)arg;
 	int __user *p = argp;
 
-	if (!pf)
-		return ppp_unattached_ioctl(current->nsproxy->net_ns,
-					pf, file, cmd, arg);
+	mutex_lock(&ppp_mutex);
+
+	pf = file->private_data;
+	if (!pf) {
+		err = ppp_unattached_ioctl(current->nsproxy->net_ns,
+					   pf, file, cmd, arg);
+		goto out;
+	}
 
 	if (cmd == PPPIOCDETACH) {
 		/*
@@ -602,7 +607,6 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		 * this fd and reopening /dev/ppp.
 		 */
 		err = -EINVAL;
-		mutex_lock(&ppp_mutex);
 		if (pf->kind == INTERFACE) {
 			ppp = PF_TO_PPP(pf);
 			rtnl_lock();
@@ -616,15 +620,13 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		} else
 			pr_warn("PPPIOCDETACH file->f_count=%ld\n",
 				atomic_long_read(&file->f_count));
-		mutex_unlock(&ppp_mutex);
-		return err;
+		goto out;
 	}
 
 	if (pf->kind == CHANNEL) {
 		struct channel *pch;
 		struct ppp_channel *chan;
 
-		mutex_lock(&ppp_mutex);
 		pch = PF_TO_CHANNEL(pf);
 
 		switch (cmd) {
@@ -646,17 +648,16 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 				err = chan->ops->ioctl(chan, cmd, arg);
 			up_read(&pch->chan_sem);
 		}
-		mutex_unlock(&ppp_mutex);
-		return err;
+		goto out;
 	}
 
 	if (pf->kind != INTERFACE) {
 		/* can't happen */
 		pr_err("PPP: not interface or channel??\n");
-		return -EINVAL;
+		err = -EINVAL;
+		goto out;
 	}
 
-	mutex_lock(&ppp_mutex);
 	ppp = PF_TO_PPP(pf);
 	switch (cmd) {
 	case PPPIOCSMRU:
@@ -831,7 +832,10 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	default:
 		err = -ENOTTY;
 	}
+
+out:
 	mutex_unlock(&ppp_mutex);
+
 	return err;
 }
 
@@ -844,7 +848,6 @@ static int ppp_unattached_ioctl(struct net *net, struct ppp_file *pf,
 	struct ppp_net *pn;
 	int __user *p = (int __user *)arg;
 
-	mutex_lock(&ppp_mutex);
 	switch (cmd) {
 	case PPPIOCNEWUNIT:
 		/* Create a new ppp unit */
@@ -894,7 +897,7 @@ static int ppp_unattached_ioctl(struct net *net, struct ppp_file *pf,
 	default:
 		err = -ENOTTY;
 	}
-	mutex_unlock(&ppp_mutex);
+
 	return err;
 }
 
-- 
cgit v0.10.2


From 7091f01e8cf6989e63c4eacb59b654fcff057901 Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Mon, 14 Mar 2016 17:30:37 -0700
Subject: phy: mdio-cavium: Add missing MODULE_* annotations.

When the code was factored out of mdio-octeon.c, the
MODULE_DESCRIPTION, MODULE_AUTHOR and MODULE_LICENSE annotations were
inadvertently omitted.  Restore them so that we don't get kernel taint
warnings upon module loading.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/mdio-cavium.c b/drivers/net/phy/mdio-cavium.c
index e796ee1..6df2fa7 100644
--- a/drivers/net/phy/mdio-cavium.c
+++ b/drivers/net/phy/mdio-cavium.c
@@ -147,3 +147,7 @@ int cavium_mdiobus_write(struct mii_bus *bus, int phy_id, int regnum, u16 val)
 	return 0;
 }
 EXPORT_SYMBOL(cavium_mdiobus_write);
+
+MODULE_DESCRIPTION("Common code for OCTEON and Thunder MDIO bus drivers");
+MODULE_AUTHOR("David Daney");
+MODULE_LICENSE("GPL");
-- 
cgit v0.10.2


From 9277a4f875adbeeb6209c0a3e3cf04c752522b2e Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Mon, 14 Mar 2016 17:30:38 -0700
Subject: net: cavium: For Kconfig THUNDER_NIC_BGX, select MDIO_THUNDER.

Previously we selected MDIO_OCTEON, which after creating the Thunder
specific MDIO bus driver is much less useful.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/Kconfig b/drivers/net/ethernet/cavium/Kconfig
index 8fb84e6..953aa40 100644
--- a/drivers/net/ethernet/cavium/Kconfig
+++ b/drivers/net/ethernet/cavium/Kconfig
@@ -35,7 +35,7 @@ config	THUNDER_NIC_BGX
 	tristate "Thunder MAC interface driver (BGX)"
 	depends on 64BIT
 	select PHYLIB
-	select MDIO_OCTEON
+	select MDIO_THUNDER
 	---help---
 	  This driver supports programming and controlling of MAC
 	  interface from NIC physical function driver.
-- 
cgit v0.10.2


From b7d3e3d3d21a09f8e45043d548224000a8efe06a Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Mon, 14 Mar 2016 17:30:39 -0700
Subject: net: thunderx: Don't leak phy device references on -EPROBE_DEFER
 condition.

It is possible, although unlikely, that probing will find the
phy_device for the first LMAC of a thunder BGX device, but then need
to fail with -EPROBE_DEFER on a subsequent LMAC.  In this case, we
need to call put_device() on each of the phy_devices that were
obtained, but will be unused due to returning -EPROBE_DEFER.

Also, since we can break out of the probing loop early, we need to
explicitly call of_node_put() outside of the loop.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
index feed231..9679515 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
@@ -974,17 +974,18 @@ static int bgx_init_acpi_phy(struct bgx *bgx)
 static int bgx_init_of_phy(struct bgx *bgx)
 {
 	struct fwnode_handle *fwn;
+	struct device_node *node = NULL;
 	u8 lmac = 0;
-	const char *mac;
 
 	device_for_each_child_node(&bgx->pdev->dev, fwn) {
 		struct phy_device *pd;
 		struct device_node *phy_np;
-		struct device_node *node = to_of_node(fwn);
+		const char *mac;
 
 		/* Should always be an OF node.  But if it is not, we
 		 * cannot handle it, so exit the loop.
 		 */
+		node = to_of_node(fwn);
 		if (!node)
 			break;
 
@@ -1005,17 +1006,30 @@ static int bgx_init_of_phy(struct bgx *bgx)
 			/* Wait until the phy drivers are available */
 			pd = of_phy_find_device(phy_np);
 			if (!pd)
-				return -EPROBE_DEFER;
+				goto defer;
 			bgx->lmac[lmac].phydev = pd;
 		}
 
 		lmac++;
-		if (lmac == MAX_LMAC_PER_BGX) {
-			of_node_put(node);
+		if (lmac == MAX_LMAC_PER_BGX)
 			break;
-		}
 	}
+	of_node_put(node);
 	return 0;
+
+defer:
+	/* We are bailing out, try not to leak device reference counts
+	 * for phy devices we may have already found.
+	 */
+	while (lmac) {
+		if (bgx->lmac[lmac].phydev) {
+			put_device(&bgx->lmac[lmac].phydev->mdio.dev);
+			bgx->lmac[lmac].phydev = NULL;
+		}
+		lmac--;
+	}
+	of_node_put(node);
+	return -EPROBE_DEFER;
 }
 
 #else
-- 
cgit v0.10.2


From f6773c5e95a6dc1af82157d4e96e412dee3abf31 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 16 Mar 2016 21:59:49 -0700
Subject: vlan: propagate gso_max_segs

vlan drivers lack proper propagation of gso_max_segs from
lower device.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 5802b90..57941d3 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -119,6 +119,7 @@ static int ipvlan_init(struct net_device *dev)
 	dev->features = phy_dev->features & IPVLAN_FEATURES;
 	dev->features |= NETIF_F_LLTX;
 	dev->gso_max_size = phy_dev->gso_max_size;
+	dev->gso_max_segs = phy_dev->gso_max_segs;
 	dev->hard_header_len = phy_dev->hard_header_len;
 
 	ipvlan_set_lockdep_class(dev);
@@ -592,6 +593,7 @@ static int ipvlan_device_event(struct notifier_block *unused,
 		list_for_each_entry(ipvlan, &port->ipvlans, pnode) {
 			ipvlan->dev->features = dev->features & IPVLAN_FEATURES;
 			ipvlan->dev->gso_max_size = dev->gso_max_size;
+			ipvlan->dev->gso_max_segs = dev->gso_max_segs;
 			netdev_features_change(ipvlan->dev);
 		}
 		break;
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 6e953e3..2bcf1f3 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -803,6 +803,7 @@ static int macvlan_init(struct net_device *dev)
 	dev->hw_features	|= NETIF_F_LRO;
 	dev->vlan_features	= lowerdev->vlan_features & MACVLAN_FEATURES;
 	dev->gso_max_size	= lowerdev->gso_max_size;
+	dev->gso_max_segs	= lowerdev->gso_max_segs;
 	dev->hard_header_len	= lowerdev->hard_header_len;
 
 	macvlan_set_lockdep_class(dev);
@@ -1532,6 +1533,7 @@ static int macvlan_device_event(struct notifier_block *unused,
 	case NETDEV_FEAT_CHANGE:
 		list_for_each_entry(vlan, &port->vlans, list) {
 			vlan->dev->gso_max_size = dev->gso_max_size;
+			vlan->dev->gso_max_segs = dev->gso_max_segs;
 			netdev_update_features(vlan->dev);
 		}
 		break;
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 69929c0..a1e273a 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -311,6 +311,7 @@ static void vlan_transfer_features(struct net_device *dev,
 	struct vlan_dev_priv *vlan = vlan_dev_priv(vlandev);
 
 	vlandev->gso_max_size = dev->gso_max_size;
+	vlandev->gso_max_segs = dev->gso_max_segs;
 
 	if (vlan_hw_offload_capable(dev->features, vlan->vlan_proto))
 		vlandev->hard_header_len = dev->hard_header_len;
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index e416a40..e7e6257 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -551,6 +551,7 @@ static int vlan_dev_init(struct net_device *dev)
 	dev->features |= real_dev->vlan_features | NETIF_F_LLTX |
 			 NETIF_F_GSO_SOFTWARE;
 	dev->gso_max_size = real_dev->gso_max_size;
+	dev->gso_max_segs = real_dev->gso_max_segs;
 	if (dev->features & NETIF_F_VLAN_FEATURES)
 		netdev_warn(real_dev, "VLAN features are set incorrectly.  Q-in-Q configurations may not work correctly.\n");
 
-- 
cgit v0.10.2


From e316ea62e3203d524ff0239a40c56d3a39ad1b5c Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 16 Mar 2016 22:52:15 -0700
Subject: tcp/dccp: remove obsolete WARN_ON() in icmp handlers

Now SYN_RECV request sockets are installed in ehash table, an ICMP
handler can find a request socket while another cpu handles an incoming
packet transforming this SYN_RECV request socket into an ESTABLISHED
socket.

We need to remove the now obsolete WARN_ON(req->sk), since req->sk
is set when a new child is created and added into listener accept queue.

If this race happens, the ICMP will do nothing special.

Fixes: 079096f103fa ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Ben Lazarus <blazarus@google.com>
Reported-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index b5672e5..9c67a96 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -204,8 +204,6 @@ void dccp_req_err(struct sock *sk, u64 seq)
 	 * ICMPs are not backlogged, hence we cannot get an established
 	 * socket here.
 	 */
-	WARN_ON(req->sk);
-
 	if (!between48(seq, dccp_rsk(req)->dreq_iss, dccp_rsk(req)->dreq_gss)) {
 		NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 	} else {
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 0b02ef7..e7528b1 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -319,8 +319,6 @@ void tcp_req_err(struct sock *sk, u32 seq, bool abort)
 	/* ICMPs are not backlogged, hence we cannot get
 	 * an established socket here.
 	 */
-	WARN_ON(req->sk);
-
 	if (seq != tcp_rsk(req)->snt_isn) {
 		NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
 	} else if (abort) {
-- 
cgit v0.10.2


From bd9e33508c5e1eb5d807d11d7bfc52125fcdb04e Mon Sep 17 00:00:00 2001
From: Schemmel Hans-Christoph <Hans-Christoph.Schemmel@gemalto.com>
Date: Thu, 17 Mar 2016 10:07:56 +0000
Subject: qmi_wwan: Added support for Gemalto's Cinterion PHxx WWAN interface
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Added support for Gemalto's Cinterion PHxx WWAN interfaces
by adding QMI_FIXED_INTF with Cinterion's VID and PID.

PHxx can have:
2 RmNet Interfaces (PID 0x0082) or
1 RmNet + 1 USB Audio interface (PID 0x0083).

Signed-off-by: Hans-Christoph Schemmel <hans-christoph.schemmel@gemalto.com>
Acked-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c
index a3a4ccf..7d717c6 100644
--- a/drivers/net/usb/qmi_wwan.c
+++ b/drivers/net/usb/qmi_wwan.c
@@ -881,6 +881,9 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x0b3c, 0xc00b, 4)},	/* Olivetti Olicard 500 */
 	{QMI_FIXED_INTF(0x1e2d, 0x0060, 4)},	/* Cinterion PLxx */
 	{QMI_FIXED_INTF(0x1e2d, 0x0053, 4)},	/* Cinterion PHxx,PXxx */
+	{QMI_FIXED_INTF(0x1e2d, 0x0082, 4)},	/* Cinterion PHxx,PXxx (2 RmNet) */
+	{QMI_FIXED_INTF(0x1e2d, 0x0082, 5)},	/* Cinterion PHxx,PXxx (2 RmNet) */
+	{QMI_FIXED_INTF(0x1e2d, 0x0083, 4)},	/* Cinterion PHxx,PXxx (1 RmNet + USB Audio)*/
 	{QMI_FIXED_INTF(0x413c, 0x81a2, 8)},	/* Dell Wireless 5806 Gobi(TM) 4G LTE Mobile Broadband Card */
 	{QMI_FIXED_INTF(0x413c, 0x81a3, 8)},	/* Dell Wireless 5570 HSPA+ (42Mbps) Mobile Broadband Card */
 	{QMI_FIXED_INTF(0x413c, 0x81a4, 8)},	/* Dell Wireless 5570e HSPA+ (42Mbps) Mobile Broadband Card */
-- 
cgit v0.10.2


From b73f96fcb49ec90c2f837719893e7b25fcdf08d8 Mon Sep 17 00:00:00 2001
From: Wu Fengguang <fengguang.wu@intel.com>
Date: Fri, 18 Mar 2016 23:27:28 +0800
Subject: net: dst_cache_per_cpu_dst_set() can be static

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/dst_cache.c b/net/core/dst_cache.c
index 3938f3f..554d364 100644
--- a/net/core/dst_cache.c
+++ b/net/core/dst_cache.c
@@ -28,8 +28,8 @@ struct dst_cache_pcpu {
 	};
 };
 
-void dst_cache_per_cpu_dst_set(struct dst_cache_pcpu *dst_cache,
-			       struct dst_entry *dst, u32 cookie)
+static void dst_cache_per_cpu_dst_set(struct dst_cache_pcpu *dst_cache,
+				      struct dst_entry *dst, u32 cookie)
 {
 	dst_release(dst_cache->dst);
 	if (dst)
@@ -39,8 +39,8 @@ void dst_cache_per_cpu_dst_set(struct dst_cache_pcpu *dst_cache,
 	dst_cache->dst = dst;
 }
 
-struct dst_entry *dst_cache_per_cpu_get(struct dst_cache *dst_cache,
-					struct dst_cache_pcpu *idst)
+static struct dst_entry *dst_cache_per_cpu_get(struct dst_cache *dst_cache,
+					       struct dst_cache_pcpu *idst)
 {
 	struct dst_entry *dst;
 
-- 
cgit v0.10.2


From e014e8468552236f0f0cb64c7c341c1dce506070 Mon Sep 17 00:00:00 2001
From: Wu Fengguang <fengguang.wu@intel.com>
Date: Sat, 19 Mar 2016 00:54:50 +0800
Subject: ovs: internal_set_rx_headroom() can be static

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index 83a5534..7c8b90b 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -138,7 +138,7 @@ internal_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	return stats;
 }
 
-void internal_set_rx_headroom(struct net_device *dev, int new_hr)
+static void internal_set_rx_headroom(struct net_device *dev, int new_hr)
 {
 	dev->needed_headroom = new_hr;
 }
-- 
cgit v0.10.2


From d07a147f00de303c799af3427d4e2d612902a7d3 Mon Sep 17 00:00:00 2001
From: David Daney <david.daney@cavium.com>
Date: Mon, 14 Mar 2016 17:57:08 -0700
Subject: netdev: Move octeon/octeon_mgmt driver to cavium directory.

No code changes.  Since OCTEON is a Cavium product, move the driver to
the vendor directory to unclutter things a bit.

Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index be67a19..2ffd634 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -139,7 +139,6 @@ config NET_NETX
 source "drivers/net/ethernet/nuvoton/Kconfig"
 source "drivers/net/ethernet/nvidia/Kconfig"
 source "drivers/net/ethernet/nxp/Kconfig"
-source "drivers/net/ethernet/octeon/Kconfig"
 source "drivers/net/ethernet/oki-semi/Kconfig"
 
 config ETHOC
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 6ffcc80..1d349e9 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -59,7 +59,6 @@ obj-$(CONFIG_NET_NETX) += netx-eth.o
 obj-$(CONFIG_NET_VENDOR_NUVOTON) += nuvoton/
 obj-$(CONFIG_NET_VENDOR_NVIDIA) += nvidia/
 obj-$(CONFIG_LPC_ENET) += nxp/
-obj-$(CONFIG_OCTEON_MGMT_ETHERNET) += octeon/
 obj-$(CONFIG_NET_VENDOR_OKI) += oki-semi/
 obj-$(CONFIG_ETHOC) += ethoc.o
 obj-$(CONFIG_NET_PACKET_ENGINE) += packetengines/
diff --git a/drivers/net/ethernet/cavium/Kconfig b/drivers/net/ethernet/cavium/Kconfig
index 953aa40..0ef232d 100644
--- a/drivers/net/ethernet/cavium/Kconfig
+++ b/drivers/net/ethernet/cavium/Kconfig
@@ -53,4 +53,15 @@ config LIQUIDIO
 	  To compile this driver as a module, choose M here: the module
 	  will be called liquidio.  This is recommended.
 
+config OCTEON_MGMT_ETHERNET
+	tristate "Octeon Management port ethernet driver (CN5XXX, CN6XXX)"
+	depends on CAVIUM_OCTEON_SOC
+	select PHYLIB
+	select MDIO_OCTEON
+	default y
+	help
+	  Enable the ethernet driver for the management
+	  port on Cavium Networks' Octeon CN57XX, CN56XX, CN55XX,
+	  CN54XX, CN52XX, and CN6XXX chips.
+
 endif # NET_VENDOR_CAVIUM
diff --git a/drivers/net/ethernet/cavium/Makefile b/drivers/net/ethernet/cavium/Makefile
index d22f886..872da9f 100644
--- a/drivers/net/ethernet/cavium/Makefile
+++ b/drivers/net/ethernet/cavium/Makefile
@@ -3,3 +3,4 @@
 #
 obj-$(CONFIG_NET_VENDOR_CAVIUM) += thunder/
 obj-$(CONFIG_NET_VENDOR_CAVIUM) += liquidio/
+obj-$(CONFIG_NET_VENDOR_CAVIUM) += octeon/
diff --git a/drivers/net/ethernet/cavium/octeon/Makefile b/drivers/net/ethernet/cavium/octeon/Makefile
new file mode 100644
index 0000000..efa41c1
--- /dev/null
+++ b/drivers/net/ethernet/cavium/octeon/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the Cavium network device drivers.
+#
+
+obj-$(CONFIG_OCTEON_MGMT_ETHERNET)	+= octeon_mgmt.o
diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
new file mode 100644
index 0000000..c177c7c
--- /dev/null
+++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c
@@ -0,0 +1,1593 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2009-2012 Cavium, Inc
+ */
+
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/etherdevice.h>
+#include <linux/capability.h>
+#include <linux/net_tstamp.h>
+#include <linux/interrupt.h>
+#include <linux/netdevice.h>
+#include <linux/spinlock.h>
+#include <linux/if_vlan.h>
+#include <linux/of_mdio.h>
+#include <linux/module.h>
+#include <linux/of_net.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/phy.h>
+#include <linux/io.h>
+
+#include <asm/octeon/octeon.h>
+#include <asm/octeon/cvmx-mixx-defs.h>
+#include <asm/octeon/cvmx-agl-defs.h>
+
+#define DRV_NAME "octeon_mgmt"
+#define DRV_VERSION "2.0"
+#define DRV_DESCRIPTION \
+	"Cavium Networks Octeon MII (management) port Network Driver"
+
+#define OCTEON_MGMT_NAPI_WEIGHT 16
+
+/* Ring sizes that are powers of two allow for more efficient modulo
+ * opertions.
+ */
+#define OCTEON_MGMT_RX_RING_SIZE 512
+#define OCTEON_MGMT_TX_RING_SIZE 128
+
+/* Allow 8 bytes for vlan and FCS. */
+#define OCTEON_MGMT_RX_HEADROOM (ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN)
+
+union mgmt_port_ring_entry {
+	u64 d64;
+	struct {
+#define RING_ENTRY_CODE_DONE 0xf
+#define RING_ENTRY_CODE_MORE 0x10
+#ifdef __BIG_ENDIAN_BITFIELD
+		u64 reserved_62_63:2;
+		/* Length of the buffer/packet in bytes */
+		u64 len:14;
+		/* For TX, signals that the packet should be timestamped */
+		u64 tstamp:1;
+		/* The RX error code */
+		u64 code:7;
+		/* Physical address of the buffer */
+		u64 addr:40;
+#else
+		u64 addr:40;
+		u64 code:7;
+		u64 tstamp:1;
+		u64 len:14;
+		u64 reserved_62_63:2;
+#endif
+	} s;
+};
+
+#define MIX_ORING1	0x0
+#define MIX_ORING2	0x8
+#define MIX_IRING1	0x10
+#define MIX_IRING2	0x18
+#define MIX_CTL		0x20
+#define MIX_IRHWM	0x28
+#define MIX_IRCNT	0x30
+#define MIX_ORHWM	0x38
+#define MIX_ORCNT	0x40
+#define MIX_ISR		0x48
+#define MIX_INTENA	0x50
+#define MIX_REMCNT	0x58
+#define MIX_BIST	0x78
+
+#define AGL_GMX_PRT_CFG			0x10
+#define AGL_GMX_RX_FRM_CTL		0x18
+#define AGL_GMX_RX_FRM_MAX		0x30
+#define AGL_GMX_RX_JABBER		0x38
+#define AGL_GMX_RX_STATS_CTL		0x50
+
+#define AGL_GMX_RX_STATS_PKTS_DRP	0xb0
+#define AGL_GMX_RX_STATS_OCTS_DRP	0xb8
+#define AGL_GMX_RX_STATS_PKTS_BAD	0xc0
+
+#define AGL_GMX_RX_ADR_CTL		0x100
+#define AGL_GMX_RX_ADR_CAM_EN		0x108
+#define AGL_GMX_RX_ADR_CAM0		0x180
+#define AGL_GMX_RX_ADR_CAM1		0x188
+#define AGL_GMX_RX_ADR_CAM2		0x190
+#define AGL_GMX_RX_ADR_CAM3		0x198
+#define AGL_GMX_RX_ADR_CAM4		0x1a0
+#define AGL_GMX_RX_ADR_CAM5		0x1a8
+
+#define AGL_GMX_TX_CLK			0x208
+#define AGL_GMX_TX_STATS_CTL		0x268
+#define AGL_GMX_TX_CTL			0x270
+#define AGL_GMX_TX_STAT0		0x280
+#define AGL_GMX_TX_STAT1		0x288
+#define AGL_GMX_TX_STAT2		0x290
+#define AGL_GMX_TX_STAT3		0x298
+#define AGL_GMX_TX_STAT4		0x2a0
+#define AGL_GMX_TX_STAT5		0x2a8
+#define AGL_GMX_TX_STAT6		0x2b0
+#define AGL_GMX_TX_STAT7		0x2b8
+#define AGL_GMX_TX_STAT8		0x2c0
+#define AGL_GMX_TX_STAT9		0x2c8
+
+struct octeon_mgmt {
+	struct net_device *netdev;
+	u64 mix;
+	u64 agl;
+	u64 agl_prt_ctl;
+	int port;
+	int irq;
+	bool has_rx_tstamp;
+	u64 *tx_ring;
+	dma_addr_t tx_ring_handle;
+	unsigned int tx_next;
+	unsigned int tx_next_clean;
+	unsigned int tx_current_fill;
+	/* The tx_list lock also protects the ring related variables */
+	struct sk_buff_head tx_list;
+
+	/* RX variables only touched in napi_poll.  No locking necessary. */
+	u64 *rx_ring;
+	dma_addr_t rx_ring_handle;
+	unsigned int rx_next;
+	unsigned int rx_next_fill;
+	unsigned int rx_current_fill;
+	struct sk_buff_head rx_list;
+
+	spinlock_t lock;
+	unsigned int last_duplex;
+	unsigned int last_link;
+	unsigned int last_speed;
+	struct device *dev;
+	struct napi_struct napi;
+	struct tasklet_struct tx_clean_tasklet;
+	struct phy_device *phydev;
+	struct device_node *phy_np;
+	resource_size_t mix_phys;
+	resource_size_t mix_size;
+	resource_size_t agl_phys;
+	resource_size_t agl_size;
+	resource_size_t agl_prt_ctl_phys;
+	resource_size_t agl_prt_ctl_size;
+};
+
+static void octeon_mgmt_set_rx_irq(struct octeon_mgmt *p, int enable)
+{
+	union cvmx_mixx_intena mix_intena;
+	unsigned long flags;
+
+	spin_lock_irqsave(&p->lock, flags);
+	mix_intena.u64 = cvmx_read_csr(p->mix + MIX_INTENA);
+	mix_intena.s.ithena = enable ? 1 : 0;
+	cvmx_write_csr(p->mix + MIX_INTENA, mix_intena.u64);
+	spin_unlock_irqrestore(&p->lock, flags);
+}
+
+static void octeon_mgmt_set_tx_irq(struct octeon_mgmt *p, int enable)
+{
+	union cvmx_mixx_intena mix_intena;
+	unsigned long flags;
+
+	spin_lock_irqsave(&p->lock, flags);
+	mix_intena.u64 = cvmx_read_csr(p->mix + MIX_INTENA);
+	mix_intena.s.othena = enable ? 1 : 0;
+	cvmx_write_csr(p->mix + MIX_INTENA, mix_intena.u64);
+	spin_unlock_irqrestore(&p->lock, flags);
+}
+
+static void octeon_mgmt_enable_rx_irq(struct octeon_mgmt *p)
+{
+	octeon_mgmt_set_rx_irq(p, 1);
+}
+
+static void octeon_mgmt_disable_rx_irq(struct octeon_mgmt *p)
+{
+	octeon_mgmt_set_rx_irq(p, 0);
+}
+
+static void octeon_mgmt_enable_tx_irq(struct octeon_mgmt *p)
+{
+	octeon_mgmt_set_tx_irq(p, 1);
+}
+
+static void octeon_mgmt_disable_tx_irq(struct octeon_mgmt *p)
+{
+	octeon_mgmt_set_tx_irq(p, 0);
+}
+
+static unsigned int ring_max_fill(unsigned int ring_size)
+{
+	return ring_size - 8;
+}
+
+static unsigned int ring_size_to_bytes(unsigned int ring_size)
+{
+	return ring_size * sizeof(union mgmt_port_ring_entry);
+}
+
+static void octeon_mgmt_rx_fill_ring(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	while (p->rx_current_fill < ring_max_fill(OCTEON_MGMT_RX_RING_SIZE)) {
+		unsigned int size;
+		union mgmt_port_ring_entry re;
+		struct sk_buff *skb;
+
+		/* CN56XX pass 1 needs 8 bytes of padding.  */
+		size = netdev->mtu + OCTEON_MGMT_RX_HEADROOM + 8 + NET_IP_ALIGN;
+
+		skb = netdev_alloc_skb(netdev, size);
+		if (!skb)
+			break;
+		skb_reserve(skb, NET_IP_ALIGN);
+		__skb_queue_tail(&p->rx_list, skb);
+
+		re.d64 = 0;
+		re.s.len = size;
+		re.s.addr = dma_map_single(p->dev, skb->data,
+					   size,
+					   DMA_FROM_DEVICE);
+
+		/* Put it in the ring.  */
+		p->rx_ring[p->rx_next_fill] = re.d64;
+		dma_sync_single_for_device(p->dev, p->rx_ring_handle,
+					   ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
+					   DMA_BIDIRECTIONAL);
+		p->rx_next_fill =
+			(p->rx_next_fill + 1) % OCTEON_MGMT_RX_RING_SIZE;
+		p->rx_current_fill++;
+		/* Ring the bell.  */
+		cvmx_write_csr(p->mix + MIX_IRING2, 1);
+	}
+}
+
+static void octeon_mgmt_clean_tx_buffers(struct octeon_mgmt *p)
+{
+	union cvmx_mixx_orcnt mix_orcnt;
+	union mgmt_port_ring_entry re;
+	struct sk_buff *skb;
+	int cleaned = 0;
+	unsigned long flags;
+
+	mix_orcnt.u64 = cvmx_read_csr(p->mix + MIX_ORCNT);
+	while (mix_orcnt.s.orcnt) {
+		spin_lock_irqsave(&p->tx_list.lock, flags);
+
+		mix_orcnt.u64 = cvmx_read_csr(p->mix + MIX_ORCNT);
+
+		if (mix_orcnt.s.orcnt == 0) {
+			spin_unlock_irqrestore(&p->tx_list.lock, flags);
+			break;
+		}
+
+		dma_sync_single_for_cpu(p->dev, p->tx_ring_handle,
+					ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
+					DMA_BIDIRECTIONAL);
+
+		re.d64 = p->tx_ring[p->tx_next_clean];
+		p->tx_next_clean =
+			(p->tx_next_clean + 1) % OCTEON_MGMT_TX_RING_SIZE;
+		skb = __skb_dequeue(&p->tx_list);
+
+		mix_orcnt.u64 = 0;
+		mix_orcnt.s.orcnt = 1;
+
+		/* Acknowledge to hardware that we have the buffer.  */
+		cvmx_write_csr(p->mix + MIX_ORCNT, mix_orcnt.u64);
+		p->tx_current_fill--;
+
+		spin_unlock_irqrestore(&p->tx_list.lock, flags);
+
+		dma_unmap_single(p->dev, re.s.addr, re.s.len,
+				 DMA_TO_DEVICE);
+
+		/* Read the hardware TX timestamp if one was recorded */
+		if (unlikely(re.s.tstamp)) {
+			struct skb_shared_hwtstamps ts;
+			u64 ns;
+
+			memset(&ts, 0, sizeof(ts));
+			/* Read the timestamp */
+			ns = cvmx_read_csr(CVMX_MIXX_TSTAMP(p->port));
+			/* Remove the timestamp from the FIFO */
+			cvmx_write_csr(CVMX_MIXX_TSCTL(p->port), 0);
+			/* Tell the kernel about the timestamp */
+			ts.hwtstamp = ns_to_ktime(ns);
+			skb_tstamp_tx(skb, &ts);
+		}
+
+		dev_kfree_skb_any(skb);
+		cleaned++;
+
+		mix_orcnt.u64 = cvmx_read_csr(p->mix + MIX_ORCNT);
+	}
+
+	if (cleaned && netif_queue_stopped(p->netdev))
+		netif_wake_queue(p->netdev);
+}
+
+static void octeon_mgmt_clean_tx_tasklet(unsigned long arg)
+{
+	struct octeon_mgmt *p = (struct octeon_mgmt *)arg;
+	octeon_mgmt_clean_tx_buffers(p);
+	octeon_mgmt_enable_tx_irq(p);
+}
+
+static void octeon_mgmt_update_rx_stats(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	unsigned long flags;
+	u64 drop, bad;
+
+	/* These reads also clear the count registers.  */
+	drop = cvmx_read_csr(p->agl + AGL_GMX_RX_STATS_PKTS_DRP);
+	bad = cvmx_read_csr(p->agl + AGL_GMX_RX_STATS_PKTS_BAD);
+
+	if (drop || bad) {
+		/* Do an atomic update. */
+		spin_lock_irqsave(&p->lock, flags);
+		netdev->stats.rx_errors += bad;
+		netdev->stats.rx_dropped += drop;
+		spin_unlock_irqrestore(&p->lock, flags);
+	}
+}
+
+static void octeon_mgmt_update_tx_stats(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	unsigned long flags;
+
+	union cvmx_agl_gmx_txx_stat0 s0;
+	union cvmx_agl_gmx_txx_stat1 s1;
+
+	/* These reads also clear the count registers.  */
+	s0.u64 = cvmx_read_csr(p->agl + AGL_GMX_TX_STAT0);
+	s1.u64 = cvmx_read_csr(p->agl + AGL_GMX_TX_STAT1);
+
+	if (s0.s.xsdef || s0.s.xscol || s1.s.scol || s1.s.mcol) {
+		/* Do an atomic update. */
+		spin_lock_irqsave(&p->lock, flags);
+		netdev->stats.tx_errors += s0.s.xsdef + s0.s.xscol;
+		netdev->stats.collisions += s1.s.scol + s1.s.mcol;
+		spin_unlock_irqrestore(&p->lock, flags);
+	}
+}
+
+/*
+ * Dequeue a receive skb and its corresponding ring entry.  The ring
+ * entry is returned, *pskb is updated to point to the skb.
+ */
+static u64 octeon_mgmt_dequeue_rx_buffer(struct octeon_mgmt *p,
+					 struct sk_buff **pskb)
+{
+	union mgmt_port_ring_entry re;
+
+	dma_sync_single_for_cpu(p->dev, p->rx_ring_handle,
+				ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
+				DMA_BIDIRECTIONAL);
+
+	re.d64 = p->rx_ring[p->rx_next];
+	p->rx_next = (p->rx_next + 1) % OCTEON_MGMT_RX_RING_SIZE;
+	p->rx_current_fill--;
+	*pskb = __skb_dequeue(&p->rx_list);
+
+	dma_unmap_single(p->dev, re.s.addr,
+			 ETH_FRAME_LEN + OCTEON_MGMT_RX_HEADROOM,
+			 DMA_FROM_DEVICE);
+
+	return re.d64;
+}
+
+
+static int octeon_mgmt_receive_one(struct octeon_mgmt *p)
+{
+	struct net_device *netdev = p->netdev;
+	union cvmx_mixx_ircnt mix_ircnt;
+	union mgmt_port_ring_entry re;
+	struct sk_buff *skb;
+	struct sk_buff *skb2;
+	struct sk_buff *skb_new;
+	union mgmt_port_ring_entry re2;
+	int rc = 1;
+
+
+	re.d64 = octeon_mgmt_dequeue_rx_buffer(p, &skb);
+	if (likely(re.s.code == RING_ENTRY_CODE_DONE)) {
+		/* A good packet, send it up. */
+		skb_put(skb, re.s.len);
+good:
+		/* Process the RX timestamp if it was recorded */
+		if (p->has_rx_tstamp) {
+			/* The first 8 bytes are the timestamp */
+			u64 ns = *(u64 *)skb->data;
+			struct skb_shared_hwtstamps *ts;
+			ts = skb_hwtstamps(skb);
+			ts->hwtstamp = ns_to_ktime(ns);
+			__skb_pull(skb, 8);
+		}
+		skb->protocol = eth_type_trans(skb, netdev);
+		netdev->stats.rx_packets++;
+		netdev->stats.rx_bytes += skb->len;
+		netif_receive_skb(skb);
+		rc = 0;
+	} else if (re.s.code == RING_ENTRY_CODE_MORE) {
+		/* Packet split across skbs.  This can happen if we
+		 * increase the MTU.  Buffers that are already in the
+		 * rx ring can then end up being too small.  As the rx
+		 * ring is refilled, buffers sized for the new MTU
+		 * will be used and we should go back to the normal
+		 * non-split case.
+		 */
+		skb_put(skb, re.s.len);
+		do {
+			re2.d64 = octeon_mgmt_dequeue_rx_buffer(p, &skb2);
+			if (re2.s.code != RING_ENTRY_CODE_MORE
+				&& re2.s.code != RING_ENTRY_CODE_DONE)
+				goto split_error;
+			skb_put(skb2,  re2.s.len);
+			skb_new = skb_copy_expand(skb, 0, skb2->len,
+						  GFP_ATOMIC);
+			if (!skb_new)
+				goto split_error;
+			if (skb_copy_bits(skb2, 0, skb_tail_pointer(skb_new),
+					  skb2->len))
+				goto split_error;
+			skb_put(skb_new, skb2->len);
+			dev_kfree_skb_any(skb);
+			dev_kfree_skb_any(skb2);
+			skb = skb_new;
+		} while (re2.s.code == RING_ENTRY_CODE_MORE);
+		goto good;
+	} else {
+		/* Some other error, discard it. */
+		dev_kfree_skb_any(skb);
+		/* Error statistics are accumulated in
+		 * octeon_mgmt_update_rx_stats.
+		 */
+	}
+	goto done;
+split_error:
+	/* Discard the whole mess. */
+	dev_kfree_skb_any(skb);
+	dev_kfree_skb_any(skb2);
+	while (re2.s.code == RING_ENTRY_CODE_MORE) {
+		re2.d64 = octeon_mgmt_dequeue_rx_buffer(p, &skb2);
+		dev_kfree_skb_any(skb2);
+	}
+	netdev->stats.rx_errors++;
+
+done:
+	/* Tell the hardware we processed a packet.  */
+	mix_ircnt.u64 = 0;
+	mix_ircnt.s.ircnt = 1;
+	cvmx_write_csr(p->mix + MIX_IRCNT, mix_ircnt.u64);
+	return rc;
+}
+
+static int octeon_mgmt_receive_packets(struct octeon_mgmt *p, int budget)
+{
+	unsigned int work_done = 0;
+	union cvmx_mixx_ircnt mix_ircnt;
+	int rc;
+
+	mix_ircnt.u64 = cvmx_read_csr(p->mix + MIX_IRCNT);
+	while (work_done < budget && mix_ircnt.s.ircnt) {
+
+		rc = octeon_mgmt_receive_one(p);
+		if (!rc)
+			work_done++;
+
+		/* Check for more packets. */
+		mix_ircnt.u64 = cvmx_read_csr(p->mix + MIX_IRCNT);
+	}
+
+	octeon_mgmt_rx_fill_ring(p->netdev);
+
+	return work_done;
+}
+
+static int octeon_mgmt_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct octeon_mgmt *p = container_of(napi, struct octeon_mgmt, napi);
+	struct net_device *netdev = p->netdev;
+	unsigned int work_done = 0;
+
+	work_done = octeon_mgmt_receive_packets(p, budget);
+
+	if (work_done < budget) {
+		/* We stopped because no more packets were available. */
+		napi_complete(napi);
+		octeon_mgmt_enable_rx_irq(p);
+	}
+	octeon_mgmt_update_rx_stats(netdev);
+
+	return work_done;
+}
+
+/* Reset the hardware to clean state.  */
+static void octeon_mgmt_reset_hw(struct octeon_mgmt *p)
+{
+	union cvmx_mixx_ctl mix_ctl;
+	union cvmx_mixx_bist mix_bist;
+	union cvmx_agl_gmx_bist agl_gmx_bist;
+
+	mix_ctl.u64 = 0;
+	cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
+	do {
+		mix_ctl.u64 = cvmx_read_csr(p->mix + MIX_CTL);
+	} while (mix_ctl.s.busy);
+	mix_ctl.s.reset = 1;
+	cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
+	cvmx_read_csr(p->mix + MIX_CTL);
+	octeon_io_clk_delay(64);
+
+	mix_bist.u64 = cvmx_read_csr(p->mix + MIX_BIST);
+	if (mix_bist.u64)
+		dev_warn(p->dev, "MIX failed BIST (0x%016llx)\n",
+			(unsigned long long)mix_bist.u64);
+
+	agl_gmx_bist.u64 = cvmx_read_csr(CVMX_AGL_GMX_BIST);
+	if (agl_gmx_bist.u64)
+		dev_warn(p->dev, "AGL failed BIST (0x%016llx)\n",
+			 (unsigned long long)agl_gmx_bist.u64);
+}
+
+struct octeon_mgmt_cam_state {
+	u64 cam[6];
+	u64 cam_mask;
+	int cam_index;
+};
+
+static void octeon_mgmt_cam_state_add(struct octeon_mgmt_cam_state *cs,
+				      unsigned char *addr)
+{
+	int i;
+
+	for (i = 0; i < 6; i++)
+		cs->cam[i] |= (u64)addr[i] << (8 * (cs->cam_index));
+	cs->cam_mask |= (1ULL << cs->cam_index);
+	cs->cam_index++;
+}
+
+static void octeon_mgmt_set_rx_filtering(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	union cvmx_agl_gmx_rxx_adr_ctl adr_ctl;
+	union cvmx_agl_gmx_prtx_cfg agl_gmx_prtx;
+	unsigned long flags;
+	unsigned int prev_packet_enable;
+	unsigned int cam_mode = 1; /* 1 - Accept on CAM match */
+	unsigned int multicast_mode = 1; /* 1 - Reject all multicast.  */
+	struct octeon_mgmt_cam_state cam_state;
+	struct netdev_hw_addr *ha;
+	int available_cam_entries;
+
+	memset(&cam_state, 0, sizeof(cam_state));
+
+	if ((netdev->flags & IFF_PROMISC) || netdev->uc.count > 7) {
+		cam_mode = 0;
+		available_cam_entries = 8;
+	} else {
+		/* One CAM entry for the primary address, leaves seven
+		 * for the secondary addresses.
+		 */
+		available_cam_entries = 7 - netdev->uc.count;
+	}
+
+	if (netdev->flags & IFF_MULTICAST) {
+		if (cam_mode == 0 || (netdev->flags & IFF_ALLMULTI) ||
+		    netdev_mc_count(netdev) > available_cam_entries)
+			multicast_mode = 2; /* 2 - Accept all multicast.  */
+		else
+			multicast_mode = 0; /* 0 - Use CAM.  */
+	}
+
+	if (cam_mode == 1) {
+		/* Add primary address. */
+		octeon_mgmt_cam_state_add(&cam_state, netdev->dev_addr);
+		netdev_for_each_uc_addr(ha, netdev)
+			octeon_mgmt_cam_state_add(&cam_state, ha->addr);
+	}
+	if (multicast_mode == 0) {
+		netdev_for_each_mc_addr(ha, netdev)
+			octeon_mgmt_cam_state_add(&cam_state, ha->addr);
+	}
+
+	spin_lock_irqsave(&p->lock, flags);
+
+	/* Disable packet I/O. */
+	agl_gmx_prtx.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
+	prev_packet_enable = agl_gmx_prtx.s.en;
+	agl_gmx_prtx.s.en = 0;
+	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, agl_gmx_prtx.u64);
+
+	adr_ctl.u64 = 0;
+	adr_ctl.s.cam_mode = cam_mode;
+	adr_ctl.s.mcst = multicast_mode;
+	adr_ctl.s.bcst = 1;     /* Allow broadcast */
+
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CTL, adr_ctl.u64);
+
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM0, cam_state.cam[0]);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM1, cam_state.cam[1]);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM2, cam_state.cam[2]);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM3, cam_state.cam[3]);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM4, cam_state.cam[4]);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM5, cam_state.cam[5]);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM_EN, cam_state.cam_mask);
+
+	/* Restore packet I/O. */
+	agl_gmx_prtx.s.en = prev_packet_enable;
+	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, agl_gmx_prtx.u64);
+
+	spin_unlock_irqrestore(&p->lock, flags);
+}
+
+static int octeon_mgmt_set_mac_address(struct net_device *netdev, void *addr)
+{
+	int r = eth_mac_addr(netdev, addr);
+
+	if (r)
+		return r;
+
+	octeon_mgmt_set_rx_filtering(netdev);
+
+	return 0;
+}
+
+static int octeon_mgmt_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	int size_without_fcs = new_mtu + OCTEON_MGMT_RX_HEADROOM;
+
+	/* Limit the MTU to make sure the ethernet packets are between
+	 * 64 bytes and 16383 bytes.
+	 */
+	if (size_without_fcs < 64 || size_without_fcs > 16383) {
+		dev_warn(p->dev, "MTU must be between %d and %d.\n",
+			 64 - OCTEON_MGMT_RX_HEADROOM,
+			 16383 - OCTEON_MGMT_RX_HEADROOM);
+		return -EINVAL;
+	}
+
+	netdev->mtu = new_mtu;
+
+	cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_MAX, size_without_fcs);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_JABBER,
+		       (size_without_fcs + 7) & 0xfff8);
+
+	return 0;
+}
+
+static irqreturn_t octeon_mgmt_interrupt(int cpl, void *dev_id)
+{
+	struct net_device *netdev = dev_id;
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	union cvmx_mixx_isr mixx_isr;
+
+	mixx_isr.u64 = cvmx_read_csr(p->mix + MIX_ISR);
+
+	/* Clear any pending interrupts */
+	cvmx_write_csr(p->mix + MIX_ISR, mixx_isr.u64);
+	cvmx_read_csr(p->mix + MIX_ISR);
+
+	if (mixx_isr.s.irthresh) {
+		octeon_mgmt_disable_rx_irq(p);
+		napi_schedule(&p->napi);
+	}
+	if (mixx_isr.s.orthresh) {
+		octeon_mgmt_disable_tx_irq(p);
+		tasklet_schedule(&p->tx_clean_tasklet);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev,
+				      struct ifreq *rq, int cmd)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	struct hwtstamp_config config;
+	union cvmx_mio_ptp_clock_cfg ptp;
+	union cvmx_agl_gmx_rxx_frm_ctl rxx_frm_ctl;
+	bool have_hw_timestamps = false;
+
+	if (copy_from_user(&config, rq->ifr_data, sizeof(config)))
+		return -EFAULT;
+
+	if (config.flags) /* reserved for future extensions */
+		return -EINVAL;
+
+	/* Check the status of hardware for tiemstamps */
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+		/* Get the current state of the PTP clock */
+		ptp.u64 = cvmx_read_csr(CVMX_MIO_PTP_CLOCK_CFG);
+		if (!ptp.s.ext_clk_en) {
+			/* The clock has not been configured to use an
+			 * external source.  Program it to use the main clock
+			 * reference.
+			 */
+			u64 clock_comp = (NSEC_PER_SEC << 32) /	octeon_get_io_clock_rate();
+			if (!ptp.s.ptp_en)
+				cvmx_write_csr(CVMX_MIO_PTP_CLOCK_COMP, clock_comp);
+			pr_info("PTP Clock: Using sclk reference at %lld Hz\n",
+				(NSEC_PER_SEC << 32) / clock_comp);
+		} else {
+			/* The clock is already programmed to use a GPIO */
+			u64 clock_comp = cvmx_read_csr(CVMX_MIO_PTP_CLOCK_COMP);
+			pr_info("PTP Clock: Using GPIO %d at %lld Hz\n",
+				ptp.s.ext_clk_in,
+				(NSEC_PER_SEC << 32) / clock_comp);
+		}
+
+		/* Enable the clock if it wasn't done already */
+		if (!ptp.s.ptp_en) {
+			ptp.s.ptp_en = 1;
+			cvmx_write_csr(CVMX_MIO_PTP_CLOCK_CFG, ptp.u64);
+		}
+		have_hw_timestamps = true;
+	}
+
+	if (!have_hw_timestamps)
+		return -EINVAL;
+
+	switch (config.tx_type) {
+	case HWTSTAMP_TX_OFF:
+	case HWTSTAMP_TX_ON:
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (config.rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		p->has_rx_tstamp = false;
+		rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL);
+		rxx_frm_ctl.s.ptp_mode = 0;
+		cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64);
+		break;
+	case HWTSTAMP_FILTER_ALL:
+	case HWTSTAMP_FILTER_SOME:
+	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
+	case HWTSTAMP_FILTER_PTP_V2_EVENT:
+	case HWTSTAMP_FILTER_PTP_V2_SYNC:
+	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+		p->has_rx_tstamp = have_hw_timestamps;
+		config.rx_filter = HWTSTAMP_FILTER_ALL;
+		if (p->has_rx_tstamp) {
+			rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL);
+			rxx_frm_ctl.s.ptp_mode = 1;
+			cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64);
+		}
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	if (copy_to_user(rq->ifr_data, &config, sizeof(config)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int octeon_mgmt_ioctl(struct net_device *netdev,
+			     struct ifreq *rq, int cmd)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	switch (cmd) {
+	case SIOCSHWTSTAMP:
+		return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd);
+	default:
+		if (p->phydev)
+			return phy_mii_ioctl(p->phydev, rq, cmd);
+		return -EINVAL;
+	}
+}
+
+static void octeon_mgmt_disable_link(struct octeon_mgmt *p)
+{
+	union cvmx_agl_gmx_prtx_cfg prtx_cfg;
+
+	/* Disable GMX before we make any changes. */
+	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
+	prtx_cfg.s.en = 0;
+	prtx_cfg.s.tx_en = 0;
+	prtx_cfg.s.rx_en = 0;
+	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, prtx_cfg.u64);
+
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+		int i;
+		for (i = 0; i < 10; i++) {
+			prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
+			if (prtx_cfg.s.tx_idle == 1 || prtx_cfg.s.rx_idle == 1)
+				break;
+			mdelay(1);
+			i++;
+		}
+	}
+}
+
+static void octeon_mgmt_enable_link(struct octeon_mgmt *p)
+{
+	union cvmx_agl_gmx_prtx_cfg prtx_cfg;
+
+	/* Restore the GMX enable state only if link is set */
+	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
+	prtx_cfg.s.tx_en = 1;
+	prtx_cfg.s.rx_en = 1;
+	prtx_cfg.s.en = 1;
+	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, prtx_cfg.u64);
+}
+
+static void octeon_mgmt_update_link(struct octeon_mgmt *p)
+{
+	union cvmx_agl_gmx_prtx_cfg prtx_cfg;
+
+	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
+
+	if (!p->phydev->link)
+		prtx_cfg.s.duplex = 1;
+	else
+		prtx_cfg.s.duplex = p->phydev->duplex;
+
+	switch (p->phydev->speed) {
+	case 10:
+		prtx_cfg.s.speed = 0;
+		prtx_cfg.s.slottime = 0;
+
+		if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+			prtx_cfg.s.burst = 1;
+			prtx_cfg.s.speed_msb = 1;
+		}
+		break;
+	case 100:
+		prtx_cfg.s.speed = 0;
+		prtx_cfg.s.slottime = 0;
+
+		if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+			prtx_cfg.s.burst = 1;
+			prtx_cfg.s.speed_msb = 0;
+		}
+		break;
+	case 1000:
+		/* 1000 MBits is only supported on 6XXX chips */
+		if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+			prtx_cfg.s.speed = 1;
+			prtx_cfg.s.speed_msb = 0;
+			/* Only matters for half-duplex */
+			prtx_cfg.s.slottime = 1;
+			prtx_cfg.s.burst = p->phydev->duplex;
+		}
+		break;
+	case 0:  /* No link */
+	default:
+		break;
+	}
+
+	/* Write the new GMX setting with the port still disabled. */
+	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, prtx_cfg.u64);
+
+	/* Read GMX CFG again to make sure the config is completed. */
+	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
+
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+		union cvmx_agl_gmx_txx_clk agl_clk;
+		union cvmx_agl_prtx_ctl prtx_ctl;
+
+		prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
+		agl_clk.u64 = cvmx_read_csr(p->agl + AGL_GMX_TX_CLK);
+		/* MII (both speeds) and RGMII 1000 speed. */
+		agl_clk.s.clk_cnt = 1;
+		if (prtx_ctl.s.mode == 0) { /* RGMII mode */
+			if (p->phydev->speed == 10)
+				agl_clk.s.clk_cnt = 50;
+			else if (p->phydev->speed == 100)
+				agl_clk.s.clk_cnt = 5;
+		}
+		cvmx_write_csr(p->agl + AGL_GMX_TX_CLK, agl_clk.u64);
+	}
+}
+
+static void octeon_mgmt_adjust_link(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	unsigned long flags;
+	int link_changed = 0;
+
+	if (!p->phydev)
+		return;
+
+	spin_lock_irqsave(&p->lock, flags);
+
+
+	if (!p->phydev->link && p->last_link)
+		link_changed = -1;
+
+	if (p->phydev->link
+	    && (p->last_duplex != p->phydev->duplex
+		|| p->last_link != p->phydev->link
+		|| p->last_speed != p->phydev->speed)) {
+		octeon_mgmt_disable_link(p);
+		link_changed = 1;
+		octeon_mgmt_update_link(p);
+		octeon_mgmt_enable_link(p);
+	}
+
+	p->last_link = p->phydev->link;
+	p->last_speed = p->phydev->speed;
+	p->last_duplex = p->phydev->duplex;
+
+	spin_unlock_irqrestore(&p->lock, flags);
+
+	if (link_changed != 0) {
+		if (link_changed > 0) {
+			pr_info("%s: Link is up - %d/%s\n", netdev->name,
+				p->phydev->speed,
+				DUPLEX_FULL == p->phydev->duplex ?
+				"Full" : "Half");
+		} else {
+			pr_info("%s: Link is down\n", netdev->name);
+		}
+	}
+}
+
+static int octeon_mgmt_init_phy(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	if (octeon_is_simulation() || p->phy_np == NULL) {
+		/* No PHYs in the simulator. */
+		netif_carrier_on(netdev);
+		return 0;
+	}
+
+	p->phydev = of_phy_connect(netdev, p->phy_np,
+				   octeon_mgmt_adjust_link, 0,
+				   PHY_INTERFACE_MODE_MII);
+
+	if (!p->phydev)
+		return -ENODEV;
+
+	return 0;
+}
+
+static int octeon_mgmt_open(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	union cvmx_mixx_ctl mix_ctl;
+	union cvmx_agl_gmx_inf_mode agl_gmx_inf_mode;
+	union cvmx_mixx_oring1 oring1;
+	union cvmx_mixx_iring1 iring1;
+	union cvmx_agl_gmx_rxx_frm_ctl rxx_frm_ctl;
+	union cvmx_mixx_irhwm mix_irhwm;
+	union cvmx_mixx_orhwm mix_orhwm;
+	union cvmx_mixx_intena mix_intena;
+	struct sockaddr sa;
+
+	/* Allocate ring buffers.  */
+	p->tx_ring = kzalloc(ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
+			     GFP_KERNEL);
+	if (!p->tx_ring)
+		return -ENOMEM;
+	p->tx_ring_handle =
+		dma_map_single(p->dev, p->tx_ring,
+			       ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
+			       DMA_BIDIRECTIONAL);
+	p->tx_next = 0;
+	p->tx_next_clean = 0;
+	p->tx_current_fill = 0;
+
+
+	p->rx_ring = kzalloc(ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
+			     GFP_KERNEL);
+	if (!p->rx_ring)
+		goto err_nomem;
+	p->rx_ring_handle =
+		dma_map_single(p->dev, p->rx_ring,
+			       ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
+			       DMA_BIDIRECTIONAL);
+
+	p->rx_next = 0;
+	p->rx_next_fill = 0;
+	p->rx_current_fill = 0;
+
+	octeon_mgmt_reset_hw(p);
+
+	mix_ctl.u64 = cvmx_read_csr(p->mix + MIX_CTL);
+
+	/* Bring it out of reset if needed. */
+	if (mix_ctl.s.reset) {
+		mix_ctl.s.reset = 0;
+		cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
+		do {
+			mix_ctl.u64 = cvmx_read_csr(p->mix + MIX_CTL);
+		} while (mix_ctl.s.reset);
+	}
+
+	if (OCTEON_IS_MODEL(OCTEON_CN5XXX)) {
+		agl_gmx_inf_mode.u64 = 0;
+		agl_gmx_inf_mode.s.en = 1;
+		cvmx_write_csr(CVMX_AGL_GMX_INF_MODE, agl_gmx_inf_mode.u64);
+	}
+	if (OCTEON_IS_MODEL(OCTEON_CN56XX_PASS1_X)
+		|| OCTEON_IS_MODEL(OCTEON_CN52XX_PASS1_X)) {
+		/* Force compensation values, as they are not
+		 * determined properly by HW
+		 */
+		union cvmx_agl_gmx_drv_ctl drv_ctl;
+
+		drv_ctl.u64 = cvmx_read_csr(CVMX_AGL_GMX_DRV_CTL);
+		if (p->port) {
+			drv_ctl.s.byp_en1 = 1;
+			drv_ctl.s.nctl1 = 6;
+			drv_ctl.s.pctl1 = 6;
+		} else {
+			drv_ctl.s.byp_en = 1;
+			drv_ctl.s.nctl = 6;
+			drv_ctl.s.pctl = 6;
+		}
+		cvmx_write_csr(CVMX_AGL_GMX_DRV_CTL, drv_ctl.u64);
+	}
+
+	oring1.u64 = 0;
+	oring1.s.obase = p->tx_ring_handle >> 3;
+	oring1.s.osize = OCTEON_MGMT_TX_RING_SIZE;
+	cvmx_write_csr(p->mix + MIX_ORING1, oring1.u64);
+
+	iring1.u64 = 0;
+	iring1.s.ibase = p->rx_ring_handle >> 3;
+	iring1.s.isize = OCTEON_MGMT_RX_RING_SIZE;
+	cvmx_write_csr(p->mix + MIX_IRING1, iring1.u64);
+
+	memcpy(sa.sa_data, netdev->dev_addr, ETH_ALEN);
+	octeon_mgmt_set_mac_address(netdev, &sa);
+
+	octeon_mgmt_change_mtu(netdev, netdev->mtu);
+
+	/* Enable the port HW. Packets are not allowed until
+	 * cvmx_mgmt_port_enable() is called.
+	 */
+	mix_ctl.u64 = 0;
+	mix_ctl.s.crc_strip = 1;    /* Strip the ending CRC */
+	mix_ctl.s.en = 1;           /* Enable the port */
+	mix_ctl.s.nbtarb = 0;       /* Arbitration mode */
+	/* MII CB-request FIFO programmable high watermark */
+	mix_ctl.s.mrq_hwm = 1;
+#ifdef __LITTLE_ENDIAN
+	mix_ctl.s.lendian = 1;
+#endif
+	cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
+
+	/* Read the PHY to find the mode of the interface. */
+	if (octeon_mgmt_init_phy(netdev)) {
+		dev_err(p->dev, "Cannot initialize PHY on MIX%d.\n", p->port);
+		goto err_noirq;
+	}
+
+	/* Set the mode of the interface, RGMII/MII. */
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && p->phydev) {
+		union cvmx_agl_prtx_ctl agl_prtx_ctl;
+		int rgmii_mode = (p->phydev->supported &
+				  (SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full)) != 0;
+
+		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
+		agl_prtx_ctl.s.mode = rgmii_mode ? 0 : 1;
+		cvmx_write_csr(p->agl_prt_ctl,	agl_prtx_ctl.u64);
+
+		/* MII clocks counts are based on the 125Mhz
+		 * reference, which has an 8nS period. So our delays
+		 * need to be multiplied by this factor.
+		 */
+#define NS_PER_PHY_CLK 8
+
+		/* Take the DLL and clock tree out of reset */
+		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
+		agl_prtx_ctl.s.clkrst = 0;
+		if (rgmii_mode) {
+			agl_prtx_ctl.s.dllrst = 0;
+			agl_prtx_ctl.s.clktx_byp = 0;
+		}
+		cvmx_write_csr(p->agl_prt_ctl,	agl_prtx_ctl.u64);
+		cvmx_read_csr(p->agl_prt_ctl); /* Force write out before wait */
+
+		/* Wait for the DLL to lock. External 125 MHz
+		 * reference clock must be stable at this point.
+		 */
+		ndelay(256 * NS_PER_PHY_CLK);
+
+		/* Enable the interface */
+		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
+		agl_prtx_ctl.s.enable = 1;
+		cvmx_write_csr(p->agl_prt_ctl, agl_prtx_ctl.u64);
+
+		/* Read the value back to force the previous write */
+		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
+
+		/* Enable the compensation controller */
+		agl_prtx_ctl.s.comp = 1;
+		agl_prtx_ctl.s.drv_byp = 0;
+		cvmx_write_csr(p->agl_prt_ctl,	agl_prtx_ctl.u64);
+		/* Force write out before wait. */
+		cvmx_read_csr(p->agl_prt_ctl);
+
+		/* For compensation state to lock. */
+		ndelay(1040 * NS_PER_PHY_CLK);
+
+		/* Default Interframe Gaps are too small.  Recommended
+		 * workaround is.
+		 *
+		 * AGL_GMX_TX_IFG[IFG1]=14
+		 * AGL_GMX_TX_IFG[IFG2]=10
+		 */
+		cvmx_write_csr(CVMX_AGL_GMX_TX_IFG, 0xae);
+	}
+
+	octeon_mgmt_rx_fill_ring(netdev);
+
+	/* Clear statistics. */
+	/* Clear on read. */
+	cvmx_write_csr(p->agl + AGL_GMX_RX_STATS_CTL, 1);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_STATS_PKTS_DRP, 0);
+	cvmx_write_csr(p->agl + AGL_GMX_RX_STATS_PKTS_BAD, 0);
+
+	cvmx_write_csr(p->agl + AGL_GMX_TX_STATS_CTL, 1);
+	cvmx_write_csr(p->agl + AGL_GMX_TX_STAT0, 0);
+	cvmx_write_csr(p->agl + AGL_GMX_TX_STAT1, 0);
+
+	/* Clear any pending interrupts */
+	cvmx_write_csr(p->mix + MIX_ISR, cvmx_read_csr(p->mix + MIX_ISR));
+
+	if (request_irq(p->irq, octeon_mgmt_interrupt, 0, netdev->name,
+			netdev)) {
+		dev_err(p->dev, "request_irq(%d) failed.\n", p->irq);
+		goto err_noirq;
+	}
+
+	/* Interrupt every single RX packet */
+	mix_irhwm.u64 = 0;
+	mix_irhwm.s.irhwm = 0;
+	cvmx_write_csr(p->mix + MIX_IRHWM, mix_irhwm.u64);
+
+	/* Interrupt when we have 1 or more packets to clean.  */
+	mix_orhwm.u64 = 0;
+	mix_orhwm.s.orhwm = 0;
+	cvmx_write_csr(p->mix + MIX_ORHWM, mix_orhwm.u64);
+
+	/* Enable receive and transmit interrupts */
+	mix_intena.u64 = 0;
+	mix_intena.s.ithena = 1;
+	mix_intena.s.othena = 1;
+	cvmx_write_csr(p->mix + MIX_INTENA, mix_intena.u64);
+
+	/* Enable packet I/O. */
+
+	rxx_frm_ctl.u64 = 0;
+	rxx_frm_ctl.s.ptp_mode = p->has_rx_tstamp ? 1 : 0;
+	rxx_frm_ctl.s.pre_align = 1;
+	/* When set, disables the length check for non-min sized pkts
+	 * with padding in the client data.
+	 */
+	rxx_frm_ctl.s.pad_len = 1;
+	/* When set, disables the length check for VLAN pkts */
+	rxx_frm_ctl.s.vlan_len = 1;
+	/* When set, PREAMBLE checking is  less strict */
+	rxx_frm_ctl.s.pre_free = 1;
+	/* Control Pause Frames can match station SMAC */
+	rxx_frm_ctl.s.ctl_smac = 0;
+	/* Control Pause Frames can match globally assign Multicast address */
+	rxx_frm_ctl.s.ctl_mcst = 1;
+	/* Forward pause information to TX block */
+	rxx_frm_ctl.s.ctl_bck = 1;
+	/* Drop Control Pause Frames */
+	rxx_frm_ctl.s.ctl_drp = 1;
+	/* Strip off the preamble */
+	rxx_frm_ctl.s.pre_strp = 1;
+	/* This port is configured to send PREAMBLE+SFD to begin every
+	 * frame.  GMX checks that the PREAMBLE is sent correctly.
+	 */
+	rxx_frm_ctl.s.pre_chk = 1;
+	cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64);
+
+	/* Configure the port duplex, speed and enables */
+	octeon_mgmt_disable_link(p);
+	if (p->phydev)
+		octeon_mgmt_update_link(p);
+	octeon_mgmt_enable_link(p);
+
+	p->last_link = 0;
+	p->last_speed = 0;
+	/* PHY is not present in simulator. The carrier is enabled
+	 * while initializing the phy for simulator, leave it enabled.
+	 */
+	if (p->phydev) {
+		netif_carrier_off(netdev);
+		phy_start_aneg(p->phydev);
+	}
+
+	netif_wake_queue(netdev);
+	napi_enable(&p->napi);
+
+	return 0;
+err_noirq:
+	octeon_mgmt_reset_hw(p);
+	dma_unmap_single(p->dev, p->rx_ring_handle,
+			 ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
+			 DMA_BIDIRECTIONAL);
+	kfree(p->rx_ring);
+err_nomem:
+	dma_unmap_single(p->dev, p->tx_ring_handle,
+			 ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
+			 DMA_BIDIRECTIONAL);
+	kfree(p->tx_ring);
+	return -ENOMEM;
+}
+
+static int octeon_mgmt_stop(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	napi_disable(&p->napi);
+	netif_stop_queue(netdev);
+
+	if (p->phydev)
+		phy_disconnect(p->phydev);
+	p->phydev = NULL;
+
+	netif_carrier_off(netdev);
+
+	octeon_mgmt_reset_hw(p);
+
+	free_irq(p->irq, netdev);
+
+	/* dma_unmap is a nop on Octeon, so just free everything.  */
+	skb_queue_purge(&p->tx_list);
+	skb_queue_purge(&p->rx_list);
+
+	dma_unmap_single(p->dev, p->rx_ring_handle,
+			 ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
+			 DMA_BIDIRECTIONAL);
+	kfree(p->rx_ring);
+
+	dma_unmap_single(p->dev, p->tx_ring_handle,
+			 ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
+			 DMA_BIDIRECTIONAL);
+	kfree(p->tx_ring);
+
+	return 0;
+}
+
+static int octeon_mgmt_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+	union mgmt_port_ring_entry re;
+	unsigned long flags;
+	int rv = NETDEV_TX_BUSY;
+
+	re.d64 = 0;
+	re.s.tstamp = ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) != 0);
+	re.s.len = skb->len;
+	re.s.addr = dma_map_single(p->dev, skb->data,
+				   skb->len,
+				   DMA_TO_DEVICE);
+
+	spin_lock_irqsave(&p->tx_list.lock, flags);
+
+	if (unlikely(p->tx_current_fill >= ring_max_fill(OCTEON_MGMT_TX_RING_SIZE) - 1)) {
+		spin_unlock_irqrestore(&p->tx_list.lock, flags);
+		netif_stop_queue(netdev);
+		spin_lock_irqsave(&p->tx_list.lock, flags);
+	}
+
+	if (unlikely(p->tx_current_fill >=
+		     ring_max_fill(OCTEON_MGMT_TX_RING_SIZE))) {
+		spin_unlock_irqrestore(&p->tx_list.lock, flags);
+		dma_unmap_single(p->dev, re.s.addr, re.s.len,
+				 DMA_TO_DEVICE);
+		goto out;
+	}
+
+	__skb_queue_tail(&p->tx_list, skb);
+
+	/* Put it in the ring.  */
+	p->tx_ring[p->tx_next] = re.d64;
+	p->tx_next = (p->tx_next + 1) % OCTEON_MGMT_TX_RING_SIZE;
+	p->tx_current_fill++;
+
+	spin_unlock_irqrestore(&p->tx_list.lock, flags);
+
+	dma_sync_single_for_device(p->dev, p->tx_ring_handle,
+				   ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
+				   DMA_BIDIRECTIONAL);
+
+	netdev->stats.tx_packets++;
+	netdev->stats.tx_bytes += skb->len;
+
+	/* Ring the bell.  */
+	cvmx_write_csr(p->mix + MIX_ORING2, 1);
+
+	netdev->trans_start = jiffies;
+	rv = NETDEV_TX_OK;
+out:
+	octeon_mgmt_update_tx_stats(netdev);
+	return rv;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void octeon_mgmt_poll_controller(struct net_device *netdev)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	octeon_mgmt_receive_packets(p, 16);
+	octeon_mgmt_update_rx_stats(netdev);
+}
+#endif
+
+static void octeon_mgmt_get_drvinfo(struct net_device *netdev,
+				    struct ethtool_drvinfo *info)
+{
+	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
+	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
+	strlcpy(info->bus_info, "N/A", sizeof(info->bus_info));
+}
+
+static int octeon_mgmt_get_settings(struct net_device *netdev,
+				    struct ethtool_cmd *cmd)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	if (p->phydev)
+		return phy_ethtool_gset(p->phydev, cmd);
+
+	return -EOPNOTSUPP;
+}
+
+static int octeon_mgmt_set_settings(struct net_device *netdev,
+				    struct ethtool_cmd *cmd)
+{
+	struct octeon_mgmt *p = netdev_priv(netdev);
+
+	if (!capable(CAP_NET_ADMIN))
+		return -EPERM;
+
+	if (p->phydev)
+		return phy_ethtool_sset(p->phydev, cmd);
+
+	return -EOPNOTSUPP;
+}
+
+static int octeon_mgmt_nway_reset(struct net_device *dev)
+{
+	struct octeon_mgmt *p = netdev_priv(dev);
+
+	if (!capable(CAP_NET_ADMIN))
+		return -EPERM;
+
+	if (p->phydev)
+		return phy_start_aneg(p->phydev);
+
+	return -EOPNOTSUPP;
+}
+
+static const struct ethtool_ops octeon_mgmt_ethtool_ops = {
+	.get_drvinfo = octeon_mgmt_get_drvinfo,
+	.get_settings = octeon_mgmt_get_settings,
+	.set_settings = octeon_mgmt_set_settings,
+	.nway_reset = octeon_mgmt_nway_reset,
+	.get_link = ethtool_op_get_link,
+};
+
+static const struct net_device_ops octeon_mgmt_ops = {
+	.ndo_open =			octeon_mgmt_open,
+	.ndo_stop =			octeon_mgmt_stop,
+	.ndo_start_xmit =		octeon_mgmt_xmit,
+	.ndo_set_rx_mode =		octeon_mgmt_set_rx_filtering,
+	.ndo_set_mac_address =		octeon_mgmt_set_mac_address,
+	.ndo_do_ioctl =			octeon_mgmt_ioctl,
+	.ndo_change_mtu =		octeon_mgmt_change_mtu,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller =		octeon_mgmt_poll_controller,
+#endif
+};
+
+static int octeon_mgmt_probe(struct platform_device *pdev)
+{
+	struct net_device *netdev;
+	struct octeon_mgmt *p;
+	const __be32 *data;
+	const u8 *mac;
+	struct resource *res_mix;
+	struct resource *res_agl;
+	struct resource *res_agl_prt_ctl;
+	int len;
+	int result;
+
+	netdev = alloc_etherdev(sizeof(struct octeon_mgmt));
+	if (netdev == NULL)
+		return -ENOMEM;
+
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+
+	platform_set_drvdata(pdev, netdev);
+	p = netdev_priv(netdev);
+	netif_napi_add(netdev, &p->napi, octeon_mgmt_napi_poll,
+		       OCTEON_MGMT_NAPI_WEIGHT);
+
+	p->netdev = netdev;
+	p->dev = &pdev->dev;
+	p->has_rx_tstamp = false;
+
+	data = of_get_property(pdev->dev.of_node, "cell-index", &len);
+	if (data && len == sizeof(*data)) {
+		p->port = be32_to_cpup(data);
+	} else {
+		dev_err(&pdev->dev, "no 'cell-index' property\n");
+		result = -ENXIO;
+		goto err;
+	}
+
+	snprintf(netdev->name, IFNAMSIZ, "mgmt%d", p->port);
+
+	result = platform_get_irq(pdev, 0);
+	if (result < 0)
+		goto err;
+
+	p->irq = result;
+
+	res_mix = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (res_mix == NULL) {
+		dev_err(&pdev->dev, "no 'reg' resource\n");
+		result = -ENXIO;
+		goto err;
+	}
+
+	res_agl = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	if (res_agl == NULL) {
+		dev_err(&pdev->dev, "no 'reg' resource\n");
+		result = -ENXIO;
+		goto err;
+	}
+
+	res_agl_prt_ctl = platform_get_resource(pdev, IORESOURCE_MEM, 3);
+	if (res_agl_prt_ctl == NULL) {
+		dev_err(&pdev->dev, "no 'reg' resource\n");
+		result = -ENXIO;
+		goto err;
+	}
+
+	p->mix_phys = res_mix->start;
+	p->mix_size = resource_size(res_mix);
+	p->agl_phys = res_agl->start;
+	p->agl_size = resource_size(res_agl);
+	p->agl_prt_ctl_phys = res_agl_prt_ctl->start;
+	p->agl_prt_ctl_size = resource_size(res_agl_prt_ctl);
+
+
+	if (!devm_request_mem_region(&pdev->dev, p->mix_phys, p->mix_size,
+				     res_mix->name)) {
+		dev_err(&pdev->dev, "request_mem_region (%s) failed\n",
+			res_mix->name);
+		result = -ENXIO;
+		goto err;
+	}
+
+	if (!devm_request_mem_region(&pdev->dev, p->agl_phys, p->agl_size,
+				     res_agl->name)) {
+		result = -ENXIO;
+		dev_err(&pdev->dev, "request_mem_region (%s) failed\n",
+			res_agl->name);
+		goto err;
+	}
+
+	if (!devm_request_mem_region(&pdev->dev, p->agl_prt_ctl_phys,
+				     p->agl_prt_ctl_size, res_agl_prt_ctl->name)) {
+		result = -ENXIO;
+		dev_err(&pdev->dev, "request_mem_region (%s) failed\n",
+			res_agl_prt_ctl->name);
+		goto err;
+	}
+
+	p->mix = (u64)devm_ioremap(&pdev->dev, p->mix_phys, p->mix_size);
+	p->agl = (u64)devm_ioremap(&pdev->dev, p->agl_phys, p->agl_size);
+	p->agl_prt_ctl = (u64)devm_ioremap(&pdev->dev, p->agl_prt_ctl_phys,
+					   p->agl_prt_ctl_size);
+	spin_lock_init(&p->lock);
+
+	skb_queue_head_init(&p->tx_list);
+	skb_queue_head_init(&p->rx_list);
+	tasklet_init(&p->tx_clean_tasklet,
+		     octeon_mgmt_clean_tx_tasklet, (unsigned long)p);
+
+	netdev->priv_flags |= IFF_UNICAST_FLT;
+
+	netdev->netdev_ops = &octeon_mgmt_ops;
+	netdev->ethtool_ops = &octeon_mgmt_ethtool_ops;
+
+	mac = of_get_mac_address(pdev->dev.of_node);
+
+	if (mac)
+		memcpy(netdev->dev_addr, mac, ETH_ALEN);
+	else
+		eth_hw_addr_random(netdev);
+
+	p->phy_np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
+
+	result = dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (result)
+		goto err;
+
+	netif_carrier_off(netdev);
+	result = register_netdev(netdev);
+	if (result)
+		goto err;
+
+	dev_info(&pdev->dev, "Version " DRV_VERSION "\n");
+	return 0;
+
+err:
+	free_netdev(netdev);
+	return result;
+}
+
+static int octeon_mgmt_remove(struct platform_device *pdev)
+{
+	struct net_device *netdev = platform_get_drvdata(pdev);
+
+	unregister_netdev(netdev);
+	free_netdev(netdev);
+	return 0;
+}
+
+static const struct of_device_id octeon_mgmt_match[] = {
+	{
+		.compatible = "cavium,octeon-5750-mix",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, octeon_mgmt_match);
+
+static struct platform_driver octeon_mgmt_driver = {
+	.driver = {
+		.name		= "octeon_mgmt",
+		.of_match_table = octeon_mgmt_match,
+	},
+	.probe		= octeon_mgmt_probe,
+	.remove		= octeon_mgmt_remove,
+};
+
+extern void octeon_mdiobus_force_mod_depencency(void);
+
+static int __init octeon_mgmt_mod_init(void)
+{
+	/* Force our mdiobus driver module to be loaded first. */
+	octeon_mdiobus_force_mod_depencency();
+	return platform_driver_register(&octeon_mgmt_driver);
+}
+
+static void __exit octeon_mgmt_mod_exit(void)
+{
+	platform_driver_unregister(&octeon_mgmt_driver);
+}
+
+module_init(octeon_mgmt_mod_init);
+module_exit(octeon_mgmt_mod_exit);
+
+MODULE_DESCRIPTION(DRV_DESCRIPTION);
+MODULE_AUTHOR("David Daney");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
diff --git a/drivers/net/ethernet/octeon/Kconfig b/drivers/net/ethernet/octeon/Kconfig
deleted file mode 100644
index a7aa280..0000000
--- a/drivers/net/ethernet/octeon/Kconfig
+++ /dev/null
@@ -1,14 +0,0 @@
-#
-# Cavium network device configuration
-#
-
-config OCTEON_MGMT_ETHERNET
-	tristate "Octeon Management port ethernet driver (CN5XXX, CN6XXX)"
-	depends on CAVIUM_OCTEON_SOC
-	select PHYLIB
-	select MDIO_OCTEON
-	default y
-	---help---
-	  This option enables the ethernet driver for the management
-	  port on Cavium Networks' Octeon CN57XX, CN56XX, CN55XX,
-	  CN54XX, CN52XX, and CN6XXX chips.
diff --git a/drivers/net/ethernet/octeon/Makefile b/drivers/net/ethernet/octeon/Makefile
deleted file mode 100644
index efa41c1..0000000
--- a/drivers/net/ethernet/octeon/Makefile
+++ /dev/null
@@ -1,5 +0,0 @@
-#
-# Makefile for the Cavium network device drivers.
-#
-
-obj-$(CONFIG_OCTEON_MGMT_ETHERNET)	+= octeon_mgmt.o
diff --git a/drivers/net/ethernet/octeon/octeon_mgmt.c b/drivers/net/ethernet/octeon/octeon_mgmt.c
deleted file mode 100644
index c177c7c..0000000
--- a/drivers/net/ethernet/octeon/octeon_mgmt.c
+++ /dev/null
@@ -1,1593 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2009-2012 Cavium, Inc
- */
-
-#include <linux/platform_device.h>
-#include <linux/dma-mapping.h>
-#include <linux/etherdevice.h>
-#include <linux/capability.h>
-#include <linux/net_tstamp.h>
-#include <linux/interrupt.h>
-#include <linux/netdevice.h>
-#include <linux/spinlock.h>
-#include <linux/if_vlan.h>
-#include <linux/of_mdio.h>
-#include <linux/module.h>
-#include <linux/of_net.h>
-#include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/phy.h>
-#include <linux/io.h>
-
-#include <asm/octeon/octeon.h>
-#include <asm/octeon/cvmx-mixx-defs.h>
-#include <asm/octeon/cvmx-agl-defs.h>
-
-#define DRV_NAME "octeon_mgmt"
-#define DRV_VERSION "2.0"
-#define DRV_DESCRIPTION \
-	"Cavium Networks Octeon MII (management) port Network Driver"
-
-#define OCTEON_MGMT_NAPI_WEIGHT 16
-
-/* Ring sizes that are powers of two allow for more efficient modulo
- * opertions.
- */
-#define OCTEON_MGMT_RX_RING_SIZE 512
-#define OCTEON_MGMT_TX_RING_SIZE 128
-
-/* Allow 8 bytes for vlan and FCS. */
-#define OCTEON_MGMT_RX_HEADROOM (ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN)
-
-union mgmt_port_ring_entry {
-	u64 d64;
-	struct {
-#define RING_ENTRY_CODE_DONE 0xf
-#define RING_ENTRY_CODE_MORE 0x10
-#ifdef __BIG_ENDIAN_BITFIELD
-		u64 reserved_62_63:2;
-		/* Length of the buffer/packet in bytes */
-		u64 len:14;
-		/* For TX, signals that the packet should be timestamped */
-		u64 tstamp:1;
-		/* The RX error code */
-		u64 code:7;
-		/* Physical address of the buffer */
-		u64 addr:40;
-#else
-		u64 addr:40;
-		u64 code:7;
-		u64 tstamp:1;
-		u64 len:14;
-		u64 reserved_62_63:2;
-#endif
-	} s;
-};
-
-#define MIX_ORING1	0x0
-#define MIX_ORING2	0x8
-#define MIX_IRING1	0x10
-#define MIX_IRING2	0x18
-#define MIX_CTL		0x20
-#define MIX_IRHWM	0x28
-#define MIX_IRCNT	0x30
-#define MIX_ORHWM	0x38
-#define MIX_ORCNT	0x40
-#define MIX_ISR		0x48
-#define MIX_INTENA	0x50
-#define MIX_REMCNT	0x58
-#define MIX_BIST	0x78
-
-#define AGL_GMX_PRT_CFG			0x10
-#define AGL_GMX_RX_FRM_CTL		0x18
-#define AGL_GMX_RX_FRM_MAX		0x30
-#define AGL_GMX_RX_JABBER		0x38
-#define AGL_GMX_RX_STATS_CTL		0x50
-
-#define AGL_GMX_RX_STATS_PKTS_DRP	0xb0
-#define AGL_GMX_RX_STATS_OCTS_DRP	0xb8
-#define AGL_GMX_RX_STATS_PKTS_BAD	0xc0
-
-#define AGL_GMX_RX_ADR_CTL		0x100
-#define AGL_GMX_RX_ADR_CAM_EN		0x108
-#define AGL_GMX_RX_ADR_CAM0		0x180
-#define AGL_GMX_RX_ADR_CAM1		0x188
-#define AGL_GMX_RX_ADR_CAM2		0x190
-#define AGL_GMX_RX_ADR_CAM3		0x198
-#define AGL_GMX_RX_ADR_CAM4		0x1a0
-#define AGL_GMX_RX_ADR_CAM5		0x1a8
-
-#define AGL_GMX_TX_CLK			0x208
-#define AGL_GMX_TX_STATS_CTL		0x268
-#define AGL_GMX_TX_CTL			0x270
-#define AGL_GMX_TX_STAT0		0x280
-#define AGL_GMX_TX_STAT1		0x288
-#define AGL_GMX_TX_STAT2		0x290
-#define AGL_GMX_TX_STAT3		0x298
-#define AGL_GMX_TX_STAT4		0x2a0
-#define AGL_GMX_TX_STAT5		0x2a8
-#define AGL_GMX_TX_STAT6		0x2b0
-#define AGL_GMX_TX_STAT7		0x2b8
-#define AGL_GMX_TX_STAT8		0x2c0
-#define AGL_GMX_TX_STAT9		0x2c8
-
-struct octeon_mgmt {
-	struct net_device *netdev;
-	u64 mix;
-	u64 agl;
-	u64 agl_prt_ctl;
-	int port;
-	int irq;
-	bool has_rx_tstamp;
-	u64 *tx_ring;
-	dma_addr_t tx_ring_handle;
-	unsigned int tx_next;
-	unsigned int tx_next_clean;
-	unsigned int tx_current_fill;
-	/* The tx_list lock also protects the ring related variables */
-	struct sk_buff_head tx_list;
-
-	/* RX variables only touched in napi_poll.  No locking necessary. */
-	u64 *rx_ring;
-	dma_addr_t rx_ring_handle;
-	unsigned int rx_next;
-	unsigned int rx_next_fill;
-	unsigned int rx_current_fill;
-	struct sk_buff_head rx_list;
-
-	spinlock_t lock;
-	unsigned int last_duplex;
-	unsigned int last_link;
-	unsigned int last_speed;
-	struct device *dev;
-	struct napi_struct napi;
-	struct tasklet_struct tx_clean_tasklet;
-	struct phy_device *phydev;
-	struct device_node *phy_np;
-	resource_size_t mix_phys;
-	resource_size_t mix_size;
-	resource_size_t agl_phys;
-	resource_size_t agl_size;
-	resource_size_t agl_prt_ctl_phys;
-	resource_size_t agl_prt_ctl_size;
-};
-
-static void octeon_mgmt_set_rx_irq(struct octeon_mgmt *p, int enable)
-{
-	union cvmx_mixx_intena mix_intena;
-	unsigned long flags;
-
-	spin_lock_irqsave(&p->lock, flags);
-	mix_intena.u64 = cvmx_read_csr(p->mix + MIX_INTENA);
-	mix_intena.s.ithena = enable ? 1 : 0;
-	cvmx_write_csr(p->mix + MIX_INTENA, mix_intena.u64);
-	spin_unlock_irqrestore(&p->lock, flags);
-}
-
-static void octeon_mgmt_set_tx_irq(struct octeon_mgmt *p, int enable)
-{
-	union cvmx_mixx_intena mix_intena;
-	unsigned long flags;
-
-	spin_lock_irqsave(&p->lock, flags);
-	mix_intena.u64 = cvmx_read_csr(p->mix + MIX_INTENA);
-	mix_intena.s.othena = enable ? 1 : 0;
-	cvmx_write_csr(p->mix + MIX_INTENA, mix_intena.u64);
-	spin_unlock_irqrestore(&p->lock, flags);
-}
-
-static void octeon_mgmt_enable_rx_irq(struct octeon_mgmt *p)
-{
-	octeon_mgmt_set_rx_irq(p, 1);
-}
-
-static void octeon_mgmt_disable_rx_irq(struct octeon_mgmt *p)
-{
-	octeon_mgmt_set_rx_irq(p, 0);
-}
-
-static void octeon_mgmt_enable_tx_irq(struct octeon_mgmt *p)
-{
-	octeon_mgmt_set_tx_irq(p, 1);
-}
-
-static void octeon_mgmt_disable_tx_irq(struct octeon_mgmt *p)
-{
-	octeon_mgmt_set_tx_irq(p, 0);
-}
-
-static unsigned int ring_max_fill(unsigned int ring_size)
-{
-	return ring_size - 8;
-}
-
-static unsigned int ring_size_to_bytes(unsigned int ring_size)
-{
-	return ring_size * sizeof(union mgmt_port_ring_entry);
-}
-
-static void octeon_mgmt_rx_fill_ring(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	while (p->rx_current_fill < ring_max_fill(OCTEON_MGMT_RX_RING_SIZE)) {
-		unsigned int size;
-		union mgmt_port_ring_entry re;
-		struct sk_buff *skb;
-
-		/* CN56XX pass 1 needs 8 bytes of padding.  */
-		size = netdev->mtu + OCTEON_MGMT_RX_HEADROOM + 8 + NET_IP_ALIGN;
-
-		skb = netdev_alloc_skb(netdev, size);
-		if (!skb)
-			break;
-		skb_reserve(skb, NET_IP_ALIGN);
-		__skb_queue_tail(&p->rx_list, skb);
-
-		re.d64 = 0;
-		re.s.len = size;
-		re.s.addr = dma_map_single(p->dev, skb->data,
-					   size,
-					   DMA_FROM_DEVICE);
-
-		/* Put it in the ring.  */
-		p->rx_ring[p->rx_next_fill] = re.d64;
-		dma_sync_single_for_device(p->dev, p->rx_ring_handle,
-					   ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
-					   DMA_BIDIRECTIONAL);
-		p->rx_next_fill =
-			(p->rx_next_fill + 1) % OCTEON_MGMT_RX_RING_SIZE;
-		p->rx_current_fill++;
-		/* Ring the bell.  */
-		cvmx_write_csr(p->mix + MIX_IRING2, 1);
-	}
-}
-
-static void octeon_mgmt_clean_tx_buffers(struct octeon_mgmt *p)
-{
-	union cvmx_mixx_orcnt mix_orcnt;
-	union mgmt_port_ring_entry re;
-	struct sk_buff *skb;
-	int cleaned = 0;
-	unsigned long flags;
-
-	mix_orcnt.u64 = cvmx_read_csr(p->mix + MIX_ORCNT);
-	while (mix_orcnt.s.orcnt) {
-		spin_lock_irqsave(&p->tx_list.lock, flags);
-
-		mix_orcnt.u64 = cvmx_read_csr(p->mix + MIX_ORCNT);
-
-		if (mix_orcnt.s.orcnt == 0) {
-			spin_unlock_irqrestore(&p->tx_list.lock, flags);
-			break;
-		}
-
-		dma_sync_single_for_cpu(p->dev, p->tx_ring_handle,
-					ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
-					DMA_BIDIRECTIONAL);
-
-		re.d64 = p->tx_ring[p->tx_next_clean];
-		p->tx_next_clean =
-			(p->tx_next_clean + 1) % OCTEON_MGMT_TX_RING_SIZE;
-		skb = __skb_dequeue(&p->tx_list);
-
-		mix_orcnt.u64 = 0;
-		mix_orcnt.s.orcnt = 1;
-
-		/* Acknowledge to hardware that we have the buffer.  */
-		cvmx_write_csr(p->mix + MIX_ORCNT, mix_orcnt.u64);
-		p->tx_current_fill--;
-
-		spin_unlock_irqrestore(&p->tx_list.lock, flags);
-
-		dma_unmap_single(p->dev, re.s.addr, re.s.len,
-				 DMA_TO_DEVICE);
-
-		/* Read the hardware TX timestamp if one was recorded */
-		if (unlikely(re.s.tstamp)) {
-			struct skb_shared_hwtstamps ts;
-			u64 ns;
-
-			memset(&ts, 0, sizeof(ts));
-			/* Read the timestamp */
-			ns = cvmx_read_csr(CVMX_MIXX_TSTAMP(p->port));
-			/* Remove the timestamp from the FIFO */
-			cvmx_write_csr(CVMX_MIXX_TSCTL(p->port), 0);
-			/* Tell the kernel about the timestamp */
-			ts.hwtstamp = ns_to_ktime(ns);
-			skb_tstamp_tx(skb, &ts);
-		}
-
-		dev_kfree_skb_any(skb);
-		cleaned++;
-
-		mix_orcnt.u64 = cvmx_read_csr(p->mix + MIX_ORCNT);
-	}
-
-	if (cleaned && netif_queue_stopped(p->netdev))
-		netif_wake_queue(p->netdev);
-}
-
-static void octeon_mgmt_clean_tx_tasklet(unsigned long arg)
-{
-	struct octeon_mgmt *p = (struct octeon_mgmt *)arg;
-	octeon_mgmt_clean_tx_buffers(p);
-	octeon_mgmt_enable_tx_irq(p);
-}
-
-static void octeon_mgmt_update_rx_stats(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	unsigned long flags;
-	u64 drop, bad;
-
-	/* These reads also clear the count registers.  */
-	drop = cvmx_read_csr(p->agl + AGL_GMX_RX_STATS_PKTS_DRP);
-	bad = cvmx_read_csr(p->agl + AGL_GMX_RX_STATS_PKTS_BAD);
-
-	if (drop || bad) {
-		/* Do an atomic update. */
-		spin_lock_irqsave(&p->lock, flags);
-		netdev->stats.rx_errors += bad;
-		netdev->stats.rx_dropped += drop;
-		spin_unlock_irqrestore(&p->lock, flags);
-	}
-}
-
-static void octeon_mgmt_update_tx_stats(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	unsigned long flags;
-
-	union cvmx_agl_gmx_txx_stat0 s0;
-	union cvmx_agl_gmx_txx_stat1 s1;
-
-	/* These reads also clear the count registers.  */
-	s0.u64 = cvmx_read_csr(p->agl + AGL_GMX_TX_STAT0);
-	s1.u64 = cvmx_read_csr(p->agl + AGL_GMX_TX_STAT1);
-
-	if (s0.s.xsdef || s0.s.xscol || s1.s.scol || s1.s.mcol) {
-		/* Do an atomic update. */
-		spin_lock_irqsave(&p->lock, flags);
-		netdev->stats.tx_errors += s0.s.xsdef + s0.s.xscol;
-		netdev->stats.collisions += s1.s.scol + s1.s.mcol;
-		spin_unlock_irqrestore(&p->lock, flags);
-	}
-}
-
-/*
- * Dequeue a receive skb and its corresponding ring entry.  The ring
- * entry is returned, *pskb is updated to point to the skb.
- */
-static u64 octeon_mgmt_dequeue_rx_buffer(struct octeon_mgmt *p,
-					 struct sk_buff **pskb)
-{
-	union mgmt_port_ring_entry re;
-
-	dma_sync_single_for_cpu(p->dev, p->rx_ring_handle,
-				ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
-				DMA_BIDIRECTIONAL);
-
-	re.d64 = p->rx_ring[p->rx_next];
-	p->rx_next = (p->rx_next + 1) % OCTEON_MGMT_RX_RING_SIZE;
-	p->rx_current_fill--;
-	*pskb = __skb_dequeue(&p->rx_list);
-
-	dma_unmap_single(p->dev, re.s.addr,
-			 ETH_FRAME_LEN + OCTEON_MGMT_RX_HEADROOM,
-			 DMA_FROM_DEVICE);
-
-	return re.d64;
-}
-
-
-static int octeon_mgmt_receive_one(struct octeon_mgmt *p)
-{
-	struct net_device *netdev = p->netdev;
-	union cvmx_mixx_ircnt mix_ircnt;
-	union mgmt_port_ring_entry re;
-	struct sk_buff *skb;
-	struct sk_buff *skb2;
-	struct sk_buff *skb_new;
-	union mgmt_port_ring_entry re2;
-	int rc = 1;
-
-
-	re.d64 = octeon_mgmt_dequeue_rx_buffer(p, &skb);
-	if (likely(re.s.code == RING_ENTRY_CODE_DONE)) {
-		/* A good packet, send it up. */
-		skb_put(skb, re.s.len);
-good:
-		/* Process the RX timestamp if it was recorded */
-		if (p->has_rx_tstamp) {
-			/* The first 8 bytes are the timestamp */
-			u64 ns = *(u64 *)skb->data;
-			struct skb_shared_hwtstamps *ts;
-			ts = skb_hwtstamps(skb);
-			ts->hwtstamp = ns_to_ktime(ns);
-			__skb_pull(skb, 8);
-		}
-		skb->protocol = eth_type_trans(skb, netdev);
-		netdev->stats.rx_packets++;
-		netdev->stats.rx_bytes += skb->len;
-		netif_receive_skb(skb);
-		rc = 0;
-	} else if (re.s.code == RING_ENTRY_CODE_MORE) {
-		/* Packet split across skbs.  This can happen if we
-		 * increase the MTU.  Buffers that are already in the
-		 * rx ring can then end up being too small.  As the rx
-		 * ring is refilled, buffers sized for the new MTU
-		 * will be used and we should go back to the normal
-		 * non-split case.
-		 */
-		skb_put(skb, re.s.len);
-		do {
-			re2.d64 = octeon_mgmt_dequeue_rx_buffer(p, &skb2);
-			if (re2.s.code != RING_ENTRY_CODE_MORE
-				&& re2.s.code != RING_ENTRY_CODE_DONE)
-				goto split_error;
-			skb_put(skb2,  re2.s.len);
-			skb_new = skb_copy_expand(skb, 0, skb2->len,
-						  GFP_ATOMIC);
-			if (!skb_new)
-				goto split_error;
-			if (skb_copy_bits(skb2, 0, skb_tail_pointer(skb_new),
-					  skb2->len))
-				goto split_error;
-			skb_put(skb_new, skb2->len);
-			dev_kfree_skb_any(skb);
-			dev_kfree_skb_any(skb2);
-			skb = skb_new;
-		} while (re2.s.code == RING_ENTRY_CODE_MORE);
-		goto good;
-	} else {
-		/* Some other error, discard it. */
-		dev_kfree_skb_any(skb);
-		/* Error statistics are accumulated in
-		 * octeon_mgmt_update_rx_stats.
-		 */
-	}
-	goto done;
-split_error:
-	/* Discard the whole mess. */
-	dev_kfree_skb_any(skb);
-	dev_kfree_skb_any(skb2);
-	while (re2.s.code == RING_ENTRY_CODE_MORE) {
-		re2.d64 = octeon_mgmt_dequeue_rx_buffer(p, &skb2);
-		dev_kfree_skb_any(skb2);
-	}
-	netdev->stats.rx_errors++;
-
-done:
-	/* Tell the hardware we processed a packet.  */
-	mix_ircnt.u64 = 0;
-	mix_ircnt.s.ircnt = 1;
-	cvmx_write_csr(p->mix + MIX_IRCNT, mix_ircnt.u64);
-	return rc;
-}
-
-static int octeon_mgmt_receive_packets(struct octeon_mgmt *p, int budget)
-{
-	unsigned int work_done = 0;
-	union cvmx_mixx_ircnt mix_ircnt;
-	int rc;
-
-	mix_ircnt.u64 = cvmx_read_csr(p->mix + MIX_IRCNT);
-	while (work_done < budget && mix_ircnt.s.ircnt) {
-
-		rc = octeon_mgmt_receive_one(p);
-		if (!rc)
-			work_done++;
-
-		/* Check for more packets. */
-		mix_ircnt.u64 = cvmx_read_csr(p->mix + MIX_IRCNT);
-	}
-
-	octeon_mgmt_rx_fill_ring(p->netdev);
-
-	return work_done;
-}
-
-static int octeon_mgmt_napi_poll(struct napi_struct *napi, int budget)
-{
-	struct octeon_mgmt *p = container_of(napi, struct octeon_mgmt, napi);
-	struct net_device *netdev = p->netdev;
-	unsigned int work_done = 0;
-
-	work_done = octeon_mgmt_receive_packets(p, budget);
-
-	if (work_done < budget) {
-		/* We stopped because no more packets were available. */
-		napi_complete(napi);
-		octeon_mgmt_enable_rx_irq(p);
-	}
-	octeon_mgmt_update_rx_stats(netdev);
-
-	return work_done;
-}
-
-/* Reset the hardware to clean state.  */
-static void octeon_mgmt_reset_hw(struct octeon_mgmt *p)
-{
-	union cvmx_mixx_ctl mix_ctl;
-	union cvmx_mixx_bist mix_bist;
-	union cvmx_agl_gmx_bist agl_gmx_bist;
-
-	mix_ctl.u64 = 0;
-	cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
-	do {
-		mix_ctl.u64 = cvmx_read_csr(p->mix + MIX_CTL);
-	} while (mix_ctl.s.busy);
-	mix_ctl.s.reset = 1;
-	cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
-	cvmx_read_csr(p->mix + MIX_CTL);
-	octeon_io_clk_delay(64);
-
-	mix_bist.u64 = cvmx_read_csr(p->mix + MIX_BIST);
-	if (mix_bist.u64)
-		dev_warn(p->dev, "MIX failed BIST (0x%016llx)\n",
-			(unsigned long long)mix_bist.u64);
-
-	agl_gmx_bist.u64 = cvmx_read_csr(CVMX_AGL_GMX_BIST);
-	if (agl_gmx_bist.u64)
-		dev_warn(p->dev, "AGL failed BIST (0x%016llx)\n",
-			 (unsigned long long)agl_gmx_bist.u64);
-}
-
-struct octeon_mgmt_cam_state {
-	u64 cam[6];
-	u64 cam_mask;
-	int cam_index;
-};
-
-static void octeon_mgmt_cam_state_add(struct octeon_mgmt_cam_state *cs,
-				      unsigned char *addr)
-{
-	int i;
-
-	for (i = 0; i < 6; i++)
-		cs->cam[i] |= (u64)addr[i] << (8 * (cs->cam_index));
-	cs->cam_mask |= (1ULL << cs->cam_index);
-	cs->cam_index++;
-}
-
-static void octeon_mgmt_set_rx_filtering(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	union cvmx_agl_gmx_rxx_adr_ctl adr_ctl;
-	union cvmx_agl_gmx_prtx_cfg agl_gmx_prtx;
-	unsigned long flags;
-	unsigned int prev_packet_enable;
-	unsigned int cam_mode = 1; /* 1 - Accept on CAM match */
-	unsigned int multicast_mode = 1; /* 1 - Reject all multicast.  */
-	struct octeon_mgmt_cam_state cam_state;
-	struct netdev_hw_addr *ha;
-	int available_cam_entries;
-
-	memset(&cam_state, 0, sizeof(cam_state));
-
-	if ((netdev->flags & IFF_PROMISC) || netdev->uc.count > 7) {
-		cam_mode = 0;
-		available_cam_entries = 8;
-	} else {
-		/* One CAM entry for the primary address, leaves seven
-		 * for the secondary addresses.
-		 */
-		available_cam_entries = 7 - netdev->uc.count;
-	}
-
-	if (netdev->flags & IFF_MULTICAST) {
-		if (cam_mode == 0 || (netdev->flags & IFF_ALLMULTI) ||
-		    netdev_mc_count(netdev) > available_cam_entries)
-			multicast_mode = 2; /* 2 - Accept all multicast.  */
-		else
-			multicast_mode = 0; /* 0 - Use CAM.  */
-	}
-
-	if (cam_mode == 1) {
-		/* Add primary address. */
-		octeon_mgmt_cam_state_add(&cam_state, netdev->dev_addr);
-		netdev_for_each_uc_addr(ha, netdev)
-			octeon_mgmt_cam_state_add(&cam_state, ha->addr);
-	}
-	if (multicast_mode == 0) {
-		netdev_for_each_mc_addr(ha, netdev)
-			octeon_mgmt_cam_state_add(&cam_state, ha->addr);
-	}
-
-	spin_lock_irqsave(&p->lock, flags);
-
-	/* Disable packet I/O. */
-	agl_gmx_prtx.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
-	prev_packet_enable = agl_gmx_prtx.s.en;
-	agl_gmx_prtx.s.en = 0;
-	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, agl_gmx_prtx.u64);
-
-	adr_ctl.u64 = 0;
-	adr_ctl.s.cam_mode = cam_mode;
-	adr_ctl.s.mcst = multicast_mode;
-	adr_ctl.s.bcst = 1;     /* Allow broadcast */
-
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CTL, adr_ctl.u64);
-
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM0, cam_state.cam[0]);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM1, cam_state.cam[1]);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM2, cam_state.cam[2]);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM3, cam_state.cam[3]);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM4, cam_state.cam[4]);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM5, cam_state.cam[5]);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_ADR_CAM_EN, cam_state.cam_mask);
-
-	/* Restore packet I/O. */
-	agl_gmx_prtx.s.en = prev_packet_enable;
-	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, agl_gmx_prtx.u64);
-
-	spin_unlock_irqrestore(&p->lock, flags);
-}
-
-static int octeon_mgmt_set_mac_address(struct net_device *netdev, void *addr)
-{
-	int r = eth_mac_addr(netdev, addr);
-
-	if (r)
-		return r;
-
-	octeon_mgmt_set_rx_filtering(netdev);
-
-	return 0;
-}
-
-static int octeon_mgmt_change_mtu(struct net_device *netdev, int new_mtu)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	int size_without_fcs = new_mtu + OCTEON_MGMT_RX_HEADROOM;
-
-	/* Limit the MTU to make sure the ethernet packets are between
-	 * 64 bytes and 16383 bytes.
-	 */
-	if (size_without_fcs < 64 || size_without_fcs > 16383) {
-		dev_warn(p->dev, "MTU must be between %d and %d.\n",
-			 64 - OCTEON_MGMT_RX_HEADROOM,
-			 16383 - OCTEON_MGMT_RX_HEADROOM);
-		return -EINVAL;
-	}
-
-	netdev->mtu = new_mtu;
-
-	cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_MAX, size_without_fcs);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_JABBER,
-		       (size_without_fcs + 7) & 0xfff8);
-
-	return 0;
-}
-
-static irqreturn_t octeon_mgmt_interrupt(int cpl, void *dev_id)
-{
-	struct net_device *netdev = dev_id;
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	union cvmx_mixx_isr mixx_isr;
-
-	mixx_isr.u64 = cvmx_read_csr(p->mix + MIX_ISR);
-
-	/* Clear any pending interrupts */
-	cvmx_write_csr(p->mix + MIX_ISR, mixx_isr.u64);
-	cvmx_read_csr(p->mix + MIX_ISR);
-
-	if (mixx_isr.s.irthresh) {
-		octeon_mgmt_disable_rx_irq(p);
-		napi_schedule(&p->napi);
-	}
-	if (mixx_isr.s.orthresh) {
-		octeon_mgmt_disable_tx_irq(p);
-		tasklet_schedule(&p->tx_clean_tasklet);
-	}
-
-	return IRQ_HANDLED;
-}
-
-static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev,
-				      struct ifreq *rq, int cmd)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	struct hwtstamp_config config;
-	union cvmx_mio_ptp_clock_cfg ptp;
-	union cvmx_agl_gmx_rxx_frm_ctl rxx_frm_ctl;
-	bool have_hw_timestamps = false;
-
-	if (copy_from_user(&config, rq->ifr_data, sizeof(config)))
-		return -EFAULT;
-
-	if (config.flags) /* reserved for future extensions */
-		return -EINVAL;
-
-	/* Check the status of hardware for tiemstamps */
-	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
-		/* Get the current state of the PTP clock */
-		ptp.u64 = cvmx_read_csr(CVMX_MIO_PTP_CLOCK_CFG);
-		if (!ptp.s.ext_clk_en) {
-			/* The clock has not been configured to use an
-			 * external source.  Program it to use the main clock
-			 * reference.
-			 */
-			u64 clock_comp = (NSEC_PER_SEC << 32) /	octeon_get_io_clock_rate();
-			if (!ptp.s.ptp_en)
-				cvmx_write_csr(CVMX_MIO_PTP_CLOCK_COMP, clock_comp);
-			pr_info("PTP Clock: Using sclk reference at %lld Hz\n",
-				(NSEC_PER_SEC << 32) / clock_comp);
-		} else {
-			/* The clock is already programmed to use a GPIO */
-			u64 clock_comp = cvmx_read_csr(CVMX_MIO_PTP_CLOCK_COMP);
-			pr_info("PTP Clock: Using GPIO %d at %lld Hz\n",
-				ptp.s.ext_clk_in,
-				(NSEC_PER_SEC << 32) / clock_comp);
-		}
-
-		/* Enable the clock if it wasn't done already */
-		if (!ptp.s.ptp_en) {
-			ptp.s.ptp_en = 1;
-			cvmx_write_csr(CVMX_MIO_PTP_CLOCK_CFG, ptp.u64);
-		}
-		have_hw_timestamps = true;
-	}
-
-	if (!have_hw_timestamps)
-		return -EINVAL;
-
-	switch (config.tx_type) {
-	case HWTSTAMP_TX_OFF:
-	case HWTSTAMP_TX_ON:
-		break;
-	default:
-		return -ERANGE;
-	}
-
-	switch (config.rx_filter) {
-	case HWTSTAMP_FILTER_NONE:
-		p->has_rx_tstamp = false;
-		rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL);
-		rxx_frm_ctl.s.ptp_mode = 0;
-		cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64);
-		break;
-	case HWTSTAMP_FILTER_ALL:
-	case HWTSTAMP_FILTER_SOME:
-	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
-	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
-	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
-	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
-	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
-	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
-	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
-	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
-	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
-	case HWTSTAMP_FILTER_PTP_V2_EVENT:
-	case HWTSTAMP_FILTER_PTP_V2_SYNC:
-	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
-		p->has_rx_tstamp = have_hw_timestamps;
-		config.rx_filter = HWTSTAMP_FILTER_ALL;
-		if (p->has_rx_tstamp) {
-			rxx_frm_ctl.u64 = cvmx_read_csr(p->agl + AGL_GMX_RX_FRM_CTL);
-			rxx_frm_ctl.s.ptp_mode = 1;
-			cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64);
-		}
-		break;
-	default:
-		return -ERANGE;
-	}
-
-	if (copy_to_user(rq->ifr_data, &config, sizeof(config)))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int octeon_mgmt_ioctl(struct net_device *netdev,
-			     struct ifreq *rq, int cmd)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	switch (cmd) {
-	case SIOCSHWTSTAMP:
-		return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd);
-	default:
-		if (p->phydev)
-			return phy_mii_ioctl(p->phydev, rq, cmd);
-		return -EINVAL;
-	}
-}
-
-static void octeon_mgmt_disable_link(struct octeon_mgmt *p)
-{
-	union cvmx_agl_gmx_prtx_cfg prtx_cfg;
-
-	/* Disable GMX before we make any changes. */
-	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
-	prtx_cfg.s.en = 0;
-	prtx_cfg.s.tx_en = 0;
-	prtx_cfg.s.rx_en = 0;
-	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, prtx_cfg.u64);
-
-	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
-		int i;
-		for (i = 0; i < 10; i++) {
-			prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
-			if (prtx_cfg.s.tx_idle == 1 || prtx_cfg.s.rx_idle == 1)
-				break;
-			mdelay(1);
-			i++;
-		}
-	}
-}
-
-static void octeon_mgmt_enable_link(struct octeon_mgmt *p)
-{
-	union cvmx_agl_gmx_prtx_cfg prtx_cfg;
-
-	/* Restore the GMX enable state only if link is set */
-	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
-	prtx_cfg.s.tx_en = 1;
-	prtx_cfg.s.rx_en = 1;
-	prtx_cfg.s.en = 1;
-	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, prtx_cfg.u64);
-}
-
-static void octeon_mgmt_update_link(struct octeon_mgmt *p)
-{
-	union cvmx_agl_gmx_prtx_cfg prtx_cfg;
-
-	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
-
-	if (!p->phydev->link)
-		prtx_cfg.s.duplex = 1;
-	else
-		prtx_cfg.s.duplex = p->phydev->duplex;
-
-	switch (p->phydev->speed) {
-	case 10:
-		prtx_cfg.s.speed = 0;
-		prtx_cfg.s.slottime = 0;
-
-		if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
-			prtx_cfg.s.burst = 1;
-			prtx_cfg.s.speed_msb = 1;
-		}
-		break;
-	case 100:
-		prtx_cfg.s.speed = 0;
-		prtx_cfg.s.slottime = 0;
-
-		if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
-			prtx_cfg.s.burst = 1;
-			prtx_cfg.s.speed_msb = 0;
-		}
-		break;
-	case 1000:
-		/* 1000 MBits is only supported on 6XXX chips */
-		if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
-			prtx_cfg.s.speed = 1;
-			prtx_cfg.s.speed_msb = 0;
-			/* Only matters for half-duplex */
-			prtx_cfg.s.slottime = 1;
-			prtx_cfg.s.burst = p->phydev->duplex;
-		}
-		break;
-	case 0:  /* No link */
-	default:
-		break;
-	}
-
-	/* Write the new GMX setting with the port still disabled. */
-	cvmx_write_csr(p->agl + AGL_GMX_PRT_CFG, prtx_cfg.u64);
-
-	/* Read GMX CFG again to make sure the config is completed. */
-	prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG);
-
-	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
-		union cvmx_agl_gmx_txx_clk agl_clk;
-		union cvmx_agl_prtx_ctl prtx_ctl;
-
-		prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
-		agl_clk.u64 = cvmx_read_csr(p->agl + AGL_GMX_TX_CLK);
-		/* MII (both speeds) and RGMII 1000 speed. */
-		agl_clk.s.clk_cnt = 1;
-		if (prtx_ctl.s.mode == 0) { /* RGMII mode */
-			if (p->phydev->speed == 10)
-				agl_clk.s.clk_cnt = 50;
-			else if (p->phydev->speed == 100)
-				agl_clk.s.clk_cnt = 5;
-		}
-		cvmx_write_csr(p->agl + AGL_GMX_TX_CLK, agl_clk.u64);
-	}
-}
-
-static void octeon_mgmt_adjust_link(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	unsigned long flags;
-	int link_changed = 0;
-
-	if (!p->phydev)
-		return;
-
-	spin_lock_irqsave(&p->lock, flags);
-
-
-	if (!p->phydev->link && p->last_link)
-		link_changed = -1;
-
-	if (p->phydev->link
-	    && (p->last_duplex != p->phydev->duplex
-		|| p->last_link != p->phydev->link
-		|| p->last_speed != p->phydev->speed)) {
-		octeon_mgmt_disable_link(p);
-		link_changed = 1;
-		octeon_mgmt_update_link(p);
-		octeon_mgmt_enable_link(p);
-	}
-
-	p->last_link = p->phydev->link;
-	p->last_speed = p->phydev->speed;
-	p->last_duplex = p->phydev->duplex;
-
-	spin_unlock_irqrestore(&p->lock, flags);
-
-	if (link_changed != 0) {
-		if (link_changed > 0) {
-			pr_info("%s: Link is up - %d/%s\n", netdev->name,
-				p->phydev->speed,
-				DUPLEX_FULL == p->phydev->duplex ?
-				"Full" : "Half");
-		} else {
-			pr_info("%s: Link is down\n", netdev->name);
-		}
-	}
-}
-
-static int octeon_mgmt_init_phy(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	if (octeon_is_simulation() || p->phy_np == NULL) {
-		/* No PHYs in the simulator. */
-		netif_carrier_on(netdev);
-		return 0;
-	}
-
-	p->phydev = of_phy_connect(netdev, p->phy_np,
-				   octeon_mgmt_adjust_link, 0,
-				   PHY_INTERFACE_MODE_MII);
-
-	if (!p->phydev)
-		return -ENODEV;
-
-	return 0;
-}
-
-static int octeon_mgmt_open(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	union cvmx_mixx_ctl mix_ctl;
-	union cvmx_agl_gmx_inf_mode agl_gmx_inf_mode;
-	union cvmx_mixx_oring1 oring1;
-	union cvmx_mixx_iring1 iring1;
-	union cvmx_agl_gmx_rxx_frm_ctl rxx_frm_ctl;
-	union cvmx_mixx_irhwm mix_irhwm;
-	union cvmx_mixx_orhwm mix_orhwm;
-	union cvmx_mixx_intena mix_intena;
-	struct sockaddr sa;
-
-	/* Allocate ring buffers.  */
-	p->tx_ring = kzalloc(ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
-			     GFP_KERNEL);
-	if (!p->tx_ring)
-		return -ENOMEM;
-	p->tx_ring_handle =
-		dma_map_single(p->dev, p->tx_ring,
-			       ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
-			       DMA_BIDIRECTIONAL);
-	p->tx_next = 0;
-	p->tx_next_clean = 0;
-	p->tx_current_fill = 0;
-
-
-	p->rx_ring = kzalloc(ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
-			     GFP_KERNEL);
-	if (!p->rx_ring)
-		goto err_nomem;
-	p->rx_ring_handle =
-		dma_map_single(p->dev, p->rx_ring,
-			       ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
-			       DMA_BIDIRECTIONAL);
-
-	p->rx_next = 0;
-	p->rx_next_fill = 0;
-	p->rx_current_fill = 0;
-
-	octeon_mgmt_reset_hw(p);
-
-	mix_ctl.u64 = cvmx_read_csr(p->mix + MIX_CTL);
-
-	/* Bring it out of reset if needed. */
-	if (mix_ctl.s.reset) {
-		mix_ctl.s.reset = 0;
-		cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
-		do {
-			mix_ctl.u64 = cvmx_read_csr(p->mix + MIX_CTL);
-		} while (mix_ctl.s.reset);
-	}
-
-	if (OCTEON_IS_MODEL(OCTEON_CN5XXX)) {
-		agl_gmx_inf_mode.u64 = 0;
-		agl_gmx_inf_mode.s.en = 1;
-		cvmx_write_csr(CVMX_AGL_GMX_INF_MODE, agl_gmx_inf_mode.u64);
-	}
-	if (OCTEON_IS_MODEL(OCTEON_CN56XX_PASS1_X)
-		|| OCTEON_IS_MODEL(OCTEON_CN52XX_PASS1_X)) {
-		/* Force compensation values, as they are not
-		 * determined properly by HW
-		 */
-		union cvmx_agl_gmx_drv_ctl drv_ctl;
-
-		drv_ctl.u64 = cvmx_read_csr(CVMX_AGL_GMX_DRV_CTL);
-		if (p->port) {
-			drv_ctl.s.byp_en1 = 1;
-			drv_ctl.s.nctl1 = 6;
-			drv_ctl.s.pctl1 = 6;
-		} else {
-			drv_ctl.s.byp_en = 1;
-			drv_ctl.s.nctl = 6;
-			drv_ctl.s.pctl = 6;
-		}
-		cvmx_write_csr(CVMX_AGL_GMX_DRV_CTL, drv_ctl.u64);
-	}
-
-	oring1.u64 = 0;
-	oring1.s.obase = p->tx_ring_handle >> 3;
-	oring1.s.osize = OCTEON_MGMT_TX_RING_SIZE;
-	cvmx_write_csr(p->mix + MIX_ORING1, oring1.u64);
-
-	iring1.u64 = 0;
-	iring1.s.ibase = p->rx_ring_handle >> 3;
-	iring1.s.isize = OCTEON_MGMT_RX_RING_SIZE;
-	cvmx_write_csr(p->mix + MIX_IRING1, iring1.u64);
-
-	memcpy(sa.sa_data, netdev->dev_addr, ETH_ALEN);
-	octeon_mgmt_set_mac_address(netdev, &sa);
-
-	octeon_mgmt_change_mtu(netdev, netdev->mtu);
-
-	/* Enable the port HW. Packets are not allowed until
-	 * cvmx_mgmt_port_enable() is called.
-	 */
-	mix_ctl.u64 = 0;
-	mix_ctl.s.crc_strip = 1;    /* Strip the ending CRC */
-	mix_ctl.s.en = 1;           /* Enable the port */
-	mix_ctl.s.nbtarb = 0;       /* Arbitration mode */
-	/* MII CB-request FIFO programmable high watermark */
-	mix_ctl.s.mrq_hwm = 1;
-#ifdef __LITTLE_ENDIAN
-	mix_ctl.s.lendian = 1;
-#endif
-	cvmx_write_csr(p->mix + MIX_CTL, mix_ctl.u64);
-
-	/* Read the PHY to find the mode of the interface. */
-	if (octeon_mgmt_init_phy(netdev)) {
-		dev_err(p->dev, "Cannot initialize PHY on MIX%d.\n", p->port);
-		goto err_noirq;
-	}
-
-	/* Set the mode of the interface, RGMII/MII. */
-	if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && p->phydev) {
-		union cvmx_agl_prtx_ctl agl_prtx_ctl;
-		int rgmii_mode = (p->phydev->supported &
-				  (SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full)) != 0;
-
-		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
-		agl_prtx_ctl.s.mode = rgmii_mode ? 0 : 1;
-		cvmx_write_csr(p->agl_prt_ctl,	agl_prtx_ctl.u64);
-
-		/* MII clocks counts are based on the 125Mhz
-		 * reference, which has an 8nS period. So our delays
-		 * need to be multiplied by this factor.
-		 */
-#define NS_PER_PHY_CLK 8
-
-		/* Take the DLL and clock tree out of reset */
-		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
-		agl_prtx_ctl.s.clkrst = 0;
-		if (rgmii_mode) {
-			agl_prtx_ctl.s.dllrst = 0;
-			agl_prtx_ctl.s.clktx_byp = 0;
-		}
-		cvmx_write_csr(p->agl_prt_ctl,	agl_prtx_ctl.u64);
-		cvmx_read_csr(p->agl_prt_ctl); /* Force write out before wait */
-
-		/* Wait for the DLL to lock. External 125 MHz
-		 * reference clock must be stable at this point.
-		 */
-		ndelay(256 * NS_PER_PHY_CLK);
-
-		/* Enable the interface */
-		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
-		agl_prtx_ctl.s.enable = 1;
-		cvmx_write_csr(p->agl_prt_ctl, agl_prtx_ctl.u64);
-
-		/* Read the value back to force the previous write */
-		agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl);
-
-		/* Enable the compensation controller */
-		agl_prtx_ctl.s.comp = 1;
-		agl_prtx_ctl.s.drv_byp = 0;
-		cvmx_write_csr(p->agl_prt_ctl,	agl_prtx_ctl.u64);
-		/* Force write out before wait. */
-		cvmx_read_csr(p->agl_prt_ctl);
-
-		/* For compensation state to lock. */
-		ndelay(1040 * NS_PER_PHY_CLK);
-
-		/* Default Interframe Gaps are too small.  Recommended
-		 * workaround is.
-		 *
-		 * AGL_GMX_TX_IFG[IFG1]=14
-		 * AGL_GMX_TX_IFG[IFG2]=10
-		 */
-		cvmx_write_csr(CVMX_AGL_GMX_TX_IFG, 0xae);
-	}
-
-	octeon_mgmt_rx_fill_ring(netdev);
-
-	/* Clear statistics. */
-	/* Clear on read. */
-	cvmx_write_csr(p->agl + AGL_GMX_RX_STATS_CTL, 1);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_STATS_PKTS_DRP, 0);
-	cvmx_write_csr(p->agl + AGL_GMX_RX_STATS_PKTS_BAD, 0);
-
-	cvmx_write_csr(p->agl + AGL_GMX_TX_STATS_CTL, 1);
-	cvmx_write_csr(p->agl + AGL_GMX_TX_STAT0, 0);
-	cvmx_write_csr(p->agl + AGL_GMX_TX_STAT1, 0);
-
-	/* Clear any pending interrupts */
-	cvmx_write_csr(p->mix + MIX_ISR, cvmx_read_csr(p->mix + MIX_ISR));
-
-	if (request_irq(p->irq, octeon_mgmt_interrupt, 0, netdev->name,
-			netdev)) {
-		dev_err(p->dev, "request_irq(%d) failed.\n", p->irq);
-		goto err_noirq;
-	}
-
-	/* Interrupt every single RX packet */
-	mix_irhwm.u64 = 0;
-	mix_irhwm.s.irhwm = 0;
-	cvmx_write_csr(p->mix + MIX_IRHWM, mix_irhwm.u64);
-
-	/* Interrupt when we have 1 or more packets to clean.  */
-	mix_orhwm.u64 = 0;
-	mix_orhwm.s.orhwm = 0;
-	cvmx_write_csr(p->mix + MIX_ORHWM, mix_orhwm.u64);
-
-	/* Enable receive and transmit interrupts */
-	mix_intena.u64 = 0;
-	mix_intena.s.ithena = 1;
-	mix_intena.s.othena = 1;
-	cvmx_write_csr(p->mix + MIX_INTENA, mix_intena.u64);
-
-	/* Enable packet I/O. */
-
-	rxx_frm_ctl.u64 = 0;
-	rxx_frm_ctl.s.ptp_mode = p->has_rx_tstamp ? 1 : 0;
-	rxx_frm_ctl.s.pre_align = 1;
-	/* When set, disables the length check for non-min sized pkts
-	 * with padding in the client data.
-	 */
-	rxx_frm_ctl.s.pad_len = 1;
-	/* When set, disables the length check for VLAN pkts */
-	rxx_frm_ctl.s.vlan_len = 1;
-	/* When set, PREAMBLE checking is  less strict */
-	rxx_frm_ctl.s.pre_free = 1;
-	/* Control Pause Frames can match station SMAC */
-	rxx_frm_ctl.s.ctl_smac = 0;
-	/* Control Pause Frames can match globally assign Multicast address */
-	rxx_frm_ctl.s.ctl_mcst = 1;
-	/* Forward pause information to TX block */
-	rxx_frm_ctl.s.ctl_bck = 1;
-	/* Drop Control Pause Frames */
-	rxx_frm_ctl.s.ctl_drp = 1;
-	/* Strip off the preamble */
-	rxx_frm_ctl.s.pre_strp = 1;
-	/* This port is configured to send PREAMBLE+SFD to begin every
-	 * frame.  GMX checks that the PREAMBLE is sent correctly.
-	 */
-	rxx_frm_ctl.s.pre_chk = 1;
-	cvmx_write_csr(p->agl + AGL_GMX_RX_FRM_CTL, rxx_frm_ctl.u64);
-
-	/* Configure the port duplex, speed and enables */
-	octeon_mgmt_disable_link(p);
-	if (p->phydev)
-		octeon_mgmt_update_link(p);
-	octeon_mgmt_enable_link(p);
-
-	p->last_link = 0;
-	p->last_speed = 0;
-	/* PHY is not present in simulator. The carrier is enabled
-	 * while initializing the phy for simulator, leave it enabled.
-	 */
-	if (p->phydev) {
-		netif_carrier_off(netdev);
-		phy_start_aneg(p->phydev);
-	}
-
-	netif_wake_queue(netdev);
-	napi_enable(&p->napi);
-
-	return 0;
-err_noirq:
-	octeon_mgmt_reset_hw(p);
-	dma_unmap_single(p->dev, p->rx_ring_handle,
-			 ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
-			 DMA_BIDIRECTIONAL);
-	kfree(p->rx_ring);
-err_nomem:
-	dma_unmap_single(p->dev, p->tx_ring_handle,
-			 ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
-			 DMA_BIDIRECTIONAL);
-	kfree(p->tx_ring);
-	return -ENOMEM;
-}
-
-static int octeon_mgmt_stop(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	napi_disable(&p->napi);
-	netif_stop_queue(netdev);
-
-	if (p->phydev)
-		phy_disconnect(p->phydev);
-	p->phydev = NULL;
-
-	netif_carrier_off(netdev);
-
-	octeon_mgmt_reset_hw(p);
-
-	free_irq(p->irq, netdev);
-
-	/* dma_unmap is a nop on Octeon, so just free everything.  */
-	skb_queue_purge(&p->tx_list);
-	skb_queue_purge(&p->rx_list);
-
-	dma_unmap_single(p->dev, p->rx_ring_handle,
-			 ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE),
-			 DMA_BIDIRECTIONAL);
-	kfree(p->rx_ring);
-
-	dma_unmap_single(p->dev, p->tx_ring_handle,
-			 ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
-			 DMA_BIDIRECTIONAL);
-	kfree(p->tx_ring);
-
-	return 0;
-}
-
-static int octeon_mgmt_xmit(struct sk_buff *skb, struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-	union mgmt_port_ring_entry re;
-	unsigned long flags;
-	int rv = NETDEV_TX_BUSY;
-
-	re.d64 = 0;
-	re.s.tstamp = ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) != 0);
-	re.s.len = skb->len;
-	re.s.addr = dma_map_single(p->dev, skb->data,
-				   skb->len,
-				   DMA_TO_DEVICE);
-
-	spin_lock_irqsave(&p->tx_list.lock, flags);
-
-	if (unlikely(p->tx_current_fill >= ring_max_fill(OCTEON_MGMT_TX_RING_SIZE) - 1)) {
-		spin_unlock_irqrestore(&p->tx_list.lock, flags);
-		netif_stop_queue(netdev);
-		spin_lock_irqsave(&p->tx_list.lock, flags);
-	}
-
-	if (unlikely(p->tx_current_fill >=
-		     ring_max_fill(OCTEON_MGMT_TX_RING_SIZE))) {
-		spin_unlock_irqrestore(&p->tx_list.lock, flags);
-		dma_unmap_single(p->dev, re.s.addr, re.s.len,
-				 DMA_TO_DEVICE);
-		goto out;
-	}
-
-	__skb_queue_tail(&p->tx_list, skb);
-
-	/* Put it in the ring.  */
-	p->tx_ring[p->tx_next] = re.d64;
-	p->tx_next = (p->tx_next + 1) % OCTEON_MGMT_TX_RING_SIZE;
-	p->tx_current_fill++;
-
-	spin_unlock_irqrestore(&p->tx_list.lock, flags);
-
-	dma_sync_single_for_device(p->dev, p->tx_ring_handle,
-				   ring_size_to_bytes(OCTEON_MGMT_TX_RING_SIZE),
-				   DMA_BIDIRECTIONAL);
-
-	netdev->stats.tx_packets++;
-	netdev->stats.tx_bytes += skb->len;
-
-	/* Ring the bell.  */
-	cvmx_write_csr(p->mix + MIX_ORING2, 1);
-
-	netdev->trans_start = jiffies;
-	rv = NETDEV_TX_OK;
-out:
-	octeon_mgmt_update_tx_stats(netdev);
-	return rv;
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void octeon_mgmt_poll_controller(struct net_device *netdev)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	octeon_mgmt_receive_packets(p, 16);
-	octeon_mgmt_update_rx_stats(netdev);
-}
-#endif
-
-static void octeon_mgmt_get_drvinfo(struct net_device *netdev,
-				    struct ethtool_drvinfo *info)
-{
-	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
-	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
-	strlcpy(info->bus_info, "N/A", sizeof(info->bus_info));
-}
-
-static int octeon_mgmt_get_settings(struct net_device *netdev,
-				    struct ethtool_cmd *cmd)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	if (p->phydev)
-		return phy_ethtool_gset(p->phydev, cmd);
-
-	return -EOPNOTSUPP;
-}
-
-static int octeon_mgmt_set_settings(struct net_device *netdev,
-				    struct ethtool_cmd *cmd)
-{
-	struct octeon_mgmt *p = netdev_priv(netdev);
-
-	if (!capable(CAP_NET_ADMIN))
-		return -EPERM;
-
-	if (p->phydev)
-		return phy_ethtool_sset(p->phydev, cmd);
-
-	return -EOPNOTSUPP;
-}
-
-static int octeon_mgmt_nway_reset(struct net_device *dev)
-{
-	struct octeon_mgmt *p = netdev_priv(dev);
-
-	if (!capable(CAP_NET_ADMIN))
-		return -EPERM;
-
-	if (p->phydev)
-		return phy_start_aneg(p->phydev);
-
-	return -EOPNOTSUPP;
-}
-
-static const struct ethtool_ops octeon_mgmt_ethtool_ops = {
-	.get_drvinfo = octeon_mgmt_get_drvinfo,
-	.get_settings = octeon_mgmt_get_settings,
-	.set_settings = octeon_mgmt_set_settings,
-	.nway_reset = octeon_mgmt_nway_reset,
-	.get_link = ethtool_op_get_link,
-};
-
-static const struct net_device_ops octeon_mgmt_ops = {
-	.ndo_open =			octeon_mgmt_open,
-	.ndo_stop =			octeon_mgmt_stop,
-	.ndo_start_xmit =		octeon_mgmt_xmit,
-	.ndo_set_rx_mode =		octeon_mgmt_set_rx_filtering,
-	.ndo_set_mac_address =		octeon_mgmt_set_mac_address,
-	.ndo_do_ioctl =			octeon_mgmt_ioctl,
-	.ndo_change_mtu =		octeon_mgmt_change_mtu,
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller =		octeon_mgmt_poll_controller,
-#endif
-};
-
-static int octeon_mgmt_probe(struct platform_device *pdev)
-{
-	struct net_device *netdev;
-	struct octeon_mgmt *p;
-	const __be32 *data;
-	const u8 *mac;
-	struct resource *res_mix;
-	struct resource *res_agl;
-	struct resource *res_agl_prt_ctl;
-	int len;
-	int result;
-
-	netdev = alloc_etherdev(sizeof(struct octeon_mgmt));
-	if (netdev == NULL)
-		return -ENOMEM;
-
-	SET_NETDEV_DEV(netdev, &pdev->dev);
-
-	platform_set_drvdata(pdev, netdev);
-	p = netdev_priv(netdev);
-	netif_napi_add(netdev, &p->napi, octeon_mgmt_napi_poll,
-		       OCTEON_MGMT_NAPI_WEIGHT);
-
-	p->netdev = netdev;
-	p->dev = &pdev->dev;
-	p->has_rx_tstamp = false;
-
-	data = of_get_property(pdev->dev.of_node, "cell-index", &len);
-	if (data && len == sizeof(*data)) {
-		p->port = be32_to_cpup(data);
-	} else {
-		dev_err(&pdev->dev, "no 'cell-index' property\n");
-		result = -ENXIO;
-		goto err;
-	}
-
-	snprintf(netdev->name, IFNAMSIZ, "mgmt%d", p->port);
-
-	result = platform_get_irq(pdev, 0);
-	if (result < 0)
-		goto err;
-
-	p->irq = result;
-
-	res_mix = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (res_mix == NULL) {
-		dev_err(&pdev->dev, "no 'reg' resource\n");
-		result = -ENXIO;
-		goto err;
-	}
-
-	res_agl = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	if (res_agl == NULL) {
-		dev_err(&pdev->dev, "no 'reg' resource\n");
-		result = -ENXIO;
-		goto err;
-	}
-
-	res_agl_prt_ctl = platform_get_resource(pdev, IORESOURCE_MEM, 3);
-	if (res_agl_prt_ctl == NULL) {
-		dev_err(&pdev->dev, "no 'reg' resource\n");
-		result = -ENXIO;
-		goto err;
-	}
-
-	p->mix_phys = res_mix->start;
-	p->mix_size = resource_size(res_mix);
-	p->agl_phys = res_agl->start;
-	p->agl_size = resource_size(res_agl);
-	p->agl_prt_ctl_phys = res_agl_prt_ctl->start;
-	p->agl_prt_ctl_size = resource_size(res_agl_prt_ctl);
-
-
-	if (!devm_request_mem_region(&pdev->dev, p->mix_phys, p->mix_size,
-				     res_mix->name)) {
-		dev_err(&pdev->dev, "request_mem_region (%s) failed\n",
-			res_mix->name);
-		result = -ENXIO;
-		goto err;
-	}
-
-	if (!devm_request_mem_region(&pdev->dev, p->agl_phys, p->agl_size,
-				     res_agl->name)) {
-		result = -ENXIO;
-		dev_err(&pdev->dev, "request_mem_region (%s) failed\n",
-			res_agl->name);
-		goto err;
-	}
-
-	if (!devm_request_mem_region(&pdev->dev, p->agl_prt_ctl_phys,
-				     p->agl_prt_ctl_size, res_agl_prt_ctl->name)) {
-		result = -ENXIO;
-		dev_err(&pdev->dev, "request_mem_region (%s) failed\n",
-			res_agl_prt_ctl->name);
-		goto err;
-	}
-
-	p->mix = (u64)devm_ioremap(&pdev->dev, p->mix_phys, p->mix_size);
-	p->agl = (u64)devm_ioremap(&pdev->dev, p->agl_phys, p->agl_size);
-	p->agl_prt_ctl = (u64)devm_ioremap(&pdev->dev, p->agl_prt_ctl_phys,
-					   p->agl_prt_ctl_size);
-	spin_lock_init(&p->lock);
-
-	skb_queue_head_init(&p->tx_list);
-	skb_queue_head_init(&p->rx_list);
-	tasklet_init(&p->tx_clean_tasklet,
-		     octeon_mgmt_clean_tx_tasklet, (unsigned long)p);
-
-	netdev->priv_flags |= IFF_UNICAST_FLT;
-
-	netdev->netdev_ops = &octeon_mgmt_ops;
-	netdev->ethtool_ops = &octeon_mgmt_ethtool_ops;
-
-	mac = of_get_mac_address(pdev->dev.of_node);
-
-	if (mac)
-		memcpy(netdev->dev_addr, mac, ETH_ALEN);
-	else
-		eth_hw_addr_random(netdev);
-
-	p->phy_np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
-
-	result = dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
-	if (result)
-		goto err;
-
-	netif_carrier_off(netdev);
-	result = register_netdev(netdev);
-	if (result)
-		goto err;
-
-	dev_info(&pdev->dev, "Version " DRV_VERSION "\n");
-	return 0;
-
-err:
-	free_netdev(netdev);
-	return result;
-}
-
-static int octeon_mgmt_remove(struct platform_device *pdev)
-{
-	struct net_device *netdev = platform_get_drvdata(pdev);
-
-	unregister_netdev(netdev);
-	free_netdev(netdev);
-	return 0;
-}
-
-static const struct of_device_id octeon_mgmt_match[] = {
-	{
-		.compatible = "cavium,octeon-5750-mix",
-	},
-	{},
-};
-MODULE_DEVICE_TABLE(of, octeon_mgmt_match);
-
-static struct platform_driver octeon_mgmt_driver = {
-	.driver = {
-		.name		= "octeon_mgmt",
-		.of_match_table = octeon_mgmt_match,
-	},
-	.probe		= octeon_mgmt_probe,
-	.remove		= octeon_mgmt_remove,
-};
-
-extern void octeon_mdiobus_force_mod_depencency(void);
-
-static int __init octeon_mgmt_mod_init(void)
-{
-	/* Force our mdiobus driver module to be loaded first. */
-	octeon_mdiobus_force_mod_depencency();
-	return platform_driver_register(&octeon_mgmt_driver);
-}
-
-static void __exit octeon_mgmt_mod_exit(void)
-{
-	platform_driver_unregister(&octeon_mgmt_driver);
-}
-
-module_init(octeon_mgmt_mod_init);
-module_exit(octeon_mgmt_mod_exit);
-
-MODULE_DESCRIPTION(DRV_DESCRIPTION);
-MODULE_AUTHOR("David Daney");
-MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
-- 
cgit v0.10.2


From fe3a5f6c795810edb1646a840fec3c8c350c2a4e Mon Sep 17 00:00:00 2001
From: Simon Horman <simon.horman@netronome.com>
Date: Tue, 15 Mar 2016 12:00:35 +0900
Subject: openvswitch: allow output of MPLS packets on tunnel vports

Currently output of MPLS packets on tunnel vports is not allowed by Open
vSwitch. This is because historically encapsulation was done in such a way
that the inner_protocol field of the skb needed to hold the inner protocol
for both MPLS and tunnel encapsulation in order for GSO segmentation to be
performed correctly.

Since b2acd1dc3949 ("openvswitch: Use regular GRE net_device instead of
vport") Open vSwitch makes use of lwt to output to tunnel netdevs which
perform encapsulation. As no drivers expose support for MPLS offloads this
means that GSO packets are segmented in software by validate_xmit_skb(),
which is called from __dev_queue_xmit(), before tunnel encapsulation occurs.
This means that the inner protocol of MPLS is no longer needed by the time
encapsulation occurs and the contention on the inner_protocol field of the
skb no longer occurs.

Thus it is now safe to output MPLS to tunnel vports.

Signed-off-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jesse Gross <jesse@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index 58b8efc..689c172 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -2044,9 +2044,6 @@ static int validate_set(const struct nlattr *a,
 		break;
 
 	case OVS_KEY_ATTR_TUNNEL:
-		if (eth_p_mpls(eth_type))
-			return -EINVAL;
-
 		if (masked)
 			return -EINVAL; /* Masked tunnel set not supported. */
 
-- 
cgit v0.10.2


From 977bc20cf66367a530d17696f7e1783510a115b7 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 15 Mar 2016 10:18:49 +0300
Subject: mediatek: checking for IS_ERR() instead of NULL

of_phy_connect() returns NULL on error, it never returns error pointers.

Fixes: 656e705243fd ('net-next: mediatek: add support for MT7623 ethernet')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: John Crispin <blogic@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 1e6c549..fd19d2d 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -186,9 +186,9 @@ static int mtk_phy_connect_node(struct mtk_eth *eth, struct mtk_mac *mac,
 
 	phydev = of_phy_connect(eth->netdev[mac->id], phy_node,
 				mtk_phy_link_adjust, 0, phy_mode);
-	if (IS_ERR(phydev)) {
+	if (!phydev) {
 		dev_err(eth->dev, "could not connect to PHY\n");
-		return PTR_ERR(phydev);
+		return -ENODEV;
 	}
 
 	dev_info(eth->dev,
-- 
cgit v0.10.2


From 48e77422a3cb74e04384090e21235f4916278373 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 15 Mar 2016 10:19:04 +0300
Subject: mediatek: unlock on error in mtk_tx_map()

There was a missing unlock on the error path.

Fixes: 656e705243fd ('net-next: mediatek: add support for MT7623 ethernet')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: John Crispin <blogic@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index fd19d2d..7f2126b 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -661,6 +661,8 @@ err_dma:
 		itxd = mtk_qdma_phys_to_virt(ring, itxd->txd2);
 	} while (itxd != txd);
 
+	spin_unlock_irqrestore(&eth->page_lock, flags);
+
 	return -ENOMEM;
 }
 
-- 
cgit v0.10.2


From cc2fa619a738a052eb90ccbbbc48947a2e2aa1d3 Mon Sep 17 00:00:00 2001
From: Phil Reid <preid@electromag.com.au>
Date: Tue, 15 Mar 2016 15:34:33 +0800
Subject: net: stmmac: Don't search for phys if mdio node is defined.

If a dt mdio entry has been added least assume that we wont
search for phys attached. The DT and of_mdiobus_register already do
this. This stops DSA phys being found and phys created for them, as
this is handled by the DSA driver.

Signed-off-by: Phil Reid <preid@electromag.com.au>
Acked-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index efb54f3..ea76129 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -243,6 +243,9 @@ int stmmac_mdio_register(struct net_device *ndev)
 		goto bus_register_fail;
 	}
 
+	if (priv->plat->phy_node || mdio_node)
+		goto bus_register_done;
+
 	found = 0;
 	for (addr = 0; addr < PHY_MAX_ADDR; addr++) {
 		struct phy_device *phydev = mdiobus_get_phy(new_bus, addr);
@@ -298,6 +301,7 @@ int stmmac_mdio_register(struct net_device *ndev)
 		return -ENODEV;
 	}
 
+bus_register_done:
 	priv->mii = new_bus;
 
 	return 0;
-- 
cgit v0.10.2


From 2c9a266afefe137bff06bbe0fc48b4d3b3cb348c Mon Sep 17 00:00:00 2001
From: Manish Chopra <manish.chopra@qlogic.com>
Date: Tue, 15 Mar 2016 07:13:45 -0400
Subject: qlge: Fix receive packets drop.

When running small packets [length < 256 bytes] traffic, packets were
being dropped due to invalid data in those packets which were
delivered by the driver upto the stack. Using pci_dma_sync_single_for_cpu
ensures copying latest and updated data into skb from the receive buffer.

Signed-off-by: Sony Chacko <sony.chacko@qlogic.com>
Signed-off-by: Manish Chopra <manish.chopra@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_main.c b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
index 9979764..b28e73e 100644
--- a/drivers/net/ethernet/qlogic/qlge/qlge_main.c
+++ b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
@@ -1648,7 +1648,18 @@ static void ql_process_mac_rx_skb(struct ql_adapter *qdev,
 		return;
 	}
 	skb_reserve(new_skb, NET_IP_ALIGN);
+
+	pci_dma_sync_single_for_cpu(qdev->pdev,
+				    dma_unmap_addr(sbq_desc, mapaddr),
+				    dma_unmap_len(sbq_desc, maplen),
+				    PCI_DMA_FROMDEVICE);
+
 	memcpy(skb_put(new_skb, length), skb->data, length);
+
+	pci_dma_sync_single_for_device(qdev->pdev,
+				       dma_unmap_addr(sbq_desc, mapaddr),
+				       dma_unmap_len(sbq_desc, maplen),
+				       PCI_DMA_FROMDEVICE);
 	skb = new_skb;
 
 	/* Frame error, so drop the packet. */
-- 
cgit v0.10.2


From 38c848c73180fb1d142a39e664b6d9663751f6b2 Mon Sep 17 00:00:00 2001
From: Yoshihiro Kaneko <ykaneko0929@gmail.com>
Date: Wed, 16 Mar 2016 00:52:16 +0900
Subject: ravb: fix result value overwrite

The result value is overwritten by a return value of
ravb_ptp_interrupt().

Signed-off-by: Yoshihiro Kaneko <ykaneko0929@gmail.com>
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 8f2c4fb..4e1a7db 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -757,8 +757,8 @@ static irqreturn_t ravb_interrupt(int irq, void *dev_id)
 		result = IRQ_HANDLED;
 	}
 
-	if (iss & ISS_CGIS)
-		result = ravb_ptp_interrupt(ndev);
+	if ((iss & ISS_CGIS) && ravb_ptp_interrupt(ndev) == IRQ_HANDLED)
+		result = IRQ_HANDLED;
 
 	mmiowb();
 	spin_unlock(&priv->lock);
-- 
cgit v0.10.2


From 1e6bb1a3540fec3ef112b9a89dda88e684c3ff59 Mon Sep 17 00:00:00 2001
From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Tue, 15 Mar 2016 19:52:04 +0200
Subject: bnx2x: Prevent false warning for lack of FC NPIV

Not all adapters have FC-NPIV configured. If bnx2fc is used with such an
adapter, driver would read irrelevant data from the the nvram and log
"FC-NPIV table with bad length..." In system logs.

Simply accept that reading '0' as the feature offset in nvram indicates
the feature isn't there and return.

Reported-by: Andrew Patterson <andrew.patterson@hpe.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index b597c32..d465bd7 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -14885,6 +14885,10 @@ static int bnx2x_get_fc_npiv(struct net_device *dev,
 	}
 
 	offset = SHMEM2_RD(bp, fc_npiv_nvram_tbl_addr[BP_PORT(bp)]);
+	if (!offset) {
+		DP(BNX2X_MSG_MCP, "No FC-NPIV in NVRAM\n");
+		goto out;
+	}
 	DP(BNX2X_MSG_MCP, "Offset of FC-NPIV in NVRAM: %08x\n", offset);
 
 	/* Read the table contents from nvram */
-- 
cgit v0.10.2


From 31762eaa0d0804d34e297daad57cda45cbc6c961 Mon Sep 17 00:00:00 2001
From: Aaron Young <aaron.young@oracle.com>
Date: Tue, 15 Mar 2016 11:35:37 -0700
Subject: ldmvsw: Split sunvnet driver into common code

  Split sunvnet.c into sunvnet.c and sunvnet_common.c.

  Details:

  Since the sunvnet and ldmvsw drivers will both use common sunvnet code,
  move the functions (and support functions) anticipated to be common code
  from sunvnet.c to sunvnet_common.c. Similarly, sunvnet.h was renamed to
  sunvnet_common.h. The sunvnet_common.c code will be compiled into the
  kernel and act as a library of functions that are linked by either
  (or both) drivers when loaded.

  Function names for external functions in sunvnet_common.c (to be
  called by both the sunvnet and ldmvsw drivers) were tagged with a "_common"
  suffix to clearly designate them as common functions.

  No functional changes as of yet... just moved code verbatim to the new
  sunvnet_common.c/h files.

  Makefile/Kconfig support added to build sunvnet_common.c file. The code
  is included in the kernel if SUN_LDOMS is defined/selected.

  NOTE - per the SubmittingPatches documentation, since the code was just
  moved from one file another, the code was NOT checkpatch'd in this commit
  to aid in review.

  Signed-off-by: Aaron Young <aaron.young@oracle.com>
  Signed-off-by: Rashmi Narasimhan <rashmi.narasimhan@oracle.com>
  Reviewed-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
  Reviewed-by: Alexandre Chartre <Alexandre.Chartre@oracle.com>

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/sun/Kconfig b/drivers/net/ethernet/sun/Kconfig
index dee94b6..aa58c11e 100644
--- a/drivers/net/ethernet/sun/Kconfig
+++ b/drivers/net/ethernet/sun/Kconfig
@@ -69,6 +69,11 @@ config CASSINI
 	  Support for the Sun Cassini chip, aka Sun GigaSwift Ethernet. See also
 	  <http://docs.oracle.com/cd/E19113-01/giga.ether.pci/817-4341-10/817-4341-10.pdf>.
 
+config SUNVNET_COMMON
+	bool
+	depends on SUN_LDOMS
+	default y if SUN_LDOMS
+
 config SUNVNET
 	tristate "Sun Virtual Network support"
 	depends on SUN_LDOMS
diff --git a/drivers/net/ethernet/sun/Makefile b/drivers/net/ethernet/sun/Makefile
index 1e620ff..7b622aa 100644
--- a/drivers/net/ethernet/sun/Makefile
+++ b/drivers/net/ethernet/sun/Makefile
@@ -7,5 +7,6 @@ obj-$(CONFIG_SUNQE) += sunqe.o
 obj-$(CONFIG_SUNBMAC) += sunbmac.o
 obj-$(CONFIG_SUNGEM) += sungem.o
 obj-$(CONFIG_CASSINI) += cassini.o
+obj-$(CONFIG_SUNVNET_COMMON) += sunvnet_common.o
 obj-$(CONFIG_SUNVNET) += sunvnet.o
 obj-$(CONFIG_NIU) += niu.o
diff --git a/drivers/net/ethernet/sun/sunvnet.c b/drivers/net/ethernet/sun/sunvnet.c
index 942a95d..5b911352 100644
--- a/drivers/net/ethernet/sun/sunvnet.c
+++ b/drivers/net/ethernet/sun/sunvnet.c
@@ -17,8 +17,6 @@
 #include <linux/mutex.h>
 #include <linux/highmem.h>
 #include <linux/if_vlan.h>
-#define CREATE_TRACE_POINTS
-#include <trace/events/sunvnet.h>
 
 #if IS_ENABLED(CONFIG_IPV6)
 #include <linux/icmpv6.h>
@@ -31,7 +29,12 @@
 #include <asm/vio.h>
 #include <asm/ldc.h>
 
-#include "sunvnet.h"
+#include "sunvnet_common.h"
+
+/* length of time before we decide the hardware is borked,
+ * and dev->tx_timeout() should be called to fix the problem
+ */
+#define VNET_TX_TIMEOUT			(5 * HZ)
 
 #define DRV_MODULE_NAME		"sunvnet"
 #define DRV_MODULE_VERSION	"1.0"
@@ -44,16 +47,6 @@ MODULE_DESCRIPTION("Sun LDOM virtual network driver");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_MODULE_VERSION);
 
-#define	VNET_MAX_TXQS		16
-
-/* Heuristic for the number of times to exponentially backoff and
- * retry sending an LDC trigger when EAGAIN is encountered
- */
-#define	VNET_MAX_RETRIES	10
-
-static int __vnet_tx_trigger(struct vnet_port *port, u32 start);
-static void vnet_port_reset(struct vnet_port *port);
-
 /* Ordered from largest major to lowest */
 static struct vio_version vnet_versions[] = {
 	{ .major = 1, .minor = 8 },
@@ -62,1578 +55,6 @@ static struct vio_version vnet_versions[] = {
 	{ .major = 1, .minor = 0 },
 };
 
-static inline u32 vnet_tx_dring_avail(struct vio_dring_state *dr)
-{
-	return vio_dring_avail(dr, VNET_TX_RING_SIZE);
-}
-
-static int vnet_handle_unknown(struct vnet_port *port, void *arg)
-{
-	struct vio_msg_tag *pkt = arg;
-
-	pr_err("Received unknown msg [%02x:%02x:%04x:%08x]\n",
-	       pkt->type, pkt->stype, pkt->stype_env, pkt->sid);
-	pr_err("Resetting connection\n");
-
-	ldc_disconnect(port->vio.lp);
-
-	return -ECONNRESET;
-}
-
-static int vnet_port_alloc_tx_ring(struct vnet_port *port);
-
-static int vnet_send_attr(struct vio_driver_state *vio)
-{
-	struct vnet_port *port = to_vnet_port(vio);
-	struct net_device *dev = port->vp->dev;
-	struct vio_net_attr_info pkt;
-	int framelen = ETH_FRAME_LEN;
-	int i, err;
-
-	err = vnet_port_alloc_tx_ring(to_vnet_port(vio));
-	if (err)
-		return err;
-
-	memset(&pkt, 0, sizeof(pkt));
-	pkt.tag.type = VIO_TYPE_CTRL;
-	pkt.tag.stype = VIO_SUBTYPE_INFO;
-	pkt.tag.stype_env = VIO_ATTR_INFO;
-	pkt.tag.sid = vio_send_sid(vio);
-	if (vio_version_before(vio, 1, 2))
-		pkt.xfer_mode = VIO_DRING_MODE;
-	else
-		pkt.xfer_mode = VIO_NEW_DRING_MODE;
-	pkt.addr_type = VNET_ADDR_ETHERMAC;
-	pkt.ack_freq = 0;
-	for (i = 0; i < 6; i++)
-		pkt.addr |= (u64)dev->dev_addr[i] << ((5 - i) * 8);
-	if (vio_version_after(vio, 1, 3)) {
-		if (port->rmtu) {
-			port->rmtu = min(VNET_MAXPACKET, port->rmtu);
-			pkt.mtu = port->rmtu;
-		} else {
-			port->rmtu = VNET_MAXPACKET;
-			pkt.mtu = port->rmtu;
-		}
-		if (vio_version_after_eq(vio, 1, 6))
-			pkt.options = VIO_TX_DRING;
-	} else if (vio_version_before(vio, 1, 3)) {
-		pkt.mtu = framelen;
-	} else { /* v1.3 */
-		pkt.mtu = framelen + VLAN_HLEN;
-	}
-
-	pkt.cflags = 0;
-	if (vio_version_after_eq(vio, 1, 7) && port->tso) {
-		pkt.cflags |= VNET_LSO_IPV4_CAPAB;
-		if (!port->tsolen)
-			port->tsolen = VNET_MAXTSO;
-		pkt.ipv4_lso_maxlen = port->tsolen;
-	}
-
-	pkt.plnk_updt = PHYSLINK_UPDATE_NONE;
-
-	viodbg(HS, "SEND NET ATTR xmode[0x%x] atype[0x%x] addr[%llx] "
-	       "ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] mtu[%llu] "
-	       "cflags[0x%04x] lso_max[%u]\n",
-	       pkt.xfer_mode, pkt.addr_type,
-	       (unsigned long long)pkt.addr,
-	       pkt.ack_freq, pkt.plnk_updt, pkt.options,
-	       (unsigned long long)pkt.mtu, pkt.cflags, pkt.ipv4_lso_maxlen);
-
-
-	return vio_ldc_send(vio, &pkt, sizeof(pkt));
-}
-
-static int handle_attr_info(struct vio_driver_state *vio,
-			    struct vio_net_attr_info *pkt)
-{
-	struct vnet_port *port = to_vnet_port(vio);
-	u64	localmtu;
-	u8	xfer_mode;
-
-	viodbg(HS, "GOT NET ATTR xmode[0x%x] atype[0x%x] addr[%llx] "
-	       "ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] mtu[%llu] "
-	       " (rmtu[%llu]) cflags[0x%04x] lso_max[%u]\n",
-	       pkt->xfer_mode, pkt->addr_type,
-	       (unsigned long long)pkt->addr,
-	       pkt->ack_freq, pkt->plnk_updt, pkt->options,
-	       (unsigned long long)pkt->mtu, port->rmtu, pkt->cflags,
-	       pkt->ipv4_lso_maxlen);
-
-	pkt->tag.sid = vio_send_sid(vio);
-
-	xfer_mode = pkt->xfer_mode;
-	/* for version < 1.2, VIO_DRING_MODE = 0x3 and no bitmask */
-	if (vio_version_before(vio, 1, 2) && xfer_mode == VIO_DRING_MODE)
-		xfer_mode = VIO_NEW_DRING_MODE;
-
-	/* MTU negotiation:
-	 *	< v1.3 - ETH_FRAME_LEN exactly
-	 *	> v1.3 - MIN(pkt.mtu, VNET_MAXPACKET, port->rmtu) and change
-	 *			pkt->mtu for ACK
-	 *	= v1.3 - ETH_FRAME_LEN + VLAN_HLEN exactly
-	 */
-	if (vio_version_before(vio, 1, 3)) {
-		localmtu = ETH_FRAME_LEN;
-	} else if (vio_version_after(vio, 1, 3)) {
-		localmtu = port->rmtu ? port->rmtu : VNET_MAXPACKET;
-		localmtu = min(pkt->mtu, localmtu);
-		pkt->mtu = localmtu;
-	} else { /* v1.3 */
-		localmtu = ETH_FRAME_LEN + VLAN_HLEN;
-	}
-	port->rmtu = localmtu;
-
-	/* LSO negotiation */
-	if (vio_version_after_eq(vio, 1, 7))
-		port->tso &= !!(pkt->cflags & VNET_LSO_IPV4_CAPAB);
-	else
-		port->tso = false;
-	if (port->tso) {
-		if (!port->tsolen)
-			port->tsolen = VNET_MAXTSO;
-		port->tsolen = min(port->tsolen, pkt->ipv4_lso_maxlen);
-		if (port->tsolen < VNET_MINTSO) {
-			port->tso = false;
-			port->tsolen = 0;
-			pkt->cflags &= ~VNET_LSO_IPV4_CAPAB;
-		}
-		pkt->ipv4_lso_maxlen = port->tsolen;
-	} else {
-		pkt->cflags &= ~VNET_LSO_IPV4_CAPAB;
-		pkt->ipv4_lso_maxlen = 0;
-	}
-
-	/* for version >= 1.6, ACK packet mode we support */
-	if (vio_version_after_eq(vio, 1, 6)) {
-		pkt->xfer_mode = VIO_NEW_DRING_MODE;
-		pkt->options = VIO_TX_DRING;
-	}
-
-	if (!(xfer_mode | VIO_NEW_DRING_MODE) ||
-	    pkt->addr_type != VNET_ADDR_ETHERMAC ||
-	    pkt->mtu != localmtu) {
-		viodbg(HS, "SEND NET ATTR NACK\n");
-
-		pkt->tag.stype = VIO_SUBTYPE_NACK;
-
-		(void) vio_ldc_send(vio, pkt, sizeof(*pkt));
-
-		return -ECONNRESET;
-	} else {
-		viodbg(HS, "SEND NET ATTR ACK xmode[0x%x] atype[0x%x] "
-		       "addr[%llx] ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] "
-		       "mtu[%llu] (rmtu[%llu]) cflags[0x%04x] lso_max[%u]\n",
-		       pkt->xfer_mode, pkt->addr_type,
-		       (unsigned long long)pkt->addr,
-		       pkt->ack_freq, pkt->plnk_updt, pkt->options,
-		       (unsigned long long)pkt->mtu, port->rmtu, pkt->cflags,
-		       pkt->ipv4_lso_maxlen);
-
-		pkt->tag.stype = VIO_SUBTYPE_ACK;
-
-		return vio_ldc_send(vio, pkt, sizeof(*pkt));
-	}
-
-}
-
-static int handle_attr_ack(struct vio_driver_state *vio,
-			   struct vio_net_attr_info *pkt)
-{
-	viodbg(HS, "GOT NET ATTR ACK\n");
-
-	return 0;
-}
-
-static int handle_attr_nack(struct vio_driver_state *vio,
-			    struct vio_net_attr_info *pkt)
-{
-	viodbg(HS, "GOT NET ATTR NACK\n");
-
-	return -ECONNRESET;
-}
-
-static int vnet_handle_attr(struct vio_driver_state *vio, void *arg)
-{
-	struct vio_net_attr_info *pkt = arg;
-
-	switch (pkt->tag.stype) {
-	case VIO_SUBTYPE_INFO:
-		return handle_attr_info(vio, pkt);
-
-	case VIO_SUBTYPE_ACK:
-		return handle_attr_ack(vio, pkt);
-
-	case VIO_SUBTYPE_NACK:
-		return handle_attr_nack(vio, pkt);
-
-	default:
-		return -ECONNRESET;
-	}
-}
-
-static void vnet_handshake_complete(struct vio_driver_state *vio)
-{
-	struct vio_dring_state *dr;
-
-	dr = &vio->drings[VIO_DRIVER_RX_RING];
-	dr->snd_nxt = dr->rcv_nxt = 1;
-
-	dr = &vio->drings[VIO_DRIVER_TX_RING];
-	dr->snd_nxt = dr->rcv_nxt = 1;
-}
-
-/* The hypervisor interface that implements copying to/from imported
- * memory from another domain requires that copies are done to 8-byte
- * aligned buffers, and that the lengths of such copies are also 8-byte
- * multiples.
- *
- * So we align skb->data to an 8-byte multiple and pad-out the data
- * area so we can round the copy length up to the next multiple of
- * 8 for the copy.
- *
- * The transmitter puts the actual start of the packet 6 bytes into
- * the buffer it sends over, so that the IP headers after the ethernet
- * header are aligned properly.  These 6 bytes are not in the descriptor
- * length, they are simply implied.  This offset is represented using
- * the VNET_PACKET_SKIP macro.
- */
-static struct sk_buff *alloc_and_align_skb(struct net_device *dev,
-					   unsigned int len)
-{
-	struct sk_buff *skb = netdev_alloc_skb(dev, len+VNET_PACKET_SKIP+8+8);
-	unsigned long addr, off;
-
-	if (unlikely(!skb))
-		return NULL;
-
-	addr = (unsigned long) skb->data;
-	off = ((addr + 7UL) & ~7UL) - addr;
-	if (off)
-		skb_reserve(skb, off);
-
-	return skb;
-}
-
-static inline void vnet_fullcsum(struct sk_buff *skb)
-{
-	struct iphdr *iph = ip_hdr(skb);
-	int offset = skb_transport_offset(skb);
-
-	if (skb->protocol != htons(ETH_P_IP))
-		return;
-	if (iph->protocol != IPPROTO_TCP &&
-	    iph->protocol != IPPROTO_UDP)
-		return;
-	skb->ip_summed = CHECKSUM_NONE;
-	skb->csum_level = 1;
-	skb->csum = 0;
-	if (iph->protocol == IPPROTO_TCP) {
-		struct tcphdr *ptcp = tcp_hdr(skb);
-
-		ptcp->check = 0;
-		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
-		ptcp->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
-						skb->len - offset, IPPROTO_TCP,
-						skb->csum);
-	} else if (iph->protocol == IPPROTO_UDP) {
-		struct udphdr *pudp = udp_hdr(skb);
-
-		pudp->check = 0;
-		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
-		pudp->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
-						skb->len - offset, IPPROTO_UDP,
-						skb->csum);
-	}
-}
-
-static int vnet_rx_one(struct vnet_port *port, struct vio_net_desc *desc)
-{
-	struct net_device *dev = port->vp->dev;
-	unsigned int len = desc->size;
-	unsigned int copy_len;
-	struct sk_buff *skb;
-	int maxlen;
-	int err;
-
-	err = -EMSGSIZE;
-	if (port->tso && port->tsolen > port->rmtu)
-		maxlen = port->tsolen;
-	else
-		maxlen = port->rmtu;
-	if (unlikely(len < ETH_ZLEN || len > maxlen)) {
-		dev->stats.rx_length_errors++;
-		goto out_dropped;
-	}
-
-	skb = alloc_and_align_skb(dev, len);
-	err = -ENOMEM;
-	if (unlikely(!skb)) {
-		dev->stats.rx_missed_errors++;
-		goto out_dropped;
-	}
-
-	copy_len = (len + VNET_PACKET_SKIP + 7U) & ~7U;
-	skb_put(skb, copy_len);
-	err = ldc_copy(port->vio.lp, LDC_COPY_IN,
-		       skb->data, copy_len, 0,
-		       desc->cookies, desc->ncookies);
-	if (unlikely(err < 0)) {
-		dev->stats.rx_frame_errors++;
-		goto out_free_skb;
-	}
-
-	skb_pull(skb, VNET_PACKET_SKIP);
-	skb_trim(skb, len);
-	skb->protocol = eth_type_trans(skb, dev);
-
-	if (vio_version_after_eq(&port->vio, 1, 8)) {
-		struct vio_net_dext *dext = vio_net_ext(desc);
-
-		skb_reset_network_header(skb);
-
-		if (dext->flags & VNET_PKT_HCK_IPV4_HDRCKSUM) {
-			if (skb->protocol == ETH_P_IP) {
-				struct iphdr *iph = ip_hdr(skb);
-
-				iph->check = 0;
-				ip_send_check(iph);
-			}
-		}
-		if ((dext->flags & VNET_PKT_HCK_FULLCKSUM) &&
-		    skb->ip_summed == CHECKSUM_NONE) {
-			if (skb->protocol == htons(ETH_P_IP)) {
-				struct iphdr *iph = ip_hdr(skb);
-				int ihl = iph->ihl * 4;
-
-				skb_reset_transport_header(skb);
-				skb_set_transport_header(skb, ihl);
-				vnet_fullcsum(skb);
-			}
-		}
-		if (dext->flags & VNET_PKT_HCK_IPV4_HDRCKSUM_OK) {
-			skb->ip_summed = CHECKSUM_PARTIAL;
-			skb->csum_level = 0;
-			if (dext->flags & VNET_PKT_HCK_FULLCKSUM_OK)
-				skb->csum_level = 1;
-		}
-	}
-
-	skb->ip_summed = port->switch_port ? CHECKSUM_NONE : CHECKSUM_PARTIAL;
-
-	dev->stats.rx_packets++;
-	dev->stats.rx_bytes += len;
-	napi_gro_receive(&port->napi, skb);
-	return 0;
-
-out_free_skb:
-	kfree_skb(skb);
-
-out_dropped:
-	dev->stats.rx_dropped++;
-	return err;
-}
-
-static int vnet_send_ack(struct vnet_port *port, struct vio_dring_state *dr,
-			 u32 start, u32 end, u8 vio_dring_state)
-{
-	struct vio_dring_data hdr = {
-		.tag = {
-			.type		= VIO_TYPE_DATA,
-			.stype		= VIO_SUBTYPE_ACK,
-			.stype_env	= VIO_DRING_DATA,
-			.sid		= vio_send_sid(&port->vio),
-		},
-		.dring_ident		= dr->ident,
-		.start_idx		= start,
-		.end_idx		= end,
-		.state			= vio_dring_state,
-	};
-	int err, delay;
-	int retries = 0;
-
-	hdr.seq = dr->snd_nxt;
-	delay = 1;
-	do {
-		err = vio_ldc_send(&port->vio, &hdr, sizeof(hdr));
-		if (err > 0) {
-			dr->snd_nxt++;
-			break;
-		}
-		udelay(delay);
-		if ((delay <<= 1) > 128)
-			delay = 128;
-		if (retries++ > VNET_MAX_RETRIES) {
-			pr_info("ECONNRESET %x:%x:%x:%x:%x:%x\n",
-				port->raddr[0], port->raddr[1],
-				port->raddr[2], port->raddr[3],
-				port->raddr[4], port->raddr[5]);
-			break;
-		}
-	} while (err == -EAGAIN);
-
-	if (err <= 0 && vio_dring_state == VIO_DRING_STOPPED) {
-		port->stop_rx_idx = end;
-		port->stop_rx = true;
-	} else {
-		port->stop_rx_idx = 0;
-		port->stop_rx = false;
-	}
-
-	return err;
-}
-
-static struct vio_net_desc *get_rx_desc(struct vnet_port *port,
-					struct vio_dring_state *dr,
-					u32 index)
-{
-	struct vio_net_desc *desc = port->vio.desc_buf;
-	int err;
-
-	err = ldc_get_dring_entry(port->vio.lp, desc, dr->entry_size,
-				  (index * dr->entry_size),
-				  dr->cookies, dr->ncookies);
-	if (err < 0)
-		return ERR_PTR(err);
-
-	return desc;
-}
-
-static int put_rx_desc(struct vnet_port *port,
-		       struct vio_dring_state *dr,
-		       struct vio_net_desc *desc,
-		       u32 index)
-{
-	int err;
-
-	err = ldc_put_dring_entry(port->vio.lp, desc, dr->entry_size,
-				  (index * dr->entry_size),
-				  dr->cookies, dr->ncookies);
-	if (err < 0)
-		return err;
-
-	return 0;
-}
-
-static int vnet_walk_rx_one(struct vnet_port *port,
-			    struct vio_dring_state *dr,
-			    u32 index, int *needs_ack)
-{
-	struct vio_net_desc *desc = get_rx_desc(port, dr, index);
-	struct vio_driver_state *vio = &port->vio;
-	int err;
-
-	BUG_ON(desc == NULL);
-	if (IS_ERR(desc))
-		return PTR_ERR(desc);
-
-	if (desc->hdr.state != VIO_DESC_READY)
-		return 1;
-
-	dma_rmb();
-
-	viodbg(DATA, "vio_walk_rx_one desc[%02x:%02x:%08x:%08x:%llx:%llx]\n",
-	       desc->hdr.state, desc->hdr.ack,
-	       desc->size, desc->ncookies,
-	       desc->cookies[0].cookie_addr,
-	       desc->cookies[0].cookie_size);
-
-	err = vnet_rx_one(port, desc);
-	if (err == -ECONNRESET)
-		return err;
-	trace_vnet_rx_one(port->vio._local_sid, port->vio._peer_sid,
-			  index, desc->hdr.ack);
-	desc->hdr.state = VIO_DESC_DONE;
-	err = put_rx_desc(port, dr, desc, index);
-	if (err < 0)
-		return err;
-	*needs_ack = desc->hdr.ack;
-	return 0;
-}
-
-static int vnet_walk_rx(struct vnet_port *port, struct vio_dring_state *dr,
-			u32 start, u32 end, int *npkts, int budget)
-{
-	struct vio_driver_state *vio = &port->vio;
-	int ack_start = -1, ack_end = -1;
-	bool send_ack = true;
-
-	end = (end == (u32) -1) ? vio_dring_prev(dr, start)
-				: vio_dring_next(dr, end);
-
-	viodbg(DATA, "vnet_walk_rx start[%08x] end[%08x]\n", start, end);
-
-	while (start != end) {
-		int ack = 0, err = vnet_walk_rx_one(port, dr, start, &ack);
-		if (err == -ECONNRESET)
-			return err;
-		if (err != 0)
-			break;
-		(*npkts)++;
-		if (ack_start == -1)
-			ack_start = start;
-		ack_end = start;
-		start = vio_dring_next(dr, start);
-		if (ack && start != end) {
-			err = vnet_send_ack(port, dr, ack_start, ack_end,
-					    VIO_DRING_ACTIVE);
-			if (err == -ECONNRESET)
-				return err;
-			ack_start = -1;
-		}
-		if ((*npkts) >= budget) {
-			send_ack = false;
-			break;
-		}
-	}
-	if (unlikely(ack_start == -1))
-		ack_start = ack_end = vio_dring_prev(dr, start);
-	if (send_ack) {
-		port->napi_resume = false;
-		trace_vnet_tx_send_stopped_ack(port->vio._local_sid,
-					       port->vio._peer_sid,
-					       ack_end, *npkts);
-		return vnet_send_ack(port, dr, ack_start, ack_end,
-				     VIO_DRING_STOPPED);
-	} else  {
-		trace_vnet_tx_defer_stopped_ack(port->vio._local_sid,
-						port->vio._peer_sid,
-						ack_end, *npkts);
-		port->napi_resume = true;
-		port->napi_stop_idx = ack_end;
-		return 1;
-	}
-}
-
-static int vnet_rx(struct vnet_port *port, void *msgbuf, int *npkts,
-		   int budget)
-{
-	struct vio_dring_data *pkt = msgbuf;
-	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_RX_RING];
-	struct vio_driver_state *vio = &port->vio;
-
-	viodbg(DATA, "vnet_rx stype_env[%04x] seq[%016llx] rcv_nxt[%016llx]\n",
-	       pkt->tag.stype_env, pkt->seq, dr->rcv_nxt);
-
-	if (unlikely(pkt->tag.stype_env != VIO_DRING_DATA))
-		return 0;
-	if (unlikely(pkt->seq != dr->rcv_nxt)) {
-		pr_err("RX out of sequence seq[0x%llx] rcv_nxt[0x%llx]\n",
-		       pkt->seq, dr->rcv_nxt);
-		return 0;
-	}
-
-	if (!port->napi_resume)
-		dr->rcv_nxt++;
-
-	/* XXX Validate pkt->start_idx and pkt->end_idx XXX */
-
-	return vnet_walk_rx(port, dr, pkt->start_idx, pkt->end_idx,
-			    npkts, budget);
-}
-
-static int idx_is_pending(struct vio_dring_state *dr, u32 end)
-{
-	u32 idx = dr->cons;
-	int found = 0;
-
-	while (idx != dr->prod) {
-		if (idx == end) {
-			found = 1;
-			break;
-		}
-		idx = vio_dring_next(dr, idx);
-	}
-	return found;
-}
-
-static int vnet_ack(struct vnet_port *port, void *msgbuf)
-{
-	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-	struct vio_dring_data *pkt = msgbuf;
-	struct net_device *dev;
-	struct vnet *vp;
-	u32 end;
-	struct vio_net_desc *desc;
-	struct netdev_queue *txq;
-
-	if (unlikely(pkt->tag.stype_env != VIO_DRING_DATA))
-		return 0;
-
-	end = pkt->end_idx;
-	vp = port->vp;
-	dev = vp->dev;
-	netif_tx_lock(dev);
-	if (unlikely(!idx_is_pending(dr, end))) {
-		netif_tx_unlock(dev);
-		return 0;
-	}
-
-	/* sync for race conditions with vnet_start_xmit() and tell xmit it
-	 * is time to send a trigger.
-	 */
-	trace_vnet_rx_stopped_ack(port->vio._local_sid,
-				  port->vio._peer_sid, end);
-	dr->cons = vio_dring_next(dr, end);
-	desc = vio_dring_entry(dr, dr->cons);
-	if (desc->hdr.state == VIO_DESC_READY && !port->start_cons) {
-		/* vnet_start_xmit() just populated this dring but missed
-		 * sending the "start" LDC message to the consumer.
-		 * Send a "start" trigger on its behalf.
-		 */
-		if (__vnet_tx_trigger(port, dr->cons) > 0)
-			port->start_cons = false;
-		else
-			port->start_cons = true;
-	} else {
-		port->start_cons = true;
-	}
-	netif_tx_unlock(dev);
-
-	txq = netdev_get_tx_queue(dev, port->q_index);
-	if (unlikely(netif_tx_queue_stopped(txq) &&
-		     vnet_tx_dring_avail(dr) >= VNET_TX_WAKEUP_THRESH(dr)))
-		return 1;
-
-	return 0;
-}
-
-static int vnet_nack(struct vnet_port *port, void *msgbuf)
-{
-	/* XXX just reset or similar XXX */
-	return 0;
-}
-
-static int handle_mcast(struct vnet_port *port, void *msgbuf)
-{
-	struct vio_net_mcast_info *pkt = msgbuf;
-
-	if (pkt->tag.stype != VIO_SUBTYPE_ACK)
-		pr_err("%s: Got unexpected MCAST reply [%02x:%02x:%04x:%08x]\n",
-		       port->vp->dev->name,
-		       pkt->tag.type,
-		       pkt->tag.stype,
-		       pkt->tag.stype_env,
-		       pkt->tag.sid);
-
-	return 0;
-}
-
-/* Got back a STOPPED LDC message on port. If the queue is stopped,
- * wake it up so that we'll send out another START message at the
- * next TX.
- */
-static void maybe_tx_wakeup(struct vnet_port *port)
-{
-	struct netdev_queue *txq;
-
-	txq = netdev_get_tx_queue(port->vp->dev, port->q_index);
-	__netif_tx_lock(txq, smp_processor_id());
-	if (likely(netif_tx_queue_stopped(txq))) {
-		struct vio_dring_state *dr;
-
-		dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-		netif_tx_wake_queue(txq);
-	}
-	__netif_tx_unlock(txq);
-}
-
-static inline bool port_is_up(struct vnet_port *vnet)
-{
-	struct vio_driver_state *vio = &vnet->vio;
-
-	return !!(vio->hs_state & VIO_HS_COMPLETE);
-}
-
-static int vnet_event_napi(struct vnet_port *port, int budget)
-{
-	struct vio_driver_state *vio = &port->vio;
-	int tx_wakeup, err;
-	int npkts = 0;
-	int event = (port->rx_event & LDC_EVENT_RESET);
-
-ldc_ctrl:
-	if (unlikely(event == LDC_EVENT_RESET ||
-		     event == LDC_EVENT_UP)) {
-		vio_link_state_change(vio, event);
-
-		if (event == LDC_EVENT_RESET) {
-			vnet_port_reset(port);
-			vio_port_up(vio);
-		}
-		port->rx_event = 0;
-		return 0;
-	}
-	/* We may have multiple LDC events in rx_event. Unroll send_events() */
-	event = (port->rx_event & LDC_EVENT_UP);
-	port->rx_event &= ~(LDC_EVENT_RESET|LDC_EVENT_UP);
-	if (event == LDC_EVENT_UP)
-		goto ldc_ctrl;
-	event = port->rx_event;
-	if (!(event & LDC_EVENT_DATA_READY))
-		return 0;
-
-	/* we dont expect any other bits than RESET, UP, DATA_READY */
-	BUG_ON(event != LDC_EVENT_DATA_READY);
-
-	tx_wakeup = err = 0;
-	while (1) {
-		union {
-			struct vio_msg_tag tag;
-			u64 raw[8];
-		} msgbuf;
-
-		if (port->napi_resume) {
-			struct vio_dring_data *pkt =
-				(struct vio_dring_data *)&msgbuf;
-			struct vio_dring_state *dr =
-				&port->vio.drings[VIO_DRIVER_RX_RING];
-
-			pkt->tag.type = VIO_TYPE_DATA;
-			pkt->tag.stype = VIO_SUBTYPE_INFO;
-			pkt->tag.stype_env = VIO_DRING_DATA;
-			pkt->seq = dr->rcv_nxt;
-			pkt->start_idx = vio_dring_next(dr, port->napi_stop_idx);
-			pkt->end_idx = -1;
-			goto napi_resume;
-		}
-		err = ldc_read(vio->lp, &msgbuf, sizeof(msgbuf));
-		if (unlikely(err < 0)) {
-			if (err == -ECONNRESET)
-				vio_conn_reset(vio);
-			break;
-		}
-		if (err == 0)
-			break;
-		viodbg(DATA, "TAG [%02x:%02x:%04x:%08x]\n",
-		       msgbuf.tag.type,
-		       msgbuf.tag.stype,
-		       msgbuf.tag.stype_env,
-		       msgbuf.tag.sid);
-		err = vio_validate_sid(vio, &msgbuf.tag);
-		if (err < 0)
-			break;
-napi_resume:
-		if (likely(msgbuf.tag.type == VIO_TYPE_DATA)) {
-			if (msgbuf.tag.stype == VIO_SUBTYPE_INFO) {
-				if (!port_is_up(port)) {
-					/* failures like handshake_failure()
-					 * may have cleaned up dring, but
-					 * NAPI polling may bring us here.
-					 */
-					err = -ECONNRESET;
-					break;
-				}
-				err = vnet_rx(port, &msgbuf, &npkts, budget);
-				if (npkts >= budget)
-					break;
-				if (npkts == 0)
-					break;
-			} else if (msgbuf.tag.stype == VIO_SUBTYPE_ACK) {
-				err = vnet_ack(port, &msgbuf);
-				if (err > 0)
-					tx_wakeup |= err;
-			} else if (msgbuf.tag.stype == VIO_SUBTYPE_NACK) {
-				err = vnet_nack(port, &msgbuf);
-			}
-		} else if (msgbuf.tag.type == VIO_TYPE_CTRL) {
-			if (msgbuf.tag.stype_env == VNET_MCAST_INFO)
-				err = handle_mcast(port, &msgbuf);
-			else
-				err = vio_control_pkt_engine(vio, &msgbuf);
-			if (err)
-				break;
-		} else {
-			err = vnet_handle_unknown(port, &msgbuf);
-		}
-		if (err == -ECONNRESET)
-			break;
-	}
-	if (unlikely(tx_wakeup && err != -ECONNRESET))
-		maybe_tx_wakeup(port);
-	return npkts;
-}
-
-static int vnet_poll(struct napi_struct *napi, int budget)
-{
-	struct vnet_port *port = container_of(napi, struct vnet_port, napi);
-	struct vio_driver_state *vio = &port->vio;
-	int processed = vnet_event_napi(port, budget);
-
-	if (processed < budget) {
-		napi_complete(napi);
-		port->rx_event &= ~LDC_EVENT_DATA_READY;
-		vio_set_intr(vio->vdev->rx_ino, HV_INTR_ENABLED);
-	}
-	return processed;
-}
-
-static void vnet_event(void *arg, int event)
-{
-	struct vnet_port *port = arg;
-	struct vio_driver_state *vio = &port->vio;
-
-	port->rx_event |= event;
-	vio_set_intr(vio->vdev->rx_ino, HV_INTR_DISABLED);
-	napi_schedule(&port->napi);
-
-}
-
-static int __vnet_tx_trigger(struct vnet_port *port, u32 start)
-{
-	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-	struct vio_dring_data hdr = {
-		.tag = {
-			.type		= VIO_TYPE_DATA,
-			.stype		= VIO_SUBTYPE_INFO,
-			.stype_env	= VIO_DRING_DATA,
-			.sid		= vio_send_sid(&port->vio),
-		},
-		.dring_ident		= dr->ident,
-		.start_idx		= start,
-		.end_idx		= (u32) -1,
-	};
-	int err, delay;
-	int retries = 0;
-
-	if (port->stop_rx) {
-		trace_vnet_tx_pending_stopped_ack(port->vio._local_sid,
-						  port->vio._peer_sid,
-						  port->stop_rx_idx, -1);
-		err = vnet_send_ack(port,
-				    &port->vio.drings[VIO_DRIVER_RX_RING],
-				    port->stop_rx_idx, -1,
-				    VIO_DRING_STOPPED);
-		if (err <= 0)
-			return err;
-	}
-
-	hdr.seq = dr->snd_nxt;
-	delay = 1;
-	do {
-		err = vio_ldc_send(&port->vio, &hdr, sizeof(hdr));
-		if (err > 0) {
-			dr->snd_nxt++;
-			break;
-		}
-		udelay(delay);
-		if ((delay <<= 1) > 128)
-			delay = 128;
-		if (retries++ > VNET_MAX_RETRIES)
-			break;
-	} while (err == -EAGAIN);
-	trace_vnet_tx_trigger(port->vio._local_sid,
-			      port->vio._peer_sid, start, err);
-
-	return err;
-}
-
-struct vnet_port *__tx_port_find(struct vnet *vp, struct sk_buff *skb)
-{
-	unsigned int hash = vnet_hashfn(skb->data);
-	struct hlist_head *hp = &vp->port_hash[hash];
-	struct vnet_port *port;
-
-	hlist_for_each_entry_rcu(port, hp, hash) {
-		if (!port_is_up(port))
-			continue;
-		if (ether_addr_equal(port->raddr, skb->data))
-			return port;
-	}
-	list_for_each_entry_rcu(port, &vp->port_list, list) {
-		if (!port->switch_port)
-			continue;
-		if (!port_is_up(port))
-			continue;
-		return port;
-	}
-	return NULL;
-}
-
-static struct sk_buff *vnet_clean_tx_ring(struct vnet_port *port,
-					  unsigned *pending)
-{
-	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-	struct sk_buff *skb = NULL;
-	int i, txi;
-
-	*pending = 0;
-
-	txi = dr->prod;
-	for (i = 0; i < VNET_TX_RING_SIZE; ++i) {
-		struct vio_net_desc *d;
-
-		--txi;
-		if (txi < 0)
-			txi = VNET_TX_RING_SIZE-1;
-
-		d = vio_dring_entry(dr, txi);
-
-		if (d->hdr.state == VIO_DESC_READY) {
-			(*pending)++;
-			continue;
-		}
-		if (port->tx_bufs[txi].skb) {
-			if (d->hdr.state != VIO_DESC_DONE)
-				pr_notice("invalid ring buffer state %d\n",
-					  d->hdr.state);
-			BUG_ON(port->tx_bufs[txi].skb->next);
-
-			port->tx_bufs[txi].skb->next = skb;
-			skb = port->tx_bufs[txi].skb;
-			port->tx_bufs[txi].skb = NULL;
-
-			ldc_unmap(port->vio.lp,
-				  port->tx_bufs[txi].cookies,
-				  port->tx_bufs[txi].ncookies);
-		} else if (d->hdr.state == VIO_DESC_FREE)
-			break;
-		d->hdr.state = VIO_DESC_FREE;
-	}
-	return skb;
-}
-
-static inline void vnet_free_skbs(struct sk_buff *skb)
-{
-	struct sk_buff *next;
-
-	while (skb) {
-		next = skb->next;
-		skb->next = NULL;
-		dev_kfree_skb(skb);
-		skb = next;
-	}
-}
-
-static void vnet_clean_timer_expire(unsigned long port0)
-{
-	struct vnet_port *port = (struct vnet_port *)port0;
-	struct sk_buff *freeskbs;
-	unsigned pending;
-
-	netif_tx_lock(port->vp->dev);
-	freeskbs = vnet_clean_tx_ring(port, &pending);
-	netif_tx_unlock(port->vp->dev);
-
-	vnet_free_skbs(freeskbs);
-
-	if (pending)
-		(void)mod_timer(&port->clean_timer,
-				jiffies + VNET_CLEAN_TIMEOUT);
-	 else
-		del_timer(&port->clean_timer);
-}
-
-static inline int vnet_skb_map(struct ldc_channel *lp, struct sk_buff *skb,
-			       struct ldc_trans_cookie *cookies, int ncookies,
-			       unsigned int map_perm)
-{
-	int i, nc, err, blen;
-
-	/* header */
-	blen = skb_headlen(skb);
-	if (blen < ETH_ZLEN)
-		blen = ETH_ZLEN;
-	blen += VNET_PACKET_SKIP;
-	blen += 8 - (blen & 7);
-
-	err = ldc_map_single(lp, skb->data-VNET_PACKET_SKIP, blen, cookies,
-			     ncookies, map_perm);
-	if (err < 0)
-		return err;
-	nc = err;
-
-	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		skb_frag_t *f = &skb_shinfo(skb)->frags[i];
-		u8 *vaddr;
-
-		if (nc < ncookies) {
-			vaddr = kmap_atomic(skb_frag_page(f));
-			blen = skb_frag_size(f);
-			blen += 8 - (blen & 7);
-			err = ldc_map_single(lp, vaddr + f->page_offset,
-					     blen, cookies + nc, ncookies - nc,
-					     map_perm);
-			kunmap_atomic(vaddr);
-		} else {
-			err = -EMSGSIZE;
-		}
-
-		if (err < 0) {
-			ldc_unmap(lp, cookies, nc);
-			return err;
-		}
-		nc += err;
-	}
-	return nc;
-}
-
-static inline struct sk_buff *vnet_skb_shape(struct sk_buff *skb, int ncookies)
-{
-	struct sk_buff *nskb;
-	int i, len, pad, docopy;
-
-	len = skb->len;
-	pad = 0;
-	if (len < ETH_ZLEN) {
-		pad += ETH_ZLEN - skb->len;
-		len += pad;
-	}
-	len += VNET_PACKET_SKIP;
-	pad += 8 - (len & 7);
-
-	/* make sure we have enough cookies and alignment in every frag */
-	docopy = skb_shinfo(skb)->nr_frags >= ncookies;
-	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		skb_frag_t *f = &skb_shinfo(skb)->frags[i];
-
-		docopy |= f->page_offset & 7;
-	}
-	if (((unsigned long)skb->data & 7) != VNET_PACKET_SKIP ||
-	    skb_tailroom(skb) < pad ||
-	    skb_headroom(skb) < VNET_PACKET_SKIP || docopy) {
-		int start = 0, offset;
-		__wsum csum;
-
-		len = skb->len > ETH_ZLEN ? skb->len : ETH_ZLEN;
-		nskb = alloc_and_align_skb(skb->dev, len);
-		if (nskb == NULL) {
-			dev_kfree_skb(skb);
-			return NULL;
-		}
-		skb_reserve(nskb, VNET_PACKET_SKIP);
-
-		nskb->protocol = skb->protocol;
-		offset = skb_mac_header(skb) - skb->data;
-		skb_set_mac_header(nskb, offset);
-		offset = skb_network_header(skb) - skb->data;
-		skb_set_network_header(nskb, offset);
-		offset = skb_transport_header(skb) - skb->data;
-		skb_set_transport_header(nskb, offset);
-
-		offset = 0;
-		nskb->csum_offset = skb->csum_offset;
-		nskb->ip_summed = skb->ip_summed;
-
-		if (skb->ip_summed == CHECKSUM_PARTIAL)
-			start = skb_checksum_start_offset(skb);
-		if (start) {
-			struct iphdr *iph = ip_hdr(nskb);
-			int offset = start + nskb->csum_offset;
-
-			if (skb_copy_bits(skb, 0, nskb->data, start)) {
-				dev_kfree_skb(nskb);
-				dev_kfree_skb(skb);
-				return NULL;
-			}
-			*(__sum16 *)(skb->data + offset) = 0;
-			csum = skb_copy_and_csum_bits(skb, start,
-						      nskb->data + start,
-						      skb->len - start, 0);
-			if (iph->protocol == IPPROTO_TCP ||
-			    iph->protocol == IPPROTO_UDP) {
-				csum = csum_tcpudp_magic(iph->saddr, iph->daddr,
-							 skb->len - start,
-							 iph->protocol, csum);
-			}
-			*(__sum16 *)(nskb->data + offset) = csum;
-
-			nskb->ip_summed = CHECKSUM_NONE;
-		} else if (skb_copy_bits(skb, 0, nskb->data, skb->len)) {
-			dev_kfree_skb(nskb);
-			dev_kfree_skb(skb);
-			return NULL;
-		}
-		(void)skb_put(nskb, skb->len);
-		if (skb_is_gso(skb)) {
-			skb_shinfo(nskb)->gso_size = skb_shinfo(skb)->gso_size;
-			skb_shinfo(nskb)->gso_type = skb_shinfo(skb)->gso_type;
-		}
-		nskb->queue_mapping = skb->queue_mapping;
-		dev_kfree_skb(skb);
-		skb = nskb;
-	}
-	return skb;
-}
-
-static u16
-vnet_select_queue(struct net_device *dev, struct sk_buff *skb,
-		  void *accel_priv, select_queue_fallback_t fallback)
-{
-	struct vnet *vp = netdev_priv(dev);
-	struct vnet_port *port = __tx_port_find(vp, skb);
-
-	if (port == NULL)
-		return 0;
-	return port->q_index;
-}
-
-static int vnet_start_xmit(struct sk_buff *skb, struct net_device *dev);
-
-static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb)
-{
-	struct net_device *dev = port->vp->dev;
-	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-	struct sk_buff *segs;
-	int maclen, datalen;
-	int status;
-	int gso_size, gso_type, gso_segs;
-	int hlen = skb_transport_header(skb) - skb_mac_header(skb);
-	int proto = IPPROTO_IP;
-
-	if (skb->protocol == htons(ETH_P_IP))
-		proto = ip_hdr(skb)->protocol;
-	else if (skb->protocol == htons(ETH_P_IPV6))
-		proto = ipv6_hdr(skb)->nexthdr;
-
-	if (proto == IPPROTO_TCP)
-		hlen += tcp_hdr(skb)->doff * 4;
-	else if (proto == IPPROTO_UDP)
-		hlen += sizeof(struct udphdr);
-	else {
-		pr_err("vnet_handle_offloads GSO with unknown transport "
-		       "protocol %d tproto %d\n", skb->protocol, proto);
-		hlen = 128; /* XXX */
-	}
-	datalen = port->tsolen - hlen;
-
-	gso_size = skb_shinfo(skb)->gso_size;
-	gso_type = skb_shinfo(skb)->gso_type;
-	gso_segs = skb_shinfo(skb)->gso_segs;
-
-	if (port->tso && gso_size < datalen)
-		gso_segs = DIV_ROUND_UP(skb->len - hlen, datalen);
-
-	if (unlikely(vnet_tx_dring_avail(dr) < gso_segs)) {
-		struct netdev_queue *txq;
-
-		txq  = netdev_get_tx_queue(dev, port->q_index);
-		netif_tx_stop_queue(txq);
-		if (vnet_tx_dring_avail(dr) < skb_shinfo(skb)->gso_segs)
-			return NETDEV_TX_BUSY;
-		netif_tx_wake_queue(txq);
-	}
-
-	maclen = skb_network_header(skb) - skb_mac_header(skb);
-	skb_pull(skb, maclen);
-
-	if (port->tso && gso_size < datalen) {
-		if (skb_unclone(skb, GFP_ATOMIC))
-			goto out_dropped;
-
-		/* segment to TSO size */
-		skb_shinfo(skb)->gso_size = datalen;
-		skb_shinfo(skb)->gso_segs = gso_segs;
-	}
-	segs = skb_gso_segment(skb, dev->features & ~NETIF_F_TSO);
-	if (IS_ERR(segs))
-		goto out_dropped;
-
-	skb_push(skb, maclen);
-	skb_reset_mac_header(skb);
-
-	status = 0;
-	while (segs) {
-		struct sk_buff *curr = segs;
-
-		segs = segs->next;
-		curr->next = NULL;
-		if (port->tso && curr->len > dev->mtu) {
-			skb_shinfo(curr)->gso_size = gso_size;
-			skb_shinfo(curr)->gso_type = gso_type;
-			skb_shinfo(curr)->gso_segs =
-				DIV_ROUND_UP(curr->len - hlen, gso_size);
-		} else
-			skb_shinfo(curr)->gso_size = 0;
-
-		skb_push(curr, maclen);
-		skb_reset_mac_header(curr);
-		memcpy(skb_mac_header(curr), skb_mac_header(skb),
-		       maclen);
-		curr->csum_start = skb_transport_header(curr) - curr->head;
-		if (ip_hdr(curr)->protocol == IPPROTO_TCP)
-			curr->csum_offset = offsetof(struct tcphdr, check);
-		else if (ip_hdr(curr)->protocol == IPPROTO_UDP)
-			curr->csum_offset = offsetof(struct udphdr, check);
-
-		if (!(status & NETDEV_TX_MASK))
-			status = vnet_start_xmit(curr, dev);
-		if (status & NETDEV_TX_MASK)
-			dev_kfree_skb_any(curr);
-	}
-
-	if (!(status & NETDEV_TX_MASK))
-		dev_kfree_skb_any(skb);
-	return status;
-out_dropped:
-	dev->stats.tx_dropped++;
-	dev_kfree_skb_any(skb);
-	return NETDEV_TX_OK;
-}
-
-static int vnet_start_xmit(struct sk_buff *skb, struct net_device *dev)
-{
-	struct vnet *vp = netdev_priv(dev);
-	struct vnet_port *port = NULL;
-	struct vio_dring_state *dr;
-	struct vio_net_desc *d;
-	unsigned int len;
-	struct sk_buff *freeskbs = NULL;
-	int i, err, txi;
-	unsigned pending = 0;
-	struct netdev_queue *txq;
-
-	rcu_read_lock();
-	port = __tx_port_find(vp, skb);
-	if (unlikely(!port)) {
-		rcu_read_unlock();
-		goto out_dropped;
-	}
-
-	if (skb_is_gso(skb) && skb->len > port->tsolen) {
-		err = vnet_handle_offloads(port, skb);
-		rcu_read_unlock();
-		return err;
-	}
-
-	if (!skb_is_gso(skb) && skb->len > port->rmtu) {
-		unsigned long localmtu = port->rmtu - ETH_HLEN;
-
-		if (vio_version_after_eq(&port->vio, 1, 3))
-			localmtu -= VLAN_HLEN;
-
-		if (skb->protocol == htons(ETH_P_IP)) {
-			struct flowi4 fl4;
-			struct rtable *rt = NULL;
-
-			memset(&fl4, 0, sizeof(fl4));
-			fl4.flowi4_oif = dev->ifindex;
-			fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos);
-			fl4.daddr = ip_hdr(skb)->daddr;
-			fl4.saddr = ip_hdr(skb)->saddr;
-
-			rt = ip_route_output_key(dev_net(dev), &fl4);
-			rcu_read_unlock();
-			if (!IS_ERR(rt)) {
-				skb_dst_set(skb, &rt->dst);
-				icmp_send(skb, ICMP_DEST_UNREACH,
-					  ICMP_FRAG_NEEDED,
-					  htonl(localmtu));
-			}
-		}
-#if IS_ENABLED(CONFIG_IPV6)
-		else if (skb->protocol == htons(ETH_P_IPV6))
-			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, localmtu);
-#endif
-		goto out_dropped;
-	}
-
-	skb = vnet_skb_shape(skb, 2);
-
-	if (unlikely(!skb))
-		goto out_dropped;
-
-	if (skb->ip_summed == CHECKSUM_PARTIAL)
-		vnet_fullcsum(skb);
-
-	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-	i = skb_get_queue_mapping(skb);
-	txq = netdev_get_tx_queue(dev, i);
-	if (unlikely(vnet_tx_dring_avail(dr) < 1)) {
-		if (!netif_tx_queue_stopped(txq)) {
-			netif_tx_stop_queue(txq);
-
-			/* This is a hard error, log it. */
-			netdev_err(dev, "BUG! Tx Ring full when queue awake!\n");
-			dev->stats.tx_errors++;
-		}
-		rcu_read_unlock();
-		return NETDEV_TX_BUSY;
-	}
-
-	d = vio_dring_cur(dr);
-
-	txi = dr->prod;
-
-	freeskbs = vnet_clean_tx_ring(port, &pending);
-
-	BUG_ON(port->tx_bufs[txi].skb);
-
-	len = skb->len;
-	if (len < ETH_ZLEN)
-		len = ETH_ZLEN;
-
-	err = vnet_skb_map(port->vio.lp, skb, port->tx_bufs[txi].cookies, 2,
-			   (LDC_MAP_SHADOW | LDC_MAP_DIRECT | LDC_MAP_RW));
-	if (err < 0) {
-		netdev_info(dev, "tx buffer map error %d\n", err);
-		goto out_dropped;
-	}
-
-	port->tx_bufs[txi].skb = skb;
-	skb = NULL;
-	port->tx_bufs[txi].ncookies = err;
-
-	/* We don't rely on the ACKs to free the skb in vnet_start_xmit(),
-	 * thus it is safe to not set VIO_ACK_ENABLE for each transmission:
-	 * the protocol itself does not require it as long as the peer
-	 * sends a VIO_SUBTYPE_ACK for VIO_DRING_STOPPED.
-	 *
-	 * An ACK for every packet in the ring is expensive as the
-	 * sending of LDC messages is slow and affects performance.
-	 */
-	d->hdr.ack = VIO_ACK_DISABLE;
-	d->size = len;
-	d->ncookies = port->tx_bufs[txi].ncookies;
-	for (i = 0; i < d->ncookies; i++)
-		d->cookies[i] = port->tx_bufs[txi].cookies[i];
-	if (vio_version_after_eq(&port->vio, 1, 7)) {
-		struct vio_net_dext *dext = vio_net_ext(d);
-
-		memset(dext, 0, sizeof(*dext));
-		if (skb_is_gso(port->tx_bufs[txi].skb)) {
-			dext->ipv4_lso_mss = skb_shinfo(port->tx_bufs[txi].skb)
-					     ->gso_size;
-			dext->flags |= VNET_PKT_IPV4_LSO;
-		}
-		if (vio_version_after_eq(&port->vio, 1, 8) &&
-		    !port->switch_port) {
-			dext->flags |= VNET_PKT_HCK_IPV4_HDRCKSUM_OK;
-			dext->flags |= VNET_PKT_HCK_FULLCKSUM_OK;
-		}
-	}
-
-	/* This has to be a non-SMP write barrier because we are writing
-	 * to memory which is shared with the peer LDOM.
-	 */
-	dma_wmb();
-
-	d->hdr.state = VIO_DESC_READY;
-
-	/* Exactly one ldc "start" trigger (for dr->cons) needs to be sent
-	 * to notify the consumer that some descriptors are READY.
-	 * After that "start" trigger, no additional triggers are needed until
-	 * a DRING_STOPPED is received from the consumer. The dr->cons field
-	 * (set up by vnet_ack()) has the value of the next dring index
-	 * that has not yet been ack-ed. We send a "start" trigger here
-	 * if, and only if, start_cons is true (reset it afterward). Conversely,
-	 * vnet_ack() should check if the dring corresponding to cons
-	 * is marked READY, but start_cons was false.
-	 * If so, vnet_ack() should send out the missed "start" trigger.
-	 *
-	 * Note that the dma_wmb() above makes sure the cookies et al. are
-	 * not globally visible before the VIO_DESC_READY, and that the
-	 * stores are ordered correctly by the compiler. The consumer will
-	 * not proceed until the VIO_DESC_READY is visible assuring that
-	 * the consumer does not observe anything related to descriptors
-	 * out of order. The HV trap from the LDC start trigger is the
-	 * producer to consumer announcement that work is available to the
-	 * consumer
-	 */
-	if (!port->start_cons) { /* previous trigger suffices */
-		trace_vnet_skip_tx_trigger(port->vio._local_sid,
-					   port->vio._peer_sid, dr->cons);
-		goto ldc_start_done;
-	}
-
-	err = __vnet_tx_trigger(port, dr->cons);
-	if (unlikely(err < 0)) {
-		netdev_info(dev, "TX trigger error %d\n", err);
-		d->hdr.state = VIO_DESC_FREE;
-		skb = port->tx_bufs[txi].skb;
-		port->tx_bufs[txi].skb = NULL;
-		dev->stats.tx_carrier_errors++;
-		goto out_dropped;
-	}
-
-ldc_start_done:
-	port->start_cons = false;
-
-	dev->stats.tx_packets++;
-	dev->stats.tx_bytes += port->tx_bufs[txi].skb->len;
-
-	dr->prod = (dr->prod + 1) & (VNET_TX_RING_SIZE - 1);
-	if (unlikely(vnet_tx_dring_avail(dr) < 1)) {
-		netif_tx_stop_queue(txq);
-		if (vnet_tx_dring_avail(dr) > VNET_TX_WAKEUP_THRESH(dr))
-			netif_tx_wake_queue(txq);
-	}
-
-	(void)mod_timer(&port->clean_timer, jiffies + VNET_CLEAN_TIMEOUT);
-	rcu_read_unlock();
-
-	vnet_free_skbs(freeskbs);
-
-	return NETDEV_TX_OK;
-
-out_dropped:
-	if (pending)
-		(void)mod_timer(&port->clean_timer,
-				jiffies + VNET_CLEAN_TIMEOUT);
-	else if (port)
-		del_timer(&port->clean_timer);
-	if (port)
-		rcu_read_unlock();
-	if (skb)
-		dev_kfree_skb(skb);
-	vnet_free_skbs(freeskbs);
-	dev->stats.tx_dropped++;
-	return NETDEV_TX_OK;
-}
-
-static void vnet_tx_timeout(struct net_device *dev)
-{
-	/* XXX Implement me XXX */
-}
-
-static int vnet_open(struct net_device *dev)
-{
-	netif_carrier_on(dev);
-	netif_tx_start_all_queues(dev);
-
-	return 0;
-}
-
-static int vnet_close(struct net_device *dev)
-{
-	netif_tx_stop_all_queues(dev);
-	netif_carrier_off(dev);
-
-	return 0;
-}
-
-static struct vnet_mcast_entry *__vnet_mc_find(struct vnet *vp, u8 *addr)
-{
-	struct vnet_mcast_entry *m;
-
-	for (m = vp->mcast_list; m; m = m->next) {
-		if (ether_addr_equal(m->addr, addr))
-			return m;
-	}
-	return NULL;
-}
-
-static void __update_mc_list(struct vnet *vp, struct net_device *dev)
-{
-	struct netdev_hw_addr *ha;
-
-	netdev_for_each_mc_addr(ha, dev) {
-		struct vnet_mcast_entry *m;
-
-		m = __vnet_mc_find(vp, ha->addr);
-		if (m) {
-			m->hit = 1;
-			continue;
-		}
-
-		if (!m) {
-			m = kzalloc(sizeof(*m), GFP_ATOMIC);
-			if (!m)
-				continue;
-			memcpy(m->addr, ha->addr, ETH_ALEN);
-			m->hit = 1;
-
-			m->next = vp->mcast_list;
-			vp->mcast_list = m;
-		}
-	}
-}
-
-static void __send_mc_list(struct vnet *vp, struct vnet_port *port)
-{
-	struct vio_net_mcast_info info;
-	struct vnet_mcast_entry *m, **pp;
-	int n_addrs;
-
-	memset(&info, 0, sizeof(info));
-
-	info.tag.type = VIO_TYPE_CTRL;
-	info.tag.stype = VIO_SUBTYPE_INFO;
-	info.tag.stype_env = VNET_MCAST_INFO;
-	info.tag.sid = vio_send_sid(&port->vio);
-	info.set = 1;
-
-	n_addrs = 0;
-	for (m = vp->mcast_list; m; m = m->next) {
-		if (m->sent)
-			continue;
-		m->sent = 1;
-		memcpy(&info.mcast_addr[n_addrs * ETH_ALEN],
-		       m->addr, ETH_ALEN);
-		if (++n_addrs == VNET_NUM_MCAST) {
-			info.count = n_addrs;
-
-			(void) vio_ldc_send(&port->vio, &info,
-					    sizeof(info));
-			n_addrs = 0;
-		}
-	}
-	if (n_addrs) {
-		info.count = n_addrs;
-		(void) vio_ldc_send(&port->vio, &info, sizeof(info));
-	}
-
-	info.set = 0;
-
-	n_addrs = 0;
-	pp = &vp->mcast_list;
-	while ((m = *pp) != NULL) {
-		if (m->hit) {
-			m->hit = 0;
-			pp = &m->next;
-			continue;
-		}
-
-		memcpy(&info.mcast_addr[n_addrs * ETH_ALEN],
-		       m->addr, ETH_ALEN);
-		if (++n_addrs == VNET_NUM_MCAST) {
-			info.count = n_addrs;
-			(void) vio_ldc_send(&port->vio, &info,
-					    sizeof(info));
-			n_addrs = 0;
-		}
-
-		*pp = m->next;
-		kfree(m);
-	}
-	if (n_addrs) {
-		info.count = n_addrs;
-		(void) vio_ldc_send(&port->vio, &info, sizeof(info));
-	}
-}
-
-static void vnet_set_rx_mode(struct net_device *dev)
-{
-	struct vnet *vp = netdev_priv(dev);
-	struct vnet_port *port;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(port, &vp->port_list, list) {
-
-		if (port->switch_port) {
-			__update_mc_list(vp, dev);
-			__send_mc_list(vp, port);
-			break;
-		}
-	}
-	rcu_read_unlock();
-}
-
-static int vnet_change_mtu(struct net_device *dev, int new_mtu)
-{
-	if (new_mtu < 68 || new_mtu > 65535)
-		return -EINVAL;
-
-	dev->mtu = new_mtu;
-	return 0;
-}
-
-static int vnet_set_mac_addr(struct net_device *dev, void *p)
-{
-	return -EINVAL;
-}
-
 static void vnet_get_drvinfo(struct net_device *dev,
 			     struct ethtool_drvinfo *info)
 {
@@ -1660,129 +81,21 @@ static const struct ethtool_ops vnet_ethtool_ops = {
 	.get_link		= ethtool_op_get_link,
 };
 
-static void vnet_port_free_tx_bufs(struct vnet_port *port)
-{
-	struct vio_dring_state *dr;
-	int i;
-
-	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-
-	if (dr->base == NULL)
-		return;
-
-	for (i = 0; i < VNET_TX_RING_SIZE; i++) {
-		struct vio_net_desc *d;
-		void *skb = port->tx_bufs[i].skb;
-
-		if (!skb)
-			continue;
-
-		d = vio_dring_entry(dr, i);
-
-		ldc_unmap(port->vio.lp,
-			  port->tx_bufs[i].cookies,
-			  port->tx_bufs[i].ncookies);
-		dev_kfree_skb(skb);
-		port->tx_bufs[i].skb = NULL;
-		d->hdr.state = VIO_DESC_FREE;
-	}
-	ldc_free_exp_dring(port->vio.lp, dr->base,
-			   (dr->entry_size * dr->num_entries),
-			   dr->cookies, dr->ncookies);
-	dr->base = NULL;
-	dr->entry_size = 0;
-	dr->num_entries = 0;
-	dr->pending = 0;
-	dr->ncookies = 0;
-}
-
-static void vnet_port_reset(struct vnet_port *port)
-{
-	del_timer(&port->clean_timer);
-	vnet_port_free_tx_bufs(port);
-	port->rmtu = 0;
-	port->tso = true;
-	port->tsolen = 0;
-}
-
-static int vnet_port_alloc_tx_ring(struct vnet_port *port)
-{
-	struct vio_dring_state *dr;
-	unsigned long len, elen;
-	int i, err, ncookies;
-	void *dring;
-
-	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
-
-	elen = sizeof(struct vio_net_desc) +
-	       sizeof(struct ldc_trans_cookie) * 2;
-	if (vio_version_after_eq(&port->vio, 1, 7))
-		elen += sizeof(struct vio_net_dext);
-	len = VNET_TX_RING_SIZE * elen;
-
-	ncookies = VIO_MAX_RING_COOKIES;
-	dring = ldc_alloc_exp_dring(port->vio.lp, len,
-				    dr->cookies, &ncookies,
-				    (LDC_MAP_SHADOW |
-				     LDC_MAP_DIRECT |
-				     LDC_MAP_RW));
-	if (IS_ERR(dring)) {
-		err = PTR_ERR(dring);
-		goto err_out;
-	}
-
-	dr->base = dring;
-	dr->entry_size = elen;
-	dr->num_entries = VNET_TX_RING_SIZE;
-	dr->prod = dr->cons = 0;
-	port->start_cons  = true; /* need an initial trigger */
-	dr->pending = VNET_TX_RING_SIZE;
-	dr->ncookies = ncookies;
-
-	for (i = 0; i < VNET_TX_RING_SIZE; ++i) {
-		struct vio_net_desc *d;
-
-		d = vio_dring_entry(dr, i);
-		d->hdr.state = VIO_DESC_FREE;
-	}
-	return 0;
-
-err_out:
-	vnet_port_free_tx_bufs(port);
-
-	return err;
-}
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static void vnet_poll_controller(struct net_device *dev)
-{
-	struct vnet *vp = netdev_priv(dev);
-	struct vnet_port *port;
-	unsigned long flags;
-
-	spin_lock_irqsave(&vp->lock, flags);
-	if (!list_empty(&vp->port_list)) {
-		port = list_entry(vp->port_list.next, struct vnet_port, list);
-		napi_schedule(&port->napi);
-	}
-	spin_unlock_irqrestore(&vp->lock, flags);
-}
-#endif
 static LIST_HEAD(vnet_list);
 static DEFINE_MUTEX(vnet_list_mutex);
 
 static const struct net_device_ops vnet_ops = {
-	.ndo_open		= vnet_open,
-	.ndo_stop		= vnet_close,
-	.ndo_set_rx_mode	= vnet_set_rx_mode,
-	.ndo_set_mac_address	= vnet_set_mac_addr,
+	.ndo_open		= sunvnet_open_common,
+	.ndo_stop		= sunvnet_close_common,
+	.ndo_set_rx_mode	= sunvnet_set_rx_mode_common,
+	.ndo_set_mac_address	= sunvnet_set_mac_addr_common,
 	.ndo_validate_addr	= eth_validate_addr,
-	.ndo_tx_timeout		= vnet_tx_timeout,
-	.ndo_change_mtu		= vnet_change_mtu,
-	.ndo_start_xmit		= vnet_start_xmit,
-	.ndo_select_queue	= vnet_select_queue,
+	.ndo_tx_timeout		= sunvnet_tx_timeout_common,
+	.ndo_change_mtu		= sunvnet_change_mtu_common,
+	.ndo_start_xmit		= sunvnet_start_xmit_common,
+	.ndo_select_queue	= sunvnet_select_queue_common,
 #ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= vnet_poll_controller,
+	.ndo_poll_controller	= sunvnet_poll_controller_common,
 #endif
 };
 
@@ -1908,15 +221,15 @@ static struct vnet *vnet_find_parent(struct mdesc_handle *hp,
 }
 
 static struct ldc_channel_config vnet_ldc_cfg = {
-	.event		= vnet_event,
+	.event		= sunvnet_event_common,
 	.mtu		= 64,
 	.mode		= LDC_MODE_UNRELIABLE,
 };
 
 static struct vio_driver_ops vnet_vio_ops = {
-	.send_attr		= vnet_send_attr,
-	.handle_attr		= vnet_handle_attr,
-	.handshake_complete	= vnet_handshake_complete,
+	.send_attr		= sunvnet_send_attr_common,
+	.handle_attr		= sunvnet_handle_attr_common,
+	.handshake_complete	= sunvnet_handshake_complete_common,
 };
 
 static void print_version(void)
@@ -1926,25 +239,6 @@ static void print_version(void)
 
 const char *remote_macaddr_prop = "remote-mac-address";
 
-static void
-vnet_port_add_txq(struct vnet_port *port)
-{
-	struct vnet *vp = port->vp;
-	int n;
-
-	n = vp->nports++;
-	n = n & (VNET_MAX_TXQS - 1);
-	port->q_index = n;
-	netif_tx_wake_queue(netdev_get_tx_queue(vp->dev, port->q_index));
-}
-
-static void
-vnet_port_rm_txq(struct vnet_port *port)
-{
-	port->vp->nports--;
-	netif_tx_stop_queue(netdev_get_tx_queue(port->vp->dev, port->q_index));
-}
-
 static int vnet_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
 {
 	struct mdesc_handle *hp;
@@ -1992,7 +286,8 @@ static int vnet_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
 	if (err)
 		goto err_out_free_port;
 
-	netif_napi_add(port->vp->dev, &port->napi, vnet_poll, NAPI_POLL_WEIGHT);
+	netif_napi_add(port->vp->dev, &port->napi, sunvnet_poll_common,
+		       NAPI_POLL_WEIGHT);
 
 	INIT_HLIST_NODE(&port->hash);
 	INIT_LIST_HEAD(&port->list);
@@ -2011,7 +306,7 @@ static int vnet_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
 		list_add_tail_rcu(&port->list, &vp->port_list);
 	hlist_add_head_rcu(&port->hash,
 			   &vp->port_hash[vnet_hashfn(port->raddr)]);
-	vnet_port_add_txq(port);
+	sunvnet_port_add_txq_common(port);
 	spin_unlock_irqrestore(&vp->lock, flags);
 
 	dev_set_drvdata(&vdev->dev, port);
@@ -2019,7 +314,7 @@ static int vnet_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
 	pr_info("%s: PORT ( remote-mac %pM%s )\n",
 		vp->dev->name, port->raddr, switch_port ? " switch-port" : "");
 
-	setup_timer(&port->clean_timer, vnet_clean_timer_expire,
+	setup_timer(&port->clean_timer, sunvnet_clean_timer_expire_common,
 		    (unsigned long)port);
 
 	napi_enable(&port->napi);
@@ -2052,9 +347,9 @@ static int vnet_port_remove(struct vio_dev *vdev)
 
 		synchronize_rcu();
 		del_timer_sync(&port->clean_timer);
-		vnet_port_rm_txq(port);
+		sunvnet_port_rm_txq_common(port);
 		netif_napi_del(&port->napi);
-		vnet_port_free_tx_bufs(port);
+		sunvnet_port_free_tx_bufs_common(port);
 		vio_ldc_free(&port->vio);
 
 		dev_set_drvdata(&vdev->dev, NULL);
diff --git a/drivers/net/ethernet/sun/sunvnet.h b/drivers/net/ethernet/sun/sunvnet.h
deleted file mode 100644
index 01ca781..0000000
--- a/drivers/net/ethernet/sun/sunvnet.h
+++ /dev/null
@@ -1,114 +0,0 @@
-#ifndef _SUNVNET_H
-#define _SUNVNET_H
-
-#include <linux/interrupt.h>
-
-#define DESC_NCOOKIES(entry_size)	\
-	((entry_size) - sizeof(struct vio_net_desc))
-
-/* length of time before we decide the hardware is borked,
- * and dev->tx_timeout() should be called to fix the problem
- */
-#define VNET_TX_TIMEOUT			(5 * HZ)
-
-/* length of time (or less) we expect pending descriptors to be marked
- * as VIO_DESC_DONE and skbs ready to be freed
- */
-#define	VNET_CLEAN_TIMEOUT		((HZ/100)+1)
-
-#define VNET_MAXPACKET			(65535ULL + ETH_HLEN + VLAN_HLEN)
-#define VNET_TX_RING_SIZE		512
-#define VNET_TX_WAKEUP_THRESH(dr)	((dr)->pending / 4)
-
-#define	VNET_MINTSO	 2048	/* VIO protocol's minimum TSO len */
-#define	VNET_MAXTSO	65535	/* VIO protocol's maximum TSO len */
-
-/* VNET packets are sent in buffers with the first 6 bytes skipped
- * so that after the ethernet header the IPv4/IPv6 headers are aligned
- * properly.
- */
-#define VNET_PACKET_SKIP		6
-
-#define VNET_MAXCOOKIES			(VNET_MAXPACKET/PAGE_SIZE + 1)
-
-struct vnet_tx_entry {
-	struct sk_buff		*skb;
-	unsigned int		ncookies;
-	struct ldc_trans_cookie	cookies[VNET_MAXCOOKIES];
-};
-
-struct vnet;
-struct vnet_port {
-	struct vio_driver_state	vio;
-
-	struct hlist_node	hash;
-	u8			raddr[ETH_ALEN];
-	unsigned		switch_port:1;
-	unsigned		tso:1;
-	unsigned		__pad:14;
-
-	struct vnet		*vp;
-
-	struct vnet_tx_entry	tx_bufs[VNET_TX_RING_SIZE];
-
-	struct list_head	list;
-
-	u32			stop_rx_idx;
-	bool			stop_rx;
-	bool			start_cons;
-
-	struct timer_list	clean_timer;
-
-	u64			rmtu;
-	u16			tsolen;
-
-	struct napi_struct	napi;
-	u32			napi_stop_idx;
-	bool			napi_resume;
-	int			rx_event;
-	u16			q_index;
-};
-
-static inline struct vnet_port *to_vnet_port(struct vio_driver_state *vio)
-{
-	return container_of(vio, struct vnet_port, vio);
-}
-
-#define VNET_PORT_HASH_SIZE	16
-#define VNET_PORT_HASH_MASK	(VNET_PORT_HASH_SIZE - 1)
-
-static inline unsigned int vnet_hashfn(u8 *mac)
-{
-	unsigned int val = mac[4] ^ mac[5];
-
-	return val & (VNET_PORT_HASH_MASK);
-}
-
-struct vnet_mcast_entry {
-	u8			addr[ETH_ALEN];
-	u8			sent;
-	u8			hit;
-	struct vnet_mcast_entry	*next;
-};
-
-struct vnet {
-	/* Protects port_list and port_hash.  */
-	spinlock_t		lock;
-
-	struct net_device	*dev;
-
-	u32			msg_enable;
-
-	struct list_head	port_list;
-
-	struct hlist_head	port_hash[VNET_PORT_HASH_SIZE];
-
-	struct vnet_mcast_entry	*mcast_list;
-
-	struct list_head	list;
-	u64			local_mac;
-
-	int			nports;
-};
-
-#endif /* _SUNVNET_H */
diff --git a/drivers/net/ethernet/sun/sunvnet_common.c b/drivers/net/ethernet/sun/sunvnet_common.c
new file mode 100644
index 0000000..49e85d0
--- /dev/null
+++ b/drivers/net/ethernet/sun/sunvnet_common.c
@@ -0,0 +1,1753 @@
+/* sunvnet.c: Sun LDOM Virtual Network Driver.
+ *
+ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/netdevice.h>
+#include <linux/ethtool.h>
+#include <linux/etherdevice.h>
+#include <linux/mutex.h>
+#include <linux/highmem.h>
+#include <linux/if_vlan.h>
+#define CREATE_TRACE_POINTS
+#include <trace/events/sunvnet.h>
+
+#if IS_ENABLED(CONFIG_IPV6)
+#include <linux/icmpv6.h>
+#endif
+
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/route.h>
+
+#include <asm/vio.h>
+#include <asm/ldc.h>
+
+#include "sunvnet_common.h"
+
+/* Heuristic for the number of times to exponentially backoff and
+ * retry sending an LDC trigger when EAGAIN is encountered
+ */
+#define	VNET_MAX_RETRIES	10
+
+static int __vnet_tx_trigger(struct vnet_port *port, u32 start);
+static void vnet_port_reset(struct vnet_port *port);
+
+static inline u32 vnet_tx_dring_avail(struct vio_dring_state *dr)
+{
+	return vio_dring_avail(dr, VNET_TX_RING_SIZE);
+}
+
+static int vnet_handle_unknown(struct vnet_port *port, void *arg)
+{
+	struct vio_msg_tag *pkt = arg;
+
+	pr_err("Received unknown msg [%02x:%02x:%04x:%08x]\n",
+	       pkt->type, pkt->stype, pkt->stype_env, pkt->sid);
+	pr_err("Resetting connection\n");
+
+	ldc_disconnect(port->vio.lp);
+
+	return -ECONNRESET;
+}
+
+static int vnet_port_alloc_tx_ring(struct vnet_port *port);
+
+int sunvnet_send_attr_common(struct vio_driver_state *vio)
+{
+	struct vnet_port *port = to_vnet_port(vio);
+	struct net_device *dev = port->vp->dev;
+	struct vio_net_attr_info pkt;
+	int framelen = ETH_FRAME_LEN;
+	int i, err;
+
+	err = vnet_port_alloc_tx_ring(to_vnet_port(vio));
+	if (err)
+		return err;
+
+	memset(&pkt, 0, sizeof(pkt));
+	pkt.tag.type = VIO_TYPE_CTRL;
+	pkt.tag.stype = VIO_SUBTYPE_INFO;
+	pkt.tag.stype_env = VIO_ATTR_INFO;
+	pkt.tag.sid = vio_send_sid(vio);
+	if (vio_version_before(vio, 1, 2))
+		pkt.xfer_mode = VIO_DRING_MODE;
+	else
+		pkt.xfer_mode = VIO_NEW_DRING_MODE;
+	pkt.addr_type = VNET_ADDR_ETHERMAC;
+	pkt.ack_freq = 0;
+	for (i = 0; i < 6; i++)
+		pkt.addr |= (u64)dev->dev_addr[i] << ((5 - i) * 8);
+	if (vio_version_after(vio, 1, 3)) {
+		if (port->rmtu) {
+			port->rmtu = min(VNET_MAXPACKET, port->rmtu);
+			pkt.mtu = port->rmtu;
+		} else {
+			port->rmtu = VNET_MAXPACKET;
+			pkt.mtu = port->rmtu;
+		}
+		if (vio_version_after_eq(vio, 1, 6))
+			pkt.options = VIO_TX_DRING;
+	} else if (vio_version_before(vio, 1, 3)) {
+		pkt.mtu = framelen;
+	} else { /* v1.3 */
+		pkt.mtu = framelen + VLAN_HLEN;
+	}
+
+	pkt.cflags = 0;
+	if (vio_version_after_eq(vio, 1, 7) && port->tso) {
+		pkt.cflags |= VNET_LSO_IPV4_CAPAB;
+		if (!port->tsolen)
+			port->tsolen = VNET_MAXTSO;
+		pkt.ipv4_lso_maxlen = port->tsolen;
+	}
+
+	pkt.plnk_updt = PHYSLINK_UPDATE_NONE;
+
+	viodbg(HS, "SEND NET ATTR xmode[0x%x] atype[0x%x] addr[%llx] "
+	       "ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] mtu[%llu] "
+	       "cflags[0x%04x] lso_max[%u]\n",
+	       pkt.xfer_mode, pkt.addr_type,
+	       (unsigned long long)pkt.addr,
+	       pkt.ack_freq, pkt.plnk_updt, pkt.options,
+	       (unsigned long long)pkt.mtu, pkt.cflags, pkt.ipv4_lso_maxlen);
+
+
+	return vio_ldc_send(vio, &pkt, sizeof(pkt));
+}
+EXPORT_SYMBOL_GPL(sunvnet_send_attr_common);
+
+static int handle_attr_info(struct vio_driver_state *vio,
+			    struct vio_net_attr_info *pkt)
+{
+	struct vnet_port *port = to_vnet_port(vio);
+	u64	localmtu;
+	u8	xfer_mode;
+
+	viodbg(HS, "GOT NET ATTR xmode[0x%x] atype[0x%x] addr[%llx] "
+	       "ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] mtu[%llu] "
+	       " (rmtu[%llu]) cflags[0x%04x] lso_max[%u]\n",
+	       pkt->xfer_mode, pkt->addr_type,
+	       (unsigned long long)pkt->addr,
+	       pkt->ack_freq, pkt->plnk_updt, pkt->options,
+	       (unsigned long long)pkt->mtu, port->rmtu, pkt->cflags,
+	       pkt->ipv4_lso_maxlen);
+
+	pkt->tag.sid = vio_send_sid(vio);
+
+	xfer_mode = pkt->xfer_mode;
+	/* for version < 1.2, VIO_DRING_MODE = 0x3 and no bitmask */
+	if (vio_version_before(vio, 1, 2) && xfer_mode == VIO_DRING_MODE)
+		xfer_mode = VIO_NEW_DRING_MODE;
+
+	/* MTU negotiation:
+	 *	< v1.3 - ETH_FRAME_LEN exactly
+	 *	> v1.3 - MIN(pkt.mtu, VNET_MAXPACKET, port->rmtu) and change
+	 *			pkt->mtu for ACK
+	 *	= v1.3 - ETH_FRAME_LEN + VLAN_HLEN exactly
+	 */
+	if (vio_version_before(vio, 1, 3)) {
+		localmtu = ETH_FRAME_LEN;
+	} else if (vio_version_after(vio, 1, 3)) {
+		localmtu = port->rmtu ? port->rmtu : VNET_MAXPACKET;
+		localmtu = min(pkt->mtu, localmtu);
+		pkt->mtu = localmtu;
+	} else { /* v1.3 */
+		localmtu = ETH_FRAME_LEN + VLAN_HLEN;
+	}
+	port->rmtu = localmtu;
+
+	/* LSO negotiation */
+	if (vio_version_after_eq(vio, 1, 7))
+		port->tso &= !!(pkt->cflags & VNET_LSO_IPV4_CAPAB);
+	else
+		port->tso = false;
+	if (port->tso) {
+		if (!port->tsolen)
+			port->tsolen = VNET_MAXTSO;
+		port->tsolen = min(port->tsolen, pkt->ipv4_lso_maxlen);
+		if (port->tsolen < VNET_MINTSO) {
+			port->tso = false;
+			port->tsolen = 0;
+			pkt->cflags &= ~VNET_LSO_IPV4_CAPAB;
+		}
+		pkt->ipv4_lso_maxlen = port->tsolen;
+	} else {
+		pkt->cflags &= ~VNET_LSO_IPV4_CAPAB;
+		pkt->ipv4_lso_maxlen = 0;
+	}
+
+	/* for version >= 1.6, ACK packet mode we support */
+	if (vio_version_after_eq(vio, 1, 6)) {
+		pkt->xfer_mode = VIO_NEW_DRING_MODE;
+		pkt->options = VIO_TX_DRING;
+	}
+
+	if (!(xfer_mode | VIO_NEW_DRING_MODE) ||
+	    pkt->addr_type != VNET_ADDR_ETHERMAC ||
+	    pkt->mtu != localmtu) {
+		viodbg(HS, "SEND NET ATTR NACK\n");
+
+		pkt->tag.stype = VIO_SUBTYPE_NACK;
+
+		(void) vio_ldc_send(vio, pkt, sizeof(*pkt));
+
+		return -ECONNRESET;
+	} else {
+		viodbg(HS, "SEND NET ATTR ACK xmode[0x%x] atype[0x%x] "
+		       "addr[%llx] ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] "
+		       "mtu[%llu] (rmtu[%llu]) cflags[0x%04x] lso_max[%u]\n",
+		       pkt->xfer_mode, pkt->addr_type,
+		       (unsigned long long)pkt->addr,
+		       pkt->ack_freq, pkt->plnk_updt, pkt->options,
+		       (unsigned long long)pkt->mtu, port->rmtu, pkt->cflags,
+		       pkt->ipv4_lso_maxlen);
+
+		pkt->tag.stype = VIO_SUBTYPE_ACK;
+
+		return vio_ldc_send(vio, pkt, sizeof(*pkt));
+	}
+
+}
+
+static int handle_attr_ack(struct vio_driver_state *vio,
+			   struct vio_net_attr_info *pkt)
+{
+	viodbg(HS, "GOT NET ATTR ACK\n");
+
+	return 0;
+}
+
+static int handle_attr_nack(struct vio_driver_state *vio,
+			    struct vio_net_attr_info *pkt)
+{
+	viodbg(HS, "GOT NET ATTR NACK\n");
+
+	return -ECONNRESET;
+}
+
+int sunvnet_handle_attr_common(struct vio_driver_state *vio, void *arg)
+{
+	struct vio_net_attr_info *pkt = arg;
+
+	switch (pkt->tag.stype) {
+	case VIO_SUBTYPE_INFO:
+		return handle_attr_info(vio, pkt);
+
+	case VIO_SUBTYPE_ACK:
+		return handle_attr_ack(vio, pkt);
+
+	case VIO_SUBTYPE_NACK:
+		return handle_attr_nack(vio, pkt);
+
+	default:
+		return -ECONNRESET;
+	}
+}
+EXPORT_SYMBOL_GPL(sunvnet_handle_attr_common);
+
+void sunvnet_handshake_complete_common(struct vio_driver_state *vio)
+{
+	struct vio_dring_state *dr;
+
+	dr = &vio->drings[VIO_DRIVER_RX_RING];
+	dr->snd_nxt = dr->rcv_nxt = 1;
+
+	dr = &vio->drings[VIO_DRIVER_TX_RING];
+	dr->snd_nxt = dr->rcv_nxt = 1;
+}
+EXPORT_SYMBOL_GPL(sunvnet_handshake_complete_common);
+
+/* The hypervisor interface that implements copying to/from imported
+ * memory from another domain requires that copies are done to 8-byte
+ * aligned buffers, and that the lengths of such copies are also 8-byte
+ * multiples.
+ *
+ * So we align skb->data to an 8-byte multiple and pad-out the data
+ * area so we can round the copy length up to the next multiple of
+ * 8 for the copy.
+ *
+ * The transmitter puts the actual start of the packet 6 bytes into
+ * the buffer it sends over, so that the IP headers after the ethernet
+ * header are aligned properly.  These 6 bytes are not in the descriptor
+ * length, they are simply implied.  This offset is represented using
+ * the VNET_PACKET_SKIP macro.
+ */
+static struct sk_buff *alloc_and_align_skb(struct net_device *dev,
+					   unsigned int len)
+{
+	struct sk_buff *skb = netdev_alloc_skb(dev, len+VNET_PACKET_SKIP+8+8);
+	unsigned long addr, off;
+
+	if (unlikely(!skb))
+		return NULL;
+
+	addr = (unsigned long) skb->data;
+	off = ((addr + 7UL) & ~7UL) - addr;
+	if (off)
+		skb_reserve(skb, off);
+
+	return skb;
+}
+
+static inline void vnet_fullcsum(struct sk_buff *skb)
+{
+	struct iphdr *iph = ip_hdr(skb);
+	int offset = skb_transport_offset(skb);
+
+	if (skb->protocol != htons(ETH_P_IP))
+		return;
+	if (iph->protocol != IPPROTO_TCP &&
+	    iph->protocol != IPPROTO_UDP)
+		return;
+	skb->ip_summed = CHECKSUM_NONE;
+	skb->csum_level = 1;
+	skb->csum = 0;
+	if (iph->protocol == IPPROTO_TCP) {
+		struct tcphdr *ptcp = tcp_hdr(skb);
+
+		ptcp->check = 0;
+		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
+		ptcp->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
+						skb->len - offset, IPPROTO_TCP,
+						skb->csum);
+	} else if (iph->protocol == IPPROTO_UDP) {
+		struct udphdr *pudp = udp_hdr(skb);
+
+		pudp->check = 0;
+		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);
+		pudp->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
+						skb->len - offset, IPPROTO_UDP,
+						skb->csum);
+	}
+}
+
+static int vnet_rx_one(struct vnet_port *port, struct vio_net_desc *desc)
+{
+	struct net_device *dev = port->vp->dev;
+	unsigned int len = desc->size;
+	unsigned int copy_len;
+	struct sk_buff *skb;
+	int maxlen;
+	int err;
+
+	err = -EMSGSIZE;
+	if (port->tso && port->tsolen > port->rmtu)
+		maxlen = port->tsolen;
+	else
+		maxlen = port->rmtu;
+	if (unlikely(len < ETH_ZLEN || len > maxlen)) {
+		dev->stats.rx_length_errors++;
+		goto out_dropped;
+	}
+
+	skb = alloc_and_align_skb(dev, len);
+	err = -ENOMEM;
+	if (unlikely(!skb)) {
+		dev->stats.rx_missed_errors++;
+		goto out_dropped;
+	}
+
+	copy_len = (len + VNET_PACKET_SKIP + 7U) & ~7U;
+	skb_put(skb, copy_len);
+	err = ldc_copy(port->vio.lp, LDC_COPY_IN,
+		       skb->data, copy_len, 0,
+		       desc->cookies, desc->ncookies);
+	if (unlikely(err < 0)) {
+		dev->stats.rx_frame_errors++;
+		goto out_free_skb;
+	}
+
+	skb_pull(skb, VNET_PACKET_SKIP);
+	skb_trim(skb, len);
+	skb->protocol = eth_type_trans(skb, dev);
+
+	if (vio_version_after_eq(&port->vio, 1, 8)) {
+		struct vio_net_dext *dext = vio_net_ext(desc);
+
+		skb_reset_network_header(skb);
+
+		if (dext->flags & VNET_PKT_HCK_IPV4_HDRCKSUM) {
+			if (skb->protocol == ETH_P_IP) {
+				struct iphdr *iph = ip_hdr(skb);
+
+				iph->check = 0;
+				ip_send_check(iph);
+			}
+		}
+		if ((dext->flags & VNET_PKT_HCK_FULLCKSUM) &&
+		    skb->ip_summed == CHECKSUM_NONE) {
+			if (skb->protocol == htons(ETH_P_IP)) {
+				struct iphdr *iph = ip_hdr(skb);
+				int ihl = iph->ihl * 4;
+
+				skb_reset_transport_header(skb);
+				skb_set_transport_header(skb, ihl);
+				vnet_fullcsum(skb);
+			}
+		}
+		if (dext->flags & VNET_PKT_HCK_IPV4_HDRCKSUM_OK) {
+			skb->ip_summed = CHECKSUM_PARTIAL;
+			skb->csum_level = 0;
+			if (dext->flags & VNET_PKT_HCK_FULLCKSUM_OK)
+				skb->csum_level = 1;
+		}
+	}
+
+	skb->ip_summed = port->switch_port ? CHECKSUM_NONE : CHECKSUM_PARTIAL;
+
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += len;
+	napi_gro_receive(&port->napi, skb);
+	return 0;
+
+out_free_skb:
+	kfree_skb(skb);
+
+out_dropped:
+	dev->stats.rx_dropped++;
+	return err;
+}
+
+static int vnet_send_ack(struct vnet_port *port, struct vio_dring_state *dr,
+			 u32 start, u32 end, u8 vio_dring_state)
+{
+	struct vio_dring_data hdr = {
+		.tag = {
+			.type		= VIO_TYPE_DATA,
+			.stype		= VIO_SUBTYPE_ACK,
+			.stype_env	= VIO_DRING_DATA,
+			.sid		= vio_send_sid(&port->vio),
+		},
+		.dring_ident		= dr->ident,
+		.start_idx		= start,
+		.end_idx		= end,
+		.state			= vio_dring_state,
+	};
+	int err, delay;
+	int retries = 0;
+
+	hdr.seq = dr->snd_nxt;
+	delay = 1;
+	do {
+		err = vio_ldc_send(&port->vio, &hdr, sizeof(hdr));
+		if (err > 0) {
+			dr->snd_nxt++;
+			break;
+		}
+		udelay(delay);
+		if ((delay <<= 1) > 128)
+			delay = 128;
+		if (retries++ > VNET_MAX_RETRIES) {
+			pr_info("ECONNRESET %x:%x:%x:%x:%x:%x\n",
+				port->raddr[0], port->raddr[1],
+				port->raddr[2], port->raddr[3],
+				port->raddr[4], port->raddr[5]);
+			break;
+		}
+	} while (err == -EAGAIN);
+
+	if (err <= 0 && vio_dring_state == VIO_DRING_STOPPED) {
+		port->stop_rx_idx = end;
+		port->stop_rx = true;
+	} else {
+		port->stop_rx_idx = 0;
+		port->stop_rx = false;
+	}
+
+	return err;
+}
+
+static struct vio_net_desc *get_rx_desc(struct vnet_port *port,
+					struct vio_dring_state *dr,
+					u32 index)
+{
+	struct vio_net_desc *desc = port->vio.desc_buf;
+	int err;
+
+	err = ldc_get_dring_entry(port->vio.lp, desc, dr->entry_size,
+				  (index * dr->entry_size),
+				  dr->cookies, dr->ncookies);
+	if (err < 0)
+		return ERR_PTR(err);
+
+	return desc;
+}
+
+static int put_rx_desc(struct vnet_port *port,
+		       struct vio_dring_state *dr,
+		       struct vio_net_desc *desc,
+		       u32 index)
+{
+	int err;
+
+	err = ldc_put_dring_entry(port->vio.lp, desc, dr->entry_size,
+				  (index * dr->entry_size),
+				  dr->cookies, dr->ncookies);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+static int vnet_walk_rx_one(struct vnet_port *port,
+			    struct vio_dring_state *dr,
+			    u32 index, int *needs_ack)
+{
+	struct vio_net_desc *desc = get_rx_desc(port, dr, index);
+	struct vio_driver_state *vio = &port->vio;
+	int err;
+
+	BUG_ON(desc == NULL);
+	if (IS_ERR(desc))
+		return PTR_ERR(desc);
+
+	if (desc->hdr.state != VIO_DESC_READY)
+		return 1;
+
+	dma_rmb();
+
+	viodbg(DATA, "vio_walk_rx_one desc[%02x:%02x:%08x:%08x:%llx:%llx]\n",
+	       desc->hdr.state, desc->hdr.ack,
+	       desc->size, desc->ncookies,
+	       desc->cookies[0].cookie_addr,
+	       desc->cookies[0].cookie_size);
+
+	err = vnet_rx_one(port, desc);
+	if (err == -ECONNRESET)
+		return err;
+	trace_vnet_rx_one(port->vio._local_sid, port->vio._peer_sid,
+			  index, desc->hdr.ack);
+	desc->hdr.state = VIO_DESC_DONE;
+	err = put_rx_desc(port, dr, desc, index);
+	if (err < 0)
+		return err;
+	*needs_ack = desc->hdr.ack;
+	return 0;
+}
+
+static int vnet_walk_rx(struct vnet_port *port, struct vio_dring_state *dr,
+			u32 start, u32 end, int *npkts, int budget)
+{
+	struct vio_driver_state *vio = &port->vio;
+	int ack_start = -1, ack_end = -1;
+	bool send_ack = true;
+
+	end = (end == (u32) -1) ? vio_dring_prev(dr, start)
+				: vio_dring_next(dr, end);
+
+	viodbg(DATA, "vnet_walk_rx start[%08x] end[%08x]\n", start, end);
+
+	while (start != end) {
+		int ack = 0, err = vnet_walk_rx_one(port, dr, start, &ack);
+		if (err == -ECONNRESET)
+			return err;
+		if (err != 0)
+			break;
+		(*npkts)++;
+		if (ack_start == -1)
+			ack_start = start;
+		ack_end = start;
+		start = vio_dring_next(dr, start);
+		if (ack && start != end) {
+			err = vnet_send_ack(port, dr, ack_start, ack_end,
+					    VIO_DRING_ACTIVE);
+			if (err == -ECONNRESET)
+				return err;
+			ack_start = -1;
+		}
+		if ((*npkts) >= budget) {
+			send_ack = false;
+			break;
+		}
+	}
+	if (unlikely(ack_start == -1))
+		ack_start = ack_end = vio_dring_prev(dr, start);
+	if (send_ack) {
+		port->napi_resume = false;
+		trace_vnet_tx_send_stopped_ack(port->vio._local_sid,
+					       port->vio._peer_sid,
+					       ack_end, *npkts);
+		return vnet_send_ack(port, dr, ack_start, ack_end,
+				     VIO_DRING_STOPPED);
+	} else  {
+		trace_vnet_tx_defer_stopped_ack(port->vio._local_sid,
+						port->vio._peer_sid,
+						ack_end, *npkts);
+		port->napi_resume = true;
+		port->napi_stop_idx = ack_end;
+		return 1;
+	}
+}
+
+static int vnet_rx(struct vnet_port *port, void *msgbuf, int *npkts,
+		   int budget)
+{
+	struct vio_dring_data *pkt = msgbuf;
+	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_RX_RING];
+	struct vio_driver_state *vio = &port->vio;
+
+	viodbg(DATA, "vnet_rx stype_env[%04x] seq[%016llx] rcv_nxt[%016llx]\n",
+	       pkt->tag.stype_env, pkt->seq, dr->rcv_nxt);
+
+	if (unlikely(pkt->tag.stype_env != VIO_DRING_DATA))
+		return 0;
+	if (unlikely(pkt->seq != dr->rcv_nxt)) {
+		pr_err("RX out of sequence seq[0x%llx] rcv_nxt[0x%llx]\n",
+		       pkt->seq, dr->rcv_nxt);
+		return 0;
+	}
+
+	if (!port->napi_resume)
+		dr->rcv_nxt++;
+
+	/* XXX Validate pkt->start_idx and pkt->end_idx XXX */
+
+	return vnet_walk_rx(port, dr, pkt->start_idx, pkt->end_idx,
+			    npkts, budget);
+}
+
+static int idx_is_pending(struct vio_dring_state *dr, u32 end)
+{
+	u32 idx = dr->cons;
+	int found = 0;
+
+	while (idx != dr->prod) {
+		if (idx == end) {
+			found = 1;
+			break;
+		}
+		idx = vio_dring_next(dr, idx);
+	}
+	return found;
+}
+
+static int vnet_ack(struct vnet_port *port, void *msgbuf)
+{
+	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+	struct vio_dring_data *pkt = msgbuf;
+	struct net_device *dev;
+	struct vnet *vp;
+	u32 end;
+	struct vio_net_desc *desc;
+	struct netdev_queue *txq;
+
+	if (unlikely(pkt->tag.stype_env != VIO_DRING_DATA))
+		return 0;
+
+	end = pkt->end_idx;
+	vp = port->vp;
+	dev = vp->dev;
+	netif_tx_lock(dev);
+	if (unlikely(!idx_is_pending(dr, end))) {
+		netif_tx_unlock(dev);
+		return 0;
+	}
+
+	/* sync for race conditions with vnet_start_xmit() and tell xmit it
+	 * is time to send a trigger.
+	 */
+	trace_vnet_rx_stopped_ack(port->vio._local_sid,
+				  port->vio._peer_sid, end);
+	dr->cons = vio_dring_next(dr, end);
+	desc = vio_dring_entry(dr, dr->cons);
+	if (desc->hdr.state == VIO_DESC_READY && !port->start_cons) {
+		/* vnet_start_xmit() just populated this dring but missed
+		 * sending the "start" LDC message to the consumer.
+		 * Send a "start" trigger on its behalf.
+		 */
+		if (__vnet_tx_trigger(port, dr->cons) > 0)
+			port->start_cons = false;
+		else
+			port->start_cons = true;
+	} else {
+		port->start_cons = true;
+	}
+	netif_tx_unlock(dev);
+
+	txq = netdev_get_tx_queue(dev, port->q_index);
+	if (unlikely(netif_tx_queue_stopped(txq) &&
+		     vnet_tx_dring_avail(dr) >= VNET_TX_WAKEUP_THRESH(dr)))
+		return 1;
+
+	return 0;
+}
+
+static int vnet_nack(struct vnet_port *port, void *msgbuf)
+{
+	/* XXX just reset or similar XXX */
+	return 0;
+}
+
+static int handle_mcast(struct vnet_port *port, void *msgbuf)
+{
+	struct vio_net_mcast_info *pkt = msgbuf;
+
+	if (pkt->tag.stype != VIO_SUBTYPE_ACK)
+		pr_err("%s: Got unexpected MCAST reply [%02x:%02x:%04x:%08x]\n",
+		       port->vp->dev->name,
+		       pkt->tag.type,
+		       pkt->tag.stype,
+		       pkt->tag.stype_env,
+		       pkt->tag.sid);
+
+	return 0;
+}
+
+/* Got back a STOPPED LDC message on port. If the queue is stopped,
+ * wake it up so that we'll send out another START message at the
+ * next TX.
+ */
+static void maybe_tx_wakeup(struct vnet_port *port)
+{
+	struct netdev_queue *txq;
+
+	txq = netdev_get_tx_queue(port->vp->dev, port->q_index);
+	__netif_tx_lock(txq, smp_processor_id());
+	if (likely(netif_tx_queue_stopped(txq))) {
+		struct vio_dring_state *dr;
+
+		dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+		netif_tx_wake_queue(txq);
+	}
+	__netif_tx_unlock(txq);
+}
+
+static inline bool port_is_up(struct vnet_port *vnet)
+{
+	struct vio_driver_state *vio = &vnet->vio;
+
+	return !!(vio->hs_state & VIO_HS_COMPLETE);
+}
+
+static int vnet_event_napi(struct vnet_port *port, int budget)
+{
+	struct vio_driver_state *vio = &port->vio;
+	int tx_wakeup, err;
+	int npkts = 0;
+	int event = (port->rx_event & LDC_EVENT_RESET);
+
+ldc_ctrl:
+	if (unlikely(event == LDC_EVENT_RESET ||
+		     event == LDC_EVENT_UP)) {
+		vio_link_state_change(vio, event);
+
+		if (event == LDC_EVENT_RESET) {
+			vnet_port_reset(port);
+			vio_port_up(vio);
+		}
+		port->rx_event = 0;
+		return 0;
+	}
+	/* We may have multiple LDC events in rx_event. Unroll send_events() */
+	event = (port->rx_event & LDC_EVENT_UP);
+	port->rx_event &= ~(LDC_EVENT_RESET|LDC_EVENT_UP);
+	if (event == LDC_EVENT_UP)
+		goto ldc_ctrl;
+	event = port->rx_event;
+	if (!(event & LDC_EVENT_DATA_READY))
+		return 0;
+
+	/* we dont expect any other bits than RESET, UP, DATA_READY */
+	BUG_ON(event != LDC_EVENT_DATA_READY);
+
+	tx_wakeup = err = 0;
+	while (1) {
+		union {
+			struct vio_msg_tag tag;
+			u64 raw[8];
+		} msgbuf;
+
+		if (port->napi_resume) {
+			struct vio_dring_data *pkt =
+				(struct vio_dring_data *)&msgbuf;
+			struct vio_dring_state *dr =
+				&port->vio.drings[VIO_DRIVER_RX_RING];
+
+			pkt->tag.type = VIO_TYPE_DATA;
+			pkt->tag.stype = VIO_SUBTYPE_INFO;
+			pkt->tag.stype_env = VIO_DRING_DATA;
+			pkt->seq = dr->rcv_nxt;
+			pkt->start_idx = vio_dring_next(dr, port->napi_stop_idx);
+			pkt->end_idx = -1;
+			goto napi_resume;
+		}
+		err = ldc_read(vio->lp, &msgbuf, sizeof(msgbuf));
+		if (unlikely(err < 0)) {
+			if (err == -ECONNRESET)
+				vio_conn_reset(vio);
+			break;
+		}
+		if (err == 0)
+			break;
+		viodbg(DATA, "TAG [%02x:%02x:%04x:%08x]\n",
+		       msgbuf.tag.type,
+		       msgbuf.tag.stype,
+		       msgbuf.tag.stype_env,
+		       msgbuf.tag.sid);
+		err = vio_validate_sid(vio, &msgbuf.tag);
+		if (err < 0)
+			break;
+napi_resume:
+		if (likely(msgbuf.tag.type == VIO_TYPE_DATA)) {
+			if (msgbuf.tag.stype == VIO_SUBTYPE_INFO) {
+				if (!port_is_up(port)) {
+					/* failures like handshake_failure()
+					 * may have cleaned up dring, but
+					 * NAPI polling may bring us here.
+					 */
+					err = -ECONNRESET;
+					break;
+				}
+				err = vnet_rx(port, &msgbuf, &npkts, budget);
+				if (npkts >= budget)
+					break;
+				if (npkts == 0)
+					break;
+			} else if (msgbuf.tag.stype == VIO_SUBTYPE_ACK) {
+				err = vnet_ack(port, &msgbuf);
+				if (err > 0)
+					tx_wakeup |= err;
+			} else if (msgbuf.tag.stype == VIO_SUBTYPE_NACK) {
+				err = vnet_nack(port, &msgbuf);
+			}
+		} else if (msgbuf.tag.type == VIO_TYPE_CTRL) {
+			if (msgbuf.tag.stype_env == VNET_MCAST_INFO)
+				err = handle_mcast(port, &msgbuf);
+			else
+				err = vio_control_pkt_engine(vio, &msgbuf);
+			if (err)
+				break;
+		} else {
+			err = vnet_handle_unknown(port, &msgbuf);
+		}
+		if (err == -ECONNRESET)
+			break;
+	}
+	if (unlikely(tx_wakeup && err != -ECONNRESET))
+		maybe_tx_wakeup(port);
+	return npkts;
+}
+
+int sunvnet_poll_common(struct napi_struct *napi, int budget)
+{
+	struct vnet_port *port = container_of(napi, struct vnet_port, napi);
+	struct vio_driver_state *vio = &port->vio;
+	int processed = vnet_event_napi(port, budget);
+
+	if (processed < budget) {
+		napi_complete(napi);
+		port->rx_event &= ~LDC_EVENT_DATA_READY;
+		vio_set_intr(vio->vdev->rx_ino, HV_INTR_ENABLED);
+	}
+	return processed;
+}
+EXPORT_SYMBOL_GPL(sunvnet_poll_common);
+
+void sunvnet_event_common(void *arg, int event)
+{
+	struct vnet_port *port = arg;
+	struct vio_driver_state *vio = &port->vio;
+
+	port->rx_event |= event;
+	vio_set_intr(vio->vdev->rx_ino, HV_INTR_DISABLED);
+	napi_schedule(&port->napi);
+
+}
+EXPORT_SYMBOL_GPL(sunvnet_event_common);
+
+static int __vnet_tx_trigger(struct vnet_port *port, u32 start)
+{
+	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+	struct vio_dring_data hdr = {
+		.tag = {
+			.type		= VIO_TYPE_DATA,
+			.stype		= VIO_SUBTYPE_INFO,
+			.stype_env	= VIO_DRING_DATA,
+			.sid		= vio_send_sid(&port->vio),
+		},
+		.dring_ident		= dr->ident,
+		.start_idx		= start,
+		.end_idx		= (u32) -1,
+	};
+	int err, delay;
+	int retries = 0;
+
+	if (port->stop_rx) {
+		trace_vnet_tx_pending_stopped_ack(port->vio._local_sid,
+						  port->vio._peer_sid,
+						  port->stop_rx_idx, -1);
+		err = vnet_send_ack(port,
+				    &port->vio.drings[VIO_DRIVER_RX_RING],
+				    port->stop_rx_idx, -1,
+				    VIO_DRING_STOPPED);
+		if (err <= 0)
+			return err;
+	}
+
+	hdr.seq = dr->snd_nxt;
+	delay = 1;
+	do {
+		err = vio_ldc_send(&port->vio, &hdr, sizeof(hdr));
+		if (err > 0) {
+			dr->snd_nxt++;
+			break;
+		}
+		udelay(delay);
+		if ((delay <<= 1) > 128)
+			delay = 128;
+		if (retries++ > VNET_MAX_RETRIES)
+			break;
+	} while (err == -EAGAIN);
+	trace_vnet_tx_trigger(port->vio._local_sid,
+			      port->vio._peer_sid, start, err);
+
+	return err;
+}
+
+static struct vnet_port *__tx_port_find(struct vnet *vp, struct sk_buff *skb)
+{
+	unsigned int hash = vnet_hashfn(skb->data);
+	struct hlist_head *hp = &vp->port_hash[hash];
+	struct vnet_port *port;
+
+	hlist_for_each_entry_rcu(port, hp, hash) {
+		if (!port_is_up(port))
+			continue;
+		if (ether_addr_equal(port->raddr, skb->data))
+			return port;
+	}
+	list_for_each_entry_rcu(port, &vp->port_list, list) {
+		if (!port->switch_port)
+			continue;
+		if (!port_is_up(port))
+			continue;
+		return port;
+	}
+	return NULL;
+}
+
+static struct sk_buff *vnet_clean_tx_ring(struct vnet_port *port,
+					  unsigned *pending)
+{
+	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+	struct sk_buff *skb = NULL;
+	int i, txi;
+
+	*pending = 0;
+
+	txi = dr->prod;
+	for (i = 0; i < VNET_TX_RING_SIZE; ++i) {
+		struct vio_net_desc *d;
+
+		--txi;
+		if (txi < 0)
+			txi = VNET_TX_RING_SIZE-1;
+
+		d = vio_dring_entry(dr, txi);
+
+		if (d->hdr.state == VIO_DESC_READY) {
+			(*pending)++;
+			continue;
+		}
+		if (port->tx_bufs[txi].skb) {
+			if (d->hdr.state != VIO_DESC_DONE)
+				pr_notice("invalid ring buffer state %d\n",
+					  d->hdr.state);
+			BUG_ON(port->tx_bufs[txi].skb->next);
+
+			port->tx_bufs[txi].skb->next = skb;
+			skb = port->tx_bufs[txi].skb;
+			port->tx_bufs[txi].skb = NULL;
+
+			ldc_unmap(port->vio.lp,
+				  port->tx_bufs[txi].cookies,
+				  port->tx_bufs[txi].ncookies);
+		} else if (d->hdr.state == VIO_DESC_FREE)
+			break;
+		d->hdr.state = VIO_DESC_FREE;
+	}
+	return skb;
+}
+
+static inline void vnet_free_skbs(struct sk_buff *skb)
+{
+	struct sk_buff *next;
+
+	while (skb) {
+		next = skb->next;
+		skb->next = NULL;
+		dev_kfree_skb(skb);
+		skb = next;
+	}
+}
+
+void sunvnet_clean_timer_expire_common(unsigned long port0)
+{
+	struct vnet_port *port = (struct vnet_port *)port0;
+	struct sk_buff *freeskbs;
+	unsigned pending;
+
+	netif_tx_lock(port->vp->dev);
+	freeskbs = vnet_clean_tx_ring(port, &pending);
+	netif_tx_unlock(port->vp->dev);
+
+	vnet_free_skbs(freeskbs);
+
+	if (pending)
+		(void)mod_timer(&port->clean_timer,
+				jiffies + VNET_CLEAN_TIMEOUT);
+	 else
+		del_timer(&port->clean_timer);
+}
+EXPORT_SYMBOL_GPL(sunvnet_clean_timer_expire_common);
+
+static inline int vnet_skb_map(struct ldc_channel *lp, struct sk_buff *skb,
+			       struct ldc_trans_cookie *cookies, int ncookies,
+			       unsigned int map_perm)
+{
+	int i, nc, err, blen;
+
+	/* header */
+	blen = skb_headlen(skb);
+	if (blen < ETH_ZLEN)
+		blen = ETH_ZLEN;
+	blen += VNET_PACKET_SKIP;
+	blen += 8 - (blen & 7);
+
+	err = ldc_map_single(lp, skb->data-VNET_PACKET_SKIP, blen, cookies,
+			     ncookies, map_perm);
+	if (err < 0)
+		return err;
+	nc = err;
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *f = &skb_shinfo(skb)->frags[i];
+		u8 *vaddr;
+
+		if (nc < ncookies) {
+			vaddr = kmap_atomic(skb_frag_page(f));
+			blen = skb_frag_size(f);
+			blen += 8 - (blen & 7);
+			err = ldc_map_single(lp, vaddr + f->page_offset,
+					     blen, cookies + nc, ncookies - nc,
+					     map_perm);
+			kunmap_atomic(vaddr);
+		} else {
+			err = -EMSGSIZE;
+		}
+
+		if (err < 0) {
+			ldc_unmap(lp, cookies, nc);
+			return err;
+		}
+		nc += err;
+	}
+	return nc;
+}
+
+static inline struct sk_buff *vnet_skb_shape(struct sk_buff *skb, int ncookies)
+{
+	struct sk_buff *nskb;
+	int i, len, pad, docopy;
+
+	len = skb->len;
+	pad = 0;
+	if (len < ETH_ZLEN) {
+		pad += ETH_ZLEN - skb->len;
+		len += pad;
+	}
+	len += VNET_PACKET_SKIP;
+	pad += 8 - (len & 7);
+
+	/* make sure we have enough cookies and alignment in every frag */
+	docopy = skb_shinfo(skb)->nr_frags >= ncookies;
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *f = &skb_shinfo(skb)->frags[i];
+
+		docopy |= f->page_offset & 7;
+	}
+	if (((unsigned long)skb->data & 7) != VNET_PACKET_SKIP ||
+	    skb_tailroom(skb) < pad ||
+	    skb_headroom(skb) < VNET_PACKET_SKIP || docopy) {
+		int start = 0, offset;
+		__wsum csum;
+
+		len = skb->len > ETH_ZLEN ? skb->len : ETH_ZLEN;
+		nskb = alloc_and_align_skb(skb->dev, len);
+		if (nskb == NULL) {
+			dev_kfree_skb(skb);
+			return NULL;
+		}
+		skb_reserve(nskb, VNET_PACKET_SKIP);
+
+		nskb->protocol = skb->protocol;
+		offset = skb_mac_header(skb) - skb->data;
+		skb_set_mac_header(nskb, offset);
+		offset = skb_network_header(skb) - skb->data;
+		skb_set_network_header(nskb, offset);
+		offset = skb_transport_header(skb) - skb->data;
+		skb_set_transport_header(nskb, offset);
+
+		offset = 0;
+		nskb->csum_offset = skb->csum_offset;
+		nskb->ip_summed = skb->ip_summed;
+
+		if (skb->ip_summed == CHECKSUM_PARTIAL)
+			start = skb_checksum_start_offset(skb);
+		if (start) {
+			struct iphdr *iph = ip_hdr(nskb);
+			int offset = start + nskb->csum_offset;
+
+			if (skb_copy_bits(skb, 0, nskb->data, start)) {
+				dev_kfree_skb(nskb);
+				dev_kfree_skb(skb);
+				return NULL;
+			}
+			*(__sum16 *)(skb->data + offset) = 0;
+			csum = skb_copy_and_csum_bits(skb, start,
+						      nskb->data + start,
+						      skb->len - start, 0);
+			if (iph->protocol == IPPROTO_TCP ||
+			    iph->protocol == IPPROTO_UDP) {
+				csum = csum_tcpudp_magic(iph->saddr, iph->daddr,
+							 skb->len - start,
+							 iph->protocol, csum);
+			}
+			*(__sum16 *)(nskb->data + offset) = csum;
+
+			nskb->ip_summed = CHECKSUM_NONE;
+		} else if (skb_copy_bits(skb, 0, nskb->data, skb->len)) {
+			dev_kfree_skb(nskb);
+			dev_kfree_skb(skb);
+			return NULL;
+		}
+		(void)skb_put(nskb, skb->len);
+		if (skb_is_gso(skb)) {
+			skb_shinfo(nskb)->gso_size = skb_shinfo(skb)->gso_size;
+			skb_shinfo(nskb)->gso_type = skb_shinfo(skb)->gso_type;
+		}
+		nskb->queue_mapping = skb->queue_mapping;
+		dev_kfree_skb(skb);
+		skb = nskb;
+	}
+	return skb;
+}
+
+u16 sunvnet_select_queue_common(struct net_device *dev, struct sk_buff *skb,
+		  void *accel_priv, select_queue_fallback_t fallback)
+{
+	struct vnet *vp = netdev_priv(dev);
+	struct vnet_port *port = __tx_port_find(vp, skb);
+
+	if (port == NULL)
+		return 0;
+	return port->q_index;
+}
+EXPORT_SYMBOL_GPL(sunvnet_select_queue_common);
+
+static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb)
+{
+	struct net_device *dev = port->vp->dev;
+	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+	struct sk_buff *segs;
+	int maclen, datalen;
+	int status;
+	int gso_size, gso_type, gso_segs;
+	int hlen = skb_transport_header(skb) - skb_mac_header(skb);
+	int proto = IPPROTO_IP;
+
+	if (skb->protocol == htons(ETH_P_IP))
+		proto = ip_hdr(skb)->protocol;
+	else if (skb->protocol == htons(ETH_P_IPV6))
+		proto = ipv6_hdr(skb)->nexthdr;
+
+	if (proto == IPPROTO_TCP)
+		hlen += tcp_hdr(skb)->doff * 4;
+	else if (proto == IPPROTO_UDP)
+		hlen += sizeof(struct udphdr);
+	else {
+		pr_err("vnet_handle_offloads GSO with unknown transport "
+		       "protocol %d tproto %d\n", skb->protocol, proto);
+		hlen = 128; /* XXX */
+	}
+	datalen = port->tsolen - hlen;
+
+	gso_size = skb_shinfo(skb)->gso_size;
+	gso_type = skb_shinfo(skb)->gso_type;
+	gso_segs = skb_shinfo(skb)->gso_segs;
+
+	if (port->tso && gso_size < datalen)
+		gso_segs = DIV_ROUND_UP(skb->len - hlen, datalen);
+
+	if (unlikely(vnet_tx_dring_avail(dr) < gso_segs)) {
+		struct netdev_queue *txq;
+
+		txq  = netdev_get_tx_queue(dev, port->q_index);
+		netif_tx_stop_queue(txq);
+		if (vnet_tx_dring_avail(dr) < skb_shinfo(skb)->gso_segs)
+			return NETDEV_TX_BUSY;
+		netif_tx_wake_queue(txq);
+	}
+
+	maclen = skb_network_header(skb) - skb_mac_header(skb);
+	skb_pull(skb, maclen);
+
+	if (port->tso && gso_size < datalen) {
+		if (skb_unclone(skb, GFP_ATOMIC))
+			goto out_dropped;
+
+		/* segment to TSO size */
+		skb_shinfo(skb)->gso_size = datalen;
+		skb_shinfo(skb)->gso_segs = gso_segs;
+	}
+	segs = skb_gso_segment(skb, dev->features & ~NETIF_F_TSO);
+	if (IS_ERR(segs))
+		goto out_dropped;
+
+	skb_push(skb, maclen);
+	skb_reset_mac_header(skb);
+
+	status = 0;
+	while (segs) {
+		struct sk_buff *curr = segs;
+
+		segs = segs->next;
+		curr->next = NULL;
+		if (port->tso && curr->len > dev->mtu) {
+			skb_shinfo(curr)->gso_size = gso_size;
+			skb_shinfo(curr)->gso_type = gso_type;
+			skb_shinfo(curr)->gso_segs =
+				DIV_ROUND_UP(curr->len - hlen, gso_size);
+		} else
+			skb_shinfo(curr)->gso_size = 0;
+
+		skb_push(curr, maclen);
+		skb_reset_mac_header(curr);
+		memcpy(skb_mac_header(curr), skb_mac_header(skb),
+		       maclen);
+		curr->csum_start = skb_transport_header(curr) - curr->head;
+		if (ip_hdr(curr)->protocol == IPPROTO_TCP)
+			curr->csum_offset = offsetof(struct tcphdr, check);
+		else if (ip_hdr(curr)->protocol == IPPROTO_UDP)
+			curr->csum_offset = offsetof(struct udphdr, check);
+
+		if (!(status & NETDEV_TX_MASK))
+			status = sunvnet_start_xmit_common(curr, dev);
+		if (status & NETDEV_TX_MASK)
+			dev_kfree_skb_any(curr);
+	}
+
+	if (!(status & NETDEV_TX_MASK))
+		dev_kfree_skb_any(skb);
+	return status;
+out_dropped:
+	dev->stats.tx_dropped++;
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
+}
+
+int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev)
+{
+	struct vnet *vp = netdev_priv(dev);
+	struct vnet_port *port = NULL;
+	struct vio_dring_state *dr;
+	struct vio_net_desc *d;
+	unsigned int len;
+	struct sk_buff *freeskbs = NULL;
+	int i, err, txi;
+	unsigned pending = 0;
+	struct netdev_queue *txq;
+
+	rcu_read_lock();
+	port = __tx_port_find(vp, skb);
+	if (unlikely(!port)) {
+		rcu_read_unlock();
+		goto out_dropped;
+	}
+
+	if (skb_is_gso(skb) && skb->len > port->tsolen) {
+		err = vnet_handle_offloads(port, skb);
+		rcu_read_unlock();
+		return err;
+	}
+
+	if (!skb_is_gso(skb) && skb->len > port->rmtu) {
+		unsigned long localmtu = port->rmtu - ETH_HLEN;
+
+		if (vio_version_after_eq(&port->vio, 1, 3))
+			localmtu -= VLAN_HLEN;
+
+		if (skb->protocol == htons(ETH_P_IP)) {
+			struct flowi4 fl4;
+			struct rtable *rt = NULL;
+
+			memset(&fl4, 0, sizeof(fl4));
+			fl4.flowi4_oif = dev->ifindex;
+			fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos);
+			fl4.daddr = ip_hdr(skb)->daddr;
+			fl4.saddr = ip_hdr(skb)->saddr;
+
+			rt = ip_route_output_key(dev_net(dev), &fl4);
+			rcu_read_unlock();
+			if (!IS_ERR(rt)) {
+				skb_dst_set(skb, &rt->dst);
+				icmp_send(skb, ICMP_DEST_UNREACH,
+					  ICMP_FRAG_NEEDED,
+					  htonl(localmtu));
+			}
+		}
+#if IS_ENABLED(CONFIG_IPV6)
+		else if (skb->protocol == htons(ETH_P_IPV6))
+			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, localmtu);
+#endif
+		goto out_dropped;
+	}
+
+	skb = vnet_skb_shape(skb, 2);
+
+	if (unlikely(!skb))
+		goto out_dropped;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		vnet_fullcsum(skb);
+
+	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+	i = skb_get_queue_mapping(skb);
+	txq = netdev_get_tx_queue(dev, i);
+	if (unlikely(vnet_tx_dring_avail(dr) < 1)) {
+		if (!netif_tx_queue_stopped(txq)) {
+			netif_tx_stop_queue(txq);
+
+			/* This is a hard error, log it. */
+			netdev_err(dev, "BUG! Tx Ring full when queue awake!\n");
+			dev->stats.tx_errors++;
+		}
+		rcu_read_unlock();
+		return NETDEV_TX_BUSY;
+	}
+
+	d = vio_dring_cur(dr);
+
+	txi = dr->prod;
+
+	freeskbs = vnet_clean_tx_ring(port, &pending);
+
+	BUG_ON(port->tx_bufs[txi].skb);
+
+	len = skb->len;
+	if (len < ETH_ZLEN)
+		len = ETH_ZLEN;
+
+	err = vnet_skb_map(port->vio.lp, skb, port->tx_bufs[txi].cookies, 2,
+			   (LDC_MAP_SHADOW | LDC_MAP_DIRECT | LDC_MAP_RW));
+	if (err < 0) {
+		netdev_info(dev, "tx buffer map error %d\n", err);
+		goto out_dropped;
+	}
+
+	port->tx_bufs[txi].skb = skb;
+	skb = NULL;
+	port->tx_bufs[txi].ncookies = err;
+
+	/* We don't rely on the ACKs to free the skb in vnet_start_xmit(),
+	 * thus it is safe to not set VIO_ACK_ENABLE for each transmission:
+	 * the protocol itself does not require it as long as the peer
+	 * sends a VIO_SUBTYPE_ACK for VIO_DRING_STOPPED.
+	 *
+	 * An ACK for every packet in the ring is expensive as the
+	 * sending of LDC messages is slow and affects performance.
+	 */
+	d->hdr.ack = VIO_ACK_DISABLE;
+	d->size = len;
+	d->ncookies = port->tx_bufs[txi].ncookies;
+	for (i = 0; i < d->ncookies; i++)
+		d->cookies[i] = port->tx_bufs[txi].cookies[i];
+	if (vio_version_after_eq(&port->vio, 1, 7)) {
+		struct vio_net_dext *dext = vio_net_ext(d);
+
+		memset(dext, 0, sizeof(*dext));
+		if (skb_is_gso(port->tx_bufs[txi].skb)) {
+			dext->ipv4_lso_mss = skb_shinfo(port->tx_bufs[txi].skb)
+					     ->gso_size;
+			dext->flags |= VNET_PKT_IPV4_LSO;
+		}
+		if (vio_version_after_eq(&port->vio, 1, 8) &&
+		    !port->switch_port) {
+			dext->flags |= VNET_PKT_HCK_IPV4_HDRCKSUM_OK;
+			dext->flags |= VNET_PKT_HCK_FULLCKSUM_OK;
+		}
+	}
+
+	/* This has to be a non-SMP write barrier because we are writing
+	 * to memory which is shared with the peer LDOM.
+	 */
+	dma_wmb();
+
+	d->hdr.state = VIO_DESC_READY;
+
+	/* Exactly one ldc "start" trigger (for dr->cons) needs to be sent
+	 * to notify the consumer that some descriptors are READY.
+	 * After that "start" trigger, no additional triggers are needed until
+	 * a DRING_STOPPED is received from the consumer. The dr->cons field
+	 * (set up by vnet_ack()) has the value of the next dring index
+	 * that has not yet been ack-ed. We send a "start" trigger here
+	 * if, and only if, start_cons is true (reset it afterward). Conversely,
+	 * vnet_ack() should check if the dring corresponding to cons
+	 * is marked READY, but start_cons was false.
+	 * If so, vnet_ack() should send out the missed "start" trigger.
+	 *
+	 * Note that the dma_wmb() above makes sure the cookies et al. are
+	 * not globally visible before the VIO_DESC_READY, and that the
+	 * stores are ordered correctly by the compiler. The consumer will
+	 * not proceed until the VIO_DESC_READY is visible assuring that
+	 * the consumer does not observe anything related to descriptors
+	 * out of order. The HV trap from the LDC start trigger is the
+	 * producer to consumer announcement that work is available to the
+	 * consumer
+	 */
+	if (!port->start_cons) { /* previous trigger suffices */
+		trace_vnet_skip_tx_trigger(port->vio._local_sid,
+					   port->vio._peer_sid, dr->cons);
+		goto ldc_start_done;
+	}
+
+	err = __vnet_tx_trigger(port, dr->cons);
+	if (unlikely(err < 0)) {
+		netdev_info(dev, "TX trigger error %d\n", err);
+		d->hdr.state = VIO_DESC_FREE;
+		skb = port->tx_bufs[txi].skb;
+		port->tx_bufs[txi].skb = NULL;
+		dev->stats.tx_carrier_errors++;
+		goto out_dropped;
+	}
+
+ldc_start_done:
+	port->start_cons = false;
+
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += port->tx_bufs[txi].skb->len;
+
+	dr->prod = (dr->prod + 1) & (VNET_TX_RING_SIZE - 1);
+	if (unlikely(vnet_tx_dring_avail(dr) < 1)) {
+		netif_tx_stop_queue(txq);
+		if (vnet_tx_dring_avail(dr) > VNET_TX_WAKEUP_THRESH(dr))
+			netif_tx_wake_queue(txq);
+	}
+
+	(void)mod_timer(&port->clean_timer, jiffies + VNET_CLEAN_TIMEOUT);
+	rcu_read_unlock();
+
+	vnet_free_skbs(freeskbs);
+
+	return NETDEV_TX_OK;
+
+out_dropped:
+	if (pending)
+		(void)mod_timer(&port->clean_timer,
+				jiffies + VNET_CLEAN_TIMEOUT);
+	else if (port)
+		del_timer(&port->clean_timer);
+	if (port)
+		rcu_read_unlock();
+	if (skb)
+		dev_kfree_skb(skb);
+	vnet_free_skbs(freeskbs);
+	dev->stats.tx_dropped++;
+	return NETDEV_TX_OK;
+}
+EXPORT_SYMBOL_GPL(sunvnet_start_xmit_common);
+
+void sunvnet_tx_timeout_common(struct net_device *dev)
+{
+	/* XXX Implement me XXX */
+}
+EXPORT_SYMBOL_GPL(sunvnet_tx_timeout_common);
+
+int sunvnet_open_common(struct net_device *dev)
+{
+	netif_carrier_on(dev);
+	netif_tx_start_all_queues(dev);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(sunvnet_open_common);
+
+int sunvnet_close_common(struct net_device *dev)
+{
+	netif_tx_stop_all_queues(dev);
+	netif_carrier_off(dev);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(sunvnet_close_common);
+
+static struct vnet_mcast_entry *__vnet_mc_find(struct vnet *vp, u8 *addr)
+{
+	struct vnet_mcast_entry *m;
+
+	for (m = vp->mcast_list; m; m = m->next) {
+		if (ether_addr_equal(m->addr, addr))
+			return m;
+	}
+	return NULL;
+}
+
+static void __update_mc_list(struct vnet *vp, struct net_device *dev)
+{
+	struct netdev_hw_addr *ha;
+
+	netdev_for_each_mc_addr(ha, dev) {
+		struct vnet_mcast_entry *m;
+
+		m = __vnet_mc_find(vp, ha->addr);
+		if (m) {
+			m->hit = 1;
+			continue;
+		}
+
+		if (!m) {
+			m = kzalloc(sizeof(*m), GFP_ATOMIC);
+			if (!m)
+				continue;
+			memcpy(m->addr, ha->addr, ETH_ALEN);
+			m->hit = 1;
+
+			m->next = vp->mcast_list;
+			vp->mcast_list = m;
+		}
+	}
+}
+
+static void __send_mc_list(struct vnet *vp, struct vnet_port *port)
+{
+	struct vio_net_mcast_info info;
+	struct vnet_mcast_entry *m, **pp;
+	int n_addrs;
+
+	memset(&info, 0, sizeof(info));
+
+	info.tag.type = VIO_TYPE_CTRL;
+	info.tag.stype = VIO_SUBTYPE_INFO;
+	info.tag.stype_env = VNET_MCAST_INFO;
+	info.tag.sid = vio_send_sid(&port->vio);
+	info.set = 1;
+
+	n_addrs = 0;
+	for (m = vp->mcast_list; m; m = m->next) {
+		if (m->sent)
+			continue;
+		m->sent = 1;
+		memcpy(&info.mcast_addr[n_addrs * ETH_ALEN],
+		       m->addr, ETH_ALEN);
+		if (++n_addrs == VNET_NUM_MCAST) {
+			info.count = n_addrs;
+
+			(void) vio_ldc_send(&port->vio, &info,
+					    sizeof(info));
+			n_addrs = 0;
+		}
+	}
+	if (n_addrs) {
+		info.count = n_addrs;
+		(void) vio_ldc_send(&port->vio, &info, sizeof(info));
+	}
+
+	info.set = 0;
+
+	n_addrs = 0;
+	pp = &vp->mcast_list;
+	while ((m = *pp) != NULL) {
+		if (m->hit) {
+			m->hit = 0;
+			pp = &m->next;
+			continue;
+		}
+
+		memcpy(&info.mcast_addr[n_addrs * ETH_ALEN],
+		       m->addr, ETH_ALEN);
+		if (++n_addrs == VNET_NUM_MCAST) {
+			info.count = n_addrs;
+			(void) vio_ldc_send(&port->vio, &info,
+					    sizeof(info));
+			n_addrs = 0;
+		}
+
+		*pp = m->next;
+		kfree(m);
+	}
+	if (n_addrs) {
+		info.count = n_addrs;
+		(void) vio_ldc_send(&port->vio, &info, sizeof(info));
+	}
+}
+
+void sunvnet_set_rx_mode_common(struct net_device *dev)
+{
+	struct vnet *vp = netdev_priv(dev);
+	struct vnet_port *port;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(port, &vp->port_list, list) {
+
+		if (port->switch_port) {
+			__update_mc_list(vp, dev);
+			__send_mc_list(vp, port);
+			break;
+		}
+	}
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL_GPL(sunvnet_set_rx_mode_common);
+
+int sunvnet_change_mtu_common(struct net_device *dev, int new_mtu)
+{
+	if (new_mtu < 68 || new_mtu > 65535)
+		return -EINVAL;
+
+	dev->mtu = new_mtu;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(sunvnet_change_mtu_common);
+
+int sunvnet_set_mac_addr_common(struct net_device *dev, void *p)
+{
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(sunvnet_set_mac_addr_common);
+
+void sunvnet_port_free_tx_bufs_common(struct vnet_port *port)
+{
+	struct vio_dring_state *dr;
+	int i;
+
+	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+
+	if (dr->base == NULL)
+		return;
+
+	for (i = 0; i < VNET_TX_RING_SIZE; i++) {
+		struct vio_net_desc *d;
+		void *skb = port->tx_bufs[i].skb;
+
+		if (!skb)
+			continue;
+
+		d = vio_dring_entry(dr, i);
+
+		ldc_unmap(port->vio.lp,
+			  port->tx_bufs[i].cookies,
+			  port->tx_bufs[i].ncookies);
+		dev_kfree_skb(skb);
+		port->tx_bufs[i].skb = NULL;
+		d->hdr.state = VIO_DESC_FREE;
+	}
+	ldc_free_exp_dring(port->vio.lp, dr->base,
+			   (dr->entry_size * dr->num_entries),
+			   dr->cookies, dr->ncookies);
+	dr->base = NULL;
+	dr->entry_size = 0;
+	dr->num_entries = 0;
+	dr->pending = 0;
+	dr->ncookies = 0;
+}
+EXPORT_SYMBOL_GPL(sunvnet_port_free_tx_bufs_common);
+
+static void vnet_port_reset(struct vnet_port *port)
+{
+	del_timer(&port->clean_timer);
+	sunvnet_port_free_tx_bufs_common(port);
+	port->rmtu = 0;
+	port->tso = true;
+	port->tsolen = 0;
+}
+
+static int vnet_port_alloc_tx_ring(struct vnet_port *port)
+{
+	struct vio_dring_state *dr;
+	unsigned long len, elen;
+	int i, err, ncookies;
+	void *dring;
+
+	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
+
+	elen = sizeof(struct vio_net_desc) +
+	       sizeof(struct ldc_trans_cookie) * 2;
+	if (vio_version_after_eq(&port->vio, 1, 7))
+		elen += sizeof(struct vio_net_dext);
+	len = VNET_TX_RING_SIZE * elen;
+
+	ncookies = VIO_MAX_RING_COOKIES;
+	dring = ldc_alloc_exp_dring(port->vio.lp, len,
+				    dr->cookies, &ncookies,
+				    (LDC_MAP_SHADOW |
+				     LDC_MAP_DIRECT |
+				     LDC_MAP_RW));
+	if (IS_ERR(dring)) {
+		err = PTR_ERR(dring);
+		goto err_out;
+	}
+
+	dr->base = dring;
+	dr->entry_size = elen;
+	dr->num_entries = VNET_TX_RING_SIZE;
+	dr->prod = dr->cons = 0;
+	port->start_cons  = true; /* need an initial trigger */
+	dr->pending = VNET_TX_RING_SIZE;
+	dr->ncookies = ncookies;
+
+	for (i = 0; i < VNET_TX_RING_SIZE; ++i) {
+		struct vio_net_desc *d;
+
+		d = vio_dring_entry(dr, i);
+		d->hdr.state = VIO_DESC_FREE;
+	}
+	return 0;
+
+err_out:
+	sunvnet_port_free_tx_bufs_common(port);
+
+	return err;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+void sunvnet_poll_controller_common(struct net_device *dev)
+{
+	struct vnet *vp = netdev_priv(dev);
+	struct vnet_port *port;
+	unsigned long flags;
+
+	spin_lock_irqsave(&vp->lock, flags);
+	if (!list_empty(&vp->port_list)) {
+		port = list_entry(vp->port_list.next, struct vnet_port, list);
+		napi_schedule(&port->napi);
+	}
+	spin_unlock_irqrestore(&vp->lock, flags);
+}
+EXPORT_SYMBOL_GPL(sunvnet_poll_controller_common);
+#endif
+
+void sunvnet_port_add_txq_common(struct vnet_port *port)
+{
+	struct vnet *vp = port->vp;
+	int n;
+
+	n = vp->nports++;
+	n = n & (VNET_MAX_TXQS - 1);
+	port->q_index = n;
+	netif_tx_wake_queue(netdev_get_tx_queue(vp->dev, port->q_index));
+}
+EXPORT_SYMBOL_GPL(sunvnet_port_add_txq_common);
+
+void sunvnet_port_rm_txq_common(struct vnet_port *port)
+{
+	port->vp->nports--;
+	netif_tx_stop_queue(netdev_get_tx_queue(port->vp->dev, port->q_index));
+}
+EXPORT_SYMBOL_GPL(sunvnet_port_rm_txq_common);
diff --git a/drivers/net/ethernet/sun/sunvnet_common.h b/drivers/net/ethernet/sun/sunvnet_common.h
new file mode 100644
index 0000000..c29c526
--- /dev/null
+++ b/drivers/net/ethernet/sun/sunvnet_common.h
@@ -0,0 +1,133 @@
+#ifndef _SUNVNETCOMMON_H
+#define _SUNVNETCOMMON_H
+
+#include <linux/interrupt.h>
+
+/* length of time (or less) we expect pending descriptors to be marked
+ * as VIO_DESC_DONE and skbs ready to be freed
+ */
+#define	VNET_CLEAN_TIMEOUT		((HZ / 100) + 1)
+
+#define VNET_MAXPACKET			(65535ULL + ETH_HLEN + VLAN_HLEN)
+#define VNET_TX_RING_SIZE		512
+#define VNET_TX_WAKEUP_THRESH(dr)	((dr)->pending / 4)
+
+#define	VNET_MINTSO	 2048	/* VIO protocol's minimum TSO len */
+#define	VNET_MAXTSO	65535	/* VIO protocol's maximum TSO len */
+
+/* VNET packets are sent in buffers with the first 6 bytes skipped
+ * so that after the ethernet header the IPv4/IPv6 headers are aligned
+ * properly.
+ */
+#define VNET_PACKET_SKIP		6
+
+#define	VNET_MAXCOOKIES			(VNET_MAXPACKET / PAGE_SIZE + 1)
+
+#define	VNET_MAX_TXQS		16
+
+struct vnet_tx_entry {
+	struct sk_buff		*skb;
+	unsigned int		ncookies;
+	struct ldc_trans_cookie	cookies[VNET_MAXCOOKIES];
+};
+
+struct vnet;
+struct vnet_port {
+	struct vio_driver_state	vio;
+
+	struct hlist_node	hash;
+	u8			raddr[ETH_ALEN];
+	unsigned		switch_port:1;
+	unsigned		tso:1;
+	unsigned		__pad:14;
+
+	struct vnet		*vp;
+
+	struct vnet_tx_entry	tx_bufs[VNET_TX_RING_SIZE];
+
+	struct list_head	list;
+
+	u32			stop_rx_idx;
+	bool			stop_rx;
+	bool			start_cons;
+
+	struct timer_list	clean_timer;
+
+	u64			rmtu;
+	u16			tsolen;
+
+	struct napi_struct	napi;
+	u32			napi_stop_idx;
+	bool			napi_resume;
+	int			rx_event;
+	u16			q_index;
+};
+
+static inline struct vnet_port *to_vnet_port(struct vio_driver_state *vio)
+{
+	return container_of(vio, struct vnet_port, vio);
+}
+
+#define VNET_PORT_HASH_SIZE	16
+#define VNET_PORT_HASH_MASK	(VNET_PORT_HASH_SIZE - 1)
+
+static inline unsigned int vnet_hashfn(u8 *mac)
+{
+	unsigned int val = mac[4] ^ mac[5];
+
+	return val & (VNET_PORT_HASH_MASK);
+}
+
+struct vnet_mcast_entry {
+	u8			addr[ETH_ALEN];
+	u8			sent;
+	u8			hit;
+	struct vnet_mcast_entry	*next;
+};
+
+struct vnet {
+	/* Protects port_list and port_hash.  */
+	spinlock_t		lock;
+
+	struct net_device	*dev;
+
+	u32			msg_enable;
+
+	struct list_head	port_list;
+
+	struct hlist_head	port_hash[VNET_PORT_HASH_SIZE];
+
+	struct vnet_mcast_entry	*mcast_list;
+
+	struct list_head	list;
+	u64			local_mac;
+
+	int			nports;
+};
+
+/* Common funcs */
+void sunvnet_clean_timer_expire_common(unsigned long port0);
+int sunvnet_open_common(struct net_device *dev);
+int sunvnet_close_common(struct net_device *dev);
+void sunvnet_set_rx_mode_common(struct net_device *dev);
+int sunvnet_set_mac_addr_common(struct net_device *dev, void *p);
+void sunvnet_tx_timeout_common(struct net_device *dev);
+int sunvnet_change_mtu_common(struct net_device *dev, int new_mtu);
+int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev);
+u16 sunvnet_select_queue_common(struct net_device *dev,
+				struct sk_buff *skb,
+				void *accel_priv,
+				select_queue_fallback_t fallback);
+#ifdef CONFIG_NET_POLL_CONTROLLER
+void sunvnet_poll_controller_common(struct net_device *dev);
+#endif
+void sunvnet_event_common(void *arg, int event);
+int sunvnet_send_attr_common(struct vio_driver_state *vio);
+int sunvnet_handle_attr_common(struct vio_driver_state *vio, void *arg);
+void sunvnet_handshake_complete_common(struct vio_driver_state *vio);
+int sunvnet_poll_common(struct napi_struct *napi, int budget);
+void sunvnet_port_free_tx_bufs_common(struct vnet_port *port);
+void sunvnet_port_add_txq_common(struct vnet_port *port);
+void sunvnet_port_rm_txq_common(struct vnet_port *port);
+
+#endif /* _SUNVNETCOMMON_H */
-- 
cgit v0.10.2


From 67d0719f06ded9488311472b3d65ad37d992c332 Mon Sep 17 00:00:00 2001
From: Aaron Young <aaron.young@oracle.com>
Date: Tue, 15 Mar 2016 11:35:38 -0700
Subject: ldmvsw: Make sunvnet_common compatible with ldmvsw

  Modify sunvnet common code and data structures to be compatible
  with both sunvnet and ldmvsw drivers.

  Details:

  Sunvnet operates on "vnet-port" nodes which appear in the Machine
  Description (MD) in a guest domain. Ldmvsw operates on "vsw-port"
  nodes which appear in the MD of a service domain.

  A difference between the sunvnet driver and the ldmvsw driver is
  the sunvnet driver creates a network interface (i.e. a struct net_device)
  for every vnet-port *parent* "network" node. Several vnet-ports may appear
  under this common parent network node - each corresponding to a common parent
  network interface.  Conversely, since bridge/vswitch software will need
  to interface with every vsw-port in a system, the ldmvsw driver creates
  a network interface (i.e. a struct net_device) for every vsw-port - not
  every parent node as with sunvnet.  This difference required some special
  handling in the common code as explained below.

  There are 2 key data structures used by the sunvnet and ldmvsw drivers
  (which are now found in sunvnet_common.h):

  1. struct vnet_port
     This structure represents a vnet-port node in sunvnet and a vsw-port
     in the ldmvsw driver.

  2. struct vnet
     This structure represents a parent "network" node in sunvnet and a parent
     "virtual-network-switch" node in ldmvsw.

  Since the sunvnet driver allocates a net_device for every parent "network"
  node, a net_device member appears in the struct vnet. Since the ldmvsw
  driver allocates a net_device for every port, a net_device member was
  added to the vnet_port. The common code distinguishes which structure
  net_device member to use by checking a 'vsw' bit that was added to the
  vnet_port structure. See the VNET_PORT_TO_NET_DEVICE() marco in
  sunvnet_common.h.

  The netdev_priv() in sunvnet is allocated as a vnet. The netdev_priv()
  in ldmvsw is a vnet_port. Therefore, any place in the common code
  where a netdev_priv() call was made, a wrapper function was implemented
  in each driver to first get the vnet and/or vnet_port (in a driver
  specific way) and pass them as newly added parameters to the common
  functions (see wrapper funcs: vnet_set_rx_mode() and vnet_poll_controller()).
  Since these wrapper functions call __tx_port_find(), __tx_port_find() was
  moved from the common code back into sunvnet.c. Note - ldmvsw.c does not
  require this function.

  These changes also required that port_is_up() be made
  into a common function and thus it was given a _common suffix and
  exported like the other common functions.

  A wrapper function was also added for vnet_start_xmit_common() to pass a
  driver-specific function arg to return the port associated with a given
  struct sk_buff and struct net_device. This was required because
  vnet_start_xmit_common() grabs a lock prior to getting the associated
  port. Using a function pointer arg allowed the code to work unchanged
  without risking changes to the non-trivial locking logic in
  vnet_start_xmit_common().

  Signed-off-by: Aaron Young <aaron.young@oracle.com>
  Signed-off-by: Rashmi Narasimhan <rashmi.narasimhan@oracle.com>
  Reviewed-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
  Reviewed-by: Alexandre Chartre <Alexandre.Chartre@oracle.com>

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/sun/sunvnet.c b/drivers/net/ethernet/sun/sunvnet.c
index 5b911352..98c5f16 100644
--- a/drivers/net/ethernet/sun/sunvnet.c
+++ b/drivers/net/ethernet/sun/sunvnet.c
@@ -1,6 +1,7 @@
 /* sunvnet.c: Sun LDOM Virtual Network Driver.
  *
  * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2016 Oracle. All rights reserved.
  */
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -84,18 +85,83 @@ static const struct ethtool_ops vnet_ethtool_ops = {
 static LIST_HEAD(vnet_list);
 static DEFINE_MUTEX(vnet_list_mutex);
 
+static struct vnet_port *__tx_port_find(struct vnet *vp, struct sk_buff *skb)
+{
+	unsigned int hash = vnet_hashfn(skb->data);
+	struct hlist_head *hp = &vp->port_hash[hash];
+	struct vnet_port *port;
+
+	hlist_for_each_entry_rcu(port, hp, hash) {
+		if (!sunvnet_port_is_up_common(port))
+			continue;
+		if (ether_addr_equal(port->raddr, skb->data))
+			return port;
+	}
+	list_for_each_entry_rcu(port, &vp->port_list, list) {
+		if (!port->switch_port)
+			continue;
+		if (!sunvnet_port_is_up_common(port))
+			continue;
+		return port;
+	}
+	return NULL;
+}
+
+/* func arg to vnet_start_xmit_common() to get the proper tx port */
+static struct vnet_port *vnet_tx_port_find(struct sk_buff *skb,
+					   struct net_device *dev)
+{
+	struct vnet *vp = netdev_priv(dev);
+
+	return __tx_port_find(vp, skb);
+}
+
+static u16 vnet_select_queue(struct net_device *dev, struct sk_buff *skb,
+			     void *accel_priv, select_queue_fallback_t fallback)
+{
+	struct vnet *vp = netdev_priv(dev);
+	struct vnet_port *port = __tx_port_find(vp, skb);
+
+	if (!port)
+		return 0;
+
+	return port->q_index;
+}
+
+/* Wrappers to common functions */
+static int vnet_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	return sunvnet_start_xmit_common(skb, dev, vnet_tx_port_find);
+}
+
+static void vnet_set_rx_mode(struct net_device *dev)
+{
+	struct vnet *vp = netdev_priv(dev);
+
+	return sunvnet_set_rx_mode_common(dev, vp);
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void vnet_poll_controller(struct net_device *dev)
+{
+	struct vnet *vp = netdev_priv(dev);
+
+	return sunvnet_poll_controller_common(dev, vp);
+}
+#endif
+
 static const struct net_device_ops vnet_ops = {
 	.ndo_open		= sunvnet_open_common,
 	.ndo_stop		= sunvnet_close_common,
-	.ndo_set_rx_mode	= sunvnet_set_rx_mode_common,
+	.ndo_set_rx_mode	= vnet_set_rx_mode,
 	.ndo_set_mac_address	= sunvnet_set_mac_addr_common,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_tx_timeout		= sunvnet_tx_timeout_common,
 	.ndo_change_mtu		= sunvnet_change_mtu_common,
-	.ndo_start_xmit		= sunvnet_start_xmit_common,
-	.ndo_select_queue	= sunvnet_select_queue_common,
+	.ndo_start_xmit		= vnet_start_xmit,
+	.ndo_select_queue	= vnet_select_queue,
 #ifdef CONFIG_NET_POLL_CONTROLLER
-	.ndo_poll_controller	= sunvnet_poll_controller_common,
+	.ndo_poll_controller	= vnet_poll_controller,
 #endif
 };
 
diff --git a/drivers/net/ethernet/sun/sunvnet_common.c b/drivers/net/ethernet/sun/sunvnet_common.c
index 49e85d0..083f41c 100644
--- a/drivers/net/ethernet/sun/sunvnet_common.c
+++ b/drivers/net/ethernet/sun/sunvnet_common.c
@@ -1,6 +1,7 @@
 /* sunvnet.c: Sun LDOM Virtual Network Driver.
  *
  * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2016 Oracle. All rights reserved.
  */
 
 #include <linux/module.h>
@@ -62,7 +63,7 @@ static int vnet_port_alloc_tx_ring(struct vnet_port *port);
 int sunvnet_send_attr_common(struct vio_driver_state *vio)
 {
 	struct vnet_port *port = to_vnet_port(vio);
-	struct net_device *dev = port->vp->dev;
+	struct net_device *dev = VNET_PORT_TO_NET_DEVICE(port);
 	struct vio_net_attr_info pkt;
 	int framelen = ETH_FRAME_LEN;
 	int i, err;
@@ -330,7 +331,7 @@ static inline void vnet_fullcsum(struct sk_buff *skb)
 
 static int vnet_rx_one(struct vnet_port *port, struct vio_net_desc *desc)
 {
-	struct net_device *dev = port->vp->dev;
+	struct net_device *dev = VNET_PORT_TO_NET_DEVICE(port);
 	unsigned int len = desc->size;
 	unsigned int copy_len;
 	struct sk_buff *skb;
@@ -633,7 +634,6 @@ static int vnet_ack(struct vnet_port *port, void *msgbuf)
 	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
 	struct vio_dring_data *pkt = msgbuf;
 	struct net_device *dev;
-	struct vnet *vp;
 	u32 end;
 	struct vio_net_desc *desc;
 	struct netdev_queue *txq;
@@ -642,8 +642,7 @@ static int vnet_ack(struct vnet_port *port, void *msgbuf)
 		return 0;
 
 	end = pkt->end_idx;
-	vp = port->vp;
-	dev = vp->dev;
+	dev = VNET_PORT_TO_NET_DEVICE(port);
 	netif_tx_lock(dev);
 	if (unlikely(!idx_is_pending(dr, end))) {
 		netif_tx_unlock(dev);
@@ -688,10 +687,11 @@ static int vnet_nack(struct vnet_port *port, void *msgbuf)
 static int handle_mcast(struct vnet_port *port, void *msgbuf)
 {
 	struct vio_net_mcast_info *pkt = msgbuf;
+	struct net_device *dev = VNET_PORT_TO_NET_DEVICE(port);
 
 	if (pkt->tag.stype != VIO_SUBTYPE_ACK)
 		pr_err("%s: Got unexpected MCAST reply [%02x:%02x:%04x:%08x]\n",
-		       port->vp->dev->name,
+		       dev->name,
 		       pkt->tag.type,
 		       pkt->tag.stype,
 		       pkt->tag.stype_env,
@@ -708,7 +708,8 @@ static void maybe_tx_wakeup(struct vnet_port *port)
 {
 	struct netdev_queue *txq;
 
-	txq = netdev_get_tx_queue(port->vp->dev, port->q_index);
+	txq = netdev_get_tx_queue(VNET_PORT_TO_NET_DEVICE(port),
+				  port->q_index);
 	__netif_tx_lock(txq, smp_processor_id());
 	if (likely(netif_tx_queue_stopped(txq))) {
 		struct vio_dring_state *dr;
@@ -719,12 +720,13 @@ static void maybe_tx_wakeup(struct vnet_port *port)
 	__netif_tx_unlock(txq);
 }
 
-static inline bool port_is_up(struct vnet_port *vnet)
+bool sunvnet_port_is_up_common(struct vnet_port *vnet)
 {
 	struct vio_driver_state *vio = &vnet->vio;
 
 	return !!(vio->hs_state & VIO_HS_COMPLETE);
 }
+EXPORT_SYMBOL_GPL(sunvnet_port_is_up_common);
 
 static int vnet_event_napi(struct vnet_port *port, int budget)
 {
@@ -797,7 +799,7 @@ ldc_ctrl:
 napi_resume:
 		if (likely(msgbuf.tag.type == VIO_TYPE_DATA)) {
 			if (msgbuf.tag.stype == VIO_SUBTYPE_INFO) {
-				if (!port_is_up(port)) {
+				if (!sunvnet_port_is_up_common(port)) {
 					/* failures like handshake_failure()
 					 * may have cleaned up dring, but
 					 * NAPI polling may bring us here.
@@ -911,28 +913,6 @@ static int __vnet_tx_trigger(struct vnet_port *port, u32 start)
 	return err;
 }
 
-static struct vnet_port *__tx_port_find(struct vnet *vp, struct sk_buff *skb)
-{
-	unsigned int hash = vnet_hashfn(skb->data);
-	struct hlist_head *hp = &vp->port_hash[hash];
-	struct vnet_port *port;
-
-	hlist_for_each_entry_rcu(port, hp, hash) {
-		if (!port_is_up(port))
-			continue;
-		if (ether_addr_equal(port->raddr, skb->data))
-			return port;
-	}
-	list_for_each_entry_rcu(port, &vp->port_list, list) {
-		if (!port->switch_port)
-			continue;
-		if (!port_is_up(port))
-			continue;
-		return port;
-	}
-	return NULL;
-}
-
 static struct sk_buff *vnet_clean_tx_ring(struct vnet_port *port,
 					  unsigned *pending)
 {
@@ -994,9 +974,9 @@ void sunvnet_clean_timer_expire_common(unsigned long port0)
 	struct sk_buff *freeskbs;
 	unsigned pending;
 
-	netif_tx_lock(port->vp->dev);
+	netif_tx_lock(VNET_PORT_TO_NET_DEVICE(port));
 	freeskbs = vnet_clean_tx_ring(port, &pending);
-	netif_tx_unlock(port->vp->dev);
+	netif_tx_unlock(VNET_PORT_TO_NET_DEVICE(port));
 
 	vnet_free_skbs(freeskbs);
 
@@ -1140,21 +1120,11 @@ static inline struct sk_buff *vnet_skb_shape(struct sk_buff *skb, int ncookies)
 	return skb;
 }
 
-u16 sunvnet_select_queue_common(struct net_device *dev, struct sk_buff *skb,
-		  void *accel_priv, select_queue_fallback_t fallback)
+static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb,
+				struct vnet_port *(*vnet_tx_port)
+				(struct sk_buff *, struct net_device *))
 {
-	struct vnet *vp = netdev_priv(dev);
-	struct vnet_port *port = __tx_port_find(vp, skb);
-
-	if (port == NULL)
-		return 0;
-	return port->q_index;
-}
-EXPORT_SYMBOL_GPL(sunvnet_select_queue_common);
-
-static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb)
-{
-	struct net_device *dev = port->vp->dev;
+	struct net_device *dev = VNET_PORT_TO_NET_DEVICE(port);
 	struct vio_dring_state *dr = &port->vio.drings[VIO_DRIVER_TX_RING];
 	struct sk_buff *segs;
 	int maclen, datalen;
@@ -1239,7 +1209,8 @@ static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb)
 			curr->csum_offset = offsetof(struct udphdr, check);
 
 		if (!(status & NETDEV_TX_MASK))
-			status = sunvnet_start_xmit_common(curr, dev);
+			status = sunvnet_start_xmit_common(curr, dev,
+							   vnet_tx_port);
 		if (status & NETDEV_TX_MASK)
 			dev_kfree_skb_any(curr);
 	}
@@ -1253,9 +1224,10 @@ out_dropped:
 	return NETDEV_TX_OK;
 }
 
-int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev)
+int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev,
+			      struct vnet_port *(*vnet_tx_port)
+			      (struct sk_buff *, struct net_device *))
 {
-	struct vnet *vp = netdev_priv(dev);
 	struct vnet_port *port = NULL;
 	struct vio_dring_state *dr;
 	struct vio_net_desc *d;
@@ -1266,14 +1238,14 @@ int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev)
 	struct netdev_queue *txq;
 
 	rcu_read_lock();
-	port = __tx_port_find(vp, skb);
+	port = vnet_tx_port(skb, dev);
 	if (unlikely(!port)) {
 		rcu_read_unlock();
 		goto out_dropped;
 	}
 
 	if (skb_is_gso(skb) && skb->len > port->tsolen) {
-		err = vnet_handle_offloads(port, skb);
+		err = vnet_handle_offloads(port, skb, vnet_tx_port);
 		rcu_read_unlock();
 		return err;
 	}
@@ -1588,9 +1560,8 @@ static void __send_mc_list(struct vnet *vp, struct vnet_port *port)
 	}
 }
 
-void sunvnet_set_rx_mode_common(struct net_device *dev)
+void sunvnet_set_rx_mode_common(struct net_device *dev, struct vnet *vp)
 {
-	struct vnet *vp = netdev_priv(dev);
 	struct vnet_port *port;
 
 	rcu_read_lock();
@@ -1717,9 +1688,8 @@ err_out:
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
-void sunvnet_poll_controller_common(struct net_device *dev)
+void sunvnet_poll_controller_common(struct net_device *dev, struct vnet *vp)
 {
-	struct vnet *vp = netdev_priv(dev);
 	struct vnet_port *port;
 	unsigned long flags;
 
@@ -1741,13 +1711,16 @@ void sunvnet_port_add_txq_common(struct vnet_port *port)
 	n = vp->nports++;
 	n = n & (VNET_MAX_TXQS - 1);
 	port->q_index = n;
-	netif_tx_wake_queue(netdev_get_tx_queue(vp->dev, port->q_index));
+	netif_tx_wake_queue(netdev_get_tx_queue(VNET_PORT_TO_NET_DEVICE(port),
+						port->q_index));
+
 }
 EXPORT_SYMBOL_GPL(sunvnet_port_add_txq_common);
 
 void sunvnet_port_rm_txq_common(struct vnet_port *port)
 {
 	port->vp->nports--;
-	netif_tx_stop_queue(netdev_get_tx_queue(port->vp->dev, port->q_index));
+	netif_tx_stop_queue(netdev_get_tx_queue(VNET_PORT_TO_NET_DEVICE(port),
+						port->q_index));
 }
 EXPORT_SYMBOL_GPL(sunvnet_port_rm_txq_common);
diff --git a/drivers/net/ethernet/sun/sunvnet_common.h b/drivers/net/ethernet/sun/sunvnet_common.h
index c29c526..bd36528 100644
--- a/drivers/net/ethernet/sun/sunvnet_common.h
+++ b/drivers/net/ethernet/sun/sunvnet_common.h
@@ -32,6 +32,13 @@ struct vnet_tx_entry {
 };
 
 struct vnet;
+
+/* Structure to describe a vnet-port or vsw-port in the MD.
+ * If the vsw bit is set, this structure represents a vswitch
+ * port, and the net_device can be found from ->dev. If the
+ * vsw bit is not set, the net_device is available from ->vp->dev.
+ * See the VNET_PORT_TO_NET_DEVICE macro below.
+ */
 struct vnet_port {
 	struct vio_driver_state	vio;
 
@@ -39,9 +46,11 @@ struct vnet_port {
 	u8			raddr[ETH_ALEN];
 	unsigned		switch_port:1;
 	unsigned		tso:1;
-	unsigned		__pad:14;
+	unsigned		vsw:1;
+	unsigned		__pad:13;
 
 	struct vnet		*vp;
+	struct net_device	*dev;
 
 	struct vnet_tx_entry	tx_bufs[VNET_TX_RING_SIZE];
 
@@ -105,21 +114,23 @@ struct vnet {
 	int			nports;
 };
 
+/* Def used by common code to get the net_device from the proper location */
+#define VNET_PORT_TO_NET_DEVICE(__port) \
+	((__port)->vsw ? (__port)->dev : (__port)->vp->dev)
+
 /* Common funcs */
 void sunvnet_clean_timer_expire_common(unsigned long port0);
 int sunvnet_open_common(struct net_device *dev);
 int sunvnet_close_common(struct net_device *dev);
-void sunvnet_set_rx_mode_common(struct net_device *dev);
+void sunvnet_set_rx_mode_common(struct net_device *dev, struct vnet *vp);
 int sunvnet_set_mac_addr_common(struct net_device *dev, void *p);
 void sunvnet_tx_timeout_common(struct net_device *dev);
 int sunvnet_change_mtu_common(struct net_device *dev, int new_mtu);
-int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev);
-u16 sunvnet_select_queue_common(struct net_device *dev,
-				struct sk_buff *skb,
-				void *accel_priv,
-				select_queue_fallback_t fallback);
+int sunvnet_start_xmit_common(struct sk_buff *skb, struct net_device *dev,
+			   struct vnet_port *(*vnet_tx_port)
+			   (struct sk_buff *, struct net_device *));
 #ifdef CONFIG_NET_POLL_CONTROLLER
-void sunvnet_poll_controller_common(struct net_device *dev);
+void sunvnet_poll_controller_common(struct net_device *dev, struct vnet *vp);
 #endif
 void sunvnet_event_common(void *arg, int event);
 int sunvnet_send_attr_common(struct vio_driver_state *vio);
@@ -127,6 +138,7 @@ int sunvnet_handle_attr_common(struct vio_driver_state *vio, void *arg);
 void sunvnet_handshake_complete_common(struct vio_driver_state *vio);
 int sunvnet_poll_common(struct napi_struct *napi, int budget);
 void sunvnet_port_free_tx_bufs_common(struct vnet_port *port);
+bool sunvnet_port_is_up_common(struct vnet_port *vnet);
 void sunvnet_port_add_txq_common(struct vnet_port *port);
 void sunvnet_port_rm_txq_common(struct vnet_port *port);
 
-- 
cgit v0.10.2


From 5d01fa0c6bd84ddf1339a3fadfefecd2c28d472e Mon Sep 17 00:00:00 2001
From: Aaron Young <aaron.young@oracle.com>
Date: Tue, 15 Mar 2016 11:35:39 -0700
Subject: ldmvsw: Add ldmvsw.c driver code

  Add ldmvsw.c driver

  Details:

  The ldmvsw driver very closely follows the sunvnet.c code and makes
  use of the sunvnet_common.c code for core functionality.

  A significant difference between sunvnet and ldmvsw driver is
  sunvnet creates a network interface for each vnet-port *parent*
  node in the MD while the ldmvsw driver creates a network interface
  for every vsw-port node in the Machine Description (MD).
  Therefore the netdev_priv() for sunvnet is a vnet structure while
  the netdev_priv() for ldmvsw is a vnet_port structure.

  Vnet_port structures allocated by ldmvsw have the vsw bit set.
  When finding the net_device associated with a port, the common code keys
  off this bit to use either the net_device found in the vnet_port or the
  net_device in the vnet structure (see the VNET_PORT_TO_NET_DEVICE() macro in
  sunvnet_common.h). This scheme allows the common code to work with
  both drivers with minimal changes.

  Similar to Xen, network interfaces created by the ldmvsw driver will always
  have a HW Addr (i.e. mac address) of FE:FF:FF:FF:FF:FF and each will be
  assigned the devname "vif<cfg_handle>.<port_id>" - where <cfg_handle> and
  <port_id> are a unique handle/port pair assigned to the associated
  vsw-port node in the MD.

  Signed-off-by: Aaron Young <aaron.young@oracle.com>
  Signed-off-by: Rashmi Narasimhan <rashmi.narasimhan@oracle.com>
  Reviewed-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
  Reviewed-by: Alexandre Chartre <Alexandre.Chartre@oracle.com>

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/arch/sparc/configs/sparc64_defconfig b/arch/sparc/configs/sparc64_defconfig
index 6b68f12..04920ab 100644
--- a/arch/sparc/configs/sparc64_defconfig
+++ b/arch/sparc/configs/sparc64_defconfig
@@ -102,6 +102,7 @@ CONFIG_SUNLANCE=m
 CONFIG_HAPPYMEAL=m
 CONFIG_SUNGEM=m
 CONFIG_SUNVNET=m
+CONFIG_LDMVSW=m
 CONFIG_NET_PCI=y
 CONFIG_E1000=m
 CONFIG_E1000E=m
diff --git a/drivers/net/ethernet/sun/Kconfig b/drivers/net/ethernet/sun/Kconfig
index aa58c11e..a4b40e3 100644
--- a/drivers/net/ethernet/sun/Kconfig
+++ b/drivers/net/ethernet/sun/Kconfig
@@ -80,6 +80,17 @@ config SUNVNET
 	---help---
 	  Support for virtual network devices under Sun Logical Domains.
 
+config LDMVSW
+	tristate "Sun4v LDoms Virtual Switch support"
+	depends on SUN_LDOMS
+	---help---
+	  Support for virtual switch devices under Sun4v Logical Domains.
+	  This driver adds a network interface for every vsw-port node
+	  found in the machine description of a service domain.
+	  Linux bridge/switch software can use these interfaces for
+	  guest domain network interconnectivity or guest domain
+	  connection to a physical network on a service domain.
+
 config NIU
 	tristate "Sun Neptune 10Gbit Ethernet support"
 	depends on PCI
diff --git a/drivers/net/ethernet/sun/Makefile b/drivers/net/ethernet/sun/Makefile
index 7b622aa..3785543 100644
--- a/drivers/net/ethernet/sun/Makefile
+++ b/drivers/net/ethernet/sun/Makefile
@@ -9,4 +9,5 @@ obj-$(CONFIG_SUNGEM) += sungem.o
 obj-$(CONFIG_CASSINI) += cassini.o
 obj-$(CONFIG_SUNVNET_COMMON) += sunvnet_common.o
 obj-$(CONFIG_SUNVNET) += sunvnet.o
+obj-$(CONFIG_LDMVSW) += ldmvsw.o
 obj-$(CONFIG_NIU) += niu.o
diff --git a/drivers/net/ethernet/sun/ldmvsw.c b/drivers/net/ethernet/sun/ldmvsw.c
new file mode 100644
index 0000000..e15bf84
--- /dev/null
+++ b/drivers/net/ethernet/sun/ldmvsw.c
@@ -0,0 +1,468 @@
+/* ldmvsw.c: Sun4v LDOM Virtual Switch Driver.
+ *
+ * Copyright (C) 2016 Oracle. All rights reserved.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/delay.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/highmem.h>
+#include <linux/if_vlan.h>
+#include <linux/init.h>
+#include <linux/kconfig.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/netdevice.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+#if defined(CONFIG_IPV6)
+#include <linux/icmpv6.h>
+#endif
+
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/route.h>
+
+#include <asm/vio.h>
+#include <asm/ldc.h>
+
+/* This driver makes use of the common code in sunvnet_common.c */
+#include "sunvnet_common.h"
+
+/* Length of time before we decide the hardware is hung,
+ * and dev->tx_timeout() should be called to fix the problem.
+ */
+#define VSW_TX_TIMEOUT			(10 * HZ)
+
+/* Static HW Addr used for the network interfaces representing vsw ports */
+static u8 vsw_port_hwaddr[ETH_ALEN] = {0xFE, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF};
+
+#define DRV_MODULE_NAME		"ldmvsw"
+#define DRV_MODULE_VERSION	"1.0"
+#define DRV_MODULE_RELDATE	"Jan 15, 2016"
+
+static char version[] =
+	DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
+MODULE_AUTHOR("Oracle");
+MODULE_DESCRIPTION("Sun4v LDOM Virtual Switch Driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_MODULE_VERSION);
+
+/* Ordered from largest major to lowest */
+static struct vio_version vsw_versions[] = {
+	{ .major = 1, .minor = 8 },
+	{ .major = 1, .minor = 7 },
+	{ .major = 1, .minor = 6 },
+	{ .major = 1, .minor = 0 },
+};
+
+static void vsw_get_drvinfo(struct net_device *dev,
+			    struct ethtool_drvinfo *info)
+{
+	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
+}
+
+static u32 vsw_get_msglevel(struct net_device *dev)
+{
+	struct vnet_port *port = netdev_priv(dev);
+
+	return port->vp->msg_enable;
+}
+
+static void vsw_set_msglevel(struct net_device *dev, u32 value)
+{
+	struct vnet_port *port = netdev_priv(dev);
+
+	port->vp->msg_enable = value;
+}
+
+static const struct ethtool_ops vsw_ethtool_ops = {
+	.get_drvinfo		= vsw_get_drvinfo,
+	.get_msglevel		= vsw_get_msglevel,
+	.set_msglevel		= vsw_set_msglevel,
+	.get_link		= ethtool_op_get_link,
+};
+
+static LIST_HEAD(vnet_list);
+static DEFINE_MUTEX(vnet_list_mutex);
+
+/* func arg to vnet_start_xmit_common() to get the proper tx port */
+static struct vnet_port *vsw_tx_port_find(struct sk_buff *skb,
+					  struct net_device *dev)
+{
+	struct vnet_port *port = netdev_priv(dev);
+
+	return port;
+}
+
+static u16 vsw_select_queue(struct net_device *dev, struct sk_buff *skb,
+			    void *accel_priv, select_queue_fallback_t fallback)
+{
+	struct vnet_port *port = netdev_priv(dev);
+
+	if (!port)
+		return 0;
+
+	return port->q_index;
+}
+
+/* Wrappers to common functions */
+static int vsw_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	return sunvnet_start_xmit_common(skb, dev, vsw_tx_port_find);
+}
+
+static void vsw_set_rx_mode(struct net_device *dev)
+{
+	struct vnet_port *port = netdev_priv(dev);
+
+	return sunvnet_set_rx_mode_common(dev, port->vp);
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void vsw_poll_controller(struct net_device *dev)
+{
+	struct vnet_port *port = netdev_priv(dev);
+
+	return sunvnet_poll_controller_common(dev, port->vp);
+}
+#endif
+
+static const struct net_device_ops vsw_ops = {
+	.ndo_open		= sunvnet_open_common,
+	.ndo_stop		= sunvnet_close_common,
+	.ndo_set_rx_mode	= vsw_set_rx_mode,
+	.ndo_set_mac_address	= sunvnet_set_mac_addr_common,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_tx_timeout		= sunvnet_tx_timeout_common,
+	.ndo_change_mtu		= sunvnet_change_mtu_common,
+	.ndo_start_xmit		= vsw_start_xmit,
+	.ndo_select_queue	= vsw_select_queue,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller    = vsw_poll_controller,
+#endif
+};
+
+static const char *local_mac_prop = "local-mac-address";
+static const char *cfg_handle_prop = "cfg-handle";
+
+static struct vnet *vsw_get_vnet(struct mdesc_handle *hp,
+				 u64 port_node,
+				 u64 *handle)
+{
+	struct vnet *vp;
+	struct vnet *iter;
+	const u64 *local_mac = NULL;
+	const u64 *cfghandle = NULL;
+	u64 a;
+
+	/* Get the parent virtual-network-switch macaddr and cfghandle */
+	mdesc_for_each_arc(a, hp, port_node, MDESC_ARC_TYPE_BACK) {
+		u64 target = mdesc_arc_target(hp, a);
+		const char *name;
+
+		name = mdesc_get_property(hp, target, "name", NULL);
+		if (!name || strcmp(name, "virtual-network-switch"))
+			continue;
+
+		local_mac = mdesc_get_property(hp, target,
+					       local_mac_prop, NULL);
+		cfghandle = mdesc_get_property(hp, target,
+					       cfg_handle_prop, NULL);
+		break;
+	}
+	if (!local_mac || !cfghandle)
+		return ERR_PTR(-ENODEV);
+
+	/* find or create associated vnet */
+	vp = NULL;
+	mutex_lock(&vnet_list_mutex);
+	list_for_each_entry(iter, &vnet_list, list) {
+		if (iter->local_mac == *local_mac) {
+			vp = iter;
+			break;
+		}
+	}
+
+	if (!vp) {
+		vp = kzalloc(sizeof(*vp), GFP_KERNEL);
+		if (unlikely(!vp)) {
+			mutex_unlock(&vnet_list_mutex);
+			return ERR_PTR(-ENOMEM);
+		}
+
+		spin_lock_init(&vp->lock);
+		INIT_LIST_HEAD(&vp->port_list);
+		INIT_LIST_HEAD(&vp->list);
+		vp->local_mac = *local_mac;
+		list_add(&vp->list, &vnet_list);
+	}
+
+	mutex_unlock(&vnet_list_mutex);
+
+	*handle = (u64)*cfghandle;
+
+	return vp;
+}
+
+static struct net_device *vsw_alloc_netdev(u8 hwaddr[],
+					   struct vio_dev *vdev,
+					   u64 handle,
+					   u64 port_id)
+{
+	struct net_device *dev;
+	struct vnet_port *port;
+	int i;
+
+	dev = alloc_etherdev_mqs(sizeof(*port), VNET_MAX_TXQS, 1);
+	if (!dev)
+		return ERR_PTR(-ENOMEM);
+	dev->needed_headroom = VNET_PACKET_SKIP + 8;
+	dev->needed_tailroom = 8;
+
+	for (i = 0; i < ETH_ALEN; i++) {
+		dev->dev_addr[i] = hwaddr[i];
+		dev->perm_addr[i] = dev->dev_addr[i];
+	}
+
+	sprintf(dev->name, "vif%d.%d", (int)handle, (int)port_id);
+
+	dev->netdev_ops = &vsw_ops;
+	dev->ethtool_ops = &vsw_ethtool_ops;
+	dev->watchdog_timeo = VSW_TX_TIMEOUT;
+
+	dev->hw_features = NETIF_F_TSO | NETIF_F_GSO | NETIF_F_GSO_SOFTWARE |
+			   NETIF_F_HW_CSUM | NETIF_F_SG;
+	dev->features = dev->hw_features;
+
+	SET_NETDEV_DEV(dev, &vdev->dev);
+
+	return dev;
+}
+
+static struct ldc_channel_config vsw_ldc_cfg = {
+	.event		= sunvnet_event_common,
+	.mtu		= 64,
+	.mode		= LDC_MODE_UNRELIABLE,
+};
+
+static struct vio_driver_ops vsw_vio_ops = {
+	.send_attr		= sunvnet_send_attr_common,
+	.handle_attr		= sunvnet_handle_attr_common,
+	.handshake_complete	= sunvnet_handshake_complete_common,
+};
+
+static void print_version(void)
+{
+	printk_once(KERN_INFO "%s", version);
+}
+
+static const char *remote_macaddr_prop = "remote-mac-address";
+static const char *id_prop = "id";
+
+static int vsw_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
+{
+	struct mdesc_handle *hp;
+	struct vnet_port *port;
+	unsigned long flags;
+	struct vnet *vp;
+	struct net_device *dev;
+	const u64 *rmac;
+	int len, i, err;
+	const u64 *port_id;
+	u64 handle;
+
+	print_version();
+
+	hp = mdesc_grab();
+
+	rmac = mdesc_get_property(hp, vdev->mp, remote_macaddr_prop, &len);
+	err = -ENODEV;
+	if (!rmac) {
+		pr_err("Port lacks %s property\n", remote_macaddr_prop);
+		mdesc_release(hp);
+		return err;
+	}
+
+	port_id = mdesc_get_property(hp, vdev->mp, id_prop, NULL);
+	err = -ENODEV;
+	if (!port_id) {
+		pr_err("Port lacks %s property\n", id_prop);
+		mdesc_release(hp);
+		return err;
+	}
+
+	/* Get (or create) the vnet associated with this port */
+	vp = vsw_get_vnet(hp, vdev->mp, &handle);
+	if (unlikely(IS_ERR(vp))) {
+		err = PTR_ERR(vp);
+		pr_err("Failed to get vnet for vsw-port\n");
+		mdesc_release(hp);
+		return err;
+	}
+
+	mdesc_release(hp);
+
+	dev = vsw_alloc_netdev(vsw_port_hwaddr, vdev, handle, *port_id);
+	if (IS_ERR(dev)) {
+		err = PTR_ERR(dev);
+		pr_err("Failed to alloc netdev for vsw-port\n");
+		return err;
+	}
+
+	port = netdev_priv(dev);
+
+	INIT_LIST_HEAD(&port->list);
+
+	for (i = 0; i < ETH_ALEN; i++)
+		port->raddr[i] = (*rmac >> (5 - i) * 8) & 0xff;
+
+	port->vp = vp;
+	port->dev = dev;
+	port->switch_port = 1;
+	port->tso = true;
+	port->tsolen = 0;
+
+	/* Mark the port as belonging to ldmvsw which directs the
+	 * the common code to use the net_device in the vnet_port
+	 * rather than the net_device in the vnet (which is used
+	 * by sunvnet). This bit is used by the VNET_PORT_TO_NET_DEVICE
+	 * macro.
+	 */
+	port->vsw = 1;
+
+	err = vio_driver_init(&port->vio, vdev, VDEV_NETWORK,
+			      vsw_versions, ARRAY_SIZE(vsw_versions),
+			      &vsw_vio_ops, dev->name);
+	if (err)
+		goto err_out_free_dev;
+
+	err = vio_ldc_alloc(&port->vio, &vsw_ldc_cfg, port);
+	if (err)
+		goto err_out_free_dev;
+
+	dev_set_drvdata(&vdev->dev, port);
+
+	netif_napi_add(dev, &port->napi, sunvnet_poll_common,
+		       NAPI_POLL_WEIGHT);
+
+	spin_lock_irqsave(&vp->lock, flags);
+	list_add_rcu(&port->list, &vp->port_list);
+	spin_unlock_irqrestore(&vp->lock, flags);
+
+	setup_timer(&port->clean_timer, sunvnet_clean_timer_expire_common,
+		    (unsigned long)port);
+
+	err = register_netdev(dev);
+	if (err) {
+		pr_err("Cannot register net device, aborting\n");
+		goto err_out_del_timer;
+	}
+
+	spin_lock_irqsave(&vp->lock, flags);
+	sunvnet_port_add_txq_common(port);
+	spin_unlock_irqrestore(&vp->lock, flags);
+
+	napi_enable(&port->napi);
+	vio_port_up(&port->vio);
+
+	netdev_info(dev, "LDOM vsw-port %pM\n", dev->dev_addr);
+
+	pr_info("%s: PORT ( remote-mac %pM%s )\n", dev->name,
+		port->raddr, " switch-port");
+
+	return 0;
+
+err_out_del_timer:
+	del_timer_sync(&port->clean_timer);
+	list_del_rcu(&port->list);
+	synchronize_rcu();
+	netif_napi_del(&port->napi);
+	dev_set_drvdata(&vdev->dev, NULL);
+	vio_ldc_free(&port->vio);
+
+err_out_free_dev:
+	free_netdev(dev);
+	return err;
+}
+
+static int vsw_port_remove(struct vio_dev *vdev)
+{
+	struct vnet_port *port = dev_get_drvdata(&vdev->dev);
+	unsigned long flags;
+
+	if (port) {
+		del_timer_sync(&port->vio.timer);
+
+		napi_disable(&port->napi);
+
+		list_del_rcu(&port->list);
+
+		synchronize_rcu();
+		del_timer_sync(&port->clean_timer);
+		spin_lock_irqsave(&port->vp->lock, flags);
+		sunvnet_port_rm_txq_common(port);
+		spin_unlock_irqrestore(&port->vp->lock, flags);
+		netif_napi_del(&port->napi);
+		sunvnet_port_free_tx_bufs_common(port);
+		vio_ldc_free(&port->vio);
+
+		dev_set_drvdata(&vdev->dev, NULL);
+
+		unregister_netdev(port->dev);
+		free_netdev(port->dev);
+	}
+
+	return 0;
+}
+
+static void vsw_cleanup(void)
+{
+	struct vnet *vp;
+
+	/* just need to free up the vnet list */
+	mutex_lock(&vnet_list_mutex);
+	while (!list_empty(&vnet_list)) {
+		vp = list_first_entry(&vnet_list, struct vnet, list);
+		list_del(&vp->list);
+		/* vio_unregister_driver() should have cleaned up port_list */
+		if (!list_empty(&vp->port_list))
+			pr_err("Ports not removed by VIO subsystem!\n");
+		kfree(vp);
+	}
+	mutex_unlock(&vnet_list_mutex);
+}
+
+static const struct vio_device_id vsw_port_match[] = {
+	{
+		.type = "vsw-port",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(vio, vsw_port_match);
+
+static struct vio_driver vsw_port_driver = {
+	.id_table	= vsw_port_match,
+	.probe		= vsw_port_probe,
+	.remove		= vsw_port_remove,
+	.name		= "vsw_port",
+};
+
+static int __init vsw_init(void)
+{
+	return vio_register_driver(&vsw_port_driver);
+}
+
+static void __exit vsw_exit(void)
+{
+	vio_unregister_driver(&vsw_port_driver);
+	vsw_cleanup();
+}
+
+module_init(vsw_init);
+module_exit(vsw_exit);
-- 
cgit v0.10.2


From dc153f850daba6eb665fbfedd349d09bcfd9bda9 Mon Sep 17 00:00:00 2001
From: Aaron Young <aaron.young@oracle.com>
Date: Tue, 15 Mar 2016 11:35:40 -0700
Subject: ldmvsw: Checkpatch sunvnet.c and sunvnet_common.c

  Checkpatch updates for sunvnet.c and sunvnet_common.c.

  Signed-off-by: Aaron Young <aaron.young@oracle.com>
  Signed-off-by: Rashmi Narasimhan <rashmi.narasimhan@oracle.com>
  Reviewed-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
  Reviewed-by: Alexandre Chartre <Alexandre.Chartre@oracle.com>

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/sun/sunvnet.c b/drivers/net/ethernet/sun/sunvnet.c
index 98c5f16..a2f9b47 100644
--- a/drivers/net/ethernet/sun/sunvnet.c
+++ b/drivers/net/ethernet/sun/sunvnet.c
@@ -66,12 +66,14 @@ static void vnet_get_drvinfo(struct net_device *dev,
 static u32 vnet_get_msglevel(struct net_device *dev)
 {
 	struct vnet *vp = netdev_priv(dev);
+
 	return vp->msg_enable;
 }
 
 static void vnet_set_msglevel(struct net_device *dev, u32 value)
 {
 	struct vnet *vp = netdev_priv(dev);
+
 	vp->msg_enable = value;
 }
 
@@ -359,7 +361,7 @@ static int vnet_port_probe(struct vio_dev *vdev, const struct vio_device_id *id)
 	INIT_LIST_HEAD(&port->list);
 
 	switch_port = 0;
-	if (mdesc_get_property(hp, vdev->mp, "switch-port", NULL) != NULL)
+	if (mdesc_get_property(hp, vdev->mp, "switch-port", NULL))
 		switch_port = 1;
 	port->switch_port = switch_port;
 	port->tso = true;
@@ -403,7 +405,6 @@ static int vnet_port_remove(struct vio_dev *vdev)
 	struct vnet_port *port = dev_get_drvdata(&vdev->dev);
 
 	if (port) {
-
 		del_timer_sync(&port->vio.timer);
 
 		napi_disable(&port->napi);
@@ -421,7 +422,6 @@ static int vnet_port_remove(struct vio_dev *vdev)
 		dev_set_drvdata(&vdev->dev, NULL);
 
 		kfree(port);
-
 	}
 	return 0;
 }
diff --git a/drivers/net/ethernet/sun/sunvnet_common.c b/drivers/net/ethernet/sun/sunvnet_common.c
index 083f41c..904a5a1 100644
--- a/drivers/net/ethernet/sun/sunvnet_common.c
+++ b/drivers/net/ethernet/sun/sunvnet_common.c
@@ -119,7 +119,6 @@ int sunvnet_send_attr_common(struct vio_driver_state *vio)
 	       pkt.ack_freq, pkt.plnk_updt, pkt.options,
 	       (unsigned long long)pkt.mtu, pkt.cflags, pkt.ipv4_lso_maxlen);
 
-
 	return vio_ldc_send(vio, &pkt, sizeof(pkt));
 }
 EXPORT_SYMBOL_GPL(sunvnet_send_attr_common);
@@ -197,24 +196,23 @@ static int handle_attr_info(struct vio_driver_state *vio,
 
 		pkt->tag.stype = VIO_SUBTYPE_NACK;
 
-		(void) vio_ldc_send(vio, pkt, sizeof(*pkt));
+		(void)vio_ldc_send(vio, pkt, sizeof(*pkt));
 
 		return -ECONNRESET;
-	} else {
-		viodbg(HS, "SEND NET ATTR ACK xmode[0x%x] atype[0x%x] "
-		       "addr[%llx] ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] "
-		       "mtu[%llu] (rmtu[%llu]) cflags[0x%04x] lso_max[%u]\n",
-		       pkt->xfer_mode, pkt->addr_type,
-		       (unsigned long long)pkt->addr,
-		       pkt->ack_freq, pkt->plnk_updt, pkt->options,
-		       (unsigned long long)pkt->mtu, port->rmtu, pkt->cflags,
-		       pkt->ipv4_lso_maxlen);
-
-		pkt->tag.stype = VIO_SUBTYPE_ACK;
-
-		return vio_ldc_send(vio, pkt, sizeof(*pkt));
 	}
 
+	viodbg(HS, "SEND NET ATTR ACK xmode[0x%x] atype[0x%x] "
+	       "addr[%llx] ackfreq[%u] plnk_updt[0x%02x] opts[0x%02x] "
+	       "mtu[%llu] (rmtu[%llu]) cflags[0x%04x] lso_max[%u]\n",
+	       pkt->xfer_mode, pkt->addr_type,
+	       (unsigned long long)pkt->addr,
+	       pkt->ack_freq, pkt->plnk_updt, pkt->options,
+	       (unsigned long long)pkt->mtu, port->rmtu, pkt->cflags,
+	       pkt->ipv4_lso_maxlen);
+
+	pkt->tag.stype = VIO_SUBTYPE_ACK;
+
+	return vio_ldc_send(vio, pkt, sizeof(*pkt));
 }
 
 static int handle_attr_ack(struct vio_driver_state *vio,
@@ -258,10 +256,12 @@ void sunvnet_handshake_complete_common(struct vio_driver_state *vio)
 	struct vio_dring_state *dr;
 
 	dr = &vio->drings[VIO_DRIVER_RX_RING];
-	dr->snd_nxt = dr->rcv_nxt = 1;
+	dr->rcv_nxt = 1;
+	dr->snd_nxt = 1;
 
 	dr = &vio->drings[VIO_DRIVER_TX_RING];
-	dr->snd_nxt = dr->rcv_nxt = 1;
+	dr->rcv_nxt = 1;
+	dr->snd_nxt = 1;
 }
 EXPORT_SYMBOL_GPL(sunvnet_handshake_complete_common);
 
@@ -283,13 +283,14 @@ EXPORT_SYMBOL_GPL(sunvnet_handshake_complete_common);
 static struct sk_buff *alloc_and_align_skb(struct net_device *dev,
 					   unsigned int len)
 {
-	struct sk_buff *skb = netdev_alloc_skb(dev, len+VNET_PACKET_SKIP+8+8);
+	struct sk_buff *skb;
 	unsigned long addr, off;
 
+	skb = netdev_alloc_skb(dev, len + VNET_PACKET_SKIP + 8 + 8);
 	if (unlikely(!skb))
 		return NULL;
 
-	addr = (unsigned long) skb->data;
+	addr = (unsigned long)skb->data;
 	off = ((addr + 7UL) & ~7UL) - addr;
 	if (off)
 		skb_reserve(skb, off);
@@ -505,7 +506,7 @@ static int vnet_walk_rx_one(struct vnet_port *port,
 	struct vio_driver_state *vio = &port->vio;
 	int err;
 
-	BUG_ON(desc == NULL);
+	BUG_ON(!desc);
 	if (IS_ERR(desc))
 		return PTR_ERR(desc);
 
@@ -540,13 +541,14 @@ static int vnet_walk_rx(struct vnet_port *port, struct vio_dring_state *dr,
 	int ack_start = -1, ack_end = -1;
 	bool send_ack = true;
 
-	end = (end == (u32) -1) ? vio_dring_prev(dr, start)
-				: vio_dring_next(dr, end);
+	end = (end == (u32)-1) ? vio_dring_prev(dr, start)
+			       : vio_dring_next(dr, end);
 
 	viodbg(DATA, "vnet_walk_rx start[%08x] end[%08x]\n", start, end);
 
 	while (start != end) {
 		int ack = 0, err = vnet_walk_rx_one(port, dr, start, &ack);
+
 		if (err == -ECONNRESET)
 			return err;
 		if (err != 0)
@@ -568,8 +570,10 @@ static int vnet_walk_rx(struct vnet_port *port, struct vio_dring_state *dr,
 			break;
 		}
 	}
-	if (unlikely(ack_start == -1))
-		ack_start = ack_end = vio_dring_prev(dr, start);
+	if (unlikely(ack_start == -1)) {
+		ack_end = vio_dring_prev(dr, start);
+		ack_start = ack_end;
+	}
 	if (send_ack) {
 		port->napi_resume = false;
 		trace_vnet_tx_send_stopped_ack(port->vio._local_sid,
@@ -749,7 +753,7 @@ ldc_ctrl:
 	}
 	/* We may have multiple LDC events in rx_event. Unroll send_events() */
 	event = (port->rx_event & LDC_EVENT_UP);
-	port->rx_event &= ~(LDC_EVENT_RESET|LDC_EVENT_UP);
+	port->rx_event &= ~(LDC_EVENT_RESET | LDC_EVENT_UP);
 	if (event == LDC_EVENT_UP)
 		goto ldc_ctrl;
 	event = port->rx_event;
@@ -759,7 +763,8 @@ ldc_ctrl:
 	/* we dont expect any other bits than RESET, UP, DATA_READY */
 	BUG_ON(event != LDC_EVENT_DATA_READY);
 
-	tx_wakeup = err = 0;
+	err = 0;
+	tx_wakeup = 0;
 	while (1) {
 		union {
 			struct vio_msg_tag tag;
@@ -776,7 +781,8 @@ ldc_ctrl:
 			pkt->tag.stype = VIO_SUBTYPE_INFO;
 			pkt->tag.stype_env = VIO_DRING_DATA;
 			pkt->seq = dr->rcv_nxt;
-			pkt->start_idx = vio_dring_next(dr, port->napi_stop_idx);
+			pkt->start_idx = vio_dring_next(dr,
+							port->napi_stop_idx);
 			pkt->end_idx = -1;
 			goto napi_resume;
 		}
@@ -860,7 +866,6 @@ void sunvnet_event_common(void *arg, int event)
 	port->rx_event |= event;
 	vio_set_intr(vio->vdev->rx_ino, HV_INTR_DISABLED);
 	napi_schedule(&port->napi);
-
 }
 EXPORT_SYMBOL_GPL(sunvnet_event_common);
 
@@ -876,7 +881,7 @@ static int __vnet_tx_trigger(struct vnet_port *port, u32 start)
 		},
 		.dring_ident		= dr->ident,
 		.start_idx		= start,
-		.end_idx		= (u32) -1,
+		.end_idx		= (u32)-1,
 	};
 	int err, delay;
 	int retries = 0;
@@ -928,7 +933,7 @@ static struct sk_buff *vnet_clean_tx_ring(struct vnet_port *port,
 
 		--txi;
 		if (txi < 0)
-			txi = VNET_TX_RING_SIZE-1;
+			txi = VNET_TX_RING_SIZE - 1;
 
 		d = vio_dring_entry(dr, txi);
 
@@ -949,8 +954,9 @@ static struct sk_buff *vnet_clean_tx_ring(struct vnet_port *port,
 			ldc_unmap(port->vio.lp,
 				  port->tx_bufs[txi].cookies,
 				  port->tx_bufs[txi].ncookies);
-		} else if (d->hdr.state == VIO_DESC_FREE)
+		} else if (d->hdr.state == VIO_DESC_FREE) {
 			break;
+		}
 		d->hdr.state = VIO_DESC_FREE;
 	}
 	return skb;
@@ -1001,7 +1007,7 @@ static inline int vnet_skb_map(struct ldc_channel *lp, struct sk_buff *skb,
 	blen += VNET_PACKET_SKIP;
 	blen += 8 - (blen & 7);
 
-	err = ldc_map_single(lp, skb->data-VNET_PACKET_SKIP, blen, cookies,
+	err = ldc_map_single(lp, skb->data - VNET_PACKET_SKIP, blen, cookies,
 			     ncookies, map_perm);
 	if (err < 0)
 		return err;
@@ -1061,7 +1067,7 @@ static inline struct sk_buff *vnet_skb_shape(struct sk_buff *skb, int ncookies)
 
 		len = skb->len > ETH_ZLEN ? skb->len : ETH_ZLEN;
 		nskb = alloc_and_align_skb(skb->dev, len);
-		if (nskb == NULL) {
+		if (!nskb) {
 			dev_kfree_skb(skb);
 			return NULL;
 		}
@@ -1138,11 +1144,11 @@ static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb,
 	else if (skb->protocol == htons(ETH_P_IPV6))
 		proto = ipv6_hdr(skb)->nexthdr;
 
-	if (proto == IPPROTO_TCP)
+	if (proto == IPPROTO_TCP) {
 		hlen += tcp_hdr(skb)->doff * 4;
-	else if (proto == IPPROTO_UDP)
+	} else if (proto == IPPROTO_UDP) {
 		hlen += sizeof(struct udphdr);
-	else {
+	} else {
 		pr_err("vnet_handle_offloads GSO with unknown transport "
 		       "protocol %d tproto %d\n", skb->protocol, proto);
 		hlen = 128; /* XXX */
@@ -1195,8 +1201,9 @@ static int vnet_handle_offloads(struct vnet_port *port, struct sk_buff *skb,
 			skb_shinfo(curr)->gso_type = gso_type;
 			skb_shinfo(curr)->gso_segs =
 				DIV_ROUND_UP(curr->len - hlen, gso_size);
-		} else
+		} else {
 			skb_shinfo(curr)->gso_size = 0;
+		}
 
 		skb_push(curr, maclen);
 		skb_reset_mac_header(curr);
@@ -1521,14 +1528,14 @@ static void __send_mc_list(struct vnet *vp, struct vnet_port *port)
 		if (++n_addrs == VNET_NUM_MCAST) {
 			info.count = n_addrs;
 
-			(void) vio_ldc_send(&port->vio, &info,
-					    sizeof(info));
+			(void)vio_ldc_send(&port->vio, &info,
+					   sizeof(info));
 			n_addrs = 0;
 		}
 	}
 	if (n_addrs) {
 		info.count = n_addrs;
-		(void) vio_ldc_send(&port->vio, &info, sizeof(info));
+		(void)vio_ldc_send(&port->vio, &info, sizeof(info));
 	}
 
 	info.set = 0;
@@ -1546,8 +1553,8 @@ static void __send_mc_list(struct vnet *vp, struct vnet_port *port)
 		       m->addr, ETH_ALEN);
 		if (++n_addrs == VNET_NUM_MCAST) {
 			info.count = n_addrs;
-			(void) vio_ldc_send(&port->vio, &info,
-					    sizeof(info));
+			(void)vio_ldc_send(&port->vio, &info,
+					   sizeof(info));
 			n_addrs = 0;
 		}
 
@@ -1556,7 +1563,7 @@ static void __send_mc_list(struct vnet *vp, struct vnet_port *port)
 	}
 	if (n_addrs) {
 		info.count = n_addrs;
-		(void) vio_ldc_send(&port->vio, &info, sizeof(info));
+		(void)vio_ldc_send(&port->vio, &info, sizeof(info));
 	}
 }
 
@@ -1566,7 +1573,6 @@ void sunvnet_set_rx_mode_common(struct net_device *dev, struct vnet *vp)
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(port, &vp->port_list, list) {
-
 		if (port->switch_port) {
 			__update_mc_list(vp, dev);
 			__send_mc_list(vp, port);
@@ -1600,7 +1606,7 @@ void sunvnet_port_free_tx_bufs_common(struct vnet_port *port)
 
 	dr = &port->vio.drings[VIO_DRIVER_TX_RING];
 
-	if (dr->base == NULL)
+	if (!dr->base)
 		return;
 
 	for (i = 0; i < VNET_TX_RING_SIZE; i++) {
@@ -1668,7 +1674,8 @@ static int vnet_port_alloc_tx_ring(struct vnet_port *port)
 	dr->base = dring;
 	dr->entry_size = elen;
 	dr->num_entries = VNET_TX_RING_SIZE;
-	dr->prod = dr->cons = 0;
+	dr->prod = 0;
+	dr->cons = 0;
 	port->start_cons  = true; /* need an initial trigger */
 	dr->pending = VNET_TX_RING_SIZE;
 	dr->ncookies = ncookies;
@@ -1713,7 +1720,6 @@ void sunvnet_port_add_txq_common(struct vnet_port *port)
 	port->q_index = n;
 	netif_tx_wake_queue(netdev_get_tx_queue(VNET_PORT_TO_NET_DEVICE(port),
 						port->q_index));
-
 }
 EXPORT_SYMBOL_GPL(sunvnet_port_add_txq_common);
 
-- 
cgit v0.10.2


From 3a461da1d03e7a857edfa6a002040d07e118c639 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Tue, 15 Mar 2016 22:41:22 +0100
Subject: cls_bpf: reset class and reuse major in da

There are two issues with the current code. First one is that we need
to set res->class to 0 in case we use non-default classid matching.

This is important for the case where cls_bpf was initially set up with
an optional binding to a default class with tcf_bind_filter(), where
the underlying qdisc implements bind_tcf() that fills res->class and
tests for it later on when doing the classification. Convention for
these cases is that after tc_classify() was called, such qdiscs (atm,
drr, qfq, cbq, hfsc, htb) first test class, and if 0, then they lookup
based on classid.

Second, there's a bug with da mode, where res->classid is only assigned
a 16 bit minor, but it needs to expand to the full 32 bit major/minor
combination instead, therefore we need to expand with the bound major.
This is fine as classes belonging to a classful qdisc must share the
same major.

Fixes: 045efa82ff56 ("cls_bpf: introduce integrated actions")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index 8dc8430..425fe6a 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -103,8 +103,9 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 		}
 
 		if (prog->exts_integrated) {
-			res->class = prog->res.class;
-			res->classid = qdisc_skb_cb(skb)->tc_classid;
+			res->class   = 0;
+			res->classid = TC_H_MAJ(prog->res.classid) |
+				       qdisc_skb_cb(skb)->tc_classid;
 
 			ret = cls_bpf_exec_opcode(filter_res);
 			if (ret == TC_ACT_UNSPEC)
@@ -114,10 +115,12 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 
 		if (filter_res == 0)
 			continue;
-
-		*res = prog->res;
-		if (filter_res != -1)
+		if (filter_res != -1) {
+			res->class   = 0;
 			res->classid = filter_res;
+		} else {
+			*res = prog->res;
+		}
 
 		ret = tcf_exts_exec(skb, &prog->exts, res);
 		if (ret < 0)
-- 
cgit v0.10.2


From 019ded3aa7c9799fbe6533baeac9aafc7063bd39 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 15 Mar 2016 22:47:14 +0100
Subject: net: mvneta: bm: clarify dependencies

MVNETA_BM has a dependency on MVNETA, so we can only select the former
if the latter is enabled. However, the code dependency is the reverse:
The mvneta module can call into the mvneta_bm module, so mvneta cannot
be a built-in if mvneta_bm is a module, or we get a link error:

drivers/net/built-in.o: In function `mvneta_remove':
drivers/net/ethernet/marvell/mvneta.c:4211: undefined reference to `mvneta_bm_pool_destroy'
drivers/net/built-in.o: In function `mvneta_bm_update_mtu':
drivers/net/ethernet/marvell/mvneta.c:1034: undefined reference to `mvneta_bm_bufs_free'

This avoids the problem by further clarifying the dependency so that
MVNETA_BM is a silent Kconfig option that gets turned on by the
new MVNETA_BM_ENABLE option. This way both the core HWBM module and
the MVNETA_BM code are always built-in when needed.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: dc35a10f68d3 ("net: mvneta: bm: add support for hardware buffer management")
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index 62d80fd..b5c6d42 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -40,10 +40,9 @@ config MVMDIO
 
 	  This driver is used by the MV643XX_ETH and MVNETA drivers.
 
-config MVNETA_BM
+config MVNETA_BM_ENABLE
 	tristate "Marvell Armada 38x/XP network interface BM support"
 	depends on MVNETA
-	select HWBM
 	---help---
 	  This driver supports auxiliary block of the network
 	  interface units in the Marvell ARMADA XP and ARMADA 38x SoC
@@ -67,6 +66,15 @@ config MVNETA
 	  driver, which should be used for the older Marvell SoCs
 	  (Dove, Orion, Discovery, Kirkwood).
 
+config MVNETA_BM
+	tristate
+	default y if MVNETA=y && MVNETA_BM_ENABLE
+	default MVNETA_BM_ENABLE
+	select HWBM
+	help
+	  MVNETA_BM must not be 'm' if MVNETA=y, so this symbol ensures
+	  that all dependencies are met.
+
 config MVPP2
 	tristate "Marvell Armada 375 network interface support"
 	depends on MACH_ARMADA_375
-- 
cgit v0.10.2


From 09c37a2c5bbc28b5fbc07a01db4bccdbd0a5d8a2 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 16 Mar 2016 01:42:49 +0100
Subject: bpf: make skb->tc_classid also readable

Currently, the tc_classid from eBPF skb context is write-only, but there's
no good reason for tc programs to limit it to write-only. For example,
it can be used to transfer its state via tail calls where the resulting
tc_classid gets filled gradually.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/core/filter.c b/net/core/filter.c
index 6fc3893..69c7b2f 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2069,16 +2069,14 @@ static bool sk_filter_is_valid_access(int off, int size,
 static bool tc_cls_act_is_valid_access(int off, int size,
 				       enum bpf_access_type type)
 {
-	if (off == offsetof(struct __sk_buff, tc_classid))
-		return type == BPF_WRITE ? true : false;
-
 	if (type == BPF_WRITE) {
 		switch (off) {
 		case offsetof(struct __sk_buff, mark):
 		case offsetof(struct __sk_buff, tc_index):
 		case offsetof(struct __sk_buff, priority):
 		case offsetof(struct __sk_buff, cb[0]) ...
-			offsetof(struct __sk_buff, cb[4]):
+		     offsetof(struct __sk_buff, cb[4]):
+		case offsetof(struct __sk_buff, tc_classid):
 			break;
 		default:
 			return false;
@@ -2195,8 +2193,10 @@ static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg,
 		ctx_off -= offsetof(struct __sk_buff, tc_classid);
 		ctx_off += offsetof(struct sk_buff, cb);
 		ctx_off += offsetof(struct qdisc_skb_cb, tc_classid);
-		WARN_ON(type != BPF_WRITE);
-		*insn++ = BPF_STX_MEM(BPF_H, dst_reg, src_reg, ctx_off);
+		if (type == BPF_WRITE)
+			*insn++ = BPF_STX_MEM(BPF_H, dst_reg, src_reg, ctx_off);
+		else
+			*insn++ = BPF_LDX_MEM(BPF_H, dst_reg, src_reg, ctx_off);
 		break;
 
 	case offsetof(struct __sk_buff, tc_index):
-- 
cgit v0.10.2


From 808c1b697c3c4dd2a7132882424c390b0d0acfb9 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 16 Mar 2016 01:42:50 +0100
Subject: bpf, dst: add and use dst_tclassid helper

We can just add a small helper dst_tclassid() for retrieving the
dst->tclassid value. It makes the code a bit better in that we can
get rid of the ifdef from filter.c by moving this into the header.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/dst.h b/include/net/dst.h
index c7329dc..5c98443 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -398,6 +398,18 @@ static inline void skb_tunnel_rx(struct sk_buff *skb, struct net_device *dev,
 	__skb_tunnel_rx(skb, dev, net);
 }
 
+static inline u32 dst_tclassid(const struct sk_buff *skb)
+{
+#ifdef CONFIG_IP_ROUTE_CLASSID
+	const struct dst_entry *dst;
+
+	dst = skb_dst(skb);
+	if (dst)
+		return dst->tclassid;
+#endif
+	return 0;
+}
+
 int dst_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb);
 static inline int dst_discard(struct sk_buff *skb)
 {
diff --git a/net/core/filter.c b/net/core/filter.c
index 69c7b2f..4c35d83 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1682,14 +1682,7 @@ static const struct bpf_func_proto bpf_get_cgroup_classid_proto = {
 
 static u64 bpf_get_route_realm(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
 {
-#ifdef CONFIG_IP_ROUTE_CLASSID
-	const struct dst_entry *dst;
-
-	dst = skb_dst((struct sk_buff *) (unsigned long) r1);
-	if (dst)
-		return dst->tclassid;
-#endif
-	return 0;
+	return dst_tclassid((struct sk_buff *) (unsigned long) r1);
 }
 
 static const struct bpf_func_proto bpf_get_route_realm_proto = {
-- 
cgit v0.10.2


From fca5fdf67de9e092fda23c9eb059ba968e7b5267 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Wed, 16 Mar 2016 01:42:51 +0100
Subject: ip_tunnels, bpf: define IP_TUNNEL_OPTS_MAX and use it

eBPF defines this as BPF_TUNLEN_MAX and OVS just uses the hard-coded
value inside struct sw_flow_key. Thus, add and use IP_TUNNEL_OPTS_MAX
for this, which makes the code a bit more generic and allows to remove
BPF_TUNLEN_MAX from eBPF code.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 5dc2e45..c35dda9 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -7,6 +7,8 @@
 #include <linux/socket.h>
 #include <linux/types.h>
 #include <linux/u64_stats_sync.h>
+#include <linux/bitops.h>
+
 #include <net/dsfield.h>
 #include <net/gro_cells.h>
 #include <net/inet_ecn.h>
@@ -57,6 +59,11 @@ struct ip_tunnel_key {
 #define IP_TUNNEL_INFO_TX	0x01	/* represents tx tunnel parameters */
 #define IP_TUNNEL_INFO_IPV6	0x02	/* key contains IPv6 addresses */
 
+/* Maximum tunnel options length. */
+#define IP_TUNNEL_OPTS_MAX					\
+	GENMASK((FIELD_SIZEOF(struct ip_tunnel_info,		\
+			      options_len) * BITS_PER_BYTE) - 1, 0)
+
 struct ip_tunnel_info {
 	struct ip_tunnel_key	key;
 #ifdef CONFIG_DST_CACHE
diff --git a/net/core/filter.c b/net/core/filter.c
index 4c35d83..b7177d0 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1904,8 +1904,6 @@ static const struct bpf_func_proto bpf_skb_set_tunnel_key_proto = {
 	.arg4_type	= ARG_ANYTHING,
 };
 
-#define BPF_TUNLEN_MAX	255
-
 static u64 bpf_skb_set_tunnel_opt(u64 r1, u64 r2, u64 size, u64 r4, u64 r5)
 {
 	struct sk_buff *skb = (struct sk_buff *) (long) r1;
@@ -1915,7 +1913,7 @@ static u64 bpf_skb_set_tunnel_opt(u64 r1, u64 r2, u64 size, u64 r4, u64 r5)
 
 	if (unlikely(info != &md->u.tun_info || (size & (sizeof(u32) - 1))))
 		return -EINVAL;
-	if (unlikely(size > BPF_TUNLEN_MAX))
+	if (unlikely(size > IP_TUNNEL_OPTS_MAX))
 		return -ENOMEM;
 
 	ip_tunnel_info_opts_set(info, from, size);
@@ -1936,13 +1934,10 @@ static const struct bpf_func_proto *
 bpf_get_skb_set_tunnel_proto(enum bpf_func_id which)
 {
 	if (!md_dst) {
-		BUILD_BUG_ON(FIELD_SIZEOF(struct ip_tunnel_info,
-					  options_len) != 1);
-
 		/* Race is not possible, since it's called from verifier
 		 * that is holding verifier mutex.
 		 */
-		md_dst = metadata_dst_alloc_percpu(BPF_TUNLEN_MAX,
+		md_dst = metadata_dst_alloc_percpu(IP_TUNNEL_OPTS_MAX,
 						   GFP_KERNEL);
 		if (!md_dst)
 			return NULL;
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index eaca244..d27276f 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -398,6 +398,12 @@ static const struct lwtunnel_encap_ops ip6_tun_lwt_ops = {
 
 void __init ip_tunnel_core_init(void)
 {
+	/* If you land here, make sure whether increasing ip_tunnel_info's
+	 * options_len is a reasonable choice with its usage in front ends
+	 * (f.e., it's part of flow keys, etc).
+	 */
+	BUILD_BUG_ON(IP_TUNNEL_OPTS_MAX != 255);
+
 	lwtunnel_encap_add_ops(&ip_tun_lwt_ops, LWTUNNEL_ENCAP_IP);
 	lwtunnel_encap_add_ops(&ip6_tun_lwt_ops, LWTUNNEL_ENCAP_IP6);
 }
diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
index 1d055c5..03378e7 100644
--- a/net/openvswitch/flow.h
+++ b/net/openvswitch/flow.h
@@ -55,7 +55,7 @@ struct ovs_tunnel_info {
 	FIELD_SIZEOF(struct sw_flow_key, recirc_id))
 
 struct sw_flow_key {
-	u8 tun_opts[255];
+	u8 tun_opts[IP_TUNNEL_OPTS_MAX];
 	u8 tun_opts_len;
 	struct ip_tunnel_key tun_key;	/* Encapsulating tunnel key. */
 	struct {
-- 
cgit v0.10.2


From 3af0d554c1ce11e9d0953381ff566271f9ab81a9 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 16 Mar 2016 10:45:10 +0300
Subject: ethernet: micrel: fix some error codes

There were two issues here:
1) dma_mapping_error() return true/false but we want to return -ENOMEM
2) If dmaengine_prep_slave_sg() failed then "err" wasn't set but
   presumably that should be -ENOMEM as well.

I changed the success path to "return 0;" instead of "return ret;" for
clarity.

Fixes: 94fe8c683cea ('ks8842: Support DMA when accessed via timberdale')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/micrel/ks8842.c b/drivers/net/ethernet/micrel/ks8842.c
index 09d2e16..cb0102d 100644
--- a/drivers/net/ethernet/micrel/ks8842.c
+++ b/drivers/net/ethernet/micrel/ks8842.c
@@ -561,8 +561,8 @@ static int __ks8842_start_new_rx_dma(struct net_device *netdev)
 		sg_init_table(sg, 1);
 		sg_dma_address(sg) = dma_map_single(adapter->dev,
 			ctl->skb->data, DMA_BUFFER_SIZE, DMA_FROM_DEVICE);
-		err = dma_mapping_error(adapter->dev, sg_dma_address(sg));
-		if (unlikely(err)) {
+		if (dma_mapping_error(adapter->dev, sg_dma_address(sg))) {
+			err = -ENOMEM;
 			sg_dma_address(sg) = 0;
 			goto out;
 		}
@@ -572,8 +572,10 @@ static int __ks8842_start_new_rx_dma(struct net_device *netdev)
 		ctl->adesc = dmaengine_prep_slave_sg(ctl->chan,
 			sg, 1, DMA_DEV_TO_MEM, DMA_PREP_INTERRUPT);
 
-		if (!ctl->adesc)
+		if (!ctl->adesc) {
+			err = -ENOMEM;
 			goto out;
+		}
 
 		ctl->adesc->callback_param = netdev;
 		ctl->adesc->callback = ks8842_dma_rx_cb;
@@ -584,7 +586,7 @@ static int __ks8842_start_new_rx_dma(struct net_device *netdev)
 		goto out;
 	}
 
-	return err;
+	return 0;
 out:
 	if (sg_dma_address(sg))
 		dma_unmap_single(adapter->dev, sg_dma_address(sg),
-- 
cgit v0.10.2


From 93e68cd6115f67d8363c94dae8206af36f6d3b00 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Wed, 16 Mar 2016 09:12:46 +0000
Subject: net: fix a comment typo

Fix a comment typo.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/uapi/linux/if.h b/include/uapi/linux/if.h
index 9cf2394..f802775 100644
--- a/include/uapi/linux/if.h
+++ b/include/uapi/linux/if.h
@@ -37,7 +37,7 @@
  * are shared for all types of net_devices. The sysfs entries are available
  * via /sys/class/net/<dev>/flags. Flags which can be toggled through sysfs
  * are annotated below, note that only a few flags can be toggled and some
- * other flags are always always preserved from the original net_device flags
+ * other flags are always preserved from the original net_device flags
  * even if you try to set them via sysfs. Flags which are always preserved
  * are kept under the flag grouping @IFF_VOLATILE. Flags which are volatile
  * are annotated below as such.
-- 
cgit v0.10.2


From 1098cee6e1b11e4fe80b2765eb54d412d4294f03 Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Wed, 16 Mar 2016 09:59:15 +0000
Subject: bonding: remove duplicate set of flag IFF_MULTICAST

Remove unnecessary set of flag IFF_MULTICAST, since ether_setup
already does this.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: Andy Gospodarek <gospo@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index b6236ff..270b39c 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -4175,7 +4175,7 @@ void bond_setup(struct net_device *bond_dev)
 	SET_NETDEV_DEVTYPE(bond_dev, &bond_type);
 
 	/* Initialize the device options */
-	bond_dev->flags |= IFF_MASTER|IFF_MULTICAST;
+	bond_dev->flags |= IFF_MASTER;
 	bond_dev->priv_flags |= IFF_BONDING | IFF_UNICAST_FLT | IFF_NO_QUEUE;
 	bond_dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
 
-- 
cgit v0.10.2


From bc0df13887b1147f4e63390d7d6767f7fa64b91b Mon Sep 17 00:00:00 2001
From: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Date: Wed, 16 Mar 2016 09:59:16 +0000
Subject: team: remove duplicate set of flag IFF_MULTICAST

Remove unnecessary set of flag IFF_MULTICAST, since ether_setup
already does this.

Signed-off-by: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 2769835..26c64d2 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -2082,7 +2082,6 @@ static void team_setup(struct net_device *dev)
 	dev->netdev_ops = &team_netdev_ops;
 	dev->ethtool_ops = &team_ethtool_ops;
 	dev->destructor	= team_destructor;
-	dev->flags |= IFF_MULTICAST;
 	dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
 	dev->priv_flags |= IFF_NO_QUEUE;
 	dev->priv_flags |= IFF_TEAM;
-- 
cgit v0.10.2


From 79d3b59a93ba25f3b2c72eb4099c189d41d30204 Mon Sep 17 00:00:00 2001
From: Robert Jarzmik <robert.jarzmik@free.fr>
Date: Wed, 16 Mar 2016 18:26:02 +0100
Subject: net: smc911x: convert pxa dma to dmaengine

Convert the dma transfers to be dmaengine based, now pxa has a dmaengine
slave driver. This makes this driver a bit more PXA agnostic.

The driver was only compile tested. The risk is quite small as no
current PXA platform I'm aware of is using smc911x driver.

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Tested-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/smsc/smc911x.c b/drivers/net/ethernet/smsc/smc911x.c
index bd64eb9..3f57110 100644
--- a/drivers/net/ethernet/smsc/smc911x.c
+++ b/drivers/net/ethernet/smsc/smc911x.c
@@ -73,6 +73,9 @@ static const char version[] =
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 
+#include <linux/dmaengine.h>
+#include <linux/dma/pxa-dma.h>
+
 #include <asm/io.h>
 
 #include "smc911x.h"
@@ -1174,18 +1177,16 @@ static irqreturn_t smc911x_interrupt(int irq, void *dev_id)
 
 #ifdef SMC_USE_DMA
 static void
-smc911x_tx_dma_irq(int dma, void *data)
+smc911x_tx_dma_irq(void *data)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct smc911x_local *lp = netdev_priv(dev);
+	struct smc911x_local *lp = data;
+	struct net_device *dev = lp->netdev;
 	struct sk_buff *skb = lp->current_tx_skb;
 	unsigned long flags;
 
 	DBG(SMC_DEBUG_FUNC, dev, "--> %s\n", __func__);
 
 	DBG(SMC_DEBUG_TX | SMC_DEBUG_DMA, dev, "TX DMA irq handler\n");
-	/* Clear the DMA interrupt sources */
-	SMC_DMA_ACK_IRQ(dev, dma);
 	BUG_ON(skb == NULL);
 	dma_unmap_single(NULL, tx_dmabuf, tx_dmalen, DMA_TO_DEVICE);
 	dev->trans_start = jiffies;
@@ -1208,18 +1209,16 @@ smc911x_tx_dma_irq(int dma, void *data)
 	    "TX DMA irq completed\n");
 }
 static void
-smc911x_rx_dma_irq(int dma, void *data)
+smc911x_rx_dma_irq(void *data)
 {
-	struct net_device *dev = (struct net_device *)data;
-	struct smc911x_local *lp = netdev_priv(dev);
+	struct smc911x_local *lp = data;
+	struct net_device *dev = lp->netdev;
 	struct sk_buff *skb = lp->current_rx_skb;
 	unsigned long flags;
 	unsigned int pkts;
 
 	DBG(SMC_DEBUG_FUNC, dev, "--> %s\n", __func__);
 	DBG(SMC_DEBUG_RX | SMC_DEBUG_DMA, dev, "RX DMA irq handler\n");
-	/* Clear the DMA interrupt sources */
-	SMC_DMA_ACK_IRQ(dev, dma);
 	dma_unmap_single(NULL, rx_dmabuf, rx_dmalen, DMA_FROM_DEVICE);
 	BUG_ON(skb == NULL);
 	lp->current_rx_skb = NULL;
@@ -1792,6 +1791,9 @@ static int smc911x_probe(struct net_device *dev)
 	unsigned int val, chip_id, revision;
 	const char *version_string;
 	unsigned long irq_flags;
+	struct dma_slave_config	config;
+	dma_cap_mask_t mask;
+	struct pxad_param param;
 
 	DBG(SMC_DEBUG_FUNC, dev, "--> %s\n", __func__);
 
@@ -1963,11 +1965,40 @@ static int smc911x_probe(struct net_device *dev)
 		goto err_out;
 
 #ifdef SMC_USE_DMA
-	lp->rxdma = SMC_DMA_REQUEST(dev, smc911x_rx_dma_irq);
-	lp->txdma = SMC_DMA_REQUEST(dev, smc911x_tx_dma_irq);
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+	param.prio = PXAD_PRIO_LOWEST;
+	param.drcmr = -1UL;
+
+	lp->rxdma =
+		dma_request_slave_channel_compat(mask, pxad_filter_fn,
+						 &param, &dev->dev, "rx");
+	lp->txdma =
+		dma_request_slave_channel_compat(mask, pxad_filter_fn,
+						 &param, &dev->dev, "tx");
 	lp->rxdma_active = 0;
 	lp->txdma_active = 0;
-	dev->dma = lp->rxdma;
+
+	memset(&config, 0, sizeof(config));
+	config.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	config.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	config.src_addr = lp->physaddr + RX_DATA_FIFO;
+	config.dst_addr = lp->physaddr + TX_DATA_FIFO;
+	config.src_maxburst = 32;
+	config.dst_maxburst = 32;
+	retval = dmaengine_slave_config(lp->rxdma, &config);
+	if (retval) {
+		dev_err(lp->dev, "dma rx channel configuration failed: %d\n",
+			retval);
+		goto err_out;
+	}
+	retval = dmaengine_slave_config(lp->txdma, &config);
+	if (retval) {
+		dev_err(lp->dev, "dma tx channel configuration failed: %d\n",
+			retval);
+		goto err_out;
+	}
 #endif
 
 	retval = register_netdev(dev);
@@ -1978,11 +2009,11 @@ static int smc911x_probe(struct net_device *dev)
 			    dev->base_addr, dev->irq);
 
 #ifdef SMC_USE_DMA
-		if (lp->rxdma != -1)
-			pr_cont(" RXDMA %d", lp->rxdma);
+		if (lp->rxdma)
+			pr_cont(" RXDMA %p", lp->rxdma);
 
-		if (lp->txdma != -1)
-			pr_cont(" TXDMA %d", lp->txdma);
+		if (lp->txdma)
+			pr_cont(" TXDMA %p", lp->txdma);
 #endif
 		pr_cont("\n");
 		if (!is_valid_ether_addr(dev->dev_addr)) {
@@ -2005,12 +2036,10 @@ static int smc911x_probe(struct net_device *dev)
 err_out:
 #ifdef SMC_USE_DMA
 	if (retval) {
-		if (lp->rxdma != -1) {
-			SMC_DMA_FREE(dev, lp->rxdma);
-		}
-		if (lp->txdma != -1) {
-			SMC_DMA_FREE(dev, lp->txdma);
-		}
+		if (lp->rxdma)
+			dma_release_channel(lp->rxdma);
+		if (lp->txdma)
+			dma_release_channel(lp->txdma);
 	}
 #endif
 	return retval;
@@ -2112,12 +2141,10 @@ static int smc911x_drv_remove(struct platform_device *pdev)
 
 #ifdef SMC_USE_DMA
 	{
-		if (lp->rxdma != -1) {
-			SMC_DMA_FREE(dev, lp->rxdma);
-		}
-		if (lp->txdma != -1) {
-			SMC_DMA_FREE(dev, lp->txdma);
-		}
+		if (lp->rxdma)
+			dma_release_channel(lp->rxdma);
+		if (lp->txdma)
+			dma_release_channel(lp->txdma);
 	}
 #endif
 	iounmap(lp->base);
diff --git a/drivers/net/ethernet/smsc/smc911x.h b/drivers/net/ethernet/smsc/smc911x.h
index 04b35f5..fa528ea 100644
--- a/drivers/net/ethernet/smsc/smc911x.h
+++ b/drivers/net/ethernet/smsc/smc911x.h
@@ -101,8 +101,8 @@ struct smc911x_local {
 #ifdef SMC_USE_DMA
 	/* DMA needs the physical address of the chip */
 	u_long physaddr;
-	int rxdma;
-	int txdma;
+	struct dma_chan *rxdma;
+	struct dma_chan *txdma;
 	int rxdma_active;
 	int txdma_active;
 	struct sk_buff *current_rx_skb;
@@ -210,27 +210,6 @@ static inline void SMC_outsl(struct smc911x_local *lp, int reg,
 
 #ifdef SMC_USE_PXA_DMA
 
-#include <mach/dma.h>
-
-/*
- * Define the request and free functions
- * These are unfortunately architecture specific as no generic allocation
- * mechanism exits
- */
-#define SMC_DMA_REQUEST(dev, handler) \
-	 pxa_request_dma(dev->name, DMA_PRIO_LOW, handler, dev)
-
-#define SMC_DMA_FREE(dev, dma) \
-	 pxa_free_dma(dma)
-
-#define SMC_DMA_ACK_IRQ(dev, dma)					\
-{									\
-	if (DCSR(dma) & DCSR_BUSERR) {					\
-		netdev_err(dev, "DMA %d bus error!\n", dma);		\
-	}								\
-	DCSR(dma) = DCSR_STARTINTR|DCSR_ENDINTR|DCSR_BUSERR;		\
-}
-
 /*
  * Use a DMA for RX and TX packets.
  */
@@ -238,6 +217,8 @@ static inline void SMC_outsl(struct smc911x_local *lp, int reg,
 
 static dma_addr_t rx_dmabuf, tx_dmabuf;
 static int rx_dmalen, tx_dmalen;
+static void smc911x_rx_dma_irq(void *data);
+static void smc911x_tx_dma_irq(void *data);
 
 #ifdef SMC_insl
 #undef SMC_insl
@@ -246,8 +227,10 @@ static int rx_dmalen, tx_dmalen;
 
 static inline void
 smc_pxa_dma_insl(struct smc911x_local *lp, u_long physaddr,
-		int reg, int dma, u_char *buf, int len)
+		int reg, struct dma_chan *dma, u_char *buf, int len)
 {
+	struct dma_async_tx_descriptor *tx;
+
 	/* 64 bit alignment is required for memory to memory DMA */
 	if ((long)buf & 4) {
 		*((u32 *)buf) = SMC_inl(lp, reg);
@@ -258,12 +241,14 @@ smc_pxa_dma_insl(struct smc911x_local *lp, u_long physaddr,
 	len *= 4;
 	rx_dmabuf = dma_map_single(lp->dev, buf, len, DMA_FROM_DEVICE);
 	rx_dmalen = len;
-	DCSR(dma) = DCSR_NODESC;
-	DTADR(dma) = rx_dmabuf;
-	DSADR(dma) = physaddr + reg;
-	DCMD(dma) = (DCMD_INCTRGADDR | DCMD_BURST32 |
-		DCMD_WIDTH4 | DCMD_ENDIRQEN | (DCMD_LENGTH & rx_dmalen));
-	DCSR(dma) = DCSR_NODESC | DCSR_RUN;
+	tx = dmaengine_prep_slave_single(dma, rx_dmabuf, rx_dmalen,
+					 DMA_DEV_TO_MEM, 0);
+	if (tx) {
+		tx->callback = smc911x_rx_dma_irq;
+		tx->callback_param = lp;
+		dmaengine_submit(tx);
+		dma_async_issue_pending(dma);
+	}
 }
 #endif
 
@@ -274,8 +259,10 @@ smc_pxa_dma_insl(struct smc911x_local *lp, u_long physaddr,
 
 static inline void
 smc_pxa_dma_outsl(struct smc911x_local *lp, u_long physaddr,
-		int reg, int dma, u_char *buf, int len)
+		int reg, struct dma_chan *dma, u_char *buf, int len)
 {
+	struct dma_async_tx_descriptor *tx;
+
 	/* 64 bit alignment is required for memory to memory DMA */
 	if ((long)buf & 4) {
 		SMC_outl(*((u32 *)buf), lp, reg);
@@ -286,12 +273,14 @@ smc_pxa_dma_outsl(struct smc911x_local *lp, u_long physaddr,
 	len *= 4;
 	tx_dmabuf = dma_map_single(lp->dev, buf, len, DMA_TO_DEVICE);
 	tx_dmalen = len;
-	DCSR(dma) = DCSR_NODESC;
-	DSADR(dma) = tx_dmabuf;
-	DTADR(dma) = physaddr + reg;
-	DCMD(dma) = (DCMD_INCSRCADDR | DCMD_BURST32 |
-		DCMD_WIDTH4 | DCMD_ENDIRQEN | (DCMD_LENGTH & tx_dmalen));
-	DCSR(dma) = DCSR_NODESC | DCSR_RUN;
+	tx = dmaengine_prep_slave_single(dma, tx_dmabuf, tx_dmalen,
+					 DMA_DEV_TO_MEM, 0);
+	if (tx) {
+		tx->callback = smc911x_tx_dma_irq;
+		tx->callback_param = lp;
+		dmaengine_submit(tx);
+		dma_async_issue_pending(dma);
+	}
 }
 #endif
 #endif	 /* SMC_USE_PXA_DMA */
-- 
cgit v0.10.2


From c6a58ffed53612be86b758df1cdb0b0f4305e9cb Mon Sep 17 00:00:00 2001
From: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Date: Wed, 16 Mar 2016 11:38:12 -0700
Subject: RDS: TCP: Add sysctl tunables for sndbuf/rcvbuf on rds-tcp socket

Add per-net sysctl tunables to set the size of sndbuf and
rcvbuf on the kernel tcp socket.

The tunables are added at /proc/sys/net/rds/tcp/rds_tcp_sndbuf
and /proc/sys/net/rds/tcp/rds_tcp_rcvbuf.

These values must be set before accept() or connect(),
and there may be an arbitrary number of existing rds-tcp
sockets when the tunable is modified. To make sure that all
connections in the netns pick up the same value for the tunable,
we reset existing rds-tcp connections in the netns, so that
they can reconnect with the new parameters.

Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index ad60299..3802785 100644
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -54,6 +54,35 @@ static struct kmem_cache *rds_tcp_conn_slab;
 
 #define RDS_TCP_DEFAULT_BUFSIZE (128 * 1024)
 
+static int rds_tcp_skbuf_handler(struct ctl_table *ctl, int write,
+				 void __user *buffer, size_t *lenp,
+				 loff_t *fpos);
+
+int rds_tcp_min_sndbuf = SOCK_MIN_SNDBUF;
+int rds_tcp_min_rcvbuf = SOCK_MIN_RCVBUF;
+
+static struct ctl_table rds_tcp_sysctl_table[] = {
+#define	RDS_TCP_SNDBUF	0
+	{
+		.procname       = "rds_tcp_sndbuf",
+		/* data is per-net pointer */
+		.maxlen         = sizeof(int),
+		.mode           = 0644,
+		.proc_handler   = rds_tcp_skbuf_handler,
+		.extra1		= &rds_tcp_min_sndbuf,
+	},
+#define	RDS_TCP_RCVBUF	1
+	{
+		.procname       = "rds_tcp_rcvbuf",
+		/* data is per-net pointer */
+		.maxlen         = sizeof(int),
+		.mode           = 0644,
+		.proc_handler   = rds_tcp_skbuf_handler,
+		.extra1		= &rds_tcp_min_rcvbuf,
+	},
+	{ }
+};
+
 /* doing it this way avoids calling tcp_sk() */
 void rds_tcp_nonagle(struct socket *sock)
 {
@@ -66,15 +95,6 @@ void rds_tcp_nonagle(struct socket *sock)
 	set_fs(oldfs);
 }
 
-/* All module specific customizations to the RDS-TCP socket should be done in
- * rds_tcp_tune() and applied after socket creation. In general these
- * customizations should be tunable via module_param()
- */
-void rds_tcp_tune(struct socket *sock)
-{
-	rds_tcp_nonagle(sock);
-}
-
 u32 rds_tcp_snd_nxt(struct rds_tcp_connection *tc)
 {
 	return tcp_sk(tc->t_sock->sk)->snd_nxt;
@@ -272,8 +292,34 @@ static int rds_tcp_netid;
 struct rds_tcp_net {
 	struct socket *rds_tcp_listen_sock;
 	struct work_struct rds_tcp_accept_w;
+	struct ctl_table_header *rds_tcp_sysctl;
+	struct ctl_table *ctl_table;
+	int sndbuf_size;
+	int rcvbuf_size;
 };
 
+/* All module specific customizations to the RDS-TCP socket should be done in
+ * rds_tcp_tune() and applied after socket creation.
+ */
+void rds_tcp_tune(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	struct net *net = sock_net(sk);
+	struct rds_tcp_net *rtn = net_generic(net, rds_tcp_netid);
+
+	rds_tcp_nonagle(sock);
+	lock_sock(sk);
+	if (rtn->sndbuf_size > 0) {
+		sk->sk_sndbuf = rtn->sndbuf_size;
+		sk->sk_userlocks |= SOCK_SNDBUF_LOCK;
+	}
+	if (rtn->rcvbuf_size > 0) {
+		sk->sk_sndbuf = rtn->rcvbuf_size;
+		sk->sk_userlocks |= SOCK_RCVBUF_LOCK;
+	}
+	release_sock(sk);
+}
+
 static void rds_tcp_accept_worker(struct work_struct *work)
 {
 	struct rds_tcp_net *rtn = container_of(work,
@@ -295,20 +341,60 @@ void rds_tcp_accept_work(struct sock *sk)
 static __net_init int rds_tcp_init_net(struct net *net)
 {
 	struct rds_tcp_net *rtn = net_generic(net, rds_tcp_netid);
+	struct ctl_table *tbl;
+	int err = 0;
 
+	memset(rtn, 0, sizeof(*rtn));
+
+	/* {snd, rcv}buf_size default to 0, which implies we let the
+	 * stack pick the value, and permit auto-tuning of buffer size.
+	 */
+	if (net == &init_net) {
+		tbl = rds_tcp_sysctl_table;
+	} else {
+		tbl = kmemdup(rds_tcp_sysctl_table,
+			      sizeof(rds_tcp_sysctl_table), GFP_KERNEL);
+		if (!tbl) {
+			pr_warn("could not set allocate syctl table\n");
+			return -ENOMEM;
+		}
+		rtn->ctl_table = tbl;
+	}
+	tbl[RDS_TCP_SNDBUF].data = &rtn->sndbuf_size;
+	tbl[RDS_TCP_RCVBUF].data = &rtn->rcvbuf_size;
+	rtn->rds_tcp_sysctl = register_net_sysctl(net, "net/rds/tcp", tbl);
+	if (!rtn->rds_tcp_sysctl) {
+		pr_warn("could not register sysctl\n");
+		err = -ENOMEM;
+		goto fail;
+	}
 	rtn->rds_tcp_listen_sock = rds_tcp_listen_init(net);
 	if (!rtn->rds_tcp_listen_sock) {
 		pr_warn("could not set up listen sock\n");
-		return -EAFNOSUPPORT;
+		unregister_net_sysctl_table(rtn->rds_tcp_sysctl);
+		rtn->rds_tcp_sysctl = NULL;
+		err = -EAFNOSUPPORT;
+		goto fail;
 	}
 	INIT_WORK(&rtn->rds_tcp_accept_w, rds_tcp_accept_worker);
 	return 0;
+
+fail:
+	if (net != &init_net)
+		kfree(tbl);
+	return err;
 }
 
 static void __net_exit rds_tcp_exit_net(struct net *net)
 {
 	struct rds_tcp_net *rtn = net_generic(net, rds_tcp_netid);
 
+	if (rtn->rds_tcp_sysctl)
+		unregister_net_sysctl_table(rtn->rds_tcp_sysctl);
+
+	if (net != &init_net && rtn->ctl_table)
+		kfree(rtn->ctl_table);
+
 	/* If rds_tcp_exit_net() is called as a result of netns deletion,
 	 * the rds_tcp_kill_sock() device notifier would already have cleaned
 	 * up the listen socket, thus there is no work to do in this function.
@@ -383,6 +469,45 @@ static struct notifier_block rds_tcp_dev_notifier = {
 	.priority = -10, /* must be called after other network notifiers */
 };
 
+/* when sysctl is used to modify some kernel socket parameters,this
+ * function  resets the RDS connections in that netns  so that we can
+ * restart with new parameters.  The assumption is that such reset
+ * events are few and far-between.
+ */
+static void rds_tcp_sysctl_reset(struct net *net)
+{
+	struct rds_tcp_connection *tc, *_tc;
+
+	spin_lock_irq(&rds_tcp_conn_lock);
+	list_for_each_entry_safe(tc, _tc, &rds_tcp_conn_list, t_tcp_node) {
+		struct net *c_net = read_pnet(&tc->conn->c_net);
+
+		if (net != c_net || !tc->t_sock)
+			continue;
+
+		rds_conn_drop(tc->conn); /* reconnect with new parameters */
+	}
+	spin_unlock_irq(&rds_tcp_conn_lock);
+}
+
+static int rds_tcp_skbuf_handler(struct ctl_table *ctl, int write,
+				 void __user *buffer, size_t *lenp,
+				 loff_t *fpos)
+{
+	struct net *net = current->nsproxy->net_ns;
+	int err;
+
+	err = proc_dointvec_minmax(ctl, write, buffer, lenp, fpos);
+	if (err < 0) {
+		pr_warn("Invalid input. Must be >= %d\n",
+			*(int *)(ctl->extra1));
+		return err;
+	}
+	if (write)
+		rds_tcp_sysctl_reset(net);
+	return 0;
+}
+
 static void rds_tcp_exit(void)
 {
 	rds_info_deregister_func(RDS_INFO_TCP_SOCKETS, rds_tcp_tc_info);
-- 
cgit v0.10.2


From a3382e408b645b4f68ec01f9c048e356c62598fb Mon Sep 17 00:00:00 2001
From: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Date: Wed, 16 Mar 2016 11:38:13 -0700
Subject: RDS: TCP: Remove unused constant

RDS_TCP_DEFAULT_BUFSIZE has been unused since commit 1edd6a14d24f
("RDS-TCP: Do not bloat sndbuf/rcvbuf in rds_tcp_tune").

Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index 3802785..61ed2a8 100644
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -52,8 +52,6 @@ static LIST_HEAD(rds_tcp_conn_list);
 
 static struct kmem_cache *rds_tcp_conn_slab;
 
-#define RDS_TCP_DEFAULT_BUFSIZE (128 * 1024)
-
 static int rds_tcp_skbuf_handler(struct ctl_table *ctl, int write,
 				 void __user *buffer, size_t *lenp,
 				 loff_t *fpos);
-- 
cgit v0.10.2


From 20ff55655a93554611fb7790c8a2d29ee4598d24 Mon Sep 17 00:00:00 2001
From: Woojung Huh <woojung.huh@microchip.com>
Date: Wed, 16 Mar 2016 22:10:40 +0000
Subject: lan78xx: handle statistics counter rollover

Update to handle statistics counter rollover.
Check statistics counter periodically and compensate it when
counter value rolls over at max (20 or 32bits).

Simple mechanism adjusts monitoring timer to allow USB auto suspend.

Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 705c180..f20890e 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -36,7 +36,7 @@
 #define DRIVER_AUTHOR	"WOOJUNG HUH <woojung.huh@microchip.com>"
 #define DRIVER_DESC	"LAN78XX USB 3.0 Gigabit Ethernet Devices"
 #define DRIVER_NAME	"lan78xx"
-#define DRIVER_VERSION	"1.0.3"
+#define DRIVER_VERSION	"1.0.4"
 
 #define TX_TIMEOUT_JIFFIES		(5 * HZ)
 #define THROTTLE_JIFFIES		(HZ / 8)
@@ -86,6 +86,9 @@
 /* default autosuspend delay (mSec)*/
 #define DEFAULT_AUTOSUSPEND_DELAY	(10 * 1000)
 
+/* statistic update interval (mSec) */
+#define STAT_UPDATE_TIMER		(1 * 1000)
+
 static const char lan78xx_gstrings[][ETH_GSTRING_LEN] = {
 	"RX FCS Errors",
 	"RX Alignment Errors",
@@ -186,6 +189,56 @@ struct lan78xx_statstage {
 	u32 eee_tx_lpi_time;
 };
 
+struct lan78xx_statstage64 {
+	u64 rx_fcs_errors;
+	u64 rx_alignment_errors;
+	u64 rx_fragment_errors;
+	u64 rx_jabber_errors;
+	u64 rx_undersize_frame_errors;
+	u64 rx_oversize_frame_errors;
+	u64 rx_dropped_frames;
+	u64 rx_unicast_byte_count;
+	u64 rx_broadcast_byte_count;
+	u64 rx_multicast_byte_count;
+	u64 rx_unicast_frames;
+	u64 rx_broadcast_frames;
+	u64 rx_multicast_frames;
+	u64 rx_pause_frames;
+	u64 rx_64_byte_frames;
+	u64 rx_65_127_byte_frames;
+	u64 rx_128_255_byte_frames;
+	u64 rx_256_511_bytes_frames;
+	u64 rx_512_1023_byte_frames;
+	u64 rx_1024_1518_byte_frames;
+	u64 rx_greater_1518_byte_frames;
+	u64 eee_rx_lpi_transitions;
+	u64 eee_rx_lpi_time;
+	u64 tx_fcs_errors;
+	u64 tx_excess_deferral_errors;
+	u64 tx_carrier_errors;
+	u64 tx_bad_byte_count;
+	u64 tx_single_collisions;
+	u64 tx_multiple_collisions;
+	u64 tx_excessive_collision;
+	u64 tx_late_collisions;
+	u64 tx_unicast_byte_count;
+	u64 tx_broadcast_byte_count;
+	u64 tx_multicast_byte_count;
+	u64 tx_unicast_frames;
+	u64 tx_broadcast_frames;
+	u64 tx_multicast_frames;
+	u64 tx_pause_frames;
+	u64 tx_64_byte_frames;
+	u64 tx_65_127_byte_frames;
+	u64 tx_128_255_byte_frames;
+	u64 tx_256_511_bytes_frames;
+	u64 tx_512_1023_byte_frames;
+	u64 tx_1024_1518_byte_frames;
+	u64 tx_greater_1518_byte_frames;
+	u64 eee_tx_lpi_transitions;
+	u64 eee_tx_lpi_time;
+};
+
 struct lan78xx_net;
 
 struct lan78xx_priv {
@@ -232,6 +285,15 @@ struct usb_context {
 #define EVENT_DEV_WAKING		6
 #define EVENT_DEV_ASLEEP		7
 #define EVENT_DEV_OPEN			8
+#define EVENT_STAT_UPDATE		9
+
+struct statstage {
+	struct mutex			access_lock;	/* for stats access */
+	struct lan78xx_statstage	saved;
+	struct lan78xx_statstage	rollover_count;
+	struct lan78xx_statstage	rollover_max;
+	struct lan78xx_statstage64	curr_stat;
+};
 
 struct lan78xx_net {
 	struct net_device	*net;
@@ -272,6 +334,7 @@ struct lan78xx_net {
 
 	unsigned		maxpacket;
 	struct timer_list	delay;
+	struct timer_list	stat_monitor;
 
 	unsigned long		data[5];
 
@@ -284,6 +347,9 @@ struct lan78xx_net {
 
 	int			fc_autoneg;
 	u8			fc_request_control;
+
+	int			delta;
+	struct statstage	stats;
 };
 
 /* use ethtool to change the level for any given device */
@@ -382,6 +448,93 @@ static int lan78xx_read_stats(struct lan78xx_net *dev,
 	return ret;
 }
 
+#define check_counter_rollover(struct1, dev_stats, member) {	\
+	if (struct1->member < dev_stats.saved.member)		\
+		dev_stats.rollover_count.member++;		\
+	}
+
+static void lan78xx_check_stat_rollover(struct lan78xx_net *dev,
+					struct lan78xx_statstage *stats)
+{
+	check_counter_rollover(stats, dev->stats, rx_fcs_errors);
+	check_counter_rollover(stats, dev->stats, rx_alignment_errors);
+	check_counter_rollover(stats, dev->stats, rx_fragment_errors);
+	check_counter_rollover(stats, dev->stats, rx_jabber_errors);
+	check_counter_rollover(stats, dev->stats, rx_undersize_frame_errors);
+	check_counter_rollover(stats, dev->stats, rx_oversize_frame_errors);
+	check_counter_rollover(stats, dev->stats, rx_dropped_frames);
+	check_counter_rollover(stats, dev->stats, rx_unicast_byte_count);
+	check_counter_rollover(stats, dev->stats, rx_broadcast_byte_count);
+	check_counter_rollover(stats, dev->stats, rx_multicast_byte_count);
+	check_counter_rollover(stats, dev->stats, rx_unicast_frames);
+	check_counter_rollover(stats, dev->stats, rx_broadcast_frames);
+	check_counter_rollover(stats, dev->stats, rx_multicast_frames);
+	check_counter_rollover(stats, dev->stats, rx_pause_frames);
+	check_counter_rollover(stats, dev->stats, rx_64_byte_frames);
+	check_counter_rollover(stats, dev->stats, rx_65_127_byte_frames);
+	check_counter_rollover(stats, dev->stats, rx_128_255_byte_frames);
+	check_counter_rollover(stats, dev->stats, rx_256_511_bytes_frames);
+	check_counter_rollover(stats, dev->stats, rx_512_1023_byte_frames);
+	check_counter_rollover(stats, dev->stats, rx_1024_1518_byte_frames);
+	check_counter_rollover(stats, dev->stats, rx_greater_1518_byte_frames);
+	check_counter_rollover(stats, dev->stats, eee_rx_lpi_transitions);
+	check_counter_rollover(stats, dev->stats, eee_rx_lpi_time);
+	check_counter_rollover(stats, dev->stats, tx_fcs_errors);
+	check_counter_rollover(stats, dev->stats, tx_excess_deferral_errors);
+	check_counter_rollover(stats, dev->stats, tx_carrier_errors);
+	check_counter_rollover(stats, dev->stats, tx_bad_byte_count);
+	check_counter_rollover(stats, dev->stats, tx_single_collisions);
+	check_counter_rollover(stats, dev->stats, tx_multiple_collisions);
+	check_counter_rollover(stats, dev->stats, tx_excessive_collision);
+	check_counter_rollover(stats, dev->stats, tx_late_collisions);
+	check_counter_rollover(stats, dev->stats, tx_unicast_byte_count);
+	check_counter_rollover(stats, dev->stats, tx_broadcast_byte_count);
+	check_counter_rollover(stats, dev->stats, tx_multicast_byte_count);
+	check_counter_rollover(stats, dev->stats, tx_unicast_frames);
+	check_counter_rollover(stats, dev->stats, tx_broadcast_frames);
+	check_counter_rollover(stats, dev->stats, tx_multicast_frames);
+	check_counter_rollover(stats, dev->stats, tx_pause_frames);
+	check_counter_rollover(stats, dev->stats, tx_64_byte_frames);
+	check_counter_rollover(stats, dev->stats, tx_65_127_byte_frames);
+	check_counter_rollover(stats, dev->stats, tx_128_255_byte_frames);
+	check_counter_rollover(stats, dev->stats, tx_256_511_bytes_frames);
+	check_counter_rollover(stats, dev->stats, tx_512_1023_byte_frames);
+	check_counter_rollover(stats, dev->stats, tx_1024_1518_byte_frames);
+	check_counter_rollover(stats, dev->stats, tx_greater_1518_byte_frames);
+	check_counter_rollover(stats, dev->stats, eee_tx_lpi_transitions);
+	check_counter_rollover(stats, dev->stats, eee_tx_lpi_time);
+
+	memcpy(&dev->stats.saved, stats, sizeof(struct lan78xx_statstage));
+}
+
+static void lan78xx_update_stats(struct lan78xx_net *dev)
+{
+	u32 *p, *count, *max;
+	u64 *data;
+	int i;
+	struct lan78xx_statstage lan78xx_stats;
+
+	if (usb_autopm_get_interface(dev->intf) < 0)
+		return;
+
+	p = (u32 *)&lan78xx_stats;
+	count = (u32 *)&dev->stats.rollover_count;
+	max = (u32 *)&dev->stats.rollover_max;
+	data = (u64 *)&dev->stats.curr_stat;
+
+	mutex_lock(&dev->stats.access_lock);
+
+	if (lan78xx_read_stats(dev, &lan78xx_stats) > 0)
+		lan78xx_check_stat_rollover(dev, &lan78xx_stats);
+
+	for (i = 0; i < (sizeof(lan78xx_stats) / (sizeof(u32))); i++)
+		data[i] = (u64)p[i] + ((u64)count[i] * ((u64)max[i] + 1));
+
+	mutex_unlock(&dev->stats.access_lock);
+
+	usb_autopm_put_interface(dev->intf);
+}
+
 /* Loop until the read is completed with timeout called with phy_mutex held */
 static int lan78xx_phy_wait_not_busy(struct lan78xx_net *dev)
 {
@@ -967,6 +1120,8 @@ static int lan78xx_link_reset(struct lan78xx_net *dev)
 			return -EIO;
 
 		phy_mac_interrupt(phydev, 0);
+
+		del_timer(&dev->stat_monitor);
 	} else if (phydev->link && !dev->link_on) {
 		dev->link_on = true;
 
@@ -1007,6 +1162,12 @@ static int lan78xx_link_reset(struct lan78xx_net *dev)
 
 		ret = lan78xx_update_flowcontrol(dev, ecmd.duplex, ladv, radv);
 		phy_mac_interrupt(phydev, 1);
+
+		if (!timer_pending(&dev->stat_monitor)) {
+			dev->delta = 1;
+			mod_timer(&dev->stat_monitor,
+				  jiffies + STAT_UPDATE_TIMER);
+		}
 	}
 
 	return ret;
@@ -1099,20 +1260,12 @@ static void lan78xx_get_stats(struct net_device *netdev,
 			      struct ethtool_stats *stats, u64 *data)
 {
 	struct lan78xx_net *dev = netdev_priv(netdev);
-	struct lan78xx_statstage lan78xx_stat;
-	u32 *p;
-	int i;
 
-	if (usb_autopm_get_interface(dev->intf) < 0)
-		return;
+	lan78xx_update_stats(dev);
 
-	if (lan78xx_read_stats(dev, &lan78xx_stat) > 0) {
-		p = (u32 *)&lan78xx_stat;
-		for (i = 0; i < (sizeof(lan78xx_stat) / (sizeof(u32))); i++)
-			data[i] = p[i];
-	}
-
-	usb_autopm_put_interface(dev->intf);
+	mutex_lock(&dev->stats.access_lock);
+	memcpy(data, &dev->stats.curr_stat, sizeof(dev->stats.curr_stat));
+	mutex_unlock(&dev->stats.access_lock);
 }
 
 static void lan78xx_get_wol(struct net_device *netdev,
@@ -2095,6 +2248,32 @@ static int lan78xx_reset(struct lan78xx_net *dev)
 	return 0;
 }
 
+static void lan78xx_init_stats(struct lan78xx_net *dev)
+{
+	u32 *p;
+	int i;
+
+	/* initialize for stats update
+	 * some counters are 20bits and some are 32bits
+	 */
+	p = (u32 *)&dev->stats.rollover_max;
+	for (i = 0; i < (sizeof(dev->stats.rollover_max) / (sizeof(u32))); i++)
+		p[i] = 0xFFFFF;
+
+	dev->stats.rollover_max.rx_unicast_byte_count = 0xFFFFFFFF;
+	dev->stats.rollover_max.rx_broadcast_byte_count = 0xFFFFFFFF;
+	dev->stats.rollover_max.rx_multicast_byte_count = 0xFFFFFFFF;
+	dev->stats.rollover_max.eee_rx_lpi_transitions = 0xFFFFFFFF;
+	dev->stats.rollover_max.eee_rx_lpi_time = 0xFFFFFFFF;
+	dev->stats.rollover_max.tx_unicast_byte_count = 0xFFFFFFFF;
+	dev->stats.rollover_max.tx_broadcast_byte_count = 0xFFFFFFFF;
+	dev->stats.rollover_max.tx_multicast_byte_count = 0xFFFFFFFF;
+	dev->stats.rollover_max.eee_tx_lpi_transitions = 0xFFFFFFFF;
+	dev->stats.rollover_max.eee_tx_lpi_time = 0xFFFFFFFF;
+
+	lan78xx_defer_kevent(dev, EVENT_STAT_UPDATE);
+}
+
 static int lan78xx_open(struct net_device *net)
 {
 	struct lan78xx_net *dev = netdev_priv(net);
@@ -2122,6 +2301,8 @@ static int lan78xx_open(struct net_device *net)
 		}
 	}
 
+	lan78xx_init_stats(dev);
+
 	set_bit(EVENT_DEV_OPEN, &dev->flags);
 
 	netif_start_queue(net);
@@ -2166,6 +2347,9 @@ int lan78xx_stop(struct net_device *net)
 {
 	struct lan78xx_net		*dev = netdev_priv(net);
 
+	if (timer_pending(&dev->stat_monitor))
+		del_timer_sync(&dev->stat_monitor);
+
 	phy_stop(net->phydev);
 	phy_disconnect(net->phydev);
 	net->phydev = NULL;
@@ -2910,6 +3094,13 @@ static void lan78xx_bh(unsigned long param)
 	}
 
 	if (netif_device_present(dev->net) && netif_running(dev->net)) {
+		/* reset update timer delta */
+		if (timer_pending(&dev->stat_monitor) && (dev->delta != 1)) {
+			dev->delta = 1;
+			mod_timer(&dev->stat_monitor,
+				  jiffies + STAT_UPDATE_TIMER);
+		}
+
 		if (!skb_queue_empty(&dev->txq_pend))
 			lan78xx_tx_bh(dev);
 
@@ -2984,6 +3175,17 @@ skip_reset:
 			usb_autopm_put_interface(dev->intf);
 		}
 	}
+
+	if (test_bit(EVENT_STAT_UPDATE, &dev->flags)) {
+		lan78xx_update_stats(dev);
+
+		clear_bit(EVENT_STAT_UPDATE, &dev->flags);
+
+		mod_timer(&dev->stat_monitor,
+			  jiffies + (STAT_UPDATE_TIMER * dev->delta));
+
+		dev->delta = min((dev->delta * 2), 50);
+	}
 }
 
 static void intr_complete(struct urb *urb)
@@ -3074,6 +3276,15 @@ static const struct net_device_ops lan78xx_netdev_ops = {
 	.ndo_vlan_rx_kill_vid	= lan78xx_vlan_rx_kill_vid,
 };
 
+static void lan78xx_stat_monitor(unsigned long param)
+{
+	struct lan78xx_net *dev;
+
+	dev = (struct lan78xx_net *)param;
+
+	lan78xx_defer_kevent(dev, EVENT_STAT_UPDATE);
+}
+
 static int lan78xx_probe(struct usb_interface *intf,
 			 const struct usb_device_id *id)
 {
@@ -3120,6 +3331,13 @@ static int lan78xx_probe(struct usb_interface *intf,
 	netdev->watchdog_timeo = TX_TIMEOUT_JIFFIES;
 	netdev->ethtool_ops = &lan78xx_ethtool_ops;
 
+	dev->stat_monitor.function = lan78xx_stat_monitor;
+	dev->stat_monitor.data = (unsigned long)dev;
+	dev->delta = 1;
+	init_timer(&dev->stat_monitor);
+
+	mutex_init(&dev->stats.access_lock);
+
 	ret = lan78xx_bind(dev, intf);
 	if (ret < 0)
 		goto out2;
@@ -3397,6 +3615,8 @@ int lan78xx_suspend(struct usb_interface *intf, pm_message_t message)
 	}
 
 	if (test_bit(EVENT_DEV_ASLEEP, &dev->flags)) {
+		del_timer(&dev->stat_monitor);
+
 		if (PMSG_IS_AUTO(message)) {
 			/* auto suspend (selective suspend) */
 			ret = lan78xx_read_reg(dev, MAC_TX, &buf);
@@ -3457,6 +3677,12 @@ int lan78xx_resume(struct usb_interface *intf)
 	int ret;
 	u32 buf;
 
+	if (!timer_pending(&dev->stat_monitor)) {
+		dev->delta = 1;
+		mod_timer(&dev->stat_monitor,
+			  jiffies + STAT_UPDATE_TIMER);
+	}
+
 	if (!--dev->suspend_count) {
 		/* resume interrupt URBs */
 		if (dev->urb_intr && test_bit(EVENT_DEV_OPEN, &dev->flags))
-- 
cgit v0.10.2


From a59f8c5b048dc938fb958c91c282c865cd845705 Mon Sep 17 00:00:00 2001
From: Woojung Huh <woojung.huh@microchip.com>
Date: Wed, 16 Mar 2016 22:10:42 +0000
Subject: lan78xx: add ndo_get_stats64

Add lan78xx_get_stats64 of ndo_get_stats64 to report
all statistics counters including errors from HW statistics.

Read from HW when auto suspend is disabled, use saved counter when
auto suspend is enabled because periodic call to ndo_get_stats64
prevents USB auto suspend.

Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index f20890e..d36d5eb 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -3261,6 +3261,54 @@ void lan78xx_tx_timeout(struct net_device *net)
 	tasklet_schedule(&dev->bh);
 }
 
+struct rtnl_link_stats64 *lan78xx_get_stats64(struct net_device *netdev,
+					      struct rtnl_link_stats64 *storage)
+{
+	struct lan78xx_net *dev = netdev_priv(netdev);
+	struct lan78xx_statstage64 stats;
+
+	/* curr_stat is updated by timer.
+	 * periodic reading from HW will prevent from entering USB auto suspend.
+	 * if autosuspend is disabled, read from HW.
+	 */
+	if (!dev->udev->dev.power.runtime_auto)
+		lan78xx_update_stats(dev);
+
+	mutex_lock(&dev->stats.access_lock);
+	memcpy(&stats, &dev->stats.curr_stat, sizeof(stats));
+	mutex_unlock(&dev->stats.access_lock);
+
+	/* calc by driver */
+	storage->rx_packets = (__u64)netdev->stats.rx_packets;
+	storage->tx_packets = (__u64)netdev->stats.tx_packets;
+	storage->rx_bytes = (__u64)netdev->stats.rx_bytes;
+	storage->tx_bytes = (__u64)netdev->stats.tx_bytes;
+
+	/* use counter */
+	storage->rx_length_errors = stats.rx_undersize_frame_errors +
+				    stats.rx_oversize_frame_errors;
+	storage->rx_crc_errors = stats.rx_fcs_errors;
+	storage->rx_frame_errors = stats.rx_alignment_errors;
+	storage->rx_fifo_errors = stats.rx_dropped_frames;
+	storage->rx_over_errors = stats.rx_oversize_frame_errors;
+	storage->rx_errors = stats.rx_fcs_errors +
+			     stats.rx_alignment_errors +
+			     stats.rx_fragment_errors +
+			     stats.rx_jabber_errors +
+			     stats.rx_undersize_frame_errors +
+			     stats.rx_oversize_frame_errors +
+			     stats.rx_dropped_frames;
+
+	storage->tx_carrier_errors = stats.tx_carrier_errors;
+	storage->tx_errors = stats.tx_fcs_errors +
+			     stats.tx_excess_deferral_errors +
+			     stats.tx_carrier_errors;
+
+	storage->multicast = stats.rx_multicast_frames;
+
+	return storage;
+}
+
 static const struct net_device_ops lan78xx_netdev_ops = {
 	.ndo_open		= lan78xx_open,
 	.ndo_stop		= lan78xx_stop,
@@ -3274,6 +3322,7 @@ static const struct net_device_ops lan78xx_netdev_ops = {
 	.ndo_set_features	= lan78xx_set_features,
 	.ndo_vlan_rx_add_vid	= lan78xx_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid	= lan78xx_vlan_rx_kill_vid,
+	.ndo_get_stats64	= lan78xx_get_stats64,
 };
 
 static void lan78xx_stat_monitor(unsigned long param)
-- 
cgit v0.10.2


From e2ad1f976b721df383ff12c12a6dcc805cbb80f3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Andreas=20F=C3=A4rber?= <afaerber@suse.de>
Date: Thu, 17 Mar 2016 00:23:37 +0100
Subject: phy: mdio-thunder: Fix some Kconfig typos
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Drop two extra occurrences of "on" in option title and help text.

Fixes: 379d7ac7ca31 ("phy: mdio-thunder: Add driver for Cavium Thunder SoC MDIO buses.")
Cc: David Daney <david.daney@cavium.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 075a4cc..6dad9a9 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -197,13 +197,13 @@ config MDIO_OCTEON
 	  drivers on some systems.
 
 config MDIO_THUNDER
-	tristate "Support for MDIO buses on on ThunderX SOCs"
+	tristate "Support for MDIO buses on ThunderX SOCs"
 	depends on 64BIT
 	depends on PCI
 	select MDIO_CAVIUM
 	help
 	  This driver supports the MDIO interfaces found on Cavium
-	  ThunderX SoCs when the MDIO bus device appears on as a PCI
+	  ThunderX SoCs when the MDIO bus device appears as a PCI
 	  device.
 
 
-- 
cgit v0.10.2


From 76e39ccf9c36352a8fc7c084ec25b735010685fd Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@mellanox.com>
Date: Thu, 17 Mar 2016 18:49:42 +0200
Subject: net/mlx4_core: Fix backward compatibility on VFs

Commit 85743f1eb345 ("net/mlx4_core: Set UAR page size to 4KB regardless
of system page size") introduced dependency where old VF drivers without
this fix fail to load if the PF driver runs with this commit.

To resolve this add a module parameter which disables that functionality
by default.  If both the PF and VFs are running with a driver with that
commit the administrator may set the module param to true.

The module parameter is called enable_4k_uar.

Fixes: 85743f1eb345 ('net/mlx4_core: Set UAR page size to 4KB ...')
Signed-off-by: Eli Cohen <eli@mellanox.com>
Tested-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 503ec23..358f723 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -105,6 +105,11 @@ module_param(enable_64b_cqe_eqe, bool, 0444);
 MODULE_PARM_DESC(enable_64b_cqe_eqe,
 		 "Enable 64 byte CQEs/EQEs when the FW supports this (default: True)");
 
+static bool enable_4k_uar;
+module_param(enable_4k_uar, bool, 0444);
+MODULE_PARM_DESC(enable_4k_uar,
+		 "Enable using 4K UAR. Should not be enabled if have VFs which do not support 4K UARs (default: false)");
+
 #define PF_CONTEXT_BEHAVIOUR_MASK	(MLX4_FUNC_CAP_64B_EQE_CQE | \
 					 MLX4_FUNC_CAP_EQE_CQE_STRIDE | \
 					 MLX4_FUNC_CAP_DMFS_A0_STATIC)
@@ -423,7 +428,11 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		/* Virtual PCI function needs to determine UAR page size from
 		 * firmware. Only master PCI function can set the uar page size
 		 */
-		dev->uar_page_shift = DEFAULT_UAR_PAGE_SHIFT;
+		if (enable_4k_uar)
+			dev->uar_page_shift = DEFAULT_UAR_PAGE_SHIFT;
+		else
+			dev->uar_page_shift = PAGE_SHIFT;
+
 		mlx4_set_num_reserved_uars(dev, dev_cap);
 	}
 
@@ -2233,11 +2242,14 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		dev->caps.max_fmr_maps = (1 << (32 - ilog2(dev->caps.num_mpts))) - 1;
 
-		/* Always set UAR page size 4KB, set log_uar_sz accordingly */
-		init_hca.log_uar_sz = ilog2(dev->caps.num_uars) +
-				      PAGE_SHIFT -
-				      DEFAULT_UAR_PAGE_SHIFT;
-		init_hca.uar_page_sz = DEFAULT_UAR_PAGE_SHIFT - 12;
+		if (enable_4k_uar) {
+			init_hca.log_uar_sz = ilog2(dev->caps.num_uars) +
+						    PAGE_SHIFT - DEFAULT_UAR_PAGE_SHIFT;
+			init_hca.uar_page_sz = DEFAULT_UAR_PAGE_SHIFT - 12;
+		} else {
+			init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
+			init_hca.uar_page_sz = PAGE_SHIFT - 12;
+		}
 
 		init_hca.mw_enabled = 0;
 		if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
-- 
cgit v0.10.2


From eee577232203842b4dcadb7ab477a298479633ed Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 17 Mar 2016 11:57:06 -0700
Subject: net: bcmgenet: fix dma api length mismatch

When un-mapping skb->data in __bcmgenet_tx_reclaim(),
we must use the length that was used in original dma_map_single(),
instead of skb->len that might be bigger (includes the frags)

We simply can store skb_len into tx_cb_ptr->dma_len and use it
at unmap time.

Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index d7e01a7..6746fd0 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -1197,7 +1197,7 @@ static unsigned int __bcmgenet_tx_reclaim(struct net_device *dev,
 			dev->stats.tx_bytes += tx_cb_ptr->skb->len;
 			dma_unmap_single(&dev->dev,
 					 dma_unmap_addr(tx_cb_ptr, dma_addr),
-					 tx_cb_ptr->skb->len,
+					 dma_unmap_len(tx_cb_ptr, dma_len),
 					 DMA_TO_DEVICE);
 			bcmgenet_free_cb(tx_cb_ptr);
 		} else if (dma_unmap_addr(tx_cb_ptr, dma_addr)) {
@@ -1308,7 +1308,7 @@ static int bcmgenet_xmit_single(struct net_device *dev,
 	}
 
 	dma_unmap_addr_set(tx_cb_ptr, dma_addr, mapping);
-	dma_unmap_len_set(tx_cb_ptr, dma_len, skb->len);
+	dma_unmap_len_set(tx_cb_ptr, dma_len, skb_len);
 	length_status = (skb_len << DMA_BUFLENGTH_SHIFT) | dma_desc_flags |
 			(priv->hw_params->qtag_mask << DMA_TX_QTAG_SHIFT) |
 			DMA_TX_APPEND_CRC;
-- 
cgit v0.10.2


From fe30937b65354c7fec244caebbdaae68e28ca797 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Thu, 17 Mar 2016 17:23:36 -0700
Subject: bonding: fix bond_get_stats()

bond_get_stats() can be called from rtnetlink (with RTNL held)
or from /proc/net/dev seq handler (with RCU held)

The logic added in commit 5f0c5f73e5ef ("bonding: make global bonding
stats more reliable") kind of assumed only one cpu could run there.

If multiple threads are reading /proc/net/dev, stats can be really
messed up after a while.

A second problem is that some fields are 32bit, so we need to properly
handle the wrap around problem.

Given that RTNL is not always held, we need to use
bond_for_each_slave_rcu().

Fixes: 5f0c5f73e5ef ("bonding: make global bonding stats more reliable")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Andy Gospodarek <gospo@cumulusnetworks.com>
Cc: Jay Vosburgh <j.vosburgh@gmail.com>
Cc: Veaceslav Falico <vfalico@gmail.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 270b39c..941ec99 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -3301,6 +3301,30 @@ static int bond_close(struct net_device *bond_dev)
 	return 0;
 }
 
+/* fold stats, assuming all rtnl_link_stats64 fields are u64, but
+ * that some drivers can provide 32bit values only.
+ */
+static void bond_fold_stats(struct rtnl_link_stats64 *_res,
+			    const struct rtnl_link_stats64 *_new,
+			    const struct rtnl_link_stats64 *_old)
+{
+	const u64 *new = (const u64 *)_new;
+	const u64 *old = (const u64 *)_old;
+	u64 *res = (u64 *)_res;
+	int i;
+
+	for (i = 0; i < sizeof(*_res) / sizeof(u64); i++) {
+		u64 nv = new[i];
+		u64 ov = old[i];
+
+		/* detects if this particular field is 32bit only */
+		if (((nv | ov) >> 32) == 0)
+			res[i] += (u32)nv - (u32)ov;
+		else
+			res[i] += nv - ov;
+	}
+}
+
 static struct rtnl_link_stats64 *bond_get_stats(struct net_device *bond_dev,
 						struct rtnl_link_stats64 *stats)
 {
@@ -3309,44 +3333,23 @@ static struct rtnl_link_stats64 *bond_get_stats(struct net_device *bond_dev,
 	struct list_head *iter;
 	struct slave *slave;
 
+	spin_lock(&bond->stats_lock);
 	memcpy(stats, &bond->bond_stats, sizeof(*stats));
 
-	bond_for_each_slave(bond, slave, iter) {
-		const struct rtnl_link_stats64 *sstats =
+	rcu_read_lock();
+	bond_for_each_slave_rcu(bond, slave, iter) {
+		const struct rtnl_link_stats64 *new =
 			dev_get_stats(slave->dev, &temp);
-		struct rtnl_link_stats64 *pstats = &slave->slave_stats;
-
-		stats->rx_packets +=  sstats->rx_packets - pstats->rx_packets;
-		stats->rx_bytes += sstats->rx_bytes - pstats->rx_bytes;
-		stats->rx_errors += sstats->rx_errors - pstats->rx_errors;
-		stats->rx_dropped += sstats->rx_dropped - pstats->rx_dropped;
-		stats->rx_nohandler += sstats->rx_nohandler - pstats->rx_nohandler;
-
-		stats->tx_packets += sstats->tx_packets - pstats->tx_packets;;
-		stats->tx_bytes += sstats->tx_bytes - pstats->tx_bytes;
-		stats->tx_errors += sstats->tx_errors - pstats->tx_errors;
-		stats->tx_dropped += sstats->tx_dropped - pstats->tx_dropped;
-
-		stats->multicast += sstats->multicast - pstats->multicast;
-		stats->collisions += sstats->collisions - pstats->collisions;
-
-		stats->rx_length_errors += sstats->rx_length_errors - pstats->rx_length_errors;
-		stats->rx_over_errors += sstats->rx_over_errors - pstats->rx_over_errors;
-		stats->rx_crc_errors += sstats->rx_crc_errors - pstats->rx_crc_errors;
-		stats->rx_frame_errors += sstats->rx_frame_errors - pstats->rx_frame_errors;
-		stats->rx_fifo_errors += sstats->rx_fifo_errors - pstats->rx_fifo_errors;
-		stats->rx_missed_errors += sstats->rx_missed_errors - pstats->rx_missed_errors;
-
-		stats->tx_aborted_errors += sstats->tx_aborted_errors - pstats->tx_aborted_errors;
-		stats->tx_carrier_errors += sstats->tx_carrier_errors - pstats->tx_carrier_errors;
-		stats->tx_fifo_errors += sstats->tx_fifo_errors - pstats->tx_fifo_errors;
-		stats->tx_heartbeat_errors += sstats->tx_heartbeat_errors - pstats->tx_heartbeat_errors;
-		stats->tx_window_errors += sstats->tx_window_errors - pstats->tx_window_errors;
+
+		bond_fold_stats(stats, new, &slave->slave_stats);
 
 		/* save off the slave stats for the next run */
-		memcpy(pstats, sstats, sizeof(*sstats));
+		memcpy(&slave->slave_stats, new, sizeof(*new));
 	}
+	rcu_read_unlock();
+
 	memcpy(&bond->bond_stats, stats, sizeof(*stats));
+	spin_unlock(&bond->stats_lock);
 
 	return stats;
 }
@@ -4160,6 +4163,7 @@ void bond_setup(struct net_device *bond_dev)
 	struct bonding *bond = netdev_priv(bond_dev);
 
 	spin_lock_init(&bond->mode_lock);
+	spin_lock_init(&bond->stats_lock);
 	bond->params = bonding_defaults;
 
 	/* Initialize pointers */
diff --git a/include/net/bonding.h b/include/net/bonding.h
index ee6c520..791800d 100644
--- a/include/net/bonding.h
+++ b/include/net/bonding.h
@@ -215,6 +215,7 @@ struct bonding {
 	 * ALB mode (6) - to sync the use and modifications of its hash table
 	 */
 	spinlock_t mode_lock;
+	spinlock_t stats_lock;
 	u8	 send_peer_notif;
 	u8       igmp_retrans;
 #ifdef CONFIG_PROC_FS
-- 
cgit v0.10.2